arXiv Papers of Vision-Language

Abstract:
Manipulating dynamic objects remains an open challenge for Vision-Language-Action (VLA) models, which, despite strong generalization in static manipulation, struggle in dynamic scenarios requiring rapid perception, temporal anticipation, and continuous control. We present DynamicVLA, a framework for dynamic object manipulation that integrates temporal reasoning and closed-loop adaptation through three key designs: 1) a compact 0.4B VLA using a convolutional vision encoder for spatially efficient, structurally faithful encoding, enabling fast multimodal inference; 2) Continuous Inference, enabling overlapping reasoning and execution for lower latency and timely adaptation to object motion; and 3) Latent-aware Action Streaming, which bridges the perception-execution gap by enforcing temporally aligned action execution. To fill the missing foundation of dynamic manipulation data, we introduce the Dynamic Object Manipulation (DOM) benchmark, built from scratch with an auto data collection pipeline that efficiently gathers 200K synthetic episodes across 2.8K scenes and 206 objects, and enables fast collection of 2K real-world episodes without teleoperation. Extensive evaluations demonstrate remarkable improvements in response speed, perception, and generalization, positioning DynamicVLA as a unified framework for general dynamic object manipulation across embodiments.

Paperid: 2, https://arxiv.org/pdf/2601.22069.pdf GitHub GitHub

Abstract:
Long-context reasoning has significantly empowered large language models (LLMs) to tackle complex tasks, yet it introduces severe efficiency bottlenecks due to the computational complexity. Existing efficient approaches often rely on complex additional training or external models for compression, which limits scalability and discards critical fine-grained information. In this paper, we propose VTC-R1, a new efficient reasoning paradigm that integrates vision-text compression into the reasoning process. Instead of processing lengthy textual traces, VTC-R1 renders intermediate reasoning segments into compact images, which are iteratively fed back into vision-language models as "optical memory." We construct a training dataset based on OpenR1-Math-220K achieving 3.4x token compression and fine-tune representative VLMs-Glyph and Qwen3-VL. Extensive experiments on benchmarks such as MATH500, AIME25, AMC23 and GPQA-D demonstrate that VTC-R1 consistently outperforms standard long-context reasoning. Furthermore, our approach significantly improves inference efficiency, achieving 2.7x speedup in end-to-end latency, highlighting its potential as a scalable solution for reasoning-intensive applications. Our code is available at https://github.com/w-yibo/VTC-R1.

Paperid: 3, https://arxiv.org/pdf/2601.21998.pdf GitHub

Abstract:
This work highlights that video world modeling, alongside vision-language pre-training, establishes a fresh and independent foundation for robot learning. Intuitively, video world models provide the ability to imagine the near future by understanding the causality between actions and visual dynamics. Inspired by this, we introduce LingBot-VA, an autoregressive diffusion framework that learns frame prediction and policy execution simultaneously. Our model features three carefully crafted designs: (1) a shared latent space, integrating vision and action tokens, driven by a Mixture-of-Transformers (MoT) architecture, (2) a closed-loop rollout mechanism, allowing for ongoing acquisition of environmental feedback with ground-truth observations, (3) an asynchronous inference pipeline, parallelizing action prediction and motor execution to support efficient control. We evaluate our model on both simulation benchmarks and real-world scenarios, where it shows significant promise in long-horizon manipulation, data efficiency in post-training, and strong generalizability to novel configurations. The code and model are made publicly available to facilitate the community.

Paperid: 4, https://arxiv.org/pdf/2601.21751.pdf GitHub

Abstract:
Vision-Language Navigation in Continuous Environments (VLN-CE) presents a core challenge: grounding high-level linguistic instructions into precise, safe, and long-horizon spatial actions. Explicit topological maps have proven to be a vital solution for providing robust spatial memory in such tasks. However, existing topological planning methods suffer from a "Granularity Rigidity" problem. Specifically, these methods typically rely on fixed geometric thresholds to sample nodes, which fails to adapt to varying environmental complexities. This rigidity leads to a critical mismatch: the model tends to over-sample in simple areas, causing computational redundancy, while under-sampling in high-uncertainty regions, increasing collision risks and compromising precision. To address this, we propose DGNav, a framework for Dynamic Topological Navigation, introducing a context-aware mechanism to modulate map density and connectivity on-the-fly. Our approach comprises two core innovations: (1) A Scene-Aware Adaptive Strategy that dynamically modulates graph construction thresholds based on the dispersion of predicted waypoints, enabling "densification on demand" in challenging environments; (2) A Dynamic Graph Transformer that reconstructs graph connectivity by fusing visual, linguistic, and geometric cues into dynamic edge weights, enabling the agent to filter out topological noise and enhancing instruction adherence. Extensive experiments on the R2R-CE and RxR-CE benchmarks demonstrate DGNav exhibits superior navigation performance and strong generalization capabilities. Furthermore, ablation studies confirm that our framework achieves an optimal trade-off between navigation efficiency and safe exploration. The code is available at https://github.com/shannanshouyin/DGNav.

Paperid: 5, https://arxiv.org/pdf/2601.21617.pdf GitHub

Abstract:
Vision-Language Models (VLMs) are advancing computational pathology with superior visual understanding capabilities. However, current systems often reduce diagnosis to directly output conclusions without verifiable evidence-linked reasoning, which severely limits clinical trust and hinders expert error rectification. To address these barriers, we construct PathReasoner, the first large-scale dataset of whole-slide image (WSI) reasoning. Unlike previous work reliant on unverified distillation, we develop a rigorous knowledge-guided generation pipeline. By leveraging medical knowledge graphs, we explicitly align structured pathological findings and clinical reasoning with diagnoses, generating over 20K high-quality instructional samples. Based on the database, we propose PathReasoner-R1, which synergizes trajectory-masked supervised fine-tuning with reasoning-oriented reinforcement learning to instill structured chain-of-thought capabilities. To ensure medical rigor, we engineer a knowledge-aware multi-granular reward function incorporating an Entity Reward mechanism strictly aligned with knowledge graphs. This effectively guides the model to optimize for logical consistency rather than mere outcome matching, thereby enhancing robustness. Extensive experiments demonstrate that PathReasoner-R1 achieves state-of-the-art performance on both PathReasoner and public benchmarks across various image scales, equipping pathology models with transparent, clinically grounded reasoning capabilities. Dataset and code are available at https://github.com/cyclexfy/PathReasoner-R1.

Paperid: 6, https://arxiv.org/pdf/2601.20867.pdf GitHub

Abstract:
Prompt tuning has achieved remarkable progress in vision-language models (VLMs) and is recently being adopted for audio-language models (ALMs). However, its generalization ability in ALMs remains largely underexplored. We observe that conventional prompt tuning for ALMs also suffers from the Base-New Tradeoff, and we identify that this issue stems from the disrupted semantic structure of the embedding space. To address this issue, we propose Semantically Expanded Prompt Tuning (SEPT)-a plug-and-play framework that explicitly regularizes the prompt embedding space by incorporating semantic neighbors generated by large language models. SEPT introduces a novel semantic expansion loss with margin constraints that promote intra-class compactness and inter-class separability, thereby enhancing the semantic structure of the prompt embedding space. For comprehensive evaluation, we establish the first benchmark setup for prompt generalization in ALMs, covering both base-to-new generalization and cross-dataset transferability. Extensive experiments demonstrate that SEPT consistently improves generalization performance across multiple prompt tuning baselines, while maintaining computational cost during inference. Codes are available in https://github.com/jhyukjang/SEPT.

Paperid: 7, https://arxiv.org/pdf/2601.20675.pdf GitHub

Abstract:
Prompt learning (PL) has emerged as an effective strategy to adapt vision-language models (VLMs), such as CLIP, for downstream tasks under limited supervision. While PL has demonstrated strong generalization on natural image datasets, its transferability to remote sensing (RS) imagery remains underexplored. RS data present unique challenges, including multi-label scenes, high intra-class variability, and diverse spatial resolutions, that hinder the direct applicability of existing PL methods. In particular, current prompt-based approaches often struggle to identify dominant semantic cues and fail to generalize to novel classes in RS scenarios. To address these challenges, we propose BiMoRS, a lightweight bi-modal prompt learning framework tailored for RS tasks. BiMoRS employs a frozen image captioning model (e.g., BLIP-2) to extract textual semantic summaries from RS images. These captions are tokenized using a BERT tokenizer and fused with high-level visual features from the CLIP encoder. A lightweight cross-attention module then conditions a learnable query prompt on the fused textual-visual representation, yielding contextualized prompts without altering the CLIP backbone. We evaluate BiMoRS on four RS datasets across three domain generalization (DG) tasks and observe consistent performance gains, outperforming strong baselines by up to 2% on average. Codes are available at https://github.com/ipankhi/BiMoRS.

Paperid: 8, https://arxiv.org/pdf/2601.20552.pdf GitHub

Abstract:
We present DeepSeek-OCR 2 to investigate the feasibility of a novel encoder-DeepEncoder V2-capable of dynamically reordering visual tokens upon image semantics. Conventional vision-language models (VLMs) invariably process visual tokens in a rigid raster-scan order (top-left to bottom-right) with fixed positional encoding when fed into LLMs. However, this contradicts human visual perception, which follows flexible yet semantically coherent scanning patterns driven by inherent logical structures. Particularly for images with complex layouts, human vision exhibits causally-informed sequential processing. Inspired by this cognitive mechanism, DeepEncoder V2 is designed to endow the encoder with causal reasoning capabilities, enabling it to intelligently reorder visual tokens prior to LLM-based content interpretation. This work explores a novel paradigm: whether 2D image understanding can be effectively achieved through two-cascaded 1D causal reasoning structures, thereby offering a new architectural approach with the potential to achieve genuine 2D reasoning. Codes and model weights are publicly accessible at http://github.com/deepseek-ai/DeepSeek-OCR-2.

Paperid: 9, https://arxiv.org/pdf/2601.20524.pdf GitHub

Abstract:
Zero-shot anomaly detection aims to detect and localise abnormal regions in the image without access to any in-domain training images. While recent approaches leverage vision-language models (VLMs), such as CLIP, to transfer high-level concept knowledge, methods based on purely vision foundation models (VFMs), like DINOv2, have lagged behind in performance. We argue that this gap stems from two practical issues: (i) limited diversity in existing auxiliary anomaly detection datasets and (ii) overly shallow VFM adaptation strategies. To address both challenges, we propose AnomalyVFM, a general and effective framework that turns any pretrained VFM into a strong zero-shot anomaly detector. Our approach combines a robust three-stage synthetic dataset generation scheme with a parameter-efficient adaptation mechanism, utilising low-rank feature adapters and a confidence-weighted pixel loss. Together, these components enable modern VFMs to substantially outperform current state-of-the-art methods. More specifically, with RADIO as a backbone, AnomalyVFM achieves an average image-level AUROC of 94.1% across 9 diverse datasets, surpassing previous methods by significant 3.3 percentage points. Project Page: https://maticfuc.github.io/anomaly_vfm/

Paperid: 10, https://arxiv.org/pdf/2601.20357.pdf GitHub

Abstract:
Speculative decoding (SD) has proven effective for accelerating LLM inference by quickly generating draft tokens and verifying them in parallel. However, SD remains largely unexplored for Large Vision-Language Models (LVLMs), which extend LLMs to process both image and text prompts. To address this gap, we benchmark existing inference methods with small draft models on 11 datasets across diverse input scenarios and observe scenario-specific performance fluctuations. Motivated by these findings, we propose Test-time Adaptive Batched Ensemble Drafting (TABED), which dynamically ensembles multiple drafts obtained via batch inference by leveraging deviations from past ground truths available in the SD setting. The dynamic ensemble method achieves an average robust walltime speedup of 1.74x over autoregressive decoding and a 5% improvement over single drafting methods, while remaining training-free and keeping ensembling costs negligible through parameter sharing. With its plug-and-play compatibility, we further enhance TABED by integrating advanced verification and alternative drafting methods. Code and custom-trained models are available at https://github.com/furiosa-ai/TABED.

Paperid: 11, https://arxiv.org/pdf/2601.20334.pdf GitHub

Abstract:
Robotic manipulation has increasingly adopted vision-language-action (VLA) models, which achieve strong performance but typically require task-specific demonstrations and fine-tuning, and often generalize poorly under domain shift. We investigate whether general-purpose large language model (LLM) agent frameworks, originally developed for software engineering, can serve as an alternative control paradigm for embodied manipulation. We introduce FAEA (Frontier Agent as Embodied Agent), which applies an LLM agent framework directly to embodied manipulation without modification. Using the same iterative reasoning that enables software agents to debug code, FAEA enables embodied agents to reason through manipulation strategies. We evaluate an unmodified frontier agent, Claude Agent SDK, across the LIBERO, ManiSkill3, and MetaWorld benchmarks. With privileged environment state access, FAEA achieves success rates of 84.9%, 85.7%, and 96%, respectively. This level of task success approaches that of VLA models trained with less than 100 demonstrations per task, without requiring demonstrations or fine-tuning. With one round of human feedback as an optional optimization, performance increases to 88.2% on LIBERO. This demonstration-free capability has immediate practical value: FAEA can autonomously explore novel scenarios in simulation and generate successful trajectories for training data augmentation in embodied learning. Our results indicate that general-purpose agents are sufficient for a class of manipulation tasks dominated by deliberative, task-level planning. This opens a path for robotics systems to leverage actively maintained agent infrastructure and benefit directly from ongoing advances in frontier models. Code is available at https://github.com/robiemusketeer/faea-sim

Paperid: 12, https://arxiv.org/pdf/2601.20279.pdf GitHub

Abstract:
Recent studies have examined attention dynamics in large vision-language models (LVLMs) to detect hallucinations. However, existing approaches remain limited in reliably distinguishing hallucinated from factually grounded outputs, as they rely solely on forward-pass attention patterns and neglect gradient-based signals that reveal how token influence propagates through the network. To bridge this gap, we introduce LVLMs-Saliency, a gradient-aware diagnostic framework that quantifies the visual grounding strength of each output token by fusing attention weights with their input gradients. Our analysis uncovers a decisive pattern: hallucinations frequently arise when preceding output tokens exhibit low saliency toward the prediction of the next token, signaling a breakdown in contextual memory retention. Leveraging this insight, we propose a dual-mechanism inference-time framework to mitigate hallucinations: (1) Saliency-Guided Rejection Sampling (SGRS), which dynamically filters candidate tokens during autoregressive decoding by rejecting those whose saliency falls below a context-adaptive threshold, thereby preventing coherence-breaking tokens from entering the output sequence; and (2) Local Coherence Reinforcement (LocoRE), a lightweight, plug-and-play module that strengthens attention from the current token to its most recent predecessors, actively counteracting the contextual forgetting behavior identified by LVLMs-Saliency. Extensive experiments across multiple LVLMs demonstrate that our method significantly reduces hallucination rates while preserving fluency and task performance, offering a robust and interpretable solution for enhancing model reliability. Code is available at: https://github.com/zhangbaijin/LVLMs-Saliency

Paperid: 13, https://arxiv.org/pdf/2601.19850.pdf GitHub

Abstract:
Robust 3D hand reconstruction in egocentric vision is challenging due to depth ambiguity, self-occlusion, and complex hand-object interactions. Prior methods mitigate these issues by scaling training data or adding auxiliary cues, but they often struggle in unseen contexts. We present EgoHandICL, the first in-context learning (ICL) framework for 3D hand reconstruction that improves semantic alignment, visual consistency, and robustness under challenging egocentric conditions. EgoHandICL introduces complementary exemplar retrieval guided by vision-language models (VLMs), an ICL-tailored tokenizer for multimodal context, and a masked autoencoder (MAE)-based architecture trained with hand-guided geometric and perceptual objectives. Experiments on ARCTIC and EgoExo4D show consistent gains over state-of-the-art methods. We also demonstrate real-world generalization and improve EgoVLM hand-object interaction reasoning by using reconstructed hands as visual prompts. Code and data: https://github.com/Nicous20/EgoHandICL

Paperid: 14, https://arxiv.org/pdf/2601.19659.pdf GitHub

Abstract:
Continual learning for pre-trained vision-language models requires balancing three competing objectives: retaining pre-trained knowledge, preserving knowledge from a sequence of learned tasks, and maintaining the plasticity to acquire new knowledge. This paper presents a simple but effective approach called KeepLoRA to effectively balance these objectives. We first analyze the knowledge retention mechanism within the model parameter space and find that general knowledge is mainly encoded in the principal subspace, while task-specific knowledge is encoded in the residual subspace. Motivated by this finding, KeepLoRA learns new tasks by restricting LoRA parameter updates in the residual subspace to prevent interfering with previously learned capabilities. Specifically, we infuse knowledge for a new task by projecting its gradient onto a subspace orthogonal to both the principal subspace of pre-trained model and the dominant directions of previous task features. Our theoretical and empirical analyses confirm that KeepLoRA balances the three objectives and achieves state-of-the-art performance. The implementation code is available at https://github.com/MaolinLuo/KeepLoRA.

Paperid: 15, https://arxiv.org/pdf/2601.19433.pdf GitHub

Abstract:
Generating immersive 3D scenes from texts is a core task in computer vision, crucial for applications in virtual reality and game development. Despite the promise of leveraging 2D diffusion priors, existing methods suffer from spatial blindness and rely on predefined trajectories that fail to exploit the inner relationships among salient objects. Consequently, these approaches are unable to comprehend the semantic layout, preventing them from exploring the scene adaptively to infer occluded content. Moreover, current inpainting models operate in 2D image space, struggling to plausibly fill holes caused by camera motion. To address these limitations, we propose RoamScene3D, a novel framework that bridges the gap between semantic guidance and spatial generation. Our method reasons about the semantic relations among objects and produces consistent and photorealistic scenes. Specifically, we employ a vision-language model (VLM) to construct a scene graph that encodes object relations, guiding the camera to perceive salient object boundaries and plan an adaptive roaming trajectory. Furthermore, to mitigate the limitations of static 2D priors, we introduce a Motion-Injected Inpainting model that is fine-tuned on a synthetic panoramic dataset integrating authentic camera trajectories, making it adaptive to camera motion. Extensive experiments demonstrate that with semantic reasoning and geometric constraints, our method significantly outperforms state-of-the-art approaches in producing consistent and photorealistic scenes. Our code is available at https://github.com/JS-CHU/RoamScene3D.

Paperid: 16, https://arxiv.org/pdf/2601.19210.pdf GitHub

Abstract:
Vision-language models (VLMs) such as CLIP have demonstrated remarkable zero-shot generalization, yet remain highly vulnerable to adversarial examples (AEs). While test-time defenses are promising, existing methods fail to provide sufficient robustness against strong attacks and are often hampered by high inference latency and task-specific applicability. To address these limitations, we start by investigating the intrinsic properties of AEs, which reveals that AEs exhibit severe feature inconsistency under progressive frequency attenuation. We further attribute this to the model's inherent spectral bias. Leveraging this insight, we propose an efficient test-time defense named Contrastive Spectral Rectification (CSR). CSR optimizes a rectification perturbation to realign the input with the natural manifold under a spectral-guided contrastive objective, which is applied input-adaptively. Extensive experiments across 16 classification benchmarks demonstrate that CSR outperforms the SOTA by an average of 18.1% against strong AutoAttack with modest inference overhead. Furthermore, CSR exhibits broad applicability across diverse visual tasks. Code is available at https://github.com/Summu77/CSR.

Paperid: 17, https://arxiv.org/pdf/2601.19204.pdf GitHub

Abstract:
Recent vision-language models have strong perceptual ability but their implicit reasoning is hard to explain and easily generates hallucinations on complex queries. Compositional methods improve interpretability, but most rely on a single agent or hand-crafted pipeline and cannot decide when to collaborate across complementary agents or compete among overlapping ones. We introduce MATA (Multi-Agent hierarchical Trainable Automaton), a multi-agent system presented as a hierarchical finite-state automaton for visual reasoning whose top-level transitions are chosen by a trainable hyper agent. Each agent corresponds to a state in the hyper automaton, and runs a small rule-based sub-automaton for reliable micro-control. All agents read and write a shared memory, yielding transparent execution history. To supervise the hyper agent's transition policy, we build transition-trajectory trees and transform to memory-to-next-state pairs, forming the MATA-SFT-90K dataset for supervised finetuning (SFT). The finetuned LLM as the transition policy understands the query and the capacity of agents, and it can efficiently choose the optimal agent to solve the task. Across multiple visual reasoning benchmarks, MATA achieves the state-of-the-art results compared with monolithic and compositional baselines. The code and dataset are available at https://github.com/ControlNet/MATA.

Paperid: 18, https://arxiv.org/pdf/2601.18798.pdf GitHub

Abstract:
ECG-Language Models (ELMs) extend recent progress in Multimodal Large Language Models (MLLMs) to automated ECG interpretation. However, most ELMs follow Vision-Language Model (VLM) designs and depend on pretrained ECG encoders, adding architectural and training complexity. Inspired by encoder-free VLMs, we introduce ELF, an encoder-free ELM that replaces the ECG encoder with a single projection layer trained jointly with the LLM. Across five datasets, ELF matches or exceeds state-of-the-art ELMs that use far more complex encoders and training pipelines. We also test whether adding architectural biases to ELF improves performance and find that the single linear projection remains competitive. Finally, we show that ELF, and potentially other ELMs, often rely more on benchmark artifacts and language priors than ECG-derived information, highlighting limitations in current evaluation practices and ELM design. All data and code is available at https://github.com/willxxy/ECG-Bench.

Paperid: 19, https://arxiv.org/pdf/2601.18190.pdf GitHub GitHub

Abstract:
Vision-Language Pre-training (VLP) models like CLIP have significantly advanced Remote Sensing Image-Text Retrieval (RSITR). However, existing methods predominantly rely on coarse-grained global alignment, which often overlooks the dense, multi-scale semantics inherent in overhead imagery. Moreover, adapting these heavy models via full fine-tuning incurs prohibitive computational costs and risks catastrophic forgetting. To address these challenges, we propose MPS-CLIP, a parameter-efficient framework designed to shift the retrieval paradigm from global matching to keyword-guided fine-grained alignment. Specifically, we leverage a Large Language Model (LLM) to extract core semantic keywords, guiding the Segment Anything Model (SamGeo) to generate semantically relevant sub-perspectives. To efficiently adapt the frozen backbone, we introduce a Gated Global Attention (G^2A) adapter, which captures global context and long-range dependencies with minimal overhead. Furthermore, a Multi-Perspective Representation (MPR) module aggregates these local cues into robust multi-perspective embeddings. The framework is optimized via a hybrid objective combining multi-perspective contrastive and weighted triplet losses, which dynamically selects maximum-response perspectives to suppress noise and enforce precise semantic matching. Extensive experiments on the RSICD and RSITMD benchmarks demonstrate that MPS-CLIP achieves state-of-the-art performance with 35.18% and 48.40% mean Recall (mR), respectively, significantly outperforming full fine-tuning baselines and recent competitive methods. Code is available at https://github.com/Lcrucial1f/MPS-CLIP.

Paperid: 20, https://arxiv.org/pdf/2601.17918.pdf GitHub

Abstract:
Large Vision-Language Models (LVLMs) hold significant promise for medical applications, yet their deployment is often constrained by insufficient alignment and reliability. While Direct Preference Optimization (DPO) has emerged as a potent framework for refining model responses, its efficacy in high-stakes medical contexts remains underexplored, lacking the rigorous empirical groundwork necessary to guide future methodological advances. To bridge this gap, we present the first comprehensive examination of diverse DPO variants within the medical domain, evaluating nine distinct formulations across two medical LVLMs: LLaVA-Med and HuatuoGPT-Vision. Our results reveal several critical limitations: current DPO approaches often yield inconsistent gains over supervised fine-tuning, with their efficacy varying significantly across different tasks and backbones. Furthermore, they frequently fail to resolve fundamental visual misinterpretation errors. Building on these insights, we present a targeted preference construction strategy as a proof-of-concept that explicitly addresses visual misinterpretation errors frequently observed in existing DPO models. This design yields a 3.6% improvement over the strongest existing DPO baseline on visual question-answering tasks. To support future research, we release our complete framework, including all training data, model checkpoints, and our codebase at https://github.com/dmis-lab/med-vlm-dpo.

Paperid: 21, https://arxiv.org/pdf/2601.17885.pdf GitHub

Abstract:
Bimanual manipulation in cluttered scenes requires policies that remain stable under occlusions, viewpoint and scene variations. Existing vision-language-action models often fail to generalize because (i) multi-view features are fused via view-agnostic token concatenation, yielding weak 3D-consistent spatial understanding, and (ii) language is injected as global conditioning, resulting in coarse instruction grounding. In this paper, we introduce PEAfowl, a perception-enhanced multi-view VLA policy for bimanual manipulation. For spatial reasoning, PEAfowl predicts per-token depth distributions, performs differentiable 3D lifting, and aggregates local cross-view neighbors to form geometrically grounded, cross-view consistent representations. For instruction grounding, we propose to replace global conditioning with a Perceiver-style text-aware readout over frozen CLIP visual features, enabling iterative evidence accumulation. To overcome noisy and incomplete commodity depth without adding inference overhead, we apply training-only depth distillation from a pretrained depth teacher to supervise the depth-distribution head, providing perception front-end with geometry-aware priors. On RoboTwin 2.0 under domain-randomized setting, PEAfowl improves the strongest baseline by 23.0 pp in success rate, and real-robot experiments further demonstrate reliable sim-to-real transfer and consistent improvements from depth distillation. Project website: https://peafowlvla.github.io/.

Paperid: 22, https://arxiv.org/pdf/2601.17814.pdf GitHub

Abstract:
Multimodal large language models (MLLMs) have advanced rapidly, yet heterogeneity in architecture, alignment strategies, and efficiency means that no single model is uniformly superior across tasks. In practical deployments, workloads span lightweight OCR to complex multimodal reasoning; using one MLLM for all queries either over-provisions compute on easy instances or sacrifices accuracy on hard ones. Query-level model selection (routing) addresses this tension, but extending routing from text-only LLMs to MLLMs is nontrivial due to modality fusion, wide variation in computational cost across models, and the absence of a standardized, budget-aware evaluation. We present MMR-Bench, a unified benchmark that isolates the multimodal routing problem and enables comparison under fixed candidate sets and cost models. MMR-Bench provides (i) a controlled environment with modality-aware inputs and variable compute budgets, (ii) a broad suite of vision-language tasks covering OCR, general VQA, and multimodal math reasoning, and (iii) strong single-model reference, oracle upper bounds, and representative routing policies. Using MMR-Bench, we show that incorporating multimodal signals improves routing quality. Empirically, these cues improve the cost-accuracy frontier and enable the routed system to exceed the strongest single model's accuracy at roughly 33% of its cost. Furthermore, policies trained on a subset of models and tasks generalize zero-shot to new datasets and text-only benchmarks without retuning, establishing MMR-Bench as a foundation for studying adaptive multimodal model selection and efficient MLLM deployment. The code will be available at: https://github.com/Hunter-Wrynn/MMR-Bench.

Paperid: 23, https://arxiv.org/pdf/2601.17706.pdf GitHub

Abstract:
Images often communicate more than they literally depict: a set of tools can suggest an occupation and a cultural artifact can suggest a tradition. This kind of indirect visual reference, known as visual metonymy, invites viewers to recover a target concept via associated cues rather than explicit depiction. In this work, we present the first computational investigation of visual metonymy. We introduce a novel pipeline grounded in semiotic theory that leverages large language models and text-to-image models to generate metonymic visual representations. Using this framework, we construct ViMET, the first visual metonymy dataset comprising 2,000 multiple-choice questions to evaluate the cognitive reasoning abilities in multimodal language models. Experimental results on our dataset reveal a significant gap between human performance (86.9%) and state-of-the-art vision-language models (65.9%), highlighting limitations in machines' ability to interpret indirect visual references. Our dataset is publicly available at: https://github.com/cincynlp/ViMET.

Paperid: 24, https://arxiv.org/pdf/2601.17657.pdf GitHub

Abstract:
Contrastive Language-Image Pre-training (CLIP) has accomplished extraordinary success for semantic understanding but inherently struggles to perceive geometric structure. Existing methods attempt to bridge this gap by querying CLIP with textual prompts, a process that is often indirect and inefficient. This paper introduces a fundamentally different approach using a dual-pathway decoder. We present SPACE-CLIP, an architecture that unlocks and interprets latent geometric knowledge directly from a frozen CLIP vision encoder, completely bypassing the text encoder and its associated textual prompts. A semantic pathway interprets high-level features, dynamically conditioned on global context using feature-wise linear modulation (FiLM). In addition, a structural pathway extracts fine-grained spatial details from early layers. These complementary streams are hierarchically fused, enabling a robust synthesis of semantic context and precise geometry. Extensive experiments on the KITTI benchmark show that SPACE-CLIP dramatically outperforms previous CLIP-based methods. Our ablation studies validate that the synergistic fusion of our dual pathways is critical to this success. SPACE-CLIP offers a new, efficient, and architecturally elegant blueprint for repurposing large-scale vision models. The proposed method is not just a standalone depth estimator, but a readily integrable spatial perception module for the next generation of embodied AI systems, such as vision-language-action (VLA) models. Our model is available at https://github.com/taewan2002/space-clip

Paperid: 25, https://arxiv.org/pdf/2601.17420.pdf GitHub

Abstract:
Existing works of reasoning segmentation often fall short in complex cases, particularly when addressing complicated queries and out-of-domain images. Inspired by the chain-of-thought reasoning, where harder problems require longer thinking steps/time, this paper aims to explore a system that can think step-by-step, look up information if needed, generate results, self-evaluate its own results, and refine the results, in the same way humans approach harder questions. We introduce CoT-Seg, a training-free framework that rethinks reasoning segmentation by combining chain-of-thought reasoning with self-correction. Instead of fine-tuning, CoT-Seg leverages the inherent reasoning ability of pre-trained MLLMs (GPT-4o) to decompose queries into meta-instructions, extract fine-grained semantics from images, and identify target objects even under implicit or complex prompts. Moreover, CoT-Seg incorporates a self-correction stage: the model evaluates its own segmentation against the original query and reasoning trace, identifies mismatches, and iteratively refines the mask. This tight integration of reasoning and correction significantly improves reliability and robustness, especially in ambiguous or error-prone cases. Furthermore, our CoT-Seg framework allows easy incorporation of retrieval-augmented reasoning, enabling the system to access external knowledge when the input lacks sufficient information. To showcase CoT-Seg's ability to handle very challenging cases ,we introduce a new dataset ReasonSeg-Hard. Our results highlight that combining chain-of-thought reasoning, self-correction, offers a powerful paradigm for vision-language integration driven segmentation.

Paperid: 26, https://arxiv.org/pdf/2601.17383.pdf GitHub

Abstract:
Large Vision-Language Models (LVLMs) are increasingly deployed in real-world intelligent systems for perception and reasoning in open physical environments. While LVLMs are known to be vulnerable to prompt injection attacks, existing methods either require access to input channels or depend on knowledge of user queries, assumptions that rarely hold in practical deployments. We propose the first Physical Prompt Injection Attack (PPIA), a black-box, query-agnostic attack that embeds malicious typographic instructions into physical objects perceivable by the LVLM. PPIA requires no access to the model, its inputs, or internal pipeline, and operates solely through visual observation. It combines offline selection of highly recognizable and semantically effective visual prompts with strategic environment-aware placement guided by spatiotemporal attention, ensuring that the injected prompts are both perceivable and influential on model behavior. We evaluate PPIA across 10 state-of-the-art LVLMs in both simulated and real-world settings on tasks including visual question answering, planning, and navigation, PPIA achieves attack success rates up to 98%, with strong robustness under varying physical conditions such as distance, viewpoint, and illumination. Our code is publicly available at https://github.com/2023cghacker/Physical-Prompt-Injection-Attack.

Paperid: 27, https://arxiv.org/pdf/2601.17197.pdf GitHub

Abstract:
Vision-language models (VLMs) have demonstrated strong reasoning abilities in literal multimodal tasks such as visual mathematics and science question answering. However, figurative language, such as sarcasm, humor, and metaphor, remains a significant challenge, as it conveys intent and emotion through subtle incongruities between expressed and intended meanings. In multimodal settings, accompanying images can amplify or invert textual meaning, demanding models that reason across modalities and account for subjectivity. We propose a three-step framework for developing efficient multimodal reasoning models that can (i) interpret multimodal figurative language, (ii) provide transparent reasoning traces, and (iii) generalize across multiple figurative styles. Experiments across four styles show that (1) incorporating reasoning traces substantially improves multimodal figurative understanding, (2) reasoning learned in one style can transfer to others, especially between related styles like sarcasm and humor, and (3) training jointly across styles yields a generalized reasoning VLM that outperforms much larger open- and closed-source models. Our findings show that lightweight VLMs with verifiable reasoning achieve robust cross-style generalization while providing inspectable reasoning traces for multimodal tasks. The code and implementation are available at https://github.com/scheshmi/CrossStyle-MMR.

Paperid: 28, https://arxiv.org/pdf/2601.17037.pdf GitHub

Abstract:
We investigated visual reasoning limitations of both multimodal large language models (MLLMs) and image generation models (IGMs) by creating a novel benchmark to systematically compare failure modes across image-to-text and text-to-image tasks, enabling cross-modal evaluation of visual understanding. Despite rapid growth in machine learning, vision language models (VLMs) still fail to understand or generate basic visual concepts such as object orientation, quantity, or spatial relationships, which highlighted gaps in elementary visual reasoning. By adapting MMVP benchmark questions into explicit and implicit prompts, we create \textit{AMVICC}, a novel benchmark for profiling failure modes across various modalities. After testing 11 MLLMs and 3 IGMs in nine categories of visual reasoning, our results show that failure modes are often shared between models and modalities, but certain failures are model-specific and modality-specific, and this can potentially be attributed to various factors. IGMs consistently struggled to manipulate specific visual components in response to prompts, especially in explicit prompts, suggesting poor control over fine-grained visual attributes. Our findings apply most directly to the evaluation of existing state-of-the-art models on structured visual reasoning tasks. This work lays the foundation for future cross-modal alignment studies, offering a framework to probe whether generation and interpretation failures stem from shared limitations to guide future improvements in unified vision-language modeling.

Paperid: 29, https://arxiv.org/pdf/2601.16973.pdf GitHub

Abstract:
Modern Vision-Language Models (VLMs) remain poorly characterized in multi-step visual interactions, particularly in how they integrate perception, memory, and action over long horizons. We introduce VisGym, a gymnasium of 17 environments for evaluating and training VLMs. The suite spans symbolic puzzles, real-image understanding, navigation, and manipulation, and provides flexible controls over difficulty, input representation, planning horizon, and feedback. We also provide multi-step solvers that generate structured demonstrations, enabling supervised finetuning. Our evaluations show that all frontier models struggle in interactive settings, achieving low success rates in both the easy (46.6%) and hard (26.0%) configurations. Our experiments reveal notable limitations: models struggle to effectively leverage long context, performing worse with an unbounded history than with truncated windows. Furthermore, we find that several text-based symbolic tasks become substantially harder once rendered visually. However, explicit goal observations, textual feedback, and exploratory demonstrations in partially observable or unknown-dynamics settings for supervised finetuning yield consistent gains, highlighting concrete failure modes and pathways for improving multi-step visual decision-making. Code, data, and models can be found at: https://visgym.github.io/.

Paperid: 30, https://arxiv.org/pdf/2601.16429.pdf GitHub

Abstract:
Existing face-swapping methods often deliver competitive results in constrained settings but exhibit substantial quality degradation when handling extreme facial poses. To improve facial pose robustness, explicit geometric features are applied, but this approach remains problematic since it introduces additional dependencies and increases computational cost. Diffusion-based methods have achieved remarkable results; however, they are impractical for real-time processing. We introduce AlphaFace, which leverages an open-source vision-language model and CLIP image and text embeddings to apply novel visual and textual semantic contrastive losses. AlphaFace enables stronger identity representation and more precise attribute preservation, all while maintaining real-time performance. Comprehensive experiments across FF++, MPIE, and LPFF demonstrate that AlphaFace surpasses state-of-the-art methods in pose-challenging cases. The project is publicly available on `https://github.com/andrewyu90/Alphaface_Official.git'.

Paperid: 31, https://arxiv.org/pdf/2601.15408.pdf GitHub

Abstract:
Medical vision-language models can automate the generation of radiology reports but struggle with accurate visual grounding and factual consistency. Existing models often misalign textual findings with visual evidence, leading to unreliable or weakly grounded predictions. We present CURE, an error-aware curriculum learning framework that improves grounding and report quality without any additional data. CURE fine-tunes a multimodal instructional model on phrase grounding, grounded report generation, and anatomy-grounded report generation using public datasets. The method dynamically adjusts sampling based on model performance, emphasizing harder samples to improve spatial and textual alignment. CURE improves grounding accuracy by +0.37 IoU, boosts report quality by +0.188 CXRFEScore, and reduces hallucinations by 18.6%. CURE is a data-efficient framework that enhances both grounding accuracy and report reliability. Code is available at https://github.com/PabloMessina/CURE and model weights at https://huggingface.co/pamessina/medgemma-4b-it-cure

Paperid: 32, https://arxiv.org/pdf/2601.15316.pdf GitHub GitHub

Abstract:
In recent years, the rapid evolution of large vision-language models (LVLMs) has driven a paradigm shift in multimodal fake news detection (MFND), transforming it from traditional feature-engineering approaches to unified, end-to-end multimodal reasoning frameworks. Early methods primarily relied on shallow fusion techniques to capture correlations between text and images, but they struggled with high-level semantic understanding and complex cross-modal interactions. The emergence of LVLMs has fundamentally changed this landscape by enabling joint modeling of vision and language with powerful representation learning, thereby enhancing the ability to detect misinformation that leverages both textual narratives and visual content. Despite these advances, the field lacks a systematic survey that traces this transition and consolidates recent developments. To address this gap, this paper provides a comprehensive review of MFND through the lens of LVLMs. We first present a historical perspective, mapping the evolution from conventional multimodal detection pipelines to foundation model-driven paradigms. Next, we establish a structured taxonomy covering model architectures, datasets, and performance benchmarks. Furthermore, we analyze the remaining technical challenges, including interpretability, temporal reasoning, and domain generalization. Finally, we outline future research directions to guide the next stage of this paradigm shift. To the best of our knowledge, this is the first comprehensive survey to systematically document and analyze the transformative role of LVLMs in combating multimodal fake news. The summary of existing methods mentioned is in our Github: \href{https://github.com/Tan-YiLong/Overview-of-Fake-News-Detection}{https://github.com/Tan-YiLong/Overview-of-Fake-News-Detection}.

Paperid: 33, https://arxiv.org/pdf/2601.15287.pdf GitHub

Abstract:
Large language models (LLMs) deliver impressive results for a variety of tasks, but state-of-the-art systems require fast GPUs with large amounts of memory. To reduce both the memory and latency of these systems, practitioners quantize their learned parameters, typically at half precision. A growing body of research focuses on preserving the model performance with more aggressive bit widths, and some work has been done to apply these strategies to other models, like vision transformers. In our study we investigate how a variety of quantization methods, including state-of-the-art GPTQ and AWQ, can be applied effectively to multimodal pipelines comprised of vision models, language models, and their connectors. We address how performance on captioning, retrieval, and question answering can be affected by bit width, quantization method, and which portion of the pipeline the quantization is used for. Results reveal that ViT and LLM exhibit comparable importance in model performance, despite significant differences in parameter size, and that lower-bit quantization of the LLM achieves high accuracy at reduced bits per weight (bpw). These findings provide practical insights for efficient deployment of MLLMs and highlight the value of exploration for understanding component sensitivities in multimodal models. Our code is available at https://github.com/gautomdas/mmq.

Paperid: 34, https://arxiv.org/pdf/2601.15224.pdf GitHub

Abstract:
Estimating task progress requires reasoning over long-horizon dynamics rather than recognizing static visual content. While modern Vision-Language Models (VLMs) excel at describing what is visible, it remains unclear whether they can infer how far a task has progressed from partial observations. To this end, we introduce Progress-Bench, a benchmark for systematically evaluating progress reasoning in VLMs. Beyond benchmarking, we further explore a human-inspired two-stage progress reasoning paradigm through both training-free prompting and training-based approach based on curated dataset ProgressLM-45K. Experiments on 14 VLMs show that most models are not yet ready for task progress estimation, exhibiting sensitivity to demonstration modality and viewpoint changes, as well as poor handling of unanswerable cases. While training-free prompting that enforces structured progress reasoning yields limited and model-dependent gains, the training-based ProgressLM-3B achieves consistent improvements even at a small model scale, despite being trained on a task set fully disjoint from the evaluation tasks. Further analyses reveal characteristic error patterns and clarify when and why progress reasoning succeeds or fails.

Paperid: 35, https://arxiv.org/pdf/2601.15115.pdf GitHub

Abstract:
Hateful videos pose serious risks by amplifying discrimination, inciting violence, and undermining online safety. Existing training-based hateful video detection methods are constrained by limited training data and lack of interpretability, while directly prompting large vision-language models often struggle to deliver reliable hate detection. To address these challenges, this paper introduces MARS, a training-free Multi-stage Adversarial ReaSoning framework that enables reliable and interpretable hateful content detection. MARS begins with the objective description of video content, establishing a neutral foundation for subsequent analysis. Building on this, it develops evidence-based reasoning that supports potential hateful interpretations, while in parallel incorporating counter-evidence reasoning to capture plausible non-hateful perspectives. Finally, these perspectives are synthesized into a conclusive and explainable decision. Extensive evaluation on two real-world datasets shows that MARS achieves up to 10% improvement under certain backbones and settings compared to other training-free approaches and outperforms state-of-the-art training-based methods on one dataset. In addition, MARS produces human-understandable justifications, thereby supporting compliance oversight and enhancing the transparency of content moderation workflows. The code is available at https://github.com/Multimodal-Intelligence-Lab-MIL/MARS.

Paperid: 36, https://arxiv.org/pdf/2601.14750.pdf GitHub

Abstract:
Chain-of-Thought (CoT) prompting has achieved remarkable success in unlocking the reasoning capabilities of Large Language Models (LLMs). Although CoT prompting enhances reasoning, its verbosity imposes substantial computational overhead. Recent works often focus exclusively on outcome alignment and lack supervision on the intermediate reasoning process. These deficiencies obscure the analyzability of the latent reasoning chain. To address these challenges, we introduce Render-of-Thought (RoT), the first framework to reify the reasoning chain by rendering textual steps into images, making the latent rationale explicit and traceable. Specifically, we leverage the vision encoders of existing Vision Language Models (VLMs) as semantic anchors to align the vision embeddings with the textual space. This design ensures plug-and-play implementation without incurring additional pre-training overhead. Extensive experiments on mathematical and logical reasoning benchmarks demonstrate that our method achieves 3-4x token compression and substantial inference acceleration compared to explicit CoT. Furthermore, it maintains competitive performance against other methods, validating the feasibility of this paradigm. Our code is available at https://github.com/TencentBAC/RoT

Paperid: 37, https://arxiv.org/pdf/2601.14732.pdf GitHub

Abstract:
AI models for drug discovery and chemical literature mining must interpret molecular images and generate outputs consistent with 3D geometry and stereochemistry. Most molecular language models rely on strings or graphs, while vision-language models often miss stereochemical details and struggle to map continuous 3D structures into discrete tokens. We propose DeepMoLM: Deep Molecular Language M odeling, a dual-view framework that grounds high-resolution molecular images in geometric invariants derived from molecular conformations. DeepMoLM preserves high-frequency evidence from 1024 $\times$ 1024 inputs, encodes conformer neighborhoods as discrete Extended 3-Dimensional Fingerprints, and fuses visual and geometric streams with cross-attention, enabling physically grounded generation without atom coordinates. DeepMoLM improves PubChem captioning with a 12.3% relative METEOR gain over the strongest generalist baseline while staying competitive with specialist methods. It produces valid numeric outputs for all property queries and attains MAE 13.64 g/mol on Molecular Weight and 37.89 on Complexity in the specialist setting. On ChEBI-20 description generation from images, it exceeds generalist baselines and matches state-of-the-art vision-language models. Code is available at https://github.com/1anj/DeepMoLM.

Paperid: 38, https://arxiv.org/pdf/2601.14602.pdf GitHub

Abstract:
Recent progress in large language models (LLMs) has shown that reasoning improves when intermediate thoughts are externalized into explicit workspaces, such as chain-of-thought traces or tool-augmented reasoning. Yet, visual language models (VLMs) lack an analogous mechanism for spatial reasoning, limiting their ability to generate images that accurately reflect geometric relations, object identities, and compositional intent. We introduce the concept of a spatial scratchpad -- a 3D reasoning substrate that bridges linguistic intent and image synthesis. Given a text prompt, our framework parses subjects and background elements, instantiates them as editable 3D meshes, and employs agentic scene planning for placement, orientation, and viewpoint selection. The resulting 3D arrangement is rendered back into the image domain with identity-preserving cues, enabling the VLM to generate spatially consistent and visually coherent outputs. Unlike prior 2D layout-based methods, our approach supports intuitive 3D edits that propagate reliably into final images. Empirically, it achieves a 32% improvement in text alignment on GenAI-Bench, demonstrating the benefit of explicit 3D reasoning for precise, controllable image generation. Our results highlight a new paradigm for vision-language models that deliberate not only in language, but also in space. Code and visualizations at https://oindrilasaha.github.io/3DScratchpad/

Paperid: 39, https://arxiv.org/pdf/2601.14406.pdf GitHub

Abstract:
Large-scale medical segmentation datasets often combine manual and pseudo-labels of uneven quality, which can compromise training and evaluation. Low-quality labels may hamper performance and make the model training less robust. To address this issue, we propose SegAE (Segmentation Assessment Engine), a lightweight vision-language model (VLM) that automatically predicts label quality across 142 anatomical structures. Trained on over four million image-label pairs with quality scores, SegAE achieves a high correlation coefficient of 0.902 with ground-truth Dice similarity and evaluates a 3D mask in 0.06s. SegAE shows several practical benefits: (I) Our analysis reveals widespread low-quality labeling across public datasets; (II) SegAE improves data efficiency and training performance in active and semi-supervised learning, reducing dataset annotation cost by one-third and quality-checking time by 70% per label. This tool provides a simple and effective solution for quality control in large-scale medical segmentation datasets. The dataset, model weights, and codes are released at https://github.com/Schuture/SegAE.

Paperid: 40, https://arxiv.org/pdf/2601.13096.pdf GitHub

Abstract:
Maritime port inspection plays a critical role in ensuring safety, regulatory compliance, and operational efficiency in complex maritime environments. However, existing inspection methods often rely on manual operations and conventional computer vision techniques that lack scalability and contextual understanding. This study introduces a novel integrated engineering framework that utilizes the synergy between Large Language Models (LLMs) and Vision Language Models (VLMs) to enable autonomous maritime port inspection using cooperative aerial and surface robotic platforms. The proposed framework replaces traditional state-machine mission planners with LLM-driven symbolic planning and improved perception pipelines through VLM-based semantic inspection, enabling context-aware and adaptive monitoring. The LLM module translates natural language mission instructions into executable symbolic plans with dependency graphs that encode operational constraints and ensure safe UAV-USV coordination. Meanwhile, the VLM module performs real-time semantic inspection and compliance assessment, generating structured reports with contextual reasoning. The framework was validated using the extended MBZIRC Maritime Simulator with realistic port infrastructure and further assessed through real-world robotic inspection trials. The lightweight on-board design ensures suitability for resource-constrained maritime platforms, advancing the development of intelligent, autonomous inspection systems. Project resources (code and videos) can be found here: https://github.com/Muhayyuddin/llm-vlm-fusion-port-inspection

Paperid: 41, https://arxiv.org/pdf/2601.12626.pdf GitHub

Abstract:
Spatio-temporal reasoning is a remarkable capability of Vision Language Models (VLMs), but the underlying mechanisms of such abilities remain largely opaque. We postulate that visual/geometrical and textual representations of spatial structure must be combined at some point in VLM computations. We search for such confluence, and ask whether the identified representation can causally explain aspects of input-output model behavior through a linear model. We show empirically that VLMs encode object locations by linearly binding \textit{spatial IDs} to textual activations, then perform reasoning via language tokens. Through rigorous causal interventions we demonstrate that these IDs, which are ubiquitous across the model, can systematically mediate model beliefs at intermediate VLM layers. Additionally, we find that spatial IDs serve as a diagnostic tool for identifying limitations in existing VLMs, and as a valuable learning signal. We extend our analysis to video VLMs and identify an analogous linear temporal ID mechanism. By characterizing our proposed spatiotemporal ID mechanism, we elucidate a previously underexplored internal reasoning process in VLMs, toward improved interpretability and the principled design of more aligned and capable models. We release our code for reproducibility: https://github.com/Raphoo/linear-mech-vlms.

Paperid: 42, https://arxiv.org/pdf/2601.12493.pdf GitHub

Abstract:
Medical Vision-language models (VLMs) have shown remarkable performances in various medical imaging domains such as histo\-pathology by leveraging pre-trained, contrastive models that exploit visual and textual information. However, histopathology images may exhibit severe domain shifts, such as staining, contamination, blurring, and noise, which may severely degrade the VLM's downstream performance. In this work, we introduce Histopath-C, a new benchmark with realistic synthetic corruptions designed to mimic real-world distribution shifts observed in digital histopathology. Our framework dynamically applies corruptions to any available dataset and evaluates Test-Time Adaptation (TTA) mechanisms on the fly. We then propose LATTE, a transductive, low-rank adaptation strategy that exploits multiple text templates, mitigating the sensitivity of histopathology VLMs to diverse text inputs. Our approach outperforms state-of-the-art TTA methods originally designed for natural images across a breadth of histopathology datasets, demonstrating the effectiveness of our proposed design for robust adaptation in histopathology images. Code and data are available at https://github.com/Mehrdad-Noori/Histopath-C.

Paperid: 43, https://arxiv.org/pdf/2601.12082.pdf GitHub

Abstract:
Assisting pathologists in the analysis of histopathological images has high clinical value, as it supports cancer detection and staging. In this context, histology foundation models have recently emerged. Among them, Vision-Language Models (VLMs) provide strong yet imperfect zero-shot predictions. We propose to refine these predictions by adapting Conditional Random Fields (CRFs) to histopathological applications, requiring no additional model training. We present HistoCRF, a CRF-based framework, with a novel definition of the pairwise potential that promotes label diversity and leverages expert annotations. We consider three experiments: without annotations, with expert annotations, and with iterative human-in-the-loop annotations that progressively correct misclassified patches. Experiments on five patch-level classification datasets covering different organs and diseases demonstrate average accuracy gains of 16.0% without annotations and 27.5% with only 100 annotations, compared to zero-shot predictions. Moreover, integrating a human in the loop reaches a further gain of 32.6% with the same number of annotations. The code will be made available on https://github.com/tgodelaine/HistoCRF.

Paperid: 44, https://arxiv.org/pdf/2601.11976.pdf GitHub

Abstract:
Multi-page Document Visual Question Answering (MP-DocVQA) remains challenging because long documents not only strain computational resources but also reduce the effectiveness of the attention mechanism in large vision-language models (LVLMs). We tackle these issues with an Adaptive Visual In-document Retrieval (AVIR) framework. A lightweight retrieval model first scores each page for question relevance. Pages are then clustered according to the score distribution to adaptively select relevant content. The clustered pages are screened again by Top-K to keep the context compact. However, for short documents, clustering reliability decreases, so we use a relevance probability threshold to select pages. The selected pages alone are fed to a frozen LVLM for answer generation, eliminating the need for model fine-tuning. The proposed AVIR framework reduces the average page count required for question answering by 70%, while achieving an ANLS of 84.58% on the MP-DocVQA dataset-surpassing previous methods with significantly lower computational cost. The effectiveness of the proposed AVIR is also verified on the SlideVQA and DUDE benchmarks. The code is available at https://github.com/Li-yachuan/AVIR.

Paperid: 45, https://arxiv.org/pdf/2601.11633.pdf GitHub

Abstract:
Despite the remarkable progress of Vision-Language Models (VLMs) in adopting "Thinking-with-Images" capabilities, accurately evaluating the authenticity of their reasoning process remains a critical challenge. Existing benchmarks mainly rely on outcome-oriented accuracy, lacking the capability to assess whether models can accurately leverage fine-grained visual cues for multi-step reasoning. To address these limitations, we propose ViEBench, a process-verifiable benchmark designed to evaluate faithful visual reasoning. Comprising 200 multi-scenario high-resolution images with expert-annotated visual evidence, ViEBench uniquely categorizes tasks by difficulty into perception and reasoning dimensions, where reasoning tasks require utilizing localized visual details with prior knowledge. To establish comprehensive evaluation criteria, we introduce a dual-axis matrix that provides fine-grained metrics through four diagnostic quadrants, enabling transparent diagnosis of model behavior across varying task complexities. Our experiments yield several interesting observations: (1) VLMs can sometimes produce correct final answers despite grounding on irrelevant regions, and (2) they may successfully locate the correct evidence but still fail to utilize it to reach accurate conclusions. Our findings demonstrate that ViEBench can serve as a more explainable and practical benchmark for comprehensively evaluating the effectiveness agentic VLMs. The codes will be released at: https://github.com/Xuchen-Li/ViEBench.

Paperid: 46, https://arxiv.org/pdf/2601.11514.pdf GitHub

Abstract:
Recent advances in 3D shape generation have achieved impressive results, but most existing methods rely on clean, unoccluded, and well-segmented inputs. Such conditions are rarely met in real-world scenarios. We present ShapeR, a novel approach for conditional 3D object shape generation from casually captured sequences. Given an image sequence, we leverage off-the-shelf visual-inertial SLAM, 3D detection algorithms, and vision-language models to extract, for each object, a set of sparse SLAM points, posed multi-view images, and machine-generated captions. A rectified flow transformer trained to effectively condition on these modalities then generates high-fidelity metric 3D shapes. To ensure robustness to the challenges of casually captured data, we employ a range of techniques including on-the-fly compositional augmentations, a curriculum training scheme spanning object- and scene-level datasets, and strategies to handle background clutter. Additionally, we introduce a new evaluation benchmark comprising 178 in-the-wild objects across 7 real-world scenes with geometry annotations. Experiments show that ShapeR significantly outperforms existing approaches in this challenging setting, achieving an improvement of 2.7x in Chamfer distance compared to state of the art.

Paperid: 47, https://arxiv.org/pdf/2601.11250.pdf GitHub

Abstract:
The rapid emergence of Vision-Language-Action models (VLAs) has a significant impact on robotics. However, their deployment remains complex due to the fragmented interfaces and the inherent communication latency in distributed setups. To address this, we introduce VLAgents, a modular policy server that abstracts VLA inferencing behind a unified Gymnasium-style protocol. Crucially, its communication layer transparently adapts to the context by supporting both zero-copy shared memory for high-speed simulation and compressed streaming for remote hardware. In this work, we present the architecture of VLAgents and validate it by integrating seven policies -- including OpenVLA and Pi Zero. In a benchmark with both local and remote communication, we further demonstrate how it outperforms the default policy servers provided by OpenVLA, OpenPi, and LeRobot. VLAgents is available at https://github.com/RobotControlStack/vlagents

Paperid: 48, https://arxiv.org/pdf/2601.10714.pdf GitHub

Abstract:
We introduce Alterbute, a diffusion-based method for editing an object's intrinsic attributes in an image. We allow changing color, texture, material, and even the shape of an object, while preserving its perceived identity and scene context. Existing approaches either rely on unsupervised priors that often fail to preserve identity or use overly restrictive supervision that prevents meaningful intrinsic variations. Our method relies on: (i) a relaxed training objective that allows the model to change both intrinsic and extrinsic attributes conditioned on an identity reference image, a textual prompt describing the target intrinsic attributes, and a background image and object mask defining the extrinsic context. At inference, we restrict extrinsic changes by reusing the original background and object mask, thereby ensuring that only the desired intrinsic attributes are altered; (ii) Visual Named Entities (VNEs) - fine-grained visual identity categories (e.g., ''Porsche 911 Carrera'') that group objects sharing identity-defining features while allowing variation in intrinsic attributes. We use a vision-language model to automatically extract VNE labels and intrinsic attribute descriptions from a large public image dataset, enabling scalable, identity-preserving supervision. Alterbute outperforms existing methods on identity-preserving object intrinsic attribute editing.

Paperid: 49, https://arxiv.org/pdf/2601.10497.pdf GitHub

Abstract:
Fine-tuning vision-language models (VLMs) such as CLIP often leads to catastrophic forgetting of pretrained knowledge. Prior work primarily aims to mitigate forgetting during adaptation; however, forgetting often remains inevitable during this process. We introduce a novel paradigm, continued fine-tuning (CFT), which seeks to recover pretrained knowledge after a zero-shot model has already been adapted. We propose a simple, model-agnostic CFT strategy (named MERGETUNE) guided by linear mode connectivity (LMC), which can be applied post hoc to existing fine-tuned models without requiring architectural changes. Given a fine-tuned model, we continue fine-tuning its trainable parameters (e.g., soft prompts or linear heads) to search for a continued model which has two low-loss paths to the zero-shot (e.g., CLIP) and the fine-tuned (e.g., CoOp) solutions. By exploiting the geometry of the loss landscape, the continued model implicitly merges the two solutions, restoring pretrained knowledge lost in the fine-tuned counterpart. A challenge is that the vanilla LMC constraint requires data replay from the pretraining task. We approximate this constraint for the zero-shot model via a second-order surrogate, eliminating the need for large-scale data replay. Experiments show that MERGETUNE improves the harmonic mean of CoOp by +5.6% on base-novel generalisation without adding parameters. On robust fine-tuning evaluations, the LMC-merged model from MERGETUNE surpasses ensemble baselines with lower inference cost, achieving further gains and state-of-the-art results when ensembled with the zero-shot model. Our code is available at https://github.com/Surrey-UP-Lab/MERGETUNE.

Paperid: 50, https://arxiv.org/pdf/2601.10477.pdf GitHub

Abstract:
As hubs of human activity, urban surfaces consist of a wealth of semantic entities. Segmenting these various entities from satellite imagery is crucial for a range of downstream applications. Current advanced segmentation models can reliably segment entities defined by physical attributes (e.g., buildings, water bodies) but still struggle with socially defined categories (e.g., schools, parks). In this work, we achieve socio-semantic segmentation by vision-language model reasoning. To facilitate this, we introduce the Urban Socio-Semantic Segmentation dataset named SocioSeg, a new resource comprising satellite imagery, digital maps, and pixel-level labels of social semantic entities organized in a hierarchical structure. Additionally, we propose a novel vision-language reasoning framework called SocioReasoner that simulates the human process of identifying and annotating social semantic entities via cross-modal recognition and multi-stage reasoning. We employ reinforcement learning to optimize this non-differentiable process and elicit the reasoning capabilities of the vision-language model. Experiments demonstrate our approach's gains over state-of-the-art models and strong zero-shot generalization. Our dataset and code are available in https://github.com/AMAP-ML/SocioReasoner.

Paperid: 51, https://arxiv.org/pdf/2601.10094.pdf GitHub

Abstract:
Recent advances in multimodal learning have significantly enhanced the reasoning capabilities of vision-language models (VLMs). However, state-of-the-art approaches rely heavily on large-scale human-annotated datasets, which are costly and time-consuming to acquire. To overcome this limitation, we introduce V-Zero, a general post-training framework that facilitates self-improvement using exclusively unlabeled images. V-Zero establishes a co-evolutionary loop by instantiating two distinct roles: a Questioner and a Solver. The Questioner learns to synthesize high-quality, challenging questions by leveraging a dual-track reasoning reward that contrasts intuitive guesses with reasoned results. The Solver is optimized using pseudo-labels derived from majority voting over its own sampled responses. Both roles are trained iteratively via Group Relative Policy Optimization (GRPO), driving a cycle of mutual enhancement. Remarkably, without a single human annotation, V-Zero achieves consistent performance gains on Qwen2.5-VL-7B-Instruct, improving visual mathematical reasoning by +1.7 and general vision-centric by +2.6, demonstrating the potential of self-improvement in multimodal systems. Code is available at https://github.com/SatonoDia/V-Zero

Paperid: 52, https://arxiv.org/pdf/2601.09708.pdf GitHub

Abstract:
Vision-Language-Action (VLA) tasks require reasoning over complex visual scenes and executing adaptive actions in dynamic environments. While recent studies on reasoning VLAs show that explicit chain-of-thought (CoT) can improve generalization, they suffer from high inference latency due to lengthy reasoning traces. We propose Fast-ThinkAct, an efficient reasoning framework that achieves compact yet performant planning through verbalizable latent reasoning. Fast-ThinkAct learns to reason efficiently with latent CoTs by distilling from a teacher, driven by a preference-guided objective to align manipulation trajectories that transfers both linguistic and visual planning capabilities for embodied control. This enables reasoning-enhanced policy learning that effectively connects compact reasoning to action execution. Extensive experiments across diverse embodied manipulation and reasoning benchmarks demonstrate that Fast-ThinkAct achieves strong performance with up to 89.3\% reduced inference latency over state-of-the-art reasoning VLAs, while maintaining effective long-horizon planning, few-shot adaptation, and failure recovery.

Paperid: 53, https://arxiv.org/pdf/2601.09663.pdf GitHub

Abstract:
Identifying individual animals in long-duration videos is essential for behavioral ecology, wildlife monitoring, and livestock management. Traditional methods require extensive manual annotation, while existing self-supervised approaches are computationally demanding and ill-suited for long sequences due to memory constraints and temporal error propagation. We introduce a highly efficient, self-supervised method that reframes animal identification as a global clustering task rather than a sequential tracking problem. Our approach assumes a known, fixed number of individuals within a single video -- a common scenario in practice -- and requires only bounding box detections and the total count. By sampling pairs of frames, using a frozen pre-trained backbone, and employing a self-bootstrapping mechanism with the Hungarian algorithm for in-batch pseudo-label assignment, our method learns discriminative features without identity labels. We adapt a Binary Cross Entropy loss from vision-language models, enabling state-of-the-art accuracy ($>$97\%) while consuming less than 1 GB of GPU memory per batch -- an order of magnitude less than standard contrastive methods. Evaluated on challenging real-world datasets (3D-POP pigeons and 8-calves feeding videos), our framework matches or surpasses supervised baselines trained on over 1,000 labeled frames, effectively removing the manual annotation bottleneck. This work enables practical, high-accuracy animal identification on consumer-grade hardware, with broad applicability in resource-constrained research settings. All code written for this paper are \href{https://huggingface.co/datasets/tonyFang04/8-calves}{here}.

Paperid: 54, https://arxiv.org/pdf/2601.09575.pdf GitHub

Abstract:
We propose OpenVoxel, a training-free algorithm for grouping and captioning sparse voxels for the open-vocabulary 3D scene understanding tasks. Given the sparse voxel rasterization (SVR) model obtained from multi-view images of a 3D scene, our OpenVoxel is able to produce meaningful groups that describe different objects in the scene. Also, by leveraging powerful Vision Language Models (VLMs) and Multi-modal Large Language Models (MLLMs), our OpenVoxel successfully build an informative scene map by captioning each group, enabling further 3D scene understanding tasks such as open-vocabulary segmentation (OVS) or referring expression segmentation (RES). Unlike previous methods, our method is training-free and does not introduce embeddings from a CLIP/BERT text encoder. Instead, we directly proceed with text-to-text search using MLLMs. Through extensive experiments, our method demonstrates superior performance compared to recent studies, particularly in complex referring expression segmentation (RES) tasks. The code will be open.

Paperid: 55, https://arxiv.org/pdf/2601.09512.pdf GitHub

Abstract:
To teach robots complex manipulation tasks, it is now a common practice to fine-tune a pre-trained vision-language-action model (VLA) on task-specific data. However, since this recipe updates existing representations, it is unsuitable for long-term operation in the real world, where robots must continually adapt to new tasks and environments while retaining the knowledge they have already acquired. Existing continual learning methods for robotics commonly require storing previous data (exemplars), struggle with long task sequences, or rely on task identifiers for deployment. To address these limitations, we propose CLARE, a general, parameter-efficient framework for exemplar-free continual learning with VLAs. CLARE introduces lightweight modular adapters into selected feedforward layers and autonomously expands the model only where necessary when learning a new task, guided by layer-wise feature similarity. During deployment, an autoencoder-based routing mechanism dynamically activates the most relevant adapters without requiring task labels. Through extensive experiments on the LIBERO benchmark, we show that CLARE achieves high performance on new tasks without catastrophic forgetting of earlier tasks, significantly outperforming even exemplar-based methods. Code and data are available at https://tum-lsy.github.io/clare.

Paperid: 56, https://arxiv.org/pdf/2601.09031.pdf GitHub

Abstract:
Humanoid robot manipulation is a crucial research area for executing diverse human-level tasks, involving high-level semantic reasoning and low-level action generation. However, precise scene understanding and sample-efficient learning from human demonstrations remain critical challenges, severely hindering the applicability and generalizability of existing frameworks. This paper presents a novel RGMP-S, Recurrent Geometric-prior Multimodal Policy with Spiking features, facilitating both high-level skill reasoning and data-efficient motion synthesis. To ground high-level reasoning in physical reality, we leverage lightweight 2D geometric inductive biases to enable precise 3D scene understanding within the vision-language model. Specifically, we construct a Long-horizon Geometric Prior Skill Selector that effectively aligns the semantic instructions with spatial constraints, ultimately achieving robust generalization in unseen environments. For the data efficiency issue in robotic action generation, we introduce a Recursive Adaptive Spiking Network. We parameterize robot-object interactions via recursive spiking for spatiotemporal consistency, fully distilling long-horizon dynamic features while mitigating the overfitting issue in sparse demonstration scenarios. Extensive experiments are conducted across the Maniskill simulation benchmark and three heterogeneous real-world robotic systems, encompassing a custom-developed humanoid, a desktop manipulator, and a commercial robotic platform. Empirical results substantiate the superiority of our method over state-of-the-art baselines and validate the efficacy of the proposed modules in diverse generalization scenarios. To facilitate reproducibility, the source code and video demonstrations are publicly available at https://github.com/xtli12/RGMP-S.git.

Paperid: 57, https://arxiv.org/pdf/2601.08557.pdf GitHub

Abstract:
Hallucinations in video-capable vision-language models (Video-VLMs) remain frequent and high-confidence, while existing uncertainty metrics often fail to align with correctness. We introduce VideoHEDGE, a modular framework for hallucination detection in video question answering that extends entropy-based reliability estimation from images to temporally structured inputs. Given a video-question pair, VideoHEDGE draws a baseline answer and multiple high-temperature generations from both clean clips and photometrically and spatiotemporally perturbed variants, then clusters the resulting textual outputs into semantic hypotheses using either Natural Language Inference (NLI)-based or embedding-based methods. Cluster-level probability masses yield three reliability scores: Semantic Entropy (SE), RadFlag, and Vision-Amplified Semantic Entropy (VASE). We evaluate VideoHEDGE on the SoccerChat benchmark using an LLM-as-a-judge to obtain binary hallucination labels. Across three 7B Video-VLMs (Qwen2-VL, Qwen2.5-VL, and a SoccerChat-finetuned model), VASE consistently achieves the highest ROC-AUC, especially at larger distortion budgets, while SE and RadFlag often operate near chance. We further show that embedding-based clustering matches NLI-based clustering in detection performance at substantially lower computational cost, and that domain fine-tuning reduces hallucination frequency but yields only modest improvements in calibration. The hedge-bench PyPI library enables reproducible and extensible benchmarking, with full code and experimental resources available at https://github.com/Simula/HEDGE#videohedge .

Paperid: 58, https://arxiv.org/pdf/2601.08420.pdf GitHub

Abstract:
In this paper, we propose a novel multimodal framework, Multimodal Language-Guided Network (MMLGNet), to align heterogeneous remote sensing modalities like Hyperspectral Imaging (HSI) and LiDAR with natural language semantics using vision-language models such as CLIP. With the increasing availability of multimodal Earth observation data, there is a growing need for methods that effectively fuse spectral, spatial, and geometric information while enabling semantic-level understanding. MMLGNet employs modality-specific encoders and aligns visual features with handcrafted textual embeddings in a shared latent space via bi-directional contrastive learning. Inspired by CLIP's training paradigm, our approach bridges the gap between high-dimensional remote sensing data and language-guided interpretation. Notably, MMLGNet achieves strong performance with simple CNN-based encoders, outperforming several established multimodal visual-only methods on two benchmark datasets, demonstrating the significant benefit of language supervision. Codes are available at https://github.com/AdityaChaudhary2913/CLIP_HSI.

Paperid: 59, https://arxiv.org/pdf/2601.08192.pdf GitHub

Abstract:
Medical image analysis increasingly relies on large vision-language models (VLMs), yet most systems remain single-pass black boxes that offer limited control over reasoning, safety, and spatial grounding. We propose R^4, an agentic framework that decomposes medical imaging workflows into four coordinated agents: a Router that configures task- and specialization-aware prompts from the image, patient history, and metadata; a Retriever that uses exemplar memory and pass@k sampling to jointly generate free-text reports and bounding boxes; a Reflector that critiques each draft-box pair for key clinical error modes (negation, laterality, unsupported claims, contradictions, missing findings, and localization errors); and a Repairer that iteratively revises both narrative and spatial outputs under targeted constraints while curating high-quality exemplars for future cases. Instantiated on chest X-ray analysis with multiple modern VLM backbones and evaluated on report generation and weakly supervised detection, R^4 consistently boosts LLM-as-a-Judge scores by roughly +1.7-+2.5 points and mAP50 by +2.5-+3.5 absolute points over strong single-VLM baselines, without any gradient-based fine-tuning. These results show that agentic routing, reflection, and repair can turn strong but brittle VLMs into more reliable and better grounded tools for clinical image interpretation. Our code can be found at: https://github.com/faiyazabdullah/MultimodalMedAgent

Paperid: 60, https://arxiv.org/pdf/2601.07812.pdf GitHub

Abstract:
Large Vision Language Models (LVLMs) have demonstrated remarkable capabilities, yet their proficiency in understanding and reasoning over multiple images remains largely unexplored. While existing benchmarks have initiated the evaluation of multi-image models, a comprehensive analysis of their core weaknesses and their causes is still lacking. In this work, we introduce MIMIC (Multi-Image Model Insights and Challenges), a new benchmark designed to rigorously evaluate the multi-image capabilities of LVLMs. Using MIMIC, we conduct a series of diagnostic experiments that reveal pervasive issues: LVLMs often fail to aggregate information across images and struggle to track or attend to multiple concepts simultaneously. To address these failures, we propose two novel complementary remedies. On the data side, we present a procedural data-generation strategy that composes single-image annotations into rich, targeted multi-image training examples. On the optimization side, we analyze layer-wise attention patterns and derive an attention-masking scheme tailored for multi-image inputs. Experiments substantially improved cross-image aggregation, while also enhancing performance on existing multi-image benchmarks, outperforming prior state of the art across tasks. Data and code will be made available at https://github.com/anurag-198/MIMIC.

Paperid: 61, https://arxiv.org/pdf/2601.06944.pdf GitHub

Abstract:
While Multimodal Large Language Models (MLLMs) have achieved remarkable progress in visual understanding, they often struggle when faced with the unstructured and ambiguous nature of human-generated sketches. This limitation is particularly pronounced in the underexplored task of visual grading, where models should not only solve a problem but also diagnose errors in hand-drawn diagrams. Such diagnostic capabilities depend on complex structural, semantic, and metacognitive reasoning. To bridge this gap, we introduce SketchJudge, a novel benchmark tailored for evaluating MLLMs as graders of hand-drawn STEM diagrams. SketchJudge encompasses 1,015 hand-drawn student responses across four domains: geometry, physics, charts, and flowcharts, featuring diverse stylistic variations and distinct error types. Evaluations on SketchJudge demonstrate that even advanced MLLMs lag significantly behind humans, validating the benchmark's effectiveness in exposing the fragility of current vision-language alignment in symbolic and noisy contexts. All data, code, and evaluation scripts are publicly available at https://github.com/yuhangsu82/SketchJudge.

Paperid: 62, https://arxiv.org/pdf/2601.06498.pdf GitHub GitHub

Abstract:
Due to the limited generalization and interpretability of deep learning classifiers, The final vetting of rare celestial object candidates still relies on expert visual inspection--a manually intensive process. In this process, astronomers leverage specialized tools to analyze spectra and construct reliable catalogs. However, this practice has become the primary bottleneck, as it is fundamentally incapable of scaling with the data deluge from modern spectroscopic surveys. To bridge this gap, we propose Spec-o3, a tool-augmented vision-language agent that performs astronomer-aligned spectral inspection via interleaved multimodal chain-of-thought reasoning. Spec-o3 is trained with a two-stage post-training recipe: cold-start supervised fine-tuning on expert inspection trajectories followed by outcome-based reinforcement learning on rare-type verification tasks. Evaluated on five rare-object identification tasks from LAMOST, Spec-o3 establishes a new State-of-the-Art, boosting the macro-F1 score from 28.3 to 76.5 with a 7B parameter base model and outperforming both proprietary VLMs and specialized deep models. Crucially, the agent demonstrates strong generalization to unseen inspection tasks across survey shifts (from LAMOST to SDSS/DESI). Expert evaluations confirm that its reasoning traces are coherent and physically consistent, supporting transparent and trustworthy decision-making. Code, data, and models are available at \href{https://github.com/Maxwell-Jia/spec-o3}{Project HomePage}.

Paperid: 63, https://arxiv.org/pdf/2601.06496.pdf GitHub

Abstract:
Spatial intelligence refers to the ability to perceive, reason about, and describe objects and their relationships within three-dimensional environments, forming a foundation for embodied perception and scene understanding. 3D captioning aims to describe 3D scenes in natural language; however, it remains challenging due to the sparsity and irregularity of point clouds and, more critically, the weak grounding and limited out-of-distribution (OOD) generalization of existing captioners across drastically different environments, including indoor and outdoor 3D scenes. To address this challenge, we propose 3D CoCa v2, a generalizable 3D captioning framework that unifies contrastive vision-language learning with 3D caption generation and further improves robustness via test-time search (TTS) without updating the captioner parameters. 3D CoCa v2 builds on a frozen CLIP-based semantic prior, a spatially-aware 3D scene encoder for geometry, and a multimodal decoder jointly optimized with contrastive and captioning objectives, avoiding external detectors or handcrafted proposals. At inference, TTS produces diverse caption candidates and performs reward-guided selection using a compact scene summary. Experiments show improvements over 3D CoCa of +1.50 CIDEr@0.5IoU on ScanRefer and +1.61 CIDEr@0.5IoU on Nr3D, and +3.8 CIDEr@0.25 in zero-shot OOD evaluation on TOD3Cap. Code will be released at https://github.com/AIGeeksGroup/3DCoCav2.

Paperid: 64, https://arxiv.org/pdf/2601.06460.pdf GitHub

Abstract:
Vision-Language Models (VLMs) are increasingly used in safety-critical applications that require reliable visual grounding. However, these models often hallucinate details that are not present in the image to satisfy user prompts. While recent datasets and benchmarks have been introduced to evaluate systematic hallucinations in VLMs, many hallucination behaviors remain insufficiently characterized. In particular, prior work primarily focuses on object presence or absence, leaving it unclear how prompt phrasing and structural constraints can systematically induce hallucinations. In this paper, we investigate how different forms of prompt pressure influence hallucination behavior. We introduce Ghost-100, a procedurally generated dataset of synthetic scenes in which key visual details are deliberately removed, enabling controlled analysis of absence-based hallucinations. Using a structured 5-Level Prompt Intensity Framework, we vary prompts from neutral queries to toxic demands and rigid formatting constraints. We evaluate three representative open-weight VLMs: MiniCPM-V 2.6-8B, Qwen2-VL-7B, and Qwen3-VL-8B. Across all three models, hallucination rates do not increase monotonically with prompt intensity. All models exhibit reductions at higher intensity levels at different thresholds, though not all show sustained reduction under maximum coercion. These results suggest that current safety alignment is more effective at detecting semantic hostility than structural coercion, revealing model-specific limitations in handling compliance pressure. Our dataset is available at: https://github.com/bli1/tone-matters

Paperid: 65, https://arxiv.org/pdf/2601.06198.pdf GitHub

Abstract:
Biryani, one of India's most celebrated dishes, exhibits remarkable regional diversity in its preparation, ingredients, and presentation. With the growing availability of online cooking videos, there is unprecedented potential to study such culinary variations using computational tools systematically. However, existing video understanding methods fail to capture the fine-grained, multimodal, and culturally grounded differences in procedural cooking videos. This work presents the first large-scale, curated dataset of biryani preparation videos, comprising 120 high-quality YouTube recordings across 12 distinct regional styles. We propose a multi-stage framework leveraging recent advances in vision-language models (VLMs) to segment videos into fine-grained procedural units and align them with audio transcripts and canonical recipe text. Building on these aligned representations, we introduce a video comparison pipeline that automatically identifies and explains procedural differences between regional variants. We construct a comprehensive question-answer (QA) benchmark spanning multiple reasoning levels to evaluate procedural understanding in VLMs. Our approach employs multiple VLMs in complementary roles, incorporates human-in-the-loop verification for high-precision tasks, and benchmarks several state-of-the-art models under zero-shot and fine-tuned settings. The resulting dataset, comparison methodology, and QA benchmark provide a new testbed for evaluating VLMs on structured, multimodal reasoning tasks and open new directions for computational analysis of cultural heritage through cooking videos. We release all data, code, and the project website at https://farzanashaju.github.io/how-does-india-cook-biryani/.

Paperid: 66, https://arxiv.org/pdf/2601.05991.pdf GitHub

Abstract:
In safety-critical domains, linguistic ambiguity can have severe consequences; a vague command like "Pass me the vial" in a surgical setting could lead to catastrophic errors. Yet, most embodied AI research overlooks this, assuming instructions are clear and focusing on execution rather than confirmation. To address this critical safety gap, we are the first to define Open-Vocabulary 3D Instruction Ambiguity Detection, a fundamental new task where a model must determine if a command has a single, unambiguous meaning within a given 3D scene. To support this research, we build Ambi3D, the large-scale benchmark for this task, featuring over 700 diverse 3D scenes and around 22k instructions. Our analysis reveals a surprising limitation: state-of-the-art 3D Large Language Models (LLMs) struggle to reliably determine if an instruction is ambiguous. To address this challenge, we propose AmbiVer, a two-stage framework that collects explicit visual evidence from multiple views and uses it to guide an vision-language model (VLM) in judging instruction ambiguity. Extensive experiments demonstrate the challenge of our task and the effectiveness of AmbiVer, paving the way for safer and more trustworthy embodied AI. Code and dataset available at https://jiayuding031020.github.io/ambi3d/.

Paperid: 67, https://arxiv.org/pdf/2601.05787.pdf GitHub

Abstract:
Vision-language models are increasingly deployed as computer-use agents (CUAs) that operate desktops and browsers. Top-performing CUAs are framework-based systems that decompose planning and execution, while end-to-end screenshot-to-action policies are easier to deploy but lag behind on benchmarks such as OSWorld-Verified. GUI datasets like OSWorld pose two bottlenecks: they expose only a few hundred interactive, verifiable tasks and environments, and expert trajectories must be gathered by interacting with these environments, making such data hard to scale. We therefore ask how reinforcement learning from verifiable rewards (RLVR) can best exploit a small pool of exist expert trajectories to train end-to-end policies. Naively mixing these off-policy traces into on-policy RLVR is brittle: even after format conversion, expert trajectories exhibit structural mismatch and distribution shift from the learner. We propose BEPA (Bi-Level Expert-to-Policy Assimilation), which turns static expert traces into policy-aligned guidance via self-rolled reachable trajectories under the base policy (LEVEL-1) and a per-task, dynamically updated cache used in RLVR (LEVEL-2). On OSWorld-Verified, BEPA improves UITARS1.5-7B success from 22.87% to 32.13% and raises a held-out split from 5.74% to 10.30%, with consistent gains on MMBench-GUI and Online-Mind2Web. Our code and data are available at: https://github.com/LEON-gittech/Verl_GUI.git

Paperid: 68, https://arxiv.org/pdf/2601.05172.pdf GitHub

Abstract:
Embodied question answering (EQA) in 3D environments often requires collecting context that is distributed across multiple viewpoints and partially occluded. However, most recent vision--language models (VLMs) are constrained to a fixed and finite set of input views, which limits their ability to acquire question-relevant context at inference time and hinders complex spatial reasoning. We propose Chain-of-View (CoV) prompting, a training-free, test-time reasoning framework that transforms a VLM into an active viewpoint reasoner through a coarse-to-fine exploration process. CoV first employs a View Selection agent to filter redundant frames and identify question-aligned anchor views. It then performs fine-grained view adjustment by interleaving iterative reasoning with discrete camera actions, obtaining new observations from the underlying 3D scene representation until sufficient context is gathered or a step budget is reached. We evaluate CoV on OpenEQA across four mainstream VLMs and obtain an average +11.56% improvement in LLM-Match, with a maximum gain of +13.62% on Qwen3-VL-Flash. CoV further exhibits test-time scaling: increasing the minimum action budget yields an additional +2.51% average improvement, peaking at +3.73% on Gemini-2.5-Flash. On ScanQA and SQA3D, CoV delivers strong performance (e.g., 116 CIDEr / 31.9 EM@1 on ScanQA and 51.1 EM@1 on SQA3D). Overall, these results suggest that question-aligned view selection coupled with open-view search is an effective, model-agnostic strategy for improving spatial reasoning in 3D EQA without additional training. Code is available on https://github.com/ziplab/CoV .

Paperid: 69, https://arxiv.org/pdf/2601.04497.pdf GitHub

Abstract:
Modern forest monitoring workflows increasingly benefit from the growing availability of high-resolution satellite imagery and advances in deep learning. Two persistent challenges in this context are accurate pixel-level change detection and meaningful semantic change captioning for complex forest dynamics. While large language models (LLMs) are being adapted for interactive data exploration, their integration with vision-language models (VLMs) for remote sensing image change interpretation (RSICI) remains underexplored. To address this gap, we introduce an LLM-driven agent for integrated forest change analysis that supports natural language querying across multiple RSICI tasks. The proposed system builds upon a multi-level change interpretation (MCI) vision-language backbone with LLM-based orchestration. To facilitate adaptation and evaluation in forest environments, we further introduce the Forest-Change dataset, which comprises bi-temporal satellite imagery, pixel-level change masks, and multi-granularity semantic change captions generated using a combination of human annotation and rule-based methods. Experimental results show that the proposed system achieves mIoU and BLEU-4 scores of 67.10% and 40.17% on the Forest-Change dataset, and 88.13% and 34.41% on LEVIR-MCI-Trees, a tree-focused subset of LEVIR-MCI benchmark for joint change detection and captioning. These results highlight the potential of interactive, LLM-driven RSICI systems to improve accessibility, interpretability, and efficiency of forest change analysis. All data and code are publicly available at https://github.com/JamesBrockUoB/ForestChat.

Paperid: 70, https://arxiv.org/pdf/2601.04453.pdf GitHub

Abstract:
World models have become central to autonomous driving, where accurate scene understanding and future prediction are crucial for safe control. Recent work has explored using vision-language models (VLMs) for planning, yet existing approaches typically treat perception, prediction, and planning as separate modules. We propose UniDrive-WM, a unified VLM-based world model that jointly performs driving-scene understanding, trajectory planning, and trajectory-conditioned future image generation within a single architecture. UniDrive-WM's trajectory planner predicts a future trajectory, which conditions a VLM-based image generator to produce plausible future frames. These predictions provide additional supervisory signals that enhance scene understanding and iteratively refine trajectory generation. We further compare discrete and continuous output representations for future image prediction, analyzing their influence on downstream driving performance. Experiments on the challenging Bench2Drive benchmark show that UniDrive-WM produces high-fidelity future images and improves planning performance by 5.9% in L2 trajectory error and 9.2% in collision rate over the previous best method. These results demonstrate the advantages of tightly integrating VLM-driven reasoning, planning, and generative world modeling for autonomous driving. The project page is available at https://unidrive-wm.github.io/UniDrive-WM .

Paperid: 71, https://arxiv.org/pdf/2601.04203.pdf GitHub

Abstract:
We present FronTalk, a benchmark for front-end code generation that pioneers the study of a unique interaction dynamic: conversational code generation with multi-modal feedback. In front-end development, visual artifacts such as sketches, mockups and annotated creenshots are essential for conveying design intent, yet their role in multi-turn code generation remains largely unexplored. To address this gap, we focus on the front-end development task and curate FronTalk, a collection of 100 multi-turn dialogues derived from real-world websites across diverse domains such as news, finance, and art. Each turn features both a textual instruction and an equivalent visual instruction, each representing the same user intent. To comprehensively evaluate model performance, we propose a novel agent-based evaluation framework leveraging a web agent to simulate users and explore the website, and thus measuring both functional correctness and user experience. Evaluation of 20 models reveals two key challenges that are under-explored systematically in the literature: (1) a significant forgetting issue where models overwrite previously implemented features, resulting in task failures, and (2) a persistent challenge in interpreting visual feedback, especially for open-source vision-language models (VLMs). We propose a strong baseline to tackle the forgetting issue with AceCoder, a method that critiques the implementation of every past instruction using an autonomous web agent. This approach significantly reduces forgetting to nearly zero and improves the performance by up to 9.3% (56.0% to 65.3%). Overall, we aim to provide a solid foundation for future research in front-end development and the general interaction dynamics of multi-turn, multi-modal code generation. Code and data are released at https://github.com/shirley-wu/frontalk

Paperid: 72, https://arxiv.org/pdf/2601.03590.pdf GitHub

Abstract:
Recent advancements in Spatial Intelligence (SI) have predominantly relied on Vision-Language Models (VLMs), yet a critical question remains: does spatial understanding originate from visual encoders or the fundamental reasoning backbone? Inspired by this question, we introduce SiT-Bench, a novel benchmark designed to evaluate the SI performance of Large Language Models (LLMs) without pixel-level input, comprises over 3,800 expert-annotated items across five primary categories and 17 subtasks, ranging from egocentric navigation and perspective transformation to fine-grained robotic manipulation. By converting single/multi-view scenes into high-fidelity, coordinate-aware textual descriptions, we challenge LLMs to perform symbolic textual reasoning rather than visual pattern matching. Evaluation results of state-of-the-art (SOTA) LLMs reveals that while models achieve proficiency in localized semantic tasks, a significant "spatial gap" remains in global consistency. Notably, we find that explicit spatial reasoning significantly boosts performance, suggesting that LLMs possess latent world-modeling potential. Our proposed dataset SiT-Bench serves as a foundational resource to foster the development of spatially-grounded LLM backbones for future VLMs and embodied agents. Our code and benchmark will be released at https://github.com/binisalegend/SiT-Bench .

Paperid: 73, https://arxiv.org/pdf/2601.02918.pdf GitHub

Abstract:
Image Quality Assessment (IQA) is a long-standing problem in computer vision. Previous methods typically focus on predicting numerical scores without explanation or providing low-level descriptions lacking precise scores. Recent reasoning-based vision language models (VLMs) have shown strong potential for IQA by jointly generating quality descriptions and scores. However, existing VLM-based IQA methods often suffer from unreliable reasoning due to their limited capability of integrating visual and textual cues. In this work, we introduce Zoom-IQA, a VLM-based IQA model to explicitly emulate key cognitive behaviors: uncertainty awareness, region reasoning, and iterative refinement. Specifically, we present a two-stage training pipeline: 1) supervised fine-tuning (SFT) on our Grounded-Rationale-IQA (GR-IQA) dataset to teach the model to ground its assessments in key regions, and 2) reinforcement learning (RL) for dynamic policy exploration, stabilized by our KL-Coverage regularizer to prevent reasoning and scoring diversity collapse, with a Progressive Re-sampling Strategy for mitigating annotation bias. Extensive experiments show that Zoom-IQA achieves improved robustness, explainability, and generalization. The application to downstream tasks, such as image restoration, further demonstrates the effectiveness of Zoom-IQA.

Paperid: 74, https://arxiv.org/pdf/2601.02854.pdf GitHub

Abstract:
As an agent-level reasoning and coordination paradigm, Multi-Agent Debate (MAD) orchestrates multiple agents through structured debate to improve answer quality and support complex reasoning. However, existing research on MAD suffers from two fundamental limitations: evaluations are conducted under fragmented and inconsistent settings, hindering fair comparison, and are largely restricted to single-modality scenarios that rely on textual inputs only. To address these gaps, we introduce M3MAD-Bench, a unified and extensible benchmark for evaluating MAD methods across Multi-domain tasks, Multi-modal inputs, and Multi-dimensional metrics. M3MAD-Bench establishes standardized protocols over five core task domains: Knowledge, Mathematics, Medicine, Natural Sciences, and Complex Reasoning, and systematically covers both pure text and vision-language datasets, enabling controlled cross-modality comparison. We evaluate MAD methods on nine base models spanning different architectures, scales, and modality capabilities. Beyond accuracy, M3MAD-Bench incorporates efficiency-oriented metrics such as token consumption and inference time, providing a holistic view of performance--cost trade-offs. Extensive experiments yield systematic insights into the effectiveness, robustness, and efficiency of MAD across text-only and multimodal scenarios. We believe M3MAD-Bench offers a reliable foundation for future research on standardized MAD evaluation. The code is available at http://github.com/liaolea/M3MAD-Bench.

Paperid: 75, https://arxiv.org/pdf/2601.02456.pdf GitHub

Abstract:
Prevalent Vision-Language-Action (VLA) models are typically built upon Multimodal Large Language Models (MLLMs) and demonstrate exceptional proficiency in semantic understanding, but they inherently lack the capability to deduce physical world dynamics. Consequently, recent approaches have shifted toward World Models, typically formulated via video prediction; however, these methods often suffer from a lack of semantic grounding and exhibit brittleness when handling prediction errors. To synergize semantic understanding with dynamic predictive capabilities, we present InternVLA-A1. This model employs a unified Mixture-of-Transformers architecture, coordinating three experts for scene understanding, visual foresight generation, and action execution. These components interact seamlessly through a unified masked self-attention mechanism. Building upon InternVL3 and Qwen3-VL, we instantiate InternVLA-A1 at 2B and 3B parameter scales. We pre-train these models on hybrid synthetic-real datasets spanning InternData-A1 and Agibot-World, covering over 533M frames. This hybrid training strategy effectively harnesses the diversity of synthetic simulation data while minimizing the sim-to-real gap. We evaluated InternVLA-A1 across 12 real-world robotic tasks and simulation benchmark. It significantly outperforms leading models like pi0 and GR00T N1.5, achieving a 14.5\% improvement in daily tasks and a 40\%-73.3\% boost in dynamic settings, such as conveyor belt sorting.

Paperid: 76, https://arxiv.org/pdf/2601.02358.pdf GitHub

Abstract:
We present VINO, a unified visual generator that performs image and video generation and editing within a single framework. Instead of relying on task-specific models or independent modules for each modality, VINO uses a shared diffusion backbone that conditions on text, images and videos, enabling a broad range of visual creation and editing tasks under one model. Specifically, VINO couples a vision-language model (VLM) with a Multimodal Diffusion Transformer (MMDiT), where multimodal inputs are encoded as interleaved conditioning tokens, and then used to guide the diffusion process. This design supports multi-reference grounding, long-form instruction following, and coherent identity preservation across static and dynamic content, while avoiding modality-specific architectural components. To train such a unified system, we introduce a multi-stage training pipeline that progressively expands a video generation base model into a unified, multi-task generator capable of both image and video input and output. Across diverse generation and editing benchmarks, VINO demonstrates strong visual quality, faithful instruction following, improved reference and attribute preservation, and more controllable multi-identity edits. Our results highlight a practical path toward scalable unified visual generation, and the promise of interleaved, in-context computation as a foundation for general-purpose visual creation.

Paperid: 77, https://arxiv.org/pdf/2601.02295.pdf GitHub

Abstract:
Current work on robot failure detection and correction typically operate in a post hoc manner, analyzing errors and applying corrections only after failures occur. This work introduces CycleVLA, a system that equips Vision-Language-Action models (VLAs) with proactive self-correction, the capability to anticipate incipient failures and recover before they fully manifest during execution. CycleVLA achieves this by integrating a progress-aware VLA that flags critical subtask transition points where failures most frequently occur, a VLM-based failure predictor and planner that triggers subtask backtracking upon predicted failure, and a test-time scaling strategy based on Minimum Bayes Risk (MBR) decoding to improve retry success after backtracking. Extensive experiments show that CycleVLA improves performance for both well-trained and under-trained VLAs, and that MBR serves as an effective zero-shot test-time scaling strategy for VLAs. Project Page: https://dannymcy.github.io/cyclevla/

Paperid: 78, https://arxiv.org/pdf/2601.02078.pdf GitHub

Abstract:
The development of robust and generalizable robot learning models is critically contingent upon the availability of large-scale, diverse training data and reliable evaluation benchmarks. Collecting data in the physical world poses prohibitive costs and scalability challenges, and prevailing simulation benchmarks frequently suffer from fragmentation, narrow scope, or insufficient fidelity to enable effective sim-to-real transfer. To address these challenges, we introduce Genie Sim 3.0, a unified simulation platform for robotic manipulation. We present Genie Sim Generator, a large language model (LLM)-powered tool that constructs high-fidelity scenes from natural language instructions. Its principal strength resides in rapid and multi-dimensional generalization, facilitating the synthesis of diverse environments to support scalable data collection and robust policy evaluation. We introduce the first benchmark that pioneers the application of LLM for automated evaluation. It leverages LLM to mass-generate evaluation scenarios and employs Vision-Language Model (VLM) to establish an automated assessment pipeline. We also release an open-source dataset comprising more than 10,000 hours of synthetic data across over 200 tasks. Through systematic experimentation, we validate the robust zero-shot sim-to-real transfer capability of our open-source dataset, demonstrating that synthetic data can server as an effective substitute for real-world data under controlled conditions for scalable policy training. For code and dataset details, please refer to: https://github.com/AgibotTech/genie_sim.

Paperid: 79, https://arxiv.org/pdf/2601.01870.pdf GitHub

Abstract:
Existing text-driven infrared and visible image fusion approaches often rely on textual information at the sentence level, which can lead to semantic noise from redundant text and fail to fully exploit the deeper semantic value of textual information. To address these issues, we propose a novel fusion approach named Entity-Guided Multi-Task learning for infrared and visible image fusion (EGMT). Our approach includes three key innovative components: (i) A principled method is proposed to extract entity-level textual information from image captions generated by large vision-language models, eliminating semantic noise from raw text while preserving critical semantic information; (ii) A parallel multi-task learning architecture is constructed, which integrates image fusion with a multi-label classification task. By using entities as pseudo-labels, the multi-label classification task provides semantic supervision, enabling the model to achieve a deeper understanding of image content and significantly improving the quality and semantic density of the fused image; (iii) An entity-guided cross-modal interactive module is also developed to facilitate the fine-grained interaction between visual and entity-level textual features, which enhances feature representation by capturing cross-modal dependencies at both inter-visual and visual-entity levels. To promote the wide application of the entity-guided image fusion framework, we release the entity-annotated version of four public datasets (i.e., TNO, RoadScene, M3FD, and MSRS). Extensive experiments demonstrate that EGMT achieves superior performance in preserving salient targets, texture details, and semantic consistency, compared to the state-of-the-art methods. The code and dataset will be publicly available at https://github.com/wyshao-01/EGMT.

Paperid: 80, https://arxiv.org/pdf/2601.01769.pdf GitHub

Abstract:
In this paper, we introduce a clinical diagnosis template-based pipeline to systematically collect and structure pathological information. In collaboration with pathologists and guided by the the College of American Pathologists (CAP) Cancer Protocols, we design a Clinical Pathology Report Template (CPRT) that ensures comprehensive and standardized extraction of diagnostic elements from pathology reports. We validate the effectiveness of our pipeline on TCGA-BRCA. First, we extract pathological features from reports using CPRT. These features are then used to build CTIS-Align, a dataset of 80k slide-description pairs from 804 WSIs for vision-language alignment training, and CTIS-Bench, a rigorously curated VQA benchmark comprising 977 WSIs and 14,879 question-answer pairs. CTIS-Bench emphasizes clinically grounded, closed-ended questions (e.g., tumor grade, receptor status) that reflect real diagnostic workflows, minimize non-visual reasoning, and require genuine slide understanding. We further propose CTIS-QA, a Slide-level Question Answering model, featuring a dual-stream architecture that mimics pathologists' diagnostic approach. One stream captures global slide-level context via clustering-based feature aggregation, while the other focuses on salient local regions through attention-guided patch perception module. Extensive experiments on WSI-VQA, CTIS-Bench, and slide-level diagnostic tasks show that CTIS-QA consistently outperforms existing state-of-the-art models across multiple metrics. Code and data are available at https://github.com/HLSvois/CTIS-QA.

Paperid: 81, https://arxiv.org/pdf/2601.01416.pdf GitHub

Abstract:
Despite notable advancements in remote sensing vision-language models (VLMs), existing models often struggle with spatial understanding, limiting their effectiveness in real-world applications. To push the boundaries of VLMs in remote sensing, we specifically address vehicle imagery captured by drones and introduce a spatially-aware dataset AirSpatial, which comprises over 206K instructions and introduces two novel tasks: Spatial Grounding and Spatial Question Answering. It is also the first remote sensing grounding dataset to provide 3DBB. To effectively leverage existing image understanding of VLMs to spatial domains, we adopt a two-stage training strategy comprising Image Understanding Pre-training and Spatial Understanding Fine-tuning. Utilizing this trained spatially-aware VLM, we develop an aerial agent, AirSpatialBot, which is capable of fine-grained vehicle attribute recognition and retrieval. By dynamically integrating task planning, image understanding, spatial understanding, and task execution capabilities, AirSpatialBot adapts to diverse query requirements. Experimental results validate the effectiveness of our approach, revealing the spatial limitations of existing VLMs while providing valuable insights. The model, code, and datasets will be released at https://github.com/VisionXLab/AirSpatialBot

Paperid: 82, https://arxiv.org/pdf/2601.01024.pdf GitHub

Abstract:
Vision Language Models (VLMs) have rapidly advanced and show strong promise for text-based person search (TBPS), a task that requires capturing fine-grained relationships between images and text to distinguish individuals. Previous methods address these challenges through local alignment, yet they are often prone to shortcut learning and spurious correlations, yielding misalignment. Moreover, injecting prior knowledge can distort intra-modality structure. Motivated by our finding that encoder attention surfaces spatially precise evidence from the earliest training epochs, and to alleviate these issues, we introduceITSELF, an attention-guided framework for implicit local alignment. At its core, Guided Representation with Attentive Bank (GRAB) converts the model's own attention into an Attentive Bank of high-saliency tokens and applies local objectives on this bank, learning fine-grained correspondences without extra supervision. To make the selection reliable and non-redundant, we introduce Multi-Layer Attention for Robust Selection (MARS), which aggregates attention across layers and performs diversity-aware top-k selection; and Adaptive Token Scheduler (ATS), which schedules the retention budget from coarse to fine over training, preserving context early while progressively focusing on discriminative details. Extensive experiments on three widely used TBPS benchmarks showstate-of-the-art performance and strong cross-dataset generalization, confirming the effectiveness and robustness of our approach without additional prior supervision. Our project is publicly available at https://trhuuloc.github.io/itself

Paperid: 83, https://arxiv.org/pdf/2601.00998.pdf GitHub

Abstract:
Remote sensing (RS) large vision-language models (LVLMs) have shown strong promise across visual grounding (VG) tasks. However, existing RS VG datasets predominantly rely on explicit referring expressions-such as relative position, relative size, and color cues-thereby constraining performance on implicit VG tasks that require scenario-specific domain knowledge. This article introduces DVGBench, a high-quality implicit VG benchmark for drones, covering six major application scenarios: traffic, disaster, security, sport, social activity, and productive activity. Each object provides both explicit and implicit queries. Based on the dataset, we design DroneVG-R1, an LVLM that integrates the novel Implicit-to-Explicit Chain-of-Thought (I2E-CoT) within a reinforcement learning paradigm. This enables the model to take advantage of scene-specific expertise, converting implicit references into explicit ones and thus reducing grounding difficulty. Finally, an evaluation of mainstream models on both explicit and implicit VG tasks reveals substantial limitations in their reasoning capabilities. These findings provide actionable insights for advancing the reasoning capacity of LVLMs for drone-based agents. The code and datasets will be released at https://github.com/zytx121/DVGBench

Paperid: 84, https://arxiv.org/pdf/2601.00993.pdf GitHub

Abstract:
Wildlife monitoring is crucial for studying biodiversity loss and climate change. Camera trap images provide a non-intrusive method for analyzing animal populations and identifying ecological patterns over time. However, manual analysis is time-consuming and resource-intensive. Deep learning, particularly foundation models, has been applied to automate wildlife identification, achieving strong performance when tested on data from the same geographical locations as their training sets. Yet, despite their promise, these models struggle to generalize to new geographical areas, leading to significant performance drops. For example, training an advanced vision-language model, such as CLIP with an adapter, on an African dataset achieves an accuracy of 84.77%. However, this performance drops significantly to 16.17% when the model is tested on an American dataset. This limitation partly arises because existing models rely predominantly on image-based representations, making them sensitive to geographical data distribution shifts, such as variation in background, lighting, and environmental conditions. To address this, we introduce WildIng, a Wildlife image Invariant representation model for geographical domain shift. WildIng integrates text descriptions with image features, creating a more robust representation to geographical domain shifts. By leveraging textual descriptions, our approach captures consistent semantic information, such as detailed descriptions of the appearance of the species, improving generalization across different geographical locations. Experiments show that WildIng enhances the accuracy of foundation models such as BioCLIP by 30% under geographical domain shift conditions. We evaluate WildIng on two datasets collected from different regions, namely America and Africa. The code and models are publicly available at https://github.com/Julian075/CATALOG/tree/WildIng.

Paperid: 85, https://arxiv.org/pdf/2601.00623.pdf GitHub

Abstract:
Direct Preference Optimization (DPO) has shown strong potential for mitigating hallucinations in Multimodal Large Language Models (MLLMs). However, existing multimodal DPO approaches often suffer from overfitting due to the difficulty imbalance in preference data. Our analysis shows that MLLMs tend to overemphasize easily distinguishable preference pairs, which hinders fine-grained hallucination suppression and degrades overall performance. To address this issue, we propose Difficulty-Aware Direct Preference Optimization (DA-DPO), a cost-effective framework designed to balance the learning process. DA-DPO consists of two main components: (1) Difficulty Estimation leverages pre-trained vision--language models with complementary generative and contrastive objectives, whose outputs are integrated via a distribution-aware voting strategy to produce robust difficulty scores without additional training; and (2) Difficulty-Aware Training reweights preference pairs based on their estimated difficulty, down-weighting easy samples while emphasizing harder ones to alleviate overfitting. This framework enables more effective preference optimization by prioritizing challenging examples, without requiring new data or extra fine-tuning stages. Extensive experiments demonstrate that DA-DPO consistently improves multimodal preference optimization, yielding stronger robustness to hallucinations and better generalization across standard benchmarks, while remaining computationally efficient. The project page is available at https://artanic30.github.io/project_pages/DA-DPO/.

Paperid: 86, https://arxiv.org/pdf/2601.00388.pdf GitHub

Abstract:
Recent advances in vision-language models have opened up new possibilities for reasoning-driven image geolocalization. However, existing approaches often rely on synthetic reasoning annotations or external image retrieval, which can limit interpretability and generalizability. In this paper, we present Geo-R, a retrieval-free framework that uncovers structured reasoning paths from existing ground-truth coordinates and optimizes geolocation accuracy via reinforcement learning. We propose the Chain of Region, a rule-based hierarchical reasoning paradigm that generates precise, interpretable supervision by mapping GPS coordinates to geographic entities (e.g., country, province, city) without relying on model-generated or synthetic labels. Building on this, we introduce a lightweight reinforcement learning strategy with coordinate-aligned rewards based on Haversine distance, enabling the model to refine predictions through spatially meaningful feedback. Our approach bridges structured geographic reasoning with direct spatial supervision, yielding improved localization accuracy, stronger generalization, and more transparent inference. Experimental results across multiple benchmarks confirm the effectiveness of Geo-R, establishing a new retrieval-free paradigm for scalable and interpretable image geolocalization. To facilitate further research and ensure reproducibility, both the model and code will be made publicly available.

Paperid: 87, https://arxiv.org/pdf/2512.24947.pdf GitHub

Abstract:
Accurate and interpretable crop disease diagnosis is essential for agricultural decision-making, yet existing methods often rely on costly supervised fine-tuning and perform poorly under domain shifts. We propose Caption--Prompt--Judge (CPJ), a training-free few-shot framework that enhances Agri-Pest VQA through structured, interpretable image captions. CPJ employs large vision-language models to generate multi-angle captions, refined iteratively via an LLM-as-Judge module, which then inform a dual-answer VQA process for both recognition and management responses. Evaluated on CDDMBench, CPJ significantly improves performance: using GPT-5-mini captions, GPT-5-Nano achieves \textbf{+22.7} pp in disease classification and \textbf{+19.5} points in QA score over no-caption baselines. The framework provides transparent, evidence-based reasoning, advancing robust and explainable agricultural diagnosis without fine-tuning. Our code and data are publicly available at: https://github.com/CPJ-Agricultural/CPJ-Agricultural-Diagnosis.

Paperid: 88, https://arxiv.org/pdf/2512.24551.pdf GitHub GitHub

Abstract:
Recent advances in text-to-video (T2V) generation have achieved good visual quality, yet synthesizing videos that faithfully follow physical laws remains an open challenge. Existing methods mainly based on graphics or prompt extension struggle to generalize beyond simple simulated environments or learn implicit physical reasoning. The scarcity of training data with rich physics interactions and phenomena is also a problem. In this paper, we first introduce a Physics-Augmented video data construction Pipeline, PhyAugPipe, that leverages a vision-language model (VLM) with chain-of-thought reasoning to collect a large-scale training dataset, PhyVidGen-135K. Then we formulate a principled Physics-aware Groupwise Direct Preference Optimization, PhyGDPO, framework that builds upon the groupwise Plackett-Luce probabilistic model to capture holistic preferences beyond pairwise comparisons. In PhyGDPO, we design a Physics-Guided Rewarding (PGR) scheme that embeds VLM-based physics rewards to steer optimization toward physical consistency. We also propose a LoRA-Switch Reference (LoRA-SR) scheme that eliminates memory-heavy reference duplication for efficient training. Experiments show that our method significantly outperforms state-of-the-art open-source methods on PhyGenBench and VideoPhy2. Please check our project page at https://caiyuanhao1998.github.io/project/PhyGDPO for more video results. Our code, models, and data will be released at https://github.com/caiyuanhao1998/Open-PhyGDPO

Paperid: 89, https://arxiv.org/pdf/2512.24022.pdf GitHub

Abstract:
Large vision-language models (VLMs) exhibit strong performance across various tasks. However, these VLMs encounter significant challenges when applied to the remote sensing domain due to the inherent differences between remote sensing images and natural images. Existing remote sensing VLMs often fail to extract fine-grained visual features and suffer from visual forgetting during deep language processing. To address this, we introduce MF-RSVLM, a Multi-Feature Fusion Remote Sensing Vision--Language Model that effectively extracts and fuses visual features for RS understanding. MF-RSVLM learns multi-scale visual representations and combines global context with local details, improving the capture of small and complex structures in RS scenes. A recurrent visual feature injection scheme ensures the language model remains grounded in visual evidence and reduces visual forgetting during generation. Extensive experiments on diverse RS benchmarks show that MF-RSVLM achieves state-of-the-art or highly competitive performance across remote sensing classification, image captioning, and VQA tasks. Our code is publicly available at https://github.com/Yunkaidang/RSVLM.

Paperid: 90, https://arxiv.org/pdf/2512.23592.pdf GitHub

Abstract:
Vision-language models (VLMs) excel at broad visual understanding but remain coarse-grained, exhibit visual biases, and miss subtle visual details. Existing training corpora reinforce this limitation by emphasizing general recognition ("Is it a cat or a dog?") over fine-grained perception. To address this, we introduce a new training corpus and task designed to enhance the perceptual abilities of VLMs. TWIN is a large-scale dataset of 561,000 image-pair queries that task models to determine whether two visually similar images depict the same object, encouraging attention to nuanced visual cues. The dataset spans a diverse range of everyday objects across contexts, viewpoints, and appearances. Fine-tuning VLMs on TWIN yields notable gains in fine-grained recognition, even on unseen domains such as art, animals, plants, and landmarks. To quantify these gains, we introduce FGVQA, a benchmark suite of 12,000 queries that repurposes fine-grained recognition and retrieval datasets from multiple domains. While existing VLMs struggle on FGVQA, when fine-tuned on TWIN they improve by up to 19.3%, without compromising performance on general VQA benchmarks. Finally, our TWIN dataset scales favorably with object annotations, and our analysis shows that scale is key to performance. We envision TWIN as a drop-in addition to open-source VLM training corpora, advancing perceptual precision of future models. Project webpage: https://glab-caltech.github.io/twin/

Paperid: 91, https://arxiv.org/pdf/2512.23453.pdf GitHub

Abstract:
Large Vision-Language Models (LVLMs) have achieved impressive progress in multi-modal understanding and generation. However, they still tend to produce hallucinated content that is inconsistent with the visual input, which limits their reliability in real-world applications. We propose \textbf{CoFi-Dec}, a training-free decoding framework that mitigates hallucinations by integrating generative self-feedback with coarse-to-fine visual conditioning. Inspired by the human visual process from global scene perception to detailed inspection, CoFi-Dec first generates two intermediate textual responses conditioned on coarse- and fine-grained views of the original image. These responses are then transformed into synthetic images using a text-to-image model, forming multi-level visual hypotheses that enrich grounding cues. To unify the predictions from these multiple visual conditions, we introduce a Wasserstein-based fusion mechanism that aligns their predictive distributions into a geometrically consistent decoding trajectory. This principled fusion reconciles high-level semantic consistency with fine-grained visual grounding, leading to more robust and faithful outputs. Extensive experiments on six hallucination-focused benchmarks show that CoFi-Dec substantially reduces both entity-level and semantic-level hallucinations, outperforming existing decoding strategies. The framework is model-agnostic, requires no additional training, and can be seamlessly applied to a wide range of LVLMs. The implementation is available at https://github.com/AI-Researcher-Team/CoFi-Dec.

Paperid: 92, https://arxiv.org/pdf/2512.23180.pdf GitHub

Abstract:
Driving World Models (DWMs) have been developing rapidly with the advances of generative models. However, existing DWMs lack 3D scene understanding capabilities and can only generate content conditioned on input data, without the ability to interpret or reason about the driving environment. Moreover, current approaches represent 3D spatial information with point cloud or BEV features do not accurately align textual information with the underlying 3D scene. To address these limitations, we propose a novel unified DWM framework based on 3D Gaussian scene representation, which enables both 3D scene understanding and multi-modal scene generation, while also enabling contextual enrichment for understanding and generation tasks. Our approach directly aligns textual information with the 3D scene by embedding rich linguistic features into each Gaussian primitive, thereby achieving early modality alignment. In addition, we design a novel task-aware language-guided sampling strategy that removes redundant 3D Gaussians and injects accurate and compact 3D tokens into LLM. Furthermore, we design a dual-condition multi-modal generation model, where the information captured by our vision-language model is leveraged as a high-level language condition in combination with a low-level image condition, jointly guiding the multi-modal generation process. We conduct comprehensive studies on the nuScenes, and NuInteract datasets to validate the effectiveness of our framework. Our method achieves state-of-the-art performance. We will release the code publicly on GitHub https://github.com/dtc111111/GaussianDWM.

Paperid: 93, https://arxiv.org/pdf/2512.23073.pdf GitHub

Abstract:
Explorations in fine-tuning Vision-Language Models (VLMs), such as Low-Rank Adaptation (LoRA) from Parameter Efficient Fine-Tuning (PEFT), have made impressive progress. However, most approaches rely on explicit weight updates, overlooking the extensive representational structures already encoded in pre-trained models that remain underutilized. Recent works have demonstrated that Mask Fine-Tuning (MFT) can be a powerful and efficient post-training paradigm for language models. Instead of updating weights, MFT assigns learnable gating scores to each weight, allowing the model to reorganize its internal subnetworks for downstream task adaptation. In this paper, we rethink fine-tuning for VLMs from a structural reparameterization perspective grounded in MFT. We apply MFT to the language and projector components of VLMs with different language backbones and compare against strong PEFT baselines. Experiments show that MFT consistently surpasses LoRA variants and even full fine-tuning, achieving high performance without altering the frozen backbone. Our findings reveal that effective adaptation can emerge not only from updating weights but also from reestablishing connections among the model's existing knowledge. Code available at: https://github.com/Ming-K9/MFT-VLM

Paperid: 94, https://arxiv.org/pdf/2512.23035.pdf GitHub

Abstract:
Semi-supervised remote sensing (RS) image semantic segmentation offers a promising solution to alleviate the burden of exhaustive annotation, yet it fundamentally struggles with pseudo-label drift, a phenomenon where confirmation bias leads to the accumulation of errors during training. In this work, we propose Co2S, a stable semi-supervised RS segmentation framework that synergistically fuses priors from vision-language models and self-supervised models. Specifically, we construct a heterogeneous dual-student architecture comprising two distinct ViT-based vision foundation models initialized with pretrained CLIP and DINOv3 to mitigate error accumulation and pseudo-label drift. To effectively incorporate these distinct priors, an explicit-implicit semantic co-guidance mechanism is introduced that utilizes text embeddings and learnable queries to provide explicit and implicit class-level guidance, respectively, thereby jointly enhancing semantic consistency. Furthermore, a global-local feature collaborative fusion strategy is developed to effectively fuse the global contextual information captured by CLIP with the local details produced by DINOv3, enabling the model to generate highly precise segmentation results. Extensive experiments on six popular datasets demonstrate the superiority of the proposed method, which consistently achieves leading performance across various partition protocols and diverse scenarios. Project page is available at https://xavierjiezou.github.io/Co2S/.

Paperid: 95, https://arxiv.org/pdf/2512.22939.pdf GitHub

Abstract:
Autonomous driving requires generating safe and reliable trajectories from complex multimodal inputs. Traditional modular pipelines separate perception, prediction, and planning, while recent end-to-end (E2E) systems learn them jointly. Vision-language models (VLMs) further enrich this paradigm by introducing cross-modal priors and commonsense reasoning, yet current VLM-based planners face three key challenges: (i) a mismatch between discrete text reasoning and continuous control, (ii) high latency from autoregressive chain-of-thought decoding, and (iii) inefficient or non-causal planners that limit real-time deployment. We propose ColaVLA, a unified vision-language-action framework that transfers reasoning from text to a unified latent space and couples it with a hierarchical, parallel trajectory decoder. The Cognitive Latent Reasoner compresses scene understanding into compact, decision-oriented meta-action embeddings through ego-adaptive selection and only two VLM forward passes. The Hierarchical Parallel Planner then generates multi-scale, causality-consistent trajectories in a single forward pass. Together, these components preserve the generalization and interpretability of VLMs while enabling efficient, accurate and safe trajectory generation. Experiments on the nuScenes benchmark show that ColaVLA achieves state-of-the-art performance in both open-loop and closed-loop settings with favorable efficiency and robustness.

Paperid: 96, https://arxiv.org/pdf/2512.22933.pdf GitHub

Abstract:
The rapid spread of multimodal misinformation poses a growing challenge for automated fact-checking systems. Existing approaches, including large vision language models (LVLMs) and deep multimodal fusion methods, often fall short due to limited reasoning and shallow evidence utilization. A key bottleneck is the lack of dedicated datasets that provide complete real-world multimodal misinformation instances accompanied by annotated reasoning processes and verifiable evidence. To address this limitation, we introduce RW-Post, a high-quality and explainable dataset for real-world multimodal fact-checking. RW-Post aligns real-world multimodal claims with their original social media posts, preserving the rich contextual information in which the claims are made. In addition, the dataset includes detailed reasoning and explicitly linked evidence, which are derived from human written fact-checking articles via a large language model assisted extraction pipeline, enabling comprehensive verification and explanation. Building upon RW-Post, we propose AgentFact, an agent-based multimodal fact-checking framework designed to emulate the human verification workflow. AgentFact consists of five specialized agents that collaboratively handle key fact-checking subtasks, including strategy planning, high-quality evidence retrieval, visual analysis, reasoning, and explanation generation. These agents are orchestrated through an iterative workflow that alternates between evidence searching and task-aware evidence filtering and reasoning, facilitating strategic decision-making and systematic evidence analysis. Extensive experimental results demonstrate that the synergy between RW-Post and AgentFact substantially improves both the accuracy and interpretability of multimodal fact-checking.

Paperid: 97, https://arxiv.org/pdf/2512.22905.pdf GitHub

Abstract:
This paper presents JavisGPT, the first unified multimodal large language model (MLLM) for Joint Audio-Video (JAV) comprehension and generation. JavisGPT adopts a concise encoder-LLM-decoder architecture, featuring a SyncFusion module for spatio-temporal audio-video fusion and synchrony-aware learnable queries to bridge a pretrained JAV-DiT generator. This design enables temporally coherent video-audio understanding and generation from multimodal instructions. We design an effective three-stage training pipeline consisting of multimodal pretraining, audio-video fine-tuning, and large-scale instruction-tuning, to progressively build multimodal comprehension and generation from existing vision-language models. To support this, we further construct JavisInst-Omni, a high-quality instruction dataset with over 200K GPT-4o-curated audio-video-text dialogues that span diverse and multi-level comprehension and generation scenarios. Extensive experiments on JAV comprehension and generation benchmarks show that JavisGPT outperforms existing MLLMs, particularly in complex and temporally synchronized settings.

Paperid: 98, https://arxiv.org/pdf/2512.22799.pdf GitHub

Abstract:
Vision-Language Tracking aims to continuously localize objects described by a visual template and a language description. Existing methods, however, are typically limited to local search, making them prone to failures under viewpoint changes, occlusions, and rapid target movements. In this work, we introduce the first global tracking framework based on Multimodal Large Language Models (VPTracker), exploiting their powerful semantic reasoning to locate targets across the entire image space. While global search improves robustness and reduces drift, it also introduces distractions from visually or semantically similar objects. To address this, we propose a location-aware visual prompting mechanism that incorporates spatial priors into the MLLM. Specifically, we construct a region-level prompt based on the target's previous location, enabling the model to prioritize region-level recognition and resort to global inference only when necessary. This design retains the advantages of global tracking while effectively suppressing interference from distracting visual content. Extensive experiments show that our approach significantly enhances tracking stability and target disambiguation under challenging scenarios, opening a new avenue for integrating MLLMs into visual tracking. Code is available at https://github.com/jcwang0602/VPTracker.

Paperid: 99, https://arxiv.org/pdf/2512.22519.pdf GitHub

Abstract:
Recent Vision-Language-Action (VLA) models have made impressive progress toward general-purpose robotic manipulation by post-training large Vision-Language Models (VLMs) for action prediction. Yet most VLAs entangle perception and control in a monolithic pipeline optimized purely for action, which can erode language-conditioned grounding. In our real-world tabletop tests, policies over-grasp when the target is absent, are distracted by clutter, and overfit to background appearance. To address these issues, we propose OBEYED-VLA (OBject-centric and gEometrY groundED VLA), a framework that explicitly disentangles perceptual grounding from action reasoning. Instead of operating directly on raw RGB, OBEYED-VLA augments VLAs with a perception module that grounds multi-view inputs into task-conditioned, object-centric, and geometry-aware observations. This module includes a VLM-based object-centric grounding stage that selects task-relevant object regions across camera views, along with a complementary geometric grounding stage that emphasizes the 3D structure of these objects over their appearance. The resulting grounded views are then fed to a pretrained VLA policy, which we fine-tune exclusively on single-object demonstrations collected without environmental clutter or non-target objects. On a real-world UR10e tabletop setup, OBEYED-VLA substantially improves robustness over strong VLA baselines across four challenging regimes and multiple difficulty levels: distractor objects, absent-target rejection, background appearance changes, and cluttered manipulation of unseen objects. Ablation studies confirm that both semantic grounding and geometry-aware grounding are critical to these gains. Overall, the results indicate that making perception an explicit, object-centric component is an effective way to strengthen and generalize VLA-based robotic manipulation.

Paperid: 100, https://arxiv.org/pdf/2512.22342.pdf GitHub

Abstract:
In most existing embodied navigation tasks, instructions are well-defined and unambiguous, such as instruction following and object searching. Under this idealized setting, agents are required solely to produce effective navigation outputs conditioned on vision and language inputs. However, real-world navigation instructions are often vague and ambiguous, requiring the agent to resolve uncertainty and infer user intent through active dialog. To address this gap, we propose Interactive Instance Object Navigation (IION), a task that requires agents not only to generate navigation actions but also to produce language outputs via active dialog, thereby aligning more closely with practical settings. IION extends Instance Object Navigation (ION) by allowing agents to freely consult an oracle in natural language while navigating. Building on this task, we present the Vision Language-Language Navigation (VL-LN) benchmark, which provides a large-scale, automatically generated dataset and a comprehensive evaluation protocol for training and assessing dialog-enabled navigation models. VL-LN comprises over 41k long-horizon dialog-augmented trajectories for training and an automatic evaluation protocol with an oracle capable of responding to agent queries. Using this benchmark, we train a navigation model equipped with dialog capabilities and show that it achieves significant improvements over the baselines. Extensive experiments and analyses further demonstrate the effectiveness and reliability of VL-LN for advancing research on dialog-enabled embodied navigation. Code and dataset: https://0309hws.github.io/VL-LN.github.io/

Paperid: 101, https://arxiv.org/pdf/2512.22188.pdf GitHub

Abstract:
Multiple Instance Learning (MIL) has enabled weakly supervised analysis of whole-slide images (WSIs) in computational pathology. However, traditional MIL approaches often lose crucial contextual information, while transformer-based variants, though more expressive, suffer from quadratic complexity and redundant computations. To address these limitations, we propose HookMIL, a context-aware and computationally efficient MIL framework that leverages compact, learnable hook tokens for structured contextual aggregation. These tokens can be initialized from (i) key-patch visual features, (ii) text embeddings from vision-language pathology models, and (iii) spatially grounded features from spatial transcriptomics-vision models. This multimodal initialization enables Hook Tokens to incorporate rich textual and spatial priors, accelerating convergence and enhancing representation quality. During training, Hook tokens interact with instances through bidirectional attention with linear complexity. To further promote specialization, we introduce a Hook Diversity Loss that encourages each token to focus on distinct histopathological patterns. Additionally, a hook-to-hook communication mechanism refines contextual interactions while minimizing redundancy. Extensive experiments on four public pathology datasets demonstrate that HookMIL achieves state-of-the-art performance, with improved computational efficiency and interpretability. Codes are available at https://github.com/lingxitong/HookMIL.

Paperid: 102, https://arxiv.org/pdf/2512.21999.pdf GitHub

Abstract:
While Vision-Language Models (VLMs) have garnered increasing attention in the AI community due to their promising practical applications, they exhibit persistent hallucination issues, generating outputs misaligned with visual inputs. Recent studies attribute these hallucinations to VLMs' over-reliance on linguistic priors and insufficient visual feature integration, proposing heuristic decoding calibration strategies to mitigate them. However, the non-trainable nature of these strategies inherently limits their optimization potential. To this end, we propose an adversarial parametric editing framework for Hallucination mitigation in VLMs, which follows an \textbf{A}ctivate-\textbf{L}ocate-\textbf{E}dit \textbf{A}dversarially paradigm. Specifically, we first construct an activation dataset that comprises grounded responses (positive samples attentively anchored in visual features) and hallucinatory responses (negative samples reflecting LLM prior bias and internal knowledge artifacts). Next, we identify critical hallucination-prone parameter clusters by analyzing differential hidden states of response pairs. Then, these clusters are fine-tuned using prompts injected with adversarial tuned prefixes that are optimized to maximize visual neglect, thereby forcing the model to prioritize visual evidence over inherent parametric biases. Evaluations on both generative and discriminative VLM tasks demonstrate the significant effectiveness of ALEAHallu in alleviating hallucinations. Our code is available at https://github.com/hujiayu1223/ALEAHallu.

Paperid: 103, https://arxiv.org/pdf/2512.21450.pdf GitHub GitHub

Abstract:
We present an RL-central framework for Language and Vision Assistants (RLLaVA) with its formulation of Markov decision process (MDP). RLLaVA decouples RL algorithmic logic from model architecture and distributed execution, supporting researchers in implementing new RL algorithms with minimal code, and to plug in a broad family of RL methods and vision-language models (VLMs) while remaining agnostic to specific training and inference engines. RLLaVA makes resource-efficient training of 1B--7B models feasible on common GPUs; notably, 4B-scale models can be trained end-to-end with full-parameter updates on a single 24GB GPU. Experiments on multi-modal and agentic tasks demonstrate that RLLaVA has task extensibility, and the models trained with it consistently improve performance over base models, competitive with other specially engineered RL frameworks. The code is available at https://github.com/TinyLoopX/RLLaVA.

Paperid: 104, https://arxiv.org/pdf/2512.21414.pdf GitHub

Abstract:
Recent tool-use frameworks powered by vision-language models (VLMs) improve image understanding by grounding model predictions with specialized tools. Broadly, these frameworks leverage VLMs and a pre-specified toolbox to decompose the prediction task into multiple tool calls (often deep learning models) which are composed to make a prediction. The dominant approach to composing tools is using text, via function calls embedded in VLM-generated code or natural language. However, these methods often perform poorly on medical image understanding, where salient information is encoded as spatially-localized features that are difficult to compose or fuse via text alone. To address this, we propose a tool-use framework for medical image understanding called the Tool Bottleneck Framework (TBF), which composes VLM-selected tools using a learned Tool Bottleneck Model (TBM). For a given image and task, TBF leverages an off-the-shelf medical VLM to select tools from a toolbox that each extract clinically-relevant features. Instead of text-based composition, these tools are composed by the TBM, which computes and fuses the tool outputs using a neural network before outputting the final prediction. We propose a simple and effective strategy for TBMs to make predictions with any arbitrary VLM tool selection. Overall, our framework not only improves tool-use in medical imaging contexts, but also yields more interpretable, clinically-grounded predictors. We evaluate TBF on tasks in histopathology and dermatology and find that these advantages enable our framework to perform on par with or better than deep learning-based classifiers, VLMs, and state-of-the-art tool-use frameworks, with particular gains in data-limited regimes. Our code is available at https://github.com/christinaliu2020/tool-bottleneck-framework.

Paperid: 105, https://arxiv.org/pdf/2512.21337.pdf GitHub

Abstract:
We expose a significant popularity bias in state-of-the-art vision-language models (VLMs), which achieve up to 34% higher accuracy on famous buildings compared to ordinary ones, indicating a reliance on memorization over generalizable understanding. To systematically investigate this, we introduce the largest open benchmark for this task: the YearGuessr dataset, a collection of 55,546 building images with multi-modal attributes from 157 countries, annotated with continuous ordinal labels of their construction year (1001-2024), GPS data, and page-view counts as a proxy for popularity. Using this dataset, we frame the construction year prediction task as ordinal regression and introduce popularity-aware interval accuracy metrics to quantify this bias. Our resulting benchmark of 30+ models, including our YearCLIP model, confirms that VLMs excel on popular, memorized items but struggle significantly with unrecognized subjects, exposing a critical flaw in their reasoning capabilities. Project page: https://sytwu.github.io/BeyondMemo/

Paperid: 106, https://arxiv.org/pdf/2512.21221.pdf GitHub

Abstract:
Retrieving images from natural language descriptions is a core task at the intersection of computer vision and natural language processing, with wide-ranging applications in search engines, media archiving, and digital content management. However, real-world image-text retrieval remains challenging due to vague or context-dependent queries, linguistic variability, and the need for scalable solutions. In this work, we propose a lightweight two-stage retrieval pipeline that leverages event-centric entity extraction to incorporate temporal and contextual signals from real-world captions. The first stage performs efficient candidate filtering using BM25 based on salient entities, while the second stage applies BEiT-3 models to capture deep multimodal semantics and rerank the results. Evaluated on the OpenEvents v1 benchmark, our method achieves a mean average precision of 0.559, substantially outperforming prior baselines. These results highlight the effectiveness of combining event-guided filtering with long-text vision-language modeling for accurate and efficient retrieval in complex, real-world scenarios. Our code is available at https://github.com/PhamPhuHoa-23/Event-Based-Image-Retrieval

Paperid: 107, https://arxiv.org/pdf/2512.21095.pdf GitHub

Abstract:
Text and formulas constitute the core informational components of many documents. Accurately and efficiently recognizing both is crucial for developing robust and generalizable document parsing systems. Recently, vision-language models (VLMs) have achieved impressive unified recognition of text and formulas. However, they are large-sized and computationally demanding, restricting their usage in many applications. In this paper, we propose UniRec-0.1B, a unified recognition model with only 0.1B parameters. It is capable of performing text and formula recognition at multiple levels, including characters, words, lines, paragraphs, and documents. To implement this task, we first establish UniRec40M, a large-scale dataset comprises 40 million text, formula and their mix samples, enabling the training of a powerful yet lightweight model. Secondly, we identify two challenges when building such a lightweight but unified expert model. They are: structural variability across hierarchies and semantic entanglement between textual and formulaic content. To tackle these, we introduce a hierarchical supervision training that explicitly guides structural comprehension, and a semantic-decoupled tokenizer that separates text and formula representations. Finally, we develop a comprehensive evaluation benchmark covering Chinese and English documents from multiple domains and with multiple levels. Experimental results on this and public benchmarks demonstrate that UniRec-0.1B outperforms both general-purpose VLMs and leading document parsing expert models, while achieving a 2-9$\times$ speedup, validating its effectiveness and efficiency. Codebase and Dataset: https://github.com/Topdu/OpenOCR.

Paperid: 108, https://arxiv.org/pdf/2512.20940.pdf GitHub

Abstract:
Vision-Language Navigation in Continuous Environments (VLN-CE) requires an embodied agent to navigate towards target in continuous environments, following natural language instructions. While current graph-based methods offer an efficient, structured approach by abstracting the environment into a topological map and simplifying the action space to waypoint selection, they lag behind methods based on Large Vision-Language Models (LVLMs) in leveraging large-scale data and advanced training paradigms. In this paper, we try to bridge this gap by introducing ETP-R1, a framework that applies the paradigm of scaling up data and Reinforcement Fine-Tuning (RFT) to a graph-based VLN-CE model. To build a strong foundation, we first construct a high-quality, large-scale pretraining dataset using the Gemini API. This dataset consists of diverse, low-hallucination instructions for topological trajectories, providing rich supervision for our graph-based policy to map language to topological paths. This foundation is further strengthened by unifying data from both R2R and RxR tasks for joint pretraining. Building on this, we introduce a three-stage training paradigm, which culminates in the first application of closed-loop, online RFT to a graph-based VLN-CE model, powered by the Group Relative Policy Optimization (GRPO) algorithm. Extensive experiments demonstrate that our approach is highly effective, establishing new state-of-the-art performance across all major metrics on both the R2R-CE and RxR-CE benchmarks. Our code is available at https://github.com/Cepillar/ETP-R1.

Paperid: 109, https://arxiv.org/pdf/2512.20934.pdf GitHub

Abstract:
Spatial reasoning in 3D scenes requires precise geometric calculations that challenge vision-language models. Visual programming addresses this by decomposing problems into steps calling specialized tools, yet existing methods rely on either fixed toolsets or speculative tool induction before solving problems, resulting in suboptimal programs and poor utilization of induced tools. We present Transductive Visual Programming (TVP), a novel framework that builds new tools from its own experience rather than speculation. TVP first solves problems using basic tools while accumulating experiential solutions into an Example Library, then abstracts recurring patterns from these programs into reusable higher-level tools for an evolving Tool Library. This allows TVP to tackle new problems with increasingly powerful tools learned from experience. On Omni3D-Bench, TVP achieves state-of-the-art performance, outperforming GPT-4o by 22% and the previous best visual programming system by 11%. Our transductively learned tools are used 5x more frequently as core program dependency than inductively created ones, demonstrating more effective tool discovery and reuse. The evolved tools also show strong generalization to unseen spatial tasks, achieving superior performance on benchmarks from SpatialScore-Hard collection without any testset-specific modification. Our work establishes experience-driven transductive tool creation as a powerful paradigm for building self-evolving visual programming agents that effectively tackle challenging spatial reasoning tasks. We release our code at https://transductive-visualprogram.github.io/.

Paperid: 110, https://arxiv.org/pdf/2512.20839.pdf GitHub

Abstract:
Vision-Language Models (VLMs) have demonstrated strong performance on multimodal reasoning tasks, but their deployment remains challenging due to high inference latency and computational cost, particularly when processing high-resolution visual inputs. While recent architectures such as FastVLM improve efficiency through optimized vision encoders, existing pipelines still rely on static visual preprocessing, leading to redundant computation for visually simple inputs. In this work, we propose an adaptive visual preprocessing method that dynamically adjusts input resolution and spatial coverage based on image content characteristics. The proposed approach combines content-aware image analysis, adaptive resolution selection, and content-aware cropping to reduce visual redundancy prior to vision encoding. Importantly, the method is integrated with FastVLM without modifying its architecture or requiring retraining. We evaluate the proposed method on a subset of the DocVQA dataset in an inference-only setting, focusing on efficiency-oriented metrics. Experimental results show that adaptive preprocessing reduces per-image inference time by over 50\%, lowers mean full generation time, and achieves a consistent reduction of more than 55\% in visual token count compared to the baseline pipeline. These findings demonstrate that input-aware preprocessing is an effective and lightweight strategy for improving deployment-oriented efficiency of vision-language models. To facilitate reproducibility, our implementation is provided as a fork of the FastVLM repository, incorporating the files for the proposed method, and is available at https://github.com/kmdavidds/mlfastlm.

Paperid: 111, https://arxiv.org/pdf/2512.19453.pdf GitHub

Abstract:
Embodied robotic AI systems designed to manage complex daily tasks rely on a task planner to understand and decompose high-level tasks. While most research focuses on enhancing the task-understanding abilities of LLMs/VLMs through fine-tuning or chain-of-thought prompting, this paper argues that defining the planned skill set is equally crucial. To handle the complexity of daily environments, the skill set should possess a high degree of generalization ability. Empirically, more abstract expressions tend to be more generalizable. Therefore, we propose to abstract the planned result as a set of meta-actions. Each meta-action comprises three components: {move/rotate, end-effector status change, relationship with the environment}. This abstraction replaces human-centric concepts, such as grasping or pushing, with the robot's intrinsic functionalities. As a result, the planned outcomes align seamlessly with the complete range of actions that the robot is capable of performing. Furthermore, to ensure that the LLM/VLM accurately produces the desired meta-action format, we employ the Retrieval-Augmented Generation (RAG) technique, which leverages a database of human-annotated planning demonstrations to facilitate in-context learning. As the system successfully completes more tasks, the database will self-augment to continue supporting diversity. The meta-action set and its integration with RAG are two novel contributions of our planner, denoted as MaP-AVR, the meta-action planner for agents composed of VLM and RAG. To validate its efficacy, we design experiments using GPT-4o as the pre-trained LLM/VLM model and OmniGibson as our robotic platform. Our approach demonstrates promising performance compared to the current state-of-the-art method. Project page: https://map-avr.github.io/.

Paperid: 112, https://arxiv.org/pdf/2512.19302.pdf GitHub

Abstract:
Large Vision-Language Models (LVLMs) hold great promise for advancing remote sensing (RS) analysis, yet existing reasoning segmentation frameworks couple linguistic reasoning and pixel prediction through end-to-end supervised fine-tuning, leading to weak geometric grounding and limited generalization across tasks. To address this, we developed Think2Seg-RS, a decoupled framework that trains an LVLM prompter to control a frozen Segment Anything Model (SAM) via structured geometric prompts. Through a mask-only reinforcement learning objective, the LVLM learns to translate abstract semantic reasoning into spatially grounded actions, achieving state-of-the-art performance on the EarthReason dataset. Remarkably, the learned prompting policy generalizes zero-shot to multiple referring segmentation benchmarks, exposing a distinct divide between semantic-level and instance-level grounding. We further found that compact segmenters outperform larger ones under semantic-level supervision, and that negative prompts are ineffective in heterogeneous aerial backgrounds. Together, these findings establish semantic-level reasoning segmentation as a new paradigm for geospatial understanding, opening the way toward unified, interpretable LVLM-driven Earth observation. Our code and model are available at https://github.com/Ricardo-XZ/Think2Seg-RS.

Paperid: 113, https://arxiv.org/pdf/2512.19070.pdf GitHub

Abstract:
Large Vision-Language Models (LVLMs) bridge the gap between visual and linguistic modalities, demonstrating strong potential across a variety of domains. However, despite significant progress, LVLMs still suffer from severe hallucination issues in object recognition tasks. These models often fail to accurately identify certain objects, leading to text generation that appears fluent but does not correspond to the visual content, which can have serious consequences in real-world applications. Recently, several methods have been proposed to alleviate LVLM hallucinations, but most focus solely on reducing hallucinations in the language modality. To mitigate hallucinations in both the language and visual modalities, we introduce Hallucination Disentangled Decoding (HDD) method that requires no training. HDD enhances the original image by segmenting it and selecting images that augment the original, while also utilizing a blank image to eliminate language prior hallucinations in both the original and segmented images. This design not only reduces the model's dependence on language priors but also enhances its visual performance. (Code: https://github.com/rickeyhhh/Hallucination-Disentangled-Decoding)

Paperid: 114, https://arxiv.org/pdf/2512.18747.pdf GitHub

Abstract:
Multimodal Large Language Models (MLLMs) deliver strong vision-language performance but at high computational cost, driven by numerous visual tokens processed by the Vision Transformer (ViT) encoder. Existing token pruning strategies are inadequate: LLM-stage token pruning overlooks the ViT's overhead, while conventional ViT token pruning, without language guidance, risks discarding textually critical visual cues and introduces feature distortions amplified by the ViT's bidirectional attention. To meet these challenges, we propose IPCV, a training-free, information-preserving compression framework for MLLM visual encoders. IPCV enables aggressive token pruning inside the ViT via Neighbor-Guided Reconstruction (NGR) that temporarily reconstructs pruned tokens to participate in attention with minimal overhead, then fully restores them before passing to the LLM. Besides, we introduce Attention Stabilization (AS) to further alleviate the negative influence from token pruning by approximating the K/V of pruned tokens. It can be directly applied to previous LLM-side token pruning methods to enhance their performance. Extensive experiments show that IPCV substantially reduces end-to-end computation and outperforms state-of-the-art training-free token compression methods across diverse image and video benchmarks. Our code is available at https://github.com/Perkzi/IPCV.

Paperid: 115, https://arxiv.org/pdf/2512.17436.pdf GitHub

Abstract:
We open-source MiMo-VL-Miloco-7B and its quantized variant MiMo-VL-Miloco-7B-GGUF, a pair of home-centric vision-language models that achieve strong performance on both home-scenario understanding and general multimodal reasoning. Built on the MiMo-VL-7B backbone, MiMo-VL-Miloco-7B is specialized for smart-home environments, attaining leading F1 scores on gesture recognition and common home-scenario understanding, while also delivering consistent gains across video benchmarks such as Video-MME, Video-MMMU, and Charades-STA, as well as language understanding benchmarks including MMMU-Pro and MMLU-Pro. In our experiments, MiMo-VL-Miloco-7B outperforms strong closed-source and open-source baselines on home-scenario understanding and several multimodal reasoning benchmarks. To balance specialization and generality, we design a two-stage training pipeline that combines supervised fine-tuning with reinforcement learning based on Group Relative Policy Optimization, leveraging efficient multi-domain data. We further incorporate chain-of-thought supervision and token-budget-aware reasoning, enabling the model to learn knowledge in a data-efficient manner while also performing reasoning efficiently. Our analysis shows that targeted home-scenario training not only enhances activity and gesture understanding, but also improves text-only reasoning with only modest trade-offs on document-centric tasks. Model checkpoints, quantized GGUF weights, and our home-scenario evaluation toolkit are publicly available at https://github.com/XiaoMi/xiaomi-mimo-vl-miloco to support research and deployment in real-world smart-home applications.

Paperid: 116, https://arxiv.org/pdf/2512.17319.pdf GitHub

Abstract:
Multimodal large language models (MLLMs) demonstrate strong perception and reasoning performance on existing remote sensing (RS) benchmarks. However, most prior benchmarks rely on low-resolution imagery, and some high-resolution benchmarks suffer from flawed reasoning-task designs. We show that text-only LLMs can perform competitively with multimodal vision-language models on RS reasoning tasks without access to images, revealing a critical mismatch between current benchmarks and the intended evaluation of visual understanding. To enable faithful assessment, we introduce RSHR-Bench, a super-high-resolution benchmark for RS visual understanding and reasoning. RSHR-Bench contains 5,329 full-scene images with a long side of at least 4,000 pixels, with up to about 3 x 10^8 pixels per image, sourced from widely used RS corpora and UAV collections. We design four task families: multiple-choice VQA, open-ended VQA, image captioning, and single-image evaluation. These tasks cover nine perception categories and four reasoning types, supporting multi-turn and multi-image dialog. To reduce reliance on language priors, we apply adversarial filtering with strong LLMs followed by rigorous human verification. Overall, we construct 3,864 VQA tasks, 3,913 image captioning tasks, and 500 fully human-written or verified single-image evaluation VQA pairs. Evaluations across open-source, closed-source, and RS-specific VLMs reveal persistent performance gaps in super-high-resolution scenarios. Code: https://github.com/Yunkaidang/RSHR

Paperid: 117, https://arxiv.org/pdf/2512.17213.pdf GitHub

Abstract:
Medical Vision-Language Models (VLMs) are prone to hallucinations, compromising clinical reliability. While reinforcement learning methods like Group Relative Policy Optimization (GRPO) offer a low-cost alignment solution, their reliance on sparse, outcome-based rewards inadvertently encourages models to "overthink" -- generating verbose, convoluted, and unverifiable Chain-of-Thought reasoning to justify answers. This focus on outcomes obscures factual errors and poses significant safety risks. To address this, we propose CheXPO-v2, a novel alignment framework that shifts from outcome to process supervision. Our core innovation is a Knowledge Graph Consistency Reward mechanism driven by Entity-Relation Matching. By explicitly parsing reasoning steps into structured "Disease, Relation, Anatomy" triplets, we provide fine-grained supervision that penalizes incoherent logic and hallucinations at the atomic level. Integrating this with a hard-example mining strategy, our approach significantly outperforms GRPO and state-of-the-art models on benchmarks like MIMIC-CXR-VQA. Crucially, CheXPO-v2 achieves new state-of-the-art accuracy using only 5k samples, demonstrating exceptional data efficiency while producing clinically sound and verifiable reasoning. The project source code is publicly available at: https://github.com/ecoxial2007/CheX-Phi4MM.

Paperid: 118, https://arxiv.org/pdf/2512.16909.pdf GitHub

Abstract:
Mobile manipulators in households must both navigate and manipulate. This requires a compact, semantically rich scene representation that captures where objects are, how they function, and which parts are actionable. Scene graphs are a natural choice, yet prior work often separates spatial and functional relations, treats scenes as static snapshots without object states or temporal updates, and overlooks information most relevant for accomplishing the current task. To address these limitations, we introduce MomaGraph, a unified scene representation for embodied agents that integrates spatial-functional relationships and part-level interactive elements. However, advancing such a representation requires both suitable data and rigorous evaluation, which have been largely missing. We thus contribute MomaGraph-Scenes, the first large-scale dataset of richly annotated, task-driven scene graphs in household environments, along with MomaGraph-Bench, a systematic evaluation suite spanning six reasoning capabilities from high-level planning to fine-grained scene understanding. Built upon this foundation, we further develop MomaGraph-R1, a 7B vision-language model trained with reinforcement learning on MomaGraph-Scenes. MomaGraph-R1 predicts task-oriented scene graphs and serves as a zero-shot task planner under a Graph-then-Plan framework. Extensive experiments demonstrate that our model achieves state-of-the-art results among open-source models, reaching 71.6% accuracy on the benchmark (+11.4% over the best baseline), while generalizing across public benchmarks and transferring effectively to real-robot experiments.

Paperid: 119, https://arxiv.org/pdf/2512.16760.pdf GitHub

Abstract:
Autonomous driving has long relied on modular "Perception-Decision-Action" pipelines, where hand-crafted interfaces and rule-based components often break down in complex or long-tailed scenarios. Their cascaded design further propagates perception errors, degrading downstream planning and control. Vision-Action (VA) models address some limitations by learning direct mappings from visual inputs to actions, but they remain opaque, sensitive to distribution shifts, and lack structured reasoning or instruction-following capabilities. Recent progress in Large Language Models (LLMs) and multimodal learning has motivated the emergence of Vision-Language-Action (VLA) frameworks, which integrate perception with language-grounded decision making. By unifying visual understanding, linguistic reasoning, and actionable outputs, VLAs offer a pathway toward more interpretable, generalizable, and human-aligned driving policies. This work provides a structured characterization of the emerging VLA landscape for autonomous driving. We trace the evolution from early VA approaches to modern VLA frameworks and organize existing methods into two principal paradigms: End-to-End VLA, which integrates perception, reasoning, and planning within a single model, and Dual-System VLA, which separates slow deliberation (via VLMs) from fast, safety-critical execution (via planners). Within these paradigms, we further distinguish subclasses such as textual vs. numerical action generators and explicit vs. implicit guidance mechanisms. We also summarize representative datasets and benchmarks for evaluating VLA-based driving systems and highlight key challenges and open directions, including robustness, interpretability, and instruction fidelity. Overall, this work aims to establish a coherent foundation for advancing human-compatible autonomous driving systems.

Paperid: 120, https://arxiv.org/pdf/2512.15885.pdf GitHub

Abstract:
Multimodal Large Language Models (MLLMs) have recently demonstrated impressive capabilities in connecting vision and language, yet their proficiency in fundamental visual reasoning tasks remains limited. This limitation can be attributed to the fact that MLLMs learn visual understanding primarily from textual descriptions, which constitute a subjective and inherently incomplete supervisory signal. Furthermore, the modest scale of multimodal instruction tuning compared to massive text-only pre-training leads MLLMs to overfit language priors while overlooking visual details. To address these issues, we introduce JARVIS, a JEPA-inspired framework for self-supervised visual enhancement in MLLMs. Specifically, we integrate the I-JEPA learning paradigm into the standard vision-language alignment pipeline of MLLMs training. Our approach leverages frozen vision foundation models as context and target encoders, while training the predictor, implemented as the early layers of an LLM, to learn structural and semantic regularities from images without relying exclusively on language supervision. Extensive experiments on standard MLLM benchmarks show that JARVIS consistently improves performance on vision-centric benchmarks across different LLM families, without degrading multimodal reasoning abilities. Our source code is publicly available at: https://github.com/aimagelab/JARVIS.

Paperid: 121, https://arxiv.org/pdf/2512.15713.pdf GitHub

Abstract:
In recent multimodal research, the diffusion paradigm has emerged as a promising alternative to the autoregressive paradigm (AR), owing to its unique decoding advantages. However, due to the capability limitations of the base diffusion language model, the performance of the diffusion vision language model (dVLM) still lags significantly behind that of mainstream models. This leads to a simple yet fundamental question: Is it possible to construct dVLMs based on existing powerful AR models? In response, we propose DiffusionVL, a dVLM family that could be translated from any powerful AR models. Through simple fine-tuning, we successfully adapt AR pre-trained models into the diffusion paradigm. This approach yields two key observations: (1) The paradigm shift from AR-based multimodal models to diffusion is remarkably effective. (2) Direct conversion of an AR language model to a dVLM is also feasible, achieving performance competitive with LLaVA-style visual-instruction-tuning. Further, we introduce a block-decoding design into dVLMs that supports arbitrary-length generation and KV cache reuse, achieving a significant inference speedup. We conduct a large number of experiments. Despite training with less than 5% of the data required by prior methods, DiffusionVL achieves a comprehensive performance improvement-a 34.4% gain on the MMMU-Pro (vision) bench and 37.5% gain on the MME (Cog.) bench-alongside a 2x inference speedup. The model and code are released at https://github.com/hustvl/DiffusionVL.

Paperid: 122, https://arxiv.org/pdf/2512.15701.pdf GitHub

Abstract:
Evaluations of image compression performance which include human preferences have generally found that naive distortion functions such as MSE are insufficiently aligned to human perception. In order to align compression models to human perception, prior work has employed differentiable perceptual losses consisting of neural networks calibrated on large-scale datasets of human psycho-visual judgments. We show that, surprisingly, state-of-the-art vision-language models (VLMs) can replicate binary human two-alternative forced choice (2AFC) judgments zero-shot when asked to reason about the differences between pairs of images. Motivated to exploit the powerful zero-shot visual reasoning capabilities of VLMs, we propose Vision-Language Models for Image Compression (VLIC), a diffusion-based image compression system designed to be post-trained with binary VLM judgments. VLIC leverages existing techniques for diffusion model post-training with preferences, rather than distilling the VLM judgments into a separate perceptual loss network. We show that calibrating this system on VLM judgments produces competitive or state-of-the-art performance on human-aligned visual compression depending on the dataset, according to perceptual metrics and large-scale user studies. We additionally conduct an extensive analysis of the VLM-based reward design and training procedure and share important insights. More visuals are available at https://kylesargent.github.io/vlic

Paperid: 123, https://arxiv.org/pdf/2512.14735.pdf GitHub

Abstract:
This paper proposes PyFi, a novel framework for pyramid-like financial image understanding that enables vision language models (VLMs) to reason through question chains in a progressive, simple-to-complex manner. At the core of PyFi is PyFi-600K, a dataset comprising 600K financial question-answer pairs organized into a reasoning pyramid: questions at the base require only basic perception, while those toward the apex demand increasing levels of capability in financial visual understanding and expertise. This data is scalable because it is synthesized without human annotations, using PyFi-adv, a multi-agent adversarial mechanism under the Monte Carlo Tree Search (MCTS) paradigm, in which, for each image, a challenger agent competes with a solver agent by generating question chains that progressively probe deeper capability levels in financial visual reasoning. Leveraging this dataset, we present fine-grained, hierarchical, and comprehensive evaluations of advanced VLMs in the financial domain. Moreover, fine-tuning Qwen2.5-VL-3B and Qwen2.5-VL-7B on the pyramid-structured question chains enables these models to answer complex financial questions by decomposing them into sub-questions with gradually increasing reasoning demands, yielding average accuracy improvements of 19.52% and 8.06%, respectively, on the dataset. All resources of code, dataset and models are available at: https://github.com/AgenticFinLab/PyFi .

Paperid: 124, https://arxiv.org/pdf/2512.14661.pdf GitHub

Abstract:
Vision-Language Models (VLMs) have demonstrated strong performance on tasks such as video captioning and visual question answering. However, their growing scale and video-level inputs lead to significant computational and memory overhead, posing challenges for real-time deployment on hardware accelerators. While prior work attempts to reduce redundancy via token pruning or merging, these methods typically operate at coarse granularity and incur high runtime overhead due to global token-level operations. In this study, we propose Focus, a Streaming Concentration Architecture that efficiently accelerates VLM inference through progressive, fine-grained redundancy elimination. Focus introduces a multilevel concentration paradigm that hierarchically compresses vision-language inputs at three levels: (1) semantic-guided token pruning based on textual prompts, (2) spatial-temporal block-level concentration using localized comparisons, and (3) vector-level redundancy removal via motion-aware matching. All concentration steps are tightly co-designed with the architecture to support streaming-friendly, on-chip execution. Focus leverages GEMM tiling, convolution-style layout, and cross-modal attention to minimize off-chip access while enabling high throughput. Implemented as a modular unit within a systolic-array accelerator, Focus achieves a 2.4x speedup and 3.3x reduction in energy, significantly outperforming state-of-the-art accelerators in both performance and energy efficiency. Full-stack implementation of Focus is open-sourced at https://github.com/dubcyfor3/Focus.

Paperid: 125, https://arxiv.org/pdf/2512.14480.pdf GitHub

Abstract:
Contrastive Language-Image Pretraining (CLIP) achieves strong generalization in vision-language tasks by aligning images and texts in a shared embedding space. However, recent findings show that CLIP-like models still underutilize fine-grained semantic signals in text, and this issue becomes even more pronounced when dealing with long and detailed captions. This stems from CLIP's training objective, which optimizes only global image-text similarity and overlooks token-level supervision - limiting its ability to achieve fine-grained visual-text alignment. To address this, we propose SuperCLIP, a simple yet effective framework that augments contrastive learning with classification-based supervision. By adding only a lightweight linear layer to the vision encoder, SuperCLIP leverages token-level cues to enhance visual-textual alignment - with just a 0.077% increase in total FLOPs, and no need for additional annotated data. Experiments show that SuperCLIP consistently improves zero-shot classification, image-text retrieval, and purely visual tasks. These gains hold regardless of whether the model is trained on original web data or rich re-captioned data, demonstrating SuperCLIP's ability to recover textual supervision in both cases. Furthermore, SuperCLIP alleviates CLIP's small-batch performance drop through classification-based supervision that avoids reliance on large batch sizes. Code and models will be made open source.

Paperid: 126, https://arxiv.org/pdf/2512.14014.pdf GitHub

Abstract:
World models have shown great utility in improving the task performance of embodied agents. While prior work largely focuses on pixel-space world models, these approaches face practical limitations in GUI settings, where predicting complex visual elements in future states is often difficult. In this work, we explore an alternative formulation of world modeling for GUI agents, where state transitions are described in natural language rather than predicting raw pixels. First, we introduce MobileWorldBench, a benchmark that evaluates the ability of vision-language models (VLMs) to function as world models for mobile GUI agents. Second, we release MobileWorld, a large-scale dataset consisting of 1.4M samples, that significantly improves the world modeling capabilities of VLMs. Finally, we propose a novel framework that integrates VLM world models into the planning framework of mobile agents, demonstrating that semantic world models can directly benefit mobile agents by improving task success rates. The code and dataset is available at https://github.com/jacklishufan/MobileWorld

Paperid: 127, https://arxiv.org/pdf/2512.13953.pdf GitHub

Abstract:
The rapid progress of text-to-image diffusion models raises significant concerns regarding the unauthorized reproduction of trademarked content. While prior work targets general concepts (e.g., styles, celebrities), it fails to address specific brand identifiers. Crucially, we note that brand recognition is multi-dimensional, extending beyond explicit logos to encompass distinctive structural features (e.g., a car's front grille). To tackle this, we introduce unbranding, a novel task for the fine-grained removal of both trademarks and subtle structural brand features, while preserving semantic coherence. To facilitate research, we construct a comprehensive benchmark dataset. Recognizing that existing brand detectors are limited to logos and fail to capture abstract trade dress (e.g., the shape of a Coca-Cola bottle), we introduce a novel evaluation metric based on Vision Language Models (VLMs). This VLM-based metric uses a question-answering framework to probe images for both explicit logos and implicit, holistic brand characteristics. Furthermore, we observe that as model fidelity increases, with newer systems (SDXL, FLUX) synthesizing brand identifiers more readily than older models (Stable Diffusion), the urgency of the unbranding challenge is starkly highlighted. Our results, validated by our VLM metric, confirm unbranding is a distinct, practically relevant problem requiring specialized techniques. Project Page: https://gmum.github.io/UNBRANDING/.

Paperid: 128, https://arxiv.org/pdf/2512.13660.pdf GitHub

Abstract:
Spatial tracing, as a fundamental embodied interaction ability for robots, is inherently challenging as it requires multi-step metric-grounded reasoning compounded with complex spatial referring and real-world metric measurement. However, existing methods struggle with this compositional task. To this end, we propose RoboTracer, a 3D-aware VLM that first achieves both 3D spatial referring and measuring via a universal spatial encoder and a regression-supervised decoder to enhance scale awareness during supervised fine-tuning (SFT). Moreover, RoboTracer advances multi-step metric-grounded reasoning via reinforcement fine-tuning (RFT) with metric-sensitive process rewards, supervising key intermediate perceptual cues to accurately generate spatial traces. To support SFT and RFT training, we introduce TraceSpatial, a large-scale dataset of 30M QA pairs, spanning outdoor/indoor/tabletop scenes and supporting complex reasoning processes (up to 9 steps). We further present TraceSpatial-Bench, a challenging benchmark filling the gap to evaluate spatial tracing. Experimental results show that RoboTracer surpasses baselines in spatial understanding, measuring, and referring, with an average success rate of 79.1%, and also achieves SOTA performance on TraceSpatial-Bench by a large margin, exceeding Gemini-2.5-Pro by 36% accuracy. Notably, RoboTracer can be integrated with various control policies to execute long-horizon, dynamic tasks across diverse robots (UR5, G1 humanoid) in cluttered real-world scenes.

Paperid: 129, https://arxiv.org/pdf/2512.13636.pdf GitHub

Abstract:
Current Vision-Language-Action (VLA) paradigms in autonomous driving primarily rely on Imitation Learning (IL), which introduces inherent challenges such as distribution shift and causal confusion. Online Reinforcement Learning offers a promising pathway to address these issues through trial-and-error learning. However, applying online reinforcement learning to VLA models in autonomous driving is hindered by inefficient exploration in continuous action spaces. To overcome this limitation, we propose MindDrive, a VLA framework comprising a large language model (LLM) with two distinct sets of LoRA parameters. The one LLM serves as a Decision Expert for scenario reasoning and driving decision-making, while the other acts as an Action Expert that dynamically maps linguistic decisions into feasible trajectories. By feeding trajectory-level rewards back into the reasoning space, MindDrive enables trial-and-error learning over a finite set of discrete linguistic driving decisions, instead of operating directly in a continuous action space. This approach effectively balances optimal decision-making in complex scenarios, human-like driving behavior, and efficient exploration in online reinforcement learning. Using the lightweight Qwen-0.5B LLM, MindDrive achieves Driving Score (DS) of 78.04 and Success Rate (SR) of 55.09% on the challenging Bench2Drive benchmark. To the best of our knowledge, this is the first work to demonstrate the effectiveness of online reinforcement learning for the VLA model in autonomous driving.

Paperid: 130, https://arxiv.org/pdf/2512.12799.pdf GitHub

Abstract:
Although multi-modal large language models (MLLMs) have shown strong capabilities across diverse domains, their application in generating fine-grained 3D perception and prediction outputs in autonomous driving remains underexplored. In this paper, we propose DrivePI, a novel spatial-aware 4D MLLM that serves as a unified Vision-Language-Action (VLA) framework that is also compatible with vision-action (VA) models. Our method jointly performs spatial understanding, 3D perception (i.e., 3D occupancy), prediction (i.e., occupancy flow), and planning (i.e., action outputs) in parallel through end-to-end optimization. To obtain both precise geometric information and rich visual appearance, our approach integrates point clouds, multi-view images, and language instructions within a unified MLLM architecture. We further develop a data engine to generate text-occupancy and text-flow QA pairs for 4D spatial understanding. Remarkably, with only a 0.5B Qwen2.5 model as MLLM backbone, DrivePI as a single unified model matches or exceeds both existing VLA models and specialized VA models. Specifically, compared to VLA models, DrivePI outperforms OpenDriveVLA-7B by 2.5% mean accuracy on nuScenes-QA and reduces collision rate by 70% over ORION (from 0.37% to 0.11%) on nuScenes. Against specialized VA models, DrivePI surpasses FB-OCC by 10.3 RayIoU for 3D occupancy on OpenOcc, reduces the mAVE from 0.591 to 0.509 for occupancy flow on OpenOcc, and achieves 32% lower L2 error than VAD (from 0.72m to 0.49m) for planning on nuScenes. Code will be available at https://github.com/happinesslz/DrivePI

Paperid: 131, https://arxiv.org/pdf/2512.12678.pdf GitHub

Abstract:
CLIP achieves strong zero-shot image-text retrieval by aligning global vision and text representations, yet it falls behind on fine-grained tasks even when fine-tuned on long, detailed captions. In this work, we propose $β$-CLIP, a multi-granular text-conditioned contrastive learning framework designed to achieve hierarchical alignment between multiple textual granularities-from full captions to sentences and phrases-and their corresponding visual regions. For each level of granularity, $β$-CLIP utilizes cross-attention to dynamically pool image patches, producing contextualized visual embeddings. To address the semantic overlap inherent in this hierarchy, we introduce the $β$-Contextualized Contrastive Alignment Loss ($β$-CAL). This objective parameterizes the trade-off between strict query-specific matching and relaxed intra-image contextualization, supporting both soft Cross-Entropy and hard Binary Cross-Entropy formulations. Through extensive experiments, we demonstrate that $β$-CLIP significantly improves dense alignment: achieving 91.8% T2I 92.3% I2T at R@1 on Urban1K and 30.9% on FG-OVD (Hard), setting state-of-the-art among methods trained without hard negatives. $β$-CLIP establishes a robust, adaptive baseline for dense vision-language correspondence. The code and models are released at https://github.com/fzohra/B-CLIP.

Paperid: 132, https://arxiv.org/pdf/2512.12069.pdf GitHub

Abstract:
Large Vision-Language Models (LVLMs) are vulnerable to a growing array of multimodal jailbreak attacks, necessitating defenses that are both generalizable to novel threats and efficient for practical deployment. Many current strategies fall short, either targeting specific attack patterns, which limits generalization, or imposing high computational overhead. While lightweight anomaly-detection methods offer a promising direction, we find that their common one-class design tends to confuse novel benign inputs with malicious ones, leading to unreliable over-rejection. To address this, we propose Representational Contrastive Scoring (RCS), a framework built on a key insight: the most potent safety signals reside within the LVLM's own internal representations. Our approach inspects the internal geometry of these representations, learning a lightweight projection to maximally separate benign and malicious inputs in safety-critical layers. This enables a simple yet powerful contrastive score that differentiates true malicious intent from mere novelty. Our instantiations, MCD (Mahalanobis Contrastive Detection) and KCD (K-nearest Contrastive Detection), achieve state-of-the-art performance on a challenging evaluation protocol designed to test generalization to unseen attack types. This work demonstrates that effective jailbreak detection can be achieved by applying simple, interpretable statistical methods to the appropriate internal representations, offering a practical path towards safer LVLM deployment. Our code is available on Github https://github.com/sarendis56/Jailbreak_Detection_RCS.

Paperid: 133, https://arxiv.org/pdf/2512.11982.pdf GitHub

Abstract:
Finding scientifically interesting phenomena through slow, manual labeling campaigns severely limits our ability to explore the billions of galaxy images produced by telescopes. In this work, we develop a pipeline to create a semantic search engine from completely unlabeled image data. Our method leverages Vision-Language Models (VLMs) to generate descriptions for galaxy images, then contrastively aligns a pre-trained multimodal astronomy foundation model with these embedded descriptions to produce searchable embeddings at scale. We find that current VLMs provide descriptions that are sufficiently informative to train a semantic search model that outperforms direct image similarity search. Our model, AION-Search, achieves state-of-the-art zero-shot performance on finding rare phenomena despite training on randomly selected images with no deliberate curation for rare cases. Furthermore, we introduce a VLM-based re-ranking method that nearly doubles the recall for our most challenging targets in the top-100 results. For the first time, AION-Search enables flexible semantic search scalable to 140 million galaxy images, enabling discovery from previously infeasible searches. More broadly, our work provides an approach for making large, unlabeled scientific image archives semantically searchable, expanding data exploration capabilities in fields from Earth observation to microscopy. The code, data, and app are publicly available at https://github.com/NolanKoblischke/AION-Search

Paperid: 134, https://arxiv.org/pdf/2512.11908.pdf GitHub GitHub

Abstract:
Contact-rich tasks pose significant challenges for robotic systems due to inherent uncertainty, complex dynamics, and the high risk of damage during interaction. Recent advances in learning-based control have shown great potential in enabling robots to acquire and generalize complex manipulation skills in such environments, but ensuring safety, both during exploration and execution, remains a critical bottleneck for reliable real-world deployment. This survey provides a comprehensive overview of safe learning-based methods for robot contact-rich tasks. We categorize existing approaches into two main domains: safe exploration and safe execution. We review key techniques, including constrained reinforcement learning, risk-sensitive optimization, uncertainty-aware modeling, control barrier functions, and model predictive safety shields, and highlight how these methods incorporate prior knowledge, task structure, and online adaptation to balance safety and efficiency. A particular emphasis of this survey is on how these safe learning principles extend to and interact with emerging robotic foundation models, especially vision-language models (VLMs) and vision-language-action models (VLAs), which unify perception, language, and control for contact-rich manipulation. We discuss both the new safety opportunities enabled by VLM/VLA-based methods, such as language-level specification of constraints and multimodal grounding of safety signals, and the amplified risks and evaluation challenges they introduce. Finally, we outline current limitations and promising future directions toward deploying reliable, safety-aligned, and foundation-model-enabled robots in complex contact-rich environments. More details and materials are available at our \href{ https://github.com/jack-sherman01/Awesome-Learning4Safe-Contact-rich-tasks}{Project GitHub Repository}.

Paperid: 135, https://arxiv.org/pdf/2512.11899.pdf GitHub

Abstract:
Large vision-language models (LVLMs) are vulnerable to typographic attacks, where misleading text within an image overrides visual understanding. Existing evaluation protocols and defenses, largely focused on object recognition, implicitly encourage ignoring text to achieve robustness; however, real-world scenarios often require joint reasoning over both objects and text (e.g., recognizing pedestrians while reading traffic signs). To address this, we introduce a novel task, Read-or-Ignore VQA (RIO-VQA), which formalizes selective text use in visual question answering (VQA): models must decide, from context, when to read text and when to ignore it. For evaluation, we present the Read-or-Ignore Benchmark (RIO-Bench), a standardized dataset and protocol that, for each real image, provides same-scene counterfactuals (read / ignore) by varying only the textual content and question type. Using RIO-Bench, we show that strong LVLMs and existing defenses fail to balance typographic robustness and text-reading capability, highlighting the need for improved approaches. Finally, RIO-Bench enables a novel data-driven defense that learns adaptive selective text use, moving beyond prior non-adaptive, text-ignoring defenses. Overall, this work reveals a fundamental misalignment between the existing evaluation scope and real-world requirements, providing a principled path toward reliable LVLMs. Our Project Page is at https://turingmotors.github.io/rio-vqa/.

Paperid: 136, https://arxiv.org/pdf/2512.11891.pdf GitHub

Abstract:
Vision-Language-Action (VLA) models have demonstrated remarkable capabilities in generalizing across diverse robotic manipulation tasks. However, deploying these models in unstructured environments remains challenging due to the critical need for simultaneous task compliance and safety assurance, particularly in preventing potential collisions during physical interactions. In this work, we introduce a Vision-Language-Safe Action (VLSA) architecture, named AEGIS, which contains a plug-and-play safety constraint (SC) layer formulated via control barrier functions. AEGIS integrates directly with existing VLA models to improve safety with theoretical guarantees, while maintaining their original instruction-following performance. To evaluate the efficacy of our architecture, we construct a comprehensive safety-critical benchmark SafeLIBERO, spanning distinct manipulation scenarios characterized by varying degrees of spatial complexity and obstacle intervention. Extensive experiments demonstrate the superiority of our method over state-of-the-art baselines. Notably, AEGIS achieves a 59.16% improvement in obstacle avoidance rate while substantially increasing the task execution success rate by 17.25%. To facilitate reproducibility and future research, we make our code, models, and the benchmark datasets publicly available at https://vlsa-aegis.github.io/.

Paperid: 137, https://arxiv.org/pdf/2512.11872.pdf GitHub

Abstract:
End-to-end autonomous driving systems based on vision-language-action (VLA) models integrate multimodal sensor inputs and language instructions to generate planning and control signals. While autoregressive large language models and continuous diffusion policies are prevalent, the potential of discrete masked diffusion for trajectory generation remains largely unexplored. This paper presents WAM-Diff, a VLA framework that employs masked diffusion to iteratively refine a discrete sequence representing future ego-trajectories. Our approach features three key innovations: a systematic adaptation of masked diffusion for autonomous driving that supports flexible, non-causal decoding orders; scalable model capacity via a sparse MoE architecture trained jointly on motion prediction and driving-oriented visual question answering (VQA); and online reinforcement learning using Group Sequence Policy Optimization (GSPO) to optimize sequence-level driving rewards. Remarkably, our model achieves 91.0 PDMS on NAVSIM-v1 and 89.7 EPDMS on NAVSIM-v2, demonstrating the effectiveness of masked diffusion for autonomous driving. The approach provides a promising alternative to autoregressive and diffusion-based policies, supporting scenario-aware decoding strategies for trajectory generation. The code for this paper will be released publicly at: https://github.com/fudan-generative-vision/WAM-Diff

Paperid: 138, https://arxiv.org/pdf/2512.11769.pdf GitHub

Abstract:
Vision-language-action (VLA) models enable impressive zero shot manipulation, but their inference stacks are often too heavy for responsive web demos or high frequency robot control on commodity GPUs. We present BLURR, a lightweight inference wrapper that can be plugged into existing VLA controllers without retraining or changing model checkpoints. Instantiated on the pi-zero VLA controller, BLURR keeps the original observation interfaces and accelerates control by combining an instruction prefix key value cache, mixed precision execution, and a single step rollout schedule that reduces per step computation. In our SimplerEnv based evaluation, BLURR maintains task success rates comparable to the original controller while significantly lowering effective FLOPs and wall clock latency. We also build an interactive web demo that allows users to switch between controllers and toggle inference options in real time while watching manipulation episodes. This highlights BLURR as a practical approach for deploying modern VLA policies under tight compute budgets.

Paperid: 139, https://arxiv.org/pdf/2512.11393.pdf GitHub

Abstract:
Humans can intuitively parallelise complex activities, but can a model learn this from observing a single person? Given one egocentric video, we introduce the N-Body Problem: how N individuals, can hypothetically perform the same set of tasks observed in this video. The goal is to maximise speed-up, but naive assignment of video segments to individuals often violates real-world constraints, leading to physically impossible scenarios like two people using the same object or occupying the same space. To address this, we formalise the N-Body Problem and propose a suite of metrics to evaluate both performance (speed-up, task coverage) and feasibility (spatial collisions, object conflicts and causal constraints). We then introduce a structured prompting strategy that guides a Vision-Language Model (VLM) to reason about the 3D environment, object usage, and temporal dependencies to produce a viable parallel execution. On 100 videos from EPIC-Kitchens and HD-EPIC, our method for N = 2 boosts action coverage by 45% over a baseline prompt for Gemini 2.5 Pro, while simultaneously slashing collision rates, object and causal conflicts by 55%, 45% and 55% respectively.

Paperid: 140, https://arxiv.org/pdf/2512.11362.pdf GitHub GitHub

Abstract:
Vision-Language-Action (VLA) models are driving a revolution in robotics, enabling machines to understand instructions and interact with the physical world. This field is exploding with new models and datasets, making it both exciting and challenging to keep pace with. This survey offers a clear and structured guide to the VLA landscape. We design it to follow the natural learning path of a researcher: we start with the basic Modules of any VLA model, trace the history through key Milestones, and then dive deep into the core Challenges that define recent research frontier. Our main contribution is a detailed breakdown of the five biggest challenges in: (1) Representation, (2) Execution, (3) Generalization, (4) Safety, and (5) Dataset and Evaluation. This structure mirrors the developmental roadmap of a generalist agent: establishing the fundamental perception-action loop, scaling capabilities across diverse embodiments and environments, and finally ensuring trustworthy deployment-all supported by the essential data infrastructure. For each of them, we review existing approaches and highlight future opportunities. We position this paper as both a foundational guide for newcomers and a strategic roadmap for experienced researchers, with the dual aim of accelerating learning and inspiring new ideas in embodied intelligence. A live version of this survey, with continuous updates, is maintained on our \href{https://suyuz1.github.io/VLA-Survey-Anatomy/}{project page}.

Paperid: 141, https://arxiv.org/pdf/2512.11218.pdf GitHub

Abstract:
The pursuit of out-of-distribution generalization in Vision-Language-Action (VLA) models is often hindered by catastrophic forgetting of the Vision-Language Model (VLM) backbone during fine-tuning. While co-training with external reasoning data helps, it requires experienced tuning and data-related overhead. Beyond such external dependencies, we identify an intrinsic cause within VLA datasets: modality imbalance, where language diversity is much lower than visual and action diversity. This imbalance biases the model toward visual shortcuts and language forgetting. To address this, we introduce BayesVLA, a Bayesian factorization that decomposes the policy into a visual-action prior, supporting seeing-to-act, and a language-conditioned likelihood, enabling prompt-to-specify. This inherently preserves generalization and promotes instruction following. We further incorporate pre- and post-contact phases to better leverage pre-trained foundation models. Information-theoretic analysis formally validates our effectiveness in mitigating shortcut learning. Extensive experiments show superior generalization to unseen instructions, objects, and environments compared to existing methods. Project page is available at: https://xukechun.github.io/papers/BayesVLA.

Paperid: 142, https://arxiv.org/pdf/2512.11061.pdf GitHub

Abstract:
Generative video models, a leading approach to world modeling, face fundamental limitations. They often violate physical and logical rules, lack interactivity, and operate as opaque black boxes ill-suited for building structured, queryable worlds. To overcome these challenges, we propose a new paradigm focused on distilling an image caption pair into a tractable, abstract representation optimized for simulation. We introduce VDAWorld, a framework where a Vision-Language Model (VLM) acts as an intelligent agent to orchestrate this process. The VLM autonomously constructs a grounded (2D or 3D) scene representation by selecting from a suite of vision tools, and accordingly chooses a compatible physics simulator (e.g., rigid body, fluid) to act upon it. VDAWorld can then infer latent dynamics from the static scene to predict plausible future states. Our experiments show that this combination of intelligent abstraction and adaptive simulation results in a versatile world model capable of producing high quality simulations across a wide range of dynamic scenarios.

Paperid: 143, https://arxiv.org/pdf/2512.10894.pdf GitHub

Abstract:
Recent vision-language model (VLM)-based approaches have achieved impressive results on SVG generation. However, because they generate only text and lack visual signals during decoding, they often struggle with complex semantics and fail to produce visually appealing or geometrically coherent SVGs. We introduce DuetSVG, a unified multimodal model that jointly generates image tokens and corresponding SVG tokens in an end-to-end manner. DuetSVG is trained on both image and SVG datasets. At inference, we apply a novel test-time scaling strategy that leverages the model's native visual predictions as guidance to improve SVG decoding quality. Extensive experiments show that our method outperforms existing methods, producing visually faithful, semantically aligned, and syntactically clean SVGs across a wide range of applications.

Paperid: 144, https://arxiv.org/pdf/2512.10619.pdf GitHub

Abstract:
Document parsing aims to transform unstructured PDF images into semi-structured data, facilitating the digitization and utilization of information in diverse domains. While vision language models (VLMs) have significantly advanced this task, achieving reliable, high-quality parsing in real-world scenarios remains challenging. Common practice often selects the top-performing model on standard benchmarks. However, these benchmarks may carry dataset-specific biases, leading to inconsistent model rankings and limited correlation with real-world performance. Moreover, benchmark metrics typically provide only overall scores, which can obscure distinct error patterns in output. This raises a key challenge: how can we reliably and comprehensively assess document parsing quality in the wild? We address this problem with DOCR-Inspector, which formalizes document parsing assessment as fine-grained error detection and analysis. Leveraging VLM-as-a-Judge, DOCR-Inspector analyzes a document image and its parsed output, identifies all errors, assigns them to one of 28 predefined types, and produces a comprehensive quality assessment. To enable this capability, we construct DOCRcase-200K for training and propose the Chain-of-Checklist reasoning paradigm to enable the hierarchical structure of parsing quality assessment. For empirical validation, we introduce DOCRcaseBench, a set of 882 real-world document parsing cases with manual annotations. On this benchmark, DOCR-Inspector-7B outperforms commercial models like Gemini 2.5 Pro, as well as leading open-source models. Further experiments demonstrate that its quality assessments provide valuable guidance for parsing results refinement, making DOCR-Inspector both a practical evaluator and a driver for advancing document parsing systems at scale. Model and code are released at: https://github.com/ZZZZZQT/DOCR-Inspector.

Paperid: 145, https://arxiv.org/pdf/2512.10384.pdf GitHub

Abstract:
Large Vision Language Models (LVLMs) have made remarkable progress, enabling sophisticated vision-language interaction and dialogue applications. However, existing benchmarks primarily focus on reasoning tasks, often neglecting fine-grained recognition, which is crucial for practical application scenarios. To address this gap, we introduce the Fine-grained Recognition Open World (FROW) benchmark, designed for detailed evaluation of LVLMs with GPT-4o. On the basis of that, we propose a novel optimization strategy from two perspectives: \textit{data construction} and \textit{training process}, to improve the performance of LVLMs. Our dataset includes mosaic data, which combines multiple short-answer responses, and open-world data, generated from real-world questions and answers using GPT-4o, creating a comprehensive framework for evaluating fine-grained recognition in LVLMs. Experiments show that mosaic data improves category recognition accuracy by 1\% and open-world data boosts FROW benchmark accuracy by 10\%-20\% and content accuracy by 6\%-12\%. Meanwhile, incorporating fine-grained data into the pre-training phase can improve the model's category recognition accuracy by up to 10\%. The benchmark will be available at https://github.com/pc-inno/FROW.

Paperid: 146, https://arxiv.org/pdf/2512.10342.pdf GitHub

Abstract:
Large-scale Vision-Language Models (VLMs) exhibit impressive complex reasoning capabilities but remain largely unexplored in visual sequential planning, i.e., executing multi-step actions towards a goal. Additionally, practical sequential planning often involves non-optimal (erroneous) steps, challenging VLMs to detect and correct such steps. We propose Corrective Sequential Planning Benchmark (CoSPlan) to evaluate VLMs in error-prone, vision-based sequential planning tasks across 4 domains: maze navigation, block rearrangement, image reconstruction,and object reorganization. CoSPlan assesses two key abilities: Error Detection (identifying non-optimal action) and Step Completion (correcting and completing action sequences to reach the goal). Despite using state-of-the-art reasoning techniques such as Chain-of-Thought and Scene Graphs, VLMs (e.g. Intern-VLM and Qwen2) struggle on CoSPlan, failing to leverage contextual cues to reach goals. Addressing this, we propose a novel training-free method, Scene Graph Incremental updates (SGI), which introduces intermediate reasoning steps between the initial and goal states. SGI helps VLMs reason about sequences, yielding an average performance gain of 5.2%. In addition to enhancing reliability in corrective sequential planning, SGI generalizes to traditional planning tasks such as Plan-Bench and VQA. Project Page : https://shroglck.github.io/cos_plan/

Paperid: 147, https://arxiv.org/pdf/2512.10244.pdf GitHub

Abstract:
Semi-supervised few-shot learning (SSFSL) formulates real-world applications like ''auto-annotation'', as it aims to learn a model over a few labeled and abundant unlabeled examples to annotate the unlabeled ones. Despite the availability of powerful open-source Vision-Language Models (VLMs) and their pretraining data, the SSFSL literature largely neglects these open-source resources. In contrast, the related area few-shot learning (FSL) has already exploited them to boost performance. Arguably, to achieve auto-annotation in the real world, SSFSL should leverage such open-source resources. To this end, we start by applying established SSL methods to finetune a VLM. Counterintuitively, they significantly underperform FSL baselines. Our in-depth analysis reveals the root cause: VLMs produce rather ''flat'' distributions of softmax probabilities. This results in zero utilization of unlabeled data and weak supervision signals. We address this issue with embarrassingly simple techniques: classifier initialization and temperature tuning. They jointly increase the confidence scores of pseudo-labels, improving the utilization rate of unlabeled data, and strengthening supervision signals. Building on this, we propose: Stage-Wise Finetuning with Temperature Tuning (SWIFT), which enables existing SSL methods to effectively finetune a VLM on limited labeled data, abundant unlabeled data, and task-relevant but noisy data retrieved from the VLM's pretraining set. Extensive experiments on five SSFSL benchmarks show that SWIFT outperforms recent FSL and SSL methods by $\sim$5 accuracy points. SWIFT even rivals supervised learning, which finetunes VLMs with the unlabeled data being labeled with ground truth!

Paperid: 148, https://arxiv.org/pdf/2512.09928.pdf GitHub

Abstract:
Vision-Language-Action (VLA) models have recently enabled robotic manipulation by grounding visual and linguistic cues into actions. However, most VLAs assume the Markov property, relying only on the current observation and thus suffering from temporal myopia that degrades long-horizon coherence. In this work, we view motion as a more compact and informative representation of temporal context and world dynamics, capturing inter-state changes while filtering static pixel-level noise. Building on this idea, we propose HiF-VLA (Hindsight, Insight, and Foresight for VLAs), a unified framework that leverages motion for bidirectional temporal reasoning. HiF-VLA encodes past dynamics through hindsight priors, anticipates future motion via foresight reasoning, and integrates both through a hindsight-modulated joint expert to enable a ''think-while-acting'' paradigm for long-horizon manipulation. As a result, HiF-VLA surpasses strong baselines on LIBERO-Long and CALVIN ABC-D benchmarks, while incurring negligible additional inference latency. Furthermore, HiF-VLA achieves substantial improvements in real-world long-horizon manipulation tasks, demonstrating its broad effectiveness in practical robotic settings.

Paperid: 149, https://arxiv.org/pdf/2512.09927.pdf GitHub GitHub

Abstract:
Vision-Language-Action (VLA) models pretrained on large-scale multimodal datasets have emerged as powerful foundations for robotic perception and control. However, their massive scale, often billions of parameters, poses significant challenges for real-time deployment, as inference becomes computationally expensive and latency-sensitive in dynamic environments. To address this, we propose Token Expand-and-Merge-VLA (TEAM-VLA), a training-free token compression framework that accelerates VLA inference while preserving task performance. TEAM-VLA introduces a dynamic token expansion mechanism that identifies and samples additional informative tokens in the spatial vicinity of attention-highlighted regions, enhancing contextual completeness. These expanded tokens are then selectively merged in deeper layers under action-aware guidance, effectively reducing redundancy while maintaining semantic coherence. By coupling expansion and merging within a single feed-forward pass, TEAM-VLA achieves a balanced trade-off between efficiency and effectiveness, without any retraining or parameter updates. Extensive experiments on LIBERO benchmark demonstrate that TEAM-VLA consistently improves inference speed while maintaining or even surpassing the task success rate of full VLA models. The code is public available on \href{https://github.com/Jasper-aaa/TEAM-VLA}{https://github.com/Jasper-aaa/TEAM-VLA}

Paperid: 150, https://arxiv.org/pdf/2512.09924.pdf GitHub

Abstract:
Video unified models exhibit strong capabilities in understanding and generation, yet they struggle with reason-informed visual editing even when equipped with powerful internal vision-language models (VLMs). We attribute this gap to two factors: 1) existing datasets are inadequate for training and evaluating reasoning-aware video editing, and 2) an inherent disconnect between the models' reasoning and editing capabilities, which prevents the rich understanding from effectively instructing the editing process. Bridging this gap requires an integrated framework that connects reasoning with visual transformation. To address this gap, we introduce the Reason-Informed Video Editing (RVE) task, which requires reasoning about physical plausibility and causal dynamics during editing. To support systematic evaluation, we construct RVE-Bench, a comprehensive benchmark with two complementary subsets: Reasoning-Informed Video Editing and In-Context Video Generation. These subsets cover diverse reasoning dimensions and real-world editing scenarios. Building upon this foundation, we propose the ReViSE, a Self-Reflective Reasoning (SRF) framework that unifies generation and evaluation within a single architecture. The model's internal VLM provides intrinsic feedback by assessing whether the edited video logically satisfies the given instruction. The differential feedback that refines the generator's reasoning behavior during training. Extensive experiments on RVE-Bench demonstrate that ReViSE significantly enhances editing accuracy and visual fidelity, achieving a 32% improvement of the Overall score in the reasoning-informed video editing subset over state-of-the-art methods.

Paperid: 151, https://arxiv.org/pdf/2512.09874.pdf GitHub

Abstract:
Correctly parsing mathematical formulas from PDFs is critical for training large language models and building scientific knowledge bases from academic literature, yet existing benchmarks either exclude formulas entirely or lack semantically-aware evaluation metrics. We introduce a novel benchmarking framework centered on synthetically generated PDFs with precise LaTeX ground truth, enabling systematic control over layout, formulas, and content characteristics. A key methodological contribution is pioneering LLM-as-a-judge for semantic formula assessment, combined with a robust two-stage matching pipeline that handles parser output inconsistencies. Through human validation on 250 formula pairs (750 ratings from 30 evaluators), we demonstrate that LLM-based evaluation achieves substantially higher correlation with human judgment (Pearson r=0.78) compared to CDM (r=0.34) and text similarity (r~0). Evaluating 20+ contemporary PDF parsers (including specialized OCR models, vision-language models, and rule-based approaches) across 100 synthetic documents with 2,000+ formulas reveals significant performance disparities. Our findings provide crucial insights for practitioners selecting parsers for downstream applications and establish a robust, scalable methodology that enables reproducible evaluation of PDF formula extraction quality. Code and benchmark data: https://github.com/phorn1/pdf-parse-bench

Paperid: 152, https://arxiv.org/pdf/2512.09841.pdf GitHub

Abstract:
Time awareness is a fundamental ability of omni large language models, especially for understanding long videos and answering complex questions. Previous approaches mainly target vision-language scenarios and focus on the explicit temporal grounding questions, such as identifying when a visual event occurs or determining what event happens at aspecific time. However, they often make insufficient use of the audio modality, and overlook implicit temporal grounding across modalities--for example, identifying what is visually present when a character speaks, or determining what is said when a visual event occurs--despite such cross-modal temporal relations being prevalent in real-world scenarios. In this paper, we propose ChronusOmni, an omni large language model designed to enhance temporal awareness for both explicit and implicit audiovisual temporal grounding. First, we interleave text-based timestamp tokens with visual and audio representations at each time unit, enabling unified temporal modeling across modalities. Second, to enforce correct temporal ordering and strengthen fine-grained temporal reasoning, we incorporate reinforcement learning with specially designed reward functions. Moreover, we construct ChronusAV, a temporally-accurate, modality-complete, and cross-modal-aligned dataset to support the training and evaluation on audiovisual temporal grounding task. Experimental results demonstrate that ChronusOmni achieves state-of-the-art performance on ChronusAV with more than 30% improvement and top results on most metrics upon other temporal grounding benchmarks. This highlights the strong temporal awareness of our model across modalities, while preserving general video and audio understanding capabilities.

Paperid: 153, https://arxiv.org/pdf/2512.09092.pdf GitHub

Abstract:
Underground mining disasters produce pervasive darkness, dust, and collapses that obscure vision and make situational awareness difficult for humans and conventional systems. To address this, we propose MDSE, Multimodal Disaster Situation Explainer, a novel vision-language framework that automatically generates detailed textual explanations of post-disaster underground scenes. MDSE has three-fold innovations: (i) Context-Aware Cross-Attention for robust alignment of visual and textual features even under severe degradation; (ii) Segmentation-aware dual pathway visual encoding that fuses global and region-specific embeddings; and (iii) Resource-Efficient Transformer-Based Language Model for expressive caption generation with minimal compute cost. To support this task, we present the Underground Mine Disaster (UMD) dataset--the first image-caption corpus of real underground disaster scenes--enabling rigorous training and evaluation. Extensive experiments on UMD and related benchmarks show that MDSE substantially outperforms state-of-the-art captioning models, producing more accurate and contextually relevant descriptions that capture crucial details in obscured environments, improving situational awareness for underground emergency response. The code is at https://github.com/mizanJewel/Multimodal-Disaster-Situation-Explainer.

Paperid: 154, https://arxiv.org/pdf/2512.08860.pdf GitHub

Abstract:
Verifiable geometric reasoning is a critical component for trustworthy and controllable agentic AI. Despite impressive capabilities, Vision-Language Models (VLMs) often fail under realistic scene changes. We present Tri-Bench, a compact benchmark of planar triangle problems that isolates relative geometric reasoning while stressing two deployment-critical factors: camera pose (planar vs. tilted) and scene context via object interference (10 everyday objects). To test verifiability and control, we evaluate four recent VLMs using a single, fixed prompt whose guardrail explicitly describes a surrounding square border, enabling correct answers via homography. We evaluate six simple tasks over binary and continuous targets, and observe that the overall accuracy with respect to 3D ground truth is modest, ~69% on average (best ~75%, worst ~64%). The same responses align even more closely with 2D projections in the image plane, where mean accuracy is ~72%. All four VLMs consistently fail, with accuracy falling to ~0%, on recognizing minority shape classes (equilateral, isosceles, right-angled triangles). Additionally, overall VLM accuracy degrades by ~4.1% under camera tilt. This demonstrates that models fail to correctly utilize the explicit frame-of-reference hint provided in the prompt and default to 2D image plane cues. Finally, we find that object interference has no significant effect on VLM accuracy.

Paperid: 155, https://arxiv.org/pdf/2512.08829.pdf GitHub

Abstract:
Window attention and linear attention represent two principal strategies for mitigating the quadratic complexity and ever-growing KV cache in Vision-Language Models (VLMs). However, we observe that window-based VLMs suffer performance degradation when sequence length exceeds the window size, while linear attention underperforms on information-intensive tasks such as OCR and document understanding. To overcome these limitations, we propose InfiniteVL, a linear-complexity VLM architecture that synergizes sliding window attention (SWA) with Gated DeltaNet. For achieving competitive multimodal performance under constrained resources, we design a three-stage training strategy comprising distillation pretraining, instruction tuning, and long-sequence SFT. Remarkably, using less than 2\% of the training data required by leading VLMs, InfiniteVL not only substantially outperforms previous linear-complexity VLMs but also matches the performance of leading Transformer-based VLMs, while demonstrating effective long-term memory retention. Compared to similar-sized Transformer-based VLMs accelerated by FlashAttention-2, InfiniteVL achieves over 3.6\times inference speedup while maintaining constant latency and memory footprint. In streaming video understanding scenarios, it sustains a stable 24 FPS real-time prefill speed while preserving long-term memory cache. Code and models are available at https://github.com/hustvl/InfiniteVL.

Paperid: 156, https://arxiv.org/pdf/2512.08233.pdf GitHub

Abstract:
Humans interpret safety not as a binary signal but as a continuous, context- and spatially-dependent notion of risk. While risk is subjective, humans form rational mental models that guide action selection in dynamic environments. This work proposes a framework for extracting implicit human risk models by introducing a novel, semantically-conditioned and spatially-varying parametrization of risk, supervised directly from safe human demonstration videos and VLM common sense. Notably, we define risk through a Bayesian formulation. The prior is furnished by a pretrained vision-language model. In order to encourage the risk estimate to be more human aligned, a likelihood function modulates the prior to produce a relative metric of risk. Specifically, the likelihood is a learned ViT that maps pretrained features, to pixel-aligned risk values. Our pipeline ingests RGB images and a query object string, producing pixel-dense risk images. These images that can then be used as value-predictors in robot planning tasks or be projected into 3D for use in conventional trajectory optimization to produce human-like motion. This learned mapping enables generalization to novel objects and contexts, and has the potential to scale to much larger training datasets. In particular, the Bayesian framework that is introduced enables fast adaptation of our model to additional observations or common sense rules. We demonstrate that our proposed framework produces contextual risk that aligns with human preferences. Additionally, we illustrate several downstream applications of the model; as a value learner for visuomotor planners or in conjunction with a classical trajectory optimization algorithm. Our results suggest that our framework is a significant step toward enabling autonomous systems to internalize human-like risk. Code and results can be found at https://riskbayesian.github.io/bayesian_risk/.

Paperid: 157, https://arxiv.org/pdf/2512.07833.pdf GitHub

Abstract:
Humans do not just see attribute similarity -- we also see relational similarity. An apple is like a peach because both are reddish fruit, but the Earth is also like a peach: its crust, mantle, and core correspond to the peach's skin, flesh, and pit. This ability to perceive and recognize relational similarity, is arguable by cognitive scientist to be what distinguishes humans from other species. Yet, all widely used visual similarity metrics today (e.g., LPIPS, CLIP, DINO) focus solely on perceptual attribute similarity and fail to capture the rich, often surprising relational similarities that humans perceive. How can we go beyond the visible content of an image to capture its relational properties? How can we bring images with the same relational logic closer together in representation space? To answer these questions, we first formulate relational image similarity as a measurable problem: two images are relationally similar when their internal relations or functions among visual elements correspond, even if their visual attributes differ. We then curate 114k image-caption dataset in which the captions are anonymized -- describing the underlying relational logic of the scene rather than its surface content. Using this dataset, we finetune a Vision-Language model to measure the relational similarity between images. This model serves as the first step toward connecting images by their underlying relational structure rather than their visible appearance. Our study shows that while relational similarity has a lot of real-world applications, existing image similarity models fail to capture it -- revealing a critical gap in visual computing.

Paperid: 158, https://arxiv.org/pdf/2512.07687.pdf GitHub

Abstract:
Multimodal Large Language Models (MLLMs) have demonstrated remarkable capabilities in vision-language understanding tasks. While these models often produce linguistically coherent output, they often suffer from hallucinations, generating descriptions that are factually inconsistent with the visual content, potentially leading to adverse consequences. Therefore, the assessment of hallucinations in MLLM has become increasingly crucial in the model development process. Contemporary methodologies predominantly depend on external LLM evaluators, which are themselves susceptible to hallucinations and may present challenges in terms of domain adaptation. In this study, we propose the hypothesis that hallucination manifests as measurable irregularities within the internal layer dynamics of MLLMs, not merely due to distributional shifts but also in the context of layer-wise analysis of specific assumptions. By incorporating such modifications, \textsc{\textsc{HalluShift++}} broadens the efficacy of hallucination detection from text-based large language models (LLMs) to encompass multimodal scenarios. Our codebase is available at https://github.com/C0mRD/HalluShift_Plus.

Paperid: 159, https://arxiv.org/pdf/2512.07564.pdf GitHub

Abstract:
Vision-language models (VLMs) frequently generate hallucinated content plausible but incorrect claims about image content. We propose a training-free self-correction framework enabling VLMs to iteratively refine responses through uncertainty-guided visual re-attention. Our method combines multidimensional uncertainty quantification (token entropy, attention dispersion, semantic consistency, claim confidence) with attention-guided cropping of under-explored regions. Operating entirely with frozen, pretrained VLMs, our framework requires no gradient updates. We validate our approach on the POPE and MMHAL BENCH benchmarks using the Qwen2.5-VL-7B [23] architecture. Experimental results demonstrate that our method reduces hallucination rates by 9.8 percentage points compared to the baseline, while improving object existence accuracy by 4.7 points on adversarial splits. Furthermore, qualitative analysis confirms that uncertainty-guided re-attention successfully grounds corrections in visual evidence where standard decoding fails. We validate our approach on Qwen2.5-VL-7B [23], with plans to extend validation across diverse architectures in future versions. We release our code and methodology to facilitate future research in trustworthy multimodal systems.

Paperid: 160, https://arxiv.org/pdf/2512.07302.pdf GitHub

Abstract:
Existing image perception methods based on VLMs generally follow a paradigm wherein models extract and analyze image content based on user-provided textual task prompts. However, such methods face limitations when applied to UAV imagery, which presents challenges like target confusion, scale variations, and complex backgrounds. These challenges arise because VLMs' understanding of image content depends on the semantic alignment between visual and textual tokens. When the task prompt is simplistic and the image content is complex, achieving effective alignment becomes difficult, limiting the model's ability to focus on task-relevant information. To address this issue, we introduce AerialVP, the first agent framework for task prompt enhancement in UAV image perception. AerialVP proactively extracts multi-dimensional auxiliary information from UAV images to enhance task prompts, overcoming the limitations of traditional VLM-based approaches. Specifically, the enhancement process includes three stages: (1) analyzing the task prompt to identify the task type and enhancement needs, (2) selecting appropriate tools from the tool repository, and (3) generating enhanced task prompts based on the analysis and selected tools. To evaluate AerialVP, we introduce AerialSense, a comprehensive benchmark for UAV image perception that includes Aerial Visual Reasoning, Aerial Visual Question Answering, and Aerial Visual Grounding tasks. AerialSense provides a standardized basis for evaluating model generalization and performance across diverse resolutions, lighting conditions, and both urban and natural scenes. Experimental results demonstrate that AerialVP significantly enhances task prompt guidance, leading to stable and substantial performance improvements in both open-source and proprietary VLMs. Our work will be available at https://github.com/lostwolves/AerialVP.

Paperid: 161, https://arxiv.org/pdf/2512.07276.pdf GitHub

Abstract:
Three-dimensional geospatial analysis is critical to applications in urban planning, climate adaptation, and environmental assessment. Current methodologies depend on costly, specialized sensors (e.g., LiDAR and multispectral), which restrict global accessibility. Existing sensor-based and rule-driven methods further struggle with tasks requiring the integration of multiple 3D cues, handling diverse queries, and providing interpretable reasoning. We hereby present Geo3DVQA, a comprehensive benchmark for evaluating vision-language models (VLMs) in height-aware, 3D geospatial reasoning using RGB-only remote sensing imagery. Unlike conventional sensor-based frameworks, Geo3DVQA emphasizes realistic scenarios that integrate elevation, sky view factors, and land cover patterns. The benchmark encompasses 110k curated question-answer pairs spanning 16 task categories across three complexity levels: single-feature inference, multi-feature reasoning, and application-level spatial analysis. The evaluation of ten state-of-the-art VLMs highlights the difficulty of RGB-to-3D reasoning. GPT-4o and Gemini-2.5-Flash achieved only 28.6% and 33.0% accuracy respectively, while domain-specific fine-tuning of Qwen2.5-VL-7B achieved 49.6% (+24.8 points). These results reveal both the limitations of current VLMs and the effectiveness of domain adaptation. Geo3DVQA introduces new challenge frontiers for scalable, accessible, and holistic 3D geospatial analysis. The dataset and code will be released upon publication at https://github.com/mm1129/Geo3DVQA.

Paperid: 162, https://arxiv.org/pdf/2512.07222.pdf GitHub

Abstract:
To address the trade-off between robustness and performance for robust VLM, we observe that function words could incur vulnerability of VLMs against cross-modal adversarial attacks, and propose Function-word De-Attention (FDA) accordingly to mitigate the impact of function words. Similar to differential amplifiers, our FDA calculates the original and the function-word cross-attention within attention heads, and differentially subtracts the latter from the former for more aligned and robust VLMs. Comprehensive experiments include 2 SOTA baselines under 6 different attacks on 2 downstream tasks, 3 datasets, and 3 models. Overall, our FDA yields an average 18/13/53% ASR drop with only 0.2/0.3/0.6% performance drops on the 3 tested models on retrieval, and a 90% ASR drop with a 0.3% performance gain on visual grounding. We demonstrate the scalability, generalization, and zero-shot performance of FDA experimentally, as well as in-depth ablation studies and analysis. Code will be made publicly at https://github.com/michaeltian108/FDA.

Paperid: 163, https://arxiv.org/pdf/2512.07155.pdf GitHub

Abstract:
Diffusion models exhibit remarkable generative ability, yet achieving smooth and semantically consistent image morphing remains a challenge. Existing approaches often yield abrupt transitions or over-saturated appearances due to the lack of adaptive structural and semantic alignments. We propose CHIMERA, a zero-shot diffusion-based framework that formulates morphing as a cached inversion-guided denoising process. To handle large semantic and appearance disparities, we propose Adaptive Cache Injection and Semantic Anchor Prompting. Adaptive Cache Injection (ACI) caches down, mid, and up blocks features from both inputs during DDIM inversion and re-injects them adaptively during denoising, enabling spatial and semantic alignment in depth- and time-adaptive manners and enabling natural feature fusion and smooth transitions. Semantic Anchor Prompting (SAP) leverages a vision-language model to generate a shared anchor prompt that serves as a semantic anchor, bridging dissimilar inputs and guiding the denoising process toward coherent results. Finally, we introduce the Global-Local Consistency Score (GLCS), a morphing-oriented metric that simultaneously evaluates the global harmonization of the two inputs and the smoothness of the local morphing transition. Extensive experiments and user studies show that CHIMERA achieves smoother and more semantically aligned transitions than existing methods, establishing a new state of the art in image morphing. The code and project page will be publicly released.

Paperid: 164, https://arxiv.org/pdf/2512.07141.pdf GitHub

Abstract:
As multimodal reasoning improves the overall capabilities of Large Vision Language Models (LVLMs), recent studies have begun to explore safety-oriented reasoning, aiming to enhance safety awareness by analyzing potential safety risks during the reasoning process before generating the final response. Although such approaches improve safety awareness and interpretability, this single-pass think-then-answer paradigm remains vulnerable to contextual or visual jailbreak attacks. This reveals a critical flaw: single-pass reasoning may overlook explicit harmful content in its own output. Our key insight is to exploit this wasted signal through reflection, which can effectively leverage the malicious content revealed in the first-pass reasoning to enable genuine self-correction and prevent unsafe generations. Motivated by this, we propose Think-Reflect-Revise (TRR), a three-stage training framework designed to enhance the safety alignment of LVLMs through policy-guided self-reflection. We first build a Reflective Safety Reasoning (ReSafe) dataset with 5,000 examples that follow a think-reflect-revise process. We then fine-tune the target model using the ReSafe dataset to initialize reflective behavior, and finally reinforce policy-guided reflection through reinforcement learning. Experimental results show that TRR substantially improves the safety performance of LVLMs across both safety-awareness benchmarks and jailbreak attack evaluations, increasing the overall safe response rate from 42.8% to 87.7% on Qwen2.5-VL-7B, while preserving stable performance on general benchmarks such as MMMU and MMStar. The project page is available at https://think-reflect-revise.github.io/.

Paperid: 165, https://arxiv.org/pdf/2512.07136.pdf GitHub GitHub

Abstract:
Multimodal human action recognition (HAR) leverages complementary sensors for activity classification. Beyond recognition, recent advances in large language models (LLMs) enable detailed descriptions and causal reasoning, motivating new tasks: human action understanding (HAU) and human action reasoning (HARn). However, most LLMs, especially large vision language models (LVLMs), struggle with non-RGB modalities such as depth, IMU, and mmWave due to the lack of large-scale data-caption resources. Existing HAR datasets mainly provide coarse data-label annotations, which are insufficient to capture fine-grained action dynamics needed for HAU and HARn. We consider two ground-truth pair types: (1) data label (discrete category) and (2) data caption (textual description). Naively generating captions from labels often lacks logical and spatiotemporal consistency. We introduce CUHK-X, a large-scale multimodal dataset and benchmark suite for HAR, HAU, and HARn. CUHK-X contains 58,445 samples covering 40 actions performed by 30 participants across two indoor environments. To improve caption consistency, we propose a prompt-based scene creation method that leverages LLMs to generate logically connected activity sequences, followed by human validation. CUHK-X includes three benchmarks with six evaluation tasks. Experiments report average accuracies of 76.52% (HAR), 40.76% (HAU), and 70.25% (HARn). CUHK-X aims to enable the community to apply and develop data-intensive learning methods for robust, multimodal human activity analysis. Project page and code: https://openaiotlab.github.io/CUHK-X/ and https://github.com/openaiotlab/CUHK-X.

Paperid: 166, https://arxiv.org/pdf/2512.07132.pdf GitHub

Abstract:
Specialized visual tools can augment large language models or vision language models with expert knowledge (e.g., grounding, spatial reasoning, medical knowledge, etc.), but knowing which tools to call (and when to call them) can be challenging. We introduce DART, a multi-agent framework that uses disagreements between multiple debating visual agents to identify useful visual tools (e.g., object detection, OCR, spatial reasoning, etc.) that can resolve inter-agent disagreement. These tools allow for fruitful multi-agent discussion by introducing new information, and by providing tool-aligned agreement scores that highlight agents in agreement with expert tools, thereby facilitating discussion. We utilize an aggregator agent to select the best answer by providing the agent outputs and tool information. We test DART on four diverse benchmarks and show that our approach improves over multi-agent debate as well as over single agent tool-calling frameworks, beating the next-strongest baseline (multi-agent debate with a judge model) by 3.4% and 2.4% on A-OKVQA and MMMU respectively. We also find that DART adapts well to new tools in applied domains, with a 1.3% improvement on the M3D medical dataset over other strong tool-calling, single agent, and multi-agent baselines. Additionally, we measure text overlap across rounds to highlight the rich discussion in DART compared to existing multi-agent methods. Finally, we study the tool call distribution, finding that diverse tools are reliably used to help resolve disagreement.

Paperid: 167, https://arxiv.org/pdf/2512.06883.pdf GitHub

Abstract:
Multimodal recommendation enhances accuracy by leveraging visual and textual signals, and its success largely depends on learning high-quality cross-modal representations. Recent advances in Large Vision-Language Models (LVLMs) offer unified multimodal representation learning, making them a promising backbone. However, applying LVLMs to recommendation remains challenging due to (i) representation misalignment, where domain gaps between item data and general pre-training lead to unaligned embedding spaces, and (ii) gradient conflicts during fine-tuning, where shared adapters cause interference and a lack of discriminative power. To address this, we propose SDA, a lightweight framework for Structural and Disentangled Adaptation, which integrates two components: Cross-Modal Structural Alignment (CMSA) and Modality-Disentangled Adaptation. CMSA aligns embeddings using intra-modal structures as a soft teacher, while MoDA mitigates gradient conflicts via expertized, gated low-rank paths to disentangle gradient flows. Experiments on three public Amazon datasets show SDA integrates seamlessly with existing multimodal and sequential recommenders, yielding average gains of 6.15% in Hit@10 and 8.64% in NDCG@10. It also achieves up to 12.83% and 18.70% gains on long-tail items with minimal inference overhead. Our code and full experimental results are available at https://github.com/RaoZhongtao/SDA.

Paperid: 168, https://arxiv.org/pdf/2512.06750.pdf GitHub

Abstract:
Image quality assessment (IQA) and image restoration are fundamental problems in low-level vision. Although IQA and restoration are closely connected conceptually, most existing work treats them in isolation. Recent advances in unified multimodal understanding-generation models demonstrate promising results and indicate that stronger understanding can improve generative performance. This motivates a single model that unifies IQA and restoration and explicitly studies how IQA can guide restoration, a setting that remains largely underexplored yet highly valuable. In this paper, we propose UARE, to our knowledge the first Unified vision-language model for image quality Assessment, Restoration, and Enhancement. Built on pretrained unified understanding and generation models, we introduce a two-stage training framework. First, a progressive, easy-to-hard schedule expands from single-type distortions to higher-order mixed degradations, enabling UARE to handle multiple degradations. Second, we perform unified fine-tuning of quality understanding and restoration with interleaved text-image data, aligning IQA signals with restoration objectives. Through multi-task co-training, UARE leverages IQA to boost restoration and enhancement performance. Extensive experiments across IQA, restoration, and enhancement tasks demonstrate the effectiveness of UARE. The code and models will be available at https://github.com/lwq20020127/UARE.

Paperid: 169, https://arxiv.org/pdf/2512.06581.pdf GitHub

Abstract:
Large vision-language models struggle with medical video understanding, where spatial precision, temporal reasoning, and clinical semantics are critical. To address this, we first introduce \textbf{MedVidBench}, a large-scale benchmark of 531,850 video-instruction pairs across 8 medical sources spanning video, segment, and frame-level tasks, curated through a rigorous quality assurance pipeline with expert-guided prompting and dual-model validation. While supervised fine-tuning on MedVidBench yields noticeable gains, standard Reinforcement Learning (RL) fails due to imbalanced reward scales across datasets, which destabilizes optimization and leads to training collapse. To overcome this, we introduce \textbf{MedGRPO}, a novel RL framework for balanced multi-dataset training with two key innovations: (1) \emph{cross-dataset reward normalization} that maps each dataset's median performance to a common reward value, ensuring fair optimization regardless of difficulty, and (2) a \emph{medical LLM judge} that evaluates caption quality on five clinical dimensions through comparative similarity scoring. Supervised fine-tuning Qwen2.5-VL-7B on MedVidBench substantially outperforms GPT-4.1 and Gemini-2.5-Flash across all tasks, demonstrating MedVidBench's efficacy, while our MedGRPO framework further improves upon the SFT baseline across grounding and captioning tasks. Our work establishes a foundational benchmark and robust training methodology for advancing vision-language models in medical domains. Our project website is available at https://yuhaosu.github.io/MedGRPO/.

Paperid: 170, https://arxiv.org/pdf/2512.06258.pdf GitHub

Abstract:
We reveal a critical yet underexplored flaw in Large Vision-Language Models (LVLMs): even when these models know the correct answer, they frequently arrive there through incorrect reasoning paths. The core issue is not a lack of knowledge, but a path selection bias within the vast reasoning search space. Although LVLMs are often capable of sampling correct solution trajectories, they disproportionately favor unstable or logically inconsistent ones, leading to erratic and unreliable outcomes. The substantial disparity between Pass@K (with large K) and Pass@1 across numerous models provides compelling evidence that such failures primarily stem from misreasoning rather than ignorance. To systematically investigate and address this issue, we propose PSO (Path-Select Optimization), a two-stage post-training framework designed to enhance both the reasoning performance and stability of existing LVLMs. In the first stage, we employ Group Relative Policy Optimization (GRPO) with template and answer-based rewards to cultivate structured, step-by-step reasoning. In the second stage, we conduct online preference optimization, where the model samples reasoning paths from GRPO-generated data, self-evaluates them, and aligns itself toward the preferred trajectories. Incorrect or suboptimal paths are concurrently stored in a Negative Replay Memory (NRM) as hard negatives, which are periodically revisited to prevent the model from repeating prior mistakes and to facilitate continual reasoning refinement. Extensive experiments show that PSO effectively prunes invalid reasoning paths, substantially enhances reasoning accuracy (with 7.4% improvements on average), and yields more stable and consistent chains of thought. Our code will be available at https://github.com/aiming-lab/PSO.

Paperid: 171, https://arxiv.org/pdf/2512.05941.pdf GitHub

Abstract:
Grounding is a fundamental capability for building graphical user interface (GUI) agents. Although existing approaches rely on large-scale bounding box supervision, they still face various challenges, such as cross-platform generalization, complex layout analysis, and fine-grained element localization. In this paper, we investigate zoom as a strong yet underexplored prior for GUI grounding, and propose a training-free method, ZoomClick. By characterizing four key properties of zoom (i.e., pre-zoom, depth, shrink size, minimal crop size), we unlock its full capabilities for dynamic spatial focusing and adaptive context switching. Experiments demonstrate that our method significantly boosts the performance of both general vision-language and specialized GUI grounding models, achieving state-of-the-art results on several mainstream benchmarks; for example, UI-Venus-72B attains a 73.1% success rate on ScreenSpot-Pro. Furthermore, we present GUIZoom-Bench, a benchmark for evaluating model adaptability to zoom, aiming to inspire future research on improving zoom for further training and test-time scaling in GUI grounding tasks.

Paperid: 172, https://arxiv.org/pdf/2512.05809.pdf GitHub

Abstract:
Vision-Language Models (VLMs) remain limited in spatial reasoning tasks that require multi-view understanding and embodied perspective shifts. Recent approaches such as MindJourney attempt to mitigate this gap through test-time scaling where a world model imagines action-conditioned trajectories and a heuristic verifier selects helpful views from such trajectories. In this work, we systematically examine how such test-time verifiers behave across benchmarks, uncovering both their promise and their pitfalls. Our uncertainty-based analyses show that MindJourney's verifier provides little meaningful calibration, and that random scoring often reduces answer entropy equally well, thus exposing systematic action biases and unreliable reward signals. To mitigate these, we introduce a Verification through Spatial Assertions (ViSA) framework that grounds the test-time reward in verifiable, frame-anchored micro-claims. This principled verifier consistently improves spatial reasoning on the SAT-Real benchmark and corrects trajectory-selection biases through more balanced exploratory behavior. However, on the challenging MMSI-Bench, none of the verifiers, including ours, achieve consistent scaling, suggesting that the current world models form an information bottleneck where imagined views fail to enrich fine-grained reasoning. Together, these findings chart the bad, good, and ugly aspects of test-time verification for world-model-based reasoning. Our code is available at https://github.com/chandar-lab/visa-for-mindjourney.

Paperid: 173, https://arxiv.org/pdf/2512.05693.pdf GitHub

Abstract:
The development of foundation models for embodied intelligence critically depends on access to large-scale, high-quality robot demonstration data. Recent approaches have sought to address this challenge by training on large collections of heterogeneous robotic datasets. However, unlike vision or language data, robotic demonstrations exhibit substantial heterogeneity across embodiments and action spaces as well as other prominent variations such as senor configurations and action control frequencies. The lack of explicit designs for handling such heterogeneity causes existing methods to struggle with integrating diverse factors, thereby limiting their generalization and leading to degraded performance when transferred to new settings. In this paper, we present HiMoE-VLA, a novel vision-language-action (VLA) framework tailored to effectively handle diverse robotic data with heterogeneity. Specifically, we introduce a Hierarchical Mixture-of-Experts (HiMoE) architecture for the action module which adaptively handles multiple sources of heterogeneity across layers and gradually abstracts them into shared knowledge representations. Through extensive experimentation with simulation benchmarks and real-world robotic platforms, HiMoE-VLA demonstrates a consistent performance boost over existing VLA baselines, achieving higher accuracy and robust generalization across diverse robots and action spaces. The code and models are publicly available at https://github.com/ZhiyingDu/HiMoE-VLA.

Paperid: 174, https://arxiv.org/pdf/2512.05524.pdf GitHub

Abstract:
Spatio-temporal scene graph generation (ST-SGG) aims to model objects and their evolving relationships across video frames, enabling interpretable representations for downstream reasoning tasks such as video captioning and visual question answering. Despite recent advancements in DETR-style single-stage ST-SGG models, they still suffer from several key limitations. First, while these models rely on attention-based learnable queries as a core component, these learnable queries are semantically uninformed and instance-agnostically initialized. Second, these models rely exclusively on unimodal visual features for predicate classification. To address these challenges, we propose VOST-SGG, a VLM-aided one-stage ST-SGG framework that integrates the common sense reasoning capabilities of vision-language models (VLMs) into the ST-SGG pipeline. First, we introduce the dual-source query initialization strategy that disentangles what to attend to from where to attend, enabling semantically grounded what-where reasoning. Furthermore, we propose a multi-modal feature bank that fuses visual, textual, and spatial cues derived from VLMs for improved predicate classification. Extensive experiments on the Action Genome dataset demonstrate that our approach achieves state-of-the-art performance, validating the effectiveness of integrating VLM-aided semantic priors and multi-modal features for ST-SGG. We will release the code at https://github.com/LUNAProject22/VOST.

Paperid: 175, https://arxiv.org/pdf/2512.05482.pdf GitHub

Abstract:
Rare-object detection remains a challenging task in autonomous driving systems, particularly when relying solely on point cloud data. Although Vision-Language Models (VLMs) exhibit strong capabilities in image understanding, their potential to enhance 3D object detection through intelligent data mining has not been fully explored. This paper proposes a novel cross-modal framework that leverages 2D VLMs to identify and mine rare objects from driving scenes, thereby improving 3D object detection performance. Our approach synthesizes complementary techniques such as object detection, semantic feature extraction, dimensionality reduction, and multi-faceted outlier detection into a cohesive, explainable pipeline that systematically identifies rare but critical objects in driving scenes. By combining Isolation Forest and t-SNE-based outlier detection methods with concept-based filtering, the framework effectively identifies semantically meaningful rare objects. A key strength of this approach lies in its ability to extract and annotate targeted rare object concepts such as construction vehicles, motorcycles, and barriers. This substantially reduces the annotation burden and focuses only on the most valuable training samples. Experiments on the nuScenes dataset demonstrate that this concept-guided data mining strategy enhances the performance of 3D object detection models while utilizing only a fraction of the training data, with particularly notable improvements for challenging object categories such as trailers and bicycles compared with the same amount of random data. This finding has substantial implications for the efficient curation of datasets in safety-critical autonomous systems.

Paperid: 176, https://arxiv.org/pdf/2512.05422.pdf GitHub

Abstract:
Unified multimodal models significantly improve visual generation by combining vision-language models (VLMs) with diffusion models. However, existing methods struggle to fully balance sufficient interaction and flexible implementation due to vast representation difference. Considering abundant and hierarchical information in VLM's layers from low-level details to high-level semantics, we propose \textbf{ParaUni}. It extracts features from variants VLM's layers in a \textbf{Para}llel way for comprehensive information interaction and retains a flexible separation architecture to enhance generation in \textbf{Uni}fied multimodal model. Concretely, visual features from all VLM's layers are fed in parallel into a Layer Integration Module (LIM), which efficiently integrates fine-grained details and semantic abstractions and provides the fused representation as a condition to the diffusion model. To further enhance performance, we reveal that these hierarchical layers respond unequally to different rewards in Reinforcement Learning (RL). Crucially, we design a Layer-wise Dynamic Adjustment Mechanism (LDAM) to facilitate multiple reward improvements that aligns the hierarchical properties of these layers using RL. Extensive experiments show ParaUni leverages complementary multi-layer features to substantially improve generation quality and shows strong potential for multiple reward advances during RL stages. Code is available at https://github.com/JosephTiTan/ParaUni.

Paperid: 177, https://arxiv.org/pdf/2512.05398.pdf GitHub

Abstract:
Estimating accurate camera poses, 3D scene geometry, and object motion from in-the-wild videos is a long-standing challenge for classical structure from motion pipelines due to the presence of dynamic objects. Recent learning-based methods attempt to overcome this challenge by training motion estimators to filter dynamic objects and focus on the static background. However, their performance is largely limited by the availability of large-scale motion segmentation datasets, resulting in inaccurate segmentation and, therefore, inferior structural 3D understanding. In this work, we introduce the Dynamic Prior (\ourmodel) to robustly identify dynamic objects without task-specific training, leveraging the powerful reasoning capabilities of Vision-Language Models (VLMs) and the fine-grained spatial segmentation capacity of SAM2. \ourmodel can be seamlessly integrated into state-of-the-art pipelines for camera pose optimization, depth reconstruction, and 4D trajectory estimation. Extensive experiments on both synthetic and real-world videos demonstrate that \ourmodel not only achieves state-of-the-art performance on motion segmentation, but also significantly improves accuracy and robustness for structural 3D understanding.

Paperid: 178, https://arxiv.org/pdf/2512.05277.pdf GitHub GitHub GitHub

Abstract:
Temporal understanding in autonomous driving (AD) remains a significant challenge, even for recent state-of-the-art (SoTA) Vision-Language Models (VLMs). Prior work has introduced datasets and benchmarks aimed at improving temporal reasoning, but these have emphasized other video content, including sports, cooking, and movies. No existing benchmark focuses exclusively on the unique challenges of temporal understanding in ego-centric AD footage. To fill this gap, the Temporal Understanding in Autonomous Driving (TAD) benchmark is presented, which evaluates VLMs' ability to capture the dynamic relationships between actions in AD. TAD comprises nearly 6,000 question-answer (QA) pairs, spanning 7 human-designed tasks. In addition, an evaluation is performed that consists of 9 closed- and open-source generalist models as well as SoTA AD specialist models. When applied to TAD, current SoTA models demonstrated substandard accuracies, largely due to imperfect fine-grained motion understanding. To improve motion understanding and overall accuracy on TAD, two novel training-free solutions are proposed: Scene-CoT, that leverages Chain-of-Thought (CoT) and TCogMap, which incorporates an ego-centric temporal cognitive map. The proposed approaches are integrated with existing VLMs and improve average accuracy on TAD by up to 17.72%. By introducing TAD, benchmarking multiple SoTA models, and proposing effective enhancements, this work aims to catalyze future research on temporal understanding in AD. The benchmark and evaluation code are available at \href{https://huggingface.co/datasets/vbdai/TAD}{Hugging Face} and \href{https://github.com/vbdi/tad_bench}{Github}, respectively.

Paperid: 179, https://arxiv.org/pdf/2512.05137.pdf GitHub

Abstract:
Vision-Language Models (VLMs) have advanced multimodal understanding, yet still struggle when targets are embedded in cluttered backgrounds requiring figure-ground segregation. To address this, we introduce ChromouVQA, a large-scale, multi-task benchmark based on Ishihara-style chromatic camouflaged images. We extend classic dot plates with multiple fill geometries and vary chromatic separation, density, size, occlusion, and rotation, recording full metadata for reproducibility. The benchmark covers nine vision-question-answering tasks, including recognition, counting, comparison, and spatial reasoning. Evaluations of humans and VLMs reveal large gaps, especially under subtle chromatic contrast or disruptive geometric fills. We also propose a model-agnostic contrastive recipe aligning silhouettes with their camouflaged renderings, improving recovery of global shapes. ChromouVQA provides a compact, controlled benchmark for reproducible evaluation and extension. Code and dataset are available at https://github.com/Chromou-VQA-Benchmark/Chromou-VQA.

Paperid: 180, https://arxiv.org/pdf/2512.05131.pdf GitHub

Abstract:
Active 3D reconstruction enables an agent to autonomously select viewpoints to efficiently obtain accurate and complete scene geometry, rather than passively reconstructing scenes from pre-collected images. However, existing active reconstruction methods often rely on hand-crafted geometric heuristics, which can lead to redundant observations without substantially improving reconstruction quality. To address this limitation, we propose AREA3D, an active reconstruction agent that leverages feed-forward 3D reconstruction models and vision-language guidance. Our framework decouples view-uncertainty modeling from the underlying feed-forward reconstructor, enabling precise uncertainty estimation without expensive online optimization. In addition, an integrated vision-language model provides high-level semantic guidance, encouraging informative and diverse viewpoints beyond purely geometric cues. Extensive experiments on both scene-level and object-level benchmarks demonstrate that AREA3D achieves state-of-the-art reconstruction accuracy, particularly in the sparse-view regime. Code will be made available at: https://github.com/TianlingXu/AREA3D .

Paperid: 181, https://arxiv.org/pdf/2512.04686.pdf GitHub

Abstract:
Cross-view correspondence is a fundamental capability for spatial understanding and embodied AI. However, it is still far from being realized in Vision-Language Models (VLMs), especially in achieving precise point-level correspondence, which is crucial for precise affordance interaction. So we propose the Cross-View Point Correspondence (CVPC) task and CrossPoint-Bench, a comprehensive benchmark with hierarchical design, inspired by the human cognitive process of "perceive", "reason", and "correspond". Our evaluation shows the state-of-the-art models (e.g., Gemini-2.5-Pro) still fall far behind humans, with a gap of over 54.65% in overall accuracy, exposing a challenge in transitioning from coarse-grained judgement to fine-grained coordinate prediction. To address this problem, we construct CrossPoint-378K, a dataset with 378K question-answering pairs across 900 scenes, focused on actionable affordance regions that better reflect real-world manipulation and interaction scenarios. Furthermore, we propose CroPond that trained on the CrossPoint-378K dataset. Our CroPond achieves state-of-the-art performance on CrossPoint-Bench, surpassing Gemini-2.5-Pro by 39.7% accuracy, which offers a foundation for advancing future work on cross-view correspondence. The benchmark, dataset, and model are publicly available at https://github.com/WangYipu2002/CrossPoint.

Paperid: 182, https://arxiv.org/pdf/2512.04425.pdf GitHub

Abstract:
Accurate and interpretable gait analysis plays a crucial role in the early detection of Parkinsons disease (PD),yet most existing approaches remain limited by single-modality inputs, low robustness, and a lack of clinical transparency. This paper presents an explainable multimodal framework that integrates RGB and Depth (RGB-D) data to recognize Parkinsonian gait patterns under realistic conditions. The proposed system employs dual YOLOv11-based encoders for modality-specific feature extraction, followed by a Multi-Scale Local-Global Extraction (MLGE) module and a Cross-Spatial Neck Fusion mechanism to enhance spatial-temporal representation. This design captures both fine-grained limb motion (e.g., reduced arm swing) and overall gait dynamics (e.g., short stride or turning difficulty), even in challenging scenarios such as low lighting or occlusion caused by clothing. To ensure interpretability, a frozen Large Language Model (LLM) is incorporated to translate fused visual embeddings and structured metadata into clinically meaningful textual explanations. Experimental evaluations on multimodal gait datasets demonstrate that the proposed RGB-D fusion framework achieves higher recognition accuracy, improved robustness to environmental variations, and clear visual-linguistic reasoning compared with single-input baselines. By combining multimodal feature learning with language-based interpretability, this study bridges the gap between visual recognition and clinical understanding, offering a novel vision-language paradigm for reliable and explainable Parkinsons disease gait analysis. Code:https://github.com/manaralnaasan/RGB-D_parkinson-LLM

Paperid: 183, https://arxiv.org/pdf/2512.04305.pdf GitHub

Abstract:
While vision-language models like CLIP have been extensively studied, their calibration, crucial for reliable predictions, has received limited attention. Although a few prior works have examined CLIP calibration in offline settings, the impact of fine-tuning CLIP in a federated learning (FL) setup remains unexplored. In this work, we investigate how FL affects CLIP calibration and propose strategies to improve reliability in this distributed setting. We first analyze Textual Prompt Tuning approaches and show that they degrade calibration metrics when operating under FL. We also evaluate existing in-training calibration techniques across four global aggregation methods, finding that they provide limited improvements. Our results suggest that the key challenge lies not only in how we aggregate or calibrate, but in which components we choose to fine-tune. Motivated by this insight, we propose $\text{FL}^2\text{oRA}$, a straightforward LoRA-based approach that naturally improves calibration in FL, and we analyze the factors behind its effectiveness. Experiments on multiple benchmarks demonstrate that $\text{FL}^2\text{oRA}$ consistently produces well-calibrated models, reducing the need for explicit calibration procedures. Codes are available at https://github.com/mainaksingha01/FL2oRA.

Paperid: 184, https://arxiv.org/pdf/2512.04069.pdf GitHub

Abstract:
Vision Language Models (VLMs) demonstrate strong qualitative visual understanding, but struggle with metrically precise spatial reasoning required for embodied applications. The agentic paradigm promises that VLMs can use a wide variety of tools that could augment these capabilities, such as depth estimators, segmentation models, and pose estimators. Yet it remains an open challenge how to realize this vision without solely relying on handcrafted prompting strategies or enforcing fixed, predefined tool pipelines that limit VLMs' ability to discover optimal tool-use patterns. Reinforcement Learning could overcome this gap, but has so far been limited to reasoning with a single visual tool due to the large search space in multi-tool reasoning. We introduce Double Interactive Reinforcement Learning (DIRL), a two-phase training framework where VLMs learn to coordinate multiple tools through interactive exploration and feedback. In the teaching phase, we combine demonstrations from a single tool specialist trained via interactive RL with traces from a frontier model using all tools. In the exploration phase, the model further refines multi-tool coordination through continued RL. Our model, SpaceTools, with tool-augmented spatial reasoning ability, achieves state-of-the-art performance on spatial understanding benchmarks (RoboSpatial-Home, BLINK, BOP-ASK) and demonstrates reliable real-world manipulation using a 7-DOF robot as a tool. DIRL provides substantial improvements over the vanilla SFT (+12% on RoboSpatial) and RL (+16% on RoboSpatial) baselines. Project page: https://spacetools.github.io/.

Paperid: 185, https://arxiv.org/pdf/2512.03558.pdf GitHub

Abstract:
The rise of Visual-Language Models (LVLMs) has unlocked new possibilities for seamlessly integrating visual and textual information. However, their ability to interpret cartographic maps remains largely unexplored. In this paper, we introduce CartoMapQA, a benchmark specifically designed to evaluate LVLMs' understanding of cartographic maps through question-answering tasks. The dataset includes over 2000 samples, each composed of a cartographic map, a question (with open-ended or multiple-choice answers), and a ground-truth answer. These tasks span key low-, mid- and high-level map interpretation skills, including symbol recognition, embedded information extraction, scale interpretation, and route-based reasoning. Our evaluation of both open-source and proprietary LVLMs reveals persistent challenges: models frequently struggle with map-specific semantics, exhibit limited geospatial reasoning, and are prone to Optical Character Recognition (OCR)-related errors. By isolating these weaknesses, CartoMapQA offers a valuable tool for guiding future improvements in LVLM architectures. Ultimately, it supports the development of models better equipped for real-world applications that depend on robust and reliable map understanding, such as navigation, geographic search, and urban planning. Our source code and data are openly available to the research community at: https://github.com/ungquanghuy-kddi/CartoMapQA.git

Paperid: 186, https://arxiv.org/pdf/2512.03508.pdf GitHub

Abstract:
Recent domain generalized semantic segmentation (DGSS) studies have achieved notable improvements by distilling semantic knowledge from Vision-Language Models (VLMs). However, they overlook the semantic misalignment between visual and textual contexts, which arises due to the rigidity of a fixed context prompt learned on a single source domain. To this end, we present a novel domain generalization framework for semantic segmentation, namely Domain-aware Prompt-driven Masked Transformer (DPMFormer). Firstly, we introduce domain-aware prompt learning to facilitate semantic alignment between visual and textual cues. To capture various domain-specific properties with a single source dataset, we propose domain-aware contrastive learning along with the texture perturbation that diversifies the observable domains. Lastly, to establish a framework resilient against diverse environmental changes, we have proposed the domain-robust consistency learning which guides the model to minimize discrepancies of prediction from original and the augmented images. Through experiments and analyses, we demonstrate the superiority of the proposed framework, which establishes a new state-of-the-art on various DGSS benchmarks. The code is available at https://github.com/jone1222/DPMFormer.

Paperid: 187, https://arxiv.org/pdf/2512.03445.pdf GitHub

Abstract:
Vision-language pretraining (VLP) has emerged as a powerful paradigm in medical image analysis, enabling representation learning from large-scale image-text pairs without relying on expensive manual annotations. However, existing methods often struggle with the noise inherent in web-collected data and the complexity of unstructured long medical texts. To address these challenges, we propose a novel VLP framework integrating a Multi-Agent data GENeration (MAGEN) system and Ontology-based Multi-Aspect Knowledge-Enhanced (O-MAKE) pretraining. First, MAGEN enhances data quality by synthesizing knowledge-enriched descriptions via a foundation model-assisted captioning and retrieval-based verification pipeline. Second, O-MAKE addresses the difficulty of learning from long, unstructured texts by decomposing them into distinct knowledge aspects. This facilitates fine-grained alignment at both global and patch levels, while explicitly modeling medical concept relationships through ontology-guided mechanisms. We validate our framework in the field of dermatology, where comprehensive experiments demonstrate the effectiveness of each component. Our approach achieves state-of-the-art zero-shot performance on disease classification and cross-modal retrieval tasks across eight datasets. Our code and the augmented dataset Derm1M-AgentAug, comprising over 400k skin-image-text pairs, will be released at https://github.com/SiyuanYan1/Derm1M.

Paperid: 188, https://arxiv.org/pdf/2512.02835.pdf GitHub

Abstract:
Reasoning-centric video object segmentation is an inherently complex task: the query often refers to dynamics, causality, and temporal interactions, rather than static appearances. Yet existing solutions generally collapse these factors into simplified reasoning with latent embeddings, rendering the reasoning chain opaque and essentially intractable. We therefore adopt an explicit decomposition perspective and introduce ReVSeg, which executes reasoning as sequential decisions in the native interface of pretrained vision language models (VLMs). Rather than folding all reasoning into a single-step prediction, ReVSeg executes three explicit operations -- semantics interpretation, temporal evidence selection, and spatial grounding -- aligning pretrained capabilities. We further employ reinforcement learning to optimize the multi-step reasoning chain, enabling the model to self-refine its decision quality from outcome-driven signals. Experimental results demonstrate that ReVSeg attains state-of-the-art performances on standard video object segmentation benchmarks and yields interpretable reasoning trajectories. Project page is available at https://clementine24.github.io/ReVSeg/ .

Paperid: 189, https://arxiv.org/pdf/2512.02787.pdf GitHub

Abstract:
Vision-Language-Action (VLA) models have recently achieved remarkable progress in robotic manipulation, yet they remain limited in failure diagnosis and learning from failures. Additionally, existing failure datasets are mostly generated programmatically in simulation, which limits their generalization to the real world. In light of these, we introduce ViFailback, a framework designed to diagnose robotic manipulation failures and provide both textual and visual correction guidance. Our framework utilizes explicit visual symbols to enhance annotation efficiency. We further release the ViFailback dataset, a large-scale collection of 58,126 Visual Question Answering (VQA) pairs along with their corresponding 5,202 real-world manipulation trajectories. Based on the dataset, we establish ViFailback-Bench, a benchmark of 11 fine-grained VQA tasks designed to assess the failure diagnosis and correction abilities of Vision-Language Models (VLMs), featuring ViFailback-Bench Lite for closed-ended and ViFailback-Bench Hard for open-ended evaluation. To demonstrate the effectiveness of our framework, we built the ViFailback-8B VLM, which not only achieves significant overall performance improvement on ViFailback-Bench but also generates visual symbols for corrective action guidance. Finally, by integrating ViFailback-8B with a VLA model, we conduct real-world robotic experiments demonstrating its ability to assist the VLA model in recovering from failures. Project Website: https://x1nyuzhou.github.io/vifailback.github.io/

Paperid: 190, https://arxiv.org/pdf/2512.02660.pdf GitHub

Abstract:
Vision-language models (VLMs) like ColPali achieve state-of-the-art document retrieval by embedding pages as images and computing fine-grained similarity between query tokens and visual patches. However, they return entire pages rather than specific regions, limiting utility for retrieval-augmented generation (RAG) where precise context is paramount. Conversely, OCR-based systems extract structured text with bounding box coordinates but lack semantic grounding for relevance assessment. We propose a hybrid architecture that unifies these paradigms: using ColPali's patch-level similarity scores as spatial relevance filters over OCR-extracted regions. We formalize the coordinate mapping between vision transformer patch grids and OCR bounding boxes, introduce intersection metrics for relevance propagation, and establish theoretical bounds on retrieval precision. Our approach operates at inference time without additional training. We release Snappy, an open-source implementation demonstrating practical applicability, with empirical evaluation ongoing.

Paperid: 191, https://arxiv.org/pdf/2512.02485.pdf GitHub

Abstract:
Vision-Language Models (VLMs) show promise in medical diagnosis, yet suffer from reasoning detachment, where linguistically fluent explanations drift from verifiable image evidence, undermining clinical trust. Recent multi-agent frameworks simulate Multidisciplinary Team (MDT) debates to mitigate single-model bias, but open-ended discussions amplify textual noise and computational cost while failing to anchor reasoning to visual evidence, the cornerstone of medical decision-making. We propose UCAgents, a hierarchical multi-agent framework enforcing unidirectional convergence through structured evidence auditing. Inspired by clinical workflows, UCAgents forbids position changes and limits agent interactions to targeted evidence verification, suppressing rhetorical drift while amplifying visual signal extraction. In UCAgents, a one-round inquiry discussion is introduced to uncover potential risks of visual-textual misalignment. This design jointly constrains visual ambiguity and textual noise, a dual-noise bottleneck that we formalize via information theory. Extensive experiments on four medical VQA benchmarks show UCAgents achieves superior accuracy (71.3% on PathVQA, +6.0% over state-of-the-art) with 87.7% lower token cost, the evaluation results further confirm that UCAgents strikes a balance between uncovering more visual evidence and avoiding confusing textual interference. These results demonstrate that UCAgents exhibits both diagnostic reliability and computational efficiency critical for real-world clinical deployment. Code is available at https://github.com/fqhank/UCAgents.

Paperid: 192, https://arxiv.org/pdf/2512.02438.pdf GitHub

Abstract:
In medical healthcare, obtaining detailed annotations is challenging, highlighting the need for robust Vision-Language Models (VLMs). Pretrained VLMs enable fine-tuning on small datasets or zero-shot inference, achieving performance comparable to task-specific models. Contrastive learning (CL) is a key paradigm for training VLMs but inherently requires large batch sizes for effective learning, making it computationally demanding and often limited to well-resourced institutions. Moreover, with limited data in healthcare, it is important to prioritize knowledge extraction from both data and models during training to improve performance. Therefore, we focus on leveraging the momentum method combined with distillation to simultaneously address computational efficiency and knowledge exploitation. Our contributions can be summarized as follows: (1) leveraging momentum self-distillation to enhance multimodal learning, and (2) integrating momentum mechanisms with gradient accumulation to enlarge the effective batch size without increasing resource consumption. Our method attains competitive performance with state-of-the-art (SOTA) approaches in zero-shot classification, while providing a substantial boost in the few-shot adaption, achieving over 90% AUC-ROC and improving retrieval tasks by 2-3%. Importantly, our method achieves high training efficiency with a single GPU while maintaining reasonable training time. Our approach aims to advance efficient multimodal learning by reducing resource requirements while improving performance over SOTA methods. The implementation of our method is available at https://github.com/phphuc612/MSD .

Paperid: 193, https://arxiv.org/pdf/2512.02340.pdf GitHub

Abstract:
Spatial reasoning is a core aspect of human intelligence that allows perception, inference and planning in 3D environments. However, current vision-language models (VLMs) struggle to maintain geometric coherence and cross-view consistency for spatial reasoning in multi-view settings. We attribute this gap to the lack of fine-grained benchmarks that isolate multi-view reasoning from single-view perception and temporal factors. To address this, we present ReMindView-Bench, a cognitively grounded benchmark for evaluating how VLMs construct, align and maintain spatial mental models across complementary viewpoints. ReMindView-Bench systematically varies viewpoint spatial pattern and query type to probe key factors of spatial cognition. Evaluations of 15 current VLMs reveals consistent failures in cross-view alignment and perspective-taking in multi-view spatial reasoning, motivating deeper analysis on the reasoning process. Explicit phase-wise analysis using LLM-as-a-judge and self-consistency prompting shows that VLMs perform well on in-frame perception but degrade sharply when integrating information across views. Implicit analysis, including linear probing and entropy dynamics, further show progressive loss of task-relevant information and uncertainty separation between correct and incorrect trajectories. These results provide a cognitively grounded diagnosis of VLM spatial reasoning and reveal how multi-view spatial mental models are formed, degraded and destabilized across reasoning phases. The ReMindView-Bench benchmark is available at https://huggingface.co/datasets/Xue0823/ReMindView-Bench, and the source codes of benchmark construction and VLM reasoning analysis are available at https://github.com/pittisl/ReMindView-Bench.

Paperid: 194, https://arxiv.org/pdf/2512.01843.pdf GitHub GitHub

Abstract:
Driven by the growing capacity and training scale, Text-to-Video (T2V) generation models have recently achieved substantial progress in video quality, length, and instruction-following capability. However, whether these models can understand physics and generate physically plausible videos remains a question. While Vision-Language Models (VLMs) have been widely used as general-purpose evaluators in various applications, they struggle to identify the physically impossible content from generated videos. To investigate this issue, we construct a \textbf{PID} (\textbf{P}hysical \textbf{I}mplausibility \textbf{D}etection) dataset, which consists of a \textit{test split} of 500 manually annotated videos and a \textit{train split} of 2,588 paired videos, where each implausible video is generated by carefully rewriting the caption of its corresponding real-world video to induce T2V models producing physically implausible content. With the constructed dataset, we introduce a lightweight fine-tuning approach, enabling VLMs to not only detect physically implausible events but also generate textual explanations on the violated physical principles. Taking the fine-tuned VLM as a physical plausibility detector and explainer, namely \textbf{PhyDetEx}, we benchmark a series of state-of-the-art T2V models to assess their adherence to physical laws. Our findings show that although recent T2V models have made notable progress toward generating physically plausible content, understanding and adhering to physical laws remains a challenging issue, especially for open-source models. Our dataset, training code, and checkpoints are available at \href{https://github.com/Zeqing-Wang/PhyDetEx}{https://github.com/Zeqing-Wang/PhyDetEx}.

Paperid: 195, https://arxiv.org/pdf/2512.01827.pdf GitHub GitHub

Abstract:
Causal thinking enables humans to understand not just what is seen, but why it happens. To replicate this capability in modern AI systems, we introduce the task of visual causal discovery. It requires models to infer cause-and-effect relations among visual entities across diverse scenarios instead of merely perceiving their presence. To this end, we first construct the Visual Causal Graph dataset (VCG-32K), a large-scale collection of over 32,000 images annotated with entity-level causal graphs, and further develop CauSight, a novel vision-language model to perform visual causal discovery through causally aware reasoning. Our training recipe integrates three components: (1) training data curation from VCG-32K, (2) Tree-of-Causal-Thought (ToCT) for synthesizing reasoning trajectories, and (3) reinforcement learning with a designed causal reward to refine the reasoning policy. Experiments show that CauSight outperforms GPT-4.1 on visual causal discovery, achieving over a threefold performance boost (21% absolute gain). Our code, model, and dataset are fully open-sourced at project page: https://github.com/OpenCausaLab/CauSight.

Paperid: 196, https://arxiv.org/pdf/2512.01422.pdf GitHub

Abstract:
Mask Diffusion Models (MDMs) have recently emerged as a promising alternative to auto-regressive models (ARMs) for vision-language tasks, owing to their flexible balance of efficiency and accuracy. In this paper, for the first time, we introduce MDMs into the Scene Text Recognition (STR) task. We show that vanilla MDM lags behind ARMs in terms of accuracy, although it improves recognition efficiency. To bridge this gap, we propose MDiff4STR, a Mask Diffusion model enhanced with two key improvement strategies tailored for STR. Specifically, we identify two key challenges in applying MDMs to STR: noising gap between training and inference, and overconfident predictions during inference. Both significantly hinder the performance of MDMs. To mitigate the first issue, we develop six noising strategies that better align training with inference behavior. For the second, we propose a token-replacement noise mechanism that provides a non-mask noise type, encouraging the model to reconsider and revise overly confident but incorrect predictions. We conduct extensive evaluations of MDiff4STR on both standard and challenging STR benchmarks, covering diverse scenarios including irregular, artistic, occluded, and Chinese text, as well as whether the use of pretraining. Across these settings, MDiff4STR consistently outperforms popular STR models, surpassing state-of-the-art ARMs in accuracy, while maintaining fast inference with only three denoising steps. Code: https://github.com/Topdu/OpenOCR.

Paperid: 197, https://arxiv.org/pdf/2512.01248.pdf GitHub

Abstract:
Table recognition (TR) aims to transform table images into semi-structured representations such as HTML or Markdown. As a core component of document parsing, TR has long relied on supervised learning, with recent efforts dominated by fine-tuning vision-language models (VLMs) using labeled data. While VLMs have brought TR to the next level, pushing performance further demands large-scale labeled data that is costly to obtain. Consequently, although proprietary models have continuously pushed the performance boundary, open-source models, often trained with limited resources and, in practice, the only viable option for many due to privacy regulations, still lag far behind. To bridge this gap, we introduce TRivia, a self-supervised fine-tuning method that enables pretrained VLMs to learn TR directly from unlabeled table images in the wild. Built upon Group Relative Policy Optimization, TRivia automatically identifies unlabeled samples that most effectively facilitate learning and eliminates the need for human annotations through a question-answering-based reward mechanism. An attention-guided module generates diverse questions for each table image, and the ability to interpret the recognition results and answer them correctly provides feedback to optimize the TR model. This closed-loop process allows the TR model to autonomously learn to recognize, structure, and reason over tables without labeled data. Leveraging this pipeline, we present TRivia-3B, an open-sourced, compact, and state-of-the-art TR model that surpasses existing systems (e.g., Gemini 2.5 Pro, MinerU2.5) on three popular benchmarks. Model and code are released at: https://github.com/opendatalab/TRivia

Paperid: 198, https://arxiv.org/pdf/2512.01048.pdf GitHub

Abstract:
Vision-language models (VLMs) have made great strides in addressing temporal understanding tasks, which involve characterizing visual changes across a sequence of images. However, recent works have suggested that when making predictions, VLMs may rely on static feature biases, such as background or object features, rather than dynamic visual changes. Static feature biases are a type of shortcut and can contribute to systematic prediction errors on downstream tasks; as a result, identifying and characterizing error-inducing static feature biases is critical prior to real-world model deployment. In this work, we introduce TRoVe, an automated approach for discovering error-inducing static feature biases learned by temporal VLMs. Given a trained VLM and an annotated validation dataset associated with a downstream classification task, TRoVe extracts candidate static features from the dataset and scores each feature by (i) the effect of the feature on classification errors as well as (ii) the extent to which the VLM relies on the feature when making predictions. In order to quantitatively evaluate TRoVe, we introduce an evaluation framework consisting of 101 trained temporal VLMs paired with ground-truth annotations for learned static feature biases. We use this framework to demonstrate that TRoVe can accurately identify error-inducing static feature biases in VLMs, achieving a 28.6% improvement over the closest baseline. Finally, we apply TRoVe to 7 off-the-shelf VLMs and 2 temporal understanding tasks, surfacing previously-unknown static feature biases and demonstrating that knowledge of learned biases can aid in improving model performance at test time. Our code is available at https://github.com/Stanford-AIMI/TRoVe.

Paperid: 199, https://arxiv.org/pdf/2512.01031.pdf GitHub

Abstract:
Vision-Language-Action models (VLAs) are becoming increasingly capable across diverse robotic tasks. However, their real-world deployment remains slow and inefficient: demonstration videos are often sped up by 5-10x to appear smooth, with noticeable action stalls and delayed reactions to environmental changes. Asynchronous inference offers a promising solution to achieve continuous and low-latency control by enabling robots to execute actions and perform inference simultaneously. However, because the robot and environment continue to evolve during inference, a temporal misalignment arises between the prediction and execution intervals. This leads to significant action instability, while existing methods either degrade accuracy or introduce runtime overhead to mitigate it. We propose VLASH, a general asynchronous inference framework for VLAs that delivers smooth, accurate, and fast reaction control without additional overhead or architectural changes. VLASH estimates the future execution-time state by rolling the robot state forward with the previously generated action chunk, thereby bridging the gap between prediction and execution. Experiments show that VLASH achieves up to 2.03x speedup and reduces reaction latency by up to 17.4x compared to synchronous inference while fully preserving the original accuracy. Moreover, it empowers VLAs to handle fast-reaction, high-precision tasks such as playing ping-pong and playing whack-a-mole, where traditional synchronous inference fails. Code is available at https://github.com/mit-han-lab/vlash

Paperid: 200, https://arxiv.org/pdf/2512.01022.pdf GitHub

Abstract:
In this paper, we explore an important yet underexplored task in robot manipulation: cycle-based manipulation, where robots need to perform cyclic or repetitive actions with an expected terminal time. These tasks are crucial in daily life, such as shaking a bottle or knocking a nail. However, few prior works have explored this task, leading to two main challenges: 1) the imitation methods often fail to complete these tasks within the expected terminal time due to the ineffective utilization of history; 2) the absence of a benchmark with sufficient data and automatic evaluation tools hinders development of effective solutions in this area. To address these challenges, we first propose the CycleManip framework to achieve cycle-based task manipulation in an end-to-end imitation manner without requiring any extra models, hierarchical structure or significant computational overhead. The core insight is to enhance effective history perception by a cost-aware sampling strategy and to improve historical understanding by multi-task learning. Second, we introduce a cycle-based task manipulation benchmark, which provides diverse cycle-based tasks, and an automatic evaluation method. Extensive experiments conducted in both simulation and real-world settings demonstrate that our method achieves high success rates in cycle-based task manipulation. The results further show strong adaptability performance in general manipulation, and the plug-and-play ability on imitation policies such as Vision-Language-Action (VLA) models. Moreover, the results show that our approach can be applied across diverse robotic platforms, including bi-arm grippers, dexterous hands, and humanoid robots.

Paperid: 201, https://arxiv.org/pdf/2512.00975.pdf GitHub

Abstract:
A generalist robotic policy needs both semantic understanding for task planning and the ability to interact with the environment through predictive capabilities. To tackle this, we present MM-ACT, a unified Vision-Language-Action (VLA) model that integrates text, image, and action in shared token space and performs generation across all three modalities. MM-ACT adopts a re-mask parallel decoding strategy for text and image generation, and employs a one-step parallel decoding strategy for action generation to improve efficiency. We introduce Context-Shared Multimodal Learning, a unified training paradigm that supervises generation in all three modalities from a shared context, enhancing action generation through cross-modal learning. Experiments were conducted on the LIBERO simulation and Franka real-robot setups as well as RoboTwin2.0 to assess in-domain and out-of-domain performances respectively. Our approach achieves a success rate of 96.3% on LIBERO, 72.0% across three tasks of real Franka, and 52.38% across eight bimanual tasks of RoboTwin2.0 with an additional gain of 9.25% from cross-modal learning. We release our codes, models and data at https://github.com/HHYHRHY/MM-ACT.

Paperid: 202, https://arxiv.org/pdf/2512.00936.pdf GitHub

Abstract:
Grounding complex, compositional visual queries with multiple objects and relationships is a fundamental challenge for vision-language models. While standard phrase grounding methods excel at localizing single objects, they lack the structural inductive bias to parse intricate relational descriptions, often failing as queries become more descriptive. To address this structural deficit, we focus on scene-graph grounding, a powerful but less-explored formulation where the query is an explicit graph of objects and their relationships. However, existing methods for this task also struggle, paradoxically showing decreased performance as the query graph grows -- failing to leverage the very information that should make grounding easier. We introduce SceneProp, a novel method that resolves this issue by reformulating scene-graph grounding as a Maximum a Posteriori (MAP) inference problem in a Markov Random Field (MRF). By performing global inference over the entire query graph, SceneProp finds the optimal assignment of image regions to nodes that jointly satisfies all constraints. This is achieved within an end-to-end framework via a differentiable implementation of the Belief Propagation algorithm. Experiments on four benchmarks show that our dedicated focus on the scene-graph grounding formulation allows SceneProp to significantly outperform prior work. Critically, its accuracy consistently improves with the size and complexity of the query graph, demonstrating for the first time that more relational context can, and should, lead to better grounding. Codes are available at https://github.com/keitaotani/SceneProp.

Paperid: 203, https://arxiv.org/pdf/2512.00343.pdf GitHub

Abstract:
Vision-language pretrained models (VLPs) such as CLIP have achieved remarkable success, but are also highly vulnerable to backdoor attacks. Given a model fine-tuned by an untrusted third party, determining whether the model has been injected with a backdoor is a critical and challenging problem. Existing detection methods usually rely on prior knowledge of training dataset, backdoor triggers and targets, or downstream classifiers, which may be impractical for real-world applications. To address this, To address this challenge, we introduce Assimilation Matters in DETection (AMDET), a novel model-level detection framework that operates without any such prior knowledge. Specifically, we first reveal the feature assimilation property in backdoored text encoders: the representations of all tokens within a backdoor sample exhibit a high similarity. Further analysis attributes this effect to the concentration of attention weights on the trigger token. Leveraging this insight, AMDET scans a model by performing gradient-based inversion on token embeddings to recover implicit features that capable of activating backdoor behaviors. Furthermore, we identify the natural backdoor feature in the OpenAI's official CLIP model, which are not intentionally injected but still exhibit backdoor-like behaviors. We then filter them out from real injected backdoor by analyzing their loss landscapes. Extensive experiments on 3,600 backdoored and benign-finetuned models with two attack paradigms and three VLP model structures show that AMDET detects backdoors with an F1 score of 89.90%. Besides, it achieves one complete detection in approximately 5 minutes on a RTX 4090 GPU and exhibits strong robustness against adaptive attacks. Code is available at: https://github.com/Robin-WZQ/AMDET

Paperid: 204, https://arxiv.org/pdf/2512.00021.pdf GitHub

Abstract:
The emergence of multi-modal foundation models has markedly transformed the technology for autonomous driving, shifting away from conventional and mostly hand-crafted design choices towards unified, foundation-model-based approaches, capable of directly inferring motion trajectories from raw sensory inputs. This new class of methods can also incorporate natural language as an additional modality, with Vision-Language-Action (VLA) models serving as a representative example. In this review, we provide a comprehensive examination of such methods through a unifying taxonomy to critically evaluate their architectural design choices, methodological strengths, and their inherent capabilities and limitations. Our survey covers 37 recently proposed approaches that span the landscape of trajectory planning with foundation models. Furthermore, we assess these approaches with respect to the openness of their source code and datasets, offering valuable information to practitioners and researchers. We provide an accompanying webpage that catalogs the methods based on our taxonomy, available at: https://github.com/fiveai/FMs-for-driving-trajectories

Paperid: 205, https://arxiv.org/pdf/2511.23469.pdf GitHub

Abstract:
Large Vision Language Models (VLMs) effectively bridge the modality gap through extensive pretraining, acquiring sophisticated visual representations aligned with language. However, it remains underexplored whether these representations, optimized for multimodal understanding tasks, harbor an inherent potential for visual generation. In this paper, we propose VGT, Visual Generation Tuning, a novel paradigm designed to stimulate the underlying capabilities of visual generation within any vision language models. By performing efficient visual generation tuning on well-pretrained VLMs, we significantly mitigate the alignment costs and accelerate the convergence of autoregressive modeling in the continuous space (20x speedup). Specifically, we dismiss the entangled pixel-level VAEs designed for diffusion transformers and formulate VGT-AE through aligning the semantic encoders from pretrained VLMs with the latent representations of pixel decoders. In image reconstruction tasks, we achieve 26.67 PSNR and 0.50 rFID at a 28x compression ratio, outperforming specialized VAEs; in visual generation tasks, we achieve state-of-the-art outcomes among autoregressive models, 0.77 on GenEval and 78.73 on DPG-Bench. Furthermore, our proposed VGT showcases significant scaling promise and is versatile for endowing any VLMs trained for multimodal understanding with the capabilities of visual generation, which paves the new avenue to explore next-generation unified multimodal foundation models. Models and codes are available at https://github.com/hustvl/VGT.

Paperid: 206, https://arxiv.org/pdf/2511.23186.pdf GitHub

Abstract:
Successful robotic grasping in cluttered environments not only requires a model to visually ground a target object but also to reason about obstructions that must be cleared beforehand. While current vision-language embodied reasoning models show emergent spatial understanding, they remain limited in terms of obstruction reasoning and accessibility planning. To bridge this gap, we present UNOGrasp, a learning-based vision-language model capable of performing visually-grounded obstruction reasoning to infer the sequence of actions needed to unobstruct the path and grasp the target object. We devise a novel multi-step reasoning process based on obstruction paths originated by the target object. We anchor each reasoning step with obstruction-aware visual cues to incentivize reasoning capability. UNOGrasp combines supervised and reinforcement finetuning through verifiable reasoning rewards. Moreover, we construct UNOBench, a large-scale dataset for both training and benchmarking, based on MetaGraspNetV2, with over 100k obstruction paths annotated by humans with obstruction ratios, contact points, and natural-language instructions. Extensive experiments and real-robot evaluations show that UNOGrasp significantly improves obstruction reasoning and grasp success across both synthetic and real-world environments, outperforming generalist and proprietary alternatives. Project website: https://tev-fbk.github.io/UnoGrasp/.

Paperid: 207, https://arxiv.org/pdf/2511.23112.pdf GitHub

Abstract:
Recent advances in Vision-Language Models (VLMs) have achieved impressive progress in multimodal mathematical reasoning. Yet, how much visual information truly contributes to reasoning remains unclear. Existing benchmarks report strong overall performance but seldom isolate the role of the image modality, leaving open whether VLMs genuinely leverage visual understanding or merely depend on linguistic priors. To address this, we present MathSight, a university-level multimodal mathematical reasoning benchmark designed to disentangle and quantify the effect of visual input. Each problem includes multiple visual variants -- original, hand-drawn, photo-captured -- and a text-only condition for controlled comparison. Experiments on state-of-the-art VLMs reveal a consistent trend: the contribution of visual information diminishes with increasing problem difficulty. Remarkably, Qwen3-VL without any image input surpasses both its multimodal variants and GPT-5, underscoring the need for benchmarks like MathSight to advance genuine vision-grounded reasoning in future models.

Paperid: 208, https://arxiv.org/pdf/2511.23034.pdf GitHub

Abstract:
Learning transferable latent actions from large-scale object manipulation videos can significantly enhance generalization in downstream robotics tasks, as such representations are agnostic to different robot embodiments. Existing approaches primarily rely on visual reconstruction objectives while neglecting physical priors, leading to sub-optimal performance in learning universal representations. To address these challenges, we propose a Universal Latent Action Learning framework that takes task instructions and multiple frames as inputs, and optimizes both future frame reconstruction and action sequence prediction. Unlike prior works, incorporating action predictions (e.g., gripper or hand trajectories and orientations) allows the model to capture richer physical priors such as real-world distances and orientations, thereby enabling seamless transferability to downstream tasks. We further decompose the latent actions into learnable motion and scene tokens to distinguish the robot's active movements from environmental changes, thus filtering out irrelevant dynamics. By distilling the learned latent actions into the latest VLA models, we achieve strong performance across both simulated (SIMPLER and LIBERO) and real-world robot settings. Notably, with only 10 real-world trajectories per task collected on a Franka robot, our approach successfully completes all five challenging tasks, demonstrating strong few-shot transferability in robotic manipulation.

Paperid: 209, https://arxiv.org/pdf/2511.22664.pdf GitHub

Abstract:
Vision-language models (VLMs), such as CLIP, have shown strong generalization under zero-shot settings, yet adapting them to downstream tasks with limited supervision remains a significant challenge. Existing multi-modal prompt learning methods typically rely on fixed, shared prompts and deterministic parameters, which limits their ability to capture instance-level variation or model uncertainty across diverse tasks and domains. To tackle this issue, we propose a novel Variational Multi-Modal Prompt Learning (VaMP) framework that enables sample-specific, uncertainty-aware prompt tuning in multi-modal representation learning. VaMP generates instance-conditioned prompts by sampling from a learned posterior distribution, allowing the model to personalize its behavior based on input content. To further enhance the integration of local and global semantics, we introduce a class-aware prior derived from the instance representation and class prototype. Building upon these, we formulate prompt tuning as variational inference over latent prompt representations and train the entire framework end-to-end through reparameterized sampling. Experiments on few-shot and domain generalization benchmarks show that VaMP achieves state-of-the-art performance, highlighting the benefits of modeling both uncertainty and task structure in our method. Project page: https://visual-ai.github.io/vamp

Paperid: 210, https://arxiv.org/pdf/2511.22645.pdf GitHub GitHub

Abstract:
Multimodal large language models (MLLMs) have undergone rapid development in advancing geospatial scene understanding. Recent studies have sought to enhance the reasoning capabilities of remote sensing MLLMs, typically through cold-start training with elaborately curated chain-of-thought (CoT) data. However, this approach not only incurs substantial annotation costs but also introduces human biases that may limit the diversity of model reasoning. To address these challenges, we propose GeoZero, a framework that enables MLLMs to perform geospatial reasoning without any predefined CoT supervision. Specifically, we construct two datasets, GeoZero-Instruct and GeoZero-Hard. GeoZero-Instruct allows the model to acquire preliminary geospatial knowledge through supervised fine-tuning, while GeoZero-Hard stimulates deep reasoning during the subsequent reinforcement learning stage. Furthermore, we introduce Answer-Anchored Group Relative Policy Optimization (A$^2$GRPO), where the reasoning process is regularized by the model's own answers, encouraging diverse yet accurate thinking. Extensive experiments on multiple remote sensing vision-language benchmarks demonstrate that GeoZero not only surpasses existing state-of-the-art methods but also fosters universal emergent reasoning capabilities across diverse geospatial tasks. Code,data,and models will be publicly available at https://github.com/MiliLab/GeoZero.

Paperid: 211, https://arxiv.org/pdf/2511.22594.pdf GitHub

Abstract:
Contrastive Language-Image Pre-training (CLIP) has demonstrated remarkable generalization ability and strong performance across a wide range of vision-language tasks. However, due to the lack of region-level supervision, CLIP exhibits limited fine-grained semantic understanding. Although several methods attempt to mitigate this issue, they unintentionally disrupt the global alignment, resulting in a persistent trade-off where improving local perception simultaneously degrades global coherence. In this paper, we propose HarmoCLIP, a novel framework designed to harmonize global and region representations within CLIP. We first identify that the absence of direct alignment between local textual and visual semantics is the fundamental cause of the trade-off. To address this, HarmoCLIP introduces an explicit fine-grained semantic supervision term that directly aligns textual segments with their corresponding visual regions, effectively bridging the image region space and the textual space. To further strengthen the representation capability at the local level, our method introduces a novel Region-Language Alignment supervision strategy that promotes fine-grained semantic learning without compromising global semantic consistency. Extensive experiments demonstrate that HarmoCLIP achieves state-of-the-art (improvement up to 69.78%) performance on the global task of retrieval and yields a substantial 3.2% improvement in Top-1 accuracy on the region task of bounding-box classification, consistently outperforming prior approaches while providing a balanced, efficient, and plug-and-play solution to the global-local trade-off in CLIP. Code is available at https://github.com/Erosist/HarmoCLIP.

Paperid: 212, https://arxiv.org/pdf/2511.22466.pdf GitHub

Abstract:
Understanding mid-level road semantics, which capture the structural and contextual cues that link low-level perception to high-level planning, is essential for reliable autonomous driving and digital map construction. However, existing benchmarks primarily target perception tasks such as detection or segmentation, overlooking the reasoning capabilities required to infer road topology and dynamic scene structure. To address this gap, we present RoadSceneBench, a lightweight yet information-rich benchmark designed to evaluate and advance visual reasoning in complex road environments. Unlike large-scale perception datasets, RoadSceneBench emphasizes relational understanding and structural consistency, encouraging models to capture the underlying logic of real-world road scenes. Furthermore, to enhance reasoning reliability, we propose Hierarchical Relational Reward Propagation with Temporal Consistency (HRRP-T), a training framework for Vision-Language Models (VLMs) in which reward signals adaptively promote spatial coherence and semantic alignment throughout the reasoning process. This paradigm enables models to move beyond static recognition toward geometry-aware and temporally consistent reasoning. Extensive experiments demonstrate that our method achieves state-of-the-art performance across diverse road configurations. RoadSceneBench thus provides a compact yet powerful foundation for studying mid-level road semantics and fostering structure-aware autonomous perception. Our dataset is available at https://github.com/XiyanLiu/RoadSceneBench.

Paperid: 213, https://arxiv.org/pdf/2511.22396.pdf GitHub

Abstract:
Recent multimodal reasoning models, inspired by DeepSeek-R1, have significantly advanced vision-language systems. However, in remote sensing (RS) tasks, we observe widespread pseudo reasoning: models narrate the process of reasoning rather than genuinely reason toward the correct answer based on visual evidence. We attribute this to the Glance Effect, where a single, coarse perception of large-scale RS imagery results in incomplete understanding and reasoning based on linguistic self-consistency instead of visual evidence. To address this, we propose RS-EoT (Remote Sensing Evidence-of-Thought), a language-driven, iterative visual evidence-seeking paradigm. To instill this paradigm, we propose SocraticAgent, a self-play multi-agent system that synthesizes reasoning traces via alternating cycles of reasoning and visual inspection. To enhance and generalize these patterns, we propose a two-stage progressive RL strategy: first, RL on fine-grained Grounding tasks to enhance RS-EoT capabilities, followed by RL on RS VQA to generalize to broader understanding scenarios. Experiments show RS-EoT achieves state-of-the-art performance on multiple RS VQA and grounding benchmarks. Analyses reveal clear iterative cycles of reasoning and evidence seeking, confirming RS-EoT mitigates the Glance Effect and enables genuine evidence-grounded reasoning. Our code, data, and models are available at https://geox-lab.github.io/Asking_like_Socrates

Paperid: 214, https://arxiv.org/pdf/2511.22235.pdf GitHub

Abstract:
The rapid development of large vision-language model (VLM) has greatly promoted the research of GUI agent. However, GUI agents still face significant challenges in handling long-horizon tasks. First, single-agent models struggle to balance high-level capabilities and low-level execution capability, facing prevalent issues of responsibility coupling and capability conflicts. Second, agents lack awareness of the task state, leading to progress loss in long-horizon tasks. To address these challenges, we propose a staged execution-feedback reinforcement learning algorithm. Unlike training a unified policy model, we focus on training high-level scheduling models. Specifically, we propose and train two agents: a Coordinator, responsible for the strategic planning and task decomposition; and a State Tracker, responsible for context compression and information management to maintain the task's state and coherence. Based on this, we built the Coordinator-Executor-State Tracker (CES) multi-agent framework, which can be integrated with any low-level Executor model, assisting the Executor in solving long-horizon tasks through task scheduling and state management. Experiments on long-horizon task benchmarks demonstrate that CES significantly enhances the system's planning and state management capabilities. Furthermore, analysis confirms that our trained high-level scheduling module is a generalizable, plug-and-play module that significantly enhances the long-horizon capabilities of various Executors. Code can be available at https://github.com/hehehahi4/CES.

Paperid: 215, https://arxiv.org/pdf/2511.22134.pdf GitHub

Abstract:
To build a generalizable Vision-Language-Action (VLA) model with strong reasoning ability, a common strategy is to first train a specialist VLA on robot demonstrations to acquire reliable manipulation skills, and then incorporate mixed annotated robot data together with multimodal data to restore broader reasoning capabilities. However, we observe that the resulting reasoning VLA often suffers from degraded action performance compared to the specialist model before fine-tuning, a phenomenon we refer to as action degeneration. To address this issue, we propose DualVLA, which enhances action performance through carefully designed post-training while still preserving reasoning capability. We first introduce a dual-layer data pruning method that removes redundant embodied reasoning, preventing it from adversely influencing action learning. To further strengthen action generation, we design a dual-teacher adaptive distillation strategy that assigns different supervision signals to different data domains while maintaining reasoning ability. To fill the evaluation gap for generalist VLAs, we also propose VLA Score, which decouples VLA capability into reasoning, intention, action, and alignment dimensions for a more fine-grained assessment. Experiments show that DualVLA achieves an average success rate of 61.0 in SimplerEnv and an average score of 65.4 across eight competitive multimodal benchmarks, demonstrating a stronger balance between precise action execution and multimodal understanding. Project Website: https://costaliya.github.io/DualVLA/.

Paperid: 216, https://arxiv.org/pdf/2511.22019.pdf GitHub

Abstract:
Vision-language models (VLMs), such as CLIP, have gained popularity for their strong open vocabulary classification performance, but they are prone to assigning high confidence scores to misclassifications, limiting their reliability in safety-critical applications. We introduce a training-free, post-hoc uncertainty estimation method for contrastive VLMs that can be used to detect erroneous predictions. The key to our approach is to measure visual feature consistency within a class, using feature projection combined with multivariate Gaussians to create class-specific probabilistic embeddings. Our method is VLM-agnostic, requires no fine-tuning, demonstrates robustness to distribution shift, and works effectively with as few as 10 training images per class. Extensive experiments on ImageNet, Flowers102, Food101, EuroSAT and DTD show state-of-the-art error detection performance, significantly outperforming both deterministic and probabilistic VLM baselines. Code is available at https://github.com/zhenxianglin/ICPE.

Paperid: 217, https://arxiv.org/pdf/2511.21982.pdf GitHub

Abstract:
The precise reading recognition of pointer meters plays a key role in smart power systems, but existing approaches remain fragile due to challenges like reflections, occlusions, dynamic viewing angles, and overly between thin pointers and scale markings. Up to now, this area still lacks large-scale datasets to support the development of robust algorithms. To address these challenges, this paper first presents a new large-scale benchmark dataset for dial reading, termed RPM-10K, which contains 10730 meter images that fully reflect the aforementioned key challenges. Built upon the dataset, we propose a novel vision-language model for pointer meter reading recognition, termed MRLM, based on physical relation injection. Instead of exhaustively learning image-level correlations, MRLM explicitly encodes the geometric and causal relationships between the pointer and the scale, aligning perception with physical reasoning in the spirit of world-model perspectives. Through cross-attentional fusion and adaptive expert selection, the model learns to interpret dial configurations and generate precise numeric readings. Extensive experiments fully validated the effectiveness of our proposed framework on the newly proposed benchmark dataset. Both the dataset and source code will be released on https://github.com/Event-AHU/DialBench

Paperid: 218, https://arxiv.org/pdf/2511.21688.pdf GitHub

Abstract:
Vision-Language Models (VLMs) still lack robustness in spatial intelligence, demonstrating poor performance on spatial understanding and reasoning tasks. We attribute this gap to the absence of a visual geometry learning process capable of reconstructing 3D space from 2D images. We present G$^2$VLM, a geometry grounded vision-language model that bridges two fundamental aspects of spatial intelligence: spatial 3D reconstruction and spatial understanding. G$^2$VLM natively leverages learned 3D visual geometry features to directly predict 3D attributes and enhance spatial reasoning tasks via in-context learning and interleaved reasoning. Our unified design is highly scalable for spatial understanding: it trains on abundant multi-view image and video data, while simultaneously leveraging the benefits of 3D visual priors that are typically only derived from hard-to-collect annotations. Experimental results demonstrate G$^2$VLM is proficient in both tasks, achieving comparable results to state-of-the-art feed-forward 3D reconstruction models and achieving better or competitive results across spatial understanding and reasoning tasks. By unifying a semantically strong VLM with low-level 3D vision tasks, we hope G$^2$VLM can serve as a strong baseline for the community and unlock more future applications, such as 3D scene editing.

Paperid: 219, https://arxiv.org/pdf/2511.20956.pdf GitHub

Abstract:
Automated radiology report generation (RRG) for breast ultrasound (BUS) is limited by the lack of paired image-report datasets and the risk of hallucinations from large language models. We propose BUSTR, a multitask vision-language framework that generates BUS reports without requiring paired image-report supervision. BUSTR constructs reports from structured descriptors (e.g., BI-RADS, pathology, histology) and radiomics features, learns descriptor-aware visual representations with a multi-head Swin encoder trained using a multitask loss over dataset-specific descriptor sets, and aligns visual and textual tokens via a dual-level objective that combines token-level cross-entropy with a cosine-similarity alignment loss between input and output representations. We evaluate BUSTR on two public BUS datasets, BrEaST and BUS-BRA, which differ in size and available descriptors. Across both datasets, BUSTR consistently improves standard natural language generation metrics and clinical efficacy metrics, particularly for key targets such as BI-RADS category and pathology. Our results show that this descriptor-aware vision model, trained with a combined token-level and alignment loss, improves both automatic report metrics and clinical efficacy without requiring paired image-report data. The source code can be found at https://github.com/AAR-UNLV/BUSTR

Paperid: 220, https://arxiv.org/pdf/2511.20937.pdf GitHub

Abstract:
Embodied cognition argues that intelligence arises from sensorimotor interaction rather than passive observation. It raises an intriguing question: do modern vision-language models (VLMs), trained largely in a disembodied manner, exhibit signs of embodied cognition? We introduce ENACT, a benchmark that casts evaluation of embodied cognition as world modeling from egocentric interaction in a visual question answering (VQA) format. Framed as a partially observable Markov decision process (POMDP) whose actions are scene graph changes, ENACT comprises two complementary sequence reordering tasks: forward world modeling (reorder shuffled observations given actions) and inverse world modeling (reorder shuffled actions given observations). While conceptually simple, solving these tasks implicitly demands capabilities central to embodied cognition-affordance recognition, action-effect reasoning, embodied awareness, and interactive, long-horizon memory from partially observable egocentric input, while avoiding low-level image synthesis that could confound the evaluation. We provide a scalable pipeline that synthesizes QA pairs from robotics simulation (BEHAVIOR) and evaluates models on 8,972 QA pairs spanning long-horizon home-scale activities. Experiments reveal a performance gap between frontier VLMs and humans that widens with interaction horizon. Models consistently perform better on the inverse task than the forward one and exhibit anthropocentric biases, including a preference for right-handed actions and degradation when camera intrinsics or viewpoints deviate from human vision. Website at https://enact-embodied-cognition.github.io/.

Paperid: 221, https://arxiv.org/pdf/2511.20841.pdf GitHub

Abstract:
To manipulate objects in novel, unstructured environments, robots need task-oriented grasps that target object parts based on the given task. Geometry-based methods often struggle with visually defined parts, occlusions, and unseen objects. We introduce OVAL-Grasp, a zero-shot open-vocabulary approach to task-oriented, affordance based grasping that uses large-language models and vision-language models to allow a robot to grasp objects at the correct part according to a given task. Given an RGB image and a task, OVAL-Grasp identifies parts to grasp or avoid with an LLM, segments them with a VLM, and generates a 2D heatmap of actionable regions on the object. During our evaluations, we found that our method outperformed two task oriented grasping baselines on experiments with 20 household objects with 3 unique tasks for each. OVAL-Grasp successfully identifies and segments the correct object part 95% of the time and grasps the correct actionable area 78.3% of the time in real-world experiments with the Fetch mobile manipulator. Additionally, OVAL-Grasp finds correct object parts under partial occlusions, demonstrating a part selection success rate of 80% in cluttered scenes. We also demonstrate OVAL-Grasp's efficacy in scenarios that rely on visual features for part selection, and show the benefit of a modular design through our ablation experiments. Our project webpage is available at https://ekjt.github.io/OVAL-Grasp/

Paperid: 222, https://arxiv.org/pdf/2511.20648.pdf GitHub

Abstract:
To act in the world, a model must name what it sees and know where it is in 3D. Today's vision-language models (VLMs) excel at open-ended 2D description and grounding, yet multi-object 3D detection remains largely missing from the VLM toolbox. We present LocateAnything3D, a VLM-native recipe that casts 3D detection as a next-token prediction problem. The key is a short, explicit Chain-of-Sight (CoS) sequence that mirrors how human reason from images: find an object in 2D, then infer its distance, size, and pose. The decoder first emits 2D detections as a visual chain-of-thought, then predicts 3D boxes under an easy-to-hard curriculum: across objects, a near-to-far order reduces early ambiguity and matches ego-centric utility; within each object, a center-from-camera, dimensions, and rotation factorization ranks information by stability and learnability. This VLM-native interface preserves open-vocabulary and visual-prompting capability without specialized heads. On the challenging Omni3D benchmark, our model achieves state-of-the-art results, with 49.89 AP_3D, surpassing the previous best by +15.51 absolute improvement even when the baseline is given ground-truth 2D boxes. It also generalizes zero-shot to held-out categories with strong robustness. By turning 3D detection into a disciplined next-token problem, LocateAnything3D offers a practical foundation for models to perceive in 3D.

Paperid: 223, https://arxiv.org/pdf/2511.20633.pdf GitHub

Abstract:
Vision-Language-Action (VLA) policies excel in aligning language, perception, and robot control. However, most VLAs are trained purely by imitation, which overfits to demonstrations, and is brittle under distribution shift. Reinforcement learning (RL) directly optimizes task reward and thus addresses this misalignment, but real-robot interaction is expensive and conventional simulators are hard to engineer and transfer. We address both data efficiency and optimization stability in VLA post-training via a learned world model and an RL procedure tailored to flow-based action heads. Specifically, we introduce Prophet, a unified action-to-video robot actuation pretrained across large-scale, heterogeneous robot data to learn reusable action-outcome dynamics. It is able to few-shot adapt to new robots, objects, and environments, yielding a rollout-ready simulator. Upon Prophet, we reinforce action policies with Flow-action-GRPO (FA-GRPO), which adapts Flow-GRPO to operate on VLA actions, and with FlowScale, a stepwise reweighting that rescales per-step gradients in the flow head. Together, Prophet, FA-GRPO, and FlowScale constitute ProphRL, a practical, data- and compute-efficient path to VLA post-training. Experiments show 5-17% success gains on public benchmarks and 24-30% gains on real robots across different VLA variants.

Paperid: 224, https://arxiv.org/pdf/2511.20439.pdf GitHub

Abstract:
In Vision Language Models (VLMs), vision tokens are quantity-heavy yet information-dispersed compared with language tokens, thus consume too much unnecessary computation. Pruning redundant vision tokens for high VLM inference efficiency has been continuously studied but all existing methods resort to indirect and non-guaranteed ways. We propose OC-VTP, a direct and guaranteed approach to select the most representative vision tokens for high-efficiency yet accuracy-preserving VLM inference. Our OC-VTP requires merely light-weight pre-training of a small object-centric vision token pruner, which can then be inserted into existing VLMs, without fine-tuning of any models on any datasets. It is gauranteed that the most representative vision tokens are kept by minimizing the error in reconstructing the original unpruned tokens from the selected ones. Across any vision pruning ratios, i.e., inference efficiency, our OC-VTP consistently helps mainstream VLMs to preserve the highest inference accuracy. Our pruning also demonstrates interesting interpretability. Our codes are available at https://github.com/GarryLarry010131/OC-VTP.

Paperid: 225, https://arxiv.org/pdf/2511.20223.pdf GitHub

Abstract:
Adversarial attacks have evolved from simply disrupting predictions on conventional task-specific models to the more complex goal of manipulating image semantics on Large Vision-Language Models (LVLMs). However, existing methods struggle with controllability and fail to precisely manipulate the semantics of specific concepts in the image. We attribute this limitation to semantic entanglement in the patch-token representations on which adversarial attacks typically operate: global context aggregated by self-attention in the vision encoder dominates individual patch features, making them unreliable handles for precise local semantic manipulation. Our systematic investigation reveals a key insight: value features (V) computed within the transformer attention block serve as much more precise handles for manipulation. We show that V suppresses global-context channels, allowing it to retain high-entropy, disentangled local semantic information. Building on this discovery, we propose V-Attack, a novel method designed for precise local semantic attacks. V-Attack targets the value features and introduces two core components: (1) a Self-Value Enhancement module to refine V's intrinsic semantic richness, and (2) a Text-Guided Value Manipulation module that leverages text prompts to locate source concept and optimize it toward a target concept. By bypassing the entangled patch features, V-Attack achieves highly effective semantic control. Extensive experiments across diverse LVLMs, including LLaVA, InternVL, DeepseekVL and GPT-4o, show that V-Attack improves the attack success rate by an average of 36% over state-of-the-art methods, exposing critical vulnerabilities in modern visual-language understanding. Our code and data are available https://github.com/Summu77/V-Attack.

Paperid: 226, https://arxiv.org/pdf/2511.19900.pdf GitHub

Abstract:
Vision-language agents have achieved remarkable progress in a variety of multimodal reasoning tasks; however, their learning remains constrained by the limitations of human-annotated supervision. Recent self-rewarding approaches attempt to overcome this constraint by allowing models to act as their own critics or reward providers. Yet, purely text-based self-evaluation struggles to verify complex visual reasoning steps and often suffers from evaluation hallucinations. To address these challenges, inspired by recent advances in tool-integrated reasoning, we propose Agent0-VL, a self-evolving vision-language agent that achieves continual improvement with tool-integrated reasoning. Agent0-VL incorporates tool usage not only into reasoning but also into self-evaluation and self-repair, enabling the model to introspect, verify, and refine its reasoning through evidence-grounded analysis. It unifies two synergistic roles within a single LVLM: a Solver that performs multi-turn tool-integrated reasoning, and a Verifier that generates structured feedback and fine-grained self-rewards through tool-grounded critique. These roles interact through a Self-Evolving Reasoning Cycle, where tool-based verification and reinforcement learning jointly align the reasoning and evaluation distributions for stable self-improvement. Through this zero-external-reward evolution, Agent0-VL aligns its reasoning and verification behaviors without any human annotation or external reward models, achieving continual self-improvement. Experiments on geometric problem solving and visual scientific analysis show that Agent0-VL achieves an 12.5% improvement over the base model. Our code is available at https://github.com/aiming-lab/Agent0.

Paperid: 227, https://arxiv.org/pdf/2511.19768.pdf GitHub

Abstract:
Large vision-language models (VLMs) have improved embodied question answering (EQA) agents by providing strong semantic priors for open-vocabulary reasoning. However, when used directly for step-level exploration, VLMs often exhibit frontier oscillations, unstable back-and-forth movements caused by overconfidence and miscalibration, leading to inefficient navigation and degraded answer quality. We propose Prune-Then-Plan, a simple and effective framework that stabilizes exploration through step-level calibration. Instead of trusting raw VLM scores, our method prunes implausible frontier choices using a Holm-Bonferroni inspired pruning procedure and then delegates final decisions to a coverage-based planner. This separation converts overconfident predictions into conservative, interpretable actions by relying on human-level judgments to calibrate the step-level behavior of VLMs. Integrated into the 3D-Mem EQA framework, our approach achieves relative improvements of up to 49% and 33% in visually grounded SPL and LLM-Match metrics respectively over baselines. Overall, our method achieves better scene coverage under equal exploration budgets on both OpenEQA and EXPRESS-Bench datasets.

Paperid: 228, https://arxiv.org/pdf/2511.19676.pdf GitHub

Abstract:
We introduce INTERLACE, a novel framework that prunes redundant layers in VLMs while maintaining performance through sample-efficient finetuning. Existing layer pruning methods lead to significant performance drop when applied to VLMs. Instead, we analyze triplets of consecutive layers to identify local redundancy, removing the most redundant of the first two layers, finetune the remaining layer to compensate for the lost capacity, and freeze the third layer to serve as a stable anchor during finetuning. We found that this interleaved finetune-freeze design enables rapid convergence with minimal data after pruning. By finetuning only a subset of layers on just 1% of the FineVision dataset for one epoch, Interlace achieves 88.9% average performance retention after dropping 25% of the network, achieving SOTA performance. Our code is available at: https://github.com/pmadinei/Interlace.git

Paperid: 229, https://arxiv.org/pdf/2511.19433.pdf GitHub

Abstract:
Vision-language-action (VLA) models have shown remarkable capabilities in robotic manipulation, but their performance is sensitive to the $\textbf{action chunk length}$ used during training, termed $\textbf{horizon}$. Our empirical study reveals an inherent trade-off: longer horizons provide stronger global foresight but degrade fine-grained accuracy, while shorter ones sharpen local control yet struggle on long-term tasks, implying fixed choice of single horizons being suboptimal. To mitigate the trade-off, we propose a $\textbf{mixture of horizons (MoH)}$ strategy. MoH rearranges the action chunk into several segments with different horizons, processes them in parallel with a shared action transformer, and fuses outputs with a light linear gate. It has three appealing benefits. 1) MoH exploits long-term foresight and short-term precision jointly within a single model, improving both performance and generalizability to complex tasks. 2) MoH is plug-and-play for full-attention action modules with minimal training or inference overhead. 3) MoH enables dynamic inference with adaptive horizons, which selects stable actions through cross-horizon consensus, achieving 2.5$\times$ higher throughput than baselines while preserving superior performance. Extensive experiments over flow-based policies $π_0$, $π_{0.5}$, and one-step regression policy $π_{\text{reg}}$ demonstrate that MoH yields consistent and significant gains on both simulations and real-world tasks. Notably, under mixed-task setting, $π_{0.5}$ with MoH reaches a new state-of-the-art with 99$\%$ average success rate on LIBERO after only $30k$ training iterations. Project page: https://github.com/Timsty1/MixtureOfHorizons

Paperid: 230, https://arxiv.org/pdf/2511.19418.pdf GitHub

Abstract:
Vision-Language Models (VLMs) excel at reasoning in linguistic space but struggle with perceptual understanding that requires dense visual perception, e.g., spatial reasoning and geometric awareness. This limitation stems from the fact that current VLMs have limited mechanisms to capture dense visual information across spatial dimensions. We introduce Chain-of-Visual-Thought (COVT), a framework that enables VLMs to reason not only in words but also through continuous visual tokens-compact latent representations that encode rich perceptual cues. Within a small budget of roughly 20 tokens, COVT distills knowledge from lightweight vision experts, capturing complementary properties such as 2D appearance, 3D geometry, spatial layout, and edge structure. During training, the VLM with COVT autoregressively predicts these visual tokens to reconstruct dense supervision signals (e.g., depth, segmentation, edges, and DINO features). At inference, the model reasons directly in the continuous visual token space, preserving efficiency while optionally decoding dense predictions for interpretability. Evaluated across more than ten diverse perception benchmarks, including CV-Bench, MMVP, RealWorldQA, MMStar, WorldMedQA, and HRBench, integrating COVT into strong VLMs such as Qwen2.5-VL and LLaVA consistently improves performance by 3% to 16% and demonstrates that compact continuous visual thinking enables more precise, grounded, and interpretable multimodal intelligence.

Paperid: 231, https://arxiv.org/pdf/2511.18921.pdf GitHub

Abstract:
Backdoor attacks undermine the reliability and trustworthiness of machine learning systems by injecting hidden behaviors that can be maliciously activated at inference time. While such threats have been extensively studied in unimodal settings, their impact on multimodal foundation models, particularly vision-language models (VLMs), remains largely underexplored. In this work, we introduce \textbf{BackdoorVLM}, the first comprehensive benchmark for systematically evaluating backdoor attacks on VLMs across a broad range of settings. It adopts a unified perspective that injects and analyzes backdoors across core vision-language tasks, including image captioning and visual question answering. BackdoorVLM organizes multimodal backdoor threats into 5 representative categories: targeted refusal, malicious injection, jailbreak, concept substitution, and perceptual hijack. Each category captures a distinct pathway through which an adversary can manipulate a model's behavior. We evaluate these threats using 12 representative attack methods spanning text, image, and bimodal triggers, tested on 2 open-source VLMs and 3 multimodal datasets. Our analysis reveals that VLMs exhibit strong sensitivity to textual instructions, and in bimodal backdoors the text trigger typically overwhelms the image trigger when forming the backdoor mapping. Notably, backdoors involving the textual modality remain highly potent, with poisoning rates as low as 1\% yielding over 90\% success across most tasks. These findings highlight significant, previously underexplored vulnerabilities in current VLMs. We hope that BackdoorVLM can serve as a useful benchmark for analyzing and mitigating multimodal backdoor threats. Code is available at: https://github.com/bin015/BackdoorVLM .

Paperid: 232, https://arxiv.org/pdf/2511.18617.pdf GitHub

Abstract:
AutoFocus-IL is a simple yet effective method to improve data efficiency and generalization in visual imitation learning by guiding policies to attend to task-relevant features rather than distractors and spurious correlations. Although saliency regularization has emerged as a promising way to achieve this, existing approaches typically require costly supervision such as human gaze data or manual saliency annotations. In contrast, AutoFocus-IL leverages vision-language models (VLMs) to automatically identify and track key objects in demonstrations, generating temporal saliency maps that highlight causal visual signals while suppressing distractors. These maps are then used to regularize behavior cloning policies, yielding stronger alignment between visual attention and task-relevant cues. Experiments in both the CARLA simulator and real-robot manipulation tasks demonstrate that AutoFocus-IL not only outperforms standard behavior cloning but also surpasses state-of-the-art baselines that assume privileged access to human supervision, such as gaze data. Code, datasets, and trained policy videos are available at https://AutoFocus-IL.github.io/.

Paperid: 233, https://arxiv.org/pdf/2511.18570.pdf GitHub

Abstract:
Understanding physical properties such as friction, stiffness, hardness, and material composition is essential for enabling robots to interact safely and effectively with their surroundings. However, existing 3D reconstruction methods focus on geometry and appearance and cannot infer these underlying physical properties. We present PhysGS, a Bayesian-inferred extension of 3D Gaussian Splatting that estimates dense, per-point physical properties from visual cues and vision--language priors. We formulate property estimation as Bayesian inference over Gaussian splats, where material and property beliefs are iteratively refined as new observations arrive. PhysGS also models aleatoric and epistemic uncertainties, enabling uncertainty-aware object and scene interpretation. Across object-scale (ABO-500), indoor, and outdoor real-world datasets, PhysGS improves accuracy of the mass estimation by up to 22.8%, reduces Shore hardness error by up to 61.2%, and lowers kinetic friction error by up to 18.1% compared to deterministic baselines. Our results demonstrate that PhysGS unifies 3D reconstruction, uncertainty modeling, and physical reasoning in a single, spatially continuous framework for dense physical property estimation. Additional results are available at https://samchopra2003.github.io/physgs.

Paperid: 234, https://arxiv.org/pdf/2511.18359.pdf GitHub

Abstract:
How do video understanding models acquire their answers? Although current Vision Language Models (VLMs) reason over complex scenes with diverse objects, action performances, and scene dynamics, understanding and controlling their internal processes remains an open challenge. Motivated by recent advancements in text-to-video (T2V) generative models, this paper introduces a logits-to-video (L2V) task alongside a model-independent approach, TRANSPORTER, to generate videos that capture the underlying rules behind VLMs' predictions. Given the high-visual-fidelity produced by T2V models, TRANSPORTER learns an optimal transport coupling to VLM's high-semantic embedding spaces. In turn, logit scores define embedding directions for conditional video generation. TRANSPORTER generates videos that reflect caption changes over diverse object attributes, action adverbs, and scene context. Quantitative and qualitative evaluations across VLMs demonstrate that L2V can provide a fidelity-rich, novel direction for model interpretability that has not been previously explored.

Paperid: 235, https://arxiv.org/pdf/2511.17964.pdf GitHub

Abstract:
Large-scale vision-language models (e.g., CLIP) have recently achieved remarkable performance in retrieval tasks, yet their potential for Video-based Visible-Infrared Person Re-Identification (VVI-ReID) remains largely unexplored. The primary challenges are narrowing the modality gap and leveraging spatiotemporal information in video sequences. To address the above issues, in this paper, we propose a novel cross-modality feature learning framework named X-ReID for VVI-ReID. Specifically, we first propose a Cross-modality Prototype Collaboration (CPC) to align and integrate features from different modalities, guiding the network to reduce the modality discrepancy. Then, a Multi-granularity Information Interaction (MII) is designed, incorporating short-term interactions from adjacent frames, long-term cross-frame information fusion, and cross-modality feature alignment to enhance temporal modeling and further reduce modality gaps. Finally, by integrating multi-granularity information, a robust sequence-level representation is achieved. Extensive experiments on two large-scale VVI-ReID benchmarks (i.e., HITSZ-VCM and BUPTCampus) demonstrate the superiority of our method over state-of-the-art methods. The source code is released at https://github.com/AsuradaYuci/X-ReID.

Paperid: 236, https://arxiv.org/pdf/2511.17889.pdf GitHub GitHub

Abstract:
Grounding natural-language instructions into continuous control for quadruped robots remains a fundamental challenge in vision language action. Existing methods struggle to bridge high-level semantic reasoning and low-level actuation, leading to unstable grounding and weak generalization in the real world. To address these issues, we present MobileVLA-R1, a unified vision-language-action framework that enables explicit reasoning and continuous control for quadruped robots. We construct MobileVLA-CoT, a large-scale dataset of multi-granularity chain-of-thought (CoT) for embodied trajectories, providing structured reasoning supervision for alignment. Built upon this foundation, we introduce a two-stage training paradigm that combines supervised CoT alignment with GRPO reinforcement learning to enhance reasoning consistency, control stability, and long-horizon execution. Extensive evaluations on VLN and VLA tasks demonstrate superior performance over strong baselines, with approximately a 5% improvement. Real-world deployment on a quadruped robot validates robust performance in complex environments. Code: https://github.com/AIGeeksGroup/MobileVLA-R1. Website: https://aigeeksgroup.github.io/MobileVLA-R1.

Paperid: 237, https://arxiv.org/pdf/2511.17448.pdf GitHub

Abstract:
Vision-Language Models (VLMs) are increasingly deployed in safety-critical applications, making their adversarial robustness a crucial concern. While adversarial knowledge distillation has shown promise in transferring robustness from teacher to student models, traditional single-teacher approaches suffer from limited knowledge diversity, slow convergence, and difficulty in balancing robustness and accuracy. To address these challenges, we propose MMT-ARD: a Multimodal Multi-Teacher Adversarial Robust Distillation framework. Our key innovation is a dual-teacher knowledge fusion architecture that collaboratively optimizes clean feature preservation and robust feature enhancement. To better handle challenging adversarial examples, we introduce a dynamic weight allocation strategy based on teacher confidence, enabling adaptive focus on harder samples. Moreover, to mitigate bias among teachers, we design an adaptive sigmoid-based weighting function that balances the strength of knowledge transfer across modalities. Extensive experiments on ImageNet and zero-shot benchmarks demonstrate that MMT-ARD improves robust accuracy by +4.32% and zero-shot accuracy by +3.5% on the ViT-B-32 model, while achieving a 2.3x increase in training efficiency over traditional single-teacher methods. These results highlight the effectiveness and scalability of MMT-ARD in enhancing the adversarial robustness of multimodal large models. Our codes are available at https://github.com/itsnotacie/MMT-ARD.

Paperid: 238, https://arxiv.org/pdf/2511.17405.pdf GitHub

Abstract:
Multiple-choice question answering (MCQA) has been a popular format for evaluating and reinforcement fine-tuning (RFT) of modern multimodal language models. Its constrained output format allows for simplified, deterministic automatic verification. However, we find that the options may leak exploitable signals, which makes the accuracy metrics unreliable for indicating real capabilities and encourages explicit or implicit answer guessing behaviors during RFT. We propose ReVeL (Rewrite and Verify by LLM), a framework that rewrites multiple-choice questions into open-form questions while keeping answers verifiable whenever possible. The framework categorizes questions according to different answer types, apply different rewriting and verification schemes, respectively. When applied for RFT, we converted 20k MCQA examples and use GRPO to finetune Qwen2.5-VL models. Models trained on ReVeL-OpenQA match MCQA accuracy on multiple-choice benchmarks and improve OpenQA accuracy by about six percentage points, indicating better data efficiency and more robust reward signals than MCQA-based training. When used for evaluation, ReVeL also reveals up to 20 percentage points of score inflation in MCQA benchmarks (relative to OpenQA), improves judging accuracy, and reduces both cost and latency. We will release code and data publicly.

Paperid: 239, https://arxiv.org/pdf/2511.17254.pdf GitHub

Abstract:
Despite their impressive performance across a wide range of tasks, Large Vision-Language Models (LVLMs) remain prone to hallucination. In this study, we propose a comprehensive intervention framework aligned with the transformer's causal architecture in LVLMs, integrating the effects of different intervention paths on hallucination. We find that hallucinations in LVLMs do not arise from a single causal path, but rather from the interplay among image-to-input-text, image-to-output-text, and text-to-text pathways. For the first time, we also find that LVLMs rely on different pathways depending on the question-answer alignment format. Building on these insights, we propose simple yet effective methods to identify and intervene on critical hallucination heads within each pathway, tailored to discriminative and generative formats. Experiments across multiple benchmarks demonstrate that our approach consistently reduces hallucinations across diverse alignment types.

Paperid: 240, https://arxiv.org/pdf/2511.16853.pdf GitHub GitHub

Abstract:
Recent progress in vision language models (VLMs) has enabled remarkable perception and reasoning capabilities, yet their potential for scientific regression in Earth Observation (EO) remains largely unexplored. Existing EO datasets mainly emphasize semantic understanding tasks such as captioning or classification, lacking benchmarks that align multimodal perception with measurable biophysical variables. To fill this gap, we present REO-Instruct, the first unified benchmark designed for both descriptive and regression tasks in EO. REO-Instruct establishes a cognitively interpretable logic chain in forest ecological scenario (human activity,land-cover classification, ecological patch counting, above-ground biomass (AGB) regression), bridging qualitative understanding and quantitative prediction. The dataset integrates co-registered Sentinel-2 and ALOS-2 imagery with structured textual annotations generated and validated through a hybrid human AI pipeline. Comprehensive evaluation protocols and baseline results across generic VLMs reveal that current models struggle with numeric reasoning, highlighting an essential challenge for scientific VLMs. REO-Instruct offers a standardized foundation for developing and assessing next-generation geospatial models capable of both description and scientific inference. The project page are publicly available at \href{https://github.com/zhu-xlab/REO-Instruct}{REO-Instruct}.

Paperid: 241, https://arxiv.org/pdf/2511.16669.pdf GitHub

Abstract:
While language models have become impactful in many real-world applications, video generation remains largely confined to entertainment. Motivated by video's inherent capacity to demonstrate physical-world information that is difficult to convey through language alone (e.g., imagine teaching someone to tie a tie using only text), we identify an underutilized opportunity to extend video as a new answer modality for Next-Event Prediction (NEP), formalized as Video-Next-Event Prediction (VNEP). While the established NEP task takes a video with a procedural or predictive question as input to predict the next event in text, VNEP requires dynamic video responses. This shift from telling to showing unlocks more intuitive and customized answers for procedural learning and creative exploration. However, this task remains challenging for existing models, as it demands an understanding of multimodal input, instruction-conditioned reasoning, and the generation of video with visual and semantic consistency. To address this, we introduce VANS, a model that leverages reinforcement learning to align a Vision-Language Model (VLM) with a Video Diffusion Model (VDM) for VNEP. The core of VANS is our proposed Joint-GRPO that orchestrates the VLM and VDM to function as a unit. Driven by a shared reward on their respective output, it optimizes the VLM to produce captions that are both accurate and friendly to visualize, while guiding the VDM to generate videos that are faithful to these captions and the input visual context. To enable this learning, we craft VANS-Data-100K, a dedicated dataset for the VNEP task. Experiments on procedural and predictive benchmarks demonstrate that VANS achieves state-of-the-art performance in both video event prediction and visualization. Codes are released in https://github.com/KlingTeam/VANS.

Paperid: 242, https://arxiv.org/pdf/2511.16595.pdf GitHub GitHub

Abstract:
We introduce TimeViper, a hybrid vision-language model designed to tackle challenges of long video understanding. Processing long videos demands both an efficient model architecture and an effective mechanism for handling extended temporal contexts. To this end, TimeViper adopts a hybrid Mamba-Transformer backbone that combines the efficiency of state-space models with the expressivity of attention mechanisms. Through this hybrid design, we reveal the vision-to-text information aggregation phenomenon, where information progressively flows from vision tokens to text tokens across increasing LLM depth, resulting in severe vision token redundancy. Motivated by this observation, we propose TransV, a token information transfer module that transfers and compresses vision tokens into instruction tokens while maintaining multimodal understanding capabilities. This design enables TimeViper to process hour-long videos exceeding 10,000 frames. Extensive experiments across multiple benchmarks demonstrate that TimeViper competes with state-of-the-art models while extending frame numbers. We further analyze attention behaviors of both Mamba and Transformer layers, offering new insights into hybrid model interpretability. This work represents an initial step towards developing, interpreting, and compressing hybrid Mamba-Transformer architectures.

Paperid: 243, https://arxiv.org/pdf/2511.16216.pdf GitHub

Abstract:
The development of Large Language Models (LLMs) increasingly depends on high-quality supervised data, yet existing instruction-tuning and RL datasets remain costly to curate and often rely on synthetic samples that introduce hallucination and limited diversity. At the same time, textbooks and exercise materials contain abundant, high-quality human-authored Question-Answer(QA) content that remains underexploited due to the difficulty of transforming raw PDFs into AI-ready supervision. Although modern OCR and vision-language models can accurately parse document structure, their outputs lack the semantic alignment required for training. We propose an automated pipeline that extracts well-formed QA and visual-QA (VQA) pairs from educational documents by combining layout-aware OCR with LLM-based semantic parsing. Experiments across diverse document types show that the method produces accurate, aligned, and low-noise QA/VQA pairs. This approach enables scalable use of real-world educational content and provides a practical alternative to synthetic data generation for improving reasoning-oriented LLM training. All code and data-processing pipelines are open-sourced at https://github.com/OpenDCAI/DataFlow.

Paperid: 244, https://arxiv.org/pdf/2511.16166.pdf GitHub GitHub

Abstract:
Long-horizon robotic manipulation remains challenging for Vision-Language-Action (VLA) models despite recent progress in zero-shot generalization and simulation-to-real-world transfer. Current VLA models suffer from stage hallucination, where agents exploit coarse evaluation signals to shortcut multi-step tasks, reporting high progress without truly completing them. We present EvoVLA, a self-supervised VLA framework that addresses this issue through three complementary components: Stage-Aligned Reward (SAR), which uses triplet contrastive learning with Gemini-generated hard negatives to prevent visual shortcuts; Pose-Based Object Exploration (POE), which grounds curiosity in relative object-gripper pose instead of raw pixels; and Long-Horizon Memory, which uses selective context retention and gated fusion to stabilize intrinsic shaping during extended rollouts. Extensive evaluations on Discoverse-L, a long-horizon manipulation benchmark with three multi-stage tasks, show that EvoVLA improves average task success by 10.2 percentage points over the strongest baseline (OpenVLA-OFT), reaching 69.2 percent. EvoVLA also achieves one-and-a-half times better sample efficiency and reduces stage hallucination from 38.5 percent to 14.8 percent. Real-world deployment on physical robots reaches an average success rate of 54.6 percent across four manipulation tasks, outperforming OpenVLA-OFT by 11 points, demonstrating effective sim-to-real transfer and strong generalization. Code: https://github.com/AIGeeksGroup/EvoVLA. Website: https://aigeeksgroup.github.io/EvoVLA.

Paperid: 245, https://arxiv.org/pdf/2511.16110.pdf GitHub

Abstract:
The growing misuse of Vision-Language Models (VLMs) has led providers to deploy multiple safeguards, including alignment tuning, system prompts, and content moderation. However, the real-world robustness of these defenses against adversarial attacks remains underexplored. We introduce Multi-Faceted Attack (MFA), a framework that systematically exposes general safety vulnerabilities in leading defense-equipped VLMs such as GPT-4o, Gemini-Pro, and Llama-4. The core component of MFA is the Attention-Transfer Attack (ATA), which hides harmful instructions inside a meta task with competing objectives. We provide a theoretical perspective based on reward hacking to explain why this attack succeeds. To improve cross-model transferability, we further introduce a lightweight transfer-enhancement algorithm combined with a simple repetition strategy that jointly bypasses both input-level and output-level filters without model-specific fine-tuning. Empirically, we show that adversarial images optimized for one vision encoder transfer broadly to unseen VLMs, indicating that shared visual representations create a cross-model safety vulnerability. Overall, MFA achieves a 58.5% success rate and consistently outperforms existing methods. On state-of-the-art commercial models, MFA reaches a 52.8% success rate, surpassing the second-best attack by 34%. These results challenge the perceived robustness of current defense mechanisms and highlight persistent safety weaknesses in modern VLMs. Code: https://github.com/cure-lab/MultiFacetedAttack

Paperid: 246, https://arxiv.org/pdf/2511.15943.pdf GitHub GitHub

Abstract:
Recent advances in image-text pretraining have significantly enhanced visual understanding by aligning visual and textual representations. Contrastive Language-Image Pretraining (CLIP) has played a pivotal role in multimodal learning. However, its focus on single-label, single-granularity alignment limits its effectiveness in complex domains such as medical imaging, where images often correspond to multiple high-level labels (e.g., disease categories) across different annotation granularities (e.g., diagnostic description, clinical explanation). To address this, we propose Multi-Granular Language Learning (MGLL), a contrastive learning framework designed to improve both multi-label and cross-granularity alignment. MGLL leverages structured multi-label supervision, integrates textual descriptions across granularities, and introduces soft-label supervision with point-wise constraints to enhance alignment. MGLL employs smooth Kullback-Leibler (KL) divergence to ensure cross-granularity consistency while maintaining computational efficiency as a plug-and-play module for vision-language models. Pretrained on our constructed large-scale multi-granular datasets and evaluated across multiple datasets, MGLL outperforms other state-of-the-art methods in downstream tasks. The code is available at \href{https://github.com/HUANGLIZI/MGLL}{https://github.com/HUANGLIZI/MGLL}.

Paperid: 247, https://arxiv.org/pdf/2511.15703.pdf GitHub

Abstract:
Abstract reasoning from minimal examples remains a core unsolved problem for frontier foundation models such as GPT-5 and Grok 4. These models still fail to infer structured transformation rules from a handful of examples, which is a key hallmark of human intelligence. The Abstraction and Reasoning Corpus for Artificial General Intelligence (ARC-AGI) provides a rigorous testbed for this capability, demanding conceptual rule induction and transfer to novel tasks. Most existing methods treat ARC-AGI as a purely textual reasoning task, overlooking the fact that humans rely heavily on visual abstraction when solving such puzzles. However, our pilot experiments reveal a paradox: naively rendering ARC-AGI grids as images degrades performance due to imprecise rule execution. This leads to our central hypothesis that vision and language possess complementary strengths across distinct reasoning stages: vision supports global pattern abstraction and verification, whereas language specializes in symbolic rule formulation and precise execution. Building on this insight, we introduce two synergistic strategies: (1) Vision-Language Synergy Reasoning (VLSR), which decomposes ARC-AGI into modality-aligned subtasks; and (2) Modality-Switch Self-Correction (MSSC), which leverages vision to verify text-based reasoning for intrinsic error correction. Extensive experiments demonstrate that our approach yields up to a 4.33\% improvement over text-only baselines across diverse flagship models and multiple ARC-AGI tasks. Our findings suggest that unifying visual abstraction with linguistic reasoning is a crucial step toward achieving generalizable, human-like intelligence in future foundation models. Source code is released at https://github.com/InternLM/ARC-VL.

Paperid: 248, https://arxiv.org/pdf/2511.15661.pdf GitHub

Abstract:
Reinforcement learning (RL) provides a principled framework for improving Vision-Language Models (VLMs) on complex reasoning tasks. However, existing RL approaches often rely on human-annotated labels or task-specific heuristics to define verifiable rewards, both of which are costly and difficult to scale. We introduce VisPlay, a self-evolving RL framework that enables VLMs to autonomously improve their reasoning abilities using large amounts of unlabeled image data. Starting from a single base VLM, VisPlay assigns the model into two interacting roles: an Image-Conditioned Questioner that formulates challenging yet answerable visual questions, and a Multimodal Reasoner that generates silver responses. These roles are jointly trained with Group Relative Policy Optimization (GRPO), which incorporates diversity and difficulty rewards to balance the complexity of generated questions with the quality of the silver answers. VisPlay scales efficiently across two model families. When trained on Qwen2.5-VL and MiMo-VL, VisPlay achieves consistent improvements in visual reasoning, compositional generalization, and hallucination reduction across eight benchmarks, including MM-Vet and MMMU, demonstrating a scalable path toward self-evolving multimodal intelligence. The project page is available at https://bruno686.github.io/VisPlay/

Paperid: 249, https://arxiv.org/pdf/2511.15311.pdf GitHub

Abstract:
3D Vision-Language Foundation Models (VLFMs) have shown strong generalization and zero-shot recognition capabilities in open-world point cloud processing tasks. However, these models often underperform in practical scenarios where data are noisy, incomplete, or drawn from a different distribution than the training data. To address this, we propose Uni-Adapter, a novel training-free online test-time adaptation (TTA) strategy for 3D VLFMs based on dynamic prototype learning. We define a 3D cache to store class-specific cluster centers as prototypes, which are continuously updated to capture intra-class variability in heterogeneous data distributions. These dynamic prototypes serve as anchors for cache-based logit computation via similarity scoring. Simultaneously, a graph-based label smoothing module captures inter-prototype similarities to enforce label consistency among similar prototypes. Finally, we unify predictions from the original 3D VLFM and the refined 3D cache using entropy-weighted aggregation for reliable adaptation. Without retraining, Uni-Adapter effectively mitigates distribution shifts, achieving state-of-the-art performance on diverse 3D benchmarks over different 3D VLFMs, improving ModelNet-40C by 10.55%, ScanObjectNN-C by 8.26%, and ShapeNet-C by 4.49% over the source 3D VLFMs. Project page: https://mehran-tam.github.io/Uni-Adapter

Paperid: 250, https://arxiv.org/pdf/2511.14901.pdf GitHub

Abstract:
As CLIP's global alignment limits its ability to capture fine-grained details, recent efforts have focused on enhancing its region-text alignment. However, current remote sensing (RS)-specific CLIP variants still inherit this limited spatial awareness. We identify two key limitations behind this: (1) current RS image-text datasets generate global captions from object-level labels, leaving the original object-level supervision underutilized; (2) despite the success of region-text alignment methods in general domain, their direct application to RS data often leads to performance degradation. To address these, we construct the first multi-granularity RS image-text dataset, MGRS-200k, featuring rich object-level textual supervision for RS region-category alignment. We further investigate existing fine-grained CLIP tuning strategies and find that current explicit region-text alignment methods, whether in a direct or indirect way, underperform due to severe degradation of CLIP's semantic coherence. Building on these, we propose FarSLIP, a Fine-grained Aligned RS Language-Image Pretraining framework. Rather than the commonly used patch-to-CLS self-distillation, FarSLIP employs patch-to-patch distillation to align local and global visual cues, which improves feature discriminability while preserving semantic coherence. Additionally, to effectively utilize region-text supervision, it employs simple CLS token-based region-category alignment rather than explicit patch-level alignment, further enhancing spatial awareness. FarSLIP features improved fine-grained vision-language alignment in RS domain and sets a new state of the art not only on RS open-vocabulary semantic segmentation, but also on image-level tasks such as zero-shot classification and image-text retrieval. Our dataset, code, and models are available at https://github.com/NJU-LHRS/FarSLIP.

Paperid: 251, https://arxiv.org/pdf/2511.14659.pdf GitHub

Abstract:
Vision--language--action (VLA) models have recently shown promising performance on a variety of embodied tasks, yet they still fall short in reliability and generalization, especially when deployed across different embodiments or real-world environments. In this work, we introduce NORA-1.5, a VLA model built from the pre-trained NORA backbone by adding to it a flow-matching-based action expert. This architectural enhancement alone yields substantial performance gains, enabling NORA-1.5 to outperform NORA and several state-of-the-art VLA models across both simulated and real-world benchmarks. To further improve robustness and task success, we develop a set of reward models for post-training VLA policies. Our rewards combine (i) an action-conditioned world model (WM) that evaluates whether generated actions lead toward the desired goal, and (ii) a deviation-from-ground-truth heuristic that distinguishes good actions from poor ones. Using these reward signals, we construct preference datasets and adapt NORA-1.5 to target embodiments through direct preference optimization (DPO). Extensive evaluations show that reward-driven post-training consistently improves performance in both simulation and real-robot settings, demonstrating significant VLA model-reliability gains through simple yet effective reward models. Our findings highlight NORA-1.5 and reward-guided post-training as a viable path toward more dependable embodied agents suitable for real-world deployment.

Paperid: 252, https://arxiv.org/pdf/2511.14631.pdf GitHub

Abstract:
We show that multi-agent systems guided by vision-language models (VLMs) improve end-to-end autonomous scientific discovery. By treating plots as verifiable checkpoints, a VLM-as-a-judge evaluates figures against dynamically generated domain-specific rubrics, enabling agents to correct their own errors and steer exploratory data analysis in real-time. Case studies in cosmology and astrochemistry demonstrate recovery from faulty reasoning paths and adaptation to new datasets without human intervention. On a 10-task benchmark for data-driven discovery, VLM-augmented systems achieve pass at 1 scores of 0.7-0.8, compared to 0.2-0.3 for code-only and 0.4-0.5 for code-and-text baselines, while also providing auditable reasoning traces that improve interpretability. Code available here: https://github.com/CMBAgents/cmbagent

Paperid: 253, https://arxiv.org/pdf/2511.14396.pdf GitHub

Abstract:
Language-conditioned manipulation facilitates human-robot interaction via behavioral cloning (BC), which learns control policies from human demonstrations and serves as a cornerstone of embodied AI. Overcoming compounding errors in sequential action decisions remains a central challenge to improving BC performance. Existing approaches mitigate compounding errors through data augmentation, expressive representation, or temporal abstraction. However, they suffer from physical discontinuities and semantic-physical misalignment, leading to inaccurate action cloning and intermittent execution. In this paper, we present Continuous vision-language-action Co-Learning with Semantic-Physical Alignment (CCoL), a novel BC framework that ensures temporally consistent execution and fine-grained semantic grounding. It generates robust and smooth action execution trajectories through continuous co-learning across vision, language, and proprioceptive inputs (e.g., robot internal states). Meanwhile, we anchor language semantics to visuomotor representations by a bidirectional cross-attention to learn contextual information for action generation, successfully overcoming the problem of semantic-physical misalignment. Extensive experiments show that CCoL achieves an average 8.0% relative improvement across three simulation suites, with up to 19.2% relative gain in human-demonstrated bimanual insertion tasks. Real-world tests on a 7-DoF robot further confirm CCoL's generalization under unseen and noisy object states.

Paperid: 254, https://arxiv.org/pdf/2511.14178.pdf GitHub

Abstract:
Vision-Language-Action (VLA) models have demonstrated significant potential in real-world robotic manipulation. However, pre-trained VLA policies still suffer from substantial performance degradation during downstream deployment. Although fine-tuning can mitigate this issue, its reliance on costly demonstration collection and intensive computation makes it impractical in real-world settings. In this work, we introduce VLA-Pilot, a plug-and-play inference-time policy steering method for zero-shot deployment of pre-trained VLA without any additional fine-tuning or data collection. We evaluate VLA-Pilot on six real-world downstream manipulation tasks across two distinct robotic embodiments, encompassing both in-distribution and out-of-distribution scenarios. Experimental results demonstrate that VLA-Pilot substantially boosts the success rates of off-the-shelf pre-trained VLA policies, enabling robust zero-shot generalization to diverse tasks and embodiments. Experimental videos and code are available at: https://rip4kobe.github.io/vla-pilot/.

Paperid: 255, https://arxiv.org/pdf/2511.14159.pdf GitHub

Abstract:
Evaluating the robustness of Large Vision-Language Models (LVLMs) is essential for their continued development and responsible deployment in real-world applications. However, existing robustness benchmarks typically focus on hallucination or misleading textual inputs, while largely overlooking the equally critical challenge posed by misleading visual inputs in assessing visual understanding. To fill this important gap, we introduce MVI-Bench, the first comprehensive benchmark specially designed for evaluating how Misleading Visual Inputs undermine the robustness of LVLMs. Grounded in fundamental visual primitives, the design of MVI-Bench centers on three hierarchical levels of misleading visual inputs: Visual Concept, Visual Attribute, and Visual Relationship. Using this taxonomy, we curate six representative categories and compile 1,248 expertly annotated VQA instances. To facilitate fine-grained robustness evaluation, we further introduce MVI-Sensitivity, a novel metric that characterizes LVLM robustness at a granular level. Empirical results across 18 state-of-the-art LVLMs uncover pronounced vulnerabilities to misleading visual inputs, and our in-depth analyses on MVI-Bench provide actionable insights that can guide the development of more reliable and robust LVLMs. The benchmark and codebase can be accessed at https://github.com/chenyil6/MVI-Bench.

Paperid: 256, https://arxiv.org/pdf/2511.14148.pdf GitHub

Abstract:
Vision-language-action (VLA) models have recently emerged as a powerful paradigm for building generalist robots. However, traditional VLA models that generate actions through flow matching (FM) typically rely on rigid and uniform time schedules, i.e., synchronous FM (SFM). Without action context awareness and asynchronous self-correction, SFM becomes unstable in long-horizon tasks, where a single action error can cascade into failure. In this work, we propose asynchronous flow matching VLA (AsyncVLA), a novel framework that introduces temporal flexibility in asynchronous FM (AFM) and enables self-correction in action generation. AsyncVLA breaks from the vanilla SFM in VLA models by generating the action tokens in a non-uniform time schedule with action context awareness. Besides, our method introduces the confidence rater to extract confidence of the initially generated actions, enabling the model to selectively refine inaccurate action tokens before execution. Moreover, we propose a unified training procedure for SFM and AFM that endows a single model with both modes, improving KV-cache utilization. Extensive experiments on robotic manipulation benchmarks demonstrate that AsyncVLA is data-efficient and exhibits self-correction ability. AsyncVLA achieves state-of-the-art results across general embodied evaluations due to its asynchronous generation in AFM. Our code is available at https://github.com/YuhuaJiang2002/AsyncVLA.

Paperid: 257, https://arxiv.org/pdf/2511.13876.pdf GitHub

Abstract:
Contrastive Language-Image Pretraining (CLIP) has demonstrated strong generalization for vision-language tasks in computer vision and medical domains, yet its text encoder accepts only up to 77 tokens, which limits its ability to represent long and information-rich radiology reports. Recent adaptations using domain-specific encoders, such as PubMedBERT or ClinicalBERT, mitigate this issue by leveraging medical corpora, but remain constrained by their limited input length (typically 512 tokens) and relatively shallow semantic understanding. To address these limitations, we propose QwenCLIP, a vision-language framework that replaces CLIP's text encoder with a large language model (LLM)-based embedding module (e.g., Qwen3-Embedding) and introduces learnable prompts to enhance cross-modal alignment. By leveraging the extended context window and richer representations of LLMs, QwenCLIP captures comprehensive medical semantics from long-form clinical text, substantially improving medical image-text alignment and downstream performance on radiology benchmarks. Our code is publicly available at https://github.com/Wxy-24/QwenCLIP.

Paperid: 258, https://arxiv.org/pdf/2511.13269.pdf GitHub

Abstract:
Vision-Language Models (VLMs), leveraging their powerful visual perception and reasoning capabilities, have been widely applied in Unmanned Aerial Vehicle (UAV) tasks. However, the spatial intelligence capabilities of existing VLMs in UAV scenarios remain largely unexplored, raising concerns about their effectiveness in navigating and interpreting dynamic environments. To bridge this gap, we introduce SpatialSky-Bench, a comprehensive benchmark specifically designed to evaluate the spatial intelligence capabilities of VLMs in UAV navigation. Our benchmark comprises two categories-Environmental Perception and Scene Understanding-divided into 13 subcategories, including bounding boxes, color, distance, height, and landing safety analysis, among others. Extensive evaluations of various mainstream open-source and closed-source VLMs reveal unsatisfactory performance in complex UAV navigation scenarios, highlighting significant gaps in their spatial capabilities. To address this challenge, we developed the SpatialSky-Dataset, a comprehensive dataset containing 1M samples with diverse annotations across various scenarios. Leveraging this dataset, we introduce Sky-VLM, a specialized VLM designed for UAV spatial reasoning across multiple granularities and contexts. Extensive experimental results demonstrate that Sky-VLM achieves state-of-the-art performance across all benchmark tasks, paving the way for the development of VLMs suitable for UAV scenarios. The source code is available at https://github.com/linglingxiansen/SpatialSKy.

Paperid: 259, https://arxiv.org/pdf/2511.13261.pdf GitHub

Abstract:
Driven by recent advances in vision language models (VLMs) and egocentric perception research, we introduce the concept of an egocentric procedural AI assistant (EgoProceAssist) tailored to step-by-step support daily procedural tasks in a first-person view. In this work, we start by identifying three core tasks: egocentric procedural error detection, egocentric procedural learning, and egocentric procedural question answering. These tasks define the essential functions of EgoProceAssist within a new taxonomy. Specifically, our work encompasses a comprehensive review of current techniques, relevant datasets, and evaluation metrics across these three core areas. To clarify the gap between the proposed EgoProceAssist and existing VLM-based AI assistants, we introduce novel experiments and provide a comprehensive evaluation of representative VLM-based methods. Based on these findings and our technical analysis, we discuss the challenges ahead and suggest future research directions. Furthermore, an exhaustive list of this study is publicly available in an active repository that continuously collects the latest work: https://github.com/z1oong/Building-Egocentric-Procedural-AI-Assistant

Paperid: 260, https://arxiv.org/pdf/2511.13099.pdf GitHub

Abstract:
Lifelong learning on Whole Slide Images (WSIs) aims to train or fine-tune a unified model sequentially on cancer-related tasks, reducing the resources and effort required for data transfer and processing, especially given the gigabyte-scale size of WSIs. In this paper, we introduce MergeSlide, a simple yet effective framework that treats lifelong learning as a model merging problem by leveraging a vision-language pathology foundation model. When a new task arrives, it is: 1) defined with class-aware prompts, 2) fine-tuned for a few epochs using an MLP-free backbone, and 3) merged into a unified model using an orthogonal continual merging strategy that preserves performance and mitigates catastrophic forgetting. For inference under the class-incremental learning (CLASS-IL) setting, where task identity is unknown, we introduce Task-to-Class Prompt-aligned (TCP) inference. Specifically, TCP first identifies the most relevant task using task-level prompts and then applies the corresponding class-aware prompts to generate predictions. To evaluate MergeSlide, we conduct experiments on a stream of six TCGA datasets. The results show that MergeSlide outperforms both rehearsal-based continual learning and vision-language zero-shot baselines. Code and data are available at https://github.com/caodoanh2001/MergeSlide.

Paperid: 261, https://arxiv.org/pdf/2511.13087.pdf GitHub GitHub

Abstract:
Graphical User Interface (GUI) grounding - the task of mapping natural language instructions to screen coordinates - is essential for autonomous agents and accessibility technologies. Existing systems rely on monolithic models or one-shot pipelines that lack modularity and fail under visual clutter and ambiguous instructions. We introduce MEGA-GUI, a multi-stage framework that separates grounding into coarse Region-of-Interest (ROI) selection and fine-grained element grounding, orchestrated by specialized vision-language agents. MEGA-GUI features a bidirectional ROI zoom algorithm that mitigates spatial dilution and a context-aware rewriting agent that reduces semantic ambiguity. Our analysis reveals complementary strengths and weaknesses across vision-language models at different visual scales, and we show that leveraging this modular structure achieves consistently higher accuracy than monolithic approaches. On the visually dense ScreenSpot-Pro benchmark, MEGA-GUI attains 73.18% accuracy, and on the semantically complex OSWorld-G benchmark it reaches 68.63%, surpassing previously reported results. Code and the Grounding Benchmark Toolkit (GBT) are available at https://github.com/samsungsds-research-papers/mega-gui.

Paperid: 262, https://arxiv.org/pdf/2511.13036.pdf GitHub

Abstract:
Contrastive Language-Image Pre-training (CLIP) has demonstrated strong generalization across a wide range of visual tasks by leveraging large-scale English-image pairs. However, its extension to low-resource languages remains limited due to the scarcity of high-quality multilingual image-text data. Existing multilingual vision-language models exhibit consistently low retrieval performance in underrepresented languages including Czech, Finnish, Croatian, Hungarian, and Romanian on the Crossmodal-3600 (XM3600) benchmark. To address this, we propose a lightweight and data-efficient framework for multilingual vision-language alignment. Our approach requires no image-text pairs or text-text pairs and freezes both the pretrained image encoder and multilingual text encoder during training. Only a compact 1.7M-parameter projection module is trained, using a contrastive loss over English representations as semantic anchors. This minimal training setup enables robust multilingual alignment even for languages with limited supervision. Extensive evaluation across multiple multilingual retrieval benchmarks confirms the effectiveness of our method, showing significant gains in five underrepresented languages where existing models typically underperform. These findings highlight the effectiveness of our pivot-based, parameter-efficient alignment strategy for inclusive multimodal learning.

Paperid: 263, https://arxiv.org/pdf/2511.12878.pdf GitHub

Abstract:
Analyzing hand-object interaction in egocentric vision facilitates VR/AR applications and human-robot policy transfer. Existing research has mostly focused on modeling the behavior paradigm of interactive actions (i.e., "how to interact"). However, the more challenging and fine-grained problem of capturing the critical moments of contact and separation between the hand and the target object (i.e., "when to interact") is still underexplored, which is crucial for immersive interactive experiences in mixed reality and robotic motion planning. Therefore, we formulate this problem as temporal interaction localization (TIL). Some recent works extract semantic masks as TIL references, but suffer from inaccurate object grounding and cluttered scenarios. Although current temporal action localization (TAL) methods perform well in detecting verb-noun action segments, they rely on category annotations during training and exhibit limited precision in localizing hand-object contact/separation moments. To address these issues, we propose a novel zero-shot approach dubbed EgoLoc to localize hand-object contact and separation timestamps in egocentric videos. EgoLoc introduces hand-dynamics-guided sampling to generate high-quality visual prompts. It exploits the vision-language model to identify contact/separation attributes, localize specific timestamps, and provide closed-loop feedback for further refinement. EgoLoc eliminates the need for object masks and verb-noun taxonomies, leading to generalizable zero-shot implementation. Comprehensive experiments on the public dataset and our novel benchmarks demonstrate that EgoLoc achieves plausible TIL for egocentric videos. It is also validated to effectively facilitate multiple downstream applications in egocentric vision and robotic manipulation tasks. Code and relevant data will be released at https://github.com/IRMVLab/EgoLoc.

Paperid: 264, https://arxiv.org/pdf/2511.12693.pdf GitHub

Abstract:
Vision-language models (VLMs) enable open-ended visual question answering but remain prone to hallucinations. We present HEDGE, a unified framework for hallucination detection that combines controlled visual perturbations, semantic clustering, and robust uncertainty metrics. HEDGE integrates sampling, distortion synthesis, clustering (entailment- and embedding-based), and metric computation into a reproducible pipeline applicable across multimodal architectures. Evaluations on VQA-RAD and KvasirVQA-x1 with three representative VLMs (LLaVA-Med, Med-Gemma, Qwen2.5-VL) reveal clear architecture- and prompt-dependent trends. Hallucination detectability is highest for unified-fusion models with dense visual tokenization (Qwen2.5-VL) and lowest for architectures with restricted tokenization (Med-Gemma). Embedding-based clustering often yields stronger separation when applied directly to the generated answers, whereas NLI-based clustering remains advantageous for LLaVA-Med and for longer, sentence-level responses. Across configurations, the VASE metric consistently provides the most robust hallucination signal, especially when paired with embedding clustering and a moderate sampling budget (n ~ 10-15). Prompt design also matters: concise, label-style outputs offer clearer semantic structure than syntactically constrained one-sentence responses. By framing hallucination detection as a geometric robustness problem shaped jointly by sampling scale, prompt structure, model architecture, and clustering strategy, HEDGE provides a principled, compute-aware foundation for evaluating multimodal reliability. The hedge-bench PyPI library enables reproducible and extensible benchmarking, with full code and experimental resources available at https://github.com/Simula/HEDGE .

Paperid: 265, https://arxiv.org/pdf/2511.12639.pdf GitHub

Abstract:
Vision-language foundation models (VLMs) have shown great potential in feature transfer and generalization across a wide spectrum of medical-related downstream tasks. However, fine-tuning these models is resource-intensive due to their large number of parameters. Prompt tuning has emerged as a viable solution to mitigate memory usage and reduce training time while maintaining competitive performance. Nevertheless, the challenge is that existing prompt tuning methods cannot precisely distinguish different kinds of medical concepts, which miss essentially specific disease-related features across various medical imaging modalities in medical image classification tasks. We find that Large Language Models (LLMs), trained on extensive text corpora, are particularly adept at providing this specialized medical knowledge. Motivated by this, we propose incorporating LLMs into the prompt tuning process. Specifically, we introduce the CILMP, Conditional Intervention of Large Language Models for Prompt Tuning, a method that bridges LLMs and VLMs to facilitate the transfer of medical knowledge into VLM prompts. CILMP extracts disease-specific representations from LLMs, intervenes within a low-rank linear subspace, and utilizes them to create disease-specific prompts. Additionally, a conditional mechanism is incorporated to condition the intervention process on each individual medical image, generating instance-adaptive prompts and thus enhancing adaptability. Extensive experiments across diverse medical image datasets demonstrate that CILMP consistently outperforms state-of-the-art prompt tuning methods, demonstrating its effectiveness. Code is available at https://github.com/usr922/cilmp.

Paperid: 266, https://arxiv.org/pdf/2511.12530.pdf GitHub

Abstract:
Keyframe selection has become essential for video understanding with vision-language models (VLMs) due to limited input tokens and the temporal sparsity of relevant information across video frames. Video understanding often relies on effective keyframes that are not only informative but also causally decisive. To this end, we propose Reinforced Causal Search with Information Bottleneck (ReaSon), a framework that formulates keyframe selection as an optimization problem with the help of a novel Causal Information Bottleneck (CIB), which explicitly defines keyframes as those satisfying both predictive sufficiency and causal necessity. Specifically, ReaSon employs a learnable policy network to select keyframes from a visually relevant pool of candidate frames to capture predictive sufficiency, and then assesses causal necessity via counterfactual interventions. Finally, a composite reward aligned with the CIB principle is designed to guide the selection policy through reinforcement learning. Extensive experiments on NExT-QA, EgoSchema, and Video-MME demonstrate that ReaSon consistently outperforms existing state-of-the-art methods under limited-frame settings, validating its effectiveness and generalization ability.

Paperid: 267, https://arxiv.org/pdf/2511.11898.pdf GitHub

Abstract:
Vision-language foundation models (VLMs) show promise for diverse imaging tasks but often underperform on medical benchmarks. Prior efforts to improve performance include model finetuning, which requires large domain-specific datasets and significant compute, or manual prompt engineering, which is hard to generalize and often inaccessible to medical institutions seeking to deploy these tools. These challenges motivate interest in approaches that draw on a model's embedded knowledge while abstracting away dependence on human-designed prompts to enable scalable, weight-agnostic performance improvements. To explore this, we adapt the Declarative Self-improving Python (DSPy) framework for structured automated prompt optimization in medical vision-language systems through a comprehensive, formal evaluation. We implement prompting pipelines for five medical imaging tasks across radiology, gastroenterology, and dermatology, evaluating 10 open-source VLMs with four prompt optimization techniques. Optimized pipelines achieved a median relative improvement of 53% over zero-shot prompting baselines, with the largest gains ranging from 300% to 3,400% on tasks where zero-shot performance is low. These results highlight the substantial potential of applying automated prompt optimization to medical AI systems, demonstrating significant gains for vision-based applications requiring accurate clinical image interpretation. By reducing dependence on prompt design to elicit intended outputs, these techniques allow clinicians to focus on patient care and clinical decision-making. Furthermore, our experiments offer scalability and preserve data privacy, demonstrating performance improvement on open-source VLMs. We publicly release our evaluation pipelines to support reproducible research on specialized medical tasks, available at https://github.com/DaneshjouLab/prompt-triage-lab.

Paperid: 268, https://arxiv.org/pdf/2511.11831.pdf GitHub

Abstract:
Large Vision-Language Models (LVLMs) typically align visual features from an encoder with a pre-trained Large Language Model (LLM). However, this makes the visual perception module a bottleneck, which constrains the overall capabilities of LVLMs. Conventional evaluation benchmarks, while rich in visual semantics, often contain unavoidable local shortcuts that can lead to an overestimation of models' perceptual abilities. Here, we introduce TopoPerception, a benchmark that leverages topological properties to rigorously evaluate the global visual perception capabilities of LVLMs across various granularities. Since topology depends on the global structure of an image and is invariant to local features, TopoPerception enables a shortcut-free assessment of global perception, fundamentally distinguishing it from semantically rich tasks. We evaluate state-of-the-art models on TopoPerception and find that even at the coarsest perceptual granularity, all models perform no better than random chance, indicating a profound inability to perceive global visual features. Notably, a consistent trend emerge within model families: more powerful models with stronger reasoning capabilities exhibit lower accuracy. This suggests that merely scaling up models is insufficient to address this deficit and may even exacerbate it. Progress may require new training paradigms or architectures. TopoPerception not only exposes a critical bottleneck in current LVLMs but also offers a lens and direction for improving their global visual perception. The data and code are publicly available at: https://github.com/Wenhao-Zhou/TopoPerception.

Paperid: 269, https://arxiv.org/pdf/2511.11526.pdf GitHub

Abstract:
Vision-Language Models (VLMs) are a new family of models that align image content with natural language. Existing approaches typically fuse either (a) early: by mixing tokens/features inside the encoders, or (b) late: by comparing pooled embeddings. Many methods also tie fusion to an autoregressive decoder. However, the hidden states of both modalities already carry rich, modality-specific structure (spatial layout in vision; syntax and semantics in text), so directly aligning these states is a natural way to match what the two modalities "think". We propose a lightweight fusion module: a few cross-only, bidirectional attention layers placed near the top of both encoders. Each layer projects the vision and text encoder hidden-state sequences into a shared space, attends across modalities, and sends gated residual updates back, with simple stabilizers to improve alignment. The encoders remain non-causal and strong for understanding, while generation stays cleanly decoupled via an optional decoder. Across standard retrieval, VQA, and visual reasoning benchmarks, BRIDGE outperforms comparable VLMs while preserving the bi-encoder efficiency of contrastive models. We make our code publicly available at https://github.com/jfeinashley/BRIDGE.

Paperid: 270, https://arxiv.org/pdf/2511.11266.pdf GitHub

Abstract:
Vision-language models have recently emerged as promising planners for autonomous driving, where success hinges on topology-aware reasoning over spatial structure and dynamic interactions from multimodal input. However, existing models are typically trained without supervision that explicitly encodes these relational dependencies, limiting their ability to infer how agents and other traffic entities influence one another from raw sensor data. In this work, we bridge this gap with a novel model-agnostic method that conditions language-based driving models on structured relational context in the form of traffic scene graphs. We serialize scene graphs at various abstraction levels and formats, and incorporate them into the models via structured prompt templates, enabling a systematic analysis of when and how relational supervision is most beneficial. Extensive evaluations on the public LangAuto benchmark show that scene graph conditioning of state-of-the-art approaches yields large and persistent improvement in driving performance. Notably, we observe up to a 15.6\% increase in driving score for LMDrive and 17.5\% for BEVDriver, indicating that models can better internalize and ground relational priors through scene graph-conditioned training, even without requiring scene graph input at test-time. Code, fine-tuned models, and our scene graph dataset are publicly available at https://github.com/iis-esslingen/GraphPilot.

Paperid: 271, https://arxiv.org/pdf/2511.11007.pdf GitHub

Abstract:
Despite the remarkable success of Vision-Language Models (VLMs), their performance on a range of complex visual tasks is often hindered by a "visual processing bottleneck": a propensity to lose grounding in visual evidence and exhibit a deficit in contextualized visual experience during prolonged generation. Drawing inspiration from human cognitive memory theory, which distinguishes short-term visually-dominant memory and long-term semantically-dominant memory, we propose VisMem, a cognitively-aligned framework that equips VLMs with dynamic latent vision memories, a short-term module for fine-grained perceptual retention and a long-term module for abstract semantic consolidation. These memories are seamlessly invoked during inference, allowing VLMs to maintain both perceptual fidelity and semantic consistency across thinking and generation. Extensive experiments across diverse visual benchmarks for understanding, reasoning, and generation reveal that VisMem delivers a significant average performance boost of 11.8% relative to the vanilla model and outperforms all counterparts, establishing a new paradigm for latent-space memory enhancement. The code will be available: https://github.com/YU-deep/VisMem.git.

Paperid: 272, https://arxiv.org/pdf/2511.10701.pdf GitHub

Abstract:
CAR-Scenes is a frame-level dataset for autonomous driving that enables training and evaluation of vision-language models (VLMs) for interpretable, scene-level understanding. We annotate 5,192 images drawn from Argoverse 1, Cityscapes, KITTI, and nuScenes using a 28-key category/sub-category knowledge base covering environment, road geometry, background-vehicle behavior, ego-vehicle behavior, vulnerable road users, sensor states, and a discrete severity scale (1-10), totaling 350+ leaf attributes. Labels are produced by a GPT-4o-assisted vision-language pipeline with human-in-the-loop verification; we release the exact prompts, post-processing rules, and per-field baseline model performance. CAR-Scenes also provides attribute co-occurrence graphs and JSONL records that support semantic retrieval, dataset triage, and risk-aware scenario mining across sources. To calibrate task difficulty, we include reproducible, non-benchmark baselines, notably a LoRA-tuned Qwen2-VL-2B with deterministic decoding, evaluated via scalar accuracy, micro-averaged F1 for list attributes, and severity MAE/RMSE on a fixed validation split. We publicly release the annotation and analysis scripts, including graph construction and evaluation scripts, to enable explainable, data-centric workflows for future intelligent vehicles. Dataset: https://github.com/Croquembouche/CAR-Scenes

Paperid: 273, https://arxiv.org/pdf/2511.10560.pdf GitHub

Abstract:
General 3D foundation models have started to lead the trend of unifying diverse vision tasks, yet most assume RGB-only inputs and ignore readily available geometric cues (e.g., camera intrinsics, poses, and depth maps). To address this issue, we introduce OmniVGGT, a novel framework that can effectively benefit from an arbitrary number of auxiliary geometric modalities during both training and inference. In our framework, a GeoAdapter is proposed to encode depth and camera intrinsics/extrinsics into a spatial foundation model. It employs zero-initialized convolutions to progressively inject geometric information without disrupting the foundation model's representation space. This design ensures stable optimization with negligible overhead, maintaining inference speed comparable to VGGT even with multiple additional inputs. Additionally, a stochastic multimodal fusion regimen is proposed, which randomly samples modality subsets per instance during training. This enables an arbitrary number of modality inputs during testing and promotes learning robust spatial representations instead of overfitting to auxiliary cues. Comprehensive experiments on monocular/multi-view depth estimation, multi-view stereo, and camera pose estimation demonstrate that OmniVGGT outperforms prior methods with auxiliary inputs and achieves state-of-the-art results even with RGB-only input. To further highlight its practical utility, we integrated OmniVGGT into vision-language-action (VLA) models. The enhanced VLA model by OmniVGGT not only outperforms the vanilla point-cloud-based baseline on mainstream benchmarks, but also effectively leverages accessible auxiliary inputs to achieve consistent gains on robotic tasks.

Paperid: 274, https://arxiv.org/pdf/2511.10518.pdf GitHub

Abstract:
Vision-Language-Action (VLA) models have advanced in robotic manipulation, yet practical deployment remains hindered by two key limitations: 1) perceptual redundancy, where irrelevant visual inputs are processed inefficiently, and 2) superficial instruction-vision alignment, which hampers semantic grounding of actions. In this paper, we propose SemanticVLA, a novel VLA framework that performs Semantic-Aligned Sparsification and Enhancement for Efficient Robotic Manipulation. Specifically: 1) To sparsify redundant perception while preserving semantic alignment, Semantic-guided Dual Visual Pruner (SD-Pruner) performs: Instruction-driven Pruner (ID-Pruner) extracts global action cues and local semantic anchors in SigLIP; Spatial-aggregation Pruner (SA-Pruner) compacts geometry-rich features into task-adaptive tokens in DINOv2. 2) To exploit sparsified features and integrate semantics with spatial geometry, Semantic-complementary Hierarchical Fuser (SH-Fuser) fuses dense patches and sparse tokens across SigLIP and DINOv2 for coherent representation. 3) To enhance the transformation from perception to action, Semantic-conditioned Action Coupler (SA-Coupler) replaces the conventional observation-to-DoF approach, yielding more efficient and interpretable behavior modeling for manipulation tasks. Extensive experiments on simulation and real-world tasks show that SemanticVLA sets a new SOTA in both performance and efficiency. SemanticVLA surpasses OpenVLA on LIBERO benchmark by 21.1% in success rate, while reducing training cost and inference latency by 3.0-fold and 2.7-fold.SemanticVLA is open-sourced and publicly available at https://github.com/JiuTian-VL/SemanticVLA

Paperid: 275, https://arxiv.org/pdf/2511.10390.pdf GitHub

Abstract:
Document parsing is a core task in document intelligence, supporting applications such as information extraction, retrieval-augmented generation, and automated document analysis. However, real-world documents often feature complex layouts with multi-level tables, embedded images or formulas, and cross-page structures, which remain challenging for existing OCR systems. We introduce MonkeyOCR v1.5, a unified vision-language framework that enhances both layout understanding and content recognition through a two-stage pipeline. The first stage employs a large multimodal model to jointly predict layout and reading order, leveraging visual information to ensure sequential consistency. The second stage performs localized recognition of text, formulas, and tables within detected regions, maintaining high visual fidelity while reducing error propagation. To address complex table structures, we propose a visual consistency-based reinforcement learning scheme that evaluates recognition quality via render-and-compare alignment, improving structural accuracy without manual annotations. Additionally, two specialized modules, Image-Decoupled Table Parsing and Type-Guided Table Merging, are introduced to enable reliable parsing of tables containing embedded images and reconstruction of tables crossing pages or columns. Comprehensive experiments on OmniDocBench v1.5 demonstrate that MonkeyOCR v1.5 achieves state-of-the-art performance, outperforming PPOCR-VL and MinerU 2.5 while showing exceptional robustness in visually complex document scenarios. A trial link can be found at https://github.com/Yuliang-Liu/MonkeyOCR .

Paperid: 276, https://arxiv.org/pdf/2511.10279.pdf GitHub

Abstract:
Despite significant progress, Vision-Language Models (VLMs) still struggle with complex visual reasoning, where multi-step dependencies cause early errors to cascade through the reasoning chain. Existing post-training paradigms are limited: Supervised Fine-Tuning (SFT) relies on costly step-level annotations, while Reinforcement Learning with Verifiable Rewards (RLVR) methods like GRPO provide only sparse, outcome-level feedback, hindering stable optimization. We introduce PROPA (Process-level Reasoning Optimization with interleaved Policy Alignment), a novel framework that integrates Monte Carlo Tree Search (MCTS) with GRPO to generate dense, process-level rewards and optimize reasoning at each intermediate step without human annotations. To overcome the cold-start problem, PROPA interleaves GRPO updates with SFT, enabling the model to learn from both successful and failed reasoning trajectories. A Process Reward Model (PRM) is further trained to guide inference-time search, aligning the test-time search with the training signal. Across seven benchmarks and four VLM backbones, PROPA consistently outperforms both SFT- and RLVR-based baselines. It achieves up to 17.0% gains on in-domain tasks and 21.0% gains on out-of-domain tasks compared to existing state-of-the-art, establishing a strong reasoning and generalization capability for visual reasoning tasks. The code isavailable at: https://github.com/YanbeiJiang/PROPA.

Paperid: 277, https://arxiv.org/pdf/2511.10098.pdf GitHub

Abstract:
Recent advances in Large Visual Language Models (LVLMs) have demonstrated impressive performance across various vision-language tasks by leveraging large-scale image-text pretraining and instruction tuning. However, the security vulnerabilities of LVLMs have become increasingly concerning, particularly their susceptibility to backdoor attacks. Existing backdoor attacks focus on single-target attacks, i.e., targeting a single malicious output associated with a specific trigger. In this work, we uncover multi-target backdoor attacks, where multiple independent triggers corresponding to different attack targets are added in a single pass of training, posing a greater threat to LVLMs in real-world applications. Executing such attacks in LVLMs is challenging since there can be many incorrect trigger-target mappings due to severe feature interference among different triggers. To address this challenge, we propose MTAttack, the first multi-target backdoor attack framework for enforcing accurate multiple trigger-target mappings in LVLMs. The core of MTAttack is a novel optimization method with two constraints, namely Proxy Space Partitioning constraint and Trigger Prototype Anchoring constraint. It jointly optimizes multiple triggers in the latent space, with each trigger independently mapping clean images to a unique proxy class while at the same time guaranteeing their separability. Experiments on popular benchmarks demonstrate a high success rate of MTAttack for multi-target attacks, substantially outperforming existing attack methods. Furthermore, our attack exhibits strong generalizability across datasets and robustness against backdoor defense strategies. These findings highlight the vulnerability of LVLMs to multi-target backdoor attacks and underscore the urgent need for mitigating such threats. Code is available at https://github.com/mala-lab/MTAttack.

Paperid: 278, https://arxiv.org/pdf/2511.09868.pdf GitHub

Abstract:
Large Vision-Language Models (LVLMs) have achieved impressive performance across a wide range of multimodal tasks. However, they still face critical challenges in modeling long-range dependencies under the usage of Rotary Positional Encoding (ROPE). Although it can facilitate precise modeling of token positions, it induces progressive attention decay as token distance increases, especially with progressive attention decay over distant token pairs, which severely impairs the model's ability to remember global context. To alleviate this issue, we propose inference-only Three-step Decay Resilience Strategies (T-DRS), comprising (1) Semantic-Driven DRS (SD-DRS), amplifying semantically meaningful but distant signals via content-aware residuals, (2) Distance-aware Control DRS (DC-DRS), which can purify attention by smoothly modulating weights based on positional distances, suppressing noise while preserving locality, and (3) re-Reinforce Distant DRS (reRD-DRS), consolidating the remaining informative remote dependencies to maintain global coherence. Together, the T-DRS recover suppressed long-range token pairs without harming local inductive biases. Extensive experiments on Vision Question Answering (VQA) benchmarks demonstrate that T-DRS can consistently improve performance in a training-free manner. The code can be accessed in https://github.com/labixiaoq-qq/Remember-me

Paperid: 279, https://arxiv.org/pdf/2511.09809.pdf GitHub

Abstract:
Vision-Language Models (VLMs) excel at zero-shot inference but often degrade under test-time domain shifts. For this reason, episodic test-time adaptation strategies have recently emerged as powerful techniques for adapting VLMs to a single unlabeled image. However, existing adaptation strategies, such as test-time prompt tuning, typically require backpropagating through large encoder weights or altering core model components. In this work, we introduce Spectrum-Aware Test-Time Steering (STS), a lightweight adaptation framework that extracts a spectral subspace from the textual embeddings to define principal semantic directions and learns to steer latent representations in a spectrum-aware manner by adapting a small number of per-sample shift parameters to minimize entropy across augmented views. STS operates entirely at inference in the latent space, without backpropagation through or modification of the frozen encoders. Building on standard evaluation protocols, our comprehensive experiments demonstrate that STS largely surpasses or compares favorably against state-of-the-art test-time adaptation methods, while introducing only a handful of additional parameters and achieving inference speeds up to 8x faster with a 12x smaller memory footprint than conventional test-time prompt tuning. The code is available at https://github.com/kdafnis/STS.

Paperid: 280, https://arxiv.org/pdf/2511.09554.pdf GitHub

Abstract:
Open-vocabulary detectors achieve impressive performance on COCO, but often fail to generalize to real-world datasets with out-of-distribution classes not typically found in their pre-training. Rather than simply fine-tuning a heavy-weight vision-language model (VLM) for new domains, we introduce RF-DETR, a light-weight specialist detection transformer that discovers accuracy-latency Pareto curves for any target dataset with weight-sharing neural architecture search (NAS). Our approach fine-tunes a pre-trained base network on a target dataset and evaluates thousands of network configurations with different accuracy-latency tradeoffs without re-training. Further, we revisit the "tunable knobs" for NAS to improve the transferability of DETRs to diverse target domains. Notably, RF-DETR significantly improves on prior state-of-the-art real-time methods on COCO and Roboflow100-VL. RF-DETR (nano) achieves 48.0 AP on COCO, beating D-FINE (nano) by 5.3 AP at similar latency, and RF-DETR (2x-large) outperforms GroundingDINO (tiny) by 1.2 AP on Roboflow100-VL while running 20x as fast. To the best of our knowledge, RF-DETR (2x-large) is the first real-time detector to surpass 60 AP on COCO. Our code is at https://github.com/roboflow/rf-detr

Paperid: 281, https://arxiv.org/pdf/2511.09540.pdf GitHub

Abstract:
Recent advances in context optimization (CoOp) guided by large language model (LLM)-distilled medical semantic priors offer a scalable alternative to manual prompt engineering and full fine-tuning for adapting biomedical CLIP-based vision-language models (VLMs). However, prompt learning in this context is challenged by semantic misalignment between LLMs and CLIP variants due to divergent training corpora and model architectures; it further lacks scalability across continuously evolving families of foundation models. More critically, pairwise multimodal alignment via conventional Euclidean-space optimization lacks the capacity to model unified representations or apply localized geometric constraints, which tends to amplify modality gaps in complex biomedical imaging and destabilize few-shot adaptation. In this work, we propose vMFCoOp, a framework that inversely estimates von Mises-Fisher (vMF) distributions on a shared Hyperspherical Manifold, aligning semantic biases between arbitrary LLMs and CLIP backbones via Unified Semantic Anchors to achieve robust biomedical prompting and superior few-shot classification. Grounded in three complementary constraints, vMFCoOp demonstrates consistent improvements across 14 medical datasets, 12 medical imaging modalities, and 13 anatomical regions, outperforming state-of-the-art methods in accuracy, generalization, and clinical applicability. This work aims to continuously expand to encompass more downstream applications, and the corresponding resources are intended to be shared through https://github.com/VinyehShaw/UniEqui.

Paperid: 282, https://arxiv.org/pdf/2511.09483.pdf GitHub GitHub

Abstract:
We present CrochetBench, a benchmark for evaluating the ability of multimodal large language models to perform fine-grained, low-level procedural reasoning in the domain of crochet. Unlike prior benchmarks that focus on high-level description or visual question answering, CrochetBench shifts the emphasis from describing to doing: models are required to recognize stitches, select structurally appropriate instructions, and generate compilable crochet procedures. We adopt the CrochetPARADE DSL as our intermediate representation, enabling structural validation and functional evaluation via execution. The benchmark covers tasks including stitch classification, instruction grounding, and both natural language and image-to-DSL translation. Across all tasks, performance sharply declines as the evaluation shifts from surface-level similarity to executable correctness, exposing limitations in long-range symbolic reasoning and 3D-aware procedural synthesis. CrochetBench offers a new lens for assessing procedural competence in multimodal models and highlights the gap between surface-level understanding and executable precision in real-world creative domains. Code is available at https://github.com/Peiyu-Georgia-Li/crochetBench.

Paperid: 283, https://arxiv.org/pdf/2511.09064.pdf GitHub

Abstract:
Vision-language pre-training models (VLPs) demonstrate strong multimodal understanding and zero-shot generalization, yet remain vulnerable to adversarial examples, raising concerns about their reliability. Recent work, Test-Time Counterattack (TTC), improves robustness by generating perturbations that maximize the embedding deviation of adversarial inputs using PGD, pushing them away from their adversarial representations. However, due to the fundamental difference in optimization objectives between adversarial attacks and counterattacks, generating counterattacks solely based on gradients with respect to the adversarial input confines the search to a narrow space. As a result, the counterattacks could overfit limited adversarial patterns and lack the diversity to fully neutralize a broad range of perturbations. In this work, we argue that enhancing the diversity and coverage of counterattacks is crucial to improving adversarial robustness in test-time defense. Accordingly, we propose Directional Orthogonal Counterattack (DOC), which augments counterattack optimization by incorporating orthogonal gradient directions and momentum-based updates. This design expands the exploration of the counterattack space and increases the diversity of perturbations, which facilitates the discovery of more generalizable counterattacks and ultimately improves the ability to neutralize adversarial perturbations. Meanwhile, we present a directional sensitivity score based on averaged cosine similarity to boost DOC by improving example discrimination and adaptively modulating the counterattack strength. Extensive experiments on 16 datasets demonstrate that DOC improves adversarial robustness under various attacks while maintaining competitive clean accuracy. Code is available at https://github.com/bookman233/DOC.

Paperid: 284, https://arxiv.org/pdf/2511.09063.pdf GitHub

Abstract:
Vision-Language Models (VLMs), with their powerful content generation capabilities, have been successfully applied to data annotation processes. However, the VLM-generated labels exhibit dual limitations: low quality (i.e., label noise) and absence of error correction mechanisms. To enhance label quality, we propose Human-Corrected Labels (HCLs), a novel setting that efficient human correction for VLM-generated noisy labels. As shown in Figure 1(b), HCL strategically deploys human correction only for instances with VLM discrepancies, achieving both higher-quality annotations and reduced labor costs. Specifically, we theoretically derive a risk-consistent estimator that incorporates both human-corrected labels and VLM predictions to train classifiers. Besides, we further propose a conditional probability method to estimate the label distribution using a combination of VLM outputs and model predictions. Extensive experiments demonstrate that our approach achieves superior classification performance and is robust to label noise, validating the effectiveness of HCL in practical weak supervision scenarios. Code https://github.com/Lilianach24/HCL.git

Paperid: 285, https://arxiv.org/pdf/2511.09018.pdf GitHub

Abstract:
Object hallucination remains a critical challenge in Large Vision-Language Models (LVLMs), where models generate content inconsistent with visual inputs. Existing language-decoder based mitigation approaches often regulate visual or textual attention independently, overlooking their interaction as two key causal factors. To address this, we propose Owl (Bi-mOdal attention reWeighting for Layer-wise hallucination mitigation), a causally-grounded framework that models hallucination process via a structural causal graph, treating decomposed visual and textual attentions as mediators. We introduce VTACR (Visual-to-Textual Attention Contribution Ratio), a novel metric that quantifies the modality contribution imbalance during decoding. Our analysis reveals that hallucinations frequently occur in low-VTACR scenarios, where textual priors dominate and visual grounding is weakened. To mitigate this, we design a fine-grained attention intervention mechanism that dynamically adjusts token- and layer-wise attention guided by VTACR signals. Finally, we propose a dual-path contrastive decoding strategy: one path emphasizes visually grounded predictions, while the other amplifies hallucinated ones -- letting visual truth shine and hallucination collapse. Experimental results on the POPE and CHAIR benchmarks show that Owl achieves significant hallucination reduction, setting a new SOTA in faithfulness while preserving vision-language understanding capability. Our code is available at https://github.com/CikZ2023/OWL

Paperid: 286, https://arxiv.org/pdf/2511.08914.pdf GitHub

Abstract:
Deploying Vision-Language Models (VLMs) on edge devices (e.g., smartphones and robots) is crucial for enabling low-latency and privacy-preserving intelligent applications. Given the resource constraints of these devices, quantization offers a promising solution by improving memory efficiency and reducing bandwidth requirements, thereby facilitating the deployment of VLMs. However, existing research has rarely explored aggressive quantization on VLMs, particularly for the models ranging from 1B to 2B parameters, which are more suitable for resource-constrained edge devices. In this paper, we propose SPEED-Q, a novel Staged Processing with Enhanced Distillation framework for VLM low-bit weight-only quantization that systematically addresses the following two critical obstacles: (1) significant discrepancies in quantization sensitivity between vision (ViT) and language (LLM) components in VLMs; (2) training instability arising from the reduced numerical precision inherent in low-bit quantization. In SPEED-Q, a staged sensitivity adaptive mechanism is introduced to effectively harmonize performance across different modalities. We further propose a distillation-enhanced quantization strategy to stabilize the training process and reduce data dependence. Together, SPEED-Q enables accurate, stable, and data-efficient quantization of complex VLMs. SPEED-Q is the first framework tailored for quantizing entire small-scale billion-parameter VLMs to low bits. Extensive experiments across multiple benchmarks demonstrate that SPEED-Q achieves up to 6x higher accuracy than existing quantization methods under 2-bit settings and consistently outperforms prior on-device VLMs under both 2-bit and 4-bit settings. Our code and models are available at https://github.com/antgroup/SPEED-Q.

Paperid: 287, https://arxiv.org/pdf/2511.08173.pdf GitHub

Abstract:
Detecting visual anomalies in diverse, multi-class real-world images is a significant challenge. We introduce \ours, a novel unsupervised multi-class visual anomaly detection framework. It integrates a Latent Diffusion Model (LDM) with a Vision-Language Model (VLM) for enhanced anomaly localization and detection. Specifically, a pre-trained VLM with a simple prompt extracts detailed image descriptions, serving as additional conditioning for LDM training. Current diffusion-based methods rely on synthetic noise generation, limiting their generalization and requiring per-class model training, which hinders scalability. \ours, however, leverages VLMs to obtain normal captions without manual annotations or additional training. These descriptions condition the diffusion model, learning a robust normal image feature representation for multi-class anomaly detection. Our method achieves competitive performance, improving the pixel-level Per-Region-Overlap (PRO) metric by up to 25 points on the Real-IAD dataset and 8 points on the COCO-AD dataset, outperforming state-of-the-art diffusion-based approaches. Code is available at https://github.com/giddyyupp/VLMDiff.

Paperid: 288, https://arxiv.org/pdf/2511.07966.pdf GitHub

Abstract:
Unsupervised domain adaptation for LiDAR-based 3D object detection (3D UDA) based on the teacher-student architecture with pseudo labels has achieved notable improvements in recent years. Although it is quite popular to collect point clouds and images simultaneously, little attention has been paid to the usefulness of image data in 3D UDA when training the models. In this paper, we propose an approach named MMAssist that improves the performance of 3D UDA with multi-modal assistance. A method is designed to align 3D features between the source domain and the target domain by using image and text features as bridges. More specifically, we project the ground truth labels or pseudo labels to the images to get a set of 2D bounding boxes. For each 2D box, we extract its image feature from a pre-trained vision backbone. A large vision-language model (LVLM) is adopted to extract the box's text description, and a pre-trained text encoder is used to obtain its text feature. During the training of the model in the source domain and the student model in the target domain, we align the 3D features of the predicted boxes with their corresponding image and text features, and the 3D features and the aligned features are fused with learned weights for the final prediction. The features between the student branch and the teacher branch in the target domain are aligned as well. To enhance the pseudo labels, we use an off-the-shelf 2D object detector to generate 2D bounding boxes from images and estimate their corresponding 3D boxes with the aid of point cloud, and these 3D boxes are combined with the pseudo labels generated by the teacher model. Experimental results show that our approach achieves promising performance compared with state-of-the-art methods in three domain adaptation tasks on three popular 3D object detection datasets. The code is available at https://github.com/liangp/MMAssist.

Paperid: 289, https://arxiv.org/pdf/2511.07929.pdf GitHub

Abstract:
Despite the remarkable performance of deep models in medical imaging, they still require source data for training, which limits their potential in light of privacy concerns. Federated learning (FL), as a decentralized learning framework that trains a shared model with multiple hospitals (a.k.a., FL clients), provides a feasible solution. However, data heterogeneity and resource costs hinder the deployment of FL models, especially when using vision language models (VLM). To address these challenges, we propose a novel contrastive language-image pre-training (CLIP) based FL approach for medical image classification (FedMedCLIP). Specifically, we introduce a masked feature adaptation module (FAM) as a communication module to reduce the communication load while freezing the CLIP encoders to reduce the computational overhead. Furthermore, we propose a masked multi-layer perceptron (MLP) as a private local classifier to adapt to the client tasks. Moreover, we design an adaptive Kullback-Leibler (KL) divergence-based distillation regularization method to enable mutual learning between FAM and MLP. Finally, we incorporate model compression to transmit the FAM parameters while using ensemble predictions for classification. Extensive experiments on four publicly available medical datasets demonstrate that our model provides feasible performance (e.g., 8\% higher compared to second best baseline on ISIC2019) with reasonable resource cost (e.g., 120$\times$ faster than FedAVG).

Paperid: 290, https://arxiv.org/pdf/2511.07820.pdf GitHub

Abstract:
Despite the rise of billion-parameter foundation models trained across thousands of GPUs, similar scaling gains have not been shown for humanoid control. Current neural controllers for humanoids remain modest in size, target a limited behavior set, and are trained on a handful of GPUs over several days. We show that scaling up model capacity, data, and compute yields a generalist humanoid controller capable of creating natural and robust whole-body movements. Specifically, we posit motion tracking as a natural and scalable task for humanoid control, leverageing dense supervision from diverse motion-capture data to acquire human motion priors without manual reward engineering. We build a foundation model for motion tracking by scaling along three axes: network size (from 1.2M to 42M parameters), dataset volume (over 100M frames, 700 hours of high-quality motion data), and compute (9k GPU hours). Beyond demonstrating the benefits of scale, we show the practical utility of our model through two mechanisms: (1) a real-time universal kinematic planner that bridges motion tracking to downstream task execution, enabling natural and interactive control, and (2) a unified token space that supports various motion input interfaces, such as VR teleoperation devices, human videos, and vision-language-action (VLA) models, all using the same policy. Scaling motion tracking exhibits favorable properties: performance improves steadily with increased compute and data diversity, and learned representations generalize to unseen motions, establishing motion tracking at scale as a practical foundation for humanoid control.

Paperid: 291, https://arxiv.org/pdf/2511.07744.pdf GitHub

Abstract:
We introduce VectorSynth, a diffusion-based framework for pixel-accurate satellite image synthesis conditioned on polygonal geographic annotations with semantic attributes. Unlike prior text- or layout-conditioned models, VectorSynth learns dense cross-modal correspondences that align imagery and semantic vector geometry, enabling fine-grained, spatially grounded edits. A vision language alignment module produces pixel-level embeddings from polygon semantics; these embeddings guide a conditional image generation framework to respect both spatial extents and semantic cues. VectorSynth supports interactive workflows that mix language prompts with geometry-aware conditioning, allowing rapid what-if simulations, spatial edits, and map-informed content generation. For training and evaluation, we assemble a collection of satellite scenes paired with pixel-registered polygon annotations spanning diverse urban scenes with both built and natural features. We observe strong improvements over prior methods in semantic fidelity and structural realism, and show that our trained vision language model demonstrates fine-grained spatial grounding. The code and data are available at https://github.com/mvrl/VectorSynth.

Paperid: 292, https://arxiv.org/pdf/2511.07290.pdf GitHub

Abstract:
The prevalence of user-generated content (UGC) on platforms such as YouTube and TikTok has rendered no-reference (NR) perceptual video quality assessment (VQA) vital for optimizing video delivery. Nonetheless, the characteristics of non-professional acquisition and the subsequent transcoding of UGC video on sharing platforms present significant challenges for NR-VQA. Although NR-VQA models attempt to infer mean opinion scores (MOS), their modeling of subjective scores for compressed content remains limited due to the absence of fine-grained perceptual annotations of artifact types. To address these challenges, we propose CAMP-VQA, a novel NR-VQA framework that exploits the semantic understanding capabilities of large vision-language models. Our approach introduces a quality-aware prompting mechanism that integrates video metadata (e.g., resolution, frame rate, bitrate) with key fragments extracted from inter-frame variations to guide the BLIP-2 pretraining approach in generating fine-grained quality captions. A unified architecture has been designed to model perceptual quality across three dimensions: semantic alignment, temporal characteristics, and spatial characteristics. These multimodal features are extracted and fused, then regressed to video quality scores. Extensive experiments on a wide variety of UGC datasets demonstrate that our model consistently outperforms existing NR-VQA methods, achieving improved accuracy without the need for costly manual fine-grained annotations. Our method achieves the best performance in terms of average rank and linear correlation (SRCC: 0.928, PLCC: 0.938) compared to state-of-the-art methods. The source code and trained models, along with a user-friendly demo, are available at: https://github.com/xinyiW915/CAMP-VQA.

Paperid: 293, https://arxiv.org/pdf/2511.07068.pdf GitHub GitHub

Abstract:
Large-scale visual out-of-distribution (OOD) detection has witnessed remarkable progress by leveraging vision-language models such as CLIP. However, a significant limitation of current methods is their reliance on a pre-defined set of in-distribution (ID) ground-truth label names (positives). These fixed label names can be unavailable, unreliable at scale, or become less relevant due to in-distribution shifts after deployment. Towards truly unsupervised OOD detection, we utilize widely available text corpora for positive label mining, bypassing the need for positives. In this paper, we utilize widely available text corpora for positive label mining under a general concept mining paradigm. Within this framework, we propose ClusterMine, a novel positive label mining method. ClusterMine is the first method to achieve state-of-the-art OOD detection performance without access to positive labels. It extracts positive concepts from a large text corpus by combining visual-only sample consistency (via clustering) and zero-shot image-text consistency. Our experimental study reveals that ClusterMine is scalable across a plethora of CLIP models and achieves state-of-the-art robustness to covariate in-distribution shifts. The code is available at https://github.com/HHU-MMBS/clustermine_wacv_official.

Paperid: 294, https://arxiv.org/pdf/2511.06925.pdf GitHub

Abstract:
Video shadow detection confronts two entwined difficulties: distinguishing shadows from complex backgrounds and modeling dynamic shadow deformations under varying illumination. To address shadow-background ambiguity, we leverage linguistic priors through the proposed Vision-language Match Module (VMM) and a Dark-aware Semantic Block (DSB), extracting text-guided features to explicitly differentiate shadows from dark objects. Furthermore, we introduce adaptive mask reweighting to downweight penumbra regions during training and apply edge masks at the final decoder stage for better supervision. For temporal modeling of variable shadow shapes, we propose a Tokenized Temporal Block (TTB) that decouples spatiotemporal learning. TTB summarizes cross-frame shadow semantics into learnable temporal tokens, enabling efficient sequence encoding with minimal computation overhead. Comprehensive Experiments on multiple benchmark datasets demonstrate state-of-the-art accuracy and real-time inference efficiency. Codes are available at https://github.com/city-cheng/DTTNet.

Paperid: 295, https://arxiv.org/pdf/2511.06653.pdf GitHub

Abstract:
Contrastive vision-language models like CLIP have achieved impressive results in image-text retrieval by aligning image and text representations in a shared embedding space. However, these models often treat text as flat sequences, limiting their ability to handle complex, compositional, and long-form descriptions. In particular, they fail to capture two essential properties of language: semantic hierarchy, which reflects the multi-level compositional structure of text, and semantic monotonicity, where richer descriptions should result in stronger alignment with visual content.To address these limitations, we propose HiMo-CLIP, a representation-level framework that enhances CLIP-style models without modifying the encoder architecture. HiMo-CLIP introduces two key components: a hierarchical decomposition (HiDe) module that extracts latent semantic components from long-form text via in-batch PCA, enabling flexible, batch-aware alignment across different semantic granularities, and a monotonicity-aware contrastive loss (MoLo) that jointly aligns global and component-level representations, encouraging the model to internalize semantic ordering and alignment strength as a function of textual completeness.These components work in concert to produce structured, cognitively-aligned cross-modal representations. Experiments on multiple image-text retrieval benchmarks show that HiMo-CLIP consistently outperforms strong baselines, particularly under long or compositional descriptions. The code is available at https://github.com/UnicomAI/HiMo-CLIP.

Paperid: 296, https://arxiv.org/pdf/2511.06316.pdf GitHub

Abstract:
Reliable geospatial information on road accidents is vital for safety analysis and infrastructure planning, yet most low- and middle-income countries continue to face a critical shortage of accurate, location-specific crash data. Existing text-based geocoding tools perform poorly in multilingual and unstructured news environments, where incomplete place descriptions and mixed Bangla-English scripts obscure spatial context. To address these limitations, this study introduces ALIGN (Accident Location Inference through Geo-Spatial Neural Reasoning)- a vision-language framework that emulates human spatial reasoning to infer accident coordinates directly from textual and map-based cues. ALIGN integrates large language and vision-language models within a multi-stage pipeline that performs optical character recognition, linguistic reasoning, and map-level verification through grid-based spatial scanning. The framework systematically evaluates each predicted location against contextual and visual evidence, ensuring interpretable, fine-grained geolocation outcomes without requiring model retraining. Applied to Bangla-language news data, ALIGN demonstrates consistent improvements over traditional geoparsing methods, accurately identifying district and sub-district-level crash sites. Beyond its technical contribution, the framework establishes a high accuracy foundation for automated crash mapping in data-scarce regions, supporting evidence-driven road-safety policymaking and the broader integration of multimodal artificial intelligence in transportation analytics. The code for this paper is open-source and available at: https://github.com/Thamed-Chowdhury/ALIGN

Paperid: 297, https://arxiv.org/pdf/2511.06251.pdf GitHub GitHub

Abstract:
User interface (UI) development requires translating design mockups into functional code, a process that remains repetitive and labor-intensive. While recent Vision-Language Models (VLMs) automate UI-to-Code generation, they generate only static HTML/CSS/JavaScript layouts lacking interactivity. To address this, we propose WebVIA, the first agentic framework for interactive UI-to-Code generation and validation. The framework comprises three components: 1) an exploration agent to capture multi-state UI screenshots; 2) a UI2Code model that generates executable interactive code; 3) a validation module that verifies the interactivity. Experiments demonstrate that WebVIA-Agent achieves more stable and accurate UI exploration than general-purpose agents (e.g., Gemini-2.5-Pro). In addition, our fine-tuned WebVIA-UI2Code models exhibit substantial improvements in generating executable and interactive HTML/CSS/JavaScript code, outperforming their base counterparts across both interactive and static UI2Code benchmarks. Our code and models are available at \href{https://zheny2751-dotcom.github.io/webvia.github.io/}{\texttt{https://webvia.github.io}}.

Paperid: 298, https://arxiv.org/pdf/2511.05275.pdf GitHub

Abstract:
Vision-language-action models (VLAs) trained on large-scale robotic datasets have demonstrated strong performance on manipulation tasks, including bimanual tasks. However, because most public datasets focus on single-arm demonstrations, adapting VLAs for bimanual tasks typically requires substantial additional bimanual data and fine-tuning. To address this challenge, we introduce TwinVLA, a modular framework that composes two copies of a pretrained single-arm VLA into a coordinated bimanual VLA. Unlike monolithic cross-embodiment models trained on mixtures of single-arm and bimanual data, TwinVLA improves both data efficiency and performance by composing pretrained single-arm policies. Across diverse bimanual tasks in real-world and simulation settings, TwinVLA outperforms a comparably-sized monolithic RDT-1B model without requiring any bimanual pretraining. Furthermore, it narrows the gap to state-of-the-art model, $π_0$ which rely on extensive proprietary bimanual data and compute cost. These results establish our modular composition approach as a data-efficient and scalable path toward high-performance bimanual manipulation, leveraging public single-arm data.

Paperid: 299, https://arxiv.org/pdf/2511.04555.pdf GitHub

Abstract:
Vision-Language-Action (VLA) models have emerged as a powerful framework that unifies perception, language, and control, enabling robots to perform diverse tasks through multimodal understanding. However, current VLA models typically contain massive parameters and rely heavily on large-scale robot data pretraining, leading to high computational costs during training, as well as limited deployability for real-time inference. Moreover, most training paradigms often degrade the perceptual representations of the vision-language backbone, resulting in overfitting and poor generalization to downstream tasks. In this work, we present Evo-1, a lightweight VLA model that reduces computation and improves deployment efficiency, while maintaining strong performance without pretraining on robot data. Evo-1 builds on a native multimodal Vision-Language model (VLM), incorporating a novel cross-modulated diffusion transformer along with an optimized integration module, together forming an effective architecture. We further introduce a two-stage training paradigm that progressively aligns action with perception, preserving the representations of the VLM. Notably, with only 0.77 billion parameters, Evo-1 achieves state-of-the-art results on the Meta-World and RoboTwin suite, surpassing the previous best models by 12.4% and 6.9%, respectively, and also attains a competitive result of 94.8% on LIBERO. In real-world evaluations, Evo-1 attains a 78% success rate with high inference frequency and low memory overhead, outperforming all baseline methods. We release code, data, and model weights to facilitate future research on lightweight and efficient VLA models.

Paperid: 300, https://arxiv.org/pdf/2511.04247.pdf GitHub

Abstract:
Multimodal co-embedding models, especially CLIP, have advanced the state of the art in zero-shot classification and multimedia information retrieval in recent years by aligning images and text in a shared representation space. However, such modals trained on a contrastive alignment can lack stability towards small input perturbations. Especially when dealing with manually expressed queries, minor variations in the query can cause large differences in the ranking of the best-matching results. In this paper, we present a systematic analysis of the effect of multiple classes of non-semantic query perturbations in an multimedia information retrieval scenario. We evaluate a diverse set of lexical, syntactic, and semantic perturbations across multiple CLIP variants using the TRECVID Ad-Hoc Video Search queries and the V3C1 video collection. Across models, we find that syntactic and semantic perturbations drive the largest instabilities, while brittleness is concentrated in trivial surface edits such as punctuation and case. Our results highlight robustness as a critical dimension for evaluating vision-language models beyond benchmark accuracy.

Paperid: 301, https://arxiv.org/pdf/2511.04123.pdf GitHub

Abstract:
Recent advances in vision-language models have facilitated progress in sketch generation. However, existing specialized methods primarily focus on generic synthesis and lack mechanisms for precise control over sketch styles. In this work, we propose a training-free framework based on diffusion models that enables explicit style guidance via textual prompts and referenced style sketches. Unlike previous style transfer methods that overwrite key and value matrices in self-attention, we incorporate the reference features as auxiliary information with linear smoothing and leverage a style-content guidance mechanism. This design effectively reduces content leakage from reference sketches and enhances synthesis quality, especially in cases with low structural similarity between reference and target sketches. Furthermore, we extend our framework to support controllable multi-style generation by integrating features from multiple reference sketches, coordinated via a joint AdaIN module. Extensive experiments demonstrate that our approach achieves high-quality sketch generation with accurate style alignment and improved flexibility in style control. The official implementation of M3S is available at https://github.com/CMACH508/M3S.

Paperid: 302, https://arxiv.org/pdf/2511.03367.pdf GitHub

Abstract:
Recent advances in large-scale vision and language models have led to significant progress in zero-shot learning tasks. Methods such as CoOp and CoCoOp have shown that replacing handcrafted prompts with learnable vectors, known as prompt learning, can result in improved performance. However, these models often struggle to generalize to entirely unseen categories. While traditional zero-shot learning techniques benefit from various data augmentation strategies, prompt learning has primarily focused on text-based modifications, leaving the potential of image-based augmentation largely unexplored. In this work, we explore how image-level augmentations, particularly those that introduce attribute-specific variations, can support and enhance prompt learning. Our analysis examines the interaction between these augmentations and soft prompt frameworks, revealing their potential to improve generalization. We also identify a limitation in existing methods, such as CoCoOp, which do not provide explicit guidance for learning prompts that focus on semantically meaningful visual features. To address this, we propose Adding Attributes to Prompt Learning, AAPL, a novel method that introduces adversarial token embeddings to decouple superficial visual variations introduced by augmentation from class-relevant semantic representations. This decoupling enables the learned prompts to concentrate on visually discriminative features that align with the target categories. We conduct comprehensive experiments on eleven benchmark datasets, and AAPL consistently outperforms existing methods across few-shot, zero-shot, cross-dataset, and domain generalization settings. Our source code is publicly available at: https://github.com/Gahyeonkim09/AAPL

Paperid: 303, https://arxiv.org/pdf/2511.02776.pdf GitHub

Abstract:
Recent progress in large-scale robotic datasets and vision-language models (VLMs) has advanced research on vision-language-action (VLA) models. However, existing VLA models still face two fundamental challenges: (i) producing precise low-level actions from high-dimensional observations, (ii) bridging domain gaps across heterogeneous data sources, including diverse robot embodiments and human demonstrations. Existing methods often encode latent variables from either visual dynamics or robotic actions to guide policy learning, but they fail to fully exploit the complementary multi-modal knowledge present in large-scale, heterogeneous datasets. In this work, we present X Robotic Model 1 (XR-1), a novel framework for versatile and scalable VLA learning across diverse robots, tasks, and environments. XR-1 introduces the \emph{Unified Vision-Motion Codes (UVMC)}, a discrete latent representation learned via a dual-branch VQ-VAE that jointly encodes visual dynamics and robotic motion. UVMC addresses these challenges by (i) serving as an intermediate representation between the observations and actions, and (ii) aligning multimodal dynamic information from heterogeneous data sources to capture complementary knowledge. To effectively exploit UVMC, we propose a three-stage training paradigm: (i) self-supervised UVMC learning, (ii) UVMC-guided pretraining on large-scale cross-embodiment robotic datasets, and (iii) task-specific post-training. We validate XR-1 through extensive real-world experiments with more than 14,000 rollouts on six different robot embodiments, spanning over 120 diverse manipulation tasks. XR-1 consistently outperforms state-of-the-art baselines such as $π_{0.5}$, $π_0$, RDT, UniVLA, and GR00T-N1.5 while demonstrating strong generalization to novel objects, background variations, distractors, and illumination changes. Our project is at https://xr-1-vla.github.io/.

Paperid: 304, https://arxiv.org/pdf/2511.02239.pdf GitHub

Abstract:
Learning generalizable policies for robotic manipulation increasingly relies on large-scale models that map language instructions to actions (L2A). However, this one-way paradigm often produces policies that execute tasks without deeper contextual understanding, limiting their ability to generalize or explain their behavior. We argue that the complementary skill of mapping actions back to language (A2L) is essential for developing more holistic grounding. An agent capable of both acting and explaining its actions can form richer internal representations and unlock new paradigms for self-supervised learning. We introduce LACY (Language-Action Cycle), a unified framework that learns such bidirectional mappings within a single vision-language model. LACY is jointly trained on three synergistic tasks: generating parameterized actions from language (L2A), explaining observed actions in language (A2L), and verifying semantic consistency between two language descriptions (L2C). This enables a self-improving cycle that autonomously generates and filters new training data through an active augmentation strategy targeting low-confidence cases, thereby improving the model without additional human labels. Experiments on pick-and-place tasks in both simulation and the real world show that LACY improves task success rates by 56.46% on average and yields more robust language-action grounding for robotic manipulation. Project page: https://vla2026.github.io/LACY/

Paperid: 305, https://arxiv.org/pdf/2511.01999.pdf GitHub

Abstract:
Vision-Language Models (VLMs) struggle to translate high-level instructions into the precise spatial affordances required for robotic manipulation. While visual Chain-of-Thought (CoT) methods exist, they are often computationally intensive. In this work, we introduce TRACE (Textual Reasoning for Affordance Coordinate Extraction), a novel methodology that integrates a textual Chain of Reasoning (CoR) into the affordance prediction process. We use this methodology to create the TRACE dataset, a large-scale collection created via an autonomous pipeline that pairs instructions with explicit textual rationales. By fine-tuning a VLM on this data, our model learns to externalize its spatial reasoning before acting. Our experiments show that our TRACE-tuned model achieves state-of-the-art performance, reaching 48.1% accuracy on the primary Where2Place (W2P) benchmark (a 9.6% relative improvement) and 55.0% on the more challenging W2P(h) subset. Crucially, an ablation study demonstrates that performance scales directly with the amount of reasoning data used, confirming the CoR's effectiveness. Furthermore, analysis of the model's attention maps reveals an interpretable reasoning process where focus shifts dynamically across reasoning steps. This work shows that training VLMs to generate a textual CoR is an effective and robust strategy for enhancing the precision, reliability, and interpretability of VLM-based robot control. Our dataset and code are available at https://github.com/jink-ucla/TRACE

Paperid: 306, https://arxiv.org/pdf/2511.01952.pdf GitHub

Abstract:
Large vision-language models (LVLMs) derive their capabilities from extensive training on vast corpora of visual and textual data. Empowered by large-scale parameters, these models often exhibit strong memorization of their training data, rendering them susceptible to membership inference attacks (MIAs). Existing MIA methods for LVLMs typically operate under white- or gray-box assumptions, by extracting likelihood-based features for the suspected data samples based on the target LVLMs. However, mainstream LVLMs generally only expose generated outputs while concealing internal computational features during inference, limiting the applicability of these methods. In this work, we propose the first black-box MIA framework for LVLMs, based on a prior knowledge-calibrated memory probing mechanism. The core idea is to assess the model memorization of the private semantic information embedded within the suspected image data, which is unlikely to be inferred from general world knowledge alone. We conducted extensive experiments across four LVLMs and three datasets. Empirical results demonstrate that our method effectively identifies training data of LVLMs in a purely black-box setting and even achieves performance comparable to gray-box and white-box methods. Further analysis reveals the robustness of our method against potential adversarial manipulations, and the effectiveness of the methodology designs. Our code and data are available at https://github.com/spmede/KCMP.

Paperid: 307, https://arxiv.org/pdf/2511.01791.pdf GitHub

Abstract:
Data scarcity remains a fundamental bottleneck for embodied intelligence. Existing approaches use large language models (LLMs) to automate gripper-based simulation generation, but they transfer poorly to dexterous manipulation, which demands more specialized environment design. Meanwhile, dexterous manipulation tasks are inherently more difficult due to their higher degrees of freedom. Massively generating feasible and trainable dexterous hand tasks remains an open challenge. To this end, we present GenDexHand, a generative simulation pipeline that autonomously produces diverse robotic tasks and environments for dexterous manipulation. GenDexHand introduces a closed-loop refinement process that adjusts object placements and scales based on vision-language model (VLM) feedback, substantially improving the average quality of generated environments. Each task is further decomposed into sub-tasks to enable sequential reinforcement learning, reducing training time and increasing success rates. Our work provides a viable path toward scalable training of diverse dexterous hand behaviors in embodied intelligence by offering a simulation-based solution to synthetic data generation. Our website: https://winniechen2002.github.io/GenDexHand/.

Paperid: 308, https://arxiv.org/pdf/2511.01718.pdf GitHub

Abstract:
Vision-language-action (VLA) models aim to understand natural language instructions and visual observations and to execute corresponding actions as an embodied agent. Recent work integrates future images into the understanding-acting loop, yielding unified VLAs that jointly understand, generate, and act -- reading text and images and producing future images and actions. However, these models either rely on external experts for modality unification or treat image generation and action prediction as separate processes, limiting the benefits of direct synergy between these tasks. Our core philosophy is to optimize generation and action jointly through a synchronous denoising process, where the iterative refinement enables actions to evolve from initialization, under constant and sufficient visual guidance. We ground this philosophy in our proposed Unified Diffusion VLA and Joint Discrete Denoising Diffusion Process (JD3P), which is a joint diffusion process that integrates multiple modalities into a single denoising trajectory to serve as the key mechanism enabling understanding, generation, and acting to be intrinsically synergistic. Our model and theory are built on a unified tokenized space of all modalities and a hybrid attention mechanism. We further propose a two-stage training pipeline and several inference-time techniques that optimize performance and efficiency. Our approach achieves state-of-the-art performance on benchmarks such as CALVIN, LIBERO, and SimplerEnv with 4$\times$ faster inference than autoregressive methods, and we demonstrate its effectiveness through in-depth analysis and real-world evaluations. Our project page is available at https://irpn-eai.github.io/UD-VLA.github.io/.

Paperid: 309, https://arxiv.org/pdf/2511.01678.pdf GitHub

Abstract:
Relighting is a crucial task with both practical demand and artistic value, and recent diffusion models have shown strong potential by enabling rich and controllable lighting effects. However, as they are typically optimized in semantic latent space, where proximity does not guarantee physical correctness in visual space, they often produce unrealistic results, such as overexposed highlights, misaligned shadows, and incorrect occlusions. We address this with UniLumos, a unified relighting framework for both images and videos that brings RGB-space geometry feedback into a flow matching backbone. By supervising the model with depth and normal maps extracted from its outputs, we explicitly align lighting effects with the scene structure, enhancing physical plausibility. Nevertheless, this feedback requires high-quality outputs for supervision in visual space, making standard multi-step denoising computationally expensive. To mitigate this, we employ path consistency learning, allowing supervision to remain effective even under few-step training regimes. To enable fine-grained relighting control and supervision, we design a structured six-dimensional annotation protocol capturing core illumination attributes. Building upon this, we propose LumosBench, a disentangled attribute-level benchmark that evaluates lighting controllability via large vision-language models, enabling automatic and interpretable assessment of relighting precision across individual dimensions. Extensive experiments demonstrate that UniLumos achieves state-of-the-art relighting quality with significantly improved physical consistency, while delivering a 20x speedup for both image and video relighting. Code is available at https://github.com/alibaba-damo-academy/Lumos-Custom.

Paperid: 310, https://arxiv.org/pdf/2511.01617.pdf GitHub

Abstract:
In the retrieval domain, candidates' fusion from heterogeneous retrievers is a long-standing challenge, particularly for complex, multi-modal data such as videos. While typical fusion techniques are training-free, they rely solely on rank or score signals, disregarding candidates' representations. This work introduces Vote-in-Context (ViC), a generalized, training-free framework that re-thinks list-wise reranking and fusion as a zero-shot reasoning task for a Vision-Language Model (VLM). The core insight is to serialize both content evidence and retriever metadata directly within the VLM's prompt, allowing the model to adaptively weigh retriever consensus against visual-linguistic content. We demonstrate the generality of this framework by applying it to the challenging domain of cross-modal video retrieval. To this end, we introduce the S-Grid, a compact serialization map that represents each video as an image grid, optionally paired with subtitles to enable list-wise reasoning over video candidates. ViC is evaluated both as a single-list reranker, where it dramatically improves the precision of individual retrievers, and as an ensemble fuser, where it consistently outperforms strong baselines like CombSUM. Across video retrieval benchmarks including ActivityNet and VATEX, the framework establishes new state-of-the-art zero-shot retrieval performance, demonstrating its effectiveness in handling complex visual and temporal signals alongside text. In zero-shot settings, ViC achieves Recall@1 scores of 87.1% (t2v) / 89.0% (v2t) on MSR-VTT and 99.6% (v2t) on VATEX, representing massive gains of up to +40 Recall@1 over previous state-of-the-art baselines. We present ViC as a simple, reproducible, and highly effective recipe for turning modern VLMs into powerful zero-shot rerankers and fusers. Code and resources are publicly available at: https://github.com/mohammad2012191/ViC

Paperid: 311, https://arxiv.org/pdf/2511.01458.pdf GitHub

Abstract:
Safety and reliability are essential for deploying Visual Question Answering (VQA) in surgery, where incorrect or ambiguous responses can harm the patient. Most surgical VQA research focuses on accuracy or linguistic quality while overlooking safety behaviors such as ambiguity awareness, referral to human experts, or triggering a second opinion. Inspired by Automatic Failure Detection (AFD), we study uncertainty estimation as a key enabler of safer decision making. We introduce Question Aligned Semantic Nearest Neighbor Entropy (QA-SNNE), a black box uncertainty estimator that incorporates question semantics into prediction confidence. It measures semantic entropy by comparing generated answers with nearest neighbors in a medical text embedding space, conditioned on the question. We evaluate five models, including domain specific Parameter-Efficient Fine-Tuned (PEFT) models and zero-shot Large Vision-Language Models (LVLMs), on EndoVis18-VQA and PitVQA. PEFT models degrade under mild paraphrasing, while LVLMs are more resilient. Across three LVLMs and two PEFT baselines, QA-SNNE improves AUROC in most in-template settings and enhances hallucination detection. The Area Under the ROC Curve (AUROC) increases by 15-38% for zero-shot models, with gains maintained under out-of-template stress. QA-SNNE offers a practical and interpretable step toward AFD in surgical VQA by linking semantic uncertainty to question context. Combining LVLM backbones with question aligned uncertainty estimation can improve safety and clinician trust. The code and model are available at https://github.com/DennisPierantozzi/QASNNE

Paperid: 312, https://arxiv.org/pdf/2511.00480.pdf GitHub

Abstract:
In this paper, we introduce FedMGP, a new paradigm for personalized federated prompt learning in vision-language models. FedMGP equips each client with multiple groups of paired textual and visual prompts, enabling the model to capture diverse, fine-grained semantic and instance-level cues. A diversity loss is introduced to drive each prompt group to specialize in distinct and complementary semantic aspects, ensuring that the groups collectively cover a broader range of local characteristics. During communication, FedMGP employs a dynamic prompt aggregation strategy based on similarity-guided probabilistic sampling: each client computes the cosine similarity between its prompt groups and the global prompts from the previous round, then samples s groups via a softmax-weighted distribution. This soft selection mechanism preferentially aggregates semantically aligned knowledge while still enabling exploration of underrepresented patterns effectively balancing the preservation of common knowledge with client-specific features. Notably, FedMGP maintains parameter efficiency by redistributing a fixed prompt capacity across multiple groups, achieving state-of-the-art performance with the lowest communication parameters among all federated prompt learning methods. Theoretical analysis shows that our dynamic aggregation strategy promotes robust global representation learning by reinforcing shared semantics while suppressing client-specific noise. Extensive experiments demonstrate that FedMGP consistently outperforms prior approaches in both personalization and domain generalization across diverse federated vision-language benchmarks. The code will be released on https://github.com/weihao-bo/FedMGP.git.

Paperid: 313, https://arxiv.org/pdf/2511.00446.pdf GitHub

Abstract:
The Contrastive Language-Image Pretraining (CLIP) model has significantly advanced vision-language modeling by aligning image-text pairs from large-scale web data through self-supervised contrastive learning. Yet, its reliance on uncurated Internet-sourced data exposes it to data poisoning and backdoor risks. While existing studies primarily investigate image-based attacks, the text modality, which is equally central to CLIP's training, remains underexplored. In this work, we introduce ToxicTextCLIP, a framework for generating high-quality adversarial texts that target CLIP during the pre-training phase. The framework addresses two key challenges: semantic misalignment caused by background inconsistency with the target class, and the scarcity of background-consistent texts. To this end, ToxicTextCLIP iteratively applies: 1) a background-aware selector that prioritizes texts with background content aligned to the target class, and 2) a background-driven augmenter that generates semantically coherent and diverse poisoned samples. Extensive experiments on classification and retrieval tasks show that ToxicTextCLIP achieves up to 95.83% poisoning success and 98.68% backdoor Hit@1, while bypassing RoCLIP, CleanCLIP and SafeCLIP defenses. The source code can be accessed via https://github.com/xinyaocse/ToxicTextCLIP/.

Paperid: 314, https://arxiv.org/pdf/2511.00391.pdf GitHub

Abstract:
Multimodal code generation has garnered significant interest within the research community. Despite the notable success of recent vision-language models (VLMs) on specialized tasks like Chart-to-code generation, their reliance on single-task training regimens fosters a narrow paradigm that hinders the development of generalized \textbf{VI}sio\textbf{N} \textbf{C}ode \textbf{I}ntelligence. In this work, we introduce \textbf{VinciCoder}, a unified multimodal code generation model that addresses this limitation via a two-stage training framework. We begin by constructing a large-scale Supervised Finetuning (SFT) corpus comprising 1.6M image-code pairs for tasks involving direct code generation and visual-based code refinement. Subsequently, we introduce a Visual Reinforcement Learning (ViRL) strategy, which employs a coarse-to-fine reward mechanism to improve visual fidelity by calculating visual similarity across local and global image patches. Extensive experiments on various multimodal code generation benchmarks demonstrate that VinciCoder achieves state-of-the-art performance, underscoring the effectiveness of our coarse-to-fine ViRL strategy. The code and model will be available at https://github.com/DocTron-hub/VinciCoder.

Paperid: 315, https://arxiv.org/pdf/2511.00095.pdf GitHub

Abstract:
The anatomical structure segmentation of the spine and adjacent structures from computed tomography (CT) images is a key step for spinal disease diagnosis and treatment. However, the segmentation of CT images is impeded by low contrast and complex vertebral boundaries. Although advanced models such as the Segment Anything Model (SAM) have shown promise in various segmentation tasks, their performance in spinal CT imaging is limited by high annotation requirements and poor domain adaptability. To address these limitations, we propose SpinalSAM-R1, a multimodal vision-language interactive system that integrates a fine-tuned SAM with DeepSeek-R1, for spine CT image segmentation. Specifically, our SpinalSAM-R1 introduces an anatomy-guided attention mechanism to improve spine segmentation performance, and a semantics-driven interaction protocol powered by DeepSeek-R1, enabling natural language-guided refinement. The SpinalSAM-R1 is fine-tuned using Low-Rank Adaptation (LoRA) for efficient adaptation. We validate our SpinalSAM-R1 on the spine anatomical structure with CT images. Experimental results suggest that our method achieves superior segmentation performance. Meanwhile, we develop a PyQt5-based interactive software, which supports point, box, and text-based prompts. The system supports 11 clinical operations with 94.3\% parsing accuracy and sub-800 ms response times. The software is released on https://github.com/6jm233333/spinalsam-r1.

Paperid: 316, https://arxiv.org/pdf/2511.00062.pdf GitHub GitHub

Authors:NVIDIA, :, Arslan Ali, Junjie Bai, Maciej Bala, Yogesh Balaji, Aaron Blakeman, Tiffany Cai, Jiaxin Cao, Tianshi Cao, Elizabeth Cha, Yu-Wei Chao, Prithvijit Chattopadhyay, Mike Chen, Yongxin Chen, Yu Chen, Shuai Cheng, Yin Cui, Jenna Diamond, Yifan Ding, Jiaojiao Fan, Linxi Fan, Liang Feng, Francesco Ferroni, Sanja Fidler, Xiao Fu, Ruiyuan Gao, Yunhao Ge, Jinwei Gu, Aryaman Gupta, Siddharth Gururani, Imad El Hanafi, Ali Hassani, Zekun Hao, Jacob Huffman, Joel Jang, Pooya Jannaty, Jan Kautz, Grace Lam, Xuan Li, Zhaoshuo Li, Maosheng Liao, Chen-Hsuan Lin, Tsung-Yi Lin, Yen-Chen Lin, Huan Ling, Ming-Yu Liu, Xian Liu, Yifan Lu, Alice Luo, Qianli Ma, Hanzi Mao, Kaichun Mo, Seungjun Nah, Yashraj Narang, Abhijeet Panaskar, Lindsey Pavao, Trung Pham, Morteza Ramezanali, Fitsum Reda, Scott Reed, Xuanchi Ren, Haonan Shao, Yue Shen, Stella Shi, Shuran Song, Bartosz Stefaniak, Shangkun Sun, Shitao Tang, Sameena Tasmeen, Lyne Tchapmi, Wei-Cheng Tseng, Jibin Varghese, Andrew Z. Wang, Hao Wang, Haoxiang Wang, Heng Wang, Ting-Chun Wang, Fangyin Wei, Jiashu Xu, Dinghao Yang, Xiaodong Yang, Haotian Ye, Seonghyeon Ye, Xiaohui Zeng, Jing Zhang, Qinsheng Zhang, Kaiwen Zheng, Andrew Zhu, Yuke Zhu

Abstract:
We introduce [Cosmos-Predict2.5], the latest generation of the Cosmos World Foundation Models for Physical AI. Built on a flow-based architecture, [Cosmos-Predict2.5] unifies Text2World, Image2World, and Video2World generation in a single model and leverages [Cosmos-Reason1], a Physical AI vision-language model, to provide richer text grounding and finer control of world simulation. Trained on 200M curated video clips and refined with reinforcement learning-based post-training, [Cosmos-Predict2.5] achieves substantial improvements over [Cosmos-Predict1] in video quality and instruction alignment, with models released at 2B and 14B scales. These capabilities enable more reliable synthetic data generation, policy evaluation, and closed-loop simulation for robotics and autonomous systems. We further extend the family with [Cosmos-Transfer2.5], a control-net style framework for Sim2Real and Real2Real world translation. Despite being 3.5$\times$ smaller than [Cosmos-Transfer1], it delivers higher fidelity and robust long-horizon video generation. Together, these advances establish [Cosmos-Predict2.5] and [Cosmos-Transfer2.5] as versatile tools for scaling embodied intelligence. To accelerate research and deployment in Physical AI, we release source code, pretrained checkpoints, and curated benchmarks under the NVIDIA Open Model License at https://github.com/nvidia-cosmos/cosmos-predict2.5 and https://github.com/nvidia-cosmos/cosmos-transfer2.5. We hope these open resources lower the barrier to adoption and foster innovation in building the next generation of embodied intelligence.

Paperid: 317, https://arxiv.org/pdf/2510.27265.pdf GitHub

Abstract:
In medical imaging, vision-language models face a critical duality: pretrained networks offer broad robustness but lack subtle, modality-specific characteristics, while fine-tuned expert models achieve high in-distribution accuracy yet falter under modality shift. Existing model-merging techniques, designed for natural-image benchmarks, are simple and efficient but fail to deliver consistent gains across diverse medical modalities; their static interpolation limits reliability in varied clinical tasks. To address this, we introduce Test-Time Task adaptive merging (T^3), a backpropagation-free framework that computes per-sample interpolation coefficients via the Jensen-Shannon divergence between the two models' output distributions. T^3 dynamically preserves local precision when models agree and defers to generalist robustness under drift. To overcome the inference costs of sample-wise merging, we further propose a batch-wise extension, T^3_B, that computes a merging coefficient across a batch of samples, dramatically reducing computational bottleneck. Recognizing the lack of a standardized medical-merging benchmark, we present a rigorous cross-evaluation protocol spanning in-domain, base-to-novel, and corruptions across four modalities. Empirically, T^3 sets new state-of-the-art in Top-1 accuracy and error reduction, outperforming strong baselines while maintaining efficiency, paving the way for adaptive MVLM deployment in clinical settings. Our code is available at https://github.com/Razaimam45/TCube.

Paperid: 318, https://arxiv.org/pdf/2510.26937.pdf GitHub

Abstract:
Large Vision-Language Models (LVLMs) have exhibited remarkable progress. However, deficiencies remain compared to human intelligence, such as hallucination and shallow pattern matching. In this work, we aim to evaluate a fundamental yet underexplored intelligence: association, a cornerstone of human cognition for creative thinking and knowledge integration. Current benchmarks, often limited to closed-ended tasks, fail to capture the complexity of open-ended association reasoning vital for real-world applications. To address this, we present MM-OPERA, a systematic benchmark with 11,497 instances across two open-ended tasks: Remote-Item Association (RIA) and In-Context Association (ICA), aligning association intelligence evaluation with human psychometric principles. It challenges LVLMs to resemble the spirit of divergent thinking and convergent associative reasoning through free-form responses and explicit reasoning paths. We deploy tailored LLM-as-a-Judge strategies to evaluate open-ended outputs, applying process-reward-informed judgment to dissect reasoning with precision. Extensive empirical studies on state-of-the-art LVLMs, including sensitivity analysis of task instances, validity analysis of LLM-as-a-Judge strategies, and diversity analysis across abilities, domains, languages, cultures, etc., provide a comprehensive and nuanced understanding of the limitations of current LVLMs in associative reasoning, paving the way for more human-like and general-purpose AI. The dataset and code are available at https://github.com/MM-OPERA-Bench/MM-OPERA.

Paperid: 319, https://arxiv.org/pdf/2510.26909.pdf GitHub

Abstract:
Vision-language models demonstrate unprecedented performance and generalization across a wide range of tasks and scenarios. Integrating these foundation models into robotic navigation systems opens pathways toward building general-purpose robots. Yet, evaluating these models' navigation capabilities remains constrained by costly real-world trials, overly simplified simulations, and limited benchmarks. We introduce NaviTrace, a high-quality Visual Question Answering benchmark where a model receives an instruction and embodiment type (human, legged robot, wheeled robot, bicycle) and must output a 2D navigation trace in image space. Across 1000 scenarios and more than 3000 expert traces, we systematically evaluate eight state-of-the-art VLMs using a newly introduced semantic-aware trace score. This metric combines Dynamic Time Warping distance, goal endpoint error, and embodiment-conditioned penalties derived from per-pixel semantics and correlates with human preferences. Our evaluation reveals consistent gap to human performance caused by poor spatial grounding and goal localization. NaviTrace establishes a scalable and reproducible benchmark for real-world robotic navigation. The benchmark and leaderboard can be found at https://leggedrobotics.github.io/navitrace_webpage/.

Paperid: 320, https://arxiv.org/pdf/2510.26865.pdf GitHub

Abstract:
Reading measurement instruments is effortless for humans and requires relatively little domain expertise, yet it remains surprisingly challenging for current vision-language models (VLMs) as we find in preliminary evaluation. In this work, we introduce MeasureBench, a benchmark on visual measurement reading covering both real-world and synthesized images of various types of measurements, along with an extensible pipeline for data synthesis. Our pipeline procedurally generates a specified type of gauge with controllable visual appearance, enabling scalable variation in key details such as pointers, scales, fonts, lighting, and clutter. Evaluation on popular proprietary and open-weight VLMs shows that even the strongest frontier VLMs struggle measurement reading in general. A consistent failure mode is indicator localization: models can read digits or labels but misidentify the key positions of pointers or alignments, leading to big numeric errors despite plausible textual reasoning. We have also conducted preliminary experiments with reinforcement learning over synthetic data, and find encouraging results on in-domain synthetic subset but less promising for real-world images. Our analysis highlights a fundamental limitation of current VLMs in fine-grained spatial grounding. We hope this resource can help future advances on visually grounded numeracy and precise spatial perception of VLMs, bridging the gap between recognizing numbers and measuring the world.

Paperid: 321, https://arxiv.org/pdf/2510.26583.pdf GitHub

Abstract:
We introduce Emu3.5, a large-scale multimodal world model that natively predicts the next state across vision and language. Emu3.5 is pre-trained end-to-end with a unified next-token prediction objective on a corpus of vision-language interleaved data containing over 10 trillion tokens, primarily derived from sequential frames and transcripts of internet videos. The model naturally accepts interleaved vision-language inputs and generates interleaved vision-language outputs. Emu3.5 is further post-trained with large-scale reinforcement learning to enhance multimodal reasoning and generation. To improve inference efficiency, we propose Discrete Diffusion Adaptation (DiDA), which converts token-by-token decoding into bidirectional parallel prediction, accelerating per-image inference by about 20x without sacrificing performance. Emu3.5 exhibits strong native multimodal capabilities, including long-horizon vision-language generation, any-to-image (X2I) generation, and complex text-rich image generation. It also exhibits generalizable world-modeling abilities, enabling spatiotemporally consistent world exploration and open-world embodied manipulation across diverse scenarios and tasks. For comparison, Emu3.5 achieves performance comparable to Gemini 2.5 Flash Image (Nano Banana) on image generation and editing tasks and demonstrates superior results on a suite of interleaved generation tasks. We open-source Emu3.5 at https://github.com/baaivision/Emu3.5 to support community research.

Paperid: 322, https://arxiv.org/pdf/2510.26536.pdf GitHub

Abstract:
The proliferation of collaborative robots across diverse tasks and embodiments presents a central challenge: achieving lifelong adaptability, scalable coordination, and robust scheduling in multi-agent systems. Existing approaches, from vision-language-action (VLA) models to hierarchical frameworks, fall short due to their reliance on limited or dividual-agent memory. This fundamentally constrains their ability to learn over long horizons, scale to heterogeneous teams, or recover from failures, highlighting the need for a unified memory representation. To address these limitations, we introduce RoboOS-NeXT, a unified memory-based framework for lifelong, scalable, and robust multi-robot collaboration. At the core of RoboOS-NeXT is the novel Spatio-Temporal-Embodiment Memory (STEM), which integrates spatial scene geometry, temporal event history, and embodiment profiles into a shared representation. This memory-centric design is integrated into a brain-cerebellum framework, where a high-level brain model performs global planning by retrieving and updating STEM, while low-level controllers execute actions locally. This closed loop between cognition, memory, and execution enables dynamic task allocation, fault-tolerant collaboration, and consistent state synchronization. We conduct extensive experiments spanning complex coordination tasks in restaurants, supermarkets, and households. Our results demonstrate that RoboOS-NeXT achieves superior performance across heterogeneous embodiments, validating its effectiveness in enabling lifelong, scalable, and robust multi-robot collaboration. Project website: https://flagopen.github.io/RoboOS/

Paperid: 323, https://arxiv.org/pdf/2510.25801.pdf GitHub

Abstract:
Reinforcement learning (RL) with verifiable rewards has recently catalyzed a wave of "MLLM-r1" approaches that bring RL to vision language models. Most representative paradigms begin with a cold start, typically employing supervised fine-tuning (SFT), to initialize the policy before RL. However, SFT-based cold start adopts the reasoning paradigm intertwined with task solution and output format, which may induce instruction-style overfitting, weakens out-of-distribution generalization, and ultimately affects downstream RL. We revisit the cold start along two views, its training method and data construction, and introduce the Generalization Factor (GF) coefficient to quantify the generalization capability under different methods. Our empirical study finds that preference-based training methods (e.g. DPO) generalizes better than SFT-based methods in cold start. Motivated by this, we propose SPECS-a Self-distilled, Preference-based Cold Start framework that decouples multimodal learning: (1) generates introspective preference data pairs via self-distillation, avoiding reliance on larger teachers or manual annotation; (2) performs preference-based training to learn, focusing on shallow, transferable surface-form criteria (format, structure, style) rather than memorizing content; and (3) hands off to RL with verifiable rewards for deep reasoning results. Experimental results across multiple multimodal benchmarks show that our decoupling learning framework yields consistent performance gains over strong baselines, improving MEGA-Bench by 4.1% and MathVista by 12.2%. Additional experiments indicate that SPECS contributes to reducing in-distribution "stuckness," improving exploration, stabilizing training, and raising the performance ceiling.

Paperid: 324, https://arxiv.org/pdf/2510.25760.pdf GitHub

Abstract:
Humans possess spatial reasoning abilities that enable them to understand spaces through multimodal observations, such as vision and sound. Large multimodal reasoning models extend these abilities by learning to perceive and reason, showing promising performance across diverse spatial tasks. However, systematic reviews and publicly available benchmarks for these models remain limited. In this survey, we provide a comprehensive review of multimodal spatial reasoning tasks with large models, categorizing recent progress in multimodal large language models (MLLMs) and introducing open benchmarks for evaluation. We begin by outlining general spatial reasoning, focusing on post-training techniques, explainability, and architecture. Beyond classical 2D tasks, we examine spatial relationship reasoning, scene and layout understanding, as well as visual question answering and grounding in 3D space. We also review advances in embodied AI, including vision-language navigation and action models. Additionally, we consider emerging modalities such as audio and egocentric video, which contribute to novel spatial understanding through new sensors. We believe this survey establishes a solid foundation and offers insights into the growing field of multimodal spatial reasoning. Updated information about this survey, codes and implementation of the open benchmarks can be found at https://github.com/zhengxuJosh/Awesome-Spatial-Reasoning.

Paperid: 325, https://arxiv.org/pdf/2510.25682.pdf GitHub GitHub

Abstract:
Unified vision-language models (UVLMs) must perform both understanding and generation within a single architecture, but these tasks rely on heterogeneous data and supervision, making it difficult to balance them during reinforcement learning (RL). We propose PairUni, a unified framework that reorganizes data into understanding-generation (UG) pairs and aligns optimization accordingly. We first use GPT-o3 to augment single-task data, generating captions for understanding samples and question-answer (QA) pairs for generation samples, forming aligned pairs from the same instance. Additionally, for each generation sample, we retrieve a semantically related understanding example to form a retrieved pair, linking different but related data points. These paired structures expose cross-task semantic correspondences and support consistent policy learning. To leverage this structure, we present Pair-GPRO, a pair-aware variant based on Group Relative Policy Optimization. It assigns a similarity score to each pair to modulate the advantage, strengthening learning from well-aligned examples and reducing task interference. We curate a high-quality dataset of 16K UG pairs named PairUG for RL fine-tuning and evaluate PairUni on the powerful Janus-Pro UVLMs. Our approach achieves balanced improvements on various UVLMs, outperforming strong UVLM RL baselines. Code: \href{https://github.com/Haochen-Wang409/PairUni}{github.com/Haochen-Wang409/PairUni}

Paperid: 326, https://arxiv.org/pdf/2510.25175.pdf GitHub

Abstract:
In recent years, test-time adaptive object detection has attracted increasing attention due to its unique advantages in online domain adaptation, which aligns more closely with real-world application scenarios. However, existing approaches heavily rely on source-derived statistical characteristics while making the strong assumption that the source and target domains share an identical category space. In this paper, we propose the first foundation model-powered test-time adaptive object detection method that eliminates the need for source data entirely and overcomes traditional closed-set limitations. Specifically, we design a Multi-modal Prompt-based Mean-Teacher framework for vision-language detector-driven test-time adaptation, which incorporates text and visual prompt tuning to adapt both language and vision representation spaces on the test data in a parameter-efficient manner. Correspondingly, we propose a Test-time Warm-start strategy tailored for the visual prompts to effectively preserve the representation capability of the vision branch. Furthermore, to guarantee high-quality pseudo-labels in every test batch, we maintain an Instance Dynamic Memory (IDM) module that stores high-quality pseudo-labels from previous test samples, and propose two novel strategies-Memory Enhancement and Memory Hallucination-to leverage IDM's high-quality instances for enhancing original predictions and hallucinating images without available pseudo-labels, respectively. Extensive experiments on cross-corruption and cross-dataset benchmarks demonstrate that our method consistently outperforms previous state-of-the-art methods, and can adapt to arbitrary cross-domain and cross-category target data. Code is available at https://github.com/gaoyingjay/ttaod_foundation.

Paperid: 327, https://arxiv.org/pdf/2510.25094.pdf GitHub

Abstract:
Zero-shot Human-Object Interaction detection aims to localize humans and objects in an image and recognize their interaction, even when specific verb-object pairs are unseen during training. Recent works have shown promising results using prompt learning with pretrained vision-language models such as CLIP, which align natural language prompts with visual features in a shared embedding space. However, existing approaches still fail to handle the visual complexity of interaction, including (1) intra-class visual diversity, where instances of the same verb appear in diverse poses and contexts, and (2) inter-class visual entanglement, where distinct verbs yield visually similar patterns. To address these challenges, we propose VDRP, a framework for Visual Diversity and Region-aware Prompt learning. First, we introduce a visual diversity-aware prompt learning strategy that injects group-wise visual variance into the context embedding. We further apply Gaussian perturbation to encourage the prompts to capture diverse visual variations of a verb. Second, we retrieve region-specific concepts from the human, object, and union regions. These are used to augment the diversity-aware prompt embeddings, yielding region-aware prompts that enhance verb-level discrimination. Experiments on the HICO-DET benchmark demonstrate that our method achieves state-of-the-art performance under four zero-shot evaluation settings, effectively addressing both intra-class diversity and inter-class visual entanglement. Code is available at https://github.com/mlvlab/VDRP.

Paperid: 328, https://arxiv.org/pdf/2510.24650.pdf GitHub

Abstract:
Site-specific disease management (SSDM) in crops has advanced rapidly through machine and deep learning (ML and DL) for real-time computer vision. Research evolved from handcrafted feature extraction to large-scale automated feature learning. With foundation models (FMs), crop disease datasets are now processed in fundamentally new ways. Unlike traditional neural networks, FMs integrate visual and textual data, interpret symptoms in text, reason about symptom-management relationships, and support interactive QA for growers and educators. Adaptive and imitation learning in robotics further enables field-based disease management. This review screened approx. 40 articles on FM applications for SSDM, focusing on large-language models (LLMs) and vision-language models (VLMs), and discussing their role in adaptive learning (AL), reinforcement learning (RL), and digital twin frameworks for targeted spraying. Key findings: (a) FMs are gaining traction with surging literature in 2023-24; (b) VLMs outpace LLMs, with a 5-10x increase in publications; (c) RL and AL are still nascent for smart spraying; (d) digital twins with RL can simulate targeted spraying virtually; (e) addressing the sim-to-real gap is critical for real-world deployment; (f) human-robot collaboration remains limited, especially in human-in-the-loop approaches where robots detect early symptoms and humans validate uncertain cases; (g) multi-modal FMs with real-time feedback will drive next-gen SSDM. For updates, resources, and contributions, visit, https://github.com/nitin-dominic/AgriPathogenDatabase, to submit papers, code, or datasets.

Paperid: 329, https://arxiv.org/pdf/2510.24214.pdf GitHub GitHub

Abstract:
Multimodal Large Language Models (MLLMs) typically process a large number of visual tokens, leading to considerable computational overhead, even though many of these tokens are redundant. Existing visual token pruning methods primarily focus on selecting the most salient tokens based on attention scores, resulting in the semantic incompleteness of the selected tokens. In this paper, we propose a novel visual token pruning strategy, called \textbf{S}aliency-\textbf{C}overage \textbf{O}riented token \textbf{P}runing for \textbf{E}fficient MLLMs (SCOPE), to jointly model both the saliency and coverage of the selected visual tokens to better preserve semantic completeness. Specifically, we introduce a set-coverage for a given set of selected tokens, computed based on the token relationships. We then define a token-coverage gain for each unselected token, quantifying how much additional coverage would be obtained by including it. By integrating the saliency score into the token-coverage gain, we propose our SCOPE score and iteratively select the token with the highest SCOPE score. We conduct extensive experiments on multiple vision-language understanding benchmarks using the LLaVA-1.5 and LLaVA-Next models. Experimental results demonstrate that our method consistently outperforms prior approaches. Our code is available at \href{https://github.com/kinredon/SCOPE}{https://github.com/kinredon/SCOPE}.

Paperid: 330, https://arxiv.org/pdf/2510.24152.pdf GitHub

Abstract:
This technical report presents our solution for the RoboSense Challenge at IROS 2025, which evaluates Vision-Language Models (VLMs) on autonomous driving scene understanding across perception, prediction, planning, and corruption detection tasks. We propose a systematic framework built on four core components. First, a Mixture-of-Prompts router classifies questions and dispatches them to task-specific expert prompts, eliminating interference across diverse question types. Second, task-specific prompts embed explicit coordinate systems, spatial reasoning rules, role-playing, Chain-of-Thought/Tree-of-Thought reasoning, and few-shot examples tailored to each task. Third, a visual assembly module composes multi-view images with object crops, magenta markers, and adaptive historical frames based on question requirements. Fourth, we configure model inference parameters (temperature, top-p, message roles) per task to optimize output quality. Implemented on Qwen2.5-VL-72B, our approach achieves 70.87% average accuracy on Phase-1 (clean data) and 72.85% on Phase-2 (corrupted data), demonstrating that structured prompting and spatial grounding substantially enhance VLM performance on safety-critical autonomous driving tasks. Code and prompt are available at https://github.com/wuaodi/UCAS-CSU-phase2.

Paperid: 331, https://arxiv.org/pdf/2510.24133.pdf GitHub GitHub

Abstract:
Despite their impressive realism, modern text-to-image models still struggle with compositionality, often failing to render accurate object counts, attributes, and spatial relations. To address this challenge, we present a training-free framework that combines an object-centric approach with self-refinement to improve layout faithfulness while preserving aesthetic quality. Specifically, we leverage large language models (LLMs) to synthesize explicit layouts from input prompts, and we inject these layouts into the image generation process, where a object-centric vision-language model (VLM) judge reranks multiple candidates to select the most prompt-aligned outcome iteratively. By unifying explicit layout-grounding with self-refine-based inference-time scaling, our framework achieves stronger scene alignment with prompts compared to recent text-to-image models. The code are available at https://github.com/gcl-inha/ReFocus.

Paperid: 332, https://arxiv.org/pdf/2510.24010.pdf GitHub

Abstract:
Foundation models have enabled rapid progress across many specialized domains by leveraging large-scale pre-training on unlabeled data, demonstrating strong generalization to a variety of downstream tasks. While such models have gained significant attention in fields like Earth Observation, their application to Mars science remains limited. A key enabler of progress in other domains has been the availability of standardized benchmarks that support systematic evaluation. In contrast, Mars science lacks such benchmarks and standardized evaluation frameworks, which have limited progress toward developing foundation models for Martian tasks. To address this gap, we introduce Mars-Bench, the first benchmark designed to systematically evaluate models across a broad range of Mars-related tasks using both orbital and surface imagery. Mars-Bench comprises 20 datasets spanning classification, segmentation, and object detection, focused on key geologic features such as craters, cones, boulders, and frost. We provide standardized, ready-to-use datasets and baseline evaluations using models pre-trained on natural images, Earth satellite data, and state-of-the-art vision-language models. Results from all analyses suggest that Mars-specific foundation models may offer advantages over general-domain counterparts, motivating further exploration of domain-adapted pre-training. Mars-Bench aims to establish a standardized foundation for developing and comparing machine learning models for Mars science. Our data, models, and code are available at: https://mars-bench.github.io/.

Paperid: 333, https://arxiv.org/pdf/2510.23511.pdf GitHub

Abstract:
In this paper, we present Dexbotic, an open-source Vision-Language-Action (VLA) model toolbox based on PyTorch. It aims to provide a one-stop VLA research service for professionals in the field of embodied intelligence. It offers a codebase that supports multiple mainstream VLA policies simultaneously, allowing users to reproduce various VLA methods with just a single environment setup. The toolbox is experiment-centric, where the users can quickly develop new VLA experiments by simply modifying the Exp script. Moreover, we provide much stronger pretrained models to achieve great performance improvements for state-of-the-art VLA policies. Dexbotic will continuously update to include more of the latest pre-trained foundation models and cutting-edge VLA models in the industry.

Paperid: 334, https://arxiv.org/pdf/2510.23482.pdf GitHub

Abstract:
Recent large vision-language models (LVLMs) can generate vision-text multimodal chain-of-thought (MCoT) traces after reinforcement fine-tuning (RFT). However, we observe that the visual information incorporated in MCoT is often inaccurate, though still yield correct answers, indicating a lack of faithfulness in the MCoT reasoning process. We attribute this unfaithfulness to the RL reward in RFT, which solely incentivizes the format of interleaved vision-text cues, ie, it encourages the model to incorporate visual information into its text reasoning steps without considering the correctness of the visual information. In this paper, we first probe the faithfulness of MCoT by measuring how much the prediction changes when its visual and textual thoughts are intervened. Surprisingly, the model's predictions remain nearly unchanged under visual intervention but change significantly under textual intervention, indicating that the visual evidence is largely ignored. To further analyze visual information, we introduce an automated LVLM-based evaluation metric that quantifies the faithfulness of visual cues from two perspectives: reliability and sufficiency. Our evaluation reveals that the visual information in current MCoT traces is simultaneously unreliable and insufficient. To address this issue, we propose a novel MCoT learning strategy termed Sufficient-Component Cause Model (SCCM) learning. This approach encourages the MCoT to generate sufficient yet minimal visual components that are independently capable of leading to correct answers. We note that the proposed SCCM is annotation-free and compatible with various RFT for MCoT in a plug-and-play manner. Empirical results demonstrate that SCCM consistently improves the visual faithfulness across a suite of fine-grained perception and reasoning benchmarks. Code is available at https://github.com/EugeneLiu01/Faithful_Thinking_with_Image.

Paperid: 335, https://arxiv.org/pdf/2510.23479.pdf GitHub

Abstract:
Vision-language alignment in multi-modal large language models (MLLMs) typically relies on supervised fine-tuning (SFT) or reinforcement learning (RL). SFT is stable and efficient but requires large-scale human annotations and cannot capture subtle preferences, while RL brings in a reward signal for training, but suffers from overhead and instability. These limitations highlight a trade-off between scalability, robustness, and alignment quality. To address this, we propose MergeMix, a training-time augmentation paradigm that bridges SFT and RL. It first applies an attention-aware image mixing via token merge with more cluster representation and spatial context, and then presents a preference-driven training paradigm for MLLMs by building preference pairs with mixed images and raw images, and optimizing via SimPO loss. As a mixup augmentation, MergeMix enhances attention consistency and efficiency, surpassing other heuristic-based methods in classification. Extensive experiments demonstrate that MergeMix achieves competitive accuracy with improved efficiency, providing a scalable approach to preference alignment in classification and MLLMs.

Paperid: 336, https://arxiv.org/pdf/2510.23253.pdf GitHub

Abstract:
As we become increasingly dependent on vision language models (VLMs) to answer questions about the world around us, there is a significant amount of research devoted to increasing both the difficulty of video question answering (VQA) datasets, and the context lengths of the models that they evaluate. The reliance on large language models as backbones has lead to concerns about potential text dominance, and the exploration of interactions between modalities is underdeveloped. How do we measure whether we're heading in the right direction, with the complexity that multi-modal models introduce? We propose a joint method of computing both feature attributions and modality scores based on Shapley values, where both the features and modalities are arbitrarily definable. Using these metrics, we compare $6$ VLM models of varying context lengths on $4$ representative datasets, focusing on multiple-choice VQA. In particular, we consider video frames and whole textual elements as equal features in the hierarchy, and the multiple-choice VQA task as an interaction between three modalities: video, question and answer. Our results demonstrate a dependence on text and show that the multiple-choice VQA task devolves into a model's ability to ignore distractors. Code available at https://github.com/sjpollard/a-video-is-not-worth-a-thousand-words.

Paperid: 337, https://arxiv.org/pdf/2510.23203.pdf GitHub

Abstract:
Accurate vertex-level contact prediction between humans and surrounding objects is a prerequisite for high fidelity human object interaction models used in robotics, AR/VR, and behavioral simulation. DECO was the first in the wild estimator for this task but is limited to binary contact maps and struggles with soft surfaces, occlusions, children, and false-positive foot contacts. We address these issues and introduce DecoDINO, a three-branch network based on DECO's framework. It uses two DINOv2 ViT-g/14 encoders, class-balanced loss weighting to reduce bias, and patch-level cross-attention for improved local reasoning. Vertex features are finally passed through a lightweight MLP with a softmax to assign semantic contact labels. We also tested a vision-language model (VLM) to integrate text features, but the simpler architecture performed better and was used instead. On the DAMON benchmark, DecoDINO (i) raises the binary-contact F1 score by 7$\%$, (ii) halves the geodesic error, and (iii) augments predictions with object-level semantic labels. Ablation studies show that LoRA fine-tuning and the dual encoders are key to these improvements. DecoDINO outperformed the challenge baseline in both tasks of the DAMON Challenge. Our code is available at https://github.com/DavidePasero/deco/tree/main.

Paperid: 338, https://arxiv.org/pdf/2510.23190.pdf GitHub

Abstract:
The widespread use of cameras in our society has created an overwhelming amount of video data, far exceeding the capacity for human monitoring. This presents a critical challenge for public safety and security, as the timely detection of anomalous or criminal events is crucial for effective response and prevention. The ability for an embodied agent to recognize unexpected events is fundamentally tied to its capacity for spatial reasoning. This paper investigates the spatial reasoning of vision-language models (VLMs) by framing anomalous action recognition as a zero-shot, language-grounded task, addressing the embodied perception challenge of interpreting dynamic 3D scenes from sparse 2D video. Specifically, we investigate whether small, pre-trained vision--LLMs can act as spatially-grounded, zero-shot anomaly detectors by converting video into text descriptions and scoring labels via textual entailment. We evaluate four open models on UCF-Crime and RWF-2000 under prompting and privacy-preserving conditions. Few-shot exemplars can improve accuracy for some models, but may increase false positives, and privacy filters -- especially full-body GAN transforms -- introduce inconsistencies that degrade accuracy. These results chart where current vision--LLMs succeed (simple, spatially salient events) and where they falter (noisy spatial cues, identity obfuscation). Looking forward, we outline concrete paths to strengthen spatial grounding without task-specific training: structure-aware prompts, lightweight spatial memory across clips, scene-graph or 3D-pose priors during description, and privacy methods that preserve action-relevant geometry. This positions zero-shot, language-grounded pipelines as adaptable building blocks for embodied, real-world video understanding. Our implementation for evaluating VLMs is publicly available at: https://github.com/pascalbenschopTU/VLLM_AnomalyRecognition

Paperid: 339, https://arxiv.org/pdf/2510.23095.pdf GitHub

Abstract:
Multimodal position encoding is essential for vision-language models, yet there has been little systematic investigation into multimodal position encoding. We conduct a comprehensive analysis of multimodal Rotary Positional Embedding (RoPE) by examining its two core components: position design and frequency allocation. Through extensive experiments, we identify three key guidelines: positional coherence, full frequency utilization, and preservation of textual priors-ensuring unambiguous layout, rich representation, and faithful transfer from the pre-trained LLM. Based on these insights, we propose Multi-Head RoPE (MHRoPE) and MRoPE-Interleave (MRoPE-I), two simple and plug-and-play variants that require no architectural changes. Our methods consistently outperform existing approaches across diverse benchmarks, with significant improvements in both general and fine-grained multimodal understanding. Code will be avaliable at https://github.com/JJJYmmm/Multimodal-RoPEs.

Paperid: 340, https://arxiv.org/pdf/2510.22370.pdf GitHub

Abstract:
In this paper, we propose Bootstrapped Language-Image Pretraining-driven Fused State Representation in Proximal Policy Optimization (BLIP-FusePPO), a novel multimodal reinforcement learning (RL) framework for autonomous lane-keeping (LK), in which semantic embeddings generated by a vision-language model (VLM) are directly fused with geometric states, LiDAR observations, and Proportional-Integral-Derivative-based (PID) control feedback within the agent observation space. The proposed method lets the agent learn driving rules that are aware of their surroundings and easy to understand by combining high-level scene understanding from the VLM with low-level control and spatial signals. Our architecture brings together semantic, geometric, and control-aware representations to make policy learning more robust. A hybrid reward function that includes semantic alignment, LK accuracy, obstacle avoidance, and speed regulation helps learning to be more efficient and generalizable. Our method is different from the approaches that only use semantic models to shape rewards. Instead, it directly embeds semantic features into the state representation. This cuts down on expensive runtime inference and makes sure that semantic guidance is always available. The simulation results show that the proposed model is better at LK stability and adaptability than the best vision-based and multimodal RL baselines in a wide range of difficult driving situations. We make our code publicly available.

Paperid: 341, https://arxiv.org/pdf/2510.22340.pdf GitHub GitHub GitHub

Abstract:
Solid geometry problem solving demands spatial mathematical reasoning that integrates spatial intelligence and symbolic reasoning. However, most existing multimodal mathematical reasoning benchmarks focus primarily on 2D plane geometry, rely on static datasets prone to data contamination and memorization, and evaluate models solely by final answers, overlooking the reasoning process. To address these limitations, we introduce DynaSolidGeo, the first dynamic benchmark for evaluating genuine spatial reasoning in Vision-Language Models (VLMs). Constructed through a semi-automatic annotation pipeline, DynaSolidGeo contains 503 expert-curated seed questions that can, in principle, dynamically generate an unbounded number of diverse multimodal text-visual instances. Beyond answer accuracy, we incorporate process evaluation based on expert-annotated reasoning chains to measure logical validity and causal coherence. Experiments across representative open-source and closed-source VLMs reveal large performance gaps, severe degradation in dynamic settings, and poor performance on tasks requiring high-level spatial intelligence, such as mental rotation and visualization. The code and dataset are available at \href{https://zgca-ai4edu.github.io/DynaSolidGeo/}{DynaSolidGeo}.

Paperid: 342, https://arxiv.org/pdf/2510.22276.pdf GitHub

Abstract:
Large-scale and high-quality image-text pair datasets play an important role in developing high-performing Vision-Language Models (VLMs). In this work, we introduce WAON, a large-scale and high-quality Japanese image-text pair dataset containing approximately 155 million examples, collected from Common Crawl. Our dataset construction pipeline employs various techniques, including filtering and deduplication, which have been shown to be effective in previous studies. To evaluate its effectiveness, we also construct WAON-Bench, a manually curated benchmark for Japanese cultural image classification, consisting of 374 classes. To assess the effectiveness of our dataset, we conduct experiments using both WAON and the Japanese subset of ReLAION, one of the most widely used vision-language datasets. We fine-tune SigLIP2, a strong multilingual model, on both datasets. The results demonstrate that WAON enhances model performance on WAON-Bench more efficiently than ReLAION and achieves higher accuracy across all evaluated benchmarks. Furthermore, the model fine-tuned on WAON achieves state-of-the-art performance on several Japanese cultural benchmarks. We release our dataset, model, and code at https://speed1313.github.io/WAON.

Paperid: 343, https://arxiv.org/pdf/2510.22201.pdf GitHub GitHub

Abstract:
Diffusion and flow matching models have emerged as powerful robot policies, enabling Vision-Language-Action (VLA) models to generalize across diverse scenes and instructions. Yet, when trained via imitation learning, their high generative capacity makes them sensitive to noise in human demonstrations: jerks, pauses, and jitter which reduce action coherence. Reduced action coherence causes instability and trajectory drift during deployment, failures that are catastrophic in fine-grained manipulation where precision is crucial. In this paper, we present Action Coherence Guidance (ACG) for VLA models, a training-free test-time guidance algorithm that improves action coherence and thereby yields performance gains. Evaluated on RoboCasa, DexMimicGen, and real-world SO-101 tasks, ACG consistently improves action coherence and boosts success rates across diverse manipulation tasks. Code and project page are available at https://github.com/DAVIAN-Robotics/ACG and https://DAVIAN-Robotics.github.io/ACG , respectively.

Paperid: 344, https://arxiv.org/pdf/2510.22127.pdf GitHub

Abstract:
Pretrained vision-language models such as CLIP achieve strong zero-shot generalization but remain vulnerable to distribution shifts caused by input corruptions. In this work, we investigate how corruptions affect CLIP's image embeddings and uncover a consistent phenomenon we term as embedding variance collapse, where both intra-class and inter-class variances shrink as corruption severity increases. We find that this collapse is closely tied to performance degradation, with inter-class variance strongly correlated with classification accuracy. To explain this phenomenon, we analyze how corruptions alter the structure of the embedding space. Our theoretical results suggest that the visual encoder tends to encode corruption-related signals, which dilute class-discriminative features and compress the representation geometry. We further show that maximizing inter-class variance, even when estimated from pseudo-labels, can provably enhance embedding quality. Based on this insight, we propose Mint, a simple test-time adaptation method that maximizes pseudo-label-based inter-class variance on the fly using a mean accumulator and a gradient accumulator. Mint operates effectively with small batch sizes and consistently improves performance across multiple corruption benchmarks and CLIP architectures. Our code is available at https://github.com/baowenxuan/Mint .

Paperid: 345, https://arxiv.org/pdf/2510.22118.pdf GitHub GitHub

Abstract:
Vision Language Models (VLMs) achieve strong performance on many vision-language tasks but often struggle with spatial reasoning$\unicode{x2014}$a prerequisite for many applications. Empirically, we find that a dataset produced by a current training data generation pipeline has a 57.6% human validation rate. These rates stem from current limitations: single-image 3D reconstruction introduces cascading modeling errors and requires wide answer tolerances, while caption-based methods require hyper-detailed annotations and suffer from generative hallucinations. We present GRAID, built on the key insight that qualitative spatial relationships can be reliably determined from 2D geometric primitives alone. By operating exclusively on 2D bounding boxes from standard object detectors, GRAID avoids both 3D reconstruction errors and generative hallucinations, resulting in datasets that are of higher quality than existing tools that produce similar datasets as validated by human evaluations. We apply our framework to the BDD100k, NuImages, and Waymo datasets, generating over 8.5 million high-quality VQA pairs creating questions spanning spatial relations, counting, ranking, and size comparisons. We evaluate one of the datasets and find it achieves 91.16% human-validated accuracy$\unicode{x2014}$compared to 57.6% on a dataset generated by recent work. Critically, we demonstrate that when trained on GRAID data, models learn spatial reasoning concepts that generalize: models fine-tuned on 6 question types improve on over 10 held-out types, with accuracy gains of 47.5% on BDD and 37.9% on NuImages for Llama 3.2B 11B, and when trained on all questions types, achieve improvements on several existing benchmarks such as BLINK. The GRAID framework, datasets, and additional information can be found $\href{this https URL}{here}$.

Paperid: 346, https://arxiv.org/pdf/2510.21817.pdf GitHub

Abstract:
Current Vision-Language-Action (VLA) models are often constrained by a rigid, static interaction paradigm, which lacks the ability to see, hear, speak, and act concurrently as well as handle real-time user interruptions dynamically. This hinders seamless embodied collaboration, resulting in an inflexible and unresponsive user experience. To address these limitations, we introduce VITA-E, a novel embodied interaction framework designed for both behavioral concurrency and nearly real-time interruption. The core of our approach is a dual-model architecture where two parallel VLA instances operate as an ``Active Model'' and a ``Standby Model'', allowing the embodied agent to observe its environment, listen to user speech, provide verbal responses, and execute actions, all concurrently and interruptibly, mimicking human-like multitasking capabilities. We further propose a ``model-as-controller'' paradigm, where we fine-tune the VLM to generate special tokens that serve as direct system-level commands, coupling the model's reasoning with the system's behavior. Experiments conducted on a physical humanoid platform demonstrate that VITA-E can reliably handle complex interactive scenarios. Our framework is compatible with various dual-system VLA models, achieving an extremely high success rate on emergency stops and speech interruptions while also successfully performing concurrent speech and action. This represents a significant step towards more natural and capable embodied assistants.

Paperid: 347, https://arxiv.org/pdf/2510.21761.pdf GitHub

Abstract:
We introduce J-ORA, a novel multimodal dataset that bridges the gap in robot perception by providing detailed object attribute annotations within Japanese human-robot dialogue scenarios. J-ORA is designed to support three critical perception tasks, object identification, reference resolution, and next-action prediction, by leveraging a comprehensive template of attributes (e.g., category, color, shape, size, material, and spatial relations). Extensive evaluations with both proprietary and open-source Vision Language Models (VLMs) reveal that incorporating detailed object attributes substantially improves multimodal perception performance compared to without object attributes. Despite the improvement, we find that there still exists a gap between proprietary and open-source VLMs. In addition, our analysis of object affordances demonstrates varying abilities in understanding object functionality and contextual relationships across different VLMs. These findings underscore the importance of rich, context-sensitive attribute annotations in advancing robot perception in dynamic environments. See project page at https://jatuhurrra.github.io/J-ORA/.

Paperid: 348, https://arxiv.org/pdf/2510.21746.pdf GitHub

Abstract:
We propose Avi, a novel 3D Vision-Language-Action (VLA) architecture that reframes robotic action generation as a problem of 3D perception and spatial reasoning, rather than low-level policy learning. While existing VLA models primarily operate on 2D visual inputs and are trained end-to-end on task-specific action policies, Avi leverages 3D point clouds and language-grounded scene understanding to compute actions through classical geometric transformations. Most notably, Avi does not train on previous action tokens, rather, we build upon a 3D Multi-modal Large Language Model (MLLM) to generate the next point cloud and explicitly calculate the actions through classical transformations. This approach enables generalizable behaviors that are robust to occlusions, camera pose variations, and changes in viewpoint. By treating the robotic decision-making process as a structured reasoning task over 3D representations, Avi bridges the gap between high-level language instructions and low-level actuation without requiring opaque policy learning. Our preliminary results highlight the potential of 3D vision-language reasoning as a foundation for scalable, robust robotic systems. Check it out at https://avi-3drobot.github.io/.

Paperid: 349, https://arxiv.org/pdf/2510.21744.pdf GitHub

Abstract:
Long-horizon robot manipulation tasks remain challenging for Vision-Language-Action (VLA) policies due to drift and exposure bias, often denoise the entire trajectory with fixed hyperparameters, causing small geometric errors to compound across stages and offering no mechanism to allocate extra test-time compute where clearances are tight. To address these challenges, we introduce FORGE-Tree, a plug-in control layer that couples a stage-aligned Diffusion Forcing (DF) head with test-time Monte Carlo Tree Diffusion (MCTD). With a frozen VLA encoder, DF aligns timesteps to subtask stages; during inference we partially denoise only a target segment while keeping other tokens frozen, turning trajectory refinement into a sequence of local edits. We then apply Monte Carlo Tree Diffusion to select the next segment to refine. A scene graph supplies priors for expansion and geometry relation-aware scoring for rollouts, yielding tree-structured denoising whose performance scales with search budget while preserving the executed prefix. Evaluation on LIBERO, FORGE-Tree improves success rate by 13.4 to 17.2 pp over the native VLA baselines with both OpenVLA and Octo-Base. Gains remain consistent under comparable compute budgets, especially on long-horizon variants. Videos available at: https://taco-group.github.io/FORGE-Tree/

Paperid: 350, https://arxiv.org/pdf/2510.21571.pdf GitHub

Abstract:
This paper presents a novel approach for pretraining robotic manipulation Vision-Language-Action (VLA) models using a large corpus of unscripted real-life video recordings of human hand activities. Treating human hand as dexterous robot end-effector, we show that "in-the-wild" egocentric human videos without any annotations can be transformed into data formats fully aligned with existing robotic V-L-A training data in terms of task granularity and labels. This is achieved by the development of a fully-automated holistic human activity analysis approach for arbitrary human hand videos. This approach can generate atomic-level hand activity segments and their language descriptions, each accompanied with framewise 3D hand motion and camera motion. We process a large volume of egocentric videos and create a hand-VLA training dataset containing 1M episodes and 26M frames. This training data covers a wide range of objects and concepts, dexterous manipulation tasks, and environment variations in real life, vastly exceeding the coverage of existing robot data. We design a dexterous hand VLA model architecture and pretrain the model on this dataset. The model exhibits strong zero-shot capabilities on completely unseen real-world observations. Additionally, fine-tuning it on a small amount of real robot action data significantly improves task success rates and generalization to novel objects in real robotic experiments. We also demonstrate the appealing scaling behavior of the model's task performance with respect to pretraining data scale. We believe this work lays a solid foundation for scalable VLA pretraining, advancing robots toward truly generalizable embodied intelligence.

Paperid: 351, https://arxiv.org/pdf/2510.21449.pdf GitHub

Abstract:
Video Anomaly Detection (VAD) aims to locate unusual activities or behaviors within videos. Recently, offline VAD has garnered substantial research attention, which has been invigorated by the progress in large language models (LLMs) and vision-language models (VLMs), offering the potential for a more nuanced understanding of anomalies. However, online VAD has seldom received attention due to real-time constraints and computational intensity. In this paper, we introduce a novel Memory-based online scoring queue scheme for Training-free VAD (MoniTor), to address the inherent complexities in online VAD. Specifically, MoniTor applies a streaming input to VLMs, leveraging the capabilities of pre-trained large-scale models. To capture temporal dependencies more effectively, we incorporate a novel prediction mechanism inspired by Long Short-Term Memory (LSTM) networks. This ensures the model can effectively model past states and leverage previous predictions to identify anomalous behaviors. Thereby, it better understands the current frame. Moreover, we design a scoring queue and an anomaly prior to dynamically store recent scores and cover all anomalies in the monitoring scenario, providing guidance for LLMs to distinguish between normal and abnormal behaviors over time. We evaluate MoniTor on two large datasets (i.e., UCF-Crime and XD-Violence) containing various surveillance and real-world scenarios. The results demonstrate that MoniTor outperforms state-of-the-art methods and is competitive with weakly supervised methods without training. Code is available at https://github.com/YsTvT/MoniTor.

Paperid: 352, https://arxiv.org/pdf/2510.21412.pdf GitHub

Abstract:
Human intelligence effortlessly interprets visual scenes along a rich spectrum of semantic dimensions. However, existing approaches to language-grounded visual concept learning are limited to a few predefined primitive axes, such as color and shape, and are typically explored in synthetic datasets. In this work, we propose a scalable framework that adaptively identifies image-related concept axes and grounds visual concepts along these axes in real-world scenes. Leveraging a pretrained vision-language model and our universal prompting strategy, our framework identifies a diverse image-related axes without any prior knowledge. Our universal concept encoder adaptively binds visual features to the discovered axes without introducing additional model parameters for each concept. To ground visual concepts along the discovered axes, we optimize a compositional anchoring objective, which ensures that each axis can be independently manipulated without affecting others. We demonstrate the effectiveness of our framework on subsets of ImageNet, CelebA-HQ, and AFHQ, showcasing superior editing capabilities across diverse real-world concepts that are too varied to be manually predefined. Our method also exhibits strong compositional generalization, outperforming existing visual concept learning and text-based editing methods. The code is available at https://github.com/whieya/Language-grounded-VCL.

Paperid: 353, https://arxiv.org/pdf/2510.21323.pdf GitHub

Abstract:
The alignment of vision-language representations endows current Vision-Language Models (VLMs) with strong multi-modal reasoning capabilities. However, the interpretability of the alignment component remains uninvestigated due to the difficulty in mapping the semantics of multi-modal representations into a unified concept set. To address this problem, we propose VL-SAE, a sparse autoencoder that encodes vision-language representations into its hidden activations. Each neuron in its hidden layer correlates to a concept represented by semantically similar images and texts, thereby interpreting these representations with a unified concept set. To establish the neuron-concept correlation, we encourage semantically similar representations to exhibit consistent neuron activations during self-supervised training. First, to measure the semantic similarity of multi-modal representations, we perform their alignment in an explicit form based on cosine similarity. Second, we construct the VL-SAE with a distance-based encoder and two modality-specific decoders to ensure the activation consistency of semantically similar representations. Experiments across multiple VLMs (e.g., CLIP, LLaVA) demonstrate the superior capability of VL-SAE in interpreting and enhancing the vision-language alignment. For interpretation, the alignment between vision and language representations can be understood by comparing their semantics with concepts. For enhancement, the alignment can be strengthened by aligning vision-language representations at the concept level, contributing to performance improvements in downstream tasks, including zero-shot image classification and hallucination elimination. Codes are available at https://github.com/ssfgunner/VL-SAE.

Paperid: 354, https://arxiv.org/pdf/2510.20812.pdf GitHub

Abstract:
Large Vision-Language Models (VLMs) have achieved remarkable progress in multimodal understanding, yet they struggle when reasoning over information-intensive images that densely interleave textual annotations with fine-grained graphical elements. The main challenges lie in precisely localizing critical cues in dense layouts and multi-hop reasoning to integrate dispersed evidence. We propose Speculative Verdict (SV), a training-free framework inspired by speculative decoding that combines multiple lightweight draft experts with a large verdict model. In the draft stage, small VLMs act as draft experts to generate reasoning paths that provide diverse localization candidates; in the verdict stage, a strong VLM synthesizes these paths to produce the final answer, minimizing computational cost while recovering correct answers. To further improve efficiency and accuracy, SV introduces a consensus expert selection mechanism that forwards only high-agreement reasoning paths to the verdict. Empirically, SV achieves consistent gains on challenging information-intensive and high-resolution visual question answering benchmarks, including InfographicVQA, ChartMuseum, ChartQAPro, and HR-Bench 4K. By synthesizing correct insights from multiple partially accurate reasoning paths, SV achieves both error correction and cost-efficiency compared to large proprietary models or training pipelines. Code is available at https://github.com/Tinaliu0123/speculative-verdict

Paperid: 355, https://arxiv.org/pdf/2510.20707.pdf GitHub GitHub

Abstract:
Recent large vision-language models (LVLMs) demonstrate remarkable capabilities in processing extended multi-modal sequences, yet the resulting key-value (KV) cache expansion creates a critical memory bottleneck that fundamentally limits deployment scalability. While existing KV cache compression methods focus on retaining high-importance KV pairs to minimize storage, they often overlook the modality-specific semantic redundancy patterns that emerge distinctively in multi-modal KV caches. In this work, we first analyze how, beyond simple importance, the KV cache in LVLMs exhibits varying levels of redundancy across attention heads. We show that relying solely on importance can only cover a subset of the full KV cache information distribution, leading to potential loss of semantic coverage. To address this, we propose \texttt{MixKV}, a novel method that mixes importance with diversity for optimized KV cache compression in LVLMs. \texttt{MixKV} adapts to head-wise semantic redundancy, selectively balancing diversity and importance when compressing KV pairs. Extensive experiments demonstrate that \texttt{MixKV} consistently enhances existing methods across multiple LVLMs. Under extreme compression (budget=64), \texttt{MixKV} improves baseline methods by an average of \textbf{5.1\%} across five multi-modal understanding benchmarks and achieves remarkable gains of \textbf{8.0\%} and \textbf{9.0\%} for SnapKV and AdaKV on GUI grounding tasks, all while maintaining comparable inference efficiency. Furthermore, \texttt{MixKV} extends seamlessly to LLMs with comparable performance gains. Our code is available at \href{https://github.com/xuyang-liu16/MixKV}{\textcolor{citeblue}{https://github.com/xuyang-liu16/MixKV}}.

Paperid: 356, https://arxiv.org/pdf/2510.20639.pdf GitHub

Abstract:
Recent progress in vision-language modeling for 3D medical imaging has been fueled by large-scale computed tomography (CT) corpora with paired free-text reports, stronger architectures, and powerful pretrained models. This has enabled applications such as automated report generation and text-conditioned 3D image synthesis. Yet, current approaches struggle with high-resolution, long-sequence volumes: contrastive pretraining often yields vision encoders that are misaligned with clinical language, and slice-wise tokenization blurs fine anatomy, reducing diagnostic performance on downstream tasks. We introduce BTB3D (Better Tokens for Better 3D), a causal convolutional encoder-decoder that unifies 2D and 3D training and inference while producing compact, frequency-aware volumetric tokens. A three-stage training curriculum enables (i) local reconstruction, (ii) overlapping-window tiling, and (iii) long-context decoder refinement, during which the model learns from short slice excerpts yet generalizes to scans exceeding 300 slices without additional memory overhead. BTB3D sets a new state-of-the-art on two key tasks: it improves BLEU scores and increases clinical F1 by 40% over CT2Rep, CT-CHAT, and Merlin for report generation; and it reduces FID by 75% and halves FVD compared to GenerateCT and MedSyn for text-to-CT synthesis, producing anatomically consistent 512*512*241 volumes. These results confirm that precise three-dimensional tokenization, rather than larger language backbones alone, is essential for scalable vision-language modeling in 3D medical imaging. The codebase is available at: https://github.com/ibrahimethemhamamci/BTB3D

Paperid: 357, https://arxiv.org/pdf/2510.20578.pdf GitHub

Abstract:
The realization of Artificial General Intelligence (AGI) necessitates Embodied AI agents capable of robust spatial perception, effective task planning, and adaptive execution in physical environments. However, current large language models (LLMs) and multimodal LLMs (MLLMs) for embodied tasks suffer from key limitations, including a significant gap between model design and agent requirements, an unavoidable trade-off between real-time latency and performance, and the use of unauthentic, offline evaluation metrics. To address these challenges, we propose EmbodiedBrain, a novel vision-language foundation model available in both 7B and 32B parameter sizes. Our framework features an agent-aligned data structure and employs a powerful training methodology that integrates large-scale Supervised Fine-Tuning (SFT) with Step-Augumented Group Relative Policy Optimization (Step-GRPO), which boosts long-horizon task success by integrating preceding steps as Guided Precursors. Furthermore, we incorporate a comprehensive reward system, including a Generative Reward Model (GRM) accelerated at the infrastructure level, to improve training efficiency. For enable thorough validation, we establish a three-part evaluation system encompassing General, Planning, and End-to-End Simulation Benchmarks, highlighted by the proposal and open-sourcing of a novel, challenging simulation environment. Experimental results demonstrate that EmbodiedBrain achieves superior performance across all metrics, establishing a new state-of-the-art for embodied foundation models. Towards paving the way for the next generation of generalist embodied agents, we open-source all of our data, model weight, and evaluating methods, which are available at https://zterobot.github.io/EmbodiedBrain.github.io.

Paperid: 358, https://arxiv.org/pdf/2510.20328.pdf GitHub GitHub

Abstract:
Humans routinely rely on memory to perform tasks, yet most robot policies lack this capability; our goal is to endow robot policies with the same ability. Naively conditioning on long observation histories is computationally expensive and brittle under covariate shift, while indiscriminate subsampling of history leads to irrelevant or redundant information. We propose a hierarchical policy framework, where the high-level policy is trained to select and track previous relevant keyframes from its experience. The high-level policy uses selected keyframes and the most recent frames when generating text instructions for a low-level policy to execute. This design is compatible with existing vision-language-action (VLA) models and enables the system to efficiently reason over long-horizon dependencies. In our experiments, we finetune Qwen2.5-VL-7B-Instruct and $π_{0.5}$ as the high-level and low-level policies respectively, using demonstrations supplemented with minimal language annotations. Our approach, MemER, outperforms prior methods on three real-world long-horizon robotic manipulation tasks that require minutes of memory. Videos and code can be found at https://jen-pan.github.io/memer/.

Paperid: 359, https://arxiv.org/pdf/2510.20134.pdf GitHub

Abstract:
Out-of-distribution (OOD) detection is critical for ensuring the reliability of deep learning models in open-world applications. While post-hoc methods are favored for their efficiency and ease of deployment, existing approaches often underexploit the rich information embedded in the model's logits space. In this paper, we propose LogitGap, a novel post-hoc OOD detection method that explicitly exploits the relationship between the maximum logit and the remaining logits to enhance the separability between in-distribution (ID) and OOD samples. To further improve its effectiveness, we refine LogitGap by focusing on a more compact and informative subset of the logit space. Specifically, we introduce a training-free strategy that automatically identifies the most informative logits for scoring. We provide both theoretical analysis and empirical evidence to validate the effectiveness of our approach. Extensive experiments on both vision-language and vision-only models demonstrate that LogitGap consistently achieves state-of-the-art performance across diverse OOD detection scenarios and benchmarks. Code is available at https://github.com/GIT-LJc/LogitGap.

Paperid: 360, https://arxiv.org/pdf/2510.19871.pdf GitHub

Abstract:
Discrete diffusion models have emerged as a promising direction for vision-language tasks, offering bidirectional context modeling and theoretical parallelization. However, their practical application is severely hindered by a train-inference discrepancy, which leads to catastrophic error cascades: initial token errors during parallel decoding pollute the generation context, triggering a chain reaction of compounding errors and leading to syntactic errors and semantic hallucinations. To address this fundamental challenge, we reframe the generation process from passive denoising to active refining. We introduce ReDiff, a refining-enhanced diffusion framework that teaches the model to identify and correct its own errors. Our approach features a two-stage training process: first, we instill a foundational revision capability by training the model to revise synthetic errors; second, we implement a novel online self-correction loop where the model is explicitly trained to revise its own flawed drafts by learning from an expert's corrections. This mistake-driven learning endows the model with the crucial ability to revisit and refine its already generated output, effectively breaking the error cascade. Extensive experiments demonstrate that ReDiff significantly improves the coherence and factual accuracy of generated content, enabling stable and efficient parallel generation far superior to traditional denoising methods. Our codes and models are available at https://rediff-hku.github.io/.

Paperid: 361, https://arxiv.org/pdf/2510.19818.pdf GitHub

Abstract:
Planning with world models offers a powerful paradigm for robotic control. Conventional approaches train a model to predict future frames conditioned on current frames and actions, which can then be used for planning. However, the objective of predicting future pixels is often at odds with the actual planning objective; strong pixel reconstruction does not always correlate with good planning decisions. This paper posits that instead of reconstructing future frames as pixels, world models only need to predict task-relevant semantic information about the future. For such prediction the paper poses world modeling as a visual question answering problem about semantic information in future frames. This perspective allows world modeling to be approached with the same tools underlying vision language models. Thus vision language models can be trained as "semantic" world models through a supervised finetuning process on image-action-text data, enabling planning for decision-making while inheriting many of the generalization and robustness properties from the pretrained vision-language models. The paper demonstrates how such a semantic world model can be used for policy improvement on open-ended robotics tasks, leading to significant generalization improvements over typical paradigms of reconstruction-based action-conditional world modeling. Website available at https://weirdlabuw.github.io/swm.

Paperid: 362, https://arxiv.org/pdf/2510.19626.pdf GitHub

Abstract:
General-purpose large Vision-Language Models (VLMs) demonstrate strong capabilities in generating detailed descriptions for natural images. However, their performance in the medical domain remains suboptimal, even for relatively straightforward tasks, primarily due to the lack of large-scale, high-quality, specialized medical imaging datasets and the neglect of the diagnostic process that progresses from coarse to fine-grained. To address the first issue, we construct the CT-RATE-VQA dataset, which has 84K QA pairs. For the second issue, we propose MedReason-R1, a medical VLM with explicit reasoning process for disease diagnosis. MedReason-R1 incorporates a novel strategy that embeds zoom-in disease region-of-interest areas into the image, highlighting the crucial role of both global localization and disease-specific details in enhancing the model's diagnostic performance. Furthermore, we introduce the GRPO reinforcement learning framework to MedReason-R1, which enables effective reasoning without relying on costly manual annotations. Compared to recent general-purpose and medical VLMs, MedReason-R1 achieves state-of-the-art performance in CT disease diagnosis while retaining generalization. The code, checkpoints, and dataset are available at: https://github.com/Leevan001/MedReason-R1

Paperid: 363, https://arxiv.org/pdf/2510.19599.pdf GitHub

Abstract:
Vision-language models (VLMs) have recently shown remarkable zero-shot performance in medical image understanding, yet their grounding ability, the extent to which textual concepts align with visual evidence, remains underexplored. In the medical domain, however, reliable grounding is essential for interpretability and clinical adoption. In this work, we present the first systematic benchmark for evaluating cross-modal interpretability in chest X-rays across seven CLIP-style VLM variants. We generate visual explanations using cross-attention and similarity-based localization maps, and quantitatively assess their alignment with radiologist-annotated regions across multiple pathologies. Our analysis reveals that: (1) while all VLM variants demonstrate reasonable localization for large and well-defined pathologies, their performance substantially degrades for small or diffuse lesions; (2) models that are pretrained on chest X-ray-specific datasets exhibit improved alignment compared to those trained on general-domain data. (3) The overall recognition ability and grounding ability of the model are strongly correlated. These findings underscore that current VLMs, despite their strong recognition ability, still fall short in clinically reliable grounding, highlighting the need for targeted interpretability benchmarks before deployment in medical practice. XBench code is available at https://github.com/Roypic/Benchmarkingattention

Paperid: 364, https://arxiv.org/pdf/2510.19559.pdf GitHub

Abstract:
Large-scale vision-language models (VLMs) such as CLIP have gained popularity for their generalizable and expressive multimodal representations. By leveraging large-scale training data with diverse textual metadata, VLMs acquire open-vocabulary capabilities, solving tasks beyond their training scope. This paper investigates the temporal awareness of VLMs, assessing their ability to position visual content in time. We introduce TIME10k, a benchmark dataset of over 10,000 images with temporal ground truth, and evaluate the time-awareness of 37 VLMs by a novel methodology. Our investigation reveals that temporal information is structured along a low-dimensional, non-linear manifold in the VLM embedding space. Based on this insight, we propose methods to derive an explicit ``timeline'' representation from the embedding space. These representations model time and its chronological progression and thereby facilitate temporal reasoning tasks. Our timeline approaches achieve competitive to superior accuracy compared to a prompt-based baseline while being computationally efficient. All code and data are available at https://tekayanidham.github.io/timeline-page/.

Paperid: 365, https://arxiv.org/pdf/2510.19400.pdf GitHub

Abstract:
Vision-language models (VLMs) are essential to Embodied AI, enabling robots to perceive, reason, and act in complex environments. They also serve as the foundation for the recent Vision-Language-Action (VLA) models. Yet most evaluations of VLMs focus on single-view settings, leaving their ability to integrate multi-view information underexplored. At the same time, multi-camera setups are increasingly standard in robotic platforms, as they provide complementary perspectives to mitigate occlusion and depth ambiguity. Whether VLMs can effectively leverage such multi-view inputs for robotic reasoning therefore remains an open question. To bridge this gap, we introduce MV-RoboBench, a benchmark specifically designed to evaluate the multi-view spatial reasoning capabilities of VLMs in robotic manipulation. MV-RoboBench consists of 1.7k manually curated QA items across eight subtasks, divided into two primary categories: spatial understanding and robotic execution. We evaluate a diverse set of existing VLMs, including both open-source and closed-source models, along with enhanced versions incorporating CoT-inspired techniques. The results show that state-of-the-art models remain far below human performance, underscoring the substantial challenges VLMs face in multi-view robotic perception. Additionally, our analysis uncovers two key findings: (i) spatial intelligence and robotic task execution are positively correlated in multi-view robotic scenarios; and (ii) strong performance on existing general-purpose single-view spatial understanding benchmarks does not reliably translate to success in the robotic spatial tasks assessed by our benchmark. We release MV-RoboBench as an open resource to foster progress in spatially grounded VLMs and VLAs, providing not only data but also a standardized evaluation protocol for multi-view embodied reasoning.

Paperid: 366, https://arxiv.org/pdf/2510.19307.pdf GitHub

Abstract:
Vision-Language Models (VLMs) have achieved remarkable progress, yet their large scale often renders them impractical for resource-constrained environments. This paper introduces Unified Reinforcement and Imitation Learning (RIL), a novel and efficient training algorithm designed to create powerful, lightweight VLMs. RIL distinctively combines the strengths of reinforcement learning with adversarial imitation learning. This enables smaller student VLMs not only to mimic the sophisticated text generation of large teacher models but also to systematically improve their generative capabilities through reinforcement signals. Key to our imitation framework is an LLM-based discriminator that adeptly distinguishes between student and teacher outputs, complemented by guidance from multiple large teacher VLMs to ensure diverse learning. This unified learning strategy, leveraging both reinforcement and imitation, empowers student models to achieve significant performance gains, making them competitive with leading closed-source VLMs. Extensive experiments on diverse vision-language benchmarks demonstrate that RIL significantly narrows the performance gap with state-of-the-art open- and closed-source VLMs and, in several instances, surpasses them.

Paperid: 367, https://arxiv.org/pdf/2510.19060.pdf GitHub

Abstract:
While vision-language models (VLMs) have advanced into detailed image description, evaluation remains a challenge. Standard metrics (e.g. CIDEr, SPICE) were designed for short texts and tuned to recognize errors that are now uncommon, such as object misidentification. In contrast, long texts require sensitivity to attribute and relation attachments and scores that localize errors to particular text spans. In this work, we introduce PoSh, a metric for detailed image description that uses scene graphs as structured rubrics to guide LLMs-as-a-Judge, producing aggregate scores grounded in fine-grained errors (e.g. mistakes in compositional understanding). PoSh is replicable, interpretable and a better proxy for human raters than existing metrics (including GPT4o-as-a-Judge). To validate PoSh, we introduce a challenging new dataset, DOCENT. This novel benchmark contains artwork, paired with expert-written references, and model-generated descriptions, augmented with granular and coarse judgments of their quality from art history students. Thus, DOCENT enables evaluating both detailed image description metrics and detailed image description itself in a challenging new domain. We show that PoSh achieves stronger correlations (+0.05 Spearman $ρ$) with the human judgments in DOCENT than the best open-weight alternatives, is robust to image type (using CapArena, an existing dataset of web imagery) and is a capable reward function, outperforming standard supervised fine-tuning. Then, using PoSh, we characterize the performance of open and closed models in describing the paintings, sketches and statues in DOCENT and find that foundation models struggle to achieve full, error-free coverage of images with rich scene dynamics, establishing a demanding new task to gauge VLM progress. Through both PoSh and DOCENT, we hope to enable advances in important areas such as assistive text generation.

Paperid: 368, https://arxiv.org/pdf/2510.18795.pdf GitHub

Abstract:
The original CLIP text encoder is limited by a maximum input length of 77 tokens, which hampers its ability to effectively process long texts and perform fine-grained semantic understanding. In addition, the CLIP text encoder lacks support for multilingual inputs. All these limitations significantly restrict its applicability across a broader range of tasks. Recent studies have attempted to replace the CLIP text encoder with an LLM-based embedder to enhance its ability in processing long texts, multilingual understanding, and fine-grained semantic comprehension. However, because the representation spaces of LLMs and the vision-language space of CLIP are pretrained independently without alignment priors, direct alignment using contrastive learning can disrupt the intrinsic vision-language alignment in the CLIP image encoder, leading to an underutilization of the knowledge acquired during pre-training. To address this challenge, we propose ProCLIP, a curriculum learning-based progressive vision-language alignment framework to effectively align the CLIP image encoder with an LLM-based embedder. Specifically, ProCLIP first distills knowledge from CLIP's text encoder into the LLM-based embedder to leverage CLIP's rich pretrained knowledge while establishing initial alignment between the LLM embedder and CLIP image encoder. Subsequently, ProCLIP further aligns the CLIP image encoder with the LLM-based embedder through image-text contrastive tuning, employing self-distillation regularization to avoid overfitting. To achieve a more effective alignment, instance semantic alignment loss and embedding structure alignment loss are employed during representation inheritance and contrastive tuning. The Code is available at https://github.com/VisionXLab/ProCLIP.

Paperid: 369, https://arxiv.org/pdf/2510.18703.pdf GitHub GitHub

Abstract:
Contrastive vision-language models such as CLIP have demonstrated strong performance across a wide range of multimodal tasks by learning from aligned image-text pairs. However, their ability to handle complex, real-world web documents remains limited, particularly in scenarios where text and images are interleaved, loosely aligned, or embedded in visual form. To address these challenges, we propose Vision-Centric Contrastive Learning (VC2L), a unified framework that models text, images, and their combinations using a single vision transformer. VC2L operates entirely in pixel space by rendering all inputs, whether textual, visual, or combined, as images, thus eliminating the need for OCR, text tokenization, or modality fusion strategy. To capture complex cross-modal relationships in multimodal web documents, VC2L employs a snippet-level contrastive learning objective that aligns consecutive multimodal segments, leveraging the inherent coherence of documents without requiring explicitly paired image-text data. To assess the effectiveness of this approach, we introduce three retrieval benchmarks, AnyCIR, SeqCIR, and CSR, designed to evaluate cross-modal retrieval, fine-grained sequential understanding, and generalization to unseen data, respectively. Empirical results show that VC2L achieves competitive or superior performance compared to CLIP-style models on both the proposed benchmarks and established datasets such as M-BEIR and MTEB. These findings underscore the potential of multimodal web data as a valuable training resource for contrastive learning and illustrate the scalability of a unified, vision-centric approach for multimodal representation learning. Code and models are available at: https://github.com/showlab/VC2L.

Paperid: 370, https://arxiv.org/pdf/2510.18632.pdf GitHub

Abstract:
Though recent advances in vision-language models (VLMs) have achieved remarkable progress across a wide range of multimodal tasks, understanding 3D spatial relationships from limited views remains a significant challenge. Previous reasoning methods typically rely on pure text (e.g., topological cognitive maps) or on 2D visual cues. However, their limited representational capacity hinders performance in specific tasks that require 3D spatial imagination. To address this limitation, we propose 3DThinker, a framework that can effectively exploits the rich geometric information embedded within images while reasoning, like humans do. Our framework is the first to enable 3D mentaling during reasoning without any 3D prior input, and it does not rely on explicitly labeled 3D data for training. Specifically, our training consists of two stages. First, we perform supervised training to align the 3D latent generated by VLM while reasoning with that of a 3D foundation model (e.g., VGGT). Then, we optimize the entire reasoning trajectory solely based on outcome signals, thereby refining the underlying 3D mentaling. Extensive experiments across multiple benchmarks show that 3DThinker consistently outperforms strong baselines and offers a new perspective toward unifying 3D representations into multimodal reasoning. Our code will be available at https://github.com/zhangquanchen/3DThinker.

Paperid: 371, https://arxiv.org/pdf/2510.18321.pdf GitHub

Abstract:
Large Vision-Language Models (LVLMs) have recently achieved impressive results in multimodal tasks such as image captioning and visual question answering. However, they remain prone to object hallucination -- generating descriptions of nonexistent or misidentified objects. Prior work has partially mitigated this via auxiliary training objectives or external modules, but challenges remain in terms of scalability, adaptability, and model independence. To address these limitations, we propose Adaptive Token Ensemble Decoding (ATED), a training-free, token-level ensemble framework that mitigates hallucination by aggregating predictions from multiple LVLMs during inference. ATED dynamically computes uncertainty-based weights for each model, reflecting their reliability at each decoding step. It also integrates diverse decoding paths to improve contextual grounding and semantic consistency. Experiments on standard hallucination detection benchmarks demonstrate that ATED significantly outperforms state-of-the-art methods, reducing hallucination without compromising fluency or relevance. Our findings highlight the benefits of adaptive ensembling and point to a promising direction for improving LVLM robustness in high-stakes applications. The code is available at https://github.com/jinlin2021/ATED.

Paperid: 372, https://arxiv.org/pdf/2510.17800.pdf GitHub

Abstract:
Large language models (LLMs) increasingly rely on long-context modeling for tasks such as document understanding, code analysis, and multi-step reasoning. However, scaling context windows to the million-token level brings prohibitive computational and memory costs, limiting the practicality of long-context LLMs. In this work, we take a different perspective-visual context scaling-to tackle this challenge. Instead of extending token-based sequences, we propose Glyph, a framework that renders long texts into images and processes them with vision-language models (VLMs). This approach substantially compresses textual input while preserving semantic information, and we further design an LLM-driven genetic search to identify optimal visual rendering configurations for balancing accuracy and compression. Through extensive experiments, we demonstrate that our method achieves 3-4x token compression while maintaining accuracy comparable to leading LLMs such as Qwen3-8B on various long-context benchmarks. This compression also leads to around 4x faster prefilling and decoding, and approximately 2x faster SFT training. Furthermore, under extreme compression, a 128K-context VLM could scale to handle 1M-token-level text tasks. In addition, the rendered text data benefits real-world multimodal tasks, such as document understanding. Our code and model are released at https://github.com/thu-coai/Glyph.

Paperid: 373, https://arxiv.org/pdf/2510.17205.pdf GitHub

Abstract:
Multimodal Large Language Models (MLLMs) have achieved strong performance across vision-language tasks, but suffer from significant computational overhead due to the quadratic growth of attention computations with the number of multimodal tokens. Though efforts have been made to prune tokens in MLLMs, \textit{they lack a fundamental understanding of how MLLMs process and fuse multimodal information.} Through systematic analysis, we uncover a \textbf{three-stage} cross-modal interaction process: (1) Shallow layers recognize task intent, with visual tokens acting as passive attention sinks; (2) Cross-modal fusion occurs abruptly in middle layers, driven by a few critical visual tokens; (3) Deep layers discard vision tokens, focusing solely on linguistic refinement. Based on these findings, we propose \emph{VisiPruner}, a training-free pruning framework that reduces up to 99\% of vision-related attention computations and 53.9\% of FLOPs on LLaVA-v1.5 7B. It significantly outperforms existing token pruning methods and generalizes across diverse MLLMs. Beyond pruning, our insights further provide actionable guidelines for training efficient MLLMs by aligning model architecture with its intrinsic layer-wise processing dynamics. Our code is available at: https://github.com/EIT-NLP/VisiPruner.

Paperid: 374, https://arxiv.org/pdf/2510.16776.pdf GitHub

Abstract:
X-ray image-based medical report generation (MRG) is a pivotal area in artificial intelligence that can significantly reduce diagnostic burdens for clinicians and patient wait times. Existing MRG models predominantly rely on Large Language Models (LLMs) to improve report generation, with limited exploration of pre-trained vision foundation models or advanced fine-tuning techniques. Mainstream frameworks either avoid fine-tuning or utilize simplistic methods like LoRA, often neglecting the potential of enhancing cross-attention mechanisms. Additionally, while Transformer-based models dominate vision-language tasks, non-Transformer architectures, such as the Mamba network, remain underexplored for medical report generation, presenting a promising avenue for future research. In this paper, we propose EMRRG, a novel X-ray report generation framework that fine-tunes pre-trained Mamba networks using parameter-efficient methods. Specifically, X-ray images are divided into patches, tokenized, and processed by an SSM-based vision backbone for feature extraction, with Partial LoRA yielding optimal performance. An LLM with a hybrid decoder generates the medical report, enabling end-to-end training and achieving strong results on benchmark datasets. Extensive experiments on three widely used benchmark datasets fully validated the effectiveness of our proposed strategies for the X-ray MRG. The source code of this paper will be released on https://github.com/Event-AHU/Medical_Image_Analysis.

Paperid: 375, https://arxiv.org/pdf/2510.16772.pdf GitHub

Abstract:
Recent research has made significant progress in localizing and editing image regions based on text. However, most approaches treat these regions in isolation, relying solely on local cues without accounting for how each part contributes to the overall visual and semantic composition. This often results in inconsistent edits, unnatural transitions, or loss of coherence across the image. In this work, we propose Region in Context, a novel framework for text-conditioned image editing that performs multilevel semantic alignment between vision and language, inspired by the human ability to reason about edits in relation to the whole scene. Our method encourages each region to understand its role within the global image context, enabling precise and harmonized changes. At its core, the framework introduces a dual-level guidance mechanism: regions are represented with full-image context and aligned with detailed region-level descriptions, while the entire image is simultaneously matched to a comprehensive scene-level description generated by a large vision-language model. These descriptions serve as explicit verbal references of the intended content, guiding both local modifications and global structure. Experiments show that it produces more coherent and instruction-aligned results. Code is available at: https://github.com/thuyvuphuong/Region-in-Context.git

Paperid: 376, https://arxiv.org/pdf/2510.16688.pdf GitHub

Abstract:
Spatial reasoning, the ability to ground language in 3D understanding, remains a persistent challenge for Vision-Language Models (VLMs). We identify two fundamental bottlenecks: inadequate 3D understanding capabilities stemming from 2D-centric pre-training, and reasoning failures induced by redundant 3D information. To address these, we first construct a Minimal Sufficient Set (MSS) of information before answering a given question: a compact selection of 3D perception results from \textit{expert models}. We introduce MSSR (Minimal Sufficient Spatial Reasoner), a dual-agent framework that implements this principle. A Perception Agent programmatically queries 3D scenes using a versatile perception toolbox to extract sufficient information, including a novel SOG (Situated Orientation Grounding) module that robustly extracts language-grounded directions. A Reasoning Agent then iteratively refines this information to pursue minimality, pruning redundant details and requesting missing ones in a closed loop until the MSS is curated. Extensive experiments demonstrate that our method, by explicitly pursuing both sufficiency and minimality, significantly improves accuracy and achieves state-of-the-art performance across two challenging benchmarks. Furthermore, our framework produces interpretable reasoning paths, offering a promising source of high-quality training data for future models. Source code is available at https://github.com/gyj155/mssr.

Paperid: 377, https://arxiv.org/pdf/2510.16500.pdf GitHub

Abstract:
A major bottleneck in off-road autonomous driving research lies in the scarcity of large-scale, high-quality datasets and benchmarks. To bridge this gap, we present ORAD-3D, which, to the best of our knowledge, is the largest dataset specifically curated for off-road autonomous driving. ORAD-3D covers a wide spectrum of terrains, including woodlands, farmlands, grasslands, riversides, gravel roads, cement roads, and rural areas, while capturing diverse environmental variations across weather conditions (sunny, rainy, foggy, and snowy) and illumination levels (bright daylight, daytime, twilight, and nighttime). Building upon this dataset, we establish a comprehensive suite of benchmark evaluations spanning five fundamental tasks: 2D free-space detection, 3D occupancy prediction, rough GPS-guided path planning, vision-language model-driven autonomous driving, and world model for off-road environments. Together, the dataset and benchmarks provide a unified and robust resource for advancing perception and planning in challenging off-road scenarios. The dataset and code will be made publicly available at https://github.com/chaytonmin/ORAD-3D.

Paperid: 378, https://arxiv.org/pdf/2510.16410.pdf GitHub

Abstract:
Bridging the gap between complex human instructions and precise 3D object grounding remains a significant challenge in vision and robotics. Existing 3D segmentation methods often struggle to interpret ambiguous, reasoning-based instructions, while 2D vision-language models that excel at such reasoning lack intrinsic 3D spatial understanding. In this paper, we introduce REALM, an innovative MLLM-agent framework that enables open-world reasoning-based segmentation without requiring extensive 3D-specific post-training. We perform segmentation directly on 3D Gaussian Splatting representations, capitalizing on their ability to render photorealistic novel views that are highly suitable for MLLM comprehension. As directly feeding one or more rendered views to the MLLM can lead to high sensitivity to viewpoint selection, we propose a novel Global-to-Local Spatial Grounding strategy. Specifically, multiple global views are first fed into the MLLM agent in parallel for coarse-level localization, aggregating responses to robustly identify the target object. Then, several close-up novel views of the object are synthesized to perform fine-grained local segmentation, yielding accurate and consistent 3D masks. Extensive experiments show that REALM achieves remarkable performance in interpreting both explicit and implicit instructions across LERF, 3D-OVS, and our newly introduced REALM3D benchmarks. Furthermore, our agent framework seamlessly supports a range of 3D interaction tasks, including object removal, replacement, and style transfer, demonstrating its practical utility and versatility. Project page: https://ChangyueShi.github.io/REALM.

Paperid: 379, https://arxiv.org/pdf/2510.16292.pdf GitHub GitHub

Abstract:
Vision-Language Models (VLMs) are integral to tasks such as image captioning and visual question answering, but their high computational cost, driven by large memory footprints and processing time, limits their scalability and real-time applicability. In this work, we propose leveraging Singular-Value Decomposition (SVD) over the joint query (Q), key (K), and value (V) weight matrices to reduce KV cache size and computational overhead. We in addition introduce an efficient rank allocation strategy that dynamically adjusts the SVD rank based on its impact on VLM accuracy, achieving a significant reduction in both memory usage and computational cost. Finally, we extend this approach by applying quantization to both VLM weights and activations, resulting in a highly efficient VLM. Our method outperforms previous approaches that rely solely on quantization or SVD by achieving more than $10\%$ accuracy improvement while consuming less hardware cost, making it better for real-time deployment on resource-constrained devices. We open source our code at \href{https://github.com/SAI-Lab-NYU/QSVD}{\texttt{https://github.com/SAI-Lab-NYU/QSVD}}.

Paperid: 380, https://arxiv.org/pdf/2510.16281.pdf GitHub

Abstract:
Reasoning Vision Language Action (VLA) models improve robotic instruction-following by generating step-by-step textual plans before low-level actions, an approach inspired by Chain-of-Thought (CoT) reasoning in language models. Yet even with a correct textual plan, the generated actions can still miss the intended outcomes in the plan, especially in out-of-distribution (OOD) scenarios. We formalize this phenomenon as a lack of embodied CoT faithfulness, and introduce a training-free, runtime policy steering method for reasoning-action alignment. Given a reasoning VLA's intermediate textual plan, our framework samples multiple candidate action sequences from the same model, predicts their outcomes via simulation, and uses a pre-trained Vision-Language Model (VLM) to select the sequence whose outcome best aligns with the VLA's own textual plan. Only executing action sequences that align with the textual reasoning turns our base VLA's natural action diversity from a source of error into a strength, boosting robustness to semantic and visual OOD perturbations and enabling novel behavior composition without costly re-training. We also contribute a reasoning-annotated extension of LIBERO-100, environment variations tailored for OOD evaluation, and demonstrate up to 15% performance gain over prior work on behavior composition tasks and scales with compute and data diversity. Project Website at: https://yilin-wu98.github.io/steering-reasoning-vla/

Paperid: 381, https://arxiv.org/pdf/2510.16263.pdf GitHub

Abstract:
The evaluation of Vision-Language-Action (VLA) agents is hindered by the coarse, end-task success metric that fails to provide precise skill diagnosis or measure robustness to real-world perturbations. This challenge is exacerbated by a fragmented data landscape that impedes reproducible research and the development of generalist models. To address these limitations, we introduce \textbf{NEBULA}, a unified ecosystem for single-arm manipulation that enables diagnostic and reproducible evaluation. NEBULA features a novel dual-axis evaluation protocol that combines fine-grained \textit{capability tests} for precise skill diagnosis with systematic \textit{stress tests} that measure robustness. A standardized API and a large-scale, aggregated dataset are provided to reduce fragmentation and support cross-dataset training and fair comparison. Using NEBULA, we demonstrate that top-performing VLAs struggle with key capabilities such as spatial reasoning and dynamic adaptation, which are consistently obscured by conventional end-task success metrics. By measuring both what an agent can do and when it does so reliably, NEBULA provides a practical foundation for robust, general-purpose embodied agents.

Paperid: 382, https://arxiv.org/pdf/2510.15710.pdf GitHub

Abstract:
Medical diagnostic applications require models that can process multimodal medical inputs (images, patient histories, lab results) and generate diverse outputs including both textual reports and visual content (annotations, segmentation masks, and images). Despite this need, existing medical AI systems disrupt this unified process: medical image understanding models interpret images but cannot generate visual outputs, while medical image generation models synthesize images but cannot provide textual explanations. This leads to gaps in data representation, feature integration, and task-level multimodal capabilities. To this end, we propose a multi-level framework that draws inspiration from diagnostic workflows through the Observation-Knowledge-Analysis (OKA) paradigm. Specifically, at the observation level, we construct UniMed-5M, a dataset comprising over 5.6M samples that reformat diverse unimodal data into multimodal pairs for foundational observation. At the knowledge level, we propose Progressive Curriculum Learning that systematically introduces medical multimodal knowledge. At the analysis level, we introduce UniMedVL, the first medical unified multimodal model for the simultaneous analysis of image understanding and generation tasks within a single architecture. UniMedVL achieves superior performance on five medical image understanding benchmarks, while matching specialized models in generation quality across eight medical imaging modalities. Crucially, our unified architecture enables bidirectional knowledge sharing: generation tasks enhance visual understanding features, demonstrating that integrating traditionally separate capabilities within a single medical framework unlocks improvements across diverse medical vision-language tasks. Code is available at https://github.com/uni-medical/UniMedVL.

Paperid: 383, https://arxiv.org/pdf/2510.14979.pdf GitHub

Abstract:
The edifice of native Vision-Language Models (VLMs) has emerged as a rising contender to typical modular VLMs, shaped by evolving model architectures and training paradigms. Yet, two lingering clouds cast shadows over its widespread exploration and promotion: (-) What fundamental constraints set native VLMs apart from modular ones, and to what extent can these barriers be overcome? (-) How to make research in native VLMs more accessible and democratized, thereby accelerating progress in the field. In this paper, we clarify these challenges and outline guiding principles for constructing native VLMs. Specifically, one native VLM primitive should: (i) effectively align pixel and word representations within a shared semantic space; (ii) seamlessly integrate the strengths of formerly separate vision and language modules; (iii) inherently embody various cross-modal properties that support unified vision-language encoding, aligning, and reasoning. Hence, we launch NEO, a novel family of native VLMs built from first principles, capable of rivaling top-tier modular counterparts across diverse real-world scenarios. With only 390M image-text examples, NEO efficiently develops visual perception from scratch while mitigating vision-language conflicts inside a dense and monolithic model crafted from our elaborate primitives. We position NEO as a cornerstone for scalable and powerful native VLMs, paired with a rich set of reusable components that foster a cost-effective and extensible ecosystem. Our code and models are publicly available at: https://github.com/EvolvingLMMs-Lab/NEO.

Paperid: 384, https://arxiv.org/pdf/2510.14978.pdf GitHub

Abstract:
Recent image editing models have achieved impressive results while following natural language editing instructions, but they rely on supervised fine-tuning with large datasets of input-target pairs. This is a critical bottleneck, as such naturally occurring pairs are hard to curate at scale. Current workarounds use synthetic training pairs that leverage the zero-shot capabilities of existing models. However, this can propagate and magnify the artifacts of the pretrained model into the final trained model. In this work, we present a new training paradigm that eliminates the need for paired data entirely. Our approach directly optimizes a few-step diffusion model by unrolling it during training and leveraging feedback from vision-language models (VLMs). For each input and editing instruction, the VLM evaluates if an edit follows the instruction and preserves unchanged content, providing direct gradients for end-to-end optimization. To ensure visual fidelity, we incorporate distribution matching loss (DMD), which constrains generated images to remain within the image manifold learned by pretrained models. We evaluate our method on standard benchmarks and include an extensive ablation study. Without any paired data, our method performs on par with various image editing diffusion models trained on extensive supervised paired data, under the few-step setting. Given the same VLM as the reward model, we also outperform RL-based techniques like Flow-GRPO.

Paperid: 385, https://arxiv.org/pdf/2510.14583.pdf GitHub

Abstract:
Vision-language models have achieved remarkable success in cross-modal understanding. Yet, these models remain limited to object-level or region-level grounding, lacking the capability for pixel-precise keypoint comprehension through natural language. We introduce a novel framework for pixel level grounding. The framework consists of two complementary components: a Point Descriptor that generates rich, contextual descriptions of individual keypoints, and a Point Localizer that regresses precise pixel coordinates from these descriptions. Unlike prior work that relies on templated prompts or keypoint names, our approach produces free-form, coarse-to-fine descriptions that situate keypoints within their visual context. Since there is no available dataset to train such a system, we introduce LlamaPointInPart, a carefully curated dataset of 20K+ image-keypoint-description triplets synthesized from multiple vision-language models, capturing multi-scale information from scene-level context to visual features around the keypoint. For cross-category generalization, we optimize the Point Descriptor on AP-10K via GRPO, using the frozen Point Localizer as a reward model to produce descriptions that maximize localization accuracy. To evaluate our results we establish a new evaluation protocol. Instead of comparing the text description produced by our method to the ground truth, we use the localizer to determine how close is the predicted point generated to the ground truth point. Experiments demonstrate superior performance compared to baseline models on LlamaPointInPart.The bidirectional nature of our framework should enable future applications in both keypoint-guided image understanding and language-guided precise localization. Our code and dataset are publicly available at https://github.com/matanr/Talking_Points.

Paperid: 386, https://arxiv.org/pdf/2510.14304.pdf GitHub

Abstract:
Large Vision-Language Models (LVLMs) have recently shown promising results on various multimodal tasks, even achieving human-comparable performance in certain cases. Nevertheless, LVLMs remain prone to hallucinations -- they often rely heavily on a single modality or memorize training data without properly grounding their outputs. To address this, we propose a training-free, tri-layer contrastive decoding with watermarking, which proceeds in three steps: (1) select a mature layer and an amateur layer among the decoding layers, (2) identify a pivot layer using a watermark-related question to assess whether the layer is visually well-grounded, and (3) apply tri-layer contrastive decoding to generate the final output. Experiments on public benchmarks such as POPE, MME and AMBER demonstrate that our method achieves state-of-the-art performance in reducing hallucinations in LVLMs and generates more visually grounded responses.

Paperid: 387, https://arxiv.org/pdf/2510.13778.pdf GitHub

Abstract:
We introduce InternVLA-M1, a unified framework for spatial grounding and robot control that advances instruction-following robots toward scalable, general-purpose intelligence. Its core idea is spatially guided vision-language-action training, where spatial grounding serves as the critical link between instructions and robot actions. InternVLA-M1 employs a two-stage pipeline: (i) spatial grounding pre-training on over 2.3M spatial reasoning data to determine ``where to act'' by aligning instructions with visual, embodiment-agnostic positions, and (ii) spatially guided action post-training to decide ``how to act'' by generating embodiment-aware actions through plug-and-play spatial prompting. This spatially guided training recipe yields consistent gains: InternVLA-M1 outperforms its variant without spatial guidance by +14.6% on SimplerEnv Google Robot, +17% on WidowX, and +4.3% on LIBERO Franka, while demonstrating stronger spatial reasoning capability in box, point, and trace prediction. To further scale instruction following, we built a simulation engine to collect 244K generalizable pick-and-place episodes, enabling a 6.2% average improvement across 200 tasks and 3K+ objects. In real-world clustered pick-and-place, InternVLA-M1 improved by 7.3%, and with synthetic co-training, achieved +20.6% on unseen objects and novel configurations. Moreover, in long-horizon reasoning-intensive scenarios, it surpassed existing works by over 10%. These results highlight spatially guided training as a unifying principle for scalable and resilient generalist robots. Code and models are available at https://github.com/InternRobotics/InternVLA-M1.

Paperid: 388, https://arxiv.org/pdf/2510.13237.pdf GitHub

Abstract:
Vision-Language-Action (VLA) models have achieved revolutionary progress in robot learning, enabling robots to execute complex physical robot tasks from natural language instructions. Despite this progress, their adversarial robustness remains underexplored. In this work, we propose both adversarial patch attack and corresponding defense strategies for VLA models. We first introduce the Embedding Disruption Patch Attack (EDPA), a model-agnostic adversarial attack that generates patches directly placeable within the camera's view. In comparison to prior methods, EDPA can be readily applied to different VLA models without requiring prior knowledge of the model architecture, or the controlled robotic manipulator. EDPA constructs these patches by (i) disrupting the semantic alignment between visual and textual latent representations, and (ii) maximizing the discrepancy of latent representations between adversarial and corresponding clean visual inputs. Through the optimization of these objectives, EDPA distorts the VLA's interpretation of visual information, causing the model to repeatedly generate incorrect actions and ultimately result in failure to complete the given robotic task. To counter this, we propose an adversarial fine-tuning scheme for the visual encoder, in which the encoder is optimized to produce similar latent representations for both clean and adversarially perturbed visual inputs. Extensive evaluations on the widely recognized LIBERO robotic simulation benchmark demonstrate that EDPA substantially increases the task failure rate of cutting-edge VLA models, while our proposed defense effectively mitigates this degradation. The codebase is accessible via the homepage at https://edpa-attack.github.io/.

Paperid: 389, https://arxiv.org/pdf/2510.13054.pdf GitHub

Abstract:
Vision-Language-Action models (VLAs) hold immense promise for enabling generalist robot manipulation. However, the best way to build them remains an open question. Current approaches often add complexity, such as modifying the existing vocabulary of a Vision-Language Model (VLM) with action tokens or introducing special action heads. Curiously, the simplest strategy of representing actions directly as text has remained largely unexplored. This work introduces VLA-0 to investigate this idea. We find that VLA-0 is not only effective; it is surprisingly powerful. With the right design, VLA-0 outperforms more involved models. On LIBERO, a popular benchmark for evaluating VLAs, VLA-0 outperforms all existing methods trained on the same robotic data, including $π_0.5$-KI, OpenVLA-OFT and SmolVLA. Furthermore, without large-scale robotics-specific training, it outperforms methods trained on large-scale robotic data, like $π_0.5$-KI, $π_0$, GR00T-N1 and MolmoAct. These findings also translate to the real world, where VLA-0 outperforms SmolVLA, a VLA model pre-trained on large-scale real data. This paper summarizes our unexpected findings and spells out the specific techniques required to unlock the high performance of this simple yet potent VLA design. Visual results, code, and trained models are provided here: https://vla0.github.io/.

Paperid: 390, https://arxiv.org/pdf/2510.12953.pdf GitHub

Abstract:
Recent medical vision-language models have shown promise on tasks such as VQA, report generation, and anomaly detection. However, most are adapted to structured adult imaging and underperform in fetal ultrasound, which poses challenges of multi-view image reasoning, numerous diseases, and image diversity. To bridge this gap, we introduce FetalMind, a medical AI system tailored to fetal ultrasound for both report generation and diagnosis. Guided by clinical workflow, we propose Salient Epistemic Disentanglement (SED), which injects an expert-curated bipartite graph into the model to decouple view-disease associations and to steer preference selection along clinically faithful steps via reinforcement learning. This design mitigates variability across diseases and heterogeneity across views, reducing learning bottlenecks while aligning the model's inference with obstetric practice. To train FetalMind at scale, we curate FetalSigma-1M dataset, the first large-scale fetal ultrasound report corpus, comprising 20K reports from twelve medical centers, addressing the scarcity of domain data. Extensive experiments show that FetalMind outperforms open- and closed-source baselines across all gestational stages, achieving +14% average gains and +61.2% higher accuracy on critical conditions while remaining efficient, stable, and scalable. Project Page: https://hexiao0275.github.io/FetalMind.

Paperid: 391, https://arxiv.org/pdf/2510.12789.pdf GitHub

Abstract:
Although recent advances in visual generation have been remarkable, most existing architectures still depend on distinct encoders for images and text. This separation constrains diffusion models' ability to perform cross-modal reasoning and knowledge transfer. Prior attempts to bridge this gap often use the last layer information from VLM, employ multiple visual encoders, or train large unified models jointly for text and image generation, which demands substantial computational resources and large-scale data, limiting its accessibility.We present UniFusion, a diffusion-based generative model conditioned on a frozen large vision-language model (VLM) that serves as a unified multimodal encoder. At the core of UniFusion is the Layerwise Attention Pooling (LAP) mechanism that extracts both high level semantics and low level details from text and visual tokens of a frozen VLM to condition a diffusion generative model. We demonstrate that LAP outperforms other shallow fusion architectures on text-image alignment for generation and faithful transfer of visual information from VLM to the diffusion model which is key for editing. We propose VLM-Enabled Rewriting Injection with Flexibile Inference (VERIFI), which conditions a diffusion transformer (DiT) only on the text tokens generated by the VLM during in-model prompt rewriting. VERIFI combines the alignment of the conditioning distribution with the VLM's reasoning capabilities for increased capabilities and flexibility at inference. In addition, finetuning on editing task not only improves text-image alignment for generation, indicative of cross-modality knowledge transfer, but also exhibits tremendous generalization capabilities. Our model when trained on single image editing, zero-shot generalizes to multiple image references further motivating the unified encoder design of UniFusion.

Paperid: 392, https://arxiv.org/pdf/2510.12784.pdf GitHub

Abstract:
Recently, remarkable progress has been made in Unified Multimodal Models (UMMs), which integrate vision-language generation and understanding capabilities within a single framework. However, a significant gap exists where a model's strong visual understanding often fails to transfer to its visual generation. A model might correctly understand an image based on user instructions, yet be unable to generate a faithful image from text prompts. This phenomenon directly raises a compelling question: Can a model achieve self-improvement by using its understanding module to reward its generation module? To bridge this gap and achieve self-improvement, we introduce SRUM, a self-rewarding post-training framework that can be directly applied to existing UMMs of various designs. SRUM creates a feedback loop where the model's own understanding module acts as an internal ``evaluator'', providing corrective signals to improve its generation module, without requiring additional human-labeled data. To ensure this feedback is comprehensive, we designed a global-local dual reward system. To tackle the inherent structural complexity of images, this system offers multi-scale guidance: a \textbf{global reward} ensures the correctness of the overall visual semantics and layout, while a \textbf{local reward} refines fine-grained, object-level fidelity. SRUM leads to powerful capabilities and shows strong generalization, boosting performance on T2I-CompBench from 82.18 to \textbf{88.37} and on T2I-ReasonBench from 43.82 to \textbf{46.75}. Overall, our work establishes a powerful new paradigm for enabling a UMMs' understanding module to guide and enhance its own generation via self-rewarding.

Paperid: 393, https://arxiv.org/pdf/2510.12283.pdf GitHub

Abstract:
Almost all previous text-to-video retrieval works ideally assume that videos are pre-trimmed with short durations containing solely text-related content. However, in practice, videos are typically untrimmed in long durations with much more complicated background content. Therefore, in this paper, we focus on the more practical yet challenging task of Partially Relevant Video Retrieval (PRVR), which aims to retrieve partially relevant untrimmed videos with the given query. To tackle this task, we propose a novel framework that distills generalization knowledge from a powerful large-scale vision-language pre-trained model and transfers it to a lightweight, task-specific PRVR network. Specifically, we introduce a Dual Learning framework with Dynamic Knowledge Distillation (DL-DKD++), where a large teacher model provides supervision to a compact dual-branch student network. The student model comprises two branches: an inheritance branch that absorbs transferable knowledge from the teacher, and an exploration branch that learns task-specific information from the PRVR dataset to address domain gaps. To further enhance learning, we incorporate a dynamic soft-target construction mechanism. By replacing rigid hard-target supervision with adaptive soft targets that evolve during training, our method enables the model to better capture the fine-grained, partial relevance between videos and queries. Experiment results demonstrate that our proposed model achieves state-of-the-art performance on TVR, ActivityNet, and Charades-STA datasets for PRVR. The code is available at https://github.com/HuiGuanLab/DL-DKD.

Paperid: 394, https://arxiv.org/pdf/2510.12190.pdf GitHub

Abstract:
Recent advances in end-to-end (E2E) autonomous driving have been enabled by training on diverse large-scale driving datasets, yet autonomous driving models still struggle in out-of-distribution (OOD) scenarios. The COOOL benchmark targets this gap by encouraging hazard understanding beyond closed taxonomies, and the 2COOOL challenge extends it to generating human-interpretable incident reports. We present a hierarchical reasoning framework for incident report generation from dashcam videos that integrates frame-level captioning, incident frame detection, and fine-grained reasoning within vision-language models (VLMs). We further improve factual accuracy and readability through model ensembling and a Blind A/B Scoring selection protocol. On the official 2COOOL open leaderboard, our method ranks 2nd among 29 teams and achieves the best CIDEr-D score, producing accurate and coherent incident narratives. These results indicate that hierarchical reasoning with VLMs is a promising direction for accident analysis and for broader understanding of safety-critical traffic events. The implementation and code are available at https://github.com/riron1206/kaggle-2COOOL-2nd-Place-Solution.

Paperid: 395, https://arxiv.org/pdf/2510.12119.pdf GitHub

Abstract:
The widespread adoption of Retrieval-Augmented Image Generation (RAIG) has raised significant concerns about the unauthorized use of private image datasets. While these systems have shown remarkable capabilities in enhancing generation quality through reference images, protecting visual datasets from unauthorized use in such systems remains a challenging problem. Traditional digital watermarking approaches face limitations in RAIG systems, as the complex feature extraction and recombination processes fail to preserve watermark signals during generation. To address these challenges, we propose ImageSentinel, a novel framework for protecting visual datasets in RAIG. Our framework synthesizes sentinel images that maintain visual consistency with the original dataset. These sentinels enable protection verification through randomly generated character sequences that serve as retrieval keys. To ensure seamless integration, we leverage vision-language models to generate the sentinel images. Experimental results demonstrate that ImageSentinel effectively detects unauthorized dataset usage while preserving generation quality for authorized applications. Code is available at https://github.com/luo-ziyuan/ImageSentinel.

Paperid: 396, https://arxiv.org/pdf/2510.11718.pdf GitHub

Abstract:
Recent advances in Large Language Models (LLMs) and Vision Language Models (VLMs) have shown significant progress in mathematical reasoning, yet they still face a critical bottleneck with problems requiring visual assistance, such as drawing auxiliary lines or plotting functions to solve the problems. Most LLMs and VLMs are constrained to text-only reasoning chains, while multimodal unified models that can generate interleaved text and images lack the necessary precision and controllability for such tasks. To address this, we propose CodePlot-CoT, a code-driven Chain-of-Thought paradigm for "thinking with images" in mathematics. Our approach leverages the VLM to generate text reasoning as well as executable plotting code, which is then rendered into images as "visual thought", to solve mathematical problems. To achieve this, we first construct Math-VR, the first large-scale, bilingual dataset and benchmark for Mathematics problems with Visual Reasoning, comprising 178K samples. Second, to create high-quality training data, we develop a state-of-the-art image-to-code converter specialized for parsing complex mathematical figures into codes. Finally, using these training data, we train the CodePlot-CoT model for solving mathematical problems. Experimental results show that our model achieves up to 21% increase over base model on our new benchmark, fully validating the efficacy of our proposed code-driven reasoning paradigm. Our work opens a new direction for multimodal mathematical reasoning and provides the community with the first large-scale dataset, comprehensive benchmark, and strong approach for such problems. To facilitate future research, we make our datasets, code, and pretrained models publicly available at https://github.com/HKU-MMLab/Math-VR-CodePlot-CoT.

Paperid: 397, https://arxiv.org/pdf/2510.11660.pdf GitHub

Abstract:
While Vision-Language-Action (VLA) models have demonstrated impressive capabilities in robotic manipulation, their performance in complex reasoning and long-horizon task planning is limited by data scarcity and model capacity. To address this, we introduce ManiAgent, an agentic architecture for general manipulation tasks that achieves end-to-end output from task descriptions and environmental inputs to robotic manipulation actions. In this framework, multiple agents involve inter-agent communication to perform environmental perception, sub-task decomposition and action generation, enabling efficient handling of complex manipulation scenarios. Evaluations show ManiAgent achieves an 86.8% success rate on the SimplerEnv benchmark and 95.8% on real-world pick-and-place tasks, enabling efficient data collection that yields VLA models with performance comparable to those trained on human-annotated datasets.The project webpage is available at https://yi-yang929.github.io/ManiAgent/.

Paperid: 398, https://arxiv.org/pdf/2510.11520.pdf GitHub

Abstract:
Walking assistance in extreme or complex environments remains a significant challenge for people with blindness or low vision (BLV), largely due to the lack of a holistic scene understanding. Motivated by the real-world needs of the BLV community, we build mmWalk, a simulated multi-modal dataset that integrates multi-view sensor and accessibility-oriented features for outdoor safe navigation. Our dataset comprises 120 manually controlled, scenario-categorized walking trajectories with 62k synchronized frames. It contains over 559k panoramic images across RGB, depth, and semantic modalities. Furthermore, to emphasize real-world relevance, each trajectory involves outdoor corner cases and accessibility-specific landmarks for BLV users. Additionally, we generate mmWalkVQA, a VQA benchmark with over 69k visual question-answer triplets across 9 categories tailored for safe and informed walking assistance. We evaluate state-of-the-art Vision-Language Models (VLMs) using zero- and few-shot settings and found they struggle with our risk assessment and navigational tasks. We validate our mmWalk-finetuned model on real-world datasets and show the effectiveness of our dataset for advancing multi-modal walking assistance.

Paperid: 399, https://arxiv.org/pdf/2510.11456.pdf GitHub

Abstract:
Existing Infrared and Visible Image Fusion (IVIF) methods typically assume high-quality inputs. However, when handing degraded images, these methods heavily rely on manually switching between different pre-processing techniques. This decoupling of degradation handling and image fusion leads to significant performance degradation. In this paper, we propose a novel VLM-Guided Degradation-Coupled Fusion network (VGDCFusion), which tightly couples degradation modeling with the fusion process and leverages vision-language models (VLMs) for degradation-aware perception and guided suppression. Specifically, the proposed Specific-Prompt Degradation-Coupled Extractor (SPDCE) enables modality-specific degradation awareness and establishes a joint modeling of degradation suppression and intra-modal feature extraction. In parallel, the Joint-Prompt Degradation-Coupled Fusion (JPDCF) facilitates cross-modal degradation perception and couples residual degradation filtering with complementary cross-modal feature fusion. Extensive experiments demonstrate that our VGDCFusion significantly outperforms existing state-of-the-art fusion approaches under various degraded image scenarios. Our code is available at https://github.com/Lmmh058/VGDCFusion.

Paperid: 400, https://arxiv.org/pdf/2510.10982.pdf GitHub

Abstract:
Recent AI regulations call for data that remain useful for innovation while resistant to misuse, balancing utility with protection at the model level. Existing approaches either perturb data to make it unlearnable or retrain models to suppress transfer, but neither governs inference by unknown models, and both typically require control over training. We propose non-transferable examples (NEs), a training-free and data-agnostic input-side usage-control mechanism. We recode inputs within a model-specific low-sensitivity subspace, preserving outputs for the authorized model while reducing performance on unauthorized models through subspace misalignment. We establish formal bounds that guarantee utility for the authorized model and quantify deviation for unauthorized ones, with the Hoffman-Wielandt inequality linking degradation to spectral differences. Empirically, NEs retain performance on diverse vision backbones and state-of-the-art vision-language models under common preprocessing, whereas non-target models collapse even with reconstruction attempts. These results establish NEs as a practical means to preserve intended data utility while preventing unauthorized exploitation. Our project is available at https://trusted-system-lab.github.io/model-specificity

Paperid: 401, https://arxiv.org/pdf/2510.10880.pdf GitHub

Abstract:
Vision-language models (VLMs) have advanced rapidly, yet their capacity for image-grounded geolocation in open-world conditions, a task that is challenging and of demand in real life, has not been comprehensively evaluated. We present EarthWhere, a comprehensive benchmark for VLM image geolocation that evaluates visual recognition, step-by-step reasoning, and evidence use. EarthWhere comprises 810 globally distributed images across two complementary geolocation scales: WhereCountry (i.e., 500 multiple-choice question-answering, with country-level answer and panoramas) and WhereStreet (i.e., 310 fine-grained street-level identification tasks requiring multi-step reasoning with optional web search). For evaluation, we adopt the final-prediction metrics: location accuracies within k km (Acc@k) for coordinates and hierarchical path scores for textual localization. Beyond this, we propose to explicitly score intermediate reasoning chains using human-verified key visual clues and a Shapley-reweighted thinking score that attributes credit to each clue's marginal contribution. We benchmark 13 state-of-the-art VLMs with web searching tools on our EarthWhere and report different types of final answer accuracies as well as the calibrated model thinking scores. Overall, Gemini-2.5-Pro achieves the best average accuracy at 56.32%, while the strongest open-weight model, GLM-4.5V, reaches 34.71%. We reveal that web search and reasoning do not guarantee improved performance when visual clues are limited, and models exhibit regional biases, achieving up to 42.7% higher scores in certain areas than others. These findings highlight not only the promise but also the persistent challenges of models to mitigate bias and achieve robust, fine-grained localization. We open-source our benchmark at https://github.com/UCSC-VLAA/EarthWhere.

Paperid: 402, https://arxiv.org/pdf/2510.10715.pdf GitHub

Abstract:
Creative generation is the synthesis of new, surprising, and valuable samples that reflect user intent yet cannot be envisioned in advance. This task aims to extend human imagination, enabling the discovery of visual concepts that exist in the unexplored spaces between familiar domains. While text-to-image diffusion models excel at rendering photorealistic scenes that faithfully match user prompts, they still struggle to generate genuinely novel content. Existing approaches to enhance generative creativity either rely on interpolation of image features, which restricts exploration to predefined categories, or require time-intensive procedures such as embedding optimization or model fine-tuning. We propose VLM-Guided Adaptive Negative-Prompting, a training-free, inference-time method that promotes creative image generation while preserving the validity of the generated object. Our approach utilizes a vision-language model (VLM) that analyzes intermediate outputs of the generation process and adaptively steers it away from conventional visual concepts, encouraging the emergence of novel and surprising outputs. We evaluate creativity through both novelty and validity, using statistical metrics in the CLIP embedding space. Through extensive experiments, we show consistent gains in creative novelty with negligible computational overhead. Moreover, unlike existing methods that primarily generate single objects, our approach extends to complex scenarios, such as generating coherent sets of creative objects and preserving creativity within elaborate compositional prompts. Our method integrates seamlessly into existing diffusion pipelines, offering a practical route to producing creative outputs that venture beyond the constraints of textual descriptions.

Paperid: 403, https://arxiv.org/pdf/2510.10671.pdf GitHub

Abstract:
Image-Language Foundation Models (ILFMs) have demonstrated remarkable success in vision-language understanding, providing transferable multimodal representations that generalize across diverse downstream image-based tasks. The advancement of video-text research has spurred growing interest in extending image-based models to the video domain. This paradigm, termed as image-to-video transfer learning, effectively mitigates the substantial data and computational demands compared to training video-language models from scratch while achieves comparable or even stronger model performance. This survey provides the first comprehensive review of this emerging field, which begins by summarizing the widely used ILFMs and their capabilities. We then systematically classify existing image-to-video transfer learning techniques into two broad root categories (frozen features and adapted features), along with numerous fine-grained subcategories, based on the paradigm for transferring image understanding capability to video tasks. Building upon the task-specific nature of image-to-video transfer, this survey methodically elaborates these strategies and details their applications across a spectrum of video-text learning tasks, ranging from fine-grained settings (e.g., spatio-temporal video grounding) to coarse-grained ones (e.g., video question answering). We further present a detailed experimental analysis to investigate the efficacy of different image-to-video transfer learning paradigms on a range of downstream video understanding tasks. Finally, we identify prevailing challenges and highlight promising directions for future research. By offering a comprehensive and structured overview, this survey aims to establish a structured roadmap for advancing video-text learning based on existing ILFM, and to inspire future research directions in this rapidly evolving domain. Github repository is available.

Paperid: 404, https://arxiv.org/pdf/2510.10487.pdf GitHub

Abstract:
Vision-Language Models (VLMs) integrate visual knowledge with the analytical capabilities of Large Language Models (LLMs) through supervised visual instruction tuning, using image-question-answer triplets. However, the potential of VLMs trained without supervised instruction remains largely unexplored. This study validates that VLMs possess inherent self-refinement capabilities, enabling them to generate high-quality supervised data without external inputs and thereby learn autonomously. Specifically, to stimulate the self-refinement ability of VLMs, we propose a self-refinement framework based on a Triangular Consistency principle: within the image-query-answer triangle, any masked elements should be consistently and accurately reconstructed. The framework involves three steps: (1) We enable the instruction generation ability of VLMs by adding multi-task instruction tuning like image$\rightarrow$question-answer or image-answer$\rightarrow$question. (2) We generate image-query-answer triplets from unlabeled images and use the Triangular Consistency principle for filtering. (3) The model is further updated using the filtered synthetic data. To investigate the underlying mechanisms behind this self-refinement capability, we conduct a theoretical analysis from a causal perspective. Using the widely recognized LLaVA-1.5 as our baseline, our experiments reveal that the model can autonomously achieve consistent, though deliberately modest, improvements across multiple benchmarks without any external supervision, such as human annotations or environmental feedback. We expect that the insights of this study on the self-refinement ability of VLMs can inspire future research on the learning mechanism of VLMs. Code is available at https://github.com/dengyl20/SRF-LLaVA-1.5.

Paperid: 405, https://arxiv.org/pdf/2510.10274.pdf GitHub

Abstract:
Successful generalist Vision-Language-Action (VLA) models rely on effective training across diverse robotic platforms with large-scale, cross-embodiment, heterogeneous datasets. To facilitate and leverage the heterogeneity in rich, diverse robotic data sources, we propose a novel Soft Prompt approach with minimally added parameters, by infusing prompt learning concepts into cross-embodiment robot learning and introducing separate sets of learnable embeddings for each distinct data source. These embeddings serve as embodiment-specific prompts, which in unity empower VLA models with effective exploitation of varying cross-embodiment features. Our new X-VLA, a neat flow-matching-based VLA architecture, relies exclusively on soft-prompted standard Transformer encoders, enjoying both scalability and simplicity. Evaluated across 6 simulations as well as 3 real-world robots, our 0.9B instantiation-X-VLA-0.9B simultaneously achieves SOTA performance over a sweep of benchmarks, demonstrating superior results on a wide axes of capabilities, from flexible dexterity to quick adaptation across embodiments, environments, and tasks. Website: https://thu-air-dream.github.io/X-VLA/

Paperid: 406, https://arxiv.org/pdf/2510.10100.pdf GitHub GitHub

Abstract:
Unsupervised Federated Learning (UFL) aims to collaboratively train a global model across distributed clients without sharing data or accessing label information. Previous UFL works have predominantly focused on representation learning and clustering tasks. Recently, vision language models (e.g., CLIP) have gained significant attention for their powerful zero-shot prediction capabilities. Leveraging this advancement, classification problems that were previously infeasible under the UFL paradigm now present promising new opportunities, yet remain largely unexplored. In this paper, we extend UFL to the classification problem with CLIP for the first time and propose a novel method, \underline{\textbf{Fed}}erated \underline{\textbf{Co}}operative \underline{\textbf{P}}seudo \underline{\textbf{L}}abeling (\textbf{FedCoPL}). Specifically, clients estimate and upload their pseudo label distribution, and the server adjusts and redistributes them to avoid global imbalance among classes. Moreover, we introduce a partial prompt aggregation protocol for effective collaboration and personalization. In particular, visual prompts containing general image features are aggregated at the server, while text prompts encoding personalized knowledge are retained locally. Extensive experiments demonstrate the superior performance of our FedCoPL compared to baseline methods. Our code is available at \href{https://github.com/krumpguo/FedCoPL}{https://github.com/krumpguo/FedCoPL}.

Paperid: 407, https://arxiv.org/pdf/2510.09608.pdf GitHub

Abstract:
Vision-language models (VLMs) could power real-time assistants and autonomous agents, but they face a critical challenge: understanding near-infinite video streams without escalating latency and memory usage. Processing entire videos with full attention leads to quadratic computational costs and poor performance on long videos. Meanwhile, simple sliding window methods are also flawed, as they either break coherence or suffer from high latency due to redundant recomputation. In this paper, we introduce StreamingVLM, a model designed for real-time, stable understanding of infinite visual input. Our approach is a unified framework that aligns training with streaming inference. During inference, we maintain a compact KV cache by reusing states of attention sinks, a short window of recent vision tokens, and a long window of recent text tokens. This streaming ability is instilled via a simple supervised fine-tuning (SFT) strategy that applies full attention on short, overlapped video chunks, which effectively mimics the inference-time attention pattern without training on prohibitively long contexts. For evaluation, we build Inf-Streams-Eval, a new benchmark with videos averaging over two hours that requires dense, per-second alignment between frames and text. On Inf-Streams-Eval, StreamingVLM achieves a 66.18% win rate against GPT-4O mini and maintains stable, real-time performance at up to 8 FPS on a single NVIDIA H100. Notably, our SFT strategy also enhances general VQA abilities without any VQA-specific fine-tuning, improving performance on LongVideoBench by +4.30 and OVOBench Realtime by +5.96. Code is available at https://github.com/mit-han-lab/streaming-vlm.

Paperid: 408, https://arxiv.org/pdf/2510.09607.pdf GitHub

Abstract:
Vision-Language Action (VLA) models significantly advance robotic manipulation by leveraging the strong perception capabilities of pretrained vision-language models (VLMs). By integrating action modules into these pretrained models, VLA methods exhibit improved generalization. However, training them from scratch is costly. In this work, we propose a simple yet effective distillation-based framework that equips VLMs with action-execution capability by transferring knowledge from pretrained small action models. Our architecture retains the original VLM structure, adding only an action token and a state encoder to incorporate physical inputs. To distill action knowledge, we adopt a two-stage training strategy. First, we perform lightweight alignment by mapping VLM hidden states into the action space of the small action model, enabling effective reuse of its pretrained action decoder and avoiding expensive pretraining. Second, we selectively fine-tune the language model, state encoder, and action modules, enabling the system to integrate multimodal inputs with precise action generation. Specifically, the action token provides the VLM with a direct handle for predicting future actions, while the state encoder allows the model to incorporate robot dynamics not captured by vision alone. This design yields substantial efficiency gains over training large VLA models from scratch. Compared with previous state-of-the-art methods, our method achieves 97.3% average success rate on LIBERO (11.8% improvement) and 93.5% on LIBERO-LONG (24.5% improvement). In real-world experiments across five manipulation tasks, our method consistently outperforms the teacher model, achieving 82.0% success rate (17% improvement), which demonstrate that action distillation effectively enables VLMs to generate precise actions while substantially reducing training costs.

Paperid: 409, https://arxiv.org/pdf/2510.09358.pdf GitHub GitHub

Abstract:
Multi-modal keyphrase prediction (MMKP) aims to advance beyond text-only methods by incorporating multiple modalities of input information to produce a set of conclusive phrases. Traditional multi-modal approaches have been proven to have significant limitations in handling the challenging absence and unseen scenarios. Additionally, we identify shortcomings in existing benchmarks that overestimate model capability due to significant overlap in training tests. In this work, we propose leveraging vision-language models (VLMs) for the MMKP task. Firstly, we use two widely-used strategies, e.g., zero-shot and supervised fine-tuning (SFT) to assess the lower bound performance of VLMs. Next, to improve the complex reasoning capabilities of VLMs, we adopt Fine-tune-CoT, which leverages high-quality CoT reasoning data generated by a teacher model to finetune smaller models. Finally, to address the "overthinking" phenomenon, we propose a dynamic CoT strategy which adaptively injects CoT data during training, allowing the model to flexibly leverage its reasoning capabilities during the inference stage. We evaluate the proposed strategies on various datasets and the experimental results demonstrate the effectiveness of the proposed approaches. The code is available at https://github.com/bytedance/DynamicCoT.

Paperid: 410, https://arxiv.org/pdf/2510.09285.pdf GitHub

Abstract:
While Reinforcement Learning with Verifiable Rewards (RLVR) has advanced the reasoning capabilities of Large Vision-Language Models (LVLMs), most existing methods in multimodal reasoning neglect the critical role of visual perception within the RLVR optimization process. In this paper, we undertake a pioneering exploration of multimodal RLVR through the novel perspective of token perception, which measures the visual dependency of each generated token. With a granular analysis of Chain-of-Thought (CoT) processes, we uncover two key insights: first, token perception in a rollout trajectory is sparsely distributed, where only a small fraction of tokens have high visual dependency for visually-grounded reasoning; second, different trajectories exhibit significant divergence in their overall visual dependency. Based on these observations, we propose Visually-Perceptive Policy Optimization (VPPO), a novel policy gradient algorithm that explicitly leverages token perception to refine the learning signal. Specifically, VPPO achieves this through a dual mechanism: it reweights a trajectory's advantage by its overall visual dependency, and focuses policy updates exclusively on perceptually pivotal tokens. On a comprehensive suite of eight perception and reasoning benchmarks, VPPO demonstrates substantial gains over leading open-source RL-tuned models, with its effectiveness consistently validated across 7B and 32B model scales. Our findings not only establish a new token-level perceptual perspective for analyzing multimodal RLVR but also present a novel and effective optimization strategy to significantly enhance the multimodal reasoning capabilities of LVLMs.

Paperid: 411, https://arxiv.org/pdf/2510.09269.pdf GitHub

Abstract:
Recent advances in vision-language-action (VLA) models have greatly improved embodied AI, enabling robots to follow natural language instructions and perform diverse tasks. However, their reliance on uncurated training datasets raises serious security concerns. Existing backdoor attacks on VLAs mostly assume white-box access and result in task failures instead of enforcing specific actions. In this work, we reveal a more practical threat: attackers can manipulate VLAs by simply injecting physical objects as triggers into the training dataset. We propose goal-oriented backdoor attacks (GoBA), where the VLA behaves normally in the absence of physical triggers but executes predefined and goal-oriented actions in the presence of physical triggers. Specifically, based on a popular VLA benchmark LIBERO, we introduce BadLIBERO that incorporates diverse physical triggers and goal-oriented backdoor actions. In addition, we propose a three-level evaluation that categorizes the victim VLA's actions under GoBA into three states: nothing to do, try to do, and success to do. Experiments show that GoBA enables the victim VLA to successfully achieve the backdoor goal in 97 percentage of inputs when the physical trigger is present, while causing zero performance degradation on clean inputs. Finally, by investigating factors related to GoBA, we find that the action trajectory and trigger color significantly influence attack performance, while trigger size has surprisingly little effect. The code and BadLIBERO dataset are accessible via the project page at https://goba-attack.github.io/.

Paperid: 412, https://arxiv.org/pdf/2510.09256.pdf GitHub

Abstract:
To determine whether using discrete semantic entropy (DSE) to reject questions likely to generate hallucinations can improve the accuracy of black-box vision-language models (VLMs) in radiologic image based visual question answering (VQA). This retrospective study evaluated DSE using two publicly available, de-identified datasets: (i) the VQA-Med 2019 benchmark (500 images with clinical questions and short-text answers) and (ii) a diagnostic radiology dataset (206 cases: 60 computed tomography scans, 60 magnetic resonance images, 60 radiographs, 26 angiograms) with corresponding ground-truth diagnoses. GPT-4o and GPT-4.1 answered each question 15 times using a temperature of 1.0. Baseline accuracy was determined using low-temperature answers (temperature 0.1). Meaning-equivalent responses were grouped using bidirectional entailment checks, and DSE was computed from the relative frequencies of the resulting semantic clusters. Accuracy was recalculated after excluding questions with DSE > 0.6 or > 0.3. p-values and 95% confidence intervals were obtained using bootstrap resampling and a Bonferroni-corrected threshold of p < .004 for statistical significance. Across 706 image-question pairs, baseline accuracy was 51.7% for GPT-4o and 54.8% for GPT-4.1. After filtering out high-entropy questions (DSE > 0.3), accuracy on the remaining questions was 76.3% (retained questions: 334/706) for GPT-4o and 63.8% (retained questions: 499/706) for GPT-4.1 (both p < .001). Accuracy gains were observed across both datasets and largely remained statistically significant after Bonferroni correction. DSE enables reliable hallucination detection in black-box VLMs by quantifying semantic inconsistency. This method significantly improves diagnostic answer accuracy and offers a filtering strategy for clinical VLM applications.

Paperid: 413, https://arxiv.org/pdf/2510.09228.pdf GitHub

Abstract:
Adverse weather conditions such as haze, rain, and snow significantly degrade the quality of images and videos, posing serious challenges to intelligent transportation systems (ITS) that rely on visual input. These degradations affect critical applications including autonomous driving, traffic monitoring, and surveillance. This survey presents a comprehensive review of image and video restoration techniques developed to mitigate weather-induced visual impairments. We categorize existing approaches into traditional prior-based methods and modern data-driven models, including CNNs, transformers, diffusion models, and emerging vision-language models (VLMs). Restoration strategies are further classified based on their scope: single-task models, multi-task/multi-weather systems, and all-in-one frameworks capable of handling diverse degradations. In addition, we discuss day and night time restoration challenges, benchmark datasets, and evaluation protocols. The survey concludes with an in-depth discussion on limitations in current research and outlines future directions such as mixed/compound-degradation restoration, real-time deployment, and agentic AI frameworks. This work aims to serve as a valuable reference for advancing weather-resilient vision systems in smart transportation environments. Lastly, to stay current with rapid advancements in this field, we will maintain regular updates of the latest relevant papers and their open-source implementations at https://github.com/ChaudharyUPES/A-comprehensive-review-on-Multi-weather-restoration

Paperid: 414, https://arxiv.org/pdf/2510.09078.pdf GitHub

Abstract:
Generative artificial intelligence (AI) has made unprecedented advances in vision language models over the past two years. During the generative process, new samples (images) are generated from an unknown high-dimensional distribution. Markov Chain Monte Carlo (MCMC) methods are particularly effective in drawing samples from such complex, high-dimensional distributions. This makes MCMC methods an integral component for models like EBMs, ensuring accurate sample generation. Gradient-based optimization is at the core of modern generative models. The update step during the optimization forms a Markov chain where the new update depends only on the current state. This allows exploration of the parameter space in a memoryless manner, thus combining the benefits of gradient-based optimization and MCMC sampling. MCMC methods have shown an equally important role in physically based rendering where complex light paths are otherwise quite challenging to sample from simple importance sampling techniques. A lot of research is dedicated towards bringing physical realism to samples (images) generated from diffusion-based generative models in a data-driven manner, however, a unified framework connecting these techniques is still missing. In this course, we take the first steps toward understanding each of these components and exploring how MCMC could potentially serve as a bridge, linking these closely related areas of research. Our course aims to provide necessary theoretical and practical tools to guide students, researchers and practitioners towards the common goal of generative physically based rendering. All Jupyter notebooks with demonstrations associated to this tutorial can be found on the project webpage: https://sinbag.github.io/mcmc/

Paperid: 415, https://arxiv.org/pdf/2510.08818.pdf GitHub

Abstract:
Video large language models (Vid-LLMs), which excel in diverse video-language tasks, can be effectively constructed by adapting image-pretrained vision-language models (VLMs). However, this adaptation remains challenging, as it requires processing dense and temporally extended visual inputs that exceed the capacity of image-based models. This paper identifies the perception bottleneck and token overload as key challenges in extending image-based VLMs to the video domain. To address these issues, we propose D-CoDe, a training-free adaptation framework that incorporates dynamic compression and question decomposition. Specifically, dynamic compression alleviates the perception bottleneck through adaptive selection of representative frames and content-aware aggregation of spatial tokens, thereby reducing redundancy while preserving informative content. In parallel, question decomposition mitigates token overload by reformulating the original query into sub-questions, guiding the model to focus on distinct aspects of the video and enabling more comprehensive understanding. Experiments demonstrate that D-CoDe effectively improves video understanding across various benchmarks. Furthermore, strong performance on the challenging long-video benchmark highlights the potential of D-CoDe in handling complex video-language tasks. Code is available at https://github.com/hukcc/D-CoDe.

Paperid: 416, https://arxiv.org/pdf/2510.08673.pdf GitHub

Abstract:
Camera-centric understanding and generation are two cornerstones of spatial intelligence, yet they are typically studied in isolation. We present Puffin, a unified camera-centric multimodal model that extends spatial awareness along the camera dimension. Puffin integrates language regression and diffusion-based generation to interpret and create scenes from arbitrary viewpoints. To bridge the modality gap between cameras and vision-language, we introduce a novel paradigm that treats camera as language, enabling thinking with camera. This guides the model to align spatially grounded visual cues with photographic terminology while reasoning across geometric context. Puffin is trained on Puffin-4M, a large-scale dataset of 4 million vision-language-camera triplets. We incorporate both global camera parameters and pixel-wise camera maps, yielding flexible and reliable spatial generation. Experiments demonstrate Puffin superior performance over specialized models for camera-centric generation and understanding. With instruction tuning, Puffin generalizes to diverse cross-view tasks such as spatial imagination, world exploration, and photography guidance. We will release the code, models, dataset pipeline, and benchmark to advance multimodal spatial intelligence research.

Paperid: 417, https://arxiv.org/pdf/2510.08668.pdf GitHub GitHub

Abstract:
Real-world clinical decision-making grapples with integrating information from diverse data modalities, including medical text, 2D/3D images, and video, leading to inefficiencies and potential diagnostic oversights. While generalist vision-language models (VLMs) offer promise, their medical development faces challenges of opaque pipelines, data scarcity, and architectural inflexibility. Here we present Hulu-Med, a transparent medical VLM that unifies understanding across all these modalities. Built upon a unified patch-based vision encoder and an LLM decoder, Hulu-Med was progressively trained on 16.7 million (M) samples to scale from 2D to 3D and video comprehension. The medical-aware token reduction enables efficient training, requiring only 4,000 to 40,000 GPU hours for 7B to 32B parameter variants. Extensive evaluation across 30 benchmarks exhibits state-of-the-art performance, surpassing leading open-source models and competing with proprietary systems in tasks spanning visual question-answering, medical report generation, and complex reasoning in multilingual and rare disease scenarios. By open-sourcing our complete pipeline, we establish that high-performance medical VLM can be achieved transparently, providing a foundational tool for accessible and impactful clinical AI. Code is released on \href{https://github.com/ZJUI-AI4H/Hulu-Med}{https://github.com/ZJUI-AI4H/Hulu-Med}.

Paperid: 418, https://arxiv.org/pdf/2510.08567.pdf GitHub

Abstract:
Vision language models (VLMs) are increasingly deployed as controllers with access to external tools for complex reasoning and decision-making, yet their effectiveness remains limited by the scarcity of high-quality multimodal trajectories and the cost of manual annotation. We address this challenge with a vision-centric agent tuning framework that automatically synthesizes multimodal trajectories, generates step-wise preference pairs, and trains a VLM controller for robust tool-use reasoning. Our pipeline first constructs M-TRACE, a large-scale dataset of 28.5K multimodal tasks with 177K verified trajectories, enabling imitation-based trajectory tuning. Building on this, we develop MATRIX Agent, a controller finetuned on M-TRACE for step-wise tool reasoning. To achieve finer alignment, we further introduce Pref-X, a set of 11K automatically generated preference pairs, and optimize MATRIX on it via step-wise preference learning. Across three benchmarks, Agent-X, GTA, and GAIA, MATRIX consistently surpasses both open- and closed-source VLMs, demonstrating scalable and effective multimodal tool use. Our data and code is avaliable at https://github.com/mbzuai-oryx/MATRIX.

Paperid: 419, https://arxiv.org/pdf/2510.08531.pdf GitHub GitHub

Abstract:
Spatial reasoning remains a fundamental challenge for Vision-Language Models (VLMs), with current approaches struggling to achieve robust performance despite recent advances. We identify that this limitation stems from a critical gap: existing methods attempt to learn spatial reasoning directly without establishing the hierarchical foundations of perception and understanding. To address this challenge, we present a comprehensive methodology for building spatial intelligence progressively. We introduce SpatialLadder-26k, a multimodal dataset containing 26,610 samples spanning object localization, single image, multi-view, and video spatial reasoning tasks, constructed through a standardized pipeline that ensures systematic coverage across modalities. Building on this dataset, we design a three-stage progressive training framework that (1) establishes spatial perception through object localization, (2) develops spatial understanding through multi-dimensional spatial tasks, and (3) strengthens complex reasoning via reinforcement learning with verifiable rewards. This approach yields SpatialLadder, a 3B-parameter model that achieves state-of-the-art performance on spatial reasoning benchmarks, with 23.4% average improvement over the base model, surpassing GPT-4o by 20.8% and Gemini-2.0-Flash by 10.1%. Notably, SpatialLadder maintains strong generalization with 7.2% improvement on out-of-domain benchmarks, demonstrating that progressive training from perception to reasoning is essential for robust spatial intelligence.

Paperid: 420, https://arxiv.org/pdf/2510.08510.pdf GitHub

Abstract:
Large Vision Language Models (LVLMs) have recently emerged as powerful architectures capable of understanding and reasoning over both visual and textual information. These models typically rely on two key components: a Vision Transformer (ViT) and a Large Language Model (LLM). ViT encodes visual content into a sequence of image tokens and serves as the perceptual front-end -- the eyes of the model. In contrast, the LLM interprets these tokens to perform high-level reasoning, generates responses, and functions as the cognitive core -- the brain of the model. However, it remains unclear which visual tokens contribute most significantly to understanding and reasoning, and how effectively these signals are propagated from ViT to the LLM. While most existing works have focused on identifying attention sinks, low-semantic tokens receiving disproportionately high attention, within the LLM, we shift the focus to the vision encoder by identifying a class of high-norm visual tokens from ViT, referred to as ViT attention sinks -- a problem that has been rarely studied but is indeed very important for LVLMs. Our findings show that these ViT sinks encapsulate high-level semantic concepts from images, allowing the LLM to perform more effective understanding and reasoning. Despite their importance, these sink tokens are often overlooked in existing LVLM architectures. To explore their contribution, we present both qualitative and quantitative analyses of the information embedded in these sink tokens. We also propose both training-free and training-based approaches to better leverage how this information is interpreted by the LLM, and to what extent. By explicitly utilizing these tokens, we demonstrate substantial improvements across a range of LVLMs and visual reasoning tasks, highlighting the untapped potential of ViT attention sinks in enhancing visual reasoning.

Paperid: 421, https://arxiv.org/pdf/2510.08464.pdf GitHub

Abstract:
Vision-Language-Action (VLA) models have advanced robotic capabilities but remain challenging to deploy on resource-limited hardware. Pruning has enabled efficient compression of large language models (LLMs), yet it is largely understudied in robotics. Surprisingly, we observe that pruning VLA models leads to drastic degradation and increased safety violations. We introduce GLUESTICK, a post-pruning recovery method that restores much of the original model's functionality while retaining sparsity benefits. Our method performs a one-time interpolation between the dense and pruned models in weight-space to compute a corrective term. This correction is used during inference by each pruned layer to recover lost capabilities with minimal overhead. GLUESTICK requires no additional training, is agnostic to the pruning algorithm, and introduces a single hyperparameter that controls the tradeoff between efficiency and accuracy. Across diverse VLA architectures and tasks in manipulation and navigation, GLUESTICK achieves competitive memory efficiency while substantially recovering success rates and reducing safety violations. Additional material can be found at: https://gluestick-vla.github.io/.

Paperid: 422, https://arxiv.org/pdf/2510.08132.pdf GitHub

Abstract:
Pre-trained Vision-Language Models (VLMs) exhibit strong generalization capabilities, enabling them to recognize a wide range of objects across diverse domains without additional training. However, they often retain irrelevant information beyond the requirements of specific downstream tasks, raising concerns about computational efficiency and potential information leakage. This has motivated growing interest in approximate unlearning, which aims to selectively remove unnecessary knowledge while preserving overall model performance. Existing approaches to approximate unlearning have primarily focused on class unlearning, where a VLM is retrained to fail to recognize specified object classes while maintaining accuracy for others. However, merely forgetting object classes is often insufficient in practical applications. For instance, an autonomous driving system should accurately recognize real cars while avoiding misrecognition of illustrated cars depicted in roadside advertisements as real cars, which could be hazardous. In this paper, we introduce Approximate Domain Unlearning (ADU), a novel problem setting that requires reducing recognition accuracy for images from specified domains (e.g., illustration) while preserving accuracy for other domains (e.g., real). ADU presents new technical challenges: due to the strong domain generalization capability of pre-trained VLMs, domain distributions are highly entangled in the feature space, making naive approaches based on penalizing target domains ineffective. To tackle this limitation, we propose a novel approach that explicitly disentangles domain distributions and adaptively captures instance-specific domain information. Extensive experiments show that our approach outperforms baselines built upon VLM tuning techniques, paving the way for practical and fine-grained unlearning in VLMs. Code: https://kodaikawamura.github.io/Domain_Unlearning/.

Paperid: 423, https://arxiv.org/pdf/2510.07791.pdf GitHub

Abstract:
Recently spatial-temporal intelligence of Visual-Language Models (VLMs) has attracted much attention due to its importance for Autonomous Driving, Embodied AI and General Artificial Intelligence. Existing spatial-temporal benchmarks mainly focus on egocentric perspective reasoning with images/video context, or geographic perspective reasoning with graphics context (eg. a map), thus fail to assess VLMs' geographic spatial-temporal intelligence with both images/video and graphics context, which is important for areas like traffic management and emergency response. To address the gaps, we introduce Geo-Temporal Reasoning benchmark (GTR-Bench), a novel challenge for geographic temporal reasoning of moving targets in a large-scale camera network. GTR-Bench is more challenging as it requires multiple perspective switches between maps and videos, joint reasoning across multiple videos with non-overlapping fields of view, and inference over spatial-temporal regions that are unobserved by any video context. Evaluations of more than 10 popular VLMs on GTR-Bench demonstrate that even the best proprietary model, Gemini-2.5-Pro (34.9%), significantly lags behind human performance (78.61%) on geo-temporal reasoning. Moreover, our comprehensive analysis on GTR-Bench reveals three primary deficiencies of current models for geo-temporal reasoning. (1) VLMs' reasoning is impaired by an imbalanced utilization of spatial-temporal context. (2) VLMs are weak in temporal forecasting, which leads to worse performance on temporal-emphasized tasks than on spatial-emphasized tasks. (3) VLMs lack the proficiency to comprehend or align the map data with multi-view video inputs. We believe GTR-Bench offers valuable insights and opens up new opportunities for research and applications in spatial-temporal intelligence. Benchmark and code will be released at https://github.com/X-Luffy/GTR-Bench.

Paperid: 424, https://arxiv.org/pdf/2510.07135.pdf GitHub

Abstract:
Remote Sensing Vision-Language Models (RSVLMs) have shown remarkable potential thanks to large-scale pretraining, achieving strong zero-shot performance on various tasks. However, their ability to generalize in low-data regimes, such as few-shot learning, remains insufficiently explored. In this work, we present the first structured benchmark for evaluating few-shot adaptation methods on RSVLMs. We conduct comprehensive experiments across ten remote sensing scene classification datasets, applying five widely used few-shot adaptation strategies to three state-of-the-art RSVLMs with varying backbones. Our findings reveal that models with similar zero-shot performance can exhibit markedly different behavior under few-shot adaptation, with some RSVLMs being inherently more amenable to such adaptation than others. The variability of performance and the absence of a clear winner among existing methods highlight the need for the development of more robust methods for few-shot adaptation tailored to RS. To facilitate future research, we provide a reproducible benchmarking framework and open-source code to systematically evaluate RSVLMs under few-shot conditions. The source code is publicly available on Github: https://github.com/elkhouryk/fewshot_RSVLMs

Paperid: 425, https://arxiv.org/pdf/2510.07134.pdf GitHub

Abstract:
Embodied Visual Tracking (EVT) is a fundamental ability that underpins practical applications, such as companion robots, guidance robots and service assistants, where continuously following moving targets is essential. Recent advances have enabled language-guided tracking in complex and unstructured scenes. However, existing approaches lack explicit spatial reasoning and effective temporal memory, causing failures under severe occlusions or in the presence of similar-looking distractors. To address these challenges, we present TrackVLA++, a novel Vision-Language-Action (VLA) model that enhances embodied visual tracking with two key modules, a spatial reasoning mechanism and a Target Identification Memory (TIM). The reasoning module introduces a Chain-of-Thought paradigm, termed Polar-CoT, which infers the target's relative position and encodes it as a compact polar-coordinate token for action prediction. Guided by these spatial priors, the TIM employs a gated update strategy to preserve long-horizon target memory, ensuring spatiotemporal consistency and mitigating target loss during extended occlusions. Extensive experiments show that TrackVLA++ achieves state-of-the-art performance on public benchmarks across both egocentric and multi-camera settings. On the challenging EVT-Bench DT split, TrackVLA++ surpasses the previous leading approach by 5.1 and 12, respectively. Furthermore, TrackVLA++ exhibits strong zero-shot generalization, enabling robust real-world tracking in dynamic and occluded scenarios.

Paperid: 426, https://arxiv.org/pdf/2510.07035.pdf GitHub

Abstract:
Molecular representation learning plays a crucial role in advancing applications such as drug discovery and material design. Existing work leverages 2D and 3D modalities of molecular information for pre-training, aiming to capture comprehensive structural and geometric insights. However, these methods require paired 2D and 3D molecular data to train the model effectively and prevent it from collapsing into a single modality, posing limitations in scenarios where a certain modality is unavailable or computationally expensive to generate. To overcome this limitation, we propose FlexMol, a flexible molecule pre-training framework that learns unified molecular representations while supporting single-modality input. Specifically, inspired by the unified structure in vision-language models, our approach employs separate models for 2D and 3D molecular data, leverages parameter sharing to improve computational efficiency, and utilizes a decoder to generate features for the missing modality. This enables a multistage continuous learning process where both modalities contribute collaboratively during training, while ensuring robustness when only one modality is available during inference. Extensive experiments demonstrate that FlexMol achieves superior performance across a wide range of molecular property prediction tasks, and we also empirically demonstrate its effectiveness with incomplete data. Our code and data are available at https://github.com/tewiSong/FlexMol.

Paperid: 427, https://arxiv.org/pdf/2510.06710.pdf GitHub

Abstract:
Recent progress in vision and language foundation models has significantly advanced multimodal understanding, reasoning, and generation, inspiring a surge of interest in extending such capabilities to embodied settings through vision-language-action (VLA) models. Yet, most VLA models are still trained with supervised fine-tuning (SFT), which struggles to generalize under distribution shifts due to error accumulation. Reinforcement learning (RL) offers a promising alternative by directly optimizing task performance through interaction, but existing attempts remain fragmented and lack a unified platform for fair and systematic comparison across model architectures and algorithmic designs. To address this gap, we introduce RLinf-VLA, a unified and efficient framework for scalable RL training of VLA models. The system adopts a highly flexible resource allocation design that addresses the challenge of integrating rendering, training, and inference in RL+VLA training. In particular, for GPU-parallelized simulators, RLinf-VLA implements a novel hybrid fine-grained pipeline allocation mode, achieving a 1.61x-1.88x speedup in training. Through a unified interface, RLinf-VLA seamlessly supports diverse VLA architectures (e.g., OpenVLA, OpenVLA-OFT), multiple RL algorithms (e.g., PPO, GRPO), and various simulators (e.g., ManiSkill, LIBERO). In simulation, a unified model achieves 98.11\% across 130 LIBERO tasks and 97.66\% across 25 ManiSkill tasks. Beyond empirical performance, our study distills a set of best practices for applying RL to VLA training and sheds light on emerging patterns in this integration. Furthermore, we present preliminary deployment on a real-world Franka robot, where RL-trained policies exhibit stronger generalization than those trained with SFT. We envision RLinf-VLA as a foundation to accelerate and standardize research on embodied intelligence.

Paperid: 428, https://arxiv.org/pdf/2510.06590.pdf GitHub

Abstract:
Visual tokenization remains a core challenge in unifying visual understanding and generation within the autoregressive paradigm. Existing methods typically employ tokenizers in discrete latent spaces to align with the tokens from large language models, where the quantization errors can limit semantic expressiveness and degrade the capability of vision-language understanding. To address this, we introduce MingTok, a new family of visual tokenizers with a continuous latent space, for unified autoregressive generation and understanding. While understanding tasks favor discriminative high-dimensional features, generation tasks prefer compact low-level codes. Thus, to reconcile these competing demands, MingTok adopts a three-stage sequential architecture involving low-level encoding, semantic expansion, and visual reconstruction. Built on top of it, Ming-UniVision eliminates the need for task-specific visual representations, and unifies diverse vision-language tasks under a single autoregrsssive prediction paradigm. By formulating both understanding and generation as next-token prediction in a shared continuous space, it seamlessly supports multi-round, in-context tasks such as iterative understanding, generation and editing. Empirically, we find that using a unified continuous visual representation reconciles the competing requirements on the tokenizers by the understanding and generation tasks, thereby leading to state-of-the-art level performance across both domains. We hope our findings will facilitate unified visual tokenization in the continuous domain. Inference code and model weights are released to benefit community.

Paperid: 429, https://arxiv.org/pdf/2510.05038.pdf GitHub

Abstract:
Multimodal encoders have pushed the boundaries of visual document retrieval, matching textual query tokens directly to image patches and achieving state-of-the-art performance on public benchmarks. Recent models relying on this paradigm have massively scaled the sizes of their query and document representations, presenting obstacles to deployment and scalability in real-world pipelines. Furthermore, purely vision-centric approaches may be constrained by the inherent modality gap still exhibited by modern vision-language models. In this work, we connect these challenges to the paradigm of hybrid retrieval, investigating whether a lightweight dense text retriever can enhance a stronger vision-centric model. Existing hybrid methods, which rely on coarse-grained fusion of ranks or scores, fail to exploit the rich interactions within each model's representation space. To address this, we introduce Guided Query Refinement (GQR), a novel test-time optimization method that refines a primary retriever's query embedding using guidance from a complementary retriever's scores. Through extensive experiments on visual document retrieval benchmarks, we demonstrate that GQR allows vision-centric models to match the performance of models with significantly larger representations, while being up to 14x faster and requiring 54x less memory. Our findings show that GQR effectively pushes the Pareto frontier for performance and efficiency in multimodal retrieval. We release our code at https://github.com/IBM/test-time-hybrid-retrieval

Paperid: 430, https://arxiv.org/pdf/2510.04898.pdf GitHub

Abstract:
Built upon language and vision foundation models with strong generalization ability and trained on large-scale robotic data, Vision-Language-Action (VLA) models have recently emerged as a promising approach to learning generalist robotic policies. However, a key drawback of existing VLAs is their extremely high inference costs. In this paper, we propose HyperVLA to address this problem. Unlike existing monolithic VLAs that activate the whole model during both training and inference, HyperVLA uses a novel hypernetwork (HN)-based architecture that activates only a small task-specific policy during inference, while still retaining the high model capacity needed to accommodate diverse multi-task behaviors during training. Successfully training an HN-based VLA is nontrivial so HyperVLA contains several key algorithm design features that improve its performance, including properly utilizing the prior knowledge from existing vision foundation models, HN normalization, and an action generation strategy. Compared to monolithic VLAs, HyperVLA achieves a similar or even higher success rate for both zero-shot generalization and few-shot adaptation, while significantly reducing inference costs. Compared to OpenVLA, a state-of-the-art VLA model, HyperVLA reduces the number of activated parameters at test time by $90\times$, and accelerates inference speed by $120\times$. Code is publicly available at https://github.com/MasterXiong/HyperVLA

Paperid: 431, https://arxiv.org/pdf/2510.04564.pdf GitHub

Abstract:
Conventional representation learning methods learn a universal representation that primarily captures dominant semantics, which may not always align with customized downstream tasks. For instance, in animal habitat analysis, researchers prioritize scene-related features, whereas universal embeddings emphasize categorical semantics, leading to suboptimal results. As a solution, existing approaches resort to supervised fine-tuning, which however incurs high computational and annotation costs. In this paper, we propose Conditional Representation Learning (CRL), aiming to extract representations tailored to arbitrary user-specified criteria. Specifically, we reveal that the semantics of a space are determined by its basis, thereby enabling a set of descriptive words to approximate the basis for a customized feature space. Building upon this insight, given a user-specified criterion, CRL first employs a large language model (LLM) to generate descriptive texts to construct the semantic basis, then projects the image representation into this conditional feature space leveraging a vision-language model (VLM). The conditional representation better captures semantics for the specific criterion, which could be utilized for multiple customized tasks. Extensive experiments on classification and retrieval tasks demonstrate the superiority and generality of the proposed CRL. The code is available at https://github.com/XLearning-SCU/2025-NeurIPS-CRL.

Paperid: 432, https://arxiv.org/pdf/2510.04246.pdf GitHub

Abstract:
Leveraging temporal context is crucial for success in partially observable robotic tasks. However, prior work in behavior cloning has demonstrated inconsistent performance gains when using multi-frame observations. In this paper, we introduce ContextVLA, a policy model that robustly improves robotic task performance by effectively leveraging multi-frame observations. Our approach is motivated by the key observation that Vision-Language-Action models (VLA), i.e., policy models built upon a Vision-Language Model (VLM), more effectively utilize multi-frame observations for action generation. This suggests that VLMs' inherent temporal understanding capability enables them to extract more meaningful context from multi-frame observations. However, the high dimensionality of video inputs introduces significant computational overhead, making VLA training and inference inefficient. To address this, ContextVLA compresses past observations into a single context token, allowing the policy to efficiently leverage temporal context for action generation. Our experiments show that ContextVLA consistently improves over single-frame VLAs and achieves the benefits of full multi-frame training but with reduced training and inference times.

Paperid: 433, https://arxiv.org/pdf/2510.03903.pdf GitHub

Abstract:
Large Vision-Language Models (LVLMs) have demonstrated impressive performance on vision-language reasoning tasks. However, their potential for zero-shot fine-grained image classification, a challenging task requiring precise differentiation between visually similar categories, remains underexplored. We present a novel method that transforms zero-shot fine-grained image classification into a visual question-answering framework, leveraging LVLMs' comprehensive understanding capabilities rather than relying on direct class name generation. We enhance model performance through a novel attention intervention technique. We also address a key limitation in existing datasets by developing more comprehensive and precise class description benchmarks. We validate the effectiveness of our method through extensive experimentation across multiple fine-grained image classification benchmarks. Our proposed method consistently outperforms the current state-of-the-art (SOTA) approach, demonstrating both the effectiveness of our method and the broader potential of LVLMs for zero-shot fine-grained classification tasks. Code and Datasets: https://github.com/Atabuzzaman/Fine-grained-classification

Paperid: 434, https://arxiv.org/pdf/2510.03827.pdf GitHub

Abstract:
LIBERO has emerged as a widely adopted benchmark for evaluating Vision-Language-Action (VLA) models; however, its current training and evaluation settings are problematic, often leading to inflated performance estimates and preventing fair model comparison. To address these issues, we introduce LIBERO-PRO, an extended LIBERO benchmark that systematically evaluates model performance under reasonable perturbations across four dimensions: manipulated objects, initial states, task instructions, and environments. Experimental results reveal that, although existing models achieve over 90% accuracy under the standard LIBERO evaluation, their performance collapses to 0.0% under our generalized setting. Crucially, this discrepancy exposes the models' reliance on rote memorization of action sequences and environment layouts from the training set, rather than genuine task understanding or environmental perception. For instance, models persist in executing grasping actions when the target object is replaced with irrelevant items, and their outputs remain unchanged even when given corrupted instructions or even messy tokens. These findings expose the severe flaws in current evaluation practices, and we call on the community to abandon misleading methodologies in favor of robust assessments of model generalization and comprehension. Our code is available at: https://github.com/Zxy-MLlab/LIBERO-PRO.

Paperid: 435, https://arxiv.org/pdf/2510.03363.pdf GitHub

Abstract:
Unsupervised anomaly detection (UAD) aims to identify image- and pixel-level anomalies using only normal training data, with wide applications such as industrial inspection and medical analysis, where anomalies are scarce due to privacy concerns and cold-start constraints. Existing methods, whether reconstruction-based (restoring normal counterparts) or embedding-based (pretrained representations), fundamentally conduct image- or feature-level matching to generate anomaly maps. Nonetheless, matching noise has been largely overlooked, limiting their detection ability. Beyond earlier focus on unimodal RGB-based UAD, recent advances expand to multimodal scenarios, e.g., RGB-3D and RGB-Text, enabled by point cloud sensing and vision-language models. Despite shared challenges, these lines remain largely isolated, hindering a comprehensive understanding and knowledge transfer. In this paper, we advocate unified UAD for both unimodal and multimodal settings in the matching perspective. Under this insight, we present Unified Cost Filtering (UCF), a generic post-hoc refinement framework for refining anomaly cost volume of any UAD model. The cost volume is constructed by matching a test sample against normal samples from the same or different modalities, followed by a learnable filtering module with multi-layer attention guidance from the test sample, mitigating matching noise and highlighting subtle anomalies. Comprehensive experiments on 22 diverse benchmarks demonstrate the efficacy of UCF in enhancing a variety of UAD methods, consistently achieving new state-of-the-art results in both unimodal (RGB) and multimodal (RGB-3D, RGB-Text) UAD scenarios. Code and models will be released at https://github.com/ZHE-SAPI/CostFilter-AD.

Paperid: 436, https://arxiv.org/pdf/2510.03142.pdf GitHub

Abstract:
Visual navigation policy is widely regarded as a promising direction, as it mimics humans by using egocentric visual observations for navigation. However, optical information of visual observations is difficult to be explicitly modeled like LiDAR point clouds or depth maps, which subsequently requires intelligent models and large-scale data. To this end, we propose to leverage the intelligence of the Vision-Language-Action (VLA) model to learn diverse navigation capabilities from synthetic expert data in a teacher-student manner. Specifically, we implement the VLA model, MM-Nav, as a multi-view VLA (with 360 observations) based on pretrained large language models and visual foundation models. For large-scale navigation data, we collect expert data from three reinforcement learning (RL) experts trained with privileged depth information in three challenging tailor-made environments for different navigation capabilities: reaching, squeezing, and avoiding. We iteratively train our VLA model using data collected online from RL experts, where the training ratio is dynamically balanced based on performance on individual capabilities. Through extensive experiments in synthetic environments, we demonstrate that our model achieves strong generalization capability. Moreover, we find that our student VLA model outperforms the RL teachers, demonstrating the synergistic effect of integrating multiple capabilities. Extensive real-world experiments further confirm the effectiveness of our method.

Paperid: 437, https://arxiv.org/pdf/2510.02790.pdf GitHub

Abstract:
Large vision-language models (LVLMs) have shown remarkable performance in visual-language understanding for downstream multimodal tasks. While their capabilities are improving, problems emerge simultaneously. Among those problems, the hallucinations have attracted much attention, which stands for the phenomenon where LVLMs generate contradictory content to their input visual and text contents. Many approaches have been proposed to deal with this issue, such as contrastive decoding and attention manipulation. However, contrastive decoding methods struggle in constructing appropriate contrastive samples, and attention manipulation methods are highly sensitive, lacking stability. In this work, we propose image head Masked Contrastive Decoding (MaskCD). Our approach utilizes the "image heads" in LVLMs, masking them to construct contrastive samples for contrastive decoding. We evaluated MaskCD on LLaVA-1.5-7b and Qwen-VL-7b, using various benchmarks such as CHAIR, POPE, AMBER and MME. The results demonstrate that MaskCD effectively alleviates the phenomenon of hallucinations and retains the general capabilities of LVLMs. Corresponding resources could be found at: https://github.com/Deng-Jingyuan/MaskCD .

Paperid: 438, https://arxiv.org/pdf/2510.02778.pdf GitHub

Abstract:
Understanding long-form videos remains a significant challenge for vision--language models (VLMs) due to their extensive temporal length and high information density. Most current multimodal large language models (MLLMs) rely on uniform sampling, which often overlooks critical moments, leading to incorrect responses to queries. In parallel, many keyframe selection approaches impose rigid temporal spacing: once a frame is chosen, an exclusion window suppresses adjacent timestamps to reduce redundancy. While effective at limiting overlap, this strategy frequently misses short, fine-grained cues near important events. Other methods instead emphasize visual diversity but neglect query relevance. We propose AdaRD-Key, a training-free keyframe sampling module for query-driven long-form video understanding. AdaRD-Key maximizes a unified Relevance--Diversity Max-Volume (RD-MV) objective, combining a query-conditioned relevance score with a log-determinant diversity component to yield informative yet non-redundant frames. To handle broad queries with weak alignment to the video, AdaRD-Key employs a lightweight relevance-aware gating mechanism; when the relevance distribution indicates weak alignment, the method seamlessly shifts into a diversity-only mode, enhancing coverage without additional supervision. Our pipeline is training-free, computationally efficient (running in real time on a single GPU), and compatible with existing VLMs in a plug-and-play manner. Extensive experiments on LongVideoBench and Video-MME demonstrate state-of-the-art performance, particularly on long-form videos. Code available at https://github.com/Xian867/AdaRD-Key.

Paperid: 439, https://arxiv.org/pdf/2510.02543.pdf GitHub

Abstract:
We investigate OCR-augmented generation with Vision Language Models (VLMs), exploring tasks in Korean and English toward multilingualism. To support research in this domain, we train and release KLOCR, a strong bilingual OCR baseline trained on 100M instances to augment VLMs with OCR ability. To complement existing VQA benchmarks, we curate KOCRBench for Korean VQA, and analyze different prompting methods. Extensive experiments show that OCR-extracted text significantly boosts performance across open source and commercial models. Our work offers new insights into OCR-augmented generation for bilingual VQA. Model, code, and data are available at https://github.com/JHLee0513/KLOCR.

Paperid: 440, https://arxiv.org/pdf/2510.02292.pdf GitHub

Abstract:
We introduce VLM-Lens, a toolkit designed to enable systematic benchmarking, analysis, and interpretation of vision-language models (VLMs) by supporting the extraction of intermediate outputs from any layer during the forward pass of open-source VLMs. VLM-Lens provides a unified, YAML-configurable interface that abstracts away model-specific complexities and supports user-friendly operation across diverse VLMs. It currently supports 16 state-of-the-art base VLMs and their over 30 variants, and is extensible to accommodate new models without changing the core logic. The toolkit integrates easily with various interpretability and analysis methods. We demonstrate its usage with two simple analytical experiments, revealing systematic differences in the hidden representations of VLMs across layers and target concepts. VLM-Lens is released as an open-sourced project to accelerate community efforts in understanding and improving VLMs.

Paperid: 441, https://arxiv.org/pdf/2510.02270.pdf GitHub

Abstract:
Unsupervised adaptation of CLIP-based vision-language models (VLMs) for fine-grained image classification requires sensitivity to microscopic local cues. While CLIP exhibits strong zero-shot transfer, its reliance on coarse global features restricts its performance on fine-grained classification tasks. Prior efforts inject fine-grained knowledge by aligning large language model (LLM) descriptions with the CLIP $\texttt{[CLS]}$ token; however, this approach overlooks spatial precision. We propose $\textbf{microCLIP}$, a self-training framework that jointly refines CLIP's visual and textual representations using fine-grained cues. At its core is Saliency-Oriented Attention Pooling (SOAP) within a lightweight TokenFusion module, which builds a saliency-guided $\texttt{[FG]}$ token from patch embeddings and fuses it with the global $\texttt{[CLS]}$ token for coarse-fine alignment. To stabilize adaptation, we introduce a two-headed LLM-derived classifier: a frozen classifier that, via multi-view alignment, provides a stable text-based prior for pseudo-labeling, and a learnable classifier initialized from LLM descriptions and fine-tuned with TokenFusion. We further develop Dynamic Knowledge Aggregation, which convexly combines fixed LLM/CLIP priors with TokenFusion's evolving logits to iteratively refine pseudo-labels. Together, these components uncover latent fine-grained signals in CLIP, yielding a consistent $2.90\%$ average accuracy gain across 13 fine-grained benchmarks while requiring only light adaptation. Our code is available at https://github.com/sathiiii/microCLIP.

Paperid: 442, https://arxiv.org/pdf/2510.02186.pdf GitHub

Abstract:
Recent attempts to transfer features from 2D Vision-Language Models (VLMs) to 3D semantic segmentation expose a persistent trade-off. Directly projecting 2D features into 3D yields noisy and fragmented predictions, whereas enforcing geometric coherence necessitates costly training pipelines and large-scale annotated 3D data. We argue that this limitation stems from the dominant segmentation-and-matching paradigm, which fails to reconcile 2D semantics with 3D geometric structure. The geometric cues are not eliminated during the 2D-to-3D transfer but remain latent within the noisy and view-aggregated features. To exploit this property, we propose GeoPurify that applies a small Student Affinity Network to purify 2D VLM-generated 3D point features using geometric priors distilled from a 3D self-supervised teacher model. During inference, we devise a Geometry-Guided Pooling module to further denoise the point cloud and ensure the semantic and structural consistency. Benefiting from latent geometric information and the learned affinity network, GeoPurify effectively mitigates the trade-off and achieves superior data efficiency. Extensive experiments on major 3D benchmarks demonstrate that GeoPurify achieves or surpasses state-of-the-art performance while utilizing only about 1.5% of the training data. Our codes and checkpoints are available at [https://github.com/tj12323/GeoPurify](https://github.com/tj12323/GeoPurify).

Paperid: 443, https://arxiv.org/pdf/2510.01718.pdf GitHub

Abstract:
Attention is a core operation in large language models (LLMs) and vision-language models (VLMs). We present BD Attention (BDA), the first lossless algorithmic reformulation of attention. BDA is enabled by a simple matrix identity from Basis Decomposition (BD), which restructures multi-head projections into a compact form while preserving exact outputs. Unlike I/O-aware system optimizations such as FlashAttention, BDA provides a mathematically guaranteed acceleration that is architecture-agnostic. On DeepSeek-V2-Lite (16B, FP16), BDA requires only 4s of offline preparation with no retraining required and, on modern GPUs, achieves 32% faster key/value projections and 25% smaller weights, while increasing end-to-end perplexity (PPL) by just 0.02% (FP16) or 0.0004% (FP32), a negligible effect on model performance. These results position BDA as the first theoretically exact method for lossless attention acceleration that is complementary to existing engineering-level optimizations. Our code is available at https://github.com/abcbdf/basis-decomposition-official.

Paperid: 444, https://arxiv.org/pdf/2510.01642.pdf GitHub

Abstract:
Recent advances in robotic manipulation have integrated low-level robotic control into Vision-Language Models (VLMs), extending them into Vision-Language-Action (VLA) models. Although state-of-the-art VLAs achieve strong performance in downstream robotic applications, supported by large-scale crowd-sourced robot training data, they still inevitably encounter failures during execution. Enabling robots to reason about and recover from unpredictable and abrupt failures remains a critical challenge. Existing robotic manipulation datasets, collected in either simulation or the real world, primarily provide only ground-truth trajectories, leaving robots unable to recover once failures occur. Moreover, the few datasets that address failure detection typically offer only textual explanations, which are difficult to utilize directly in VLA models. To address this gap, we introduce FailSafe, a novel failure generation and recovery system that automatically produces diverse failure cases paired with executable recovery actions. FailSafe can be seamlessly applied to any manipulation task in any simulator, enabling scalable creation of failure-action data. To demonstrate its effectiveness, we fine-tune LLaVa-OneVision-7B (LLaVa-OV-7B) to build FailSafe-VLM. Experimental results show that FailSafe-VLM successfully helps robotic arm detect and recover from potential failures, improving the performance of three state-of-the-art VLA models pi0-FAST, OpenVLA, OpenVLA-OFT) by up to 22.6% on average across several tasks in Maniskill. Furthermore, FailSafe-VLM could generalize across different spatial configurations, camera viewpoints, and robotic embodiments. We plan to release the FailSafe code to the community.

Paperid: 445, https://arxiv.org/pdf/2510.01623.pdf GitHub GitHub

Abstract:
Vision-Language-Action (VLA) models aim to unify perception, language understanding, and action generation, offering strong cross-task and cross-scene generalization with broad impact on embodied AI. However, current VLA models often lack explicit step-by-step reasoning, instead emitting final actions without considering affordance constraints or geometric relations. Their post-training pipelines also rarely reinforce reasoning quality, relying primarily on supervised fine-tuning with weak reward design. To address these challenges, we present VLA-R1, a reasoning-enhanced VLA that integrates Reinforcement Learning from Verifiable Rewards (RLVR) with Group Relative Policy Optimization (GRPO) to systematically optimize both reasoning and execution. Specifically, we design an RLVR-based post-training strategy with verifiable rewards for region alignment, trajectory consistency, and output formatting, thereby strengthening reasoning robustness and execution accuracy. Moreover, we develop VLA-CoT-13K, a high-quality dataset that provides chain-of-thought supervision explicitly aligned with affordance and trajectory annotations. Furthermore, extensive evaluations on in-domain, out-of-domain, simulation, and real-robot platforms demonstrate that VLA-R1 achieves superior generalization and real-world performance compared to prior VLA methods. We plan to release the model, code, and dataset following the publication of this work. Code: https://github.com/GigaAI-research/VLA-R1. Website: https://gigaai-research.github.io/VLA-R1.

Paperid: 446, https://arxiv.org/pdf/2510.01454.pdf GitHub GitHub

Abstract:
Data-efficient learning aims to eliminate redundancy in large training datasets by training models on smaller subsets of the most informative examples. While data selection has been extensively explored for vision models and large language models (LLMs), it remains underexplored for Large Vision-Language Models (LVLMs). Notably, none of existing methods can outperform random selection at different subset sizes. In this work, we propose the first principled method for data-efficient instruction tuning of LVLMs. We prove that examples with similar cross-modal attention matrices during instruction tuning have similar gradients. Thus, they influence model parameters in a similar manner and convey the same information to the model during training. Building on this insight, we propose XMAS, which clusters examples based on the trajectories of the top singular values of their attention matrices obtained from fine-tuning a small proxy LVLM. By sampling a balanced subset from these clusters, XMAS effectively removes redundancy in large-scale LVLM training data. Extensive experiments show that XMAS can discard 50% of the LLaVA-665k dataset and 85% of the Vision-Flan dataset while fully preserving performance of LLaVA-1.5-7B on 10 downstream benchmarks and speeding up its training by 1.2x. This is 30% more data reduction compared to the best baseline for LLaVA-665k. The project's website can be found at https://bigml-cs-ucla.github.io/XMAS-project-page/.

Paperid: 447, https://arxiv.org/pdf/2510.01304.pdf GitHub

Abstract:
Although current large Vision-Language Models (VLMs) have advanced in multimodal understanding and reasoning, their fundamental perceptual and reasoning abilities remain limited. Specifically, even on simple jigsaw tasks, existing VLMs perform near randomly, revealing deficiencies in core perception and reasoning capabilities. While high-quality vision-language data can enhance these capabilities, its scarcity and limited scalability impose significant constraints. To address this, we propose AGILE, an Agentic jiGsaw Interaction Learning for Enhancing visual perception and reasoning in VLMs. AGILE formulates jigsaw solving as an interactive process, enabling the model to progressively engage with the environment. At each step, the model generates executable code to perform an action based on the current state, while the environment provides fine-grained visual feedback to guide task completion. Through this iterative cycle of observation and interaction, the model incrementally improves its perceptual and reasoning capabilities via exploration and feedback. Experimental results show that AGILE not only substantially boosts performance on jigsaw tasks of varying complexity (e.g., increasing accuracy from 9.5% to 82.8% under the 2 $\times$ 2 setting) but also demonstrates strong generalization across 9 general vision tasks, achieving an average improvement of 3.1%. These results indicate notable enhancements in both perceptual and reasoning abilities. This work opens a new avenue for advancing reasoning and generalization in multimodal models and provides an efficient, scalable solution to the scarcity of multimodal reinforcement learning data. The code and datasets is available at https://github.com/yuzeng0-0/AGILE .

Paperid: 448, https://arxiv.org/pdf/2510.01174.pdf GitHub GitHub

Abstract:
While recent generative models advance pixel-space video synthesis, they remain limited in producing professional educational videos, which demand disciplinary knowledge, precise visual structures, and coherent transitions, limiting their applicability in educational scenarios. Intuitively, such requirements are better addressed through the manipulation of a renderable environment, which can be explicitly controlled via logical commands (e.g., code). In this work, we propose Code2Video, a code-centric agent framework for generating educational videos via executable Python code. The framework comprises three collaborative agents: (i) Planner, which structures lecture content into temporally coherent flows and prepares corresponding visual assets; (ii) Coder, which converts structured instructions into executable Python codes while incorporating scope-guided auto-fix to enhance efficiency; and (iii) Critic, which leverages vision-language models (VLM) with visual anchor prompts to refine spatial layout and ensure clarity. To support systematic evaluation, we build MMMC, a benchmark of professionally produced, discipline-specific educational videos. We evaluate MMMC across diverse dimensions, including VLM-as-a-Judge aesthetic scores, code efficiency, and particularly, TeachQuiz, a novel end-to-end metric that quantifies how well a VLM, after unlearning, can recover knowledge by watching the generated videos. Our results demonstrate the potential of Code2Video as a scalable, interpretable, and controllable approach, achieving 40% improvement over direct code generation and producing videos comparable to human-crafted tutorials. The code and datasets are available at https://github.com/showlab/Code2Video.

Paperid: 449, https://arxiv.org/pdf/2510.01068.pdf GitHub

Abstract:
Diffusion-based models for robotic control, including vision-language-action (VLA) and vision-action (VA) policies, have demonstrated significant capabilities. Yet their advancement is constrained by the high cost of acquiring large-scale interaction datasets. This work introduces an alternative paradigm for enhancing policy performance without additional model training. Perhaps surprisingly, we demonstrate that the composed policies can exceed the performance of either parent policy. Our contribution is threefold. First, we establish a theoretical foundation showing that the convex composition of distributional scores from multiple diffusion models can yield a superior one-step functional objective compared to any individual score. A Grönwall-type bound is then used to show that this single-step improvement propagates through entire generation trajectories, leading to systemic performance gains. Second, motivated by these results, we propose General Policy Composition (GPC), a training-free method that enhances performance by combining the distributional scores of multiple pre-trained policies via a convex combination and test-time search. GPC is versatile, allowing for the plug-and-play composition of heterogeneous policies, including VA and VLA models, as well as those based on diffusion or flow-matching, irrespective of their input visual modalities. Third, we provide extensive empirical validation. Experiments on Robomimic, PushT, and RoboTwin benchmarks, alongside real-world robotic evaluations, confirm that GPC consistently improves performance and adaptability across a diverse set of tasks. Further analysis of alternative composition operators and weighting strategies offers insights into the mechanisms underlying the success of GPC. These results establish GPC as a simple yet effective method for improving control performance by leveraging existing policies.

Paperid: 450, https://arxiv.org/pdf/2510.00695.pdf GitHub

Abstract:
Inherently, robotic manipulation tasks are history-dependent: leveraging past context could be beneficial. However, most existing Vision-Language-Action models (VLAs) have been designed without considering this aspect, i.e., they rely solely on the current observation, ignoring preceding context. In this paper, we propose HAMLET, a scalable framework to adapt VLAs to attend to the historical context during action prediction. Specifically, we introduce moment tokens that compactly encode perceptual information at each timestep. Their representations are initialized with time-contrastive learning, allowing them to better capture temporally distinctive aspects. Next, we employ a lightweight memory module that integrates the moment tokens across past timesteps into memory features, which are then leveraged for action prediction. Through empirical evaluation, we show that HAMLET successfully transforms a state-of-the-art VLA into a history-aware policy, especially demonstrating significant improvements on long-horizon tasks that require historical context. In particular, on top of GR00T N1.5, HAMLET achieves an average success rate of 76.4% on history-dependent real-world tasks, surpassing the baseline performance by 47.2%. Furthermore, HAMLET pushes prior art performance from 64.1% to 66.4% on RoboCasa Kitchen (100-demo setup) and from 95.6% to 97.7% on LIBERO, highlighting its effectiveness even under generic robot-manipulation benchmarks.

Paperid: 451, https://arxiv.org/pdf/2510.00647.pdf GitHub

Abstract:
The alt-text generation task produces concise, context-relevant descriptions of images, enabling blind and low-vision users to access online images. Despite the capabilities of large vision-language models, alt-text generation performance remains limited due to noisy user annotations, inconsistent standards, and MLLMs' insensitivity to contextual information. Previous efforts to fine-tune MLLMs using supervised fine-tuning (SFT) have struggled, as SFT relies on accurate target annotations, which are often flawed in user-generated alt-text. To address this, we propose Multi-faceted Cross-modal Direct Preference Optimization (MCM-DPO), which improves alt-text generation by learning to identify better options in preference pairs without requiring precise annotations. MCM-DPO optimizes preferences across single, paired, and multi-preference dimensions, covering textual, visual, and cross-modal factors. In light of the scarcity of high-quality annotated and preference-labeled datasets for alt-text, we constructed two large-scale, high-quality datasets named TAlt and PAlt, sourced from Twitter and Pinterest. These datasets include 202k annotated alt-text samples and 18k preference pairs that cover diverse preference dimensions, aiming to support further research in this domain. Experimental results show that our proposed MCM-DPO method consistently outperforms both DPO and SFT, establishing a new state of the art in alt-text generation. We release the code and data here: https://github.com/LVUGAI/MCM-DPO

Paperid: 452, https://arxiv.org/pdf/2510.00483.pdf GitHub

Abstract:
We introduce \textsc{MathSticks}, a benchmark for Visual Symbolic Compositional Reasoning (VSCR), which unifies visual perception, symbolic manipulation, and arithmetic consistency. Each task presents an incorrect matchstick equation that must be corrected by moving one or two sticks under strict conservation rules. The benchmark includes both text-guided and purely visual settings, systematically covering digit scale, move complexity, solution multiplicity, and operator variation, with 1.4M generated instances and a curated test set. Evaluations of 14 vision--language models reveal substantial limitations: closed-source models succeed only on simple cases, open-source models fail in the visual regime, while humans exceed 90\% accuracy. These findings establish \textsc{MathSticks} as a rigorous testbed for advancing compositional reasoning across vision and symbols. Our code and dataset are publicly available at https://github.com/Yuheng2000/MathSticks.

Paperid: 453, https://arxiv.org/pdf/2510.00458.pdf GitHub

Abstract:
Vision-language object detectors (VLODs) such as YOLO-World and Grounding DINO achieve impressive zero-shot recognition by aligning region proposals with text representations. However, their performance often degrades under domain shift. We introduce VLOD-TTA, a test-time adaptation (TTA) framework for VLODs that leverages dense proposal overlap and image-conditioned prompt scores. First, an IoU-weighted entropy objective is proposed that concentrates adaptation on spatially coherent proposal clusters and reduces confirmation bias from isolated boxes. Second, image-conditioned prompt selection is introduced, which ranks prompts by image-level compatibility and fuses the most informative prompts with the detector logits. Our benchmarking across diverse distribution shifts -- including stylized domains, driving scenes, low-light conditions, and common corruptions -- shows the effectiveness of our method on two state-of-the-art VLODs, YOLO-World and Grounding DINO, with consistent improvements over the zero-shot and TTA baselines. Code : https://github.com/imatif17/VLOD-TTA

Paperid: 454, https://arxiv.org/pdf/2510.00406.pdf GitHub

Abstract:
Vision-Language-Action (VLA) models enable embodied decision-making but rely heavily on imitation learning, leading to compounding errors and poor robustness under distribution shift. Reinforcement learning (RL) can mitigate these issues yet typically demands costly real-world interactions or suffers from sim-to-real gaps. We introduce VLA-RFT, a reinforcement fine-tuning framework that leverages a data-driven world model as a controllable simulator. Trained from real interaction data, the simulator predicts future visual observations conditioned on actions, allowing policy rollouts with dense, trajectory-level rewards derived from goal-achieving references. This design delivers an efficient and action-aligned learning signal, drastically lowering sample requirements. With fewer than 400 fine-tuning steps, VLA-RFT surpasses strong supervised baselines and achieves greater efficiency than simulator-based RL. Moreover, it exhibits strong robustness under perturbed conditions, sustaining stable task execution. Our results establish world-model-based RFT as a practical post-training paradigm to enhance the generalization and robustness of VLA models. For more details, please refer to https://vla-rft.github.io/.

Paperid: 455, https://arxiv.org/pdf/2510.00161.pdf GitHub

Abstract:
Procedural activity assistants potentially support humans in a variety of settings, from our daily lives, e.g., cooking or assembling flat-pack furniture, to professional situations, e.g., manufacturing or biological experiments. Despite its potential use cases, the system development tailored for such an assistant is still underexplored. In this paper, we propose a novel framework, called TAMA, a Tool-Augmented Multimodal Agent, for procedural activity understanding. TAMA enables interleaved multimodal reasoning by making use of multimedia-returning tools in a training-free setting. Our experimental result on the multimodal procedural QA dataset, ProMQA-Assembly, shows that our approach can improve the performance of vision-language models, especially GPT-5 and MiMo-VL. Furthermore, our ablation studies provide empirical support for the effectiveness of two features that characterize our framework, multimedia-returning tools and agentic flexible tool selection. We believe our proposed framework and experimental results facilitate the thinking with images paradigm for video and multimodal tasks, let alone the development of procedural activity assistants.

Paperid: 456, https://arxiv.org/pdf/2509.26462.pdf GitHub

Abstract:
CLIP has revolutionized zero-shot learning by enabling task generalization without fine-tuning. While prompting techniques like CoOp and CoCoOp enhance CLIP's adaptability, their effectiveness in Federated Learning (FL) remains an open challenge. Existing federated prompt learning approaches, such as FedCoOp and FedTPG, improve performance but face generalization issues, high communication costs, and reliance on a central server, limiting scalability and privacy. We propose Zero-shot Decentralized Federated Learning (ZeroDFL), a fully decentralized framework that enables zero-shot adaptation across distributed clients without a central coordinator. ZeroDFL employs an iterative prompt-sharing mechanism, allowing clients to optimize and exchange textual prompts to enhance generalization while drastically reducing communication overhead. We validate ZeroDFL on nine diverse image classification datasets, demonstrating that it consistently outperforms--or remains on par with--state-of-the-art federated prompt learning methods. More importantly, ZeroDFL achieves this performance in a fully decentralized setting while reducing communication overhead by 118x compared to FedTPG. These results highlight that our approach not only enhances generalization in federated zero-shot learning but also improves scalability, efficiency, and privacy preservation--paving the way for decentralized adaptation of large vision-language models in real-world applications.

Paperid: 457, https://arxiv.org/pdf/2509.26039.pdf GitHub

Abstract:
We extend HAMMER, a state-of-the-art model for multimodal manipulation detection, to handle global scene inconsistencies such as foreground-background (FG-BG) mismatch. While HAMMER achieves strong performance on the DGM4 dataset, it consistently fails when the main subject is contextually misplaced into an implausible background. We diagnose this limitation as a combination of label-space bias, local attention focus, and spurious text-foreground alignment. To remedy this without retraining, we propose a lightweight segmentation-guided scoring (SGS) pipeline. SGS uses person/face segmentation masks to separate foreground and background regions, extracts embeddings with a joint vision-language model, and computes region-aware coherence scores. These scores are fused with HAMMER's original prediction to improve binary detection, grounding, and token-level explanations. SGS is inference-only, incurs negligible computational overhead, and significantly enhances robustness to global manipulations. This work demonstrates the importance of region-aware reasoning in multimodal disinformation detection. We release scripts for segmentation and scoring at https://github.com/Gaganx0/HAMMER-sgs

Paperid: 458, https://arxiv.org/pdf/2509.25851.pdf GitHub

Abstract:
Multimodal symbolic logical reasoning, which aims to deduce new facts from multimodal input via formal logic, is critical in high-stakes applications such as autonomous driving and medical diagnosis, as its rigorous, deterministic reasoning helps prevent serious consequences. To evaluate such capabilities of current state-of-the-art vision language models (VLMs), we introduce the first benchmark MuSLR for multimodal symbolic logical reasoning grounded in formal logical rules. MuSLR comprises 1,093 instances across 7 domains, including 35 atomic symbolic logic and 976 logical combinations, with reasoning depths ranging from 2 to 9. We evaluate 7 state-of-the-art VLMs on MuSLR and find that they all struggle with multimodal symbolic reasoning, with the best model, GPT-4.1, achieving only 46.8%. Thus, we propose LogiCAM, a modular framework that applies formal logical rules to multimodal inputs, boosting GPT-4.1's Chain-of-Thought performance by 14.13%, and delivering even larger gains on complex logics such as first-order logic. We also conduct a comprehensive error analysis, showing that around 70% of failures stem from logical misalignment between modalities, offering key insights to guide future improvements. All data and code are publicly available at https://llm-symbol.github.io/MuSLR.

Paperid: 459, https://arxiv.org/pdf/2509.25848.pdf GitHub

Abstract:
Reasoning has emerged as a pivotal capability in Large Language Models (LLMs). Through Reinforcement Learning (RL), typically Group Relative Policy Optimization (GRPO), these models are able to solve complex tasks such as mathematics and code generation. Building on these advances, recent research has sought to extend reasoning to Vision-Language Models (VLMs), yielding promising results across diverse visual tasks. Despite this progress, our study uncovers the dual nature of multimodal reasoning: while it substantially enhances logical inference and facilitates performance on challenging problems, it may gradually impair perceptual grounding, leading to recognition failures on otherwise basic visual questions. Through further analysis, we attribute this phenomenon to visual forgetting, wherein prolonged reasoning causes the model to increasingly disregard visual input. To address this, we propose Vision-Anchored Policy Optimization (VAPO), a simple yet effective method that explicitly steers the reasoning process toward visually grounded trajectories. Our result model, VAPO-Thinker-7B, significantly strengthens the model's reliance on visual information and achieves new state-of-the-art results on a wide range of established benchmarks. Project page: https://xytian1008.github.io/VAPO/

Paperid: 460, https://arxiv.org/pdf/2509.25791.pdf GitHub

Abstract:
Electrocardiogram (ECG) is a widely used tool for assessing cardiac function due to its low cost and accessibility. Emergent research shows that ECGs can help make predictions on key outcomes traditionally derived from more complex modalities such as echocardiograms (ECHO), enabling the use of ECGs as a more accessible method to predict broader measurements of cardiac function. ECHO, in particular, are of great importance because they require considerable hospital resources while playing a key role in clinical cardiac assessment. To aid this use case, we introduce EchoingECG, a probabilistic student-teacher model that leverages uncertainty-aware ECG embeddings and ECHO supervision to improve ECG-based cardiac function prediction. Our approach integrates Probabilistic Cross-Modal Embeddings (PCME++), a probabilistic contrastive framework, with ECHO-CLIP, a vision-language pre-trained model trained on ECHO-text pairs, to distill ECHO knowledge into ECG representations. Through experiments and external validation, we showed that EchoingECG outperforms state-of-the-art foundation ECG models in zero-shot, few-shot, and fine-tune settings for ECHO predictions based on ECG. We also highlighted that variance estimation (enabled through our method) enhanced our understanding of model performance by identifying underlying regions of uncertainty within ECGs. The code is available: https://github.com/mcintoshML/EchoingECG.

Paperid: 461, https://arxiv.org/pdf/2509.25654.pdf GitHub

Abstract:
Automated textual description of remote sensing images is crucial for unlocking their full potential in diverse applications, from environmental monitoring to urban planning and disaster management. However, existing studies in remote sensing image captioning primarily focus on the image level, lacking object-level fine-grained interpretation, which prevents the full utilization and transformation of the rich semantic and structural information contained in remote sensing images. To address this limitation, we propose Geo-DLC, a novel task of object-level fine-grained image captioning for remote sensing. To support this task, we construct DE-Dataset, a large-scale dataset contains 25 categories and 261,806 annotated instances with detailed descriptions of object attributes, relationships, and contexts. Furthermore, we introduce DE-Benchmark, a LLM-assisted question-answering based evaluation suite designed to systematically measure model capabilities on the Geo-DLC task. We also present DescribeEarth, a Multi-modal Large Language Model (MLLM) architecture explicitly designed for Geo-DLC, which integrates a scale-adaptive focal strategy and a domain-guided fusion module leveraging remote sensing vision-language model features to encode high-resolution details and remote sensing category priors while maintaining global context. Our DescribeEarth model consistently outperforms state-of-the-art general MLLMs on DE-Benchmark, demonstrating superior factual accuracy, descriptive richness, and grammatical soundness, particularly in capturing intrinsic object features and surrounding environmental attributes across simple, complex, and even out-of-distribution remote sensing scenarios. All data, code and weights are released at https://github.com/earth-insights/DescribeEarth.

Paperid: 462, https://arxiv.org/pdf/2509.25541.pdf GitHub

Abstract:
Although reinforcement learning (RL) can effectively enhance the reasoning capabilities of vision-language models (VLMs), current methods remain heavily dependent on labor-intensive datasets that require extensive manual construction and verification, leading to extremely high training costs and consequently constraining the practical deployment of VLMs. To address this challenge, we propose Vision-Zero, a domain-agnostic framework enabling VLM self-improvement through competitive visual games generated from arbitrary image pairs. Specifically, Vision-Zero encompasses three main attributes: (1) Strategic Self-Play Framework: Vision-Zero trains VLMs in "Who Is the Spy"-style games, where the models engage in strategic reasoning and actions across multiple roles. Through interactive gameplay, models autonomously generate their training data without human annotation. (2) Gameplay from Arbitrary Images: Unlike existing gamified frameworks, Vision-Zero can generate games from arbitrary images, thereby enhancing the model's reasoning ability across diverse domains and showing strong generalization to different tasks. We demonstrate this versatility using three distinct types of image datasets: CLEVR-based synthetic scenes, charts, and real-world images. (3) Sustainable Performance Gain: We introduce Iterative Self-Play Policy Optimization (Iterative-SPO), a novel training algorithm that alternates between Self-Play and reinforcement learning with verifiable rewards (RLVR), mitigating the performance plateau often seen in self-play-only training and achieving sustained long-term improvements. Despite using label-free data, Vision-Zero achieves state-of-the-art performance on reasoning, chart question answering, and vision-centric understanding tasks, surpassing other annotation-based methods. Models and code has been released at https://github.com/wangqinsi1/Vision-Zero.

Paperid: 463, https://arxiv.org/pdf/2509.25390.pdf GitHub

Abstract:
We present SpinBench, a cognitively grounded diagnostic benchmark for evaluating spatial reasoning in vision language models (VLMs). SpinBench is designed around the core challenge of spatial reasoning: perspective taking, the ability to reason about how scenes and object relations change under viewpoint transformation. Since perspective taking requires multiple cognitive capabilities, such as recognizing objects across views, relative positions grounding, and mentally simulating transformations, SpinBench introduces a set of fine-grained diagnostic categories. Our categories target translation, rotation, object relative pose, and viewpoint change, and are progressively structured so that single-object simpler tasks scaffold toward the most demanding multi-object perspective-taking setting. We evaluate 37 state-of-the-art VLMs, both proprietary and open source. Results reveal systematic weaknesses: strong egocentric bias, poor rotational understanding, and inconsistencies under symmetrical and syntactic reformulations. Scaling analysis shows both smooth improvements and emergent capabilities. While human subjects achieve high accuracy (91.2\%), task difficulty as measured by human response time shows strong correlation with VLM accuracy, indicating that SpinBench captures spatial reasoning challenges shared across humans and VLMs. We believe SpinBench provides critical insights into spatial reasoning in VLMs and highlights key gaps in their ability to reason about physical space. Our website can be found at https://spinbench25.github.io/.

Paperid: 464, https://arxiv.org/pdf/2509.25160.pdf GitHub GitHub

Abstract:
Vision language models (VLMs) achieve unified modeling of images and text, enabling them to accomplish complex real-world tasks through perception, planning, and reasoning. Among these tasks, reasoning is particularly representative, with mathematical reasoning serving as a prominent example. It highlights the high-level capability of VLMs to comprehend mathematical information in images and to perform sophisticated reasoning. Recently, numerous visual mathematical reasoning benchmarks have been proposed, but they are often restricted to geometry, lack coverage of math word problems, and rarely assess reasoning across multiple images. To address these gaps, we introduce GSM8K-V, a purely visual multi-image mathematical reasoning benchmark. GSM8K-V is built by systematically mapping each sample from the widely used text-based GSM8K into visual form. Through a carefully designed automated image-generation pipeline combined with meticulous human annotation, we curate 1,319 high-quality samples. We evaluate a wide range of open-source and closed-source models on GSM8K-V. Results show that although existing VLMs have nearly saturated performance on text-based GSM8K, there remains substantial room for improvement on GSM8K-V. For example, the best-performing model, Gemini-2.5-Pro, achieves 95.22% accuracy on GSM8K but only 46.93% on GSM8K-V. We conduct a comprehensive analysis of GSM8K-V, examining the limitations of current models as well as potential directions for improvement. GSM8K-V offers a new perspective on visual mathematical reasoning and establishes a benchmark to guide the development of more robust and generalizable VLMs.

Paperid: 465, https://arxiv.org/pdf/2509.25151.pdf GitHub

Abstract:
Multimodal Large Language Models (MLLMs) have achieved impressive progress in vision-language alignment, yet they remain limited in visual-spatial reasoning. We first identify that this limitation arises from the attention mechanism: visual tokens are overshadowed by language tokens, preventing the model from consistently recognizing the same visual cues across frames. To address this challenge, we draw a novel connection between the self-expressiveness property in sparse subspace clustering and the attention mechanism in Transformers. Building on this insight, we propose VideoAnchor, a plug-and-play module that leverages subspace affinities to reinforce visual cues across frames without retraining, effectively anchoring attention to shared visual structures. Extensive experiments across benchmarks and backbone models show consistent performance gains -- $e.g.$, 3.2% and 4.6% improvements on VSI-Bench and Video-MME (spatial-related tasks) with InternVL2-8B and Qwen2.5VL-72B -- while qualitative analyses demonstrate more coherent subspace partitions and stronger visual grounding. Our codes will be made public available at https://github.com/feufhd/VideoAnchor.

Paperid: 466, https://arxiv.org/pdf/2509.25026.pdf GitHub

Abstract:
Recent advances in reinforcement learning (RL) have delivered strong reasoning capabilities in natural image domains, yet their potential for Earth Observation (EO) remains largely unexplored. EO tasks introduce unique challenges, spanning referred object detection, image or region captioning, change detection, grounding, and temporal analysis, that demand task aware reasoning. We propose a novel post training framework that incorporates task aware rewards to enable effective adaptation of reasoning based RL models to diverse EO tasks. This training strategy enhances reasoning capabilities for remote sensing images, stabilizes optimization, and improves robustness. Extensive experiments across multiple EO benchmarks show consistent performance gains over state of the art generic and specialized vision language models. Code and models will be released publicly at https://mustansarfiaz.github.io/GeoVLM-R1/ .

Paperid: 467, https://arxiv.org/pdf/2509.24709.pdf GitHub

Abstract:
The webpage-to-code task requires models to understand visual representations of webpages and generate corresponding code. However, existing benchmarks primarily focus on static screenshot-to-code tasks, thereby overlooking the dynamic interactions fundamental to real-world web applications. To address this limitation, this paper introduces IWR-Bench, a novel benchmark for evaluating the capabilities of Large Vision-Language Models (LVLMs) in interactive webpage reconstruction from video. IWR-Bench comprises 113 meticulously curated tasks from 100 real-world websites, with 1,001 actions and featuring diverse interaction complexities (e.g., web games), visual styles, and domains. Aligning with standard web development practices, each task includes not only user interaction videos but also all crawled static assets (e.g., images, videos). This benchmark evaluates models on two fundamental challenges: comprehensive multi-modal reasoning to infer interaction logic from video and assets, and advanced code generation to translate this logic into functional code. An agent-as-a-judge framework with a comprehensive metric system automatically assesses the functional correctness and visual fidelity of generated webpages. Extensive experiments on 28 LVLMs reveal a significant challenge: the best model achieves an overall score of only 36.35%, as functional correctness (24.39% IFS) lags significantly behind visual fidelity (64.25% VFS). These results highlight critical limitations in current models' ability to reason about temporal dynamics and synthesize event-driven logic, establishing IWR-Bench as a challenging frontier for vision-language research. The benchmark and evaluation code will be made publicly available. Code is available at https://github.com/L-O-I/IWR-Bench.

Paperid: 468, https://arxiv.org/pdf/2509.24473.pdf GitHub

Abstract:
Spatial intelligence spans a rich suite of abilities, including visualising and transforming shapes, mentally rotating objects, judging relational positions and containment, and estimating numerosity. However, it still remains a critical unresolved challenge for Multimodal Large Language Models (MLLMs).To fill this gap, we propose to treat Euclidean geometry problem-solving as a surrogate task. Specifically, we meticulously constructed a curated multimodal dataset, called Euclid30K, comprising approximately 30K plane and solid geometry problems. To enable the model to acquire and apply Euclidean principles from these geometry problems, we employed Group Relative Policy Optimization (GRPO) to finetune the Qwen2.5VL family and RoboBrain2.0 family, inspiring the models to identify shapes, count, and relate entities, and perform multi-step deductive reasoning using Euclidean principles. Our experiments demonstrate that the resulting models achieve substantial zero-shot gains across four spatial reasoning benchmarks (Super-CLEVR, Omni3DBench, VSI-Bench, and MindCube) without any task-specific adaptations. Notably, after training on the Euclid30K, the mean VSI-Bench accuracy of all evaluated models rose from 34.5% to 40.5%, improving by 5.5 percentage points. Among them, RoboBrain2.0-Euclid-7B achieves 49.6\% accuracy, surpassing the previous state-of-the-art model, Spatial-MLLM.To our knowledge, this is the first systematic study showing that geometry-centric fine-tuning can confer vision-language models with broadly transferable spatial skills. Code and Euclid30K dataset can be found in https://zgca-ai4edu.github.io/Euclids_Gift.

Paperid: 469, https://arxiv.org/pdf/2509.24387.pdf GitHub

Abstract:
Vision Language Navigation (VLN) requires agents to follow natural language instructions by grounding them in sequential visual observations over long horizons. Explicit reasoning could enhance temporal consistency and perception action alignment, but reasoning at fixed steps often leads to suboptimal performance and unnecessary computation. To address this, we propose AdaNav, an uncertainty-based adaptive reasoning framework for VLN. At its core is the Uncertainty Adaptive Reasoning Block (UAR), a lightweight plugin that dynamically triggers reasoning. We introduce Action Entropy as a policy prior for UAR and progressively refine it through a Heuristics to RL training method, enabling agents to learn difficulty aware reasoning policies under the strict data limitations of embodied tasks. Results show that with only 6K training samples, AdaNav achieves substantial gains over closed source models trained on million scale data, improving success rate by 20% on R2R val-unseen, 11.7% on RxR-CE, and 11.4% in real world scenes. The code is available at https://github.com/xinding-sys/AdaNav.

Paperid: 470, https://arxiv.org/pdf/2509.23931.pdf GitHub

Abstract:
The established redundancy in visual tokens within large vision-language models allows pruning to effectively reduce their substantial computational demands. Previous methods typically employ heuristic layer-specific pruning strategies where, although the number of tokens removed may differ across decoder layers, the overall pruning schedule is fixed and applied uniformly to all input samples and tasks, failing to align token elimination with the model's holistic reasoning trajectory. Cognitive science indicates that human visual processing often begins with broad exploration to accumulate evidence before narrowing focus as the target becomes distinct. Our experiments reveal an analogous pattern in these models. This observation suggests that neither a fixed pruning schedule nor a heuristic layer-wise strategy can optimally accommodate the diverse complexities inherent in different inputs. To overcome this limitation, we introduce Complexity-Adaptive Pruning (AutoPrune), a training-free, plug-and-play framework that tailors pruning policies to varying sample and task complexities. Specifically, AutoPrune quantifies the mutual information between visual and textual tokens, then projects this signal to a budget-constrained logistic retention curve. Each such logistic curve, defined by its unique shape, corresponds to the specific complexity of different tasks and can guarantee adherence to predefined computational constraints. We evaluate AutoPrune on standard vision-language tasks and on Vision-Language-Action models for autonomous driving. Notably, when applied to LLaVA-1.5-7B, our method prunes 89% of visual tokens and reduces inference FLOPs by 76.8% while retaining 96.7% of the original accuracy averaged over all tasks. This corresponds to a 9.1% improvement over the recent work PDrop, demonstrating the effectiveness. Code is available at https://github.com/AutoLab-SAI-SJTU/AutoPrune.

Paperid: 471, https://arxiv.org/pdf/2509.23823.pdf GitHub

Abstract:
Cross-platform robot control remains difficult because hardware interfaces, data formats, and control paradigms vary widely, which fragments toolchains and slows deployment. To address this, we present Control Your Robot, a modular, general-purpose framework that unifies data collection and policy deployment across diverse platforms. The system reduces fragmentation through a standardized workflow with modular design, unified APIs, and a closed-loop architecture. It supports flexible robot registration, dual-mode control with teleoperation and trajectory playback, and seamless integration from multimodal data acquisition to inference. Experiments on single-arm and dual-arm systems show efficient, low-latency data collection and effective support for policy learning with imitation learning and vision-language-action models. Policies trained on data gathered by Control Your Robot match expert demonstrations closely, indicating that the framework enables scalable and reproducible robot learning across platforms.

Paperid: 472, https://arxiv.org/pdf/2509.23770.pdf GitHub GitHub

Abstract:
The success of contrastive learning depends on the construction and utilization of high-quality positive pairs. However, current methods face critical limitations on two fronts: on the construction side, both handcrafted and generative augmentations often suffer from limited diversity and risk semantic corruption; on the learning side, the absence of a quality assessment mechanism leads to suboptimal supervision where all pairs are treated equally. To tackle these challenges, we propose GenView++, a unified framework that addresses both fronts by introducing two synergistic innovations. To improve pair construction, GenView++ introduces a multi-source adaptive view generation mechanism to synthesize diverse yet semantically coherent views by dynamically modulating generative parameters across image-conditioned, text-conditioned, and image-text-conditioned strategies. Second, a quality-driven contrastive learning mechanism assesses each pair's semantic alignment and diversity to dynamically reweight their training contribution, prioritizing high-quality pairs while suppressing redundant or misaligned pairs. Extensive experiments demonstrate the effectiveness of GenView++ across both vision and vision-language tasks. For vision representation learning, it improves MoCov2 by +2.5% on ImageNet linear classification. For vision-language learning, it raises the average zero-shot classification accuracy by +12.31% over CLIP and +5.31% over SLIP across ten datasets, and further improves Flickr30k text retrieval R@5 by +3.2%. The code is available at https://github.com/xiaojieli0903/GenViewPlusPlus.

Paperid: 473, https://arxiv.org/pdf/2509.23666.pdf GitHub

Abstract:
Early-Exit Deep Neural Networks enable adaptive inference by allowing prediction at intermediary layers, significantly reducing computational costs and latency. Most of the early exit strategies greedily exit a sample at an intermediary layer if the confidence in class prediction exceeds a predefined threshold that is set using a static validation set. This is problematic as the model might be overconfident in a wrong class. Also, they are not robust to distribution shifts encountered in deployment, which can undermine model trustworthiness and accuracy. To address these challenges, we propose UAT that adapts the threshold for exit decisions using a Multi-Armed Bandit framework, enabling online, unsupervised adjustment of exit decisions. UAT makes decisions based on a new reward function that assesses predictive certainty and its reliability to balance computational efficiency and prediction quality while penalizing unnecessary late exits. We provide guarantees on risk achieved by UAT and validate its performance on diverse tasks spanning vision-language understanding, text generation, and classification. Our framework demonstrates consistent improvements in speedup (1.70-2.10x) with a minimal performance drop (<2%) as compared to full model performance. Our source code is available at https://github.com/Div290/UAT.

Paperid: 474, https://arxiv.org/pdf/2509.23316.pdf GitHub

Abstract:
Object detection has advanced significantly in the closed-set setting, but real-world deployment remains limited by two challenges: poor generalization to unseen categories and insufficient robustness under adverse conditions. Prior research has explored these issues separately: visible-infrared detection improves robustness but lacks generalization, while open-world detection leverages vision-language alignment strategy for category diversity but struggles under extreme environments. This trade-off leaves robustness and diversity difficult to achieve simultaneously. To mitigate these issues, we propose \textbf{C3-OWD}, a curriculum cross-modal contrastive learning framework that unifies both strengths. Stage~1 enhances robustness by pretraining with RGBT data, while Stage~2 improves generalization via vision-language alignment. To prevent catastrophic forgetting between two stages, we introduce an Exponential Moving Average (EMA) mechanism that theoretically guarantees preservation of pre-stage performance with bounded parameter lag and function consistency. Experiments on FLIR, OV-COCO, and OV-LVIS demonstrate the effectiveness of our approach: C3-OWD achieves $80.1$ AP$^{50}$ on FLIR, $48.6$ AP$^{50}_{\text{Novel}}$ on OV-COCO, and $35.7$ mAP$_r$ on OV-LVIS, establishing competitive performance across both robustness and diversity evaluations. Code available at: https://github.com/justin-herry/C3-OWD.git.

Paperid: 475, https://arxiv.org/pdf/2509.23281.pdf GitHub

Abstract:
Large Language Models (LLMs) and Vision-Language Models (VLMs) are increasingly deployed in robotic environments but remain vulnerable to jailbreaking attacks that bypass safety mechanisms and drive unsafe or physically harmful behaviors in the real world. Data-driven defenses such as jailbreak classifiers show promise, yet they struggle to generalize in domains where specialized datasets are scarce, limiting their effectiveness in robotics and other safety-critical contexts. To address this gap, we introduce J-DAPT, a lightweight framework for multimodal jailbreak detection through attention-based fusion and domain adaptation. J-DAPT integrates textual and visual embeddings to capture both semantic intent and environmental grounding, while aligning general-purpose jailbreak datasets with domain-specific reference data. Evaluations across autonomous driving, maritime robotics, and quadruped navigation show that J-DAPT boosts detection accuracy to nearly 100% with minimal overhead. These results demonstrate that J-DAPT provides a practical defense for securing VLMs in robotic applications. Additional materials are made available at: https://j-dapt.github.io.

Paperid: 476, https://arxiv.org/pdf/2509.22674.pdf GitHub

Abstract:
Vision Language Models (VLMs) are improving quickly, but standard benchmarks can hide systematic failures that reduce real world trust. We introduce MATS (Multimodal Audit for Truthful Spatialization), a compact behavioral audit that measures whether models reject visually contradicted statements, and two metrics Spatial Consistency Score (SCS) and Incorrect Agreement Rate (IAR). Instruction tuned generative VLMs (LLaVA 1.5, QwenVLchat) exhibit very low SCS and high IAR, while contrastive encoders (CLIP, SigLIP) are far more robust. Activation patching causally localizes failure loci (mid to late cross attention for generative models, pooled projection components for contrastive models) and suggests concrete repair paths.

Paperid: 477, https://arxiv.org/pdf/2509.22647.pdf GitHub GitHub

Abstract:
Image captioning is a fundamental task that bridges the visual and linguistic domains, playing a critical role in pre-training Large Vision-Language Models (LVLMs). Current state-of-the-art captioning models are typically trained with Supervised Fine-Tuning (SFT), a paradigm that relies on expensive, non-scalable data annotated by humans or proprietary models. This approach often leads to models that memorize specific ground-truth answers, limiting their generality and ability to generate diverse, creative descriptions. To overcome the limitation of SFT, we propose applying the Reinforcement Learning with Verifiable Rewards (RLVR) paradigm to the open-ended task of image captioning. A primary challenge, however, is designing an objective reward function for the inherently subjective nature of what constitutes a "good" caption. We introduce Captioning Reinforcement Learning (CapRL), a novel training framework that redefines caption quality through its utility: a high-quality caption should enable a non-visual language model to accurately answer questions about the corresponding image. CapRL employs a decoupled two-stage pipeline where an LVLM generates a caption, and the objective reward is derived from the accuracy of a separate, vision-free LLM answering Multiple-Choice Questions based solely on that caption. As the first study to apply RLVR to the subjective image captioning task, we demonstrate that CapRL significantly enhances multiple settings. Pretraining on the CapRL-5M caption dataset annotated by CapRL-3B results in substantial gains across 12 benchmarks. Moreover, within the Prism Framework for caption quality evaluation, CapRL achieves performance comparable to Qwen2.5-VL-72B, while exceeding the baseline by an average margin of 8.4%. Code is available here: https://github.com/InternLM/CapRL.

Paperid: 478, https://arxiv.org/pdf/2509.22624.pdf GitHub

Abstract:
Recent Large Language Models (LLMs) and Large Vision-Language Models (LVLMs) increasingly use Reinforcement Learning (RL) for post-pretraining, such as RL with Verifiable Rewards (RLVR) for objective tasks and RL from Human Feedback (RLHF) for subjective tasks. However, RLHF incurs high costs and potential reward-policy mismatch due to reliance on human preferences, while RLVR still wastes supervision by discarding rollouts and correctness signals after each update. To address these challenges, we introduce the Synergistic Policy And Reward Co-Evolving Framework (SPARK), an efficient, on-policy, and stable method that builds on RLVR. Instead of discarding rollouts and correctness data, SPARK recycles this valuable information to simultaneously train the model itself as a generative reward model. This auxiliary training uses a mix of objectives, such as pointwise reward score, pairwise comparison, and evaluation conditioned on further-reflection responses, to teach the model to evaluate and improve its own responses. Our process eliminates the need for a separate reward model and costly human preference data. SPARK creates a positive co-evolving feedback loop: improved reward accuracy yields better policy gradients, which in turn produce higher-quality rollouts that further refine the reward model. Our unified framework supports test-time scaling via self-reflection without external reward models and their associated costs. We show that SPARK achieves significant performance gains on multiple LLM and LVLM models and multiple reasoning, reward models, and general benchmarks. For example, SPARK-VL-7B achieves an average 9.7% gain on 7 reasoning benchmarks, 12.1% on 2 reward benchmarks, and 1.5% on 8 general benchmarks over the baselines, demonstrating robustness and broad generalization.

Paperid: 479, https://arxiv.org/pdf/2509.22548.pdf GitHub GitHub

Abstract:
Vision-and-Language Navigation requires an embodied agent to navigate through unseen environments, guided by natural language instructions and a continuous video stream. Recent advances in VLN have been driven by the powerful semantic understanding of Multimodal Large Language Models. However, these methods typically rely on explicit semantic memory, such as building textual cognitive maps or storing historical visual frames. This type of method suffers from spatial information loss, computational redundancy, and memory bloat, which impede efficient navigation. Inspired by the implicit scene representation in human navigation, analogous to the left brain's semantic understanding and the right brain's spatial cognition, we propose JanusVLN, a novel VLN framework featuring a dual implicit neural memory that models spatial-geometric and visual-semantic memory as separate, compact, and fixed-size neural representations. This framework first extends the MLLM to incorporate 3D prior knowledge from the spatial-geometric encoder, thereby enhancing the spatial reasoning capabilities of models based solely on RGB input. Then, the historical key-value caches from the spatial-geometric and visual-semantic encoders are constructed into a dual implicit memory. By retaining only the KVs of tokens in the initial and sliding window, redundant computation is avoided, enabling efficient incremental updates. Extensive experiments demonstrate that JanusVLN outperforms over 20 recent methods to achieve SOTA performance. For example, the success rate improves by 10.5-35.5 compared to methods using multiple data types as input and by 3.6-10.8 compared to methods using more RGB training data. This indicates that the proposed dual implicit neural memory, as a novel paradigm, explores promising new directions for future VLN research. Ours project page: https://miv-xjtu.github.io/JanusVLN.github.io/.

Paperid: 480, https://arxiv.org/pdf/2509.22437.pdf GitHub

Abstract:
Diagrams convey symbolic information in a visual format rather than a linear stream of words, making them especially challenging for AI models to process. While recent evaluations suggest that vision-language models (VLMs) perform well on diagram-related benchmarks, their reliance on knowledge, reasoning, or modality shortcuts raises concerns about whether they genuinely understand and reason over diagrams. To address this gap, we introduce Chimera, a comprehensive test suite comprising 7,500 high-quality diagrams sourced from Wikipedia; each diagram is annotated with its symbolic content represented by semantic triples along with multi-level questions designed to assess four fundamental aspects of diagram comprehension: entity recognition, relation understanding, knowledge grounding, and visual reasoning. We use Chimera to measure the presence of three types of shortcuts in visual question answering: (1) the visual-memorization shortcut, where VLMs rely on memorized visual patterns; (2) the knowledge-recall shortcut, where models leverage memorized factual knowledge instead of interpreting the diagram; and (3) the Clever-Hans shortcut, where models exploit superficial language patterns or priors without true comprehension. We evaluate 15 open-source VLMs from 7 model families on Chimera and find that their seemingly strong performance largely stems from shortcut behaviors: visual-memorization shortcuts have slight impact, knowledge-recall shortcuts play a moderate role, and Clever-Hans shortcuts contribute significantly. These findings expose critical limitations in current VLMs and underscore the need for more robust evaluation protocols that benchmark genuine comprehension of complex visual inputs (e.g., diagrams) rather than question-answering shortcuts.

Paperid: 481, https://arxiv.org/pdf/2509.22378.pdf GitHub

Abstract:
Recently, Image-to-Music (I2M) generation has garnered significant attention, with potential applications in fields such as gaming, advertising, and multi-modal art creation. However, due to the ambiguous and subjective nature of I2M tasks, most end-to-end methods lack interpretability, leaving users puzzled about the generation results. Even methods based on emotion mapping face controversy, as emotion represents only a singular aspect of art. Additionally, most learning-based methods require substantial computational resources and large datasets for training, hindering accessibility for common users. To address these challenges, we propose the first Vision Language Model (VLM)-based I2M framework that offers high interpretability and low computational cost. Specifically, we utilize ABC notation to bridge the text and music modalities, enabling the VLM to generate music using natural language. We then apply multi-modal Retrieval-Augmented Generation (RAG) and self-refinement techniques to allow the VLM to produce high-quality music without external training. Furthermore, we leverage the generated motivations in text and the attention maps from the VLM to provide explanations for the generated results in both text and image modalities. To validate our method, we conduct both human studies and machine evaluations, where our method outperforms others in terms of music quality and music-image consistency, indicating promising results. Our code is available at https://github.com/RS2002/Image2Music .

Paperid: 482, https://arxiv.org/pdf/2509.22283.pdf GitHub GitHub

Abstract:
Rule-based reinforcement learning has been gaining popularity ever since DeepSeek-R1 has demonstrated its success through simple verifiable rewards. In the domain of document analysis, reinforcement learning is not as prevalent, even though many downstream tasks may benefit from the emerging properties of reinforcement learning, particularly the enhanced reason capabilities. We study the effects of rule-based reinforcement learning with the task of Document Image Classification which is one of the most commonly studied downstream tasks in document analysis. We find that reinforcement learning tends to have better generalisation capabilities to out-of-distritbution data, which we examine in three different scenarios, namely out-of-distribution images, unseen classes and different modalities. Our code is available at https://github.com/jungomi/vision-finetune.

Paperid: 483, https://arxiv.org/pdf/2509.22186.pdf GitHub

Authors:Junbo Niu, Zheng Liu, Zhuangcheng Gu, Bin Wang, Linke Ouyang, Zhiyuan Zhao, Tao Chu, Tianyao He, Fan Wu, Qintong Zhang, Zhenjiang Jin, Guang Liang, Rui Zhang, Wenzheng Zhang, Yuan Qu, Zhifei Ren, Yuefeng Sun, Yuanhong Zheng, Dongsheng Ma, Zirui Tang, Boyu Niu, Ziyang Miao, Hejun Dong, Siyi Qian, Junyuan Zhang, Jingzhou Chen, Fangdong Wang, Xiaomeng Zhao, Liqun Wei, Wei Li, Shasha Wang, Ruiliang Xu, Yuanyuan Cao, Lu Chen, Qianqian Wu, Huaiyu Gu, Lindong Lu, Keming Wang, Dechen Lin, Guanlin Shen, Xuanhe Zhou, Linfeng Zhang, Yuhang Zang, Xiaoyi Dong, Jiaqi Wang, Bo Zhang, Lei Bai, Pei Chu, Weijia Li, Jiang Wu, Lijun Wu, Zhenxiang Li, Guangyu Wang, Zhongying Tu, Chao Xu, Kai Chen, Yu Qiao, Bowen Zhou, Dahua Lin, Wentao Zhang, Conghui He

Abstract:
We introduce MinerU2.5, a 1.2B-parameter document parsing vision-language model that achieves state-of-the-art recognition accuracy while maintaining exceptional computational efficiency. Our approach employs a coarse-to-fine, two-stage parsing strategy that decouples global layout analysis from local content recognition. In the first stage, the model performs efficient layout analysis on downsampled images to identify structural elements, circumventing the computational overhead of processing high-resolution inputs. In the second stage, guided by the global layout, it performs targeted content recognition on native-resolution crops extracted from the original image, preserving fine-grained details in dense text, complex formulas, and tables. To support this strategy, we developed a comprehensive data engine that generates diverse, large-scale training corpora for both pretraining and fine-tuning. Ultimately, MinerU2.5 demonstrates strong document parsing ability, achieving state-of-the-art performance on multiple benchmarks, surpassing both general-purpose and domain-specific models across various recognition tasks, while maintaining significantly lower computational overhead.

Paperid: 484, https://arxiv.org/pdf/2509.22093.pdf GitHub GitHub

Abstract:
Robotic manipulation with Vision-Language-Action models requires efficient inference over long-horizon multi-modal context, where attention to dense visual tokens dominates computational cost. Existing methods optimize inference speed by reducing visual redundancy within VLA models, but they overlook the varying redundancy across robotic manipulation stages. We observe that the visual token redundancy is higher in coarse manipulation phase than in fine-grained operations, and is strongly correlated with the action dynamic. Motivated by this observation, we propose \textbf{A}ction-aware \textbf{D}ynamic \textbf{P}runing (\textbf{ADP}), a multi-modal pruning framework that integrates text-driven token selection with action-aware trajectory gating. Our method introduces a gating mechanism that conditions the pruning signal on recent action trajectories, using past motion windows to adaptively adjust token retention ratios in accordance with dynamics, thereby balancing computational efficiency and perceptual precision across different manipulation stages. Extensive experiments on the LIBERO suites and diverse real-world scenarios demonstrate that our method significantly reduces FLOPs and action inference latency (\textit{e.g.} $1.35 \times$ speed up on OpenVLA-OFT) while maintaining competitive success rates (\textit{e.g.} 25.8\% improvements with OpenVLA) compared to baselines, thereby providing a simple plug-in path to efficient robot policies that advances the efficiency and performance frontier of robotic manipulation. Our project website is: \href{https://vla-adp.github.io/}{ADP.com}.

Paperid: 485, https://arxiv.org/pdf/2509.21991.pdf GitHub

Abstract:
Efficient processing of high-resolution images is crucial for real-world vision-language applications. However, existing Large Vision-Language Models (LVLMs) incur substantial computational overhead due to the large number of vision tokens. With the advent of "thinking with images" models, reasoning now extends beyond text to the visual domain. This capability motivates our two-stage "coarse-to-fine" reasoning pipeline: first, a downsampled image is analyzed to identify task-relevant regions; then, only these regions are cropped at full resolution and processed in a subsequent reasoning stage. This approach reduces computational cost while preserving fine-grained visual details where necessary. A major challenge lies in inferring which regions are truly relevant to a given query. Recent related methods often fail in the first stage after input-image downsampling, due to perception-driven reasoning, where clear visual information is required for effective reasoning. To address this issue, we propose ERGO (Efficient Reasoning & Guided Observation) that performs reasoning-driven perception-leveraging multimodal context to determine where to focus. Our model can account for perceptual uncertainty, expanding the cropped region to cover visually ambiguous areas for answering questions. To this end, we develop simple yet effective reward components in a reinforcement learning framework for coarse-to-fine perception. Across multiple datasets, our approach delivers higher accuracy than the original model and competitive methods, with greater efficiency. For instance, ERGO surpasses Qwen2.5-VL-7B on the V* benchmark by 4.7 points while using only 23% of the vision tokens, achieving a 3x inference speedup. The code and models can be found at: https://github.com/nota-github/ERGO.

Paperid: 486, https://arxiv.org/pdf/2509.21989.pdf GitHub

Abstract:
We propose a novel approach for disentangling visual and semantic features from the backbones of pre-trained diffusion models, enabling visual correspondence in a manner analogous to the well-established semantic correspondence. While diffusion model backbones are known to encode semantically rich features, they must also contain visual features to support their image synthesis capabilities. However, isolating these visual features is challenging due to the absence of annotated datasets. To address this, we introduce an automated pipeline that constructs image pairs with annotated semantic and visual correspondences based on existing subject-driven image generation datasets, and design a contrastive architecture to separate the two feature types. Leveraging the disentangled representations, we propose a new metric, Visual Semantic Matching (VSM), that quantifies visual inconsistencies in subject-driven image generation. Empirical results show that our approach outperforms global feature-based metrics such as CLIP, DINO, and vision--language models in quantifying visual inconsistencies while also enabling spatial localization of inconsistent regions. To our knowledge, this is the first method that supports both quantification and localization of inconsistencies in subject-driven generation, offering a valuable tool for advancing this task. Project Page:https://abdo-eldesokey.github.io/mind-the-glitch/

Paperid: 487, https://arxiv.org/pdf/2509.21650.pdf GitHub

Abstract:
Mask-based pretraining has become a cornerstone of modern large-scale models across language, vision, and recently biology. Despite its empirical success, its role and limits in learning data representations have been unclear. In this work, we show that the behavior of mask-based pretraining can be directly characterized by test risk in high-dimensional minimum-norm ("ridge-less") linear regression, without relying on further model specifications. Further analysis of linear models uncovers several novel aspects of mask-based pretraining. The theoretical framework and its implications have been validated across diverse neural architectures (including MLPs, CNNs, and Transformers) applied to both vision and language tasks. Guided by our theory, we propose an embarrassingly simple yet overlooked pretraining scheme named Randomly Random Mask AutoEncoding (R$^2$MAE), which enforces capturing multi-scale features from data and is able to outperform optimal fixed mask ratio settings in our linear model framework. We implement R$^2$MAE in vision, language, DNA sequence, and single-cell models, where it consistently outperforms standard and more complicated masking schemes, leading to improvements for state-of-the-art models. Our code is available at: https://github.com/MingzeDong/r2mae

Paperid: 488, https://arxiv.org/pdf/2509.21102.pdf GitHub

Abstract:
Understanding what deep learning (DL) models learn is essential for the safe deployment of artificial intelligence (AI) in clinical settings. While previous work has focused on pixel-based explainability methods, less attention has been paid to the textual concepts learned by these models, which may better reflect the reasoning used by clinicians. We introduce Mammo-CLIP Dissect, the first concept-based explainability framework for systematically dissecting DL vision models trained for mammography. Leveraging a mammography-specific vision-language model (Mammo-CLIP) as a "dissector," our approach labels neurons at specified layers with human-interpretable textual concepts and quantifies their alignment to domain knowledge. Using Mammo-CLIP Dissect, we investigate three key questions: (1) how concept learning differs between DL vision models trained on general image datasets versus mammography-specific datasets; (2) how fine-tuning for downstream mammography tasks affects concept specialisation; and (3) which mammography-relevant concepts remain underrepresented. We show that models trained on mammography data capture more clinically relevant concepts and align more closely with radiologists' workflows than models not trained on mammography data. Fine-tuning for task-specific classification enhances the capture of certain concept categories (e.g., benign calcifications) but can reduce coverage of others (e.g., density-related features), indicating a trade-off between specialisation and generalisation. Our findings show that Mammo-CLIP Dissect provides insights into how convolutional neural networks (CNNs) capture mammography-specific knowledge. By comparing models across training data and fine-tuning regimes, we reveal how domain-specific training and task-specific adaptation shape concept learning. Code and concept set are available: https://github.com/Suaiba/Mammo-CLIP-Dissect.

Paperid: 489, https://arxiv.org/pdf/2509.20961.pdf GitHub

Abstract:
The dynamic propagation of social media has broadened the reach of financial advisory content through podcast videos, yet extracting insights from lengthy, multimodal segments (30-40 minutes) remains challenging. We introduce FASTER (Financial Advisory Summariser with Textual Embedded Relevant images), a modular framework that tackles three key challenges: (1) extracting modality-specific features, (2) producing optimized, concise summaries, and (3) aligning visual keyframes with associated textual points. FASTER employs BLIP for semantic visual descriptions, OCR for textual patterns, and Whisper-based transcription with Speaker diarization as BOS features. A modified Direct Preference Optimization (DPO)-based loss function, equipped with BOS-specific fact-checking, ensures precision, relevance, and factual consistency against the human-aligned summary. A ranker-based retrieval mechanism further aligns keyframes with summarized content, enhancing interpretability and cross-modal coherence. To acknowledge data resource scarcity, we introduce Fin-APT, a dataset comprising 470 publicly accessible financial advisory pep-talk videos for robust multimodal research. Comprehensive cross-domain experiments confirm FASTER's strong performance, robustness, and generalizability when compared to Large Language Models (LLMs) and Vision-Language Models (VLMs). By establishing a new standard for multimodal summarization, FASTER makes financial advisory content more accessible and actionable, thereby opening new avenues for research. The dataset and code are available at: https://github.com/sarmistha-D/FASTER

Paperid: 490, https://arxiv.org/pdf/2509.19760.pdf GitHub

Abstract:
Recent advances in Large Vision-Language models (LVLM) have spurred significant progress in document parsing task. Compared to traditional pipeline-based methods, end-to-end paradigms have shown their excellence in converting PDF images into structured outputs through integrated Optical Character Recognition (OCR), table recognition, mathematical formula recognition and so on. However, the absence of explicit analytical stages for document layouts and reading orders limits the LVLM's capability in handling complex document types such as multi-column newspapers or posters. To address this limitation, we propose in this report Logics-Parsing: an end-to-end LVLM-based model augmented with reinforcement learning. Our model incorporates meticulously designed reward mechanisms to optimize complex layout analysis and reading order inference. In addition, we expand the model's versatility by incorporating diverse data types such as chemical formulas and handwritten Chinese characters into supervised fine-tuning. Finally, to enable rigorous evaluation of our approach, we introduce LogicsParsingBench, a curated set of 1,078 page-level PDF images spanning nine major categories and over twenty sub-categories, which will be released later. Comprehensive experiments conducted on LogicsParsingBench have validated the efficacy and State-of-the-art (SOTA) performance of our proposed model across diverse document analysis scenarios. Project Page: https://github.com/alibaba/Logics-Parsing

Paperid: 491, https://arxiv.org/pdf/2509.19203.pdf GitHub

Abstract:
Contrastively-trained Vision-Language Models (VLMs), such as CLIP, have become the standard approach for learning discriminative vision-language representations. However, these models often exhibit shallow language understanding, manifesting bag-of-words behaviour. These limitations are reinforced by their dual-encoder design, which induces a modality gap. Additionally, the reliance on vast web-collected data corpora for training makes the process computationally expensive and introduces significant privacy concerns. To address these limitations, in this work, we challenge the necessity of vision encoders for retrieval tasks by introducing a vision-free, single-encoder retrieval pipeline. Departing from the traditional text-to-image retrieval paradigm, we migrate to a text-to-text paradigm with the assistance of VLLM-generated structured image descriptions. We demonstrate that this paradigm shift has significant advantages, including a substantial reduction of the modality gap, improved compositionality, and better performance on short and long caption queries, all attainable with only a few hours of calibration on two GPUs. Additionally, substituting raw images with textual descriptions introduces a more privacy-friendly alternative for retrieval. To further assess generalisation and address some of the shortcomings of prior compositionality benchmarks, we release two benchmarks derived from Flickr30k and COCO, containing diverse compositional queries made of short captions, which we coin subFlickr and subCOCO. Our vision-free retriever matches and often surpasses traditional multimodal models. Importantly, our approach achieves state-of-the-art zero-shot performance on multiple retrieval and compositionality benchmarks, with models as small as 0.3B parameters. Code is available at: https://github.com/IoannaNti/LexiCLIP

Paperid: 492, https://arxiv.org/pdf/2509.19018.pdf GitHub

Abstract:
Recent advances in multimodal large language models (LLMs) have led to significant progress in understanding, generation, and retrieval tasks. However, current solutions often treat these tasks in isolation or require training LLMs from scratch, resulting in high computational costs and limited generalization across modalities. In this work, we present OmniBridge, a unified and modular multimodal framework that supports vision-language understanding, generation, and retrieval within a unified architecture. OmniBridge adopts a language-centric design that reuses pretrained LLMs and introduces a lightweight bidirectional latent alignment module. To address the challenge of task interference, we propose a two-stage decoupled training strategy: supervised fine-tuning and latent space alignment for aligning LLM behavior with multimodal reasoning, and semantic-guided diffusion training to align cross-modal latent spaces via learnable query embeddings. Extensive experiments across a wide range of benchmarks demonstrate that OmniBridge achieves competitive or state-of-the-art performance in all three tasks. Moreover, our results highlight the effectiveness of latent space alignment for unifying multimodal modeling under a shared representation space. Code and models are released at https://github.com/xiao-xt/OmniBridge.

Paperid: 493, https://arxiv.org/pdf/2509.19003.pdf GitHub

Abstract:
Chain of thought reasoning has demonstrated remarkable success in large language models, yet its adaptation to vision-language reasoning remains an open challenge with unclear best practices. Existing attempts typically employ reasoning chains at a coarse-grained level, which struggles to perform fine-grained structured reasoning and, more importantly, are difficult to evaluate the reward and quality of intermediate reasoning. In this work, we delve into chain of step reasoning for vision-language models, enabling assessing reasoning step quality accurately and leading to effective reinforcement learning and inference-time scaling with fine-grained rewards. We present a simple, effective, and fully transparent framework, including the step-level reasoning data, process reward model (PRM), and reinforcement learning training. With the proposed approaches, our models set strong baselines with consistent improvements on challenging vision-language benchmarks. More importantly, we conduct a thorough empirical analysis and ablation study, unveiling the impact of each component and several intriguing properties of inference-time scaling. We believe this paper serves as a baseline for vision-language models and offers insights into more complex multimodal reasoning. Our dataset, PRM, and code will be available at https://github.com/baaivision/CoS.

Paperid: 494, https://arxiv.org/pdf/2509.18905.pdf GitHub

Abstract:
Visual Spatial Reasoning (VSR) is a core human cognitive ability and a critical requirement for advancing embodied intelligence and autonomous systems. Despite recent progress in Vision-Language Models (VLMs), achieving human-level VSR remains highly challenging due to the complexity of representing and reasoning over three-dimensional space. In this paper, we present a systematic investigation of VSR in VLMs, encompassing a review of existing methodologies across input modalities, model architectures, training strategies, and reasoning mechanisms. Furthermore, we categorize spatial intelligence into three levels of capability, ie, basic perception, spatial understanding, spatial planning, and curate SIBench, a spatial intelligence benchmark encompassing nearly 20 open-source datasets across 23 task settings. Experiments with state-of-the-art VLMs reveal a pronounced gap between perception and reasoning, as models show competence in basic perceptual tasks but consistently underperform in understanding and planning tasks, particularly in numerical estimation, multi-view reasoning, temporal dynamics, and spatial imagination. These findings underscore the substantial challenges that remain in achieving spatial intelligence, while providing both a systematic roadmap and a comprehensive benchmark to drive future research in the field. The related resources of this study are accessible at https://sibench.github.io/Awesome-Visual-Spatial-Reasoning/.

Paperid: 495, https://arxiv.org/pdf/2509.18865.pdf GitHub

Abstract:
We propose Bilateral Control-Based Imitation Learning via Vision-Language Fusion for Action Generation (Bi-VLA), a novel framework that extends bilateral control-based imitation learning to handle more than one task within a single model. Conventional bilateral control methods exploit joint angle, velocity, torque, and vision for precise manipulation but require task-specific models, limiting their generality. Bi-VLA overcomes this limitation by utilizing robot joint angle, velocity, and torque data from leader-follower bilateral control with visual features and natural language instructions through SigLIP and FiLM-based fusion. We validated Bi-VLA on two task types: one requiring supplementary language cues and another distinguishable solely by vision. Real-robot experiments showed that Bi-VLA successfully interprets vision-language combinations and improves task success rates compared to conventional bilateral control-based imitation learning. Our Bi-VLA addresses the single-task limitation of prior bilateral approaches and provides empirical evidence that combining vision and language significantly enhances versatility. Experimental results validate the effectiveness of Bi-VLA in real-world tasks. For additional material, please visit the website: https://mertcookimg.github.io/bi-vla/

Paperid: 496, https://arxiv.org/pdf/2509.18715.pdf GitHub

Abstract:
Object Re-IDentification (ReID) aims to recognize individuals across non-overlapping camera views. While recent advances have achieved remarkable progress, most existing models are constrained to either single-domain or cross-domain scenarios, limiting their real-world applicability. Single-domain models tend to overfit to domain-specific features, whereas cross-domain models often rely on diverse normalization strategies that may inadvertently suppress identity-specific discriminative cues. To address these limitations, we propose an Attribute Prompt Composition (APC) framework, which exploits textual semantics to jointly enhance discrimination and generalization. Specifically, we design an Attribute Prompt Generator (APG) consisting of a Semantic Attribute Dictionary (SAD) and a Prompt Composition Module (PCM). SAD is an over-complete attribute dictionary to provide rich semantic descriptions, while PCM adaptively composes relevant attributes from SAD to generate discriminative attribute-aware features. In addition, motivated by the strong generalization ability of Vision-Language Models (VLM), we propose a Fast-Slow Training Strategy (FSTS) to balance ReID-specific discrimination and generalizable representation learning. Specifically, FSTS adopts a Fast Update Stream (FUS) to rapidly acquire ReID-specific discriminative knowledge and a Slow Update Stream (SUS) to retain the generalizable knowledge inherited from the pre-trained VLM. Through a mutual interaction, the framework effectively focuses on ReID-relevant features while mitigating overfitting. Extensive experiments on both conventional and Domain Generalized (DG) ReID datasets demonstrate that our framework surpasses state-of-the-art methods, exhibiting superior performances in terms of both discrimination and generalization. The source code is available at https://github.com/AWangYQ/APC.

Paperid: 497, https://arxiv.org/pdf/2509.18693.pdf GitHub

Abstract:
Land-cover understanding in remote sensing increasingly demands class-agnostic systems that generalize across datasets while remaining spatially precise and interpretable. We study a geometry-first discovery-and-interpretation setting under domain shift, where candidate regions are delineated class-agnostically and supervision avoids lexical class names via anonymized identifiers. Complementary to open-set recognition and open-world learning, we focus on coupling class-agnostic mask evidence with taxonomy-grounded scene interpretation, rather than unknown rejection or continual class expansion. We propose MVT, a three-stage framework that (i) extracts boundary-faithful region masks using SAM2 with domain adaptation, (ii) performs mask-grounded semantic tagging and scene description generation via dual-step LoRA fine-tuning of multimodal LLMs, and (iii) evaluates outputs with LLM-as-judge scoring calibrated by stratified expert ratings. On cross-dataset segmentation transfer (train on OpenEarthMap, evaluate on LoveDA), domain-adapted SAM2 improves mask quality; meanwhile, dual-step MLLM fine-tuning yields more accurate taxonomy-aligned tags and more informative mask-grounded scene descriptions.

Paperid: 498, https://arxiv.org/pdf/2509.18592.pdf GitHub

Abstract:
Rapid adaptation in unseen environments is essential for scalable real-world autonomy, yet existing approaches rely on exhaustive exploration or rigid navigation policies that fail to generalize. We present VLN-Zero, a two-phase vision-language navigation framework that leverages vision-language models to efficiently construct symbolic scene graphs and enable zero-shot neurosymbolic navigation. In the exploration phase, structured prompts guide VLM-based search toward informative and diverse trajectories, yielding compact scene graph representations. In the deployment phase, a neurosymbolic planner reasons over the scene graph and environmental observations to generate executable plans, while a cache-enabled execution module accelerates adaptation by reusing previously computed task-location trajectories. By combining rapid exploration, symbolic reasoning, and cache-enabled execution, the proposed framework overcomes the computational inefficiency and poor generalization of prior vision-language navigation methods, enabling robust and scalable decision-making in unseen environments. VLN-Zero achieves 2x higher success rate compared to state-of-the-art zero-shot models, outperforms most fine-tuned baselines, and reaches goal locations in half the time with 55% fewer VLM calls on average compared to state-of-the-art models across diverse environments. Codebase, datasets, and videos for VLN-Zero are available at: https://vln-zero.github.io/.

Paperid: 499, https://arxiv.org/pdf/2509.18588.pdf GitHub

Abstract:
Recent unified models such as GPT-5 have achieved encouraging progress on vision-language tasks. However, these unified models typically fail to correctly understand ECG signals and provide accurate medical diagnoses, nor can they correctly generate ECG signals. To address these limitations, we propose UniECG, the first unified model for ECG capable of concurrently performing evidence-based ECG interpretation and text-conditioned ECG generation tasks. Through a decoupled two-stage training approach, the model first learns evidence-based interpretation skills (ECG-to-Text), and then injects ECG generation capabilities (Text-to-ECG) via latent space alignment. UniECG can autonomously choose to interpret or generate an ECG based on user input, significantly extending the capability boundaries of current ECG models. Our code and checkpoints will be made publicly available at https://github.com/PKUDigitalHealth/UniECG upon acceptance.

Paperid: 500, https://arxiv.org/pdf/2509.18282.pdf GitHub

Abstract:
Robotic manipulation policies often fail to generalize because they must simultaneously learn where to attend, what actions to take, and how to execute them. We argue that high-level reasoning about where and what can be offloaded to vision-language models (VLMs), leaving policies to specialize in how to act. We present PEEK (Policy-agnostic Extraction of Essential Keypoints), which fine-tunes VLMs to predict a unified point-based intermediate representation: 1. end-effector paths specifying what actions to take, and 2. task-relevant masks indicating where to focus. These annotations are directly overlaid onto robot observations, making the representation policy-agnostic and transferable across architectures. To enable scalable training, we introduce an automatic annotation pipeline, generating labeled data across 20+ robot datasets spanning 9 embodiments. In real-world evaluations, PEEK consistently boosts zero-shot generalization, including a 41.4x real-world improvement for a 3D policy trained only in simulation, and 2-3.5x gains for both large VLAs and small manipulation policies. By letting VLMs absorb semantic and visual complexity, PEEK equips manipulation policies with the minimal cues they need--where, what, and how. Website at https://peek-robot.github.io/.

Paperid: 501, https://arxiv.org/pdf/2509.18119.pdf GitHub

Abstract:
Building general-purpose graphical user interface (GUI) agents has become increasingly promising with the progress in vision language models. However, developing effective mobile GUI agents with reinforcement learning (RL) remains challenging due to the heavy-tailed distribution of task difficulty and the inefficiency of large-scale environment sampling. We present an online agentic reinforcement learning framework MOBILERL to enhance GUI agents in mobile environments. Its core component is the Difficulty-Adaptive GRPO (ADAGRPO) algorithm. In ADAGRPO, we design difficulty-adaptive positive replay and failure curriculum filtering to adapt the model to different task difficulties. We introduce the shortest path reward adjustment strategy to reshape rewards concerning the task length in multi-turn agentic tasks. Those strategies jointly stabilize RL training, improve sample efficiency, and generate strong performance across diverse mobile apps and tasks. We apply MOBILERL to two open models (Qwen2.5-VL-7B-Instruct and GLM-4.1V-9B-Base). The resultant MOBILERL-9B model achieves state-of-the-art results in terms of success rates on both AndroidWorld (75.8%) and AndroidLab (46.8%). The MOBILERL framework is adopted in the AutoGLM products, and also open-sourced at https://github.com/THUDM/MobileRL.

Paperid: 502, https://arxiv.org/pdf/2509.17664.pdf GitHub

Abstract:
While vision language models (VLMs) excel in 2D semantic visual understanding, their ability to quantitatively reason about 3D spatial relationships remains under-explored, due to the deficiency of 2D images' spatial representation ability. In this paper, we analyze the problem hindering VLMs' spatial understanding abilities and propose SD-VLM, a novel framework that significantly enhances fundamental spatial perception abilities of VLMs through two key contributions: (1) propose Massive Spatial Measuring and Understanding (MSMU) dataset with precise spatial annotations, and (2) introduce a simple depth positional encoding method strengthening VLMs' spatial awareness. MSMU dataset covers massive quantitative spatial tasks with 700K QA pairs, 2.5M physical numerical annotations, and 10K chain-of-thought augmented samples. We have trained SD-VLM, a strong generalist VLM which shows superior quantitative spatial measuring and understanding capability. SD-VLM not only achieves state-of-the-art performance on our proposed MSMU-Bench, but also shows spatial generalization abilities on other spatial understanding benchmarks including Q-Spatial and SpatialRGPT-Bench. Extensive experiments demonstrate that SD-VLM outperforms GPT-4o and Intern-VL3-78B by 26.91% and 25.56% respectively on MSMU-Bench. Code and models are released at https://github.com/cpystan/SD-VLM.

Paperid: 503, https://arxiv.org/pdf/2509.17598.pdf GitHub

Abstract:
Test-time adaptation (TTA) has gained increasing popularity due to its efficacy in addressing ``distribution shift'' issue while simultaneously protecting data privacy. However, most prior methods assume that a paired source domain model and target domain sharing the same label space coexist, heavily limiting their applicability. In this paper, we investigate a more general source model capable of adaptation to multiple target domains without needing shared labels. This is achieved by using a pre-trained vision-language model (VLM), \egno, CLIP, that can recognize images through matching with class descriptions. While the zero-shot performance of VLMs is impressive, they struggle to effectively capture the distinctive attributes of a target domain. To that end, we propose a novel method -- Context-aware Language-driven TTA (COLA). The proposed method incorporates a lightweight context-aware module that consists of three key components: a task-aware adapter, a context-aware unit, and a residual connection unit for exploring task-specific knowledge, domain-specific knowledge from the VLM and prior knowledge of the VLM, respectively. It is worth noting that the context-aware module can be seamlessly integrated into a frozen VLM, ensuring both minimal effort and parameter efficiency. Additionally, we introduce a Class-Balanced Pseudo-labeling (CBPL) strategy to mitigate the adverse effects caused by class imbalance. We demonstrate the effectiveness of our method not only in TTA scenarios but also in class generalisation tasks. The source code is available at https://github.com/NUDT-Bai-Group/COLA-TTA.

Paperid: 504, https://arxiv.org/pdf/2509.17562.pdf GitHub

Abstract:
Modern computer vision is converging on a closed loop in which perception, reasoning and generation mutually reinforce each other. However, this loop remains incomplete: the top-down influence of high-level reasoning on the foundational learning of low-level perceptual features is not yet underexplored. This paper addresses this gap by proposing a new paradigm for pretraining foundation models in downstream domains. We introduce Visual insTruction Pretraining (ViTP), a novel approach that directly leverages reasoning to enhance perception. ViTP embeds a Vision Transformer (ViT) backbone within a Vision-Language Model and pretrains it end-to-end using a rich corpus of visual instruction data curated from target downstream domains. ViTP is powered by our proposed Visual Robustness Learning (VRL), which compels the ViT to learn robust and domain-relevant features from a sparse set of visual tokens. Extensive experiments on 16 challenging remote sensing and medical imaging benchmarks demonstrate that ViTP establishes new state-of-the-art performance across a diverse range of downstream tasks. The code is available at https://github.com/zcablii/ViTP.

Paperid: 505, https://arxiv.org/pdf/2509.17481.pdf GitHub

Abstract:
Large Vision-Language Models (LVLMs) have recently demonstrated remarkable progress, yet hallucination remains a critical barrier, particularly in chart understanding, which requires sophisticated perceptual and cognitive abilities as well as rigorous factual accuracy. While prior work has investigated hallucinations and chart comprehension independently, their intersection remains largely unexplored. To address this gap, we present ChartHal, a benchmark that features a fine-grained taxonomy of hallucination scenarios in chart understanding, along with a human-validated dataset of 1,062 samples. Our evaluation shows that state-of-the-art LVLMs suffer from severe hallucinations on ChartHal, including proprietary models such as GPT-5 and o4-mini, which achieve only 34.46% and 22.79% accuracy, respectively. Further analysis reveals that questions involving information absent from or contradictory to charts are especially likely to trigger hallucinations, underscoring the urgent need for more robust mitigation strategies. Code and data are available at https://github.com/ymcui/ChartHal .

Paperid: 506, https://arxiv.org/pdf/2509.17321.pdf GitHub

Abstract:
Data scarcity remains one of the most limiting factors in driving progress in robotics. However, the amount of available robotics data in the wild is growing exponentially, creating new opportunities for large-scale data utilization. Reliable temporal task completion prediction could help automatically annotate and curate this data at scale. The Generative Value Learning (GVL) approach was recently proposed, leveraging the knowledge embedded in vision-language models (VLMs) to predict task progress from visual observations. Building upon GVL, we propose OpenGVL, a comprehensive benchmark for estimating task progress across diverse challenging manipulation tasks involving both robotic and human embodiments. We evaluate the capabilities of publicly available open-source foundation models, showing that open-source model families significantly underperform closed-source counterparts, achieving only approximately $70\%$ of their performance on temporal progress prediction tasks. Furthermore, we demonstrate how OpenGVL can serve as a practical tool for automated data curation and filtering, enabling efficient quality assessment of large-scale robotics datasets. We release the benchmark along with the complete codebase at \href{github.com/budzianowski/opengvl}{OpenGVL}.

Paperid: 507, https://arxiv.org/pdf/2509.17177.pdf GitHub

Abstract:
Video action recognition is a fundamental task in computer vision, but state-of-the-art models are often computationally expensive and rely on extensive video pre-training. In parallel, large-scale vision-language models like Contrastive Language-Image Pre-training (CLIP) offer powerful zero-shot capabilities on static images, while motion vectors (MV) provide highly efficient temporal information directly from compressed video streams. To synergize the strengths of these paradigms, we propose MoCLIP-Lite, a simple yet powerful two-stream late fusion framework for efficient video recognition. Our approach combines features from a frozen CLIP image encoder with features from a lightweight, supervised network trained on raw MV. During fusion, both backbones are frozen, and only a tiny Multi-Layer Perceptron (MLP) head is trained, ensuring extreme efficiency. Through comprehensive experiments on the UCF101 dataset, our method achieves a remarkable 89.2% Top-1 accuracy, significantly outperforming strong zero-shot (65.0%) and MV-only (66.5%) baselines. Our work provides a new, highly efficient baseline for video understanding that effectively bridges the gap between large static models and dynamic, low-cost motion cues. Our code and models are available at https://github.com/microa/MoCLIP-Lite.

Paperid: 509, https://arxiv.org/pdf/2509.17065.pdf GitHub

Abstract:
Echocardiography is a vital non-invasive modality for cardiac assessment, with left ventricular ejection fraction (LVEF) serving as a key indicator of heart function. Existing LVEF estimation methods depend on large-scale annotated video datasets, which are costly and limit adaptability across various clinical settings. Recent vision-language models for echocardiography, such as EchoCLIP, apply image-to-text pretraining but fail to capture crucial temporal dynamics and localized cardiac structures essential for accurate diagnosis. To address these challenges, we propose CardiacCLIP, a video-based framework that enhances LVEF prediction through attention-based frame aggregation and multi-resolution input scaling. Specifically, we introduce MFL (Multi Frame Learning), a novel attention-based mechanism for selectively fusing informative frames, and EchoZoom, a multi-scale feature extraction strategy that refines spatial representations of cardiac structures. As a novel adaptation of CLIP models for few-shot echocardiogram video analysis, our approach significantly improves diagnostic accuracy, reducing MAE by 2.07 on the EchoNet-Dynamic dataset under 1-shot setting. The code is available at https://github.com/xmed-lab/CardiacCLIP.

Paperid: 510, https://arxiv.org/pdf/2509.17024.pdf GitHub

Abstract:
Adverse Weather Image Restoration (AWIR) is a highly challenging task due to the unpredictable and dynamic nature of weather-related degradations. Traditional task-specific methods often fail to generalize to unseen or complex degradation types, while recent prompt-learning approaches depend heavily on the degradation estimation capabilities of vision-language models, resulting in inconsistent restorations. In this paper, we propose \textbf{LCDiff}, a novel framework comprising two key components: \textit{Lumina-Chroma Decomposition Network} (LCDN) and \textit{Lumina-Guided Diffusion Model} (LGDM). LCDN processes degraded images in the YCbCr color space, separately handling degradation-related luminance and degradation-invariant chrominance components. This decomposition effectively mitigates weather-induced degradation while preserving color fidelity. To further enhance restoration quality, LGDM leverages degradation-related luminance information as a guiding condition, eliminating the need for explicit degradation prompts. Additionally, LGDM incorporates a \textit{Dynamic Time Step Loss} to optimize the denoising network, ensuring a balanced recovery of both low- and high-frequency features in the image. Finally, we present DriveWeather, a comprehensive all-weather driving dataset designed to enable robust evaluation. Extensive experiments demonstrate that our approach surpasses state-of-the-art methods, setting a new benchmark in AWIR. The dataset and code are available at: https://github.com/fiwy0527/LCDiff.

Paperid: 511, https://arxiv.org/pdf/2509.16805.pdf GitHub

Abstract:
Large Vision-Language Models (LVLMs) have achieved strong performance on vision-language tasks, particularly Visual Question Answering (VQA). While prior work has explored unimodal biases in VQA, the problem of selection bias in Multiple-Choice Question Answering (MCQA), where models may favor specific option tokens (e.g., "A") or positions, remains underexplored. In this paper, we investigate both the presence and nature of selection bias in LVLMs through fine-grained MCQA benchmarks spanning easy, medium, and hard difficulty levels, defined by the semantic similarity of the options. We further propose an inference-time logit-level debiasing method that estimates an ensemble bias vector from general and contextual prompts and applies confidence-adaptive corrections to the model's output. Our method mitigates bias without retraining and is compatible with frozen LVLMs. Extensive experiments across several state-of-the-art models reveal consistent selection biases that intensify with task difficulty, and show that our mitigation approach significantly reduces bias while improving accuracy in challenging settings. This work offers new insights into the limitations of LVLMs in MCQA and presents a practical approach to improve their robustness in fine-grained visual reasoning. Datasets and code are available at: https://github.com/Atabuzzaman/Selection-Bias-of-LVLMs

Paperid: 512, https://arxiv.org/pdf/2509.16656.pdf GitHub

Abstract:
Recent advancements in 2D multimodal large language models (MLLMs) have significantly improved performance in vision-language tasks. However, extending these capabilities to 3D environments remains a distinct challenge due to the complexity of spatial reasoning. Nevertheless, existing 3D benchmarks often lack fine-grained numerical reasoning task annotations, limiting MLLMs' ability to perform precise spatial measurements and complex numerical reasoning. To address this gap, we introduce NUMINA, the first Natural Understanding benchmark for Multi-dimensional Intelligence and Numerical reasoning Abilities to enhance multimodal indoor perceptual understanding. NUMINA features multi-scale annotations and various question-answer pairs, generated using NUMINA-Flow, an automated annotation pipeline that integrates LLM rewriting and rule-based self-verification. We evaluate the performance of various state-of-the-art LLMs on NUMINA following the Chat-Scene framework, demonstrating that current LLMs struggle with multimodal numerical reasoning, particularly in performing precise computations such as distance and volume estimation, highlighting the need for further advancements in 3D models. The dataset and source codes can be obtained from https://github.com/fengshun124/NUMINA.

Paperid: 513, https://arxiv.org/pdf/2509.16611.pdf GitHub

Abstract:
Modern manufacturing demands robotic assembly systems with enhanced flexibility and reliability. However, traditional approaches often rely on programming tailored to each product by experts for fixed settings, which are inherently inflexible to product changes and lack the robustness to handle variations. As Behavior Trees (BTs) are increasingly used in robotics for their modularity and reactivity, we propose a novel hierarchical framework, Video-to-BT, that seamlessly integrates high-level cognitive planning with low-level reactive control, with BTs serving both as the structured output of planning and as the governing structure for execution. Our approach leverages a Vision-Language Model (VLM) to decompose human demonstration videos into subtasks, from which Behavior Trees are generated. During the execution, the planned BTs combined with real-time scene interpretation enable the system to operate reactively in the dynamic environment, while VLM-driven replanning is triggered upon execution failure. This closed-loop architecture ensures stability and adaptivity. We validate our framework on real-world assembly tasks through a series of experiments, demonstrating high planning reliability, robust performance in long-horizon assembly tasks, and strong generalization across diverse and perturbed conditions. Project website: https://video2bt.github.io/video2bt_page/

Paperid: 514, https://arxiv.org/pdf/2509.16609.pdf GitHub

Abstract:
Accurately assessing image complexity (IC) is critical for computer vision, yet most existing methods rely solely on visual features and often neglect high-level semantic information, limiting their accuracy and generalization. We introduce vision-text fusion for IC modeling. This approach integrates visual and textual semantic features, increasing representational diversity. It also reduces the complexity of the hypothesis space, which enhances both accuracy and generalization in complexity assessment. We propose the D2S (Describe-to-Score) framework, which generates image captions with a pre-trained vision-language model. We propose the feature alignment and entropy distribution alignment mechanisms, D2S guides semantic information to inform complexity assessment while bridging the gap between vision and text modalities. D2S utilizes multi-modal information during training but requires only the vision branch during inference, thereby avoiding multi-modal computational overhead and enabling efficient assessment. Experimental results demonstrate that D2S outperforms existing methods on the IC9600 dataset and maintains competitiveness on no-reference image quality assessment (NR-IQA) benchmark, validating the effectiveness and efficiency of multi-modal fusion in complexity-related tasks. Code is available at: https://github.com/xauat-liushipeng/D2S

Paperid: 515, https://arxiv.org/pdf/2509.16517.pdf GitHub

Abstract:
Multimodal vision-language models (VLMs) have made substantial progress in various tasks that require a combined understanding of visual and textual content, particularly in cultural understanding tasks, with the emergence of new cultural datasets. However, these datasets frequently fall short of providing cultural reasoning while underrepresenting many cultures. In this paper, we introduce the Seeing Culture Benchmark (SCB), focusing on cultural reasoning with a novel approach that requires VLMs to reason on culturally rich images in two stages: i) selecting the correct visual option with multiple-choice visual question answering (VQA), and ii) segmenting the relevant cultural artifact as evidence of reasoning. Visual options in the first stage are systematically organized into three types: those originating from the same country, those from different countries, or a mixed group. Notably, all options are derived from a singular category for each type. Progression to the second stage occurs only after a correct visual option is chosen. The SCB benchmark comprises 1,065 images that capture 138 cultural artifacts across five categories from seven Southeast Asia countries, whose diverse cultures are often overlooked, accompanied by 3,178 questions, of which 1,093 are unique and meticulously curated by human annotators. Our evaluation of various VLMs reveals the complexities involved in cross-modal cultural reasoning and highlights the disparity between visual reasoning and spatial grounding in culturally nuanced scenarios. The SCB serves as a crucial benchmark for identifying these shortcomings, thereby guiding future developments in the field of cultural reasoning. https://github.com/buraksatar/SeeingCulture

Paperid: 516, https://arxiv.org/pdf/2509.16072.pdf GitHub

Abstract:
Language-conditioned robotic manipulation in open-world settings requires not only accurate task execution but also the ability to detect failures for robust deployment in real-world environments. Although recent advances in vision-language models (VLMs) have significantly improved the spatial reasoning and task-planning capabilities of robots, they remain limited in their ability to recognize their own failures. In particular, a critical yet underexplored challenge lies in detecting semantic misalignment errors, where the robot executes a task that is semantically meaningful but inconsistent with the given instruction. To address this, we propose a method for building datasets targeting Semantic Misalignment Failures detection, from existing language-conditioned manipulation datasets. We also present I-FailSense, an open-source VLM framework with grounded arbitration designed specifically for failure detection. Our approach relies on post-training a base VLM, followed by training lightweight classification heads, called FS blocks, attached to different internal layers of the VLM and whose predictions are aggregated using an ensembling mechanism. Experiments show that I-FailSense outperforms state-of-the-art VLMs, both comparable in size and larger, in detecting semantic misalignment errors. Notably, despite being trained only on semantic misalignment detection, I-FailSense generalizes to broader robotic failure categories and effectively transfers to other simulation environments and real-world with zero-shot or minimal post-training. The datasets and models are publicly released on HuggingFace (Webpage: https://clemgris.github.io/I-FailSense/).

Paperid: 517, https://arxiv.org/pdf/2509.15968.pdf GitHub

Abstract:
Autonomous Driving (AD) systems have made notable progress, but their performance in long-tail, safety-critical scenarios remains limited. These rare cases contribute a disproportionate number of accidents. Vision-Language Action (VLA) models have strong reasoning abilities and offer a potential solution, but their effectiveness is limited by the lack of high-quality data and inefficient learning in such conditions. To address these challenges, we propose CoReVLA, a continual learning end-to-end autonomous driving framework that improves the performance in long-tail scenarios through a dual-stage process of data Collection and behavior Refinement. First, the model is jointly fine-tuned on a mixture of open-source driving QA datasets, allowing it to acquire a foundational understanding of driving scenarios. Next, CoReVLA is deployed within the Cave Automatic Virtual Environment (CAVE) simulation platform, where driver takeover data is collected from real-time interactions. Each takeover indicates a long-tail scenario that CoReVLA fails to handle reliably. Finally, the model is refined via Direct Preference Optimization (DPO), allowing it to learn directly from human preferences and thereby avoid reward hacking caused by manually designed rewards. Extensive open-loop and closed-loop experiments demonstrate that the proposed CoReVLA model can accurately perceive driving scenarios and make appropriate decisions. On the Bench2Drive benchmark, CoReVLA achieves a Driving Score (DS) of 72.18 and a Success Rate (SR) of 50%, outperforming state-of-the-art methods by 7.96 DS and 15% SR under long-tail, safety-critical scenarios. Furthermore, case studies demonstrate the model's ability to continually improve its performance in similar failure-prone scenarios by leveraging past takeover experiences. All codea and preprocessed datasets are available at: https://github.com/FanGShiYuu/CoReVLA

Paperid: 518, https://arxiv.org/pdf/2509.15490.pdf GitHub

Abstract:
Recent advances in vision-language models (VLMs) have enabled powerful multimodal reasoning, but state-of-the-art approaches typically rely on extremely large models with prohibitive computational and memory requirements. This makes their deployment challenging in resource-constrained environments such as warehouses, robotics, and industrial applications, where both efficiency and robust spatial understanding are critical. In this work, we present SmolRGPT, a compact vision-language architecture that explicitly incorporates region-level spatial reasoning by integrating both RGB and depth cues. SmolRGPT employs a three-stage curriculum that progressively align visual and language features, enables spatial relationship understanding, and adapts to task-specific datasets. We demonstrate that with only 600M parameters, SmolRGPT achieves competitive results on challenging warehouse spatial reasoning benchmarks, matching or exceeding the performance of much larger alternatives. These findings highlight the potential for efficient, deployable multimodal intelligence in real-world settings without sacrificing core spatial reasoning capabilities. The code of the experimentation will be available at: https://github.com/abtraore/SmolRGPT

Paperid: 519, https://arxiv.org/pdf/2509.15235.pdf GitHub

Abstract:
Speculative decoding is a widely adopted technique for accelerating inference in large language models (LLMs), yet its application to vision-language models (VLMs) remains underexplored, with existing methods achieving only modest speedups (<1.5x). This gap is increasingly significant as multimodal capabilities become central to large-scale models. We hypothesize that large VLMs can effectively filter redundant image information layer by layer without compromising textual comprehension, whereas smaller draft models struggle to do so. To address this, we introduce Vision-Aware Speculative Decoding (ViSpec), a novel framework tailored for VLMs. ViSpec employs a lightweight vision adaptor module to compress image tokens into a compact representation, which is seamlessly integrated into the draft model's attention mechanism while preserving original image positional information. Additionally, we extract a global feature vector for each input image and augment all subsequent text tokens with this feature to enhance multimodal coherence. To overcome the scarcity of multimodal datasets with long assistant responses, we curate a specialized training dataset by repurposing existing datasets and generating extended outputs using the target VLM with modified prompts. Our training strategy mitigates the risk of the draft model exploiting direct access to the target model's hidden states, which could otherwise lead to shortcut learning when training solely on target model outputs. Extensive experiments validate ViSpec, achieving, to our knowledge, the first substantial speedup in VLM speculative decoding. Code is available at https://github.com/KangJialiang/ViSpec.

Paperid: 520, https://arxiv.org/pdf/2509.15226.pdf GitHub

Abstract:
Medical Vision-Language Models (Med-VLMs) have demonstrated remarkable performance across diverse medical imaging tasks by leveraging large-scale image-text pretraining. However, their confidence calibration is largely unexplored, and so remains a significant challenge. As such, miscalibrated predictions can lead to overconfident errors, undermining clinical trust and decision-making reliability. To address this, we introduce CalibPrompt, the first framework to calibrate Med-VLMs during prompt tuning. CalibPrompt optimizes a small set of learnable prompts with carefully designed calibration objectives under scarce labeled data regime. First, we study a regularizer that attempts to align the smoothed accuracy with the predicted model confidences. Second, we introduce an angular separation loss to maximize textual feature proximity toward improving the reliability in confidence estimates of multimodal Med-VLMs. Extensive experiments on four publicly available Med-VLMs and five diverse medical imaging datasets reveal that CalibPrompt consistently improves calibration without drastically affecting clean accuracy. Our code is available at https://github.com/iabh1shekbasu/CalibPrompt.

Paperid: 521, https://arxiv.org/pdf/2509.15221.pdf GitHub

Abstract:
Vision-Language Models (VLMs) have enabled computer use agents (CUAs) that operate GUIs autonomously, showing great potential, yet progress is limited by the lack of large-scale, open-source computer use data and foundation models. In this work, we introduce ScaleCUA, a step toward scaling open-source CUAs. It offers a large-scale dataset spanning 6 operating systems and 3 task domains, built via a closed-loop pipeline uniting automated agents with human experts. Trained on this scaled-up data, ScaleCUA can operate seamlessly across platforms. Specifically, it delivers strong gains over baselines (+26.6 on WebArena-Lite-v2, +10.7 on ScreenSpot-Pro) and sets new state-of-the-art results (94.4% on MMBench-GUI L1-Hard, 60.6% on OSWorld-G, 47.4% on WebArena-Lite-v2). These findings underscore the power of data-driven scaling for general-purpose computer use agents. We will release data, models, and code to advance future research: https://github.com/OpenGVLab/ScaleCUA.

Paperid: 522, https://arxiv.org/pdf/2509.15212.pdf GitHub

Abstract:
This paper presents RynnVLA-001, a vision-language-action(VLA) model built upon large-scale video generative pretraining from human demonstrations. We propose a novel two-stage pretraining methodology. The first stage, Ego-Centric Video Generative Pretraining, trains an Image-to-Video model on 12M ego-centric manipulation videos to predict future frames conditioned on an initial frame and a language instruction. The second stage, Human-Centric Trajectory-Aware Modeling, extends this by jointly predicting future keypoint trajectories, thereby effectively bridging visual frame prediction with action prediction. Furthermore, to enhance action representation, we propose ActionVAE, a variational autoencoder that compresses sequences of actions into compact latent embeddings, reducing the complexity of the VLA output space. When finetuned on the same downstream robotics datasets, RynnVLA-001 achieves superior performance over state-of-the-art baselines, demonstrating that the proposed pretraining strategy provides a more effective initialization for VLA models.

Paperid: 523, https://arxiv.org/pdf/2509.15154.pdf GitHub

Abstract:
Ensuring factual consistency and reliable reasoning remains a critical challenge for medical vision-language models. We introduce MEDFACT-R1, a two-stage framework that integrates external knowledge grounding with reinforcement learning to improve the factual medical reasoning. The first stage uses pseudo-label supervised fine-tuning (SFT) to incorporate external factual expertise; while the second stage applies Group Relative Policy Optimization (GRPO) with four tailored factual reward signals to encourage self-consistent reasoning. Across three public medical QA benchmarks, MEDFACT-R1 delivers up to 22.5% absolute improvement in factual accuracy over previous state-of-the-art methods. Ablation studies highlight the necessity of pseudo-label SFT cold start and validate the contribution of each GRPO reward, underscoring the synergy between knowledge grounding and RL-driven reasoning for trustworthy medical AI. Codes are released at https://github.com/Garfieldgengliang/MEDFACT-R1.

Paperid: 524, https://arxiv.org/pdf/2509.14977.pdf GitHub

Abstract:
Ultrasound imaging has become the preferred imaging modality for early cancer screening due to its advantages of non-ionizing radiation, low cost, and real-time imaging capabilities. However, conventional ultrasound diagnosis heavily relies on physician expertise, presenting challenges of high subjectivity and low diagnostic efficiency. Vision-language models (VLMs) offer promising solutions for this issue, but existing general-purpose models demonstrate limited knowledge in ultrasound medical tasks, with poor generalization in multi-organ lesion recognition and low efficiency across multi-task diagnostics. To address these limitations, we propose EchoVLM, a vision-language model specifically designed for ultrasound medical imaging. The model employs a Mixture of Experts (MoE) architecture trained on data spanning seven anatomical regions. This design enables the model to perform multiple tasks, including ultrasound report generation, diagnosis and visual question-answering (VQA). The experimental results demonstrated that EchoVLM achieved significant improvements of 10.15 and 4.77 points in BLEU-1 scores and ROUGE-1 scores respectively compared to Qwen2-VL on the ultrasound report generation task. These findings suggest that EchoVLM has substantial potential to enhance diagnostic accuracy in ultrasound imaging, thereby providing a viable technical solution for future clinical applications. Source code and model weights are available at https://github.com/Asunatan/EchoVLM.

Paperid: 525, https://arxiv.org/pdf/2509.14837.pdf GitHub

Abstract:
Recent advances in causal interpretability have extended from language models to vision-language models (VLMs), seeking to reveal their internal mechanisms through input interventions. While textual interventions often target semantics, visual interventions typically rely on coarse pixel-level perturbations, limiting semantic insights on multimodal integration. In this study, we introduce V-SEAM, a novel framework that combines Visual Semantic Editing and Attention Modulating for causal interpretation of VLMs. V-SEAM enables concept-level visual manipulations and identifies attention heads with positive or negative contributions to predictions across three semantic levels: objects, attributes, and relationships. We observe that positive heads are often shared within the same semantic level but vary across levels, while negative heads tend to generalize broadly. Finally, we introduce an automatic method to modulate key head embeddings, demonstrating enhanced performance for both LLaVA and InstructBLIP across three diverse VQA benchmarks. Our data and code are released at: https://github.com/petergit1/V-SEAM.

Paperid: 526, https://arxiv.org/pdf/2509.14738.pdf GitHub

Abstract:
Unified vision large language models (VLLMs) have recently achieved impressive advancements in both multimodal understanding and generation, powering applications such as visual question answering and text-guided image synthesis. However, progress in unified VLLMs remains constrained by the lack of datasets that fully exploit the synergistic potential between these two core abilities. Existing datasets typically address understanding and generation in isolation, thereby limiting the performance of unified VLLMs. To bridge this critical gap, we introduce a novel dataset construction framework, UnifiedVisual, and present UnifiedVisual-240K, a high-quality dataset meticulously designed to facilitate mutual enhancement between multimodal understanding and generation. UnifiedVisual-240K seamlessly integrates diverse visual and textual inputs and outputs, enabling comprehensive cross-modal reasoning and precise text-to-image alignment. Our dataset encompasses a wide spectrum of tasks and data sources, ensuring rich diversity and addressing key shortcomings of prior resources. Extensive experiments demonstrate that models trained on UnifiedVisual-240K consistently achieve strong performance across a wide range of tasks. Notably, these models exhibit significant mutual reinforcement between multimodal understanding and generation, further validating the effectiveness of our framework and dataset. We believe UnifiedVisual represents a new growth point for advancing unified VLLMs and unlocking their full potential. Our code and datasets is available at https://github.com/fnlp-vision/UnifiedVisual.

Paperid: 527, https://arxiv.org/pdf/2509.14735.pdf GitHub

Abstract:
Multimodal large language models (MLLMs) have gained significant attention due to their impressive ability to integrate vision and language modalities. Recent advancements in MLLMs have primarily focused on improving performance through high-quality datasets, novel architectures, and optimized training strategies. However, in this paper, we identify a previously overlooked issue, language prior conflict, a mismatch between the inherent language priors of large language models (LLMs) and the language priors in training datasets. This conflict leads to suboptimal vision-language alignment, as MLLMs are prone to adapting to the language style of training samples. To address this issue, we propose a novel training method called Decoupled Proxy Alignment (DPA). DPA introduces two key innovations: (1) the use of a proxy LLM during pretraining to decouple the vision-language alignment process from language prior interference, and (2) dynamic loss adjustment based on visual relevance to strengthen optimization signals for visually relevant tokens. Extensive experiments demonstrate that DPA significantly mitigates the language prior conflict, achieving superior alignment performance across diverse datasets, model families, and scales. Our method not only improves the effectiveness of MLLM training but also shows exceptional generalization capabilities, making it a robust approach for vision-language alignment. Our code is available at https://github.com/fnlp-vision/DPA.

Paperid: 528, https://arxiv.org/pdf/2509.14687.pdf GitHub

Abstract:
The emerging field of Vision-Language-Action (VLA) for humanoid robots faces several fundamental challenges, including the high cost of data acquisition, the lack of a standardized benchmark, and the significant gap between simulation and the real world. To overcome these obstacles, we propose RealMirror, a comprehensive, open-source embodied AI VLA platform. RealMirror builds an efficient, low-cost data collection, model training, and inference system that enables end-to-end VLA research without requiring a real robot. To facilitate model evolution and fair comparison, we also introduce a dedicated VLA benchmark for humanoid robots, featuring multiple scenarios, extensive trajectories, and various VLA models. Furthermore, by integrating generative models and 3D Gaussian Splatting to reconstruct realistic environments and robot models, we successfully demonstrate zero-shot Sim2Real transfer, where models trained exclusively on simulation data can perform tasks on a real robot seamlessly, without any fine-tuning. In conclusion, with the unification of these critical components, RealMirror provides a robust framework that significantly accelerates the development of VLA models for humanoid robots. Project page: https://terminators2025.github.io/RealMirror.github.io

Paperid: 529, https://arxiv.org/pdf/2509.14630.pdf GitHub

Abstract:
Vision-language-action policies learn manipulation skills across tasks, environments and embodiments through large-scale pre-training. However, their ability to generalize to novel robot configurations remains limited. Most approaches emphasize model size, dataset scale and diversity while paying less attention to the design of action spaces. This leads to the configuration generalization problem, which requires costly adaptation. We address this challenge by formulating cross-embodiment pre-training as designing policies equivariant to embodiment configuration transformations. Building on this principle, we propose a framework that (i) establishes a embodiment equivariance theory for action space and policy design, (ii) introduces an action decoder that enforces configuration equivariance, and (iii) incorporates a geometry-aware network architecture to enhance embodiment-agnostic spatial reasoning. Extensive experiments in both simulation and real-world settings demonstrate that our approach improves pre-training effectiveness and enables efficient fine-tuning on novel robot embodiments. Our code is available at https://github.com/hhcaz/e2vla

Paperid: 530, https://arxiv.org/pdf/2509.14084.pdf GitHub

Abstract:
Zero-Shot Anomaly Detection (ZSAD) seeks to identify anomalies from arbitrary novel categories, offering a scalable and annotation-efficient solution. Traditionally, most ZSAD works have been based on the CLIP model, which performs anomaly detection by calculating the similarity between visual and text embeddings. Recently, vision foundation models such as DINOv3 have demonstrated strong transferable representation capabilities. In this work, we are the first to adapt DINOv3 for ZSAD. However, this adaptation presents two key challenges: (i) the domain bias between large-scale pretraining data and anomaly detection tasks leads to feature misalignment; and (ii) the inherent bias toward global semantics in pretrained representations often leads to subtle anomalies being misinterpreted as part of the normal foreground objects, rather than being distinguished as abnormal regions. To overcome these challenges, we introduce AD-DINOv3, a novel vision-language multimodal framework designed for ZSAD. Specifically, we formulate anomaly detection as a multimodal contrastive learning problem, where DINOv3 is employed as the visual backbone to extract patch tokens and a CLS token, and the CLIP text encoder provides embeddings for both normal and abnormal prompts. To bridge the domain gap, lightweight adapters are introduced in both modalities, enabling their representations to be recalibrated for the anomaly detection task. Beyond this baseline alignment, we further design an Anomaly-Aware Calibration Module (AACM), which explicitly guides the CLS token to attend to anomalous regions rather than generic foreground semantics, thereby enhancing discriminability. Extensive experiments on eight industrial and medical benchmarks demonstrate that AD-DINOv3 consistently matches or surpasses state-of-the-art methods.The code will be available at https://github.com/Kaisor-Yuan/AD-DINOv3.

Paperid: 531, https://arxiv.org/pdf/2509.13858.pdf GitHub

Abstract:
Dataset distillation aims to synthesize a compact dataset from the original large-scale one, enabling highly efficient learning while preserving competitive model performance. However, traditional techniques primarily capture low-level visual features, neglecting the high-level semantic and structural information inherent in images. In this paper, we propose EDITS, a novel framework that exploits the implicit textual semantics within the image data to achieve enhanced distillation. First, external texts generated by a Vision Language Model (VLM) are fused with image features through a Global Semantic Query module, forming the prior clustered buffer. Local Semantic Awareness then selects representative samples from the buffer to construct image and text prototypes, with the latter produced by guiding a Large Language Model (LLM) with meticulously crafted prompt. Ultimately, Dual Prototype Guidance strategy generates the final synthetic dataset through a diffusion model. Extensive experiments confirm the effectiveness of our method.Source code is available in: https://github.com/einsteinxia/EDITS.

Paperid: 532, https://arxiv.org/pdf/2509.13760.pdf GitHub

Abstract:
Text-to-Image (T2I) models have made remarkable progress in generating images from text prompts, but their output quality and safety still depend heavily on how prompts are phrased. Existing safety methods typically refine prompts using large language models (LLMs), but they overlook the images produced, which can result in unsafe outputs or unnecessary changes to already safe prompts. To address this, we propose an iterative prompt refinement algorithm that uses Vision Language Models (VLMs) to analyze both the input prompts and the generated images. By leveraging visual feedback, our method refines prompts more effectively, improving safety while maintaining user intent and reliability comparable to existing LLM-based approaches. Additionally, we introduce a new dataset labeled with both textual and visual safety signals using off-the-shelf multi-modal LLM, enabling supervised fine-tuning. Experimental results demonstrate that our approach produces safer outputs without compromising alignment with user intent, offering a practical solution for generating safer T2I content. Our code is available at https://github.com/ku-dmlab/IPR. \textbf{\textcolor{red}WARNING: This paper contains examples of harmful or inappropriate images generated by models.

Paperid: 533, https://arxiv.org/pdf/2509.13399.pdf GitHub

Abstract:
Instruction-based image editing has advanced rapidly, yet reliable and interpretable evaluation remains a bottleneck. Current protocols either (i) depend on paired reference images -- resulting in limited coverage and inheriting biases from prior generative models -- or (ii) rely solely on zero-shot vision-language models (VLMs), whose prompt-based assessments of instruction following, content consistency, and visual quality are often imprecise. To address this, we introduce EdiVal-Agent, an automated, scalable, and fine-grained evaluation framework for multi-turn instruction-based editing from an object-centric perspective, supported by a suite of expert tools. Given an image, EdiVal-Agent first decomposes it into semantically meaningful objects, then synthesizes diverse, context-aware editing instructions. For evaluation, it integrates VLMs with open-vocabulary object detectors to assess instruction following, uses semantic-level feature extractors to evaluate content consistency, and leverages human preference models to judge visual quality. We show that combining VLMs with object detectors yields stronger agreement with human judgments in instruction-following evaluation compared to using VLMs alone and CLIP-based metrics. Furthermore, the pipeline's modular design allows future tools to be seamlessly integrated, enhancing evaluation accuracy over time. Instantiating this pipeline, we build EdiVal-Bench, a multi-turn editing benchmark covering 9 instruction types and 11 state-of-the-art editing models spanning autoregressive (AR) (including Nano Banana, GPT-Image-1), flow-matching, and diffusion paradigms. We demonstrate that EdiVal-Agent can be used to identify existing failure modes, thereby informing the development of the next generation of editing models. Project page: https://tianyucodings.github.io/EdiVAL-page/.

Paperid: 534, https://arxiv.org/pdf/2509.13347.pdf GitHub

Abstract:
The choice of action spaces is a critical yet unresolved challenge in developing capable, end-to-end trainable agents. This paper first presents a large-scale, systematic comparison of prominent abstracted action spaces and tokenizers for Vision-Language-Action (VLA) or hierarchical agent models in the open-ended Minecraft. Our analysis reveals that no single action space is universally optimal; instead, the most effective abstraction is highly task-dependent, creating a dilemma for building generalist agents. To resolve this, we introduce Chain of Action (CoA), a novel framework that unifies high-level planning and low-level control within a single, monolithic VLA model. CoA treats an abstracted action not as a command for a separate policy, but as an intermediate reasoning step--akin to a chain of thought--that guides the generation of the final, executable action. Furthermore, we demonstrate that an All-in-One agent trained on a diverse mixture of action spaces using the CoA paradigm learns a more robust and generalizable policy. This unified agent achieves a new state-of-the-art, improving the overall task success rate over strong, specialized baselines. To foster reproducible research, we release the OpenHA (Open Hierarchical Agents) suite, which includes our comprehensive benchmark of over 800 distinct tasks, curated datasets, source code, and all pretrained model checkpoints at https://github.com/CraftJarvis/OpenHA

Paperid: 535, https://arxiv.org/pdf/2509.13175.pdf GitHub

Abstract:
The emergence of Large Language Models (LLMs) presents unprecedented opportunities to revolutionize medical contrastive vision-language pre-training. In this paper, we show how LLMs can facilitate large-scale supervised pre-training, thereby advancing vision-language alignment. We begin by demonstrate that modern LLMs can automatically extract diagnostic labels from radiology reports with remarkable precision (>96\% AUC in our experiments) without complex prompt engineering, enabling the creation of large-scale "silver-standard" datasets at a minimal cost (~\$3 for 50k CT image-report pairs). Further, we find that vision encoder trained on this "silver-standard" dataset achieves performance comparable to those trained on labels extracted by specialized BERT-based models, thereby democratizing the access to large-scale supervised pre-training. Building on this foundation, we proceed to reveal that supervised pre-training fundamentally improves contrastive vision-language alignment. Our approach achieves state-of-the-art performance using only a 3D ResNet-18 with vanilla CLIP training, including 83.8\% AUC for zero-shot diagnosis on CT-RATE, 77.3\% AUC on RAD-ChestCT, and substantial improvements in cross-modal retrieval (MAP@50=53.7\% for image-image, Recall@100=52.2\% for report-image). These results demonstrate the potential of utilizing LLMs to facilitate {\bf more performant and scalable} medical AI systems. Our code is avaiable at https://github.com/SadVoxel/More-performant-and-scalable.

Paperid: 536, https://arxiv.org/pdf/2509.12594.pdf GitHub

Abstract:
We present LightVLA, a simple yet effective differentiable token pruning framework for vision-language-action (VLA) models. While VLA models have shown impressive capability in executing real-world robotic tasks, their deployment on resource-constrained platforms is often bottlenecked by the heavy attention-based computation over large sets of visual tokens. LightVLA addresses this challenge through adaptive, performance-driven pruning of visual tokens: It generates dynamic queries to evaluate visual token importance, and adopts Gumbel softmax to enable differentiable token selection. Through fine-tuning, LightVLA learns to preserve the most informative visual tokens while pruning tokens which do not contribute to task execution, thereby improving efficiency and performance simultaneously. Notably, LightVLA requires no heuristic magic numbers and introduces no additional trainable parameters, making it compatible with modern inference frameworks. Experimental results demonstrate that LightVLA outperforms different VLA models and existing token pruning methods across diverse tasks on the LIBERO benchmark, achieving higher success rates with substantially reduced computational overhead. Specifically, LightVLA reduces FLOPs and latency by 59.1% and 38.2% respectively, with a 2.6% improvement in task success rate. Meanwhile, we also investigate the learnable query-based token pruning method LightVLA* with additional trainable parameters, which also achieves satisfactory performance. Our work reveals that as VLA pursues optimal performance, LightVLA spontaneously learns to prune tokens from a performance-driven perspective. To the best of our knowledge, LightVLA is the first work to apply adaptive visual token pruning to VLA tasks with the collateral goals of efficiency and performance, marking a significant step toward more efficient, powerful and practical real-time robotic systems.

Paperid: 537, https://arxiv.org/pdf/2509.12129.pdf GitHub

Abstract:
Navigation is a fundamental capability in embodied AI, representing the intelligence required to perceive and interact within physical environments following language instructions. Despite significant progress in large Vision-Language Models (VLMs), which exhibit remarkable zero-shot performance on general vision-language tasks, their generalization ability in embodied navigation remains largely confined to narrow task settings and embodiment-specific architectures. In this work, we introduce a cross-embodiment and cross-task Navigation Foundation Model (NavFoM), trained on eight million navigation samples that encompass quadrupeds, drones, wheeled robots, and vehicles, and spanning diverse tasks such as vision-and-language navigation, object searching, target tracking, and autonomous driving. NavFoM employs a unified architecture that processes multimodal navigation inputs from varying camera configurations and navigation horizons. To accommodate diverse camera setups and temporal horizons, NavFoM incorporates identifier tokens that embed camera view information of embodiments and the temporal context of tasks. Furthermore, to meet the demands of real-world deployment, NavFoM controls all observation tokens using a dynamically adjusted sampling strategy under a limited token length budget. Extensive evaluations on public benchmarks demonstrate that our model achieves state-of-the-art or highly competitive performance across multiple navigation tasks and embodiments without requiring task-specific fine-tuning. Additional real-world experiments further confirm the strong generalization capability and practical applicability of our approach.

Paperid: 538, https://arxiv.org/pdf/2509.12040.pdf GitHub GitHub

Abstract:
Open-Vocabulary Remote Sensing Image Segmentation (OVRSIS), an emerging task that adapts Open-Vocabulary Segmentation (OVS) to the remote sensing (RS) domain, remains underexplored due to the absence of a unified evaluation benchmark and the domain gap between natural and RS images. To bridge these gaps, we first establish a standardized OVRSIS benchmark (\textbf{OVRSISBench}) based on widely-used RS segmentation datasets, enabling consistent evaluation across methods. Using this benchmark, we comprehensively evaluate several representative OVS/OVRSIS models and reveal their limitations when directly applied to remote sensing scenarios. Building on these insights, we propose \textbf{RSKT-Seg}, a novel open-vocabulary segmentation framework tailored for remote sensing. RSKT-Seg integrates three key components: (1) a Multi-Directional Cost Map Aggregation (RS-CMA) module that captures rotation-invariant visual cues by computing vision-language cosine similarities across multiple directions; (2) an Efficient Cost Map Fusion (RS-Fusion) transformer, which jointly models spatial and semantic dependencies with a lightweight dimensionality reduction strategy; and (3) a Remote Sensing Knowledge Transfer (RS-Transfer) module that injects pre-trained knowledge and facilitates domain adaptation via enhanced upsampling. Extensive experiments on the benchmark show that RSKT-Seg consistently outperforms strong OVS baselines by +3.8 mIoU and +5.9 mACC, while achieving 2x faster inference through efficient aggregation. Our code is \href{https://github.com/LiBingyu01/RSKT-Seg}{\textcolor{blue}{here}}.

Paperid: 539, https://arxiv.org/pdf/2509.11839.pdf GitHub GitHub

Abstract:
Recent Vision-Language-Action models show potential to generalize across embodiments but struggle to quickly align with a new robot's action space when high-quality demonstrations are scarce, especially for bipedal humanoids. We present TrajBooster, a cross-embodiment framework that leverages abundant wheeled-humanoid data to boost bipedal VLA. Our key idea is to use end-effector trajectories as a morphology-agnostic interface. TrajBooster (i) extracts 6D dual-arm end-effector trajectories from real-world wheeled humanoids, (ii) retargets them in simulation to Unitree G1 with a whole-body controller trained via a heuristic-enhanced harmonized online DAgger to lift low-dimensional trajectory references into feasible high-dimensional whole-body actions, and (iii) forms heterogeneous triplets that couple source vision/language with target humanoid-compatible actions to post-pre-train a VLA, followed by only 10 minutes of teleoperation data collection on the target humanoid domain. Deployed on Unitree G1, our policy achieves beyond-tabletop household tasks, enabling squatting, cross-height manipulation, and coordinated whole-body motion with markedly improved robustness and generalization. Results show that TrajBooster allows existing wheeled-humanoid data to efficiently strengthen bipedal humanoid VLA performance, reducing reliance on costly same-embodiment data while enhancing action space understanding and zero-shot skill transfer capabilities. For more details, For more details, please refer to our \href{https://jiachengliu3.github.io/TrajBooster/}.

Paperid: 540, https://arxiv.org/pdf/2509.11815.pdf GitHub

Abstract:
Speculative decoding is a powerful way to accelerate autoregressive large language models (LLMs), but directly porting it to vision-language models (VLMs) faces unique systems constraints: the prefill stage is dominated by visual tokens whose count scales with image resolution and video length, inflating both compute and memory, especially the key-value (KV) cache. We study speculative decoding for VLMs and introduce SpecVLM, a practical system that (1) establishes a strong EAGLE-2-style baseline, EagleVLM, delivering 1.5--2.3x end-to-end speedups over full autoregressive inference, and (2) further accelerates VLM inference with an elastic visual compressor that adaptively selects among pruning, pooling, convolution, and resampler primitives to balance FLOPs/parameters and accuracy per input. To avoid costly offline distillation corpora, we propose an online-logit distillation protocol that trains the draft model with on-the-fly teacher logits and penultimate features using a combined cross-entropy and Smooth L1 objective, eliminating storage and preprocessing while remaining compute-efficient. This protocol reveals a training-time scaling effect: longer online training monotonically increases the draft model's average accepted length, improving speculative efficiency. Empirically, SpecVLM achieves additional acceleration, culminating in 2.5--2.9x end-to-end speedups within 5 epochs across LLaVA and MMMU, consistently over resolutions and task difficulties, while preserving the target model's output distribution (lossless decoding). Our code is available at https://github.com/haiduo/SpecVLM.

Paperid: 541, https://arxiv.org/pdf/2509.11287.pdf GitHub

Abstract:
Large Vision-Language Models (LVLMs) suffer from serious hallucination problems, where the model-generated responses are inconsistent with the visual inputs. Existing hallucination mitigation methods are mainly based on preference alignment and require external human annotations or auxiliary models for preference data collection, which increase costs and limit sustainable improvement. To tackle these challenges, we propose Autonomous Preference Alignment via Self-Injection (APASI), a novel and generalizable method that mitigates hallucinations without external dependencies. APASI leverages the target LVLM to self-inject hallucinations into a generated response, creating a pair of responses with varying preference levels. During the self-injection process, the dis-preferred response is generated based on three key observations of hallucinations, ensuring it simulates real hallucination patterns. This fidelity offers an accurate learning signal for hallucination mitigation. Moreover, APASI incorporates an iterative alignment training strategy combined with curriculum learning to periodically update the preference data with increasing challenge, enabling stable and continuous enhancement of the LVLM. Extensive experiments across six benchmarks show that APASI not only effectively mitigates hallucinations for three baseline models but also achieves comparable or even superior performance to alignment-based methods with external dependency, thereby demonstrating its effectiveness and generalization capability. The code is available at https://github.com/davidluciolu/APASI.

Paperid: 542, https://arxiv.org/pdf/2509.10416.pdf GitHub

Abstract:
We present TASC, a Task-Aware Shared Control framework for teleoperated manipulation that infers task-level user intent and provides assistance throughout the task. To support everyday tasks without predefined knowledge, TASC constructs an open-vocabulary interaction graph from visual input to represent functional object relationships, and infers user intent accordingly. A shared control policy then provides rotation assistance during both grasping and object interaction, guided by spatial constraints predicted by a vision-language model. Our method addresses two key challenges in general-purpose, long-horizon shared control: (1) understanding and inferring task-level user intent, and (2) generalizing assistance across diverse objects and tasks. Experiments in both simulation and the real world demonstrate that TASC improves task efficiency and reduces user input effort compared to prior methods. To the best of our knowledge, this is the first shared control framework that supports everyday manipulation tasks with zero-shot generalization. The code that supports our experiments is publicly available at https://github.com/fitz0401/tasc.

Paperid: 543, https://arxiv.org/pdf/2509.10260.pdf GitHub

Abstract:
Text-to-image (T2I) generation has achieved remarkable progress in instruction following and aesthetics. However, a persistent challenge is the prevalence of physical artifacts, such as anatomical and structural flaws, which severely degrade perceptual quality and limit application. Given the diversity and complexity of these artifacts, a systematic and fine-grained evaluation framework is required, which is lacking in current benchmarks. To fill this gap, we introduce MagicMirror, a comprehensive framework for artifacts assessment. We first establish a detailed taxonomy of generated image artifacts. Guided by this taxonomy, we manually annotate MagicData340K, the first human-annotated large-scale dataset of 340K generated images with fine-grained artifact labels. Building on this dataset, we train MagicAssessor, a Vision-Language Model (VLM) that provides detailed assessments and corresponding labels. To overcome challenges like class imbalance and reward hacking, we design a novel data sampling strategy and a multi-level reward system for Group Relative Policy Optimization (GRPO). Finally, we leverage MagicAssessor to construct MagicBench, an automated benchmark for evaluating the image artifacts of current T2I models. Our evaluation with MagicBench reveals that despite their widespread adoption, even top-tier models like GPT-image-1 are consistently plagued by significant artifacts, highlighting artifact reduction as a critical frontier for future T2I development. Project page: https://wj-inf.github.io/MagicMirror-page/.

Paperid: 544, https://arxiv.org/pdf/2509.10059.pdf GitHub

Abstract:
Mathematical reasoning is critical for tasks such as precise distance and area computations, trajectory estimations, and spatial analysis in unmanned aerial vehicle (UAV) based remote sensing, yet current vision-language models (VLMs) have not been adequately tested in this domain. To address this gap, we introduce AVI-Math, the first benchmark to rigorously evaluate multimodal mathematical reasoning in aerial vehicle imagery, moving beyond simple counting tasks to include domain-specific knowledge in areas such as geometry, logic, and algebra. The dataset comprises 3,773 high-quality vehicle-related questions captured from UAV views, covering 6 mathematical subjects and 20 topics. The data, collected at varying altitudes and from multiple UAV angles, reflects real-world UAV scenarios, ensuring the diversity and complexity of the constructed mathematical problems. In this paper, we benchmark 14 prominent VLMs through a comprehensive evaluation and demonstrate that, despite their success on previous multimodal benchmarks, these models struggle with the reasoning tasks in AVI-Math. Our detailed analysis highlights significant limitations in the mathematical reasoning capabilities of current VLMs and suggests avenues for future research. Furthermore, we explore the use of Chain-of-Thought prompting and fine-tuning techniques, which show promise in addressing the reasoning challenges in AVI-Math. Our findings not only expose the limitations of VLMs in mathematical reasoning but also offer valuable insights for advancing UAV-based trustworthy VLMs in real-world applications. The code, and datasets will be released at https://github.com/VisionXLab/avi-math

Paperid: 545, https://arxiv.org/pdf/2509.10026.pdf GitHub GitHub

Abstract:
As large vision language models (VLMs) advance, their capabilities in multilingual visual question answering (mVQA) have significantly improved. Chain-of-thought (CoT) reasoning has been proven to enhance interpretability and complex reasoning. However, most existing approaches rely primarily on textual CoT and provide limited support for multilingual multimodal reasoning, constraining their deployment in real-world applications. To address this gap, we introduce \textbf{LaV-CoT}, the first Language-aware Visual CoT framework with Multi-Aspect Reward Optimization. LaV-CoT incorporates an interpretable multi-stage reasoning pipeline consisting of Text Summary with Bounding Box (BBox), Language Identification, Spatial Object-level Captioning, and Step-by-step Logical Reasoning. Following this reasoning pipeline, we design an automated data curation method that generates multilingual CoT annotations through iterative generation, correction, and refinement, enabling scalable and high-quality training data. To improve reasoning and generalization, LaV-CoT adopts a two-stage training paradigm combining Supervised Fine-Tuning (SFT) with Language-aware Group Relative Policy Optimization (GRPO), guided by verifiable multi-aspect rewards including language consistency, structural accuracy, and semantic alignment. Extensive evaluations on public datasets including MMMB, Multilingual MMBench, and MTVQA show that LaV-CoT achieves up to ~9.5% accuracy improvements over open-source baselines of similar size and even surpasses models with 2$\times$ larger scales by ~2.6%. Moreover, LaV-CoT outperforms advanced proprietary models such as GPT-4o-0513 and Gemini-2.5-flash. We further conducted an online A/B test to validate our method on real-world data, highlighting its effectiveness for industrial deployment. Our code is available at this link: \href{https://github.com/HJNVR/LaV-CoT}

Paperid: 546, https://arxiv.org/pdf/2509.09674.pdf GitHub

Abstract:
Vision-Language-Action (VLA) models have recently emerged as a powerful paradigm for robotic manipulation. Despite substantial progress enabled by large-scale pretraining and supervised fine-tuning (SFT), these models face two fundamental challenges: (i) the scarcity and high cost of large-scale human-operated robotic trajectories required for SFT scaling, and (ii) limited generalization to tasks involving distribution shift. Recent breakthroughs in Large Reasoning Models (LRMs) demonstrate that reinforcement learning (RL) can dramatically enhance step-by-step reasoning capabilities, raising a natural question: Can RL similarly improve the long-horizon step-by-step action planning of VLA? In this work, we introduce SimpleVLA-RL, an efficient RL framework tailored for VLA models. Building upon veRL, we introduce VLA-specific trajectory sampling, scalable parallelization, multi-environment rendering, and optimized loss computation. When applied to OpenVLA-OFT, SimpleVLA-RL achieves SoTA performance on LIBERO and even outperforms $Ï_0$ on RoboTwin 1.0\&2.0 with the exploration-enhancing strategies we introduce. SimpleVLA-RL not only reduces dependence on large-scale data and enables robust generalization, but also remarkably surpasses SFT in real-world tasks. Moreover, we identify a novel phenomenon ``pushcut'' during RL training, wherein the policy discovers previously unseen patterns beyond those seen in the previous training process. Github: https://github.com/PRIME-RL/SimpleVLA-RL

Paperid: 547, https://arxiv.org/pdf/2509.09397.pdf GitHub

Abstract:
Medical vision-language models (VLMs) offer promise for clinical decision support, yet their reliability under distribution shifts remains a major concern for safe deployment. These models often learn task-agnostic correlations due to variability in imaging protocols and free-text reports, limiting their generalizability and increasing the risk of failure in real-world settings. We propose DRiFt, a structured feature decoupling framework that explicitly separates clinically relevant signals from task-agnostic noise using parameter-efficient tuning (LoRA) and learnable prompt tokens. To enhance cross-modal alignment and reduce uncertainty, we curate high-quality, clinically grounded image-text pairs by generating captions for a diverse medical dataset. Our approach improves in-distribution performance by +11.4% Top-1 accuracy and +3.3% Macro-F1 over prior prompt-based methods, while maintaining strong robustness across unseen datasets. Ablation studies reveal that disentangling task-relevant features and careful alignment significantly enhance model generalization and reduce unpredictable behavior under domain shift. These insights contribute toward building safer, more trustworthy VLMs for clinical use. The code is available at https://github.com/rumaima/DRiFt.

Paperid: 548, https://arxiv.org/pdf/2509.09372.pdf GitHub GitHub GitHub

Abstract:
Vision-Language-Action (VLA) models typically bridge the gap between perceptual and action spaces by pre-training a large-scale Vision-Language Model (VLM) on robotic data. While this approach greatly enhances performance, it also incurs significant training costs. In this paper, we investigate how to effectively bridge vision-language (VL) representations to action (A). We introduce VLA-Adapter, a novel paradigm designed to reduce the reliance of VLA models on large-scale VLMs and extensive pre-training. To this end, we first systematically analyze the effectiveness of various VL conditions and present key findings on which conditions are essential for bridging perception and action spaces. Based on these insights, we propose a lightweight Policy module with Bridge Attention, which autonomously injects the optimal condition into the action space. In this way, our method achieves high performance using only a 0.5B-parameter backbone, without any robotic data pre-training. Extensive experiments on both simulated and real-world robotic benchmarks demonstrate that VLA-Adapter not only achieves state-of-the-art level performance, but also offers the fast inference speed reported to date. Furthermore, thanks to the proposed advanced bridging paradigm, VLA-Adapter enables the training of a powerful VLA model in just 8 hours on a single consumer-grade GPU, greatly lowering the barrier to deploying the VLA model. Project page: https://vla-adapter.github.io/.

Paperid: 549, https://arxiv.org/pdf/2509.09311.pdf GitHub

Abstract:
Vision-language models (VLMs) have enabled strong zero-shot classification through image-text alignment. Yet, their purely visual inference capabilities remain under-explored. In this work, we conduct a comprehensive evaluation of both language-guided and vision-only image classification with a diverse set of dual-encoder VLMs, including both well-established and recent models such as SigLIP 2 and RADIOv2.5. The performance is compared in a standard setup on the ImageNet-1k validation set and its label-corrected variant. The key factors affecting accuracy are analysed, including prompt design, class diversity, the number of neighbours in k-NN, and reference set size. We show that language and vision offer complementary strengths, with some classes favouring textual prompts and others better handled by visual similarity. To exploit this complementarity, we introduce a simple, learning-free fusion method based on per-class precision that improves classification performance. The code is available at: https://github.com/gonikisgo/bmvc2025-vlm-image-recognition.

Paperid: 550, https://arxiv.org/pdf/2509.09254.pdf GitHub

Abstract:
Recent advances in large vision-language models (LVLMs) have demonstrated strong performance on general-purpose medical tasks. However, their effectiveness in specialized domains such as dentistry remains underexplored. In particular, panoramic X-rays, a widely used imaging modality in oral radiology, pose interpretative challenges due to dense anatomical structures and subtle pathological cues, which are not captured by existing medical benchmarks or instruction datasets. To this end, we introduce MMOral, the first large-scale multimodal instruction dataset and benchmark tailored for panoramic X-ray interpretation. MMOral consists of 20,563 annotated images paired with 1.3 million instruction-following instances across diverse task types, including attribute extraction, report generation, visual question answering, and image-grounded dialogue. In addition, we present MMOral-Bench, a comprehensive evaluation suite covering five key diagnostic dimensions in dentistry. We evaluate 64 LVLMs on MMOral-Bench and find that even the best-performing model, i.e., GPT-4o, only achieves 41.45% accuracy, revealing significant limitations of current models in this domain. To promote the progress of this specific domain, we also propose OralGPT, which conducts supervised fine-tuning (SFT) upon Qwen2.5-VL-7B with our meticulously curated MMOral instruction dataset. Remarkably, a single epoch of SFT yields substantial performance enhancements for LVLMs, e.g., OralGPT demonstrates a 24.73% improvement. Both MMOral and OralGPT hold significant potential as a critical foundation for intelligent dentistry and enable more clinically impactful multimodal AI systems in the dental field. The dataset, model, benchmark, and evaluation suite are available at https://github.com/isbrycee/OralGPT.

Paperid: 551, https://arxiv.org/pdf/2509.09116.pdf GitHub

Abstract:
Foundation segmentation models achieve reasonable leaf instance extraction from top-view crop images without training (i.e., zero-shot). However, segmenting entire plant individuals with each consisting of multiple overlapping leaves remains challenging. This problem is referred to as a hierarchical segmentation task, typically requiring annotated training datasets, which are often species-specific and require notable human labor. To address this, we introduce ZeroPlantSeg, a zero-shot segmentation for rosette-shaped plant individuals from top-view images. We integrate a foundation segmentation model, extracting leaf instances, and a vision-language model, reasoning about plants' structures to extract plant individuals without additional training. Evaluations on datasets with multiple plant species, growth stages, and shooting environments demonstrate that our method surpasses existing zero-shot methods and achieves better cross-domain performance than supervised methods. Implementations are available at https://github.com/JunhaoXing/ZeroPlantSeg.

Paperid: 552, https://arxiv.org/pdf/2509.09014.pdf GitHub

Abstract:
Urdu, spoken by over 250 million people, remains critically under-served in multimodal and vision-language research. The absence of large-scale, high-quality datasets has limited the development of Urdu-capable systems and reinforced biases in multilingual vision-language models trained primarily on high-resource languages. To address this gap, we present COCO-Urdu, a large-scale image-caption dataset derived from MS COCO, containing 59,000 images and 319,000 Urdu captions selected through stratified sampling to preserve the original distribution. Captions were translated using SeamlessM4T v2 and validated with a hybrid multimodal quality estimation framework that integrates COMET-Kiwi for translation quality, CLIP-based similarity for visual grounding, and BERTScore with back-translation for semantic consistency; low-scoring captions were iteratively refined using open-source large language models. We further benchmark COCO-Urdu on BLEU, SacreBLEU, and chrF, reporting consistently strong results. To the best of our knowledge, COCO-Urdu is the largest publicly available Urdu captioning dataset. By releasing both the dataset and the quality estimation pipeline, we aim to reduce language bias in multimodal research and establish a foundation for inclusive vision-language systems.

Paperid: 553, https://arxiv.org/pdf/2509.08897.pdf GitHub

Abstract:
With the rapid advancement of multimodal retrieval and its application in LLMs and multimodal LLMs, increasingly complex retrieval tasks have emerged. Existing methods predominantly rely on task-specific fine-tuning of vision-language models and are limited to single-modality queries or documents. In this paper, we propose ReT-2, a unified retrieval model that supports multimodal queries, composed of both images and text, and searches across multimodal document collections where text and images coexist. ReT-2 leverages multi-layer representations and a recurrent Transformer architecture with LSTM-inspired gating mechanisms to dynamically integrate information across layers and modalities, capturing fine-grained visual and textual details. We evaluate ReT-2 on the challenging M2KR and M-BEIR benchmarks across different retrieval configurations. Results demonstrate that ReT-2 consistently achieves state-of-the-art performance across diverse settings, while offering faster inference and reduced memory usage compared to prior approaches. When integrated into retrieval-augmented generation pipelines, ReT-2 also improves downstream performance on Encyclopedic-VQA and InfoSeek datasets. Our source code and trained models are publicly available at: https://github.com/aimagelab/ReT-2

Paperid: 554, https://arxiv.org/pdf/2509.08820.pdf GitHub

Abstract:
Robotic chemists promise to both liberate human experts from repetitive tasks and accelerate scientific discovery, yet remain in their infancy. Chemical experiments involve long-horizon procedures over hazardous and deformable substances, where success requires not only task completion but also strict compliance with experimental norms. To address these challenges, we propose \textit{RoboChemist}, a dual-loop framework that integrates Vision-Language Models (VLMs) with Vision-Language-Action (VLA) models. Unlike prior VLM-based systems (e.g., VoxPoser, ReKep) that rely on depth perception and struggle with transparent labware, and existing VLA systems (e.g., RDT, pi0) that lack semantic-level feedback for complex tasks, our method leverages a VLM to serve as (1) a planner to decompose tasks into primitive actions, (2) a visual prompt generator to guide VLA models, and (3) a monitor to assess task success and regulatory compliance. Notably, we introduce a VLA interface that accepts image-based visual targets from the VLM, enabling precise, goal-conditioned control. Our system successfully executes both primitive actions and complete multi-step chemistry protocols. Results show 23.57% higher average success rate and a 0.298 average increase in compliance rate over state-of-the-art VLA baselines, while also demonstrating strong generalization to objects and tasks.

Paperid: 555, https://arxiv.org/pdf/2509.08757.pdf GitHub

Abstract:
Robot navigation in dynamic, human-centered environments requires socially-compliant decisions grounded in robust scene understanding. Recent Vision-Language Models (VLMs) exhibit promising capabilities such as object recognition, common-sense reasoning, and contextual understanding-capabilities that align with the nuanced requirements of social robot navigation. However, it remains unclear whether VLMs can accurately understand complex social navigation scenes (e.g., inferring the spatial-temporal relations among agents and human intentions), which is essential for safe and socially compliant robot navigation. While some recent works have explored the use of VLMs in social robot navigation, no existing work systematically evaluates their ability to meet these necessary conditions. In this paper, we introduce the Social Navigation Scene Understanding Benchmark (SocialNav-SUB), a Visual Question Answering (VQA) dataset and benchmark designed to evaluate VLMs for scene understanding in real-world social robot navigation scenarios. SocialNav-SUB provides a unified framework for evaluating VLMs against human and rule-based baselines across VQA tasks requiring spatial, spatiotemporal, and social reasoning in social robot navigation. Through experiments with state-of-the-art VLMs, we find that while the best-performing VLM achieves an encouraging probability of agreeing with human answers, it still underperforms simpler rule-based approach and human consensus baselines, indicating critical gaps in social scene understanding of current VLMs. Our benchmark sets the stage for further research on foundation models for social robot navigation, offering a framework to explore how VLMs can be tailored to meet real-world social robot navigation needs. An overview of this paper along with the code and data can be found at https://larg.github.io/socialnav-sub .

Paperid: 556, https://arxiv.org/pdf/2509.08715.pdf GitHub

Abstract:
As multimodal large language models (MLLMs) advance, their large-scale architectures pose challenges for deployment in resource-constrained environments. In the age of large models, where energy efficiency, computational scalability and environmental sustainability are paramount, the development of lightweight and high-performance models is critical for real-world applications. As such, we propose a lightweight MLLM framework for end-to-end visual question answering. Our proposed approach centres on BreezeCLIP, a compact yet powerful vision-language encoder optimised for efficient multimodal understanding. With only 1.2 billion parameters overall, our model significantly reduces computational cost while achieving performance comparable to standard-size MLLMs. Experiments conducted on multiple datasets further validate its effectiveness in balancing accuracy and efficiency. The modular and extensible design enables generalisation to broader multimodal tasks. The proposed lightweight vision-language framework is denoted as BcQLM (BreezeCLIP-enhanced Q-Gated Multimodal Language Model). It offers a promising path toward deployable MLLMs under practical hardware constraints. The source code is available at https://github.com/thico0224/BcQLM.

Paperid: 557, https://arxiv.org/pdf/2509.08311.pdf GitHub

Abstract:
Medical vision-language pre-training shows great potential in learning representative features from massive paired radiographs and reports. However, in computed tomography (CT) scans, the distribution of lesions which contain intricate structures is characterized by spatial sparsity. Besides, the complex and implicit relationships between different pathological descriptions in each sentence of the report and their corresponding sub-regions in radiographs pose additional challenges. In this paper, we propose a Similarity-Driven Cross-Granularity Pre-training (SimCroP) framework on chest CTs, which combines similarity-driven alignment and cross-granularity fusion to improve radiograph interpretation. We first leverage multi-modal masked modeling to optimize the encoder for understanding precise low-level semantics from radiographs. Then, similarity-driven alignment is designed to pre-train the encoder to adaptively select and align the correct patches corresponding to each sentence in reports. The cross-granularity fusion module integrates multimodal information across instance level and word-patch level, which helps the model better capture key pathology structures in sparse radiographs, resulting in improved performance for multi-scale downstream tasks. SimCroP is pre-trained on a large-scale paired CT-reports dataset and validated on image classification and segmentation tasks across five public datasets. Experimental results demonstrate that SimCroP outperforms both cutting-edge medical self-supervised learning methods and medical vision-language pre-training methods. Codes and models are available at https://github.com/ToniChopp/SimCroP.

Paperid: 558, https://arxiv.org/pdf/2509.07966.pdf GitHub

Abstract:
Visual reasoning over structured data such as tables is a critical capability for modern vision-language models (VLMs), yet current benchmarks remain limited in scale, diversity, or reasoning depth, especially when it comes to rendered table images. Addressing this gap, we introduce Visual-TableQA, a large-scale, open-domain multimodal dataset specifically designed to evaluate and enhance visual reasoning over complex tabular data. Our generation pipeline is modular, scalable, and fully autonomous, involving multiple reasoning LLMs collaborating across distinct roles: generation, validation, and inspiration. Visual-TableQA comprises 2.5k richly structured LaTeX-rendered tables and 6k reasoning-intensive QA pairs, all produced at a cost of under USD 100. To promote diversity and creativity, our pipeline performs multi-model collaborative data generation via cross-model prompting ('inspiration') and LLM-jury filtering. Stronger models seed layouts and topics that weaker models elaborate, collectively distilling diverse reasoning patterns and visual structures into the dataset. Empirical results show that models fine-tuned on Visual-TableQA generalize robustly to external benchmarks, outperforming several proprietary models despite the dataset's synthetic nature. The full pipeline and resources are publicly available at https://github.com/AI-4-Everyone/Visual-TableQA.

Paperid: 559, https://arxiv.org/pdf/2509.07962.pdf GitHub

Abstract:
Many robotic manipulation tasks require sensing and responding to force signals such as torque to assess whether the task has been successfully completed and to enable closed-loop control. However, current Vision-Language-Action (VLA) models lack the ability to integrate such subtle physical feedback. In this work, we explore Torque-aware VLA models, aiming to bridge this gap by systematically studying the design space for incorporating torque signals into existing VLA architectures. We identify and evaluate several strategies, leading to three key findings. First, introducing torque adapters into the decoder consistently outperforms inserting them into the encoder.Third, inspired by joint prediction and planning paradigms in autonomous driving, we propose predicting torque as an auxiliary output, which further improves performance. This strategy encourages the model to build a physically grounded internal representation of interaction dynamics. Extensive quantitative and qualitative experiments across contact-rich manipulation benchmarks validate our findings.

Paperid: 560, https://arxiv.org/pdf/2509.07666.pdf GitHub

Abstract:
Document Understanding is a foundational AI capability with broad applications, and Document Question Answering (DocQA) is a key evaluation task. Traditional methods convert the document into text for processing by Large Language Models (LLMs), but this process strips away critical multi-modal information like figures. While Large Vision-Language Models (LVLMs) address this limitation, their constrained input size makes multi-page document comprehension infeasible. Retrieval-augmented generation (RAG) methods mitigate this by selecting relevant pages, but they rely solely on semantic relevance, ignoring logical connections between pages and the query, which is essential for reasoning. To this end, we propose MoLoRAG, a logic-aware retrieval framework for multi-modal, multi-page document understanding. By constructing a page graph that captures contextual relationships between pages, a lightweight VLM performs graph traversal to retrieve relevant pages, including those with logical connections often overlooked. This approach combines semantic and logical relevance to deliver more accurate retrieval. After retrieval, the top-$K$ pages are fed into arbitrary LVLMs for question answering. To enhance flexibility, MoLoRAG offers two variants: a training-free solution for easy deployment and a fine-tuned version to improve logical relevance checking. Experiments on four DocQA datasets demonstrate average improvements of 9.68% in accuracy over LVLM direct inference and 7.44% in retrieval precision over baselines. Codes and datasets are released at https://github.com/WxxShirley/MoLoRAG.

Paperid: 561, https://arxiv.org/pdf/2509.06951.pdf GitHub

Abstract:
Executing language-conditioned tasks in dynamic visual environments remains a central challenge in embodied AI. Existing Vision-Language-Action (VLA) models predominantly adopt reactive state-to-action mappings, often leading to short-sighted behaviors and poor robustness in dynamic scenes. In this paper, we introduce F1, a pretrained VLA framework which integrates the visual foresight generation into decision-making pipeline. F1 adopts a Mixture-of-Transformer architecture with dedicated modules for perception, foresight generation, and control, thereby bridging understanding, generation, and actions. At its core, F1 employs a next-scale prediction mechanism to synthesize goal-conditioned visual foresight as explicit planning targets. By forecasting plausible future visual states, F1 reformulates action generation as a foresight-guided inverse dynamics problem, enabling actions that implicitly achieve visual goals. To endow F1 with robust and generalizable capabilities, we propose a three-stage training recipe on an extensive dataset comprising over 330k trajectories across 136 diverse tasks. This training scheme enhances modular reasoning and equips the model with transferable visual foresight, which is critical for complex and dynamic environments. Extensive evaluations on real-world tasks and simulation benchmarks demonstrate F1 consistently outperforms existing approaches, achieving substantial gains in both task success rate and generalization ability.

Paperid: 562, https://arxiv.org/pdf/2509.06819.pdf GitHub

Abstract:
Learning-based controllers, such as diffusion policies and vision-language action models, often generate low-frequency or discontinuous robot state changes. Achieving smooth reference tracking requires a low-level controller that converts high-level targets commands into joint torques, enabling compliant behavior during contact interactions. We present CRISP, a lightweight C++ implementation of compliant Cartesian and joint-space controllers for the ROS2 control standard, designed for seamless integration with high-level learning-based policies as well as teleoperation. The controllers are compatible with any manipulator that exposes a joint-torque interface. Through our Python and Gymnasium interfaces, CRISP provides a unified pipeline for recording data from hardware and simulation and deploying high-level learning-based policies seamlessly, facilitating rapid experimentation. The system has been validated on hardware with the Franka Robotics FR3 and in simulation with the Kuka IIWA14 and Kinova Gen3. Designed for rapid integration, flexible deployment, and real-time performance, our implementation provides a unified pipeline for data collection and policy execution, lowering the barrier to applying learning-based methods on ROS2-compatible manipulators. Detailed documentation is available at the project website - https://utiasDSL.github.io/crisp_controllers.

Paperid: 563, https://arxiv.org/pdf/2509.06771.pdf GitHub

Abstract:
Dark humor in online memes poses unique challenges due to its reliance on implicit, sensitive, and culturally contextual cues. To address the lack of resources and methods for detecting dark humor in multimodal content, we introduce a novel dataset of 4,379 Reddit memes annotated for dark humor, target category (gender, mental health, violence, race, disability, and other), and a three-level intensity rating (mild, moderate, severe). Building on this resource, we propose a reasoning-augmented framework that first generates structured explanations for each meme using a Large Vision-Language Model (VLM). Through a Role-Reversal Self-Loop, VLM adopts the author's perspective to iteratively refine its explanations, ensuring completeness and alignment. We then extract textual features from both the OCR transcript and the self-refined reasoning via a text encoder, while visual features are obtained using a vision transformer. A Tri-stream Cross-Reasoning Network (TCRNet) fuses these three streams, text, image, and reasoning, via pairwise attention mechanisms, producing a unified representation for classification. Experimental results demonstrate that our approach outperforms strong baselines across three tasks: dark humor detection, target identification, and intensity prediction. The dataset, annotations, and code are released to facilitate further research in multimodal humor understanding and content moderation. Code and Dataset are available at: https://github.com/Sai-Kartheek-Reddy/D-Humor-Dark-Humor-Understanding-via-Multimodal-Open-ended-Reasoning

Paperid: 564, https://arxiv.org/pdf/2509.06336.pdf GitHub

Abstract:
Recent face anti-spoofing (FAS) methods have shown remarkable cross-domain performance by employing vision-language models like CLIP. However, existing CLIP-based FAS models do not fully exploit CLIP's patch embedding tokens, failing to detect critical spoofing clues. Moreover, these models rely on a single text prompt per class (e.g., 'live' or 'fake'), which limits generalization. To address these issues, we propose MVP-FAS, a novel framework incorporating two key modules: Multi-View Slot attention (MVS) and Multi-Text Patch Alignment (MTPA). Both modules utilize multiple paraphrased texts to generate generalized features and reduce dependence on domain-specific text. MVS extracts local detailed spatial features and global context from patch embeddings by leveraging diverse texts with multiple perspectives. MTPA aligns patches with multiple text representations to improve semantic robustness. Extensive experiments demonstrate that MVP-FAS achieves superior generalization performance, outperforming previous state-of-the-art methods on cross-domain datasets. Code: https://github.com/Elune001/MVP-FAS.

Paperid: 565, https://arxiv.org/pdf/2509.05978.pdf GitHub

Abstract:
Vision-language models have demonstrated impressive capabilities in generating 2D images under various conditions; however, the success of these models is largely enabled by extensive, readily available pretrained foundation models. Critically, comparable pretrained models do not exist for 3D, significantly limiting progress. As a result, the potential of vision-language models to produce high-resolution 3D counterfactual medical images conditioned solely on natural language remains unexplored. Addressing this gap would enable powerful clinical and research applications, such as personalized counterfactual explanations, simulation of disease progression, and enhanced medical training by visualizing hypothetical conditions in realistic detail. Our work takes a step toward this challenge by introducing a framework capable of generating high-resolution 3D counterfactual medical images of synthesized patients guided by free-form language prompts. We adapt state-of-the-art 3D diffusion models with enhancements from Simple Diffusion and incorporate augmented conditioning to improve text alignment and image quality. To our knowledge, this is the first demonstration of a language-guided native-3D diffusion model applied to neurological imaging, where faithful three-dimensional modeling is essential. On two neurological MRI datasets, our framework simulates varying counterfactual lesion loads in Multiple Sclerosis and cognitive states in Alzheimer's disease, generating high-quality images while preserving subject fidelity. Our results lay the groundwork for prompt-driven disease progression analysis in 3D medical imaging. Project link - https://lesupermomo.github.io/imagining-alternatives/.

Paperid: 566, https://arxiv.org/pdf/2509.05154.pdf GitHub

Abstract:
Vision-language models and their adaptations to image segmentation tasks present enormous potential for producing highly accurate and interpretable results. However, implementations based on CLIP and BiomedCLIP are still lagging behind more sophisticated architectures such as CRIS. In this work, instead of focusing on text prompt engineering as is the norm, we attempt to narrow this gap by showing how to ensemble vision-language segmentation models (VLSMs) with a low-complexity CNN. By doing so, we achieve a significant Dice score improvement of 6.3% on the BKAI polyp dataset using the ensembled BiomedCLIPSeg, while other datasets exhibit gains ranging from 1% to 6%. Furthermore, we provide initial results on additional four radiology and non-radiology datasets. We conclude that ensembling works differently across these datasets (from outperforming to underperforming the CRIS model), indicating a topic for future investigation by the community. The code is available at https://github.com/juliadietlmeier/VLSM-Ensemble.

Paperid: 567, https://arxiv.org/pdf/2509.04996.pdf GitHub

Abstract:
Developing efficient Vision-Language-Action (VLA) policies is crucial for practical robotics deployment, yet current approaches face prohibitive computational costs and resource requirements. Existing diffusion-based VLA policies require multi-billion-parameter models and massive datasets to achieve strong performance. We tackle this efficiency challenge with two contributions: intermediate-modality fusion, which reallocates capacity to the diffusion head by pruning up to $50\%$ of LLM layers, and action-specific Global-AdaLN conditioning, which cuts parameters by $20\%$ through modular adaptation. We integrate these advances into a novel 950 M-parameter VLA called FLOWER. Pretrained in just 200 H100 GPU hours, FLOWER delivers competitive performance with bigger VLAs across $190$ tasks spanning ten simulation and real-world benchmarks and demonstrates robustness across diverse robotic embodiments. In addition, FLOWER achieves a new SoTA of 4.53 on the CALVIN ABC benchmark. Demos, code and pretrained weights are available at https://intuitive-robots.github.io/flower_vla/.

Paperid: 568, https://arxiv.org/pdf/2509.04758.pdf GitHub

Abstract:
This paper proposes dynamic human group detection in videos. For detecting complex groups, not only the local appearance features of in-group members but also the global context of the scene are important. Such local and global appearance features in each frame are extracted using a Vision-Language Model (VLM) augmented for group detection in our method. For further improvement, the group structure should be consistent over time. While previous methods are stabilized on the assumption that groups are not changed in a video, our method detects dynamically changing groups by global optimization using a graph with all frames' groupness probabilities estimated by our groupness-augmented CLIP features. Our experimental results demonstrate that our method outperforms state-of-the-art group detection methods on public datasets. Code: https://github.com/irajisamurai/VLM-GroupDetection.git

Paperid: 569, https://arxiv.org/pdf/2509.04448.pdf GitHub

Abstract:
Multimodal misinformation, encompassing textual, visual, and cross-modal distortions, poses an increasing societal threat that is amplified by generative AI. Existing methods typically focus on a single type of distortion and struggle to generalize to unseen scenarios. In this work, we observe that different distortion types share common reasoning capabilities while also requiring task-specific skills. We hypothesize that joint training across distortion types facilitates knowledge sharing and enhances the model's ability to generalize. To this end, we introduce TRUST-VL, a unified and explainable vision-language model for general multimodal misinformation detection. TRUST-VL incorporates a novel Question-Aware Visual Amplifier module, designed to extract task-specific visual features. To support training, we also construct TRUST-Instruct, a large-scale instruction dataset containing 198K samples featuring structured reasoning chains aligned with human fact-checking workflows. Extensive experiments on both in-domain and zero-shot benchmarks demonstrate that TRUST-VL achieves state-of-the-art performance, while also offering strong generalization and interpretability.

Paperid: 570, https://arxiv.org/pdf/2509.03961.pdf GitHub

Abstract:
Although deep learning has advanced remote sensing change detection (RSCD), most methods rely solely on image modality, limiting feature representation, change pattern modeling, and generalization especially under illumination and noise disturbances. To address this, we propose MMChange, a multimodal RSCD method that combines image and text modalities to enhance accuracy and robustness. An Image Feature Refinement (IFR) module is introduced to highlight key regions and suppress environmental noise. To overcome the semantic limitations of image features, we employ a vision language model (VLM) to generate semantic descriptions of bitemporal images. A Textual Difference Enhancement (TDE) module then captures fine grained semantic shifts, guiding the model toward meaningful changes. To bridge the heterogeneity between modalities, we design an Image Text Feature Fusion (ITFF) module that enables deep cross modal integration. Extensive experiments on LEVIRCD, WHUCD, and SYSUCD demonstrate that MMChange consistently surpasses state of the art methods across multiple metrics, validating its effectiveness for multimodal RSCD. Code is available at: https://github.com/yikuizhai/MMChange.

Paperid: 571, https://arxiv.org/pdf/2509.03740.pdf GitHub

Abstract:
Vision-language models (VLMs) like CLIP have shown impressive zero-shot and few-shot learning capabilities across diverse applications. However, adapting these models to new fine-grained domains remains difficult due to reliance on prompt engineering and the high cost of full model fine-tuning. Existing adaptation approaches rely on augmented components, such as prompt tokens and adapter modules, which could limit adaptation quality, destabilize the model, and compromise the rich knowledge learned during pretraining. In this work, we present CLIP-SVD, a novel multi-modal and parameter-efficient adaptation technique that leverages Singular Value Decomposition (SVD) to modify the internal parameter space of CLIP without injecting additional modules. Specifically, we fine-tune only the singular values of the CLIP parameter matrices to rescale the basis vectors for domain adaptation while retaining the pretrained model. This design enables enhanced adaptation performance using only 0.04% of the model's total parameters and better preservation of its generalization ability. CLIP-SVD achieves state-of-the-art classification results on 11 natural and 10 biomedical datasets, outperforming previous methods in both accuracy and generalization under few-shot settings. Additionally, we leverage a natural language-based approach to analyze the effectiveness and dynamics of the CLIP adaptation to allow interpretability of CLIP-SVD. The code is publicly available at https://github.com/HealthX-Lab/CLIP-SVD.

Paperid: 572, https://arxiv.org/pdf/2509.03383.pdf GitHub

Abstract:
The integration of vision-language-action (VLA) models into embodied AI (EAI) robots is rapidly advancing their ability to perform complex, long-horizon tasks in humancentric environments. However, EAI systems introduce critical security risks: a compromised VLA model can directly translate adversarial perturbations on sensory input into unsafe physical actions. Traditional safety definitions and methodologies from the machine learning community are no longer sufficient. EAI systems raise new questions, such as what constitutes safety, how to measure it, and how to design effective attack and defense mechanisms in physically grounded, interactive settings. In this work, we present the first systematic study of adversarial safety attacks on embodied AI systems, grounded in ISO standards for human-robot interactions. We (1) formalize a principled taxonomy of safety violations (critical, dangerous, risky) based on physical constraints such as separation distance, velocity, and collision boundaries; (2) introduce ANNIEBench, a benchmark of nine safety-critical scenarios with 2,400 video-action sequences for evaluating embodied safety; and (3) ANNIE-Attack, a task-aware adversarial framework with an attack leader model that decomposes long-horizon goals into frame-level perturbations. Our evaluation across representative EAI models shows attack success rates exceeding 50% across all safety categories. We further demonstrate sparse and adaptive attack strategies and validate the real-world impact through physical robot experiments. These results expose a previously underexplored but highly consequential attack surface in embodied AI systems, highlighting the urgent need for security-driven defenses in the physical AI era. Code is available at https://github.com/RLCLab/Annie.

Paperid: 573, https://arxiv.org/pdf/2509.01907.pdf GitHub

Abstract:
Remote sensing is critical for disaster monitoring, yet existing datasets lack temporal image pairs and detailed textual annotations. While single-snapshot imagery dominates current resources, it fails to capture dynamic disaster impacts over time. To address this gap, we introduce the Remote Sensing Change Caption (RSCC) dataset, a large-scale benchmark comprising 62,315 pre-/post-disaster image pairs (spanning earthquakes, floods, wildfires, and more) paired with rich, human-like change captions. By bridging the temporal and semantic divide in remote sensing data, RSCC enables robust training and evaluation of vision-language models for disaster-aware bi-temporal understanding. Our results highlight RSCC's ability to facilitate detailed disaster-related analysis, paving the way for more accurate, interpretable, and scalable vision-language applications in remote sensing. Code and dataset are available at https://github.com/Bili-Sakura/RSCC.

Paperid: 574, https://arxiv.org/pdf/2509.01658.pdf GitHub

Abstract:
Mobile manipulation stands as a core challenge in robotics, enabling robots to assist humans across varied tasks and dynamic daily environments. Conventional mobile manipulation approaches often struggle to generalize across different tasks and environments due to the lack of large-scale training. However, recent advances in manipulation foundation models demonstrate impressive generalization capability on a wide range of fixed-base manipulation tasks, which are still limited to a fixed setting. Therefore, we devise a plug-in module named MoTo, which can be combined with any off-the-shelf manipulation foundation model to empower them with mobile manipulation ability. Specifically, we propose an interaction-aware navigation policy to generate robot docking points for generalized mobile manipulation. To enable zero-shot ability, we propose an interaction keypoints framework via vision-language models (VLM) under multi-view consistency for both target object and robotic arm following instructions, where fixed-base manipulation foundation models can be employed. We further propose motion planning objectives for the mobile base and robot arm, which minimize the distance between the two keypoints and maintain the physical feasibility of trajectories. In this way, MoTo guides the robot to move to the docking points where fixed-base manipulation can be successfully performed, and leverages VLM generation and trajectory optimization to achieve mobile manipulation in a zero-shot manner, without any requirement on mobile manipulation expert data. Extensive experimental results on OVMM and real-world demonstrate that MoTo achieves success rates of 2.68% and 16.67% higher than the state-of-the-art mobile manipulation methods, respectively, without requiring additional training data.

Paperid: 575, https://arxiv.org/pdf/2509.01552.pdf GitHub

Abstract:
Large vision-language models (LVLMs) have demonstrated remarkable capabilities in multimodal understanding tasks. However, the increasing demand for high-resolution image and long-video understanding results in substantial token counts, leading to reduced inference efficiency. Token compression offers a direct solution by reducing the number of tokens to be processed, thereby improving computational efficiency. Through extensive analysis, we identify two critical limitations in existing inner-LLM token compression methods: positional bias and incompatibility with efficient operators, which hinder their practical deployment for LVLM acceleration. This paper presents the first approach from a token variation perspective, revealing that visual token variations within LLMs exhibit task-agnostic properties. We propose Variation-aware Vision Token Dropping (\textit{i.e.}, \textbf{V$^2$Drop}), which progressively removes visual tokens with minimal variation during LVLM inference, thereby enhancing computational efficiency. Extensive experiments across multiple models and benchmarks demonstrate that our V$^2$Drop is able to maintain \textbf{94.0\%} and \textbf{98.6\%} of the original model performance for image and video understanding tasks respectively, while reducing LLM generation latency by \textbf{31.5\%} and \textbf{74.2\%}. When combined with efficient operators, V$^2$Drop further reduces GPU peak memory usage.

Paperid: 576, https://arxiv.org/pdf/2509.01259.pdf GitHub

Abstract:
Image captioning systems often produce generic descriptions that fail to capture event-level semantics which are crucial for applications like news reporting and digital archiving. We present ReCap, a novel pipeline for event-enriched image retrieval and captioning that incorporates broader contextual information from relevant articles to generate narrative-rich, factually grounded captions. Our approach addresses the limitations of standard vision-language models that typically focus on visible content while missing temporal, social, and historical contexts. ReCap comprises three integrated components: (1) a robust two-stage article retrieval system using DINOv2 embeddings with global feature similarity for initial candidate selection followed by patch-level mutual nearest neighbor similarity re-ranking; (2) a context extraction framework that synthesizes information from article summaries, generic captions, and original source metadata; and (3) a large language model-based caption generation system with Semantic Gaussian Normalization to enhance fluency and relevance. Evaluated on the OpenEvents V1 dataset as part of Track 1 in the EVENTA 2025 Grand Challenge, ReCap achieved a strong overall score of 0.54666, ranking 2nd on the private test set. These results highlight ReCap's effectiveness in bridging visual perception with real-world knowledge, offering a practical solution for context-aware image understanding in high-stakes domains. The code is available at https://github.com/Noridom1/EVENTA2025-Event-Enriched-Image-Captioning.

Paperid: 577, https://arxiv.org/pdf/2509.01215.pdf GitHub

Abstract:
High-quality labeled data is essential for training accurate document conversion models, particularly in domains with complex formats such as tables, formulas, and multi-column text. However, manual annotation is both costly and time-consuming, while automatic labeling using existing models often lacks accuracy in handling such challenging scenarios. Consequently, training student models by distilling outputs from teacher models can significantly limit their performance in real-world applications. In this paper, we propose a fully automated, distillation-free framework comprising two stages for constructing high-quality document extraction datasets and models capable of handling diverse document formats and layouts. In the first stage, we introduce a method for generating large-scale, diverse synthetic data, which enables a model to extract key elements in a unified format with strong initial performance. In the second stage, we present a self-improvement approach that further adapts the model, initially trained on synthetic data, to real-world documents. Specifically, we first use the fine-tuned model to annotate real documents, then apply a suite of filtering strategies to verify annotation quality, and finally retrain the model on the verified dataset. By iteratively repeating this process, we progressively enhance both the model's conversion capabilities and the quality of the generated data. We train a public POINTS-1.5 model to obtain POINTS-Reader, which surpasses many existing public and proprietary models of comparable or larger size. Our model is available at https://github.com/Tencent/POINTS-Reader.

Paperid: 578, https://arxiv.org/pdf/2509.01106.pdf GitHub

Abstract:
We introduce Robix, a unified model that integrates robot reasoning, task planning, and natural language interaction within a single vision-language architecture. Acting as the high-level cognitive layer in a hierarchical robot system, Robix dynamically generates atomic commands for the low-level controller and verbal responses for human interaction, enabling robots to follow complex instructions, plan long-horizon tasks, and interact naturally with human within an end-to-end framework. Robix further introduces novel capabilities such as proactive dialogue, real-time interruption handling, and context-aware commonsense reasoning during task execution. At its core, Robix leverages chain-of-thought reasoning and adopts a three-stage training strategy: (1) continued pretraining to enhance foundational embodied reasoning abilities including 3D spatial understanding, visual grounding, and task-centric reasoning; (2) supervised finetuning to model human-robot interaction and task planning as a unified reasoning-action sequence; and (3) reinforcement learning to improve reasoning-action consistency and long-horizon task coherence. Extensive experiments demonstrate that Robix outperforms both open-source and commercial baselines (e.g., GPT-4o and Gemini 2.5 Pro) in interactive task execution, demonstrating strong generalization across diverse instruction types (e.g., open-ended, multi-stage, constrained, invalid, and interrupted) and various user-involved tasks such as table bussing, grocery shopping, and dietary filtering.

Paperid: 579, https://arxiv.org/pdf/2509.00751.pdf GitHub

Abstract:
Event-based image retrieval from free-form captions presents a significant challenge: models must understand not only visual features but also latent event semantics, context, and real-world knowledge. Conventional vision-language retrieval approaches often fall short when captions describe abstract events, implicit causality, temporal context, or contain long, complex narratives. To tackle these issues, we introduce a multi-stage retrieval framework combining dense article retrieval, event-aware language model reranking, and efficient image collection, followed by caption-guided semantic matching and rank-aware selection. We leverage Qwen3 for article search, Qwen3-Reranker for contextual alignment, and Qwen2-VL for precise image scoring. To further enhance performance and robustness, we fuse outputs from multiple configurations using Reciprocal Rank Fusion (RRF). Our system achieves the top-1 score on the private test set of Track 2 in the EVENTA 2025 Grand Challenge, demonstrating the effectiveness of combining language-based reasoning and multimodal retrieval for complex, real-world image understanding. The code is available at https://github.com/vdkhoi20/EVENT-Retriever.

Paperid: 580, https://arxiv.org/pdf/2509.00576.pdf GitHub

Abstract:
We present Galaxea Open-World Dataset, a large-scale, diverse collection of robot behaviors recorded in authentic human living and working environments. All demonstrations are gathered using a consistent robotic embodiment, paired with precise subtask-level language annotations to facilitate both training and evaluation. Building on this dataset, we introduce G0, a dual-system framework that couples a Vision-Language Model (VLM) for multimodal planning with a Vision-Language-Action (VLA) model for fine-grained execution. G0 is trained using a three-stage curriculum: cross-embodiment pre-training, single-embodiment pre-training, and task-specific post-training. A comprehensive benchmark spanning tabletop manipulation, few-shot learning, and long-horizon mobile manipulation, demonstrates the effectiveness of our approach. In particular, we find that the single-embodiment pre-training stage, together with the Galaxea Open-World Dataset, plays a critical role in achieving strong performance.

Paperid: 581, https://arxiv.org/pdf/2509.00305.pdf GitHub GitHub

Abstract:
Transductive few-shot learning has triggered an abundant literature focusing on vision-only models, but is still at a nascent stage within the recent context of foundational vision-language models (VLMs). Only a few recent methods addressed the problem, pointing to the potential of tranduction in VLMs and to the need for VLM-tailored methods. Building on this momentum, we leverage information-theoretic concepts and recent progress in parameter-efficient fine-tuning (PEFT), developing a highly competitive transductive few-shot CLIP method. Specifically, we introduce a novel Language-aware Information MaximizatiOn (LIMO) loss integrating three complementary terms: (i) the mutual information between the vision inputs and the textual class descriptions; (ii) a Kullback-Leibler (KL) divergence penalizing deviation of the network's probabilistic outputs from the text-driven zero-shot predictions; and (iii) a standard cross-entropy loss based on the labeled shots. Furthermore, we challenge the commonly followed fine-tuning practices in the context of transductive few-shot learning, and explore PEFT strategies, completely overlooked in this context. Surprisingly, we observe substantial boosts in performances, which points to the potential of adapting a subset of the model's parameters in the transductive few-shot setting. We report comprehensive evaluations, which show that LIMO outperforms the very recent transductive few-shot CLIP methods by a large margin and yields significant gains over the best-performing inductive methods. Our code is publicly available at:\[ \href{https://github.com/ghassenbaklouti/LIMO}{\text{here}} \]

Paperid: 582, https://arxiv.org/pdf/2508.21809.pdf GitHub

Abstract:
Understanding objects in videos in terms of fine-grained localization masks and detailed semantic properties is a fundamental task in video understanding. In this paper, we propose VoCap, a flexible video model that consumes a video and a prompt of various modalities (text, box or mask), and produces a spatio-temporal masklet with a corresponding object-centric caption. As such our model addresses simultaneously the tasks of promptable video object segmentation, referring expression segmentation, and object captioning. Since obtaining data for this task is tedious and expensive, we propose to annotate an existing large-scale segmentation dataset (SAV) with pseudo object captions. We do so by preprocessing videos with their ground-truth masks to highlight the object of interest and feed this to a large Vision Language Model (VLM). For an unbiased evaluation, we collect manual annotations on the validation set. We call the resulting dataset SAV-Caption. We train our VoCap model at scale on a SAV-Caption together with a mix of other image and video datasets. Our model yields state-of-the-art results on referring expression video object segmentation, is competitive on semi-supervised video object segmentation, and establishes a benchmark for video object captioning. Our dataset will be made available at https://github.com/google-deepmind/vocap.

Paperid: 583, https://arxiv.org/pdf/2508.21046.pdf GitHub GitHub

Abstract:
Recent Vision-Language-Action (VLA) models built on pre-trained Vision-Language Models (VLMs) require extensive post-training, resulting in high computational overhead that limits scalability and deployment.We propose CogVLA, a Cognition-Aligned Vision-Language-Action framework that leverages instruction-driven routing and sparsification to improve both efficiency and performance. CogVLA draws inspiration from human multimodal coordination and introduces a 3-stage progressive architecture. 1) Encoder-FiLM based Aggregation Routing (EFA-Routing) injects instruction information into the vision encoder to selectively aggregate and compress dual-stream visual tokens, forming a instruction-aware latent representation. 2) Building upon this compact visual encoding, LLM-FiLM based Pruning Routing (LFP-Routing) introduces action intent into the language model by pruning instruction-irrelevant visually grounded tokens, thereby achieving token-level sparsity. 3) To ensure that compressed perception inputs can still support accurate and coherent action generation, we introduce V-L-A Coupled Attention (CAtten), which combines causal vision-language attention with bidirectional action parallel decoding. Extensive experiments on the LIBERO benchmark and real-world robotic tasks demonstrate that CogVLA achieves state-of-the-art performance with success rates of 97.4% and 70.0%, respectively, while reducing training costs by 2.5-fold and decreasing inference latency by 2.8-fold compared to OpenVLA. CogVLA is open-sourced and publicly available at https://github.com/JiuTian-VL/CogVLA.

Paperid: 584, https://arxiv.org/pdf/2508.20783.pdf GitHub

Abstract:
A fundamental aspect of the semantics of natural language is that novel meanings can be formed from the composition of previously known parts. Vision-language models (VLMs) have made significant progress in recent years, however, there is evidence that they are unable to perform this kind of composition. For example, given an image of a red cube and a blue cylinder, a VLM such as CLIP is likely to incorrectly label the image as a red cylinder or a blue cube, indicating it represents the image as a `bag-of-words' and fails to capture compositional semantics. Diffusion models have recently gained significant attention for their impressive generative abilities, and zero-shot classifiers based on diffusion models have been shown to perform competitively with CLIP in certain compositional tasks. In this work we explore whether the generative Diffusion Classifier has improved compositional generalisation abilities compared to discriminative models. We assess three models -- Diffusion Classifier, CLIP, and ViLT -- on their ability to bind objects with attributes and relations in both zero-shot learning (ZSL) and generalised zero-shot learning (GZSL) settings. Our results show that the Diffusion Classifier and ViLT perform well at concept binding tasks, but that all models struggle significantly with the relational GZSL task, underscoring the broader challenges VLMs face with relational reasoning. Analysis of CLIP embeddings suggests that the difficulty may stem from overly similar representations of relational concepts such as left and right. Code and dataset are available at: https://github.com/otmive/diffusion_classifier_clip

Paperid: 585, https://arxiv.org/pdf/2508.20029.pdf GitHub

Abstract:
In dynamic environments, unfamiliar objects and distribution shifts are often encountered, which challenge the generalization abilities of the deployed trained models. This work addresses Incremental Test Time Adaptation of Vision Language Models, tackling scenarios where unseen classes and unseen domains continuously appear during testing. Unlike traditional Test Time Adaptation approaches, where the test stream comes only from a predefined set of classes, our framework allows models to adapt simultaneously to both covariate and label shifts, actively incorporating new classes as they emerge. Towards this goal, we establish a new benchmark for ITTA, integrating single image TTA methods for VLMs with active labeling techniques that query an oracle for samples potentially representing unseen classes during test time. We propose a segmentation assisted active labeling module, termed SegAssist, which is training free and repurposes the segmentation capabilities of VLMs to refine active sample selection, prioritizing samples likely to belong to unseen classes. Extensive experiments on several benchmark datasets demonstrate the potential of SegAssist to enhance the performance of VLMs in real world scenarios, where continuous adaptation to emerging data is essential. Project-page:https://manogna-s.github.io/segassist/

Paperid: 586, https://arxiv.org/pdf/2508.19944.pdf GitHub

Abstract:
Understanding and reasoning over text within visual contexts poses a significant challenge for Vision-Language Models (VLMs), given the complexity and diversity of real-world scenarios. To address this challenge, text-rich Visual Question Answering (VQA) datasets and benchmarks have emerged for high-resource languages like English. However, a critical gap persists for low-resource languages such as Korean, where the lack of comprehensive benchmarks hinders robust model evaluation and comparison. To bridge this gap, we introduce KRETA, a benchmark for Korean Reading and rEasoning in Text-rich VQA Attuned to diverse visual contexts. KRETA facilitates an in-depth evaluation of both visual text understanding and reasoning capabilities, while also supporting a multifaceted assessment across 15 domains and 26 image types. Additionally, we introduce a semi-automated VQA generation pipeline specifically optimized for text-rich settings, leveraging refined stepwise image decomposition and a rigorous seven-metric evaluation protocol to ensure data quality. While KRETA is tailored for Korean, we hope our adaptable and extensible pipeline will facilitate the development of similar benchmarks in other languages, thereby accelerating multilingual VLM research. The code and dataset for KRETA are available at https://github.com/tabtoyou/KRETA.

Paperid: 587, https://arxiv.org/pdf/2508.19485.pdf GitHub

Abstract:
Gas leaks pose serious threats to human health and contribute significantly to atmospheric pollution, drawing increasing public concern. However, the lack of effective detection methods hampers timely and accurate identification of gas leaks. While some vision-based techniques leverage infrared videos for leak detection, the blurry and non-rigid nature of gas clouds often limits their effectiveness. To address these challenges, we propose a novel framework called Joint Vision-Language Gas leak Segmentation (JVLGS), which integrates the complementary strengths of visual and textual modalities to enhance gas leak representation and segmentation. Recognizing that gas leaks are sporadic and many video frames may contain no leak at all, our method incorporates a post-processing step to reduce false positives caused by noise and non-target objects, an issue that affects many existing approaches. Extensive experiments conducted across diverse scenarios show that JVLGS significantly outperforms state-of-the-art gas leak segmentation methods. We evaluate our model under both supervised and few-shot learning settings, and it consistently achieves strong performance in both, whereas competing methods tend to perform well in only one setting or poorly in both. Code available at: https://github.com/GeekEagle/JVLGS

Paperid: 588, https://arxiv.org/pdf/2508.19322.pdf GitHub

Abstract:
Agentic AI is advancing rapidly, yet truly autonomous medical-imaging triage, where a system decides when to stop, escalate, or defer under real constraints, remains relatively underexplored. To address this gap, we introduce AT-CXR, an uncertainty-aware agent for chest X-rays. The system estimates per-case confidence and distributional fit, then follows a stepwise policy to issue an automated decision or abstain with a suggested label for human intervention. We evaluate two router designs that share the same inputs and actions: a deterministic rule-based router and an LLM-decided router. Across five-fold evaluation on a balanced subset of NIH ChestX-ray14 dataset, both variants outperform strong zero-shot vision-language models and state-of-the-art supervised classifiers, achieving higher full-coverage accuracy and superior selective-prediction performance, evidenced by a lower area under the risk-coverage curve (AURC) and a lower error rate at high coverage, while operating with lower latency that meets practical clinical constraints. The two routers provide complementary operating points, enabling deployments to prioritize maximal throughput or maximal accuracy. Our code is available at https://github.com/XLIAaron/uncertainty-aware-cxr-agent.

Paperid: 589, https://arxiv.org/pdf/2508.19298.pdf GitHub

Abstract:
Large Vision Language Models (LVLMs) have demonstrated remarkable capabilities across various downstream tasks, including biometric face recognition (FR) with description. However, demographic biases remain a critical concern in FR, as these foundation models often fail to perform equitably across diverse demographic groups, considering ethnicity/race, gender, and age. Therefore, through our work DemoBias, we conduct an empirical evaluation to investigate the extent of demographic biases in LVLMs for biometric FR with textual token generation tasks. We fine-tuned and evaluated three widely used pre-trained LVLMs: LLaVA, BLIP-2, and PaliGemma on our own generated demographic-balanced dataset. We utilize several evaluation metrics, like group-specific BERTScores and the Fairness Discrepancy Rate, to quantify and trace the performance disparities. The experimental results deliver compelling insights into the fairness and reliability of LVLMs across diverse demographic groups. Our empirical study uncovered demographic biases in LVLMs, with PaliGemma and LLaVA exhibiting higher disparities for Hispanic/Latino, Caucasian, and South Asian groups, whereas BLIP-2 demonstrated comparably consistent. Repository: https://github.com/Sufianlab/DemoBias.

Paperid: 590, https://arxiv.org/pdf/2508.19236.pdf GitHub

Abstract:
Temporal context is essential for robotic manipulation because such tasks are inherently non-Markovian, yet mainstream VLA models typically overlook it and struggle with long-horizon, temporally dependent tasks. Cognitive science suggests that humans rely on working memory to buffer short-lived representations for immediate control, while the hippocampal system preserves verbatim episodic details and semantic gist of past experience for long-term memory. Inspired by these mechanisms, we propose MemoryVLA, a Cognition-Memory-Action framework for long-horizon robotic manipulation. A pretrained VLM encodes the observation into perceptual and cognitive tokens that form working memory, while a Perceptual-Cognitive Memory Bank stores low-level details and high-level semantics consolidated from it. Working memory retrieves decision-relevant entries from the bank, adaptively fuses them with current tokens, and updates the bank by merging redundancies. Using these tokens, a memory-conditioned diffusion action expert yields temporally aware action sequences. We evaluate MemoryVLA on 150+ simulation and real-world tasks across three robots. On SimplerEnv-Bridge, Fractal, and LIBERO-5 suites, it achieves 71.9%, 72.7%, and 96.5% success rates, respectively, all outperforming state-of-the-art baselines CogACT and pi-0, with a notable +14.6 gain on Bridge. On 12 real-world tasks spanning general skills and long-horizon temporal dependencies, MemoryVLA achieves 84.0% success rate, with long-horizon tasks showing a +26 improvement over state-of-the-art baseline. Project Page: https://shihao1895.github.io/MemoryVLA

Paperid: 591, https://arxiv.org/pdf/2508.19024.pdf GitHub

Abstract:
Partially Relevant Video Retrieval (PRVR) is a practical yet challenging task that involves retrieving videos based on queries relevant to only specific segments. While existing works follow the paradigm of developing models to process unimodal features, powerful pretrained vision-language models like CLIP remain underexplored in this field. To bridge this gap, we propose ProPy, a model with systematic architectural adaption of CLIP specifically designed for PRVR. Drawing insights from the semantic relevance of multi-granularity events, ProPy introduces two key innovations: (1) A Prompt Pyramid structure that organizes event prompts to capture semantics at multiple granularity levels, and (2) An Ancestor-Descendant Interaction Mechanism built on the pyramid that enables dynamic semantic interaction among events. With these designs, ProPy achieves SOTA performance on three public datasets, outperforming previous models by significant margins. Code is available at https://github.com/BUAAPY/ProPy.

Paperid: 592, https://arxiv.org/pdf/2508.18805.pdf GitHub

Abstract:
Vision-Language Models (VLMs) are increasingly deployed in real-world applications, but their high inference cost makes them vulnerable to resource consumption attacks. Prior attacks attempt to extend VLM output sequences by optimizing adversarial images, thereby increasing inference costs. However, these extended outputs often introduce irrelevant abnormal content, compromising attack stealthiness. This trade-off between effectiveness and stealthiness poses a major limitation for existing attacks. To address this challenge, we propose \textit{Hidden Tail}, a stealthy resource consumption attack that crafts prompt-agnostic adversarial images, inducing VLMs to generate maximum-length outputs by appending special tokens invisible to users. Our method employs a composite loss function that balances semantic preservation, repetitive special token induction, and suppression of the end-of-sequence (EOS) token, optimized via a dynamic weighting strategy. Extensive experiments show that \textit{Hidden Tail} outperforms existing attacks, increasing output length by up to 19.2$\times$ and reaching the maximum token limit, while preserving attack stealthiness. These results highlight the urgent need to improve the robustness of VLMs against efficiency-oriented adversarial threats. Our code is available at https://github.com/zhangrui4041/Hidden_Tail.

Paperid: 593, https://arxiv.org/pdf/2508.18649.pdf GitHub

Abstract:
Safeguarding vision-language models (VLMs) is a critical challenge, as existing methods often suffer from over-defense, which harms utility, or rely on shallow alignment, failing to detect complex threats that require deep reasoning. To this end, we introduce PRISM (Principled Reasoning for Integrated Safety in Multimodality), a system2-like framework that aligns VLMs by embedding a structured, safety-aware reasoning process. Our framework consists of two key components: PRISM-CoT, a dataset that teaches safety-aware chain-of-thought reasoning, and PRISM-DPO, generated via Monte Carlo Tree Search (MCTS) to further refine this reasoning through Direct Preference Optimization to help obtain a delicate safety boundary. Comprehensive evaluations demonstrate PRISM's effectiveness, achieving remarkably low attack success rates including 0.15% on JailbreakV-28K for Qwen2-VL and 90% improvement over the previous best method on VLBreak for LLaVA-1.5. PRISM also exhibits strong robustness against adaptive attacks, significantly increasing computational costs for adversaries, and generalizes effectively to out-of-distribution challenges, reducing attack success rates to just 8.70% on the challenging multi-image MIS benchmark. Remarkably, this robust defense is achieved while preserving, and in some cases enhancing, model utility. To promote reproducibility, we have made our code, data, and model weights available at https://github.com/SaFoLab-WISC/PRISM.

Paperid: 594, https://arxiv.org/pdf/2508.18463.pdf GitHub

Abstract:
Detecting anomalies in surveillance footage is inherently challenging due to their unpredictable and context-dependent nature. This work introduces a novel context-aware zero-shot anomaly detection framework that identifies abnormal events without exposure to anomaly examples during training. The proposed hybrid architecture combines TimeSformer, DPC, and CLIP to model spatiotemporal dynamics and semantic context. TimeSformer serves as the vision backbone to extract rich spatial-temporal features, while DPC forecasts future representations to identify temporal deviations. Furthermore, a CLIP-based semantic stream enables concept-level anomaly detection through context-specific text prompts. These components are jointly trained using InfoNCE and CPC losses, aligning visual inputs with their temporal and semantic representations. A context-gating mechanism further enhances decision-making by modulating predictions with scene-aware cues or global video features. By integrating predictive modeling with vision-language understanding, the system can generalize to previously unseen behaviors in complex environments. This framework bridges the gap between temporal reasoning and semantic context in zero-shot anomaly detection for surveillance. The code for this research has been made available at https://github.com/NK-II/Context-Aware-Zero-Shot-Anomaly-Detection-in-Surveillance.

Paperid: 595, https://arxiv.org/pdf/2508.18269.pdf GitHub

Abstract:
Many Vision-Language-Action (VLA) models are built upon an internal world model trained via direct next-frame prediction ($v_t \rightarrow v_{t+1}$). This paradigm, however, presents a fundamental challenge: it \textbf{conflates} the task of predicting physical motion with that of rendering static appearance, forcing a single mechanism to handle both. This inherent coupling often leads to physically implausible forecasts and inefficient policy learning. To address this limitation, we introduce the \textbf{Visual Chain of Thought (Visual CoT)}, a framework that disentangles these processes by compelling the model to first reason about \textbf{motion dynamics} before generating the future frame's \textbf{visual appearance}. We instantiate this principle by proposing \textbf{FlowVLA}, an autoregressive Transformer that explicitly materializes this reasoning process as ``$v_t \rightarrow f_t \rightarrow v_{t+1}$'', where $f_t$ is an intermediate optical flow prediction. By forcing the model to first commit to a motion plan ($f_t$), FlowVLA learns disentangled dynamics, resulting in more coherent visual predictions and significantly more efficient policy learning. Experiments on challenging robotics manipulation benchmarks demonstrate that FlowVLA achieves state-of-the-art performance with substantially improved sample efficiency, pointing toward a more principled foundation for world modeling in VLAs. Project page: https://irpn-lab.github.io/FlowVLA/

Paperid: 596, https://arxiv.org/pdf/2508.18268.pdf GitHub

Abstract:
Bimanual manipulation has been widely applied in household services and manufacturing, which enables the complex task completion with coordination requirements. Recent diffusion-based policy learning approaches have achieved promising performance in modeling action distributions for bimanual manipulation. However, they ignored the physical safety constraints of bimanual manipulation, which leads to the dangerous behaviors with damage to robots and objects. To this end, we propose a test-time trajectory optimization framework named SafeBimanual for any pre-trained diffusion-based bimanual manipulation policies, which imposes the safety constraints on bimanual actions to avoid dangerous robot behaviors with improved success rate. Specifically, we design diverse cost functions for safety constraints in different dual-arm cooperation patterns including avoidance of tearing objects and collision between arms and objects, which optimizes the manipulator trajectories with guided sampling of diffusion denoising process. Moreover, we employ a vision-language model (VLM) to schedule the cost functions by specifying keypoints and corresponding pairwise relationship, so that the optimal safety constraint is dynamically generated in the entire bimanual manipulation process. SafeBimanual demonstrates superiority on 8 simulated tasks in RoboTwin with a 13.7% increase in success rate and a 18.8% reduction in unsafe interactions over state-of-the-art diffusion-based methods. Extensive experiments on 4 real-world tasks further verify its practical value by improving the success rate by 32.5%.

Paperid: 597, https://arxiv.org/pdf/2508.18040.pdf GitHub

Abstract:
Vision language model (VLM)-based mobile agents show great potential for assisting users in performing instruction-driven tasks. However, these agents typically struggle with personalized instructions -- those containing ambiguous, user-specific context -- a challenge that has been largely overlooked in previous research. In this paper, we define personalized instructions and introduce PerInstruct, a novel human-annotated dataset covering diverse personalized instructions across various mobile scenarios. Furthermore, given the limited personalization capabilities of existing mobile agents, we propose PerPilot, a plug-and-play framework powered by large language models (LLMs) that enables mobile agents to autonomously perceive, understand, and execute personalized user instructions. PerPilot identifies personalized elements and autonomously completes instructions via two complementary approaches: memory-based retrieval and reasoning-based exploration. Experimental results demonstrate that PerPilot effectively handles personalized tasks with minimal user intervention and progressively improves its performance with continued use, underscoring the importance of personalization-aware reasoning for next-generation mobile agents. The dataset and code are available at: https://github.com/xinwang-nwpu/PerPilot

Paperid: 598, https://arxiv.org/pdf/2508.17398.pdf GitHub

Abstract:
Dashboards are powerful visualization tools for data-driven decision-making, integrating multiple interactive views that allow users to explore, filter, and navigate data. Unlike static charts, dashboards support rich interactivity, which is essential for uncovering insights in real-world analytical workflows. However, existing question-answering benchmarks for data visualizations largely overlook this interactivity, focusing instead on static charts. This limitation severely constrains their ability to evaluate the capabilities of modern multimodal agents designed for GUI-based reasoning. To address this gap, we introduce DashboardQA, the first benchmark explicitly designed to assess how vision-language GUI agents comprehend and interact with real-world dashboards. The benchmark includes 112 interactive dashboards from Tableau Public and 405 question-answer pairs with interactive dashboards spanning five categories: multiple-choice, factoid, hypothetical, multi-dashboard, and conversational. By assessing a variety of leading closed- and open-source GUI agents, our analysis reveals their key limitations, particularly in grounding dashboard elements, planning interaction trajectories, and performing reasoning. Our findings indicate that interactive dashboard reasoning is a challenging task overall for all the VLMs evaluated. Even the top-performing agents struggle; for instance, the best agent based on Gemini-Pro-2.5 achieves only 38.69% accuracy, while the OpenAI CUA agent reaches just 22.69%, demonstrating the benchmark's significant difficulty. We release DashboardQA at https://github.com/vis-nlp/DashboardQA

Paperid: 599, https://arxiv.org/pdf/2508.17298.pdf GitHub

Abstract:
Compositional visual reasoning has emerged as a key research frontier in multimodal AI, aiming to endow machines with the human-like ability to decompose visual scenes, ground intermediate concepts, and perform multi-step logical inference. While early surveys focus on monolithic vision-language models or general multimodal reasoning, a dedicated synthesis of the rapidly expanding compositional visual reasoning literature is still missing. We fill this gap with a comprehensive survey spanning 2023 to 2025 that systematically reviews 260+ papers from top venues (CVPR, ICCV, NeurIPS, ICML, ACL, etc.). We first formalize core definitions and describe why compositional approaches offer advantages in cognitive alignment, semantic fidelity, robustness, interpretability, and data efficiency. Next, we trace a five-stage paradigm shift: from prompt-enhanced language-centric pipelines, through tool-enhanced LLMs and tool-enhanced VLMs, to recently minted chain-of-thought reasoning and unified agentic VLMs, highlighting their architectural designs, strengths, and limitations. We then catalog 60+ benchmarks and corresponding metrics that probe compositional visual reasoning along dimensions such as grounding accuracy, chain-of-thought faithfulness, and high-resolution perception. Drawing on these analyses, we distill key insights, identify open challenges (e.g., limitations of LLM-based reasoning, hallucination, a bias toward deductive reasoning, scalable supervision, tool integration, and benchmark limitations), and outline future directions, including world-model integration, human-AI collaborative reasoning, and richer evaluation protocols. By offering a unified taxonomy, historical roadmap, and critical outlook, this survey aims to serve as a foundational reference and inspire the next generation of compositional visual reasoning research.

Paperid: 600, https://arxiv.org/pdf/2508.17188.pdf GitHub

Abstract:
Multi-agent systems built upon large language models (LLMs) have demonstrated remarkable capabilities in tackling complex compositional tasks. In this work, we apply this paradigm to the paper-to-poster generation problem, a practical yet time-consuming process faced by researchers preparing for conferences. While recent approaches have attempted to automate this task, most neglect core design and aesthetic principles, resulting in posters that require substantial manual refinement. To address these design limitations, we propose PosterGen, a multi-agent framework that mirrors the workflow of professional poster designers. It consists of four collaborative specialized agents: (1) Parser and Curator agents extract content from the paper and organize storyboard; (2) Layout agent maps the content into a coherent spatial layout; (3) Stylist agents apply visual design elements such as color and typography; and (4) Renderer composes the final poster. Together, these agents produce posters that are both semantically grounded and visually appealing. To evaluate design quality, we introduce a vision-language model (VLM)-based rubric that measures layout balance, readability, and aesthetic coherence. Experimental results show that PosterGen consistently matches in content fidelity, and significantly outperforms existing methods in visual designs, generating posters that are presentation-ready with minimal human refinements.

Paperid: 601, https://arxiv.org/pdf/2508.17037.pdf GitHub

Abstract:
The proliferation of digital food content has intensified the need for robust and accurate systems capable of fine-grained visual understanding and retrieval. In this work, we address the challenging task of food image-to-text matching, a critical component in applications such as dietary monitoring, smart kitchens, and restaurant automation. We propose F4-ITS: Fine-grained Feature Fusion for Food Image-Text Search, a training-free, vision-language model (VLM)-guided framework that significantly improves retrieval performance through enhanced multi-modal feature representations. Our approach introduces two key contributions: (1) a uni-directional(and bi-directional) multi-modal fusion strategy that combines image embeddings with VLM-generated textual descriptions to improve query expressiveness, and (2) a novel feature-based re-ranking mechanism for top-k retrieval, leveraging predicted food ingredients to refine results and boost precision. Leveraging open-source image-text encoders, we demonstrate substantial gains over standard baselines - achieving ~10% and ~7.7% improvements in top-1 retrieval under dense and sparse caption scenarios, and a ~28.6% gain in top-k ingredient-level retrieval. Additionally, we show that smaller models (e.g., ViT-B/32) can match or outperform larger counterparts (e.g., ViT-H, ViT-G, ViT-bigG) when augmented with textual fusion, highlighting the effectiveness of our method in resource-constrained settings. Code and test datasets will be made publicly available at: https://github.com/mailcorahul/f4-its

Paperid: 602, https://arxiv.org/pdf/2508.16943.pdf GitHub GitHub

Abstract:
We introduce HumanoidVerse, a novel framework for vision-language guided humanoid control that enables a single physically simulated robot to perform long-horizon, multi-object rearrangement tasks across diverse scenes. Unlike prior methods that operate in fixed settings with single-object interactions, our approach supports consecutive manipulation of multiple objects, guided only by natural language instructions and egocentric camera RGB observations. HumanoidVerse is trained via a multi-stage curriculum using a dual-teacher distillation pipeline, enabling fluid transitions between sub-tasks without requiring environment resets. To support this, we construct a large-scale dataset comprising 350 multi-object tasks spanning four room layouts. Extensive experiments in the Isaac Gym simulator demonstrate that our method significantly outperforms prior state-of-the-art in both task success rate and spatial precision, and generalizes well to unseen environments and instructions. Our work represents a key step toward robust, general-purpose humanoid agents capable of executing complex, sequential tasks under real-world sensory constraints. The video visualization results can be found on the project page: https://haozhuo-zhang.github.io/HumanoidVerse-project-page/.

Paperid: 603, https://arxiv.org/pdf/2508.16932.pdf GitHub

Abstract:
Recent advances in NeRF and 3DGS have significantly enhanced the efficiency and quality of 3D content synthesis. However, efficient personalization of generated 3D content remains a critical challenge. Current 3D personalization approaches predominantly rely on knowledge distillation-based methods, which require computationally expensive retraining procedures. To address this challenge, we propose \textbf{Invert3D}, a novel framework for convenient 3D content personalization. Nowadays, vision-language models such as CLIP enable direct image personalization through aligned vision-text embedding spaces. However, the inherent structural differences between 3D content and 2D images preclude direct application of these techniques to 3D personalization. Our approach bridges this gap by establishing alignment between 3D representations and text embedding spaces. Specifically, we develop a camera-conditioned 3D-to-text inverse mechanism that projects 3D contents into a 3D embedding aligned with text embeddings. This alignment enables efficient manipulation and personalization of 3D content through natural language prompts, eliminating the need for computationally retraining procedures. Extensive experiments demonstrate that Invert3D achieves effective personalization of 3D content. Our work is available at: https://github.com/qsong2001/Invert3D.

Paperid: 604, https://arxiv.org/pdf/2508.16762.pdf GitHub

Abstract:
As Vision-Language Models (VLMs) achieve widespread deployment across diverse cultural contexts, ensuring their cultural competence becomes critical for responsible AI systems. While prior work has evaluated cultural awareness in text-only models and VLM object recognition tasks, no research has systematically assessed how VLMs adapt outputs when cultural identity cues are embedded in both textual prompts and visual inputs during generative tasks. We present the first comprehensive evaluation of VLM cultural competence through multimodal story generation, developing a novel multimodal framework that perturbs cultural identity and evaluates 5 contemporary VLMs on a downstream task: story generation. Our analysis reveals significant cultural adaptation capabilities, with rich culturally-specific vocabulary spanning names, familial terms, and geographic markers. However, we uncover concerning limitations: cultural competence varies dramatically across architectures, some models exhibit inverse cultural alignment, and automated metrics show architectural bias contradicting human assessments. Cross-modal evaluation shows that culturally distinct outputs are indeed detectable through visual-semantic similarity (28.7% within-nationality vs. 0.2% cross-nationality recall), yet visual-cultural understanding remains limited. In essence, we establish the promise and challenges of cultural competence in multimodal AI. We publicly release our codebase and data: https://github.com/ArkaMukherjee0/mmCultural

Paperid: 605, https://arxiv.org/pdf/2508.16652.pdf GitHub

Abstract:
Vision-Language Models (VLMs) have shown remarkable performance in integrating visual and textual information for tasks such as image captioning and visual question answering. However, these models struggle with compositional generalization and object binding, which limit their ability to handle novel combinations of objects and their attributes. Our work explores the root causes of these failures using mechanistic interpretability techniques. We show evidence that individual neurons in the MLP layers of CLIP's vision encoder represent multiple features, and this "superposition" directly hinders its compositional feature representation which consequently affects compositional reasoning and object binding capabilities. We hope this study will serve as an initial step toward uncovering the mechanistic roots of compositional failures in VLMs. The code and supporting results can be found https://github.com/Mystic-Slice/Do-VLMs-Have-Bad-Eyes.

Paperid: 606, https://arxiv.org/pdf/2508.16463.pdf GitHub

Abstract:
The advent of pre-trained Vision-Language Models (VLMs) has significantly transformed Continual Learning (CL), mainly due to their zero-shot classification abilities. Such proficiency makes VLMs well-suited for real-world applications, enabling robust performance on novel unseen classes without requiring adaptation. However, fine-tuning remains essential when downstream tasks deviate significantly from the pre-training domain. Prior CL approaches primarily focus on preserving the zero-shot capabilities of VLMs during incremental fine-tuning on a downstream task. We take a step further by devising an approach that transforms preservation into enhancement of the zero-shot capabilities of VLMs. Our approach, named MoDular Embedding Recomposition (MoDER), introduces a modular framework that trains multiple textual experts, each specialized in a single seen class, and stores them in a foundational hub. At inference time, for each unseen class, we query the hub and compose the retrieved experts to synthesize a refined prototype that improves classification. We show the effectiveness of our method across two popular zero-shot incremental protocols, Class-IL and MTIL, comprising a total of 14 datasets. The codebase is available at https://github.com/aimagelab/mammoth.

Paperid: 607, https://arxiv.org/pdf/2508.15418.pdf GitHub

Abstract:
The development of Large Speech-Language Models (LSLMs) has been slowed by fragmented architectures and a lack of transparency, hindering the systematic comparison and reproducibility of research. Unlike in the vision-language domain, the LSLM field suffers from the common practice of releasing model weights without their corresponding training data and configurations. To address these critical gaps, we introduce LLaSO, the first fully open, end-to-end framework for large-scale speech-language modeling. LLaSO provides the community with three essential resources: (1) LLaSO-Align, a 12M-instance speech-text alignment corpus; (2) LLaSO-Instruct, a 13.5M-instance multi-task instruction-tuning dataset; and (3) LLaSO-Eval, a reproducible benchmark for standardized evaluation. To validate our framework, we build and release LLaSO-Base, a 3.8B-parameter reference model trained exclusively on our public data. It achieves a normalized score of 0.72, establishing a strong, reproducible baseline that surpasses comparable models. Our analysis reveals that while broader training coverage enhances performance, significant generalization gaps persist on unseen tasks, particularly in pure audio scenarios. By releasing the complete stack of data, benchmarks, and models, LLaSO establishes a foundational open standard to unify research efforts and accelerate community-driven progress in LSLMs. We release the code, dataset, pretrained models, and results in https://github.com/EIT-NLP/LLaSO.

Paperid: 608, https://arxiv.org/pdf/2508.15236.pdf GitHub

Abstract:
Anomaly detection is an emerging approach in digital pathology for its ability to efficiently and effectively utilize data for disease diagnosis. While supervised learning approaches deliver high accuracy, they rely on extensively annotated datasets, suffering from data scarcity in digital pathology. Unsupervised anomaly detection, however, offers a viable alternative by identifying deviations from normal tissue distributions without requiring exhaustive annotations. Recently, denoising diffusion probabilistic models have gained popularity in unsupervised anomaly detection, achieving promising performance in both natural and medical imaging datasets. Building on this, we incorporate a vision-language model with a diffusion model for unsupervised anomaly detection in digital pathology, utilizing histopathology prompts during reconstruction. Our approach employs a set of pathology-related keywords associated with normal tissues to guide the reconstruction process, facilitating the differentiation between normal and abnormal tissues. To evaluate the effectiveness of the proposed method, we conduct experiments on a gastric lymph node dataset from a local hospital and assess its generalization ability under domain shift using a public breast lymph node dataset. The experimental results highlight the potential of the proposed method for unsupervised anomaly detection across various organs in digital pathology. Code: https://github.com/QuIIL/AnoPILaD.

Paperid: 609, https://arxiv.org/pdf/2508.15222.pdf GitHub

Abstract:
We study sketch-to-diagram generation: converting rough hand sketches into precise, compositional diagrams. Diffusion models excel at photorealism but struggle with the spatial precision, alignment, and symbolic structure required for flowcharts. We introduce See it. Say it. Sorted., a training-free agentic system that couples a Vision-Language Model (VLM) with Large Language Models (LLMs) to produce editable Scalable Vector Graphics (SVG) programs. The system runs an iterative loop in which a Critic VLM proposes a small set of qualitative, relational edits; multiple candidate LLMs synthesize SVG updates with diverse strategies (conservative->aggressive, alternative, focused); and a Judge VLM selects the best candidate, ensuring stable improvement. This design prioritizes qualitative reasoning over brittle numerical estimates, preserves global constraints (e.g., alignment, connectivity), and naturally supports human-in-the-loop corrections. On 10 sketches derived from flowcharts in published papers, our method more faithfully reconstructs layout and structure than two frontier closed-source image generation LLMs (GPT-5 and Gemini-2.5-Pro), accurately composing primitives (e.g., multi-headed arrows) without inserting unwanted text. Because outputs are programmatic SVGs, the approach is readily extensible to presentation tools (e.g., PowerPoint) via APIs and can be specialized with improved prompts and task-specific tools. The codebase is open-sourced at https://github.com/hantaoZhangrichard/see_it_say_it_sorted.git.

Paperid: 610, https://arxiv.org/pdf/2508.14197.pdf GitHub

Abstract:
Symmetry is one of the most fundamental geometric cues in computer vision, and detecting it has been an ongoing challenge. With the recent advances in vision-language models,~i.e., CLIP, we investigate whether a pre-trained CLIP model can aid symmetry detection by leveraging the additional symmetry cues found in the natural image descriptions. We propose CLIPSym, which leverages CLIP's image and language encoders and a rotation-equivariant decoder based on a hybrid of Transformer and $G$-Convolution to detect rotation and reflection symmetries. To fully utilize CLIP's language encoder, we have developed a novel prompting technique called Semantic-Aware Prompt Grouping (SAPG), which aggregates a diverse set of frequent object-based prompts to better integrate the semantic cues for symmetry detection. Empirically, we show that CLIPSym outperforms the current state-of-the-art on three standard symmetry detection datasets (DENDI, SDRW, and LDRS). Finally, we conduct detailed ablations verifying the benefits of CLIP's pre-training, the proposed equivariant decoder, and the SAPG technique. The code is available at https://github.com/timyoung2333/CLIPSym.

Paperid: 611, https://arxiv.org/pdf/2508.14160.pdf GitHub

Abstract:
We introduce RynnEC, a video multimodal large language model designed for embodied cognition. Built upon a general-purpose vision-language foundation model, RynnEC incorporates a region encoder and a mask decoder, enabling flexible region-level video interaction. Despite its compact architecture, RynnEC achieves state-of-the-art performance in object property understanding, object segmentation, and spatial reasoning. Conceptually, it offers a region-centric video paradigm for the brain of embodied agents, providing fine-grained perception of the physical world and enabling more precise interactions. To mitigate the scarcity of annotated 3D datasets, we propose an egocentric video based pipeline for generating embodied cognition data. Furthermore, we introduce RynnEC-Bench, a region-centered benchmark for evaluating embodied cognitive capabilities. We anticipate that RynnEC will advance the development of general-purpose cognitive cores for embodied agents and facilitate generalization across diverse embodied tasks. The code, model checkpoints, and benchmark are available at: https://github.com/alibaba-damo-academy/RynnEC

Paperid: 612, https://arxiv.org/pdf/2508.14153.pdf GitHub GitHub

Abstract:
Text-prompted image segmentation enables fine-grained visual understanding and is critical for applications such as human-computer interaction and robotics. However, existing supervised fine-tuning methods typically ignore explicit chain-of-thought (CoT) reasoning at test time, which limits their ability to generalize to unseen prompts and domains. To address this issue, we introduce LENS, a scalable reinforcement-learning framework that jointly optimizes the reasoning process and segmentation in an end-to-end manner. We propose unified reinforcement-learning rewards that span sentence-, box-, and segment-level cues, encouraging the model to generate informative CoT rationales while refining mask quality. Using a publicly available 3-billion-parameter vision-language model, i.e., Qwen2.5-VL-3B-Instruct, LENS achieves an average cIoU of 81.2% on the RefCOCO, RefCOCO+, and RefCOCOg benchmarks, outperforming the strong fine-tuned method, i.e., GLaMM, by up to 5.6%. These results demonstrate that RL-driven CoT reasoning serves as a robust prior for text-prompted segmentation and offers a practical path toward more generalizable Segment Anything models. Code is available at https://github.com/hustvl/LENS.

Paperid: 613, https://arxiv.org/pdf/2508.13744.pdf GitHub

Abstract:
Large Vision-Language Models (LVLMs) demonstrate strong performance on single-image tasks. However, we observe that their performance degrades significantly when handling multi-image inputs. This occurs because visual cues from different images become entangled in the model's output. We refer to this phenomenon as cross-image information leakage. To address this issue, we propose FOCUS, a training-free and architecture-agnostic decoding strategy that mitigates cross-image information leakage during inference. FOCUS sequentially masks all but one image with random noise, guiding the model to focus on the single clean image. We repeat this process across all target images to obtain logits under partially masked contexts. These logits are aggregated and then contrastively refined using a noise-only reference input, which suppresses the leakage and yields more accurate outputs. FOCUS consistently improves performance across four multi-image benchmarks and diverse LVLM families. This demonstrates that FOCUS offers a general and practical solution for enhancing multi-image reasoning without additional training or architectural modifications.

Paperid: 614, https://arxiv.org/pdf/2508.13713.pdf GitHub

Abstract:
Every day, a large amount of educational content is uploaded online across different areas, including agriculture and gardening. When these videos or materials are grouped meaningfully, they can make learning easier and more effective. One promising way to organize and enrich such content is through the Metaverse, which allows users to explore educational experiences in an interactive and immersive environment. However, searching for relevant Metaverse scenarios and finding those matching users' interests remains a challenging task. A first step in this direction has been done recently, but existing datasets are small and not sufficient for training advanced models. In this work, we make two main contributions: first, we introduce a new dataset containing 457 agricultural-themed virtual museums (AgriMuseums), each enriched with textual descriptions; and second, we propose a hierarchical vision-language model to represent and retrieve relevant AgriMuseums using natural language queries. In our experimental setting, the proposed method achieves up to about 62\% R@1 and 78\% MRR, confirming its effectiveness, and it also leads to improvements on existing benchmarks by up to 6\% R@1 and 11\% MRR. Moreover, an extensive evaluation validates our design choices. Code and dataset are available at https://github.com/aliabdari/Agricultural_Metaverse_Retrieval .

Paperid: 615, https://arxiv.org/pdf/2508.13560.pdf GitHub

Abstract:
Recent vision-language models (e.g., CLIP) have demonstrated remarkable class-generalizable ability to unseen classes in few-shot anomaly segmentation (FSAS), leveraging supervised prompt learning or fine-tuning on seen classes. However, their cross-category generalization largely depends on prior knowledge of real seen anomaly samples. In this paper, we propose a novel framework, namely DictAS, which enables a unified model to detect visual anomalies in unseen object categories without any retraining on the target data, only employing a few normal reference images as visual prompts. The insight behind DictAS is to transfer dictionary lookup capabilities to the FSAS task for unseen classes via self-supervised learning, instead of merely memorizing the normal and abnormal feature patterns from the training set. Specifically, DictAS mainly consists of three components: (1) Dictionary Construction - to simulate the index and content of a real dictionary using features from normal reference images. (2) Dictionary Lookup - to retrieve queried region features from the dictionary via a sparse lookup strategy. When a query feature cannot be retrieved, it is classified as an anomaly. (3) Query Discrimination Regularization - to enhance anomaly discrimination by making abnormal features harder to retrieve from the dictionary. To achieve this, Contrastive Query Constraint and Text Alignment Constraint are further proposed. Extensive experiments on seven public industrial and medical datasets demonstrate that DictAS consistently outperforms state-of-the-art FSAS methods.

Paperid: 616, https://arxiv.org/pdf/2508.13229.pdf GitHub

Abstract:
Vision-Language Models (VLMs) struggle with complex image annotation tasks, such as emotion classification and context-driven object detection, which demand sophisticated reasoning. Standard Supervised Fine-Tuning (SFT) focuses solely on annotation outcomes, ignoring underlying rationales, while Visual Reinforcement Fine-Tuning (Visual-RFT) produces inconsistent Chains of Thought (CoTs) due to the absence of high-quality, verified CoTs during pre-training. We introduce RISE (Reason-Inspire-Strengthen-Expertise), a two-stage framework to overcome these limitations. In the Reason stage (RISE-CoT), a reinforcement learning-driven "annotation-reasoning-annotation" closed-loop generates visually grounded, logically consistent CoTs by verifying their ability to reconstruct original annotations without direct leakage. The Inspire and Strengthen stage (RISE-R1) leverages a high-quality CoT subset, filtered by RISE-CoT rewards, for supervised fine-tuning, followed by reinforcement fine-tuning to produce interpretable reasoning and accurate annotations, achieving Expertise in complex visual tasks. Evaluated on complex and simple image annotation tasks, RISE-trained Qwen2-VL-2B outperforms SFT and Visual-RFT, achieving robust performance and enhanced explainability. RISE offers a self-supervised solution for advancing VLM reasoning without requiring manually annotated CoTs.Code and resources are available at: https://github.com/HSH55/RISE.

Paperid: 617, https://arxiv.org/pdf/2508.13073.pdf GitHub

Abstract:
Robotic manipulation, a key frontier in robotics and embodied AI, requires precise motor control and multimodal understanding, yet traditional rule-based methods fail to scale or generalize in unstructured, novel environments. In recent years, Vision-Language-Action (VLA) models, built upon Large Vision-Language Models (VLMs) pretrained on vast image-text datasets, have emerged as a transformative paradigm. This survey provides the first systematic, taxonomy-oriented review of large VLM-based VLA models for robotic manipulation. We begin by clearly defining large VLM-based VLA models and delineating two principal architectural paradigms: (1) monolithic models, encompassing single-system and dual-system designs with differing levels of integration; and (2) hierarchical models, which explicitly decouple planning from execution via interpretable intermediate representations. Building on this foundation, we present an in-depth examination of large VLM-based VLA models: (1) integration with advanced domains, including reinforcement learning, training-free optimization, learning from human videos, and world model integration; (2) synthesis of distinctive characteristics, consolidating architectural traits, operational strengths, and the datasets and benchmarks that support their development; (3) identification of promising directions, including memory mechanisms, 4D perception, efficient adaptation, multi-agent cooperation, and other emerging capabilities. This survey consolidates recent advances to resolve inconsistencies in existing taxonomies, mitigate research fragmentation, and fill a critical gap through the systematic integration of studies at the intersection of large VLMs and robotic manipulation. We provide a regularly updated project page to document ongoing progress: https://github.com/JiuTian-VL/Large-VLM-based-VLA-for-Robotic-Manipulation

Paperid: 618, https://arxiv.org/pdf/2508.13043.pdf GitHub

Abstract:
Novel view synthesis from images, for example, with 3D Gaussian splatting, has made great progress. Rendering fidelity and speed are now ready even for demanding virtual reality applications. However, the problem of assisting humans in collecting the input images for these rendering algorithms has received much less attention. High-quality view synthesis requires uniform and dense view sampling. Unfortunately, these requirements are not easily addressed by human camera operators, who are in a hurry, impatient, or lack understanding of the scene structure and the photographic process. Existing approaches to guide humans during image acquisition concentrate on single objects or neglect view-dependent material characteristics. We propose a novel situated visualization technique for scanning at multiple scales. During the scanning of a scene, our method identifies important objects that need extended image coverage to properly represent view-dependent appearance. To this end, we leverage semantic segmentation and category identification, ranked by a vision-language model. Spherical proxies are generated around highly ranked objects to guide the user during scanning. Our results show superior performance in real scenes compared to conventional view sampling strategies.

Paperid: 619, https://arxiv.org/pdf/2508.12680.pdf GitHub

Abstract:
Despite their success, current training pipelines for reasoning VLMs focus on a limited range of tasks, such as mathematical and logical reasoning. As a result, these models face difficulties in generalizing their reasoning capabilities to a wide range of domains, primarily due to the scarcity of readily available and verifiable reward data beyond these narrowly defined areas. Moreover, integrating data from multiple domains is challenging, as the compatibility between domain-specific datasets remains uncertain. To address these limitations, we build a comprehensive RL-ready visual reasoning dataset from 46 data sources across 8 dimensions, covering a wide range of tasks such as infographic, mathematical, spatial, cross-image, graphic user interface, medical, common sense and general science. We propose an influence function based data selection and difficulty based filtering strategy to identify high-quality training samples from this dataset. Subsequently, we train the VLM, referred to as Vision-G1, using multi-round RL with a data curriculum to iteratively improve its visual reasoning capabilities. Our model achieves state-of-the-art performance across various visual reasoning benchmarks, outperforming similar-sized VLMs and even proprietary models like GPT-4o and Gemini-1.5 Flash. The model, code and dataset are publicly available at https://github.com/yuh-zha/Vision-G1.

Paperid: 620, https://arxiv.org/pdf/2508.12668.pdf GitHub

Abstract:
WÃ¶lfflin's five principles offer a structured approach to analyzing stylistic variations for formal analysis. However, no existing metric effectively predicts all five principles in visual art. Computationally evaluating the visual aspects of a painting requires a metric that can interpret key elements such as color, composition, and thematic choices. Recent advancements in vision-language models (VLMs) have demonstrated their ability to evaluate abstract image attributes, making them promising candidates for this task. In this work, we investigate whether CLIP, pre-trained on large-scale data, can understand and predict WÃ¶lfflin's principles. Our findings indicate that it does not inherently capture such nuanced stylistic elements. To address this, we fine-tune CLIP on annotated datasets of real art images to predict a score for each principle. We evaluate our model, WP-CLIP, on GAN-generated paintings and the Pandora-18K art dataset, demonstrating its ability to generalize across diverse artistic styles. Our results highlight the potential of VLMs for automated art analysis.

Paperid: 621, https://arxiv.org/pdf/2508.12587.pdf GitHub

Abstract:
Many reasoning techniques for large multimodal models adapt language model approaches, such as Chain-of-Thought (CoT) prompting, which express reasoning as word sequences. While effective for text, these methods are suboptimal for multimodal contexts, struggling to align audio, visual, and textual information dynamically. To explore an alternative paradigm, we propose the Multimodal Chain of Continuous Thought (MCOUT), which enables reasoning directly in a joint latent space rather than in natural language. In MCOUT, the reasoning state is represented as a continuous hidden vector, iteratively refined and aligned with visual and textual embeddings, inspired by human reflective cognition. We develop two variants: MCOUT-Base, which reuses the language model`s last hidden state as the continuous thought for iterative reasoning, and MCOUT-Multi, which integrates multimodal latent attention to strengthen cross-modal alignment between visual and textual features. Experiments on benchmarks including MMMU, ScienceQA, and MMStar show that MCOUT consistently improves multimodal reasoning, yielding up to 8.23% accuracy gains over strong baselines and improving BLEU scores up to 8.27% across multiple-choice and open-ended tasks. These findings highlight latent continuous reasoning as a promising direction for advancing LMMs beyond language-bound CoT, offering a scalable framework for human-like reflective multimodal inference. Code is available at https://github.com/Hanhpt23/OmniMod.

Paperid: 622, https://arxiv.org/pdf/2508.12512.pdf GitHub GitHub GitHub

Abstract:
Vision Language Models (VLMs) integrate visual and text modalities to enable multimodal understanding and generation. These models typically combine a Vision Transformer (ViT) as an image encoder and a Large Language Model (LLM) for text generation. LoRA (Low-Rank Adaptation) is an efficient fine-tuning method to adapt pre-trained models to new tasks by introducing low-rank updates to their weights. While LoRA has emerged as a powerful technique for fine-tuning large models by introducing low-rank updates, current implementations assume a fixed rank, potentially limiting flexibility and efficiency across diverse tasks. This paper introduces \textit{LangVision-LoRA-NAS}, a novel framework that integrates Neural Architecture Search (NAS) with LoRA to optimize VLMs for variable-rank adaptation. Our approach leverages NAS to dynamically search for the optimal LoRA rank configuration tailored to specific multimodal tasks, balancing performance and computational efficiency. Through extensive experiments using the LLaMA-3.2-11B model on several datasets, LangVision-LoRA-NAS demonstrates notable improvement in model performance while reducing fine-tuning costs. Our Base and searched fine-tuned models on LLaMA-3.2-11B-Vision-Instruct can be found \href{https://huggingface.co/collections/krishnateja95/llama-32-11b-vision-instruct-langvision-lora-nas-6786cac480357a6a6fcc59ee}{\textcolor{blue}{here}} and the code for LangVision-LoRA-NAS can be found \href{https://github.com/krishnateja95/LangVision-NAS}{\textcolor{blue}{here}}.

Paperid: 623, https://arxiv.org/pdf/2508.12349.pdf GitHub

Abstract:
Analyzing hand-object interaction in egocentric vision facilitates VR/AR applications and human-robot policy transfer. Existing research has mostly focused on modeling the behavior paradigm of interactive actions (i.e., ``how to interact''). However, the more challenging and fine-grained problem of capturing the critical moments of contact and separation between the hand and the target object (i.e., ``when to interact'') is still underexplored, which is crucial for immersive interactive experiences in mixed reality and robotic motion planning. Therefore, we formulate this problem as temporal interaction localization (TIL). Some recent works extract semantic masks as TIL references, but suffer from inaccurate object grounding and cluttered scenarios. Although current temporal action localization (TAL) methods perform well in detecting verb-noun action segments, they rely on category annotations during training and exhibit limited precision in localizing hand-object contact/separation moments. To address these issues, we propose a novel zero-shot approach dubbed EgoLoc to localize hand-object contact and separation timestamps in egocentric videos. EgoLoc introduces hand-dynamics-guided sampling to generate high-quality visual prompts. It exploits the vision-language model to identify contact/separation attributes, localize specific timestamps, and provide closed-loop feedback for further refinement. EgoLoc eliminates the need for object masks and verb-noun taxonomies, leading to generalizable zero-shot implementation. Comprehensive experiments on the public dataset and our novel benchmarks demonstrate that EgoLoc achieves plausible TIL for egocentric videos. It is also validated to effectively facilitate multiple downstream applications in egocentric vision and robotic manipulation tasks. Code and relevant data will be released at https://github.com/IRMVLab/EgoLoc.

Paperid: 624, https://arxiv.org/pdf/2508.12341.pdf GitHub

Abstract:
With the rapid advancement of image generation techniques, robust forgery detection has become increasingly imperative to ensure the trustworthiness of digital media. Recent research indicates that the learned semantic concepts of pre-trained models are critical for identifying fake images. However, the misalignment between the forgery and semantic concept spaces hinders the model's forgery detection performance. To address this problem, we propose a novel Semantic Discrepancy-aware Detector (SDD) that leverages reconstruction learning to align the two spaces at a fine-grained visual level. By exploiting the conceptual knowledge embedded in the pre-trained vision language model, we specifically design a semantic token sampling module to mitigate the space shifts caused by features irrelevant to both forgery traces and semantic concepts. A concept-level forgery discrepancy learning module, built upon a visual reconstruction paradigm, is proposed to strengthen the interaction between visual semantic concepts and forgery traces, effectively capturing discrepancies under the concepts' guidance. Finally, the low-level forgery feature enhancemer integrates the learned concept level forgery discrepancies to minimize redundant forgery information. Experiments conducted on two standard image forgery datasets demonstrate the efficacy of the proposed SDD, which achieves superior results compared to existing methods. The code is available at https://github.com/wzy1111111/SSD.

Paperid: 625, https://arxiv.org/pdf/2508.12137.pdf GitHub

Abstract:
Large-scale contrastive pre-training produces powerful Vision-and-Language Models (VLMs) capable of generating representations (embeddings) effective for a wide variety of visual and multimodal tasks. However, these pretrained embeddings remain suboptimal for fine-grained open-set visual retrieval, where state-of-the-art results require fine-tuning the vision encoder using annotated domain-specific samples. Naively performing such fine-tuning typically leads to catastrophic forgetting, severely diminishing the model's general-purpose visual and cross-modal capabilities. In this work, we propose a fine-tuning method explicitly designed to achieve optimal balance between fine-grained domain adaptation and retention of the pretrained VLM's broad multimodal knowledge. Drawing inspiration from continual learning literature, we systematically analyze standard regularization techniques aimed at knowledge retention and propose an efficient and effective combination strategy. Additionally, we address the commonly overlooked yet critical aspects of validation set design and hyperparameter tuning to ensure reproducibility and robust generalization across datasets and pretrained models. We extensively evaluate our method on both fine-grained and coarse-grained image-image and image-text retrieval benchmarks. Our approach consistently achieves strong results, notably retaining the visual-text alignment without utilizing any text data or the original text encoder during fine-tuning. Code and model checkpoints: https://github.com/nikosips/infusing .

Paperid: 626, https://arxiv.org/pdf/2508.11919.pdf GitHub

Abstract:
Vision-language models have shown significant promise in remote sensing applications, particularly for land-use and land-cover (LULC) via zero-shot classification and retrieval. However, current approaches face two key challenges: reliance on large spatial tiles that increase computational cost, and dependence on text-based supervision, which is often not readily available. In this work, we present TimeSenCLIP, a lightweight framework that reevaluate the role of spatial context by evaluating the effectiveness of a single pixel by leveraging its temporal and spectral dimensions, for classifying LULC and ecosystem types. By leveraging spectral and temporal information from Sentinel-2 imagery and cross-view learning with geo-tagged ground-level photos, we minimises the need for caption-based training while preserving semantic alignment between overhead (satellite) and ground perspectives. Our approach is grounded in the LUCAS and Sen4Map datasets, and evaluated on classification tasks including LULC, crop type, and ecosystem type. We demonstrate that single pixel inputs, when combined with temporal and spectral cues, are sufficient for thematic mapping, offering a scalable and efficient alternative for large-scale remote sensing applications. Code is available at https://github.com/pallavijain-pj/TimeSenCLIP

Paperid: 627, https://arxiv.org/pdf/2508.11801.pdf GitHub

Abstract:
Attribute Value Extraction (AVE) is important for structuring product information in e-commerce. However, existing AVE datasets are primarily limited to text-to-text or image-to-text settings, lacking support for product videos, diverse attribute coverage, and public availability. To address these gaps, we introduce VideoAVE, the first publicly available video-to-text e-commerce AVE dataset across 14 different domains and covering 172 unique attributes. To ensure data quality, we propose a post-hoc CLIP-based Mixture of Experts filtering system (CLIP-MoE) to remove the mismatched video-product pairs, resulting in a refined dataset of 224k training data and 25k evaluation data. In order to evaluate the usability of the dataset, we further establish a comprehensive benchmark by evaluating several state-of-the-art video vision language models (VLMs) under both attribute-conditioned value prediction and open attribute-value pair extraction tasks. Our results analysis reveals that video-to-text AVE remains a challenging problem, particularly in open settings, and there is still room for developing more advanced VLMs capable of leveraging effective temporal information. The dataset and benchmark code for VideoAVE are available at: https://github.com/gjiaying/VideoAVE

Paperid: 628, https://arxiv.org/pdf/2508.11673.pdf GitHub

Abstract:
Multimodal Biomedical Image Incremental Learning (MBIIL) is essential for handling diverse tasks and modalities in the biomedical domain, as training separate models for each modality or task significantly increases inference costs. Existing incremental learning methods focus on task expansion within a single modality, whereas MBIIL seeks to train a unified model incrementally across modalities. The MBIIL faces two challenges: I) How to preserve previously learned knowledge during incremental updates? II) How to effectively leverage knowledge acquired from existing modalities to support new modalities? To address these challenges, we propose MSLoRA-CR, a method that fine-tunes Modality-Specific LoRA modules while incorporating Contrastive Regularization to enhance intra-modality knowledge sharing and promote inter-modality knowledge differentiation. Our approach builds upon a large vision-language model (LVLM), keeping the pretrained model frozen while incrementally adapting new LoRA modules for each modality or task. Experiments on the incremental learning of biomedical images demonstrate that MSLoRA-CR outperforms both the state-of-the-art (SOTA) approach of training separate models for each modality and the general incremental learning method (incrementally fine-tuning LoRA). Specifically, MSLoRA-CR achieves a 1.88% improvement in overall performance compared to unconstrained incremental learning methods while maintaining computational efficiency. Our code is publicly available at https://github.com/VentusAislant/MSLoRA_CR.

Paperid: 629, https://arxiv.org/pdf/2508.11479.pdf GitHub

Abstract:
Open-vocabulary Object Goal Navigation requires an embodied agent to reach objects described by free-form language, including categories never seen during training. Existing end-to-end policies overfit small simulator datasets, achieving high success on training scenes but failing to generalize and exhibiting unsafe behaviour (frequent collisions). We introduce OVSegDT, a lightweight transformer policy that tackles these issues with two synergistic components. The first component is the semantic branch, which includes an encoder for the target binary mask and an auxiliary segmentation loss function, grounding the textual goal and providing precise spatial cues. The second component consists of a proposed Entropy-Adaptive Loss Modulation, a per-sample scheduler that continuously balances imitation and reinforcement signals according to the policy entropy, eliminating brittle manual phase switches. These additions cut the sample complexity of training by 33%, and reduce collision count in two times while keeping inference cost low (130M parameters, RGB-only input). On HM3D-OVON, our model matches the performance on unseen categories to that on seen ones and establishes state-of-the-art results (40.1% SR, 20.9% SPL on val unseen) without depth, odometry, or large vision-language models. Code is available at https://github.com/CognitiveAISystems/OVSegDT.

Paperid: 630, https://arxiv.org/pdf/2508.11256.pdf GitHub

Abstract:
Dense visual perception tasks have been constrained by their reliance on predefined categories, limiting their applicability in real-world scenarios where visual concepts are unbounded. While Vision-Language Models (VLMs) like CLIP have shown promise in open-vocabulary tasks, their direct application to dense perception often leads to suboptimal performance due to limitations in local feature representation. In this work, we present our observation that CLIP's image tokens struggle to effectively aggregate information from spatially or semantically related regions, resulting in features that lack local discriminability and spatial consistency. To address this issue, we propose DeCLIP, a novel framework that enhances CLIP by decoupling the self-attention module to obtain ``content'' and ``context'' features respectively. \revise{The context features are enhanced by jointly distilling semantic correlations from Vision Foundation Models (VFMs) and object integrity cues from diffusion models, thereby enhancing spatial consistency. In parallel, the content features are aligned with image crop representations and constrained by region correlations from VFMs to improve local discriminability. Extensive experiments demonstrate that DeCLIP establishes a solid foundation for open-vocabulary dense perception, consistently achieving state-of-the-art performance across a broad spectrum of tasks, including 2D detection and segmentation, 3D instance segmentation, video instance segmentation, and 6D object pose estimation.} Code is available at https://github.com/xiaomoguhz/DeCLIP

Paperid: 631, https://arxiv.org/pdf/2508.11058.pdf GitHub

Abstract:
The advancement of 3D vision-language (3D VL) learning is hindered by several limitations in existing 3D VL datasets: they rarely necessitate reasoning beyond a close range of objects in single viewpoint, and annotations often link instructions to single objects, missing richer contextual alignments between multiple objects. This significantly curtails the development of models capable of deep, multi-view 3D scene understanding over distant objects. To address these challenges, we introduce MV-ScanQA, a novel 3D question answering dataset where 68% of questions explicitly require integrating information from multiple views (compared to less than 7% in existing datasets), thereby rigorously testing multi-view compositional reasoning. To facilitate the training of models for such demanding scenarios, we present TripAlign dataset, a large-scale and low-cost 2D-3D-language pre-training corpus containing 1M <2D view, set of 3D objects, text> triplets that explicitly aligns groups of contextually related objects with text, providing richer, view-grounded multi-object multimodal alignment signals than previous single-object annotations. We further develop LEGO, a baseline method for the multi-view reasoning challenge in MV-ScanQA, transferring knowledge from pre-trained 2D LVLMs to 3D domain with TripAlign. Empirically, LEGO pre-trained on TripAlign achieves state-of-the-art performance not only on the proposed MV-ScanQA, but also on existing benchmarks for 3D dense captioning and question answering. Datasets and code are available at https://matthewdm0816.github.io/tripalign-mvscanqa.

Paperid: 632, https://arxiv.org/pdf/2508.10427.pdf GitHub

Abstract:
Vision-Language Models (VLMs) have been applied to autonomous driving to support decision-making in complex real-world scenarios. However, their training on static, web-sourced image-text pairs fundamentally limits the precise spatiotemporal reasoning required to understand and predict dynamic traffic scenes. We address this critical gap with STRIDE-QA, a large-scale visual question answering (VQA) dataset for physically grounded reasoning from an ego-centric perspective. Constructed from 100 hours of multi-sensor driving data in Tokyo, capturing diverse and challenging conditions, STRIDE-QA is the largest VQA dataset for spatiotemporal reasoning in urban driving, offering 16 million QA pairs over 285K frames. Grounded by dense, automatically generated annotations including 3D bounding boxes, segmentation masks, and multi-object tracks, the dataset uniquely supports both object-centric and ego-centric reasoning through three novel QA tasks that require spatial localization and temporal prediction. Our benchmarks demonstrate that existing VLMs struggle significantly, achieving near-zero scores on prediction consistency. In contrast, VLMs fine-tuned on STRIDE-QA exhibit dramatic performance gains, achieving 55% success in spatial localization and 28% consistency in future motion prediction, compared to near-zero scores from general-purpose VLMs. Therefore, STRIDE-QA establishes a comprehensive foundation for developing more reliable VLMs for safety-critical autonomous systems.

Paperid: 633, https://arxiv.org/pdf/2508.10333.pdf GitHub

Abstract:
Recent advances in Vision-Language-Action (VLA) models have enabled robotic agents to integrate multimodal understanding with action execution. However, our empirical analysis reveals that current VLAs struggle to allocate visual attention to target regions. Instead, visual attention is always dispersed. To guide the visual attention grounding on the correct target, we propose ReconVLA, a reconstructive VLA model with an implicit grounding paradigm. Conditioned on the model's visual outputs, a diffusion transformer aims to reconstruct the gaze region of the image, which corresponds to the target manipulated objects. This process prompts the VLA model to learn fine-grained representations and accurately allocate visual attention, thus effectively leveraging task-specific visual information and conducting precise manipulation. Moreover, we curate a large-scale pretraining dataset comprising over 100k trajectories and 2 million data samples from open-source robotic datasets, further boosting the model's generalization in visual reconstruction. Extensive experiments in simulation and the real world demonstrate the superiority of our implicit grounding method, showcasing its capabilities of precise manipulation and generalization. Our project page is https://zionchow.github.io/ReconVLA/.

Paperid: 634, https://arxiv.org/pdf/2508.10113.pdf GitHub

Abstract:
As the oldest mature writing system, Oracle Bone Script (OBS) has long posed significant challenges for archaeological decipherment due to its rarity, abstractness, and pictographic diversity. Current deep learning-based methods have made exciting progress on the OBS decipherment task, but existing approaches often ignore the intricate connections between glyphs and the semantics of OBS. This results in limited generalization and interpretability, especially when addressing zero-shot settings and undeciphered OBS. To this end, we propose an interpretable OBS decipherment method based on Large Vision-Language Models, which synergistically combines radical analysis and pictograph-semantic understanding to bridge the gap between glyphs and meanings of OBS. Specifically, we propose a progressive training strategy that guides the model from radical recognition and analysis to pictographic analysis and mutual analysis, thus enabling reasoning from glyph to meaning. We also design a Radical-Pictographic Dual Matching mechanism informed by the analysis results, significantly enhancing the model's zero-shot decipherment performance. To facilitate model training, we propose the Pictographic Decipherment OBS Dataset, which comprises 47,157 Chinese characters annotated with OBS images and pictographic analysis texts. Experimental results on public benchmarks demonstrate that our approach achieves state-of-the-art Top-10 accuracy and superior zero-shot decipherment capabilities. More importantly, our model delivers logical analysis processes, possibly providing archaeologically valuable reference results for undeciphered OBS, and thus has potential applications in digital humanities and historical research. The dataset and code will be released in https://github.com/PKXX1943/PD-OBS.

Paperid: 635, https://arxiv.org/pdf/2508.09981.pdf GitHub

Abstract:
Large Vision-Language Models (VLMs) exhibit impressive multi-modal capabilities but suffer from prohibitive computational and memory demands, due to their long visual token sequences and massive parameter sizes. To address these issues, recent works have proposed training-free compression methods. However, existing efforts often suffer from three major limitations: (1) Current approaches do not decompose techniques into comparable modules, hindering fair evaluation across spatial and temporal redundancy. (2) Evaluation confined to simple single-turn tasks, failing to reflect performance in realistic scenarios. (3) Isolated use of individual compression techniques, without exploring their joint potential. To overcome these gaps, we introduce LLMC+, a comprehensive VLM compression benchmark with a versatile, plug-and-play toolkit. LLMC+ supports over 20 algorithms across five representative VLM families and enables systematic study of token-level and model-level compression. Our benchmark reveals that: (1) Spatial and temporal redundancies demand distinct technical strategies. (2) Token reduction methods degrade significantly in multi-turn dialogue and detail-sensitive tasks. (3) Combining token and model compression achieves extreme compression with minimal performance loss. We believe LLMC+ will facilitate fair evaluation and inspire future research in efficient VLM. Our code is available at https://github.com/ModelTC/LightCompress.

Paperid: 636, https://arxiv.org/pdf/2508.09967.pdf GitHub

Abstract:
Recent advances in histopathology vision-language foundation models (VLFMs) have shown promise in addressing data scarcity for whole slide image (WSI) classification via zero-shot adaptation. However, these methods remain outperformed by conventional multiple instance learning (MIL) approaches trained on large datasets, motivating recent efforts to enhance VLFM-based WSI classification through fewshot learning paradigms. While existing few-shot methods improve diagnostic accuracy with limited annotations, their reliance on conventional classifier designs introduces critical vulnerabilities to data scarcity. To address this problem, we propose a Meta-Optimized Classifier (MOC) comprising two core components: (1) a meta-learner that automatically optimizes a classifier configuration from a mixture of candidate classifiers and (2) a classifier bank housing diverse candidate classifiers to enable a holistic pathological interpretation. Extensive experiments demonstrate that MOC outperforms prior arts in multiple few-shot benchmarks. Notably, on the TCGA-NSCLC benchmark, MOC improves AUC by 10.4% over the state-of-the-art few-shot VLFM-based methods, with gains up to 26.25% under 1-shot conditions, offering a critical advancement for clinical deployments where diagnostic training data is severely limited. Code is available at https://github.com/xmed-lab/MOC.

Paperid: 637, https://arxiv.org/pdf/2508.09779.pdf GitHub

Abstract:
Large Vision-Language Models (LVLMs) have demonstrated remarkable performance across multi-modal tasks by scaling model size and training data. However, these dense LVLMs incur significant computational costs and motivate the exploration of sparse Mixture of Experts (MoE) architectures. While MoE improve parameter efficiency, effectively applying MoE to simultaneously model modality-specific features and cross-modal associations in LVLMs remains challenging. In this work, we propose to incorporate Mixture of Intra- and Inter-Modality Experts (MoIIE) to LVLMs. For each token, expert routing is guided by its modality, directing tokens to their respective intra-modality experts as well as a shared pool of inter-modality experts, enabling the model to jointly learn rich intra-modal features and cross-modal interactions. We further introduce an effective and straightforward two-stage training strategy, which facilitates the direct activation of both MoE and multi-modal capabilities. Extensive experiments across different data scales and LLM backbone demonstrate the effectiveness, efficiency and generality of our approach. Notably, our MoIIE models with 5.5B and 11.3B activated parameters match or even surpass the performance of existing advanced open-source MoE-LLMs based multi-modal models that involve more activated parameters. The code is available at https://github.com/AlenjandroWang/MoIIE.

Paperid: 638, https://arxiv.org/pdf/2508.09715.pdf GitHub

Abstract:
The rapid growth of multimodal medical imaging data presents significant storage and transmission challenges, particularly in resource-constrained clinical settings. We propose NEURAL, a novel framework that addresses this by using semantics-guided data compression. Our approach repurposes cross-attention scores between the image and its radiological report from a fine-tuned generative vision-language model to structurally prune chest X-rays, preserving only diagnostically critical regions. This process transforms the image into a highly compressed, graph representation. This unified graph-based representation fuses the pruned visual graph with a knowledge graph derived from the clinical report, creating a universal data structure that simplifies downstream modeling. Validated on the MIMIC-CXR and CheXpert Plus dataset for pneumonia detection, NEURAL achieves a 93.4-97.7\% reduction in image data size while maintaining a high diagnostic performance of 0.88-0.95 AUC, outperforming other baseline models that use uncompressed data. By creating a persistent, task-agnostic data asset, NEURAL resolves the trade-off between data size and clinical utility, enabling efficient workflows and teleradiology without sacrificing performance. Our NEURAL code is available at https://github.com/basiralab/NEURAL.

Paperid: 639, https://arxiv.org/pdf/2508.09444.pdf GitHub

Abstract:
Vision-Language Navigation in Continuous Environments (VLN-CE) requires agents to follow natural language instructions through free-form 3D spaces. Existing VLN-CE approaches typically use a two-stage waypoint planning framework, where a high-level waypoint predictor generates the navigable waypoints, and then a navigation planner suggests the intermediate goals in the high-level action space. However, this two-stage decomposition framework suffers from: (1) global sub-optimization due to the proxy objective in each stage, and (2) a performance bottleneck caused by the strong reliance on the quality of the first-stage predicted waypoints. To address these limitations, we propose DAgger Diffusion Navigation (DifNav), an end-to-end optimized VLN-CE policy that unifies the traditional two stages, i.e. waypoint generation and planning, into a single diffusion policy. Notably, DifNav employs a conditional diffusion policy to directly model multi-modal action distributions over future actions in continuous navigation space, eliminating the need for a waypoint predictor while enabling the agent to capture multiple possible instruction-following behaviors. To address the issues of compounding error in imitation learning and enhance spatial reasoning in long-horizon navigation tasks, we employ DAgger for online policy training and expert trajectory augmentation, and use the aggregated data to further fine-tune the policy. This approach significantly improves the policy's robustness and its ability to recover from error states. Extensive experiments on benchmark datasets demonstrate that, even without a waypoint predictor, the proposed method substantially outperforms previous state-of-the-art two-stage waypoint-based models in terms of navigation performance. Our code is available at: https://github.com/Tokishx/DifNav.

Paperid: 640, https://arxiv.org/pdf/2508.09205.pdf GitHub

Abstract:
Explaining deep learning models is essential for clinical integration of medical image analysis systems. A good explanation highlights if a model depends on spurious features that undermines generalization and harms a subset of patients or, conversely, may present novel biological insights. Although techniques like GradCAM can identify influential features, they are measurement tools that do not themselves form an explanation. We propose a human-machine-VLM interaction system tailored to explaining classifiers in computational pathology, including multi-instance learning for whole-slide images. Our proof of concept comprises (1) an AI-integrated slide viewer to run sliding-window experiments to test claims of an explanation, and (2) quantification of an explanation's predictiveness using general-purpose vision-language models. The results demonstrate that this allows us to qualitatively test claims of explanations and can quantifiably distinguish competing explanations. This offers a practical path from explainable AI to explained AI in digital pathology and beyond. Code and prompts are available at https://github.com/nki-ai/x2x.

Paperid: 641, https://arxiv.org/pdf/2508.09178.pdf GitHub

Abstract:
Industrial anomaly detection is a critical component of modern manufacturing, yet the scarcity of defective samples restricts traditional detection methods to scenario-specific applications. Although Vision-Language Models (VLMs) demonstrate significant advantages in generalization capabilities, their performance in industrial anomaly detection remains limited. To address this challenge, we propose IAD-R1, a universal post-training framework applicable to VLMs of different architectures and parameter scales, which substantially enhances their anomaly detection capabilities. IAD-R1 employs a two-stage training strategy: the Perception Activation Supervised Fine-Tuning (PA-SFT) stage utilizes a meticulously constructed high-quality Chain-of-Thought dataset (Expert-AD) for training, enhancing anomaly perception capabilities and establishing reasoning-to-answer correlations; the Structured Control Group Relative Policy Optimization (SC-GRPO) stage employs carefully designed reward functions to achieve a capability leap from "Anomaly Perception" to "Anomaly Interpretation". Experimental results demonstrate that IAD-R1 achieves significant improvements across 7 VLMs, the largest improvement was on the DAGM dataset, with average accuracy 43.3% higher than the 0.5B baseline. Notably, the 0.5B parameter model trained with IAD-R1 surpasses commercial models including GPT-4.1 and Claude-Sonnet-4 in zero-shot settings, demonstrating the effectiveness and superiority of IAD-R1. The dataset, code, and all model weights will be publicly available at https://github.com/Yanhui-Lee/IAD-R1.

Paperid: 642, https://arxiv.org/pdf/2508.09071.pdf GitHub

Abstract:
Vision-Language-Action (VLA) models have emerged as a promising approach for enabling robots to follow language instructions and predict corresponding actions. However, current VLA models mainly rely on 2D visual inputs, neglecting the rich geometric information in the 3D physical world, which limits their spatial awareness and adaptability. In this paper, we present GeoVLA, a novel VLA framework that effectively integrates 3D information to advance robotic manipulation. It uses a vision-language model (VLM) to process images and language instructions,extracting fused vision-language embeddings. In parallel, it converts depth maps into point clouds and employs a customized point encoder, called Point Embedding Network, to generate 3D geometric embeddings independently. These produced embeddings are then concatenated and processed by our proposed spatial-aware action expert, called 3D-enhanced Action Expert, which combines information from different sensor modalities to produce precise action sequences. Through extensive experiments in both simulation and real-world environments, GeoVLA demonstrates superior performance and robustness. It achieves state-of-the-art results in the LIBERO and ManiSkill2 simulation benchmarks and shows remarkable robustness in real-world tasks requiring height adaptability, scale awareness and viewpoint invariance.

Paperid: 643, https://arxiv.org/pdf/2508.09032.pdf GitHub

Abstract:
Vision-Language-Action models have demonstrated remarkable capabilities in predicting agent movements within virtual environments and real-world scenarios based on visual observations and textual instructions. Although recent research has focused on enhancing spatial and temporal understanding independently, this paper presents a novel approach that integrates both aspects through visual prompting. We introduce a method that projects visual traces of key points from observations onto depth maps, enabling models to capture both spatial and temporal information simultaneously. The experiments in SimplerEnv show that the mean number of tasks successfully solved increased for 4% compared to SpatialVLA and 19% compared to TraceVLA. Furthermore, we show that this enhancement can be achieved with minimal training data, making it particularly valuable for real-world applications where data collection is challenging. The project page is available at https://ampiromax.github.io/ST-VLA.

Paperid: 644, https://arxiv.org/pdf/2508.08706.pdf GitHub

Abstract:
Recent vision-language-action (VLA) models build upon vision-language foundations, and have achieved promising results and exhibit the possibility of task generalization in robot manipulation. However, due to the heterogeneity of tactile sensors and the difficulty of acquiring tactile data, current VLA models significantly overlook the importance of tactile perception and fail in contact-rich tasks. To address this issue, this paper proposes OmniVTLA, a novel architecture involving tactile sensing. Specifically, our contributions are threefold. First, our OmniVTLA features a dual-path tactile encoder framework. This framework enhances tactile perception across diverse vision-based and force-based tactile sensors by using a pretrained vision transformer (ViT) and a semantically-aligned tactile ViT (SA-ViT). Second, we introduce ObjTac, a comprehensive force-based tactile dataset capturing textual, visual, and tactile information for 56 objects across 10 categories. With 135K tri-modal samples, ObjTac supplements existing visuo-tactile datasets. Third, leveraging this dataset, we train a semantically-aligned tactile encoder to learn a unified tactile representation, serving as a better initialization for OmniVTLA. Real-world experiments demonstrate substantial improvements over state-of-the-art VLA baselines, achieving 96.9% success rates with grippers, (21.9% higher over baseline) and 100% success rates with dexterous hands (6.2% higher over baseline) in pick-and-place tasks. Besides, OmniVTLA significantly reduces task completion time and generates smoother trajectories through tactile sensing compared to existing VLA. Our ObjTac dataset can be found at https://readerek.github.io/Objtac.github.io

Paperid: 645, https://arxiv.org/pdf/2508.08701.pdf GitHub

Abstract:
Deep learning models for visual recognition often exhibit systematic errors due to underrepresented semantic subpopulations. Although existing debugging frameworks can pinpoint these failures by identifying key failure attributes, repairing the model effectively remains difficult. Current solutions often rely on manually designed prompts to generate synthetic training images -- an approach prone to distribution shift and semantic errors. To overcome these challenges, we introduce a model repair module that builds on an interpretable failure attribution pipeline. Our approach uses a conditional text-to-image model to generate semantically faithful and targeted images for failure cases. To preserve the quality and relevance of the generated samples, we further employ a large vision-language model (LVLM) to filter the outputs, enforcing alignment with the original data distribution and maintaining semantic consistency. By retraining vision models with this rare-case-augmented synthetic dataset, we significantly reduce errors associated with rare cases. Our experiments demonstrate that this targeted repair strategy improves model robustness without introducing new bugs. Code is available at https://github.com/oxu2/SafeFix

Paperid: 646, https://arxiv.org/pdf/2508.08240.pdf GitHub

Abstract:
Language-guided long-horizon mobile manipulation has long been a grand challenge in embodied semantic reasoning, generalizable manipulation, and adaptive locomotion. Three fundamental limitations hinder progress: First, although large language models have improved spatial reasoning and task planning through semantic priors, existing implementations remain confined to tabletop scenarios, failing to address the constrained perception and limited actuation ranges of mobile platforms. Second, current manipulation strategies exhibit insufficient generalization when confronted with the diverse object configurations encountered in open-world environments. Third, while crucial for practical deployment, the dual requirement of maintaining high platform maneuverability alongside precise end-effector control in unstructured settings remains understudied. In this work, we present ODYSSEY, a unified mobile manipulation framework for agile quadruped robots equipped with manipulators, which seamlessly integrates high-level task planning with low-level whole-body control. To address the challenge of egocentric perception in language-conditioned tasks, we introduce a hierarchical planner powered by a vision-language model, enabling long-horizon instruction decomposition and precise action execution. At the control level, our novel whole-body policy achieves robust coordination across challenging terrains. We further present the first benchmark for long-horizon mobile manipulation, evaluating diverse indoor and outdoor scenarios. Through successful sim-to-real transfer, we demonstrate the system's generalization and robustness in real-world deployments, underscoring the practicality of legged manipulators in unstructured environments. Our work advances the feasibility of generalized robotic assistants capable of complex, dynamic tasks. Our project page: https://kaijwang.github.io/odyssey.github.io/

Paperid: 647, https://arxiv.org/pdf/2508.08189.pdf GitHub

Abstract:
Recent advances at the intersection of reinforcement learning (RL) and visual intelligence have enabled agents that not only perceive complex visual scenes but also reason, generate, and act within them. This survey offers a critical and up-to-date synthesis of the field. We first formalize visual RL problems and trace the evolution of policy-optimization strategies from RLHF to verifiable reward paradigms, and from Proximal Policy Optimization to Group Relative Policy Optimization. We then organize more than 200 representative works into four thematic pillars: multi-modal large language models, visual generation, unified model frameworks, and vision-language-action models. For each pillar we examine algorithmic design, reward engineering, benchmark progress, and we distill trends such as curriculum-driven training, preference-aligned diffusion, and unified reward modeling. Finally, we review evaluation protocols spanning set-level fidelity, sample-level preference, and state-level stability, and we identify open challenges that include sample efficiency, generalization, and safe deployment. Our goal is to provide researchers and practitioners with a coherent map of the rapidly expanding landscape of visual RL and to highlight promising directions for future inquiry. Resources are available at: https://github.com/weijiawu/Awesome-Visual-Reinforcement-Learning.

Paperid: 648, https://arxiv.org/pdf/2508.08038.pdf GitHub

Abstract:
Depth estimation, essential for autonomous driving, seeks to interpret the 3D environment surrounding vehicles. The development of radar sensors, known for their cost-efficiency and robustness, has spurred interest in radar-camera fusion-based solutions. However, existing algorithms fuse features from these modalities without accounting for weather conditions, despite radars being known to be more robust than cameras under adverse weather. Additionally, while Vision-Language models have seen rapid advancement, utilizing language descriptions alongside other modalities for depth estimation remains an open challenge. This paper first introduces a text-generation strategy along with feature extraction and fusion techniques that can assist monocular depth estimation pipelines, leading to improved accuracy across different algorithms on the KITTI dataset. Building on this, we propose TRIDE, a radar-camera fusion algorithm that enhances text feature extraction by incorporating radar point information. To address the impact of weather on sensor performance, we introduce a weather-aware fusion block that adaptively adjusts radar weighting based on current weather conditions. Our method, benchmarked on the nuScenes dataset, demonstrates performance gains over the state-of-the-art, achieving a 12.87% improvement in MAE and a 9.08% improvement in RMSE. Code: https://github.com/harborsarah/TRIDE

Paperid: 649, https://arxiv.org/pdf/2508.07925.pdf GitHub

Abstract:
Video Temporal Grounding (VTG) aims to extract relevant video segments based on a given natural language query. Recently, zero-shot VTG methods have gained attention by leveraging pretrained vision-language models (VLMs) to localize target moments without additional training. However, existing approaches suffer from semantic fragmentation, where temporally continuous frames sharing the same semantics are split across multiple segments. When segments are fragmented, it becomes difficult to predict an accurate target moment that aligns with the text query. Also, they rely on skewed similarity distributions for localization, making it difficult to select the optimal segment. Furthermore, they heavily depend on the use of LLMs which require expensive inferences. To address these limitations, we propose a \textit{TAG}, a simple yet effective Temporal-Aware approach for zero-shot video temporal Grounding, which incorporates temporal pooling, temporal coherence clustering, and similarity adjustment. Our proposed method effectively captures the temporal context of videos and addresses distorted similarity distributions without training. Our approach achieves state-of-the-art results on Charades-STA and ActivityNet Captions benchmark datasets without rely on LLMs. Our code is available at https://github.com/Nuetee/TAG

Paperid: 650, https://arxiv.org/pdf/2508.07863.pdf GitHub

Abstract:
Human motion generation has emerged as a critical technology with transformative potential for real-world applications. However, existing vision-language-motion models (VLMMs) face significant limitations that hinder their practical deployment. We identify controllability as a main bottleneck, manifesting in five key aspects: inadequate response to diverse human commands, limited pose initialization capabilities, poor performance on long-term sequences, insufficient handling of unseen scenarios, and lack of fine-grained control over individual body parts. To overcome these limitations, we present Being-M0.5, the first real-time, controllable VLMM that achieves state-of-the-art performance across multiple motion generation tasks. Our approach is built upon HuMo100M, the largest and most comprehensive human motion dataset to date, comprising over 5 million self-collected motion sequences, 100 million multi-task instructional instances, and detailed part-level annotations that address a critical gap in existing datasets. We introduce a novel part-aware residual quantization technique for motion tokenization that enables precise, granular control over individual body parts during generation. Extensive experimental validation demonstrates Being-M0.5's superior performance across diverse motion benchmarks, while comprehensive efficiency analysis confirms its real-time capabilities. Our contributions include design insights and detailed computational analysis to guide future development of practical motion generators. We believe that HuMo100M and Being-M0.5 represent significant advances that will accelerate the adoption of motion generation technologies in real-world applications. The project page is available at https://beingbeyond.github.io/Being-M0.5.

Paperid: 651, https://arxiv.org/pdf/2508.07835.pdf GitHub

Abstract:
Recent advances in Vision-Language Models (VLMs) in histopathology, such as CONCH and QuiltNet, have demonstrated impressive zero-shot classification capabilities across various tasks. However, their general-purpose design may lead to suboptimal performance in specific downstream applications. While supervised fine-tuning methods address this issue, they require manually labeled samples for adaptation. This paper investigates annotation-free adaptation of VLMs through continued pretraining on domain- and task-relevant image-caption pairs extracted from existing databases. Our experiments on two VLMs, CONCH and QuiltNet, across three downstream tasks reveal that these pairs substantially enhance both zero-shot and few-shot performance. Notably, with larger training sizes, continued pretraining matches the performance of few-shot methods while eliminating manual labeling. Its effectiveness, task-agnostic design, and annotation-free workflow make it a promising pathway for adapting VLMs to new histopathology tasks. Code is available at https://github.com/DeepMicroscopy/Annotation-free-VLM-specialization.

Paperid: 652, https://arxiv.org/pdf/2508.07833.pdf GitHub

Abstract:
Vision Language Models (VLMs) encode multimodal inputs over large, complex, and difficult-to-interpret architectures, which limit transparency and trust. We propose a Multimodal Inversion for Model Interpretation and Conceptualization (MIMIC) framework to visualize the internal representations of VLMs by synthesizing visual concepts corresponding to internal encodings. MIMIC uses a joint VLM-based inversion and a feature alignment objective to account for VLM's autoregressive processing. It additionally includes a triplet of regularizers for spatial alignment, natural image smoothness, and semantic realism. We quantitatively and qualitatively evaluate MIMIC by inverting visual concepts over a range of varying-length free-form VLM output texts. Reported results include both standard visual quality metrics as well as semantic text-based metrics. To the best of our knowledge, this is the first model inversion approach addressing visual interpretations of VLM concepts.

Paperid: 653, https://arxiv.org/pdf/2508.07770.pdf GitHub

Abstract:
We introduce AgentWorld, an interactive simulation platform for developing household mobile manipulation capabilities. Our platform combines automated scene construction that encompasses layout generation, semantic asset placement, visual material configuration, and physics simulation, with a dual-mode teleoperation system supporting both wheeled bases and humanoid locomotion policies for data collection. The resulting AgentWorld Dataset captures diverse tasks ranging from primitive actions (pick-and-place, push-pull, etc.) to multistage activities (serve drinks, heat up food, etc.) across living rooms, bedrooms, and kitchens. Through extensive benchmarking of imitation learning methods including behavior cloning, action chunking transformers, diffusion policies, and vision-language-action models, we demonstrate the dataset's effectiveness for sim-to-real transfer. The integrated system provides a comprehensive solution for scalable robotic skill acquisition in complex home environments, bridging the gap between simulation-based training and real-world deployment. The code, datasets will be available at https://yizhengzhang1.github.io/agent_world/

Paperid: 654, https://arxiv.org/pdf/2508.07260.pdf GitHub

Abstract:
Personalizing Vision-Language Models (VLMs) to transform them into daily assistants has emerged as a trending research direction. However, leading companies like OpenAI continue to increase model size and develop complex designs such as the chain of thought (CoT). While large VLMs are proficient in complex multi-modal understanding, their high training costs and limited access via paid APIs restrict direct personalization. Conversely, small VLMs are easily personalized and freely available, but they lack sufficient reasoning capabilities. Inspired by this, we propose a novel collaborative framework named Small-Large Collaboration (SLC) for large VLM personalization, where the small VLM is responsible for generating personalized information, while the large model integrates this personalized information to deliver accurate responses. To effectively incorporate personalized information, we develop a test-time reflection strategy, preventing the potential hallucination of the small VLM. Since SLC only needs to train a meta personalized small VLM for the large VLMs, the overall process is training-efficient. To the best of our knowledge, this is the first training-efficient framework that supports both open-source and closed-source large VLMs, enabling broader real-world personalized applications. We conduct thorough experiments across various benchmarks and large VLMs to demonstrate the effectiveness of the proposed SLC framework. The code will be released at https://github.com/Hhankyangg/SLC.

Paperid: 655, https://arxiv.org/pdf/2508.06944.pdf GitHub GitHub

Abstract:
Large Language Models (LLMs) are typically fine-tuned for reasoning tasks through a two-stage pipeline of Supervised Fine-Tuning (SFT) followed by Reinforcement Learning (RL), a process fraught with catastrophic forgetting and suboptimal trade-offs between imitation and exploration. Recent single-stage methods attempt to unify SFT and RL using heuristics, but lack a principled mechanism for dynamically balancing the two paradigms. In this paper, we reframe this challenge through the theoretical lens of \textbf{implicit rewards}, viewing SFT and RL not as distinct methods but as complementary reward signals. We introduce \textbf{Adaptive Meta Fine-Tuning (AMFT)}, a novel single-stage algorithm that learns the optimal balance between SFT's implicit, path-level reward and RL's explicit, outcome-based reward. The core of AMFT is a \textbf{meta-gradient adaptive weight controller} that treats the SFT-RL balance as a learnable parameter, dynamically optimizing it to maximize long-term task performance. This forward-looking approach, regularized by policy entropy for stability, autonomously discovers an effective training curriculum. We conduct a comprehensive evaluation on challenging benchmarks spanning mathematical reasoning, abstract visual reasoning (General Points), and vision-language navigation (V-IRL). AMFT consistently establishes a new state-of-the-art and demonstrats superior generalization on out-of-distribution (OOD) tasks. Ablation studies and training dynamic analysis confirm that the meta-learning controller is crucial for AMFT's stability, sample efficiency, and performance, offering a more principled and effective paradigm for LLM alignment. Our codes are open-sourced via https://github.com/hlxtsyj/AMFT.

Paperid: 656, https://arxiv.org/pdf/2508.06763.pdf GitHub

Abstract:
Multimodal large language models (MLLMs) have achieved remarkable progress across a range of vision-language tasks and demonstrate strong potential for traffic accident understanding. However, existing MLLMs in this domain primarily focus on coarse-grained image-level or video-level comprehension and often struggle to handle fine-grained visual details or localized scene components, limiting their applicability in complex accident scenarios. To address these limitations, we propose SafePLUG, a novel framework that empowers MLLMs with both Pixel-Level Understanding and temporal Grounding for comprehensive traffic accident analysis. SafePLUG supports both arbitrary-shaped visual prompts for region-aware question answering and pixel-level segmentation based on language instructions, while also enabling the recognition of temporally anchored events in traffic accident scenarios. To advance the development of MLLMs for traffic accident understanding, we curate a new dataset containing multimodal question-answer pairs centered on diverse accident scenarios, with detailed pixel-level annotations and temporal event boundaries. Experimental results show that SafePLUG achieves strong performance on multiple tasks, including region-based question answering, pixel-level segmentation, temporal event localization, and accident event understanding. These capabilities lay a foundation for fine-grained understanding of complex traffic scenes, with the potential to improve driving safety and enhance situational awareness in smart transportation systems. The code, dataset, and model checkpoints will be made publicly available at: https://zihaosheng.github.io/SafePLUG

Paperid: 657, https://arxiv.org/pdf/2508.06530.pdf GitHub

Abstract:
Large Vision-Language Models (LVLMs), empowered by the success of Large Language Models (LLMs), have achieved impressive performance across domains. Despite the great advances in LVLMs, they still suffer from the unavailable object hallucination issue, which tends to generate objects inconsistent with the image content. The most commonly used Polling-based Object Probing Evaluation (POPE) benchmark evaluates this issue by sampling negative categories according to category-level statistics, \textit{e.g.}, category frequencies and co-occurrence. However, with the continuous advancement of LVLMs, the POPE benchmark has shown diminishing effectiveness in assessing object hallucination, as it employs a simplistic sampling strategy that overlooks image-specific information and restricts distractors to negative object categories only. In this paper, we introduce the Hallucination searching-based Object Probing Evaluation (HOPE) benchmark, aiming to generate the most misleading distractors (\textit{i.e.}, non-existent objects or incorrect image descriptions) that can trigger hallucination in LVLMs, which serves as a means to more rigorously assess their immunity to hallucination. To explore the image-specific information, the content-aware hallucination searching leverages Contrastive Language-Image Pre-Training (CLIP) to approximate the predictive behavior of LVLMs by selecting negative objects with the highest predicted likelihood as distractors. To expand the scope of hallucination assessment, the description-based hallucination searching constructs highly misleading distractors by pairing true objects with false descriptions. Experimental results show that HOPE leads to a precision drop of at least 9\% and up to 23\% across various state-of-the-art LVLMs, significantly outperforming POPE in exposing hallucination vulnerabilities. The code is available at https://github.com/xiemk/HOPE.

Paperid: 658, https://arxiv.org/pdf/2508.06496.pdf GitHub

Abstract:
An ensemble of trained multimodal encoders and vision-language models (VLMs) has become a standard approach for visual question answering (VQA) tasks. However, such models often fail to produce responses with the detailed precision necessary for complex, domain-specific applications such as medical VQA. Our representation model, BIND: BLIVA Integrated with Dense Encoding, extends prior multimodal work by refining the joint embedding space through dense, query-token-based encodings inspired by contrastive pretraining techniques. This refined encoder powers Med-GRIM, a model designed for medical VQA tasks that leverages graph-based retrieval and prompt engineering to integrate domain-specific knowledge. Rather than relying on compute-heavy fine-tuning of vision and language models on specific datasets, Med-GRIM applies a low-compute, modular workflow with small language models (SLMs) for efficiency. Med-GRIM employs prompt-based retrieval to dynamically inject relevant knowledge, ensuring both accuracy and robustness in its responses. By assigning distinct roles to each agent within the VQA system, Med-GRIM achieves large language model performance at a fraction of the computational cost. Additionally, to support scalable research in zero-shot multimodal medical applications, we introduce DermaGraph, a novel Graph-RAG dataset comprising diverse dermatological conditions. This dataset facilitates both multimodal and unimodal querying. The code and dataset are available at: https://github.com/Rakesh-123-cryp/Med-GRIM.git

Paperid: 659, https://arxiv.org/pdf/2508.06036.pdf GitHub

Abstract:
In this paper, we present our solution for the semi-supervised learning track (MER-SEMI) in MER2025. We propose a comprehensive framework, grounded in the principle that "more is better," to construct a robust Mixture of Experts (MoE) emotion recognition system. Our approach integrates a diverse range of input modalities as independent experts, including novel signals such as knowledge from large Vision-Language Models (VLMs) and temporal Action Unit (AU) information. To effectively utilize unlabeled data, we introduce a consensus-based pseudo-labeling strategy, generating high-quality labels from the agreement between a baseline model and Gemini, which are then used in a two-stage training paradigm. Finally, we employ a multi-expert voting ensemble combined with a rule-based re-ranking process to correct prediction bias and better align the outputs with human preferences. Evaluated on the MER2025-SEMI challenge dataset, our method achieves an F1-score of 0.8772 on the test set, ranking 2nd in the track. Our code is available at https://github.com/zhuyjan/MER2025-MRAC25.

Paperid: 660, https://arxiv.org/pdf/2508.05898.pdf GitHub

Abstract:
Pretrained vision-language models (VLMs) like CLIP show strong zero-shot performance but struggle with generalization under distribution shifts. Test-Time Adaptation (TTA) addresses this by adapting VLMs to unlabeled test data in new domains. While some TTA methods rely on prompt-tuning, training-free cache-based approaches are preferred for efficiency. However, current cache-based TTA models store only a limited set of high-confidence samples, restricting the decision boundary to these samples and ignoring the influence of other incoming test data. To address this, we propose Efficient Test-Time Adaptation (ETTA), introducing a Recursive Updating module that integrates all incoming test samples, progressively refining the decision boundary. This strategy mimics an unbounded cache, dynamically updating contextual embeddings for improved accuracy with minimal memory and computational overhead. ETTA also includes an Adaptive Ensemble module to reduce prompt dependency in image-to-text scores by dynamically selecting optimal prompts for each class. Furthermore, ETTA adaptively combines scores from both modules based on confidence levels, leveraging their complementary strengths. Extensive experiments on two benchmarks confirm that ETTA surpasses the state-of-the-art TTA models in computational complexity and accuracy, setting a new standard for effective, efficient test-time adaptation. The code has been released at https://github.com/hamidreza-dastmalchi/ETTA.

Paperid: 661, https://arxiv.org/pdf/2508.05669.pdf GitHub

Abstract:
Accurately extracting and representing the structure of tabular data from financial documents remains a critical challenge in document understanding, particularly for regulatory and analytical use cases. This study addresses the complexity of converting financial tables from Malaysian audited financial reports into Markdown format, a task complicated by rotated layouts, multi-level headers, and implicit structural cues. We propose a fine-tuned vision-language model (VLM), based on Qwen2.5-VL-7B, optimized for high-fidelity Markdown generation from document images. Our approach includes a curated dataset of 2,152 image-text pairs with augmentations and a supervised fine-tuning strategy using LoRA. To assess performance, we evaluated our model on 100 out-of-sample tables using a dual framework: a criteria-based LLM-as-a-judge for fine-grained accuracy and our novel Markdown Tree-Edit-Distance-based Similarity (TEDS) metric for holistic structural fidelity. Our model achieves a 92.20% overall accuracy on the criteria-based assessment and a 96.53% Markdown TEDS score. This performance significantly surpasses its Qwen2.5-VL-7B base model, larger-scale VLMs, and specialized reasoning-enabled models. Compared to these self-hosted alternatives, it also significantly reduces inference time. Furthermore, its accuracy exceeds that of widely used proprietary models such as OpenAI's GPT-4o and Gemini 2.5 Flash. These results demonstrate that domain-specific fine-tuning provides an effective and efficient method to bridge the gap between unstructured financial documents and downstream automation, rivalling much larger and more general models without their computational overhead.

Paperid: 662, https://arxiv.org/pdf/2508.05606.pdf GitHub

Abstract:
Chain-of-Thought (CoT) reasoning has been widely adopted to enhance Large Language Models (LLMs) by decomposing complex tasks into simpler, sequential subtasks. However, extending CoT to vision-language reasoning tasks remains challenging, as it often requires interpreting transitions of visual states to support reasoning. Existing methods often struggle with this due to limited capacity of modeling visual state transitions or incoherent visual trajectories caused by fragmented architectures. To overcome these limitations, we propose Uni-CoT, a Unified Chain-of-Thought framework that enables coherent and grounded multimodal reasoning within a single unified model. The key idea is to leverage a model capable of both image understanding and generation to reason over visual content and model evolving visual states. However, empowering a unified model to achieve that is non-trivial, given the high computational cost and the burden of training. To address this, Uni-CoT introduces a novel two-level reasoning paradigm: A Macro-Level CoT for high-level task planning and A Micro-Level CoT for subtask execution. This design significantly reduces the computational overhead. Furthermore, we introduce a structured training paradigm that combines interleaved image-text supervision for macro-level CoT with multi-task objectives for micro-level CoT. Together, these innovations allow Uni-CoT to perform scalable and coherent multi-modal reasoning. Furthermore, thanks to our design, all experiments can be efficiently completed using only 8 A100 GPUs with 80GB VRAM each. Experimental results on reasoning-driven image generation benchmark (WISE) and editing benchmarks (RISE and KRIS) indicates that Uni-CoT demonstrates SOTA performance and strong generalization, establishing Uni-CoT as a promising solution for multi-modal reasoning. Project Page and Code: https://sais-fuxi.github.io/projects/uni-cot/

Paperid: 663, https://arxiv.org/pdf/2508.05547.pdf GitHub GitHub

Abstract:
Vision-Language Models (VLMs) have demonstrated remarkable generalization capabilities across a wide range of tasks. However, their performance often remains suboptimal when directly applied to specific downstream scenarios without task-specific adaptation. To enhance their utility while preserving data efficiency, recent research has increasingly focused on unsupervised adaptation methods that do not rely on labeled data. Despite the growing interest in this area, there remains a lack of a unified, task-oriented survey dedicated to unsupervised VLM adaptation. To bridge this gap, we present a comprehensive and structured overview of the field. We propose a taxonomy based on the availability and nature of unlabeled visual data, categorizing existing approaches into four key paradigms: Data-Free Transfer (no data), Unsupervised Domain Transfer (abundant data), Episodic Test-Time Adaptation (batch data), and Online Test-Time Adaptation (streaming data). Within this framework, we analyze core methodologies and adaptation strategies associated with each paradigm, aiming to establish a systematic understanding of the field. Additionally, we review representative benchmarks across diverse applications and highlight open challenges and promising directions for future research. An actively maintained repository of relevant literature is available at https://github.com/tim-learn/Awesome-LabelFree-VLMs.

Paperid: 664, https://arxiv.org/pdf/2508.05221.pdf GitHub

Abstract:
Vision-language tracking has received increasing attention in recent years, as textual information can effectively address the inflexibility and inaccuracy associated with specifying the target object to be tracked. Existing works either directly fuse the fixed language with vision features or simply modify using attention, however, their performance is still limited. Recently, some researchers have explored using text generation to adapt to the variations in the target during tracking, however, these works fail to provide insights into the model's reasoning process and do not fully leverage the advantages of large models, which further limits their overall performance. To address the aforementioned issues, this paper proposes a novel reasoning-based vision-language tracking framework, named ReasoningTrack, based on a pre-trained vision-language model Qwen2.5-VL. Both SFT (Supervised Fine-Tuning) and reinforcement learning GRPO are used for the optimization of reasoning and language generation. We embed the updated language descriptions and feed them into a unified tracking backbone network together with vision features. Then, we adopt a tracking head to predict the specific location of the target object. In addition, we propose a large-scale long-term vision-language tracking benchmark dataset, termed TNLLT, which contains 200 video sequences. 20 baseline visual trackers are re-trained and evaluated on this dataset, which builds a solid foundation for the vision-language visual tracking task. Extensive experiments on multiple vision-language tracking benchmark datasets fully validated the effectiveness of our proposed reasoning-based natural language generation strategy. The source code of this paper will be released on https://github.com/Event-AHU/Open_VLTrack

Paperid: 665, https://arxiv.org/pdf/2508.05202.pdf GitHub

Abstract:
Spectral information has long been recognized as a critical cue in remote sensing observations. Although numerous vision-language models have been developed for pixel-level interpretation, spectral information remains underutilized, resulting in suboptimal performance, particularly in multispectral scenarios. To address this limitation, we construct a vision-language instruction-following dataset named SPIE, which encodes spectral priors of land-cover objects into textual attributes recognizable by large language models (LLMs), based on classical spectral index computations. Leveraging this dataset, we propose SPEX, a multimodal LLM designed for instruction-driven land cover extraction. To this end, we introduce several carefully designed components and training strategies, including multiscale feature aggregation, token context condensation, and multispectral visual pre-training, to achieve precise and flexible pixel-level interpretation. To the best of our knowledge, SPEX is the first multimodal vision-language model dedicated to land cover extraction in spectral remote sensing imagery. Extensive experiments on five public multispectral datasets demonstrate that SPEX consistently outperforms existing state-of-the-art methods in extracting typical land cover categories such as vegetation, buildings, and water bodies. Moreover, SPEX is capable of generating textual explanations for its predictions, thereby enhancing interpretability and user-friendliness. Code will be released at: https://github.com/MiliLab/SPEX.

Paperid: 666, https://arxiv.org/pdf/2508.05186.pdf GitHub GitHub

Abstract:
Recent vision-language-action (VLA) models for multi-task robotic manipulation commonly rely on static viewpoints and shared visual encoders, which limit 3D perception and cause task interference, hindering robustness and generalization. In this work, we propose Task-Aware View Planning (TAVP), a framework designed to overcome these challenges by integrating active view planning with task-specific representation learning. TAVP employs an efficient exploration policy, accelerated by a novel pseudo-environment, to actively acquire informative views. Furthermore, we introduce a Mixture-of-Experts (MoE) visual encoder to disentangle features across different tasks, boosting both representation fidelity and task generalization. By learning to see the world in a task-aware way, TAVP generates more complete and discriminative visual representations, demonstrating significantly enhanced action prediction across a wide array of manipulation challenges. Extensive experiments on RLBench tasks show that our proposed TAVP model achieves superior performance over state-of-the-art fixed-view approaches. Visual results and code are provided at: https://hcplab-sysu.github.io/TAVP.

Paperid: 667, https://arxiv.org/pdf/2508.04868.pdf GitHub GitHub

Abstract:
Transformer-based object detectors often struggle with occlusions, fine-grained localization, and computational inefficiency caused by fixed queries and dense attention. We propose DAMM, Dual-stream Attention with Multi-Modal queries, a novel framework introducing both query adaptation and structured cross-attention for improved accuracy and efficiency. DAMM capitalizes on three types of queries: appearance-based queries from vision-language models, positional queries using polygonal embeddings, and random learned queries for general scene coverage. Furthermore, a dual-stream cross-attention module separately refines semantic and spatial features, boosting localization precision in cluttered scenes. We evaluated DAMM on four challenging benchmarks, and it achieved state-of-the-art performance in average precision (AP) and recall, demonstrating the effectiveness of multi-modal query adaptation and dual-stream attention. Source code is at: \href{https://github.com/DET-LIP/DAMM}{GitHub}.

Paperid: 668, https://arxiv.org/pdf/2508.04700.pdf GitHub

Abstract:
Repurposing large vision-language models (LVLMs) as computer use agents (CUAs) has led to substantial breakthroughs, primarily driven by human-labeled data. However, these models often struggle with novel and specialized software, particularly in scenarios lacking human annotations. To address this challenge, we propose SEAgent, an agentic self-evolving framework enabling CUAs to autonomously evolve through interactions with unfamiliar software. Specifically, SEAgent empowers computer-use agents to autonomously master novel software environments via experiential learning, where agents explore new software, learn through iterative trial-and-error, and progressively tackle auto-generated tasks organized from simple to complex. To achieve this goal, we design a World State Model for step-wise trajectory assessment, along with a Curriculum Generator that generates increasingly diverse and challenging tasks. The agent's policy is updated through experiential learning, comprised of adversarial imitation of failure actions and Group Relative Policy Optimization (GRPO) on successful ones. Furthermore, we introduce a specialist-to-generalist training strategy that integrates individual experiential insights from specialist agents, facilitating the development of a stronger generalist CUA capable of continuous autonomous evolution. This unified agent ultimately achieves performance surpassing ensembles of individual specialist agents on their specialized software. We validate the effectiveness of SEAgent across five novel software environments within OS-World. Our approach achieves a significant improvement of 23.2% in success rate, from 11.3% to 34.5%, over a competitive open-source CUA, i.e., UI-TARS.

Paperid: 669, https://arxiv.org/pdf/2508.04681.pdf GitHub

Abstract:
Learning action models from real-world human-centric interaction datasets is important towards building general-purpose intelligent assistants with efficiency. However, most existing datasets only offer specialist interaction category and ignore that AI assistants perceive and act based on first-person acquisition. We urge that both the generalist interaction knowledge and egocentric modality are indispensable. In this paper, we embed the manual-assisted task into a vision-language-action framework, where the assistant provides services to the instructor following egocentric vision and commands. With our hybrid RGB-MoCap system, pairs of assistants and instructors engage with multiple objects and the scene following GPT-generated scripts. Under this setting, we accomplish InterVLA, the first large-scale human-object-human interaction dataset with 11.4 hours and 1.2M frames of multimodal data, spanning 2 egocentric and 5 exocentric videos, accurate human/object motions and verbal commands. Furthermore, we establish novel benchmarks on egocentric human motion estimation, interaction synthesis, and interaction prediction with comprehensive analysis. We believe that our InterVLA testbed and the benchmarks will foster future works on building AI agents in the physical world.

Paperid: 670, https://arxiv.org/pdf/2508.04572.pdf GitHub GitHub

Abstract:
In this work, we address the problem of grounding abnormalities in medical images, where the goal is to localize clinical findings based on textual descriptions. While generalist Vision-Language Models (VLMs) excel in natural grounding tasks, they often struggle in the medical domain due to rare, compositional, and domain-specific terms that are poorly aligned with visual patterns. Specialized medical VLMs address this challenge via large-scale domain pretraining, but at the cost of substantial annotation and computational resources. To overcome these limitations, we propose \textbf{Knowledge to Sight (K2Sight)}, a framework that introduces structured semantic supervision by decomposing clinical concepts into interpretable visual attributes, such as shape, density, and anatomical location. These attributes are distilled from domain ontologies and encoded into concise instruction-style prompts, which guide region-text alignment during training. Unlike conventional report-level supervision, our approach explicitly bridges domain knowledge and spatial structure, enabling data-efficient training of compact models. We train compact models with 0.23B and 2B parameters using only 1.5\% of the data required by state-of-the-art medical VLMs. Despite their small size and limited training data, these models achieve performance on par with or better than 7B+ medical VLMs, with up to 9.82\% improvement in $mAP_{50}$. Code and models: \href{https://lijunrio.github.io/K2Sight/}{\textcolor{SOTAPink}{https://lijunrio.github.io/K2Sight/}}.

Paperid: 671, https://arxiv.org/pdf/2508.04453.pdf GitHub

Abstract:
Large Vision-Language Models (LVLMs) have experienced significant advancements in recent years. However, their performance still falls short in tasks requiring deep visual perception, such as identifying subtle differences between images. A potential cause is the scarcity of visual knowledge in popular instruction-tuning corpora, resulting in inadequate visual perception and reasoning capabilities. To address this challenge, we introduce a self-improvement framework grounded in a novel visual knowledge-intensive task, \underline{C}ausality-driven \underline{V}isual object \underline{C}ompletion (CVC). This task requires LVLMs to infer the masked object in an image based on its \textit{causal} relationships with the other visible information. We first obtain rich examples cheaply through our automated instance construction pipeline, without relying on sophisticated LVLMs (\textit{e.g.}, GPT-4V) or human assistance. Then, LVLMs effectively self-improve through trial and error learning using these created instances. Our experiments demonstrate substantial gains across four challenging specialized tasks and four widely-used comprehensive benchmarks. Especially on specialized tasks, our method achieves an average improvement of 5.4\% and 4.0\% compared to the corresponding baselines when utilizing LLaVA-1.5-7B and LLaVA-1.5-13B, respectively. The code is available at https://github.com/XMUDeepLIT/CVC.

Paperid: 672, https://arxiv.org/pdf/2508.04369.pdf GitHub

Abstract:
Multimodal Large Language Models (MLLMs) have demonstrated significant progress in vision-language tasks, yet they still face challenges when processing long-duration video inputs. The limitation arises from MLLMs' context limit and training costs, necessitating sparse frame sampling before feeding videos into MLLMs. However, building a trainable sampling method remains challenging due to the unsupervised and non-differentiable nature of sparse frame sampling in Video-MLLMs. To address these problems, we propose Temporal Sampling Policy Optimization (TSPO), advancing MLLMs' long-form video-language understanding via reinforcement learning. Specifically, we first propose a trainable event-aware temporal agent, which captures event-query correlation for performing probabilistic keyframe selection. Then, we propose the TSPO reinforcement learning paradigm, which models keyframe selection and language generation as a joint decision-making process, enabling end-to-end group relative optimization for the temporal sampling policy. Furthermore, we propose a dual-style long video training data construction pipeline, balancing comprehensive temporal understanding and key segment localization. Finally, we incorporate rule-based answering accuracy and temporal locating reward mechanisms to optimize the temporal sampling policy. Comprehensive experiments show that our TSPO achieves state-of-the-art performance across multiple long video understanding benchmarks, and shows transferable ability across different cutting-edge Video-MLLMs. Our code is available at https://github.com/Hui-design/TSPO

Paperid: 673, https://arxiv.org/pdf/2508.04227.pdf GitHub

Abstract:
Vision-language models (VLMs) have achieved impressive performance across diverse multimodal tasks by leveraging large-scale pre-training. However, enabling them to learn continually from non-stationary data remains a major challenge, as their cross-modal alignment and generalization capabilities are particularly vulnerable to catastrophic forgetting. Unlike traditional unimodal continual learning (CL), VLMs face unique challenges such as cross-modal feature drift, parameter interference due to shared architectures, and zero-shot capability erosion. This survey offers the first focused and systematic review of continual learning for VLMs (VLM-CL). We begin by identifying the three core failure modes that degrade performance in VLM-CL. Based on these, we propose a challenge-driven taxonomy that maps solutions to their target problems: (1) \textit{Multi-Modal Replay Strategies} address cross-modal drift through explicit or implicit memory mechanisms; (2) \textit{Cross-Modal Regularization} preserves modality alignment during updates; and (3) \textit{Parameter-Efficient Adaptation} mitigates parameter interference with modular or low-rank updates. We further analyze current evaluation protocols, datasets, and metrics, highlighting the need for better benchmarks that capture VLM-specific forgetting and compositional generalization. Finally, we outline open problems and future directions, including continual pre-training and compositional zero-shot learning. This survey aims to serve as a comprehensive and diagnostic reference for researchers developing lifelong vision-language systems. All resources are available at: https://github.com/YuyangSunshine/Awesome-Continual-learning-of-Vision-Language-Models.

Paperid: 674, https://arxiv.org/pdf/2508.04043.pdf GitHub

Abstract:
Visual transformation reasoning (VTR) is a vital cognitive capability that empowers intelligent agents to understand dynamic scenes, model causal relationships, and predict future states, and thereby guiding actions and laying the foundation for advanced intelligent systems. However, existing benchmarks suffer from a sim-to-real gap, limited task complexity, and incomplete reasoning coverage, limiting their practical use in real-world scenarios. To address these limitations, we introduce VisualTrans, the first comprehensive benchmark specifically designed for VTR in real-world human-object interaction scenarios. VisualTrans encompasses 12 semantically diverse manipulation tasks and systematically evaluates three essential reasoning dimensions - spatial, procedural, and quantitative - through 6 well-defined subtask types. The benchmark features 472 high-quality question-answer pairs in various formats, including multiple-choice, open-ended counting, and target enumeration. We introduce a scalable data construction pipeline built upon first-person manipulation videos, which integrates task selection, image pair extraction, automated metadata annotation with large multimodal models, and structured question generation. Human verification ensures the final benchmark is both high-quality and interpretable. Evaluations of various state-of-the-art vision-language models show strong performance in static spatial tasks. However, they reveal notable shortcomings in dynamic, multi-step reasoning scenarios, particularly in areas like intermediate state recognition and transformation sequence planning. These findings highlight fundamental weaknesses in temporal modeling and causal reasoning, providing clear directions for future research aimed at developing more capable and generalizable VTR systems. The dataset and code are available at https://github.com/WangYipu2002/VisualTrans.

Paperid: 675, https://arxiv.org/pdf/2508.03742.pdf GitHub

Abstract:
Vision-language pre-training (VLP) has great potential for developing multifunctional and general medical diagnostic capabilities. However, aligning medical images with a low signal-to-noise ratio (SNR) to reports with a high SNR presents a semantic density gap, leading to visual alignment bias. In this paper, we propose boosting vision semantic density to improve alignment effectiveness. On one hand, we enhance visual semantics through disease-level vision contrastive learning, which strengthens the model's ability to differentiate between normal and abnormal samples for each anatomical structure. On the other hand, we introduce an anatomical normality modeling method to model the distribution of normal samples for each anatomy, leveraging VQ-VAE for reconstructing normal vision embeddings in the latent space. This process amplifies abnormal signals by leveraging distribution shifts in abnormal samples, enhancing the model's perception and discrimination of abnormal attributes. The enhanced visual representation effectively captures the diagnostic-relevant semantics, facilitating more efficient and accurate alignment with the diagnostic report. We conduct extensive experiments on two chest CT datasets, CT-RATE and Rad-ChestCT, and an abdominal CT dataset, MedVL-CT69K, and comprehensively evaluate the diagnosis performance across multiple tasks in the chest and abdominal CT scenarios, achieving state-of-the-art zero-shot performance. Notably, our method achieved an average AUC of 84.9% across 54 diseases in 15 organs, significantly surpassing existing methods. Additionally, we demonstrate the superior transfer learning capabilities of our pre-trained model. Code is available at https://github.com/alibaba-damo-academy/ViSD-Boost.

Paperid: 676, https://arxiv.org/pdf/2508.03654.pdf GitHub

Abstract:
Sarcasm is a complex linguistic phenomenon that involves a disparity between literal and intended meanings, making it challenging for sentiment analysis and other emotion-sensitive tasks. While traditional sarcasm detection methods primarily focus on text, recent approaches have incorporated multimodal information. However, the application of Large Visual Language Models (LVLMs) in Multimodal Sarcasm Analysis (MSA) remains underexplored. In this paper, we evaluate LVLMs in MSA tasks, specifically focusing on Multimodal Sarcasm Detection and Multimodal Sarcasm Explanation. Through comprehensive experiments, we identify key limitations, such as insufficient visual understanding and a lack of conceptual knowledge. To address these issues, we propose a training-free framework that integrates in-depth object extraction and external conceptual knowledge to improve the model's ability to interpret and explain sarcasm in multimodal contexts. The experimental results on multiple models show the effectiveness of our proposed framework. The code is available at https://github.com/cp-cp/LVLM-MSA.

Paperid: 677, https://arxiv.org/pdf/2508.03447.pdf GitHub

Abstract:
Recently, large pre-trained vision-language models have shown remarkable performance in zero-shot anomaly detection (ZSAD). With fine-tuning on a single auxiliary dataset, the model enables cross-category anomaly detection on diverse datasets covering industrial defects and medical lesions. Compared to manually designed prompts, prompt learning eliminates the need for expert knowledge and trial-and-error. However, it still faces the following challenges: (i) static learnable tokens struggle to capture the continuous and diverse patterns of normal and anomalous states, limiting generalization to unseen categories; (ii) fixed textual labels provide overly sparse category information, making the model prone to overfitting to a specific semantic subspace. To address these issues, we propose Conditional Prompt Synthesis (CoPS), a novel framework that synthesizes dynamic prompts conditioned on visual features to enhance ZSAD performance. Specifically, we extract representative normal and anomaly prototypes from fine-grained patch features and explicitly inject them into prompts, enabling adaptive state modeling. Given the sparsity of class labels, we leverage a variational autoencoder to model semantic image features and implicitly fuse varied class tokens into prompts. Additionally, integrated with our spatially-aware alignment mechanism, extensive experiments demonstrate that CoPS surpasses state-of-the-art methods by 2.5% AUROC in both classification and segmentation across 13 industrial and medical datasets. Code will be available at https://github.com/cqylunlun/CoPS.

Paperid: 678, https://arxiv.org/pdf/2508.03404.pdf GitHub

Abstract:
Existing vision-language models (VLMs), whether generalists or specialists, remain constrained by their parameter scale, lack robust self-correction capabilities, and underperform in tasks involving long visual contexts and complex reasoning, resulting in suboptimal performance on document-based tasks. To address this, we propose MACT, a Multi-Agent Collaboration framework with Test-Time scaling, tailored for visual document understanding and visual question answering (VQA). It comprises four distinct small-scale agents, i.e., planning, execution, judgment, and answer agents, with clearly defined roles and effective collaboration. Notably, the judgment agent exclusively verifies correctness and redirects to prior agents for revisions, outperforming conventional correction strategies. To further expand the capability boundaries of the framework, we propose mixed reward modeling that balances agent-specific abilities and global collaboration, as well as agent-wise hybrid test-time scaling, which customizes different scaling strategies for each agent based on their functions. Evaluated on benchmarks spanning both document-based and non-document-based settings, our MACT shows superior performance with a smaller parameter scale without sacrificing the ability of general and mathematical tasks. Especially, it stands out in benchmarks involving long visual contexts and complicated reasoning. The three variants of MACT consistently hold the top three positions in average scores, leading in 13 of the 15 benchmarks. Code will be available at: https://github.com/YU-deep/MACT.git.

Paperid: 679, https://arxiv.org/pdf/2508.03138.pdf GitHub

Abstract:
Robots operating in human-centric or hazardous environments must proactively anticipate and mitigate dangers beyond basic obstacle detection. Traditional navigation systems often depend on static maps, which struggle to account for dynamic risks, such as a person emerging from a suddenly opening door. As a result, these systems tend to be reactive rather than anticipatory when handling dynamic hazards. Recent advancements in pre-trained large language models and vision-language models (VLMs) create new opportunities for proactive hazard avoidance. In this work, we propose a zero-shot language-as-cost mapping framework that leverages VLMs to interpret visual scenes, assess potential dynamic risks, and assign risk-aware navigation costs preemptively, enabling robots to anticipate hazards before they materialize. By integrating this language-based cost map with a geometric obstacle map, the robot not only identifies existing obstacles but also anticipates and proactively plans around potential hazards arising from environmental dynamics. Experiments in simulated and diverse dynamic environments demonstrate that the proposed method significantly improves navigation success rates and reduces hazard encounters, compared to reactive baseline planners. Code and supplementary materials are available at https://github.com/Taekmino/LaC.

Paperid: 680, https://arxiv.org/pdf/2508.03127.pdf GitHub

Abstract:
Vision language models (VLMs) that enable natural language interaction with satellite imagery can democratize Earth observation by accelerating expert workflows, making data accessible to non-specialists, and enabling planet-scale automation. However, existing datasets focus mainly on short-term, high-resolution imagery from a limited number of satellites, overlooking low-resolution, multi-satellite, long-term archives, such as Landsat, that are essential for affordable and bias-robust global monitoring. We address this gap with Landsat30-AU, a large-scale vision-language dataset built from 30-meter resolution imagery collected by four Landsat satellites (5, 7, 8, and 9) over Australia, spanning more than 36 years. The dataset includes two components: Landsat30-AU-Cap, containing $196,262$ image-caption pairs, and Landsat30-AU-VQA, comprising 17,725 human-verified visual question answering (VQA) samples across eight remote sensing domains. Both datasets are curated through a bootstrapped pipeline that leverages generic VLMs with iterative refinement and human verification to ensure quality. Our evaluation of eight VLMs on our benchmark reveals that off-the-shelf models struggle to understand satellite imagery. The open-source remote-sensing VLM EarthDial achieves only 0.07 SPIDEr in captioning and a VQA accuracy of 0.48, highlighting the limitations of current approaches. Encouragingly, lightweight fine-tuning of Qwen2.5-VL-7B on Landsat30-AU improves captioning performance from 0.11 to 0.31 SPIDEr and boosts VQA accuracy from 0.74 to 0.87. Code and data are available at https://github.com/papersubmit1/landsat30-au.

Paperid: 681, https://arxiv.org/pdf/2508.02671.pdf GitHub

Abstract:
For CLIP-based prompt tuning, introducing more data as additional knowledge for enhancing fine-tuning process is proved to be an effective approach. Existing data amplification strategies for prompt tuning typically rely on external knowledge (e.g., large language models or pre-structured knowledge bases), resulting in higher costs for data collection and processing, while generally ignoring further utilization of features in image modality. To address this, we propose Augmentation-driven Prompt Tuning (AugPT), a self-contained distillation-based prompt tuning approach using only internal augmentation on raw dataset to better exploit known features. Specifically, AugPT employs self-supervised augmentation on unlabeled images in the training set, and introduces a novel gating mechanism based on consensus test, reusing the pre-trained prompt tuning backbone model to spontaneously filter noisy samples, further enhancing the quality of augmented views. Extensive experiments validate that AugPT simultaneously enhances model performance and generalization capability without using appended external knowledge. The code of AugPT is available at: https://github.com/JREion/AugPT .

Paperid: 682, https://arxiv.org/pdf/2508.02258.pdf GitHub

Abstract:
Although Vision Language Models (VLMs) have shown strong generalization in medical imaging, pathology presents unique challenges due to ultra-high resolution, complex tissue structures, and nuanced clinical semantics. These factors make pathology VLMs prone to hallucinations, i.e., generating outputs inconsistent with visual evidence, which undermines clinical trust. Existing RAG approaches in this domain largely depend on text-based knowledge bases, limiting their ability to leverage diagnostic visual cues. To address this, we propose Patho-AgenticRAG, a multimodal RAG framework with a database built on page-level embeddings from authoritative pathology textbooks. Unlike traditional text-only retrieval systems, it supports joint text-image search, enabling direct retrieval of textbook pages that contain both the queried text and relevant visual cues, thus avoiding the loss of critical image-based information. Patho-AgenticRAG also supports reasoning, task decomposition, and multi-turn search interactions, improving accuracy in complex diagnostic scenarios. Experiments show that Patho-AgenticRAG significantly outperforms existing multimodal models in complex pathology tasks like multiple-choice diagnosis and visual question answering. Our project is available at the Patho-AgenticRAG repository: https://github.com/Wenchuan-Zhang/Patho-AgenticRAG.

Paperid: 683, https://arxiv.org/pdf/2508.02047.pdf GitHub

Abstract:
Obtaining high-quality fine-grained annotations for traffic signs is critical for accurate and safe decision-making in autonomous driving. Widely used datasets, such as Mapillary, often provide only coarse-grained labels - without distinguishing semantically important types such as stop signs or speed limit signs. To this end, we present a new validation set for traffic signs derived from the Mapillary dataset called Mapillary Vistas Validation for Traffic Signs (MVV), where we decompose composite traffic signs into granular, semantically meaningful categories. The dataset includes pixel-level instance masks and has been manually annotated by expert annotators to ensure label fidelity. Further, we benchmark several state-of-the-art VLMs against the self-supervised DINOv2 model on this dataset and show that DINOv2 consistently outperforms all VLM baselines-not only on traffic sign recognition, but also on heavily represented categories like vehicles and humans. Our analysis reveals significant limitations in current vision-language models for fine-grained visual understanding and establishes DINOv2 as a strong baseline for dense semantic matching in autonomous driving scenarios. This dataset and evaluation framework pave the way for more reliable, interpretable, and scalable perception systems. Code and data are available at: https://github.com/nec-labs-ma/relabeling

Paperid: 684, https://arxiv.org/pdf/2508.01617.pdf GitHub

Abstract:
Autoregressive models (ARMs) have long dominated the landscape of biomedical vision-language models (VLMs). Recently, masked diffusion models such as LLaDA have emerged as promising alternatives, yet their application in the biomedical domain remains largely underexplored. To bridge this gap, we introduce \textbf{LLaDA-MedV}, the first large language diffusion model tailored for biomedical image understanding through vision instruction tuning. LLaDA-MedV achieves relative performance gains of 7.855\% over LLaVA-Med and 1.867\% over LLaDA-V in the open-ended biomedical visual conversation task, and sets new state-of-the-art accuracy on the closed-form subset of three VQA benchmarks: 84.93\% on VQA-RAD, 92.31\% on SLAKE, and 95.15\% on PathVQA. Furthermore, a detailed comparison with LLaVA-Med suggests that LLaDA-MedV is capable of generating reasonably longer responses by explicitly controlling response length, which can lead to more informative outputs. We also conduct an in-depth analysis of both the training and inference stages, highlighting the critical roles of initialization weight selection, fine-tuning strategies, and the interplay between sampling steps and response repetition. The code and model weight is released at https://github.com/LLM-VLM-GSL/LLaDA-MedV.

Paperid: 685, https://arxiv.org/pdf/2508.01558.pdf GitHub

Abstract:
Pre-trained Vision-Language Models (VLMs) have been exploited in various Computer Vision tasks (e.g., few-shot recognition) via model adaptation, such as prompt tuning and adapters. However, existing adaptation methods are designed by human experts, requiring significant time cost and experience. Inspired by recent advances in Large Language Models (LLMs) based code generation, we propose an Evolutionary Vision-Language Model Adaptation (EvoVLMA) method to automatically search training-free efficient adaptation algorithms for VLMs. We recognize feature selection and logits computation as the key functions in training-free VLM adaptation, and propose a two-stage LLM-assisted evolutionary algorithm for optimizing these parts in a sequential manner, effectively addressing the challenge posed by the expansive search space through a divide-and-conquer strategy. Besides, to enhance the stability and efficiency of searching process, we propose low-precision code conversion, web based code execution and process monitoring, leading to a highly effective automatic algorithm design system. Extensive experiments demonstrate that the algorithms found by EvoVLMA can obtain promising results compared to previous manually-designed ones. More specifically, in the 8-shot image classification setting, the classical APE algorithm can be improved by 1.91 points in recognition accuracy. This research opens new possibilities for automating the optimization of adaptation algorithms of pre-trained multimodal models. Code is available at: https://github.com/kding1225/EvoVLMA

Paperid: 686, https://arxiv.org/pdf/2508.01555.pdf GitHub

Abstract:
With the advancement of remote sensing satellite technology and the rapid progress of deep learning, remote sensing change detection (RSCD) has become a key technique for regional monitoring. Traditional change detection (CD) methods and deep learning-based approaches have made significant contributions to change analysis and detection, however, many outstanding methods still face limitations in the exploration and application of multimodal data. To address this, we propose the multimodal graph-conditioned vision-language reconstruction network (MGCR-Net) to further explore the semantic interaction capabilities of multimodal data. Multimodal large language models (MLLM) have attracted widespread attention for their outstanding performance in computer vision, particularly due to their powerful visual-language understanding and dialogic interaction capabilities. Specifically, we design a MLLM-based optimization strategy to generate multimodal textual data from the original CD images, which serve as textual input to MGCR. Visual and textual features are extracted through a dual encoder framework. For the first time in the RSCD task, we introduce a multimodal graph-conditioned vision-language reconstruction mechanism, which is integrated with graph attention to construct a semantic graph-conditioned reconstruction module (SGCM), this module generates vision-language (VL) tokens through graph-based conditions and enables cross-dimensional interaction between visual and textual features via multihead attention. The reconstructed VL features are then deeply fused using the language vision transformer (LViT), achieving fine-grained feature alignment and high-level semantic interaction. Experimental results on four public datasets demonstrate that MGCR achieves superior performance compared to mainstream CD methods. Our code is available on https://github.com/cn-xvkong/MGCR

Paperid: 687, https://arxiv.org/pdf/2508.01548.pdf GitHub

Abstract:
Visual token compression is critical for Large Vision-Language Models (LVLMs) to efficiently process high-resolution inputs. Existing methods that typically adopt fixed compression ratios cannot adapt to scenes of varying complexity, often causing imprecise pruning that discards informative visual tokens and results in degraded model performance. To address this issue, we introduce a dynamic pruning framework, GlimpsePrune, inspired by human cognition. It takes a data-driven ''glimpse'' and prunes irrelevant visual tokens in a single forward pass before answer generation. This approach prunes 92.6% of visual tokens while on average fully retaining the baseline performance on free-form VQA tasks. The reduced computational cost also enables more effective fine-tuning: an enhanced GlimpsePrune+ achieves 110% of the baseline performance while maintaining a similarly high pruning rate. Our work paves a new way for building more powerful and efficient LVLMs.

Paperid: 688, https://arxiv.org/pdf/2508.01225.pdf GitHub

Abstract:
In zero-shot setting, test-time adaptation adjusts pre-trained models using unlabeled data from the test phase to enhance performance on unknown test distributions. Existing cache-enhanced TTA methods rely on a low-entropy criterion to select samples for prototype construction, assuming intra-class compactness. However, low-entropy samples may be unreliable under distribution shifts, and the resulting prototypes may not ensure compact intra-class distributions. This study identifies a positive correlation between cache-enhanced performance and intra-class compactness. Based on this observation, we propose a Multi-Cache enhanced Prototype-based Test-Time Adaptation (MCP) featuring three caches: an entropy cache for initializing prototype representations with low-entropy samples, an align cache for integrating visual and textual information to achieve compact intra-class distributions, and a negative cache for prediction calibration using high-entropy samples. We further developed MCP++, a framework incorporating cross-modal prototype alignment and residual learning, introducing prototype residual fine-tuning. Comparative and ablation experiments across 15 downstream tasks demonstrate that the proposed method and framework achieve state-of-the-art generalization performance. Project Page available at: https://zhaihaotian.github.io/MCP-ICCV25/

Paperid: 689, https://arxiv.org/pdf/2508.01008.pdf GitHub

Abstract:
We present ROVI, a high-quality synthetic dataset for instance-grounded text-to-image generation, created by labeling 1M curated web images. Our key innovation is a strategy called re-captioning, focusing on the pre-detection stage, where a VLM (Vision-Language Model) generates comprehensive visual descriptions that are then processed by an LLM (Large Language Model) to extract a flat list of potential categories for OVDs (Open-Vocabulary Detectors) to detect. This approach yields a global prompt inherently linked to instance annotations while capturing secondary visual elements humans typically overlook. Evaluations show that ROVI exceeds existing detection datasets in image quality and resolution while containing two orders of magnitude more categories with an open-vocabulary nature. For demonstrative purposes, a text-to-image model GLIGEN trained on ROVI significantly outperforms state-of-the-art alternatives in instance grounding accuracy, prompt fidelity, and aesthetic quality. Our dataset and reproducible pipeline are available at https://github.com/CihangPeng/ROVI.

Paperid: 690, https://arxiv.org/pdf/2508.00553.pdf GitHub

Abstract:
Vision-Language Models (VLMs) encode images into lengthy sequences of visual tokens, leading to excessive computational overhead and limited inference efficiency. While prior efforts prune or merge tokens to address this issue, they often rely on special tokens (e.g., CLS) or require task-specific training, hindering scalability across architectures. In this paper, we propose HiPrune, a training-free and model-agnostic token Pruning framework that exploits the Hierarchical attention structure within vision encoders. We identify that middle layers attend to object-centric regions, while deep layers capture global contextual features. Based on this observation, HiPrune selects three types of informative tokens: (1) Anchor tokens with high attention in object-centric layers, (2) Buffer tokens adjacent to anchors for spatial continuity, and (3) Register tokens with strong attention in deep layers for global summarization. Our method requires no retraining and integrates seamlessly with any ViT-based VLM. Extensive experiments on LLaVA-1.5, LLaVA-NeXT, and Qwen2.5-VL demonstrate that HiPrune achieves state-of-the-art pruning performance, preserving up to 99.3% task accuracy with only 33.3% tokens, and maintaining 99.5% accuracy with just 11.1% tokens. Meanwhile, it reduces inference FLOPs and latency by up to 9$\times$, showcasing strong generalization across models and tasks. Code is available at https://github.com/Danielement321/HiPrune.

Paperid: 691, https://arxiv.org/pdf/2508.00400.pdf GitHub

Abstract:
We present Sari Sandbox, a high-fidelity, photorealistic 3D retail store simulation for benchmarking embodied agents against human performance in shopping tasks. Addressing a gap in retail-specific sim environments for embodied agent training, Sari Sandbox features over 250 interactive grocery items across three store configurations, controlled via an API. It supports both virtual reality (VR) for human interaction and a vision language model (VLM)-powered embodied agent. We also introduce SariBench, a dataset of annotated human demonstrations across varied task difficulties. Our sandbox enables embodied agents to navigate, inspect, and manipulate retail items, providing baselines against human performance. We conclude with benchmarks, performance analysis, and recommendations for enhancing realism and scalability. The source code can be accessed via https://github.com/upeee/sari-sandbox-env.

Paperid: 692, https://arxiv.org/pdf/2508.00399.pdf GitHub

Abstract:
Recent advances in vision-language models (VLMs) have enabled impressive generalization across diverse video understanding tasks under zero-shot settings. However, their capabilities in high-stakes industrial domains-where recognizing both routine operations and safety-critical anomalies is essential-remain largely underexplored. To address this gap, we introduce iSafetyBench, a new video-language benchmark specifically designed to evaluate model performance in industrial environments across both normal and hazardous scenarios. iSafetyBench comprises 1,100 video clips sourced from real-world industrial settings, annotated with open-vocabulary, multi-label action tags spanning 98 routine and 67 hazardous action categories. Each clip is paired with multiple-choice questions for both single-label and multi-label evaluation, enabling fine-grained assessment of VLMs in both standard and safety-critical contexts. We evaluate eight state-of-the-art video-language models under zero-shot conditions. Despite their strong performance on existing video benchmarks, these models struggle with iSafetyBench-particularly in recognizing hazardous activities and in multi-label scenarios. Our results reveal significant performance gaps, underscoring the need for more robust, safety-aware multimodal models for industrial applications. iSafetyBench provides a first-of-its-kind testbed to drive progress in this direction. The dataset is available at: https://github.com/iSafetyBench/data.

Paperid: 693, https://arxiv.org/pdf/2508.00395.pdf GitHub

Abstract:
Prompt tuning (PT), as an emerging resource-efficient fine-tuning paradigm, has showcased remarkable effectiveness in improving the task-specific transferability of vision-language models. This paper delves into a previously overlooked information asymmetry issue in PT, where the visual modality mostly conveys more context than the object-oriented textual modality. Correspondingly, coarsely aligning these two modalities could result in the biased attention, driving the model to merely focus on the context area. To address this, we propose DAPT, an effective PT framework based on an intuitive decouple-before-align concept. First, we propose to explicitly decouple the visual modality into the foreground and background representation via exploiting coarse-and-fine visual segmenting cues, and then both of these decoupled patterns are aligned with the original foreground texts and the hand-crafted background classes, thereby symmetrically strengthening the modal alignment. To further enhance the visual concentration, we propose a visual pull-push regularization tailored for the foreground-background patterns, directing the original visual representation towards unbiased attention on the region-of-interest object. We demonstrate the power of architecture-free DAPT through few-shot learning, base-to-novel generalization, and data-efficient learning, all of which yield superior performance across prevailing benchmarks. Our code will be released at https://github.com/Ferenas/DAPT.

Paperid: 694, https://arxiv.org/pdf/2508.00367.pdf GitHub

Abstract:
Transformers have demonstrated remarkable success across vision, language, and video. Yet, increasing task complexity has led to larger models and more tokens, raising the quadratic cost of self-attention and the overhead of GPU memory access. To reduce the computation cost of self-attention, prior work has proposed token compression techniques that drop redundant or less informative tokens. Meanwhile, fused attention kernels such as FlashAttention have been developed to alleviate memory overhead by avoiding attention map construction and its associated I/O to HBM. This, however, makes it incompatible with most training-free token compression methods, which rely on attention maps to determine token importance. Here, we propose Representation Shift, a training-free, model-agnostic metric that measures the degree of change in each token's representation. This seamlessly integrates token compression with FlashAttention, without attention maps or retraining. Our method further generalizes beyond Transformers to CNNs and state space models. Extensive experiments show that Representation Shift enables effective token compression compatible with FlashAttention, yielding significant speedups of up to 5.5% and 4.4% in video-text retrieval and video QA, respectively. Code is available at https://github.com/mlvlab/Representation-Shift.

Paperid: 695, https://arxiv.org/pdf/2508.00097.pdf GitHub

Abstract:
The rapid advancement of Vision-Language-Action models has created an urgent need for large-scale, high-quality robot demonstration datasets. Although teleoperation is the predominant method for data collection, current approaches suffer from limited scalability, complex setup procedures, and suboptimal data quality. This paper presents XRoboToolkit, a cross-platform framework for extended reality based robot teleoperation built on the OpenXR standard. The system features low-latency stereoscopic visual feedback, optimization-based inverse kinematics, and support for diverse tracking modalities including head, controller, hand, and auxiliary motion trackers. XRoboToolkit's modular architecture enables seamless integration across robotic platforms and simulation environments, spanning precision manipulators, mobile robots, and dexterous hands. We demonstrate the framework's effectiveness through precision manipulation tasks and validate data quality by training VLA models that exhibit robust autonomous performance.

Paperid: 696, https://arxiv.org/pdf/2507.23478.pdf GitHub GitHub

Abstract:
Large vision-language models (VLMs) have made significant strides in 2D visual understanding tasks, sparking interest in extending these capabilities to 3D scene understanding. However, current 3D VLMs often struggle with robust reasoning and generalization due to limitations in high-quality spatial data and the static nature of viewpoint assumptions. To address these challenges, we propose 3D-R1, a foundation model that enhances the reasoning capabilities of 3D VLMs. Specifically, we first construct a high-quality synthetic dataset with CoT, named Scene-30K, leveraging existing 3D-VL datasets and a data engine based on Gemini 2.5 Pro. It serves as cold-start initialization data for 3D-R1. Moreover, we leverage RLHF policy such as GRPO in the reinforcement learning training process to enhance reasoning capabilities and introduce three reward functions: a perception reward, a semantic similarity reward and a format reward to maintain detection accuracy and answer semantic precision. Furthermore, we introduce a dynamic view selection strategy that adaptively chooses the most informative perspectives for 3D scene understanding. Extensive experiments demonstrate that 3D-R1 delivers an average improvement of 10% across various 3D scene benchmarks, highlighting its effectiveness in enhancing reasoning and generalization in 3D scene understanding. Code: https://github.com/AIGeeksGroup/3D-R1. Website: https://aigeeksgroup.github.io/3D-R1.

Paperid: 697, https://arxiv.org/pdf/2507.23362.pdf GitHub

Abstract:
Although large vision-language models (LVLMs) have demonstrated impressive capabilities in multi-modal understanding and reasoning, their practical applications are still limited by massive model parameters and high computational costs. Recent efforts from natural language processing (NLP) have shown the effectiveness of layer pruning, offering a plausible training-free compression solution. However, due to the modality divergence between vision and language, it is unclear whether these NLP techniques are still effective in LVLMs. In this paper, we empirically prove that directly applying these layer pruning methods to LVLMs is ineffective. Through extensive experiments, we find that non-essential vision-language (VL) tokens and inter-layer feature gaps pose critical challenges to pruning layers in LVLMs. Based on these insights, we propose a novel framework Short-LVLM (SVL) that can utilize important VL tokens and mitigate the layer-wise feature gaps. Notably, Short-LVLM not only achieves a superior trade-off between performance and efficiency but also exhibits several potential advantages, i.e., training-free, model-agnostic, and highly compatible. The code for this work is publicly available at https://github.com/ASGO-MM/Short-LVLM.

Paperid: 698, https://arxiv.org/pdf/2507.23070.pdf GitHub

Abstract:
Fine-grained image classification, the task of distinguishing between visually similar subcategories within a broader category (e.g., bird species, car models, flower types), is a challenging computer vision problem. Traditional approaches rely heavily on fixed vocabularies and closed-set classification paradigms, limiting their scalability and adaptability in real-world settings where novel classes frequently emerge. Recent research has demonstrated that combining large language models (LLMs) with vision-language models (VLMs) makes open-set recognition possible without the need for predefined class labels. However, the existing methods are often limited in harnessing the power of LLMs at the classification phase, and also rely heavily on the guessed class names provided by an LLM without thorough analysis and refinement. To address these bottlenecks, we propose our training-free method, Enriched-FineR (or E-FineR for short), which demonstrates state-of-the-art results in fine-grained visual recognition while also offering greater interpretability, highlighting its strong potential in real-world scenarios and new domains where expert annotations are difficult to obtain. Additionally, we demonstrate the application of our proposed approach to zero-shot and few-shot classification, where it demonstrated performance on par with the existing SOTA while being training-free and not requiring human interventions. Overall, our vocabulary-free framework supports the shift in image classification from rigid label prediction to flexible, language-driven understanding, enabling scalable and generalizable systems for real-world applications. Well-documented code is available on https://github.com/demidovd98/e-finer.

Paperid: 699, https://arxiv.org/pdf/2507.22958.pdf GitHub

Abstract:
This paper introduces a novel benchmark, EGE-Math Solutions Assessment Benchmark, for evaluating Vision-Language Models (VLMs) on their ability to assess hand-written mathematical solutions. Unlike existing benchmarks that focus on problem solving, our approach centres on understanding student solutions, identifying mistakes, and assigning grades according to fixed criteria. We compile 122 scanned solutions from the Russian Unified State Exam (EGE) together with official expert grades, and evaluate seven modern VLMs from Google, OpenAI, Arcee AI, and Alibaba Cloud in three inference modes. The results reveal current limitations in mathematical reasoning and human-rubric alignment, opening new research avenues in AI-assisted assessment. You can find code in https://github.com/Karifannaa/Auto-check-EGE-math

Paperid: 700, https://arxiv.org/pdf/2507.22827.pdf GitHub

Abstract:
Automating the transformation of user interface (UI) designs into front-end code holds significant promise for accelerating software development and democratizing design workflows. While recent large language models (LLMs) have demonstrated progress in text-to-code generation, many existing approaches rely solely on natural language prompts, limiting their effectiveness in capturing spatial layout and visual design intent. In contrast, UI development in practice is inherently multimodal, often starting from visual sketches or mockups. To address this gap, we introduce a modular multi-agent framework that performs UI-to-code generation in three interpretable stages: grounding, planning, and generation. The grounding agent uses a vision-language model to detect and label UI components, the planning agent constructs a hierarchical layout using front-end engineering priors, and the generation agent produces HTML/CSS code via adaptive prompt-based synthesis. This design improves robustness, interpretability, and fidelity over end-to-end black-box methods. Furthermore, we extend the framework into a scalable data engine that automatically produces large-scale image-code pairs. Using these synthetic examples, we fine-tune and reinforce an open-source VLM, yielding notable gains in UI understanding and code quality. Extensive experiments demonstrate that our approach achieves state-of-the-art performance in layout accuracy, structural coherence, and code correctness. Our code is made publicly available at https://github.com/leigest519/ScreenCoder.

Paperid: 701, https://arxiv.org/pdf/2507.22802.pdf GitHub

Abstract:
Accurate fetal biometric measurements, such as abdominal circumference, play a vital role in prenatal care. However, obtaining high-quality ultrasound images for these measurements heavily depends on the expertise of sonographers, posing a significant challenge in low-income countries due to the scarcity of trained personnel. To address this issue, we leverage FetalCLIP, a vision-language model pretrained on a curated dataset of over 210,000 fetal ultrasound image-caption pairs, to perform automated fetal ultrasound image quality assessment (IQA) on blind-sweep ultrasound data. We introduce FetalCLIP$_{CLS}$, an IQA model adapted from FetalCLIP using Low-Rank Adaptation (LoRA), and evaluate it on the ACOUSLIC-AI dataset against six CNN and Transformer baselines. FetalCLIP$_{CLS}$ achieves the highest F1 score of 0.757. Moreover, we show that an adapted segmentation model, when repurposed for classification, further improves performance, achieving an F1 score of 0.771. Our work demonstrates how parameter-efficient fine-tuning of fetal ultrasound foundation models can enable task-specific adaptations, advancing prenatal care in resource-limited settings. The experimental code is available at: https://github.com/donglihe-hub/FetalCLIP-IQA.

Paperid: 702, https://arxiv.org/pdf/2507.22346.pdf GitHub GitHub

Abstract:
Accurate interpretation of land-cover changes in multi-temporal satellite imagery is critical for real-world scenarios. However, existing methods typically provide only one-shot change masks or static captions, limiting their ability to support interactive, query-driven analysis. In this work, we introduce remote sensing image change analysis (RSICA) as a new paradigm that combines the strengths of change detection and visual question answering to enable multi-turn, instruction-guided exploration of changes in bi-temporal remote sensing images. To support this task, we construct ChangeChat-105k, a large-scale instruction-following dataset, generated through a hybrid rule-based and GPT-assisted process, covering six interaction types: change captioning, classification, quantification, localization, open-ended question answering, and multi-turn dialogues. Building on this dataset, we propose DeltaVLM, an end-to-end architecture tailored for interactive RSICA. DeltaVLM features three innovations: (1) a fine-tuned bi-temporal vision encoder to capture temporal differences; (2) a visual difference perception module with a cross-semantic relation measuring (CSRM) mechanism to interpret changes; and (3) an instruction-guided Q-former to effectively extract query-relevant difference information from visual changes, aligning them with textual instructions. We train DeltaVLM on ChangeChat-105k using a frozen large language model, adapting only the vision and alignment modules to optimize efficiency. Extensive experiments and ablation studies demonstrate that DeltaVLM achieves state-of-the-art performance on both single-turn captioning and multi-turn interactive change analysis, outperforming existing multimodal large language models and remote sensing vision-language models. Code, dataset and pre-trained weights are available at https://github.com/hanlinwu/DeltaVLM.

Paperid: 703, https://arxiv.org/pdf/2507.22264.pdf GitHub

Abstract:
Contrastive Language-Image Pre-training (CLIP)~\citep{radford2021learning} has emerged as a pivotal model in computer vision and multimodal learning, achieving state-of-the-art performance at aligning visual and textual representations through contrastive learning. However, CLIP struggles with potential information misalignment in many image-text datasets and suffers from entangled representation. On the one hand, short captions for a single image in datasets like MSCOCO may describe disjoint regions in the image, leaving the model uncertain about which visual features to retain or disregard. On the other hand, directly aligning long captions with images can lead to the retention of entangled details, preventing the model from learning disentangled, atomic concepts -- ultimately limiting its generalization on certain downstream tasks involving short prompts. In this paper, we establish theoretical conditions that enable flexible alignment between textual and visual representations across varying levels of granularity. Specifically, our framework ensures that a model can not only \emph{preserve} cross-modal semantic information in its entirety but also \emph{disentangle} visual representations to capture fine-grained textual concepts. Building on this foundation, we introduce \ours, a novel approach that identifies and aligns the most relevant visual and textual representations in a modular manner. Superior performance across various tasks demonstrates its capability to handle information misalignment and supports our identification theory. The code is available at https://github.com/Mid-Push/SmartCLIP.

Paperid: 704, https://arxiv.org/pdf/2507.22101.pdf GitHub

Abstract:
Crops, fisheries and livestock form the backbone of global food production, essential to feed the ever-growing global population. However, these sectors face considerable challenges, including climate variability, resource limitations, and the need for sustainable management. Addressing these issues requires efficient, accurate, and scalable technological solutions, highlighting the importance of artificial intelligence (AI). This survey presents a systematic and thorough review of more than 200 research works covering conventional machine learning approaches, advanced deep learning techniques (e.g., vision transformers), and recent vision-language foundation models (e.g., CLIP) in the agriculture domain, focusing on diverse tasks such as crop disease detection, livestock health management, and aquatic species monitoring. We further cover major implementation challenges such as data variability and experimental aspects: datasets, performance evaluation metrics, and geographical focus. We finish the survey by discussing potential open research directions emphasizing the need for multimodal data integration, efficient edge-device deployment, and domain-adaptable AI models for diverse farming environments. Rapid growth of evolving developments in this field can be actively tracked on our project page: https://github.com/umair1221/AI-in-Agriculture

Paperid: 705, https://arxiv.org/pdf/2507.22051.pdf GitHub

Abstract:
Animating metaphoric visualizations brings data to life, enhancing the comprehension of abstract data encodings and fostering deeper engagement. However, creators face significant challenges in designing these animations, such as crafting motions that align semantically with the metaphors, maintaining faithful data representation during animation, and seamlessly integrating interactivity. We propose a human-AI co-creation workflow that facilitates creating animations for SVG-based metaphoric visualizations. Users can initially derive animation clips for data elements from vision-language models (VLMs) and subsequently coordinate their timelines based on entity order, attribute values, spatial layout, or randomness. Our design decisions were informed by a formative study with experienced designers (N=8). We further developed a prototype, DataSway, and conducted a user study (N=14) to evaluate its creativity support and usability. A gallery with 6 cases demonstrates its capabilities and applications in web-based hypermedia. We conclude with implications for future research on bespoke data visualization animation.

Paperid: 706, https://arxiv.org/pdf/2507.22003.pdf GitHub

Abstract:
Large Vision-Language Models (LVLMs) have demonstrated remarkable capabilities in visual understanding and multimodal reasoning. However, LVLMs frequently exhibit hallucination phenomena, manifesting as the generated textual responses that demonstrate inconsistencies with the provided visual content. Existing hallucination mitigation methods are predominantly text-centric, the challenges of visual-semantic alignment significantly limit their effectiveness, especially when confronted with fine-grained visual understanding scenarios. To this end, this paper presents ViHallu, a Vision-Centric Hallucination mitigation framework that enhances visual-semantic alignment through Visual Variation Image Generation and Visual Instruction Construction. ViHallu introduces visual variation images with controllable visual alterations while maintaining the overall image structure. These images, combined with carefully constructed visual instructions, enable LVLMs to better understand fine-grained visual content through fine-tuning, allowing models to more precisely capture the correspondence between visual content and text, thereby enhancing visual-semantic alignment. Extensive experiments on multiple benchmarks show that ViHallu effectively enhances models' fine-grained visual understanding while significantly reducing hallucination tendencies. Furthermore, we release ViHallu-Instruction, a visual instruction dataset specifically designed for hallucination mitigation and visual-semantic alignment. Code is available at https://github.com/oliviadzy/ViHallu.

Paperid: 707, https://arxiv.org/pdf/2507.21786.pdf GitHub GitHub

Abstract:
Vision-language pre-trained models (VLMs) such as CLIP have demonstrated remarkable zero-shot generalization, and prompt learning has emerged as an efficient alternative to full fine-tuning. However, existing methods often struggle with generalization to novel classes, a phenomenon attributed to overfitting on seen classes and forgetting general knowledge. Furthermore, recent approaches that improve generalization often introduce complex architectures or heavy computational overhead. In this paper, we propose a Multiple Semantic-Guided Context Optimization (MSGCoOp) framework to enhance few-shot generalization while maintaining computational efficiency. Our approach leverages an ensemble of parallel learnable context vectors to capture diverse semantic aspects. To enrich these prompts, we introduce a semantic guidance mechanism that aligns them with comprehensive class descriptions automatically generated by a Large Language Model (LLM). Furthermore, a diversity regularization loss encourages the prompts to learn complementary and orthogonal features, preventing them from collapsing into redundant representations. Extensive experiments on 11 benchmark datasets show that MSGCoOp significantly improves performance on base-to-novel generalization, achieving an average harmonic mean improvement of 1.10\% over the strong KgCoOp baseline. Our method also demonstrates enhanced robustness in cross-domain generalization tasks. Our code is avaliable at: \href{https://github.com/Rain-Bus/MSGCoOp}{https://github.com/Rain-Bus/MSGCoOp}.

Paperid: 708, https://arxiv.org/pdf/2507.21778.pdf GitHub

Abstract:
The detection of micro-expression Action Units (AUs) is a formidable challenge in affective computing, pivotal for decoding subtle, involuntary human emotions. While Large Language Models (LLMs) demonstrate profound reasoning abilities, their application to the fine-grained, low-intensity domain of micro-expression AU detection remains unexplored. This paper pioneers this direction by introducing \textbf{AU-LLM}, a novel framework that for the first time uses LLM to detect AUs in micro-expression datasets with subtle intensities and the scarcity of data. We specifically address the critical vision-language semantic gap, the \textbf{Enhanced Fusion Projector (EFP)}. The EFP employs a Multi-Layer Perceptron (MLP) to intelligently fuse mid-level (local texture) and high-level (global semantics) visual features from a specialized 3D-CNN backbone into a single, information-dense token. This compact representation effectively empowers the LLM to perform nuanced reasoning over subtle facial muscle movements.Through extensive evaluations on the benchmark CASME II and SAMM datasets, including stringent Leave-One-Subject-Out (LOSO) and cross-domain protocols, AU-LLM establishes a new state-of-the-art, validating the significant potential and robustness of LLM-based reasoning for micro-expression analysis. The codes are available at https://github.com/ZS-liu-JLU/AU-LLMs.

Paperid: 709, https://arxiv.org/pdf/2507.21745.pdf GitHub

Abstract:
Recent advances in large language and vision-language models have enabled strong reasoning capabilities, yet they remain impractical for specialized domains like remote sensing, where annotated data is scarce and expensive. We present the first few-shot reinforcement learning with verifiable reward (RLVR) framework for satellite imagery that eliminates the need for caption supervision--relying solely on lightweight, rule-based binary or IoU-based rewards. Adapting the "1-shot RLVR" paradigm from language models to vision-language models, we employ policy-gradient optimization with as few as one curated example to align model outputs for satellite reasoning tasks. Comprehensive experiments across multiple remote sensing benchmarks--including classification, visual question answering, and grounding--show that even a single example yields substantial improvements over the base model. Scaling to 128 examples matches or exceeds models trained on thousands of annotated samples. While the extreme one-shot setting can induce mild, task-specific overfitting, our approach consistently demonstrates robust generalization and efficiency across diverse tasks. Further, we find that prompt design and loss weighting significantly influence training stability and final accuracy. Our method enables cost-effective and data-efficient development of domain-specialist vision-language reasoning models, offering a pragmatic recipe for data-scarce fields: start from a compact VLM, curate a handful of reward-checkable cases, and train via RLVR.

Paperid: 710, https://arxiv.org/pdf/2507.21585.pdf GitHub

Abstract:
In this work, we study how vision-language models (VLMs) can be utilized to enhance the safety for the autonomous driving system, including perception, situational understanding, and path planning. However, existing research has largely overlooked the evaluation of these models in traffic safety-critical driving scenarios. To bridge this gap, we create the benchmark (SafeDrive228K) and propose a new baseline based on VLM with knowledge graph-based retrieval-augmented generation (SafeDriveRAG) for visual question answering (VQA). Specifically, we introduce SafeDrive228K, the first large-scale multimodal question-answering benchmark comprising 228K examples across 18 sub-tasks. This benchmark encompasses a diverse range of traffic safety queries, from traffic accidents and corner cases to common safety knowledge, enabling a thorough assessment of the comprehension and reasoning abilities of the models. Furthermore, we propose a plug-and-play multimodal knowledge graph-based retrieval-augmented generation approach that employs a novel multi-scale subgraph retrieval algorithm for efficient information retrieval. By incorporating traffic safety guidelines collected from the Internet, this framework further enhances the model's capacity to handle safety-critical situations. Finally, we conduct comprehensive evaluations on five mainstream VLMs to assess their reliability in safety-sensitive driving tasks. Experimental results demonstrate that integrating RAG significantly improves performance, achieving a +4.73% gain in Traffic Accidents tasks, +8.79% in Corner Cases tasks and +14.57% in Traffic Safety Commonsense across five mainstream VLMs, underscoring the potential of our proposed benchmark and methodology for advancing research in traffic safety. Our source code and data are available at https://github.com/Lumos0507/SafeDriveRAG.

Paperid: 711, https://arxiv.org/pdf/2507.21503.pdf GitHub

Abstract:
Recently Multimodal Large Language Models (MLLMs) have achieved considerable advancements in vision-language tasks, yet produce potentially harmful or untrustworthy content. Despite substantial work investigating the trustworthiness of language models, MMLMs' capability to act honestly, especially when faced with visually unanswerable questions, remains largely underexplored. This work presents the first systematic assessment of honesty behaviors across various MLLMs. We ground honesty in models' response behaviors to unanswerable visual questions, define four representative types of such questions, and construct MoHoBench, a large-scale MMLM honest benchmark, consisting of 12k+ visual question samples, whose quality is guaranteed by multi-stage filtering and human verification. Using MoHoBench, we benchmarked the honesty of 28 popular MMLMs and conducted a comprehensive analysis. Our findings show that: (1) most models fail to appropriately refuse to answer when necessary, and (2) MMLMs' honesty is not solely a language modeling issue, but is deeply influenced by visual information, necessitating the development of dedicated methods for multimodal honesty alignment. Therefore, we implemented initial alignment methods using supervised and preference learning to improve honesty behavior, providing a foundation for future work on trustworthy MLLMs. Our data and code can be found at https://github.com/DSTTSD/MoHoBench.

Paperid: 712, https://arxiv.org/pdf/2507.21494.pdf GitHub

Abstract:
Test-time adaptation with pre-trained vision-language models has gained increasing attention for addressing distribution shifts during testing. Among these approaches, memory-based algorithms stand out due to their training-free nature and ability to leverage historical test data. However, existing test-time adaptation methods are typically designed for a single domain with abundant data. In decentralized settings such as federated learning, applying these methods individually to each client suffers from limited test data, while directly sharing a single global memory via the server prevents proper personalization to each client's unique distribution. To address this, we propose Latte, a novel framework where each client maintains a local memory to store embeddings from its own historical test data and an external memory to store class prototypes from other relevant clients. During communication, each client retrieves prototypes from similar clients under the server's coordination to expand its memory. For local adaptation, Latte utilizes both embedding similarity and uncertainty to enhance model performance. Our theoretical analysis shows that Latte effectively leverages in-distribution clients while remaining robust to out-of-distribution clients. Extensive experiments on domain adaptation and corruption benchmarks validate that Latte achieves superior performance in decentralized settings, while introducing only negligible communication and computation costs. Our code is available at https://github.com/baowenxuan/Latte .

Paperid: 713, https://arxiv.org/pdf/2507.20920.pdf GitHub

Abstract:
Referring Image Segmentation (RIS), which aims to segment specific objects based on natural language descriptions, plays an essential role in vision-language understanding. Despite its progress in remote sensing applications, RIS in Low-Altitude Drone (LAD) scenarios remains underexplored. Existing datasets and methods are typically designed for high-altitude and static-view imagery. They struggle to handle the unique characteristics of LAD views, such as diverse viewpoints and high object density. To fill this gap, we present RIS-LAD, the first fine-grained RIS benchmark tailored for LAD scenarios. This dataset comprises 13,871 carefully annotated image-text-mask triplets collected from realistic drone footage, with a focus on small, cluttered, and multi-viewpoint scenes. It highlights new challenges absent in previous benchmarks, such as category drift caused by tiny objects and object drift under crowded same-class objects. To tackle these issues, we propose the Semantic-Aware Adaptive Reasoning Network (SAARN). Rather than uniformly injecting all linguistic features, SAARN decomposes and routes semantic information to different stages of the network. Specifically, the Category-Dominated Linguistic Enhancement (CDLE) aligns visual features with object categories during early encoding, while the Adaptive Reasoning Fusion Module (ARFM) dynamically selects semantic cues across scales to improve reasoning in complex scenes. The experimental evaluation reveals that RIS-LAD presents substantial challenges to state-of-the-art RIS algorithms, and also demonstrates the effectiveness of our proposed model in addressing these challenges. The dataset and code will be publicly released soon at: https://github.com/AHideoKuzeA/RIS-LAD/.

Paperid: 714, https://arxiv.org/pdf/2507.20842.pdf GitHub

Abstract:
Vision encoders serve as the cornerstone of multimodal understanding. Single-encoder architectures like CLIP exhibit inherent constraints in generalizing across diverse multimodal tasks, while recent multi-encoder fusion methods introduce prohibitive computational overhead to achieve superior performance using complementary visual representations from multiple vision encoders. To address this, we propose a progressive pruning framework, namely Multi-Encoder collaboraTivE tOken pRuning (METEOR), that eliminates redundant visual tokens across the encoding, fusion, and decoding stages for multi-encoder MLLMs. For multi-vision encoding, we discard redundant tokens within each encoder via a rank guided collaborative token assignment strategy. Subsequently, for multi-vision fusion, we combine the visual features from different encoders while reducing cross-encoder redundancy with cooperative pruning. Finally, we propose an adaptive token pruning method in the LLM decoding stage to further discard irrelevant tokens based on the text prompts with dynamically adjusting pruning ratios for specific task demands. To our best knowledge, this is the first successful attempt that achieves an efficient multi-encoder based vision language model with multi-stage pruning strategies. Extensive experiments on 11 benchmarks demonstrate the effectiveness of our proposed approach. Compared with EAGLE, a typical multi-encoder MLLMs, METEOR reduces 76% visual tokens with only 0.3% performance drop in average. The code is available at https://github.com/YuchenLiu98/METEOR.

Paperid: 715, https://arxiv.org/pdf/2507.20745.pdf GitHub

Abstract:
Low-Rank Adaptation (LoRA) and its variants have delivered strong capability in Parameter-Efficient Transfer Learning (PETL) by minimizing trainable parameters and benefiting from reparameterization. However, their projection matrices remain unrestricted during training, causing high representation redundancy and diminishing the effectiveness of feature adaptation in the resulting subspaces. While existing methods mitigate this by manually adjusting the rank or implicitly applying channel-wise masks, they lack flexibility and generalize poorly across various datasets and architectures. Hence, we propose ReSoRA, a method that explicitly models redundancy between mapping subspaces and adaptively Regularizes Subspace redundancy of Low-Rank Adaptation. Specifically, it theoretically decomposes the low-rank submatrices into multiple equivalent subspaces and systematically applies de-redundancy constraints to the feature distributions across different projections. Extensive experiments validate that our proposed method consistently facilitates existing state-of-the-art PETL methods across various backbones and datasets in vision-language retrieval and standard visual classification benchmarks. Besides, as a training supervision, ReSoRA can be seamlessly integrated into existing approaches in a plug-and-play manner, with no additional inference costs. Code is publicly available at: https://github.com/Lucenova/ReSoRA.

Paperid: 716, https://arxiv.org/pdf/2507.20630.pdf GitHub

Abstract:
Large Vision-Language Models (LVLMs) have advanced multimodal learning but face high computational costs due to the large number of visual tokens, motivating token pruning to improve inference efficiency. The key challenge lies in identifying which tokens are truly important. Most existing approaches rely on attention-based criteria to estimate token importance. However, they inherently suffer from certain limitations, such as positional bias. In this work, we explore a new perspective on token importance based on token transitions in LVLMs. We observe that the transition of token representations provides a meaningful signal of semantic information. Based on this insight, we propose TransPrune, a training-free and efficient token pruning method. Specifically, TransPrune progressively prunes tokens by assessing their importance through a combination of Token Transition Variation (TTV)-which measures changes in both the magnitude and direction of token representations-and Instruction-Guided Attention (IGA), which measures how strongly the instruction attends to image tokens via attention. Extensive experiments demonstrate that TransPrune achieves comparable multimodal performance to original LVLMs, such as LLaVA-v1.5 and LLaVA-Next, across eight benchmarks, while reducing inference TFLOPs by more than half. Moreover, TTV alone can serve as an effective criterion without relying on attention, achieving performance comparable to attention-based methods. The code will be made publicly available upon acceptance of the paper at https://github.com/liaolea/TransPrune.

Paperid: 717, https://arxiv.org/pdf/2507.20519.pdf GitHub

Abstract:
Precise automated understanding of agricultural tasks such as disease identification is essential for sustainable crop production. Recent advances in vision-language models (VLMs) are expected to further expand the range of agricultural tasks by facilitating human-model interaction through easy, text-based communication. Here, we introduce AgroBench (Agronomist AI Benchmark), a benchmark for evaluating VLM models across seven agricultural topics, covering key areas in agricultural engineering and relevant to real-world farming. Unlike recent agricultural VLM benchmarks, AgroBench is annotated by expert agronomists. Our AgroBench covers a state-of-the-art range of categories, including 203 crop categories and 682 disease categories, to thoroughly evaluate VLM capabilities. In our evaluation on AgroBench, we reveal that VLMs have room for improvement in fine-grained identification tasks. Notably, in weed identification, most open-source VLMs perform close to random. With our wide range of topics and expert-annotated categories, we analyze the types of errors made by VLMs and suggest potential pathways for future VLM development. Our dataset and code are available at https://dahlian00.github.io/AgroBenchPage/ .

Paperid: 718, https://arxiv.org/pdf/2507.20174.pdf GitHub

Abstract:
Real-world applications, such as autonomous driving and humanoid robot manipulation, require precise spatial perception. However, it remains underexplored how Vision-Language Models (VLMs) recognize spatial relationships and perceive spatial movement. In this work, we introduce a spatial evaluation pipeline and construct a corresponding benchmark. Specifically, we categorize spatial understanding into two main types: absolute spatial understanding, which involves querying the absolute spatial position (e.g., left, right) of an object within an image, and 3D spatial understanding, which includes movement and rotation. Notably, our dataset is entirely synthetic, enabling the generation of test samples at a low cost while also preventing dataset contamination. We conduct experiments on multiple state-of-the-art VLMs and observe that there is significant room for improvement in their spatial understanding abilities. Explicitly, in our experiments, humans achieve near-perfect performance on all tasks, whereas current VLMs attain human-level performance only on the two simplest tasks. For the remaining tasks, the performance of VLMs is distinctly lower than that of humans. In fact, the best-performing Vision-Language Models even achieve near-zero scores on multiple tasks. The dataset and code are available on https://github.com/kong13661/LRR-Bench.

Paperid: 719, https://arxiv.org/pdf/2507.20156.pdf GitHub

Abstract:
Vision-language models (VLMs) extend the conventional large language models by integrating visual data, enabling richer multimodal reasoning and significantly broadens the practical applications of AI. However, including visual inputs also brings new challenges in maintaining data quality. Empirical evidence consistently shows that carefully curated and representative training examples often yield superior results compared to simply increasing the quantity of data. Inspired by this observation, we introduce a streamlined data filtration framework that employs a compact VLM, fine-tuned on a high-quality image-caption annotated dataset. This model effectively evaluates and filters potential training samples based on caption and image quality and alignment. Unlike previous approaches, which typically add auxiliary filtration modules on top of existing full-scale VLMs, our method exclusively utilizes the inherent evaluative capability of a purpose-built small VLM. This strategy eliminates the need for extra modules and reduces training overhead. Our lightweight model efficiently filters out inaccurate, noisy web data, improving image-text alignment and caption linguistic fluency. Experimental results show that datasets underwent high-precision filtration using our compact VLM perform on par with, or even surpass, larger and noisier datasets gathered through high-volume web crawling. Thus, our method provides a lightweight yet robust solution for building high-quality vision-language training corpora. \\ \textbf{Availability and implementation:} Our compact VLM filtration model, training data, utility scripts, and Supplementary data (Appendices) are freely available at https://github.com/daulettoibazar/Compact_VLM_Filter.

Paperid: 720, https://arxiv.org/pdf/2507.20145.pdf GitHub

Abstract:
Document Understanding (DU) in long-contextual scenarios with complex layouts remains a significant challenge in vision-language research. Although Large Vision-Language Models (LVLMs) excel at short-context DU tasks, their performance declines in long-context settings. A key limitation is the scarcity of fine-grained training data, particularly for low-resource languages such as Arabic. Existing state-of-the-art techniques rely heavily on human annotation, which is costly and inefficient. We propose a fully automated, multi-agent interactive framework to generate long-context questions efficiently. Our approach efficiently generates high-quality single- and multi-page questions for extensive English and Arabic documents, covering hundreds of pages across diverse domains. This facilitates the development of LVLMs with enhanced long-context understanding ability. Experimental results in this work have shown that our generated English and Arabic questions (\textbf{AraEngLongBench}) are quite challenging to major open- and close-source LVLMs. The code and data proposed in this work can be found in https://github.com/wangk0b/Multi_Agentic_QA_Long_Doc.git. Sample Question and Answer (QA) pairs and structured system prompts can be found in the Appendix.

Paperid: 721, https://arxiv.org/pdf/2507.20136.pdf GitHub

Abstract:
This paper presents the technical solution developed by team CRUISE for the KDD Cup 2025 Meta Comprehensive RAG Benchmark for Multi-modal, Multi-turn (CRAG-MM) challenge. The challenge aims to address a critical limitation of modern Vision Language Models (VLMs): their propensity to hallucinate, especially when faced with egocentric imagery, long-tail entities, and complex, multi-hop questions. This issue is particularly problematic in real-world applications where users pose fact-seeking queries that demand high factual accuracy across diverse modalities. To tackle this, we propose a robust, multi-stage framework that prioritizes factual accuracy and truthfulness over completeness. Our solution integrates a lightweight query router for efficiency, a query-aware retrieval and summarization pipeline, a dual-pathways generation and a post-hoc verification. This conservative strategy is designed to minimize hallucinations, which incur a severe penalty in the competition's scoring metric. Our approach achieved 3rd place in Task 1, demonstrating the effectiveness of prioritizing answer reliability in complex multi-modal RAG systems. Our implementation is available at https://github.com/Breezelled/KDD-Cup-2025-Meta-CRAG-MM .

Paperid: 722, https://arxiv.org/pdf/2507.20025.pdf GitHub

Abstract:
Learning visual representations is foundational for a broad spectrum of downstream tasks. Although recent vision-language contrastive models, such as CLIP and SigLIP, have achieved impressive zero-shot performance via large-scale vision-language alignment, their reliance on global representations constrains their effectiveness for dense prediction tasks, such as grounding, OCR, and segmentation. To address this gap, we introduce Region-Aware Cluster Discrimination (RICE), a novel method that enhances region-level visual and OCR capabilities. We first construct a billion-scale candidate region dataset and propose a Region Transformer layer to extract rich regional semantics. We further design a unified region cluster discrimination loss that jointly supports object and OCR learning within a single classification framework, enabling efficient and scalable distributed training on large-scale data. Extensive experiments show that RICE consistently outperforms previous methods on tasks, including segmentation, dense detection, and visual perception for Multimodal Large Language Models (MLLMs). The pre-trained models have been released at https://github.com/deepglint/MVT.

Paperid: 723, https://arxiv.org/pdf/2507.19891.pdf GitHub

Abstract:
We propose Reverse Contrast Attention (RCA), a plug-in method that enhances object localization in vision-language transformers without retraining. RCA reweights final-layer attention by suppressing extremes and amplifying mid-level activations to let semantically relevant but subdued tokens guide predictions. We evaluate it on Open Vocabulary Referring Object Detection (OV-RefOD), introducing FitAP, a confidence-free average precision metric based on IoU and box area. RCA improves FitAP in 11 out of 15 open-source VLMs, with gains up to $+26.6\%$. Effectiveness aligns with attention sharpness and fusion timing; while late-fusion models benefit consistently, models like $\texttt{DeepSeek-VL2}$ also improve, pointing to capacity and disentanglement as key factors. RCA offers both interpretability and performance gains for multimodal transformers. Codes and dataset are available from https://github.com/earl-juanico/rca

Paperid: 724, https://arxiv.org/pdf/2507.19875.pdf GitHub

Abstract:
Vision-language tracking aims to locate the target object in the video sequence using a template patch and a language description provided in the initial frame. To achieve robust tracking, especially in complex long-term scenarios that reflect real-world conditions as recently highlighted by MGIT, it is essential not only to characterize the target features but also to utilize the context features related to the target. However, the visual and textual target-context cues derived from the initial prompts generally align only with the initial target state. Due to their dynamic nature, target states are constantly changing, particularly in complex long-term sequences. It is intractable for these cues to continuously guide Vision-Language Trackers (VLTs). Furthermore, for the text prompts with diverse expressions, our experiments reveal that existing VLTs struggle to discern which words pertain to the target or the context, complicating the utilization of textual cues. In this work, we present a novel tracker named ATCTrack, which can obtain multimodal cues Aligned with the dynamic target states through comprehensive Target-Context feature modeling, thereby achieving robust tracking. Specifically, (1) for the visual modality, we propose an effective temporal visual target-context modeling approach that provides the tracker with timely visual cues. (2) For the textual modality, we achieve precise target words identification solely based on textual content, and design an innovative context words calibration method to adaptively utilize auxiliary context words. (3) We conduct extensive experiments on mainstream benchmarks and ATCTrack achieves a new SOTA performance. The code and models will be released at: https://github.com/XiaokunFeng/ATCTrack.

Paperid: 725, https://arxiv.org/pdf/2507.19847.pdf GitHub GitHub

Abstract:
Out-of-distribution (OOD) detection is crucial for building reliable machine learning models. Although negative prompt tuning has enhanced the OOD detection capabilities of vision-language models, these tuned models often suffer from reduced generalization performance on unseen classes and styles. To address this challenge, we propose a novel method called Knowledge Regularized Negative Feature Tuning (KR-NFT), which integrates an innovative adaptation architecture termed Negative Feature Tuning (NFT) and a corresponding knowledge-regularization (KR) optimization strategy. Specifically, NFT applies distribution-aware transformations to pre-trained text features, effectively separating positive and negative features into distinct spaces. This separation maximizes the distinction between in-distribution (ID) and OOD images. Additionally, we introduce image-conditional learnable factors through a lightweight meta-network, enabling dynamic adaptation to individual images and mitigating sensitivity to class and style shifts. Compared to traditional negative prompt tuning, NFT demonstrates superior efficiency and scalability. To optimize this adaptation architecture, the KR optimization strategy is designed to enhance the discrimination between ID and OOD sets while mitigating pre-trained knowledge forgetting. This enhances OOD detection performance on trained ID classes while simultaneously improving OOD detection on unseen ID datasets. Notably, when trained with few-shot samples from ImageNet dataset, KR-NFT not only improves ID classification accuracy and OOD detection but also significantly reduces the FPR95 by 5.44\% under an unexplored generalization setting with unseen ID categories. Codes can be found at \href{https://github.com/ZhuWenjie98/KRNFT}.

Paperid: 726, https://arxiv.org/pdf/2507.19492.pdf GitHub

Abstract:
Chart-to-code reconstruction -- the task of recovering executable plotting scripts from chart images -- provides important insights into a model's ability to ground data visualizations in precise, machine-readable form. Yet many existing multimodal benchmarks largely focus primarily on answering questions about charts or summarizing them. To bridge this gap, we present ChartGen, a fully-automated pipeline for code-guided synthetic chart generation. Starting from seed chart images, ChartGen (i) prompts a vision-language model (VLM) to reconstruct each image into a python script, and (ii) iteratively augments that script with a code-oriented large language model (LLM). Using ChartGen, we create 222.5K unique chart-image code pairs from 13K seed chart images, and present an open-source synthetic chart dataset covering 27 chart types, 11 plotting libraries, and multiple data modalities (image, code, text, CSV, DocTags). From this corpus, we curate a held-out chart-to-code evaluation subset of 4.3K chart image-code pairs, and evaluate six open-weight VLMs (3B - 26B parameters), highlighting substantial room for progress. We release the pipeline, prompts, and the dataset to help accelerate efforts towards robust chart understanding and vision-conditioned code generation: https://github.com/SD122025/ChartGen/

Paperid: 727, https://arxiv.org/pdf/2507.19064.pdf GitHub

Abstract:
In this paper, we study a practical but less-touched problem in Vision-Language Models (VLMs), \ie, negation understanding. Specifically, many real-world applications require models to explicitly identify what is false or non-existent, \eg, radiologists may search for images that exclude specific conditions. Despite the impressive transferability of VLMs through large-scale training, they suffer from a critical limitation that fails to handle negation. To address this challenge, existing methods attribute its root cause to the scarcity of negation training data and propose to fine-tune VLMs on massive data containing explicit negation. Undoubtedly, such data-centric solutions demand substantial data and computational resources, limiting their sustainable widespread adoption. To tackle negation in a low-carbon manner, we empirically observe that the key obstacle lies in the dual-concept shifts between the affirmation and negation distributions. Therefore, we propose a Negation-Aware Test-Time Adaptation (NEAT) method to efficiently adjust distribution-related parameters during inference. In brief, NEAT can reduce distribution shift in consistent semantics while eliminating false distributional consistency in unrelated semantics. Extensive experiments on the various negation understanding tasks verify the effectiveness of the proposed method. Remarkably, with less than 0.01\% of trainable parameters, NEAT achieves comparable or superior performance to state-of-the-art post-training approaches. Our code is available at https://github.com/hhc1997/NEAT.

Paperid: 728, https://arxiv.org/pdf/2507.19054.pdf GitHub

Abstract:
Mixed modality search -- retrieving information across a heterogeneous corpus composed of images, texts, and multimodal documents -- is an important yet underexplored real-world application. In this work, we investigate how contrastive vision-language models, such as CLIP, perform on the mixed modality search task. Our analysis reveals a critical limitation: these models exhibit a pronounced modality gap in the embedding space, where image and text embeddings form distinct clusters, leading to intra-modal ranking bias and inter-modal fusion failure. To address this issue, we propose GR-CLIP, a lightweight post-hoc calibration method that removes the modality gap in CLIP's embedding space. Evaluated on MixBench -- the first benchmark specifically designed for mixed modality search -- GR-CLIP improves NDCG@10 by up to 26 percentage points over CLIP, surpasses recent vision-language generative embedding models by 4 percentage points, while using 75x less compute.

Paperid: 729, https://arxiv.org/pdf/2507.18743.pdf GitHub

Abstract:
Vision Language Models (VLMs) have achieved remarkable breakthroughs in the field of remote sensing in recent years. Synthetic Aperture Radar (SAR) imagery, with its all-weather capability, is essential in remote sensing, yet the lack of large-scale, high-quality SAR image-text datasets hinders its semantic understanding. In this paper, we construct SAR-TEXT, a large-scale and high-quality dataset consisting of over 130,000 SAR image-text pairs. To construct the SAR-TEXT dataset, we design the SAR-Narrator framework, which generates textual descriptions for SAR images through a multi-stage strategy. To verify the effectiveness of the SAR-TEXT dataset, we conduct experiments on three typical vision-language tasks: image-text retrieval, image captioning, and visual question answering (VQA). Specifically, we construct three representative models on SAR-TEXT: SAR-RS-CLIP, SAR-RS-CoCa, and SAR-GPT. SAR-RS-CLIP achieves notable improvements in retrieval performance, boosting average recall by 12.97% and 10.0% on the OSdataset_512 and HRSID test sets, respectively. In the captioning task, SAR-RS-CoCa achieves significant improvements over the original CoCa models in terms of BLEU-4, SPICE, and CIDEr scores. In the VQA task, SAR-GPT outperforms baseline and single-stage models on multiple SAR-VQA datasets, demonstrating stronger semantic understanding and reasoning ability, as further confirmed by qualitative results. It is worth noting that, as a flexible captioning tool, SAR-Narrator can be readily adopted by the community to construct larger-scale SAR image-text datasets. All code, pretrained models, and the SAR-Text dataset are publicly available at: https://github.com/YiguoHe/SAR-TEXT.

Paperid: 730, https://arxiv.org/pdf/2507.18064.pdf GitHub

Abstract:
Most existing low-light image enhancement (LLIE) methods rely on pre-trained model priors, low-light inputs, or both, while neglecting the semantic guidance available from normal-light images. This limitation hinders their effectiveness in complex lighting conditions. In this paper, we propose VLM-IMI, a novel framework that leverages large vision-language models (VLMs) with iterative and manual instructions (IMIs) for LLIE. VLM-IMI incorporates textual descriptions of the desired normal-light content as enhancement cues, enabling semantically informed restoration. To effectively integrate cross-modal priors, we introduce an instruction prior fusion module, which dynamically aligns and fuses image and text features, promoting the generation of detailed and semantically coherent outputs. During inference, we adopt an iterative and manual instruction strategy to refine textual instructions, progressively improving visual quality. This refinement enhances structural fidelity, semantic alignment, and the recovery of fine details under extremely low-light conditions. Extensive experiments across diverse scenarios demonstrate that VLM-IMI outperforms state-of-the-art methods in both quantitative metrics and perceptual quality. The source code is available at https://github.com/sunxiaoran01/VLM-IMI.

Paperid: 731, https://arxiv.org/pdf/2507.18043.pdf GitHub

Abstract:
Inference-time steering methods offer a lightweight alternative to fine-tuning large language models (LLMs) and vision-language models (VLMs) by modifying internal activations at test time without updating model weights. However, most existing approaches rely on fixed, global intervention vectors, overlook the causal influence of individual input tokens, and fail to leverage informative gradients from the model's logits, particularly in multimodal settings where visual and textual inputs contribute unevenly. To address these limitations, we introduce GrAInS, an inference-time steering approach that operates across both language-only and vision-language models and tasks. GrAInS uses contrastive, gradient-based attribution via Integrated Gradients to identify the top-k most influential tokens, both positively and negatively attributed based on their contribution to preferred versus dispreferred outputs. These tokens are then used to construct directional steering vectors that capture semantic shifts from undesirable to desirable behavior. During inference, GrAInS adjusts hidden activations at transformer layers guided by token-level attribution signals, and normalizes activations to preserve representational scale. This enables fine-grained, interpretable, and modular control over model behavior, without retraining or auxiliary supervision. Empirically, GrAInS consistently outperforms both fine-tuning and existing steering baselines: it achieves a 13.22% accuracy gain on TruthfulQA using Llama-3.1-8B, reduces hallucination rates on MMHal-Bench from 0.624 to 0.514 with LLaVA-1.6-7B, and improves alignment win rates on SPA-VL by 8.11%, all while preserving the model's fluency and general capabilities.

Paperid: 732, https://arxiv.org/pdf/2507.18033.pdf GitHub

Abstract:
Pre-trained large language models (LLMs) have demonstrated strong common-sense reasoning abilities, making them promising for robotic navigation and planning tasks. However, despite recent progress, bridging the gap between language descriptions and actual robot actions in the open-world, beyond merely invoking limited predefined motion primitives, remains an open challenge. In this work, we aim to enable robots to interpret and decompose complex language instructions, ultimately synthesizing a sequence of trajectory points to complete diverse navigation tasks given open-set instructions and open-set objects. We observe that multi-modal large language models (MLLMs) exhibit strong cross-modal understanding when processing free-form language instructions, demonstrating robust scene comprehension. More importantly, leveraging their code-generation capability, MLLMs can interact with vision-language perception models to generate compositional 2D bird-eye-view value maps, effectively integrating semantic knowledge from MLLMs with spatial information from maps to reinforce the robot's spatial understanding. To further validate our approach, we effectively leverage large-scale autonomous vehicle datasets (AVDs) to validate our proposed zero-shot vision-language navigation framework in outdoor navigation tasks, demonstrating its capability to execute a diverse range of free-form natural language navigation instructions while maintaining robustness against object detection errors and linguistic ambiguities. Furthermore, we validate our system on a Husky robot in both indoor and outdoor scenes, demonstrating its real-world robustness and applicability. Supplementary videos are available at https://trailab.github.io/OpenNav-website/

Paperid: 733, https://arxiv.org/pdf/2507.17456.pdf GitHub

Abstract:
Human-Object Interaction (HOI) detection aims to identify humans and objects within images and interpret their interactions. Existing HOI methods rely heavily on large datasets with manual annotations to learn interactions from visual cues. These annotations are labor-intensive to create, prone to inconsistency, and limit scalability to new domains and rare interactions. We argue that recent advances in Vision-Language Models (VLMs) offer untapped potential, particularly in enhancing interaction representation. While prior work has injected such potential and even proposed training-free methods, there remain key gaps. Consequently, we propose a novel training-free HOI detection framework for Dynamic Scoring with enhanced semantics (DYSCO) that effectively utilizes textual and visual interaction representations within a multimodal registry, enabling robust and nuanced interaction understanding. This registry incorporates a small set of visual cues and uses innovative interaction signatures to improve the semantic alignment of verbs, facilitating effective generalization to rare interactions. Additionally, we propose a unique multi-head attention mechanism that adaptively weights the contributions of the visual and textual features. Experimental results demonstrate that our DYSCO surpasses training-free state-of-the-art models and is competitive with training-based approaches, particularly excelling in rare interactions. Code is available at https://github.com/francescotonini/dysco.

Paperid: 734, https://arxiv.org/pdf/2507.17294.pdf GitHub GitHub

Abstract:
Tactile feedback is generally recognized to be crucial for effective interaction with the physical world. However, state-of-the-art Vision-Language-Action (VLA) models lack the ability to interpret and use tactile signals, limiting their effectiveness in contact-rich tasks. Incorporating tactile feedback into these systems is challenging due to the absence of large multi-modal datasets. We present VLA-Touch, an approach that enhances generalist robot policies with tactile sensing \emph{without fine-tuning} the base VLA. Our method introduces two key innovations: (1) a pipeline that leverages a pretrained tactile-language model that provides semantic tactile feedback for high-level task planning, and (2) a diffusion-based controller that refines VLA-generated actions with tactile signals for contact-rich manipulation. Through real-world experiments, we demonstrate that our dual-level integration of tactile feedback improves task planning efficiency while enhancing execution precision. Code is open-sourced at \href{https://github.com/jxbi1010/VLA-Touch}{this URL}.

Paperid: 735, https://arxiv.org/pdf/2507.16815.pdf GitHub

Abstract:
Vision-language-action (VLA) reasoning tasks require agents to interpret multimodal instructions, perform long-horizon planning, and act adaptively in dynamic environments. Existing approaches typically train VLA models in an end-to-end fashion, directly mapping inputs to actions without explicit reasoning, which hinders their ability to plan over multiple steps or adapt to complex task variations. In this paper, we propose ThinkAct, a dual-system framework that bridges high-level reasoning with low-level action execution via reinforced visual latent planning. ThinkAct trains a multimodal LLM to generate embodied reasoning plans guided by reinforcing action-aligned visual rewards based on goal completion and trajectory consistency. These reasoning plans are compressed into a visual plan latent that conditions a downstream action model for robust action execution on target environments. Extensive experiments on embodied reasoning and robot manipulation benchmarks demonstrate that ThinkAct enables few-shot adaptation, long-horizon planning, and self-correction behaviors in complex embodied AI tasks.

Paperid: 736, https://arxiv.org/pdf/2507.16716.pdf GitHub

Abstract:
The application of Vision-language foundation models (VLFMs) to remote sensing (RS) imagery has garnered significant attention due to their superior capability in various downstream tasks. A key challenge lies in the scarcity of high-quality, large-scale, image-text paired training data. Recently, several works introduced extensive image-text datasets for RS and trained their VLFMs. However, due to the rudimentary methods used for generating captions, the quality of datasets is suboptimal, requiring larger volumes of training data, while only yielding modest performance improvements. In this paper, we propose a two-stage method named MpGI(Multi-Perspective Generation and Integration) for generating high-quality text captions for RS images. Firstly, we generate distinct and detailed descriptions from different perspectives using Rule-MLLM(Multimodal Large Language Model) Relay Generation and MLLMs generation methods. Next, we utilize Large Language Models (LLMs) to integrate these diverse descriptions into comprehensive captions, capturing details from multiple perspectives. Finally, we have created the HQRS-IT-210K dataset, including about 210,000 RS images and 1.3 million captions. We fine-tuned two VLFMs using our dataset: CLIP, a discriminative model, and CoCa, an image-to-text generative model. This process resulted in our proposed HQRS-CLIP and RS-CoCa models. Experimental results demonstrate that HQRS-CLIP surpassed the previous SOTA RS CLIP model in various downstream tasks while using only 4.2\% of the training data. RS-CoCa outperforms other advanced approaches across benchmark datasets and can generate captions for RS images that rival or even exceed manual annotations. Dataset, pre-trained models, and codes will be released at https://github.com/YiguoHe/HQRS-210K-and-HQRS-CLIP.

Paperid: 737, https://arxiv.org/pdf/2507.16524.pdf GitHub

Abstract:
New era has unlocked exciting possibilities for extending Large Language Models (LLMs) to tackle 3D vision-language tasks. However, most existing 3D multimodal LLMs (MLLMs) rely on compressing holistic 3D scene information or segmenting independent objects to perform these tasks, which limits their spatial awareness due to insufficient representation of the richness inherent in 3D scenes. To overcome these limitations, we propose Spatial 3D-LLM, a 3D MLLM specifically designed to enhance spatial awareness for 3D vision-language tasks by enriching the spatial embeddings of 3D scenes. Spatial 3D-LLM integrates an LLM backbone with a progressive spatial awareness scheme that progressively captures spatial information as the perception field expands, generating location-enriched 3D scene embeddings to serve as visual prompts. Furthermore, we introduce two novel tasks: 3D object distance measurement and 3D layout editing, and construct a 3D instruction dataset, MODEL, to evaluate the model's spatial awareness capabilities. Experimental results demonstrate that Spatial 3D-LLM achieves state-of-the-art performance across a wide range of 3D vision-language tasks, revealing the improvements stemmed from our progressive spatial awareness scheme of mining more profound spatial information. Our code is available at https://github.com/bjshuyuan/Spatial-3D-LLM.

Paperid: 738, https://arxiv.org/pdf/2507.15852.pdf GitHub GitHub

Abstract:
Video Object Segmentation (VOS) is a core task in computer vision, requiring models to track and segment target objects across video frames. Despite notable advances with recent efforts, current techniques still lag behind human capabilities in handling drastic visual variations, occlusions, and complex scene changes. This limitation arises from their reliance on appearance matching, neglecting the human-like conceptual understanding of objects that enables robust identification across temporal dynamics. Motivated by this gap, we propose Segment Concept (SeC), a concept-driven segmentation framework that shifts from conventional feature matching to the progressive construction and utilization of high-level, object-centric representations. SeC employs Large Vision-Language Models (LVLMs) to integrate visual cues across diverse frames, constructing robust conceptual priors. During inference, SeC forms a comprehensive semantic representation of the target based on processed frames, realizing robust segmentation of follow-up frames. Furthermore, SeC adaptively balances LVLM-based semantic reasoning with enhanced feature matching, dynamically adjusting computational efforts based on scene complexity. To rigorously assess VOS methods in scenarios demanding high-level conceptual reasoning and robust semantic understanding, we introduce the Semantic Complex Scenarios Video Object Segmentation benchmark (SeCVOS). SeCVOS comprises 160 manually annotated multi-scenario videos designed to challenge models with substantial appearance variations and dynamic scene transformations. In particular, SeC achieves an 11.8-point improvement over SAM 2.1 on SeCVOS, establishing a new state-of-the-art in concept-aware video object segmentation.

Paperid: 739, https://arxiv.org/pdf/2507.15807.pdf GitHub

Abstract:
Multimodal Large Language Models (MLLMs), built on powerful language backbones, have enabled Multimodal In-Context Learning (MICL)-adapting to new tasks from a few multimodal demonstrations consisting of images, questions, and answers. Despite showing noticeable improvement on standard vision-language datasets, current MLLMs struggle to leverage visual information in the demonstrations. Specifically, they tend to neglect visual cues and over-rely on textual patterns, leading to mere text imitation rather than genuine multimodal adaptation. This behavior makes MICL still unimodal and largely restricts its practical utility. More importantly, this limitation is often concealed by the improved performance on tasks that do not require understanding the visual context. As a result, how to effectively enhance MICL ability and reliably evaluate the MICL performance remains underexplored. To address these issues, we first introduce Dynamic Attention Reallocation (DARA), an efficient fine-tuning strategy that encourages models to attend to the visual context by rebalancing attention across visual and textual tokens. In addition, we present TrueMICL, an MICL-dedicated dataset with both support and test sets that explicitly requires the integration of multimodal information-particularly visual content-for correct task completion. Extensive experiments demonstrate the effectiveness of our holistic solution, showcasing substantial improvements in the true multimodal in-context learning capabilities. Code and datasets are available at https://chenxshuo.github.io/true-micl-colm .

Paperid: 740, https://arxiv.org/pdf/2507.15597.pdf GitHub

Abstract:
We introduce Being-H0, a dexterous Vision-Language-Action model (VLA) trained on large-scale human videos. Existing VLAs struggle with complex manipulation tasks requiring high dexterity and generalize poorly to novel scenarios and tasks, primarily due to their reliance on synthetic data with significant sim-to-real gaps or teleoperated demonstrations lacking scale and diversity. To address this data bottleneck, we propose leveraging human hands as a foundation manipulator, capitalizing on the rich dexterity and scalability present in web data. Our approach centers on physical instruction tuning, a novel training paradigm that combines large-scale VLA pretraining from human videos, physical space alignment for 3D reasoning, and post-training adaptation for robotic tasks. Additionally, we introduce a part-level motion tokenization method which achieves millimeter-level reconstruction accuracy to model precise hand trajectories for action learning. To support our proposed paradigm, we further develop a comprehensive data curation pipeline that integrates heterogeneous sources -- including motion capture, VR, and RGB-only videos -- into a large-scale dataset with millions of motion-based instructional instances. We empirically show the excellence of Being-H0 in hand motion generation and instruction following, and it also scales well with model and data sizes. Importantly, we observe the expected gains of Being-H0 in real-world robotic manipulation as physical instruction tuning is applied. More details are available at https://beingbeyond.github.io/Being-H0.

Paperid: 741, https://arxiv.org/pdf/2507.15576.pdf GitHub GitHub

Abstract:
Terahertz (THz) imaging enables non-invasive analysis for applications such as security screening and material classification, but effective image classification remains challenging due to limited annotations, low resolution, and visual ambiguity. We introduce In-Context Learning (ICL) with Vision-Language Models (VLMs) as a flexible, interpretable alternative that requires no fine-tuning. Using a modality-aligned prompting framework, we adapt two open-weight VLMs to the THz domain and evaluate them under zero-shot and one-shot settings. Our results show that ICL improves classification and interpretability in low-data regimes. This is the first application of ICL-enhanced VLMs to THz imaging, offering a promising direction for resource-constrained scientific domains. Code: \href{https://github.com/Nicolas-Poggi/Project_THz_Classification/tree/main}{GitHub repository}.

Paperid: 742, https://arxiv.org/pdf/2507.15542.pdf GitHub

Abstract:
Zero-shot human-object interaction (HOI) detection remains a challenging task, particularly in generalizing to unseen actions. Existing methods address this challenge by tapping Vision-Language Models (VLMs) to access knowledge beyond the training data. However, they either struggle to distinguish actions involving the same object or demonstrate limited generalization to unseen classes. In this paper, we introduce HOLa (Zero-Shot HOI Detection with Low-Rank Decomposed VLM Feature Adaptation), a novel approach that both enhances generalization to unseen classes and improves action distinction. In training, HOLa decomposes VLM text features for given HOI classes via low-rank factorization, producing class-shared basis features and adaptable weights. These features and weights form a compact HOI representation that preserves shared information across classes, enhancing generalization to unseen classes. Subsequently, we refine action distinction by adapting weights for each HOI class and introducing human-object tokens to enrich visual interaction representations. To further distinguish unseen actions, we guide the weight adaptation with LLM-derived action regularization. Experimental results show that our method sets a new state-of-the-art across zero-shot HOI settings on HICO-DET, achieving an unseen-class mAP of 27.91 in the unseen-verb setting. Our code is available at https://github.com/ChelsieLei/HOLa.

Paperid: 743, https://arxiv.org/pdf/2507.15480.pdf GitHub GitHub

Abstract:
Pretrained vision-language models (VLMs), such as CLIP, achieve remarkable zero-shot performance, yet their downstream potential hinges on effective fine-tuning. Most adaptation methods typically focus on refining representation from separate modalities (text or vision) but neglect the critical role of their fused representations in the decision-making process, \emph{\ie} rational matrix that drives the final prediction. To bridge the gap, we propose a simple yet effective \textbf{R}ational \textbf{Ada}ptaion ({RAda}) to explicitly exploit the final fused representation during fine-tuning. RAda employs a learned mask, obtained from a lightweight attention layer attached at the end of a VLM, to dynamically calibrate the contribution of each element in the rational matrix, enabling targeted adjustments to the final cross-modal interactions without incurring costly modifications to intermediate features. Experiments in different settings (i.e., updating, or freezing pretrained encoders in adaptation, and test-time training that can only access the unlabeled test data) show that RAda serves as a versatile fine-tuning technique, improving the baseline with minimal code and performing comparably against current arts in most settings. Code is available at \href{https://github.com/khufia/RAda/tree/main}{github.com/khufia/RAda}.

Paperid: 744, https://arxiv.org/pdf/2507.15321.pdf GitHub

Abstract:
Depth estimation is a fundamental task in computer vision with diverse applications. Recent advancements in deep learning have led to powerful depth foundation models (DFMs), yet their evaluation remains challenging due to inconsistencies in existing protocols. Traditional benchmarks rely on alignment-based metrics that introduce biases, favor certain depth representations, and complicate fair comparisons. In this work, we propose BenchDepth, a new benchmark that evaluates DFMs through five carefully selected downstream proxy tasks: depth completion, stereo matching, monocular feed-forward 3D scene reconstruction, SLAM, and vision-language spatial understanding. Unlike conventional evaluation protocols, our approach assesses DFMs based on their practical utility in real-world applications, bypassing problematic alignment procedures. We benchmark eight state-of-the-art DFMs and provide an in-depth analysis of key findings and observations. We hope our work sparks further discussion in the community on best practices for depth model evaluation and paves the way for future research and advancements in depth estimation.

Paperid: 745, https://arxiv.org/pdf/2507.15266.pdf GitHub

Abstract:
Scene understanding and risk-aware attentions are crucial for human drivers to make safe and effective driving decisions. To imitate this cognitive ability in urban autonomous driving while ensuring the transparency and interpretability, we propose a vision-language model (VLM)-enhanced unified decision-making and motion control framework, named VLM-UDMC. This framework incorporates scene reasoning and risk-aware insights into an upper-level slow system, which dynamically reconfigures the optimal motion planning for the downstream fast system. The reconfiguration is based on real-time environmental changes, which are encoded through context-aware potential functions. More specifically, the upper-level slow system employs a two-step reasoning policy with Retrieval-Augmented Generation (RAG), leveraging foundation models to process multimodal inputs and retrieve contextual knowledge, thereby generating risk-aware insights. Meanwhile, a lightweight multi-kernel decomposed LSTM provides real-time trajectory predictions for heterogeneous traffic participants by extracting smoother trend representations for short-horizon trajectory prediction. The effectiveness of the proposed VLM-UDMC framework is verified via both simulations and real-world experiments with a full-size autonomous vehicle. It is demonstrated that the presented VLM-UDMC effectively leverages scene understanding and attention decomposition for rational driving decisions, thus improving the overall urban driving performance. Our open-source project is available at https://github.com/henryhcliu/vlmudmc.git.

Paperid: 746, https://arxiv.org/pdf/2507.15227.pdf GitHub

Abstract:
Interpretability is critical in high-stakes domains such as medical imaging, where understanding model decisions is essential for clinical adoption. In this work, we introduce Sparse Autoencoder (SAE)-based interpretability to breast imaging by analyzing {Mammo-CLIP}, a vision--language foundation model pretrained on large-scale mammogram image--report pairs. We train a patch-level \texttt{Mammo-SAE} on Mammo-CLIP to identify and probe latent features associated with clinically relevant breast concepts such as \textit{mass} and \textit{suspicious calcification}. Our findings reveal that top activated class level latent neurons in the SAE latent space often tend to align with ground truth regions, and also uncover several confounding factors influencing the model's decision-making process. Additionally, we analyze which latent neurons the model relies on during downstream finetuning for improving the breast concept prediction. This study highlights the promise of interpretable SAE latent representations in providing deeper insight into the internal workings of foundation models at every layer for breast imaging. The code will be released at https://krishnakanthnakka.github.io/MammoSAE/

Paperid: 747, https://arxiv.org/pdf/2507.14976.pdf GitHub

Abstract:
Pre-trained Vision-Language Models (VLMs) such as CLIP have shown excellent generalization abilities. However, adapting these large-scale models to downstream tasks while preserving their generalization capabilities remains challenging. Although prompt learning methods have shown promise, they suffer from two fundamental bottlenecks that limit generalization: (a) modality isolation, and (b) hierarchical semantic decay. To address these limitations, we propose HiCroPL, a Hierarchical Cross-modal Prompt Learning framework that establishes bidirectional knowledge flow between text and vision modalities, enabling them to refine their semantics mutually. HiCroPL routes knowledge flows by leveraging the complementary strengths of text and vision. In early layers, text prompts inject relatively clear semantics into visual prompts through a hierarchical knowledge mapper, enhancing the representation of low-level visual semantics. In later layers, visual prompts encoding specific task-relevant objects flow back to refine text prompts, enabling deeper alignment. Crucially, our hierarchical knowledge mapper allows representations at multi-scales to be fused, ensuring that deeper representations retain transferable shallow semantics thereby enhancing generalization. We further introduce a lightweight layer-specific knowledge proxy to enable efficient cross-modal interactions. Extensive evaluations across four tasks demonstrate HiCroPL's superior performance, achieving state-of-the-art results on 11 benchmarks with significant improvements. Code is available at: https://github.com/zzeoZheng/HiCroPL.

Paperid: 748, https://arxiv.org/pdf/2507.14298.pdf GitHub

Abstract:
Recent methods for customizing Large Vision Language Models (LVLMs) for domain-specific tasks have shown promising results in scientific chart comprehension. However, existing approaches face two major limitations: First, they rely on paired data from only a few chart types, limiting generalization to wide range of chart types. Secondly, they lack targeted pre-training for chart-data alignment, which hampers the model's understanding of underlying data. In this paper, we introduce ChartScope, an LVLM optimized for in-depth chart comprehension across diverse chart types. We propose an efficient data generation pipeline that synthesizes paired data for a wide range of chart types, along with a novel Dual-Path training strategy that enabling the model to succinctly capture essential data details while preserving robust reasoning capabilities by incorporating reasoning over the underlying data. Lastly, we establish ChartDQA, a new benchmark for evaluating not only question-answering at different levels but also underlying data understanding. Experimental results demonstrate that ChartScope significantly enhances comprehension on a wide range of chart types. The code and data are available at https://davidhalladay.github.io/chartscope_demo.

Paperid: 749, https://arxiv.org/pdf/2507.14049.pdf GitHub GitHub

Abstract:
Vision-Language Models (VLMs) have emerged as a promising approach to address the data scarcity challenge in robotics, enabling the development of generalizable visuomotor control policies. While models like OpenVLA showcase the potential of this paradigm, deploying large-scale VLMs on resource-constrained mobile manipulation systems remains a significant hurdle. This paper introduces Edge VLA (EVLA), a novel approach designed to significantly enhance the inference speed of Vision-Language-Action (VLA) models. EVLA maintains the representational power of these models while enabling real-time performance on edge devices. We achieve this through two key innovations: 1) Eliminating the autoregressive requirement for end-effector position prediction, leading to a 7x speedup in inference, and 2) Leveraging the efficiency of Small Language Models (SLMs), demonstrating comparable training performance to larger models with significantly reduced computational demands. Our early results demonstrate that EVLA achieves comparable training characteristics to OpenVLA while offering substantial gains in inference speed and memory efficiency. We release our model checkpoints and training \href{https://github.com/kscalelabs/evla }{codebase} to foster further research.

Paperid: 750, https://arxiv.org/pdf/2507.13363.pdf GitHub

Abstract:
Modern 3D object detection datasets are constrained by narrow class taxonomies and costly manual annotations, limiting their ability to scale to open-world settings. In contrast, 2D vision-language models trained on web-scale image-text pairs exhibit rich semantic understanding and support open-vocabulary detection via natural language prompts. In this work, we leverage the maturity and category diversity of 2D foundation models to perform open-vocabulary 3D object detection without any human-annotated 3D labels. Our pipeline uses a 2D vision-language detector to generate text-conditioned proposals, which are segmented with SAM and back-projected into 3D using camera geometry and either LiDAR or monocular pseudo-depth. We introduce a geometric inflation strategy based on DBSCAN clustering and Rotating Calipers to infer 3D bounding boxes without training. To simulate adverse real-world conditions, we construct Pseudo-nuScenes, a fog-augmented, RGB-only variant of the nuScenes dataset. Experiments demonstrate that our method achieves competitive localization performance across multiple settings, including LiDAR-based and purely RGB-D inputs, all while remaining training-free and open-vocabulary. Our results highlight the untapped potential of 2D foundation models for scalable 3D perception. We open-source our code and resources at https://github.com/atharv0goel/open-world-3D-det.

Paperid: 751, https://arxiv.org/pdf/2507.13362.pdf GitHub

Abstract:
This study investigates the spatial reasoning capabilities of vision-language models (VLMs) through Chain-of-Thought (CoT) prompting and reinforcement learning. We begin by evaluating the impact of different prompting strategies and find that simple CoT formats, where the model generates a reasoning step before the answer, not only fail to help, but can even harm the model's original performance. In contrast, structured multi-stage prompting based on scene graphs (SceneGraph CoT) significantly improves spatial reasoning accuracy. Furthermore, to improve spatial reasoning ability, we fine-tune models using Group Relative Policy Optimization (GRPO) on the SAT dataset and evaluate their performance on CVBench. Compared to supervised fine-tuning (SFT), GRPO achieves higher accuracy on Pass@1 evaluations and demonstrates superior robustness under out-of-distribution (OOD) conditions. In particular, we find that SFT overfits to surface-level linguistic patterns and may degrade performance when test-time phrasing changes (e.g., from "closer to" to "farther from"). GRPO, on the other hand, generalizes more reliably and maintains stable performance under such shifts. Our findings provide insights into how reinforcement learning and structured prompting improve the spatial reasoning capabilities and generalization behavior of modern VLMs. All code is open source at: https://github.com/Yvonne511/spatial-vlm-investigator

Paperid: 752, https://arxiv.org/pdf/2507.13348.pdf GitHub GitHub

Abstract:
Recent advancements in vision-language models (VLMs) have improved performance by increasing the number of visual tokens, which are often significantly longer than text tokens. However, we observe that most real-world scenarios do not require such an extensive number of visual tokens. While the performance drops significantly in a small subset of OCR-related tasks, models still perform accurately in most other general VQA tasks with only 1/4 resolution. Therefore, we propose to dynamically process distinct samples with different resolutions, and present a new paradigm for visual token compression, namely, VisionThink. It starts with a downsampled image and smartly decides whether it is sufficient for problem solving. Otherwise, the model could output a special token to request the higher-resolution image. Compared to existing Efficient VLM methods that compress tokens using fixed pruning ratios or thresholds, VisionThink autonomously decides whether to compress tokens case by case. As a result, it demonstrates strong fine-grained visual understanding capability on OCR-related tasks, and meanwhile saves substantial visual tokens on simpler tasks. We adopt reinforcement learning and propose the LLM-as-Judge strategy to successfully apply RL to general VQA tasks. Moreover, we carefully design a reward function and penalty mechanism to achieve a stable and reasonable image resize call ratio. Extensive experiments demonstrate the superiority, efficiency, and effectiveness of our method. Our code is available at https://github.com/dvlab-research/VisionThink.

Paperid: 753, https://arxiv.org/pdf/2507.12768.pdf GitHub

Abstract:
Vision-language-action (VLA) models have shown promise on task-conditioned control in complex settings such as bimanual manipulation. However, the heavy reliance on task-specific human demonstrations limits their generalization and incurs high data acquisition costs. In this work, we present a new notion of task-agnostic action paradigm that decouples action execution from task-specific conditioning, enhancing scalability, efficiency, and cost-effectiveness. To address the data collection challenges posed by this paradigm -- such as low coverage density, behavioral redundancy, and safety risks -- we introduce ATARA (Automated Task-Agnostic Random Actions), a scalable self-supervised framework that accelerates collection by over $ 30\times $ compared to human teleoperation. To further enable effective learning from task-agnostic data, which often suffers from distribution mismatch and irrelevant trajectories, we propose AnyPos, an inverse dynamics model equipped with Arm-Decoupled Estimation and a Direction-Aware Decoder (DAD). We additionally integrate a video-conditioned action validation module to verify the feasibility of learned policies across diverse manipulation tasks. Extensive experiments show that the AnyPos-ATARA pipeline yields a 51% improvement in test accuracy and achieves 30-40% higher success rates in downstream tasks such as lifting, pick-and-place, and clicking, using replay-based video validation. Project Page: https://embodiedfoundation.github.io/vidar_anypos

Paperid: 754, https://arxiv.org/pdf/2507.12698.pdf GitHub

Abstract:
Medical image synthesis presents unique challenges due to the inherent complexity and high-resolution details required in clinical contexts. Traditional generative architectures such as Generative Adversarial Networks (GANs) or Variational Auto Encoder (VAEs) have shown great promise for high-resolution image generation but struggle with preserving fine-grained details that are key for accurate diagnosis. To address this issue, we introduce Pixel Perfect MegaMed, the first vision-language foundation model to synthesize images at resolutions of 1024x1024. Our method deploys a multi-scale transformer architecture designed specifically for ultra-high resolution medical image generation, enabling the preservation of both global anatomical context and local image-level details. By leveraging vision-language alignment techniques tailored to medical terminology and imaging modalities, Pixel Perfect MegaMed bridges the gap between textual descriptions and visual representations at unprecedented resolution levels. We apply our model to the CheXpert dataset and demonstrate its ability to generate clinically faithful chest X-rays from text prompts. Beyond visual quality, these high-resolution synthetic images prove valuable for downstream tasks such as classification, showing measurable performance gains when used for data augmentation, particularly in low-data regimes. Our code is accessible through the project website - https://tehraninasab.github.io/pixelperfect-megamed.

Paperid: 755, https://arxiv.org/pdf/2507.12644.pdf GitHub

Abstract:
Tool design and use reflect the ability to understand and manipulate the physical world through creativity, planning, and foresight. As such, these capabilities are often regarded as measurable indicators of intelligence across biological species. While much of today's research on robotic intelligence focuses on generating better controllers, inventing smarter tools offers a complementary form of physical intelligence: shifting the onus of problem-solving onto the tool's design. Given the vast and impressive common-sense, reasoning, and creative capabilities of today's foundation models, we investigate whether these models can provide useful priors to automatically design and effectively wield such tools? We present VLMgineer, a framework that harnesses the code generation abilities of vision language models (VLMs) together with evolutionary search to iteratively co-design physical tools and the action plans that operate them to perform a task. We evaluate VLMgineer on a diverse new benchmark of everyday manipulation scenarios that demand creative tool design and use. Across this suite, VLMgineer consistently discovers tools and policies that solve tasks more effectively and innovatively, transforming challenging robotics problems into straightforward executions. It also outperforms VLM-generated designs from human specifications and existing human-crafted tools for everyday tasks. To facilitate future research on automated tool invention, we will release our benchmark and code.

Paperid: 756, https://arxiv.org/pdf/2507.12621.pdf GitHub

Abstract:
Traditional volume visualization (VolVis) methods, like direct volume rendering, suffer from rigid transfer function designs and high computational costs. Although novel view synthesis approaches enhance rendering efficiency, they require additional learning effort for non-experts and lack support for semantic-level interaction. To bridge this gap, we propose NLI4VolVis, an interactive system that enables users to explore, query, and edit volumetric scenes using natural language. NLI4VolVis integrates multi-view semantic segmentation and vision-language models to extract and understand semantic components in a scene. We introduce a multi-agent large language model architecture equipped with extensive function-calling tools to interpret user intents and execute visualization tasks. The agents leverage external tools and declarative VolVis commands to interact with the VolVis engine powered by 3D editable Gaussians, enabling open-vocabulary object querying, real-time scene editing, best-view selection, and 2D stylization. We validate our system through case studies and a user study, highlighting its improved accessibility and usability in volumetric data exploration. We strongly recommend readers check our case studies, demo video, and source code at https://nli4volvis.github.io/.

Paperid: 757, https://arxiv.org/pdf/2507.12508.pdf GitHub

Abstract:
Spatial reasoning in 3D space is central to human cognition and indispensable for embodied tasks such as navigation and manipulation. However, state-of-the-art vision-language models (VLMs) struggle frequently with tasks as simple as anticipating how a scene will look after an egocentric motion: they perceive 2D images but lack an internal model of 3D dynamics. We therefore propose MindJourney, a test-time scaling framework that grants a VLM with this missing capability by coupling it to a controllable world model based on video diffusion. The VLM iteratively sketches a concise camera trajectory, while the world model synthesizes the corresponding view at each step. The VLM then reasons over this multi-view evidence gathered during the interactive exploration. Without any fine-tuning, our MindJourney achieves over an average 8% performance boost on the representative spatial reasoning benchmark SAT, showing that pairing VLMs with world models for test-time scaling offers a simple, plug-and-play route to robust 3D reasoning. Meanwhile, our method also improves upon the test-time inference VLMs trained through reinforcement learning, which demonstrates the potential of our method that utilizes world models for test-time scaling.

Paperid: 758, https://arxiv.org/pdf/2507.12499.pdf GitHub GitHub

Abstract:
End-to-end autonomous driving has emerged as a promising approach to unify perception, prediction, and planning within a single framework, reducing information loss and improving adaptability. However, existing methods often rely on fixed and sparse trajectory supervision, limiting their ability to capture the hierarchical reasoning process that human drivers naturally employ. To bridge this gap, we propose ReAL-AD, a Reasoning-Augmented Learning framework that structures decision-making in autonomous driving based on the three-tier human cognitive model: Driving Strategy, Driving Decision, and Driving Operation, where Vision-Language Models (VLMs) are incorporated to enhance situational awareness and structured reasoning across these levels. Specifically, we introduce: (1) the Strategic Reasoning Injector, which formulates high-level driving strategies by interpreting complex traffic contexts from VLM-generated insights; (2) the Tactical Reasoning Integrator, which refines strategic intent into interpretable tactical choices such as lane changes, overtaking, and speed adjustments; and (3) the Hierarchical Trajectory Decoder, which progressively translates tactical decisions into precise control actions for smooth and human-like trajectory execution. Extensive evaluations show that integrating our framework improves planning accuracy and safety by over 30%, making end-to-end autonomous driving more interpretable and aligned with human-like hierarchical reasoning. The project page can be found at: \href{https://4dvlab.github.io/project_page/realad}{\texttt{4dvlab.github.io/project\_page/realad}}

Paperid: 759, https://arxiv.org/pdf/2507.12441.pdf GitHub

Abstract:
Recent progress has been made in region-aware vision-language modeling, particularly with the emergence of the Describe Anything Model (DAM). DAM is capable of generating detailed descriptions of any specific image areas or objects without the need for additional localized image-text alignment supervision. We hypothesize that such region-level descriptive capability is beneficial for the task of Visual Question Answering (VQA), especially in challenging scenarios involving images with dense text. In such settings, the fine-grained extraction of textual information is crucial to producing correct answers. Motivated by this, we introduce DAM-QA, a framework with a tailored evaluation protocol, developed to investigate and harness the region-aware capabilities from DAM for the text-rich VQA problem that requires reasoning over text-based information within images. DAM-QA incorporates a mechanism that aggregates answers from multiple regional views of image content, enabling more effective identification of evidence that may be tied to text-related elements. Experiments on six VQA benchmarks show that our approach consistently outperforms the baseline DAM, with a notable 7+ point gain on DocVQA. DAM-QA also achieves the best overall performance among region-aware models with fewer parameters, significantly narrowing the gap with strong generalist VLMs. These results highlight the potential of DAM-like models for text-rich and broader VQA tasks when paired with efficient usage and integration strategies. Our code is publicly available at https://github.com/Linvyl/DAM-QA.git.

Paperid: 760, https://arxiv.org/pdf/2507.12440.pdf GitHub

Abstract:
Real robot data collection for imitation learning has led to significant advancements in robotic manipulation. However, the requirement for robot hardware in the process fundamentally constrains the scale of the data. In this paper, we explore training Vision-Language-Action (VLA) models using egocentric human videos. The benefit of using human videos is not only for their scale but more importantly for the richness of scenes and tasks. With a VLA trained on human video that predicts human wrist and hand actions, we can perform Inverse Kinematics and retargeting to convert the human actions to robot actions. We fine-tune the model using a few robot manipulation demonstrations to obtain the robot policy, namely EgoVLA. We propose a simulation benchmark called Ego Humanoid Manipulation Benchmark, where we design diverse bimanual manipulation tasks with demonstrations. We fine-tune and evaluate EgoVLA with Ego Humanoid Manipulation Benchmark and show significant improvements over baselines and ablate the importance of human data. Videos can be found on our website: https://rchalyang.github.io/EgoVLA

Paperid: 761, https://arxiv.org/pdf/2507.12236.pdf GitHub

Abstract:
Phrase grounding, i.e., mapping natural language phrases to specific image regions, holds significant potential for disease localization in medical imaging through clinical reports. While current state-of-the-art methods rely on discriminative, self-supervised contrastive models, we demonstrate that generative text-to-image diffusion models, leveraging cross-attention maps, can achieve superior zero-shot phrase grounding performance. Contrary to prior assumptions, we show that fine-tuning diffusion models with a frozen, domain-specific language model, such as CXR-BERT, substantially outperforms domain-agnostic counterparts. This setup achieves remarkable improvements, with mIoU scores doubling those of current discriminative methods. These findings highlight the underexplored potential of generative models for phrase grounding tasks. To further enhance performance, we introduce Bimodal Bias Merging (BBM), a novel post-processing technique that aligns text and image biases to identify regions of high certainty. BBM refines cross-attention maps, achieving even greater localization accuracy. Our results establish generative approaches as a more effective paradigm for phrase grounding in the medical imaging domain, paving the way for more robust and interpretable applications in clinical practice. The source code and model weights are available at https://github.com/Felix-012/generate_to_ground.

Paperid: 762, https://arxiv.org/pdf/2507.12060.pdf GitHub

Abstract:
Face anti-spoofing (FAS) aims to construct a robust system that can withstand diverse attacks. While recent efforts have concentrated mainly on cross-domain generalization, two significant challenges persist: limited semantic understanding of attack types and training redundancy across domains. We address the first by integrating vision-language models (VLMs) to enhance the perception of visual input. For the second challenge, we employ a meta-domain strategy to learn a unified model that generalizes well across multiple domains. Our proposed InstructFLIP is a novel instruction-tuned framework that leverages VLMs to enhance generalization via textual guidance trained solely on a single domain. At its core, InstructFLIP explicitly decouples instructions into content and style components, where content-based instructions focus on the essential semantics of spoofing, and style-based instructions consider variations related to the environment and camera characteristics. Extensive experiments demonstrate the effectiveness of InstructFLIP by outperforming SOTA models in accuracy and substantially reducing training redundancy across diverse domains in FAS. Project website is available at https://kunkunlin1221.github.io/InstructFLIP.

Paperid: 763, https://arxiv.org/pdf/2507.10548.pdf GitHub

Abstract:
Recent advanced vision-language models(VLMs) have demonstrated strong performance on passive, offline image and video understanding tasks. However, their effectiveness in embodied settings, which require online interaction and active scene understanding remains limited. In such scenarios, an agent perceives the environment from a first-person perspective, with each action dynamically shaping subsequent observations. Even state-of-the-art models such as GPT-4o, Claude 3.5 Sonnet, and Gemini 2.5 Pro struggle in open-environment interactions, exhibiting clear limitations in spatial reasoning and long-horizon planning. To address this gap, we introduce EmRACE-3K, a dataset of over 3,000 language-guided tasks situated in diverse, photorealistic environments constructed using Unreal Engine and the UnrealCV-Zoo framework. The tasks encompass a wide range of embodied challenges, including navigation, object manipulation, and multi-stage goal execution. Each task unfolds as a multi-step trajectory, pairing first-person visual observations with high-level instructions, grounded actions, and natural language rationales that express the agent's intent at every step. Using EmRACE-3K, we establish a benchmark to evaluate the embodied reasoning capabilities of VLMs across three key dimensions: Exploration, Dynamic Spatial-Semantic Reasoning, and Multi-stage Goal Execution. In zero-shot settings, all models achieve success rates below 20%, underscoring the challenge posed by our benchmark and the current limitations of VLMs in interactive environments. To demonstrate the utility of EmRACE-3K, we further fine-tune Qwen2.5-VL-7B using supervised learning followed by reinforcement learning. This approach yields substantial improvements across all three challenge categories, highlighting the dataset's effectiveness in enabling the development of embodied reasoning capabilities.

Paperid: 764, https://arxiv.org/pdf/2507.10225.pdf GitHub

Abstract:
Pre-trained vision-language models have exhibited remarkable abilities in detecting out-of-distribution (OOD) samples. However, some challenging OOD samples, which lie close to in-distribution (InD) data in image feature space, can still lead to misclassification. The emergence of foundation models like diffusion models and multimodal large language models (MLLMs) offers a potential solution to this issue. In this work, we propose SynOOD, a novel approach that harnesses foundation models to generate synthetic, challenging OOD data for fine-tuning CLIP models, thereby enhancing boundary-level discrimination between InD and OOD samples. Our method uses an iterative in-painting process guided by contextual prompts from MLLMs to produce nuanced, boundary-aligned OOD samples. These samples are refined through noise adjustments based on gradients from OOD scores like the energy score, effectively sampling from the InD/OOD boundary. With these carefully synthesized images, we fine-tune the CLIP image encoder and negative label features derived from the text encoder to strengthen connections between near-boundary OOD samples and a set of negative labels. Finally, SynOOD achieves state-of-the-art performance on the large-scale ImageNet benchmark, with minimal increases in parameters and runtime. Our approach significantly surpasses existing methods, and the code is available at https://github.com/Jarvisgivemeasuit/SynOOD.

Paperid: 765, https://arxiv.org/pdf/2507.10202.pdf GitHub

Abstract:
Multimodal Large Language Models (MLLMs) have demonstrated remarkable capabilities in vision-language understanding, reasoning, and generation. However, they struggle with tasks requiring fine-grained localization and reasoning in high-resolution images. This constraint stems from the fact that MLLMs are fine-tuned with fixed image resolution to align with the pre-trained image encoder used in MLLM. Consequently, feeding high-resolution images directly into MLLMs leads to poor generalization due to a train-test resolution discrepancy, while downsampling these images-although ensuring consistency-compromises fine-grained visual details and ultimately degrades performance. To address this challenge, we propose Extract Candidate then Predict (ECP), a novel training-free, task-agnostic two-stage framework designed to enhance MLLM performance on high-resolution images. The key intuition behind ECP is that while MLLMs struggle with high-resolution images, their predictions on downsampled images still contain implicit localization cues. By first identifying candidate region using the coarse prediction and then predicting the final output based on candidate region, ECP effectively preserves fine-grained details while mitigating the challenges posed by high-resolution data. We validate our framework on 4K GUI grounding and 4K, 8K MLLM perception, achieving +21.3%, +5.8%, +5.2% absolute improvement compared to baseline respectively, demonstrating its effectiveness. Code is available at https://github.com/yenncye/ECP.

Paperid: 766, https://arxiv.org/pdf/2507.09795.pdf GitHub

Abstract:
Recent advancements in Vision-Language Models like CLIP have enabled zero-shot OOD detection by leveraging both image and textual label information. Among these, negative label-based methods such as NegLabel and CSP have shown promising results by utilizing a lexicon of words to define negative labels for distinguishing OOD samples. However, these methods suffer from detecting in-distribution samples as OOD due to negative labels that are subcategories of in-distribution labels or proper nouns. They also face limitations in handling images that match multiple in-distribution and negative labels. We propose NegRefine, a novel negative label refinement framework for zero-shot OOD detection. By introducing a filtering mechanism to exclude subcategory labels and proper nouns from the negative label set and incorporating a multi-matching-aware scoring function that dynamically adjusts the contributions of multiple labels matching an image, NegRefine ensures a more robust separation between in-distribution and OOD samples. We evaluate NegRefine on large-scale benchmarks, including ImageNet-1K. The code is available at https://github.com/ah-ansari/NegRefine.

Paperid: 767, https://arxiv.org/pdf/2507.09524.pdf GitHub

Abstract:
Recent advancements in unpaired dehazing, particularly those using GANs, show promising performance in processing real-world hazy images. However, these methods tend to face limitations due to the generator's limited transport mapping capability, which hinders the full exploitation of their effectiveness in unpaired training paradigms. To address these challenges, we propose DehazeSB, a novel unpaired dehazing framework based on the SchrÃ¶dinger Bridge. By leveraging optimal transport (OT) theory, DehazeSB directly bridges the distributions between hazy and clear images. This enables optimal transport mappings from hazy to clear images in fewer steps, thereby generating high-quality results. To ensure the consistency of structural information and details in the restored images, we introduce detail-preserving regularization, which enforces pixel-level alignment between hazy inputs and dehazed outputs. Furthermore, we propose a novel prompt learning to leverage pre-trained CLIP models in distinguishing hazy images and clear ones, by learning a haze-aware vision-language alignment. Extensive experiments on multiple real-world datasets demonstrate our method's superiority. Code: https://github.com/ywxjm/DehazeSB.

Paperid: 768, https://arxiv.org/pdf/2507.09500.pdf GitHub

Abstract:
Vision-language models (VLMs) exhibit remarkable zero-shot capabilities but struggle with distribution shifts in downstream tasks when labeled data is unavailable, which has motivated the development of Test-Time Adaptation (TTA) to improve VLMs' performance during inference without annotations. Among various TTA approaches, cache-based methods show promise by preserving historical knowledge from low-entropy samples in a dynamic cache and fostering efficient adaptation. However, these methods face two critical reliability challenges: (1) entropy often becomes unreliable under distribution shifts, causing error accumulation in the cache and degradation in adaptation performance; (2) the final predictions may be unreliable due to inflexible decision boundaries that fail to accommodate large downstream shifts. To address these challenges, we propose a Reliable Test-time Adaptation (ReTA) method that integrates two complementary strategies to enhance reliability from two perspectives. First, to mitigate the unreliability of entropy as a sample selection criterion for cache construction, we introduce Consistency-aware Entropy Reweighting (CER), which incorporates consistency constraints to weight entropy during cache updating. While conventional approaches rely solely on low entropy for cache prioritization and risk introducing noise, our method leverages predictive consistency to maintain a high-quality cache and facilitate more robust adaptation. Second, we present Diversity-driven Distribution Calibration (DDC), which models class-wise text embeddings as multivariate Gaussian distributions, enabling adaptive decision boundaries for more accurate predictions across visually diverse content. Extensive experiments demonstrate that ReTA consistently outperforms state-of-the-art methods, particularly under real-world distribution shifts. Code: https://github.com/Evelyn1ywliang/ReTA.

Paperid: 769, https://arxiv.org/pdf/2507.09222.pdf GitHub

Abstract:
Foundation models like CLIP and SAM have advanced computer vision and medical imaging via low-shot transfer learning, aiding CADD with limited data. However, their deployment faces two key challenges. \textit{distribution shift} where pre-training and post-training data distributions differ (e.g., due to inter-center image acquisition) and \textit{confidence misalignment}, which leads to overconfident errors. These issues surface differently, vision-language models (e.g., CLIP) suffer from 2D embedding shift (image-text misalignment), while medical models (e.g., SAM) encounter 3D domain shifts (e.g., scanner variation) and voxel-wise calibration need. Existing solutions are domain-specific. We propose \textbf{StaRFM}, a fusion of Fisher information penalty (FIP) and confidence misalignment penalty (CMP) tackling both challenges. It applies FIP, extended to 3D via patch-wise regularization, to reduce embedding shift, and CMP, reformulated for voxel-level predictions, to calibrate segmentation uncertainty. We derive PAC-Bayes bounds. FIP controls generalization via the Fisher-Rao norm, and CMP reduces calibration error via Brier score minimization. StaRFM surpasses baselines by \texttt{+}3.5\% accuracy and 28\% lower ECE on 19 vision datasets (e.g., ImageNet, Office-Home), achieves +4.2\% DSC over SAM-FT and 4.8mm HD95 on medical benchmarks (e.g., BraTS, ATLAS), and reduces cross-domain gaps by up to 20\%. The framework is plug-and-play, requiring minimal architectural changes. Code and models are available at: \href{https://anonymous.4open.science/r/StaRFM-C0CD/}{\textcolor{blue}{\underline{StaRFM}}}

Paperid: 770, https://arxiv.org/pdf/2507.09184.pdf GitHub

Abstract:
Hallucinations pose a significant challenge in Large Vision Language Models (LVLMs), with misalignment between multimodal features identified as a key contributing factor. This paper reveals the negative impact of the long-term decay in Rotary Position Encoding (RoPE), used for positional modeling in LVLMs, on multimodal alignment. Concretely, under long-term decay, instruction tokens exhibit uneven perception of image tokens located at different positions within the two-dimensional space: prioritizing image tokens from the bottom-right region since in the one-dimensional sequence, these tokens are positionally closer to the instruction tokens. This biased perception leads to insufficient image-instruction interaction and suboptimal multimodal alignment. We refer to this phenomenon as image alignment bias. To enhance instruction's perception of image tokens at different spatial locations, we propose MCA-LLaVA, based on Manhattan distance, which extends the long-term decay to a two-dimensional, multi-directional spatial decay. MCA-LLaVA integrates the one-dimensional sequence order and two-dimensional spatial position of image tokens for positional modeling, mitigating hallucinations by alleviating image alignment bias. Experimental results of MCA-LLaVA across various hallucination and general benchmarks demonstrate its effectiveness and generality. The code can be accessed in https://github.com/ErikZ719/MCA-LLaVA.

Paperid: 771, https://arxiv.org/pdf/2507.09155.pdf GitHub GitHub

Abstract:
This work presents OPENXRD, an open-book pipeline designed for crystallography question answering, which integrates textual prompts with concise supporting content generated by GPT-4.5. Instead of using scanned textbooks, which may lead to copyright issues, OPENXRD generates compact, domain-specific references that help smaller models understand key concepts in X-ray diffraction (XRD). We evaluate OPENXRD on a well-defined set of 217 expert-level XRD questions by comparing different vision-language models, including GPT-4 and LLaVA-based frameworks such as Mistral, LLaMA, and QWEN, under both closed-book (without supporting material) and open-book (with supporting material) conditions. Our experimental results show significant accuracy improvements in models that use the GPT-4.5-generated summaries, particularly those with limited prior training in crystallography. OPENXRD uses knowledge from larger models to fill knowledge gaps in crystallography and shows that AI-generated texts can help smaller models reason more effectively in scientific tasks. While the current version of OPENXRD focuses on text-based inputs, we also explore future extensions such as adding real crystal diagrams or diffraction patterns to improve interpretation in specialized materials science contexts. Overall, OPENXRD shows that specialized open-book systems can be useful in materials science and provides a foundation for broader natural language processing (NLP) tools in critical scientific fields.

Paperid: 772, https://arxiv.org/pdf/2507.09139.pdf GitHub

Abstract:
Human pose estimation traditionally relies on architectures that encode keypoint priors, limiting their generalization to novel poses or unseen keypoints. Recent language-guided approaches like LocLLM reformulate keypoint localization as a vision-language task, enabling zero-shot generalization through textual descriptions. However, LocLLM's linear projector fails to capture complex spatial-textual interactions critical for high-precision localization. To address this, we propose PoseLLM, the first Large Language Model (LLM)-based pose estimation framework that replaces the linear projector with a nonlinear MLP vision-language connector. This lightweight two-layer MLP with GELU activation enables hierarchical cross-modal feature transformation, enhancing the fusion of visual patches and textual keypoint descriptions. Trained exclusively on COCO data, PoseLLM achieves 77.8 AP on the COCO validation set, outperforming LocLLM by +0.4 AP, while maintaining strong zero-shot generalization on Human-Art and MPII. Our work demonstrates that a simple yet powerful nonlinear connector significantly boosts localization accuracy without sacrificing generalization, advancing the state-of-the-art in language-guided pose estimation. Code is available at https://github.com/Ody-trek/PoseLLM.

Paperid: 773, https://arxiv.org/pdf/2507.09118.pdf GitHub

Abstract:
Continual learning aims to enable models to learn sequentially from continuously incoming data while retaining performance on previously learned tasks. With the Contrastive Language-Image Pre-trained model (CLIP) exhibiting strong capabilities across various downstream tasks, there has been growing interest in leveraging CLIP for continual learning in such scenarios. Most existing works overlook the inherent modality gap in CLIP, a key factor in its generalization and adaptability. In this paper, we analyze the variations in the modality gap during the fine-tuning of vision-language pre-trained models. Our observations reveal that the modality gap effectively reflects the extent to which pre-trained knowledge is preserved. Based on these insights, we propose a simple yet effective method, MG-CLIP, that improves CLIP's performance in class-incremental learning. Our approach leverages modality gap preservation to mitigate forgetting and modality gap compensation to enhance the capacity for new data, introducing a novel modality-gap-based perspective for continual learning. Extensive experiments on multiple benchmarks demonstrate that our method outperforms existing approaches without requiring additional replay data. Our code is available at https://github.com/linlany/MindtheGap.

Paperid: 774, https://arxiv.org/pdf/2507.08982.pdf GitHub

Abstract:
Recent years have witnessed remarkable progress in developing Vision-Language Models (VLMs) capable of processing both textual and visual inputs. These models have demonstrated impressive performance, leading to their widespread adoption in various applications. However, this widespread raises serious concerns regarding user privacy, particularly when models inadvertently process or expose private visual information. In this work, we frame the preservation of privacy in VLMs as an adversarial attack problem. We propose a novel attack strategy that selectively conceals information within designated Region Of Interests (ROIs) in an image, effectively preventing VLMs from accessing sensitive content while preserving the semantic integrity of the remaining image. Unlike conventional adversarial attacks that often disrupt the entire image, our method maintains high coherence in unmasked areas. Experimental results across three state-of-the-art VLMs namely LLaVA, Instruct-BLIP, and BLIP2-T5 demonstrate up to 98% reduction in detecting targeted ROIs, while maintaining global image semantics intact, as confirmed by high similarity scores between clean and adversarial outputs. We believe that this work contributes to a more privacy conscious use of multimodal models and offers a practical tool for further research, with the source code publicly available at: https://github.com/hbrachemi/Vlm_defense-attack.

Paperid: 775, https://arxiv.org/pdf/2507.08979.pdf GitHub

Abstract:
We introduce Projection-based Reduction of Implicit Spurious bias in vision-language Models (PRISM), a new data-free and task-agnostic solution for bias mitigation in VLMs like CLIP. VLMs often inherit and amplify biases in their training data, leading to skewed predictions. PRISM is designed to debias VLMs without relying on predefined bias categories or additional external data. It operates in two stages: first, an LLM is prompted with simple class prompts to generate scene descriptions that contain spurious correlations. Next, PRISM uses our novel contrastive-style debiasing loss to learn a projection that maps the embeddings onto a latent space that minimizes spurious correlations while preserving the alignment between image and text embeddings.Extensive experiments demonstrate that PRISM outperforms current debiasing methods on the commonly used Waterbirds and CelebA datasets We make our code public at: https://github.com/MahdiyarMM/PRISM.

Paperid: 776, https://arxiv.org/pdf/2507.08851.pdf GitHub

Abstract:
Understanding open-world semantics is critical for robotic planning and control, particularly in unstructured outdoor environments. Existing vision-language mapping approaches typically rely on object-centric segmentation priors, which often fail outdoors due to semantic ambiguities and indistinct class boundaries. We propose OTAS - an Open-vocabulary Token Alignment method for outdoor Segmentation. OTAS addresses the limitations of open-vocabulary segmentation models by extracting semantic structure directly from the output tokens of pre-trained vision models. By clustering semantically similar structures across single and multiple views and grounding them in language, OTAS reconstructs a geometrically consistent feature field that supports open-vocabulary segmentation queries. Our method operates in a zero-shot manner, without scene-specific fine-tuning, and achieves real-time performance of up to ~17 fps. On the Off-Road Freespace Detection dataset, OTAS yields a modest IoU improvement over fine-tuned and open-vocabulary 2D segmentation baselines. In 3D segmentation on TartanAir, it achieves up to a 151% relative IoU improvement compared to existing open-vocabulary mapping methods. Real-world reconstructions further demonstrate OTAS' applicability to robotic deployment. Code and a ROS 2 node are available at https://otas-segmentation.github.io/.

Paperid: 777, https://arxiv.org/pdf/2507.08607.pdf GitHub GitHub

Abstract:
Vision-language models (VLMs) such as CLIP achieve strong zero-shot recognition but degrade significantly under \textit{temporally evolving distribution shifts} common in real-world scenarios (e.g., gradual illumination or seasonal changes). Existing continual test-time adaptation (CTTA) methods are typically built around sudden and severe distribution shifts and neglect temporal continuity, leading to three core defects: limited memory cache restricts long-range distribution modeling, causing catastrophic forgetting; entropy-based confidence becomes unreliable under temporal drift, worsening error accumulation; and static visual representations misalign with evolving inputs. We formalize this practical problem as \textit{Continual-Temporal Test-Time Adaptation (CT-TTA)}, where test distributions evolve gradually over time. To address it, we propose \textit{BayesTTA}, a Bayesian adaptation framework that enforces temporally consistent predictions and dynamically aligns visual representations. Specifically, BayesTTA incrementally estimates class-conditional Gaussian mixture distributions without storing raw data, adaptively selects covariance structures through statistical hypothesis testing, and performs calibrated inference using Gaussian discriminant analysis (GDA). These calibrated predictions supervise self-paced adaptation of normalization layers, ensuring efficient and stable representation alignment. We establish a comprehensive CT-TTA benchmark across four temporally evolving datasets and further evaluate generalization on ten standard TTA datasets. Extensive experiments show that BayesTTA consistently outperforms state-of-the-art methods, achieving significant gains while maintaining efficiency. Code is available at \href{https://github.com/cuishuang99/BayesTTA}{https://github.com/cuishuang99/BayesTTA}.

Paperid: 778, https://arxiv.org/pdf/2507.08496.pdf GitHub

Abstract:
While large language models (LLMs) have advanced procedural planning for embodied AI systems through strong reasoning abilities, the integration of multimodal inputs and counterfactual reasoning remains underexplored. To tackle these challenges, we introduce LLaPa, a vision-language model framework designed for multimodal procedural planning. LLaPa generates executable action sequences from textual task descriptions and visual environmental images using vision-language models (VLMs). Furthermore, we enhance LLaPa with two auxiliary modules to improve procedural planning. The first module, the Task-Environment Reranker (TER), leverages task-oriented segmentation to create a task-sensitive feature space, aligning textual descriptions with visual environments and emphasizing critical regions for procedural execution. The second module, the Counterfactual Activities Retriever (CAR), identifies and emphasizes potential counterfactual conditions, enhancing the model's reasoning capability in counterfactual scenarios. Extensive experiments on ActPlan-1K and ALFRED benchmarks demonstrate that LLaPa generates higher-quality plans with superior LCS and correctness, outperforming advanced models. The code and models are available https://github.com/sunshibo1234/LLaPa.

Paperid: 779, https://arxiv.org/pdf/2507.08224.pdf GitHub

Abstract:
Large language models (LLMs) have shown promise in robotic procedural planning, yet their human-centric reasoning often omits the low-level, grounded details needed for robotic execution. Vision-language models (VLMs) offer a path toward more perceptually grounded plans, but current methods either rely on expensive, large-scale models or are constrained to narrow simulation settings. We introduce SelfReVision, a lightweight and scalable self-improvement framework for vision-language procedural planning. SelfReVision enables small VLMs to iteratively critique, revise, and verify their own plans-without external supervision or teacher models-drawing inspiration from chain-of-thought prompting and self-instruct paradigms. Through this self-distillation loop, models generate higher-quality, execution-ready plans that can be used both at inference and for continued fine-tuning. Using models varying from 3B to 72B, our results show that SelfReVision not only boosts performance over weak base VLMs but also outperforms models 100X the size, yielding improved control in downstream embodied tasks.

Paperid: 780, https://arxiv.org/pdf/2507.07966.pdf GitHub

Abstract:
We introduce a full-stack framework that scales up reasoning in vision-language models (VLMs) to long videos, leveraging reinforcement learning. We address the unique challenges of long video reasoning by integrating three critical components: (1) a large-scale dataset, LongVideo-Reason, comprising 104K long video QA pairs with high-quality reasoning annotations across diverse domains such as sports, games, and vlogs; (2) a two-stage training pipeline that extends VLMs with chain-of-thought supervised fine-tuning (CoT-SFT) and reinforcement learning (RL); and (3) a training infrastructure for long video RL, named Multi-modal Reinforcement Sequence Parallelism (MR-SP), which incorporates sequence parallelism and a vLLM-based engine tailored for long video, using cached video embeddings for efficient rollout and prefilling. In our experiments, LongVILA-R1-7B achieves strong performance on video benchmarks, reaching 65.1% and 71.1% accuracy on VideoMME without and with subtitles, respectively, and consistently outperforming LongVILA-7B across multiple benchmarks. Moreover, LongVILA-R1-7B supports processing up to 8,192 video frames per video, and configurable FPS settings. Notably, our MR-SP system achieves up to 2.1x speedup on long video RL training. In addition, we release our training system for public availability that supports RL training on various modalities (video, text, and audio), various models (VILA and Qwen series), and even image and video generation models. On a single A100 node (8 GPUs), it supports RL training on hour-long videos (e.g., 3,600 frames).

Paperid: 781, https://arxiv.org/pdf/2507.07939.pdf GitHub

Abstract:
While Vision-Language Models (VLMs) have shown promising progress in general multimodal tasks, they often struggle in industrial anomaly detection and reasoning, particularly in delivering interpretable explanations and generalizing to unseen categories. This limitation stems from the inherently domain-specific nature of anomaly detection, which hinders the applicability of existing VLMs in industrial scenarios that require precise, structured, and context-aware analysis. To address these challenges, we propose SAGE, a VLM-based framework that enhances anomaly reasoning through Self-Guided Fact Enhancement (SFE) and Entropy-aware Direct Preference Optimization (E-DPO). SFE integrates domain-specific knowledge into visual reasoning via fact extraction and fusion, while E-DPO aligns model outputs with expert preferences using entropy-aware optimization. Additionally, we introduce AD-PL, a preference-optimized dataset tailored for industrial anomaly reasoning, consisting of 28,415 question-answering instances with expert-ranked responses. To evaluate anomaly reasoning models, we develop Multiscale Logical Evaluation (MLE), a quantitative framework analyzing model logic and consistency. SAGE demonstrates superior performance on industrial anomaly datasets under zero-shot and one-shot settings. The code, model and dataset are available at https://github.com/amoreZgx1n/SAGE.

Paperid: 782, https://arxiv.org/pdf/2507.07781.pdf GitHub

Abstract:
The integration of language and 3D perception is critical for embodied AI and robotic systems to perceive, understand, and interact with the physical world. Spatial reasoning, a key capability for understanding spatial relationships between objects, remains underexplored in current 3D vision-language research. Existing datasets often mix semantic cues (e.g., object name) with spatial context, leading models to rely on superficial shortcuts rather than genuinely interpreting spatial relationships. To address this gap, we introduce S\textsc{urprise}3D, a novel dataset designed to evaluate language-guided spatial reasoning segmentation in complex 3D scenes. S\textsc{urprise}3D consists of more than 200k vision language pairs across 900+ detailed indoor scenes from ScanNet++ v2, including more than 2.8k unique object classes. The dataset contains 89k+ human-annotated spatial queries deliberately crafted without object name, thereby mitigating shortcut biases in spatial understanding. These queries comprehensively cover various spatial reasoning skills, such as relative position, narrative perspective, parametric perspective, and absolute distance reasoning. Initial benchmarks demonstrate significant challenges for current state-of-the-art expert 3D visual grounding methods and 3D-LLMs, underscoring the necessity of our dataset and the accompanying 3D Spatial Reasoning Segmentation (3D-SRS) benchmark suite. S\textsc{urprise}3D and 3D-SRS aim to facilitate advancements in spatially aware AI, paving the way for effective embodied interaction and robotic planning. The code and datasets can be found in https://github.com/liziwennba/SUPRISE.

Paperid: 783, https://arxiv.org/pdf/2507.07620.pdf GitHub

Abstract:
Reliable Uncertainty Quantification (UQ) and failure prediction remain open challenges for Vision-Language Models (VLMs). We introduce ViLU, a new Vision-Language Uncertainty quantification framework that contextualizes uncertainty estimates by leveraging all task-relevant textual representations. ViLU constructs an uncertainty-aware multi-modal representation by integrating the visual embedding, the predicted textual embedding, and an image-conditioned textual representation via cross-attention. Unlike traditional UQ methods based on loss prediction, ViLU trains an uncertainty predictor as a binary classifier to distinguish correct from incorrect predictions using a weighted binary cross-entropy loss, making it loss-agnostic. In particular, our proposed approach is well-suited for post-hoc settings, where only vision and text embeddings are available without direct access to the model itself. Extensive experiments on diverse datasets show the significant gains of our method compared to state-of-the-art failure prediction methods. We apply our method to standard classification datasets, such as ImageNet-1k, as well as large-scale image-caption datasets like CC12M and LAION-400M. Ablation studies highlight the critical role of our architecture and training in achieving effective uncertainty quantification. Our code is publicly available and can be found here: https://github.com/ykrmm/ViLU.

Paperid: 784, https://arxiv.org/pdf/2507.07317.pdf GitHub

Abstract:
Recent advances in instruction-guided image editing underscore the need for effective automated evaluation. While Vision-Language Models (VLMs) have been explored as judges, open-source models struggle with alignment, and proprietary models lack transparency and cost efficiency. Additionally, no public training datasets exist to fine-tune open-source VLMs, only small benchmarks with diverse evaluation schemes. To address this, we introduce ADIEE, an automated dataset creation approach which is then used to train a scoring model for instruction-guided image editing evaluation. We generate a large-scale dataset with over 100K samples and use it to fine-tune a LLaVA-NeXT-8B model modified to decode a numeric score from a custom token. The resulting scorer outperforms all open-source VLMs and Gemini-Pro 1.5 across all benchmarks, achieving a 0.0696 (+17.24%) gain in score correlation with human ratings on AURORA-Bench, and improving pair-wise comparison accuracy by 4.03% (+7.21%) on GenAI-Bench and 4.75% (+9.35%) on AURORA-Bench, respectively, compared to the state-of-the-art. The scorer can act as a reward model, enabling automated best edit selection and model fine-tuning. Notably, the proposed scorer can boost MagicBrush model's average evaluation score on ImagenHub from 5.90 to 6.43 (+8.98%). Our code and models are available at https://github.com/SherryXTChen/ADIEE.git.

Paperid: 785, https://arxiv.org/pdf/2507.07254.pdf GitHub

Abstract:
Modern deep learning implementations for medical imaging usually rely on large labeled datasets. These datasets are often difficult to obtain due to privacy concerns, high costs, and even scarcity of cases. In this paper, a label-efficient strategy is proposed for chest X-ray diagnosis that seeks to reflect real-world hospital scenarios. The experiments use the NIH Chest X-ray14 dataset and a pre-trained CLIP ViT-B/32 model. The model is adapted via partial fine-tuning of its visual encoder and then evaluated using zero-shot and few-shot learning with 1-16 labeled examples per disease class. The tests demonstrate that CLIP's pre-trained vision-language features can be effectively adapted to few-shot medical imaging tasks, achieving over 20\% improvement in mean AUC score as compared to the zero-shot baseline. The key aspect of this work is to attempt to simulate internal hospital workflows, where image archives exist but annotations are sparse. This work evaluates a practical and scalable solution for both common and rare disease diagnosis. Additionally this research is intended for academic and experimental purposes only and has not been peer reviewed yet. All code is found at https://github.com/heet007-code/CLIP-disease-xray.

Paperid: 786, https://arxiv.org/pdf/2507.07135.pdf GitHub

Abstract:
The composed image retrieval (CIR) task is to retrieve target images given a reference image and a modification text. Recent methods for CIR leverage large pretrained vision-language models (VLMs) and achieve good performance on general-domain concepts like color and texture. However, they still struggle with application domains like fashion, because the rich and diverse vocabulary used in fashion requires specific fine-grained vision and language understanding. An additional difficulty is the lack of large-scale fashion datasets with detailed and relevant annotations, due to the expensive cost of manual annotation by specialists. To address these challenges, we introduce FACap, a large-scale, automatically constructed fashion-domain CIR dataset. It leverages web-sourced fashion images and a two-stage annotation pipeline powered by a VLM and a large language model (LLM) to generate accurate and detailed modification texts. Then, we propose a new CIR model FashionBLIP-2, which fine-tunes the general-domain BLIP-2 model on FACap with lightweight adapters and multi-head query-candidate matching to better account for fine-grained fashion-specific information. FashionBLIP-2 is evaluated with and without additional fine-tuning on the Fashion IQ benchmark and the enhanced evaluation dataset enhFashionIQ, leveraging our pipeline to obtain higher-quality annotations. Experimental results show that the combination of FashionBLIP-2 and pretraining with FACap significantly improves the model's performance in fashion CIR especially for retrieval with fine-grained modification texts, demonstrating the value of our dataset and approach in a highly demanding environment such as e-commerce websites. Code is available at https://fgxaos.github.io/facap-paper-website/.

Paperid: 787, https://arxiv.org/pdf/2507.07125.pdf GitHub

Abstract:
Unsupervised domain adaptation (UDA) involves learning class semantics from labeled data within a source domain that generalize to an unseen target domain. UDA methods are particularly impactful for semantic segmentation, where annotations are more difficult to collect than in image classification. Despite recent advances in large-scale vision-language representation learning, UDA methods for segmentation have not taken advantage of the domain-agnostic properties of text. To address this, we present a novel Covariance-based Pixel-Text loss, CoPT, that uses domain-agnostic text embeddings to learn domain-invariant features in an image segmentation encoder. The text embeddings are generated through our LLM Domain Template process, where an LLM is used to generate source and target domain descriptions that are fed to a frozen CLIP model and combined. In experiments on four benchmarks we show that a model trained using CoPT achieves the new state of the art performance on UDA for segmentation. The code can be found at https://github.com/cfmata/CoPT.

Paperid: 788, https://arxiv.org/pdf/2507.07104.pdf GitHub

Abstract:
Building state-of-the-art Vision-Language Models (VLMs) with strong captioning capabilities typically necessitates training on billions of high-quality image-text pairs, requiring millions of GPU hours. This paper introduces the Vision-Language-Vision (VLV) auto-encoder framework, which strategically leverages key pretrained components: a vision encoder, the decoder of a Text-to-Image (T2I) diffusion model, and subsequently, a Large Language Model (LLM). Specifically, we establish an information bottleneck by regularizing the language representation space, achieved through freezing the pretrained T2I diffusion decoder. Our VLV pipeline effectively distills knowledge from the text-conditioned diffusion model using continuous embeddings, demonstrating comprehensive semantic understanding via high-quality reconstructions. Furthermore, by fine-tuning a pretrained LLM to decode the intermediate language representations into detailed descriptions, we construct a state-of-the-art (SoTA) captioner comparable to leading models like GPT-4o and Gemini 2.0 Flash. Our method demonstrates exceptional cost-efficiency and significantly reduces data requirements; by primarily utilizing single-modal images for training and maximizing the utility of existing pretrained models (image encoder, T2I diffusion model, and LLM), it circumvents the need for massive paired image-text datasets, keeping the total training expenditure under $1,000 USD.

Paperid: 789, https://arxiv.org/pdf/2507.07079.pdf GitHub

Abstract:
Despite the rapid advances in Text-to-Image (T2I) generation models, their evaluation remains challenging in domains like fashion, involving complex compositional generation. Recent automated T2I evaluation methods leverage pre-trained vision-language models to measure cross-modal alignment. However, our preliminary study reveals that they are still limited in assessing rich entity-attribute semantics, facing challenges in attribute confusion, i.e., when attributes are correctly depicted but associated to the wrong entities. To address this, we build on a Visual Question Answering (VQA) localization strategy targeting one single entity at a time across both visual and textual modalities. We propose a localized human evaluation protocol and introduce a novel automatic metric, Localized VQAScore (L-VQAScore), that combines visual localization with VQA probing both correct (reflection) and miss-localized (leakage) attribute generation. On a newly curated dataset featuring challenging compositional alignment scenarios, L-VQAScore outperforms state-of-the-art T2I evaluation methods in terms of correlation with human judgments, demonstrating its strength in capturing fine-grained entity-attribute associations. We believe L-VQAScore can be a reliable and scalable alternative to subjective evaluations.

Paperid: 790, https://arxiv.org/pdf/2507.06761.pdf GitHub

Abstract:
Manchu, a critically endangered language essential for understanding early modern Eastern Eurasian history, lacks effective OCR systems that can handle real-world historical documents. This study develops high-performing OCR systems by fine-tuning three open-source vision-language models (LLaMA-3.2-11B, Qwen2.5-VL-7B, Qwen2.5-VL-3B) on 60,000 synthetic Manchu word images using parameter-efficient training. LLaMA-3.2-11B achieved exceptional performance with 98.3\% word accuracy and 0.0024 character error rate on synthetic data, while crucially maintaining 93.1\% accuracy on real-world handwritten documents. Comparative evaluation reveals substantial advantages over traditional approaches: while a CRNN baseline achieved 99.8\% synthetic accuracy, it suffered severe degradation to 72.5\% on real documents. Our approach demonstrates effective synthetic-to-real domain transfer, providing a cost-effective solution deployable on accessible infrastructure. This work establishes a transferable framework for endangered language OCR that removes technical and financial barriers in digital humanities, enabling historians and linguists to process historical archives without specialized computing resources. Code and model weights are available at https://github.com/mic7ch1/ManchuAI-OCR.

Paperid: 791, https://arxiv.org/pdf/2507.06679.pdf GitHub

Abstract:
Recent advances in large vision-language models (VLMs) have shown remarkable progress in solving the text-promptable object counting problem. Representative methods typically specify text prompts with object category information in images. This however is insufficient for training the model to accurately distinguish the number of objects in the counting task. To this end, we propose QUANet, which introduces novel quantity-oriented text prompts with a vision-text quantity alignment loss to enhance the model's quantity awareness. Moreover, we propose a dual-stream adaptive counting decoder consisting of a Transformer stream, a CNN stream, and a number of Transformer-to-CNN enhancement adapters (T2C-adapters) for density map prediction. The T2C-adapters facilitate the effective knowledge communication and aggregation between the Transformer and CNN streams. A cross-stream quantity ranking loss is proposed in the end to optimize the ranking orders of predictions from the two streams. Extensive experiments on standard benchmarks such as FSC-147, CARPK, PUCPR+, and ShanghaiTech demonstrate our model's strong generalizability for zero-shot class-agnostic counting. Code is available at https://github.com/viscom-tongji/QUANet

Paperid: 792, https://arxiv.org/pdf/2507.06603.pdf GitHub

Abstract:
Long-term action recognition (LTAR) is challenging due to extended temporal spans with complex atomic action correlations and visual confounders. Although vision-language models (VLMs) have shown promise, they often rely on statistical correlations instead of causal mechanisms. Moreover, existing causality-based methods address modal-specific biases but lack cross-modal causal modeling, limiting their utility in VLM-based LTAR. This paper proposes \textbf{C}ross-\textbf{M}odal \textbf{D}ual-\textbf{C}ausal \textbf{L}earning (CMDCL), which introduces a structural causal model to uncover causal relationships between videos and label texts. CMDCL addresses cross-modal biases in text embeddings via textual causal intervention and removes confounders inherent in the visual modality through visual causal intervention guided by the debiased text. These dual-causal interventions enable robust action representations to address LTAR challenges. Experimental results on three benchmarks including Charades, Breakfast and COIN, demonstrate the effectiveness of the proposed model. Our code is available at https://github.com/xushaowu/CMDCL.

Paperid: 793, https://arxiv.org/pdf/2507.06231.pdf GitHub

Abstract:
Referring Remote Sensing Image Segmentation provides a flexible and fine-grained framework for remote sensing scene analysis via vision-language collaborative interpretation. Current approaches predominantly utilize a three-stage pipeline encompassing dual-modal encoding, cross-modal interaction, and pixel decoding. These methods demonstrate significant limitations in managing complex semantic relationships and achieving precise cross-modal alignment, largely due to their coupled processing mechanism that conflates target localization with boundary delineation. This architectural coupling amplifies error propagation under semantic ambiguity while restricting model generalizability and interpretability. To address these issues, we propose RSRefSeg 2, a decoupling paradigm that reformulates the conventional workflow into a collaborative dual-stage framework: coarse localization followed by fine segmentation. RSRefSeg 2 integrates CLIP's cross-modal alignment strength with SAM's segmentation generalizability through strategic foundation model collaboration. Specifically, CLIP is employed as the dual-modal encoder to activate target features within its pre-aligned semantic space and generate localization prompts. To mitigate CLIP's misactivation challenges in multi-entity scenarios described by referring texts, a cascaded second-order prompter is devised, which enhances precision through implicit reasoning via decomposition of text embeddings into complementary semantic subspaces. These optimized semantic prompts subsequently direct the SAM to generate pixel-level refined masks, thereby completing the semantic transmission pipeline. Extensive experiments (RefSegRS, RRSIS-D, and RISBench) demonstrate that RSRefSeg 2 surpasses contemporary methods in segmentation accuracy (+~3% gIoU) and complex semantic interpretation. Code is available at: https://github.com/KyanChen/RSRefSeg2.

Paperid: 794, https://arxiv.org/pdf/2507.06140.pdf GitHub

Abstract:
Low-dose computed tomography (LDCT) reduces radiation exposure but often degrades image quality, potentially compromising diagnostic accuracy. Existing deep learning-based denoising methods focus primarily on pixel-level mappings, overlooking the potential benefits of high-level semantic guidance. Recent advances in vision-language models (VLMs) suggest that language can serve as a powerful tool for capturing structured semantic information, offering new opportunities to improve LDCT reconstruction. In this paper, we introduce LangMamba, a Language-driven Mamba framework for LDCT denoising that leverages VLM-derived representations to enhance supervision from normal-dose CT (NDCT). LangMamba follows a two-stage learning strategy. First, we pre-train a Language-guided AutoEncoder (LangAE) that leverages frozen VLMs to map NDCT images into a semantic space enriched with anatomical information. Second, we synergize LangAE with two key components to guide LDCT denoising: Semantic-Enhanced Efficient Denoiser (SEED), which enhances NDCT-relevant local semantic while capturing global features with efficient Mamba mechanism, and Language-engaged Dual-space Alignment (LangDA) Loss, which ensures that denoised images align with NDCT in both perceptual and semantic spaces. Extensive experiments on two public datasets demonstrate that LangMamba outperforms conventional state-of-the-art methods, significantly improving detail preservation and visual fidelity. Remarkably, LangAE exhibits strong generalizability to unseen datasets, thereby reducing training costs. Furthermore, LangDA loss improves explainability by integrating language-guided insights into image reconstruction and offers a plug-and-play fashion. Our findings shed new light on the potential of language as a supervisory signal to advance LDCT denoising. The code is publicly available on https://github.com/hao1635/LangMamba.

Paperid: 795, https://arxiv.org/pdf/2507.06072.pdf GitHub

Abstract:
Accurate driving behavior recognition and reasoning are critical for autonomous driving video understanding. However, existing methods often tend to dig out the shallow causal, fail to address spurious correlations across modalities, and ignore the ego-vehicle level causality modeling. To overcome these limitations, we propose a novel Multimodal Causal Analysis Model (MCAM) that constructs latent causal structures between visual and language modalities. Firstly, we design a multi-level feature extractor to capture long-range dependencies. Secondly, we design a causal analysis module that dynamically models driving scenarios using a directed acyclic graph (DAG) of driving states. Thirdly, we utilize a vision-language transformer to align critical visual features with their corresponding linguistic expressions. Extensive experiments on the BDD-X, and CoVLA datasets demonstrate that MCAM achieves SOTA performance in visual-language causal relationship learning. Furthermore, the model exhibits superior capability in capturing causal characteristics within video sequences, showcasing its effectiveness for autonomous driving applications. The code is available at https://github.com/SixCorePeach/MCAM.

Paperid: 796, https://arxiv.org/pdf/2507.05394.pdf GitHub

Abstract:
Vision-Language Models (VLMs) like CLIP have demonstrated remarkable generalization in zero- and few-shot settings, but adapting them efficiently to decentralized, heterogeneous data remains a challenge. While prompt tuning has emerged as a popular parameter-efficient approach in personalized federated learning, existing methods often sacrifice generalization in favor of personalization, struggling particularly on unseen classes or domains. In this work, we propose pFedMMA, the first personalized federated learning framework that leverages multi-modal adapters for vision-language tasks. Each adapter contains modality-specific up- and down-projection layers alongside a globally shared projection that aligns cross-modal features. Our asymmetric optimization strategy allows clients to locally adapt to personalized data distributions while collaboratively training the shared projection to improve global generalization. This design is also communication-efficient, as only the shared component is exchanged during rounds. Through extensive experiments across eleven datasets, including domain- and label-shift scenarios, we show that pFedMMA achieves state-of-the-art trade-offs between personalization and generalization, outperforming recent federated prompt tuning methods. The code is available at https://github.com/sajjad-ucsb/pFedMMA.

Paperid: 797, https://arxiv.org/pdf/2507.05240.pdf GitHub GitHub

Abstract:
Vision-and-Language Navigation (VLN) in real-world settings requires agents to process continuous visual streams and generate actions with low latency grounded in language instructions. While Video-based Large Language Models (Video-LLMs) have driven recent progress, current VLN methods based on Video-LLM often face trade-offs among fine-grained visual understanding, long-term context modeling and computational efficiency. We introduce StreamVLN, a streaming VLN framework that employs a hybrid slow-fast context modeling strategy to support multi-modal reasoning over interleaved vision, language and action inputs. The fast-streaming dialogue context facilitates responsive action generation through a sliding-window of active dialogues, while the slow-updating memory context compresses historical visual states using a 3D-aware token pruning strategy. With this slow-fast design, StreamVLN achieves coherent multi-turn dialogue through efficient KV cache reuse, supporting long video streams with bounded context size and inference cost. Experiments on VLN-CE benchmarks demonstrate state-of-the-art performance with stable low latency, ensuring robustness and efficiency in real-world deployment. The project page is: \href{https://streamvln.github.io/}{https://streamvln.github.io/}.

Paperid: 798, https://arxiv.org/pdf/2507.05211.pdf GitHub

Abstract:
Unified segmentation of 3D point clouds is crucial for scene understanding, but is hindered by its sparse structure, limited annotations, and the challenge of distinguishing fine-grained object classes in complex environments. Existing methods often struggle to capture rich semantic and contextual information due to limited supervision and a lack of diverse multimodal cues, leading to suboptimal differentiation of classes and instances. To address these challenges, we propose VDG-Uni3DSeg, a novel framework that integrates pre-trained vision-language models (e.g., CLIP) and large language models (LLMs) to enhance 3D segmentation. By leveraging LLM-generated textual descriptions and reference images from the internet, our method incorporates rich multimodal cues, facilitating fine-grained class and instance separation. We further design a Semantic-Visual Contrastive Loss to align point features with multimodal queries and a Spatial Enhanced Module to model scene-wide relationships efficiently. Operating within a closed-set paradigm that utilizes multimodal knowledge generated offline, VDG-Uni3DSeg achieves state-of-the-art results in semantic, instance, and panoptic segmentation, offering a scalable and practical solution for 3D understanding. Our code is available at https://github.com/Hanzy1996/VDG-Uni3DSeg.

Paperid: 799, https://arxiv.org/pdf/2507.05165.pdf GitHub

Abstract:
Social networks can be a valuable source of information during crisis events. In particular, users can post a stream of multimodal data that can be critical for real-time humanitarian response. However, effectively extracting meaningful information from this large and noisy data stream and effectively integrating heterogeneous data remains a formidable challenge. In this work, we explore vision language models (VLMs) and advanced fusion strategies to enhance the classification of crisis data in three different tasks. We incorporate LLaVA-generated text to improve text-image alignment. Additionally, we leverage Contrastive Language-Image Pretraining (CLIP)-based vision and text embeddings, which, without task-specific fine-tuning, outperform traditional models. To further refine multimodal fusion, we employ Guided Cross Attention (Guided CA) and combine it with the Differential Attention mechanism to enhance feature alignment by emphasizing critical information while filtering out irrelevant content. Our results show that while Differential Attention improves classification performance, Guided CA remains highly effective in aligning multimodal features. Extensive experiments on the CrisisMMD benchmark data set demonstrate that the combination of pretrained VLMs, enriched textual descriptions, and adaptive fusion strategies consistently outperforms state-of-the-art models in classification accuracy, contributing to more reliable and interpretable models for three different tasks that are crucial for disaster response. Our code is available at https://github.com/Munia03/Multimodal_Crisis_Event.

Paperid: 800, https://arxiv.org/pdf/2507.05116.pdf GitHub

Abstract:
Recent large-scale Vision Language Action (VLA) models have shown superior performance in robotic manipulation tasks guided by natural language. However, current VLA models suffer from two drawbacks: (i) generation of massive tokens leading to high inference latency and increased training cost, and (ii) insufficient utilization of generated actions resulting in potential performance loss. To address these issues, we develop a training framework to finetune VLA models for generating significantly fewer action tokens with high parallelism, effectively reducing inference latency and training cost. Furthermore, we introduce an inference optimization technique with a novel voting-based ensemble strategy to combine current and previous action predictions, improving the utilization of generated actions and overall performance. Our results demonstrate that we achieve superior performance compared with state-of-the-art VLA models, achieving significantly higher success rates and 39$\times$ faster inference than OpenVLA with 46 Hz throughput on edge platforms, demonstrating practical deployability. The code is available at https://github.com/LukeLIN-web/VOTE.

Paperid: 801, https://arxiv.org/pdf/2507.05108.pdf GitHub

Abstract:
Historical documents represent an invaluable cultural heritage, yet have undergone significant degradation over time through tears, water erosion, and oxidation. Existing Historical Document Restoration (HDR) methods primarily focus on single modality or limited-size restoration, failing to meet practical needs. To fill this gap, we present a full-page HDR dataset (FPHDR) and a novel automated HDR solution (AutoHDR). Specifically, FPHDR comprises 1,633 real and 6,543 synthetic images with character-level and line-level locations, as well as character annotations in different damage grades. AutoHDR mimics historians' restoration workflows through a three-stage approach: OCR-assisted damage localization, vision-language context text prediction, and patch autoregressive appearance restoration. The modular architecture of AutoHDR enables seamless human-machine collaboration, allowing for flexible intervention and optimization at each restoration stage. Experiments demonstrate AutoHDR's remarkable performance in HDR. When processing severely damaged documents, our method improves OCR accuracy from 46.83% to 84.05%, with further enhancement to 94.25% through human-machine collaboration. We believe this work represents a significant advancement in automated historical document restoration and contributes substantially to cultural heritage preservation. The model and dataset are available at https://github.com/SCUT-DLVCLab/AutoHDR.

Paperid: 802, https://arxiv.org/pdf/2507.05020.pdf GitHub

Abstract:
Surgical AI often involves multiple tasks within a single procedure, like phase recognition or assessing the Critical View of Safety in laparoscopic cholecystectomy. Traditional models, built for one task at a time, lack flexibility, requiring a separate model for each. To address this, we introduce MML-SurgAdapt, a unified multi-task framework with Vision-Language Models (VLMs), specifically CLIP, to handle diverse surgical tasks through natural language supervision. A key challenge in multi-task learning is the presence of partial annotations when integrating different tasks. To overcome this, we employ Single Positive Multi-Label (SPML) learning, which traditionally reduces annotation burden by training models with only one positive label per instance. Our framework extends this approach to integrate data from multiple surgical tasks within a single procedure, enabling effective learning despite incomplete or noisy annotations. We demonstrate the effectiveness of our model on a combined dataset consisting of Cholec80, Endoscapes2023, and CholecT50, utilizing custom prompts. Extensive evaluation shows that MML-SurgAdapt performs comparably to task-specific benchmarks, with the added advantage of handling noisy annotations. It also outperforms the existing SPML frameworks for the task. By reducing the required labels by 23%, our approach proposes a more scalable and efficient labeling process, significantly easing the annotation burden on clinicians. To our knowledge, this is the first application of SPML to integrate data from multiple surgical tasks, presenting a novel and generalizable solution for multi-task learning in surgical computer vision. Implementation is available at: https://github.com/CAMMA-public/MML-SurgAdapt

Paperid: 803, https://arxiv.org/pdf/2507.04789.pdf GitHub

Abstract:
Recent advances in vision-language models (VLMs) have significantly improved performance in embodied tasks such as goal decomposition and visual comprehension. However, providing accurate rewards for robotic manipulation without fine-tuning VLMs remains challenging due to the absence of domain-specific robotic knowledge in pre-trained datasets and high computational costs that hinder real-time applicability. To address this, we propose $\mathrm{T}^2$-VLM, a novel training-free, temporally consistent framework that generates accurate rewards through tracking the status changes in VLM-derived subgoals. Specifically, our method first queries the VLM to establish spatially aware subgoals and an initial completion estimate before each round of interaction. We then employ a Bayesian tracking algorithm to update the goal completion status dynamically, using subgoal hidden states to generate structured rewards for reinforcement learning (RL) agents. This approach enhances long-horizon decision-making and improves failure recovery capabilities with RL. Extensive experiments indicate that $\mathrm{T}^2$-VLM achieves state-of-the-art performance in two robot manipulation benchmarks, demonstrating superior reward accuracy with reduced computation consumption. We believe our approach not only advances reward generation techniques but also contributes to the broader field of embodied AI. Project website: https://t2-vlm.github.io/.

Paperid: 804, https://arxiv.org/pdf/2507.04511.pdf GitHub

Abstract:
Pre-trained vision-language models (VLMs) have advanced out-of-distribution (OOD) detection recently. However, existing CLIP-based methods often focus on learning OOD-related knowledge to improve OOD detection, showing limited generalization or reliance on external large-scale auxiliary datasets. In this study, instead of delving into the intricate OOD-related knowledge, we propose an innovative CLIP-based framework based on Forced prompt leArning (FA), designed to make full use of the In-Distribution (ID) knowledge and ultimately boost the effectiveness of OOD detection. Our key insight is to learn a prompt (i.e., forced prompt) that contains more diversified and richer descriptions of the ID classes beyond the textual semantics of class labels. Specifically, it promotes better discernment for ID images, by forcing more notable semantic similarity between ID images and the learnable forced prompt. Moreover, we introduce a forced coefficient, encouraging the forced prompt to learn more comprehensive and nuanced descriptions of the ID classes. In this way, FA is capable of achieving notable improvements in OOD detection, even when trained without any external auxiliary datasets, while maintaining an identical number of trainable parameters as CoOp. Extensive empirical evaluations confirm our method consistently outperforms current state-of-the-art methods. Code is available at https://github.com/0xFAFA/FA.

Paperid: 805, https://arxiv.org/pdf/2507.04036.pdf GitHub

Abstract:
We present PresentAgent, a multimodal agent that transforms long-form documents into narrated presentation videos. While existing approaches are limited to generating static slides or text summaries, our method advances beyond these limitations by producing fully synchronized visual and spoken content that closely mimics human-style presentations. To achieve this integration, PresentAgent employs a modular pipeline that systematically segments the input document, plans and renders slide-style visual frames, generates contextual spoken narration with large language models and Text-to-Speech models, and seamlessly composes the final video with precise audio-visual alignment. Given the complexity of evaluating such multimodal outputs, we introduce PresentEval, a unified assessment framework powered by Vision-Language Models that comprehensively scores videos across three critical dimensions: content fidelity, visual clarity, and audience comprehension through prompt-based evaluation. Our experimental validation on a curated dataset of 30 document-presentation pairs demonstrates that PresentAgent approaches human-level quality across all evaluation metrics. These results highlight the significant potential of controllable multimodal agents in transforming static textual materials into dynamic, effective, and accessible presentation formats. Code will be available at https://github.com/AIGeeksGroup/PresentAgent.

Paperid: 806, https://arxiv.org/pdf/2507.03916.pdf GitHub

Abstract:
Slide animations, such as fade-in, fly-in, and wipe, are critical for audience engagement, efficient information delivery, and vivid visual expression. However, most AI-driven slide-generation tools still lack native animation support, and existing vision-language models (VLMs) struggle with animation tasks due to the absence of public datasets and limited temporal-reasoning capabilities. To address this gap, we release the first public dataset for slide-animation modeling: 12,000 triplets of natural-language descriptions, animation JSON files, and rendered videos, collectively covering every built-in PowerPoint effect. Using this resource, we fine-tune Qwen-2.5-VL-7B with Low-Rank Adaptation (LoRA) and achieve consistent improvements over GPT-4.1 and Gemini-2.5-Pro in BLEU-4, ROUGE-L, SPICE, and our Coverage-Order-Detail Assessment (CODA) metric, which evaluates action coverage, temporal order, and detail fidelity. On a manually created test set of slides, the LoRA model increases BLEU-4 by around 60%, ROUGE-L by 30%, and shows significant improvements in CODA-detail. This demonstrates that low-rank adaptation enables reliable temporal reasoning and generalization beyond synthetic data. Overall, our dataset, LoRA-enhanced model, and CODA metric provide a rigorous benchmark and foundation for future research on VLM-based dynamic slide generation.

Paperid: 807, https://arxiv.org/pdf/2507.03283.pdf GitHub GitHub

Abstract:
Molecular property prediction is a fundamental task in computational chemistry with critical applications in drug discovery and materials science. While recent works have explored Large Language Models (LLMs) for this task, they primarily rely on textual molecular representations such as SMILES/SELFIES, which can be ambiguous and structurally less informative. In this work, we introduce MolVision, a novel approach that leverages Vision-Language Models (VLMs) by integrating both molecular structure as images and textual descriptions to enhance property prediction. We construct a benchmark spanning ten diverse datasets, covering classification, regression and description tasks. Evaluating nine different VLMs in zero-shot, few-shot, and fine-tuned settings, we find that visual information improves prediction performance, particularly when combined with efficient fine-tuning strategies such as LoRA. Our results reveal that while visual information alone is insufficient, multimodal fusion significantly enhances generalization across molecular properties. Adaptation of vision encoder for molecular images in conjunction with LoRA further improves the performance. The code and data is available at : $\href{https://molvision.github.io/MolVision/}{https://molvision.github.io/MolVision/}$.

Paperid: 808, https://arxiv.org/pdf/2507.03123.pdf GitHub

Abstract:
Hallucination is a long-standing problem that has been actively investigated in Vision-Language Models (VLMs). Existing research commonly attributes hallucinations to technical limitations or sycophancy bias, where the latter means the models tend to generate incorrect answers to align with user expectations. However, these explanations primarily focus on technical or externally driven factors, may have neglected the possibility that hallucination behaviours might mirror cognitive biases observed in human psychology. In this work, we introduce a psychological taxonomy, categorizing VLMs' hallucination behaviours, including sycophancy, logical inconsistency, and a newly identified VLMs behaviour: authority bias. To systematically analyze these behaviours, we design AIpsych, a scalable benchmark that reveals psychological tendencies in model response patterns. Leveraging this benchmark, we investigate how variations in model architecture and parameter size influence model behaviour when responding to strategically manipulated questions. Our experiments reveal that as model size increases, VLMs exhibit stronger sycophantic tendencies but reduced authority bias, suggesting increasing competence but a potential erosion of response integrity. A human subject study further validates our hypotheses and highlights key behavioural differences between VLMs and human respondents. This work suggests a new perspective for understanding hallucination in VLMs and highlights the importance of integrating psychological principles into model evaluation.The benchmark is available at https://github.com/lxrswdd/AIpsych.

Paperid: 809, https://arxiv.org/pdf/2507.02994.pdf GitHub

Abstract:
Medical Image Grounding (MIG), which involves localizing specific regions in medical images based on textual descriptions, requires models to not only perceive regions but also deduce spatial relationships of these regions. Existing Vision-Language Models (VLMs) for MIG often rely on Supervised Fine-Tuning (SFT) with large amounts of Chain-of-Thought (CoT) reasoning annotations, which are expensive and time-consuming to acquire. Recently, DeepSeek-R1 demonstrated that Large Language Models (LLMs) can acquire reasoning abilities through Group Relative Policy Optimization (GRPO) without requiring CoT annotations. In this paper, we adapt the GRPO reinforcement learning framework to VLMs for Medical Image Grounding. We propose the Spatial-Semantic Rewarded Group Relative Policy Optimization to train the model without CoT reasoning annotations. Specifically, we introduce Spatial-Semantic Rewards, which combine spatial accuracy reward and semantic consistency reward to provide nuanced feedback for both spatially positive and negative completions. Additionally, we propose to use the Chain-of-Box template, which integrates visual information of referring bounding boxes into the reasoning process, enabling the model to explicitly reason about spatial regions during intermediate steps. Experiments on three datasets MS-CXR, ChestX-ray8, and M3D-RefSeg demonstrate that our method achieves state-of-the-art performance in Medical Image Grounding. Ablation studies further validate the effectiveness of each component in our approach. Code, checkpoints, and datasets are available at https://github.com/bio-mlhui/MedGround-R1

Paperid: 810, https://arxiv.org/pdf/2507.02948.pdf GitHub

Abstract:
Autonomous driving has seen significant progress, driven by extensive real-world data. However, in long-tail scenarios, accurately predicting the safety of the ego vehicle's future motion remains a major challenge due to uncertainties in dynamic environments and limitations in data coverage. In this work, we aim to explore whether it is possible to enhance the motion risk prediction capabilities of Vision-Language Models (VLM) by synthesizing high-risk motion data. Specifically, we introduce a Bird's-Eye View (BEV) based motion simulation method to model risks from three aspects: the ego-vehicle, other vehicles, and the environment. This allows us to synthesize plug-and-play, high-risk motion data suitable for VLM training, which we call DriveMRP-10K. Furthermore, we design a VLM-agnostic motion risk estimation framework, named DriveMRP-Agent. This framework incorporates a novel information injection strategy for global context, ego-vehicle perspective, and trajectory projection, enabling VLMs to effectively reason about the spatial relationships between motion waypoints and the environment. Extensive experiments demonstrate that by fine-tuning with DriveMRP-10K, our DriveMRP-Agent framework can significantly improve the motion risk prediction performance of multiple VLM baselines, with the accident recognition accuracy soaring from 27.13% to 88.03%. Moreover, when tested via zero-shot evaluation on an in-house real-world high-risk motion dataset, DriveMRP-Agent achieves a significant performance leap, boosting the accuracy from base_model's 29.42% to 68.50%, which showcases the strong generalization capabilities of our method in real-world scenarios.

Paperid: 811, https://arxiv.org/pdf/2507.02909.pdf GitHub

Abstract:
Prior Vision Language Model (VLM) token pruning reduces computation by eliminating attention and feed-forward operations for pruned tokens while maintaining all operations for critical tokens. However, this binary approach conflates token/operation redundancy - critical operations may be removed along with discarded tokens, while preserved tokens retain all potentially redundant operations. To surgically eliminate redundant operations while preserving critical ones, we propose Greedily Sorted Operation Pruning (GSOP), a data-driven method that directly prunes operations rather than tokens. GSOP first decomposes a VLM decoder's computations into atomic operations along three dimensions: token groups, layer positions, and computation modules. GSOP determines the pruning order of operations through greedy sorting: GSOP iteratively selects the redundant operation that incurs minimal performance drop considering previously pruned operations. Different computational budgets can be accommodated without re-searching by simply pruning operations according to this order until the desired budget is met. GSOP enhances sorting efficiency through: a) leveraging historical operation rankings to avoid redundant evaluations; b) excluding the ``free-to-prune" and ``danger-to-prune" operations from sorting. GSOP achieves compelling efficiency-performance tradeoffs, reducing computation by 70% with only 4% performance loss while maintaining up to 18% higher performance than state-of-the-art methods when transferred across diverse VLMs and tasks. Real GPU efficiency evaluations confirm its practical value. The code is in https://github.com/zxcvfd13502/GSOP.

Paperid: 812, https://arxiv.org/pdf/2507.02844.pdf GitHub

Abstract:
With the emergence of strong vision language capabilities, multimodal large language models (MLLMs) have demonstrated tremendous potential for real-world applications. However, the security vulnerabilities exhibited by the visual modality pose significant challenges to deploying such models in open-world environments. Recent studies have successfully induced harmful responses from target MLLMs by encoding harmful textual semantics directly into visual inputs. However, in these approaches, the visual modality primarily serves as a trigger for unsafe behavior, often exhibiting semantic ambiguity and lacking grounding in realistic scenarios. In this work, we define a novel setting: vision-centric jailbreak, where visual information serves as a necessary component in constructing a complete and realistic jailbreak context. Building on this setting, we propose the VisCo (Visual Contextual) Attack. VisCo fabricates contextual dialogue using four distinct vision-focused strategies, dynamically generating auxiliary images when necessary to construct a vision-centric jailbreak scenario. To maximize attack effectiveness, it incorporates automatic toxicity obfuscation and semantic refinement to produce a final attack prompt that reliably triggers harmful responses from the target black-box MLLMs. Specifically, VisCo achieves a toxicity score of 4.78 and an Attack Success Rate (ASR) of 85% on MM-SafetyBench against GPT-4o, significantly outperforming the baseline, which achieves a toxicity score of 2.48 and an ASR of 22.2%. Code: https://github.com/Dtc7w3PQ/Visco-Attack.

Paperid: 813, https://arxiv.org/pdf/2507.01949.pdf GitHub

Authors:Kwai Keye Team, Biao Yang, Bin Wen, Changyi Liu, Chenglong Chu, Chengru Song, Chongling Rao, Chuan Yi, Da Li, Dunju Zang, Fan Yang, Guorui Zhou, Hao Peng, Haojie Ding, Jiaming Huang, Jiangxia Cao, Jiankang Chen, Jingyun Hua, Jin Ouyang, Kaibing Chen, Kaiyu Jiang, Kaiyu Tang, Kun Gai, Shengnan Zhang, Siyang Mao, Sui Huang, Tianke Zhang, Tingting Gao, Wei Chen, Wei Yuan, Xiangyu Wu, Xiao Hu, Xingyu Lu, Yang Zhou, Yi-Fan Zhang, Yiping Yang, Yulong Chen, Zhenhua Wu, Zhenyu Li, Zhixin Ling, Ziming Li, Dehua Ma, Di Xu, Haixuan Gao, Hang Li, Jiawei Guo, Jing Wang, Lejian Ren, Muhao Wei, Qianqian Wang, Qigen Hu, Shiyao Wang, Tao Yu, Xinchen Luo, Yan Li, Yiming Liang, Yuhang Hu, Zeyi Lu, Zhuoran Yang, Zixing Zhang

Abstract:
While Multimodal Large Language Models (MLLMs) demonstrate remarkable capabilities on static images, they often fall short in comprehending dynamic, information-dense short-form videos, a dominant medium in today's digital landscape. To bridge this gap, we introduce \textbf{Kwai Keye-VL}, an 8-billion-parameter multimodal foundation model engineered for leading-edge performance in short-video understanding while maintaining robust general-purpose vision-language abilities. The development of Keye-VL rests on two core pillars: a massive, high-quality dataset exceeding 600 billion tokens with a strong emphasis on video, and an innovative training recipe. This recipe features a four-stage pre-training process for solid vision-language alignment, followed by a meticulous two-phase post-training process. The first post-training stage enhances foundational capabilities like instruction following, while the second phase focuses on stimulating advanced reasoning. In this second phase, a key innovation is our five-mode ``cold-start'' data mixture, which includes ``thinking'', ``non-thinking'', ``auto-think'', ``think with image'', and high-quality video data. This mixture teaches the model to decide when and how to reason. Subsequent reinforcement learning (RL) and alignment steps further enhance these reasoning capabilities and correct abnormal model behaviors, such as repetitive outputs. To validate our approach, we conduct extensive evaluations, showing that Keye-VL achieves state-of-the-art results on public video benchmarks and remains highly competitive on general image-based tasks (Figure 1). Furthermore, we develop and release the \textbf{KC-MMBench}, a new benchmark tailored for real-world short-video scenarios, where Keye-VL shows a significant advantage.

Paperid: 814, https://arxiv.org/pdf/2507.01790.pdf GitHub

Abstract:
AI models are increasingly required to be multimodal, integrating disparate input streams into a coherent state representation on which subsequent behaviors and actions can be based. This paper seeks to understand how such models behave when input streams present conflicting information. Focusing specifically on vision-language models, we provide inconsistent inputs (e.g., an image of a dog paired with the caption "A photo of a cat") and ask the model to report the information present in one of the specific modalities (e.g., "What does the caption say / What is in the image?"). We find that models often favor one modality over the other, e.g., reporting the image regardless of what the caption says, but that different models differ in which modality they favor. We find evidence that the behaviorally preferred modality is evident in the internal representational structure of the model, and that specific attention heads can restructure the representations to favor one modality over the other. Moreover, we find modality-agnostic "router heads" which appear to promote answers about the modality requested in the instruction, and which can be manipulated or transferred in order to improve performance across datasets and modalities. Together, the work provides essential steps towards identifying and controlling if and how models detect and resolve conflicting signals within complex multimodal environments.

Paperid: 815, https://arxiv.org/pdf/2507.01738.pdf GitHub

Abstract:
Referring Image Segmentation (RIS) is a challenging task that aims to segment objects in an image based on natural language expressions. While prior studies have predominantly concentrated on improving vision-language interactions and achieving fine-grained localization, a systematic analysis of the fundamental bottlenecks in existing RIS frameworks remains underexplored. To bridge this gap, we propose DeRIS, a novel framework that decomposes RIS into two key components: perception and cognition. This modular decomposition facilitates a systematic analysis of the primary bottlenecks impeding RIS performance. Our findings reveal that the predominant limitation lies not in perceptual deficiencies, but in the insufficient multi-modal cognitive capacity of current models. To mitigate this, we propose a Loopback Synergy mechanism, which enhances the synergy between the perception and cognition modules, thereby enabling precise segmentation while simultaneously improving robust image-text comprehension. Additionally, we analyze and introduce a simple non-referent sample conversion data augmentation to address the long-tail distribution issue related to target existence judgement in general scenarios. Notably, DeRIS demonstrates inherent adaptability to both non- and multi-referents scenarios without requiring specialized architectural modifications, enhancing its general applicability. The codes and models are available at https://github.com/Dmmm1997/DeRIS.

Paperid: 816, https://arxiv.org/pdf/2507.01544.pdf GitHub

Abstract:
Scientific applications of machine learning often rely on small, specialized models tuned to particular domains. Such models often achieve excellent performance, but lack flexibility. Foundation models offer versatility, but typically underperform specialized approaches, especially on non-traditional modalities and long-tail domains. We propose MARVIS (Modality Adaptive Reasoning over VISualizations), a training-free method that enables even small vision-language models to predict any data modality with high accuracy. MARVIS transforms latent embedding spaces into visual representations and then leverages the spatial and fine-grained reasoning skills of VLMs to successfully interpret and utilize them. MARVIS achieves competitive performance on vision, audio, biological, and tabular domains using a single 3B parameter model, achieving results that beat Gemini by 16\% on average and approach specialized methods, without exposing personally identifiable information (P.I.I.) or requiring any domain-specific training. We open source our code and datasets at https://github.com/penfever/marvis

Paperid: 817, https://arxiv.org/pdf/2507.01504.pdf GitHub

Abstract:
The collection and release of street-level recordings as Open Data play a vital role in advancing autonomous driving systems and AI research. However, these datasets pose significant privacy risks, particularly for pedestrians, due to the presence of Personally Identifiable Information (PII) that extends beyond biometric traits such as faces. In this paper, we present cRID, a novel cross-modal framework combining Large Vision-Language Models, Graph Attention Networks, and representation learning to detect textual describable clues of PII and enhance person re-identification (Re-ID). Our approach focuses on identifying and leveraging interpretable features, enabling the detection of semantically meaningful PII beyond low-level appearance cues. We conduct a systematic evaluation of PII presence in person image datasets. Our experiments show improved performance in practical cross-dataset Re-ID scenarios, notably from Market-1501 to CUHK03-np (detected), highlighting the framework's practical utility. Code is available at https://github.com/RAufschlaeger/cRID.

Paperid: 818, https://arxiv.org/pdf/2507.01409.pdf GitHub

Abstract:
An image captioning model flexibly switching its language pattern, e.g., descriptiveness and length, should be useful since it can be applied to diverse applications. However, despite the dramatic improvement in generative vision-language models, fine-grained control over the properties of generated captions is not easy due to two reasons: (i) existing models are not given the properties as a condition during training and (ii) existing models cannot smoothly transition its language pattern from one state to the other. Given this challenge, we propose a new approach, CaptionSmiths, to acquire a single captioning model that can handle diverse language patterns. First, our approach quantifies three properties of each caption, length, descriptiveness, and uniqueness of a word, as continuous scalar values, without human annotation. Given the values, we represent the conditioning via interpolation between two endpoint vectors corresponding to the extreme states, e.g., one for a very short caption and one for a very long caption. Empirical results demonstrate that the resulting model can smoothly change the properties of the output captions and show higher lexical alignment than baselines. For instance, CaptionSmiths reduces the error in controlling caption length by 506\% despite better lexical alignment. Code will be available on https://github.com/omron-sinicx/captionsmiths.

Paperid: 819, https://arxiv.org/pdf/2507.01016.pdf GitHub

Abstract:
In this paper, we introduce an innovative vector quantization based action tokenizer built upon the largest-scale action trajectory dataset to date, leveraging over 100 times more data than previous approaches. This extensive dataset enables our tokenizer to capture rich spatiotemporal dynamics, resulting in a model that not only accelerates inference but also generates smoother and more coherent action outputs. Once trained, the tokenizer can be seamlessly adapted to a wide range of downstream tasks in a zero-shot manner, from short-horizon reactive behaviors to long-horizon planning. A key finding of our work is that the domain gap between synthetic and real action trajectories is marginal, allowing us to effectively utilize a vast amount of synthetic data during training without compromising real-world performance. To validate our approach, we conducted extensive experiments in both simulated environments and on real robotic platforms. The results demonstrate that as the volume of synthetic trajectory data increases, the performance of our tokenizer on downstream tasks improves significantly-most notably, achieving up to a 30% higher success rate on two real-world tasks in long-horizon scenarios. These findings highlight the potential of our action tokenizer as a robust and scalable solution for real-time embodied intelligence systems, paving the way for more efficient and reliable robotic control in diverse application domains.Project website: https://xiaoxiao0406.github.io/vqvla.github.io

Paperid: 820, https://arxiv.org/pdf/2507.01006.pdf GitHub

Authors:V Team, Wenyi Hong, Wenmeng Yu, Xiaotao Gu, Guo Wang, Guobing Gan, Haomiao Tang, Jiale Cheng, Ji Qi, Junhui Ji, Lihang Pan, Shuaiqi Duan, Weihan Wang, Yan Wang, Yean Cheng, Zehai He, Zhe Su, Zhen Yang, Ziyang Pan, Aohan Zeng, Baoxu Wang, Bin Chen, Boyan Shi, Changyu Pang, Chenhui Zhang, Da Yin, Fan Yang, Guoqing Chen, Jiazheng Xu, Jiale Zhu, Jiali Chen, Jing Chen, Jinhao Chen, Jinghao Lin, Jinjiang Wang, Junjie Chen, Leqi Lei, Letian Gong, Leyi Pan, Mingdao Liu, Mingde Xu, Mingzhi Zhang, Qinkai Zheng, Sheng Yang, Shi Zhong, Shiyu Huang, Shuyuan Zhao, Siyan Xue, Shangqin Tu, Shengbiao Meng, Tianshu Zhang, Tianwei Luo, Tianxiang Hao, Tianyu Tong, Wenkai Li, Wei Jia, Xiao Liu, Xiaohan Zhang, Xin Lyu, Xinyue Fan, Xuancheng Huang, Yanling Wang, Yadong Xue, Yanfeng Wang, Yanzi Wang, Yifan An, Yifan Du, Yiming Shi, Yiheng Huang, Yilin Niu, Yuan Wang, Yuanchang Yue, Yuchen Li, Yutao Zhang, Yuting Wang, Yu Wang, Yuxuan Zhang, Zhao Xue, Zhenyu Hou, Zhengxiao Du, Zihan Wang, Peng Zhang, Debing Liu, Bin Xu, Juanzi Li, Minlie Huang, Yuxiao Dong, Jie Tang

Abstract:
We present GLM-4.1V-Thinking and GLM-4.5V, a family of vision-language models (VLMs) designed to advance general-purpose multimodal understanding and reasoning. In this report, we share our key findings in the development of the reasoning-centric training framework. We first develop a capable vision foundation model with significant potential through large-scale pre-training, which arguably sets the upper bound for the final performance. We then propose Reinforcement Learning with Curriculum Sampling (RLCS) to unlock the full potential of the model, leading to comprehensive capability enhancement across a diverse range of tasks, including STEM problem solving, video understanding, content recognition, coding, grounding, GUI-based agents, and long document interpretation. In a comprehensive evaluation across 42 public benchmarks, GLM-4.5V achieves state-of-the-art performance on nearly all tasks among open-source models of similar size, and demonstrates competitive or even superior results compared to closed-source models such as Gemini-2.5-Flash on challenging tasks including Coding and GUI Agents. Meanwhile, the smaller GLM-4.1V-9B-Thinking remains highly competitive-achieving superior results to the much larger Qwen2.5-VL-72B on 29 benchmarks. We open-source both GLM-4.1V-9B-Thinking and GLM-4.5V. Code, models and more information are released at https://github.com/zai-org/GLM-V.

Paperid: 821, https://arxiv.org/pdf/2507.00898.pdf GitHub GitHub

Abstract:
Recent Large Vision-Language Models (LVLMs) have introduced a new paradigm for understanding and reasoning about image input through textual responses. Although they have achieved remarkable performance across a range of multi-modal tasks, they face the persistent challenge of hallucination, which introduces practical weaknesses and raises concerns about their reliable deployment in real-world applications. Existing work has explored contrastive decoding approaches to mitigate this issue, where the output of the original LVLM is compared and contrasted with that of a perturbed version. However, these methods require two or more queries that slow down LVLM response generation, making them less suitable for real-time applications. To overcome this limitation, we propose ONLY, a training-free decoding approach that requires only a single query and a one-layer intervention during decoding, enabling efficient real-time deployment. Specifically, we enhance textual outputs by selectively amplifying crucial textual information using a text-to-visual entropy ratio for each token. Extensive experimental results demonstrate that our proposed ONLY consistently outperforms state-of-the-art methods across various benchmarks while requiring minimal implementation effort and computational cost. Code is available at https://github.com/zifuwan/ONLY.

Paperid: 822, https://arxiv.org/pdf/2507.00721.pdf GitHub

Abstract:
Zero-shot domain adaptation (ZSDA) presents substantial challenges due to the lack of images in the target domain. Previous approaches leverage Vision-Language Models (VLMs) to tackle this challenge, exploiting their zero-shot learning capabilities. However, these methods primarily address domain distribution shifts and overlook the misalignment between the detection task and VLMs, which rely on manually crafted prompts. To overcome these limitations, we propose the unified prompt and representation enhancement (UPRE) framework, which jointly optimizes both textual prompts and visual representations. Specifically, our approach introduces a multi-view domain prompt that combines linguistic domain priors with detection-specific knowledge, and a visual representation enhancement module that produces domain style variations. Furthermore, we introduce multi-level enhancement strategies, including relative domain distance and positive-negative separation, which align multi-modal representations at the image level and capture diverse visual representations at the instance level, respectively. Extensive experiments conducted on nine benchmark datasets demonstrate the superior performance of our framework in ZSDA detection scenarios. Code is available at https://github.com/AMAP-ML/UPRE.

Paperid: 823, https://arxiv.org/pdf/2507.00586.pdf GitHub

Abstract:
Academic emotion analysis plays a crucial role in evaluating students' engagement and cognitive states during the learning process. This paper addresses the challenge of automatically recognizing academic emotions through facial expressions in real-world learning environments. While significant progress has been made in facial expression recognition for basic emotions, academic emotion recognition remains underexplored, largely due to the scarcity of publicly available datasets. To bridge this gap, we introduce RAER, a novel dataset comprising approximately 2,700 video clips collected from around 140 students in diverse, natural learning contexts such as classrooms, libraries, laboratories, and dormitories, covering both classroom sessions and individual study. Each clip was annotated independently by approximately ten annotators using two distinct sets of academic emotion labels with varying granularity, enhancing annotation consistency and reliability. To our knowledge, RAER is the first dataset capturing diverse natural learning scenarios. Observing that annotators naturally consider context cues-such as whether a student is looking at a phone or reading a book-alongside facial expressions, we propose CLIP-CAER (CLIP-based Context-aware Academic Emotion Recognition). Our method utilizes learnable text prompts within the vision-language model CLIP to effectively integrate facial expression and context cues from videos. Experimental results demonstrate that CLIP-CAER substantially outperforms state-of-the-art video-based facial expression recognition methods, which are primarily designed for basic emotions, emphasizing the crucial role of context in accurately recognizing academic emotions. Project page: https://zgsfer.github.io/CAER

Paperid: 824, https://arxiv.org/pdf/2507.00525.pdf GitHub

Abstract:
Interpretable communication is essential for safe and trustworthy autonomous driving, yet current vision-language models (VLMs) often operate under idealized assumptions and struggle to capture user intent in real-world scenarios. Existing driving-oriented VQA datasets are limited to full-scene descriptions or waypoint prediction, preventing the assessment of whether VLMs can respond to localized user-driven queries. We introduce Box-QAymo, a box-referring dataset and benchmark designed to both evaluate and finetune VLMs on spatial and temporal reasoning over user-specified objects. Users express intent by drawing bounding boxes, offering a fast and intuitive interface for focused queries in complex scenes. Specifically, we propose a hierarchical evaluation protocol that begins with binary sanity-check questions to assess basic model capacities, and progresses to (1) attribute prediction for box-referred objects, (2) motion understanding of target instances, and (3) spatiotemporal motion reasoning over inter-object dynamics across frames. To support this, we crowd-sourced fine-grained object classes and visual attributes that reflect the complexity drivers encounter, and extract object trajectories to construct temporally grounded QA pairs. Rigorous quality control through negative sampling, temporal consistency checks, and difficulty-aware balancing guarantee dataset robustness and diversity. Our comprehensive evaluation reveals significant limitations in current VLMs when queried about perception questions, highlighting the gap in achieving real-world performance. This work provides a foundation for developing more robust and interpretable autonomous driving systems that can communicate effectively with users under real-world conditions. Project page and dataset are available at https://djamahl99.github.io/qaymo-pages/.

Paperid: 825, https://arxiv.org/pdf/2506.24102.pdf GitHub

Abstract:
Multimodal Large Language Models (MLLMs) demonstrate a complex understanding of scenes, benefiting from large-scale and high-quality datasets. Most existing caption datasets lack the ground locations and relations for visual entities. Several grounded caption datasets face the problems of missing detailed descriptions, relations, and massive object descriptions on high-resolution images. To fill this gap for the community, we present DenseWorld-1M, the first massive, detailed, dense grounded caption dataset in the real world. We design a three-stage labeling pipeline, containing open-world perception, detailed object caption generation, and dense caption merging. The first stage obtains entity-level masks and labels. The second stage generates the object-level, detailed captions with the guidance of masks and labels from the first stage. The final stage merges object captions and masks into spatial and relational dense captions. To accelerate the labeling process and improve caption quality, we present two VLM models: the Detailed Region Caption model and the Spatial Caption Merging model. Extensive experiments on various settings, including vision-language understanding, visual grounding, and region caption generation, demonstrate the effectiveness of our DenseWorld-1M dataset and labeling models.

Paperid: 826, https://arxiv.org/pdf/2506.24044.pdf GitHub GitHub

Abstract:
The rapid progress of multimodal large language models (MLLM) has paved the way for Vision-Language-Action (VLA) paradigms, which integrate visual perception, natural language understanding, and control within a single policy. Researchers in autonomous driving are actively adapting these methods to the vehicle domain. Such models promise autonomous vehicles that can interpret high-level instructions, reason about complex traffic scenes, and make their own decisions. However, the literature remains fragmented and is rapidly expanding. This survey offers the first comprehensive overview of VLA for Autonomous Driving (VLA4AD). We (i) formalize the architectural building blocks shared across recent work, (ii) trace the evolution from early explainer to reasoning-centric VLA models, and (iii) compare over 20 representative models according to VLA's progress in the autonomous driving domain. We also consolidate existing datasets and benchmarks, highlighting protocols that jointly measure driving safety, accuracy, and explanation quality. Finally, we detail open challenges - robustness, real-time efficiency, and formal verification - and outline future directions of VLA4AD. This survey provides a concise yet complete reference for advancing interpretable socially aligned autonomous vehicles. Github repo is available at \href{https://github.com/JohnsonJiang1996/Awesome-VLA4AD}{SicongJiang/Awesome-VLA4AD}.

Paperid: 827, https://arxiv.org/pdf/2506.24016.pdf GitHub

Abstract:
Recent advances in large language models and vision-language models have led to growing interest in explainable evaluation metrics for image captioning. However, these metrics generate explanations without standardized criteria, and the overall quality of the generated explanations remains unverified. In this paper, we propose EXPERT, a reference-free evaluation metric that provides structured explanations based on three fundamental criteria: fluency, relevance, and descriptiveness. By constructing large-scale datasets of high-quality structured explanations, we develop a two-stage evaluation template to effectively supervise a vision-language model for both scoring and explanation generation. EXPERT achieves state-of-the-art results on benchmark datasets while providing significantly higher-quality explanations than existing metrics, as validated through comprehensive human evaluation. Our code and datasets are available at https://github.com/hjkim811/EXPERT.

Paperid: 828, https://arxiv.org/pdf/2506.24000.pdf GitHub

Abstract:
Test-time adaptation (TTA) methods have gained significant attention for enhancing the performance of vision-language models (VLMs) such as CLIP during inference, without requiring additional labeled data. However, current TTA researches generally suffer from major limitations such as duplication of baseline results, limited evaluation metrics, inconsistent experimental settings, and insufficient analysis. These problems hinder fair comparisons between TTA methods and obscure their practical strengths and weaknesses. To address these challenges, we introduce TTA-VLM, a comprehensive benchmark for evaluating TTA methods on VLMs. Our benchmark implements 8 episodic TTA and 7 online TTA methods within a unified and reproducible framework, and evaluates them across 15 widely used datasets. Unlike prior studies focused solely on CLIP, we extend the evaluation to SigLIP--a model trained with a Sigmoid loss--and include training-time tuning methods such as CoOp, MaPLe, and TeCoA to assess generality. Beyond classification accuracy, TTA-VLM incorporates various evaluation metrics, including robustness, calibration, out-of-distribution detection, and stability, enabling a more holistic assessment of TTA methods. Through extensive experiments, we find that 1) existing TTA methods produce limited gains compared to the previous pioneering work; 2) current TTA methods exhibit poor collaboration with training-time fine-tuning methods; 3) accuracy gains frequently come at the cost of reduced model trustworthiness. We release TTA-VLM to provide fair comparison and comprehensive evaluation of TTA methods for VLMs, and we hope it encourages the community to develop more reliable and generalizable TTA strategies.

Paperid: 829, https://arxiv.org/pdf/2506.23919.pdf GitHub

Abstract:
Generalization remains a fundamental challenge in robotic manipulation. To tackle this challenge, recent Vision-Language-Action (VLA) models build policies on top of Vision-Language Models (VLMs), seeking to transfer their open-world semantic knowledge. However, their zero-shot capability lags significantly behind the base VLMs, as the instruction-vision-action data is too limited to cover diverse scenarios, tasks, and robot embodiments. In this work, we present Goal-VLA, a zero-shot framework that leverages Image-Generative VLMs as world models to generate desired goal states, from which the target object pose is derived to enable generalizable manipulation. The key insight is that object state representation is the golden interface, naturally separating a manipulation system into high-level and low-level policies. This representation abstracts away explicit action annotations, allowing the use of highly generalizable VLMs while simultaneously providing spatial cues for training-free low-level control. To further improve robustness, we introduce a Reflection-through-Synthesis process that iteratively validates and refines the generated goal image before execution. Both simulated and real-world experiments demonstrate that our \name achieves strong performance and inspiring generalizability in manipulation tasks. Supplementary materials are available at https://nus-lins-lab.github.io/goalvlaweb/.

Paperid: 830, https://arxiv.org/pdf/2506.23856.pdf GitHub

Abstract:
Despite the great promise of Prompt Tuning (PT) in adapting large Vision-Language Pretrained Models (VLPMs) to downstream tasks, they often struggle to overcome the Base-New Tradeoff (BNT) dilemma: as VLPMs are better tuned to a base task, their ability to generalize to new tasks diminishes. Recent work on conditional PT addresses this problem by replacing static prompts with dynamic Visual Image Information (VII)-conditioned prompts, improving the model's generalization to new tasks to some extent. In this work, we first identify a critical issue with existing conditional PT methods: using VII as the "condition" of prompts yields suboptimal performance, and even random noise-conditioned prompts can outperform the VII-conditioned counterparts. On further analysis, we find that learning dynamic prompts conditioned on Textual Class Information (TCI) is the key to solving the BNT problem. Motivated by this, we then propose Class-adaptive Prompt Tuning (CaPT), which enables fast adaptation of tuned models to new classes by learning TCI-conditioned prompts from base classes. Remarkably, CaPT can be used as a plugin to mitigate the BNT problem for existing unconditional PT schemes. Extensive experiments on 11 datasets show that CaPT consistently improves the performance of five strong unconditional PT baselines with negligible additional computational cost. Additionally, by integrating CaPT with our recently proposed DePT framework, we devise a new conditional PT approach, termed DeCaPT, which outperforms the H ACC of the state-of-the-art conditional PT scheme by 3.49%, averaged over the 11 datasets. Code: https://github.com/Koorye/CaPT.

Paperid: 831, https://arxiv.org/pdf/2506.23822.pdf GitHub

Abstract:
Large-scale vision-language models (VLMs), such as CLIP, have achieved remarkable success in zero-shot learning (ZSL) by leveraging large-scale visual-text pair datasets. However, these methods often lack interpretability, as they compute the similarity between an entire query image and the embedded category words, making it difficult to explain their predictions. One approach to address this issue is to develop interpretable models by integrating language, where classifiers are built using discrete attributes, similar to human perception. This introduces a new challenge: how to effectively align local visual features with corresponding attributes based on pre-trained VLMs. To tackle this, we propose LaZSL, a locally-aligned vision-language model for interpretable ZSL. LaZSL employs local visual-semantic alignment via optimal transport to perform interaction between visual regions and their associated attributes, facilitating effective alignment and providing interpretable similarity without the need for additional training. Extensive experiments demonstrate that our method offers several advantages, including enhanced interpretability, improved accuracy, and strong domain generalization. Codes available at: https://github.com/shiming-chen/LaZSL.

Paperid: 832, https://arxiv.org/pdf/2506.23785.pdf GitHub

Abstract:
We propose VisTex-OVLM, a novel image prompted object detection method that introduces visual textualization -- a process that projects a few visual exemplars into the text feature space to enhance Object-level Vision-Language Models' (OVLMs) capability in detecting rare categories that are difficult to describe textually and nearly absent from their pre-training data, while preserving their pre-trained object-text alignment. Specifically, VisTex-OVLM leverages multi-scale textualizing blocks and a multi-stage fusion strategy to integrate visual information from visual exemplars, generating textualized visual tokens that effectively guide OVLMs alongside text prompts. Unlike previous methods, our method maintains the original architecture of OVLM, maintaining its generalization capabilities while enhancing performance in few-shot settings. VisTex-OVLM demonstrates superior performance across open-set datasets which have minimal overlap with OVLM's pre-training data and achieves state-of-the-art results on few-shot benchmarks PASCAL VOC and MSCOCO. The code will be released at https://github.com/WitGotFlg/VisTex-OVLM.

Paperid: 833, https://arxiv.org/pdf/2506.23663.pdf GitHub

Abstract:
In real-world vision-language applications, practitioners increasingly rely on large, pretrained foundation models rather than custom-built solutions, despite limited transparency regarding their training data and processes. While these models achieve impressive performance on general benchmarks, their effectiveness can decline notably under specialized domain shifts, such as unique imaging conditions or environmental variations. In this work, we introduce Deepbench, a framework designed to assess domain-specific robustness of vision-language models (VLMs). Deepbench leverages a large language model (LLM) to generate realistic, context-aware image corruptions tailored to specific deployment domains without requiring labeled data. We evaluate a range of contrastive vision-language architectures and architectural variants across six real-world domains and observe substantial variability in robustness, highlighting the need for targeted, domain-aware evaluation. Deepbench is released as open-source software to support further research into domain-aware robustness assessment.

Paperid: 834, https://arxiv.org/pdf/2506.23580.pdf GitHub

Abstract:
Dataset distillation (DD) condenses large datasets into compact yet informative substitutes, preserving performance comparable to the original dataset while reducing storage, transmission costs, and computational consumption. However, previous DD methods mainly focus on distilling information from images, often overlooking the semantic information inherent in the data. The disregard for context hinders the model's generalization ability, particularly in tasks involving complex datasets, which may result in illogical outputs or the omission of critical objects. In this study, we integrate vision-language methods into DD by introducing text prototypes to distill language information and collaboratively synthesize data with image prototypes, thereby enhancing dataset distillation performance. Notably, the text prototypes utilized in this study are derived from descriptive text information generated by an open-source large language model. This framework demonstrates broad applicability across datasets without pre-existing text descriptions, expanding the potential of dataset distillation beyond traditional image-based approaches. Compared to other methods, the proposed approach generates logically coherent images containing target objects, achieving state-of-the-art validation performance and demonstrating robust generalization. Source code and generated data are available in https://github.com/zou-yawen/Dataset-Distillation-via-Vision-Language-Category-Prototype/

Paperid: 835, https://arxiv.org/pdf/2506.23491.pdf GitHub

Abstract:
In this paper, we present ZonUI-3B, a lightweight Vision-Language Model (VLM) that can be fully trained on a single consumer-grade GPU (RTX 4090) while delivering performance comparable to significantly larger models on GUI grounding tasks. The model incorporates several key innovations: (i) combine cross-platform, multi-resolution dataset of 24K examples from diverse sources including mobile, desktop, and web GUI screenshots to effectively address data scarcity in high-resolution desktop environments; (ii) a two-stage fine-tuning strategy, where initial cross-platform training establishes robust GUI understanding, followed by specialized fine-tuning on high-resolution data to significantly enhance model adaptability; and (iii) data curation and redundancy reduction strategies, demonstrating that randomly sampling a smaller subset with reduced redundancy achieves performance comparable to larger datasets, emphasizing data diversity over sheer volume. Empirical evaluation on standard GUI grounding benchmarks, including ScreenSpot, ScreenSpot-v2, and the challenging ScreenSpot-Pro, highlights ZonUI-3B's exceptional accuracy, achieving 84.9% on ScreenSpot and 86.4% on ScreenSpot-v2, surpassing prior models under 4B parameters. Ablation studies validate the critical role of balanced sampling and two-stage fine-tuning in enhancing robustness, particularly in high-resolution desktop scenarios. The ZonUI-3B is available at: https://github.com/Han1018/ZonUI-3B

Paperid: 836, https://arxiv.org/pdf/2506.23352.pdf GitHub

Abstract:
The advancement of 3D language fields has enabled intuitive interactions with 3D scenes via natural language. However, existing approaches are typically limited to small-scale environments, lacking the scalability and compositional reasoning capabilities necessary for large, complex urban settings. To overcome these limitations, we propose GeoProg3D, a visual programming framework that enables natural language-driven interactions with city-scale high-fidelity 3D scenes. GeoProg3D consists of two key components: (i) a Geography-aware City-scale 3D Language Field (GCLF) that leverages a memory-efficient hierarchical 3D model to handle large-scale data, integrated with geographic information for efficiently filtering vast urban spaces using directional cues, distance measurements, elevation data, and landmark references; and (ii) Geographical Vision APIs (GV-APIs), specialized geographic vision tools such as area segmentation and object detection. Our framework employs large language models (LLMs) as reasoning engines to dynamically combine GV-APIs and operate GCLF, effectively supporting diverse geographic vision tasks. To assess performance in city-scale reasoning, we introduce GeoEval3D, a comprehensive benchmark dataset containing 952 query-answer pairs across five challenging tasks: grounding, spatial reasoning, comparison, counting, and measurement. Experiments demonstrate that GeoProg3D significantly outperforms existing 3D language fields and vision-language models across multiple tasks. To our knowledge, GeoProg3D is the first framework enabling compositional geographic reasoning in high-fidelity city-scale 3D environments via natural language. The code is available at https://snskysk.github.io/GeoProg3D/.

Paperid: 837, https://arxiv.org/pdf/2506.23309.pdf GitHub GitHub

Abstract:
In contemporary surgical research and practice, accurately comprehending 3D surgical scenes with text-promptable capabilities is particularly crucial for surgical planning and real-time intra-operative guidance, where precisely identifying and interacting with surgical tools and anatomical structures is paramount. However, existing works focus on surgical vision-language model (VLM), 3D reconstruction, and segmentation separately, lacking support for real-time text-promptable 3D queries. In this paper, we present SurgTPGS, a novel text-promptable Gaussian Splatting method to fill this gap. We introduce a 3D semantics feature learning strategy incorporating the Segment Anything model and state-of-the-art vision-language models. We extract the segmented language features for 3D surgical scene reconstruction, enabling a more in-depth understanding of the complex surgical environment. We also propose semantic-aware deformation tracking to capture the seamless deformation of semantic features, providing a more precise reconstruction for both texture and semantic features. Furthermore, we present semantic region-aware optimization, which utilizes regional-based semantic information to supervise the training, particularly promoting the reconstruction quality and semantic smoothness. We conduct comprehensive experiments on two real-world surgical datasets to demonstrate the superiority of SurgTPGS over state-of-the-art methods, highlighting its potential to revolutionize surgical practices. SurgTPGS paves the way for developing next-generation intelligent surgical systems by enhancing surgical precision and safety. Our code is available at: https://github.com/lastbasket/SurgTPGS.

Paperid: 838, https://arxiv.org/pdf/2506.23115.pdf GitHub

Abstract:
Multimodal embedding models, built upon causal Vision Language Models (VLMs), have shown promise in various tasks. However, current approaches face three key limitations: the use of causal attention in VLM backbones is suboptimal for embedding tasks; scalability issues due to reliance on high-quality labeled paired data for contrastive learning; and limited diversity in training objectives and data. To address these issues, we propose MoCa, a two-stage framework for transforming pre-trained VLMs into effective bidirectional multimodal embedding models. The first stage, Modality-aware Continual Pre-training, introduces a joint reconstruction objective that simultaneously denoises interleaved text and image inputs, enhancing bidirectional context-aware reasoning. The second stage, Heterogeneous Contrastive Fine-tuning, leverages diverse, semantically rich multimodal data beyond simple image-caption pairs to enhance generalization and alignment. Our method addresses the stated limitations by introducing bidirectional attention through continual pre-training, scaling effectively with massive unlabeled datasets via joint reconstruction objectives, and utilizing diverse multimodal data for enhanced representation robustness. Experiments demonstrate that MoCa consistently improves performance across MMEB and ViDoRe-v2 benchmarks, achieving new state-of-the-art results, and exhibits strong scalability with both model size and training data on MMEB.

Paperid: 839, https://arxiv.org/pdf/2506.23061.pdf GitHub

Abstract:
Empowering Small-scale Vision-Language Models (SVLMs) with reliable thinking capabilities remains fundamentally challenging due to their limited parameter capacity and weak instruction-following abilities. Existing training paradigms, including Supervised Fine-Tuning (SFT) and Reinforcement Learning with Verifiable Reward (RLVR), impose substantial demands on the base VLM, exceeding the capabilities of SVLMs. Consequently, directly applying these paradigms to SVLMs often suffers from severe pseudo thinking traces and advantage collapse, ultimately undermining both thinking reliability and task performance. A natural solution is to combine SFT and RLVR, leveraging their complementarity to reduce the dependence on model capacity. However, the widely adopted two-stage training paradigm still performs poorly on SVLMs, as their tendency toward sub-optimal convergence hinders the trade-off and limits the benefits of the combination. To address this, we propose DyME, a novel training paradigm that Dynamically selects between Memorization (via SFT) and Exploration (via RLVR) modes at each optimization step, ensuring that every update contributes to the trade-off. Extensive experiments across diverse domains demonstrate that DyME consistently achieves this balance, and thus delivers substantial performance improvements. These results establish DyME as a practical and effective solution for empowering SVLMs with reliable thinking capabilities. GitHub: https://github.com/HKUST-LongGroup/DyME

Paperid: 840, https://arxiv.org/pdf/2506.22979.pdf GitHub

Abstract:
Generalized Few-Shot Semantic Segmentation (GFSS) aims to extend a segmentation model to novel classes with only a few annotated examples while maintaining performance on base classes. Recently, pretrained vision-language models (VLMs) such as CLIP have been leveraged in GFSS to improve generalization on novel classes through multi-modal prototypes learning. However, existing prototype-based methods are inherently deterministic, limiting the adaptability of learned prototypes to diverse samples, particularly for novel classes with scarce annotations. To address this, we propose FewCLIP, a probabilistic prototype calibration framework over multi-modal prototypes from the pretrained CLIP, thus providing more adaptive prototype learning for GFSS. Specifically, FewCLIP first introduces a prototype calibration mechanism, which refines frozen textual prototypes with learnable visual calibration prototypes, leading to a more discriminative and adaptive representation. Furthermore, unlike deterministic prototype learning techniques, FewCLIP introduces distribution regularization over these calibration prototypes. This probabilistic formulation ensures structured and uncertainty-aware prototype learning, effectively mitigating overfitting to limited novel class data while enhancing generalization. Extensive experimental results on PASCAL-5$^i$ and COCO-20$^i$ datasets demonstrate that our proposed FewCLIP significantly outperforms state-of-the-art approaches across both GFSS and class-incremental setting. The code is available at https://github.com/jliu4ai/FewCLIP.

Paperid: 841, https://arxiv.org/pdf/2506.22900.pdf GitHub

Abstract:
Medical visual question answering (MedVQA) plays a vital role in clinical decision-making by providing contextually rich answers to image-based queries. Although vision-language models (VLMs) are widely used for this task, they often generate factually incorrect answers. Retrieval-augmented generation addresses this challenge by providing information from external sources, but risks retrieving irrelevant context, which can degrade the reasoning capabilities of VLMs. Re-ranking retrievals, as introduced in existing approaches, enhances retrieval relevance by focusing on query-text alignment. However, these approaches neglect the visual or multimodal context, which is particularly crucial for medical diagnosis. We propose MOTOR, a novel multimodal retrieval and re-ranking approach that leverages grounded captions and optimal transport. It captures the underlying relationships between the query and the retrieved context based on textual and visual information. Consequently, our approach identifies more clinically relevant contexts to augment the VLM input. Empirical analysis and human expert evaluation demonstrate that MOTOR achieves higher accuracy on MedVQA datasets, outperforming state-of-the-art methods by an average of 6.45%. Code is available at https://github.com/BioMedIA-MBZUAI/MOTOR.

Paperid: 842, https://arxiv.org/pdf/2506.22819.pdf GitHub

Abstract:
Vision-language models (VLM) have demonstrated impressive performance in image recognition by leveraging self-supervised training on large datasets. Their performance can be further improved by adapting to the test sample using test-time prompt tuning (TPT). Unfortunately, the singular focus of TPT approaches on improving the accuracy suffers from tunnel vision, and leads to degradation in confidence calibration. This limits the applicability of TPT in critical applications. We make three contributions in this work. (1) We posit that random or naive initialization of prompts leads to overfitting on a particular test sample, and is the main reason for miscalibration of the VLM after TPT. To mitigate the problem, we propose careful initialization of test time prompt using prior knowledge about the target label attributes from a large language model (LLM); (2) To further maintain the quality of prompts during \tpt, we propose a novel regularization loss to reduce intraclass distance, and increase inter-class distance between the learnt Through extensive experiments on different CLIP architectures and 15 datasets, we show that our approach can effectively improve the calibration after TPT. We report an average expected calibration error (ECE) of 4.11 with our method, TCA, compared to 11.7 for vanilla TPT, 6.12 for C-TPT (ICLR'24), 6.78 for DiffTPT (CVPR'23), and 8.43 for PromptAlign (NeurIPS'23). The code is publicly accessible at: https://github.com/rhebbalaguppe/TCA_PromptWithoutPanic.

Paperid: 843, https://arxiv.org/pdf/2506.22274.pdf GitHub GitHub

Abstract:
Natural scenes provide us with rich contexts for object recognition and reference. In particular, knowing what type of scene one is looking at generates expectations about which objects will occur, and what their spatial configuration should be. Do Vision-Language Models (VLMs) learn to rely on scene contexts in a similar way, when generating references to objects? To address this question, we introduce the \textit{Common Objects Out-of-Context (COOCO)} dataset and test to what extent VLMs rely on scene context to refer to objects under different degrees of scene-object congruency, and different perturbations. Our findings show that models leverage scene context adaptively, depending on both the semantic relatedness between object and scene and the level of noise. In particular, models rely more on context under high target-scene congruence or when objects are degraded. Attention analysis reveals that successful object categorisation involves increased focus on the target in mid-level layers, especially under moderate noise, suggesting that VLMs dynamically balance local and contextual information for reference generation. We make our dataset, code and models available at \href{https://github.com/cs-nlp-uu/scenereg}{https://github.com/cs-nlp-uu/scenereg}.

Paperid: 844, https://arxiv.org/pdf/2506.22149.pdf GitHub

Abstract:
The rise of imaging techniques such as optical coherence tomography (OCT) and advances in deep learning (DL) have enabled clinicians and researchers to streamline retinal disease staging. A popular DL approach is self-supervised learning (SSL), where models learn from vast amounts of unlabeled data, avoiding costly annotation. SSL has allowed the development of foundation models (FMs), large models that can be used for a variety of downstream tasks. However, existing FMs for OCT, trained solely on image data, lack a comprehensive and robust semantic understanding of images, as evidenced by their downstream performance (especially for complex tasks), and thus require supervised fine-tuning (which may be unfeasible) to better adapt to specific applications and populations. To address this, we propose RetFiner, an SSL vision-language refinement scheme that improves the representations of existing FMs and enables their efficient and direct adaptation to specific populations for improved downstream performance. Our method uses a diverse set of training objectives which take advantage of the rich supervisory signal found in textual data. We tested RetFiner on the retinal FMs RETFound, UrFound, and VisionFM, showing significant improvements in linear probing performance on seven highly diverse OCT classification tasks, with an average increase of 5.8, 3.9, and 2.1 percentage points over their baselines, respectively. Our code and model weights are publicly available at https://github.com/ronnief1/RetFiner.

Paperid: 845, https://arxiv.org/pdf/2506.21833.pdf GitHub

Abstract:
Forward-mode automatic differentiation (FmAD) and zero-order (ZO) optimization have been proposed as memory-efficient alternatives to backpropagation (BP) for gradient computation, especially in low-resource settings. However, their practical benefits remain unclear due to two key gaps: a lack of comparison against memory-efficient BP variants, such as activation checkpointing, and a lack of a unified theoretical analysis. This work presents a comprehensive theoretical and empirical comparison of BP, FmAD, and ZO methods. Our theoretical analysis shows that while FmAD, and ZO can reduce memory usage, they incur significant costs in accuracy, convergence speed, and computation compared to BP with checkpointing. These drawbacks worsen with larger models or constrained perturbation budgets. Empirical experiments on large language and vision-language models show that BP with checkpointing outperforms FmAD and ZO variants, including those enhanced with variance reduction, achieving up to 31.1% higher accuracy, 34.8% faster convergence, and 3.8x fewer computations at comparable memory usage. Our results highlight fundamental limitations of FmAD and ZO, and reaffirm BP with checkpointing as the most effective strategy for model training under memory-constrained settings. Our code is available at https://github.com/Astuary/The_Cost_of_Avoiding_Backpropagation.

Paperid: 846, https://arxiv.org/pdf/2506.21601.pdf GitHub

Abstract:
Multi-vector document retrieval systems, such as ColPali, excel in fine-grained matching for complex queries but incur significant storage and computational costs due to their reliance on high-dimensional patch embeddings and late-interaction scoring. To address these challenges, we propose HPC-ColPali, a Hierarchical Patch Compression framework that enhances the efficiency of ColPali while preserving its retrieval accuracy. Our approach integrates three innovative techniques: (1) K-Means quantization, which compresses patch embeddings into 1-byte centroid indices, achieving up to 32$\times$ storage reduction; (2) attention-guided dynamic pruning, utilizing Vision-Language Model attention weights to retain only the top-$p\%$ most salient patches, reducing late-interaction computation by up to 60\% with less than 2\% nDCG@10 loss; and (3) optional binary encoding of centroid indices into $b$-bit strings ($b=\lceil\log_2 K\rceil$), enabling rapid Hamming distance-based similarity search for resource-constrained environments. Evaluated on the ViDoRe and SEC-Filings datasets, HPC-ColPali achieves 30--50\% lower query latency under HNSW indexing while maintaining high retrieval precision. When integrated into a Retrieval-Augmented Generation pipeline for legal summarization, it reduces hallucination rates by 30\% and halves end-to-end latency. These advancements establish HPC-ColPali as a scalable and efficient solution for multi-vector document retrieval across diverse applications. Code is available at https://github.com/DngBack/HPC-ColPali.

Paperid: 847, https://arxiv.org/pdf/2506.21546.pdf GitHub

Abstract:
Recent progress in vision-language segmentation has significantly advanced grounded visual understanding. However, these models often exhibit hallucinations by producing segmentation masks for objects not grounded in the image content or by incorrectly labeling irrelevant regions. Existing evaluation protocols for segmentation hallucination primarily focus on label or textual hallucinations without manipulating the visual context, limiting their capacity to diagnose critical failures. In response, we introduce HalluSegBench, the first benchmark specifically designed to evaluate hallucinations in visual grounding through the lens of counterfactual visual reasoning. Our benchmark consists of a novel dataset of 1340 counterfactual instance pairs spanning 281 unique object classes, and a set of newly introduced metrics that quantify hallucination sensitivity under visually coherent scene edits. Experiments on HalluSegBench with state-of-the-art vision-language segmentation models reveal that vision-driven hallucinations are significantly more prevalent than label-driven ones, with models often persisting in false segmentation, highlighting the need for counterfactual reasoning to diagnose grounding fidelity.

Paperid: 848, https://arxiv.org/pdf/2506.21539.pdf GitHub

Abstract:
We present WorldVLA, an autoregressive action world model that unifies action and image understanding and generation. Our WorldVLA intergrates Vision-Language-Action (VLA) model and world model in one single framework. The world model predicts future images by leveraging both action and image understanding, with the purpose of learning the underlying physics of the environment to improve action generation. Meanwhile, the action model generates the subsequent actions based on image observations, aiding in visual understanding and in turn helps visual generation of the world model. We demonstrate that WorldVLA outperforms standalone action and world models, highlighting the mutual enhancement between the world model and the action model. In addition, we find that the performance of the action model deteriorates when generating sequences of actions in an autoregressive manner. This phenomenon can be attributed to the model's limited generalization capability for action prediction, leading to the propagation of errors from earlier actions to subsequent ones. To address this issue, we propose an attention mask strategy that selectively masks prior actions during the generation of the current action, which shows significant performance improvement in the action chunk generation task.

Paperid: 849, https://arxiv.org/pdf/2506.21476.pdf GitHub

Abstract:
Learning the hierarchical structure of data in vision-language models is a significant challenge. Previous works have attempted to address this challenge by employing entailment learning. However, these approaches fail to model the transitive nature of entailment explicitly, which establishes the relationship between order and semantics within a representation space. In this work, we introduce Radial Cross-Modal Embeddings (RCME), a framework that enables the explicit modeling of transitivity-enforced entailment. Our proposed framework optimizes for the partial order of concepts within vision-language models. By leveraging our framework, we develop a hierarchical vision-language foundation model capable of representing the hierarchy in the Tree of Life. Our experiments on hierarchical species classification and hierarchical retrieval tasks demonstrate the enhanced performance of our models compared to the existing state-of-the-art models. Our code and models are open-sourced at https://vishu26.github.io/RCME/index.html.

Paperid: 850, https://arxiv.org/pdf/2506.21317.pdf GitHub

Abstract:
Current vision-language models (VLMs) are well-adapted for general visual understanding tasks. However, they perform inadequately when handling complex visual tasks related to human poses and actions due to the lack of specialized vision-language instruction-following data. We introduce a method for generating such data by integrating human keypoints with traditional visual features such as captions and bounding boxes, enabling more precise understanding of human-centric scenes. Our approach constructs a dataset comprising 200,328 samples tailored to fine-tune models for human-centric tasks, focusing on three areas: conversation, detailed description, and complex reasoning. We establish an Extended Human Pose and Action Understanding Benchmark (E-HPAUB) to assess model performance on human pose and action understanding. We fine-tune the LLaVA-1.5-7B model using this dataset and evaluate our resulting LLaVA-Pose model on the benchmark, achieving significant improvements. Experimental results show an overall improvement of 33.2% compared to the original LLaVA-1.5-7B model. These findings highlight the effectiveness of keypoint-integrated data in enhancing multimodal models for human-centric visual understanding. Code is available at https://github.com/Ody-trek/LLaVA-Pose.

Paperid: 851, https://arxiv.org/pdf/2506.21001.pdf GitHub

Abstract:
Challenges such as the lack of high-quality annotations, long-tailed data distributions, and inconsistent staining styles pose significant obstacles to training neural networks to detect abnormal cells in cytopathology robustly. This paper proposes a style-aligned image composition (SAIC) method that composes high-fidelity and style-preserved pathological images to enhance the effectiveness and robustness of detection models. Without additional training, SAIC first selects an appropriate candidate from the abnormal cell bank based on attribute guidance. Then, it employs a high-frequency feature reconstruction to achieve a style-aligned and high-fidelity composition of abnormal cells and pathological backgrounds. Finally, it introduces a large vision-language model to filter high-quality synthesis images. Experimental results demonstrate that incorporating SAIC-synthesized images effectively enhances the performance and robustness of abnormal cell detection for tail categories and styles, thereby improving overall detection performance. The comprehensive quality evaluation further confirms the generalizability and practicality of SAIC in clinical application scenarios. Our code will be released at https://github.com/Joey-Qi/SAIC.

Paperid: 852, https://arxiv.org/pdf/2506.20966.pdf GitHub

Abstract:
Vision-language-action (VLA) models extend vision-language models (VLM) by integrating action generation modules for robotic manipulation. Leveraging strengths of VLM in vision perception and instruction understanding, VLA models exhibit promising generalization across diverse manipulation tasks. However, applications demanding high precision and accuracy reveal performance gaps without further adaptation. Evidence from multiple domains highlights the critical role of post-training to align foundational models with downstream applications, spurring extensive research on post-training VLA models. VLA model post-training aims to address the challenge of improving an embodiment's ability to interact with the environment for the given tasks, analogous to the process of humans motor skills acquisition. Accordingly, this paper reviews post-training strategies for VLA models through the lens of human motor learning, focusing on three dimensions: environments, embodiments, and tasks. A structured taxonomy is introduced aligned with human learning mechanisms: (1) enhancing environmental perception, (2) improving embodiment awareness, (3) deepening task comprehension, and (4) multi-component integration. Finally, key challenges and trends in post-training VLA models are identified, establishing a conceptual framework to guide future research. This work delivers both a comprehensive overview of current VLA model post-training methods from a human motor learning perspective and practical insights for VLA model development. (Project website: https://github.com/AoqunJin/Awesome-VLA-Post-Training)

Paperid: 853, https://arxiv.org/pdf/2506.20566.pdf GitHub

Abstract:
Real-time human perception is crucial for effective human-robot interaction (HRI). Large vision-language models (VLMs) offer promising generalizable perceptual capabilities but often suffer from high latency, which negatively impacts user experience and limits VLM applicability in real-world scenarios. To systematically study VLM capabilities in human perception for HRI and performance-latency trade-offs, we introduce HRIBench, a visual question-answering (VQA) benchmark designed to evaluate VLMs across a diverse set of human perceptual tasks critical for HRI. HRIBench covers five key domains: (1) non-verbal cue understanding, (2) verbal instruction understanding, (3) human-robot object relationship understanding, (4) social navigation, and (5) person identification. To construct HRIBench, we collected data from real-world HRI environments to curate questions for non-verbal cue understanding, and leveraged publicly available datasets for the remaining four domains. We curated 200 VQA questions for each domain, resulting in a total of 1000 questions for HRIBench. We then conducted a comprehensive evaluation of both state-of-the-art closed-source and open-source VLMs (N=11) on HRIBench. Our results show that, despite their generalizability, current VLMs still struggle with core perceptual capabilities essential for HRI. Moreover, none of the models within our experiments demonstrated a satisfactory performance-latency trade-off suitable for real-time deployment, underscoring the need for future research on developing smaller, low-latency VLMs with improved human perception capabilities. HRIBench and our results can be found in this Github repository: https://github.com/interaction-lab/HRIBench.

Paperid: 854, https://arxiv.org/pdf/2506.20332.pdf GitHub

Abstract:
Vision-language model-based mobile agents have gained the ability to not only understand complex instructions and mobile screenshots, but also optimize their action outputs via thinking and reasoning, benefiting from reinforcement learning, such as Group Relative Policy Optimization (GRPO). However, existing research centers on offline reinforcement learning training or online optimization using action-level rewards, which limits the agent's dynamic interaction with the environment. This often results in agents settling into local optima, thereby weakening their ability for exploration and error action correction. To address these challenges, we introduce an approach called Mobile-R1, which employs interactive multi-turn reinforcement learning with task-level rewards for mobile agents. Our training framework consists of three stages: initial format finetuning, single-step online training via action-level reward, followed by online training via task-level reward based on multi-turn trajectories. This strategy is designed to enhance the exploration and error correction capabilities of Mobile-R1, leading to significant performance improvements. Moreover, we have collected a dataset covering 28 Chinese applications with 24,521 high-quality manual annotations and established a new benchmark with 500 trajectories. We will open source all resources, including the dataset, benchmark, model weight, and codes: https://mobile-r1.github.io/Mobile-R1/.

Paperid: 855, https://arxiv.org/pdf/2506.20254.pdf GitHub

Abstract:
The complexity and diversity of surgical workflows, driven by heterogeneous operating room settings, institutional protocols, and anatomical variability, present a significant challenge in developing generalizable models for cross-institutional and cross-procedural surgical understanding. While recent surgical foundation models pretrained on large-scale vision-language data offer promising transferability, their zero-shot performance remains constrained by domain shifts, limiting their utility in unseen surgical environments. To address this, we introduce Surgical Phase Anywhere (SPA), a lightweight framework for versatile surgical workflow understanding that adapts foundation models to institutional settings with minimal annotation. SPA leverages few-shot spatial adaptation to align multi-modal embeddings with institution-specific surgical scenes and phases. It also ensures temporal consistency through diffusion modeling, which encodes task-graph priors derived from institutional procedure protocols. Finally, SPA employs dynamic test-time adaptation, exploiting the mutual agreement between multi-modal phase prediction streams to adapt the model to a given test video in a self-supervised manner, enhancing the reliability under test-time distribution shifts. SPA is a lightweight adaptation framework, allowing hospitals to rapidly customize phase recognition models by defining phases in natural language text, annotating a few images with the phase labels, and providing a task graph defining phase transitions. The experimental results show that the SPA framework achieves state-of-the-art performance in few-shot surgical phase recognition across multiple institutions and procedures, even outperforming full-shot models with 32-shot labeled data. Code is available at https://github.com/CAMMA-public/SPA

Paperid: 856, https://arxiv.org/pdf/2506.19513.pdf GitHub

Abstract:
Despite the remarkable multimodal capabilities of Large Vision-Language Models (LVLMs), discrepancies often occur between visual inputs and textual outputs--a phenomenon we term visual hallucination. This critical reliability gap poses substantial risks in safety-critical Artificial Intelligence (AI) applications, necessitating a comprehensive evaluation benchmark and effective detection methods. Firstly, we observe that existing visual-centric hallucination benchmarks mainly assess LVLMs from a perception perspective, overlooking hallucinations arising from advanced reasoning capabilities. We develop the Perception-Reasoning Evaluation Hallucination (PRE-HAL) dataset, which enables the systematic evaluation of both perception and reasoning capabilities of LVLMs across multiple visual semantics, such as instances, scenes, and relations. Comprehensive evaluation with this new benchmark exposed more visual vulnerabilities, particularly in the more challenging task of relation reasoning. To address this issue, we propose, to the best of our knowledge, the first Dempster-Shafer theory (DST)-based visual hallucination detection method for LVLMs through uncertainty estimation. This method aims to efficiently capture the degree of conflict in high-level features at the model inference phase. Specifically, our approach employs simple mass functions to mitigate the computational complexity of evidence combination on power sets. We conduct an extensive evaluation of state-of-the-art LVLMs, LLaVA-v1.5, mPLUG-Owl2 and mPLUG-Owl3, with the new PRE-HAL benchmark. Experimental results indicate that our method outperforms five baseline uncertainty metrics, achieving average AUROC improvements of 4%, 10%, and 7% across three LVLMs. Our code is available at https://github.com/HT86159/Evidential-Conflict.

Paperid: 857, https://arxiv.org/pdf/2506.19320.pdf GitHub

Abstract:
Traditional fundus image analysis models focus on single-modal tasks, ignoring fundus modality complementarity, which limits their versatility. Recently, retinal foundation models have emerged, but most still remain modality-specific. Integrating multiple fundus imaging modalities into a single foundation model is valuable. However, in dynamic environments, data from different modalities often arrive incrementally, necessitating continual pre-training. To address this, we propose RetCoP, the first continual vision-language pre-training framework in the fundus domain, which incrementally integrates image and text features from different imaging modalities into a single unified foundation model. To mitigate catastrophic forgetting in continual pre-training, we introduce a rehearsal strategy utilizing representative image-text pairs and an off-diagonal information distillation approach. The former allows the model to revisit knowledge from previous stages, while the latter explicitly preserves the alignment between image and text representations. Experiments show that RetCoP outperforms all the compared methods, achieving the best generalization and lowest forgetting rate. The code can be found at https://github.com/Yuang-Yao/RetCoP.

Paperid: 858, https://arxiv.org/pdf/2506.18544.pdf GitHub

Abstract:
Recently, detecting logical anomalies is becoming a more challenging task compared to detecting structural ones. Existing encoder decoder based methods typically compress inputs into low-dimensional bottlenecks on the assumption that the compression process can effectively suppress the transmission of logical anomalies to the decoder. However, logical anomalies present a particular difficulty because, while their local features often resemble normal semantics, their global semantics deviate significantly from normal patterns. Thanks to the generalisation capabilities inherent in neural networks, these abnormal semantic features can propagate through low-dimensional bottlenecks. This ultimately allows the decoder to reconstruct anomalous images with misleading fidelity. To tackle the above challenge, we propose a novel normality prior guided multi-semantic fusion network for unsupervised anomaly detection. Instead of feeding the compressed bottlenecks to the decoder directly, we introduce the multi-semantic features of normal samples into the reconstruction process. To this end, we first extract abstract global semantics of normal cases by a pre-trained vision-language network, then the learnable semantic codebooks are constructed to store representative feature vectors of normal samples by vector quantisation. Finally, the above multi-semantic features are fused and employed as input to the decoder to guide the reconstruction of anomalies to approximate normality. Extensive experiments are conducted to validate the effectiveness of our proposed method, and it achieves the SOTA performance on the MVTec LOCO AD dataset with improvements of 5.7% in pixel-sPRO and 2.6% in image-AUROC. The source code is available at https://github.com/Xmh-L/NPGMF.

Paperid: 859, https://arxiv.org/pdf/2506.18378.pdf GitHub

Abstract:
Modern Vision-Language Models (VLMs) exhibit unprecedented capabilities in cross-modal semantic understanding between visual and textual modalities. Given the intrinsic need for multi-modal integration in clinical applications, VLMs have emerged as a promising solution for a wide range of medical image analysis tasks. However, adapting general-purpose VLMs to medical domain poses numerous challenges, such as large domain gaps, complicated pathological variations, and diversity and uniqueness of different tasks. The central purpose of this review is to systematically summarize recent advances in adapting VLMs for medical image analysis, analyzing current challenges, and recommending promising yet urgent directions for further investigations. We begin by introducing core learning strategies for medical VLMs, including pretraining, fine-tuning, and prompt learning. We then categorize five major VLM adaptation strategies for medical image analysis. These strategies are further analyzed across eleven medical imaging tasks to illustrate their current practical implementations. Furthermore, we analyze key challenges that impede the effective adaptation of VLMs to clinical applications and discuss potential directions for future research. We also provide an open-access repository of related literature to facilitate further research, available at https://github.com/haonenglin/Awesome-VLM-for-MIA. It is anticipated that this article can help researchers who are interested in harnessing VLMs in medical image analysis tasks have a better understanding on their capabilities and limitations, as well as current technical barriers, to promote their innovative, robust, and safe application in clinical practice.

Paperid: 860, https://arxiv.org/pdf/2506.18372.pdf GitHub

Abstract:
We introduce OpenEvents V1a large-scale benchmark dataset designed to advance event-centric vision-language understanding. Unlike conventional image captioning and retrieval datasets that focus on surface-level descriptions, OpenEvents V1 dataset emphasizes contextual and temporal grounding through three primary tasks: (1) generating rich, event-aware image captions, (2) retrieving event-relevant news articles from image queries, and (3) retrieving event-relevant images from narrative-style textual queries. The dataset comprises over 200,000 news articles and 400,000 associated images sourced from CNN and The Guardian, spanning diverse domains and time periods. We provide extensive baseline results and standardized evaluation protocols for all tasks. OpenEvents V1 establishes a robust foundation for developing multimodal AI systems capable of deep reasoning over complex real-world events. The dataset is publicly available at https://ltnghia.github.io/eventa/openevents-v1.

Paperid: 861, https://arxiv.org/pdf/2506.17629.pdf GitHub

Abstract:
Embodied Visual Reasoning (EVR) seeks to follow complex, free-form instructions based on egocentric video, enabling semantic understanding and spatiotemporal reasoning in dynamic environments. Despite its promising potential, EVR encounters significant challenges stemming from the diversity of complex instructions and the intricate spatiotemporal dynamics in long-term egocentric videos. Prior solutions either employ Large Language Models (LLMs) over static video captions, which often omit critical visual details, or rely on end-to-end Vision-Language Models (VLMs) that struggle with stepwise compositional reasoning. Consider the complementary strengths of LLMs in reasoning and VLMs in perception, we propose CLiViS. It is a novel training-free framework that leverages LLMs for high-level task planning and orchestrates VLM-driven open-world visual perception to iteratively update the scene context. Building on this synergy, the core of CLiViS is a dynamic Cognitive Map that evolves throughout the reasoning process. This map constructs a structured representation of the embodied scene, bridging low-level perception and high-level reasoning. Extensive experiments across multiple benchmarks demonstrate the effectiveness and generality of CLiViS, especially in handling long-term visual dependencies. Code is available at https://github.com/Teacher-Tom/CLiViS.

Paperid: 862, https://arxiv.org/pdf/2506.17590.pdf GitHub

Abstract:
Understanding the short-term motion of vulnerable road users (VRUs) like pedestrians and cyclists is critical for safe autonomous driving, especially in urban scenarios with ambiguous or high-risk behaviors. While vision-language models (VLMs) have enabled open-vocabulary perception, their utility for fine-grained intent reasoning remains underexplored. Notably, no existing benchmark evaluates multi-class intent prediction in safety-critical situations, To address this gap, we introduce DRAMA-X, a fine-grained benchmark constructed from the DRAMA dataset via an automated annotation pipeline. DRAMA-X contains 5,686 accident-prone frames labeled with object bounding boxes, a nine-class directional intent taxonomy, binary risk scores, expert-generated action suggestions for the ego vehicle, and descriptive motion summaries. These annotations enable a structured evaluation of four interrelated tasks central to autonomous decision-making: object detection, intent prediction, risk assessment, and action suggestion. As a reference baseline, we propose SGG-Intent, a lightweight, training-free framework that mirrors the ego vehicle's reasoning pipeline. It sequentially generates a scene graph from visual input using VLM-backed detectors, infers intent, assesses risk, and recommends an action using a compositional reasoning stage powered by a large language model. We evaluate a range of recent VLMs, comparing performance across all four DRAMA-X tasks. Our experiments demonstrate that scene-graph-based reasoning enhances intent prediction and risk assessment, especially when contextual cues are explicitly modeled.

Paperid: 863, https://arxiv.org/pdf/2506.17503.pdf GitHub

Abstract:
Medical vision-language models (VLMs) have demonstrated unprecedented transfer capabilities and are being increasingly adopted for data-efficient image classification. Despite its growing popularity, its reliability aspect remains largely unexplored. This work explores the split conformal prediction (SCP) framework to provide trustworthiness guarantees when transferring such models based on a small labeled calibration set. Despite its potential, the generalist nature of the VLMs' pre-training could negatively affect the properties of the predicted conformal sets for specific tasks. While common practice in transfer learning for discriminative purposes involves an adaptation stage, we observe that deploying such a solution for conformal purposes is suboptimal since adapting the model using the available calibration data breaks the rigid exchangeability assumptions for test data in SCP. To address this issue, we propose transductive split conformal adaptation (SCA-T), a novel pipeline for transfer learning on conformal scenarios, which performs an unsupervised transductive adaptation jointly on calibration and test data. We present comprehensive experiments utilizing medical VLMs across various image modalities, transfer tasks, and non-conformity scores. Our framework offers consistent gains in efficiency and conditional coverage compared to SCP, maintaining the same empirical guarantees.

Paperid: 864, https://arxiv.org/pdf/2506.17500.pdf GitHub

Abstract:
Vision-language models (VLMs) are gaining attention in medical image analysis. These are pre-trained on large, heterogeneous data sources, yielding rich and transferable representations. Notably, the combination of modality-specialized VLMs with few-shot adaptation has provided fruitful results, enabling the efficient deployment of high-performing solutions. However, previous works on this topic make strong assumptions about the distribution of adaptation data, which are unrealistic in the medical domain. First, prior art assumes access to a balanced support set, a condition that breaks the natural imbalance in disease prevalence found in real-world scenarios. Second, these works typically assume the presence of an additional validation set to fix critical hyper-parameters, which is highly data-inefficient. This work challenges these favorable deployment scenarios and introduces a realistic, imbalanced, validation-free adaptation setting. Our extensive benchmark across various modalities and downstream tasks demonstrates that current methods systematically compromise their performance when operating under realistic conditions, occasionally even performing worse than zero-shot inference. Also, we introduce a training-free linear probe that adaptively blends visual and textual supervision. Detailed studies demonstrate that the proposed solver is a strong, efficient baseline, enabling robust adaptation in challenging scenarios.

Paperid: 865, https://arxiv.org/pdf/2506.16805.pdf GitHub

Abstract:
Humans exhibit a remarkable ability to recognize co-visibility-the 3D regions simultaneously visible in multiple images-even when these images are sparsely distributed across a complex scene. This ability is foundational to 3D vision, robotic perception, and relies not only on low-level feature matching but also on high-level spatial reasoning and cognitive integration. Yet, it remains unclear whether current vision models can replicate this human-level proficiency. In this work, we introduce the Co-VisiON benchmark, designed to evaluate human-inspired co-visibility reasoning across more than 1,000 sparse-view indoor scenarios. Our results show that while co-visibility is often approached as a low-level feature-matching task, it remains challenging for existing vision models under sparse conditions. Notably, a proprietary vision-language model surpasses all vision-only baselines, but all models fall significantly short of human performance. This gap underscores the limitations of current architectures and motivates the need for models that integrate spatial and semantic information in a human-like manner. Inspired by human visual cognition, we propose a novel multi-view baseline, Covis, which achieves top performance among pure vision models and narrows the gap to the proprietary VLM. We hope our benchmark and findings will spur further advancements in developing vision models capable of robust, cognitively inspired reasoning in challenging, sparse environments. Our dataset and source code can be found at https://ai4ce.github.io/CoVISION.

Paperid: 866, https://arxiv.org/pdf/2506.16703.pdf GitHub

Abstract:
The increasingly complex and diverse planetary exploration environment requires more adaptable and flexible rover navigation strategy. In this study, we propose a VLM-empowered multi-mode system to achieve efficient while safe autonomous navigation for planetary rovers. Vision-Language Model (VLM) is used to parse scene information by image inputs to achieve a human-level understanding of terrain complexity. Based on the complexity classification, the system switches to the most suitable navigation mode, composing of perception, mapping and planning modules designed for different terrain types, to traverse the terrain ahead before reaching the next waypoint. By integrating the local navigation system with a map server and a global waypoint generation module, the rover is equipped to handle long-distance navigation tasks in complex scenarios. The navigation system is evaluated in various simulation environments. Compared to the single-mode conservative navigation method, our multi-mode system is able to bootstrap the time and energy efficiency in a long-distance traversal with varied type of obstacles, enhancing efficiency by 79.5%, while maintaining its avoidance capabilities against terrain hazards to guarantee rover safety. More system information is shown at https://chengsn1234.github.io/multi-mode-planetary-navigation/.

Paperid: 867, https://arxiv.org/pdf/2506.16652.pdf GitHub

Abstract:
Natural language instructions for robotic manipulation tasks often exhibit ambiguity and vagueness. For instance, the instruction "Hang a mug on the mug tree" may involve multiple valid actions if there are several mugs and branches to choose from. Existing language-conditioned policies typically rely on end-to-end models that jointly handle high-level semantic understanding and low-level action generation, which can result in suboptimal performance due to their lack of modularity and interpretability. To address these challenges, we introduce a novel robotic manipulation framework that can accomplish tasks specified by potentially ambiguous natural language. This framework employs a Vision-Language Model (VLM) to interpret abstract concepts in natural language instructions and generates task-specific code - an interpretable and executable intermediate representation. The generated code interfaces with the perception module to produce 3D attention maps that highlight task-relevant regions by integrating spatial and semantic information, effectively resolving ambiguities in instructions. Through extensive experiments, we identify key limitations of current imitation learning methods, such as poor adaptation to language and environmental variations. We show that our approach excels across challenging manipulation tasks involving language ambiguity, contact-rich manipulation, and multi-object interactions.

Paperid: 868, https://arxiv.org/pdf/2506.16012.pdf GitHub

Abstract:
Developing embodied agents capable of performing complex interactive tasks in real-world scenarios remains a fundamental challenge in embodied AI. Although recent advances in simulation platforms have greatly enhanced task diversity to train embodied Vision Language Models (VLMs), most platforms rely on simplified robot morphologies and bypass the stochastic nature of low-level execution, which limits their transferability to real-world robots. To address these issues, we present a physics-based simulation platform DualTHOR for complex dual-arm humanoid robots, built upon an extended version of AI2-THOR. Our simulator includes real-world robot assets, a task suite for dual-arm collaboration, and inverse kinematics solvers for humanoid robots. We also introduce a contingency mechanism that incorporates potential failures through physics-based low-level execution, bridging the gap to real-world scenarios. Our simulator enables a more comprehensive evaluation of the robustness and generalization of VLMs in household environments. Extensive evaluations reveal that current VLMs struggle with dual-arm coordination and exhibit limited robustness in realistic environments with contingencies, highlighting the importance of using our simulator to develop more capable VLMs for embodied tasks. The code is available at https://github.com/ds199895/DualTHOR.git.

Paperid: 869, https://arxiv.org/pdf/2506.15681.pdf GitHub

Abstract:
Recent advancements in vision-language models (VLMs) have leveraged large language models (LLMs) to achieve performance on par with closed-source systems like GPT-4V. However, deploying these models in real-world scenarios, particularly on resource-constrained devices, remains challenging due to their substantial computational demands. This has spurred interest in distilling knowledge from large VLMs into smaller, more efficient counterparts. A key challenge arises here from the diversity of VLM architectures, which are built on different LLMs and employ varying token types-differing in vocabulary size, token splits, and token index ordering. To address this challenge of limitation to a specific VLM type, we present Generation after Recalibration (GenRecal), a novel, general-purpose distillation framework for VLMs. GenRecal incorporates a Recalibrator that aligns and adapts feature representations between heterogeneous VLMs, enabling effective knowledge transfer across different types of VLMs. Through extensive experiments on multiple challenging benchmarks, we demonstrate that GenRecal significantly improves baseline performances, eventually outperforming large-scale open- and closed-source VLMs.

Paperid: 870, https://arxiv.org/pdf/2506.15583.pdf GitHub

Abstract:
Vision-Language Models (VLMs) generate discourse-level, multi-sentence visual descriptions, challenging text scene graph parsers built for single-sentence caption-to-graph mapping. Current approaches typically merge sentence-level parsing outputs for discourse input, often missing phenomena like cross-sentence coreference, resulting in fragmented graphs and degraded downstream VLM task performance. We introduce a new task, Discourse-level text Scene Graph parsing (DiscoSG), and release DiscoSG-DS, a dataset of 400 expert-annotated and 8,430 synthesised multi-sentence caption-graph pairs. Each caption averages 9 sentences, and each graph contains at least 3 times more triples than those in existing datasets. Fine-tuning GPT-4o on DiscoSG-DS yields over 40% higher SPICE than the strongest sentence-merging baseline. However, its high inference cost and licensing restrict open-source use, and smaller fine-tuned open-source models (e.g., Flan-T5) perform poorly on dense graph generation. To bridge this gap, we propose DiscoSG-Refiner, which drafts a base graph using a seed parser and iteratively refines it with a second model, improving robustness for complex graph generation. Using two small fine-tuned Flan-T5-Base models, DiscoSG-Refiner improves SPICE by approximately 30% over the baseline while achieving 86 times faster inference than GPT-4o. It also delivers consistent gains on downstream VLM tasks, including discourse-level caption evaluation and hallucination detection, outperforming alternative parsers. Code and data are available at https://github.com/ShaoqLin/DiscoSG .

Paperid: 871, https://arxiv.org/pdf/2506.15318.pdf GitHub GitHub

Abstract:
Pathology image classification plays a crucial role in accurate medical diagnosis and treatment planning. Training high-performance models for this task typically requires large-scale annotated datasets, which are both expensive and time-consuming to acquire. Active Learning (AL) offers a solution by iteratively selecting the most informative samples for annotation, thereby reducing the labeling effort. However, most AL methods are designed under the assumption of a closed-set scenario, where all the unannotated images belong to target classes. In real-world clinical environments, the unlabeled pool often contains a substantial amount of Out-Of-Distribution (OOD) data, leading to low efficiency of annotation in traditional AL methods. Furthermore, most existing AL methods start with random selection in the first query round, leading to a significant waste of labeling costs in open-set scenarios. To address these challenges, we propose OpenPath, a novel open-set active learning approach for pathological image classification leveraging a pre-trained Vision-Language Model (VLM). In the first query, we propose task-specific prompts that combine target and relevant non-target class prompts to effectively select In-Distribution (ID) and informative samples from the unlabeled pool. In subsequent queries, Diverse Informative ID Sampling (DIS) that includes Prototype-based ID candidate Selection (PIS) and Entropy-Guided Stochastic Sampling (EGSS) is proposed to ensure both purity and informativeness in a query, avoiding the selection of OOD samples. Experiments on two public pathology image datasets show that OpenPath significantly enhances the model's performance due to its high purity of selected samples, and outperforms several state-of-the-art open-set AL methods. The code is available at \href{https://github.com/HiLab-git/OpenPath}{https://github.com/HiLab-git/OpenPath}..

Paperid: 872, https://arxiv.org/pdf/2506.14808.pdf GitHub

Abstract:
Vision language models (VLMs) respond to user-crafted text prompts and visual inputs, and are applied to numerous real-world problems. VLMs integrate visual modalities with large language models (LLMs), which are well known to be prompt-sensitive. Hence, it is crucial to determine whether VLMs inherit this instability to varying prompts. We therefore investigate which prompt variations VLMs are most sensitive to and which VLMs are most agnostic to prompt variations. To this end, we introduce PARC (Prompt Analysis via Reliability and Calibration), a VLM prompt sensitivity analysis framework built on three pillars: (1) plausible prompt variations in both the language and vision domain, (2) a novel model reliability score with built-in guarantees, and (3) a calibration step that enables dataset- and prompt-spanning prompt variation analysis. Regarding prompt variations, PARC's evaluation shows that VLMs mirror LLM language prompt sensitivity in the vision domain, and most destructive variations change the expected answer. Regarding models, outstandingly robust VLMs among 22 evaluated models come from the InternVL2 family. We further find indications that prompt sensitivity is linked to training data. The code will be at https://github.com/NVlabs/PARC.

Paperid: 873, https://arxiv.org/pdf/2506.14727.pdf GitHub

Abstract:
Assistive teleoperation, where control is shared between a human and a robot, enables efficient and intuitive human-robot collaboration in diverse and unstructured environments. A central challenge in real-world assistive teleoperation is for the robot to infer a wide range of human intentions from user control inputs and to assist users with correct actions. Existing methods are either confined to simple, predefined scenarios or restricted to task-specific data distributions at training, limiting their support for real-world assistance. We introduce Casper, an assistive teleoperation system that leverages commonsense knowledge embedded in pre-trained visual language models (VLMs) for real-time intent inference and flexible skill execution. Casper incorporates an open-world perception module for a generalized understanding of novel objects and scenes, a VLM-powered intent inference mechanism that leverages commonsense reasoning to interpret snippets of teleoperated user input, and a skill library that expands the scope of prior assistive teleoperation systems to support diverse, long-horizon mobile manipulation tasks. Extensive empirical evaluation, including human studies and system ablations, demonstrates that Casper improves task performance, reduces human cognitive load, and achieves higher user satisfaction than direct teleoperation and assistive teleoperation baselines. More information is available at https://ut-austin-rpl.github.io/casper/

Paperid: 874, https://arxiv.org/pdf/2506.14629.pdf GitHub

Abstract:
Mosquito-borne diseases pose a major global health risk, requiring early detection and proactive control of breeding sites to prevent outbreaks. In this paper, we present VisText-Mosquito, a multimodal dataset that integrates visual and textual data to support automated detection, segmentation, and reasoning for mosquito breeding site analysis. The dataset includes 1,828 annotated images for object detection, 142 images for water surface segmentation, and natural language reasoning texts linked to each image. The YOLOv9s model achieves the highest precision of 0.92926 and mAP@50 of 0.92891 for object detection, while YOLOv11n-Seg reaches a segmentation precision of 0.91587 and mAP@50 of 0.79795. For reasoning generation, we tested a range of large vision-language models (LVLMs) in both zero-shot and few-shot settings. Our fine-tuned Mosquito-LLaMA3-8B model achieved the best results, with a final loss of 0.0028, a BLEU score of 54.7, BERTScore of 0.91, and ROUGE-L of 0.85. This dataset and model framework emphasize the theme "Prevention is Better than Cure", showcasing how AI-based detection can proactively address mosquito-borne disease risks. The dataset and implementation code are publicly available at GitHub: https://github.com/adnanul-islam-jisun/VisText-Mosquito

Paperid: 875, https://arxiv.org/pdf/2506.14507.pdf GitHub

Abstract:
Foundation models have revolutionized robotics by providing rich semantic representations without task-specific training. While many approaches integrate pretrained vision-language models (VLMs) with specialized navigation architectures, the fundamental question remains: can these pretrained embeddings alone successfully guide navigation without additional fine-tuning or specialized modules? We present a minimalist framework that decouples this question by training a behavior cloning policy directly on frozen vision-language embeddings from demonstrations collected by a privileged expert. Our approach achieves a 74% success rate in navigation to language-specified targets, compared to 100% for the state-aware expert, though requiring 3.2 times more steps on average. This performance gap reveals that pretrained embeddings effectively support basic language grounding but struggle with long-horizon planning and spatial reasoning. By providing this empirical baseline, we highlight both the capabilities and limitations of using foundation models as drop-in representations for embodied tasks, offering critical insights for robotics researchers facing practical design tradeoffs between system complexity and performance in resource-constrained scenarios. Our code is available at https://github.com/oadamharoon/text2nav

Paperid: 876, https://arxiv.org/pdf/2506.14317.pdf GitHub

Abstract:
Dexterous grasping in cluttered scenes presents significant challenges due to diverse object geometries, occlusions, and potential collisions. Existing methods primarily focus on single-object grasping or grasp-pose prediction without interaction, which are insufficient for complex, cluttered scenes. Recent vision-language-action models offer a potential solution but require extensive real-world demonstrations, making them costly and difficult to scale. To address these limitations, we revisit the sim-to-real transfer pipeline and develop key techniques that enable zero-shot deployment in reality while maintaining robust generalization. We propose ClutterDexGrasp, a two-stage teacher-student framework for closed-loop target-oriented dexterous grasping in cluttered scenes. The framework features a teacher policy trained in simulation using clutter density curriculum learning, incorporating both a geometry and spatially-embedded scene representation and a novel comprehensive safety curriculum, enabling general, dynamic, and safe grasping behaviors. Through imitation learning, we distill the teacher's knowledge into a student 3D diffusion policy (DP3) that operates on partial point cloud observations. To the best of our knowledge, this represents the first zero-shot sim-to-real closed-loop system for target-oriented dexterous grasping in cluttered scenes, demonstrating robust performance across diverse objects and layouts. More details and videos are available at https://clutterdexgrasp.github.io/.

Paperid: 877, https://arxiv.org/pdf/2506.14136.pdf GitHub

Abstract:
In this paper, we construct two research objectives: i) explore the learned embedding space of BiomedCLIP, an open-source large vision language model, to analyse meaningful class separations, and ii) quantify the limitations of BiomedCLIP when applied to a highly imbalanced, out-of-distribution multi-label medical dataset. We experiment on IU-xray dataset, which exhibits the aforementioned criteria, and evaluate BiomedCLIP in classifying images (radiographs) in three contexts: zero-shot inference, full finetuning, and linear probing. The results show that the model under zero-shot settings over-predicts all labels, leading to poor precision and inter-class separability. Full fine-tuning improves classification of distinct diseases, while linear probing detects overlapping features. We demonstrate visual understanding of the model using Grad-CAM heatmaps and compare with 15 annotations by a radiologist. We highlight the need for careful adaptations of the models to foster reliability and applicability in a real-world setting. The code for the experiments in this work is available and maintained on GitHub.

Paperid: 878, https://arxiv.org/pdf/2506.13759.pdf GitHub

Abstract:
In this work, we provide a systematic survey of Discrete Diffusion Language Models (dLLMs) and Discrete Diffusion Multimodal Language Models (dMLLMs). Unlike autoregressive (AR) models, dLLMs and dMLLMs adopt a multi-token, parallel decoding paradigm using full attention and a denoising-based generation strategy. This paradigm naturally enables parallel generation, fine-grained output control, and dynamic perception. These capabilities are previously difficult to achieve with AR models. A growing number of industrial-scale proprietary d(M)LLMs, as well as a large number of open-source academic d(M)LLMs, have demonstrated performance comparable to their autoregressive counterparts, while achieving up to 10$\times$ acceleration in inference speed. These developments position discrete diffusion models as a promising alternative to intelligence based on the traditional autoregressive approach. In this work, we present a comprehensive overview of the research in the dLLM and dMLLM domains. We trace the historical development of dLLMs and dMLLMs, formalize the underlying mathematical frameworks, list commonly-used modeling methods, and categorize representative models. We further analyze key techniques for training, inference, quantization. We also discuss the trustworthy issues and summarize emerging applications across language, vision-language, and biological domains and etc.. We conclude by discussing future directions for research and deployment. Relative papers are collected in https://github.com/LiQiiiii/Awesome-Discrete-Diffusion-LLM_MLLM

Paperid: 879, https://arxiv.org/pdf/2506.13751.pdf GitHub

Abstract:
Vision-language-action (VLA) models have demonstrated strong semantic understanding and zero-shot generalization, yet most existing systems assume an accurate low-level controller with hand-crafted action "vocabulary" such as end-effector pose or root velocity. This assumption confines prior work to quasi-static tasks and precludes the agile, whole-body behaviors required by humanoid whole-body control (WBC) tasks. To capture this gap in the literature, we start by introducing the first sim-to-real-ready, vision-language, closed-loop benchmark for humanoid WBC, comprising over 150 tasks from 10 categories. We then propose LeVERB: Latent Vision-Language-Encoded Robot Behavior, a hierarchical latent instruction-following framework for humanoid vision-language WBC, the first of its kind. At the top level, a vision-language policy learns a latent action vocabulary from synthetically rendered kinematic demonstrations; at the low level, a reinforcement-learned WBC policy consumes these latent verbs to generate dynamics-level commands. In our benchmark, LeVERB can zero-shot attain a 80% success rate on simple visual navigation tasks, and 58.5% success rate overall, outperforming naive hierarchical whole-body VLA implementation by 7.8 times.

Paperid: 880, https://arxiv.org/pdf/2506.13725.pdf GitHub

Abstract:
In recent years, Vision-Language-Action (VLA) models have become a vital research direction in robotics due to their impressive multimodal understanding and generalization capabilities. Despite the progress, their practical deployment is severely constrained by inference speed bottlenecks, particularly in high-frequency and dexterous manipulation tasks. While recent studies have explored Jacobi decoding as a more efficient alternative to traditional autoregressive decoding, its practical benefits are marginal due to the lengthy iterations. To address it, we introduce consistency distillation training to predict multiple correct action tokens in each iteration, thereby achieving acceleration. Besides, we design mixed-label supervision to mitigate the error accumulation during distillation. Although distillation brings acceptable speedup, we identify that certain inefficient iterations remain a critical bottleneck. To tackle this, we propose an early-exit decoding strategy that moderately relaxes convergence conditions, which further improves average inference efficiency. Experimental results show that the proposed method achieves more than 4 times inference acceleration across different baselines while maintaining high task success rates in both simulated and real-world robot tasks. These experiments validate that our approach provides an efficient and general paradigm for accelerating multimodal decision-making in robotics. Our project page is available at https://irpn-eai.github.io/CEED-VLA/.

Paperid: 881, https://arxiv.org/pdf/2506.13638.pdf GitHub

Abstract:
Model editing aims to efficiently update a pre-trained model's knowledge without the need for time-consuming full retraining. While existing pioneering editing methods achieve promising results, they primarily focus on editing single-modal language models (LLMs). However, for vision-language models (VLMs), which involve multiple modalities, the role and impact of each modality on editing performance remain largely unexplored. To address this gap, we explore the impact of textual and visual modalities on model editing and find that: (1) textual and visual representations reach peak sensitivity at different layers, reflecting their varying importance; and (2) editing both modalities can efficiently update knowledge, but this comes at the cost of compromising the model's original capabilities. Based on our findings, we propose DualEdit, an editor that modifies both textual and visual modalities at their respective key layers. Additionally, we introduce a gating module within the more sensitive textual modality, allowing DualEdit to efficiently update new knowledge while preserving the model's original information. We evaluate DualEdit across multiple VLM backbones and benchmark datasets, demonstrating its superiority over state-of-the-art VLM editing baselines as well as adapted LLM editing methods on different evaluation metrics. Codes are available at https://github.com/zhiyiscs/DualEdit

Paperid: 882, https://arxiv.org/pdf/2506.13051.pdf GitHub

Abstract:
Evaluating foundation models for crystallographic reasoning requires benchmarks that isolate generalization behavior while enforcing physical constraints. This work introduces a multiscale multicrystal dataset with two physically grounded evaluation protocols to stress-test multimodal generative models. The Spatial-Exclusion benchmark withholds all supercells of a given radius from a diverse dataset, enabling controlled assessments of spatial interpolation and extrapolation. The Compositional-Exclusion benchmark omits all samples of a specific chemical composition, probing generalization across stoichiometries. Nine vision--language foundation models are prompted with crystallographic images and textual context to generate structural annotations. Responses are evaluated via (i) relative errors in lattice parameters and density, (ii) a physics-consistency index penalizing volumetric violations, and (iii) a hallucination score capturing geometric outliers and invalid space-group predictions. These benchmarks establish a reproducible, physically informed framework for assessing generalization, consistency, and reliability in large-scale multimodal models. Dataset and code are available at https://github.com/KurbanIntelligenceLab/StressTestingMMFMinCR.

Paperid: 883, https://arxiv.org/pdf/2506.13045.pdf GitHub

Abstract:
The rapid advancement of generative models has empowered modern AI systems to comprehend and produce highly sophisticated content, even achieving human-level performance in specific domains. However, these models are fundamentally constrained by \emph{catastrophic forgetting}, \ie~a persistent challenge where models experience performance degradation on previously learned tasks when adapting to new tasks. To address this practical limitation, numerous approaches have been proposed to enhance the adaptability and scalability of generative AI in real-world applications. In this work, we present a comprehensive survey of continual learning methods for mainstream generative AI models, encompassing large language models, multimodal large language models, vision-language-action models, and diffusion models. Drawing inspiration from the memory mechanisms of the human brain, we systematically categorize these approaches into three paradigms: architecture-based, regularization-based, and replay-based methods, while elucidating their underlying methodologies and motivations. We further analyze continual learning setups for different generative models, including training objectives, benchmarks, and core backbones, thereby providing deeper insights into the field. The project page of this paper is available at https://github.com/Ghy0501/Awesome-Continual-Learning-in-Generative-Models.

Paperid: 884, https://arxiv.org/pdf/2506.12776.pdf GitHub

Abstract:
Vision-Language Models (VLMs) face significant challenges when dealing with the diverse resolutions and aspect ratios of real-world images, as most existing models rely on fixed, low-resolution inputs. While recent studies have explored integrating native resolution visual encoding to improve model performance, such efforts remain fragmented and lack a systematic framework within the open-source community. Moreover, existing benchmarks fall short in evaluating VLMs under varied visual conditions, often neglecting resolution as a critical factor. To address the "Resolution Dilemma" stemming from both model design and benchmark limitations, we introduce RC-Bench, a novel benchmark specifically designed to systematically evaluate VLM capabilities under extreme visual conditions, with an emphasis on resolution and aspect ratio variations. In conjunction, we propose NativeRes-LLaVA, an open-source training framework that empowers VLMs to effectively process images at their native resolutions and aspect ratios. Based on RC-Bench and NativeRes-LLaVA, we conduct comprehensive experiments on existing visual encoding strategies. The results show that Native Resolution Visual Encoding significantly improves the performance of VLMs on RC-Bench as well as other resolution-centric benchmarks. Code is available at https://github.com/Niujunbo2002/NativeRes-LLaVA.

Paperid: 885, https://arxiv.org/pdf/2506.11989.pdf GitHub

Abstract:
Test-time scaling offers a promising way to improve the reasoning performance of vision-language large models (VLLMs) without additional training. In this paper, we explore a simple but effective approach for applying test-time scaling to radiology report generation. Specifically, we introduce a lightweight Thought Graph Traversal (TGT) framework that guides the model to reason through organ-specific findings in a medically coherent order. This framework integrates structured medical priors into the prompt, enabling deeper and more logical analysis with no changes to the underlying model. To further enhance reasoning depth, we apply a reasoning budget forcing strategy that adjusts the model's inference depth at test time by dynamically extending its generation process. This simple yet powerful combination allows a frozen radiology VLLM to self-correct and generate more accurate, consistent chest X-ray reports. Our method outperforms baseline prompting approaches on standard benchmarks, and also reveals dataset biases through traceable reasoning paths. Code and prompts are open-sourced for reproducibility at https://github.com/glerium/Thought-Graph-Traversal.

Paperid: 886, https://arxiv.org/pdf/2506.11526.pdf GitHub

Abstract:
For autonomous vehicles, safe navigation in complex environments depends on handling a broad range of diverse and rare driving scenarios. Simulation- and scenario-based testing have emerged as key approaches to development and validation of autonomous driving systems. Traditional scenario generation relies on rule-based systems, knowledge-driven models, and data-driven synthesis, often producing limited diversity and unrealistic safety-critical cases. With the emergence of foundation models, which represent a new generation of pre-trained, general-purpose AI models, developers can process heterogeneous inputs (e.g., natural language, sensor data, HD maps, and control actions), enabling the synthesis and interpretation of complex driving scenarios. In this paper, we conduct a survey about the application of foundation models for scenario generation and scenario analysis in autonomous driving (as of May 2025). Our survey presents a unified taxonomy that includes large language models, vision-language models, multimodal large language models, diffusion models, and world models for the generation and analysis of autonomous driving scenarios. In addition, we review the methodologies, open-source datasets, simulation platforms, and benchmark challenges, and we examine the evaluation metrics tailored explicitly to scenario generation and analysis. Finally, the survey concludes by highlighting the open challenges and research questions, and outlining promising future research directions. All reviewed papers are listed in a continuously maintained repository, which contains supplementary materials and is available at https://github.com/TUM-AVS/FM-for-Scenario-Generation-Analysis.

Paperid: 887, https://arxiv.org/pdf/2506.11515.pdf GitHub

Abstract:
Two-Tower Vision--Language Models (VLMs) have demonstrated strong performance across various downstream VL tasks. While BridgeTower further enhances performance by building bridges between encoders, it \textit{(i)} suffers from ineffective layer-by-layer utilization of unimodal representations, \textit{(ii)} restricts the flexible exploitation of different levels of unimodal semantic knowledge, and \textit{(iii)} is limited to the evaluation on traditional low-resolution datasets only with the Two-Tower VLM architecture. In this work, we propose Manager, a lightweight, efficient and effective plugin that adaptively aggregates insights from different levels of pre-trained unimodal experts to facilitate more comprehensive VL alignment and fusion. First, under the Two-Tower VLM architecture, we introduce ManagerTower, a novel VLM that introduces the manager in each cross-modal layer. Whether with or without VL pre-training, ManagerTower outperforms previous strong baselines and achieves superior performance on 4 downstream VL tasks. Moreover, we extend our exploration to the latest Multimodal Large Language Model (MLLM) architecture. We demonstrate that LLaVA-OV-Manager significantly boosts the zero-shot performance of LLaVA-OV across different categories of capabilities, images, and resolutions on 20 downstream datasets, whether the multi-grid algorithm is enabled or not. In-depth analysis reveals that both our manager and the multi-grid algorithm can be viewed as a plugin that improves the visual representation by capturing more diverse visual details from two orthogonal perspectives (depth and width). Their synergy can mitigate the semantic ambiguity caused by the multi-grid algorithm and further improve performance. Code and models are available at https://github.com/LooperXX/ManagerTower.

Paperid: 888, https://arxiv.org/pdf/2506.11147.pdf GitHub

Abstract:
Medical Visual Question Answering (Med-VQA) holds significant potential for clinical decision support, yet existing efforts primarily focus on 2D imaging with limited task diversity. This paper presents 3D-RAD, a large-scale dataset designed to advance 3D Med-VQA using radiology CT scans. The 3D-RAD dataset encompasses six diverse VQA tasks: anomaly detection, image observation, medical computation, existence detection, static temporal diagnosis, and longitudinal temporal diagnosis. It supports both open- and closed-ended questions while introducing complex reasoning challenges, including computational tasks and multi-stage temporal analysis, to enable comprehensive benchmarking. Extensive evaluations demonstrate that existing vision-language models (VLMs), especially medical VLMs exhibit limited generalization, particularly in multi-temporal tasks, underscoring the challenges of real-world 3D diagnostic reasoning. To drive future advancements, we release a high-quality training set 3D-RAD-T of 136,195 expert-aligned samples, showing that fine-tuning on this dataset could significantly enhance model performance. Our dataset and code, aiming to catalyze multimodal medical AI research and establish a robust foundation for 3D medical visual understanding, are publicly available at https://github.com/Tang-xiaoxiao/M3D-RAD.

Paperid: 889, https://arxiv.org/pdf/2506.11031.pdf GitHub

Abstract:
As image generators produce increasingly realistic images, concerns about potential misuse continue to grow. Supervised detection relies on large, curated datasets and struggles to generalize across diverse generators. In this work, we investigate the use of pre-trained Vision-Language Models (VLMs) for zero-shot detection of AI-generated images. While off-the-shelf VLMs exhibit some task-specific reasoning and chain-of-thought prompting offers gains, we show that task-aligned prompting elicits more focused reasoning and significantly improves performance without fine-tuning. Specifically, prefixing the model's response with the phrase "Let's examine the style and the synthesis artifacts" -- a method we call zero-shot-s$^2$ -- boosts Macro F1 scores by 8%-29%. These gains are consistent for two widely used open-source models and across three recent, diverse datasets spanning human faces, objects, and animals with images generated by 16 different models -- demonstrating strong generalization. We further evaluate the approach across three additional model sizes and observe improvements in most dataset-model combinations -- suggesting robustness to model scale. Surprisingly, self-consistency, a behavior previously observed in language reasoning, where aggregating answers from diverse reasoning paths improves performance, also holds in this setting. Even here, zero-shot-s$^2$ scales better than chain-of-thought in most cases -- indicating that it elicits more useful diversity. Our findings show that task-aligned prompts elicit more focused reasoning and enhance latent capabilities in VLMs, like the detection of AI-generated images -- offering a simple, generalizable, and explainable alternative to supervised methods. Our code is publicly available on github: https://github.com/Zoher15/Zero-shot-s2.

Paperid: 890, https://arxiv.org/pdf/2506.10967.pdf GitHub GitHub GitHub

Abstract:
In multimodal large language models (MLLMs), the length of input visual tokens is often significantly greater than that of their textual counterparts, leading to a high inference cost. Many works aim to address this issue by removing redundant visual tokens. However, current approaches either rely on attention-based pruning, which retains numerous duplicate tokens, or use similarity-based pruning, overlooking the instruction relevance, consequently causing suboptimal performance. In this paper, we go beyond attention or similarity by proposing a novel visual token pruning method named CDPruner, which maximizes the conditional diversity of retained tokens. We first define the conditional similarity between visual tokens conditioned on the instruction, and then reformulate the token pruning problem with determinantal point process (DPP) to maximize the conditional diversity of the selected subset. The proposed CDPruner is training-free and model-agnostic, allowing easy application to various MLLMs. Extensive experiments across diverse MLLMs show that CDPruner establishes new state-of-the-art on various vision-language benchmarks. By maximizing conditional diversity through DPP, the selected subset better represents the input images while closely adhering to user instructions, thereby preserving strong performance even with high reduction ratios. When applied to LLaVA, CDPruner reduces FLOPs by 95\% and CUDA latency by 78\%, while maintaining 94\% of the original accuracy. Our code is available at https://github.com/Theia-4869/CDPruner.

Paperid: 891, https://arxiv.org/pdf/2506.10826.pdf GitHub

Abstract:
A fundamental requirement for real-world robotic deployment is the ability to understand and respond to natural language instructions. Existing language-conditioned manipulation tasks typically assume that instructions are perfectly aligned with the environment. This assumption limits robustness and generalization in realistic scenarios where instructions may be ambiguous, irrelevant, or infeasible. To address this problem, we introduce RAtional MAnipulation (RAMA), a new benchmark that challenges models with both unseen executable instructions and defective ones that should be rejected. In RAMA, we construct a dataset with over 14,000 samples, including diverse defective instructions spanning six dimensions: visual, physical, semantic, motion, safety, and out-of-context. We further propose the Rational Vision-Language-Action model (RationalVLA). It is a dual system for robotic arms that integrates the high-level vision-language model with the low-level manipulation policy by introducing learnable latent space embeddings. This design enables RationalVLA to reason over instructions, reject infeasible commands, and execute manipulation effectively. Experiments demonstrate that RationalVLA outperforms state-of-the-art baselines on RAMA by a 14.5% higher success rate and 0.94 average task length, while maintaining competitive performance on standard manipulation tasks. Real-world trials further validate its effectiveness and robustness in practical applications. Our project page is https://irpn-eai.github.io/RationalVLA.

Paperid: 892, https://arxiv.org/pdf/2506.10741.pdf GitHub

Abstract:
Generating aesthetic posters is more challenging than simple design images: it requires not only precise text rendering but also the seamless integration of abstract artistic content, striking layouts, and overall stylistic harmony. To address this, we propose PosterCraft, a unified framework that abandons prior modular pipelines and rigid, predefined layouts, allowing the model to freely explore coherent, visually compelling compositions. PosterCraft employs a carefully designed, cascaded workflow to optimize the generation of high-aesthetic posters: (i) large-scale text-rendering optimization on our newly introduced Text-Render-2M dataset; (ii) region-aware supervised fine-tuning on HQ-Poster100K; (iii) aesthetic-text-reinforcement learning via best-of-n preference optimization; and (iv) joint vision-language feedback refinement. Each stage is supported by a fully automated data-construction pipeline tailored to its specific needs, enabling robust training without complex architectural modifications. Evaluated on multiple experiments, PosterCraft significantly outperforms open-source baselines in rendering accuracy, layout coherence, and overall visual appeal-approaching the quality of SOTA commercial systems. Our code, models, and datasets can be found in the Project page: https://ephemeral182.github.io/PosterCraft

Paperid: 893, https://arxiv.org/pdf/2506.10730.pdf GitHub

Abstract:
Recently, the rapid advancements of vision-language models, such as CLIP, leads to significant progress in zero-/few-shot anomaly detection (ZFSAD) tasks. However, most existing CLIP-based ZFSAD methods commonly assume prior knowledge of categories and rely on carefully crafted prompts tailored to specific scenarios. While such meticulously designed text prompts effectively capture semantic information in the textual space, they fall short of distinguishing normal and anomalous instances within the joint embedding space. Moreover, these ZFSAD methods are predominantly explored in industrial scenarios, with few efforts conducted to medical tasks. To this end, we propose an innovative framework for ZFSAD tasks in medical domain, denoted as IQE-CLIP. We reveal that query embeddings, which incorporate both textual and instance-aware visual information, are better indicators for abnormalities. Specifically, we first introduce class-based prompting tokens and learnable prompting tokens for better adaptation of CLIP to the medical domain. Then, we design an instance-aware query module (IQM) to extract region-level contextual information from both text prompts and visual features, enabling the generation of query embeddings that are more sensitive to anomalies. Extensive experiments conducted on six medical datasets demonstrate that IQE-CLIP achieves state-of-the-art performance on both zero-shot and few-shot tasks. We release our code and data at https://github.com/hongh0/IQE-CLIP/.

Paperid: 894, https://arxiv.org/pdf/2506.10609.pdf GitHub

Abstract:
Scene text retrieval has made significant progress with the assistance of accurate text localization. However, existing approaches typically require costly bounding box annotations for training. Besides, they mostly adopt a customized retrieval strategy but struggle to unify various types of queries to meet diverse retrieval needs. To address these issues, we introduce Muti-query Scene Text retrieval with Attention Recycling (MSTAR), a box-free approach for scene text retrieval. It incorporates progressive vision embedding to dynamically capture the multi-grained representation of texts and harmonizes free-style text queries with style-aware instructions. Additionally, a multi-instance matching module is integrated to enhance vision-language alignment. Furthermore, we build the Multi-Query Text Retrieval (MQTR) dataset, the first benchmark designed to evaluate the multi-query scene text retrieval capability of models, comprising four query types and 16k images. Extensive experiments demonstrate the superiority of our method across seven public datasets and the MQTR dataset. Notably, MSTAR marginally surpasses the previous state-of-the-art model by 6.4% in MAP on Total-Text while eliminating box annotation costs. Moreover, on the MQTR benchmark, MSTAR significantly outperforms the previous models by an average of 8.5%. The code and datasets are available at https://github.com/yingift/MSTAR.

Paperid: 895, https://arxiv.org/pdf/2506.10416.pdf GitHub

Abstract:
What happens when we push audio-visual alignment to its absolute limits? To systematically investigate this question, we needed datasets with granular alignment quality annotations, but existing datasets treat alignment as binary, either synchronized or not. To address this limitation, we developed a comprehensive dataset featuring detailed alignment scores that reveal the hidden spectrum of audio-visual perceptual correspondence. Using these precise scores, we create "superaligned" representations by training exclusively on the most perfectly matched audio-visual pairs, then conduct our systematic investigation into how this extreme alignment transforms perceptual model behavior across retrieval and generation tasks. The encoders under study fall into two main groups consisting of image-centric encoders that were pretrained using visual modalities as intermediary hubs for connecting modalities, and text-centric encoders that were pretrained with direct audio-language alignment. We first measure the baseline performance of these encoders on two key tasks, namely cross-modal retrieval and text description generation in vision-language models. Subsequently, we realign all encoders with the CLIP space using highly coherent audio-visual data and observe the performance changes. Our findings reveal that the initial architectural type of the encoder determines how it responds to the alignment process. Image-centric encoders, which are inherently designed for alignment, demonstrate exceptional performance in cross-modal retrieval, but this intensive alignment causes compression of unique linguistic information and reduces the quality of their text description generation in vision-language models. In contrast, text-centric encoders, which possess stronger linguistic authenticity, are able to maintain a better balance between the two objectives.

Paperid: 896, https://arxiv.org/pdf/2506.10286.pdf GitHub

Abstract:
Hallucinations pose a significant challenge to the reliability of large vision-language models, making their detection essential for ensuring accuracy in critical applications. Current detection methods often rely on computationally intensive models, leading to high latency and resource demands. Their definitive outcomes also fail to account for real-world scenarios where the line between hallucinated and truthful information is unclear. To address these issues, we propose HalLoc, a dataset designed for efficient, probabilistic hallucination detection. It features 150K token-level annotated samples, including hallucination types, across Visual Question Answering (VQA), instruction-following, and image captioning tasks. This dataset facilitates the development of models that detect hallucinations with graded confidence, enabling more informed user interactions. Additionally, we introduce a baseline model trained on HalLoc, offering low-overhead, concurrent hallucination detection during generation. The model can be seamlessly integrated into existing VLMs, improving reliability while preserving efficiency. The prospect of a robust plug-and-play hallucination detection module opens new avenues for enhancing the trustworthiness of vision-language models in real-world applications. The HalLoc dataset and code are publicly available at: https://github.com/dbsltm/cvpr25_halloc.

Paperid: 897, https://arxiv.org/pdf/2506.10182.pdf GitHub

Abstract:
Personalized vision-language retrieval seeks to recognize new concepts (e.g. "my dog Fido") from only a few examples. This task is challenging because it requires not only learning a new concept from a few images, but also integrating the personal and general knowledge together to recognize the concept in different contexts. In this paper, we show how to effectively adapt the internal representation of a vision-language dual encoder model for personalized vision-language retrieval. We find that regularized low-rank adaption of a small set of parameters in the language encoder's final layer serves as a highly effective alternative to textual inversion for recognizing the personal concept while preserving general knowledge. Additionally, we explore strategies for combining parameters of multiple learned personal concepts, finding that parameter addition is effective. To evaluate how well general knowledge is preserved in a finetuned representation, we introduce a metric that measures image retrieval accuracy based on captions generated by a vision language model (VLM). Our approach achieves state-of-the-art accuracy on two benchmarks for personalized image retrieval with natural language queries - DeepFashion2 and ConCon-Chi - outperforming the prior art by 4%-22% on personal retrievals.

Paperid: 898, https://arxiv.org/pdf/2506.09937.pdf GitHub

Abstract:
While vision-language-action models (VLAs) have shown promising robotic behaviors across a diverse set of manipulation tasks, they achieve limited success rates when deployed on novel tasks out-of-the-box. To allow these policies to safely interact with their environments, we need a failure detector that gives a timely alert such that the robot can stop, backtrack, or ask for help. However, existing failure detectors are trained and tested only on one or a few specific tasks, while VLAs require the detector to generalize and detect failures also in unseen tasks and novel environments. In this paper, we introduce the multitask failure detection problem and propose SAFE, a failure detector for generalist robot policies such as VLAs. We analyze the VLA feature space and find that VLAs have sufficient high-level knowledge about task success and failure, which is generic across different tasks. Based on this insight, we design SAFE to learn from VLA internal features and predict a single scalar indicating the likelihood of task failure. SAFE is trained on both successful and failed rollouts, and is evaluated on unseen tasks. SAFE is compatible with different policy architectures. We test it on OpenVLA, $Ï_0$, and $Ï_0$-FAST in both simulated and real-world environments extensively. We compare SAFE with diverse baselines and show that SAFE achieves state-of-the-art failure detection performance and the best trade-off between accuracy and detection time using conformal prediction. More qualitative results can be found at https://vla-safe.github.io/.

Paperid: 899, https://arxiv.org/pdf/2506.09930.pdf GitHub

Abstract:
One promise that Vision-Language-Action (VLA) models hold over traditional imitation learning for robotics is to leverage the broad generalization capabilities of large Vision-Language Models (VLMs) to produce versatile, "generalist" robot policies. However, current evaluations of VLAs remain insufficient. Traditional imitation learning benchmarks are unsuitable due to the lack of language instructions. Emerging benchmarks for VLAs that incorporate language often come with limited evaluation tasks and do not intend to investigate how much VLM pretraining truly contributes to the generalization capabilities of the downstream robotic policy. Meanwhile, much research relies on real-world robot setups designed in isolation by different institutions, which creates a barrier for reproducibility and accessibility. To address this gap, we introduce a unified probing suite of 50 simulation-based tasks across 10 subcategories spanning language instruction, vision, and objects. We systematically evaluate several state-of-the-art VLA architectures on this suite to understand their generalization capability. Our results show that while VLM backbones endow VLAs with robust perceptual understanding and high level planning, which we refer to as good intentions, this does not reliably translate into precise motor execution: when faced with out-of-distribution observations, policies often exhibit coherent intentions, but falter in action execution. Moreover, finetuning on action data can erode the original VLM's generalist reasoning abilities. We release our task suite and evaluation code to serve as a standardized benchmark for future VLAs and to drive research on closing the perception-to-action gap. More information, including the source code, can be found at https://ai4ce.github.io/INT-ACT/

Paperid: 900, https://arxiv.org/pdf/2506.09634.pdf GitHub

Abstract:
Automated 3D CT diagnosis empowers clinicians to make timely, evidence-based decisions by enhancing diagnostic accuracy and workflow efficiency. While multimodal large language models (MLLMs) exhibit promising performance in visual-language understanding, existing methods mainly focus on 2D medical images, which fundamentally limits their ability to capture complex 3D anatomical structures. This limitation often leads to misinterpretation of subtle pathologies and causes diagnostic hallucinations. In this paper, we present Hybrid Spatial Encoding Network (HSENet), a framework that exploits enriched 3D medical visual cues by effective visual perception and projection for accurate and robust vision-language understanding. Specifically, HSENet employs dual-3D vision encoders to perceive both global volumetric contexts and fine-grained anatomical details, which are pre-trained by dual-stage alignment with diagnostic reports. Furthermore, we propose Spatial Packer, an efficient multimodal projector that condenses high-resolution 3D spatial regions into a compact set of informative visual tokens via centroid-based compression. By assigning spatial packers with dual-3D vision encoders, HSENet can seamlessly perceive and transfer hybrid visual representations to LLM's semantic space, facilitating accurate diagnostic text generation. Experimental results demonstrate that our method achieves state-of-the-art performance in 3D language-visual retrieval (39.85% of R@100, +5.96% gain), 3D medical report generation (24.01% of BLEU-4, +8.01% gain), and 3D visual question answering (73.60% of Major Class Accuracy, +1.99% gain), confirming its effectiveness. Our code is available at https://github.com/YanzhaoShi/HSENet.

Paperid: 901, https://arxiv.org/pdf/2506.09522.pdf GitHub

Abstract:
Large Vision-Language Models (LVLMs) have demonstrated remarkable performance across various multimodal tasks by integrating visual perception with language understanding. However, conventional decoding strategies of LVLMs often fail to successfully utilize visual information, leading to visually ungrounded responses. While various approaches have been proposed to address this limitation, they typically require additional training, multi-step inference procedures, or external model dependencies. This paper introduces ReVisiT, a simple yet effective decoding method that references vision tokens to guide the text generation process in LVLMs. Our approach leverages the semantic information embedded within vision tokens by projecting them into the text token distribution space, and dynamically selecting the most relevant vision token at each decoding step through constrained divergence minimization. This selected vision token is then used to refine the output distribution to better incorporate visual semantics. Experiments on three LVLM hallucination benchmarks with two recent LVLMs demonstrate that ReVisiT consistently enhances visual grounding with minimal computational overhead. Moreover, our method achieves competitive or superior results relative to state-of-the-art baselines while reducing computational costs for up to $2\times$.

Paperid: 902, https://arxiv.org/pdf/2506.09353.pdf GitHub

Abstract:
Large Vision-Language Models (LVLMs) have achieved impressive progress across various applications but remain vulnerable to malicious queries that exploit the visual modality. Existing alignment approaches typically fail to resist malicious queries while preserving utility on benign ones effectively. To address these challenges, we propose Deep Aligned Visual Safety Prompt (DAVSP), which is built upon two key innovations. First, we introduce the Visual Safety Prompt, which appends a trainable padding region around the input image. It preserves visual features and expands the optimization space. Second, we propose Deep Alignment, a novel approach to train the visual safety prompt through supervision in the model's activation space. It enhances the inherent ability of LVLMs to perceive malicious queries, achieving deeper alignment than prior works. Extensive experiments across five benchmarks on two representative LVLMs demonstrate that DAVSP effectively resists malicious queries while preserving benign input utility. Furthermore, DAVSP exhibits great cross-model generation ability. Ablation studies further reveal that both the Visual Safety Prompt and Deep Alignment are essential components, jointly contributing to its overall effectiveness. The code is publicly available at https://github.com/zhangyitonggg/DAVSP.

Paperid: 903, https://arxiv.org/pdf/2506.09081.pdf GitHub

Abstract:
We present FlagEvalMM, an open-source evaluation framework designed to comprehensively assess multimodal models across a diverse range of vision-language understanding and generation tasks, such as visual question answering, text-to-image/video generation, and image-text retrieval. We decouple model inference from evaluation through an independent evaluation service, thus enabling flexible resource allocation and seamless integration of new tasks and models. Moreover, FlagEvalMM utilizes advanced inference acceleration tools (e.g., vLLM, SGLang) and asynchronous data loading to significantly enhance evaluation efficiency. Extensive experiments show that FlagEvalMM offers accurate and efficient insights into model strengths and limitations, making it a valuable tool for advancing multimodal research. The framework is publicly accessible at https://github.com/flageval-baai/FlagEvalMM.

Paperid: 904, https://arxiv.org/pdf/2506.09049.pdf GitHub

Abstract:
Coordinating multiple embodied agents in dynamic environments remains a core challenge in artificial intelligence, requiring both perception-driven reasoning and scalable cooperation strategies. While recent works have leveraged large language models (LLMs) for multi-agent planning, a few have begun to explore vision-language models (VLMs) for visual reasoning. However, these VLM-based approaches remain limited in their support for diverse embodiment types. In this work, we introduce VIKI-Bench, the first hierarchical benchmark tailored for embodied multi-agent cooperation, featuring three structured levels: agent activation, task planning, and trajectory perception. VIKI-Bench includes diverse robot embodiments, multi-view visual observations, and structured supervision signals to evaluate reasoning grounded in visual inputs. To demonstrate the utility of VIKI-Bench, we propose VIKI-R, a two-stage framework that fine-tunes a pretrained vision-language model (VLM) using Chain-of-Thought annotated demonstrations, followed by reinforcement learning under multi-level reward signals. Our extensive experiments show that VIKI-R significantly outperforms baselines method across all task levels. Furthermore, we show that reinforcement learning enables the emergence of compositional cooperation patterns among heterogeneous agents. Together, VIKI-Bench and VIKI-R offer a unified testbed and method for advancing multi-agent, visual-driven cooperation in embodied AI systems.

Paperid: 905, https://arxiv.org/pdf/2506.09040.pdf GitHub

Abstract:
Typical large vision-language models (LVLMs) apply autoregressive supervision solely to textual sequences, without fully incorporating the visual modality into the learning process. This results in three key limitations: (1) an inability to utilize images without accompanying captions, (2) the risk that captions omit critical visual details, and (3) the challenge that certain vision-centric content cannot be adequately conveyed through text. As a result, current LVLMs often prioritize vision-to-language alignment while potentially overlooking fine-grained visual information. While some prior works have explored autoregressive image generation, effectively leveraging autoregressive visual supervision to enhance image understanding remains an open challenge. In this paper, we introduce Autoregressive Semantic Visual Reconstruction (ASVR), which enables joint learning of visual and textual modalities within a unified autoregressive framework. We show that autoregressively reconstructing the raw visual appearance of images does not enhance and may even impair multimodal understanding. In contrast, autoregressively reconstructing the semantic representation of images consistently improves comprehension. Notably, we find that even when models are given continuous image features as input, they can effectively reconstruct discrete semantic tokens, resulting in stable and consistent improvements across a wide range of multimodal understanding benchmarks. Our approach delivers significant performance gains across varying data scales (556k-2M) and types of LLM bacbones. Specifically, ASVR improves LLaVA-1.5 by 5% in average scores across 14 multimodal benchmarks. The code is available at https://github.com/AlenjandroWang/ASVR.

Paperid: 906, https://arxiv.org/pdf/2506.08990.pdf GitHub

Abstract:
Medical vision-language alignment through cross-modal contrastive learning shows promising performance in image-text matching tasks, such as retrieval and zero-shot classification. However, conventional cross-modal contrastive learning (CLIP-based) methods suffer from suboptimal visual representation capabilities, which also limits their effectiveness in vision-language alignment. In contrast, although the models pretrained via multimodal masked modeling struggle with direct cross-modal matching, they excel in visual representation. To address this contradiction, we propose ALTA (ALign Through Adapting), an efficient medical vision-language alignment method that utilizes only about 8% of the trainable parameters and less than 1/5 of the computational consumption required for masked record modeling. ALTA achieves superior performance in vision-language matching tasks like retrieval and zero-shot classification by adapting the pretrained vision model from masked record modeling. Additionally, we integrate temporal-multiview radiograph inputs to enhance the information consistency between radiographs and their corresponding descriptions in reports, further improving the vision-language alignment. Experimental evaluations show that ALTA outperforms the best-performing counterpart by over 4% absolute points in text-to-image accuracy and approximately 6% absolute points in image-to-text retrieval accuracy. The adaptation of vision-language models during efficient alignment also promotes better vision and language understanding. Code is publicly available at https://github.com/DopamineLcy/ALTA.

Paperid: 907, https://arxiv.org/pdf/2506.08887.pdf GitHub

Abstract:
The parameter-efficient adaptation of the image-text pretraining model CLIP for video-text retrieval is a prominent area of research. While CLIP is focused on image-level vision-language matching, video-text retrieval demands comprehensive understanding at the video level. Three key discrepancies emerge in the transfer from image-level to video-level: vision, language, and alignment. However, existing methods mainly focus on vision while neglecting language and alignment. In this paper, we propose Discrepancy Reduction in Vision, Language, and Alignment (DiscoVLA), which simultaneously mitigates all three discrepancies. Specifically, we introduce Image-Video Features Fusion to integrate image-level and video-level features, effectively tackling both vision and language discrepancies. Additionally, we generate pseudo image captions to learn fine-grained image-level alignment. To mitigate alignment discrepancies, we propose Image-to-Video Alignment Distillation, which leverages image-level alignment knowledge to enhance video-level alignment. Extensive experiments demonstrate the superiority of our DiscoVLA. In particular, on MSRVTT with CLIP (ViT-B/16), DiscoVLA outperforms previous methods by 1.5% in R@1, reaching a final score of 50.5% R@1. The code is available at https://github.com/LunarShen/DsicoVLA.

Paperid: 908, https://arxiv.org/pdf/2506.08849.pdf GitHub

Abstract:
Medical ultrasonography is an essential imaging technique for examining superficial organs and tissues, including lymph nodes, breast, and thyroid. It employs high-frequency ultrasound waves to generate detailed images of the internal structures of the human body. However, manually contouring regions of interest in these images is a labor-intensive task that demands expertise and often results in inconsistent interpretations among individuals. Vision-language foundation models, which have excelled in various computer vision applications, present new opportunities for enhancing ultrasound image analysis. Yet, their performance is hindered by the significant differences between natural and medical imaging domains. This research seeks to overcome these challenges by developing domain adaptation methods for vision-language foundation models. In this study, we explore the fine-tuning pipeline for vision-language foundation models by utilizing large language model as text refiner with special-designed adaptation strategies and task-driven heads. Our approach has been extensively evaluated on six ultrasound datasets and two tasks: segmentation and classification. The experimental results show that our method can effectively improve the performance of vision-language foundation models for ultrasound image analysis, and outperform the existing state-of-the-art vision-language and pure foundation models. The source code of this study is available at https://github.com/jinggqu/NextGen-UIA.

Paperid: 909, https://arxiv.org/pdf/2506.08440.pdf GitHub

Abstract:
Recent advances in Vision-Language-Action (VLA) model have demonstrated strong generalization capabilities across diverse scenes, tasks, and robotic platforms when pretrained at large-scale datasets. However, these models still require task-specific fine-tuning in novel environments, a process that relies almost exclusively on supervised fine-tuning (SFT) using static trajectory datasets. Such approaches neither allow robot to interact with environment nor do they leverage feedback from live execution. Also, their success is critically dependent on the size and quality of the collected trajectories. Reinforcement learning (RL) offers a promising alternative by enabling closed-loop interaction and aligning learned policies directly with task objectives. In this work, we draw inspiration from the ideas of GRPO and propose the Trajectory-wise Group Relative Policy Optimization (TGRPO) method. By fusing step-level and trajectory-level advantage signals, this method improves GRPO's group-level advantage estimation, thereby making the algorithm more suitable for online reinforcement learning training of VLA. Experimental results on ten manipulation tasks from the libero-object benchmark demonstrate that TGRPO consistently outperforms various baseline methods, capable of generating more robust and efficient policies across multiple tested scenarios. Our source codes are available at: https://github.com/hahans/TGRPO

Paperid: 910, https://arxiv.org/pdf/2506.08185.pdf GitHub

Abstract:
Surgeons exhibit distinct operating styles shaped by training, experience, and motor behavior-yet most surgical AI systems overlook this personalization signal. We propose a novel agentic modeling approach for surgeon-specific behavior prediction in robotic surgery, combining a discrete diffusion framework with a vision-language-action (VLA) pipeline. Gesture prediction is framed as a structured sequence denoising task, conditioned on multimodal inputs including surgical video, intent language, and personalized embeddings of surgeon identity and skill. These embeddings are encoded through natural language prompts using third-party language models, allowing the model to retain individual behavioral style without exposing explicit identity. We evaluate our method on the JIGSAWS dataset and demonstrate that it accurately reconstructs gesture sequences while learning meaningful motion fingerprints unique to each surgeon. To quantify the privacy implications of personalization, we perform membership inference attacks and find that more expressive embeddings improve task performance but simultaneously increase susceptibility to identity leakage. These findings demonstrate that while personalized embeddings improve performance, they also increase vulnerability to identity leakage, revealing the importance of balancing personalization with privacy risk in surgical modeling. Code is available at: https://github.com/huixin-zhan-ai/Surgeon_style_fingerprinting.

Paperid: 911, https://arxiv.org/pdf/2506.08071.pdf GitHub

Abstract:
Popular text-to-image (T2I) systems are trained on web-scraped data, which is heavily Amero and Euro-centric, underrepresenting the cultures of the Global South. To analyze these biases, we introduce CuRe, a novel and scalable benchmarking and scoring suite for cultural representativeness that leverages the marginal utility of attribute specification to T2I systems as a proxy for human judgments. Our CuRe benchmark dataset has a novel categorical hierarchy built from the crowdsourced Wikimedia knowledge graph, with 300 cultural artifacts across 32 cultural subcategories grouped into six broad cultural axes (food, art, fashion, architecture, celebrations, and people). Our dataset's categorical hierarchy enables CuRe scorers to evaluate T2I systems by analyzing their response to increasing the informativeness of text conditioning, enabling fine-grained cultural comparisons. We empirically observe much stronger correlations of our class of scorers to human judgments of perceptual similarity, image-text alignment, and cultural diversity across image encoders (SigLIP 2, AIMV2 and DINOv2), vision-language models (OpenCLIP, SigLIP 2, Gemini 2.0 Flash) and state-of-the-art text-to-image systems, including three variants of Stable Diffusion (1.5, XL, 3.5 Large), FLUX.1 [dev], Ideogram 2.0, and DALL-E 3. The code and dataset is open-sourced and available at https://aniketrege.github.io/cure/.

Paperid: 912, https://arxiv.org/pdf/2506.08010.pdf GitHub

Abstract:
We investigate the mechanism underlying a previously identified phenomenon in Vision Transformers -- the emergence of high-norm tokens that lead to noisy attention maps. We observe that in multiple models (e.g., CLIP, DINOv2), a sparse set of neurons is responsible for concentrating high-norm activations on outlier tokens, leading to irregular attention patterns and degrading downstream visual processing. While the existing solution for removing these outliers involves retraining models from scratch with additional learned register tokens, we use our findings to create a training-free approach to mitigate these artifacts. By shifting the high-norm activations from our discovered register neurons into an additional untrained token, we can mimic the effect of register tokens on a model already trained without registers. We demonstrate that our method produces cleaner attention and feature maps, enhances performance over base models across multiple downstream visual tasks, and achieves results comparable to models explicitly trained with register tokens. We then extend test-time registers to off-the-shelf vision-language models to improve their interpretability. Our results suggest that test-time registers effectively take on the role of register tokens at test-time, offering a training-free solution for any pre-trained model released without them.

Paperid: 913, https://arxiv.org/pdf/2506.07785.pdf GitHub

Abstract:
Recent advancements in Large Vision Language Models (LVLMs) have significantly improved performance in Visual Question Answering (VQA) tasks through multimodal Retrieval-Augmented Generation (RAG). However, existing methods still face challenges, such as the scarcity of knowledge with reasoning examples and erratic responses from retrieved knowledge. To address these issues, in this study, we propose a multimodal RAG framework, termed RCTS, which enhances LVLMs by constructing a Reasoning Context-enriched knowledge base and a Tree Search re-ranking method. Specifically, we introduce a self-consistent evaluation mechanism to enrich the knowledge base with intrinsic reasoning patterns. We further propose a Monte Carlo Tree Search with Heuristic Rewards (MCTS-HR) to prioritize the most relevant examples. This ensures that LVLMs can leverage high-quality contextual reasoning for better and more consistent responses. Extensive experiments demonstrate that our framework achieves state-of-the-art performance on multiple VQA datasets, significantly outperforming In-Context Learning (ICL) and Vanilla-RAG methods. It highlights the effectiveness of our knowledge base and re-ranking method in improving LVLMs. Our code is available at https://github.com/yannqi/RCTS-RAG.

Paperid: 914, https://arxiv.org/pdf/2506.07631.pdf GitHub

Abstract:
Large Vision-Language Models (VLMs) now generate highly detailed, paragraphlength image captions, yet evaluating their factual accuracy remains challenging. Current methods often miss fine-grained errors, being designed for shorter texts or lacking datasets with verified inaccuracies. We introduce DOCCI-Critique, a benchmark with 1,400 VLM-generated paragraph captions (100 images, 14 VLMs) featuring over 10,216 sentence-level human annotations of factual correctness and explanatory rationales for errors, all within paragraph context. Building on this, we develop VNLI-Critique, a model for automated sentence-level factuality classification and critique generation. We highlight three key applications: (1) VNLI-Critique demonstrates robust generalization, validated by state-of-the-art performance on the M-HalDetect benchmark and strong results in CHOCOLATE claim verification. (2) The VNLI-Critique driven AutoRater for DOCCI-Critique provides reliable VLM rankings, showing excellent alignment with human factuality judgments (e.g., 0.98 Spearman). (3) An innovative Critic-and-Revise pipeline, where critiques from VNLI-Critique guide LLM-based corrections, achieves substantial improvements in caption factuality (e.g., a 46% gain on DetailCaps-4870). Our work offers a crucial benchmark alongside practical tools, designed to significantly elevate the standards for fine-grained evaluation and foster the improvement of VLM image understanding. Project page: https://google.github.io/unblocking-detail-caption

Paperid: 915, https://arxiv.org/pdf/2506.07553.pdf GitHub

Abstract:
Optical Chemical Structure Recognition (OCSR) is crucial for digitizing chemical knowledge by converting molecular images into machine-readable formats. While recent vision-language models (VLMs) have shown potential in this task, their image-captioning approach often struggles with complex molecular structures and inconsistent annotations. To overcome these challenges, we introduce GTR-Mol-VLM, a novel framework featuring two key innovations: (1) the Graph Traversal as Visual Chain of Thought mechanism that emulates human reasoning by incrementally parsing molecular graphs through sequential atom-bond predictions, and (2) the data-centric principle of Faithfully Recognize What You've Seen, which addresses the mismatch between abbreviated structures in images and their expanded annotations. To support model development, we constructed GTR-CoT-1.3M, a large-scale instruction-tuning dataset with meticulously corrected annotations, and introduced MolRec-Bench, the first benchmark designed for a fine-grained evaluation of graph-parsing accuracy in OCSR. Comprehensive experiments demonstrate that GTR-Mol-VLM achieves superior results compared to specialist models, chemistry-domain VLMs, and commercial general-purpose VLMs. Notably, in scenarios involving molecular images with functional group abbreviations, GTR-Mol-VLM outperforms the second-best baseline by approximately 14 percentage points, both in SMILES-based and graph-based metrics. We hope that this work will drive OCSR technology to more effectively meet real-world needs, thereby advancing the fields of cheminformatics and AI for Science. We will release GTR-CoT at https://github.com/opendatalab/GTR-CoT.

Paperid: 916, https://arxiv.org/pdf/2506.07530.pdf GitHub

Abstract:
Vision-Language-Action (VLA) models have shown impressive capabilities across a wide range of robotics manipulation tasks. However, their growing model size poses significant challenges for deployment on resource-constrained robotic systems. While 1-bit pretraining has proven effective for enhancing the inference efficiency of large language models with minimal performance loss, its application to VLA models remains underexplored. In this work, we present BitVLA, the first 1-bit VLA model for robotics manipulation, in which every parameter is ternary, i.e., {-1, 0, 1}. To further reduce the memory footprint of the vision encoder, we propose the distillation-aware training strategy that compresses the full-precision encoder to 1.58-bit weights. During this process, a full-precision encoder serves as a teacher model to better align latent representations. Despite the lack of large-scale robotics pretraining, BitVLA achieves performance comparable to the state-of-the-art model OpenVLA-OFT with 4-bit post-training quantization on the LIBERO benchmark, while consuming only 29.8% of the memory. These results highlight BitVLA's promise for deployment on memory-constrained edge devices. We release the code and model weights in https://github.com/ustcwhy/BitVLA.

Paperid: 917, https://arxiv.org/pdf/2506.07509.pdf GitHub

Abstract:
Recent advances in agentic and physical artificial intelligence (AI) have largely focused on ground-based platforms such as humanoid and wheeled robots, leaving aerial robots relatively underexplored. Meanwhile, state-of-the-art unmanned aerial vehicle (UAV) multimodal vision-language systems typically rely on closed-source models accessible only to well-resourced organizations. To democratize natural language control of autonomous drones, we present an open-source agentic framework that integrates PX4-based flight control, Robot Operating System 2 (ROS 2) middleware, and locally hosted models using Ollama. We evaluate performance both in simulation and on a custom quadcopter platform, benchmarking four large language model (LLM) families for command generation and three vision-language model (VLM) families for scene understanding.

Paperid: 918, https://arxiv.org/pdf/2506.07484.pdf GitHub

Abstract:
Prompt tuning, which adapts vision-language models by freezing model parameters and optimizing only the prompt, has proven effective for task-specific adaptations. The core challenge in prompt tuning is improving specialization for a specific task and generalization for unseen domains. However, frozen encoders often produce misaligned features, leading to confusion between classes and limiting specialization. To overcome this issue, we propose a confusion-aware loss (CoA-loss) that improves specialization by refining the decision boundaries between confusing classes. Additionally, we mathematically demonstrate that a mixture model can enhance generalization without compromising specialization. This is achieved using confidence-aware weights (CoA-weights), which adjust the weights of each prediction in the mixture model based on its confidence within the class domains. Extensive experiments show that CoCoA-Mix, a mixture model with CoA-loss and CoA-weights, outperforms state-of-the-art methods by enhancing specialization and generalization. Our code is publicly available at https://github.com/url-kaist/CoCoA-Mix.

Paperid: 919, https://arxiv.org/pdf/2506.07138.pdf GitHub GitHub

Abstract:
Large multimodal models (LMMs) suffer significant computational challenges due to the high cost of Large Language Models (LLMs) and the quadratic complexity of processing long vision token sequences. In this paper, we explore the spatial redundancy among vision tokens and shorten the length of vision token sequences for inference acceleration. Specifically, we propose a Spatial Token Fusion (STF) method to learn compact vision tokens for short vision token sequence, where spatial-adjacent tokens are fused into one. Meanwhile, weight-frozen vision encoder can not well adapt to the demand of extensive downstream vision-language tasks. To this end, we further introduce a Multi-Block Token Fusion (MBTF) module to supplement multi-granularity features for the reduced token sequence. Overall, we combine STF and MBTF module to balance token reduction and information preservation, thereby improving inference efficiency without sacrificing multimodal reasoning capabilities. Experimental results demonstrate that our method based on LLaVA-1.5 achieves comparable or even superior performance to the baseline on 8 popular vision-language benchmarks with only $25\%$ vision tokens of baseline. The source code and trained weights are available at https://github.com/visresearch/LLaVA-STF.

Paperid: 920, https://arxiv.org/pdf/2506.06884.pdf GitHub

Abstract:
In recent years, Vision-Language Models (VLMs) have shown remarkable performance improvements in Vision-Language tasks. However, their large size poses challenges for real-world applications where inference latency is a concern. To tackle this issue, we propose employing Early Exit (EE) strategies in VLMs. However, training exit classifiers in VLMs is challenging, particularly with limited labeled training data. To address this, we introduce FREE, an adversarial training approach within a GAN-based framework. Here, each exit consists of a transformer layer and a classifier. The transformer layer is adversarially trained to produce feature representations similar to the final layer, while a feature classifier serves as the discriminator. Our method focuses on performing input-adaptive inference that increases inference speed with minimal drop in performance. Experimental results demonstrate the effectiveness of our approach in enhancing accuracy and model robustness by mitigating overthinking and the phenomenon of mid-crisis that we highlight. We experimentally validate that our method speeds up the inference process by more than 1.51x while retaining comparable performance. The source code is available at https://github.com/Div290/FREE.

Paperid: 921, https://arxiv.org/pdf/2506.06279.pdf GitHub

Abstract:
Recent advancements in Large Vision-Language Models built upon Large Language Models have established aligning visual features with LLM representations as the dominant paradigm. However, inherited LLM architectural designs introduce suboptimal characteristics for multimodal processing. First, LVLMs exhibit a bimodal distribution in attention allocation, leading to the progressive neglect of middle visual content as context expands. Second, conventional positional encoding schemes fail to preserve vital 2D structural relationships when processing dynamic high-resolution images. To address these limitations, we propose CoMemo - a dual-path architecture that combines a Context image path with an image Memory path for visual processing, effectively alleviating visual information neglect. Additionally, we introduce RoPE-DHR, a novel positional encoding mechanism that employs thumbnail-based positional aggregation to maintain 2D spatial awareness while mitigating remote decay in extended sequences. Evaluations across seven benchmarks,including long-context comprehension, multi-image reasoning, and visual question answering, demonstrate CoMemo's superior performance compared to conventional LVLM architectures. Project page is available at https://lalbj.github.io/projects/CoMemo/.

Paperid: 922, https://arxiv.org/pdf/2506.06218.pdf GitHub

Abstract:
We introduce STSBench, a scenario-based framework to benchmark the holistic understanding of vision-language models (VLMs) for autonomous driving. The framework automatically mines pre-defined traffic scenarios from any dataset using ground-truth annotations, provides an intuitive user interface for efficient human verification, and generates multiple-choice questions for model evaluation. Applied to the NuScenes dataset, we present STSnu, the first benchmark that evaluates the spatio-temporal reasoning capabilities of VLMs based on comprehensive 3D perception. Existing benchmarks typically target off-the-shelf or fine-tuned VLMs for images or videos from a single viewpoint and focus on semantic tasks such as object recognition, dense captioning, risk assessment, or scene understanding. In contrast, STSnu evaluates driving expert VLMs for end-to-end driving, operating on videos from multi-view cameras or LiDAR. It specifically assesses their ability to reason about both ego-vehicle actions and complex interactions among traffic participants, a crucial capability for autonomous vehicles. The benchmark features 43 diverse scenarios spanning multiple views and frames, resulting in 971 human-verified multiple-choice questions. A thorough evaluation uncovers critical shortcomings in existing models' ability to reason about fundamental traffic dynamics in complex environments. These findings highlight the urgent need for architectural advances that explicitly model spatio-temporal reasoning. By addressing a core gap in spatio-temporal evaluation, STSBench enables the development of more robust and explainable VLMs for autonomous driving.

Paperid: 923, https://arxiv.org/pdf/2506.06076.pdf GitHub

Abstract:
Vision-language models (VLMs) pre-trained at large scale have shown unprecedented transferability capabilities and are being progressively integrated into medical image analysis. Although its discriminative potential has been widely explored, its reliability aspect remains overlooked. This work investigates their behavior under the increasingly popular split conformal prediction (SCP) framework, which theoretically guarantees a given error level on output sets by leveraging a labeled calibration set. However, the zero-shot performance of VLMs is inherently limited, and common practice involves few-shot transfer learning pipelines, which cannot absorb the rigid exchangeability assumptions of SCP. To alleviate this issue, we propose full conformal adaptation, a novel setting for jointly adapting and conformalizing pre-trained foundation models, which operates transductively over each test data point using a few-shot adaptation set. Moreover, we complement this framework with SS-Text, a novel training-free linear probe solver for VLMs that alleviates the computational cost of such a transductive approach. We provide comprehensive experiments using 3 different modality-specialized medical VLMs and 9 adaptation tasks. Our framework requires exactly the same data as SCP, and provides consistent relative improvements of up to 27% on set efficiency while maintaining the same coverage guarantees.

Paperid: 924, https://arxiv.org/pdf/2506.05765.pdf GitHub

Abstract:
Humans are susceptible to optical illusions, which serve as valuable tools for investigating sensory and cognitive processes. Inspired by human vision studies, research has begun exploring whether machines, such as large vision language models (LVLMs), exhibit similar susceptibilities to visual illusions. However, studies often have used non-abstract images and have not distinguished actual and apparent features, leading to ambiguous assessments of machine cognition. To address these limitations, we introduce a visual question answering (VQA) dataset, categorized into genuine and fake illusions, along with corresponding control images. Genuine illusions present discrepancies between actual and apparent features, whereas fake illusions have the same actual and apparent features even though they look illusory due to the similar geometric configuration. We evaluate the performance of LVLMs for genuine and fake illusion VQA tasks and investigate whether the models discern actual and apparent features. Our findings indicate that although LVLMs may appear to recognize illusions by correctly answering questions about both feature types, they predict the same answers for both Genuine Illusion and Fake Illusion VQA questions. This suggests that their responses might be based on prior knowledge of illusions rather than genuine visual understanding. The dataset is available at https://github.com/ynklab/FILM

Paperid: 925, https://arxiv.org/pdf/2506.05412.pdf GitHub

Abstract:
Gaze-referential inference--the ability to infer what others are looking at--is a critical component of a theory of mind that underpins natural human-AI interaction. In a controlled study, we evaluated this skill across 111 Vision Language Models (VLMs) using photos taken with manipulated difficulty and variability, comparing performance with that of human participants (N = 65), and analyzed behaviors using mixed-effects models. We found that 94 of the 111 VLMs failed to do better than random guessing, while humans achieved near-ceiling accuracy. VLMs even respond with each choice almost equally frequently. Are they randomly guessing? Although most VLMs struggle, when we zoom in on five of the top-tier VLMs with above-chance performance, we find that their performance declined with increasing task difficulty but varied only slightly across different prompts and scene objects. These behavioral features cannot be explained by considering them as random guessers. Instead, they likely use a combination of heuristics and guessing such that their performance is subject to the task difficulty but robust to perceptual variations. This suggests that VLMs, lacking gaze inference capability, have yet to become technologies that can naturally interact with humans, but the potential remains.

Paperid: 926, https://arxiv.org/pdf/2506.05218.pdf GitHub

Abstract:
We introduce MonkeyOCR, a vision-language model for document parsing that advances the state of the art by leveraging a Structure-Recognition-Relation (SRR) triplet paradigm. This design simplifies what would otherwise be a complex multi-tool pipeline (as in MinerU's modular approach) and avoids the inefficiencies of processing full pages with giant end-to-end models (e.g., large multimodal LLMs like Qwen-VL). In SRR, document parsing is abstracted into three fundamental questions - "Where is it?" (structure), "What is it?" (recognition), and "How is it organized?" (relation) - corresponding to layout analysis, content identification, and logical ordering. This focused decomposition balances accuracy and speed: it enables efficient, scalable processing without sacrificing precision. To train and evaluate this approach, we introduce the MonkeyDoc (the most comprehensive document parsing dataset to date), with 3.9 million instances spanning over ten document types in both Chinese and English. Experiments show that MonkeyOCR outperforms MinerU by an average of 5.1%, with particularly notable improvements on challenging content such as formulas (+15.0%) and tables (+8.6%). Remarkably, our 3B-parameter model surpasses much larger and top-performing models, including Qwen2.5-VL (72B) and Gemini 2.5 Pro, achieving state-of-the-art average performance on English document parsing tasks. In addition, MonkeyOCR processes multi-page documents significantly faster (0.84 pages per second compared to 0.65 for MinerU and 0.12 for Qwen2.5-VL-7B). The 3B model can be efficiently deployed for inference on a single NVIDIA 3090 GPU. Code and models will be released at https://github.com/Yuliang-Liu/MonkeyOCR.

Paperid: 927, https://arxiv.org/pdf/2506.04713.pdf GitHub

Abstract:
Pretrained VLMs achieve strong performance on downstream tasks when adapted with just a few labeled examples. As the adapted models inevitably encounter out-of-distribution (OOD) test data that deviates from the in-distribution (ID) task-specific training data, enhancing OOD generalization in few-shot adaptation is critically important. We study robust few-shot VLM adaptation, aiming to increase both ID and OOD accuracy. By comparing different adaptation methods (e.g., prompt tuning, linear probing, contrastive finetuning, and full finetuning), we uncover three key findings: (1) finetuning with proper hyperparameters significantly outperforms the popular VLM adaptation methods prompt tuning and linear probing; (2) visual encoder-only finetuning achieves better efficiency and accuracy than contrastively finetuning both visual and textual encoders; (3) finetuning the top layers of the visual encoder provides the best balance between ID and OOD accuracy. Building on these findings, we propose partial finetuning of the visual encoder empowered with two simple augmentation techniques: (1) retrieval augmentation which retrieves task-relevant data from the VLM's pretraining dataset to enhance adaptation, and (2) adversarial perturbation which promotes robustness during finetuning. Results show that the former/latter boosts OOD/ID accuracy while slightly sacrificing the ID/OOD accuracy. Yet, perhaps understandably, naively combining the two does not maintain their best OOD/ID accuracy. We address this dilemma with the developed SRAPF, Stage-wise Retrieval Augmentation-based Adversarial Partial Finetuning. SRAPF consists of two stages: (1) partial finetuning the visual encoder using both ID and retrieved data, and (2) adversarial partial finetuning with few-shot ID data. Extensive experiments demonstrate that SRAPF achieves the state-of-the-art ID and OOD accuracy on the ImageNet OOD benchmarks.

Paperid: 928, https://arxiv.org/pdf/2506.04704.pdf GitHub

Abstract:
Despite emerging efforts to enhance the safety of Vision-Language Models (VLMs), current approaches face two main shortcomings. 1) Existing safety-tuning datasets and benchmarks only partially consider how image-text interactions can yield harmful content, often overlooking contextually unsafe outcomes from seemingly benign pairs. This narrow coverage leaves VLMs vulnerable to jailbreak attacks in unseen configurations. 2) Prior methods rely primarily on data-centric tuning, with limited architectural innovations to intrinsically strengthen safety. We address these gaps by introducing a holistic safety dataset and benchmark, HoliSafe, that spans all five safe/unsafe image-text combinations, providing a more robust basis for both training and evaluation. We further propose SafeLLaVA, a novel VLM augmented with a learnable safety meta token and a dedicated safety head. The meta token encodes harmful visual cues during training, intrinsically guiding the language model toward safer responses, while the safety head offers interpretable harmfulness classification aligned with refusal rationales. Experiments show that SafeLLaVA, trained on HoliSafe, achieves state-of-the-art safety performance across multiple VLM benchmarks. Additionally, the HoliSafe benchmark itself reveals critical vulnerabilities in existing models. We hope that HoliSafe and SafeLLaVA will spur further research into robust and interpretable VLM safety, expanding future avenues for multimodal alignment.

Paperid: 929, https://arxiv.org/pdf/2506.04308.pdf GitHub

Abstract:
Spatial referring is a fundamental capability of embodied robots to interact with the 3D physical world. However, even with the powerful pretrained vision language models (VLMs), recent approaches are still not qualified to accurately understand the complex 3D scenes and dynamically reason about the instruction-indicated locations for interaction. To this end, we propose RoboRefer, a 3D-aware VLM that can first achieve precise spatial understanding by integrating a disentangled but dedicated depth encoder via supervised fine-tuning (SFT). Moreover, RoboRefer advances generalized multi-step spatial reasoning via reinforcement fine-tuning (RFT), with metric-sensitive process reward functions tailored for spatial referring tasks. To support SFT and RFT training, we introduce RefSpatial, a large-scale dataset of 20M QA pairs (2x prior), covering 31 spatial relations (vs. 15 prior) and supporting complex reasoning processes (up to 5 steps). In addition, we introduce RefSpatial-Bench, a challenging benchmark filling the gap in evaluating spatial referring with multi-step reasoning. Experiments show that SFT-trained RoboRefer achieves state-of-the-art spatial understanding, with an average success rate of 89.6%. RFT-trained RoboRefer further outperforms all other baselines by a large margin, even surpassing Gemini-2.5-Pro by 17.4% in average accuracy on RefSpatial-Bench. Notably, RoboRefer can be integrated with various control policies to execute long-horizon, dynamic tasks across diverse robots (e,g., UR5, G1 humanoid) in cluttered real-world scenes.

Paperid: 930, https://arxiv.org/pdf/2506.04158.pdf GitHub

Abstract:
While diffusion models have achieved remarkable success in text-to-image generation, they encounter significant challenges with instruction-driven image editing. Our research highlights a key challenge: these models particularly struggle with structurally inconsistent edits that involve substantial layout changes. To mitigate this gap, we introduce Image Editing As Programs (IEAP), a unified image editing framework built upon the Diffusion Transformer (DiT) architecture. At its core, IEAP approaches instructional editing through a reductionist lens, decomposing complex editing instructions into sequences of atomic operations. Each operation is implemented via a lightweight adapter sharing the same DiT backbone and is specialized for a specific type of edit. Programmed by a vision-language model (VLM)-based agent, these operations collaboratively support arbitrary and structurally inconsistent transformations. By modularizing and sequencing edits in this way, IEAP generalizes robustly across a wide range of editing tasks, from simple adjustments to substantial structural changes. Extensive experiments demonstrate that IEAP significantly outperforms state-of-the-art methods on standard benchmarks across various editing scenarios. In these evaluations, our framework delivers superior accuracy and semantic fidelity, particularly for complex, multi-step instructions. Codes are available at https://github.com/YujiaHu1109/IEAP.

Paperid: 931, https://arxiv.org/pdf/2506.04070.pdf GitHub GitHub

Abstract:
Navigation instruction generation for visually impaired (VI) individuals (NIG-VI) is critical yet relatively underexplored. This study, hence, focuses on producing precise, in-situ, step-by-step navigation instructions that are practically usable by VI users. Concretely, we propose LaF-GRPO (LLM-as-Follower GRPO), where an LLM simulates VI user responses to generate rewards guiding the Vision-Language Model (VLM) post-training. This enhances instruction usability while reducing costly real-world data needs. To facilitate training and testing, we introduce NIG4VI, a 27k-sample open-sourced benchmark. It provides diverse navigation scenarios with accurate spatial coordinates, supporting detailed, open-ended in-situ instruction generation. Experiments on NIG4VI show the effectiveness of LaF-GRPO by quantitative metrics (e.g., Zero-(LaF-GRPO) boosts BLEU +14\%; SFT+(LaF-GRPO) METEOR 0.542 vs. GPT-4o's 0.323) and yields more intuitive, safer instructions. Code and benchmark are available at \href{https://github.com/YiyiyiZhao/NIG4VI}{https://github.com/YiyiyiZhao/NIG4VI}.

Paperid: 932, https://arxiv.org/pdf/2506.04005.pdf GitHub

Abstract:
Recent advances in few-shot adaptation for Vision-Language Models (VLMs) have greatly expanded their ability to generalize across tasks using only a few labeled examples. However, existing approaches primarily build upon the strong zero-shot priors of these models by leveraging carefully designed, task-specific prompts. This dependence on predefined class names can restrict their applicability, especially in scenarios where exact class names are unavailable or difficult to specify. To address this limitation, we introduce vocabulary-free few-shot learning for VLMs, a setting where target class instances - that is, images - are available but their corresponding names are not. We propose Similarity Mapping (SiM), a simple yet effective baseline that classifies target instances solely based on similarity scores with a set of generic prompts (textual or visual), eliminating the need for carefully handcrafted prompts. Although conceptually straightforward, SiM demonstrates strong performance, operates with high computational efficiency (learning the mapping typically takes less than one second), and provides interpretability by linking target classes to generic prompts. We believe that our approach could serve as an important baseline for future research in vocabulary-free few-shot learning. Code is available at https://github.com/MaxZanella/vocabulary-free-FSL.

Paperid: 933, https://arxiv.org/pdf/2506.03662.pdf GitHub

Abstract:
Locating human-object interaction (HOI) actions within video serves as the foundation for multiple downstream tasks, such as human behavior analysis and human-robot skill transfer. Current temporal action localization methods typically rely on annotated action and object categories of interactions for optimization, which leads to domain bias and low deployment efficiency. Although some recent works have achieved zero-shot temporal action localization (ZS-TAL) with large vision-language models (VLMs), their coarse-grained estimations and open-loop pipelines hinder further performance improvements for temporal interaction localization (TIL). To address these issues, we propose a novel zero-shot TIL approach dubbed EgoLoc to locate the timings of grasp actions for human-object interaction in egocentric videos. EgoLoc introduces a self-adaptive sampling strategy to generate reasonable visual prompts for VLM reasoning. By absorbing both 2D and 3D observations, it directly samples high-quality initial guesses around the possible contact/separation timestamps of HOI according to 3D hand velocities, leading to high inference accuracy and efficiency. In addition, EgoLoc generates closed-loop feedback from visual and dynamic cues to further refine the localization results. Comprehensive experiments on the publicly available dataset and our newly proposed benchmark demonstrate that EgoLoc achieves better temporal interaction localization for egocentric videos compared to state-of-the-art baselines. We will release our code and relevant data as open-source at https://github.com/IRMVLab/EgoLoc.

Paperid: 934, https://arxiv.org/pdf/2506.03614.pdf GitHub

Abstract:
One way to mitigate risks in vision-language models (VLMs) is to remove dangerous samples in their training data. However, such data moderation can be easily bypassed when harmful images are split into small, benign-looking patches, scattered across many training samples. VLMs may then learn to piece these fragments together during training and generate harmful responses at inference, either from full images or text references. For instance, if trained on image patches from a bloody scene paired with the descriptions "safe," VLMs may later describe, the full image or a text reference to the scene, as "safe." We define the core ability of VLMs enabling this attack as $\textit{visual stitching}$ -- the ability to integrate visual information spread across multiple training samples that share the same textual descriptions. In our work, we first demonstrate visual stitching abilities in common open-source VLMs on three datasets where each image is labeled with a unique synthetic ID: we split each $(\texttt{image}, \texttt{ID})$ pair into $\{(\texttt{patch}, \texttt{ID})\}$ pairs at different granularity for finetuning, and we find that tuned models can verbalize the correct IDs from full images or text reference. Building on this, we simulate the adversarial data poisoning scenario mentioned above by using patches from dangerous images and replacing IDs with text descriptions like ``safe'' or ``unsafe'', demonstrating how harmful content can evade moderation in patches and later be reconstructed through visual stitching, posing serious VLM safety risks. Code is available at https://github.com/ZHZisZZ/visual-stitching.

Paperid: 935, https://arxiv.org/pdf/2506.03569.pdf GitHub

Authors:Core Team, Zihao Yue, Zhenru Lin, Yifan Song, Weikun Wang, Shuhuai Ren, Shuhao Gu, Shicheng Li, Peidian Li, Liang Zhao, Lei Li, Kainan Bao, Hao Tian, Hailin Zhang, Gang Wang, Dawei Zhu, Cici, Chenhong He, Bowen Ye, Bowen Shen, Zihan Zhang, Zihan Jiang, Zhixian Zheng, Zhichao Song, Zhenbo Luo, Yue Yu, Yudong Wang, Yuanyuan Tian, Yu Tu, Yihan Yan, Yi Huang, Xu Wang, Xinzhe Xu, Xingchen Song, Xing Zhang, Xing Yong, Xin Zhang, Xiangwei Deng, Wenyu Yang, Wenhan Ma, Weiwei Lv, Weiji Zhuang, Wei Liu, Sirui Deng, Shuo Liu, Shimao Chen, Shihua Yu, Shaohui Liu, Shande Wang, Rui Ma, Qiantong Wang, Peng Wang, Nuo Chen, Menghang Zhu, Kangyang Zhou, Kang Zhou, Kai Fang, Jun Shi, Jinhao Dong, Jiebao Xiao, Jiaming Xu, Huaqiu Liu, Hongshen Xu, Heng Qu, Haochen Zhao, Hanglong Lv, Guoan Wang, Duo Zhang, Dong Zhang, Di Zhang, Chong Ma, Chang Liu, Can Cai, Bingquan Xia

Abstract:
We open-source MiMo-VL-7B-SFT and MiMo-VL-7B-RL, two powerful vision-language models delivering state-of-the-art performance in both general visual understanding and multimodal reasoning. MiMo-VL-7B-RL outperforms Qwen2.5-VL-7B on 35 out of 40 evaluated tasks, and scores 59.4 on OlympiadBench, surpassing models with up to 78B parameters. For GUI grounding applications, it sets a new standard with 56.1 on OSWorld-G, even outperforming specialized models such as UI-TARS. Our training combines four-stage pre-training (2.4 trillion tokens) with Mixed On-policy Reinforcement Learning (MORL) integrating diverse reward signals. We identify the importance of incorporating high-quality reasoning data with long Chain-of-Thought into pre-training stages, and the benefits of mixed RL despite challenges in simultaneous multi-domain optimization. We also contribute a comprehensive evaluation suite covering 50+ tasks to promote reproducibility and advance the field. The model checkpoints and full evaluation suite are available at https://github.com/XiaomiMiMo/MiMo-VL.

Paperid: 936, https://arxiv.org/pdf/2506.03517.pdf GitHub

Abstract:
Direct Preference Optimization (DPO) has recently been applied as a post-training technique for text-to-video diffusion models. To obtain training data, annotators are asked to provide preferences between two videos generated from independent noise. However, this approach prohibits fine-grained comparisons, and we point out that it biases the annotators towards low-motion clips as they often contain fewer visual artifacts. In this work, we introduce DenseDPO, a method that addresses these shortcomings by making three contributions. First, we create each video pair for DPO by denoising corrupted copies of a ground truth video. This results in aligned pairs with similar motion structures while differing in local details, effectively neutralizing the motion bias. Second, we leverage the resulting temporal alignment to label preferences on short segments rather than entire clips, yielding a denser and more precise learning signal. With only one-third of the labeled data, DenseDPO greatly improves motion generation over vanilla DPO, while matching it in text alignment, visual quality, and temporal consistency. Finally, we show that DenseDPO unlocks automatic preference annotation using off-the-shelf Vision Language Models (VLMs): GPT accurately predicts segment-level preferences similar to task-specifically fine-tuned video reward models, and DenseDPO trained on these labels achieves performance close to using human labels.

Paperid: 937, https://arxiv.org/pdf/2506.03350.pdf GitHub

Abstract:
The emergence of vision-language-action models (VLAs) for end-to-end control is reshaping the field of robotics by enabling the fusion of multimodal sensory inputs at the billion-parameter scale. The capabilities of VLAs stem primarily from their architectures, which are often based on frontier large language models (LLMs). However, LLMs are known to be susceptible to adversarial misuse, and given the significant physical risks inherent to robotics, questions remain regarding the extent to which VLAs inherit these vulnerabilities. Motivated by these concerns, in this work we initiate the study of adversarial attacks on VLA-controlled robots. Our main algorithmic contribution is the adaptation and application of LLM jailbreaking attacks to obtain complete control authority over VLAs. We find that textual attacks, which are applied once at the beginning of a rollout, facilitate full reachability of the action space of commonly used VLAs and often persist over longer horizons. This differs significantly from LLM jailbreaking literature, as attacks in the real world do not have to be semantically linked to notions of harm. We make all code available at https://github.com/eliotjones1/robogcg .

Paperid: 938, https://arxiv.org/pdf/2506.03270.pdf GitHub

Abstract:
While recent advances in vision-language models (VLMs) have accelerated the development of language-guided robot planners, their black-box nature often lacks safety guarantees and interpretability crucial for real-world deployment. Conversely, classical symbolic planners offer rigorous safety verification but require significant expert knowledge for setup. To bridge the current gap, this paper proposes ViLaIn-TAMP, a hybrid planning framework for enabling verifiable, interpretable, and autonomous robot behaviors. ViLaIn-TAMP comprises three main components: (1) ViLaIn (Vision-Language Interpreter) - A prior framework that converts multimodal inputs into structured problem specifications using off-the-shelf VLMs without additional domain-specific training, (2) a modular Task and Motion Planning (TAMP) system that grounds these specifications in actionable trajectory sequences through symbolic and geometric constraint reasoning and can utilize learning-based skills for key manipulation phases, and (3) a corrective planning module which receives concrete feedback on failed solution attempts from the motion and task planning components and can feed adapted logic and geometric feasibility constraints back to ViLaIn to improve and further refine the specification. We evaluate our framework on several challenging manipulation tasks in a cooking domain. We demonstrate that the proposed closed-loop corrective architecture exhibits a more than 30% higher mean success rate for ViLaIn-TAMP compared to without corrective planning.

Paperid: 939, https://arxiv.org/pdf/2506.03198.pdf GitHub GitHub GitHub GitHub

Abstract:
Action Quality Assessment (AQA) -- the task of quantifying how well an action is performed -- has great potential for detecting errors in gym weight training, where accurate feedback is critical to prevent injuries and maximize gains. Existing AQA datasets, however, are limited to single-view competitive sports and RGB video, lacking multimodal signals and professional assessment of fitness actions. We introduce FLEX, the first large-scale, multimodal, multiview dataset for fitness AQA that incorporates surface electromyography (sEMG). FLEX contains over 7,500 multiview recordings of 20 weight-loaded exercises performed by 38 subjects of diverse skill levels, with synchronized RGB video, 3D pose, sEMG, and physiological signals. Expert annotations are organized into a Fitness Knowledge Graph (FKG) linking actions, key steps, error types, and feedback, supporting a compositional scoring function for interpretable quality assessment. FLEX enables multimodal fusion, cross-modal prediction -- including the novel Video$\rightarrow$EMG task -- and biomechanically oriented representation learning. Building on the FKG, we further introduce FLEX-VideoQA, a structured question-answering benchmark with hierarchical queries that drive cross-modal reasoning in vision-language models. Baseline experiments demonstrate that multimodal inputs, multiview video, and fine-grained annotations significantly enhance AQA performance. FLEX thus advances AQA toward richer multimodal settings and provides a foundation for AI-powered fitness assessment and coaching. Dataset and code are available at \href{https://github.com/HaoYin116/FLEX}{https://github.com/HaoYin116/FLEX}. Link to Project \href{https://haoyin116.github.io/FLEX_Dataset}{page}.

Paperid: 940, https://arxiv.org/pdf/2506.03190.pdf GitHub

Abstract:
Improving the generalization ability of Vision-Language Pre-trained Models (VLMs) under test-time data distribution shifts remains a critical challenge. The existing Test-Time Adaptation (TTA) methods fall short in fully leveraging the model's internal knowledge, particularly in dynamically adapting to complex and hierarchical visual semantic information. In this paper, we propose Memory-Infused Prompt Tuning (MINT), a novel framework to address this issue. Inspired by human associative memory theory, MINT introduces a Memory Prompt Bank (MPB), which stores learnable key-value prompt pairs that work as a memory of previously seen samples. During the test time, relevant prompt pairs in the MPB are retrieved by the hierarchical visual features of test images to dynamically assemble Associative Prompts. The associative prompts are then injected into the image encoder for fine-grained, customized visual contextual guidance. MINT also utilizes learnable text prompts. MINT thus enables rapid, precise VLM adaptation at test time by leveraging this MPB-acquired memory, without source data or retraining. The code is available at https://github.com/Jamieyi2004/MINT.

Paperid: 941, https://arxiv.org/pdf/2506.03135.pdf GitHub

Abstract:
Spatial reasoning is a key aspect of cognitive psychology and remains a bottleneck for current vision-language models (VLMs). While extensive research has aimed to evaluate or improve VLMs' understanding of basic spatial relations, such as distinguishing left from right, near from far, and object counting, these tasks cover only the most elementary layer of spatial reasoning and are largely approaching saturation in the latest reasoning models. In this work, we introduce OmniSpatial, a comprehensive and challenging benchmark for spatial reasoning, grounded in cognitive psychology. OmniSpatial covers four major categories: dynamic reasoning, complex spatial logic, spatial interaction, and perspective-taking, with 50 fine-grained subcategories. Through careful manual annotation, we construct over 8.4K question-answer pairs. Extensive experiments show that both open- and closed-source VLMs exhibit significant limitations in comprehensive spatial reasoning. We also explore two strategies-PointGraph (explicit scene graph cues) and SpatialCoT (novel-view chain-of-thought)-to bolster spatial reasoning.

Paperid: 942, https://arxiv.org/pdf/2506.03097.pdf GitHub

Abstract:
Emerging embodied AI applications, such as wearable cameras and autonomous agents, have underscored the need for robust reasoning from first person video streams. We introduce EgoVLM, a vision-language model specifically designed to integrate visual comprehension and spatial-temporal reasoning within egocentric video contexts. EgoVLM is fine-tuned via Group Relative Policy Optimization (GRPO), a reinforcement learning method adapted to align model outputs with human-like reasoning steps. Following DeepSeek R1-Zero's approach, we directly tune using RL without any supervised fine-tuning phase on chain-of-thought (CoT) data. We evaluate EgoVLM on egocentric video question answering benchmarks and show that domain-specific training substantially improves performance over general-purpose VLMs. Our EgoVLM-3B, trained exclusively on non-CoT egocentric data, outperforms the base Qwen2.5-VL 3B and 7B models by 14.33 and 13.87 accuracy points on the EgoSchema benchmark, respectively. By explicitly generating reasoning traces, EgoVLM enhances interpretability, making it well-suited for downstream applications. Furthermore, we introduce a novel keyframe-based reward that incorporates salient frame selection to guide reinforcement learning optimization. This reward formulation opens a promising avenue for future exploration in temporally grounded egocentric reasoning.

Paperid: 943, https://arxiv.org/pdf/2506.02535.pdf GitHub

Abstract:
Video Anomaly Detection (VAD) methods based on reconstruction or prediction face two critical challenges: (1) strong generalization capability often results in accurate reconstruction or prediction of abnormal events, making it difficult to distinguish normal from abnormal patterns; (2) reliance only on low-level appearance and motion cues limits their ability to identify high-level semantic in abnormal events from complex scenes. To address these limitations, we propose a novel VAD framework with two key innovations. First, to suppress excessive generalization, we introduce the Sparse Feature Filtering Module (SFFM) that employs bottleneck filters to dynamically and adaptively remove abnormal information from features. Unlike traditional memory modules, it does not need to memorize the normal prototypes across the training dataset. Further, we design the Mixture of Experts (MoE) architecture for SFFM. Each expert is responsible for extracting specialized principal features during running time, and different experts are selectively activated to ensure the diversity of the learned principal features. Second, to overcome the neglect of semantics in existing methods, we integrate a Vision-Language Model (VLM) to generate textual descriptions for video clips, enabling comprehensive joint modeling of semantic, appearance, and motion cues. Additionally, we enforce modality consistency through semantic similarity constraints and motion frame-difference contrastive loss. Extensive experiments on multiple public datasets validate the effectiveness of our multimodal joint modeling framework and sparse feature filtering paradigm. Project page at https://qzfm.github.io/sfn_vad_project_page/.

Paperid: 944, https://arxiv.org/pdf/2506.02161.pdf GitHub

Abstract:
The rapid advancements of Text-to-Image (T2I) models have ushered in a new phase of AI-generated content, marked by their growing ability to interpret and follow user instructions. However, existing T2I model evaluation benchmarks fall short in limited prompt diversity and complexity, as well as coarse evaluation metrics, making it difficult to evaluate the fine-grained alignment performance between textual instructions and generated images. In this paper, we present TIIF-Bench (Text-to-Image Instruction Following Benchmark), aiming to systematically assess T2I models' ability in interpreting and following intricate textual instructions. TIIF-Bench comprises a set of 5000 prompts organized along multiple dimensions, which are categorized into three levels of difficulties and complexities. To rigorously evaluate model robustness to varying prompt lengths, we provide a short and a long version for each prompt with identical core semantics. Two critical attributes, i.e., text rendering and style control, are introduced to evaluate the precision of text synthesis and the aesthetic coherence of T2I models. In addition, we collect 100 high-quality designer level prompts that encompass various scenarios to comprehensively assess model performance. Leveraging the world knowledge encoded in large vision language models, we propose a novel computable framework to discern subtle variations in T2I model outputs. Through meticulous benchmarking of mainstream T2I models on TIIF-Bench, we analyze the pros and cons of current T2I models and reveal the limitations of current T2I benchmarks. Project Page: https://a113n-w3i.github.io/TIIF_Bench/.

Paperid: 945, https://arxiv.org/pdf/2506.01949.pdf GitHub

Abstract:
Recent diffusion models have advanced image editing by enhancing visual quality and control, supporting broad applications across creative and personalized domains. However, current image editing largely overlooks multi-object scenarios, where precise control over object categories, counts, and spatial layouts remains a significant challenge. To address this, we introduce a new task, quantity-and-layout consistent image editing (QL-Edit), which aims to enable fine-grained control of object quantity and spatial structure in complex scenes. We further propose IMAGHarmony, a structure-aware framework that incorporates harmony-aware attention (HA) to integrate multimodal semantics, explicitly modeling object counts and layouts to enhance editing accuracy and structural consistency. In addition, we observe that diffusion models are susceptible to initial noise and exhibit strong preferences for specific noise patterns. Motivated by this, we present a preference-guided noise selection (PNS) strategy that chooses semantically aligned initial noise samples based on vision-language matching, thereby improving generation stability and layout consistency in multi-object editing. To support evaluation, we construct HarmonyBench, a comprehensive benchmark covering diverse quantity and layout control scenarios. Extensive experiments demonstrate that IMAGHarmony consistently outperforms state-of-the-art methods in structural alignment and semantic accuracy. The code and model are available at https://github.com/muzishen/IMAGHarmony.

Paperid: 946, https://arxiv.org/pdf/2506.01908.pdf GitHub

Abstract:
Understanding real-world videos with complex semantics and long temporal dependencies remains a fundamental challenge in computer vision. Recent progress in multimodal large language models (MLLMs) has demonstrated strong capabilities in vision-language tasks, while reinforcement learning tuning (RLT) has further improved their reasoning abilities. In this work, we explore RLT as a post-training strategy to enhance the video-specific reasoning capabilities of MLLMs. Built upon the Group Relative Policy Optimization (GRPO) framework, we propose a dual-reward formulation that supervises both semantic and temporal reasoning through discrete and continuous reward signals. To facilitate effective preference-based optimization, we introduce a variance-aware data selection strategy based on repeated inference to identify samples that provide informative learning signals. We evaluate our approach across eight representative video understanding tasks, including VideoQA, Temporal Video Grounding, and Grounded VideoQA. Our method consistently outperforms supervised fine-tuning and existing RLT baselines, achieving superior performance with significantly less training data. These results underscore the importance of reward design and data selection in advancing reasoning-centric video understanding with MLLMs. Notably, The initial code release (two months ago) has now been expanded with updates, including optimized reward mechanisms and additional datasets. The latest version is available at https://github.com/appletea233/Temporal-R1 .

Paperid: 947, https://arxiv.org/pdf/2506.01844.pdf GitHub

Abstract:
Vision-language models (VLMs) pretrained on large-scale multimodal datasets encode rich visual and linguistic knowledge, making them a strong foundation for robotics. Rather than training robotic policies from scratch, recent approaches adapt VLMs into vision-language-action (VLA) models that enable natural language-driven perception and control. However, existing VLAs are typically massive--often with billions of parameters--leading to high training costs and limited real-world deployability. Moreover, they rely on academic and industrial datasets, overlooking the growing availability of community-collected data from affordable robotic platforms. In this work, we present SmolVLA, a small, efficient, and community-driven VLA that drastically reduces both training and inference costs, while retaining competitive performance. SmolVLA is designed to be trained on a single GPU and deployed on consumer-grade GPUs or even CPUs. To further improve responsiveness, we introduce an asynchronous inference stack decoupling perception and action prediction from action execution, allowing higher control rates with chunked action generation. Despite its compact size, SmolVLA achieves performance comparable to VLAs that are 10x larger. We evaluate SmolVLA on a range of both simulated as well as real-world robotic benchmarks and release all code, pretrained models, and training data.

Paperid: 948, https://arxiv.org/pdf/2506.01724.pdf GitHub

Abstract:
Pretrained on web-scale open data, VLMs offer powerful capabilities for solving downstream tasks after being adapted to task-specific labeled data. Yet, data labeling can be expensive and may demand domain expertise. Active Learning (AL) aims to reduce this expense by strategically selecting the most informative data for labeling and model training. Recent AL methods have explored VLMs but have not leveraged publicly available open data, such as VLM's pretraining data. In this work, we leverage such data by retrieving task-relevant examples to augment the task-specific examples. As expected, incorporating them significantly improves AL. Given that our method exploits open-source VLM and open data, we refer to it as Active Learning with Open Resources (ALOR). Additionally, most VLM-based AL methods use prompt tuning (PT) for model adaptation, likely due to its ability to directly utilize pretrained parameters and the assumption that doing so reduces the risk of overfitting to limited labeled data. We rigorously compare popular adaptation approaches, including linear probing (LP), finetuning (FT), and contrastive tuning (CT). We reveal two key findings: (1) All adaptation approaches benefit from incorporating retrieved data, and (2) CT resoundingly outperforms other approaches across AL methods. Further analysis of retrieved data reveals a naturally imbalanced distribution of task-relevant classes, exposing inherent biases within the VLM. This motivates our novel Tail First Sampling (TFS) strategy for AL, an embarrassingly simple yet effective method that prioritizes sampling data from underrepresented classes to label. Extensive experiments demonstrate that our final method, contrastively finetuning VLM on both retrieved and TFS-selected labeled data, significantly outperforms existing methods.

Paperid: 949, https://arxiv.org/pdf/2506.01701.pdf GitHub GitHub

Abstract:
In this paper, we present InfoMax, a novel data pruning method, also known as coreset selection, designed to maximize the information content of selected samples while minimizing redundancy. By doing so, InfoMax enhances the overall informativeness of the coreset. The information of individual samples is measured by importance scores, which capture their influence or difficulty in model learning. To quantify redundancy, we use pairwise sample similarities, based on the premise that similar samples contribute similarly to the learning process. We formalize the coreset selection problem as a discrete quadratic programming (DQP) task, with the objective of maximizing the total information content, represented as the sum of individual sample contributions minus the redundancies introduced by similar samples within the coreset. To ensure practical scalability, we introduce an efficient gradient-based solver, complemented by sparsification techniques applied to the similarity matrix and dataset partitioning strategies. This enables InfoMax to seamlessly scale to datasets with millions of samples. Extensive experiments demonstrate the superior performance of InfoMax in various data pruning tasks, including image classification, vision-language pre-training, and instruction tuning for large language models. Code is available at https://github.com/hrtan/InfoMax.

Paperid: 950, https://arxiv.org/pdf/2506.01608.pdf GitHub

Abstract:
Understanding behavior requires datasets that capture humans while carrying out complex tasks. The kitchen is an excellent environment for assessing human motor and cognitive function, as many complex actions are naturally exhibited in kitchens from chopping to cleaning. Here, we introduce the EPFL-Smart-Kitchen-30 dataset, collected in a noninvasive motion capture platform inside a kitchen environment. Nine static RGB-D cameras, inertial measurement units (IMUs) and one head-mounted HoloLens~2 headset were used to capture 3D hand, body, and eye movements. The EPFL-Smart-Kitchen-30 dataset is a multi-view action dataset with synchronized exocentric, egocentric, depth, IMUs, eye gaze, body and hand kinematics spanning 29.7 hours of 16 subjects cooking four different recipes. Action sequences were densely annotated with 33.78 action segments per minute. Leveraging this multi-modal dataset, we propose four benchmarks to advance behavior understanding and modeling through 1) a vision-language benchmark, 2) a semantic text-to-motion generation benchmark, 3) a multi-modal action recognition benchmark, 4) a pose-based action segmentation benchmark. We expect the EPFL-Smart-Kitchen-30 dataset to pave the way for better methods as well as insights to understand the nature of ecologically-valid human behavior. Code and data are available at https://github.com/amathislab/EPFL-Smart-Kitchen

Paperid: 951, https://arxiv.org/pdf/2506.01551.pdf GitHub

Abstract:
Building Vision-Language Navigation (VLN) agents which can navigate following natural language instructions is a long-standing goal in human-robot interaction applications. Recent studies have revealed the potential of training open-source Large Language Models (LLMs) to unleash LLMs' reasoning ability for improving navigation, and simultaneously mitigate the domain gap between LLMs' training corpus and the VLN task. However, these approaches primarily adopt direct input-output mapping paradigms, causing the mapping learning difficult and the navigational decisions unexplainable. Chain-of-Thought (CoT) training is a promising way to improve both navigational decision accuracy and interpretability, while the complexity of the navigation task makes the perfect CoT labels unavailable and may lead to overfitting through pure CoT supervised fine-tuning. In this paper, we propose a novel sElf-improving embodied reasoning framework for boosting LLM-based vision-language Navigation, dubbed EvolveNav. Our EvolveNav consists of two stages: (1) Formalized CoT Supervised Fine-Tuning, where we train the model with formalized CoT labels to both activate the model's navigational reasoning capabilities and increase the reasoning speed; (2) Self-Reflective Post-Training, where the model is iteratively trained with its own reasoning outputs as self-enriched CoT labels to enhance the supervision diversity. A self-reflective auxiliary task is also introduced to encourage learning correct reasoning patterns by contrasting with wrong ones. Experimental results on the popular VLN benchmarks demonstrate the superiority of EvolveNav over previous LLM-based VLN approaches. Code is available at https://github.com/expectorlin/EvolveNav.

Paperid: 952, https://arxiv.org/pdf/2506.01411.pdf GitHub

Abstract:
The Pedestrian Attribute Recognition (PAR) task aims to identify various detailed attributes of an individual, such as clothing, accessories, and gender. To enhance PAR performance, a model must capture features ranging from coarse-grained global attributes (e.g., for identifying gender) to fine-grained local details (e.g., for recognizing accessories) that may appear in diverse regions. Recent research suggests that body part representation can enhance the model's robustness and accuracy, but these methods are often restricted to attribute classes within fixed horizontal regions, leading to degraded performance when attributes appear in varying or unexpected body locations. In this paper, we propose Visual and Textual Attribute Alignment with Attribute Prompting for Pedestrian Attribute Recognition, dubbed as ViTA-PAR, to enhance attribute recognition through specialized multimodal prompting and vision-language alignment. We introduce visual attribute prompts that capture global-to-local semantics, enabling diverse attribute representations. To enrich textual embeddings, we design a learnable prompt template, termed person and attribute context prompting, to learn person and attributes context. Finally, we align visual and textual attribute features for effective fusion. ViTA-PAR is validated on four PAR benchmarks, achieving competitive performance with efficient inference. We release our code and model at https://github.com/mlnjeongpark/ViTA-PAR.

Paperid: 953, https://arxiv.org/pdf/2506.01064.pdf GitHub

Abstract:
Recent advances in large vision-language models (LVLMs) have showcased their remarkable capabilities across a wide range of multimodal vision-language tasks. However, these models remain vulnerable to visual adversarial attacks, which can substantially compromise their performance. In this paper, we introduce F3, a novel adversarial purification framework that employs a counterintuitive ``fighting fire with fire'' strategy: intentionally introducing simple perturbations to adversarial examples to mitigate their harmful effects. Specifically, F3 leverages cross-modal attentions derived from randomly perturbed adversary examples as reference targets. By injecting noise into these adversarial examples, F3 effectively refines their attention, resulting in cleaner and more reliable model outputs. Remarkably, this seemingly paradoxical approach of employing noise to counteract adversarial attacks yields impressive purification results. Furthermore, F3 offers several distinct advantages: it is training-free and straightforward to implement, and exhibits significant computational efficiency improvements compared to existing purification methods. These attributes render F3 particularly suitable for large-scale industrial applications where both robust performance and operational efficiency are critical priorities. The code is available at https://github.com/btzyd/F3.

Paperid: 954, https://arxiv.org/pdf/2506.00930.pdf GitHub

Abstract:
Vision-language models (VLMs) aligned with general human objectives, such as being harmless and hallucination-free, have become valuable assistants of humans in managing visual tasks. However, people with diversified backgrounds have different cognition even in the same situation. Consequently, they may have personalized expectations for VLM assistants. This highlights the urgent need to align VLM assistants with personalized situated cognition for real-world assistance. To study this problem, we first simplify it by characterizing individuals based on the sociological concept of Role-Set. Then, we propose to evaluate the individuals' actions to examine whether the personalized alignment is achieved. Further, we construct a benchmark named PCogAlignBench, which includes 18k instances and 20 individuals with different Role-Sets. Finally, we present a framework called PCogAlign, which constructs a cognition-aware and action-based reward model for personalized alignment. Experimental results and human evaluations demonstrate the reliability of the PCogAlignBench and the effectiveness of our proposed PCogAlign. We will open-source the constructed benchmark and code at https://github.com/NLPGM/PCogAlign.

Paperid: 955, https://arxiv.org/pdf/2506.00868.pdf GitHub GitHub

Abstract:
The rapid advancement of GenAI technology over the past few years has significantly contributed towards highly realistic deepfake content generation. Despite ongoing efforts, the research community still lacks a large-scale and reasoning capability driven deepfake benchmark dataset specifically tailored for person-centric object, context and scene manipulations. In this paper, we address this gap by introducing MultiFakeVerse, a large scale person-centric deepfake dataset, comprising 845,286 images generated through manipulation suggestions and image manipulations both derived from vision-language models (VLM). The VLM instructions were specifically targeted towards modifications to individuals or contextual elements of a scene that influence human perception of importance, intent, or narrative. This VLM-driven approach enables semantic, context-aware alterations such as modifying actions, scenes, and human-object interactions rather than synthetic or low-level identity swaps and region-specific edits that are common in existing datasets. Our experiments reveal that current state-of-the-art deepfake detection models and human observers struggle to detect these subtle yet meaningful manipulations. The code and dataset are available on \href{https://github.com/Parul-Gupta/MultiFakeVerse}{GitHub}.

Paperid: 956, https://arxiv.org/pdf/2506.00835.pdf GitHub

Abstract:
Fine-grained video captioning aims to generate detailed, temporally coherent descriptions of video content. However, existing methods struggle to capture subtle video dynamics and rich detailed information. In this paper, we leverage preference learning to enhance the performance of vision-language models in fine-grained video captioning, while mitigating several limitations inherent to direct preference optimization (DPO). First, we propose a pipeline for constructing preference pairs that leverages the intrinsic properties of VLMs along with partial assistance from large language models, achieving an optimal balance between cost and data quality. Second, we propose Synergistic Preference Optimization (SynPO), a novel optimization method offering significant advantages over DPO and its variants. SynPO prevents negative preferences from dominating the optimization, explicitly preserves the model's language capability to avoid deviation of the optimization objective, and improves training efficiency by eliminating the need for the reference model. We extensively evaluate SynPO not only on video captioning benchmarks (e.g., VDC, VDD, VATEX) but also across well-established NLP tasks, including general language understanding and preference evaluation, using diverse pretrained models. Results demonstrate that SynPO consistently outperforms DPO variants while achieving 20\% improvement in training efficiency. Code is available at https://github.com/longmalongma/SynPO

Paperid: 957, https://arxiv.org/pdf/2506.00302.pdf GitHub

Abstract:
Most materials science datasets are limited to atomic geometries (e.g., XYZ files), restricting their utility for multimodal learning and comprehensive data-centric analysis. These constraints have historically impeded the adoption of advanced machine learning techniques in the field. This work introduces MultiCrystalSpectrumSet (MCS-Set), a curated framework that expands materials datasets by integrating atomic structures with 2D projections and structured textual annotations, including lattice parameters and coordination metrics. MCS-Set enables two key tasks: (1) multimodal property and summary prediction, and (2) constrained crystal generation with partial cluster supervision. Leveraging a human-in-the-loop pipeline, MCS-Set combines domain expertise with standardized descriptors for high-quality annotation. Evaluations using state-of-the-art language and vision-language models reveal substantial modality-specific performance gaps and highlight the importance of annotation quality for generalization. MCS-Set offers a foundation for benchmarking multimodal models, advancing annotation practices, and promoting accessible, versatile materials science datasets. The dataset and implementations are available at https://github.com/KurbanIntelligenceLab/MultiCrystalSpectrumSet.

Paperid: 958, https://arxiv.org/pdf/2505.24872.pdf GitHub

Abstract:
Recent advancements in reinforcement learning with verifiable rewards have pushed the boundaries of the visual reasoning capabilities in large vision-language models (LVLMs). However, training LVLMs with reinforcement fine-tuning (RFT) is computationally expensive, posing a significant challenge to scaling model size. In this work, we propose ProxyThinker, an inference-time technique that enables large models to inherit the visual reasoning capabilities from small, slow-thinking visual reasoners without any training. By subtracting the output distributions of base models from those of RFT reasoners, ProxyThinker modifies the decoding dynamics and successfully elicits the slow-thinking reasoning demonstrated by the emerged sophisticated behaviors such as self-verification and self-correction. ProxyThinker consistently boosts performance on challenging visual benchmarks on spatial, mathematical, and multi-disciplinary reasoning, enabling untuned base models to compete with the performance of their full-scale RFT counterparts. Furthermore, our implementation efficiently coordinates multiple language models with parallelism techniques and achieves up to 38 $\times$ faster inference compared to previous decoding-time methods, paving the way for the practical deployment of ProxyThinker. Code is available at https://github.com/MrZilinXiao/ProxyThinker.

Paperid: 959, https://arxiv.org/pdf/2505.24867.pdf GitHub

Abstract:
Recent advances in vision-language models (VLMs) have made impressive strides in understanding spatio-temporal relationships in videos. However, when spatial information is obscured, these models struggle to capture purely temporal patterns. We introduce $\textbf{SpookyBench}$, a benchmark where information is encoded solely in temporal sequences of noise-like frames, mirroring natural phenomena from biological signaling to covert communication. Interestingly, while humans can recognize shapes, text, and patterns in these sequences with over 98% accuracy, state-of-the-art VLMs achieve 0% accuracy. This performance gap highlights a critical limitation: an over-reliance on frame-level spatial features and an inability to extract meaning from temporal cues. Furthermore, when trained in data sets with low spatial signal-to-noise ratios (SNR), temporal understanding of models degrades more rapidly than human perception, especially in tasks requiring fine-grained temporal reasoning. Overcoming this limitation will require novel architectures or training paradigms that decouple spatial dependencies from temporal processing. Our systematic analysis shows that this issue persists across model scales and architectures. We release SpookyBench to catalyze research in temporal pattern recognition and bridge the gap between human and machine video understanding. Dataset and code has been made available on our project website: https://timeblindness.github.io/.

Paperid: 960, https://arxiv.org/pdf/2505.24693.pdf GitHub

Abstract:
Vision-language models pre-trained at large scale have shown unprecedented adaptability and generalization to downstream tasks. Although its discriminative potential has been widely explored, its reliability and uncertainty are still overlooked. In this work, we investigate the capabilities of CLIP models under the split conformal prediction paradigm, which provides theoretical guarantees to black-box models based on a small, labeled calibration set. In contrast to the main body of literature on conformal predictors in vision classifiers, foundation models exhibit a particular characteristic: they are pre-trained on a one-time basis on an inaccessible source domain, different from the transferred task. This domain drift negatively affects the efficiency of the conformal sets and poses additional challenges. To alleviate this issue, we propose Conf-OT, a transfer learning setting that operates transductive over the combined calibration and query sets. Solving an optimal transport problem, the proposed method bridges the domain gap between pre-training and adaptation without requiring additional data splits but still maintaining coverage guarantees. We comprehensively explore this conformal prediction strategy on a broad span of 15 datasets and three non-conformity scores. Conf-OT provides consistent relative improvements of up to 20% on set efficiency while being 15 times faster than popular transductive approaches.

Paperid: 961, https://arxiv.org/pdf/2505.24423.pdf GitHub

Abstract:
Large language models and vision-language models (which we jointly call LMs) have transformed NLP and CV, demonstrating remarkable potential across various fields. However, their capabilities in affective analysis (i.e. sentiment analysis and emotion detection) remain underexplored. This gap is largely due to the absence of comprehensive evaluation benchmarks, and the inherent complexity of affective analysis tasks. In this paper, we introduce MMAFFBen, the first extensive open-source benchmark for multilingual multimodal affective analysis. MMAFFBen encompasses text, image, and video modalities across 35 languages, covering four key affective analysis tasks: sentiment polarity, sentiment intensity, emotion classification, and emotion intensity. Moreover, we construct the MMAFFIn dataset for fine-tuning LMs on affective analysis tasks, and further develop MMAFFLM-3b and MMAFFLM-7b based on it. We evaluate various representative LMs, including GPT-4o-mini, providing a systematic comparison of their affective understanding capabilities. This project is available at https://github.com/lzw108/MMAFFBen.

Paperid: 962, https://arxiv.org/pdf/2505.23883.pdf GitHub

Abstract:
Foundation models trained at scale exhibit remarkable emergent behaviors, learning new capabilities beyond their initial training objectives. We find such emergent behaviors in biological vision models via large-scale contrastive vision-language training. To achieve this, we first curate TreeOfLife-200M, comprising 214 million images of living organisms, the largest and most diverse biological organism image dataset to date. We then train BioCLIP 2 on TreeOfLife-200M to distinguish different species. Despite the narrow training objective, BioCLIP 2 yields extraordinary accuracy when applied to various biological visual tasks such as habitat classification and trait prediction. We identify emergent properties in the learned embedding space of BioCLIP 2. At the inter-species level, the embedding distribution of different species aligns closely with functional and ecological meanings (e.g., beak sizes and habitats). At the intra-species level, instead of being diminished, the intra-species variations (e.g., life stages and sexes) are preserved and better separated in subspaces orthogonal to inter-species distinctions. We provide formal proof and analyses to explain why hierarchical supervision and contrastive objectives encourage these emergent properties. Crucially, our results reveal that these properties become increasingly significant with larger-scale training data, leading to a biologically meaningful embedding space.

Paperid: 963, https://arxiv.org/pdf/2505.23766.pdf GitHub

Abstract:
Recent advances in multimodal large language models (MLLMs) have demonstrated remarkable capabilities in vision-language tasks, yet they often struggle with vision-centric scenarios where precise visual focus is needed for accurate reasoning. In this paper, we introduce Argus to address these limitations with a new visual attention grounding mechanism. Our approach employs object-centric grounding as visual chain-of-thought signals, enabling more effective goal-conditioned visual attention during multimodal reasoning tasks. Evaluations on diverse benchmarks demonstrate that Argus excels in both multimodal reasoning tasks and referring object grounding tasks. Extensive analysis further validates various design choices of Argus, and reveals the effectiveness of explicit language-guided visual region-of-interest engagement in MLLMs, highlighting the importance of advancing multimodal intelligence from a visual-centric perspective. Project page: https://yunzeman.github.io/argus/

Paperid: 964, https://arxiv.org/pdf/2505.23762.pdf GitHub

Abstract:
The rapid advancement of large Vision-Language Models (VLMs) has propelled the development of pure-vision-based GUI Agents, capable of perceiving and operating Graphical User Interfaces (GUI) to autonomously fulfill user instructions. However, existing approaches usually adopt an offline learning framework, which faces two core limitations: (1) heavy reliance on high-quality manual annotations for element grounding and action supervision, and (2) limited adaptability to dynamic and interactive environments. To address these limitations, we propose ZeroGUI, a scalable, online learning framework for automating GUI Agent training at Zero human cost. Specifically, ZeroGUI integrates (i) VLM-based automatic task generation to produce diverse training goals from the current environment state, (ii) VLM-based automatic reward estimation to assess task success without hand-crafted evaluation functions, and (iii) two-stage online reinforcement learning to continuously interact with and learn from GUI environments. Experiments on two advanced GUI Agents (UI-TARS and Aguvis) demonstrate that ZeroGUI significantly boosts performance across OSWorld and AndroidLab environments. The code is available at https://github.com/OpenGVLab/ZeroGUI.

Paperid: 965, https://arxiv.org/pdf/2505.23757.pdf GitHub GitHub

Abstract:
Vision-Language-Action (VLA) models for autonomous driving show promise but falter in unstructured corner case scenarios, largely due to a scarcity of targeted benchmarks. To address this, we introduce Impromptu VLA. Our core contribution is the Impromptu VLA Dataset: over 80,000 meticulously curated video clips, distilled from over 2M source clips sourced from 8 open-source large-scale datasets. This dataset is built upon our novel taxonomy of four challenging unstructured categories and features rich, planning-oriented question-answering annotations and action trajectories. Crucially, experiments demonstrate that VLAs trained with our dataset achieve substantial performance gains on established benchmarks--improving closed-loop NeuroNCAP scores and collision rates, and reaching near state-of-the-art L2 accuracy in open-loop nuScenes trajectory prediction. Furthermore, our Q&A suite serves as an effective diagnostic, revealing clear VLM improvements in perception, prediction, and planning. Our code, data and models are available at https://github.com/ahydchh/Impromptu-VLA.

Paperid: 966, https://arxiv.org/pdf/2505.23745.pdf GitHub

Abstract:
Vision-Language Models (VLMs) have demonstrated strong capabilities in aligning visual and textual modalities, enabling a wide range of applications in multimodal understanding and generation. While they excel in zero-shot and transfer learning scenarios, VLMs remain susceptible to misclassification, often yielding confident yet incorrect predictions. This limitation poses a significant risk in safety-critical domains, where erroneous predictions can lead to severe consequences. In this work, we introduce TrustVLM, a training-free framework designed to address the critical challenge of estimating when VLM's predictions can be trusted. Motivated by the observed modality gap in VLMs and the insight that certain concepts are more distinctly represented in the image embedding space, we propose a novel confidence-scoring function that leverages this space to improve misclassification detection. We rigorously evaluate our approach across 17 diverse datasets, employing 4 architectures and 2 VLMs, and demonstrate state-of-the-art performance, with improvements of up to 51.87% in AURC, 9.14% in AUROC, and 32.42% in FPR95 compared to existing baselines. By improving the reliability of the model without requiring retraining, TrustVLM paves the way for safer deployment of VLMs in real-world applications. The code is available at https://github.com/EPFL-IMOS/TrustVLM.

Paperid: 967, https://arxiv.org/pdf/2505.23704.pdf GitHub

Abstract:
VOT remains a fundamental yet challenging task in computer vision due to dynamic appearance changes, occlusions, and background clutter. Traditional trackers, relying primarily on visual cues, often struggle in such complex scenarios. Recent advancements in VLMs have shown promise in semantic understanding for tasks like open-vocabulary detection and image captioning, suggesting their potential for VOT. However, the direct application of VLMs to VOT is hindered by critical limitations: the absence of a rich and comprehensive textual representation that semantically captures the target object's nuances, limiting the effective use of language information; inefficient fusion mechanisms that fail to optimally integrate visual and textual features, preventing a holistic understanding of the target; and a lack of temporal modeling of the target's evolving appearance in the language domain, leading to a disconnect between the initial description and the object's subsequent visual changes. To bridge these gaps and unlock the full potential of VLMs for VOT, we propose CLDTracker, a novel Comprehensive Language Description framework for robust visual Tracking. Our tracker introduces a dual-branch architecture consisting of a textual and a visual branch. In the textual branch, we construct a rich bag of textual descriptions derived by harnessing the powerful VLMs such as CLIP and GPT-4V, enriched with semantic and contextual cues to address the lack of rich textual representation. Experiments on six standard VOT benchmarks demonstrate that CLDTracker achieves SOTA performance, validating the effectiveness of leveraging robust and temporally-adaptive vision-language representations for tracking. Code and models are publicly available at: https://github.com/HamadYA/CLDTracker

Paperid: 968, https://arxiv.org/pdf/2505.23606.pdf GitHub

Abstract:
Unified generation models aim to handle diverse tasks across modalities -- such as text generation, image generation, and vision-language reasoning -- within a single architecture and decoding paradigm. Autoregressive unified models suffer from slow inference due to sequential decoding, and non-autoregressive unified models suffer from weak generalization due to limited pretrained backbones. We introduce Muddit, a unified discrete diffusion transformer that enables fast and parallel generation across both text and image modalities. Unlike prior unified diffusion models trained from scratch, Muddit integrates strong visual priors from a pretrained text-to-image backbone with a lightweight text decoder, enabling flexible and high-quality multimodal generation under a unified architecture. Empirical results show that Muddit achieves competitive or superior performance compared to significantly larger autoregressive models in both quality and efficiency. The work highlights the potential of purely discrete diffusion, when equipped with strong visual priors, as a scalable and effective backbone for unified generation.

Paperid: 969, https://arxiv.org/pdf/2505.23566.pdf GitHub

Abstract:
Handwritten Mathematical Expression Recognition (HMER) remains a persistent challenge in Optical Character Recognition (OCR) due to the inherent freedom of symbol layout and variability in handwriting styles. Prior methods have faced performance bottlenecks, proposing isolated architectural modifications that are difficult to integrate coherently into a unified framework. Meanwhile, recent advances in pretrained vision-language models (VLMs) have demonstrated strong cross-task generalization, offering a promising foundation for developing unified solutions. In this paper, we introduce Uni-MuMER, which fully fine-tunes a VLM for the HMER task without modifying its architecture, effectively injecting domain-specific knowledge into a generalist framework. Our method integrates three data-driven tasks: Tree-Aware Chain-of-Thought (Tree-CoT) for structured spatial reasoning, Error-Driven Learning (EDL) for reducing confusion among visually similar characters, and Symbol Counting (SC) for improving recognition consistency in long expressions. Experiments on the CROHME and HME100K datasets show that Uni-MuMER achieves new state-of-the-art performance, surpassing the best lightweight specialized model SSAN by 16.31% and the top-performing VLM Gemini2.5-flash by 24.42% in the zero-shot setting. Our datasets, models, and code are open-sourced at: https://github.com/BFlameSwift/Uni-MuMER

Paperid: 970, https://arxiv.org/pdf/2505.23558.pdf GitHub

Abstract:
Inference time scaling drives extended reasoning to enhance the performance of Vision-Language Models (VLMs), thus forming powerful Vision-Language Reasoning Models (VLRMs). However, long reasoning dilutes visual tokens, causing visual information to receive less attention and may trigger hallucinations. Although introducing text-only reflection processes shows promise in language models, we demonstrate that it is insufficient to suppress hallucinations in VLMs. To address this issue, we introduce Qwen-LookAgain (Qwen-LA), a novel VLRM designed to mitigate hallucinations by incorporating a vision-text reflection process that guides the model to re-attention visual information during reasoning. We first propose a reinforcement learning method Balanced Reflective Policy Optimization (BRPO), which guides the model to decide when to generate vision-text reflection on its own and balance the number and length of reflections. Then, we formally prove that VLRMs lose attention to visual tokens as reasoning progresses, and demonstrate that supplementing visual information during reflection enhances visual attention. Therefore, during training and inference, Visual Token COPY and Visual Token ROUTE are introduced to force the model to re-attention visual information at the visual level, addressing the limitations of text-only reflection. Experiments on multiple visual QA datasets and hallucination metrics indicate that Qwen-LA achieves leading accuracy performance while reducing hallucinations. Our code is available at: https://github.com/Liar406/Look_Again

Paperid: 971, https://arxiv.org/pdf/2505.23484.pdf GitHub

Abstract:
Video captions play a crucial role in text-to-video generation tasks, as their quality directly influences the semantic coherence and visual fidelity of the generated videos. Although large vision-language models (VLMs) have demonstrated significant potential in caption generation, existing benchmarks inadequately address fine-grained evaluation, particularly in capturing spatial-temporal details critical for video generation. To address this gap, we introduce the Fine-grained Video Caption Evaluation Benchmark (VCapsBench), the first large-scale fine-grained benchmark comprising 5,677 (5K+) videos and 109,796 (100K+) question-answer pairs. These QA-pairs are systematically annotated across 21 fine-grained dimensions (e.g., camera movement, and shot type) that are empirically proven critical for text-to-video generation. We further introduce three metrics (Accuracy (AR), Inconsistency Rate (IR), Coverage Rate (CR)), and an automated evaluation pipeline leveraging large language model (LLM) to verify caption quality via contrastive QA-pairs analysis. By providing actionable insights for caption optimization, our benchmark can advance the development of robust text-to-video models. The dataset and codes are available at website: https://github.com/GXYM/VCapsBench.

Paperid: 972, https://arxiv.org/pdf/2505.23271.pdf GitHub

Abstract:
Continual learning with vision-language models like CLIP offers a pathway toward scalable machine learning systems by leveraging its transferable representations. Existing CLIP-based methods adapt the pre-trained image encoder by adding multiple sets of learnable parameters, with each task using a partial set of parameters. This requires selecting the expected parameters for input images during inference, which is prone to error that degrades performance. To address this problem, we introduce LADA (Label-specific ADApter). Instead of partitioning parameters across tasks, LADA appends lightweight, label-specific memory units to the frozen CLIP image encoder, enabling discriminative feature generation by aggregating task-agnostic knowledge. To prevent catastrophic forgetting, LADA employs feature distillation for seen classes, preventing their features from being interfered with by new classes. Positioned after the image encoder, LADA prevents gradient flow to the frozen CLIP parameters, ensuring efficient training. Extensive results show that LADA achieves state-of-the-art performance in continual learning settings. The implementation code is available at https://github.com/MaolinLuo/LADA.

Paperid: 973, https://arxiv.org/pdf/2505.23189.pdf GitHub

Abstract:
Embodied visual tracking is a fundamental skill in Embodied AI, enabling an agent to follow a specific target in dynamic environments using only egocentric vision. This task is inherently challenging as it requires both accurate target recognition and effective trajectory planning under conditions of severe occlusion and high scene dynamics. Existing approaches typically address this challenge through a modular separation of recognition and planning. In this work, we propose TrackVLA, a Vision-Language-Action (VLA) model that learns the synergy between object recognition and trajectory planning. Leveraging a shared LLM backbone, we employ a language modeling head for recognition and an anchor-based diffusion model for trajectory planning. To train TrackVLA, we construct an Embodied Visual Tracking Benchmark (EVT-Bench) and collect diverse difficulty levels of recognition samples, resulting in a dataset of 1.7 million samples. Through extensive experiments in both synthetic and real-world environments, TrackVLA demonstrates SOTA performance and strong generalizability. It significantly outperforms existing methods on public benchmarks in a zero-shot manner while remaining robust to high dynamics and occlusion in real-world scenarios at 10 FPS inference speed. Our project page is: https://pku-epic.github.io/TrackVLA-web.

Paperid: 974, https://arxiv.org/pdf/2505.23143.pdf GitHub

Abstract:
Artificial intelligence (AI)-based chest X-ray (CXR) interpretation assistants have demonstrated significant progress and are increasingly being applied in clinical settings. However, contemporary medical AI models often adhere to a simplistic input-to-output paradigm, directly processing an image and an instruction to generate a result, where the instructions may be integral to the model's architecture. This approach overlooks the modeling of the inherent diagnostic reasoning in chest X-ray interpretation. Such reasoning is typically sequential, where each interpretive stage considers the images, the current task, and the contextual information from previous stages. This oversight leads to several shortcomings, including misalignment with clinical scenarios, contextless reasoning, and untraceable errors. To fill this gap, we construct CXRTrek, a new multi-stage visual question answering (VQA) dataset for CXR interpretation. The dataset is designed to explicitly simulate the diagnostic reasoning process employed by radiologists in real-world clinical settings for the first time. CXRTrek covers 8 sequential diagnostic stages, comprising 428,966 samples and over 11 million question-answer (Q&A) pairs, with an average of 26.29 Q&A pairs per sample. Building on the CXRTrek dataset, we propose a new vision-language large model (VLLM), CXRTrekNet, specifically designed to incorporate the clinical reasoning flow into the VLLM framework. CXRTrekNet effectively models the dependencies between diagnostic stages and captures reasoning patterns within the radiological context. Trained on our dataset, the model consistently outperforms existing medical VLLMs on the CXRTrek benchmarks and demonstrates superior generalization across multiple tasks on five diverse external datasets. The dataset and model can be found in our repository (https://github.com/guanjinquan/CXRTrek).

Paperid: 975, https://arxiv.org/pdf/2505.23010.pdf GitHub

Abstract:
High-resolution (HR) remote sensing imagery plays a vital role in a wide range of applications, including urban planning and environmental monitoring. However, due to limitations in sensors and data transmission links, the images acquired in practice often suffer from resolution degradation. Remote Sensing Image Super-Resolution (RSISR) aims to reconstruct HR images from low-resolution (LR) inputs, providing a cost-effective and efficient alternative to direct HR image acquisition. Existing RSISR methods primarily focus on low-level characteristics in pixel space, while neglecting the high-level understanding of remote sensing scenes. This may lead to semantically inconsistent artifacts in the reconstructed results. Motivated by this observation, our work aims to explore the role of high-level semantic knowledge in improving RSISR performance. We propose a Semantic-Guided Super-Resolution framework, SeG-SR, which leverages Vision-Language Models (VLMs) to extract semantic knowledge from input images and uses it to guide the super resolution (SR) process. Specifically, we first design a Semantic Feature Extraction Module (SFEM) that utilizes a pretrained VLM to extract semantic knowledge from remote sensing images. Next, we propose a Semantic Localization Module (SLM), which derives a series of semantic guidance from the extracted semantic knowledge. Finally, we develop a Learnable Modulation Module (LMM) that uses semantic guidance to modulate the features extracted by the SR network, effectively incorporating high-level scene understanding into the SR pipeline. We validate the effectiveness and generalizability of SeG-SR through extensive experiments: SeG-SR achieves state-of-the-art performance on two datasets and consistently delivers performance improvements across various SR architectures. Codes can be found at https://github.com/Mr-Bamboo/SeG-SR.

Paperid: 976, https://arxiv.org/pdf/2505.22946.pdf GitHub

Abstract:
Negation is a fundamental linguistic phenomenon that can entirely reverse the meaning of a sentence. As vision language models (VLMs) continue to advance and are deployed in high-stakes applications, assessing their ability to comprehend negation becomes essential. To address this, we introduce NegVQA, a visual question answering (VQA) benchmark consisting of 7,379 two-choice questions covering diverse negation scenarios and image-question distributions. We construct NegVQA by leveraging large language models to generate negated versions of questions from existing VQA datasets. Evaluating 20 state-of-the-art VLMs across seven model families, we find that these models struggle significantly with negation, exhibiting a substantial performance drop compared to their responses to the original questions. Furthermore, we uncover a U-shaped scaling trend, where increasing model size initially degrades performance on NegVQA before leading to improvements. Our benchmark reveals critical gaps in VLMs' negation understanding and offers insights into future VLM development. Project page available at https://yuhui-zh15.github.io/NegVQA/.

Paperid: 977, https://arxiv.org/pdf/2505.22805.pdf GitHub

Abstract:
In order to navigate safely and reliably in off-road and unstructured environments, robots must detect anomalies that are out-of-distribution (OOD) with respect to the training data. We present an analysis-by-synthesis approach for pixel-wise anomaly detection without making any assumptions about the nature of OOD data. Given an input image, we use a generative diffusion model to synthesize an edited image that removes anomalies while keeping the remaining image unchanged. Then, we formulate anomaly detection as analyzing which image segments were modified by the diffusion model. We propose a novel inference approach for guided diffusion by analyzing the ideal guidance gradient and deriving a principled approximation that bootstraps the diffusion model to predict guidance gradients. Our editing technique is purely test-time that can be integrated into existing workflows without the need for retraining or fine-tuning. Finally, we use a combination of vision-language foundation models to compare pixels in a learned feature space and detect semantically meaningful edits, enabling accurate anomaly detection for off-road navigation. Project website: https://siddancha.github.io/anomalies-by-diffusion-synthesis/

Paperid: 978, https://arxiv.org/pdf/2505.22664.pdf GitHub

Abstract:
Vision language models (VLMs) typically pair a modestly sized vision encoder with a large language model (LLM), e.g., Llama-70B, making the decoder the primary computational burden during training. To reduce costs, a potential promising strategy is to first train the vision encoder using a small language model before transferring it to the large one. We construct small "surrogate models" that share the same embedding space and representation language as the large target LLM by directly inheriting its shallow layers. Vision encoders trained on the surrogate can then be directly transferred to the larger model, a process we call zero-shot grafting -- when plugged directly into the full-size target LLM, the grafted pair surpasses the encoder-surrogate pair and, on some benchmarks, even performs on par with full decoder training with the target LLM. Furthermore, our surrogate training approach reduces overall VLM training costs by ~45% when using Llama-70B as the decoder. The code is at https://github.com/facebookresearch/zero.

Paperid: 979, https://arxiv.org/pdf/2505.22663.pdf GitHub

Abstract:
Stylized abstraction synthesizes visually exaggerated yet semantically faithful representations of subjects, balancing recognizability with perceptual distortion. Unlike image-to-image translation, which prioritizes structural fidelity, stylized abstraction demands selective retention of identity cues while embracing stylistic divergence, especially challenging for out-of-distribution individuals. We propose a training-free framework that generates stylized abstractions from a single image using inference-time scaling in vision-language models (VLLMs) to extract identity-relevant features, and a novel cross-domain rectified flow inversion strategy that reconstructs structure based on style-dependent priors. Our method adapts structural restoration dynamically through style-aware temporal scheduling, enabling high-fidelity reconstructions that honor both subject and style. It supports multi-round abstraction-aware generation without fine-tuning. To evaluate this task, we introduce StyleBench, a GPT-based human-aligned metric suited for abstract styles where pixel-level similarity fails. Experiments across diverse abstraction (e.g., LEGO, knitted dolls, South Park) show strong generalization to unseen identities and styles in a fully open-source setup.

Paperid: 980, https://arxiv.org/pdf/2505.22654.pdf GitHub

Abstract:
Recent Large Vision-Language Models (LVLMs) have advanced multi-modal understanding by incorporating finer-grained visual perception and encoding. However, such methods incur significant computational costs due to longer visual token sequences, posing challenges for real-time deployment. To mitigate this, prior studies have explored pruning unimportant visual tokens either at the output layer of the visual encoder or at the early layers of the language model. In this work, we revisit these design choices and reassess their effectiveness through comprehensive empirical studies of how visual tokens are processed throughout the visual encoding and language decoding stages. Guided by these insights, we propose VScan, a two-stage visual token reduction framework that addresses token redundancy by: (1) integrating complementary global and local scans with token merging during visual encoding, and (2) introducing pruning at intermediate layers of the language model. Extensive experimental results across four LVLMs validate the effectiveness of VScan in accelerating inference and demonstrate its superior performance over current state-of-the-arts on sixteen benchmarks. Notably, when applied to LLaVA-NeXT-7B, VScan achieves a 2.91$\times$ speedup in prefilling and a 10$\times$ reduction in FLOPs, while retaining 95.4\% of the original performance. Code is available at https://github.com/Tencent/SelfEvolvingAgent/tree/main/VScan.

Paperid: 981, https://arxiv.org/pdf/2505.21979.pdf GitHub

Authors:Fakhraddin Alwajih, Samar Mohamed Magdy, Abdellah El Mekki, Omer Nacar, Youssef Nafea, Safaa Taher Abdelfadil, Abdulfattah Mohammed Yahya, Hamzah Luqman, Nada Almarwani, Samah Aloufi, Baraah Qawasmeh, Houdaifa Atou, Serry Sibaee, Hamzah A. Alsayadi, Walid Al-Dhabyani, Maged S. Al-shaibani, Aya El aatar, Nour Qandos, Rahaf Alhamouri, Samar Ahmad, Razan Khassib, Lina Hamad, Mohammed Anwar AL-Ghrawi, Fatimah Alshamari, Cheikh Malainine, Doaa Qawasmeh, Aminetou Yacoub, Tfeil moilid, Ruwa AbuHweidi, Ahmed Aboeitta, Vatimetou Mohamed Lemin, Reem Abdel-Salam, Ahlam Bashiti, Adel Ammar, Aisha Alansari, Ahmed Ashraf, Nora Alturayeif, Sara Shatnawi, Alcides Alcoba Inciarte, AbdelRahim A. Elmadany, Mohamedou cheikh tourad, Ismail Berrada, Mustafa Jarrar, Shady Shehata, Muhammad Abdul-Mageed

Abstract:
Mainstream large vision-language models (LVLMs) inherently encode cultural biases, highlighting the need for diverse multimodal datasets. To address this gap, we introduce Pearl, a large-scale Arabic multimodal dataset and benchmark explicitly designed for cultural understanding. Constructed through advanced agentic workflows and extensive human-in-the-loop annotations by 45 annotators from across the Arab world, Pearl comprises over K multimodal examples spanning ten culturally significant domains covering all Arab countries. We further provide two robust evaluation benchmarks Pearl and Pearl-Lite along with a specialized subset Pearl-X explicitly developed to assess nuanced cultural variations. Comprehensive evaluations on state-of-the-art open and proprietary LVLMs demonstrate that reasoning-centric instruction alignment substantially improves models' cultural grounding compared to conventional scaling methods. Pearl establishes a foundational resource for advancing culturally-informed multimodal modeling research. All datasets and benchmarks are publicly available.

Paperid: 982, https://arxiv.org/pdf/2505.21698.pdf GitHub

Abstract:
Recent vision-language foundation models deliver state-of-the-art results on natural image classification but falter on medical images due to pronounced domain shifts. At the same time, training a medical foundation model requires substantial resources, including extensive annotated data and high computational capacity. To bridge this gap with minimal overhead, we introduce MedBridge, a lightweight multimodal adaptation framework that re-purposes pretrained VLMs for accurate medical image diagnosis. MedBridge comprises three key components. First, a Focal Sampling module that extracts high-resolution local regions to capture subtle pathological features and compensate for the limited input resolution of general-purpose VLMs. Second, a Query Encoder (QEncoder) injects a small set of learnable queries that attend to the frozen feature maps of VLM, aligning them with medical semantics without retraining the entire backbone. Third, a Mixture of Experts mechanism, driven by learnable queries, harnesses the complementary strength of diverse VLMs to maximize diagnostic performance. We evaluate MedBridge on five medical imaging benchmarks across three key adaptation tasks, demonstrating its superior performance in both cross-domain and in-domain adaptation settings, even under varying levels of training data availability. Notably, MedBridge achieved over 6-15% improvement in AUC compared to state-of-the-art VLM adaptation methods in multi-label thoracic disease diagnosis, underscoring its effectiveness in leveraging foundation models for accurate and data-efficient medical diagnosis. Our code is available at https://github.com/ai-med/MedBridge.

Paperid: 983, https://arxiv.org/pdf/2505.21547.pdf GitHub

Abstract:
Large Vision-Language Models (LVLMs) with discrete image tokenizers unify multimodal representations by encoding visual inputs into a finite set of tokens. Despite their effectiveness, we find that these models still hallucinate non-existent objects. We hypothesize that this may be due to visual priors induced during training: When certain image tokens frequently co-occur in the same spatial regions and represent shared objects, they become strongly associated with the verbalizations of those objects. As a result, the model may hallucinate by evoking visually absent tokens that often co-occur with present ones. To test this assumption, we construct a co-occurrence graph of image tokens using a segmentation dataset and employ a Graph Neural Network (GNN) with contrastive learning followed by a clustering method to group tokens that frequently co-occur in similar visual contexts. We find that hallucinations predominantly correspond to clusters whose tokens dominate the input, and more specifically, that the visually absent tokens in those clusters show much higher correlation with hallucinated objects compared to tokens present in the image. Based on this observation, we propose a hallucination mitigation method that suppresses the influence of visually absent tokens by modifying latent image embeddings during generation. Experiments show our method reduces hallucinations while preserving expressivity. Code is available at https://github.com/weixingW/CGC-VTD/tree/main

Paperid: 984, https://arxiv.org/pdf/2505.21500.pdf GitHub

Abstract:
Vision-language models (VLMs) have demonstrated remarkable capabilities in understanding and reasoning about visual content, but significant challenges persist in tasks requiring cross-viewpoint understanding and spatial reasoning. We identify a critical limitation: current VLMs excel primarily at egocentric spatial reasoning (from the camera's perspective) but fail to generalize to allocentric viewpoints when required to adopt another entity's spatial frame of reference. We introduce ViewSpatial-Bench, the first comprehensive benchmark designed specifically for multi-viewpoint spatial localization recognition evaluation across five distinct task types, supported by an automated 3D annotation pipeline that generates precise directional labels. Comprehensive evaluation of diverse VLMs on ViewSpatial-Bench reveals a significant performance disparity: models demonstrate reasonable performance on camera-perspective tasks but exhibit reduced accuracy when reasoning from a human viewpoint. By fine-tuning VLMs on our multi-perspective spatial dataset, we achieve an overall performance improvement of 46.24% across tasks, highlighting the efficacy of our approach. Our work establishes a crucial benchmark for spatial intelligence in embodied AI systems and provides empirical evidence that modeling 3D spatial relationships enhances VLMs' corresponding spatial comprehension capabilities.

Paperid: 985, https://arxiv.org/pdf/2505.21499.pdf GitHub

Abstract:
Vision-Language Model (VLM) based Web Agents represent a significant step towards automating complex tasks by simulating human-like interaction with websites. However, their deployment in uncontrolled web environments introduces significant security vulnerabilities. Existing research on adversarial environmental injection attacks often relies on unrealistic assumptions, such as direct HTML manipulation, knowledge of user intent, or access to agent model parameters, limiting their practical applicability. In this paper, we propose AdInject, a novel and real-world black-box attack method that leverages the internet advertising delivery to inject malicious content into the Web Agent's environment. AdInject operates under a significantly more realistic threat model than prior work, assuming a black-box agent, static malicious content constraints, and no specific knowledge of user intent. AdInject includes strategies for designing malicious ad content aimed at misleading agents into clicking, and a VLM-based ad content optimization technique that infers potential user intents from the target website's context and integrates these intents into the ad content to make it appear more relevant or critical to the agent's task, thus enhancing attack effectiveness. Experimental evaluations demonstrate the effectiveness of AdInject, attack success rates exceeding 60% in most scenarios and approaching 100% in certain cases. This strongly demonstrates that prevalent advertising delivery constitutes a potent and real-world vector for environment injection attacks against Web Agents. This work highlights a critical vulnerability in Web Agent security arising from real-world environment manipulation channels, underscoring the urgent need for developing robust defense mechanisms against such threats. Our code is available at https://github.com/NicerWang/AdInject.

Paperid: 986, https://arxiv.org/pdf/2505.21465.pdf GitHub

Abstract:
Currently, a prevalent approach for enhancing Vision-Language Models (VLMs) performance is to encode both the high-resolution version and the thumbnail of an image simultaneously. While effective, this method generates a large number of image tokens. When combined with the widely used Rotary Position Embedding (RoPE), its long-term decay property hinders the interaction between high-resolution tokens and thumbnail tokens, as well as between text and image. To address these issues, we propose ID-Align, which alleviates these problems by reordering position IDs. In this method, high-resolution tokens inherit IDs from their corresponding thumbnail token while constraining the overexpansion of positional indices. Our experiments conducted within the LLaVA-Next framework demonstrate that ID-Align achieves significant improvements, including a 6.09% enhancement on MMBench's relation reasoning tasks and notable gains across multiple benchmarks. Our code is available at the following link: https://github.com/zooblastlbz/ID-Align.

Paperid: 987, https://arxiv.org/pdf/2505.21279.pdf GitHub

Abstract:
Recent advancements in vision-language models (VLMs) have spurred increased interest in Device-Control Agents (DC agents), such as utilizing in-the-wild device control to manage graphical user interfaces. Conventional methods for assessing the capabilities of DC agents, such as computing step-wise action accuracy and overall task success rates, provide a macroscopic view of DC agents' performance; however, they fail to offer microscopic insights into potential errors that may occur in real-world applications. Conducting a finer-grained performance evaluation of DC agents presents significant challenges. This study introduces a new perspective on evaluation methods for DC agents by proposing the XBOUND evaluation method, which employs the calculation of a novel Explore Metric to delineate the capability boundaries of DC agents. Compared to previous evaluation methods, XBOUND focuses on individual states to assess the proficiency of DC agents in mastering these states. Furthermore, we have developed a ``pseudo'' episode tree dataset derived from Android Control test data. Utilizing this dataset and XBOUND, we comprehensively evaluate the OS-Atlas and UI-TARS series, examining both the overall and specific performance across five common tasks. Additionally, we select representative cases to highlight the current deficiencies and limitations inherent in both series. Code is available at https://github.com/sqzhang-lazy/XBOUND.

Paperid: 988, https://arxiv.org/pdf/2505.21061.pdf GitHub GitHub

Abstract:
Aligning large VLMs with human preferences is a challenging task, as methods like RLHF and DPO often overfit to textual information or exacerbate hallucinations. Although augmenting negative image samples partially addresses these pitfalls, no prior work has employed listwise preference optimization for VLMs, due to the complexity and cost of constructing listwise image samples. In this work, we propose LPOI, the first object-aware listwise preference optimization developed for reducing hallucinations in VLMs. LPOI identifies and masks a critical object in the image, and then interpolates the masked region between the positive and negative images to form a sequence of incrementally more complete images. The model is trained to rank these images in ascending order of object visibility, effectively reducing hallucinations while retaining visual fidelity. LPOI requires no extra annotations beyond standard pairwise preference data, as it automatically constructs the ranked lists through object masking and interpolation. Comprehensive experiments on MMHalBench, AMBER, and Object HalBench confirm that LPOI outperforms existing preference optimization methods in reducing hallucinations and enhancing VLM performance. We make the code available at https://github.com/fatemehpesaran310/lpoi.

Paperid: 989, https://arxiv.org/pdf/2505.20981.pdf GitHub GitHub GitHub

Abstract:
Autonomous Vehicles (AVs) collect and pseudo-label terabytes of multi-modal data localized to HD maps during normal fleet testing. However, identifying interesting and safety-critical scenarios from uncurated driving logs remains a significant challenge. Traditional scenario mining techniques are error-prone and prohibitively time-consuming, often relying on hand-crafted structured queries. In this work, we revisit spatio-temporal scenario mining through the lens of recent vision-language models (VLMs) to detect whether a described scenario occurs in a driving log and, if so, precisely localize it in both time and space. To address this problem, we introduce RefAV, a large-scale dataset of 10,000 diverse natural language queries that describe complex multi-agent interactions relevant to motion planning derived from 1000 driving logs in the Argoverse 2 Sensor dataset. We evaluate several referential multi-object trackers and present an empirical analysis of our baselines. Notably, we find that naively repurposing off-the-shelf VLMs yields poor performance, suggesting that scenario mining presents unique challenges. Lastly, we discuss our recent CVPR 2025 competition and share insights from the community. Our code and dataset are available at https://github.com/CainanD/RefAV/ and https://argoverse.github.io/user-guide/tasks/scenario_mining.html

Paperid: 990, https://arxiv.org/pdf/2505.20783.pdf GitHub

Abstract:
Path planning is a critical component in autonomous drone operations, enabling safe and efficient navigation through complex environments. Recent advances in foundation models, particularly large language models (LLMs) and vision-language models (VLMs), have opened new opportunities for enhanced perception and intelligent decision-making in robotics. However, their practical applicability and effectiveness in global path planning remain relatively unexplored. This paper proposes foundation model-guided path planners (FM-Planner) and presents a comprehensive benchmarking study and practical validation for drone path planning. Specifically, we first systematically evaluate eight representative LLM and VLM approaches using standardized simulation scenarios. To enable effective real-time navigation, we then design an integrated LLM-Vision planner that combines semantic reasoning with visual perception. Furthermore, we deploy and validate the proposed path planner through real-world experiments under multiple configurations. Our findings provide valuable insights into the strengths, limitations, and feasibility of deploying foundation models in real-world drone applications and providing practical implementations in autonomous flight. Project site: https://github.com/NTU-ICG/FM-Planner.

Paperid: 991, https://arxiv.org/pdf/2505.20753.pdf GitHub

Abstract:
Large Multimodal Models (LMMs) have recently demonstrated remarkable visual understanding performance on both vision-language and vision-centric tasks. However, they often fall short in integrating advanced, task-specific capabilities for compositional reasoning, which hinders their progress toward truly competent general vision models. To address this, we present a unified visual reasoning mechanism that enables LMMs to solve complicated compositional problems by leveraging their intrinsic capabilities (e.g. grounding and visual understanding capabilities). Different from the previous shortcut learning mechanism, our approach introduces a human-like understanding-thinking-answering process, allowing the model to complete all steps in a single pass forwarding without the need for multiple inferences or external tools. This design bridges the gap between foundational visual capabilities and general question answering, encouraging LMMs to generate faithful and traceable responses for complex visual reasoning. Meanwhile, we curate 334K visual instruction samples covering both general scenes and text-rich scenes and involving multiple foundational visual capabilities. Our trained model, Griffon-R, has the ability of end-to-end automatic understanding, self-thinking, and reasoning answers. Comprehensive experiments show that Griffon-R not only achieves advancing performance on complex visual reasoning benchmarks including VSR and CLEVR, but also enhances multimodal capabilities across various benchmarks like MMBench and ScienceQA. Data, models, and codes will be release at https://github.com/jefferyZhan/Griffon/tree/master/Griffon-R soon.

Paperid: 992, https://arxiv.org/pdf/2505.20728.pdf GitHub

Abstract:
Spatial reasoning is a core component of human cognition, enabling individuals to perceive, comprehend, and interact with the physical world. It relies on a nuanced understanding of spatial structures and inter-object relationships, serving as the foundation for complex reasoning and decision-making. To investigate whether current vision-language models (VLMs) exhibit similar capability, we introduce Jigsaw-Puzzles, a novel benchmark consisting of 1,100 carefully curated real-world images with high spatial complexity. Based on this dataset, we design five tasks to rigorously evaluate VLMs' spatial perception, structural understanding, and reasoning capabilities, while deliberately minimizing reliance on domain-specific knowledge to better isolate and assess the general spatial reasoning capability. We conduct a comprehensive evaluation across 24 state-of-the-art VLMs. The results show that even the strongest model, Gemini-2.5-Pro, achieves only 77.14% overall accuracy and performs particularly poorly on the Order Generation task, with only 30.00% accuracy, far below the performance exceeding 90% achieved by human participants. This persistent gap underscores the need for continued progress, positioning Jigsaw-Puzzles as a challenging and diagnostic benchmark for advancing spatial reasoning research in VLMs. Our project page is at https://zesen01.github.io/jigsaw-puzzles.

Paperid: 993, https://arxiv.org/pdf/2505.20639.pdf GitHub

Abstract:
Open-Ended object Detection (OED) is a novel and challenging task that detects objects and generates their category names in a free-form manner, without requiring additional vocabularies during inference. However, the existing OED models, such as GenerateU, require large-scale datasets for training, suffer from slow convergence, and exhibit limited performance. To address these issues, we present a novel and efficient Open-Det framework, consisting of four collaborative parts. Specifically, Open-Det accelerates model training in both the bounding box and object name generation process by reconstructing the Object Detector and the Object Name Generator. To bridge the semantic gap between Vision and Language modalities, we propose a Vision-Language Aligner with V-to-L and L-to-V alignment mechanisms, incorporating with the Prompts Distiller to transfer knowledge from the VLM into VL-prompts, enabling accurate object name generation for the LLM. In addition, we design a Masked Alignment Loss to eliminate contradictory supervision and introduce a Joint Loss to enhance classification, resulting in more efficient training. Compared to GenerateU, Open-Det, using only 1.5% of the training data (0.077M vs. 5.077M), 20.8% of the training epochs (31 vs. 149), and fewer GPU resources (4 V100 vs. 16 A100), achieves even higher performance (+1.0% in APr). The source codes are available at: https://github.com/Med-Process/Open-Det.

Paperid: 994, https://arxiv.org/pdf/2505.20612.pdf GitHub

Abstract:
Vision-language models (VLMs) trained on internet-scale data achieve remarkable zero-shot detection performance on common objects like car, truck, and pedestrian. However, state-of-the-art models still struggle to generalize to out-of-distribution classes, tasks and imaging modalities not typically found in their pre-training. Rather than simply re-training VLMs on more visual data, we argue that one should align VLMs to new concepts with annotation instructions containing a few visual examples and rich textual descriptions. To this end, we introduce Roboflow100-VL, a large-scale collection of 100 multi-modal object detection datasets with diverse concepts not commonly found in VLM pre-training. We evaluate state-of-the-art models on our benchmark in zero-shot, few-shot, semi-supervised, and fully-supervised settings, allowing for comparison across data regimes. Notably, we find that VLMs like GroundingDINO and Qwen2.5-VL achieve less than 2% zero-shot accuracy on challenging medical imaging datasets within Roboflow100-VL, demonstrating the need for few-shot concept alignment. Lastly, we discuss our recent CVPR 2025 Foundational FSOD competition and share insights from the community. Notably, the winning team significantly outperforms our baseline by 16.8 mAP! Our code and dataset are available at https://github.com/roboflow/rf100-vl/ and https://universe.roboflow.com/rf100-vl/

Paperid: 995, https://arxiv.org/pdf/2505.20569.pdf GitHub

Abstract:
Despite significant advancements in Large Vision-Language Models, Object Hallucination (OH) remains a persistent challenge. Building upon prior studies on contrastive decoding that address this issue without requiring additional model training, we introduce RVCD (Retrieval Visual Contrastive Decoding), an advanced method to suppress OH. RVCD leverages both negative and positive images at the logit level, explicitly referencing AI-generated images designed to represent a single concept. Our approach demonstrates substantial improvements over existing decoding-based methods.

Paperid: 996, https://arxiv.org/pdf/2505.20469.pdf GitHub

Abstract:
Recent advances in 3D reconstruction techniques and vision-language models have fueled significant progress in 3D semantic understanding, a capability critical to robotics, autonomous driving, and virtual/augmented reality. However, methods that rely on 2D priors are prone to a critical challenge: cross-view semantic inconsistencies induced by occlusion, image blur, and view-dependent variations. These inconsistencies, when propagated via projection supervision, deteriorate the quality of 3D Gaussian semantic fields and introduce artifacts in the rendered outputs. To mitigate this limitation, we propose CCL-LGS, a novel framework that enforces view-consistent semantic supervision by integrating multi-view semantic cues. Specifically, our approach first employs a zero-shot tracker to align a set of SAM-generated 2D masks and reliably identify their corresponding categories. Next, we utilize CLIP to extract robust semantic encodings across views. Finally, our Contrastive Codebook Learning (CCL) module distills discriminative semantic features by enforcing intra-class compactness and inter-class distinctiveness. In contrast to previous methods that directly apply CLIP to imperfect masks, our framework explicitly resolves semantic conflicts while preserving category discriminability. Extensive experiments demonstrate that CCL-LGS outperforms previous state-of-the-art methods. Our project page is available at https://epsilontl.github.io/CCL-LGS/.

Paperid: 997, https://arxiv.org/pdf/2505.20444.pdf GitHub

Abstract:
Vision-Language Models (VLMs) have made significant progress in multimodal tasks. However, their performance often deteriorates in long-context scenarios, particularly long videos. While Rotary Position Embedding (RoPE) has been widely adopted for length generalization in Large Language Models (LLMs), extending vanilla RoPE to capture the intricate spatial-temporal dependencies in videos remains an unsolved challenge. Existing methods typically allocate different frequencies within RoPE to encode 3D positional information. However, these allocation strategies mainly rely on heuristics, lacking in-depth theoretical analysis. In this paper, we first study how different allocation strategies impact the long-context capabilities of VLMs. Our analysis reveals that current multimodal RoPEs fail to reliably capture semantic similarities over extended contexts. To address this issue, we propose HoPE, a Hybrid of Position Embedding designed to improve the long-context capabilities of VLMs. HoPE introduces a hybrid frequency allocation strategy for reliable semantic modeling over arbitrarily long context, and a dynamic temporal scaling mechanism to facilitate robust learning and flexible inference across diverse context lengths. Extensive experiments across four video benchmarks on long video understanding and retrieval tasks demonstrate that HoPE consistently outperforms existing methods, confirming its effectiveness. Code is available at https://github.com/hrlics/HoPE.

Paperid: 998, https://arxiv.org/pdf/2505.20426.pdf GitHub

Abstract:
Understanding perspective is fundamental to human visual perception, yet the extent to which multimodal large language models (MLLMs) internalize perspective geometry remains unclear. We introduce MMPerspective, the first benchmark specifically designed to systematically evaluate MLLMs' understanding of perspective through 10 carefully crafted tasks across three complementary dimensions: Perspective Perception, Reasoning, and Robustness. Our benchmark comprises 2,711 real-world and synthetic image instances with 5,083 question-answer pairs that probe key capabilities, such as vanishing point perception and counting, perspective type reasoning, line relationship understanding in 3D space, invariance to perspective-preserving transformations, etc. Through a comprehensive evaluation of 43 state-of-the-art MLLMs, we uncover significant limitations: while models demonstrate competence on surface-level perceptual tasks, they struggle with compositional reasoning and maintaining spatial consistency under perturbations. Our analysis further reveals intriguing patterns between model architecture, scale, and perspective capabilities, highlighting both robustness bottlenecks and the benefits of chain-of-thought prompting. MMPerspective establishes a valuable testbed for diagnosing and advancing spatial understanding in vision-language systems. Resources available at: https://yunlong10.github.io/MMPerspective/

Paperid: 999, https://arxiv.org/pdf/2505.20362.pdf GitHub

Abstract:
The rapid advancement of vision-language models (VLMs) has brought a lot of attention to their safety alignment. However, existing methods have primarily focused on model undersafety, where the model responds to hazardous queries, while neglecting oversafety, where the model refuses to answer safe queries. In this paper, we introduce the concept of $\textit{safety calibration}$, which systematically addresses both undersafety and oversafety. Specifically, we present $\textbf{VSCBench}$, a novel dataset of 3,600 image-text pairs that are visually or textually similar but differ in terms of safety, which is designed to evaluate safety calibration across image-centric and text-centric scenarios. Based on our benchmark, we evaluate safety calibration across eleven widely used VLMs. Our extensive experiments revealed major issues with both undersafety and oversafety. We further investigated four approaches to improve the model's safety calibration. We found that even though some methods effectively calibrated the models' safety problems, these methods also lead to the degradation of models' utility. This trade-off underscores the urgent need for advanced calibration methods, and our benchmark provides a valuable tool for evaluating future approaches. Our code and data are available at https://github.com/jiahuigeng/VSCBench.git.

Paperid: 1000, https://arxiv.org/pdf/2505.20275.pdf GitHub

Abstract:
Recent advancements in generative models have enabled high-fidelity text-to-image generation. However, open-source image-editing models still lag behind their proprietary counterparts, primarily due to limited high-quality data and insufficient benchmarks. To overcome these limitations, we introduce ImgEdit, a large-scale, high-quality image-editing dataset comprising 1.2 million carefully curated edit pairs, which contain both novel and complex single-turn edits, as well as challenging multi-turn tasks. To ensure the data quality, we employ a multi-stage pipeline that integrates a cutting-edge vision-language model, a detection model, a segmentation model, alongside task-specific in-painting procedures and strict post-processing. ImgEdit surpasses existing datasets in both task novelty and data quality. Using ImgEdit, we train ImgEdit-E1, an editing model using Vision Language Model to process the reference image and editing prompt, which outperforms existing open-source models on multiple tasks, highlighting the value of ImgEdit and model design. For comprehensive evaluation, we introduce ImgEdit-Bench, a benchmark designed to evaluate image editing performance in terms of instruction adherence, editing quality, and detail preservation. It includes a basic testsuite, a challenging single-turn suite, and a dedicated multi-turn suite. We evaluate both open-source and proprietary models, as well as ImgEdit-E1, providing deep analysis and actionable insights into the current behavior of image-editing models. The source data are publicly available on https://github.com/PKU-YuanGroup/ImgEdit.

Paperid: 1001, https://arxiv.org/pdf/2505.20129.pdf GitHub GitHub

Abstract:
Despite recent advances in multimodal content generation enabled by vision-language models (VLMs), their ability to reason about and generate structured 3D scenes remains largely underexplored. This limitation constrains their utility in spatially grounded tasks such as embodied AI, immersive simulations, and interactive 3D applications. We introduce a new paradigm that enables VLMs to generate, understand, and edit complex 3D environments by injecting a continually evolving spatial context. Constructed from multimodal input, this context consists of three components: a scene portrait that provides a high-level semantic blueprint, a semantically labeled point cloud capturing object-level geometry, and a scene hypergraph that encodes rich spatial relationships, including unary, binary, and higher-order constraints. Together, these components provide the VLM with a structured, geometry-aware working memory that integrates its inherent multimodal reasoning capabilities with structured 3D understanding for effective spatial reasoning. Building on this foundation, we develop an agentic 3D scene generation pipeline in which the VLM iteratively reads from and updates the spatial context. The pipeline features high-quality asset generation with geometric restoration, environment setup with automatic verification, and ergonomic adjustment guided by the scene hypergraph. Experiments show that our framework can handle diverse and challenging inputs, achieving a level of generalization not observed in prior work. Further results demonstrate that injecting spatial context enables VLMs to perform downstream tasks such as interactive scene editing and path planning, suggesting strong potential for spatially intelligent systems in computer graphics, 3D vision, and embodied applications. Project page: https://spatctxvlm.github.io/project_page/.

Paperid: 1002, https://arxiv.org/pdf/2505.20024.pdf GitHub GitHub

Abstract:
Due to the powerful vision-language reasoning and generalization abilities, multimodal large language models (MLLMs) have garnered significant attention in the field of end-to-end (E2E) autonomous driving. However, their application to closed-loop systems remains underexplored, and current MLLM-based methods have not shown clear superiority to mainstream E2E imitation learning approaches. In this work, we propose ReasonPlan, a novel MLLM fine-tuning framework designed for closed-loop driving through holistic reasoning with a self-supervised Next Scene Prediction task and supervised Decision Chain-of-Thought process. This dual mechanism encourages the model to align visual representations with actionable driving context, while promoting interpretable and causally grounded decision making. We curate a planning-oriented decision reasoning dataset, namely PDR, comprising 210k diverse and high-quality samples. Our method outperforms the mainstream E2E imitation learning method by a large margin of 19% L2 and 16.1 driving score on Bench2Drive benchmark. Furthermore, ReasonPlan demonstrates strong zero-shot generalization on unseen DOS benchmark, highlighting its adaptability in handling zero-shot corner cases. Code and dataset will be found in https://github.com/Liuxueyi/ReasonPlan.

Paperid: 1003, https://arxiv.org/pdf/2505.19812.pdf GitHub

Abstract:
Traditional approaches to adapting multi-modal large language models (MLLMs) to new tasks have relied heavily on fine-tuning. This paper introduces Efficient Multi-Modal Long Context Learning (EMLoC), a novel training-free alternative that embeds demonstration examples directly into the model input. EMLoC offers a more efficient, flexible, and scalable solution for task adaptation. Because extremely lengthy inputs introduce prohibitive computational and memory overhead, EMLoC contributes a chunk-wise compression mechanism combined with layer-wise adaptive pruning. It condenses long-context multimodal inputs into compact, task-specific memory representations. By adaptively pruning tokens at each layer under a Jensen-Shannon divergence constraint, our method achieves a dramatic reduction in inference complexity without sacrificing performance. This approach is the first to seamlessly integrate compression and pruning techniques for multi-modal long-context learning, offering a scalable and efficient solution for real-world applications. Extensive experiments on diverse vision-language benchmarks demonstrate that EMLoC achieves performance on par with or superior to naive long-context approaches. Our results highlight the potential of EMLoC as a groundbreaking framework for efficient and flexible adaptation of multi-modal models in resource-constrained environments. Codes are publicly available at https://github.com/Zehong-Ma/EMLoC.

Paperid: 1004, https://arxiv.org/pdf/2505.19536.pdf GitHub

Abstract:
Large vision-language models (LVLMs) excel at multimodal understanding but suffer from high computational costs due to redundant vision tokens. Existing pruning methods typically rely on single-layer attention scores to rank and prune redundant visual tokens to solve this inefficiency. However, as the interaction between tokens and layers is complicated, this raises a basic question: Is such a simple single-layer criterion sufficient to identify redundancy? To answer this question, we rethink the emergence of redundant visual tokens from a fundamental perspective: information flow, which models the interaction between tokens and layers by capturing how information moves between tokens across layers. We find (1) the CLS token acts as an information relay, which can simplify the complicated flow analysis; (2) the redundancy emerges progressively and dynamically via layer-wise attention concentration; and (3) relying solely on attention scores from single layers can lead to contradictory redundancy identification. Based on this, we propose FlowCut, an information-flow-aware pruning framework, mitigating the insufficiency of the current criterion for identifying redundant tokens and better aligning with the model's inherent behaviors. Extensive experiments show that FlowCut achieves superior results, outperforming SoTA by 1.6% on LLaVA-1.5-7B with 88.9% token reduction, and by 4.3% on LLaVA-NeXT-7B with 94.4% reduction, delivering 3.2x speed-up in the prefilling stage. Our code is available at https://github.com/TungChintao/FlowCut

Paperid: 1005, https://arxiv.org/pdf/2505.19503.pdf GitHub

Abstract:
Recent methods for zero-shot Human-Object Interaction (HOI) detection typically leverage the generalization ability of large Vision-Language Model (VLM), i.e., CLIP, on unseen categories, showing impressive results on various zero-shot settings. However, existing methods struggle to adapt CLIP representations for human-object pairs, as CLIP tends to overlook fine-grained information necessary for distinguishing interactions. To address this issue, we devise, LAIN, a novel zero-shot HOI detection framework enhancing the locality and interaction awareness of CLIP representations. The locality awareness, which involves capturing fine-grained details and the spatial structure of individual objects, is achieved by aggregating the information and spatial priors of adjacent neighborhood patches. The interaction awareness, which involves identifying whether and how a human is interacting with an object, is achieved by capturing the interaction pattern between the human and the object. By infusing locality and interaction awareness into CLIP representation, LAIN captures detailed information about the human-object pairs. Our extensive experiments on existing benchmarks show that LAIN outperforms previous methods on various zero-shot settings, demonstrating the importance of locality and interaction awareness for effective zero-shot HOI detection.

Paperid: 1006, https://arxiv.org/pdf/2505.19490.pdf GitHub

Abstract:
Designing complex computer-aided design (CAD) models is often time-consuming due to challenges such as computational inefficiency and the difficulty of generating precise models. We propose a novel language-guided framework for industrial design automation to address these issues, integrating large language models (LLMs) with computer-automated design (CAutoD).Through this framework, CAD models are automatically generated from parameters and appearance descriptions, supporting the automation of design tasks during the detailed CAD design phase. Our approach introduces three key innovations: (1) a semi-automated data annotation pipeline that leverages LLMs and vision-language large models (VLLMs) to generate high-quality parameters and appearance descriptions; (2) a Transformer-based CAD generator (TCADGen) that predicts modeling sequences via dual-channel feature aggregation; (3) an enhanced CAD modeling generation model, called CADLLM, that is designed to refine the generated sequences by incorporating the confidence scores from TCADGen. Experimental results demonstrate that the proposed approach outperforms traditional methods in both accuracy and efficiency, providing a powerful tool for automating industrial workflows and generating complex CAD models from textual prompts. The code is available at https://jianxliao.github.io/cadllm-page/

Paperid: 1007, https://arxiv.org/pdf/2505.19312.pdf GitHub GitHub

Abstract:
The rapid advancement of unsupervised representation learning and large-scale pre-trained vision-language models has significantly improved cross-modal retrieval tasks. However, existing multi-modal information retrieval (MMIR) studies lack a comprehensive exploration of document-level retrieval and suffer from the absence of cross-domain datasets at this granularity. To address this limitation, we introduce DocMMIR, a novel multi-modal document retrieval framework designed explicitly to unify diverse document formats and domains, including Wikipedia articles, scientific papers (arXiv), and presentation slides, within a comprehensive retrieval scenario. We construct a large-scale cross-domain multimodal benchmark, comprising 450K samples, which systematically integrates textual and visual information. Our comprehensive experimental analysis reveals substantial limitations in current state-of-the-art MLLMs (CLIP, BLIP2, SigLIP-2, ALIGN) when applied to our tasks, with only CLIP demonstrating reasonable zero-shot performance. Furthermore, we conduct a systematic investigation of training strategies, including cross-modal fusion methods and loss functions, and develop a tailored approach to train CLIP on our benchmark. This results in a +31% improvement in MRR@10 compared to the zero-shot baseline. All our data and code are released in https://github.com/J1mL1/DocMMIR.

Paperid: 1008, https://arxiv.org/pdf/2505.19255.pdf GitHub

Abstract:
Reinforcement Learning Finetuning (RFT) has significantly advanced the reasoning capabilities of large language models (LLMs) by enabling long chains of thought, self-correction, and effective tool use. While recent works attempt to extend RFT to vision-language models (VLMs), these efforts largely produce text-only reasoning conditioned on static image inputs, falling short of true multimodal reasoning in the response. In contrast, test-time methods like Visual Sketchpad incorporate visual steps but lack training mechanisms. We introduce VTool-R1, the first framework that trains VLMs to generate multimodal chains of thought by interleaving text and intermediate visual reasoning steps. VTool-R1 integrates Python-based visual editing tools into the RFT process, enabling VLMs to learn when and how to generate visual reasoning steps that benefit final reasoning. Trained with outcome-based rewards tied to task accuracy, our approach elicits strategic visual tool use for reasoning without relying on process-based supervision. Experiments on structured visual question answering over charts and tables show that VTool-R1 enhances reasoning performance by teaching VLMs to "think with images" and generate multimodal chain of thoughts with tools.

Paperid: 1009, https://arxiv.org/pdf/2505.19235.pdf GitHub

Abstract:
Vision-Language Models (VLMs) excel across diverse tasks but suffer from high inference costs in time and memory. Token sparsity mitigates inefficiencies in token usage, while neuron sparsity reduces high-dimensional computations, both offering promising solutions to enhance efficiency. Recently, these two sparsity paradigms have evolved largely in parallel, fostering the prevailing assumption that they function independently. However, a fundamental yet underexplored question remains: Do they truly operate in isolation, or is there a deeper underlying interplay that has yet to be uncovered? In this paper, we conduct the first comprehensive investigation into this question. By introducing and analyzing the matching mechanism between Core Neurons and Core Tokens, we found that key neurons and tokens for inference mutually influence and reinforce each other. Building on this insight, we propose CoreMatching, a co-adaptive sparse inference framework, which leverages the synergy between token and neuron sparsity to enhance inference efficiency. Through theoretical analysis and efficiency evaluations, we demonstrate that the proposed method surpasses state-of-the-art baselines on ten image understanding tasks and three hardware devices. Notably, on the NVIDIA Titan Xp, it achieved 5x FLOPs reduction and a 10x overall speedup. Code is released at https://github.com/wangqinsi1/2025-ICML-CoreMatching/tree/main.

Paperid: 1010, https://arxiv.org/pdf/2505.19201.pdf GitHub

Abstract:
Speculative decoding (SD) has emerged as a powerful method for accelerating autoregressive generation in large language models (LLMs), yet its integration into vision-language models (VLMs) remains underexplored. We introduce DREAM, a novel speculative decoding framework tailored for VLMs that combines three key innovations: (1) a cross-attention-based mechanism to inject intermediate features from the target model into the draft model for improved alignment, (2) adaptive intermediate feature selection based on attention entropy to guide efficient draft model training, and (3) visual token compression to reduce draft model latency. DREAM enables efficient, accurate, and parallel multimodal decoding with significant throughput improvement. Experiments across a diverse set of recent popular VLMs, including LLaVA, Pixtral, SmolVLM and Gemma3, demonstrate up to 3.6x speedup over conventional decoding and significantly outperform prior SD baselines in both inference throughput and speculative draft acceptance length across a broad range of multimodal benchmarks. The code is publicly available at: https://github.com/SAI-Lab-NYU/DREAM.git

Paperid: 1011, https://arxiv.org/pdf/2505.19091.pdf GitHub

Abstract:
Recent advancements in Large Vision-Language Models (VLMs), have greatly enhanced their capability to jointly process text and images. However, despite extensive benchmarks evaluating visual comprehension (e.g., diagrams, color schemes, OCR tasks...), there is limited assessment of VLMs' ability to read and reason about text-rich images effectively. To fill this gap, we introduce ReadBench, a multimodal benchmark specifically designed to evaluate the reading comprehension capabilities of VLMs. ReadBench transposes contexts from established text-only benchmarks into images of text while keeping textual prompts and questions intact. Evaluating leading VLMs with ReadBench, we find minimal-but-present performance degradation on short, text-image inputs, while performance sharply declines for longer, multi-page contexts. Our experiments further reveal that text resolution has negligible effects on multimodal performance. These findings highlight needed improvements in VLMs, particularly their reasoning over visually presented extensive textual content, a capability critical for practical applications. ReadBench is available at https://github.com/answerdotai/ReadBench .

Paperid: 1012, https://arxiv.org/pdf/2505.18915.pdf GitHub

Abstract:
Vision-Language Models (VLMs) have shown promise in various 2D visual tasks, yet their readiness for 3D clinical diagnosis remains unclear due to stringent demands for recognition precision, reasoning ability, and domain knowledge. To systematically evaluate these dimensions, we present DeepTumorVQA, a diagnostic visual question answering (VQA) benchmark targeting abdominal tumors in CT scans. It comprises 9,262 CT volumes (3.7M slices) from 17 public datasets, with 395K expert-level questions spanning four categories: Recognition, Measurement, Visual Reasoning, and Medical Reasoning. DeepTumorVQA introduces unique challenges, including small tumor detection and clinical reasoning across 3D anatomy. Benchmarking four advanced VLMs (RadFM, M3D, Merlin, CT-CHAT), we find current models perform adequately on measurement tasks but struggle with lesion recognition and reasoning, and are still not meeting clinical needs. Two key insights emerge: (1) large-scale multimodal pretraining plays a crucial role in DeepTumorVQA testing performance, making RadFM stand out among all VLMs. (2) Our dataset exposes critical differences in VLM components, where proper image preprocessing and design of vision modules significantly affect 3D perception. To facilitate medical multimodal research, we have released DeepTumorVQA as a rigorous benchmark: https://github.com/Schuture/DeepTumorVQA.

Paperid: 1013, https://arxiv.org/pdf/2505.18700.pdf GitHub

Abstract:
Recent advances in Visual Language Models (VLMs) have demonstrated exceptional performance in visual reasoning tasks. However, geo-localization presents unique challenges, requiring the extraction of multigranular visual cues from images and their integration with external world knowledge for systematic reasoning. Current approaches to geo-localization tasks often lack robust reasoning mechanisms and explainability, limiting their effectiveness. To address these limitations, we propose the Geo Reason Enhancement (GRE) Suite, a novel framework that augments VLMs with structured reasoning chains for accurate and interpretable location inference. The GRE Suite is systematically developed across three key dimensions: dataset, model, and benchmark. First, we introduce GRE30K, a high-quality geo-localization reasoning dataset designed to facilitate fine-grained visual and contextual analysis. Next, we present the GRE model, which employs a multi-stage reasoning strategy to progressively infer scene attributes, local details, and semantic features, thereby narrowing down potential geographic regions with enhanced precision. Finally, we construct the Geo Reason Evaluation Benchmark (GREval-Bench), a comprehensive evaluation framework that assesses VLMs across diverse urban, natural, and landmark scenes to measure both coarse-grained (e.g., country, continent) and fine-grained (e.g., city, street) localization performance. Experimental results demonstrate that GRE significantly outperforms existing methods across all granularities of geo-localization tasks, underscoring the efficacy of reasoning-augmented VLMs in complex geographic inference. Code and data will be released at https://github.com/Thorin215/GRE.

Paperid: 1014, https://arxiv.org/pdf/2505.18612.pdf GitHub

Abstract:
Personalized text-to-image generation aims to synthesize images of user-provided concepts in diverse contexts. Despite recent progress in multi-concept personalization, most are limited to object concepts and struggle to customize abstract concepts (e.g., pose, lighting). Some methods have begun exploring multi-concept personalization supporting abstract concepts, but they require test-time fine-tuning for each new concept, which is time-consuming and prone to overfitting on limited training images. In this work, we propose a novel tuning-free method for multi-concept personalization that can effectively customize both object and abstract concepts without test-time fine-tuning. Our method builds upon the modulation mechanism in pretrained Diffusion Transformers (DiTs) model, leveraging the localized and semantically meaningful properties of the modulation space. Specifically, we propose a novel module, Mod-Adapter, to predict concept-specific modulation direction for the modulation process of concept-related text tokens. It incorporates vision-language cross-attention for extracting concept visual features, and Mixture-of-Experts (MoE) layers that adaptively map the concept features into the modulation space. Furthermore, to mitigate the training difficulty caused by the large gap between the concept image space and the modulation space, we introduce a VLM-guided pretraining strategy that leverages the strong image understanding capabilities of vision-language models to provide semantic supervision signals. For a comprehensive comparison, we extend a standard benchmark by incorporating abstract concepts. Our method achieves state-of-the-art performance in multi-concept personalization, supported by quantitative, qualitative, and human evaluations.

Paperid: 1015, https://arxiv.org/pdf/2505.18600.pdf GitHub

Abstract:
Modern single-image super-resolution (SISR) models deliver photo-realistic results at the scale factors on which they are trained, but collapse when asked to magnify far beyond that regime. We address this scalability bottleneck with Chain-of-Zoom (CoZ), a model-agnostic framework that factorizes SISR into an autoregressive chain of intermediate scale-states with multi-scale-aware prompts. CoZ repeatedly re-uses a backbone SR model, decomposing the conditional probability into tractable sub-problems to achieve extreme resolutions without additional training. Because visual cues diminish at high magnifications, we augment each zoom step with multi-scale-aware text prompts generated by a vision-language model (VLM). The prompt extractor itself is fine-tuned using Generalized Reward Policy Optimization (GRPO) with a critic VLM, aligning text guidance towards human preference. Experiments show that a standard 4x diffusion SR model wrapped in CoZ attains beyond 256x enlargement with high perceptual quality and fidelity. Project Page: https://bryanswkim.github.io/chain-of-zoom/ .

Paperid: 1016, https://arxiv.org/pdf/2505.18525.pdf GitHub

Abstract:
3D medical image segmentation is vital for clinical diagnosis and treatment but is challenged by high-dimensional data and complex spatial dependencies. Traditional single-modality networks, such as CNNs and Transformers, are often limited by computational inefficiency and constrained contextual modeling in 3D settings. We introduce a novel multimodal framework that leverages Mamba and Kolmogorov-Arnold Networks (KAN) as an efficient backbone for long-sequence modeling. Our approach features three key innovations: First, an EGSC (Enhanced Gated Spatial Convolution) module captures spatial information when unfolding 3D images into 1D sequences. Second, we extend Group-Rational KAN (GR-KAN), a Kolmogorov-Arnold Networks variant with rational basis functions, into 3D-Group-Rational KAN (3D-GR-KAN) for 3D medical imaging - its first application in this domain - enabling superior feature representation tailored to volumetric data. Third, a dual-branch text-driven strategy leverages CLIP's text embeddings: one branch swaps one-hot labels for semantic vectors to preserve inter-organ semantic relationships, while the other aligns images with detailed organ descriptions to enhance semantic alignment. Experiments on the Medical Segmentation Decathlon (MSD) and KiTS23 datasets show our method achieving state-of-the-art performance, surpassing existing approaches in accuracy and efficiency. This work highlights the power of combining advanced sequence modeling, extended network architectures, and vision-language synergy to push forward 3D medical image segmentation, delivering a scalable solution for clinical use. The source code is openly available at https://github.com/yhy-whu/TK-Mamba.

Paperid: 1017, https://arxiv.org/pdf/2505.18291.pdf GitHub GitHub

Abstract:
Large multimodal foundation models, particularly in the domains of language and vision, have significantly advanced various tasks, including robotics, autonomous driving, information retrieval, and grounding. However, many of these models perceive objects as indivisible, overlooking the components that constitute them. Understanding these components and their associated affordances provides valuable insights into an object's functionality, which is fundamental for performing a wide range of tasks. In this work, we introduce a novel real-world benchmark, InstructPart, comprising hand-labeled part segmentation annotations and task-oriented instructions to evaluate the performance of current models in understanding and executing part-level tasks within everyday contexts. Through our experiments, we demonstrate that task-oriented part segmentation remains a challenging problem, even for state-of-the-art Vision-Language Models (VLMs). In addition to our benchmark, we introduce a simple baseline that achieves a twofold performance improvement through fine-tuning with our dataset. With our dataset and benchmark, we aim to facilitate research on task-oriented part segmentation and enhance the applicability of VLMs across various domains, including robotics, virtual reality, information retrieval, and other related fields. Project website: https://zifuwan.github.io/InstructPart/.

Paperid: 1018, https://arxiv.org/pdf/2505.18229.pdf GitHub

Abstract:
With the rapid advancement of low-altitude remote sensing and Vision-Language Models (VLMs), Embodied Agents based on Unmanned Aerial Vehicles (UAVs) have shown significant potential in autonomous tasks. However, current evaluation methods for UAV-Embodied Agents (UAV-EAs) remain constrained by the lack of standardized benchmarks, diverse testing scenarios and open system interfaces. To address these challenges, we propose BEDI (Benchmark for Embodied Drone Intelligence), a systematic and standardized benchmark designed for evaluating UAV-EAs. Specifically, we introduce a novel Dynamic Chain-of-Embodied-Task paradigm based on the perception-decision-action loop, which decomposes complex UAV tasks into standardized, measurable subtasks. Building on this paradigm, we design a unified evaluation framework encompassing five core sub-skills: semantic perception, spatial perception, motion control, tool utilization, and task planning. Furthermore, we construct a hybrid testing platform that integrates static real-world environments with dynamic virtual scenarios, enabling comprehensive performance assessment of UAV-EAs across varied contexts. The platform also offers open and standardized interfaces, allowing researchers to customize tasks and extend scenarios, thereby enhancing flexibility and scalability in the evaluation process. Finally, through empirical evaluations of several state-of-the-art (SOTA) VLMs, we reveal their limitations in embodied UAV tasks, underscoring the critical role of the BEDI benchmark in advancing embodied intelligence research and model optimization. By filling the gap in systematic and standardized evaluation within this field, BEDI facilitates objective model comparison and lays a robust foundation for future development in this field. Our benchmark will be released at https://github.com/lostwolves/BEDI .

Paperid: 1019, https://arxiv.org/pdf/2505.18227.pdf GitHub

Abstract:
In Transformer architectures, tokens\textemdash discrete units derived from raw data\textemdash are formed by segmenting inputs into fixed-length chunks. Each token is then mapped to an embedding, enabling parallel attention computations while preserving the input's essential information. Due to the quadratic computational complexity of transformer self-attention mechanisms, token reduction has primarily been used as an efficiency strategy. This is especially true in single vision and language domains, where it helps balance computational costs, memory usage, and inference latency. Despite these advances, this paper argues that token reduction should transcend its traditional efficiency-oriented role in the era of large generative models. Instead, we position it as a fundamental principle in generative modeling, critically influencing both model architecture and broader applications. Specifically, we contend that across vision, language, and multimodal systems, token reduction can: (i) facilitate deeper multimodal integration and alignment, (ii) mitigate "overthinking" and hallucinations, (iii) maintain coherence over long inputs, and (iv) enhance training stability, etc. We reframe token reduction as more than an efficiency measure. By doing so, we outline promising future directions, including algorithm design, reinforcement learning-guided token reduction, token optimization for in-context learning, and broader ML and scientific domains. We highlight its potential to drive new model architectures and learning strategies that improve robustness, increase interpretability, and better align with the objectives of generative modeling.

Paperid: 1020, https://arxiv.org/pdf/2505.18129.pdf GitHub

Abstract:
Reinforcement learning (RL) has significantly advanced the reasoning capabilities of vision-language models (VLMs). However, the use of RL beyond reasoning tasks remains largely unexplored, especially for perceptionintensive tasks like object detection and grounding. We propose V-Triune, a Visual Triple Unified Reinforcement Learning system that enables VLMs to jointly learn visual reasoning and perception tasks within a single training pipeline. V-Triune comprises triple complementary components: Sample-Level Data Formatting (to unify diverse task inputs), Verifier-Level Reward Computation (to deliver custom rewards via specialized verifiers) , and Source-Level Metric Monitoring (to diagnose problems at the data-source level). We further introduce a novel Dynamic IoU reward, which provides adaptive, progressive, and definite feedback for perception tasks handled by V-Triune. Our approach is instantiated within off-the-shelf RL training framework using open-source 7B and 32B backbone models. The resulting model, dubbed Orsta (One RL to See Them All), demonstrates consistent improvements across both reasoning and perception tasks. This broad capability is significantly shaped by its training on a diverse dataset, constructed around four representative visual reasoning tasks (Math, Puzzle, Chart, and Science) and four visual perception tasks (Grounding, Detection, Counting, and OCR). Subsequently, Orsta achieves substantial gains on MEGA-Bench Core, with improvements ranging from +2.1 to an impressive +14.1 across its various 7B and 32B model variants, with performance benefits extending to a wide range of downstream tasks. These results highlight the effectiveness and scalability of our unified RL approach for VLMs. The V-Triune system, along with the Orsta models, is publicly available at https://github.com/MiniMax-AI.

Paperid: 1021, https://arxiv.org/pdf/2505.18087.pdf GitHub

Abstract:
Recent progress in Large Vision-Language Models (LVLMs) has enabled promising applications in medical tasks, such as report generation and visual question answering. However, existing benchmarks focus mainly on the final diagnostic answer, offering limited insight into whether models engage in clinically meaningful reasoning. To address this, we present CheXStruct and CXReasonBench, a structured pipeline and benchmark built on the publicly available MIMIC-CXR-JPG dataset. CheXStruct automatically derives a sequence of intermediate reasoning steps directly from chest X-rays, such as segmenting anatomical regions, deriving anatomical landmarks and diagnostic measurements, computing diagnostic indices, and applying clinical thresholds. CXReasonBench leverages this pipeline to evaluate whether models can perform clinically valid reasoning steps and to what extent they can learn from structured guidance, enabling fine-grained and transparent assessment of diagnostic reasoning. The benchmark comprises 18,988 QA pairs across 12 diagnostic tasks and 1,200 cases, each paired with up to 4 visual inputs, and supports multi-path, multi-stage evaluation including visual grounding via anatomical region selection and diagnostic measurements. Even the strongest of 10 evaluated LVLMs struggle with structured reasoning and generalization, often failing to link abstract knowledge with anatomically grounded visual interpretation. The code is available at https://github.com/ttumyche/CXReasonBench

Paperid: 1022, https://arxiv.org/pdf/2505.17982.pdf GitHub

Abstract:
Vision-language models (VLMs) have recently been integrated into multiple instance learning (MIL) frameworks to address the challenge of few-shot, weakly supervised classification of whole slide images (WSIs). A key trend involves leveraging multi-scale information to better represent hierarchical tissue structures. However, existing methods often face two key limitations: (1) insufficient modeling of interactions within the same modalities across scales (e.g., 5x and 20x) and (2) inadequate alignment between visual and textual modalities on the same scale. To address these gaps, we propose HiVE-MIL, a hierarchical vision-language framework that constructs a unified graph consisting of (1) parent-child links between coarse (5x) and fine (20x) visual/textual nodes to capture hierarchical relationships, and (2) heterogeneous intra-scale edges linking visual and textual nodes on the same scale. To further enhance semantic consistency, HiVE-MIL incorporates a two-stage, text-guided dynamic filtering mechanism that removes weakly correlated patch-text pairs, and introduces a hierarchical contrastive loss to align textual semantics across scales. Extensive experiments on TCGA breast, lung, and kidney cancer datasets demonstrate that HiVE-MIL consistently outperforms both traditional MIL and recent VLM-based MIL approaches, achieving gains of up to 4.1% in macro F1 under 16-shot settings. Our results demonstrate the value of jointly modeling hierarchical structure and multimodal alignment for efficient and scalable learning from limited pathology data. The code is available at https://github.com/bryanwong17/HiVE-MIL

Paperid: 1023, https://arxiv.org/pdf/2505.17812.pdf GitHub

Abstract:
Large Vision-Language Models (LVLMs) have achieved remarkable success but continue to struggle with object hallucination (OH), generating outputs inconsistent with visual inputs. While previous work has proposed methods to reduce OH, the visual decision-making mechanisms that lead to hallucinations remain poorly understood. In this paper, we propose VaLSe, a Vision-aware Latent Steering framework that adopts an interpretation-then-mitigation strategy to address OH in LVLMs. By tackling dual challenges of modeling complex vision-language interactions and eliminating spurious activation artifacts, VaLSe can generate visual contribution maps that trace how specific visual inputs influence individual output tokens. These maps reveal the model's vision-aware focus regions, which are then used to perform latent space steering, realigning internal representations toward semantically relevant content and reducing hallucinated outputs. Extensive experiments demonstrate that VaLSe is a powerful interpretability tool and an effective method for enhancing model robustness against OH across multiple benchmarks. Furthermore, our analysis uncovers limitations in existing OH evaluation metrics, underscoring the need for more nuanced, interpretable, and visually grounded OH benchmarks in future work. Code is available at: https://github.com/Ziwei-Zheng/VaLSe.

Paperid: 1024, https://arxiv.org/pdf/2505.17727.pdf GitHub

Abstract:
Safety-critical scenarios are rare yet pivotal for evaluating and enhancing the robustness of autonomous driving systems. While existing methods generate safety-critical driving trajectories, simulations, or single-view videos, they fall short of meeting the demands of advanced end-to-end autonomous systems (E2E AD), which require real-world, multi-view video data. To bridge this gap, we introduce SafeMVDrive, the first framework designed to generate high-quality, safety-critical, multi-view driving videos grounded in real-world domains. SafeMVDrive strategically integrates a safety-critical trajectory generator with an advanced multi-view video generator. To tackle the challenges inherent in this integration, we first enhance scene understanding ability of the trajectory generator by incorporating visual context -- which is previously unavailable to such generator -- and leveraging a GRPO-finetuned vision-language model to achieve more realistic and context-aware trajectory generation. Second, recognizing that existing multi-view video generators struggle to render realistic collision events, we introduce a two-stage, controllable trajectory generation mechanism that produces collision-evasion trajectories, ensuring both video quality and safety-critical fidelity. Finally, we employ a diffusion-based multi-view video generator to synthesize high-quality safety-critical driving videos from the generated trajectories. Experiments conducted on an E2E AD planner demonstrate a significant increase in collision rate when tested with our generated data, validating the effectiveness of SafeMVDrive in stress-testing planning modules. Our code, examples, and datasets are publicly available at: https://zhoujiawei3.github.io/SafeMVDrive/.

Paperid: 1025, https://arxiv.org/pdf/2505.17685.pdf GitHub GitHub

Abstract:
Vision-Language-Action (VLA) models are increasingly used for end-to-end driving due to their world knowledge and reasoning ability. Most prior work, however, inserts textual chains-of-thought (CoT) as intermediate steps tailored to the current scene. Such symbolic compressions can blur spatio-temporal relations and discard fine visual cues, creating a cross-modal gap between perception and planning. We propose FSDrive, a visual spatio-temporal CoT framework that enables VLAs to think in images. The model first acts as a world model to generate a unified future frame that overlays coarse but physically-plausible priors-future lane dividers and 3D boxes-on the predicted future image. This unified frame serves as the visual CoT, capturing both spatial structure and temporal evolution. The same VLA then functions as an inverse-dynamics model, planning trajectories from current observations and the visual CoT. To equip VLAs with image generation while preserving understanding, we introduce a unified pre-training paradigm that expands the vocabulary to include visual tokens and jointly optimizes VQA (for semantics) and future-frame prediction (for dynamics). A progressive easy-to-hard scheme first predicts lane/box priors to enforce physical constraints, then completes full future frames for fine details. On nuScenes and NAVSIM, FSDrive improves trajectory accuracy and reduces collisions under both ST-P3 and UniAD metrics, and attains competitive FID for future-frame generation despite using lightweight autoregression. It also advances scene understanding on DriveLM. Together, these results indicate that visual CoT narrows the cross-modal gap and yields safer, more anticipatory planning. Code is available at https://github.com/MIV-XJTU/FSDrive.

Paperid: 1026, https://arxiv.org/pdf/2505.17674.pdf GitHub

Abstract:
Spiking Neural Networks (SNNs) provide an energy-efficient way to extract 3D spatio-temporal features. However, existing SNNs still exhibit a significant performance gap compared to Artificial Neural Networks (ANNs) due to inadequate pre-training strategies. These limitations manifest as restricted generalization ability, task specificity, and a lack of multimodal understanding, particularly in challenging tasks such as multimodal question answering and zero-shot 3D classification. To overcome these challenges, we propose a Spike-based Vision-Language (SVL) pretraining framework that empowers SNNs with open-world 3D understanding while maintaining spike-driven efficiency. SVL introduces two key components: (i) Multi-scale Triple Alignment (MTA) for label-free triplet-based contrastive learning across 3D, image, and text modalities, and (ii) Re-parameterizable Vision-Language Integration (Rep-VLI) to enable lightweight inference without relying on large text encoders. Extensive experiments show that SVL achieves a top-1 accuracy of 85.4% in zero-shot 3D classification, surpassing advanced ANN models, and consistently outperforms prior SNNs on downstream tasks, including 3D classification (+6.1%), DVS action recognition (+2.1%), 3D detection (+1.1%), and 3D segmentation (+2.1%) with remarkable efficiency. Moreover, SVL enables SNNs to perform open-world 3D question answering, sometimes outperforming ANNs. To the best of our knowledge, SVL represents the first scalable, generalizable, and hardware-friendly paradigm for 3D open-world understanding, effectively bridging the gap between SNNs and ANNs in complex open-world understanding tasks. Code is available https://github.com/bollossom/SVL.

Paperid: 1027, https://arxiv.org/pdf/2505.17609.pdf GitHub

Abstract:
Current large vision-language models (LVLMs) typically employ a connector module to link visual features with text embeddings of large language models (LLMs) and use end-to-end training to achieve multi-modal understanding in a unified process. Effective alignment needs high-quality pre-training data and a carefully designed training process. Current LVLMs face challenges when addressing complex vision-language reasoning tasks, with their reasoning capabilities notably lagging behind those of LLMs. This paper proposes a paradigm shift: instead of training end-to-end vision-language reasoning models, we advocate for developing a decoupled reasoning framework based on existing visual interpretation specialists and text-based reasoning LLMs. Our approach leverages (1) a dedicated vision-language model to transform the visual content of images into textual descriptions and (2) an LLM to perform reasoning according to the visual-derived text and the original question. This method presents a cost-efficient solution for multi-modal model development by optimizing existing models to work collaboratively, avoiding end-to-end development of vision-language models from scratch. By transforming images into language model-compatible text representations, it facilitates future low-cost and flexible upgrades to upcoming powerful LLMs. We introduce an outcome-rewarded joint-tuning strategy to optimize the cooperation between the visual interpretation and linguistic reasoning model. Evaluation results on vision-language benchmarks demonstrate that the decoupled reasoning framework outperforms recent LVLMs. Our approach yields particularly significant performance gains on visually intensive geometric mathematics problems. The code is available: https://github.com/guozix/DVLR.

Paperid: 1028, https://arxiv.org/pdf/2505.17440.pdf GitHub

Abstract:
Large Vision-Language Models (LVLMs) have demonstrated remarkable capabilities in multimodal understanding and generation, yet their vulnerability to adversarial attacks raises significant robustness concerns. While existing effective attacks always focus on task-specific white-box settings, these approaches are limited in the context of LVLMs, which are designed for diverse downstream tasks and require expensive full-model gradient computations. Motivated by the pivotal role and wide adoption of the vision encoder in LVLMs, we propose a simple yet effective Vision Encoder Attack (VEAttack), which targets the vision encoder of LVLMs only. Specifically, we propose to generate adversarial examples by minimizing the cosine similarity between the clean and perturbed visual features, without accessing the following large language models, task information, and labels. It significantly reduces the computational overhead while eliminating the task and label dependence of traditional white-box attacks in LVLMs. To make this simple attack effective, we propose to perturb images by optimizing image tokens instead of the classification token. We provide both empirical and theoretical evidence that VEAttack can easily generalize to various tasks. VEAttack has achieved a performance degradation of 94.5% on image caption task and 75.7% on visual question answering task. We also reveal some key observations to provide insights into LVLM attack/defense: 1) hidden layer variations of LLM, 2) token attention differential, 3) MÃ¶bius band in transfer attack, 4) low sensitivity to attack steps. The code is available at https://github.com/hfmei/VEAttack-LVLM

Paperid: 1029, https://arxiv.org/pdf/2505.17061.pdf GitHub

Abstract:
Large Vision-Language Models (LVLMs) have exhibited impressive capabilities across various visual tasks, yet they remain hindered by the persistent challenge of hallucinations. To address this critical issue, we propose Mixture of Decoding (MoD), a novel approach for hallucination mitigation that dynamically adapts decoding strategies by evaluating the correctness of the model's attention on image tokens. Specifically, MoD measures the consistency between outputs generated from the original image tokens and those derived from the model's attended image tokens, to distinguish the correctness aforementioned. If the outputs are consistent, indicating correct attention, MoD employs a complementary strategy to amplify critical information. Conversely, if the outputs are inconsistent, suggesting erroneous attention, MoD utilizes a contrastive strategy to suppress misleading information. Extensive experiments demonstrate that MoD significantly outperforms existing decoding methods across multiple mainstream benchmarks, effectively mitigating hallucinations in LVLMs. The code is available at https://github.com/xlchen0205/MoD.

Paperid: 1030, https://arxiv.org/pdf/2505.17019.pdf GitHub GitHub

Abstract:
Metaphorical comprehension in images remains a critical challenge for AI systems, as existing models struggle to grasp the nuanced cultural, emotional, and contextual implications embedded in visual content. While multimodal large language models (MLLMs) excel in basic Visual Question Answer (VQA) tasks, they struggle with a fundamental limitation on image implication tasks: contextual gaps that obscure the relationships between different visual elements and their abstract meanings. Inspired by the human cognitive process, we propose Let Androids Dream (LAD), a novel framework for image implication understanding and reasoning. LAD addresses contextual missing through the three-stage framework: (1) Perception: converting visual information into rich and multi-level textual representations, (2) Search: iteratively searching and integrating cross-domain knowledge to resolve ambiguity, and (3) Reasoning: generating context-alignment image implication via explicit reasoning. Our framework with the lightweight GPT-4o-mini model achieves SOTA performance compared to 15+ MLLMs on English image implication benchmark and a huge improvement on Chinese benchmark, performing comparable with the GPT-4o model on Multiple-Choice Question (MCQ) and outperforms 36.7% on Open-Style Question (OSQ). Additionally, our work provides new insights into how AI can more effectively interpret image implications, advancing the field of vision-language reasoning and human-AI interaction. Our project is publicly available at https://github.com/MING-ZCH/Let-Androids-Dream-of-Electric-Sheep.

Paperid: 1031, https://arxiv.org/pdf/2505.17016.pdf GitHub

Abstract:
We introduce RIPT-VLA, a simple and scalable reinforcement-learning-based interactive post-training paradigm that fine-tunes pretrained Vision-Language-Action (VLA) models using only sparse binary success rewards. Existing VLA training pipelines rely heavily on offline expert demonstration data and supervised imitation, limiting their ability to adapt to new tasks and environments under low-data regimes. RIPT-VLA addresses this by enabling interactive post-training with a stable policy optimization algorithm based on dynamic rollout sampling and leave-one-out advantage estimation. RIPT-VLA has the following characteristics. First, it applies to various VLA models, resulting in an improvement on the lightweight QueST model by 21.2%, and the 7B OpenVLA-OFT model to an unprecedented 97.5% success rate. Second, it is computationally efficient and data-efficient: with only one demonstration, RIPT-VLA enables an unworkable SFT model (4%) to succeed with a 97% success rate within 15 iterations. Furthermore, we demonstrate that the policy learned by RIPT-VLA generalizes across different tasks and scenarios and is robust to the initial state context. These results highlight RIPT-VLA as a practical and effective paradigm for post-training VLA models through minimal supervision.

Paperid: 1032, https://arxiv.org/pdf/2505.16854.pdf GitHub

Abstract:
Reinforcement Learning (RL) has proven to be an effective post-training strategy for enhancing reasoning in vision-language models (VLMs). Group Relative Policy Optimization (GRPO) is a recent prominent method that encourages models to generate complete reasoning traces before answering, leading to increased token usage and computational cost. Inspired by the human-like thinking process-where people skip reasoning for easy questions but think carefully when needed-we explore how to enable VLMs to first decide when reasoning is necessary. To realize this, we propose TON, a two-stage training strategy: (i) a supervised fine-tuning (SFT) stage with a simple yet effective 'thought dropout' operation, where reasoning traces are randomly replaced with empty thoughts. This introduces a think-or-not format that serves as a cold start for selective reasoning; (ii) a GRPO stage that enables the model to freely explore when to think or not, while maximizing task-aware outcome rewards. Experimental results show that TON can reduce the completion length by up to 90% compared to vanilla GRPO, without sacrificing performance or even improving it. Further evaluations across diverse vision-language tasks-covering a range of reasoning difficulties under both 3B and 7B models-consistently reveal that the model progressively learns to bypass unnecessary reasoning steps as training advances. These findings shed light on the path toward human-like reasoning patterns in reinforcement learning approaches. Our code is available at https://github.com/kokolerk/TON.

Paperid: 1033, https://arxiv.org/pdf/2505.16832.pdf GitHub GitHub

Abstract:
While foundation models (FMs), such as diffusion models and large vision-language models (LVLMs), have been widely applied in educational contexts, their ability to generate pedagogically effective visual explanations remains limited. Most existing approaches focus primarily on textual reasoning, overlooking the critical role of structured and interpretable visualizations in supporting conceptual understanding. To better assess the visual reasoning capabilities of FMs in educational settings, we introduce EduVisBench, a multi-domain, multi-level benchmark. EduVisBench features diverse STEM problem sets requiring visually grounded solutions, along with a fine-grained evaluation rubric informed by pedagogical theory. Our empirical analysis reveals that existing models frequently struggle with the inherent challenge of decomposing complex reasoning and translating it into visual representations aligned with human cognitive processes. To address these limitations, we propose EduVisAgent, a multi-agent collaborative framework that coordinates specialized agents for instructional planning, reasoning decomposition, metacognitive prompting, and visualization design. Experimental results show that EduVisAgent substantially outperforms all baselines, achieving a 40.2% improvement and delivering more educationally aligned visualizations. EduVisBench and EduVisAgent are available at https://github.com/aiming-lab/EduVisBench and https://github.com/aiming-lab/EduVisAgent.

Paperid: 1034, https://arxiv.org/pdf/2505.16815.pdf GitHub

Abstract:
Embodied AI has developed rapidly in recent years, but it is still mainly deployed in laboratories, with various distortions in the Real-world limiting its application. Traditionally, Image Quality Assessment (IQA) methods are applied to predict human preferences for distorted images; however, there is no IQA method to assess the usability of an image in embodied tasks, namely, the perceptual quality for robots. To provide accurate and reliable quality indicators for future embodied scenarios, we first propose the topic: IQA for Embodied AI. Specifically, we (1) based on the Mertonian system and meta-cognitive theory, constructed a perception-cognition-decision-execution pipeline and defined a comprehensive subjective score collection process; (2) established the Embodied-IQA database, containing over 36k reference/distorted image pairs, with more than 5m fine-grained annotations provided by Vision Language Models/Vision Language Action-models/Real-world robots; (3) trained and validated the performance of mainstream IQA methods on Embodied-IQA, demonstrating the need to develop more accurate quality indicators for Embodied AI. We sincerely hope that through evaluation, we can promote the application of Embodied AI under complex distortions in the Real-world. Project page: https://github.com/lcysyzxdxc/EmbodiedIQA

Paperid: 1035, https://arxiv.org/pdf/2505.16774.pdf GitHub

Abstract:
Large language models (LLMs) have demonstrated strong instruction-following capabilities in text-based tasks. However, this ability often deteriorates in multimodal models after alignment with non-text modalities such as images or audio. While several recent efforts have investigated instruction-following performance in text and vision-language models, instruction-following in audio-based large language models remains largely unexplored. To bridge this gap, we introduce IFEval-Audio, a novel evaluation dataset designed to assess the ability to follow instructions in an audio LLM. IFEval-Audio contains 280 audio-instruction-answer triples across six diverse dimensions: Content, Capitalization, Symbol, List Structure, Length, and Format. Each example pairs an audio input with a text instruction, requiring the model to generate an output that follows a specified structure. We benchmark state-of-the-art audio LLMs on their ability to follow audio-involved instructions. The dataset is released publicly to support future research in this emerging area.

Paperid: 1036, https://arxiv.org/pdf/2505.16647.pdf GitHub

Abstract:
We investigate fine-tuning Vision-Language Models (VLMs) for multi-task medical image understanding, focusing on detection, localization, and counting of findings in medical images. Our objective is to evaluate whether instruction-tuned VLMs can simultaneously improve these tasks, with the goal of enhancing diagnostic accuracy and efficiency. Using MedMultiPoints, a multimodal dataset with annotations from endoscopy (polyps and instruments) and microscopy (sperm cells), we reformulate each task into instruction-based prompts suitable for vision-language reasoning. We fine-tune Qwen2.5-VL-7B-Instruct using Low-Rank Adaptation (LoRA) across multiple task combinations. Results show that multi-task training improves robustness and accuracy. For example, it reduces the Count Mean Absolute Error (MAE) and increases Matching Accuracy in the Counting + Pointing task. However, trade-offs emerge, such as more zero-case point predictions, indicating reduced reliability in edge cases despite overall performance gains. Our study highlights the potential of adapting general-purpose VLMs to specialized medical tasks via prompt-driven fine-tuning. This approach mirrors clinical workflows, where radiologists simultaneously localize, count, and describe findings - demonstrating how VLMs can learn composite diagnostic reasoning patterns. The model produces interpretable, structured outputs, offering a promising step toward explainable and versatile medical AI. Code, model weights, and scripts will be released for reproducibility at https://github.com/simula/PointDetectCount.

Paperid: 1037, https://arxiv.org/pdf/2505.16640.pdf GitHub

Abstract:
Vision-Language-Action (VLA) models have advanced robotic control by enabling end-to-end decision-making directly from multimodal inputs. However, their tightly coupled architectures expose novel security vulnerabilities. Unlike traditional adversarial perturbations, backdoor attacks represent a stealthier, persistent, and practically significant threat-particularly under the emerging Training-as-a-Service paradigm-but remain largely unexplored in the context of VLA models. To address this gap, we propose BadVLA, a backdoor attack method based on Objective-Decoupled Optimization, which for the first time exposes the backdoor vulnerabilities of VLA models. Specifically, it consists of a two-stage process: (1) explicit feature-space separation to isolate trigger representations from benign inputs, and (2) conditional control deviations that activate only in the presence of the trigger, while preserving clean-task performance. Empirical results on multiple VLA benchmarks demonstrate that BadVLA consistently achieves near-100% attack success rates with minimal impact on clean task accuracy. Further analyses confirm its robustness against common input perturbations, task transfers, and model fine-tuning, underscoring critical security vulnerabilities in current VLA deployments. Our work offers the first systematic investigation of backdoor vulnerabilities in VLA models, highlighting an urgent need for secure and trustworthy embodied model design practices. We have released the project page at https://badvla-project.github.io/.

Paperid: 1038, https://arxiv.org/pdf/2505.16416.pdf GitHub

Abstract:
Rotary Position Embedding (RoPE) is a widely adopted technique for encoding relative positional information in large language models (LLMs). However, when extended to large vision-language models (LVLMs), its variants introduce unintended cross-modal positional biases. Specifically, they enforce relative positional dependencies between text token indices and image tokens, causing spurious alignments. This issue arises because image tokens representing the same content but located at different spatial positions are assigned distinct positional biases, leading to inconsistent cross-modal associations. To address this, we propose Per-Token Distance (PTD) - a simple yet effective metric for quantifying the independence of positional encodings across modalities. Informed by this analysis, we introduce Circle-RoPE, a novel encoding scheme that maps image token indices onto a circular trajectory orthogonal to the linear path of text token indices, forming a cone-like structure. This configuration ensures that each text token maintains an equal distance to all image tokens, reducing artificial cross-modal biases while preserving intra-image spatial information. To further enhance performance, we propose a staggered layer strategy that applies different RoPE variants across layers. This design leverages the complementary strengths of each RoPE variant, thereby enhancing the model's overall performance. Our experimental results demonstrate that our method effectively preserves spatial information from images while reducing relative positional bias, offering a more robust and flexible positional encoding framework for LVLMs. The code is available at [https://github.com/lose4578/CircleRoPE](https://github.com/lose4578/CircleRoPE).

Paperid: 1039, https://arxiv.org/pdf/2505.16411.pdf GitHub

Abstract:
Despite their remarkable progress in multimodal understanding tasks, large vision language models (LVLMs) often suffer from "hallucinations", generating texts misaligned with the visual context. Existing methods aimed at reducing hallucinations through inference time intervention incur a significant increase in latency. To mitigate this, we present SPIN, a task-agnostic attention-guided head suppression strategy that can be seamlessly integrated during inference, without incurring any significant compute or latency overhead. We investigate whether hallucination in LVLMs can be linked to specific model components. Our analysis suggests that hallucinations can be attributed to a dynamic subset of attention heads in each layer. Leveraging this insight, for each text query token, we selectively suppress attention heads that exhibit low attention to image tokens, keeping the top-K attention heads intact. Extensive evaluations on visual question answering and image description tasks demonstrate the efficacy of SPIN in reducing hallucination scores up to 2.7x while maintaining F1, and improving throughput by 1.8x compared to existing alternatives. Code is available at https://github.com/YUECHE77/SPIN.

Paperid: 1040, https://arxiv.org/pdf/2505.16377.pdf GitHub

Abstract:
Reinforcement learning (RL)-based autonomous driving policy learning faces critical limitations such as low sample efficiency and poor generalization; its reliance on online interactions and trial-and-error learning is especially unacceptable in safety-critical scenarios. Existing methods including safe RL often fail to capture the true semantic meaning of "safety" in complex driving contexts, leading to either overly conservative driving behavior or constraint violations. To address these challenges, we propose VL-SAFE, a world model-based safe RL framework with Vision-Language model (VLM)-as-safety-guidance paradigm, designed for offline safe policy learning. Specifically, we construct offline datasets containing data collected by expert agents and labeled with safety scores derived from VLMs. A world model is trained to generate imagined rollouts together with safety estimations, allowing the agent to perform safe planning without interacting with the real environment. Based on these imagined trajectories and safety evaluations, actor-critic learning is conducted under VLM-based safety guidance to optimize the driving policy more safely and efficiently. Extensive evaluations demonstrate that VL-SAFE achieves superior sample efficiency, generalization, safety, and overall performance compared to existing baselines. To the best of our knowledge, this is the first work that introduces a VLM-guided world model-based approach for safe autonomous driving. The demo video and code can be accessed at: https://ys-qu.github.io/vlsafe-website/

Paperid: 1041, https://arxiv.org/pdf/2505.16282.pdf GitHub

Abstract:
Training large language models (LLMs) as interactive agents for controlling graphical user interfaces (GUIs) presents a unique challenge to optimize long-horizon action sequences with multimodal feedback from complex environments. While recent works have advanced multi-turn reinforcement learning (RL) for reasoning and tool-using capabilities in LLMs, their application to GUI-based agents remains relatively underexplored due to the difficulty of sparse rewards, delayed feedback, and high rollout costs. In this paper, we investigate end-to-end policy optimization for vision-language-based GUI agents with the aim of improving performance on complex, long-horizon computer tasks. We propose Agentic Replay Policy Optimization (ARPO), an end-to-end RL approach that augments Group Relative Policy Optimization (GRPO) with a replay buffer to reuse the successful experience across training iterations. To further stabilize the training process, we propose a task selection strategy that filters tasks based on baseline agent performance, allowing the agent to focus on learning from informative interactions. Additionally, we compare ARPO with offline preference optimization approaches, highlighting the advantages of policy-based methods in GUI environments. Experiments on the OSWorld benchmark demonstrate that ARPO achieves competitive results, establishing a new performance baseline for LLM-based GUI agents trained via reinforcement learning. Our findings underscore the effectiveness of reinforcement learning for training multi-turn, vision-language GUI agents capable of managing complex real-world UI interactions. Codes and models:https://github.com/dvlab-research/ARPO.git.

Paperid: 1042, https://arxiv.org/pdf/2505.16278.pdf GitHub

Abstract:
End-to-end autonomous driving (E2E-AD) demands effective processing of multi-view sensory data and robust handling of diverse and complex driving scenarios, particularly rare maneuvers such as aggressive turns. Recent success of Mixture-of-Experts (MoE) architecture in Large Language Models (LLMs) demonstrates that specialization of parameters enables strong scalability. In this work, we propose DriveMoE, a novel MoE-based E2E-AD framework, with a Scene-Specialized Vision MoE and a Skill-Specialized Action MoE. DriveMoE is built upon our $Ï_0$ Vision-Language-Action (VLA) baseline (originally from the embodied AI field), called Drive-$Ï_0$. Specifically, we add Vision MoE to Drive-$Ï_0$ by training a router to select relevant cameras according to the driving context dynamically. This design mirrors human driving cognition, where drivers selectively attend to crucial visual cues rather than exhaustively processing all visual information. In addition, we add Action MoE by training another router to activate specialized expert modules for different driving behaviors. Through explicit behavioral specialization, DriveMoE is able to handle diverse scenarios without suffering from modes averaging like existing models. In Bench2Drive closed-loop evaluation experiments, DriveMoE achieves state-of-the-art (SOTA) performance, demonstrating the effectiveness of combining vision and action MoE in autonomous driving tasks. We will release our code and models of DriveMoE and Drive-$Ï_0$.

Paperid: 1043, https://arxiv.org/pdf/2505.16146.pdf GitHub

Abstract:
Large vision-language models (LVLMs) have achieved remarkable performance on multimodal tasks. However, they still suffer from hallucinations, generating text inconsistent with visual input, posing significant risks in real-world applications. Existing approaches to address this issue focus on incorporating external knowledge bases, alignment training, or decoding strategies, all of which require substantial computational cost and time. Recent works try to explore more efficient alternatives by adjusting LVLMs' internal representations. Although promising, these methods may cause hallucinations to be insufficiently suppressed or lead to excessive interventions that negatively affect normal semantics. In this work, we leverage sparse autoencoders (SAEs) to identify semantic directions closely associated with faithfulness or hallucination, extracting more precise and disentangled hallucination-related representations. Our analysis demonstrates that interventions along the identified faithful direction can mitigate hallucinations, while those along the hallucinatory direction can exacerbate them. Building on these insights, we propose Steering LVLMs via SAE Latent Directions (SSL), a plug-and-play method based on SAE-derived latent directions to mitigate hallucinations in LVLMs. Extensive experiments demonstrate that SSL significantly outperforms existing decoding approaches in mitigating hallucinations, while maintaining transferability across different model architectures with negligible additional time overhead. The code is available at https://github.com/huazhenglin2003/SSL.

Paperid: 1044, https://arxiv.org/pdf/2505.15966.pdf GitHub

Abstract:
Chain-of-thought reasoning has significantly improved the performance of Large Language Models (LLMs) across various domains. However, this reasoning process has been confined exclusively to textual space, limiting its effectiveness in visually intensive tasks. To address this limitation, we introduce the concept of reasoning in the pixel-space. Within this novel framework, Vision-Language Models (VLMs) are equipped with a suite of visual reasoning operations, such as zoom-in and select-frame. These operations enable VLMs to directly inspect, interrogate, and infer from visual evidences, thereby enhancing reasoning fidelity for visual tasks. Cultivating such pixel-space reasoning capabilities in VLMs presents notable challenges, including the model's initially imbalanced competence and its reluctance to adopt the newly introduced pixel-space operations. We address these challenges through a two-phase training approach. The first phase employs instruction tuning on synthesized reasoning traces to familiarize the model with the novel visual operations. Following this, a reinforcement learning (RL) phase leverages a curiosity-driven reward scheme to balance exploration between pixel-space reasoning and textual reasoning. With these visual operations, VLMs can interact with complex visual inputs, such as information-rich images or videos to proactively gather necessary information. We demonstrate that this approach significantly improves VLM performance across diverse visual reasoning benchmarks. Our 7B model, \model, achieves 84\% on V* bench, 74\% on TallyQA-Complex, and 84\% on InfographicsVQA, marking the highest accuracy achieved by any open-source model to date. These results highlight the importance of pixel-space reasoning and the effectiveness of our framework.

Paperid: 1045, https://arxiv.org/pdf/2505.15952.pdf GitHub

Abstract:
With video games now generating the highest revenues in the entertainment industry, optimizing game development workflows has become essential for the sector's sustained growth. Recent advancements in Vision-Language Models (VLMs) offer considerable potential to automate and enhance various aspects of game development, particularly Quality Assurance (QA), which remains one of the industry's most labor-intensive processes with limited automation options. To accurately evaluate the performance of VLMs in video game QA tasks and determine their effectiveness in handling real-world scenarios, there is a clear need for standardized benchmarks, as existing benchmarks are insufficient to address the specific requirements of this domain. To bridge this gap, we introduce VideoGameQA-Bench, a comprehensive benchmark that covers a wide array of game QA activities, including visual unit testing, visual regression testing, needle-in-a-haystack tasks, glitch detection, and bug report generation for both images and videos of various games. Code and data are available at: https://asgaardlab.github.io/videogameqa-bench/

Paperid: 1046, https://arxiv.org/pdf/2505.15870.pdf GitHub

Abstract:
Commuting Origin-destination~(OD) flows, capturing daily population mobility of citizens, are vital for sustainable development across cities around the world. However, it is challenging to obtain the data due to the high cost of travel surveys and privacy concerns. Surprisingly, we find that satellite imagery, publicly available across the globe, contains rich urban semantic signals to support high-quality OD flow generation, with over 98\% expressiveness of traditional multisource hard-to-collect urban sociodemographic, economics, land use, and point of interest data. This inspires us to design a novel data generator, GlODGen, which can generate OD flow data for any cities of interest around the world. Specifically, GlODGen first leverages Vision-Language Geo-Foundation Models to extract urban semantic signals related to human mobility from satellite imagery. These features are then combined with population data to form region-level representations, which are used to generate OD flows via graph diffusion models. Extensive experiments on 4 continents and 6 representative cities show that GlODGen has great generalizability across diverse urban environments on different continents and can generate OD flow data for global cities highly consistent with real-world mobility data. We implement GlODGen as an automated tool, seamlessly integrating data acquisition and curation, urban semantic feature extraction, and OD flow generation together. It has been released at https://github.com/tsinghua-fib-lab/generate-od-pubtools.

Paperid: 1047, https://arxiv.org/pdf/2505.15660.pdf GitHub

Abstract:
The generalization capabilities of vision-language-action (VLA) models to unseen tasks are crucial to achieving general-purpose robotic manipulation in open-world settings. However, the cross-task generalization capabilities of existing VLA models remain significantly underexplored. To address this gap, we introduce AGNOSTOS, a novel simulation benchmark designed to rigorously evaluate cross-task zero-shot generalization in manipulation. AGNOSTOS comprises 23 unseen manipulation tasks for testing, distinct from common training task distributions, and incorporates two levels of generalization difficulty to assess robustness. Our systematic evaluation reveals that current VLA models, despite being trained on diverse datasets, struggle to generalize effectively to these unseen tasks. To overcome this limitation, we propose Cross-Task In-Context Manipulation (X-ICM), a method that conditions large language models (LLMs) on in-context demonstrations from seen tasks to predict action sequences for unseen tasks. Additionally, we introduce a dynamics-guided sample selection strategy that identifies relevant demonstrations by capturing cross-task dynamics. On AGNOSTOS, X-ICM significantly improves cross-task zero-shot generalization performance over leading VLAs. We believe AGNOSTOS and X-ICM will serve as valuable tools for advancing general-purpose robotic manipulation.

Paperid: 1048, https://arxiv.org/pdf/2505.15576.pdf GitHub

Abstract:
Vision-Language Models (VLMs) are essential for multimodal tasks, especially compositional reasoning (CR) tasks, which require distinguishing fine-grained semantic differences between visual and textual embeddings. However, existing methods primarily fine-tune the model by generating text-based hard negative samples, neglecting the importance of image-based negative samples, which results in insufficient training of the visual encoder and ultimately impacts the overall performance of the model. Moreover, negative samples are typically treated uniformly, without considering their difficulty levels, and the alignment of positive samples is insufficient, which leads to challenges in aligning difficult sample pairs. To address these issues, we propose Adaptive Hard Negative Perturbation Learning (AHNPL). AHNPL translates text-based hard negatives into the visual domain to generate semantically disturbed image-based negatives for training the model, thereby enhancing its overall performance. AHNPL also introduces a contrastive learning approach using a multimodal hard negative loss to improve the model's discrimination of hard negatives within each modality and a dynamic margin loss that adjusts the contrastive margin according to sample difficulty to enhance the distinction of challenging sample pairs. Experiments on three public datasets demonstrate that our method effectively boosts VLMs' performance on complex CR tasks. The source code is available at https://github.com/nynu-BDAI/AHNPL.

Paperid: 1049, https://arxiv.org/pdf/2505.15435.pdf GitHub GitHub

Abstract:
Reasoning about temporal causality, particularly irreversible transformations of objects governed by real-world knowledge (e.g., fruit decay and human aging), is a fundamental aspect of human visual understanding. Unlike temporal perception based on simple event sequences, this form of reasoning requires a deeper comprehension of how object states change over time. Although the current powerful Vision-Language Models (VLMs) have demonstrated impressive performance on a wide range of downstream tasks, their capacity to reason about temporal causality remains underexplored. To address this gap, we introduce \textbf{TimeCausality}, a novel benchmark specifically designed to evaluate the causal reasoning ability of VLMs in the temporal dimension. Based on our TimeCausality, we find that while the current SOTA open-source VLMs have achieved performance levels comparable to closed-source models like GPT-4o on various standard visual question answering tasks, they fall significantly behind on our benchmark compared with their closed-source competitors. Furthermore, even GPT-4o exhibits a marked drop in performance on TimeCausality compared to its results on other tasks. These findings underscore the critical need to incorporate temporal causality into the evaluation and development of VLMs, and they highlight an important challenge for the open-source VLM community moving forward. Code and Data are available at \href{https://github.com/Zeqing-Wang/TimeCausality }{TimeCausality}.

Paperid: 1050, https://arxiv.org/pdf/2505.15425.pdf GitHub

Abstract:
Medical Vision-Language Models (MVLMs) have achieved par excellence generalization in medical image analysis, yet their performance under noisy, corrupted conditions remains largely untested. Clinical imaging is inherently susceptible to acquisition artifacts and noise; however, existing evaluations predominantly assess generally clean datasets, overlooking robustness -- i.e., the model's ability to perform under real-world distortions. To address this gap, we first introduce MediMeta-C, a corruption benchmark that systematically applies several perturbations across multiple medical imaging datasets. Combined with MedMNIST-C, this establishes a comprehensive robustness evaluation framework for MVLMs. We further propose RobustMedCLIP, a visual encoder adaptation of a pretrained MVLM that incorporates few-shot tuning to enhance resilience against corruptions. Through extensive experiments, we benchmark 5 major MVLMs across 5 medical imaging modalities, revealing that existing models exhibit severe degradation under corruption and struggle with domain-modality tradeoffs. Our findings highlight the necessity of diverse training and robust adaptation strategies, demonstrating that efficient low-rank adaptation when paired with few-shot tuning, improves robustness while preserving generalization across modalities.

Paperid: 1051, https://arxiv.org/pdf/2505.15389.pdf GitHub

Abstract:
Rapid deployment of vision-language models (VLMs) magnifies safety risks, yet most evaluations rely on artificial images. This study asks: How safe are current VLMs when confronted with meme images that ordinary users share? To investigate this question, we introduce MemeSafetyBench, a 50,430-instance benchmark pairing real meme images with both harmful and benign instructions. Using a comprehensive safety taxonomy and LLM-based instruction generation, we assess multiple VLMs across single and multi-turn interactions. We investigate how real-world memes influence harmful outputs, the mitigating effects of conversational context, and the relationship between model scale and safety metrics. Our findings demonstrate that VLMs are more vulnerable to meme-based harmful prompts than to synthetic or typographic images. Memes significantly increase harmful responses and decrease refusals compared to text-only inputs. Though multi-turn interactions provide partial mitigation, elevated vulnerability persists. These results highlight the need for ecologically valid evaluations and stronger safety mechanisms. MemeSafetyBench is publicly available at https://github.com/oneonlee/Meme-Safety-Bench.

Paperid: 1052, https://arxiv.org/pdf/2505.15232.pdf GitHub

Abstract:
3D scene understanding plays a fundamental role in vision applications such as robotics, autonomous driving, and augmented reality. However, advancing learning-based 3D scene understanding remains challenging due to two key limitations: (1) the large scale and complexity of 3D scenes lead to higher computational costs and slower training compared to 2D counterparts; and (2) high-quality annotated 3D datasets are significantly scarcer than those available for 2D vision. These challenges underscore the need for more efficient learning paradigms. In this work, we propose DC-Scene, a data-centric framework tailored for 3D scene understanding, which emphasizes enhancing data quality and training efficiency. Specifically, we introduce a CLIP-driven dual-indicator quality (DIQ) filter, combining vision-language alignment scores with caption-loss perplexity, along with a curriculum scheduler that progressively expands the training pool from the top 25% to 75% of scene-caption pairs. This strategy filters out noisy samples and significantly reduces dependence on large-scale labeled 3D data. Extensive experiments on ScanRefer and Nr3D demonstrate that DC-Scene achieves state-of-the-art performance (86.1 CIDEr with the top-75% subset vs. 85.4 with the full dataset) while reducing training cost by approximately two-thirds, confirming that a compact set of high-quality samples can outperform exhaustive training. Code will be available at https://github.com/AIGeeksGroup/DC-Scene.

Paperid: 1053, https://arxiv.org/pdf/2505.15197.pdf GitHub

Abstract:
When humans speak, gestures help convey communicative intentions, such as adding emphasis or describing concepts. However, current co-speech gesture generation methods rely solely on superficial linguistic cues (\textit{e.g.} speech audio or text transcripts), neglecting to understand and leverage the communicative intention that underpins human gestures. This results in outputs that are rhythmically synchronized with speech but are semantically shallow. To address this gap, we introduce \textbf{Intentional-Gesture}, a novel framework that casts gesture generation as an intention-reasoning task grounded in high-level communicative functions. % First, we curate the \textbf{InG} dataset by augmenting BEAT-2 with gesture-intention annotations (\textit{i.e.}, text sentences summarizing intentions), which are automatically annotated using large vision-language models. Next, we introduce the \textbf{Intentional Gesture Motion Tokenizer} to leverage these intention annotations. It injects high-level communicative functions (\textit{e.g.}, intentions) into tokenized motion representations to enable intention-aware gesture synthesis that are both temporally aligned and semantically meaningful, achieving new state-of-the-art performance on the BEAT-2 benchmark. Our framework offers a modular foundation for expressive gesture generation in digital humans and embodied AI. Project Page: https://andypinxinliu.github.io/Intentional-Gesture

Paperid: 1054, https://arxiv.org/pdf/2505.15008.pdf GitHub

Abstract:
Selective classification enhances the reliability of predictive models by allowing them to abstain from making uncertain predictions. In this work, we revisit the design of optimal selection functions through the lens of the Neyman--Pearson lemma, a classical result in statistics that characterizes the optimal rejection rule as a likelihood ratio test. We show that this perspective not only unifies the behavior of several post-hoc selection baselines, but also motivates new approaches to selective classification which we propose here. A central focus of our work is the setting of covariate shift, where the input distribution at test time differs from that at training. This realistic and challenging scenario remains relatively underexplored in the context of selective classification. We evaluate our proposed methods across a range of vision and language tasks, including both supervised learning and vision-language models. Our experiments demonstrate that our Neyman--Pearson-informed methods consistently outperform existing baselines, indicating that likelihood ratio-based selection offers a robust mechanism for improving selective classification under covariate shifts. Our code is publicly available at https://github.com/clear-nus/sc-likelihood-ratios.

Paperid: 1055, https://arxiv.org/pdf/2505.14671.pdf GitHub GitHub

Abstract:
Personalized models have demonstrated remarkable success in understanding and generating concepts provided by users. However, existing methods use separate concept tokens for understanding and generation, treating these tasks in isolation. This may result in limitations for generating images with complex prompts. For example, given the concept $\langle bo\rangle$, generating "$\langle bo\rangle$ wearing its hat" without additional textual descriptions of its hat. We call this kind of generation personalized knowledge-driven generation. To address the limitation, we present UniCTokens, a novel framework that effectively integrates personalized information into a unified vision language model (VLM) for understanding and generation. UniCTokens trains a set of unified concept tokens to leverage complementary semantics, boosting two personalized tasks. Moreover, we propose a progressive training strategy with three stages: understanding warm-up, bootstrapping generation from understanding, and deepening understanding from generation to enhance mutual benefits between both tasks. To quantitatively evaluate the unified VLM personalization, we present UnifyBench, the first benchmark for assessing concept understanding, concept generation, and knowledge-driven generation. Experimental results on UnifyBench indicate that UniCTokens shows competitive performance compared to leading methods in concept understanding, concept generation, and achieving state-of-the-art results in personalized knowledge-driven generation. Our research demonstrates that enhanced understanding improves generation, and the generation process can yield valuable insights into understanding. Our code and dataset will be released at: \href{https://github.com/arctanxarc/UniCTokens}{https://github.com/arctanxarc/UniCTokens}.

Paperid: 1056, https://arxiv.org/pdf/2505.14646.pdf GitHub

Abstract:
Efficient creation of accurate and editable 3D CAD models is critical in engineering design, significantly impacting cost and time-to-market in product innovation. Current manual workflows remain highly time-consuming and demand extensive user expertise. While recent developments in AI-driven CAD generation show promise, existing models are limited by incomplete representations of CAD operations, inability to generalize to real-world images, and low output accuracy. This paper introduces CAD-Coder, an open-source Vision-Language Model (VLM) explicitly fine-tuned to generate editable CAD code (CadQuery Python) directly from visual input. Leveraging a novel dataset that we created--GenCAD-Code, consisting of over 163k CAD-model image and code pairs--CAD-Coder outperforms state-of-the-art VLM baselines such as GPT-4.5 and Qwen2.5-VL-72B, achieving a 100% valid syntax rate and the highest accuracy in 3D solid similarity. Notably, our VLM demonstrates some signs of generalizability, successfully generating CAD code from real-world images and executing CAD operations unseen during fine-tuning. The performance and adaptability of CAD-Coder highlights the potential of VLMs fine-tuned on code to streamline CAD workflows for engineers and designers. CAD-Coder is publicly available at: https://github.com/anniedoris/CAD-Coder.

Paperid: 1057, https://arxiv.org/pdf/2505.14362.pdf GitHub

Abstract:
Large Vision-Language Models (VLMs) have shown strong capabilities in multimodal understanding and reasoning, yet they are primarily constrained by text-based reasoning processes. However, achieving seamless integration of visual and textual reasoning which mirrors human cognitive processes remains a significant challenge. In particular, effectively incorporating advanced visual input processing into reasoning mechanisms is still an open question. Thus, in this paper, we explore the interleaved multimodal reasoning paradigm and introduce DeepEyes, a model with "thinking with images" capabilities incentivized through end-to-end reinforcement learning without the need for cold-start SFT. Notably, this ability emerges natively within the model itself, leveraging its inherent grounding ability as a tool instead of depending on separate specialized models. Specifically, we propose a tool-use-oriented data selection mechanism and a reward strategy to encourage successful tool-assisted reasoning trajectories. DeepEyes achieves significant performance gains on fine-grained perception and reasoning benchmarks and also demonstrates improvement in grounding, hallucination, and mathematical reasoning tasks. Interestingly, we observe the distinct evolution of tool-calling behavior from initial exploration to efficient and accurate exploitation, and diverse thinking patterns that closely mirror human visual reasoning processes. Code is available at https://github.com/Visual-Agent/DeepEyes.

Paperid: 1058, https://arxiv.org/pdf/2505.14246.pdf GitHub

Abstract:
A key trend in Large Reasoning Models (e.g., OpenAI's o3) is the native agentic ability to use external tools such as web browsers for searching and writing/executing code for image manipulation to think with images. In the open-source research community, while significant progress has been made in language-only agentic abilities such as function calling and tool integration, the development of multi-modal agentic capabilities that involve truly thinking with images, and their corresponding benchmarks, are still less explored. This work highlights the effectiveness of Visual Agentic Reinforcement Fine-Tuning (Visual-ARFT) for enabling flexible and adaptive reasoning abilities for Large Vision-Language Models (LVLMs). With Visual-ARFT, open-source LVLMs gain the ability to browse websites for real-time information updates and write code to manipulate and analyze input images through cropping, rotation, and other image processing techniques. We also present a Multi-modal Agentic Tool Bench (MAT) with two settings (MAT-Search and MAT-Coding) designed to evaluate LVLMs' agentic search and coding abilities. Our experimental results demonstrate that Visual-ARFT outperforms its baseline by +18.6% F1 / +13.0% EM on MAT-Coding and +10.3% F1 / +8.7% EM on MAT-Search, ultimately surpassing GPT-4o. Visual-ARFT also achieves +29.3 F1% / +25.9% EM gains on existing multi-hop QA benchmarks such as 2Wiki and HotpotQA, demonstrating strong generalization capabilities. Our findings suggest that Visual-ARFT offers a promising path toward building robust and generalizable multimodal agents.

Paperid: 1059, https://arxiv.org/pdf/2505.13921.pdf GitHub

Abstract:
Large Language Models (LLMs) demonstrate strong reasoning and task planning capabilities but remain fundamentally limited in physical interaction modeling. Existing approaches integrate perception via Vision-Language Models (VLMs) or adaptive decision-making through Reinforcement Learning (RL), but they fail to capture dynamic object interactions or require task-specific training, limiting their real-world applicability. We introduce APEX (Anticipatory Physics-Enhanced Execution), a framework that equips LLMs with physics-driven foresight for real-time task planning. APEX constructs structured graphs to identify and model the most relevant dynamic interactions in the environment, providing LLMs with explicit physical state updates. Simultaneously, APEX provides low-latency forward simulations of physically feasible actions, allowing LLMs to select optimal strategies based on predictive outcomes rather than static observations. We evaluate APEX on three benchmarks designed to assess perception, prediction, and decision-making: (1) Physics Reasoning Benchmark, testing causal inference and object motion prediction; (2) Tetris, evaluating whether physics-informed prediction enhances decision-making performance in long-horizon planning tasks; (3) Dynamic Obstacle Avoidance, assessing the immediate integration of perception and action feasibility analysis. APEX significantly outperforms standard LLMs and VLM-based models, demonstrating the necessity of explicit physics reasoning for bridging the gap between language-based intelligence and real-world task execution. The source code and experiment setup are publicly available at https://github.com/hwj20/APEX_EXP .

Paperid: 1060, https://arxiv.org/pdf/2505.13888.pdf GitHub

Abstract:
Leveraging pretrained Vision-Language Models (VLMs) to map language instruction and visual observations to raw low-level actions, Vision-Language-Action models (VLAs) hold great promise for achieving general-purpose robotic systems. Despite their advancements, existing VLAs tend to spuriously correlate task-irrelevant visual features with actions, limiting their generalization capacity beyond the training data. To tackle this challenge, we propose Intrinsic Spatial Reasoning (InSpire), a simple yet effective approach that mitigates the adverse effects of spurious correlations by boosting the spatial reasoning ability of VLAs. Specifically, InSpire redirects the VLA's attention to task-relevant factors by prepending the question "In which direction is the [object] relative to the robot?" to the language instruction and aligning the answer "right/left/up/down/front/back/grasped" and predicted actions with the ground-truth. Notably, InSpire can be used as a plugin to enhance existing autoregressive VLAs, requiring no extra training data or interaction with other large models. Extensive experimental results in both simulation and real-world environments demonstrate the effectiveness and flexibility of our approach. Our code, pretrained models and demos are publicly available at: https://Koorye.github.io/proj/Inspire.

Paperid: 1061, https://arxiv.org/pdf/2505.13669.pdf GitHub

Abstract:
Cross-view geo-localisation identifies coarse geographical position of an automated vehicle by matching a ground-level image to a geo-tagged satellite image from a database. Despite the advancements in Cross-view geo-localisation, significant challenges still persist such as similar looking scenes which makes it challenging to find the correct match as the top match. Existing approaches reach high recall rates but they still fail to rank the correct image as the top match. To address this challenge, this paper proposes GeoVLM, a novel approach which uses the zero-shot capabilities of vision language models to enable cross-view geo-localisation using interpretable cross-view language descriptions. GeoVLM is a trainable reranking approach which improves the best match accuracy of cross-view geo-localisation. GeoVLM is evaluated on standard benchmark VIGOR and University-1652 and also through real-life driving environments using Cross-View United Kingdom, a new benchmark dataset introduced in this paper. The results of the paper show that GeoVLM improves retrieval performance of cross-view geo-localisation compared to the state-of-the-art methods with the help of explainable natural language descriptions. The code is available at https://github.com/CAV-Research-Lab/GeoVLM

Paperid: 1062, https://arxiv.org/pdf/2505.13427.pdf GitHub

Abstract:
While Multimodal Large Language Models (MLLMs) have achieved impressive progress in vision-language understanding, they still struggle with complex multi-step reasoning, often producing logically inconsistent or partially correct solutions. A key limitation lies in the lack of fine-grained supervision over intermediate reasoning steps. To address this, we propose MM-PRM, a process reward model trained within a fully automated, scalable framework. We first build MM-Policy, a strong multimodal model trained on diverse mathematical reasoning data. Then, we construct MM-K12, a curated dataset of 10,000 multimodal math problems with verifiable answers, which serves as seed data. Leveraging a Monte Carlo Tree Search (MCTS)-based pipeline, we generate over 700k step-level annotations without human labeling. The resulting PRM is used to score candidate reasoning paths in the Best-of-N inference setup and achieves significant improvements across both in-domain (MM-K12 test set) and out-of-domain (OlympiadBench, MathVista, etc.) benchmarks. Further analysis confirms the effectiveness of soft labels, smaller learning rates, and path diversity in optimizing PRM performance. MM-PRM demonstrates that process supervision is a powerful tool for enhancing the logical robustness of multimodal reasoning systems. We release all our codes and data at https://github.com/ModalMinds/MM-PRM.

Paperid: 1063, https://arxiv.org/pdf/2505.13426.pdf GitHub

Abstract:
Vision-Language Models (VLMs) excel in many direct multimodal tasks but struggle to translate this prowess into effective decision-making within interactive, visually rich environments like games. This ``knowing-doing'' gap significantly limits their potential as autonomous agents, as leading VLMs often performing badly in simple games. To address this, we introduce VLM-Gym, a curated reinforcement learning (RL) environment featuring diverse visual games with unified interfaces and adjustable, compositional difficulty, specifically designed for scalable multi-game parallel training. Leveraging VLM-Gym, we train G0 models using pure RL-driven self-evolution, which demonstrate emergent perception and reasoning patterns. To further mitigate challenges arising from game diversity, we develop G1 models. G1 incorporates a perception-enhanced cold start prior to RL fine-tuning. Our resulting G1 models consistently surpass their teacher across all games and outperform leading proprietary models like Claude-3.7-Sonnet-Thinking. Systematic analysis reveals an intriguing finding: perception and reasoning abilities mutually bootstrap each other throughout the RL training process. Source code including VLM-Gym and RL training are released at https://github.com/chenllliang/G1 to foster future research in advancing VLMs as capable interactive agents.

Paperid: 1064, https://arxiv.org/pdf/2505.13317.pdf GitHub

Abstract:
Semi-supervised learning (SSL) alleviates the cost of data labeling process by exploiting unlabeled data, and has achieved promising results on various tasks such as image classification. Meanwhile, the Pretrain-Finetuning paradigm has garnered significant attention in recent years, and exploiting pre-trained models could also reduce the requirement of labeled data in downstream tasks. Therefore, a question naturally occurs: \emph{When the labeled data is scarce in the target tasks, should we exploit unlabeled data or pre-trained models?} To answer this question, we select pre-trained Vision-Language Models (VLMs) as representative pretrain-finetuning instances and propose \textit{Few-shot SSL} -- a framework that enables fair comparison between these two paradigms by controlling the amount of labeled data used. Extensive experiments across various settings demonstrate that pre-trained VLMs generally outperform SSL methods in nearly all cases, except when the data has low resolution or lacks clear semantic structure. Therefore, we encourage future SSL research to compare with pre-trained models and explore deeper integration, such as using pre-trained knowledge to enhance pseudo-labeling. To support future research, we release our unified reproduction and evaluation framework. Codes are available \href{https://anonymous.4open.science/r/Rethinking-SSL-and-Pretrain-Finetuning-5566 }{here}.

Paperid: 1065, https://arxiv.org/pdf/2505.13302.pdf GitHub

Abstract:
Large language models are increasingly integrated into news recommendation systems, raising concerns about their role in spreading misinformation. In humans, visual content is known to boost credibility and shareability of information, yet its effect on vision-language models (VLMs) remains unclear. We present the first study examining how images influence VLMs' propensity to reshare news content, whether this effect varies across model families, and how persona conditioning and content attributes modulate this behavior. To support this analysis, we introduce two methodological contributions: a jailbreaking-inspired prompting strategy that elicits resharing decisions from VLMs while simulating users with antisocial traits and political alignments; and a multimodal dataset of fact-checked political news from PolitiFact, paired with corresponding images and ground-truth veracity labels. Experiments across model families reveal that image presence increases resharing rates by 4.8% for true news and 15.0% for false news. Persona conditioning further modulates this effect: Dark Triad traits amplify resharing of false news, whereas Republican-aligned profiles exhibit reduced veracity sensitivity. Of all the tested models, only Claude-3-Haiku demonstrates robustness to visual misinformation. These findings highlight emerging risks in multimodal model behavior and motivate the development of tailored evaluation frameworks and mitigation strategies for personalized AI systems. Code and dataset are available at: https://github.com/3lis/misinfo_vlm

Paperid: 1066, https://arxiv.org/pdf/2505.13233.pdf GitHub GitHub

Abstract:
Pretrained vision-language models (VLMs), e.g., CLIP, demonstrate impressive zero-shot capabilities on downstream tasks. Prior research highlights the crucial role of visual augmentation techniques, like random cropping, in alignment with fine-grained class descriptions generated by large language models (LLMs), significantly enhancing zero-shot performance by incorporating multi-view information. However, the inherent randomness of these augmentations can inevitably introduce background artifacts and cause models to overly focus on local details, compromising global semantic understanding. To address these issues, we propose an \textbf{A}ttention-\textbf{B}ased \textbf{S}election (\textbf{ABS}) method from local details to global context, which applies attention-guided cropping in both raw images and feature space, supplement global semantic information through strategic feature selection. Additionally, we introduce a soft matching technique to effectively filter LLM descriptions for better alignment. \textbf{ABS} achieves state-of-the-art performance on out-of-distribution generalization and zero-shot classification tasks. Notably, \textbf{ABS} is training-free and even rivals few-shot and test-time adaptation methods. Our code is available at \href{https://github.com/BIT-DA/ABS}{\textcolor{darkgreen}{https://github.com/BIT-DA/ABS}}.

Paperid: 1067, https://arxiv.org/pdf/2505.13061.pdf GitHub

Abstract:
3D visual illusion is a perceptual phenomenon where a two-dimensional plane is manipulated to simulate three-dimensional spatial relationships, making a flat artwork or object look three-dimensional in the human visual system. In this paper, we reveal that the machine visual system is also seriously fooled by 3D visual illusions, including monocular and binocular depth estimation. In order to explore and analyze the impact of 3D visual illusion on depth estimation, we collect a large dataset containing almost 3k scenes and 200k images to train and evaluate SOTA monocular and binocular depth estimation methods. We also propose a robust depth estimation framework that uses common sense from a vision-language model to adaptively select reliable depth from binocular disparity and monocular depth. Experiments show that SOTA monocular, binocular, and multi-view depth estimation approaches are all fooled by various 3D visual illusions, while our method achieves SOTA performance.

Paperid: 1068, https://arxiv.org/pdf/2505.13031.pdf GitHub

Abstract:
Recent text-to-image systems face limitations in handling multimodal inputs and complex reasoning tasks. We introduce MindOmni, a unified multimodal large language model that addresses these challenges by incorporating reasoning generation through reinforcement learning. MindOmni leverages a three-phase training strategy: i) design of a unified vision language model with a decoder-only diffusion module, ii) supervised fine-tuning with Chain-of-Thought (CoT) instruction data, and iii) our proposed Reasoning Generation Policy Optimization (RGPO) algorithm, utilizing multimodal feedback to effectively guide policy updates. Experimental results demonstrate that MindOmni outperforms existing models, achieving impressive performance on both understanding and generation benchmarks, meanwhile showcasing advanced fine-grained reasoning generation capabilities, especially with mathematical reasoning instruction. All codes will be made public at https://github.com/TencentARC/MindOmni

Paperid: 1069, https://arxiv.org/pdf/2505.12912.pdf GitHub

Abstract:
Pre-trained vision-language models such as contrastive language-image pre-training (CLIP) have demonstrated a remarkable generalizability, which has enabled a wide range of applications represented by zero-shot classification. However, vision-language models still suffer when they face datasets with large gaps from training ones, i.e., distribution shifts. We found that CLIP is especially vulnerable to sensor degradation, a type of realistic distribution shift caused by sensor conditions such as weather, light, or noise. Collecting a new dataset from a test distribution for fine-tuning highly costs since sensor degradation occurs unexpectedly and has a range of variety. Thus, we investigate test-time adaptation (TTA) of zero-shot classification, which enables on-the-fly adaptation to the test distribution with unlabeled test data. Existing TTA methods for CLIP mainly focus on modifying image and text embeddings or predictions to address distribution shifts. Although these methods can adapt to domain shifts, such as fine-grained labels spaces or different renditions in input images, they fail to adapt to distribution shifts caused by sensor degradation. We found that this is because image embeddings are "corrupted" in terms of uniformity, a measure related to the amount of information. To make models robust to sensor degradation, we propose a novel method called uniformity-aware information-balanced TTA (UnInfo). To address the corruption of image embeddings, we introduce uniformity-aware confidence maximization, information-aware loss balancing, and knowledge distillation from the exponential moving average (EMA) teacher. Through experiments, we demonstrate that our UnInfo improves accuracy under sensor degradation by retaining information in terms of uniformity.

Paperid: 1070, https://arxiv.org/pdf/2505.12650.pdf GitHub

Abstract:
Machine learning-based interatomic potentials and force fields depend critically on accurate atomic structures, yet such data are scarce due to the limited availability of experimentally resolved crystals. Although atomic-resolution electron microscopy offers a potential source of structural data, converting these images into simulation-ready formats remains labor-intensive and error-prone, creating a bottleneck for model training and validation. We introduce AutoMat, an end-to-end, agent-assisted pipeline that automatically transforms scanning transmission electron microscopy (STEM) images into atomic crystal structures and predicts their physical properties. AutoMat combines pattern-adaptive denoising, physics-guided template retrieval, symmetry-aware atomic reconstruction, fast relaxation and property prediction via MatterSim, and coordinated orchestration across all stages. We propose the first dedicated STEM2Mat-Bench for this task and evaluate performance using lattice RMSD, formation energy MAE, and structure-matching success rate. By orchestrating external tool calls, AutoMat enables a text-only LLM to outperform vision-language models in this domain, achieving closed-loop reasoning throughout the pipeline. In large-scale experiments over 450 structure samples, AutoMat substantially outperforms existing multimodal large language models and tools. These results validate both AutoMat and STEM2Mat-Bench, marking a key step toward bridging microscopy and atomistic simulation in materials science.The code and dataset are publicly available at https://github.com/yyt-2378/AutoMat and https://huggingface.co/datasets/yaotianvector/STEM2Mat.

Paperid: 1071, https://arxiv.org/pdf/2505.12493.pdf GitHub

Abstract:
Training effective Vision Language Models (VLMs) for GUI agents typically relies on supervised fine-tuning (SFT) over large-scale annotated datasets, where the collection process is labor-intensive and error-prone. In this work, we propose a self-supervised inverse dynamics task to enable VLMs to learn from GUI transition pairs by inferring the action that caused that transition. This training task offers two advantages: (1) It enables VLMs to ignore variations unrelated to user actions (e.g., background refreshes, ads) and to focus on true affordances such as buttons and input fields within complex GUIs. (2) The training data can be easily obtained from existing GUI trajectories without requiring human annotation, and it can be easily scaled through automatic offline exploration. Using this training task, we propose UI-shift, a framework for enhancing VLM-based GUI agents through self-supervised reinforcement learning (RL). With only 2K training samples sourced from existing datasets, two VLMs -- Qwen2.5-VL-3B and Qwen2.5-VL-7B -- trained with UI-Shift achieve competitive or superior performance on grounding tasks (ScreenSpot-series benchmarks) and GUI automation tasks (AndroidControl), compared to SFT baselines and GUI-specific models that explicitly elicit reasoning abilities during RL. Our findings suggest a potential direction for enhancing VLMs for GUI agents by leveraging more self-supervised training data in the future. Code, model, and data are available at: https://github.com/UbiquitousLearning/UIShift

Paperid: 1072, https://arxiv.org/pdf/2505.12448.pdf GitHub

Abstract:
Despite impressive advancements in Visual-Language Models (VLMs) for multi-modal tasks, their reliance on RGB inputs limits precise spatial understanding. Existing methods for integrating spatial cues, such as point clouds or depth, either require specialized sensors or fail to effectively exploit depth information for higher-order reasoning. To this end, we propose a novel Spatial Sense and Reasoning method, dubbed SSR, a novel framework that transforms raw depth data into structured, interpretable textual rationales. These textual rationales serve as meaningful intermediate representations to significantly enhance spatial reasoning capabilities. Additionally, we leverage knowledge distillation to compress the generated rationales into compact latent embeddings, which facilitate resource-efficient and plug-and-play integration into existing VLMs without retraining. To enable comprehensive evaluation, we introduce a new dataset named SSR-CoT, a million-scale visual-language reasoning dataset enriched with intermediate spatial reasoning annotations, and present SSRBench, a comprehensive multi-task benchmark. Extensive experiments on multiple benchmarks demonstrate SSR substantially improves depth utilization and enhances spatial reasoning, thereby advancing VLMs toward more human-like multi-modal understanding. Our project page is at https://yliu-cs.github.io/SSR.

Paperid: 1073, https://arxiv.org/pdf/2505.12044.pdf GitHub

Abstract:
Attention with bias, which extends standard attention by introducing prior knowledge as an additive bias matrix to the query-key scores, has been widely deployed in vision, language, protein-folding and other advanced scientific models, underscoring its status as a key evolution of this foundational module. However, introducing bias terms creates a severe efficiency bottleneck in attention computation. It disrupts the tightly fused memory-compute pipeline that underlies the speed of accelerators like FlashAttention, thereby stripping away most of their performance gains and leaving biased attention computationally expensive. Surprisingly, despite its common usage, targeted efficiency optimization for attention with bias remains absent, which seriously hinders its application in complex tasks. Diving into the computation of FlashAttention, we prove that its optimal efficiency is determined by the rank of the attention weight matrix. Inspired by this theoretical result, this paper presents FlashBias based on the low-rank compressed sensing theory, which can provide fast-exact computation for many widely used attention biases and a fast-accurate approximation for biases in general formalizations. FlashBias can fully take advantage of the extremely optimized matrix multiplication operation in modern GPUs, achieving 1.5$\times$ speedup for Pairformer in AlphaFold 3, and over 2$\times$ speedup for attention with bias in vision and language models without loss of accuracy. Code is available at this repository: https://github.com/thuml/FlashBias.

Paperid: 1074, https://arxiv.org/pdf/2505.11842.pdf GitHub

Abstract:
The increasing deployment of Large Vision-Language Models (LVLMs) raises safety concerns under potential malicious inputs. However, existing multimodal safety evaluations primarily focus on model vulnerabilities exposed by static image inputs, ignoring the temporal dynamics of video that may induce distinct safety risks. To bridge this gap, we introduce Video-SafetyBench, the first comprehensive benchmark designed to evaluate the safety of LVLMs under video-text attacks. It comprises 2,264 video-text pairs spanning 48 fine-grained unsafe categories, each pairing a synthesized video with either a harmful query, which contains explicit malice, or a benign query, which appears harmless but triggers harmful behavior when interpreted alongside the video. To generate semantically accurate videos for safety evaluation, we design a controllable pipeline that decomposes video semantics into subject images (what is shown) and motion text (how it moves), which jointly guide the synthesis of query-relevant videos. To effectively evaluate uncertain or borderline harmful outputs, we propose RJScore, a novel LLM-based metric that incorporates the confidence of judge models and human-aligned decision threshold calibration. Extensive experiments show that benign-query video composition achieves average attack success rates of 67.2%, revealing consistent vulnerabilities to video-induced attacks. We believe Video-SafetyBench will catalyze future research into video-based safety evaluation and defense strategies.

Paperid: 1075, https://arxiv.org/pdf/2505.11815.pdf GitHub

Abstract:
Current research has explored vision-language models for multi-modal embedding tasks, such as information retrieval, visual grounding, and classification. However, real-world scenarios often involve diverse modality combinations between queries and targets, such as text and image to text, text and image to text and image, and text to text and image. These diverse combinations pose significant challenges for existing models, as they struggle to align all modality combinations within a unified embedding space during training, which degrades performance at inference. To address this limitation, we propose UniMoCo, a novel vision-language model architecture designed for multi-modal embedding tasks. UniMoCo introduces a modality-completion module that generates visual features from textual inputs, ensuring modality completeness for both queries and targets. Additionally, we develop a specialized training strategy to align embeddings from both original and modality-completed inputs, ensuring consistency within the embedding space. This enables the model to robustly handle a wide range of modality combinations across embedding tasks. Experiments show that UniMoCo outperforms previous methods while demonstrating consistent robustness across diverse settings. More importantly, we identify and quantify the inherent bias in conventional approaches caused by imbalance of modality combinations in training data, which can be mitigated through our modality-completion paradigm. The code is available at https://github.com/HobbitQia/UniMoCo.

Paperid: 1076, https://arxiv.org/pdf/2505.11404.pdf GitHub

Abstract:
Recent advances in vision language models (VLMs) have enabled broad progress in the general medical field. However, pathology still remains a more challenging subdomain, with current pathology specific VLMs exhibiting limitations in both diagnostic accuracy and reasoning plausibility. Such shortcomings are largely attributable to the nature of current pathology datasets, which are primarily composed of image description pairs that lack the depth and structured diagnostic paradigms employed by real world pathologists. In this study, we leverage pathology textbooks and real world pathology experts to construct high-quality, reasoning-oriented datasets. Building on this, we introduce Patho-R1, a multimodal RL-based pathology Reasoner, trained through a three-stage pipeline: (1) continued pretraining on 3.5 million image-text pairs for knowledge infusion; (2) supervised fine-tuning on 500k high-quality Chain-of-Thought samples for reasoning incentivizing; (3) reinforcement learning using Group Relative Policy Optimization and Decoupled Clip and Dynamic sAmpling Policy Optimization strategies for multimodal reasoning quality refinement. To further assess the alignment quality of our dataset, we propose Patho-CLIP, trained on the same figure-caption corpus used for continued pretraining. Comprehensive experimental results demonstrate that both Patho-CLIP and Patho-R1 achieve robust performance across a wide range of pathology-related tasks, including zero-shot classification, cross-modal retrieval, Visual Question Answering, and Multiple Choice Question. Our project is available at the Patho-R1 repository: https://github.com/Wenchuan-Zhang/Patho-R1.

Paperid: 1077, https://arxiv.org/pdf/2505.11326.pdf GitHub GitHub

Abstract:
Vision-language models (VLMs) have shown remarkable progress in offline tasks such as image captioning and video question answering. However, real-time interactive environments impose new demands on VLMs, requiring them to generate utterances that are not only semantically accurate but also precisely timed. We identify two core capabilities necessary for such settings -- $\textit{perceptual updating}$ and $\textit{contingency awareness}$ -- and propose a new benchmark task, $\textbf{Temporally-Grounded Language Generation (TGLG)}$, to evaluate them. TGLG requires models to generate utterances in response to streaming video such that both content and timing align with dynamic visual input. To support this benchmark, we curate evaluation datasets from sports broadcasting and egocentric human interaction domains, and introduce a new metric, $\textbf{TRACE}$, to evaluate TGLG by jointly measuring semantic similarity and temporal alignment. Finally, we present $\textbf{Vision-Language Model with Time-Synchronized Interleaving (VLM-TSI)}$, a model that interleaves visual and linguistic tokens in a time-synchronized manner, enabling real-time language generation without relying on turn-based assumptions. Experimental results show that VLM-TSI significantly outperforms a strong baseline, yet overall performance remains modest -- highlighting the difficulty of TGLG and motivating further research in real-time VLMs. Code and data available $\href{https://github.com/yukw777/tglg}{here}$.

Paperid: 1078, https://arxiv.org/pdf/2505.10769.pdf GitHub

Abstract:
Accurate segmentation of regions of interest in biomedical images holds substantial value in image analysis. Although several foundation models for biomedical segmentation have currently achieved excellent performance on certain datasets, they typically demonstrate sub-optimal performance on unseen domain data. We owe the deficiency to lack of vision-language knowledge before segmentation. Multimodal Large Language Models (MLLMs) bring outstanding understanding and reasoning capabilities to multimodal tasks, which inspires us to leverage MLLMs to inject Vision-Language Knowledge (VLK), thereby enabling vision models to demonstrate superior generalization capabilities on cross-domain datasets. In this paper, we propose using MLLMs to guide SAM in learning microscopy crose-domain data, unifying Segment Anything in Microscopy, named uLLSAM. Specifically, we propose the Vision-Language Semantic Alignment (VLSA) module, which injects VLK into Segment Anything Model (SAM). We find that after SAM receives global VLK prompts, its performance improves significantly, but there are deficiencies in boundary contour perception. Therefore, we further propose Semantic Boundary Regularization (SBR) to prompt SAM. Our method achieves performance improvements of 7.71% in Dice and 12.10% in SA across 9 in-domain microscopy datasets, achieving state-of-the-art performance. Our method also demonstrates improvements of 6.79% in Dice and 10.08% in SA across 10 out-ofdomain datasets, exhibiting strong generalization capabilities. Code is available at https://github.com/ieellee/uLLSAM.

Paperid: 1079, https://arxiv.org/pdf/2505.09926.pdf GitHub

Abstract:
Universal visual anomaly detection aims to identify anomalies from novel or unseen vision domains without additional fine-tuning, which is critical in open scenarios. Recent studies have demonstrated that pre-trained vision-language models like CLIP exhibit strong generalization with just zero or a few normal images. However, existing methods struggle with designing prompt templates, complex token interactions, or requiring additional fine-tuning, resulting in limited flexibility. In this work, we present a simple yet effective method called AdaptCLIP based on two key insights. First, adaptive visual and textual representations should be learned alternately rather than jointly. Second, comparative learning between query and normal image prompt should incorporate both contextual and aligned residual features, rather than relying solely on residual features. AdaptCLIP treats CLIP models as a foundational service, adding only three simple adapters, visual adapter, textual adapter, and prompt-query adapter, at its input or output ends. AdaptCLIP supports zero-/few-shot generalization across domains and possesses a training-free manner on target domains once trained on a base dataset. AdaptCLIP achieves state-of-the-art performance on 12 anomaly detection benchmarks from industrial and medical domains, significantly outperforming existing competitive methods. We will make the code and model of AdaptCLIP available at https://github.com/gaobb/AdaptCLIP.

Paperid: 1080, https://arxiv.org/pdf/2505.09731.pdf GitHub

Abstract:
Vision language models (VLMs) exhibit vast knowledge of the physical world, including intuition of physical and spatial properties, affordances, and motion. With fine-tuning, VLMs can also natively produce robot trajectories. We demonstrate that eliciting wrenches, not trajectories, allows VLMs to explicitly reason about forces and leads to zero-shot generalization in a series of manipulation tasks without pretraining. We achieve this by overlaying a consistent visual representation of relevant coordinate frames on robot-attached camera images to augment our query. First, we show how this addition enables a versatile motion control framework evaluated across four tasks (opening and closing a lid, pushing a cup or chair) spanning prismatic and rotational motion, an order of force and position magnitude, different camera perspectives, annotation schemes, and two robot platforms over 220 experiments, resulting in 51% success across the four tasks. Then, we demonstrate that the proposed framework enables VLMs to continually reason about interaction feedback to recover from task failure or incompletion, with and without human supervision. Finally, we observe that prompting schemes with visual annotation and embodied reasoning can bypass VLM safeguards. We characterize prompt component contribution to harmful behavior elicitation and discuss its implications for developing embodied reasoning. Our code, videos, and data are available at: https://scalingforce.github.io/.

Paperid: 1081, https://arxiv.org/pdf/2505.09659.pdf GitHub

Abstract:
Spiking Large Language Models (LLMs) have emerged as an energy-efficient alternative to conventional LLMs through their event-driven computation. To effectively obtain spiking LLMs, researchers develop different ANN-to-SNN conversion methods by leveraging pre-trained ANN parameters while inheriting the energy efficiency of SNN. However, existing conversion methods struggle with extreme activation outliers and incompatible nonlinear operations of ANN-based LLMs. To address this, we propose a loss-less ANN-SNN conversion for fully spike-driven LLMs, termed LAS. Specifically, LAS introduces two novel neurons to convert the activation outlier and nonlinear operation of ANN-based LLMs. Moreover, LAS tailors the spike-equivalent Transformer components for spiking LLMs, which can ensure full spiking conversion without any loss of performance. Experimental results on six language models and two vision-language models demonstrate that LAS achieves loss-less conversion. Notably, on OPT-66B, LAS even improves the accuracy of 2\% on the WSC task. In addition, the parameter and ablation studies further verify the effectiveness of LAS. The source code is available at https://github.com/lc783/LAS

Paperid: 1082, https://arxiv.org/pdf/2505.09265.pdf GitHub

Abstract:
Zero- and few-shot visual anomaly segmentation relies on powerful vision-language models that detect unseen anomalies using manually designed textual prompts. However, visual representations are inherently independent of language. In this paper, we explore the potential of a pure visual foundation model as an alternative to widely used vision-language models for universal visual anomaly segmentation. We present a novel paradigm that unifies anomaly segmentation into change segmentation. This paradigm enables us to leverage large-scale synthetic image pairs, featuring object-level and local region changes, derived from existing image datasets, which are independent of target anomaly datasets. We propose a one-prompt Meta-learning framework for Universal Anomaly Segmentation (MetaUAS) that is trained on this synthetic dataset and then generalizes well to segment any novel or unseen visual anomalies in the real world. To handle geometrical variations between prompt and query images, we propose a soft feature alignment module that bridges paired-image change perception and single-image semantic segmentation. This is the first work to achieve universal anomaly segmentation using a pure vision model without relying on special anomaly detection datasets and pre-trained visual-language models. Our method effectively and efficiently segments any anomalies with only one normal image prompt and enjoys training-free without guidance from language. Our MetaUAS significantly outperforms previous zero-shot, few-shot, and even full-shot anomaly segmentation methods. The code and pre-trained models are available at https://github.com/gaobb/MetaUAS.

Paperid: 1083, https://arxiv.org/pdf/2505.09118.pdf GitHub

Abstract:
Traditional scene graphs primarily focus on spatial relationships, limiting vision-language models' (VLMs) ability to reason about complex interactions in visual scenes. This paper addresses two key challenges: (1) conventional detection-to-construction methods produce unfocused, contextually irrelevant relationship sets, and (2) existing approaches fail to form persistent memories for generalizing interaction reasoning to new scenes. We propose Interaction-augmented Scene Graph Reasoning (ISGR), a framework that enhances VLMs' interactional reasoning through three complementary components. First, our dual-stream graph constructor combines SAM-powered spatial relation extraction with interaction-aware captioning to generate functionally salient scene graphs with spatial grounding. Second, we employ targeted interaction queries to activate VLMs' latent knowledge of object functionalities, converting passive recognition into active reasoning about how objects work together. Finally, we introduce a lone-term memory reinforcement learning strategy with a specialized interaction-focused reward function that transforms transient patterns into long-term reasoning heuristics. Extensive experiments demonstrate that our approach significantly outperforms baseline methods on interaction-heavy reasoning benchmarks, with particularly strong improvements on complex scene understanding tasks. The source code can be accessed at https://github.com/open_upon_acceptance.

Paperid: 1084, https://arxiv.org/pdf/2505.09092.pdf GitHub

Abstract:
Lane Keeping Assist (LKA) is widely adopted in modern vehicles, yet its real-world performance remains underexplored due to proprietary systems and limited data access. This paper presents OpenLKA, the first open, large-scale dataset for LKA evaluation and improvement. It includes 400 hours of driving data from 62 production vehicle models, collected through extensive road testing in Tampa, Florida and global contributions from the Comma.ai driving community. The dataset spans a wide range of challenging scenarios, including complex road geometries, degraded lane markings, adverse weather, lighting conditions and surrounding traffic. The dataset is multimodal, comprising: i) full CAN bus streams, decoded using custom reverse-engineered DBC files to extract key LKA events (e.g., system disengagements, lane detection failures); ii) synchronized high-resolution dash-cam video; iii) real-time outputs from Openpilot, providing accurate estimates of road curvature and lane positioning; iv) enhanced scene annotations generated by Vision Language Models, describing lane visibility, pavement quality, weather, lighting, and traffic conditions. By integrating vehicle-internal signals with high-fidelity perception and rich semantic context, OpenLKA provides a comprehensive platform for benchmarking the real-world performance of production LKA systems, identifying safety-critical operational scenarios, and assessing the readiness of current road infrastructure for autonomous driving. The dataset is publicly available at: https://github.com/OpenLKA/OpenLKA.

Paperid: 1085, https://arxiv.org/pdf/2505.08971.pdf GitHub

Abstract:
In standard large vision-language models (LVLMs) pre-training, the model typically maximizes the joint probability of the caption conditioned on the image via next-token prediction (NTP); however, since only a small subset of caption tokens directly relates to the visual content, this naive NTP unintentionally fits the model to noise and increases the risk of hallucination. We present PRIOR, a simple vision-language pre-training approach that addresses this issue by prioritizing image-related tokens through differential weighting in the NTP loss, drawing from the importance sampling framework. PRIOR introduces a reference model-a text-only large language model (LLM) trained on the captions without image inputs, to weight each token based on its probability for LVLMs training. Intuitively, tokens that are directly related to the visual inputs are harder to predict without the image and thus receive lower probabilities from the text-only reference LLM. During training, we implement a token-specific re-weighting term based on the importance scores to adjust each token's loss. We implement PRIOR in two distinct settings: LVLMs with visual encoders and LVLMs without visual encoders. We observe 19% and 8% average relative improvement, respectively, on several vision-language benchmarks compared to NTP. In addition, PRIOR exhibits superior scaling properties, as demonstrated by significantly higher scaling coefficients, indicating greater potential for performance gains compared to NTP given increasing compute and data.

Paperid: 1086, https://arxiv.org/pdf/2505.08910.pdf GitHub

Abstract:
In recent times, we have seen a rapid development of large Vision-Language Models (VLMs). They have shown impressive results on academic benchmarks, primarily in widely spoken languages but lack performance on low-resource languages and varied cultural contexts. To address these limitations, we introduce Maya, an open-source Multilingual VLM. Our contributions are: 1) a multilingual image-text pretraining dataset in eight languages, based on the LLaVA pretraining dataset; and 2) a multilingual image-text model supporting these languages, enhancing cultural and linguistic comprehension in vision-language tasks. Code available at https://github.com/nahidalam/maya.

Paperid: 1087, https://arxiv.org/pdf/2505.08725.pdf GitHub GitHub

Abstract:
The Large Visual-Language Models (LVLMs) have significantly advanced image understanding. Their comprehension and reasoning capabilities enable promising applications in autonomous driving scenarios. However, existing research typically focuses on front-view perspectives and partial objects within scenes, struggling to achieve comprehensive scene understanding. Meanwhile, existing LVLMs suffer from the lack of mapping relationship between 2D and 3D and insufficient integration of 3D object localization and instruction understanding. To tackle these limitations, we first introduce NuInteract, a large-scale dataset with over 1.5M multi-view image language pairs spanning dense scene captions and diverse interactive tasks. Furthermore, we propose DriveMonkey, a simple yet effective framework that seamlessly integrates LVLMs with a spatial processor using a series of learnable queries. The spatial processor, designed as a plug-and-play component, can be initialized with pre-trained 3D detectors to improve 3D perception. Our experiments show that DriveMonkey outperforms general LVLMs, especially achieving a 9.86% notable improvement on the 3D visual grounding task. The dataset and code will be released at https://github.com/zc-zhao/DriveMonkey.

Paperid: 1088, https://arxiv.org/pdf/2505.07817.pdf GitHub

Abstract:
We present LangToMo, a vision-language-action framework structured as a dual-system architecture that uses pixel motion forecasts as intermediate representations. Our high-level System 2, an image diffusion model, generates text-conditioned pixel motion sequences from a single frame to guide robot control. Pixel motion-a universal, interpretable, and motion-centric representation-can be extracted from videos in a weakly-supervised manner, enabling diffusion model training on any video-caption data. Treating generated pixel motion as learned universal representations, our low level System 1 module translates these into robot actions via motion-to-action mapping functions, which can be either hand-crafted or learned with minimal supervision. System 2 operates as a high-level policy applied at sparse temporal intervals, while System 1 acts as a low-level policy at dense temporal intervals. This hierarchical decoupling enables flexible, scalable, and generalizable robot control under both unsupervised and supervised settings, bridging the gap between language, motion, and action. Checkout https://kahnchana.github.io/LangToMo

Paperid: 1089, https://arxiv.org/pdf/2505.07538.pdf GitHub

Abstract:
We completely discard the conventional spatial prior in image representation and introduce a novel discrete visual tokenizer: Self-consistency Tokenizer (Selftok). At its design core, we compose an autoregressive (AR) prior -- mirroring the causal structure of language -- into visual tokens by using the reverse diffusion process of image generation. The AR property makes Selftok fundamentally distinct from traditional spatial tokens in the following two key ways: - Selftok offers an elegant and minimalist approach to unify diffusion and AR for vision-language models (VLMs): By representing images with Selftok tokens, we can train a VLM using a purely discrete autoregressive architecture -- like that in LLMs -- without requiring additional modules or training objectives. - We theoretically show that the AR prior satisfies the Bellman equation, whereas the spatial prior does not. Therefore, Selftok supports reinforcement learning (RL) for visual generation with effectiveness comparable to that achieved in LLMs. Besides the AR property, Selftok is also a SoTA tokenizer that achieves a favorable trade-off between high-quality reconstruction and compression rate. We use Selftok to build a pure AR VLM for both visual comprehension and generation tasks. Impressively, without using any text-image training pairs, a simple policy gradient RL working in the visual tokens can significantly boost the visual generation benchmark, surpassing all the existing models by a large margin. Therefore, we believe that Selftok effectively addresses the long-standing challenge that visual tokens cannot support effective RL. When combined with the well-established strengths of RL in LLMs, this brings us one step closer to realizing a truly multimodal LLM. Project Page: https://selftok-team.github.io/report/.

Paperid: 1090, https://arxiv.org/pdf/2505.07219.pdf GitHub GitHub

Abstract:
Generalizing an object detector trained on a single domain to multiple unseen domains is a challenging task. Existing methods typically introduce image or feature augmentation to diversify the source domain to raise the robustness of the detector. Vision-Language Model (VLM)-based augmentation techniques have been proven to be effective, but they require that the detector's backbone has the same structure as the image encoder of VLM, limiting the detector framework selection. To address this problem, we propose Language-Driven Dual Style Mixing (LDDS) for single-domain generalization, which diversifies the source domain by fully utilizing the semantic information of the VLM. Specifically, we first construct prompts to transfer style semantics embedded in the VLM to an image translation network. This facilitates the generation of style diversified images with explicit semantic information. Then, we propose image-level style mixing between the diversified images and source domain images. This effectively mines the semantic information for image augmentation without relying on specific augmentation selections. Finally, we propose feature-level style mixing in a double-pipeline manner, allowing feature augmentation to be model-agnostic and can work seamlessly with the mainstream detector frameworks, including the one-stage, two-stage, and transformer-based detectors. Extensive experiments demonstrate the effectiveness of our approach across various benchmark datasets, including real to cartoon and normal to adverse weather tasks. The source code and pre-trained models will be publicly available at https://github.com/qinhongda8/LDDS.

Paperid: 1091, https://arxiv.org/pdf/2505.07164.pdf GitHub

Abstract:
Visual emotion analysis, which has gained considerable attention in the field of affective computing, aims to predict the dominant emotions conveyed by an image. Despite advancements in visual emotion analysis with the emergence of vision-language models, we observed that instruction-tuned vision-language models and conventional vision models exhibit complementary strengths in visual emotion analysis, as vision-language models excel in certain cases, whereas vision models perform better in others. This finding highlights the need to integrate these capabilities to enhance the performance of visual emotion analysis. To bridge this gap, we propose EmoVLM-KD, an instruction-tuned vision-language model augmented with a lightweight module distilled from conventional vision models. Instead of deploying both models simultaneously, which incurs high computational costs, we transfer the predictive patterns of a conventional vision model into the vision-language model using a knowledge distillation framework. Our approach first fine-tunes a vision-language model on emotion-specific instruction data and then attaches a distilled module to its visual encoder while keeping the vision-language model frozen. Predictions from the vision language model and the distillation module are effectively balanced by a gate module, which subsequently generates the final outcome. Extensive experiments show that EmoVLM-KD achieves state-of-the-art performance on multiple visual emotion analysis benchmark datasets, outperforming the existing methods while maintaining computational efficiency. The code is available in https://github.com/sange1104/EmoVLM-KD.

Paperid: 1092, https://arxiv.org/pdf/2505.07007.pdf GitHub

Abstract:
Micro-expressions (MEs) are crucial psychological responses with significant potential for affective computing. However, current automatic micro-expression recognition (MER) research primarily focuses on discrete emotion classification, neglecting a convincing analysis of the subtle dynamic movements and inherent emotional cues. The rapid progress in multimodal large language models (MLLMs), known for their strong multimodal comprehension and language generation abilities, offers new possibilities. MLLMs have shown success in various vision-language tasks, indicating their potential to understand MEs comprehensively, including both fine-grained motion patterns and underlying emotional semantics. Nevertheless, challenges remain due to the subtle intensity and short duration of MEs, as existing MLLMs are not designed to capture such delicate frame-level facial dynamics. In this paper, we propose a novel Micro-Expression Large Language Model (MELLM), which incorporates a subtle facial motion perception strategy with the strong inference capabilities of MLLMs, representing the first exploration of MLLMs in the domain of ME analysis. Specifically, to explicitly guide the MLLM toward motion-sensitive regions, we construct an interpretable motion-enhanced color map by fusing onset-apex optical flow dynamics with the corresponding grayscale onset frame as the model input. Additionally, specialized fine-tuning strategies are incorporated to further enhance the model's visual perception of MEs. Furthermore, we construct an instruction-description dataset based on Facial Action Coding System (FACS) annotations and emotion labels to train our MELLM. Comprehensive evaluations across multiple benchmark datasets demonstrate that our model exhibits superior robustness and generalization capabilities in ME understanding (MEU). Code is available at https://github.com/zyzhangUstc/MELLM.

Paperid: 1093, https://arxiv.org/pdf/2505.07001.pdf GitHub

Abstract:
Vision-Language Models (VLMs) are becoming increasingly popular in the medical domain, bridging the gap between medical images and clinical language. Existing VLMs demonstrate an impressive ability to comprehend medical images and text queries to generate detailed, descriptive diagnostic medical reports. However, hallucination--the tendency to generate descriptions that are inconsistent with the visual content--remains a significant issue in VLMs, with particularly severe implications in the medical field. To facilitate VLM research on gastrointestinal (GI) image analysis and study hallucination, we curate a multimodal image-text GI dataset: Gut-VLM. This dataset is created using a two-stage pipeline: first, descriptive medical reports of Kvasir-v2 images are generated using ChatGPT, which introduces some hallucinated or incorrect texts. In the second stage, medical experts systematically review these reports, and identify and correct potential inaccuracies to ensure high-quality, clinically reliable annotations. Unlike traditional datasets that contain only descriptive texts, our dataset also features tags identifying hallucinated sentences and their corresponding corrections. A common approach to reducing hallucination in VLM is to finetune the model on a small-scale, problem-specific dataset. However, we take a different strategy using our dataset. Instead of finetuning the VLM solely for generating textual reports, we finetune it to detect and correct hallucinations, an approach we call hallucination-aware finetuning. Our results show that this approach is better than simply finetuning for descriptive report generation. Additionally, we conduct an extensive evaluation of state-of-the-art VLMs across several metrics, establishing a benchmark. GitHub Repo: https://github.com/bhattarailab/Hallucination-Aware-VLM.

Paperid: 1094, https://arxiv.org/pdf/2505.06729.pdf GitHub

Abstract:
Vision-Language Models (VLMs) have been increasingly integrated into object navigation tasks for their rich prior knowledge and strong reasoning abilities. However, applying VLMs to navigation poses two key challenges: effectively representing complex environment information and determining \textit{when and how} to query VLMs. Insufficient environment understanding and over-reliance on VLMs (e.g. querying at every step) can lead to unnecessary backtracking and reduced navigation efficiency, especially in continuous environments. To address these challenges, we propose a novel framework that constructs a multi-layer representation of the environment during navigation. This representation consists of viewpoint, object nodes, and room nodes. Viewpoints and object nodes facilitate intra-room exploration and accurate target localization, while room nodes support efficient inter-room planning. Building on this representation, we propose a novel two-stage navigation policy, integrating high-level planning guided by VLM reasoning with low-level VLM-assisted exploration to efficiently locate a goal object. We evaluated our approach on three simulated benchmarks (HM3D, RoboTHOR, and MP3D), and achieved state-of-the-art performance on both the success rate ($\mathord{\uparrow}\, 7.1\%$) and navigation efficiency ($\mathord{\uparrow}\, 12.5\%$). We further validate our method on a real robot platform, demonstrating strong robustness across 15 object navigation tasks in 10 different indoor environments. Project page is available at https://zwandering.github.io/STRIVE.github.io/ .

Paperid: 1095, https://arxiv.org/pdf/2505.06152.pdf GitHub

Abstract:
Medical vision-language models (VLMs) have shown promise as clinical assistants across various medical fields. However, specialized dermatology VLM capable of delivering professional and detailed diagnostic analysis remains underdeveloped, primarily due to less specialized text descriptions in current dermatology multimodal datasets. To address this issue, we propose MM-Skin, the first large-scale multimodal dermatology dataset that encompasses 3 imaging modalities, including clinical, dermoscopic, and pathological and nearly 10k high-quality image-text pairs collected from professional textbooks. In addition, we generate over 27k diverse, instruction-following vision question answering (VQA) samples (9 times the size of current largest dermatology VQA dataset). Leveraging public datasets and MM-Skin, we developed SkinVL, a dermatology-specific VLM designed for precise and nuanced skin disease interpretation. Comprehensive benchmark evaluations of SkinVL on VQA, supervised fine-tuning (SFT) and zero-shot classification tasks across 8 datasets, reveal its exceptional performance for skin diseases in comparison to both general and medical VLM models. The introduction of MM-Skin and SkinVL offers a meaningful contribution to advancing the development of clinical dermatology VLM assistants. MM-Skin is available at https://github.com/ZwQ803/MM-Skin

Paperid: 1096, https://arxiv.org/pdf/2505.06111.pdf GitHub

Abstract:
A generalist robot should perform effectively across various environments. However, most existing approaches heavily rely on scaling action-annotated data to enhance their capabilities. Consequently, they are often limited to single physical specification and struggle to learn transferable knowledge across different embodiments and environments. To confront these limitations, we propose UniVLA, a new framework for learning cross-embodiment vision-language-action (VLA) policies. Our key innovation is to derive task-centric action representations from videos with a latent action model. This enables us to exploit extensive data across a wide spectrum of embodiments and perspectives. To mitigate the effect of task-irrelevant dynamics, we incorporate language instructions and establish a latent action model within the DINO feature space. Learned from internet-scale videos, the generalist policy can be deployed to various robots through efficient latent action decoding. We obtain state-of-the-art results across multiple manipulation and navigation benchmarks, as well as real-robot deployments. UniVLA achieves superior performance over OpenVLA with less than 1/20 of pretraining compute and 1/10 of downstream data. Continuous performance improvements are observed as heterogeneous data, even including human videos, are incorporated into the training pipeline. The results underscore UniVLA's potential to facilitate scalable and efficient robot policy learning.

Paperid: 1097, https://arxiv.org/pdf/2505.05528.pdf GitHub GitHub

Abstract:
As Contrastive Language-Image Pre-training (CLIP) models are increasingly adopted for diverse downstream tasks and integrated into large vision-language models (VLMs), their susceptibility to adversarial perturbations has emerged as a critical concern. In this work, we introduce \textbf{X-Transfer}, a novel attack method that exposes a universal adversarial vulnerability in CLIP. X-Transfer generates a Universal Adversarial Perturbation (UAP) capable of deceiving various CLIP encoders and downstream VLMs across different samples, tasks, and domains. We refer to this property as \textbf{super transferability}--a single perturbation achieving cross-data, cross-domain, cross-model, and cross-task adversarial transferability simultaneously. This is achieved through \textbf{surrogate scaling}, a key innovation of our approach. Unlike existing methods that rely on fixed surrogate models, which are computationally intensive to scale, X-Transfer employs an efficient surrogate scaling strategy that dynamically selects a small subset of suitable surrogates from a large search space. Extensive evaluations demonstrate that X-Transfer significantly outperforms previous state-of-the-art UAP methods, establishing a new benchmark for adversarial transferability across CLIP models. The code is publicly available in our \href{https://github.com/HanxunH/XTransferBench}{GitHub repository}.

Paperid: 1098, https://arxiv.org/pdf/2505.05464.pdf GitHub

Abstract:
Vision-Language Models (VLMs) combine visual perception with the general capabilities, such as reasoning, of Large Language Models (LLMs). However, the mechanisms by which these two abilities can be combined and contribute remain poorly understood. In this work, we explore to compose perception and reasoning through model merging that connects parameters of different models. Unlike previous works that often focus on merging models of the same kind, we propose merging models across modalities, enabling the incorporation of the reasoning capabilities of LLMs into VLMs. Through extensive experiments, we demonstrate that model merging offers a successful pathway to transfer reasoning abilities from LLMs to VLMs in a training-free manner. Moreover, we utilize the merged models to understand the internal mechanism of perception and reasoning and how merging affects it. We find that perception capabilities are predominantly encoded in the early layers of the model, whereas reasoning is largely facilitated by the middle-to-late layers. After merging, we observe that all layers begin to contribute to reasoning, whereas the distribution of perception abilities across layers remains largely unchanged. These observations shed light on the potential of model merging as a tool for multimodal integration and interpretation.

Paperid: 1099, https://arxiv.org/pdf/2505.05343.pdf GitHub

Abstract:
Large-scale vision-language models demonstrate strong multimodal alignment and generalization across diverse tasks. Among them, CLIP stands out as one of the most successful approaches. In this work, we extend the application of CLIP to sound source localization, proposing a self-supervised method operates without explicit text input. We introduce a framework that maps audios into tokens compatible with CLIP's text encoder, producing audio-driven embeddings. These embeddings are used to generate sounding region masks, from which visual features are extracted and aligned with the audio embeddings through a contrastive audio-visual correspondence objective. Our findings show that alignment knowledge of pre-trained multimodal foundation model enables our method to generate more complete and compact localization for sounding objects. We further propose an LLM-guided extension that distills object-aware audio-visual scene understanding into the model during training to enhance alignment. Extensive experiments across five diverse tasks demonstrate that our method, in all variants, outperforms state-of-the-art approaches and achieves strong generalization in zero-shot settings.

Paperid: 1100, https://arxiv.org/pdf/2505.05189.pdf GitHub

Abstract:
Prompt learning is one of the most effective paradigms for adapting pre-trained vision-language models (VLMs) to the biomedical image classification tasks in few shot scenarios. However, most of the current prompt learning methods only used the text prompts and ignored the particular structures (such as the complex anatomical structures and subtle pathological features) in the biomedical images. In this work, we propose Biomed-DPT, a knowledge-enhanced dual modality prompt tuning technique. In designing the text prompt, Biomed-DPT constructs a dual prompt including the template-driven clinical prompts and the large language model (LLM)-driven domain-adapted prompts, then extracts the clinical knowledge from the domain-adapted prompts through the knowledge distillation technique. In designing the vision prompt, Biomed-DPT introduces the zero vector as a soft prompt to leverage attention re-weighting so that the focus on non-diagnostic regions and the recognition of non-critical pathological features are avoided. Biomed-DPT achieves an average classification accuracy of 66.14\% across 11 biomedical image datasets covering 9 modalities and 10 organs, with performance reaching 78.06\% in base classes and 75.97\% in novel classes, surpassing the Context Optimization (CoOp) method by 6.20\%, 3.78\%, and 8.04\%, respectively. Our code are available at \underline{https://github.com/Kanyooo/Biomed-DPT}.

Paperid: 1101, https://arxiv.org/pdf/2505.05180.pdf GitHub

Abstract:
Prompt tuning adapts Vision-Language Models like CLIP to open-world tasks with minimal training costs. In this direction, one typical paradigm evaluates model performance separately on known classes (i.e., base domain) and unseen classes (i.e., new domain). However, real-world scenarios require models to handle inputs without prior domain knowledge. This practical challenge has spurred the development of open-world prompt tuning, which demands a unified evaluation of two stages: 1) detecting whether an input belongs to the base or new domain (P1), and 2) classifying the sample into its correct class (P2). What's more, as domain distributions are generally unknown, a proper metric should be insensitive to varying base/new sample ratios (P3). However, we find that current metrics, including HM, overall accuracy, and AUROC, fail to satisfy these three properties simultaneously. To bridge this gap, we propose OpenworldAUC, a unified metric that jointly assesses detection and classification through pairwise instance comparisons. To optimize OpenworldAUC effectively, we introduce Gated Mixture-of-Prompts (GMoP), which employs domain-specific prompts and a gating mechanism to dynamically balance detection and classification. Theoretical guarantees ensure generalization of GMoP under practical conditions. Experiments on 15 benchmarks in open-world scenarios show GMoP achieves SOTA performance on OpenworldAUC and other metrics. We release the code at https://github.com/huacong/OpenworldAUC

Paperid: 1102, https://arxiv.org/pdf/2505.04410.pdf GitHub

Abstract:
Dense visual prediction tasks have been constrained by their reliance on predefined categories, limiting their applicability in real-world scenarios where visual concepts are unbounded. While Vision-Language Models (VLMs) like CLIP have shown promise in open-vocabulary tasks, their direct application to dense prediction often leads to suboptimal performance due to limitations in local feature representation. In this work, we present our observation that CLIP's image tokens struggle to effectively aggregate information from spatially or semantically related regions, resulting in features that lack local discriminability and spatial consistency. To address this issue, we propose DeCLIP, a novel framework that enhances CLIP by decoupling the self-attention module to obtain ``content'' and ``context'' features respectively. The ``content'' features are aligned with image crop representations to improve local discriminability, while ``context'' features learn to retain the spatial correlations under the guidance of vision foundation models, such as DINO. Extensive experiments demonstrate that DeCLIP significantly outperforms existing methods across multiple open-vocabulary dense prediction tasks, including object detection and semantic segmentation. Code is available at \textcolor{magenta}{https://github.com/xiaomoguhz/DeCLIP}.

Paperid: 1103, https://arxiv.org/pdf/2505.03912.pdf GitHub

Abstract:
Dual-system VLA (Vision-Language-Action) architectures have become a hot topic in embodied intelligence research, but there is a lack of sufficient open-source work for further performance analysis and optimization. To address this problem, this paper will summarize and compare the structural designs of existing dual-system architectures, and conduct systematic empirical evaluations on the core design elements of existing dual-system architectures. Ultimately, it will provide a low-cost open-source model for further exploration. Of course, this project will continue to update with more experimental conclusions and open-source models with improved performance for everyone to choose from. Project page: https://openhelix-robot.github.io/.

Paperid: 1104, https://arxiv.org/pdf/2505.03792.pdf GitHub

Abstract:
Online fine-tuning vision-language model (VLM) agents with reinforcement learning (RL) has shown promise for equipping agents with multi-step, goal-oriented capabilities in dynamic environments. However, their open-ended textual action space and non-end-to-end nature of action generation present significant challenges to effective online exploration in RL, e.g., explosion of the exploration space. We propose a novel online fine-tuning method, Counterfactual Soft Reinforcement Learning (CoSo), better suited to the textual output space of VLM agents. Compared to prior methods that assign uniform uncertainty to all tokens, CoSo leverages counterfactual reasoning to dynamically assess the causal influence of individual tokens on post-processed actions. By prioritizing the exploration of action-critical tokens while reducing the impact of semantically redundant or low-impact tokens, CoSo enables a more targeted and efficient online rollout process. We provide theoretical analysis proving CoSo's convergence and policy improvement guarantees, and extensive empirical evaluations supporting CoSo's effectiveness. Our results across a diverse set of agent tasks, including Android device control, card gaming, and embodied AI, highlight its remarkable ability to enhance exploration efficiency and deliver consistent performance gains. The code is available at https://github.com/langfengQ/CoSo.

Paperid: 1105, https://arxiv.org/pdf/2505.03242.pdf GitHub

Abstract:
Natural language goes beyond dryly describing visual content. It contains rich abstract concepts to express feeling, creativity and properties that cannot be directly perceived. Yet, current research in Vision Language Models (VLMs) has not shed light on abstract-oriented language. Our research breaks new ground by uncovering its wide presence and under-estimated value, with extensive analysis. Particularly, we focus our investigation on the fashion domain, a highly-representative field with abstract expressions. By analyzing recent large-scale multimodal fashion datasets, we find that abstract terms have a dominant presence, rivaling the concrete ones, providing novel information, and being useful in the retrieval task. However, a critical challenge emerges: current general-purpose or fashion-specific VLMs are pre-trained with databases that lack sufficient abstract words in their text corpora, thus hindering their ability to effectively represent abstract-oriented language. We propose a training-free and model-agnostic method, Abstract-to-Concrete Translator (ACT), to shift abstract representations towards well-represented concrete ones in the VLM latent space, using pre-trained models and existing multimodal databases. On the text-to-image retrieval task, despite being training-free, ACT outperforms the fine-tuned VLMs in both same- and cross-dataset settings, exhibiting its effectiveness with a strong generalization capability. Moreover, the improvement introduced by ACT is consistent with various VLMs, making it a plug-and-play solution.

Paperid: 1106, https://arxiv.org/pdf/2505.02971.pdf GitHub

Abstract:
Adversarial attacks have been fairly explored for computer vision and vision-language models. However, the avenue of adversarial attack for the vision language segmentation models (VLSMs) is still under-explored, especially for medical image analysis. Thus, we have investigated the robustness of VLSMs against adversarial attacks for 2D medical images with different modalities with radiology, photography, and endoscopy. The main idea of this project was to assess the robustness of the fine-tuned VLSMs specially in the medical domain setting to address the high risk scenario. First, we have fine-tuned pre-trained VLSMs for medical image segmentation with adapters. Then, we have employed adversarial attacks -- projected gradient descent (PGD) and fast gradient sign method (FGSM) -- on that fine-tuned model to determine its robustness against adversaries. We have reported models' performance decline to analyze the adversaries' impact. The results exhibit significant drops in the DSC and IoU scores after the introduction of these adversaries. Furthermore, we also explored universal perturbation but were not able to find for the medical images. \footnote{https://github.com/anjilab/secure-private-ai}

Paperid: 1107, https://arxiv.org/pdf/2505.02829.pdf GitHub

Abstract:
Segmentation models can recognize a pre-defined set of objects in images. However, models that can reason over complex user queries that implicitly refer to multiple objects of interest are still in their infancy. Recent advances in reasoning segmentation--generating segmentation masks from complex, implicit query text--demonstrate that vision-language models can operate across an open domain and produce reasonable outputs. However, our experiments show that such models struggle with complex remote-sensing imagery. In this work, we introduce LISAt, a vision-language model designed to describe complex remote-sensing scenes, answer questions about them, and segment objects of interest. We trained LISAt on a new curated geospatial reasoning-segmentation dataset, GRES, with 27,615 annotations over 9,205 images, and a multimodal pretraining dataset, PreGRES, containing over 1 million question-answer pairs. LISAt outperforms existing geospatial foundation models such as RS-GPT4V by over 10.04 % (BLEU-4) on remote-sensing description tasks, and surpasses state-of-the-art open-domain models on reasoning segmentation tasks by 143.36 % (gIoU). Our model, datasets, and code are available at https://lisat-bair.github.io/LISAt/

Paperid: 1108, https://arxiv.org/pdf/2505.02370.pdf GitHub

Abstract:
Due to the challenges of manually collecting accurate editing data, existing datasets are typically constructed using various automated methods, leading to noisy supervision signals caused by the mismatch between editing instructions and original-edited image pairs. Recent efforts attempt to improve editing models through generating higher-quality edited images, pre-training on recognition tasks, or introducing vision-language models (VLMs) but fail to resolve this fundamental issue. In this paper, we offer a novel solution by constructing more effective editing instructions for given image pairs. This includes rectifying the editing instructions to better align with the original-edited image pairs and using contrastive editing instructions to further enhance their effectiveness. Specifically, we find that editing models exhibit specific generation attributes at different inference steps, independent of the text. Based on these prior attributes, we define a unified guide for VLMs to rectify editing instructions. However, there are some challenging editing scenarios that cannot be resolved solely with rectified instructions. To this end, we further construct contrastive supervision signals with positive and negative instructions and introduce them into the model training using triplet loss, thereby further facilitating supervision effectiveness. Our method does not require the VLM modules or pre-training tasks used in previous work, offering a more direct and efficient way to provide better supervision signals, and providing a novel, simple, and effective solution for instruction-based image editing. Results on multiple benchmarks demonstrate that our method significantly outperforms existing approaches. Compared with previous SOTA SmartEdit, we achieve 9.19% improvements on the Real-Edit benchmark with 30x less training data and 13x smaller model size.

Paperid: 1109, https://arxiv.org/pdf/2505.02325.pdf GitHub

Abstract:
Learning discriminative 3D representations that generalize well to unknown testing categories is an emerging requirement for many real-world 3D applications. Existing well-established methods often struggle to attain this goal due to insufficient 3D training data from broader concepts. Meanwhile, pre-trained large vision-language models (e.g., CLIP) have shown remarkable zero-shot generalization capabilities. Yet, they are limited in extracting suitable 3D representations due to substantial gaps between their 2D training and 3D testing distributions. To address these challenges, we propose Testing-time Distribution Alignment (TeDA), a novel framework that adapts a pretrained 2D vision-language model CLIP for unknown 3D object retrieval at test time. To our knowledge, it is the first work that studies the test-time adaptation of a vision-language model for 3D feature learning. TeDA projects 3D objects into multi-view images, extracts features using CLIP, and refines 3D query embeddings with an iterative optimization strategy by confident query-target sample pairs in a self-boosting manner. Additionally, TeDA integrates textual descriptions generated by a multimodal language model (InternVL) to enhance 3D object understanding, leveraging CLIP's aligned feature space to fuse visual and textual cues. Extensive experiments on four open-set 3D object retrieval benchmarks demonstrate that TeDA greatly outperforms state-of-the-art methods, even those requiring extensive training. We also experimented with depth maps on Objaverse-LVIS, further validating its effectiveness. Code is available at https://github.com/wangzhichuan123/TeDA.

Paperid: 1110, https://arxiv.org/pdf/2505.02278.pdf GitHub

Abstract:
Vision-language pretraining on large datasets of images-text pairs is one of the main building blocks of current Vision-Language Models. While with additional training, these models excel in various downstream tasks, including visual question answering, image captioning, and visual commonsense reasoning. However, a notable weakness of pretrained models like CLIP, is their inability to perform entity grounding and compositional image and text matching~\cite{Jiang2024ComCLIP, yang2023amc, Rajabi2023GroundedVSR, learninglocalizeCVPR24}. In this work we propose a novel learning-free zero-shot augmentation of CLIP embeddings that has favorable compositional properties. We compute separate embeddings of sub-images of object entities and relations that are localized by the state of the art open vocabulary detectors and dynamically adjust the baseline global image embedding. % The final embedding is obtained by computing a weighted combination of the sub-image embeddings. The resulting embedding is then utilized for similarity computation with text embedding, resulting in a average 1.5\% improvement in image-text matching accuracy on the Visual Genome and SVO Probes datasets~\cite{krishna2017visualgenome, svo}. Notably, the enhanced embeddings demonstrate superior retrieval performance, thus achieving significant gains on the Flickr30K and MS-COCO retrieval benchmarks~\cite{flickr30ke, mscoco}, improving the state-of-the-art Recall@1 by 12\% and 0.4\%, respectively. Our code is available at https://github.com/madhukarreddyvongala/GroundingCLIP.

Paperid: 1111, https://arxiv.org/pdf/2505.01709.pdf GitHub

Abstract:
Operating robots in open-ended scenarios with diverse tasks is a crucial research and application direction in robotics. While recent progress in natural language processing and large multimodal models has enhanced robots' ability to understand complex instructions, robot manipulation still faces the procedural skill dilemma and the declarative skill dilemma in open environments. Existing methods often compromise cognitive and executive capabilities. To address these challenges, in this paper, we propose RoBridge, a hierarchical intelligent architecture for general robotic manipulation. It consists of a high-level cognitive planner (HCP) based on a large-scale pre-trained vision-language model (VLM), an invariant operable representation (IOR) serving as a symbolic bridge, and a generalist embodied agent (GEA). RoBridge maintains the declarative skill of VLM and unleashes the procedural skill of reinforcement learning, effectively bridging the gap between cognition and execution. RoBridge demonstrates significant performance improvements over existing baselines, achieving a 75% success rate on new tasks and an 83% average success rate in sim-to-real generalization using only five real-world data samples per task. This work represents a significant step towards integrating cognitive reasoning with physical execution in robotic systems, offering a new paradigm for general robotic manipulation.

Paperid: 1112, https://arxiv.org/pdf/2505.01481.pdf GitHub

Abstract:
Vision-Language Models (VLMs) have achieved strong results in video understanding, yet a key question remains: do they truly comprehend visual content or only learn shallow correlations between vision and language? Real visual understanding, especially of physics and common sense, is essential for AI systems that interact with the physical world. Current evaluations mostly use real-world videos similar to training data, so high benchmark scores may not reflect real reasoning ability. To address this, we propose negative-control tests using videos that depict physically impossible or logically inconsistent events. We introduce VideoHallu, a synthetic dataset of physics- and commonsense-violating scenes generated with Veo2, Sora, and Kling. It includes expert-annotated question-answer pairs across four categories of violations. Tests of leading VLMs (Qwen-2.5-VL, Video-R1, VideoChat-R1) show that, despite strong results on benchmarks such as MVBench and MMVU, they often miss these violations, exposing gaps in visual reasoning. Reinforcement learning fine-tuning on VideoHallu improves recognition of such violations without reducing standard benchmark performance. Our data is available at https://github.com/zli12321/VideoHallu.git.

Paperid: 1113, https://arxiv.org/pdf/2505.01431.pdf GitHub

Abstract:
Camouflaged object segmentation presents unique challenges compared to traditional segmentation tasks, primarily due to the high similarity in patterns and colors between camouflaged objects and their backgrounds. Effective solutions to this problem have significant implications in critical areas such as pest control, defect detection, and lesion segmentation in medical imaging. Prior research has predominantly emphasized supervised or unsupervised pre-training methods, leaving zero-shot approaches significantly underdeveloped. Existing zero-shot techniques commonly utilize the Segment Anything Model (SAM) in automatic mode or rely on vision-language models to generate cues for segmentation; however, their performances remain unsatisfactory, due to the similarity of the camouflaged object and the background. This work studies how to avoid training by integrating large pre-trained models like SAM-2 and Owl-v2 with temporal information into a modular pipeline. Evaluated on the MoCA-Mask dataset, our approach achieves outstanding performance improvements, significantly outperforming existing zero-shot methods by raising the F-measure ($F_Î²^w$) from 0.296 to 0.628. Our approach also surpasses supervised methods, increasing the F-measure from 0.476 to 0.628. Additionally, evaluation on the MoCA-Filter dataset demonstrates an increase in the success rate from 0.628 to 0.697 when compared with FlowSAM, a supervised transfer method. A thorough ablation study further validates the individual contributions of each component. Besides our main contributions, we also highlight inconsistencies in previous work regarding metrics and settings. Code can be found in https://github.com/weathon/vcos.

Paperid: 1114, https://arxiv.org/pdf/2505.00684.pdf GitHub GitHub

Abstract:
We introduce RegionFocus, a visual test-time scaling approach for Vision Language Model Agents. Understanding webpages is challenging due to the visual complexity of GUI images and the large number of interface elements, making accurate action selection difficult. Our approach dynamically zooms in on relevant regions, reducing background clutter and improving grounding accuracy. To support this process, we propose an image-as-map mechanism that visualizes key landmarks at each step, providing a transparent action record and enables the agent to effectively choose among action candidates. Even with a simple region selection strategy, we observe significant performance gains of 28+\% on Screenspot-pro and 24+\% on WebVoyager benchmarks on top of two state-of-the-art open vision language model agents, UI-TARS and Qwen2.5-VL, highlighting the effectiveness of visual test-time scaling in interactive settings. We achieve a new state-of-the-art grounding performance of 61.6\% on the ScreenSpot-Pro benchmark by applying RegionFocus to a Qwen2.5-VL-72B model. Our code will be released publicly at https://github.com/tiangeluo/RegionFocus.

Paperid: 1115, https://arxiv.org/pdf/2505.00284.pdf GitHub

Abstract:
Vision-Language Models (VLMs) have demonstrated significant potential for end-to-end autonomous driving. However, the field still lacks a practical platform that enables dynamic model updates, rapid validation, fair comparison, and intuitive performance assessment. To that end, we introduce LightEMMA, a Lightweight End-to-End Multimodal Model for Autonomous driving. LightEMMA provides a unified, VLM-based autonomous driving framework without ad hoc customizations, enabling easy integration with evolving state-of-the-art commercial and open-source models. We construct twelve autonomous driving agents using various VLMs and evaluate their performance on the challenging nuScenes prediction task, comprehensively assessing computational metrics and providing critical insights. Illustrative examples show that, although VLMs exhibit strong scenario interpretation capabilities, their practical performance in autonomous driving tasks remains a concern. Additionally, increased model complexity and extended reasoning do not necessarily lead to better performance, emphasizing the need for further improvements and task-specific designs. The code is available at https://github.com/michigan-traffic-lab/LightEMMA.

Paperid: 1116, https://arxiv.org/pdf/2505.00254.pdf GitHub

Abstract:
AI-driven video analytics has become increasingly important across diverse domains. However, existing systems are often constrained to specific, predefined tasks, limiting their adaptability in open-ended analytical scenarios. The recent emergence of Vision Language Models (VLMs) as transformative technologies offers significant potential for enabling open-ended video understanding, reasoning, and analytics. Nevertheless, their limited context windows present challenges when processing ultra-long video content, which is prevalent in real-world applications. To address this, we introduce AVA, a VLM-powered system designed for open-ended, advanced video analytics. AVA incorporates two key innovations: (1) the near real-time construction of Event Knowledge Graphs (EKGs) for efficient indexing of long or continuous video streams, and (2) an agentic retrieval-generation mechanism that leverages EKGs to handle complex and diverse queries. Comprehensive evaluations on public benchmarks, LVBench and VideoMME-Long, demonstrate that AVA achieves state-of-the-art performance, attaining 62.3% and 64.1% accuracy, respectively-significantly surpassing existing VLM and video Retrieval-Augmented Generation (RAG) systems. Furthermore, to evaluate video analytics in ultra-long and open-world video scenarios, we introduce a new benchmark, AVA-100. This benchmark comprises 8 videos, each exceeding 10 hours in duration, along with 120 manually annotated, diverse, and complex question-answer pairs. On AVA-100, AVA achieves top-tier performance with an accuracy of 75.8%. The source code of AVA is available at https://github.com/I-ESC/Project-Ava. The AVA-100 benchmark can be accessed at https://huggingface.co/datasets/iesc/Ava-100.

Paperid: 1117, https://arxiv.org/pdf/2504.21344.pdf GitHub

Abstract:
Machine learning models have utilized semantic features, deep features, or both to assess lung nodule malignancy. However, their reliance on manual annotation during inference, limited interpretability, and sensitivity to imaging variations hinder their application in real-world clinical settings. Thus, this research aims to integrate semantic features derived from radiologists' assessments of nodules, guiding the model to learn clinically relevant, robust, and explainable imaging features for predicting lung cancer. We obtained 938 low-dose CT scans from the National Lung Screening Trial (NLST) with 1,246 nodules and semantic features. Additionally, the Lung Image Database Consortium dataset contains 1,018 CT scans, with 2,625 lesions annotated for nodule characteristics. Three external datasets were obtained from UCLA Health, the LUNGx Challenge, and the Duke Lung Cancer Screening. We fine-tuned a pretrained Contrastive Language-Image Pretraining (CLIP) model with a parameter-efficient fine-tuning approach to align imaging and semantic text features and predict the one-year lung cancer diagnosis. Our model outperformed state-of-the-art (SOTA) models in the NLST test set with an AUROC of 0.901 and AUPRC of 0.776. It also showed robust results in external datasets. Using CLIP, we also obtained predictions on semantic features through zero-shot inference, such as nodule margin (AUROC: 0.812), nodule consistency (0.812), and pleural attachment (0.840). Our approach surpasses the SOTA models in predicting lung cancer across datasets collected from diverse clinical settings, providing explainable outputs, aiding clinicians in comprehending the underlying meaning of model predictions. This approach also prevents the model from learning shortcuts and generalizes across clinical settings. The code is available at https://github.com/luotingzhuang/CLIP_nodule.

Paperid: 1118, https://arxiv.org/pdf/2504.21063.pdf GitHub

Abstract:
Federated domain generalization (FedDG) aims to learn a globally generalizable model from decentralized clients with heterogeneous data while preserving privacy. Recent studies have introduced prompt learning to adapt vision-language models (VLMs) in FedDG by learning a single global prompt. However, such a one-prompt-fits-all learning paradigm typically leads to performance degradation on personalized samples. Although the mixture of experts (MoE) offers a promising solution for specialization, existing MoE-based methods suffer from coarse image-level expert assignment and high communication costs from parameterized routers. To address these limitations, we propose TRIP, a Token-level prompt mixture with parameter-free routing framework for FedDG, which treats multiple prompts as distinct experts. Unlike existing image-level routing designs, TRIP assigns different tokens within an image to specific experts. To ensure communication efficiency, TRIP incorporates a parameter-free routing mechanism based on token clustering and optimal transport. The instance-specific prompt is then synthesized by aggregating experts, weighted by the number of tokens assigned to each. Additionally, TRIP develops an unbiased learning strategy for prompt experts, leveraging the VLM's zero-shot generalization capability. Extensive experiments across four benchmarks demonstrate that TRIP achieves optimal generalization results, with communication of only 1K parameters per round. Our code is available at https://github.com/GongShuai8210/TRIP.

Paperid: 1119, https://arxiv.org/pdf/2504.20860.pdf GitHub

Abstract:
In federated learning, textual prompt tuning adapts Vision-Language Models (e.g., CLIP) by tuning lightweight input tokens (or prompts) on local client data, while keeping network weights frozen. After training, only the prompts are shared by the clients with the central server for aggregation. However, textual prompt tuning suffers from overfitting to known concepts, limiting its generalizability to unseen concepts. To address this limitation, we propose Multimodal Visual Prompt Tuning (FedMVP) that conditions the prompts on multimodal contextual information - derived from the input image and textual attribute features of a class. At the core of FedMVP is a PromptFormer module that synergistically aligns textual and visual features through a cross-attention mechanism. The dynamically generated multimodal visual prompts are then input to the frozen vision encoder of CLIP, and trained with a combination of CLIP similarity loss and a consistency loss. Extensive evaluation on 20 datasets, spanning three generalization settings, demonstrates that FedMVP not only preserves performance on in-distribution classes and domains, but also displays higher generalizability to unseen classes and domains, surpassing state-of-the-art methods by a notable margin of +1.57% - 2.26%. Code is available at https://github.com/mainaksingha01/FedMVP.

Paperid: 1120, https://arxiv.org/pdf/2504.20343.pdf GitHub

Abstract:
Medical image reporting (MIR) aims to generate structured clinical descriptions from radiological images. Existing methods struggle with fine-grained feature extraction, multimodal alignment, and generalization across diverse imaging types, often relying on vanilla transformers and focusing primarily on chest X-rays. We propose MicarVLMoE, a vision-language mixture-of-experts model with gated cross-aligned fusion, designed to address these limitations. Our architecture includes: (i) a multiscale vision encoder (MSVE) for capturing anatomical details at varying resolutions, (ii) a multihead dual-branch latent attention (MDLA) module for vision-language alignment through latent bottleneck representations, and (iii) a modulated mixture-of-experts (MoE) decoder for adaptive expert specialization. We extend MIR to CT scans, retinal imaging, MRI scans, and gross pathology images, reporting state-of-the-art results on COVCTR, MMR, PGROSS, and ROCO datasets. Extensive experiments and ablations confirm improved clinical accuracy, cross-modal alignment, and model interpretability. Code is available at https://github.com/AI-14/micar-vl-moe.

Paperid: 1121, https://arxiv.org/pdf/2504.19742.pdf GitHub

Abstract:
The presence of species provides key insights into the ecological properties of a location such as land cover, climatic conditions or even soil properties. We propose a method to predict such ecological properties directly from remote sensing (RS) images by aligning them with species habitat descriptions. We introduce the EcoWikiRS dataset, consisting of high-resolution aerial images, the corresponding geolocated species observations, and, for each species, the textual descriptions of their habitat from Wikipedia. EcoWikiRS offers a scalable way of supervision for RS vision language models (RS-VLMs) for ecology. This is a setting with weak and noisy supervision, where, for instance, some text may describe properties that are specific only to part of the species' niche or is irrelevant to a specific image. We tackle this by proposing WINCEL, a weighted version of the InfoNCE loss. We evaluate our model on the task of ecosystem zero-shot classification by following the habitat definitions from the European Nature Information System (EUNIS). Our results show that our approach helps in understanding RS images in a more ecologically meaningful manner. The code and the dataset are available at https://github.com/eceo-epfl/EcoWikiRS.

Paperid: 1122, https://arxiv.org/pdf/2504.19524.pdf GitHub

Abstract:
Industrial Anomaly Detection (IAD) is critical for ensuring product quality by identifying defects. Traditional methods such as feature embedding and reconstruction-based approaches require large datasets and struggle with scalability. Existing vision-language models (VLMs) and Multimodal Large Language Models (MLLMs) address some limitations but rely on mask annotations, leading to high implementation costs and false positives. Additionally, industrial datasets like MVTec-AD and VisA suffer from severe class imbalance, with defect samples constituting only 23.8% and 11.1% of total data respectively. To address these challenges, we propose a reward function that dynamically prioritizes rare defect patterns during training to handle class imbalance. We also introduce a mask-free reasoning framework using Chain of Thought (CoT) and Group Relative Policy Optimization (GRPO) mechanisms, enabling anomaly detection directly from raw images without annotated masks. This approach generates interpretable step-by-step explanations for defect localization. Our method achieves state-of-the-art performance, outperforming prior approaches by 36% in accuracy on MVTec-AD and 16% on VisA. By eliminating mask dependency and reducing costs while providing explainable outputs, this work advances industrial anomaly detection and supports scalable quality control in manufacturing. Code to reproduce the experiment is available at https://github.com/LilaKen/LR-IAD.

Paperid: 1123, https://arxiv.org/pdf/2504.19127.pdf GitHub GitHub

Abstract:
There has long been a belief that high-level semantics learning can benefit various downstream computer vision tasks. However, in the low-light image enhancement (LLIE) community, existing methods learn a brutal mapping between low-light and normal-light domains without considering the semantic information of different regions, especially in those extremely dark regions that suffer from severe information loss. To address this issue, we propose a new deep semantic prior-guided framework (DeepSPG) based on Retinex image decomposition for LLIE to explore informative semantic knowledge via a pre-trained semantic segmentation model and multimodal learning. Notably, we incorporate both image-level semantic prior and text-level semantic prior and thus formulate a multimodal learning framework with combinatorial deep semantic prior guidance for LLIE. Specifically, we incorporate semantic knowledge to guide the enhancement process via three designs: an image-level semantic prior guidance by leveraging hierarchical semantic features from a pre-trained semantic segmentation model; a text-level semantic prior guidance by integrating natural language semantic constraints via a pre-trained vision-language model; a multi-scale semantic-aware structure that facilitates effective semantic feature incorporation. Eventually, our proposed DeepSPG demonstrates superior performance compared to state-of-the-art methods across five benchmark datasets. The implementation details and code are publicly available at https://github.com/Wenyuzhy/DeepSPG.

Paperid: 1124, https://arxiv.org/pdf/2504.18856.pdf GitHub

Abstract:
In Computational Pathology (CPath), the introduction of Vision-Language Models (VLMs) has opened new avenues for research, focusing primarily on aligning image-text pairs at a single magnification level. However, this approach might not be sufficient for tasks like cancer subtype classification, tissue phenotyping, and survival analysis due to the limited level of detail that a single-resolution image can provide. Addressing this, we propose a novel multi-resolution paradigm leveraging Whole Slide Images (WSIs) to extract histology patches at multiple resolutions and generate corresponding textual descriptions through advanced CPath VLM. We introduce visual-textual alignment at multiple resolutions as well as cross-resolution alignment to establish more effective text-guided visual representations. Cross-resolution alignment using a multimodal encoder enhances the model's ability to capture context from multiple resolutions in histology images. Our model aims to capture a broader range of information, supported by novel loss functions, enriches feature representation, improves discriminative ability, and enhances generalization across different resolutions. Pre-trained on a comprehensive TCGA dataset with 34 million image-language pairs at various resolutions, our fine-tuned model outperforms state-of-the-art (SOTA) counterparts across multiple datasets and tasks, demonstrating its effectiveness in CPath. The code is available on GitHub at: https://github.com/BasitAlawode/MR-PLIP

Paperid: 1125, https://arxiv.org/pdf/2504.18589.pdf GitHub GitHub

Abstract:
Recent advancements in Large Vision-Language Models (LVLMs) have significantly enhanced their ability to integrate visual and linguistic information, achieving near-human proficiency in tasks like object recognition, captioning, and visual question answering. However, current benchmarks typically focus on knowledge-centric evaluations that assess domain-specific expertise, often neglecting the core ability to reason about fundamental mathematical elements and visual concepts. We identify a gap in evaluating elementary-level math problems, which rely on explicit visual dependencies-requiring models to discern, integrate, and reason across multiple images while incorporating commonsense knowledge, all of which are crucial for advancing toward broader AGI capabilities. To address this gap, we introduce VCBENCH, a comprehensive benchmark for multimodal mathematical reasoning with explicit visual dependencies. VCBENCH includes 1,720 problems across six cognitive domains, featuring 6,697 images (averaging 3.9 per question) to ensure multi-image reasoning. We evaluate 26 state-of-the-art LVLMs on VCBENCH, revealing substantial performance disparities, with even the top models unable to exceed 50% accuracy. Our findings highlight the ongoing challenges in visual-mathematical integration and suggest avenues for future LVLM advancements. The project can be found at https://alibaba-damo-academy.github.io/VCBench/.

Paperid: 1126, https://arxiv.org/pdf/2504.17432.pdf GitHub

Abstract:
The Contrastive Language-Image Pre-training (CLIP) framework has become a widely used approach for multimodal representation learning, particularly in image-text retrieval and clustering. However, its efficacy is constrained by three key limitations: (1) text token truncation, (2) isolated image-text encoding, and (3) deficient compositionality due to bag-of-words behavior. While recent Multimodal Large Language Models (MLLMs) have demonstrated significant advances in generalized vision-language understanding, their potential for learning transferable multimodal representations remains underexplored.In this work, we present UniME (Universal Multimodal Embedding), a novel two-stage framework that leverages MLLMs to learn discriminative representations for diverse downstream tasks. In the first stage, we perform textual discriminative knowledge distillation from a powerful LLM-based teacher model to enhance the embedding capability of the MLLMÅ language component. In the second stage, we introduce hard negative enhanced instruction tuning to further advance discriminative representation learning. Specifically, we initially mitigate false negative contamination and then sample multiple hard negatives per instance within each batch, forcing the model to focus on challenging samples. This approach not only improves discriminative power but also enhances instruction-following ability in downstream tasks. We conduct extensive experiments on the MMEB benchmark and multiple retrieval tasks, including short and long caption retrieval and compositional retrieval. Results demonstrate that UniME achieves consistent performance improvement across all tasks, exhibiting superior discriminative and compositional capabilities.

Paperid: 1127, https://arxiv.org/pdf/2504.17040.pdf GitHub

Abstract:
We present DyMU, an efficient, training-free framework that dynamically reduces the computational burden of vision-language models (VLMs) while maintaining high task performance. Our approach comprises two key components. First, Dynamic Token Merging (DToMe) reduces the number of visual token embeddings by merging similar tokens based on image complexity, addressing the inherent inefficiency of fixed-length outputs in vision transformers. Second, Virtual Token Unmerging (VTU) simulates the expected token sequence for large language models (LLMs) by efficiently reconstructing the attention dynamics of a full sequence, thus preserving the downstream performance without additional fine-tuning. Unlike previous approaches, our method dynamically adapts token compression to the content of the image and operates completely training-free, making it readily applicable to most state-of-the-art VLM architectures. Extensive experiments on image and video understanding tasks demonstrate that DyMU can reduce the average visual token count by 32%-85% while achieving comparable performance to full-length models across diverse VLM architectures, including the recently popularized AnyRes-based visual encoders. Furthermore, through qualitative analyses, we demonstrate that DToMe effectively adapts token reduction based on image complexity and, unlike existing systems, provides users more control over computational costs. Project page: https://mikewangwzhl.github.io/dymu/.

Paperid: 1128, https://arxiv.org/pdf/2504.16801.pdf GitHub

Abstract:
Contrastive Language-Image Pre-training (CLIP) has achieved success on multiple downstream tasks by aligning image and text modalities. However, the nature of global contrastive learning limits CLIP's ability to comprehend compositional concepts, such as relations and attributes. Although recent studies employ global hard negative samples to improve compositional understanding, these methods significantly compromise the model's inherent general capabilities by forcibly distancing textual negative samples from images in the embedding space. To overcome this limitation, we introduce a Decoupled Global-Local Alignment (DeGLA) framework that improves compositional understanding while substantially mitigating losses in general capabilities. To optimize the retention of the model's inherent capabilities, we incorporate a self-distillation mechanism within the global alignment process, aligning the learnable image-text encoder with a frozen teacher model derived from an exponential moving average. Under the constraint of self-distillation, it effectively mitigates the catastrophic forgetting of pretrained knowledge during fine-tuning. To improve compositional understanding, we first leverage the in-context learning capability of Large Language Models (LLMs) to construct about 2M high-quality negative captions across five types. Subsequently, we propose the Image-Grounded Contrast (IGC) loss and Text-Grounded Contrast (TGC) loss to enhance vision-language compositionally. Extensive experimental results demonstrate the effectiveness of the DeGLA framework. Compared to previous state-of-the-art methods, DeGLA achieves an average enhancement of 3.5% across the VALSE, SugarCrepe, and ARO benchmarks. Concurrently, it obtains an average performance improvement of 13.0% on zero-shot classification tasks across eleven datasets. Our code will be released at https://github.com/xiaoxing2001/DeGLA

Paperid: 1129, https://arxiv.org/pdf/2504.16433.pdf GitHub

Abstract:
In recent years, large-scale vision-language models (VLMs) like CLIP have gained attention for their zero-shot inference using instructional text prompts. While these models excel in general computer vision, their potential for domain generalization in remote sensing (RS) remains underexplored. Existing approaches enhance prompt learning by generating visual prompt tokens but rely on full-image features, introducing noise and background artifacts that vary within a class, causing misclassification. To address this, we propose FrogDogNet, a novel prompt learning framework integrating Fourier frequency filtering and self-attention to improve RS scene classification and domain generalization. FrogDogNet selectively retains invariant low-frequency components while eliminating noise and irrelevant backgrounds, ensuring robust feature representation across domains. The model first extracts significant features via projection and self-attention, then applies frequency-based filtering to preserve essential structural information for prompt learning. Extensive experiments on four RS datasets and three domain generalization tasks show that FrogDogNet consistently outperforms state-of-the-art prompt learning methods, demonstrating superior adaptability across domain shifts. Our findings highlight the effectiveness of frequency-based invariant feature retention in generalization, paving the way for broader applications. Our code is available at https://github.com/HariseetharamG/FrogDogNet

Paperid: 1130, https://arxiv.org/pdf/2504.16082.pdf GitHub

Abstract:
We propose MR. Video, an agentic long video understanding framework that demonstrates the simple yet effective MapReduce principle for processing long videos: (1) Map: independently and densely perceiving short video clips, and (2) Reduce: jointly aggregating information from all clips. Compared with sequence-to-sequence vision-language models (VLMs), MR. Video performs detailed short video perception without being limited by context length. Compared with existing video agents that typically rely on sequential key segment selection, the Map operation enables simpler and more scalable sequence parallel perception of short video segments. Its Reduce step allows for more comprehensive context aggregation and reasoning, surpassing explicit key segment retrieval. This MapReduce principle is applicable to both VLMs and video agents, and we use LLM agents to validate its effectiveness. In practice, MR. Video employs two MapReduce stages: (A) Captioning: generating captions for short video clips (map), then standardizing repeated characters and objects into shared names (reduce); (B) Analysis: for each user question, analyzing relevant information from individual short videos (map), and integrating them into a final answer (reduce). MR. Video achieves over 10% accuracy improvement on the challenging LVBench compared to state-of-the-art VLMs and video agents. Code is available at: https://github.com/ziqipang/MR-Video

Paperid: 1131, https://arxiv.org/pdf/2504.16030.pdf GitHub

Abstract:
Recent video large language models (Video LLMs) often depend on costly human annotations or proprietary model APIs (e.g., GPT-4o) to produce training data, which limits their training at scale. In this paper, we explore large-scale training for Video LLM with cheap automatic speech recognition (ASR) transcripts. Specifically, we propose a novel streaming training approach that densely interleaves the ASR words and video frames according to their timestamps. Compared to previous studies in vision-language representation with ASR, our method naturally fits the streaming characteristics of ASR, thus enabling the model to learn temporally-aligned, fine-grained vision-language modeling. To support the training algorithm, we introduce a data production pipeline to process YouTube videos and their closed captions (CC, same as ASR), resulting in Live-CC-5M dataset for pre-training and Live-WhisperX-526K dataset for high-quality supervised fine-tuning (SFT). Remarkably, even without SFT, the ASR-only pre-trained LiveCC-7B-Base model demonstrates competitive general video QA performance and exhibits a new capability in real-time video commentary. To evaluate this, we carefully design a new LiveSports-3K benchmark, using LLM-as-a-judge to measure the free-form commentary. Experiments show our final LiveCC-7B-Instruct model can surpass advanced 72B models (Qwen2.5-VL-72B-Instruct, LLaVA-Video-72B) in commentary quality even working in a real-time mode. Meanwhile, it achieves state-of-the-art results at the 7B/8B scale on popular video QA benchmarks such as VideoMME and OVOBench, demonstrating the broad generalizability of our approach. All resources of this paper have been released at https://showlab.github.io/livecc.

Paperid: 1132, https://arxiv.org/pdf/2504.15545.pdf GitHub

Abstract:
In histopathology, tissue sections are typically stained using common H&E staining or special stains (MAS, PAS, PASM, etc.) to clearly visualize specific tissue structures. The rapid advancement of deep learning offers an effective solution for generating virtually stained images, significantly reducing the time and labor costs associated with traditional histochemical staining. However, a new challenge arises in separating the fundamental visual characteristics of tissue sections from the visual differences induced by staining agents. Additionally, virtual staining often overlooks essential pathological knowledge and the physical properties of staining, resulting in only style-level transfer. To address these issues, we introduce, for the first time in virtual staining tasks, a pathological vision-language large model (VLM) as an auxiliary tool. We integrate contrastive learnable prompts, foundational concept anchors for tissue sections, and staining-specific concept anchors to leverage the extensive knowledge of the pathological VLM. This approach is designed to describe, frame, and enhance the direction of virtual staining. Furthermore, we have developed a data augmentation method based on the constraints of the VLM. This method utilizes the VLM's powerful image interpretation capabilities to further integrate image style and structural information, proving beneficial in high-precision pathological diagnostics. Extensive evaluations on publicly available multi-domain unpaired staining datasets demonstrate that our method can generate highly realistic images and enhance the accuracy of downstream tasks, such as glomerular detection and segmentation. Our code is available at: https://github.com/CZZZZZZZZZZZZZZZZZ/VPGAN-HARBOR

Paperid: 1133, https://arxiv.org/pdf/2504.15485.pdf GitHub

Abstract:
Recognizing and reasoning about occluded (partially or fully hidden) objects is vital to understanding visual scenes, as occlusions frequently occur in real-world environments and act as obstacles for spatial comprehension. To test models' ability to reason about multiple occluded objects, we introduce a novel task, Counting Amodally for Patterns Through Unseen REgions (CAPTURe), which requires a model to count objects arranged in a pattern by inferring how the pattern continues behind an occluder (an object which blocks parts of the scene). CAPTURe requires both recognizing visual patterns and reasoning, making it a useful testbed for evaluating vision-language models (VLMs) on whether they understand occluded patterns and possess spatial understanding skills. By requiring models to reason about occluded objects, CAPTURe also tests VLMs' ability to form world models that would allow them to fill in missing information. CAPTURe consists of two parts: (1) CAPTURe-real, with manually filtered images of real objects in patterns and (2) CAPTURe-synthetic, a controlled diagnostic with generated patterned images. We evaluate four strong VLMs (GPT-4o, Intern-VL2, Molmo, and Qwen2-VL) on CAPTURe, finding that models struggle to count on both occluded and unoccluded patterns. Crucially, we find that models perform worse with occlusion, suggesting that VLMs are also deficient in inferring unseen spatial relationships: even the strongest VLMs like GPT-4o fail to count with occlusion. In contrast, we find that humans achieve very little error on CAPTURe. We also find that providing auxiliary information of occluded object locations increases performance, underscoring that the model error comes both from an inability to handle occlusion as well as difficulty in counting in images. Code and data: https://github.com/atinpothiraj/CAPTURe

Paperid: 1134, https://arxiv.org/pdf/2504.15135.pdf GitHub

Abstract:
Entity linking (EL) aligns textual mentions with their corresponding entities in a knowledge base, facilitating various applications such as semantic search and question answering. Recent advances in multimodal entity linking (MEL) have shown that combining text and images can reduce ambiguity and improve alignment accuracy. However, most existing MEL methods overlook the rich structural information available in the form of knowledge-graph (KG) triples. In this paper, we propose KGMEL, a novel framework that leverages KG triples to enhance MEL. Specifically, it operates in three stages: (1) Generation: Produces high-quality triples for each mention by employing vision-language models based on its text and images. (2) Retrieval: Learns joint mention-entity representations, via contrastive learning, that integrate text, images, and (generated or KG) triples to retrieve candidate entities for each mention. (3) Reranking: Refines the KG triples of the candidate entities and employs large language models to identify the best-matching entity for the mention. Extensive experiments on benchmark datasets demonstrate that KGMEL outperforms existing methods. Our code and datasets are available at: https://github.com/juyeonnn/KGMEL.

Paperid: 1135, https://arxiv.org/pdf/2504.14988.pdf GitHub

Abstract:
Recent advancements in Large Vision-Language Models (LVLMs) have demonstrated remarkable multimodal perception capabilities, garnering significant attention. While numerous evaluation studies have emerged, assessing LVLMs both holistically and on specialized tasks, fine-grained image tasks-fundamental to computer vision-remain largely unexplored. To fill this gap, we introduce a comprehensive fine-grained evaluation benchmark, i.e., FG-BMK, comprising 1.01 million questions and 0.33 million images. Our evaluation systematically examines LVLMs from both human-oriented and machine-oriented perspectives, focusing on their semantic recognition and fine-grained feature representation capabilities. Through extensive experiments on twelve representative LVLMs/VLMs, we uncover key findings regarding the influence of training paradigms, modality alignment, perturbation susceptibility, and fine-grained category reasoning on task performance. This work provides critical insights into the limitations of current LVLMs and offers guidance for future data construction and model design in the development of more advanced LVLMs. Our code is open-source and available at https://github.com/SEU-VIPGroup/FG-BMK.

Paperid: 1136, https://arxiv.org/pdf/2504.13945.pdf GitHub

Abstract:
The rapid advancement of large vision-language models (LVLMs) has significantly propelled applications in document understanding, particularly in optical character recognition (OCR) and multilingual translation. However, current evaluations of LVLMs, like the widely used OCRBench, mainly focus on verifying the correctness of their short-text responses and long-text responses with simple layout, while the evaluation of their ability to understand long texts with complex layout design is highly significant but largely overlooked. In this paper, we propose Menu OCR and Translation Benchmark (MOTBench), a specialized evaluation framework emphasizing the pivotal role of menu translation in cross-cultural communication. MOTBench requires LVLMs to accurately recognize and translate each dish, along with its price and unit items on a menu, providing a comprehensive assessment of their visual understanding and language processing capabilities. Our benchmark is comprised of a collection of Chinese and English menus, characterized by intricate layouts, a variety of fonts, and culturally specific elements across different languages, along with precise human annotations. Experiments show that our automatic evaluation results are highly consistent with professional human evaluation. We evaluate a range of publicly available state-of-the-art LVLMs, and through analyzing their output to identify the strengths and weaknesses in their performance, offering valuable insights to guide future advancements in LVLM development. MOTBench is available at https://github.com/gitwzl/MOTBench.

Paperid: 1137, https://arxiv.org/pdf/2504.13650.pdf GitHub

Abstract:
Medical Large Vision-Language Models (Med-LVLMs) demonstrate significant potential in healthcare, but their reliance on general medical data and coarse-grained global visual understanding limits them in intelligent ophthalmic diagnosis. Currently, intelligent ophthalmic diagnosis faces three major challenges: (i) Data. The lack of deeply annotated, high-quality, multi-modal ophthalmic visual instruction data; (ii) Benchmark. The absence of a comprehensive and systematic benchmark for evaluating diagnostic performance; (iii) Model. The difficulty of adapting holistic visual architectures to fine-grained, region-specific ophthalmic lesion identification. In this paper, we propose the Eyecare Kit, which systematically tackles the aforementioned three key challenges with the tailored dataset, benchmark and model: First, we construct a multi-agent data engine with real-life ophthalmology data to produce Eyecare-100K, a high-quality ophthalmic visual instruction dataset. Subsequently, we design Eyecare-Bench, a benchmark that comprehensively evaluates the overall performance of LVLMs on intelligent ophthalmic diagnosis tasks across multiple dimensions. Finally, we develop the EyecareGPT, optimized for fine-grained ophthalmic visual understanding thoroughly, which incorporates an adaptive resolution mechanism and a layer-wise dense connector. Extensive experimental results indicate that the EyecareGPT achieves state-of-the-art performance in a range of ophthalmic tasks, underscoring its significant potential for the advancement of open research in intelligent ophthalmic diagnosis. Our project is available at https://github.com/DCDmllm/EyecareGPT.

Paperid: 1138, https://arxiv.org/pdf/2504.13406.pdf GitHub

Abstract:
Multi-agent collaboration holds great promise for enhancing the safety, reliability, and mobility of autonomous driving systems by enabling information sharing among multiple connected agents. However, existing multi-agent communication approaches are hindered by limitations of existing communication media, including high bandwidth demands, agent heterogeneity, and information loss. To address these challenges, we introduce LangCoop, a new paradigm for collaborative autonomous driving that leverages natural language as a compact yet expressive medium for inter-agent communication. LangCoop features two key innovations: Mixture Model Modular Chain-of-thought (M$^3$CoT) for structured zero-shot vision-language reasoning and Natural Language Information Packaging (LangPack) for efficiently packaging information into concise, language-based messages. Through extensive experiments conducted in the CARLA simulations, we demonstrate that LangCoop achieves a remarkable 96\% reduction in communication bandwidth (< 2KB per message) compared to image-based communication, while maintaining competitive driving performance in the closed-loop evaluation. Our project page and code are at https://xiangbogaobarry.github.io/LangCoop/.

Paperid: 1139, https://arxiv.org/pdf/2504.13399.pdf GitHub

Abstract:
Detecting anomalous hazards in visual data, particularly in video streams, is a critical challenge in autonomous driving. Existing models often struggle with unpredictable, out-of-label hazards due to their reliance on predefined object categories. In this paper, we propose a multimodal approach that integrates vision-language reasoning with zero-shot object detection to improve hazard identification and explanation. Our pipeline consists of a Vision-Language Model (VLM), a Large Language Model (LLM), in order to detect hazardous objects within a traffic scene. We refine object detection by incorporating OpenAI's CLIP model to match predicted hazards with bounding box annotations, improving localization accuracy. To assess model performance, we create a ground truth dataset by denoising and extending the foundational COOOL (Challenge-of-Out-of-Label) anomaly detection benchmark dataset with complete natural language descriptions for hazard annotations. We define a means of hazard detection and labeling evaluation on the extended dataset using cosine similarity. This evaluation considers the semantic similarity between the predicted hazard description and the annotated ground truth for each video. Additionally, we release a set of tools for structuring and managing large-scale hazard detection datasets. Our findings highlight the strengths and limitations of current vision-language-based approaches, offering insights into future improvements in autonomous hazard detection systems. Our models, scripts, and data can be found at https://github.com/mi3labucm/COOOLER.git

Paperid: 1140, https://arxiv.org/pdf/2504.13181.pdf GitHub

Abstract:
We introduce Perception Encoder (PE), a state-of-the-art vision encoder for image and video understanding trained via simple vision-language learning. Traditionally, vision encoders have relied on a variety of pretraining objectives, each tailored to specific downstream tasks such as classification, captioning, or localization. Surprisingly, after scaling our carefully tuned image pretraining recipe and refining with our robust video data engine, we find that contrastive vision-language training alone can produce strong, general embeddings for all of these downstream tasks. There is only one caveat: these embeddings are hidden within the intermediate layers of the network. To draw them out, we introduce two alignment methods: language alignment for multimodal language modeling, and spatial alignment for dense prediction. Together, our PE family of models achieves best-in-class results on a wide variety of tasks, including (1) zero-shot image and video classification and retrieval, simultaneously obtaining 86.6 average zero-shot ImageNet robustness and 76.9 zero-shot Kinetics-400 video classification; (2) document, image, and video Q&A, enabling 94.6 DocVQA, 80.9 InfographicVQA, and 82.7 PerceptionTest with an 8B LLM; and (3) spatial tasks such as detection, tracking, and depth estimation, setting a new COCO state-of-the-art of 66.0 box mAP. To foster further research, we release our models, code, and novel dataset of synthetically and human-annotated videos: https://github.com/facebookresearch/perception_models

Paperid: 1141, https://arxiv.org/pdf/2504.13180.pdf GitHub

Abstract:
Vision-language models are integral to computer vision research, yet many high-performing models remain closed-source, obscuring their data, design and training recipe. The research community has responded by using distillation from black-box models to label training data, achieving strong benchmark results, at the cost of measurable scientific progress. However, without knowing the details of the teacher model and its data sources, scientific progress remains difficult to measure. In this paper, we study building a Perception Language Model (PLM) in a fully open and reproducible framework for transparent research in image and video understanding. We analyze standard training pipelines without distillation from proprietary models and explore large-scale synthetic data to identify critical data gaps, particularly in detailed video understanding. To bridge these gaps, we release 2.8M human-labeled instances of fine-grained video question-answer pairs and spatio-temporally grounded video captions. Additionally, we introduce PLM-VideoBench, a suite for evaluating challenging video understanding tasks focusing on the ability to reason about "what", "where", "when", and "how" of a video. We make our work fully reproducible by providing data, training recipes, code & models. https://github.com/facebookresearch/perception_models

Paperid: 1142, https://arxiv.org/pdf/2504.13120.pdf GitHub

Abstract:
The ability to combine existing concepts into novel ideas stands as a fundamental hallmark of human intelligence. Recent advances in Vision-Language Models (VLMs) like GPT-4V and DALLE-3 have sparked debate about whether their outputs reflect combinational creativity--defined by M. A. Boden (1998) as synthesizing novel ideas through combining existing concepts--or sophisticated pattern matching of training data. Drawing inspiration from cognitive science, we investigate the combinational creativity of VLMs from the lens of concept blending. We propose the Identification-Explanation-Implication (IEI) framework, which decomposes creative processes into three levels: identifying input spaces, extracting shared attributes, and deriving novel semantic implications. To validate this framework, we curate CreativeMashup, a high-quality dataset of 666 artist-generated visual mashups annotated according to the IEI framework. Through extensive experiments, we demonstrate that in comprehension tasks, best VLMs have surpassed average human performance while falling short of expert-level understanding; in generation tasks, incorporating our IEI framework into the generation pipeline significantly enhances the creative quality of VLMs' outputs. Our findings establish both a theoretical foundation for evaluating artificial creativity and practical guidelines for improving creative generation in VLMs.

Paperid: 1143, https://arxiv.org/pdf/2504.12717.pdf GitHub

Abstract:
Contrastive language image pre-training (CLIP) is an essential component of building modern vision-language foundation models. While CLIP demonstrates remarkable zero-shot performance on downstream tasks, the multi-modal feature spaces still suffer from a modality gap, which is a gap between image and text feature clusters and limits downstream task performance. Although existing works attempt to address the modality gap by modifying pre-training or fine-tuning, they struggle with heavy training costs with large datasets or degradations of zero-shot performance. This paper presents CLIP-Refine, a post-pre-training method for CLIP models at a phase between pre-training and fine-tuning. CLIP-Refine aims to align the feature space with 1 epoch training on small image-text datasets without zero-shot performance degradations. To this end, we introduce two techniques: random feature alignment (RaFA) and hybrid contrastive-distillation (HyCD). RaFA aligns the image and text features to follow a shared prior distribution by minimizing the distance to random reference vectors sampled from the prior. HyCD updates the model with hybrid soft labels generated by combining ground-truth image-text pair labels and outputs from the pre-trained CLIP model. This contributes to achieving both maintaining the past knowledge and learning new knowledge to align features. Our extensive experiments with multiple classification and retrieval tasks show that CLIP-Refine succeeds in mitigating the modality gap and improving the zero-shot performance.

Paperid: 1144, https://arxiv.org/pdf/2504.12256.pdf GitHub

Abstract:
Over the past years, advances in artificial intelligence (AI) have demonstrated how AI can solve many perception and generation tasks, such as image classification and text writing, yet reasoning remains a challenge. This paper introduces the FLIP dataset, a benchmark for evaluating AI reasoning capabilities based on human verification tasks on the Idena blockchain. FLIP challenges present users with two orderings of 4 images, requiring them to identify the logically coherent one. By emphasizing sequential reasoning, visual storytelling, and common sense, FLIP provides a unique testbed for multimodal AI systems. Our experiments evaluate state-of-the-art models, leveraging both vision-language models (VLMs) and large language models (LLMs). Results reveal that even the best open-sourced and closed-sourced models achieve maximum accuracies of 75.5% and 77.9%, respectively, in zero-shot settings, compared to human performance of 95.3%. Captioning models aid reasoning models by providing text descriptions of images, yielding better results than when using the raw images directly, 69.6% vs. 75.2% for Gemini 1.5 Pro. Combining the predictions from 15 models in an ensemble increases the accuracy to 85.2%. These findings highlight the limitations of existing reasoning models and the need for robust multimodal benchmarks like FLIP. The full codebase and dataset will be available at https://github.com/aplesner/FLIP-Reasoning-Challenge.

Paperid: 1145, https://arxiv.org/pdf/2504.11669.pdf GitHub

Abstract:
Recent advances in Source-Free Unsupervised Video Domain Adaptation (SFUVDA) leverage vision-language models to enhance pseudo-label generation. However, challenges such as noisy pseudo-labels and over-confident predictions limit their effectiveness in adapting well across domains. We propose Co-STAR, a novel framework that integrates curriculum learning with collaborative self-training between a source-trained teacher and a contrastive vision-language model (CLIP). Our curriculum learning approach employs a reliability-based weight function that measures bidirectional prediction alignment between the teacher and CLIP, balancing between confident and uncertain predictions. This function preserves uncertainty for difficult samples, while prioritizing reliable pseudo-labels when the predictions from both models closely align. To further improve adaptation, we propose Adaptive Curriculum Regularization, which modifies the learning priority of samples in a probabilistic, adaptive manner based on their confidence scores and prediction stability, mitigating overfitting to noisy and over-confident samples. Extensive experiments across multiple video domain adaptation benchmarks demonstrate that Co-STAR consistently outperforms state-of-the-art SFUVDA methods. Code is available at: https://github.com/Plrbear/Co-Star

Paperid: 1146, https://arxiv.org/pdf/2504.11368.pdf GitHub

Abstract:
Medical image segmentation remains challenging due to the high cost of pixel-level annotations for training. In the context of weak supervision, clinician gaze data captures regions of diagnostic interest; however, its sparsity limits its use for segmentation. In contrast, vision-language models (VLMs) provide semantic context through textual descriptions but lack the explanation precision required. Recognizing that neither source alone suffices, we propose a teacher-student framework that integrates both gaze and language supervision, leveraging their complementary strengths. Our key insight is that gaze data indicates where clinicians focus during diagnosis, while VLMs explain why those regions are significant. To implement this, the teacher model first learns from gaze points enhanced by VLM-generated descriptions of lesion morphology, establishing a foundation for guiding the student model. The teacher then directs the student through three strategies: (1) Multi-scale feature alignment to fuse visual cues with textual semantics; (2) Confidence-weighted consistency constraints to focus on reliable predictions; (3) Adaptive masking to limit error propagation in uncertain areas. Experiments on the Kvasir-SEG, NCI-ISBI, and ISIC datasets show that our method achieves Dice scores of 80.78%, 80.53%, and 84.22%, respectively-improving 3-5% over gaze baselines without increasing the annotation burden. By preserving correlations among predictions, gaze data, and lesion descriptions, our framework also maintains clinical interpretability. This work illustrates how integrating human visual attention with AI-generated semantic context can effectively overcome the limitations of individual weak supervision signals, thereby advancing the development of deployable, annotation-efficient medical AI systems. Code is available at: https://github.com/jingkunchen/FGI.git.

Paperid: 1147, https://arxiv.org/pdf/2504.11257.pdf GitHub

Abstract:
Recent advancements in Large Vision-Language Models are accelerating the development of Graphical User Interface (GUI) agents that utilize human-like vision perception capabilities to enhance productivity on digital devices. Compared to approaches predicated on GUI metadata, which are platform-dependent and vulnerable to implementation variations, vision-based approaches offer broader applicability. In this vision-based paradigm, the GUI instruction grounding, which maps user instruction to the location of corresponding element on the given screenshot, remains a critical challenge, particularly due to limited public training dataset and resource-intensive manual instruction data annotation. In this paper, we delve into unexplored challenges in this task including element-to-screen ratio, unbalanced element type, and implicit instruction. To address these challenges, we introduce a large-scale data synthesis pipeline UI-E2I-Synth for generating varying complex instruction datasets using GPT-4o instead of human annotators. Furthermore, we propose a new GUI instruction grounding benchmark UI-I2E-Bench, which is designed to address the limitations of existing benchmarks by incorporating diverse annotation aspects. Our model, trained on the synthesized data, achieves superior performance in GUI instruction grounding, demonstrating the advancements of proposed data synthesis pipeline. The proposed benchmark, accompanied by extensive analyses, provides practical insights for future research in GUI grounding. We will release corresponding artifacts at https://microsoft.github.io/FIVE-UI-Evol/ .

Paperid: 1148, https://arxiv.org/pdf/2504.11195.pdf GitHub

Abstract:
Vision-language models (VLMs), such as CLIP, have gained significant popularity as foundation models, with numerous fine-tuning methods developed to enhance performance on downstream tasks. However, due to their inherent vulnerability and the common practice of selecting from a limited set of open-source models, VLMs suffer from a higher risk of adversarial attacks than traditional vision models. Existing defense techniques typically rely on adversarial fine-tuning during training, which requires labeled data and lacks of flexibility for downstream tasks. To address these limitations, we propose robust test-time prompt tuning (R-TPT), which mitigates the impact of adversarial attacks during the inference stage. We first reformulate the classic marginal entropy objective by eliminating the term that introduces conflicts under adversarial conditions, retaining only the pointwise entropy minimization. Furthermore, we introduce a plug-and-play reliability-based weighted ensembling strategy, which aggregates useful information from reliable augmented views to strengthen the defense. R-TPT enhances defense against adversarial attacks without requiring labeled training data while offering high flexibility for inference tasks. Extensive experiments on widely used benchmarks with various attacks demonstrate the effectiveness of R-TPT. The code is available in https://github.com/TomSheng21/R-TPT.

Paperid: 1149, https://arxiv.org/pdf/2504.11038.pdf GitHub

Abstract:
In typical multimodal tasks, such as Visual Question Answering (VQA), adversarial attacks targeting a specific image and question can lead large vision-language models (LVLMs) to provide incorrect answers. However, it is common for a single image to be associated with multiple questions, and LVLMs may still answer other questions correctly even for an adversarial image attacked by a specific question. To address this, we introduce the query-agnostic visual attack (QAVA), which aims to create robust adversarial examples that generate incorrect responses to unspecified and unknown questions. Compared to traditional adversarial attacks focused on specific images and questions, QAVA significantly enhances the effectiveness and efficiency of attacks on images when the question is unknown, achieving performance comparable to attacks on known target questions. Our research broadens the scope of visual adversarial attacks on LVLMs in practical settings, uncovering previously overlooked vulnerabilities, particularly in the context of visual adversarial threats. The code is available at https://github.com/btzyd/qava.

Paperid: 1150, https://arxiv.org/pdf/2504.10514.pdf GitHub

Abstract:
Color plays an important role in human perception and usually provides critical clues in visual reasoning. However, it is unclear whether and how vision-language models (VLMs) can perceive, understand, and leverage color as humans. This paper introduces ColorBench, an innovative benchmark meticulously crafted to assess the capabilities of VLMs in color understanding, including color perception, reasoning, and robustness. By curating a suite of diverse test scenarios, with grounding in real applications, ColorBench evaluates how these models perceive colors, infer meanings from color-based cues, and maintain consistent performance under varying color transformations. Through an extensive evaluation of 32 VLMs with varying language models and vision encoders, our paper reveals some undiscovered findings: (i) The scaling law (larger models are better) still holds on ColorBench, while the language model plays a more important role than the vision encoder. (ii) However, the performance gaps across models are relatively small, indicating that color understanding has been largely neglected by existing VLMs. (iii) CoT reasoning improves color understanding accuracies and robustness, though they are vision-centric tasks. (iv) Color clues are indeed leveraged by VLMs on ColorBench but they can also mislead models in some tasks. These findings highlight the critical limitations of current VLMs and underscore the need to enhance color comprehension. Our ColorBenchcan serve as a foundational tool for advancing the study of human-level color understanding of multimodal AI.

Paperid: 1151, https://arxiv.org/pdf/2504.10465.pdf GitHub

Abstract:
Multimodal Large Language Models (MLLMs) achieve remarkable performance for fine-grained pixel-level understanding tasks. However, all the works rely heavily on extra components, such as vision encoder (CLIP), segmentation experts, leading to high system complexity and limiting model scaling. In this work, our goal is to explore a highly simplified MLLM without introducing extra components. Our work is motivated by the recent works on Single trAnsformer as a unified vIsion-Language Model (SAIL) design, where these works jointly learn vision tokens and text tokens in transformers. We present Pixel-SAIL, a single transformer for pixel-wise MLLM tasks. In particular, we present three technical improvements on the plain baseline. First, we design a learnable upsampling module to refine visual token features. Secondly, we propose a novel visual prompt injection strategy to enable the single transformer to understand visual prompt inputs and benefit from the early fusion of visual prompt embeddings and vision tokens. Thirdly, we introduce a vision expert distillation strategy to efficiently enhance the single transformer's fine-grained feature extraction capability. In addition, we have collected a comprehensive pixel understanding benchmark (PerBench), using a manual check. It includes three tasks: detailed object description, visual prompt-based question answering, and visual-text referring segmentation. Extensive experiments on four referring segmentation benchmarks, one visual prompt benchmark, and our PerBench show that our Pixel-SAIL achieves comparable or even better results with a much simpler pipeline. Code and model will be released at https://github.com/magic-research/Sa2VA.

Paperid: 1152, https://arxiv.org/pdf/2504.10462.pdf GitHub

Abstract:
This paper introduces SAIL, a single transformer unified multimodal large language model (MLLM) that integrates raw pixel encoding and language decoding within a singular architecture. Unlike existing modular MLLMs, which rely on a pre-trained vision transformer (ViT), SAIL eliminates the need for a separate vision encoder, presenting a more minimalist architecture design. Instead of introducing novel architectural components, SAIL adapts mix-attention mechanisms and multimodal positional encodings to better align with the distinct characteristics of visual and textual modalities. We systematically compare SAIL's properties-including scalability, cross-modal information flow patterns, and visual representation capabilities-with those of modular MLLMs. By scaling both training data and model size, SAIL achieves performance comparable to modular MLLMs. Notably, the removal of pretrained ViT components enhances SAIL's scalability and results in significantly different cross-modal information flow patterns. Moreover, SAIL demonstrates strong visual representation capabilities, achieving results on par with ViT-22B in vision tasks such as semantic segmentation. Code and models are available at https://github.com/bytedance/SAIL.

Paperid: 1153, https://arxiv.org/pdf/2504.10127.pdf GitHub

Abstract:
Graphical User Interface (GUI) agents offer cross-platform solutions for automating complex digital tasks, with significant potential to transform productivity workflows. However, their performance is often constrained by the scarcity of high-quality trajectory data. To address this limitation, we propose training Vision Language Models (VLMs) on data-rich, reasoning-intensive tasks during a dedicated mid-training stage, and then examine how incorporating these tasks facilitates generalization to GUI planning scenarios. Specifically, we explore a range of tasks with readily available instruction-tuning data, including GUI perception, multimodal reasoning, and textual reasoning. Through extensive experiments across 11 mid-training tasks, we demonstrate that: (1) Task generalization proves highly effective, yielding substantial improvements across most settings. For instance, multimodal mathematical reasoning enhances performance on AndroidWorld by an absolute 6.3%. Remarkably, text-only mathematical data significantly boosts GUI web agent performance, achieving a 5.6% improvement on WebArena and 5.4% improvement on AndroidWorld, underscoring notable cross-modal generalization from text-based to visual domains; (2) Contrary to prior assumptions, GUI perception data - previously considered closely aligned with GUI agent tasks and widely utilized for training - has a comparatively limited impact on final performance; (3) Building on these insights, we identify the most effective mid-training tasks and curate optimized mixture datasets, resulting in absolute performance gains of 8.0% on WebArena and 12.2% on AndroidWorld. Our work provides valuable insights into cross-domain knowledge transfer for GUI agents and offers a practical approach to addressing data scarcity challenges in this emerging field. The code, data and models will be available at https://github.com/hkust-nlp/GUIMid.

Paperid: 1154, https://arxiv.org/pdf/2504.10044.pdf GitHub

Abstract:
Anime video generation faces significant challenges due to the scarcity of anime data and unusual motion patterns, leading to issues such as motion distortion and flickering artifacts, which result in misalignment with human preferences. Existing reward models, designed primarily for real-world videos, fail to capture the unique appearance and consistency requirements of anime. In this work, we propose a pipeline to enhance anime video generation by leveraging human feedback for better alignment. Specifically, we construct the first multi-dimensional reward dataset for anime videos, comprising 30k human-annotated samples that incorporating human preferences for both visual appearance and visual consistency. Based on this, we develop AnimeReward, a powerful reward model that employs specialized vision-language models for different evaluation dimensions to guide preference alignment. Furthermore, we introduce Gap-Aware Preference Optimization (GAPO), a novel training method that explicitly incorporates preference gaps into the optimization process, enhancing alignment performance and efficiency. Extensive experiment results show that AnimeReward outperforms existing reward models, and the inclusion of GAPO leads to superior alignment in both quantitative benchmarks and human evaluations, demonstrating the effectiveness of our pipeline in enhancing anime video quality. Our code and dataset are publicly available at https://github.com/bilibili/Index-anisora.

Paperid: 1155, https://arxiv.org/pdf/2504.09925.pdf GitHub

Abstract:
We introduce FUSION, a family of multimodal large language models (MLLMs) with a fully vision-language alignment and integration paradigm. Unlike existing methods that primarily rely on late-stage modality interaction during LLM decoding, our approach achieves deep, dynamic integration throughout the entire processing pipeline. To this end, we propose Text-Guided Unified Vision Encoding, incorporating textual information in vision encoding to achieve pixel-level integration. We further design Context-Aware Recursive Alignment Decoding that recursively aggregates visual features conditioned on textual context during decoding, enabling fine-grained, question-level semantic integration. To guide feature mapping and mitigate modality discrepancies, we develop Dual-Supervised Semantic Mapping Loss. Additionally, we construct a Synthesized Language-Driven Question-Answer (QA) dataset through a new data synthesis method, prioritizing high-quality QA pairs to optimize text-guided feature integration. Building on these foundations, we train FUSION at two scales-3B, 8B-and demonstrate that our full-modality integration approach significantly outperforms existing methods with only 630 vision tokens. Notably, FUSION 3B surpasses Cambrian-1 8B and Florence-VL 8B on most benchmarks. FUSION 3B continues to outperform Cambrian-1 8B even when limited to 300 vision tokens. Our ablation studies show that FUSION outperforms LLaVA-NeXT on over half of the benchmarks under same configuration without dynamic resolution, highlighting the effectiveness of our approach. We release our code, model weights, and dataset. https://github.com/starriver030515/FUSION

Paperid: 1156, https://arxiv.org/pdf/2504.09724.pdf GitHub

Abstract:
Vision-language models (VLMs) integrate visual and textual information, enabling a wide range of applications such as image captioning and visual question answering, making them crucial for modern AI systems. However, their high computational demands pose challenges for real-time applications. This has led to a growing focus on developing efficient vision language models. In this survey, we review key techniques for optimizing VLMs on edge and resource-constrained devices. We also explore compact VLM architectures, frameworks and provide detailed insights into the performance-memory trade-offs of efficient VLMs. Furthermore, we establish a GitHub repository at https://github.com/MPSCUMBC/Efficient-Vision-Language-Models-A-Survey to compile all surveyed papers, which we will actively update. Our objective is to foster deeper research in this area.

Paperid: 1157, https://arxiv.org/pdf/2504.09518.pdf GitHub

Abstract:
3D captioning, which aims to describe the content of 3D scenes in natural language, remains highly challenging due to the inherent sparsity of point clouds and weak cross-modal alignment in existing methods. To address these challenges, we propose 3D CoCa, a novel unified framework that seamlessly combines contrastive vision-language learning with 3D caption generation in a single architecture. Our approach leverages a frozen CLIP vision-language backbone to provide rich semantic priors, a spatially-aware 3D scene encoder to capture geometric context, and a multi-modal decoder to generate descriptive captions. Unlike prior two-stage methods that rely on explicit object proposals, 3D CoCa jointly optimizes contrastive and captioning objectives in a shared feature space, eliminating the need for external detectors or handcrafted proposals. This joint training paradigm yields stronger spatial reasoning and richer semantic grounding by aligning 3D and textual representations. Extensive experiments on the ScanRefer and Nr3D benchmarks demonstrate that 3D CoCa significantly outperforms current state-of-the-arts by 10.2% and 5.76% in CIDEr at 0.5IoU, respectively. Code will be available at https://github.com/AIGeeksGroup/3DCoCa.

Paperid: 1158, https://arxiv.org/pdf/2504.09480.pdf GitHub

Abstract:
Vision-Language Model (VLM) have gained widespread adoption in Open-Vocabulary (OV) object detection and segmentation tasks. Despite they have shown promise on OV-related tasks, their effectiveness in conventional vision tasks has thus far been unevaluated. In this work, we present the systematic review of VLM-based detection and segmentation, view VLM as the foundational model and conduct comprehensive evaluations across multiple downstream tasks for the first time: 1) The evaluation spans eight detection scenarios (closed-set detection, domain adaptation, crowded objects, etc.) and eight segmentation scenarios (few-shot, open-world, small object, etc.), revealing distinct performance advantages and limitations of various VLM architectures across tasks. 2) As for detection tasks, we evaluate VLMs under three finetuning granularities: \textit{zero prediction}, \textit{visual fine-tuning}, and \textit{text prompt}, and further analyze how different finetuning strategies impact performance under varied task. 3) Based on empirical findings, we provide in-depth analysis of the correlations between task characteristics, model architectures, and training methodologies, offering insights for future VLM design. 4) We believe that this work shall be valuable to the pattern recognition experts working in the fields of computer vision, multimodal learning, and vision foundation models by introducing them to the problem, and familiarizing them with the current status of the progress while providing promising directions for future research. A project associated with this review and evaluation has been created at https://github.com/better-chao/perceptual_abilities_evaluation.

Paperid: 1159, https://arxiv.org/pdf/2504.08066.pdf GitHub

Abstract:
AI is increasingly playing a pivotal role in transforming how scientific discoveries are made. We introduce The AI Scientist-v2, an end-to-end agentic system capable of producing the first entirely AI generated peer-review-accepted workshop paper. This system iteratively formulates scientific hypotheses, designs and executes experiments, analyzes and visualizes data, and autonomously authors scientific manuscripts. Compared to its predecessor (v1, Lu et al., 2024 arXiv:2408.06292), The AI Scientist-v2 eliminates the reliance on human-authored code templates, generalizes effectively across diverse machine learning domains, and leverages a novel progressive agentic tree-search methodology managed by a dedicated experiment manager agent. Additionally, we enhance the AI reviewer component by integrating a Vision-Language Model (VLM) feedback loop for iterative refinement of content and aesthetics of the figures. We evaluated The AI Scientist-v2 by submitting three fully autonomous manuscripts to a peer-reviewed ICLR workshop. Notably, one manuscript achieved high enough scores to exceed the average human acceptance threshold, marking the first instance of a fully AI-generated paper successfully navigating a peer review. This accomplishment highlights the growing capability of AI in conducting all aspects of scientific research. We anticipate that further advancements in autonomous scientific discovery technologies will profoundly impact human knowledge generation, enabling unprecedented scalability in research productivity and significantly accelerating scientific breakthroughs, greatly benefiting society at large. We have open-sourced the code at https://github.com/SakanaAI/AI-Scientist-v2 to foster the future development of this transformative technology. We also discuss the role of AI in science, including AI safety.

Paperid: 1160, https://arxiv.org/pdf/2504.07615.pdf GitHub

Abstract:
Recently DeepSeek R1 has shown that reinforcement learning (RL) can substantially improve the reasoning capabilities of Large Language Models (LLMs) through a simple yet effective design. The core of R1 lies in its rule-based reward formulation, which leverages tasks with deterministic ground-truth answers to enable precise and stable reward computation. In the visual domain, we similarly observe that a wide range of visual understanding tasks are inherently equipped with well-defined ground-truth annotations. This property makes them naturally compatible with rule-based reward mechanisms. Motivated by this observation, we investigate the extension of R1-style reinforcement learning to Vision-Language Models (VLMs), aiming to enhance their visual reasoning capabilities. To this end, we develop VLM-R1, a dedicated framework designed to harness RL for improving VLMs' performance on general vision-language tasks. Using this framework, we further explore the feasibility of applying RL to visual domain. Experimental results indicate that the RL-based model not only delivers competitive performance on visual understanding tasks but also surpasses Supervised Fine-Tuning (SFT) in generalization ability. Furthermore, we conduct comprehensive ablation studies that uncover a series of noteworthy insights, including the presence of reward hacking in object detection, the emergence of the "OD aha moment", the impact of training data quality, and the scaling behavior of RL across different model sizes. Through these analyses, we aim to deepen the understanding of how reinforcement learning enhances the capabilities of vision-language models, and we hope our findings and open-source contributions will support continued progress in the vision-language RL community. Our code and model are available at https://github.com/om-ai-lab/VLM-R1

Paperid: 1161, https://arxiv.org/pdf/2504.07521.pdf GitHub GitHub

Abstract:
Most existing emotion analysis emphasizes which emotion arises (e.g., happy, sad, angry) but neglects the deeper why. We propose Emotion Interpretation (EI), focusing on causal factors-whether explicit (e.g., observable objects, interpersonal interactions) or implicit (e.g., cultural context, off-screen events)-that drive emotional responses. Unlike traditional emotion recognition, EI tasks require reasoning about triggers instead of mere labeling. To facilitate EI research, we present EIBench, a large-scale benchmark encompassing 1,615 basic EI samples and 50 complex EI samples featuring multifaceted emotions. Each instance demands rationale-based explanations rather than straightforward categorization. We further propose a Coarse-to-Fine Self-Ask (CFSA) annotation pipeline, which guides Vision-Language Models (VLLMs) through iterative question-answer rounds to yield high-quality labels at scale. Extensive evaluations on open-source and proprietary large language models under four experimental settings reveal consistent performance gaps-especially for more intricate scenarios-underscoring EI's potential to enrich empathetic, context-aware AI applications. Our benchmark and methods are publicly available at: https://github.com/Lum1104/EIBench, offering a foundation for advanced multimodal causal analysis and next-generation affective computing.

Paperid: 1162, https://arxiv.org/pdf/2504.07491.pdf GitHub

Authors:Kimi Team, Angang Du, Bohong Yin, Bowei Xing, Bowen Qu, Bowen Wang, Cheng Chen, Chenlin Zhang, Chenzhuang Du, Chu Wei, Congcong Wang, Dehao Zhang, Dikang Du, Dongliang Wang, Enming Yuan, Enzhe Lu, Fang Li, Flood Sung, Guangda Wei, Guokun Lai, Han Zhu, Hao Ding, Hao Hu, Hao Yang, Hao Zhang, Haoning Wu, Haotian Yao, Haoyu Lu, Heng Wang, Hongcheng Gao, Huabin Zheng, Jiaming Li, Jianlin Su, Jianzhou Wang, Jiaqi Deng, Jiezhong Qiu, Jin Xie, Jinhong Wang, Jingyuan Liu, Junjie Yan, Kun Ouyang, Liang Chen, Lin Sui, Longhui Yu, Mengfan Dong, Mengnan Dong, Nuo Xu, Pengyu Cheng, Qizheng Gu, Runjie Zhou, Shaowei Liu, Sihan Cao, Tao Yu, Tianhui Song, Tongtong Bai, Wei Song, Weiran He, Weixiao Huang, Weixin Xu, Xiaokun Yuan, Xingcheng Yao, Xingzhe Wu, Xinhao Li, Xinxing Zu, Xinyu Zhou, Xinyuan Wang, Y. Charles, Yan Zhong, Yang Li, Yangyang Hu, Yanru Chen, Yejie Wang, Yibo Liu, Yibo Miao, Yidao Qin, Yimin Chen, Yiping Bao, Yiqin Wang, Yongsheng Kang, Yuanxin Liu, Yuhao Dong, Yulun Du, Yuxin Wu, Yuzhi Wang, Yuzi Yan, Zaida Zhou, Zhaowei Li, Zhejun Jiang, Zheng Zhang, Zhilin Yang, Zhiqi Huang, Zihao Huang, Zijia Zhao, Ziwei Chen, Zongyu Lin

Abstract:
We present Kimi-VL, an efficient open-source Mixture-of-Experts (MoE) vision-language model (VLM) that offers advanced multimodal reasoning, long-context understanding, and strong agent capabilities - all while activating only 2.8B parameters in its language decoder (Kimi-VL-A3B). Kimi-VL demonstrates strong performance across challenging domains: as a general-purpose VLM, Kimi-VL excels in multi-turn agent tasks (e.g., OSWorld), matching flagship models. Furthermore, it exhibits remarkable capabilities across diverse challenging vision language tasks, including college-level image and video comprehension, OCR, mathematical reasoning, and multi-image understanding. In comparative evaluations, it effectively competes with cutting-edge efficient VLMs such as GPT-4o-mini, Qwen2.5-VL-7B, and Gemma-3-12B-IT, while surpassing GPT-4o in several key domains. Kimi-VL also advances in processing long contexts and perceiving clearly. With a 128K extended context window, Kimi-VL can process diverse long inputs, achieving impressive scores of 64.5 on LongVideoBench and 35.1 on MMLongBench-Doc. Its native-resolution vision encoder, MoonViT, further allows it to see and understand ultra-high-resolution visual inputs, achieving 83.2 on InfoVQA and 34.5 on ScreenSpot-Pro, while maintaining lower computational cost for common tasks. Building upon Kimi-VL, we introduce an advanced long-thinking variant: Kimi-VL-Thinking-2506. Developed through long chain-of-thought (CoT) supervised fine-tuning (SFT) and reinforcement learning (RL), the latest model exhibits strong long-horizon reasoning capabilities (64.0 on MMMU, 46.3 on MMMU-Pro, 56.9 on MathVision, 80.1 on MathVista, 65.2 on VideoMMMU) while obtaining robust general abilities. Code and models are publicly accessible at https://github.com/MoonshotAI/Kimi-VL.

Paperid: 1163, https://arxiv.org/pdf/2504.06925.pdf GitHub

Abstract:
Automatic dietary assessment based on food images remains a challenge, requiring precise food detection, segmentation, and classification. Vision-Language Models (VLMs) offer new possibilities by integrating visual and textual reasoning. In this study, we evaluate six state-of-the-art VLMs (ChatGPT, Gemini, Claude, Moondream, DeepSeek, and LLaVA), analyzing their capabilities in food recognition at different levels. For the experimental framework, we introduce the FoodNExTDB, a unique food image database that contains 9,263 expert-labeled images across 10 categories (e.g., "protein source"), 62 subcategories (e.g., "poultry"), and 9 cooking styles (e.g., "grilled"). In total, FoodNExTDB includes 50k nutritional labels generated by seven experts who manually annotated all images in the database. Also, we propose a novel evaluation metric, Expert-Weighted Recall (EWR), that accounts for the inter-annotator variability. Results show that closed-source models outperform open-source ones, achieving over 90% EWR in recognizing food products in images containing a single product. Despite their potential, current VLMs face challenges in fine-grained food recognition, particularly in distinguishing subtle differences in cooking styles and visually similar food items, which limits their reliability for automatic dietary assessment. The FoodNExTDB database is publicly available at https://github.com/AI4Food/FoodNExtDB.

Paperid: 1164, https://arxiv.org/pdf/2504.06863.pdf GitHub

Abstract:
Moving object segmentation plays a vital role in understanding dynamic visual environments. While existing methods rely on multi-frame image sequences to identify moving objects, single-image MOS is critical for applications like motion intention prediction and handling camera frame drops. However, segmenting moving objects from a single image remains challenging for existing methods due to the absence of temporal cues. To address this gap, we propose MovSAM, the first framework for single-image moving object segmentation. MovSAM leverages a Multimodal Large Language Model (MLLM) enhanced with Chain-of-Thought (CoT) prompting to search the moving object and generate text prompts based on deep thinking for segmentation. These prompts are cross-fused with visual features from the Segment Anything Model (SAM) and a Vision-Language Model (VLM), enabling logic-driven moving object segmentation. The segmentation results then undergo a deep thinking refinement loop, allowing MovSAM to iteratively improve its understanding of the scene context and inter-object relationships with logical reasoning. This innovative approach enables MovSAM to segment moving objects in single images by considering scene understanding. We implement MovSAM in the real world to validate its practical application and effectiveness for autonomous driving scenarios where the multi-frame methods fail. Furthermore, despite the inherent advantage of multi-frame methods in utilizing temporal information, MovSAM achieves state-of-the-art performance across public MOS benchmarks, reaching 92.5\% on J\&F. Our implementation will be available at https://github.com/IRMVLab/MovSAM.

Paperid: 1165, https://arxiv.org/pdf/2504.06397.pdf GitHub

Abstract:
Human pose and shape (HPS) estimation presents challenges in diverse scenarios such as crowded scenes, person-person interactions, and single-view reconstruction. Existing approaches lack mechanisms to incorporate auxiliary "side information" that could enhance reconstruction accuracy in such challenging scenarios. Furthermore, the most accurate methods rely on cropped person detections and cannot exploit scene context while methods that process the whole image often fail to detect people and are less accurate than methods that use crops. While recent language-based methods explore HPS reasoning through large language or vision-language models, their metric accuracy is well below the state of the art. In contrast, we present PromptHMR, a transformer-based promptable method that reformulates HPS estimation through spatial and semantic prompts. Our method processes full images to maintain scene context and accepts multiple input modalities: spatial prompts like bounding boxes and masks, and semantic prompts like language descriptions or interaction labels. PromptHMR demonstrates robust performance across challenging scenarios: estimating people from bounding boxes as small as faces in crowded scenes, improving body shape estimation through language descriptions, modeling person-person interactions, and producing temporally coherent motions in videos. Experiments on benchmarks show that PromptHMR achieves state-of-the-art performance while offering flexible prompt-based control over the HPS estimation process.

Paperid: 1166, https://arxiv.org/pdf/2504.06389.pdf GitHub GitHub

Abstract:
Domain Adaptation (DA) and Semi-supervised Learning (SSL) converge in Semi-supervised Domain Adaptation (SSDA), where the objective is to transfer knowledge from a source domain to a target domain using a combination of limited labeled target samples and abundant unlabeled target data. Although intuitive, a simple amalgamation of DA and SSL is suboptimal in semantic segmentation due to two major reasons: (1) previous methods, while able to learn good segmentation boundaries, are prone to confuse classes with similar visual appearance due to limited supervision; and (2) skewed and imbalanced training data distribution preferring source representation learning whereas impeding from exploring limited information about tailed classes. Language guidance can serve as a pivotal semantic bridge, facilitating robust class discrimination and mitigating visual ambiguities by leveraging the rich semantic relationships encoded in pre-trained language models to enhance feature representations across domains. Therefore, we propose the first language-guided SSDA setting for semantic segmentation in this work. Specifically, we harness the semantic generalization capabilities inherent in vision-language models (VLMs) to establish a synergistic framework within the SSDA paradigm. To address the inherent class-imbalance challenges in long-tailed distributions, we introduce class-balanced segmentation loss formulations that effectively regularize the learning process. Through extensive experimentation across diverse domain adaptation scenarios, our approach demonstrates substantial performance improvements over contemporary state-of-the-art (SoTA) methodologies. Code is available: \href{https://github.com/hritam-98/SemiDAViL}{GitHub}.

Paperid: 1167, https://arxiv.org/pdf/2504.06010.pdf GitHub

Abstract:
Multimodal misinformation, such as miscaptioned images, where captions misrepresent an image's origin, context, or meaning, poses a growing challenge in the digital age. To support fact-checkers, researchers have focused on developing datasets and methods for multimodal misinformation detection (MMD). Due to the scarcity of large-scale annotated MMD datasets, recent approaches rely on synthetic training data created via out-of-context pairings or named entity manipulations (e.g., altering names, dates, or locations). However, these often yield simplistic examples that lack real-world complexity, limiting model robustness. Meanwhile, Large Vision-Language Models (LVLMs) remain underexplored for generating diverse and realistic synthetic data for MMD. To address, we introduce "Miscaption This!", a collection of LVLM-generated miscaptioned image datasets. Additionally, we introduce "Latent Multimodal Reconstruction" (LAMAR), a network trained to reconstruct the embeddings of truthful captions, providing a strong auxiliary signal to guide detection. We explore various training strategies (end-to-end vs. large-scale pre-training) and integration mechanisms (direct, mask, gate, and attention). Extensive experiments show that models trained on "MisCaption This!" generalize better to real-world misinformation while LAMAR achieves new state-of-the-art on both NewsCLIPpings and VERITE benchmarks; highlighting the value of LVLM-generated data and reconstruction-based networks for advancing MMD. Our code is available at https://github.com/stevejpapad/miscaptioned-image-reconstruction

Paperid: 1168, https://arxiv.org/pdf/2504.05506.pdf GitHub

Abstract:
Charts are ubiquitous, as people often use them to analyze data, answer questions, and discover critical insights. However, performing complex analytical tasks with charts requires significant perceptual and cognitive effort. Chart Question Answering (CQA) systems automate this process by enabling models to interpret and reason with visual representations of data. However, existing benchmarks like ChartQA lack real-world diversity and have recently shown performance saturation with modern large vision-language models (LVLMs). To address these limitations, we introduce ChartQAPro, a new benchmark that includes 1,341 charts from 157 diverse sources, spanning various chart types, including infographics and dashboards, and featuring 1,948 questions in various types, such as multiple-choice, conversational, hypothetical, and unanswerable questions, to better reflect real-world challenges. Our evaluations with 21 models show a substantial performance drop for LVLMs on ChartQAPro; e.g., Claude Sonnet 3.5 scores 90.5% on ChartQA but only 55.81% on ChartQAPro, underscoring the complexity of chart reasoning. We complement our findings with detailed error analyses and ablation studies, identifying key challenges and opportunities for advancing LVLMs in chart understanding and reasoning. We release ChartQAPro at https://github.com/vis-nlp/ChartQAPro.

Paperid: 1169, https://arxiv.org/pdf/2504.05227.pdf GitHub

Abstract:
Vision-language pre-training has recently gained popularity as it allows learning rich feature representations using large-scale data sources. This paradigm has quickly made its way into the medical image analysis community. In particular, there is an impressive amount of recent literature developing vision-language models for radiology. However, the available medical datasets with image-text supervision are scarce, and medical concepts are fine-grained, involving expert knowledge that existing vision-language models struggle to encode. In this paper, we propose to take a prudent step back from the literature and revisit supervised, unimodal pre-training, using fine-grained labels instead. We conduct an extensive comparison demonstrating that unimodal pre-training is highly competitive and better suited to integrating heterogeneous data sources. Our results also question the potential of recent vision-language models for open-vocabulary generalization, which have been evaluated using optimistic experimental settings. Finally, we study novel alternatives to better integrate fine-grained labels and noisy text supervision.

Paperid: 1170, https://arxiv.org/pdf/2504.05225.pdf GitHub

Abstract:
Model Predictive Control (MPC) is a widely adopted control paradigm that leverages predictive models to estimate future system states and optimize control inputs accordingly. However, while MPC excels in planning and control, it lacks the capability for environmental perception, leading to failures in complex and unstructured scenarios. To address this limitation, we introduce Vision-Language Model Predictive Control (VLMPC), a robotic manipulation planning framework that integrates the perception power of vision-language models (VLMs) with MPC. VLMPC utilizes a conditional action sampling module that takes a goal image or language instruction as input and leverages VLM to generate candidate action sequences. These candidates are fed into a video prediction model that simulates future frames based on the actions. In addition, we propose an enhanced variant, Traj-VLMPC, which replaces video prediction with motion trajectory generation to reduce computational complexity while maintaining accuracy. Traj-VLMPC estimates motion dynamics conditioned on the candidate actions, offering a more efficient alternative for long-horizon tasks and real-time applications. Both VLMPC and Traj-VLMPC select the optimal action sequence using a VLM-based hierarchical cost function that captures both pixel-level and knowledge-level consistency between the current observation and the task input. We demonstrate that both approaches outperform existing state-of-the-art methods on public benchmarks and achieve excellent performance in various real-world robotic manipulation tasks. Code is available at https://github.com/PPjmchen/VLMPC.

Paperid: 1171, https://arxiv.org/pdf/2504.04740.pdf GitHub

Abstract:
Compositionality, or correctly recognizing scenes as compositions of atomic visual concepts, remains difficult for multimodal large language models (MLLMs). Even state of the art MLLMs such as GPT-4o can make mistakes in distinguishing compositions like "dog chasing cat" vs "cat chasing dog". While on Winoground, a benchmark for measuring such reasoning, MLLMs have made significant progress, they are still far from a human's performance. We show that compositional reasoning in these models can be improved by elucidating such concepts via data, where a model is trained to prefer the correct caption for an image over a close but incorrect one. We introduce SCRAMBLe: Synthetic Compositional Reasoning Augmentation of MLLMs with Binary preference Learning, an approach for preference tuning open-weight MLLMs on synthetic preference data generated in a fully automated manner from existing image-caption data. SCRAMBLe holistically improves these MLLMs' compositional reasoning capabilities which we can see through significant improvements across multiple vision language compositionality benchmarks, as well as smaller but significant improvements on general question answering tasks. As a sneak peek, SCRAMBLe tuned Molmo-7B model improves on Winoground from 49.5% to 54.8% (best reported to date), while improving by ~1% on more general visual question answering tasks. Code for SCRAMBLe along with tuned models and our synthetic training dataset is available at https://github.com/samarth4149/SCRAMBLe.

Paperid: 1172, https://arxiv.org/pdf/2504.04517.pdf GitHub

Abstract:
Foundation models pretrained on extensive datasets, such as GroundingDINO and LAE-DINO, have performed remarkably in the cross-domain few-shot object detection (CD-FSOD) task. Through rigorous few-shot training, we found that the integration of image-based data augmentation techniques and grid-based sub-domain search strategy significantly enhances the performance of these foundation models. Building upon GroundingDINO, we employed several widely used image augmentation methods and established optimization objectives to effectively navigate the expansive domain space in search of optimal sub-domains. This approach facilitates efficient few-shot object detection and introduces an approach to solving the CD-FSOD problem by efficiently searching for the optimal parameter configuration from the foundation model. Our findings substantially advance the practical deployment of vision-language models in data-scarce environments, offering critical insights into optimizing their cross-domain generalization capabilities without labor-intensive retraining. Code is available at https://github.com/jaychempan/ETS.

Paperid: 1173, https://arxiv.org/pdf/2504.04323.pdf GitHub

Abstract:
Medical image analysis is essential in modern healthcare. Deep learning has redirected research focus toward complex medical multimodal tasks, including report generation and visual question answering. Traditional task-specific models often fall short in handling these challenges. Large vision-language models (LVLMs) offer new solutions for solving such tasks. In this study, we build on the popular LLaVA framework to systematically explore model architectures and training strategies for both 2D and 3D medical LVLMs. We present extensive empirical findings and practical guidance. To support reproducibility and future research, we release a modular codebase, MedM-VL, and two pre-trained models: MedM-VL-2D for 2D medical image analysis and MedM-VL-CT-Chest for 3D CT-based applications. The code is available at: https://github.com/MSIIP/MedM-VL

Paperid: 1174, https://arxiv.org/pdf/2504.03970.pdf GitHub

Abstract:
We introduce VideoComp, a benchmark and learning framework for advancing video-text compositionality understanding, aimed at improving vision-language models (VLMs) in fine-grained temporal alignment. Unlike existing benchmarks focused on static image-text compositionality or isolated single-event videos, our benchmark targets alignment in continuous multi-event videos. Leveraging video-text datasets with temporally localized event captions (e.g. ActivityNet-Captions, YouCook2), we construct two compositional benchmarks, ActivityNet-Comp and YouCook2-Comp. We create challenging negative samples with subtle temporal disruptions such as reordering, action word replacement, partial captioning, and combined disruptions. These benchmarks comprehensively test models' compositional sensitivity across extended, cohesive video-text sequences. To improve model performance, we propose a hierarchical pairwise preference loss that strengthens alignment with temporally accurate pairs and gradually penalizes increasingly disrupted ones, encouraging fine-grained compositional learning. To mitigate the limited availability of densely annotated video data, we introduce a pretraining strategy that concatenates short video-caption pairs to simulate multi-event sequences. We evaluate video-text foundational models and large multimodal models (LMMs) on our benchmark, identifying both strengths and areas for improvement in compositionality. Overall, our work provides a comprehensive framework for evaluating and enhancing model capabilities in achieving fine-grained, temporally coherent video-text alignment.

Paperid: 1175, https://arxiv.org/pdf/2504.03801.pdf GitHub

Abstract:
Multi-label Recognition (MLR) involves assigning multiple labels to each data instance in an image, offering advantages over single-label classification in complex scenarios. However, it faces the challenge of annotating all relevant categories, often leading to uncertain annotations, such as unseen or incomplete labels. Recent Vision and Language Pre-training (VLP) based methods have made significant progress in tackling zero-shot MLR tasks by leveraging rich vision-language correlations. However, the correlation between multi-label semantics has not been fully explored, and the learned visual features often lack essential semantic information. To overcome these limitations, we introduce a Semantic-guided Representation Learning approach (SigRL) that enables the model to learn effective visual and textual representations, thereby improving the downstream alignment of visual images and categories. Specifically, we first introduce a graph-based multi-label correlation module (GMC) to facilitate information exchange between labels, enriching the semantic representation across the multi-label texts. Next, we propose a Semantic Visual Feature Reconstruction module (SVFR) to enhance the semantic information in the visual representation by integrating the learned textual representation during reconstruction. Finally, we optimize the image-text matching capability of the VLP model using both local and global features to achieve zero-shot MLR. Comprehensive experiments are conducted on several MLR benchmarks, encompassing both zero-shot MLR (with unseen labels) and single positive multi-label learning (with limited labels), demonstrating the superior performance of our approach compared to state-of-the-art methods. The code is available at https://github.com/MVL-Lab/SigRL.

Paperid: 1176, https://arxiv.org/pdf/2504.03748.pdf GitHub

Abstract:
The rapid emergence of Vision-Language Models (VLMs) has significantly advanced multimodal understanding, enabling applications in scene comprehension and visual reasoning. While these models have been primarily evaluated and developed for front-view image understanding, their capabilities in interpreting top-down images have received limited attention, partly due to the scarcity of diverse top-down datasets and the challenges in collecting such data. In contrast, top-down vision provides explicit spatial overviews and improved contextual understanding of scenes, making it particularly valuable for tasks like autonomous navigation, aerial imaging, and spatial planning. In this work, we address this gap by introducing TDBench, a comprehensive benchmark for VLMs in top-down image understanding. TDBench is constructed from public top-down view datasets and high-quality simulated images, including diverse real-world and synthetic scenarios. TDBench consists of visual question-answer pairs across ten evaluation dimensions of image understanding. Moreover, we conduct four case studies that commonly happen in real-world scenarios but are less explored. By revealing the strengths and limitations of existing VLM through evaluation results, we hope TDBench to provide insights for motivating future research. Project homepage: https://github.com/Columbia-ICSL/TDBench

Paperid: 1177, https://arxiv.org/pdf/2504.03724.pdf GitHub

Abstract:
We propose Fuzzy Group Relative Policy Reward (FGRPR), a novel framework that integrates Group Relative Policy Optimization (GRPO) with a fuzzy reward function to enhance learning efficiency. Unlike the conventional binary 0/1 accuracy reward, our fuzzy reward model provides nuanced incentives, encouraging more precise outputs. Experimental results demonstrate that GRPO with a standard 0/1 accuracy reward underperforms compared to supervised fine-tuning (SFT). In contrast, FGRPR, applied to Qwen2.5-VL(3B and 7B), surpasses all baseline models, including GPT4o, LLaMA2(90B), and SFT, across five in-domain datasets. On an out-of-domain dataset, FGRPR achieves performance comparable to SFT but excels when target values are larger, as its fuzzy reward function assigns higher rewards to closer approximations. This approach is broadly applicable to tasks where the precision of the answer is critical. Code and data: https://github.com/yeyimilk/CrowdVLM-R1

Paperid: 1178, https://arxiv.org/pdf/2504.03254.pdf GitHub

Abstract:
Synthetic Aperture Radar (SAR) is a crucial remote sensing technology, enabling all-weather, day-and-night observation with strong surface penetration for precise and continuous environmental monitoring and analysis. However, SAR image interpretation remains challenging due to its complex physical imaging mechanisms and significant visual disparities from human perception. Recently, Vision-Language Models (VLMs) have demonstrated remarkable success in RGB image understanding, offering powerful open-vocabulary interpretation and flexible language interaction. However, their application to SAR images is severely constrained by the absence of SAR-specific knowledge in their training distributions, leading to suboptimal performance. To address this limitation, we introduce SARLANG-1M, a large-scale benchmark tailored for multimodal SAR image understanding, with a primary focus on integrating SAR with textual modality. SARLANG-1M comprises more than 1 million high-quality SAR image-text pairs collected from over 59 cities worldwide. It features hierarchical resolutions (ranging from 0.1 to 25 meters), fine-grained semantic descriptions (including both concise and detailed captions), diverse remote sensing categories (1,696 object types and 16 land cover classes), and multi-task question-answering pairs spanning seven applications and 1,012 question types. Extensive experiments on mainstream VLMs demonstrate that fine-tuning with SARLANG-1M significantly enhances their performance in SAR image interpretation, reaching performance comparable to human experts. The dataset and code will be made publicly available at https://github.com/Jimmyxichen/SARLANG-1M.

Paperid: 1179, https://arxiv.org/pdf/2504.03193.pdf GitHub

Abstract:
Vision Foundation Models (VFMs) and Vision-Language Models (VLMs) have gained traction in Domain Generalized Semantic Segmentation (DGSS) due to their strong generalization capabilities. However, existing DGSS methods often rely exclusively on either VFMs or VLMs, overlooking their complementary strengths. VFMs (e.g., DINOv2) excel at capturing fine-grained features, while VLMs (e.g., CLIP) provide robust text alignment but struggle with coarse granularity. Despite their complementary strengths, effectively integrating VFMs and VLMs with attention mechanisms is challenging, as the increased patch tokens complicate long-sequence modeling. To address this, we propose MFuser, a novel Mamba-based fusion framework that efficiently combines the strengths of VFMs and VLMs while maintaining linear scalability in sequence length. MFuser consists of two key components: MVFuser, which acts as a co-adapter to jointly fine-tune the two models by capturing both sequential and spatial dynamics; and MTEnhancer, a hybrid attention-Mamba module that refines text embeddings by incorporating image priors. Our approach achieves precise feature locality and strong text alignment without incurring significant computational overhead. Extensive experiments demonstrate that MFuser significantly outperforms state-of-the-art DGSS methods, achieving 68.20 mIoU on synthetic-to-real and 71.87 mIoU on real-to-real benchmarks. The code is available at https://github.com/devinxzhang/MFuser.

Paperid: 1180, https://arxiv.org/pdf/2504.03096.pdf GitHub

Abstract:
In this work, we focus on scaling open-vocabulary action detection. Existing approaches for action detection are predominantly limited to closed-set scenarios and rely on complex, parameter-heavy architectures. Extending these models to the open-vocabulary setting poses two key challenges: (1) the lack of large-scale datasets with many action classes for robust training, and (2) parameter-heavy adaptations to a pretrained vision-language contrastive model to convert it for detection, risking overfitting the additional non-pretrained parameters to base action classes. Firstly, we introduce an encoder-only multimodal model for video action detection, reducing the reliance on parameter-heavy additions for video action detection. Secondly, we introduce a simple weakly supervised training strategy to exploit an existing closed-set action detection dataset for pretraining. Finally, we depart from the ill-posed base-to-novel benchmark used by prior works in open-vocabulary action detection and devise a new benchmark to evaluate on existing closed-set action detection datasets without ever using them for training, showing novel results to serve as baselines for future work. Our code is available at https://siatheindochinese.github.io/sia_act_page/ .

Paperid: 1181, https://arxiv.org/pdf/2504.02876.pdf GitHub

Abstract:
Visual grounding focuses on detecting objects from images based on language expressions. Recent Large Vision-Language Models (LVLMs) have significantly advanced visual grounding performance by training large models with large-scale datasets. However, the problem remains challenging, especially when similar objects appear in the input image. For example, an LVLM may not be able to differentiate Diet Coke and regular Coke in an image. In this case, if additional reference images of Diet Coke and regular Coke are available, it can help the visual grounding of similar objects. In this work, we introduce a new task named Multimodal Reference Visual Grounding (MRVG). In this task, a model has access to a set of reference images of objects in a database. Based on these reference images and a language expression, the model is required to detect a target object from a query image. We first introduce a new dataset to study the MRVG problem. Then we introduce a novel method, named MRVG-Net, to solve this visual grounding problem. We show that by efficiently using reference images with few-shot object detection and using Large Language Models (LLMs) for object matching, our method achieves superior visual grounding performance compared to the state-of-the-art LVLMs such as Qwen2.5-VL-72B. Our approach bridges the gap between few-shot detection and visual grounding, unlocking new capabilities for visual understanding, which has wide applications in robotics. Project page with our video, code, and dataset: https://irvlutd.github.io/MultiGrounding

Paperid: 1182, https://arxiv.org/pdf/2504.02862.pdf GitHub

Abstract:
Large Vision-Language Models (LVLMs) are gradually becoming the foundation for many artificial intelligence applications. However, understanding their internal working mechanisms has continued to puzzle researchers, which in turn limits the further enhancement of their capabilities. In this paper, we seek to investigate how multimodal knowledge evolves and eventually induces natural languages in LVLMs. We design a series of novel strategies for analyzing internal knowledge within LVLMs, and delve into the evolution of multimodal knowledge from three levels, including single token probabilities, token probability distributions, and feature encodings. In this process, we identify two key nodes in knowledge evolution: the critical layers and the mutation layers, dividing the evolution process into three stages: rapid evolution, stabilization, and mutation. Our research is the first to reveal the trajectory of knowledge evolution in LVLMs, providing a fresh perspective for understanding their underlying mechanisms. Our codes are available at https://github.com/XIAO4579/Vlm-interpretability.

Paperid: 1183, https://arxiv.org/pdf/2504.02823.pdf GitHub

Abstract:
Advancements in Computer-Aided Screening (CAS) systems are essential for improving the detection of security threats in X-ray baggage scans. However, current datasets are limited in representing real-world, sophisticated threats and concealment tactics, and existing approaches are constrained by a closed-set paradigm with predefined labels. To address these challenges, we introduce STCray, the first multimodal X-ray baggage security dataset, comprising 46,642 image-caption paired scans across 21 threat categories, generated using an X-ray scanner for airport security. STCray is meticulously developed with our specialized protocol that ensures domain-aware, coherent captions, that lead to the multi-modal instruction following data in X-ray baggage security. This allows us to train a domain-aware visual AI assistant named STING-BEE that supports a range of vision-language tasks, including scene comprehension, referring threat localization, visual grounding, and visual question answering (VQA), establishing novel baselines for multi-modal learning in X-ray baggage security. Further, STING-BEE shows state-of-the-art generalization in cross-domain settings. Code, data, and models are available at https://divs1159.github.io/STING-BEE/.

Paperid: 1184, https://arxiv.org/pdf/2504.02821.pdf GitHub

Abstract:
Given that interpretability and steerability are crucial to AI safety, Sparse Autoencoders (SAEs) have emerged as a tool to enhance them in Large Language Models (LLMs). In this work, we extend the application of SAEs to Vision-Language Models (VLMs), such as CLIP, and introduce a comprehensive framework for evaluating monosemanticity at the neuron-level in vision representations. To ensure that our evaluation aligns with human perception, we propose a benchmark derived from a large-scale user study. Our experimental results reveal that SAEs trained on VLMs significantly enhance the monosemanticity of individual neurons, with sparsity and wide latents being the most influential factors. Notably, we demonstrate that applying SAE interventions on CLIP's vision encoder directly steers multimodal LLM outputs (e.g., LLaVA), without any modifications to the underlying model. These findings emphasize the practicality and efficacy of SAEs as an unsupervised tool for enhancing both interpretability and control of VLMs. Code is available at https://github.com/ExplainableML/sae-for-vlm.

Paperid: 1185, https://arxiv.org/pdf/2504.02587.pdf GitHub

Abstract:
Reinforcement learning (RL) has recently shown strong potential in improving the reasoning capabilities of large language models and is now being actively extended to vision-language models (VLMs). However, existing RL applications in VLMs often rely on heavily engineered frameworks that hinder reproducibility and accessibility, while lacking standardized evaluation protocols, making it difficult to compare results or interpret training dynamics. This work introduces a transparent, from-scratch framework for RL in VLMs, offering a minimal yet functional four-step pipeline validated across multiple models and datasets. In addition, a standardized evaluation scheme is proposed to assess training dynamics and reflective behaviors. Extensive experiments on visual reasoning tasks uncover key empirical findings: response length is sensitive to random seeds, reflection correlates with output length, and RL consistently outperforms supervised fine-tuning (SFT) in generalization, even with high-quality data. These findings, together with the proposed framework, aim to establish a reproducible baseline and support broader engagement in RL-based VLM research.

Paperid: 1186, https://arxiv.org/pdf/2504.02477.pdf GitHub

Abstract:
Robot vision has greatly benefited from advancements in multimodal fusion techniques and vision-language models (VLMs). We systematically review the applications of multimodal fusion in key robotic vision tasks, including semantic scene understanding, simultaneous localization and mapping (SLAM), 3D object detection, navigation and localization, and robot manipulation. We compare VLMs based on large language models (LLMs) with traditional multimodal fusion methods, analyzing their advantages, limitations, and synergies. Additionally, we conduct an in-depth analysis of commonly used datasets, evaluating their applicability and challenges in real-world robotic scenarios. Furthermore, we identify critical research challenges such as cross-modal alignment, efficient fusion strategies, real-time deployment, and domain adaptation, and propose future research directions, including self-supervised learning for robust multimodal representations, transformer-based fusion architectures, and scalable multimodal frameworks. Through a comprehensive review, comparative analysis, and forward-looking discussion, we provide a valuable reference for advancing multimodal perception and interaction in robotic vision. A comprehensive list of studies in this survey is available at https://github.com/Xiaofeng-Han-Res/MF-RV.

Paperid: 1187, https://arxiv.org/pdf/2504.02438.pdf GitHub

Abstract:
Long-form video processing fundamentally challenges vision-language models (VLMs) due to the high computational costs of handling extended temporal sequences. Existing token pruning and feature merging methods often sacrifice critical temporal dependencies or dilute semantic information. We introduce differential distillation, a principled approach that systematically preserves task-relevant information while suppressing redundancy. Based on this principle, we develop ViLAMP, a hierarchical video-language model that processes hour-long videos at "mixed precision" through two key mechanisms: (1) differential keyframe selection that maximizes query relevance while maintaining temporal distinctiveness at the frame level and (2) differential feature merging that preserves query-salient features in non-keyframes at the patch level. Hence, ViLAMP retains full information in keyframes while reducing non-keyframes to their most salient features, resembling mixed-precision training. Extensive experiments demonstrate ViLAMP's superior performance across four video understanding benchmarks, particularly on long-form content. Notably, ViLAMP can process ultra-long videos (up to 10K frames) on a single NVIDIA A100 GPU, achieving substantial computational efficiency while maintaining state-of-the-art performance. Code and model are available at https://github.com/steven-ccq/ViLAMP.

Paperid: 1188, https://arxiv.org/pdf/2504.01954.pdf GitHub

Abstract:
Referring expression segmentation (RES) aims at segmenting the entities' masks that match the descriptive language expression. While traditional RES methods primarily address object-level grounding, real-world scenarios demand a more versatile framework that can handle multiple levels of target granularity, such as multi-object, single object or part-level references. This introduces great challenges due to the diverse and nuanced ways users describe targets. However, existing datasets and models mainly focus on designing grounding specialists for object-level target localization, lacking the necessary data resources and unified frameworks for the more practical multi-grained RES. In this paper, we take a step further towards visual granularity unified RES task. To overcome the limitation of data scarcity, we introduce a new multi-granularity referring expression segmentation (MRES) task, alongside the RefCOCOm benchmark, which includes part-level annotations for advancing finer-grained visual understanding. In addition, we create MRES-32M, the largest visual grounding dataset, comprising over 32.2M masks and captions across 1M images, specifically designed for part-level vision-language grounding. To tackle the challenges of multi-granularity RES, we propose UniRES++, a unified multimodal large language model that integrates object-level and part-level RES tasks. UniRES++ incorporates targeted designs for fine-grained visual feature exploration. With the joint model architecture and parameters, UniRES++ achieves state-of-the-art performance across multiple benchmarks, including RefCOCOm for MRES, gRefCOCO for generalized RES, and RefCOCO, RefCOCO+, RefCOCOg for classic RES. To foster future research into multi-grained visual grounding, our RefCOCOm benchmark, MRES-32M dataset and model UniRES++ will be publicly available at https://github.com/Rubics-Xuan/MRES.

Paperid: 1189, https://arxiv.org/pdf/2504.01838.pdf GitHub

Abstract:
Artificial Intelligence (AI) in skin disease diagnosis has improved significantly, but a major concern is that these models frequently show biased performance across subgroups, especially regarding sensitive attributes such as skin color. To address these issues, we propose a novel generative AI-based framework, namely, Dermatology Diffusion Transformer (DermDiT), which leverages text prompts generated via Vision Language Models and multimodal text-image learning to generate new dermoscopic images. We utilize large vision language models to generate accurate and proper prompts for each dermoscopic image which helps to generate synthetic images to improve the representation of underrepresented groups (patient, disease, etc.) in highly imbalanced datasets for clinical diagnoses. Our extensive experimentation showcases the large vision language models providing much more insightful representations, that enable DermDiT to generate high-quality images. Our code is available at https://github.com/Munia03/DermDiT

Paperid: 1190, https://arxiv.org/pdf/2504.01561.pdf GitHub

Abstract:
Accurate segmentation of lesions plays a critical role in medical image analysis and diagnosis. Traditional segmentation approaches that rely solely on visual features often struggle with the inherent uncertainty in lesion distribution and size. To address these issues, we propose STPNet, a Scale-aware Text Prompt Network that leverages vision-language modeling to enhance medical image segmentation. Our approach utilizes multi-scale textual descriptions to guide lesion localization and employs retrieval-segmentation joint learning to bridge the semantic gap between visual and linguistic modalities. Crucially, STPNet retrieves relevant textual information from a specialized medical text repository during training, eliminating the need for text input during inference while retaining the benefits of cross-modal learning. We evaluate STPNet on three datasets: COVID-Xray, COVID-CT, and Kvasir-SEG. Experimental results show that our vision-language approach outperforms state-of-the-art segmentation methods, demonstrating the effectiveness of incorporating textual semantic knowledge into medical image analysis. The code has been made publicly on https://github.com/HUANGLIZI/STPNet.

Paperid: 1191, https://arxiv.org/pdf/2504.01321.pdf GitHub

Abstract:
Transformer has recently demonstrated great potential in improving vision-language (VL) tracking algorithms. However, most of the existing VL trackers rely on carefully designed mechanisms to perform the multi-stage multi-modal fusion. Additionally, direct multi-modal fusion without alignment ignores distribution discrepancy between modalities in feature space, potentially leading to suboptimal representations. In this work, we propose COST, a contrastive one-stage transformer fusion framework for VL tracking, aiming to learn semantically consistent and unified VL representations. Specifically, we introduce a contrastive alignment strategy that maximizes mutual information (MI) between a video and its corresponding language description. This enables effective cross-modal alignment, yielding semantically consistent features in the representation space. By leveraging a visual-linguistic transformer, we establish an efficient multi-modal fusion and reasoning mechanism, empirically demonstrating that a simple stack of transformer encoders effectively enables unified VL representations. Moreover, we contribute a newly collected VL tracking benchmark dataset for small object tracking, named VL-SOT500, with bounding boxes and language descriptions. Our dataset comprises two challenging subsets, VL-SOT230 and VL-SOT270, dedicated to evaluating generic and high-speed small object tracking, respectively. Small object tracking is notoriously challenging due to weak appearance and limited features, and this dataset is, to the best of our knowledge, the first to explore the usage of language cues to enhance visual representation for small object tracking. Extensive experiments demonstrate that COST achieves state-of-the-art performance on five existing VL tracking datasets, as well as on our proposed VL-SOT500 dataset. Source codes and dataset will be made publicly available.

Paperid: 1192, https://arxiv.org/pdf/2504.01308.pdf GitHub

Abstract:
Vision-Language Models (VLMs) extend the capabilities of Large Language Models (LLMs) by incorporating visual information, yet they remain vulnerable to jailbreak attacks, especially when processing noisy or corrupted images. Although existing VLMs adopt security measures during training to mitigate such attacks, vulnerabilities associated with noise-augmented visual inputs are overlooked. In this work, we identify that missing noise-augmented training causes critical security gaps: many VLMs are susceptible to even simple perturbations such as Gaussian noise. To address this challenge, we propose Robust-VLGuard, a multimodal safety dataset with aligned / misaligned image-text pairs, combined with noise-augmented fine-tuning that reduces attack success rates while preserving functionality of VLM. For stronger optimization-based visual perturbation attacks, we propose DiffPure-VLM, leveraging diffusion models to convert adversarial perturbations into Gaussian-like noise, which can be defended by VLMs with noise-augmented safety fine-tuning. Experimental results demonstrate that the distribution-shifting property of diffusion model aligns well with our fine-tuned VLMs, significantly mitigating adversarial perturbations across varying intensities. The dataset and code are available at https://github.com/JarvisUSTC/DiffPure-RobustVLM.

Paperid: 1193, https://arxiv.org/pdf/2503.24129.pdf GitHub

Abstract:
The platonic representation hypothesis suggests that vision and language embeddings become more homogeneous as model and dataset sizes increase. In particular, pairwise distances within each modality become more similar. This suggests that as foundation models mature, it may become possible to match vision and language embeddings in a fully unsupervised fashion, i.e. without parallel data. We present the first feasibility study, and investigate conformity of existing vision and language foundation models in the context of unsupervised, or "blind", matching. First, we formulate unsupervised matching as a quadratic assignment problem and introduce a novel heuristic that outperforms previous solvers. We also develop a technique to find optimal matching problems, for which a non-trivial match is very likely. Second, we conduct an extensive study deploying a range of vision and language models on four datasets. Our analysis reveals that for many problem instances, vision and language representations can be indeed matched without supervision. This finding opens up the exciting possibility of embedding semantic knowledge into other modalities virtually annotation-free. As a proof of concept, we showcase an unsupervised classifier, which achieves non-trivial classification accuracy without any image-text annotation.

Paperid: 1194, https://arxiv.org/pdf/2503.23907.pdf GitHub

Abstract:
Image Aesthetic Assessment (IAA) is a long-standing and challenging research task. However, its subset, Human Image Aesthetic Assessment (HIAA), has been scarcely explored. To bridge this research gap, our work pioneers a holistic implementation framework tailored for HIAA. Specifically, we introduce HumanBeauty, the first dataset purpose-built for HIAA, which comprises 108k high-quality human images with manual annotations. To achieve comprehensive and fine-grained HIAA, 50K human images are manually collected through a rigorous curation process and annotated leveraging our trailblazing 12-dimensional aesthetic standard, while the remaining 58K with overall aesthetic labels are systematically filtered from public datasets. Based on the HumanBeauty database, we propose HumanAesExpert, a powerful Vision Language Model for aesthetic evaluation of human images. We innovatively design an Expert head to incorporate human knowledge of aesthetic sub-dimensions while jointly utilizing the Language Modeling (LM) and Regression heads. This approach empowers our model to achieve superior proficiency in both overall and fine-grained HIAA. Furthermore, we introduce a MetaVoter, which aggregates scores from all three heads, to effectively balance the capabilities of each head, thereby realizing improved assessment precision. Extensive experiments demonstrate that our HumanAesExpert models deliver significantly better performance in HIAA than other state-of-the-art models. Project webpage: https://humanaesexpert.github.io/HumanAesExpert/

Paperid: 1195, https://arxiv.org/pdf/2503.23730.pdf GitHub

Abstract:
The recent emergence of Large Vision-Language Models(VLMs) has resulted in a variety of different benchmarks for evaluating such models. Despite this, we observe that most existing evaluation methods suffer from the fact that they either require the model to choose from pre-determined responses, sacrificing open-endedness, or evaluate responses using a judge model, resulting in subjective and unreliable evaluation. In addition, we observe a lack of benchmarks for VLMs in the Korean language, which are necessary as a separate metric from more common English language benchmarks, as the performance of generative language models can differ significantly based on the language being used. Therefore, we present KOFFVQA, a general-purpose free-form visual question answering benchmark in the Korean language for the evaluation of VLMs. Our benchmark consists of 275 carefully crafted questions each paired with an image and grading criteria covering 10 different aspects of VLM performance. The grading criteria eliminate the problem of unreliability by allowing the judge model to grade each response based on a pre-determined set of rules. By defining the evaluation criteria in an objective manner, even a small open-source model can be used to evaluate models on our benchmark reliably. In addition to evaluating a large number of existing VLMs on our benchmark, we also experimentally verify that our method of using pre-existing grading criteria for evaluation is much more reliable than existing methods. Our evaluation code is available at https://github.com/maum-ai/KOFFVQA

Paperid: 1196, https://arxiv.org/pdf/2503.23573.pdf GitHub

Abstract:
Vision-language models (VLMs) are prone to object hallucinations, where they erroneously indicate the presenceof certain objects in an image. Existing benchmarks quantify hallucinations using relatively small, labeled datasets. However, this approach is i) insufficient to assess hallucinations that arise in open-world settings, where VLMs are widely used, and ii) inadequate for detecting systematic errors in VLMs. We propose DASH (Detection and Assessment of Systematic Hallucinations), an automatic, large-scale pipeline designed to identify systematic hallucinations of VLMs on real-world images in an open-world setting. A key component is DASH-OPT for image-based retrieval, where we optimize over the ''natural image manifold'' to generate images that mislead the VLM. The output of DASH consists of clusters of real and semantically similar images for which the VLM hallucinates an object. We apply DASH to PaliGemma and two LLaVA-NeXT models across 380 object classes and, in total, find more than 19k clusters with 950k images. We study the transfer of the identified systematic hallucinations to other VLMs and show that fine-tuning PaliGemma with the model-specific images obtained with DASH mitigates object hallucinations. Code and data are available at https://YanNeu.github.io/DASH.

Paperid: 1197, https://arxiv.org/pdf/2503.23509.pdf GitHub

Abstract:
Referring Video Object Segmentation (RVOS) aims to segment target objects throughout a video based on a text description. This task has attracted increasing attention in the field of computer vision due to its promising applications in video editing and human-agent interaction. Recently, ReferDINO has demonstrated promising performance in this task by adapting object-level vision-language knowledge from pretrained foundational image models. In this report, we further enhance its capabilities by incorporating the advantages of SAM2 in mask quality and object consistency. In addition, to effectively balance performance between single-object and multi-object scenarios, we introduce a conditional mask fusion strategy that adaptively fuses the masks from ReferDINO and SAM2. Our solution, termed ReferDINO-Plus, achieves 60.43 $\mathcal{J}\&\mathcal{F}$ on MeViS test set, securing 2nd place in the MeViS PVUW challenge at CVPR 2025. The code is available at: https://github.com/iSEE-Laboratory/ReferDINO-Plus.

Paperid: 1198, https://arxiv.org/pdf/2503.23495.pdf GitHub GitHub

Abstract:
Understanding the representation shift on Vision Language Models like CLIP under different augmentations provides valuable insights on Mechanistic Interpretability. In this study, we show the shift on CLIP's embeddings on 9 common augmentation techniques: noise, blur, color jitter, scale and rotate, flip, elastic and perspective transforms, random brightness and contrast, and coarse dropout of pixel blocks. We scrutinize the embedding shifts under similarity on attention map, patch, edge, detail preservation, cosine similarity, L2 distance, pairwise distance and dendrogram clusters and provide qualitative analysis on sample images. Our findings suggest certain augmentations like noise, perspective transform and shift scaling have higher degree of drastic impact on embedding shift. This study provides a concrete foundation for future work on VLM's robustness for mechanical interpretation and adversarial data defense. The code implementation for this study can be found on \href{https://github.com/ashimdahal/clip-shift-analysis}{https://github.com/ashimdahal/clip-shift-analysis}.

Paperid: 1199, https://arxiv.org/pdf/2503.23368.pdf GitHub

Abstract:
Video diffusion models (VDMs) have advanced significantly in recent years, enabling the generation of highly realistic videos and drawing the attention of the community in their potential as world simulators. However, despite their capabilities, VDMs often fail to produce physically plausible videos due to an inherent lack of understanding of physics, resulting in incorrect dynamics and event sequences. To address this limitation, we propose a novel two-stage image-to-video generation framework that explicitly incorporates physics with vision and language informed physical prior. In the first stage, we employ a Vision Language Model (VLM) as a coarse-grained motion planner, integrating chain-of-thought and physics-aware reasoning to predict a rough motion trajectories/changes that approximate real-world physical dynamics while ensuring the inter-frame consistency. In the second stage, we use the predicted motion trajectories/changes to guide the video generation of a VDM. As the predicted motion trajectories/changes are rough, noise is added during inference to provide freedom to the VDM in generating motion with more fine details. Extensive experimental results demonstrate that our framework can produce physically plausible motion, and comparative evaluations highlight the notable superiority of our approach over existing methods. More video results are available on our Project Page: https://madaoer.github.io/projects/physically_plausible_video_generation.

Paperid: 1200, https://arxiv.org/pdf/2503.23312.pdf GitHub

Abstract:
Conversational recommender systems engage users in dialogues to refine their needs and provide more personalized suggestions. Although textual information suffices for many domains, visually driven categories such as fashion or home decor potentially require detailed visual information related to color, style, or design. To address this challenge, we propose LaViC (Large Vision-Language Conversational Recommendation Framework), a novel approach that integrates compact image representations into dialogue-based recommendation systems. LaViC leverages a large vision-language model in a two-stage process: (1) visual knowledge self-distillation, which condenses product images from hundreds of tokens into a small set of visual tokens in a self-distillation manner, significantly reducing computational overhead, and (2) recommendation prompt tuning, which enables the model to incorporate both dialogue context and distilled visual tokens, providing a unified mechanism for capturing textual and visual features. To support rigorous evaluation of visually-aware conversational recommendation, we construct a new dataset by aligning Reddit conversations with Amazon product listings across multiple visually oriented categories (e.g., fashion, beauty, and home). This dataset covers realistic user queries and product appearances in domains where visual details are crucial. Extensive experiments demonstrate that LaViC significantly outperforms text-only conversational recommendation methods and open-source vision-language baselines. Moreover, LaViC achieves competitive or superior accuracy compared to prominent proprietary baselines (e.g., GPT-3.5-turbo, GPT-4o-mini, and GPT-4o), demonstrating the necessity of explicitly using visual data for capturing product attributes and showing the effectiveness of our vision-language integration. Our code and dataset are available at https://github.com/jeon185/LaViC.

Paperid: 1201, https://arxiv.org/pdf/2503.23131.pdf GitHub GitHub

Abstract:
Recently, Vision Language Models (VLMs) have increasingly emphasized document visual grounding to achieve better human-computer interaction, accessibility, and detailed understanding. However, its application to visualizations such as charts remains under-explored due to the inherent complexity of interleaved visual-numerical relationships in chart images. Existing chart understanding methods primarily focus on answering questions without explicitly identifying the visual elements that support their predictions. To bridge this gap, we introduce RefChartQA, a novel benchmark that integrates Chart Question Answering (ChartQA) with visual grounding, enabling models to refer elements at multiple granularities within chart images. Furthermore, we conduct a comprehensive evaluation by instruction-tuning 5 state-of-the-art VLMs across different categories. Our experiments demonstrate that incorporating spatial awareness via grounding improves response accuracy by over 15%, reducing hallucinations, and improving model reliability. Additionally, we identify key factors influencing text-spatial alignment, such as architectural improvements in TinyChart, which leverages a token-merging module for enhanced feature fusion. Our dataset is open-sourced for community development and further advancements. All models and code will be publicly available at https://github.com/moured/RefChartQA.

Paperid: 1202, https://arxiv.org/pdf/2503.22976.pdf GitHub

Abstract:
Recent advances in LVLMs have improved vision-language understanding, but they still struggle with spatial perception, limiting their ability to reason about complex 3D scenes. Unlike previous approaches that incorporate 3D representations into models to improve spatial understanding, we aim to unlock the potential of VLMs by leveraging spatially relevant image data. To this end, we introduce a novel 2D spatial data generation and annotation pipeline built upon scene data with 3D ground-truth. This pipeline enables the creation of a diverse set of spatial tasks, ranging from basic perception tasks to more complex reasoning tasks. Leveraging this pipeline, we construct SPAR-7M, a large-scale dataset generated from thousands of scenes across multiple public datasets. In addition, we introduce SPAR-Bench, a benchmark designed to offer a more comprehensive evaluation of spatial capabilities compared to existing spatial benchmarks, supporting both single-view and multi-view inputs. Training on both SPAR-7M and large-scale 2D datasets enables our models to achieve state-of-the-art performance on 2D spatial benchmarks. Further fine-tuning on 3D task-specific datasets yields competitive results, underscoring the effectiveness of our dataset in enhancing spatial reasoning.

Paperid: 1203, https://arxiv.org/pdf/2503.22655.pdf GitHub

Abstract:
Training vision-language models (VLMs) typically requires large-scale, high-quality image-text pairs, but collecting or synthesizing such data is costly. In contrast, text data is abundant and inexpensive, prompting the question: can high-quality multimodal training data be synthesized purely from text? To tackle this, we propose a cross-integrated three-stage multimodal data synthesis framework, which generates two datasets: Unicorn-1.2M and Unicorn-471K-Instruction. In Stage 1: Diverse Caption Data Synthesis, we construct 1.2M semantically diverse high-quality captions by expanding sparse caption seeds using large language models (LLMs). In Stage 2: Instruction-Tuning Data Generation, we further process 471K captions into multi-turn instruction-tuning tasks to support complex reasoning. Finally, in Stage 3: Modality Representation Transfer, these textual captions representations are transformed into visual representations, resulting in diverse synthetic image representations. This three-stage process enables us to construct Unicorn-1.2M for pretraining and Unicorn-471K-Instruction for instruction-tuning, without relying on real images. By eliminating the dependency on real images while maintaining data quality and diversity, our framework offers a cost-effective and scalable solution for VLMs training. Code is available at https://github.com/Yu-xm/Unicorn.git.

Paperid: 1204, https://arxiv.org/pdf/2503.22263.pdf GitHub

Abstract:
The increasing emphasis on privacy and data security has driven the adoption of federated learning, a decentralized approach to train machine learning models without sharing raw data. Prompt learning, which fine-tunes prompt embeddings of pretrained models, offers significant advantages in federated settings by reducing computational costs and communication overheads while leveraging the strong performance and generalization capabilities of vision-language models such as CLIP. This paper addresses the intersection of federated learning and prompt learning, particularly for vision-language models. In this work, we introduce a comprehensive framework, named FLIP, to evaluate federated prompt learning algorithms. FLIP assesses the performance of 8 state-of-the-art federated prompt learning methods across 4 federated learning protocols and 12 open datasets, considering 6 distinct evaluation scenarios. Our findings demonstrate that prompt learning maintains strong generalization performance in both in-distribution and out-of-distribution settings with minimal resource consumption. This work highlights the effectiveness of federated prompt learning in environments characterized by data scarcity, unseen classes, and cross-domain distributional shifts. We open-source the code for all implemented algorithms in FLIP to facilitate further research in this domain.

Paperid: 1205, https://arxiv.org/pdf/2503.22093.pdf GitHub

Abstract:
Vision Language Models (VLMs) have demonstrated strong reasoning capabilities in Visual Question Answering (VQA) tasks; however, their ability to perform Theory of Mind (ToM) tasks, such as inferring human intentions, beliefs, and mental states, remains underexplored. We propose an open-ended question framework to evaluate VLMs' performance across diverse categories of ToM tasks. We curated and annotated a benchmark dataset of 30 images and evaluated the performance of four VLMs of varying sizes. Our results show that the GPT-4 model outperformed all the others, with only one smaller model, GPT-4o-mini, achieving comparable performance. We observed that VLMs often struggle to infer intentions in complex scenarios such as bullying or cheating. Our findings reveal that smaller models can sometimes infer correct intentions despite relying on incorrect visual cues. The dataset is available at https://github.com/ximingwen/ToM-AAAI25-Multimodal.

Paperid: 1206, https://arxiv.org/pdf/2503.22081.pdf GitHub

Abstract:
The rapid advancement of remote sensing foundation models, particularly vision and multimodal models, has significantly enhanced the capabilities of intelligent geospatial data interpretation. These models combine various data modalities, such as optical, radar, and LiDAR imagery, with textual and geographic information, enabling more comprehensive analysis and understanding of remote sensing data. The integration of multiple modalities allows for improved performance in tasks like object detection, land cover classification, and change detection, which are often challenged by the complex and heterogeneous nature of remote sensing data. However, despite these advancements, several challenges remain. The diversity in data types, the need for large-scale annotated datasets, and the complexity of multimodal fusion techniques pose significant obstacles to the effective deployment of these models. Moreover, the computational demands of training and fine-tuning multimodal models require significant resources, further complicating their practical application in remote sensing image interpretation tasks. This paper provides a comprehensive review of the state-of-the-art in vision and multimodal foundation models for remote sensing, focusing on their architecture, training methods, datasets and application scenarios. We discuss the key challenges these models face, such as data alignment, cross-modal transfer learning, and scalability, while also identifying emerging research directions aimed at overcoming these limitations. Our goal is to provide a clear understanding of the current landscape of remote sensing foundation models and inspire future research that can push the boundaries of what these models can achieve in real-world applications. The list of resources collected by the paper can be found in the https://github.com/IRIP-BUAA/A-Review-for-remote-sensing-vision-language-models.

Paperid: 1207, https://arxiv.org/pdf/2503.21840.pdf GitHub

Abstract:
Introduction: This study provides a comprehensive performance assessment of vision-language models (VLMs) against established convolutional neural networks (CNNs) and classic machine learning models (CMLs) for computer-aided detection (CADe) and computer-aided diagnosis (CADx) of colonoscopy polyp images. Method: We analyzed 2,258 colonoscopy images with corresponding pathology reports from 428 patients. We preprocessed all images using standardized techniques (resizing, normalization, and augmentation) and implemented a rigorous comparative framework evaluating 11 distinct models: ResNet50, 4 CMLs (random forest, support vector machine, logistic regression, decision tree), two specialized contrastive vision language encoders (CLIP, BiomedCLIP), and three general-purpose VLMs ( GPT-4 Gemini-1.5-Pro, Claude-3-Opus). Our performance assessment focused on two clinical tasks: polyp detection (CADe) and classification (CADx). Result: In polyp detection, ResNet50 achieved the best performance (F1: 91.35%, AUROC: 0.98), followed by BiomedCLIP (F1: 88.68%, AUROC: [AS1] ). GPT-4 demonstrated comparable effectiveness to traditional machine learning approaches (F1: 81.02%, AUROC: [AS2] ), outperforming other general-purpose VLMs. For polyp classification, performance rankings remained consistent but with lower overall metrics. ResNet50 maintained the highest efficacy (weighted F1: 74.94%), while GPT-4 demonstrated moderate capability (weighted F1: 41.18%), significantly exceeding other VLMs (Claude-3-Opus weighted F1: 25.54%, Gemini 1.5 Pro weighted F1: 6.17%). Conclusion: CNNs remain superior for both CADx and CADe tasks. However, VLMs like BioMedCLIP and GPT-4 may be useful for polyp detection tasks where training CNNs is not feasible.

Paperid: 1208, https://arxiv.org/pdf/2503.21839.pdf GitHub

Abstract:
We investigate a critical yet under-explored question in Large Vision-Language Models (LVLMs): Do LVLMs genuinely comprehend interleaved image-text in the document? Existing document understanding benchmarks often assess LVLMs using question-answer formats, which are information-sparse and difficult to guarantee the coverage of long-range dependencies. To address this issue, we introduce a novel and challenging Multimodal Document Summarization Benchmark (M-DocSum-Bench), which comprises 500 high-quality arXiv papers, along with interleaved multimodal summaries aligned with human preferences. M-DocSum-Bench is a reference-based generation task and necessitates the generation of interleaved image-text summaries using provided reference images, thereby simultaneously evaluating capabilities in understanding, reasoning, localization, and summarization within complex multimodal document scenarios. To facilitate this benchmark, we develop an automated framework to construct summaries and propose a fine-grained evaluation method called M-DocEval. Moreover, we further develop a robust summarization baseline, i.e., M-DocSum-7B, by progressive two-stage training with diverse instruction and preference data. The extensive results on our M-DocSum-Bench reveal that the leading LVLMs struggle to maintain coherence and accurately integrate information within long and interleaved contexts, often exhibiting confusion between similar images and a lack of robustness. Notably, M-DocSum-7B achieves state-of-the-art performance compared to larger and closed-source models (including GPT-4o, Gemini Pro, Claude-3.5-Sonnet and Qwen2.5-VL-72B, etc.), demonstrating the potential of LVLMs for improved interleaved image-text understanding. The code, data, and models are available at https://github.com/stepfun-ai/M-DocSum-Bench.

Paperid: 1209, https://arxiv.org/pdf/2503.21761.pdf GitHub

Abstract:
This paper presents a unified approach to understanding dynamic scenes from casual videos. Large pretrained vision foundation models, such as vision-language, video depth prediction, motion tracking, and segmentation models, offer promising capabilities. However, training a single model for comprehensive 4D understanding remains challenging. We introduce Uni4D, a multi-stage optimization framework that harnesses multiple pretrained models to advance dynamic 3D modeling, including static/dynamic reconstruction, camera pose estimation, and dense 3D motion tracking. Our results show state-of-the-art performance in dynamic 4D modeling with superior visual quality. Notably, Uni4D requires no retraining or fine-tuning, highlighting the effectiveness of repurposing visual foundation models for 4D understanding.

Paperid: 1210, https://arxiv.org/pdf/2503.21483.pdf GitHub

Abstract:
Large video-language models (VLMs) have demonstrated promising progress in various video understanding tasks. However, their effectiveness in long-form video analysis is constrained by limited context windows. Traditional approaches, such as uniform frame sampling, often inevitably allocate resources to irrelevant content, diminishing their effectiveness in real-world scenarios. In this paper, we introduce BOLT, a method to BOost Large VLMs without additional Training through a comprehensive study of frame selection strategies. First, to enable a more realistic evaluation of VLMs in long-form video understanding, we propose a multi-source retrieval evaluation setting. Our findings reveal that uniform sampling performs poorly in noisy contexts, underscoring the importance of selecting the right frames. Second, we explore several frame selection strategies based on query-frame similarity and analyze their effectiveness at inference time. Our results show that inverse transform sampling yields the most significant performance improvement, increasing accuracy on the Video-MME benchmark from 53.8% to 56.1% and MLVU benchmark from 58.9% to 63.4%. Our code is available at https://github.com/sming256/BOLT.

Paperid: 1211, https://arxiv.org/pdf/2503.20752.pdf GitHub

Abstract:
Visual reasoning abilities play a crucial role in understanding complex multimodal data, advancing both domain-specific applications and artificial general intelligence (AGI). Existing methods improve VLM reasoning via Chain-of-Thought (CoT) supervised fine-tuning, using meticulously annotated training data to enhance visual reasoning capabilities. However, this training paradigm may lead to overfitting and cognitive rigidity, restricting the model's ability to transfer visual reasoning skills across domains and limiting its real-world applicability. To address these limitations, we propose Reason-RFT, a novel reinforcement fine-tuning framework that significantly enhances generalization capabilities in visual reasoning tasks. Reason-RFT introduces a two-phase training framework for visual reasoning: (1) Supervised Fine-Tuning (SFT) with curated Chain-of-Thought (CoT) data activates the reasoning potential of Vision-Language Models (VLMs), followed by (2) Group Relative Policy Optimization (GRPO)-based reinforcement learning that generates multiple reasoning-response pairs, significantly enhancing generalization in visual reasoning tasks. To evaluate Reason-RFT's visual reasoning capabilities, we reconstructed a comprehensive dataset spanning visual counting, structure perception, and spatial transformation. Experimental results demonstrate Reasoning-RFT's three key advantages: (1) Performance Enhancement: achieving state-of-the-art results across multiple tasks, outperforming most mainstream open-source and proprietary models; (2) Generalization Superiority: consistently maintaining robust performance across diverse tasks and domains, outperforming alternative training paradigms; (3) Data Efficiency: excelling in few-shot learning scenarios while surpassing full-dataset SFT baselines. Project website: https://tanhuajie.github.io/ReasonRFT

Paperid: 1212, https://arxiv.org/pdf/2503.20612.pdf GitHub GitHub

Abstract:
Recent pre-trained vision-language models (PT-VLMs) often face a Multi-Domain Task Incremental Learning (MTIL) scenario in practice, where several classes and domains of multi-modal tasks are incrementally arrived. Without access to previously seen tasks and unseen tasks, memory-constrained MTIL suffers from forward and backward forgetting. To alleviate the above challenges, parameter-efficient fine-tuning techniques (PEFT), such as prompt tuning, are employed to adapt the PT-VLM to the diverse incrementally learned tasks. To achieve effective new task adaptation, existing methods only consider the effect of PEFT strategy selection, but neglect the influence of PEFT parameter setting (e.g., prompting). In this paper, we tackle the challenge of optimizing prompt designs for diverse tasks in MTIL and propose an Instance-Aware Prompting (IAP) framework. Specifically, our Instance-Aware Gated Prompting (IA-GP) strategy enhances adaptation to new tasks while mitigating forgetting by adaptively assigning prompts across transformer layers at the instance level. Our Instance-Aware Class-Distribution-Driven Prompting (IA-CDDP) improves the task adaptation process by determining an accurate task-label-related confidence score for each instance. Experimental evaluations across 11 datasets, using three performance metrics, demonstrate the effectiveness of our proposed method. The source codes are available at https://github.com/FerdinandZJU/IAP.

Paperid: 1213, https://arxiv.org/pdf/2503.20322.pdf GitHub

Abstract:
Multimodal large language models (MLLMs) have demonstrated impressive performance in various vision-language (VL) tasks, but their expensive computations still limit the real-world application. To address this issue, recent efforts aim to compress the visual features to save the computational costs of MLLMs. However, direct visual compression methods, e.g. efficient projectors, inevitably destroy the visual semantics in MLLM, especially in difficult samples. To overcome this shortcoming, we propose a novel dynamic pyramid network (DPN) for efficient MLLMs. Specifically, DPN formulates MLLM as a hierarchical structure where visual features are gradually compressed with increasing depth. In this case, even with a high compression ratio, fine-grained visual information can still be perceived in shallow layers. To maximize the benefit of DPN, we further propose an innovative Dynamic Pooling Experts (DPE) that can dynamically choose the optimal visual compression rate according to input features. With this design, harder samples will be assigned larger computations, thus preserving the model performance. To validate our approach, we conduct extensive experiments on two popular MLLMs and ten benchmarks. Experimental results show that DPN can save up to 56% average FLOPs on LLaVA while further achieving +0.74% performance gains. Besides, the generalization ability of DPN is also validated on the existing high-resolution MLLM called LLaVA-HR. The source code will be released at https://github.com/aihao2000/DPN-LLaVA.

Paperid: 1214, https://arxiv.org/pdf/2503.20271.pdf GitHub

Abstract:
Process-supervised reward models serve as a fine-grained function that provides detailed step-wise feedback to model responses, facilitating effective selection of reasoning trajectories for complex tasks. Despite its advantages, evaluation on PRMs remains less explored, especially in the multimodal domain. To address this gap, this paper first benchmarks current vision large language models (VLLMs) as two types of reward models: output reward models (ORMs) and process reward models (PRMs) on multiple vision-language benchmarks, which reveal that neither ORM nor PRM consistently outperforms across all tasks, and superior VLLMs do not necessarily yield better rewarding performance. To further advance evaluation, we introduce ViLBench, a vision-language benchmark designed to require intensive process reward signals. Notably, OpenAI's GPT-4o with Chain-of-Thought (CoT) achieves only 27.3% accuracy, indicating the benchmark's challenge for current VLLMs. Lastly, we preliminarily showcase a promising pathway towards bridging the gap between general VLLMs and reward models -- by collecting 73.6K vision-language process reward data using an enhanced tree-search algorithm, our 3B model is able to achieve an average improvement of 3.3% over standard CoT and up to 2.5% compared to its untrained counterpart on ViLBench by selecting OpenAI o1's generations. We release the implementations at https://ucsc-vlaa.github.io/ViLBench with our code, model, and data.

Paperid: 1215, https://arxiv.org/pdf/2503.20047.pdf GitHub

Abstract:
Vision-language models (VLMs) have shown promise in 2D medical image analysis, but extending them to 3D remains challenging due to the high computational demands of volumetric data and the difficulty of aligning 3D spatial features with clinical text. We present Med3DVLM, a 3D VLM designed to address these challenges through three key innovations: (1) DCFormer, an efficient encoder that uses decomposed 3D convolutions to capture fine-grained spatial features at scale; (2) SigLIP, a contrastive learning strategy with pairwise sigmoid loss that improves image-text alignment without relying on large negative batches; and (3) a dual-stream MLP-Mixer projector that fuses low- and high-level image features with text embeddings for richer multi-modal representations. We evaluate our model on the M3D dataset, which includes radiology reports and VQA data for 120,084 3D medical images. Results show that Med3DVLM achieves superior performance across multiple benchmarks. For image-text retrieval, it reaches 61.00% R@1 on 2,000 samples, significantly outperforming the current state-of-the-art M3D model (19.10%). For report generation, it achieves a METEOR score of 36.42% (vs. 14.38%). In open-ended visual question answering (VQA), it scores 36.76% METEOR (vs. 33.58%), and in closed-ended VQA, it achieves 79.95% accuracy (vs. 75.78%). These results highlight Med3DVLM's ability to bridge the gap between 3D imaging and language, enabling scalable, multi-task reasoning across clinical applications. Our code is publicly available at https://github.com/mirthAI/Med3DVLM.

Paperid: 1216, https://arxiv.org/pdf/2503.19839.pdf GitHub

Abstract:
Currently, instruction-based image editing methods have made significant progress by leveraging the powerful cross-modal understanding capabilities of vision language models (VLMs). However, they still face challenges in three key areas: 1) complex scenarios; 2) semantic consistency; and 3) fine-grained editing. To address these issues, we propose FireEdit, an innovative Fine-grained Instruction-based image editing framework that exploits a REgion-aware VLM. FireEdit is designed to accurately comprehend user instructions and ensure effective control over the editing process. Specifically, we enhance the fine-grained visual perception capabilities of the VLM by introducing additional region tokens. Relying solely on the output of the LLM to guide the diffusion model may lead to suboptimal editing results. Therefore, we propose a Time-Aware Target Injection module and a Hybrid Visual Cross Attention module. The former dynamically adjusts the guidance strength at various denoising stages by integrating timestep embeddings with the text embeddings. The latter enhances visual details for image editing, thereby preserving semantic consistency between the edited result and the source image. By combining the VLM enhanced with fine-grained region tokens and the time-dependent diffusion model, FireEdit demonstrates significant advantages in comprehending editing instructions and maintaining high semantic consistency. Extensive experiments indicate that our approach surpasses the state-of-the-art instruction-based image editing methods. Our project is available at https://zjgans.github.io/fireedit.github.io.

Paperid: 1217, https://arxiv.org/pdf/2503.19707.pdf GitHub

Abstract:
Vision-Language Models (VLMs) have recently emerged as powerful tools, excelling in tasks that integrate visual and textual comprehension, such as image captioning, visual question answering, and image-text retrieval. However, existing benchmarks for VLMs include spatial components, which often fail to isolate spatial reasoning from related tasks such as object detection or semantic comprehension. In this paper, we address these deficiencies with a multi-faceted approach towards understanding spatial reasoning. Informed by the diverse and multi-dimensional nature of human spatial reasoning abilities, we present a detailed analysis that first delineates the core elements of spatial reasoning: spatial relations, orientation and navigation, mental rotation, and spatial visualization, and then assesses the performance of these models in both synthetic and real-world images, bridging controlled and naturalistic contexts. We analyze 13 state-of-the-art Vision-Language Models, uncovering pivotal insights into their spatial reasoning performance. Our results reveal profound shortcomings in current VLMs, with average accuracy across the 13 models approximating random chance, highlighting spatial reasoning as a persistent obstacle. This work not only exposes the pressing need to advance spatial reasoning within VLMs but also establishes a solid platform for future exploration. Code available on GitHub (https://github.com/stogiannidis/srbench) and dataset available on HuggingFace (https://huggingface.co/datasets/stogiannidis/srbench).

Paperid: 1218, https://arxiv.org/pdf/2503.19653.pdf GitHub

Abstract:
This paper identifies OpenSDI, a challenge for spotting diffusion-generated images in open-world settings. In response to this challenge, we define a new benchmark, the OpenSDI dataset (OpenSDID), which stands out from existing datasets due to its diverse use of large vision-language models that simulate open-world diffusion-based manipulations. Another outstanding feature of OpenSDID is its inclusion of both detection and localization tasks for images manipulated globally and locally by diffusion models. To address the OpenSDI challenge, we propose a Synergizing Pretrained Models (SPM) scheme to build up a mixture of foundation models. This approach exploits a collaboration mechanism with multiple pretrained foundation models to enhance generalization in the OpenSDI context, moving beyond traditional training by synergizing multiple pretrained models through prompting and attending strategies. Building on this scheme, we introduce MaskCLIP, an SPM-based model that aligns Contrastive Language-Image Pre-Training (CLIP) with Masked Autoencoder (MAE). Extensive evaluations on OpenSDID show that MaskCLIP significantly outperforms current state-of-the-art methods for the OpenSDI challenge, achieving remarkable relative improvements of 14.23% in IoU (14.11% in F1) and 2.05% in accuracy (2.38% in F1) compared to the second-best model in localization and detection tasks, respectively. Our dataset and code are available at https://github.com/iamwangyabin/OpenSDI.

Paperid: 1219, https://arxiv.org/pdf/2503.19404.pdf GitHub

Abstract:
Recent years have witnessed remarkable advances in Large Vision-Language Models (LVLMs), which have achieved human-level performance across various complex vision-language tasks. Following LLaVA's paradigm, mainstream LVLMs typically employ a shallow MLP for visual-language alignment through a two-stage training process: pretraining for cross-modal alignment followed by instruction tuning. While this approach has proven effective, the underlying mechanisms of how MLPs bridge the modality gap remain poorly understood. Although some research has explored how LLMs process transformed visual tokens, few studies have investigated the fundamental alignment mechanism. Furthermore, the MLP adapter requires retraining whenever switching LLM backbones. To address these limitations, we first investigate the working principles of MLP adapters and discover that they learn to project visual embeddings into subspaces spanned by corresponding text embeddings progressively. Based on this insight, we propose LangBridge, a novel adapter that explicitly maps visual tokens to linear combinations of LLM vocabulary embeddings. This innovative design enables pretraining-free adapter transfer across different LLMs while maintaining performance. Our experimental results demonstrate that a LangBridge adapter pre-trained on Qwen2-0.5B can be directly applied to larger models such as LLaMA3-8B or Qwen2.5-14B while maintaining competitive performance. Overall, LangBridge enables interpretable vision-language alignment by grounding visual representations in LLM vocab embedding, while its plug-and-play design ensures efficient reuse across multiple LLMs with nearly no performance degradation. See our project page at https://jiaqiliao77.github.io/LangBridge.github.io/

Paperid: 1220, https://arxiv.org/pdf/2503.19355.pdf GitHub

Abstract:
Spatio-temporal reasoning is essential in understanding real-world environments in various fields, eg, autonomous driving and sports analytics. Recent advances have improved the spatial reasoning ability of Vision-Language Models (VLMs) by introducing large-scale data, but these models still struggle to analyze kinematic elements like traveled distance and speed of moving objects. To bridge this gap, we construct a spatio-temporal reasoning dataset and benchmark involving kinematic instruction tuning, referred to as STKit and STKit-Bench. They consist of real-world videos with 3D annotations, detailing object motion dynamics: traveled distance, speed, movement direction, inter-object distance comparisons, and relative movement direction. To further scale such data construction to videos without 3D labels, we propose an automatic pipeline to generate pseudo-labels using 4D reconstruction in real-world scale. With our kinematic instruction tuning data for spatio-temporal reasoning, we present ST-VLM, a VLM enhanced for spatio-temporal reasoning, which exhibits outstanding performance on STKit-Bench. Furthermore, we show that ST-VLM generalizes robustly across diverse domains and tasks, outperforming baselines on other spatio-temporal benchmarks (eg, ActivityNet, TVQA+). Finally, by integrating learned spatio-temporal reasoning with existing abilities, ST-VLM enables complex multi-step reasoning. Project page: https://ikodoh.github.io/ST-VLM.

Paperid: 1221, https://arxiv.org/pdf/2503.19311.pdf GitHub

Abstract:
This study addresses the technical bottlenecks in handling long text and the "hallucination" issue caused by insufficient short text information in remote sensing vision-language foundation models (VLFM). We propose a novel vision-language foundation model, LRSCLIP, and a multimodal dataset, LRS2M. The main contributions are as follows: (1) By integrating multi-source remote sensing data and adopting a large language model labeling strategy, we construct the LRS2M dataset, which contains 2 million image-text pairs, providing both short and long texts for the first time, thus solving the problem of semantic granularity limitations in existing datasets; (2) The design of the LRSCLIP architecture based on Long-CLIP's KPS module, which extends CLIP's text processing capacity and achieves fine-grained cross-modal feature alignment through a dual-text loss weighting mechanism. Experimental results show that LRSCLIP improves retrieval accuracy by 10\%-20\% over the Long-CLIP baseline in the zero-shot long-text cross-modal retrieval task. For the zero-shot short-text cross-modal retrieval task, LRSCLIP achieves improvements over the current best model, GeoRSCLIP, with increases of 0.17\%, 0.67\%, and 0.92\% in Text to Image R@1, Image to Text R@1, and mR on RSITMD, respectively, and 0.04\%, 2.93\%, and 1.28\% on RSICD. In the zero-shot image classification task (average accuracy=75.75\%) and semantic localization task (Rmi=0.7653), LRSCLIP achieves state-of-the-art performance. These results validate the dual advantages of fine-grained semantic understanding and global feature matching in LRSCLIP. This work provides a new benchmark model and data support for remote sensing multimodal learning. The related code has been open source and is available at https://github.com/MitsuiChen14/LRSCLIP.

Paperid: 1222, https://arxiv.org/pdf/2503.19012.pdf GitHub

Abstract:
The task of translating visible-to-infrared images (V2IR) is inherently challenging due to three main obstacles: 1) achieving semantic-aware translation, 2) managing the diverse wavelength spectrum in infrared imagery, and 3) the scarcity of comprehensive infrared datasets. Current leading methods tend to treat V2IR as a conventional image-to-image synthesis challenge, often overlooking these specific issues. To address this, we introduce DiffV2IR, a novel framework for image translation comprising two key elements: a Progressive Learning Module (PLM) and a Vision-Language Understanding Module (VLUM). PLM features an adaptive diffusion model architecture that leverages multi-stage knowledge learning to infrared transition from full-range to target wavelength. To improve V2IR translation, VLUM incorporates unified Vision-Language Understanding. We also collected a large infrared dataset, IR-500K, which includes 500,000 infrared images compiled by various scenes and objects under various environmental conditions. Through the combination of PLM, VLUM, and the extensive IR-500K dataset, DiffV2IR markedly improves the performance of V2IR. Experiments validate DiffV2IR's excellence in producing high-quality translations, establishing its efficacy and broad applicability. The code, dataset, and DiffV2IR model will be available at https://github.com/LidongWang-26/DiffV2IR.

Paperid: 1223, https://arxiv.org/pdf/2503.18968.pdf GitHub

Abstract:
In modern medicine, clinical diagnosis relies on the comprehensive analysis of primarily textual and visual data, drawing on medical expertise to ensure systematic and rigorous reasoning. Recent advances in large Vision-Language Models (VLMs) and agent-based methods hold great potential for medical diagnosis, thanks to the ability to effectively integrate multi-modal patient data. However, they often provide direct answers and draw empirical-driven conclusions without quantitative analysis, which reduces their reliability and clinical usability. We propose MedAgent-Pro, a new agentic reasoning paradigm that follows the diagnosis principle in modern medicine, to decouple the process into sequential components for step-by-step, evidence-based reasoning. Our MedAgent-Pro workflow presents a hierarchical diagnostic structure to mirror this principle, consisting of disease-level standardized plan generation and patient-level personalized step-by-step reasoning. To support disease-level planning, an RAG-based agent is designed to retrieve medical guidelines to ensure alignment with clinical standards. For patient-level reasoning, we propose to integrate professional tools such as visual models to enable quantitative assessments. Meanwhile, we propose to verify the reliability of each step to achieve evidence-based diagnosis, enforcing rigorous logical reasoning and a well-founded conclusion. Extensive experiments across a wide range of anatomical regions, imaging modalities, and diseases demonstrate the superiority of MedAgent-Pro to mainstream VLMs, agentic systems and state-of-the-art expert models. Ablation studies and human evaluation by clinical experts further validate its robustness and clinical relevance. Code is available at https://github.com/jinlab-imvr/MedAgent-Pro.

Paperid: 1224, https://arxiv.org/pdf/2503.18854.pdf GitHub

Abstract:
Current vision-language models (VLMs) show exceptional abilities across diverse tasks, such as visual question answering. To enhance user experience, recent studies investigate VLM personalization to understand user-provided concepts. However, they mainly focus on single-concept personalization, neglecting the existence and interplay of multiple concepts, which limits real-world applicability. This paper proposes the first multi-concept personalization paradigm, MC-LLaVA. Specifically, MC-LLaVA employs a multi-concept instruction tuning strategy, effectively integrating multiple concepts in a single training step. To reduce the costs related to joint training, we propose a personalized textual prompt that uses visual token information to initialize concept tokens. Additionally, we introduce a personalized visual prompt during inference, aggregating location confidence maps for enhanced recognition and grounding capabilities. To advance multi-concept personalization research, we further contribute a high-quality instruction tuning dataset. We carefully collect images with multiple characters and objects from movies and manually generate question-answer samples for multi-concept scenarios, featuring superior diversity. Comprehensive qualitative and quantitative experiments demonstrate that MC-LLaVA can achieve impressive multi-concept personalized responses, paving the way for VLMs to become better user-specific assistants. The code and dataset will be publicly available at https://github.com/arctanxarc/MC-LLaVA}.

Paperid: 1225, https://arxiv.org/pdf/2503.18680.pdf GitHub

Abstract:
Efficiently searching for relevant case studies is critical in architectural design, as designers rely on precedent examples to guide or inspire their ongoing projects. However, traditional text-based search tools struggle to capture the inherently visual and complex nature of architectural knowledge, often leading to time-consuming and imprecise exploration. This paper introduces ArchSeek, an innovative case study search system with recommendation capability, tailored for architecture design professionals. Powered by the visual understanding capabilities from vision-language models and cross-modal embeddings, it enables text and image queries with fine-grained control, and interaction-based design case recommendations. It offers architects a more efficient, personalized way to discover design inspirations, with potential applications across other visually driven design fields. The source code is available at https://github.com/danruili/ArchSeek.

Paperid: 1226, https://arxiv.org/pdf/2503.18556.pdf GitHub

Abstract:
Despite the significant success of Large Vision-Language models(LVLMs), these models still suffer hallucinations when describing images, generating answers that include non-existent objects. It is reported that these models tend to over-focus on certain irrelevant image tokens that do not contain critical information for answering the question and distort the output. To address this, we propose an Instruction-Aligned Visual Attention(IAVA) approach, which identifies irrelevant tokens by comparing changes in attention weights under two different instructions. By applying contrastive decoding, we dynamically adjust the logits generated from original image tokens and irrelevant image tokens, reducing the model's over-attention to irrelevant information. The experimental results demonstrate that IAVA consistently outperforms existing decoding techniques on benchmarks such as MME, POPE, and TextVQA in mitigating object hallucinations. Our IAVA approach is available online at https://github.com/Lee-lab558/IAVA.

Paperid: 1227, https://arxiv.org/pdf/2503.18484.pdf GitHub

Abstract:
Existing multilingual benchmarks for Large Vision Language Models (LVLMs) suffer from limitations including language-specific content biases, disjointed multimodal input formats, and a lack of safety evaluation. To address these gaps, we propose PM4Bench, the first Parallel Multilingual Multi-Modal Multi-task Benchmark for LVLMs. PM4Bench features a parallel corpus design across 10 languages, enabling fair and accurate cross-lingual comparisons. It includes the vision setting where text and queries are embedded in images, requiring LVLMs to simultaneously "see", "read", and "think", aligning with real-world applications. Additionally, PM\textsuperscript{4}Bench incorporates safety evaluations, addressing critical oversight in existing multilingual benchmarks. Using PM4Bench, we evaluate 11 mainstream LVLMs, revealing significant cross-linguistic performance disparities, particularly in vision settings, and identifying OCR capability as a key determinant of these imbalances. We will release PM4Bench at https://github.com/opendatalab/PM4Bench .

Paperid: 1228, https://arxiv.org/pdf/2503.18483.pdf GitHub

Abstract:
Concept-based models can map black-box representations to human-understandable concepts, which makes the decision-making process more transparent and then allows users to understand the reason behind predictions. However, domain-specific concepts often impact the final predictions, which subsequently undermine the model generalization capabilities, and prevent the model from being used in high-stake applications. In this paper, we propose a novel Language-guided Concept-Erasing (LanCE) framework. In particular, we empirically demonstrate that pre-trained vision-language models (VLMs) can approximate distinct visual domain shifts via domain descriptors while prompting large Language Models (LLMs) can easily simulate a wide range of descriptors of unseen visual domains. Then, we introduce a novel plug-in domain descriptor orthogonality (DDO) regularizer to mitigate the impact of these domain-specific concepts on the final predictions. Notably, the DDO regularizer is agnostic to the design of concept-based models and we integrate it into several prevailing models. Through evaluation of domain generalization on four standard benchmarks and three newly introduced benchmarks, we demonstrate that DDO can significantly improve the out-of-distribution (OOD) generalization over the previous state-of-the-art concept-based models.Our code is available at https://github.com/joeyz0z/LanCE.

Paperid: 1229, https://arxiv.org/pdf/2503.18476.pdf GitHub

Abstract:
Large Vision-Language Models (VLMs), such as GPT-4, have achieved remarkable success across various fields. However, there are few studies on 3D indoor scene generation with VLMs. This paper considers this task as a planning problem subject to spatial and layout common sense constraints. To solve the problem with a VLM, we propose a new global-local tree search algorithm. Globally, the method places each object sequentially and explores multiple placements during each placement process, where the problem space is represented as a tree. To reduce the depth of the tree, we decompose the scene structure hierarchically, i.e. room level, region level, floor object level, and supported object level. The algorithm independently generates the floor objects in different regions and supported objects placed on different floor objects. Locally, we also decompose the sub-task, the placement of each object, into multiple steps. The algorithm searches the tree of problem space. To leverage the VLM model to produce positions of objects, we discretize the top-down view space as a dense grid and fill each cell with diverse emojis to make to cells distinct. We prompt the VLM with the emoji grid and the VLM produces a reasonable location for the object by describing the position with the name of emojis. The quantitative and qualitative experimental results illustrate our approach generates more plausible 3D scenes than state-of-the-art approaches. Our source code is available at https://github.com/dw-dengwei/TreeSearchGen .

Paperid: 1230, https://arxiv.org/pdf/2503.18470.pdf GitHub

Abstract:
We present MetaSpatial, the first reinforcement learning (RL)-based framework designed to enhance 3D spatial reasoning in vision-language models (VLMs), enabling real-time 3D scene generation without the need for hard-coded optimizations. MetaSpatial addresses two core challenges: (i) the lack of internalized 3D spatial reasoning in VLMs, which limits their ability to generate realistic layouts, and (ii) the inefficiency of traditional supervised fine-tuning (SFT) for layout generation tasks, as perfect ground truth annotations are unavailable. Our key innovation is a multi-turn RL-based optimization mechanism that integrates physics-aware constraints and rendered image evaluations, ensuring generated 3D layouts are coherent, physically plausible, and aesthetically consistent. Methodologically, MetaSpatial introduces an adaptive, iterative reasoning process, where the VLM refines spatial arrangements over multiple turns by analyzing rendered outputs, improving scene coherence progressively. Empirical evaluations demonstrate that MetaSpatial significantly enhances the spatial consistency and formatting stability of various scale models. Post-training, object placements are more realistic, aligned, and functionally coherent, validating the effectiveness of RL for 3D spatial reasoning in metaverse, AR/VR, digital twins, and game development applications. Our code, data, and training pipeline are publicly available at https://github.com/PzySeere/MetaSpatial.

Paperid: 1231, https://arxiv.org/pdf/2503.18435.pdf GitHub

Abstract:
Chart understanding requires models to effectively analyze and reason about numerical data, textual elements, and complex visual components. Our observations reveal that the perception capabilities of existing large vision-language models (LVLMs) constitute a critical bottleneck in this process. In this study, we delve into this perception bottleneck by decomposing it into two components: the vision encoder bottleneck, where the visual representation may fail to encapsulate the correct information, and the extraction bottleneck, where the language model struggles to extract the necessary information from the provided visual representations. Through comprehensive experiments, we find that (1) the information embedded within visual representations is substantially richer than what is typically captured by linear extractors, such as the widely used retrieval accuracy metric; (2) While instruction tuning effectively enhances the extraction capability of LVLMs, the vision encoder remains a critical bottleneck, demanding focused attention and improvement. Therefore, we further enhance the visual encoder to mitigate the vision encoder bottleneck under a contrastive learning framework. Empirical results demonstrate that our approach significantly mitigates the perception bottleneck and improves the ability of LVLMs to comprehend charts. Code is publicly available at https://github.com/hkust-nlp/Vision4Chart.

Paperid: 1232, https://arxiv.org/pdf/2503.18407.pdf GitHub

Abstract:
Vision-language models bridge visual and linguistic understanding and have proven to be powerful for video recognition tasks. Existing approaches primarily rely on parameter-efficient fine-tuning of image-text pre-trained models, yet they often suffer from limited interpretability and poor generalization due to inadequate temporal modeling. To address these, we propose a simple yet effective video-to-text discretization framework. Our method repurposes the frozen text encoder to construct a visual codebook from video class labels due to the many-to-one contrastive alignment between visual and textual embeddings in multimodal pretraining. This codebook effectively transforms temporal visual data into textual tokens via feature lookups and offers interpretable video representations through explicit video modeling. Then, to enhance robustness against irrelevant or noisy frames, we introduce a confidence-aware fusion module that dynamically weights keyframes by assessing their semantic relevance via the codebook. Furthermore, our method incorporates learnable text prompts to conduct adaptive codebook updates. Extensive experiments on HMDB-51, UCF-101, SSv2, and Kinetics-400 have validated the superiority of our approach, achieving more competitive improvements over state-of-the-art methods. The code will be publicly available at https://github.com/isxinxin/VTD-CLIP.

Paperid: 1233, https://arxiv.org/pdf/2503.18349.pdf GitHub

Abstract:
Human-object interaction (HOI) synthesis is crucial for applications in animation, simulation, and robotics. However, existing approaches either rely on expensive motion capture data or require manual reward engineering, limiting their scalability and generalizability. In this work, we introduce the first unified physics-based HOI framework that leverages Vision-Language Models (VLMs) to enable long-horizon interactions with diverse object types, including static, dynamic, and articulated objects. We introduce VLM-Guided Relative Movement Dynamics (RMD), a fine-grained spatio-temporal bipartite representation that automatically constructs goal states and reward functions for reinforcement learning. By encoding structured relationships between human and object parts, RMD enables VLMs to generate semantically grounded, interaction-aware motion guidance without manual reward tuning. To support our methodology, we present Interplay, a novel dataset with thousands of long-horizon static and dynamic interaction plans. Extensive experiments demonstrate that our framework outperforms existing methods in synthesizing natural, human-like motions across both simple single-task and complex multi-task scenarios. For more details, please refer to our project webpage: https://vlm-rmd.github.io/.

Paperid: 1234, https://arxiv.org/pdf/2503.18160.pdf GitHub

Abstract:
Though CLIP-based prompt tuning significantly enhances pre-trained Vision-Language Models, existing research focuses on reconstructing the model architecture, e.g., additional loss calculation and meta-networks. These approaches generally lead to increased complexity and extended training cost. To maintain the efficiency of the tuning process, we propose plug-and-play Model-Agnostic Optimization (MAO) for prompt tuning. Without altering any components of the prompt tuning backbone, we introduce a Data-Driven Enhancement framework to optimize the distribution of the initial data, and incorporate an Alterable Regularization module to boost the task-specific feature processing pipeline, thereby improving overall performance while maintaining low computational cost. Extensive experiments on MAO demonstrate its outstanding performance and efficiency. The code of MAO is available at: https://github.com/JREion/M.A.O .

Paperid: 1235, https://arxiv.org/pdf/2503.18065.pdf GitHub

Abstract:
Data scarcity is a long-standing challenge in the Vision-Language Navigation (VLN) field, which extremely hinders the generalization of agents to unseen environments. Previous works primarily rely on additional simulator data or web-collected images/videos to improve the generalization. However, the simulator environments still face limited diversity, and the web-collected data often requires extensive labor to remove the noise. In this paper, we propose a Rewriting-driven AugMentation (RAM) paradigm for VLN, which directly creates the unseen observation-instruction pairs via rewriting human-annotated training data. Benefiting from our rewriting mechanism, new observation-instruction can be obtained in both simulator-free and labor-saving manners to promote generalization. Specifically, we first introduce Object-Enriched Observation Rewriting, where we combine Vision-Language Models (VLMs) and Large Language Models (LLMs) to derive rewritten object-enriched scene descriptions, enabling observation synthesis with diverse objects and spatial layouts via Text-to-Image Generation Models (T2IMs). Then, we propose Observation-Contrast Instruction Rewriting, which generates observation-aligned rewritten instructions by requiring LLMs to reason the difference between original and new observations. We further develop a mixing-then-focusing training strategy with a random observation cropping scheme, effectively enhancing data distribution diversity while suppressing augmentation data noise during training. Experiments on both the discrete environments (R2R, REVERIE, and R4R datasets) and continuous environments (R2R-CE dataset) show the superior performance and impressive generalization ability of our method. Code is available at https://github.com/SaDil13/VLN-RAM.

Paperid: 1236, https://arxiv.org/pdf/2503.18052.pdf GitHub

Abstract:
Recognizing arbitrary or previously unseen categories is essential for comprehensive real-world 3D scene understanding. Currently, all existing methods rely on 2D or textual modalities during training or together at inference. This highlights the clear absence of a model capable of processing 3D data alone for learning semantics end-to-end, along with the necessary data to train such a model. Meanwhile, 3D Gaussian Splatting (3DGS) has emerged as the de facto standard for 3D scene representation across various vision tasks. However, effectively integrating semantic reasoning into 3DGS in a generalizable manner remains an open challenge. To address these limitations, we introduce SceneSplat, to our knowledge the first large-scale 3D indoor scene understanding approach that operates natively on 3DGS. Furthermore, we propose a self-supervised learning scheme that unlocks rich 3D feature learning from unlabeled scenes. To power the proposed methods, we introduce SceneSplat-7K, the first large-scale 3DGS dataset for indoor scenes, comprising 7916 scenes derived from seven established datasets, such as ScanNet and Matterport3D. Generating SceneSplat-7K required computational resources equivalent to 150 GPU days on an L4 GPU, enabling standardized benchmarking for 3DGS-based reasoning for indoor scenes. Our exhaustive experiments on SceneSplat-7K demonstrate the significant benefit of the proposed method over the established baselines.

Paperid: 1237, https://arxiv.org/pdf/2503.18013.pdf GitHub

Abstract:
Large Vision-Language Models (LVLMs) typically follow a two-stage training paradigm-pretraining and supervised fine-tuning. Recently, preference optimization, derived from the language domain, has emerged as an effective post-training reinforcement strategy to enhance capabilities of LVLMs. However, constructing high-quality human-annotated preference data and developing robust reward models to mimic these preferences are both costly and challenging. Motivated by this observation, we propose Vision-R1, a novel vision-guided R1-like reinforcement learning algorithm for LVLMs that rewards models with definitive vision feedback. It only leverages curated instruction data, eliminating the need for specialized reward models and handcrafted preference datasets. We incorporate a criterion-driven reward function that further integrates multi-dimensional feedback to evaluate model completions comprehensively based on the vision task logic. Furthermore, we introduce a progressive rule refinement strategy that dynamically adjusts the reward criteria during training, enabling continuous model improvement and mitigating reward hacking. Extensive experiments on both in-distribution and out-of-distribution benchmarks demonstrate that fine-tuning the 7B LVLMs with Vision-R1 achieves consistent performance gains, with even up to 50% improvement and surpassing the state-of-the-art 10x size model.

Paperid: 1238, https://arxiv.org/pdf/2503.17736.pdf GitHub

Abstract:
Large Vision-Language Models (LVLMs) have made significant progress in the field of video understanding recently. However, current benchmarks uniformly lean on text prompts for evaluation, which often necessitate complex referential language and fail to provide precise spatial and temporal references. This limitation diminishes the experience and efficiency of human-model interaction. To address this limitation, we propose the Video Visual Prompt Benchmark(V2P-Bench), a comprehensive benchmark specifically designed to evaluate LVLMs' video understanding capabilities in multimodal human-model interaction scenarios. V2P-Bench includes 980 unique videos and 1,172 QA pairs, covering 5 main tasks and 12 dimensions, facilitating instance-level fine-grained understanding aligned with human cognition. Benchmarking results reveal that even the most powerful models perform poorly on V2P-Bench (65.4% for GPT-4o and 67.9% for Gemini-1.5-Pro), significantly lower than the human experts' 88.3%, highlighting the current shortcomings of LVLMs in understanding video visual prompts. We hope V2P-Bench will serve as a foundation for advancing multimodal human-model interaction and video understanding evaluation. Project page: https://github.com/gaotiexinqu/V2P-Bench.

Paperid: 1239, https://arxiv.org/pdf/2503.17489.pdf GitHub

Abstract:
Evaluating generative foundation models on open-ended multimodal understanding (MMU) and generation (MMG) tasks across diverse modalities (e.g., images, audio, video) poses significant challenges due to the complexity of cross-modal interactions. To this end, the idea of utilizing Multimodal LLMs (MLLMs) as automated judges has emerged, with encouraging results in assessing vision-language understanding tasks. Moving further, this paper extends MLLM-as-a-Judge across modalities to a unified manner by introducing two benchmarks, TaskAnything and JudgeAnything, to respectively evaluate the overall performance and judging capabilities of MLLMs across any-to-any modality tasks. Specifically, TaskAnything evaluates the MMU and MMG capabilities across 15 any-to-any modality categories, employing 1,500 queries curated from well-established benchmarks. Furthermore, JudgeAnything evaluates the judging capabilities of 5 advanced (e.g., GPT-4o and Gemini-2.0-Flash) from the perspectives of Pair Comparison and Score Evaluation, providing a standardized testbed that incorporates human judgments and detailed rubrics. Our extensive experiments reveal that while these MLLMs show promise in assessing MMU (i.e., achieving an average of 66.55% in Pair Comparison setting and 42.79% in Score Evaluation setting), they encounter significant challenges with MMG tasks (i.e., averaging only 53.37% in Pair Comparison setting and 30.05% in Score Evaluation setting), exposing cross-modality biases and hallucination issues. To address this, we present OmniArena, an automated platform for evaluating omni-models and multimodal reward models. Our work highlights the need for fairer evaluation protocols and stronger alignment with human preferences. The source code and dataset are publicly available at: https://urrealhero.github.io/judgeanythingweb/.

Paperid: 1240, https://arxiv.org/pdf/2503.17406.pdf GitHub

Abstract:
With the recent rise of large language models, vision-language models, and other general foundation models, there is growing potential for multimodal, multi-task robotics that can operate in diverse environments given natural language input. One such application is indoor navigation using natural language instructions. However, despite recent progress, this problem remains challenging due to the 3D spatial reasoning and semantic understanding required. Additionally, the language used may be imperfect or misaligned with the scene, further complicating the task. To address this challenge, we curate a benchmark dataset, IRef-VLA, for Interactive Referential Vision and Language-guided Action in 3D Scenes with imperfect references. IRef-VLA is the largest real-world dataset for the referential grounding task, consisting of over 11.5K scanned 3D rooms from existing datasets, 7.6M heuristically generated semantic relations, and 4.7M referential statements. Our dataset also contains semantic object and room annotations, scene graphs, navigable free space annotations, and is augmented with statements where the language has imperfections or ambiguities. We verify the generalizability of our dataset by evaluating with state-of-the-art models to obtain a performance baseline and also develop a graph-search baseline to demonstrate the performance bound and generation of alternatives using scene-graph knowledge. With this benchmark, we aim to provide a resource for 3D scene understanding that aids the development of robust, interactive navigation systems. The dataset and all source code is publicly released at https://github.com/HaochenZ11/IRef-VLA.

Paperid: 1241, https://arxiv.org/pdf/2503.17353.pdf GitHub

Abstract:
Many high-impact machine learning tasks involve multi-dimensional data such as images, volumetric medical scans, and multivariate time-series. Yet, most neural architectures flatten these inputs, discarding critical cross-dimension information. We introduce $\textbf{NdLinear}$, a novel linear transformation that circumvents this destructive flattening by operating directly on tensors. NdLinear applies transformations separately along each data dimension, thereby preserving the native data structure. Extensive experiments demonstrate NdLinear's capacity to significantly enhance representational power, achieve dramatic parameter reductions (often by orders of magnitude), and maintain a favorable computational profile. For instance, when applied to Large Language Model finetuning, our $\textbf{NdLinear-LoRA}$ delivers comparable or improved accuracy on reasoning tasks using up to $9\times$ fewer trainable parameters than standard LoRA. These broad advantages of NdLinear are consistently validated across diverse neural architectures (CNNs, RNNs, Transformers, MLPs) and data domains, including vision, language, time-series, and tabular tasks. As a versatile, drop-in replacement for standard linear layers, NdLinear processes data in its original N-dimensional form, offering a foundational component for developing more efficient and powerful next-generation neural architectures.

Paperid: 1242, https://arxiv.org/pdf/2503.17352.pdf GitHub

Abstract:
We introduce OpenVLThinker, one of the first open-source large vision-language models (LVLMs) to exhibit sophisticated chain-of-thought reasoning, achieving notable performance gains on challenging visual reasoning tasks. While text-based reasoning models (e.g., Deepseek R1) show promising results in text-only tasks, distilling their reasoning into LVLMs via supervised fine-tuning (SFT) often results in performance degradation due to imprecise visual grounding. Conversely, purely reinforcement learning (RL)-based methods face a large search space, hindering the emergence of reflective behaviors in smaller models (e.g., 7B LVLMs). Surprisingly, alternating between SFT and RL ultimately results in significant performance improvements after a few iterations. Our analysis reveals that the base model rarely exhibits reasoning behaviors initially, but SFT effectively surfaces these latent actions and narrows the RL search space, accelerating the development of reasoning capabilities. Each subsequent RL stage further refines the model's reasoning skills, producing higher-quality SFT data for continued self-improvement. OpenVLThinker-7B consistently advances performance across six benchmarks demanding mathematical and general reasoning, notably improving MathVista by 3.8%, EMMA by 2.4%, and HallusionBench by 1.6%. Beyond demonstrating the synergy between SFT and RL for complex reasoning tasks, our findings provide early evidence towards achieving R1-style reasoning in multimodal contexts. The code, model and data are held at https://github.com/yihedeng9/OpenVLThinker.

Paperid: 1243, https://arxiv.org/pdf/2503.17238.pdf GitHub

Abstract:
In this paper, we address the challenge of few-shot classification in histopathology whole slide images (WSIs) by utilizing foundational vision-language models (VLMs) and slide-level prompt learning. Given the gigapixel scale of WSIs, conventional multiple instance learning (MIL) methods rely on aggregation functions to derive slide-level (bag-level) predictions from patch representations, which require extensive bag-level labels for training. In contrast, VLM-based approaches excel at aligning visual embeddings of patches with candidate class text prompts but lack essential pathological prior knowledge. Our method distinguishes itself by utilizing pathological prior knowledge from language models to identify crucial local tissue types (patches) for WSI classification, integrating this within a VLM-based MIL framework. Our approach effectively aligns patch images with tissue types, and we fine-tune our model via prompt learning using only a few labeled WSIs per category. Experimentation on real-world pathological WSI datasets and ablation studies highlight our method's superior performance over existing MIL- and VLM-based methods in few-shot WSI classification tasks. Our code is publicly available at https://github.com/LTS5/SLIP.

Paperid: 1244, https://arxiv.org/pdf/2503.17142.pdf GitHub

Abstract:
Vision-Language Models (VLMs) learn a shared feature space for text and images, enabling the comparison of inputs of different modalities. While prior works demonstrated that VLMs organize natural language representations into regular structures encoding composite meanings, it remains unclear if compositional patterns also emerge in the visual embedding space. In this work, we investigate compositionality in the image domain, where the analysis of compositional properties is challenged by noise and sparsity of visual data. We address these problems and propose a framework, called Geodesically Decomposable Embeddings (GDE), that approximates image representations with geometry-aware compositional structures in the latent space. We demonstrate that visual embeddings of pre-trained VLMs exhibit a compositional arrangement, and evaluate the effectiveness of this property in the tasks of compositional classification and group robustness. GDE achieves stronger performance in compositional classification compared to its counterpart method that assumes linear geometry of the latent space. Notably, it is particularly effective for group robustness, where we achieve higher results than task-specific solutions. Our results indicate that VLMs can automatically develop a human-like form of compositional reasoning in the visual domain, making their underlying processes more interpretable. Code is available at https://github.com/BerasiDavide/vlm_image_compositionality.

Paperid: 1245, https://arxiv.org/pdf/2503.17110.pdf GitHub

Abstract:
Deep learning has become an essential part of computer vision, with deep neural networks (DNNs) excelling in predictive performance. However, they often fall short in other critical quality dimensions, such as robustness, calibration, or fairness. While existing studies have focused on a subset of these quality dimensions, none have explored a more general form of "well-behavedness" of DNNs. With this work, we address this gap by simultaneously studying nine different quality dimensions for image classification. Through a large-scale study, we provide a bird's-eye view by analyzing 326 backbone models and how different training paradigms and model architectures affect the quality dimensions. We reveal various new insights such that (i) vision-language models exhibit high fairness on ImageNet-1k classification and strong robustness against domain changes; (ii) self-supervised learning is an effective training paradigm to improve almost all considered quality dimensions; and (iii) the training dataset size is a major driver for most of the quality dimensions. We conclude our study by introducing the QUBA score (Quality Understanding Beyond Accuracy), a novel metric that ranks models across multiple dimensions of quality, enabling tailored recommendations based on specific user needs.

Paperid: 1246, https://arxiv.org/pdf/2503.16945.pdf GitHub

Abstract:
Vision-Language Models (VLMs) like CLIP offer promising solutions for Dynamic Facial Expression Recognition (DFER) but face challenges such as inefficient full fine-tuning, high complexity, and poor alignment between textual and visual representations. Additionally, existing methods struggle with ineffective temporal modeling. To address these issues, we propose PE-CLIP, a parameter-efficient fine-tuning (PEFT) framework that adapts CLIP for DFER while significantly reducing trainable parameters while maintaining high accuracy. PE-CLIP introduces two specialized adapters: a Temporal Dynamic Adapter (TDA) and a Shared Adapter (ShA). The TDA is a GRU-based module with dynamic scaling that captures sequential dependencies while emphasizing informative temporal features and suppressing irrelevant variations. The ShA is a lightweight adapter that refines representations within both textual and visual encoders, ensuring consistency and efficiency. Additionally, we integrate Multi-modal Prompt Learning (MaPLe), introducing learnable prompts for visual and action unit-based textual inputs, enhancing semantic alignment between modalities and enabling efficient CLIP adaptation for dynamic tasks. We evaluate PE-CLIP on two benchmark datasets, DFEW and FERV39K, achieving competitive performance compared to state-of-the-art methods while requiring fewer trainable parameters. By balancing efficiency and accuracy, PE-CLIP sets a new benchmark in resource-efficient DFER. The source code of the proposed PE-CLIP will be publicly available at https://github.com/Ibtissam-SAADI/PE-CLIP .

Paperid: 1247, https://arxiv.org/pdf/2503.16929.pdf GitHub

Abstract:
Video Large Language Models (Video LLMs) have achieved significant success by leveraging a two-stage paradigm: pretraining on large-scale video-text data for vision-language alignment, followed by supervised fine-tuning (SFT) for task-specific capabilities. However, existing approaches struggle with temporal reasoning due to weak temporal correspondence in the data and reliance on the next-token prediction paradigm during training. To address these limitations, we propose TEMPLE (TEMporal Preference Learning), a systematic framework that enhances Video LLMs' temporal reasoning capabilities through Direct Preference Optimization (DPO). To facilitate this, we introduce an automated preference data generation pipeline that systematically constructs preference pairs by selecting videos that are rich in temporal information, designing video-specific perturbation strategies, and finally evaluating model responses on clean and perturbed video inputs. Our temporal alignment features two key innovations: curriculum learning which that progressively increases perturbation difficulty to improve model robustness and adaptability; and "Pre-SFT Alignment'', applying preference optimization before instruction tuning to prioritize fine-grained temporal comprehension. Extensive experiments demonstrate that our approach consistently improves Video LLM performance across multiple benchmarks with a relatively small set of self-generated DPO data. We further analyze the transferability of DPO data across architectures and the role of difficulty scheduling in optimization. Our findings highlight our TEMPLE as a scalable and efficient complement to SFT-based methods, paving the way for developing reliable Video LLMs. Code is available at https://github.com/lscpku/TEMPLE.

Paperid: 1248, https://arxiv.org/pdf/2503.16873.pdf GitHub

Abstract:
Multi-label classification is crucial for comprehensive image understanding, yet acquiring accurate annotations is challenging and costly. To address this, a recent study suggests exploiting unsupervised multi-label classification leveraging CLIP, a powerful vision-language model. Despite CLIP's proficiency, it suffers from view-dependent predictions and inherent bias, limiting its effectiveness. We propose a novel method that addresses these issues by leveraging multiple views near target objects, guided by Class Activation Mapping (CAM) of the classifier, and debiasing pseudo-labels derived from CLIP predictions. Our Classifier-guided CLIP Distillation (CCD) enables selecting multiple local views without extra labels and debiasing predictions to enhance classification performance. Experimental results validate our method's superiority over existing techniques across diverse datasets. The code is available at https://github.com/k0u-id/CCD.

Paperid: 1249, https://arxiv.org/pdf/2503.16788.pdf GitHub

Abstract:
Reasoning capabilities have significantly improved the performance of vision-language models (VLMs) in domains such as mathematical problem-solving, coding, and visual question-answering. However, their impact on real-world applications remains unclear. This paper presents the first empirical study on the effectiveness of reasoning-enabled VLMs in mobile GUI agents, a domain that requires interpreting complex screen layouts, understanding user instructions, and executing multi-turn interactions. We evaluate two pairs of commercial models--Gemini 2.0 Flash and Claude 3.7 Sonnet--comparing their base and reasoning-enhanced versions across two static benchmarks (ScreenSpot and AndroidControl) and one interactive environment (AndroidWorld). We surprisingly find the Claude 3.7 Sonnet reasoning model achieves state-of-the-art performance on AndroidWorld. However, reasoning VLMs generally offer marginal improvements over non-reasoning models on static benchmarks and even degrade performance in some agent setups. Notably, reasoning and non-reasoning VLMs fail on different sets of tasks, suggesting that reasoning does have an impact, but its benefits and drawbacks counterbalance each other. We attribute these inconsistencies to the limitations of benchmarks and VLMs. Based on the findings, we provide insights for further enhancing mobile GUI agents in terms of benchmarks, VLMs, and their adaptability in dynamically invoking reasoning VLMs. The experimental data are publicly available at https://github.com/LlamaTouch/VLM-Reasoning-Traces.

Paperid: 1250, https://arxiv.org/pdf/2503.16707.pdf GitHub

Abstract:
The lack of a large-scale 3D-text corpus has led recent works to distill open-vocabulary knowledge from vision-language models (VLMs). However, these methods typically rely on a single VLM to align the feature spaces of 3D models within a common language space, which limits the potential of 3D models to leverage the diverse spatial and semantic capabilities encapsulated in various foundation models. In this paper, we propose Cross-modal and Uncertainty-aware Agglomeration for Open-vocabulary 3D Scene Understanding dubbed CUA-O3D, the first model to integrate multiple foundation models-such as CLIP, DINOv2, and Stable Diffusion-into 3D scene understanding. We further introduce a deterministic uncertainty estimation to adaptively distill and harmonize the heterogeneous 2D feature embeddings from these models. Our method addresses two key challenges: (1) incorporating semantic priors from VLMs alongside the geometric knowledge of spatially-aware vision foundation models, and (2) using a novel deterministic uncertainty estimation to capture model-specific uncertainties across diverse semantic and geometric sensitivities, helping to reconcile heterogeneous representations during training. Extensive experiments on ScanNetV2 and Matterport3D demonstrate that our method not only advances open-vocabulary segmentation but also achieves robust cross-domain alignment and competitive spatial perception capabilities. The code will be available at: https://github.com/TyroneLi/CUA_O3D.

Paperid: 1251, https://arxiv.org/pdf/2503.16413.pdf GitHub

Abstract:
We present 3D Spatial MultiModal Memory (M3), a multimodal memory system designed to retain information about medium-sized static scenes through video sources for visual perception. By integrating 3D Gaussian Splatting techniques with foundation models, M3 builds a multimodal memory capable of rendering feature representations across granularities, encompassing a wide range of knowledge. In our exploration, we identify two key challenges in previous works on feature splatting: (1) computational constraints in storing high-dimensional features for each Gaussian primitive, and (2) misalignment or information loss between distilled features and foundation model features. To address these challenges, we propose M3 with key components of principal scene components and Gaussian memory attention, enabling efficient training and inference. To validate M3, we conduct comprehensive quantitative evaluations of feature similarity and downstream tasks, as well as qualitative visualizations to highlight the pixel trace of Gaussian memory attention. Our approach encompasses a diverse range of foundation models, including vision-language models (VLMs), perception models, and large multimodal and language models (LMMs/LLMs). Furthermore, to demonstrate real-world applicability, we deploy M3's feature field in indoor scenes on a quadruped robot. Notably, we claim that M3 is the first work to address the core compression challenges in 3D feature distillation.

Paperid: 1252, https://arxiv.org/pdf/2503.16365.pdf GitHub

Abstract:
Recently, action-based decision-making in open-world environments has gained significant attention. Visual Language Action (VLA) models, pretrained on large-scale web datasets, have shown promise in decision-making tasks. However, previous work has primarily focused on action post-training, often neglecting enhancements to the foundational model itself. In response, we introduce a novel approach, Act from Visual Language Post-Training, which refines Visual Language Models (VLMs) through visual and linguistic guidance in a self-supervised manner. This enhancement improves the models' capabilities in world knowledge, visual recognition, and spatial grounding in open-world environments. Following the above post-training paradigms, we obtain the first VLA models in Minecraft that can follow human instructions on over 1k different atomic tasks, including crafting, smelting, cooking, mining, and killing. Our experiments demonstrate that post-training on non-trajectory tasks leads to a significant 40% improvement over the best agent baseline on a diverse set of atomic tasks. Furthermore, we demonstrate that our approach surpasses traditional imitation learning-based policies in Minecraft, achieving state-of-the-art performance. We have open-sourced the code, models, and datasets to foster further research. The project page can be found in https://craftjarvis.github.io/JarvisVLA.

Paperid: 1253, https://arxiv.org/pdf/2503.16282.pdf GitHub

Abstract:
Generalized few-shot 3D point cloud segmentation (GFS-PCS) adapts models to new classes with few support samples while retaining base class segmentation. Existing GFS-PCS methods enhance prototypes via interacting with support or query features but remain limited by sparse knowledge from few-shot samples. Meanwhile, 3D vision-language models (3D VLMs), generalizing across open-world novel classes, contain rich but noisy novel class knowledge. In this work, we introduce a GFS-PCS framework that synergizes dense but noisy pseudo-labels from 3D VLMs with precise yet sparse few-shot samples to maximize the strengths of both, named GFS-VL. Specifically, we present a prototype-guided pseudo-label selection to filter low-quality regions, followed by an adaptive infilling strategy that combines knowledge from pseudo-label contexts and few-shot samples to adaptively label the filtered, unlabeled areas. Additionally, we design a novel-base mix strategy to embed few-shot samples into training scenes, preserving essential context for improved novel class learning. Moreover, recognizing the limited diversity in current GFS-PCS benchmarks, we introduce two challenging benchmarks with diverse novel classes for comprehensive generalization evaluation. Experiments validate the effectiveness of our framework across models and datasets. Our approach and benchmarks provide a solid foundation for advancing GFS-PCS in the real world. The code is at https://github.com/ZhaochongAn/GFS-VL

Paperid: 1254, https://arxiv.org/pdf/2503.16120.pdf GitHub

Abstract:
Multi-species animal pose estimation has emerged as a challenging yet critical task, hindered by substantial visual diversity and uncertainty. This paper challenges the problem by efficient prompt learning for Vision-Language Pretrained (VLP) models, \textit{e.g.} CLIP, aiming to resolve the cross-species generalization problem. At the core of the solution lies in the prompt designing, probabilistic prompt modeling and cross-modal adaptation, thereby enabling prompts to compensate for cross-modal information and effectively overcome large data variances under unbalanced data distribution. To this end, we propose a novel probabilistic prompting approach to fully explore textual descriptions, which could alleviate the diversity issues caused by long-tail property and increase the adaptability of prompts on unseen category instance. Specifically, we first introduce a set of learnable prompts and propose a diversity loss to maintain distinctiveness among prompts, thus representing diverse image attributes. Diverse textual probabilistic representations are sampled and used as the guidance for the pose estimation. Subsequently, we explore three different cross-modal fusion strategies at spatial level to alleviate the adverse impacts of visual uncertainty. Extensive experiments on multi-species animal pose benchmarks show that our method achieves the state-of-the-art performance under both supervised and zero-shot settings. The code is available at https://github.com/Raojiyong/PPAP.

Paperid: 1255, https://arxiv.org/pdf/2503.15973.pdf GitHub

Abstract:
Pre-trained on tremendous image-text pairs, vision-language models like CLIP have demonstrated promising zero-shot generalization across numerous image-based tasks. However, extending these capabilities to video tasks remains challenging due to limited labeled video data and high training costs. Recent video prompting methods attempt to adapt CLIP for video tasks by introducing learnable prompts, but they typically rely on a single static prompt for all video sequences, overlooking the diverse temporal dynamics and spatial variations that exist across frames. This limitation significantly hinders the model's ability to capture essential temporal information for effective video understanding. To address this, we propose an integrated Spatial-TempOral dynamic Prompting (STOP) model which consists of two complementary modules, the intra-frame spatial prompting and inter-frame temporal prompting. Our intra-frame spatial prompts are designed to adaptively highlight discriminative regions within each frame by leveraging intra-frame attention and temporal variation, allowing the model to focus on areas with substantial temporal dynamics and capture fine-grained spatial details. Additionally, to highlight the varying importance of frames for video understanding, we further introduce inter-frame temporal prompts, dynamically inserting prompts between frames with high temporal variance as measured by frame similarity. This enables the model to prioritize key frames and enhances its capacity to understand temporal dependencies across sequences. Extensive experiments on various video benchmarks demonstrate that STOP consistently achieves superior performance against state-of-the-art methods. The code is available at https://github.com/zhoujiahuan1991/CVPR2025-STOP.

Paperid: 1256, https://arxiv.org/pdf/2503.15969.pdf GitHub

Abstract:
Vision-language models for Earth observation (EO) typically rely on the visual spectrum of data as the only model input, thus failing to leverage the rich spectral information available in the multispectral channels recorded by satellites. Therefore, we introduce Llama3-MS-CLIP, the first vision-language model pre-trained with contrastive learning on a large-scale multispectral dataset and report on the performance gains due to the extended spectral range. Furthermore, we present the largest-to-date image-caption dataset for multispectral data, consisting of one million Sentinel-2 samples and corresponding textual descriptions generated using Llama3-LLaVA-Next and Overture Maps data. We develop a scalable captioning pipeline, which is validated by domain experts. We evaluate Llama3-MS-CLIP on multispectral zero-shot image classification and retrieval using three datasets of varying complexity. Our results demonstrate that Llama3-MS-CLIP significantly outperforms other RGB-based approaches, improving classification accuracy by +6.77% on average and retrieval performance by +4.63% mAP compared to the second-best model. Our results emphasize the relevance of multispectral vision-language learning. The image-caption dataset, code, and model weights are available at https://github.com/IBM/MS-CLIP.

Paperid: 1257, https://arxiv.org/pdf/2503.15949.pdf GitHub

Abstract:
Referring medical image segmentation targets delineating lesions indicated by textual descriptions. Aligning visual and textual cues is challenging due to their distinct data properties. Inspired by large-scale pre-trained vision-language models, we propose CausalCLIPSeg, an end-to-end framework for referring medical image segmentation that leverages CLIP. Despite not being trained on medical data, we enforce CLIP's rich semantic space onto the medical domain by a tailored cross-modal decoding method to achieve text-to-pixel alignment. Furthermore, to mitigate confounding bias that may cause the model to learn spurious correlations instead of meaningful causal relationships, CausalCLIPSeg introduces a causal intervention module which self-annotates confounders and excavates causal features from inputs for segmentation judgments. We also devise an adversarial min-max game to optimize causal features while penalizing confounding ones. Extensive experiments demonstrate the state-of-the-art performance of our proposed method. Code is available at https://github.com/WUTCM-Lab/CausalCLIPSeg.

Paperid: 1258, https://arxiv.org/pdf/2503.15940.pdf GitHub

Abstract:
Automated radiology report generation aims to expedite the tedious and error-prone reporting process for radiologists. While recent works have made progress, learning to align medical images and textual findings remains challenging due to the relative scarcity of labeled medical data. For example, datasets for this task are much smaller than those used for image captioning in computer vision. In this work, we propose to transfer representations from CLIP, a large-scale pre-trained vision-language model, to better capture cross-modal semantics between images and texts. However, directly applying CLIP is suboptimal due to the domain gap between natural images and radiology. To enable efficient adaptation, we introduce UniCrossAdapter, lightweight adapter modules that are incorporated into CLIP and fine-tuned on the target task while keeping base parameters fixed. The adapters are distributed across modalities and their interaction to enhance vision-language alignment. Experiments on two public datasets demonstrate the effectiveness of our approach, advancing state-of-the-art in radiology report generation. The proposed transfer learning framework provides a means of harnessing semantic knowledge from large-scale pre-trained models to tackle data-scarce medical vision-language tasks. Code is available at https://github.com/chauncey-tow/MRG-CLIP.

Paperid: 1259, https://arxiv.org/pdf/2503.15784.pdf GitHub

Abstract:
Vision-Language Foundation Models (VLFM) have shown a tremendous increase in performance in terms of generating high-resolution, photorealistic natural images. While VLFMs show a rich understanding of semantic content across modalities, they often struggle with fine-grained alignment tasks that require precise correspondence between image regions and textual descriptions, a limitation in medical imaging, where accurate localization and detection of clinical features are essential for diagnosis and analysis. To address this issue, we propose a multi-stage architecture where a pre-trained VLFM (e.g. Stable Diffusion) provides a cursory semantic understanding, while a reinforcement learning (RL) algorithm refines the alignment through an iterative process that optimizes for understanding semantic context. The reward signal is designed to align the semantic information of the text with synthesized images. Experiments on the public ISIC2019 skin lesion dataset demonstrate that the proposed method improves (a) the quality of the generated images, and (b) the alignment with the text prompt over the original fine-tuned Stable Diffusion baseline. We also show that the synthesized samples could be used to improve disease classifier performance for underrepresented subgroups through augmentation. Our code is accessible through the project website: https://parhamsaremi.github.io/rl4med-ddpo

Paperid: 1260, https://arxiv.org/pdf/2503.15406.pdf GitHub

Abstract:
We introduce Visual Persona, a foundation model for text-to-image full-body human customization that, given a single in-the-wild human image, generates diverse images of the individual guided by text descriptions. Unlike prior methods that focus solely on preserving facial identity, our approach captures detailed full-body appearance, aligning with text descriptions for body structure and scene variations. Training this model requires large-scale paired human data, consisting of multiple images per individual with consistent full-body identities, which is notoriously difficult to obtain. To address this, we propose a data curation pipeline leveraging vision-language models to evaluate full-body appearance consistency, resulting in Visual Persona-500K, a dataset of 580k paired human images across 100k unique identities. For precise appearance transfer, we introduce a transformer encoder-decoder architecture adapted to a pre-trained text-to-image diffusion model, which augments the input image into distinct body regions, encodes these regions as local appearance features, and projects them into dense identity embeddings independently to condition the diffusion model for synthesizing customized images. Visual Persona consistently surpasses existing approaches, generating high-quality, customized images from in-the-wild inputs. Extensive ablation studies validate design choices, and we demonstrate the versatility of Visual Persona across various downstream tasks.

Paperid: 1261, https://arxiv.org/pdf/2503.15404.pdf GitHub

Abstract:
Vision Transformers (ViTs) have been widely applied in various computer vision and vision-language tasks. To gain insights into their robustness in practical scenarios, transferable adversarial examples on ViTs have been extensively studied. A typical approach to improving adversarial transferability is by refining the surrogate model. However, existing work on ViTs has restricted their surrogate refinement to backward propagation. In this work, we instead focus on Forward Propagation Refinement (FPR) and specifically refine two key modules of ViTs: attention maps and token embeddings. For attention maps, we propose Attention Map Diversification (AMD), which diversifies certain attention maps and also implicitly imposes beneficial gradient vanishing during backward propagation. For token embeddings, we propose Momentum Token Embedding (MTE), which accumulates historical token embeddings to stabilize the forward updates in both the Attention and MLP blocks. We conduct extensive experiments with adversarial examples transferred from ViTs to various CNNs and ViTs, demonstrating that our FPR outperforms the current best (backward) surrogate refinement by up to 7.0\% on average. We also validate its superiority against popular defenses and its compatibility with other transfer methods. Codes and appendix are available at https://github.com/RYC-98/FPR.

Paperid: 1262, https://arxiv.org/pdf/2503.15337.pdf GitHub

Abstract:
Identifying multiple novel classes in an image, known as open-vocabulary multi-label recognition, is a challenging task in computer vision. Recent studies explore the transfer of powerful vision-language models such as CLIP. However, these approaches face two critical challenges: (1) The local semantics of CLIP are disrupted due to its global pre-training objectives, resulting in unreliable regional predictions. (2) The matching property between image regions and candidate labels has been neglected, relying instead on naive feature aggregation such as average pooling, which leads to spurious predictions from irrelevant regions. In this paper, we present RAM (Recover And Match), a novel framework that effectively addresses the above issues. To tackle the first problem, we propose Ladder Local Adapter (LLA) to enforce refocusing on local regions, recovering local semantics in a memory-friendly way. For the second issue, we propose Knowledge-Constrained Optimal Transport (KCOT) to suppress meaningless matching to non-GT labels by formulating the task as an optimal transport problem. As a result, RAM achieves state-of-the-art performance on various datasets from three distinct domains, and shows great potential to boost the existing methods. Code: https://github.com/EricTan7/RAM.

Paperid: 1263, https://arxiv.org/pdf/2503.15166.pdf GitHub

Abstract:
Machine unlearning methods have become increasingly important for selective concept removal in large pre-trained models. While recent work has explored unlearning in Euclidean contrastive vision-language models, the effectiveness of concept removal in hyperbolic spaces remains unexplored. This paper investigates machine unlearning in hyperbolic contrastive learning by adapting Alignment Calibration to MERU, a model that embeds images and text in hyperbolic space to better capture semantic hierarchies. Through systematic experiments and ablation studies, we demonstrate that hyperbolic geometry offers distinct advantages for concept removal, achieving near perfect forgetting with reasonable performance on retained concepts, particularly when scaling to multiple concept removal. Our approach introduces hyperbolic-specific components including entailment calibration and norm regularization that leverage the unique properties of hyperbolic space. Comparative analysis with Euclidean models reveals fundamental differences in unlearning dynamics, with hyperbolic unlearning reorganizing the semantic hierarchy while Euclidean approaches merely disconnect cross-modal associations. These findings not only advance machine unlearning techniques but also provide insights into the geometric properties that influence concept representation and removal in multimodal models. Source code available at https://github.com/alex-pv01/HAC

Paperid: 1264, https://arxiv.org/pdf/2503.15024.pdf GitHub

Abstract:
Recently, the rapid development of AIGC has significantly boosted the diversities of fake media spread in the Internet, posing unprecedented threats to social security, politics, law, and etc. To detect the ever-increasingly diverse malicious fake media in the new era of AIGC, recent studies have proposed to exploit Large Vision Language Models (LVLMs) to design robust forgery detectors due to their impressive performance on a wide range of multimodal tasks. However, it still lacks a comprehensive benchmark designed to comprehensively assess LVLMs' discerning capabilities on forgery media. To fill this gap, we present Forensics-Bench, a new forgery detection evaluation benchmark suite to assess LVLMs across massive forgery detection tasks, requiring comprehensive recognition, location and reasoning capabilities on diverse forgeries. Forensics-Bench comprises 63,292 meticulously curated multi-choice visual questions, covering 112 unique forgery detection types from 5 perspectives: forgery semantics, forgery modalities, forgery tasks, forgery types and forgery models. We conduct thorough evaluations on 22 open-sourced LVLMs and 3 proprietary models GPT-4o, Gemini 1.5 Pro, and Claude 3.5 Sonnet, highlighting the significant challenges of comprehensive forgery detection posed by Forensics-Bench. We anticipate that Forensics-Bench will motivate the community to advance the frontier of LVLMs, striving for all-around forgery detectors in the era of AIGC. The deliverables will be updated at https://Forensics-Bench.github.io/.

Paperid: 1265, https://arxiv.org/pdf/2503.14911.pdf GitHub

Abstract:
The emergence of vision-language models has transformed medical AI, enabling unprecedented advances in diagnostic capability and clinical applications. However, progress in dermatology has lagged behind other medical domains due to the lack of standard image-text pairs. Existing dermatological datasets are limited in both scale and depth, offering only single-label annotations across a narrow range of diseases instead of rich textual descriptions, and lacking the crucial clinical context needed for real-world applications. To address these limitations, we present Derm1M, the first large-scale vision-language dataset for dermatology, comprising 1,029,761 image-text pairs. Built from diverse educational resources and structured around a standard ontology collaboratively developed by experts, Derm1M provides comprehensive coverage for over 390 skin conditions across four hierarchical levels and 130 clinical concepts with rich contextual information such as medical history, symptoms, and skin tone. To demonstrate Derm1M potential in advancing both AI research and clinical application, we pretrained a series of CLIP-like models, collectively called DermLIP, on this dataset. The DermLIP family significantly outperforms state-of-the-art foundation models on eight diverse datasets across multiple tasks, including zero-shot skin disease classification, clinical and artifacts concept identification, few-shot/full-shot learning, and cross-modal retrieval. Our dataset and code will be publicly available at https://github.com/SiyuanYan1/Derm1M upon acceptance.

Paperid: 1266, https://arxiv.org/pdf/2503.14607.pdf GitHub

Abstract:
In this paper, we introduce MapBench-the first dataset specifically designed for human-readable, pixel-based map-based outdoor navigation, curated from complex path finding scenarios. MapBench comprises over 1600 pixel space map path finding problems from 100 diverse maps. In MapBench, LVLMs generate language-based navigation instructions given a map image and a query with beginning and end landmarks. For each map, MapBench provides Map Space Scene Graph (MSSG) as an indexing data structure to convert between natural language and evaluate LVLM-generated results. We demonstrate that MapBench significantly challenges state-of-the-art LVLMs both zero-shot prompting and a Chain-of-Thought (CoT) augmented reasoning framework that decomposes map navigation into sequential cognitive processes. Our evaluation of both open-source and closed-source LVLMs underscores the substantial difficulty posed by MapBench, revealing critical limitations in their spatial reasoning and structured decision-making capabilities. We release all the code and dataset in https://github.com/taco-group/MapBench.

Paperid: 1267, https://arxiv.org/pdf/2503.14526.pdf GitHub

Abstract:
Vision-language-action (VLA) models present a promising paradigm by training policies directly on real robot datasets like Open X-Embodiment. However, the high cost of real-world data collection hinders further data scaling, thereby restricting the generalizability of VLAs. In this paper, we introduce ReBot, a novel real-to-sim-to-real approach for scaling real robot datasets and adapting VLA models to target domains, which is the last-mile deployment challenge in robot manipulation. Specifically, ReBot replays real-world robot trajectories in simulation to diversify manipulated objects (real-to-sim), and integrates the simulated movements with inpainted real-world background to synthesize physically realistic and temporally consistent robot videos (sim-to-real). Our approach has several advantages: 1) it enjoys the benefit of real data to minimize the sim-to-real gap; 2) it leverages the scalability of simulation; and 3) it can generalize a pretrained VLA to a target domain with fully automated data pipelines. Extensive experiments in both simulation and real-world environments show that ReBot significantly enhances the performance and robustness of VLAs. For example, in SimplerEnv with the WidowX robot, ReBot improved the in-domain performance of Octo by 7.2% and OpenVLA by 21.8%, and out-of-domain generalization by 19.9% and 9.4%, respectively. For real-world evaluation with a Franka robot, ReBot increased the success rates of Octo by 17% and OpenVLA by 20%. More information can be found at: https://yuffish.github.io/rebot/

Paperid: 1268, https://arxiv.org/pdf/2503.14421.pdf GitHub

Abstract:
The ever growing realism and quality of generated videos makes it increasingly harder for humans to spot deepfake content, who need to rely more and more on automatic deepfake detectors. However, deepfake detectors are also prone to errors, and their decisions are not explainable, leaving humans vulnerable to deepfake-based fraud and misinformation. To this end, we introduce ExDDV, the first dataset and benchmark for Explainable Deepfake Detection in Video. ExDDV comprises around 5.4K real and deepfake videos that are manually annotated with text descriptions (to explain the artifacts) and clicks (to point out the artifacts). We evaluate a number of vision-language models on ExDDV, performing experiments with various fine-tuning and in-context learning strategies. Our results show that text and click supervision are both required to develop robust explainable models for deepfake videos, which are able to localize and describe the observed artifacts. Our novel dataset and code to reproduce the results are available at https://github.com/vladhondru25/ExDDV.

Paperid: 1269, https://arxiv.org/pdf/2503.13964.pdf GitHub

Abstract:
Document Question Answering (DocQA) is a very common task. Existing methods using Large Language Models (LLMs) or Large Vision Language Models (LVLMs) and Retrieval Augmented Generation (RAG) often prioritize information from a single modal, failing to effectively integrate textual and visual cues. These approaches struggle with complex multi-modal reasoning, limiting their performance on real-world documents. We present MDocAgent (A Multi-Modal Multi-Agent Framework for Document Understanding), a novel RAG and multi-agent framework that leverages both text and image. Our system employs five specialized agents: a general agent, a critical agent, a text agent, an image agent and a summarizing agent. These agents engage in multi-modal context retrieval, combining their individual insights to achieve a more comprehensive understanding of the document's content. This collaborative approach enables the system to synthesize information from both textual and visual components, leading to improved accuracy in question answering. Preliminary experiments on five benchmarks like MMLongBench, LongDocURL demonstrate the effectiveness of our MDocAgent, achieve an average improvement of 12.1% compared to current state-of-the-art method. This work contributes to the development of more robust and comprehensive DocQA systems capable of handling the complexities of real-world documents containing rich textual and visual information. Our data and code are available at https://github.com/aiming-lab/MDocAgent.

Paperid: 1270, https://arxiv.org/pdf/2503.13891.pdf GitHub

Abstract:
Large Vision-Language Models (LVLMs) have shown promising performance in vision-language understanding and reasoning tasks. However, their visual understanding behaviors remain underexplored. A fundamental question arises: to what extent do LVLMs rely on visual input, and which image regions contribute to their responses? It is non-trivial to interpret the free-form generation of LVLMs due to their complicated visual architecture (e.g., multiple encoders and multi-resolution) and variable-length outputs. In this paper, we extend existing heatmap visualization methods (e.g., iGOS++) to support LVLMs for open-ended visual question answering. We propose a method to select visually relevant tokens that reflect the relevance between generated answers and input image. Furthermore, we conduct a comprehensive analysis of state-of-the-art LVLMs on benchmarks designed to require visual information to answer. Our findings offer several insights into LVLM behavior, including the relationship between focus region and answer correctness, differences in visual attention across architectures, and the impact of LLM scale on visual understanding. The code and data are available at https://github.com/bytedance/LVLM_Interpretation.

Paperid: 1271, https://arxiv.org/pdf/2503.13881.pdf GitHub

Abstract:
The fusion of Large Language Models with vision models is pioneering new possibilities in user-interactive vision-language tasks. A notable application is reasoning segmentation, where models generate pixel-level segmentation masks by comprehending implicit meanings in human instructions. However, seamless human-AI interaction demands more than just object-level recognition; it requires understanding both objects and the functions of their detailed parts, particularly in multi-target scenarios. For example, when instructing a robot to \textit{turn on the TV"}, there could be various ways to accomplish this command. Recognizing multiple objects capable of turning on the TV, such as the TV itself or a remote control (multi-target), provides more flexible options and aids in finding the optimized scenario. Furthermore, understanding specific parts of these objects, like the TV's button or the remote's button (part-level), is important for completing the action. Unfortunately, current reasoning segmentation datasets predominantly focus on a single target object-level reasoning, which limits the detailed recognition of an object's parts in multi-target contexts. To address this gap, we construct a large-scale dataset called Multi-target and Multi-granularity Reasoning (MMR). MMR comprises 194K complex and implicit instructions that consider multi-target, object-level, and part-level aspects, based on pre-existing image-mask sets. This dataset supports diverse and context-aware interactions by hierarchically providing object and part information. Moreover, we propose a straightforward yet effective framework for multi-target, object-level, and part-level reasoning segmentation. Experimental results on MMR show that the proposed method can reason effectively in multi-target and multi-granularity scenarios, while the existing reasoning segmentation model still has room for improvement.

Paperid: 1272, https://arxiv.org/pdf/2503.13652.pdf GitHub

Abstract:
Vision-language models (VLMs) (e.g. CLIP, LLaVA) are trained on large-scale, lightly curated web datasets, leading them to learn unintended correlations between semantic concepts and unrelated visual signals. These associations degrade model accuracy by causing predictions to rely on incidental patterns rather than genuine visual understanding. Prior work has weaponized these correlations as an attack vector to manipulate model predictions, such as inserting a deceiving class text onto the image in a "typographic" attack. These attacks succeed due to VLMs' text-heavy bias-a result of captions that echo visible words rather than describing content. However, this attack has focused solely on text that matches the target class exactly, overlooking a broader range of correlations, including non-matching text and graphical symbols, which arise from the abundance of branding content in web-scale data. To address this gap, we introduce "artifact-based" attacks: a novel class of manipulations that mislead models using both non-matching text and graphical elements. Unlike typographic attacks, these artifacts are not predefined, making them simultaneously harder to defend against and more challenging to find. We address this by framing artifact attacks as a search problem and demonstrate their effectiveness across five datasets, with some artifacts reinforcing each other to reach 100% attack success rates. These attacks transfer across models with up to 90% effectiveness, making it possible to attack unseen models. To defend against these attacks, we extend prior work's artifact aware prompting to the graphical setting. We see a moderate reduction of success rates of up to 15% relative to standard prompts, suggesting a promising direction for enhancing model robustness. Code: https://github.com/mqraitem/Web-Artifact-Attacks

Paperid: 1273, https://arxiv.org/pdf/2503.13446.pdf GitHub

Abstract:
Mobile manipulation is the fundamental challenge for robotics to assist humans with diverse tasks and environments in everyday life. However, conventional mobile manipulation approaches often struggle to generalize across different tasks and environments because of the lack of large-scale training. In contrast, recent advances in vision-language-action (VLA) models have shown impressive generalization capabilities, but these foundation models are developed for fixed-base manipulation tasks. Therefore, we propose an efficient policy adaptation framework named MoManipVLA to transfer pre-trained VLA models of fix-base manipulation to mobile manipulation, so that high generalization ability across tasks and environments can be achieved in mobile manipulation policy. Specifically, we utilize pre-trained VLA models to generate waypoints of the end-effector with high generalization ability. We design motion planning objectives for the mobile base and the robot arm, which aim at maximizing the physical feasibility of the trajectory. Finally, we present an efficient bi-level objective optimization framework for trajectory generation, where the upper-level optimization predicts waypoints for base movement to enhance the manipulator policy space, and the lower-level optimization selects the optimal end-effector trajectory to complete the manipulation task. In this way, MoManipVLA can adjust the position of the robot base in a zero-shot manner, thus making the waypoints predicted from the fixed-base VLA models feasible. Extensive experimental results on OVMM and the real world demonstrate that MoManipVLA achieves a 4.2% higher success rate than the state-of-the-art mobile manipulation, and only requires 50 training cost for real world deployment due to the strong generalization ability in the pre-trained VLA models.

Paperid: 1274, https://arxiv.org/pdf/2503.13443.pdf GitHub

Abstract:
The Base-New Trade-off (BNT) problem universally exists during the optimization of CLIP-based prompt tuning, where continuous fine-tuning on base (target) classes leads to a simultaneous decrease of generalization ability on new (unseen) classes. Existing approaches attempt to regulate the prompt tuning process to balance BNT by appending constraints. However, imposed on the same target prompt, these constraints fail to fully avert the mutual exclusivity between the optimization directions for base and new. As a novel solution to this challenge, we propose the plug-and-play Dual-Prompt Collaboration (DPC) framework, the first that decoupling the optimization processes of base and new tasks at the prompt level. Specifically, we clone a learnable parallel prompt based on the backbone prompt, and introduce a variable Weighting-Decoupling framework to independently control the optimization directions of dual prompts specific to base or new tasks, thus avoiding the conflict in generalization. Meanwhile, we propose a Dynamic Hard Negative Optimizer, utilizing dual prompts to construct a more challenging optimization task on base classes for enhancement. For interpretability, we prove the feature channel invariance of the prompt vector during the optimization process, providing theoretical support for the Weighting-Decoupling of DPC. Extensive experiments on multiple backbones demonstrate that DPC can significantly improve base performance without introducing any external knowledge beyond the base classes, while maintaining generalization to new classes. Code is available at: https://github.com/JREion/DPC.

Paperid: 1275, https://arxiv.org/pdf/2503.13440.pdf GitHub GitHub

Abstract:
With the advancement of RNN models with linear complexity, the quadratic complexity challenge of transformers has the potential to be overcome. Notably, the emerging Mamba-2 has demonstrated competitive performance, bridging the gap between RNN models and transformers. However, due to sequential processing and vanishing gradients, RNN models struggle to capture long-range dependencies, limiting contextual understanding. This results in slow convergence, high resource demands, and poor performance on downstream understanding and complex reasoning tasks. In this work, we present a hybrid model MaTVLM by substituting a portion of the transformer decoder layers in a pre-trained VLM with Mamba-2 layers. Leveraging the inherent relationship between attention and Mamba-2, we initialize Mamba-2 with corresponding attention weights to accelerate convergence. Subsequently, we employ a single-stage distillation process, using the pre-trained VLM as the teacher model to transfer knowledge to the MaTVLM, further enhancing convergence speed and performance. Furthermore, we investigate the impact of differential distillation loss within our training framework. We evaluate the MaTVLM on multiple benchmarks, demonstrating competitive performance against the teacher model and existing VLMs while surpassing both Mamba-based VLMs and models of comparable parameter scales. Remarkably, the MaTVLM achieves up to 3.6x faster inference than the teacher model while reducing GPU memory consumption by 27.5%, all without compromising performance. Code and models are released at http://github.com/hustvl/MaTVLM.

Paperid: 1276, https://arxiv.org/pdf/2503.13377.pdf GitHub

Abstract:
Temporal Video Grounding (TVG), the task of locating specific video segments based on language queries, is a core challenge in long-form video understanding. While recent Large Vision-Language Models (LVLMs) have shown early promise in tackling TVG through supervised fine-tuning (SFT), their abilities to generalize remain limited. To address this, we propose a novel post-training framework that enhances the generalization capabilities of LVLMs via reinforcement learning (RL). Specifically, our contributions span three key directions: (1) Time-R1: we introduce a reasoning-guided post-training framework via RL with verifiable reward to enhance the capabilities of LVLMs on the TVG task. (2) TimeRFT: we explore data-efficient post-training strategies on our curated RL-friendly dataset, which trains the model to progressively comprehend difficult samples, leading to better generalization. (3) TVGBench: we carefully construct a small yet comprehensive benchmark for LVLM evaluation, assessing 11 types of queries and featuring balanced distributions across both videos and queries. Extensive experiments demonstrate that Time-R1 achieves state-of-the-art performance across multiple downstream datasets using only 2.5K training data, while improving its general video understanding capabilities.

Paperid: 1277, https://arxiv.org/pdf/2503.13082.pdf GitHub GitHub

Abstract:
Performing robotic grasping from a cluttered bin based on human instructions is a challenging task, as it requires understanding both the nuances of free-form language and the spatial relationships between objects. Vision-Language Models (VLMs) trained on web-scale data, such as GPT-4o, have demonstrated remarkable reasoning capabilities across both text and images. But can they truly be used for this task in a zero-shot setting? And what are their limitations? In this paper, we explore these research questions via the free-form language-based robotic grasping task, and propose a novel method, FreeGrasp, leveraging the pre-trained VLMs' world knowledge to reason about human instructions and object spatial arrangements. Our method detects all objects as keypoints and uses these keypoints to annotate marks on images, aiming to facilitate GPT-4o's zero-shot spatial reasoning. This allows our method to determine whether a requested object is directly graspable or if other objects must be grasped and removed first. Since no existing dataset is specifically designed for this task, we introduce a synthetic dataset FreeGraspData by extending the MetaGraspNetV2 dataset with human-annotated instructions and ground-truth grasping sequences. We conduct extensive analyses with both FreeGraspData and real-world validation with a gripper-equipped robotic arm, demonstrating state-of-the-art performance in grasp reasoning and execution. Project website: https://tev-fbk.github.io/FreeGrasp/.

Paperid: 1278, https://arxiv.org/pdf/2503.13047.pdf GitHub

Abstract:
Directly generating planning results from raw sensors has become increasingly prevalent due to its adaptability and robustness in complex scenarios. Scene representation, as a key module in the pipeline, has traditionally relied on conventional perception, which focus on the global scene. However, in driving scenarios, human drivers typically focus only on regions that directly impact driving, which often coincide with those required for end-to-end autonomous driving. In this paper, a novel end-to-end autonomous driving method called InsightDrive is proposed, which organizes perception by language-guided scene representation. We introduce an instance-centric scene tokenizer that transforms the surrounding environment into map- and object-aware instance tokens. Scene attention language descriptions, which highlight key regions and obstacles affecting the ego vehicle's movement, are generated by a vision-language model that leverages the cognitive reasoning capabilities of foundation models. We then align scene descriptions with visual features using the vision-language model, guiding visual attention through these descriptions to give effectively scene representation. Furthermore, we employ self-attention and cross-attention mechanisms to model the ego-agents and ego-map relationships to comprehensively build the topological relationships of the scene. Finally, based on scene understanding, we jointly perform motion prediction and planning. Extensive experiments on the widely used nuScenes benchmark demonstrate that the proposed InsightDrive achieves state-of-the-art performance in end-to-end autonomous driving. The code is available at https://github.com/songruiqi/InsightDrive

Paperid: 1279, https://arxiv.org/pdf/2503.12999.pdf GitHub GitHub GitHub

Abstract:
Vision-Language Models (VLMs) have demonstrated exceptional performance in various multi-modal tasks. Recently, there has been an increasing interest in improving the personalization capabilities of VLMs. To better integrate user-provided concepts into VLMs, many methods use positive and negative samples to fine-tune these models. However, the scarcity of user-provided positive samples and the low quality of retrieved negative samples pose challenges for fine-tuning. To reveal the relationship between sample and model performance, we systematically investigate the impact of positive and negative samples (easy and hard) and their diversity on VLM personalization tasks. Based on the detailed analysis, we introduce Concept-as-Tree (CaT), which represents a concept as a tree structure, thereby enabling the data generation of positive and negative samples with varying difficulty and diversity for VLM personalization. With a well-designed data filtering strategy, our CaT framework can ensure the quality of generated data, constituting a powerful pipeline. We perform thorough experiments with various VLM personalization baselines to assess the effectiveness of the pipeline, alleviating the lack of positive samples and the low quality of negative samples. Our results demonstrate that CaT equipped with the proposed data filter significantly enhances the personalization capabilities of VLMs across the MyVLM, Yo'LLaVA, and MC-LLaVA datasets. To our knowledge, this work is the first controllable synthetic data pipeline for VLM personalization. The code is released at $\href{https://github.com/zengkaiya/CaT}{\text{https://github.com/zengkaiya/CaT}}$.

Paperid: 1280, https://arxiv.org/pdf/2503.12972.pdf GitHub

Abstract:
Multimodal reasoning in Large Language Models (LLMs) struggles with incomplete knowledge and hallucination artifacts, challenges that textual Knowledge Graphs (KGs) only partially mitigate due to their modality isolation. While Multimodal Knowledge Graphs (MMKGs) promise enhanced cross-modal understanding, their practical construction is impeded by semantic narrowness of manual text annotations and inherent noise in visual-semantic entity linkages. In this paper, we propose Vision-align-to-Language integrated Knowledge Graph (VaLiK), a novel approach for constructing MMKGs that enhances LLMs reasoning through cross-modal information supplementation. Specifically, we cascade pre-trained Vision-Language Models (VLMs) to align image features with text, transforming them into descriptions that encapsulate image-specific information. Furthermore, we developed a cross-modal similarity verification mechanism to quantify semantic consistency, effectively filtering out noise introduced during feature alignment. Even without manually annotated image captions, the refined descriptions alone suffice to construct the MMKG. Compared to conventional MMKGs construction paradigms, our approach achieves substantial storage efficiency gains while maintaining direct entity-to-image linkage capability. Experimental results on multimodal reasoning tasks demonstrate that LLMs augmented with VaLiK outperform previous state-of-the-art models. Our code is published at https://github.com/Wings-Of-Disaster/VaLiK.

Paperid: 1281, https://arxiv.org/pdf/2503.12955.pdf GitHub

Abstract:
We propose a new task to benchmark human-in-scene understanding for embodied agents: Human-In-Scene Question Answering (HIS-QA). Given a human motion within a 3D scene, HIS-QA requires the agent to comprehend human states and behaviors, reason about its surrounding environment, and answer human-related questions within the scene. To support this new task, we present HIS-Bench, a multimodal benchmark that systematically evaluates HIS understanding across a broad spectrum, from basic perception to commonsense reasoning and planning. Our evaluation of various vision-language models on HIS-Bench reveals significant limitations in their ability to handle HIS-QA tasks. To this end, we propose HIS-GPT, the first foundation model for HIS understanding. HIS-GPT integrates 3D scene context and human motion dynamics into large language models while incorporating specialized mechanisms to capture human-scene interactions. Extensive experiments demonstrate that HIS-GPT sets a new state-of-the-art on HIS-QA tasks. We hope this work inspires future research on human behavior analysis in 3D scenes, advancing embodied AI and world models. The codes and data: https://github.com/ZJHTerry18/HumanInScene.

Paperid: 1282, https://arxiv.org/pdf/2503.12927.pdf GitHub

Abstract:
Neuroblastoma (NB), a leading cause of childhood cancer mortality, exhibits significant histopathological variability, necessitating precise subtyping for accurate prognosis and treatment. Traditional diagnostic methods rely on subjective evaluations that are time-consuming and inconsistent. To address these challenges, we introduce MMLNB, a multi-modal learning (MML) model that integrates pathological images with generated textual descriptions to improve classification accuracy and interpretability. The approach follows a two-stage process. First, we fine-tune a Vision-Language Model (VLM) to enhance pathology-aware text generation. Second, the fine-tuned VLM generates textual descriptions, using a dual-branch architecture to independently extract visual and textual features. These features are fused via Progressive Robust Multi-Modal Fusion (PRMF) Block for stable training. Experimental results show that the MMLNB model is more accurate than the single modal model. Ablation studies demonstrate the importance of multi-modal fusion, fine-tuning, and the PRMF mechanism. This research creates a scalable AI-driven framework for digital pathology, enhancing reliability and interpretability in NB subtyping classification. Our source code is available at https://github.com/HovChen/MMLNB.

Paperid: 1283, https://arxiv.org/pdf/2503.12874.pdf GitHub

Abstract:
Large pre-trained vision-language models (VLMs), such as CLIP, demonstrate impressive generalization but remain highly vulnerable to adversarial examples (AEs). Previous work has explored robust text prompts through adversarial training, achieving some improvement in both robustness and generalization. However, they primarily rely on singlegradient direction perturbations (e.g., PGD) to generate AEs, which lack diversity, resulting in limited improvement in adversarial robustness. To address these limitations, we propose an evolution-based region adversarial prompt tuning method called ER-APT, which combines gradient methods with genetic evolution to generate more diverse and challenging AEs. In each training iteration, we first generate AEs using traditional gradient-based methods. Subsequently, a genetic evolution mechanism incorporating selection, mutation, and crossover is applied to optimize the AEs, ensuring a broader and more aggressive perturbation distribution.The final evolved AEs are used for prompt tuning, achieving region-based adversarial optimization instead of conventional single-point adversarial prompt tuning. We also propose a dynamic loss weighting method to adjust prompt learning efficiency for accuracy and robustness. Experimental evaluations on various benchmark datasets demonstrate the superiority of our proposed method, outperforming stateof-the-art APT methods. The code is released at https://github.com/jiaxiaojunQAQ/ER-APT.

Paperid: 1284, https://arxiv.org/pdf/2503.12866.pdf GitHub

Abstract:
Vision-language models (VLMs) encounter considerable challenges when adapting to domain shifts stemming from changes in data distribution. Test-time adaptation (TTA) has emerged as a promising approach to enhance VLM performance under such conditions. In practice, test data often arrives in batches, leading to increasing interest in the transductive TTA setting. However, existing TTA methods primarily focus on individual test samples, overlooking crucial cross-sample correlations within a batch. While recent ViT-based TTA methods have introduced batch-level adaptation, they remain suboptimal for VLMs due to inadequate integration of the text modality. To address these limitations, we propose a novel transductive TTA framework, Supportive Clique-based Attribute Prompting (SCAP), which effectively combines visual and textual information to enhance adaptation by generating fine-grained attribute prompts across test batches. SCAP first forms supportive cliques of test samples in an unsupervised manner based on visual similarity and learns an attribute prompt for each clique, capturing shared attributes critical for adaptation. For each test sample, SCAP aggregates attribute prompts from its associated cliques, providing enriched contextual information. To ensure adaptability over time, we incorporate a retention module that dynamically updates attribute prompts and their associated attributes as new data arrives. Comprehensive experiments across multiple benchmarks demonstrate that SCAP outperforms existing state-of-the-art methods, significantly advancing VLM generalization under domain shifts. Our code is available at https://github.com/zhoujiahuan1991/CVPR2025-SCAP.

Paperid: 1285, https://arxiv.org/pdf/2503.12609.pdf GitHub

Abstract:
We propose VISO-Grasp, a novel vision-language-informed system designed to systematically address visibility constraints for grasping in severely occluded environments. By leveraging Foundation Models (FMs) for spatial reasoning and active view planning, our framework constructs and updates an instance-centric representation of spatial relationships, enhancing grasp success under challenging occlusions. Furthermore, this representation facilitates active Next-Best-View (NBV) planning and optimizes sequential grasping strategies when direct grasping is infeasible. Additionally, we introduce a multi-view uncertainty-driven grasp fusion mechanism that refines grasp confidence and directional uncertainty in real-time, ensuring robust and stable grasp execution. Extensive real-world experiments demonstrate that VISO-Grasp achieves a success rate of $87.5\%$ in target-oriented grasping with the fewest grasp attempts outperforming baselines. To the best of our knowledge, VISO-Grasp is the first unified framework integrating FMs into target-aware active view planning and 6-DoF grasping in environments with severe occlusions and entire invisibility constraints. Code is available at: https://github.com/YitianShi/vMF-Contact

Paperid: 1286, https://arxiv.org/pdf/2503.12533.pdf GitHub

Abstract:
Building autonomous robotic agents capable of achieving human-level performance in real-world embodied tasks is an ultimate goal in humanoid robot research. Recent advances have made significant progress in high-level cognition with Foundation Models (FMs) and low-level skill development for humanoid robots. However, directly combining these components often results in poor robustness and efficiency due to compounding errors in long-horizon tasks and the varied latency of different modules. We introduce Being-0, a hierarchical agent framework that integrates an FM with a modular skill library. The FM handles high-level cognitive tasks such as instruction understanding, task planning, and reasoning, while the skill library provides stable locomotion and dexterous manipulation for low-level control. To bridge the gap between these levels, we propose a novel Connector module, powered by a lightweight vision-language model (VLM). The Connector enhances the FM's embodied capabilities by translating language-based plans into actionable skill commands and dynamically coordinating locomotion and manipulation to improve task success. With all components, except the FM, deployable on low-cost onboard computation devices, Being-0 achieves efficient, real-time performance on a full-sized humanoid robot equipped with dexterous hands and active vision. Extensive experiments in large indoor environments demonstrate Being-0's effectiveness in solving complex, long-horizon tasks that require challenging navigation and manipulation subtasks. For further details and videos, visit https://beingbeyond.github.io/Being-0.

Paperid: 1287, https://arxiv.org/pdf/2503.12532.pdf GitHub

Abstract:
Developing AI agents to autonomously manipulate graphical user interfaces is a long challenging task. Recent advances in data scaling law inspire us to train computer-use agents with a scaled instruction set, yet using behavior cloning to train agents still requires immense high-quality trajectories. To meet the scalability need, we designed STEVE, a step verification pipeline for computer-use agent training. First, we establish a large instruction set for computer-use agents and collect trajectory data with some suboptimal agents. GPT-4o is used to verify the correctness of each step in the trajectories based on the screens before and after the action execution, assigning each step with a binary label. Last, we adopt the Kahneman and Tversky Optimization to optimize the agent from the binary stepwise labels. Extensive experiments manifest that our agent outperforms supervised finetuning by leveraging both positive and negative actions within a trajectory. Also, STEVE enables us to train a 7B vision-language model as a computer-use agent, achieving leading performance in the challenging live desktop environment WinAgentArena with great efficiency at a reduced cost. Code and data: https://github.com/FanbinLu/STEVE.

Paperid: 1288, https://arxiv.org/pdf/2503.12496.pdf GitHub

Abstract:
The rise of Large Vision-Language Models (LVLMs) has significantly advanced video understanding. However, efficiently processing long videos remains a challenge due to the ``Sampling Dilemma'': low-density sampling risks missing critical information, while high-density sampling introduces redundancy. To address this issue, we introduce LSDBench, the first benchmark designed to evaluate LVLMs on long-video tasks by constructing high Necessary Sampling Density (NSD) questions, where NSD represents the minimum sampling density required to accurately answer a given question. LSDBench focuses on dense, short-duration actions to rigorously assess the sampling strategies employed by LVLMs. To tackle the challenges posed by high-NSD questions, we propose a novel Reasoning-Driven Hierarchical Sampling (RHS) framework, which combines global localization of question-relevant cues with local dense sampling for precise inference. Additionally, we develop a lightweight Semantic-Guided Frame Selector to prioritize informative frames, enabling RHS to achieve comparable or superior performance with significantly fewer sampled frames. Together, our LSDBench and RHS framework address the unique challenges of high-NSD long-video tasks, setting a new standard for evaluating and improving LVLMs in this domain. Our benchmark and evaluation codes has been released at: https://github.com/dvlab-research/LSDBench

Paperid: 1289, https://arxiv.org/pdf/2503.12149.pdf GitHub

Abstract:
With the advent of large vision-language models (LVLMs) demonstrating increasingly human-like abilities, a pivotal question emerges: do different LVLMs interpret multimodal sarcasm differently, and can a single model grasp sarcasm from multiple perspectives like humans? To explore this, we introduce an analytical framework using systematically designed prompts on existing multimodal sarcasm datasets. Evaluating 12 state-of-the-art LVLMs over 2,409 samples, we examine interpretive variations within and across models, focusing on confidence levels, alignment with dataset labels, and recognition of ambiguous "neutral" cases. We further validate our findings on a diverse 100-sample mini-benchmark, incorporating multiple datasets, expanded prompt variants, and representative commercial LVLMs. Our findings reveal notable discrepancies -- across LVLMs and within the same model under varied prompts. While classification-oriented prompts yield higher internal consistency, models diverge markedly when tasked with interpretive reasoning. These results challenge binary labeling paradigms by highlighting sarcasm's subjectivity. We advocate moving beyond rigid annotation schemes toward multi-perspective, uncertainty-aware modeling, offering deeper insights into multimodal sarcasm comprehension. Our code and data are available at: https://github.com/CoderChen01/LVLMSarcasmAnalysis

Paperid: 1290, https://arxiv.org/pdf/2503.12127.pdf GitHub

Abstract:
Addressing the retrieval of unsafe content from vision-language models such as CLIP is an important step towards real-world integration. Current efforts have relied on unlearning techniques that try to erase the model's knowledge of unsafe concepts. While effective in reducing unwanted outputs, unlearning limits the model's capacity to discern between safe and unsafe content. In this work, we introduce a novel approach that shifts from unlearning to an awareness paradigm by leveraging the inherent hierarchical properties of the hyperbolic space. We propose to encode safe and unsafe content as an entailment hierarchy, where both are placed in different regions of hyperbolic space. Our HySAC, Hyperbolic Safety-Aware CLIP, employs entailment loss functions to model the hierarchical and asymmetrical relations between safe and unsafe image-text pairs. This modelling, ineffective in standard vision-language models due to their reliance on Euclidean embeddings, endows the model with awareness of unsafe content, enabling it to serve as both a multimodal unsafe classifier and a flexible content retriever, with the option to dynamically redirect unsafe queries toward safer alternatives or retain the original output. Extensive experiments show that our approach not only enhances safety recognition but also establishes a more adaptable and interpretable framework for content moderation in vision-language models. Our source code is available at https://github.com/aimagelab/HySAC.

Paperid: 1291, https://arxiv.org/pdf/2503.12067.pdf GitHub

Abstract:
Deep Neural Networks (DNNs) have achieved notable performance in the fields of computer vision and natural language processing with various applications in both academia and industry. However, with recent advancements in DNNs and transformer models with a tremendous number of parameters, deploying these large models on edge devices causes serious issues such as high runtime and memory consumption. This is especially concerning with the recent large-scale foundation models, Vision-Language Models (VLMs), and Large Language Models (LLMs). Knowledge Distillation (KD) is one of the prominent techniques proposed to address the aforementioned problems using a teacher-student architecture. More specifically, a lightweight student model is trained using additional knowledge from a cumbersome teacher model. In this work, a comprehensive survey of knowledge distillation methods is proposed. This includes reviewing KD from different aspects: distillation sources, distillation schemes, distillation algorithms, distillation by modalities, applications of distillation, and comparison among existing methods. In contrast to most existing surveys, which are either outdated or simply update former surveys, this work proposes a comprehensive survey with a new point of view and representation structure that categorizes and investigates the most recent methods in knowledge distillation. This survey considers various critically important subcategories, including KD for diffusion models, 3D inputs, foundational models, transformers, and LLMs. Furthermore, existing challenges in KD and possible future research directions are discussed. Github page of the project: https://github.com/IPL-Sharif/KD_Survey

Paperid: 1292, https://arxiv.org/pdf/2503.11832.pdf GitHub

Abstract:
Recent vision-language models (VLMs) have made remarkable strides in generative modeling with multimodal inputs, particularly text and images. However, their susceptibility to generating harmful content when exposed to unsafe queries raises critical safety concerns. While current alignment strategies primarily rely on supervised safety fine-tuning with curated datasets, we identify a fundamental limitation we call the "safety mirage" where supervised fine-tuning inadvertently reinforces spurious correlations between superficial textual patterns and safety responses, rather than fostering deep, intrinsic mitigation of harm. We show that these spurious correlations leave fine-tuned VLMs vulnerable even to a simple one-word modification-based attack, where substituting a single word in text queries with a spurious correlation-inducing alternative can effectively bypass safeguards. Additionally, these correlations contribute to the over prudence, causing fine-tuned VLMs to refuse benign queries unnecessarily. To address this issue, we show machine unlearning (MU) as a powerful alternative to supervised safety fine-tuning as it avoids biased feature-label mappings and directly removes harmful knowledge from VLMs while preserving their general capabilities. Extensive evaluations across safety benchmarks show that under one-word attacks, MU-based alignment reduces the attack success rate by up to 60.17% and cuts unnecessary rejections by over 84.20%. Codes are available at https://github.com/OPTML-Group/VLM-Safety-MU. WARNING: There exist AI generations that may be offensive in nature.

Paperid: 1293, https://arxiv.org/pdf/2503.11368.pdf GitHub

Abstract:
Generating high-quality physically based rendering (PBR) materials is important to achieve realistic rendering in the downstream tasks, yet it remains challenging due to the intertwined effects of materials and lighting. While existing methods have made breakthroughs by incorporating material decomposition in the 3D generation pipeline, they tend to bake highlights into albedo and ignore spatially varying properties of metallicity and roughness. In this work, we present PBR3DGen, a two-stage mesh generation method with high-quality PBR materials that integrates the novel multi-view PBR material estimation model and a 3D PBR mesh reconstruction model. Specifically, PBR3DGen leverages vision language models (VLM) to guide multi-view diffusion, precisely capturing the spatial distribution and inherent attributes of reflective-metalness material. Additionally, we incorporate view-dependent illumination-aware conditions as pixel-aware priors to enhance spatially varying material properties. Furthermore, our reconstruction model reconstructs high-quality mesh with PBR materials. Experimental results demonstrate that PBR3DGen significantly outperforms existing methods, achieving new state-of-the-art results for PBR estimation and mesh generation. More results and visualization can be found on our project page: https://pbr3dgen1218.github.io/.

Paperid: 1294, https://arxiv.org/pdf/2503.11070.pdf GitHub

Abstract:
This paper introduces a holistic vision-language foundation model tailored for remote sensing, named Falcon. Falcon offers a unified, prompt-based paradigm that effectively executes comprehensive and complex remote sensing tasks. Falcon demonstrates powerful understanding and reasoning abilities at the image, region, and pixel levels. Specifically, given simple natural language instructions and remote sensing images, Falcon can produce impressive results in text form across 14 distinct tasks, i.e., image classification, object detection, segmentation, image captioning, and etc. To facilitate Falcon's training and empower its representation capacity to encode rich spatial and semantic information, we developed Falcon_SFT, a large-scale, multi-task, instruction-tuning dataset in the field of remote sensing. The Falcon_SFT dataset consists of approximately 78 million high-quality data samples, covering 5.6 million multi-spatial resolution and multi-view remote sensing images with diverse instructions. It features hierarchical annotations and undergoes manual sampling verification to ensure high data quality and reliability. Extensive comparative experiments are conducted, which verify that Falcon achieves remarkable performance over 67 datasets and 14 tasks, despite having only 0.7B parameters. We release the complete dataset, code, and model weights at https://github.com/TianHuiLab/Falcon, hoping to help further develop the open-source community.

Paperid: 1295, https://arxiv.org/pdf/2503.10857.pdf GitHub

Abstract:
Despite the promising results of large multimodal models (LMMs) in complex vision-language tasks that require knowledge, reasoning, and perception abilities together, we surprisingly found that these models struggle with simple tasks on infographics that require perception only. As existing benchmarks primarily focus on end tasks that require various abilities, they provide limited, fine-grained insights into the limitations of the models' perception abilities. To address this gap, we leverage the theory of graphical perception, an approach used to study how humans decode visual information encoded on charts and graphs, to develop an evaluation framework for analyzing gaps in LMMs' perception abilities in charts. With automated task generation and response evaluation designs, our framework enables comprehensive and controlled testing of LMMs' graphical perception across diverse chart types, visual elements, and task types. We apply our framework to evaluate and diagnose the perception capabilities of state-of-the-art LMMs at three granularity levels (chart, visual element, and pixel). Our findings underscore several critical limitations of current state-of-the-art LMMs, including GPT-4o: their inability to (1) generalize across chart types, (2) understand fundamental visual elements, and (3) cross reference values within a chart. These insights provide guidance for future improvements in perception abilities of LMMs. The evaluation framework and labeled data are publicly available at https://github.com/microsoft/lmm-graphical-perception.

Paperid: 1296, https://arxiv.org/pdf/2503.10635.pdf GitHub GitHub

Abstract:
Despite promising performance on open-source large vision-language models (LVLMs), transfer-based targeted attacks often fail against black-box commercial LVLMs. Analyzing failed adversarial perturbations reveals that the learned perturbations typically originate from a uniform distribution and lack clear semantic details, resulting in unintended responses. This critical absence of semantic information leads commercial LVLMs to either ignore the perturbation entirely or misinterpret its embedded semantics, thereby causing the attack to fail. To overcome these issues, we notice that identifying core semantic objects is a key objective for models trained with various datasets and methodologies. This insight motivates our approach that refines semantic clarity by encoding explicit semantic details within local regions, thus ensuring interoperability and capturing finer-grained features, and by concentrating modifications on semantically rich areas rather than applying them uniformly. To achieve this, we propose a simple yet highly effective solution: at each optimization step, the adversarial image is cropped randomly by a controlled aspect ratio and scale, resized, and then aligned with the target image in the embedding space. Experimental results confirm our hypothesis. Our adversarial examples crafted with local-aggregated perturbations focused on crucial regions exhibit surprisingly good transferability to commercial LVLMs, including GPT-4.5, GPT-4o, Gemini-2.0-flash, Claude-3.5-sonnet, Claude-3.7-sonnet, and even reasoning models like o1, Claude-3.7-thinking and Gemini-2.0-flash-thinking. Our approach achieves success rates exceeding 90% on GPT-4.5, 4o, and o1, significantly outperforming all prior state-of-the-art attack methods. Our optimized adversarial examples under different configurations and training code are available at https://github.com/VILA-Lab/M-Attack.

Paperid: 1297, https://arxiv.org/pdf/2503.10602.pdf GitHub

Abstract:
Object Hallucination (OH) has been acknowledged as one of the major trustworthy challenges in Large Vision-Language Models (LVLMs). Recent advancements in Large Language Models (LLMs) indicate that internal states, such as hidden states, encode the "overall truthfulness" of generated responses. However, it remains under-explored how internal states in LVLMs function and whether they could serve as "per-token" hallucination indicators, which is essential for mitigating OH. In this paper, we first conduct an in-depth exploration of LVLM internal states in relation to OH issues and discover that (1) LVLM internal states are high-specificity per-token indicators of hallucination behaviors. Moreover, (2) different LVLMs encode universal patterns of hallucinations in common latent subspaces, indicating that there exist "generic truthful directions" shared by various LVLMs. Based on these discoveries, we propose Truthful-Guided Pre-Intervention (TruthPrInt) that first learns the truthful direction of LVLM decoding and then applies truthful-guided inference-time intervention during LVLM decoding. We further propose ComnHallu to enhance both cross-LVLM and cross-data hallucination detection transferability by constructing and aligning hallucination latent subspaces. We evaluate TruthPrInt in extensive experimental settings, including in-domain and out-of-domain scenarios, over popular LVLMs and OH benchmarks. Experimental results indicate that TruthPrInt significantly outperforms state-of-the-art methods. Codes will be available at https://github.com/jinhaoduan/TruthPrInt.

Paperid: 1298, https://arxiv.org/pdf/2503.10596.pdf GitHub

Abstract:
Pixel grounding, encompassing tasks such as Referring Expression Segmentation (RES), has garnered considerable attention due to its immense potential for bridging the gap between vision and language modalities. However, advancements in this domain are currently constrained by limitations inherent in existing datasets, including limited object categories, insufficient textual diversity, and a scarcity of high-quality annotations. To mitigate these limitations, we introduce GroundingSuite, which comprises: (1) an automated data annotation framework leveraging multiple Vision-Language Model (VLM) agents; (2) a large-scale training dataset encompassing 9.56 million diverse referring expressions and their corresponding segmentations; and (3) a meticulously curated evaluation benchmark consisting of 3,800 images. The GroundingSuite training dataset facilitates substantial performance improvements, enabling models trained on it to achieve state-of-the-art results. Specifically, a cIoU of 68.9 on gRefCOCO and a gIoU of 55.3 on RefCOCOm. Moreover, the GroundingSuite annotation framework demonstrates superior efficiency compared to the current leading data annotation method, i.e., $4.5 \times$ faster than GLaMM.

Paperid: 1299, https://arxiv.org/pdf/2503.10127.pdf GitHub GitHub

Abstract:
In this paper, we propose a unified layout planning and image generation model, PlanGen, which can pre-plan spatial layout conditions before generating images. Unlike previous diffusion-based models that treat layout planning and layout-to-image as two separate models, PlanGen jointly models the two tasks into one autoregressive transformer using only next-token prediction. PlanGen integrates layout conditions into the model as context without requiring specialized encoding of local captions and bounding box coordinates, which provides significant advantages over the previous embed-and-pool operations on layout conditions, particularly when dealing with complex layouts. Unified prompting allows PlanGen to perform multitasking training related to layout, including layout planning, layout-to-image generation, image layout understanding, etc. In addition, PlanGen can be seamlessly expanded to layout-guided image manipulation thanks to the well-designed modeling, with teacher-forcing content manipulation policy and negative layout guidance. Extensive experiments verify the effectiveness of our PlanGen in multiple layoutrelated tasks, showing its great potential. Code is available at: https://360cvgroup.github.io/PlanGen.

Paperid: 1300, https://arxiv.org/pdf/2503.10110.pdf GitHub

Abstract:
Motion planning involves determining a sequence of robot configurations to reach a desired pose, subject to movement and safety constraints. Traditional motion planning finds collision-free paths, but this is overly restrictive in clutter, where it may not be possible for a robot to accomplish a task without contact. In addition, contacts range from relatively benign (e.g., brushing a soft pillow) to more dangerous (e.g., toppling a glass vase). Due to this diversity, it is difficult to characterize which contacts may be acceptable or unacceptable. In this paper, we propose IMPACT, a novel motion planning framework that uses Vision-Language Models (VLMs) to infer environment semantics, identifying which parts of the environment can best tolerate contact based on object properties and locations. Our approach uses the VLM's outputs to produce a dense 3D "cost map" that encodes contact tolerances and seamlessly integrates with standard motion planners. We perform experiments using 20 simulation and 10 real-world scenes and assess using task success rate, object displacements, and feedback from human evaluators. Our results over 3620 simulation and 200 real-world trials suggest that IMPACT enables efficient contact-rich motion planning in cluttered settings while outperforming alternative methods and ablations. Supplementary material is available at https://impact-planning.github.io/.

Paperid: 1301, https://arxiv.org/pdf/2503.10080.pdf GitHub

Abstract:
Recently, vision-language models (e.g. CLIP) have demonstrated remarkable performance in zero-shot anomaly detection (ZSAD). By leveraging auxiliary data during training, these models can directly perform cross-category anomaly detection on target datasets, such as detecting defects on industrial product surfaces or identifying tumors in organ tissues. Existing approaches typically construct text prompts through either manual design or the optimization of learnable prompt vectors. However, these methods face several challenges: 1) handcrafted prompts require extensive expert knowledge and trial-and-error; 2) single-form learnable prompts struggle to capture complex anomaly semantics; and 3) an unconstrained prompt space limits generalization to unseen categories. To address these issues, we propose Bayesian Prompt Flow Learning (Bayes-PFL), which models the prompt space as a learnable probability distribution from a Bayesian perspective. Specifically, a prompt flow module is designed to learn both image-specific and image-agnostic distributions, which are jointly utilized to regularize the text prompt space and improve the model's generalization on unseen categories. These learned distributions are then sampled to generate diverse text prompts, effectively covering the prompt space. Additionally, a residual cross-model attention (RCA) module is introduced to better align dynamic text embeddings with fine-grained image features. Extensive experiments on 15 industrial and medical datasets demonstrate our method's superior performance. The code is available at https://github.com/xiaozhen228/Bayes-PFL.

Paperid: 1302, https://arxiv.org/pdf/2503.09527.pdf GitHub

Abstract:
Recent advances in Vision-Language-Action models (VLAs) have expanded the capabilities of embodied intelligence. However, significant challenges remain in real-time decision-making in complex 3D environments, which demand second-level responses, high-resolution perception, and tactical reasoning under dynamic conditions. To advance the field, we introduce CombatVLA, an efficient VLA model optimized for combat tasks in 3D action role-playing games(ARPGs). Specifically, our CombatVLA is a 3B model trained on video-action pairs collected by an action tracker, where the data is formatted as action-of-thought (AoT) sequences. Thereafter, CombatVLA seamlessly integrates into an action execution framework, allowing efficient inference through our truncated AoT strategy. Experimental results demonstrate that CombatVLA not only outperforms all existing models on the combat understanding benchmark but also achieves a 50-fold acceleration in game combat. Moreover, it has a higher task success rate than human players. We will open-source all resources, including the action tracker, dataset, benchmark, model weights, training code, and the implementation of the framework at https://combatvla.github.io/.

Paperid: 1303, https://arxiv.org/pdf/2503.09423.pdf GitHub

Abstract:
We study the task of language-conditioned pick and place in clutter, where a robot should grasp a target object in open clutter and move it to a specified place. Some approaches learn end-to-end policies with features from vision foundation models, requiring large datasets. Others combine foundation models in a zero-shot setting, suffering from cascading errors. In addition, they primarily leverage vision and language foundation models, focusing less on action priors. In this paper, we aim to develop an effective policy by integrating foundation priors from vision, language, and action. We propose A$^2$, an action prior alignment method that aligns unconditioned action priors with 3D vision-language priors by learning one attention layer. The alignment formulation enables our policy to train with less data and preserve zero-shot generalization capabilities. We show that a shared policy for both pick and place actions enhances the performance for each task, and introduce a policy adaptation scheme to accommodate the multi-modal nature of actions. Extensive experiments in simulation and the real-world show that our policy achieves higher task success rates with fewer steps for both pick and place tasks in clutter, effectively generalizing to unseen objects and language instructions. Videos and codes are available at https://xukechun.github.io/papers/A2.

Paperid: 1304, https://arxiv.org/pdf/2503.09348.pdf GitHub GitHub

Abstract:
Large multimodal models (LMMs) have demonstrated significant potential as generalists in vision-language (VL) tasks. However, there remains a significant gap between state-of-the-art LMMs and human performance when it comes to complex tasks that require a combination of fundamental VL capabilities, as well as tasks involving the grounding of complex instructions. To thoroughly investigate the human-LMM gap and its underlying causes, we propose MOAT, a diverse benchmark with complex real-world VL tasks that are challenging for LMMs. Specifically, the tasks in MOAT require LMMs to engage in generalist problem solving by integrating fundamental VL capabilities such as reading text, counting, understanding spatial relations, grounding textual and visual instructions, etc. All these abilities fit into a taxonomy proposed by us that contains 10 fundamental VL capabilities, enabling MOAT to provide a fine-grained view of LMMs' strengths and weaknesses. Besides, MOAT is the first benchmark to explicitly evaluate LMMs' ability to ground complex text and visual instructions, which is essential to many real-world applications. We evaluate over 20 proprietary and open source LMMs, as well as humans, on MOAT, and found that humans achieved 82.7% accuracy while the best performing LMM (OpenAI o1) achieved only 38.8%. To guide future model development, we analyze common trends in our results and discuss the underlying causes of observed performance gaps between LMMs and humans, focusing on which VL capability forms the bottleneck in complex tasks, whether test time scaling improves performance on MOAT, and how tiling harms LMMs' capability to count. Code and data are available at https://cambrian-yzt.github.io/MOAT.

Paperid: 1305, https://arxiv.org/pdf/2503.09279.pdf GitHub

Abstract:
Video Detailed Captioning (VDC) is a crucial task for vision-language bridging, enabling fine-grained descriptions of complex video content. In this paper, we first comprehensively benchmark current state-of-the-art approaches and systematically identified two critical limitations: biased capability towards specific captioning aspect and misalignment with human preferences. To address these deficiencies, we propose Cockatiel, a novel three-stage training pipeline that ensembles synthetic and human-aligned training for improving VDC performance. In the first stage, we derive a scorer from a meticulously annotated dataset to select synthetic captions high-performing on certain fine-grained video-caption alignment and human-preferred while disregarding others. Then, we train Cockatiel-13B, using this curated dataset to infuse it with assembled model strengths and human preferences. Finally, we further distill Cockatiel-8B from Cockatiel-13B for the ease of usage. Extensive quantitative and qualitative experiments reflect the effectiveness of our method, as we not only set new state-of-the-art performance on VDCSCORE in a dimension-balanced way but also surpass leading alternatives on human preference by a large margin as depicted by the human evaluation results.

Paperid: 1306, https://arxiv.org/pdf/2503.09146.pdf GitHub

Abstract:
Despite recent advances in Video Large Language Models (VideoLLMs), effectively understanding long-form videos remains a significant challenge. Perceiving lengthy videos containing thousands of frames poses substantial computational burden. To mitigate this issue, this paper introduces Generative Frame Sampler (GenS), a plug-and-play module integrated with VideoLLMs to facilitate efficient lengthy video perception. Built upon a lightweight VideoLLM, GenS leverages its inherent vision-language capabilities to identify question-relevant frames. To facilitate effective retrieval, we construct GenS-Video-150K, a large-scale video instruction dataset with dense frame relevance annotations. Extensive experiments demonstrate that GenS consistently boosts the performance of various VideoLLMs, including open-source models (Qwen2-VL-7B, Aria-25B, VILA-40B, LLaVA-Video-7B/72B) and proprietary assistants (GPT-4o, Gemini). When equipped with GenS, open-source VideoLLMs achieve impressive state-of-the-art results on long-form video benchmarks: LLaVA-Video-72B reaches 66.8 (+4.3) on LongVideoBench and 77.0 (+2.7) on MLVU, while Aria obtains 39.2 on HourVideo surpassing the Gemini-1.5-pro by 1.9 points. We will release all datasets and models at https://generative-sampler.github.io.

Paperid: 1307, https://arxiv.org/pdf/2503.09013.pdf GitHub

Abstract:
Universal adverse weather removal (UAWR) seeks to address various weather degradations within a unified framework. Recent methods are inspired by prompt learning using pre-trained vision-language models (e.g., CLIP), leveraging degradation-aware prompts to facilitate weather-free image restoration, yielding significant improvements. In this work, we propose CyclicPrompt, an innovative cyclic prompt approach designed to enhance the effectiveness, adaptability, and generalizability of UAWR. CyclicPrompt Comprises two key components: 1) a composite context prompt that integrates weather-related information and context-aware representations into the network to guide restoration. This prompt differs from previous methods by marrying learnable input-conditional vectors with weather-specific knowledge, thereby improving adaptability across various degradations. 2) The erase-and-paste mechanism, after the initial guided restoration, substitutes weather-specific knowledge with constrained restoration priors, inducing high-quality weather-free concepts into the composite prompt to further fine-tune the restoration process. Therefore, we can form a cyclic "Prompt-Restore-Prompt" pipeline that adeptly harnesses weather-specific knowledge, textual contexts, and reliable textures. Extensive experiments on synthetic and real-world datasets validate the superior performance of CyclicPrompt. The code is available at: https://github.com/RongxinL/CyclicPrompt.

Paperid: 1308, https://arxiv.org/pdf/2503.08906.pdf GitHub

Abstract:
Vision-language models (VLMs) such as CLIP demonstrate strong performance but struggle when adapted to downstream tasks. Prompt learning has emerged as an efficient and effective strategy to adapt VLMs while preserving their pre-trained knowledge. However, existing methods still lead to overfitting and degrade zero-shot generalization. To address this challenge, we propose an optimal transport (OT)-guided prompt learning framework that mitigates forgetting by preserving the structural consistency of feature distributions between pre-trained and fine-tuned models. Unlike conventional point-wise constraints, OT naturally captures cross-instance relationships and expands the feasible parameter space for prompt tuning, allowing a better trade-off between adaptation and generalization. Our approach enforces joint constraints on both vision and text representations, ensuring a holistic feature alignment. Extensive experiments on benchmark datasets demonstrate that our simple yet effective method can outperform existing prompt learning strategies in base-to-novel generalization, cross-dataset evaluation, and domain generalization without additional augmentation or ensemble techniques. The code is available at https://github.com/ChongQingNoSubway/Prompt-OT

Paperid: 1309, https://arxiv.org/pdf/2503.08510.pdf GitHub

Abstract:
Class-Incremental Learning (CIL) enables learning systems to continuously adapt to evolving data streams. With the advancement of pre-training, leveraging pre-trained vision-language models (e.g., CLIP) offers a promising starting point for CIL. However, CLIP makes decisions by matching visual embeddings to class names, overlooking the rich contextual information conveyed through language. For instance, the concept of ``cat'' can be decomposed into features like tail, fur, and face for recognition. Besides, since the model is continually updated, these detailed features are overwritten in CIL, requiring external knowledge for compensation. In this paper, we introduce ExterNal knowledGe INjEction (ENGINE) for CLIP-based CIL. To enhance knowledge transfer from outside the dataset, we propose a dual-branch injection tuning framework that encodes informative knowledge from both visual and textual modalities. The visual branch is enhanced with data augmentation to enrich the visual features, while the textual branch leverages GPT-4 to rewrite discriminative descriptors. In addition to this on-the-fly knowledge injection, we also implement post-tuning knowledge by re-ranking the prediction results during inference. With the injected knowledge, the model can better capture informative features for downstream tasks as data evolves. Extensive experiments demonstrate the state-of-the-art performance of ENGINE. Code is available at: https://github.com/LAMDA-CL/ICCV25-ENGINE

Paperid: 1310, https://arxiv.org/pdf/2503.08497.pdf GitHub

Abstract:
Large-scale pre-trained Vision-Language Models (VLMs) have become essential for transfer learning across diverse tasks. However, adapting these models with limited few-shot data often leads to overfitting, diminishing their performance on new tasks. To tackle this issue, we propose a novel Multi-Modal Representation Learning (MMRL) framework that introduces a shared, learnable, and modality-agnostic representation space. MMRL projects the space tokens to text and image representation tokens, facilitating more effective multi-modal interactions. Unlike previous approaches that solely optimize class token features, MMRL integrates representation tokens at higher layers of the encoders--where dataset-specific features are more prominent--while preserving generalized knowledge in the lower layers. During training, both representation and class features are optimized, with trainable projection layer applied to the representation tokens, whereas the class token projection layer remains frozen to retain pre-trained knowledge. Furthermore, a regularization term is introduced to align the class features and text features with the zero-shot features from the frozen VLM, thereby safeguarding the model's generalization capacity. For inference, a decoupling strategy is employed, wherein both representation and class features are utilized for base classes, while only the class features, which retain more generalized knowledge, are used for new tasks. Extensive experiments across 15 datasets demonstrate that MMRL outperforms state-of-the-art methods, achieving a balanced trade-off between task-specific adaptation and generalization. Code is available at https://github.com/yunncheng/MMRL.

Paperid: 1311, https://arxiv.org/pdf/2503.08336.pdf GitHub

Abstract:
Embodied outdoor scene understanding forms the foundation for autonomous agents to perceive, analyze, and react to dynamic driving environments. However, existing 3D understanding is predominantly based on 2D Vision-Language Models (VLMs), which collect and process limited scene-aware contexts. In contrast, compared to the 2D planar visual information, point cloud sensors such as LiDAR provide rich depth and fine-grained 3D representations of objects. Even better the emerging 4D millimeter-wave radar detects the motion trend, velocity, and reflection intensity of each object. The integration of these two modalities provides more flexible querying conditions for natural language, thereby supporting more accurate 3D visual grounding. To this end, we propose a novel method called TPCNet, the first outdoor 3D visual grounding model upon the paradigm of prompt-guided point cloud sensor combination, including both LiDAR and radar sensors. To optimally combine the features of these two sensors required by the prompt, we design a multi-fusion paradigm called Two-Stage Heterogeneous Modal Adaptive Fusion. Specifically, this paradigm initially employs Bidirectional Agent Cross-Attention (BACA), which feeds both-sensor features, characterized by global receptive fields, to the text features for querying. Moreover, we design a Dynamic Gated Graph Fusion (DGGF) module to locate the regions of interest identified by the queries. To further enhance accuracy, we devise an C3D-RECHead, based on the nearest object edge to the ego-vehicle. Experimental results demonstrate that our TPCNet, along with its individual modules, achieves the state-of-the-art performance on both the Talk2Radar and Talk2Car datasets. We release the code at https://github.com/GuanRunwei/TPCNet.

Paperid: 1312, https://arxiv.org/pdf/2503.08229.pdf GitHub

Abstract:
Large pre-trained vision-language models (VLMs) offer a promising approach to leveraging human language for enhancing downstream tasks. However, VLMs such as CLIP face significant limitation: its performance is highly sensitive to prompt template design. Although prompt learning methods can address the sensitivity issue by replacing natural language prompts with learnable ones, they are incomprehensible to humans. Ensuring consistent performance across various prompt templates enables models to adapt seamlessly to diverse phrasings, enhancing their ability to handle downstream tasks without requiring extensive prompt engineering. In this work, we introduce the RobustPrompt Benchmark, a systematic benchmark to evaluate robustness to different prompt templates for VLMs. It includes a dataset with hundreds of carefully designed prompt templates, divided into six types, covering a wide variety of commonly used templates. Beside the benchmark, we propose Modeling Variants of Prompts (MVP), a simple yet effective method that mitigates sensitivity by modeling variants of prompt structures. The innovation of MVP lies in decoupling prompts into templates and class names, and using Variational Autoencoders (VAE) to model the distribution of diverse prompt structures. Experiments across 11 datasets demonstrate that MVP can greatly enhance model robustness to variations in input prompts without a drop in performance. The code is available at https://github.com/liaolea/MVP.

Paperid: 1313, https://arxiv.org/pdf/2503.08048.pdf GitHub GitHub

Abstract:
Recently, Probabilistic Language-Image Pre-Training (ProLIP) has been proposed to tackle the multiplicity issue of vision-language (VL) tasks. Despite their success in probabilistic representation learning at a scale, the ProLIP models cannot handle long context texts longer than 64 context length, which limits their ability to capture rich contextual information from longer text sequences. To address this issue, this paper proposes a fine-tuning strategy for ProLIP to accept longer texts, e.g., 256 text tokens. Experimental results on Urban-1k and the DataComp evaluation suite show that the proposed LongProLIP recipe can improve understanding of long contexts while minimizing the negative effect of fine-tuning.We also observe a trade-off between the long context understanding (measured by Urban-1k) and general zero-shot capability (measured by evaluation datasets by DataComp). Code is available at https://github.com/naver-ai/prolip

Paperid: 1314, https://arxiv.org/pdf/2503.08038.pdf GitHub

Abstract:
In this paper, we delve deeper into the Kullback-Leibler (KL) Divergence loss and mathematically prove that it is equivalent to the Decoupled Kullback-Leibler (DKL) Divergence loss that consists of (1) a weighted Mean Square Error (wMSE) loss and (2) a Cross-Entropy loss incorporating soft labels. Thanks to the decoupled structure of DKL loss, we have identified two areas for improvement. Firstly, we address the limitation of KL loss in scenarios like knowledge distillation by breaking its asymmetric optimization property along with a smoother weight function. This modification effectively alleviates convergence challenges in optimization, particularly for classes with high predicted scores in soft labels. Secondly, we introduce class-wise global information into KL/DKL to reduce bias arising from individual samples. With these two enhancements, we derive the Generalized Kullback-Leibler (GKL) Divergence loss and evaluate its effectiveness by conducting experiments on CIFAR-10/100, ImageNet, and vision-language datasets, focusing on adversarial training, and knowledge distillation tasks. Specifically, we achieve new state-of-the-art adversarial robustness on the public leaderboard -- RobustBench and competitive knowledge distillation performance across CIFAR/ImageNet models and CLIP models, demonstrating the substantial practical merits. Our code is available at https://github.com/jiequancui/DKL.

Paperid: 1315, https://arxiv.org/pdf/2503.08019.pdf GitHub

Abstract:
As the computational needs of Large Vision-Language Models (LVLMs) increase, visual token pruning has proven effective in improving inference speed and memory efficiency. Traditional pruning methods in LVLMs predominantly focus on attention scores to determine token relevance, overlooking critical aspects such as spatial position and token similarity. To this end, we introduce AdaptPrune, a novel plug-and-play training-free pruning method that builds on conventional attention-based pruning by integrating spatial distance and token similarity with an adaptive NMS approach. Our method is based on several observed phenomena in large models: the positional bias in the model's image attention and the redundancy of token information ignored by previous approaches. By integrating attention, spatial, and similarity information, our approach ensures a comprehensive evaluation of token importance and substantially refines the pruning decisions. Our method has been extensively tested across various LVLMs and benchmarks, confirming its robustness and adaptability. The results demonstrate that AdaptPrune consistently outperforms existing methods across various pruning ratios. Code is available at https://github.com/bzluan/AdaptPrune.

Paperid: 1316, https://arxiv.org/pdf/2503.07920.pdf GitHub

Authors:Samuel Cahyawijaya, Holy Lovenia, Joel Ruben Antony Moniz, Tack Hwa Wong, Mohammad Rifqi Farhansyah, Thant Thiri Maung, Frederikus Hudi, David Anugraha, Muhammad Ravi Shulthan Habibi, Muhammad Reza Qorib, Amit Agarwal, Joseph Marvin Imperial, Hitesh Laxmichand Patel, Vicky Feliren, Bahrul Ilmi Nasution, Manuel Antonio Rufino, Genta Indra Winata, Rian Adam Rajagede, Carlos Rafael Catalan, Mohamed Fazli Imam, Priyaranjan Pattnayak, Salsabila Zahirah Pranida, Kevin Pratama, Yeshil Bangera, Adisai Na-Thalang, Patricia Nicole Monderin, Yueqi Song, Christian Simon, Lynnette Hui Xian Ng, Richardy Lobo' Sapan, Taki Hasan Rafi, Bin Wang, Supryadi, Kanyakorn Veerakanjana, Piyalitt Ittichaiwong, Matthew Theodore Roque, Karissa Vincentio, Takdanai Kreangphet, Phakphum Artkaew, Kadek Hendrawan Palgunadi, Yanzhi Yu, Rochana Prih Hastuti, William Nixon, Mithil Bangera, Adrian Xuan Wei Lim, Aye Hninn Khine, Hanif Muhammad Zhafran, Teddy Ferdinan, Audra Aurora Izzani, Ayushman Singh, Evan, Jauza Akbar Krito, Michael Anugraha, Fenal Ashokbhai Ilasariya, Haochen Li, John Amadeo Daniswara, Filbert Aurelian Tjiaranata, Eryawan Presma Yulianrifat, Can Udomcharoenchaikit, Fadil Risdian Ansori, Mahardika Krisna Ihsani, Giang Nguyen, Anab Maulana Barik, Dan John Velasco, Rifo Ahmad Genadi, Saptarshi Saha, Chengwei Wei, Isaiah Flores, Kenneth Ko Han Chen, Anjela Gail Santos, Wan Shen Lim, Kaung Si Phyo, Tim Santos, Meisyarah Dwiastuti, Jiayun Luo, Jan Christian Blaise Cruz, Ming Shan Hee, Ikhlasul Akmal Hanif, M. Alif Al Hakim, Muhammad Rizky Sya'ban, Kun Kerdthaisong, Lester James V. Miranda, Fajri Koto, Tirana Noor Fatyanosa, Alham Fikri Aji, Jostin Jerico Rosal, Jun Kevin, Robert Wijaya, Onno P. Kampman, Ruochen Zhang, BÃ¶rje F. Karlsson, Peerat Limkonchotiwat

Abstract:
Southeast Asia (SEA) is a region of extraordinary linguistic and cultural diversity, yet it remains significantly underrepresented in vision-language (VL) research. This often results in artificial intelligence (AI) models that fail to capture SEA cultural nuances. To fill this gap, we present SEA-VL, an open-source initiative dedicated to developing high-quality, culturally relevant data for SEA languages. By involving contributors from SEA countries, SEA-VL aims to ensure better cultural relevance and diversity, fostering greater inclusivity of underrepresented languages in VL research. Beyond crowdsourcing, our initiative goes one step further in the exploration of the automatic collection of culturally relevant images through crawling and image generation. First, we find that image crawling achieves approximately ~85% cultural relevance while being more cost- and time-efficient than crowdsourcing. Second, despite the substantial progress in generative vision models, synthetic images remain unreliable in accurately reflecting SEA cultures. The generated images often fail to reflect the nuanced traditions and cultural contexts of the region. Collectively, we gather 1.28M SEA culturally-relevant images, more than 50 times larger than other existing datasets. Through SEA-VL, we aim to bridge the representation gap in SEA, fostering the development of more inclusive AI systems that authentically represent diverse cultures across SEA.

Paperid: 1317, https://arxiv.org/pdf/2503.07635.pdf GitHub

Abstract:
Video question grounding (VideoQG) requires models to answer the questions and simultaneously infer the relevant video segments to support the answers. However, existing VideoQG methods usually suffer from spurious cross-modal correlations, leading to a failure to identify the dominant visual scenes that align with the intended question. Moreover, vision-language models exhibit unfaithful generalization performance and lack robustness on challenging downstream tasks such as VideoQG. In this work, we propose a novel VideoQG framework named Cross-modal Causal Relation Alignment (CRA), to eliminate spurious correlations and improve the causal consistency between question-answering and video temporal grounding. Our CRA involves three essential components: i) Gaussian Smoothing Grounding (GSG) module for estimating the time interval via cross-modal attention, which is de-noised by an adaptive Gaussian filter, ii) Cross-Modal Alignment (CMA) enhances the performance of weakly supervised VideoQG by leveraging bidirectional contrastive learning between estimated video segments and QA features, iii) Explicit Causal Intervention (ECI) module for multimodal deconfounding, which involves front-door intervention for vision and back-door intervention for language. Extensive experiments on two VideoQG datasets demonstrate the superiority of our CRA in discovering visually grounded content and achieving robust question reasoning. Codes are available at https://github.com/WissingChen/CRA-GQA.

Paperid: 1318, https://arxiv.org/pdf/2503.07608.pdf GitHub

Abstract:
OpenAI o1 and DeepSeek R1 achieve or even surpass human expert-level performance in complex domains like mathematics and science, with reinforcement learning (RL) and reasoning playing a crucial role. In autonomous driving, recent end-to-end models have greatly improved planning performance but still struggle with long-tailed problems due to limited common sense and reasoning abilities. Some studies integrate vision-language models (VLMs) into autonomous driving, but they typically rely on pre-trained models with simple supervised fine-tuning (SFT) on driving data, without further exploration of training strategies or optimizations specifically tailored for planning. In this paper, we propose AlphaDrive, a RL and reasoning framework for VLMs in autonomous driving. AlphaDrive introduces four GRPO-based RL rewards tailored for planning and employs a two-stage planning reasoning training strategy that combines SFT with RL. As a result, AlphaDrive significantly improves both planning performance and training efficiency compared to using only SFT or without reasoning. Moreover, we are also excited to discover that, following RL training, AlphaDrive exhibits some emergent multimodal planning capabilities, which is critical for improving driving safety and efficiency. To the best of our knowledge, AlphaDrive is the first to integrate GRPO-based RL with planning reasoning into autonomous driving. Code will be released to facilitate future research.

Paperid: 1319, https://arxiv.org/pdf/2503.07593.pdf GitHub

Abstract:
Open-vocabulary 3D object detection (OV-3DOD) aims at localizing and classifying novel objects beyond closed sets. The recent success of vision-language models (VLMs) has demonstrated their remarkable capabilities to understand open vocabularies. Existing works that leverage VLMs for 3D object detection (3DOD) generally resort to representations that lose the rich scene context required for 3D perception. To address this problem, we propose in this paper a hierarchical framework, named HCMA, to simultaneously learn local object and global scene information for OV-3DOD. Specifically, we first design a Hierarchical Data Integration (HDI) approach to obtain coarse-to-fine 3D-image-text data, which is fed into a VLM to extract object-centric knowledge. To facilitate the association of feature hierarchies, we then propose an Interactive Cross-Modal Alignment (ICMA) strategy to establish effective intra-level and inter-level feature connections. To better align features across different levels, we further propose an Object-Focusing Context Adjustment (OFCA) module to refine multi-level features by emphasizing object-related features. Extensive experiments demonstrate that the proposed method outperforms SOTA methods on the existing OV-3DOD benchmarks. It also achieves promising OV-3DOD results even without any 3D annotations.

Paperid: 1320, https://arxiv.org/pdf/2503.07591.pdf GitHub

Abstract:
Visual instruction tuning (VIT) for large vision-language models (LVLMs) requires training on expansive datasets of image-instruction pairs, which can be costly. Recent efforts in VIT data selection aim to select a small subset of high-quality image-instruction pairs, reducing VIT runtime while maintaining performance comparable to full-scale training. However, a major challenge often overlooked is that generating instructions from unlabeled images for VIT is highly expensive. Most existing VIT datasets rely heavily on human annotations or paid services like the GPT API, which limits users with constrained resources from creating VIT datasets for custom applications. To address this, we introduce Pre-Instruction Data Selection (PreSel), a more practical data selection paradigm that directly selects the most beneficial unlabeled images and generates instructions only for the selected images. PreSel first estimates the relative importance of each vision task within VIT datasets to derive task-wise sampling budgets. It then clusters image features within each task, selecting the most representative images with the budget. This approach reduces computational overhead for both instruction generation during VIT data formation and LVLM fine-tuning. By generating instructions for only 15% of the images, PreSel achieves performance comparable to full-data VIT on the LLaVA-1.5 and Vision-Flan datasets. The link to our project page: https://bardisafa.github.io/PreSel

Paperid: 1321, https://arxiv.org/pdf/2503.07588.pdf GitHub

Abstract:
Efficient vision-language understanding of large Remote Sensing Images (RSIs) is meaningful but challenging. Current Large Vision-Language Models (LVLMs) typically employ limited pre-defined grids to process images, leading to information loss when handling gigapixel RSIs. Conversely, using unlimited grids significantly increases computational costs. To preserve image details while reducing computational complexity, we propose a text-guided token pruning method with Dynamic Image Pyramid (DIP) integration. Our method introduces: (i) a Region Focus Module (RFM) that leverages text-aware region localization capability to identify critical vision tokens, and (ii) a coarse-to-fine image tile selection and vision token pruning strategy based on DIP, which is guided by RFM outputs and avoids directly processing the entire large imagery. Additionally, existing benchmarks for evaluating LVLMs' perception ability on large RSI suffer from limited question diversity and constrained image sizes. We construct a new benchmark named LRS-VQA, which contains 7,333 QA pairs across 8 categories, with image length up to 27,328 pixels. Our method outperforms existing high-resolution strategies on four datasets using the same data. Moreover, compared to existing token reduction methods, our approach demonstrates higher efficiency under high-resolution settings. Dataset and code are in https://github.com/VisionXLab/LRS-VQA.

Paperid: 1322, https://arxiv.org/pdf/2503.07575.pdf GitHub

Abstract:
This research investigates both explicit and implicit social biases exhibited by Vision-Language Models (VLMs). The key distinction between these bias types lies in the level of awareness: explicit bias refers to conscious, intentional biases, while implicit bias operates subconsciously. To analyze explicit bias, we directly pose questions to VLMs related to gender and racial differences: (1) Multiple-choice questions based on a given image (e.g., "What is the education level of the person in the image?") (2) Yes-No comparisons using two images (e.g., "Is the person in the first image more educated than the person in the second image?") For implicit bias, we design tasks where VLMs assist users but reveal biases through their responses: (1) Image description tasks: Models are asked to describe individuals in images, and we analyze disparities in textual cues across demographic groups. (2) Form completion tasks: Models draft a personal information collection form with 20 attributes, and we examine correlations among selected attributes for potential biases. We evaluate Gemini-1.5, GPT-4V, GPT-4o, LLaMA-3.2-Vision and LLaVA-v1.6. Our code and data are publicly available at https://github.com/uscnlp-lime/VisBias.

Paperid: 1323, https://arxiv.org/pdf/2503.07503.pdf GitHub

Abstract:
Reasoning segmentation is a challenging vision-language task that aims to output the segmentation mask with respect to a complex, implicit, and even non-visual query text. Previous works incorporated multimodal Large Language Models (MLLMs) with segmentation models to approach the difficult problem. However, their segmentation quality often falls short in complex cases, particularly when dealing with out-of-domain objects with intricate structures, blurry boundaries, occlusions, or high similarity with surroundings. In this paper, we introduce ThinkFirst, a training-free reasoning segmentation framework that leverages GPT's chain of thought to address these challenging cases. Our approach allows GPT-4o or other powerful MLLMs to generate a detailed, chain-of-thought description of an image. This summarized description is then passed to a language-instructed segmentation assistant to aid the segmentation process. Our framework allows users to easily interact with the segmentation agent using multimodal inputs, such as easy text and image scribbles, for successive refinement or communication. We evaluate the performance of ThinkFirst on diverse objects. Extensive experiments show that, this zero-shot-CoT approach significantly improves the vanilla reasoning segmentation agent, both qualitatively and quantitatively, while being less sensitive or critical to user-supplied prompts after Thinking First.

Paperid: 1324, https://arxiv.org/pdf/2503.07485.pdf GitHub

Abstract:
Lane topology extraction involves detecting lanes and traffic elements and determining their relationships, a key perception task for mapless autonomous driving. This task requires complex reasoning, such as determining whether it is possible to turn left into a specific lane. To address this challenge, we introduce neuro-symbolic methods powered by vision-language foundation models (VLMs). Existing approaches have notable limitations: (1) Dense visual prompting with VLMs can achieve strong performance but is costly in terms of both financial resources and carbon footprint, making it impractical for robotics applications. (2) Neuro-symbolic reasoning methods for 3D scene understanding fail to integrate visual inputs when synthesizing programs, making them ineffective in handling complex corner cases. To this end, we propose a fast-slow neuro-symbolic lane topology extraction algorithm, named Chameleon, which alternates between a fast system that directly reasons over detected instances using synthesized programs and a slow system that utilizes a VLM with a chain-of-thought design to handle corner cases. Chameleon leverages the strengths of both approaches, providing an affordable solution while maintaining high performance. We evaluate the method on the OpenLane-V2 dataset, showing consistent improvements across various baseline detectors. Our code, data, and models are publicly available at https://github.com/XR-Lee/neural-symbolic

Paperid: 1325, https://arxiv.org/pdf/2503.07478.pdf GitHub

Abstract:
Although large visual-language models (LVLMs) have demonstrated strong performance in multimodal tasks, errors may occasionally arise due to biases during the reasoning process. Recently, reward models (RMs) have become increasingly pivotal in the reasoning process. Specifically, process RMs evaluate each reasoning step, outcome RMs focus on the assessment of reasoning results, and critique RMs perform error analysis on the entire reasoning process, followed by corrections. However, existing benchmarks for vision-language RMs (VLRMs) typically assess only a single aspect of their capabilities (e.g., distinguishing between two answers), thus limiting the all-round evaluation and restricting the development of RMs in the visual-language domain. To address this gap, we propose a comprehensive and challenging benchmark, dubbed as VLRMBench, encompassing 12,634 questions. VLRMBench is constructed based on three distinct types of datasets, covering mathematical reasoning, hallucination understanding, and multi-image understanding. We design 12 tasks across three major categories, focusing on evaluating VLRMs in the aspects of process understanding, outcome judgment, and critique generation. Extensive experiments are conducted on 21 open-source models and 5 advanced closed-source models, highlighting the challenges posed by VLRMBench. For instance, in the `Forecasting Future', a binary classification task, the advanced GPT-4o achieves only a 76.0% accuracy. Additionally, we perform comprehensive analytical studies, offering valuable insights for the future development of VLRMs. We anticipate that VLRMBench will serve as a pivotal benchmark in advancing VLRMs. Code and datasets will be available at https://github.com/JCruan519/VLRMBench.

Paperid: 1326, https://arxiv.org/pdf/2503.07413.pdf GitHub

Abstract:
Multimodal Large Language Models (MLLMs) demonstrate robust zero-shot capabilities across diverse vision-language tasks after training on mega-scale datasets. However, dense prediction tasks, such as semantic segmentation and keypoint detection, pose significant challenges for MLLMs when represented solely as text outputs. Simultaneously, current MLLMs utilizing latent embeddings for visual task decoding generally demonstrate limited adaptability to both multi-task learning and multi-granularity scenarios. In this work, we present REF-VLM, an end-to-end framework for unified training of various visual decoding tasks. To address complex visual decoding scenarios, we introduce the Triplet-Based Referring Paradigm (TRP), which explicitly decouples three critical dimensions in visual decoding tasks through a triplet structure: concepts, decoding types, and targets. TRP employs symbolic delimiters to enforce structured representation learning, enhancing the parsability and interpretability of model outputs. Additionally, we construct Visual-Task Instruction Following Dataset (VTInstruct), a large-scale multi-task dataset containing over 100 million multimodal dialogue samples across 25 task types. Beyond text inputs and outputs, VT-Instruct incorporates various visual prompts such as point, box, scribble, and mask, and generates outputs composed of text and visual units like box, keypoint, depth and mask. The combination of different visual prompts and visual units generates a wide variety of task types, expanding the applicability of REF-VLM significantly. Both qualitative and quantitative experiments demonstrate that our REF-VLM outperforms other MLLMs across a variety of standard benchmarks. The code, dataset, and demo available at https://github.com/MacavityT/REF-VLM.

Paperid: 1327, https://arxiv.org/pdf/2503.06626.pdf GitHub

Abstract:
We propose DiffCLIP, a novel vision-language model that extends the differential attention mechanism to CLIP architectures. Differential attention was originally developed for large language models to amplify relevant context while canceling out noisy information. In this work, we integrate this mechanism into CLIP's dual encoder (image and text) framework. With minimal additional parameters, DiffCLIP achieves superior performance on image-text understanding tasks. Across zero-shot classification, retrieval, and robustness benchmarks, DiffCLIP consistently outperforms baseline CLIP models. Notably, these gains come with negligible computational overhead, demonstrating that differential attention can significantly enhance multi-modal representations without sacrificing efficiency. Code can be found at https://github.com/hammoudhasan/DiffCLIP.

Paperid: 1328, https://arxiv.org/pdf/2503.06621.pdf GitHub

Abstract:
The consistency between the semantic information provided by the multi-modal reference and the tracked object is crucial for visual-language (VL) tracking. However, existing VL tracking frameworks rely on static multi-modal references to locate dynamic objects, which can lead to semantic discrepancies and reduce the robustness of the tracker. To address this issue, we propose a novel vision-language tracking framework, named DUTrack, which captures the latest state of the target by dynamically updating multi-modal references to maintain consistency. Specifically, we introduce a Dynamic Language Update Module, which leverages a large language model to generate dynamic language descriptions for the object based on visual features and object category information. Then, we design a Dynamic Template Capture Module, which captures the regions in the image that highly match the dynamic language descriptions. Furthermore, to ensure the efficiency of description generation, we design an update strategy that assesses changes in target displacement, scale, and other factors to decide on updates. Finally, the dynamic template and language descriptions that record the latest state of the target are used to update the multi-modal references, providing more accurate reference information for subsequent inference and enhancing the robustness of the tracker. DUTrack achieves new state-of-the-art performance on four mainstream vision-language and two vision-only tracking benchmarks, including LaSOT, LaSOT$_{\rm{ext}}$, TNL2K, OTB99-Lang, GOT-10K, and UAV123. Code and models are available at https://github.com/GXNU-ZhongLab/DUTrack.

Paperid: 1329, https://arxiv.org/pdf/2503.06312.pdf GitHub

Abstract:
Earth observation (EO) spans a broad spectrum of modalities, including optical, radar, multispectral, and hyperspectral data, each capturing distinct environmental signals. However, current vision-language models in EO, particularly CLIP-based variants, remain confined to individual modalities, limiting generalization and scalability across diverse tasks. We present DOFA-CLIP (Dynamic-One-For-All CLIP), a unified vision-language foundation model that dynamically adapts to EO modalities with flexible spectral configurations through a single Transformer backbone. Our approach introduces three key contributions: 1) the construction of GeoLangBind-2M, a large-scale EO image-text dataset covering six heterogeneous modalities with rich natural language descriptions; 2) a novel training strategy called VECT (Vision-models Enhanced Contrastive Text-image pretraining), which enhances the spatial awareness of CLIP features with multiple vision foundation models; and 3) a Modality-aware Knowledge Agglomeration (MaKA) module that refines feature distillation with modality-specific awareness. DOFA-CLIP achieves state-of-the-art zero-shot performance across a wide range of EO benchmarks, including unseen modalities and a diverse number of input spectral bands. Together, these contributions establish a scalable foundation for multimodal EO understanding and open new avenues for integrating heterogeneous EO data with large language models. Code and datasets will be released. Code and datasets are publicly available.

Paperid: 1330, https://arxiv.org/pdf/2503.06169.pdf GitHub

Abstract:
Vision-Language Models (VLMs) have advanced multi-modal tasks like image captioning, visual question answering, and reasoning. However, they often generate hallucinated outputs inconsistent with the visual context or prompt, limiting reliability in critical applications like autonomous driving and medical imaging. Existing studies link hallucination to statistical biases, language priors, and biased feature learning but lack a structured causal understanding. In this work, we introduce a causal perspective to analyze and mitigate hallucination in VLMs. We hypothesize that hallucination arises from unintended direct influences of either the vision or text modality, bypassing proper multi-modal fusion. To address this, we construct a causal graph for VLMs and employ counterfactual analysis to estimate the Natural Direct Effect (NDE) of vision, text, and their cross-modal interaction on the output. We systematically identify and mitigate these unintended direct effects to ensure that responses are primarily driven by genuine multi-modal fusion. Our approach consists of three steps: (1) designing structural causal graphs to distinguish correct fusion pathways from spurious modality shortcuts, (2) estimating modality-specific and cross-modal NDE using perturbed image representations, hallucinated text embeddings, and degraded visual inputs, and (3) implementing a test-time intervention module to dynamically adjust the model's dependence on each modality. Experimental results demonstrate that our method significantly reduces hallucination while preserving task performance, providing a robust and interpretable framework for improving VLM reliability. To enhance accessibility and reproducibility, our code is publicly available at https://github.com/TREE985/Treble-Counterfactual-VLMs.

Paperid: 1331, https://arxiv.org/pdf/2503.06030.pdf GitHub

Abstract:
Computed tomography (CT) is extensively used for accurate visualization and segmentation of organs and lesions. While deep learning models such as convolutional neural networks (CNNs) and vision transformers (ViTs) have significantly improved CT image analysis, their performance often declines when applied to diverse, real-world clinical data. Although foundation models offer a broader and more adaptable solution, their potential is limited due to the challenge of obtaining large-scale, voxel-level annotations for medical images. In response to these challenges, prompting-based models using visual or text prompts have emerged. Visual-prompting methods, such as the Segment Anything Model (SAM), still require significant manual input and can introduce ambiguity when applied to clinical scenarios. Instead, foundation models that use text prompts offer a more versatile and clinically relevant approach. Notably, current text-prompt models, such as the CLIP-Driven Universal Model, are limited to text prompts already encountered during training and struggle to process the complex and diverse scenarios of real-world clinical applications. Instead of fine-tuning models trained from natural imaging, we propose OpenVocabCT, a vision-language model pretrained on large-scale 3D CT images for universal text-driven segmentation. Using the large-scale CT-RATE dataset, we decompose the diagnostic reports into fine-grained, organ-level descriptions using large language models for multi-granular contrastive learning. We evaluate our OpenVocabCT on downstream segmentation tasks across nine public datasets for organ and tumor segmentation, demonstrating the superior performance of our model compared to existing methods. All code, datasets, and models will be publicly released at https://github.com/ricklisz/OpenVocabCT.

Paperid: 1332, https://arxiv.org/pdf/2503.05776.pdf GitHub

Abstract:
Despite the remarkable performance of vision language models (VLMs) such as Contrastive Language Image Pre-training (CLIP), the large size of these models is a considerable obstacle to their use in federated learning (FL) systems where the parameters of local client models need to be transferred to a global server for aggregation. Another challenge in FL is the heterogeneity of data from different clients, which affects the generalization performance of the solution. In addition, natural pre-trained VLMs exhibit poor generalization ability in the medical datasets, suggests there exists a domain gap. To solve these issues, we introduce a novel method for the Federated Adversarial Adaptation (FAA) of CLIP. Our method, named FAA-CLIP, handles the large communication costs of CLIP using a light-weight feature adaptation module (FAM) for aggregation, effectively adapting this VLM to each client's data while greatly reducing the number of parameters to transfer. By keeping CLIP frozen and only updating the FAM parameters, our method is also computationally efficient. Unlike existing approaches, our FAA-CLIP method directly addresses the problem of domain shifts across clients via a domain adaptation (DA) module. This module employs a domain classifier to predict if a given sample is from the local client or the global server, allowing the model to learn domain-invariant representations. Extensive experiments on six different datasets containing both natural and medical images demonstrate that FAA-CLIP can generalize well on both natural and medical datasets compared to recent FL approaches. Our codes are available at https://github.com/AIPMLab/FAA-CLIP.

Paperid: 1333, https://arxiv.org/pdf/2503.05383.pdf GitHub

Abstract:
We introduce Attentive VLM Agent (AVA), a multimodal StarCraft II agent that aligns artificial agent perception with the human gameplay experience. Traditional frameworks such as SMAC rely on abstract state representations that diverge significantly from human perception, limiting the ecological validity of agent behavior. Our agent addresses this limitation by incorporating RGB visual inputs and natural language observations that more closely simulate human cognitive processes during gameplay. The AVA architecture consists of three integrated components: (1) a vision-language model enhanced with specialized self-attention mechanisms for strategic unit targeting and battlefield assessment, (2) a retrieval-augmented generation system that leverages domain-specific StarCraft II knowledge to inform tactical decisions, and (3) a dynamic role-based task distribution system that enables coordinated multi-agent behavior. The experimental evaluation in our proposed AVACraft environment, which contains 21 multimodal StarCraft II scenarios, demonstrates that AVA powered by foundation models (specifically Qwen-VL and GPT-4o) can execute complex tactical maneuvers without explicit training, achieving comparable performance to traditional MARL methods that require substantial training iterations. This work establishes a foundation for developing human-aligned StarCraft II agents and advances the broader research agenda of multimodal game AI. Our implementation is available at https://github.com/camel-ai/VLM-Play-StarCraft2.

Paperid: 1334, https://arxiv.org/pdf/2503.04982.pdf GitHub

Abstract:
Despite recent efforts in understanding the compression impact on large language models (LLMs) in terms of their downstream task performance and trustworthiness on relatively simpler uni-modal benchmarks (for example, question answering, common sense reasoning), their detailed study on multi-modal Large Vision-Language Models (LVLMs) is yet to be unveiled. Towards mitigating this gap, we present LVLM-Compress-Bench, a framework to first thoroughly study the broad impact of compression on the generative performance of LVLMs with multi-modal input driven tasks. In specific, we consider two major classes of compression for autoregressive models, namely KV cache and weight compression, for the dynamically growing intermediate cache and static weights, respectively. We use four LVLM variants of the popular LLaVA framework to present our analysis via integrating various state-of-the-art KV and weight compression methods including uniform, outlier-reduced, and group quantization for the KV cache and weights. With this framework we demonstrate on ten different multi-modal datasets with different capabilities including recognition, knowledge, language generation, spatial awareness, visual reasoning, hallucination and visual illusion identification, toxicity, stereotypes and bias. In specific, our framework demonstrates the compression impact on both general and ethically critical metrics leveraging a combination of real world and synthetic datasets to encompass diverse societal intersectional attributes. Extensive experimental evaluations yield diverse and intriguing observations on the behavior of LVLMs at different quantization budget of KV and weights, in both maintaining and losing performance as compared to the baseline model with FP16 data format. Code will be open-sourced at https://github.com/opengear-project/LVLM-compress-bench.

Paperid: 1335, https://arxiv.org/pdf/2503.04724.pdf GitHub

Abstract:
Recent advancements in speech-to-speech dialogue systems leverage LLMs for multimodal interactions, yet they remain hindered by fine-tuning requirements, high computational overhead, and text-speech misalignment. Existing speech-enabled LLMs often degrade conversational quality by modifying the LLM, thereby compromising its linguistic capabilities. In contrast, we propose LLMVoX, a lightweight 30M-parameter, LLM-agnostic, autoregressive streaming TTS system that generates high-quality speech with low latency, while fully preserving the capabilities of the base LLM. Our approach achieves a significantly lower Word Error Rate compared to speech-enabled LLMs, while operating at comparable latency and UTMOS score. By decoupling speech synthesis from LLM processing via a multi-queue token streaming system, LLMVoX supports seamless, infinite-length dialogues. Its plug-and-play design also facilitates extension to various tasks with different backbones. Furthermore, LLMVoX generalizes to new languages with only dataset adaptation, attaining a low Character Error Rate on an Arabic speech task. Additionally, we have integrated LLMVoX with a Vision-Language Model to create an omni-model with speech, text, and vision capabilities, without requiring additional multimodal training. Our code base and project page is available at https://mbzuai-oryx.github.io/LLMVoX .

Paperid: 1336, https://arxiv.org/pdf/2503.04592.pdf GitHub

Abstract:
Remote Sensing Image Captioning (RSIC) is a cross-modal field bridging vision and language, aimed at automatically generating natural language descriptions of features and scenes in remote sensing imagery. Despite significant advances in developing sophisticated methods and large-scale datasets for training vision-language models (VLMs), two critical challenges persist: the scarcity of non-English descriptive datasets and the lack of multilingual capability evaluation for models. These limitations fundamentally impede the progress and practical deployment of RSIC, particularly in the era of large VLMs. To address these challenges, this paper presents several significant contributions to the field. First, we introduce and analyze BRSIC (Bilingual Remote Sensing Image Captioning), a comprehensive bilingual dataset that enriches three established English RSIC datasets with Chinese descriptions, encompassing 13,634 images paired with 68,170 bilingual captions. Building upon this foundation, we develop a systematic evaluation framework that addresses the prevalent inconsistency in evaluation protocols, enabling rigorous assessment of model performance through standardized retraining procedures on BRSIC. Furthermore, we present an extensive empirical study of eight state-of-the-art large vision-language models (LVLMs), examining their capabilities across multiple paradigms including zero-shot inference, supervised fine-tuning, and multi-lingual training. This comprehensive evaluation provides crucial insights into the strengths and limitations of current LVLMs in handling multilingual remote sensing tasks. Additionally, our cross-dataset transfer experiments reveal interesting findings. The code and data will be available at https://github.com/mrazhou/BRSIC.

Paperid: 1337, https://arxiv.org/pdf/2503.04250.pdf GitHub

Abstract:
We present Vinci, a vision-language system designed to provide real-time, comprehensive AI assistance on portable devices. At its core, Vinci leverages EgoVideo-VL, a novel model that integrates an egocentric vision foundation model with a large language model (LLM), enabling advanced functionalities such as scene understanding, temporal grounding, video summarization, and future planning. To enhance its utility, Vinci incorporates a memory module for processing long video streams in real time while retaining contextual history, a generation module for producing visual action demonstrations, and a retrieval module that bridges egocentric and third-person perspectives to provide relevant how-to videos for skill acquisition. Unlike existing systems that often depend on specialized hardware, Vinci is hardware-agnostic, supporting deployment across a wide range of devices, including smartphones and wearable cameras. In our experiments, we first demonstrate the superior performance of EgoVideo-VL on multiple public benchmarks, showcasing its vision-language reasoning and contextual understanding capabilities. We then conduct a series of user studies to evaluate the real-world effectiveness of Vinci, highlighting its adaptability and usability in diverse scenarios. We hope Vinci can establish a new framework for portable, real-time egocentric AI systems, empowering users with contextual and actionable insights. Including the frontend, backend, and models, all codes of Vinci are available at https://github.com/OpenGVLab/vinci.

Paperid: 1338, https://arxiv.org/pdf/2503.04163.pdf GitHub

Abstract:
The emergence of vision-language-action (VLA) models has given rise to foundation models for robot manipulation. Although these models have achieved significant improvements, their generalization in multi-task manipulation remains limited. This study proposes a VLA model-expert collaboration framework that leverages a limited number of expert actions to enhance VLA model performance. This approach reduces expert workload relative to manual operation while simultaneously improving the reliability and generalization of VLA models. Furthermore, manipulation data collected during collaboration can further refine the VLA model, while human participants concurrently enhance their skills. This bi-directional learning loop boosts the overall performance of the collaboration system. Experimental results across various VLA models demonstrate the effectiveness of the proposed system in collaborative manipulation and learning, as evidenced by improved success rates across tasks. Additionally, validation using a brain-computer interface (BCI) indicates that the collaboration system enhances the efficiency of low-speed action systems by involving VLA model during manipulation. These promising results pave the way for advancing human-robot interaction in the era of foundation models for robotics. (Project website: https://aoqunjin.github.io/Expert-VLA/)

Paperid: 1339, https://arxiv.org/pdf/2503.03987.pdf GitHub

Abstract:
Recently, Multimodal Large Language Models (MLLMs) have gained significant attention for their remarkable ability to process and analyze non-textual data, such as images, videos, and audio. Notably, several adaptations of general-domain MLLMs to the medical field have been explored, including LLaVA-Med. However, these medical adaptations remain insufficiently advanced in understanding and interpreting retinal images. In contrast, medical experts emphasize the importance of quantitative analyses for disease detection and interpretation. This underscores a gap between general-domain and medical-domain MLLMs: while general-domain MLLMs excel in broad applications, they lack the specialized knowledge necessary for precise diagnostic and interpretative tasks in the medical field. To address these challenges, we introduce \textit{RetinalGPT}, a multimodal conversational assistant for clinically preferred quantitative analysis of retinal images. Specifically, we achieve this by compiling a large retinal image dataset, developing a novel data pipeline, and employing customized visual instruction tuning to enhance both retinal analysis and enrich medical knowledge. In particular, RetinalGPT outperforms MLLM in the generic domain by a large margin in the diagnosis of retinal diseases in 8 benchmark retinal datasets. Beyond disease diagnosis, RetinalGPT features quantitative analyses and lesion localization, representing a pioneering step in leveraging LLMs for an interpretable and end-to-end clinical research framework. The code is available at https://github.com/Retinal-Research/RetinalGPT

Paperid: 1340, https://arxiv.org/pdf/2503.03734.pdf GitHub

Abstract:
Vision-Language-Action (VLA) models aim to predict robotic actions based on visual observations and language instructions. Existing approaches require fine-tuning pre-trained visionlanguage models (VLMs) as visual and language features are independently fed into downstream policies, degrading the pre-trained semantic alignments. We propose OTTER, a novel VLA architecture that leverages these existing alignments through explicit, text-aware visual feature extraction. Instead of processing all visual features, OTTER selectively extracts and passes only task-relevant visual features that are semantically aligned with the language instruction to the policy transformer. This allows OTTER to keep the pre-trained vision-language encoders frozen. Thereby, OTTER preserves and utilizes the rich semantic understanding learned from large-scale pre-training, enabling strong zero-shot generalization capabilities. In simulation and real-world experiments, OTTER significantly outperforms existing VLA models, demonstrating strong zeroshot generalization to novel objects and environments. Video, code, checkpoints, and dataset: https://ottervla.github.io/.

Paperid: 1341, https://arxiv.org/pdf/2503.03196.pdf GitHub

Abstract:
Graphical User Interface (GUI) agents show amazing abilities in assisting human-computer interaction, automating human user's navigation on digital devices. An ideal GUI agent is expected to achieve high accuracy, low latency, and compatibility for different GUI platforms. Recent vision-based approaches have shown promise by leveraging advanced Vision Language Models (VLMs). While they generally meet the requirements of compatibility and low latency, these vision-based GUI agents tend to have low accuracy due to their limitations in element grounding. To address this issue, we propose $\textbf{SpiritSight}$, a vision-based, end-to-end GUI agent that excels in GUI navigation tasks across various GUI platforms. First, we create a multi-level, large-scale, high-quality GUI dataset called $\textbf{GUI-Lasagne}$ using scalable methods, empowering SpiritSight with robust GUI understanding and grounding capabilities. Second, we introduce the $\textbf{Universal Block Parsing (UBP)}$ method to resolve the ambiguity problem in dynamic high-resolution of visual inputs, further enhancing SpiritSight's ability to ground GUI objects. Through these efforts, SpiritSight agent outperforms other advanced methods on diverse GUI benchmarks, demonstrating its superior capability and compatibility in GUI navigation tasks. Models and datasets are available at https://hzhiyuan.github.io/SpiritSight-Agent.

Paperid: 1342, https://arxiv.org/pdf/2503.02865.pdf GitHub

Abstract:
In this paper, we introduce FairSense-AI: a multimodal framework designed to detect and mitigate bias in both text and images. By leveraging Large Language Models (LLMs) and Vision-Language Models (VLMs), FairSense-AI uncovers subtle forms of prejudice or stereotyping that can appear in content, providing users with bias scores, explanatory highlights, and automated recommendations for fairness enhancements. In addition, FairSense-AI integrates an AI risk assessment component that aligns with frameworks like the MIT AI Risk Repository and NIST AI Risk Management Framework, enabling structured identification of ethical and safety concerns. The platform is optimized for energy efficiency via techniques such as model pruning and mixed-precision computation, thereby reducing its environmental footprint. Through a series of case studies and applications, we demonstrate how FairSense-AI promotes responsible AI use by addressing both the social dimension of fairness and the pressing need for sustainability in large-scale AI deployments. https://vectorinstitute.github.io/FairSense-AI, https://pypi.org/project/fair-sense-ai/ (Sustainability , Responsible AI , Large Language Models , Vision Language Models , Ethical AI , Green AI)

Paperid: 1343, https://arxiv.org/pdf/2503.02597.pdf GitHub

Abstract:
Recent Multimodal Large Language Models (MLLMs) have demonstrated significant progress in perceiving and reasoning over multimodal inquiries, ushering in a new research era for foundation models. However, vision-language misalignment in MLLMs has emerged as a critical challenge, where the textual responses generated by these models are not factually aligned with the given text-image inputs. Existing efforts to address vision-language misalignment have focused on developing specialized vision-language connectors or leveraging visual instruction tuning from diverse domains. In this paper, we tackle this issue from a fundamental yet unexplored perspective by revisiting the core architecture of MLLMs. Most MLLMs are typically built on decoder-only LLMs consisting of a causal attention mechanism, which limits the ability of the earlier modalities (e.g., images) to incorporate information from the latter modalities (e.g., text). To address this problem, we propose \MapleLeaf AKI, a novel MLLM that unlocks causal attention into modality-mutual attention (MMA) to enable image tokens to attend to text tokens. This simple yet effective design allows AKI to achieve superior performance in 12 multimodal understanding benchmarks (+7.2% on average) without introducing additional parameters and increasing training time. Our MMA design is intended to be generic, allowing for application across various modalities, and scalable to accommodate diverse multimodal scenarios. The code and model are publicly available at https://github.com/sony/aki to encourage further advancements in MLLMs across various directions.

Paperid: 1344, https://arxiv.org/pdf/2503.02579.pdf GitHub

Abstract:
Operating rooms (ORs) are complex, high-stakes environments requiring precise understanding of interactions among medical staff, tools, and equipment for enhancing surgical assistance, situational awareness, and patient safety. Current datasets fall short in scale, realism and do not capture the multimodal nature of OR scenes, limiting progress in OR modeling. To this end, we introduce MM-OR, a realistic and large-scale multimodal spatiotemporal OR dataset, and the first dataset to enable multimodal scene graph generation. MM-OR captures comprehensive OR scenes containing RGB-D data, detail views, audio, speech transcripts, robotic logs, and tracking data and is annotated with panoptic segmentations, semantic scene graphs, and downstream task labels. Further, we propose MM2SG, the first multimodal large vision-language model for scene graph generation, and through extensive experiments, demonstrate its ability to effectively leverage multimodal inputs. Together, MM-OR and MM2SG establish a new benchmark for holistic OR understanding, and open the path towards multimodal scene analysis in complex, high-stakes environments. Our code, and data is available at https://github.com/egeozsoy/MM-OR.

Paperid: 1345, https://arxiv.org/pdf/2503.02358.pdf GitHub

Abstract:
Large Vision Language Models (LVLMs) have demonstrated remarkable abilities in understanding and reasoning about both visual and textual information. However, existing evaluation methods for LVLMs, primarily based on benchmarks like Visual Question Answering and image captioning, often fail to capture the full scope of LVLMs' capabilities. These benchmarks are limited by issues such as inadequate assessment of detailed visual perception, data contamination, and a lack of focus on multi-turn reasoning. To address these challenges, we propose \method{}, a game-based evaluation framework designed to provide a comprehensive assessment of LVLMs' cognitive and reasoning skills in structured environments. \method{} uses a set of games to evaluate LVLMs on four core tasks: Perceiving, Question Answering, Rule Following, and End-to-End Playing, with each target task designed to assess specific abilities, including visual perception, reasoning, decision-making, etc. Based on this framework, we conduct extensive experiments that explore the limitations of current LVLMs, such as handling long structured outputs and perceiving detailed and dense elements. Code and data are publicly available at https://github.com/xinke-wang/LVLM-Playground.

Paperid: 1346, https://arxiv.org/pdf/2503.02247.pdf GitHub

Abstract:
Object Goal Navigation-requiring an agent to locate a specific object in an unseen environment-remains a core challenge in embodied AI. Although recent progress in Vision-Language Model (VLM)-based agents has demonstrated promising perception and decision-making abilities through prompting, none has yet established a fully modular world model design that reduces risky and costly interactions with the environment by predicting the future state of the world. We introduce WMNav, a novel World Model-based Navigation framework powered by Vision-Language Models (VLMs). It predicts possible outcomes of decisions and builds memories to provide feedback to the policy module. To retain the predicted state of the environment, WMNav proposes the online maintained Curiosity Value Map as part of the world model memory to provide dynamic configuration for navigation policy. By decomposing according to a human-like thinking process, WMNav effectively alleviates the impact of model hallucination by making decisions based on the feedback difference between the world model plan and observation. To further boost efficiency, we implement a two-stage action proposer strategy: broad exploration followed by precise localization. Extensive evaluation on HM3D and MP3D validates WMNav surpasses existing zero-shot benchmarks in both success rate and exploration efficiency (absolute improvement: +3.2% SR and +3.2% SPL on HM3D, +13.5% SR and +1.1% SPL on MP3D). Project page: https://b0b8k1ng.github.io/WMNav/.

Paperid: 1347, https://arxiv.org/pdf/2503.02034.pdf GitHub

Abstract:
Medical imaging plays a pivotal role in modern healthcare, with computed tomography pulmonary angiography (CTPA) being a critical tool for diagnosing pulmonary embolism and other thoracic conditions. However, the complexity of interpreting CTPA scans and generating accurate radiology reports remains a significant challenge. This paper introduces Abn-BLIP (Abnormality-aligned Bootstrapping Language-Image Pretraining), an advanced diagnosis model designed to align abnormal findings to generate the accuracy and comprehensiveness of radiology reports. By leveraging learnable queries and cross-modal attention mechanisms, our model demonstrates superior performance in detecting abnormalities, reducing missed findings, and generating structured reports compared to existing methods. Our experiments show that Abn-BLIP outperforms state-of-the-art medical vision-language models and 3D report generation methods in both accuracy and clinical relevance. These results highlight the potential of integrating multimodal learning strategies for improving radiology reporting. The source code is available at https://github.com/zzs95/abn-blip.

Paperid: 1348, https://arxiv.org/pdf/2503.01785.pdf GitHub

Abstract:
Reinforcement Fine-Tuning (RFT) in Large Reasoning Models like OpenAI o1 learns from feedback on its answers, which is especially useful in applications when fine-tuning data is scarce. Recent open-source work like DeepSeek-R1 demonstrates that reinforcement learning with verifiable reward is one key direction in reproducing o1. While the R1-style model has demonstrated success in language models, its application in multi-modal domains remains under-explored. This work introduces Visual Reinforcement Fine-Tuning (Visual-RFT), which further extends the application areas of RFT on visual tasks. Specifically, Visual-RFT first uses Large Vision-Language Models (LVLMs) to generate multiple responses containing reasoning tokens and final answers for each input, and then uses our proposed visual perception verifiable reward functions to update the model via the policy optimization algorithm such as Group Relative Policy Optimization (GRPO). We design different verifiable reward functions for different perception tasks, such as the Intersection over Union (IoU) reward for object detection. Experimental results on fine-grained image classification, few-shot object detection, reasoning grounding, as well as open-vocabulary object detection benchmarks show the competitive performance and advanced generalization ability of Visual-RFT compared with Supervised Fine-tuning (SFT). For example, Visual-RFT improves accuracy by $24.3\%$ over the baseline in one-shot fine-grained image classification with around 100 samples. In few-shot object detection, Visual-RFT also exceeds the baseline by $21.9$ on COCO's two-shot setting and $15.4$ on LVIS. Our Visual-RFT represents a paradigm shift in fine-tuning LVLMs, offering a data-efficient, reward-driven approach that enhances reasoning and adaptability for domain-specific tasks.

Paperid: 1349, https://arxiv.org/pdf/2503.01773.pdf GitHub

Abstract:
Large Vision Language Models (VLMs) have long struggled with spatial reasoning tasks. Surprisingly, even simple spatial reasoning tasks, such as recognizing "under" or "behind" relationships between only two objects, pose significant challenges for current VLMs. In this work, we study the spatial reasoning challenge from the lens of mechanistic interpretability, diving into the model's internal states to examine the interactions between image and text tokens. By tracing attention distribution over the image through out intermediate layers, we observe that successful spatial reasoning correlates strongly with the model's ability to align its attention distribution with actual object locations, particularly differing between familiar and unfamiliar spatial relationships. Motivated by these findings, we propose ADAPTVIS based on inference-time confidence scores to sharpen the attention on highly relevant regions when confident, while smoothing and broadening the attention window to consider a wider context when confidence is lower. This training-free decoding method shows significant improvement (e.g., up to a 50 absolute point improvement) on spatial reasoning benchmarks such as WhatsUp and VSR with negligible cost. We make code and data publicly available for research purposes at https://github.com/shiqichen17/AdaptVis.

Paperid: 1350, https://arxiv.org/pdf/2503.01616.pdf GitHub

Abstract:
This paper introduces RoboDexVLM, an innovative framework for robot task planning and grasp detection tailored for a collaborative manipulator equipped with a dexterous hand. Previous methods focus on simplified and limited manipulation tasks, which often neglect the complexities associated with grasping a diverse array of objects in a long-horizon manner. In contrast, our proposed framework utilizes a dexterous hand capable of grasping objects of varying shapes and sizes while executing tasks based on natural language commands. The proposed approach has the following core components: First, a robust task planner with a task-level recovery mechanism that leverages vision-language models (VLMs) is designed, which enables the system to interpret and execute open-vocabulary commands for long sequence tasks. Second, a language-guided dexterous grasp perception algorithm is presented based on robot kinematics and formal methods, tailored for zero-shot dexterous manipulation with diverse objects and commands. Comprehensive experimental results validate the effectiveness, adaptability, and robustness of RoboDexVLM in handling long-horizon scenarios and performing dexterous grasping. These results highlight the framework's ability to operate in complex environments, showcasing its potential for open-vocabulary dexterous manipulation. Our open-source project page can be found at https://henryhcliu.github.io/robodexvlm.

Paperid: 1351, https://arxiv.org/pdf/2503.01342.pdf GitHub

Abstract:
Generalist models have achieved remarkable success in both language and vision-language tasks, showcasing the potential of unified modeling. However, effectively integrating fine-grained perception tasks like detection and segmentation into these models remains a significant challenge. This is primarily because these tasks often rely heavily on task-specific designs and architectures that can complicate the modeling process. To address this challenge, we present \ours, a framework that \textbf{U}nifies \textbf{F}ine-grained visual perception tasks through an \textbf{O}pen-ended language interface. By transforming all perception targets into the language space, \ours unifies object-level detection, pixel-level segmentation, and image-level vision-language tasks into a single model. Additionally, we introduce a novel embedding retrieval approach that relies solely on the language interface to support segmentation tasks. Our framework bridges the gap between fine-grained perception and vision-language tasks, significantly simplifying architectural design and training strategies while achieving comparable or superior performance to methods with intricate task-specific designs. After multi-task training on five standard visual perception datasets, \ours outperforms the previous state-of-the-art generalist models by 12.3 mAP on COCO instance segmentation and 3.3 mIoU on ADE20K semantic segmentation. Furthermore, our method seamlessly integrates with existing MLLMs, effectively combining fine-grained perception capabilities with their advanced language abilities, thereby enabling more challenging tasks such as reasoning segmentation. Code and models are available at https://github.com/nnnth/UFO.

Paperid: 1352, https://arxiv.org/pdf/2503.01208.pdf GitHub

Abstract:
Multi-Modal Large Language Models (MLLMs) have exhibited remarkable performance on various vision-language tasks such as Visual Question Answering (VQA). Despite accumulating evidence of privacy concerns associated with task-relevant content, it remains unclear whether MLLMs inadvertently memorize private content that is entirely irrelevant to the training tasks. In this paper, we investigate how randomly generated task-irrelevant private content can become spuriously correlated with downstream objectives due to partial mini-batch training dynamics, thus causing inadvertent memorization. Concretely, we randomly generate task-irrelevant watermarks into VQA fine-tuning images at varying probabilities and propose a novel probing framework to determine whether MLLMs have inadvertently encoded such content. Our experiments reveal that MLLMs exhibit notably different training behaviors in partial mini-batch settings with task-irrelevant watermarks embedded. Furthermore, through layer-wise probing, we demonstrate that MLLMs trigger distinct representational patterns when encountering previously seen task-irrelevant knowledge, even if this knowledge does not influence their output during prompting. Our code is available at https://github.com/illusionhi/ProbingPrivacy.

Paperid: 1353, https://arxiv.org/pdf/2503.01062.pdf GitHub

Abstract:
While internet-scale image and textual data have enabled strong generalization in Vision-Language Models (VLMs), the absence of internet-scale control data has impeded the development of similar generalization in standard reinforcement learning (RL) agents. Although VLMs are fundamentally limited in their ability to solve control tasks due to their lack of action-conditioned training data, their capacity for image understanding allows them to provide valuable feedback in RL tasks by recognizing successful outcomes. A key challenge in Reinforcement Learning from AI Feedback (RLAIF) is determining how best to integrate VLM-derived signals into the learning process. We explore this question in the context of offline RL and introduce a class of methods called Sub-Trajectory Filtered Optimization (SFO). We identify three key insights. First, trajectory length plays a crucial role in offline RL, as full-trajectory preference learning exacerbates the stitching problem, necessitating the use of sub-trajectories. Second, even in Markovian environments, a non-Markovian reward signal from a sequence of images is required to assess trajectory improvement, as VLMs do not interpret control actions and must rely on visual cues over time. Third, a simple yet effective approach--filtered and weighted behavior cloning--consistently outperforms more complex RLHF-based methods. We propose Sub-Trajectory Filtered Behavior Cloning (SFBC), a method that leverages VLM feedback on sub-trajectories while incorporating a retrospective filtering mechanism that removes sub-trajectories preceding failures to improve robustness and prevent turbulence. Please enjoy our airport puns.

Paperid: 1354, https://arxiv.org/pdf/2503.01020.pdf GitHub

Abstract:
Recent advances in medical vision-language models (VLMs) demonstrate impressive performance in image classification tasks, driven by their strong zero-shot generalization capabilities. However, given the high variability and complexity inherent in medical imaging data, the ability of these models to detect out-of-distribution (OOD) data in this domain remains underexplored. In this work, we conduct the first systematic investigation into the OOD detection potential of medical VLMs. We evaluate state-of-the-art VLM-based OOD detection methods across a diverse set of medical VLMs, including both general and domain-specific purposes. To accurately reflect real-world challenges, we introduce a cross-modality evaluation pipeline for benchmarking full-spectrum OOD detection, rigorously assessing model robustness against both semantic shifts and covariate shifts. Furthermore, we propose a novel hierarchical prompt-based method that significantly enhances OOD detection performance. Extensive experiments are conducted to validate the effectiveness of our approach. The codes are available at https://github.com/PyJulie/Medical-VLMs-OOD-Detection.

Paperid: 1355, https://arxiv.org/pdf/2503.00778.pdf GitHub

Abstract:
Inferring the affordance of an object and grasping it in a task-oriented manner is crucial for robots to successfully complete manipulation tasks. Affordance indicates where and how to grasp an object by taking its functionality into account, serving as the foundation for effective task-oriented grasping. However, current task-oriented methods often depend on extensive training data that is confined to specific tasks and objects, making it difficult to generalize to novel objects and complex scenes. In this paper, we introduce AffordGrasp, a novel open-vocabulary grasping framework that leverages the reasoning capabilities of vision-language models (VLMs) for in-context affordance reasoning. Unlike existing methods that rely on explicit task and object specifications, our approach infers tasks directly from implicit user instructions, enabling more intuitive and seamless human-robot interaction in everyday scenarios. Building on the reasoning outcomes, our framework identifies task-relevant objects and grounds their part-level affordances using a visual grounding module. This allows us to generate task-oriented grasp poses precisely within the affordance regions of the object, ensuring both functional and context-aware robotic manipulation. Extensive experiments demonstrate that AffordGrasp achieves state-of-the-art performance in both simulation and real-world scenarios, highlighting the effectiveness of our method. We believe our approach advances robotic manipulation techniques and contributes to the broader field of embodied AI. Project website: https://eqcy.github.io/affordgrasp/.

Paperid: 1356, https://arxiv.org/pdf/2503.00726.pdf GitHub

Abstract:
3D scene reconstruction is essential for applications in virtual reality, robotics, and autonomous driving, enabling machines to understand and interact with complex environments. Traditional 3D Gaussian Splatting techniques rely on images captured from multiple viewpoints to achieve optimal performance, but this dependence limits their use in scenarios where only a single image is available. In this work, we introduce FlashDreamer, a novel approach for reconstructing a complete 3D scene from a single image, significantly reducing the need for multi-view inputs. Our approach leverages a pre-trained vision-language model to generate descriptive prompts for the scene, guiding a diffusion model to produce images from various perspectives, which are then fused to form a cohesive 3D reconstruction. Extensive experiments show that our method effectively and robustly expands single-image inputs into a comprehensive 3D scene, extending monocular 3D reconstruction capabilities without further training. Our code is available https://github.com/CharlieSong1999/FlashDreamer/tree/main.

Paperid: 1357, https://arxiv.org/pdf/2503.00641.pdf GitHub

Abstract:
Post-hoc importance attribution methods are a popular tool for "explaining" Deep Neural Networks (DNNs) and are inherently based on the assumption that the explanations can be applied independently of how the models were trained. Contrarily, in this work we bring forward empirical evidence that challenges this very notion. Surprisingly, we discover a strong dependency on and demonstrate that the training details of a pre-trained model's classification layer (less than 10 percent of model parameters) play a crucial role, much more than the pre-training scheme itself. This is of high practical relevance: (1) as techniques for pre-training models are becoming increasingly diverse, understanding the interplay between these techniques and attribution methods is critical; (2) it sheds light on an important yet overlooked assumption of post-hoc attribution methods which can drastically impact model explanations and how they are interpreted eventually. With this finding we also present simple yet effective adjustments to the classification layers, that can significantly enhance the quality of model explanations. We validate our findings across several visual pre-training frameworks (fully-supervised, self-supervised, contrastive vision-language training) and analyse how they impact explanations for a wide range of attribution methods on a diverse set of evaluation metrics.

Paperid: 1358, https://arxiv.org/pdf/2503.00361.pdf GitHub

Abstract:
Large Vision-Language Models (LVLMs) have obtained impressive performance in visual content understanding and multi-modal reasoning. Unfortunately, these large models suffer from serious hallucination problems and tend to generate fabricated responses. Recently, several Contrastive Decoding (CD) strategies have been proposed to alleviate hallucination by introducing disturbed inputs. Although great progress has been made, these CD strategies mostly apply a one-size-fits-all approach for all input conditions. In this paper, we revisit this process through extensive experiments. Related results show that hallucination causes are hybrid and each generative step faces a unique hallucination challenge. Leveraging these meaningful insights, we introduce a simple yet effective Octopus-like framework that enables the model to adaptively identify hallucination types and create a dynamic CD workflow. Our Octopus framework not only outperforms existing methods across four benchmarks but also demonstrates excellent deployability and expansibility. Code is available at https://github.com/LijunZhang01/Octopus.

Paperid: 1359, https://arxiv.org/pdf/2503.00329.pdf GitHub

Abstract:
Visual embedding models excel at zero-shot tasks like visual retrieval and classification. However, these models cannot be used for tasks that contain ambiguity or require user instruction. These tasks necessitate an embedding model which outputs can use a natural language instruction to control the representation of a visual embedding. Existing CLIP-based approaches embed images and text independently, and fuse the result. We find that this results in weak interactions between modalities, and poor user control over the representation. We introduce ABC, an open-source multimodal embedding model that uses a vision-language model backbone to deeply integrate image features with natural language instructions. ABC achieves best-for-size performance on MSCOCO image-to-text retrieval and is the top performing model on classification and VQA tasks in the Massive Multimodal Embedding Benchmark. With a strongly unified vision-language representation, ABC can use natural language to solve subtle and potentially ambiguous visual retrieval problems. To evaluate this capability, we design CtrlBench, a benchmark that requires interleaving textual instructions with image content for correct retrieval. ABC advances the state of visual embeddings, outputting high-quality visual representations with natural language control. Our model and datasets are available at our project page: https://tiger-ai-lab.github.io/ABC/

Paperid: 1360, https://arxiv.org/pdf/2503.00196.pdf GitHub

Abstract:
Developing reliable and generalizable deep learning systems for medical imaging faces significant obstacles due to spurious correlations, data imbalances, and limited text annotations in datasets. Addressing these challenges requires architectures that are robust to the unique complexities posed by medical imaging data. Rapid advancements in vision-language foundation models within the natural image domain prompt the question of how they can be adapted for medical imaging tasks. In this work, we present PRISM, a framework that leverages foundation models to generate high-resolution, language-guided medical image counterfactuals using Stable Diffusion. Our approach demonstrates unprecedented precision in selectively modifying spurious correlations (the medical devices) and disease features, enabling the removal and addition of specific attributes while preserving other image characteristics. Through extensive evaluation, we show how PRISM advances counterfactual generation and enables the development of more robust downstream classifiers for clinically deployable solutions. To facilitate broader adoption and research, we make our code publicly available at https://github.com/Amarkr1/PRISM.

Paperid: 1361, https://arxiv.org/pdf/2502.21291.pdf GitHub

Abstract:
Despite significant progress in diffusion-based image generation, subject-driven generation and instruction-based editing remain challenging. Existing methods typically treat them separately, struggling with limited high-quality data and poor generalization. However, both tasks require capturing complex visual variations while maintaining consistency between inputs and outputs. Inspired by this, we propose MIGE, a unified framework that standardizes task representations using multimodal instructions. It first treats subject-driven generation as creation on a blank canvas and instruction-based editing as modification of an existing image, establishing a shared input-output formulation, then introduces a novel multimodal encoder that maps free-form multimodal instructions into a unified vision-language space, integrating visual and semantic features through a feature fusion mechanism. This unification enables joint training of both tasks, providing two key advantages: (1) Cross-Task Enhancement: by leveraging shared visual and semantic representations, joint training improves instruction adherence and visual consistency in both subject-driven generation and instruction-based editing. (2) Generalization: learning in a unified format facilitates cross-task knowledge transfer, enabling MIGE to generalize to novel compositional tasks, including instruction-based subject-driven editing. Experiments show that MIGE excels in both subject-driven generation and instruction-based editing while setting a SOTA in the new task of instruction-based subject-driven editing. Code and model have been publicly available at https://github.com/Eureka-Maggie/MIGE.

Paperid: 1362, https://arxiv.org/pdf/2502.20780.pdf GitHub GitHub

Abstract:
The increasing use of vision-language models (VLMs) in healthcare applications presents great challenges related to hallucinations, in which the models may generate seemingly plausible results that are in fact incorrect. Such hallucinations can jeopardize clinical decision making, potentially harming the diagnosis and treatments. In this work, we propose MedHallTune, a large-scale benchmark designed specifically to evaluate and mitigate hallucinations in medical VLMs. Comprising over 100,000 images and 1,000,000 instruction pairs, MedHallTune includes both hallucination and non-hallucination samples, each with ground-truth annotations. We conduct a comprehensive evaluation of current medical and general VLMs using MedHallTune, assessing their performance across key metrics, including clinical accuracy, relevance, detail level, and risk level. The experimental results show that fine-tuning with MedHallTune successfully improves the ability of several existing models to manage hallucinations and boost their zero-shot performance on downstream visual-question-answering (VQA) tasks, making them more reliable for practical medical applications. Our work contributes to the development of more trustworthy VLMs. Codes and dataset will be available at \href{https://github.com/russellyq/MedHallTune}{MedHallTune}.

Paperid: 1363, https://arxiv.org/pdf/2502.20625.pdf GitHub

Abstract:
Zero-shot object counting aims to count instances of arbitrary object categories specified by text descriptions. Existing methods typically rely on vision-language models like CLIP, but often exhibit limited sensitivity to text prompts. We present T2ICount, a diffusion-based framework that leverages rich prior knowledge and fine-grained visual understanding from pretrained diffusion models. While one-step denoising ensures efficiency, it leads to weakened text sensitivity. To address this challenge, we propose a Hierarchical Semantic Correction Module that progressively refines text-image feature alignment, and a Representational Regional Coherence Loss that provides reliable supervision signals by leveraging the cross-attention maps extracted from the denosing U-Net. Furthermore, we observe that current benchmarks mainly focus on majority objects in images, potentially masking models' text sensitivity. To address this, we contribute a challenging re-annotated subset of FSC147 for better evaluation of text-guided counting ability. Extensive experiments demonstrate that our method achieves superior performance across different benchmarks. Code is available at https://github.com/cha15yq/T2ICount.

Paperid: 1364, https://arxiv.org/pdf/2502.20587.pdf GitHub

Abstract:
Vision Language Models (VLMs) have achieved remarkable success in a wide range of vision applications of increasing complexity and scales, yet choosing the right VLM model size involves a trade-off between response quality and cost. While smaller VLMs are cheaper to run, they typically produce responses only marginally better than random guessing on benchmarks such as MMMU. In this paper, we propose Cache of Thought (CoT), a master apprentice framework for collaborative inference between large and small VLMs. CoT manages high quality query results from large VLMs (master) in a cache, which are then selected via a novel multi modal retrieval and in-context learning to aid the performance of small VLMs (apprentice). We extensively evaluate CoT on various widely recognized and challenging general reasoning benchmarks, and show that CoT increases overall reasoning performance by up to 7.7% under the same budget, and specifically boosts the performance of apprentice VLMs by up to 36.6%. Our code is available at https://github.com/UIUC-MONET/Cache-of-Thoughts

Paperid: 1365, https://arxiv.org/pdf/2502.20578.pdf GitHub

Abstract:
Sparse autoencoders (SAEs) are useful for detecting and steering interpretable features in neural networks, with particular potential for understanding complex multimodal representations. Given their ability to uncover interpretable features, SAEs are particularly valuable for analyzing large-scale vision-language models (e.g., CLIP and SigLIP), which are fundamental building blocks in modern systems yet remain challenging to interpret and control. However, current SAE methods are limited by optimizing both reconstruction quality and sparsity simultaneously, as they rely on either activation suppression or rigid sparsity constraints. To this end, we introduce Matryoshka SAE (MSAE), a new architecture that learns hierarchical representations at multiple granularities simultaneously, enabling a direct optimization of both metrics without compromise. MSAE establishes a new state-of-the-art Pareto frontier between reconstruction quality and sparsity for CLIP, achieving 0.99 cosine similarity and less than 0.1 fraction of variance unexplained while maintaining ~80% sparsity. Finally, we demonstrate the utility of MSAE as a tool for interpreting and controlling CLIP by extracting over 120 semantic concepts from its representation to perform concept-based similarity search and bias analysis in downstream tasks like CelebA. We make the codebase available at https://github.com/WolodjaZ/MSAE.

Paperid: 1366, https://arxiv.org/pdf/2502.20104.pdf GitHub

Abstract:
Referring Expression Comprehension (REC) is a foundational cross-modal task that evaluates the interplay of language understanding, image comprehension, and language-to-image grounding. It serves as an essential testing ground for Multimodal Large Language Models (MLLMs). To advance this field, we introduced a new REC dataset in our previous conference paper, characterized by two key features. First, it is designed with controllable difficulty levels, requiring multi-level fine-grained reasoning across object categories, attributes, and multi-hop relationships. Second, it incorporates negative text and images generated through fine-grained editing and augmentation, explicitly testing a model's ability to reject scenarios where the target object is absent, an often overlooked yet critical challenge in existing datasets. In this extended work, we propose two new methods to tackle the challenges of fine-grained REC by combining the strengths of Specialist Models and MLLMs. The first method adaptively assigns simple cases to faster, lightweight models and reserves complex ones for powerful MLLMs, balancing accuracy and efficiency. The second method lets a specialist generate a set of possible object regions, and the MLLM selects the most plausible one using its reasoning ability. These collaborative strategies lead to significant improvements on our dataset and other challenging benchmarks. Our results show that combining specialized and general-purpose models offers a practical path toward solving complex real-world vision-language tasks. Our dataset and code are available at https://github.com/sleepyshep/FineCops-Ref.

Paperid: 1367, https://arxiv.org/pdf/2502.19645.pdf GitHub

Abstract:
Recent vision-language-action models (VLAs) build upon pretrained vision-language models and leverage diverse robot datasets to demonstrate strong task execution, language following ability, and semantic generalization. Despite these successes, VLAs struggle with novel robot setups and require fine-tuning to achieve good performance, yet how to most effectively fine-tune them is unclear given many possible strategies. In this work, we study key VLA adaptation design choices such as different action decoding schemes, action representations, and learning objectives for fine-tuning, using OpenVLA as our representative base model. Our empirical analysis informs an Optimized Fine-Tuning (OFT) recipe that integrates parallel decoding, action chunking, a continuous action representation, and a simple L1 regression-based learning objective to altogether improve inference efficiency, policy performance, and flexibility in the model's input-output specifications. We propose OpenVLA-OFT, an instantiation of this recipe, which sets a new state of the art on the LIBERO simulation benchmark, significantly boosting OpenVLA's average success rate across four task suites from 76.5% to 97.1% while increasing action generation throughput by 26$\times$. In real-world evaluations, our fine-tuning recipe enables OpenVLA to successfully execute dexterous, high-frequency control tasks on a bimanual ALOHA robot and outperform other VLAs ($Ï_0$ and RDT-1B) fine-tuned using their default recipes, as well as strong imitation learning policies trained from scratch (Diffusion Policy and ACT) by up to 15% (absolute) in average success rate. We release code for OFT and pretrained model checkpoints at https://openvla-oft.github.io/.

Paperid: 1368, https://arxiv.org/pdf/2502.18816.pdf GitHub

Abstract:
Significant progress has been achieved on the improvement and downstream usages of the Contrastive Language-Image Pre-training (CLIP) vision-language model, while less attention is paid to the interpretation of CLIP. We propose a Gradient-based visual and textual Explanation method for CLIP (Grad-ECLIP), which interprets the matching result of CLIP for specific input image-text pair. By decomposing the architecture of the encoder and discovering the relationship between the matching similarity and intermediate spatial features, Grad-ECLIP produces effective heat maps that show the influence of image regions or words on the CLIP results. Different from the previous Transformer interpretation methods that focus on the utilization of self-attention maps, which are typically extremely sparse in CLIP, we produce high-quality visual explanations by applying channel and spatial weights on token features. Qualitative and quantitative evaluations verify the effectiveness and superiority of Grad-ECLIP compared with the state-of-the-art methods. Furthermore, a series of analysis are conducted based on our visual and textual explanation results, from which we explore the working mechanism of image-text matching, the strengths and limitations in attribution identification of CLIP, and the relationship between the concreteness/abstractness of a word and its usage in CLIP. Finally, based on the ability of explanation map that indicates text-specific saliency region of input image, we also propose an application with Grad-ECLIP, which is adopted to boost the fine-grained alignment in the CLIP fine-tuning. The code of Grad-ECLIP is available here: https://github.com/Cyang-Zhao/Grad-Eclip.

Paperid: 1369, https://arxiv.org/pdf/2502.18176.pdf GitHub

Abstract:
In this paper, we aim to build an adversarially robust zero-shot image classifier. We ground our work on CLIP, a vision-language pre-trained encoder model that can perform zero-shot classification by matching an image with text prompts ``a photo of a .''. Purification is the path we choose since it does not require adversarial training on specific attack types and thus can cope with any foreseen attacks. We then formulate purification risk as the KL divergence between the joint distributions of the purification process of denoising the adversarial samples and the attack process of adding perturbations to benign samples, through bidirectional Stochastic Differential Equations (SDEs). The final derived results inspire us to explore purification in the multi-modal latent space of CLIP. We propose two variants for our CLIPure approach: CLIPure-Diff which models the likelihood of images' latent vectors with the DiffusionPrior module in DaLLE-2 (modeling the generation process of CLIP's latent vectors), and CLIPure-Cos which models the likelihood with the cosine similarity between the embeddings of an image and ``a photo of a.''. As far as we know, CLIPure is the first purification method in multi-modal latent space and CLIPure-Cos is the first purification method that is not based on generative models, which substantially improves defense efficiency. We conducted extensive experiments on CIFAR-10, ImageNet, and 13 datasets that previous CLIP-based defense methods used for evaluating zero-shot classification robustness. Results show that CLIPure boosts the SOTA robustness by a large margin, e.g., from 71.7% to 91.1% on CIFAR10, from 59.6% to 72.6% on ImageNet, and 108% relative improvements of average robustness on the 13 datasets over previous SOTA. The code is available at https://github.com/TMLResearchGroup-CAS/CLIPure.

Paperid: 1370, https://arxiv.org/pdf/2502.17812.pdf GitHub

Abstract:
Large language models (LLMs) have been increasingly used in time series analysis. However, the potential of multimodal LLMs (MLLMs), particularly vision-language models, for time series remains largely under-explored. One natural way for humans to detect time series anomalies is through visualization and textual description. Motivated by this, we raise a critical and practical research question: Can multimodal LLMs perform time series anomaly detection? To answer this, we propose VisualTimeAnomaly benchmark to evaluate MLLMs in time series anomaly detection (TSAD). Our approach transforms time series numerical data into the image format and feed these images into various MLLMs, including proprietary models (GPT-4o and Gemini-1.5) and open-source models (LLaVA-NeXT and Qwen2-VL), each with one larger and one smaller variant. In total, VisualTimeAnomaly contains 12.4k time series images spanning 3 scenarios and 3 anomaly granularities with 9 anomaly types across 8 MLLMs. Starting with the univariate case (point- and range-wise anomalies), we extend our evaluation to more practical scenarios, including multivariate and irregular time series scenarios, and variate-wise anomalies. Our study reveals several key insights: 1) MLLMs detect range- and variate-wise anomalies more effectively than point-wise anomalies. 2) MLLMs are highly robust to irregular time series, even with 25% of the data missing. 3) Open-source MLLMs perform comparably to proprietary models in TSAD. While open-source MLLMs excel on univariate time series, proprietary MLLMs demonstrate superior effectiveness on multivariate time series. To the best of our knowledge, this is the first work to comprehensively investigate MLLMs for TSAD, particularly for multivariate and irregular time series scenarios. We release our dataset and code at https://github.com/mllm-ts/VisualTimeAnomaly to support future research.

Paperid: 1371, https://arxiv.org/pdf/2502.17358.pdf GitHub

Abstract:
How can we verify whether copyrighted content was used to train a large vision-language model (VLM) without direct access to its training data? Motivated by the hypothesis that a VLM is able to recognize images from its training corpus, we propose DIS-CO, a novel approach to infer the inclusion of copyrighted content during the model's development. By repeatedly querying a VLM with specific frames from targeted copyrighted material, DIS-CO extracts the content's identity through free-form text completions. To assess its effectiveness, we introduce MovieTection, a benchmark comprising 14,000 frames paired with detailed captions, drawn from films released both before and after a model's training cutoff. Our results show that DIS-CO significantly improves detection performance, nearly doubling the average AUC of the best prior method on models with logits available. Our findings also highlight a broader concern: all tested models appear to have been exposed to some extent to copyrighted content. Our code and data are available at https://github.com/avduarte333/DIS-CO

Paperid: 1372, https://arxiv.org/pdf/2502.16680.pdf GitHub

Abstract:
As a novel and challenging task, referring segmentation combines computer vision and natural language processing to localize and segment objects based on textual descriptions. While referring image segmentation (RIS) has been extensively studied in natural images, little attention has been given to aerial imagery, particularly from unmanned aerial vehicles (UAVs). The unique challenges of UAV imagery, including complex spatial scales, occlusions, and varying object orientations, render existing RIS approaches ineffective. A key limitation has been the lack of UAV-specific datasets, as manually annotating pixel-level masks and generating textual descriptions is labour-intensive and time-consuming. To address this gap, we design an automatic labelling pipeline that leverages pre-existing UAV segmentation datasets and Multimodal Large Language Models (MLLM) for generating textual descriptions. Furthermore, we propose Aerial Referring Transformer (AeroReformer), a novel framework for UAV referring image segmentation (UAV-RIS), featuring a Vision-Language Cross-Attention Module (VLCAM) for effective cross-modal understanding and a Rotation-Aware Multi-Scale Fusion (RAMSF) decoder to enhance segmentation accuracy in aerial scenes. Extensive experiments on two newly developed datasets demonstrate the superiority of AeroReformer over existing methods, establishing a new benchmark for UAV-RIS. The datasets and code will be publicly available at: https://github.com/lironui/AeroReformer.

Paperid: 1373, https://arxiv.org/pdf/2502.16412.pdf GitHub

Abstract:
This paper comprehensively reviews anomaly synthesis methodologies. Existing surveys focus on limited techniques, missing an overall field view and understanding method interconnections. In contrast, our study offers a unified review, covering about 40 representative methods across Hand-crafted, Distribution-hypothesis-based, Generative models (GM)-based, and Vision-language models (VLM)-based synthesis. We introduce the first industrial anomaly synthesis (IAS) taxonomy. Prior works lack formal classification or use simplistic taxonomies, hampering structured comparisons and trend identification. Our taxonomy provides a fine-grained framework reflecting methodological progress and practical implications, grounding future research. Furthermore, we explore cross-modality synthesis and large-scale VLM. Previous surveys overlooked multimodal data and VLM in anomaly synthesis, limiting insights into their advantages. Our survey analyzes their integration, benefits, challenges, and prospects, offering a roadmap to boost IAS with multimodal learning. More resources are available at https://github.com/M-3LAB/awesome-anomaly-synthesis.

Paperid: 1374, https://arxiv.org/pdf/2502.16025.pdf GitHub

Abstract:
The feature maps of vision encoders are fundamental to myriad modern AI tasks, ranging from core perception algorithms (e.g. semantic segmentation, object detection, depth perception, etc.) to modern multimodal understanding in vision-language models (VLMs). Currently, in computer vision, the frontier of general purpose vision backbones is Vision Transformers (ViT), typically trained using contrastive loss (e.g. CLIP). A key problem with most off-the-shelf ViTs, particularly CLIP, is that these models are inflexibly low resolution. Most run at $224 \times 224$px, while the "high-resolution" versions are around $378-448$px, but still inflexible. We introduce a novel method to coherently and cheaply upsample the feature maps of low-resolution vision encoders while picking up on fine-grained details that would otherwise be lost due to resolution. We demonstrate the effectiveness of this approach on core perception tasks as well as within agglomerative model training using RADIO as a way of providing richer targets for distillation. Code available at https://github.com/NVlabs/FeatSharp .

Paperid: 1375, https://arxiv.org/pdf/2502.15969.pdf GitHub

Abstract:
Despite strong performance on vision-language tasks, Multimodal Large Language Models (MLLMs) struggle with mathematical problem-solving, with both open-source and state-of-the-art models falling short of human performance on visual-math benchmarks. To systematically examine visual-mathematical reasoning in MLLMs, we (1) evaluate their understanding of geometric primitives, (2) test multi-step reasoning, and (3) explore a potential solution to improve visual reasoning capabilities. Our findings reveal fundamental shortcomings in shape recognition, with top models achieving under 50% accuracy in identifying regular polygons. We analyze these failures through the lens of dual-process theory and show that MLLMs rely on System 1 (intuitive, memorized associations) rather than System 2 (deliberate reasoning). Consequently, MLLMs fail to count the sides of both familiar and novel shapes, suggesting they have neither learned the concept of sides nor effectively process visual inputs. Finally, we propose Visually Cued Chain-of-Thought (VC-CoT) prompting, which enhances multi-step mathematical reasoning by explicitly referencing visual annotations in diagrams, boosting GPT-4o's accuracy on an irregular polygon side-counting task from 7% to 93%. Our findings suggest that System 2 reasoning in MLLMs remains an open problem, and visually-guided prompting is essential for successfully engaging visual reasoning. Code available at: https://github.com/rsinghlab/Shape-Blind.

Paperid: 1376, https://arxiv.org/pdf/2502.15309.pdf GitHub GitHub

Abstract:
In real-world scenarios, environment changes caused by human or agent activities make it extremely challenging for robots to perform various long-term tasks. Recent works typically struggle to effectively understand and adapt to dynamic environments due to the inability to update their environment representations in memory according to environment changes and lack of fine-grained reconstruction of the environments. To address these challenges, we propose DynamicGSG, a dynamic, high-fidelity, open-vocabulary scene graph construction system leveraging Gaussian splatting. DynamicGSG builds hierarchical scene graphs using advanced vision language models to represent the spatial and semantic relationships between objects in the environments, utilizes a joint feature loss we designed to supervise Gaussian instance grouping while optimizing the Gaussian maps, and locally updates the Gaussian scene graphs according to real environment changes for long-term environment adaptation. Experiments and ablation studies demonstrate the performance and efficacy of our proposed method in terms of semantic segmentation, language-guided object retrieval, and reconstruction quality. Furthermore, we validate the dynamic updating capabilities of our system in real laboratory environments. The source code and supplementary experimental materials will be released at:~\href{https://github.com/GeLuzhou/Dynamic-GSG}{https://github.com/GeLuzhou/Dynamic-GSG}.

Paperid: 1377, https://arxiv.org/pdf/2502.15119.pdf GitHub

Abstract:
Ensuring safety in autonomous driving systems remains a critical challenge, particularly in handling rare but potentially catastrophic safety-critical scenarios. While existing research has explored generating safety-critical scenarios for autonomous vehicle (AV) testing, there is limited work on effectively incorporating these scenarios into policy learning to enhance safety. Furthermore, developing training curricula that adapt to an AV's evolving behavioral patterns and performance bottlenecks remains largely unexplored. To address these challenges, we propose CurricuVLM, a novel framework that leverages Vision-Language Models (VLMs) to enable personalized curriculum learning for autonomous driving agents. Our approach uniquely exploits VLMs' multimodal understanding capabilities to analyze agent behavior, identify performance weaknesses, and dynamically generate tailored training scenarios for curriculum adaptation. Through comprehensive analysis of unsafe driving situations with narrative descriptions, CurricuVLM performs in-depth reasoning to evaluate the AV's capabilities and identify critical behavioral patterns. The framework then synthesizes customized training scenarios targeting these identified limitations, enabling effective and personalized curriculum learning. Extensive experiments on the Waymo Open Motion Dataset show that CurricuVLM outperforms state-of-the-art baselines across both regular and safety-critical scenarios, achieving superior performance in terms of navigation success, driving efficiency, and safety metrics. Further analysis reveals that CurricuVLM serves as a general approach that can be integrated with various RL algorithms to enhance autonomous driving systems. The code and demo video are available at: https://zihaosheng.github.io/CurricuVLM/.

Paperid: 1378, https://arxiv.org/pdf/2502.14881.pdf GitHub

Abstract:
With the rapid advancement of Large Vision-Language Models (LVLMs), ensuring their safety has emerged as a crucial area of research. This survey provides a comprehensive analysis of LVLM safety, covering key aspects such as attacks, defenses, and evaluation methods. We introduce a unified framework that integrates these interrelated components, offering a holistic perspective on the vulnerabilities of LVLMs and the corresponding mitigation strategies. Through an analysis of the LVLM lifecycle, we introduce a classification framework that distinguishes between inference and training phases, with further subcategories to provide deeper insights. Furthermore, we highlight limitations in existing research and outline future directions aimed at strengthening the robustness of LVLMs. As part of our research, we conduct a set of safety evaluations on the latest LVLM, Deepseek Janus-Pro, and provide a theoretical analysis of the results. Our findings provide strategic recommendations for advancing LVLM safety and ensuring their secure and reliable deployment in high-stakes, real-world applications. This survey aims to serve as a cornerstone for future research, facilitating the development of models that not only push the boundaries of multimodal intelligence but also adhere to the highest standards of security and ethical integrity. Furthermore, to aid the growing research in this field, we have created a public repository to continuously compile and update the latest work on LVLM safety: https://github.com/XuankunRong/Awesome-LVLM-Safety .

Paperid: 1379, https://arxiv.org/pdf/2502.14846.pdf GitHub

Abstract:
Reasoning about images with rich text, such as charts and documents, is a critical application of vision-language models (VLMs). However, VLMs often struggle in these domains due to the scarcity of diverse text-rich vision-language data. To address this challenge, we present CoSyn, a framework that leverages the coding capabilities of text-only large language models (LLMs) to automatically create synthetic text-rich multimodal data. Given input text describing a target domain (e.g., "nutrition fact labels"), CoSyn prompts an LLM to generate code (Python, HTML, LaTeX, etc.) for rendering synthetic images. With the underlying code as textual representations of the synthetic images, CoSyn can generate high-quality instruction-tuning data, again relying on a text-only LLM. Using CoSyn, we constructed a dataset comprising 400K images and 2.7M rows of vision-language instruction-tuning data. Comprehensive experiments on seven benchmarks demonstrate that models trained on our synthetic data achieve state-of-the-art performance among competitive open-source models, including Llama 3.2, and surpass proprietary models such as GPT-4V and Gemini 1.5 Flash. Furthermore, CoSyn can produce synthetic pointing data, enabling VLMs to ground information within input images, showcasing its potential for developing multimodal agents capable of acting in real-world environments.

Paperid: 1380, https://arxiv.org/pdf/2502.14834.pdf GitHub

Abstract:
Existing Large Vision-Language Models (LVLMs) can process inputs with context lengths up to 128k visual and text tokens, yet they struggle to generate coherent outputs beyond 1,000 words. We find that the primary limitation is the absence of long output examples during supervised fine-tuning (SFT). To tackle this issue, we introduce LongWriter-V-22k, a SFT dataset comprising 22,158 examples, each with multiple input images, an instruction, and corresponding outputs ranging from 0 to 10,000 words. Moreover, to achieve long outputs that maintain high-fidelity to the input images, we employ Direct Preference Optimization (DPO) to the SFT model. Given the high cost of collecting human feedback for lengthy outputs (e.g., 3,000 words), we propose IterDPO, which breaks long outputs into segments and uses iterative corrections to form preference pairs with the original outputs. Additionally, we develop MMLongBench-Write, a benchmark featuring six tasks to evaluate the long-generation capabilities of VLMs. Our 7B parameter model, trained with LongWriter-V-22k and IterDPO, achieves impressive performance on this benchmark, outperforming larger proprietary models like GPT-4o. Code and data: https://github.com/THU-KEG/LongWriter-V

Paperid: 1381, https://arxiv.org/pdf/2502.14786.pdf GitHub

Abstract:
We introduce SigLIP 2, a family of new multilingual vision-language encoders that build on the success of the original SigLIP. In this second iteration, we extend the original image-text training objective with several prior, independently developed techniques into a unified recipe -- this includes captioning-based pretraining, self-supervised losses (self-distillation, masked prediction) and online data curation. With these changes, SigLIP 2 models outperform their SigLIP counterparts at all model scales in core capabilities, including zero-shot classification, image-text retrieval, and transfer performance when extracting visual representations for Vision-Language Models (VLMs). Furthermore, the new training recipe leads to significant improvements on localization and dense prediction tasks. We also train variants which support multiple resolutions and preserve the input's native aspect ratio. Finally, we train on a more diverse data-mixture that includes de-biasing techniques, leading to much better multilingual understanding and improved fairness. To allow users to trade off inference cost with performance, we release model checkpoints at four sizes: ViT-B (86M), L (303M), So400m (400M), and g (1B).

Paperid: 1382, https://arxiv.org/pdf/2502.14778.pdf GitHub

Abstract:
Large Multimodal Models (LMMs) have demonstrated strong performance in English, but their effectiveness in Japanese remains limited due to the lack of high-quality training data. Current Japanese LMMs often rely on translated English datasets, restricting their ability to capture Japan-specific cultural knowledge. To address this, we explore the potential of Japanese PDF data as a training resource, an area that remains largely underutilized. We introduce a fully automated pipeline that leverages pretrained models to extract image-text pairs from PDFs through layout analysis, OCR, and vision-language pairing, removing the need for manual annotation. Additionally, we construct instruction data from extracted image-text pairs to enrich the training data. To evaluate the effectiveness of PDF-derived data, we train Japanese LMMs and assess their performance on the Japanese LMM Benchmark. Our results demonstrate substantial improvements, with performance gains ranging from 2.1% to 13.8% on Heron-Bench. Further analysis highlights the impact of PDF-derived data on various factors, such as model size and language models, reinforcing its value as a multimodal resource for Japanese LMMs.

Paperid: 1383, https://arxiv.org/pdf/2502.14744.pdf GitHub

Abstract:
The integration of additional modalities increases the susceptibility of large vision-language models (LVLMs) to safety risks, such as jailbreak attacks, compared to their language-only counterparts. While existing research primarily focuses on post-hoc alignment techniques, the underlying safety mechanisms within LVLMs remain largely unexplored. In this work , we investigate whether LVLMs inherently encode safety-relevant signals within their internal activations during inference. Our findings reveal that LVLMs exhibit distinct activation patterns when processing unsafe prompts, which can be leveraged to detect and mitigate adversarial inputs without requiring extensive fine-tuning. Building on this insight, we introduce HiddenDetect, a novel tuning-free framework that harnesses internal model activations to enhance safety. Experimental results show that {HiddenDetect} surpasses state-of-the-art methods in detecting jailbreak attacks against LVLMs. By utilizing intrinsic safety-aware patterns, our method provides an efficient and scalable solution for strengthening LVLM robustness against multimodal threats. Our code will be released publicly at https://github.com/leigest519/HiddenDetect.

Paperid: 1384, https://arxiv.org/pdf/2502.14638.pdf GitHub

Abstract:
Image geo-localization is the task of predicting the specific location of an image and requires complex reasoning across visual, geographical, and cultural contexts. While prior Vision Language Models (VLMs) have the best accuracy at this task, there is a dearth of high-quality datasets and models for analytical reasoning. We first create NaviClues, a high-quality dataset derived from GeoGuessr, a popular geography game, to supply examples of expert reasoning from language. Using this dataset, we present Navig, a comprehensive image geo-localization framework integrating global and fine-grained image information. By reasoning with language, Navig reduces the average distance error by 14% compared to previous state-of-the-art models while requiring fewer than 1000 training samples. Our dataset and code are available at https://github.com/SparrowZheyuan18/Navig/.

Paperid: 1385, https://arxiv.org/pdf/2502.14604.pdf GitHub

Abstract:
Test-time adaptation (TTA) aims to address distribution shifts between source and target data by relying solely on target data during testing. In open-world scenarios, models often encounter noisy samples, i.e., samples outside the in-distribution (ID) label space. Leveraging the zero-shot capability of pre-trained vision-language models (VLMs), this paper introduces Zero-Shot Noisy TTA (ZS-NTTA), focusing on adapting the model to target data with noisy samples during test-time in a zero-shot manner. We find existing TTA methods underperform under ZS-NTTA, often lagging behind even the frozen model. We conduct comprehensive experiments to analyze this phenomenon, revealing that the negative impact of unfiltered noisy data outweighs the benefits of clean data during model updating. Also, adapting a classifier for ID classification and noise detection hampers both sub-tasks. Built on this, we propose a framework that decouples the classifier and detector, focusing on developing an individual detector while keeping the classifier frozen. Technically, we introduce the Adaptive Noise Detector (AdaND), which utilizes the frozen model's outputs as pseudo-labels to train a noise detector. To handle clean data streams, we further inject Gaussian noise during adaptation, preventing the detector from misclassifying clean samples as noisy. Beyond the ZS-NTTA, AdaND can also improve the zero-shot out-of-distribution (ZS-OOD) detection ability of VLMs. Experiments show that AdaND outperforms in both ZS-NTTA and ZS-OOD detection. On ImageNet, AdaND achieves a notable improvement of $8.32\%$ in harmonic mean accuracy ($\text{Acc}_\text{H}$) for ZS-NTTA and $9.40\%$ in FPR95 for ZS-OOD detection, compared to SOTA methods. Importantly, AdaND is computationally efficient and comparable to the model-frozen method. The code is publicly available at: https://github.com/tmlr-group/ZS-NTTA.

Paperid: 1386, https://arxiv.org/pdf/2502.14191.pdf GitHub GitHub

Abstract:
Reward models play an essential role in training vision-language models (VLMs) by assessing output quality to enable aligning with human preferences. Despite their importance, the research community lacks comprehensive open benchmarks for evaluating multimodal reward models in VLMs. To address this gap, we introduce Multimodal RewardBench, an expert-annotated benchmark covering six domains: general correctness, preference, knowledge, reasoning, safety, and visual question-answering. Our dataset comprises 5,211 annotated (prompt, chosen response, rejected response) triplets collected from various VLMs. In evaluating a range of VLM judges, we find that even the top-performing models, Gemini 1.5 Pro and Claude 3.5 Sonnet, achieve only 72% overall accuracy. Notably, most models struggle in the reasoning and safety domains. These findings suggest that Multimodal RewardBench offers a challenging testbed for advancing reward model development across multiple domains. We release the benchmark at https://github.com/facebookresearch/multimodal_rewardbench.

Paperid: 1387, https://arxiv.org/pdf/2502.14149.pdf GitHub

Abstract:
Vision-Language Models (VLMs) in visual question answering (VQA) offer a unique opportunity to enhance intra-operative decision-making, promote intuitive interactions, and significantly advancing surgical education. However, the development of VLMs for surgical VQA is challenging due to limited datasets and the risk of overfitting and catastrophic forgetting during full fine-tuning of pretrained weights. While parameter-efficient techniques like Low-Rank Adaptation (LoRA) and Matrix of Rank Adaptation (MoRA) address adaptation challenges, their uniform parameter distribution overlooks the feature hierarchy in deep networks, where earlier layers, that learn general features, require more parameters than later ones. This work introduces PitVQA++ with an open-ended PitVQA dataset and vector matrix-low-rank adaptation (Vector-MoLoRA), an innovative VLM fine-tuning approach for adapting GPT-2 to pituitary surgery. Open-Ended PitVQA comprises around 101,803 frames from 25 procedural videos with 745,972 question-answer sentence pairs, covering key surgical elements such as phase and step recognition, context understanding, tool detection, localization, and interactions recognition. Vector-MoLoRA incorporates the principles of LoRA and MoRA to develop a matrix-low-rank adaptation strategy that employs vector ranking to allocate more parameters to earlier layers, gradually reducing them in the later layers. Our approach, validated on the Open-Ended PitVQA and EndoVis18-VQA datasets, effectively mitigates catastrophic forgetting while significantly enhancing performance over recent baselines. Furthermore, our risk-coverage analysis highlights its enhanced reliability and trustworthiness in handling uncertain predictions. Our source code and dataset is available at~\url{https://github.com/HRL-Mike/PitVQA-Plus}.

Paperid: 1388, https://arxiv.org/pdf/2502.14044.pdf GitHub

Abstract:
Large Multimodal Models (LMMs), or Vision-Language Models (VLMs), have shown impressive capabilities in a wide range of visual tasks. However, they often struggle with fine-grained visual reasoning, failing to identify domain-specific objectives and provide justifiable explanations for their predictions. To address the above challenge, we propose a novel visual rejection sampling framework to improve the cognition and explainability of LMMs using self-synthesized data. Specifically, visual fine-tuning requires images, queries, and target answers. Our approach begins by synthesizing interpretable answers that include human-verifiable visual features. These features are based on expert-defined concepts, and carefully selected based on their alignment with the image content. After each round of fine-tuning, we apply a reward model-free filtering mechanism to select the highest-quality interpretable answers for the next round of tuning. This iterative process of synthetic data generation and fine-tuning progressively improves the model's ability to generate accurate and reasonable explanations. Experimental results demonstrate the effectiveness of our method in improving both the accuracy and explainability of specialized visual classification tasks.

Paperid: 1389, https://arxiv.org/pdf/2502.13146.pdf GitHub

Abstract:
The emergence of large Vision Language Models (VLMs) has broadened the scope and capabilities of single-modal Large Language Models (LLMs) by integrating visual modalities, thereby unlocking transformative cross-modal applications in a variety of real-world scenarios. Despite their impressive performance, VLMs are prone to significant hallucinations, particularly in the form of cross-modal inconsistencies. Building on the success of Reinforcement Learning from Human Feedback (RLHF) in aligning LLMs, recent advancements have focused on applying direct preference optimization (DPO) on carefully curated datasets to mitigate these issues. Yet, such approaches typically introduce preference signals in a brute-force manner, neglecting the crucial role of visual information in the alignment process. In this paper, we introduce Re-Align, a novel alignment framework that leverages image retrieval to construct a dual-preference dataset, effectively incorporating both textual and visual preference signals. We further introduce rDPO, an extension of the standard direct preference optimization that incorporates an additional visual preference objective during fine-tuning. Our experimental results demonstrate that Re-Align not only mitigates hallucinations more effectively than previous methods but also yields significant performance gains in general visual question-answering (VQA) tasks. Moreover, we show that Re-Align maintains robustness and scalability across a wide range of VLM sizes and architectures. This work represents a significant step forward in aligning multimodal LLMs, paving the way for more reliable and effective cross-modal applications. We release all the code in https://github.com/taco-group/Re-Align.

Paperid: 1390, https://arxiv.org/pdf/2502.13130.pdf GitHub

Abstract:
We present Magma, a foundation model that serves multimodal AI agentic tasks in both the digital and physical worlds. Magma is a significant extension of vision-language (VL) models in that it not only retains the VL understanding ability (verbal intelligence) of the latter, but is also equipped with the ability to plan and act in the visual-spatial world (spatial-temporal intelligence) and complete agentic tasks ranging from UI navigation to robot manipulation. To endow the agentic capabilities, Magma is pretrained on large amounts of heterogeneous datasets spanning from images, videos to robotics data, where the actionable visual objects (e.g., clickable buttons in GUI) in images are labeled by Set-of-Mark (SoM) for action grounding, and the object movements (e.g., the trace of human hands or robotic arms) in videos are labeled by Trace-of-Mark (ToM) for action planning. Extensive experiments show that SoM and ToM reach great synergy and facilitate the acquisition of spatial-temporal intelligence for our Magma model, which is fundamental to a wide range of tasks as shown in Fig.1. In particular, Magma creates new state-of-the-art results on UI navigation and robotic manipulation tasks, outperforming previous models that are specifically tailored to these tasks. On image and video-related multimodal tasks, Magma also compares favorably to popular large multimodal models that are trained on much larger datasets. We make our model and code public for reproducibility at https://microsoft.github.io/Magma.

Paperid: 1391, https://arxiv.org/pdf/2502.13061.pdf GitHub

Abstract:
Hateful memes have become a significant concern on the Internet, necessitating robust automated detection systems. While Large Multimodal Models (LMMs) have shown promise in hateful meme detection, they face notable challenges like sub-optimal performance and limited out-of-domain generalization capabilities. Recent studies further reveal the limitations of both supervised fine-tuning (SFT) and in-context learning when applied to LMMs in this setting. To address these issues, we propose a robust adaptation framework for hateful meme detection that enhances in-domain accuracy and cross-domain generalization while preserving the general vision-language capabilities of LMMs. Analysis reveals that our approach achieves improved robustness under adversarial attacks compared to SFT models. Experiments on six meme classification datasets show that our approach achieves state-of-the-art performance, outperforming larger agentic systems. Moreover, our method generates higher-quality rationales for explaining hateful content compared to standard SFT, enhancing model interpretability. Code available at https://github.com/JingbiaoMei/RGCL

Paperid: 1392, https://arxiv.org/pdf/2502.12513.pdf GitHub GitHub

Abstract:
After pre-training on extensive image-text pairs, Contrastive Language-Image Pre-training (CLIP) demonstrates promising performance on a wide variety of benchmarks. However, a substantial volume of multimodal interleaved documents remains underutilized for contrastive vision-language representation learning. To fully leverage these unpaired documents, we initially establish a Real-World Data Extraction pipeline to extract high-quality images and texts. Then we design a hierarchical retrieval method to efficiently associate each image with multiple semantically relevant realistic texts. To further enhance fine-grained visual information, we propose an image semantic augmented generation module for synthetic text production. Furthermore, we employ a semantic balance sampling strategy to improve dataset diversity, enabling better learning of long-tail concepts. Based on these innovations, we construct RealSyn, a dataset combining realistic and synthetic texts, available in three scales: 15M, 30M, and 100M. We compare our dataset with other widely used datasets of equivalent scale for CLIP training. Models pre-trained on RealSyn consistently achieve state-of-the-art performance across various downstream tasks, including linear probe, zero-shot transfer, zero-shot robustness, and zero-shot retrieval. Furthermore, extensive experiments confirm that RealSyn significantly enhances contrastive vision-language representation learning and demonstrates robust scalability. To facilitate future research, the RealSyn dataset and pretrained model weights are released at https://github.com/deepglint/RealSyn.

Paperid: 1393, https://arxiv.org/pdf/2502.11651.pdf GitHub

Abstract:
Large vision-language models (LVLMs) have shown great promise in medical applications, particularly in visual question answering (MedVQA) and diagnosis from medical images. However, existing datasets and models often fail to consider critical aspects of medical diagnostics, such as the integration of historical records and the analysis of disease progression over time. In this paper, we introduce MMXU (Multimodal and MultiX-ray Understanding), a novel dataset for MedVQA that focuses on identifying changes in specific regions between two patient visits. Unlike previous datasets that primarily address single-image questions, MMXU enables multi-image questions, incorporating both current and historical patient data. We demonstrate the limitations of current LVLMs in identifying disease progression on MMXU-\textit{test}, even those that perform well on traditional benchmarks. To address this, we propose a MedRecord-Augmented Generation (MAG) approach, incorporating both global and regional historical records. Our experiments show that integrating historical records significantly enhances diagnostic accuracy by at least 20\%, bridging the gap between current LVLMs and human expert performance. Additionally, we fine-tune models with MAG on MMXU-\textit{dev}, which demonstrates notable improvements. We hope this work could illuminate the avenue of advancing the use of LVLMs in medical diagnostics by emphasizing the importance of historical context in interpreting medical images. Our dataset is released at github: https://github.com/linjiemu/MMXU.

Paperid: 1394, https://arxiv.org/pdf/2502.11492.pdf GitHub

Abstract:
Vision Language Models (VLMs) have achieved remarkable progress in multimodal tasks, yet they often struggle with visual arithmetic, seemingly simple capabilities like object counting or length comparison, which are essential for relevant complex tasks like chart understanding and geometric reasoning. In this work, we first investigate the root causes of this deficiency through a suite of probing tasks focusing on basic visual arithmetic. Our analysis reveals that while pre-trained vision encoders typically capture sufficient information, the text decoder often fails to decode it correctly for arithmetic reasoning. To address this, we propose CogAlign, a novel post-training strategy inspired by Piaget's theory of cognitive development. CogAlign trains VLMs to recognize invariant properties under visual transformations. We demonstrate that this approach significantly improves the performance of three diverse VLMs on our proposed probing tasks. Furthermore, CogAlign enhances performance by an average of 4.6% on CHOCOLATE and 2.9% on MATH-VISION, outperforming or matching supervised fine-tuning methods while requiring only 60% less training data. These results highlight the effectiveness and generalizability of CogAlign in improving fundamental visual arithmetic capabilities and their transfer to downstream tasks.

Paperid: 1395, https://arxiv.org/pdf/2502.11361.pdf GitHub GitHub

Abstract:
Detecting disinformation that blends manipulated text and images has become increasingly challenging, as AI tools make synthetic content easy to generate and disseminate. While most existing AI safety benchmarks focus on single modality misinformation (i.e., false content shared without intent to deceive), intentional multimodal disinformation, such as propaganda or conspiracy theories that imitate credible news, remains largely unaddressed. We introduce the Vision-Language Disinformation Detection Benchmark (VLDBench), the first large-scale resource supporting both unimodal (text-only) and multimodal (text + image) disinformation detection. VLDBench comprises approximately 62,000 labeled text-image pairs across 13 categories, curated from 58 news outlets. Using a semi-automated pipeline followed by expert review, 22 domain experts invested over 500 hours to produce high-quality annotations with substantial inter-annotator agreement. Evaluations of state-of-the-art Large Language Models (LLMs) and Vision-Language Models (VLMs) on VLDBench show that incorporating visual cues improves detection accuracy by 5 to 35 percentage points over text-only models. VLDBench provides data and code for evaluation, fine-tuning, and robustness testing to support disinformation analysis. Developed in alignment with AI governance frameworks (e.g., the MIT AI Risk Repository), VLDBench offers a principled foundation for advancing trustworthy disinformation detection in multimodal media. Project: https://vectorinstitute.github.io/VLDBench/ Dataset: https://huggingface.co/datasets/vector-institute/VLDBench Code: https://github.com/VectorInstitute/VLDBench

Paperid: 1396, https://arxiv.org/pdf/2502.10810.pdf GitHub

Abstract:
Despite the significant advancements of Large Vision-Language Models (LVLMs) on established benchmarks, there remains a notable gap in suitable evaluation regarding their applicability in the emerging domain of long-context streaming video understanding. Current benchmarks for video understanding typically emphasize isolated single-instance text inputs and fail to evaluate the capacity to sustain temporal reasoning throughout the entire duration of video streams. To address these limitations, we introduce SVBench, a pioneering benchmark with temporal multi-turn question-answering chains specifically designed to thoroughly assess the capabilities of streaming video understanding of current LVLMs. We design a semi-automated annotation pipeline to obtain 49,979 Question-Answer (QA) pairs of 1,353 streaming videos, which includes generating QA chains that represent a series of consecutive multi-turn dialogues over video segments and constructing temporal linkages between successive QA chains. Our experimental results, obtained from 14 models in dialogue and streaming evaluations, reveal that while the closed-source GPT-4o outperforms others, most open-source LVLMs struggle with long-context streaming video understanding. We also construct a StreamingChat model, which significantly outperforms open-source LVLMs on our SVBench and achieves comparable performance on diverse vision-language benchmarks. We expect SVBench to advance the research of streaming video understanding by providing a comprehensive and in-depth analysis of current LVLMs. Our benchmark and model can be accessed at https://yzy-bupt.github.io/SVBench.

Paperid: 1397, https://arxiv.org/pdf/2502.10458.pdf GitHub GitHub

Abstract:
This paper presents ThinkDiff, a novel alignment paradigm that empowers text-to-image diffusion models with multimodal in-context understanding and reasoning capabilities by integrating the strengths of vision-language models (VLMs). Existing multimodal diffusion finetuning methods largely focus on pixel-level reconstruction rather than in-context reasoning, and are constrained by the complexity and limited availability of reasoning-based datasets. ThinkDiff addresses these challenges by leveraging vision-language training as a proxy task, aligning VLMs with the decoder of an encoder-decoder large language model (LLM) instead of a diffusion decoder. This proxy task builds on the observation that the $\textbf{LLM decoder}$ shares the same input feature space with $\textbf{diffusion decoders}$ that use the corresponding $\textbf{LLM encoder}$ for prompt embedding. As a result, aligning VLMs with diffusion decoders can be simplified through alignment with the LLM decoder. Without complex training and datasets, ThinkDiff effectively unleashes understanding, reasoning, and composing capabilities in diffusion models. Experiments demonstrate that ThinkDiff significantly improves accuracy from 19.2% to 46.3% on the challenging CoBSAT benchmark for multimodal in-context reasoning generation, with only 5 hours of training on 4 A100 GPUs. Additionally, ThinkDiff demonstrates exceptional performance in composing multiple images and texts into logically coherent images. Project page: https://mizhenxing.github.io/ThinkDiff.

Paperid: 1398, https://arxiv.org/pdf/2502.10090.pdf GitHub

Abstract:
Humans possess an extraordinary ability to understand and execute complex manipulation tasks by interpreting abstract instruction manuals. For robots, however, this capability remains a substantial challenge, as they cannot interpret abstract instructions and translate them into executable actions. In this paper, we present Manual2Skill, a novel framework that enables robots to perform complex assembly tasks guided by high-level manual instructions. Our approach leverages a Vision-Language Model (VLM) to extract structured information from instructional images and then uses this information to construct hierarchical assembly graphs. These graphs represent parts, subassemblies, and the relationships between them. To facilitate task execution, a pose estimation model predicts the relative 6D poses of components at each assembly step. At the same time, a motion planning module generates actionable sequences for real-world robotic implementation. We demonstrate the effectiveness of Manual2Skill by successfully assembling several real-world IKEA furniture items. This application highlights its ability to manage long-horizon manipulation tasks with both efficiency and precision, significantly enhancing the practicality of robot learning from instruction manuals. This work marks a step forward in advancing robotic systems capable of understanding and executing complex manipulation tasks in a manner akin to human capabilities.Project Page: https://owensun2004.github.io/Furniture-Assembly-Web/

Paperid: 1399, https://arxiv.org/pdf/2502.09838.pdf GitHub

Abstract:
We present HealthGPT, a powerful Medical Large Vision-Language Model (Med-LVLM) that integrates medical visual comprehension and generation capabilities within a unified autoregressive paradigm. Our bootstrapping philosophy is to progressively adapt heterogeneous comprehension and generation knowledge to pre-trained large language models (LLMs). This is achieved through a novel heterogeneous low-rank adaptation (H-LoRA) technique, which is complemented by a tailored hierarchical visual perception approach and a three-stage learning strategy. To effectively learn the HealthGPT, we devise a comprehensive medical domain-specific comprehension and generation dataset called VL-Health. Experimental results demonstrate exceptional performance and scalability of HealthGPT in medical visual unified tasks. Our project can be accessed at https://github.com/DCDmllm/HealthGPT.

Paperid: 1400, https://arxiv.org/pdf/2502.09057.pdf GitHub

Abstract:
We propose general visual inspection model using Vision-Language Model~(VLM) with few-shot images of non-defective or defective products, along with explanatory texts that serve as inspection criteria. Although existing VLM exhibit high performance across various tasks, they are not trained on specific tasks such as visual inspection. Thus, we construct a dataset consisting of diverse images of non-defective and defective products collected from the web, along with unified formatted output text, and fine-tune VLM. For new products, our method employs In-Context Learning, which allows the model to perform inspections with an example of non-defective or defective image and the corresponding explanatory texts with visual prompts. This approach eliminates the need to collect a large number of training samples and re-train the model for each product. The experimental results show that our method achieves high performance, with MCC of 0.804 and F1-score of 0.950 on MVTec AD in a one-shot manner. Our code is available at~https://github.com/ia-gu/Vision-Language-In-Context-Learning-Driven-Few-Shot-Visual-Inspection-Model.

Paperid: 1401, https://arxiv.org/pdf/2502.08193.pdf GitHub

Abstract:
Large Vision-Language Models (LVLMs) are susceptible to typographic attacks, which are misclassifications caused by an attack text that is added to an image. In this paper, we introduce a multi-image setting for studying typographic attacks, broadening the current emphasis of the literature on attacking individual images. Specifically, our focus is on attacking image sets without repeating the attack query. Such non-repeating attacks are stealthier, as they are more likely to evade a gatekeeper than attacks that repeat the same attack text. We introduce two attack strategies for the multi-image setting, leveraging the difficulty of the target image, the strength of the attack text, and text-image similarity. Our text-image similarity approach improves attack success rates by 21% over random, non-specific methods on the CLIP model using ImageNet while maintaining stealth in a multi-image scenario. An additional experiment demonstrates transferability, i.e., text-image similarity calculated using CLIP transfers when attacking InstructBLIP.

Paperid: 1402, https://arxiv.org/pdf/2502.08168.pdf GitHub

Abstract:
As a powerful all-weather Earth observation tool, synthetic aperture radar (SAR) remote sensing enables critical military reconnaissance, maritime surveillance, and infrastructure monitoring. Although Vision language models (VLMs) have made remarkable progress in natural language processing and image understanding, their applications remain limited in professional domains due to insufficient domain expertise. This paper innovatively proposes the first large-scale multimodal dialogue dataset for SAR images, named SARChat-2M, which contains approximately 2 million high-quality image-text pairs, encompasses diverse scenarios with detailed target annotations. This dataset not only supports several key tasks such as visual understanding and object detection tasks, but also has unique innovative aspects: this study develop a visual-language dataset and benchmark for the SAR domain, enabling and evaluating VLMs' capabilities in SAR image interpretation, which provides a paradigmatic framework for constructing multimodal datasets across various remote sensing vertical domains. Through experiments on 16 mainstream VLMs, the effectiveness of the dataset has been fully verified. The project will be released at https://github.com/JimmyMa99/SARChat.

Paperid: 1403, https://arxiv.org/pdf/2502.07183.pdf GitHub

Abstract:
Guide dog robots offer promising solutions to enhance mobility and safety for visually impaired individuals, addressing the limitations of traditional guide dogs, particularly in perceptual intelligence and communication. With the emergence of Vision-Language Models (VLMs), robots are now capable of generating natural language descriptions of their surroundings, aiding in safer decision-making. However, existing VLMs often struggle to accurately interpret and convey spatial relationships, which is crucial for navigation in complex environments such as street crossings. We introduce the Space-Aware Instruction Tuning (SAIT) dataset and the Space-Aware Benchmark (SA-Bench) to address the limitations of current VLMs in understanding physical environments. Our automated data generation pipeline focuses on the virtual path to the destination in 3D space and the surroundings, enhancing environmental comprehension and enabling VLMs to provide more accurate guidance to visually impaired individuals. We also propose an evaluation protocol to assess VLM effectiveness in delivering walking guidance. Comparative experiments demonstrate that our space-aware instruction-tuned model outperforms state-of-the-art algorithms. We have fully open-sourced the SAIT dataset and SA-Bench, along with the related code, at https://github.com/byungokhan/Space-awareVLM

Paperid: 1404, https://arxiv.org/pdf/2502.06884.pdf GitHub

Abstract:
Large Language and Vision-Language Models (LLMs/VLMs) are increasingly used in safety-critical applications, yet their opaque decision-making complicates risk assessment and reliability. Uncertainty quantification (UQ) helps assess prediction confidence and enables abstention when uncertainty is high. Conformal prediction (CP), a leading UQ method, provides statistical guarantees but relies on static thresholds, which fail to adapt to task complexity and evolving data distributions, leading to suboptimal trade-offs in accuracy, coverage, and informativeness. To address this, we propose learnable conformal abstention, integrating reinforcement learning (RL) with CP to optimize abstention thresholds dynamically. By treating CP thresholds as adaptive actions, our approach balances multiple objectives, minimizing prediction set size while maintaining reliable coverage. Extensive evaluations across diverse LLM/VLM benchmarks show our method outperforms Least Ambiguous Classifiers (LAC) and Adaptive Prediction Sets (APS), improving accuracy by up to 3.2%, boosting AUROC for hallucination detection by 22.19%, enhancing uncertainty-guided selective generation (AUARC) by 21.17%, and reducing calibration error by 70%-85%. These improvements hold across multiple models and datasets while consistently meeting the 90% coverage target, establishing our approach as a more effective and flexible solution for reliable decision-making in safety-critical applications. The code is available at: {https://github.com/sinatayebati/vlm-uncertainty}.

Paperid: 1405, https://arxiv.org/pdf/2502.06814.pdf GitHub GitHub

Abstract:
We introduce Lavender, a simple supervised fine-tuning (SFT) method that boosts the performance of advanced vision-language models (VLMs) by leveraging state-of-the-art image generation models such as Stable Diffusion. Specifically, Lavender aligns the text-vision attention in the VLM transformer with the equivalent used by Stable Diffusion during SFT, instead of adapting separate encoders. This alignment enriches the model's visual understanding and significantly boosts performance across in- and out-of-distribution tasks. Lavender requires just 0.13 million training examples, 2.5% of typical large-scale SFT datasets, and fine-tunes on standard hardware (8 GPUs) in a single day. It consistently improves state-of-the-art open-source multimodal LLMs (e.g., Llama-3.2-11B, MiniCPM-Llama3-v2.5), achieving up to 30% gains and a 68% boost on challenging out-of-distribution medical QA tasks. By efficiently transferring the visual expertise of image generators with minimal supervision, Lavender offers a scalable solution for more accurate vision-language systems. All code, training data, and models will be shared at https://astrazeneca.github.io/vlm/.

Paperid: 1406, https://arxiv.org/pdf/2502.06788.pdf GitHub

Abstract:
Existing encoder-free vision-language models (VLMs) are rapidly narrowing the performance gap with their encoder-based counterparts, highlighting the promising potential for unified multimodal systems with structural simplicity and efficient deployment. We systematically clarify the performance gap between VLMs using pre-trained vision encoders, discrete tokenizers, and minimalist visual layers from scratch, deeply excavating the under-examined characteristics of encoder-free VLMs. We develop efficient strategies for encoder-free VLMs that rival mainstream encoder-based ones. After an in-depth investigation, we launch EVEv2.0, a new and improved family of encoder-free VLMs. We show that: (i) Properly decomposing and hierarchically associating vision and language within a unified model reduces interference between modalities. (ii) A well-designed training strategy enables effective optimization for encoder-free VLMs. Through extensive evaluation, our EVEv2.0 represents a thorough study for developing a decoder-only architecture across modalities, demonstrating superior data efficiency and strong vision-reasoning capability. Code is publicly available at: https://github.com/baaivision/EVE.

Paperid: 1407, https://arxiv.org/pdf/2502.06593.pdf GitHub

Abstract:
Recent advancements in generative AI have made text-guided image inpainting - adding, removing, or altering image regions using textual prompts - widely accessible. However, generating semantically correct photorealistic imagery, typically requires carefully-crafted prompts and iterative refinement by evaluating the realism of the generated content - tasks commonly performed by humans. To automate the generative process, we propose Semantically Aligned and Uncertainty Guided AI Image Inpainting (SAGI), a model-agnostic pipeline, to sample prompts from a distribution that closely aligns with human perception and to evaluate the generated content and discard instances that deviate from such a distribution, which we approximate using pretrained large language models and vision-language models. By applying this pipeline on multiple state-of-the-art inpainting models, we create the SAGI Dataset (SAGI-D), currently the largest and most diverse dataset of AI-generated inpaintings, comprising over 95k inpainted images and a human-evaluated subset. Our experiments show that semantic alignment significantly improves image quality and aesthetics, while uncertainty guidance effectively identifies realistic manipulations - human ability to distinguish inpainted images from real ones drops from 74% to 35% in terms of accuracy, after applying our pipeline. Moreover, using SAGI-D for training several image forensic approaches increases in-domain detection performance on average by 37.4% and out-of-domain generalization by 26.1% in terms of IoU, also demonstrating its utility in countering malicious exploitation of generative AI. Code and dataset are available at https://mever-team.github.io/SAGI/

Paperid: 1408, https://arxiv.org/pdf/2502.06445.pdf GitHub

Abstract:
This paper introduces an open-source benchmark for evaluating Vision-Language Models (VLMs) on Optical Character Recognition (OCR) tasks in dynamic video environments. We present a curated dataset containing 1,477 manually annotated frames spanning diverse domains, including code editors, news broadcasts, YouTube videos, and advertisements. Three state of the art VLMs - Claude-3, Gemini-1.5, and GPT-4o are benchmarked against traditional OCR systems such as EasyOCR and RapidOCR. Evaluation metrics include Word Error Rate (WER), Character Error Rate (CER), and Accuracy. Our results highlight the strengths and limitations of VLMs in video-based OCR tasks, demonstrating their potential to outperform conventional OCR models in many scenarios. However, challenges such as hallucinations, content security policies, and sensitivity to occluded or stylized text remain. The dataset and benchmarking framework are publicly available to foster further research.

Paperid: 1409, https://arxiv.org/pdf/2502.06390.pdf GitHub

Abstract:
Vision-Language Models (VLMs) have gained considerable prominence in recent years due to their remarkable capability to effectively integrate and process both textual and visual information. This integration has significantly enhanced performance across a diverse spectrum of applications, such as scene perception and robotics. However, the deployment of VLMs has also given rise to critical safety and security concerns, necessitating extensive research to assess the potential vulnerabilities these VLM systems may harbor. In this work, we present an in-depth survey of the attack strategies tailored for VLMs. We categorize these attacks based on their underlying objectives - namely jailbreak, camouflage, and exploitation - while also detailing the various methodologies employed for data manipulation of VLMs. Meanwhile, we outline corresponding defense mechanisms that have been proposed to mitigate these vulnerabilities. By discerning key connections and distinctions among the diverse types of attacks, we propose a compelling taxonomy for VLM attacks. Moreover, we summarize the evaluation metrics that comprehensively describe the characteristics and impact of different attacks on VLMs. Finally, we conclude with a discussion of promising future research directions that could further enhance the robustness and safety of VLMs, emphasizing the importance of ongoing exploration in this critical area of study. To facilitate community engagement, we maintain an up-to-date project page, accessible at: https://github.com/AobtDai/VLM_Attack_Paper_List.

Paperid: 1410, https://arxiv.org/pdf/2502.06130.pdf GitHub GitHub

Abstract:
While recent Large Vision-Language Models (LVLMs) have shown remarkable performance in multi-modal tasks, they are prone to generating hallucinatory text responses that do not align with the given visual input, which restricts their practical applicability in real-world scenarios. In this work, inspired by the observation that the text-to-image generation process is the inverse of image-conditioned response generation in LVLMs, we explore the potential of leveraging text-to-image generative models to assist in mitigating hallucinations in LVLMs. We discover that generative models can offer valuable self-feedback for mitigating hallucinations at both the response and token levels. Building on this insight, we introduce self-correcting Decoding with Generative Feedback (DeGF), a novel training-free algorithm that incorporates feedback from text-to-image generative models into the decoding process to effectively mitigate hallucinations in LVLMs. Specifically, DeGF generates an image from the initial response produced by LVLMs, which acts as an auxiliary visual reference and provides self-feedback to verify and correct the initial response through complementary or contrastive decoding. Extensive experimental results validate the effectiveness of our approach in mitigating diverse types of hallucinations, consistently surpassing state-of-the-art methods across six benchmarks. Code is available at https://github.com/zhangce01/DeGF.

Paperid: 1411, https://arxiv.org/pdf/2502.06019.pdf GitHub

Abstract:
Recently, test-time adaptation has garnered attention as a method for tuning models without labeled data. The conventional modus operandi for adapting pre-trained vision-language models (VLMs) during test-time primarily focuses on tuning learnable prompts; however, this approach overlooks potential distribution shifts in the visual representations themselves. In this work, we address this limitation by introducing Test-Time Noise Tuning (TNT), a novel method for handling unpredictable shifts in the visual space. TNT leverages, for the first time, a noise adaptation strategy that optimizes learnable noise directly in the visual input space, enabling adaptive feature learning from a single test sample. We further introduce a novel approach for inter-view representation alignment by explicitly enforcing coherence in embedding distances, ensuring consistent feature representations across views. Combined with scaled logits and confident view selection at inference, TNT substantially enhances VLM generalization and calibration, achieving average gains of +7.38% on natural distributions benchmark and +0.80% on cross-dataset evaluations over zero-shot CLIP. These improvements lay a strong foundation for adaptive out-of-distribution handling.

Paperid: 1412, https://arxiv.org/pdf/2502.05855.pdf GitHub

Abstract:
Enabling robots to perform diverse tasks across varied environments is a central challenge in robot learning. While vision-language-action (VLA) models have shown promise for generalizable robot skills, realizing their full potential requires addressing limitations in action representation and efficient training. Current VLA models often focus on scaling the vision-language model (VLM) component, while the action space representation remains a critical bottleneck. This paper introduces DexVLA, a novel framework designed to enhance the efficiency and generalization capabilities of VLAs for complex, long-horizon tasks across diverse robot embodiments. DexVLA features a novel diffusion-based action expert, scaled to one billion parameters, designed for cross-embodiment learning. A novel embodiment curriculum learning strategy facilitates efficient training: (1) pre-training the diffusion expert that is separable from the VLA on cross-embodiment data, (2) aligning the VLA model to specific embodiments, and (3) post-training for rapid adaptation to new tasks. We conduct comprehensive experiments across multiple embodiments, including single-arm, bimanual, and dexterous hand, demonstrating DexVLA's adaptability to challenging tasks without task-specific adaptation, its ability to learn dexterous skills on novel embodiments with limited data, and its capacity to complete complex, long-horizon tasks using only direct language prompting, such as laundry folding. In all settings, our method demonstrates superior performance compared to state-of-the-art models like Octo, OpenVLA, and Diffusion Policy.

Paperid: 1413, https://arxiv.org/pdf/2502.05450.pdf GitHub

Abstract:
Vision-Language-Action (VLA) models have shown substantial potential in real-world robotic manipulation. However, fine-tuning these models through supervised learning struggles to achieve robust performance due to limited, inconsistent demonstrations, especially in contact-rich environments. In this paper, we propose a reinforced fine-tuning approach for VLA models, named ConRFT, which consists of offline and online fine-tuning with a unified consistency-based training objective, to address these challenges. In the offline stage, our method integrates behavior cloning and Q-learning to effectively extract policy from a small set of demonstrations and stabilize value estimating. In the online stage, the VLA model is further fine-tuned via consistency policy, with human interventions to ensure safe exploration and high sample efficiency. We evaluate our approach on eight diverse real-world manipulation tasks. It achieves an average success rate of 96.3% within 45-90 minutes of online fine-tuning, outperforming prior supervised methods with a 144% improvement in success rate and 1.9x shorter episode length. This work highlights the potential of integrating reinforcement learning to enhance the performance of VLA models for real-world robotic applications. Videos and code are available at our project website https://cccedric.github.io/conrft/.

Paperid: 1414, https://arxiv.org/pdf/2502.05206.pdf GitHub

Authors:Xingjun Ma, Yifeng Gao, Yixu Wang, Ruofan Wang, Xin Wang, Ye Sun, Yifan Ding, Hengyuan Xu, Yunhao Chen, Yunhan Zhao, Hanxun Huang, Yige Li, Yutao Wu, Jiaming Zhang, Xiang Zheng, Yang Bai, Zuxuan Wu, Xipeng Qiu, Jingfeng Zhang, Yiming Li, Xudong Han, Haonan Li, Jun Sun, Cong Wang, Jindong Gu, Baoyuan Wu, Siheng Chen, Tianwei Zhang, Yang Liu, Mingming Gong, Tongliang Liu, Shirui Pan, Cihang Xie, Tianyu Pang, Yinpeng Dong, Ruoxi Jia, Yang Zhang, Shiqing Ma, Xiangyu Zhang, Neil Gong, Chaowei Xiao, Sarah Erfani, Tim Baldwin, Bo Li, Masashi Sugiyama, Dacheng Tao, James Bailey, Yu-Gang Jiang

Abstract:
The rapid advancement of large models, driven by their exceptional abilities in learning and generalization through large-scale pre-training, has reshaped the landscape of Artificial Intelligence (AI). These models are now foundational to a wide range of applications, including conversational AI, recommendation systems, autonomous driving, content generation, medical diagnostics, and scientific discovery. However, their widespread deployment also exposes them to significant safety risks, raising concerns about robustness, reliability, and ethical implications. This survey provides a systematic review of current safety research on large models, covering Vision Foundation Models (VFMs), Large Language Models (LLMs), Vision-Language Pre-training (VLP) models, Vision-Language Models (VLMs), Diffusion Models (DMs), and large-model-powered Agents. Our contributions are summarized as follows: (1) We present a comprehensive taxonomy of safety threats to these models, including adversarial attacks, data poisoning, backdoor attacks, jailbreak and prompt injection attacks, energy-latency attacks, data and model extraction attacks, and emerging agent-specific threats. (2) We review defense strategies proposed for each type of attacks if available and summarize the commonly used datasets and benchmarks for safety research. (3) Building on this, we identify and discuss the open challenges in large model safety, emphasizing the need for comprehensive safety evaluations, scalable and effective defense mechanisms, and sustainable data practices. More importantly, we highlight the necessity of collective efforts from the research community and international collaboration. Our work can serve as a useful reference for researchers and practitioners, fostering the ongoing development of comprehensive defense systems and platforms to safeguard AI models.

Paperid: 1415, https://arxiv.org/pdf/2502.05177.pdf GitHub

Abstract:
We introduce Long-VITA, a simple yet effective large multi-modal model for long-context visual-language understanding tasks. It is adept at concurrently processing and analyzing modalities of image, video, and text over 4K frames or 1M tokens while delivering advanced performances on short-context multi-modal tasks. We propose an effective multi-modal training schema that starts with large language models and proceeds through vision-language alignment, general knowledge learning, and two sequential stages of long-sequence fine-tuning. We further implement context-parallelism distributed inference and logits-masked language modeling head to scale Long-VITA to infinitely long inputs of images and texts during model inference. Regarding training data, Long-VITA is built on a mix of 17M samples from public datasets only and demonstrates the state-of-the-art performance on various multi-modal benchmarks, compared against recent cutting-edge models with internal data. Long-VITA is fully reproducible and supports both NPU and GPU platforms for training and testing. By leveraging our inference designs, Long-VITA models achieve a remarkable 2x prefill speedup and 4x context length extension in single node with 8 GPUs. We hope Long-VITA can serve as a competitive baseline and offer valuable insights for the open-source community in advancing long-context multi-modal understanding.

Paperid: 1416, https://arxiv.org/pdf/2502.05091.pdf GitHub

Abstract:
Vision-language models (VLMs) have been widely applied to 2D medical image analysis due to their ability to align visual and textual representations. However, extending VLMs to 3D imaging remains computationally challenging. Existing 3D VLMs often rely on Vision Transformers (ViTs), which are computationally expensive due to the quadratic complexity of self-attention, or on 3D convolutions, which require large numbers of parameters and FLOPs as kernel size increases. We introduce DCFormer, an efficient 3D image encoder that factorizes 3D convolutions into three parallel 1D convolutions along the depth, height, and width dimensions. This design preserves spatial information while significantly reducing computational cost. Integrated into a CLIP-based vision-language framework, DCFormer is trained and evaluated on CT-RATE, a dataset of 50,188 paired 3D chest CT volumes and radiology reports. In zero-shot and fine-tuned detection of 18 pathologies, as well as in image-text retrieval tasks, DCFormer consistently outperforms state-of-the-art 3D vision encoders, including CT-ViT, ViT, ConvNeXt, PoolFormer, and TransUNet. These results highlight DCFormer's potential for scalable, clinically deployable 3D medical VLMs. Our code is available at: https://github.com/mirthAI/DCFormer.

Paperid: 1417, https://arxiv.org/pdf/2502.04757.pdf GitHub

Abstract:
Current Vision Language Models (VLMs) remain vulnerable to malicious prompts that induce harmful outputs. Existing safety benchmarks for VLMs primarily rely on automated evaluation methods, but these methods struggle to detect implicit harmful content or produce inaccurate evaluations. Therefore, we found that existing benchmarks have low levels of harmfulness, ambiguous data, and limited diversity in image-text pair combinations. To address these issues, we propose the ELITE benchmark, a high-quality safety evaluation benchmark for VLMs, underpinned by our enhanced evaluation method, the ELITE evaluator. The ELITE evaluator explicitly incorporates a toxicity score to accurately assess harmfulness in multimodal contexts, where VLMs often provide specific, convincing, but unharmful descriptions of images. We filter out ambiguous and low-quality image-text pairs from existing benchmarks using the ELITE evaluator and generate diverse combinations of safe and unsafe image-text pairs. Our experiments demonstrate that the ELITE evaluator achieves superior alignment with human evaluations compared to prior automated methods, and the ELITE benchmark offers enhanced benchmark quality and diversity. By introducing ELITE, we pave the way for safer, more robust VLMs, contributing essential tools for evaluating and mitigating safety risks in real-world applications.

Paperid: 1418, https://arxiv.org/pdf/2502.04405.pdf GitHub

Abstract:
Spiking Large Language Models have been shown as a good alternative to LLMs in various scenarios. Existing methods for creating Spiking LLMs, i.e., direct training and ANN-SNN conversion, often suffer from performance degradation and relatively high computational costs. To address these issues, we propose a novel Fast ANN-SNN conversion strategy (FAS) that transforms LLMs into spiking LLMs in two stages. The first stage employs a full-parameter fine-tuning of pre-trained models, so it does not need any direct training from scratch. The second stage introduces a coarse-to-fine calibration method to reduce conversion errors and improve accuracy. Experiments on both language and vision-language tasks across four different scales of LLMs demonstrate that FAS can achieve state-of-the-art performance yet with significantly reduced inference latency and computational costs. Notably, FAS only takes eight timesteps to achieve an accuracy of 3\% higher than that of the OPT-7B model, while reducing energy consumption by 96.63\%. The source code is available at https://github.com/lc783/FAS

Paperid: 1419, https://arxiv.org/pdf/2502.04395.pdf GitHub

Abstract:
Recent advancements in time series forecasting have explored augmenting models with text or vision modalities to improve accuracy. While text provides contextual understanding, it often lacks fine-grained temporal details. Conversely, vision captures intricate temporal patterns but lacks semantic context, limiting the complementary potential of these modalities. To address this, we propose \method, a novel multimodal framework that leverages pre-trained Vision-Language Models (VLMs) to bridge temporal, visual, and textual modalities for enhanced forecasting. Our framework comprises three key components: (1) a Retrieval-Augmented Learner, which extracts enriched temporal features through memory bank interactions; (2) a Vision-Augmented Learner, which encodes time series as informative images; and (3) a Text-Augmented Learner, which generates contextual textual descriptions. These components collaborate with frozen pre-trained VLMs to produce multimodal embeddings, which are then fused with temporal features for final prediction. Extensive experiments demonstrate that Time-VLM achieves superior performance, particularly in few-shot and zero-shot scenarios, thereby establishing a new direction for multimodal time series forecasting. Code is available at https://github.com/CityMind-Lab/ICML25-TimeVLM.

Paperid: 1420, https://arxiv.org/pdf/2502.04389.pdf GitHub GitHub

Abstract:
Diagrams play a crucial role in visually conveying complex relationships and processes within business documentation. Despite recent advances in Vision-Language Models (VLMs) for various image understanding tasks, accurately identifying and extracting the structures and relationships depicted in diagrams continues to pose significant challenges. This study addresses these challenges by proposing a text-driven approach that bypasses reliance on VLMs' visual recognition capabilities. Instead, it utilizes the editable source files--such as xlsx, pptx or docx--where diagram elements (e.g., shapes, lines, annotations) are preserved as textual metadata. In our proof-of-concept, we extracted diagram information from xlsx-based system design documents and transformed the extracted shape data into textual input for Large Language Models (LLMs). This approach allowed the LLM to analyze relationships and generate responses to business-oriented questions without the bottleneck of image-based processing. Experimental comparisons with a VLM-based method demonstrated that the proposed text-driven framework yielded more accurate answers for questions requiring detailed comprehension of diagram structures.The results obtained in this study are not limited to the tested .xlsx files but can also be extended to diagrams in other documents with source files, such as Office pptx and docx formats. These findings highlight the feasibility of circumventing VLM constraints through direct textual extraction from original source files. By enabling robust diagram understanding through LLMs, our method offers a promising path toward enhanced workflow efficiency and information analysis in real-world business scenarios.

Paperid: 1421, https://arxiv.org/pdf/2502.04263.pdf GitHub

Abstract:
Pre-trained multi-modal Vision-Language Models like CLIP are widely used off-the-shelf for a variety of applications. In this paper, we show that the common practice of individually exploiting the text or image encoders of these powerful multi-modal models is highly suboptimal for intra-modal tasks like image-to-image retrieval. We argue that this is inherently due to the CLIP-style inter-modal contrastive loss that does not enforce any intra-modal constraints, leading to what we call intra-modal misalignment. To demonstrate this, we leverage two optimization-based modality inversion techniques that map representations from their input modality to the complementary one without any need for auxiliary data or additional trained adapters. We empirically show that, in the intra-modal tasks of image-to-image and text-to-text retrieval, approaching these tasks inter-modally significantly improves performance with respect to intra-modal baselines on more than fifteen datasets. Additionally, we demonstrate that approaching a native inter-modal task (e.g. zero-shot image classification) intra-modally decreases performance, further validating our findings. Finally, we show that incorporating an intra-modal term in the pre-training objective or narrowing the modality gap between the text and image feature embedding spaces helps reduce the intra-modal misalignment. The code is publicly available at: https://github.com/miccunifi/Cross-the-Gap.

Paperid: 1422, https://arxiv.org/pdf/2502.03997.pdf GitHub

Abstract:
Computer Aided Design (CAD) is indispensable across various industries. \emph{Text-based CAD editing}, which automates the modification of CAD models based on textual instructions, holds great potential but remains underexplored. Existing methods primarily focus on design variation generation or text-based CAD generation, either lacking support for text-based control or neglecting existing CAD models as constraints. We introduce \emph{CAD-Editor}, the first framework for text-based CAD editing. To address the challenge of demanding triplet data with accurate correspondence for training, we propose an automated data synthesis pipeline. This pipeline utilizes design variation models to generate pairs of original and edited CAD models and employs Large Vision-Language Models (LVLMs) to summarize their differences into editing instructions. To tackle the composite nature of text-based CAD editing, we propose a locate-then-infill framework that decomposes the task into two focused sub-tasks: locating regions requiring modification and infilling these regions with appropriate edits. Large Language Models (LLMs) serve as the backbone for both sub-tasks, leveraging their capabilities in natural language understanding and CAD knowledge. Experiments show that CAD-Editor achieves superior performance both quantitatively and qualitatively. The code is available at \url {https://github.com/microsoft/CAD-Editor}.

Paperid: 1423, https://arxiv.org/pdf/2502.03628.pdf GitHub

Abstract:
Large Vision-Language Models (LVLMs) can reason effectively over both textual and visual inputs, but they tend to hallucinate syntactically coherent yet visually ungrounded contents. In this paper, we investigate the internal dynamics of hallucination by examining the tokens logits ranking throughout the generation process, revealing three key patterns in how LVLMs process information: (1) gradual visual information loss - visually grounded tokens gradually become less favored throughout generation, and (2) early excitation - semantically meaningful tokens achieve peak activation in the layers earlier than the final layer. (3) hidden genuine information - visually grounded tokens though not being eventually decoded still retain relatively high rankings at inference. Based on these insights, we propose VISTA (Visual Information Steering with Token-logit Augmentation), a training-free inference-time intervention framework that reduces hallucination while promoting genuine information. VISTA works by combining two complementary approaches: reinforcing visual information in activation space and leveraging early layer activations to promote semantically meaningful decoding. Compared to existing methods, VISTA requires no external supervision and is applicable to various decoding strategies. Extensive experiments show that VISTA on average reduces hallucination by about 40% on evaluated open-ended generation task, and it consistently outperforms existing methods on four benchmarks across four architectures under three decoding strategies. Code is available at https://github.com/LzVv123456/VISTA.

Paperid: 1424, https://arxiv.org/pdf/2502.02589.pdf GitHub

Abstract:
This paper introduces the COCONut-PanCap dataset, created to enhance panoptic segmentation and grounded image captioning. Building upon the COCO dataset with advanced COCONut panoptic masks, this dataset aims to overcome limitations in existing image-text datasets that often lack detailed, scene-comprehensive descriptions. The COCONut-PanCap dataset incorporates fine-grained, region-level captions grounded in panoptic segmentation masks, ensuring consistency and improving the detail of generated captions. Through human-edited, densely annotated descriptions, COCONut-PanCap supports improved training of vision-language models (VLMs) for image understanding and generative models for text-to-image tasks. Experimental results demonstrate that COCONut-PanCap significantly boosts performance across understanding and generation tasks, offering complementary benefits to large-scale datasets. This dataset sets a new benchmark for evaluating models on joint panoptic segmentation and grounded captioning tasks, addressing the need for high-quality, detailed image-text annotations in multi-modal learning.

Paperid: 1425, https://arxiv.org/pdf/2502.02548.pdf GitHub

Abstract:
We tackle open-vocabulary 3D scene understanding by introducing a novel data generation pipeline and training framework. Our method addresses three critical requirements for effective training: precise 3D region segmentation, comprehensive textual descriptions, and sufficient dataset scale. By leveraging state-of-the-art open-vocabulary image segmentation models and region-aware Vision-Language Models, we develop an automatic pipeline that generates high-quality 3D mask-text pairs. Applying this pipeline to multiple 3D scene datasets, we create Mosaic3D-5.6M, a dataset of over 30K annotated scenes with 5.6M mask-text pairs, significantly larger than existing datasets. Building upon this data, we propose Mosaic3D, a foundation model combining a 3D encoder trained with contrastive learning and a lightweight mask decoder for open-vocabulary 3D semantic and instance segmentation. Our approach achieves state-of-the-art results on open-vocabulary 3D semantic and instance segmentation tasks including ScanNet200, Matterport3D, and ScanNet++, with ablation studies validating the effectiveness of our large-scale training data.

Paperid: 1426, https://arxiv.org/pdf/2502.01977.pdf GitHub

Abstract:
User interface understanding with vision-language models (VLMs) has received much attention due to its potential for enhancing software automation. However, existing datasets used to build UI-VLMs either only contain large-scale context-free element annotations or contextualized functional descriptions for elements at a small scale. In this work, we propose the \textbf{AutoGUI} pipeline for automatically annotating UI elements with detailed functionality descriptions at scale. Specifically, we leverage large language models (LLMs) to infer element functionality by comparing UI state changes before and after simulated interactions. To improve annotation quality, we propose LLM-aided rejection and verification, eliminating invalid annotations without human labor. We construct a high-quality AutoGUI-704k dataset using the proposed pipeline, featuring diverse and detailed functionality annotations that are hardly provided by previous datasets. Human evaluation shows that we achieve annotation correctness comparable to a trained human annotator. Extensive experiments show that our dataset remarkably enhances VLM's UI grounding capabilities and exhibits significant scaling effects. We also show the interesting potential use of our dataset in UI agent tasks. Please view our project at https://autogui-project.github.io/.

Paperid: 1427, https://arxiv.org/pdf/2502.01828.pdf GitHub

Abstract:
While generative robot policies have demonstrated significant potential in learning complex, multimodal behaviors from demonstrations, they still exhibit diverse failures at deployment-time. Policy steering offers an elegant solution to reducing the chance of failure by using an external verifier to select from low-level actions proposed by an imperfect generative policy. Here, one might hope to use a Vision Language Model (VLM) as a verifier, leveraging its open-world reasoning capabilities. However, off-the-shelf VLMs struggle to understand the consequences of low-level robot actions as they are represented fundamentally differently than the text and images the VLM was trained on. In response, we propose FOREWARN, a novel framework to unlock the potential of VLMs as open-vocabulary verifiers for runtime policy steering. Our key idea is to decouple the VLM's burden of predicting action outcomes (foresight) from evaluation (forethought). For foresight, we leverage a latent world model to imagine future latent states given diverse low-level action plans. For forethought, we align the VLM with these predicted latent states to reason about the consequences of actions in its native representation--natural language--and effectively filter proposed plans. We validate our framework across diverse robotic manipulation tasks, demonstrating its ability to bridge representational gaps and provide robust, generalizable policy steering. Videos can be found on the project website: https://yilin-wu98.github.io/forewarn/.

Paperid: 1428, https://arxiv.org/pdf/2502.01576.pdf GitHub

Abstract:
Multi-modal Large Language Models (MLLMs) excel in vision-language tasks but remain vulnerable to visual adversarial perturbations that can induce hallucinations, manipulate responses, or bypass safety mechanisms. Existing methods seek to mitigate these risks by applying constrained adversarial fine-tuning to CLIP vision encoders on ImageNet-scale data, ensuring their generalization ability is preserved. However, this limited adversarial training restricts robustness and broader generalization. In this work, we explore an alternative approach of leveraging existing vision classification models that have been adversarially pre-trained on large-scale data. Our analysis reveals two principal contributions: (1) the extensive scale and diversity of adversarial pre-training enables these models to demonstrate superior robustness against diverse adversarial threats, ranging from imperceptible perturbations to advanced jailbreaking attempts, without requiring additional adversarial training, and (2) end-to-end MLLM integration with these robust models facilitates enhanced adaptation of language components to robust visual features, outperforming existing plug-and-play methodologies on complex reasoning tasks. Through systematic evaluation across visual question-answering, image captioning, and jail-break attacks, we demonstrate that MLLMs trained with these robust models achieve superior adversarial robustness while maintaining favorable clean performance. Our framework achieves 2x and 1.5x average robustness gains in captioning and VQA tasks, respectively, and delivers over 10% improvement against jailbreak attacks. Code and pretrained models will be available at https://github.com/HashmatShadab/Robust-LLaVA.

Paperid: 1429, https://arxiv.org/pdf/2502.01167.pdf GitHub

Abstract:
The introduction of robots into everyday scenarios necessitates algorithms capable of monitoring the execution of tasks. In this paper, we propose ConditionNET, an approach for learning the preconditions and effects of actions in a fully data-driven manner. We develop an efficient vision-language model and introduce additional optimization objectives during training to optimize for consistent feature representations. ConditionNET explicitly models the dependencies between actions, preconditions, and effects, leading to improved performance. We evaluate our model on two robotic datasets, one of which we collected for this paper, containing 406 successful and 138 failed teleoperated demonstrations of a Franka Emika Panda robot performing tasks like pouring and cleaning the counter. We show in our experiments that ConditionNET outperforms all baselines on both anomaly detection and phase prediction tasks. Furthermore, we implement an action monitoring system on a real robot to demonstrate the practical applicability of the learned preconditions and effects. Our results highlight the potential of ConditionNET for enhancing the reliability and adaptability of robots in real-world environments. The data is available on the project website: https://dsliwowski1.github.io/ConditionNET_page.

Paperid: 1430, https://arxiv.org/pdf/2502.01051.pdf GitHub

Abstract:
Preference optimization for diffusion models aims to align them with human preferences for images. Previous methods typically use Vision-Language Models (VLMs) as pixel-level reward models to approximate human preferences. However, when used for step-level preference optimization, these models face challenges in handling noisy images of different timesteps and require complex transformations into pixel space. In this work, we show that pre-trained diffusion models are naturally suited for step-level reward modeling in the noisy latent space, as they are explicitly designed to process latent images at various noise levels. Accordingly, we propose the Latent Reward Model (LRM), which repurposes components of the diffusion model to predict preferences of latent images at arbitrary timesteps. Building on LRM, we introduce Latent Preference Optimization (LPO), a step-level preference optimization method conducted directly in the noisy latent space. Experimental results indicate that LPO significantly improves the model's alignment with general, aesthetic, and text-image alignment preferences, while achieving a 2.5-28x training speedup over existing preference optimization methods. Our code and models are available at https://github.com/Kwai-Kolors/LPO.

Paperid: 1431, https://arxiv.org/pdf/2502.00954.pdf GitHub

Abstract:
The rise of vision-language foundation models marks an advancement in bridging the gap between human and machine capabilities in 3D scene reasoning. Existing 3D reasoning benchmarks assume real-time scene accessibility, which is impractical due to the high cost of frequent scene updates. To this end, we introduce Hypothetical 3D Reasoning, namely Hypo3D, a benchmark designed to evaluate models' ability to reason without access to real-time scene data. Models need to imagine the scene state based on a provided change description before reasoning. Hypo3D is formulated as a 3D Visual Question Answering (VQA) benchmark, comprising 7,727 context changes across 700 indoor scenes, resulting in 14,885 question-answer pairs. An anchor-based world frame is established for all scenes, ensuring consistent reference to a global frame for directional terms in context changes and QAs. Extensive experiments show that state-of-the-art foundation models struggle to reason in hypothetically changed scenes. This reveals a substantial performance gap compared to humans, particularly in scenarios involving movement changes and directional reasoning. Even when the context change is irrelevant to the question, models often incorrectly adjust their answers. Project website: https://matchlab-imperial.github.io/Hypo3D/

Paperid: 1432, https://arxiv.org/pdf/2502.00719.pdf GitHub

Abstract:
Segment Anything Model (SAM) represents a large-scale segmentation model that enables powerful zero-shot capabilities with flexible prompts. While SAM can segment any object in zero-shot, it requires user-provided prompts for each target image and does not attach any label information to masks. Few-shot segmentation models addressed these issues by inputting annotated reference images as prompts to SAM and can segment specific objects in target images without user-provided prompts. Previous SAM-based few-shot segmentation models only use annotated reference images as prompts, resulting in limited accuracy due to a lack of reference information. In this paper, we propose a novel few-shot segmentation model, Vision and Language reference Prompt into SAM (VLP-SAM), that utilizes the visual information of the reference images and the semantic information of the text labels by inputting not only images but also language as reference information. In particular, VLP-SAM is a simple and scalable structure with minimal learnable parameters, which inputs prompt embeddings with vision-language information into SAM using a multimodal vision-language model. To demonstrate the effectiveness of VLP-SAM, we conducted experiments on the PASCAL-5i and COCO-20i datasets, and achieved high performance in the few-shot segmentation task, outperforming the previous state-of-the-art model by a large margin (6.3% and 9.5% in mIoU, respectively). Furthermore, VLP-SAM demonstrates its generality in unseen objects that are not included in the training data. Our code is available at https://github.com/kosukesakurai1/VLP-SAM.

Paperid: 1433, https://arxiv.org/pdf/2502.00372.pdf GitHub

Abstract:
Visual Grounding (VG) tasks, such as referring expression detection and segmentation tasks are important for linking visual entities to context, especially in complex reasoning tasks that require detailed query interpretation. This paper explores VG beyond basic perception, highlighting challenges for methods that require reasoning like human cognition. Recent advances in large language methods (LLMs) and Vision-Language methods (VLMs) have improved abilities for visual comprehension, contextual understanding, and reasoning. These methods are mainly split into end-to-end and compositional methods, with the latter offering more flexibility. Compositional approaches that integrate LLMs and foundation models show promising performance but still struggle with complex reasoning with language-based logical representations. To address these limitations, we propose NAVER, a compositional visual grounding method that integrates explicit probabilistic logic reasoning within a finite-state automaton, equipped with a self-correcting mechanism. This design improves robustness and interpretability in inference through explicit logic reasoning. Our results show that NAVER achieves SoTA performance comparing to recent end-to-end and compositional baselines. The code is available at https://github.com/ControlNet/NAVER .

Paperid: 1434, https://arxiv.org/pdf/2501.19252.pdf GitHub

Abstract:
The remarkable progress in text-to-video diffusion models enables photorealistic generations, although the contents of the generated video often include unnatural movement or deformation, reverse playback, and motionless scenes. Recently, an alignment problem has attracted huge attention, where we steer the output of diffusion models based on some quantity on the goodness of the content. Because there is a large room for improvement of perceptual quality along the frame direction, we should address which metrics we should optimize and how we can optimize them in the video generation. In this paper, we propose diffusion latent beam search with lookahead estimator, which can select a better diffusion latent to maximize a given alignment reward, at inference time. We then point out that the improvement of perceptual video quality considering the alignment to prompts requires reward calibration by weighting existing metrics. This is because when humans or vision language models evaluate outputs, many previous metrics to quantify the naturalness of video do not always correlate with evaluation. We demonstrate that our method improves the perceptual quality evaluated on the calibrated reward, VLMs, and human assessment, without model parameter update, and outputs the best generation compared to greedy search and best-of-N sampling under much more efficient computational cost. The experiments highlight that our method is beneficial to many capable generative models, and provide a practical guideline that we should prioritize the inference-time compute allocation into lookahead steps for reward estimation over search budget or denoising steps.

Paperid: 1435, https://arxiv.org/pdf/2501.18463.pdf GitHub

Abstract:
Out-of-distribution (OOD) detection is a task that detects OOD samples during inference to ensure the safety of deployed models. However, conventional benchmarks have reached performance saturation, making it difficult to compare recent OOD detection methods. To address this challenge, we introduce three novel OOD detection benchmarks that enable a deeper understanding of method characteristics and reflect real-world conditions. First, we present ImageNet-X, designed to evaluate performance under challenging semantic shifts. Second, we propose ImageNet-FS-X for full-spectrum OOD detection, assessing robustness to covariate shifts (feature distribution shifts). Finally, we propose Wilds-FS-X, which extends these evaluations to real-world datasets, offering a more comprehensive testbed. Our experiments reveal that recent CLIP-based OOD detection methods struggle to varying degrees across the three proposed benchmarks, and none of them consistently outperforms the others. We hope the community goes beyond specific benchmarks and includes more challenging conditions reflecting real-world scenarios. The code is https://github.com/hoshi23/OOD-X-Benchmarks.

Paperid: 1436, https://arxiv.org/pdf/2501.16458.pdf GitHub

Abstract:
Cloth folding is a complex task due to the inevitable self-occlusions of clothes, their complicated dynamics, and the disparate materials, geometries, and textures that garments can have. In this work, we learn folding actions conditioned on text commands. Translating high-level, abstract instructions into precise robotic actions requires sophisticated language understanding and manipulation capabilities. To do that, we leverage a pre-trained vision-language model and repurpose it to predict manipulation actions. Our model, BiFold, can take context into account and achieves state-of-the-art performance on an existing language-conditioned folding benchmark. To address the lack of annotated bimanual folding data, we introduce a novel dataset with automatically parsed actions and language-aligned instructions, enabling better learning of text-conditioned manipulation. BiFold attains the best performance on our dataset and demonstrates strong generalization to new instructions, garments, and environments.

Paperid: 1437, https://arxiv.org/pdf/2501.16037.pdf GitHub

Abstract:
This paper presents a novel approach for hazard analysis in dashcam footage, addressing the detection of driver reactions to hazards, the identification of hazardous objects, and the generation of descriptive captions. We first introduce a method for detecting driver reactions through speed and sound anomaly detection, leveraging unsupervised learning techniques. For hazard detection, we employ a set of heuristic rules as weak classifiers, which are combined using an ensemble method. This ensemble approach is further refined with differential privacy to mitigate overconfidence, ensuring robustness despite the lack of labeled data. Lastly, we use state-of-the-art vision-language models for hazard captioning, generating descriptive labels for the detected hazards. Our method achieved the highest scores in the Challenge on Out-of-Label in Autonomous Driving, demonstrating its effectiveness across all three tasks. Source codes are publicly available at https://github.com/ffyyytt/COOOL_2025.

Paperid: 1438, https://arxiv.org/pdf/2501.15798.pdf GitHub

Abstract:
Vision-language pretraining (VLP) has been investigated to generalize across diverse downstream tasks for fundus image analysis. Although recent methods showcase promising achievements, they significantly rely on large-scale private image-text data but pay less attention to the pretraining manner, which limits their further advancements. In this work, we introduce MM-Retinal V2, a high-quality image-text paired dataset comprising CFP, FFA, and OCT image modalities. Then, we propose a novel fundus vision-language pretraining model, namely KeepFIT V2, which is pretrained by integrating knowledge from the elite data spark into categorical public datasets. Specifically, a preliminary textual pretraining is adopted to equip the text encoder with primarily ophthalmic textual knowledge. Moreover, a hybrid image-text knowledge injection module is designed for knowledge transfer, which is essentially based on a combination of global semantic concepts from contrastive learning and local appearance details from generative learning. Extensive experiments across zero-shot, few-shot, and linear probing settings highlight the generalization and transferability of KeepFIT V2, delivering performance competitive to state-of-the-art fundus VLP models trained on large-scale private image-text datasets. Our dataset and model are publicly available via https://github.com/lxirich/MM-Retinal.

Paperid: 1439, https://arxiv.org/pdf/2501.15503.pdf GitHub

Abstract:
The classification and recognition of maritime objects are crucial for enhancing maritime safety, monitoring, and intelligent sea environment prediction. However, existing unsupervised methods for maritime object classification often struggle with the long-tail data distributions in both object categories and weather conditions. In this paper, we construct a dataset named AIMO produced by large-scale generative models with diverse weather conditions and balanced object categories, and collect a dataset named RMO with real-world images where long-tail issue exists. We propose a novel domain adaptation approach that leverages AIMO (source domain) to address the problem of limited labeled data, unbalanced distribution and domain shift in RMO (target domain), enhance the generalization of source features with the Vision-Language Models such as CLIP, and propose a difficulty score for curriculum learning to optimize training process. Experimental results shows that the proposed method significantly improves the classification accuracy, particularly for samples within rare object categories and weather conditions. Datasets and codes will be publicly available at https://github.com/honoria0204/AIMO.

Paperid: 1440, https://arxiv.org/pdf/2501.15438.pdf GitHub

Abstract:
Detecting hate speech in online content is essential to ensuring safer digital spaces. While significant progress has been made in text and meme modalities, video-based hate speech detection remains under-explored, hindered by a lack of annotated datasets and the high cost of video annotation. This gap is particularly problematic given the growing reliance on large models, which demand substantial amounts of training data. To address this challenge, we leverage meme datasets as both a substitution and an augmentation strategy for training hateful video detection models. Our approach introduces a human-assisted reannotation pipeline to align meme dataset labels with video datasets, ensuring consistency with minimal labeling effort. Using two state-of-the-art vision-language models, we demonstrate that meme data can substitute for video data in resource-scarce scenarios and augment video datasets to achieve further performance gains. Our results consistently outperform state-of-the-art benchmarks, showcasing the potential of cross-modal transfer learning for advancing hateful video detection. Dataset and code are available at https://github.com/Social-AI-Studio/CrossModalTransferLearning.

Paperid: 1441, https://arxiv.org/pdf/2501.15415.pdf GitHub

Abstract:
Understanding the chemical structure from a graphical representation of a molecule is a challenging image caption task that would greatly benefit molecule-centric scientific discovery. Variations in molecular images and caption subtasks pose a significant challenge in both image representation learning and task modeling. Yet, existing methods only focus on a specific caption task that translates a molecular image into its graph structure, i.e., OCSR. In this paper, we propose the Optical Chemical Structure Understanding (OCSU) task, which extends low-level recognition to multilevel understanding and aims to translate chemical structure diagrams into readable strings for both machine and chemist. To facilitate the development of OCSU technology, we explore both OCSR-based and OCSR-free paradigms. We propose DoubleCheck to enhance OCSR performance via attentive feature enhancement for local ambiguous atoms. It can be cascaded with existing SMILES-based molecule understanding methods to achieve OCSU. Meanwhile, Mol-VL is a vision-language model end-to-end optimized for OCSU. We also construct Vis-CheBI20, the first large-scale OCSU dataset. Through comprehensive experiments, we demonstrate the proposed approaches excel at providing chemist-readable caption for chemical structure diagrams, which provide solid baselines for further research. Our code, model, and data are open-sourced at https://github.com/PharMolix/OCSU.

Paperid: 1442, https://arxiv.org/pdf/2501.14607.pdf GitHub

Abstract:
Referring video object segmentation (RVOS) aims to segment target objects throughout a video based on a text description. This is challenging as it involves deep vision-language understanding, pixel-level dense prediction and spatiotemporal reasoning. Despite notable progress in recent years, existing methods still exhibit a noticeable gap when considering all these aspects. In this work, we propose \textbf{ReferDINO}, a strong RVOS model that inherits region-level vision-language alignment from foundational visual grounding models, and is further endowed with pixel-level dense perception and cross-modal spatiotemporal reasoning. In detail, ReferDINO integrates two key components: 1) a grounding-guided deformable mask decoder that utilizes location prediction to progressively guide mask prediction through differentiable deformation mechanisms; 2) an object-consistent temporal enhancer that injects pretrained time-varying text features into inter-frame interaction to capture object-aware dynamic changes. Moreover, a confidence-aware query pruning strategy is designed to accelerate object decoding without compromising model performance. Extensive experimental results on five benchmarks demonstrate that our ReferDINO significantly outperforms previous methods (e.g., +3.9% (\mathcal{J}&\mathcal{F}) on Ref-YouTube-VOS) with real-time inference speed (51 FPS).

Paperid: 1443, https://arxiv.org/pdf/2501.14369.pdf GitHub

Abstract:
Research on continual learning in multi-modal tasks has been receiving increasing attention. However, most existing work overlooks the explicit cross-modal and cross-task interactions. In this paper, we innovatively propose the Low-rank Prompt Interaction (LPI) to address this general problem of multi-modal understanding, which considers both cross-modal and cross-task interactions. Specifically, as for the former, we employ multi-modal correlation modules for corresponding Transformer layers. Considering that the training parameters scale to the number of layers and tasks, we propose low-rank interaction-augmented decomposition to avoid memory explosion while enhancing the cross-modal association through sharing and separating common-specific low-rank factors. In addition, due to the multi-modal semantic differences carried by the low-rank initialization, we adopt hierarchical low-rank contrastive learning to ensure training robustness. As for the latter, we initially employ a visual analysis and identify that different tasks have clear distinctions in proximity. Therefore, we introduce explicit task contrastive constraints in the prompt learning process based on task semantic distances. Experiments on two retrieval tasks show performance improvements with the introduction of a minimal number of parameters, demonstrating the effectiveness of our method. Code is available at https://github.com/Kelvin-ywc/LPI.

Paperid: 1444, https://arxiv.org/pdf/2501.13982.pdf GitHub

Abstract:
Visual reprogramming (VR) reuses pre-trained vision models for downstream image classification tasks by adding trainable noise patterns to inputs. When applied to vision-language models (e.g., CLIP), existing VR approaches follow the same pipeline used in vision models (e.g., ResNet, ViT), where ground-truth class labels are inserted into fixed text templates to guide the optimization of VR patterns. This label-based approach, however, overlooks the rich information and diverse attribute-guided textual representations that CLIP can exploit, which may lead to the misclassification of samples. In this paper, we propose Attribute-based Visual Reprogramming (AttrVR) for CLIP, utilizing descriptive attributes (DesAttrs) and distinctive attributes (DistAttrs), which respectively represent common and unique feature descriptions for different classes. Besides, as images of the same class may reflect different attributes after VR, AttrVR iteratively refines patterns using the $k$-nearest DesAttrs and DistAttrs for each image sample, enabling more dynamic and sample-specific optimization. Theoretically, AttrVR is shown to reduce intra-class variance and increase inter-class separation. Empirically, it achieves superior performance in 12 downstream tasks for both ViT-based and ResNet-based CLIP. The success of AttrVR facilitates more effective integration of VR from unimodal vision models into vision-language models. Our code is available at https://github.com/tmlr-group/AttrVR.

Paperid: 1445, https://arxiv.org/pdf/2501.13851.pdf GitHub

Abstract:
Memes have emerged as a powerful form of communication, integrating visual and textual elements to convey humor, satire, and cultural messages. Existing research has focused primarily on aspects such as emotion classification, meme generation, propagation, interpretation, figurative language, and sociolinguistics, but has often overlooked deeper meme comprehension and meme-text retrieval. To address these gaps, this study introduces ClassicMemes-50-templates (CM50), a large-scale dataset consisting of over 33,000 memes, centered around 50 popular meme templates. We also present an automated knowledge-grounded annotation pipeline leveraging large vision-language models to produce high-quality image captions, meme captions, and literary device labels overcoming the labor intensive demands of manual annotation. Additionally, we propose a meme-text retrieval CLIP model (mtrCLIP) that utilizes cross-modal embedding to enhance meme analysis, significantly improving retrieval performance. Our contributions include:(1) a novel dataset for large-scale meme study, (2) a scalable meme annotation framework, and (3) a fine-tuned CLIP for meme-text retrieval, all aimed at advancing the understanding and analysis of memes at scale.

Paperid: 1446, https://arxiv.org/pdf/2501.13106.pdf GitHub

Abstract:
In this paper, we propose VideoLLaMA3, a more advanced multimodal foundation model for image and video understanding. The core design philosophy of VideoLLaMA3 is vision-centric. The meaning of "vision-centric" is two-fold: the vision-centric training paradigm and vision-centric framework design. The key insight of our vision-centric training paradigm is that high-quality image-text data is crucial for both image and video understanding. Instead of preparing massive video-text datasets, we focus on constructing large-scale and high-quality image-text datasets. VideoLLaMA3 has four training stages: 1) Vision Encoder Adaptation, which enables vision encoder to accept images of variable resolutions as input; 2) Vision-Language Alignment, which jointly tunes the vision encoder, projector, and LLM with large-scale image-text data covering multiple types (including scene images, documents, charts) as well as text-only data. 3) Multi-task Fine-tuning, which incorporates image-text SFT data for downstream tasks and video-text data to establish a foundation for video understanding. 4) Video-centric Fine-tuning, which further improves the model's capability in video understanding. As for the framework design, to better capture fine-grained details in images, the pretrained vision encoder is adapted to encode images of varying sizes into vision tokens with corresponding numbers, rather than a fixed number of tokens. For video inputs, we reduce the number of vision tokens according to their similarity so that the representation of videos will be more precise and compact. Benefit from vision-centric designs, VideoLLaMA3 achieves compelling performances in both image and video understanding benchmarks.

Paperid: 1447, https://arxiv.org/pdf/2501.12418.pdf GitHub

Abstract:
Vision-Language Models (VLMs) have demonstrated remarkable capabilities in understanding multimodal inputs and have been widely integrated into Retrieval-Augmented Generation (RAG) based conversational systems. While current VLM-powered chatbots can provide textual source references in their responses, they exhibit significant limitations in referencing contextually relevant images during conversations. In this paper, we introduce Contextual Image Reference -- the ability to appropriately reference relevant images from retrieval documents based on conversation context -- and systematically investigate VLMs' capability in this aspect. We conduct the first evaluation for contextual image referencing, comprising a dedicated testing dataset and evaluation metrics. Furthermore, we propose ImageRef-VL, a method that significantly enhances open-source VLMs' image referencing capabilities through instruction fine-tuning on a large-scale, manually curated multimodal conversation dataset. Experimental results demonstrate that ImageRef-VL not only outperforms proprietary models but also achieves an 88% performance improvement over state-of-the-art open-source VLMs in contextual image referencing tasks. Our code is available at https://github.com/bytedance/ImageRef-VL.

Paperid: 1448, https://arxiv.org/pdf/2501.12368.pdf GitHub

Abstract:
Despite the promising performance of Large Vision Language Models (LVLMs) in visual understanding, they occasionally generate incorrect outputs. While reward models (RMs) with reinforcement learning or test-time scaling offer the potential for improving generation quality, a critical gap remains: publicly available multi-modal RMs for LVLMs are scarce, and the implementation details of proprietary models are often unclear. We bridge this gap with InternLM-XComposer2.5-Reward (IXC-2.5-Reward), a simple yet effective multi-modal reward model that aligns LVLMs with human preferences. To ensure the robustness and versatility of IXC-2.5-Reward, we set up a high-quality multi-modal preference corpus spanning text, image, and video inputs across diverse domains, such as instruction following, general understanding, text-rich documents, mathematical reasoning, and video understanding. IXC-2.5-Reward achieves excellent results on the latest multi-modal reward model benchmark and shows competitive performance on text-only reward model benchmarks. We further demonstrate three key applications of IXC-2.5-Reward: (1) Providing a supervisory signal for RL training. We integrate IXC-2.5-Reward with Proximal Policy Optimization (PPO) yields IXC-2.5-Chat, which shows consistent improvements in instruction following and multi-modal open-ended dialogue; (2) Selecting the best response from candidate responses for test-time scaling; and (3) Filtering outlier or noisy samples from existing image and video instruction tuning training data. To ensure reproducibility and facilitate further research, we have open-sourced all model weights and training recipes at https://github.com/InternLM/InternLM-XComposer/tree/main/InternLM-XComposer-2.5-Reward

Paperid: 1449, https://arxiv.org/pdf/2501.12266.pdf GitHub

Abstract:
The main challenges limiting the adoption of deep learning-based solutions in medical workflows are the availability of annotated data and the lack of interpretability of such systems. Concept Bottleneck Models (CBMs) tackle the latter by constraining the final disease prediction on a set of predefined and human-interpretable concepts. However, the increased interpretability achieved through these concept-based explanations implies a higher annotation burden. Moreover, if a new concept needs to be added, the whole system needs to be retrained. Inspired by the remarkable performance shown by Large Vision-Language Models (LVLMs) in few-shot settings, we propose a simple, yet effective, methodology, CBVLM, which tackles both of the aforementioned challenges. First, for each concept, we prompt the LVLM to answer if the concept is present in the input image. Then, we ask the LVLM to classify the image based on the previous concept predictions. Moreover, in both stages, we incorporate a retrieval module responsible for selecting the best examples for in-context learning. By grounding the final diagnosis on the predicted concepts, we ensure explainability, and by leveraging the few-shot capabilities of LVLMs, we drastically lower the annotation cost. We validate our approach with extensive experiments across four medical datasets and twelve LVLMs (both generic and medical) and show that CBVLM consistently outperforms CBMs and task-specific supervised methods without requiring any training and using just a few annotated examples. More information on our project page: https://cristianopatricio.github.io/CBVLM/.

Paperid: 1450, https://arxiv.org/pdf/2501.12206.pdf GitHub GitHub

Abstract:
Large Vision Language Models (LVLMs) have demonstrated remarkable capabilities in understanding and describing visual content, achieving state-of-the-art performance across various vision-language tasks. However, these models often generate descriptions containing objects or details that are absent in the input image, a phenomenon commonly known as hallucination. Our work investigates the key reasons behind this issue by analyzing the pattern of self-attention in transformer layers. We find that hallucinations often arise from the progressive weakening of attention weight to visual tokens in the deeper layers of the LLM. Some previous works naively boost the attention of all visual tokens to mitigate this issue, resulting in suboptimal hallucination reduction. To address this, we identify two critical sets of visual tokens that facilitate the transfer of visual information from the vision encoder to the LLM. Local tokens encode grounded information about objects present in an image, while summary tokens capture the overall aggregated representation of the image. Importantly, these two sets of tokens require different levels of weight enhancement. To this end, we propose \textbf{PAINT} (\textbf{P}aying \textbf{A}ttention to \textbf{IN}formed \textbf{T}okens), a plug-and-play framework that intervenes in the self-attention mechanism of the LLM, selectively boosting the attention weights of local and summary tokens with experimentally learned margins. Evaluation on the MSCOCO image captioning dataset demonstrate that our approach reduces hallucination rates by up to 62.3\% compared to baseline models while maintaining accuracy. Code is available at \href{https://github.com/hasanar1f/PAINT}{https://github.com/hasanar1f/PAINT}

Paperid: 1451, https://arxiv.org/pdf/2501.11485.pdf GitHub

Abstract:
Detecting out-of-distribution (OOD) data is crucial in real-world machine learning applications, particularly in safety-critical domains. Existing methods often leverage language information from vision-language models (VLMs) to enhance OOD detection by improving confidence estimation through rich class-wise text information. However, when building OOD detection score upon on in-distribution (ID) text-image affinity, existing works either focus on each ID class or whole ID label sets, overlooking inherent ID classes' connection. We find that the semantic information across different ID classes is beneficial for effective OOD detection. We thus investigate the ability of image-text comprehension among different semantic-related ID labels in VLMs and propose a novel post-hoc strategy called SimLabel. SimLabel enhances the separability between ID and OOD samples by establishing a more robust image-class similarity metric that considers consistency over a set of similar class labels. Extensive experiments demonstrate the superior performance of SimLabel on various zero-shot OOD detection benchmarks. The proposed model is also extended to various VLM-backbones, demonstrating its good generalization ability. Our demonstration and implementation codes are available at: https://github.com/ShuZou-1/SimLabel.

Paperid: 1452, https://arxiv.org/pdf/2501.11403.pdf GitHub

Abstract:
The web has become a crucial source of information, but it is also used to spread disinformation, often conveyed through multiple modalities like images and text. The identification of inconsistent cross-modal information, in particular entities such as persons, locations, and events, is critical to detect disinformation. Previous works either identify out-of-context disinformation by assessing the consistency of images to the whole document, neglecting relations of individual entities, or focus on generic entities that are not relevant to news. So far, only few approaches have addressed the task of validating entity consistency between images and text in news. However, the potential of large vision-language models (LVLMs) has not been explored yet. In this paper, we propose an LVLM-based framework for verifying Cross-modal Entity Consistency~(LVLM4CEC), to assess whether persons, locations and events in news articles are consistent across both modalities. We suggest effective prompting strategies for LVLMs for entity verification that leverage reference images crawled from web. Moreover, we extend three existing datasets for the task of entity verification in news providing manual ground-truth data. Our results show the potential of LVLMs for automating cross-modal entity verification, showing improved accuracy in identifying persons and events when using evidence images. Moreover, our method outperforms a baseline for location and event verification in documents. The datasets and source code are available on GitHub at https://github.com/TIBHannover/LVLM4CEC.

Paperid: 1453, https://arxiv.org/pdf/2501.11175.pdf GitHub

Abstract:
The growing popularity of Contrastive Language-Image Pretraining (CLIP) has led to its widespread application in various visual downstream tasks. To enhance CLIP's effectiveness and versatility, efficient few-shot adaptation techniques have been widely adopted. Among these approaches, training-free methods, particularly caching methods exemplified by Tip-Adapter, have gained attention for their lightweight adaptation without the need for additional fine-tuning. In this paper, we revisit Tip-Adapter from a kernel perspective, showing that caching methods function as local adapters and are connected to a well-established kernel literature. Drawing on this insight, we offer a theoretical understanding of how these methods operate and suggest multiple avenues for enhancing the Tip-Adapter baseline. Notably, our analysis shows the importance of incorporating global information in local adapters. Therefore, we subsequently propose a global method that learns a proximal regularizer in a reproducing kernel Hilbert space (RKHS) using CLIP as a base learner. Our method, which we call ProKeR (Proximal Kernel ridge Regression), has a closed form solution and achieves state-of-the-art performances across 11 datasets in the standard few-shot adaptation benchmark.

Paperid: 1454, https://arxiv.org/pdf/2501.10967.pdf GitHub

Abstract:
Vision-language Models (VLMs) have shown remarkable capabilities in advancing general artificial intelligence, yet the irrational encoding of visual positions persists in inhibiting the models' comprehensive perception performance across different levels of granularity. In this work, we propose Pyramid-descent Visual Position Encoding (PyPE), a novel approach designed to enhance the perception of visual tokens within VLMs. By assigning visual position indexes from the periphery to the center and expanding the central receptive field incrementally, PyPE addresses the limitations of traditional raster-scan methods and mitigates the long-term decay effects induced by Rotary Position Embedding (RoPE). Our method reduces the relative distance between interrelated visual elements and instruction tokens, promoting a more rational allocation of attention weights and allowing for a multi-granularity perception of visual elements and countering the over-reliance on anchor tokens. Extensive experimental evaluations demonstrate that PyPE consistently improves the general capabilities of VLMs across various sizes. Code is available at https://github.com/SakuraTroyChen/PyPE.

Paperid: 1455, https://arxiv.org/pdf/2501.10775.pdf GitHub

Abstract:
Medical vision-language pretraining (VLP) that leverages naturally-paired medical image-report data is crucial for medical image analysis. However, existing methods struggle to accurately characterize associations between images and diseases, leading to inaccurate or incomplete diagnostic results. In this work, we propose MedFILIP, a fine-grained VLP model, introduces medical image-specific knowledge through contrastive learning, specifically: 1) An information extractor based on a large language model is proposed to decouple comprehensive disease details from reports, which excels in extracting disease deals through flexible prompt engineering, thereby effectively reducing text complexity while retaining rich information at a tiny cost. 2) A knowledge injector is proposed to construct relationships between categories and visual attributes, which help the model to make judgments based on image features, and fosters knowledge extrapolation to unfamiliar disease categories. 3) A semantic similarity matrix based on fine-grained annotations is proposed, providing smoother, information-richer labels, thus allowing fine-grained image-text alignment. 4) We validate MedFILIP on numerous datasets, e.g., RSNA-Pneumonia, NIH ChestX-ray14, VinBigData, and COVID-19. For single-label, multi-label, and fine-grained classification, our model achieves state-of-the-art performance, the classification accuracy has increased by a maximum of 6.69\%. The code is available in https://github.com/PerceptionComputingLab/MedFILIP.

Paperid: 1456, https://arxiv.org/pdf/2501.10318.pdf GitHub

Abstract:
Benefiting from recent advancements in large language models and modality alignment techniques, existing Large Vision-Language Models(LVLMs) have achieved prominent performance across a wide range of scenarios. However, the excessive computational complexity limits the widespread use of these models in practical applications. We argue that one main bottleneck in computational complexity is caused by the involvement of redundant vision sequences in model computation. This is inspired by a reassessment of the efficiency of vision and language information transmission in the language decoder of LVLMs. Then, we propose a novel hierarchical vision-language interaction mechanism called Hierarchical Vision injection for Mixture Attention (HiMix). In HiMix, only the language sequence undergoes full forward propagation, while the vision sequence interacts with the language at specific stages within each language decoder layer. It is striking that our approach significantly reduces computational complexity with minimal performance loss. Specifically, HiMix achieves a 10x reduction in the computational cost of the language decoder across multiple LVLM models while maintaining comparable performance. This highlights the advantages of our method, and we hope our research brings new perspectives to the field of vision-language understanding. Project Page: https://xuange923.github.io/HiMix

Paperid: 1457, https://arxiv.org/pdf/2501.09695.pdf GitHub

Abstract:
Hallucination remains a major challenge for Large Vision-Language Models (LVLMs). Direct Preference Optimization (DPO) has gained increasing attention as a simple solution to hallucination issues. It directly learns from constructed preference pairs that reflect the severity of hallucinations in responses to the same prompt and image. Nonetheless, different data construction methods in existing works bring notable performance variations. We identify a crucial factor here: outcomes are largely contingent on whether the constructed data aligns on-policy w.r.t the initial (reference) policy of DPO. Theoretical analysis suggests that learning from off-policy data is impeded by the presence of KL-divergence between the updated policy and the reference policy. From the perspective of dataset distribution, we systematically summarize the inherent flaws in existing algorithms that employ DPO to address hallucination issues. To alleviate the problems, we propose On-Policy Alignment (OPA)-DPO framework, which uniquely leverages expert feedback to correct hallucinated responses and aligns both the original and expert-revised responses in an on-policy manner. Notably, with only 4.8k data, OPA-DPO achieves an additional reduction in the hallucination rate of LLaVA-1.5-7B: 13.26% on the AMBER benchmark and 5.39% on the Object-Hal benchmark, compared to the previous SOTA algorithm trained with 16k samples. Our implementation is available at https://github.com/zhyang2226/OPA-DPO.

Paperid: 1458, https://arxiv.org/pdf/2501.09446.pdf GitHub

Abstract:
This paper investigates the robustness of vision-language models against adversarial visual perturbations and introduces a novel ``double visual defense" to enhance this robustness. Unlike previous approaches that resort to lightweight adversarial fine-tuning of a pre-trained CLIP model, we perform large-scale adversarial vision-language pre-training from scratch using web-scale data. We then strengthen the defense by incorporating adversarial visual instruction tuning. The resulting models from each stage, $Î$CLIP and $Î^2$LLaVA, show substantially enhanced zero-shot robustness and set a new state-of-the-art in adversarial defense for vision-language models. For example, the adversarial robustness of $Î$CLIP surpasses that of the previous best models on ImageNet-1k by ~20%. %For example, $Î$CLIP surpasses the previous best models on ImageNet-1k by ~20% in terms of adversarial robustness. Similarly, compared to prior art, $Î^2$LLaVA brings a ~30% robustness improvement to image captioning task and a ~20% robustness improvement to visual question answering task. Furthermore, our models exhibit stronger zero-shot recognition capability, fewer hallucinations, and superior reasoning performance compared to baselines. Our project page is https://doublevisualdefense.github.io/.

Paperid: 1459, https://arxiv.org/pdf/2501.09167.pdf GitHub

Abstract:
Vision Language Models (VLMs) demonstrate significant potential as embodied AI agents for various mobility applications. However, a standardized, closed-loop benchmark for evaluating their spatial reasoning and sequential decision-making capabilities is lacking. To address this, we present MetaVQA: a comprehensive benchmark designed to assess and enhance VLMs' understanding of spatial relationships and scene dynamics through Visual Question Answering (VQA) and closed-loop simulations. MetaVQA leverages Set-of-Mark prompting and top-down view ground-truth annotations from nuScenes and Waymo datasets to automatically generate extensive question-answer pairs based on diverse real-world traffic scenarios, ensuring object-centric and context-rich instructions. Our experiments show that fine-tuning VLMs with the MetaVQA dataset significantly improves their spatial reasoning and embodied scene comprehension in safety-critical simulations, evident not only in improved VQA accuracies but also in emerging safety-aware driving maneuvers. In addition, the learning demonstrates strong transferability from simulation to real-world observation. Code and data will be publicly available at https://metadriverse.github.io/metavqa .

Paperid: 1460, https://arxiv.org/pdf/2501.08861.pdf GitHub

Abstract:
Autonomous driving is a challenging task that requires perceiving and understanding the surrounding environment for safe trajectory planning. While existing vision-based end-to-end models have achieved promising results, these methods are still facing the challenges of vision understanding, decision reasoning and scene generalization. To solve these issues, a generative planning with 3D-vision language pre-training model named GPVL is proposed for end-to-end autonomous driving. The proposed paradigm has two significant aspects. On one hand, a 3D-vision language pre-training module is designed to bridge the gap between visual perception and linguistic understanding in the bird's eye view. On the other hand, a cross-modal language model is introduced to generate holistic driving decisions and fine-grained trajectories with perception and navigation information in an auto-regressive manner. Experiments on the challenging nuScenes dataset demonstrate that the proposed scheme achieves excellent performances compared with state-of-the-art methods. Besides, the proposed GPVL presents strong generalization ability and real-time potential when handling high-level commands in various scenarios. It is believed that the effective, robust and efficient performance of GPVL is crucial for the practical application of future autonomous driving systems. Code is available at https://github.com/ltp1995/GPVL

Paperid: 1461, https://arxiv.org/pdf/2501.08313.pdf GitHub GitHub

Authors:MiniMax, Aonian Li, Bangwei Gong, Bo Yang, Boji Shan, Chang Liu, Cheng Zhu, Chunhao Zhang, Congchao Guo, Da Chen, Dong Li, Enwei Jiao, Gengxin Li, Guojun Zhang, Haohai Sun, Houze Dong, Jiadai Zhu, Jiaqi Zhuang, Jiayuan Song, Jin Zhu, Jingtao Han, Jingyang Li, Junbin Xie, Junhao Xu, Junjie Yan, Kaishun Zhang, Kecheng Xiao, Kexi Kang, Le Han, Leyang Wang, Lianfei Yu, Liheng Feng, Lin Zheng, Linbo Chai, Long Xing, Meizhi Ju, Mingyuan Chi, Mozhi Zhang, Peikai Huang, Pengcheng Niu, Pengfei Li, Pengyu Zhao, Qi Yang, Qidi Xu, Qiexiang Wang, Qin Wang, Qiuhui Li, Ruitao Leng, Shengmin Shi, Shuqi Yu, Sichen Li, Songquan Zhu, Tao Huang, Tianrun Liang, Weigao Sun, Weixuan Sun, Weiyu Cheng, Wenkai Li, Xiangjun Song, Xiao Su, Xiaodong Han, Xinjie Zhang, Xinzhu Hou, Xu Min, Xun Zou, Xuyang Shen, Yan Gong, Yingjie Zhu, Yipeng Zhou, Yiran Zhong, Yongyi Hu, Yuanxiang Fan, Yue Yu, Yufeng Yang, Yuhao Li, Yunan Huang, Yunji Li, Yunpeng Huang, Yunzhi Xu, Yuxin Mao, Zehan Li, Zekang Li, Zewei Tao, Zewen Ying, Zhaoyang Cong, Zhen Qin, Zhenhua Fan, Zhihang Yu, Zhuo Jiang, Zijia Wu

Abstract:
We introduce MiniMax-01 series, including MiniMax-Text-01 and MiniMax-VL-01, which are comparable to top-tier models while offering superior capabilities in processing longer contexts. The core lies in lightning attention and its efficient scaling. To maximize computational capacity, we integrate it with Mixture of Experts (MoE), creating a model with 32 experts and 456 billion total parameters, of which 45.9 billion are activated for each token. We develop an optimized parallel strategy and highly efficient computation-communication overlap techniques for MoE and lightning attention. This approach enables us to conduct efficient training and inference on models with hundreds of billions of parameters across contexts spanning millions of tokens. The context window of MiniMax-Text-01 can reach up to 1 million tokens during training and extrapolate to 4 million tokens during inference at an affordable cost. Our vision-language model, MiniMax-VL-01 is built through continued training with 512 billion vision-language tokens. Experiments on both standard and in-house benchmarks show that our models match the performance of state-of-the-art models like GPT-4o and Claude-3.5-Sonnet while offering 20-32 times longer context window. We publicly release MiniMax-01 at https://github.com/MiniMax-AI.

Paperid: 1462, https://arxiv.org/pdf/2501.07554.pdf GitHub GitHub

Abstract:
Video editing models have advanced significantly, but evaluating their performance remains challenging. Traditional metrics, such as CLIP text and image scores, often fall short: text scores are limited by inadequate training data and hierarchical dependencies, while image scores fail to assess temporal consistency. We present SST-EM (Semantic, Spatial, and Temporal Evaluation Metric), a novel evaluation framework that leverages modern Vision-Language Models (VLMs), Object Detection, and Temporal Consistency checks. SST-EM comprises four components: (1) semantic extraction from frames using a VLM, (2) primary object tracking with Object Detection, (3) focused object refinement via an LLM agent, and (4) temporal consistency assessment using a Vision Transformer (ViT). These components are integrated into a unified metric with weights derived from human evaluations and regression analysis. The name SST-EM reflects its focus on Semantic, Spatial, and Temporal aspects of video evaluation. SST-EM provides a comprehensive evaluation of semantic fidelity and temporal smoothness in video editing. The source code is available in the \textbf{\href{https://github.com/custommetrics-sst/SST_CustomEvaluationMetrics.git}{GitHub Repository}}.

Paperid: 1463, https://arxiv.org/pdf/2501.07525.pdf GitHub

Abstract:
Automated chest radiographs interpretation requires both accurate disease classification and detailed radiology report generation, presenting a significant challenge in the clinical workflow. Current approaches either focus on classification accuracy at the expense of interpretability or generate detailed but potentially unreliable reports through image captioning techniques. In this study, we present RadAlign, a novel framework that combines the predictive accuracy of vision-language models (VLMs) with the reasoning capabilities of large language models (LLMs). Inspired by the radiologist's workflow, RadAlign first employs a specialized VLM to align visual features with key medical concepts, achieving superior disease classification with an average AUC of 0.885 across multiple diseases. These recognized medical conditions, represented as text-based concepts in the aligned visual-language space, are then used to prompt LLM-based report generation. Enhanced by a retrieval-augmented generation mechanism that grounds outputs in similar historical cases, RadAlign delivers superior report quality with a GREEN score of 0.678, outperforming state-of-the-art methods' 0.634. Our framework maintains strong clinical interpretability while reducing hallucinations, advancing automated medical imaging and report analysis through integrated predictive and generative AI. Code is available at https://github.com/difeigu/RadAlign.

Paperid: 1464, https://arxiv.org/pdf/2501.06553.pdf GitHub

Abstract:
Large Vision-Language Models (LVLMs) may produce outputs that are unfaithful to reality, also known as visual hallucinations (VH), which significantly impedes their real-world usage. To alleviate VH, various decoding strategies have been proposed to enhance visual information. However, many of these methods may require secondary decoding and rollback, which significantly reduces inference speed. In this work, we propose an efficient plug-and-play decoding algorithm via Visual-Aware Sparsification (VASparse) from the perspective of token sparsity for mitigating VH. VASparse is inspired by empirical observations: (1) the sparse activation of attention in LVLMs, and (2) visual-agnostic tokens sparsification exacerbates VH. Based on these insights, we propose a novel token sparsification strategy that balances efficiency and trustworthiness. Specifically, VASparse implements a visual-aware token selection strategy during decoding to reduce redundant tokens while preserving visual context effectively. Additionally, we innovatively introduce a sparse-based visual contrastive decoding method to recalibrate the distribution of hallucinated outputs without the time overhead associated with secondary decoding. Subsequently, VASparse recalibrates attention scores to penalize attention sinking of LVLMs towards text tokens. Extensive experiments across four popular benchmarks confirm the effectiveness of VASparse in mitigating VH across different LVLM families without requiring additional training or post-processing. Impressively, VASparse achieves state-of-the-art performance for mitigating VH while maintaining competitive decoding speed. Code is available at https://github.com/mengchuang123/VASparse-github.

Paperid: 1465, https://arxiv.org/pdf/2501.06173.pdf GitHub

Abstract:
Recent video generation models have shown promising results in producing high-quality video clips lasting several seconds. However, these models face challenges in generating long sequences that convey clear and informative events, limiting their ability to support coherent narrations. In this paper, we present a large-scale cooking video dataset designed to advance long-form narrative generation in the cooking domain. We validate the quality of our proposed dataset in terms of visual fidelity and textual caption accuracy using state-of-the-art Vision-Language Models (VLMs) and video generation models, respectively. We further introduce a Long Narrative Video Director to enhance both visual and semantic coherence in generated videos and emphasize the role of aligning visual embeddings to achieve improved overall video quality. Our method demonstrates substantial improvements in generating visually detailed and semantically aligned keyframes, supported by finetuning techniques that integrate text and image embeddings within the video generation process. Project page: https://videoauteur.github.io/

Paperid: 1466, https://arxiv.org/pdf/2501.06132.pdf GitHub

Abstract:
The increasing demand for flexible and efficient urban transportation solutions has spotlighted the limitations of traditional Demand Responsive Transport (DRT) systems, particularly in accommodating diverse passenger needs and dynamic urban environments. Autonomous Mobility-on-Demand (AMoD) systems have emerged as a promising alternative, leveraging connected and autonomous vehicles (CAVs) to provide responsive and adaptable services. However, existing methods primarily focus on either vehicle scheduling or path planning, which often simplify complex urban layouts and neglect the necessity for simultaneous coordination and mutual avoidance among CAVs. This oversimplification poses significant challenges to the deployment of AMoD systems in real-world scenarios. To address these gaps, we propose CoDriveVLM, a novel framework that integrates high-fidelity simultaneous dispatching and cooperative motion planning for future AMoD systems. Our method harnesses Vision-Language Models (VLMs) to enhance multi-modality information processing, and this enables comprehensive dispatching and collision risk evaluation. The VLM-enhanced CAV dispatching coordinator is introduced to effectively manage complex and unforeseen AMoD conditions, thus supporting efficient scheduling decision-making. Furthermore, we propose a scalable decentralized cooperative motion planning method via consensus alternating direction method of multipliers (ADMM) focusing on collision risk evaluation and decentralized trajectory optimization. Simulation results demonstrate the feasibility and robustness of CoDriveVLM in various traffic conditions, showcasing its potential to significantly improve the fidelity and effectiveness of AMoD systems in future urban transportation networks. The code is available at https://github.com/henryhcliu/CoDriveVLM.git.

Paperid: 1467, https://arxiv.org/pdf/2501.06031.pdf GitHub

Abstract:
Transductive zero-shot learning with vision-language models leverages image-image similarities within the dataset to achieve better classification accuracy compared to the inductive setting. However, there is little work that explores the structure of the language space in this context. We propose GTA-CLIP, a novel technique that incorporates supervision from language models for joint transduction in language and vision spaces. Our approach is iterative and consists of three steps: (i) incrementally exploring the attribute space by querying language models, (ii) an attribute-augmented transductive inference procedure, and (iii) fine-tuning the language and vision encoders based on inferred labels within the dataset. Through experiments with CLIP encoders, we demonstrate that GTA-CLIP, yields an average performance improvement of 8.6% and 3.7% across 12 datasets and 3 encoders, over CLIP and transductive CLIP respectively in the zero-shot setting. We also observe similar improvements in a few-shot setting. We present ablation studies that demonstrate the value of each step and visualize how the vision and language spaces evolve over iterations driven by the transductive learning.

Paperid: 1468, https://arxiv.org/pdf/2501.05901.pdf GitHub

Abstract:
Recently, vision-language models have made remarkable progress, demonstrating outstanding capabilities in various tasks such as image captioning and video understanding. We introduce Valley2, a novel multimodal large language model designed to enhance performance across all domains and extend the boundaries of practical applications in e-commerce and short video scenarios. Notably, Valley2 achieves state-of-the-art (SOTA) performance on e-commerce benchmarks, surpassing open-source models of similar size by a large margin (79.66 vs. 72.76). Additionally, Valley2 ranks second on the OpenCompass leaderboard among models with fewer than 10B parameters, with an impressive average score of 67.4. The code and model weights are open-sourced at https://github.com/bytedance/Valley.

Paperid: 1469, https://arxiv.org/pdf/2501.05728.pdf GitHub

Abstract:
Attribute classification is crucial for identifying specific characteristics within image regions. Vision-Language Models (VLMs) have been effective in zero-shot tasks by leveraging their general knowledge from large-scale datasets. Recent studies demonstrate that transformer-based models with class-wise queries can effectively address zero-shot multi-label classification. However, poor utilization of the relationship between seen and unseen attributes makes the model lack generalizability. Additionally, attribute classification generally involves many attributes, making maintaining the model's scalability difficult. To address these issues, we propose Super-class guided transFormer (SugaFormer), a novel framework that leverages super-classes to enhance scalability and generalizability for zero-shot attribute classification. SugaFormer employs Super-class Query Initialization (SQI) to reduce the number of queries, utilizing common semantic information from super-classes, and incorporates Multi-context Decoding (MD) to handle diverse visual cues. To strengthen generalizability, we introduce two knowledge transfer strategies that utilize VLMs. During training, Super-class guided Consistency Regularization (SCR) aligns model's features with VLMs using super-class guided prompts, and during inference, Zero-shot Retrieval-based Score Enhancement (ZRSE) refines predictions for unseen attributes. Extensive experiments demonstrate that SugaFormer achieves state-of-the-art performance across three widely-used attribute classification benchmarks under zero-shot, and cross-dataset transfer settings. Our code is available at https://github.com/mlvlab/SugaFormer.

Paperid: 1470, https://arxiv.org/pdf/2501.05711.pdf GitHub

Abstract:
Large Vision Language Models (LVLMs) have demonstrated impressive capabilities in video understanding, yet their adoption for Activities of Daily Living (ADL) remains limited by their inability to capture fine-grained interactions and spatial relationships. To address this, we aim to leverage the complementary nature of egocentric views to enhance LVLM's understanding of exocentric ADL videos. Consequently, we propose ego2exo knowledge distillation to learn ego-augmented exp representations. While effective, this approach requires paired ego-exo videos, which are impractical to collect at scale. To address this, we propose Skeleton-guided Synthetic Ego Generation (SK-EGO), which leverages human skeleton motion to generate synthetic ego views from exocentric videos. To enhance the ego representation of LVLMs trained on synthetic data, we develop a domain-agnostic bootstrapped ego2exo strategy that effectively transfers knowledge from real ego-exo pairs to synthetic ego-exo pairs, while mitigating domain misalignment. We find that the exo representations of our ego-augmented LVLMs successfully learn to extract ego-perspective cues, demonstrated through comprehensive evaluation on six ADL benchmarks and our proposed Ego-in-Exo PerceptionMCQ benchmark designed specifically to assess egocentric understanding from exocentric videos. Code, models, and data will be open-sourced at https://github.com/dominickrei/EgoExo4ADL.

Paperid: 1471, https://arxiv.org/pdf/2501.05179.pdf GitHub GitHub

Abstract:
Large vision-language models (LVLMs) excel at visual understanding, but face efficiency challenges due to quadratic complexity in processing long multi-modal contexts. While token compression can reduce computational costs, existing approaches are designed for single-view LVLMs and fail to consider the unique multi-view characteristics of high-resolution LVLMs with dynamic cropping. Existing methods treat all tokens uniformly, but our analysis reveals that global thumbnails can naturally guide the compression of local crops by providing holistic context for informativeness evaluation. In this paper, we first analyze dynamic cropping strategy, revealing both the complementary nature between thumbnails and crops, and the distinctive characteristics across different crops. Based on our observations, we propose "Global Compression Commander" (GlobalCom$^2$), a novel plug-and-play token compression framework for HR-LVLMs. GlobalCom$^2$ leverages thumbnail as the "commander" to guide the compression of local crops, adaptively preserving informative details while eliminating redundancy. Extensive experiments show that GlobalCom$^2$ maintains over 90% performance while compressing 90% visual tokens, reducing FLOPs and peak memory to 9.1% and 60%. Our code is available at https://github.com/xuyang-liu16/GlobalCom2.

Paperid: 1472, https://arxiv.org/pdf/2501.05031.pdf GitHub

Abstract:
The enhancement of generalization in robots by large vision-language models (LVLMs) is increasingly evident. Therefore, the embodied cognitive abilities of LVLMs based on egocentric videos are of great interest. However, current datasets for embodied video question answering lack comprehensive and systematic evaluation frameworks. Critical embodied cognitive issues, such as robotic self-cognition, dynamic scene perception, and hallucination, are rarely addressed. To tackle these challenges, we propose ECBench, a high-quality benchmark designed to systematically evaluate the embodied cognitive abilities of LVLMs. ECBench features a diverse range of scene video sources, open and varied question formats, and 30 dimensions of embodied cognition. To ensure quality, balance, and high visual dependence, ECBench uses class-independent meticulous human annotation and multi-round question screening strategies. Additionally, we introduce ECEval, a comprehensive evaluation system that ensures the fairness and rationality of the indicators. Utilizing ECBench, we conduct extensive evaluations of proprietary, open-source, and task-specific LVLMs. ECBench is pivotal in advancing the embodied cognitive capabilities of LVLMs, laying a solid foundation for developing reliable core models for embodied agents. All data and code are available at https://github.com/Rh-Dang/ECBench.

Paperid: 1473, https://arxiv.org/pdf/2501.04897.pdf GitHub

Abstract:
Online Continual Learning (OCL) is a critical area in machine learning, focusing on enabling models to adapt to evolving data streams in real-time while addressing challenges such as catastrophic forgetting and the stability-plasticity trade-off. This study conducts the first comprehensive Systematic Literature Review (SLR) on OCL, analyzing 81 approaches, extracting over 1,000 features (specific tasks addressed by these approaches), and identifying more than 500 components (sub-models within approaches, including algorithms and tools). We also review 83 datasets spanning applications like image classification, object detection, and multimodal vision-language tasks. Our findings highlight key challenges, including reducing computational overhead, developing domain-agnostic solutions, and improving scalability in resource-constrained environments. Furthermore, we identify promising directions for future research, such as leveraging self-supervised learning for multimodal and sequential data, designing adaptive memory mechanisms that integrate sparse retrieval and generative replay, and creating efficient frameworks for real-world applications with noisy or evolving task boundaries. By providing a rigorous and structured synthesis of the current state of OCL, this review offers a valuable resource for advancing this field and addressing its critical challenges and opportunities. The complete SLR methodology steps and extracted data are publicly available through the provided link: https://github.com/kiyan-rezaee/ Systematic-Literature-Review-on-Online-Continual-Learning

Paperid: 1474, https://arxiv.org/pdf/2501.04671.pdf GitHub

Abstract:
While chain-of-thought (CoT) prompting improves reasoning in large language models, its effectiveness in vision-language models (VLMs) remains limited due to over-reliance on textual cues and memorized knowledge. To investigate the visual reasoning capabilities of VLMs in complex real-world scenarios, we introduce DrivingVQA, a visual question answering dataset derived from driving theory exams, which contains 3,931 multiple-choice problems with expert-written explanations and grounded entities relevant to the reasoning process. Leveraging this dataset, we propose RIV-CoT, a Retrieval-Based Interleaved Visual Chain-of-Thought method that enables VLMs to reason using visual crops corresponding to these relevant entities. Our experiments demonstrate that RIV-CoT improves answer accuracy by 3.1% and reasoning accuracy by 4.6% over vanilla CoT prompting. Furthermore, we demonstrate that our method effectively scales to the larger A-OKVQA reasoning dataset by leveraging automatically generated pseudo-labels, outperforming CoT prompting.

Paperid: 1475, https://arxiv.org/pdf/2501.04352.pdf GitHub

Abstract:
Online test-time adaptation (OTTA) of vision-language models (VLMs) has recently garnered increased attention to take advantage of data observed along a stream to improve future predictions. Unfortunately, existing methods rely on dataset-specific hyperparameters, significantly limiting their adaptability to unseen tasks. In response, we propose Online Gaussian Adaptation (OGA), a novel method that models the likelihoods of visual features using Gaussian distributions and incorporates zero-shot priors into an interpretable Maximum A Posteriori (MAP) estimation framework with fixed hyper-parameters across all datasets. We demonstrate that OGA outperforms state-of-the-art methods on most datasets and runs. Additionally, we show that combining OTTA with popular few-shot techniques (a practical yet overlooked setting in prior research) is highly beneficial. Furthermore, our experimental study reveals that common OTTA evaluation protocols, which average performance over at most three runs per dataset, are inadequate due to the substantial variability observed across runs for all OTTA methods. Therefore, we advocate for more rigorous evaluation practices, including increasing the number of runs and considering additional quantitative metrics, such as our proposed Expected Tail Accuracy (ETA), calculated as the average accuracy in the worst 10% of runs. We hope these contributions will encourage more rigorous and diverse evaluation practices in the OTTA community. Code is available at https://github.com/cfuchs2023/OGA .

Paperid: 1476, https://arxiv.org/pdf/2501.04322.pdf GitHub

Abstract:
Multimodal vision language models (VLMs) have made significant progress with the support of continuously increasing model sizes and data volumes. Running VLMs on edge devices has become a challenge for their widespread application. There are several efficient VLM efforts, but they often sacrifice linguistic capabilities to enhance multimodal abilities, or require extensive training. To address this quandary,we introduce the innovative framework of Efficient Vision Language Models with Elastic Visual Experts (Eve). By strategically incorporating adaptable visual expertise at multiple stages of training, Eve strikes a balance between preserving linguistic abilities and augmenting multimodal capabilities. This balanced approach results in a versatile model with only 1.8B parameters that delivers significant improvements in both multimodal and linguistic tasks. Notably, in configurations below 3B parameters, Eve distinctly outperforms in language benchmarks and achieves state-of-the-art results 68.87% in VLM Benchmarks. Additionally, its multimodal accuracy outstrips that of the larger 7B LLaVA-1.5 model. Our code is available at https://github.com/rangmiao/Eve.

Paperid: 1477, https://arxiv.org/pdf/2501.04155.pdf GitHub

Abstract:
Vision-language models (VLMs) are highly effective but often underperform on specialized tasks; for example, Llava-1.5 struggles with chart and diagram understanding due to scarce task-specific training data. Existing training data, sourced from general-purpose datasets, fails to capture the nuanced details needed for these tasks. We introduce MM-Gen, a scalable method that generates task-specific, high-quality synthetic text for candidate images by leveraging stronger models. MM-Gen employs a three-stage targeted process: partitioning data into subgroups, generating targeted text based on task descriptions, and filtering out redundant and outlier data. Fine-tuning VLMs with data generated by MM-Gen leads to significant performance gains, including 29% on spatial reasoning and 15% on diagram understanding for Llava-1.5 (7B). Compared to human-curated caption data, MM-Gen achieves up to 1.6x better improvements for the original models, proving its effectiveness in enhancing task-specific VLM performance and bridging the gap between general-purpose datasets and specialized requirements. Code available at https://github.com/sjoshi804/MM-Gen.

Paperid: 1478, https://arxiv.org/pdf/2501.04001.pdf GitHub

Abstract:
This work presents Sa2VA, the first unified model for dense grounded understanding of both images and videos. Unlike existing multi-modal large language models, which are often limited to specific modalities and tasks, Sa2VA supports a wide range of image and video tasks, including referring segmentation and conversation, with minimal one-shot instruction tuning. Sa2VA combines SAM-2, a foundation video segmentation model, with LLaVA, an advanced vision-language model, and unifies text, image, and video into a shared LLM token space. Using the LLM, Sa2VA generates instruction tokens that guide SAM-2 in producing precise masks, enabling a grounded, multi-modal understanding of both static and dynamic visual content. Additionally, we introduce Ref-SAV, an auto-labeled dataset containing over 72k object expressions in complex video scenes, designed to boost model performance. We also manually validate 2k video objects in the Ref-SAV datasets to benchmark referring video object segmentation in complex environments. Experiments show that Sa2VA achieves state-of-the-art across multiple tasks, particularly in referring video object segmentation, highlighting its potential for complex real-world applications.

Paperid: 1479, https://arxiv.org/pdf/2501.03992.pdf GitHub

Abstract:
Vector graphics are essential in design, providing artists with a versatile medium for creating resolution-independent and highly editable visual content. Recent advancements in vision-language and diffusion models have fueled interest in text-to-vector graphics generation. However, existing approaches often suffer from over-parameterized outputs or treat the layered structure - a core feature of vector graphics - as a secondary goal, diminishing their practical use. Recognizing the importance of layered SVG representations, we propose NeuralSVG, an implicit neural representation for generating vector graphics from text prompts. Inspired by Neural Radiance Fields (NeRFs), NeuralSVG encodes the entire scene into the weights of a small MLP network, optimized using Score Distillation Sampling (SDS). To encourage a layered structure in the generated SVG, we introduce a dropout-based regularization technique that strengthens the standalone meaning of each shape. We additionally demonstrate that utilizing a neural representation provides an added benefit of inference-time control, enabling users to dynamically adapt the generated SVG based on user-provided inputs, all with a single learned representation. Through extensive qualitative and quantitative evaluations, we demonstrate that NeuralSVG outperforms existing methods in generating structured and flexible SVG.

Paperid: 1480, https://arxiv.org/pdf/2501.03729.pdf GitHub

Abstract:
The zero-shot capabilities of Vision-Language Models (VLMs) have been widely leveraged to improve predictive performance. However, previous works on transductive or test-time adaptation (TTA) often make strong assumptions about the data distribution, such as the presence of all classes. Our work challenges these favorable deployment scenarios, and introduces a more realistic evaluation framework, including: (i) a variable number of effective classes for adaptation within a single batch, and (ii) non-i.i.d. batches of test samples in online adaptation settings. We provide comprehensive evaluations, comparisons, and ablation studies that demonstrate how current transductive or TTA methods for VLMs systematically compromise the models' initial zero-shot robustness across various realistic scenarios, favoring performance gains under advantageous assumptions about the test samples' distributions. Furthermore, we introduce StatA, a versatile method that could handle a wide range of deployment scenarios, including those with a variable number of effective classes at test time. Our approach incorporates a novel regularization term designed specifically for VLMs, which acts as a statistical anchor preserving the initial text-encoder knowledge, particularly in low-data regimes. Code available at https://github.com/MaxZanella/StatA.

Paperid: 1481, https://arxiv.org/pdf/2501.02189.pdf GitHub

Abstract:
Multimodal Vision Language Models (VLMs) have emerged as a transformative topic at the intersection of computer vision and natural language processing, enabling machines to perceive and reason about the world through both visual and textual modalities. For example, models such as CLIP, Claude, and GPT-4V demonstrate strong reasoning and understanding abilities on visual and textual data and beat classical single modality vision models on zero-shot classification [93]. With their rapid advancements in research and growing popularity in various applications, we provide a comprehensive survey of VLMs. Specifically, we provide a systematic overview of VLMs in the following aspects: [1] model information of the major VLMs developed up to 2025; [2] the transition of VLM architectures and the newest VLM alignment methods; [3] summary and categorization of the popular benchmarks and evaluation metrics of VLMs; [4] the challenges and issues faced by current VLMs such as hallucination, alignment, fairness, and safety. Detailed collections including papers and model repository links are listed in https://github.com/zli12321/Vision-Language-Models-Overview.

Paperid: 1482, https://arxiv.org/pdf/2501.02029.pdf GitHub

Abstract:
With the integration of an additional modality, large vision-language models (LVLMs) exhibit greater vulnerability to safety risks (e.g., jailbreaking) compared to their language-only predecessors. Although recent studies have devoted considerable effort to the post-hoc alignment of LVLMs, the inner safety mechanisms remain largely unexplored. In this paper, we discover that internal activations of LVLMs during the first token generation can effectively identify malicious prompts across different attacks. This inherent safety perception is governed by sparse attention heads, which we term ``safety heads." Further analysis reveals that these heads act as specialized shields against malicious prompts; ablating them leads to higher attack success rates, while the model's utility remains unaffected. By locating these safety heads and concatenating their activations, we construct a straightforward but powerful malicious prompt detector that integrates seamlessly into the generation process with minimal extra inference overhead. Despite its simple structure of a logistic regression model, the detector surprisingly exhibits strong zero-shot generalization capabilities. Experiments across various prompt-based attacks confirm the effectiveness of leveraging safety heads to protect LVLMs. Code is available at \url{https://github.com/Ziwei-Zheng/SAHs}.

Paperid: 1483, https://arxiv.org/pdf/2501.01986.pdf GitHub

Abstract:
The increasing demand to process long and high-resolution videos significantly burdens Large Vision-Language Models (LVLMs) due to the enormous number of visual tokens. Existing token reduction methods primarily prune tokens based on importance metrics, such as cumulative attention scores. However, even important tokens may exhibit high redundancy caused by similarity among adjacent video frames and repetitive visual elements. To address this limitation, we propose FrameFusion, a novel token reduction approach integrating similarity-based merging with importance-based pruning. We conduct a thorough study on token similarity characteristics, revealing three key insights: (1) spatially corresponding visual tokens between adjacent frames have higher cosine similarities compared to other token pairs; (2) high token similarities prominently decrease in deeper model layers; and (3) token similarity rankings are highly consistent across different layers. Guided by these observations, FrameFusion computes token similarities exclusively between corresponding visual tokens from adjacent frames, applies token merging at initial successive layers followed by pruning in deeper layers, and adopts a cascaded merging strategy to further enhance efficiency. We evaluate FrameFusion comprehensively across six diverse LVLMs, ranging from 2B to 72B parameters, using five video benchmarks encompassing video retrieval, question-answering, and spatial-temporal understanding tasks. Experiments show that FrameFusion reduces visual tokens by 70%, achieving 1.6-3.6x end-to-end speedups, with an average performance impact of less than 3%. Our code is available at: https://github.com/thu-nics/FrameFusion.

Paperid: 1484, https://arxiv.org/pdf/2501.01957.pdf GitHub

Abstract:
Recent Multimodal Large Language Models (MLLMs) have typically focused on integrating visual and textual modalities, with less emphasis placed on the role of speech in enhancing interaction. However, speech plays a crucial role in multimodal dialogue systems, and implementing high-performance in both vision and speech tasks remains a significant challenge due to the fundamental modality differences. In this paper, we propose a carefully designed multi-stage training methodology that progressively trains LLM to understand both visual and speech information, ultimately enabling fluent vision and speech interaction. Our approach not only preserves strong vision-language capacity, but also enables efficient speech-to-speech dialogue capabilities without separate ASR and TTS modules, significantly accelerating multimodal end-to-end response speed. By comparing our method against state-of-the-art counterparts across benchmarks for image, video, and speech tasks, we demonstrate that our model is equipped with both strong visual and speech capabilities, making near real-time vision and speech interaction.

Paperid: 1485, https://arxiv.org/pdf/2501.01926.pdf GitHub

Abstract:
Large vision-language models (LVLMs) have shown remarkable capabilities in visual-language understanding for downstream multi-modal tasks. Despite their success, LVLMs still suffer from generating hallucinations in complex generation tasks, leading to inconsistencies between visual inputs and generated content. To address this issue, some approaches have introduced inference-time interventions, such as contrastive decoding and attention rectification, to reduce overreliance on language priors. However, these approaches overlook hallucinations stemming from spurious inter-modality correlations. In this paper, we propose an Inter-Modality Correlation Calibration Decoding (IMCCD) method to mitigate hallucinations in LVLMs in a training-free manner. In this method, we design a Cross-Modal Value-Enhanced Decoding(CMVED) module to alleviate hallucination by a novel contrastive decoding mechanism. During the estimation of distorted distribution, CMVED masks the value vectors associated with significant cross-modal attention weights, which address both uni-modality overreliance and misleading inter-modality correlations. Additionally, a Content-Driven Attention Refinement(CDAR) module refines cross-modal attention weights, guiding LVLMs to focus on important visual content. Experimental results on diverse hallucination benchmarks validate the superiority of our method over existing state-of-the-art techniques in reducing hallucinations in LVLM text generation. Our code will be available at https://github.com/lijm48/IMCCD.

Paperid: 1486, https://arxiv.org/pdf/2501.01767.pdf GitHub

Abstract:
Logical image understanding involves interpreting and reasoning about the relationships and consistency within an image's visual content. This capability is essential in applications such as industrial inspection, where logical anomaly detection is critical for maintaining high-quality standards and minimizing costly recalls. Previous research in anomaly detection (AD) has relied on prior knowledge for designing algorithms, which often requires extensive manual annotations, significant computing power, and large amounts of data for training. Autoregressive, multimodal Vision Language Models (AVLMs) offer a promising alternative due to their exceptional performance in visual reasoning across various domains. Despite this, their application to logical AD remains unexplored. In this work, we investigate using AVLMs for logical AD and demonstrate that they are well-suited to the task. Combining AVLMs with format embedding and a logic reasoner, we achieve SOTA performance on public benchmarks, MVTec LOCO AD, with an AUROC of 86.0% and F1-max of 83.7%, along with explanations of anomalies. This significantly outperforms the existing SOTA method by a large margin.

Paperid: 1487, https://arxiv.org/pdf/2501.01709.pdf GitHub

Abstract:
Visual encoders are fundamental components in vision-language models (VLMs), each showcasing unique strengths derived from various pre-trained visual foundation models. To leverage the various capabilities of these encoders, recent studies incorporate multiple encoders within a single VLM, leading to a considerable increase in computational cost. In this paper, we present Mixture-of-Visual-Encoder Knowledge Distillation (MoVE-KD), a novel framework that distills the unique proficiencies of multiple vision encoders into a single, efficient encoder model. Specifically, to mitigate conflicts and retain the unique characteristics of each teacher encoder, we employ low-rank adaptation (LoRA) and mixture-of-experts (MoEs) to selectively activate specialized knowledge based on input features, enhancing both adaptability and efficiency. To regularize the KD process and enhance performance, we propose an attention-based distillation strategy that adaptively weighs the different encoders and emphasizes valuable visual tokens, reducing the burden of replicating comprehensive but distinct features from multiple teachers. Comprehensive experiments on popular VLMs, such as LLaVA and LLaVA-NeXT, validate the effectiveness of our method. Our code is available at: https://github.com/hey-cjj/MoVE-KD.

Paperid: 1488, https://arxiv.org/pdf/2501.00958.pdf GitHub

Abstract:
Compared to image-text pair data, interleaved corpora enable Vision-Language Models (VLMs) to understand the world more naturally like humans. However, such existing datasets are crawled from webpage, facing challenges like low knowledge density, loose image-text relations, and poor logical coherence between images. On the other hand, the internet hosts vast instructional videos (e.g., online geometry courses) that are widely used by humans to learn foundational subjects, yet these valuable resources remain underexplored in VLM training. In this paper, we introduce a high-quality \textbf{multimodal textbook} corpus with richer foundational knowledge for VLM pretraining. It collects over 2.5 years of instructional videos, totaling 22,000 class hours. We first use an LLM-proposed taxonomy to systematically gather instructional videos. Then we progressively extract and refine visual (keyframes), audio (ASR), and textual knowledge (OCR) from the videos, and organize as an image-text interleaved corpus based on temporal order. Compared to its counterparts, our video-centric textbook offers more coherent context, richer knowledge, and better image-text alignment. Experiments demonstrate its superb pretraining performance, particularly in knowledge- and reasoning-intensive tasks like ScienceQA and MathVista. Moreover, VLMs pre-trained on our textbook exhibit outstanding interleaved context awareness, leveraging visual and textual cues in their few-shot context for task solving. Our code are available at https://github.com/DAMO-NLP-SG/multimodal_textbook.

Paperid: 1489, https://arxiv.org/pdf/2412.21080.pdf GitHub

Abstract:
We introduce Vinci, a real-time embodied smart assistant built upon an egocentric vision-language model. Designed for deployment on portable devices such as smartphones and wearable cameras, Vinci operates in an "always on" mode, continuously observing the environment to deliver seamless interaction and assistance. Users can wake up the system and engage in natural conversations to ask questions or seek assistance, with responses delivered through audio for hands-free convenience. With its ability to process long video streams in real-time, Vinci can answer user queries about current observations and historical context while also providing task planning based on past interactions. To further enhance usability, Vinci integrates a video generation module that creates step-by-step visual demonstrations for tasks that require detailed guidance. We hope that Vinci can establish a robust framework for portable, real-time egocentric AI systems, empowering users with contextual and actionable insights. We release the complete implementation for the development of the device in conjunction with a demo web platform to test uploaded videos at https://github.com/OpenGVLab/vinci.

Paperid: 1490, https://arxiv.org/pdf/2412.20916.pdf GitHub

Abstract:
Although significant progress has been made in enhancing visibility, retrieving texture details, and mitigating noise in Low-Light (LL) images, the challenge persists in applying current Low-Light Image Enhancement (LLIE) methods to real-world scenarios, primarily due to the diverse illumination conditions encountered. Furthermore, the quest for generating enhancements that are visually realistic and attractive remains an underexplored realm. In response to these challenges, we introduce a novel \textbf{LLIE} framework with the guidance of \textbf{G}enerative \textbf{P}erceptual \textbf{P}riors (\textbf{GPP-LLIE}) derived from vision-language models (VLMs). Specifically, we first propose a pipeline that guides VLMs to assess multiple visual attributes of the LL image and quantify the assessment to output the global and local perceptual priors. Subsequently, to incorporate these generative perceptual priors to benefit LLIE, we introduce a transformer-based backbone in the diffusion process, and develop a new layer normalization (\textit{\textbf{GPP-LN}}) and an attention mechanism (\textit{\textbf{LPP-Attn}}) guided by global and local perceptual priors. Extensive experiments demonstrate that our model outperforms current SOTA methods on paired LL datasets and exhibits superior generalization on real-world data. The code is released at \url{https://github.com/LowLevelAI/GPP-LLIE}.

Paperid: 1491, https://arxiv.org/pdf/2412.19794.pdf GitHub

Abstract:
Multimodal Large Language Models (MLLMs), are recent advancement of Vision-Language Models (VLMs) that have driven major advances in video understanding. However, their vulnerability to adversarial tampering and manipulations remains underexplored. To address this gap, we introduce \textbf{MVTamperBench}, a benchmark that systematically evaluates MLLM robustness against five prevalent tampering techniques: rotation, masking, substitution, repetition, and dropping; based on real-world visual tampering scenarios such as surveillance interference, social media content edits, and misinformation injection. MVTamperBench comprises ~3.4K original videos, expanded into over ~17K tampered clips covering 19 distinct video manipulation tasks. This benchmark challenges models to detect manipulations in spatial and temporal coherence. We evaluate 45 recent MLLMs from 15+ model families. We reveal substantial variability in resilience across tampering types and show that larger parameter counts do not necessarily guarantee robustness. MVTamperBench sets a new benchmark for developing tamper-resilient MLLM in safety-critical applications, including detecting clickbait, preventing harmful content distribution, and enforcing policies on media platforms. We release all code, data, and benchmark to foster open research in trustworthy video understanding. Code: https://amitbcp.github.io/MVTamperBench/ Data: https://huggingface.co/datasets/Srikant86/MVTamperBench

Paperid: 1492, https://arxiv.org/pdf/2412.19723.pdf GitHub

Abstract:
Graphical User Interface (GUI) agents powered by Vision-Language Models (VLMs) have demonstrated human-like computer control capability. Despite their utility in advancing digital automation, a critical bottleneck persists: collecting high-quality trajectory data for training. Common practices for collecting such data rely on human supervision or synthetic data generation through executing pre-defined tasks, which are either resource-intensive or unable to guarantee data quality. Moreover, these methods suffer from limited data diversity and significant gaps between synthetic data and real-world environments. To address these challenges, we propose OS-Genesis, a novel GUI data synthesis pipeline that reverses the conventional trajectory collection process. Instead of relying on pre-defined tasks, OS-Genesis enables agents first to perceive environments and perform step-wise interactions, then retrospectively derive high-quality tasks to enable trajectory-level exploration. A trajectory reward model is then employed to ensure the quality of the generated trajectories. We demonstrate that training GUI agents with OS-Genesis significantly improves their performance on highly challenging online benchmarks. In-depth analysis further validates OS-Genesis's efficiency and its superior data quality and diversity compared to existing synthesis methods. Our codes, data, and checkpoints are available at https://qiushisun.github.io/OS-Genesis-Home/.

Paperid: 1493, https://arxiv.org/pdf/2412.19648.pdf GitHub

Abstract:
Vision-Language Tracking (VLT) aims to localize a target in video sequences using a visual template and language description. While textual cues enhance tracking potential, current datasets typically contain much more image data than text, limiting the ability of VLT methods to align the two modalities effectively. To address this imbalance, we propose a novel plug-and-play method named CTVLT that leverages the strong text-image alignment capabilities of foundation grounding models. CTVLT converts textual cues into interpretable visual heatmaps, which are easier for trackers to process. Specifically, we design a textual cue mapping module that transforms textual cues into target distribution heatmaps, visually representing the location described by the text. Additionally, the heatmap guidance module fuses these heatmaps with the search image to guide tracking more effectively. Extensive experiments on mainstream benchmarks demonstrate the effectiveness of our approach, achieving state-of-the-art performance and validating the utility of our method for enhanced VLT.

Paperid: 1494, https://arxiv.org/pdf/2412.19509.pdf GitHub

Abstract:
Vision-Language Models (VLMs) have enabled a variety of real-world applications. The large parameter size of VLMs brings large memory and computation overhead which poses significant challenges for deployment. Post-Training Quantization (PTQ) is an effective technique to reduce the memory and computation overhead. Existing PTQ methods mainly focus on large language models (LLMs), without considering the differences across other modalities. In this paper, we discover that there is a significant difference in sensitivity between language and vision tokens in large VLMs. Therefore, treating tokens from different modalities equally, as in existing PTQ methods, may over-emphasize the insensitive modalities, leading to significant accuracy loss. To deal with the above issue, we propose a simple yet effective method, Modality-Balanced Quantization (MBQ), for large VLMs. Specifically, MBQ incorporates the different sensitivities across modalities during the calibration process to minimize the reconstruction loss for better quantization parameters. Extensive experiments show that MBQ can significantly improve task accuracy by up to 4.4% and 11.6% under W3 and W4A8 quantization for 7B to 70B VLMs, compared to SOTA baselines. Additionally, we implement a W3 GPU kernel that fuses the dequantization and GEMV operators, achieving a 1.4x speedup on LLaVA-onevision-7B on the RTX 4090. The code is available at https://github.com/thu-nics/MBQ.

Paperid: 1495, https://arxiv.org/pdf/2412.19492.pdf GitHub

Abstract:
Recently, deep learning based methods have revolutionized remote sensing image segmentation. However, these methods usually rely on a pre-defined semantic class set, thus needing additional image annotation and model training when adapting to new classes. More importantly, they are unable to segment arbitrary semantic classes. In this work, we introduce Open-Vocabulary Remote Sensing Image Semantic Segmentation (OVRSISS), which aims to segment arbitrary semantic classes in remote sensing images. To address the lack of OVRSISS datasets, we develop LandDiscover50K, a comprehensive dataset of 51,846 images covering 40 diverse semantic classes. In addition, we propose a novel framework named GSNet that integrates domain priors from special remote sensing models and versatile capabilities of general vision-language models. Technically, GSNet consists of a Dual-Stream Image Encoder (DSIE), a Query-Guided Feature Fusion (QGFF), and a Residual Information Preservation Decoder (RIPD). DSIE first captures comprehensive features from both special models and general models in dual streams. Then, with the guidance of variable vocabularies, QGFF integrates specialist and generalist features, enabling them to complement each other. Finally, RIPD is proposed to aggregate multi-source features for more accurate mask predictions. Experiments show that our method outperforms other methods by a large margin, and our proposed LandDiscover50K improves the performance of OVRSISS methods. The proposed dataset and method will be made publicly available at https://github.com/yecy749/GSNet.

Paperid: 1496, https://arxiv.org/pdf/2412.19331.pdf GitHub

Abstract:
Recent advances in Large Vision-Language Models (LVLMs) have enabled general-purpose vision tasks through visual instruction tuning. While existing LVLMs can generate segmentation masks from text prompts for single images, they struggle with segmentation-grounded reasoning across images, especially at finer granularities such as object parts. In this paper, we introduce the new task of part-focused semantic co-segmentation, which involves identifying and segmenting common objects, as well as common and unique object parts across images. To address this task, we present CALICO, the first LVLM designed for multi-image part-level reasoning segmentation. CALICO features two key components, a novel Correspondence Extraction Module that identifies semantic part-level correspondences, and Correspondence Adaptation Modules that embed this information into the LVLM to facilitate multi-image understanding in a parameter-efficient manner. To support training and evaluation, we curate MixedParts, a large-scale multi-image segmentation dataset containing $\sim$2.4M samples across $\sim$44K images spanning diverse object and part categories. Experimental results demonstrate that CALICO, with just 0.3% of its parameters finetuned, achieves strong performance on this challenging task.

Paperid: 1497, https://arxiv.org/pdf/2412.19178.pdf GitHub

Abstract:
Cross-modal (e.g. image-text, video-text) retrieval is an important task in information retrieval and multimodal vision-language understanding field. Temporal understanding makes video-text retrieval more challenging than image-text retrieval. However, we find that the widely used video-text benchmarks have shortcomings in comprehensively assessing abilities of models, especially in temporal understanding, causing large-scale image-text pre-trained models can already achieve comparable zero-shot performance with video-text pre-trained models. In this paper, we introduce RTime, a novel temporal-emphasized video-text retrieval dataset. We first obtain videos of actions or events with significant temporality, and then reverse these videos to create harder negative samples. We then recruit annotators to judge the significance and reversibility of candidate videos, and write captions for qualified videos. We further adopt GPT-4 to extend more captions based on human-written captions. Our RTime dataset currently consists of 21k videos with 10 captions per video, totalling about 122 hours. Based on RTime, we propose three retrieval benchmark tasks: RTime-Origin, RTime-Hard, and RTime-Binary. We further enhance the use of harder-negatives in model training, and benchmark a variety of video-text models on RTime. Extensive experiment analysis proves that RTime indeed poses new and higher challenges to video-text retrieval. We release our RTime dataset\footnote{\url{https://github.com/qyr0403/Reversed-in-Time}} to further advance video-text retrieval and multimodal understanding research.

Paperid: 1498, https://arxiv.org/pdf/2412.19153.pdf GitHub

Abstract:
To use assistive robots in everyday life, a remote control system with common devices, such as 2D devices, is helpful to control the robots anytime and anywhere as intended. Hand-drawn sketches are one of the intuitive ways to control robots with 2D devices. However, since similar sketches have different intentions from scene to scene, existing work needs additional modalities to set the sketches' semantics. This requires complex operations for users and leads to decreasing usability. In this paper, we propose Sketch-MoMa, a teleoperation system using the user-given hand-drawn sketches as instructions to control a robot. We use Vision-Language Models (VLMs) to understand the user-given sketches superimposed on an observation image and infer drawn shapes and low-level tasks of the robot. We utilize the sketches and the generated shapes for recognition and motion planning of the generated low-level tasks for precise and intuitive operations. We validate our approach using state-of-the-art VLMs with 7 tasks and 5 sketch shapes. We also demonstrate that our approach effectively specifies the detailed motions, such as how to grasp and how much to rotate. Moreover, we show the competitive usability of our approach compared with the existing 2D interface through a user experiment with 14 participants.

Paperid: 1499, https://arxiv.org/pdf/2412.19048.pdf GitHub

Abstract:
A crucial component in many deep learning applications, such as Frequently Asked Questions (FAQ) and Retrieval-Augmented Generation (RAG), is dense retrieval. In this process, embedding models transform raw text into numerical vectors. However, the embedding models that currently excel on text embedding benchmarks, like the Massive Text Embedding Benchmark (MTEB), often have numerous parameters and high vector dimensionality. This poses challenges for their application in real-world scenarios. To address this issue, we propose a novel multi-stage distillation framework that enables a smaller student embedding model to distill multiple larger teacher embedding models through three carefully designed losses. Meanwhile, we utilize Matryoshka Representation Learning (MRL) to reduce the vector dimensionality of the student embedding model effectively. Our student model named Jasper with 2 billion parameters, built upon the Stella embedding model, obtained the No.3 position on the MTEB leaderboard (as of December 24, 2024), achieving an average 71.54 score across 56 datasets. We have released the model and data on the Hugging Face Hub (https://huggingface.co/infgrad/jasper_en_vision_language_v1) (https://huggingface.co/datasets/infgrad/jasper_text_distill_dataset), and the training codes are available in this project repository (https://github.com/NLPJCL/RAG-Retrieval).

Paperid: 1500, https://arxiv.org/pdf/2412.19021.pdf GitHub

Abstract:
Open-vocabulary Scene Graph Generation (OV-SGG) overcomes the limitations of the closed-set assumption by aligning visual relationship representations with open-vocabulary textual representations. This enables the identification of novel visual relationships, making it applicable to real-world scenarios with diverse relationships. However, existing OV-SGG methods are constrained by fixed text representations, limiting diversity and accuracy in image-text alignment. To address these challenges, we propose the Relation-Aware Hierarchical Prompting (RAHP) framework, which enhances text representation by integrating subject-object and region-specific relation information. Our approach utilizes entity clustering to address the complexity of relation triplet categories, enabling the effective integration of subject-object information. Additionally, we utilize a large language model (LLM) to generate detailed region-aware prompts, capturing fine-grained visual interactions and improving alignment between visual and textual modalities. RAHP also introduces a dynamic selection mechanism within Vision-Language Models (VLMs), which adaptively selects relevant text prompts based on the visual content, reducing noise from irrelevant prompts. Extensive experiments on the Visual Genome and Open Images v6 datasets demonstrate that our framework consistently achieves state-of-the-art performance, demonstrating its effectiveness in addressing the challenges of open-vocabulary scene graph generation. The code is available at: https://github.com/Leon022/RAHP

Paperid: 1501, https://arxiv.org/pdf/2412.18450.pdf GitHub

Abstract:
A 3D scene graph represents a compact scene model by capturing both the objects present and the semantic relationships between them, making it a promising structure for robotic applications. To effectively interact with users, an embodied intelligent agent should be able to answer a wide range of natural language queries about the surrounding 3D environment. Large Language Models (LLMs) are beneficial solutions for user-robot interaction due to their natural language understanding and reasoning abilities. Recent methods for learning scene representations have shown that adapting these representations to the 3D world can significantly improve the quality of LLM responses. However, existing methods typically rely only on geometric information, such as object coordinates, and overlook the rich semantic relationships between objects. In this work, we propose 3DGraphLLM, a method for constructing a learnable representation of a 3D scene graph that explicitly incorporates semantic relationships. This representation is used as input to LLMs for performing 3D vision-language tasks. In our experiments on popular ScanRefer, Multi3DRefer, ScanQA, Sqa3D, and Scan2cap datasets, we demonstrate that our approach outperforms baselines that do not leverage semantic relationships between objects. The code is publicly available at https://github.com/CognitiveAISystems/3DGraphLLM.

Paperid: 1502, https://arxiv.org/pdf/2412.18404.pdf GitHub

Abstract:
Recent vision-language foundation models still frequently produce outputs misaligned with their inputs, evidenced by object hallucination in captioning and prompt misalignment in the text-to-image generation model. Recent studies have explored methods for identifying misaligned elements, aiming not only to enhance interpretability but also to improve model performance. However, current approaches primarily rely on large foundation models in a zero-shot manner or fine-tuned models with human annotations, which limits scalability due to significant computational costs. This work proposes a novel approach, dubbed CLIP4DM, for detecting dense misalignments from pre-trained CLIP, specifically focusing on pinpointing misaligned words between image and text. We carefully revamp the gradient-based attribution computation method, enabling negative gradient of individual text tokens to indicate misalignment. We also propose F-CLIPScore, which aggregates misaligned attributions with a global alignment score. We evaluate our method on various dense misalignment detection benchmarks, covering various image and text domains and misalignment types. Our method demonstrates state-of-the-art performance among zero-shot models and competitive performance with fine-tuned models while maintaining superior efficiency. Our qualitative examples show that our method has a unique strength to detect entity-level objects, intangible objects, and attributes that can not be easily detected for existing works. We conduct ablation studies and analyses to highlight the strengths and limitations of our approach. Our code is publicly available at https://github.com/naver-ai/CLIP4DM.

Paperid: 1503, https://arxiv.org/pdf/2412.18303.pdf GitHub

Abstract:
Vision-language models (VLMs) have revolutionized machine learning by leveraging large pre-trained models to tackle various downstream tasks. Although label, training, and data efficiency have improved, many state-of-the-art VLMs still require task-specific hyperparameter tuning and fail to fully exploit test samples. To overcome these challenges, we propose a graph-based approach for label-efficient adaptation and inference. Our method dynamically constructs a graph over text prompts, few-shot examples, and test samples, using label propagation for inference without task-specific tuning. Unlike existing zero-shot label propagation techniques, our approach requires no additional unlabeled support set and effectively leverages the test sample manifold through dynamic graph expansion. We further introduce a context-aware feature re-weighting mechanism to improve task adaptation accuracy. Additionally, our method supports efficient graph expansion, enabling real-time inductive inference. Extensive evaluations on downstream tasks, such as fine-grained categorization and out-of-distribution generalization, demonstrate the effectiveness of our approach. The source code is available at https://github.com/Yushu-Li/ECALP.

Paperid: 1504, https://arxiv.org/pdf/2412.18292.pdf GitHub

Abstract:
Understanding how humans cooperatively utilize semantic knowledge to explore unfamiliar environments and decide on navigation directions is critical for house service multi-robot systems. Previous methods primarily focused on single-robot centralized planning strategies, which severely limited exploration efficiency. Recent research has considered decentralized planning strategies for multiple robots, assigning separate planning models to each robot, but these approaches often overlook communication costs. In this work, we propose Multimodal Chain-of-Thought Co-Navigation (MCoCoNav), a modular approach that utilizes multimodal Chain-of-Thought to plan collaborative semantic navigation for multiple robots. MCoCoNav combines visual perception with Vision Language Models (VLMs) to evaluate exploration value through probabilistic scoring, thus reducing time costs and achieving stable outputs. Additionally, a global semantic map is used as a communication bridge, minimizing communication overhead while integrating observational results. Guided by scores that reflect exploration trends, robots utilize this map to assess whether to explore new frontier points or revisit history nodes. Experiments on HM3D_v0.2 and MP3D demonstrate the effectiveness of our approach. Our code is available at https://github.com/FrankZxShen/MCoCoNav.git.

Paperid: 1505, https://arxiv.org/pdf/2412.18072.pdf GitHub

Abstract:
With advances in foundational and vision-language models, and effective fine-tuning techniques, a large number of both general and special-purpose models have been developed for a variety of visual tasks. Despite the flexibility and accessibility of these models, no single model is able to handle all tasks and/or applications that may be envisioned by potential users. Recent approaches, such as visual programming and multimodal LLMs with integrated tools aim to tackle complex visual tasks, by way of program synthesis. However, such approaches overlook user constraints (e.g., performance / computational needs), produce test-time sample-specific solutions that are difficult to deploy, and, sometimes, require low-level instructions that maybe beyond the abilities of a naive user. To address these limitations, we introduce MMFactory, a universal framework that includes model and metrics routing components, acting like a solution search engine across various available models. Based on a task description and few sample input-output pairs and (optionally) resource and/or performance constraints, MMFactory can suggest a diverse pool of programmatic solutions by instantiating and combining visio-lingual tools from its model repository. In addition to synthesizing these solutions, MMFactory also proposes metrics and benchmarks performance / resource characteristics, allowing users to pick a solution that meets their unique design constraints. From the technical perspective, we also introduced a committee-based solution proposer that leverages multi-agent LLM conversation to generate executable, diverse, universal, and robust solutions for the user. Experimental results show that MMFactory outperforms existing methods by delivering state-of-the-art solutions tailored to user problem specifications. Project page is available at https://davidhalladay.github.io/mmfactory_demo.

Paperid: 1506, https://arxiv.org/pdf/2412.17810.pdf GitHub

Abstract:
The attention operator is arguably the key distinguishing factor of transformer architectures, which have demonstrated state-of-the-art performance on a variety of tasks. However, transformer attention operators often impose a significant computational burden, with the computational complexity scaling quadratically with the number of tokens. In this work, we propose a novel transformer attention operator whose computational complexity scales linearly with the number of tokens. We derive our network architecture by extending prior work which has shown that a transformer style architecture naturally arises by "white-box" architecture design, where each layer of the network is designed to implement an incremental optimization step of a maximal coding rate reduction objective (MCR$^2$). Specifically, we derive a novel variational form of the MCR$^2$ objective and show that the architecture that results from unrolled gradient descent of this variational objective leads to a new attention module called Token Statistics Self-Attention (TSSA). TSSA has linear computational and memory complexity and radically departs from the typical attention architecture that computes pairwise similarities between tokens. Experiments on vision, language, and long sequence tasks show that simply swapping TSSA for standard self-attention, which we refer to as the Token Statistics Transformer (ToST), achieves competitive performance with conventional transformers while being significantly more computationally efficient and interpretable. Our results also somewhat call into question the conventional wisdom that pairwise similarity style attention mechanisms are critical to the success of transformer architectures. Code will be available at https://github.com/RobinWu218/ToST.

Paperid: 1507, https://arxiv.org/pdf/2412.17800.pdf GitHub

Abstract:
Enabling models to recognize vast open-world categories has been a longstanding pursuit in object detection. By leveraging the generalization capabilities of vision-language models, current open-world detectors can recognize a broader range of vocabularies, despite being trained on limited categories. However, when the scale of the category vocabularies during training expands to a real-world level, previous classifiers aligned with coarse class names significantly reduce the recognition performance of these detectors. In this paper, we introduce Prova, a multi-modal prototype classifier for vast-vocabulary object detection. Prova extracts comprehensive multi-modal prototypes as initialization of alignment classifiers to tackle the vast-vocabulary object recognition failure problem. On V3Det, this simple method greatly enhances the performance among one-stage, two-stage, and DETR-based detectors with only additional projection layers in both supervised and open-vocabulary settings. In particular, Prova improves Faster R-CNN, FCOS, and DINO by 3.3, 6.2, and 2.9 AP respectively in the supervised setting of V3Det. For the open-vocabulary setting, Prova achieves a new state-of-the-art performance with 32.8 base AP and 11.0 novel AP, which is of 2.6 and 4.3 gain over the previous methods.

Paperid: 1508, https://arxiv.org/pdf/2412.17787.pdf GitHub

Abstract:
Recent Large Vision-Language Models (LVLMs) have shown promising reasoning capabilities on text-rich images from charts, tables, and documents. However, the abundant text within such images may increase the model's sensitivity to language. This raises the need to evaluate LVLM performance on cross-lingual text-rich visual inputs, where the language in the image differs from the language of the instructions. To address this, we introduce XT-VQA (Cross-Lingual Text-Rich Visual Question Answering), a benchmark designed to assess how LVLMs handle language inconsistency between image text and questions. XT-VQA integrates five existing text-rich VQA datasets and a newly collected dataset, XPaperQA, covering diverse scenarios that require faithful recognition and comprehension of visual information despite language inconsistency. Our evaluation of prominent LVLMs on XT-VQA reveals a significant drop in performance for cross-lingual scenarios, even for models with multilingual capabilities. A mutual information analysis suggests that this performance gap stems from cross-lingual questions failing to adequately activate relevant visual information. To mitigate this issue, we propose MVCL-MI (Maximization of Vision-Language Cross-Lingual Mutual Information), where a visual-text cross-lingual alignment is built by maximizing mutual information between the model's outputs and visual information. This is achieved by distilling knowledge from monolingual to cross-lingual settings through KL divergence minimization, where monolingual output logits serve as a teacher. Experimental results on the XT-VQA demonstrate that MVCL-MI effectively reduces the visual-text cross-lingual performance disparity while preserving the inherent capabilities of LVLMs, shedding new light on the potential practice for improving LVLMs. Codes are available at: https://github.com/Stardust-y/XTVQA.git

Paperid: 1509, https://arxiv.org/pdf/2412.17741.pdf GitHub GitHub

Abstract:
Current Large Multimodal Models (LMMs) empowered visual grounding typically rely on $\texttt{}$ tokens as a text prompt to jointly optimize the vision-language model (e.g., LLaVA) and the downstream task-specific model (e.g., SAM). However, we observe that little research has looked into how it works.In this work, we first visualize the similarity maps, which are obtained by computing the semantic similarity between the $\texttt{}$ token and the image token embeddings derived from the last hidden layer in both the LLaVA encoder and SAM decoder. Intriguingly, we have found that a striking consistency holds in terms of activation responses in the similarity map, which reveals that what the $\texttt{}$ token contributes to is semantic similarity within image-text pairs. Specifically, the $\texttt{}$ token, a placeholder expanded in text vocabulary, extensively queries among individual tokenized image patches to match the semantics of an object from text to the paired image, while the Large Language Models (LLMs) are being fine-tuned. Upon the above findings, we present READ, which facilitates LMMs' resilient $\textbf{REA}$soning capability of where to atten$\textbf{D}$ under the guidance of highly activated points borrowed from similarity maps. Remarkably, READ features an intuitive design, Similarity as Points module (SasP), which can be seamlessly applied to $\texttt{}$-like paradigms in a plug-and-play fashion. Also, extensive experiments have been conducted on ReasonSeg and RefCOCO(+/g) datasets. To validate whether READ suffers from catastrophic forgetting of previous skills after fine-tuning, we further assess its generation ability on an augmented FP-RefCOCO(+/g) dataset. All codes and models are publicly available at https://github.com/rui-qian/READ.

Paperid: 1510, https://arxiv.org/pdf/2412.17619.pdf GitHub

Abstract:
Few-shot anomaly detection (FSAD) aims to detect unseen anomaly regions with the guidance of very few normal support images from the same class. Existing FSAD methods usually find anomalies by directly designing complex text prompts to align them with visual features under the prevailing large vision-language model paradigm. However, these methods, almost always, neglect intrinsic contextual information in visual features, e.g., the interaction relationships between different vision layers, which is an important clue for detecting anomalies comprehensively. To this end, we propose a kernel-aware graph prompt learning framework, termed as KAG-prompt, by reasoning the cross-layer relations among visual features for FSAD. Specifically, a kernel-aware hierarchical graph is built by taking the different layer features focusing on anomalous regions of different sizes as nodes, meanwhile, the relationships between arbitrary pairs of nodes stand for the edges of the graph. By message passing over this graph, KAG-prompt can capture cross-layer contextual information, thus leading to more accurate anomaly prediction. Moreover, to integrate the information of multiple important anomaly signals in the prediction map, we propose a novel image-level scoring method based on multi-level information fusion. Extensive experiments on MVTecAD and VisA datasets show that KAG-prompt achieves state-of-the-art FSAD results for image-level/pixel-level anomaly detection. Code is available at https://github.com/CVL-hub/KAG-prompt.git.

Paperid: 1511, https://arxiv.org/pdf/2412.16211.pdf GitHub

Abstract:
The current state-of-the-art video generative models can produce commercial-grade videos with highly realistic details. However, they still struggle to coherently present multiple sequential events in the stories specified by the prompts, which is foreseeable an essential capability for future long video generation scenarios. For example, top T2V generative models still fail to generate a video of the short simple story 'how to put an elephant into a refrigerator.' While existing detail-oriented benchmarks primarily focus on fine-grained metrics like aesthetic quality and spatial-temporal consistency, they fall short of evaluating models' abilities to handle event-level story presentation. To address this gap, we introduce StoryEval, a story-oriented benchmark specifically designed to assess text-to-video (T2V) models' story-completion capabilities. StoryEval features 423 prompts spanning 7 classes, each representing short stories composed of 2-4 consecutive events. We employ advanced vision-language models, such as GPT-4V and LLaVA-OV-Chat-72B, to verify the completion of each event in the generated videos, applying a unanimous voting method to enhance reliability. Our methods ensure high alignment with human evaluations, and the evaluation of 11 models reveals its challenge, with none exceeding an average story-completion rate of 50%. StoryEval provides a new benchmark for advancing T2V models and highlights the challenges and opportunities in developing next-generation solutions for coherent story-driven video generation.

Paperid: 1512, https://arxiv.org/pdf/2412.15588.pdf GitHub

Abstract:
Compositional generalization is crucial for artificial intelligence agents to solve complex vision-language reasoning tasks. Neuro-symbolic approaches have demonstrated promise in capturing compositional structures, but they face critical challenges: (a) reliance on predefined predicates for symbolic representations that limit adaptability, (b) difficulty in extracting predicates from raw data, and (c) using non-differentiable operations for combining primitive concepts. To address these issues, we propose NeSyCoCo, a neuro-symbolic framework that leverages large language models (LLMs) to generate symbolic representations and map them to differentiable neural computations. NeSyCoCo introduces three innovations: (a) augmenting natural language inputs with dependency structures to enhance the alignment with symbolic representations, (b) employing distributed word representations to link diverse, linguistically motivated logical predicates to neural modules, and (c) using the soft composition of normalized predicate scores to align symbolic and differentiable reasoning. Our framework achieves state-of-the-art results on the ReaSCAN and CLEVR-CoGenT compositional generalization benchmarks and demonstrates robust performance with novel concepts in the CLEVR-SYN benchmark.

Paperid: 1513, https://arxiv.org/pdf/2412.15544.pdf GitHub

Abstract:
In recent years, reinforcement learning (RL)-based methods for learning driving policies have gained increasing attention in the autonomous driving community and have achieved remarkable progress in various driving scenarios. However, traditional RL approaches rely on manually engineered rewards, which require extensive human effort and often lack generalizability. To address these limitations, we propose \textbf{VLM-RL}, a unified framework that integrates pre-trained Vision-Language Models (VLMs) with RL to generate reward signals using image observation and natural language goals. The core of VLM-RL is the contrasting language goal (CLG)-as-reward paradigm, which uses positive and negative language goals to generate semantic rewards. We further introduce a hierarchical reward synthesis approach that combines CLG-based semantic rewards with vehicle state information, improving reward stability and offering a more comprehensive reward signal. Additionally, a batch-processing technique is employed to optimize computational efficiency during training. Extensive experiments in the CARLA simulator demonstrate that VLM-RL outperforms state-of-the-art baselines, achieving a 10.5\% reduction in collision rate, a 104.6\% increase in route completion rate, and robust generalization to unseen driving scenarios. Furthermore, VLM-RL can seamlessly integrate almost any standard RL algorithms, potentially revolutionizing the existing RL paradigm that relies on manual reward engineering and enabling continuous performance improvements. The demo video and code can be accessed at: https://zilin-huang.github.io/VLM-RL-website.

Paperid: 1514, https://arxiv.org/pdf/2412.15209.pdf GitHub

Abstract:
Despite significant advancements in Large Vision-Language Models (LVLMs), existing pixel-grounding models operate on single-image settings, limiting their ability to perform detailed, fine-grained comparisons across multiple images. Conversely, current multi-image understanding models lack pixel-level grounding. Our work addresses this gap by introducing the task of multi-image pixel-grounded reasoning segmentation, and PRIMA, a novel LVLM that integrates pixel-level grounding with robust multi-image reasoning capabilities to produce contextually rich, pixel-grounded explanations. Central to PRIMA is an efficient vision module that queries fine-grained visual representations across multiple images, reducing TFLOPs by $25.3\%$. To support training and evaluation, we curate $M^4Seg$, a new reasoning segmentation benchmark consisting of $\sim$224K question-answer pairs that require fine-grained visual understanding across multiple images. Experimental results demonstrate PRIMA outperforms state-of-the-art baselines.

Paperid: 1515, https://arxiv.org/pdf/2412.15206.pdf GitHub GitHub

Abstract:
Recent advancements in large vision language models (VLMs) tailored for autonomous driving (AD) have shown strong scene understanding and reasoning capabilities, making them undeniable candidates for end-to-end driving systems. However, limited work exists on studying the trustworthiness of DriveVLMs -- a critical factor that directly impacts public transportation safety. In this paper, we introduce AutoTrust, a comprehensive trustworthiness benchmark for large vision-language models in autonomous driving (DriveVLMs), considering diverse perspectives -- including trustfulness, safety, robustness, privacy, and fairness. We constructed the largest visual question-answering dataset for investigating trustworthiness issues in driving scenarios, comprising over 10k unique scenes and 18k queries. We evaluated six publicly available VLMs, spanning from generalist to specialist, from open-source to commercial models. Our exhaustive evaluations have unveiled previously undiscovered vulnerabilities of DriveVLMs to trustworthiness threats. Specifically, we found that the general VLMs like LLaVA-v1.6 and GPT-4o-mini surprisingly outperform specialized models fine-tuned for driving in terms of overall trustworthiness. DriveVLMs like DriveLM-Agent are particularly vulnerable to disclosing sensitive information. Additionally, both generalist and specialist VLMs remain susceptible to adversarial attacks and struggle to ensure unbiased decision-making across diverse environments and populations. Our findings call for immediate and decisive action to address the trustworthiness of DriveVLMs -- an issue of critical importance to public safety and the welfare of all citizens relying on autonomous transportation systems. Our benchmark is publicly available at \url{https://github.com/taco-group/AutoTrust}, and the leaderboard is released at \url{https://taco-group.github.io/AutoTrust/}.

Paperid: 1516, https://arxiv.org/pdf/2412.15190.pdf GitHub

Abstract:
Automated analysis of vast Earth observation data via interactive Vision-Language Models (VLMs) can unlock new opportunities for environmental monitoring, disaster response, and {resource management}. Existing generic VLMs do not perform well on Remote Sensing data, while the recent Geo-spatial VLMs remain restricted to a fixed resolution and few sensor modalities. In this paper, we introduce EarthDial, a conversational assistant specifically designed for Earth Observation (EO) data, transforming complex, multi-sensory Earth observations into interactive, natural language dialogues. EarthDial supports multi-spectral, multi-temporal, and multi-resolution imagery, enabling a wide range of remote sensing tasks, including classification, detection, captioning, question answering, visual reasoning, and visual grounding. To achieve this, we introduce an extensive instruction tuning dataset comprising over 11.11M instruction pairs covering RGB, Synthetic Aperture Radar (SAR), and multispectral modalities such as Near-Infrared (NIR) and infrared. Furthermore, EarthDial handles bi-temporal and multi-temporal sequence analysis for applications like change detection. Our extensive experimental results on 44 downstream datasets demonstrate that EarthDial outperforms existing generic and domain-specific models, achieving better generalization across various EO tasks. Our source codes and pre-trained models are at https://github.com/hiyamdebary/EarthDial.

Paperid: 1517, https://arxiv.org/pdf/2412.14680.pdf GitHub

Abstract:
Open-set object detection (OSOD) is highly desirable for robotic manipulation in unstructured environments. However, existing OSOD methods often fail to meet the requirements of robotic applications due to their high computational burden and complex deployment. To address this issue, this paper proposes a light-weight framework called Decoupled OSOD (DOSOD), which is a practical and highly efficient solution to support real-time OSOD tasks in robotic systems. Specifically, DOSOD builds upon the YOLO-World pipeline by integrating a vision-language model (VLM) with a detector. A Multilayer Perceptron (MLP) adaptor is developed to transform text embeddings extracted by the VLM into a joint space, within which the detector learns the region representations of class-agnostic proposals. Cross-modality features are directly aligned in the joint space, avoiding the complex feature interactions and thereby improving computational efficiency. DOSOD operates like a traditional closed-set detector during the testing phase, effectively bridging the gap between closed-set and open-set detection. Compared to the baseline YOLO-World, the proposed DOSOD significantly enhances real-time performance while maintaining comparable accuracy. The slight DOSOD-S model achieves a Fixed AP of $26.7\%$, compared to $26.2\%$ for YOLO-World-v1-S and $22.7\%$ for YOLO-World-v2-S, using similar backbones on the LVIS minival dataset. Meanwhile, the FPS of DOSOD-S is $57.1\%$ higher than YOLO-World-v1-S and $29.6\%$ higher than YOLO-World-v2-S. Meanwhile, we demonstrate that the DOSOD model facilitates the deployment of edge devices. The codes and models are publicly available at https://github.com/D-Robotics-AI-Lab/DOSOD.

Paperid: 1518, https://arxiv.org/pdf/2412.14672.pdf GitHub

Abstract:
Large Vision Language Models (LVLMs) have achieved significant progress in integrating visual and textual inputs for multimodal reasoning. However, a recurring challenge is ensuring these models utilize visual information as effectively as linguistic content when both modalities are necessary to formulate an accurate answer. We hypothesize that hallucinations arise due to the lack of effective visual grounding in current LVLMs. Furthermore, current vision-language benchmarks are not specifically measuring the degree to which the answer require the visual input. This limitation makes it challenging to confirm that the image is truly necessary, particularly in tasks like visual question answering. In this work, we introduce FiVL, a novel method for constructing datasets designed to train LVLMs for enhanced visual grounding and also evaluate their effectiveness in achieving it. We demonstrate the value of our datasets through three approaches. First, we introduce a novel training task based on our augmented training dataset, resulting in better performance than the baseline. Second, we present benchmarks to assess the model's ability to use image as substantive evidence, rather than relying solely on linguistic priors. Finally, we identify attention heads with the strongest vision-language alignment, enabling explainability on visual-driven hallucinations. The code is available at https://github.com/IntelLabs/fivl.

Paperid: 1519, https://arxiv.org/pdf/2412.14480.pdf GitHub

Abstract:
In Embodied Question Answering (EQA), agents must explore and develop a semantic understanding of an unseen environment to answer a situated question with confidence. This problem remains challenging in robotics, due to the difficulties in obtaining useful semantic representations, updating these representations online, and leveraging prior world knowledge for efficient planning and exploration. To address these limitations, we propose GraphEQA, a novel approach that utilizes real-time 3D metric-semantic scene graphs (3DSGs) and task relevant images as multi-modal memory for grounding Vision-Language Models (VLMs) to perform EQA tasks in unseen environments. We employ a hierarchical planning approach that exploits the hierarchical nature of 3DSGs for structured planning and semantics-guided exploration. We evaluate GraphEQA in simulation on two benchmark datasets, HM-EQA and OpenEQA, and demonstrate that it outperforms key baselines by completing EQA tasks with higher success rates and fewer planning steps. We further demonstrate GraphEQA in multiple real-world home and office environments.

Paperid: 1520, https://arxiv.org/pdf/2412.13947.pdf GitHub

Abstract:
In this study, we define and tackle zero shot "real" classification by description, a novel task that evaluates the ability of Vision-Language Models (VLMs) like CLIP to classify objects based solely on descriptive attributes, excluding object class names. This approach highlights the current limitations of VLMs in understanding intricate object descriptions, pushing these models beyond mere object recognition. To facilitate this exploration, we introduce a new challenge and release description data for six popular fine-grained benchmarks, which omit object names to encourage genuine zero-shot learning within the research community. Additionally, we propose a method to enhance CLIP's attribute detection capabilities through targeted training using ImageNet21k's diverse object categories, paired with rich attribute descriptions generated by large language models. Furthermore, we introduce a modified CLIP architecture that leverages multiple resolutions to improve the detection of fine-grained part attributes. Through these efforts, we broaden the understanding of part-attribute recognition in CLIP, improving its performance in fine-grained classification tasks across six popular benchmarks, as well as in the PACO dataset, a widely used benchmark for object-attribute recognition. Code is available at: https://github.com/ethanbar11/grounding_ge_public.

Paperid: 1521, https://arxiv.org/pdf/2412.13877.pdf GitHub

Abstract:
In this paper, we introduce RoboMIND (Multi-embodiment Intelligence Normative Data for Robot Manipulation), a dataset containing 107k demonstration trajectories across 479 diverse tasks involving 96 object classes. RoboMIND is collected through human teleoperation and encompasses comprehensive robotic-related information, including multi-view observations, proprioceptive robot state information, and linguistic task descriptions. To ensure data consistency and reliability for imitation learning, RoboMIND is built on a unified data collection platform and a standardized protocol, covering four distinct robotic embodiments: the Franka Emika Panda, the UR5e, the AgileX dual-arm robot, and a humanoid robot with dual dexterous hands. Our dataset also includes 5k real-world failure demonstrations, each accompanied by detailed causes, enabling failure reflection and correction during policy learning. Additionally, we created a digital twin environment in the Isaac Sim simulator, replicating the real-world tasks and assets, which facilitates the low-cost collection of additional training data and enables efficient evaluation. To demonstrate the quality and diversity of our dataset, we conducted extensive experiments using various imitation learning methods for single-task settings and state-of-the-art Vision-Language-Action (VLA) models for multi-task scenarios. By leveraging RoboMIND, the VLA models achieved high manipulation success rates and demonstrated strong generalization capabilities. To the best of our knowledge, RoboMIND is the largest multi-embodiment teleoperation dataset collected on a unified platform, providing large-scale and high-quality robotic training data. Our project is at https://x-humanoid-robomind.github.io/.

Paperid: 1522, https://arxiv.org/pdf/2412.13817.pdf GitHub

Abstract:
Recent studies have shown that large vision-language models (LVLMs) often suffer from the issue of object hallucinations (OH). To mitigate this issue, we introduce an efficient method that edits the model weights based on an unsafe subspace, which we call HalluSpace in this paper. With truthful and hallucinated text prompts accompanying the visual content as inputs, the HalluSpace can be identified by extracting the hallucinated embedding features and removing the truthful representations in LVLMs. By orthogonalizing the model weights, input features will be projected into the Null space of the HalluSpace to reduce OH, based on which we name our method Nullu. We reveal that HalluSpaces generally contain prior information in the large language models (LLMs) applied to build LVLMs, which have been shown as essential causes of OH in previous studies. Therefore, null space projection suppresses the LLMs' priors to filter out the hallucinated features, resulting in contextually accurate outputs. Experiments show that our method can effectively mitigate OH across different LVLM families without extra inference costs and also show strong performance in general LVLM benchmarks. Code is released at https://github.com/Ziwei-Zheng/Nullu.

Paperid: 1523, https://arxiv.org/pdf/2412.13303.pdf GitHub

Abstract:
Scaling the input image resolution is essential for enhancing the performance of Vision Language Models (VLMs), particularly in text-rich image understanding tasks. However, popular visual encoders such as ViTs become inefficient at high resolutions due to the large number of tokens and high encoding latency caused by stacked self-attention layers. At different operational resolutions, the vision encoder of a VLM can be optimized along two axes: reducing encoding latency and minimizing the number of visual tokens passed to the LLM, thereby lowering overall latency. Based on a comprehensive efficiency analysis of the interplay between image resolution, vision latency, token count, and LLM size, we introduce FastVLM, a model that achieves an optimized trade-off between latency, model size and accuracy. FastVLM incorporates FastViTHD, a novel hybrid vision encoder designed to output fewer tokens and significantly reduce encoding time for high-resolution images. Unlike previous methods, FastVLM achieves the optimal balance between visual token count and image resolution solely by scaling the input image, eliminating the need for additional token pruning and simplifying the model design. In the LLaVA-1.5 setup, FastVLM achieves 3.2$\times$ improvement in time-to-first-token (TTFT) while maintaining similar performance on VLM benchmarks compared to prior works. Compared to LLaVa-OneVision at the highest resolution (1152$\times$1152), FastVLM achieves better performance on key benchmarks like SeedBench, MMMU and DocVQA, using the same 0.5B LLM, but with 85$\times$ faster TTFT and a vision encoder that is 3.4$\times$ smaller. Code and models are available at https://github.com/apple/ml-fastvlm.

Paperid: 1524, https://arxiv.org/pdf/2412.12932.pdf GitHub

Abstract:
Large Vision-Language Models (LVLMs) have recently demonstrated amazing success in multi-modal tasks, including advancements in Multi-modal Chain-of-Thought (MCoT) reasoning. Despite these successes, current benchmarks still follow a traditional paradigm with multi-modal input and text-modal output, which leads to significant drawbacks such as missing visual operations and vague expressions. Motivated by this, we introduce a novel Chain of Multi-modal Thought (CoMT) benchmark to address these limitations. Different from the traditional MCoT benchmark, CoMT requires both multi-modal input and multi-modal reasoning output, aiming to mimic human-like reasoning that inherently integrates visual operation. Specifically, CoMT consists of four categories: (1) Visual Creation, (2) Visual Deletion, (3) Visual Update, and (4) Visual Selection to comprehensively explore complex visual operations and concise expression in real scenarios. We evaluate various LVLMs and strategies on CoMT, revealing some key insights into the capabilities and limitations of the current approaches. We hope that CoMT can inspire more research on introducing multi-modal generation into the reasoning process.

Paperid: 1525, https://arxiv.org/pdf/2412.12798.pdf GitHub GitHub

Abstract:
Instance segmentation algorithms in remote sensing are typically based on conventional methods, limiting their application to seen scenarios and closed-set predictions. In this work, we propose a novel task called zero-shot remote sensing instance segmentation, aimed at identifying aerial objects that are absent from training data. Challenges arise when classifying aerial categories with high inter-class similarity and intra-class variance. Besides, the domain gap between vision-language models' pretraining datasets and remote sensing datasets hinders the zero-shot capabilities of the pretrained model when it is directly applied to remote sensing images. To address these challenges, we propose a $\textbf{Z}$ero-Sh$\textbf{o}$t $\textbf{R}$emote Sensing $\textbf{I}$nstance Segmentation framework, dubbed $\textbf{ZoRI}$. Our approach features a discrimination-enhanced classifier that uses refined textual embeddings to increase the awareness of class disparities. Instead of direct fine-tuning, we propose a knowledge-maintained adaptation strategy that decouples semantic-related information to preserve the pretrained vision-language alignment while adjusting features to capture remote sensing domain-specific visual cues. Additionally, we introduce a prior-injected prediction with cache bank of aerial visual prototypes to supplement the semantic richness of text embeddings and seamlessly integrate aerial representations, adapting to the remote sensing domain. We establish new experimental protocols and benchmarks, and extensive experiments convincingly demonstrate that ZoRI achieves the state-of-art performance on the zero-shot remote sensing instance segmentation task. Our code is available at https://github.com/HuangShiqi128/ZoRI.

Paperid: 1526, https://arxiv.org/pdf/2412.12693.pdf GitHub

Abstract:
Current vision-language models may grasp basic spatial cues and simple directions (e.g. left, right, front, back), but struggle with the multi-dimensional spatial reasoning necessary for human-like understanding and real-world applications. To address this gap, we develop SPHERE (Spatial Perception and Hierarchical Evaluation of REasoning), a hierarchical evaluation framework supported by a new human-annotated dataset. SPHERE systematically probes models across increasing levels of complexity, from fundamental skills to multi-skill integration and high-level reasoning that combines spatial, visual, and logical understanding. Benchmark evaluation of state-of-the-art models reveals significant deficiencies, especially in reasoning about distance and proximity, understanding both egocentric and allocentric perspectives, and applying spatial logic in physical contexts. These findings expose critical blind spots in existing models and underscore the need for more advanced spatial reasoning techniques, driving the development of vision-language models that align more closely with human spatial cognition. The SPHERE benchmark is available at https://github.com/zwenyu/SPHERE-VLM.

Paperid: 1527, https://arxiv.org/pdf/2412.12222.pdf GitHub

Abstract:
Traffic accidents are a global safety concern, resulting in numerous fatalities each year. A considerable number of these deaths are caused by animal-vehicle collisions (AVCs), which not only endanger human lives but also present serious risks to animal populations. This paper presents an innovative self-training methodology aimed at detecting rare animals, such as the cassowary in Australia, whose survival is threatened by road accidents. The proposed method addresses critical real-world challenges, including acquiring and labelling sensor data for rare animal species in resource-limited environments. It achieves this by leveraging cloud and edge computing, and automatic data labelling to improve the detection performance of the field-deployed model iteratively. Our approach introduces Label-Augmentation Non-Maximum Suppression (LA-NMS), which incorporates a vision-language model (VLM) to enable automated data labelling. During a five-month deployment, we confirmed the method's robustness and effectiveness, resulting in improved object detection accuracy and increased prediction confidence. The source code is available: https://github.com/acfr/CassDetect

Paperid: 1528, https://arxiv.org/pdf/2412.11917.pdf GitHub

Abstract:
Accurately describing images with text is a foundation of explainable AI. Vision-Language Models (VLMs) like CLIP have recently addressed this by aligning images and texts in a shared embedding space, expressing semantic similarities between vision and language embeddings. VLM classification can be improved with descriptions generated by Large Language Models (LLMs). However, it is difficult to determine the contribution of actual description semantics, as the performance gain may also stem from a semantic-agnostic ensembling effect, where multiple modified text prompts act as a noisy test-time augmentation for the original one. We propose an alternative evaluation scenario to decide if a performance boost of LLM-generated descriptions is caused by such a noise augmentation effect or rather by genuine description semantics. The proposed scenario avoids noisy test-time augmentation and ensures that genuine, distinctive descriptions cause the performance boost. Furthermore, we propose a training-free method for selecting discriminative descriptions that work independently of classname-ensembling effects. Our approach identifies descriptions that effectively differentiate classes within a local CLIP label neighborhood, improving classification accuracy across seven datasets. Additionally, we provide insights into the explainability of description-based image classification with VLMs.

Paperid: 1529, https://arxiv.org/pdf/2412.11892.pdf GitHub

Abstract:
In this paper, we present CAD2Program, a new method for reconstructing 3D parametric models from 2D CAD drawings. Our proposed method is inspired by recent successes in vision-language models (VLMs), and departs from traditional methods which rely on task-specific data representations and/or algorithms. Specifically, on the input side, we simply treat the 2D CAD drawing as a raster image, regardless of its original format, and encode the image with a standard ViT model. We show that such an encoding scheme achieves competitive performance against existing methods that operate on vector-graphics inputs, while imposing substantially fewer restrictions on the 2D drawings. On the output side, our method auto-regressively predicts a general-purpose language describing 3D parametric models in text form. Compared to other sequence modeling methods for CAD which use domain-specific sequence representations with fixed-size slots, our text-based representation is more flexible, and can be easily extended to arbitrary geometric entities and semantic or functional properties. Experimental results on a large-scale dataset of cabinet models demonstrate the effectiveness of our method.

Paperid: 1530, https://arxiv.org/pdf/2412.11863.pdf GitHub

Abstract:
Despite their proficiency in general tasks, Multi-modal Large Language Models (MLLMs) struggle with automatic Geometry Problem Solving (GPS), which demands understanding diagrams, interpreting symbols, and performing complex reasoning. This limitation arises from their pre-training on natural images and texts, along with the lack of automated verification in the problem-solving process. Besides, current geometric specialists are limited by their task-specific designs, making them less effective for broader geometric problems. To this end, we present GeoX, a multi-modal large model focusing on geometric understanding and reasoning tasks. Given the significant differences between geometric diagram-symbol and natural image-text, we introduce unimodal pre-training to develop a diagram encoder and symbol decoder, enhancing the understanding of geometric images and corpora. Furthermore, we introduce geometry-language alignment, an effective pre-training paradigm that bridges the modality gap between unimodal geometric experts. We propose a Generator-And-Sampler Transformer (GS-Former) to generate discriminative queries and eliminate uninformative representations from unevenly distributed geometric signals. Finally, GeoX benefits from visual instruction tuning, empowering it to take geometric images and questions as input and generate verifiable solutions. Experiments show that GeoX outperforms both generalists and geometric specialists on publicly recognized benchmarks, such as GeoQA, UniGeo, Geometry3K, and PGPS9k.

Paperid: 1531, https://arxiv.org/pdf/2412.11509.pdf GitHub

Abstract:
Prompt tuning (PT) has long been recognized as an effective and efficient paradigm for transferring large pre-trained vision-language models (VLMs) to downstream tasks by learning a tiny set of context vectors. Nevertheless, in this work, we reveal that freezing the parameters of VLMs during learning the context vectors neither facilitates the transferability of pre-trained knowledge nor improves the memory and time efficiency significantly. Upon further investigation, we find that reducing both the length and width of the feature-gradient propagation flows of the full fine-tuning (FT) baseline is key to achieving effective and efficient knowledge transfer. Motivated by this, we propose Skip Tuning, a novel paradigm for adapting VLMs to downstream tasks. Unlike existing PT or adapter-based methods, Skip Tuning applies Layer-wise Skipping (LSkip) and Class-wise Skipping (CSkip) upon the FT baseline without introducing extra context vectors or adapter modules. Extensive experiments across a wide spectrum of benchmarks demonstrate the superior effectiveness and efficiency of our Skip Tuning over both PT and adapter-based methods. Code: https://github.com/Koorye/SkipTuning.

Paperid: 1532, https://arxiv.org/pdf/2412.11464.pdf GitHub

Abstract:
Open-vocabulary image segmentation has been advanced through the synergy between mask generators and vision-language models like Contrastive Language-Image Pre-training (CLIP). Previous approaches focus on generating masks while aligning mask features with text embeddings during training. In this paper, we observe that relying on generated low-quality masks can weaken the alignment of vision and language in regional representations. This motivates us to present a new fine-tuning framework, named MaskCLIP++, which uses ground-truth masks instead of generated masks to enhance the mask classification capability of CLIP. Due to the limited diversity of image segmentation datasets with mask annotations, we propose incorporating a consistency alignment principle during fine-tuning, which alleviates categorical bias toward the fine-tuning dataset. After low-cost fine-tuning, MaskCLIP++ significantly improves the mask classification performance on multi-domain datasets. Combining with the mask generator in previous state-of-the-art mask-based open vocabulary segmentation methods, we achieve performance improvements of +1.7, +2.3, +2.1, +3.1, and +0.3 mIoU on the A-847, PC-459, A-150, PC-59, and PAS-20 datasets, respectively. Code is avaliable at https://github.com/HVision-NKU/MaskCLIPpp .

Paperid: 1533, https://arxiv.org/pdf/2412.11077.pdf GitHub

Abstract:
Composed Image Retrieval (CIR) aims to retrieve target images that closely resemble a reference image while integrating user-specified textual modifications, thereby capturing user intent more precisely. Existing training-free zero-shot CIR (ZS-CIR) methods often employ a two-stage process: they first generate a caption for the reference image and then use Large Language Models for reasoning to obtain a target description. However, these methods suffer from missing critical visual details and limited reasoning capabilities, leading to suboptimal retrieval performance. To address these challenges, we propose a novel, training-free one-stage method, One-Stage Reflective Chain-of-Thought Reasoning for ZS-CIR (OSrCIR), which employs Multimodal Large Language Models to retain essential visual information in a single-stage reasoning process, eliminating the information loss seen in two-stage methods. Our Reflective Chain-of-Thought framework further improves interpretative accuracy by aligning manipulation intent with contextual cues from reference images. OSrCIR achieves performance gains of 1.80% to 6.44% over existing training-free methods across multiple tasks, setting new state-of-the-art results in ZS-CIR and enhancing its utility in vision-language applications. Our code will be available at https://github.com/Pter61/osrcir2024/.

Paperid: 1534, https://arxiv.org/pdf/2412.10680.pdf GitHub

Abstract:
Universal Cross-Domain Retrieval (UCDR) retrieves relevant images from unseen domains and classes without semantic labels, ensuring robust generalization. Existing methods commonly employ prompt tuning with pre-trained vision-language models but are inherently limited by static prompts, reducing adaptability. We propose UCDR-Adapter, which enhances pre-trained models with adapters and dynamic prompt generation through a two-phase training strategy. First, Source Adapter Learning integrates class semantics with domain-specific visual knowledge using a Learnable Textual Semantic Template and optimizes Class and Domain Prompts via momentum updates and dual loss functions for robust alignment. Second, Target Prompt Generation creates dynamic prompts by attending to masked source prompts, enabling seamless adaptation to unseen domains and classes. Unlike prior approaches, UCDR-Adapter dynamically adapts to evolving data distributions, enhancing both flexibility and generalization. During inference, only the image branch and generated prompts are used, eliminating reliance on textual inputs for highly efficient retrieval. Extensive benchmark experiments show that UCDR-Adapter consistently outperforms ProS in most cases and other state-of-the-art methods on UCDR, U(c)CDR, and U(d)CDR settings.

Paperid: 1535, https://arxiv.org/pdf/2412.10594.pdf GitHub

Abstract:
Human perception of similarity across uni- and multimodal inputs is highly complex, making it challenging to develop automated metrics that accurately mimic it. General purpose vision-language models, such as CLIP and large multi-modal models (LMMs), can be applied as zero-shot perceptual metrics, and several recent works have developed models specialized in narrow perceptual tasks. However, the extent to which existing perceptual metrics align with human perception remains unclear. To investigate this question, we introduce UniSim-Bench, a benchmark encompassing 7 multi-modal perceptual similarity tasks, with a total of 25 datasets. Our evaluation reveals that while general-purpose models perform reasonably well on average, they often lag behind specialized models on individual tasks. Conversely, metrics fine-tuned for specific tasks fail to generalize well to unseen, though related, tasks. As a first step towards a unified multi-task perceptual similarity metric, we fine-tune both encoder-based and generative vision-language models on a subset of the UniSim-Bench tasks. This approach yields the highest average performance, and in some cases, even surpasses taskspecific models. Nevertheless, these models still struggle with generalization to unseen tasks, highlighting the ongoing challenge of learning a robust, unified perceptual similarity metric capable of capturing the human notion of similarity. The code and models are available at https://github.com/SaraGhazanfari/UniSim.

Paperid: 1536, https://arxiv.org/pdf/2412.10372.pdf GitHub GitHub

Abstract:
Vision-Language Models (VLMs) trained via contrastive learning have achieved notable success in natural image tasks. However, their application in the medical domain remains limited due to the scarcity of openly accessible, large-scale medical image-text datasets. Existing medical VLMs either train on closed-source proprietary or relatively small open-source datasets that do not generalize well. Similarly, most models remain specific to a single or limited number of medical imaging domains, again restricting their applicability to other modalities. To address this gap, we introduce UniMed, a large-scale, open-source multi-modal medical dataset comprising over 5.3 million image-text pairs across six diverse imaging modalities: X-ray, CT, MRI, Ultrasound, Pathology, and Fundus. UniMed is developed using a data-collection framework that leverages Large Language Models (LLMs) to transform modality-specific classification datasets into image-text formats while incorporating existing image-text data from the medical domain, facilitating scalable VLM pretraining. Using UniMed, we trained UniMed-CLIP, a unified VLM for six modalities that significantly outperforms existing generalist VLMs and matches modality-specific medical VLMs, achieving notable gains in zero-shot evaluations. For instance, UniMed-CLIP improves over BiomedCLIP (trained on proprietary data) by an absolute gain of +12.61, averaged over 21 datasets, while using 3x less training data. To facilitate future research, we release UniMed dataset, training codes, and models at https://github.com/mbzuai-oryx/UniMed-CLIP.

Paperid: 1537, https://arxiv.org/pdf/2412.10302.pdf GitHub

Abstract:
We present DeepSeek-VL2, an advanced series of large Mixture-of-Experts (MoE) Vision-Language Models that significantly improves upon its predecessor, DeepSeek-VL, through two key major upgrades. For the vision component, we incorporate a dynamic tiling vision encoding strategy designed for processing high-resolution images with different aspect ratios. For the language component, we leverage DeepSeekMoE models with the Multi-head Latent Attention mechanism, which compresses Key-Value cache into latent vectors, to enable efficient inference and high throughput. Trained on an improved vision-language dataset, DeepSeek-VL2 demonstrates superior capabilities across various tasks, including but not limited to visual question answering, optical character recognition, document/table/chart understanding, and visual grounding. Our model series is composed of three variants: DeepSeek-VL2-Tiny, DeepSeek-VL2-Small and DeepSeek-VL2, with 1.0B, 2.8B and 4.5B activated parameters respectively. DeepSeek-VL2 achieves competitive or state-of-the-art performance with similar or fewer activated parameters compared to existing open-source dense and MoE-based models. Codes and pre-trained models are publicly accessible at https://github.com/deepseek-ai/DeepSeek-VL2.

Paperid: 1538, https://arxiv.org/pdf/2412.09936.pdf GitHub

Abstract:
The obesity phenomenon, known as the heavy issue, is a leading cause of preventable chronic diseases worldwide. Traditional calorie estimation tools often rely on specific data formats or complex pipelines, limiting their practicality in real-world scenarios. Recently, vision-language models (VLMs) have excelled in understanding real-world contexts and enabling conversational interactions, making them ideal for downstream tasks such as ingredient analysis. However, applying VLMs to calorie estimation requires domain-specific data and alignment strategies. To this end, we curated CalData, a 330K image-text pair dataset tailored for ingredient recognition and calorie estimation, combining a large-scale recipe dataset with detailed nutritional instructions for robust vision-language training. Built upon this dataset, we present CaLoRAify, a novel VLM framework aligning ingredient recognition and calorie estimation via training with visual-text pairs. During inference, users only need a single monocular food image to estimate calories while retaining the flexibility of agent-based conversational interaction. With Low-rank Adaptation (LoRA) and Retrieve-augmented Generation (RAG) techniques, our system enhances the performance of foundational VLMs in the vertical domain of calorie estimation. Our code and data are fully open-sourced at https://github.com/KennyYao2001/16824-CaLORAify.

Paperid: 1539, https://arxiv.org/pdf/2412.09616.pdf GitHub

Abstract:
Vision-Language Models (VLMs) have shown promising capabilities in handling various multimodal tasks, yet they struggle in long-context scenarios, particularly in tasks involving videos, high-resolution images, or lengthy image-text documents. In our work, we first conduct an empirical analysis of the long-context capabilities of VLMs using our augmented long-context multimodal datasets. Our findings reveal that directly applying the positional encoding mechanism used for textual tokens to visual tokens is suboptimal, and VLM performance degrades sharply when the position encoding exceeds the model's context window. To address this, we propose Variable Visual Position Encoding (V2PE), a novel positional encoding approach that employs variable and smaller increments for visual tokens, enabling more efficient management of long multimodal sequences. Our experiments demonstrate the effectiveness of V2PE to enhances VLMs' ability to effectively understand and reason over long multimodal contexts. We further integrate V2PE with our augmented long-context multimodal datasets to fine-tune the open-source VLM, InternVL2. The fine-tuned model achieves strong performance on both standard and long-context multimodal tasks. Notably, when the sequence length of the training dataset is increased to 256K tokens, the model is capable of processing multimodal sequences up to 1M tokens, highlighting its potential for real-world long-context applications.

Paperid: 1540, https://arxiv.org/pdf/2412.09530.pdf GitHub

Abstract:
The application of Large Vision-Language Models (LVLMs) for analyzing images and videos is an exciting and rapidly evolving field. In recent years, we've seen significant growth in high-quality image-text datasets for fine-tuning image understanding, but there is still a lack of comparable datasets for videos. Additionally, many VideoLLMs are extensions of single-image VLMs, which may not efficiently handle the complexities of longer videos. In this study, we introduce a large-scale synthetic dataset created from proprietary models, using carefully designed prompts to tackle a wide range of questions. We also explore a dynamic visual token compression architecture that strikes a balance between computational efficiency and performance. Our proposed \model{} achieves state-of-the-art results across various video tasks and shows impressive generalization, setting new baselines in multi-image understanding. Notably, \model{} delivers an absolute improvement of 2.7\% over LLaVA-OneVision on VideoMME and 10.7\% on MuirBench. Codes are available at https://github.com/Hon-Wong/ByteVideoLLM

Paperid: 1541, https://arxiv.org/pdf/2412.09442.pdf GitHub GitHub

Abstract:
Textual-based prompt learning methods primarily employ multiple learnable soft prompts and hard class tokens in a cascading manner as text inputs, aiming to align image and text (category) spaces for downstream tasks. However, current training is restricted to aligning images with predefined known categories and cannot be associated with unknown categories. In this work, we propose utilizing universal attributes as a bridge to enhance the alignment between images and unknown categories. Specifically, we introduce an Attribute-anchored Textual Prompt learning method for vision-language models, named ATPrompt. This approach expands the learning space of soft prompts from the original one-dimensional category level into the multi-dimensional attribute level by incorporating multiple attribute tokens into the learnable soft prompts. Through this modification, we transform the text prompt from a category-centric form to an attribute-category hybrid form. Additionally, we introduce a straightforward differentiable attribute search method to identify representative and suitable attributes for downstream tasks. As an easy-to-use plug-in technique, ATPrompt can seamlessly replace the existing basic prompt format in textual-based methods, providing general improvements at a negligible computational cost. Extensive experiments across 11 datasets validate the effectiveness of our method. Code is publicly available at https://github.com/zhengli97/ATPrompt.

Paperid: 1542, https://arxiv.org/pdf/2412.09286.pdf GitHub

Abstract:
Robots are increasingly deployed across diverse domains to tackle tasks requiring novel skills. However, current robot learning algorithms for acquiring novel skills often rely on demonstration datasets or environment interactions, resulting in high labor costs and potential safety risks. To address these challenges, this study proposes DemoGen, a skill-learning framework that enables robots to acquire novel skills from natural language instructions. DemoGen leverages the vision-language model and the video diffusion model to generate demonstration videos of novel skills, which enabling robots to learn new skills effectively. Experimental evaluations in the MetaWorld simulation environments demonstrate the pipeline's capability to generate high-fidelity and reliable demonstrations. Using the generated demonstrations, various skill learning algorithms achieve an accomplishment rate three times the original on novel tasks. These results highlight a novel approach to robot learning, offering a foundation for the intuitive and intelligent acquisition of novel robotic skills. (Project website: https://aoqunjin.github.io/LNSLGD/)

Paperid: 1543, https://arxiv.org/pdf/2412.09082.pdf GitHub

Abstract:
Existing Vision-Language Navigation (VLN) methods primarily focus on single-stage navigation, limiting their effectiveness in multi-stage and long-horizon tasks within complex and dynamic environments. To address these limitations, we propose a novel VLN task, named Long-Horizon Vision-Language Navigation (LH-VLN), which emphasizes long-term planning and decision consistency across consecutive subtasks. Furthermore, to support LH-VLN, we develop an automated data generation platform NavGen, which constructs datasets with complex task structures and improves data utility through a bidirectional, multi-granularity generation approach. To accurately evaluate complex tasks, we construct the Long-Horizon Planning and Reasoning in VLN (LHPR-VLN) benchmark consisting of 3,260 tasks with an average of 150 task steps, serving as the first dataset specifically designed for the long-horizon vision-language navigation task. Furthermore, we propose Independent Success Rate (ISR), Conditional Success Rate (CSR), and CSR weight by Ground Truth (CGT) metrics, to provide fine-grained assessments of task completion. To improve model adaptability in complex tasks, we propose a novel Multi-Granularity Dynamic Memory (MGDM) module that integrates short-term memory blurring with long-term memory retrieval to enable flexible navigation in dynamic environments. Our platform, benchmark and method supply LH-VLN with a robust data generation pipeline, comprehensive model evaluation dataset, reasonable metrics, and a novel VLN model, establishing a foundational framework for advancing LH-VLN.

Paperid: 1544, https://arxiv.org/pdf/2412.08614.pdf GitHub

Abstract:
Generating detailed captions comprehending text-rich visual content in images has received growing attention for Large Vision-Language Models (LVLMs). However, few studies have developed benchmarks specifically tailored for detailed captions to measure their accuracy and comprehensiveness. In this paper, we introduce a detailed caption benchmark, termed as CompreCap, to evaluate the visual context from a directed scene graph view. Concretely, we first manually segment the image into semantically meaningful regions (i.e., semantic segmentation mask) according to common-object vocabulary, while also distinguishing attributes of objects within all those regions. Then directional relation labels of these objects are annotated to compose a directed scene graph that can well encode rich compositional information of the image. Based on our directed scene graph, we develop a pipeline to assess the generated detailed captions from LVLMs on multiple levels, including the object-level coverage, the accuracy of attribute descriptions, the score of key relationships, etc. Experimental results on the CompreCap dataset confirm that our evaluation method aligns closely with human evaluation scores across LVLMs.

Paperid: 1545, https://arxiv.org/pdf/2412.08261.pdf GitHub

Abstract:
We aim to develop a model-based planning framework for world models that can be scaled with increasing model and data budgets for general-purpose manipulation tasks with only language and vision inputs. To this end, we present FLow-centric generative Planning (FLIP), a model-based planning algorithm on visual space that features three key modules: 1. a multi-modal flow generation model as the general-purpose action proposal module; 2. a flow-conditioned video generation model as the dynamics module; and 3. a vision-language representation learning model as the value module. Given an initial image and language instruction as the goal, FLIP can progressively search for long-horizon flow and video plans that maximize the discounted return to accomplish the task. FLIP is able to synthesize long-horizon plans across objects, robots, and tasks with image flows as the general action representation, and the dense flow information also provides rich guidance for long-horizon video generation. In addition, the synthesized flow and video plans can guide the training of low-level control policies for robot execution. Experiments on diverse benchmarks demonstrate that FLIP can improve both the success rates and quality of long-horizon video plan synthesis and has the interactive world model property, opening up wider applications for future works.Video demos are on our website: https://nus-lins-lab.github.io/flipweb/.

Paperid: 1546, https://arxiv.org/pdf/2412.08176.pdf GitHub

Abstract:
Despite the efficiency of prompt learning in transferring vision-language models (VLMs) to downstream tasks, existing methods mainly learn the prompts in a coarse-grained manner where the learned prompt vectors are shared across all categories. Consequently, the tailored prompts often fail to discern class-specific visual concepts, thereby hindering the transferred performance for classes that share similar or complex visual attributes. Recent advances mitigate this challenge by leveraging external knowledge from Large Language Models (LLMs) to furnish class descriptions, yet incurring notable inference costs. In this paper, we introduce TextRefiner, a plug-and-play method to refine the text prompts of existing methods by leveraging the internal knowledge of VLMs. Particularly, TextRefiner builds a novel local cache module to encapsulate fine-grained visual concepts derivedfrom local tokens within the image branch. By aggregating and aligning the cached visual descriptions with the original output of the text branch, TextRefiner can efficiently refine and enrich the learned prompts from existing methods without relying on any external expertise. For example, it improves the performance of CoOp from 71.66 % to 76.94 % on 11 benchmarks, surpassing CoCoOp which introduces instance-wise features for text prompts. Equipped with TextRefiner, PromptKD achieves state-of-the-art performance and is efficient in inference. Our code is relesed at https://github.com/xjjxmu/TextRefiner

Paperid: 1547, https://arxiv.org/pdf/2412.08125.pdf GitHub

Abstract:
Existing Large Vision-Language Models (LVLMs) excel at matching concepts across multi-modal inputs but struggle with compositional concepts and high-level relationships between entities. This paper introduces Progressive multi-granular Vision-Language alignments (PromViL), a novel framework to enhance LVLMs' ability in performing grounded compositional visual reasoning tasks. Our approach constructs a hierarchical structure of multi-modal alignments, ranging from simple to complex concepts. By progressively aligning textual descriptions with corresponding visual regions, our model learns to leverage contextual information from lower levels to inform higher-level reasoning. To facilitate this learning process, we introduce a data generation process that creates a novel dataset derived from Visual Genome, providing a wide range of nested compositional vision-language pairs. Experimental results demonstrate that our PromViL framework significantly outperforms baselines on various visual grounding and compositional question answering tasks. The code is available at: https://github.com/lqh52/PromViL.

Paperid: 1548, https://arxiv.org/pdf/2412.07626.pdf GitHub

Abstract:
Document content extraction is a critical task in computer vision, underpinning the data needs of large language models (LLMs) and retrieval-augmented generation (RAG) systems. Despite recent progress, current document parsing methods have not been fairly and comprehensively evaluated due to the narrow coverage of document types and the simplified, unrealistic evaluation procedures in existing benchmarks. To address these gaps, we introduce OmniDocBench, a novel benchmark featuring high-quality annotations across nine document sources, including academic papers, textbooks, and more challenging cases such as handwritten notes and densely typeset newspapers. OmniDocBench supports flexible, multi-level evaluations--ranging from an end-to-end assessment to the task-specific and attribute--based analysis using 19 layout categories and 15 attribute labels. We conduct a thorough evaluation of both pipeline-based methods and end-to-end vision-language models, revealing their strengths and weaknesses across different document types. OmniDocBench sets a new standard for the fair, diverse, and fine-grained evaluation in document parsing. Dataset and code are available at https://github.com/opendatalab/OmniDocBench.

Paperid: 1549, https://arxiv.org/pdf/2412.07518.pdf GitHub

Abstract:
Large vision-language models (LVLMs) have demonstrated remarkable capabilities in multimodal understanding and generation tasks. However, these models occasionally generate hallucinatory texts, resulting in descriptions that seem reasonable but do not correspond to the image. This phenomenon can lead to wrong driving decisions of the autonomous driving system. To address this challenge, this paper proposes HCOENet, a plug-and-play chain-of-thought correction method designed to eliminate object hallucinations and generate enhanced descriptions for critical objects overlooked in the initial response. Specifically, HCOENet employs a cross-checking mechanism to filter entities and directly extracts critical objects from the given image, enriching the descriptive text. Experimental results on the POPE benchmark demonstrate that HCOENet improves the F1-score of the Mini-InternVL-4B and mPLUG-Owl3 models by 12.58% and 4.28%, respectively. Additionally, qualitative results using images collected in open campus scene further highlight the practical applicability of the proposed method. Compared with the GPT-4o model, HCOENet achieves comparable descriptive performance while significantly reducing costs. Finally, two novel semantic understanding datasets, CODA_desc and nuScenes_desc, are created for traffic scenarios to support future research. The codes and datasets are publicly available at https://github.com/fjq-tongji/HCOENet.

Paperid: 1550, https://arxiv.org/pdf/2412.07472.pdf GitHub

Abstract:
Recent advances in embodied agents with multimodal perception and reasoning capabilities based on large vision-language models (LVLMs), excel in autonomously interacting either real or cyber worlds, helping people make intelligent decisions in complex environments. However, the current works are normally optimized by golden action trajectories or ideal task-oriented solutions toward a definitive goal. This paradigm considers limited user-oriented factors, which could be the reason for their performance reduction in a wide range of personal assistant applications. To address this, we propose Chain-of-User-Thought (COUT), a novel embodied reasoning paradigm that takes a chain of thought from basic action thinking to explicit and implicit personalized preference thought to incorporate personalized factors into autonomous agent learning. To target COUT, we introduce SmartAgent, an agent framework perceiving cyber environments and reasoning personalized requirements as 1) interacting with GUI to access an item pool, 2) generating users' explicit requirements implied by previous actions, and 3) recommending items to fulfill users' implicit requirements. To demonstrate SmartAgent's capabilities, we also create a brand-new dataset SmartSpot that offers a full-stage personalized action-involved environment. To our best knowledge, our work is the first to formulate the COUT process, serving as a preliminary attempt towards embodied personalized agent learning. Our extensive experiments on SmartSpot illuminate SmartAgent's functionality among a series of embodied and personalized sub-tasks. We will release code and data upon paper notification at https://github.com/tsinghua-fib-lab/SmartAgent.

Paperid: 1551, https://arxiv.org/pdf/2412.07112.pdf GitHub

Abstract:
The rapid development of large Vision-Language Models (VLMs) has led to impressive results on academic benchmarks, primarily in widely spoken languages. However, significant gaps remain in the ability of current VLMs to handle low-resource languages and varied cultural contexts, largely due to a lack of high-quality, diverse, and safety-vetted data. Consequently, these models often struggle to understand low-resource languages and cultural nuances in a manner free from toxicity. To address these limitations, we introduce Maya, an open-source Multimodal Multilingual model. Our contributions are threefold: 1) a multilingual image-text pretraining dataset in eight languages, based on the LLaVA pretraining dataset; 2) a thorough analysis of toxicity within the LLaVA dataset, followed by the creation of a novel toxicity-free version across eight languages; and 3) a multilingual image-text model supporting these languages, enhancing cultural and linguistic comprehension in vision-language tasks. Code available at https://github.com/nahidalam/maya.

Paperid: 1552, https://arxiv.org/pdf/2412.06775.pdf GitHub

Abstract:
While large vision-language models (LVLMs) have shown impressive capabilities in generating plausible responses correlated with input visual contents, they still suffer from hallucinations, where the generated text inaccurately reflects visual contents. To address this, recent approaches apply contrastive decoding to calibrate the model's response via contrasting output distributions with original and visually distorted samples, demonstrating promising hallucination mitigation in a training-free manner. However, the potential of changing information in visual inputs is not well-explored, so a deeper investigation into the behaviors of visual contrastive decoding is of great interest. In this paper, we first explore various methods for contrastive decoding to change visual contents, including image downsampling and editing. Downsampling images reduces the detailed textual information while editing yields new contents in images, providing new aspects as visual contrastive samples. To further study benefits by using different contrastive samples, we analyze probability-level metrics, including entropy and distribution distance. Interestingly, the effect of these samples in mitigating hallucinations varies a lot across LVLMs and benchmarks. Based on our analysis, we propose a simple yet effective method to combine contrastive samples, offering a practical solution for applying contrastive decoding across various scenarios. Extensive experiments are conducted to validate the proposed fusion method among different benchmarks.

Paperid: 1553, https://arxiv.org/pdf/2412.06474.pdf GitHub

Abstract:
Large vision-language models (LVLMs) demonstrate remarkable capabilities in multimodal tasks but are prone to misinterpreting visual inputs, often resulting in hallucinations and unreliable outputs. To address these challenges, we propose Dropout Decoding, a novel inference-time approach that quantifies the uncertainty of visual tokens and selectively masks uncertain tokens to improve decoding. Our method measures the uncertainty of each visual token by projecting it onto the text space and decomposing it into aleatoric and epistemic components. Specifically, we focus on epistemic uncertainty, which captures perception-related errors more effectively. Inspired by dropout regularization, we introduce uncertainty-guided token dropout, which applies the dropout principle to input visual tokens instead of model parameters, and during inference rather than training. By aggregating predictions from an ensemble of masked decoding contexts, Dropout Decoding robustly mitigates errors arising from visual token misinterpretations. Evaluations on benchmarks including CHAIR, THRONE, and MMBench demonstrate that Dropout Decoding significantly reduces object hallucinations (OH) and enhances both reliability and quality of LVLM outputs across diverse visual contexts.

Paperid: 1554, https://arxiv.org/pdf/2412.06458.pdf GitHub

Abstract:
Although Large Vision-Language Models (LVLMs) have achieved impressive results, their high computational costs pose a significant barrier to wide application. To enhance inference efficiency, most existing approaches can be categorized as parameter-dependent or token-dependent strategies to reduce computational demands. However, parameter-dependent methods require retraining LVLMs to recover performance while token-dependent strategies struggle to consistently select the most relevant tokens. In this paper, we systematically analyze the above challenges and provide a series of valuable insights for inference acceleration. Based on these findings, we propose a novel framework, the Pruning All-Rounder (PAR). Different from previous works, PAR develops a meta-router to adaptively organize pruning flows across both tokens and layers. With a self-supervised learning manner, our method achieves a superior balance between performance and efficiency. Notably, PAR is highly flexible, offering multiple pruning versions to address a range of acceleration scenarios. The code for this work is publicly available at https://github.com/ASGO-MM/Pruning-All-Rounder.

Paperid: 1555, https://arxiv.org/pdf/2412.06286.pdf GitHub

Abstract:
Object detection in art is a valuable tool for the digital humanities, as it allows for faster identification of objects in artistic and historical images compared to humans. However, annotating such images poses significant challenges due to the need for specialized domain expertise. We present NADA (no annotations for detection in art), a pipeline that leverages diffusion models' art-related knowledge for object detection in paintings without the need for full bounding box supervision. Our method, which supports both weakly-supervised and zero-shot scenarios and does not require any fine-tuning of its pretrained components, consists of a class proposer based on large vision-language models and a class-conditioned detector based on Stable Diffusion. NADA is evaluated on two artwork datasets, ArtDL 2.0 and IconArt, outperforming prior work in weakly-supervised detection, while being the first work for zero-shot object detection in art. Code is available at https://github.com/patrick-john-ramos/nada

Paperid: 1556, https://arxiv.org/pdf/2412.06263.pdf GitHub

Abstract:
In this paper, we introduce iLLaVA, a simple method that can be seamlessly deployed upon current Large Vision-Language Models (LVLMs) to greatly increase the throughput with nearly lossless model performance, without a further requirement to train. iLLaVA achieves this by finding and gradually merging the redundant tokens with an accurate and fast algorithm, which can merge hundreds of tokens within only one step. While some previous methods have explored directly pruning or merging tokens in the inference stage to accelerate models, our method excels in both performance and throughput by two key designs. First, while most previous methods only try to save the computations of Large Language Models (LLMs), our method accelerates the forward pass of both image encoders and LLMs in LVLMs, which both occupy a significant part of time during inference. Second, our method recycles the beneficial information from the pruned tokens into existing tokens, which avoids directly dropping context tokens like previous methods to cause performance loss. iLLaVA can nearly 2$\times$ the throughput, and reduce the memory costs by half with only a 0.2\% - 0.5\% performance drop across models of different scales including 7B, 13B and 34B. On tasks across different domains including single-image, multi-images and videos, iLLaVA demonstrates strong generalizability with consistently promising efficiency. We finally offer abundant visualizations to show the merging processes of iLLaVA in each step, which show insights into the distribution of computing resources in LVLMs. Code is available at https://github.com/hulianyuyy/iLLaVA.

Paperid: 1557, https://arxiv.org/pdf/2412.06244.pdf GitHub GitHub

Abstract:
Pre-trained vision-language models (VLMs), such as CLIP, have demonstrated impressive zero-shot recognition capability, but still underperform in dense prediction tasks. Self-distillation recently is emerging as a promising approach for fine-tuning VLMs to better adapt to local regions without requiring extensive annotations. However, previous state-of-the-art approaches often suffer from significant `foreground bias', where models tend to wrongly identify background regions as foreground objects. To alleviate this issue, we propose DenseVLM, a framework designed to learn unbiased region-language alignment from powerful pre-trained VLM representations. To alleviate this issue, we propose DenseVLM, a framework designed to learn unbiased region-language alignment from powerful pre-trained VLM representations. DenseVLM leverages the pre-trained VLM to retrieve categories for unlabeled regions and then decouples the interference between foreground and background features. We show that DenseVLM can directly replace the original VLM in open-vocabulary object detection and image segmentation methods, leading to notable performance improvements. Furthermore, it exhibits promising zero-shot scalability when training on more extensive and diverse datasets. Our code is available at https://github.com/HVision-NKU/DenseVLM.

Paperid: 1558, https://arxiv.org/pdf/2412.06224.pdf GitHub

Abstract:
A practical navigation agent must be capable of handling a wide range of interaction demands, such as following instructions, searching objects, answering questions, tracking people, and more. Existing models for embodied navigation fall short of serving as practical generalists in the real world, as they are often constrained by specific task configurations or pre-defined maps with discretized waypoints. In this work, we present Uni-NaVid, the first video-based vision-language-action (VLA) model designed to unify diverse embodied navigation tasks and enable seamless navigation for mixed long-horizon tasks in unseen real-world environments. Uni-NaVid achieves this by harmonizing the input and output data configurations for all commonly used embodied navigation tasks and thereby integrating all tasks in one model. For training Uni-NaVid, we collect 3.6 million navigation data samples in total from four essential navigation sub-tasks and foster synergy in learning across them. Extensive experiments on comprehensive navigation benchmarks clearly demonstrate the advantages of unification modeling in Uni-NaVid and show it achieves state-of-the-art performance. Additionally, real-world experiments confirm the model's effectiveness and efficiency, shedding light on its strong generalizability.

Paperid: 1559, https://arxiv.org/pdf/2412.06141.pdf GitHub

Abstract:
The advancement of Large Vision-Language Models (LVLMs) has propelled their application in the medical field. However, Medical LVLMs (Med-LVLMs) encounter factuality challenges due to modality misalignment, where the models prioritize textual knowledge over visual input, leading to hallucinations that contradict information in medical images. Previous attempts to enhance modality alignment in Med-LVLMs through preference optimization have inadequately mitigated clinical relevance in preference data, making these samples easily distinguishable and reducing alignment effectiveness. To address this challenge, we propose MMedPO, a novel multimodal medical preference optimization approach that considers the clinical relevance of preference samples to enhance Med-LVLM alignment. MMedPO curates multimodal preference data by introducing two types of dispreference: (1) plausible hallucinations injected through target Med-LVLMs or GPT-4o to produce medically inaccurate responses, and (2) lesion region neglect achieved through local lesion-noising, disrupting visual understanding of critical areas. We then calculate clinical relevance for each sample based on scores from multiple Med-LLMs and visual tools, and integrate these scores into the preference optimization process as weights, enabling effective alignment. Our experiments demonstrate that MMedPO significantly enhances factual accuracy in Med-LVLMs, achieving substantial improvements over existing preference optimization methods by averaging 14.2% and 51.7% across the Med-VQA and report generation tasks. Our code are available in https://github.com/aiming-lab/MMedPO.

Paperid: 1560, https://arxiv.org/pdf/2412.06014.pdf GitHub

Abstract:
Vision-language models (VLMs), such as CLIP and SigLIP, have found remarkable success in classification, retrieval, and generative tasks. For this, VLMs deterministically map images and text descriptions to a joint latent space in which their similarity is assessed using the cosine similarity. However, a deterministic mapping of inputs fails to capture uncertainties over concepts arising from domain shifts when used in downstream tasks. In this work, we propose post-hoc uncertainty estimation in VLMs that does not require additional training. Our method leverages a Bayesian posterior approximation over the last layers in VLMs and analytically quantifies uncertainties over cosine similarities. We demonstrate its effectiveness for uncertainty quantification and support set selection in active learning. Compared to baselines, we obtain improved and well-calibrated predictive uncertainties, interpretable uncertainty estimates, and sample-efficient active learning. Our results show promise for safety-critical applications of large-scale models.

Paperid: 1561, https://arxiv.org/pdf/2412.05939.pdf GitHub

Abstract:
Multimodal Large Language Models (MLLMs) excel in vision--language tasks by pre-training solely on coarse-grained concept annotations (e.g., image captions). We hypothesize that integrating fine-grained concept annotations (e.g., object labels and object regions) will further improve performance, as both data granularities complement each other in terms of breadth and depth in concept representation. We introduce a new dataset featuring Multimodal Multi-Grained Concept annotations (MMGiC) for MLLMs. In constructing MMGiC, we explore the impact of different data recipes on multimodal comprehension and generation. Our analyses reveal that multi-grained concept annotations integrate and complement each other, under our structured template and a general MLLM framework. We clearly explore and demonstrate the potential of MMGiC to help MLLMs better locate and learn concepts, aligning vision and language at multiple granularities. We further validate our hypothesis by investigating the fair comparison and effective collaboration between MMGiC and image--caption data on 12 multimodal comprehension and generation benchmarks, e.g., their appropriate combination achieve 3.95% and 2.34% absolute improvements over image--caption data alone on POPE and SEED-Bench. Code, data and models will be available at https://github.com/LooperXX/MMGiC.

Paperid: 1562, https://arxiv.org/pdf/2412.05876.pdf GitHub

Abstract:
3D medical image analysis is pivotal in numerous clinical applications. However, the scarcity of labeled data and limited generalization capabilities hinder the advancement of AI-empowered models. Radiology reports are easily accessible and can serve as weakly-supervised signals. However, large-scale vision-language pre-training (VLP) remains underexplored in 3D medical image analysis. Specifically, the insufficient investigation into multi-grained radiology semantics and their correlations across patients leads to underutilization of large-scale volume-report data. Considering intra-patient cross-modal semantic consistency and inter-patient semantic correlations, we propose a multi-task VLP method, MG-3D, pre-trained on large-scale data (47.1K), addressing the challenges by the following two aspects: 1) Establishing the correspondence between volume semantics and multi-grained medical knowledge of each patient with cross-modal global alignment and complementary modality-guided local reconstruction, ensuring intra-patient features of different modalities cohesively represent the same semantic content; 2) Correlating inter-patient visual semantics based on fine-grained report correlations across patients, and keeping sensitivity to global individual differences via contrastive learning, enhancing the discriminative feature representation. Furthermore, we delve into the scaling law to explore potential performance improvements. Comprehensive evaluations across nine uni- and cross-modal clinical tasks are carried out to assess model efficacy. Extensive experiments on both internal and external datasets demonstrate the superior transferability, scalability, and generalization of MG-3D, showcasing its potential in advancing feature representation for 3D medical image analysis. Code will be available: https://github.com/Xuefeng-Ni/MG-3D.

Paperid: 1563, https://arxiv.org/pdf/2412.05819.pdf GitHub

Abstract:
Multimodal Large Language Models (MLLMs) have recently demonstrated strong performance across a wide range of vision-language tasks, garnering significant attention in the computer vision. However, their efficient deployment remains a substantial challenge due to high computational costs and memory requirements. Recognizing the redundancy of information within the vision modality, recent studies have explored methods for compressing visual tokens in MLLMs to enhance efficiency in a training-free manner. Despite their effectiveness, existing methods like Fast rely on the attention between visual tokens and prompt text tokens as the importance indicator, overlooking the relevance to response text and thus introducing perception bias. In this paper, we demonstrate that in MLLMs, the [CLS] token in the visual encoder inherently knows which visual tokens are important for MLLMs. Building on this prior, we introduce a simple yet effective method for train-free visual token compression, called VTC-CLS. Firstly, it leverages the attention score of the [CLS] token on visual tokens as an importance indicator for pruning visual tokens. Besides, we also explore ensembling the importance scores derived by the [CLS] token from different layers to capture the key visual information more comprehensively. Extensive experiments demonstrate that our VTC-CLS achieves the state-of-the-art performance across various tasks compared with baseline methods. It also brings notably less computational costs in a training-free manner, highlighting its effectiveness and superiority. Code and models are available at \url{https://github.com/THU-MIG/VTC-CLS}.

Paperid: 1564, https://arxiv.org/pdf/2412.05679.pdf GitHub GitHub

Abstract:
Remote Sensing Vision-Language Models (RS VLMs) have made much progress in the tasks of remote sensing (RS) image comprehension. While performing well in multi-modal reasoning and multi-turn conversations, the existing models lack pixel-level understanding and struggle with multi-image inputs. In this work, we propose RSUniVLM, a unified, end-to-end RS VLM designed for comprehensive vision understanding across multiple granularity, including image-level, region-level, and pixel-level tasks. RSUniVLM also performs effectively in multi-image analysis, with instances of change detection and change captioning. To enhance the model's ability to capture visual information at different levels without increasing model size, we design a novel architecture called Granularity-oriented Mixture of Experts to constraint the model to about 1 billion parameters. We also construct a large-scale RS instruction-following dataset based on a variety of existing datasets in both RS and general domain, encompassing various tasks such as object localization, visual question answering, and semantic segmentation. Substantial experiments have been conducted to validate the superiority of the proposed RSUniVLM up to state-of-the-art across various RS tasks. Code and model will be available at \href{https://github.com/xuliu-cyber/RSUniVLM}{here}.

Paperid: 1565, https://arxiv.org/pdf/2412.05293.pdf GitHub

Abstract:
Out-of-Distribution (OOD) detection is crucial when deploying machine learning models in open-world applications. The core challenge in OOD detection is mitigating the model's overconfidence on OOD data. While recent methods using auxiliary outlier datasets or synthesizing outlier features have shown promising OOD detection performance, they are limited due to costly data collection or simplified assumptions. In this paper, we propose a novel OOD detection framework FodFoM that innovatively combines multiple foundation models to generate two types of challenging fake outlier images for classifier training. The first type is based on BLIP-2's image captioning capability, CLIP's vision-language knowledge, and Stable Diffusion's image generation ability. Jointly utilizing these foundation models constructs fake outlier images which are semantically similar to but different from in-distribution (ID) images. For the second type, GroundingDINO's object detection ability is utilized to help construct pure background images by blurring foreground ID objects in ID images. The proposed framework can be flexibly combined with multiple existing OOD detection methods. Extensive empirical evaluations show that image classifiers with the help of constructed fake images can more accurately differentiate real OOD images from ID ones. New state-of-the-art OOD detection performance is achieved on multiple benchmarks. The code is available at \url{https://github.com/Cverchen/ACMMM2024-FodFoM}.

Paperid: 1566, https://arxiv.org/pdf/2412.04652.pdf GitHub

Abstract:
KV cache pruning has emerged as a promising technique for reducing memory and computation costs in long-context auto-regressive generation. Existing methods for vision-language models (VLMs) typically rely on self-attention scores from large language models (LLMs) to identify and prune irrelevant tokens. However, these approaches overlook the inherent distributional discrepancies between modalities, often leading to inaccurate token importance estimation and the over-pruning of critical visual tokens. To address this, we propose decomposing attention scores into intra-modality attention (within the same modality) and inter-modality attention (across modalities), enabling more precise KV cache pruning by independently managing these distinct attention types. Additionally, we introduce an n-softmax function to counteract distribution shifts caused by pruning, preserving the original smoothness of attention scores and ensuring stable performance. Our final training-free method, \textbf{C}ross-\textbf{S}elf \textbf{P}runing (CSP), achieves competitive performance compared to models with full KV caches while significantly outperforming previous pruning methods. Extensive evaluations on MileBench, a benchmark encompassing 29 multimodal datasets, demonstrate CSP's effectiveness, achieving up to a 41\% performance improvement on challenging tasks like conversational embodied dialogue while reducing the KV cache budget by 13.6\%. The code is available at https://github.com/TerryPei/CSP

Paperid: 1567, https://arxiv.org/pdf/2412.04616.pdf GitHub

Abstract:
How well are unimodal vision and language models aligned? Although prior work have approached answering this question, their assessment methods do not directly translate to how these models are used in practical vision-language tasks. In this paper, we propose a direct assessment method, inspired by linear probing, to assess vision-language alignment. We identify that the degree of alignment of the SSL vision models depends on their SSL training objective, and we find that the clustering quality of SSL representations has a stronger impact on alignment performance than their linear separability. Next, we introduce Swift Alignment of Image and Language (SAIL), a efficient transfer learning framework that aligns pretrained unimodal vision and language models for downstream vision-language tasks. Since SAIL leverages the strengths of pretrained unimodal models, it requires significantly fewer (6%) paired image-text data for the multimodal alignment compared to models like CLIP which are trained from scratch. SAIL training only requires a single A100 GPU, 5 hours of training and can accommodate a batch size up to 32,768. SAIL achieves 73.4% zero-shot accuracy on ImageNet (vs. CLIP's 72.7%) and excels in zero-shot retrieval, complex reasoning, and semantic segmentation. Additionally, SAIL improves the language-compatibility of vision encoders that in turn enhance the performance of multimodal large language models. The entire codebase and model weights are open-source: https://lezhang7.github.io/sail.github.io/

Paperid: 1568, https://arxiv.org/pdf/2412.04533.pdf GitHub GitHub

Abstract:
Recent open-vocabulary segmentation methods adopt mask generators to predict segmentation masks and leverage pre-trained vision-language models, e.g., CLIP, to classify these masks via mask pooling. Although these approaches show promising results, it is counterintuitive that accurate masks often fail to yield accurate classification results through pooling CLIP image embeddings within the mask regions. In this paper, we reveal the performance limitations of mask pooling and introduce Mask-Adapter, a simple yet effective method to address these challenges in open-vocabulary segmentation. Compared to directly using proposal masks, our proposed Mask-Adapter extracts semantic activation maps from proposal masks, providing richer contextual information and ensuring alignment between masks and CLIP. Additionally, we propose a mask consistency loss that encourages proposal masks with similar IoUs to obtain similar CLIP embeddings to enhance models' robustness to varying predicted masks. Mask-Adapter integrates seamlessly into open-vocabulary segmentation methods based on mask pooling in a plug-and-play manner, delivering more accurate classification results. Extensive experiments across several zero-shot benchmarks demonstrate significant performance gains for the proposed Mask-Adapter on several well-established methods. Notably, Mask-Adapter also extends effectively to SAM and achieves impressive results on several open-vocabulary segmentation datasets. Code and models are available at https://github.com/hustvl/MaskAdapter.

Paperid: 1569, https://arxiv.org/pdf/2412.04467.pdf GitHub

Abstract:
Recent advancements in vision-language models have enhanced performance by increasing the length of visual tokens, making them much longer than text tokens and significantly raising computational costs. However, we observe that the visual tokens generated by popular vision encoders, such as CLIP and SigLIP, contain significant redundancy. To address this, we introduce VisionZip, a simple yet effective method that selects a set of informative tokens for input to the language model, reducing visual token redundancy and improving efficiency while maintaining model performance. The proposed VisionZip can be widely applied to image and video understanding tasks and is well-suited for multi-turn dialogues in real-world scenarios, where previous methods tend to underperform. Experimental results show that VisionZip outperforms the previous state-of-the-art method by at least 5% performance gains across nearly all settings. Moreover, our method significantly enhances model inference speed, improving the prefilling time by 8x and enabling the LLaVA-Next 13B model to infer faster than the LLaVA-Next 7B model while achieving better results. Furthermore, we analyze the causes of this redundancy and encourage the community to focus on extracting better visual features rather than merely increasing token length. Our code is available at https://github.com/dvlab-research/VisionZip .

Paperid: 1570, https://arxiv.org/pdf/2412.04455.pdf GitHub

Abstract:
Automatic detection and prevention of open-set failures are crucial in closed-loop robotic systems. Recent studies often struggle to simultaneously identify unexpected failures reactively after they occur and prevent foreseeable ones proactively. To this end, we propose Code-as-Monitor (CaM), a novel paradigm leveraging the vision-language model (VLM) for both open-set reactive and proactive failure detection. The core of our method is to formulate both tasks as a unified set of spatio-temporal constraint satisfaction problems and use VLM-generated code to evaluate them for real-time monitoring. To enhance the accuracy and efficiency of monitoring, we further introduce constraint elements that abstract constraint-related entities or their parts into compact geometric elements. This approach offers greater generality, simplifies tracking, and facilitates constraint-aware visual programming by leveraging these elements as visual prompts. Experiments show that CaM achieves a 28.7% higher success rate and reduces execution time by 31.8% under severe disturbances compared to baselines across three simulators and a real-world setting. Moreover, CaM can be integrated with open-loop control policies to form closed-loop systems, enabling long-horizon tasks in cluttered scenes with dynamic environments.

Paperid: 1571, https://arxiv.org/pdf/2412.04429.pdf GitHub

Abstract:
Vision-language models (VLMs) like CLIP have been cherished for their ability to perform zero-shot visual recognition on open-vocabulary concepts. This is achieved by selecting the object category whose textual representation bears the highest similarity with the query image. While successful in some domains, this method struggles with identifying fine-grained entities as well as generalizing to unseen concepts that are not captured by the training distribution. Recent works attempt to mitigate these challenges by integrating category descriptions at test time, albeit yielding modest improvements. We attribute these limited gains to a fundamental misalignment between image and description representations, which is rooted in the pretraining structure of CLIP. In this paper, we propose GRAIN, a new pretraining strategy aimed at aligning representations at both fine and coarse levels simultaneously. Our approach learns to jointly ground textual descriptions in image regions along with aligning overarching captions with global image representations. To drive this pre-training, we leverage frozen Multimodal Large Language Models (MLLMs) to derive large-scale synthetic annotations. We demonstrate the enhanced zero-shot performance of our model compared to current state-of-the art methods across 11 diverse image classification datasets. Additionally, we introduce Products-2023, a newly curated, manually labeled dataset featuring novel concepts, and showcase our model's ability to recognize these concepts by benchmarking on it. Significant improvements achieved by our model on other downstream tasks like retrieval further highlight the superior quality of representations learned by our approach. Code available at https://github.com/shaunak27/grain-clip .

Paperid: 1572, https://arxiv.org/pdf/2412.04424.pdf GitHub

Abstract:
We present Florence-VL, a new family of multimodal large language models (MLLMs) with enriched visual representations produced by Florence-2, a generative vision foundation model. Unlike the widely used CLIP-style vision transformer trained by contrastive learning, Florence-2 can capture different levels and aspects of visual features, which are more versatile to be adapted to diverse downstream tasks. We propose a novel feature-fusion architecture and an innovative training recipe that effectively integrates Florence-2's visual features into pretrained LLMs, such as Phi 3.5 and LLama 3. In particular, we propose "depth-breath fusion (DBFusion)" to fuse the visual features extracted from different depths and under multiple prompts. Our model training is composed of end-to-end pretraining of the whole model followed by finetuning of the projection layer and the LLM, on a carefully designed recipe of diverse open-source datasets that include high-quality image captions and instruction-tuning pairs. Our quantitative analysis and visualization of Florence-VL's visual features show its advantages over popular vision encoders on vision-language alignment, where the enriched depth and breath play important roles. Florence-VL achieves significant improvements over existing state-of-the-art MLLMs across various multi-modal and vision-centric benchmarks covering general VQA, perception, hallucination, OCR, Chart, knowledge-intensive understanding, etc. To facilitate future research, our models and the complete training recipe are open-sourced. https://github.com/JiuhaiChen/Florence-VL

Paperid: 1573, https://arxiv.org/pdf/2412.04332.pdf GitHub GitHub

Abstract:
We present Liquid, an auto-regressive generation paradigm that seamlessly integrates visual comprehension and generation by tokenizing images into discrete codes and learning these code embeddings alongside text tokens within a shared feature space for both vision and language. Unlike previous multimodal large language model (MLLM), Liquid achieves this integration using a single large language model (LLM), eliminating the need for external pretrained visual embeddings such as CLIP. For the first time, Liquid uncovers a scaling law that performance drop unavoidably brought by the unified training of visual and language tasks diminishes as the model size increases. Furthermore, the unified token space enables visual generation and comprehension tasks to mutually enhance each other, effectively removing the typical interference seen in earlier models. We show that existing LLMs can serve as strong foundations for Liquid, saving 100x in training costs while outperforming Chameleon in multimodal capabilities and maintaining language performance comparable to mainstream LLMs like LLAMA2. Liquid also outperforms models like SD v2.1 and SD-XL (FID of 5.47 on MJHQ-30K), excelling in both vision-language and text-only tasks. This work demonstrates that LLMs such as Qwen2.5 and GEMMA2 are powerful multimodal generators, offering a scalable solution for enhancing both vision-language understanding and generation. The code and models will be released at https://github.com/FoundationVision/Liquid.

Paperid: 1574, https://arxiv.org/pdf/2412.03704.pdf GitHub

Abstract:
Despite significant advancements in vision-language models (VLMs), there lacks effective approaches to enhance response quality by scaling inference-time computation. This capability is known to be a core step towards the self-improving models in recent large language model studies. In this paper, we present Vision Value Model (VisVM) that can guide VLM inference-time search to generate responses with better visual comprehension. Specifically, VisVM not only evaluates the generated sentence quality in the current search step, but also anticipates the quality of subsequent sentences that may result from the current step, thus providing a long-term value. In this way, VisVM steers VLMs away from generating sentences prone to hallucinations or insufficient detail, thereby producing higher quality responses. Experimental results demonstrate that VisVM-guided search significantly enhances VLMs' ability to generate descriptive captions with richer visual details and fewer hallucinations, compared with greedy decoding and search methods with other visual reward signals. Furthermore, we find that self-training the model with the VisVM-guided captions improve VLM's performance across a wide range of multimodal benchmarks, indicating the potential for developing self-improving VLMs. Our value model and code are available at https://github.com/si0wang/VisVM.

Paperid: 1575, https://arxiv.org/pdf/2412.03561.pdf GitHub

Abstract:
CLIP has shown impressive results in aligning images and texts at scale. However, its ability to capture detailed visual features remains limited because CLIP matches images and texts at a global level. To address this issue, we propose FLAIR, Fine-grained Language-informed Image Representations, an approach that utilizes long and detailed image descriptions to learn localized image embeddings. By sampling diverse sub-captions that describe fine-grained details about an image, we train our vision-language model to produce not only global embeddings but also text-specific image representations. Our model introduces text-conditioned attention pooling on top of local image tokens to produce fine-grained image representations that excel at retrieving detailed image content. We achieve state-of-the-art performance on both, existing multimodal retrieval benchmarks, as well as, our newly introduced fine-grained retrieval task which evaluates vision-language models' ability to retrieve partial image content. Furthermore, our experiments demonstrate the effectiveness of FLAIR trained on 30M image-text pairs in capturing fine-grained visual information, including zero-shot semantic segmentation, outperforming models trained on billions of pairs. Code is available at https://github.com/ExplainableML/flair .

Paperid: 1576, https://arxiv.org/pdf/2412.03409.pdf GitHub

Abstract:
Recently, large vision-language models (LVLMs) have rapidly gained popularity for their strong generation and reasoning capabilities given diverse multimodal inputs. However, these models incur significant computational and memory overhead during inference, which greatly hinders the efficient deployment in practical scenarios. The extensive key-value (KV) cache, necessitated by the lengthy input and output sequences, notably contributes to the high inference cost. Based on this, recent works have investigated ways to reduce the KV cache size for higher efficiency. Although effective, they generally overlook the distinct importance distributions of KV vectors across layers and maintain the same cache size for each layer during the next token prediction. This results in the significant contextual information loss for certain layers, leading to notable performance decline. To address this, we present PrefixKV. It reframes the challenge of determining KV cache sizes for all layers into the task of searching for the optimal global prefix configuration. With an adaptive layer-wise KV retention recipe based on binary search, the maximum contextual information can thus be preserved in each layer, facilitating the generation. Extensive experiments demonstrate that our method achieves the state-of-the-art performance compared with others. It exhibits superior inference efficiency and generation quality trade-offs, showing promising potential for practical applications. Code is available at https://github.com/THU-MIG/PrefixKV.

Paperid: 1577, https://arxiv.org/pdf/2412.03297.pdf GitHub

Abstract:
This work addresses composed image retrieval in the context of domain conversion, where the content of a query image is retrieved in the domain specified by the query text. We show that a strong vision-language model provides sufficient descriptive power without additional training. The query image is mapped to the text input space using textual inversion. Unlike common practice that invert in the continuous space of text tokens, we use the discrete word space via a nearest-neighbor search in a text vocabulary. With this inversion, the image is softly mapped across the vocabulary and is made more robust using retrieval-based augmentation. Database images are retrieved by a weighted ensemble of text queries combining mapped words with the domain text. Our method outperforms prior art by a large margin on standard and newly introduced benchmarks. Code: https://github.com/NikosEfth/freedom

Paperid: 1578, https://arxiv.org/pdf/2412.02573.pdf GitHub

Abstract:
The interpretation of multi-temporal remote sensing imagery is critical for monitoring Earth's dynamic processes-yet previous change detection methods, which produce binary or semantic masks, fall short of providing human-readable insights into changes. Recent advances in Vision-Language Models (VLMs) have opened a new frontier by fusing visual and linguistic modalities, enabling spatio-temporal vision-language understanding: models that not only capture spatial and temporal dependencies to recognize changes but also provide a richer interactive semantic analysis of temporal images (e.g., generate descriptive captions and answer natural-language queries). In this survey, we present the first comprehensive review of RS-STVLMs. The survey covers the evolution of models from early task-specific models to recent general foundation models that leverage powerful large language models. We discuss progress in representative tasks, such as change captioning, change question answering, and change grounding. Moreover, we systematically dissect the fundamental components and key technologies underlying these models, and review the datasets and evaluation metrics that have driven the field. By synthesizing task-level insights with a deep dive into shared architectural patterns, we aim to illuminate current achievements and chart promising directions for future research in spatio-temporal vision-language understanding for remote sensing. We will keep tracing related works at https://github.com/Chen-Yang-Liu/Awesome-RS-SpatioTemporal-VLMs

Paperid: 1579, https://arxiv.org/pdf/2412.02565.pdf GitHub

Abstract:
Despite significant advances in vision-language understanding, implementing image segmentation within multimodal architectures remains a fundamental challenge in modern artificial intelligence systems. Existing vision-language models, which primarily rely on backbone architectures or CLIP-based embedding learning, demonstrate inherent limitations in fine-grained spatial localization and operational capabilities. This paper introduces SJTU: Spatial Judgments in Multimodal Models - Towards Unified Segmentation through Coordinate Detection, a framework that leverages spatial coordinate understanding to bridge vision-language interaction and precise segmentation, enabling accurate target identification through natural language instructions. The framework presents an approach for integrating segmentation techniques with vision-language models through spatial inference in multimodal space. By utilizing normalized coordinate detection for bounding boxes and transforming them into actionable segmentation outputs, we establish a connection between spatial and language representations in multimodal architectures. Experimental results demonstrate superior performance across benchmark datasets, achieving IoU scores of 0.5958 on COCO 2017 and 0.6758 on Pascal VOC. Testing on a single NVIDIA RTX 3090 GPU with 512x512 resolution images yields an average inference time of 7 seconds per image, demonstrating the framework's effectiveness in both accuracy and practical deployability. The project code is available at https://github.com/jw-chae/SJTU

Paperid: 1580, https://arxiv.org/pdf/2412.02531.pdf GitHub

Abstract:
Remote sensing scene classification (RSSC) is a critical task with diverse applications in land use and resource management. While unimodal image-based approaches show promise, they often struggle with limitations such as high intra-class variance and inter-class similarity. Incorporating textual information can enhance classification by providing additional context and semantic understanding, but manual text annotation is labor-intensive and costly. In this work, we propose a novel RSSC framework that integrates text descriptions generated by large vision-language models (VLMs) as an auxiliary modality without incurring expensive manual annotation costs. To fully leverage the latent complementarities between visual and textual data, we propose a dual cross-attention-based network to fuse these modalities into a unified representation. Extensive experiments with both quantitative and qualitative evaluation across five RSSC datasets demonstrate that our framework consistently outperforms baseline models. We also verify the effectiveness of VLM-generated text descriptions compared to human-annotated descriptions. Additionally, we design a zero-shot classification scenario to show that the learned multimodal representation can be effectively utilized for unseen class classification. This research opens new opportunities for leveraging textual information in RSSC tasks and provides a promising multimodal fusion structure, offering insights and inspiration for future studies. Code is available at: https://github.com/CJR7/MultiAtt-RSSC

Paperid: 1581, https://arxiv.org/pdf/2412.01824.pdf GitHub

Abstract:
In-context generation is a key component of large language models' (LLMs) open-task generalization capability. By leveraging a few examples as context, LLMs can perform both in-domain and out-of-domain tasks. Recent advancements in auto-regressive vision-language models (VLMs) built upon LLMs have showcased impressive performance in text-to-image generation. However, the potential of in-context learning for general image generation tasks remains largely unexplored. To address this, we introduce X-Prompt, a purely auto-regressive large-vision language model designed to deliver competitive performance across a wide range of both seen and unseen image generation tasks, all within a unified in-context learning framework. X-Prompt incorporates a specialized design that efficiently compresses valuable features from in-context examples, supporting longer in-context token sequences and improving its ability to generalize to unseen tasks. A unified training task for both text and image prediction enables X-Prompt to handle general image generation with enhanced task awareness from in-context examples. Extensive experiments validate the model's performance across diverse seen image generation tasks and its capacity to generalize to previously unseen tasks.

Paperid: 1582, https://arxiv.org/pdf/2412.01822.pdf GitHub

Abstract:
The recent surge in high-quality visual instruction tuning samples from closed-source vision-language models (VLMs) such as GPT-4V has accelerated the release of open-source VLMs across various model sizes. However, scaling VLMs to improve performance using larger models brings significant computational challenges, especially for deployment on resource-constrained devices like mobile platforms and robots. To address this, we propose VLsI: Verbalized Layers-to-Interactions, a new VLM family in 2B and 7B model sizes, which prioritizes efficiency without compromising accuracy. VLsI leverages a unique, layer-wise distillation process, introducing intermediate "verbalizers" that map features from each layer to natural language space, allowing smaller VLMs to flexibly align with the reasoning processes of larger VLMs. This approach mitigates the training instability often encountered in output imitation and goes beyond typical final-layer tuning by aligning the small VLMs' layer-wise progression with that of the large ones. We validate VLsI across ten challenging vision-language benchmarks, achieving notable performance gains (11.0% for 2B and 17.4% for 7B) over GPT-4V without the need for model scaling, merging, or architectural changes.

Paperid: 1583, https://arxiv.org/pdf/2412.01818.pdf GitHub GitHub GitHub

Abstract:
Large vision-language models (LVLMs) generally contain significantly more visual tokens than their textual counterparts, resulting in a considerable computational burden. Recent efforts have been made to tackle this issue by pruning visual tokens early within the language model. Most existing works use attention scores between text and visual tokens to assess the importance of visual tokens. However, in this study, we first analyze the text-visual attention in the language model and find that this score is not an ideal indicator for token pruning. Based on the analysis, We propose VisPruner, a plug-and-play method that utilizes visual cues for more effective token pruning in LVLMs. Specifically, we first use visual attention to select a limited number of significant tokens. Then, we remove duplicate tokens from the remaining ones based on their similarity. By retaining diverse tokens alongside the initially selected important tokens, we maximally preserve the visual information of the input image. Experimental results demonstrate that our VisPruner sustains strong performance across various VLM architectures and reduction ratios, significantly outperforming existing methods based on text-visual attention. Notably, without any training, VisPruner can reduce the FLOPs of LLaVA-1.5-7B by 91% and inference latency by 75%, while maintaining comparable performance. Our code is available at https://github.com/Theia-4869/VisPruner.

Paperid: 1584, https://arxiv.org/pdf/2412.01814.pdf GitHub

Abstract:
Vision-Language Models (VLMs) trained with contrastive loss have achieved significant advancements in various vision and language tasks. However, the global nature of the contrastive loss makes VLMs focus predominantly on foreground objects, neglecting other crucial information in the image, which limits their effectiveness in downstream tasks. To address these challenges, we propose COSMOS: CrOSs-MOdality Self-distillation for vision-language pre-training that integrates a novel text-cropping strategy and cross-attention module into a self-supervised learning framework. We create global and local views of images and texts (i.e., multi-modal augmentations), which are essential for self-distillation in VLMs. We further introduce a cross-attention module, enabling COSMOS to learn comprehensive cross-modal representations optimized via a cross-modality self-distillation loss. COSMOS consistently outperforms previous strong baselines on various zero-shot downstream tasks, including retrieval, classification, and semantic segmentation. Additionally, it surpasses CLIP-based models trained on larger datasets in visual perception and contextual understanding tasks. Code is available at https://github.com/ExplainableML/cosmos.

Paperid: 1585, https://arxiv.org/pdf/2412.01807.pdf GitHub

Abstract:
TL;DR: Gaussian Splatting is a widely adopted approach for 3D scene representation, offering efficient, high-quality reconstruction and rendering. A key reason for its success is the simplicity of representing scenes with sets of Gaussians, making it interpretable and adaptable. To enhance understanding beyond visual representation, recent approaches extend Gaussian Splatting with semantic vision-language features, enabling open-set tasks. Typically, these language features are aggregated from multiple 2D views, however, existing methods rely on cumbersome techniques, resulting in high computational costs and longer training times. In this work, we show that the complicated pipelines for language 3D Gaussian Splatting are simply unnecessary. Instead, we follow a probabilistic formulation of Language Gaussian Splatting and apply Occam's razor to the task at hand, leading to a highly efficient weighted multi-view feature aggregation technique. Doing so offers us state-of-the-art results with a speed-up of two orders of magnitude without any compression, allowing for easy scene manipulation. Project Page: https://insait-institute.github.io/OccamLGS/

Paperid: 1586, https://arxiv.org/pdf/2412.01558.pdf GitHub

Abstract:
Video Highlight Detection and Moment Retrieval (HD/MR) are essential in video analysis. Recent joint prediction transformer models often overlook their cross-task dynamics and video-text alignment and refinement. Moreover, most models typically use limited, uni-directional attention mechanisms, resulting in weakly integrated representations and suboptimal performance in capturing the interdependence between video and text modalities. Although large-language and vision-language models (LLM/LVLMs) have gained prominence across various domains, their application in this field remains relatively underexplored. Here we propose VideoLights, a novel HD/MR framework addressing these limitations through (i) Convolutional Projection and Feature Refinement modules with an alignment loss for better video-text feature alignment, (ii) Bi-Directional Cross-Modal Fusion network for strongly coupled query-aware clip representations, and (iii) Uni-directional joint-task feedback mechanism enhancing both tasks through correlation. In addition, (iv) we introduce hard positive/negative losses for adaptive error penalization and improved learning, and (v) leverage LVLMs like BLIP-2 for enhanced multimodal feature integration and intelligent pretraining using synthetic data generated from LVLMs. Comprehensive experiments on QVHighlights, TVSum, and Charades-STA benchmarks demonstrate state-of-the-art performance. Codes and models are available at https://github.com/dpaul06/VideoLights .

Paperid: 1587, https://arxiv.org/pdf/2412.01284.pdf GitHub

Abstract:
Text-to-image generation models have revolutionized content creation, but diffusion-based vision-language models still face challenges in precisely controlling the shape, appearance, and positional placement of objects in generated images using text guidance alone. Existing global image editing models rely on additional masks or images as guidance to achieve layout control, often requiring retraining of the model. While local object-editing models allow modifications to object shapes, they lack the capability to control object positions. To address these limitations, we propose the Mask-free Training-free Object-Level Layout Control Diffusion Model (MFTF), which provides precise control over object positions without requiring additional masks or images. The MFTF model supports both single-object and multi-object positional adjustments, such as translation and rotation, while enabling simultaneous layout control and object semantic editing. The MFTF model employs a parallel denoising process for both the source and target diffusion models. During this process, attention masks are dynamically generated from the cross-attention layers of the source diffusion model and applied to queries from the self-attention layers to isolate objects. These queries, generated in the source diffusion model, are then adjusted according to the layout control parameters and re-injected into the self-attention layers of the target diffusion model. This approach ensures accurate and precise positional control of objects. Project source code available at https://github.com/syang-genai/MFTF.

Paperid: 1588, https://arxiv.org/pdf/2412.01282.pdf GitHub

Abstract:
Vision-Language Models (VLMs) bring powerful understanding and reasoning capabilities to multimodal tasks. Meanwhile, the great need for capable aritificial intelligence on mobile devices also arises, such as the AI assistant software. Some efforts try to migrate VLMs to edge devices to expand their application scope. Simplifying the model structure is a common method, but as the model shrinks, the trade-off between performance and size becomes more and more difficult. Knowledge distillation (KD) can help models improve comprehensive capabilities without increasing size or data volume. However, most of the existing large model distillation techniques only consider applications on single-modal LLMs, or only use teachers to create new data environments for students. None of these methods take into account the distillation of the most important cross-modal alignment knowledge in VLMs. We propose a method called Align-KD to guide the student model to learn the cross-modal matching that occurs at the shallow layer. The teacher also helps student learn the projection of vision token into text embedding space based on the focus of text. Under the guidance of Align-KD, the 1.7B MobileVLM V2 model can learn rich knowledge from the 7B teacher model with light design of training loss, and achieve an average score improvement of 2.0 across 6 benchmarks under two training subsets respectively. Code is available at: https://github.com/fqhank/Align-KD.

Paperid: 1589, https://arxiv.org/pdf/2412.01250.pdf GitHub

Abstract:
Language-driven instance object navigation assumes that human users initiate the task by providing a detailed description of the target instance to the embodied agent. While this description is crucial for distinguishing the target from visually similar instances in a scene, providing it prior to navigation can be demanding for human. To bridge this gap, we introduce Collaborative Instance object Navigation (CoIN), a new task setting where the agent actively resolve uncertainties about the target instance during navigation in natural, template-free, open-ended dialogues with human. We propose a novel training-free method, Agent-user Interaction with UncerTainty Awareness (AIUTA), which operates independently from the navigation policy, and focuses on the human-agent interaction reasoning with Vision-Language Models (VLMs) and Large Language Models (LLMs). First, upon object detection, a Self-Questioner model initiates a self-dialogue within the agent to obtain a complete and accurate observation description with a novel uncertainty estimation technique. Then, an Interaction Trigger module determines whether to ask a question to the human, continue or halt navigation, minimizing user input. For evaluation, we introduce CoIN-Bench, with a curated dataset designed for challenging multi-instance scenarios. CoIN-Bench supports both online evaluation with humans and reproducible experiments with simulated user-agent interactions. On CoIN-Bench, we show that AIUTA serves as a competitive baseline, while existing language-driven instance navigation methods struggle in complex multi-instance scenes. Code and benchmark will be available upon acceptance at https://intelligolabs.github.io/CoIN/

Paperid: 1590, https://arxiv.org/pdf/2412.01132.pdf GitHub

Abstract:
Recent advances in video question answering (VideoQA) offer promising applications, especially in traffic monitoring, where efficient video interpretation is critical. Within ITS, answering complex, real-time queries like "How many red cars passed in the last 10 minutes?" or "Was there an incident between 3:00 PM and 3:05 PM?" enhances situational awareness and decision-making. Despite progress in vision-language models, VideoQA remains challenging, especially in dynamic environments involving multiple objects and intricate spatiotemporal relationships. This study evaluates state-of-the-art VideoQA models using non-benchmark synthetic and real-world traffic sequences. The framework leverages GPT-4o to assess accuracy, relevance, and consistency across basic detection, temporal reasoning, and decomposition queries. VideoLLaMA-2 excelled with 57% accuracy, particularly in compositional reasoning and consistent answers. However, all models, including VideoLLaMA-2, faced limitations in multi-object tracking, temporal coherence, and complex scene interpretation, highlighting gaps in current architectures. These findings underscore VideoQA's potential in traffic monitoring but also emphasize the need for improvements in multi-object tracking, temporal reasoning, and compositional capabilities. Enhancing these areas could make VideoQA indispensable for incident detection, traffic flow management, and responsive urban planning. The study's code and framework are open-sourced for further exploration: https://github.com/joe-rabbit/VideoQA_Pilot_Study

Paperid: 1591, https://arxiv.org/pdf/2412.00947.pdf GitHub

Abstract:
Large Vision Language Models (LVLMs) have achieved remarkable performance in various vision-language tasks. However, it is still unclear how accurately LVLMs can perceive visual information in images. In particular, the capability of LVLMs to perceive geometric information, such as shape, angle, and size, remains insufficiently analyzed, although the perception of these properties is crucial for tasks that require a detailed visual understanding. In this work, we introduce VisOnlyQA, a dataset for evaluating the geometric perception of LVLMs, and reveal that LVLMs often cannot accurately perceive basic geometric information in images, while human performance is nearly perfect. VisOnlyQA consists of 12 tasks that directly ask about geometric information in geometric shapes, charts, chemical structures, and 3D shapes. Our experiments highlight the following findings: (i) State-of-the-art LVLMs struggle with basic geometric perception. 23 LVLMs we evaluate, including GPT-4o and Gemini 2.5 Pro, work poorly on VisOnlyQA. (ii) Additional training data does not resolve this issue. Fine-tuning on the training set of VisOnlyQA is not always effective, even for in-distribution tasks. (iii) LLM may be the bottleneck. LVLMs using stronger LLMs exhibit better geometric perception on VisOnlyQA, while it does not require complex reasoning, suggesting that the way LVLMs process information from visual encoders is a bottleneck. The datasets, code, and model responses are provided at https://github.com/psunlpgroup/VisOnlyQA.

Paperid: 1592, https://arxiv.org/pdf/2412.00876.pdf GitHub

Abstract:
Multimodal Large Language Models (MLLMs) have achieved remarkable success in vision understanding, reasoning, and interaction. However, the inference computation and memory increase progressively with the generation of output tokens during decoding, directly affecting the efficacy of MLLMs. Existing methods attempt to reduce the vision context redundancy to achieve efficient MLLMs. Unfortunately, the efficiency benefits of the vision context reduction in the prefill stage gradually diminish during the decoding stage. To address this problem, we proposed a dynamic vision-language context sparsification framework Dynamic-LLaVA, which dynamically reduces the redundancy of vision context in the prefill stage and decreases the memory and computation overhead of the generated language context during decoding. Dynamic-LLaVA designs a tailored sparsification inference scheme for different inference modes, i.e., prefill, decoding with and without KV cache, to achieve efficient inference of MLLMs. In practice, Dynamic-LLaVA can reduce computation consumption by $\sim$75\% in the prefill stage. Meanwhile, throughout the entire generation process of MLLMs, Dynamic-LLaVA reduces the $\sim$50\% computation consumption under decoding without KV cache, while saving $\sim$50\% GPU memory overhead when decoding with KV cache, due to the vision-language context sparsification. Extensive experiments also demonstrate that Dynamic-LLaVA achieves efficient inference for MLLMs with negligible understanding and generation ability degradation or even performance gains compared to the full-context inference baselines. Code is available at https://github.com/Osilly/dynamic_llava .

Paperid: 1593, https://arxiv.org/pdf/2412.00727.pdf GitHub

Abstract:
Vision-Language models like CLIP have been shown to be highly effective at linking visual perception and natural language understanding, enabling sophisticated image-text capabilities, including strong retrieval and zero-shot classification performance. Their widespread use, as well as the fact that CLIP models are trained on image-text pairs from the web, make them both a worthwhile and relatively easy target for backdoor attacks. As training foundational models, such as CLIP, from scratch is very expensive, this paper focuses on cleaning potentially poisoned models via fine-tuning. We first show that existing cleaning techniques are not effective against simple structured triggers used in Blended or BadNet backdoor attacks, exposing a critical vulnerability for potential real-world deployment of these models. Then, we introduce PAR, Perturb and Recover, a surprisingly simple yet effective mechanism to remove backdoors from CLIP models. Through extensive experiments across different encoders and types of backdoor attacks, we show that PAR achieves high backdoor removal rate while preserving good standard performance. Finally, we illustrate that our approach is effective even only with synthetic text-image pairs, i.e. without access to real training data. The code and models are available at https://github.com/nmndeep/PerturbAndRecover.

Paperid: 1594, https://arxiv.org/pdf/2412.00622.pdf GitHub

Abstract:
The zero-shot performance of object detectors degrades when tested on different modalities, such as infrared and depth. While recent work has explored image translation techniques to adapt detectors to new modalities, these methods are limited to a single modality and apply only to traditional detectors. Recently, vision-language detectors, such as YOLO-World and Grounding DINO, have shown promising zero-shot capabilities, however, they have not yet been adapted for other visual modalities. Traditional fine-tuning approaches compromise the zero-shot capabilities of the detectors. The visual prompt strategies commonly used for classification with vision-language models apply the same linear prompt translation to each image, making them less effective. To address these limitations, we propose ModPrompt, a visual prompt strategy to adapt vision-language detectors to new modalities without degrading zero-shot performance. In particular, an encoder-decoder visual prompt strategy is proposed, further enhanced by the integration of inference-friendly modality prompt decoupled residual, facilitating a more robust adaptation. Empirical benchmarking results show our method for modality adaptation on two vision-language detectors, YOLO-World and Grounding DINO, and on challenging infrared (LLVIP, FLIR) and depth (NYUv2) datasets, achieving performance comparable to full fine-tuning while preserving the model's zero-shot capability. Code available at: https://github.com/heitorrapela/ModPrompt.

Paperid: 1595, https://arxiv.org/pdf/2412.00473.pdf GitHub

Abstract:
With the significant advancement of Large Vision-Language Models (VLMs), concerns about their potential misuse and abuse have grown rapidly. Previous studies have highlighted VLMs' vulnerability to jailbreak attacks, where carefully crafted inputs can lead the model to produce content that violates ethical and legal standards. However, existing methods struggle against state-of-the-art VLMs like GPT-4o, due to the over-exposure of harmful content and lack of stealthy malicious guidance. In this work, we propose a novel jailbreak attack framework: Multi-Modal Linkage (MML) Attack. Drawing inspiration from cryptography, MML utilizes an encryption-decryption process across text and image modalities to mitigate over-exposure of malicious information. To align the model's output with malicious intent covertly, MML employs a technique called "evil alignment", framing the attack within a video game production scenario. Comprehensive experiments demonstrate MML's effectiveness. Specifically, MML jailbreaks GPT-4o with attack success rates of 97.80% on SafeBench, 98.81% on MM-SafeBench and 99.07% on HADES-Dataset. Our code is available at https://github.com/wangyu-ovo/MML.

Paperid: 1596, https://arxiv.org/pdf/2412.00447.pdf GitHub

Abstract:
Large Vision Language Models (LVLMs) have achieved significant success across multi-modal tasks. However, the computational cost of processing long visual tokens can be prohibitively expensive on resource-limited devices. Previous methods have identified redundancy in visual tokens within the Large Language Model (LLM) decoder layers and have mitigated this by pruning tokens using a pre-defined or fixed ratio, thereby reducing computational overhead. Nonetheless, we observe that the impact of pruning ratio varies across different LLM layers and instances (image-prompt pairs). Therefore, it is essential to develop a layer-wise and instance-wise vision token pruning strategy to balance computational cost and model performance effectively. We propose ATP-LLaVA, a novel approach that adaptively determines instance-specific token pruning ratios for each LLM layer. Specifically, we introduce an Adaptive Token Pruning (ATP) module, which computes the importance score and pruning threshold based on input instance adaptively. The ATP module can be seamlessly integrated between any two LLM layers with negligible computational overhead. Additionally, we develop a Spatial Augmented Pruning (SAP) strategy that prunes visual tokens with both token redundancy and spatial modeling perspectives. Our approach reduces the average token count by 75% while maintaining performance, with only a minimal 1.9% degradation across seven widely used benchmarks. The project page can be accessed via https://yxxxb.github.io/ATP-LLaVA-page/.

Paperid: 1597, https://arxiv.org/pdf/2412.00171.pdf GitHub

Abstract:
Existing robot policies predominantly adopt the task-centric approach, requiring end-to-end task data collection. This results in limited generalization to new tasks and difficulties in pinpointing errors within long-horizon, multi-stage tasks. To address this, we propose RoboMatrix, a skill-centric hierarchical framework designed for scalable robot task planning and execution in open-world environments. RoboMatrix extracts general meta-skills from diverse complex tasks, enabling the completion of unseen tasks through skill composition. Its architecture consists of a high-level scheduling layer that utilizes large language models (LLMs) for task decomposition, an intermediate skill layer housing meta-skill models, and a low-level hardware layer for robot control. A key innovation of our work is the introduction of the first unified vision-language-action (VLA) model capable of seamlessly integrating both movement and manipulation within one model. This is achieved by combining vision and language prompts to generate discrete actions. Experimental results demonstrate that RoboMatrix achieves a 50% higher success rate than task-centric baselines when applied to unseen objects, scenes, and tasks. To advance open-world robotics research, we will open-source code, hardware designs, model weights, and datasets at https://github.com/WayneMao/RoboMatrix.

Paperid: 1598, https://arxiv.org/pdf/2411.19688.pdf GitHub

Abstract:
Vision-Language Models (VLMs) have great potential in medical tasks, like Visual Question Answering (VQA), where they could act as interactive assistants for both patients and clinicians. Yet their robustness to distribution shifts on unseen data remains a key concern for safe deployment. Evaluating such robustness requires a controlled experimental setup that allows for systematic insights into the model's behavior. However, we demonstrate that current setups fail to offer sufficiently thorough evaluations. To address this gap, we introduce a novel framework, called SURE-VQA, centered around three key requirements to overcome current pitfalls and systematically analyze VLM robustness: 1) Since robustness on synthetic shifts does not necessarily translate to real-world shifts, it should be measured on real-world shifts that are inherent to the VQA data; 2) Traditional token-matching metrics often fail to capture underlying semantics, necessitating the use of large language models (LLMs) for more accurate semantic evaluation; 3) Model performance often lacks interpretability due to missing sanity baselines, thus meaningful baselines should be reported that allow assessing the multimodal impact on the VLM. To demonstrate the relevance of this framework, we conduct a study on the robustness of various Fine-Tuning (FT) methods across three medical datasets with four types of distribution shifts. Our study highlights key insights into robustness: 1) No FT method consistently outperforms others in robustness, and 2) robustness trends are more stable across FT methods than across distribution shifts. Additionally, we find that simple sanity baselines that do not use the image data can perform surprisingly well and confirm LoRA as the best-performing FT method on in-distribution data. Code is provided at https://github.com/IML-DKFZ/sure-vqa.

Paperid: 1599, https://arxiv.org/pdf/2411.19666.pdf GitHub

Abstract:
The field of computational pathology has been transformed with recent advances in foundation models that encode histopathology region-of-interests (ROIs) into versatile and transferable feature representations via self-supervised learning (SSL). However, translating these advancements to address complex clinical challenges at the patient and slide level remains constrained by limited clinical data in disease-specific cohorts, especially for rare clinical conditions. We propose TITAN, a multimodal whole slide foundation model pretrained using 335,645 WSIs via visual self-supervised learning and vision-language alignment with corresponding pathology reports and 423,122 synthetic captions generated from a multimodal generative AI copilot for pathology. Without any finetuning or requiring clinical labels, TITAN can extract general-purpose slide representations and generate pathology reports that generalize to resource-limited clinical scenarios such as rare disease retrieval and cancer prognosis. We evaluate TITAN on diverse clinical tasks and find that TITAN outperforms both ROI and slide foundation models across machine learning settings such as linear probing, few-shot and zero-shot classification, rare cancer retrieval and cross-modal retrieval, and pathology report generation.

Paperid: 1600, https://arxiv.org/pdf/2411.19331.pdf GitHub

Abstract:
Open-Vocabulary Segmentation (OVS) aims at segmenting images from free-form textual concepts without predefined training classes. While existing vision-language models such as CLIP can generate segmentation masks by leveraging coarse spatial information from Vision Transformers, they face challenges in spatial localization due to their global alignment of image and text features. Conversely, self-supervised visual models like DINO excel in fine-grained visual encoding but lack integration with language. To bridge this gap, we present Talk2DINO, a novel hybrid approach that combines the spatial accuracy of DINOv2 with the language understanding of CLIP. Our approach aligns the textual embeddings of CLIP to the patch-level features of DINOv2 through a learned mapping function without the need to fine-tune the underlying backbones. At training time, we exploit the attention maps of DINOv2 to selectively align local visual patches with textual embeddings. We show that the powerful semantic and localization abilities of Talk2DINO can enhance the segmentation process, resulting in more natural and less noisy segmentations, and that our approach can also effectively distinguish foreground objects from the background. Experimental results demonstrate that Talk2DINO achieves state-of-the-art performance across several unsupervised OVS benchmarks. Source code and models are publicly available at: https://lorebianchi98.github.io/Talk2DINO/.

Paperid: 1601, https://arxiv.org/pdf/2411.19325.pdf GitHub

Abstract:
While numerous recent benchmarks focus on evaluating generic Vision-Language Models (VLMs), they do not effectively address the specific challenges of geospatial applications. Generic VLM benchmarks are not designed to handle the complexities of geospatial data, an essential component for applications such as environmental monitoring, urban planning, and disaster management. Key challenges in the geospatial domain include temporal change detection, large-scale object counting, tiny object detection, and understanding relationships between entities in remote sensing imagery. To bridge this gap, we present GEOBench-VLM, a comprehensive benchmark specifically designed to evaluate VLMs on geospatial tasks, including scene understanding, object counting, localization, fine-grained categorization, segmentation, and temporal analysis. Our benchmark features over 10,000 manually verified instructions and spanning diverse visual conditions, object types, and scales. We evaluate several state-of-the-art VLMs to assess performance on geospatial-specific challenges. The results indicate that although existing VLMs demonstrate potential, they face challenges when dealing with geospatial-specific tasks, highlighting the room for further improvements. Notably, the best-performing LLaVa-OneVision achieves only 41.7% accuracy on MCQs, slightly more than GPT-4o, which is approximately double the random guess performance. Our benchmark is publicly available at https://github.com/The-AI-Alliance/GEO-Bench-VLM .

Paperid: 1602, https://arxiv.org/pdf/2411.19067.pdf GitHub

Abstract:
Referring Image Segmentation (RIS) is an advanced vision-language task that involves identifying and segmenting objects within an image as described by free-form text descriptions. While previous studies focused on aligning visual and language features, exploring training techniques, such as data augmentation, remains underexplored. In this work, we explore effective data augmentation for RIS and propose a novel training framework called Masked Referring Image Segmentation (MaskRIS). We observe that the conventional image augmentations fall short of RIS, leading to performance degradation, while simple random masking significantly enhances the performance of RIS. MaskRIS uses both image and text masking, followed by Distortion-aware Contextual Learning (DCL) to fully exploit the benefits of the masking strategy. This approach can improve the model's robustness to occlusions, incomplete information, and various linguistic complexities, resulting in a significant performance improvement. Experiments demonstrate that MaskRIS can easily be applied to various RIS models, outperforming existing methods in both fully supervised and weakly supervised settings. Finally, MaskRIS achieves new state-of-the-art performance on RefCOCO, RefCOCO+, and RefCOCOg datasets. Code is available at https://github.com/naver-ai/maskris.

Paperid: 1603, https://arxiv.org/pdf/2411.18676.pdf GitHub

Abstract:
Language-conditioned robot models have the potential to enable robots to perform a wide range of tasks based on natural language instructions. However, assessing their safety and effectiveness remains challenging because it is difficult to test all the different ways a single task can be phrased. Current benchmarks have two key limitations: they rely on a limited set of human-generated instructions, missing many challenging cases, and focus only on task performance without assessing safety, such as avoiding damage. To address these gaps, we introduce Embodied Red Teaming (ERT), a new evaluation method that generates diverse and challenging instructions to test these models. ERT uses automated red teaming techniques with Vision Language Models (VLMs) to create contextually grounded, difficult instructions. Experimental results show that state-of-the-art language-conditioned robot models fail or behave unsafely on ERT-generated instructions, underscoring the shortcomings of current benchmarks in evaluating real-world performance and safety. Code and videos are available at: https://s-karnik.github.io/embodied-red-team-project-page.

Paperid: 1604, https://arxiv.org/pdf/2411.18672.pdf GitHub

Abstract:
Medical vision-language models often struggle with generating accurate quantitative measurements in radiology reports, leading to hallucinations that undermine clinical reliability. We introduce FactCheXcker, a modular framework that de-hallucinates radiology report measurements by leveraging an improved query-code-update paradigm. Specifically, FactCheXcker employs specialized modules and the code generation capabilities of large language models to solve measurement queries generated based on the original report. After extracting measurable findings, the results are incorporated into an updated report. We evaluate FactCheXcker on endotracheal tube placement, which accounts for an average of 78% of report measurements, using the MIMIC-CXR dataset and 11 medical report-generation models. Our results show that FactCheXcker significantly reduces hallucinations, improves measurement precision, and maintains the quality of the original reports. Specifically, FactCheXcker improves the performance of 10/11 models and achieves an average improvement of 135.0% in reducing measurement hallucinations measured by mean absolute error. Code is available at https://github.com/rajpurkarlab/FactCheXcker.

Paperid: 1605, https://arxiv.org/pdf/2411.18659.pdf GitHub

Abstract:
Large vision-language models (LVLMs) have demonstrated exceptional performance on complex multimodal tasks. However, they continue to suffer from significant hallucination issues, including object, attribute, and relational hallucinations. To accurately detect these hallucinations, we investigated the variations in cross-modal attention patterns between hallucination and non-hallucination states. Leveraging these distinctions, we developed a lightweight detector capable of identifying hallucinations. Our proposed method, Detecting Hallucinations by Cross-modal Attention Patterns (DHCP), is straightforward and does not require additional LVLM training or extra LVLM inference steps. Experimental results show that DHCP achieves remarkable performance in hallucination detection. By offering novel insights into the identification and analysis of hallucinations in LVLMs, DHCP contributes to advancing the reliability and trustworthiness of these models. The code is available at https://github.com/btzyd/DHCP.

Paperid: 1606, https://arxiv.org/pdf/2411.18651.pdf GitHub

Abstract:
Humans recognize objects after observing only a few examples, a remarkable capability enabled by their inherent language understanding of the real-world environment. Developing verbalized and interpretable representation can significantly improve model generalization in low-data settings. In this work, we propose Verbalized Representation Learning (VRL), a novel approach for automatically extracting human-interpretable features for object recognition using few-shot data. Our method uniquely captures inter-class differences and intra-class commonalities in the form of natural language by employing a Vision-Language Model (VLM) to identify key discriminative features between different classes and shared characteristics within the same class. These verbalized features are then mapped to numeric vectors through the VLM. The resulting feature vectors can be further utilized to train and infer with downstream classifiers. Experimental results show that, at the same model scale, VRL achieves a 24% absolute improvement over prior state-of-the-art methods while using 95% less data and a smaller mode. Furthermore, compared to human-labeled attributes, the features learned by VRL exhibit a 20% absolute gain when used for downstream classification tasks. Code is available at: https://github.com/joeyy5588/VRL/tree/main.

Paperid: 1607, https://arxiv.org/pdf/2411.18620.pdf GitHub

Abstract:
The recent advancements in auto-regressive multimodal large language models (MLLMs) have demonstrated promising progress for vision-language tasks. While there exists a variety of studies investigating the processing of linguistic information within large language models, little is currently known about the inner working mechanism of MLLMs and how linguistic and visual information interact within these models. In this study, we aim to fill this gap by examining the information flow between different modalities -- language and vision -- in MLLMs, focusing on visual question answering. Specifically, given an image-question pair as input, we investigate where in the model and how the visual and linguistic information are combined to generate the final prediction. Conducting experiments with a series of models from the LLaVA series, we find that there are two distinct stages in the process of integration of the two modalities. In the lower layers, the model first transfers the more general visual features of the whole image into the representations of (linguistic) question tokens. In the middle layers, it once again transfers visual information about specific objects relevant to the question to the respective token positions of the question. Finally, in the higher layers, the resulting multimodal representation is propagated to the last position of the input sequence for the final prediction. Overall, our findings provide a new and comprehensive perspective on the spatial and functional aspects of image and language processing in the MLLMs, thereby facilitating future research into multimodal information localization and editing. Our code and collected dataset are released here: https://github.com/FightingFighting/cross-modal-information-flow-in-MLLM.git.

Paperid: 1608, https://arxiv.org/pdf/2411.18211.pdf GitHub

Abstract:
Rapid development of large language models (LLMs) has significantly advanced multimodal large language models (LMMs), particularly in vision-language tasks. However, existing video-language models often overlook precise temporal localization and struggle with videos of varying lengths. We introduce TimeMarker, a versatile Video-LLM designed for high-quality dialogue based on video content, emphasizing temporal localization. TimeMarker integrates Temporal Separator Tokens to enhance temporal awareness, accurately marking specific moments within videos. It employs the AnyLength mechanism for dynamic frame sampling and adaptive token merging, enabling effective handling of both short and long videos. Additionally, TimeMarker utilizes diverse datasets, including further transformed temporal-related video QA datasets, to bolster its temporal understanding capabilities. Image and interleaved data are also employed to further enhance the model's semantic perception ability. Evaluations demonstrate that TimeMarker achieves state-of-the-art performance across multiple benchmarks, excelling in both short and long video categories. Our project page is at \url{https://github.com/TimeMarker-LLM/TimeMarker/}.

Paperid: 1609, https://arxiv.org/pdf/2411.18145.pdf GitHub

Abstract:
The rapid advancement of Large Vision-Language Models (VLMs), both general-domain models and those specifically tailored for remote sensing, has demonstrated exceptional perception and reasoning capabilities in Earth observation tasks. However, a benchmark for systematically evaluating their capabilities in this domain is still lacking. To bridge this gap, we propose CHOICE, an extensive benchmark designed to objectively evaluate the hierarchical remote sensing capabilities of VLMs. Focusing on 2 primary capability dimensions essential to remote sensing: perception and reasoning, we further categorize 6 secondary dimensions and 23 leaf tasks to ensure a well-rounded assessment coverage. CHOICE guarantees the quality of all 10,507 problems through a rigorous process of data collection from 50 globally distributed cities, question construction and quality control. The newly curated data and the format of multiple-choice questions with definitive answers allow for an objective and straightforward performance assessment. Our evaluation of 3 proprietary and 21 open-source VLMs highlights their critical limitations within this specialized context. We hope that CHOICE will serve as a valuable resource and offer deeper insights into the challenges and potential of VLMs in the field of remote sensing. We will release CHOICE at https://github.com/ShawnAn-WHU/CHOICE.

Paperid: 1610, https://arxiv.org/pdf/2411.17465.pdf GitHub GitHub

Abstract:
Building Graphical User Interface (GUI) assistants holds significant promise for enhancing human workflow productivity. While most agents are language-based, relying on closed-source API with text-rich meta-information (e.g., HTML or accessibility tree), they show limitations in perceiving UI visuals as humans do, highlighting the need for GUI visual agents. In this work, we develop a vision-language-action model in digital world, namely ShowUI, which features the following innovations: (i) UI-Guided Visual Token Selection to reduce computational costs by formulating screenshots as an UI connected graph, adaptively identifying their redundant relationship and serve as the criteria for token selection during self-attention blocks; (ii) Interleaved Vision-Language-Action Streaming that flexibly unifies diverse needs within GUI tasks, enabling effective management of visual-action history in navigation or pairing multi-turn query-action sequences per screenshot to enhance training efficiency; (iii) Small-scale High-quality GUI Instruction-following Datasets by careful data curation and employing a resampling strategy to address significant data type imbalances. With above components, ShowUI, a lightweight 2B model using 256K data, achieves a strong 75.1% accuracy in zero-shot screenshot grounding. Its UI-guided token selection further reduces 33% of redundant visual tokens during training and speeds up the performance by 1.4x. Navigation experiments across web Mind2Web, mobile AITW, and online MiniWob environments further underscore the effectiveness and potential of our model in advancing GUI visual agents. The models are available at https://github.com/showlab/ShowUI.

Paperid: 1611, https://arxiv.org/pdf/2411.16828.pdf GitHub

Abstract:
Previous works show that noisy, web-crawled image-text pairs may limit vision-language pretraining like CLIP and propose learning with synthetic captions as a promising alternative. Our work continues this effort, introducing two simple yet effective designs to better leverage richly described synthetic captions. Firstly, by observing a strong inverse effect in learning with synthetic captions -- the short synthetic captions can generally lead to MUCH higher performance than full-length ones -- we therefore fed only partial synthetic captions to the text encoder. Secondly, we incorporate an autoregressive captioner to mimic the recaptioning process -- by conditioning on the paired image input and web-crawled text description, the captioner learns to predict the full-length synthetic caption generated by advanced MLLMs. Experiments show that our framework significantly improves zero-shot performance in cross-modal retrieval tasks, setting new SOTA results on MSCOCO and Flickr30K. Moreover, such trained vision encoders can enhance the visual capability of LLaVA, showing strong improvements on a range of MLLM benchmarks. Our project page is https://ucsc-vlaa.github.io/CLIPS/.

Paperid: 1612, https://arxiv.org/pdf/2411.16740.pdf GitHub

Abstract:
Large multimodal models (LMMs) have achieved impressive progress in vision-language understanding, yet they face limitations in real-world applications requiring complex reasoning over a large number of images. Existing benchmarks for multi-image question-answering are limited in scope, each question is paired with only up to 30 images, which does not fully capture the demands of large-scale retrieval tasks encountered in the real-world usages. To reduce these gaps, we introduce two document haystack benchmarks, dubbed DocHaystack and InfoHaystack, designed to evaluate LMM performance on large-scale visual document retrieval and understanding. Additionally, we propose V-RAG, a novel, vision-centric retrieval-augmented generation (RAG) framework that leverages a suite of multimodal vision encoders, each optimized for specific strengths, and a dedicated question-document relevance module. V-RAG sets a new standard, with a 9% and 11% improvement in Recall@1 on the challenging DocHaystack-1000 and InfoHaystack-1000 benchmarks, respectively, compared to the previous best baseline models. Additionally, integrating V-RAG with LMMs enables them to efficiently operate across thousands of images, yielding significant improvements on our DocHaystack and InfoHaystack benchmarks. Our code and datasets are available at https://github.com/Vision-CAIR/dochaystacks

Paperid: 1613, https://arxiv.org/pdf/2411.16724.pdf GitHub

Abstract:
Hallucinations in Large Vision-Language Models (LVLMs) significantly undermine their reliability, motivating researchers to explore the causes of hallucination. However, most studies primarily focus on the language aspect rather than the visual. In this paper, we address how LVLMs process visual information and whether this process causes hallucination. Firstly, we use the attention lens to identify the stages at which LVLMs handle visual data, discovering that the middle layers are crucial. Moreover, we find that these layers can be further divided into two stages: ''visual information enrichment'' and ''semantic refinement'' which respectively propagate visual data to object tokens and interpret it through text. By analyzing attention patterns during the visual information enrichment stage, we find that real tokens consistently receive higher attention weights than hallucinated ones, serving as a strong indicator of hallucination. Further examination of multi-head attention maps reveals that hallucination tokens often result from heads interacting with inconsistent objects. Based on these insights, we propose a simple inference-time method that adjusts visual attention by integrating information across various heads. Extensive experiments demonstrate that this approach effectively mitigates hallucinations in mainstream LVLMs without additional training costs. Code is available at https://github.com/ZhangqiJiang07/middle_layers_indicating_hallucinations.

Paperid: 1614, https://arxiv.org/pdf/2411.16721.pdf GitHub

Abstract:
Vision Language Models (VLMs) can produce unintended and harmful content when exposed to adversarial attacks, particularly because their vision capabilities create new vulnerabilities. Existing defenses, such as input preprocessing, adversarial training, and response evaluation-based methods, are often impractical for real-world deployment due to their high costs. To address this challenge, we propose ASTRA, an efficient and effective defense by adaptively steering models away from adversarial feature directions to resist VLM attacks. Our key procedures involve finding transferable steering vectors representing the direction of harmful response and applying adaptive activation steering to remove these directions at inference time. To create effective steering vectors, we randomly ablate the visual tokens from the adversarial images and identify those most strongly associated with jailbreaks. These tokens are then used to construct steering vectors. During inference, we perform the adaptive steering method that involves the projection between the steering vectors and calibrated activation, resulting in little performance drops on benign inputs while strongly avoiding harmful outputs under adversarial inputs. Extensive experiments across multiple models and baselines demonstrate our state-of-the-art performance and high efficiency in mitigating jailbreak risks. Additionally, ASTRA exhibits good transferability, defending against unseen attacks (i.e., structured-based attack, perturbation-based attack with project gradient descent variants, and text-only attack). Our code is available at \url{https://github.com/ASTRAL-Group/ASTRA}.

Paperid: 1615, https://arxiv.org/pdf/2411.16503.pdf GitHub

Abstract:
Diffusion models have achieved impressive success in generating photorealistic images, but challenges remain in ensuring precise semantic alignment with input prompts. Optimizing the initial noisy latent offers a more efficient alternative to modifying model architectures or prompt engineering for improving semantic alignment. A latest approach, InitNo, refines the initial noisy latent by leveraging attention maps; however, these maps capture only limited information, and the effectiveness of InitNo is highly dependent on the initial starting point, as it tends to converge on a local optimum near this point. To this end, this paper proposes leveraging the language comprehension capabilities of large vision-language models (LVLMs) to guide the optimization of the initial noisy latent, and introduces the Noise Diffusion process, which updates the noisy latent to generate semantically faithful images while preserving distribution consistency. Furthermore, we provide a theoretical analysis of the condition under which the update improves semantic faithfulness. Experimental results demonstrate the effectiveness and adaptability of our framework, consistently enhancing semantic alignment across various diffusion models. The code is available at https://github.com/Bomingmiao/NoiseDiffusion.

Paperid: 1616, https://arxiv.org/pdf/2411.16044.pdf GitHub GitHub

Abstract:
Multimodal Large Language Models (MLLMs) have demonstrated impressive capabilities in vision-language understanding. Recently, with the integration of test-time scaling techniques, these models have also shown strong potential in visual reasoning. However, most existing reasoning approaches remain text-level in nature: MLLMs are prompted to explore various combinations of textual tokens via their underlying language model, while the visual input remains fixed throughout the reasoning process. This paradigm limits the model's ability to fully exploit rich visual information, particularly when dealing with images containing numerous fine-grained elements. In such cases, vision-level reasoning becomes crucial - where models dynamically zoom into specific regions of the image to gather detailed visual cues necessary for accurate decision-making. In this paper, we propose Zoom Eye, a training-free, model-agnostic tree search algorithm tailored for vision-level reasoning. Zoom Eye treats an image as a hierarchical tree structure, where each child node represents a zoomed-in sub-region of its parent, and the root corresponds to the full image. The algorithm enables MLLMs to simulate human-like zooming behavior by navigating from root to leaf nodes in search of task-relevant visual evidence. We experiment on a series of high-resolution benchmarks and the results demonstrate that Zoom Eye consistently improves the performance of multiple MLLMs by a large margin (e.g., InternVL2.5-8B increases by 15.71% and 17.69% on HR-Bench) and also enables small 3-8B MLLMs to outperform strong large models such as GPT-4o. Code: https://github.com/om-ai-lab/ZoomEye

Paperid: 1617, https://arxiv.org/pdf/2411.15922.pdf GitHub GitHub

Abstract:
Recent advances in All-in-One (AiO) RGB image restoration have demonstrated the effectiveness of prompt learning in handling multiple degradations within a single model. However, extending these approaches to hyperspectral image (HSI) restoration is challenging due to the domain gap between RGB and HSI features, information loss in visual prompts under severe composite degradations, and difficulties in capturing HSI-specific degradation patterns via text prompts. In this paper, we propose PromptHSI, the first universal AiO HSI restoration framework that addresses these challenges. By incorporating frequency-aware feature modulation, which utilizes frequency analysis to narrow down the restoration search space and employing vision-language model (VLM)-guided prompt learning, our approach decomposes text prompts into intensity and bias controllers that effectively guide the restoration process while mitigating domain discrepancies. Extensive experiments demonstrate that our unified architecture excels at both fine-grained recovery and global information restoration across diverse degradation scenarios, highlighting its significant potential for practical remote sensing applications. The source code is available at https://github.com/chingheng0808/PromptHSI.

Paperid: 1618, https://arxiv.org/pdf/2411.15869.pdf GitHub

Abstract:
Recent advancements in pre-trained vision-language models like CLIP, have enabled the task of open-vocabulary segmentation. CLIP demonstrates impressive zero-shot capabilities in various downstream tasks that require holistic image understanding. However, due to its image-level pre-training, CLIP struggles to capture local details, resulting in poor performance in segmentation tasks. Our analysis reveals that anomaly tokens emerge during the forward pass, drawing excessive attention from normal patch tokens, thereby diminishing spatial awareness. To address this issue, we propose Self-Calibrated CLIP (SC-CLIP), a training-free method that calibrates CLIP to produce finer representations while preserving its original generalization ability, without introducing new parameters or relying on additional backbones. Specifically, we first identify and resolve the anomaly tokens to mitigate their negative impact. Next, we enhance feature discriminability and attention correlation by leveraging the semantic consistency found in CLIP's intermediate features. Furthermore, we explore how to effectively employ multi-level feature fusion under the training-free setting. Collectively, these strategies enhance CLIP's feature representation with greater granularity and coherence. Experimental results demonstrate the effectiveness of SC-CLIP, achieving state-of-the-art results across all datasets and surpassing previous methods by 9.5%. Notably, SC-CLIP boosts the performance of vanilla CLIP ViT-L/14 by 6.8 times. Our source code is available at https://github.com/SuleBai/SC-CLIP.

Paperid: 1619, https://arxiv.org/pdf/2411.15851.pdf GitHub

Abstract:
While vision-language models like CLIP have shown remarkable success in open-vocabulary tasks, their application is currently confined to image-level tasks, and they still struggle with dense predictions. Recent works often attribute such deficiency in dense predictions to the self-attention layers in the final block, and have achieved commendable results by modifying the original query-key attention to self-correlation attention, (e.g., query-query and key-key attention). However, these methods overlook the cross-correlation attention (query-key) properties, which capture the rich spatial correspondence. In this paper, we reveal that the cross-correlation of the self-attention in CLIP's non-final layers also exhibits localization properties. Therefore, we propose the Residual Cross-correlation Self-attention (RCS) module, which leverages the cross-correlation self-attention from intermediate layers to remold the attention in the final block. The RCS module effectively reorganizes spatial information, unleashing the localization potential within CLIP for dense vision-language inference. Furthermore, to enhance the focus on regions of the same categories and local consistency, we propose the Semantic Feedback Refinement (SFR) module, which utilizes semantic segmentation maps to further adjust the attention scores. By integrating these two strategies, our method, termed ResCLIP, can be easily incorporated into existing approaches as a plug-and-play module, significantly boosting their performance in dense vision-language inference. Extensive experiments across multiple standard benchmarks demonstrate that our method surpasses state-of-the-art training-free methods, validating the effectiveness of the proposed approach. Code is available at https://github.com/yvhangyang/ResCLIP.

Paperid: 1620, https://arxiv.org/pdf/2411.15839.pdf GitHub GitHub

Abstract:
Large Vision-Language Models (LVLMs) have demonstrated remarkable capabilities in multimodal task reasoning. However, they often generate responses that appear plausible yet do not accurately reflect the visual content, a phenomenon known as hallucination. Recent approaches have introduced training-free methods to mitigate hallucinations by adjusting the decoding strategy during the inference stage, typically attributing hallucinations to the language model itself. Our analysis, however, reveals that distortions in the visual encoding process significantly affect the model's reasoning capabilities. Specifically, earlier visual layers may retain key features but gradually distort as the information propagates toward the output layer. Building on these insights, we propose a novel hallucination-mitigation method from the visual encoding perspective: \textbf{V}isu\textbf{a}l \textbf{L}ayer Fus\textbf{i}on Contrastive \textbf{D}ecoding (\textbf{VaLiD}). This method utilizes uncertainty to guide the visual layer selection, correcting distortions in the visual encoding process and thereby enhancing the reliability of the generated content. Experimental results demonstrate the effectiveness of VaLiD in mitigating hallucinations across various benchmarks, achieving state-of-the-art performance when compared to baseline methods. Codes are available at \href{https://github.com/RicardoLuL/VaLiD_LVLMs_hallucinations}{Github}.

Paperid: 1621, https://arxiv.org/pdf/2411.15735.pdf GitHub

Abstract:
Test-time adaptation with pre-trained vision-language models (VLMs) has attracted increasing attention for tackling the issue of distribution shift during the test phase. While prior methods have shown effectiveness in addressing distribution shift by adjusting classification logits, they are not optimal due to keeping text features unchanged. To address this issue, we introduce a new approach called Test-time Alignment-Enhanced Adapter (TAEA), which trains an adapter with test samples to adjust text features during the test phase. We can enhance the text-to-image alignment prediction by utilizing an adapter to adapt text features. Furthermore, we also propose to adopt the negative cache from TDA as enhancement module, which further improves the performance of TAEA. Our approach outperforms the state-of-the-art TTA method of pre-trained VLMs by an average of 0.75% on the out-of-distribution benchmark and 2.5% on the cross-domain benchmark, with an acceptable training time. Code will be available at https://github.com/BaoshunWq/clip-TAEA.

Paperid: 1622, https://arxiv.org/pdf/2411.15714.pdf GitHub

Abstract:
Recently, Vision Language Models (VLMs) have experienced significant advancements, yet these models still face challenges in spatial hierarchical reasoning within indoor scenes. In this study, we introduce ROOT, a VLM-based system designed to enhance the analysis of indoor scenes. Specifically, we first develop an iterative object perception algorithm using GPT-4V to detect object entities within indoor scenes. This is followed by employing vision foundation models to acquire additional meta-information about the scene, such as bounding boxes. Building on this foundational data, we propose a specialized VLM, SceneVLM, which is capable of generating spatial hierarchical scene graphs and providing distance information for objects within indoor environments. This information enhances our understanding of the spatial arrangement of indoor scenes. To train our SceneVLM, we collect over 610,000 images from various public indoor datasets and implement a scene data generation pipeline with a semi-automated technique to establish relationships and estimate distances among indoor objects. By utilizing this enriched data, we conduct various training recipes and finish SceneVLM. Our experiments demonstrate that \rootname facilitates indoor scene understanding and proves effective in diverse downstream applications, such as 3D scene generation and embodied AI. The code will be released at \url{https://github.com/harrytea/ROOT}.

Paperid: 1623, https://arxiv.org/pdf/2411.15605.pdf GitHub

Abstract:
Understanding deep models is crucial for deploying them in safety-critical applications. We introduce GIFT, a framework for deriving post-hoc, global, interpretable, and faithful textual explanations for vision classifiers. GIFT starts from local faithful visual counterfactual explanations and employs (vision) language models to translate those into global textual explanations. Crucially, GIFT provides a verification stage measuring the causal effect of the proposed explanations on the classifier decision. Through experiments across diverse datasets, including CLEVR, CelebA, and BDD, we demonstrate that GIFT effectively reveals meaningful insights, uncovering tasks, concepts, and biases used by deep vision classifiers. The framework is released at https://github.com/valeoai/GIFT.

Paperid: 1624, https://arxiv.org/pdf/2411.15232.pdf GitHub

Abstract:
Recent advancements in vision-language models (VLMs), such as CLIP, have demonstrated substantial success in self-supervised representation learning for vision tasks. However, effectively adapting VLMs to downstream applications remains challenging, as their accuracy often depends on time-intensive and expertise-demanding prompt engineering, while full model fine-tuning is costly. This is particularly true for biomedical images, which, unlike natural images, typically suffer from limited annotated datasets, unintuitive image contrasts, and nuanced visual features. Recent prompt learning techniques, such as Context Optimization (CoOp) intend to tackle these issues, but still fall short in generalizability. Meanwhile, explorations in prompt learning for biomedical image analysis are still highly limited. In this work, we propose BiomedCoOp, a novel prompt learning framework that enables efficient adaptation of BiomedCLIP for accurate and highly generalizable few-shot biomedical image classification. Our approach achieves effective prompt context learning by leveraging semantic consistency with average prompt ensembles from Large Language Models (LLMs) and knowledge distillation with a statistics-based prompt selection strategy. We conducted comprehensive validation of our proposed framework on 11 medical datasets across 9 modalities and 10 organs against existing state-of-the-art methods, demonstrating significant improvements in both accuracy and generalizability. The code is publicly available at https://github.com/HealthX-Lab/BiomedCoOp.

Paperid: 1625, https://arxiv.org/pdf/2411.14901.pdf GitHub

Abstract:
Large language models (LLMs) excel at retrieving information from lengthy text, but their vision-language counterparts (VLMs) face difficulties with hour-long videos, especially for temporal grounding. Specifically, these VLMs are constrained by frame limitations, often losing essential temporal details needed for accurate event localization in extended video content. We propose ReVisionLLM, a recursive vision-language model designed to locate events in hour-long videos. Inspired by human search strategies, our model initially targets broad segments of interest, progressively revising its focus to pinpoint exact temporal boundaries. Our model can seamlessly handle videos of vastly different lengths, from minutes to hours. We also introduce a hierarchical training strategy that starts with short clips to capture distinct events and progressively extends to longer videos. To our knowledge, ReVisionLLM is the first VLM capable of temporal grounding in hour-long videos, outperforming previous state-of-the-art methods across multiple datasets by a significant margin (+2.6% R1@0.1 on MAD). The code is available at https://github.com/Tanveer81/ReVisionLLM.

Paperid: 1626, https://arxiv.org/pdf/2411.14794.pdf GitHub

Abstract:
The advancement of Large Vision Language Models (LVLMs) has significantly improved multimodal understanding, yet challenges remain in video reasoning tasks due to the scarcity of high-quality, large-scale datasets. Existing video question-answering (VideoQA) datasets often rely on costly manual annotations with insufficient granularity or automatic construction methods with redundant frame-by-frame analysis, limiting their scalability and effectiveness for complex reasoning. To address these challenges, we introduce VideoEspresso, a novel dataset that features VideoQA pairs preserving essential spatial details and temporal coherence, along with multimodal annotations of intermediate reasoning steps. Our construction pipeline employs a semantic-aware method to reduce redundancy, followed by generating QA pairs using GPT-4o. We further develop video Chain-of-Thought (CoT) annotations to enrich reasoning processes, guiding GPT-4o in extracting logical relationships from QA pairs and video content. To exploit the potential of high-quality VideoQA pairs, we propose a Hybrid LVLMs Collaboration framework, featuring a Frame Selector and a two-stage instruction fine-tuned reasoning LVLM. This framework adaptively selects core frames and performs CoT reasoning using multimodal evidence. Evaluated on our proposed benchmark with 14 tasks against 9 popular LVLMs, our method outperforms existing baselines on most tasks, demonstrating superior video reasoning capabilities. Our code and dataset will be released at: https://github.com/hshjerry/VideoEspresso

Paperid: 1627, https://arxiv.org/pdf/2411.13918.pdf GitHub

Abstract:
Deep neural networks, while achieving remarkable success across diverse tasks, demand significant resources, including computation, GPU memory, bandwidth, storage, and energy. Network quantization, as a standard compression and acceleration technique, reduces storage costs and enables potential inference acceleration by discretizing network weights and activations into a finite set of integer values. However, current quantization methods are often complex and sensitive, requiring extensive task-specific hyperparameters, where even a single misconfiguration can impair model performance, limiting generality across different models and tasks. In this paper, we propose Quantization without Tears (QwT), a method that simultaneously achieves quantization speed, accuracy, simplicity, and generality. The key insight of QwT is to incorporate a lightweight additional structure into the quantized network to mitigate information loss during quantization. This structure consists solely of a small set of linear layers, keeping the method simple and efficient. More importantly, it provides a closed-form solution, allowing us to improve accuracy effortlessly under 2 minutes. Extensive experiments across various vision, language, and multimodal tasks demonstrate that QwT is both highly effective and versatile. In fact, our approach offers a robust solution for network quantization that combines simplicity, accuracy, and adaptability, which provides new insights for the design of novel quantization paradigms. The code is publicly available at https://github.com/wujx2001/QwT

Paperid: 1628, https://arxiv.org/pdf/2411.13754.pdf GitHub

Abstract:
Complex visual reasoning and question answering (VQA) is a challenging task that requires compositional multi-step processing and higher-level reasoning capabilities beyond the immediate recognition and localization of objects and events. Here, we introduce a fully neural Iterative and Parallel Reasoning Mechanism (IPRM) that combines two distinct forms of computation -- iterative and parallel -- to better address complex VQA scenarios. Specifically, IPRM's "iterative" computation facilitates compositional step-by-step reasoning for scenarios wherein individual operations need to be computed, stored, and recalled dynamically (e.g. when computing the query "determine the color of pen to the left of the child in red t-shirt sitting at the white table"). Meanwhile, its "parallel" computation allows for the simultaneous exploration of different reasoning paths and benefits more robust and efficient execution of operations that are mutually independent (e.g. when counting individual colors for the query: "determine the maximum occurring color amongst all t-shirts"). We design IPRM as a lightweight and fully-differentiable neural module that can be conveniently applied to both transformer and non-transformer vision-language backbones. It notably outperforms prior task-specific methods and transformer-based attention modules across various image and video VQA benchmarks testing distinct complex reasoning capabilities such as compositional spatiotemporal reasoning (AGQA), situational reasoning (STAR), multi-hop reasoning generalization (CLEVR-Humans) and causal event linking (CLEVRER-Humans). Further, IPRM's internal computations can be visualized across reasoning steps, aiding interpretability and diagnosis of its errors.

Paperid: 1629, https://arxiv.org/pdf/2411.13591.pdf GitHub

Abstract:
Graphical User Interface (GUI) grounding plays a crucial role in enhancing the capabilities of Vision-Language Model (VLM) agents. While general VLMs, such as GPT-4V, demonstrate strong performance across various tasks, their proficiency in GUI grounding remains suboptimal. Recent studies have focused on fine-tuning these models specifically for zero-shot GUI grounding, yielding significant improvements over baseline performance. We introduce a visual prompting framework that employs an iterative narrowing mechanism to further improve the performance of both general and fine-tuned models in GUI grounding. For evaluation, we tested our method on a comprehensive benchmark comprising various UI platforms and provided the code to reproduce our results.

Paperid: 1630, https://arxiv.org/pdf/2411.13587.pdf GitHub

Abstract:
Recently in robotics, Vision-Language-Action (VLA) models have emerged as a transformative approach, enabling robots to execute complex tasks by integrating visual and linguistic inputs within an end-to-end learning framework. Despite their significant capabilities, VLA models introduce new attack surfaces. This paper systematically evaluates their robustness. Recognizing the unique demands of robotic execution, our attack objectives target the inherent spatial and functional characteristics of robotic systems. In particular, we introduce two untargeted attack objectives that leverage spatial foundations to destabilize robotic actions, and a targeted attack objective that manipulates the robotic trajectory. Additionally, we design an adversarial patch generation approach that places a small, colorful patch within the camera's view, effectively executing the attack in both digital and physical environments. Our evaluation reveals a marked degradation in task success rates, with up to a 100\% reduction across a suite of simulated robotic tasks, highlighting critical security gaps in current VLA architectures. By unveiling these vulnerabilities and proposing actionable evaluation metrics, we advance both the understanding and enhancement of safety for VLA-based robotic systems, underscoring the necessity for continuously developing robust defense strategies prior to physical-world deployments.

Paperid: 1631, https://arxiv.org/pdf/2411.13243.pdf GitHub

Abstract:
Existing methodologies in open vocabulary 3D semantic segmentation primarily concentrate on establishing a unified feature space encompassing 3D, 2D, and textual modalities. Nevertheless, traditional techniques such as global feature alignment or vision-language model distillation tend to impose only approximate correspondence, struggling notably with delineating fine-grained segmentation boundaries. To address this gap, we propose a more meticulous mask-level alignment between 3D features and the 2D-text embedding space through a cross-modal mask reasoning framework, XMask3D. In our approach, we developed a mask generator based on the denoising UNet from a pre-trained diffusion model, leveraging its capability for precise textual control over dense pixel representations and enhancing the open-world adaptability of the generated masks. We further integrate 3D global features as implicit conditions into the pre-trained 2D denoising UNet, enabling the generation of segmentation masks with additional 3D geometry awareness. Subsequently, the generated 2D masks are employed to align mask-level 3D representations with the vision-language feature space, thereby augmenting the open vocabulary capability of 3D geometry embeddings. Finally, we fuse complementary 2D and 3D mask features, resulting in competitive performance across multiple benchmarks for 3D open vocabulary semantic segmentation. Code is available at https://github.com/wangzy22/XMask3D.

Paperid: 1632, https://arxiv.org/pdf/2411.13112.pdf GitHub

Abstract:
Accurate spatial reasoning in outdoor environments - covering geometry, object pose, and inter-object relationships - is fundamental to downstream tasks such as mapping, motion forecasting, and high-level planning in autonomous driving. We introduce SURDS, a large-scale benchmark designed to systematically evaluate the spatial reasoning capabilities of vision language models (VLMs). Built on the nuScenes dataset, SURDS comprises 41,080 vision-question-answer training instances and 9,250 evaluation samples, spanning six spatial categories: orientation, depth estimation, pixel-level localization, pairwise distance, lateral ordering, and front-behind relations. We benchmark leading general-purpose VLMs, including GPT, Gemini, and Qwen, revealing persistent limitations in fine-grained spatial understanding. To address these deficiencies, we go beyond static evaluation and explore whether alignment techniques can improve spatial reasoning performance. Specifically, we propose a reinforcement learning-based alignment scheme leveraging spatially grounded reward signals - capturing both perception-level accuracy (location) and reasoning consistency (logic). We further incorporate final-answer correctness and output-format rewards to guide fine-grained policy adaptation. Our GRPO-aligned variant achieves an overall score of 40.80 in the SURDS benchmark. Notably, it outperforms proprietary systems such as GPT-4o (13.30) and Gemini-2.0-flash (35.71). To our best knowledge, this is the first study to demonstrate that reinforcement learning-based alignment can significantly and consistently enhance the spatial reasoning capabilities of VLMs in real-world driving contexts. We release the SURDS benchmark, evaluation toolkit, and GRPO alignment code through: https://github.com/XiandaGuo/Drive-MLLM.

Paperid: 1633, https://arxiv.org/pdf/2411.12044.pdf GitHub

Abstract:
Recent advances in foundational Vision Language Models (VLMs) have reshaped the evaluation paradigm in computer vision tasks. These foundational models, especially CLIP, have accelerated research in open-vocabulary computer vision tasks, including Open-Vocabulary Semantic Segmentation (OVSS). Although the initial results are promising, the dense prediction capabilities of VLMs still require further improvement. In this study, we enhance the semantic segmentation performance of CLIP by introducing new modules and modifications: 1) architectural changes in the last layer of ViT and the incorporation of attention maps from the middle layers with the last layer, 2) Image Engineering: applying data augmentations to enrich input image representations, and 3) using Large Language Models (LLMs) to generate definitions and synonyms for each class name to leverage CLIP's open-vocabulary capabilities. Our training-free method, ITACLIP, outperforms current state-of-the-art approaches on segmentation benchmarks such as COCO-Stuff, COCO-Object, Pascal Context, and Pascal VOC. Our code is available at https://github.com/m-arda-aydn/ITACLIP.

Paperid: 1634, https://arxiv.org/pdf/2411.11904.pdf GitHub

Abstract:
Remote sensing (RS) visual grounding aims to use natural language expression to locate specific objects (in the form of the bounding box or segmentation mask) in RS images, enhancing human interaction with intelligent RS interpretation systems. Early research in this area was primarily based on horizontal bounding boxes (HBBs), but as more diverse RS datasets have become available, tasks involving oriented bounding boxes (OBBs) and segmentation masks have emerged. In practical applications, different targets require different grounding types: HBB can localize an object's position, OBB provides its orientation, and mask depicts its shape. However, existing specialized methods are typically tailored to a single type of RS visual grounding task and are hard to generalize across tasks. In contrast, large vision-language models (VLMs) exhibit powerful multi-task learning capabilities but struggle to handle dense prediction tasks like segmentation. This paper proposes GeoGround, a novel framework that unifies support for HBB, OBB, and mask RS visual grounding tasks, allowing flexible output selection. Rather than customizing the architecture of VLM, our work aims to elegantly support pixel-level visual grounding output through the Text-Mask technique. We define prompt-assisted and geometry-guided learning to enhance consistency across different signals. Experimental results show that GeoGround demonstrates strong performance across four RS visual grounding tasks, matching the performance of specialized methods on multiple benchmarks. Code available at https://github.com/zytx121/GeoGround

Paperid: 1635, https://arxiv.org/pdf/2411.11706.pdf GitHub

Paperid: 1636, https://arxiv.org/pdf/2411.11496.pdf GitHub

Abstract:
Recent advances in Large Vision-Language Models (LVLMs) have showcased strong reasoning abilities across multiple modalities, achieving significant breakthroughs in various real-world applications. Despite this great success, the safety guardrail of LVLMs may not cover the unforeseen domains introduced by the visual modality. Existing studies primarily focus on eliciting LVLMs to generate harmful responses via carefully crafted image-based jailbreaks designed to bypass alignment defenses. In this study, we reveal that a safe image can be exploited to achieve the same jailbreak consequence when combined with additional safe images and prompts. This stems from two fundamental properties of LVLMs: universal reasoning capabilities and safety snowball effect. Building on these insights, we propose Safety Snowball Agent (SSA), a novel agent-based framework leveraging agents' autonomous and tool-using abilities to jailbreak LVLMs. SSA operates through two principal stages: (1) initial response generation, where tools generate or retrieve jailbreak images based on potential harmful intents, and (2) harmful snowballing, where refined subsequent prompts induce progressively harmful outputs. Our experiments demonstrate that \ours can use nearly any image to induce LVLMs to produce unsafe content, achieving high success jailbreaking rates against the latest LVLMs. Unlike prior works that exploit alignment flaws, \ours leverages the inherent properties of LVLMs, presenting a profound challenge for enforcing safety in generative multimodal systems. Our code is avaliable at \url{https://github.com/gzcch/Safety_Snowball_Agent}.

Paperid: 1637, https://arxiv.org/pdf/2411.11479.pdf GitHub

Abstract:
The recent progress in Vision-Language Models (VLMs) has broadened the scope of multimodal applications. However, evaluations often remain limited to functional tasks, neglecting abstract dimensions such as personality traits and human values. To address this gap, we introduce Value-Spectrum, a novel Visual Question Answering (VQA) benchmark aimed at assessing VLMs based on Schwartz's value dimensions that capture core human values guiding people's preferences and actions. We design a VLM agent pipeline to simulate video browsing and construct a vector database comprising over 50,000 short videos from TikTok, YouTube Shorts, and Instagram Reels. These videos span multiple months and cover diverse topics, including family, health, hobbies, society, technology, etc. Benchmarking on Value-Spectrum highlights notable variations in how VLMs handle value-oriented content. Beyond identifying VLMs' intrinsic preferences, we also explore the ability of VLM agents to adopt specific personas when explicitly prompted, revealing insights into the adaptability of the model in role-playing scenarios. These findings highlight the potential of Value-Spectrum as a comprehensive evaluation set for tracking VLM preferences in value-based tasks and abilities to simulate diverse personas. The complete code and data are available at: https://github.com/Jeremyyny/Value-Spectrum.

Paperid: 1638, https://arxiv.org/pdf/2411.10922.pdf GitHub

Abstract:
Action detection aims to detect (recognize and localize) human actions spatially and temporally in videos. Existing approaches focus on the closed-set setting where an action detector is trained and tested on videos from a fixed set of action categories. However, this constrained setting is not viable in an open world where test videos inevitably come beyond the trained action categories. In this paper, we address the practical yet challenging Open-Vocabulary Action Detection (OVAD) problem. It aims to detect any action in test videos while training a model on a fixed set of action categories. To achieve such an open-vocabulary capability, we propose a novel method OpenMixer that exploits the inherent semantics and localizability of large vision-language models (VLM) within the family of query-based detection transformers (DETR). Specifically, the OpenMixer is developed by spatial and temporal OpenMixer blocks (S-OMB and T-OMB), and a dynamically fused alignment (DFA) module. The three components collectively enjoy the merits of strong generalization from pre-trained VLMs and end-to-end learning from DETR design. Moreover, we established OVAD benchmarks under various settings, and the experimental results show that the OpenMixer performs the best over baselines for detecting seen and unseen actions. We release the codes, models, and dataset splits at https://github.com/Cogito2012/OpenMixer.

Paperid: 1639, https://arxiv.org/pdf/2411.10440.pdf GitHub

Abstract:
Large language models have demonstrated substantial advancements in reasoning capabilities. However, current Vision-Language Models (VLMs) often struggle to perform systematic and structured reasoning, especially when handling complex visual question-answering tasks. In this work, we introduce LLaVA-CoT, a large VLM designed to conduct autonomous multistage reasoning. Unlike chain-of-thought prompting, LLaVA-CoT independently engages in sequential stages of summarization, visual interpretation, logical reasoning, and conclusion generation. This structured approach enables LLaVA-CoT to achieve marked improvements on reasoning-intensive tasks. To accomplish this, we construct the LLaVA-CoT-100k dataset, integrating samples from various visual question answering sources and providing structured reasoning annotations. Besides, we propose a test-time stage-wise retracing search method (SWIRES), which enables effective and efficient test-time scaling. Remarkably, with only 100k training samples and test-time scaling, LLaVA-CoT not only outperforms its base model by 9.4% on a wide range of multimodal reasoning benchmarks, but also surpasses the performance of larger and even closed-source models, such as Gemini-1.5-pro, GPT-4o-mini, and Llama-3.2-90B-Vision-Instruct. The code, dataset, and pre-trained weights are publicly available at https://github.com/PKU-YuanGroup/LLaVA-CoT.

Paperid: 1640, https://arxiv.org/pdf/2411.10161.pdf GitHub

Abstract:
Existing Image Quality Assessment (IQA) methods achieve remarkable success in analyzing quality for overall image, but few works explore quality analysis for Regions of Interest (ROIs). The quality analysis of ROIs can provide fine-grained guidance for image quality improvement and is crucial for scenarios focusing on region-level quality. This paper proposes a novel network, SEAGULL, which can SEe and Assess ROIs quality with GUidance from a Large vision-Language model. SEAGULL incorporates a vision-language model (VLM), masks generated by Segment Anything Model (SAM) to specify ROIs, and a meticulously designed Mask-based Feature Extractor (MFE) to extract global and local tokens for specified ROIs, enabling accurate fine-grained IQA for ROIs. Moreover, this paper constructs two ROI-based IQA datasets, SEAGULL-100w and SEAGULL-3k, for training and evaluating ROI-based IQA. SEAGULL-100w comprises about 100w synthetic distortion images with 33 million ROIs for pre-training to improve the model's ability of regional quality perception, and SEAGULL-3k contains about 3k authentic distortion ROIs to enhance the model's ability to perceive real world distortions. After pre-training on SEAGULL-100w and fine-tuning on SEAGULL-3k, SEAGULL shows remarkable performance on fine-grained ROI quality assessment. Code and datasets are publicly available at the https://github.com/chencn2020/Seagull.

Paperid: 1641, https://arxiv.org/pdf/2411.09894.pdf GitHub

Abstract:
Whole slide image (WSI) analysis is gaining prominence within the medical imaging field. Recent advances in pathology foundation models have shown the potential to extract powerful feature representations from WSIs for downstream tasks. However, these foundation models are usually designed for general-purpose pathology image analysis and may not be optimal for specific downstream tasks or cancer types. In this work, we present Concept Anchor-guided Task-specific Feature Enhancement (CATE), an adaptable paradigm that can boost the expressivity and discriminativeness of pathology foundation models for specific downstream tasks. Based on a set of task-specific concepts derived from the pathology vision-language model with expert-designed prompts, we introduce two interconnected modules to dynamically calibrate the generic image features extracted by foundation models for certain tasks or cancer types. Specifically, we design a Concept-guided Information Bottleneck module to enhance task-relevant characteristics by maximizing the mutual information between image features and concept anchors while suppressing superfluous information. Moreover, a Concept-Feature Interference module is proposed to utilize the similarity between calibrated features and concept anchors to further generate discriminative task-specific features. The extensive experiments on public WSI datasets demonstrate that CATE significantly enhances the performance and generalizability of MIL models. Additionally, heatmap and umap visualization results also reveal the effectiveness and interpretability of CATE. The source code is available at https://github.com/HKU-MedAI/CATE.

Paperid: 1642, https://arxiv.org/pdf/2411.09301.pdf GitHub

Abstract:
Automatically and rapidly understanding Earth's surface is fundamental to our grasp of the living environment and informed decision-making. This underscores the need for a unified system with comprehensive capabilities in analyzing Earth's surface to address a wide range of human needs. The emergence of multimodal large language models (MLLMs) has great potential in boosting the efficiency and convenience of intelligent Earth observation. These models can engage in human-like conversations, serve as unified platforms for understanding images, follow diverse instructions, and provide insightful feedbacks. In this study, we introduce LHRS-Bot-Nova, an MLLM specialized in understanding remote sensing (RS) images, designed to expertly perform a wide range of RS understanding tasks aligned with human instructions. LHRS-Bot-Nova features an enhanced vision encoder and a novel bridge layer, enabling efficient visual compression and better language-vision alignment. To further enhance RS-oriented vision-language alignment, we propose a large-scale RS image-caption dataset, generated through feature-guided image recaptioning. Additionally, we introduce an instruction dataset specifically designed to improve spatial recognition abilities. Extensive experiments demonstrate superior performance of LHRS-Bot-Nova across various RS image understanding tasks. We also evaluate different MLLM performances in complex RS perception and instruction following using a complicated multi-choice question evaluation benchmark, providing a reliable guide for future model selection and improvement. Data, code, and models will be available at https://github.com/NJU-LHRS/LHRS-Bot.

Paperid: 1643, https://arxiv.org/pdf/2411.09022.pdf GitHub

Abstract:
Large Language Models (LLMs) have demonstrated promising reasoning capabilities in robotics; however, their application in multi-robot systems remains limited, particularly in handling task dependencies. This paper introduces DART-LLM, a novel framework that employs Directed Acyclic Graphs (DAGs) to model task dependencies, enabling the decomposition of natural language instructions into well-coordinated subtasks for multi-robot execution. DART-LLM comprises four key components: a Question-Answering (QA) LLM module for dependency-aware task decomposition, a Breakdown Function module for robot assignment, an Actuation module for execution, and a Vision-Language Model (VLM)-based object detector for environmental perception, achieving end-to-end task execution. Experimental results across three task complexity levels demonstrate that DART-LLM achieves state-of-the-art performance, significantly outperforming the baseline across all evaluation metrics. Among the tested models, DeepSeek-r1-671B achieves the highest success rate, whereas Llama-3.1-8B exhibits superior response time reliability. Ablation studies further confirm that explicit dependency modeling notably enhances the performance of smaller models, facilitating efficient deployment on resource-constrained platforms. Please refer to the project website https://wyd0817.github.io/project-dart-llm/ for videos and code.

Paperid: 1644, https://arxiv.org/pdf/2411.07184.pdf GitHub

Abstract:
3D part segmentation is a crucial and challenging task in 3D perception, playing a vital role in applications such as robotics, 3D generation, and 3D editing. Recent methods harness the powerful Vision Language Models (VLMs) for 2D-to-3D knowledge distillation, achieving zero-shot 3D part segmentation. However, these methods are limited by their reliance on text prompts, which restricts the scalability to large-scale unlabeled datasets and the flexibility in handling part ambiguities. In this work, we introduce SAMPart3D, a scalable zero-shot 3D part segmentation framework that segments any 3D object into semantic parts at multiple granularities, without requiring predefined part label sets as text prompts. For scalability, we use text-agnostic vision foundation models to distill a 3D feature extraction backbone, allowing scaling to large unlabeled 3D datasets to learn rich 3D priors. For flexibility, we distill scale-conditioned part-aware 3D features for 3D part segmentation at multiple granularities. Once the segmented parts are obtained from the scale-conditioned part-aware 3D features, we use VLMs to assign semantic labels to each part based on the multi-view renderings. Compared to previous methods, our SAMPart3D can scale to the recent large-scale 3D object dataset Objaverse and handle complex, non-ordinary objects. Additionally, we contribute a new 3D part segmentation benchmark to address the lack of diversity and complexity of objects and parts in existing benchmarks. Experiments show that our SAMPart3D significantly outperforms existing zero-shot 3D part segmentation methods, and can facilitate various applications such as part-level editing and interactive segmentation.

Paperid: 1645, https://arxiv.org/pdf/2411.06921.pdf GitHub

Abstract:
Pre-trained vision-language models (e.g., CLIP) have shown powerful zero-shot transfer capabilities. But they still struggle with domain shifts and typically require labeled data to adapt to downstream tasks, which could be costly. In this work, we aim to leverage unlabeled data that naturally spans multiple domains to enhance the transferability of vision-language models. Under this unsupervised multi-domain setting, we have identified inherent model bias within CLIP, notably in its visual and text encoders. Specifically, we observe that CLIP's visual encoder tends to prioritize encoding domain over discriminative category information, meanwhile its text encoder exhibits a preference for domain-relevant classes. To mitigate this model bias, we propose a training-free and label-free feature calibration method, Unsupervised Multi-domain Feature Calibration (UMFC). UMFC estimates image-level biases from domain-specific features and text-level biases from the direction of domain transition. These biases are subsequently subtracted from original image and text features separately, to render them domain-invariant. We evaluate our method on multiple settings including transductive learning and test-time adaptation. Extensive experiments show that our method outperforms CLIP and performs on par with the state-of-the-arts that need additional annotations or optimization. Our code is available at https://github.com/GIT-LJc/UMFC.

Paperid: 1646, https://arxiv.org/pdf/2411.06657.pdf GitHub

Abstract:
In the past several years there has been an explosion of available models for vision-language tasks. Unfortunately, the literature still leaves open a number of questions related to best practices in designing and training such models. In this paper we seek to answer several questions related to the pretraining of vision-language encoders through meta-analysis. In our first set of experiments, we show that we can save significant compute at no cost to downstream performance, by freezing large parts of vision-language models during pretraining. In our second set of experiments we examine the effect of basing a VL transformer on a vision model versus a text model. Additionally, we introduce a VL modeling platform called Renaissance that we use to conduct all of the experiments. This program offers a great deal of flexibility in creating, training and evaluating transformer encoders for VL modeling. The source code for Renaissance can be found at https://github.com/bsu-slim/renaissance.

Paperid: 1647, https://arxiv.org/pdf/2411.06287.pdf GitHub

Abstract:
Despite the importance of shape perception in human vision, early neural image classifiers relied less on shape information for object recognition than other (often spurious) features. While recent research suggests that current large Vision-Language Models (VLMs) exhibit more reliance on shape, we find them to still be seriously limited in this regard. To quantify such limitations, we introduce IllusionBench, a dataset that challenges current cutting-edge VLMs to decipher shape information when the shape is represented by an arrangement of visual elements in a scene. Our extensive evaluations reveal that, while these shapes are easily detectable by human annotators, current VLMs struggle to recognize them, indicating important avenues for future work in developing more robust visual perception systems. The full dataset and codebase are available at: \url{https://arshiahemmat.github.io/illusionbench/}

Paperid: 1648, https://arxiv.org/pdf/2411.06071.pdf GitHub

Abstract:
Zero-shot anomaly detection (ZSAD) is crucial for detecting anomalous patterns in target datasets without using training samples, specifically in scenarios where there are distributional differences between the target domain and training data or where data scarcity arises because of restricted access. Although recently pretrained vision-language models demonstrate strong zero-shot performance across various visual tasks, they focus on learning class semantics, which makes their direct application to ZSAD challenging. To address this scenario, we propose GlocalCLIP, which uniquely separates global and local prompts and jointly optimizes them. This approach enables the object-agnostic glocal semantic prompt to effectively capture general normal and anomalous patterns without dependency on specific objects in the image. We refine the text prompts for more precise adjustments by utilizing deep-text prompt tuning in the text encoder. In the vision encoder, we apply V-V attention layers to capture detailed local image features. Finally, we introduce glocal contrastive learning to improve the complementary learning of global and local prompts, effectively detecting anomalous patterns across various domains. The generalization performance of GlocalCLIP in ZSAD was demonstrated on 15 real-world datasets from both the industrial and medical domains, achieving superior performance compared to existing methods. Code will be made available at https://github.com/YUL-git/GlocalCLIP.

Paperid: 1649, https://arxiv.org/pdf/2411.05755.pdf GitHub

Abstract:
We present VLMnav, an embodied framework to transform a Vision-Language Model (VLM) into an end-to-end navigation policy. In contrast to prior work, we do not rely on a separation between perception, planning, and control; instead, we use a VLM to directly select actions in one step. Surprisingly, we find that a VLM can be used as an end-to-end policy zero-shot, i.e., without any fine-tuning or exposure to navigation data. This makes our approach open-ended and generalizable to any downstream navigation task. We run an extensive study to evaluate the performance of our approach in comparison to baseline prompting methods. In addition, we perform a design analysis to understand the most impactful design decisions. Visual examples and code for our project can be found at https://jirl-upenn.github.io/VLMnav/

Paperid: 1650, https://arxiv.org/pdf/2411.05609.pdf GitHub

Abstract:
The main challenges hindering the adoption of deep learning-based systems in clinical settings are the scarcity of annotated data and the lack of interpretability and trust in these systems. Concept Bottleneck Models (CBMs) offer inherent interpretability by constraining the final disease prediction on a set of human-understandable concepts. However, this inherent interpretability comes at the cost of greater annotation burden. Additionally, adding new concepts requires retraining the entire system. In this work, we introduce a novel two-step methodology that addresses both of these challenges. By simulating the two stages of a CBM, we utilize a pretrained Vision Language Model (VLM) to automatically predict clinical concepts, and an off-the-shelf Large Language Model (LLM) to generate disease diagnoses based on the predicted concepts. Furthermore, our approach supports test-time human intervention, enabling corrections to predicted concepts, which improves final diagnoses and enhances transparency in decision-making. We validate our approach on three skin lesion datasets, demonstrating that it outperforms traditional CBMs and state-of-the-art explainable methods, all without requiring any training and utilizing only a few annotated examples. The code is available at https://github.com/CristianoPatricio/2-step-concept-based-skin-diagnosis.

Paperid: 1651, https://arxiv.org/pdf/2411.05079.pdf GitHub

Abstract:
Despite advancements in text-to-image models, generating images that precisely align with textual descriptions remains challenging due to misalignment in training data. In this paper, we analyze the critical role of caption precision and recall in text-to-image model training. Our analysis of human-annotated captions shows that both precision and recall are important for text-image alignment, but precision has a more significant impact. Leveraging these insights, we utilize Large Vision Language Models to generate synthetic captions for training. Models trained with these synthetic captions show similar behavior to those trained on human-annotated captions, underscores the potential for synthetic data in text-to-image training.

Paperid: 1652, https://arxiv.org/pdf/2411.04713.pdf GitHub

Abstract:
High-quality training triplets (instruction, original image, edited image) are essential for instruction-based image editing. Predominant training datasets (e.g., InsPix2Pix) are created using text-to-image generative models (e.g., Stable Diffusion, DALL-E) which are not trained for image editing. Accordingly, these datasets suffer from inaccurate instruction following, poor detail preserving, and generation artifacts. In this paper, we propose to address the training data quality issue with multi-perspective reward data instead of refining the ground-truth image quality. 1) we first design a quantitative metric system based on best-in-class LVLM (Large Vision Language Model), i.e., GPT-4o in our case, to evaluate the generation quality from 3 perspectives, namely, instruction following, detail preserving, and generation quality. For each perspective, we collected quantitative score in $0\sim 5$ and text descriptive feedback on the specific failure points in ground-truth edited images, resulting in a high-quality editing reward dataset, i.e., RewardEdit20K. 2) We further proposed a novel training framework to seamlessly integrate the metric output, regarded as multi-reward, into editing models to learn from the imperfect training triplets. During training, the reward scores and text descriptions are encoded as embeddings and fed into both the latent space and the U-Net of the editing models as auxiliary conditions. 3) We also build a challenging evaluation benchmark with real-world images/photos and diverse editing instructions, named Real-Edit. Experiments indicate that our multi-reward conditioned model outperforms its no-reward counterpart on two popular editing pipelines, i.e., InsPix2Pix and SmartEdit. Code is released at https://github.com/bytedance/Multi-Reward-Editing.

Paperid: 1653, https://arxiv.org/pdf/2411.03888.pdf GitHub

Abstract:
Warning: this paper contains content that may be offensive or upsetting Hate speech moderation on global platforms poses unique challenges due to the multimodal and multilingual nature of content, along with the varying cultural perceptions. How well do current vision-language models (VLMs) navigate these nuances? To investigate this, we create the first multimodal and multilingual parallel hate speech dataset, annotated by a multicultural set of annotators, called Multi3Hate. It contains 300 parallel meme samples across 5 languages: English, German, Spanish, Hindi, and Mandarin. We demonstrate that cultural background significantly affects multimodal hate speech annotation in our dataset. The average pairwise agreement among countries is just 74%, significantly lower than that of randomly selected annotator groups. Our qualitative analysis indicates that the lowest pairwise label agreement-only 67% between the USA and India-can be attributed to cultural factors. We then conduct experiments with 5 large VLMs in a zero-shot setting, finding that these models align more closely with annotations from the US than with those from other cultures, even when the memes and prompts are presented in the dominant language of the other culture. Code and dataset are available at https://github.com/MinhDucBui/Multi3Hate.

Paperid: 1654, https://arxiv.org/pdf/2411.03540.pdf GitHub

Abstract:
With the recent rise of Large Language Models (LLMs), Vision-Language Models (VLMs), and other general foundation models, there is growing potential for multimodal, multi-task embodied agents that can operate in diverse environments given only natural language as input. One such application area is indoor navigation using natural language instructions. However, despite recent progress, this problem remains challenging due to the spatial reasoning and semantic understanding required, particularly in arbitrary scenes that may contain many objects belonging to fine-grained classes. To address this challenge, we curate the largest real-world dataset for Vision and Language-guided Action in 3D Scenes (VLA-3D), consisting of over 11.5K scanned 3D indoor rooms from existing datasets, 23.5M heuristically generated semantic relations between objects, and 9.7M synthetically generated referential statements. Our dataset consists of processed 3D point clouds, semantic object and room annotations, scene graphs, navigable free space annotations, and referential language statements that specifically focus on view-independent spatial relations for disambiguating objects. The goal of these features is to aid the downstream task of navigation, especially on real-world systems where some level of robustness must be guaranteed in an open world of changing scenes and imperfect language. We benchmark our dataset with current state-of-the-art models to obtain a performance baseline. All code to generate and visualize the dataset is publicly released, see https://github.com/HaochenZ11/VLA-3D. With the release of this dataset, we hope to provide a resource for progress in semantic 3D scene understanding that is robust to changes and one which will aid the development of interactive indoor navigation systems.

Paperid: 1655, https://arxiv.org/pdf/2411.03409.pdf GitHub

Abstract:
The complexity of the real world demands robotic systems that can intelligently adapt to unseen situations. We present STEER, a robot learning framework that bridges high-level, commonsense reasoning with precise, flexible low-level control. Our approach translates complex situational awareness into actionable low-level behavior through training language-grounded policies with dense annotation. By structuring policy training around fundamental, modular manipulation skills expressed in natural language, STEER exposes an expressive interface for humans or Vision-Language Models (VLMs) to intelligently orchestrate the robot's behavior by reasoning about the task and context. Our experiments demonstrate the skills learned via STEER can be combined to synthesize novel behaviors to adapt to new situations or perform completely new tasks without additional data collection or training.

Paperid: 1656, https://arxiv.org/pdf/2411.03313.pdf GitHub

Abstract:
We introduce SuperClass, a super simple classification method for vision-language pre-training on image-text data. Unlike its contrastive counterpart CLIP who contrast with a text encoder, SuperClass directly utilizes tokenized raw text as supervised classification labels, without the need for additional text filtering or selection. Due to the absence of the text encoding as contrastive target, SuperClass does not require a text encoder and does not need to maintain a large batch size as CLIP does. SuperClass demonstrated superior performance on various downstream tasks, including classic computer vision benchmarks and vision language downstream tasks. We further explored the scaling behavior of SuperClass on model size, training length, or data size, and reported encouraging results and comparisons to CLIP. https://github.com/x-cls/superclass

Paperid: 1657, https://arxiv.org/pdf/2411.03312.pdf GitHub

Abstract:
Vision Language Models (VLMs) have demonstrated strong capabilities across various visual understanding and reasoning tasks, driven by incorporating image representations into the token inputs of Large Language Models (LLMs). However, their real-world deployment is often constrained by high latency during inference due to the substantial compute required by the LLM to process the large number of input tokens, predominantly arising from the image. To reduce inference costs, one can either downsize the LLM or reduce the number of input tokens needed to represent the image, the latter of which has been the focus of many recent efforts around token compression. However, it is unclear what the optimal trade-off is given a fixed inference budget. We first characterize this optimal trade-off between the number of visual tokens and LLM parameters by establishing scaling laws that capture variations in performance with these two factors. Our results reveal a surprising trend: for visual reasoning tasks, the inference-optimal behavior in VLMs is achieved by using the largest LLM that fits within the inference budget while minimizing visual token count - often to a single token. While the token reduction literature has mainly focused on maintaining base model performance by modestly reducing the token count (e.g., $5-10\times$), our results indicate that the compute-optimal inference regime requires operating under even higher token compression ratios. Based on these insights, we take the first steps toward designing token compression algorithms tailored for high-compression settings, utilizing prompt-based compression of tokens. Our work underscores the performance and efficiency benefits of operating in low visual token regimes and the importance of developing tailored token reduction algorithms for such conditions. Code is available at https://github.com/locuslab/llava-token-compression.

Paperid: 1658, https://arxiv.org/pdf/2411.02902.pdf GitHub

Abstract:
Large vision-language models (VLLMs) exhibit promising capabilities for processing multi-modal tasks across various application scenarios. However, their emergence also raises significant data security concerns, given the potential inclusion of sensitive information, such as private photos and medical records, in their training datasets. Detecting inappropriately used data in VLLMs remains a critical and unresolved issue, mainly due to the lack of standardized datasets and suitable methodologies. In this study, we introduce the first membership inference attack (MIA) benchmark tailored for various VLLMs to facilitate training data detection. Then, we propose a novel MIA pipeline specifically designed for token-level image detection. Lastly, we present a new metric called MaxRÃ©nyi-K%, which is based on the confidence of the model output and applies to both text and image data. We believe that our work can deepen the understanding and methodology of MIAs in the context of VLLMs. Our code and datasets are available at https://github.com/LIONS-EPFL/VL-MIA.

Paperid: 1659, https://arxiv.org/pdf/2411.02733.pdf GitHub

Abstract:
With the rapid development of large vision language models (LVLMs), these models have shown excellent results in various multimodal tasks. Since LVLMs are prone to hallucinations and there are currently few datasets and evaluation methods specifically designed for remote sensing, their performance is typically poor when applied to remote sensing tasks. To address these issues, this paper introduces a high quality remote sensing LVLMs dataset, DDFAV, created using data augmentation and data mixing strategies. Next, a training instruction set is produced based on some high-quality remote sensing images selected from the proposed dataset. Finally, we develop a remote sensing LVLMs hallucination evaluation method RSPOPE based on the proposed dataset and evaluate the zero-shot capabilities of different LVLMs. Our proposed dataset, instruction set, and evaluation method files are available at https://github.com/HaodongLi2024/rspope.

Paperid: 1660, https://arxiv.org/pdf/2411.02712.pdf GitHub

Abstract:
Large vision-language models (LVLMs) suffer from hallucination, resulting in misalignment between the output textual response and the input visual content. Recent research indicates that the over-reliance on the Large Language Model (LLM) backbone, as one cause of the LVLM hallucination, inherently introduces bias from language priors, leading to insufficient context attention to the visual inputs. We tackle this issue of hallucination by mitigating such over-reliance through preference learning. We propose Vision-guided Direct Preference Optimization (V-DPO) to enhance visual context learning at training time. To interpret the effectiveness and generalizability of V-DPO on different types of training data, we construct a synthetic dataset containing both response- and image-contrast preference pairs, compared against existing human-annotated hallucination samples. Our approach achieves significant improvements compared with baseline methods across various hallucination benchmarks. Our analysis indicates that V-DPO excels in learning from image-contrast preference data, demonstrating its superior ability to elicit and understand nuances of visual context. Our code is publicly available at https://github.com/YuxiXie/V-DPO.

Paperid: 1661, https://arxiv.org/pdf/2411.02669.pdf GitHub

Abstract:
Vision-language pre-training (VLP) models excel at interpreting both images and text but remain vulnerable to multimodal adversarial examples (AEs). Advancing the generation of transferable AEs, which succeed across unseen models, is key to developing more robust and practical VLP models. Previous approaches augment image-text pairs to enhance diversity within the adversarial example generation process, aiming to improve transferability by expanding the contrast space of image-text features. However, these methods focus solely on diversity around the current AEs, yielding limited gains in transferability. To address this issue, we propose to increase the diversity of AEs by leveraging the intersection regions along the adversarial trajectory during optimization. Specifically, we propose sampling from adversarial evolution triangles composed of clean, historical, and current adversarial examples to enhance adversarial diversity. We provide a theoretical analysis to demonstrate the effectiveness of the proposed adversarial evolution triangle. Moreover, we find that redundant inactive dimensions can dominate similarity calculations, distorting feature matching and making AEs model-dependent with reduced transferability. Hence, we propose to generate AEs in the semantic image-text feature contrast space, which can project the original feature space into a semantic corpus subspace. The proposed semantic-aligned subspace can reduce the image feature redundancy, thereby improving adversarial transferability. Extensive experiments across different datasets and models demonstrate that the proposed method can effectively improve adversarial transferability and outperform state-of-the-art adversarial attack methods. The code is released at https://github.com/jiaxiaojunQAQ/SA-AET.

Paperid: 1662, https://arxiv.org/pdf/2411.02359.pdf GitHub

Abstract:
MLLMs have demonstrated remarkable comprehension and reasoning capabilities with complex language and visual data. These advances have spurred the vision of establishing a generalist robotic MLLM proficient in understanding complex human instructions and accomplishing various embodied tasks. However, developing MLLMs for real-world robots is challenging due to the typically limited computation and memory capacities available on robotic platforms. In contrast, the inference of MLLMs involves storing billions of parameters and performing tremendous computation, imposing significant hardware demands. In our paper, we propose a Dynamic Early-Exit Framework for Robotic Vision-Language-Action Model (DeeR-VLA, or simply DeeR) that automatically adjusts the size of the activated MLLM based on each situation at hand. The approach leverages a multi-exit architecture in MLLMs, which allows the model to terminate processing once a proper size of the model has been activated for a specific situation, thus avoiding further redundant computation. Additionally, we develop novel algorithms that establish early-termination criteria for DeeR, conditioned on predefined demands such as average computational cost (i.e., power consumption), as well as peak computational consumption (i.e., latency) and GPU memory usage. These enhancements ensure that DeeR operates efficiently under varying resource constraints while maintaining competitive performance. On the CALVIN robot manipulation benchmark, DeeR demonstrates significant reductions in computational costs of LLM by 5.2-6.5x and GPU memory of LLM by 2-6x without compromising performance. Code and checkpoints are available at https://github.com/yueyang130/DeeR-VLA.

Paperid: 1663, https://arxiv.org/pdf/2411.00871.pdf GitHub

Abstract:
Large Language Models (LLMs) have demonstrated remarkable generalization and instruction-following capabilities with instruction tuning. The advancements in LLMs and instruction tuning have led to the development of Large Vision-Language Models (LVLMs). However, the competency of the LLMs and instruction tuning have been less explored in the molecular domain. Thus, we propose LLaMo: Large Language Model-based Molecular graph assistant, which is an end-to-end trained large molecular graph-language model. To bridge the discrepancy between the language and graph modalities, we present the multi-level graph projector that transforms graph representations into graph tokens by abstracting the output representations of each GNN layer and motif representations with the cross-attention mechanism. We also introduce machine-generated molecular graph instruction data to instruction-tune the large molecular graph-language model for general-purpose molecule and language understanding. Our extensive experiments demonstrate that LLaMo shows the best performance on diverse tasks, such as molecular description generation, property prediction, and IUPAC name prediction. The code of LLaMo is available at https://github.com/mlvlab/LLaMo.

Paperid: 1664, https://arxiv.org/pdf/2411.00827.pdf GitHub

Abstract:
As large Vision-Language Models (VLMs) gain prominence, ensuring their safe deployment has become critical. Recent studies have explored VLM robustness against jailbreak attacks-techniques that exploit model vulnerabilities to elicit harmful outputs. However, the limited availability of diverse multimodal data has constrained current approaches to rely heavily on adversarial or manually crafted images derived from harmful text datasets, which often lack effectiveness and diversity across different contexts. In this paper, we propose IDEATOR, a novel jailbreak method that autonomously generates malicious image-text pairs for black-box jailbreak attacks. IDEATOR is grounded in the insight that VLMs themselves could serve as powerful red team models for generating multimodal jailbreak prompts. Specifically, IDEATOR leverages a VLM to create targeted jailbreak texts and pairs them with jailbreak images generated by a state-of-the-art diffusion model. Extensive experiments demonstrate IDEATOR's high effectiveness and transferability, achieving a 94% attack success rate (ASR) in jailbreaking MiniGPT-4 with an average of only 5.34 queries, and high ASRs of 82%, 88%, and 75% when transferred to LLaVA, InstructBLIP, and Chameleon, respectively. Building on IDEATOR's strong transferability and automated process, we introduce the VLJailbreakBench, a safety benchmark comprising 3,654 multimodal jailbreak samples. Our benchmark results on 11 recently released VLMs reveal significant gaps in safety alignment. For instance, our challenge set achieves ASRs of 46.31% on GPT-4o and 19.65% on Claude-3.5-Sonnet, underscoring the urgent need for stronger defenses. VLJailbreakBench is publicly available at https://roywang021.github.io/VLJailbreakBench.

Paperid: 1665, https://arxiv.org/pdf/2410.24018.pdf GitHub

Abstract:
Visual reprogramming (VR) leverages the intrinsic capabilities of pretrained vision models by adapting their input or output interfaces to solve downstream tasks whose labels (i.e., downstream labels) might be totally different from the labels associated with the pretrained models (i.e., pretrained labels). When adapting the output interface, label mapping methods transform the pretrained labels to downstream labels by establishing a gradient-free one-to-one correspondence between the two sets of labels. However, in this paper, we reveal that one-to-one mappings may overlook the complex relationship between pretrained and downstream labels. Motivated by this observation, we propose a Bayesian-guided Label Mapping (BLM) method. BLM constructs an iteratively-updated probabilistic label mapping matrix, with each element quantifying a pairwise relationship between pretrained and downstream labels. The assignment of values to the constructed matrix is guided by Bayesian conditional probability, considering the joint distribution of the downstream labels and the labels predicted by the pretrained model on downstream samples. Experiments conducted on both pretrained vision models (e.g., ResNeXt) and vision-language models (e.g., CLIP) demonstrate the superior performance of BLM over existing label mapping methods. The success of BLM also offers a probabilistic lens through which to understand and analyze the effectiveness of VR. Our code is available at https://github.com/tmlr-group/BayesianLM.

Paperid: 1666, https://arxiv.org/pdf/2410.23996.pdf GitHub

Abstract:
Multimodal representation learning seeks to relate and decompose information inherent in multiple modalities. By disentangling modality-specific information from information that is shared across modalities, we can improve interpretability and robustness and enable downstream tasks such as the generation of counterfactual outcomes. Separating the two types of information is challenging since they are often deeply entangled in many real-world applications. We propose Disentangled Self-Supervised Learning (DisentangledSSL), a novel self-supervised approach for learning disentangled representations. We present a comprehensive analysis of the optimality of each disentangled representation, particularly focusing on the scenario not covered in prior work where the so-called Minimum Necessary Information (MNI) point is not attainable. We demonstrate that DisentangledSSL successfully learns shared and modality-specific features on multiple synthetic and real-world datasets and consistently outperforms baselines on various downstream tasks, including prediction tasks for vision-language data, as well as molecule-phenotype retrieval tasks for biological data. The code is available at https://github.com/uhlerlab/DisentangledSSL.

Paperid: 1667, https://arxiv.org/pdf/2410.23904.pdf GitHub

Abstract:
Detecting Human-Object Interactions (HOI) in zero-shot settings, where models must handle unseen classes, poses significant challenges. Existing methods that rely on aligning visual encoders with large Vision-Language Models (VLMs) to tap into the extensive knowledge of VLMs, require large, computationally expensive models and encounter training difficulties. Adapting VLMs with prompt learning offers an alternative to direct alignment. However, fine-tuning on task-specific datasets often leads to overfitting to seen classes and suboptimal performance on unseen classes, due to the absence of unseen class labels. To address these challenges, we introduce a novel prompt learning-based framework for Efficient Zero-Shot HOI detection (EZ-HOI). First, we introduce Large Language Model (LLM) and VLM guidance for learnable prompts, integrating detailed HOI descriptions and visual semantics to adapt VLMs to HOI tasks. However, because training datasets contain seen-class labels alone, fine-tuning VLMs on such datasets tends to optimize learnable prompts for seen classes instead of unseen ones. Therefore, we design prompt learning for unseen classes using information from related seen classes, with LLMs utilized to highlight the differences between unseen and related seen classes. Quantitative evaluations on benchmark datasets demonstrate that our EZ-HOI achieves state-of-the-art performance across various zero-shot settings with only 10.35% to 33.95% of the trainable parameters compared to existing methods. Code is available at https://github.com/ChelsieLei/EZ-HOI.

Paperid: 1668, https://arxiv.org/pdf/2410.23643.pdf GitHub

Abstract:
Careful robot manipulation in every-day cluttered environments requires an accurate understanding of the 3D scene, in order to grasp and place objects stably and reliably and to avoid colliding with other objects. In general, we must construct such a 3D interpretation of a complex scene based on limited input, such as a single RGB-D image. We describe SceneComplete, a system for constructing a complete, segmented, 3D model of a scene from a single view. SceneComplete is a novel pipeline for composing general-purpose pretrained perception modules (vision-language, segmentation, image-inpainting, image-to-3D, visual-descriptors and pose-estimation) to obtain highly accurate results. We demonstrate its accuracy and effectiveness with respect to ground-truth models in a large benchmark dataset and show that its accurate whole-object reconstruction enables robust grasp proposal generation, including for a dexterous hand. We release the code on our website https://scenecomplete.github.io/.

Paperid: 1669, https://arxiv.org/pdf/2410.23114.pdf GitHub GitHub

Abstract:
Despite the outstanding performance in vision-language reasoning, Large Vision-Language Models (LVLMs) might generate hallucinated contents that do not exist in the given image. Most existing LVLM hallucination benchmarks are constrained to evaluate the object-related hallucinations. However, the potential hallucination on the relations between two objects, i.e., relation hallucination, still lacks investigation. To remedy that, we design a unified framework to measure the object and relation hallucination in LVLMs simultaneously. The core idea of our framework is to evaluate hallucinations via (object, relation, object) triplets extracted from LVLMs' responses, making it easily generalizable to different vision-language tasks. Based on our framework, we further introduce Tri-HE, a novel Triplet-level Hallucination Evaluation benchmark which can be used to study both object and relation hallucination at the same time. With comprehensive evaluations on Tri-HE, we observe that the relation hallucination issue is even more serious than object hallucination among existing LVLMs, highlighting a previously neglected problem towards reliable LVLMs. Moreover, based on our findings, we design a simple training-free approach that effectively mitigates hallucinations for LVLMs. Our dataset and code for the reproduction of our experiments are available publicly at https://github.com/wujunjie1998/Tri-HE.

Paperid: 1670, https://arxiv.org/pdf/2410.22927.pdf GitHub

Abstract:
Reliable re-identification of individuals within large wildlife populations is crucial for biological studies, ecological research, and wildlife conservation. Classic computer vision techniques offer a promising direction for Animal Re-identification (Animal ReID), but their backbones' close-set nature limits their applicability and generalizability. Despite the demonstrated effectiveness of vision-language models like CLIP in re-identifying persons and vehicles, their application to Animal ReID remains limited due to unique challenges, such as the various visual representations of animals, including variations in poses and forms. To address these limitations, we leverage CLIP's cross-modal capabilities to introduce a two-stage framework, the \textbf{Indiv}idual \textbf{A}nimal \textbf{ID}entity-Driven (IndivAID) framework, specifically designed for Animal ReID. In the first stage, IndivAID trains a text description generator by extracting individual semantic information from each image, generating both image-specific and individual-specific textual descriptions that fully capture the diverse visual concepts of each individual across animal images. In the second stage, IndivAID refines its learning of visual concepts by dynamically incorporating individual-specific textual descriptions with an integrated attention module to further highlight discriminative features of individuals for Animal ReID. Evaluation against state-of-the-art methods across eight benchmark datasets and a real-world Stoat dataset demonstrates IndivAID's effectiveness and applicability. Code is available at \url{https://github.com/ywu840/IndivAID}.

Paperid: 1671, https://arxiv.org/pdf/2410.22888.pdf GitHub

Abstract:
Visual Language Models (VLMs) are vulnerable to adversarial attacks, especially those from adversarial images, which is however under-explored in literature. To facilitate research on this critical safety problem, we first construct a new laRge-scale Adervsarial images dataset with Diverse hArmful Responses (RADAR), given that existing datasets are either small-scale or only contain limited types of harmful responses. With the new RADAR dataset, we further develop a novel and effective iN-time Embedding-based AdveRSarial Image DEtection (NEARSIDE) method, which exploits a single vector that distilled from the hidden states of VLMs, which we call the attacking direction, to achieve the detection of adversarial images against benign ones in the input. Extensive experiments with two victim VLMs, LLaVA and MiniGPT-4, well demonstrate the effectiveness, efficiency, and cross-model transferrability of our proposed method. Our code is available at https://github.com/mob-scu/RADAR-NEARSIDE

Paperid: 1672, https://arxiv.org/pdf/2410.22362.pdf GitHub

Abstract:
Recently, the diffusion-based generative paradigm has achieved impressive general image generation capabilities with text prompts due to its accurate distribution modeling and stable training process. However, generating diverse remote sensing (RS) images that are tremendously different from general images in terms of scale and perspective remains a formidable challenge due to the lack of a comprehensive remote sensing image generation dataset with various modalities, ground sample distances (GSD), and scenes. In this paper, we propose a Multi-modal, Multi-GSD, Multi-scene Remote Sensing (MMM-RS) dataset and benchmark for text-to-image generation in diverse remote sensing scenarios. Specifically, we first collect nine publicly available RS datasets and conduct standardization for all samples. To bridge RS images to textual semantic information, we utilize a large-scale pretrained vision-language model to automatically output text prompts and perform hand-crafted rectification, resulting in information-rich text-image pairs (including multi-modal images). In particular, we design some methods to obtain the images with different GSD and various environments (e.g., low-light, foggy) in a single sample. With extensive manual screening and refining annotations, we ultimately obtain a MMM-RS dataset that comprises approximately 2.1 million text-image pairs. Extensive experimental results verify that our proposed MMM-RS dataset allows off-the-shelf diffusion models to generate diverse RS images across various modalities, scenes, weather conditions, and GSD. The dataset is available at https://github.com/ljl5261/MMM-RS.

Paperid: 1673, https://arxiv.org/pdf/2410.22332.pdf GitHub

Abstract:
Sim2real for robotic manipulation is difficult due to the challenges of simulating complex contacts and generating realistic task distributions. To tackle the latter problem, we introduce ManipGen, which leverages a new class of policies for sim2real transfer: local policies. Locality enables a variety of appealing properties including invariances to absolute robot and object pose, skill ordering, and global scene configuration. We combine these policies with foundation models for vision, language and motion planning and demonstrate SOTA zero-shot performance of our method to Robosuite benchmark tasks in simulation (97%). We transfer our local policies from simulation to reality and observe they can solve unseen long-horizon manipulation tasks with up to 8 stages with significant pose, object and scene configuration variation. ManipGen outperforms SOTA approaches such as SayCan, OpenVLA, LLMTrajGen and VoxPoser across 50 real-world manipulation tasks by 36%, 76%, 62% and 60% respectively. Video results at https://mihdalal.github.io/manipgen/

Paperid: 1674, https://arxiv.org/pdf/2410.22317.pdf GitHub

Abstract:
With the advent of large pre-trained vision-language models such as CLIP, prompt learning methods aim to enhance the transferability of the CLIP model. They learn the prompt given few samples from the downstream task given the specific class names as prior knowledge, which we term as semantic-aware classification. However, in many realistic scenarios, we only have access to few samples and knowledge of the class names (e.g., when considering instances of classes). This challenging scenario represents the semantic-agnostic discriminative case. Text-to-Image (T2I) personalization methods aim to adapt T2I models to unseen concepts by learning new tokens and endowing these tokens with the capability of generating the learned concepts. These methods do not require knowledge of class names as a semantic-aware prior. Therefore, in this paper, we first explore Textual Inversion and reveal that the new concept tokens possess both generation and classification capabilities by regarding each category as a single concept. However, learning classifiers from single-concept textual inversion is limited since the learned tokens are suboptimal for the discriminative tasks. To mitigate this issue, we propose Multi-Class textual inversion, which includes a discriminative regularization term for the token updating process. Using this technique, our method MC-TI achieves stronger Semantic-Agnostic Classification while preserving the generation capability of these modifier tokens given only few samples per category. In the experiments, we extensively evaluate MC-TI on 12 datasets covering various scenarios, which demonstrates that MC-TI achieves superior results in terms of both classification and generation outcomes.

Paperid: 1675, https://arxiv.org/pdf/2410.22313.pdf GitHub GitHub

Abstract:
End-to-end autonomous driving demonstrates strong planning capabilities with large-scale data but still struggles in complex, rare scenarios due to limited commonsense. In contrast, Large Vision-Language Models (LVLMs) excel in scene understanding and reasoning. The path forward lies in merging the strengths of both approaches. Previous methods using LVLMs to predict trajectories or control signals yield suboptimal results, as LVLMs are not well-suited for precise numerical predictions. This paper presents Senna, an autonomous driving system combining an LVLM (Senna-VLM) with an end-to-end model (Senna-E2E). Senna decouples high-level planning from low-level trajectory prediction. Senna-VLM generates planning decisions in natural language, while Senna-E2E predicts precise trajectories. Senna-VLM utilizes a multi-image encoding approach and multi-view prompts for efficient scene understanding. Besides, we introduce planning-oriented QAs alongside a three-stage training strategy, which enhances Senna-VLM's planning performance while preserving commonsense. Extensive experiments on two datasets show that Senna achieves state-of-the-art planning performance. Notably, with pre-training on a large-scale dataset DriveX and fine-tuning on nuScenes, Senna significantly reduces average planning error by 27.12% and collision rate by 33.33% over model without pre-training. We believe Senna's cross-scenario generalization and transferability are essential for achieving fully autonomous driving. Code and models will be released at https://github.com/hustvl/Senna.

Paperid: 1676, https://arxiv.org/pdf/2410.21969.pdf GitHub

Abstract:
Medical Vision-Language Pretraining (MedVLP) shows promise in learning generalizable and transferable visual representations from paired and unpaired medical images and reports. MedVLP can provide useful features to downstream tasks and facilitate adapting task-specific models to new setups using fewer examples. However, existing MedVLP methods often differ in terms of datasets, preprocessing, and finetuning implementations. This pose great challenges in evaluating how well a MedVLP method generalizes to various clinically-relevant tasks due to the lack of unified, standardized, and comprehensive benchmark. To fill this gap, we propose BenchX, a unified benchmark framework that enables head-to-head comparison and systematical analysis between MedVLP methods using public chest X-ray datasets. Specifically, BenchX is composed of three components: 1) Comprehensive datasets covering nine datasets and four medical tasks; 2) Benchmark suites to standardize data preprocessing, train-test splits, and parameter selection; 3) Unified finetuning protocols that accommodate heterogeneous MedVLP methods for consistent task adaptation in classification, segmentation, and report generation, respectively. Utilizing BenchX, we establish baselines for nine state-of-the-art MedVLP methods and found that the performance of some early MedVLP methods can be enhanced to surpass more recent ones, prompting a revisiting of the developments and conclusions from prior works in MedVLP. Our code are available at https://github.com/yangzhou12/BenchX.

Paperid: 1677, https://arxiv.org/pdf/2410.21802.pdf GitHub

Abstract:
Due to the impressive zero-shot capabilities, pre-trained vision-language models (e.g. CLIP), have attracted widespread attention and adoption across various domains. Nonetheless, CLIP has been observed to be susceptible to adversarial examples. Through experimental analysis, we have observed a phenomenon wherein adversarial perturbations induce shifts in text-guided attention. Building upon this observation, we propose a simple yet effective strategy: Text-Guided Attention for Zero-Shot Robustness (TGA-ZSR). This framework incorporates two components: the Attention Refinement module and the Attention-based Model Constraint module. Our goal is to maintain the generalization of the CLIP model and enhance its adversarial robustness: The Attention Refinement module aligns the text-guided attention obtained from the target model via adversarial examples with the text-guided attention acquired from the original model via clean examples. This alignment enhances the model's robustness. Additionally, the Attention-based Model Constraint module acquires text-guided attention from both the target and original models using clean examples. Its objective is to maintain model performance on clean samples while enhancing overall robustness. The experiments validate that our method yields a 9.58% enhancement in zero-shot robust accuracy over the current state-of-the-art techniques across 16 datasets. Our code is available at https://github.com/zhyblue424/TGA-ZSR.

Paperid: 1678, https://arxiv.org/pdf/2410.21220.pdf GitHub

Abstract:
Search engines enable the retrieval of unknown information with texts. However, traditional methods fall short when it comes to understanding unfamiliar visual content, such as identifying an object that the model has never seen before. This challenge is particularly pronounced for large vision-language models (VLMs): if the model has not been exposed to the object depicted in an image, it struggles to generate reliable answers to the user's question regarding that image. Moreover, as new objects and events continuously emerge, frequently updating VLMs is impractical due to heavy computational burdens. To address this limitation, we propose Vision Search Assistant, a novel framework that facilitates collaboration between VLMs and web agents. This approach leverages VLMs' visual understanding capabilities and web agents' real-time information access to perform open-world Retrieval-Augmented Generation via the web. By integrating visual and textual representations through this collaboration, the model can provide informed responses even when the image is novel to the system. Extensive experiments conducted on both open-set and closed-set QA benchmarks demonstrate that the Vision Search Assistant significantly outperforms the other models and can be widely applied to existing VLMs.

Paperid: 1679, https://arxiv.org/pdf/2410.20971.pdf GitHub

Abstract:
In this paper, we focus on black-box defense for VLMs against jailbreak attacks. Existing black-box defense methods are either unimodal or bimodal. Unimodal methods enhance either the vision or language module of the VLM, while bimodal methods robustify the model through text-image representation realignment. However, these methods suffer from two limitations: 1) they fail to fully exploit the cross-modal information, or 2) they degrade the model performance on benign inputs. To address these limitations, we propose a novel blue-team method BlueSuffix that defends target VLMs against jailbreak attacks without compromising its performance under black-box setting. BlueSuffix includes three key components: 1) a visual purifier against jailbreak images, 2) a textual purifier against jailbreak texts, and 3) a blue-team suffix generator using reinforcement fine-tuning for enhancing cross-modal robustness. We empirically show on four VLMs (LLaVA, MiniGPT-4, InstructionBLIP, and Gemini) and four safety benchmarks (Harmful Instruction, AdvBench, MM-SafetyBench, and RedTeam-2K) that BlueSuffix outperforms the baseline defenses by a significant margin. Our BlueSuffix opens up a promising direction for defending VLMs against jailbreak attacks. Code is available at https://github.com/Vinsonzyh/BlueSuffix.

Paperid: 1680, https://arxiv.org/pdf/2410.20494.pdf GitHub

Abstract:
Multimodal information extraction (MIE) is crucial for scientific literature, where valuable data is often spread across text, figures, and tables. In materials science, extracting structured information from research articles can accelerate the discovery of new materials. However, the multimodal nature and complex interconnections of scientific content present challenges for traditional text-based methods. We introduce \textsc{MatViX}, a benchmark consisting of $324$ full-length research articles and $1,688$ complex structured JSON files, carefully curated by domain experts. These JSON files are extracted from text, tables, and figures in full-length documents, providing a comprehensive challenge for MIE. We introduce an evaluation method to assess the accuracy of curve similarity and the alignment of hierarchical structures. Additionally, we benchmark vision-language models (VLMs) in a zero-shot manner, capable of processing long contexts and multimodal inputs, and show that using a specialized model (DePlot) can improve performance in extracting curves. Our results demonstrate significant room for improvement in current models. Our dataset and evaluation code are available\footnote{\url{https://matvix-bench.github.io/}}.

Paperid: 1681, https://arxiv.org/pdf/2410.20263.pdf GitHub

Abstract:
Embodied Question Answering (EQA) is an essential yet challenging task for robot assistants. Large vision-language models (VLMs) have shown promise for EQA, but existing approaches either treat it as static video question answering without active exploration or restrict answers to a closed set of choices. These limitations hinder real-world applicability, where a robot must explore efficiently and provide accurate answers in open-vocabulary settings. To overcome these challenges, we introduce EfficientEQA, a novel framework that couples efficient exploration with free-form answer generation. EfficientEQA features three key innovations: (1) Semantic-Value-Weighted Frontier Exploration (SFE) with Verbalized Confidence (VC) from a black-box VLM to prioritize semantically important areas to explore, enabling the agent to gather relevant information faster; (2) a BLIP relevancy-based mechanism to stop adaptively by flagging highly relevant observations as outliers to indicate whether the agent has collected enough information; and (3) a Retrieval-Augmented Generation (RAG) method for the VLM to answer accurately based on pertinent images from the agent's observation history without relying on predefined choices. Our experimental results show that EfficientEQA achieves over 15% higher answer accuracy and requires over 20% fewer exploration steps than state-of-the-art methods. Our code is available at: https://github.com/chengkaiAcademyCity/EfficientEQA

Paperid: 1682, https://arxiv.org/pdf/2410.20149.pdf GitHub GitHub

Abstract:
Recent research has shown that pre-trained vision-language models are effective at identifying out-of-distribution (OOD) samples by using negative labels as guidance. However, employing consistent negative labels across different OOD datasets often results in semantic misalignments, as these text labels may not accurately reflect the actual space of OOD images. To overcome this issue, we introduce \textit{adaptive negative proxies}, which are dynamically generated during testing by exploring actual OOD images, to align more closely with the underlying OOD label space and enhance the efficacy of negative proxy guidance. Specifically, our approach utilizes a feature memory bank to selectively cache discriminative features from test images, representing the targeted OOD distribution. This facilitates the creation of proxies that can better align with specific OOD datasets. While task-adaptive proxies average features to reflect the unique characteristics of each dataset, the sample-adaptive proxies weight features based on their similarity to individual test samples, exploring detailed sample-level nuances. The final score for identifying OOD samples integrates static negative labels with our proposed adaptive proxies, effectively combining textual and visual knowledge for enhanced performance. Our method is training-free and annotation-free, and it maintains fast testing speed. Extensive experiments across various benchmarks demonstrate the effectiveness of our approach, abbreviated as AdaNeg. Notably, on the large-scale ImageNet benchmark, our AdaNeg significantly outperforms existing methods, with a 2.45\% increase in AUROC and a 6.48\% reduction in FPR95. Codes are available at \url{https://github.com/YBZh/OpenOOD-VLM}.

Paperid: 1683, https://arxiv.org/pdf/2410.19925.pdf GitHub

Abstract:
Generative large language models (LLMs) exhibit impressive capabilities, which can be further augmented by integrating a pre-trained vision model into the original LLM to create a multimodal LLM (MLLM). However, this integration often significantly decreases performance on natural language understanding and generation tasks, compared to the original LLM. This study investigates this issue using the LLaVA MLLM, treating the integration as a continual learning problem. We evaluate five continual learning methods to mitigate forgetting and identify a technique that enhances visual understanding while minimizing linguistic performance loss. Our approach reduces linguistic performance degradation by up to 15% over the LLaVA recipe, while maintaining high multimodal accuracy. We also demonstrate the robustness of our method through continual learning on a sequence of vision-language tasks, effectively preserving linguistic skills while acquiring new multimodal capabilities. Project webpage: https://shikhar-srivastava.github.io/cl-for-improving-mllms

Paperid: 1684, https://arxiv.org/pdf/2410.19808.pdf GitHub

Abstract:
Pre-trained large language models (LLMs) have been reliably integrated with visual input for multimodal tasks. The widespread adoption of instruction-tuned image-to-text vision-language assistants (VLAs) like LLaVA and InternVL necessitates evaluating gender biases. We study gender bias in 22 popular open-source VLAs with respect to personality traits, skills, and occupations. Our results show that VLAs replicate human biases likely present in the data, such as real-world occupational imbalances. Similarly, they tend to attribute more skills and positive personality traits to women than to men, and we see a consistent tendency to associate negative personality traits with men. To eliminate the gender bias in these models, we find that fine-tuning-based debiasing methods achieve the best trade-off between debiasing and retaining performance on downstream tasks. We argue for pre-deploying gender bias assessment in VLAs and motivate further development of debiasing strategies to ensure equitable societal outcomes. Code is available at https://github.com/ExplainableML/vla-gender-bias.

Paperid: 1686, https://arxiv.org/pdf/2410.19313.pdf GitHub

Abstract:
FP8 training has emerged as a promising method for improving training efficiency. Existing frameworks accelerate training by applying FP8 computation to linear layers while leaving optimizer states and activations in higher precision, which fails to fully optimize memory usage. This paper introduces COAT (Compressing Optimizer States and Activations for FP8 Training), a novel FP8 training framework designed to significantly reduce memory footprint when training large models. COAT addresses current limitations through two key innovations: (1) Dynamic Range Expansion, which aligns optimizer state distributions more closely with the FP8 representation range, thereby reducing quantization error, and (2) Mixed-Granularity Activation Quantization, which optimizes activation memory using a combination of per-tensor and per-group quantization strategies. Experiments demonstrate that COAT effectively reduces end-to-end training memory footprint by 1.54x compared to BF16 while achieving nearly lossless performance across various tasks, such as Large Language Model pretraining and fine-tuning and Vision Language Model training. COAT also achieves a 1.43x end-to-end training speedup compared to BF16, performing on par with or surpassing TransformerEngine's speedup. COAT enables efficient full-parameter training of large models on fewer GPUs, and facilitates doubling the batch size in distributed training settings, providing a practical solution for scaling large-scale model training. The code is available at https://github.com/NVlabs/COAT.

Paperid: 1687, https://arxiv.org/pdf/2410.19294.pdf GitHub

Abstract:
Vision-language models, such as CLIP, have shown impressive generalization capacities when using appropriate text descriptions. While optimizing prompts on downstream labeled data has proven effective in improving performance, these methods entail labor costs for annotations and are limited by their quality. Additionally, since CLIP is pre-trained on highly imbalanced Web-scale data, it suffers from inherent label bias that leads to suboptimal performance. To tackle the above challenges, we propose a label-Free prompt distribution learning and bias correction framework, dubbed as **Frolic**, which boosts zero-shot performance without the need for labeled data. Specifically, our Frolic learns distributions over prompt prototypes to capture diverse visual representations and adaptively fuses these with the original CLIP through confidence matching. This fused model is further enhanced by correcting label bias via a label-free logit adjustment. Notably, our method is not only training-free but also circumvents the necessity for hyper-parameter tuning. Extensive experimental results across 16 datasets demonstrate the efficacy of our approach, particularly outperforming the state-of-the-art by an average of $2.6\%$ on 10 datasets with CLIP ViT-B/16 and achieving an average margin of $1.5\%$ on ImageNet and its five distribution shifts with CLIP ViT-B/16. Codes are available in https://github.com/zhuhsingyuu/Frolic.

Paperid: 1688, https://arxiv.org/pdf/2410.18857.pdf GitHub

Abstract:
Vision-language models (VLMs) embed aligned image-text pairs into a joint space but often rely on deterministic embeddings, assuming a one-to-one correspondence between images and texts. This oversimplifies real-world relationships, which are inherently many-to-many, with multiple captions describing a single image and vice versa. We introduce Probabilistic Language-Image Pre-training (ProLIP), the first probabilistic VLM pre-trained on a billion-scale image-text dataset using only probabilistic objectives, achieving a strong zero-shot capability (e.g., 74.6% ImageNet zero-shot accuracy with ViT-B/16). ProLIP efficiently estimates uncertainty by an "uncertainty token" without extra parameters. We also introduce a novel inclusion loss that enforces distributional inclusion relationships between image-text pairs and between original and masked inputs. Experiments demonstrate that, by leveraging uncertainty estimates, ProLIP benefits downstream tasks and aligns with intuitive notions of uncertainty, e.g., shorter texts being more uncertain and more general inputs including specific ones. Utilizing text uncertainties, we further improve ImageNet accuracy from 74.6% to 75.8% (under a few-shot setting), supporting the practical advantages of our probabilistic approach. The code is available at https://github.com/naver-ai/prolip

Paperid: 1689, https://arxiv.org/pdf/2410.17856.pdf GitHub

Abstract:
Vision-language models (VLMs) have excelled in multimodal tasks, but adapting them to embodied decision-making in open-world environments presents challenges. One critical issue is bridging the gap between discrete entities in low-level observations and the abstract concepts required for effective planning. A common solution is building hierarchical agents, where VLMs serve as high-level reasoners that break down tasks into executable sub-tasks, typically specified using language. However, language suffers from the inability to communicate detailed spatial information. We propose visual-temporal context prompting, a novel communication protocol between VLMs and policy models. This protocol leverages object segmentation from past observations to guide policy-environment interactions. Using this approach, we train ROCKET-1, a low-level policy that predicts actions based on concatenated visual observations and segmentation masks, supported by real-time object tracking from SAM-2. Our method unlocks the potential of VLMs, enabling them to tackle complex tasks that demand spatial reasoning. Experiments in Minecraft show that our approach enables agents to achieve previously unattainable tasks, with a $\mathbf{76}\%$ absolute improvement in open-world interaction performance. Codes and demos are now available on the project page: https://craftjarvis.github.io/ROCKET-1.

Paperid: 1690, https://arxiv.org/pdf/2410.17779.pdf GitHub

Abstract:
Recent advancements in multimodal fusion have witnessed the remarkable success of vision-language (VL) models, which excel in various multimodal applications such as image captioning and visual question answering. However, building VL models requires substantial hardware resources, where efficiency is restricted by two key factors: the extended input sequence of the language model with vision features demands more computational operations, and a large number of additional learnable parameters increase memory complexity. These challenges significantly restrict the broader applicability of such models. To bridge this gap, we propose ADEM-VL, an efficient vision-language method that tunes VL models based on pretrained large language models (LLMs) by adopting a parameter-free cross-attention mechanism for similarity measurements in multimodal fusion. This approach only requires embedding vision features into the language space, significantly reducing the number of trainable parameters and accelerating both training and inference speeds. To enhance representation learning in fusion module, we introduce an efficient multiscale feature generation scheme that requires only a single forward pass through the vision encoder. Moreover, we propose an adaptive fusion scheme that dynamically discards less relevant visual information for each text token based on its attention score. This ensures that the fusion process prioritizes the most pertinent visual features. With experiments on various tasks including visual question answering, image captioning, and instruction-following, we demonstrate that our framework outperforms existing approaches. Specifically, our method surpasses existing methods by an average accuracy of 0.77% on ScienceQA dataset, with reduced training and inference latency, demonstrating the superiority of our framework. The code is available at https://github.com/Hao840/ADEM-VL.

Paperid: 1691, https://arxiv.org/pdf/2410.17637.pdf GitHub

Abstract:
Visual preference alignment involves training Large Vision-Language Models (LVLMs) to predict human preferences between visual inputs. This is typically achieved by using labeled datasets of chosen/rejected pairs and employing optimization algorithms like direct preference optimization (DPO). Existing visual alignment methods, primarily designed for single-image scenarios, struggle to effectively handle the complexity of multi-image tasks due to the scarcity of diverse training data and the high cost of annotating chosen/rejected pairs. We present Multi-Image Augmented Direct Preference Optimization (MIA-DPO), a visual preference alignment approach that effectively handles multi-image inputs. MIA-DPO mitigates the scarcity of diverse multi-image training data by extending single-image data with unrelated images arranged in grid collages or pic-in-pic formats, significantly reducing the costs associated with multi-image data annotations. Our observation reveals that attention values of LVLMs vary considerably across different images. We use attention values to identify and filter out rejected responses the model may have mistakenly focused on. Our attention-aware selection for constructing the chosen/rejected pairs without relying on (i) human annotation, (ii) extra data, and (iii) external models or APIs. MIA-DPO is compatible with various architectures and outperforms existing methods on five multi-image benchmarks, achieving an average performance boost of 3.0% on LLaVA-v1.5 and 4.3% on the recent InternLM-XC2.5. Moreover, MIA-DPO has a minimal effect on the model's ability to understand single images.

Paperid: 1692, https://arxiv.org/pdf/2410.17247.pdf GitHub GitHub

Abstract:
In large vision-language models (LVLMs), images serve as inputs that carry a wealth of information. As the idiom "A picture is worth a thousand words" implies, representing a single image in current LVLMs can require hundreds or even thousands of tokens. This results in significant computational costs, which grow quadratically as input image resolution increases, thereby severely impacting the efficiency of both training and inference. Previous approaches have attempted to reduce the number of image tokens either before or within the early layers of LVLMs. However, these strategies inevitably result in the loss of crucial image information, ultimately diminishing model performance. To address this challenge, we conduct an empirical study revealing that all visual tokens are necessary for LVLMs in the shallow layers, and token redundancy progressively increases in the deeper layers of the model. To this end, we propose PyramidDrop, a visual redundancy reduction strategy for LVLMs to boost their efficiency in both training and inference with neglectable performance loss. Specifically, we partition the LVLM into several stages and drop part of the image tokens at the end of each stage with a pre-defined ratio, creating pyramid-like visual tokens across model layers. The dropping is based on a lightweight similarity calculation with a negligible time overhead. Extensive experiments demonstrate that PyramidDrop can achieve a 40% training time and 55% inference FLOPs acceleration of LLaVA-NeXT with comparable performance. Besides, the PyramidDrop could also serve as a plug-and-play strategy for inference acceleration without training, with better performance and lower inference cost than counterparts. Code is available at https://github.com/Cooperx521/PyramidDrop.

Paperid: 1693, https://arxiv.org/pdf/2410.17241.pdf GitHub

Abstract:
Colonoscopy is currently one of the most sensitive screening methods for colorectal cancer. This study investigates the frontiers of intelligent colonoscopy techniques and their prospective implications for multimodal medical applications. With this goal, we begin by assessing the current data-centric and model-centric landscapes through four tasks for colonoscopic scene perception, including classification, detection, segmentation, and vision-language understanding. This assessment enables us to identify domain-specific challenges and reveals that multimodal research in colonoscopy remains open for further exploration. To embrace the coming multimodal era, we establish three foundational initiatives: a large-scale multimodal instruction tuning dataset ColonINST, a colonoscopy-designed multimodal language model ColonGPT, and a multimodal benchmark. To facilitate ongoing monitoring of this rapidly evolving field, we provide a public website for the latest updates: https://github.com/ai4colonoscopy/IntelliScope.

Paperid: 1694, https://arxiv.org/pdf/2410.16261.pdf GitHub

Abstract:
Multimodal large language models (MLLMs) have demonstrated impressive performance in vision-language tasks across a broad spectrum of domains. However, the large model scale and associated high computational costs pose significant challenges for training and deploying MLLMs on consumer-grade GPUs or edge devices, thereby hindering their widespread application. In this work, we introduce Mini-InternVL, a series of MLLMs with parameters ranging from 1B to 4B, which achieves 90% of the performance with only 5% of the parameters. This significant improvement in efficiency and effectiveness makes our models more accessible and applicable in various real-world scenarios. To further promote the adoption of our models, we develop a unified adaptation framework for Mini-InternVL, which enables our models to transfer and outperform specialized models in downstream tasks, including autonomous driving, medical images, and remote sensing. We believe that our study can provide valuable insights and resources to advance the development of efficient and effective MLLMs. Code is available at https://github.com/OpenGVLab/InternVL.

Paperid: 1695, https://arxiv.org/pdf/2410.16163.pdf GitHub

Abstract:
Large Multimodal Models (LMMs) have achieved significant breakthroughs in various vision-language and vision-centric tasks based on auto-regressive modeling. However, these models typically focus on either vision-centric tasks, such as visual grounding and region description, or vision-language tasks, like image caption and multi-scenario VQAs. None of the LMMs have yet comprehensively unified both types of tasks within a single model, as seen in Large Language Models in the natural language processing field. Furthermore, even with abundant multi-task instruction-following data, directly stacking these data for universal capabilities extension remains challenging. To address these issues, we introduce a novel multi-dimension curated and consolidated multimodal dataset, named CCMD-8M, which overcomes the data barriers of unifying vision-centric and vision-language tasks through multi-level data curation and multi-task consolidation. More importantly, we present Griffon-G, a general large multimodal model that addresses both vision-centric and vision-language tasks within a single end-to-end paradigm. Griffon-G resolves the training collapse issue encountered during the joint optimization of these tasks, achieving better training efficiency. Evaluations across multimodal benchmarks, general Visual Question Answering (VQA) tasks, scene text-centric VQA tasks, document-related VQA tasks, Referring Expression Comprehension, and object detection demonstrate that Griffon-G surpasses the advanced LMMs and achieves expert-level performance in complicated vision-centric tasks.

Paperid: 1696, https://arxiv.org/pdf/2410.15926.pdf GitHub

Abstract:
Recent Large Vision Language Models (LVLMs) present remarkable zero-shot conversational and reasoning capabilities given multimodal queries. Nevertheless, they suffer from object hallucination, a phenomenon where LVLMs are prone to generate textual responses not factually aligned with image inputs. Our pilot study reveals that object hallucination is closely tied with Rotary Position Encoding (RoPE), a widely adopted positional dependency modeling design in existing LVLMs. Due to the long-term decay in RoPE, LVLMs tend to hallucinate more when relevant visual cues are distant from instruction tokens in the multimodal input sequence. Additionally, we observe a similar effect when reversing the sequential order of visual tokens during multimodal alignment. Our tests indicate that long-term decay in RoPE poses challenges to LVLMs while capturing visual-instruction interactions across long distances. We propose Concentric Causal Attention (CCA), a simple yet effective positional alignment strategy that mitigates the impact of RoPE long-term decay in LVLMs by naturally reducing relative distance between visual and instruction tokens. With CCA, visual tokens can better interact with instruction tokens, thereby enhancing model's perception capability and alleviating object hallucination. Without bells and whistles, our positional alignment method surpasses existing hallucination mitigation strategies by large margins on multiple object hallucination benchmarks.

Paperid: 1697, https://arxiv.org/pdf/2410.15881.pdf GitHub

Abstract:
Vision-language supervision has made remarkable strides in learning visual representations from textual guidance. In digital pathology, vision-language models (VLM), pre-trained on curated datasets of histological image-captions, have been adapted to downstream tasks, such as region of interest classification. Zero-shot transfer for slide-level prediction has been formulated by MI-Zero, but it exhibits high variability depending on the textual prompts. Inspired by prototypical learning, we propose MI-VisionShot, a training-free adaptation method on top of VLMs to predict slide-level labels in few-shot learning scenarios. Our framework takes advantage of the excellent representation learning of VLM to create prototype-based classifiers under a multiple-instance setting by retrieving the most discriminative patches within each slide. Experimentation through different settings shows the ability of MI-VisionShot to surpass zero-shot transfer with lower variability, even in low-shot scenarios. Code coming soon at thttps://github.com/cvblab/MIVisionShot.

Paperid: 1698, https://arxiv.org/pdf/2410.15780.pdf GitHub

Abstract:
Historical maps provide valuable information and knowledge about the past. However, as they often feature non-standard projections, hand-drawn styles, and artistic elements, it is challenging for non-experts to identify and interpret them. While existing image captioning methods have achieved remarkable success on natural images, their performance on maps is suboptimal as maps are underrepresented in their pre-training process. Despite the recent advance of GPT-4 in text recognition and map captioning, it still has a limited understanding of maps, as its performance wanes when texts (e.g., titles and legends) in maps are missing or inaccurate. Besides, it is inefficient or even impractical to fine-tune the model with users' own datasets. To address these problems, we propose a novel and lightweight map-captioning counterpart. Specifically, we fine-tune the state-of-the-art vision-language model CLIP to generate captions relevant to historical maps and enrich the captions with GPT-3.5 to tell a brief story regarding where, what, when and why of a given map. We propose a novel decision tree architecture to only generate captions relevant to the specified map type. Our system shows invariance to text alterations in maps. The system can be easily adapted and extended to other map types and scaled to a larger map captioning system. The code is open-sourced at https://github.com/claudaff/automatic-map-storytelling.

Paperid: 1699, https://arxiv.org/pdf/2410.15364.pdf GitHub

Abstract:
Current approaches for open-vocabulary scene graph generation (OVSGG) use vision-language models such as CLIP and follow a standard zero-shot pipeline -- computing similarity between the query image and the text embeddings for each category (i.e., text classifiers). In this work, we argue that the text classifiers adopted by existing OVSGG methods, i.e., category-/part-level prompts, are scene-agnostic as they remain unchanged across contexts. Using such fixed text classifiers not only struggles to model visual relations with high variance, but also falls short in adapting to distinct contexts. To plug these intrinsic shortcomings, we devise SDSGG, a scene-specific description based OVSGG framework where the weights of text classifiers are adaptively adjusted according to the visual content. In particular, to generate comprehensive and diverse descriptions oriented to the scene, an LLM is asked to play different roles (e.g., biologist and engineer) to analyze and discuss the descriptive features of a given scene from different views. Unlike previous efforts simply treating the generated descriptions as mutually equivalent text classifiers, SDSGG is equipped with an advanced renormalization mechanism to adjust the influence of each text classifier based on its relevance to the presented scene (this is what the term "specific" means). Furthermore, to capture the complicated interplay between subjects and objects, we propose a new lightweight module called mutual visual adapter. It refines CLIP's ability to recognize relations by learning an interaction-aware semantic space. Extensive experiments on prevalent benchmarks show that SDSGG outperforms top-leading methods by a clear margin.

Paperid: 1700, https://arxiv.org/pdf/2410.15270.pdf GitHub

Abstract:
Despite rapid progress in large vision-language models (LVLMs), existing video caption benchmarks remain limited in evaluating their alignment with human understanding. Most rely on a single annotation per video and lexical similarity-based metrics, failing to capture the variability in human perception and the cognitive importance of events. These limitations hinder accurate diagnosis of model capabilities in producing coherent, complete, and human-aligned descriptions. To address this, we introduce FIOVA (Five-In-One Video Annotations), a human-centric benchmark tailored for evaluation. It comprises 3,002 real-world videos (about 33.6s each), each annotated independently by five annotators. This design enables modeling of semantic diversity and inter-subjective agreement, offering a richer foundation for measuring human-machine alignment. We further propose FIOVA-DQ, an event-level evaluation metric that incorporates cognitive weights derived from annotator consensus, providing fine-grained assessment of event relevance and semantic coverage. Leveraging FIOVA, we conduct a comprehensive evaluation of nine representative LVLMs and introduce a complexity-aware analysis framework based on inter-annotator variation (CV). This reveals consistency gaps across difficulty levels and identifies structural issues such as event under-description and template convergence. Our results highlight FIOVA's diagnostic value for understanding LVLM behavior under varying complexity, setting a new standard for cognitively aligned evaluation in long-video captioning. The benchmark, annotations, metric, and model outputs are publicly released to support future evaluation-driven research in video understanding. More detailed information can be found at https://huuuuusy.github.io/fiova/.

Paperid: 1701, https://arxiv.org/pdf/2410.14975.pdf GitHub

Abstract:
With the recent emergence of foundation models trained on internet-scale data and demonstrating remarkable generalization capabilities, such foundation models have become more widely adopted, leading to an expanding range of application domains. Despite this rapid proliferation, the trustworthiness of foundation models remains underexplored. Specifically, the out-of-distribution detection (OoDD) capabilities of large vision-language models (LVLMs), such as GPT-4o, which are trained on massive multi-modal data, have not been sufficiently addressed. The disparity between their demonstrated potential and practical reliability raises concerns regarding the safe and trustworthy deployment of foundation models. To address this gap, we evaluate and analyze the OoDD capabilities of various proprietary and open-source LVLMs. Our investigation contributes to a better understanding of how these foundation models represent confidence scores through their generated natural language responses. Furthermore, we propose a self-guided prompting approach, termed Reflexive Guidance (ReGuide), aimed at enhancing the OoDD capability of LVLMs by leveraging self-generated image-adaptive concept suggestions. Experimental results demonstrate that our ReGuide enhances the performance of current LVLMs in both image classification and OoDD tasks. The lists of sampled images, along with the prompts and responses for each sample are available at https://github.com/daintlab/ReGuide.

Paperid: 1702, https://arxiv.org/pdf/2410.14669.pdf GitHub

Abstract:
Vision-language models (VLMs) have made significant progress in recent visual-question-answering (VQA) benchmarks that evaluate complex visio-linguistic reasoning. However, are these models truly effective? In this work, we show that VLMs still struggle with natural images and questions that humans can easily answer, which we term natural adversarial samples. We also find it surprisingly easy to generate these VQA samples from natural image-text corpora using off-the-shelf models like CLIP and ChatGPT. We propose a semi-automated approach to collect a new benchmark, NaturalBench, for reliably evaluating VLMs with 10,000 human-verified VQA samples. Crucially, we adopt a $\textbf{vision-centric}$ design by pairing each question with two images that yield different answers, preventing blind solutions from answering without using the images. This makes NaturalBench more challenging than previous benchmarks that can be solved with commonsense priors. We evaluate 53 state-of-the-art VLMs on NaturalBench, showing that models like LLaVA-OneVision, Cambrian-1, Llama3.2-Vision, Molmo, Qwen2-VL, and even GPT-4o lag 50%-70% behind human performance (over 90%). We analyze why NaturalBench is hard from two angles: (1) Compositionality: Solving NaturalBench requires diverse visio-linguistic skills, including understanding attribute bindings, object relationships, and advanced reasoning like logic and counting. To this end, unlike prior work that uses a single tag per sample, we tag each NaturalBench sample with 1 to 8 skill tags for fine-grained evaluation. (2) Biases: NaturalBench exposes severe biases in VLMs, as models often choose the same answer regardless of the image. Lastly, we apply our benchmark curation method to diverse data sources, including long captions (over 100 words) and non-English languages like Chinese and Hindi, highlighting its potential for dynamic evaluations of VLMs.

Paperid: 1703, https://arxiv.org/pdf/2410.14340.pdf GitHub GitHub

Abstract:
Precise action localization in untrimmed video is vital for fields such as professional sports and minimally invasive surgery, where the delineation of particular motions in recordings can dramatically enhance analysis. But in many cases, large scale datasets with video-label pairs for localization are unavailable, limiting the opportunity to fine-tune video-understanding models. Recent developments in large vision-language models (LVLM) address this need with impressive zero-shot capabilities in a variety of video understanding tasks. However, the adaptation of LVLMs, with their powerful visual question answering capabilities, to zero-shot localization in long-form video is still relatively unexplored. To this end, we introduce a true Zero-shot Action Localization method (ZEAL). Specifically, we leverage the built-in action knowledge of a large language model (LLM) to inflate actions into detailed descriptions of the archetypal start and end of the action. These descriptions serve as queries to LVLM for generating frame-level confidence scores which can be aggregated to produce localization outputs. The simplicity and flexibility of our method lends it amenable to more capable LVLMs as they are developed, and we demonstrate remarkable results in zero-shot action localization on a challenging benchmark, without any training. Our code is publicly available at $\href{https://github.com/josaklil-ai/zeal}{github.com/josaklil-ai/zeal}$.

Paperid: 1704, https://arxiv.org/pdf/2410.14179.pdf GitHub

Abstract:
Multimodal Large Language Models (MLLMs) have demonstrated impressive abilities across various tasks, including visual question answering and chart comprehension, yet existing benchmarks for chart-related tasks fall short in capturing the complexity of real-world multi-chart scenarios. Current benchmarks primarily focus on single-chart tasks, neglecting the multi-hop reasoning required to extract and integrate information from multiple charts, which is essential in practical applications. To fill this gap, we introduce MultiChartQA, a benchmark that evaluates MLLMs' capabilities in four key areas: direct question answering, parallel question answering, comparative reasoning, and sequential reasoning. Our evaluation of a wide range of MLLMs reveals significant performance gaps compared to humans. These results highlight the challenges in multi-chart comprehension and the potential of MultiChartQA to drive advancements in this field. Our code and data are available at https://github.com/Zivenzhu/Multi-chart-QA

Paperid: 1705, https://arxiv.org/pdf/2410.13861.pdf GitHub GitHub

Abstract:
Recent advancements in multimodal foundation models have yielded significant progress in vision-language understanding. Initial attempts have also explored the potential of multimodal large language models (MLLMs) for visual content generation. However, existing works have insufficiently addressed the varying granularity demands of different image generation tasks within a unified MLLM paradigm - from the diversity required in text-to-image generation to the precise controllability needed in image manipulation. In this work, we propose PUMA, emPowering Unified MLLM with Multi-grAnular visual generation. PUMA unifies multi-granular visual features as both inputs and outputs of MLLMs, elegantly addressing the different granularity requirements of various image generation tasks within a unified MLLM framework. Following multimodal pretraining and task-specific instruction tuning, PUMA demonstrates proficiency in a wide range of multimodal tasks. This work represents a significant step towards a truly unified MLLM capable of adapting to the granularity demands of various visual tasks. The code and model will be released in https://github.com/rongyaofang/PUMA.

Paperid: 1706, https://arxiv.org/pdf/2410.13860.pdf GitHub

Abstract:
3D visual grounding is crucial for robots, requiring integration of natural language and 3D scene understanding. Traditional methods depending on supervised learning with 3D point clouds are limited by scarce datasets. Recently zero-shot methods leveraging LLMs have been proposed to address the data issue. While effective, these methods only use object-centric information, limiting their ability to handle complex queries. In this work, we present VLM-Grounder, a novel framework using vision-language models (VLMs) for zero-shot 3D visual grounding based solely on 2D images. VLM-Grounder dynamically stitches image sequences, employs a grounding and feedback scheme to find the target object, and uses a multi-view ensemble projection to accurately estimate 3D bounding boxes. Experiments on ScanRefer and Nr3D datasets show VLM-Grounder outperforms previous zero-shot methods, achieving 51.6% Acc@0.25 on ScanRefer and 48.0% Acc on Nr3D, without relying on 3D geometry or object priors. Codes are available at https://github.com/OpenRobotLab/VLM-Grounder .

Paperid: 1707, https://arxiv.org/pdf/2410.13823.pdf GitHub

Abstract:
Deep generative models have significantly advanced medical imaging analysis by enhancing dataset size and quality. Beyond mere data augmentation, our research in this paper highlights an additional, significant capacity of deep generative models: their ability to reveal and demonstrate patterns in medical images. We employ a generative structure with hybrid conditions, combining clinical data and segmentation masks to guide the image synthesis process. Furthermore, we innovatively transformed the tabular clinical data into textual descriptions. This approach simplifies the handling of missing values and also enables us to leverage large pre-trained vision-language models that investigate the relations between independent clinical entries and comprehend general terms, such as gender and smoking status. Our approach differs from and presents a more challenging task than traditional medical report-guided synthesis due to the less visual correlation of our clinical information with the images. To overcome this, we introduce a text-visual embedding mechanism that strengthens the conditions, ensuring the network effectively utilizes the provided information. Our pipeline is generalizable to both GAN-based and diffusion models. Experiments on chest CT, particularly focusing on the smoking status, demonstrated a consistent intensity shift in the lungs which is in agreement with clinical observations, indicating the effectiveness of our method in capturing and visualizing the impact of specific attributes on medical image patterns. Our methods offer a new avenue for the early detection and precise visualization of complex clinical conditions with deep generative models. All codes are https://github.com/junzhin/DGM-VLC.

Paperid: 1708, https://arxiv.org/pdf/2410.13085.pdf GitHub

Abstract:
Artificial Intelligence (AI) has demonstrated significant potential in healthcare, particularly in disease diagnosis and treatment planning. Recent progress in Medical Large Vision-Language Models (Med-LVLMs) has opened up new possibilities for interactive diagnostic tools. However, these models often suffer from factual hallucination, which can lead to incorrect diagnoses. Fine-tuning and retrieval-augmented generation (RAG) have emerged as methods to address these issues. However, the amount of high-quality data and distribution shifts between training data and deployment data limit the application of fine-tuning methods. Although RAG is lightweight and effective, existing RAG-based approaches are not sufficiently general to different medical domains and can potentially cause misalignment issues, both between modalities and between the model and the ground truth. In this paper, we propose a versatile multimodal RAG system, MMed-RAG, designed to enhance the factuality of Med-LVLMs. Our approach introduces a domain-aware retrieval mechanism, an adaptive retrieved contexts selection method, and a provable RAG-based preference fine-tuning strategy. These innovations make the RAG process sufficiently general and reliable, significantly improving alignment when introducing retrieved contexts. Experimental results across five medical datasets (involving radiology, ophthalmology, pathology) on medical VQA and report generation demonstrate that MMed-RAG can achieve an average improvement of 43.8% in the factual accuracy of Med-LVLMs. Our data and code are available in https://github.com/richard-peng-xia/MMed-RAG.

Paperid: 1709, https://arxiv.org/pdf/2410.12790.pdf GitHub GitHub

Abstract:
Test-time adaptation, which enables models to generalize to diverse data with unlabeled test samples, holds significant value in real-world scenarios. Recently, researchers have applied this setting to advanced pre-trained vision-language models (VLMs), developing approaches such as test-time prompt tuning to further extend their practical applicability. However, these methods typically focus solely on adapting VLMs from a single modality and fail to accumulate task-specific knowledge as more samples are processed. To address this, we introduce Dual Prototype Evolving (DPE), a novel test-time adaptation approach for VLMs that effectively accumulates task-specific knowledge from multi-modalities. Specifically, we create and evolve two sets of prototypes--textual and visual--to progressively capture more accurate multi-modal representations for target classes during test time. Moreover, to promote consistent multi-modal representations, we introduce and optimize learnable residuals for each test sample to align the prototypes from both modalities. Extensive experimental results on 15 benchmark datasets demonstrate that our proposed DPE consistently outperforms previous state-of-the-art methods while also exhibiting competitive computational efficiency. Code is available at https://github.com/zhangce01/DPE-CLIP.

Paperid: 1710, https://arxiv.org/pdf/2410.12773.pdf GitHub GitHub

Abstract:
Humanoid robots, with their human-like embodiment, have the potential to integrate seamlessly into human environments. Critical to their coexistence and cooperation with humans is the ability to understand natural language communications and exhibit human-like behaviors. This work focuses on generating diverse whole-body motions for humanoid robots from language descriptions. We leverage human motion priors from extensive human motion datasets to initialize humanoid motions and employ the commonsense reasoning capabilities of Vision Language Models (VLMs) to edit and refine these motions. Our approach demonstrates the capability to produce natural, expressive, and text-aligned humanoid motions, validated through both simulated and real-world experiments. More videos can be found at https://ut-austin-rpl.github.io/Harmon/.

Paperid: 1711, https://arxiv.org/pdf/2410.12694.pdf GitHub

Abstract:
Recent advancements in Vision Language Models (VLMs) have demonstrated remarkable promise in generating visually grounded responses. However, their application in the medical domain is hindered by unique challenges. For instance, most VLMs rely on a single method of visual grounding, whereas complex medical tasks demand more versatile approaches. Additionally, while most VLMs process only 2D images, a large portion of medical images are 3D. The lack of medical data further compounds these obstacles. To address these challenges, we present VividMed, a vision language model with versatile visual grounding for medicine. Our model supports generating both semantic segmentation masks and instance-level bounding boxes, and accommodates various imaging modalities, including both 2D and 3D data. We design a three-stage training procedure and an automatic data synthesis pipeline based on open datasets and models. Besides visual grounding tasks, VividMed also excels in other common downstream tasks, including Visual Question Answering (VQA) and report generation. Ablation studies empirically show that the integration of visual grounding ability leads to improved performance on these tasks. Our code is publicly available at https://github.com/function2-llx/MMMM.

Paperid: 1712, https://arxiv.org/pdf/2410.12332.pdf GitHub

Abstract:
While multimodal large language models (MLLMs) have demonstrated extraordinary vision-language understanding capabilities, their abilities to solve instance-level visual-language problems beyond a single image warrant further exploration. To assess these unproven abilities of MLLMs, this paper proposes a new visual grounding task called multi-context visual grounding, which aims to localize instances of interest across multiple images based on open-ended text prompts. In order to facilitate this research, we construct a new dataset MC-Bench that features 2K high-quality and manually annotated samples. Each sample consists of an instance-level labeled image pair and a corresponding text prompt that indicates the target instances in the images. These text prompts are highly open-ended and follow three distinct styles, covering 20 practical skills. We benchmark over 20 state-of-the-art MLLMs and foundation models with potential multi-context visual grounding capabilities, along with our developed simple yet effective agentic baseline and a finetuned baseline by multi-context instruction tuning. Our evaluation reveals a non-trivial performance gap between existing MLLMs and humans, along with some insightful observations that suggest potential future directions. We hope that MC-Bench and our empirical findings encourage the research community to further advance the untapped potentials of MLLMs in instance-level tasks, particularly in multi-image contexts. Project page: https://xuyunqiu.github.io/MC-Bench.

Paperid: 1713, https://arxiv.org/pdf/2410.11989.pdf GitHub

Abstract:
Enabling mobile robots to perform long-term tasks in dynamic real-world environments is a formidable challenge, especially when the environment changes frequently due to human-robot interactions or the robot's own actions. Traditional methods typically assume static scenes, which limits their applicability in the continuously changing real world. To overcome these limitations, we present DovSG, a novel mobile manipulation framework that leverages dynamic open-vocabulary 3D scene graphs and a language-guided task planning module for long-term task execution. DovSG takes RGB-D sequences as input and utilizes vision-language models (VLMs) for object detection to obtain high-level object semantic features. Based on the segmented objects, a structured 3D scene graph is generated for low-level spatial relationships. Furthermore, an efficient mechanism for locally updating the scene graph, allows the robot to adjust parts of the graph dynamically during interactions without the need for full scene reconstruction. This mechanism is particularly valuable in dynamic environments, enabling the robot to continually adapt to scene changes and effectively support the execution of long-term tasks. We validated our system in real-world environments with varying degrees of manual modifications, demonstrating its effectiveness and superior performance in long-term tasks. Our project page is available at: https://bjhyzj.github.io/dovsg-web.

Paperid: 1714, https://arxiv.org/pdf/2410.11829.pdf GitHub

Abstract:
Despite significant advancements in Multimodal Large Language Models (MLLMs) for understanding complex human intentions through cross-modal interactions, capturing intricate image details remains challenging. Previous methods integrating multiple vision encoders to enhance visual detail introduce redundancy and computational overhead. We observe that most MLLMs utilize only the last-layer feature map of the vision encoder for visual representation, neglecting the rich fine-grained information in shallow feature maps. To address this issue, we propose \modelname, a simple yet effective multi-layer feature fuser that efficiently integrates deep and shallow features from Vision Transformers (ViTs). Specifically, it leverages semantically aligned deep features as queries to dynamically extract missing details from shallow features, thus preserving semantic alignment while enriching the representation with fine-grained information. Applied to the LLaVA-1.5 model, \modelname~achieves significant improvements in visual representation and benchmark performance, providing a more flexible and lightweight solution compared to multi-encoder ensemble methods. The code and model have been released at https://github.com/yuecao0119/MMFuser.

Paperid: 1715, https://arxiv.org/pdf/2410.11761.pdf GitHub

Abstract:
Despite the progress made by multimodal large language models (MLLMs) in computational pathology, they remain limited by a predominant focus on patch-level analysis, missing essential contextual information at the whole-slide level. The lack of large-scale instruction datasets and the gigapixel scale of whole slide images (WSIs) pose significant developmental challenges. In this paper, we present SlideChat, the first vision-language assistant capable of understanding gigapixel whole-slide images, exhibiting excellent multimodal conversational capability and response complex instruction across diverse pathology scenarios. To support its development, we created SlideInstruction, the largest instruction-following dataset for WSIs consisting of 4.2K WSI captions and 176K VQA pairs with multiple categories. Furthermore, we propose SlideBench, a multimodal benchmark that incorporates captioning and VQA tasks to assess SlideChat's capabilities in varied clinical settings such as microscopy, diagnosis. Compared to both general and specialized MLLMs, SlideChat exhibits exceptional capabilities achieving state-of-the-art performance on 18 of 22 tasks. For example, it achieved an overall accuracy of 81.17% on SlideBench-VQA (TCGA), and 54.15% on SlideBench-VQA (BCNB). Our code, data, and model is publicly accessible at https://uni-medical.github.io/SlideChat.github.io.

Paperid: 1716, https://arxiv.org/pdf/2410.11374.pdf GitHub

Abstract:
The development of vision-language and generative models has significantly advanced text-guided image editing, which seeks the preservation of core elements in the source image while implementing modifications based on the target text. However, existing metrics have a context-blindness problem, indiscriminately applying the same evaluation criteria on completely different pairs of source image and target text, biasing towards either modification or preservation. Directional CLIP similarity, the only metric that considers both source image and target text, is also biased towards modification aspects and attends to irrelevant editing regions of the image. We propose AugCLIP, a context-aware metric that adaptively coordinates preservation and modification aspects, depending on the specific context of a given source image and target text. This is done by deriving the CLIP representation of an ideally edited image, that preserves the source image with necessary modifications to align with target text. More specifically, using a multi-modal large language model, AugCLIP augments the textual descriptions of the source and target, then calculates a modification vector through a hyperplane that separates source and target attributes in CLIP space. Extensive experiments on five benchmark datasets, encompassing a diverse range of editing scenarios, show that AugCLIP aligns remarkably well with human evaluation standards, outperforming existing metrics. The code is available at https://github.com/augclip/augclip_eval.

Paperid: 1717, https://arxiv.org/pdf/2410.11201.pdf GitHub

Abstract:
Prompt learning has proven effective in adapting vision language models for downstream tasks. However, existing methods usually append learnable prompt tokens solely with the category names to obtain textual features, which fails to fully leverage the rich context indicated in the category name. To address this issue, we propose the Tree of Attributes Prompt learning (TAP), which first instructs LLMs to generate a tree of attributes with a "concept - attribute - description" structure for each category, and then learn the hierarchy with vision and text prompt tokens. Unlike existing methods that merely augment category names with a set of unstructured descriptions, our approach essentially distills structured knowledge graphs associated with class names from LLMs. Furthermore, our approach introduces text and vision prompts designed to explicitly learn the corresponding visual attributes, effectively serving as domain experts. Additionally, the general and diverse descriptions generated based on the class names may be wrong or absent in the specific given images. To address this misalignment, we further introduce a vision-conditional pooling module to extract instance-specific text features. Extensive experimental results demonstrate that our approach outperforms state-of-the-art methods on the zero-shot base-to-novel generalization, cross-dataset transfer, as well as few-shot classification across 11 diverse datasets. Code is available at https://github.com/HHenryD/TAP.

Paperid: 1718, https://arxiv.org/pdf/2410.10799.pdf GitHub

Abstract:
Building a foundation model for 3D vision is a complex challenge that remains unsolved. Towards that goal, it is important to understand the 3D reasoning capabilities of current models as well as identify the gaps between these models and humans. Therefore, we construct a new 3D visual understanding benchmark named UniQA-3D. UniQA-3D covers fundamental 3D vision tasks in the Visual Question Answering (VQA) format. We evaluate state-of-the-art Vision-Language Models (VLMs), specialized models, and human subjects on it. Our results show that VLMs generally perform poorly, while the specialized models are accurate but not robust, failing under geometric perturbations. In contrast, human vision continues to be the most reliable 3D visual system. We further demonstrate that neural networks align more closely with human 3D vision mechanisms compared to classical computer vision methods, and Transformer-based networks such as ViT align more closely with human 3D vision mechanisms than CNNs. We hope our study will benefit the future development of foundation models for 3D vision. Code is available at https://github.com/princeton-vl/UniQA-3D .

Paperid: 1719, https://arxiv.org/pdf/2410.10594.pdf GitHub

Abstract:
Retrieval-augmented generation (RAG) is an effective technique that enables large language models (LLMs) to utilize external knowledge sources for generation. However, current RAG systems are solely based on text, rendering it impossible to utilize vision information like layout and images that play crucial roles in real-world multi-modality documents. In this paper, we introduce VisRAG, which tackles this issue by establishing a vision-language model (VLM)-based RAG pipeline. In this pipeline, instead of first parsing the document to obtain text, the document is directly embedded using a VLM as an image and then retrieved to enhance the generation of a VLM. Compared to traditional text-based RAG, VisRAG maximizes the retention and utilization of the data information in the original documents, eliminating the information loss introduced during the parsing process. We collect both open-source and synthetic data to train the retriever in VisRAG and explore a variety of generation methods. Experiments demonstrate that VisRAG outperforms traditional RAG in both the retrieval and generation stages, achieving a 20--40% end-to-end performance gain over traditional text-based RAG pipeline. Further analysis reveals that VisRAG is efficient in utilizing training data and demonstrates strong generalization capability, positioning it as a promising solution for RAG on multi-modality documents. Our code and data are available at https://github.com/openbmb/visrag.

Paperid: 1720, https://arxiv.org/pdf/2410.10573.pdf GitHub

Abstract:
Whole Slide Image (WSI) classification has very significant applications in clinical pathology, e.g., tumor identification and cancer diagnosis. Currently, most research attention is focused on Multiple Instance Learning (MIL) using static datasets. One of the most obvious weaknesses of these methods is that they cannot efficiently preserve and utilize previously learned knowledge. With any new data arriving, classification models are required to be re-trained on both previous and current new data. To overcome this shortcoming and break through traditional vision modality, this paper proposes the first Vision-Language-based framework with Queryable Prototype Multiple Instance Learning (QPMIL-VL) specially designed for incremental WSI classification. This framework mainly consists of two information processing branches: one is for generating bag-level features by prototype-guided aggregation of instance features, while the other is for enhancing class features through a combination of class ensemble, tunable vector and class similarity loss. The experiments on four public WSI datasets demonstrate that our QPMIL-VL framework is effective for incremental WSI classification and often significantly outperforms other compared methods, achieving state-of-the-art (SOTA) performance. Our source code is publicly available at https://github.com/can-can-ya/QPMIL-VL.

Paperid: 1721, https://arxiv.org/pdf/2410.10563.pdf GitHub

Abstract:
We present MEGA-Bench, an evaluation suite that scales multimodal evaluation to over 500 real-world tasks, to address the highly heterogeneous daily use cases of end users. Our objective is to optimize for a set of high-quality data samples that cover a highly diverse and rich set of multimodal tasks, while enabling cost-effective and accurate model evaluation. In particular, we collected 505 realistic tasks encompassing over 8,000 samples from 16 expert annotators to extensively cover the multimodal task space. Instead of unifying these problems into standard multi-choice questions (like MMMU, MMBench, and MMT-Bench), we embrace a wide range of output formats like numbers, phrases, code, \LaTeX, coordinates, JSON, free-form, etc. To accommodate these formats, we developed over 40 metrics to evaluate these tasks. Unlike existing benchmarks, MEGA-Bench offers a fine-grained capability report across multiple dimensions (e.g., application, input type, output format, skill), allowing users to interact with and visualize model capabilities in depth. We evaluate a wide variety of frontier vision-language models on MEGA-Bench to understand their capabilities across these dimensions.

Paperid: 1722, https://arxiv.org/pdf/2410.10441.pdf GitHub

Abstract:
Vision-language large models have achieved remarkable success in various multi-modal tasks, yet applying them to video understanding remains challenging due to the inherent complexity and computational demands of video data. While training-based video-LLMs deliver high performance, they often require substantial resources for training and inference. Conversely, training-free approaches offer a more efficient alternative by adapting pre-trained image-LLMs models for video tasks without additional training, but they face inference efficiency bottlenecks due to the large number of visual tokens generated from video frames. In this work, we present a novel prompt-guided visual perception framework (abbreviated as Free Video-LLM) for efficient inference of training-free video LLMs. The proposed framework decouples spatial-temporal dimension and performs temporal frame sampling and spatial RoI cropping respectively based on task-specific prompts. Our method effectively reduces the number of visual tokens while maintaining high performance across multiple video question-answering benchmarks. Extensive experiments demonstrate that our approach achieves competitive results with significantly fewer tokens, offering an optimal trade-off between accuracy and computational efficiency compared to state-of-the-art video LLMs. The code will be available at https://github.com/contrastive/FreeVideoLLM.

Paperid: 1723, https://arxiv.org/pdf/2410.10308.pdf GitHub

Abstract:
Concept activation vector (CAV) has attracted broad research interest in explainable AI, by elegantly attributing model predictions to specific concepts. However, the training of CAV often necessitates a large number of high-quality images, which are expensive to curate and thus limited to a predefined set of concepts. To address this issue, we propose Language-Guided CAV (LG-CAV) to harness the abundant concept knowledge within the certain pre-trained vision-language models (e.g., CLIP). This method allows training any CAV without labeled data, by utilizing the corresponding concept descriptions as guidance. To bridge the gap between vision-language model and the target model, we calculate the activation values of concept descriptions on a common pool of images (probe images) with vision-language model and utilize them as language guidance to train the LG-CAV. Furthermore, after training high-quality LG-CAVs related to all the predicted classes in the target model, we propose the activation sample reweighting (ASR), serving as a model correction technique, to improve the performance of the target model in return. Experiments on four datasets across nine architectures demonstrate that LG-CAV achieves significantly superior quality to previous CAV methods given any concept, and our model correction method achieves state-of-the-art performance compared to existing concept-based methods. Our code is available at https://github.com/hqhQAQ/LG-CAV.

Paperid: 1724, https://arxiv.org/pdf/2410.10289.pdf GitHub

Abstract:
Current zero-shot anomaly detection (ZSAD) methods show remarkable success in prompting large pre-trained vision-language models to detect anomalies in a target dataset without using any dataset-specific training or demonstration. However, these methods are often focused on crafting/learning prompts that capture only coarse-grained semantics of abnormality, e.g., high-level semantics like "damaged", "imperfect", or "defective" on carpet. They therefore have limited capability in recognizing diverse abnormality details with distinctive visual appearance, e.g., specific defect types like color stains, cuts, holes, and threads on carpet. To address this limitation, we propose FAPrompt, a novel framework designed to learn Fine-grained Abnormality Prompts for more accurate ZSAD. To this end, we introduce a novel compound abnormality prompting module in FAPrompt to learn a set of complementary, decomposed abnormality prompts, where each abnormality prompt is formed by a compound of shared normal tokens and a few learnable abnormal tokens. On the other hand, the fine-grained abnormality patterns can be very different from one dataset to another. To enhance their cross-dataset generalization, we further introduce a data-dependent abnormality prior module that learns to derive abnormality features from each query/test image as a sample-wise abnormality prior to ground the abnormality prompts in a given target dataset. Comprehensive experiments conducted across 19 real-world datasets, covering both industrial defects and medical anomalies, demonstrate that FAPrompt substantially outperforms state-of-the-art methods by at least 3%-5% AUC/AP in both image- and pixel-level ZSAD tasks. Code is available at https://github.com/mala-lab/FAPrompt.

Paperid: 1725, https://arxiv.org/pdf/2410.10139.pdf GitHub

Abstract:
Interleaved multimodal comprehension and generation, enabling models to produce and interpret both images and text in arbitrary sequences, have become a pivotal area in multimodal learning. Despite significant advancements, the evaluation of this capability remains insufficient. Existing benchmarks suffer from limitations in data scale, scope, and evaluation depth, while current evaluation metrics are often costly or biased, lacking in reliability for practical applications. To address these challenges, we introduce MMIE, a large-scale knowledge-intensive benchmark for evaluating interleaved multimodal comprehension and generation in Large Vision-Language Models (LVLMs). MMIE comprises 20K meticulously curated multimodal queries, spanning 3 categories, 12 fields, and 102 subfields, including mathematics, coding, physics, literature, health, and arts. It supports both interleaved inputs and outputs, offering a mix of multiple-choice and open-ended question formats to evaluate diverse competencies. Moreover, we propose a reliable automated evaluation metric, leveraging a scoring model fine-tuned with human-annotated data and systematic evaluation criteria, aimed at reducing bias and improving evaluation accuracy. Extensive experiments demonstrate the effectiveness of our benchmark and metrics in providing a comprehensive evaluation of interleaved LVLMs. Specifically, we evaluate eight LVLMs, revealing that even the best models show significant room for improvement, with most achieving only moderate results. We believe MMIE will drive further advancements in the development of interleaved LVLMs. We publicly release our benchmark and code in https://mmie-bench.github.io/.

Paperid: 1726, https://arxiv.org/pdf/2410.10114.pdf GitHub

Abstract:
Federated prompt learning benefits federated learning with CLIP-like Vision-Language Model's (VLM's) robust representation learning ability through prompt learning. However, current federated prompt learning methods are habitually restricted to the traditional FL paradigm, where the participating clients are generally only allowed to download a single globally aggregated model from the server. While justifiable for training full-sized models under federated settings, in this work, we argue that this paradigm is ill-suited for lightweight prompts. By facilitating the clients to download multiple pre-aggregated prompts as fixed non-local experts, we propose Personalized Federated Mixture of Adaptive Prompts (pFedMoAP), a novel FL framework that personalizes the prompt learning process through the lens of Mixture of Experts (MoE). pFedMoAP implements a local attention-based gating network that learns to generate enhanced text features for better alignment with local image data, benefiting from both local and downloaded non-local adaptive prompt experts. Extensive experiments on 9 datasets under various federated settings demonstrate the efficacy of the proposed pFedMoAP algorithm. The code is available at https://github.com/ljaiverson/pFedMoAP.

Paperid: 1727, https://arxiv.org/pdf/2410.10112.pdf GitHub

Abstract:
Evaluating large vision-language models (LVLMs) is very expensive, due to high computational cost and the wide variety of tasks. The good news is that if we already have some observed performance scores, we may be able to infer unknown ones. In this study, we propose a new framework for predicting unknown performance scores based on observed ones from other LVLMs or tasks. We first formulate the performance prediction as a matrix completion task. Specifically, we construct a sparse performance matrix $\boldsymbol{R}$, where each entry $R_{mn}$ represents the performance score of the $m$-th model on the $n$-th dataset. By applying probabilistic matrix factorization (PMF) with Markov chain Monte Carlo (MCMC), we can complete the performance matrix, i.e., predict unknown scores. Additionally, we estimate the uncertainty of performance prediction based on MCMC. Practitioners can evaluate their models on untested tasks with higher uncertainty first, which quickly reduces the prediction errors. We further introduce several improvements to enhance PMF for scenarios with sparse observed performance scores. Our experiments demonstrate the accuracy of PMF in predicting unknown scores, the reliability of uncertainty estimates in ordering evaluations, and the effectiveness of our enhancements for handling sparse data.

Paperid: 1728, https://arxiv.org/pdf/2410.10034.pdf GitHub

Abstract:
We address the challenge of representing long captions in vision-language models, such as CLIP. By design these models are limited by fixed, absolute positional encodings, restricting inputs to a maximum of 77 tokens and hindering performance on tasks requiring longer descriptions. Although recent work has attempted to overcome this limit, their proposed approaches struggle to model token relationships over longer distances and simply extend to a fixed new token length. Instead, we propose a generalizable method, named TULIP, able to upgrade the token length to any length for CLIP-like models. We do so by improving the architecture with relative position encodings, followed by a training procedure that (i) distills the original CLIP text encoder into an encoder with relative position encodings and (ii) enhances the model for aligning longer captions with images. By effectively encoding captions longer than the default 77 tokens, our model outperforms baselines on cross-modal tasks such as retrieval and text-to-image generation. The code repository is available at https://github.com/ivonajdenkoska/tulip.

Paperid: 1729, https://arxiv.org/pdf/2410.09855.pdf GitHub

Abstract:
Multimodal Large Language Models (MLLMs) have shown exceptional capabilities in vision-language tasks; however, effectively integrating image segmentation into these models remains a significant challenge. In this paper, we introduce Text4Seg, a novel text-as-mask paradigm that casts image segmentation as a text generation problem, eliminating the need for additional decoders and significantly simplifying the segmentation process. Our key innovation is semantic descriptors, a new textual representation of segmentation masks where each image patch is mapped to its corresponding text label. This unified representation allows seamless integration into the auto-regressive training pipeline of MLLMs for easier optimization. We demonstrate that representing an image with $16\times16$ semantic descriptors yields competitive segmentation performance. To enhance efficiency, we introduce the Row-wise Run-Length Encoding (R-RLE), which compresses redundant text sequences, reducing the length of semantic descriptors by 74% and accelerating inference by $3\times$, without compromising performance. Extensive experiments across various vision tasks, such as referring expression segmentation and comprehension, show that Text4Seg achieves state-of-the-art performance on multiple datasets by fine-tuning different MLLM backbones. Our approach provides an efficient, scalable solution for vision-centric tasks within the MLLM framework.

Paperid: 1730, https://arxiv.org/pdf/2410.09733.pdf GitHub

Abstract:
The advent of large Vision-Language Models (VLMs) has significantly advanced multimodal understanding, enabling more sophisticated and accurate integration of visual and textual information across various tasks, including image and video captioning, visual question answering, and cross-modal retrieval. Despite VLMs' superior capabilities, researchers lack a comprehensive understanding of their compositionality -- the ability to understand and produce novel combinations of known visual and textual components. Prior benchmarks provide only a relatively rough compositionality evaluation from the perspectives of objects, relations, and attributes while neglecting deeper reasoning about object interactions, counting, and complex compositions. However, compositionality is a critical ability that facilitates coherent reasoning and understanding across modalities for VLMs. To address this limitation, we propose MMCOMPOSITION, a novel human-annotated benchmark for comprehensively and accurately evaluating VLMs' compositionality. Our proposed benchmark serves as a complement to these earlier works. With MMCOMPOSITION, we can quantify and explore the compositionality of the mainstream VLMs. Surprisingly, we find GPT-4o's compositionality inferior to the best open-source model, and we analyze the underlying reasons. Our experimental analysis reveals the limitations of VLMs in fine-grained compositional perception and reasoning, and points to areas for improvement in VLM design and training. Resources available at: https://hanghuacs.github.io/MMComposition/

Paperid: 1731, https://arxiv.org/pdf/2410.09416.pdf GitHub

Abstract:
This study evaluates the capability of Vision-Language Models (VLMs) in image data annotation by comparing their performance on the CelebA dataset in terms of quality and cost-effectiveness against manual annotation. Annotations from the state-of-the-art LLaVA-NeXT model on 1000 CelebA images are in 79.5% agreement with the original human annotations. Incorporating re-annotations of disagreed cases into a majority vote boosts AI annotation consistency to 89.1% and even higher for more objective labels. Cost assessments demonstrate that AI annotation significantly reduces expenditures compared to traditional manual methods -- representing less than 1% of the costs for manual annotation in the CelebA dataset. These findings support the potential of VLMs as a viable, cost-effective alternative for specific annotation tasks, reducing both financial burden and ethical concerns associated with large-scale manual data annotation. The AI annotations and re-annotations utilized in this study are available on https://github.com/evev2024/EVEV2024_CelebA.

Paperid: 1732, https://arxiv.org/pdf/2410.09132.pdf GitHub

Abstract:
Multimodal Attributed Graphs (MAGs) are ubiquitous in real-world applications, encompassing extensive knowledge through multimodal attributes attached to nodes (e.g., texts and images) and topological structure representing node interactions. Despite its potential to advance diverse research fields like social networks and e-commerce, MAG representation learning (MAGRL) remains underexplored due to the lack of standardized datasets and evaluation frameworks. In this paper, we first propose MAGB, a comprehensive MAG benchmark dataset, featuring curated graphs from various domains with both textual and visual attributes. Based on MAGB dataset, we further systematically evaluate two mainstream MAGRL paradigms: $\textit{GNN-as-Predictor}$, which integrates multimodal attributes via Graph Neural Networks (GNNs), and $\textit{VLM-as-Predictor}$, which harnesses Vision Language Models (VLMs) for zero-shot reasoning. Extensive experiments on MAGB reveal following critical insights: $\textit{(i)}$ Modality significances fluctuate drastically with specific domain characteristics. $\textit{(ii)}$ Multimodal embeddings can elevate the performance ceiling of GNNs. However, intrinsic biases among modalities may impede effective training, particularly in low-data scenarios. $\textit{(iii)}$ VLMs are highly effective at generating multimodal embeddings that alleviate the imbalance between textual and visual attributes. These discoveries, which illuminate the synergy between multimodal attributes and graph topologies, contribute to reliable benchmarks, paving the way for future MAG research. The MAGB dataset and evaluation pipeline are publicly available at https://github.com/sktsherlock/MAGB.

Paperid: 1733, https://arxiv.org/pdf/2410.08791.pdf GitHub

Abstract:
The rapid growth in machine learning models, especially in natural language processing and computer vision, has led to challenges when running these models on hardware with limited resources. This paper introduces Superpipeline, a new framework designed to optimize the execution of large AI models on constrained hardware during both training and inference. Our approach involves dynamically managing model execution by dividing models into individual layers and efficiently transferring these layers between GPU and CPU memory. Superpipeline reduces GPU memory usage by up to 60% in our experiments while maintaining model accuracy and acceptable processing speeds. This allows models that would otherwise exceed available GPU memory to run effectively. Unlike existing solutions that focus mainly on inference or specific model types, Superpipeline can be applied to large language models (LLMs), vision-language models (VLMs), and vision-based models. We tested Superpipeline's performance across various models and hardware setups. The method includes two key parameters that allow fine-tuning the balance between GPU memory use and processing speed. Importantly, Superpipeline does not require retraining or changing model parameters, ensuring that the original model's output remains unchanged. Superpipeline's simplicity and flexibility make it useful for researchers and professionals working with advanced AI models on limited hardware. It enables the use of larger models or bigger batch sizes on existing hardware, potentially speeding up innovation across many machine learning applications. This work marks an important step toward making advanced AI models more accessible and optimizing their deployment in resource-limited environments. The code for Superpipeline is available at https://github.com/abbasiReza/super-pipeline.

Paperid: 1734, https://arxiv.org/pdf/2410.08611.pdf GitHub

Abstract:
A straightforward pipeline for zero-shot out-of-distribution (OOD) detection involves selecting potential OOD labels from an extensive semantic pool and then leveraging a pre-trained vision-language model to perform classification on both in-distribution (ID) and OOD labels. In this paper, we theorize that enhancing performance requires expanding the semantic pool, while increasing the expected probability of selected OOD labels being activated by OOD samples, and ensuring low mutual dependence among the activations of these OOD labels. A natural expansion manner is to adopt a larger lexicon; however, the inevitable introduction of numerous synonyms and uncommon words fails to meet the above requirements, indicating that viable expansion manners move beyond merely selecting words from a lexicon. Since OOD detection aims to correctly classify input images into ID/OOD class groups, we can "make up" OOD label candidates which are not standard class names but beneficial for the process. Observing that the original semantic pool is comprised of unmodified specific class names, we correspondingly construct a conjugated semantic pool (CSP) consisting of modified superclass names, each serving as a cluster center for samples sharing similar properties across different categories. Consistent with our established theory, expanding OOD label candidates with the CSP satisfies the requirements and outperforms existing works by 7.89% in FPR95. Codes are available in https://github.com/MengyuanChen21/NeurIPS2024-CSP.

Paperid: 1735, https://arxiv.org/pdf/2410.08405.pdf GitHub

Abstract:
Significant progress has been made in advancing large multimodal conversational models (LMMs), capitalizing on vast repositories of image-text data available online. Despite this progress, these models often encounter substantial domain gaps, hindering their ability to engage in complex conversations across new domains. Recent efforts have aimed to mitigate this issue, albeit relying on domain-specific image-text data to curate instruction-tuning data. However, many domains, such as agriculture, lack such vision-language data. In this work, we propose an approach to construct instruction-tuning data that harnesses vision-only data for the agriculture domain. We utilize diverse agricultural datasets spanning multiple domains, curate class-specific information, and employ large language models (LLMs) to construct an expert-tuning set, resulting in a 70k expert-tuning dataset called AgroInstruct. Subsequently, we expert-tuned and created AgroGPT, an efficient LMM that can hold complex agriculture-related conversations and provide useful insights. We also develop AgroEvals for evaluation and compare {AgroGPT's} performance with large open and closed-source models. {AgroGPT} excels at identifying fine-grained agricultural concepts, can act as an agriculture expert, and provides helpful information for multimodal agriculture questions. The code, datasets, and models are available at https://github.com/awaisrauf/agroGPT.

Paperid: 1736, https://arxiv.org/pdf/2410.08119.pdf GitHub

Abstract:
In this paper, we propose a post-training quantization framework of large vision-language models (LVLMs) for efficient multi-modal inference. Conventional quantization methods sequentially search the layer-wise rounding functions by minimizing activation discretization errors, which fails to acquire optimal quantization strategy without considering cross-layer dependency. On the contrary, we mine the cross-layer dependency that significantly influences discretization errors of the entire vision-language model, and embed this dependency into optimal quantization strategy searching with low search cost. Specifically, we observe the strong correlation between the activation entropy and the cross-layer dependency concerning output discretization errors. Therefore, we employ the entropy as the proxy to partition blocks optimally, which aims to achieve satisfying trade-offs between discretization errors and the search cost. Moreover, we optimize the visual encoder to disentangle the cross-layer dependency for fine-grained decomposition of search space, so that the search cost is further reduced without harming the quantization accuracy. Experimental results demonstrate that our method compresses the memory by 2.78x and increase generate speed by 1.44x about 13B LLaVA model without performance degradation on diverse multi-modal reasoning tasks. Code is available at https://github.com/ChangyuanWang17/QVLM.

Paperid: 1737, https://arxiv.org/pdf/2410.07540.pdf GitHub GitHub

Abstract:
submucosal dissection (ESD) enables rapid resection of large lesions, minimizing recurrence rates and improving long-term overall survival. Despite these advantages, ESD is technically challenging and carries high risks of complications, necessitating skilled surgeons and precise instruments. Recent advancements in Large Visual-Language Models (LVLMs) offer promising decision support and predictive planning capabilities for robotic systems, which can augment the accuracy of ESD and reduce procedural risks. However, existing datasets for multi-level fine-grained ESD surgical motion understanding are scarce and lack detailed annotations. In this paper, we design a hierarchical decomposition of ESD motion granularity and introduce a multi-level surgical motion dataset (CoPESD) for training LVLMs as the robotic \textbf{Co}-\textbf{P}ilot of \textbf{E}ndoscopic \textbf{S}ubmucosal \textbf{D}issection. CoPESD includes 17,679 images with 32,699 bounding boxes and 88,395 multi-level motions, from over 35 hours of ESD videos for both robot-assisted and conventional surgeries. CoPESD enables granular analysis of ESD motions, focusing on the complex task of submucosal dissection. Extensive experiments on the LVLMs demonstrate the effectiveness of CoPESD in training LVLMs to predict following surgical robotic motions. As the first multimodal ESD motion dataset, CoPESD supports advanced research in ESD instruction-following and surgical automation. The dataset is available at \href{https://github.com/gkw0010/CoPESD}{https://github.com/gkw0010/CoPESD.}}

Paperid: 1738, https://arxiv.org/pdf/2410.07167.pdf GitHub GitHub

Abstract:
We present the Modality Integration Rate (MIR), an effective, robust, and generalized metric to indicate the multi-modal pre-training quality of Large Vision Language Models (LVLMs). Large-scale pre-training plays a critical role in building capable LVLMs, while evaluating its training quality without the costly supervised fine-tuning stage is under-explored. Loss, perplexity, and in-context evaluation results are commonly used pre-training metrics for Large Language Models (LLMs), while we observed that these metrics are less indicative when aligning a well-trained LLM with a new modality. Due to the lack of proper metrics, the research of LVLMs in the critical pre-training stage is hindered greatly, including the training data choice, efficient module design, etc. In this paper, we propose evaluating the pre-training quality from the inter-modal distribution distance perspective and present MIR, the Modality Integration Rate, which is 1) \textbf{Effective} to represent the pre-training quality and show a positive relation with the benchmark performance after supervised fine-tuning. 2) \textbf{Robust} toward different training/evaluation data. 3) \textbf{Generalize} across training configurations and architecture choices. We conduct a series of pre-training experiments to explore the effectiveness of MIR and observe satisfactory results that MIR is indicative about training data selection, training strategy schedule, and model architecture design to get better pre-training results. We hope MIR could be a helpful metric for building capable LVLMs and inspire the following research about modality alignment in different areas. Our code is at: https://github.com/shikiw/Modality-Integration-Rate.

Paperid: 1739, https://arxiv.org/pdf/2410.07157.pdf GitHub

Abstract:
In this paper, we approach an overlooked yet critical task Graph2Image: generating images from multimodal attributed graphs (MMAGs). This task poses significant challenges due to the explosion in graph size, dependencies among graph entities, and the need for controllability in graph conditions. To address these challenges, we propose a graph context-conditioned diffusion model called InstructG2I. InstructG2I first exploits the graph structure and multimodal information to conduct informative neighbor sampling by combining personalized page rank and re-ranking based on vision-language features. Then, a Graph-QFormer encoder adaptively encodes the graph nodes into an auxiliary set of graph prompts to guide the denoising process of diffusion. Finally, we propose graph classifier-free guidance, enabling controllable generation by varying the strength of graph guidance and multiple connected edges to a node. Extensive experiments conducted on three datasets from different domains demonstrate the effectiveness and controllability of our approach. The code is available at https://github.com/PeterGriffinJin/InstructG2I.

Paperid: 1740, https://arxiv.org/pdf/2410.07133.pdf GitHub

Abstract:
Recent advancements in generation models have showcased remarkable capabilities in generating fantastic content. However, most of them are trained on proprietary high-quality data, and some models withhold their parameters and only provide accessible application programming interfaces (APIs), limiting their benefits for downstream tasks. To explore the feasibility of training a text-to-image generation model comparable to advanced models using publicly available resources, we introduce EvolveDirector. This framework interacts with advanced models through their public APIs to obtain text-image data pairs to train a base model. Our experiments with extensive data indicate that the model trained on generated data of the advanced model can approximate its generation capability. However, it requires large-scale samples of 10 million or more. This incurs significant expenses in time, computational resources, and especially the costs associated with calling fee-based APIs. To address this problem, we leverage pre-trained large vision-language models (VLMs) to guide the evolution of the base model. VLM continuously evaluates the base model during training and dynamically updates and refines the training dataset by the discrimination, expansion, deletion, and mutation operations. Experimental results show that this paradigm significantly reduces the required data volume. Furthermore, when approaching multiple advanced models, EvolveDirector can select the best samples generated by them to learn powerful and balanced abilities. The final trained model Edgen is demonstrated to outperform these advanced models. The code and model weights are available at https://github.com/showlab/EvolveDirector.

Paperid: 1741, https://arxiv.org/pdf/2410.06625.pdf GitHub

Abstract:
Vision Language Models (VLMs) have become essential backbones for multimodal intelligence, yet significant safety challenges limit their real-world application. While textual inputs are often effectively safeguarded, adversarial visual inputs can easily bypass VLM defense mechanisms. Existing defense methods are either resource-intensive, requiring substantial data and compute, or fail to simultaneously ensure safety and usefulness in responses. To address these limitations, we propose a novel two-phase inference-time alignment framework, Evaluating Then Aligning (ETA): 1) Evaluating input visual contents and output responses to establish a robust safety awareness in multimodal settings, and 2) Aligning unsafe behaviors at both shallow and deep levels by conditioning the VLMs' generative distribution with an interference prefix and performing sentence-level best-of-N to search the most harmless and helpful generation paths. Extensive experiments show that ETA outperforms baseline methods in terms of harmlessness, helpfulness, and efficiency, reducing the unsafe rate by 87.5% in cross-modality attacks and achieving 96.6% win-ties in GPT-4 helpfulness evaluation. The code is publicly available at https://github.com/DripNowhy/ETA.

Paperid: 1742, https://arxiv.org/pdf/2410.06456.pdf GitHub

Abstract:
Large vision language models (VLMs) combine large language models with vision encoders, demonstrating promise across various tasks. However, they often underperform in task-specific applications due to domain gaps between pre-training and fine-tuning. We introduce VITask, a novel framework that enhances task-specific adaptability of VLMs by integrating task-specific models (TSMs). VITask employs three key strategies: exemplar prompting (EP), response distribution alignment (RDA), and contrastive response tuning (CRT) to improve the task-specific performance of VLMs by adjusting their response distributions. EP allows TSM features to guide VLMs, while RDA enables VLMs to adapt without TSMs during inference by learning from exemplar-prompted models. CRT further optimizes the ranking of correct image-response pairs, thereby reducing the risk of generating undesired responses. Experiments on 12 medical diagnosis datasets across 9 imaging modalities show that VITask outperforms both vanilla instruction-tuned VLMs and TSMs, showcasing its ability to integrate complementary features from both models effectively. Additionally, VITask offers practical advantages such as flexible TSM integration and robustness to incomplete instructions, making it a versatile and efficient solution for task-specific VLM tuning. Our code are available at https://github.com/baiyang4/VITask.

Paperid: 1743, https://arxiv.org/pdf/2410.06234.pdf GitHub

Abstract:
Large vision and language assistants have enabled new capabilities for interpreting natural images. These approaches have recently been adapted to earth observation data, but they are only able to handle single image inputs, limiting their use for many real-world tasks. In this work, we develop a new vision and language assistant called TEOChat that can engage in conversations about temporal sequences of earth observation data. To train TEOChat, we curate an instruction-following dataset composed of many single image and temporal tasks including building change and damage assessment, semantic change detection, and temporal scene classification. We show that TEOChat can perform a wide variety of spatial and temporal reasoning tasks, substantially outperforming previous vision and language assistants, and even achieving comparable or better performance than several specialist models trained to perform specific tasks. Furthermore, TEOChat achieves impressive zero-shot performance on a change detection and change question answering dataset, outperforms GPT-4o and Gemini 1.5 Pro on multiple temporal tasks, and exhibits stronger single image capabilities than a comparable single image instruction-following model on scene classification, visual question answering, and captioning. We publicly release our data, model, and code at https://github.com/ermongroup/TEOChat .

Paperid: 1744, https://arxiv.org/pdf/2410.06154.pdf GitHub

Abstract:
In this work, we propose GLOV, which enables Large Language Models (LLMs) to act as implicit optimizers for Vision-Language Models (VLMs) to enhance downstream vision tasks. GLOV prompts an LLM with the downstream task description, querying it for suitable VLM prompts (e.g., for zero-shot classification with CLIP). These prompts are ranked according to their fitness for the downstream vision task. In each respective optimization step, the ranked prompts are fed as in-context examples (with their accuracies) to equip the LLM with the knowledge of the type of prompts preferred by the downstream VLM. Furthermore, we explicitly guide the LLM's generation at each optimization step by adding an offset vector -- calculated from the embedding differences between previous positive and negative solutions -- to the intermediate layer of the network for the next generation. This offset vector biases the LLM generation toward the type of language the downstream VLM prefers, resulting in enhanced performance on the downstream vision tasks. We comprehensively evaluate our GLOV on two tasks: object recognition and the critical task of enhancing VLM safety. Our GLOV shows performance improvement by up to 15.0% and 57.5% for dual-encoder (e.g., CLIP) and encoder-decoder (e.g., LlaVA) models for object recognition and reduces the attack success rate (ASR) on state-of-the-art VLMs by up to $60.7\%$.

Paperid: 1745, https://arxiv.org/pdf/2410.05650.pdf GitHub

Abstract:
Open-vocabulary detection (OVD) aims to detect novel objects without instance-level annotations to achieve open-world object detection at a lower cost. Existing OVD methods mainly rely on the powerful open-vocabulary image-text alignment capability of Vision-Language Pretrained Models (VLM) such as CLIP. However, CLIP is trained on image-text pairs and lacks the perceptual ability for local regions within an image, resulting in the gap between image and region representations. Directly using CLIP for OVD causes inaccurate region classification. We find the image-region gap is primarily caused by the deformation of region feature maps during region of interest (RoI) extraction. To mitigate the inaccurate region classification in OVD, we propose a new Shape-Invariant Adapter named SIA-OVD to bridge the image-region gap in the OVD task. SIA-OVD learns a set of feature adapters for regions with different shapes and designs a new adapter allocation mechanism to select the optimal adapter for each region. The adapted region representations can align better with text representations learned by CLIP. Extensive experiments demonstrate that SIA-OVD effectively improves the classification accuracy for regions by addressing the gap between images and regions caused by shape deformation. SIA-OVD achieves substantial improvements over representative methods on the COCO-OVD benchmark. The code is available at https://github.com/PKU-ICST-MIPL/SIA-OVD_ACMMM2024.

Paperid: 1746, https://arxiv.org/pdf/2410.05363.pdf GitHub

Abstract:
Text-to-video (T2V) models like Sora have made significant strides in visualizing complex prompts, which is increasingly viewed as a promising path towards constructing the universal world simulator. Cognitive psychologists believe that the foundation for achieving this goal is the ability to understand intuitive physics. However, the capacity of these models to accurately represent intuitive physics remains largely unexplored. To bridge this gap, we introduce PhyGenBench, a comprehensive \textbf{Phy}sics \textbf{Gen}eration \textbf{Ben}chmark designed to evaluate physical commonsense correctness in T2V generation. PhyGenBench comprises 160 carefully crafted prompts across 27 distinct physical laws, spanning four fundamental domains, which could comprehensively assesses models' understanding of physical commonsense. Alongside PhyGenBench, we propose a novel evaluation framework called PhyGenEval. This framework employs a hierarchical evaluation structure utilizing appropriate advanced vision-language models and large language models to assess physical commonsense. Through PhyGenBench and PhyGenEval, we can conduct large-scale automated assessments of T2V models' understanding of physical commonsense, which align closely with human feedback. Our evaluation results and in-depth analysis demonstrate that current models struggle to generate videos that comply with physical commonsense. Moreover, simply scaling up models or employing prompt engineering techniques is insufficient to fully address the challenges presented by PhyGenBench (e.g., dynamic scenarios). We hope this study will inspire the community to prioritize the learning of physical commonsense in these models beyond entertainment applications. We will release the data and codes at https://github.com/OpenGVLab/PhyGenBench

Paperid: 1747, https://arxiv.org/pdf/2410.05270.pdf GitHub

Abstract:
We consider the problem of adapting a contrastively pretrained vision-language model like CLIP (Radford et al., 2021) for few-shot classification. The literature addresses this problem by learning a linear classifier of the frozen visual features, optimizing word embeddings, or learning external feature adapters. We introduce an alternative way for few-shot CLIP adaptation without adding ''external'' parameters to optimize. We find that simply fine-tuning the embedding projection matrix of the vision encoder leads to better performance than all baselines. Furthermore, we show that regularizing training with the distance between the fine-tuned and pretrained matrices adds reliability for adapting CLIP, making the results stable across different learning rates in the ''validation-free'' setting. This simple approach, coined ProLIP, yields state-of-the-art performance on 11 few-shot classification benchmarks, few-shot cross-dataset transfer, domain generalization, and base-to-new class generalization. We also show that ProLIP significantly outperforms prompt tuning when extended to another task of test-time adaptation, while being one order of magnitude faster to train. Code will be made available at: https://github.com/astra-vision/ProLIP .

Paperid: 1748, https://arxiv.org/pdf/2410.05239.pdf GitHub

Abstract:
Vision-Language Models (VLMs) have shown impressive performance in vision tasks, but adapting them to new domains often requires expensive fine-tuning. Prompt tuning techniques, including textual, visual, and multimodal prompting, offer efficient alternatives by leveraging learnable prompts. However, their application to Vision-Language Segmentation Models (VLSMs) and evaluation under significant domain shifts remain unexplored. This work presents an open-source benchmarking framework, TuneVLSeg, to integrate various unimodal and multimodal prompt tuning techniques into VLSMs, making prompt tuning usable for downstream segmentation datasets with any number of classes. TuneVLSeg includes $6$ prompt tuning strategies on various prompt depths used in $2$ VLSMs totaling of $8$ different combinations. We test various prompt tuning on $8$ diverse medical datasets, including $3$ radiology datasets (breast tumor, echocardiograph, chest X-ray pathologies) and $5$ non-radiology datasets (polyp, ulcer, skin cancer), and two natural domain segmentation datasets. Our study found that textual prompt tuning struggles under significant domain shifts, from natural-domain images to medical data. Furthermore, visual prompt tuning, with fewer hyperparameters than multimodal prompt tuning, often achieves performance competitive to multimodal approaches, making it a valuable first attempt. Our work advances the understanding and applicability of different prompt-tuning techniques for robust domain-specific segmentation. The source code is available at https://github.com/naamiinepal/tunevlseg.

Paperid: 1749, https://arxiv.org/pdf/2410.05227.pdf GitHub

Abstract:
High-quality video generation, encompassing text-to-video (T2V), image-to-video (I2V), and video-to-video (V2V) generation, holds considerable significance in content creation to benefit anyone express their inherent creativity in new ways and world simulation to modeling and understanding the world. Models like SORA have advanced generating videos with higher resolution, more natural motion, better vision-language alignment, and increased controllability, particularly for long video sequences. These improvements have been driven by the evolution of model architectures, shifting from UNet to more scalable and parameter-rich DiT models, along with large-scale data expansion and refined training strategies. However, despite the emergence of DiT-based closed-source and open-source models, a comprehensive investigation into their capabilities and limitations remains lacking. Furthermore, the rapid development has made it challenging for recent benchmarks to fully cover SORA-like models and recognize their significant advancements. Additionally, evaluation metrics often fail to align with human preferences.

Paperid: 1750, https://arxiv.org/pdf/2410.05051.pdf GitHub

Abstract:
In this paper, we propose HE-Drive: the first human-like-centric end-to-end autonomous driving system to generate trajectories that are both temporally consistent and comfortable. Recent studies have shown that imitation learning-based planners and learning-based trajectory scorers can effectively generate and select accuracy trajectories that closely mimic expert demonstrations. However, such trajectory planners and scorers face the dilemma of generating temporally inconsistent and uncomfortable trajectories. To solve the above problems, Our HE-Drive first extracts key 3D spatial representations through sparse perception, which then serves as conditional inputs for a Conditional Denoising Diffusion Probabilistic Models (DDPMs)-based motion planner to generate temporal consistency multi-modal trajectories. A Vision-Language Models (VLMs)-guided trajectory scorer subsequently selects the most comfortable trajectory from these candidates to control the vehicle, ensuring human-like end-to-end driving. Experiments show that HE-Drive not only achieves state-of-the-art performance (i.e., reduces the average collision rate by 71% than VAD) and efficiency (i.e., 1.9X faster than SparseDrive) on the challenging nuScenes and OpenScene datasets but also provides the most comfortable driving experience on real-world data.For more information, visit the project website: https://jmwang0117.github.io/HE-Drive/.

Paperid: 1751, https://arxiv.org/pdf/2410.04652.pdf GitHub

Abstract:
Seamless integration of virtual and physical worlds in augmented reality benefits from the system semantically "understanding" the physical environment. AR research has long focused on the potential of context awareness, demonstrating novel capabilities that leverage the semantics in the 3D environment for various object-level interactions. Meanwhile, the computer vision community has made leaps in neural vision-language understanding to enhance environment perception for autonomous tasks. In this work, we introduce a multimodal 3D object representation that unifies both semantic and linguistic knowledge with the geometric representation, enabling user-guided machine learning involving physical objects. We first present a fast multimodal 3D reconstruction pipeline that brings linguistic understanding to AR by fusing CLIP vision-language features into the environment and object models. We then propose "in-situ" machine learning, which, in conjunction with the multimodal representation, enables new tools and interfaces for users to interact with physical spaces and objects in a spatially and linguistically meaningful manner. We demonstrate the usefulness of the proposed system through two real-world AR applications on Magic Leap 2: a) spatial search in physical environments with natural language and b) an intelligent inventory system that tracks object changes over time. We also make our full implementation and demo data available at (https://github.com/cy-xu/spatially_aware_AI) to encourage further exploration and research in spatially aware AI.

Paperid: 1752, https://arxiv.org/pdf/2410.04417.pdf GitHub

Abstract:
In vision-language models (VLMs), visual tokens usually bear a significant amount of computational overhead despite sparsity of information in them when compared to text tokens. To address this, most existing methods learn a network to prune redundant visual tokens using certain training data. Differently, we propose a text-guided training-free token optimization mechanism dubbed SparseVLM that eliminates the need of extra parameters or fine-tuning costs. Given that visual tokens complement text tokens in VLM's linguistic reasoning, we select relevant text tokens to rate the significance of visual tokens using self-attention matrices and, then, prune visual tokens using the proposed strategy to maximize sparsity while retaining information. In particular, we introduce a rank-based strategy to adaptively determine the sparsification ratio for each layer, alongside a token recycling method that compresses pruned tokens into more compact representations. Experimental results show that SparseVLM increases the efficiency of various VLMs in a number of image and video understanding tasks. For example, LLaVA when equipped with SparseVLM achieves 54% reduction in FLOPs, 37% decrease in CUDA latency while maintaining 97% of its original accuracy. Our code is available at https://github.com/Gumpest/SparseVLMs.

Paperid: 1753, https://arxiv.org/pdf/2410.04345.pdf GitHub

Abstract:
Humans perform visual perception at multiple levels, including low-level object recognition and high-level semantic interpretation such as behavior understanding. Subtle differences in low-level details can lead to substantial changes in high-level perception. For example, substituting the shopping bag held by a person with a gun suggests violent behavior, implying criminal or violent activity. Despite significant advancements in various multimodal tasks, Large Visual-Language Models (LVLMs) remain unexplored in their capabilities to conduct such multi-level visual perceptions. To investigate the perception gap between LVLMs and humans, we introduce MVP-Bench, the first visual-language benchmark systematically evaluating both low- and high-level visual perception of LVLMs. We construct MVP-Bench across natural and synthetic images to investigate how manipulated content influences model perception. Using MVP-Bench, we diagnose the visual perception of 10 open-source and 2 closed-source LVLMs, showing that high-level perception tasks significantly challenge existing LVLMs. The state-of-the-art GPT-4o only achieves an accuracy of $56\%$ on Yes/No questions, compared with $74\%$ in low-level scenarios. Furthermore, the performance gap between natural and manipulated images indicates that current LVLMs do not generalize in understanding the visual semantics of synthetic images as humans do. Our data and code are publicly available at https://github.com/GuanzhenLi/MVP-Bench.

Paperid: 1754, https://arxiv.org/pdf/2410.04107.pdf GitHub

Abstract:
Large Vision-Language Models (LVLMs) have achieved remarkable progress on visual perception and linguistic interpretation. Despite their impressive capabilities across various tasks, LVLMs still suffer from the issue of hallucination, which involves generating content that is incorrect or unfaithful to the visual or textual inputs. Traditional benchmarks, such as MME and POPE, evaluate hallucination in LVLMs within the scope of Visual Question Answering (VQA) using answerable questions. However, some questions are unanswerable due to insufficient information in the images, and the performance of LVLMs on such unanswerable questions remains underexplored. To bridge this research gap, we propose TUBench, a benchmark specifically designed to evaluate the reliability of LVLMs using unanswerable questions. TUBench comprises an extensive collection of high-quality, unanswerable questions that are meticulously crafted using ten distinct strategies. To thoroughly evaluate LVLMs, the unanswerable questions in TUBench are based on images from four diverse domains as visual contexts: screenshots of code snippets, natural images, geometry diagrams, and screenshots of statistical tables. These unanswerable questions are tailored to test LVLMs' trustworthiness in code reasoning, commonsense reasoning, geometric reasoning, and mathematical reasoning related to tables, respectively. We conducted a comprehensive quantitative evaluation of 28 leading foundational models on TUBench, with Gemini-1.5-Pro, the top-performing model, achieving an average accuracy of 69.2%, and GPT-4o, the third-ranked model, reaching 66.7% average accuracy, in determining whether questions are answerable. TUBench is available at https://github.com/NLPCode/TUBench.

Paperid: 1755, https://arxiv.org/pdf/2410.03659.pdf GitHub

Abstract:
Large Vision-Language Models (LVLMs) have demonstrated impressive capabilities for capturing and reasoning over multimodal inputs. However, these models are prone to parametric knowledge conflicts, which arise from inconsistencies of represented knowledge between their vision and language components. In this paper, we formally define the problem of $\textbf{cross-modality parametric knowledge conflict}$ and present a systematic approach to detect, interpret, and mitigate them. We introduce a pipeline that identifies conflicts between visual and textual answers, showing a persistently high conflict rate across modalities in recent LVLMs regardless of the model size. We further investigate how these conflicts interfere with the inference process and propose a contrastive metric to discern the conflicting samples from the others. Building on these insights, we develop a novel dynamic contrastive decoding method that removes undesirable logits inferred from the less confident modality components based on answer confidence. For models that do not provide logits, we also introduce two prompt-based strategies to mitigate the conflicts. Our methods achieve promising improvements in accuracy on both the ViQuAE and InfoSeek datasets. Specifically, using LLaVA-34B, our proposed dynamic contrastive decoding improves an average accuracy of 2.24%.

Paperid: 1756, https://arxiv.org/pdf/2410.03161.pdf GitHub GitHub

Abstract:
In recent years, zero-shot and few-shot learning in visual grounding have garnered considerable attention, largely due to the success of large-scale vision-language pre-training on expansive datasets such as LAION-5B and DataComp-1B. However, the continuous expansion of these datasets presents significant challenges, particularly with respect to data availability and computational overhead, thus creating a bottleneck in the advancement of low-shot learning capabilities. In this paper, we propose IMAGE, Interpretative MAsking with Gaussian radiation modEling, aimed at enhancing vocabulary grounding in low-shot learning scenarios without necessitating an increase in dataset size. Drawing inspiration from cognitive science and the recent success of masked autoencoders (MAE), our method leverages adaptive masking on salient regions of the feature maps generated by the vision backbone. This enables the model to learn robust, generalized representations through the reconstruction of occluded information, thereby facilitating effective attention to both local and global features. We evaluate the efficacy of our approach on benchmark datasets, including COCO and ODinW, demonstrating its superior performance in zero-shot and few-shot tasks. Experimental results consistently show that IMAGE outperforms baseline models, achieving enhanced generalization and improved performance in low-shot scenarios. These findings highlight the potential of adaptive feature manipulation through attention mechanisms and Gaussian modeling as a promising alternative to approaches that rely on the continual scaling of dataset sizes for the advancement of zero-shot and few-shot learning. Our code is publicly available at https://github.com/git-lenny/IMAGE.

Paperid: 1757, https://arxiv.org/pdf/2410.02874.pdf GitHub

Abstract:
Although there is a growing demand for cooking behaviours as one of the expected tasks for robots, a series of cooking behaviours based on new recipe descriptions by robots in the real world has not yet been realised. In this study, we propose a robot system that integrates real-world executable robot cooking behaviour planning using the Large Language Model (LLM) and classical planning of PDDL descriptions, and food ingredient state recognition learning from a small number of data using the Vision-Language model (VLM). We succeeded in experiments in which PR2, a dual-armed wheeled robot, performed cooking from arranged new recipes in a real-world environment, and confirmed the effectiveness of the proposed system.

Paperid: 1758, https://arxiv.org/pdf/2410.02762.pdf GitHub

Abstract:
We investigate the internal representations of vision-language models (VLMs) to address hallucinations, a persistent challenge despite advances in model size and training. We project VLMs' internal image representations to their language vocabulary and observe more confident output probabilities on real objects than hallucinated objects. We additionally use these output probabilities to spatially localize real objects. Building on this approach, we introduce a knowledge erasure algorithm that removes hallucinations by linearly orthogonalizing image features with respect to hallucinated object features. We show that targeted edits to a model's latent representations can reduce hallucinations by up to 25.7% on the COCO2014 dataset while preserving performance. Our findings demonstrate how a deeper understanding of VLMs' latent representations can enhance reliability and enable novel capabilities, such as zero-shot segmentation.

Paperid: 1759, https://arxiv.org/pdf/2410.02193.pdf GitHub

Abstract:
Vision-Language Models (VLM) can generate plausible high-level plans when prompted with a goal, the context, an image of the scene, and any planning constraints. However, there is no guarantee that the predicted actions are geometrically and kinematically feasible for a particular robot embodiment. As a result, many prerequisite steps such as opening drawers to access objects are often omitted in their plans. Robot task and motion planners can generate motion trajectories that respect the geometric feasibility of actions and insert physically necessary actions, but do not scale to everyday problems that require common-sense knowledge and involve large state spaces comprised of many variables. We propose VLM-TAMP, a hierarchical planning algorithm that leverages a VLM to generate goth semantically-meaningful and horizon-reducing intermediate subgoals that guide a task and motion planner. When a subgoal or action cannot be refined, the VLM is queried again for replanning. We evaluate VLM- TAMP on kitchen tasks where a robot must accomplish cooking goals that require performing 30-50 actions in sequence and interacting with up to 21 objects. VLM-TAMP substantially outperforms baselines that rigidly and independently execute VLM-generated action sequences, both in terms of success rates (50 to 100% versus 0%) and average task completion percentage (72 to 100% versus 15 to 45%). See project site https://zt-yang.github.io/vlm-tamp-robot/ for more information.

Paperid: 1760, https://arxiv.org/pdf/2410.01971.pdf GitHub

Abstract:
Vision-language-action (VLA) models trained on large-scale internet data and robot demonstrations have the potential to serve as generalist robot policies. However, despite their large-scale training, VLAs are often brittle to task-irrelevant visual details such as distractor objects or background colors. We introduce Bring Your Own VLA (BYOVLA): a run-time intervention scheme that (1) dynamically identifies regions of the input image that the model is sensitive to, and (2) minimally alters task-irrelevant regions to reduce the model's sensitivity using automated image editing tools. Our approach is compatible with any off the shelf VLA without model fine-tuning or access to the model's weights. Hardware experiments on language-instructed manipulation tasks demonstrate that BYOVLA enables state-of-the-art VLA models to nearly retain their nominal performance in the presence of distractor objects and backgrounds, which otherwise degrade task success rates by up to 40%. Website with additional information, videos, and code: https://aasherh.github.io/byovla/ .

Paperid: 1761, https://arxiv.org/pdf/2410.01912.pdf GitHub GitHub

Abstract:
This work tackles the information loss bottleneck of vector-quantization (VQ) autoregressive image generation by introducing a novel model architecture called the 2-Dimensional Autoregression (DnD) Transformer. The DnD-Transformer predicts more codes for an image by introducing a new autoregression direction, \textit{model depth}, along with the sequence length direction. Compared to traditional 1D autoregression and previous work utilizing similar 2D image decomposition such as RQ-Transformer, the DnD-Transformer is an end-to-end model that can generate higher quality images with the same backbone model size and sequence length, opening a new optimization perspective for autoregressive image generation. Furthermore, our experiments reveal that the DnD-Transformer's potential extends beyond generating natural images. It can even generate images with rich text and graphical elements in a self-supervised manner, demonstrating an understanding of these combined modalities. This has not been previously demonstrated for popular vision generative models such as diffusion models, showing a spark of vision-language intelligence when trained solely on images. Code, datasets and models are open at https://github.com/chenllliang/DnD-Transformer.

Paperid: 1762, https://arxiv.org/pdf/2410.01744.pdf GitHub GitHub

Abstract:
Text-rich images, where text serves as the central visual element guiding the overall understanding, are prevalent in real-world applications, such as presentation slides, scanned documents, and webpage snapshots. Tasks involving multiple text-rich images are especially challenging, as they require not only understanding the content of individual images but reasoning about inter-relationships and logical flows across multiple visual inputs. Despite the importance of these scenarios, current multimodal large language models (MLLMs) struggle to handle such tasks due to two key challenges: (1) the scarcity of high-quality instruction tuning datasets for text-rich multi-image scenarios, and (2) the difficulty in balancing image resolution with visual feature sequence length. To address these challenges, we propose Leopard, an MLLM tailored for handling vision-language tasks involving multiple text-rich images. First, we curated about one million high-quality multimodal instruction-tuning data, tailored to text-rich, multi-image scenarios. Second, we proposed an adaptive high-resolution multi-image encoding module to dynamically optimize the allocation of visual sequence length based on the original aspect ratios and resolutions of images. Experiments on a diverse set of benchmarks reveal that our model consistently outperforms state-of-the-art systems, such as Llama-3.2 and Qwen2-VL, in challenging text-rich, multi-image evaluations. Remarkably, our approach achieves outstanding performance using only 1.2M training instances, all of which are fully open-sourced, demonstrating both high efficiency and effectiveness compared to models trained on large-scale in-house data. Our code and data are available at https://github.com/tencent-ailab/Leopard.

Paperid: 1763, https://arxiv.org/pdf/2410.01620.pdf GitHub

Abstract:
The prevalence of vision-threatening eye diseases is a significant global burden, with many cases remaining undiagnosed or diagnosed too late for effective treatment. Large vision-language models (LVLMs) have the potential to assist in understanding anatomical information, diagnosing eye diseases, and drafting interpretations and follow-up plans, thereby reducing the burden on clinicians and improving access to eye care. However, limited benchmarks are available to assess LVLMs' performance in ophthalmology-specific applications. In this study, we introduce LMOD, a large-scale multimodal ophthalmology benchmark consisting of 21,993 instances across (1) five ophthalmic imaging modalities: optical coherence tomography, color fundus photographs, scanning laser ophthalmoscopy, lens photographs, and surgical scenes; (2) free-text, demographic, and disease biomarker information; and (3) primary ophthalmology-specific applications such as anatomical information understanding, disease diagnosis, and subgroup analysis. In addition, we benchmarked 13 state-of-the-art LVLM representatives from closed-source, open-source, and medical domains. The results demonstrate a significant performance drop for LVLMs in ophthalmology compared to other domains. Systematic error analysis further identified six major failure modes: misclassification, failure to abstain, inconsistent reasoning, hallucination, assertions without justification, and lack of domain-specific knowledge. In contrast, supervised neural networks specifically trained on these tasks as baselines demonstrated high accuracy. These findings underscore the pressing need for benchmarks in the development and validation of ophthalmology-specific LVLMs.

Paperid: 1764, https://arxiv.org/pdf/2410.01407.pdf GitHub GitHub

Abstract:
Capitalizing on vast amount of image-text data, large-scale vision-language pre-training has demonstrated remarkable zero-shot capabilities and has been utilized in several applications. However, models trained on general everyday web-crawled data often exhibit sub-optimal performance for specialized domains, likely due to domain shift. Recent works have tackled this problem for some domains (e.g., healthcare) by constructing domain-specialized image-text data. However, constructing a dedicated large-scale image-text dataset for sustainable area of agriculture and livestock is still open to research. Further, this domain desires fine-grained feature learning due to the subtle nature of the downstream tasks (e.g, nutrient deficiency detection, livestock breed classification). To address this we present AgriCLIP, a vision-language foundational model dedicated to the domain of agriculture and livestock. First, we propose a large-scale dataset, named ALive, that leverages customized prompt generation strategy to overcome the scarcity of expert annotations. Our ALive dataset covers crops, livestock, and fishery, with around 600,000 image-text pairs. Second, we propose a training pipeline that integrates both contrastive and self-supervised learning to learn both global semantic and local fine-grained domain-specialized features. Experiments on diverse set of 20 downstream tasks demonstrate the effectiveness of AgriCLIP framework, achieving an absolute gain of 7.8\% in terms of average zero-shot classification accuracy, over the standard CLIP adaptation via domain-specialized ALive dataset. Our ALive dataset and code can be accessible at \href{https://github.com/umair1221/AgriCLIP/tree/main}{Github}.

Paperid: 1765, https://arxiv.org/pdf/2410.00982.pdf GitHub

Abstract:
Accurately identifying, understanding and describing traffic safety-critical events (SCEs), including crashes, tire strikes, and near-crashes, is crucial for advanced driver assistance systems, automated driving systems, and traffic safety. As SCEs are rare events, most general vision-language models (VLMs) have not been trained sufficiently to link SCE videos and narratives, which could lead to hallucinations and missing key safety characteristics. Here, we introduce ScVLM, a novel hybrid methodology that integrates supervised and contrastive learning techniques to classify the severity and types of SCEs, as well as to generate narrative descriptions of SCEs. This approach utilizes classification to enhance VLMs' comprehension of driving videos and improve the rationality of event descriptions. The proposed approach is trained on and evaluated by more than 8,600 SCEs from the Second Strategic Highway Research Program Naturalistic Driving Study dataset, the largest publicly accessible driving dataset with videos and SCE annotations. The results demonstrate the superiority of the proposed approach in generating contextually accurate event descriptions and mitigating VLM hallucinations. The code will be available at https://github.com/datadrivenwheels/ScVLM.

Paperid: 1766, https://arxiv.org/pdf/2410.00309.pdf GitHub

Abstract:
Social dynamics in close human interactions pose significant challenges for Human Mesh Estimation (HME), particularly due to the complexity of physical contacts and the scarcity of training data. Addressing these challenges, we introduce a novel data generation method that utilizes Large Vision Language Models (LVLMs) to annotate contact maps which guide test-time optimization to produce paired image and pseudo-ground truth meshes. This methodology not only alleviates the annotation burden but also enables the assembly of a comprehensive dataset specifically tailored for close interactions in HME. Our Ask Pose Unite (APU) dataset, comprising over 6.2k human mesh pairs in contact covering diverse interaction types, is curated from images depicting naturalistic person-to-person scenes. We empirically show that using our dataset to train a diffusion-based contact prior, used as guidance during optimization, improves mesh estimation on unseen interactions. Our work addresses longstanding challenges of data scarcity for close interactions in HME enhancing the field's capabilities of handling complex interaction scenarios.

Paperid: 1767, https://arxiv.org/pdf/2409.20424.pdf GitHub

Abstract:
Recent advances in Vision-Language Models (VLMs) and the scarcity of high-quality multi-modal alignment data have inspired numerous researches on synthetic VLM data generation. The conventional norm in VLM data construction uses a mixture of specialists in caption and OCR, or stronger VLM APIs and expensive human annotation. In this paper, we present World to Code (W2C), a meticulously curated multi-modal data construction pipeline that organizes the final generation output into a Python code format. The pipeline leverages the VLM itself to extract cross-modal information via different prompts and filter the generated outputs again via a consistency filtering strategy. Experiments have demonstrated the high quality of W2C by improving various existing visual question answering and visual grounding benchmarks across different VLMs. Further analysis also demonstrates that the new code parsing ability of VLMs presents better cross-modal equivalence than the commonly used detail caption ability. Our code is available at https://github.com/foundation-multimodal-models/World2Code.

Paperid: 1768, https://arxiv.org/pdf/2409.20353.pdf GitHub

Abstract:
Machine learning models are increasingly being deployed in real-world contexts. However, systematic studies on their transferability to specific and critical applications are underrepresented in the research literature. An important example is visual anomaly detection (VAD) for robotic power line inspection. While existing VAD methods perform well in controlled environments, real-world scenarios present diverse and unexpected anomalies that current datasets fail to capture. To address this gap, we introduce $\textit{CableInspect-AD}$, a high-quality, publicly available dataset created and annotated by domain experts from Hydro-QuÃ©bec, a Canadian public utility. This dataset includes high-resolution images with challenging real-world anomalies, covering defects with varying severity levels. To address the challenges of collecting diverse anomalous and nominal examples for setting a detection threshold, we propose an enhancement to the celebrated PatchCore algorithm. This enhancement enables its use in scenarios with limited labeled data. We also present a comprehensive evaluation protocol based on cross-validation to assess models' performances. We evaluate our $\textit{Enhanced-PatchCore}$ for few-shot and many-shot detection, and Vision-Language Models for zero-shot detection. While promising, these models struggle to detect all anomalies, highlighting the dataset's value as a challenging benchmark for the broader research community. Project page: https://mila-iqia.github.io/cableinspect-ad/.

Paperid: 1769, https://arxiv.org/pdf/2409.19967.pdf GitHub

Abstract:
Text-to-image diffusion models particularly Stable Diffusion, have revolutionized the field of computer vision. However, the synthesis quality often deteriorates when asked to generate images that faithfully represent complex prompts involving multiple attributes and objects. While previous studies suggest that blended text embeddings lead to improper attribute binding, few have explored this in depth. In this work, we critically examine the limitations of the CLIP text encoder in understanding attributes and investigate how this affects diffusion models. We discern a phenomenon of attribute bias in the text space and highlight a contextual issue in padding embeddings that entangle different concepts. We propose \textbf{Magnet}, a novel training-free approach to tackle the attribute binding problem. We introduce positive and negative binding vectors to enhance disentanglement, further with a neighbor strategy to increase accuracy. Extensive experiments show that Magnet significantly improves synthesis quality and binding accuracy with negligible computational cost, enabling the generation of unconventional and unnatural concepts.

Paperid: 1770, https://arxiv.org/pdf/2409.19846.pdf GitHub

Abstract:
Large-scale vision-language models like CLIP have demonstrated impressive open-vocabulary capabilities for image-level tasks, excelling in recognizing what objects are present. However, they struggle with pixel-level recognition tasks like semantic segmentation, which additionally require understanding where the objects are located. In this work, we propose a novel method, PixelCLIP, to adapt the CLIP image encoder for pixel-level understanding by guiding the model on where, which is achieved using unlabeled images and masks generated from vision foundation models such as SAM and DINO. To address the challenges of leveraging masks without semantic labels, we devise an online clustering algorithm using learnable class names to acquire general semantic concepts. PixelCLIP shows significant performance improvements over CLIP and competitive results compared to caption-supervised methods in open-vocabulary semantic segmentation. Project page is available at https://cvlab-kaist.github.io/PixelCLIP

Paperid: 1771, https://arxiv.org/pdf/2409.19840.pdf GitHub

Abstract:
In our study, we explore methods for detecting unwanted content lurking in visual datasets. We provide a theoretical analysis demonstrating that a model capable of successfully partitioning visual data can be obtained using only textual data. Based on the analysis, we propose Hassle-Free Textual Training (HFTT), a streamlined method capable of acquiring detectors for unwanted visual content, using only synthetic textual data in conjunction with pre-trained vision-language models. HFTT features an innovative objective function that significantly reduces the necessity for human involvement in data annotation. Furthermore, HFTT employs a clever textual data synthesis method, effectively emulating the integration of unknown visual data distribution into the training process at no extra cost. The unique characteristics of HFTT extend its utility beyond traditional out-of-distribution detection, making it applicable to tasks that address more abstract concepts. We complement our analyses with experiments in out-of-distribution detection and hateful image detection. Our codes are available at https://github.com/Saehyung-Lee/HFTT

Paperid: 1772, https://arxiv.org/pdf/2409.19806.pdf GitHub

Abstract:
Audio-Language Models (ALMs) have recently achieved remarkable success in zero-shot audio recognition tasks, which match features of audio waveforms with class-specific text prompt features, inspired by advancements in Vision-Language Models (VLMs). Given the sensitivity of zero-shot performance to the choice of hand-crafted text prompts, many prompt learning techniques have been developed for VLMs. We explore the efficacy of these approaches in ALMs and propose a novel method, Prompt Learning in Audio Language Models (PALM), which optimizes the feature space of the text encoder branch. Unlike existing methods that work in the input space, our approach results in greater training efficiency. We demonstrate the effectiveness of our approach on 11 audio recognition datasets, encompassing a variety of speech-processing tasks, and compare the results with three baselines in a few-shot learning setup. Our method is either on par with or outperforms other approaches while being computationally less demanding. Code is available at https://asif-hanif.github.io/palm/

Paperid: 1773, https://arxiv.org/pdf/2409.19474.pdf GitHub

Abstract:
Despite significant advancements and pervasive use of vision-language models, a paucity of studies has addressed their ethical implications. These models typically require extensive training data, often from hastily reviewed text and image datasets, leading to highly imbalanced datasets and ethical concerns. Additionally, models initially trained in English are frequently fine-tuned for other languages, such as the CLIP model, which can be expanded with more data to enhance capabilities but can add new biases. The CAPIVARA, a CLIP-based model adapted to Portuguese, has shown strong performance in zero-shot tasks. In this paper, we evaluate four different types of discriminatory practices within visual-language models and introduce FairPIVARA, a method to reduce them by removing the most affected dimensions of feature embeddings. The application of FairPIVARA has led to a significant reduction of up to 98% in observed biases while promoting a more balanced word distribution within the model. Our model and code are available at: https://github.com/hiaac-nlp/FairPIVARA.

Paperid: 1774, https://arxiv.org/pdf/2409.18674.pdf GitHub

Abstract:
Predicting and explaining the private information contained in an image in human-understandable terms is a complex and contextual task. This task is challenging even for large language models. To facilitate the understanding of privacy decisions, we propose to predict image privacy based on a set of natural language content descriptors. These content descriptors are associated with privacy scores that reflect how people perceive image content. We generate descriptors with our novel Image-guided Topic Modeling (ITM) approach. ITM leverages, via multimodality alignment, both vision information and image textual descriptions from a vision language model. We use the ITM-generated descriptors to learn a privacy predictor, Priv$\times$ITM, whose decisions are interpretable by design. Our Priv$\times$ITM classifier outperforms the reference interpretable method by 5 percentage points in accuracy and performs comparably to the current non-interpretable state-of-the-art model.

Paperid: 1775, https://arxiv.org/pdf/2409.18125.pdf GitHub

Abstract:
Recent advancements in Large Multimodal Models (LMMs) have greatly enhanced their proficiency in 2D visual understanding tasks, enabling them to effectively process and understand images and videos. However, the development of LMMs with 3D scene understanding capabilities has been hindered by the lack of large-scale 3D vision-language datasets and powerful 3D encoders. In this paper, we introduce a simple yet effective framework called LLaVA-3D. Leveraging the strong 2D visual understanding priors from LLaVA, our LLaVA-3D efficiently adapts LLaVA for 3D scene understanding without compromising 2D understanding capabilities. To achieve this, we utilize the 3D position embeddings to enhance the 2D CLIP Patches with 3D spatial context information and construct 3D patches. By integrating the 3D position embeddings into 2D LMMs and employing joint 2D and 3D vision-language instruction tuning, we establish a unified architecture for both 2D visual understanding and 3D scene understanding. In contrast to previous 3D LMMs, LLaVA-3D supports decoding accurate 3D spatial perception outputs, e.g., 3D bounding boxes, directly from these 3D patches, without relying on the time-consuming off-the-shelf 3D segmentors. Experimental results show that LLaVA-3D converges 3.5x faster than existing 3D LMMs when trained on 3D vision-language datasets. Moreover, LLaVA-3D not only achieves state-of-the-art performance across various 3D tasks but also maintains comparable 2D visual understanding and vision-language conversation capabilities with LLaVA.

Paperid: 1776, https://arxiv.org/pdf/2409.17805.pdf GitHub

Abstract:
Prompt learning has surfaced as an effective approach to enhance the performance of Vision-Language Models (VLMs) like CLIP when applied to downstream tasks. However, current learnable prompt tokens are primarily used for the single phase of adapting to tasks (i.e., adapting prompt), easily leading to overfitting risks. In this work, we propose a novel Cascade Prompt Learning CasPL framework to enable prompt learning to serve both generic and specific expertise (i.e., boosting and adapting prompt) simultaneously. Specifically, CasPL is a new learning paradigm comprising two distinct phases of learnable prompts: the first boosting prompt is crafted to extract domain-general knowledge from a senior larger CLIP teacher model by aligning their predicted logits using extensive unlabeled domain images. The second adapting prompt is then cascaded with the frozen first set to fine-tune the downstream tasks, following the approaches employed in prior research. In this manner, CasPL can effectively capture both domain-general and task-specific representations into explicitly different gradual groups of prompts, thus potentially alleviating overfitting issues in the target domain. It's worth noting that CasPL serves as a plug-and-play module that can seamlessly integrate into any existing prompt learning approach. CasPL achieves a significantly better balance between performance and inference speed, which is especially beneficial for deploying smaller VLM models in resource-constrained environments. Compared to the previous state-of-the-art method PromptSRC, CasPL shows an average improvement of 1.85% for base classes, 3.44% for novel classes, and 2.72% for the harmonic mean over 11 image classification datasets. Code is publicly available at: https://github.com/megvii-research/CasPL.

Paperid: 1777, https://arxiv.org/pdf/2409.17663.pdf GitHub

Abstract:
Recent concept-based interpretable models have succeeded in providing meaningful explanations by pre-defined concept sets. However, the dependency on the pre-defined concepts restricts the application because of the limited number of concepts for explanations. This paper proposes a novel interpretable deep neural network called explanation bottleneck models (XBMs). XBMs generate a text explanation from the input without pre-defined concepts and then predict a final task prediction based on the generated explanation by leveraging pre-trained vision-language encoder-decoder models. To achieve both the target task performance and the explanation quality, we train XBMs through the target task loss with the regularization penalizing the explanation decoder via the distillation from the frozen pre-trained decoder. Our experiments, including a comparison to state-of-the-art concept bottleneck models, confirm that XBMs provide accurate and fluent natural language explanations without pre-defined concept sets. Code is available at https://github.com/yshinya6/xbm/.

Paperid: 1778, https://arxiv.org/pdf/2409.17519.pdf GitHub

Abstract:
In order for robots to autonomously navigate and operate in diverse environments, it is essential for them to recognize the state of their environment. On the other hand, the environmental state recognition has traditionally involved distinct methods tailored to each state to be recognized. In this study, we perform a unified environmental state recognition for robots through the spoken language with pre-trained large-scale vision-language models. We apply Visual Question Answering and Image-to-Text Retrieval, which are tasks of Vision-Language Models. We show that with our method, it is possible to recognize not only whether a room door is open/closed, but also whether a transparent door is open/closed and whether water is running in a sink, without training neural networks or manual programming. In addition, the recognition accuracy can be improved by selecting appropriate texts from the set of prepared texts based on black-box optimization. For each state recognition, only the text set and its weighting need to be changed, eliminating the need to prepare multiple different models and programs, and facilitating the management of source code and computer resource. We experimentally demonstrate the effectiveness of our method and apply it to the recognition behavior on a mobile robot, Fetch.

Paperid: 1779, https://arxiv.org/pdf/2409.17313.pdf GitHub

Abstract:
This study presents a novel evaluation framework for the Vision-Language Navigation (VLN) task. It aims to diagnose current models for various instruction categories at a finer-grained level. The framework is structured around the context-free grammar (CFG) of the task. The CFG serves as the basis for the problem decomposition and the core premise of the instruction categories design. We propose a semi-automatic method for CFG construction with the help of Large-Language Models (LLMs). Then, we induct and generate data spanning five principal instruction categories (i.e. direction change, landmark recognition, region recognition, vertical movement, and numerical comprehension). Our analysis of different models reveals notable performance discrepancies and recurrent issues. The stagnation of numerical comprehension, heavy selective biases over directional concepts, and other interesting findings contribute to the development of future language-guided navigation systems.

Paperid: 1780, https://arxiv.org/pdf/2409.17143.pdf GitHub

Abstract:
Compared with Large Language Models (LLMs), Large Vision-Language Models (LVLMs) can also accept images as input, thus showcasing more interesting emergent capabilities and demonstrating impressive performance on various vision-language tasks. Motivated by text prompting in LLMs, visual prompting has been explored to enhance LVLMs' capabilities of perceiving visual information. However, previous visual prompting techniques solely process visual inputs without considering text queries, limiting the models' ability to follow text instructions to complete tasks. To fill this gap, in this work, we propose a new prompting technique named Attention Prompting on Image, which just simply overlays a text-query-guided attention heatmap on the original input image and effectively enhances LVLM on various tasks. Specifically, we generate an attention heatmap for the input image dependent on the text query with an auxiliary model like CLIP. Then the heatmap simply multiplies the pixel values of the original image to obtain the actual input image for the LVLM. Extensive experiments on various vison-language benchmarks verify the effectiveness of our technique. For example, Attention Prompting on Image improves LLaVA-1.5 by 3.8% and 2.9% on MM-Vet and LLaVA-Wild benchmarks, respectively.

Paperid: 1781, https://arxiv.org/pdf/2409.17080.pdf GitHub

Abstract:
Large vision-language models (VLMs) have become state-of-the-art for many computer vision tasks, with in-context learning (ICL) as a popular adaptation strategy for new ones. But can VLMs learn novel concepts purely from visual demonstrations, or are they limited to adapting to the output format of ICL examples? We propose a new benchmark we call Spatial Visual Ambiguity Tasks (SVAT) that challenges state-of-the-art VLMs to learn new visuospatial tasks in-context. We find that VLMs fail to do this zero-shot, and sometimes continue to fail after finetuning. However, adding simpler data to the training by curriculum learning leads to improved ICL performance.

Paperid: 1782, https://arxiv.org/pdf/2409.16902.pdf GitHub GitHub

Abstract:
Over the past decade, significant progress has been made in visual object tracking, largely due to the availability of large-scale datasets. However, these datasets have primarily focused on open-air scenarios and have largely overlooked underwater animal tracking-especially the complex challenges posed by camouflaged marine animals. To bridge this gap, we take a step forward by proposing the first large-scale multi-modal underwater camouflaged object tracking dataset, namely UW-COT220. Based on the proposed dataset, this work first comprehensively evaluates current advanced visual object tracking methods, including SAM- and SAM2-based trackers, in challenging underwater environments, \eg, coral reefs. Our findings highlight the improvements of SAM2 over SAM, demonstrating its enhanced ability to handle the complexities of underwater camouflaged objects. Furthermore, we propose a novel vision-language tracking framework called VL-SAM2, based on the video foundation model SAM2. Extensive experimental results demonstrate that the proposed VL-SAM2 achieves state-of-the-art performance across underwater and open-air object tracking datasets. The dataset and codes are available at~{\color{magenta}{https://github.com/983632847/Awesome-Multimodal-Object-Tracking}}.

Paperid: 1783, https://arxiv.org/pdf/2409.16718.pdf GitHub

Abstract:
Recent advances in fine-tuning Vision-Language Models (VLMs) have witnessed the success of prompt tuning and adapter tuning, while the classic model fine-tuning on inherent parameters seems to be overlooked. It is believed that fine-tuning the parameters of VLMs with few-shot samples corrupts the pre-trained knowledge since fine-tuning the CLIP model even degrades performance. In this paper, we revisit this viewpoint, and propose a new perspective: fine-tuning the specific parameters instead of all will uncover the power of classic model fine-tuning on VLMs. Through our meticulous study, we propose ClipFit, a simple yet effective method to fine-tune CLIP without introducing any overhead of extra parameters. We demonstrate that by only fine-tuning the specific bias terms and normalization layers, ClipFit can improve the performance of zero-shot CLIP by 7.27\% average harmonic mean accuracy. Lastly, to understand how fine-tuning in CLIPFit affects the pre-trained models, we conducted extensive experimental analyses w.r.t. changes in internal parameters and representations. We found that low-level text bias layers and the first layer normalization layer change much more than other layers. The code is available at \url{https://github.com/minglllli/CLIPFit}.

Paperid: 1784, https://arxiv.org/pdf/2409.16597.pdf GitHub

Abstract:
Recently, Multimodal Large Language Models (MLLMs) have made significant progress in the video comprehension field. Despite remarkable content reasoning and instruction following capabilities they demonstrated, the hallucination problem of these VideoLLMs is less explored compared with its counterpart in the image domain. To mitigate this gap, we propose EventHallusion, a novel benchmark that focuses on assessing the VideoLLMs' hallucination toward event, the crux of video analysis. From a hallucination attribution perspective, our EventHallusion benchmark is curated to assess a VideoLLM's susceptibility toward language priors and vision-language biases. On the other hand, we also propose a simple yet effective method, called Temporal Contrastive Decoding (TCD), to tackle the hallucination problems of VideoLLMs. The proposed TCD method rectifies the model's bias toward its priors during the decoding stage by comparing the original video with a modified version, in which temporal cues are disrupted. Through comprehensive evaluation of eight open-source and two closed-source VideoLLMs on the proposed EventHallusion benchmark, we observe that the open-source models suffer significantly from hallucination problems, whereas the closed-source ones perform markedly better. By further equipping open-source VideoLLMs with the proposed TCD approach, evident performance improvements are achieved across most metrics in the EventHallusion benchmark. Our codes and benchmark data are available at https://github.com/Stevetich/EventHallusion.

Paperid: 1785, https://arxiv.org/pdf/2409.16159.pdf GitHub GitHub

Abstract:
The comic domain is rapidly advancing with the development of single- and multi-page analysis and synthesis models. Recent benchmarks and datasets have been introduced to support and assess models' capabilities in tasks such as detection (panels, characters, text), linking (character re-identification and speaker identification), and analysis of comic elements (e.g., dialog transcription). However, to provide a comprehensive understanding of the storyline, a model must not only extract elements but also understand their relationships and generate highly informative captions. In this work, we propose a pipeline that leverages Vision-Language Models (VLMs) to obtain dense, grounded captions. To construct our pipeline, we introduce an attribute-retaining metric that assesses whether all important attributes are identified in the caption. Additionally, we created a densely annotated test set to fairly evaluate open-source VLMs and select the best captioning model according to our metric. Our pipeline generates dense captions with bounding boxes that are quantitatively and qualitatively superior to those produced by specifically trained models, without requiring any additional training. Using this pipeline, we annotated over 2 million panels across 13,000 books, which will be available on the project page https://github.com/emanuelevivoli/ComiCap.

Paperid: 1786, https://arxiv.org/pdf/2409.15256.pdf GitHub

Abstract:
Large Vision-Language Models (LVLMs) evolve rapidly as Large Language Models (LLMs) was equipped with vision modules to create more human-like models. However, we should carefully evaluate their applications in different domains, as they may possess undesired biases. Our work studies the potential behavioral biases of LVLMs from a behavioral finance perspective, an interdisciplinary subject that jointly considers finance and psychology. We propose an end-to-end framework, from data collection to new evaluation metrics, to assess LVLMs' reasoning capabilities and the dynamic behaviors manifested in two established human financial behavioral biases: recency bias and authority bias. Our evaluations find that recent open-source LVLMs such as LLaVA-NeXT, MobileVLM-V2, Mini-Gemini, MiniCPM-Llama3-V 2.5 and Phi-3-vision-128k suffer significantly from these two biases, while the proprietary model GPT-4o is negligibly impacted. Our observations highlight directions in which open-source models can improve. The code is available at https://github.com/mydcxiao/vlm_behavioral_fin.

Paperid: 1787, https://arxiv.org/pdf/2409.14963.pdf GitHub

Abstract:
In smart retail applications, the large number of products and their frequent turnover necessitate reliable zero-shot object classification methods. The zero-shot assumption is essential to avoid the need for re-training the classifier every time a new product is introduced into stock or an existing product undergoes rebranding. In this paper, we make three key contributions. Firstly, we introduce the MIMEX dataset, comprising 28 distinct product categories. Unlike existing datasets in the literature, MIMEX focuses on fine-grained product classification and includes a diverse range of retail products. Secondly, we benchmark the zero-shot object classification performance of state-of-the-art vision-language models (VLMs) on the proposed MIMEX dataset. Our experiments reveal that these models achieve unsatisfactory fine-grained classification performance, highlighting the need for specialized approaches. Lastly, we propose a novel ensemble approach that integrates embeddings from CLIP and DINOv2 with dimensionality reduction techniques to enhance classification performance. By combining these components, our ensemble approach outperforms VLMs, effectively capturing visual cues crucial for fine-grained product discrimination. Additionally, we introduce a class adaptation method that utilizes visual prototyping with limited samples in scenarios with scarce labeled data, addressing a critical need in retail environments where product variety frequently changes. To encourage further research into zero-shot object classification for smart retail applications, we will release both the MIMEX dataset and benchmark to the research community. Interested researchers can contact the authors for details on the terms and conditions of use. The code is available: https://github.com/AnilOsmanTur/Zero-shot-Retail-Product-Classification.

Paperid: 1788, https://arxiv.org/pdf/2409.14713.pdf GitHub

Abstract:
The success of visual instruction tuning has accelerated the development of large language and vision models (LLVMs). Following the scaling laws of instruction-tuned large language models (LLMs), LLVMs either have further increased their sizes, reaching 26B, 34B, and even 80B parameters. While this increase in model size has yielded significant performance gains, it demands substantially more hardware resources for both training and inference. Consequently, there naturally exists a strong need for efficient LLVMs that achieve the performance of larger models while being smaller in size. To achieve this need, we present a new efficient LLVM family with model sizes of 0.5B, 1.8B, 3.8B, and 7B parameters, Phantom, which significantly enhances learning capabilities within limited structures. By temporarily increasing the latent hidden dimension during multi-head self-attention (MHSA), we make LLVMs prepare to look and understand much more vision-language knowledge on the latent, without substantially increasing physical model sizes. To maximize its advantage, we introduce Phantom Optimization (PO) using both autoregressive supervised fine-tuning (SFT) and direct preference optimization (DPO)-like concept, which effectively follows correct answers while eliminating incorrect and ambiguous ones. Phantom outperforms numerous larger open- and closed-source LLVMs, positioning itself as a leading solution in the landscape of efficient LLVMs.

Paperid: 1789, https://arxiv.org/pdf/2409.14484.pdf GitHub

Abstract:
Recent studies have shown that Vision Language Large Models (VLLMs) may output content not relevant to the input images. This problem, called the hallucination phenomenon, undoubtedly degrades VLLM performance. Therefore, various anti-hallucination techniques have been proposed to make model output more reasonable and accurate. Despite their successes, from extensive tests we found that augmenting the prompt (e.g. word appending, rewriting, and spell error etc.) may change model output and make the output hallucinate again. To cure this drawback, we propose a new instruct-tuning framework called Prompt Augmentation and Caption Utilization (PACU) to boost VLLM's generation ability under the augmented prompt scenario. Concretely, on the one hand, PACU exploits existing LLMs to augment and evaluate diverse prompts automatically. The resulting high-quality prompts are utilized to enhance VLLM's ability to process different prompts. On the other hand, PACU exploits image captions to jointly work with image features as well as the prompts for response generation. When the visual feature is inaccurate, LLM can capture useful information from the image captions for response generation. Extensive experiments on hallucination evaluation and prompt-augmented datasets demonstrate that our PACU method can work well with existing schemes to effectively boost VLLM model performance. Code is available in https://github.com/zhaominyiz/PACU.

Paperid: 1790, https://arxiv.org/pdf/2409.14215.pdf GitHub

Abstract:
As Vision-Language Models (VLMs) advance, human-centered Assistive Technologies (ATs) for helping People with Visual Impairments (PVIs) are evolving into generalists, capable of performing multiple tasks simultaneously. However, benchmarking VLMs for ATs remains under-explored. To bridge this gap, we first create a novel AT benchmark (@Bench). Guided by a pre-design user study with PVIs, our benchmark includes the five most crucial vision-language tasks: Panoptic Segmentation, Depth Estimation, Optical Character Recognition (OCR), Image Captioning, and Visual Question Answering (VQA). Besides, we propose a novel AT model (@Model) that addresses all tasks simultaneously and can be expanded to more assistive functions for helping PVIs. Our framework exhibits outstanding performance across tasks by integrating multi-modal information, and it offers PVIs a more comprehensive assistance. Extensive experiments prove the effectiveness and generalizability of our framework.

Paperid: 1791, https://arxiv.org/pdf/2409.14163.pdf GitHub

Abstract:
Source-free domain generalization (SFDG) tackles the challenge of adapting models to unseen target domains without access to source domain data. To deal with this challenging task, recent advances in SFDG have primarily focused on leveraging the text modality of vision-language models such as CLIP. These methods involve developing a transferable linear classifier based on diverse style features extracted from the text and learned prompts or deriving domain-unified text representations from domain banks. However, both style features and domain banks have limitations in capturing comprehensive domain knowledge. In this work, we propose Prompt-Driven Text Adapter (PromptTA) method, which is designed to better capture the distribution of style features and employ resampling to ensure thorough coverage of domain knowledge. To further leverage this rich domain information, we introduce a text adapter that learns from these style features for efficient domain information storage. Extensive experiments conducted on four benchmark datasets demonstrate that PromptTA achieves state-of-the-art performance. The code is available at https://github.com/zhanghr2001/PromptTA.

Paperid: 1792, https://arxiv.org/pdf/2409.14083.pdf GitHub

Abstract:
Large Vision-Language Models (LVLMs) have become pivotal at the intersection of computer vision and natural language processing. However, the full potential of LVLMs Retrieval-Augmented Generation (RAG) capabilities remains underutilized. Existing works either focus solely on the text modality or are limited to specific tasks. Moreover, most LVLMs struggle to selectively utilize retrieved information and are sensitive to irrelevant or misleading references. To address these challenges, we propose a self-refinement framework designed to teach LVLMs to Selectively Utilize Retrieved Information (SURf). Specifically, when given questions that are incorrectly answered by the LVLM backbone, we obtain references that help correct the answers (positive references) and those that do not (negative references). We then fine-tune the LVLM backbone using a combination of these positive and negative references. Our experiments across three tasks and seven datasets demonstrate that our framework significantly enhances LVLMs ability to effectively utilize retrieved multimodal references and improves their robustness against irrelevant or misleading information. The source code is available at https://github.com/GasolSun36/SURf.

Paperid: 1793, https://arxiv.org/pdf/2409.13592.pdf GitHub

Abstract:
Understanding satire and humor is a challenging task for even current Vision-Language models. In this paper, we propose the challenging tasks of Satirical Image Detection (detecting whether an image is satirical), Understanding (generating the reason behind the image being satirical), and Completion (given one half of the image, selecting the other half from 2 given options, such that the complete image is satirical) and release a high-quality dataset YesBut, consisting of 2547 images, 1084 satirical and 1463 non-satirical, containing different artistic styles, to evaluate those tasks. Each satirical image in the dataset depicts a normal scenario, along with a conflicting scenario which is funny or ironic. Despite the success of current Vision-Language Models on multimodal tasks such as Visual QA and Image Captioning, our benchmarking experiments show that such models perform poorly on the proposed tasks on the YesBut Dataset in Zero-Shot Settings w.r.t both automated as well as human evaluation. Additionally, we release a dataset of 119 real, satirical photographs for further research. The dataset and code are available at https://github.com/abhi1nandy2/yesbut_dataset.

Paperid: 1794, https://arxiv.org/pdf/2409.13496.pdf GitHub

Abstract:
Autonomous vehicles and robots often struggle with reliable visual perception at night due to the low illumination and motion blur caused by the long exposure time of RGB cameras. Existing methods address this challenge by sequentially connecting the off-the-shelf pretrained low-light enhancement and deblurring models. Unfortunately, these methods often lead to noticeable artifacts (\eg, color distortions) in the over-exposed regions or make it hardly possible to learn the motion cues of the dark regions. In this paper, we interestingly find vision-language models, \eg, Contrastive Language-Image Pretraining (CLIP), can comprehensively perceive diverse degradation levels at night. In light of this, we propose a novel transformer-based joint learning framework, named DAP-LED, which can jointly achieve low-light enhancement and deblurring, benefiting downstream tasks, such as depth estimation, segmentation, and detection in the dark. The key insight is to leverage CLIP to adaptively learn the degradation levels from images at night. This subtly enables learning rich semantic information and visual representation for optimization of the joint tasks. To achieve this, we first introduce a CLIP-guided cross-fusion module to obtain multi-scale patch-wise degradation heatmaps from the image embeddings. Then, the heatmaps are fused via the designed CLIP-enhanced transformer blocks to retain useful degradation information for effective model optimization. Experimental results show that, compared to existing methods, our DAP-LED achieves state-of-the-art performance in the dark. Meanwhile, the enhanced results are demonstrated to be effective for three downstream tasks. For demo and more results, please check the project page: \url{https://vlislab22.github.io/dap-led/}.

Paperid: 1795, https://arxiv.org/pdf/2409.13162.pdf GitHub

Abstract:
Detecting anomalies within point clouds is crucial for various industrial applications, but traditional unsupervised methods face challenges due to data acquisition costs, early-stage production constraints, and limited generalization across product categories. To overcome these challenges, we introduce the Multi-View Projection (MVP) framework, leveraging pre-trained Vision-Language Models (VLMs) to detect anomalies. Specifically, MVP projects point cloud data into multi-view depth images, thereby translating point cloud anomaly detection into image anomaly detection. Following zero-shot image anomaly detection methods, pre-trained VLMs are utilized to detect anomalies on these depth images. Given that pre-trained VLMs are not inherently tailored for zero-shot point cloud anomaly detection and may lack specificity, we propose the integration of learnable visual and adaptive text prompting techniques to fine-tune these VLMs, thereby enhancing their detection performance. Extensive experiments on the MVTec 3D-AD and Real3D-AD demonstrate our proposed MVP framework's superior zero-shot anomaly detection performance and the prompting techniques' effectiveness. Real-world evaluations on automotive plastic part inspection further showcase that the proposed method can also be generalized to practical unseen scenarios. The code is available at https://github.com/hustCYQ/MVP-PCLIP.

Paperid: 1796, https://arxiv.org/pdf/2409.12889.pdf GitHub

Abstract:
Recently, large language model (LLM)-based agents have made significant advances across various fields. One of the most popular research areas involves applying these agents to video games. Traditionally, these methods have relied on game APIs to access in-game environmental and action data. However, this approach is limited by the availability of APIs and does not reflect how humans play games. With the advent of vision language models (VLMs), agents now have enhanced visual understanding capabilities, enabling them to interact with games using only visual inputs. Despite these advances, current approaches still face challenges in action-oriented tasks, particularly in action role-playing games (ARPGs), where reinforcement learning methods are prevalent but suffer from poor generalization and require extensive training. To address these limitations, we select an ARPG, ``Black Myth: Wukong'', as a research platform to explore the capability boundaries of existing VLMs in scenarios requiring visual-only input and complex action output. We define 12 tasks within the game, with 75% focusing on combat, and incorporate several state-of-the-art VLMs into this benchmark. Additionally, we will release a human operation dataset containing recorded gameplay videos and operation logs, including mouse and keyboard actions. Moreover, we propose a novel VARP (Vision Action Role-Playing) agent framework, consisting of an action planning system and a visual trajectory system. Our framework demonstrates the ability to perform basic tasks and succeed in 90% of easy and medium-level combat scenarios. This research aims to provide new insights and directions for applying multimodal agents in complex action game environments. The code and datasets will be made available at https://varp-agent.github.io/.

Paperid: 1797, https://arxiv.org/pdf/2409.12191.pdf GitHub GitHub

Abstract:
We present the Qwen2-VL Series, an advanced upgrade of the previous Qwen-VL models that redefines the conventional predetermined-resolution approach in visual processing. Qwen2-VL introduces the Naive Dynamic Resolution mechanism, which enables the model to dynamically process images of varying resolutions into different numbers of visual tokens. This approach allows the model to generate more efficient and accurate visual representations, closely aligning with human perceptual processes. The model also integrates Multimodal Rotary Position Embedding (M-RoPE), facilitating the effective fusion of positional information across text, images, and videos. We employ a unified paradigm for processing both images and videos, enhancing the model's visual perception capabilities. To explore the potential of large multimodal models, Qwen2-VL investigates the scaling laws for large vision-language models (LVLMs). By scaling both the model size-with versions at 2B, 8B, and 72B parameters-and the amount of training data, the Qwen2-VL Series achieves highly competitive performance. Notably, the Qwen2-VL-72B model achieves results comparable to leading models such as GPT-4o and Claude3.5-Sonnet across various multimodal benchmarks, outperforming other generalist models. Code is available at https://github.com/QwenLM/Qwen2-VL .

Paperid: 1798, https://arxiv.org/pdf/2409.11919.pdf GitHub

Abstract:
Vision Language Models (VLMs) have demonstrated remarkable capabilities in various open-vocabulary tasks, yet their zero-shot performance lags behind task-specific fine-tuned models, particularly in complex tasks like Referring Expression Comprehension (REC). Fine-tuning usually requires 'white-box' access to the model's architecture and weights, which is not always feasible due to proprietary or privacy concerns. In this work, we propose LLM-wrapper, a method for 'black-box' adaptation of VLMs for the REC task using Large Language Models (LLMs). LLM-wrapper capitalizes on the reasoning abilities of LLMs, improved with a light fine-tuning, to select the most relevant bounding box matching the referring expression, from candidates generated by a zero-shot black-box VLM. Our approach offers several advantages: it enables the adaptation of closed-source models without needing access to their internal workings, it is versatile as it works with any VLM, it transfers to new VLMs and datasets, and it allows for the adaptation of an ensemble of VLMs. We evaluate LLM-wrapper on multiple datasets using different VLMs and LLMs, demonstrating significant performance improvements and highlighting the versatility of our method. While LLM-wrapper is not meant to directly compete with standard white-box fine-tuning, it offers a practical and effective alternative for black-box VLM adaptation. Code and checkpoints are available at https://github.com/valeoai/LLM_wrapper .

Paperid: 1799, https://arxiv.org/pdf/2409.11363.pdf GitHub

Abstract:
AI agents have the potential to aid users on a variety of consequential tasks, including conducting scientific research. To spur the development of useful agents, we need benchmarks that are challenging, but more crucially, directly correspond to real-world tasks of interest. This paper introduces such a benchmark, designed to measure the accuracy of AI agents in tackling a crucial yet surprisingly challenging aspect of scientific research: computational reproducibility. This task, fundamental to the scientific process, involves reproducing the results of a study using the provided code and data. We introduce CORE-Bench (Computational Reproducibility Agent Benchmark), a benchmark consisting of 270 tasks based on 90 scientific papers across three disciplines (computer science, social science, and medicine). Tasks in CORE-Bench consist of three difficulty levels and include both language-only and vision-language tasks. We provide an evaluation system to measure the accuracy of agents in a fast and parallelizable way, saving days of evaluation time for each run compared to a sequential implementation. We evaluated two baseline agents: the general-purpose AutoGPT and a task-specific agent called CORE-Agent. We tested both variants using two underlying language models: GPT-4o and GPT-4o-mini. The best agent achieved an accuracy of 21% on the hardest task, showing the vast scope for improvement in automating routine scientific tasks. Having agents that can reproduce existing work is a necessary step towards building agents that can conduct novel research and could verify and improve the performance of other research agents. We hope that CORE-Bench can improve the state of reproducibility and spur the development of future research agents.

Paperid: 1800, https://arxiv.org/pdf/2409.10921.pdf GitHub

Abstract:
Exploring the narratives conveyed by fine-art paintings is a challenge in image captioning, where the goal is to generate descriptions that not only precisely represent the visual content but also offer a in-depth interpretation of the artwork's meaning. The task is particularly complex for artwork images due to their diverse interpretations and varied aesthetic principles across different artistic schools and styles. In response to this, we present KALE Knowledge-Augmented vision-Language model for artwork Elaborations), a novel approach that enhances existing vision-language models by integrating artwork metadata as additional knowledge. KALE incorporates the metadata in two ways: firstly as direct textual input, and secondly through a multimodal heterogeneous knowledge graph. To optimize the learning of graph representations, we introduce a new cross-modal alignment loss that maximizes the similarity between the image and its corresponding metadata. Experimental results demonstrate that KALE achieves strong performance (when evaluated with CIDEr, in particular) over existing state-of-the-art work across several artwork datasets. Source code of the project is available at https://github.com/Yanbei-Jiang/Artwork-Interpretation.

Paperid: 1801, https://arxiv.org/pdf/2409.10419.pdf GitHub

Abstract:
Robots interacting with humans through natural language can unlock numerous applications such as Referring Grasp Synthesis (RGS). Given a text query, RGS determines a stable grasp pose to manipulate the referred object in the robot's workspace. RGS comprises two steps: visual grounding and grasp pose estimation. Recent studies leverage powerful Vision-Language Models (VLMs) for visually grounding free-flowing natural language in real-world robotic execution. However, comparisons in complex, cluttered environments with multiple instances of the same object are lacking. This paper introduces HiFi-CS, featuring hierarchical application of Featurewise Linear Modulation (FiLM) to fuse image and text embeddings, enhancing visual grounding for complex attribute rich text queries encountered in robotic grasping. Visual grounding associates an object in 2D/3D space with natural language input and is studied in two scenarios: Closed and Open Vocabulary. HiFi-CS features a lightweight decoder combined with a frozen VLM and outperforms competitive baselines in closed vocabulary settings while being 100x smaller in size. Our model can effectively guide open-set object detectors like GroundedSAM to enhance open-vocabulary performance. We validate our approach through real-world RGS experiments using a 7-DOF robotic arm, achieving 90.33\% visual grounding accuracy in 15 tabletop scenes. Our codebase is provided here: https://github.com/vineet2104/hifics

Paperid: 1802, https://arxiv.org/pdf/2409.09502.pdf GitHub GitHub

Abstract:
Vision-language models have recently evolved into versatile systems capable of high performance across a range of tasks, such as document understanding, visual question answering, and grounding, often in zero-shot settings. Comics Understanding, a complex and multifaceted field, stands to greatly benefit from these advances. Comics, as a medium, combine rich visual and textual narratives, challenging AI models with tasks that span image classification, object detection, instance segmentation, and deeper narrative comprehension through sequential panels. However, the unique structure of comics -- characterized by creative variations in style, reading order, and non-linear storytelling -- presents a set of challenges distinct from those in other visual-language domains. In this survey, we present a comprehensive review of Comics Understanding from both dataset and task perspectives. Our contributions are fivefold: (1) We analyze the structure of the comics medium, detailing its distinctive compositional elements; (2) We survey the widely used datasets and tasks in comics research, emphasizing their role in advancing the field; (3) We introduce the Layer of Comics Understanding (LoCU) framework, a novel taxonomy that redefines vision-language tasks within comics and lays the foundation for future work; (4) We provide a detailed review and categorization of existing methods following the LoCU framework; (5) Finally, we highlight current research challenges and propose directions for future exploration, particularly in the context of vision-language models applied to comics. This survey is the first to propose a task-oriented framework for comics intelligence and aims to guide future research by addressing critical gaps in data availability and task definition. A project associated with this survey is available at https://github.com/emanuelevivoli/awesome-comics-understanding.

Paperid: 1803, https://arxiv.org/pdf/2409.09369.pdf GitHub

Abstract:
Histopathology Whole-Slide Images (WSIs) provide an important tool to assess cancer prognosis in computational pathology (CPATH). While existing survival analysis (SA) approaches have made exciting progress, they are generally limited to adopting highly-expressive network architectures and only coarse-grained patient-level labels to learn visual prognostic representations from gigapixel WSIs. Such learning paradigm suffers from critical performance bottlenecks, when facing present scarce training data and standard multi-instance learning (MIL) framework in CPATH. To overcome it, this paper, for the first time, proposes a new Vision-Language-based SA (VLSA) paradigm. Concretely, (1) VLSA is driven by pathology VL foundation models. It no longer relies on high-capability networks and shows the advantage of data efficiency. (2) In vision-end, VLSA encodes textual prognostic prior and then employs it as auxiliary signals to guide the aggregating of visual prognostic features at instance level, thereby compensating for the weak supervision in MIL. Moreover, given the characteristics of SA, we propose i) ordinal survival prompt learning to transform continuous survival labels into textual prompts; and ii) ordinal incidence function as prediction target to make SA compatible with VL-based prediction. Notably, VLSA's predictions can be interpreted intuitively by our Shapley values-based method. The extensive experiments on five datasets confirm the effectiveness of our scheme. Our VLSA could pave a new way for SA in CPATH by offering weakly-supervised MIL an effective means to learn valuable prognostic clues from gigapixel WSIs. Our source code is available at https://github.com/liupei101/VLSA.

Paperid: 1804, https://arxiv.org/pdf/2409.09306.pdf GitHub

Abstract:
Current vision-language multimodal models are well-adapted for general visual understanding tasks. However, they perform inadequately when handling complex visual tasks related to human poses and actions due to the lack of specialized vision-language instruction-following data. We introduce a method for generating such data by integrating human keypoints with traditional visual features such as captions and bounding boxes, enabling more precise understanding of human-centric scenes. Our approach constructs a dataset comprising 200,328 samples tailored to fine-tune models for human-centric tasks, focusing on three areas: conversation, detailed description, and complex reasoning. We establish a benchmark called Human Pose and Action Understanding Benchmark (HPAUB) to assess model performance on human pose and action understanding. We fine-tune the LLaVA-1.5-7B model using this dataset and evaluate it on the benchmark, achieving significant improvements. Experimental results show an overall improvement of 21.18% compared to the original LLaVA-1.5-7B model. These findings highlight the effectiveness of keypoint-integrated data in enhancing multimodal models. Code is available at https://github.com/Ody-trek/Keypoint-Instruction-Tuning.

Paperid: 1805, https://arxiv.org/pdf/2409.09276.pdf GitHub

Abstract:
Tactile perception is vital, especially when distinguishing visually similar objects. We propose an approach to incorporate tactile data into a Vision-Language Model (VLM) for visuo-tactile zero-shot object recognition. Our approach leverages the zero-shot capability of VLMs to infer tactile properties from the names of tactilely similar objects. The proposed method translates tactile data into a textual description solely by annotating object names for each tactile sequence during training, making it adaptable to various contexts with low training costs. The proposed method was evaluated on the FoodReplica and Cube datasets, demonstrating its effectiveness in recognizing objects that are difficult to distinguish by vision alone.

Paperid: 1806, https://arxiv.org/pdf/2409.08582.pdf GitHub

Abstract:
Remote sensing (RS) change analysis is vital for monitoring Earth's dynamic processes by detecting alterations in images over time. Traditional change detection excels at identifying pixel-level changes but lacks the ability to contextualize these alterations. While recent advancements in change captioning offer natural language descriptions of changes, they do not support interactive, user-specific queries. To address these limitations, we introduce ChangeChat, the first bitemporal vision-language model (VLM) designed specifically for RS change analysis. ChangeChat utilizes multimodal instruction tuning, allowing it to handle complex queries such as change captioning, category-specific quantification, and change localization. To enhance the model's performance, we developed the ChangeChat-87k dataset, which was generated using a combination of rule-based methods and GPT-assisted techniques. Experiments show that ChangeChat offers a comprehensive, interactive solution for RS change analysis, achieving performance comparable to or even better than state-of-the-art (SOTA) methods on specific tasks, and significantly surpassing the latest general-domain model, GPT-4. Code and pre-trained weights are available at https://github.com/hanlinwu/ChangeChat.

Paperid: 1807, https://arxiv.org/pdf/2409.07683.pdf GitHub

Abstract:
Open-vocabulary image semantic segmentation (OVS) seeks to segment images into semantic regions across an open set of categories. Existing OVS methods commonly depend on foundational vision-language models and utilize similarity computation to tackle OVS tasks. However, these approaches are predominantly tailored to natural images and struggle with the unique characteristics of remote sensing images, such as rapidly changing orientations and significant scale variations. These challenges complicate OVS tasks in earth vision, requiring specialized approaches. To tackle this dilemma, we propose the first OVS framework specifically designed for remote sensing imagery, drawing inspiration from the distinct remote sensing traits. Particularly, to address the varying orientations, we introduce a rotation-aggregative similarity computation module that generates orientation-adaptive similarity maps as initial semantic maps. These maps are subsequently refined at both spatial and categorical levels to produce more accurate semantic maps. Additionally, to manage significant scale changes, we integrate multi-scale image features into the upsampling process, resulting in the final scale-aware semantic masks. To advance OVS in earth vision and encourage reproducible research, we establish the first open-sourced OVS benchmark for remote sensing imagery, including four public remote sensing datasets. Extensive experiments on this benchmark demonstrate our proposed method achieves state-of-the-art performance. All codes and datasets are available at https://github.com/caoql98/OVRS.

Paperid: 1808, https://arxiv.org/pdf/2409.07353.pdf GitHub

Abstract:
Large Vision-Language Models (LVLMs), trained on multimodal big datasets, have significantly advanced AI by excelling in vision-language tasks. However, these models remain vulnerable to adversarial attacks, particularly jailbreak attacks, which bypass safety protocols and cause the model to generate misleading or harmful responses. This vulnerability stems from both the inherent susceptibilities of LLMs and the expanded attack surface introduced by the visual modality. We propose Sim-CLIP+, a novel defense mechanism that adversarially fine-tunes the CLIP vision encoder by leveraging a Siamese architecture. This approach maximizes cosine similarity between perturbed and clean samples, facilitating resilience against adversarial manipulations. Sim-CLIP+ offers a plug-and-play solution, allowing seamless integration into existing LVLM architectures as a robust vision encoder. Unlike previous defenses, our method requires no structural modifications to the LVLM and incurs minimal computational overhead. Sim-CLIP+ demonstrates effectiveness against both gradient-based adversarial attacks and various jailbreak techniques. We evaluate Sim-CLIP+ against three distinct jailbreak attack strategies and perform clean evaluations using standard downstream datasets, including COCO for image captioning and OKVQA for visual question answering. Extensive experiments demonstrate that Sim-CLIP+ maintains high clean accuracy while substantially improving robustness against both gradient-based adversarial attacks and jailbreak techniques. Our code and robust vision encoders are available at https://github.com/speedlab-git/Robust-Encoder-against-Jailbreak-attack.git.

Paperid: 1809, https://arxiv.org/pdf/2409.06809.pdf GitHub

Abstract:
In this paper, we introduce DetailCLIP: A Detail-Oriented CLIP to address the limitations of contrastive learning-based vision-language models, particularly CLIP, in handling detail-oriented and fine-grained tasks like segmentation. While CLIP and its variants excel in the global alignment of image and text representations, they often struggle to capture the fine-grained details necessary for precise segmentation. To overcome these challenges, we propose a novel framework that employs patch-level comparison of self-distillation and pixel-level reconstruction losses, enhanced with an attention-based token removal mechanism. This approach selectively retains semantically relevant tokens, enabling the model to focus on the image's critical regions aligned with the specific functions of our model, including textual information processing, patch comparison, and image reconstruction, ensuring that the model learns high-level semantics and detailed visual features. Our experiments demonstrate that DetailCLIP surpasses existing CLIP-based and traditional self-supervised learning (SSL) models in segmentation accuracy and exhibits superior generalization across diverse datasets. DetailCLIP represents a significant advancement in vision-language modeling, offering a robust solution for tasks that demand high-level semantic understanding and detailed feature extraction. https://github.com/KishoreP1/DetailCLIP.

Paperid: 1810, https://arxiv.org/pdf/2409.06481.pdf GitHub

Abstract:
Negation is a fundamental linguistic concept used by humans to convey information that they do not desire. Despite this, minimal research has focused on negation within text-guided image editing. This lack of research means that vision-language models (VLMs) for image editing may struggle to understand negation, implying that they struggle to provide accurate results. One barrier to achieving human-level intelligence is the lack of a standard collection by which research into negation can be evaluated. This paper presents the first large-scale dataset, Negative Instruction (NeIn), for studying negation within instruction-based image editing. Our dataset comprises 366,957 quintuplets, i.e., source image, original caption, selected object, negative sentence, and target image in total, including 342,775 queries for training and 24,182 queries for benchmarking image editing methods. Specifically, we automatically generate NeIn based on a large, existing vision-language dataset, MS-COCO, via two steps: generation and filtering. During the generation phase, we leverage two VLMs, BLIP and InstructPix2Pix (fine-tuned on MagicBrush dataset), to generate NeIn's samples and the negative clauses that expresses the content of the source image. In the subsequent filtering phase, we apply BLIP and LLaVA-NeXT to remove erroneous samples. Additionally, we introduce an evaluation protocol to assess the negation understanding for image editing models. Extensive experiments using our dataset across multiple VLMs for text-guided image editing demonstrate that even recent state-of-the-art VLMs struggle to understand negative queries.

Paperid: 1811, https://arxiv.org/pdf/2409.05405.pdf GitHub

Abstract:
In the real world, where information is abundant and diverse across different modalities, understanding and utilizing various data types to improve retrieval systems is a key focus of research. Multimodal composite retrieval integrates diverse modalities such as text, image and audio, etc. to provide more accurate, personalized, and contextually relevant results. To facilitate a deeper understanding of this promising direction, this survey explores multimodal composite editing and retrieval in depth, covering image-text composite editing, image-text composite retrieval, and other multimodal composite retrieval. In this survey, we systematically organize the application scenarios, methods, benchmarks, experiments, and future directions. Multimodal learning is a hot topic in large model era, and have also witnessed some surveys in multimodal learning and vision-language models with transformers published in the PAMI journal. To the best of our knowledge, this survey is the first comprehensive review of the literature on multimodal composite retrieval, which is a timely complement of multimodal fusion to existing reviews. To help readers' quickly track this field, we build the project page for this survey, which can be found at https://github.com/fuxianghuang1/Multimodal-Composite-Editing-and-Retrieval.

Paperid: 1812, https://arxiv.org/pdf/2409.05076.pdf GitHub

Abstract:
Large Vision-Language Models (LVLMs) have demonstrated their powerful multimodal capabilities. However, they also face serious safety problems, as adversaries can induce robustness issues in LVLMs through the use of well-designed adversarial examples. Therefore, LVLMs are in urgent need of detection tools for adversarial examples to prevent incorrect responses. In this work, we first discover that LVLMs exhibit regular attention patterns for clean images when presented with probe questions. We propose an unconventional method named PIP, which utilizes the attention patterns of one randomly selected irrelevant probe question (e.g., "Is there a clock?") to distinguish adversarial examples from clean examples. Regardless of the image to be tested and its corresponding question, PIP only needs to perform one additional inference of the image to be tested and the probe question, and then achieves successful detection of adversarial examples. Even under black-box attacks and open dataset scenarios, our PIP, coupled with a simple SVM, still achieves more than 98% recall and a precision of over 90%. Our PIP is the first attempt to detect adversarial attacks on LVLMs via simple irrelevant probe questions, shedding light on deeper understanding and introspection within LVLMs. The code is available at https://github.com/btzyd/pip.

Paperid: 1813, https://arxiv.org/pdf/2409.04796.pdf GitHub GitHub

Abstract:
Out-of-Distribution (OOD) detection, aiming to distinguish outliers from known categories, has gained prominence in practical scenarios. Recently, the advent of vision-language models (VLM) has heightened interest in enhancing OOD detection for VLM through few-shot tuning. However, existing methods mainly focus on optimizing global prompts, ignoring refined utilization of local information with regard to outliers. Motivated by this, we freeze global prompts and introduce Local-Prompt, a novel coarse-to-fine tuning paradigm to emphasize regional enhancement with local prompts. Our method comprises two integral components: global prompt guided negative augmentation and local prompt enhanced regional regularization. The former utilizes frozen, coarse global prompts as guiding cues to incorporate negative augmentation, thereby leveraging local outlier knowledge. The latter employs trainable local prompts and a regional regularization to capture local information effectively, aiding in outlier identification. We also propose regional-related metric to empower the enrichment of OOD detection. Moreover, since our approach explores enhancing local prompts only, it can be seamlessly integrated with trained global prompts during inference to boost the performance. Comprehensive experiments demonstrate the effectiveness and potential of our method. Notably, our method reduces average FPR95 by 5.17% against state-of-the-art method in 4-shot tuning on challenging ImageNet-1k dataset, even outperforming 16-shot results of previous methods. Code is released at https://github.com/AuroraZengfh/Local-Prompt.

Paperid: 1814, https://arxiv.org/pdf/2409.03868.pdf GitHub GitHub

Abstract:
Integrating image and text data through multi-modal learning has emerged as a new approach in medical imaging research, following its successful deployment in computer vision. While considerable efforts have been dedicated to establishing medical foundation models and their zero-shot transfer to downstream tasks, the popular few-shot setting remains relatively unexplored. Following on from the currently strong emergence of this setting in computer vision, we introduce the first structured benchmark for adapting medical vision-language models (VLMs) in a strict few-shot regime and investigate various adaptation strategies commonly used in the context of natural images. Furthermore, we evaluate a simple generalization of the linear-probe adaptation baseline, which seeks an optimal blending of the visual prototypes and text embeddings via learnable class-wise multipliers. Surprisingly, such a text-informed linear probe yields competitive performances in comparison to convoluted prompt-learning and adapter-based strategies, while running considerably faster and accommodating the black-box setting. Our extensive experiments span three different medical modalities and specialized foundation models, nine downstream tasks, and several state-of-the-art few-shot adaptation methods. We made our benchmark and code publicly available to trigger further developments in this emergent subject: \url{https://github.com/FereshteShakeri/few-shot-MedVLMs}.

Paperid: 1815, https://arxiv.org/pdf/2409.03757.pdf GitHub GitHub

Abstract:
Complex 3D scene understanding has gained increasing attention, with scene encoding strategies playing a crucial role in this success. However, the optimal scene encoding strategies for various scenarios remain unclear, particularly compared to their image-based counterparts. To address this issue, we present a comprehensive study that probes various visual encoding models for 3D scene understanding, identifying the strengths and limitations of each model across different scenarios. Our evaluation spans seven vision foundation encoders, including image-based, video-based, and 3D foundation models. We evaluate these models in four tasks: Vision-Language Scene Reasoning, Visual Grounding, Segmentation, and Registration, each focusing on different aspects of scene understanding. Our evaluations yield key findings: DINOv2 demonstrates superior performance, video models excel in object-level tasks, diffusion models benefit geometric tasks, and language-pretrained models show unexpected limitations in language-related tasks. These insights challenge some conventional understandings, provide novel perspectives on leveraging visual foundation models, and highlight the need for more flexible encoder selection in future vision-language and scene-understanding tasks. Code: https://github.com/YunzeMan/Lexicon3D

Paperid: 1816, https://arxiv.org/pdf/2409.03583.pdf GitHub GitHub

Abstract:
In many real-world applications, the frequency distribution of class labels for training data can exhibit a long-tailed distribution, which challenges traditional approaches of training deep neural networks that require heavy amounts of balanced data. Gathering and labeling data to balance out the class label distribution can be both costly and time-consuming. Many existing solutions that enable ensemble learning, re-balancing strategies, or fine-tuning applied to deep neural networks are limited by the inert problem of few class samples across a subset of classes. Recently, vision-language models like CLIP have been observed as effective solutions to zero-shot or few-shot learning by grasping a similarity between vision and language features for image and text pairs. Considering that large pre-trained vision-language models may contain valuable side textual information for minor classes, we propose to leverage text supervision to tackle the challenge of long-tailed learning. Concretely, we propose a novel text-guided mixup technique that takes advantage of the semantic relations between classes recognized by the pre-trained text encoder to help alleviate the long-tailed problem. Our empirical study on benchmark long-tailed tasks demonstrates the effectiveness of our proposal with a theoretical guarantee. Our code is available at https://github.com/rsamf/text-guided-mixup.

Paperid: 1817, https://arxiv.org/pdf/2409.03525.pdf GitHub

Abstract:
Open-vocabulary segmentation poses significant challenges, as it requires segmenting and recognizing objects across an open set of categories in unconstrained environments. Building on the success of powerful vision-language (ViL) foundation models, such as CLIP, recent efforts sought to harness their zero-short capabilities to recognize unseen categories. Despite notable performance improvements, these models still encounter the critical issue of generating precise mask proposals for unseen categories and scenarios, resulting in inferior segmentation performance eventually. To address this challenge, we introduce a novel approach, FrozenSeg, designed to integrate spatial knowledge from a localization foundation model (e.g., SAM) and semantic knowledge extracted from a ViL model (e.g., CLIP), in a synergistic framework. Taking the ViL model's visual encoder as the feature backbone, we inject the space-aware feature into the learnable queries and CLIP features within the transformer decoder. In addition, we devise a mask proposal ensemble strategy for further improving the recall rate and mask quality. To fully exploit pre-trained knowledge while minimizing training overhead, we freeze both foundation models, focusing optimization efforts solely on a lightweight transformer decoder for mask proposal generation-the performance bottleneck. Extensive experiments demonstrate that FrozenSeg advances state-of-the-art results across various segmentation benchmarks, trained exclusively on COCO panoptic data, and tested in a zero-shot manner. Code is available at https://github.com/chenxi52/FrozenSeg.

Paperid: 1818, https://arxiv.org/pdf/2409.03109.pdf GitHub

Abstract:
We introduce FIDAVL: Fake Image Detection and Attribution using a Vision-Language Model. FIDAVL is a novel and efficient mul-titask approach inspired by the synergies between vision and language processing. Leveraging the benefits of zero-shot learning, FIDAVL exploits the complementarity between vision and language along with soft prompt-tuning strategy to detect fake images and accurately attribute them to their originating source models. We conducted extensive experiments on a comprehensive dataset comprising synthetic images generated by various state-of-the-art models. Our results demonstrate that FIDAVL achieves an encouraging average detection accuracy of 95.42% and F1-score of 95.47% while also obtaining noteworthy performance metrics, with an average F1-score of 92.64% and ROUGE-L score of 96.50% for attributing synthetic images to their respective source generation models. The source code of this work will be publicly released at https://github.com/Mamadou-Keita/FIDAVL.

Paperid: 1819, https://arxiv.org/pdf/2409.02914.pdf GitHub

Abstract:
Large Vision-Language Models (LVLMs) have recently garnered significant attention, with many efforts aimed at harnessing their general knowledge to enhance the interpretability and robustness of autonomous driving models. However, LVLMs typically rely on large, general-purpose datasets and lack the specialized expertise required for professional and safe driving. Existing vision-language driving datasets focus primarily on scene understanding and decision-making, without providing explicit guidance on traffic rules and driving skills, which are critical aspects directly related to driving safety. To bridge this gap, we propose IDKB, a large-scale dataset containing over one million data items collected from various countries, including driving handbooks, theory test data, and simulated road test data. Much like the process of obtaining a driver's license, IDKB encompasses nearly all the explicit knowledge needed for driving from theory to practice. In particular, we conducted comprehensive tests on 15 LVLMs using IDKB to assess their reliability in the context of autonomous driving and provided extensive analysis. We also fine-tuned popular models, achieving notable performance improvements, which further validate the significance of our dataset. The project page can be found at: \url{https://4dvlab.github.io/project_page/idkb.html}

Paperid: 1820, https://arxiv.org/pdf/2409.02882.pdf GitHub

Abstract:
Few-shot image classifiers are designed to recognize and classify new data with minimal supervision and limited data but often show reliance on spurious correlations between classes and spurious attributes, known as spurious bias. Spurious correlations commonly hold in certain samples and few-shot classifiers can suffer from spurious bias induced from them. There is an absence of an automatic benchmarking system to assess the robustness of few-shot classifiers against spurious bias. In this paper, we propose a systematic and rigorous benchmark framework, termed FewSTAB, to fairly demonstrate and quantify varied degrees of robustness of few-shot classifiers to spurious bias. FewSTAB creates few-shot evaluation tasks with biased attributes so that using them for predictions can demonstrate poor performance. To construct these tasks, we propose attribute-based sample selection strategies based on a pre-trained vision-language model, eliminating the need for manual dataset curation. This allows FewSTAB to automatically benchmark spurious bias using any existing test data. FewSTAB offers evaluation results in a new dimension along with a new design guideline for building robust classifiers. Moreover, it can benchmark spurious bias in varied degrees and enable designs for varied degrees of robustness. Its effectiveness is demonstrated through experiments on ten few-shot learning methods across three datasets. We hope our framework can inspire new designs of robust few-shot classifiers. Our code is available at https://github.com/gtzheng/FewSTAB.

Paperid: 1821, https://arxiv.org/pdf/2409.02729.pdf GitHub

Abstract:
In medical image classification, supervised learning is challenging due to the scarcity of labeled medical images. To address this, we leverage the visual-textual alignment within Vision-Language Models (VLMs) to enable unsupervised learning of a medical image classifier. In this work, we propose \underline{Med}ical \underline{Un}supervised \underline{A}daptation (\texttt{MedUnA}) of VLMs, where the LLM-generated descriptions for each class are encoded into text embeddings and matched with class labels via a cross-modal adapter. This adapter attaches to a visual encoder of \texttt{MedCLIP} and aligns the visual embeddings through unsupervised learning, driven by a contrastive entropy-based loss and prompt tuning. Thereby, improving performance in scenarios where textual information is more abundant than labeled images, particularly in the healthcare domain. Unlike traditional VLMs, \texttt{MedUnA} uses \textbf{unpaired images and text} for learning representations and enhances the potential of VLMs beyond traditional constraints. We evaluate the performance on three chest X-ray datasets and two multi-class datasets (diabetic retinopathy and skin lesions), showing significant accuracy gains over the zero-shot baseline. Our code is available at https://github.com/rumaima/meduna.

Paperid: 1822, https://arxiv.org/pdf/2409.02253.pdf GitHub

Abstract:
Large foundation models have revolutionized the field, yet challenges remain in optimizing multi-modal models for specialized visual tasks. We propose a novel, generalizable methodology to identify preferred image distributions for black-box Vision-Language Models (VLMs) by measuring output consistency across varied input prompts. Applying this to different rendering types of 3D objects, we demonstrate its efficacy across various domains requiring precise interpretation of complex structures, with a focus on Computer-Aided Design (CAD) as an exemplar field. We further refine VLM outputs using in-context learning with human feedback, significantly enhancing explanation quality. To address the lack of benchmarks in specialized domains, we introduce CAD-VQA, a new dataset for evaluating VLMs on CAD-related visual question answering tasks. Our evaluation of state-of-the-art VLMs on CAD-VQA establishes baseline performance levels, providing a framework for advancing VLM capabilities in complex visual reasoning tasks across various fields requiring expert-level visual interpretation. We release the dataset and evaluation codes at \url{https://github.com/asgsaeid/cad_vqa}.

Paperid: 1823, https://arxiv.org/pdf/2409.01883.pdf GitHub

Abstract:
The development of vision-language models (VLMs) for histo-pathology has shown promising new usages and zero-shot performances. However, current approaches, which decompose large slides into smaller patches, focus solely on inductive classification, i.e., prediction for each patch is made independently of the other patches in the target test data. We extend the capability of these large models by introducing a transductive approach. By using text-based predictions and affinity relationships among patches, our approach leverages the strong zero-shot capabilities of these new VLMs without any additional labels. Our experiments cover four histopathology datasets and five different VLMs. Operating solely in the embedding space (i.e., in a black-box setting), our approach is highly efficient, processing $10^5$ patches in just a few seconds, and shows significant accuracy improvements over inductive zero-shot classification. Code available at https://github.com/FereshteShakeri/Histo-TransCLIP.

Paperid: 1824, https://arxiv.org/pdf/2409.01835.pdf GitHub

Abstract:
Although foundational vision-language models (VLMs) have proven to be very successful for various semantic discrimination tasks, they still struggle to perform faithfully for fine-grained categorization. Moreover, foundational models trained on one domain do not generalize well on a different domain without fine-tuning. We attribute these to the limitations of the VLM's semantic representations and attempt to improve their fine-grained visual awareness using generative modeling. Specifically, we propose two novel methods: Generative Class Prompt Learning (GCPL) and Contrastive Multi-class Prompt Learning (CoMPLe). Utilizing text-to-image diffusion models, GCPL significantly improves the visio-linguistic synergy in class embeddings by conditioning on few-shot exemplars with learnable class prompts. CoMPLe builds on this foundation by introducing a contrastive learning component that encourages inter-class separation during the generative optimization process. Our empirical results demonstrate that such a generative class prompt learning approach substantially outperform existing methods, offering a better alternative to few shot image recognition challenges. The source code will be made available at: https://github.com/soumitri2001/GCPL.

Paperid: 1825, https://arxiv.org/pdf/2409.01821.pdf GitHub

Abstract:
Adapting pre-trained models to new tasks can exhibit varying effectiveness across datasets. Visual prompting, a state-of-the-art parameter-efficient transfer learning method, can significantly improve the performance of out-of-distribution tasks. On the other hand, linear probing, a standard transfer learning method, can sometimes become the best approach. We propose a log-likelihood ratio (LLR) approach to analyze the comparative benefits of visual prompting and linear probing. By employing the LLR score alongside resource-efficient visual prompts approximations, our cost-effective measure attains up to a 100-fold reduction in run time compared to full training, while achieving prediction accuracies up to 91%. The source code is available at https://github.com/IBM/VP-LLR.

Paperid: 1826, https://arxiv.org/pdf/2409.01652.pdf GitHub

Abstract:
Representing robotic manipulation tasks as constraints that associate the robot and the environment is a promising way to encode desired robot behaviors. However, it remains unclear how to formulate the constraints such that they are 1) versatile to diverse tasks, 2) free of manual labeling, and 3) optimizable by off-the-shelf solvers to produce robot actions in real-time. In this work, we introduce Relational Keypoint Constraints (ReKep), a visually-grounded representation for constraints in robotic manipulation. Specifically, ReKep is expressed as Python functions mapping a set of 3D keypoints in the environment to a numerical cost. We demonstrate that by representing a manipulation task as a sequence of Relational Keypoint Constraints, we can employ a hierarchical optimization procedure to solve for robot actions (represented by a sequence of end-effector poses in SE(3)) with a perception-action loop at a real-time frequency. Furthermore, in order to circumvent the need for manual specification of ReKep for each new task, we devise an automated procedure that leverages large vision models and vision-language models to produce ReKep from free-form language instructions and RGB-D observations. We present system implementations on a wheeled single-arm platform and a stationary dual-arm platform that can perform a large variety of manipulation tasks, featuring multi-stage, in-the-wild, bimanual, and reactive behaviors, all without task-specific data or environment models. Website at https://rekep-robot.github.io/.

Paperid: 1827, https://arxiv.org/pdf/2409.01389.pdf GitHub

Abstract:
How do vision-language (VL) transformer models ground verb phrases and do they integrate contextual and world knowledge in this process? We introduce the CV-Probes dataset, containing image-caption pairs involving verb phrases that require both social knowledge and visual context to interpret (e.g., "beg"), as well as pairs involving verb phrases that can be grounded based on information directly available in the image (e.g., "sit"). We show that VL models struggle to ground VPs that are strongly context-dependent. Further analysis using explainable AI techniques shows that such models may not pay sufficient attention to the verb token in the captions. Our results suggest a need for improved methodologies in VL model training and evaluation. The code and dataset will be available https://github.com/ivana-13/CV-Probes.

Paperid: 1828, https://arxiv.org/pdf/2409.01109.pdf GitHub

Abstract:
Out-of-Distribution (OOD) detection in computer vision is a crucial research area, with related benchmarks playing a vital role in assessing the generalizability of models and their applicability in real-world scenarios. However, existing OOD benchmarks in the literature suffer from two main limitations: (1) they often overlook semantic shift as a potential challenge, and (2) their scale is limited compared to the large datasets used to train modern models. To address these gaps, we introduce SOOD-ImageNet, a novel dataset comprising around 1.6M images across 56 classes, designed for common computer vision tasks such as image classification and semantic segmentation under OOD conditions, with a particular focus on the issue of semantic shift. We ensured the necessary scalability and quality by developing an innovative data engine that leverages the capabilities of modern vision-language models, complemented by accurate human checks. Through extensive training and evaluation of various models on SOOD-ImageNet, we showcase its potential to significantly advance OOD research in computer vision. The project page is available at https://github.com/bach05/SOODImageNet.git.

Paperid: 1829, https://arxiv.org/pdf/2409.00698.pdf GitHub

Abstract:
Vision-Language Models for remote sensing have shown promising uses thanks to their extensive pretraining. However, their conventional usage in zero-shot scene classification methods still involves dividing large images into patches and making independent predictions, i.e., inductive inference, thereby limiting their effectiveness by ignoring valuable contextual information. Our approach tackles this issue by utilizing initial predictions based on text prompting and patch affinity relationships from the image encoder to enhance zero-shot capabilities through transductive inference, all without the need for supervision and at a minor computational cost. Experiments on 10 remote sensing datasets with state-of-the-art Vision-Language Models demonstrate significant accuracy improvements over inductive zero-shot classification. Our source code is publicly available on Github: https://github.com/elkhouryk/RS-TransCLIP

Paperid: 1830, https://arxiv.org/pdf/2409.00556.pdf GitHub

Abstract:
Automatic image anomaly detection is important for quality inspection in the manufacturing industry. The usual unsupervised anomaly detection approach is to train a model for each object class using a dataset of normal samples. However, a more realistic problem is zero-/few-shot anomaly detection where zero or only a few normal samples are available. This makes the training of object-specific models challenging. Recently, large foundation vision-language models have shown strong zero-shot performance in various downstream tasks. While these models have learned complex relationships between vision and language, they are not specifically designed for the tasks of anomaly detection. In this paper, we propose the Few-shot/zero-shot Anomaly Detection Engine (FADE) which leverages the vision-language CLIP model and adjusts it for the purpose of industrial anomaly detection. Specifically, we improve language-guided anomaly segmentation 1) by adapting CLIP to extract multi-scale image patch embeddings that are better aligned with language and 2) by automatically generating an ensemble of text prompts related to industrial anomaly detection. 3) We use additional vision-based guidance from the query and reference images to further improve both zero-shot and few-shot anomaly detection. On the MVTec-AD (and VisA) dataset, FADE outperforms other state-of-the-art methods in anomaly segmentation with pixel-AUROC of 89.6% (91.5%) in zero-shot and 95.4% (97.5%) in 1-normal-shot. Code is available at https://github.com/BMVC-FADE/BMVC-FADE.

Paperid: 1831, https://arxiv.org/pdf/2409.00397.pdf GitHub

Abstract:
Multi-Target Domain Adaptation (MTDA) entails learning domain-invariant information from a single source domain and applying it to multiple unlabeled target domains. Yet, existing MTDA methods predominantly focus on addressing domain shifts within visual features, often overlooking semantic features and struggling to handle unknown classes, resulting in what is known as Open-Set (OS) MTDA. While large-scale vision-language foundation models like CLIP show promise, their potential for MTDA remains largely unexplored. This paper introduces COSMo, a novel method that learns domain-agnostic prompts through source domain-guided prompt learning to tackle the MTDA problem in the prompt space. By leveraging a domain-specific bias network and separate prompts for known and unknown classes, COSMo effectively adapts across domain and class shifts. To the best of our knowledge, COSMo is the first method to address Open-Set Multi-Target DA (OSMTDA), offering a more realistic representation of real-world scenarios and addressing the challenges of both open-set and multi-target DA. COSMo demonstrates an average improvement of $5.1\%$ across three challenging datasets: Mini-DomainNet, Office-31, and Office-Home, compared to other related DA methods adapted to operate within the OSMTDA setting. Code is available at: https://github.com/munish30monga/COSMo

Paperid: 1832, https://arxiv.org/pdf/2409.00147.pdf GitHub

Abstract:
The rapid development of large language models (LLMs) has spurred extensive research into their domain-specific capabilities, particularly mathematical reasoning. However, most open-source LLMs focus solely on mathematical reasoning, neglecting the integration with visual injection, despite the fact that many mathematical tasks rely on visual inputs such as geometric diagrams, charts, and function plots. To fill this gap, we introduce \textbf{MultiMath-7B}, a multimodal large language model that bridges the gap between math and vision. \textbf{MultiMath-7B} is trained through a four-stage process, focusing on vision-language alignment, visual and math instruction-tuning, and process-supervised reinforcement learning. We also construct a novel, diverse and comprehensive multimodal mathematical dataset, \textbf{MultiMath-300K}, which spans K-12 levels with image captions and step-wise solutions. MultiMath-7B achieves state-of-the-art (SOTA) performance among open-source models on existing multimodal mathematical benchmarks and also excels on text-only mathematical benchmarks. Our model and dataset are available at {\textcolor{blue}{\url{https://github.com/pengshuai-rin/MultiMath}}}.

Paperid: 1833, https://arxiv.org/pdf/2409.00084.pdf GitHub

Abstract:
Background and Aims: This study evaluates the medical reasoning performance of large language models (LLMs) and vision language models (VLMs) in gastroenterology. Methods: We used 300 gastroenterology board exam-style multiple-choice questions, 138 of which contain images to systematically assess the impact of model configurations and parameters and prompt engineering strategies utilizing GPT-3.5. Next, we assessed the performance of proprietary and open-source LLMs (versions), including GPT (3.5, 4, 4o, 4omini), Claude (3, 3.5), Gemini (1.0), Mistral, Llama (2, 3, 3.1), Mixtral, and Phi (3), across different interfaces (web and API), computing environments (cloud and local), and model precisions (with and without quantization). Finally, we assessed accuracy using a semiautomated pipeline. Results: Among the proprietary models, GPT-4o (73.7%) and Claude3.5-Sonnet (74.0%) achieved the highest accuracy, outperforming the top open-source models: Llama3.1-405b (64%), Llama3.1-70b (58.3%), and Mixtral-8x7b (54.3%). Among the quantized open-source models, the 6-bit quantized Phi3-14b (48.7%) performed best. The scores of the quantized models were comparable to those of the full-precision models Llama2-7b, Llama2--13b, and Gemma2-9b. Notably, VLM performance on image-containing questions did not improve when the images were provided and worsened when LLM-generated captions were provided. In contrast, a 10% increase in accuracy was observed when images were accompanied by human-crafted image descriptions. Conclusion: In conclusion, while LLMs exhibit robust zero-shot performance in medical reasoning, the integration of visual data remains a challenge for VLMs. Effective deployment involves carefully determining optimal model configurations, encouraging users to consider either the high performance of proprietary models or the flexible adaptability of open-source models.

Paperid: 1834, https://arxiv.org/pdf/2511.21150.pdf

Abstract:
Visual encoding followed by token condensing has become the standard architectural paradigm in multi-modal large language models (MLLMs). Many recent MLLMs increasingly favor global native- resolution visual encoding over slice-based methods. To investigate this trend, we systematically compare their behavior on vision-language understanding and attention patterns, revealing that global encoding enhances overall capability but at the expense of greater computational overhead. To address this issue, we present LLaVA-UHD v3, an MLLM centered upon our proposed Progressive Visual Compression (PVC) method, which can be seamlessly integrated into standard Vision Transformer (ViT) to enable efficient native-resolution encoding. The PVC approach consists of two key modules: (i) refined patch embedding, which supports flexible patch-size scaling for fine-grained visual model- ing, (ii) windowed token compression, hierarchically deployed across ViT layers to progressively aggregate local token representations. Jointly modulated by these two modules, a widely pretrained ViT can be reconfigured into an efficient architecture while largely preserving generality. Evaluated across extensive benchmarks, the transformed ViT, termed ViT-UHD, demonstrates competitive performance with MoonViT while reducing TTFT (time-to-first-token) by 2.4x, when developed within an identical MLLM architecture. Building upon ViT-UHD, LLaVA-UHD v3 also achieves competitive performance to Qwen2-VL, while further reducing TTFT by 1.9x. We will release all code and checkpoints to support future research on efficient MLLMs.

Paperid: 1835, https://arxiv.org/pdf/2510.09733.pdf

Abstract:
Visual retrieval-augmented generation (VRAG) augments vision-language models (VLMs) with external visual knowledge to ground reasoning and reduce hallucinations. Yet current VRAG systems often fail to reliably perceive and integrate evidence across multiple images, leading to weak grounding and erroneous conclusions. In this paper, we propose EVisRAG, an end-to-end framework that learns to reason with evidence-guided multi-image to address this issue. The model first observes retrieved images and records per-image evidence, then derives the final answer from the aggregated evidence. To train EVisRAG effectively, we introduce Reward-Scoped Group Relative Policy Optimization (RS-GRPO), which binds fine-grained rewards to scope-specific tokens to jointly optimize visual perception and reasoning abilities of VLMs. Experimental results on multiple visual question answering benchmarks demonstrate that EVisRAG delivers substantial end-to-end gains over backbone VLM with 27\% improvements on average. Further analysis shows that, powered by RS-GRPO, EVisRAG improves answer accuracy by precisely perceiving and localizing question-relevant evidence across multiple images and deriving the final answer from that evidence, much like a real detective.

Paperid: 1836, https://arxiv.org/pdf/2412.13871.pdf

Abstract:
Vision transformers (ViTs) are widely employed in multimodal large language models (MLLMs) for visual encoding. However, they exhibit inferior performance on tasks regarding fine-grained visual perception. We attribute this to the limitations of ViTs in capturing diverse multi-modal visual levels, such as low-level details. To address this issue, we present LLaVA-UHD v2, an MLLM with advanced perception abilities by introducing a well-designed vision-language projector, the Hierarchical window (Hiwin) transformer. Hiwin transformer enhances MLLM's ability to capture diverse multi-modal visual granularities, by incorporating our constructed high-resolution semantic pyramid. Specifically, Hiwin transformer comprises two key modules: (i) a visual detail injection module, which progressively injects low-level visual details into high-level language-aligned semantics features, thereby forming an inverse semantic pyramid (ISP), and (ii) a hierarchical window attention module, which leverages cross-scale windows to condense multi-level semantics from the ISP. Extensive experiments show that LLaVA-UHD v2 outperforms compared MLLMs on a wide range of benchmarks. Notably, our design achieves an average boost of 3.7% across 14 benchmarks compared with the baseline method, 9.3% on DocVQA for instance. All the data and code will be publicly available to facilitate future research.

Paperid: 1837, https://arxiv.org/pdf/2510.26474.pdf

Abstract:
Self-improvement has emerged as a mainstream paradigm for advancing the reasoning capabilities of large vision-language models (LVLMs), where models explore and learn from successful trajectories iteratively. However, we identify a critical issue during this process: the model excels at generating high-quality trajectories for simple queries (i.e., head data) but struggles with more complex ones (i.e., tail data). This leads to an imbalanced optimization that drives the model to prioritize simple reasoning skills, while hindering its ability to tackle more complex reasoning tasks. Over iterations, this imbalance becomes increasingly pronounced--a dynamic we term the "Matthew effect"--which ultimately hinders further model improvement and leads to performance bottlenecks. To counteract this challenge, we introduce four efficient strategies from two perspectives: distribution-reshaping and trajectory-resampling, to achieve head-tail re-balancing during the exploration-and-learning self-improvement process. Extensive experiments on Qwen2-VL-7B-Instruct and InternVL2.5-4B models across visual reasoning tasks demonstrate that our methods consistently improve visual reasoning capabilities, outperforming vanilla self-improvement by 3.86 points on average.

Paperid: 1838, https://arxiv.org/pdf/2505.13886.pdf

Abstract:
Vision-language reinforcement learning (RL) has primarily focused on narrow domains (e.g. geometry or chart reasoning). This leaves broader training scenarios and resources underexplored, limiting the exploration and learning of Vision Language Models (VLMs) through RL. We find video games inherently provide rich visual elements and mechanics that are easy to verify. To fully use the multimodal and verifiable reward in video games, we propose Game-RL, constructing diverse game tasks for RL training to boost VLMs general reasoning ability. To obtain training data, we propose Code2Logic, a novel approach that adapts game code to synthesize game reasoning task data, thus obtaining the GameQA dataset of 30 games and 158 tasks with controllable difficulty gradation. Unexpectedly, RL training solely on GameQA enables multiple VLMs to achieve performance improvements across 7 diverse vision-language benchmarks, demonstrating the value of Game-RL for enhancing VLMs' general reasoning. Furthermore, this suggests that video games may serve as valuable scenarios and resources to boost general reasoning abilities. Our code, dataset and models are available at the GitHub repository.

Paperid: 1839, https://arxiv.org/pdf/2511.17962.pdf

Abstract:
Developing a robust visual quality assessment (VQualA) large multi-modal model (LMM) requires achieving versatility, powerfulness, and transferability. However, existing VQualA LMMs typically focus on a single task and rely on full-parameter fine-tuning, which makes them prone to overfitting on specific modalities or task types, thereby limiting their generalization capacity and transferability. To address this, we propose a vision-encoder-centered generative pre-training pipeline and develop the VITAL-Series LMMs. (1) We adopt a machine-executed annotation-scrutiny paradigm, constructing over 4.5M vision-language (VL) pairs-the largest VQualA training dataset to date. (2) We employ a multi-task training workflow that simultaneously enhances the model's quantitative scoring precision and strengthens its capability for quality interpretation across both image and video modalities. (3) Building upon the vision encoder, we realize an efficient model zoo extension: the model zoo exhibits strong zero-shot performance, and each paired decoder requires only a swift warm-up using less than 1/1000 of the pre-training data to achieve performance comparable to the fully trained counterpart. Overall, our work lays a cornerstone for advancing toward the foundation LMM for VQualA.

Paperid: 1840, https://arxiv.org/pdf/2501.00848.pdf

Abstract:
Current Visual Language Models (VLMs) show impressive image understanding but struggle with visual illusions, especially in real-world scenarios. Existing benchmarks focus on classical cognitive illusions, which have been learned by state-of-the-art (SOTA) VLMs, revealing issues such as hallucinations and limited perceptual abilities. To address this gap, we introduce IllusionBench, a comprehensive visual illusion dataset that encompasses not only classic cognitive illusions but also real-world scene illusions. This dataset features 1,051 images, 5,548 question-answer pairs, and 1,051 golden text descriptions that address the presence, causes, and content of the illusions. We evaluate ten SOTA VLMs on this dataset using true-or-false, multiple-choice, and open-ended tasks. In addition to real-world illusions, we design trap illusions that resemble classical patterns but differ in reality, highlighting hallucination issues in SOTA models. The top-performing model, GPT-4o, achieves 80.59% accuracy on true-or-false tasks and 76.75% on multiple-choice questions, but still lags behind human performance. In the semantic description task, GPT-4o's hallucinations on classical illusions result in low scores for trap illusions, even falling behind some open-source models. IllusionBench is, to the best of our knowledge, the largest and most comprehensive benchmark for visual illusions in VLMs to date.

Paperid: 1841, https://arxiv.org/pdf/2508.15904.pdf

Abstract:
Rare cancers comprise 20-25% of all malignancies but face major diagnostic challenges due to limited expert availability-especially in pediatric oncology, where they represent over 70% of cases. While pathology vision-language (VL) foundation models show promising zero-shot capabilities for common cancer subtyping, their clinical performance for rare cancers remains limited. Existing multi-instance learning (MIL) methods rely only on visual features, overlooking cross-modal knowledge and compromising interpretability critical for rare cancer diagnosis. To address this limitation, we propose PathPT, a novel framework that fully exploits the potential of vision-language pathology foundation models through spatially-aware visual aggregation and task-specific prompt tuning. Unlike conventional MIL, PathPT converts WSI-level supervision into fine-grained tile-level guidance by leveraging the zero-shot capabilities of VL models, thereby preserving localization on cancerous regions and enabling cross-modal reasoning through prompts aligned with histopathological semantics. We benchmark PathPT on eight rare cancer datasets(four adult and four pediatric) spanning 56 subtypes and 2,910 WSIs, as well as three common cancer datasets, evaluating four state-of-the-art VL models and four MIL frameworks under three few-shot settings. Results show that PathPT consistently delivers superior performance, achieving substantial gains in subtyping accuracy and cancerous region grounding ability. This work advances AI-assisted diagnosis for rare cancers, offering a scalable solution for improving subtyping accuracy in settings with limited access to specialized expertise.

Paperid: 1842, https://arxiv.org/pdf/2503.13938.pdf

Abstract:
Traffic scene understanding is essential for intelligent transportation systems and autonomous driving, ensuring safe and efficient vehicle operation. While recent advancements in VLMs have shown promise for holistic scene understanding, the application of VLMs to traffic scenarios, particularly using BEV maps, remains under explored. Existing methods often suffer from limited task design and narrow data amount, hindering comprehensive scene understanding. To address these challenges, we introduce ChatBEV-QA, a novel BEV VQA benchmark contains over 137k questions, designed to encompass a wide range of scene understanding tasks, including global scene understanding, vehicle-lane interactions, and vehicle-vehicle interactions. This benchmark is constructed using an novel data collection pipeline that generates scalable and informative VQA data for BEV maps. We further fine-tune a specialized vision-language model ChatBEV, enabling it to interpret diverse question prompts and extract relevant context-aware information from BEV maps. Additionally, we propose a language-driven traffic scene generation pipeline, where ChatBEV facilitates map understanding and text-aligned navigation guidance, significantly enhancing the generation of realistic and consistent traffic scenarios. The dataset, code and the fine-tuned model will be released.

Paperid: 1843, https://arxiv.org/pdf/2412.13126.pdf

Abstract:
Deep learning has enabled the development of highly robust foundation models for various pathological tasks across diverse diseases and patient cohorts. Among these models, vision-language pre-training, which leverages large-scale paired data to align pathology image and text embedding spaces, and provides a novel zero-shot paradigm for downstream tasks. However, existing models have been primarily data-driven and lack the incorporation of domain-specific knowledge, which limits their performance in cancer diagnosis, especially for rare tumor subtypes. To address this limitation, we establish a Knowledge-enhanced Pathology (KEEP) foundation model that harnesses disease knowledge to facilitate vision-language pre-training. Specifically, we first construct a disease knowledge graph (KG) that covers 11,454 human diseases with 139,143 disease attributes, including synonyms, definitions, and hypernym relations. We then systematically reorganize the millions of publicly available noisy pathology image-text pairs, into 143K well-structured semantic groups linked through the hierarchical relations of the disease KG. To derive more nuanced image and text representations, we propose a novel knowledge-enhanced vision-language pre-training approach that integrates disease knowledge into the alignment within hierarchical semantic groups instead of unstructured image-text pairs. Validated on 18 diverse benchmarks with more than 14,000 whole slide images (WSIs), KEEP achieves state-of-the-art performance in zero-shot cancer diagnostic tasks. Notably, for cancer detection, KEEP demonstrates an average sensitivity of 89.8% at a specificity of 95.0% across 7 cancer types. For cancer subtyping, KEEP achieves a median balanced accuracy of 0.456 in subtyping 30 rare brain cancers, indicating strong generalizability for diagnosing rare tumors.

Paperid: 1844, https://arxiv.org/pdf/2411.04923.pdf

Abstract:
Fine-grained alignment between videos and text is challenging due to complex spatial and temporal dynamics in videos. Existing video-based Large Multimodal Models (LMMs) handle basic conversations but struggle with precise pixel-level grounding in videos. To address this, we introduce VideoGLaMM, a LMM designed for fine-grained pixel-level grounding in videos based on user-provided textual inputs. Our design seamlessly connects three key components: a Large Language Model, a dual vision encoder that emphasizes both spatial and temporal details, and a spatio-temporal decoder for accurate mask generation. This connection is facilitated via tunable V-L and L-V adapters that enable close Vision-Language (VL) alignment. The architecture is trained to synchronize both spatial and temporal elements of video content with textual instructions. To enable fine-grained grounding, we curate a multimodal dataset featuring detailed visually-grounded conversations using a semiautomatic annotation pipeline, resulting in a diverse set of 38k video-QA triplets along with 83k objects and 671k masks. We evaluate VideoGLaMM on three challenging tasks: Grounded Conversation Generation, Visual Grounding, and Referring Video Segmentation. Experimental results show that our model consistently outperforms existing approaches across all three tasks.

Paperid: 1845, https://arxiv.org/pdf/2512.11612.pdf

Abstract:
Image Compression for Machines (ICM) has emerged as a pivotal research direction in the field of visual data compression. However, with the rapid evolution of machine intelligence, the target of compression has shifted from task-specific virtual models to Embodied agents operating in real-world environments. To address the communication constraints of Embodied AI in multi-agent systems and ensure real-time task execution, this paper introduces, for the first time, the scientific problem of Embodied Image Compression. We establish a standardized benchmark, EmbodiedComp, to facilitate systematic evaluation under ultra-low bitrate conditions in a closed-loop setting. Through extensive empirical studies in both simulated and real-world settings, we demonstrate that existing Vision-Language-Action models (VLAs) fail to reliably perform even simple manipulation tasks when compressed below the Embodied bitrate threshold. We anticipate that EmbodiedComp will catalyze the development of domain-specific compression tailored for Embodied agents , thereby accelerating the Embodied AI deployment in the Real-world.

Paperid: 1846, https://arxiv.org/pdf/2508.06553.pdf

Abstract:
Embodied intelligence is advancing rapidly, driving the need for efficient evaluation. Current benchmarks typically rely on interactive simulated environments or real-world setups, which are costly, fragmented, and hard to scale. To address this, we introduce StaticEmbodiedBench, a plug-and-play benchmark that enables unified evaluation using static scene representations. Covering 42 diverse scenarios and 8 core dimensions, it supports scalable and comprehensive assessment through a simple interface. Furthermore, we evaluate 19 Vision-Language Models (VLMs) and 11 Vision-Language-Action models (VLAs), establishing the first unified static leaderboard for Embodied intelligence. Moreover, we release a subset of 200 samples from our benchmark to accelerate the development of embodied intelligence.

Paperid: 1847, https://arxiv.org/pdf/2506.00893.pdf

Abstract:
Affordance theory suggests that environments inherently provide action possibilities shaping perception and behavior. While Multimodal Large Language Models (MLLMs) achieve strong performance in vision-language tasks, their ability to perceive affordance, which is crucial for intuitive and safe interactions, remains underexplored. To address this, we introduce **A4Bench**, a novel benchmark designed to evaluate the affordance perception abilities of MLLMs across two dimensions: 1) Constitutive Affordance, assessing understanding of inherent object properties through 1,282 questionanswer pairs spanning nine sub-disciplines, and 2) Transformative Affordance, probing dynamic and contextual nuances (e.g., misleading, time-dependent, cultural, or individual-specific affordance) with 718 challenging question-answer pairs. We evaluate 17 MLLMs (nine proprietary and eight open-source) and compare them to human performance. Results show that proprietary models generally outperform open-source ones, yet all models perform far below humans, especially in transformative affordance. Furthermore, even top-performing models, such as Gemini-2.0-Pro (18.05% overall exact match accuracy), significantly lag behind human performance (best: 85.34%, worst: 81.25%). These findings highlight critical gaps in environmental understanding of MLLMs and provide a foundation for advancing AI systems toward more robust, context-aware interactions.

Paperid: 1848, https://arxiv.org/pdf/2511.22076.pdf

Abstract:
The Internet of Agents (IoA) is rapidly gaining prominence as a foundational architecture for interconnected intelligent systems, designed to facilitate seamless discovery, communication, and collaborative reasoning among a vast network of Artificial Intelligence (AI) agents. Powered by Large Language and Vision-Language Models, IoA enables the development of interactive, rational agents capable of complex cooperation, moving far beyond traditional isolated models. IoA involves physical entities, i.e., Wireless Agents (WAs) with limited onboard resources, which need to offload their compute-intensive agentic AI services to nearby servers. Such servers can be Mobile Agents (MAs), e.g., vehicle agents, or Fixed Agents (FAs), e.g., end-side units agents. Given their fixed geographical locations and stable connectivity, FAs can serve as reliable communication gateways and task aggregation points. This stability allows them to effectively coordinate with and offload to an Aerial Agent (AA) tier, which has an advantage not affordable for highly mobile MAs with dynamic connectivity limitations. As such, we propose a two-tier optimization approach. The first tier employs a multi-leader multi-follower Stackelberg game. In the game, MAs and FAs act as the leaders who set resource prices. WAs are the followers to determine task offloading ratios. However, when FAs become overloaded, they can further offload tasks to available aerial resources. Therefore, the second tier introduces a Double Dutch Auction model where overloaded FAs act as the buyers to request resources, and AAs serve as the sellers for resource provision. We then develop a diffusion-based Deep Reinforcement Learning algorithm to solve the model. Numerical results demonstrate the superiority of our proposed scheme in facilitating task offloading.

Paperid: 1849, https://arxiv.org/pdf/2510.21093.pdf

Abstract:
Recently, large models have shown significant potential for smart healthcare. However, the deployment of Large Vision-Language Models (LVLMs) for clinical services is currently hindered by three critical challenges: a tendency to hallucinate answers not grounded in visual evidence, the inefficiency of fixed-depth reasoning, and the difficulty of multi-institutional collaboration. To address these challenges, in this paper, we develop MedAlign, a novel framework to ensure visually accurate LVLM responses for Medical Visual Question Answering (Med-VQA). Specifically, we first propose a multimodal Direct Preference Optimization (mDPO) objective to explicitly align preference learning with visual context. We then design a Retrieval-Aware Mixture-of-Experts (RA-MoE) architecture that utilizes image and text similarity to route queries to a specialized and context-augmented LVLM (i.e., an expert), thereby mitigating hallucinations in LVLMs. To achieve adaptive reasoning and facilitate multi-institutional collaboration, we propose a federated governance mechanism, where the selected expert, fine-tuned on clinical datasets based on mDPO, locally performs iterative Chain-of-Thought (CoT) reasoning via the local meta-cognitive uncertainty estimator. Extensive experiments on three representative Med-VQA datasets demonstrate that MedAlign achieves state-of-the-art performance, outperforming strong retrieval-augmented baselines by up to $11.85\%$ in F1-score, and simultaneously reducing the average reasoning length by $51.60\%$ compared with fixed-depth CoT approaches.

Paperid: 1850, https://arxiv.org/pdf/2509.06461.pdf

Abstract:
Vision-Language Models (VLMs) have demonstrated remarkable success across diverse visual tasks, yet their performance degrades in complex visual environments. While existing enhancement approaches require additional training, rely on external segmentation tools, or operate at coarse-grained levels, they overlook the innate ability within VLMs. To bridge this gap, we investigate VLMs' attention patterns and discover that: (1) visual complexity strongly correlates with attention entropy, negatively impacting reasoning performance; (2) attention progressively refines from global scanning in shallow layers to focused convergence in deeper layers, with convergence degree determined by visual complexity. (3) Theoretically, we prove that the contrast of attention maps between general queries and task-specific queries enables the decomposition of visual signal into semantic signals and visual noise components. Building on these insights, we propose Contrastive Attention Refinement for Visual Enhancement (CARVE), a training-free method that extracts task-relevant visual signals through attention contrasting at the pixel level. Extensive experiments demonstrate that CARVE consistently enhances performance, achieving up to 75% improvement on open-source models. Our work provides critical insights into the interplay between visual complexity and attention mechanisms, offering an efficient pathway for improving visual reasoning with contrasting attention.

Paperid: 1851, https://arxiv.org/pdf/2505.07176.pdf

Abstract:
With the rapid proliferation of large language models and vision-language models, AI agents have evolved from isolated, task-specific systems into autonomous, interactive entities capable of perceiving, reasoning, and acting without human intervention. As these agents proliferate across virtual and physical environments, from virtual assistants to embodied robots, the need for a unified, agent-centric infrastructure becomes paramount. In this survey, we introduce the Internet of Agents (IoA) as a foundational framework that enables seamless interconnection, dynamic discovery, and collaborative orchestration among heterogeneous agents at scale. We begin by presenting a general IoA architecture, highlighting its hierarchical organization, distinguishing features relative to the traditional Internet, and emerging applications. Next, we analyze the key operational enablers of IoA, including capability notification and discovery, adaptive communication protocols, dynamic task matching, consensus and conflict-resolution mechanisms, and incentive models. Finally, we identify open research directions toward building resilient and trustworthy IoA ecosystems.

Paperid: 1852, https://arxiv.org/pdf/2504.14326.pdf

Abstract:
Mobile metaverses have attracted significant attention from both academia and industry, which are envisioned as the next-generation Internet, providing users with immersive and ubiquitous metaverse services through mobile devices. Driven by Large Language Models (LLMs) and Vision-Language Models (VLMs), Artificial Intelligence (AI) agents hold the potential to empower the creation, maintenance, and evolution of mobile metaverses. Currently, AI agents are primarily constructed using cloud-based LLMs and VLMs. However, several challenges hinder their effective implementation, including high service latency and potential sensitive data leakage during perception and processing. In this paper, we develop an edge-cloud collaboration-based federated AI agent construction framework in mobile metaverses. Specifically, Edge Servers (ESs), acting as agent infrastructures, collaboratively create agent modules in a distributed manner. The cloud server then integrates these modules into AI agents and deploys them at the edge, thereby enabling low-latency AI agent services for users. Considering that ESs may exhibit dynamic levels of willingness to participate in federated AI agent construction, we design a two-period dynamic contract model to continuously motivate ESs to participate in agent module creation, effectively addressing the dynamic information asymmetry between the cloud server and the ESs. Furthermore, we propose an Enhanced Diffusion Model-based Soft Actor-Critic (EDMSAC) algorithm to efficiently generate optimal dynamic contracts, in which dynamic structured pruning is applied to DM-based actor networks to enhance denoising efficiency and policy learning performance. Extensive simulations demonstrate the effectiveness and superiority of the EDMSAC algorithm and the proposed contract model.

Paperid: 1853, https://arxiv.org/pdf/2502.13464.pdf

Abstract:
Commonsense plausibility estimation is critical for evaluating language models (LMs), yet existing generative approaches--reliant on likelihoods or verbalized judgments--struggle with fine-grained discrimination. In this paper, we propose ComPaSS, a novel discriminative framework that quantifies commonsense plausibility by measuring semantic shifts when augmenting sentences with commonsense-related information. Plausible augmentations induce minimal shifts in semantics, while implausible ones result in substantial deviations. Evaluations on two types of fine-grained commonsense plausibility estimation tasks across different backbones, including LLMs and vision-language models (VLMs), show that ComPaSS consistently outperforms baselines. It demonstrates the advantage of discriminative approaches over generative methods in fine-grained commonsense plausibility evaluation. Experiments also show that (1) VLMs yield superior performance to LMs, when integrated with ComPaSS, on vision-grounded commonsense tasks. (2) contrastive pre-training sharpens backbone models' ability to capture semantic nuances, thereby further enhancing ComPaSS.

Paperid: 1854, https://arxiv.org/pdf/2601.10781.pdf

Abstract:
Future motion representations, such as optical flow, offer immense value for control and generative tasks. However, forecasting generalizable spatially dense motion representations remains a key challenge, and learning such forecasting from noisy, real-world data remains relatively unexplored. We introduce FOFPred, a novel language-conditioned optical flow forecasting model featuring a unified Vision-Language Model (VLM) and Diffusion architecture. This unique combination enables strong multimodal reasoning with pixel-level generative fidelity for future motion prediction. Our model is trained on web-scale human activity data-a highly scalable but unstructured source. To extract meaningful signals from this noisy video-caption data, we employ crucial data preprocessing techniques and our unified architecture with strong image pretraining. The resulting trained model is then extended to tackle two distinct downstream tasks in control and generation. Evaluations across robotic manipulation and video generation under language-driven settings establish the cross-domain versatility of FOFPred, confirming the value of a unified VLM-Diffusion architecture and scalable learning from diverse web data for future optical flow prediction.

Paperid: 1855, https://arxiv.org/pdf/2512.18007.pdf

Abstract:
Vision-Language-Action (VLA) models have achieved remarkable progress in robotic manipulation by mapping multimodal observations and instructions directly to actions. However, they typically mimic expert trajectories without predictive motion reasoning, which limits their ability to reason about what actions to take. To address this limitation, we propose joint learning with motion image diffusion, a novel strategy that enhances VLA models with motion reasoning capabilities. Our method extends the VLA architecture with a dual-head design: while the action head predicts action chunks as in vanilla VLAs, an additional motion head, implemented as a Diffusion Transformer (DiT), predicts optical-flow-based motion images that capture future dynamics. The two heads are trained jointly, enabling the shared VLM backbone to learn representations that couple robot control with motion knowledge. This joint learning builds temporally coherent and physically grounded representations without modifying the inference pathway of standard VLAs, thereby maintaining test-time latency. Experiments in both simulation and real-world environments demonstrate that joint learning with motion image diffusion improves the success rate of pi-series VLAs to 97.5% on the LIBERO benchmark and 58.0% on the RoboTwin benchmark, yielding a 23% improvement in real-world performance and validating its effectiveness in enhancing the motion reasoning capability of large-scale VLAs.

Paperid: 1856, https://arxiv.org/pdf/2412.05479.pdf

Abstract:
While open-source vision-language models perform well on simple question-answering, they still struggle with complex questions that require both perceptual and reasoning capabilities. We propose LATTE, a family of vision-language models that have LeArned to Think wiTh vision spEcialists. By offloading perception to state-of-the-art vision models, our approach enables vision-language models to focus solely on reasoning over high-quality perceptual information. To train LATTE, we synthesize and filter a large dataset of 293K multi-modal reasoning traces over perceptual outputs of vision specialists. LATTE trained on this data achieves significant 4-5% gains over baselines across 6 benchmarks covering both perception and reasoning abilities. Ablation studies reveal that the effectiveness of multi-modal reasoning traces depends on the data sources, formats, and quality of thoughts.

Paperid: 1857, https://arxiv.org/pdf/2411.07461.pdf

Abstract:
We introduce BLIP3-KALE, a dataset of 218 million image-text pairs that bridges the gap between descriptive synthetic captions and factual web-scale alt-text. KALE augments synthetic dense image captions with web-scale alt-text to generate factually grounded image captions. Our two-stage approach leverages large vision-language models and language models to create knowledge-augmented captions, which are then used to train a specialized VLM for scaling up the dataset. We train vision-language models on KALE and demonstrate improvements on vision-language tasks. Our experiments show the utility of KALE for training more capable and knowledgeable multimodal models. We release the KALE dataset at https://huggingface.co/datasets/Salesforce/blip3-kale

Paperid: 1858, https://arxiv.org/pdf/2509.20146.pdf

Abstract:
Recent benchmarks for medical Large Vision-Language Models (LVLMs) emphasize leaderboard accuracy, overlooking reliability and safety. We study sycophancy -- models' tendency to uncritically echo user-provided information -- in high-stakes clinical settings. We introduce EchoBench, a benchmark to systematically evaluate sycophancy in medical LVLMs. It contains 2,122 images across 18 departments and 20 modalities with 90 prompts that simulate biased inputs from patients, medical students, and physicians. We evaluate medical-specific, open-source, and proprietary LVLMs. All exhibit substantial sycophancy; the best proprietary model (Claude 3.7 Sonnet) still shows 45.98% sycophancy, and GPT-4.1 reaches 59.15%. Many medical-specific models exceed 95% sycophancy despite only moderate accuracy. Fine-grained analyses by bias type, department, perceptual granularity, and modality identify factors that increase susceptibility. We further show that higher data quality/diversity and stronger domain knowledge reduce sycophancy without harming unbiased accuracy. EchoBench also serves as a testbed for mitigation: simple prompt-level interventions (negative prompting, one-shot, few-shot) produce consistent reductions and motivate training- and decoding-time strategies. Our findings highlight the need for robust evaluation beyond accuracy and provide actionable guidance toward safer, more trustworthy medical LVLMs.

Paperid: 1859, https://arxiv.org/pdf/2505.19892.pdf

Abstract:
Foundation models update slowly due to resource-intensive training, whereas domain-specific models evolve rapidly between releases. Model merging seeks to combine multiple expert models into a single, more capable model, reducing storage and serving costs while supporting decentralized development. Despite its potential, previous studies have primarily focused on merging visual classification models or Large Language Models (LLMs) for code and math tasks. Recently, Multimodal LLMs (MLLMs) that extend LLMs through large-scale multimodal training have gained traction. However, there lacks a benchmark for model merging research that clearly divides the tasks for MLLM training and evaluation. In this paper, $\textbf{(i)}$ we introduce a model merging benchmark for MLLMs, which includes multiple tasks such as VQA, Geometry, Chart, OCR, and Grounding, studying both LoRA and full fine-tuning models. Moreover, we explore how model merging can combine different modalities (e.g., vision-language, audio-language, and video-language models), moving toward the Omni-language model. $\textbf{(ii)}$ We implement 10 model merging algorithms on the benchmark. Furthermore, we propose a novel method that removes noise from task vectors and robustly optimizes the merged vector based on a loss defined over task vector interactions, achieving an average performance gain of 2.48%. $\textbf{(iii)}$ We find that model merging offers a promising way for building improved MLLMs without requiring training data. Our results also demonstrate that the complementarity among multiple modalities outperforms individual modalities.

Paperid: 1860, https://arxiv.org/pdf/2510.22102.pdf

Abstract:
Multimodal large language models (MLLMs) excel at vision-language tasks such as VQA and document understanding, yet precise coordinate prediction remains challenging. High-resolution inputs exacerbate this difficulty by producing long token sequences that weaken positional encodings and introduce directional biases in coordinate outputs. We investigate this phenomenon by analyzing how MLLMs behave when visual positional encodings (VPEs) are deliberately perturbed through shuffling. Our analysis reveals that such perturbations induce predictable, non-random coordinate biases rather than random errors, suggesting that models rely on internal positional priors when spatial grounding signals are degraded. Crucially, we observe similar directional error patterns in natural high-resolution datasets, indicating that positional encoding failures are a key bottleneck for accurate coordinate prediction at scale. To address this issue, we propose Vision-PE Shuffle Guidance (VPSG), a training-free test-time method that leverages the directional nature of these biases for correction. VPSG runs auxiliary decoding with shuffled VPEs to isolate position-unconditioned tendencies, then uses this as negative evidence to guide digit prediction while preserving coordinate format through a lightweight finite-state machine. Experiments on ScreenSpot-Pro demonstrate reliable improvements, highlighting positional encoding robustness as a critical factor for spatial reasoning in MLLMs.

Paperid: 1861, https://arxiv.org/pdf/2510.12287.pdf

Abstract:
Vision Language Models (VLMs) have achieved impressive progress in multimodal reasoning; yet, they remain vulnerable to hallucinations, where outputs are not grounded in visual evidence. In this paper, we investigate a previously overlooked setting: logo hallucination, where models generate brand names or textual content despite logos containing no visible words. Using curated splits of pure symbols, hybrids, and text-bearing logos, as well as the challenging Hard-60 subset, we systematically measure hallucination across leading VLMs. We further probe robustness through nine structured perturbations and show that hallucinations persist even under strong distortions, with occlusion exposing the sharpest weaknesses. Embedding-level analysis with open-weight LLaVA demonstrates that hallucination is tied to a small subset of projector dimensions, and targeted ablation substantially reduces errors while preserving OCR accuracy. Together, these findings reveal that VLMs often rely on symbolic priors rather than genuine glyph perception, particularly for iconic circular logos, and that projector subspaces play a decisive role in this failure mode. Our work contributes both a novel diagnostic lens and actionable mitigation insights, highlighting projector disentanglement and OCR-guided decoding as promising directions for building more trustworthy multimodal systems.

Paperid: 1862, https://arxiv.org/pdf/2510.06292.pdf

Abstract:
While Large Vision-Language Models (LVLMs) achieve strong performance in multimodal tasks, hallucinations continue to hinder their reliability. Among the three categories of hallucinations, which include object, attribute, and relation, relation hallucinations account for the largest proportion but have received the least attention. To address this issue, we propose ChainMPQ (Multi-Perspective Questions guided Interleaved Chain of Image and Text), a training-free method that improves relational inference in LVLMs by utilizing accumulated textual and visual memories. ChainMPQ first extracts subject and object keywords from the question to enhance the corresponding image regions. It then constructs multi-perspective questions that focus on the three core components of a relationship: the subject, the object, and the relation that links them. These questions are sequentially input to the model, with textual and visual memories from earlier steps providing supporting context for subsequent ones, thereby forming an interleaved chain of images and text that guides progressive relational reasoning. Experiments on multiple LVLMs and benchmarks show that ChainMPQ substantially reduces relation hallucinations, while ablation studies further validate the effectiveness of its three core modules.

Paperid: 1863, https://arxiv.org/pdf/2510.04560.pdf

Abstract:
Recent advances demonstrate that multimodal large language models (MLLMs) exhibit strong multimodal in-context learning (ICL) capabilities, enabling them to adapt to novel vision-language tasks from a few contextual examples. However, existing ICL approaches face challenges in reconciling scalability with robustness across diverse tasks and noisy contextual examples: manually selecting examples produces clean contexts but is labor-intensive and task-specific, while similarity-based retrieval improves scalability but could introduce irrelevant or structurally inconsistent samples that degrade ICL performance. To address these limitations, we propose ContextNav, the first agentic framework that integrates the scalability of automated retrieval with the quality and adaptiveness of human-like curation, enabling noise-robust and dynamically optimized contextualization for multimodal ICL. ContextNav unifies context management and noise-robust contextualization within a closed-loop workflow driven by graph-based orchestration. Specifically, it builds a resource-aware multimodal embedding pipeline, maintains a retrievable vector database, and applies agentic retrieval and structural alignment to construct noise-resilient contexts. An Operational Grammar Graph (OGG) further supports adaptive workflow planning and optimization, enabling the agent to refine its operational strategies based on downstream ICL feedback. Experimental results demonstrate that ContextNav achieves state-of-the-art performance across various datasets, underscoring the promise of agentic workflows for advancing scalable and robust contextualization in multimodal ICL.

Paperid: 1864, https://arxiv.org/pdf/2509.24133.pdf

Abstract:
Grounding natural language queries in graphical user interfaces (GUIs) presents a challenging task that requires models to comprehend diverse UI elements across various applications and systems, while also accurately predicting the spatial coordinates for the intended operation. To tackle this problem, we propose GMS: Generalist Scanner Meets Specialist Locator, a synergistic coarse-to-fine framework that effectively improves GUI grounding performance. GMS leverages the complementary strengths of general vision-language models (VLMs) and small, task-specific GUI grounding models by assigning them distinct roles within the framework. Specifically, the general VLM acts as a 'Scanner' to identify potential regions of interest, while the fine-tuned grounding model serves as a 'Locator' that outputs precise coordinates within these regions. This design is inspired by how humans perform GUI grounding, where the eyes scan the interface and the brain focuses on interpretation and localization. Our whole framework consists of five stages and incorporates hierarchical search with cross-modal communication to achieve promising prediction results. Experimental results on the ScreenSpot-Pro dataset show that while the 'Scanner' and 'Locator' models achieve only $2.0\%$ and $3.7\%$ accuracy respectively when used independently, their integration within GMS framework yields an overall accuracy of $35.7\%$, representing a $10 \times$ improvement. Additionally, GMS significantly outperforms other strong baselines under various settings, demonstrating its robustness and potential for general-purpose GUI grounding.

Paperid: 1865, https://arxiv.org/pdf/2509.23789.pdf

Abstract:
Chain-of-Thought (CoT) techniques have significantly enhanced reasoning in Vision-Language Models (VLMs). Extending this paradigm, Visual CoT integrates explicit visual edits, such as cropping or annotating regions of interest, into the reasoning process, achieving superior multimodal performance. However, the robustness of Visual CoT-based VLMs against image-level noise remains unexplored. In this paper, we present the first systematic evaluation of Visual CoT robustness under visual perturbations. Our benchmark spans 12 image corruption types across 4 Visual Question Answering (VQA) datasets, enabling a comprehensive comparison between VLMs that use Visual CoT, and VLMs that do not. The results reveal that integrating Visual CoT consistently improves absolute accuracy regardless of whether the input images are clean or corrupted by noise; however, it also increases sensitivity to input perturbations, resulting in sharper performance degradation compared to standard VLMs. Through extensive analysis, we identify the intermediate reasoning components of Visual CoT, i.e., the edited image patches , as the primary source of fragility. Building on this analysis, we propose a plug-and-play robustness enhancement method that integrates Grounding DINO model into the Visual CoT pipeline, providing high-confidence local visual cues to stabilize reasoning. Our work reveals clear fragility patterns in Visual CoT and offers an effective, architecture-agnostic solution for enhancing visual robustness.

Paperid: 1866, https://arxiv.org/pdf/2508.10264.pdf

Abstract:
Large Vision-Language Models (LVLMs) have shown strong performance across multimodal tasks. However, they often produce hallucinations -- text that is inconsistent with visual input, due to the limited ability to verify information in different regions of the image. To address this, we propose Multi-Region Fusion Decoding (MRFD), a training-free decoding method that improves factual grounding by modeling inter-region consistency. MRFD identifies salient regions using cross-attention, generates initial responses for each, and computes reliability weights based on Jensen-Shannon Divergence (JSD) among the responses. These weights guide a consistency-aware fusion of per-region predictions, using region-aware prompts inspired by Chain-of-Thought reasoning. Experiments across multiple LVLMs and benchmarks show that MRFD significantly reduces hallucinations and improves response factuality without requiring model updates.

Paperid: 1867, https://arxiv.org/pdf/2508.01678.pdf

Abstract:
Vision-Language Models (VLMs) often suffer from hallucination, partly due to challenges in aligning multimodal information. We propose Prompt-in-Image, a simple method that embeds textual instructions directly into images. This removes the need for separate text inputs and forces the model to process all content through the visual channel. We evaluate this method on three popular open-source VLMs: Qwen2.5-VL, LLaVA-1.5, and InstructBLIP. The results reveal sharp differences. Prompt-in-Image improves Qwen2.5-VL's performance, increasing POPE accuracy by 4.1 percent (from 80.2 percent to 84.3 percent) and also reducing hallucination rates on MS-COCO. In contrast, LLaVA-1.5 and InstructBLIP experience a severe performance drop, with accuracy falling from around 84 percent to near-random levels. Through detailed analysis, we found that CLIP-based encoders in LLaVA and InstructBLIP exhibit excessive attention bias toward embedded text regions, disrupting visual understanding. In contrast, Qwen's vision encoder handles text-embedded images robustly. Crucially, Prompt-in-Image reduces Qwen's modality gap, enhancing cross-modal alignment by unifying information processing through a single modality.

Paperid: 1868, https://arxiv.org/pdf/2507.20890.pdf

Abstract:
Img2LaTeX is a practically significant task that involves converting mathematical expressions or tabular data from images into LaTeX code. In recent years, vision-language models (VLMs) have demonstrated strong performance across a variety of visual understanding tasks, owing to their generalization capabilities. While some studies have explored the use of VLMs for the Img2LaTeX task, their performance often falls short of expectations. Empirically, VLMs sometimes struggle with fine-grained visual elements, leading to inaccurate LaTeX predictions. To address this challenge, we propose $A^2R^2$: Advancing Img2LaTeX Conversion via Visual Reasoning with Attention-Guided Refinement, a framework that effectively integrates attention localization and iterative refinement within a visual reasoning framework, enabling VLMs to perform self-correction and progressively improve prediction quality. For effective evaluation, we introduce a new dataset, Img2LaTex-Hard-1K, consisting of 1,100 carefully curated and challenging examples designed to rigorously evaluate the capabilities of VLMs within this task domain. Extensive experimental results demonstrate that: (1) $A^2R^2$ significantly improves model performance across six evaluation metrics spanning both textual and visual levels, consistently outperforming other baseline methods; (2) Increasing the number of inference rounds yields notable performance gains, underscoring the potential of $A^2R^2$ in test-time scaling scenarios; (3) Ablation studies and human evaluations validate the practical effectiveness of our approach, as well as the strong synergy among its core components during inference.

Paperid: 1869, https://arxiv.org/pdf/2507.00008.pdf

Abstract:
Grounding natural language queries in graphical user interfaces (GUIs) poses unique challenges due to the diversity of visual elements, spatial clutter, and the ambiguity of language. In this paper, we introduce DiMo-GUI, a training-free framework for GUI grounding that leverages two core strategies: dynamic visual grounding and modality-aware optimization. Instead of treating the GUI as a monolithic image, our method splits the input into textual elements and iconic elements, allowing the model to reason over each modality independently using general-purpose vision-language models. When predictions are ambiguous or incorrect, DiMo-GUI dynamically focuses attention by generating candidate focal regions centered on the model's initial predictions and incrementally zooms into subregions to refine the grounding result. This hierarchical refinement process helps disambiguate visually crowded layouts without the need for additional training or annotations. We evaluate our approach on standard GUI grounding benchmarks and demonstrate consistent improvements over baseline inference pipelines, highlighting the effectiveness of combining modality separation with region-focused reasoning.

Paperid: 1870, https://arxiv.org/pdf/2506.02803.pdf

Abstract:
Vision-language models (VLMs) excel in semantic tasks but falter at a core human capability: detecting hidden content in optical illusions or AI-generated images through perceptual adjustments like zooming. We introduce HC-Bench, a benchmark of 112 images with hidden text, objects, and illusions, revealing that leading VLMs achieve near-zero accuracy (0-5.36%)-even with explicit prompting. Humans resolve such ambiguities instinctively, yet VLMs fail due to an overreliance on high-level semantics. Strikingly, we propose SemVink (Semantic Visual Thinking) by simply scaling images to low resolutions (32-128 pixels), which unlocks >99% accuracy by eliminating redundant visual noise. This exposes a critical architectural flaw: VLMs prioritize abstract reasoning over low-level visual operations crucial for real-world robustness. Our work urges a shift toward hybrid models integrating multi-scale processing, bridging the gap between computational vision and human cognition for applications in medical imaging, security, and beyond.

Paperid: 1871, https://arxiv.org/pdf/2504.01589.pdf

Abstract:
Vision-language models (VLMs) have advanced rapidly in processing multimodal information, but their ability to reconcile conflicting signals across modalities remains underexplored. This work investigates how VLMs process ASCII art, a unique medium where textual elements collectively form visual patterns, potentially creating semantic-visual conflicts. We introduce a novel evaluation framework that systematically challenges five state-of-the-art models (including GPT-4o, Claude, and Gemini) using adversarial ASCII art, where character-level semantics deliberately contradict global visual patterns. Our experiments reveal a strong text-priority bias: VLMs consistently prioritize textual information over visual patterns, with visual recognition ability declining dramatically as semantic complexity increases. Various mitigation attempts through visual parameter tuning and prompt engineering yielded only modest improvements, suggesting that this limitation requires architectural-level solutions. These findings uncover fundamental flaws in how current VLMs integrate multimodal information, providing important guidance for future model development while highlighting significant implications for content moderation systems vulnerable to adversarial examples.

Paperid: 1872, https://arxiv.org/pdf/2503.23768.pdf

Abstract:
Modern Vision-Language Models (VLMs) exhibit remarkable visual and linguistic capabilities, achieving impressive performance in various tasks such as image recognition and object localization. However, their effectiveness in fine-grained tasks remains an open question. In everyday scenarios, individuals encountering design materials, such as magazines, typography tutorials, research papers, or branding content, may wish to identify aesthetically pleasing fonts used in the text. Given their multimodal capabilities and free accessibility, many VLMs are often considered potential tools for font recognition. This raises a fundamental question: Do VLMs truly possess the capability to recognize fonts? To investigate this, we introduce the Font Recognition Benchmark (FRB), a compact and well-structured dataset comprising 15 commonly used fonts. FRB includes two versions: (i) an easy version, where 10 sentences are rendered in different fonts, and (ii) a hard version, where each text sample consists of the names of the 15 fonts themselves, introducing a stroop effect that challenges model perception. Through extensive evaluation of various VLMs on font recognition tasks, we arrive at the following key findings: (i) Current VLMs exhibit limited font recognition capabilities, with many state-of-the-art models failing to achieve satisfactory performance and being easily affected by the stroop effect introduced by textual information. (ii) Few-shot learning and Chain-of-Thought (CoT) prompting provide minimal benefits in improving font recognition accuracy across different VLMs. (iii) Attention analysis sheds light on the inherent limitations of VLMs in capturing semantic features.

Paperid: 1873, https://arxiv.org/pdf/2503.11750.pdf

Abstract:
In the realm of large vision-language models (LVLMs), adversarial jailbreak attacks serve as a red-teaming approach to identify safety vulnerabilities of these models and their associated defense mechanisms. However, we identify a critical limitation: not every adversarial optimization step leads to a positive outcome, and indiscriminately accepting optimization results at each step may reduce the overall attack success rate. To address this challenge, we introduce HKVE (Hierarchical Key-Value Equalization), an innovative jailbreaking framework that selectively accepts gradient optimization results based on the distribution of attention scores across different layers, ensuring that every optimization step positively contributes to the attack. Extensive experiments demonstrate HKVE's significant effectiveness, achieving attack success rates of 75.08% on MiniGPT4, 85.84% on LLaVA and 81.00% on Qwen-VL, substantially outperforming existing methods by margins of 20.43\%, 21.01\% and 26.43\% respectively. Furthermore, making every step effective not only leads to an increase in attack success rate but also allows for a reduction in the number of iterations, thereby lowering computational costs. Warning: This paper contains potentially harmful example data.

Paperid: 1874, https://arxiv.org/pdf/2503.11619.pdf

Abstract:
Deploying large vision-language models (LVLMs) introduces a unique vulnerability: susceptibility to malicious attacks via visual inputs. However, existing defense methods suffer from two key limitations: (1) They solely focus on textual defenses, fail to directly address threats in the visual domain where attacks originate, and (2) the additional processing steps often incur significant computational overhead or compromise model performance on benign tasks. Building on these insights, we propose ESIII (Embedding Security Instructions Into Images), a novel methodology for transforming the visual space from a source of vulnerability into an active defense mechanism. Initially, we embed security instructions into defensive images through gradient-based optimization, obtaining security instructions in the visual dimension. Subsequently, we integrate security instructions from visual and textual dimensions with the input query. The collaboration between security instructions from different dimensions ensures comprehensive security protection. Extensive experiments demonstrate that our approach effectively fortifies the robustness of LVLMs against such attacks while preserving their performance on standard benign tasks and incurring an imperceptible increase in time costs.

Paperid: 1875, https://arxiv.org/pdf/2502.02977.pdf

Abstract:
Vision-language models like CLIP excel at recognizing the single, prominent object in a scene. However, they struggle in complex scenes containing multiple objects. We identify a fundamental reason for this limitation: VLM feature space exhibits excessive mutual feature information (MFI), where the features of one class contain substantial information about other, unrelated classes. This high MFI becomes evident during class-specific queries, as unrelated objects are activated alongside the queried class. To address this limitation, we propose DCLIP, an efficient framework that learns an optimal level of mutual information while adding only minimal learnable parameters to a frozen VLM. DCLIP uses two complementary losses: a novel MFI Loss that regulates class feature similarity to prevent excessive overlap while preserving necessary shared information, and the Asymmetric Loss (ASL) that aligns image features with the disentangled text features. Through this disentanglement, DCLIP reduces excessive inter-class similarity by 30%. On multi-label recognition, DCLIP performs favorably over SOTA approaches on VOC2007 and COCO-14 while using 75% fewer training parameters. For zero-shot semantic segmentation, it shows improved performance across six benchmark datasets. These results highlight the importance of feature disentanglement for multi-object perception in VLMs.

Paperid: 1876, https://arxiv.org/pdf/2411.14460.pdf

Abstract:
Structured data, such as tables, graphs, and databases, play a critical role in plentiful NLP tasks such as question answering and dialogue system. Recently, inspired by Vision-Language Models, Graph Neutral Networks (GNNs) have been introduced as an additional modality into the input of Large Language Models (LLMs) to improve their performance on Structured Knowledge Grounding (SKG) tasks. However, those GNN-enhanced LLMs have the following limitations: (1) They employ diverse GNNs to model varying types of structured data, rendering them unable to uniformly process various forms of structured data. (2) The pretraining of GNNs is coupled with specific LLMs, which prevents GNNs from fully aligning with the textual space and limits their adaptability to other LLMs. To address these issues, we propose \textbf{L}arge \textbf{L}anguage and \textbf{S}tructured Data \textbf{A}ssistant (LLaSA), a general framework for enhancing LLMs' ability to handle structured data. Specifically, we represent various types of structured data in a unified hypergraph format, and use self-supervised learning to pretrain a hypergraph encoder, and a G-Former compressing encoded hypergraph representations with cross-attention. The compressed hypergraph representations are appended to the serialized inputs during training and inference stages of LLMs. Experimental results on multiple SKG tasks show that our pretrained hypergraph encoder can adapt to various LLMs and enhance their ability to process different types of structured data. Besides, LLaSA, with LoRA fine-tuning, outperforms previous SOTA method using full parameters tuning.

Paperid: 1877, https://arxiv.org/pdf/2511.17411.pdf

Abstract:
Robotic Foundation Models (RFMs) hold great promise as generalist, end-to-end systems for robot control. Yet their ability to generalize across new environments, tasks, and embodiments remains limited. We argue that a major bottleneck lies in their foundations: most RFMs are built by fine-tuning internet-pretrained Vision-Language Models (VLMs). However, these VLMs are trained on 2D image-language tasks and lack the 3D spatial reasoning inherently required for embodied control in the 3D world. Bridging this gap directly with large-scale robotic data is costly and difficult to scale. Instead, we propose to enrich easy-to-collect non-robotic image data with 3D annotations and enhance a pretrained VLM with 3D understanding capabilities. Following this strategy, we train SPEAR-VLM, a 3D-aware VLM that infers object coordinates in 3D space from a single 2D image. Building on SPEAR-VLM, we introduce our main contribution, $~\textbf{SPEAR-1}$: a robotic foundation model that integrates grounded 3D perception with language-instructed embodied control. Trained on $\sim$45M frames from 24 Open X-Embodiment datasets, SPEAR-1 outperforms or matches state-of-the-art models such as $π_0$-FAST and $π_{0.5}$, while it uses 20$\times$ fewer robot demonstrations. This carefully-engineered training strategy unlocks new VLM capabilities and as a consequence boosts the reliability of embodied control beyond what is achievable with only robotic data. We make our model weights and 3D-annotated datasets publicly available.

Paperid: 1878, https://arxiv.org/pdf/2509.19958.pdf

Abstract:
Recent advances in generalist robot manipulation leverage pre-trained Vision-Language Models (VLMs) and large-scale robot demonstrations to tackle diverse tasks in a zero-shot manner. A key challenge remains: scaling high-quality, action-labeled robot demonstration data, which existing methods rely on for robustness and generalization. To address this, we propose a method that benefits from videos without action labels - featuring humans and/or robots in action - enhancing open-vocabulary performance and enabling data-efficient learning of new tasks. Our method extracts dense, dynamic 3D point clouds at the hand or gripper location and uses a proposed 3D dynamics predictor for self-supervision. This predictor is then tuned to an action predictor using a smaller labeled dataset for action alignment. We show that our method not only learns from unlabeled human and robot demonstrations - improving downstream generalist robot policies - but also enables robots to learn new tasks without action labels (i.e., out-of-action generalization) in both real-world and simulated settings.

Paperid: 1879, https://arxiv.org/pdf/2507.00886.pdf

Abstract:
As multimodal language models advance, their application to 3D scene understanding is a fast-growing frontier, driving the development of 3D Vision-Language Models (VLMs). Current methods show strong dependence on object detectors, introducing processing bottlenecks and limitations in taxonomic flexibility. To address these limitations, we propose a scene-centric 3D VLM for 3D Gaussian splat scenes that employs language- and task-aware scene representations. Our approach directly embeds rich linguistic features into the 3D scene representation by associating language with each Gaussian primitive, achieving early modality alignment. To process the resulting dense representations, we introduce a dual sparsifier that distills them into compact, task-relevant tokens via task-guided and location-guided pathways, producing sparse, task-aware global and local scene tokens. Notably, we present the first Gaussian splatting-based VLM, leveraging photorealistic 3D representations derived from standard RGB images, demonstrating strong generalization: it improves performance of prior 3D VLM five folds, in out-of-the-domain settings.

Paperid: 1880, https://arxiv.org/pdf/2506.22032.pdf

Abstract:
Zero-shot Semantic Segmentation (ZSS) aims to segment both seen and unseen classes using supervision from only seen classes. Beyond adaptation-based methods, distillation-based approaches transfer vision-language alignment of vision-language model, e.g., CLIP, to segmentation models. However, such knowledge transfer remains challenging due to: (1) the difficulty of aligning vision-based features with the textual space, which requires combining spatial precision with vision-language alignment; and (2) the semantic gap between CLIP's global representations and the local, fine-grained features of segmentation models. To address challenge (1), we propose Chimera-Seg, which integrates a segmentation backbone as the body and a CLIP-based semantic head as the head, like the Chimera in Greek mythology, combining spatial precision with vision-language alignment. Specifically, Chimera-Seg comprises a trainable segmentation model and a CLIP Semantic Head (CSH), which maps dense features into the CLIP-aligned space. The CSH incorporates a frozen subnetwork and fixed projection layers from the CLIP visual encoder, along with lightweight trainable components. The partial module from CLIP visual encoder, paired with the segmentation model, retains segmentation capability while easing the mapping to CLIP's semantic space. To address challenge (2), we propose Selective Global Distillation (SGD), which distills knowledge from dense features exhibiting high similarity to the CLIP CLS token, while gradually reducing the number of features used for alignment as training progresses. Besides, we also use a Semantic Alignment Module (SAM) to further align dense visual features with semantic embeddings extracted from the frozen CLIP text encoder. Experiments on two benchmarks show improvements of 0.9% and 1.2% in hIoU.

Paperid: 1881, https://arxiv.org/pdf/2506.01667.pdf

Abstract:
Large Multimodal Models (LMMs) have demonstrated strong performance in various vision-language tasks. However, they often struggle to comprehensively understand Earth Observation (EO) data, which is critical for monitoring the environment and the effects of human activity on it. In this work, we present EarthMind, a novel vision-language framework for multi-granular and multi-sensor EO data understanding. EarthMind features two core components: (1) Spatial Attention Prompting (SAP), which reallocates attention within the LLM to enhance pixel-level understanding; and (2) Cross-modal Fusion, which aligns heterogeneous modalities into a shared space and adaptively reweighs tokens based on their information density for effective fusion. To facilitate multi-sensor fusion evaluation, we propose EarthMind-Bench, a comprehensive benchmark with over 2,000 human-annotated multi-sensor image-question pairs, covering a wide range of perception and reasoning tasks. Extensive experiments demonstrate the effectiveness of EarthMind. It achieves state-of-the-art performance on EarthMind-Bench, surpassing GPT-4o despite being only 4B in scale. Moreover, EarthMind outperforms existing methods on multiple public EO benchmarks, showcasing its potential to handle both multi-granular and multi-sensor challenges in a unified framework.

Paperid: 1882, https://arxiv.org/pdf/2505.05023.pdf

Abstract:
Zero-shot Semantic Segmentation (ZSS) aims to segment categories that are not annotated during training. While fine-tuning vision-language models has achieved promising results, these models often overfit to seen categories due to the lack of supervision for unseen classes. As an alternative to fully supervised approaches, query-based segmentation has shown great latent in ZSS, as it enables object localization without relying on explicit labels. However, conventional Hungarian matching, a core component in query-based frameworks, needs full supervision and often misclassifies unseen categories as background in the setting of ZSS. To address this issue, we propose Split Matching (SM), a novel assignment strategy that decouples Hungarian matching into two components: one for seen classes in annotated regions and another for latent classes in unannotated regions (referred to as unseen candidates). Specifically, we partition the queries into seen and candidate groups, enabling each to be optimized independently according to its available supervision. To discover unseen candidates, we cluster CLIP dense features to generate pseudo masks and extract region-level embeddings using CLS tokens. Matching is then conducted separately for the two groups based on both class-level similarity and mask-level consistency. Additionally, we introduce a Multi-scale Feature Enhancement (MFE) module that refines decoder features through residual multi-scale aggregation, improving the model's ability to capture spatial details across resolutions. SM is the first to introduce decoupled Hungarian matching under the inductive ZSS setting, and achieves state-of-the-art performance on two standard benchmarks.

Paperid: 1883, https://arxiv.org/pdf/2501.08982.pdf

Abstract:
Localizing textual descriptions within large-scale 3D scenes presents inherent ambiguities, such as identifying all traffic lights in a city. Addressing this, we introduce a method to generate distributions of camera poses conditioned on textual descriptions, facilitating robust reasoning for broadly defined concepts. Our approach employs a diffusion-based architecture to refine noisy 6DoF camera poses towards plausible locations, with conditional signals derived from pre-trained text encoders. Integration with the pretrained Vision-Language Model, CLIP, establishes a strong linkage between text descriptions and pose distributions. Enhancement of localization accuracy is achieved by rendering candidate poses using 3D Gaussian splatting, which corrects misaligned samples through visual reasoning. We validate our method's superiority by comparing it against standard distribution estimation methods across five large-scale datasets, demonstrating consistent outperformance. Code, datasets and more information will be publicly available at our project page.

Paperid: 1884, https://arxiv.org/pdf/2412.01370.pdf

Abstract:
Museums serve as repositories of cultural heritage and historical artifacts from diverse epochs, civilizations, and regions, preserving well-documented collections that encapsulate vast knowledge, which, when systematically structured into large-scale datasets, can train specialized models. Visitors engage with exhibits through curiosity and questions, making expert domain-specific models essential for interactive query resolution and gaining historical insights. Understanding exhibits from images requires analyzing visual features and linking them to historical knowledge to derive meaningful correlations. We facilitate such reasoning by (a) collecting and curating a large-scale dataset of 65M images and 200M question-answer pairs for exhibits from all around the world; (b) training large vision-language models (VLMs) on the collected dataset; (c) benchmarking their ability on five visual question answering tasks, specifically designed to reflect real-world inquiries and challenges observed in museum settings. The complete dataset is labeled by museum experts, ensuring the quality and the practical significance of the labels. We train two VLMs from different categories: BLIP with vision-language aligned embeddings, but lacking the expressive power of large language models, and the LLaVA model, a powerful instruction-tuned LLM enriched with vision-language reasoning capabilities. Through extensive experiments, we find that while both model types effectively answer visually grounded questions, large vision-language models excel in queries requiring deeper historical context and reasoning. We further demonstrate the necessity of fine-tuning models on large-scale domain-specific datasets by showing that our fine-tuned models significantly outperform current SOTA VLMs in answering questions related to specific attributes, highlighting their limitations in handling complex, nuanced queries.

Paperid: 1885, https://arxiv.org/pdf/2409.15250.pdf

Abstract:
Recent progress in large language models and access to large-scale robotic datasets has sparked a paradigm shift in robotics models transforming them into generalists able to adapt to various tasks, scenes, and robot modalities. A large step for the community are open Vision Language Action models which showcase strong performance in a wide variety of tasks. In this work, we study the visual generalization capabilities of three existing robotic foundation models, and propose a corresponding evaluation framework. Our study shows that the existing models do not exhibit robustness to visual out-of-domain scenarios. This is potentially caused by limited variations in the training data and/or catastrophic forgetting, leading to domain limitations in the vision foundation models. We further explore OpenVLA, which uses two pre-trained vision foundation models and is, therefore, expected to generalize to out-of-domain experiments. However, we showcase catastrophic forgetting by DINO-v2 in OpenVLA through its failure to fulfill the task of depth regression. To overcome the aforementioned issue of visual catastrophic forgetting, we propose a gradual backbone reversal approach founded on model merging. This enables OpenVLA -- which requires the adaptation of the visual backbones during initial training -- to regain its visual generalization ability. Regaining this capability enables our ReVLA model to improve over OpenVLA by a factor of 77\% and 66\% for grasping and lifting in visual OOD tasks. Comprehensive evaluations, episode rollouts and model weights are available on the ReVLA Page

Paperid: 1886, https://arxiv.org/pdf/2601.06559.pdf

Abstract:
Grounding events in videos serves as a fundamental capability in video analysis. While Vision-Language Models (VLMs) are increasingly employed for this task, existing approaches predominantly train models to associate events with timestamps in the forward video only. This paradigm hinders VLMs from capturing the inherent temporal structure and directionality of events, thereby limiting robustness and generalization. To address this limitation, inspired by the arrow of time in physics, which characterizes the intrinsic directionality of temporal processes, we propose ArrowGEV, a reinforcement learning framework that explicitly models temporal directionality in events to improve both event grounding and temporal directionality understanding in VLMs. Specifically, we categorize events into time-sensitive (e.g., putting down a bag) and time-insensitive (e.g., holding a towel in the left hand). The former denote events whose reversal substantially alters their meaning, while the latter remain semantically unchanged under reversal. For time-sensitive events, ArrowGEV introduces a reward that encourages VLMs to discriminate between forward and backward videos, whereas for time-insensitive events, it enforces consistent grounding across both directions. Extensive experiments demonstrate that ArrowGEV not only improves grounding precision and temporal directionality recognition, but also enhances general video understanding and reasoning ability.

Paperid: 1887, https://arxiv.org/pdf/2506.04039.pdf

Abstract:
Large Visual Language Models (LVLMs) have demonstrated impressive capabilities across multiple tasks. However, their trustworthiness is often challenged by hallucinations, which can be attributed to the modality misalignment and the inherent hallucinations of their underlying Large Language Models (LLMs) backbone. Existing preference alignment methods focus on aligning model responses with human preferences while neglecting image-text modality alignment, resulting in over-reliance on LLMs and hallucinations. In this paper, we propose Entity-centric Multimodal Preference Optimization (EMPO), which achieves enhanced modality alignment compared to existing human preference alignment methods. Besides, to overcome the scarcity of high-quality multimodal preference data, we utilize open-source instruction datasets to automatically construct high-quality preference data across three aspects: image, instruction, and response. Experiments on two human preference datasets and five multimodal hallucination benchmarks demonstrate the effectiveness of EMPO, e.g., reducing hallucination rates by 85.9\% on Object-HalBench and 49.8\% on MM-HalBench.

Paperid: 1888, https://arxiv.org/pdf/2508.04567.pdf

Abstract:
As scaling up training data has significantly improved the general multimodal capabilities of Large Vision-Language Models (LVLMs), they still suffer from the hallucination issue, generating text that is inconsistent with the visual input. This phenomenon motivates us to systematically investigate the role of training data in hallucination. We introduce a new benchmark, POPEv2, which consists of counterfactual images collected from the training data of LVLMs with certain objects masked. Through comprehensive evaluation on POPEv2, we find that current LVLMs suffer from training bias: they fail to fully leverage their training data and hallucinate more frequently on images seen during training. Specifically, they perform poorly on counterfactual images, often incorrectly answering ``Yes'' to questions about masked objects. To understand this issue, we conduct probing experiments on the models' internal components, revealing that this training bias is primarily located in the language modeling (LM) head. Based on these findings, we propose Obliviate, an efficient and lightweight unlearning method designed to mitigate object hallucination via training bias unlearning. Obliviate identifies the discrepancy between ground-truth labels and model outputs on the training data as a proxy for bias and adopts a parameter- and data-efficient fine-tuning strategy that only updates the LM head. Extensive experiments demonstrate the effectiveness of our approach. While only reusing the training data and updating approximately 2\% of the parameters, Obliviate significantly reduces hallucination across both discriminative and generative tasks. Furthermore, it demonstrates strong scalability with respect to both model size (2B to 72B) and training data volume, and exhibits promising generalization to hallucination types beyond object-level hallucination. Our code and data will be publicly released.

Paperid: 1889, https://arxiv.org/pdf/2502.11427.pdf

Abstract:
Visual instruction tuning has become the predominant technology in eliciting the multimodal task-solving capabilities of large vision-language models (LVLMs). Despite the success, as visual instructions require images as the input, it would leave the gap in inheriting the task-solving capabilities from the backbone LLMs, and make it costly to collect a large-scale dataset. To address it, we propose ViFT, a visual instruction-free fine-tuning framework for LVLMs. In ViFT, we only require the text-only instructions and image caption data during training, to separately learn the task-solving and visual perception abilities. During inference, we extract and combine the representations of the text and image inputs, for fusing the two abilities to fulfill multimodal tasks. Experimental results demonstrate that ViFT can achieve state-of-the-art performance on several visual reasoning and visual instruction following benchmarks, with rather less training data. Our code and data will be publicly released.

Paperid: 1890, https://arxiv.org/pdf/2508.05748.pdf

Abstract:
Web agents such as Deep Research have demonstrated superhuman cognitive abilities, capable of solving highly challenging information-seeking problems. However, most research remains primarily text-centric, overlooking visual information in the real world. This makes multimodal Deep Research highly challenging, as such agents require much stronger reasoning abilities in perception, logic, knowledge, and the use of more sophisticated tools compared to text-based agents. To address this limitation, we introduce WebWatcher, a multi-modal Agent for Deep Research equipped with enhanced visual-language reasoning capabilities. It leverages high-quality synthetic multimodal trajectories for efficient cold start training, utilizes various tools for deep reasoning, and further enhances generalization through reinforcement learning. To better evaluate the capabilities of multimodal agents, we propose BrowseComp-VL, a benchmark with BrowseComp-style that requires complex information retrieval involving both visual and textual information. Experimental results show that WebWatcher significantly outperforms proprietary baseline, RAG workflow and open-source agents in four challenging VQA benchmarks, which paves the way for solving complex multimodal information-seeking tasks.

Paperid: 1891, https://arxiv.org/pdf/2511.00540.pdf

Abstract:
Industrial Anomaly Detection (IAD) is critical for enhancing operational safety, ensuring product quality, and optimizing manufacturing efficiency across global industries. However, the IAD algorithms are severely constrained by the limitations of existing public benchmarks. Current datasets exhibit restricted category diversity and insufficient scale, frequently resulting in metric saturation and limited model transferability to real-world scenarios. To address this gap, we introduce Real-IAD Variety, the largest and most diverse IAD benchmark, comprising 198,960 high-resolution images across 160 distinct object categories. Its diversity is ensured through comprehensive coverage of 28 industries, 24 material types, and 22 color variations. Our comprehensive experimental analysis validates the benchmark's substantial challenge: state-of-the-art multi-class unsupervised anomaly detection methods experience significant performance degradation when scaled from 30 to 160 categories. Crucially, we demonstrate that vision-language models exhibit remarkable robustness to category scale-up, with minimal performance variation across different category counts, significantly enhancing generalization capabilities in diverse industrial contexts. The unprecedented scale and complexity of Real-IAD Variety position it as an essential resource for training and evaluating next-generation foundation models for anomaly detection. By providing this comprehensive benchmark with rigorous evaluation protocols across multi-class unsupervised, multi-view, and zero-/few-shot settings, we aim to accelerate research beyond domain-specific constraints, enabling the development of scalable, general-purpose anomaly detection systems. Real-IAD Variety will be made publicly available to facilitate innovation in this critical field.

Paperid: 1892, https://arxiv.org/pdf/2509.19870.pdf

Abstract:
Vision-Language-Action (VLA) models are driving rapid progress in robotics by enabling agents to interpret multimodal inputs and execute complex, long-horizon tasks. However, their safety and robustness against adversarial attacks remain largely underexplored. In this work, we identify and formalize a critical adversarial vulnerability in which adversarial images can "freeze" VLA models and cause them to ignore subsequent instructions. This threat effectively disconnects the robot's digital mind from its physical actions, potentially inducing inaction during critical interventions. To systematically study this vulnerability, we propose FreezeVLA, a novel attack framework that generates and evaluates action-freezing attacks via min-max bi-level optimization. Experiments on three state-of-the-art VLA models and four robotic benchmarks show that FreezeVLA attains an average attack success rate of 76.2%, significantly outperforming existing methods. Moreover, adversarial images generated by FreezeVLA exhibit strong transferability, with a single image reliably inducing paralysis across diverse language prompts. Our findings expose a critical safety risk in VLA models and highlight the urgent need for robust defense mechanisms.

Paperid: 1893, https://arxiv.org/pdf/2508.18381.pdf

Abstract:
Large vision-language models (LVLMs) have demonstrated exceptional capabilities in understanding visual information with human languages but also exhibit an imbalance in multilingual capabilities. In this work, we delve into the multilingual working pattern of LVLMs and identify a salient correlation between the multilingual understanding ability of LVLMs and language-specific neuron activations in shallow layers. Building on this insight, we introduce PLAST, a training recipe that achieves efficient multilingual enhancement for LVLMs by Precise LAnguage-Specific layers fine-Tuning. PLAST first identifies layers involved in multilingual understanding by monitoring language-specific neuron activations. These layers are then precisely fine-tuned with question-translation pairs to achieve multilingual alignment. Our empirical results on MM-Bench and MMMB demonstrate that PLAST effectively improves the multilingual capabilities of LVLMs and achieves significant efficiency with only 14% of the parameters tuned. Further analysis reveals that PLAST can be generalized to low-resource and complex visual reasoning tasks, facilitating the language-specific visual information engagement in shallow layers.

Paperid: 1894, https://arxiv.org/pdf/2508.06163.pdf

Abstract:
Model merging has emerged as a compelling data-free paradigm for multi-task learning, enabling the fusion of multiple fine-tuned models into a single, powerful entity. A key technique in merging methods is sparsification, which prunes redundant parameters from task vectors to mitigate interference. However, prevailing approaches employ a ``one-size-fits-all'' strategy, applying a uniform sparsity ratio that overlooks the inherent structural and statistical heterogeneity of model parameters. This often leads to a suboptimal trade-off, where critical parameters are inadvertently pruned while less useful ones are retained. To address this limitation, we introduce \textbf{TADrop} (\textbf{T}ensor-wise \textbf{A}daptive \textbf{Drop}), an adaptive sparsification strategy that respects this heterogeneity. Instead of a global ratio, TADrop assigns a tailored sparsity level to each parameter tensor based on its distributional properties. The core intuition is that tensors with denser, more redundant distributions can be pruned aggressively, while sparser, more critical ones are preserved. As a simple and plug-and-play module, we validate TADrop by integrating it with foundational, classic, and SOTA merging methods. Extensive experiments across diverse tasks (vision, language, and multimodal) and models (ViT, BEiT) demonstrate that TADrop consistently and significantly boosts their performance. For instance, when enhancing a leading merging method, it achieves an average performance gain of 2.0\% across 8 ViT-B/32 tasks. TADrop provides a more effective way to mitigate parameter interference by tailoring sparsification to the model's structure, offering a new baseline for high-performance model merging.

Paperid: 1895, https://arxiv.org/pdf/2507.07151.pdf

Abstract:
Despite the impressive capabilities of multimodal large language models (MLLMs) in vision-language tasks, they are prone to hallucinations in real-world scenarios. This paper investigates the hallucination phenomenon in MLLMs from the perspective of modality conflict. Unlike existing works focusing on the conflicts between model responses and inputs, we study the inherent conflicts in inputs from different modalities that place MLLMs in a dilemma and directly lead to hallucinations. We formally define the modality conflict and construct a dataset named Multimodal Modality Conflict (MMMC) to simulate this phenomenon in vision-language tasks. Three methods based on prompt engineering, supervised fine-tuning, and reinforcement learning are proposed to alleviate the hallucination caused by modality conflict. Extensive experiments are conducted on the MMMC dataset to analyze the merits and demerits of these methods. Our results show that the reinforcement learning method achieves the best performance in mitigating the hallucination under modality conflict, while the supervised fine-tuning method shows promising and stable performance. Our work sheds light on the unnoticed modality conflict that leads to hallucinations and provides more insights into the robustness of MLLMs.

Paperid: 1896, https://arxiv.org/pdf/2505.24870.pdf

Abstract:
Humans can intuitively compose and arrange scenes in the 3D space for photography. However, can advanced AI image generators plan scenes with similar 3D spatial awareness when creating images from text or image prompts? We present GenSpace, a novel benchmark and evaluation pipeline to comprehensively assess the spatial awareness of current image generation models. Furthermore, standard evaluations using general Vision-Language Models (VLMs) frequently fail to capture the detailed spatial errors. To handle this challenge, we propose a specialized evaluation pipeline and metric, which reconstructs 3D scene geometry using multiple visual foundation models and provides a more accurate and human-aligned metric of spatial faithfulness. Our findings show that while AI models create visually appealing images and can follow general instructions, they struggle with specific 3D details like object placement, relationships, and measurements. We summarize three core limitations in the spatial perception of current state-of-the-art image generation models: 1) Object Perspective Understanding, 2) Egocentric-Allocentric Transformation and 3) Metric Measurement Adherence, highlighting possible directions for improving spatial intelligence in image generation.

Paperid: 1897, https://arxiv.org/pdf/2504.13055.pdf

Abstract:
Recent advances in reinforcement learning (RL) have strengthened the reasoning capabilities of vision-language models (VLMs). However, enhancing policy exploration to better scale test-time compute remains largely underexplored. In addition, VLMs continue to struggle with imperfect visual perception, which in turn affects the subsequent reasoning process. To this end, we propose NoisyRollout, a simple yet effective data augmentation method that mixes trajectories from both clean and moderately distorted images during RL training. By injecting targeted diversity in visual perception and the resulting reasoning patterns, NoisyRollout promotes better policy exploration through vision-oriented inductive biases, ultimately leading to more robust reasoning behaviors. We further adopt a noise annealing schedule that gradually reduces distortion strength over training, leveraging noisy signals early on while ensuring training stability in later stages. Crucially, our method is easy-to-adopt--requiring no additional training cost and no modifications to the RL objective. Extensive experiments on $2$ distinct training datasets demonstrate that NoisyRollout achieves state-of-the-art performance among open-source RL-tuned models across $5$ out-of-domain reasoning and perception benchmarks. Furthermore, we validate the effectiveness of NoisyRollout across model sizes ($7$B and $32$B) and data scales (from $1$K to $6$K), highlighting its generalizability and scalability.

Paperid: 1898, https://arxiv.org/pdf/2601.04266.pdf

Abstract:
Vision-Language-Action (VLA) models are widely deployed in safety-critical embodied AI applications such as robotics. However, their complex multimodal interactions also expose new security vulnerabilities. In this paper, we investigate a backdoor threat in VLA models, where malicious inputs cause targeted misbehavior while preserving performance on clean data. Existing backdoor methods predominantly rely on inserting visible triggers into visual modality, which suffer from poor robustness and low insusceptibility in real-world settings due to environmental variability. To overcome these limitations, we introduce the State Backdoor, a novel and practical backdoor attack that leverages the robot arm's initial state as the trigger. To optimize trigger for insusceptibility and effectiveness, we design a Preference-guided Genetic Algorithm (PGA) that efficiently searches the state space for minimal yet potent triggers. Extensive experiments on five representative VLA models and five real-world tasks show that our method achieves over 90% attack success rate without affecting benign task performance, revealing an underexplored vulnerability in embodied AI systems.

Paperid: 1899, https://arxiv.org/pdf/2510.10028.pdf

Abstract:
The rapid advancement of Low-Altitude Economy Networks (LAENets) has enabled a variety of applications, including aerial surveillance, environmental sensing, and semantic data collection. To support these scenarios, unmanned aerial vehicles (UAVs) equipped with onboard vision-language models (VLMs) offer a promising solution for real-time multimodal inference. However, ensuring both inference accuracy and communication efficiency remains a significant challenge due to limited onboard resources and dynamic network conditions. In this paper, we first propose a UAV-enabled LAENet system model that jointly captures UAV mobility, user-UAV communication, and the onboard visual question answering (VQA) pipeline. Based on this model, we formulate a mixed-integer non-convex optimization problem to minimize task latency and power consumption under user-specific accuracy constraints. To solve the problem, we design a hierarchical optimization framework composed of two parts: (i) an Alternating Resolution and Power Optimization (ARPO) algorithm for resource allocation under accuracy constraints, and (ii) a Large Language Model-augmented Reinforcement Learning Approach (LLaRA) for adaptive UAV trajectory optimization. The large language model (LLM) serves as an expert in refining reward design of reinforcement learning in an offline fashion, introducing no additional latency in real-time decision-making. Numerical results demonstrate the efficacy of our proposed framework in improving inference performance and communication efficiency under dynamic LAENet conditions.

Paperid: 1900, https://arxiv.org/pdf/2410.12662.pdf

Abstract:
Vision-language alignment in Large Vision-Language Models (LVLMs) successfully enables LLMs to understand visual input. However, we find that existing vision-language alignment methods fail to transfer the existing safety mechanism for text in LLMs to vision, which leads to vulnerabilities in toxic image. To explore the cause of this problem, we give the insightful explanation of where and how the safety mechanism of LVLMs operates and conduct comparative analysis between text and vision. We find that the hidden states at the specific transformer layers play a crucial role in the successful activation of safety mechanism, while the vision-language alignment at hidden states level in current methods is insufficient. This results in a semantic shift for input images compared to text in hidden states, therefore misleads the safety mechanism. To address this, we propose a novel Text-Guided vision-language Alignment method (TGA) for LVLMs. TGA retrieves the texts related to input vision and uses them to guide the projection of vision into the hidden states space in LLMs. Experiments show that TGA not only successfully transfers the safety mechanism for text in basic LLMs to vision in vision-language alignment for LVLMs without any safety fine-tuning on the visual modality but also maintains the general performance on various vision tasks (Safe and Good).

Paperid: 1901, https://arxiv.org/pdf/2510.25889.pdf

Abstract:
Vision-Language-Action (VLA) models enable robots to understand and perform complex tasks from multimodal input. Although recent work explores using reinforcement learning (RL) to automate the laborious data collection process in scaling supervised fine-tuning (SFT), applying large-scale RL to flow-based VLAs (e.g., $π_0$, $π_{0.5}$) remains challenging due to intractable action log-likelihoods from iterative denoising. We address this challenge with $π_{\text{RL}}$, an open-source framework for training flow-based VLAs in parallel simulation. $π_{\text{RL}}$ implements two RL algorithms: (1) {Flow-Noise} models the denoising process as a discrete-time MDP with a learnable noise network for exact log-likelihood computation. (2) {Flow-SDE} integrates denoising with agent-environment interaction, formulating a two-layer MDP that employs ODE-to-SDE conversion for efficient RL exploration. We evaluate $π_{\text{RL}}$ on LIBERO and ManiSkill benchmarks. On LIBERO, $π_{\text{RL}}$ boosts few-shot SFT models $π_0$ and $π_{0.5}$ from 57.6% to 97.6% and from 77.1% to 98.3%, respectively. In ManiSkill, we train $π_{\text{RL}}$ in 320 parallel environments, improving $π_0$ from 41.6% to 85.7% and $π_{0.5}$ from 40.0% to 84.8% across 4352 pick-and-place tasks, demonstrating scalable multitask RL under heterogeneous simulation. Overall, $π_{\text{RL}}$ achieves significant performance gains and stronger generalization over SFT-models, validating the effectiveness of online RL for flow-based VLAs.

Paperid: 1902, https://arxiv.org/pdf/2507.22424.pdf

Abstract:
Vision-Language-Action (VLA) models have made substantial progress by leveraging the robust capabilities of Visual Language Models (VLMs). However, VLMs' significant parameter size and autoregressive (AR) decoding nature impose considerable computational demands on VLA models. While Speculative Decoding (SD) has shown efficacy in accelerating Large Language Models (LLMs) by incorporating efficient drafting and parallel verification, allowing multiple tokens to be generated in one forward pass, its application to VLA models remains unexplored. This work introduces Spec-VLA, an SD framework designed to accelerate VLA models. Due to the difficulty of the action prediction task and the greedy decoding mechanism of the VLA models, the direct application of the advanced SD framework to the VLA prediction task yields a minor speed improvement. To boost the generation speed, we propose an effective mechanism to relax acceptance utilizing the relative distances represented by the action tokens of the VLA model. Empirical results across diverse test scenarios affirm the effectiveness of the Spec-VLA framework, and further analysis substantiates the impact of our proposed strategies, which enhance the acceptance length by 44%, achieving 1.42 times speedup compared with the OpenVLA baseline, without compromising the success rate. The success of the Spec-VLA framework highlights the potential for broader application of speculative execution in VLA prediction scenarios.

Paperid: 1903, https://arxiv.org/pdf/2506.02387.pdf

Abstract:
Recent advancements in Vision Language Models (VLMs) have expanded their capabilities to interactive agent tasks, yet existing benchmarks remain limited to single-agent or text-only environments. In contrast, real-world scenarios often involve multiple agents interacting within rich visual and linguistic contexts, posing challenges with both multimodal observations and strategic interactions. To bridge this gap, we introduce Visual Strategic Bench (VS-Bench), a multimodal benchmark that evaluates VLMs for strategic reasoning and decision-making in multi-agent environments. VS-Bench comprises eight vision-grounded environments spanning cooperative, competitive, and mixed-motive interactions, designed to assess agents' ability to predict others' future moves and optimize for long-term objectives. We consider two complementary evaluation dimensions, including offline evaluation of strategic reasoning by next-action prediction accuracy and online evaluation of decision-making by normalized episode return. Extensive experiments of fourteen leading VLMs reveal a significant gap between current models and optimal performance, with the best models attaining 47.8% prediction accuracy and 24.3% normalized return. We further conduct in-depth analyses on multimodal observations, test-time scaling, social behaviors, and failure cases of VLM agents. By standardizing the evaluation and highlighting the limitations of existing models, we envision VS-Bench as a foundation for future research on strategic multimodal agents. Code and data are available at https://vs-bench.github.io.

Paperid: 1904, https://arxiv.org/pdf/2505.19789.pdf

Abstract:
Large Vision-Language Action (VLA) models have shown significant potential for embodied AI. However, their predominant training via supervised fine-tuning (SFT) limits generalization due to susceptibility to compounding errors under distribution shifts. Reinforcement learning (RL) offers a path to overcome these limitations by optimizing for task objectives via trial-and-error, yet a systematic understanding of its specific generalization benefits for VLAs compared to SFT is lacking. To address this, our study introduces a comprehensive benchmark for evaluating VLA generalization and systematically investigates the impact of RL fine-tuning across diverse visual, semantic, and execution dimensions. Our extensive experiments reveal that RL fine-tuning, particularly with PPO, significantly enhances generalization in semantic understanding and execution robustness over SFT, while maintaining comparable visual robustness. We identify PPO as a more effective RL algorithm for VLAs than LLM-derived methods like DPO and GRPO. We also develop a simple recipe for efficient PPO training on VLAs, and demonstrate its practical utility for improving VLA generalization. The project page is at https://rlvla.github.io

Paperid: 1905, https://arxiv.org/pdf/2410.22707.pdf

Abstract:
State recognition of the environment and objects, such as the open/closed state of doors and the on/off of lights, is indispensable for robots that perform daily life support and security tasks. Until now, state recognition methods have been based on training neural networks from manual annotations, preparing special sensors for the recognition, or manually programming to extract features from point clouds or raw images. In contrast, we propose a robotic state recognition method using a pre-trained vision-language model, which is capable of Image-to-Text Retrieval (ITR) tasks. We prepare several kinds of language prompts in advance, calculate the similarity between these prompts and the current image by ITR, and perform state recognition. By applying the optimal weighting to each prompt using black-box optimization, state recognition can be performed with higher accuracy. Experiments show that this theory enables a variety of state recognitions by simply preparing multiple prompts without retraining neural networks or manual programming. In addition, since only prompts and their weights need to be prepared for each recognizer, there is no need to prepare multiple models, which facilitates resource management. It is possible to recognize the open/closed state of transparent doors, the state of whether water is running or not from a faucet, and even the qualitative state of whether a kitchen is clean or not, which have been challenging so far, through language.

Paperid: 1906, https://arxiv.org/pdf/2509.17044.pdf

Abstract:
Accurate crop disease diagnosis is essential for sustainable agriculture and global food security. Existing methods, which primarily rely on unimodal models such as image-based classifiers and object detectors, are limited in their ability to incorporate domain-specific agricultural knowledge and lack support for interactive, language-based understanding. Recent advances in large language models (LLMs) and large vision-language models (LVLMs) have opened new avenues for multimodal reasoning. However, their performance in agricultural contexts remains limited due to the absence of specialized datasets and insufficient domain adaptation. In this work, we propose AgriDoctor, a modular and extensible multimodal framework designed for intelligent crop disease diagnosis and agricultural knowledge interaction. As a pioneering effort to introduce agent-based multimodal reasoning into the agricultural domain, AgriDoctor offers a novel paradigm for building interactive and domain-adaptive crop health solutions. It integrates five core components: a router, classifier, detector, knowledge retriever and LLMs. To facilitate effective training and evaluation, we construct AgriMM, a comprehensive benchmark comprising 400000 annotated disease images, 831 expert-curated knowledge entries, and 300000 bilingual prompts for intent-driven tool selection. Extensive experiments demonstrate that AgriDoctor, trained on AgriMM, significantly outperforms state-of-the-art LVLMs on fine-grained agricultural tasks, establishing a new paradigm for intelligent and sustainable farming applications.

Paperid: 1907, https://arxiv.org/pdf/2506.09965.pdf

Abstract:
As textual reasoning with large language models (LLMs) has advanced significantly, there has been growing interest in enhancing the multimodal reasoning capabilities of large vision-language models (LVLMs). However, existing methods primarily approach multimodal reasoning in a straightforward, text-centric manner, where both reasoning and answer derivation are conducted purely through text, with the only difference being the presence of multimodal input. As a result, these methods often encounter fundamental limitations in spatial reasoning tasks that demand precise geometric understanding and continuous spatial tracking-capabilities that humans achieve through mental visualization and manipulation. To address the limitations, we propose drawing to reason in space, a novel paradigm that enables LVLMs to reason through elementary drawing operations in the visual space. By equipping models with basic drawing operations, including annotating bounding boxes and drawing auxiliary lines, we empower them to express and analyze spatial relationships through direct visual manipulation, meanwhile avoiding the performance ceiling imposed by specialized perception tools in previous tool-integrated reasoning approaches. To cultivate this capability, we develop a three-stage training framework: cold-start training with synthetic data to establish basic drawing abilities, reflective rejection sampling to enhance self-reflection behaviors, and reinforcement learning to directly optimize for target rewards. Extensive experiments demonstrate that our model, named VILASR, consistently outperforms existing methods across diverse spatial reasoning benchmarks, involving maze navigation, static spatial reasoning, video-based reasoning, and multi-view-based reasoning tasks, with an average improvement of 18.4%.

Paperid: 1908, https://arxiv.org/pdf/2511.16917.pdf

Abstract:
We present UniModel, a unified generative model that jointly supports visual understanding and visual generation within a single pixel-to-pixel diffusion framework. Our goal is to achieve unification along three axes: the model, the tasks, and the representations. At the representation level, we eliminate modality discrepancies by mapping both text and images into a shared visual space: textual prompts are rendered as painted text images on a clean canvas, and all inputs and outputs are treated purely as RGB pixels. This yields a fully vision-native formulation of multimodal learning. At the task level, a broad range of vision-language problems are cast as pixel-to-pixel transformations in this visual space. For understanding tasks, the model takes an RGB image and produces a painted text image that visually encodes the semantic prediction. For generation tasks, painted text images serve as visual conditions that guide realistic and semantically aligned image synthesis. Captioning and text-to-image generation thus become different directions of the same underlying visual translation process. At the model level, we instantiate a single Unified Diffusion Transformer trained with rectified flow in pixel space. A shared backbone jointly learns bidirectional mappings between natural images and painted text images, with lightweight task embeddings to specify the desired direction. Experiments on text-to-image synthesis and image-to-text understanding demonstrate strong cross-modal alignment and emergent controllability such as cycle-consistent image-caption-image loops. Our initial exploration suggests that unifying model, tasks, and representations in a single visual space is a promising paradigm for general-purpose multimodal intelligence.

Paperid: 1909, https://arxiv.org/pdf/2509.02055.pdf

Abstract:
Vision-Language-Action (VLA) models pre-trained on large, diverse datasets show remarkable potential for general-purpose robotic manipulation. However, a primary bottleneck remains in adapting these models to downstream tasks, especially when the robot's embodiment or the task itself differs from the pre-training data. This discrepancy leads to a significant mismatch in action distributions, demanding extensive data and compute for effective fine-tuning. To address this challenge, we introduce \textbf{Align-Then-stEer (\texttt{ATE})}, a novel, data-efficient, and plug-and-play adaptation framework. \texttt{ATE} first aligns disparate action spaces by constructing a unified latent space, where a variational autoencoder constrained by reverse KL divergence embeds adaptation actions into modes of the pre-training action latent distribution. Subsequently, it steers the diffusion- or flow-based VLA's generation process during fine-tuning via a guidance mechanism that pushes the model's output distribution towards the target domain. We conduct extensive experiments on cross-embodiment and cross-task manipulation in both simulation and real world. Compared to direct fine-tuning of representative VLAs, our method improves the average multi-task success rate by up to \textbf{9.8\%} in simulation and achieves a striking \textbf{32\% success rate gain} in a real-world cross-embodiment setting. Our work presents a general and lightweight solution that greatly enhances the practicality of deploying VLA models to new robotic platforms and tasks.

Paperid: 1910, https://arxiv.org/pdf/2505.24156.pdf

Abstract:
Learning a generalizable bimanual manipulation policy is extremely challenging for embodied agents due to the large action space and the need for coordinated arm movements. Existing approaches rely on Vision-Language-Action (VLA) models to acquire bimanual policies. However, transferring knowledge from single-arm datasets or pre-trained VLA models often fails to generalize effectively, primarily due to the scarcity of bimanual data and the fundamental differences between single-arm and bimanual manipulation. In this paper, we propose a novel bimanual foundation policy by fine-tuning the leading text-to-video models to predict robot trajectories and training a lightweight diffusion policy for action generation. Given the lack of embodied knowledge in text-to-video models, we introduce a two-stage paradigm that fine-tunes independent text-to-flow and flow-to-video models derived from a pre-trained text-to-video model. Specifically, optical flow serves as an intermediate variable, providing a concise representation of subtle movements between images. The text-to-flow model predicts optical flow to concretize the intent of language instructions, and the flow-to-video model leverages this flow for fine-grained video prediction. Our method mitigates the ambiguity of language in single-stage text-to-video prediction and significantly reduces the robot-data requirement by avoiding direct use of low-level actions. In experiments, we collect high-quality manipulation data for real dual-arm robot, and the results of simulation and real-world experiments demonstrate the effectiveness of our method.

Paperid: 1911, https://arxiv.org/pdf/2512.13030.pdf

Abstract:
While a general embodied agent must function as a unified system, current methods are built on isolated models for understanding, world modeling, and control. This fragmentation prevents unifying multimodal generative capabilities and hinders learning from large-scale, heterogeneous data. In this paper, we propose Motus, a unified latent action world model that leverages existing general pretrained models and rich, sharable motion information. Motus introduces a Mixture-of-Transformer (MoT) architecture to integrate three experts (i.e., understanding, video generation, and action) and adopts a UniDiffuser-style scheduler to enable flexible switching between different modeling modes (i.e., world models, vision-language-action models, inverse dynamics models, video generation models, and video-action joint prediction models). Motus further leverages the optical flow to learn latent actions and adopts a recipe with three-phase training pipeline and six-layer data pyramid, thereby extracting pixel-level "delta action" and enabling large-scale action pretraining. Experiments show that Motus achieves superior performance against state-of-the-art methods in both simulation (a +15% improvement over X-VLA and a +45% improvement over Pi0.5) and real-world scenarios(improved by +11~48%), demonstrating unified modeling of all functionalities and priors significantly benefits downstream robotic tasks.

Paperid: 1912, https://arxiv.org/pdf/2511.04307.pdf

Abstract:
We introduce GUI-360$^\circ$, a large-scale, comprehensive dataset and benchmark suite designed to advance computer-using agents (CUAs). CUAs present unique challenges and is constrained by three persistent gaps: a scarcity of real-world CUA tasks, the lack of automated collection-and-annotation pipelines for multi-modal trajectories, and the absence of a unified benchmark that jointly evaluates GUI grounding, screen parsing, and action prediction. GUI-360$^\circ$ addresses these gaps with an LLM-augmented, largely automated pipeline for query sourcing, environment-template construction, task instantiation, batched execution, and LLM-driven quality filtering. The released corpus contains over 1.2M executed action steps across thousands of trajectories in popular Windows office applications, and includes full-resolution screenshots, accessibility metadata when available, instantiated goals, intermediate reasoning traces, and both successful and failed action trajectories. The dataset supports three canonical tasks, GUI grounding, screen parsing, and action prediction, and a hybrid GUI+API action space that reflects modern agent designs. Benchmarking state-of-the-art vision--language models on GUI-360$^\circ$ reveals substantial out-of-the-box shortcomings in grounding and action prediction; supervised fine-tuning and reinforcement learning yield significant gains but do not close the gap to human-level reliability. We release GUI-360$^\circ$ and accompanying code to facilitate reproducible research and accelerate progress on robust desktop CUAs. The full dataset has been made public on https://huggingface.co/datasets/vyokky/GUI-360.

Paperid: 1913, https://arxiv.org/pdf/2506.01616.pdf

Abstract:
The emergence of multimodal LLM-based agents (MLAs) has transformed interaction paradigms by seamlessly integrating vision, language, action and dynamic environments, enabling unprecedented autonomous capabilities across GUI applications ranging from web automation to mobile systems. However, MLAs introduce critical trustworthiness challenges that extend far beyond traditional language models' limitations, as they can directly modify digital states and trigger irreversible real-world consequences. Existing benchmarks inadequately tackle these unique challenges posed by MLAs' actionable outputs, long-horizon uncertainty and multimodal attack vectors. In this paper, we introduce MLA-Trust, the first comprehensive and unified framework that evaluates the MLA trustworthiness across four principled dimensions: truthfulness, controllability, safety and privacy. We utilize websites and mobile applications as realistic testbeds, designing 34 high-risk interactive tasks and curating rich evaluation datasets. Large-scale experiments involving 13 state-of-the-art agents reveal previously unexplored trustworthiness vulnerabilities unique to multimodal interactive scenarios. For instance, proprietary and open-source GUI-interacting MLAs pose more severe trustworthiness risks than static MLLMs, particularly in high-stakes domains; the transition from static MLLMs into interactive MLAs considerably compromises trustworthiness, enabling harmful content generation in multi-step interactions that standalone MLLMs would typically prevent; multi-step execution, while enhancing the adaptability of MLAs, involves latent nonlinear risk accumulation across successive interactions, circumventing existing safeguards and resulting in unpredictable derived risks. Moreover, we present an extensible toolbox to facilitate continuous evaluation of MLA trustworthiness across diverse interactive environments.

Paperid: 1914, https://arxiv.org/pdf/2505.00742.pdf

Abstract:
Recent advancements in multimodal large language models (MLLMs) have broadened the scope of vision-language tasks, excelling in applications like image captioning and interactive question-answering. However, these models struggle with accurately processing visual data, particularly in tasks requiring precise object recognition and fine visual details. Stringent token limits often result in the omission of critical information, hampering performance. To address these limitations, we introduce \SysName, a novel visual prompting mechanism designed to enhance MLLM performance while preserving essential visual details within token limits. \SysName features three key innovations: a prompt-aware strategy that dynamically highlights relevant image regions, a spatial-preserving orchestration schema that maintains object integrity, and a budget-aware prompting method that balances global context with crucial visual details. Comprehensive evaluations across multiple datasets demonstrate that \SysName consistently outperforms baseline methods, achieving up to a $26.9\%$ improvement in accuracy while significantly reducing token consumption.

Paperid: 1915, https://arxiv.org/pdf/2502.18906.pdf

Abstract:
Training Vision-Language Models (VLMs) for Graphical User Interfaces (GUI) agents via Reinforcement Learning (RL) faces critical challenges: environment-based RL requires costly interactions, while environment-free methods struggle with distribution shift and reward generalization. We propose an environment-free RL framework that decouples value estimation from policy optimization by leveraging a pretrained Value Environment Model (VEM). VEM predicts state-action values directly from offline data, distilling human-like priors about GUI interaction outcomes without requiring next-state prediction or environmental feedback. This avoids compounding errors and enhances resilience to UI changes by focusing on semantic reasoning (e.g., Does this action advance the user's goal?). The framework operates in two stages: (1) pretraining VEM to estimate long-term action utilities and (2) guiding policy exploration with frozen VEM signals, enabling layout-agnostic GUI automation. Evaluated on Android-in-the-Wild benchmarks, VEM achieves state-of-the-art performance in both offline and online settings, outperforming environment-free baselines significantly and matching environment-based approaches without interaction costs. Importantly, VEM demonstrates that semantic-aware value estimation can achieve comparable performance with online-trained methods.

Paperid: 1916, https://arxiv.org/pdf/2501.04561.pdf

Abstract:
Recent advancements in omnimodal learning have significantly improved understanding and generation across images, text, and speech, yet these developments remain predominantly confined to proprietary models. The lack of high-quality omnimodal datasets and the challenges of real-time emotional speech synthesis have notably hindered progress in open-source research. To address these limitations, we introduce \name, a two-stage training framework that integrates omnimodal alignment and speech generation to develop a state-of-the-art omnimodal large language model. In the alignment phase, a pre-trained speech model undergoes further training on text-image tasks, enabling (near) zero-shot generalization from vision to speech, outperforming models trained on tri-modal datasets. In the speech generation phase, a lightweight decoder is trained on speech tasks with direct preference optimization, enabling real-time emotional speech synthesis with high fidelity. Experiments show that \name surpasses state-of-the-art models across omnimodal, vision-language, and speech-language benchmarks. It achieves a 4-point absolute improvement on OmniBench over the leading open-source model VITA, despite using 5x fewer training samples and a smaller model size (7B vs. 7x8B). Additionally, \name achieves real-time speech generation with <1s latency at non-autoregressive mode, reducing inference time by 5x compared to autoregressive methods, and improves emotion classification accuracy by 7.7\%

Paperid: 1917, https://arxiv.org/pdf/2411.08768.pdf

Abstract:
Video recordings of user activities, particularly desktop recordings, offer a rich source of data for understanding user behaviors and automating processes. However, despite advancements in Vision-Language Models (VLMs) and their increasing use in video analysis, extracting user actions from desktop recordings remains an underexplored area. This paper addresses this gap by proposing two novel VLM-based methods for user action extraction: the Direct Frame-Based Approach (DF), which inputs sampled frames directly into VLMs, and the Differential Frame-Based Approach (DiffF), which incorporates explicit frame differences detected via computer vision techniques. We evaluate these methods using a basic self-curated dataset and an advanced benchmark adapted from prior work. Our results show that the DF approach achieves an accuracy of 70% to 80% in identifying user actions, with the extracted action sequences being re-playable though Robotic Process Automation. We find that while VLMs show potential, incorporating explicit UI changes can degrade performance, making the DF approach more reliable. This work represents the first application of VLMs for extracting user action sequences from desktop recordings, contributing new methods, benchmarks, and insights for future research.

Paperid: 1918, https://arxiv.org/pdf/2409.05840.pdf

Abstract:
The development of Multimodal Large Language Models (MLLMs) has seen significant advancements with increasing demands in various fields (e.g., multimodal agents, embodied intelligence). While model-driven approaches attempt to enhance MLLMs capabilities through diverse architectures, the gains have become increasingly marginal. Conversely, data-driven methods, which scale up image-text instruction data, are more effective but face limited data diversity and complexity challenges. The absence of high-quality data constitutes a significant development barrier for MLLMs. To address the data quality bottleneck, we propose MMEvol, a novel multimodal instruction data evolution framework. This framework iteratively improve data quality through a refined combination of fine-grained perception, cognitive reasoning, and interaction evolution, generating a more complex and diverse image-text instruction dataset that empowers MLLMs with enhanced capabilities. Beginning with an initial set of instructions, SEED-163K, we utilize MMEvol to systematically broaden the diversity of instruction types, extend visual reasoning steps to improve cognitive reasoning abilities, and thoroughly explore fine-grained information within images to enhance visual understanding and robustness. To comprehensively evaluate the effectiveness of our approach, we conduct extensive qualitative analysis and quantitative experiments across 13 vision-language tasks. Compared to baseline models trained with the initial seed data, the results demonstrate that our method achieves an average accuracy improvement of 3.1 percentage points. Furthermore, our approach reaches state-of-the-art (SOTA) performance in nine tasks using significantly less data compared to state-of-the-art models.

Paperid: 1919, https://arxiv.org/pdf/2503.15369.pdf

Abstract:
While multimodal large language models demonstrate strong performance in complex reasoning tasks, they pose significant challenges related to model complexity during deployment, especially for resource-limited devices. In this paper, we propose an automatic pruning method for large vision-language models to enhance the efficiency of multimodal reasoning. Conventional methods rely on the training data of the original model to select the proper pruning ratio for different network components. However, these methods are impractical for large vision-language models due to the unaffordable search costs caused by web-scale training corpus. In contrast, our approach only leverages a small number of samples to search for the desired pruning policy by maximizing its generalization ability on unknown training data while maintaining the model accuracy, which enables the achievement of an optimal trade-off between accuracy and efficiency for large visual language models. Specifically, we formulate the generalization gap of the pruning strategy using the structural risk minimization principle. Based on both task performance and generalization capability, we iteratively search for the optimal pruning policy within a given search space and optimize the vision projector to evolve the search space with higher upper bound of performance. We conduct extensive experiments on the ScienceQA, Vizwiz, MM-vet, and LLaVA-Bench datasets for the task of visual question answering. Using only 64 samples for pruning policy search, EfficientLLaVA achieves an accuracy of 83.05% on ScienceQA, along with a $\times$ 1.8 speedup compared to the dense LLaVA-v1.5-7B model.

Paperid: 1920, https://arxiv.org/pdf/2601.21610.pdf

Abstract:
Digital watermarking is essential for securing generated images from diffusion models. Accurate watermark evaluation is critical for algorithm development, yet existing methods have significant limitations: they lack a unified framework for both residual and semantic watermarks, provide results without interpretability, neglect comprehensive security considerations, and often use inappropriate metrics for semantic watermarks. To address these gaps, we propose WMVLM, the first unified and interpretable evaluation framework for diffusion model image watermarking via vision-language models (VLMs). We redefine quality and security metrics for each watermark type: residual watermarks are evaluated by artifact strength and erasure resistance, while semantic watermarks are assessed through latent distribution shifts. Moreover, we introduce a three-stage training strategy to progressively enable the model to achieve classification, scoring, and interpretable text generation. Experiments show WMVLM outperforms state-of-the-art VLMs with strong generalization across datasets, diffusion models, and watermarking methods.

Paperid: 1921, https://arxiv.org/pdf/2512.05111.pdf

Abstract:
Reward models are critical for aligning vision-language systems with human preferences, yet current approaches suffer from hallucination, weak visual grounding, and an inability to use tools for verification, limiting their reliability on complex multimodal reasoning tasks. We present ARM-Thinker, an A}gentic multimodal Reward Model that autonomously invokes external tools (e.g., image cropping, doc page retrieval) to ground judgments in verifiable evidence, replacing static, non-interactive reward scoring. This enables the model to verify fine-grained visual details, cross-reference multi-page evidence, and validate reasoning claims, which are capabilities absent in existing reward models. We train ARM-Thinker with multi-stage reinforcement learning, jointly optimizing tool-calling decisions and judgment accuracy. To evaluate agentic reward modeling, we introduce ARMBench-VL, comprising three benchmarks that assess fine-grained visual grounding (image-level tools), multi-page document understanding (retrieval tools), and instruction following (text-level verification). ARM-Thinker achieves +16.2% average improvement on reward modeling benchmarks, +9.6% on tool-use tasks, and outperforms baselines on multimodal math and logical reasoning benchmarks. Our results demonstrate that agentic capabilities significantly enhance both accuracy and interpretability of reward models.

Paperid: 1922, https://arxiv.org/pdf/2510.27606.pdf

Abstract:
Spatial understanding remains a weakness of Large Vision-Language Models (LVLMs). Existing supervised fine-tuning (SFT) and recent reinforcement learning with verifiable rewards (RLVR) pipelines depend on costly supervision, specialized tools, or constrained environments that limit scale. We introduce Spatial-SSRL, a self-supervised RL paradigm that derives verifiable signals directly from ordinary RGB or RGB-D images. Spatial-SSRL automatically formulates five pretext tasks that capture 2D and 3D spatial structure: shuffled patch reordering, flipped patch recognition, cropped patch inpainting, regional depth ordering, and relative 3D position prediction. These tasks provide ground-truth answers that are easy to verify and require no human or LVLM annotation. Training on our tasks substantially improves spatial reasoning while preserving general visual capabilities. On seven spatial understanding benchmarks in both image and video settings, Spatial-SSRL delivers average accuracy gains of 4.63% (3B) and 3.89% (7B) over the Qwen2.5-VL baselines. Our results show that simple, intrinsic supervision enables RLVR at scale and provides a practical route to stronger spatial intelligence in LVLMs.

Paperid: 1923, https://arxiv.org/pdf/2509.23838.pdf

Abstract:
Semi-supervised Video Object Segmentation aims to segment a specified target throughout a video sequence, initialized by a first-frame mask. Previous methods rely heavily on appearance-based pattern matching and thus exhibit limited robustness against challenges such as drastic visual changes, occlusions, and scene shifts. This failure is often attributed to a lack of high-level conceptual understanding of the target. The recently proposed Segment Concept (SeC) framework mitigated this limitation by using a Large Vision-Language Model (LVLM) to establish a deep semantic understanding of the object for more persistent segmentation. In this work, we evaluate its zero-shot performance on the challenging coMplex video Object SEgmentation v2 (MOSEv2) dataset. Without any fine-tuning on the training set, SeC achieved 39.7 \JFn on the test set and ranked 2nd place in the Complex VOS track of the 7th Large-scale Video Object Segmentation Challenge.

Paperid: 1924, https://arxiv.org/pdf/2508.19652.pdf

Abstract:
Vision-Language Models (VLMs) often suffer from visual hallucinations, saying things that are not actually in the image, and language shortcuts, where they skip the visual part and just rely on text priors. These issues arise because most post-training methods for VLMs rely on simple verifiable answer matching and supervise only final outputs, leaving intermediate visual reasoning without explicit guidance. As a result, VLMs receive sparse visual signals and often learn to prioritize language-based reasoning over visual perception. To mitigate this, some existing methods add visual supervision using human annotations or distilled labels from external large models. However, human annotations are labor-intensive and costly, and because external signals cannot adapt to the evolving policy, they cause distributional shifts that can lead to reward hacking. In this paper, we introduce Vision-SR1, a self-rewarding method that improves visual reasoning without relying on external visual supervisions via reinforcement learning. Vision-SR1 decomposes VLM reasoning into two stages: visual perception and language reasoning. The model is first prompted to produce self-contained visual perceptions that are sufficient to answer the question without referring back the input image. To validate this self-containment, the same VLM model is then re-prompted to perform language reasoning using only the generated perception as input to compute reward. This self-reward is combined with supervision on final outputs, providing a balanced training signal that strengthens both visual perception and language reasoning. Our experiments demonstrate that Vision-SR1 improves visual reasoning, mitigates visual hallucinations, and reduces reliance on language shortcuts across diverse vision-language tasks.

Paperid: 1925, https://arxiv.org/pdf/2506.08052.pdf

Abstract:
Although end-to-end autonomous driving has made remarkable progress, its performance degrades significantly in rare and long-tail scenarios. Recent approaches attempt to address this challenge by leveraging the rich world knowledge of Vision-Language Models (VLMs), but these methods suffer from several limitations: (1) a significant domain gap between the pre-training data of VLMs and real-world driving data, (2) a dimensionality mismatch between the discrete language space and the continuous action space, and (3) imitation learning tends to capture the average behavior present in the dataset, which may be suboptimal even dangerous. In this paper, we propose ReCogDrive, an autonomous driving system that integrates VLMs with diffusion planner, which adopts a three-stage paradigm for training. In the first stage, we use a large-scale driving question-answering datasets to train the VLMs, mitigating the domain discrepancy between generic content and real-world driving scenarios. In the second stage, we employ a diffusion-based planner to perform imitation learning, mapping representations from the latent language space to continuous driving actions. Finally, we fine-tune the diffusion planner using reinforcement learning with NAVSIM non-reactive simulator, enabling the model to generate safer, more human-like driving trajectories. We evaluate our approach on the planning-oriented NAVSIM benchmark, achieving a PDMS of 89.6 and setting a new state-of-the-art that surpasses the previous vision-only SOTA by 5.6 PDMS.

Paperid: 1926, https://arxiv.org/pdf/2506.07497.pdf

Abstract:
We present Genesis, a unified framework for joint generation of multi-view driving videos and LiDAR sequences with spatio-temporal and cross-modal consistency. Genesis employs a two-stage architecture that integrates a DiT-based video diffusion model with 3D-VAE encoding, and a BEV-aware LiDAR generator with NeRF-based rendering and adaptive sampling. Both modalities are directly coupled through a shared latent space, enabling coherent evolution across visual and geometric domains. To guide the generation with structured semantics, we introduce DataCrafter, a captioning module built on vision-language models that provides scene-level and instance-level supervision. Extensive experiments on the nuScenes benchmark demonstrate that Genesis achieves state-of-the-art performance across video and LiDAR metrics (FVD 16.95, FID 4.24, Chamfer 0.611), and benefits downstream tasks including segmentation and 3D detection, validating the semantic fidelity and practical utility of the generated data.

Paperid: 1927, https://arxiv.org/pdf/2501.02955.pdf

Abstract:
In recent years, vision language models (VLMs) have made significant advancements in video understanding. However, a crucial capability - fine-grained motion comprehension - remains under-explored in current benchmarks. To address this gap, we propose MotionBench, a comprehensive evaluation benchmark designed to assess the fine-grained motion comprehension of video understanding models. MotionBench evaluates models' motion-level perception through six primary categories of motion-oriented question types and includes data collected from diverse sources, ensuring a broad representation of real-world video content. Experimental results reveal that existing VLMs perform poorly in understanding fine-grained motions. To enhance VLM's ability to perceive fine-grained motion within a limited sequence length of LLM, we conduct extensive experiments reviewing VLM architectures optimized for video feature compression and propose a novel and efficient Through-Encoder (TE) Fusion method. Experiments show that higher frame rate inputs and TE Fusion yield improvements in motion understanding, yet there is still substantial room for enhancement. Our benchmark aims to guide and motivate the development of more capable video understanding models, emphasizing the importance of fine-grained motion comprehension. Project page: https://motion-bench.github.io .

Paperid: 1928, https://arxiv.org/pdf/2511.21631.pdf

Authors:Shuai Bai, Yuxuan Cai, Ruizhe Chen, Keqin Chen, Xionghui Chen, Zesen Cheng, Lianghao Deng, Wei Ding, Chang Gao, Chunjiang Ge, Wenbin Ge, Zhifang Guo, Qidong Huang, Jie Huang, Fei Huang, Binyuan Hui, Shutong Jiang, Zhaohai Li, Mingsheng Li, Mei Li, Kaixin Li, Zicheng Lin, Junyang Lin, Xuejing Liu, Jiawei Liu, Chenglong Liu, Yang Liu, Dayiheng Liu, Shixuan Liu, Dunjie Lu, Ruilin Luo, Chenxu Lv, Rui Men, Lingchen Meng, Xuancheng Ren, Xingzhang Ren, Sibo Song, Yuchong Sun, Jun Tang, Jianhong Tu, Jianqiang Wan, Peng Wang, Pengfei Wang, Qiuyue Wang, Yuxuan Wang, Tianbao Xie, Yiheng Xu, Haiyang Xu, Jin Xu, Zhibo Yang, Mingkun Yang, Jianxin Yang, An Yang, Bowen Yu, Fei Zhang, Hang Zhang, Xi Zhang, Bo Zheng, Humen Zhong, Jingren Zhou, Fan Zhou, Jing Zhou, Yuanzhi Zhu, Ke Zhu

Abstract:
We introduce Qwen3-VL, the most capable vision-language model in the Qwen series to date, achieving superior performance across a broad range of multimodal benchmarks. It natively supports interleaved contexts of up to 256K tokens, seamlessly integrating text, images, and video. The model family includes both dense (2B/4B/8B/32B) and mixture-of-experts (30B-A3B/235B-A22B) variants to accommodate diverse latency-quality trade-offs. Qwen3-VL delivers three core pillars: (i) markedly stronger pure-text understanding, surpassing comparable text-only backbones in several cases; (ii) robust long-context comprehension with a native 256K-token window for both text and interleaved multimodal inputs, enabling faithful retention, retrieval, and cross-referencing across long documents and videos; and (iii) advanced multimodal reasoning across single-image, multi-image, and video tasks, demonstrating leading performance on comprehensive evaluations such as MMMU and visual-math benchmarks (e.g., MathVista and MathVision). Architecturally, we introduce three key upgrades: (i) an enhanced interleaved-MRoPE for stronger spatial-temporal modeling across images and video; (ii) DeepStack integration, which effectively leverages multi-level ViT features to tighten vision-language alignment; and (iii) text-based time alignment for video, evolving from T-RoPE to explicit textual timestamp alignment for more precise temporal grounding. Under comparable token budgets and latency constraints, Qwen3-VL achieves superior performance in both dense and Mixture-of-Experts (MoE) architectures. We envision Qwen3-VL serving as a foundational engine for image-grounded reasoning, agentic decision-making, and multimodal code intelligence in real-world workflows.

Paperid: 1929, https://arxiv.org/pdf/2601.14339.pdf

Abstract:
Cross-view spatial reasoning is essential for embodied AI, underpinning spatial understanding, mental simulation and planning in complex environments. Existing benchmarks primarily emphasize indoor or street settings, overlooking the unique challenges of open-ended urban spaces characterized by rich semantics, complex geometries, and view variations. To address this, we introduce CityCube, a systematic benchmark designed to probe cross-view reasoning capabilities of current VLMs in urban settings. CityCube integrates four viewpoint dynamics to mimic camera movements and spans a wide spectrum of perspectives from multiple platforms, e.g., vehicles, drones and satellites. For a comprehensive assessment, it features 5,022 meticulously annotated multi-view QA pairs categorized into five cognitive dimensions and three spatial relation expressions. A comprehensive evaluation of 33 VLMs reveals a significant performance disparity with humans: even large-scale models struggle to exceed 54.1% accuracy, remaining 34.2% below human performance. By contrast, small-scale fine-tuned VLMs achieve over 60.0% accuracy, highlighting the necessity of our benchmark. Further analyses indicate the task correlations and fundamental cognitive disparity between VLMs and human-like reasoning.

Paperid: 1930, https://arxiv.org/pdf/2507.15428.pdf

Abstract:
Egomotion videos are first-person recordings where the view changes continuously due to the agent's movement. As they serve as the primary visual input for embodied AI agents, making egomotion video reasoning more efficient is therefore essential for real-world deployment. Recent advances in vision-language models have enabled strong multimodal reasoning capabilities, but their computational cost remains prohibitive for long, redundant video inputs. Existing token pruning methods, typically designed for third-person videos, fail to leverage the spatiotemporal continuity and motion constraints inherent in egomotion settings. To address this, we propose EgoPrune, a training-free token pruning method tailored for egomotion video reasoning. EgoPrune comprises three components: a keyframe selector adapted from EmbodiedR for temporally efficient sampling; Perspective-Aware Redundancy Filtering (PARF), which aligns visual tokens using perspective transformations and removes redundant tokens; and a Maximal Marginal Relevance (MMR)-based token selector that jointly considers visual-text relevance and intra-frame diversity. Experiments on two egomotion video benchmarks show that EgoPrune consistently outperforms prior training-free methods across various pruning ratios while significantly reducing FLOPs, memory usage, and latency. Moreover, we deploy EgoPrune on an embodied agent equipped with a Jetson Orin NX 16GB edge device, demonstrating its real-world efficiency and suitability for on-device egomotion video reasoning.

Paperid: 1931, https://arxiv.org/pdf/2505.22038.pdf

Abstract:
Large Vision-Language Models (LVLMs) have shown impressive performance across multi-modal tasks by encoding images into thousands of tokens. However, the large number of image tokens results in significant computational overhead, and the use of dynamic high-resolution inputs further increases this burden. Previous approaches have attempted to reduce the number of image tokens through token pruning, typically by selecting tokens based on attention scores or image token diversity. Through empirical studies, we observe that existing methods often overlook the joint impact of pruning on both the current layer's output (local) and the outputs of subsequent layers (global), leading to suboptimal pruning decisions. To address this challenge, we propose Balanced Token Pruning (BTP), a plug-and-play method for pruning vision tokens. Specifically, our method utilizes a small calibration set to divide the pruning process into multiple stages. In the early stages, our method emphasizes the impact of pruning on subsequent layers, whereas in the deeper stages, the focus shifts toward preserving the consistency of local outputs. Extensive experiments across various LVLMs demonstrate the broad effectiveness of our approach on multiple benchmarks. Our method achieves a 78% compression rate while preserving 96.7% of the original models' performance on average.

Paperid: 1932, https://arxiv.org/pdf/2504.12680.pdf

Abstract:
Humans can perceive and reason about spatial relationships from sequential visual observations, such as egocentric video streams. However, how pretrained models acquire such abilities, especially high-level reasoning, remains unclear. This paper introduces Embodied-R, a collaborative framework combining large-scale Vision-Language Models (VLMs) for perception and small-scale Language Models (LMs) for reasoning. Using Reinforcement Learning (RL) with a novel reward system considering think-answer logical consistency, the model achieves slow-thinking capabilities with limited computational resources. After training on only 5k embodied video samples, Embodied-R with a 3B LM matches state-of-the-art multimodal reasoning models (OpenAI-o1, Gemini-2.5-pro) on both in-distribution and out-of-distribution embodied spatial reasoning tasks. Embodied-R also exhibits emergent thinking patterns such as systematic analysis and contextual integration. We further explore research questions including response length, training on VLM, strategies for reward design, and differences in model generalization after SFT (Supervised Fine-Tuning) and RL training.

Paperid: 1933, https://arxiv.org/pdf/2503.06157.pdf

Abstract:
Large multimodal models exhibit remarkable intelligence, yet their embodied cognitive abilities during motion in open-ended urban 3D space remain to be explored. We introduce a benchmark to evaluate whether video-large language models (Video-LLMs) can naturally process continuous first-person visual observations like humans, enabling recall, perception, reasoning, and navigation. We have manually control drones to collect 3D embodied motion video data from real-world cities and simulated environments, resulting in 1.5k video clips. Then we design a pipeline to generate 5.2k multiple-choice questions. Evaluations of 17 widely-used Video-LLMs reveal current limitations in urban embodied cognition. Correlation analysis provides insight into the relationships between different tasks, showing that causal reasoning has a strong correlation with recall, perception, and navigation, while the abilities for counterfactual and associative reasoning exhibit lower correlation with other tasks. We also validate the potential for Sim-to-Real transfer in urban embodiment through fine-tuning.

Paperid: 1934, https://arxiv.org/pdf/2502.14504.pdf

Abstract:
Large Vision-Language Models (LVLMs) have demonstrated remarkable capabilities across a range of multimodal tasks. However, their inference efficiency is constrained by the large number of visual tokens processed during decoding. To address this challenge, we propose Per-Layer Per-Head Vision Token Pruning (PLPHP), a two-level fine-grained pruning method including Layer-Level Retention Rate Allocation and Head-Level Vision Token Pruning. Motivated by the Vision Token Re-attention phenomenon across decoder layers, we dynamically adjust token retention rates layer by layer. Layers that exhibit stronger attention to visual information preserve more vision tokens, while layers with lower vision attention are aggressively pruned. Furthermore, PLPHP applies pruning at the attention head level, enabling different heads within the same layer to independently retain critical context. Experiments on multiple benchmarks demonstrate that PLPHP delivers an 18% faster decoding speed and reduces the Key-Value Cache (KV Cache) size by over 50%, all at the cost of 0.46% average performance drop, while also achieving notable performance improvements in multi-image tasks. These results highlight the effectiveness of fine-grained token pruning and contribute to advancing the efficiency and scalability of LVLMs. Our source code will be made publicly available.

Paperid: 1935, https://arxiv.org/pdf/2510.03896.pdf

Abstract:
Although Vision-Language Models (VLM) have demonstrated impressive planning and reasoning capabilities, translating these abilities into the physical world introduces significant challenges. Conventional Vision-Language-Action (VLA) models, which integrate reasoning and action into a monolithic architecture, generalize poorly because they are constrained by scarce, narrow-domain data. While recent dual-system approaches attempt to decouple "thinking" from "acting", they are often constrained by semantic ambiguities within the action module. This ambiguity makes large-scale, cross-task training infeasible. Consequently, these systems typically necessitate fine-tuning on newly collected data when deployed to novel environments, and the cooperation mechanism between the two systems remains ill-defined. To address these limitations, we introduce, for the first time, a framework centered around a generalizable action expert. Our approach utilizes sparse 3D trajectories as an intermediate representation, effectively bridging the high-level planning capabilities of the VLM with the low-level physical action module. During the planning phase, the VLM is only required to generate coarse 3D waypoints. These waypoints are then processed by our generalizable action expert, which refines them into dense, executable action sequences by sampling real-time point cloud observations of the environment. To promote training efficiency and robust generalization, we introduce a novel "Action Pre-training, Pointcloud Fine-tuning" paradigm. Our method combines the broad generalization capabilities of VLMs in visual understanding and planning with the fine-grained, action-level generalization of action expert.

Paperid: 1936, https://arxiv.org/pdf/2510.03895.pdf

Abstract:
Vision-Language-Action (VLA) models represent a pivotal advance in embodied intelligence, yet they confront critical barriers to real-world deployment, most notably catastrophic forgetting. This issue stems from their overreliance on continuous action sequences or action chunks, which inadvertently create isolated data silos that disrupt knowledge retention across tasks. To tackle these challenges, we propose the Narrowing of Trajectory VLA (NoTVLA) framework: a novel approach that narrows its focus to sparse trajectories, thereby avoiding the catastrophic forgetting associated with dense trajectory fine-tuning. A key innovation of NoTVLA lies in its trajectory planning strategy: instead of centering on the target object's trajectory, it leverages temporal compression and spatial reasoning pruning specifically for the robot end effector's trajectory. Furthermore, training is conducted using these sparse trajectories rather than dense action trajectories, an optimization that delivers remarkable practical advantages with better performance in zero-shot. In multi-task evaluation scenarios, NoTVLA achieves superior performance and generalization compared to pi0 while operating under two critical constraints: it uses over an order of magnitude less computing power than pi0 and requires no wrist-mounted camera. This design ensures that NoTVLA's operational accuracy closely approximates that of single-task expert models. Crucially, it also preserves the model's inherent language capabilities, enabling zero-shot generalization in specific scenarios, supporting unified model deployment across multiple robot platforms, and fostering a degree of generalization even when perceiving tasks from novel perspectives.

Paperid: 1937, https://arxiv.org/pdf/2412.16583.pdf

Abstract:
The rapid evolution of Vision Language Models (VLMs) has catalyzed significant advancements in artificial intelligence, expanding research across various disciplines, including Earth Observation (EO). While VLMs have enhanced image understanding and data processing within EO, their applications have predominantly focused on image content description. This limited focus overlooks their potential in geographic and scientific regression tasks, which are essential for diverse EO applications. To bridge this gap, this paper introduces a novel benchmark dataset, called \textbf{REO-Instruct} to unify regression and generation tasks specifically for the EO domain. Comprising 1.6 million multimodal EO imagery and language pairs, this dataset is designed to support both biomass regression and image content interpretation tasks. Leveraging this dataset, we develop \textbf{REO-VLM}, a groundbreaking model that seamlessly integrates regression capabilities with traditional generative functions. By utilizing language-driven reasoning to incorporate scientific domain knowledge, REO-VLM goes beyond solely relying on EO imagery, enabling comprehensive interpretation of complex scientific attributes from EO data. This approach establishes new performance benchmarks and significantly enhances the capabilities of environmental monitoring and resource management.

Paperid: 1938, https://arxiv.org/pdf/2409.09564.pdf

Abstract:
Currently, inspired by the success of vision-language models (VLMs), an increasing number of researchers are focusing on improving VLMs and have achieved promising results. However, most existing methods concentrate on optimizing the connector and enhancing the language model component, while neglecting improvements to the vision encoder itself. In contrast, we propose Text Guided LLaVA (TG-LLaVA) in this paper, which optimizes VLMs by guiding the vision encoder with text, offering a new and orthogonal optimization direction. Specifically, inspired by the purpose-driven logic inherent in human behavior, we use learnable latent embeddings as a bridge to analyze textual instruction and add the analysis results to the vision encoder as guidance, refining it. Subsequently, another set of latent embeddings extracts additional detailed text-guided information from high-resolution local patches as auxiliary information. Finally, with the guidance of text, the vision encoder can extract text-related features, similar to how humans focus on the most relevant parts of an image when considering a question. This results in generating better answers. Experiments on various datasets validate the effectiveness of the proposed method. Remarkably, without the need for additional training data, our propsoed method can bring more benefits to the baseline (LLaVA-1.5) compared with other concurrent methods. Furthermore, the proposed method consistently brings improvement in different settings.

Paperid: 1939, https://arxiv.org/pdf/2511.00076.pdf

Abstract:
While Vision-language Models (VLMs) have demonstrated strong semantic capabilities, their ability to interpret the underlying geometric structure of visual information is less explored. Pictographic characters, which combine visual form with symbolic structure, provide an ideal test case for this capability. We formulate this visual recognition challenge in the mathematical domain, where each character is represented by an executable program of geometric primitives. This is framed as a program synthesis task, training a VLM to decompile raster images into programs composed of Bézier curves. Our model, acting as a "visual decompiler", demonstrates performance superior to strong zero-shot baselines, including GPT-4o. The most significant finding is that when trained solely on modern Chinese characters, the model is able to reconstruct ancient Oracle Bone Script in a zero-shot context. This generalization provides strong evidence that the model acquires an abstract and transferable geometric grammar, moving beyond pixel-level pattern recognition to a more structured form of visual understanding.

Paperid: 1940, https://arxiv.org/pdf/2507.08710.pdf

Abstract:
We propose a novel embedding-based captioning metric termed as L-CLIPScore that can be used for efficiently evaluating caption quality and training captioning model. L-CLIPScore is calculated from a lightweight CLIP (L-CLIP), which is a dual-encoder architecture compressed and distilled from CLIP. To compress, we apply two powerful techniques which are weight multiplexing and matrix decomposition for reducing the parameters of encoders and word embedding matrix, respectively. To distill, we design a novel multi-modal Similarity Regulator (SR) loss to transfer more vision-language alignment knowledge. Specifically, SR loss amplifies the multi-modal embedding similarity if the given image-text pair is matched and diminishes the similarity if the pair is non-matched. By compressing and distilling by this novel SR loss, our L-CLIP achieves comparable multi-modal alignment ability to the original CLIP while it requires fewer computation resources and running time. We carry out exhaustive experiments to validate the efficiency and effectiveness of L-CLIPScore when using it as the judge to evaluate caption quality. We also discover that when using L-CLIPScore as the supervisor to train the captioning model, it should be mixed up by an n-gram-based metric and meanwhile analyze why using L-CLIPScore only will cause fail training.

Paperid: 1941, https://arxiv.org/pdf/2506.17046.pdf

Abstract:
Multimodal Large Language Models (MLLMs) have demonstrated significant advances across numerous vision-language tasks. Due to their strong image-text alignment capability, MLLMs can effectively understand image-text pairs with clear meanings. However, effectively resolving the inherent ambiguities in natural language and visual contexts remains challenging. Existing multimodal benchmarks typically overlook linguistic and visual ambiguities, relying mainly on unimodal context for disambiguation and thus failing to exploit the mutual clarification potential between modalities. To bridge this gap, we introduce MUCAR, a novel and challenging benchmark designed explicitly for evaluating multimodal ambiguity resolution across multilingual and cross-modal scenarios. MUCAR includes: (1) a multilingual dataset where ambiguous textual expressions are uniquely resolved by corresponding visual contexts, and (2) a dual-ambiguity dataset that systematically pairs ambiguous images with ambiguous textual contexts, with each combination carefully constructed to yield a single, clear interpretation through mutual disambiguation. Extensive evaluations involving 19 state-of-the-art multimodal models--encompassing both open-source and proprietary architectures--reveal substantial gaps compared to human-level performance, highlighting the need for future research into more sophisticated cross-modal ambiguity comprehension methods, further pushing the boundaries of multimodal reasoning.

Paperid: 1942, https://arxiv.org/pdf/2503.23733.pdf

Abstract:
Recently, model merging methods have demonstrated powerful strengths in combining abilities on various tasks from multiple Large Language Models (LLMs). While previous model merging methods mainly focus on merging homogeneous models with identical architecture, they meet challenges when dealing with Multimodal Large Language Models (MLLMs) with inherent heterogeneous property, including differences in model architecture and the asymmetry in the parameter space. In this work, we propose AdaMMS, a novel model merging method tailored for heterogeneous MLLMs. Our method tackles the challenges in three steps: mapping, merging and searching. Specifically, we first design mapping function between models to apply model merging on MLLMs with different architecture. Then we apply linear interpolation on model weights to actively adapt the asymmetry in the heterogeneous MLLMs. Finally in the hyper-parameter searching step, we propose an unsupervised hyper-parameter selection method for model merging. As the first model merging method capable of merging heterogeneous MLLMs without labeled data, extensive experiments on various model combinations demonstrated that AdaMMS outperforms previous model merging methods on various vision-language benchmarks.

Paperid: 1943, https://arxiv.org/pdf/2503.14161.pdf

Abstract:
Vision-Language Models (VLMs) have recently witnessed significant progress in visual comprehension. As the permitting length of image context grows, VLMs can now comprehend a broader range of views and spaces. Current benchmarks provide insightful analysis of VLMs in tasks involving complex visual instructions following, multi-image understanding and spatial reasoning. However, they usually focus on spatially irrelevant images or discrete images captured from varied viewpoints. The compositional characteristic of images captured from a static viewpoint remains underestimated. We term this characteristic as Continuous Space Perception. When observing a scene from a static viewpoint while shifting orientations, it produces a series of spatially continuous images, enabling the reconstruction of the entire space. In this paper, we present CoSpace, a multi-image visual understanding benchmark designed to assess the Continuous Space perception ability for VLMs. CoSpace contains 2,918 images and 1,626 question-answer pairs, covering seven types of tasks. We conduct evaluation across 19 proprietary and open-source VLMs. Results reveal that there exist pitfalls on the continuous space perception ability for most of the evaluated models, including proprietary ones. Interestingly, we find that the main discrepancy between open-source and proprietary models lies not in accuracy but in the consistency of responses. We believe that enhancing the ability of continuous space perception is essential for VLMs to perform effectively in real-world tasks and encourage further research to advance this capability.

Paperid: 1944, https://arxiv.org/pdf/2510.11027.pdf

Abstract:
While significant research has focused on developing embodied reasoning capabilities using Vision-Language Models (VLMs) or integrating advanced VLMs into Vision-Language-Action (VLA) models for end-to-end robot control, few studies directly address the critical gap between upstream VLM-based reasoning and downstream VLA policy learning. In this work, we take an initial step toward bridging embodied reasoning with VLA policy learning by introducing Vlaser - a Vision-Language-Action Model with synergistic embodied reasoning capability, which is a foundational vision-language model designed to integrate high-level reasoning with low-level control for embodied agents. Built upon the high-quality Vlaser-6M dataset, Vlaser achieves state-of-the-art performance across a range of embodied reasoning benchmarks - including spatial reasoning, embodied grounding, embodied QA, and task planning. Furthermore, we systematically examine how different VLM initializations affect supervised VLA fine-tuning, offering novel insights into mitigating the domain shift between internet-scale pre-training data and embodied-specific policy learning data. Based on these insights, our approach achieves state-of-the-art results on the WidowX benchmark and competitive performance on the Google Robot benchmark.

Paperid: 1945, https://arxiv.org/pdf/2508.18265.pdf

Authors:Weiyun Wang, Zhangwei Gao, Lixin Gu, Hengjun Pu, Long Cui, Xingguang Wei, Zhaoyang Liu, Linglin Jing, Shenglong Ye, Jie Shao, Zhaokai Wang, Zhe Chen, Hongjie Zhang, Ganlin Yang, Haomin Wang, Qi Wei, Jinhui Yin, Wenhao Li, Erfei Cui, Guanzhou Chen, Zichen Ding, Changyao Tian, Zhenyu Wu, Jingjing Xie, Zehao Li, Bowen Yang, Yuchen Duan, Xuehui Wang, Zhi Hou, Haoran Hao, Tianyi Zhang, Songze Li, Xiangyu Zhao, Haodong Duan, Nianchen Deng, Bin Fu, Yinan He, Yi Wang, Conghui He, Botian Shi, Junjun He, Yingtong Xiong, Han Lv, Lijun Wu, Wenqi Shao, Kaipeng Zhang, Huipeng Deng, Biqing Qi, Jiaye Ge, Qipeng Guo, Wenwei Zhang, Songyang Zhang, Maosong Cao, Junyao Lin, Kexian Tang, Jianfei Gao, Haian Huang, Yuzhe Gu, Chengqi Lyu, Huanze Tang, Rui Wang, Haijun Lv, Wanli Ouyang, Limin Wang, Min Dou, Xizhou Zhu, Tong Lu, Dahua Lin, Jifeng Dai, Weijie Su, Bowen Zhou, Kai Chen, Yu Qiao, Wenhai Wang, Gen Luo

Abstract:
We introduce InternVL 3.5, a new family of open-source multimodal models that significantly advances versatility, reasoning capability, and inference efficiency along the InternVL series. A key innovation is the Cascade Reinforcement Learning (Cascade RL) framework, which enhances reasoning through a two-stage process: offline RL for stable convergence and online RL for refined alignment. This coarse-to-fine training strategy leads to substantial improvements on downstream reasoning tasks, e.g., MMMU and MathVista. To optimize efficiency, we propose a Visual Resolution Router (ViR) that dynamically adjusts the resolution of visual tokens without compromising performance. Coupled with ViR, our Decoupled Vision-Language Deployment (DvD) strategy separates the vision encoder and language model across different GPUs, effectively balancing computational load. These contributions collectively enable InternVL3.5 to achieve up to a +16.0\% gain in overall reasoning performance and a 4.05$\times$ inference speedup compared to its predecessor, i.e., InternVL3. In addition, InternVL3.5 supports novel capabilities such as GUI interaction and embodied agency. Notably, our largest model, i.e., InternVL3.5-241B-A28B, attains state-of-the-art results among open-source MLLMs across general multimodal, reasoning, text, and agentic tasks -- narrowing the performance gap with leading commercial models like GPT-5. All models and code are publicly released.

Paperid: 1946, https://arxiv.org/pdf/2506.18385.pdf

Abstract:
Recent benchmarks and datasets have been proposed to improve spatial reasoning in vision-language models (VLMs), yet existing open resources remain limited in scale, visual diversity, and instruction expressiveness. In this work, we introduce InternSpatial, the largest open-source dataset for spatial reasoning in VLMs, along with InternSpatial-Bench, a corresponding evaluation benchmark designed to assess spatial understanding under diverse instruction formats. InternSpatial comprises 12 million QA pairs spanning both single-view and multi-view settings, drawn from diverse visual environments and supporting 19 instruction formats that reflect varied query styles. For evaluation, we propose InternSpatial-Bench for single-view tasks and expand multi-view reasoning by introducing a novel rotation angle prediction task that has not been explored in prior work. Experimental results show that models trained on InternSpatial achieve 12.1% improvement on InternSpatial-Bench and 10.7% on VSI-Bench, while maintaining strong performance on general-purpose benchmarks. We hope these resources will support the development of spatially capable VLMs in practical applications such as robotics and embodied AI.

Paperid: 1947, https://arxiv.org/pdf/2412.16158.pdf

Abstract:
The rapid advance of Large Language Models (LLMs) has catalyzed the development of Vision-Language Models (VLMs). Monolithic VLMs, which avoid modality-specific encoders, offer a promising alternative to the compositional ones but face the challenge of inferior performance. Most existing monolithic VLMs require tuning pre-trained LLMs to acquire vision abilities, which may degrade their language capabilities. To address this dilemma, this paper presents a novel high-performance monolithic VLM named HoVLE. We note that LLMs have been shown capable of interpreting images, when image embeddings are aligned with text embeddings. The challenge for current monolithic VLMs actually lies in the lack of a holistic embedding module for both vision and language inputs. Therefore, HoVLE introduces a holistic embedding module that converts visual and textual inputs into a shared space, allowing LLMs to process images in the same way as texts. Furthermore, a multi-stage training strategy is carefully designed to empower the holistic embedding module. It is first trained to distill visual features from a pre-trained vision encoder and text embeddings from the LLM, enabling large-scale training with unpaired random images and text tokens. The whole model further undergoes next-token prediction on multi-modal data to align the embeddings. Finally, an instruction-tuning stage is incorporated. Our experiments show that HoVLE achieves performance close to leading compositional models on various benchmarks, outperforming previous monolithic models by a large margin. Model available at https://huggingface.co/OpenGVLab/HoVLE.

Paperid: 1948, https://arxiv.org/pdf/2411.17837.pdf

Abstract:
Oracle bone script (OBS), as China's earliest mature writing system, present significant challenges in automatic recognition due to their complex pictographic structures and divergence from modern Chinese characters. We introduce OracleSage, a novel cross-modal framework that integrates hierarchical visual understanding with graph-based semantic reasoning. Specifically, we propose (1) a Hierarchical Visual-Semantic Understanding module that enables multi-granularity feature extraction through progressive fine-tuning of LLaVA's visual backbone, (2) a Graph-based Semantic Reasoning Framework that captures relationships between visual components and semantic concepts through dynamic message passing, and (3) OracleSem, a semantically enriched OBS dataset with comprehensive pictographic and semantic annotations. Experimental results demonstrate that OracleSage significantly outperforms state-of-the-art vision-language models. This research establishes a new paradigm for ancient text interpretation while providing valuable technical support for archaeological studies.

Paperid: 1949, https://arxiv.org/pdf/2512.17206.pdf

Abstract:
Exploration capacity shapes both inference-time performance and reinforcement learning (RL) training for large (vision-) language models, as stochastic sampling often yields redundant reasoning paths with little high-level diversity. This paper proposes Reasoning Palette, a novel latent-modulation framework that endows the model with a stochastic latent variable for strategic contextualization, guiding its internal planning prior to token generation. This latent context is inferred from the mean-pooled embedding of a question-answer pair via a variational autoencoder (VAE), where each sampled latent potentially encodes a distinct reasoning context. During inference, a sampled latent is decoded into learnable token prefixes and prepended to the input prompt, modulating the model's internal reasoning trajectory. In this way, the model performs internal sampling over reasoning strategies prior to output generation, which shapes the style and structure of the entire response sequence. A brief supervised fine-tuning (SFT) warm-up phase allows the model to adapt to this latent conditioning. Within RL optimization, Reasoning Palette facilitates structured exploration by enabling on-demand injection for diverse reasoning modes, significantly enhancing exploration efficiency and sustained learning capability. Experiments across multiple reasoning benchmarks demonstrate that our method enables interpretable and controllable control over the (vision-) language model's strategic behavior, thereby achieving consistent performance gains over standard RL methods.

Paperid: 1950, https://arxiv.org/pdf/2502.11718.pdf

Abstract:
The evaluation of factual accuracy in large vision language models (LVLMs) has lagged behind their rapid development, making it challenging to fully reflect these models' knowledge capacity and reliability. In this paper, we introduce the first factuality-based visual question-answering benchmark in Chinese, named ChineseSimpleVQA, aimed at assessing the visual factuality of LVLMs across 8 major topics and 56 subtopics. The key features of this benchmark include a focus on the Chinese language, diverse knowledge types, a multi-hop question construction, high-quality data, static consistency, and easy-to-evaluate through short answers. Moreover, we contribute a rigorous data construction pipeline and decouple the visual factuality into two parts: seeing the world (i.e., object recognition) and discovering knowledge. This decoupling allows us to analyze the capability boundaries and execution mechanisms of LVLMs. Subsequently, we evaluate 34 advanced open-source and closed-source models, revealing critical performance gaps within this field. Our evaluation-friendly code and data have already been open-sourced.

Paperid: 1951, https://arxiv.org/pdf/2601.10527.pdf

Abstract:
The rapid evolution of Large Language Models (LLMs) and Multimodal Large Language Models (MLLMs) has driven major gains in reasoning, perception, and generation across language and vision, yet whether these advances translate into comparable improvements in safety remains unclear, partly due to fragmented evaluations that focus on isolated modalities or threat models. In this report, we present an integrated safety evaluation of six frontier models--GPT-5.2, Gemini 3 Pro, Qwen3-VL, Grok 4.1 Fast, Nano Banana Pro, and Seedream 4.5--assessing each across language, vision-language, and image generation using a unified protocol that combines benchmark, adversarial, multilingual, and compliance evaluations. By aggregating results into safety leaderboards and model profiles, we reveal a highly uneven safety landscape: while GPT-5.2 demonstrates consistently strong and balanced performance, other models exhibit clear trade-offs across benchmark safety, adversarial robustness, multilingual generalization, and regulatory compliance. Despite strong results under standard benchmarks, all models remain highly vulnerable under adversarial testing, with worst-case safety rates dropping below 6%. Text-to-image models show slightly stronger alignment in regulated visual risk categories, yet remain fragile when faced with adversarial or semantically ambiguous prompts. Overall, these findings highlight that safety in frontier models is inherently multidimensional--shaped by modality, language, and evaluation design--underscoring the need for standardized, holistic safety assessments to better reflect real-world risk and guide responsible deployment.

Paperid: 1952, https://arxiv.org/pdf/2507.23373.pdf

Abstract:
Large Vision-Language Models like CLIP have become a powerful foundation for Unsupervised Domain Adaptation due to their strong zero-shot generalization. State-of-the-art methods typically leverage CLIP to generate pseudo-labels for the target domain, then fine-tune the model to learn domain-invariant features. However, these methods attempt to align source and target domains using all pseudo-labeled data simultaneously. This one-shot alignment struggles with noisy, hard-to-classify samples, leading to error propagation and suboptimal feature learning. The problem is even more amplified in the multi-source scenario, where diverse domain gaps and varying noise levels across multiple source domains further destabilize the alignment process. To address this issue, in this work, we propose a progressive alignment strategy for adapting CLIP to unlabeled downstream task. Our method begins by training the model on a high-confidence subset of target samples, allowing it to first learn a well-aligned representation from the most reliable data. As training progresses, it gradually incorporates more challenging samples, guiding the model to refine its understanding without being overwhelmed by initial label noise. This progressive approach effectively mitigates confirmation bias and promotes a more robust convergence, allowing for the learning of genuinely domain-invariant features. We name our approach MP^2A and test it on three popular UDA benchmarks, namely ImageCLEF, Office-Home, and the most challenging DomainNet. Experiments showcase that MP^2A achieves state-of-the-art performance when compared with most recent CLIP-based MS-UDA approaches, demonstrating the effectiveness of our approach.

Paperid: 1953, https://arxiv.org/pdf/2412.18194.pdf

Abstract:
General-purposed embodied agents are designed to understand the users' natural instructions or intentions and act precisely to complete universal tasks. Recently, methods based on foundation models especially Vision-Language-Action models (VLAs) have shown a substantial potential to solve language-conditioned manipulation (LCM) tasks well. However, existing benchmarks do not adequately meet the needs of VLAs and relative algorithms. To better define such general-purpose tasks in the context of LLMs and advance the research in VLAs, we present VLABench, an open-source benchmark for evaluating universal LCM task learning. VLABench provides 100 carefully designed categories of tasks, with strong randomization in each category of task and a total of 2000+ objects. VLABench stands out from previous benchmarks in four key aspects: 1) tasks requiring world knowledge and common sense transfer, 2) natural language instructions with implicit human intentions rather than templates, 3) long-horizon tasks demanding multi-step reasoning, and 4) evaluation of both action policies and language model capabilities. The benchmark assesses multiple competencies including understanding of mesh\&texture, spatial relationship, semantic instruction, physical laws, knowledge transfer and reasoning, etc. To support the downstream finetuning, we provide high-quality training data collected via an automated framework incorporating heuristic skills and prior information. The experimental results indicate that both the current state-of-the-art pretrained VLAs and the workflow based on VLMs face challenges in our tasks.

Paperid: 1954, https://arxiv.org/pdf/2509.21787.pdf

Abstract:
The rise in harmful online content not only distorts public discourse but also poses significant challenges to maintaining a healthy digital environment. In response to this, we introduce a multimodal dataset uniquely crafted for identifying hate in digital content. Central to our methodology is the innovative application of watermarked, stability-enhanced, stable diffusion techniques combined with the Digital Attention Analysis Module (DAAM). This combination is instrumental in pinpointing the hateful elements within images, thereby generating detailed hate attention maps, which are used to blur these regions from the image, thereby removing the hateful sections of the image. We release this data set as a part of the dehate shared task. This paper also describes the details of the shared task. Furthermore, we present DeHater, a vision-language model designed for multimodal dehatification tasks. Our approach sets a new standard in AI-driven image hate detection given textual prompts, contributing to the development of more ethical AI applications in social media.

Paperid: 1955, https://arxiv.org/pdf/2503.06937.pdf

Abstract:
Zero-Shot Object Navigation (ZSON) requires agents to navigate to objects specified via open-ended natural language without predefined categories or prior environmental knowledge. While recent methods leverage foundation models or multi-modal maps, they often rely on 2D representations and greedy strategies or require additional training or modules with high computation load, limiting performance in complex environments and real applications. We propose WTRP-Searcher, a novel framework that formulates ZSON as a Weighted Traveling Repairman Problem (WTRP), minimizing the weighted waiting time of viewpoints. Using a Vision-Language Model (VLM), we score viewpoints based on object-description similarity, projected onto a 2D map with depth information. An open-vocabulary detector identifies targets, dynamically updating goals, while a 3D embedding feature map enhances spatial awareness and environmental recall. WTRP-Searcher outperforms existing methods, offering efficient global planning and improved performance in complex ZSON tasks. Code and design will be open-sourced upon acceptance.

Paperid: 1956, https://arxiv.org/pdf/2412.17304.pdf

Abstract:
We build upon time-series classification by leveraging the capabilities of Vision Language Models (VLMs). We find that VLMs produce competitive results after two or less epochs of fine-tuning. We develop a novel approach that incorporates graphical data representations as images in conjunction with numerical data. This approach is rooted in the hypothesis that graphical representations can provide additional contextual information that numerical data alone may not capture. Additionally, providing a graphical representation can circumvent issues such as limited context length faced by LLMs. To further advance this work, we implemented a scalable end-to-end pipeline for training on different scenarios, allowing us to isolate the most effective strategies for transferring learning capabilities from LLMs to Time Series Classification (TSC) tasks. Our approach works with univariate and multivariate time-series data. In addition, we conduct extensive and practical experiments to show how this approach works for time-series classification and generative labels.

Paperid: 1957, https://arxiv.org/pdf/2409.09269.pdf

Abstract:
Visual Question-Answering (VQA) has become key to user experience, particularly after improved generalization capabilities of Vision-Language Models (VLMs). But evaluating VLMs for an application requirement using a standardized framework in practical settings is still challenging. This paper aims to solve that using an end-to-end framework. We present VQA360 - a novel dataset derived from established VQA benchmarks, annotated with task types, application domains, and knowledge types, for a comprehensive evaluation. We also introduce GoEval, a multimodal evaluation metric developed using GPT-4o, achieving a correlation factor of 56.71% with human judgments. Our experiments with state-of-the-art VLMs reveal that no single model excels universally, thus, making a right choice a key design decision. Proprietary models such as Gemini-1.5-Pro and GPT-4o-mini generally outperform others, but open-source models like InternVL-2-8B and CogVLM-2-Llama-3-19B also demonstrate competitive strengths, while providing additional advantages. Our framework can also be extended to other tasks.

Paperid: 1958, https://arxiv.org/pdf/2601.02046.pdf

Abstract:
Text-to-image (T2I) diffusion models such as SDXL and FLUX have achieved impressive photorealism, yet small-scale distortions remain pervasive in limbs, face, text and so on. Existing refinement approaches either perform costly iterative re-generation or rely on vision-language models (VLMs) with weak spatial grounding, leading to semantic drift and unreliable local edits. To close this gap, we propose Agentic Retoucher, a hierarchical decision-driven framework that reformulates post-generation correction as a human-like perception-reasoning-action loop. Specifically, we design (1) a perception agent that learns contextual saliency for fine-grained distortion localization under text-image consistency cues, (2) a reasoning agent that performs human-aligned inferential diagnosis via progressive preference alignment, and (3) an action agent that adaptively plans localized inpainting guided by user preference. This design integrates perceptual evidence, linguistic reasoning, and controllable correction into a unified, self-corrective decision process. To enable fine-grained supervision and quantitative evaluation, we further construct GenBlemish-27K, a dataset of 6K T2I images with 27K annotated artifact regions across 12 categories. Extensive experiments demonstrate that Agentic Retoucher consistently outperforms state-of-the-art methods in perceptual quality, distortion localization and human preference alignment, establishing a new paradigm for self-corrective and perceptually reliable T2I generation.

Paperid: 1959, https://arxiv.org/pdf/2510.08508.pdf

Abstract:
Real-world videos often suffer from complex degradations, such as noise, compression artifacts, and low-light distortions, due to diverse acquisition and transmission conditions. Existing restoration methods typically require professional manual selection of specialized models or rely on monolithic architectures that fail to generalize across varying degradations. Inspired by expert experience, we propose MoA-VR, the first \underline{M}ixture-\underline{o}f-\underline{A}gents \underline{V}ideo \underline{R}estoration system that mimics the reasoning and processing procedures of human professionals through three coordinated agents: Degradation Identification, Routing and Restoration, and Restoration Quality Assessment. Specifically, we construct a large-scale and high-resolution video degradation recognition benchmark and build a vision-language model (VLM) driven degradation identifier. We further introduce a self-adaptive router powered by large language models (LLMs), which autonomously learns effective restoration strategies by observing tool usage patterns. To assess intermediate and final processed video quality, we construct the \underline{Res}tored \underline{V}ideo \underline{Q}uality (Res-VQ) dataset and design a dedicated VLM-based video quality assessment (VQA) model tailored for restoration tasks. Extensive experiments demonstrate that MoA-VR effectively handles diverse and compound degradations, consistently outperforming existing baselines in terms of both objective metrics and perceptual quality. These results highlight the potential of integrating multimodal intelligence and modular reasoning in general-purpose video restoration systems.

Paperid: 1960, https://arxiv.org/pdf/2509.04334.pdf

Abstract:
Image geolocalization aims to predict the geographic location of images captured anywhere on Earth, but its global nature presents significant challenges. Current evaluation methodologies suffer from two major limitations. First, data leakage: advanced approaches often rely on large vision-language models (LVLMs) to predict image locations, yet these models are frequently pretrained on the test datasets, compromising the accuracy of evaluating a model's actual geolocalization capability. Second, existing metrics primarily rely on exact geographic coordinates to assess predictions, which not only neglects the reasoning process but also raises privacy concerns when user-level location data is required. To address these issues, we propose GeoArena, a first open platform for evaluating LVLMs on worldwide image geolocalization tasks, offering true in-the-wild and human-centered benchmarking. GeoArena enables users to upload in-the-wild images for a more diverse evaluation corpus, and it leverages pairwise human judgments to determine which model output better aligns with human expectations. Our platform has been deployed online for two months, during which we collected over thousands voting records. Based on this data, we conduct a detailed analysis and establish a leaderboard of different LVLMs on the image geolocalization task.

Paperid: 1961, https://arxiv.org/pdf/2507.16814.pdf

Abstract:
Enhancing large vision-language models (LVLMs) with visual slow-thinking reasoning is crucial for solving complex multimodal tasks. However, since LVLMs are mainly trained with vision-language alignment, it is difficult to adopt on-policy reinforcement learning (RL) to develop the slow thinking ability because the rollout space is restricted by its initial abilities. Off-policy RL offers a way to go beyond the current policy, but directly distilling trajectories from external models may cause visual hallucinations due to mismatched visual perception abilities across models. To address these issues, this paper proposes SOPHIA, a simple and scalable Semi-Off-Policy RL for vision-language slow-tHInking reAsoning. SOPHIA builds a semi-off-policy behavior model by combining on-policy visual understanding from a trainable LVLM with off-policy slow-thinking reasoning from a language model, assigns outcome-based rewards to reasoning, and propagates visual rewards backward. Then LVLM learns slow-thinking reasoning ability from the obtained reasoning trajectories using propagated rewards via off-policy RL algorithms. Extensive experiments with InternVL2.5 and InternVL3.0 with 8B and 38B sizes show the effectiveness of SOPHIA. Notably, SOPHIA improves InternVL3.0-38B by 8.50% in average, reaching state-of-the-art performance among open-source LVLMs on multiple multimodal reasoning benchmarks, and even outperforms some closed-source models (e.g., GPT-4.1) on the challenging MathVision and OlympiadBench, achieving 49.08% and 49.95% pass@1 accuracy, respectively. Analysis shows SOPHIA outperforms supervised fine-tuning and direct on-policy RL methods, offering a better policy initialization for further on-policy training.

Paperid: 1962, https://arxiv.org/pdf/2505.13731.pdf

Abstract:
Worldwide image geolocalization-the task of predicting GPS coordinates from images taken anywhere on Earth-poses a fundamental challenge due to the vast diversity in visual content across regions. While recent approaches adopt a two-stage pipeline of retrieving candidates and selecting the best match, they typically rely on simplistic similarity heuristics and point-wise supervision, failing to model spatial relationships among candidates. In this paper, we propose GeoRanker, a distance-aware ranking framework that leverages large vision-language models to jointly encode query-candidate interactions and predict geographic proximity. In addition, we introduce a multi-order distance loss that ranks both absolute and relative distances, enabling the model to reason over structured spatial relationships. To support this, we curate GeoRanking, the first dataset explicitly designed for geographic ranking tasks with multimodal candidate information. GeoRanker achieves state-of-the-art results on two well-established benchmarks (IM2GPS3K and YFCC4K), significantly outperforming current best methods.

Paperid: 1963, https://arxiv.org/pdf/2601.18340.pdf

Abstract:
Despite the remarkable progress in text-driven video editing, generating coherent non-rigid deformations remains a critical challenge, often plagued by physical distortion and temporal flicker. To bridge this gap, we propose NRVBench, the first dedicated and comprehensive benchmark designed to evaluate non-rigid video editing. First, we curate a high-quality dataset consisting of 180 non-rigid motion videos from six physics-based categories, equipped with 2,340 fine-grained task instructions and 360 multiple-choice questions. Second, we propose NRVE-Acc, a novel evaluation metric based on Vision-Language Models that can rigorously assess physical compliance, temporal consistency, and instruction alignment, overcoming the limitations of general metrics in capturing complex dynamics. Third, we introduce a training-free baseline, VM-Edit, which utilizes a dual-region denoising mechanism to achieve structure-aware control, balancing structural preservation and dynamic deformation. Extensive experiments demonstrate that while current methods have shortcomings in maintaining physical plausibility, our method achieves excellent performance across both standard and proposed metrics. We believe the benchmark could serve as a standard testing platform for advancing physics-aware video editing.

Paperid: 1964, https://arxiv.org/pdf/2601.10710.pdf

Abstract:
Vision-Language Models (VLMs) create a severe visual feature bottleneck by using a crude, asymmetric connection that links only the output of the vision encoder to the input of the large language model (LLM). This static architecture fundamentally limits the ability of LLMs to achieve comprehensive alignment with hierarchical visual knowledge, compromising their capacity to accurately integrate local details with global semantics into coherent reasoning. To resolve this, we introduce Cross-Layer Injection (CLI), a novel and lightweight framework that forges a dynamic many-to-many bridge between the two modalities. CLI consists of two synergistic, parameter-efficient components: an Adaptive Multi-Projection (AMP) module that harmonizes features from diverse vision layers, and an Adaptive Gating Fusion (AGF) mechanism that empowers the LLM to selectively inject the most relevant visual information based on its real-time decoding context. We validate the effectiveness and versatility of CLI by integrating it into LLaVA-OneVision and LLaVA-1.5. Extensive experiments on 18 diverse benchmarks demonstrate significant performance improvements, establishing CLI as a scalable paradigm that unlocks deeper multimodal understanding by granting LLMs on-demand access to the full visual hierarchy.

Paperid: 1965, https://arxiv.org/pdf/2512.15411.pdf

Abstract:
While leveraging abundant human videos and simulated robot data poses a scalable solution to the scarcity of real-world robot data, the generalization capability of existing vision-language-action models (VLAs) remains limited by mismatches in camera views, visual appearance, and embodiment morphologies. To overcome this limitation, we propose MiVLA, a generalizable VLA empowered by human-robot mutual imitation pre-training, which leverages inherent behavioral similarity between human hands and robotic arms to build a foundation of strong behavioral priors for both human actions and robotic control. Specifically, our method utilizes kinematic rules with left/right hand coordinate systems for bidirectional alignment between human and robot action spaces. Given human or simulated robot demonstrations, MiVLA is trained to forecast behavior trajectories for one embodiment, and imitate behaviors for another one unseen in the demonstration. Based on this mutual imitation, it integrates the behavioral fidelity of real-world human data with the manipulative diversity of simulated robot data into a unified model, thereby enhancing the generalization capability for downstream tasks. Extensive experiments conducted on both simulation and real-world platforms with three robots (ARX, PiPer and LocoMan), demonstrate that MiVLA achieves strong improved generalization capability, outperforming state-of-the-art VLAs (e.g., $\boldsymbolπ_{0}$, $\boldsymbolπ_{0.5}$ and H-RDT) by 25% in simulation, and 14% in real-world robot control tasks.

Paperid: 1966, https://arxiv.org/pdf/2510.24795.pdf

Abstract:
Vision-Language-Action models (VLAs) represent a significant frontier in embodied intelligence, aiming to bridge digital knowledge with physical-world interaction. While these models have demonstrated remarkable generalist capabilities, their deployment is severely hampered by the substantial computational and data requirements inherent to their underlying large-scale foundation models. Motivated by the urgent need to address these challenges, this survey presents the first comprehensive review of Efficient Vision-Language-Action models (Efficient VLAs) across the entire data-model-training process. Specifically, we introduce a unified taxonomy to systematically organize the disparate efforts in this domain, categorizing current techniques into three core pillars: (1) Efficient Model Design, focusing on efficient architectures and model compression; (2) Efficient Training, which reduces computational burdens during model learning; and (3) Efficient Data Collection, which addresses the bottlenecks in acquiring and utilizing robotic data. Through a critical review of state-of-the-art methods within this framework, this survey not only establishes a foundational reference for the community but also summarizes representative applications, delineates key challenges, and charts a roadmap for future research. We maintain a continuously updated project page to track our latest developments: https://evla-survey.github.io/

Paperid: 1967, https://arxiv.org/pdf/2510.16907.pdf

Abstract:
A key challenge in training Vision-Language Model (VLM) agents, compared to Language Model (LLM) agents, lies in the shift from textual states to complex visual observations. This transition introduces partial observability and demands robust world modeling. We ask: Can VLM agents construct internal world models through explicit visual state reasoning? To address this question, we architecturally enforce and reward the agent's reasoning process via reinforcement learning (RL), formulating it as a Partially Observable Markov Decision Process (POMDP). We find that decomposing the agent's reasoning into State Estimation ("what is the current state?") and Transition Modeling ("what comes next?") is critical for success, as demonstrated through five reasoning strategies. Our investigation into how agents represent internal beliefs reveals that the optimal representation is task-dependent: Natural Language excels at capturing semantic relationships in general tasks, while Structured formats are indispensable for precise manipulation and control. Building on these insights, we design a World Modeling Reward that provides dense, turn-level supervision for accurate state prediction, and introduce Bi-Level General Advantage Estimation (Bi-Level GAE) for turn-aware credit assignment. Through this form of visual state reasoning, a 3B-parameter model achieves a score of 0.82 across five diverse agent benchmarks, representing a 3$\times$ improvement over its untrained counterpart (0.21) and outperforming proprietary reasoning models such as GPT-5 (0.75), Gemini 2.5 Pro (0.67) and Claude 4.5 (0.62). All experiments are conducted within our VAGEN framework, a scalable system for training and analyzing multi-turn VLM agents in diverse visual environments. Code and data are publicly available at https://vagen-ai.github.io.

Paperid: 1968, https://arxiv.org/pdf/2509.21984.pdf

Abstract:
Large Vision-Language Models (LVLMs) have achieved remarkable success across a wide range of multimodal tasks, yet their robustness to spatial variations remains insufficiently understood. In this work, we present a systematic study of the spatial bias of LVLMs, focusing on how models respond when identical key visual information is placed at different locations within an image. Through a carefully designed probing dataset, we demonstrate that current LVLMs often produce inconsistent outputs under such spatial shifts, revealing a fundamental limitation in their spatial-semantic understanding. Further analysis shows that this phenomenon originates not from the vision encoder, which reliably perceives and interprets visual content across positions, but from the unbalanced design of position embeddings in the language model component. In particular, the widely adopted position embedding strategies, such as RoPE, introduce imbalance during cross-modal interaction, leading image tokens at different positions to exert unequal influence on semantic understanding. To mitigate this issue, we introduce Balanced Position Assignment (BaPA), a simple yet effective mechanism that assigns identical position embeddings to all image tokens, promoting a more balanced integration of visual information. Extensive experiments show that BaPA enhances the spatial robustness of LVLMs without retraining and further boosts their performance across diverse multimodal benchmarks when combined with lightweight fine-tuning. Further analysis of information flow reveals that BaPA yields balanced attention, enabling more holistic visual understanding.

Paperid: 1969, https://arxiv.org/pdf/2507.09491.pdf

Abstract:
Existing video benchmarks often resemble image-based benchmarks, with question types like "What actions does the person perform throughout the video?" or "What color is the woman's dress in the video?" For these, models can often answer by scanning just a few key frames, without deep temporal reasoning. This limits our ability to assess whether large vision-language models (LVLMs) can truly think with videos rather than perform superficial frame-level analysis. To address this, we introduce GLIMPSE, a benchmark specifically designed to evaluate whether LVLMs can genuinely think with videos. Unlike prior benchmarks, GLIMPSE emphasizes comprehensive video understanding beyond static image cues. It consists of 3,269 videos and over 4,342 highly visual-centric questions across 11 categories, including Trajectory Analysis, Temporal Reasoning, and Forensics Detection. All questions are carefully crafted by human annotators and require watching the entire video and reasoning over full video context-this is what we mean by thinking with video. These questions cannot be answered by scanning selected frames or relying on text alone. In human evaluations, GLIMPSE achieves 94.82% accuracy, but current LVLMs face significant challenges. Even the best-performing model, GPT-o3, reaches only 66.43%, highlighting that LVLMs still struggle to move beyond surface-level reasoning to truly think with videos.

Paperid: 1970, https://arxiv.org/pdf/2506.10128.pdf

Abstract:
Reinforcement learning (RL) has shown great effectiveness for fine-tuning large language models (LLMs) using tasks that are challenging yet easily verifiable, such as math reasoning or code generation. However, extending this success to visual perception in vision-language models (VLMs) has been impeded by the scarcity of vision-centric tasks that are simultaneously challenging and unambiguously verifiable. To this end, we introduce ViCrit (Visual Caption Hallucination Critic), an RL proxy task that trains VLMs to localize a subtle, synthetic visual hallucination injected into paragraphs of human-written image captions. Starting from a 200-word captions, we inject a single, subtle visual description error-altering a few words on objects, attributes, counts, or spatial relations-and task the model to pinpoint the corrupted span given the image and the modified caption. This formulation preserves the full perceptual difficulty while providing a binary, exact-match reward that is easy to compute and unambiguous. Models trained with the ViCrit Task exhibit substantial gains across a variety of VL benchmarks. Crucially, the improvements transfer beyond natural-image training data to abstract image reasoning and visual math, showing promises of learning to perceive rather than barely memorizing seen objects. To facilitate evaluation, we further introduce ViCrit-Bench, a category-balanced diagnostic benchmark that systematically probes perception errors across diverse image domains and error types. Together, our results demonstrate that fine-grained hallucination criticism is an effective and generalizable objective for enhancing visual perception in VLMs.

Paperid: 1971, https://arxiv.org/pdf/2505.19702.pdf

Abstract:
Recent advances in large language models have significantly improved textual reasoning through the effective use of Chain-of-Thought (CoT) and reinforcement learning. However, extending these successes to vision-language tasks remains challenging due to inherent limitations in text-only CoT, such as visual hallucinations and insufficient multimodal integration. In this paper, we introduce Point-RFT, a multimodal reasoning framework explicitly designed to leverage visually grounded CoT reasoning for visual document understanding. Our approach consists of two stages: First, we conduct format finetuning using a curated dataset of 71K diverse visual reasoning problems, each annotated with detailed, step-by-step rationales explicitly grounded to corresponding visual elements. Second, we employ reinforcement finetuning targeting visual document understanding. On ChartQA, our approach improves accuracy from 70.88% (format-finetuned baseline) to 90.04%, surpassing the 83.92% accuracy achieved by reinforcement finetuning relying solely on text-based CoT. The result shows that our grounded CoT is more effective for multimodal reasoning compared with the text-only CoT. Moreover, Point-RFT exhibits superior generalization capability across several out-of-domain visual document reasoning benchmarks, including CharXiv, PlotQA, IconQA, TabMWP, etc., and highlights its potential in complex real-world scenarios.

Paperid: 1972, https://arxiv.org/pdf/2505.05456.pdf

Abstract:
Spatial intelligence (SI) represents a cognitive ability encompassing the visualization, manipulation, and reasoning about spatial relationships, underpinning disciplines from neuroscience to robotics. We introduce SITE, a benchmark dataset towards SI Thorough Evaluation in a standardized format of multi-choice visual question-answering, designed to assess large vision-language models' spatial intelligence across diverse visual modalities (single-image, multi-image, and video) and SI factors (figural to environmental scales, spatial visualization and orientation, intrinsic and extrinsic, static and dynamic). Our approach to curating the benchmark combines a bottom-up survey about 31 existing datasets and a top-down strategy drawing upon three classification systems in cognitive science, which prompt us to design two novel types of tasks about view-taking and dynamic scenes. Extensive experiments reveal that leading models fall behind human experts especially in spatial orientation, a fundamental SI factor. Moreover, we demonstrate a positive correlation between a model's spatial reasoning proficiency and its performance on an embodied AI task.

Paperid: 1973, https://arxiv.org/pdf/2504.07934.pdf

Abstract:
We introduce ThinkLite-VL, a family of visual reasoning models that achieve state-of-the-art (SoTA) performance using an order of magnitude fewer training samples, relying purely on reinforcement fine-tuning (RFT) self-improvement without any knowledge distillation. Our central insight is that sample difficulty critically influences RFT effectiveness: appropriately challenging examples can drive substantial reasoning improvements, even in low-data regimes. However, quantifying sample difficulty in a reliable and scalable manner remains non-trivial. To address this, we repurpose Monte Carlo Tree Search (MCTS) to measure sample difficulty via the number of reasoning iterations a vision-language model (VLM) requires to solve each instance. This MCTS-based selection procedure identifies samples that induce deeper reasoning while remaining solvable, allowing us to filter a high-quality subset from 70k open-source examples spanning math, natural image understanding, and chart comprehension. Using this approach, we select just 11k challenging samples for RFT on Qwen2.5-VL-7B-Instruct and 7.5k samples for Qwen2.5-VL-72B-Instruct. The resulting models, ThinkLite-VL-7B and ThinkLite-VL-72B, significantly outperform their respective base models across eight visual reasoning benchmarks. In particular, ThinkLite-VL-7B improves the average performance of Qwen2.5-VL-7B-Instruct by 7\% and surpasses all existing 7B-level models, as well as much larger models such as GPT-4o, O1 and Qwen2.5-VL-72B, achieving a new SoTA score of 75.1 on MathVista. ThinkLite-VL-72B further advances the SoTA frontier, achieving an accuracy of 79.7 on MathVista and an average benchmark improvement of 4.42 over the open-source SOTA. These results demonstrate that MCTS-guided difficulty filtering provides a scalable and effective path toward data-efficient self-improvement in multimodal reasoning.

Paperid: 1974, https://arxiv.org/pdf/2503.08216.pdf

Abstract:
Despite their success, Large Vision-Language Models (LVLMs) remain vulnerable to hallucinations. While existing studies attribute the cause of hallucinations to insufficient visual attention to image tokens, our findings indicate that hallucinations also arise from interference from instruction tokens during decoding. Intuitively, certain instruction tokens continuously distort LVLMs' visual perception during decoding, hijacking their visual attention toward less discriminative visual regions. This distortion prevents them integrating broader contextual information from images, ultimately leading to hallucinations. We term this phenomenon 'Attention Hijacking', where disruptive instruction tokens act as 'Attention Hijackers'. To address this, we propose a novel, training-free strategy namely Attention HIjackers Detection and Disentanglement (AID), designed to isolate the influence of Hijackers, enabling LVLMs to rely on their context-aware intrinsic attention map. Specifically, AID consists of three components: First, Attention Hijackers Detection identifies Attention Hijackers by calculating instruction-driven visual salience. Next, Attention Disentanglement mechanism is proposed to mask the visual attention of these identified Hijackers, and thereby mitigate their disruptive influence on subsequent tokens. Finally, Re-Disentanglement recalculates the balance between instruction-driven and image-driven visual salience to avoid over-masking effects. Extensive experiments demonstrate that AID significantly reduces hallucination across various LVLMs on several benchmarks.

Paperid: 1975, https://arxiv.org/pdf/2502.14296.pdf

Authors:Yue Huang, Chujie Gao, Siyuan Wu, Haoran Wang, Xiangqi Wang, Yujun Zhou, Yanbo Wang, Jiayi Ye, Jiawen Shi, Qihui Zhang, Yuan Li, Han Bao, Zhaoyi Liu, Tianrui Guan, Dongping Chen, Ruoxi Chen, Kehan Guo, Andy Zou, Bryan Hooi Kuen-Yew, Caiming Xiong, Elias Stengel-Eskin, Hongyang Zhang, Hongzhi Yin, Huan Zhang, Huaxiu Yao, Jaehong Yoon, Jieyu Zhang, Kai Shu, Kaijie Zhu, Ranjay Krishna, Swabha Swayamdipta, Taiwei Shi, Weijia Shi, Xiang Li, Yiwei Li, Yuexing Hao, Zhihao Jia, Zhize Li, Xiuying Chen, Zhengzhong Tu, Xiyang Hu, Tianyi Zhou, Jieyu Zhao, Lichao Sun, Furong Huang, Or Cohen Sasson, Prasanna Sattigeri, Anka Reuel, Max Lamparth, Yue Zhao, Nouha Dziri, Yu Su, Huan Sun, Heng Ji, Chaowei Xiao, Mohit Bansal, Nitesh V. Chawla, Jian Pei, Jianfeng Gao, Michael Backes, Philip S. Yu, Neil Zhenqiang Gong, Pin-Yu Chen, Bo Li, Dawn Song, Xiangliang Zhang

Abstract:
Generative Foundation Models (GenFMs) have emerged as transformative tools. However, their widespread adoption raises critical concerns regarding trustworthiness across dimensions. This paper presents a comprehensive framework to address these challenges through three key contributions. First, we systematically review global AI governance laws and policies from governments and regulatory bodies, as well as industry practices and standards. Based on this analysis, we propose a set of guiding principles for GenFMs, developed through extensive multidisciplinary collaboration that integrates technical, ethical, legal, and societal perspectives. Second, we introduce TrustGen, the first dynamic benchmarking platform designed to evaluate trustworthiness across multiple dimensions and model types, including text-to-image, large language, and vision-language models. TrustGen leverages modular components--metadata curation, test case generation, and contextual variation--to enable adaptive and iterative assessments, overcoming the limitations of static evaluation methods. Using TrustGen, we reveal significant progress in trustworthiness while identifying persistent challenges. Finally, we provide an in-depth discussion of the challenges and future directions for trustworthy GenFMs, which reveals the complex, evolving nature of trustworthiness, highlighting the nuanced trade-offs between utility and trustworthiness, and consideration for various downstream applications, identifying persistent challenges and providing a strategic roadmap for future research. This work establishes a holistic framework for advancing trustworthiness in GenAI, paving the way for safer and more responsible integration of GenFMs into critical applications. To facilitate advancement in the community, we release the toolkit for dynamic evaluation.

Paperid: 1976, https://arxiv.org/pdf/2412.13540.pdf

Abstract:
Large Vision-Language Models (LVLMs) have demonstrated remarkable performance across diverse tasks. Despite great success, recent studies show that LVLMs encounter substantial limitations when engaging with visual graphs. To study the reason behind these limitations, we propose VGCure, a comprehensive benchmark covering 22 tasks for examining the fundamental graph understanding and reasoning capacities of LVLMs. Extensive evaluations conducted on 14 LVLMs reveal that LVLMs are weak in basic graph understanding and reasoning tasks, particularly those concerning relational or structurally complex information. Based on this observation, we propose a structure-aware fine-tuning framework to enhance LVLMs with structure learning abilities through three self-supervised learning tasks. Experiments validate the effectiveness of our method in improving LVLMs' performance on fundamental and downstream graph learning tasks, as well as enhancing their robustness against complex visual graphs.

Paperid: 1977, https://arxiv.org/pdf/2410.21259.pdf

Abstract:
Large Vision-Language Models (LVLMs) have become essential for advancing the integration of visual and linguistic information. However, the evaluation of LVLMs presents significant challenges as the evaluation benchmark always demands lots of human cost for its construction, and remains static, lacking flexibility once constructed. Even though automatic evaluation has been explored in textual modality, the visual modality remains under-explored. As a result, in this work, we address a question: "Can LVLMs themselves be used to benchmark each other in the visual automatically domain?". We introduce AutoBench-V, an automated framework for serving evaluation on demand, i.e., benchmarking LVLMs based on specific aspects of model capability. AutoBench-V leverages text-to-image models to generate relevant image samples and then utilizes LVLMs to orchestrate visual question-answering (VQA) tasks, completing the evaluation process efficiently and flexibly. Through an extensive evaluation of nine popular LVLMs across five demanded user inputs (i.e., evaluation capabilities), the framework shows effectiveness and reliability.

Paperid: 1978, https://arxiv.org/pdf/2410.14182.pdf

Abstract:
Artificial Intelligence (AI) is revolutionizing scientific research, yet its growing integration into laboratory environments presents critical safety challenges. While large language models (LLMs) increasingly assist in tasks ranging from procedural guidance to autonomous experiment orchestration, an "illusion of understanding" may lead researchers to overestimate their reliability. Such overreliance is particularly dangerous in high-stakes laboratory settings, where failures in hazard identification or risk assessment can result in severe accidents. To address these concerns, we propose the Laboratory Safety Benchmark (LabSafety Bench), a comprehensive framework that evaluates large language models and vision language models (VLMs) on their ability to identify potential hazards, assess risks, and predict the consequences of unsafe actions in lab environments. LabSafety Bench comprises 765 multiple-choice questions aligned with US Occupational Safety and Health Administration (OSHA) protocols, along with 404 realistic laboratory scenarios featuring dual evaluation tasks: the Hazards Identification Test and the Consequence Identification Test, with 3128 open-ended questions in total. Evaluations across eight proprietary models, seven open-weight LLMs, and four VLMs reveal that, despite advanced performance on structured assessments, no model achieves the safety threshold required for reliable operation -- none scoring above 70% on the Hazards Identification Test. Moreover, while proprietary models tend to excel in multiple-choice evaluations, their performance in open-ended, real-world scenario responses is comparable to that of open-source models. These findings underscore the urgent need for specialized evaluation frameworks to ensure the safe and responsible deployment of AI in laboratory settings.

Paperid: 1979, https://arxiv.org/pdf/2506.23590.pdf

Abstract:
Although Large Vision-Language Models (LVLMs) have demonstrated powerful capabilities in interpreting visual information, they frequently produce content that deviates from visual information, leading to object hallucination. To tackle this, recent works mostly depend on expensive manual annotations and training cost, or significantly increase inference time. In this work, we observe that LVLMs' attention to visual information is significantly stronger when answering caption queries compared to non-caption queries. Inspired by this phenomenon, we propose Caption-sensitive Attention Intervention (CAI), a training-free, plug-and-play hallucination mitigation method that leverages the attention activation pattern in response to caption queries to enhance LVLMs' visual perception capability. Extensive experimental results across four benchmarks covering both discriminative and generative tasks, demonstrate that CAI achieves state-of-the-art (SOTA) hallucination mitigating performance only with minimal additional inference cost.

Paperid: 1980, https://arxiv.org/pdf/2510.13276.pdf

Abstract:
The rapid advancement of large vision language models (LVLMs) has led to a significant expansion of their context windows. However, an extended context window does not guarantee the effective utilization of the context, posing a critical challenge for real-world applications. Current evaluations of such long-context faithfulness are predominantly focused on the text-only domain, while multimodal assessments remain limited to short contexts. To bridge this gap, we introduce MMLongCite, a comprehensive benchmark designed to evaluate the fidelity of LVLMs in long-context scenarios. MMLongCite comprises 8 distinct tasks spanning 6 context length intervals and incorporates diverse modalities, including text, images, and videos. Our evaluation of state-of-the-art LVLMs reveals their limited faithfulness in handling long multimodal contexts. Furthermore, we provide an in-depth analysis of how context length and the position of crucial content affect the faithfulness of these models.

Paperid: 1981, https://arxiv.org/pdf/2505.19678.pdf

Abstract:
Large Vision-Language Models (LVLMs) are susceptible to hallucinations, where generated responses seem semantically plausible yet exhibit little or no relevance to the input image. Previous studies reveal that this issue primarily stems from LVLMs' over-reliance on language priors while disregarding the visual information during decoding. To alleviate this issue, we introduce a novel Conditional Pointwise Mutual Information (C-PMI) calibrated decoding strategy, which adaptively strengthens the mutual dependency between generated texts and input images to mitigate hallucinations. Unlike existing methods solely focusing on text token sampling, we propose to jointly model the contributions of visual and textual tokens to C-PMI, formulating hallucination mitigation as a bi-level optimization problem aimed at maximizing mutual information. To solve it, we design a token purification mechanism that dynamically regulates the decoding process by sampling text tokens remaining maximally relevant to the given image, while simultaneously refining image tokens most pertinent to the generated response. Extensive experiments across various benchmarks reveal that the proposed method significantly reduces hallucinations in LVLMs while preserving decoding efficiency.

Paperid: 1982, https://arxiv.org/pdf/2502.19269.pdf

Abstract:
While pre-trained Vision-Language Models (VLMs) such as CLIP exhibit impressive representational capabilities for multimodal data, recent studies have revealed their vulnerability to backdoor attacks. To alleviate the threat, existing defense strategies primarily focus on fine-tuning the entire suspicious model. However, the substantial model parameters increase the difficulty of reaching a stable and consistent optimization direction, limiting their resistance against state-of-the-art attacks and often resulting in a degradation of clean accuracy. To address this challenge, we propose Class-wise Backdoor Prompt Tuning (CBPT), an efficient and effective defense mechanism that operates on text prompts to indirectly purify poisoned CLIP. Specifically, we first employ the advanced contrastive learning via carefully crafted positive and negative samples, to effectively invert the backdoor triggers that are potentially adopted by the attacker. Once the dummy trigger is established, we leverage three well-designed loss functions to optimize these class-wise text prompts, modifying the model's decision boundary and further reclassifying the feature regions affected by backdoor triggers. Extensive experiments demonstrate that CBPT significantly mitigates backdoor threats while preserving model utility, e.g. an average Clean Accuracy (CA) of 58.83% and an Attack Success Rate (ASR) of 0.39% across seven mainstream backdoor attacks. These results underscore the superiority of our prompt purifying design to strengthen CLIP's robustness against backdoor attacks.

Paperid: 1983, https://arxiv.org/pdf/2410.15430.pdf

Abstract:
Adaptation of pretrained vision-language models such as CLIP to various downstream tasks have raised great interest in recent researches. Previous works have proposed a variety of test-time adaptation (TTA) methods to achieve strong generalization without any knowledge of the target domain. However, existing training-required TTA approaches like TPT necessitate entropy minimization that involves large computational overhead, while training-free methods like TDA overlook the potential for information mining from the test samples themselves. In this paper, we break down the design of existing popular training-required and training-free TTA methods and bridge the gap between them within our framework. Specifically, we maintain a light-weight key-value memory for feature retrieval from instance-agnostic historical samples and instance-aware boosting samples. The historical samples are filtered from the testing data stream and serve to extract useful information from the target distribution, while the boosting samples are drawn from regional bootstrapping and capture the knowledge of the test sample itself. We theoretically justify the rationality behind our method and empirically verify its effectiveness on both the out-of-distribution and the cross-domain datasets, showcasing its applicability in real-world situations.

Paperid: 1984, https://arxiv.org/pdf/2601.05248.pdf

Abstract:
Vision-Language-Action (VLA) models have recently demonstrated strong generalization capabilities in robotic manipulation. Some existing VLA approaches attempt to improve action accuracy by explicitly generating linguistic reasoning traces or future visual observations before action execution. However, explicit reasoning typically incurs non-negligible inference latency, which constrains the temporal resolution required for robotic manipulation. Moreover, such reasoning is confined to the linguistic space, imposing a representational bottleneck that struggles to faithfully capture ineffable physical attributes. To mitigate these limitations, we propose LaST$_0$, a framework that enables efficient reasoning before acting through a Latent Spatio-Temporal Chain-of-Thought (CoT), capturing fine-grained physical and robotic dynamics that are often difficult to verbalize. Specifically, we introduce a token-efficient latent CoT space that models future visual dynamics, 3D structural information, and robot proprioceptive states, and further extends these representations across time to enable temporally consistent implicit reasoning trajectories. Furthermore, LaST$_0$ adopts a dual-system architecture implemented via a Mixture-of-Transformers design, where a reasoning expert conducts low-frequency latent inference and an acting expert generates high-frequency actions conditioned on robotics-oriented latent representations. To facilitate coordination, LaST$_0$ is trained with heterogeneous operation frequencies, enabling adaptive switching between reasoning and action inference rates during deployment. Across ten simulated and six real-world manipulation tasks, LaST$_0$ improves mean success rates by 8% and 13% over prior VLA methods, respectively, while achieving substantially faster inference. Project website: https://sites.google.com/view/last0

Paperid: 1985, https://arxiv.org/pdf/2512.24653.pdf

Abstract:
While data-driven imitation learning has revolutionized robotic manipulation, current approaches remain constrained by the scarcity of large-scale, diverse real-world demonstrations. Consequently, the ability of existing models to generalize across long-horizon bimanual tasks and mobile manipulation in unstructured environments remains limited. To bridge this gap, we present RoboMIND 2.0, a comprehensive real-world dataset comprising over 310K dual-arm manipulation trajectories collected across six distinct robot embodiments and 739 complex tasks. Crucially, to support research in contact-rich and spatially extended tasks, the dataset incorporates 12K tactile-enhanced episodes and 20K mobile manipulation trajectories. Complementing this physical data, we construct high-fidelity digital twins of our real-world environments, releasing an additional 20K-trajectory simulated dataset to facilitate robust sim-to-real transfer. To fully exploit the potential of RoboMIND 2.0, we propose MIND-2 system, a hierarchical dual-system frame-work optimized via offline reinforcement learning. MIND-2 integrates a high-level semantic planner (MIND-2-VLM) to decompose abstract natural language instructions into grounded subgoals, coupled with a low-level Vision-Language-Action executor (MIND-2-VLA), which generates precise, proprioception-aware motor actions.

Paperid: 1986, https://arxiv.org/pdf/2512.02013.pdf

Abstract:
Vision-Language-Action (VLA) models have recently emerged, demonstrating strong generalization in robotic scene understanding and manipulation. However, when confronted with long-horizon tasks that require defined goal states, such as LEGO assembly or object rearrangement, existing VLA models still face challenges in coordinating high-level planning with precise manipulation. Therefore, we aim to endow a VLA model with the capability to infer the "how" process from the "what" outcomes, transforming goal states into executable procedures. In this paper, we introduce ManualVLA, a unified VLA framework built upon a Mixture-of-Transformers (MoT) architecture, enabling coherent collaboration between multimodal manual generation and action execution. Unlike prior VLA models that directly map sensory inputs to actions, we first equip ManualVLA with a planning expert that generates intermediate manuals consisting of images, position prompts, and textual instructions. Building upon these multimodal manuals, we design a Manual Chain-of-Thought (ManualCoT) reasoning process that feeds them into the action expert, where each manual step provides explicit control conditions, while its latent representation offers implicit guidance for accurate manipulation. To alleviate the burden of data collection, we develop a high-fidelity digital-twin toolkit based on 3D Gaussian Splatting, which automatically generates manual data for planning expert training. ManualVLA demonstrates strong real-world performance, achieving an average success rate 32% higher than the previous hierarchical SOTA baseline on LEGO assembly and object rearrangement tasks.

Paperid: 1987, https://arxiv.org/pdf/2511.03929.pdf

Authors:NVIDIA, :, Amala Sanjay Deshmukh, Kateryna Chumachenko, Tuomas Rintamaki, Matthieu Le, Tyler Poon, Danial Mohseni Taheri, Ilia Karmanov, Guilin Liu, Jarno Seppanen, Guo Chen, Karan Sapra, Zhiding Yu, Adi Renduchintala, Charles Wang, Peter Jin, Arushi Goel, Mike Ranzinger, Lukas Voegtle, Philipp Fischer, Timo Roman, Wei Ping, Boxin Wang, Zhuolin Yang, Nayeon Lee, Shaokun Zhang, Fuxiao Liu, Zhiqi Li, Di Zhang, Greg Heinrich, Hongxu Yin, Song Han, Pavlo Molchanov, Parth Mannan, Yao Xu, Jane Polak Scowcroft, Tom Balough, Subhashree Radhakrishnan, Paris Zhang, Sean Cha, Ratnesh Kumar, Zaid Pervaiz Bhat, Jian Zhang, Darragh Hanley, Pritam Biswas, Jesse Oliver, Kevin Vasques, Roger Waleffe, Duncan Riach, Oluwatobi Olabiyi, Ameya Sunil Mahabaleshwarkar, Bilal Kartal, Pritam Gundecha, Khanh Nguyen, Alexandre Milesi, Eugene Khvedchenia, Ran Zilberstein, Ofri Masad, Natan Bagrov, Nave Assaf, Tomer Asida, Daniel Afrimi, Amit Zuker, Netanel Haber, Zhiyu Cheng, Jingyu Xin, Di Wu, Nik Spirin, Maryam Moosaei, Roman Ageev, Vanshil Atul Shah, Yuting Wu, Daniel Korzekwa, Unnikrishnan Kizhakkemadam Sreekumar, Wanli Jiang, Padmavathy Subramanian, Alejandra Rico, Sandip Bhaskar, Saeid Motiian, Kedi Wu, Annie Surla, Chia-Chih Chen, Hayden Wolff, Matthew Feinberg, Melissa Corpuz, Marek Wawrzos, Eileen Long, Aastha Jhunjhunwala, Paul Hendricks, Farzan Memarian, Benika Hall, Xin-Yu Wang, David Mosallanezhad, Soumye Singhal, Luis Vega, Katherine Cheung, Krzysztof Pawelec, Michael Evans, Katherine Luna, Jie Lou, Erick Galinkin, Akshay Hazare, Kaustubh Purandare, Ann Guan, Anna Warno, Chen Cui, Yoshi Suhara, Shibani Likhite, Seph Mard, Meredith Price, Laya Sleiman, Saori Kaji, Udi Karpas, Kari Briski, Joey Conway, Michael Lightstone, Jan Kautz, Mohammad Shoeybi, Mostofa Patwary, Jonathen Cohen, Oleksii Kuchaiev, Andrew Tao, Bryan Catanzaro

Abstract:
We introduce Nemotron Nano V2 VL, the latest model of the Nemotron vision-language series designed for strong real-world document understanding, long video comprehension, and reasoning tasks. Nemotron Nano V2 VL delivers significant improvements over our previous model, Llama-3.1-Nemotron-Nano-VL-8B, across all vision and text domains through major enhancements in model architecture, datasets, and training recipes. Nemotron Nano V2 VL builds on Nemotron Nano V2, a hybrid Mamba-Transformer LLM, and innovative token reduction techniques to achieve higher inference throughput in long document and video scenarios. We are releasing model checkpoints in BF16, FP8, and FP4 formats and sharing large parts of our datasets, recipes and training code.

Paperid: 1988, https://arxiv.org/pdf/2509.26642.pdf

Abstract:
Vision-language-action models (VLAs) have shown generalization capabilities in robotic manipulation tasks by inheriting from vision-language models (VLMs) and learning action generation. Most VLA models focus on interpreting vision and language to generate actions, whereas robots must perceive and interact within the spatial-physical world. This gap highlights the need for a comprehensive understanding of robotic-specific multisensory information, which is crucial for achieving complex and contact-rich control. To this end, we introduce a multisensory language-action (MLA) model that collaboratively perceives heterogeneous sensory modalities and predicts future multisensory objectives to facilitate physical world modeling. Specifically, to enhance perceptual representations, we propose an encoder-free multimodal alignment scheme that innovatively repurposes the large language model itself as a perception module, directly interpreting multimodal cues by aligning 2D images, 3D point clouds, and tactile tokens through positional correspondence. To further enhance MLA's understanding of physical dynamics, we design a future multisensory generation post-training strategy that enables MLA to reason about semantic, geometric, and interaction information, providing more robust conditions for action generation. For evaluation, the MLA model outperforms the previous state-of-the-art 2D and 3D VLA methods by 12% and 24% in complex, contact-rich real-world tasks, respectively, while also demonstrating improved generalization to unseen configurations. Project website: https://sites.google.com/view/open-mla

Paperid: 1989, https://arxiv.org/pdf/2509.25681.pdf

Abstract:
Vision-Language-Action (VLA) models are emerging as a next-generation paradigm for robotics. We introduce dVLA, a diffusion-based VLA that leverages a multimodal chain-of-thought to unify visual perception, language reasoning, and robotic control in a single system. dVLA jointly optimizes perception, language understanding, and action under a single diffusion objective, enabling stronger cross-modal reasoning and better generalization to novel instructions and objects. For practical deployment, we mitigate inference latency by incorporating two acceleration strategies, a prefix attention mask and KV caching, yielding up to around times speedup at test-time inference. We evaluate dVLA in both simulation and the real world: on the LIBERO benchmark, it achieves state-of-the-art performance with a 96.4% average success rate, consistently surpassing both discrete and continuous action policies; on a real Franka robot, it succeeds across a diverse task suite, including a challenging bin-picking task that requires multi-step planning, demonstrating robust real-world performance. Together, these results underscore the promise of unified diffusion frameworks for practical, high-performance VLA robotics.

Paperid: 1990, https://arxiv.org/pdf/2509.22642.pdf

Abstract:
Humans develop an understanding of intuitive physics through active interaction with the world. This approach is in stark contrast to current video models, such as Sora, which rely on passive observation and therefore struggle with grasping physical causality. This observation leads to our central hypothesis: authentic physical intuition of the world model must be grounded in extensive, causally rich interactions with the real world. To test this hypothesis, we present WoW, a 14-billion-parameter generative world model trained on 2 million robot interaction trajectories. Our findings reveal that the model's understanding of physics is a probabilistic distribution of plausible outcomes, leading to stochastic instabilities and physical hallucinations. Furthermore, we demonstrate that this emergent capability can be actively constrained toward physical realism by SOPHIA, where vision-language model agents evaluate the DiT-generated output and guide its refinement by iteratively evolving the language instructions. In addition, a co-trained Inverse Dynamics Model translates these refined plans into executable robotic actions, thus closing the imagination-to-action loop. We establish WoWBench, a new benchmark focused on physical consistency and causal reasoning in video, where WoW achieves state-of-the-art performance in both human and autonomous evaluation, demonstrating strong ability in physical causality, collision dynamics, and object permanence. Our work provides systematic evidence that large-scale, real-world interaction is a cornerstone for developing physical intuition in AI. Models, data, and benchmarks will be open-sourced.

Paperid: 1991, https://arxiv.org/pdf/2509.13317.pdf

Abstract:
We present Spatial Region 3D (SR-3D) aware vision-language model that connects single-view 2D images and multi-view 3D data through a shared visual token space. SR-3D supports flexible region prompting, allowing users to annotate regions with bounding boxes, segmentation masks on any frame, or directly in 3D, without the need for exhaustive multi-frame labeling. We achieve this by enriching 2D visual features with 3D positional embeddings, which allows the 3D model to draw upon strong 2D priors for more accurate spatial reasoning across frames, even when objects of interest do not co-occur within the same view. Extensive experiments on both general 2D vision language and specialized 3D spatial benchmarks demonstrate that SR-3D achieves state-of-the-art performance, underscoring its effectiveness for unifying 2D and 3D representation space on scene understanding. Moreover, we observe applicability to in-the-wild videos without sensory 3D inputs or ground-truth 3D annotations, where SR-3D accurately infers spatial relationships and metric measurements.

Paperid: 1992, https://arxiv.org/pdf/2508.12778.pdf

Abstract:
Medical large vision-language Models (Med-LVLMs) have shown promise in clinical applications but suffer from factual inaccuracies and unreliable outputs, posing risks in real-world diagnostics. While retrieval-augmented generation has emerged as a potential solution, current medical multimodal RAG systems are unable to perform effective retrieval across heterogeneous sources. The irrelevance of retrieved reports affects the factuality of analysis, while insufficient knowledge affects the credibility of clinical decision-making. To bridge the gap, we construct MedAtlas, which includes extensive multimodal report repositories and diverse text corpora. Based on it, we present HeteroRAG, a novel framework that enhances Med-LVLMs through heterogeneous knowledge sources. The framework introduces Modality-specific CLIPs for effective report retrieval and a Multi-corpora Query Generator for dynamically constructing queries for diverse corpora. Incorporating knowledge from such multifaceted sources, Med-LVLM is then trained with Heterogeneous Knowledge Preference Tuning to achieve cross-modality and multi-source knowledge alignment. Extensive experiments across 12 datasets and 3 modalities demonstrate that the proposed HeteroRAG achieves state-of-the-art performance in most medical vision language benchmarks, significantly improving factual accuracy and reliability of Med-LVLMs.

Paperid: 1993, https://arxiv.org/pdf/2506.18897.pdf

Abstract:
Video Generation Models (VGMs) have become powerful backbones for Vision-Language-Action (VLA) models, leveraging large-scale pretraining for robust dynamics modeling. However, current methods underutilize their distribution modeling capabilities for predicting future states. Two challenges hinder progress: integrating generative processes into feature learning is both technically and conceptually underdeveloped, and naive frame-by-frame video diffusion is computationally inefficient for real-time robotics. To address these, we propose Manipulate in Dream (MinD), a dual-system world model for real-time, risk-aware planning. MinD uses two asynchronous diffusion processes: a low-frequency visual generator (LoDiff) that predicts future scenes and a high-frequency diffusion policy (HiDiff) that outputs actions. Our key insight is that robotic policies do not require fully denoised frames but can rely on low-resolution latents generated in a single denoising step. To connect early predictions to actions, we introduce DiffMatcher, a video-action alignment module with a novel co-training strategy that synchronizes the two diffusion models. MinD achieves a 63% success rate on RL-Bench, 60% on real-world Franka tasks, and operates at 11.3 FPS, demonstrating the efficiency of single-step latent features for control signals. Furthermore, MinD identifies 74% of potential task failures in advance, providing real-time safety signals for monitoring and intervention. This work establishes a new paradigm for efficient and reliable robotic manipulation using generative world models.

Paperid: 1994, https://arxiv.org/pdf/2506.01953.pdf

Abstract:
Generalized policy and execution efficiency constitute the two critical challenges in robotic manipulation. While recent foundation policies benefit from the common-sense reasoning capabilities of internet-scale pretrained vision-language models (VLMs), they often suffer from low execution frequency. To mitigate this dilemma, dual-system approaches, inspired by Kahneman's theory, have been proposed to leverage a VLM-based System 2 model handling high-level reasoning and a separate System 1 action model ensuring real-time control. However, existing designs maintain both systems as separate models, limiting System 1 from fully leveraging the rich pretrained knowledge from the VLM-based System 2. In this work, we propose Fast-in-Slow (FiS), a unified dual-system vision-language-action (VLA) model that embeds the System 1 execution module within the VLM-based System 2 by partially sharing parameters. This innovative paradigm not only enables high-frequency execution in System 1 but also facilitates coordination between the reasoning and execution components within a single foundation model of System 2. Given their fundamentally distinct roles within FiS-VLA, we design the two systems to incorporate heterogeneous modality inputs alongside asynchronous operating frequencies, enabling both fast and precise manipulation. To enable coordination between the two systems, a dual-aware co-training strategy is proposed that equips System 1 with action generation capabilities while preserving System 2's contextual reasoning representation. For evaluation, FiS-VLA outperforms previous state-of-the-art methods by 8% in simulation and 11% in real-world tasks in terms of average success rate, while achieving a 117.7 Hz control frequency with action chunk set to eight. Project web page: fast-in-slow.github.io.

Paperid: 1995, https://arxiv.org/pdf/2505.02166.pdf

Abstract:
In robotic, task goals can be conveyed through various modalities, such as language, goal images, and goal videos. However, natural language can be ambiguous, while images or videos may offer overly detailed specifications. To tackle these challenges, we introduce CrayonRobo that leverages comprehensive multi-modal prompts that explicitly convey both low-level actions and high-level planning in a simple manner. Specifically, for each key-frame in the task sequence, our method allows for manual or automatic generation of simple and expressive 2D visual prompts overlaid on RGB images. These prompts represent the required task goals, such as the end-effector pose and the desired movement direction after contact. We develop a training strategy that enables the model to interpret these visual-language prompts and predict the corresponding contact poses and movement directions in SE(3) space. Furthermore, by sequentially executing all key-frame steps, the model can complete long-horizon tasks. This approach not only helps the model explicitly understand the task objectives but also enhances its robustness on unseen tasks by providing easily interpretable prompts. We evaluate our method in both simulated and real-world environments, demonstrating its robust manipulation capabilities.

Paperid: 1996, https://arxiv.org/pdf/2503.10631.pdf

Abstract:
A fundamental objective of manipulation policy design is to endow robots to comprehend human instructions, reason about scene cues, and execute generalized actions in dynamic environments. Recent autoregressive vision-language-action (VLA) methods inherit common-sense reasoning capabilities from vision-language models (VLMs) for next action-token prediction. However, these methods quantize actions into discrete bins, which disrupts the continuity required for precise control. In contrast, existing diffusion-based VLA methods incorporate an additional diffusion head to predict continuous actions solely conditioned on feature representations extracted by the VLM, without fully leveraging the VLM's pretrained reasoning capabilities through token-level generation. To address these limitations, we introduce HybridVLA, a unified framework that absorbs the continuous nature of diffusion-based actions and the contextual reasoning of autoregression within a single large language model. To mitigate interference between the two generation paradigms, we propose a collaborative training recipe that seamlessly incorporates diffusion denoising into the next-token prediction process. With this recipe, we find these two action prediction methods not only reinforce each other but also exhibit varying strength across different tasks. Therefore, we design a collaborative action ensemble mechanism that adaptively fuses both predictions, leading to more robust control. HybridVLA outperforms previous state-of-the-art VLA methods by 14\% and 19\% in mean success rate on simulation and real-world tasks, respectively, while demonstrating stable manipulation in unseen configurations.

Paperid: 1997, https://arxiv.org/pdf/2412.07679.pdf

Abstract:
Agglomerative models have recently emerged as a powerful approach to training vision foundation models, leveraging multi-teacher distillation from existing models such as CLIP, DINO, and SAM. This strategy enables the efficient creation of robust models, combining the strengths of individual teachers while significantly reducing computational and resource demands. In this paper, we thoroughly analyze state-of-the-art agglomerative models, identifying critical challenges including resolution mode shifts, teacher imbalance, idiosyncratic teacher artifacts, and an excessive number of output tokens. To address these issues, we propose several novel solutions: multi-resolution training, mosaic augmentation, and improved balancing of teacher loss functions. Specifically, in the context of Vision Language Models, we introduce a token compression technique to maintain high-resolution information within a fixed token count. We release our top-performing variants at multiple scales (-B, -L, -H, and -g), along with inference code and pretrained weights

Paperid: 1998, https://arxiv.org/pdf/2601.07291.pdf

Abstract:
Watermarking has emerged as a pivotal solution for content traceability and intellectual property protection in Large Vision-Language Models (LVLMs). However, vision-agnostic watermarks introduce visually irrelevant tokens and disrupt visual grounding by enforcing indiscriminate pseudo-random biases, while some semantic-aware methods incur prohibitive inference latency due to rejection sampling. In this paper, we propose the VIsual Semantic Adaptive Watermark (VISA-Mark), a novel framework that embeds detectable signals while strictly preserving visual fidelity. Our approach employs a lightweight, efficiently trained prefix-tuner to extract dynamic Visual-Evidence Weights, which quantify the evidentiary support for candidate tokens based on the visual input. These weights guide an adaptive vocabulary partitioning and logits perturbation mechanism, concentrating watermark strength specifically on visually-supported tokens. By actively aligning the watermark with visual evidence, VISA-Mark effectively maintains visual fidelity. Empirical results confirm that VISA-Mark outperforms conventional methods with a 7.8% improvement in visual consistency (Chair-I) and superior semantic fidelity. The framework maintains highly competitive detection accuracy (96.88% AUC) and robust attack resilience (99.3%) without sacrificing inference efficiency, effectively establishing a new standard for reliability-preserving multimodal watermarking.

Paperid: 1999, https://arxiv.org/pdf/2601.05159.pdf

Abstract:
Object hallucination critically undermines the reliability of Multimodal Large Language Models, often stemming from a fundamental failure in cognitive introspection, where models blindly trust linguistic priors over specific visual evidence. Existing mitigations remain limited: contrastive decoding approaches operate superficially without rectifying internal semantic misalignments, while current latent steering methods rely on static vectors that lack instance-specific precision. We introduce Vision-Language Introspection (VLI), a training-free inference framework that simulates a metacognitive self-correction process. VLI first performs Attributive Introspection to diagnose hallucination risks via probabilistic conflict detection and localize the causal visual anchors. It then employs Interpretable Bi-Causal Steering to actively modulate the inference process, dynamically isolating visual evidence from background noise while neutralizing blind confidence through adaptive calibration. VLI achieves state-of-the-art performance on advanced models, reducing object hallucination rates by 12.67% on MMHal-Bench and improving accuracy by 5.8% on POPE.

Paperid: 2000, https://arxiv.org/pdf/2509.23883.pdf

Abstract:
Visual Document Retrieval (VDR), the task of retrieving visually-rich document pages using queries that combine visual and textual cues, is crucial for numerous real-world applications. Recent state-of-the-art methods leverage Large Vision-Language Models (LVLMs) in a multi-vector paradigm, representing each document as patch-level embeddings to capture fine-grained details. While highly effective, this approach introduces a critical challenge: prohibitive storage overhead, as storing hundreds of vectors per page makes large-scale deployment costly and impractical. To address this, we introduce DocPruner, the first framework to employ adaptive patch-level embedding pruning for VDR to effectively reduce the storage overhead. DocPruner leverages the intra-document patch attention distribution to dynamically identify and discard redundant embeddings for each document. This adaptive mechanism enables a significant 50-60% reduction in storage for leading multi-vector VDR models with negligible degradation in document retrieval performance. Extensive experiments across more than ten representative datasets validate that DocPruner offers a robust, flexible, and effective solution for building storage-efficient, large-scale VDR systems.

Paperid: 2001, https://arxiv.org/pdf/2508.09945.pdf

Abstract:
Multimodal large language models (MLLMs) have significantly advanced the integration of visual and textual understanding. However, their ability to generate code from multimodal inputs remains limited. In this work, we introduce VisCodex, a unified framework that seamlessly merges vision and coding language models to empower MLLMs with strong multimodal code generation abilities. Leveraging a task vector-based model merging technique, we integrate a state-of-the-art coding LLM into a strong vision-language backbone, while preserving both visual comprehension and advanced coding skills. To support training and evaluation, we introduce the Multimodal Coding Dataset (MCD), a large-scale and diverse collection of 598k samples, including high-quality HTML code, chart image-code pairs, image-augmented StackOverflow QA, and algorithmic problems. Furthermore, we propose InfiBench-V, a novel and challenging benchmark specifically designed to assess models on visually-rich, real-world programming questions that demand a nuanced understanding of both textual and visual contexts. Extensive experiments show that VisCodex achieves state-of-the-art performance among open-source MLLMs and approaches proprietary models like GPT-4o, highlighting the effectiveness of our model merging strategy and new datasets.

Paperid: 2002, https://arxiv.org/pdf/2507.14067.pdf

Abstract:
Vision-language models demand watermarking solutions that protect intellectual property without compromising multimodal coherence. Existing text watermarking methods disrupt visual-textual alignment through biased token selection and static strategies, leaving semantic-critical concepts vulnerable. We propose VLA-Mark, a vision-aligned framework that embeds detectable watermarks while preserving semantic fidelity through cross-modal coordination. Our approach integrates multiscale visual-textual alignment metrics, combining localized patch affinity, global semantic coherence, and contextual attention patterns, to guide watermark injection without model retraining. An entropy-sensitive mechanism dynamically balances watermark strength and semantic preservation, prioritizing visual grounding during low-uncertainty generation phases. Experiments show 7.4% lower PPL and 26.6% higher BLEU than conventional methods, with near-perfect detection (98.8% AUC). The framework demonstrates 96.1\% attack resilience against attacks such as paraphrasing and synonym substitution, while maintaining text-visual consistency, establishing new standards for quality-preserving multimodal watermarking

Paperid: 2003, https://arxiv.org/pdf/2412.09613.pdf

Abstract:
Large Vision-Language Models (VLMs) have been extended to understand both images and videos. Visual token compression is leveraged to reduce the considerable token length of visual inputs. To meet the needs of different tasks, existing high-performance models usually process images and videos separately with different token compression strategies, limiting the capabilities of combining images and videos. To this end, we extend each image into a "static" video and introduce a unified token compression strategy called Progressive Visual Token Compression (PVC), where the tokens of each frame are progressively encoded and adaptively compressed to supplement the information not extracted from previous frames. Video tokens are efficiently compressed with exploiting the inherent temporal redundancy. Images are repeated as static videos, and the spatial details can be gradually supplemented in multiple frames. PVC unifies the token compressing of images and videos. With a limited number of tokens per frame (64 tokens by default), spatial details and temporal changes can still be preserved. Experiments show that our model achieves state-of-the-art performance across various video understanding benchmarks, including long video tasks and fine-grained short video tasks. Meanwhile, our unified token compression strategy incurs no performance loss on image benchmarks, particularly in detail-sensitive tasks.

Paperid: 2004, https://arxiv.org/pdf/2502.17651.pdf

Abstract:
Chart generation aims to generate code to produce charts satisfying the desired visual properties, e.g., texts, layout, color, and type. It has great potential to empower the automatic professional report generation in financial analysis, research presentation, education, and healthcare. In this work, we build a vision-language model (VLM) based multi-agent framework for effective automatic chart generation. Generating high-quality charts requires both strong visual design skills and precise coding capabilities that embed the desired visual properties into code. Such a complex multi-modal reasoning process is difficult for direct prompting of VLMs. To resolve these challenges, we propose METAL, a multi-agent framework that decomposes the task of chart generation into the iterative collaboration among specialized agents. METAL achieves 5.2% improvement over the current best result in the chart generation task. The METAL framework exhibits the phenomenon of test-time scaling: its performance increases monotonically as the logarithmic computational budget grows from 512 to 8192 tokens. In addition, we find that separating different modalities during the critique process of METAL boosts the self-correction capability of VLMs in the multimodal context.

Paperid: 2005, https://arxiv.org/pdf/2412.02172.pdf

Abstract:
The ability of large vision-language models (LVLMs) to critique and correct their reasoning is an essential building block towards their self-improvement. However, a systematic analysis of such capabilities in LVLMs is still lacking. We propose VISCO, the first benchmark to extensively analyze the fine-grained critique and correction capabilities of LVLMs. Compared to existing work that uses a single scalar value to critique the entire reasoning [4], VISCO features dense and fine-grained critique, requiring LVLMs to evaluate the correctness of each step in the chain-of-thought and provide natural language explanations to support their judgments. Extensive evaluation of 24 LVLMs demonstrates that human-written critiques significantly enhance the performance after correction, showcasing the potential of the self-improvement strategy. However, the model-generated critiques are less helpful and sometimes detrimental to the performance, suggesting that critique is the crucial bottleneck. We identified three common patterns in critique failures: failure to critique visual perception, reluctance to "say no", and exaggerated assumption of error propagation. To address these issues, we propose an effective LookBack strategy that revisits the image to verify each piece of information in the initial reasoning. LookBack significantly improves critique and correction performance by up to 13.5%.

Paperid: 2006, https://arxiv.org/pdf/2410.08182.pdf

Abstract:
Existing multimodal retrieval benchmarks primarily focus on evaluating whether models can retrieve and utilize external textual knowledge for question answering. However, there are scenarios where retrieving visual information is either more beneficial or easier to access than textual data. In this paper, we introduce a multimodal retrieval-augmented generation benchmark, MRAG-Bench, in which we systematically identify and categorize scenarios where visually augmented knowledge is better than textual knowledge, for instance, more images from varying viewpoints. MRAG-Bench consists of 16,130 images and 1,353 human-annotated multiple-choice questions across 9 distinct scenarios. With MRAG-Bench, we conduct an evaluation of 10 open-source and 4 proprietary large vision-language models (LVLMs). Our results show that all LVLMs exhibit greater improvements when augmented with images compared to textual knowledge, confirming that MRAG-Bench is vision-centric. Additionally, we conduct extensive analysis with MRAG-Bench, which offers valuable insights into retrieval-augmented LVLMs. Notably, the top-performing model, GPT-4o, faces challenges in effectively leveraging retrieved knowledge, achieving only a 5.82% improvement with ground-truth information, in contrast to a 33.16% improvement observed in human participants. These findings highlight the importance of MRAG-Bench in encouraging the community to enhance LVLMs' ability to utilize retrieved visual knowledge more effectively.

Paperid: 2007, https://arxiv.org/pdf/2510.17771.pdf

Abstract:
Vision-Language Models (VLMs) achieve strong results on multimodal tasks such as visual question answering, yet they can still fail even when the correct visual evidence is present. In this work, we systematically investigate whether these failures arise from not perceiving the evidence or from not leveraging it effectively. By examining layer-wise attention dynamics, we find that shallow layers focus primarily on text, while deeper layers sparsely but reliably attend to localized evidence regions. Surprisingly, VLMs often perceive the visual evidence when outputting incorrect answers, a phenomenon we term ``seeing but not believing'' that widely exists in major VLM families. Building on this, we introduce an inference-time intervention that highlights deep-layer evidence regions through selective attention-based masking. It requires no training and consistently improves accuracy across multiple families, including LLaVA, Qwen, Gemma, and InternVL. These results show that VLMs encode reliable evidence internally but under-utilize it, making such signals explicit can bridge the gap between perception and reasoning, advancing the diagnostic understanding and reliability of VLMs.

Paperid: 2008, https://arxiv.org/pdf/2509.21979.pdf

Abstract:
Vision language models(VLMs) are increasingly integrated into clinical workflows, but they often exhibit sycophantic behavior prioritizing alignment with user phrasing social cues or perceived authority over evidence based reasoning. This study evaluate clinical sycophancy in medical visual question answering through a novel clinically grounded benchmark. We propose a medical sycophancy dataset construct from PathVQA, SLAKE, and VQA-RAD stratified by different type organ system and modality. Using psychologically motivated pressure templates including various sycophancy. In our adversarial experiments on various VLMs, we found that these models are generally vulnerable, exhibiting significant variations in the occurrence of adversarial responses, with weak correlations to the model accuracy or size. Imitation and expert provided corrections were found to be the most effective triggers, suggesting that the models possess a bias mechanism independent of visual evidence. To address this, we propose Visual Information Purification for Evidence based Response (VIPER) a lightweight mitigation strategy that filters non evidentiary content for example social pressures and then generates constrained evidence first answers. This framework reduces sycophancy by an average amount outperforming baselines while maintaining interpretability. Our benchmark analysis and mitigation framework lay the groundwork for robust deployment of medical VLMs in real world clinician interactions emphasizing the need for evidence anchored defenses.

Paperid: 2009, https://arxiv.org/pdf/2508.13103.pdf

Abstract:
Vision-Language-Action (VLA) models frequently encounter challenges in generalizing to real-world environments due to inherent discrepancies between observation and action spaces. Although training data are collected from diverse camera perspectives, the models typically predict end-effector poses within the robot base coordinate frame, resulting in spatial inconsistencies. To mitigate this limitation, we introduce the Observation-Centric VLA (OC-VLA) framework, which grounds action predictions directly in the camera observation space. Leveraging the camera's extrinsic calibration matrix, OC-VLA transforms end-effector poses from the robot base coordinate system into the camera coordinate system, thereby unifying prediction targets across heterogeneous viewpoints. This lightweight, plug-and-play strategy ensures robust alignment between perception and action, substantially improving model resilience to camera viewpoint variations. The proposed approach is readily compatible with existing VLA architectures, requiring no substantial modifications. Comprehensive evaluations on both simulated and real-world robotic manipulation tasks demonstrate that OC-VLA accelerates convergence, enhances task success rates, and improves cross-view generalization. The code will be publicly available.

Paperid: 2010, https://arxiv.org/pdf/2507.02074.pdf

Abstract:
Crash detection from video feeds is a critical problem in intelligent transportation systems. Recent developments in large language models (LLMs) and vision-language models (VLMs) have transformed how we process, reason about, and summarize multimodal information. This paper surveys recent methods leveraging LLMs for crash detection from video data. We present a structured taxonomy of fusion strategies, summarize key datasets, analyze model architectures, compare performance benchmarks, and discuss ongoing challenges and opportunities. Our review provides a foundation for future research in this fast-growing intersection of video understanding and foundation models.

Paperid: 2011, https://arxiv.org/pdf/2506.15480.pdf

Abstract:
Large language models (LLMs) are often supplemented with external knowledge to provide information not encoded in their parameters or to reduce hallucination. In such cases, we expect the model to generate responses by grounding its response in the provided external context. However, prior work has shown that simply appending context at inference time does not ensure grounded generation. To address this, we propose Context-INformed Grounding Supervision (CINGS), a post-training supervision in which the model is trained with relevant context prepended to the response, while computing the loss only over the response tokens and masking out the context. Our experiments demonstrate that models trained with CINGS exhibit stronger grounding in both textual and visual domains compared to standard instruction-tuned models. In the text domain, CINGS outperforms other training methods across 11 information-seeking datasets and is complementary to inference-time grounding techniques. In the vision-language domain, replacing a vision-language model's LLM backbone with a CINGS-trained model reduces hallucinations across four benchmarks and maintains factual consistency throughout the generated response. This improved grounding comes without degradation in general downstream performance. Finally, we analyze the mechanism underlying the enhanced grounding in CINGS and find that it induces a shift in the model's prior knowledge and behavior, implicitly encouraging greater reliance on the external context.

Paperid: 2012, https://arxiv.org/pdf/2506.06275.pdf

Authors:Emmanouil Zaranis, AntÃ³nio Farinhas, Saul Santos, Beatriz Canaverde, Miguel Moura Ramos, Aditya K Surikuchi, AndrÃ© Viveiros, Baohao Liao, Elena Bueno-Benito, Nithin Sivakumaran, Pavlo Vasylenko, Shoubin Yu, Sonal Sannigrahi, Wafaa Mohammed, Ben Peters, Danae SÃ¡nchez Villegas, Elias Stengel-Eskin, Giuseppe Attanasio, Jaehong Yoon, Stella Frank, Alessandro Suglia, Chrysoula Zerva, Desmond Elliott, Mariella Dimiccoli, Mohit Bansal, Oswald Lanz, Raffaella Bernardi, Raquel FernÃ¡ndez, Sandro Pezzelle, Vlad Niculae, AndrÃ© F. T. Martins

Abstract:
Despite recent progress in vision-language models (VLMs), holistic understanding of long-form video content remains a significant challenge, partly due to limitations in current benchmarks. Many focus on peripheral, ``needle-in-a-haystack'' details, encouraging context-insensitive retrieval over deep comprehension. Others rely on large-scale, semi-automatically generated questions (often produced by language models themselves) that are easier for models to answer but fail to reflect genuine understanding. In this paper, we introduce MF$^2$, a new benchmark for evaluating whether models can comprehend, consolidate, and recall key narrative information from full-length movies (50-170 minutes long). MF$^2$ includes over 50 full-length, open-licensed movies, each paired with manually constructed sets of claim pairs -- one true (fact) and one plausible but false (fib), totalling over 850 pairs. These claims target core narrative elements such as character motivations and emotions, causal chains, and event order, and refer to memorable moments that humans can recall without rewatching the movie. Instead of multiple-choice formats, we adopt a binary claim evaluation protocol: for each pair, models must correctly identify both the true and false claims. This reduces biases like answer ordering and enables a more precise assessment of reasoning. Our experiments demonstrate that both open-weight and closed state-of-the-art models fall well short of human performance, underscoring the relative ease of the task for humans and their superior ability to retain and reason over critical narrative information -- an ability current VLMs lack.

Paperid: 2013, https://arxiv.org/pdf/2505.23043.pdf

Abstract:
Recent advancements in unified vision-language models (VLMs), which integrate both visual understanding and generation capabilities, have attracted significant attention. The underlying hypothesis is that a unified architecture with mixed training on both understanding and generation tasks can enable mutual enhancement between understanding and generation. However, this hypothesis remains underexplored in prior works on unified VLMs. To address this gap, this paper systematically investigates the generalization across understanding and generation tasks in unified VLMs. Specifically, we design a dataset closely aligned with real-world scenarios to facilitate extensive experiments and quantitative evaluations. We evaluate multiple unified VLM architectures to validate our findings. Our key findings are as follows. First, unified VLMs trained with mixed data exhibit mutual benefits in understanding and generation tasks across various architectures, and this mutual benefits can scale up with increased data. Second, better alignment between multimodal input and output spaces will lead to better generalization. Third, the knowledge acquired during generation tasks can transfer to understanding tasks, and this cross-task generalization occurs within the base language model, beyond modality adapters. Our findings underscore the critical necessity of unifying understanding and generation in VLMs, offering valuable insights for the design and optimization of unified VLMs.

Paperid: 2014, https://arxiv.org/pdf/2505.13957.pdf

Abstract:
Multimodal Retrieval-Augmented Generation (MRAG) systems enhance LMMs by integrating external multimodal databases, but introduce unexplored privacy vulnerabilities. While text-based RAG privacy risks have been studied, multimodal data presents unique challenges. We provide the first systematic analysis of MRAG privacy vulnerabilities across vision-language and speech-language modalities. Using a novel compositional structured prompt attack in a black-box setting, we demonstrate how attackers can extract private information by manipulating queries. Our experiments reveal that LMMs can both directly generate outputs resembling retrieved content and produce descriptions that indirectly expose sensitive information, highlighting the urgent need for robust privacy-preserving MRAG techniques.

Paperid: 2015, https://arxiv.org/pdf/2505.08617.pdf

Abstract:
While humans can flexibly leverage interactive visual cognition for complex problem-solving, enabling Large Vision-Language Models (LVLMs) to learn similarly adaptive behaviors with visual tools remains challenging. A significant hurdle is the current lack of standardized infrastructure, which hinders integrating diverse tools, generating rich interaction data, and training robust agents effectively. To address these gaps, we introduce OpenThinkIMG, the first open-source, comprehensive end-to-end framework for tool-augmented LVLMs. It features standardized vision tool interfaces, scalable trajectory generation for policy initialization, and a flexible training environment. Furthermore, considering supervised fine-tuning (SFT) on static demonstrations offers limited policy generalization for dynamic tool invocation, we propose a novel reinforcement learning (RL) framework V-ToolRL to train LVLMs to learn adaptive policies for invoking external vision tools. V-ToolRL enables LVLMs to autonomously discover optimal tool-usage strategies by directly optimizing for task success using feedback from tool interactions. We empirically validate V-ToolRL on challenging chart reasoning tasks. Our RL-trained agent, built upon a Qwen2-VL-2B, significantly outperforms its SFT-initialized counterpart (+28.83 points) and surpasses established supervised tool-learning baselines like Taco and CogCom by an average of +12.7 points. Notably, it also surpasses prominent closed-source models like GPT-4.1 by +8.68 accuracy points. We hope OpenThinkIMG can serve as a foundational framework for advancing dynamic, tool-augmented visual reasoning, helping the community develop AI agents that can genuinely "think with images".

Paperid: 2016, https://arxiv.org/pdf/2504.11468.pdf

Abstract:
This work revisits the dominant supervised fine-tuning (SFT) then reinforcement learning (RL) paradigm for training Large Vision-Language Models (LVLMs), and reveals a key finding: SFT can significantly undermine subsequent RL by inducing ``pseudo reasoning paths'' imitated from expert models. While these paths may resemble the native reasoning paths of RL models, they often involve prolonged, hesitant, less informative steps, and incorrect reasoning. To systematically study this effect, we introduce VLAA-Thinking, a new multimodal dataset designed to support reasoning in LVLMs. Constructed via a six-step pipeline involving captioning, reasoning distillation, answer rewrite and verification, VLAA-Thinking comprises high-quality, step-by-step visual reasoning traces for SFT, along with a more challenging RL split from the same data source. Using this dataset, we conduct extensive experiments comparing SFT, RL and their combinations. Results show that while SFT helps models learn reasoning formats, it often locks aligned models into imitative, rigid reasoning modes that impede further learning. In contrast, building on the Group Relative Policy Optimization (GRPO) with a novel mixed reward module integrating both perception and cognition signals, our RL approach fosters more genuine, adaptive reasoning behavior. Notably, our model VLAA-Thinker, based on Qwen2.5VL 3B, achieves top-1 performance on Open LMM Reasoning Leaderboard (https://huggingface.co/spaces/opencompass/Open_LMM_Reasoning_Leaderboard) among 4B scale LVLMs, surpassing the previous state-of-the-art by 1.8%. We hope our findings provide valuable insights in developing reasoning-capable LVLMs and can inform future research in this area.

Paperid: 2017, https://arxiv.org/pdf/2503.19757.pdf

Abstract:
While recent vision-language-action models trained on diverse robot datasets exhibit promising generalization capabilities with limited in-domain data, their reliance on compact action heads to predict discretized or continuous actions constrains adaptability to heterogeneous action spaces. We present Dita, a scalable framework that leverages Transformer architectures to directly denoise continuous action sequences through a unified multimodal diffusion process. Departing from prior methods that condition denoising on fused embeddings via shallow networks, Dita employs in-context conditioning -- enabling fine-grained alignment between denoised actions and raw visual tokens from historical observations. This design explicitly models action deltas and environmental nuances. By scaling the diffusion action denoiser alongside the Transformer's scalability, Dita effectively integrates cross-embodiment datasets across diverse camera perspectives, observation scenes, tasks, and action spaces. Such synergy enhances robustness against various variances and facilitates the successful execution of long-horizon tasks. Evaluations across extensive benchmarks demonstrate state-of-the-art or comparative performance in simulation. Notably, Dita achieves robust real-world adaptation to environmental variances and complex long-horizon tasks through 10-shot finetuning, using only third-person camera inputs. The architecture establishes a versatile, lightweight and open-source baseline for generalist robot policy learning. Project Page: https://robodita.github.io.

Paperid: 2018, https://arxiv.org/pdf/2410.15959.pdf

Abstract:
Recent large vision-language-action models pretrained on diverse robot datasets have demonstrated the potential for generalizing to new environments with a few in-domain data. However, those approaches usually predict individual discretized or continuous action by a small action head, which limits the ability in handling diverse action spaces. In contrast, we model the continuous action sequence with a large multi-modal diffusion transformer, dubbed as Diffusion Transformer Policy, in which we directly denoise action chunks by a large transformer model rather than a small action head for action embedding. By leveraging the scaling capability of transformers, the proposed approach can effectively model continuous end-effector actions across large diverse robot datasets, and achieve better generalization performance. Extensive experiments demonstrate the effectiveness and generalization of Diffusion Transformer Policy on Maniskill2, Libero, Calvin and SimplerEnv, as well as the real-world Franka arm, achieving consistent better performance on Real-to-Sim benchmark SimplerEnv, real-world Franka Arm and Libero compared to OpenVLA and Octo. Specifically, without bells and whistles, the proposed approach achieves state-of-the-art performance with only a single third-view camera stream in the Calvin task ABC->D, improving the average number of tasks completed in a row of 5 to 3.6, and the pretraining stage significantly facilitates the success sequence length on the Calvin by over 1.2.

Paperid: 2019, https://arxiv.org/pdf/2510.22765.pdf

Abstract:
The rapid development of Vision-language models (VLMs) enables open-ended perception and reasoning. Recent works have started to investigate how to adapt general-purpose VLMs into personalized assistants. Even commercial models such as ChatGPT now support model personalization by incorporating user-specific information. However, existing methods either learn a set of concept tokens or train a VLM to utilize user-specific information. However, both pipelines struggle to generate accurate answers as personalized assistants. We introduce Jarvis, an innovative framework for a personalized AI assistant through personal KV-Cache retrieval, which stores user-specific information in the KV-Caches of both textual and visual tokens. The textual tokens are created by summarizing user information into metadata, while the visual tokens are produced by extracting distinct image patches from the user's images. When answering a question, Jarvis first retrieves related KV-Caches from personal storage and uses them to ensure accuracy in responses. We also introduce a fine-grained benchmark built with the same distinct image patch mining pipeline, emphasizing accurate question answering based on fine-grained user-specific information. Jarvis is capable of providing more accurate responses, particularly when they depend on specific local details. Jarvis achieves state-of-the-art results in both visual question answering and text-only tasks across multiple datasets, indicating a practical path toward personalized AI assistants. The code and dataset will be released.

Paperid: 2020, https://arxiv.org/pdf/2509.25032.pdf

Abstract:
As robots transition from controlled settings to unstructured human environments, building generalist agents that can reliably follow natural language instructions remains a central challenge. Progress in robust mobile manipulation requires large-scale multimodal datasets that capture contact-rich and long-horizon tasks, yet existing resources lack synchronized force-torque sensing, hierarchical annotations, and explicit failure cases. We address this gap with the AIRoA MoMa Dataset, a large-scale real-world multimodal dataset for mobile manipulation. It includes synchronized RGB images, joint states, six-axis wrist force-torque signals, and internal robot states, together with a novel two-layer annotation schema of sub-goals and primitive actions for hierarchical learning and error analysis. The initial dataset comprises 25,469 episodes (approx. 94 hours) collected with the Human Support Robot (HSR) and is fully standardized in the LeRobot v2.1 format. By uniquely integrating mobile manipulation, contact-rich interaction, and long-horizon structure, AIRoA MoMa provides a critical benchmark for advancing the next generation of Vision-Language-Action models. The first version of our dataset is now available at https://huggingface.co/datasets/airoa-org/airoa-moma .

Paperid: 2021, https://arxiv.org/pdf/2509.23224.pdf

Abstract:
To improve efficiency and temporal coherence, Vision-Language-Action (VLA) models often predict action chunks; however, this action chunking harms reactivity under inference delay and long horizons. We introduce Asynchronous Action Chunk Correction (A2C2), which is a lightweight real-time chunk correction head that runs every control step and adds a time-aware correction to any off-the-shelf VLA's action chunk. The module combines the latest observation, the predicted action from VLA (base action), a positional feature that encodes the index of the base action within the chunk, and some features from the base policy, then outputs a per-step correction. This preserves the base model's competence while restoring closed-loop responsiveness. The approach requires no retraining of the base policy and is orthogonal to asynchronous execution schemes such as Real Time Chunking (RTC). On the dynamic Kinetix task suite (12 tasks) and LIBERO Spatial, our method yields consistent success rate improvements across increasing delays and execution horizons (+23% point and +7% point respectively, compared to RTC), and also improves robustness for long horizons even with zero injected delay. Since the correction head is small and fast, there is minimal overhead compared to the inference of large VLA models. These results indicate that A2C2 is an effective, plug-in mechanism for deploying high-capacity chunking policies in real-time control.

Paperid: 2022, https://arxiv.org/pdf/2506.22881.pdf

Abstract:
Density ratio estimation is a core concept in statistical machine learning because it provides a unified mechanism for tasks such as importance weighting, divergence estimation, and likelihood-free inference, but its potential in vision and language models has not been fully explored. Modern vision-language encoders such as CLIP and SigLIP are trained with contrastive objectives that implicitly optimize log density ratios between joint and marginal image-text distributions, which implicitly learn similarity scores proportional to log density ratios. However, prior work has largely focused on their embedding utility, and the density-ratio structure induced by contrastive learning has not been systematically examined or exploited in multimodal applications. To address this gap, we reinterpret CLIP-style models as pretrained and general-purpose density ratio estimators and show that this perspective enables new algorithmic capabilities. We present a unified explanation of how contrastive objectives estimate density ratios and propose two practical applications: Importance Weight Learning and KL divergence estimation. Our Importance Weight Learning method requires only a single additional prompt and improves F1 scores by up to 7 points. We further show that CLIP-based density ratios support estimation of KL divergences that quantify how conditioning on an image or text alters the distribution of the other modality. Through qualitative examples and an N-gram analysis of captions, we find that these divergences capture semantic diversity and mode structure in multimodal data. Leveraging this property, we introduce a simple KL-guided data curation method that achieves performance competitive with LAION2B filtering.

Paperid: 2023, https://arxiv.org/pdf/2504.09130.pdf

Abstract:
Recent advancements in Large Vision-Language Models have showcased remarkable capabilities. However, they often falter when confronted with complex reasoning tasks that humans typically address through visual aids and deliberate, step-by-step thinking. While existing methods have explored text-based slow thinking or rudimentary visual assistance, they fall short of capturing the intricate, interleaved nature of human visual-verbal reasoning processes. To overcome these limitations and inspired by the mechanisms of slow thinking in human cognition, we introduce VisuoThink, a novel framework that seamlessly integrates visuospatial and linguistic domains. VisuoThink facilitates multimodal slow thinking by enabling progressive visual-textual reasoning and incorporates test-time scaling through look-ahead tree search. Extensive experiments demonstrate that VisuoThink significantly enhances reasoning capabilities via inference-time scaling, even without fine-tuning, achieving state-of-the-art performance in tasks involving geometry and spatial reasoning.

Paperid: 2024, https://arxiv.org/pdf/2412.18424.pdf

Abstract:
Large vision language models (LVLMs) have improved the document understanding capabilities remarkably, enabling the handling of complex document elements, longer contexts, and a wider range of tasks. However, existing document understanding benchmarks have been limited to handling only a small number of pages and fail to provide a comprehensive analysis of layout elements locating. In this paper, we first define three primary task categories: Long Document Understanding, numerical Reasoning, and cross-element Locating, and then propose a comprehensive benchmark, LongDocURL, integrating above three primary tasks and comprising 20 sub-tasks categorized based on different primary tasks and answer evidences. Furthermore, we develop a semi-automated construction pipeline and collect 2,325 high-quality question-answering pairs, covering more than 33,000 pages of documents, significantly outperforming existing benchmarks. Subsequently, we conduct comprehensive evaluation experiments on both open-source and closed-source models across 26 different configurations, revealing critical performance gaps in this field.

Paperid: 2025, https://arxiv.org/pdf/2410.21169.pdf

Abstract:
Document parsing is essential for converting unstructured and semi-structured documents such as contracts, academic papers, and invoices into structured, machine-readable data. Document parsing reliable structured data from unstructured inputs, providing huge convenience for numerous applications. Especially with recent achievements in Large Language Models, document parsing plays an indispensable role in both knowledge base construction and training data generation. This survey presents a comprehensive review of the current state of document parsing, covering key methodologies, from modular pipeline systems to end-to-end models driven by large vision-language models. Core components such as layout detection, content extraction (including text, tables, and mathematical expressions), and multi-modal data integration are examined in detail. Additionally, this paper discusses the challenges faced by modular document parsing systems and vision-language models in handling complex layouts, integrating multiple modules, and recognizing high-density text. It outlines future research directions and emphasizes the importance of developing larger and more diverse datasets.

Paperid: 2026, https://arxiv.org/pdf/2601.19798.pdf

Abstract:
Despite the significant advancements represented by Vision-Language Models (VLMs), current architectures often exhibit limitations in retaining fine-grained visual information, leading to coarse-grained multimodal comprehension. We attribute this deficiency to a suboptimal training paradigm inherent in prevailing VLMs, which exhibits a text-dominant optimization bias by conceptualizing visual signals merely as passive conditional inputs rather than supervisory targets. To mitigate this, we introduce Youtu-VL, a framework leveraging the Vision-Language Unified Autoregressive Supervision (VLUAS) paradigm, which fundamentally shifts the optimization objective from ``vision-as-input'' to ``vision-as-target.'' By integrating visual tokens directly into the prediction stream, Youtu-VL applies unified autoregressive supervision to both visual details and linguistic content. Furthermore, we extend this paradigm to encompass vision-centric tasks, enabling a standard VLM to perform vision-centric tasks without task-specific additions. Extensive empirical evaluations demonstrate that Youtu-VL achieves competitive performance on both general multimodal tasks and vision-centric tasks, establishing a robust foundation for the development of comprehensive generalist visual agents.

Paperid: 2027, https://arxiv.org/pdf/2512.08186.pdf

Abstract:
While recent large vision-language models (VLMs) have improved generalization in vision-language navigation (VLN), existing methods typically rely on end-to-end pipelines that map vision-language inputs directly to short-horizon discrete actions. Such designs often produce fragmented motions, incur high latency, and struggle with real-world challenges like dynamic obstacle avoidance. We propose DualVLN, the first dual-system VLN foundation model that synergistically integrates high-level reasoning with low-level action execution. System 2, a VLM-based global planner, "grounds slowly" by predicting mid-term waypoint goals via image-grounded reasoning. System 1, a lightweight, multi-modal conditioning Diffusion Transformer policy, "moves fast" by leveraging both explicit pixel goals and latent features from System 2 to generate smooth and accurate trajectories. The dual-system design enables robust real-time control and adaptive local decision-making in complex, dynamic environments. By decoupling training, the VLM retains its generalization, while System 1 achieves interpretable and effective local navigation. DualVLN outperforms prior methods across all VLN benchmarks and real-world experiments demonstrate robust long-horizon planning and real-time adaptability in dynamic environments.

Paperid: 2028, https://arxiv.org/pdf/2511.04570.pdf

Abstract:
"Thinking with Text" and "Thinking with Images" paradigm significantly improve the reasoning ability of large language models (LLMs) and Vision Language Models (VLMs). However, these paradigms have inherent limitations. (1) Images capture only single moments and fail to represent dynamic processes or continuous changes, and (2) The separation of text and vision as distinct modalities, hindering unified multimodal understanding and generation. To overcome these limitations, we introduce "Thinking with Video", a new paradigm that leverages video generation models, such as Sora-2, to bridge visual and textual reasoning in a unified temporal framework. To support this exploration, we developed the Video Thinking Benchmark (VideoThinkBench). VideoThinkBench encompasses two task categories: (1) vision-centric tasks (e.g., Eyeballing Puzzles), and (2) text-centric tasks (e.g., subsets of GSM8K, MMMU). Our evaluation establishes Sora-2 as a capable reasoner. On vision-centric tasks, Sora-2 is generally comparable to state-of-the-art (SOTA) VLMs, and even surpasses VLMs on several tasks, such as Eyeballing Games. On text-centric tasks, Sora-2 achieves 92% accuracy on MATH, and 75.53% accuracy on MMMU. Furthermore, we systematically analyse the source of these abilities. We also find that self-consistency and in-context learning can improve Sora-2's performance. In summary, our findings demonstrate that the video generation model is the potential unified multimodal understanding and generation model, positions "thinking with video" as a unified multimodal reasoning paradigm.

Paperid: 2029, https://arxiv.org/pdf/2510.18596.pdf

Abstract:
Computer-using agents (CUAs) enable task completion through natural interaction with operating systems and software interfaces. While script-based verifiers are widely adopted for evaluation, they suffer from limited scalability and inability to provide step-wise assessment. Reward models offer promising alternatives, but their effectiveness on CUA evaluation remains largely underexplored. To address this gap, we present CUARewardBench, comprising four key contributions: (1) First-ever Comprehensive CUA Reward Benchmark: We introduce the first benchmark for evaluating both outcome reward models (ORM) and process reward models (PRM) on CUA tasks, enabling systematic assessment across trajectory-level and step-level evaluation. (2) Diverse, Practical and Reliable Dataset: CUARewardBench encompasses trajectories from 10 software categories and 7 agent architectures with varying performance levels (25.9%-50.8% success rates). All trajectories are expertly annotated through carefully designed protocols, with rigorous quality control to ensure reliability and practical applicability. (3) Comprehensive Analysis and Insights: Through extensive experiments across 7 vision-language models and 3 prompt templates, we reveal critical limitations of current CUA RMs, including insufficient visual reasoning capabilities, knowledge deficiencies, and the superiority of general VLMs over specialized CUA models for reward evaluation. (4) Unanimous Prompt Ensemble (UPE): Based on the insights from our comprehensive analysis, we propose UPE, a novel ensemble method that significantly enhances reward model reliability through strict unanimous voting and strategic prompt-template configurations. UPE achieves 89.8% precision and 93.3% NPV for ORM, and 81.7% precision and 85.1% NPV for PRM, substantially outperforming single VLMs and traditional ensemble approaches.

Paperid: 2030, https://arxiv.org/pdf/2508.05211.pdf

Abstract:
Large Multimodal Models (LMMs) excel in visual-language tasks by leveraging numerous visual tokens for fine-grained visual information, but this token redundancy results in significant computational costs. Previous research aimed at reducing visual tokens during inference typically leverages importance maps derived from attention scores among vision-only tokens or vision-language tokens to prune tokens across one or multiple pruning stages. Despite this progress, pruning frameworks and strategies remain simplistic and insufficiently explored, often resulting in substantial performance degradation. In this paper, we propose VFlowOpt, a token pruning framework that introduces an importance map derivation process and a progressive pruning module with a recycling mechanism. The hyperparameters of its pruning strategy are further optimized by a visual information flow-guided method. Specifically, we compute an importance map for image tokens based on their attention-derived context relevance and patch-level information entropy. We then decide which tokens to retain or prune and aggregate the pruned ones as recycled tokens to avoid potential information loss. Finally, we apply a visual information flow-guided method that regards the last token in the LMM as the most representative signal of text-visual interactions. This method minimizes the discrepancy between token representations in LMMs with and without pruning, thereby enabling superior pruning strategies tailored to different LMMs. Experiments demonstrate that VFlowOpt can prune 90% of visual tokens while maintaining comparable performance, leading to an 89% reduction in KV-Cache memory and 3.8 times faster inference.

Paperid: 2031, https://arxiv.org/pdf/2507.17520.pdf

Abstract:
To operate effectively in the real world, robots must integrate multimodal reasoning with precise action generation. However, existing vision-language-action (VLA) models often sacrifice one for the other, narrow their abilities to task-specific manipulation data, and suffer catastrophic forgetting of pre-trained vision-language capabilities. To bridge this gap, we introduce InstructVLA, an end-to-end VLA model that preserves the flexible reasoning of large vision-language models (VLMs) while delivering leading manipulation performance. InstructVLA introduces a novel training paradigm, Vision-Language-Action Instruction Tuning (VLA-IT), which employs multimodal training with mixture-of-experts adaptation to jointly optimize textual reasoning and action generation on both standard VLM corpora and a curated 650K-sample VLA-IT dataset. On in-domain SimplerEnv tasks, InstructVLA achieves 30.5% improvement over SpatialVLA. To evaluate generalization, we introduce SimplerEnv-Instruct, an 80-task benchmark requiring closed-loop control and high-level instruction understanding, where it outperforms a fine-tuned OpenVLA by 92% and an action expert aided by GPT-4o by 29%. Additionally, InstructVLA surpasses baseline VLMs on multimodal tasks and exhibits inference-time scaling by leveraging textual reasoning to boost manipulation performance in both simulated and real-world settings. These results demonstrate InstructVLA's potential for bridging intuitive and steerable human-robot interaction with efficient policy learning.

Paperid: 2032, https://arxiv.org/pdf/2506.19816.pdf

Abstract:
Recent vision-language-action (VLA) models built on pretrained vision-language models (VLMs) have demonstrated strong generalization across manipulation tasks. However, they remain constrained by a single-frame observation paradigm and cannot fully benefit from the motion information offered by aggregated multi-frame historical observations, as the large vision-language backbone introduces substantial computational cost and inference latency. We propose CronusVLA, a unified framework that extends single-frame VLA models to the multi-frame paradigm through an efficient post-training stage. CronusVLA comprises three key components: (1) single-frame pretraining on large-scale embodied datasets with autoregressive action tokens prediction, which establishes an embodied vision-language foundation; (2) multi-frame encoding, adapting the prediction of vision-language backbones from discrete action tokens to motion features during post-training, and aggregating motion features from historical frames into a feature chunking; (3) cross-frame decoding, which maps the feature chunking to accurate actions via a shared decoder with cross-attention. By reducing redundant token computation and caching past motion features, CronusVLA achieves efficient inference. As an application of motion features, we further propose an action adaptation mechanism based on feature-action retrieval to improve model performance during finetuning. CronusVLA achieves state-of-the-art performance on SimplerEnv with 70.9% success rate, and 12.7% improvement over OpenVLA on LIBERO. Real-world Franka experiments also show the strong performance and robustness.

Paperid: 2033, https://arxiv.org/pdf/2504.21530.pdf

Abstract:
Recent advancements in robotic manipulation have highlighted the potential of intermediate representations for improving policy generalization. In this work, we explore grounding masks as an effective intermediate representation, balancing two key advantages: (1) effective spatial guidance that specifies target objects and placement areas while also conveying information about object shape and size, and (2) broad generalization potential driven by large-scale vision-language models pretrained on diverse grounding datasets. We introduce RoboGround, a grounding-aware robotic manipulation system that leverages grounding masks as an intermediate representation to guide policy networks in object manipulation tasks. To further explore and enhance generalization, we propose an automated pipeline for generating large-scale, simulated data with a diverse set of objects and instructions. Extensive experiments show the value of our dataset and the effectiveness of grounding masks as intermediate guidance, significantly enhancing the generalization abilities of robot policies.

Paperid: 2034, https://arxiv.org/pdf/2502.01401.pdf

Abstract:
3D visual grounding (3DVG) is challenging due to the need to understand 3D spatial relations. While supervised approaches have achieved superior performance, they are constrained by the scarcity and high annotation costs of 3D vision-language datasets. Training-free approaches based on LLMs/VLMs eliminate the need for large-scale training data, but they either incur prohibitive grounding time and token costs or have unsatisfactory accuracy. To address the challenges, we introduce a novel method for training-free 3D visual grounding, namely Language-to-Space Programming (LaSP). LaSP introduces LLM-generated codes to analyze 3D spatial relations among objects, along with a pipeline that evaluates and optimizes the codes automatically. Experimental results demonstrate that LaSP achieves 52.9% accuracy on the Nr3D benchmark, ranking among the best training-free methods. Moreover, it substantially reduces the grounding time and token costs, offering a balanced trade-off between performance and efficiency.

Paperid: 2035, https://arxiv.org/pdf/2410.21311.pdf

Abstract:
Large Vision-Language Models (LVLMs) have achieved remarkable performance in many vision-language tasks, yet their capabilities in fine-grained visual understanding remain insufficiently evaluated. Existing benchmarks either contain limited fine-grained evaluation samples that are mixed with other data, or are confined to object-level assessments in natural images. To holistically assess LVLMs' fine-grained visual understanding capabilities, we propose using document images with multi-granularity and multi-modal information to supplement natural images. In this light, we construct MMDocBench, a benchmark with various OCR-free document understanding tasks for the evaluation of fine-grained visual perception and reasoning abilities. MMDocBench defines 15 main tasks with 4,338 QA pairs and 11,353 supporting regions, covering various document images such as research papers, receipts, financial reports, Wikipedia tables, charts, and infographics. Based on MMDocBench, we conduct extensive experiments using 13 open-source and 3 proprietary advanced LVLMs, assessing their strengths and weaknesses across different tasks and document image types. The benchmark, task instructions, and evaluation code will be made publicly available.

Paperid: 2036, https://arxiv.org/pdf/2601.14968.pdf

Abstract:
Most existing time series classification methods adopt a discriminative paradigm that maps input sequences directly to one-hot encoded class labels. While effective, this paradigm struggles to incorporate contextual features and fails to capture semantic relationships among classes. To address these limitations, we propose InstructTime, a novel framework that reformulates time series classification as a multimodal generative task. Specifically, continuous numerical sequences, contextual textual features, and task instructions are treated as multimodal inputs, while class labels are generated as textual outputs by tuned language models. To bridge the modality gap, InstructTime introduces a time series discretization module that converts continuous sequences into discrete temporal tokens, together with an alignment projection layer and a generative self-supervised pre-training strategy to enhance cross-modal representation alignment. Building upon this framework, we further propose InstructTime++, which extends InstructTime by incorporating implicit feature modeling to compensate for the limited inductive bias of language models. InstructTime++ leverages specialized toolkits to mine informative implicit patterns from raw time series and contextual inputs, including statistical feature extraction and vision-language-based image captioning, and translates them into textual descriptions for seamless integration. Extensive experiments on multiple benchmark datasets demonstrate the superior performance of InstructTime++.

Paperid: 2037, https://arxiv.org/pdf/2512.06032.pdf

Abstract:
This paper investigates the fundamental discontinuity between the latest two Segment Anything Models: SAM2 and SAM3. We explain why the expertise in prompt-based segmentation of SAM2 does not transfer to the multimodal concept-driven paradigm of SAM3. SAM2 operates through spatial prompts points, boxes, and masks yielding purely geometric and temporal segmentation. In contrast, SAM3 introduces a unified vision-language architecture capable of open-vocabulary reasoning, semantic grounding, contrastive alignment, and exemplar-based concept understanding. We structure this analysis through five core components: (1) a Conceptual Break Between Prompt-Based and Concept-Based Segmentation, contrasting spatial prompt semantics of SAM2 with multimodal fusion and text-conditioned mask generation of SAM3; (2) Architectural Divergence, detailing pure vision-temporal design of SAM2 versus integration of vision-language encoders, geometry and exemplar encoders, fusion modules, DETR-style decoders, object queries, and ambiguity-handling via Mixture-of-Experts in SAM3; (3) Dataset and Annotation Differences, contrasting SA-V video masks with multimodal concept-annotated corpora of SAM3; (4) Training and Hyperparameter Distinctions, showing why SAM2 optimization knowledge does not apply to SAM3; and (5) Evaluation, Metrics, and Failure Modes, outlining the transition from geometric IoU metrics to semantic, open-vocabulary evaluation. Together, these analyses establish SAM3 as a new class of segmentation foundation model and chart future directions for the emerging concept-driven segmentation era.

Paperid: 2038, https://arxiv.org/pdf/2511.01755.pdf

Abstract:
Visual grounding in 3D is the key for embodied agents to localize language-referred objects in open-world environments. However, existing benchmarks are limited to indoor focus, single-platform constraints, and small scale. We introduce 3EED, a multi-platform, multi-modal 3D grounding benchmark featuring RGB and LiDAR data from vehicle, drone, and quadruped platforms. We provide over 128,000 objects and 22,000 validated referring expressions across diverse outdoor scenes -- 10x larger than existing datasets. We develop a scalable annotation pipeline combining vision-language model prompting with human verification to ensure high-quality spatial grounding. To support cross-platform learning, we propose platform-aware normalization and cross-modal alignment techniques, and establish benchmark protocols for in-domain and cross-platform evaluations. Our findings reveal significant performance gaps, highlighting the challenges and opportunities of generalizable 3D grounding. The 3EED dataset and benchmark toolkit are released to advance future research in language-driven 3D embodied perception.

Paperid: 2039, https://arxiv.org/pdf/2511.00908.pdf

Abstract:
Visual geo-localization requires extensive geographic knowledge and sophisticated reasoning to determine image locations without GPS metadata. Traditional retrieval methods are constrained by database coverage and quality. Recent Large Vision-Language Models (LVLMs) enable direct location reasoning from image content, yet individual models struggle with diverse geographic regions and complex scenes. Existing multi-agent systems improve performance through model collaboration but treat all agent interactions uniformly. They lack mechanisms to handle conflicting predictions effectively. We propose \textbf{GraphGeo}, a multi-agent debate framework using heterogeneous graph neural networks for visual geo-localization. Our approach models diverse debate relationships through typed edges, distinguishing supportive collaboration, competitive argumentation, and knowledge transfer. We introduce a dual-level debate mechanism combining node-level refinement and edge-level argumentation modeling. A cross-level topology refinement strategy enables co-evolution between graph structure and agent representations. Experiments on multiple benchmarks demonstrate GraphGeo significantly outperforms state-of-the-art methods. Our framework transforms cognitive conflicts between agents into enhanced geo-localization accuracy through structured debate.

Paperid: 2040, https://arxiv.org/pdf/2509.24183.pdf

Abstract:
GUI agents powered by vision-language models (VLMs) show promise in automating complex digital tasks. However, their effectiveness in real-world applications is often limited by scarce training data and the inherent complexity of these tasks, which frequently require long-tailed knowledge covering rare, unseen scenarios. We propose RAG-GUI , a lightweight VLM that leverages web tutorials at inference time. RAG-GUI is first warm-started via supervised finetuning (SFT) and further refined through self-guided rejection sampling finetuning (RSF). Designed to be model-agnostic, RAG-GUI functions as a generic plug-in that enhances any VLM-based agent. Evaluated across three distinct tasks, it consistently outperforms baseline agents and surpasses other inference baselines by 2.6% to 13.3% across two model sizes, demonstrating strong generalization and practical plug-and-play capabilities in real-world scenarios.

Paperid: 2041, https://arxiv.org/pdf/2508.19294.pdf

Abstract:
The fusion of language and vision in large vision-language models (LVLMs) has revolutionized deep learning-based object detection by enhancing adaptability, contextual reasoning, and generalization beyond traditional architectures. This in-depth review presents a structured exploration of the state-of-the-art in LVLMs, systematically organized through a three-step research review process. First, we discuss the functioning of vision language models (VLMs) for object detection, describing how these models harness natural language processing (NLP) and computer vision (CV) techniques to revolutionize object detection and localization. We then explain the architectural innovations, training paradigms, and output flexibility of recent LVLMs for object detection, highlighting how they achieve advanced contextual understanding for object detection. The review thoroughly examines the approaches used in integration of visual and textual information, demonstrating the progress made in object detection using VLMs that facilitate more sophisticated object detection and localization strategies. This review presents comprehensive visualizations demonstrating LVLMs' effectiveness in diverse scenarios including localization and segmentation, and then compares their real-time performance, adaptability, and complexity to traditional deep learning systems. Based on the review, its is expected that LVLMs will soon meet or surpass the performance of conventional methods in object detection. The review also identifies a few major limitations of the current LVLM modes, proposes solutions to address those challenges, and presents a clear roadmap for the future advancement in this field. We conclude, based on this study, that the recent advancement in LVLMs have made and will continue to make a transformative impact on object detection and robotic applications in the future.

Paperid: 2042, https://arxiv.org/pdf/2507.12465.pdf

Abstract:
3D modeling is moving from virtual to physical. Existing 3D generation primarily emphasizes geometries and textures while neglecting physical-grounded modeling. Consequently, despite the rapid development of 3D generative models, the synthesized 3D assets often overlook rich and important physical properties, hampering their real-world application in physical domains like simulation and embodied AI. As an initial attempt to address this challenge, we propose \textbf{PhysX-3D}, an end-to-end paradigm for physical-grounded 3D asset generation. 1) To bridge the critical gap in physics-annotated 3D datasets, we present PhysXNet - the first physics-grounded 3D dataset systematically annotated across five foundational dimensions: absolute scale, material, affordance, kinematics, and function description. In particular, we devise a scalable human-in-the-loop annotation pipeline based on vision-language models, which enables efficient creation of physics-first assets from raw 3D assets.2) Furthermore, we propose \textbf{PhysXGen}, a feed-forward framework for physics-grounded image-to-3D asset generation, injecting physical knowledge into the pre-trained 3D structural space. Specifically, PhysXGen employs a dual-branch architecture to explicitly model the latent correlations between 3D structures and physical properties, thereby producing 3D assets with plausible physical predictions while preserving the native geometry quality. Extensive experiments validate the superior performance and promising generalization capability of our framework. All the code, data, and models will be released to facilitate future research in generative physical AI.

Paperid: 2043, https://arxiv.org/pdf/2507.00951.pdf

Abstract:
Can machines truly think, reason and act in domains like humans? This enduring question continues to shape the pursuit of Artificial General Intelligence (AGI). Despite the growing capabilities of models such as GPT-4.5, DeepSeek, Claude 3.5 Sonnet, Phi-4, and Grok 3, which exhibit multimodal fluency and partial reasoning, these systems remain fundamentally limited by their reliance on token-level prediction and lack of grounded agency. This paper offers a cross-disciplinary synthesis of AGI development, spanning artificial intelligence, cognitive neuroscience, psychology, generative models, and agent-based systems. We analyze the architectural and cognitive foundations of general intelligence, highlighting the role of modular reasoning, persistent memory, and multi-agent coordination. In particular, we emphasize the rise of Agentic RAG frameworks that combine retrieval, planning, and dynamic tool use to enable more adaptive behavior. We discuss generalization strategies, including information compression, test-time adaptation, and training-free methods, as critical pathways toward flexible, domain-agnostic intelligence. Vision-Language Models (VLMs) are reexamined not just as perception modules but as evolving interfaces for embodied understanding and collaborative task completion. We also argue that true intelligence arises not from scale alone but from the integration of memory and reasoning: an orchestration of modular, interactive, and self-improving components where compression enables adaptive behavior. Drawing on advances in neurosymbolic systems, reinforcement learning, and cognitive scaffolding, we explore how recent architectures begin to bridge the gap between statistical learning and goal-directed cognition. Finally, we identify key scientific, technical, and ethical challenges on the path to AGI.

Paperid: 2044, https://arxiv.org/pdf/2506.17901.pdf

Abstract:
Multimodal Large Language Models (MLLMs) excel in vision-language tasks, such as image captioning and visual question answering. However, they often suffer from over-reliance on spurious correlations, primarily due to linguistic priors that distract the model from leveraging actual visual information. To address these issues, we introduce MMGrounded-PostAlign, a post-multimodal alignment framework designed to enhance the visual understanding capabilities and mitigate the hallucinations of MLLMs. Our framework incorporates a multimodal grounding module for both visual grounding, which identifies the referred object in the image, and textual grounding, which generates the rationale for the final answer, ensuring that outputs are anchored in both visual and textual evidence. To mitigate the hallucinations, we introduce a negative rejection mechanism in the visual grounding module to distinguish grounded entities from non-existent objects influenced by linguistic biases. On the textual grounding side, we propose a selective reasoning mechanism that adjusts the model's reasoning strategy based on query complexity. Extensive evaluations are conducted on benchmarks such as POPE, HaloQuest, VQAv2, MME, and MMBench showing significant improvements in fine-grained visual understanding and hallucination suppression.

Paperid: 2045, https://arxiv.org/pdf/2505.10468.pdf

Abstract:
This study critically distinguishes between AI Agents and Agentic AI, offering a structured conceptual taxonomy, application mapping, and challenge analysis to clarify their divergent design philosophies and capabilities. We begin by outlining the search strategy and foundational definitions, characterizing AI Agents as modular systems driven by Large Language Models (LLMs) and Large Image Models (LIMs) for narrow, task-specific automation. Generative AI is positioned as a precursor, with AI Agents advancing through tool integration, prompt engineering, and reasoning enhancements. In contrast, Agentic AI systems represent a paradigmatic shift marked by multi-agent collaboration, dynamic task decomposition, persistent memory, and orchestrated autonomy. Through a sequential evaluation of architectural evolution, operational mechanisms, interaction styles, and autonomy levels, we present a comparative analysis across both paradigms. Application domains such as customer support, scheduling, and data summarization are contrasted with Agentic AI deployments in research automation, robotic coordination, and medical decision support. We further examine unique challenges in each paradigm including hallucination, brittleness, emergent behavior, and coordination failure and propose targeted solutions such as ReAct loops, RAG, orchestration layers, and causal modeling. This work aims to provide a definitive roadmap for developing robust, scalable, and explainable AI agent and Agentic AI-driven systems. >AI Agents, Agent-driven, Vision-Language-Models, Agentic AI Decision Support System, Agentic-AI Applications

Paperid: 2046, https://arxiv.org/pdf/2505.04769.pdf

Abstract:
Vision-Language-Action (VLA) models mark a transformative advancement in artificial intelligence, aiming to unify perception, natural language understanding, and embodied action within a single computational framework. This foundational review presents a comprehensive synthesis of recent advancements in Vision-Language-Action models, systematically organized across five thematic pillars that structure the landscape of this rapidly evolving field. We begin by establishing the conceptual foundations of VLA systems, tracing their evolution from cross-modal learning architectures to generalist agents that tightly integrate vision-language models (VLMs), action planners, and hierarchical controllers. Our methodology adopts a rigorous literature review framework, covering over 80 VLA models published in the past three years. Key progress areas include architectural innovations, parameter-efficient training strategies, and real-time inference accelerations. We explore diverse application domains such as humanoid robotics, autonomous vehicles, medical and industrial robotics, precision agriculture, and augmented reality navigation. The review further addresses major challenges across real-time control, multimodal action representation, system scalability, generalization to unseen tasks, and ethical deployment risks. Drawing from the state-of-the-art, we propose targeted solutions including agentic AI adaptation, cross-embodiment generalization, and unified neuro-symbolic planning. In our forward-looking discussion, we outline a future roadmap where VLA models, VLMs, and agentic AI converge to power socially aligned, adaptive, and general-purpose embodied agents. This work serves as a foundational reference for advancing intelligent, real-world robotics and artificial general intelligence. >Vision-language-action, Agentic AI, AI Agents, Vision-language Models

Paperid: 2047, https://arxiv.org/pdf/2504.20419.pdf

Abstract:
Automation in agriculture plays a vital role in addressing challenges related to crop monitoring and disease management, particularly through early detection systems. This study investigates the effectiveness of combining multimodal Large Language Models (LLMs), specifically GPT-4o, with Convolutional Neural Networks (CNNs) for automated plant disease classification using leaf imagery. Leveraging the PlantVillage dataset, we systematically evaluate model performance across zero-shot, few-shot, and progressive fine-tuning scenarios. A comparative analysis between GPT-4o and the widely used ResNet-50 model was conducted across three resolutions (100, 150, and 256 pixels) and two plant species (apple and corn). Results indicate that fine-tuned GPT-4o models achieved slightly better performance compared to the performance of ResNet-50, achieving up to 98.12% classification accuracy on apple leaf images, compared to 96.88% achieved by ResNet-50, with improved generalization and near-zero training loss. However, zero-shot performance of GPT-4o was significantly lower, underscoring the need for minimal training. Additional evaluations on cross-resolution and cross-plant generalization revealed the models' adaptability and limitations when applied to new domains. The findings highlight the promise of integrating multimodal LLMs into automated disease detection pipelines, enhancing the scalability and intelligence of precision agriculture systems while reducing the dependence on large, labeled datasets and high-resolution sensor infrastructure. Large Language Models, Vision Language Models, LLMs and CNNs, Disease Detection with Vision Language Models, VLMs

Paperid: 2048, https://arxiv.org/pdf/2504.18738.pdf

Abstract:
This review provides a systematic analysis of comprehensive survey of 3D object detection with vision-language models(VLMs) , a rapidly advancing area at the intersection of 3D vision and multimodal AI. By examining over 100 research papers, we provide the first systematic analysis dedicated to 3D object detection with vision-language models. We begin by outlining the unique challenges of 3D object detection with vision-language models, emphasizing differences from 2D detection in spatial reasoning and data complexity. Traditional approaches using point clouds and voxel grids are compared to modern vision-language frameworks like CLIP and 3D LLMs, which enable open-vocabulary detection and zero-shot generalization. We review key architectures, pretraining strategies, and prompt engineering methods that align textual and 3D features for effective 3D object detection with vision-language models. Visualization examples and evaluation benchmarks are discussed to illustrate performance and behavior. Finally, we highlight current challenges, such as limited 3D-language datasets and computational demands, and propose future research directions to advance 3D object detection with vision-language models. >Object Detection, Vision-Language Models, Agents, VLMs, LLMs, AI

Paperid: 2049, https://arxiv.org/pdf/2504.04744.pdf

Abstract:
Grounding 3D object affordance is a task that locates objects in 3D space where they can be manipulated, which links perception and action for embodied intelligence. For example, for an intelligent robot, it is necessary to accurately ground the affordance of an object and grasp it according to human instructions. In this paper, we introduce a novel task that grounds 3D object affordance based on language instructions, visual observations and interactions, which is inspired by cognitive science. We collect an Affordance Grounding dataset with Points, Images and Language instructions (AGPIL) to support the proposed task. In the 3D physical world, due to observation orientation, object rotation, or spatial occlusion, we can only get a partial observation of the object. So this dataset includes affordance estimations of objects from full-view, partial-view, and rotation-view perspectives. To accomplish this task, we propose LMAffordance3D, the first multi-modal, language-guided 3D affordance grounding network, which applies a vision-language model to fuse 2D and 3D spatial features with semantic features. Comprehensive experiments on AGPIL demonstrate the effectiveness and superiority of our method on this task, even in unseen experimental settings. Our project is available at https://sites.google.com/view/lmaffordance3d.

Paperid: 2050, https://arxiv.org/pdf/2501.04003.pdf

Abstract:
Recent advancements in Vision-Language Models (VLMs) have sparked interest in their use for autonomous driving, particularly in generating interpretable driving decisions through natural language. However, the assumption that VLMs inherently provide visually grounded, reliable, and interpretable explanations for driving remains largely unexamined. To address this gap, we introduce DriveBench, a benchmark dataset designed to evaluate VLM reliability across 17 settings (clean, corrupted, and text-only inputs), encompassing 19,200 frames, 20,498 question-answer pairs, three question types, four mainstream driving tasks, and a total of 12 popular VLMs. Our findings reveal that VLMs often generate plausible responses derived from general knowledge or textual cues rather than true visual grounding, especially under degraded or missing visual inputs. This behavior, concealed by dataset imbalances and insufficient evaluation metrics, poses significant risks in safety-critical scenarios like autonomous driving. We further observe that VLMs struggle with multi-modal reasoning and display heightened sensitivity to input corruptions, leading to inconsistencies in performance. To address these challenges, we propose refined evaluation metrics that prioritize robust visual grounding and multi-modal understanding. Additionally, we highlight the potential of leveraging VLMs' awareness of corruptions to enhance their reliability, offering a roadmap for developing more trustworthy and interpretable decision-making systems in real-world autonomous driving contexts. The benchmark toolkit is publicly accessible.

Paperid: 2051, https://arxiv.org/pdf/2411.06284.pdf

Abstract:
This survey and application guide to multimodal large language models(MLLMs) explores the rapidly developing field of MLLMs, examining their architectures, applications, and impact on AI and Generative Models. Starting with foundational concepts, we delve into how MLLMs integrate various data types, including text, images, video and audio, to enable complex AI systems for cross-modal understanding and generation. It covers essential topics such as training methods, architectural components, and practical applications in various fields, from visual storytelling to enhanced accessibility. Through detailed case studies and technical analysis, the text examines prominent MLLM implementations while addressing key challenges in scalability, robustness, and cross-modal learning. Concluding with a discussion of ethical considerations, responsible AI development, and future directions, this authoritative resource provides both theoretical frameworks and practical insights. It offers a balanced perspective on the opportunities and challenges in the development and deployment of MLLMs, and is highly valuable for researchers, practitioners, and students interested in the intersection of natural language processing and computer vision.

Paperid: 2052, https://arxiv.org/pdf/2410.12564.pdf

Abstract:
Benefiting from the revolutionary advances in large language models (LLMs) and foundational vision models, large vision-language models (LVLMs) have also made significant progress. However, current benchmarks focus on tasks that evaluating only a single aspect of LVLM capabilities (e.g., recognition, detection, understanding). These tasks fail to fully demonstrate LVLMs' potential in complex application scenarios. To comprehensively assess the performance of existing LVLMs, we propose a more challenging task called the Flow Text with Image Insertion task (FTII). This task requires LVLMs to simultaneously possess outstanding abilities in image comprehension, instruction understanding, and long-text interpretation. Specifically, given several text paragraphs and a set of candidate images, as the text paragraphs accumulate, the LVLMs are required to select the most suitable image from the candidates to insert after the corresponding paragraph. Constructing a benchmark for such a task is highly challenging, particularly in determining the sequence of flowing text and images. To address this challenge, we turn to professional news reports, which naturally contain a gold standard for image-text sequences. Based on this, we introduce the Flow Text with Image Insertion Benchmark (FTII-Bench), which includes 318 high-quality Chinese image-text news articles and 307 high-quality English image-text news articles, covering 10 different news domains. Using these 625 high-quality articles, we construct problems of two different types with multiple levels of difficulty. Furthermore, we establish two different evaluation pipelines based on the CLIP model and existing LVLMs. We evaluate 9 open-source and 2 closed-source LVLMs as well as 2 CLIP-based models. Results indicate that even the most advanced models (e.g., GPT-4o) face significant challenges when tackling the FTII task.

Paperid: 2053, https://arxiv.org/pdf/2410.02730.pdf

Abstract:
Large Vision-Language Models (LVLMs) have achieved significant progress in tasks like visual question answering and document understanding. However, their potential to comprehend embodied environments and navigate within them remains underexplored. In this work, we first study the challenge of open-vocabulary object navigation by introducing DivScene, a large-scale dataset with 4,614 houses across 81 scene types and 5,707 kinds of target objects. Our dataset provides a much greater diversity of target objects and scene types than existing datasets, enabling a comprehensive task evaluation. We evaluated various methods with LVLMs and LLMs on our dataset and found that current models still fall short of open-vocab object navigation ability. Then, we fine-tuned LVLMs to predict the next action with CoT explanations. We observe that LVLM's navigation ability can be improved substantially with only BFS-generated shortest paths without any human supervision, surpassing GPT-4o by over 20% in success rates.

Paperid: 2054, https://arxiv.org/pdf/2409.13174.pdf

Abstract:
Recently, driven by advancements in Multimodal Large Language Models (MLLMs), Vision Language Action Models (VLAMs) are being proposed to achieve better performance in open-vocabulary scenarios for robotic manipulation tasks. Since manipulation tasks involve direct interaction with the physical world, ensuring robustness and safety during the execution of this task is always a very critical issue. In this paper, by synthesizing current safety research on MLLMs and the specific application scenarios of the manipulation task in the physical world, we comprehensively evaluate VLAMs in the face of potential physical threats. Specifically, we propose the Physical Vulnerability Evaluating Pipeline (PVEP) that can incorporate as many visual modal physical threats as possible for evaluating the physical robustness of VLAMs. The physical threats in PVEP specifically include Out-of-Distribution, Typography-based Visual Prompt, and Adversarial Patch Attacks. By comparing the performance fluctuations of VLAMs before and after being attacked, we provide generalizable \textbf{\textit{Analyses}} of how VLAMs respond to different physical threats.

Paperid: 2055, https://arxiv.org/pdf/2409.07703.pdf

Abstract:
Large Language Models (LLMs) and Large Vision-Language Models (LVLMs) have demonstrated impressive language/vision reasoning abilities, igniting the recent trend of building agents for targeted applications such as shopping assistants or AI software engineers. Recently, many data science benchmarks have been proposed to investigate their performance in the data science domain. However, existing data science benchmarks still fall short when compared to real-world data science applications due to their simplified settings. To bridge this gap, we introduce DSBench, a comprehensive benchmark designed to evaluate data science agents with realistic tasks. This benchmark includes 466 data analysis tasks and 74 data modeling tasks, sourced from Eloquence and Kaggle competitions. DSBench offers a realistic setting by encompassing long contexts, multimodal task backgrounds, reasoning with large data files and multi-table structures, and performing end-to-end data modeling tasks. Our evaluation of state-of-the-art LLMs, LVLMs, and agents shows that they struggle with most tasks, with the best agent solving only 34.12% of data analysis tasks and achieving a 34.74% Relative Performance Gap (RPG). These findings underscore the need for further advancements in developing more practical, intelligent, and autonomous data science agents.

Paperid: 2056, https://arxiv.org/pdf/2601.21187.pdf

Abstract:
Efficiently enhancing the reasoning capabilities of Vision-Language Models (VLMs) by merging them with Large Reasoning Models (LRMs) has emerged as a promising direction. However, existing methods typically operate at a coarse-grained layer level, which often leads to a trade-off between injecting reasoning capabilities and preserving visual capabilities. To address this limitation, we propose {FRISM} (Fine-grained Reasoning Injection via Subspace-level model Merging), a fine-grained reasoning injection framework based on subspace-level model merging. Observing that reasoning capabilities are encoded in distinct subspaces, FRISM decomposes LRM task vectors via Singular Value Decomposition (SVD) and adaptively tunes the scaling coefficients of each subspace through learning to realize fine-grained reasoning injection. Furthermore, we introduce a label-free self-distillation learning strategy with a dual-objective optimization using common vision-language perception datasets. Extensive experiments demonstrate that FRISM effectively improves reasoning capabilities without compromising the model's original visual capabilities by consistently achieving state-of-the-art performance across diverse visual reasoning benchmarks.

Paperid: 2057, https://arxiv.org/pdf/2512.22983.pdf

Abstract:
Recent advances in vision, language, and multimodal learning have substantially accelerated progress in robotic foundation models, with robot manipulation remaining a central and challenging problem. This survey examines robot manipulation from an algorithmic perspective and organizes recent learning-based approaches within a unified abstraction of high-level planning and low-level control. At the high level, we extend the classical notion of task planning to include reasoning over language, code, motion, affordances, and 3D representations, emphasizing their role in structured and long-horizon decision making. At the low level, we propose a training-paradigm-oriented taxonomy for learning-based control, organizing existing methods along input modeling, latent representation learning, and policy learning. Finally, we identify open challenges and prospective research directions related to scalability, data efficiency, multimodal physical interaction, and safety. Together, these analyses aim to clarify the design space of modern foundation models for robotic manipulation.

Paperid: 2058, https://arxiv.org/pdf/2511.12149.pdf

Abstract:
Vision-Language-Action (VLA) models enable robots to interpret natural-language instructions and perform diverse tasks, yet their integration of perception, language, and control introduces new safety vulnerabilities. Despite growing interest in attacking such models, the effectiveness of existing techniques remains unclear due to the absence of a unified evaluation framework. One major issue is that differences in action tokenizers across VLA architectures hinder reproducibility and fair comparison. More importantly, most existing attacks have not been validated in real-world scenarios. To address these challenges, we propose AttackVLA, a unified framework that aligns with the VLA development lifecycle, covering data construction, model training, and inference. Within this framework, we implement a broad suite of attacks, including all existing attacks targeting VLAs and multiple adapted attacks originally developed for vision-language models, and evaluate them in both simulation and real-world settings. Our analysis of existing attacks reveals a critical gap: current methods tend to induce untargeted failures or static action states, leaving targeted attacks that drive VLAs to perform precise long-horizon action sequences largely unexplored. To fill this gap, we introduce BackdoorVLA, a targeted backdoor attack that compels a VLA to execute an attacker-specified long-horizon action sequence whenever a trigger is present. We evaluate BackdoorVLA in both simulated benchmarks and real-world robotic settings, achieving an average targeted success rate of 58.4% and reaching 100% on selected tasks. Our work provides a standardized framework for evaluating VLA vulnerabilities and demonstrates the potential for precise adversarial manipulation, motivating further research on securing VLA-based embodied systems.

Paperid: 2059, https://arxiv.org/pdf/2510.14902.pdf

Abstract:
Current vision-language-action (VLA) models, pre-trained on large-scale robotic data, exhibit strong multi-task capabilities and generalize well to variations in visual and language instructions for manipulation. However, their success rate drops significantly when faced with object concepts outside the training data, such as unseen object descriptions and textures in the dataset. To address this, we propose a novel agentic framework, VLA^2, which leverages OpenVLA as the execution backbone and effectively leverages external modules such as web retrieval and object detection to provide visual and textual knowledge about target objects to the VLA. This approach mitigates generalization failure when handling out-of-distribution objects. Based on the LIBERO simulation environment, we introduced novel objects and object descriptions to construct a new evaluation benchmark with three difficulty levels to test the effectiveness of our method. Our framework successfully outperformed the current state-of-the-art models on our designed hard-level generalization benchmark. Compared to the standalone OpenVLA baseline, VLA^2 achieves a 44.2% improvement in the success rate in the hard-level benchmark and an average improvement of 20.2% in all customized environments without any performance degradation on in-domain tasks. Project website: https://vla-2.github.io.

Paperid: 2060, https://arxiv.org/pdf/2510.12276.pdf

Abstract:
Vision-language-action (VLA) models have recently shown strong potential in enabling robots to follow language instructions and execute precise actions. However, most VLAs are built upon vision-language models pretrained solely on 2D data, which lack accurate spatial awareness and hinder their ability to operate in the 3D physical world. Existing solutions attempt to incorporate explicit 3D sensor inputs such as depth maps or point clouds, but these approaches face challenges due to sensor noise, hardware heterogeneity, and incomplete depth coverage in existing datasets. Alternative methods that estimate 3D cues from 2D images also suffer from the limited performance of depth estimators.We propose Spatial Forcing (SF), a simple yet effective alignment strategy that implicitly forces VLA models to develop spatial comprehension capabilities without relying on explicit 3D inputs or depth estimators. SF aligns intermediate visual embeddings of VLAs with geometric representations produced by pretrained 3D foundation models. By enforcing alignment at intermediate layers, SF guides VLAs to encode richer spatial representations that enhance action precision.Extensive experiments in simulation and real-world environments demonstrate that SF achieves state-of-the-art results, surpassing both 2D- and 3D-based VLAs. SF further accelerates training by up to 3.8x and improves data efficiency across diverse robotic tasks. Project page is at https://spatial-forcing.github.io/

Paperid: 2061, https://arxiv.org/pdf/2510.10932.pdf

Abstract:
With the growing deployment of Vision-Language-Action (VLA) models in real-world embodied AI systems, their increasing vulnerability to backdoor attacks poses a serious safety threat. A backdoored VLA agent can be covertly triggered by a pre-injected backdoor to execute adversarial actions, potentially causing system failures or even physical harm. Although backdoor attacks on VLA models have been explored, prior work has focused only on untargeted attacks, leaving the more practically threatening scenario of targeted manipulation unexamined. In this paper, we study targeted backdoor attacks on VLA models and introduce TabVLA, a novel framework that enables such attacks via black-box fine-tuning. TabVLA explores two deployment-relevant inference-time threat models: input-stream editing and in-scene triggering. It formulates poisoned data generation as an optimization problem to improve attack effectivess. Experiments with OpenVLA-7B on the LIBERO benchmark reveal that the vision channel is the principal attack surface: targeted backdoors succeed with minimal poisoning, remain robust across variations in trigger design, and are degraded only by positional mismatches between fine-tuning and inference triggers. We also investigate a potential detection-based defense against TabVLA, which reconstructs latent visual triggers from the input stream to flag activation-conditioned backdoor samples. Our work highlights the vulnerability of VLA models to targeted backdoor manipulation and underscores the need for more advanced defenses.

Paperid: 2062, https://arxiv.org/pdf/2509.24304.pdf

Abstract:
While Large Vision-Language Models (LVLMs) have achieved substantial progress in video understanding, their application to long video reasoning is hindered by uniform frame sampling and static textual reasoning, which are inefficient and struggle to handle visually intensive video tasks. To overcome these challenges, in this paper, we introduce the concept of thinking with long videos and propose a novel framework FrameThinker. Within this framework, LVLMs are able to iteratively interrogate video content. Developing such video reasoning capabilities in LVLMs presents notable challenges, particularly in adapting the model to new video actions (e.g. select frame), and designing reward functions to guide LVLMs to adopt the newly introduced action. To solve these challenges, we propose a two-phase training strategy, first employing Supervised Fine-Tuning (SFT) to instill fundamental action capabilities, followed by Reinforcement Learning (RL) to optimize a strategic decision-making policy. Notably, in this RL phase, we conduct an in-depth and comprehensive exploration of the reward design for each action and format reward. Extensive experiments on reasoning benchmarks like Video-Holmes, LongVideo-Reason, and long-video understanding benchmarks such as LongVideoBench, MLVU, VideoMME, and LVBench, demonstrate that FrameThinker achieves a significant average improvement of +10.4% over baselines while drastically reducing the number of processed frames. Most notably, our 7B model, FrameThinker establishes a new state-of-the-art on LongVideo-Reason, achieving 76.1% accuracy using an average of only 20.6 frames. This not only outperforms the competitive LongVILA-R1 (72.0%) but does so with over 20x fewer frames (vs. 512), demonstrating unparalleled efficiency and effectiveness.

Paperid: 2063, https://arxiv.org/pdf/2509.06992.pdf

Abstract:
Federated Prompt Tuning (FPT) is an efficient method for cross-client collaborative fine-tuning of large Vision-Language Models (VLMs). However, models tuned using FPT are vulnerable to adversarial attacks, leading to misclassification in downstream tasks. In this work, we introduce Federated Adversarial Prompt Tuning (\textbf{FedAPT}), a novel method designed to enhance the adversarial robustness of FPT. We identify a key issue in FedAPT under non-independent and identically distributed (non-IID) settings: a \textit{class information gap} between clients and the global model. Clients rely solely on limited local label information to generate adversarial samples for training, while the global model must defend against adversarial attacks from global labels. To address this issue, we propose a \textbf{class-aware prompt generator} that generates visual prompts from text prompts. This generator is guided by a \emph{Global Label Embedding} (serving as a ``beacon") which encodes cross-client label information to create more globally-aligned visual prompts. Additionally, we propose a \textbf{cross-layer generator sharing} strategy to enhance prompt coupling across different layers of the model, further boosting adversarial robustness. Extensive experiments on multiple image classification datasets demonstrate the superiority of FedAPT in improving adversarial robustness, outperforming existing methods by a large margin. FedAPT also exhibits exceptional generalization in cross-domain and cross-dataset scenarios, indicating its effectiveness in real-world applications.

Paperid: 2064, https://arxiv.org/pdf/2508.19958.pdf

Abstract:
Vision-Language-Action (VLA) models have become a cornerstone in robotic policy learning, leveraging large-scale multimodal data for robust and scalable control. However, existing VLA frameworks primarily address short-horizon tasks, and their effectiveness on long-horizon, multi-step robotic manipulation remains limited due to challenges in skill chaining and subtask dependencies. In this work, we introduce Long-VLA, the first end-to-end VLA model specifically designed for long-horizon robotic tasks. Our approach features a novel phase-aware input masking strategy that adaptively segments each subtask into moving and interaction phases, enabling the model to focus on phase-relevant sensory cues and enhancing subtask compatibility. This unified strategy preserves the scalability and data efficiency of VLA training, and our architecture-agnostic module can be seamlessly integrated into existing VLA models. We further propose the L-CALVIN benchmark to systematically evaluate long-horizon manipulation. Extensive experiments on both simulated and real-world tasks demonstrate that Long-VLA significantly outperforms prior state-of-the-art methods, establishing a new baseline for long-horizon robotic control.

Paperid: 2065, https://arxiv.org/pdf/2507.04447.pdf

Abstract:
Recent advances in vision-language-action (VLA) models have shown promise in integrating image generation with action prediction to improve generalization and reasoning in robot manipulation. However, existing methods are limited to challenging image-based forecasting, which suffers from redundant information and lacks comprehensive and critical world knowledge, including dynamic, spatial and semantic information. To address these limitations, we propose DreamVLA, a novel VLA framework that integrates comprehensive world knowledge forecasting to enable inverse dynamics modeling, thereby establishing a perception-prediction-action loop for manipulation tasks. Specifically, DreamVLA introduces a dynamic-region-guided world knowledge prediction, integrated with the spatial and semantic cues, which provide compact yet comprehensive representations for action planning. This design aligns with how humans interact with the world by first forming abstract multimodal reasoning chains before acting. To mitigate interference among the dynamic, spatial and semantic information during training, we adopt a block-wise structured attention mechanism that masks their mutual attention, preventing information leakage and keeping each representation clean and disentangled. Moreover, to model the conditional distribution over future actions, we employ a diffusion-based transformer that disentangles action representations from shared latent features. Extensive experiments on both real-world and simulation environments demonstrate that DreamVLA achieves 76.7% success rate on real robot tasks and 4.44 average length on the CALVIN ABC-D benchmarks.

Paperid: 2066, https://arxiv.org/pdf/2506.12706.pdf

Abstract:
Vision-Language Models (VLMs) such as CLIP have demonstrated remarkable capabilities in understanding relationships between visual and textual data through joint embedding spaces. Despite their effectiveness, these models remain vulnerable to adversarial attacks, particularly in the image modality, posing significant security concerns. Building upon our previous work on Adversarial Prompt Tuning (AdvPT), which introduced learnable text prompts to enhance adversarial robustness in VLMs without extensive parameter training, we present a significant extension by introducing the Neural Augmentor framework for Multi-modal Adversarial Prompt Tuning (NAP-Tuning).Our key innovations include: (1) extending AdvPT from text-only to multi-modal prompting across both text and visual modalities, (2) expanding from single-layer to multi-layer prompt architectures, and (3) proposing a novel architecture-level redesign through our Neural Augmentor approach, which implements feature purification to directly address the distortions introduced by adversarial attacks in feature space. Our NAP-Tuning approach incorporates token refiners that learn to reconstruct purified features through residual connections, allowing for modality-specific and layer-specific feature correction.Comprehensive experiments demonstrate that NAP-Tuning significantly outperforms existing methods across various datasets and attack types. Notably, our approach shows significant improvements over the strongest baselines under the challenging AutoAttack benchmark, outperforming them by 33.5% on ViT-B16 and 33.0% on ViT-B32 architectures while maintaining competitive clean accuracy.

Paperid: 2067, https://arxiv.org/pdf/2505.21200.pdf

Abstract:
Vision-Language-Action (VLA) models have emerged as a powerful paradigm for general-purpose robot control through natural language instructions. However, their high inference cost-stemming from large-scale token computation and autoregressive decoding-poses significant challenges for real-time deployment and edge applications. While prior work has primarily focused on architectural optimization, we take a different perspective by identifying a dual form of redundancy in VLA models: (i) high similarity across consecutive action steps, and (ii) substantial redundancy in visual tokens. Motivated by these observations, we propose FlashVLA, the first training-free and plug-and-play acceleration framework that enables action reuse in VLA models. FlashVLA improves inference efficiency through a token-aware action reuse mechanism that avoids redundant decoding across stable action steps, and an information-guided visual token selection strategy that prunes low-contribution tokens. Extensive experiments on the LIBERO benchmark show that FlashVLA reduces FLOPs by 55.7% and latency by 36.0%, with only a 0.7% drop in task success rate. These results demonstrate the effectiveness of FlashVLA in enabling lightweight, low-latency VLA inference without retraining.

Paperid: 2068, https://arxiv.org/pdf/2505.15875.pdf

Abstract:
With more open-source models available for diverse tasks, model merging has gained attention by combining models into one, reducing training, storage, and inference costs. Current research mainly focuses on model merging for full fine-tuning, overlooking the popular LoRA. However, our empirical analysis reveals that: a) existing merging methods designed for full fine-tuning perform poorly on LoRA; b) LoRA modules show much larger parameter magnitude variance than full fine-tuned weights; c) greater parameter magnitude variance correlates with worse merging performance. Considering that large magnitude variances cause deviations in the distribution of the merged parameters, resulting in information loss and performance degradation, we propose a Decoupled and Orthogonal merging approach(DO-Merging). By separating parameters into magnitude and direction components and merging them independently, we reduce the impact of magnitude differences on the directional alignment of the merged models, thereby preserving task information. Furthermore, we introduce a data-free, layer-wise gradient descent method with orthogonal constraints to mitigate interference during the merging of direction components. We provide theoretical guarantees for both the decoupling and orthogonal components. And we validate through extensive experiments across vision, language, and multi-modal domains that our proposed DO-Merging can achieve significantly higher performance than existing merging methods at a minimal cost. Notably, each component can be flexibly integrated with existing methods, offering near free-lunch improvements across tasks.

Paperid: 2069, https://arxiv.org/pdf/2505.11214.pdf

Abstract:
Vision-Language-Action (VLA) models have recently become highly prominent in the field of robotics. Leveraging vision-language foundation models trained on large-scale internet data, the VLA model can generate robotic actions directly from visual observations and human instructions through a single end-to-end neural network. Despite their effectiveness, current VLA models usually accept only one form of human prompting, language instructions, which may constrain their applicability in open-ended human-robot interactions. For example, a user might expect the robot to retrieve an object shown in an image, follow an instruction written on the whiteboard, or imitate a behavior demonstrated in a video, rather than relying solely on language-based descriptions. To address this gap, we introduce OE-VLA, which explores the potential of VLA models for open-ended multimodal instructions. Extensive results demonstrate that our OE-VLA not only achieves comparable performance to traditional VLA models with linguistic input but also delivers impressive results across four additional categories of open-ended tasks. The proposed methodology could significantly expand the applications of VLA models across various everyday scenarios and facilitate human-robot interaction.

Paperid: 2070, https://arxiv.org/pdf/2505.07395.pdf

Abstract:
Vision-Language-Action (VLA) models have shown great potential in general robotic decision-making tasks via imitation learning. However, the variable quality of training data often constrains the performance of these models. On the other hand, offline Reinforcement Learning (RL) excels at learning robust policy models from mixed-quality data. In this paper, we introduce Reinforced robot GPT (ReinboT), a novel end-to-end VLA model that integrates the RL principle of maximizing cumulative reward. ReinboT achieves a deeper understanding of the data quality distribution by predicting dense returns that capture the nuances of manipulation tasks. The dense return prediction capability enables the robot to generate more robust decision-making actions, oriented towards maximizing future benefits. Extensive experiments show that ReinboT achieves state-of-the-art performance on the CALVIN mixed-quality dataset and exhibits superior few-shot learning and out-of-distribution generalization capabilities in real-world tasks.

Paperid: 2071, https://arxiv.org/pdf/2504.06263.pdf

Abstract:
Scalable Vector Graphics (SVG) is an important image format widely adopted in graphic design because of their resolution independence and editability. The study of generating high-quality SVG has continuously drawn attention from both designers and researchers in the AIGC community. However, existing methods either produces unstructured outputs with huge computational cost or is limited to generating monochrome icons of over-simplified structures. To produce high-quality and complex SVG, we propose OmniSVG, a unified framework that leverages pre-trained Vision-Language Models (VLMs) for end-to-end multimodal SVG generation. By parameterizing SVG commands and coordinates into discrete tokens, OmniSVG decouples structural logic from low-level geometry for efficient training while maintaining the expressiveness of complex SVG structure. To further advance the development of SVG synthesis, we introduce MMSVG-2M, a multimodal dataset with two million richly annotated SVG assets, along with a standardized evaluation protocol for conditional SVG generation tasks. Extensive experiments show that OmniSVG outperforms existing methods and demonstrates its potential for integration into professional SVG design workflows.

Paperid: 2072, https://arxiv.org/pdf/2503.12821.pdf

Abstract:
Large Vision-Language Models (LVLMs) have achieved significant progress in combining visual comprehension with language generation. Despite this success, the training data of LVLMs still suffers from Long-Tail (LT) problems, where the data distribution is highly imbalanced. Previous works have mainly focused on traditional VLM architectures, i.e., CLIP or ViT, and specific tasks such as recognition and classification. Nevertheless, the exploration of LVLM (e.g. LLaVA) and more general tasks (e.g. Visual Question Answering and Visual Reasoning) remains under-explored. In this paper, we first conduct an in-depth analysis of the LT issues in LVLMs and identify two core causes: the overrepresentation of head concepts and the underrepresentation of tail concepts. Based on the above observation, we propose an $\textbf{A}$daptive $\textbf{D}$ata $\textbf{R}$efinement Framework ($\textbf{ADR}$), which consists of two stages: $\textbf{D}$ata $\textbf{R}$ebalancing ($\textbf{DR}$) and $\textbf{D}$ata $\textbf{S}$ynthesis ($\textbf{DS}$). In the DR stage, we adaptively rebalance the redundant data based on entity distributions, while in the DS stage, we leverage Denoising Diffusion Probabilistic Models (DDPMs) and scarce images to supplement underrepresented portions. Through comprehensive evaluations across eleven benchmarks, our proposed ADR effectively mitigates the long-tail problem in the training data, improving the average performance of LLaVA 1.5 relatively by 4.36%, without increasing the training data volume.

Paperid: 2073, https://arxiv.org/pdf/2503.08007.pdf

Abstract:
Developing versatile quadruped robots that can smoothly perform various actions and tasks in real-world environments remains a significant challenge. This paper introduces a novel vision-language-action (VLA) model, mixture of robotic experts (MoRE), for quadruped robots that aim to introduce reinforcement learning (RL) for fine-tuning large-scale VLA models with a large amount of mixed-quality data. MoRE integrates multiple low-rank adaptation modules as distinct experts within a dense multi-modal large language model (MLLM), forming a sparse-activated mixture-of-experts model. This design enables the model to effectively adapt to a wide array of downstream tasks. Moreover, we employ a reinforcement learning-based training objective to train our model as a Q-function after deeply exploring the structural properties of our tasks. Effective learning from automatically collected mixed-quality data enhances data efficiency and model performance. Extensive experiments demonstrate that MoRE outperforms all baselines across six different skills and exhibits superior generalization capabilities in out-of-distribution scenarios. We further validate our method in real-world scenarios, confirming the practicality of our approach and laying a solid foundation for future research on multi-task learning in quadruped robots.

Paperid: 2074, https://arxiv.org/pdf/2502.13508.pdf

Abstract:
Vision-language-action models (VLAs) have become increasingly popular in robot manipulation for their end-to-end design and remarkable performance. However, existing VLAs rely heavily on vision-language models (VLMs) that only support text-based instructions, neglecting the more natural speech modality for human-robot interaction. Traditional speech integration methods usually involves a separate speech recognition system, which complicates the model and introduces error propagation. Moreover, the transcription procedure would lose non-semantic information in the raw speech, such as voiceprint, which may be crucial for robots to successfully complete customized tasks. To overcome above challenges, we propose VLAS, a novel end-to-end VLA that integrates speech recognition directly into the robot policy model. VLAS allows the robot to understand spoken commands through inner speech-text alignment and produces corresponding actions to fulfill the task. We also present two new datasets, SQA and CSI, to support a three-stage tuning process for speech instructions, which empowers VLAS with the ability of multimodal interaction across text, image, speech, and robot actions. Taking a step further, a voice retrieval-augmented generation (RAG) paradigm is designed to enable our model to effectively handle tasks that require individual-specific knowledge. Our extensive experiments show that VLAS can effectively accomplish robot manipulation tasks with diverse speech commands, offering a seamless and customized interaction experience.

Paperid: 2075, https://arxiv.org/pdf/2502.09268.pdf

Abstract:
With the rapid development of embodied artificial intelligence, significant progress has been made in vision-language-action (VLA) models for general robot decision-making. However, the majority of existing VLAs fail to account for the inevitable external perturbations encountered during deployment. These perturbations introduce unforeseen state information to the VLA, resulting in inaccurate actions and consequently, a significant decline in generalization performance. The classic internal model control (IMC) principle demonstrates that a closed-loop system with an internal model that includes external input signals can accurately track the reference input and effectively offset the disturbance. We propose a novel closed-loop VLA method GEVRM that integrates the IMC principle to enhance the robustness of robot visual manipulation. The text-guided video generation model in GEVRM can generate highly expressive future visual planning goals. Simultaneously, we evaluate perturbations by simulating responses, which are called internal embeddings and optimized through prototype contrastive learning. This allows the model to implicitly infer and distinguish perturbations from the external environment. The proposed GEVRM achieves state-of-the-art performance on both standard and perturbed CALVIN benchmarks and shows significant improvements in realistic robot tasks.

Paperid: 2076, https://arxiv.org/pdf/2502.08869.pdf

Abstract:
Time series analysis has witnessed the inspiring development from traditional autoregressive models, deep learning models, to recent Transformers and Large Language Models (LLMs). Efforts in leveraging vision models for time series analysis have also been made along the way but are less visible to the community due to the predominant research on sequence modeling in this domain. However, the discrepancy between continuous time series and the discrete token space of LLMs, and the challenges in explicitly modeling the correlations of variates in multivariate time series have shifted some research attentions to the equally successful Large Vision Models (LVMs) and Vision Language Models (VLMs). To fill the blank in the existing literature, this survey discusses the advantages of vision models over LLMs in time series analysis. It provides a comprehensive and in-depth overview of the existing methods, with dual views of detailed taxonomy that answer the key research questions including how to encode time series as images and how to model the imaged time series for various tasks. Additionally, we address the challenges in the pre- and post-processing steps involved in this framework and outline future directions to further advance time series analysis with vision models.

Paperid: 2077, https://arxiv.org/pdf/2412.15576.pdf

Abstract:
This paper addresses the inherent inference latency challenges associated with deploying multimodal large language models (MLLM) in quadruped vision-language-action (QUAR-VLA) tasks. Our investigation reveals that conventional parameter reduction techniques ultimately impair the performance of the language foundation model during the action instruction tuning phase, making them unsuitable for this purpose. We introduce a novel latency-free quadruped MLLM model, dubbed QUART-Online, designed to enhance inference efficiency without degrading the performance of the language foundation model. By incorporating Action Chunk Discretization (ACD), we compress the original action representation space, mapping continuous action values onto a smaller set of discrete representative vectors while preserving critical information. Subsequently, we fine-tune the MLLM to integrate vision, language, and compressed actions into a unified semantic space. Experimental results demonstrate that QUART-Online operates in tandem with the existing MLLM system, achieving real-time inference in sync with the underlying controller frequency, significantly boosting the success rate across various tasks by 65%. Our project page is https://quart-online.github.io.

Paperid: 2078, https://arxiv.org/pdf/2411.15244.pdf

Abstract:
Large pre-trained Vision-Language Models (VLMs) such as Contrastive Language-Image Pre-training (CLIP) have been shown to be susceptible to adversarial attacks, raising concerns about their deployment in safety-critical applications like autonomous driving and medical diagnosis. One promising approach for robustifying pre-trained VLMs is Adversarial Prompt Tuning (APT), which applies adversarial training during the process of prompt tuning. However, existing APT methods are mostly single-modal methods that design prompt(s) for only the visual or textual modality, limiting their effectiveness in either robustness or clean accuracy. In this work, we propose Adversarial Prompt Distillation (APD), a bimodal knowledge distillation framework that enhances APT by integrating it with multi-modal knowledge transfer. APD optimizes prompts for both visual and textual modalities while distilling knowledge from a clean pre-trained teacher CLIP model. Extensive experiments on multiple benchmark datasets demonstrate the superiority of our APD method over the current state-of-the-art APT methods in terms of both adversarial robustness and clean accuracy. The effectiveness of APD also validates the possibility of using a non-robust teacher to improve the generalization and robustness of fine-tuned VLMs.

Paperid: 2079, https://arxiv.org/pdf/2510.10231.pdf

Abstract:
The rapid advancement of AI-generated content (AIGC) has enabled the synthesis of visually convincing images; however, many such outputs exhibit subtle \textbf{semantic anomalies}, including unrealistic object configurations, violations of physical laws, or commonsense inconsistencies, which compromise the overall plausibility of the generated scenes. Detecting these semantic-level anomalies is essential for assessing the trustworthiness of AIGC media, especially in AIGC image analysis, explainable deepfake detection and semantic authenticity assessment. In this paper, we formalize \textbf{semantic anomaly detection and reasoning} for AIGC images and introduce \textbf{AnomReason}, a large-scale benchmark with structured annotations as quadruples \emph{(Name, Phenomenon, Reasoning, Severity)}. Annotations are produced by a modular multi-agent pipeline (\textbf{AnomAgent}) with lightweight human-in-the-loop verification, enabling scale while preserving quality. At construction time, AnomAgent processed approximately 4.17\,B GPT-4o tokens, providing scale evidence for the resulting structured annotations. We further show that models fine-tuned on AnomReason achieve consistent gains over strong vision-language baselines under our proposed semantic matching metric (\textit{SemAP} and \textit{SemF1}). Applications to {explainable deepfake detection} and {semantic reasonableness assessment of image generators} demonstrate practical utility. In summary, AnomReason and AnomAgent serve as a foundation for measuring and improving the semantic plausibility of AI-generated images. We will release code, metrics, data, and task-aligned models to support reproducible research on semantic authenticity and interpretable AIGC forensics.

Paperid: 2080, https://arxiv.org/pdf/2506.16730.pdf

Abstract:
Infrared and visible image fusion (IVF) aims to combine complementary information from both image modalities, producing more informative and comprehensive outputs. Recently, text-guided IVF has shown great potential due to its flexibility and versatility. However, the effective integration and utilization of textual semantic information remains insufficiently studied. To tackle these challenges, we introduce textual semantics at two levels: the mask semantic level and the text semantic level, both derived from textual descriptions extracted by large Vision-Language Models (VLMs). Building on this, we propose Textual Semantic Guidance for infrared and visible image fusion, termed TeSG, which guides the image synthesis process in a way that is optimized for downstream tasks such as detection and segmentation. Specifically, TeSG consists of three core components: a Semantic Information Generator (SIG), a Mask-Guided Cross-Attention (MGCA) module, and a Text-Driven Attentional Fusion (TDAF) module. The SIG generates mask and text semantics based on textual descriptions. The MGCA module performs initial attention-based fusion of visual features from both infrared and visible images, guided by mask semantics. Finally, the TDAF module refines the fusion process with gated attention driven by text semantics. Extensive experiments demonstrate the competitiveness of our approach, particularly in terms of performance on downstream tasks, compared to existing state-of-the-art methods.

Paperid: 2081, https://arxiv.org/pdf/2505.24875.pdf

Abstract:
Although chain-of-thought reasoning and reinforcement learning (RL) have driven breakthroughs in NLP, their integration into generative vision models remains underexplored. We introduce ReasonGen-R1, a two-stage framework that first imbues an autoregressive image generator with explicit text-based "thinking" skills via supervised fine-tuning on a newly generated reasoning dataset of written rationales, and then refines its outputs using Group Relative Policy Optimization. To enable the model to reason through text before generating images, We automatically generate and release a corpus of model crafted rationales paired with visual prompts, enabling controlled planning of object layouts, styles, and scene compositions. Our GRPO algorithm uses reward signals from a pretrained vision language model to assess overall visual quality, optimizing the policy in each update. Evaluations on GenEval, DPG, and the T2I benchmark demonstrate that ReasonGen-R1 consistently outperforms strong baselines and prior state-of-the-art models. More: aka.ms/reasongen.

Paperid: 2082, https://arxiv.org/pdf/2504.16083.pdf

Abstract:
The integration of long-context capabilities with visual understanding unlocks unprecedented potential for Vision Language Models (VLMs). However, the quadratic attention complexity during the pre-filling phase remains a significant obstacle to real-world deployment. To overcome this limitation, we introduce MMInference (Multimodality Million tokens Inference), a dynamic sparse attention method that accelerates the prefilling stage for long-context multi-modal inputs. First, our analysis reveals that the temporal and spatial locality of video input leads to a unique sparse pattern, the Grid pattern. Simultaneously, VLMs exhibit markedly different sparse distributions across different modalities. We introduce a permutation-based method to leverage the unique Grid pattern and handle modality boundary issues. By offline search the optimal sparse patterns for each head, MMInference constructs the sparse distribution dynamically based on the input. We also provide optimized GPU kernels for efficient sparse computations. Notably, MMInference integrates seamlessly into existing VLM pipelines without any model modifications or fine-tuning. Experiments on multi-modal benchmarks-including Video QA, Captioning, VisionNIAH, and Mixed-Modality NIAH-with state-of-the-art long-context VLMs (LongVila, LlavaVideo, VideoChat-Flash, Qwen2.5-VL) show that MMInference accelerates the pre-filling stage by up to 8.3x at 1M tokens while maintaining accuracy. Our code is available at https://aka.ms/MMInference.

Paperid: 2083, https://arxiv.org/pdf/2503.23024.pdf

Abstract:
Driven by the great success of Large Language Models (LLMs) in the 2D image domain, their applications in 3D scene understanding has emerged as a new trend. A key difference between 3D and 2D is that the situation of an egocentric observer in 3D scenes can change, resulting in different descriptions (e.g., ''left" or ''right"). However, current LLM-based methods overlook the egocentric perspective and simply use datasets from a global viewpoint. To address this issue, we propose a novel approach to automatically generate a situation-aware dataset by leveraging the scanning trajectory during data collection and utilizing Vision-Language Models (VLMs) to produce high-quality captions and question-answer pairs. Furthermore, we introduce a situation grounding module to explicitly predict the position and orientation of observer's viewpoint, thereby enabling LLMs to ground situation description in 3D scenes. We evaluate our approach on several benchmarks, demonstrating that our method effectively enhances the 3D situational awareness of LLMs while significantly expanding existing datasets and reducing manual effort.

Paperid: 2084, https://arxiv.org/pdf/2412.19142.pdf

Abstract:
Recent works in 3D multimodal learning have made remarkable progress. However, typically 3D multimodal models are only capable of handling point clouds. Compared to the emerging 3D representation technique, 3D Gaussian Splatting (3DGS), the spatially sparse point cloud cannot depict the texture information of 3D objects, resulting in inferior reconstruction capabilities. This limitation constrains the potential of point cloud-based 3D multimodal representation learning. In this paper, we present CLIP-GS, a novel multimodal representation learning framework grounded in 3DGS. We introduce the GS Tokenizer to generate serialized gaussian tokens, which are then processed through transformer layers pre-initialized with weights from point cloud models, resulting in the 3DGS embeddings. CLIP-GS leverages contrastive loss between 3DGS and the visual-text embeddings of CLIP, and we introduce an image voting loss to guide the directionality and convergence of gradient optimization. Furthermore, we develop an efficient way to generate triplets of 3DGS, images, and text, facilitating CLIP-GS in learning unified multimodal representations. Leveraging the well-aligned multimodal representations, CLIP-GS demonstrates versatility and outperforms point cloud-based models on various 3D tasks, including multimodal retrieval, zero-shot, and few-shot classification.

Paperid: 2085, https://arxiv.org/pdf/2411.18203.pdf

Abstract:
Vision-language models (VLMs) have shown remarkable advancements in multimodal reasoning tasks. However, they still often generate inaccurate or irrelevant responses due to issues like hallucinated image understandings or unrefined reasoning paths. To address these challenges, we introduce Critic-V, a novel framework inspired by the Actor-Critic paradigm to boost the reasoning capability of VLMs. This framework decouples the reasoning process and critic process by integrating two independent components: the Reasoner, which generates reasoning paths based on visual and textual inputs, and the Critic, which provides constructive critique to refine these paths. In this approach, the Reasoner generates reasoning responses according to text prompts, which can evolve iteratively as a policy based on feedback from the Critic. This interaction process was theoretically driven by a reinforcement learning framework where the Critic offers natural language critiques instead of scalar rewards, enabling more nuanced feedback to boost the Reasoner's capability on complex reasoning tasks. The Critic model is trained using Direct Preference Optimization (DPO), leveraging a preference dataset of critiques ranked by Rule-based Reward~(RBR) to enhance its critic capabilities. Evaluation results show that the Critic-V framework significantly outperforms existing methods, including GPT-4V, on 5 out of 8 benchmarks, especially regarding reasoning accuracy and efficiency. Combining a dynamic text-based policy for the Reasoner and constructive feedback from the preference-optimized Critic enables a more reliable and context-sensitive multimodal reasoning process. Our approach provides a promising solution to enhance the reliability of VLMs, improving their performance in real-world reasoning-heavy multimodal applications such as autonomous driving and embodied intelligence.

Paperid: 2086, https://arxiv.org/pdf/2409.08598.pdf

Abstract:
Existing facial expression recognition (FER) methods typically fine-tune a pre-trained visual encoder using discrete labels. However, this form of supervision limits to specify the emotional concept of different facial expressions. In this paper, we observe that the rich knowledge in text embeddings, generated by vision-language models, is a promising alternative for learning discriminative facial expression representations. Inspired by this, we propose a novel knowledge-enhanced FER method with an emotional-to-neutral transformation. Specifically, we formulate the FER problem as a process to match the similarity between a facial expression representation and text embeddings. Then, we transform the facial expression representation to a neutral representation by simulating the difference in text embeddings from textual facial expression to textual neutral. Finally, a self-contrast objective is introduced to pull the facial expression representation closer to the textual facial expression, while pushing it farther from the neutral representation. We conduct evaluation with diverse pre-trained visual encoders including ResNet-18 and Swin-T on four challenging facial expression datasets. Extensive experiments demonstrate that our method significantly outperforms state-of-the-art FER methods. The code will be publicly available.

Paperid: 2087, https://arxiv.org/pdf/2511.11984.pdf

Abstract:
Fine-grained glomerular subtyping is central to kidney biopsy interpretation, but clinically valuable labels are scarce and difficult to obtain. Existing computational pathology approaches instead tend to evaluate coarse diseased classification under full supervision with image-only models, so it remains unclear how vision-language models (VLMs) should be adapted for clinically meaningful subtyping under data constraints. In this work, we model fine-grained glomerular subtyping as a clinically realistic few-shot problem and systematically evaluate both pathology-specialized and general-purpose vision-language models under this setting. We assess not only classification performance (accuracy, AUC, F1) but also the geometry of the learned representations, examining feature alignment between image and text embeddings and the separability of glomerular subtypes. By jointly analyzing shot count, model architecture and domain knowledge, and adaptation strategy, this study provides guidance for future model selection and training under real clinical data constraints. Our results indicate that pathology-specialized vision-language backbones, when paired with the vanilla fine-tuning, are the most effective starting point. Even with only 4-8 labeled examples per glomeruli subtype, these models begin to capture distinctions and show substantial gains in discrimination and calibration, though additional supervision continues to yield incremental improvements. We also find that the discrimination between positive and negative examples is as important as image-text alignment. Overall, our results show that supervision level and adaptation strategy jointly shape both diagnostic performance and multimodal structure, providing guidance for model selection, adaptation strategies, and annotation investment.

Paperid: 2088, https://arxiv.org/pdf/2510.20278.pdf

Abstract:
In recent years, Pretrained Large Models(PLMs) researchers proposed large-small model collaboration frameworks, leveraged easily trainable small models to assist large models, aim to(1) significantly reduce computational resource consumption while maintaining comparable accuracy, and (2) enhance large model performance in specialized domain tasks. However, this collaborative paradigm suffers from issues such as significant accuracy degradation, exacerbated catastrophic forgetting, and amplified hallucination problems induced by small model knowledge. To address these challenges, we propose a KAN-based Collaborative Model (KCM) as an improved approach to large-small model collaboration. The KAN utilized in KCM represents an alternative neural network architecture distinct from conventional MLPs. Compared to MLPs, KAN offers superior visualizability and interpretability while mitigating catastrophic forgetting. We deployed KCM in large-small model collaborative systems across three scenarios: language, vision, and vision-language cross-modal tasks. The experimental results demonstrate that, compared with pure large model approaches, the large-small model collaboration framework utilizing KCM as the collaborative model significantly reduces the number of large model inference calls while maintaining near-identical task accuracy, thereby substantially lowering computational resource consumption. Concurrently, the KAN-based small collaborative model markedly mitigates catastrophic forgetting, leading to significant accuracy improvements for long-tail data. The results reveal that KCM demonstrates superior performance across all metrics compared to MLP-based small collaborative models (MCM).

Paperid: 2089, https://arxiv.org/pdf/2510.04217.pdf

Abstract:
Multimodal large language models (MLLMs) have demonstrated remarkable capabilities across vision-language tasks, yet their large-scale deployment raises pressing concerns about memorized private data, outdated knowledge, and harmful content. Existing unlearning approaches for MLLMs typically adapt training-based strategies such as gradient ascent or preference optimization, but these methods are computationally expensive, irreversible, and often distort retained knowledge. In this work, we propose MLLMEraser, an input-aware, training-free framework for test-time unlearning. Our approach leverages activation steering to enable dynamic knowledge erasure without parameter updates. Specifically, we construct a multimodal erasure direction by contrasting adversarially perturbed, knowledge-recall image-text pairs with knowledge-erasure counterparts, capturing both textual and visual discrepancies. To prevent unnecessary interference, we further design an input-aware steering mechanism that adaptively determines when and how the erasure direction should be applied, preserving utility on retained knowledge while enforcing forgetting on designated content. Experiments on LLaVA-1.5 and Qwen-2.5-VL demonstrate that MLLMEraser consistently outperforms state-of-the-art MLLM unlearning baselines, achieving stronger forgetting performance with lower computational cost and minimal utility degradation.

Paperid: 2090, https://arxiv.org/pdf/2509.07596.pdf

Abstract:
Gender bias in vision-language foundation models (VLMs) raises concerns about their safe deployment and is typically evaluated using benchmarks with gender annotations on real-world images. However, as these benchmarks often contain spurious correlations between gender and non-gender features, such as objects and backgrounds, we identify a critical oversight in gender bias evaluation: Do spurious features distort gender bias evaluation? To address this question, we systematically perturb non-gender features across four widely used benchmarks (COCO-gender, FACET, MIAP, and PHASE) and various VLMs to quantify their impact on bias evaluation. Our findings reveal that even minimal perturbations, such as masking just 10% of objects or weakly blurring backgrounds, can dramatically alter bias scores, shifting metrics by up to 175% in generative VLMs and 43% in CLIP variants. This suggests that current bias evaluations often reflect model responses to spurious features rather than gender bias, undermining their reliability. Since creating spurious feature-free benchmarks is fundamentally challenging, we recommend reporting bias metrics alongside feature-sensitivity measurements to enable a more reliable bias assessment.

Paperid: 2091, https://arxiv.org/pdf/2508.20570.pdf

Abstract:
Typographic attacks exploit multi-modal systems by injecting text into images, leading to targeted misclassifications, malicious content generation and even Vision-Language Model jailbreaks. In this work, we analyze how CLIP vision encoders behave under typographic attacks, locating specialized attention heads in the latter half of the model's layers that causally extract and transmit typographic information to the cls token. Building on these insights, we introduce a method to defend CLIP models against typographic attacks by selectively ablating a typographic circuit, consisting of attention heads. Without requiring finetuning, our method improves performance by up to 19.6% on a typographic variant of ImageNet-100, while reducing standard ImageNet-100 accuracy by less than 1%. Notably, our training-free approach remains competitive with current state-of-the-art typographic defenses that rely on finetuning. To this end, we release a family of dyslexic CLIP models which are significantly more robust against typographic attacks. These models serve as suitable drop-in replacements for a broad range of safety-critical applications, where the risks of text-based manipulation outweigh the utility of text recognition.

Paperid: 2092, https://arxiv.org/pdf/2508.20345.pdf

Abstract:
Recent advances in medical vision-language models (VLMs) open up remarkable opportunities for clinical applications such as automated report generation, copilots for physicians, and uncertainty quantification. However, despite their promise, medical VLMs introduce serious security concerns, most notably risks of Protected Health Information (PHI) exposure, data leakage, and vulnerability to cyberthreats - which are especially critical in hospital environments. Even when adopted for research or non-clinical purposes, healthcare organizations must exercise caution and implement safeguards. To address these challenges, we present MedFoundationHub, a graphical user interface (GUI) toolkit that: (1) enables physicians to manually select and use different models without programming expertise, (2) supports engineers in efficiently deploying medical VLMs in a plug-and-play fashion, with seamless integration of Hugging Face open-source models, and (3) ensures privacy-preserving inference through Docker-orchestrated, operating system agnostic deployment. MedFoundationHub requires only an offline local workstation equipped with a single NVIDIA A6000 GPU, making it both secure and accessible within the typical resources of academic research labs. To evaluate current capabilities, we engaged board-certified pathologists to deploy and assess five state-of-the-art VLMs (Google-MedGemma3-4B, Qwen2-VL-7B-Instruct, Qwen2.5-VL-7B-Instruct, and LLaVA-1.5-7B/13B). Expert evaluation covered colon cases and renal cases, yielding 1015 clinician-model scoring events. These assessments revealed recurring limitations, including off-target answers, vague reasoning, and inconsistent pathology terminology.

Paperid: 2093, https://arxiv.org/pdf/2508.15960.pdf

Abstract:
Vision-language models (VLMs) have shown considerable potential in digital pathology, yet their effectiveness remains limited for fine-grained, disease-specific classification tasks such as distinguishing between glomerular subtypes. The subtle morphological variations among these subtypes, combined with the difficulty of aligning visual patterns with precise clinical terminology, make automated diagnosis in renal pathology particularly challenging. In this work, we explore how large pretrained VLMs can be effectively adapted to perform fine-grained glomerular classification, even in scenarios where only a small number of labeled examples are available. In this work, we introduce Glo-VLMs, a systematic framework designed to explore the adaptation of VLMs to fine-grained glomerular classification in data-constrained settings. Our approach leverages curated pathology images alongside clinical text prompts to facilitate joint image-text representation learning for nuanced renal pathology subtypes. By assessing various VLMs architectures and adaptation strategies under a few-shot learning paradigm, we explore how both the choice of method and the amount of labeled data impact model performance in clinically relevant scenarios. To ensure a fair comparison, we evaluate all models using standardized multi-class metrics, aiming to clarify the practical requirements and potential of large pretrained models for specialized clinical research applications. As a result, fine-tuning the VLMs achieved 0.7416 accuracy, 0.9045 macro-AUC, and 0.5277 F1-score with only 8 shots per class, demonstrating that even with highly limited supervision, foundation models can be effectively adapted for fine-grained medical image classification.

Paperid: 2094, https://arxiv.org/pdf/2506.08927.pdf

Abstract:
Recent research in vision-language models (VLMs) has centered around the possibility of equipping them with implicit long-form chain-of-thought reasoning -- akin to the success observed in language models -- via distillation and reinforcement learning. But what about the non-reasoning models already trained and deployed across the internet? Should we simply abandon them, or is there hope for a search mechanism that can elicit hidden knowledge and induce long reasoning traces -- without any additional training or supervision? In this paper, we explore this possibility using a Monte Carlo Tree Search (MCTS)-inspired algorithm, which injects subquestion-subanswer pairs into the model's output stream. We show that framing reasoning as a search process -- where subquestions act as latent decisions within a broader inference trajectory -- helps the model "connect the dots" between fragmented knowledge and produce extended reasoning traces in non-reasoning models. We evaluate our method across three benchmarks and observe consistent improvements. Notably, our approach yields a 2% overall improvement on MMMU-PRO, including a significant 9% gain in Liberal Arts.

Paperid: 2095, https://arxiv.org/pdf/2409.11402.pdf

Abstract:
We introduce NVLM 1.0, a family of frontier-class multimodal large language models (LLMs) that achieve state-of-the-art results on vision-language tasks, rivaling the leading proprietary models (e.g., GPT-4o) and open-access models (e.g., Llama 3-V 405B and InternVL 2). Remarkably, NVLM 1.0 shows improved text-only performance over its LLM backbone after multimodal training. In terms of model design, we perform a comprehensive comparison between decoder-only multimodal LLMs (e.g., LLaVA) and cross-attention-based models (e.g., Flamingo). Based on the strengths and weaknesses of both approaches, we propose a novel architecture that enhances both training efficiency and multimodal reasoning capabilities. Furthermore, we introduce a 1-D tile-tagging design for tile-based dynamic high-resolution images, which significantly boosts performance on multimodal reasoning and OCR-related tasks. Regarding training data, we meticulously curate and provide detailed information on our multimodal pretraining and supervised fine-tuning datasets. Our findings indicate that dataset quality and task diversity are more important than scale, even during the pretraining phase, across all architectures. Notably, we develop production-grade multimodality for the NVLM-1.0 models, enabling them to excel in vision-language tasks while maintaining and even improving text-only performance compared to their LLM backbones. To achieve this, we craft and integrate a high-quality text-only dataset into multimodal training, alongside a substantial amount of multimodal math and reasoning data, leading to enhanced math and coding capabilities across modalities. To advance research in the field, we release the model weights at https://huggingface.co/nvidia/NVLM-D-72B and will open-source the training code for the community soon.

Paperid: 2096, https://arxiv.org/pdf/2412.08687.pdf

Abstract:
With the growing adoption and capabilities of vision-language models (VLMs) comes the need for benchmarks that capture authentic user-VLM interactions. In response, we create VisionArena, a dataset of 230K real-world conversations between users and VLMs. Collected from Chatbot Arena - an open-source platform where users interact with VLMs and submit preference votes - VisionArena spans 73K unique users, 45 VLMs, and 138 languages. Our dataset contains three subsets: VisionArena-Chat, 200k single and multi-turn conversations between a user and a VLM; VisionArena-Battle, 30K conversations comparing two anonymous VLMs with user preference votes; and VisionArena-Bench, an automatic benchmark of 500 diverse user prompts that efficiently approximate the live Chatbot Arena model rankings. Additionally, we highlight the types of question asked by users, the influence of response style on preference, and areas where models often fail. We find open-ended tasks like captioning and humor are highly style-dependent, and current VLMs struggle with spatial reasoning and planning tasks. Lastly, we show finetuning the same base model on VisionArena-Chat outperforms Llava-Instruct-158K, with a 17-point gain on MMMU and a 46-point gain on the WildVision benchmark. Dataset at https://huggingface.co/lmarena-ai

Paperid: 2097, https://arxiv.org/pdf/2511.22441.pdf

Abstract:
Images shared on social media often expose geographic cues. While early geolocation methods required expert effort and lacked generalization, the rise of Large Vision Language Models (LVLMs) now enables accurate geolocation even for ordinary users. However, existing approaches are not optimized for this task. To explore the full potential and associated privacy risks, we present Geo-Detective, an agent that mimics human reasoning and tool use for image geolocation inference. It follows a procedure with four steps that adaptively selects strategies based on image difficulty and is equipped with specialized tools such as visual reverse search, which emulates how humans gather external geographic clues. Experimental results show that GEO-Detective outperforms baseline large vision language models (LVLMs) overall, particularly on images lacking visible geographic features. In country level geolocation tasks, it achieves an improvement of over 11.1% compared to baseline LLMs, and even at finer grained levels, it still provides around a 5.2% performance gain. Meanwhile, when equipped with external clues, GEO-Detective becomes more likely to produce accurate predictions, reducing the "unknown" prediction rate by more than 50.6%. We further explore multiple defense strategies and find that Geo-Detective exhibits stronger robustness, highlighting the need for more effective privacy safeguards.

Paperid: 2098, https://arxiv.org/pdf/2511.00181.pdf

Abstract:
The rapid evolution of AI-generated images poses unprecedented challenges to information integrity and media authenticity. Existing detection approaches suffer from fundamental limitations: traditional classifiers lack interpretability and fail to generalize across evolving generative models, while vision-language models (VLMs), despite their promise, remain constrained to single-shot analysis and pixel-level reasoning. To address these challenges, we introduce AIFo (Agent-based Image Forensics), a novel training-free framework that emulates human forensic investigation through multi-agent collaboration. Unlike conventional methods, our framework employs a set of forensic tools, including reverse image search, metadata extraction, pre-trained classifiers, and VLM analysis, coordinated by specialized LLM-based agents that collect, synthesize, and reason over cross-source evidence. When evidence is conflicting or insufficient, a structured multi-agent debate mechanism allows agents to exchange arguments and reach a reliable conclusion. Furthermore, we enhance the framework with a memory-augmented reasoning module that learns from historical cases to improve future detection accuracy. Our comprehensive evaluation spans 6,000 images across both controlled laboratory settings and challenging real-world scenarios, including images from modern generative platforms and diverse online sources. AIFo achieves 97.05% accuracy, substantially outperforming traditional classifiers and state-of-the-art VLMs. These results demonstrate that agent-based procedural reasoning offers a new paradigm for more robust, interpretable, and adaptable AI-generated image detection.

Paperid: 2099, https://arxiv.org/pdf/2509.25717.pdf

Abstract:
Direct Preference Optimization (DPO) has recently been extended from text-only models to vision-language models. However, existing methods rely on oversimplified pairwise comparisons, generating a single negative image via basic perturbations or similarity-based retrieval, which fail to capture the complex nature of multimodal preferences, inducing optimization bias and hallucinations. To address this issue, we propose MISP-DPO, the first framework to incorporate multiple, semantically diverse negative images in multimodal DPO via the Plackett-Luce model. Our method embeds prompts and candidate images in CLIP (Contrastive Language-Image Pretraining) space and applies a sparse autoencoder to uncover semantic deviations into interpretable factors. Negative samples are selected based on reconstruction difficulty, semantic deviation from the positive, and mutual diversity, yielding broader and more informative supervision. To handle multi-negative comparisons, we adopt a Plackett-Luce objective and introduce an importance sampling strategy that improves training efficiency. Experiments across five diverse benchmarks demonstrate that MISP-DPO consistently improves multimodal alignment over prior methods, validating the effectiveness of semantic-aware, multi-negative sampling in preference-based learning.

Paperid: 2100, https://arxiv.org/pdf/2509.04744.pdf

Abstract:
Recent advances in Multimodal Large Language Models (MLLMs) have demonstrated impressive capabilities across various vision-language tasks. However, their reasoning abilities in the multimodal symbolic music domain remain largely unexplored. We introduce WildScore, the first in-the-wild multimodal symbolic music reasoning and analysis benchmark, designed to evaluate MLLMs' capacity to interpret real-world music scores and answer complex musicological queries. Each instance in WildScore is sourced from genuine musical compositions and accompanied by authentic user-generated questions and discussions, capturing the intricacies of practical music analysis. To facilitate systematic evaluation, we propose a systematic taxonomy, comprising both high-level and fine-grained musicological ontologies. Furthermore, we frame complex music reasoning as multiple-choice question answering, enabling controlled and scalable assessment of MLLMs' symbolic music understanding. Empirical benchmarking of state-of-the-art MLLMs on WildScore reveals intriguing patterns in their visual-symbolic reasoning, uncovering both promising directions and persistent challenges for MLLMs in symbolic music reasoning and analysis. We release the dataset and code.

Paperid: 2101, https://arxiv.org/pdf/2507.22617.pdf

Abstract:
Recent advances in text-to-image diffusion models have enabled the creation of a new form of digital art: optical illusions--visual tricks that create different perceptions of reality. However, adversaries may misuse such techniques to generate hateful illusions, which embed specific hate messages into harmless scenes and disseminate them across web communities. In this work, we take the first step toward investigating the risks of scalable hateful illusion generation and the potential for bypassing current content moderation models. Specifically, we generate 1,860 optical illusions using Stable Diffusion and ControlNet, conditioned on 62 hate messages. Of these, 1,571 are hateful illusions that successfully embed hate messages, either overtly or subtly, forming the Hateful Illusion dataset. Using this dataset, we evaluate the performance of six moderation classifiers and nine vision language models (VLMs) in identifying hateful illusions. Experimental results reveal significant vulnerabilities in existing moderation models: the detection accuracy falls below 0.245 for moderation classifiers and below 0.102 for VLMs. We further identify a critical limitation in their vision encoders, which mainly focus on surface-level image details while overlooking the secondary layer of information, i.e., hidden messages. To address this risk, we explore preliminary mitigation measures and identify the most effective approaches from the perspectives of image transformations and training-level strategies.

Paperid: 2102, https://arxiv.org/pdf/2507.11155.pdf

Abstract:
Vision-language models (VLMs) are increasingly applied to identify unsafe or inappropriate images due to their internal ethical standards and powerful reasoning abilities. However, it is still unclear whether they can recognize various unsafe concepts when presented in different modalities, such as text and images. To address this, we first compile the UnsafeConcepts dataset, featuring 75 unsafe concepts, i.e., ``Swastika,'' ``Sexual Harassment,'' and ``Assaults,'' along with associated 1.5K images. We then conduct a systematic evaluation of VLMs' perception (concept recognition) and alignment (ethical reasoning) capabilities. We assess eight popular VLMs and find that, although most VLMs accurately perceive unsafe concepts, they sometimes mistakenly classify these concepts as safe. We also identify a consistent modality gap among open-source VLMs in distinguishing between visual and textual unsafe concepts. To bridge this gap, we introduce a simplified reinforcement learning (RL)-based approach using proximal policy optimization (PPO) to strengthen the ability to identify unsafe concepts from images. Our approach uses reward scores based directly on VLM responses, bypassing the need for collecting human-annotated preference data to train a new reward model. Experimental results show that our approach effectively enhances VLM alignment on images while preserving general capabilities. It outperforms baselines such as supervised fine-tuning (SFT) and direct preference optimization (DPO). We hope our dataset, evaluation findings, and proposed alignment solution contribute to the community's efforts in advancing safe VLMs.

Paperid: 2103, https://arxiv.org/pdf/2507.06899.pdf

Abstract:
Graphical User Interface (GUI) agents powered by Large Vision-Language Models (LVLMs) have emerged as a revolutionary approach to automating human-machine interactions, capable of autonomously operating personal devices (e.g., mobile phones) or applications within the device to perform complex real-world tasks in a human-like manner. However, their close integration with personal devices raises significant security concerns, with many threats, including backdoor attacks, remaining largely unexplored. This work reveals that the visual grounding of GUI agent-mapping textual plans to GUI elements-can introduce vulnerabilities, enabling new types of backdoor attacks. With backdoor attack targeting visual grounding, the agent's behavior can be compromised even when given correct task-solving plans. To validate this vulnerability, we propose VisualTrap, a method that can hijack the grounding by misleading the agent to locate textual plans to trigger locations instead of the intended targets. VisualTrap uses the common method of injecting poisoned data for attacks, and does so during the pre-training of visual grounding to ensure practical feasibility of attacking. Empirical results show that VisualTrap can effectively hijack visual grounding with as little as 5% poisoned data and highly stealthy visual triggers (invisible to the human eye); and the attack can be generalized to downstream tasks, even after clean fine-tuning. Moreover, the injected trigger can remain effective across different GUI environments, e.g., being trained on mobile/web and generalizing to desktop environments. These findings underscore the urgent need for further research on backdoor attack risks in GUI agents.

Paperid: 2104, https://arxiv.org/pdf/2503.13383.pdf

Abstract:
The hypothesis that pretrained large language models (LLMs) necessitate only minimal supervision during the fine-tuning (SFT) stage (Zhou et al., 2024) has been substantiated by recent advancements in data curation and selection research. However, their stability and generalizability are compromised due to the vulnerability to experimental setups and validation protocols, falling short of surpassing random sampling (Diddee & Ippolito, 2024; Xia et al., 2024b). Built upon LLMs, multi-modal LLMs (MLLMs), combined with the sheer token volume and heightened heterogeneity of data sources, amplify both the significance and complexity of data selection. To harvest multi-modal instructional data in a robust and efficient manner, we re-define the granularity of the quality metric by decomposing it into 14 vision-language-related capabilities, and introduce multi-modal rich scorers to evaluate the capabilities of each data candidate. To promote diversity, in light of the inherent objective of the alignment stage, we take interaction style as diversity indicator and use a multi-modal rich styler to identify data instruction patterns. In doing so, our multi-modal rich scorers and styler (mmSSR) guarantee that high-scoring information is conveyed to users in diversified forms. Free from embedding-based clustering or greedy sampling, mmSSR efficiently scales to millions of data with varying budget constraints, supports customization for general or specific capability acquisition, and facilitates training-free generalization to new domains for curation. Across 10+ experimental settings, validated by 14 multi-modal benchmarks, we demonstrate consistent improvements over random sampling, baseline strategies and state-of-the-art selection methods, achieving 99.1% of full performance with only 30% of the 2.6M data.

Paperid: 2105, https://arxiv.org/pdf/2411.00304.pdf

Abstract:
In recent times, Vision-Language Models (VLMs) have been trained under two predominant paradigms. Generative training has enabled Multimodal Large Language Models (MLLMs) to tackle various complex tasks, yet issues such as hallucinations and weak object discrimination persist. Discriminative training, exemplified by models like CLIP, excels in zero-shot image-text classification and retrieval, yet struggles with complex scenarios requiring fine-grained semantic differentiation. This paper addresses these challenges by proposing a unified approach that integrates the strengths of both paradigms. Considering interleaved image-text sequences as the general format of input samples, we introduce a structure-induced training strategy that imposes semantic relationships between input samples and the MLLM's hidden state. This approach enhances the MLLM's ability to capture global semantics and distinguish fine-grained semantics. By leveraging dynamic sequence alignment within the Dynamic Time Warping framework and integrating a novel kernel for fine-grained semantic differentiation, our method effectively balances generative and discriminative tasks. Extensive experiments demonstrate the effectiveness of our approach, achieving state-of-the-art results in multiple generative tasks, especially those requiring cognitive and discrimination abilities. Additionally, our method surpasses discriminative benchmarks in interleaved and fine-grained retrieval tasks. By employing a retrieval-augmented generation strategy, our approach further enhances performance in some generative tasks within one model, offering a promising direction for future research in vision-language modeling.

Paperid: 2106, https://arxiv.org/pdf/2410.06967.pdf

Abstract:
Large vision-language models (LVLMs) have been rapidly developed and widely used in various fields, but the (potential) stereotypical bias in the model is largely unexplored. In this study, we present a pioneering measurement framework, $\texttt{ModSCAN}$, to $\underline{SCAN}$ the stereotypical bias within LVLMs from both vision and language $\underline{Mod}$alities. $\texttt{ModSCAN}$ examines stereotypical biases with respect to two typical stereotypical attributes (gender and race) across three kinds of scenarios: occupations, descriptors, and persona traits. Our findings suggest that 1) the currently popular LVLMs show significant stereotype biases, with CogVLM emerging as the most biased model; 2) these stereotypical biases may stem from the inherent biases in the training dataset and pre-trained models; 3) the utilization of specific prompt prefixes (from both vision and language modalities) performs well in reducing stereotypical biases. We believe our work can serve as the foundation for understanding and addressing stereotypical bias in LVLMs.

Paperid: 2107, https://arxiv.org/pdf/2512.21220.pdf

Abstract:
Embodied agents powered by vision-language models (VLMs) are increasingly capable of executing complex real-world tasks, yet they remain vulnerable to hazardous instructions that may trigger unsafe behaviors. Runtime safety guardrails, which intercept hazardous actions during task execution, offer a promising solution due to their flexibility. However, existing defenses often rely on static rule filters or prompt-level control, which struggle to address implicit risks arising in dynamic, temporally dependent, and context-rich environments. To address this, we propose RoboSafe, a hybrid reasoning runtime safeguard for embodied agents through executable predicate-based safety logic. RoboSafe integrates two complementary reasoning processes on a Hybrid Long-Short Safety Memory. We first propose a Backward Reflective Reasoning module that continuously revisits recent trajectories in short-term memory to infer temporal safety predicates and proactively triggers replanning when violations are detected. We then propose a Forward Predictive Reasoning module that anticipates upcoming risks by generating context-aware safety predicates from the long-term safety memory and the agent's multimodal observations. Together, these components form an adaptive, verifiable safety logic that is both interpretable and executable as code. Extensive experiments across multiple agents demonstrate that RoboSafe substantially reduces hazardous actions (-36.8% risk occurrence) compared with leading baselines, while maintaining near-original task performance. Real-world evaluations on physical robotic arms further confirm its practicality. Code will be released upon acceptance.

Paperid: 2108, https://arxiv.org/pdf/2510.25668.pdf

Abstract:
Vision-language models (VLMs) excel at interpreting text-rich images but struggle with long, visually complex documents that demand analysis and integration of information spread across multiple pages. Existing approaches typically rely on fixed reasoning templates or rigid pipelines, which force VLMs into a passive role and hinder both efficiency and generalization. We present Active Long-DocumEnt Navigation (ALDEN), a multi-turn reinforcement learning framework that fine-tunes VLMs as interactive agents capable of actively navigating long, visually rich documents. ALDEN introduces a novel fetch action that directly accesses the page by index, complementing the classic search action and better exploiting document structure. For dense process supervision and efficient training, we propose a rule-based cross-level reward that provides both turn- and token-level signals. To address the empirically observed training instability caused by numerous visual tokens from long documents, we further propose a visual-semantic anchoring mechanism that applies a dual-path KL-divergence constraint to stabilize visual and textual representations separately during training. Trained on a corpus constructed from three open-source datasets, ALDEN achieves state-of-the-art performance on five long-document benchmarks. Overall, ALDEN marks a step beyond passive document reading toward agents that autonomously navigate and reason across long, visually rich documents, offering a robust path to more accurate and efficient long-document understanding.

Paperid: 2109, https://arxiv.org/pdf/2510.10073.pdf

Abstract:
Large vision-language model (LVLM)-based web agents are emerging as powerful tools for automating complex online tasks. However, when deployed in real-world environments, they face serious security risks, motivating the design of security evaluation benchmarks. Existing benchmarks provide only partial coverage, typically restricted to narrow scenarios such as user-level prompt manipulation, and thus fail to capture the broad range of agent vulnerabilities. To address this gap, we present \tool{}, the first holistic benchmark for evaluating the security of LVLM-based web agents. \tool{} first introduces a unified evaluation suite comprising six simulated but realistic web environments (\eg, e-commerce platforms, community forums) and includes 2,970 high-quality trajectories spanning diverse tasks and attack settings. The suite defines a structured taxonomy of six attack vectors spanning both user-level and environment-level manipulations. In addition, we introduce a multi-layered evaluation protocol that analyzes agent failures across three critical dimensions: internal reasoning, behavioral trajectory, and task outcome, facilitating a fine-grained risk analysis that goes far beyond simple success metrics. Using this benchmark, we conduct large-scale experiments on 9 representative LVLMs, which fall into three categories: general-purpose, agent-specialized, and GUI-grounded. Our results show that all tested agents are consistently vulnerable to subtle adversarial manipulations and reveal critical trade-offs between model specialization and security. By providing (1) a comprehensive benchmark suite with diverse environments and a multi-layered evaluation pipeline, and (2) empirical insights into the security challenges of modern LVLM-based web agents, \tool{} establishes a foundation for advancing trustworthy web agent deployment.

Paperid: 2110, https://arxiv.org/pdf/2510.08003.pdf

Abstract:
Composed Image Retrieval (CIR), which aims to find a target image from a reference image and a modification text, presents the core challenge of performing unified reasoning across visual and semantic modalities. While current approaches based on Vision-Language Models (VLMs, e.g., CLIP) and more recent Multimodal Large Language Models (MLLMs, e.g., Qwen-VL) have shown progress, they predominantly function as ``black boxes." This inherent opacity not only prevents users from understanding the retrieval rationale but also restricts the models' ability to follow complex, fine-grained instructions. To overcome these limitations, we introduce CIR-CoT, the first end-to-end retrieval-oriented MLLM designed to integrate explicit Chain-of-Thought (CoT) reasoning. By compelling the model to first generate an interpretable reasoning chain, CIR-CoT enhances its ability to capture crucial cross-modal interactions, leading to more accurate retrieval while making its decision process transparent. Since existing datasets like FashionIQ and CIRR lack the necessary reasoning data, a key contribution of our work is the creation of structured CoT annotations using a three-stage process involving a caption, reasoning, and conclusion. Our model is then fine-tuned to produce this structured output before encoding its final retrieval intent into a dedicated embedding. Comprehensive experiments show that CIR-CoT achieves highly competitive performance on in-domain datasets (FashionIQ, CIRR) and demonstrates remarkable generalization on the out-of-domain CIRCO dataset, establishing a new path toward more effective and trustworthy retrieval systems.

Paperid: 2111, https://arxiv.org/pdf/2509.06996.pdf

Abstract:
Writing is a universal cultural technology that reuses vision for symbolic communication. Humans display striking resilience: we readily recognize words even when characters are fragmented, fused, or partially occluded. This paper investigates whether advanced vision language models (VLMs) share this resilience. We construct two psychophysics inspired benchmarks across distinct writing systems, Chinese logographs and English alphabetic words, by splicing, recombining, and overlaying glyphs to yield ''visible but unreadable'' stimuli for models while remaining legible to humans. Despite strong performance on clean text, contemporary VLMs show a severe drop under these perturbations, frequently producing unrelated or incoherent outputs. The pattern suggests a structural limitation: models heavily leverage generic visual invariances but under rely on compositional priors needed for robust literacy. We release stimuli generation code, prompts, and evaluation protocols to facilitate transparent replication and follow up work. Our findings motivate architectures and training strategies that encode symbol segmentation, composition, and binding across scripts, and they delineate concrete challenges for deploying multimodal systems in education, accessibility, cultural heritage, and security.

Paperid: 2112, https://arxiv.org/pdf/2508.06534.pdf

Abstract:
Evaluating and ensuring the adversarial robustness of autonomous driving (AD) systems is a critical and unresolved challenge. This paper introduces MetAdv, a novel adversarial testing platform that enables realistic, dynamic, and interactive evaluation by tightly integrating virtual simulation with physical vehicle feedback. At its core, MetAdv establishes a hybrid virtual-physical sandbox, within which we design a three-layer closed-loop testing environment with dynamic adversarial test evolution. This architecture facilitates end-to-end adversarial evaluation, ranging from high-level unified adversarial generation, through mid-level simulation-based interaction, to low-level execution on physical vehicles. Additionally, MetAdv supports a broad spectrum of AD tasks, algorithmic paradigms (e.g., modular deep learning pipelines, end-to-end learning, vision-language models). It supports flexible 3D vehicle modeling and seamless transitions between simulated and physical environments, with built-in compatibility for commercial platforms such as Apollo and Tesla. A key feature of MetAdv is its human-in-the-loop capability: besides flexible environmental configuration for more customized evaluation, it enables real-time capture of physiological signals and behavioral feedback from drivers, offering new insights into human-machine trust under adversarial conditions. We believe MetAdv can offer a scalable and unified framework for adversarial assessment, paving the way for safer AD.

Paperid: 2113, https://arxiv.org/pdf/2508.02028.pdf

Abstract:
Vision-Language Models (VLMs) have recently emerged as a promising paradigm in autonomous driving (AD). However, current performance evaluation protocols for VLM-based AD systems (ADVLMs) are predominantly confined to open-loop settings with static inputs, neglecting the more realistic and informative closed-loop setting that captures interactive behavior, feedback resilience, and real-world safety. To address this, we introduce Bench2ADVLM, a unified hierarchical closed-loop evaluation framework for real-time, interactive assessment of ADVLMs across both simulation and physical platforms. Inspired by dual-process theories of cognition, we first adapt diverse ADVLMs to simulation environments via a dual-system adaptation architecture. In this design, heterogeneous high-level driving commands generated by target ADVLMs (fast system) are interpreted by a general-purpose VLM (slow system) into standardized mid-level control actions suitable for execution in simulation. To bridge the gap between simulation and reality, we design a physical control abstraction layer that translates these mid-level actions into low-level actuation signals, enabling, for the first time, closed-loop testing of ADVLMs on physical vehicles. To enable more comprehensive evaluation, Bench2ADVLM introduces a self-reflective scenario generation module that automatically explores model behavior and uncovers potential failure modes for safety-critical scenario generation. Overall, Bench2ADVLM establishes a hierarchical evaluation pipeline that seamlessly integrates high-level abstract reasoning, mid-level simulation actions, and low-level real-world execution. Experiments on diverse scenarios across multiple state-of-the-art ADVLMs and physical platforms validate the diagnostic strength of our framework, revealing that existing ADVLMs still exhibit limited performance under closed-loop conditions.

Paperid: 2114, https://arxiv.org/pdf/2506.16218.pdf

Abstract:
Federated prompt learning (FPL) for vision-language models is a powerful approach to collaboratively adapt models across distributed clients while preserving data privacy. However, existing FPL approaches suffer from a trade-off between performance and robustness, particularly in out-of-distribution (OOD) shifts, limiting their reliability in real-world scenarios. The inherent in-distribution (ID) data heterogeneity among different clients makes it more challenging to maintain this trade-off. To fill this gap, we introduce a Federated OOD-aware Context Optimization (FOCoOp) framework, which captures diverse distributions among clients using ID global prompts, local prompts, and OOD prompts. Specifically, FOCoOp leverages three sets of prompts to create both class-level and distribution-level separations, which adapt to OOD shifts through bi-level distributionally robust optimization. Additionally, FOCoOp improves the discrimination consistency among clients, i.e., calibrating global prompts, seemingly OOD prompts, and OOD prompts by semi-unbalanced optimal transport. The extensive experiments on real-world datasets demonstrate that FOCoOp effectively captures decentralized heterogeneous distributions and enhances robustness of different OOD shifts. The project is available at GitHub.

Paperid: 2115, https://arxiv.org/pdf/2506.14697.pdf

Abstract:
The rapid advancement of vision-language models (VLMs) and their integration into embodied agents have unlocked powerful capabilities for decision-making. However, as these systems are increasingly deployed in real-world environments, they face mounting safety concerns, particularly when responding to hazardous instructions. In this work, we propose AGENTSAFE, the first comprehensive benchmark for evaluating the safety of embodied VLM agents under hazardous instructions. AGENTSAFE simulates realistic agent-environment interactions within a simulation sandbox and incorporates a novel adapter module that bridges the gap between high-level VLM outputs and low-level embodied controls. Specifically, it maps recognized visual entities to manipulable objects and translates abstract planning into executable atomic actions in the environment. Building on this, we construct a risk-aware instruction dataset inspired by Asimovs Three Laws of Robotics, including base risky instructions and mutated jailbroken instructions. The benchmark includes 45 adversarial scenarios, 1,350 hazardous tasks, and 8,100 hazardous instructions, enabling systematic testing under adversarial conditions ranging from perception, planning, and action execution stages.

Paperid: 2116, https://arxiv.org/pdf/2506.01344.pdf

Abstract:
Flowcharts are a critical tool for visualizing decision-making processes. However, their non-linear structure and complex visual-textual relationships make it challenging to interpret them using LLMs, as vision-language models frequently hallucinate nonexistent connections and decision paths when analyzing these diagrams. This leads to compromised reliability for automated flowchart processing in critical domains such as logistics, health, and engineering. We introduce the task of Fine-grained Flowchart Attribution, which traces specific components grounding a flowchart referring LLM response. Flowchart Attribution ensures the verifiability of LLM predictions and improves explainability by linking generated responses to the flowchart's structure. We propose FlowPathAgent, a neurosymbolic agent that performs fine-grained post hoc attribution through graph-based reasoning. It first segments the flowchart, then converts it into a structured symbolic graph, and then employs an agentic approach to dynamically interact with the graph, to generate attribution paths. Additionally, we present FlowExplainBench, a novel benchmark for evaluating flowchart attributions across diverse styles, domains, and question types. Experimental results show that FlowPathAgent mitigates visual hallucinations in LLM answers over flowchart QA, outperforming strong baselines by 10-14% on our proposed FlowExplainBench dataset.

Paperid: 2117, https://arxiv.org/pdf/2501.13563.pdf

Abstract:
Vision-language models (VLMs) have significantly advanced autonomous driving (AD) by enhancing reasoning capabilities; however, these models remain highly susceptible to adversarial attacks. While existing research has explored white-box attacks to some extent, the more practical and challenging black-box scenarios remain largely underexplored due to their inherent difficulty. In this paper, we take the first step toward designing black-box adversarial attacks specifically targeting VLMs in AD. We identify two key challenges for achieving effective black-box attacks in this context: the effectiveness across driving reasoning chains in AD systems and the dynamic nature of driving scenarios. To address this, we propose Cascading Adversarial Disruption (CAD). It first introduces Decision Chain Disruption, which targets low-level reasoning breakdown by generating and injecting deceptive semantics, ensuring the perturbations remain effective across the entire decision-making chain. Building on this, we present Risky Scene Induction, which addresses dynamic adaptation by leveraging a surrogate VLM to understand and construct high-level risky scenarios that are likely to result in critical errors in the current driving contexts. Extensive experiments conducted on multiple AD VLMs and benchmarks demonstrate that CAD achieves state-of-the-art attack effectiveness, significantly outperforming existing methods (+13.43% on average). Moreover, we validate its practical applicability through real-world attacks on AD vehicles powered by VLMs, where the route completion rate drops by 61.11% and the vehicle crashes directly into the obstacle vehicle with adversarial patches. Finally, we release CADA dataset, comprising 18,808 adversarial visual-question-answer pairs, to facilitate further evaluation and research in this critical domain. Our codes and dataset will be available after paper's acceptance.

Paperid: 2118, https://arxiv.org/pdf/2411.18275.pdf

Abstract:
Vision-language models (VLMs) have significantly advanced autonomous driving (AD) by enhancing reasoning capabilities. However, these models remain highly vulnerable to adversarial attacks. While existing research has primarily focused on general VLM attacks, the development of attacks tailored to the safety-critical AD context has been largely overlooked. In this paper, we take the first step toward designing adversarial attacks specifically targeting VLMs in AD, exposing the substantial risks these attacks pose within this critical domain. We identify two unique challenges for effective adversarial attacks on AD VLMs: the variability of textual instructions and the time-series nature of visual scenarios. To this end, we propose ADvLM, the first visual adversarial attack framework specifically designed for VLMs in AD. Our framework introduces Semantic-Invariant Induction, which uses a large language model to create a diverse prompt library of textual instructions with consistent semantic content, guided by semantic entropy. Building on this, we introduce Scenario-Associated Enhancement, an approach where attention mechanisms select key frames and perspectives within driving scenarios to optimize adversarial perturbations that generalize across the entire scenario. Extensive experiments on several AD VLMs over multiple benchmarks show that ADvLM achieves state-of-the-art attack effectiveness. Moreover, real-world attack studies further validate its applicability and potential in practice.

Paperid: 2119, https://arxiv.org/pdf/2411.09293.pdf

Abstract:
Existing face super-resolution (FSR) methods have made significant advancements, but they primarily super-resolve face with limited visual information, original pixel-wise space in particular, commonly overlooking the pluralistic clues, like the higher-order depth and semantics, as well as non-visual inputs (text caption and description). Consequently, these methods struggle to produce a unified and meaningful representation from the input face. We suppose that introducing the language-vision pluralistic representation into unexplored potential embedding space could enhance FSR by encoding and exploiting the complementarity across language-vision prior. This motivates us to propose a new framework called LLV-FSR, which marries the power of large vision-language model and higher-order visual prior with the challenging task of FSR. Specifically, besides directly absorbing knowledge from original input, we introduce the pre-trained vision-language model to generate pluralistic priors, involving the image caption, descriptions, face semantic mask and depths. These priors are then employed to guide the more critical feature representation, facilitating realistic and high-quality face super-resolution. Experimental results demonstrate that our proposed framework significantly improves both the reconstruction quality and perceptual quality, surpassing the SOTA by 0.43dB in terms of PSNR on the MMCelebA-HQ dataset.

Paperid: 2120, https://arxiv.org/pdf/2410.15461.pdf

Abstract:
Video generation models have made significant progress in simulating future states, showcasing their potential as world simulators in embodied scenarios. However, existing models often lack robust understanding, limiting their ability to perform multi-step predictions or handle Out-of-Distribution (OOD) scenarios. To address this challenge, we propose the Reflection of Generation (RoG), a set of intermediate reasoning strategies designed to enhance video prediction. It leverages the complementary strengths of pre-trained vision-language and video generation models, enabling them to function as a world model in embodied scenarios. To support RoG, we introduce Embodied Video Anticipation Benchmark(EVA-Bench), a comprehensive benchmark that evaluates embodied world models across diverse tasks and scenarios, utilizing both in-domain and OOD datasets. Building on this foundation, we devise a world model, Embodied Video Anticipator (EVA), that follows a multistage training paradigm to generate high-fidelity video frames and apply an autoregressive strategy to enable adaptive generalization for longer video sequences. Extensive experiments demonstrate the efficacy of EVA in various downstream tasks like video generation and robotics, thereby paving the way for large-scale pre-trained models in real-world video prediction applications. The video demos are available at \hyperlink{https://sites.google.com/view/icml-eva}{https://sites.google.com/view/icml-eva}.

Paperid: 2121, https://arxiv.org/pdf/2410.13859.pdf

Abstract:
Despite the significant progress in multimodal large language models (MLLMs), their high computational cost remains a barrier to real-world deployment. Inspired by the mixture of depths (MoDs) in natural language processing, we aim to address this limitation from the perspective of ``activated tokens''. Our key insight is that if most tokens are redundant for the layer computation, then can be skipped directly via the MoD layer. However, directly converting the dense layers of MLLMs to MoD layers leads to substantial performance degradation. To address this issue, we propose an innovative MoD adaptation strategy for existing MLLMs called $Î³$-MoD. In $Î³$-MoD, a novel metric is proposed to guide the deployment of MoDs in the MLLM, namely rank of attention maps (ARank). Through ARank, we can effectively identify which layer is redundant and should be replaced with the MoD layer. Based on ARank, we further propose two novel designs to maximize the computational sparsity of MLLM while maintaining its performance, namely shared vision-language router and masked routing learning. With these designs, more than 90% dense layers of the MLLM can be effectively converted to the MoD ones. To validate our method, we apply it to three popular MLLMs, and conduct extensive experiments on 9 benchmark datasets. Experimental results not only validate the significant efficiency benefit of $Î³$-MoD to existing MLLMs but also confirm its generalization ability on various MLLMs. For example, with a minor performance drop, i.e., -1.5%, $Î³$-MoD can reduce the training and inference time of LLaVA-HR by 31.0% and 53.2%, respectively.

Paperid: 2122, https://arxiv.org/pdf/2511.21984.pdf

Abstract:
Text-prompted foundation models for medical image segmentation offer an intuitive way to delineate anatomical structures from natural language queries, but their predictions often lack spatial precision and degrade under domain shift. In contrast, visual-prompted models achieve strong segmentation performance across diverse modalities by leveraging spatial cues of precise bounding-box (bbox) prompts to guide the segmentation of target lesions. However, it is costly and challenging to obtain the precise visual prompts in clinical practice. We propose PPBoost (Progressive Prompt-Boosting), a framework that bridges these limitations by transforming weak text-derived signals into strong, spatially grounded visual prompts, operating under a strict zero-shot regime with no image- or pixel-level segmentation labels. PPBoost first uses a vision-language model to produce initial pseudo-bboxes conditioned on the textual object descriptions and applies an uncertainty-aware criterion to filter unreliable predictions. The retained image-bboxes pairs are then leveraged to train a pseudo-labeled detector, producing the high-quality bboxes for the query images. During inference, PPBoost further refines the generated bboxes by appropriately expanding them to tightly cover the target anatomical structures. The enhanced spatially-grounding bbox prompts guide existing segmentation models to generate final dense masks, effectively amplifying weak text cues into strong spatial guidance. Across three datasets spanning diverse modalities and anatomies, PPBoost consistently improves Dice and Normalized Surface Distance over text- and visual-prompted baselines and, notably, surpasses few-shot segmentation models without using labeled data. PPBoost can generalize to multiple typical visual segmentation model backbones.

Paperid: 2123, https://arxiv.org/pdf/2510.09822.pdf

Abstract:
Real-world vision-language applications demand varying levels of perceptual granularity. However, most existing visual large language models (VLLMs), such as LLaVA, pre-assume a fixed resolution for downstream tasks, which leads to subpar performance. To address this problem, we first conduct a comprehensive and pioneering investigation into the resolution preferences of different vision-language tasks, revealing a correlation between resolution preferences with image complexity, and uncertainty variance of the VLLM at different image input resolutions. Building on this insight, we propose an empirical formula to determine the optimal resolution for a given vision-language task, combining these two factors. Second, based on rigorous experiments, we propose a novel parameter-efficient fine-tuning technique to extend the visual input resolution of pre-trained VLLMs to the identified optimal resolution. Extensive experiments on various vision-language tasks validate the effectiveness of our method.

Paperid: 2124, https://arxiv.org/pdf/2410.13121.pdf

Abstract:
Vision-Language Models (VLMs) often generate plausible but incorrect responses to visual queries. However, reliably quantifying the effect of such hallucinations in free-form responses to open-ended queries is challenging as it requires visually verifying each claim within the response. We propose Programmatic VLM Evaluation (PROVE), a new benchmarking paradigm for evaluating VLM responses to open-ended queries. To construct PROVE, we provide a large language model (LLM) with a high-fidelity scene-graph representation constructed from a hyper-detailed image caption, and prompt it to generate diverse question-answer (QA) pairs, as well as programs that can be executed over the scene graph object to verify each QA pair. We thus construct a benchmark of 10.5k challenging but visually grounded QA pairs. Next, to evaluate free-form model responses to queries in PROVE, we propose a programmatic evaluation strategy that measures both the helpfulness and truthfulness of a response within a unified scene graph-based framework. We benchmark the helpfulness-truthfulness trade-offs of a range of VLMs on PROVE, finding that very few are in-fact able to achieve a good balance between the two. Project page: \url{https://prove-explorer.netlify.app/}.

Paperid: 2125, https://arxiv.org/pdf/2601.08871.pdf

Abstract:
Balancing dialogue, music, and sound effects with accompanying video is crucial for immersive storytelling, yet current audio mixing workflows remain largely manual and labor-intensive. While recent advancements have introduced the visually guided acoustic highlighting task, which implicitly rebalances audio sources using multimodal guidance, it remains unclear which visual aspects are most effective as conditioning signals.We address this gap through a systematic study of whether deep video understanding improves audio remixing. Using textual descriptions as a proxy for visual analysis, we prompt large vision-language models to extract six types of visual-semantic aspects, including object and character appearance, emotion, camera focus, tone, scene background, and inferred sound-related cues. Through extensive experiments, camera focus, tone, and scene background consistently yield the largest improvements in perceptual mix quality over state-of-the-art baselines. Our findings (i) identify which visual-semantic cues most strongly support coherent and visually aligned audio remixing, and (ii) outline a practical path toward automating cinema-grade sound design using lightweight guidance derived from large vision-language models.

Paperid: 2126, https://arxiv.org/pdf/2511.16857.pdf

Abstract:
Vision Language Models (VLMs) have achieved impressive performance on spatial reasoning benchmarks, yet these evaluations mask critical weaknesses in understanding object interactions. Current benchmarks test high level relationships ('left of,' 'behind', etc.) but ignore fine-grained spatial understanding needed for real world applications: precise 3D localization, physical compatibility between objects, object affordances and multi step spatial planning. In this work, we present BOP-ASK, a novel large scale dataset for object interaction reasoning for both training and benchmarking. Our data generation pipeline leverages 6D object poses from the Benchmark for Object Pose Estimation (BOP) datasets from which we derive fine grained annotations such as grasp poses, referred object poses, path planning trajectories, relative spatial and depth relationships, and object-to-object relationships. BOP-ASK comprises over 150k images and 33M question answer pairs spanning six tasks (four novel), providing a rich resource for training and evaluating VLMs. We evaluate proprietary and open sourced VLMs, and conduct human evaluations on BOP-ASK-core, a contributed test benchmark. We also release BOP-ASK-lab, an out-of-distribution benchmark with images not sourced from BOP, enabling testing of generalization. Our experiments demonstrate that models trained on BOP-ASK outperform baselines and exhibit emergent capabilities such as precise object and grasp pose estimation, trajectory planning, and fine-grained object-centric spatial reasoning in cluttered environments. We will publicly release our datasets and dataset generation pipeline.

Paperid: 2127, https://arxiv.org/pdf/2511.15613.pdf

Abstract:
Test-time thinking (that is, generating explicit intermediate reasoning chains) is known to boost performance in large language models and has recently shown strong gains for large vision language models (LVLMs). However, despite these promising results, there is still no systematic analysis of how thinking actually affects visual reasoning. We provide the first such analysis with a large scale, controlled comparison of thinking for LVLMs, evaluating ten variants from the InternVL3.5 and Qwen3-VL families on MMMU-val under generous token budgets and multi pass decoding. We show that more thinking is not always better; long chains often yield long wrong trajectories that ignore the image and underperform the same models run in standard instruct mode. A deeper analysis reveals that certain short lookback phrases, which explicitly refer back to the image, are strongly enriched in successful trajectories and correlate with better visual grounding. Building on this insight, we propose uncertainty guided lookback, a training free decoding strategy that combines an uncertainty signal with adaptive lookback prompts and breadth search. Our method improves overall MMMU performance, delivers the largest gains in categories where standard thinking is weak, and outperforms several strong decoding baselines, setting a new state of the art under fixed model families and token budgets. We further show that this decoding strategy generalizes, yielding consistent improvements on five additional benchmarks, including two broad multimodal suites and math focused visual reasoning datasets.

Paperid: 2128, https://arxiv.org/pdf/2510.22282.pdf

Abstract:
Harnessing publicly available, large-scale web data, such as street view and satellite imagery, urban socio-economic sensing is of paramount importance for achieving global sustainable development goals. With the emergence of Large Vision-Language Models (LVLMs), new opportunities have arisen to solve this task by treating it as a multi-modal perception and understanding problem. However, recent studies reveal that LVLMs still struggle with accurate and interpretable socio-economic predictions from visual data. To address these limitations and maximize the potential of LVLMs, we introduce \textbf{CityRiSE}, a novel framework for \textbf{R}eason\textbf{i}ng urban \textbf{S}ocio-\textbf{E}conomic status in LVLMs through pure reinforcement learning (RL). With carefully curated multi-modal data and verifiable reward design, our approach guides the LVLM to focus on semantically meaningful visual cues, enabling structured and goal-oriented reasoning for generalist socio-economic status prediction. Experiments demonstrate that CityRiSE with emergent reasoning process significantly outperforms existing baselines, improving both prediction accuracy and generalization across diverse urban contexts, particularly for prediction on unseen cities and unseen indicators. This work highlights the promise of combining RL and LVLMs for interpretable and generalist urban socio-economic sensing.

Paperid: 2129, https://arxiv.org/pdf/2510.13219.pdf

Abstract:
In computer vision, Visual Prompting (VP) and Visual Prompt Tuning (VPT) have recently emerged as lightweight and effective alternatives to full fine-tuning for adapting large-scale vision models within the ``pretrain-then-finetune'' paradigm. However, despite rapid progress, their conceptual boundaries remain blurred, as VP and VPT are frequently used interchangeably in current research, reflecting a lack of systematic distinction between these techniques and their respective applications. In this survey, we revisit the designs of VP and VPT from first principles, and conceptualize them within a unified framework termed Prompt-based Adaptation (PA). We provide a taxonomy that categorizes existing methods into learnable, generative, and non-learnable prompts, and further organizes them by injection granularity -- pixel-level and token-level. Beyond the core methodologies, we examine PA's integrations across diverse domains, including medical imaging, 3D point clouds, and vision-language tasks, as well as its role in test-time adaptation and trustworthy AI. We also summarize current benchmarks and identify key challenges and future directions. To the best of our knowledge, we are the first comprehensive survey dedicated to PA's methodologies and applications in light of their distinct characteristics. Our survey aims to provide a clear roadmap for researchers and practitioners in all area to understand and explore the evolving landscape of PA-related research.

Paperid: 2130, https://arxiv.org/pdf/2507.17353.pdf

Abstract:
Accurate road damage detection is crucial for timely infrastructure maintenance and public safety, but existing vision-only datasets and models lack the rich contextual understanding that textual information can provide. To address this limitation, we introduce RoadBench, the first multimodal benchmark for comprehensive road damage understanding. This dataset pairs high resolution images of road damages with detailed textual descriptions, providing a richer context for model training. We also present RoadCLIP, a novel vision language model that builds upon CLIP by integrating domain specific enhancements. It includes a disease aware positional encoding that captures spatial patterns of road defects and a mechanism for injecting road-condition priors to refine the model's understanding of road damages. We further employ a GPT driven data generation pipeline to expand the image to text pairs in RoadBench, greatly increasing data diversity without exhaustive manual annotation. Experiments demonstrate that RoadCLIP achieves state of the art performance on road damage recognition tasks, significantly outperforming existing vision-only models by 19.2%. These results highlight the advantages of integrating visual and textual information for enhanced road condition analysis, setting new benchmarks for the field and paving the way for more effective infrastructure monitoring through multimodal learning.

Paperid: 2131, https://arxiv.org/pdf/2507.16518.pdf

Abstract:
Recent advances in multimodal large language models (MLLMs) have shown impressive reasoning capabilities. However, further enhancing existing MLLMs necessitates high-quality vision-language datasets with carefully curated task complexities, which are both costly and challenging to scale. Although recent self-improving models that iteratively refine themselves offer a feasible solution, they still suffer from two core challenges: (i) most existing methods augment visual or textual data separately, resulting in discrepancies in data complexity (e.g., over-simplified diagrams paired with redundant textual descriptions); and (ii) the evolution of data and models is also separated, leading to scenarios where models are exposed to tasks with mismatched difficulty levels. To address these issues, we propose C2-Evo, an automatic, closed-loop self-improving framework that jointly evolves both training data and model capabilities. Specifically, given a base dataset and a base model, C2-Evo enhances them by a cross-modal data evolution loop and a data-model evolution loop. The former loop expands the base dataset by generating complex multimodal problems that combine structured textual sub-problems with iteratively specified geometric diagrams, while the latter loop adaptively selects the generated problems based on the performance of the base model, to conduct supervised fine-tuning and reinforcement learning alternately. Consequently, our method continuously refines its model and training data, and consistently obtains considerable performance gains across multiple mathematical reasoning benchmarks. Our code, models, and datasets will be released.

Paperid: 2132, https://arxiv.org/pdf/2506.06535.pdf

Abstract:
Robotic manipulation of unseen objects via natural language commands remains challenging. Language driven robotic grasping (LDRG) predicts stable grasp poses from natural language queries and RGB-D images. We propose MapleGrasp, a novel framework that leverages mask-guided feature pooling for efficient vision-language driven grasping. Our two-stage training first predicts segmentation masks from CLIP-based vision-language features. The second stage pools features within these masks to generate pixel-level grasp predictions, improving efficiency, and reducing computation. Incorporating mask pooling results in a 7% improvement over prior approaches on the OCID-VLG benchmark. Furthermore, we introduce RefGraspNet, an open-source dataset eight times larger than existing alternatives, significantly enhancing model generalization for open-vocabulary grasping. MapleGrasp scores a strong grasping accuracy of 89\% when compared with competing methods in the RefGraspNet benchmark. Our method achieves comparable performance to larger Vision-Language-Action models on the LIBERO benchmark, and shows significantly better generalization to unseen tasks. Real-world experiments on a Franka arm demonstrate 73% success rate with unseen objects, surpassing competitive baselines by 11%. Code is provided in our github repository.

Paperid: 2133, https://arxiv.org/pdf/2506.05318.pdf

Abstract:
Remarkable progress in 2D Vision-Language Models (VLMs) has spurred interest in extending them to 3D settings for tasks like 3D Question Answering, Dense Captioning, and Visual Grounding. Unlike 2D VLMs that typically process images through an image encoder, 3D scenes, with their intricate spatial structures, allow for diverse model architectures. Based on their encoder design, this paper categorizes recent 3D VLMs into 3D object-centric, 2D image-based, and 3D scene-centric approaches. Despite the architectural similarity of 3D scene-centric VLMs to their 2D counterparts, they have exhibited comparatively lower performance compared with the latest 3D object-centric and 2D image-based approaches. To understand this gap, we conduct an in-depth analysis, revealing that 3D scene-centric VLMs show limited reliance on the 3D scene encoder, and the pre-train stage appears less effective than in 2D VLMs. Furthermore, we observe that data scaling benefits are less pronounced on larger datasets. Our investigation suggests that while these models possess cross-modal alignment capabilities, they tend to over-rely on linguistic cues and overfit to frequent answer distributions, thereby diminishing the effective utilization of the 3D encoder. To address these limitations and encourage genuine 3D scene understanding, we introduce a novel 3D Relevance Discrimination QA dataset designed to disrupt shortcut learning and improve 3D understanding. Our findings highlight the need for advanced evaluation and improved strategies for better 3D understanding in 3D VLMs.

Paperid: 2134, https://arxiv.org/pdf/2505.15373.pdf

Abstract:
Mapping and understanding complex 3D environments is fundamental to how autonomous systems perceive and interact with the physical world, requiring both precise geometric reconstruction and rich semantic comprehension. While existing 3D semantic mapping systems excel at reconstructing and identifying predefined object instances, they lack the flexibility to efficiently build semantic maps with open-vocabulary during online operation. Although recent vision-language models have enabled open-vocabulary object recognition in 2D images, they haven't yet bridged the gap to 3D spatial understanding. The critical challenge lies in developing a training-free unified system that can simultaneously construct accurate 3D maps while maintaining semantic consistency and supporting natural language interactions in real time. In this paper, we develop a zero-shot framework that seamlessly integrates GPU-accelerated geometric reconstruction with open-vocabulary vision-language models through online instance-level semantic embedding fusion, guided by hierarchical object association with spatial indexing. Our training-free system achieves superior performance through incremental processing and unified geometric-semantic updates, while robustly handling 2D segmentation inconsistencies. The proposed general-purpose 3D scene understanding framework can be used for various tasks including zero-shot 3D instance retrieval, segmentation, and object detection to reason about previously unseen objects and interpret natural language queries. The project page is available at https://razer-3d.github.io.

Paperid: 2135, https://arxiv.org/pdf/2505.05804.pdf

Abstract:
Localized image captioning has made significant progress with models like the Describe Anything Model (DAM), which can generate detailed region-specific descriptions without explicit region-text supervision. However, such capabilities have yet to be widely applied to specialized domains like medical imaging, where diagnostic interpretation relies on subtle regional findings rather than global understanding. To mitigate this gap, we propose MedDAM, the first comprehensive framework leveraging large vision-language models for region-specific captioning in medical images. MedDAM employs medical expert-designed prompts tailored to specific imaging modalities and establishes a robust evaluation benchmark comprising a customized assessment protocol, data pre-processing pipeline, and specialized QA template library. This benchmark evaluates both MedDAM and other adaptable large vision-language models, focusing on clinical factuality through attribute-level verification tasks, thereby circumventing the absence of ground-truth region-caption pairs in medical datasets. Extensive experiments on the VinDr-CXR, LIDC-IDRI, and SkinCon datasets demonstrate MedDAM's superiority over leading peers (including GPT-4o, Claude 3.7 Sonnet, LLaMA-3.2 Vision, Qwen2.5-VL, GPT-4Rol, and OMG-LLaVA) in the task, revealing the importance of region-level semantic alignment in medical image understanding and establishing MedDAM as a promising foundation for clinical vision-language integration.

Paperid: 2136, https://arxiv.org/pdf/2505.05800.pdf

Abstract:
Robotic manipulation in 3D requires learning an $N$ degree-of-freedom joint space trajectory of a robot manipulator. Robots must possess semantic and visual perception abilities to transform real-world mappings of their workspace into the low-level control necessary for object manipulation. Recent work has demonstrated the capabilities of fine-tuning large Vision-Language Models (VLMs) to learn the mapping between RGB images, language instructions, and joint space control. These models typically take as input RGB images of the workspace and language instructions, and are trained on large datasets of teleoperated robot demonstrations. In this work, we explore methods to improve the scene context awareness of a popular recent Vision-Language-Action model by integrating chain-of-thought reasoning, depth perception, and task-oriented region of interest detection. Our experiments in the LIBERO simulation environment show that our proposed model, 3D-CAVLA, improves the success rate across various LIBERO task suites, achieving an average success rate of 98.1$\%$. We also evaluate the zero-shot capabilities of our method, demonstrating that 3D scene awareness leads to robust learning and adaptation for completely unseen tasks. 3D-CAVLA achieves an absolute improvement of 8.8$\%$ on unseen tasks. We will open-source our code and the unseen tasks dataset to promote community-driven research here: https://3d-cavla.github.io

Paperid: 2137, https://arxiv.org/pdf/2504.04633.pdf

Abstract:
Multimodal in-context learning (ICL) equips Large Vision-language Models (LVLMs) with the ability to adapt to new tasks via multiple user-provided demonstrations, without requiring any model parameter updates. However, its effectiveness is constrained by the token-intensive nature of multimodal inputs and the complexity of cross-modal few-shot reasoning, which together hinder LVLMs from extracting useful patterns from demonstrations. To address these challenges, we propose \textbf{M$^2$IV}, a novel representation engineering approach that replaces explicit token-level demonstrations with a set of learnable Multimodal In-context Vectors directly injected into the residual streams of LVLMs. By analyzing the distinct roles of multi-head attention (MHA) and multi-layer perceptrons (MLP) in the ICL process, we design a training strategy that enables M$^2$IV to perform fine-grained semantic distillation and robust cross-modal representation learning. M$^2$IV not only improves performance across diverse tasks and LVLMs but also significantly reduces token overhead, enabling graceful scaling to many-shot scenarios. To further enhance usability, we introduce \textbf{VLibrary}, a repository that stores trained M$^2$IVs for flexible retrieval and injection. With VLibrary, users can steer pre-trained LVLMs in a customized manner that meets diverse requirements. Extensive experiments demonstrate that M$^2$IV consistently outperforms vanilla ICL and prior representation engineering baselines, achieving an average accuracy gain of 3.74\% with substantial improvements in overall efficiency.

Paperid: 2138, https://arxiv.org/pdf/2502.17860.pdf

Abstract:
Recent advancements in multi-modal 3D pre-training methods have shown promising efficacy in learning joint representations of text, images, and point clouds. However, adopting point clouds as 3D representation fails to fully capture the intricacies of the 3D world and exhibits a noticeable gap between the discrete points and the dense 2D pixels of images. To tackle this issue, we propose UniGS, integrating 3D Gaussian Splatting (3DGS) into multi-modal pre-training to enhance the 3D representation. We first rely on the 3DGS representation to model the 3D world as a collection of 3D Gaussians with color and opacity, incorporating all the information of the 3D scene while establishing a strong connection with 2D images. Then, to achieve Language-Image-3D pertaining, UniGS starts with a pre-trained vision-language model to establish a shared visual and textual space through extensive real-world image-text pairs. Subsequently, UniGS employs a 3D encoder to align the optimized 3DGS with the Language-Image representations to learn unified multi-modal representations. To facilitate the extraction of global explicit 3D features by the 3D encoder and achieve better cross-modal alignment, we additionally introduce a novel Gaussian-Aware Guidance module that guides the learning of fine-grained representations of the 3D domain. Through extensive experiments across the Objaverse, ABO, MVImgNet and SUN RGBD datasets with zero-shot classification, text-driven retrieval and open-world understanding tasks, we demonstrate the effectiveness of UniGS in learning a more general and stronger aligned multi-modal representation. Specifically, UniGS achieves leading results across different 3D tasks with remarkable improvements over previous SOTA, Uni3D, including on zero-shot classification (+9.36%), text-driven retrieval (+4.3%) and open-world understanding (+7.92%).

Paperid: 2139, https://arxiv.org/pdf/2501.18624.pdf

Abstract:
Vision-Language Models (VLMs), built on pre-trained vision encoders and large language models (LLMs), have shown exceptional multi-modal understanding and dialog capabilities, positioning them as catalysts for the next technological revolution. However, while most VLM research focuses on enhancing multi-modal interaction, the risks of data misuse and leakage have been largely unexplored. This prompts the need for a comprehensive investigation of such risks in VLMs. In this paper, we conduct the first analysis of misuse and leakage detection in VLMs through the lens of membership inference attack (MIA). In specific, we focus on the instruction tuning data of VLMs, which is more likely to contain sensitive or unauthorized information. To address the limitation of existing MIA methods, we introduce a novel approach that infers membership based on a set of samples and their sensitivity to temperature, a unique parameter in VLMs. Based on this, we propose four membership inference methods, each tailored to different levels of background knowledge, ultimately arriving at the most challenging scenario. Our comprehensive evaluations show that these methods can accurately determine membership status, e.g., achieving an AUC greater than 0.8 targeting a small set consisting of only 5 samples on LLaVA.

Paperid: 2140, https://arxiv.org/pdf/2412.03324.pdf

Abstract:
Vision-language models (VLMs) have shown remarkable success across various multi-modal tasks, yet large VLMs encounter significant efficiency challenges due to processing numerous visual tokens. A promising approach to accelerating large VLM inference is using partial information, such as attention maps from specific layers, to assess token importance and prune less essential tokens. However, our study reveals three key insights: (i) Partial attention information is insufficient for accurately identifying critical visual tokens, resulting in suboptimal performance, especially at low token retention ratios; (ii) Global attention information, such as the attention map aggregated across all layers, more effectively preserves essential tokens and maintains comparable performance under aggressive pruning. However, the attention maps from all layers requires a full inference pass, which increases computational load and is therefore impractical in existing methods; and (iii) The global attention map aggregated from a small VLM closely resembles that of a large VLM, suggesting an efficient alternative. Based on these findings, we introduce a \textbf{training-free} method, \underline{\textbf{S}}mall VLM \underline{\textbf{G}}uidance for accelerating \underline{\textbf{L}}arge VLMs (\textbf{SGL}). Specifically, we employ the attention map aggregated from a small VLM to guide visual token pruning in a large VLM. Additionally, an early exiting mechanism is developed to fully use the small VLM's predictions, dynamically invoking the larger VLM only when necessary, yielding a superior trade-off between accuracy and computation. Extensive evaluations across 11 benchmarks demonstrate the effectiveness and generalizability of SGL, achieving up to 91\% pruning ratio for visual tokens while retaining competitive performance.

Paperid: 2141, https://arxiv.org/pdf/2410.06239.pdf

Abstract:
Enabling robots to autonomously navigate unknown, complex, dynamic environments and perform diverse tasks remains a fundamental challenge in developing robust autonomous physical agents. These agents must effectively perceive their surroundings while leveraging world knowledge for decision-making. Although recent approaches utilize vision-language and large language models for scene understanding and planning, they often rely on offline processing, offboard compute, make simplifying assumptions about the environment and perception, limiting real-world applicability. We present a novel framework for real-time onboard autonomous navigation in unknown environments that change over time by integrating multi-level abstraction in both perception and planning pipelines. Our system fuses data from multiple onboard sensors for localization and mapping and integrates it with open-vocabulary semantics to generate hierarchical scene graphs from continuously updated semantic object map. The LLM-based planner uses these graphs to create multi-step plans that guide low-level controllers in executing navigation tasks specified in natural language. The system's real-time operation enables the LLM to adjust its plans based on updates to the scene graph and task execution status, ensuring continuous adaptation to new situations or when the current plan cannot accomplish the task, a key advantage over static or rule-based systems. We demonstrate our system's efficacy on a quadruped navigating dynamic environments, showcasing its adaptability and robustness in diverse scenarios.

Paperid: 2142, https://arxiv.org/pdf/2409.18042.pdf

Abstract:
GPT-4o, an omni-modal model that enables vocal conversations with diverse emotions and tones, marks a milestone for omni-modal foundation models. However, empowering Large Language Models to perceive and generate images, texts, and speeches end-to-end with publicly available data remains challenging for the open-source community. Existing vision-language models rely on external tools for speech processing, while speech-language models still suffer from limited or totally without vision-understanding capabilities. To address this gap, we propose the EMOVA (EMotionally Omni-present Voice Assistant), to enable Large Language Models with end-to-end speech abilities while maintaining the leading vision-language performance. With a semantic-acoustic disentangled speech tokenizer, we surprisingly notice that omni-modal alignment can further enhance vision-language and speech abilities compared with the bi-modal aligned counterparts. Moreover, a lightweight style module is introduced for the flexible speech style controls including emotions and pitches. For the first time, EMOVA achieves state-of-the-art performance on both the vision-language and speech benchmarks, and meanwhile, supporting omni-modal spoken dialogue with vivid emotions.

Paperid: 2143, https://arxiv.org/pdf/2601.21531.pdf

Abstract:
Visual token compression is widely used to accelerate large vision-language models (LVLMs) by pruning or merging visual tokens, yet its adversarial robustness remains unexplored. We show that existing encoder-based attacks can substantially overestimate the robustness of compressed LVLMs, due to an optimization-inference mismatch: perturbations are optimized on the full-token representation, while inference is performed through a token-compression bottleneck. To address this gap, we propose the Compression-AliGnEd attack (CAGE), which aligns perturbation optimization with compression inference without assuming access to the deployed compression mechanism or its token budget. CAGE combines (i) expected feature disruption, which concentrates distortion on tokens likely to survive across plausible budgets, and (ii) rank distortion alignment, which actively aligns token distortions with rank scores to promote the retention of highly distorted evidence. Across diverse representative plug-and-play compression mechanisms and datasets, our results show that CAGE consistently achieves lower robust accuracy than the baseline. This work highlights that robustness assessments ignoring compression can be overly optimistic, calling for compression-aware security evaluation and defenses for efficient LVLMs.

Paperid: 2144, https://arxiv.org/pdf/2509.20196.pdf

Abstract:
Visual language modeling for automated driving is emerging as a promising research direction with substantial improvements in multimodal reasoning capabilities. Despite its advanced reasoning abilities, VLM-AD remains vulnerable to serious security threats from adversarial attacks, which involve misleading model decisions through carefully crafted perturbations. Existing attacks have obvious challenges: 1) Physical adversarial attacks primarily target vision modules. They are difficult to directly transfer to VLM-AD systems because they typically attack low-level perceptual components. 2) Adversarial attacks against VLM-AD have largely concentrated on the digital level. To address these challenges, we propose the first Universal Camouflage Attack (UCA) framework for VLM-AD. Unlike previous methods that focus on optimizing the logit layer, UCA operates in the feature space to generate physically realizable camouflage textures that exhibit strong generalization across different user commands and model architectures. Motivated by the observed vulnerability of encoder and projection layers in VLM-AD, UCA introduces a feature divergence loss (FDL) that maximizes the representational discrepancy between clean and adversarial images. In addition, UCA incorporates a multi-scale learning strategy and adjusts the sampling ratio to enhance its adaptability to changes in scale and viewpoint diversity in real-world scenarios, thereby improving training stability. Extensive experiments demonstrate that UCA can induce incorrect driving commands across various VLM-AD models and driving scenarios, significantly surpassing existing state-of-the-art attack methods (improving 30\% in 3-P metrics). Furthermore, UCA exhibits strong attack robustness under diverse viewpoints and dynamic conditions, indicating high potential for practical deployment.

Paperid: 2145, https://arxiv.org/pdf/2506.13205.pdf

Abstract:
Mobile agents powered by vision-language models (VLMs) are increasingly adopted for tasks such as UI automation and camera-based assistance. These agents are typically fine-tuned using small-scale, user-collected data, making them susceptible to stealthy training-time threats. This work introduces VIBMA, the first clean-text backdoor attack targeting VLM-based mobile agents. The attack injects malicious behaviors into the model by modifying only the visual input while preserving textual prompts and instructions, achieving stealth through the complete absence of textual anomalies. Once the agent is fine-tuned on this poisoned data, adding a predefined visual pattern (trigger) at inference time activates the attacker-specified behavior (backdoor). Our attack aligns the training gradients of poisoned samples with those of an attacker-specified target instance, effectively embedding backdoor-specific features into the poisoned data. To ensure the robustness and stealthiness of the attack, we design three trigger variants that better resemble real-world scenarios: static patches, dynamic motion patterns, and low-opacity blended content. Extensive experiments on six Android applications and three mobile-compatible VLMs demonstrate that our attack achieves high success rates (ASR up to 94.67%) while preserving clean-task behavior (FSR up to 95.85%). We further conduct ablation studies to understand how key design factors impact attack reliability and stealth. These findings is the first to reveal the security vulnerabilities of mobile agents and their susceptibility to backdoor injection, underscoring the need for robust defenses in mobile agent adaptation pipelines.

Paperid: 2146, https://arxiv.org/pdf/2506.04743.pdf

Abstract:
Vision-Language Models (VLMs) have achieved remarkable performance in image captioning, but recent studies show they are vulnerable to backdoor attacks. Attackers can inject imperceptible perturbations-such as local pixel triggers or global semantic phrases-into the training data, causing the model to generate malicious, attacker-controlled captions for specific inputs. These attacks are hard to detect and defend due to their stealthiness and cross-modal nature. By analyzing attack samples, we identify two key vulnerabilities: (1) abnormal attention concentration on specific image regions, and (2) semantic drift and incoherence in generated captions. To counter this, we propose Semantic Reward Defense (SRD), a reinforcement learning framework that mitigates backdoor behavior without prior knowledge of triggers. SRD uses a Deep Q-Network to learn policies for applying discrete perturbations (e.g., occlusion, color masking) to sensitive image regions, aiming to disrupt the activation of malicious pathways. We design a semantic fidelity score as the reward signal, which jointly evaluates semantic consistency and linguistic fluency of the output, guiding the agent toward generating robust yet faithful captions. Experiments across mainstream VLMs and datasets show SRD reduces attack success rates to 5.6%, while preserving caption quality on clean inputs with less than 10% performance drop. SRD offers a trigger-agnostic, interpretable defense paradigm against stealthy backdoor threats in multimodal generative models.

Paperid: 2147, https://arxiv.org/pdf/2505.15510.pdf

Abstract:
Large Vision-Language Models (LVLMs) have achieved significant success in multimodal tasks, with multimodal chain-of-thought (MCoT) further enhancing performance and interpretability. Recent MCoT methods fall into two categories: (i) Textual-MCoT (T-MCoT), which takes multimodal input and produces textual output; and (ii) Interleaved-MCoT (I-MCoT), which generates interleaved image-text outputs. Despite advances in both approaches, the mechanisms driving these improvements are not fully understood. To fill this gap, we first reveal that MCoT boosts LVLMs by incorporating visual thoughts, which convey image information to the reasoning process regardless of the MCoT format, depending only on clarity and conciseness of expression. Furthermore, to explore visual thoughts systematically, we define four distinct forms of visual thought expressions and analyze them comprehensively. Our findings demonstrate that these forms differ in clarity and conciseness, yielding varying levels of MCoT improvement. Additionally, we explore the internal nature of visual thoughts, finding that visual thoughts serve as intermediaries between the input image and reasoning to deeper transformer layers, enabling more advanced visual information transmission. We hope that the visual thoughts can inspire further breakthroughs for future MCoT research.

Paperid: 2148, https://arxiv.org/pdf/2503.11557.pdf

Abstract:
Visual reasoning is central to human cognition, enabling individuals to interpret and abstractly understand their environment. Although recent Multimodal Large Language Models (MLLMs) have demonstrated impressive performance across language and vision-language tasks, existing benchmarks primarily measure recognition-based skills and inadequately assess true visual reasoning capabilities. To bridge this critical gap, we introduce VERIFY, a benchmark explicitly designed to isolate and rigorously evaluate the visual reasoning capabilities of state-of-the-art MLLMs. VERIFY compels models to reason primarily from visual information, providing minimal textual context to reduce reliance on domain-specific knowledge and linguistic biases. Each problem is accompanied by a human-annotated reasoning path, making it the first to provide in-depth evaluation of model decision-making processes. Additionally, we propose novel metrics that assess visual reasoning fidelity beyond mere accuracy, highlighting critical imbalances in current model reasoning patterns. Our comprehensive benchmarking of leading MLLMs uncovers significant limitations, underscoring the need for a balanced and holistic approach to both perception and reasoning. For more teaser and testing, visit our project page (https://verify-eqh.pages.dev/).

Paperid: 2149, https://arxiv.org/pdf/2411.15421.pdf

Abstract:
Surgical practice involves complex visual interpretation, procedural skills, and advanced medical knowledge, making surgical vision-language pretraining (VLP) particularly challenging due to this complexity and the limited availability of annotated data. To address the gap, we propose OphCLIP, a hierarchical retrieval-augmented vision-language pretraining framework specifically designed for ophthalmic surgical workflow understanding. OphCLIP leverages the OphVL dataset we constructed, a large-scale and comprehensive collection of over 375K hierarchically structured video-text pairs with tens of thousands of different combinations of attributes (surgeries, phases/operations/actions, instruments, medications, as well as more advanced aspects like the causes of eye diseases, surgical objectives, and postoperative recovery recommendations, etc). These hierarchical video-text correspondences enable OphCLIP to learn both fine-grained and long-term visual representations by aligning short video clips with detailed narrative descriptions and full videos with structured titles, capturing intricate surgical details and high-level procedural insights, respectively. Our OphCLIP also designs a retrieval-augmented pretraining framework to leverage the underexplored large-scale silent surgical procedure videos, automatically retrieving semantically relevant content to enhance the representation learning of narrative videos. Evaluation across 11 datasets for phase recognition and multi-instrument identification shows OphCLIP's robust generalization and superior performance.

Paperid: 2150, https://arxiv.org/pdf/2411.14522.pdf

Abstract:
Despite significant advancements in general AI, its effectiveness in the medical domain is limited by the lack of specialized medical knowledge. To address this, we formulate GMAI-VL-5.5M, a multimodal medical dataset created by converting hundreds of specialized medical datasets with various annotations into high-quality image-text pairs. This dataset offers comprehensive task coverage, diverse modalities, and rich image-text data. Building upon this dataset, we develop GMAI-VL, a general medical vision-language model, with a three-stage training strategy that enhances the integration of visual and textual information. This approach significantly improves the model's ability to process multimodal data, supporting accurate diagnoses and clinical decision-making. Experiments show that GMAI-VL achieves state-of-the-art performance across various multimodal medical tasks, including visual question answering and medical image diagnosis.

Paperid: 2151, https://arxiv.org/pdf/2511.12003.pdf

Abstract:
Aiming to identify precise evidence sources from visual documents, visual evidence attribution for visual document retrieval-augmented generation (VD-RAG) ensures reliable and verifiable predictions from vision-language models (VLMs) in multimodal question answering. Most existing methods adopt end-to-end training to facilitate intuitive answer verification. However, they lack fine-grained supervision and progressive traceability throughout the reasoning process. In this paper, we introduce the Chain-of-Evidence (CoE) paradigm for VD-RAG. CoE unifies Chain-of-Thought (CoT) reasoning and visual evidence attribution by grounding reference elements in reasoning steps to specific regions with bounding boxes and page indexes. To enable VLMs to generate such evidence-grounded reasoning, we propose Look As You Think (LAT), a reinforcement learning framework that trains models to produce verifiable reasoning paths with consistent attribution. During training, LAT evaluates the attribution consistency of each evidence region and provides rewards only when the CoE trajectory yields correct answers, encouraging process-level self-verification. Experiments on vanilla Qwen2.5-VL-7B-Instruct with Paper- and Wiki-VISA benchmarks show that LAT consistently improves the vanilla model in both single- and multi-image settings, yielding average gains of 8.23% in soft exact match (EM) and 47.0% in IoU@0.5. Meanwhile, LAT not only outperforms the supervised fine-tuning baseline, which is trained to directly produce answers with attribution, but also exhibits stronger generalization across domains.

Paperid: 2152, https://arxiv.org/pdf/2508.09230.pdf

Abstract:
Vision Language Model (VLM)-based agents are stateful, autonomous entities capable of perceiving and interacting with their environments through vision and language. Multi-agent systems comprise specialized agents who collaborate to solve a (complex) task. A core security property is robustness, stating that the system should maintain its integrity under adversarial attacks. However, the design of existing multi-agent systems lacks the robustness consideration, as a successful exploit against one agent can spread and infect other agents to undermine the entire system's assurance. To address this, we propose a new defense approach, Cowpox, to provably enhance the robustness of multi-agent systems. It incorporates a distributed mechanism, which improves the recovery rate of agents by limiting the expected number of infections to other agents. The core idea is to generate and distribute a special cure sample that immunizes an agent against the attack before exposure and helps recover the already infected agents. We demonstrate the effectiveness of Cowpox empirically and provide theoretical robustness guarantees.

Paperid: 2153, https://arxiv.org/pdf/2507.18311.pdf

Abstract:
Large Vision-Language Models (LVLMs) have shown impressive capabilities across a range of tasks that integrate visual and textual understanding, such as image captioning and visual question answering. These models are trained on large-scale image and video datasets paired with text, enabling them to bridge visual perception and natural language processing. However, their application to scientific domains, especially in interpreting complex field data commonly used in the natural sciences, remains underexplored. In this work, we introduce FieldLVLM, a novel framework designed to improve large vision-language models' understanding of field data. FieldLVLM consists of two main components: a field-aware language generation strategy and a data-compressed multimodal model tuning. The field-aware language generation strategy leverages a special-purpose machine learning pipeline to extract key physical features from field data, such as flow classification, Reynolds number, and vortex patterns. This information is then converted into structured textual descriptions that serve as a dataset. The data-compressed multimodal model tuning focuses on LVLMs with these generated datasets, using a data compression strategy to reduce the complexity of field inputs and retain only the most informative values. This ensures compatibility with the models language decoder and guides its learning more effectively. Experimental results on newly proposed benchmark datasets demonstrate that FieldLVLM significantly outperforms existing methods in tasks involving scientific field data. Our findings suggest that this approach opens up new possibilities for applying large vision-language models to scientific research, helping bridge the gap between large models and domain-specific discovery.

Paperid: 2154, https://arxiv.org/pdf/2505.24342.pdf

Abstract:
Understanding what emotions images evoke in their viewers is a foundational goal in human-centric visual computing. While recent advances in vision-language models (VLMs) have shown promise for visual emotion analysis (VEA), several key challenges remain unresolved. Emotional cues in images are often abstract, overlapping, and entangled, making them difficult to model and interpret. Moreover, VLMs struggle to align these complex visual patterns with emotional semantics due to limited supervision and sparse emotional grounding. Finally, existing approaches lack structured affective knowledge to resolve ambiguity and ensure consistent emotional reasoning across diverse visual domains. To address these limitations, we propose \textbf{K-EVER\textsuperscript{2}}, a knowledge-enhanced framework for emotion reasoning and retrieval. Our approach introduces a semantically structured formulation of visual emotion cues and integrates external affective knowledge through multimodal alignment. Without relying on handcrafted labels or direct emotion supervision, K-EVER\textsuperscript{2} achieves robust and interpretable emotion predictions across heterogeneous image types. We validate our framework on three representative benchmarks, Emotion6, EmoSet, and M-Disaster, covering social media imagery, human-centric scenes, and disaster contexts. K-EVER\textsuperscript{2} consistently outperforms strong CNN and VLM baselines, achieving up to a \textbf{19\% accuracy gain} for specific emotions and a \textbf{12.3\% average accuracy gain} across all emotion categories. Our results demonstrate a scalable and generalizable solution for advancing emotional understanding of visual content.

Paperid: 2155, https://arxiv.org/pdf/2505.05937.pdf

Abstract:
As a critical psychological stress response, micro-expressions (MEs) are fleeting and subtle facial movements revealing genuine emotions. Automatic ME recognition (MER) holds valuable applications in fields such as criminal investigation and psychological diagnosis. The Facial Action Coding System (FACS) encodes expressions by identifying activations of specific facial action units (AUs), serving as a key reference for ME analysis. However, current MER methods typically limit AU utilization to defining regions of interest (ROIs) or relying on specific prior knowledge, often resulting in limited performance and poor generalization. To address this, we integrate the CLIP model's powerful cross-modal semantic alignment capability into MER and propose a novel approach namely MER-CLIP. Specifically, we convert AU labels into detailed textual descriptions of facial muscle movements, guiding fine-grained spatiotemporal ME learning by aligning visual dynamics and textual AU-based representations. Additionally, we introduce an Emotion Inference Module to capture the nuanced relationships between ME patterns and emotions with higher-level semantic understanding. To mitigate overfitting caused by the scarcity of ME data, we put forward LocalStaticFaceMix, an effective data augmentation strategy blending facial images to enhance facial diversity while preserving critical ME features. Finally, comprehensive experiments on four benchmark ME datasets confirm the superiority of MER-CLIP. Notably, UF1 scores on CAS(ME)3 reach 0.7832, 0.6544, and 0.4997 for 3-, 4-, and 7-class classification tasks, significantly outperforming previous methods.

Paperid: 2156, https://arxiv.org/pdf/2505.02013.pdf

Abstract:
Reliable face forgery detection algorithms are crucial for countering the growing threat of deepfake-driven disinformation. Previous research has demonstrated the potential of Multimodal Large Language Models (MLLMs) in identifying manipulated faces. However, existing methods typically depend on either the Large Language Model (LLM) alone or an external detector to generate classification results, which often leads to sub-optimal integration of visual and textual modalities. In this paper, we propose VLF-FFD, a novel Vision-Language Fusion solution for MLLM-enhanced Face Forgery Detection. Our key contributions are twofold. First, we present EFF++, a frame-level, explainability-driven extension of the widely used FaceForensics++ (FF++) dataset. In EFF++, each manipulated video frame is paired with a textual annotation that describes both the forgery artifacts and the specific manipulation technique applied, enabling more effective and informative MLLM training. Second, we design a Vision-Language Fusion Network (VLF-Net) that promotes bidirectional interaction between visual and textual features, supported by a three-stage training pipeline to fully leverage its potential. VLF-FFD achieves state-of-the-art (SOTA) performance in both cross-dataset and intra-dataset evaluations, underscoring its exceptional effectiveness in face forgery detection.

Paperid: 2157, https://arxiv.org/pdf/2503.22285.pdf

Abstract:
Enabling object detectors to recognize out-of-distribution (OOD) objects is vital for building reliable systems. A primary obstacle stems from the fact that models frequently do not receive supervisory signals from unfamiliar data, leading to overly confident predictions regarding OOD objects. Despite previous progress that estimates OOD uncertainty based on the detection model and in-distribution (ID) samples, we explore using pre-trained vision-language representations for object-level OOD detection. We first discuss the limitations of applying image-level CLIP-based OOD detection methods to object-level scenarios. Building upon these insights, we propose RUNA, a novel framework that leverages a dual encoder architecture to capture rich contextual information and employs a regional uncertainty alignment mechanism to distinguish ID from OOD objects effectively. We introduce a few-shot fine-tuning approach that aligns region-level semantic representations to further improve the model's capability to discriminate between similar objects. Our experiments show that RUNA substantially surpasses state-of-the-art methods in object-level OOD detection, particularly in challenging scenarios with diverse and complex object instances.

Paperid: 2158, https://arxiv.org/pdf/2501.04670.pdf

Abstract:
Recent advancements in multimodal large language models (MLLM) have shown a strong ability in visual perception, reasoning abilities, and vision-language understanding. However, the visual matching ability of MLLMs is rarely studied, despite finding the visual correspondence of objects is essential in computer vision. Our research reveals that the matching capabilities in recent MLLMs still exhibit systematic shortcomings, even with current strong MLLMs models, GPT-4o. In particular, we construct a Multimodal Visual Matching (MMVM) benchmark to fairly benchmark over 30 different MLLMs. The MMVM benchmark is built from 15 open-source datasets and Internet videos with manual annotation. We categorize the data samples of MMVM benchmark into eight aspects based on the required cues and capabilities to more comprehensively evaluate and analyze current MLLMs. In addition, we have designed an automatic annotation pipeline to generate the MMVM SFT dataset, including 220K visual matching data with reasoning annotation. To our knowledge, this is the first visual corresponding dataset and benchmark for the MLLM community. Finally, we present CoLVA, a novel contrastive MLLM with two novel technical designs: fine-grained vision expert with object-level contrastive learning and instruction augmentation strategy. The former learns instance discriminative tokens, while the latter further improves instruction following ability. CoLVA-InternVL2-4B achieves an overall accuracy (OA) of 49.80\% on the MMVM benchmark, surpassing GPT-4o and the best open-source MLLM, Qwen2VL-72B, by 7.15\% and 11.72\% OA, respectively. These results demonstrate the effectiveness of our MMVM SFT dataset and our novel technical designs. Code, benchmark, dataset, and models will be released.

Paperid: 2159, https://arxiv.org/pdf/2412.00114.pdf

Abstract:
Large vision-language models (LVLMs) have shown remarkable capabilities in interpreting visual content. While existing works demonstrate these models' vulnerability to deliberately placed adversarial texts, such texts are often easily identifiable as anomalous. In this paper, we present the first approach to generate scene-coherent typographic adversarial attacks that mislead advanced LVLMs while maintaining visual naturalness through the capability of the LLM-based agent. Our approach addresses three critical questions: what adversarial text to generate, where to place it within the scene, and how to integrate it seamlessly. We propose a training-free, multi-modal LLM-driven scene-coherent typographic adversarial planning (SceneTAP) that employs a three-stage process: scene understanding, adversarial planning, and seamless integration. The SceneTAP utilizes chain-of-thought reasoning to comprehend the scene, formulate effective adversarial text, strategically plan its placement, and provide detailed instructions for natural integration within the image. This is followed by a scene-coherent TextDiffuser that executes the attack using a local diffusion mechanism. We extend our method to real-world scenarios by printing and placing generated patches in physical environments, demonstrating its practical implications. Extensive experiments show that our scene-coherent adversarial text successfully misleads state-of-the-art LVLMs, including ChatGPT-4o, even after capturing new images of physical setups. Our evaluations demonstrate a significant increase in attack success rates while maintaining visual naturalness and contextual appropriateness. This work highlights vulnerabilities in current vision-language models to sophisticated, scene-coherent adversarial attacks and provides insights into potential defense mechanisms.

Paperid: 2160, https://arxiv.org/pdf/2410.08695.pdf

Abstract:
Large Vision-Language Models (LVLMs) have demonstrated remarkable capabilities across multimodal tasks such as visual perception and reasoning, leading to good performance on various multimodal evaluation benchmarks. However, these benchmarks keep a static nature and overlap with the pre-training data, resulting in fixed complexity constraints and data contamination issues. This raises the concern regarding the validity of the evaluation. To address these two challenges, we introduce a dynamic multimodal evaluation protocol called Vision-Language Bootstrapping (VLB). VLB provides a robust and comprehensive assessment for LVLMs with reduced data contamination and flexible complexity. To this end, VLB dynamically generates new visual question-answering samples through a multimodal bootstrapping module that modifies both images and language, while ensuring that newly generated samples remain consistent with the original ones by a judge module. By composing various bootstrapping strategies, VLB offers dynamic variants of existing benchmarks with diverse complexities, enabling the evaluation to co-evolve with the ever-evolving capabilities of LVLMs. Extensive experimental results across multiple benchmarks, including SEEDBench, MMBench, and MME, show that VLB significantly reduces data contamination and exposes performance limitations of LVLMs.

Paperid: 2161, https://arxiv.org/pdf/2601.01618.pdf

Abstract:
Long-horizon robotic manipulation is increasingly important for real-world deployment, requiring spatial disambiguation in complex layouts and temporal resilience under dynamic interaction. However, existing end-to-end and hierarchical Vision-Language-Action (VLA) policies often rely on text-only cues while keeping plan intent latent, which undermines referential grounding in cluttered or underspecified scenes, impedes effective task decomposition of long-horizon goals with close-loop interaction, and limits causal explanation by obscuring the rationale behind action choices. To address these issues, we first introduce Visual Sketch, an implausible visual intermediate that renders points, boxes, arrows, and typed relations in the robot's current views to externalize spatial intent, connect language to scene geometry. Building on Visual Sketch, we present Action-Sketcher, a VLA framework that operates in a cyclic See-Think-Sketch-Act workflow coordinated by adaptive token-gated strategy for reasoning triggers, sketch revision, and action issuance, thereby supporting reactive corrections and human interaction while preserving real-time action prediction. To enable scalable training and evaluation, we curate diverse corpus with interleaved images, text, Visual Sketch supervision, and action sequences, and train Action-Sketcher with a multi-stage curriculum recipe that combines interleaved sequence alignment for modality unification, language-to-sketch consistency for precise linguistic grounding, and imitation learning augmented with sketch-to-action reinforcement for robustness. Extensive experiments on cluttered scenes and multi-object tasks, in simulation and on real-world tasks, show improved long-horizon success, stronger robustness to dynamic scene changes, and enhanced interpretability via editable sketches and step-wise plans. Project website: https://action-sketcher.github.io

Paperid: 2162, https://arxiv.org/pdf/2512.22208.pdf

Abstract:
Recently, Large Language Models (LLMs) have undergone a significant transformation, marked by a rapid rise in both their popularity and capabilities. Leading this evolution are proprietary LLMs like GPT-4 and GPT-o1, which have captured widespread attention in the AI community due to their remarkable performance and versatility. Simultaneously, open-source LLMs, such as LLaMA and Mistral, have made great contributions to the ever-increasing popularity of LLMs due to the ease to customize and deploy the models across diverse applications. Moxin 7B is introduced as a fully open-source LLM developed in accordance with the Model Openness Framework, which moves beyond the simple sharing of model weights to embrace complete transparency in training, datasets, and implementation detail, thus fostering a more inclusive and collaborative research environment that can sustain a healthy open-source ecosystem. To further equip Moxin with various capabilities in different tasks, we develop three variants based on Moxin, including Moxin-VLM, Moxin-VLA, and Moxin-Chinese, which target the vision-language, vision-language-action, and Chinese capabilities, respectively. Experiments show that our models achieve superior performance in various evaluations. We adopt open-source framework and open data for the training. We release our models, along with the available data and code to derive these models.

Paperid: 2163, https://arxiv.org/pdf/2511.17366.pdf

Abstract:
Building a generalist robot that can perceive, reason, and act across diverse tasks remains an open challenge, especially for dexterous manipulation. A major bottleneck lies in the scarcity of large-scale, action-annotated data for dexterous skills, as teleoperation is difficult and costly. Human data, with its vast scale and diverse manipulation behaviors, provides rich priors for learning robotic actions. While prior works have explored leveraging human demonstrations, they are often constrained by limited scenarios and a large visual gap between human and robots. To eliminate these limitations, we propose METIS, a vision-language-action (VLA) model for dexterous manipulation pretrained on multi-source egocentric datasets. We first construct EgoAtlas, which integrates large-scale human and robotic data from multiple sources, all unified under a consistent action space. We further extract motion-aware dynamics, a compact and discretized motion representation, which provides efficient and expressive supervision for VLA training. Built upon them, METIS integrates reasoning and acting into a unified framework, enabling effective deployment to downstream dexterous manipulation tasks. Our method demonstrates exceptional dexterous manipulation capabilities, achieving highest average success rate in six real-world tasks. Experimental results also highlight the superior generalization and robustness to out-of-distribution scenarios. These findings emphasize METIS as a promising step toward a generalist model for dexterous manipulation.

Paperid: 2164, https://arxiv.org/pdf/2510.20696.pdf

Abstract:
Multimodal large language models (MLLMs) that integrate visual and textual reasoning leverage chain-of-thought (CoT) prompting to tackle complex visual tasks, yet continue to exhibit visual hallucinations and an over-reliance on textual priors. We present a systematic diagnosis of state-of-the-art vision-language models using a three-stage evaluation framework, uncovering key failure modes. To address these, we propose an agent-based architecture that combines LLM reasoning with lightweight visual modules, enabling fine-grained analysis and iterative refinement of reasoning chains. Our results highlight future visual reasoning models should focus on integrating a broader set of specialized tools for analyzing visual content. Our system achieves significant gains (+10.3 on MMMU, +6.0 on MathVista over a 7B baseline), matching or surpassing much larger models. We will release our framework and evaluation suite to facilitate future research.

Paperid: 2165, https://arxiv.org/pdf/2510.14952.pdf

Abstract:
Natural language offers a natural interface for humanoid robots, but existing language-guided humanoid locomotion pipelines remain cumbersome and unreliable. They typically decode human motion, retarget it to robot morphology, and then track it with a physics-based controller. However, this multi-stage process is prone to cumulative errors, introduces high latency, and yields weak coupling between semantics and control. These limitations call for a more direct pathway from language to action, one that eliminates fragile intermediate stages. Therefore, we present RoboGhost, a retargeting-free framework that directly conditions humanoid policies on language-grounded motion latents. By bypassing explicit motion decoding and retargeting, RoboGhost enables a diffusion-based policy to denoise executable actions directly from noise, preserving semantic intent and supporting fast, reactive control. A hybrid causal transformer-diffusion motion generator further ensures long-horizon consistency while maintaining stability and diversity, yielding rich latent representations for precise humanoid behavior. Extensive experiments demonstrate that RoboGhost substantially reduces deployment latency, improves success rates and tracking accuracy, and produces smooth, semantically aligned locomotion on real humanoids. Beyond text, the framework naturally extends to other modalities such as images, audio, and music, providing a general foundation for vision-language-action humanoid systems.

Paperid: 2166, https://arxiv.org/pdf/2510.11020.pdf

Abstract:
Auxiliary lines are essential for solving complex geometric problems but remain challenging for large vision-language models (LVLMs). Rather than editing diagrams to draw auxiliary lines, which current image editing models struggle to render with geometric precision, we generate textual descriptions of auxiliary-line constructions to better align with the representational strengths of LVLMs. To bridge the gap between textual descriptions and spatial structure, we propose a reinforcement learning framework that enhances diagram-text alignment. At the core of our approach is a cross-modal reward that evaluates how well the generated auxiliary-line description for an original diagram matches a ground-truth auxiliary-line diagram. Built on this reward, we present GeoVLMath, an open-source LVLM tailored to auxiliary-line reasoning in solid geometry. This fine-grained signal drives a GRPO-based RL stage, yielding precise diagram-text alignment. To support training, we develop a scalable data creation pipeline and construct AuxSolidMath, a dataset of 3,018 real-exam geometry problems with paired diagrams and aligned textual fields. At the 3B and 7B scales, GeoVLMath achieves competitive and often superior performance compared with strong open-source and proprietary LVLMs on auxiliary-line reasoning benchmarks.

Paperid: 2167, https://arxiv.org/pdf/2510.09667.pdf

Abstract:
Existing Vision-Language-Action (VLA) models can be broadly categorized into diffusion-based and auto-regressive (AR) approaches: diffusion models capture continuous action distributions but rely on computationally heavy iterative denoising. In contrast, AR models enable efficient optimization and flexible sequence construction, making them better suited for large-scale pretraining. To further improve AR efficiency, particularly when action chunks induce extended and high-dimensional sequences, prior work applies entropy-guided and token-frequency techniques to shorten the sequence length. However, such compression struggled with \textit{poor reconstruction or inefficient compression}. Motivated by this, we introduce an Omni Swift Action Tokenizer, which learns a compact, transferable action representation. Specifically, we first normalize value ranges and temporal horizons to obtain a consistent representation with B-Spline encoding. Then, we apply multi-stage residual quantization to the position, rotation, and gripper subspaces, producing compressed discrete tokens with coarse-to-fine granularity for each part. After pre-training on the large-scale dataset Droid, the resulting discrete tokenization shortens the training sequence by 6.8$\times$, and lowers the target entropy. To further explore the potential of OmniSAT, we develop a cross-embodiment learning strategy that builds on the unified action-pattern space and jointly leverages robot and human demonstrations. It enables scalable auxiliary supervision from heterogeneous egocentric videos. Across diverse real-robot and simulation experiments, OmniSAT encompasses higher compression while preserving reconstruction quality, enabling faster AR training convergence and model performance.

Paperid: 2168, https://arxiv.org/pdf/2510.07181.pdf

Abstract:
Vision-Language Models (VLMs) have shown remarkable capabilities in spatial reasoning, yet they remain fundamentally limited to qualitative precision and lack the computational precision required for real-world robotics. Current approaches fail to leverage metric cues from depth sensors and camera calibration, instead reducing geometric problems to pattern recognition tasks that cannot deliver the centimeter-level accuracy essential for robotic manipulation. We present TIGeR (Tool-Integrated Geometric Reasoning), a novel framework that transforms VLMs from perceptual estimators to geometric computers by enabling them to generate and execute precise geometric computations through external tools. Rather than attempting to internalize complex geometric operations within neural networks, TIGeR empowers models to recognize geometric reasoning requirements, synthesize appropriate computational code, and invoke specialized libraries for exact calculations. To support this paradigm, we introduce TIGeR-300K, a comprehensive tool-invocation-oriented dataset covering point transformations, pose estimation, and spatial compatibility verification, complete with tool invocation sequences and intermediate computations. Through a two-stage training pipeline combining supervised fine-tuning (SFT) and reinforcement fine-tuning (RFT) with our proposed hierarchical reward design, TIGeR achieves SOTA performance on geometric reasoning benchmarks while demonstrating centimeter-level precision in real-world robotic manipulation tasks.

Paperid: 2169, https://arxiv.org/pdf/2509.13375.pdf

Abstract:
Vision-Language Models (VLMs), such as CLIP, have demonstrated remarkable zero-shot out-of-distribution (OOD) detection capabilities, vital for reliable AI systems. Despite this promising capability, a comprehensive understanding of (1) why they work so effectively, (2) what advantages do they have over single-modal methods, and (3) how is their behavioral robustness -- remains notably incomplete within the research community. This paper presents a systematic empirical analysis of VLM-based OOD detection using in-distribution (ID) and OOD prompts. (1) Mechanisms: We systematically characterize and formalize key operational properties within the VLM embedding space that facilitate zero-shot OOD detection. (2) Advantages: We empirically quantify the superiority of these models over established single-modal approaches, attributing this distinct advantage to the VLM's capacity to leverage rich semantic novelty. (3) Sensitivity: We uncovers a significant and previously under-explored asymmetry in their robustness profile: while exhibiting resilience to common image noise, these VLM-based methods are highly sensitive to prompt phrasing. Our findings contribute a more structured understanding of the strengths and critical vulnerabilities inherent in VLM-based OOD detection, offering crucial, empirically-grounded guidance for developing more robust and reliable future designs.

Paperid: 2170, https://arxiv.org/pdf/2509.11480.pdf

Abstract:
Vision-Language-Action (VLA) models have emerged as powerful generalist policies for robotic control, yet their performance scaling across model architectures and hardware platforms, as well as their associated power budgets, remain poorly understood. This work presents an evaluation of five representative VLA models -- spanning state-of-the-art baselines and two newly proposed architectures -- targeting edge and datacenter GPU platforms. Using the LIBERO benchmark, we measure accuracy alongside system-level metrics, including latency, throughput, and peak memory usage, under varying edge power constraints and high-performance datacenter GPU configurations. Our results identify distinct scaling trends: (1) architectural choices, such as action tokenization and model backbone size, strongly influence throughput and memory footprint; (2) power-constrained edge devices exhibit non-linear performance degradation, with some configurations matching or exceeding older datacenter GPUs; and (3) high-throughput variants can be achieved without significant accuracy loss. These findings provide actionable insights when selecting and optimizing VLAs across a range of deployment constraints. Our work challenges current assumptions about the superiority of datacenter hardware for robotic inference.

Paperid: 2171, https://arxiv.org/pdf/2508.13305.pdf

Abstract:
Vision-Language Models (VLMs) have emerged as a promising paradigm in autonomous driving (AD), offering a unified framework for perception, reasoning, and decision-making by jointly modeling visual inputs and natural language instructions. However, their deployment is hindered by the significant computational overhead incurred when processing high-resolution, multi-view images, a standard setup in AD systems with six or more synchronized cameras. This overhead stems from the large number of visual tokens generated during encoding, increasing inference latency and memory consumption due to the quadratic complexity of self-attention. To address these challenges, we propose Prune2Drive, a plug-and-play visual token pruning framework for multi-view VLMs in autonomous driving. Prune2Drive introduces two core innovations: (i) a diversity-aware token selection mechanism inspired by farthest point sampling, which prioritizes semantic and spatial coverage across views rather than relying solely on attention scores, and (ii) a view-adaptive pruning controller that learns optimal pruning ratios for each camera view based on their importance to downstream driving tasks. Unlike prior methods, Prune2Drive does not require model retraining or access to attention maps, making it compatible with modern efficient attention implementations. Extensive experiments on two large-scale multi-view driving benchmarks, DriveLM and DriveLMM-o1, show that Prune2Drive achieves significant speedups and memory savings while maintaining or improving task performance. When retaining only 10% of the visual tokens, our method achieves a 6.40$\times$ speedup in the prefilling phase and consumes 13.4% of the original FLOPs, with only a 3% performance drop on the DriveLM benchmark.

Paperid: 2172, https://arxiv.org/pdf/2507.02029.pdf

Authors:BAAI RoboBrain Team, Mingyu Cao, Huajie Tan, Yuheng Ji, Xiansheng Chen, Minglan Lin, Zhiyu Li, Zhou Cao, Pengwei Wang, Enshen Zhou, Yi Han, Yingbo Tang, Xiangqi Xu, Wei Guo, Yaoxu Lyu, Yijie Xu, Jiayu Shi, Mengfei Du, Cheng Chi, Mengdi Zhao, Xiaoshuai Hao, Junkai Zhao, Xiaojie Zhang, Shanyu Rong, Huaihai Lyu, Zhengliang Cai, Yankai Fu, Ning Chen, Bolun Zhang, Lingfeng Zhang, Shuyi Zhang, Dong Liu, Xi Feng, Songjing Wang, Xiaodan Liu, Yance Jiao, Mengsi Lyu, Zhuo Chen, Chenrui He, Yulong Ao, Xue Sun, Zheqi He, Jingshu Zheng, Xi Yang, Donghai Shi, Kunchang Xie, Bochao Zhang, Shaokai Nie, Chunlei Men, Yonghua Lin, Zhongyuan Wang, Tiejun Huang, Shanghang Zhang

Abstract:
We introduce RoboBrain 2.0, our latest generation of embodied vision-language foundation models, designed to unify perception, reasoning, and planning for complex embodied tasks in physical environments. It comes in two variants: a lightweight 7B model and a full-scale 32B model, featuring a heterogeneous architecture with a vision encoder and a language model. Despite its compact size, RoboBrain 2.0 achieves strong performance across a wide spectrum of embodied reasoning tasks. On both spatial and temporal benchmarks, the 32B variant achieves leading results, surpassing prior open-source and proprietary models. In particular, it supports key real-world embodied AI capabilities, including spatial understanding (e.g., affordance prediction, spatial referring, trajectory forecasting) and temporal decision-making (e.g., closed-loop interaction, multi-agent long-horizon planning, and scene graph updating). This report details the model architecture, data construction, multi-stage training strategies, infrastructure and practical applications. We hope RoboBrain 2.0 advances embodied AI research and serves as a practical step toward building generalist embodied agents. The code, checkpoint and benchmark are available at https://superrobobrain.github.io.

Paperid: 2173, https://arxiv.org/pdf/2506.08817.pdf

Abstract:
Video content comprehension is essential for various applications, ranging from video analysis to interactive systems. Despite advancements in large-scale vision-language models (VLMs), these models often struggle to capture the nuanced, spatiotemporal details essential for thorough video analysis. To address this gap, we introduce Video-CoT, a groundbreaking dataset designed to enhance spatiotemporal understanding using Chain-of-Thought (CoT) methodologies. Video-CoT contains 192,000 fine-grained spa-tiotemporal question-answer pairs and 23,000 high-quality CoT-annotated samples, providing a solid foundation for evaluating spatiotemporal understanding in video comprehension. Additionally, we provide a comprehensive benchmark for assessing these tasks, with each task featuring 750 images and tailored evaluation metrics. Our extensive experiments reveal that current VLMs face significant challenges in achieving satisfactory performance, high-lighting the difficulties of effective spatiotemporal understanding. Overall, the Video-CoT dataset and benchmark open new avenues for research in multimedia understanding and support future innovations in intelligent systems requiring advanced video analysis capabilities. By making these resources publicly available, we aim to encourage further exploration in this critical area. Project website:https://video-cot.github.io/ .

Paperid: 2174, https://arxiv.org/pdf/2505.21432.pdf

Abstract:
Humans practice slow thinking before performing actual actions when handling complex tasks in the physical world. This thinking paradigm, recently, has achieved remarkable advancement in boosting Large Language Models (LLMs) to solve complex tasks in digital domains. However, the potential of slow thinking remains largely unexplored for robotic foundation models interacting with the physical world. In this work, we propose Hume: a dual-system Vision-Language-Action (VLA) model with value-guided System-2 thinking and cascaded action denoising, exploring human-like thinking capabilities of Vision-Language-Action models for dexterous robot control. System 2 of Hume implements value-Guided thinking by extending a Vision-Language-Action Model backbone with a novel value-query head to estimate the state-action value of predicted actions. The value-guided thinking is conducted by repeat sampling multiple action candidates and selecting one according to state-action value. System 1 of Hume is a lightweight reactive visuomotor policy that takes System 2 selected action and performs cascaded action denoising for dexterous robot control. At deployment time, System 2 performs value-guided thinking at a low frequency while System 1 asynchronously receives the System 2 selected action candidate and predicts fluid actions in real time. We show that Hume outperforms the existing state-of-the-art Vision-Language-Action models across multiple simulation benchmark and real-robot deployments.

Paperid: 2175, https://arxiv.org/pdf/2505.00275.pdf

Abstract:
Chronic diseases, including diabetes, hypertension, asthma, HIV-AIDS, epilepsy, and tuberculosis, necessitate rigorous adherence to medication to avert disease progression, manage symptoms, and decrease mortality rates. Adherence is frequently undermined by factors including patient behavior, caregiver support, elevated medical costs, and insufficient healthcare infrastructure. We propose AdCare-VLM, a specialized Video-LLaVA-based multimodal large vision language model (LVLM) aimed at visual question answering (VQA) concerning medication adherence through patient videos. We employ a private dataset comprising 806 custom-annotated tuberculosis (TB) medication monitoring videos, which have been labeled by clinical experts, to fine-tune the model for adherence pattern detection. We present LLM-TB-VQA, a detailed medical adherence VQA dataset that encompasses positive, negative, and ambiguous adherence cases. Our method identifies correlations between visual features, such as the clear visibility of the patient's face, medication, water intake, and the act of ingestion, and their associated medical concepts in captions. This facilitates the integration of aligned visual-linguistic representations and improves multimodal interactions. Experimental results indicate that our method surpasses parameter-efficient fine-tuning (PEFT) enabled VLM models, such as LLaVA-V1.5 and Chat-UniVi, with absolute improvements ranging from 3.1% to 3.54% across pre-trained, regular, and low-rank adaptation (LoRA) configurations. Comprehensive ablation studies and attention map visualizations substantiate our approach, enhancing interpretability.

Paperid: 2176, https://arxiv.org/pdf/2502.18041.pdf

Abstract:
Vision-Language Navigation (VLN) aims to guide agents by leveraging language instructions and visual cues, playing a pivotal role in embodied AI. Indoor VLN has been extensively studied, whereas outdoor aerial VLN remains underexplored. The potential reason is that outdoor aerial view encompasses vast areas, making data collection more challenging, which results in a lack of benchmarks. To address this problem, we propose OpenFly, a platform comprising various rendering engines, a versatile toolchain, and a large-scale benchmark for aerial VLN. Firstly, we integrate diverse rendering engines and advanced techniques for environment simulation, including Unreal Engine, GTA V, Google Earth, and 3D Gaussian Splatting (3D GS). Particularly, 3D GS supports real-to-sim rendering, further enhancing the realism of our environments. Secondly, we develop a highly automated toolchain for aerial VLN data collection, streamlining point cloud acquisition, scene semantic segmentation, flight trajectory creation, and instruction generation. Thirdly, based on the toolchain, we construct a large-scale aerial VLN dataset with 100k trajectories, covering diverse heights and lengths across 18 scenes. Moreover, we propose OpenFly-Agent, a keyframe-aware VLN model emphasizing key observations during flight. For benchmarking, extensive experiments and analyses are conducted, evaluating several recent VLN methods and showcasing the superiority of our OpenFly platform and agent. The toolchain, dataset, and codes will be open-sourced.

Paperid: 2177, https://arxiv.org/pdf/2501.15830.pdf

Abstract:
In this paper, we claim that spatial understanding is the keypoint in robot manipulation, and propose SpatialVLA to explore effective spatial representations for the robot foundation model. Specifically, we introduce Ego3D Position Encoding to inject 3D information into the input observations of the visual-language-action model, and propose Adaptive Action Grids to represent spatial robot movement actions with adaptive discretized action grids, facilitating learning generalizable and transferrable spatial action knowledge for cross-robot control. SpatialVLA is first pre-trained on top of a vision-language model with 1.1 Million real-world robot episodes, to learn a generalist manipulation policy across multiple robot environments and tasks. After pre-training, SpatialVLA is directly applied to perform numerous tasks in a zero-shot manner. The superior results in both simulation and real-world robots demonstrate its advantage of inferring complex robot motion trajectories and its strong in-domain multi-task generalization ability. We further show the proposed Adaptive Action Grids offer a new and effective way to fine-tune the pre-trained SpatialVLA model for new simulation and real-world setups, where the pre-learned action grids are re-discretized to capture robot-specific spatial action movements of new setups. The superior results from extensive evaluations demonstrate the exceptional in-distribution generalization and out-of-distribution adaptation capability, highlighting the crucial benefit of the proposed spatial-aware representations for generalist robot policy learning. All the details and codes will be open-sourced.

Paperid: 2178, https://arxiv.org/pdf/2412.06845.pdf

Abstract:
Recently, Large Language Models (LLMs) have undergone a significant transformation, marked by a rapid rise in both their popularity and capabilities. Leading this evolution are proprietary LLMs like GPT-4 and GPT-o1, which have captured widespread attention in the AI community due to their remarkable performance and versatility. Simultaneously, open-source LLMs, such as LLaMA, have made great contributions to the ever-increasing popularity of LLMs due to the ease to customize and deploy the models across diverse applications. Although open-source LLMs present unprecedented opportunities for innovation and research, the commercialization of LLMs has raised concerns about transparency, reproducibility, and safety. Many open-source LLMs fail to meet fundamental transparency requirements by withholding essential components like training code and data, which may hinder further innovations on LLMs. To mitigate this issue, we introduce Moxin 7B, a fully open-source LLM developed, adhering to principles of open science, open source, open data, and open access. We release the pre-training code and configurations, training and fine-tuning datasets, and intermediate and final checkpoints, aiming to make continuous commitments to fully open-source LLMs. After pre-training the base model, we finetune the Moxin Base model with SOTA post-training framework and instruction data to obtain Moxin Instruct model. To improve the reasoning capability, we further finetune our Instruct model with chain-of-thought data distilled from DeepSeek R1, and then use Group Relative Policy Optimization (GRPO) following DeepSeek R1 to finetune our model, leading to the Moxin Reasoning model. Moreover, we develop our vision language model based on our Moxin model. Experiments show that our models achieve superior performance in various evaluations such as zero-shot evaluation, few-shot evaluation, and CoT evaluation.

Paperid: 2179, https://arxiv.org/pdf/2411.16253.pdf

Abstract:
Open-vocabulary 3D scene understanding is indispensable for embodied agents. Recent works leverage pretrained vision-language models (VLMs) for object segmentation and project them to point clouds to build 3D maps. Despite progress, a point cloud is a set of unordered coordinates that requires substantial storage space and does not directly convey occupancy information or spatial relation, making existing methods inefficient for downstream tasks, e.g., path planning and complex text-based object retrieval. To address these issues, we propose Octree-Graph, a novel scene representation for open-vocabulary 3D scene understanding. Specifically, a Chronological Group-wise Segment Merging (CGSM) strategy and an Instance Feature Aggregation (IFA) algorithm are first designed to get 3D instances and corresponding semantic features. Subsequently, an adaptive-octree structure is developed that stores semantics and depicts the occupancy of an object adjustably according to its shape. Finally, the Octree-Graph is constructed where each adaptive-octree acts as a graph node, and edges describe the spatial relations among nodes. Extensive experiments on various tasks are conducted on several widely-used datasets, demonstrating the versatility and effectiveness of our method.

Paperid: 2180, https://arxiv.org/pdf/2409.16604.pdf

Abstract:
Despite the impressive advancements made in recent low-light image enhancement techniques, the scarcity of paired data has emerged as a significant obstacle to further advancements. This work proposes a mean-teacher-based semi-supervised low-light enhancement (Semi-LLIE) framework that integrates the unpaired data into model training. The mean-teacher technique is a prominent semi-supervised learning method, successfully adopted for addressing high-level and low-level vision tasks. However, two primary issues hinder the naive mean-teacher method from attaining optimal performance in low-light image enhancement. Firstly, pixel-wise consistency loss is insufficient for transferring realistic illumination distribution from the teacher to the student model, which results in color cast in the enhanced images. Secondly, cutting-edge image enhancement approaches fail to effectively cooperate with the mean-teacher framework to restore detailed information in dark areas due to their tendency to overlook modeling structured information within local regions. To mitigate the above issues, we first introduce a semantic-aware contrastive loss to faithfully transfer the illumination distribution, contributing to enhancing images with natural colors. Then, we design a Mamba-based low-light image enhancement backbone to effectively enhance Mamba's local region pixel relationship representation ability with a multi-scale feature learning scheme, facilitating the generation of images with rich textural details. Further, we propose novel perceptive loss based on the large-scale vision-language Recognize Anything Model (RAM) to help generate enhanced images with richer textual details. The experimental results indicate that our Semi-LLIE surpasses existing methods in both quantitative and qualitative metrics.

Paperid: 2181, https://arxiv.org/pdf/2601.19507.pdf

Abstract:
Large vision-language models (LVLMs) exhibit remarkable capabilities in cross-modal tasks but face significant safety challenges, which undermine their reliability in real-world applications. Efforts have been made to build LVLM safety evaluation benchmarks to uncover their vulnerability. However, existing benchmarks are hindered by their labor-intensive construction process, static complexity, and limited discriminative power. Thus, they may fail to keep pace with rapidly evolving models and emerging risks. To address these limitations, we propose VLSafetyBencher, the first automated system for LVLM safety benchmarking. VLSafetyBencher introduces four collaborative agents: Data Preprocessing, Generation, Augmentation, and Selection agents to construct and select high-quality samples. Experiments validates that VLSafetyBencher can construct high-quality safety benchmarks within one week at a minimal cost. The generated benchmark effectively distinguish safety, with a safety rate disparity of 70% between the most and least safe models.

Paperid: 2182, https://arxiv.org/pdf/2511.19221.pdf

Abstract:
Autonomous driving heavily relies on accurate and robust spatial perception. Many failures arise from inaccuracies and instability, especially in long-tail scenarios and complex interactions. However, current vision-language models are weak at spatial grounding and understanding, and VLA systems built on them therefore show limited perception and localization ability. To address these challenges, we introduce Percept-WAM, a perception-enhanced World-Awareness-Action Model that is the first to implicitly integrate 2D/3D scene understanding abilities within a single vision-language model (VLM). Instead of relying on QA-style spatial reasoning, Percept-WAM unifies 2D/3D perception tasks into World-PV and World-BEV tokens, which encode both spatial coordinates and confidence. We propose a grid-conditioned prediction mechanism for dense object perception, incorporating IoU-aware scoring and parallel autoregressive decoding, improving stability in long-tail, far-range, and small-object scenarios. Additionally, Percept-WAM leverages pretrained VLM parameters to retain general intelligence (e.g., logical reasoning) and can output perception results and trajectory control outputs directly. Experiments show that Percept-WAM matches or surpasses classical detectors and segmenters on downstream perception benchmarks, achieving 51.7/58.9 mAP on COCO 2D detection and nuScenes BEV 3D detection. When integrated with trajectory decoders, it further improves planning performance on nuScenes and NAVSIM, e.g., surpassing DiffusionDrive by 2.1 in PMDS on NAVSIM. Qualitative results further highlight its strong open-vocabulary and long-tail generalization.

Paperid: 2183, https://arxiv.org/pdf/2510.12693.pdf

Abstract:
Recent advances in embodied AI highlight the potential of vision language models (VLMs) as agents capable of perception, reasoning, and interaction in complex environments. However, top-performing systems rely on large-scale models that are costly to deploy, while smaller VLMs lack the necessary knowledge and skills to succeed. To bridge this gap, we present \textit{Embodied Reasoning Agent (ERA)}, a two-stage framework that integrates prior knowledge learning and online reinforcement learning (RL). The first stage, \textit{Embodied Prior Learning}, distills foundational knowledge from three types of data: (1) Trajectory-Augmented Priors, which enrich existing trajectory data with structured reasoning generated by stronger models; (2) Environment-Anchored Priors, which provide in-environment knowledge and grounding supervision; and (3) External Knowledge Priors, which transfer general knowledge from out-of-environment datasets. In the second stage, we develop an online RL pipeline that builds on these priors to further enhance agent performance. To overcome the inherent challenges in agent RL, including long horizons, sparse rewards, and training instability, we introduce three key designs: self-summarization for context management, dense reward shaping, and turn-level policy optimization. Extensive experiments on both high-level planning (EB-ALFRED) and low-level control (EB-Manipulation) tasks demonstrate that ERA-3B surpasses both prompting-based large models and previous training-based baselines. Specifically, it achieves overall improvements of 8.4\% on EB-ALFRED and 19.4\% on EB-Manipulation over GPT-4o, and exhibits strong generalization to unseen tasks. Overall, ERA offers a practical path toward scalable embodied intelligence, providing methodological insights for future embodied AI systems.

Paperid: 2184, https://arxiv.org/pdf/2510.04401.pdf

Abstract:
Vision-Language Models (VLMs) have become a central focus of today's AI community, owing to their impressive abilities gained from training on large-scale vision-language data from the Web. These models have demonstrated strong performance across diverse tasks, including image understanding, video understanding, complex visual reasoning, and embodied AI. Despite these noteworthy successes, a fundamental question remains: Can VLMs count objects correctly? In this paper, we introduce a simple yet effective benchmark, VLMCountBench, designed under a minimalist setting with only basic geometric shapes (e.g., triangles, circles) and their compositions, focusing exclusively on counting tasks without interference from other factors. We adopt strict independent variable control and systematically study the effects of simple properties such as color, size, and prompt refinement in a controlled ablation. Our empirical results reveal that while VLMs can count reliably when only one shape type is present, they exhibit substantial failures when multiple shape types are combined (i.e., compositional counting). This highlights a fundamental empirical limitation of current VLMs and motivates important directions for future research.

Paperid: 2185, https://arxiv.org/pdf/2507.18107.pdf

Abstract:
Text-to-video (T2V) models have shown remarkable performance in generating visually reasonable scenes, while their capability to leverage world knowledge for ensuring semantic consistency and factual accuracy remains largely understudied. In response to this challenge, we propose T2VWorldBench, the first systematic evaluation framework for evaluating the world knowledge generation abilities of text-to-video models, covering 6 major categories, 60 subcategories, and 1,200 prompts across a wide range of domains, including physics, nature, activity, culture, causality, and object. To address both human preference and scalable evaluation, our benchmark incorporates both human evaluation and automated evaluation using vision-language models (VLMs). We evaluated the 10 most advanced text-to-video models currently available, ranging from open source to commercial models, and found that most models are unable to understand world knowledge and generate truly correct videos. These findings point out a critical gap in the capability of current text-to-video models to leverage world knowledge, providing valuable research opportunities and entry points for constructing models with robust capabilities for commonsense reasoning and factual generation.

Paperid: 2186, https://arxiv.org/pdf/2506.13888.pdf

Abstract:
Reinforcement Fine-Tuning (RFT) with verifiable rewards has advanced large language models but remains underexplored for Vision-Language (VL) models. The Vision-Language Reward Model (VL-RM) is key to aligning VL models by providing structured feedback, yet training effective VL-RMs faces two major challenges. First, the bootstrapping dilemma arises as high-quality training data depends on already strong VL models, creating a cycle where self-generated supervision reinforces existing biases. Second, modality bias and negative example amplification occur when VL models hallucinate incorrect visual attributes, leading to flawed preference data that further misguides training. To address these issues, we propose an iterative training framework leveraging vision experts, Chain-of-Thought (CoT) rationales, and Margin-based Rejection Sampling. Our approach refines preference datasets, enhances structured critiques, and iteratively improves reasoning. Experiments across VL-RM benchmarks demonstrate superior performance in hallucination detection and multimodal reasoning, advancing VL model alignment with reinforcement learning.

Paperid: 2187, https://arxiv.org/pdf/2412.07779.pdf

Abstract:
As multi-modal large language models (MLLMs) are increasingly applied to complex reasoning tasks, the diversity and quality of reasoning paths become crucial factors affecting their performance. Although current methods aim to enhance reasoning quality through path expansion, they often neglect the diversity of reasoning paths and effective information sharing, leading to local optima and inefficiency. To address these challenges, we propose Evolution of Thought (EoT), a multi-objective framework designed to improve reasoning by fostering both high-quality and diverse reasoning paths. Specifically, we introduce the Non-dominated Sorting Genetic Algorithm II for multi-objective optimization, utilizing crossover and mutation operators to promote greater diversity in reasoning solutions. Additionally, we propose a Condensation-Aggregation mechanism to cluster and eliminate redundant paths, facilitate improved information sharing among parent nodes, and ultimately enhance both the efficiency and quality of the reasoning process. Validation experiments on various vision-language and language reasoning tasks demonstrate that EoT achieves superior reasoning performance and efficiency compared to other competitive baselines. Our study provides a novel perspective on the design of heuristic reasoning frameworks for MLLMs.

Paperid: 2188, https://arxiv.org/pdf/2512.10548.pdf

Abstract:
Multimodal large language models (MLLMs) have achieved remarkable progress on various vision-language tasks, yet their visual perception remains limited. Humans, in comparison, perceive complex scenes efficiently by dynamically scanning and focusing on salient regions in a sequential "blink-like" process. Motivated by this strategy, we first investigate whether MLLMs exhibit similar behavior. Our pilot analysis reveals that MLLMs naturally attend to different visual regions across layers and that selectively allocating more computation to salient tokens can enhance visual perception. Building on this insight, we propose Blink, a dynamic visual token resolution framework that emulates the human-inspired process within a single forward pass. Specifically, Blink includes two modules: saliency-guided scanning and dynamic token resolution. It first estimates the saliency of visual tokens in each layer based on the attention map, and extends important tokens through a plug-and-play token super-resolution (TokenSR) module. In the next layer, it drops the extended tokens when they lose focus. This dynamic mechanism balances broad exploration and fine-grained focus, thereby enhancing visual perception adaptively and efficiently. Extensive experiments validate Blink, demonstrating its effectiveness in enhancing visual perception and multimodal understanding.

Paperid: 2189, https://arxiv.org/pdf/2512.03542.pdf

Abstract:
Multimodal Large Language Models (MLLMs) excel in numerous vision-language tasks yet suffer from hallucinations, producing content inconsistent with input visuals, that undermine reliability in precision-sensitive domains. This issue stems from a fundamental problem of visual neglect, where models fail to adequately prioritize input images. Existing methods typically alleviate hallucinations by intervening in the attention score or output logits, focusing on "how to intervene" but overlooking the prerequisite "when to intervene", which leads to the "over-intervention" problem and subsequently introduces new hallucinations and unnecessary computational overhead. To address this gap, we first investigate the mechanism of visual neglect and reveal it can be accurately detected via head-level activation patterns in MLLMs. We thus propose V-ITI, a lightweight visual inference-time intervention framework integrating a Visual Neglect Detector that identifies visual neglect via head-level discriminative probes and a Visual Recall Intervenor that modulates activations with prestored visual activation information only when the visual neglect is detected. Extensive experiments across eight benchmarks and different MLLM families demonstrate that V-ITI consistently mitigates vision-related hallucinations while preserving general task performance.

Paperid: 2190, https://arxiv.org/pdf/2510.15349.pdf

Abstract:
Document parsing from scanned images into structured formats remains a significant challenge due to its complexly intertwined elements such as text paragraphs, figures, formulas, and tables. Existing supervised fine-tuning methods often struggle to generalize across diverse document types, leading to poor performance, particularly on out-of-distribution data. This issue is further exacerbated by the limited availability of high-quality training data for layout-aware parsing tasks. To address these challenges, we introduce LayoutRL, a reinforcement learning framework that optimizes layout understanding through composite rewards integrating normalized edit distance, paragraph count accuracy, and reading order preservation. To support this training, we construct the Infinity-Doc-400K dataset, which we use to train Infinity-Parser, a vision-language model demonstrating robust generalization across various domains. Extensive evaluations on benchmarks including OmniDocBench, olmOCR-Bench, PubTabNet, and FinTabNet show that Infinity-Parser consistently achieves state-of-the-art performance across a broad range of document types, languages, and structural complexities, substantially outperforming both specialized document parsing systems and general-purpose vision-language models. We will release our code, dataset, and model to facilitate reproducible research in document parsing.

Paperid: 2191, https://arxiv.org/pdf/2510.09507.pdf

Abstract:
The ability to use, understand, and create tools is a hallmark of human intelligence, enabling sophisticated interaction with the physical world. For any general-purpose intelligent agent to achieve true versatility, it must also master these fundamental skills. While modern Multimodal Large Language Models (MLLMs) leverage their extensive common knowledge for high-level planning in embodied AI and in downstream Vision-Language-Action (VLA) models, the extent of their true understanding of physical tools remains unquantified. To bridge this gap, we present PhysToolBench, the first benchmark dedicated to evaluating the comprehension of physical tools by MLLMs. Our benchmark is structured as a Visual Question Answering (VQA) dataset comprising over 1,000 image-text pairs. It assesses capabilities across three distinct difficulty levels: (1) Tool Recognition: Requiring the recognition of a tool's primary function. (2) Tool Understanding: Testing the ability to grasp the underlying principles of a tool's operation. (3) Tool Creation: Challenging the model to fashion a new tool from surrounding objects when conventional options are unavailable. Our comprehensive evaluation of 32 MLLMs-spanning proprietary, open-source, specialized embodied, and backbones in VLAs-reveals a significant deficiency in tool understanding. Furthermore, we provide an in-depth analysis and propose preliminary solutions. Code and dataset are publicly available.

Paperid: 2192, https://arxiv.org/pdf/2509.23827.pdf

Abstract:
Artificial Intelligence have profoundly transformed the technological landscape in recent years. Large Language Models (LLMs) have demonstrated impressive abilities in reasoning, text comprehension, contextual pattern recognition, and integrating language with visual understanding. While these advances offer significant benefits, they also reveal critical limitations in the models' ability to grasp the notion of privacy. There is hence substantial interest in determining if and how these models can understand and enforce privacy principles, particularly given the lack of supporting resources to test such a task. In this work, we address these challenges by examining how legal frameworks can inform the capabilities of these emerging technologies. To this end, we introduce a comprehensive, multi-level Visual Privacy Taxonomy that captures a wide range of privacy issues, designed to be scalable and adaptable to existing and future research needs. Furthermore, we evaluate the capabilities of several state-of-the-art Vision-Language Models (VLMs), revealing significant inconsistencies in their understanding of contextual privacy. Our work contributes both a foundational taxonomy for future research and a critical benchmark of current model limitations, demonstrating the urgent need for more robust, privacy-aware AI systems.

Paperid: 2193, https://arxiv.org/pdf/2509.21980.pdf

Abstract:
With the rise in popularity of smart glasses, users' attention has been integrated into Vision-Language Models (VLMs) to streamline multi-modal querying in daily scenarios. However, leveraging gaze data to model users' attention may introduce ambiguity challenges: (1) users' verbal questions become ambiguous by using pronouns or skipping context, (2) humans' gaze patterns can be noisy and exhibit complex spatiotemporal relationships with their spoken questions. Previous works only consider single image as visual modality input, failing to capture the dynamic nature of the user's attention. In this work, we introduce GLARIFY, a novel method to leverage spatiotemporal gaze information to enhance the model's effectiveness in real-world applications. Initially, we analyzed hundreds of querying samples with the gaze modality to demonstrate the noisy nature of users' gaze patterns. We then utilized GPT-4o to design an automatic data synthesis pipeline to generate the GLARIFY-Ambi dataset, which includes a dedicated chain-of-thought (CoT) process to handle noisy gaze patterns. Finally, we designed a heatmap module to incorporate gaze information into cutting-edge VLMs while preserving their pretrained knowledge. We evaluated GLARIFY using a hold-out test set. Experiments demonstrate that GLARIFY significantly outperforms baselines. By robustly aligning VLMs with human attention, GLARIFY paves the way for a usable and intuitive interaction paradigm with a visual assistant.

Paperid: 2194, https://arxiv.org/pdf/2508.21112.pdf

Abstract:
The human ability to seamlessly perform multimodal reasoning and physical interaction in the open world is a core goal for general-purpose embodied intelligent systems. Recent vision-language-action (VLA) models, which are co-trained on large-scale robot and visual-text data, have demonstrated notable progress in general robot control. However, they still fail to achieve human-level flexibility in interleaved reasoning and interaction. In this work, introduce EO-Robotics, consists of EO-1 model and EO-Data1.5M dataset. EO-1 is a unified embodied foundation model that achieves superior performance in multimodal embodied reasoning and robot control through interleaved vision-text-action pre-training. The development of EO-1 is based on two key pillars: (i) a unified architecture that processes multimodal inputs indiscriminately (image, text, video, and action), and (ii) a massive, high-quality multimodal embodied reasoning dataset, EO-Data1.5M, which contains over 1.5 million samples with emphasis on interleaved vision-text-action comprehension. EO-1 is trained through synergies between auto-regressive decoding and flow matching denoising on EO-Data1.5M, enabling seamless robot action generation and multimodal embodied reasoning. Extensive experiments demonstrate the effectiveness of interleaved vision-text-action learning for open-world understanding and generalization, validated through a variety of long-horizon, dexterous manipulation tasks across multiple embodiments. This paper details the architecture of EO-1, the data construction strategy of EO-Data1.5M, and the training methodology, offering valuable insights for developing advanced embodied foundation models.

Paperid: 2195, https://arxiv.org/pdf/2507.00469.pdf

Abstract:
Frontier vision-language models (VLMs) have made remarkable improvements in video understanding tasks. However, real-world videos typically exist as continuously evolving data streams (e.g., dynamic scenes captured by wearable glasses), necessitating models to continually adapt to shifting data distributions and novel scenarios. Considering the prohibitive computational costs of fine-tuning models on new tasks, usually, a small subset of parameters is updated while the bulk of the model remains frozen. This poses new challenges to existing continual learning frameworks in the context of large multimodal foundation models, i.e., catastrophic forgetting and update conflict. While the foundation models struggle with parameter-efficient continual learning, the hippocampus in the human brain has evolved highly efficient mechanisms for memory formation and consolidation. Inspired by the rapid Binding and pattern separation mechanisms in the hippocampus, in this work, we propose Bisecle for video-language continual learning, where a multi-directional supervision module is used to capture more cross-modal relationships and a contrastive prompt learning scheme is designed to isolate task-specific knowledge to facilitate efficient memory storage. Binding and separation processes further strengthen the ability of VLMs to retain complex experiences, enabling robust and efficient continual learning in video understanding tasks. We perform a thorough evaluation of the proposed Bisecle, demonstrating its ability to mitigate forgetting and enhance cross-task generalization on several VideoQA benchmarks.

Paperid: 2196, https://arxiv.org/pdf/2506.03197.pdf

Abstract:
Automated parsing of scanned documents into richly structured, machine-readable formats remains a critical bottleneck in Document AI, as traditional multi-stage pipelines suffer from error propagation and limited adaptability to diverse layouts. We introduce layoutRL, an end-to-end reinforcement learning framework that trains models to be explicitly layout-aware by optimizing a composite reward of normalized edit distance, paragraph count accuracy, and reading order preservation. Leveraging our newly released dataset, Infinity-Doc-55K, which combines 55K high-fidelity synthetic scanned document parsing data with expert-filtered real-world documents, we instantiate layoutRL in a vision-language-model-based parser called Infinity-Parser. Evaluated on English and Chinese benchmarks for OCR, table and formula extraction, and reading order detection, Infinity-Parser achieves new state-of-the-art performance in both accuracy and structural fidelity, outpacing specialist pipelines and general-purpose vision-language models. We will publicly release our code and dataset to accelerate progress in robust document understanding.

Paperid: 2197, https://arxiv.org/pdf/2501.12632.pdf

Abstract:
Weakly supervised object localization (WSOL) using classification models trained with only image-class labels remains an important challenge in computer vision. Given their reliance on classification objectives, traditional WSOL methods like class activation mapping focus on the most discriminative object parts, often missing the full spatial extent. In contrast, recent WSOL methods based on vision-language models like CLIP require ground truth classes or external classifiers to produce a localization map, limiting their deployment in downstream tasks. Moreover, methods like GenPromp attempt to address these issues but introduce considerable complexity due to their reliance on conditional denoising processes and intricate prompt learning. This paper introduces Text Distillation for Localization (TeD-Loc), an approach that directly distills knowledge from CLIP text embeddings into the model backbone and produces patch-level localization. Multiple instance learning of these image patches allows for accurate localization and classification using one model without requiring external classifiers. Such integration of textual and visual modalities addresses the longstanding challenge of achieving accurate localization and classification concurrently, as WSOL methods in the literature typically converge at different epochs. Extensive experiments show that leveraging text embeddings and localization cues provides a cost-effective WSOL model. TeD-Loc improves Top-1 LOC accuracy over state-of-the-art models by about 5% on both CUB and ILSVRC datasets, while significantly reducing computational complexity compared to GenPromp.

Paperid: 2198, https://arxiv.org/pdf/2410.08500.pdf

Abstract:
Aerial Vision-and-Language Navigation (VLN) is a novel task enabling Unmanned Aerial Vehicles (UAVs) to navigate in outdoor environments through natural language instructions and visual cues. However, it remains challenging due to the complex spatial relationships in aerial scenes.In this paper, we propose a training-free, zero-shot framework for aerial VLN tasks, where the large language model (LLM) is leveraged as the agent for action prediction. Specifically, we develop a novel Semantic-Topo-Metric Representation (STMR) to enhance the spatial reasoning capabilities of LLMs. This is achieved by extracting and projecting instruction-related semantic masks onto a top-down map, which presents spatial and topological information about surrounding landmarks and grows during the navigation process. At each step, a local map centered at the UAV is extracted from the growing top-down map, and transformed into a ma trix representation with distance metrics, serving as the text prompt to LLM for action prediction in response to the given instruction. Experiments conducted in real and simulation environments have proved the effectiveness and robustness of our method, achieving absolute success rate improvements of 26.8% and 5.8% over current state-of-the-art methods on simple and complex navigation tasks, respectively. The dataset and code will be released soon.

Paperid: 2199, https://arxiv.org/pdf/2601.21821.pdf

Abstract:
Recent advances in Vision Language Models (VLMs) have driven significant progress in visual reasoning. However, open-source VLMs still lag behind proprietary systems, largely due to the lack of high-quality reasoning data. Existing datasets offer limited coverage of challenging domains such as STEM diagrams and visual puzzles, and lack consistent, long-form Chain-of-Thought (CoT) annotations essential for eliciting strong reasoning capabilities. To bridge this gap, we introduce MMFineReason, a large-scale multimodal reasoning dataset comprising 1.8M samples and 5.1B solution tokens, featuring high-quality reasoning annotations distilled from Qwen3-VL-235B-A22B-Thinking. The dataset is established via a systematic three-stage pipeline: (1) large-scale data collection and standardization, (2) CoT rationale generation, and (3) comprehensive selection based on reasoning quality and difficulty awareness. The resulting dataset spans STEM problems, visual puzzles, games, and complex diagrams, with each sample annotated with visually grounded reasoning traces. We fine-tune Qwen3-VL-Instruct on MMFineReason to develop MMFineReason-2B/4B/8B versions. Our models establish new state-of-the-art results for their size class. Notably, MMFineReason-4B succesfully surpasses Qwen3-VL-8B-Thinking, and MMFineReason-8B even outperforms Qwen3-VL-30B-A3B-Thinking while approaching Qwen3-VL-32B-Thinking, demonstrating remarkable parameter efficiency. Crucially, we uncover a "less is more" phenomenon via our difficulty-aware filtering strategy: a subset of just 7\% (123K samples) achieves performance comparable to the full dataset. Notably, we reveal a synergistic effect where reasoning-oriented data composition simultaneously boosts general capabilities.

Paperid: 2200, https://arxiv.org/pdf/2512.17306.pdf

Abstract:
Recent advances in large Vision-Language Models (VLMs) have exhibited strong reasoning capabilities on complex visual tasks by thinking with images in their Chain-of-Thought (CoT), which is achieved by actively invoking tools to analyze visual inputs rather than merely perceiving them. However, existing models often struggle to reflect on and correct themselves when attempting incorrect reasoning trajectories. To address this limitation, we propose DRIM, a model that enables deep but reliable multi-turn reasoning when thinking with images in its multimodal CoT. Our pipeline comprises three stages: data construction, cold-start SFT and RL. Based on a high-resolution image dataset, we construct high-difficulty and verifiable visual question-answer pairs, where solving each task requires multi-turn tool calls to reach the correct answer. In the SFT stage, we collect tool trajectories as cold-start data, guiding a multi-turn reasoning pattern. In the RL stage, we introduce redundancy-penalized policy optimization, which incentivizes the model to develop a self-reflective reasoning pattern. The basic idea is to impose judgment on reasoning trajectories and penalize those that produce incorrect answers without sufficient multi-scale exploration. Extensive experiments demonstrate that DRIM achieves superior performance on visual understanding benchmarks.

Paperid: 2201, https://arxiv.org/pdf/2512.06835.pdf

Abstract:
Recent vision-language models (VLMs) achieve remarkable reasoning through reinforcement learning (RL), which provides a feasible solution for realizing continuous self-evolving large vision-language models (LVLMs) in the era of experience. However, RL for VLMs requires abundant high-quality multimodal data, especially challenging in specialized domains like chemistry, earth sciences, and multimodal mathematics. Existing strategies such as synthetic data and self-rewarding mechanisms suffer from limited distributions and alignment difficulties, ultimately causing reward hacking: models exploit high-reward patterns, collapsing policy entropy and destabilizing training. We propose DoGe (Decouple to Generalize), a dual-decoupling framework that guides models to first learn from context rather than problem solving by refocusing on the problem context scenarios overlooked by synthetic data methods. By decoupling learning process into dual components (Thinker and Solver), we reasonably quantify the reward signals of this process and propose a two-stage RL post-training approach from freely exploring context to practically solving tasks. Second, to increase the diversity of training data, DoGe constructs an evolving curriculum learning pipeline: an expanded native domain knowledge corpus and an iteratively evolving seed problems pool. Experiments show that our method consistently outperforms the baseline across various benchmarks, providing a scalable pathway for realizing self-evolving LVLMs.

Paperid: 2202, https://arxiv.org/pdf/2511.13420.pdf

Abstract:
Most research on hallucinations in Large Vision-Language Models (LVLMs) focuses on factual description tasks that prohibit any output absent from the image. However, little attention has been paid to hallucinations in voluntary imagination tasks, e.g., story writing, where the models are expected to generate novel content beyond the given image. In these tasks, it is inappropriate to simply regard such imagined novel content as hallucinations. To address this limitation, we introduce Voluntary-imagined Object Presence Evaluation (VOPE)-a novel method to assess LVLMs' hallucinations in voluntary imagination tasks via presence evaluation. Specifically, VOPE poses recheck-based questions to evaluate how an LVLM interprets the presence of the imagined objects in its own response. The consistency between the model's interpretation and the object's presence in the image is then used to determine whether the model hallucinates when generating the response. We apply VOPE to several mainstream LVLMs and hallucination mitigation methods, revealing two key findings: (1) most LVLMs hallucinate heavily during voluntary imagination, and their performance in presence evaluation is notably poor on imagined objects; (2) existing hallucination mitigation methods show limited effect in voluntary imagination tasks, making this an important direction for future research.

Paperid: 2203, https://arxiv.org/pdf/2511.02384.pdf

Abstract:
Large-scale chemical reaction datasets are crucial for AI research in chemistry. However, existing chemical reaction data often exist as images within papers, making them not machine-readable and unusable for training machine learning models. In response to this challenge, we propose the RxnCaption framework for the task of chemical Reaction Diagram Parsing (RxnDP). Our framework reformulates the traditional coordinate prediction driven parsing process into an image captioning problem, which Large Vision-Language Models (LVLMs) handle naturally. We introduce a strategy termed "BBox and Index as Visual Prompt" (BIVP), which uses our state-of-the-art molecular detector, MolYOLO, to pre-draw molecular bounding boxes and indices directly onto the input image. This turns the downstream parsing into a natural-language description problem. Extensive experiments show that the BIVP strategy significantly improves structural extraction quality while simplifying model design. We further construct the RxnCaption-11k dataset, an order of magnitude larger than prior real-world literature benchmarks, with a balanced test subset across four layout archetypes. Experiments demonstrate that RxnCaption-VL achieves state-of-the-art performance on multiple metrics. We believe our method, dataset, and models will advance structured information extraction from chemical literature and catalyze broader AI applications in chemistry. We will release data, models, and code on GitHub.

Paperid: 2204, https://arxiv.org/pdf/2506.11820.pdf

Abstract:
Vision-Language Translation (VLT) is a challenging task that requires accurately recognizing multilingual text embedded in images and translating it into the target language with the support of visual context. While recent Large Vision-Language Models (LVLMs) have demonstrated strong multilingual and visual understanding capabilities, there is a lack of systematic evaluation and understanding of their performance on VLT. In this work, we present a comprehensive study of VLT from three key perspectives: data quality, model architecture, and evaluation metrics. (1) We identify critical limitations in existing datasets, particularly in semantic and cultural fidelity, and introduce AibTrans -- a multilingual, parallel, human-verified dataset with OCR-corrected annotations. (2) We benchmark 11 commercial LVLMs/LLMs and 6 state-of-the-art open-source models across end-to-end and cascaded architectures, revealing their OCR dependency and contrasting generation versus reasoning behaviors. (3) We propose Density-Aware Evaluation to address metric reliability issues under varying contextual complexity, introducing the DA Score as a more robust measure of translation quality. Building upon these findings, we establish a new evaluation benchmark for VLT. Notably, we observe that fine-tuning LVLMs on high-resource language pairs degrades cross-lingual performance, and we propose a balanced multilingual fine-tuning strategy that effectively adapts LVLMs to VLT without sacrificing their generalization ability.

Paperid: 2205, https://arxiv.org/pdf/2503.16566.pdf

Abstract:
The rapid evolution of Large Vision-Language Models (LVLMs) has highlighted the necessity for comprehensive evaluation frameworks that assess these models across diverse dimensions. While existing benchmarks focus on specific aspects such as perceptual abilities, cognitive capabilities, and safety against adversarial attacks, they often lack the breadth and depth required to provide a holistic understanding of LVLMs' strengths and limitations. To address this gap, we introduce REVAL, a comprehensive benchmark designed to evaluate the \textbf{RE}liability and \textbf{VAL}ue of LVLMs. REVAL encompasses over 144K image-text Visual Question Answering (VQA) samples, structured into two primary sections: Reliability, which assesses truthfulness (\eg, perceptual accuracy and hallucination tendencies) and robustness (\eg, resilience to adversarial attacks, typographic attacks, and image corruption), and Values, which evaluates ethical concerns (\eg, bias and moral understanding), safety issues (\eg, toxicity and jailbreak vulnerabilities), and privacy problems (\eg, privacy awareness and privacy leakage). We evaluate 26 models, including mainstream open-source LVLMs and prominent closed-source models like GPT-4o and Gemini-1.5-Pro. Our findings reveal that while current LVLMs excel in perceptual tasks and toxicity avoidance, they exhibit significant vulnerabilities in adversarial scenarios, privacy preservation, and ethical reasoning. These insights underscore critical areas for future improvements, guiding the development of more secure, reliable, and ethically aligned LVLMs. REVAL provides a robust framework for researchers to systematically assess and compare LVLMs, fostering advancements in the field.

Paperid: 2206, https://arxiv.org/pdf/2501.19060.pdf

Abstract:
Vision-language models (VLMs), such as CLIP, have demonstrated exceptional generalization capabilities and can quickly adapt to downstream tasks through prompt fine-tuning. Unfortunately, in classification tasks involving non-training classes, known as open-vocabulary setting, fine-tuned VLMs often overfit to train classes, resulting in a misalignment between confidence scores and actual accuracy on unseen classes, which significantly undermines their reliability in real-world deployments. Existing confidence calibration methods typically require training parameters or analyzing features from the training dataset, restricting their ability to generalize unseen classes without corresponding train data. Moreover, VLM-specific calibration methods rely solely on text features from train classes as calibration indicators, which inherently limits their ability to calibrate train classes. To address these challenges, we propose an effective multimodal calibration method Contrast-Aware Calibration (CAC). Building on the original CLIP's zero-shot adaptability and the conclusion from empirical analysis that poor intra-class and inter-class discriminative ability on unseen classes is the root cause, we calculate calibration weights based on the contrastive difference between the original and fine-tuned CLIP. This method not only adapts to calibrating unseen classes but also overcomes the limitations of previous VLM calibration methods that could not calibrate train classes. In experiments involving 11 datasets with 5 fine-tuning methods, CAC consistently achieved the best calibration effect on both train and unseen classes without sacrificing accuracy and inference speed.

Paperid: 2207, https://arxiv.org/pdf/2501.18271.pdf

Abstract:
Pre-trained Vision-Language Models (VLMs) are becoming increasingly popular across various visual tasks, and several open-sourced VLM variants have been released. However, selecting the best-performing pre-trained VLM for a specific downstream task is challenging since no single VLM can achieve promising performance on all downstream tasks, and evaluating all available VLMs is impossible due to time and data limitations. To address this problem, this paper proposes a novel paradigm to select and reuse VLM for downstream tasks, called Model Label Learning (MLL). The proposal contains three key modules: \emph{model labeling}, which assigns labels to each VLM to describe their specialty and utility; \emph{model selection}, which matches the requirements of the target task with model labels; and \emph{model reuse}, which applies selected VLMs to the target task in an ensemble manner. The proposal is highly computationally efficient and growable since the model labeling process is completed target task independent and the ability could grow with the number of candidate VLMs. We also introduce a new benchmark for evaluating VLM selection methods, including 49 VLMs and 17 target task datasets. Experimental results clearly demonstrate the effectiveness of the proposed method for selecting and reusing VLMs.

Paperid: 2208, https://arxiv.org/pdf/2412.20718.pdf

Abstract:
Recently, large foundation models, including large language models (LLMs) and large vision-language models (LVLMs), have become essential tools in critical fields such as law, finance, and healthcare. As these models increasingly integrate into our daily life, it is necessary to conduct moral evaluation to ensure that their outputs align with human values and remain within moral boundaries. Previous works primarily focus on LLMs, proposing moral datasets and benchmarks limited to text modality. However, given the rapid development of LVLMs, there is still a lack of multimodal moral evaluation methods. To bridge this gap, we introduce M$^3$oralBench, the first MultiModal Moral Benchmark for LVLMs. M$^3$oralBench expands the everyday moral scenarios in Moral Foundations Vignettes (MFVs) and employs the text-to-image diffusion model, SD3.0, to create corresponding scenario images. It conducts moral evaluation across six moral foundations of Moral Foundations Theory (MFT) and encompasses tasks in moral judgement, moral classification, and moral response, providing a comprehensive assessment of model performance in multimodal moral understanding and reasoning. Extensive experiments on 10 popular open-source and closed-source LVLMs demonstrate that M$^3$oralBench is a challenging benchmark, exposing notable moral limitations in current models. Our benchmark is publicly available.

Paperid: 2209, https://arxiv.org/pdf/2412.19496.pdf

Abstract:
Large Vision-Language Models (LVLMs) exhibit impressive potential across various tasks but also face significant privacy risks, limiting their practical applications. Current researches on privacy assessment for LVLMs is limited in scope, with gaps in both assessment dimensions and privacy categories. To bridge this gap, we propose Multi-P$^2$A, a comprehensive benchmark for evaluating the privacy preservation capabilities of LVLMs in terms of privacy awareness and leakage. Privacy awareness measures the model's ability to recognize the privacy sensitivity of input data, while privacy leakage assesses the risk of the model unintentionally disclosing privacy information in its output. We design a range of sub-tasks to thoroughly evaluate the model's privacy protection offered by LVLMs. Multi-P$^2$A covers 26 categories of personal privacy, 15 categories of trade secrets, and 18 categories of state secrets, totaling 31,962 samples. Based on Multi-P$^2$A, we evaluate the privacy preservation capabilities of 21 open-source and 2 closed-source LVLMs. Our results reveal that current LVLMs generally pose a high risk of facilitating privacy breaches, with vulnerabilities varying across personal privacy, trade secret, and state secret.

Paperid: 2210, https://arxiv.org/pdf/2412.12232.pdf

Abstract:
Deep generative models have achieved promising results in image generation, and various generative model hubs, e.g., Hugging Face and Civitai, have been developed that enable model developers to upload models and users to download models. However, these model hubs lack advanced model management and identification mechanisms, resulting in users only searching for models through text matching, download sorting, etc., making it difficult to efficiently find the model that best meets user requirements. In this paper, we propose a novel setting called Generative Model Identification (GMI), which aims to enable the user to identify the most appropriate generative model(s) for the user's requirements from a large number of candidate models efficiently. To our best knowledge, it has not been studied yet. In this paper, we introduce a comprehensive solution consisting of three pivotal modules: a weighted Reduced Kernel Mean Embedding (RKME) framework for capturing the generated image distribution and the relationship between images and prompts, a pre-trained vision-language model aimed at addressing dimensionality challenges, and an image interrogator designed to tackle cross-modality issues. Extensive empirical results demonstrate the proposal is both efficient and effective. For example, users only need to submit a single example image to describe their requirements, and the model platform can achieve an average top-4 identification accuracy of more than 80%.

Paperid: 2211, https://arxiv.org/pdf/2412.03002.pdf

Abstract:
Vision Language Models (VLMs) have exhibited remarkable generalization capabilities, yet their robustness in dynamic real-world scenarios remains largely unexplored. To systematically evaluate VLMs' robustness to real-world 3D variations, we propose AdvDreamer, the first framework capable of generating physically reproducible Adversarial 3D Transformation (Adv-3DT) samples from single-view observations. In AdvDreamer, we integrate three key innovations: Firstly, to characterize real-world 3D variations with limited prior knowledge precisely, we design a zero-shot Monocular Pose Manipulation pipeline built upon generative 3D priors. Secondly, to ensure the visual quality of worst-case Adv-3DT samples, we propose a Naturalness Reward Model that provides continuous naturalness regularization during adversarial optimization, effectively preventing convergence to hallucinated or unnatural elements. Thirdly, to enable systematic evaluation across diverse VLM architectures and visual-language tasks, we introduce the Inverse Semantic Probability loss as the adversarial optimization objective, which solely operates in the fundamental visual-textual alignment space. Based on the captured Adv-3DT samples with high aggressiveness and transferability, we establish MM3DTBench, the first VQA benchmark dataset tailored to evaluate VLM robustness under challenging 3D variations. Extensive evaluations of representative VLMs with varying architectures reveal that real-world 3D variations can pose severe threats to model performance across various tasks.

Paperid: 2212, https://arxiv.org/pdf/2507.13405.pdf

Abstract:
Recently, many benchmarks and datasets have been developed to evaluate Vision-Language Models (VLMs) using visual question answering (VQA) pairs, and models have shown significant accuracy improvements. However, these benchmarks rarely test the model's ability to accurately complete visual entailment, for instance, accepting or refuting a hypothesis based on the image. To address this, we propose COREVQA (Crowd Observations and Reasoning Entailment), a benchmark of 5608 image and synthetically generated true/false statement pairs, with images derived from the CrowdHuman dataset, to provoke visual entailment reasoning on challenging crowded images. Our results show that even the top-performing VLMs achieve accuracy below 80%, with other models performing substantially worse (39.98%-69.95%). This significant performance gap reveals key limitations in VLMs' ability to reason over certain types of image-question pairs in crowded scenes.

Paperid: 2213, https://arxiv.org/pdf/2507.04227.pdf

Abstract:
Mobile GUI agents are designed to autonomously execute diverse device-control tasks by interpreting and interacting with mobile screens. Despite notable advancements, their resilience in real-world scenarios where screen content may be partially manipulated by untrustworthy third parties remains largely unexplored. Owing to their black-box and autonomous nature, these agents are vulnerable to manipulations that could compromise user devices. In this work, we present the first systematic investigation into the vulnerabilities of mobile GUI agents. We introduce a scalable attack simulation framework AgentHazard, which enables flexible and targeted modifications of screen content within existing applications. Leveraging this framework, we develop a comprehensive benchmark suite comprising both a dynamic task execution environment and a static dataset of vision-language-action tuples, totaling over 3,000 attack scenarios. The dynamic environment encompasses 58 reproducible tasks in an emulator with various types of hazardous UI content, while the static dataset is constructed from 210 screenshots collected from 14 popular commercial apps. Importantly, our content modifications are designed to be feasible for unprivileged third parties. We evaluate 7 widely-used mobile GUI agents and 5 common backbone models using our benchmark. Our findings reveal that all examined agents are significantly influenced by misleading third-party content (with an average misleading rate of 28.8% in human-crafted attack scenarios) and that their vulnerabilities are closely linked to the employed perception modalities and backbone LLMs. Furthermore, we assess training-based mitigation strategies, highlighting both the challenges and opportunities for enhancing the robustness of mobile GUI agents. Our code and data will be released at https://agenthazard.github.io.

Paperid: 2214, https://arxiv.org/pdf/2506.21230.pdf

Abstract:
Large Vision-Language Models (LVLMs) show promise for embodied planning tasks but struggle with complex scenarios involving unfamiliar environments and multi-step goals. Current approaches rely on environment-agnostic imitation learning that disconnects instructions from environmental contexts, causing models to struggle with context-sensitive instructions and rely on supplementary cues rather than visual reasoning during long-horizon interactions. In this work, we propose World-Aware Planning Narrative Enhancement (WAP), a framework that infuses LVLMs with comprehensive environmental understanding through four cognitive capabilities (visual appearance modeling, spatial reasoning, functional abstraction, and syntactic grounding) while developing and evaluating models using only raw visual observations through curriculum learning. Evaluations on the EB-ALFRED benchmark demonstrate substantial improvements, with Qwen2.5-VL achieving a 60.7 absolute improvement in task success rates, particularly in commonsense reasoning (+60.0) and long-horizon planning (+70.0). Notably, our enhanced open-source models outperform proprietary systems like GPT-4o and Claude-3.5-Sonnet by a large margin.

Paperid: 2215, https://arxiv.org/pdf/2506.16263.pdf

Abstract:
Vision-Language-Action (VLA) models have emerged as a prominent research area, showcasing significant potential across a variety of applications. However, their performance in endoscopy robotics, particularly endoscopy capsule robots that perform actions within the digestive system, remains unexplored. The integration of VLA models into endoscopy robots allows more intuitive and efficient interactions between human operators and medical devices, improving both diagnostic accuracy and treatment outcomes. In this work, we design CapsDT, a Diffusion Transformer model for capsule robot manipulation in the stomach. By processing interleaved visual inputs, and textual instructions, CapsDT can infer corresponding robotic control signals to facilitate endoscopy tasks. In addition, we developed a capsule endoscopy robot system, a capsule robot controlled by a robotic arm-held magnet, addressing different levels of four endoscopy tasks and creating corresponding capsule robot datasets within the stomach simulator. Comprehensive evaluations on various robotic tasks indicate that CapsDT can serve as a robust vision-language generalist, achieving state-of-the-art performance in various levels of endoscopy tasks while achieving a 26.25% success rate in real-world simulation manipulation.

Paperid: 2216, https://arxiv.org/pdf/2505.21549.pdf

Abstract:
We present Distill CLIP (DCLIP), a fine-tuned variant of the CLIP model that enhances multimodal image-text retrieval while preserving the original model's strong zero-shot classification capabilities. CLIP models are typically constrained by fixed image resolutions and limited context, which can hinder their effectiveness in retrieval tasks that require fine-grained cross-modal understanding. DCLIP addresses these challenges through a meta teacher-student distillation framework, where a cross-modal transformer teacher is fine-tuned to produce enriched embeddings via bidirectional cross-attention between YOLO-extracted image regions and corresponding textual spans. These semantically and spatially aligned global representations guide the training of a lightweight student model using a hybrid loss that combines contrastive learning and cosine similarity objectives. Despite being trained on only ~67,500 samples curated from MSCOCO, Flickr30k, and Conceptual Captions-just a fraction of CLIP's original dataset-DCLIP significantly improves image-text retrieval metrics (Recall@K, MAP), while retaining approximately 94% of CLIP's zero-shot classification performance. These results demonstrate that DCLIP effectively mitigates the trade-off between task specialization and generalization, offering a resource-efficient, domain-adaptive, and detail-sensitive solution for advanced vision-language tasks. Code available at https://anonymous.4open.science/r/DCLIP-B772/README.md.

Paperid: 2217, https://arxiv.org/pdf/2505.15206.pdf

Abstract:
In endoscopic procedures, autonomous tracking of abnormal regions and following circumferential cutting markers can significantly reduce the cognitive burden on endoscopists. However, conventional model-based pipelines are fragile for each component (e.g., detection, motion planning) requires manual tuning and struggles to incorporate high-level endoscopic intent, leading to poor generalization across diverse scenes. Vision-Language-Action (VLA) models, which integrate visual perception, language grounding, and motion planning within an end-to-end framework, offer a promising alternative by semantically adapting to surgeon prompts without manual recalibration. Despite their potential, applying VLA models to robotic endoscopy presents unique challenges due to the complex and dynamic anatomical environments of the gastrointestinal (GI) tract. To address this, we introduce EndoVLA, designed specifically for continuum robots in GI interventions. Given endoscopic images and surgeon-issued tracking prompts, EndoVLA performs three core tasks: (1) polyp tracking, (2) delineation and following of abnormal mucosal regions, and (3) adherence to circular markers during circumferential cutting. To tackle data scarcity and domain shifts, we propose a dual-phase strategy comprising supervised fine-tuning on our EndoVLA-Motion dataset and reinforcement fine-tuning with task-aware rewards. Our approach significantly improves tracking performance in endoscopy and enables zero-shot generalization in diverse scenes and complex sequential tasks.

Paperid: 2218, https://arxiv.org/pdf/2505.14035.pdf

Abstract:
Toxicity detection in multimodal text-image content faces growing challenges, especially with multimodal implicit toxicity, where each modality appears benign on its own but conveys hazard when combined. Multimodal implicit toxicity appears not only as formal statements in social platforms but also prompts that can lead to toxic dialogs from Large Vision-Language Models (LVLMs). Despite the success in unimodal text or image moderation, toxicity detection for multimodal content, particularly the multimodal implicit toxicity, remains underexplored. To fill this gap, we comprehensively build a taxonomy for multimodal implicit toxicity (MMIT) and introduce an MMIT-dataset, comprising 2,100 multimodal statements and prompts across 7 risk categories (31 sub-categories) and 5 typical cross-modal correlation modes. To advance the detection of multimodal implicit toxicity, we build ShieldVLM, a model which identifies implicit toxicity in multimodal statements, prompts and dialogs via deliberative cross-modal reasoning. Experiments show that ShieldVLM outperforms existing strong baselines in detecting both implicit and explicit toxicity. The model and dataset will be publicly available to support future researches. Warning: This paper contains potentially sensitive contents.

Paperid: 2219, https://arxiv.org/pdf/2504.20199.pdf

Abstract:
Vision-language models (VLMs) achieve remarkable success in single-image tasks. However, real-world scenarios often involve intricate multi-image inputs, leading to a notable performance decline as models struggle to disentangle critical information scattered across complex visual features. In this work, we propose Focus-Centric Visual Chain, a novel paradigm that enhances VLMs'perception, comprehension, and reasoning abilities in multi-image scenarios. To facilitate this paradigm, we propose Focus-Centric Data Synthesis, a scalable bottom-up approach for synthesizing high-quality data with elaborate reasoning paths. Through this approach, We construct VISC-150K, a large-scale dataset with reasoning data in the form of Focus-Centric Visual Chain, specifically designed for multi-image tasks. Experimental results on seven multi-image benchmarks demonstrate that our method achieves average performance gains of 3.16% and 2.24% across two distinct model architectures, without compromising the general vision-language capabilities. our study represents a significant step toward more robust and capable vision-language systems that can handle complex visual scenarios.

Paperid: 2220, https://arxiv.org/pdf/2503.23130.pdf

Abstract:
The DeepSeek models have shown exceptional performance in general scene understanding, question-answering (QA), and text generation tasks, owing to their efficient training paradigm and strong reasoning capabilities. In this study, we investigate the dialogue capabilities of the DeepSeek model in robotic surgery scenarios, focusing on tasks such as Single Phrase QA, Visual QA, and Detailed Description. The Single Phrase QA tasks further include sub-tasks such as surgical instrument recognition, action understanding, and spatial position analysis. We conduct extensive evaluations using publicly available datasets, including EndoVis18 and CholecT50, along with their corresponding dialogue data. Our empirical study shows that, compared to existing general-purpose multimodal large language models, DeepSeek-VL2 performs better on complex understanding tasks in surgical scenes. Additionally, although DeepSeek-V3 is purely a language model, we find that when image tokens are directly inputted, the model demonstrates better performance on single-sentence QA tasks. However, overall, the DeepSeek models still fall short of meeting the clinical requirements for understanding surgical scenes. Under general prompts, DeepSeek models lack the ability to effectively analyze global surgical concepts and fail to provide detailed insights into surgical scenarios. Based on our observations, we argue that the DeepSeek models are not ready for vision-language tasks in surgical contexts without fine-tuning on surgery-specific datasets.

Paperid: 2221, https://arxiv.org/pdf/2411.15268.pdf

Abstract:
Despite the recent breakthroughs achieved by Large Vision Language Models (LVLMs) in understanding and responding to complex visual-textual contexts, their inherent hallucination tendencies limit their practical application in real-world scenarios that demand high levels of precision. Existing methods typically either fine-tune the LVLMs using additional data, which incurs extra costs in manual annotation and computational resources or perform comparisons at the decoding stage, which may eliminate useful language priors for reasoning while introducing inference time overhead. Therefore, we propose ICT, a lightweight, training-free method that calculates an intervention direction to shift the model's focus towards different levels of visual information, enhancing its attention to high-level and fine-grained visual details. During the forward pass stage, the intervention is applied to the attention heads that encode the overall image information and the fine-grained object details, effectively mitigating the phenomenon of overly language priors, and thereby alleviating hallucinations. Extensive experiments demonstrate that ICT achieves strong performance with a small amount of data and generalizes well across different datasets and models. Our code will be public.

Paperid: 2222, https://arxiv.org/pdf/2409.14818.pdf

Abstract:
Recently, mobile AI agents based on VLMs have been gaining increasing attention. These works typically utilize VLM as a foundation, fine-tuning it with instruction-based mobile datasets. However, these VLMs are typically pre-trained on general-domain data, which often results in a lack of fundamental capabilities specific to the mobile domain. Therefore, they may struggle to recognize specific UI elements and understand intra-UI fine-grained information. In addition, the current fine-tuning task focuses on interacting with the most relevant element for the given instruction. These fine-tuned VLMs may still ignore the relationships between UI pages, neglect the roles of elements in page transitions and lack inter-UI understanding. To address issues, we propose a VLM called MobileVLM, which includes two additional pre-training stages to enhance both intra- and inter-UI understanding. We defined four UI-based pre-training tasks, enabling the model to better perceive fine-grained elements and capture page transition actions. To address the lack of mobile pre-training data, we built a large Chinese mobile dataset Mobile3M from scratch, which contains 3 million UI pages, and real-world transition actions, forming a directed graph structure. Experimental results show MobileVLM excels on both our test set and public mobile benchmarks, outperforming existing VLMs.

Paperid: 2223, https://arxiv.org/pdf/2512.22288.pdf

Abstract:
Recently, Masked Diffusion Models (MDMs) have shown promising potential across vision, language, and cross-modal generation. However, a notable discrepancy exists between their training and inference procedures. In particular, MDM inference is a multi-step, iterative process governed not only by the model itself but also by various schedules that dictate the token-decoding trajectory (e.g., how many tokens to decode at each step). In contrast, MDMs are typically trained using a simplified, single-step BERT-style objective that masks a subset of tokens and predicts all of them simultaneously. This step-level simplification fundamentally disconnects the training paradigm from the trajectory-level nature of inference, leaving the inference schedules never optimized during training. In this paper, we introduce Co-GRPO, which reformulates MDM generation as a unified Markov Decision Process (MDP) that jointly incorporates both the model and the inference schedule. By applying Group Relative Policy Optimization at the trajectory level, Co-GRPO cooperatively optimizes model parameters and schedule parameters under a shared reward, without requiring costly backpropagation through the multi-step generation process. This holistic optimization aligns training with inference more thoroughly and substantially improves generation quality. Empirical results across four benchmarks-ImageReward, HPS, GenEval, and DPG-Bench-demonstrate the effectiveness of our approach. For more details, please refer to our project page: https://co-grpo.github.io/ .

Paperid: 2224, https://arxiv.org/pdf/2510.27391.pdf

Abstract:
Modality alignment is critical for vision-language models (VLMs) to effectively integrate information across modalities. However, existing methods extract hierarchical features from text while representing each image with a single feature, leading to asymmetric and suboptimal alignment. To address this, we propose Alignment across Trees, a method that constructs and aligns tree-like hierarchical features for both image and text modalities. Specifically, we introduce a semantic-aware visual feature extraction framework that applies a cross-attention mechanism to visual class tokens from intermediate Transformer layers, guided by textual cues to extract visual features with coarse-to-fine semantics. We then embed the feature trees of the two modalities into hyperbolic manifolds with distinct curvatures to effectively model their hierarchical structures. To align across the heterogeneous hyperbolic manifolds with different curvatures, we formulate a KL distance measure between distributions on heterogeneous manifolds, and learn an intermediate manifold for manifold alignment by minimizing the distance. We prove the existence and uniqueness of the optimal intermediate manifold. Experiments on taxonomic open-set classification tasks across multiple image datasets demonstrate that our method consistently outperforms strong baselines under few-shot and cross-domain settings.

Paperid: 2225, https://arxiv.org/pdf/2510.26098.pdf

Abstract:
Large vision language models (VLMs) have advanced graphical user interface (GUI) task automation but still lag behind humans. We hypothesize this gap stems from missing core GUI knowledge, which existing training schemes (such as supervised fine tuning and reinforcement learning) alone cannot fully address. By analyzing common failure patterns in GUI task execution, we distill GUI knowledge into three dimensions: (1) interface perception, knowledge about recognizing widgets and system states; (2) interaction prediction, knowledge about reasoning action state transitions; and (3) instruction understanding, knowledge about planning, verifying, and assessing task completion progress. We further introduce GUI Knowledge Bench, a benchmark with multiple choice and yes/no questions across six platforms (Web, Android, MacOS, Windows, Linux, IOS) and 292 applications. Our evaluation shows that current VLMs identify widget functions but struggle with perceiving system states, predicting actions, and verifying task completion. Experiments on real world GUI tasks further validate the close link between GUI knowledge and task success. By providing a structured framework for assessing GUI knowledge, our work supports the selection of VLMs with greater potential prior to downstream training and provides insights for building more capable GUI agents.

Paperid: 2226, https://arxiv.org/pdf/2510.17777.pdf

Abstract:
Vision Language Models (VLMs) have rapidly advanced in integrating visual and textual reasoning, powering applications across high-resolution image understanding, long-video analysis, and multi-turn conversation. However, their scalability remains limited by the growing number of visual tokens that dominate inference latency. We present SparseVILA, a new paradigm for efficient VLM inference that decouples visual sparsity across the prefilling and decoding stages. SparseVILA distributes sparsity across stages by pruning redundant visual tokens during prefill and retrieving only query-relevant tokens during decoding. This decoupled design matches leading prefill pruning methods while preserving multi-turn fidelity by retaining most of the visual cache so that query-aware tokens can be retrieved at each conversation round. Built on an AWQ-optimized inference pipeline, SparseVILA achieves up to 4.0 times faster prefilling, 2.5 times faster decoding, and an overall 2.6 times end-to-end speedup on long-context video tasks -- while improving accuracy on document-understanding and reasoning tasks. By decoupling query-agnostic pruning and query-aware retrieval, SparseVILA establishes a new direction for efficient multimodal inference, offering a training-free, architecture-agnostic framework for accelerating large VLMs without sacrificing capability.

Paperid: 2227, https://arxiv.org/pdf/2509.23866.pdf

Abstract:
Vision-language model (VLM) based GUI agents show promise for automating complex desktop and mobile tasks, but face significant challenges in applying reinforcement learning (RL): (1) slow multi-turn interactions with GUI environments for policy rollout, and (2) insufficient high-quality agent-environment interactions for policy learning. To address these challenges, we propose DART, a Decoupled Agentic RL Training framework for GUI agents, which coordinates heterogeneous modules in a highly decoupled manner. DART separates the training system into four asynchronous modules: environment cluster, rollout service, data manager, and trainer. This design enables non-blocking communication, asynchronous training, rollout-wise trajectory sampling, and per-worker model synchronization, significantly improving the system efficiency: 1.6*GPU utilization for rollout, 1.9* training throughput, and 5.5* environment utilization. To facilitate effective learning from abundant samples, we introduce an adaptive data curation scheme: (1) pre-collecting successful trajectories for challenging tasks to supplement sparse success in online sampling; (2) dynamically adjusting rollout numbers and trajectory lengths based on task difficulty; (3) training selectively on high-entropy steps to prioritize critical decisions; (4) stabilizing learning via truncated importance sampling for policy mismatch between policy rollout and updating. On the OSWorld benchmark, DART-GUI-7B achieves a 42.13% task success rate, a 14.61% absolute gain over the base model, and 7.34% higher than open-source SOTA. We will fully open-source our training framework, data, and model checkpoints via computer-use-agents.github.io/dart-gui, which we believe is a timely contribution to the open-source community of agentic RL training.

Paperid: 2228, https://arxiv.org/pdf/2509.15704.pdf

Abstract:
Large Vision-Language Models (LVLMs) have recently demonstrated strong multimodal understanding, yet their fine-grained visual perception is often constrained by low input resolutions. A common remedy is to partition high-resolution images into multiple sub-images for separate encoding, but this approach drastically inflates the number of visual tokens and introduces prohibitive inference overhead. To overcome this challenge, we propose Pyramid Token Pruning (PTP), a training-free strategy that hierarchically integrates bottom-up visual saliency at both region and token levels with top-down instruction-guided relevance. Inspired by human visual cognition, PTP selectively preserves more tokens from salient regions while further emphasizing those most relevant to task instructions. Extensive experiments on 13 diverse benchmarks show that PTP substantially reduces computational cost, memory usage, and inference latency, with negligible performance degradation.

Paperid: 2229, https://arxiv.org/pdf/2509.13067.pdf

Abstract:
By cropping high-resolution images into local tiles and encoding them independently, High-Resolution Large Vision-Language Models (HR-LVLMs) have demonstrated remarkable fine-grained visual understanding capabilities. However, this divide-and-conquer paradigm significantly increases the number of visual tokens, resulting in substantial computational and memory overhead. To better understand and address this challenge, we empirically investigate visual token utilization in HR-LVLMs and uncover three key findings: (1) the local tiles have varying importance, jointly determined by visual saliency and task relevance; (2) the CLS token in CLIP-based vision encoders exhibits a two-stage attention pattern across layers, with each stage attending to different types of visual tokens; (3) the visual tokens emphasized at different stages encode information at varying levels of granularity, playing complementary roles within LVLMs. Building on these insights, we propose HERO, a High-resolution visual token early dropping framework that integrates content-adaptive token budget allocation with function-aware token selection. By accurately estimating tile-level importance and selectively retaining visual tokens with complementary roles, HERO achieves superior efficiency-accuracy trade-offs across diverse benchmarks and model scales, all in a training-free manner. This study provides both empirical insights and practical solutions toward efficient inference in HR-LVLMs.

Paperid: 2230, https://arxiv.org/pdf/2508.05167.pdf

Abstract:
Multimodal Large Language Models (MLLMs) are becoming integral to autonomous driving (AD) systems due to their strong vision-language reasoning capabilities. However, MLLMs are vulnerable to adversarial attacks, particularly adversarial patch attacks, which can pose serious threats in real-world scenarios. Existing patch-based attack methods are primarily designed for object detection models and perform poorly when transferred to MLLM-based systems due to the latter's complex architectures and reasoning abilities. To address these limitations, we propose PhysPatch, a physically realizable and transferable adversarial patch framework tailored for MLLM-based AD systems. PhysPatch jointly optimizes patch location, shape, and content to enhance attack effectiveness and real-world applicability. It introduces a semantic-based mask initialization strategy for realistic placement, an SVD-based local alignment loss with patch-guided crop-resize to improve transferability, and a potential field-based mask refinement method. Extensive experiments across open-source, commercial, and reasoning-capable MLLMs demonstrate that PhysPatch significantly outperforms prior methods in steering MLLM-based AD systems toward target-aligned perception and planning outputs. Moreover, PhysPatch consistently places adversarial patches in physically feasible regions of AD scenes, ensuring strong real-world applicability and deployability.

Paperid: 2231, https://arxiv.org/pdf/2506.05198.pdf

Abstract:
Understanding what and how neural networks memorize during training is crucial, both from the perspective of unintentional memorization of potentially sensitive information and from the standpoint of effective knowledge acquisition for real-world, knowledge-intensive tasks. While previous studies primarily investigate memorization within a single modality, such as text memorization in large language models or image memorization in diffusion models, unified multimodal models are becoming increasingly prevalent in practical applications. In this work, we focus on the unique characteristics of cross-modality memorization and conduct a systematic study centered on vision-language models. To facilitate controlled experiments, we first introduce a synthetic persona dataset comprising diverse synthetic person images and textual descriptions. We quantify factual knowledge memorization and cross-modal transferability by training models on a single modality and evaluating their performance in the other. Our results reveal that facts learned in one modality transfer to the other, but a significant gap exists between recalling information in the source and target modalities. Furthermore, we observe that this gap exists across various scenarios, including more capable models, machine unlearning, and the multi-hop case. At the end, we propose a baseline method to mitigate this challenge. We hope our study can inspire future research on developing more robust multimodal learning techniques to enhance cross-modal transferability.

Paperid: 2232, https://arxiv.org/pdf/2506.02555.pdf

Abstract:
Foundation models have achieved transformative success across biomedical domains by enabling holistic understanding of multimodal data. However, their application in surgery remains underexplored. Surgical intelligence presents unique challenges - requiring surgical visual perception, temporal analysis, and reasoning. Existing general-purpose vision-language models fail to address these needs due to insufficient domain-specific supervision and the lack of a large-scale high-quality surgical database. To bridge this gap, we propose SurgVLM, one of the first large vision-language foundation models for surgical intelligence, where this single universal model can tackle versatile surgical tasks. To enable this, we construct a large-scale multimodal surgical database, SurgVLM-DB, comprising over 1.81 million frames with 7.79 million conversations, spanning more than 16 surgical types and 18 anatomical structures. We unify and reorganize 23 public datasets across 10 surgical tasks, followed by standardizing labels and doing hierarchical vision-language alignment to facilitate comprehensive coverage of gradually finer-grained surgical tasks, from visual perception, temporal analysis, to high-level reasoning. Building upon this comprehensive dataset, we propose SurgVLM, which is built upon Qwen2.5-VL, and undergoes instruction tuning to 10+ surgical tasks. We further construct a surgical multimodal benchmark, SurgVLM-Bench, for method evaluation. SurgVLM-Bench consists of 6 popular and widely-used datasets in surgical domain, covering several crucial downstream tasks. Based on SurgVLM-Bench, we evaluate the performance of our SurgVLM (3 SurgVLM variants: SurgVLM-7B, SurgVLM-32B, and SurgVLM-72B), and conduct comprehensive comparisons with 14 mainstream commercial VLMs (e.g., GPT-4o, Gemini 2.0 Flash, Qwen2.5-Max).

Paperid: 2233, https://arxiv.org/pdf/2506.00568.pdf

Abstract:
Computer-Aided Design (CAD) plays a pivotal role in industrial manufacturing. Orthographic projection reasoning underpins the entire CAD workflow, encompassing design, manufacturing, and simulation. However, prevailing deep-learning approaches employ standard 3D reconstruction pipelines as an alternative, which often introduce imprecise dimensions and limit the parametric editability required for CAD workflows. Recently, some researchers adopt vision-language models (VLMs), particularly supervised fine-tuning (SFT), to tackle CAD-related challenges. SFT shows promise but often devolves into pattern memorization, yielding poor out-of-distribution performance on complex reasoning tasks. To address these gaps, we introduce CReFT-CAD, a two-stage fine-tuning paradigm that first employs a curriculum-driven reinforcement learning stage with difficulty-aware rewards to build reasoning ability steadily, and then applies supervised post-tuning to hone instruction following and semantic extraction. Complementing this, we release TriView2CAD, the first large-scale, open-source benchmark for orthographic projection reasoning, comprising 200,000 synthetic and 3,000 real-world orthographic projections with precise dimension annotations and six interoperable data modalities. We benchmark leading VLMs on orthographic projection reasoning and demonstrate that CReFT-CAD substantially improves reasoning accuracy and out-of-distribution generalizability in real-world scenarios, offering valuable insights for advancing CAD reasoning research.

Paperid: 2234, https://arxiv.org/pdf/2505.15436.pdf

Abstract:
Vision language models (VLMs) have achieved impressive performance across a variety of computer vision tasks. However, the multimodal reasoning capability has not been fully explored in existing models. In this paper, we propose a Chain-of-Focus (CoF) method that allows VLMs to perform adaptive focusing and zooming in on key image regions based on obtained visual cues and the given questions, achieving efficient multimodal reasoning. To enable this CoF capability, we present a two-stage training pipeline, including supervised fine-tuning (SFT) and reinforcement learning (RL). In the SFT stage, we construct the MM-CoF dataset, comprising 3K samples derived from a visual agent designed to adaptively identify key regions to solve visual tasks with different image resolutions and questions. We use MM-CoF to fine-tune the Qwen2.5-VL model for cold start. In the RL stage, we leverage the outcome accuracies and formats as rewards to update the Qwen2.5-VL model, enabling further refining the search and reasoning strategy of models without human priors. Our model achieves significant improvements on multiple benchmarks. On the V* benchmark that requires strong visual reasoning capability, our model outperforms existing VLMs by 5% among 8 image resolutions ranging from 224 to 4K, demonstrating the effectiveness of the proposed CoF method and facilitating the more efficient deployment of VLMs in practical applications.

Paperid: 2235, https://arxiv.org/pdf/2504.21561.pdf

Abstract:
Multimodal agents, which integrate a controller e.g., a vision language model) with external tools, have demonstrated remarkable capabilities in tackling complex multimodal tasks. Existing approaches for training these agents, both supervised fine-tuning and reinforcement learning, depend on extensive human-annotated task-answer pairs and tool trajectories. However, for complex multimodal tasks, such annotations are prohibitively expensive or impractical to obtain. In this paper, we propose an iterative tool usage exploration method for multimodal agents without any pre-collected data, namely SPORT, via step-wise preference optimization to refine the trajectories of tool usage. Our method enables multimodal agents to autonomously discover effective tool usage strategies through self-exploration and optimization, eliminating the bottleneck of human annotation. SPORT has four iterative components: task synthesis, step sampling, step verification, and preference tuning. We first synthesize multimodal tasks using language models. Then, we introduce a novel trajectory exploration scheme, where step sampling and step verification are executed alternately to solve synthesized tasks. In step sampling, the agent tries different tools and obtains corresponding results. In step verification, we employ a verifier to provide AI feedback to construct step-wise preference data. The data is subsequently used to update the controller for tool usage through preference tuning, producing a SPORT agent. By interacting with real environments, the SPORT agent gradually evolves into a more refined and capable system. Evaluation in the GTA and GAIA benchmarks shows that the SPORT agent achieves 6.41% and 3.64% improvements, underscoring the generalization and effectiveness introduced by our method. The project page is https://SPORT-Agents.github.io.

Paperid: 2236, https://arxiv.org/pdf/2503.22020.pdf

Abstract:
Vision-language-action models (VLAs) have shown potential in leveraging pretrained vision-language models and diverse robot demonstrations for learning generalizable sensorimotor control. While this paradigm effectively utilizes large-scale data from both robotic and non-robotic sources, current VLAs primarily focus on direct input--output mappings, lacking the intermediate reasoning steps crucial for complex manipulation tasks. As a result, existing VLAs lack temporal planning or reasoning capabilities. In this paper, we introduce a method that incorporates explicit visual chain-of-thought (CoT) reasoning into vision-language-action models (VLAs) by predicting future image frames autoregressively as visual goals before generating a short action sequence to achieve these goals. We introduce CoT-VLA, a state-of-the-art 7B VLA that can understand and generate visual and action tokens. Our experimental results demonstrate that CoT-VLA achieves strong performance, outperforming the state-of-the-art VLA model by 17% in real-world manipulation tasks and 6% in simulation benchmarks. Project website: https://cot-vla.github.io/

Paperid: 2237, https://arxiv.org/pdf/2503.18147.pdf

Abstract:
Computer-Aided Design (CAD) plays a pivotal role in industrial manufacturing, yet 2D Parametric Primitive Analysis (PPA) remains underexplored due to two key challenges: structural constraint reasoning and advanced semantic understanding. To tackle these challenges, we first propose an Efficient Hybrid Parametrization (EHP) for better representing 2D engineering drawings. EHP contains four types of atomic component i.e., point, line, circle, and arc). Additionally, we propose PHT-CAD, a novel 2D PPA framework that harnesses the modality alignment and reasoning capabilities of Vision-Language Models (VLMs) for precise engineering drawing analysis. In PHT-CAD, we introduce four dedicated regression heads to predict corresponding atomic components. To train PHT-CAD, a three-stage training paradigm Progressive Hierarchical Tuning (PHT) is proposed to progressively enhance PHT-CAD's capability to perceive individual primitives, infer structural constraints, and align annotation layers with their corresponding geometric representations. Considering that existing datasets lack complete annotation layers and real-world engineering drawings, we introduce ParaCAD, the first large-scale benchmark that explicitly integrates both the geometric and annotation layers. ParaCAD comprises over 10 million annotated drawings for training and 3,000 real-world industrial drawings with complex topological structures and physical constraints for test. Extensive experiments demonstrate the effectiveness of PHT-CAD and highlight the practical significance of ParaCAD in advancing 2D PPA research.

Paperid: 2238, https://arxiv.org/pdf/2502.19958.pdf

Abstract:
Person re-identification (Re-ID) is a crucial task in computer vision, aiming to recognize individuals across non-overlapping camera views. While recent advanced vision-language models (VLMs) excel in logical reasoning and multi-task generalization, their applications in Re-ID tasks remain limited. They either struggle to perform accurate matching based on identity-relevant features or assist image-dominated branches as auxiliary semantics. In this paper, we propose a novel framework ChatReID, that shifts the focus towards a text-side-dominated retrieval paradigm, enabling flexible and interactive re-identification. To integrate the reasoning abilities of language models into Re-ID pipelines, We first present a large-scale instruction dataset, which contains more than 8 million prompts to promote the model fine-tuning. Next. we introduce a hierarchical progressive tuning strategy, which endows Re-ID ability through three stages of tuning, i.e., from person attribute understanding to fine-grained image retrieval and to multi-modal task reasoning. Extensive experiments across ten popular benchmarks demonstrate that ChatReID outperforms existing methods, achieving state-of-the-art performance in all Re-ID tasks. More experiments demonstrate that ChatReID not only has the ability to recognize fine-grained details but also to integrate them into a coherent reasoning process.

Paperid: 2239, https://arxiv.org/pdf/2501.14276.pdf

Abstract:
As the demand for high-resolution image processing in Large Vision-Language Models (LVLMs) grows, sub-image partitioning has become a popular approach for mitigating visual information loss associated with fixed-resolution processing. However, existing partitioning methods uniformly process sub-images, resulting in suboptimal image understanding. In this work, we reveal that the sub-images with higher semantic relevance to the entire image encapsulate richer visual information for preserving the model's visual understanding ability. Therefore, we propose the Global Semantic-guided Weight Allocator (GSWA) module, which dynamically allocates weights to sub-images based on their relative information density, emulating human visual attention mechanisms. This approach enables the model to focus on more informative regions, overcoming the limitations of uniform treatment. We integrate GSWA into the InternVL2-2B framework to create SleighVL, a lightweight yet high-performing model. Extensive experiments demonstrate that SleighVL outperforms models with comparable parameters and remains competitive with larger models. Our work provides a promising direction for more efficient and contextually aware high-resolution image processing in LVLMs, advancing multimodal system development.

Paperid: 2240, https://arxiv.org/pdf/2501.08443.pdf

Abstract:
Large Vision-Language Models (LVLMs) have achieved remarkable success in a wide range of multimodal tasks by integrating pre-trained vision encoders and large language models. However, current LVLMs primarily rely on visual features extracted from the final layers of the vision encoder, overlooking the complementary information available in shallower layers. While recent approaches have explored the use of multilayer visual features in LVLMs, they tend to be task-agnostic and fail to examine the dependencies of hierarchical visual features on specific tasks. To address these gaps, we systematically investigate the contributions of visual features from different encoder layers using 18 benchmarks spanning 6 task categories. Our findings reveal that multilayer features provide complementary strengths with varying task dependencies, and uniform fusion leads to suboptimal performance. Building on these insights, we propose the instruction-guided vision aggregator, a module that dynamically integrates multi-layer visual features based on textual instructions, without increasing the number of visual tokens. Extensive evaluations demonstrate the superior performance of our method. Additionally, an in-depth analysis of the aggregator's behavior highlights the dominance of mid-to-high-level features in semantic-rich tasks and the critical role of low-level features in fine-grained perception.

Paperid: 2241, https://arxiv.org/pdf/2412.15606.pdf

Abstract:
The advancement of large language models (LLMs) prompts the development of multi-modal agents, which are used as a controller to call external tools, providing a feasible way to solve practical tasks. In this paper, we propose a multi-modal agent tuning method that automatically generates multi-modal tool-usage data and tunes a vision-language model (VLM) as the controller for powerful tool-usage reasoning. To preserve the data quality, we prompt the GPT-4o mini model to generate queries, files, and trajectories, followed by query-file and trajectory verifiers. Based on the data synthesis pipeline, we collect the MM-Traj dataset that contains 20K tasks with trajectories of tool usage. Then, we develop the T3-Agent via \underline{T}rajectory \underline{T}uning on VLMs for \underline{T}ool usage using MM-Traj. Evaluations on the GTA and GAIA benchmarks show that the T3-Agent consistently achieves improvements on two popular VLMs: MiniCPM-V-8.5B and {Qwen2-VL-7B}, which outperforms untrained VLMs by $20\%$, showing the effectiveness of the proposed data synthesis pipeline, leading to high-quality data for tool-usage capabilities.

Paperid: 2242, https://arxiv.org/pdf/2412.09353.pdf

Abstract:
Recent work has empirically shown that Vision-Language Models (VLMs) struggle to fully understand the compositional properties of the human language, usually modeling an image caption as a "bag of words". As a result, they perform poorly on compositional tasks, which require a deeper understanding of the different entities of a sentence (subject, verb, etc.) jointly with their mutual relationships in order to be solved. In this paper, we model the dependency relations among textual and visual tokens using a Causal Graphical Model (CGM), built using a dependency parser, and we train a decoder conditioned by the VLM visual encoder. Differently from standard autoregressive or parallel predictions, our decoder's generative process is partially-ordered following the CGM structure. This structure encourages the decoder to learn only the main causal dependencies in a sentence discarding spurious correlations. Using extensive experiments on five compositional benchmarks, we show that our method significantly outperforms all the state-of-the-art compositional approaches by a large margin, and it also improves over methods trained using much larger datasets.

Paperid: 2243, https://arxiv.org/pdf/2411.18111.pdf

Abstract:
Large Vision-Language Models (LVLMs) that incorporate visual models and Large Language Models (LLMs) have achieved impressive results across various cross-modal understanding and reasoning tasks. In recent years, person re-identification (ReID) has also started to explore cross-modal semantics to improve the accuracy of identity recognition. However, effectively utilizing LVLMs for ReID remains an open challenge. While LVLMs operate under a generative paradigm by predicting the next output word, ReID requires the extraction of discriminative identity features to match pedestrians across cameras. In this paper, we propose LVLM-ReID, a novel framework that harnesses the strengths of LVLMs to promote ReID. Specifically, we employ instructions to guide the LVLM in generating one pedestrian semantic token that encapsulates key appearance semantics from the person image. This token is further refined through our Semantic-Guided Interaction (SGI) module, establishing a reciprocal interaction between the semantic token and visual tokens. Ultimately, the reinforced semantic token serves as the pedestrian identity representation. Our framework integrates the semantic understanding and generation capabilities of LVLMs into end-to-end ReID training, allowing LVLMs to capture rich semantic cues from pedestrian images during both training and inference. Our method achieves competitive results on multiple benchmarks without additional image-text annotations, demonstrating the potential of LVLM-generated semantics to advance person ReID and offering a promising direction for future research.

Paperid: 2244, https://arxiv.org/pdf/2411.12915.pdf

Abstract:
Generalist vision language models (VLMs) have made significant strides in computer vision, but they fall short in specialized fields like healthcare, where expert knowledge is essential. In traditional computer vision tasks, creative or approximate answers may be acceptable, but in healthcare, precision is paramount.Current large multimodal models like Gemini and GPT-4o are insufficient for medical tasks due to their reliance on memorized internet knowledge rather than the nuanced expertise required in healthcare. VLMs are usually trained in three stages: vision pre-training, vision-language pre-training, and instruction fine-tuning (IFT). IFT has been typically applied using a mixture of generic and healthcare data. In contrast, we propose that for medical VLMs, a fourth stage of specialized IFT is necessary, which focuses on medical data and includes information from domain expert models. Domain expert models developed for medical use are crucial because they are specifically trained for certain clinical tasks, e.g. to detect tumors and classify abnormalities through segmentation and classification, which learn fine-grained features of medical data$-$features that are often too intricate for a VLM to capture effectively especially in radiology. This paper introduces a new framework, VILA-M3, for medical VLMs that utilizes domain knowledge via expert models. Through our experiments, we show an improved state-of-the-art (SOTA) performance with an average improvement of ~9% over the prior SOTA model Med-Gemini and ~6% over models trained on the specific tasks. Our approach emphasizes the importance of domain expertise in creating precise, reliable VLMs for medical applications.

Paperid: 2245, https://arxiv.org/pdf/2512.03405.pdf

Abstract:
Understanding visual differences between dynamic scenes requires the comparative perception of compositional, spatial, and temporal changes--a capability that remains underexplored in existing vision-language systems. While prior work on Image Difference Captioning (IDC) has enabled models to describe semantic changes between static images, these approaches fail to capture motion continuity, event evolution, or editing consistency over time. We introduce the ViDiC (Video Difference Captioning) task and its corresponding ViDiC-1K dataset, designed to evaluate the ability of Multimodal Large Language Models (MLLMs) to provide fine-grained descriptions of similarities and differences between video pairs. ViDiC-1K comprises 1,000 curated video pairs annotated with over 4,000 comparative checklist items, covering seven categories: subject, style, background, cinematography, motion, location, and playback techniques. To ensure reliable evaluation, we propose a dual-checklist framework that measures the accuracy of similarity and difference separately, based on the LLM-as-a-Judge protocol. Experiments on nineteen representative multimodal models reveal a significant performance gap in their comparative description and difference perception abilities. We hope ViDiC-1K can be a challenging benchmark that lays a solid foundation for advancing video understanding, edit awareness, and comparative reasoning in multimodal intelligence.

Paperid: 2246, https://arxiv.org/pdf/2511.21272.pdf

Abstract:
With Transformers achieving outstanding performance on individual remote sensing (RS) tasks, we are now approaching the realization of a unified model that excels across multiple tasks through multi-task learning (MTL). Compared to single-task approaches, MTL methods offer improved generalization, enhanced scalability, and greater practical applicability. Recently, vision language models (VLMs) have achieved promising results in RS image understanding, grounding, and ultra-high-resolution (UHR) image reasoning, respectively. Moreover, the unified text-based interface demonstrates significant potential for MTL. Hence, in this work, we present RSCoVLM, a simple yet flexible VLM baseline for RS MTL. Firstly, we create the data curation engine, including data acquisition, offline processing and integrating, as well as online loading and weighting. This data engine effectively addresses complex RS data enviroment and generates flexible vision-language conversations. Furthermore, we propose a unified dynamic-resolution strategy to address the diverse image scales inherent in RS imagery. For UHR images, we introduce the Zoom-in Chain mechanism together with its corresponding dataset, LRS-VQA-Zoom. The strategies are flexible and effectively mitigate the computational burdens. Additionally, we significantly enhance the model's object detection capability and propose a novel evaluation protocol that ensures fair comparison between VLMs and conventional detection models. Extensive experiments demonstrate that RSCoVLM achieves state-of-the-art performance across diverse tasks, outperforming existing RS VLMs and even rivaling specialized expert models. All the training and evaluating tools, model weights, and datasets have been fully open-sourced to support reproducibility. We expect that this baseline will promote further progress toward general-purpose RS models.

Paperid: 2247, https://arxiv.org/pdf/2509.16645.pdf

Abstract:
Vision-Language Models (VLMs), with their strong reasoning and planning capabilities, are widely used in embodied decision-making (EDM) tasks in embodied agents, such as autonomous driving and robotic manipulation. Recent research has increasingly explored adversarial attacks on VLMs to reveal their vulnerabilities. However, these attacks either rely on overly strong assumptions, requiring full knowledge of the victim VLM, which is impractical for attacking VLM-based agents, or exhibit limited effectiveness. The latter stems from disrupting most semantic information in the image, which leads to a misalignment between the perception and the task context defined by system prompts. This inconsistency interrupts the VLM's reasoning process, resulting in invalid outputs that fail to affect interactions in the physical world. To this end, we propose a fine-grained adversarial attack framework, ADVEDM, which modifies the VLM's perception of only a few key objects while preserving the semantics of the remaining regions. This attack effectively reduces conflicts with the task context, making VLMs output valid but incorrect decisions and affecting the actions of agents, thus posing a more substantial safety threat in the physical world. We design two variants of based on this framework, ADVEDM-R and ADVEDM-A, which respectively remove the semantics of a specific object from the image and add the semantics of a new object into the image. The experimental results in both general scenarios and EDM tasks demonstrate fine-grained control and excellent attack performance.

Paperid: 2248, https://arxiv.org/pdf/2506.18883.pdf

Abstract:
This paper presents a computational model for universal video temporal grounding, which accurately localizes temporal moments in videos based on natural language queries (e.g., questions or descriptions). Unlike existing methods that are often limited to specific video domains or durations, we propose UniTime, a robust and universal video grounding model leveraging the strong vision-language understanding capabilities of generative Multi-modal Large Language Models (MLLMs). Our model effectively handles videos of diverse views, genres, and lengths while comprehending complex language queries. The key contributions include: (i) We consider steering strong MLLMs for temporal grounding in videos. To enable precise timestamp outputs, we incorporate temporal information by interleaving timestamp tokens with video tokens. (ii) By training the model to handle videos with different input granularities through adaptive frame scaling, our approach achieves robust temporal grounding for both short and long videos. (iii) Comprehensive experiments show that UniTime outperforms state-of-the-art approaches in both zero-shot and dataset-specific finetuned settings across five public temporal grounding benchmarks. (iv) When employed as a preliminary moment retriever for long-form video question-answering (VideoQA), UniTime significantly improves VideoQA accuracy, highlighting its value for complex video understanding tasks.

Paperid: 2249, https://arxiv.org/pdf/2505.02152.pdf

Abstract:
Vision-Language-Action (VLA) models have shown great promise for generalist robotic manipulation in the physical world. However, existing models are restricted to robot observations and text-only instructions, lacking the flexibility of interleaved multimodal instructions enabled by recent advances in foundation models in the digital world. In this paper, we present Interleave-VLA, the first framework capable of comprehending interleaved image-text instructions and directly generating continuous action sequences in the physical world. It offers a flexible, model-agnostic paradigm that extends state-of-the-art VLA models with minimal modifications and strong zero-shot generalization. A key challenge in realizing Interleave-VLA is the absence of large-scale interleaved embodied datasets. To bridge this gap, we develop an automatic pipeline that converts text-only instructions from real-world datasets in Open X-Embodiment into interleaved image-text instructions, resulting in the first large-scale real-world interleaved embodied dataset with 210k episodes. Through comprehensive evaluation on simulation benchmarks and real-robot experiments, we demonstrate that Interleave-VLA offers significant benefits: 1) it improves out-of-domain generalization to unseen objects by 2-3x compared to state-of-the-art baselines, 2) supports flexible task interfaces, and 3) handles diverse user-provided image instructions in a zero-shot manner, such as hand-drawn sketches. We further analyze the factors behind Interleave-VLA's strong zero-shot performance, showing that the interleaved paradigm effectively leverages heterogeneous datasets and diverse instruction images, including those from the Internet, which demonstrates strong potential for scaling up. Our model and dataset will be open-sourced.

Paperid: 2250, https://arxiv.org/pdf/2412.01720.pdf

Abstract:
With the rapid advancement of multimodal information retrieval, increasingly complex retrieval tasks have emerged. Existing methods predominately rely on task-specific fine-tuning of vision-language models, often those trained with image-text contrastive learning. In this paper, we explore the possibility of re-purposing generative Large Multimodal Models (LMMs) for retrieval. This approach enables unifying all retrieval tasks under the same formulation and, more importantly, allows for extrapolation towards unseen retrieval tasks without additional training. Our contributions can be summarised in the following aspects: (i) We introduce LamRA, a versatile framework designed to empower LMMs with sophisticated retrieval and reranking capabilities. (ii) For retrieval, we adopt a two-stage training strategy comprising language-only pre-training and multimodal instruction tuning to progressively enhance LMM's retrieval performance. (iii) For reranking, we employ joint training for both pointwise and listwise reranking, offering two distinct ways to further boost the retrieval performance. (iv) Extensive experimental results underscore the efficacy of our method in handling more than ten retrieval tasks, demonstrating robust performance in both supervised and zero-shot settings, including scenarios involving previously unseen retrieval tasks.

Paperid: 2251, https://arxiv.org/pdf/2411.11683.pdf

Abstract:
Robotic manipulation in the physical world is increasingly empowered by \textit{large language models} (LLMs) and \textit{vision-language models} (VLMs), leveraging their understanding and perception capabilities. Recently, various attacks against such robotic policies have been proposed, with backdoor attacks drawing considerable attention for their high stealth and strong persistence capabilities. However, existing backdoor efforts are limited to simulators and suffer from physical-world realization. To address this, we propose \textit{TrojanRobot}, a highly stealthy and broadly effective robotic backdoor attack in the physical world. Specifically, we introduce a module-poisoning approach by embedding a backdoor module into the modular robotic policy, enabling backdoor control over the policy's visual perception module thereby backdooring the entire robotic policy. Our vanilla implementation leverages a backdoor-finetuned VLM to serve as the backdoor module. To enhance its generalization in physical environments, we propose a prime implementation, leveraging the LVLM-as-a-backdoor paradigm and developing three types of prime attacks, \ie, \textit{permutation}, \textit{stagnation}, and \textit{intentional} attacks, thus achieving finer-grained backdoors. Extensive experiments on the UR3e manipulator with 18 task instructions using robotic policies based on four VLMs demonstrate the broad effectiveness and physical-world stealth of TrojanRobot. Our attack's video demonstrations are available via a github link https://trojanrobot.github.io.

Paperid: 2252, https://arxiv.org/pdf/2601.16394.pdf

Abstract:
Referring Expression Segmentation (RES) is a core vision-language segmentation task that enables pixel-level understanding of targets via free-form linguistic expressions, supporting critical applications such as human-robot interaction and augmented reality. Despite the progress of Multimodal Large Language Model (MLLM)-based approaches, existing RES methods still suffer from two key limitations: first, the coarse bounding boxes from MLLMs lead to redundant or non-discriminative point prompts; second, the prevalent reliance on textual coordinate reasoning is unreliable, as it fails to distinguish targets from visually similar distractors. To address these issues, we propose \textbf{\model}, a novel RES framework integrating \textbf{E}ntropy-\textbf{B}ased Point \textbf{D}iscovery (\textbf{EBD}) and \textbf{V}ision-\textbf{B}ased \textbf{R}easoning (\textbf{VBR}). Specifically, EBD identifies high-information candidate points by modeling spatial uncertainty within coarse bounding boxes, treating point selection as an information maximization process. VBR verifies point correctness through joint visual-semantic alignment, abandoning text-only coordinate inference for more robust validation. Built on these components, \model implements a coarse-to-fine workflow: bounding box initialization, entropy-guided point discovery, vision-based validation, and mask decoding. Extensive evaluations on four benchmark datasets (RefCOCO, RefCOCO+, RefCOCOg, and ReasonSeg) demonstrate that \model achieves new state-of-the-art performance across all four benchmarks, highlighting its effectiveness in generating accurate and semantically grounded segmentation masks with minimal prompts.

Paperid: 2253, https://arxiv.org/pdf/2601.04404.pdf

Abstract:
Driven by applications in autonomous driving robotics and augmented reality 3D object annotation presents challenges beyond 2D annotation including spatial complexity occlusion and viewpoint inconsistency Existing approaches based on single models often struggle to address these issues effectively We propose Tri MARF a novel framework that integrates tri modal inputs including 2D multi view images textual descriptions and 3D point clouds within a multi agent collaborative architecture to enhance large scale 3D annotation Tri MARF consists of three specialized agents a vision language model agent for generating multi view descriptions an information aggregation agent for selecting optimal descriptions and a gating agent that aligns textual semantics with 3D geometry for refined captioning Extensive experiments on Objaverse LVIS Objaverse XL and ABO demonstrate that Tri MARF substantially outperforms existing methods achieving a CLIPScore of 88 point 7 compared to prior state of the art methods retrieval accuracy of 45 point 2 and 43 point 8 on ViLT R at 5 and a throughput of up to 12000 objects per hour on a single NVIDIA A100 GPU

Paperid: 2254, https://arxiv.org/pdf/2601.01798.pdf

Abstract:
Face verification systems have seen substantial advancements; however, they often lack transparency in their decision-making processes. In this paper, we introduce an innovative Vision-Language Model (VLM) for Face Verification, which not only accurately determines if two face images depict the same individual but also explicitly explains the rationale behind its decisions. Our model is uniquely trained using two complementary explanation styles: (1) concise explanations that summarize the key factors influencing its decision, and (2) comprehensive explanations detailing the specific differences observed between the images. We adapt and enhance a state-of-the-art modeling approach originally designed for audio-based differentiation to suit visual inputs effectively. This cross-modal transfer significantly improves our model's accuracy and interpretability. The proposed VLM integrates sophisticated feature extraction techniques with advanced reasoning capabilities, enabling clear articulation of its verification process. Our approach demonstrates superior performance, surpassing baseline methods and existing models. These findings highlight the immense potential of vision language models in face verification set up, contributing to more transparent, reliable, and explainable face verification systems.

Paperid: 2255, https://arxiv.org/pdf/2512.24426.pdf

Abstract:
Recent reasoning-augmented Vision-Language-Action (VLA) models have improved the interpretability of end-to-end autonomous driving by generating intermediate reasoning traces. Yet these models primarily describe what they perceive and intend to do, rarely questioning whether their planned actions are safe or appropriate. This work introduces Counterfactual VLA (CF-VLA), a self-reflective VLA framework that enables the model to reason about and revise its planned actions before execution. CF-VLA first generates time-segmented meta-actions that summarize driving intent, and then performs counterfactual reasoning conditioned on both the meta-actions and the visual context. This step simulates potential outcomes, identifies unsafe behaviors, and outputs corrected meta-actions that guide the final trajectory generation. To efficiently obtain such self-reflective capabilities, we propose a rollout-filter-label pipeline that mines high-value scenes from a base (non-counterfactual) VLA's rollouts and labels counterfactual reasoning traces for subsequent training rounds. Experiments on large-scale driving datasets show that CF-VLA improves trajectory accuracy by up to 17.6%, enhances safety metrics by 20.5%, and exhibits adaptive thinking: it only enables counterfactual reasoning in challenging scenarios. By transforming reasoning traces from one-shot descriptions to causal self-correction signals, CF-VLA takes a step toward self-reflective autonomous driving agents that learn to think before they act.

Paperid: 2256, https://arxiv.org/pdf/2512.22539.pdf

Abstract:
While Vision-Language-Action models (VLAs) are rapidly advancing towards generalist robot policies, it remains difficult to quantitatively understand their limits and failure modes. To address this, we introduce a comprehensive benchmark called VLA-Arena. We propose a novel structured task design framework to quantify difficulty across three orthogonal axes: (1) Task Structure, (2) Language Command, and (3) Visual Observation. This allows us to systematically design tasks with fine-grained difficulty levels, enabling a precise measurement of model capability frontiers. For Task Structure, VLA-Arena's 170 tasks are grouped into four dimensions: Safety, Distractor, Extrapolation, and Long Horizon. Each task is designed with three difficulty levels (L0-L2), with fine-tuning performed exclusively on L0 to assess general capability. Orthogonal to this, language (W0-W4) and visual (V0-V4) perturbations can be applied to any task to enable a decoupled analysis of robustness. Our extensive evaluation of state-of-the-art VLAs reveals several critical limitations, including a strong tendency toward memorization over generalization, asymmetric robustness, a lack of consideration for safety constraints, and an inability to compose learned skills for long-horizon tasks. To foster research addressing these challenges and ensure reproducibility, we provide the complete VLA-Arena framework, including an end-to-end toolchain from task definition to automated evaluation and the VLA-Arena-S/M/L datasets for fine-tuning. Our benchmark, data, models, and leaderboard are available at https://vla-arena.github.io.

Paperid: 2257, https://arxiv.org/pdf/2512.20561.pdf

Abstract:
Large vision-language models (VLMs) typically process hundreds or thousands of visual tokens per image or video frame, incurring quadratic attention cost and substantial redundancy. Existing token reduction methods often ignore the textual query or rely on deep attention maps, whose instability under aggressive pruning leads to degraded semantic alignment. We propose FlashVLM, a text guided visual token selection framework that dynamically adapts visual inputs to the query. Instead of relying on noisy attention weights, FlashVLM computes an explicit cross modal similarity between projected image tokens and normalized text embeddings in the language model space. This extrinsic relevance is fused with intrinsic visual saliency using log domain weighting and temperature controlled sharpening. In addition, a diversity preserving partition retains a minimal yet representative set of background tokens to maintain global context. Under identical token budgets and evaluation protocols, FlashVLM achieves beyond lossless compression, slightly surpassing the unpruned baseline while pruning up to 77.8 percent of visual tokens on LLaVA 1.5, and maintaining 92.8 percent accuracy even under 94.4 percent compression. Extensive experiments on 14 image and video benchmarks demonstrate that FlashVLM delivers state of the art efficiency performance trade offs while maintaining strong robustness and generalization across mainstream VLMs.

Paperid: 2258, https://arxiv.org/pdf/2512.10226.pdf

Abstract:
Recent Vision-Language-Action (VLA) models for autonomous driving explore inference-time reasoning as a way to improve driving performance and safety in challenging scenarios. Most prior work uses natural language to express chain-of-thought (CoT) reasoning before producing driving actions. However, text may not be the most efficient representation for reasoning. In this work, we present Latent-CoT-Drive (LCDrive): a model that expresses CoT in a latent language that captures possible outcomes of the driving actions being considered. Our approach unifies CoT reasoning and decision making by representing both in an action-aligned latent space. Instead of natural language, the model reasons by interleaving (1) action-proposal tokens, which use the same vocabulary as the model's output actions; and (2) world model tokens, which are grounded in a learned latent world model and express future outcomes of these actions. We cold start latent CoT by supervising the model's action proposals and world model tokens based on ground-truth future rollouts of the scene. We then post-train with closed-loop reinforcement learning to strengthen reasoning capabilities. On a large-scale end-to-end driving benchmark, LCDrive achieves faster inference, better trajectory quality, and larger improvements from interactive reinforcement learning compared to both non-reasoning and text-reasoning baselines.

Paperid: 2259, https://arxiv.org/pdf/2512.08240.pdf

Abstract:
Vision-language models (VLMs) have transformed multimodal reasoning, but feeding hundreds of visual patch tokens into LLMs incurs quadratic computational costs, straining memory and context windows. Traditional approaches face a trade-off: continuous compression dilutes high-level semantics such as object identities, while discrete quantization loses fine-grained details such as textures. We introduce HTC-VLM, a hybrid framework that disentangles semantics and appearance through dual channels, i.e., a continuous pathway for fine-grained details via ViT patches and a discrete pathway for symbolic anchors using MGVQ quantization projected to four tokens. These are fused into a 580-token hybrid sequence and compressed into a single voco token via a disentanglement attention mask and bottleneck, ensuring efficient and grounded representations. HTC-VLM achieves an average performance retention of 87.2 percent across seven benchmarks (GQA, VQAv2, MMBench, MME, POPE, SEED-Bench, ScienceQA-Image), outperforming the leading continuous baseline at 81.0 percent with a 580-to-1 compression ratio. Attention analyses show that the compressed token prioritizes the discrete anchor, validating its semantic guidance. Our work demonstrates that a minimalist hybrid design can resolve the efficiency-fidelity dilemma and advance scalable VLMs.

Paperid: 2260, https://arxiv.org/pdf/2512.08228.pdf

Abstract:
The ability to perform Chain-of-Thought (CoT) reasoning marks a major milestone for multimodal models (MMs), enabling them to solve complex visual reasoning problems. Yet a critical question remains: is such reasoning genuinely grounded in visual evidence and logically coherent? Existing benchmarks emphasize generation but neglect verification, i.e., the capacity to assess whether a reasoning chain is both visually consistent and logically valid. To fill this gap, we introduce MM-CoT, a diagnostic benchmark specifically designed to probe the visual grounding and logical coherence of CoT reasoning in MMs. Instead of generating free-form explanations, models must select the sole event chain that satisfies two orthogonal constraints: (i) visual consistency, ensuring all steps are anchored in observable evidence, and (ii) logical coherence, ensuring causal and commonsense validity. Adversarial distractors are engineered to violate one of these constraints, exposing distinct reasoning failures. We evaluate leading vision-language models on MM-CoT and find that even the most advanced systems struggle, revealing a sharp discrepancy between generative fluency and true reasoning fidelity. MM-CoT shows low correlation with existing benchmarks, confirming that it measures a unique combination of visual grounding and logical reasoning. This benchmark provides a foundation for developing future models that reason not just plausibly, but faithfully and coherently within the visual world.

Paperid: 2261, https://arxiv.org/pdf/2512.04459.pdf

Abstract:
The autonomous driving community is increasingly focused on addressing the challenges posed by out-of-distribution (OOD) driving scenarios. A dominant research trend seeks to enhance end-to-end (E2E) driving systems by integrating vision-language models (VLMs), leveraging their rich world knowledge and reasoning abilities to improve generalization across diverse environments. However, most existing VLMs or vision-language agents (VLAs) are built upon autoregressive (AR) models. In this paper, we observe that existing AR-based VLMs -- limited by causal attention and sequential token generation -- often fail to maintain consistency and controllability between high-level reasoning and low-level planning. In contrast, recent discrete diffusion VLMs equipped with bidirectional attention exhibit superior controllability and reliability through iterative denoising. Building on these observations, we introduce dVLM-AD, a diffusion-based vision-language model that unifies perception, structured reasoning, and low-level planning for end-to-end driving. Evaluated on nuScenes and WOD-E2E, dVLM-AD yields more consistent reasoning-action pairs and achieves planning performance comparable to existing driving VLM/VLA systems despite a modest backbone, outperforming AR-based baselines with a 9 percent improvement in behavior-trajectory consistency and a 6 percent increase in RFS on long-tail WOD-E2E scenarios. These results suggest a controllable and reliable pathway for scalable end-to-end driving.

Paperid: 2262, https://arxiv.org/pdf/2512.00903.pdf

Abstract:
Vision-Language-Action (VLA) models built on pretrained Vision-Language Models (VLMs) show strong potential but are limited in practicality due to their large parameter counts. To mitigate this issue, using a lightweight VLM has been explored, but it compromises spatiotemporal reasoning. Although some methods suggest that incorporating additional 3D inputs can help, they usually rely on large VLMs to fuse 3D and 2D inputs and still lack temporal understanding. Therefore, we propose SwiftVLA, an architecture that enhances a compact model with 4D understanding while preserving design efficiency. Specifically, our approach features a pretrained 4D visual geometry transformer with a temporal cache that extracts 4D features from 2D images. Then, to enhance the VLM's ability to exploit both 2D images and 4D features, we introduce Fusion Tokens, a set of learnable tokens trained with a future prediction objective to generate unified representations for action generation. Finally, we introduce a mask-and-reconstruct strategy that masks 4D inputs to the VLM and trains the VLA to reconstruct them, enabling the VLM to learn effective 4D representations and allowing the 4D branch to be dropped at inference with minimal performance loss. Experiments in real and simulated environments show that SwiftVLA outperforms lightweight baselines and rivals VLAs up to 7 times larger, achieving comparable performance on edge devices while being 18 times faster and reducing memory footprint by 12 times.

Paperid: 2263, https://arxiv.org/pdf/2511.21542.pdf

Abstract:
Vision-Language-Action (VLA) models offer a unified framework for robotic manipulation by integrating visual perception, language understanding, and control generation. Yet existing VLA models still struggle to generalize across diverse tasks, scenes, and camera viewpoints, and often produce coarse or unstable actions. We introduce E0, a continuized discrete diffusion framework that formulates action generation as iterative denoising over quantized action tokens. Compared with continuous diffusion policies, E0 offers two key advantages: (1) discrete action tokens align naturally with the symbolic structure of pretrained VLM/VLA backbones, enabling stronger semantic conditioning; and 2. discrete diffusion matches the true quantized nature of real-world robot control-whose hardware constraints (e.g., encoder resolution, control frequency, actuation latency) inherently discretize continuous signals-and therefore benefits from a Bayes-optimal denoiser that models the correct discrete action distribution, leading to stronger generalization. Compared with discrete autoregressive and mask-based discrete diffusion models, E0 supports a significantly larger and finer-grained action vocabulary and avoids the distributional mismatch introduced by masking-based corruptions-yielding more accurate fine-grained action control. We further introduce a spherical viewpoint perturbation augmentation method to improve robustness to camera shifts without additional data. Experiments on LIBERO, VLABench, and ManiSkill show that E0 achieves state-of-the-art performance across 14 diverse environments, outperforming strong baselines by 10.7% on average. Real-world evaluation on a Franka arm confirms that E0 delivers precise, robust, and transferable manipulation, establishing discrete diffusion as a promising direction for generalizable VLA policy learning.

Paperid: 2264, https://arxiv.org/pdf/2511.19861.pdf

Abstract:
World models are emerging as a foundational paradigm for scalable, data-efficient embodied AI. In this work, we present GigaWorld-0, a unified world model framework designed explicitly as a data engine for Vision-Language-Action (VLA) learning. GigaWorld-0 integrates two synergistic components: GigaWorld-0-Video, which leverages large-scale video generation to produce diverse, texture-rich, and temporally coherent embodied sequences under fine-grained control of appearance, camera viewpoint, and action semantics; and GigaWorld-0-3D, which combines 3D generative modeling, 3D Gaussian Splatting reconstruction, physically differentiable system identification, and executable motion planning to ensure geometric consistency and physical realism. Their joint optimization enables the scalable synthesis of embodied interaction data that is visually compelling, spatially coherent, physically plausible, and instruction-aligned. Training at scale is made feasible through our efficient GigaTrain framework, which exploits FP8-precision and sparse attention to drastically reduce memory and compute requirements. We conduct comprehensive evaluations showing that GigaWorld-0 generates high-quality, diverse, and controllable data across multiple dimensions. Critically, VLA model (e.g., GigaBrain-0) trained on GigaWorld-0-generated data achieve strong real-world performance, significantly improving generalization and task success on physical robots without any real-world interaction during training.

Paperid: 2265, https://arxiv.org/pdf/2511.00088.pdf

Authors:NVIDIA, :, Yan Wang, Wenjie Luo, Junjie Bai, Yulong Cao, Tong Che, Ke Chen, Yuxiao Chen, Jenna Diamond, Yifan Ding, Wenhao Ding, Liang Feng, Greg Heinrich, Jack Huang, Peter Karkus, Boyi Li, Pinyi Li, Tsung-Yi Lin, Dongran Liu, Ming-Yu Liu, Langechuan Liu, Zhijian Liu, Jason Lu, Yunxiang Mao, Pavlo Molchanov, Lindsey Pavao, Zhenghao Peng, Mike Ranzinger, Ed Schmerling, Shida Shen, Yunfei Shi, Sarah Tariq, Ran Tian, Tilman Wekel, Xinshuo Weng, Tianjun Xiao, Eric Yang, Xiaodong Yang, Yurong You, Xiaohui Zeng, Wenyuan Zhang, Boris Ivanovic, Marco Pavone

Abstract:
End-to-end architectures trained via imitation learning have advanced autonomous driving by scaling model size and data, yet performance remains brittle in safety-critical long-tail scenarios where supervision is sparse and causal understanding is limited. To address this, we introduce Alpamayo-R1 (AR1), a vision-language-action model (VLA) that integrates Chain of Causation reasoning with trajectory planning to enhance decision-making in complex driving scenarios. Our approach features three key innovations: (1) the Chain of Causation (CoC) dataset, built through a hybrid auto-labeling and human-in-the-loop pipeline producing decision-grounded, causally linked reasoning traces aligned with driving behaviors; (2) a modular VLA architecture combining Cosmos-Reason, a Vision-Language Model pre-trained for Physical AI applications, with a diffusion-based trajectory decoder that generates dynamically feasible plans in real time; (3) a multi-stage training strategy using supervised fine-tuning to elicit reasoning and reinforcement learning (RL) to optimize reasoning quality via large reasoning model feedback and enforce reasoning-action consistency. Evaluation shows AR1 achieves up to a 12% improvement in planning accuracy on challenging cases compared to a trajectory-only baseline, with a 35% reduction in off-road rate and 25% reduction in close encounter rate in closed-loop simulation. RL post-training improves reasoning quality by 45% as measured by a large reasoning model critic and reasoning-action consistency by 37%. Model scaling from 0.5B to 7B parameters shows consistent improvements. On-vehicle road tests confirm real-time performance (99 ms latency) and successful urban deployment. By bridging interpretable reasoning with precise control, AR1 demonstrates a practical path towards Level 4 autonomous driving. We plan to release AR1 models and a subset of the CoC in a future update.

Paperid: 2266, https://arxiv.org/pdf/2510.22391.pdf

Abstract:
Large Vision-Language Models (VLMs) face an inherent contradiction in image captioning: their powerful single-step generation capabilities often lead to a myopic decision-making process. This makes it difficult to maintain global narrative coherence while capturing rich details, a limitation that is particularly pronounced in tasks that require multi-step and complex scene description. To overcome this fundamental challenge, we redefine image captioning as a goal-oriented hierarchical refinement planning problem, and further propose a novel framework, named Top-Down Semantic Refinement (TDSR), which models the generation process as a Markov Decision Process (MDP). However, planning within the vast state space of a VLM presents a significant computational hurdle. Our core contribution, therefore, is the design of a highly efficient Monte Carlo Tree Search (MCTS) algorithm tailored for VLMs. By incorporating a visual-guided parallel expansion and a lightweight value network, our TDSR reduces the call frequency to the expensive VLM by an order of magnitude without sacrificing planning quality. Furthermore, an adaptive early stopping mechanism dynamically matches computational overhead to the image's complexity. Extensive experiments on multiple benchmarks, including DetailCaps, COMPOSITIONCAP, and POPE, demonstrate that our TDSR, as a plug-and-play module, can significantly enhance the performance of existing VLMs (e.g., LLaVA-1.5, Qwen2.5-VL) by achieving state-of-the-art or highly competitive results in fine-grained description, compositional generalization, and hallucination suppression.

Paperid: 2267, https://arxiv.org/pdf/2510.19430.pdf

Abstract:
Training Vision-Language-Action (VLA) models for generalist robots typically requires large-scale real-world robot data, which is expensive and time-consuming to collect. The inefficiency of physical data collection severely limits the scalability, and generalization capacity of current VLA systems. To address this challenge, we introduce GigaBrain-0, a novel VLA foundation model empowered by world model-generated data (e.g., video generation, real2real transfer, human transfer, view transfer, sim2real transfer data). By leveraging world models to generate diverse data at scale, GigaBrain-0 significantly reduces reliance on real robot data while improving cross-task generalization. Our approach further improves policy robustness through RGBD input modeling and embodied Chain-of-Thought (CoT) supervision, enabling the model to reason about spatial geometry, object states, and long-horizon dependencies during task execution. This leads to substantial gains in real-world performance on dexterous, long-horizon, and mobile manipulation tasks. Extensive experiments demonstrate that GigaBrain-0 achieves superior generalization across variations in appearances (e.g., textures, colors), object placements, and camera viewpoints. Additionally, we present GigaBrain-0-Small, an optimized lightweight variant designed to run efficiently on devices such as the NVIDIA Jetson AGX Orin.

Paperid: 2268, https://arxiv.org/pdf/2510.11978.pdf

Abstract:
Preference-based finetuning of vision--language models (VLMs) is brittle: trivially wrong negatives inject uninformative gradients that destabilize training. We recast alignment as \textbf{learning-dynamics--aware optimization} and introduce \textbf{Cooling-Weighted DPO (CW-DPO)}, a two-stage recipe that explicitly models and exploits the training trajectory. \textbf{Stage 1} performs supervised finetuning with \textbf{gentle negatives}: \textbf{low-weight smoothed supervision} that regularizes the base policy and curbs overconfidence without explicit penalties. \textbf{Stage 2} applies a DPO objective in which the \textbf{negative term is scaled by a cooling weight} computed from the model's \textbf{average token log-probability} on each negative, suppressing uninformative gradients from easy or off-distribution samples while preserving signal from hard negatives. In practice, we emphasize \textbf{on-policy negatives} and allow \textbf{mixed negatives} by blending a controllable fraction of dataset negatives to maintain contrast freshness. Throughout, we instrument training with $Δ\!\log p$ probes on positives and negatives as first-class signals for early stopping, curriculum design, and failure diagnosis. Across diverse VLM tasks, CW-DPO yields \textbf{more stable optimization}, \textbf{better calibration}, and \textbf{higher pairwise win-rates} than SFT-only and vanilla DPO, while \textbf{converging in fewer steps}. Ablations isolate the \textbf{cooling-weight mechanism} as the primary driver of these gains and show complementary benefits from mixing on-policy and dataset negatives. Taken together, our results show that \textbf{smoothing learning dynamics before cooling preferences} is a simple, general principle for robust VLM alignment.

Paperid: 2269, https://arxiv.org/pdf/2510.10117.pdf

Abstract:
Multimodal abductive reasoning--the generation and selection of explanatory hypotheses from partial observations--is a cornerstone of intelligence. Current evaluations of this ability in vision-language models (VLMs) are largely confined to static, single-agent tasks. Inspired by Dixit, we introduce DixitWorld, a comprehensive evaluation suite designed to deconstruct this challenge. DIXITWORLD features two core components: DixitArena, a dynamic, multi-agent environment that evaluates both hypothesis generation (a "storyteller" crafting cryptic clues) and hypothesis selection ("listeners" choosing the target image from decoys) under imperfect information; and DixitBench, a static QA benchmark that isolates the listener's task for efficient, controlled evaluation. Results from DixitArena reveal distinct, role-dependent behaviors: smaller open-source models often excel as creative storytellers, producing imaginative yet less discriminative clues, whereas larger proprietary models demonstrate superior overall performance, particularly as listeners. Performance on DixitBench strongly correlates with listener results in DixitArena, validating it as a reliable proxy for hypothesis selection. Our findings reveal a key trade-off between generative creativity and discriminative understanding in multimodal abductive reasoning, a central challenge for developing more balanced and capable vision-language agents.

Paperid: 2270, https://arxiv.org/pdf/2510.00037.pdf

Abstract:
In Vision-Language-Action (VLA) models, robustness to real-world perturbations is critical for deployment. Existing methods target simple visual disturbances, overlooking the broader multi-modal perturbations that arise in actions, instructions, environments, and observations. Here, we first evaluate the robustness of mainstream VLAs under 17 perturbations across four modalities. We find (1) actions as the most fragile modality, (2) Existing visual-robust VLA do not gain robustness in other modality, and (3) pi0 demonstrates superior robustness with a diffusion-based action head. To build multi-modal robust VLAs, we propose RobustVLA against perturbations in VLA inputs and outputs. For output robustness, we perform offline robust optimization against worst-case action noise that maximizes mismatch in flow matching objective. This can be seen as adversarial training, label smoothing, and outlier penalization. For input robustness, we enforce consistent actions across input variations that preserve task semantics. To account for multiple perturbations, we formulate robustness as a multi-armed bandit problem and apply an upper confidence bound algorithm to automatically identify the most harmful noise. Experiments on LIBERO demonstrate our RobustVLA delivers absolute gains over baselines of 12.6% on the pi0 backbone and 10.4% on the OpenVLA backbone across all 17 perturbations, achieving 50.6x faster inference than existing visual-robust VLAs, and a 10.4% gain under mixed perturbations. Our RobustVLA is particularly effective on real-world FR5 robot with limited demonstrations, showing absolute gains by 65.6% under perturbations of four modalities.

Paperid: 2271, https://arxiv.org/pdf/2509.22407.pdf

Abstract:
Vision-language-action (VLA) models increasingly rely on diverse training data to achieve robust generalization. However, collecting large-scale real-world robot manipulation data across varied object appearances and environmental conditions remains prohibitively time-consuming and expensive. To overcome this bottleneck, we propose Embodied Manipulation Media Adaptation (EMMA), a VLA policy enhancement framework that integrates a generative data engine with an effective training pipeline. We introduce DreamTransfer, a diffusion Transformer-based framework for generating multi-view consistent, geometrically grounded embodied manipulation videos. DreamTransfer enables text-controlled visual editing of robot videos, transforming foreground, background, and lighting conditions without compromising 3D structure or geometrical plausibility. Furthermore, we explore hybrid training with real and generated data, and introduce AdaMix, a hard-sample-aware training strategy that dynamically reweights training batches to focus optimization on perceptually or kinematically challenging samples. Extensive experiments show that videos generated by DreamTransfer significantly outperform prior video generation methods in multi-view consistency, geometric fidelity, and text-conditioning accuracy. Crucially, VLAs trained with generated data enable robots to generalize to unseen object categories and novel visual domains using only demonstrations from a single appearance. In real-world robotic manipulation tasks with zero-shot visual domains, our approach achieves over a 200% relative performance gain compared to training on real data alone, and further improves by 13% with AdaMix, demonstrating its effectiveness in boosting policy generalization.

Paperid: 2272, https://arxiv.org/pdf/2509.22199.pdf

Abstract:
Vision Language Action (VLA) models derive their generalization capability from diverse training data, yet collecting embodied robot interaction data remains prohibitively expensive. In contrast, human demonstration videos are far more scalable and cost-efficient to collect, and recent studies confirm their effectiveness in training VLA models. However, a significant domain gap persists between human videos and robot-executed videos, including unstable camera viewpoints, visual discrepancies between human hands and robotic arms, and differences in motion dynamics. To bridge this gap, we propose MimicDreamer, a framework that turns fast, low-cost human demonstrations into robot-usable supervision by jointly aligning vision, viewpoint, and actions to directly support policy training. For visual alignment, we propose H2R Aligner, a video diffusion model that generates high-fidelity robot demonstration videos by transferring motion from human manipulation footage. For viewpoint stabilization, EgoStabilizer is proposed, which canonicalizes egocentric videos via homography and inpaints occlusions and distortions caused by warping. For action alignment, we map human hand trajectories to the robot frame and apply a constrained inverse kinematics solver to produce feasible, low-jitter joint commands with accurate pose tracking. Empirically, VLA models trained purely on our synthesized human-to-robot videos achieve few-shot execution on real robots. Moreover, scaling training with human data significantly boosts performance compared to models trained solely on real robot data; our approach improves the average success rate by 14.7\% across six representative manipulation tasks.

Paperid: 2273, https://arxiv.org/pdf/2508.11176.pdf

Abstract:
Adapter-based approaches have garnered attention for fine-tuning pre-trained Vision-Language Models (VLMs) on few-shot classification tasks. These methods strive to develop a lightweight module that better aligns visual and (category) textual representations, thereby enhancing performance on downstream few-shot learning tasks. However, existing adapters generally learn/align (category) textual-visual modalities via explicit spatial proximity in the underlying embedding space, which i) fails to capture the inherent one-to-many associations between categories and image samples and ii) struggles to establish accurate associations between the unknown categories and images. To address these issues, inspired by recent works on hyperbolic learning, we develop a novel Latent Hierarchical Adapter (LatHAdapter) for fine-tuning VLMs on downstream few-shot classification tasks. The core of LatHAdapter is to exploit the latent semantic hierarchy of downstream training data and employ it to provide richer, fine-grained guidance for the adapter learning process. Specifically, LatHAdapter first introduces some learnable `attribute' prompts as the bridge to align categories and images. Then, it projects the categories, attribute prompts, and images within each batch in a hyperbolic space, and employs hierarchical regularization to learn the latent semantic hierarchy of them, thereby fully modeling the inherent one-to-many associations among categories, learnable attributes, and image samples. Extensive experiments on four challenging few-shot tasks show that the proposed LatHAdapter consistently outperforms many other fine-tuning approaches, particularly in adapting known classes and generalizing to unknown classes.

Paperid: 2274, https://arxiv.org/pdf/2507.19362.pdf

Abstract:
Large Vision-Language Models (LVLMs) have transformed image captioning, shifting from concise captions to detailed descriptions. We introduce LOTUS, a leaderboard for evaluating detailed captions, addressing three main gaps in existing evaluations: lack of standardized criteria, bias-aware assessments, and user preference considerations. LOTUS comprehensively evaluates various aspects, including caption quality (e.g., alignment, descriptiveness), risks (\eg, hallucination), and societal biases (e.g., gender bias) while enabling preference-oriented evaluations by tailoring criteria to diverse user preferences. Our analysis of recent LVLMs reveals no single model excels across all criteria, while correlations emerge between caption detail and bias risks. Preference-oriented evaluations demonstrate that optimal model selection depends on user priorities.

Paperid: 2275, https://arxiv.org/pdf/2506.17267.pdf

Abstract:
Recent advances in vision-language models (VLMs) have greatly improved cross-modal semantic understanding, yet significant limitations remain in fine-grained discrimination and deep causal reasoning tasks. Existing VLMs often rely on superficial statistical correlations, lacking the ability to capture the underlying causal logic between visual and textual content. To address this, we propose CounterFactual Vision-Language Fine-tuning (CF-VLM), a novel framework that enhances the causal reasoning capabilities of VLMs through the targeted use of counterfactual samples. CF-VLM introduces three complementary training objectives: maintaining foundational cross-modal alignment, reinforcing the uniqueness and stability of factual scene representations against coherent counterfactuals, and sharpening the model's sensitivity to minimal but critical causal edits. Extensive experiments demonstrate that CF-VLM consistently outperforms strong baselines and state-of-the-art methods on compositional reasoning and generalization benchmarks. Furthermore, it shows promise in mitigating visual hallucinations, indicating improved factual consistency. Our CF-VLM provides a robust foundation for deploying VLMs in high-stakes, real-world scenarios requiring reliable reasoning and interpretability.

Paperid: 2276, https://arxiv.org/pdf/2506.10639.pdf

Abstract:
Recent progress in diffusion models has greatly enhanced video generation quality, yet these models still require fine-tuning to improve specific dimensions like instance preservation, motion rationality, composition, and physical plausibility. Existing fine-tuning approaches often rely on human annotations and large-scale computational resources, limiting their practicality. In this work, we propose GigaVideo-1, an efficient fine-tuning framework that advances video generation without additional human supervision. Rather than injecting large volumes of high-quality data from external sources, GigaVideo-1 unlocks the latent potential of pre-trained video diffusion models through automatic feedback. Specifically, we focus on two key aspects of the fine-tuning process: data and optimization. To improve fine-tuning data, we design a prompt-driven data engine that constructs diverse, weakness-oriented training samples. On the optimization side, we introduce a reward-guided training strategy, which adaptively weights samples using feedback from pre-trained vision-language models with a realism constraint. We evaluate GigaVideo-1 on the VBench-2.0 benchmark using Wan2.1 as the baseline across 17 evaluation dimensions. Experiments show that GigaVideo-1 consistently improves performance on almost all the dimensions with an average gain of about 4% using only 4 GPU-hours. Requiring no manual annotations and minimal real data, GigaVideo-1 demonstrates both effectiveness and efficiency. Code, model, and data will be publicly available.

Paperid: 2277, https://arxiv.org/pdf/2505.23399.pdf

Abstract:
We propose GAM-Agent, a game-theoretic multi-agent framework for enhancing vision-language reasoning. Unlike prior single-agent or monolithic models, GAM-Agent formulates the reasoning process as a non-zero-sum game between base agents--each specializing in visual perception subtasks--and a critical agent that verifies logic consistency and factual correctness. Agents communicate via structured claims, evidence, and uncertainty estimates. The framework introduces an uncertainty-aware controller to dynamically adjust agent collaboration, triggering multi-round debates when disagreement or ambiguity is detected. This process yields more robust and interpretable predictions. Experiments on four challenging benchmarks--MMMU, MMBench, MVBench, and V*Bench--demonstrate that GAM-Agent significantly improves performance across various VLM backbones. Notably, GAM-Agent boosts the accuracy of small-to-mid scale models (e.g., Qwen2.5-VL-7B, InternVL3-14B) by 5--6\%, and still enhances strong models like GPT-4o by up to 2--3\%. Our approach is modular, scalable, and generalizable, offering a path toward reliable and explainable multi-agent multimodal reasoning.

Paperid: 2278, https://arxiv.org/pdf/2503.11519.pdf

Abstract:
Current Cross-Modality Generation Models (GMs) demonstrate remarkable capabilities in various generative tasks. Given the ubiquity and information richness of vision modality inputs in real-world scenarios, Cross-Vision tasks, encompassing Vision-Language Perception (VLP) and Image-to-Image (I2I), have attracted significant attention. Large Vision Language Models (LVLMs) and I2I Generation Models (GMs) are employed to handle VLP and I2I tasks, respectively. Previous research indicates that printing typographic words into input images significantly induces LVLMs and I2I GMs to produce disruptive outputs that are semantically aligned with those words. Additionally, visual prompts, as a more sophisticated form of typography, are also revealed to pose security risks to various applications of cross-vision tasks. However, the specific characteristics of the threats posed by visual prompts remain underexplored. In this paper, to comprehensively investigate the performance impact induced by Typographic Visual Prompt Injection (TVPI) in various LVLMs and I2I GMs, we propose the Typographic Visual Prompts Injection Dataset and thoroughly evaluate the TVPI security risks on various open-source and closed-source LVLMs and I2I GMs under visual prompts with different target semantics, deepening the understanding of TVPI threats.

Paperid: 2279, https://arxiv.org/pdf/2503.03480.pdf

Abstract:
Vision-language-action models (VLAs) show potential as generalist robot policies. However, these models pose extreme safety challenges during real-world deployment, including the risk of harm to the environment, the robot itself, and humans. How can safety constraints be explicitly integrated into VLAs? We address this by exploring an integrated safety approach (ISA), systematically modeling safety requirements, then actively eliciting diverse unsafe behaviors, effectively constraining VLA policies via safe reinforcement learning, and rigorously assuring their safety through targeted evaluations. Leveraging the constrained Markov decision process (CMDP) paradigm, ISA optimizes VLAs from a min-max perspective against elicited safety risks. Thus, policies aligned through this comprehensive approach achieve the following key features: (I) effective safety-performance trade-offs, this exploration yields an 83.58% safety improvement compared to the current state-of-the-art method, while also maintaining task performance (+3.85%). (II) strong safety assurance, with the ability to mitigate long-tail risks and handle extreme failure scenarios. (III) robust generalization of learned safety behaviors to various out-of-distribution perturbations. Our data, models and newly proposed benchmark environment are available at https://pku-safevla.github.io.

Paperid: 2280, https://arxiv.org/pdf/2502.20750.pdf

Abstract:
Large vision-language models show tremendous potential in understanding visual information through human languages. However, they are prone to suffer from object hallucination, i.e., the generated image descriptions contain objects that do not exist in the image. In this paper, we reveal that object hallucination can be attributed to overconfidence in irrelevant visual features when soft visual tokens map to the LLM's word embedding space. Specifically, by figuring out the semantic similarity between visual tokens and LLM's word embedding, we observe that the smoothness of similarity distribution strongly correlates with the emergence of object hallucinations. To mitigate hallucinations, we propose using the Variational Information Bottleneck (VIB) to alleviate overconfidence by introducing stochastic noise, facilitating the constraining of irrelevant information. Furthermore, we propose an entropy-based noise-controlling strategy to enable the injected noise to be adaptively constrained regarding the smoothness of the similarity distribution. We adapt the proposed AdaVIB across distinct model architectures. Experimental results demonstrate that the proposed AdaVIB mitigates object hallucinations by effectively alleviating the overconfidence in irrelevant visual features, with consistent improvements on two object hallucination benchmarks.

Paperid: 2281, https://arxiv.org/pdf/2410.07854.pdf

Abstract:
Adapter-based tuning methods have shown significant potential in transferring knowledge from pre-trained Vision-Language Models to the downstream tasks. However, after reviewing existing adapters, we find they generally fail to fully explore the interactions between different modalities in constructing task-specific knowledge. Also, existing works usually only focus on similarity matching between positive text prompts, making it challenging to distinguish the classes with high similar visual contents. To address these issues, in this paper, we propose a novel Heterogeneous Graph Adapter to achieve tuning VLMs for the downstream tasks. To be specific, we first construct a unified heterogeneous graph mode, which contains i) visual nodes, positive text nodes and negative text nodes, and ii) several types of edge connections to comprehensively model the intra-modality, inter-modality and inter-class structure knowledge together. Next, we employ a specific Heterogeneous Graph Neural Network to excavate multi-modality structure knowledge for adapting both visual and textual features for the downstream tasks. Finally, after HeGraphAdapter, we construct both text-based and visual-based classifiers simultaneously to comprehensively enhance the performance of the CLIP model. Experimental results on 11 benchmark datasets demonstrate the effectiveness and benefits of the proposed HeGraphAdapter.

Paperid: 2282, https://arxiv.org/pdf/2601.08325.pdf

Abstract:
Recent advances in robot manipulation have leveraged pre-trained vision-language models (VLMs) and explored integrating 3D spatial signals into these models for effective action prediction, giving rise to the promising vision-language-action (VLA) paradigm. However, most existing approaches overlook the importance of active perception: they typically rely on static, wrist-mounted cameras that provide an end-effector-centric viewpoint. As a result, these models are unable to adaptively select optimal viewpoints or resolutions during task execution, which significantly limits their performance in long-horizon tasks and fine-grained manipulation scenarios. To address these limitations, we propose ActiveVLA, a novel vision-language-action framework that empowers robots with active perception capabilities for high-precision, fine-grained manipulation. ActiveVLA adopts a coarse-to-fine paradigm, dividing the process into two stages: (1) Critical region localization. ActiveVLA projects 3D inputs onto multi-view 2D projections, identifies critical 3D regions, and supports dynamic spatial awareness. (2) Active perception optimization. Drawing on the localized critical regions, ActiveVLA uses an active view selection strategy to choose optimal viewpoints. These viewpoints aim to maximize amodal relevance and diversity while minimizing occlusions. Additionally, ActiveVLA applies a 3D zoom-in to improve resolution in key areas. Together, these steps enable finer-grained active perception for precise manipulation. Extensive experiments demonstrate that ActiveVLA achieves precise 3D manipulation and outperforms state-of-the-art baselines on three simulation benchmarks. Moreover, ActiveVLA transfers seamlessly to real-world scenarios, enabling robots to learn high-precision tasks in complex environments.

Paperid: 2283, https://arxiv.org/pdf/2512.13072.pdf

Abstract:
Multimodal biomedical Vision-Language Models (VLMs) exhibit immense potential in the field of Continual Learning (CL). However, they confront a core dilemma: how to preserve fine-grained intra-modality features while bridging the significant domain gap across different modalities. To address this challenge, we propose a comprehensive framework. Leveraging our 18-million multimodal and comprehensive medical retrieval database derived from PubMed scientific papers, we pioneer the integration of Retrieval-Augmented Generation (RAG) into CL. Specifically, we employ a multi-modal, multi-layer RAG system that provides real-time guidance for model fine-tuning through dynamic, on-demand knowledge retrieval. Building upon this, we introduce a dynamic knowledge distillation framework. This framework precisely resolves the aforementioned core dilemma by dynamically modulating the importance of the parameter space, the granularity of the distilled knowledge, and the data distribution of the reference dataset in accordance with the required level of detail. To thoroughly validate the clinical value of our strategy, we have designed a more rigorous \textbf{M}edical Generalist Task Incremental Learning (MGTIL) benchmark. This benchmark is engineered to simultaneously evaluate the model's capacity for adaptation to significant domain shifts, retention of subtle intra-domain features, and real-time learning of novel and complex medical tasks. Extensive experimental results demonstrate that our proposed method achieves state-of-the-art (SOTA) performance across all metrics. The code is provided in the supplementary materials.

Paperid: 2284, https://arxiv.org/pdf/2512.08282.pdf

Abstract:
Recent advances in Video-to-Audio (V2A) generation have achieved impressive perceptual quality and temporal synchronization, yet most models remain appearance-driven, capturing visual-acoustic correlations without considering the physical factors that shape real-world sounds. We present Physics-Aware Video-to-Audio Synthesis (PAVAS), a method that incorporates physical reasoning into a latent diffusion-based V2A generation through the Physics-Driven Audio Adapter (Phy-Adapter). The adapter receives object-level physical parameters estimated by the Physical Parameter Estimator (PPE), which uses a Vision-Language Model (VLM) to infer the moving-object mass and a segmentation-based dynamic 3D reconstruction module to recover its motion trajectory for velocity computation. These physical cues enable the model to synthesize sounds that reflect underlying physical factors. To assess physical realism, we curate VGG-Impact, a benchmark focusing on object-object interactions, and introduce Audio-Physics Correlation Coefficient (APCC), an evaluation metric that measures consistency between physical and auditory attributes. Comprehensive experiments show that PAVAS produces physically plausible and perceptually coherent audio, outperforming existing V2A models in both quantitative and qualitative evaluations. Visit https://physics-aware-video-to-audio-synthesis.github.io for demo videos.

Paperid: 2285, https://arxiv.org/pdf/2512.00756.pdf

Abstract:
With the advancement of computational resources, Large Vision-Language Models (LVLMs) exhibit impressive Perception and Reasoning (P&R) performance on Graphical User Interface (GUI) tasks. However, although they demonstrate strong P&R capabilities in English GUI scenarios, their performance in multilingual settings has received little attention, which limits their global applications. Moreover, existing studies on GUI tasks lack fine-grained analyses, including widget functions and elements' spatial relationships, which are fundamental for more targeted improvements. To tackle these issues, we propose MPR-GUI-Bench, a Multilingual fine-grained Perception and Reasoning GUI Benchmark to evaluate GUI agents' P&R capabilities. Evaluation results demonstrate that LVLMs exhibit significantly worse P&R performance in non-English languages than in English. To address these gaps, we propose GUI-XLI, a GUI Cross-Lingual Intervention method that applies interventions to the hidden states at P&R capability-related layers to mitigate the gaps between English and other languages, building on previous research showing that the hidden states of different language inputs exhibit significant differences in the latent space. Experimental results indicate that our method improves GUI agents' multilingual P&R capability by 6.5% on average.

Paperid: 2286, https://arxiv.org/pdf/2512.00582.pdf

Abstract:
Satire, a form of artistic expression combining humor with implicit critique, holds significant social value by illuminating societal issues. Despite its cultural and societal significance, satire comprehension, particularly in purely visual forms, remains a challenging task for current vision-language models. This task requires not only detecting satire but also deciphering its nuanced meaning and identifying the implicated entities. Existing models often fail to effectively integrate local entity relationships with global context, leading to misinterpretation, comprehension biases, and hallucinations. To address these limitations, we propose SatireDecoder, a training-free framework designed to enhance satirical image comprehension. Our approach proposes a multi-agent system performing visual cascaded decoupling to decompose images into fine-grained local and global semantic representations. In addition, we introduce a chain-of-thought reasoning strategy guided by uncertainty analysis, which breaks down the complex satire comprehension process into sequential subtasks with minimized uncertainty. Our method significantly improves interpretive accuracy while reducing hallucinations. Experimental results validate that SatireDecoder outperforms existing baselines in comprehending visual satire, offering a promising direction for vision-language reasoning in nuanced, high-level semantic tasks.

Paperid: 2287, https://arxiv.org/pdf/2511.23231.pdf

Abstract:
Large Language Models (LLMs) and Large Vision-Language Models (LVLMs) demonstrate strong reasoning capabilities, yet their performance in English significantly outperforms that in low-resource languages, raising fairness concerns in multilingual applications. Existing approaches either rely on costly multilingual training or employ prompting with external translation tools, both of which are resource-intensive and sensitive to translation quality. To address these limitations, we propose a training-free inference-time method to enhance Multilingual Reasoning capabilities via Representation Engineering (MRRE) without using any additional training data or tools. MRRE sequentially injects two precomputed vectors at specific layers during inference processing: cross-lingual reasoning enhancement vectors, which steer non-English reasoning representations toward English space to unlock multilingual reasoning, and target-language output anchoring vectors, which restore the distribution of the target language to preserve input-output language consistency. Comprehensive experiments across six advanced LLMs and LVLMs on four reasoning benchmarks demonstrate that MRRE consistently enhances non-English reasoning by an average gain of 5.48% and up to 7.54% in low-resource languages (Thai and Swahili), while improving input-output language consistency by 3.78%.

Paperid: 2288, https://arxiv.org/pdf/2511.22850.pdf

Abstract:
Document understanding is a long standing practical task. Vision Language Models (VLMs) have gradually become a primary approach in this domain, demonstrating effective performance on single page tasks. However, their effectiveness diminishes when handling long documents. In such scenarios, clues are often scattered across multiple pages and modalities, and redundancy from lengthy inputs can impair the models judgment. While retrieval augmented generation mitigates this issue by filtering for question relevant content, the retrieved results still contain substantial redundancy. To address these limitations, we propose SLEUTH, a multi agent framework. Concretely, SLEUTH orchestrates a retriever and four collaborative agents in a coarse to fine process. The framework identifies key textual and visual clues within the retrieved pages, filters for salient visual evidence such as tables and charts, and analyzes the query to devise a reasoning strategy. It ultimately synthesizes a distilled, evidence dense multimodal context to generate the final prediction. SLEUTH is model agnostic and scalable. When paired with advanced VLM backbones, it consistently improves performance on multiple long document benchmarks, achieving state of the art results. Ablation studies verify each modules effectiveness and confirm the benefits of our hierarchical refinement paradigm.

Paperid: 2289, https://arxiv.org/pdf/2507.04047.pdf

Abstract:
Embodied scene understanding requires not only comprehending visual-spatial information that has been observed but also determining where to explore next in the 3D physical world. Existing 3D Vision-Language (3D-VL) models primarily focus on grounding objects in static observations from 3D reconstruction, such as meshes and point clouds, but lack the ability to actively perceive and explore their environment. To address this limitation, we introduce \underline{\textbf{M}}ove \underline{\textbf{t}}o \underline{\textbf{U}}nderstand (\textbf{\model}), a unified framework that integrates active perception with \underline{\textbf{3D}} vision-language learning, enabling embodied agents to effectively explore and understand their environment. This is achieved by three key innovations: 1) Online query-based representation learning, enabling direct spatial memory construction from RGB-D frames, eliminating the need for explicit 3D reconstruction. 2) A unified objective for grounding and exploring, which represents unexplored locations as frontier queries and jointly optimizes object grounding and frontier selection. 3) End-to-end trajectory learning that combines \textbf{V}ision-\textbf{L}anguage-\textbf{E}xploration pre-training over a million diverse trajectories collected from both simulated and real-world RGB-D sequences. Extensive evaluations across various embodied navigation and question-answering benchmarks show that MTU3D outperforms state-of-the-art reinforcement learning and modular navigation approaches by 14\%, 23\%, 9\%, and 2\% in success rate on HM3D-OVON, GOAT-Bench, SG3D, and A-EQA, respectively. \model's versatility enables navigation using diverse input modalities, including categories, language descriptions, and reference images. These findings highlight the importance of bridging visual grounding and exploration for embodied intelligence.

Paperid: 2290, https://arxiv.org/pdf/2507.01424.pdf

Abstract:
Recent advancements in vision-language models (VLMs) for common-sense reasoning have led to the development of vision-language-action (VLA) models, enabling robots to perform generalized manipulation. Although existing autoregressive VLA methods design a specific architecture like dual-system to leverage large-scale pretrained knowledge, they tend to capture static information, often neglecting the dynamic aspects vital for embodied tasks. To this end, we propose TriVLA, a unified Vision-Language-Action model with a triple-system architecture for general robot control. The vision-language module (System 2) interprets the environment through vision and language instructions. The dynamics perception module (System 3) inherently produces visual representations that encompass both current static information and predicted future dynamics, thereby providing valuable guidance for policy learning. TriVLA utilizes pre-trained VLM model and fine-tunes pre-trained video foundation model on robot datasets along with internet human manipulation data. The subsequent policy learning module (System 1) generates fluid motor actions in real time. Experimental evaluation demonstrates that TriVLA operates at approximately 36 Hz and surpasses state-of-the-art imitation learning baselines on standard simulation benchmarks as well as challenging real-world manipulation tasks.

Paperid: 2291, https://arxiv.org/pdf/2506.22434.pdf

Abstract:
This work explores enabling Chain-of-Thought (CoT) reasoning to link visual cues across multiple images. A straightforward solution is to adapt rule-based reinforcement learning for Vision-Language Models (VLMs). However, such methods typically rely on manually curated question-answer pairs, which can be particularly challenging when dealing with fine grained visual details and complex logic across images. Inspired by self-supervised visual representation learning, we observe that images contain inherent constraints that can serve as supervision. Based on this insight, we construct image triplets comprising two augmented views of the same image and a third, similar but distinct image. During training, the model is prompted to generate a reasoning process to compare these images (i.e., determine same or different). Then we optimize the model with rule-based reinforcement learning. Due to the high visual similarity and the presence of augmentations, the model must attend to subtle visual changes and perform logical reasoning to succeed. Experiments show that, although trained solely on visual comparison tasks, the learned reasoning ability generalizes effectively to a wide range of questions. Without relying on any human-annotated question-answer pairs, our method achieves significant improvements on multi-image reasoning benchmarks and shows strong performance on general vision tasks.

Paperid: 2292, https://arxiv.org/pdf/2506.11073.pdf

Abstract:
Large Vision-Language Models (LVLMs) have demonstrated impressive multimodal abilities but remain prone to multilingual object hallucination, with a higher likelihood of generating responses inconsistent with the visual input when utilizing queries in non-English languages compared to English. Most existing approaches to address these rely on pretraining or fine-tuning, which are resource-intensive. In this paper, inspired by observing the disparities in cross-modal attention patterns across languages, we propose Cross-Lingual Attention Intervention for Mitigating multilingual object hallucination (CLAIM) in LVLMs, a novel near training-free method by aligning attention patterns. CLAIM first identifies language-specific cross-modal attention heads, then estimates language shift vectors from English to the target language, and finally intervenes in the attention outputs during inference to facilitate cross-lingual visual perception capability alignment. Extensive experiments demonstrate that CLAIM achieves an average improvement of 13.56% (up to 30% in Spanish) on the POPE and 21.75% on the hallucination subsets of the MME benchmark across various languages. Further analysis reveals that multilingual attention divergence is most prominent in intermediate layers, highlighting their critical role in multilingual scenarios.

Paperid: 2293, https://arxiv.org/pdf/2506.09935.pdf

Abstract:
Developing 3D-VL generalists capable of understanding 3D scenes and following natural language instructions to perform a wide range of tasks has been a long-standing goal in the 3D-VL community. Despite recent progress, 3D-VL models still lag behind their 2D counterparts in capability and robustness, falling short of the generalist standard. A key obstacle to developing 3D-VL generalists lies in data scalability, hindered by the lack of an efficient scene representation. We propose LEO-VL, a 3D-VL model built upon condensed feature grid (CFG), an efficient scene representation that bridges 2D perception and 3D spatial structure while significantly reducing token overhead. This efficiency unlocks large-scale training towards 3D-VL generalist, for which we curate over 700k high-quality 3D-VL data spanning four domains of real-world indoor scenes and five tasks such as captioning and dialogue. LEO-VL achieves state-of-the-art performance on a variety of 3D QA benchmarks, including SQA3D, MSQA, and Beacon3D. Ablation studies confirm the efficiency of our representation, the importance of task and scene diversity, and the validity of our data curation principle. Furthermore, we introduce SceneDPO, a novel post-training objective that enhances the robustness of 3D-VL models. We hope our findings contribute to the advancement of scalable and robust 3D-VL generalists.

Paperid: 2294, https://arxiv.org/pdf/2503.23297.pdf

Abstract:
Open-vocabulary 3D visual grounding and reasoning aim to localize objects in a scene based on implicit language descriptions, even when they are occluded. This ability is crucial for tasks such as vision-language navigation and autonomous robotics. However, current methods struggle because they rely heavily on fine-tuning with 3D annotations and mask proposals, which limits their ability to handle diverse semantics and common knowledge required for effective reasoning. In this work, we propose ReasonGrounder, an LVLM-guided framework that uses hierarchical 3D feature Gaussian fields for adaptive grouping based on physical scale, enabling open-vocabulary 3D grounding and reasoning. ReasonGrounder interprets implicit instructions using large vision-language models (LVLM) and localizes occluded objects through 3D Gaussian splatting. By incorporating 2D segmentation masks from the SAM and multi-view CLIP embeddings, ReasonGrounder selects Gaussian groups based on object scale, enabling accurate localization through both explicit and implicit language understanding, even in novel, occluded views. We also contribute ReasoningGD, a new dataset containing over 10K scenes and 2 million annotations for evaluating open-vocabulary 3D grounding and amodal perception under occlusion. Experiments show that ReasonGrounder significantly improves 3D grounding accuracy in real-world scenarios.

Paperid: 2295, https://arxiv.org/pdf/2503.22420.pdf

Abstract:
Existing 3D vision-language (3D-VL) benchmarks fall short in evaluating 3D-VL models, creating a "mist" that obscures rigorous insights into model capabilities and 3D-VL tasks. This mist persists due to three key limitations. First, flawed test data, like ambiguous referential text in the grounding task, can yield incorrect and unreliable test results. Second, oversimplified metrics such as simply averaging accuracy per question answering (QA) pair, cannot reveal true model capability due to their vulnerability to language variations. Third, existing benchmarks isolate the grounding and QA tasks, disregarding the underlying coherence that QA should be based on solid grounding capabilities. To unveil the "mist", we propose Beacon3D, a benchmark for 3D-VL grounding and QA tasks, delivering a perspective shift in the evaluation of 3D-VL understanding. Beacon3D features (i) high-quality test data with precise and natural language, (ii) object-centric evaluation with multiple tests per object to ensure robustness, and (iii) a novel chain-of-analysis paradigm to address language robustness and model performance coherence across grounding and QA. Our evaluation of state-of-the-art 3D-VL models on Beacon3D reveals that (i) object-centric evaluation elicits true model performance and particularly weak generalization in QA; (ii) grounding-QA coherence remains fragile in current 3D-VL models, and (iii) incorporating large language models (LLMs) to 3D-VL models, though as a prevalent practice, hinders grounding capabilities and has yet to elevate QA capabilities. We hope Beacon3D and our comprehensive analysis could benefit the 3D-VL community towards faithful developments.

Paperid: 2296, https://arxiv.org/pdf/2503.12545.pdf

Abstract:
Multimodal large language models (MLLMs) have achieved remarkable success in vision-language tasks, but their reliance on vast, internet-sourced data raises significant privacy and security concerns. Machine unlearning (MU) has emerged as a critical technique to address these issues, enabling the selective removal of targeted information from pre-trained models without costly retraining. However, the evaluation of MU for MLLMs remains inadequate. Existing benchmarks often lack a comprehensive scope, focusing narrowly on entities while overlooking the unlearning of broader visual concepts and the inherent semantic coupling between them. To bridge this gap, we introduce, PEBench, a novel benchmark designed to facilitate a thorough assessment of MU in MLLMs. PEBench features a fictitious dataset of personal entities and corresponding event scenes to evaluate unlearning across these distinct yet entangled concepts. We leverage this benchmark to evaluate five MU methods, revealing their unique strengths and weaknesses. Our findings show that unlearning one concept can unintentionally degrade performance on related concepts within the same image, a challenge we term cross-concept interference. Furthermore, we demonstrate the difficulty of unlearning person and event concepts simultaneously and propose an effective method to mitigate these conflicting objectives. The source code and benchmark are publicly available at https://pebench.github.io.

Paperid: 2297, https://arxiv.org/pdf/2410.16400.pdf

Abstract:
While vision-language models (VLMs) have demonstrated remarkable performance across various tasks combining textual and visual information, they continue to struggle with fine-grained visual perception tasks that require detailed pixel-level analysis. Effectively eliciting comprehensive reasoning from VLMs on such intricate visual elements remains an open challenge. In this paper, we present VipAct, an agent framework that enhances VLMs by integrating multi-agent collaboration and vision expert models, enabling more precise visual understanding and comprehensive reasoning. VipAct consists of an orchestrator agent, which manages task requirement analysis, planning, and coordination, along with specialized agents that handle specific tasks such as image captioning and vision expert models that provide high-precision perceptual information. This multi-agent approach allows VLMs to better perform fine-grained visual perception tasks by synergizing planning, reasoning, and tool use. We evaluate VipAct on benchmarks featuring a diverse set of visual perception tasks, with experimental results demonstrating significant performance improvements over state-of-the-art baselines across all tasks. Furthermore, comprehensive ablation studies reveal the critical role of multi-agent collaboration in eliciting more detailed System-2 reasoning and highlight the importance of image input for task planning. Additionally, our error analysis identifies patterns of VLMs' inherent limitations in visual perception, providing insights into potential future improvements. VipAct offers a flexible and extensible framework, paving the way for more advanced visual perception systems across various real-world applications.

Paperid: 2298, https://arxiv.org/pdf/2410.08584.pdf

Abstract:
The efficiency of large vision-language models (LVLMs) is constrained by the computational bottleneck of the attention mechanism during the prefill phase and the memory bottleneck of fetching the key-value (KV) cache in the decoding phase, particularly in scenarios involving high-resolution images or videos. Visual content often exhibits substantial redundancy, resulting in highly sparse attention maps within LVLMs. This sparsity can be leveraged to accelerate attention computation or compress the KV cache through various approaches. However, most studies focus on addressing only one of these bottlenecks and do not adequately support dynamic adjustment of sparsity concerning distinct layers or tasks. In this paper, we present ZipVL, an efficient inference framework designed for LVLMs through a dynamic ratio allocation strategy of important tokens. This ratio is adaptively determined based on the layer-specific distribution of attention scores, rather than fixed hyper-parameters, thereby improving efficiency for less complex tasks while maintaining high performance for more challenging ones. Then we select important tokens based on their normalized attention scores and perform sparse attention mechanism solely on those important tokens, reducing the latency in the prefill phase. Tokens deemed less important will be discarded to reduce KV cache size, alleviating the memory bottleneck in the decoding phase. Our experiments demonstrate that ZipVL can accelerate the prefill phase by 2.3$\times$ and improve decoding throughput by 2.8$\times$, with a minimal accuracy reduction of only 0.5\% on VQAv2 benchmark over LLaVA-Next-13B model, effectively enhancing the generation efficiency of LVLMs.

Paperid: 2299, https://arxiv.org/pdf/2410.07391.pdf

Abstract:
There is increasing interest in tracking the capabilities of general intelligence foundation models. This study benchmarks leading large language models and vision language models against human performance on the Wechsler Adult Intelligence Scale (WAIS-IV), a comprehensive, population-normed assessment of underlying human cognition and intellectual abilities, with a focus on the domains of VerbalComprehension (VCI), Working Memory (WMI), and Perceptual Reasoning (PRI). Most models demonstrated exceptional capabilities in the storage, retrieval, and manipulation of tokens such as arbitrary sequences of letters and numbers, with performance on the Working Memory Index (WMI) greater or equal to the 99.5th percentile when compared to human population normative ability. Performance on the Verbal Comprehension Index (VCI) which measures retrieval of acquired information, and linguistic understanding about the meaning of words and their relationships to each other, also demonstrated consistent performance at or above the 98th percentile. Despite these broad strengths, we observed consistently poor performance on the Perceptual Reasoning Index (PRI; range 0.1-10th percentile) from multimodal models indicating profound inability to interpret and reason on visual information. Smaller and older model versions consistently performed worse, indicating that training data, parameter count and advances in tuning are resulting in significant advances in cognitive ability.

Paperid: 2300, https://arxiv.org/pdf/2410.02052.pdf

Abstract:
Autonomous agents have demonstrated significant potential in automating complex multistep decision-making tasks. However, even state-of-the-art vision-language models (VLMs), such as GPT-4o, still fall short of human-level performance, particularly in intricate web environments and long-horizon tasks. To address these limitations, we present ExACT, an approach to combine test-time search and self-learning to build o1-like models for agentic applications. We first introduce Reflective Monte Carlo Tree Search (R-MCTS), a novel test time algorithm designed to enhance AI agents' ability to explore decision space on the fly. R-MCTS extends traditional MCTS by 1) incorporating contrastive reflection, allowing agents to learn from past interactions and dynamically improve their search efficiency; and 2) using multi-agent debate for reliable state evaluation. Next, we introduce Exploratory Learning, a novel learning strategy to teach agents to search at inference time without relying on any external search algorithms. On the challenging VisualWebArena benchmark, our GPT-4o based R-MCTS agent achieves a 6% to 30% relative improvement across various tasks compared to the previous state-of-the-art. Additionally, we show that the knowledge and experience gained from test-time search can be effectively transferred back to GPT-4o via fine-tuning. After Exploratory Learning, GPT-4o 1) demonstrates the ability to explore the environment, evaluate a state, and backtrack to viable ones when it detects that the current state cannot lead to success, and 2) matches 87% of R-MCTS's performance while using significantly less compute. Notably, our work demonstrates the compute scaling properties in both training - data collection with R-MCTS - and testing time. These results suggest a promising research direction to enhance VLMs' capabilities for agentic applications via test-time search and self-learning.

Paperid: 2301, https://arxiv.org/pdf/2409.02389.pdf

Abstract:
Situation awareness is essential for understanding and reasoning about 3D scenes in embodied AI agents. However, existing datasets and benchmarks for situated understanding are limited in data modality, diversity, scale, and task scope. To address these limitations, we propose Multi-modal Situated Question Answering (MSQA), a large-scale multi-modal situated reasoning dataset, scalably collected leveraging 3D scene graphs and vision-language models (VLMs) across a diverse range of real-world 3D scenes. MSQA includes 251K situated question-answering pairs across 9 distinct question categories, covering complex scenarios within 3D scenes. We introduce a novel interleaved multi-modal input setting in our benchmark to provide text, image, and point cloud for situation and question description, resolving ambiguity in previous single-modality convention (e.g., text). Additionally, we devise the Multi-modal Situated Next-step Navigation (MSNN) benchmark to evaluate models' situated reasoning for navigation. Comprehensive evaluations on MSQA and MSNN highlight the limitations of existing vision-language models and underscore the importance of handling multi-modal interleaved inputs and situation modeling. Experiments on data scaling and cross-domain transfer further demonstrate the efficacy of leveraging MSQA as a pre-training dataset for developing more powerful situated reasoning models.

Paperid: 2302, https://arxiv.org/pdf/2510.23968.pdf

Abstract:
Vision-language models (VLMs) have shown strong promise for medical image analysis, but most remain opaque, offering predictions without the transparent, stepwise reasoning clinicians rely on. We present a framework that brings chain-of-thought (CoT) reasoning to chest X-ray interpretation. Inspired by reasoning-first training paradigms, our approach is designed to learn how experts reason, not just what they conclude, by aligning intermediate steps with observable image evidence and radiology workflow. Beyond accuracy, the explicit reasoning traces support clinical auditability: they reveal why a conclusion was reached, which alternatives were considered, and where uncertainty remains, enabling quality assurance, error analysis, and safer human-AI collaboration. Our model couples high-fidelity visual encoding with a two-stage training recipe: a reasoning-style supervised fine-tuning (SFT) followed by reinforcement learning (RL) that uses verifiable rewards over a list of X-ray abnormalities. The model outputs reasoning that mirrors radiologists systematic thought process, uncertainty, and differential diagnosis. In out-of-distribution evaluation, the approach achieves competitive multi-label classification while improving interpretability. In a reader study with expert radiologists, full reasoning traces increased confidence, supported error auditing, and reduced time to finalize reports. We release code and the model NV-Reason-CXR-3B to support community progress toward trustworthy, explainable AI in chest radiography and other medical imaging tasks where reasoning quality is as critical as prediction quality.

Paperid: 2303, https://arxiv.org/pdf/2510.10154.pdf

Abstract:
The dominant paradigm for training Large Vision-Language Models (LVLMs) in navigation relies on imitating expert trajectories. This approach reduces the complex navigation task to a sequence-to-sequence replication of a single correct path, fundamentally limiting the agent's ability to explore and generalize. In this work, we argue for and introduce a new paradigm: a shift from Path Imitation to Decision Understanding. The goal of this paradigm is to build agents that do not just follow, but truly understand how to navigate. We materialize this through two core contributions: first, we introduce Compass-Data-22k, a novel 22k-trajectory dataset.Its Reinforcement Fine-Tuning (RFT) subset provides a panoramic view of the decision landscape by annotating all feasible actions with A* geodesic distances. Second, we design a novel gap-aware hybrid reward function that dynamically adapts its feedback to decision certainty, shifting between decisive signals for optimal actions and nuanced scores to encourage exploration. Integrated into an SFT-then-RFT recipe, our CompassNav agent is trained not to memorize static routes, but to develop an internal ``compass'' that constantly intuits the direction to the goal by evaluating the relative quality of all possible moves. This approach enables our 7B agent to set a new state-of-the-art on Goal navigation benchmarks, outperforming even larger proprietary models, and achieve robust real-world goal navigation on a physical robot.

Paperid: 2304, https://arxiv.org/pdf/2507.11003.pdf

Abstract:
With the advent of vision-language models (e.g., CLIP) in zero- and few-shot settings, CLIP has been widely applied to zero-shot anomaly detection (ZSAD) in recent research, where the rare classes are essential and expected in many applications. This study introduces \textbf{FiSeCLIP} for ZSAD with training-free \textbf{CLIP}, combining the feature matching with the cross-modal alignment. Testing with the entire dataset is impractical, while batch-based testing better aligns with real industrial needs, and images within a batch can serve as mutual reference points. Accordingly, FiSeCLIP utilizes other images in the same batch as reference information for the current image. However, the lack of labels for these references can introduce ambiguity, we apply text information to \textbf{fi}lter out noisy features. In addition, we further explore CLIP's inherent potential to restore its local \textbf{se}mantic correlation, adapting it for fine-grained anomaly detection tasks to enable a more accurate filtering process. Our approach exhibits superior performance for both anomaly classification and segmentation on anomaly detection benchmarks, building a stronger baseline for the direction, e.g., on MVTec-AD, FiSeCLIP outperforms the SOTA AdaCLIP by +4.6\%$\uparrow$/+5.7\%$\uparrow$ in segmentation metrics AU-ROC/$F_1$-max.

Paperid: 2305, https://arxiv.org/pdf/2505.21969.pdf

Abstract:
Adaptive navigation in unfamiliar environments is crucial for household service robots but remains challenging due to the need for both low-level path planning and high-level scene understanding. While recent vision-language model (VLM) based zero-shot approaches reduce dependence on prior maps and scene-specific training data, they face significant limitations: spatiotemporal discontinuity from discrete observations, unstructured memory representations, and insufficient task understanding leading to navigation failures. We propose DORAEMON (Decentralized Ontology-aware Reliable Agent with Enhanced Memory Oriented Navigation), a novel cognitive-inspired framework consisting of Ventral and Dorsal Streams that mimics human navigation capabilities. The Dorsal Stream implements the Hierarchical Semantic-Spatial Fusion and Topology Map to handle spatiotemporal discontinuities, while the Ventral Stream combines RAG-VLM and Policy-VLM to improve decision-making. Our approach also develops Nav-Ensurance to ensure navigation safety and efficiency. We evaluate DORAEMON on the HM3D, MP3D, and GOAT datasets, where it achieves state-of-the-art performance on both success rate (SR) and success weighted by path length (SPL) metrics, significantly outperforming existing methods. We also introduce a new evaluation metric (AORI) to assess navigation intelligence better. Comprehensive experiments demonstrate DORAEMON's effectiveness in zero-shot autonomous navigation without requiring prior map building or pre-training.

Paperid: 2306, https://arxiv.org/pdf/2505.12644.pdf

Abstract:
In surrogate ensemble attacks, using more surrogate models yields higher transferability but lower resource efficiency. This practical trade-off between transferability and efficiency has largely limited existing attacks despite many pre-trained models are easily accessible online. In this paper, we argue that such a trade-off is caused by an unnecessary common assumption, i.e., all models should be identical across iterations. By lifting this assumption, we can use as many surrogates as we want to unleash transferability without sacrificing efficiency. Concretely, we propose Selective Ensemble Attack (SEA), which dynamically selects diverse models (from easily accessible pre-trained models) across iterations based on our new interpretation of decoupling within-iteration and cross-iteration model diversity.In this way, the number of within-iteration models is fixed for maintaining efficiency, while only cross-iteration model diversity is increased for higher transferability. Experiments on ImageNet demonstrate the superiority of SEA in various scenarios. For example, when dynamically selecting 4 from 20 accessible models, SEA yields 8.5% higher transferability than existing attacks under the same efficiency. The superiority of SEA also generalizes to real-world systems, such as commercial vision APIs and large vision-language models. Overall, SEA opens up the possibility of adaptively balancing transferability and efficiency according to specific resource requirements.

Paperid: 2307, https://arxiv.org/pdf/2503.09149.pdf

Abstract:
Efficient long-video understanding~(LVU) remains a challenging task in computer vision. Current long-context vision-language models~(LVLMs) suffer from information loss due to compression and brute-force downsampling. While retrieval-augmented generation (RAG) methods mitigate this issue, their applicability is limited due to explicit query dependency. To overcome this challenge, we introduce a novel memory-enhanced RAG-based approach called MemVid, which is inspired by the cognitive memory of human beings. Our approach operates in four basic steps: 1) memorizing holistic video information, 2) reasoning about the task's information needs based on memory, 3) retrieving critical moments based on the information needs, and 4) focusing on the retrieved moments to produce the final answer. To enhance the system's memory-grounded reasoning capabilities while achieving optimal end-to-end performance, we propose a curriculum learning strategy. This approach begins with supervised learning on well-annotated reasoning results, then progressively explores and reinforces more plausible reasoning outcomes through reinforcement learning. We perform extensive evaluations on popular LVU benchmarks, including MLVU, VideoMME and LVBench. In our experiments, MemVid demonstrates superior efficiency and effectiveness compared to both LVLMs and RAG methods.

Paperid: 2308, https://arxiv.org/pdf/2412.04453.pdf

Abstract:
This paper proposes to solve the problem of Vision-and-Language Navigation with legged robots, which not only provides a flexible way for humans to command but also allows the robot to navigate through more challenging and cluttered scenes. However, it is non-trivial to translate human language instructions all the way to low-level leg joint actions. We propose NaVILA, a 2-level framework that unifies a Vision-Language-Action model (VLA) with locomotion skills. Instead of directly predicting low-level actions from VLA, NaVILA first generates mid-level actions with spatial information in the form of language, (e.g., "moving forward 75cm"), which serves as an input for a visual locomotion RL policy for execution. NaVILA substantially improves previous approaches on existing benchmarks. The same advantages are demonstrated in our newly developed benchmarks with IsaacLab, featuring more realistic scenes, low-level controls, and real-world robot experiments. We show more results at https://navila-bot.github.io/

Paperid: 2309, https://arxiv.org/pdf/2410.12183.pdf

Abstract:
Vision-language foundation models (such as CLIP) have recently shown their power in transfer learning, owing to large-scale image-text pre-training. However, target domain data in the downstream tasks can be highly different from the pre-training phase, which makes it hard for such a single model to generalize well. Alternatively, there exists a wide range of expert models that contain diversified vision and/or language knowledge pre-trained on different modalities, tasks, networks, and datasets. Unfortunately, these models are "isolated agents" with heterogeneous structures, and how to integrate their knowledge for generalizing CLIP-like models has not been fully explored. To bridge this gap, we propose a general and concise TransAgent framework, which transports the knowledge of the isolated agents in a unified manner, and effectively guides CLIP to generalize with multi-source knowledge distillation. With such a distinct framework, we flexibly collaborate with 11 heterogeneous agents to empower vision-language foundation models, without further cost in the inference phase. Finally, our TransAgent achieves state-of-the-art performance on 11 visual recognition datasets. Under the same low-shot setting, it outperforms the popular CoOp with around 10% on average, and 20% on EuroSAT which contains large domain shifts.

Paperid: 2310, https://arxiv.org/pdf/2409.05381.pdf

Abstract:
Image Quality Assessment (IQA) remains an unresolved challenge in computer vision due to complex distortions, diverse image content, and limited data availability. Existing Blind IQA (BIQA) methods largely rely on extensive human annotations, which are labor-intensive and costly due to the demanding nature of creating IQA datasets. To reduce this dependency, we propose the Gradient-Regulated Meta-Prompt IQA Framework (GRMP-IQA), designed to efficiently adapt the visual-language pre-trained model, CLIP, to IQA tasks, achieving high accuracy even with limited data. GRMP-IQA consists of two core modules: (i) Meta-Prompt Pre-training Module and (ii) Quality-Aware Gradient Regularization. The Meta Prompt Pre-training Module leverages a meta-learning paradigm to pre-train soft prompts with shared meta-knowledge across different distortions, enabling rapid adaptation to various IQA tasks. On the other hand, the Quality-Aware Gradient Regularization is designed to adjust the update gradients during fine-tuning, focusing the model's attention on quality-relevant features and preventing overfitting to semantic information. Extensive experiments on standard BIQA datasets demonstrate the superior performance to the state-of-the-art BIQA methods under limited data setting. Notably, utilizing just 20% of the training data, GRMP-IQA is competitive with most existing fully supervised BIQA approaches.

Paperid: 2311, https://arxiv.org/pdf/2601.03905.pdf

Abstract:
Agents built on vision-language models increasingly face tasks that demand anticipating future states rather than relying on short-horizon reasoning. Generative world models offer a promising remedy: agents could use them as external simulators to foresee outcomes before acting. This paper empirically examines whether current agents can leverage such world models as tools to enhance their cognition. Across diverse agentic and visual question answering tasks, we observe that some agents rarely invoke simulation (fewer than 1%), frequently misuse predicted rollouts (approximately 15%), and often exhibit inconsistent or even degraded performance (up to 5%) when simulation is available or enforced. Attribution analysis further indicates that the primary bottleneck lies in the agents' capacity to decide when to simulate, how to interpret predicted outcomes, and how to integrate foresight into downstream reasoning. These findings underscore the need for mechanisms that foster calibrated, strategic interaction with world models, paving the way toward more reliable anticipatory cognition in future agent systems.

Paperid: 2312, https://arxiv.org/pdf/2512.14068.pdf

Abstract:
Block-wise discrete diffusion offers an attractive balance between parallel generation and causal dependency modeling, making it a promising backbone for vision-language modeling. However, its practical adoption has been limited by high training cost, slow convergence, and instability, which have so far kept it behind strong autoregressive (AR) baselines. We present \textbf{SDAR-VL}, the first systematic application of block-wise discrete diffusion to large-scale vision-language understanding (VLU), together with an \emph{integrated framework for efficient and stable training}. This framework unifies three components: (1) \textbf{Asynchronous Block-wise Noise Scheduling} to diversify supervision within each batch; (2) \textbf{Effective Mask Ratio Scaling} for unbiased loss normalization under stochastic masking; and (3) a \textbf{Progressive Beta Noise Curriculum} that increases effective mask coverage while preserving corruption diversity. Experiments on 21 single-image, multi-image, and video benchmarks show that SDAR-VL consistently improves \emph{training efficiency}, \emph{convergence stability}, and \emph{task performance} over conventional block diffusion. On this evaluation suite, SDAR-VL sets a new state of the art among diffusion-based vision-language models and, under matched settings, matches or surpasses strong AR baselines such as LLaVA-OneVision as well as the global diffusion baseline LLaDA-V, establishing block-wise diffusion as a practical backbone for VLU.

Paperid: 2313, https://arxiv.org/pdf/2511.23311.pdf

Abstract:
Large-scale Vision Language Models (LVLMs) exhibit advanced capabilities in tasks that require visual information, including object detection. These capabilities have promising applications in various industrial domains, such as autonomous driving. For example, LVLMs can generate safety-oriented descriptions of videos captured by road-facing cameras. However, ensuring comprehensive safety requires monitoring driver-facing views as well to detect risky events, such as the use of mobiles while driving. Thus, the ability to process synchronized inputs is necessary from both driver-facing and road-facing cameras. In this study, we develop models and investigate the capabilities of LVLMs by constructing a dataset and evaluating their performance on this dataset. Our experimental results demonstrate that while pre-trained LVLMs have limited effectiveness, fine-tuned LVLMs can generate accurate and safety-aware driving instructions. Nonetheless, several challenges remain, particularly in detecting subtle or complex events in the video. Our findings and error analysis provide valuable insights that can contribute to the improvement of LVLM-based systems in this domain.

Paperid: 2314, https://arxiv.org/pdf/2511.11421.pdf

Abstract:
Class-Incremental Learning (CIL) aims to continually learn new categories without forgetting previously acquired knowledge. Vision-language models such as CLIP offer strong transferable representations via multi-modal supervision, making them promising for CIL. However, applying CLIP to CIL poses two major challenges: (1) adapting to downstream tasks often requires additional learnable modules, increasing model complexity and susceptibility to forgetting; and (2) while multi-modal representations offer complementary strengths, existing methods have yet to fully realize their potential in effectively integrating visual and textual modalities. To address these issues, we propose BOFA (Bridge-layer Orthogonal Fusion for Adaptation), a novel framework for CIL. BOFA confines all model adaptation exclusively to CLIP's existing cross-modal bridge-layer, thereby adding no extra parameters or inference cost. To prevent forgetting within this layer, it leverages Orthogonal Low-Rank Fusion, a mechanism that constrains parameter updates to a low-rank ``safe subspace" mathematically constructed to be orthogonal to past task features. This ensures stable knowledge accumulation without data replay. Furthermore, BOFA employs a cross-modal hybrid prototype that synergizes stable textual prototypes with visual counterparts derived from our stably adapted bridge-layer, enhancing classification performance. Extensive experiments on standard benchmarks show that BOFA achieves superior accuracy and efficiency compared to existing methods.

Paperid: 2315, https://arxiv.org/pdf/2511.08480.pdf

Abstract:
Vision-language models advance multimodal representation learning by acquiring transferable semantic embeddings, thereby substantially enhancing performance across a range of vision-language tasks, including cross-modal retrieval, clustering, and classification. An effective embedding is expected to comprehensively preserve the semantic content of the input while simultaneously emphasizing features that are discriminative for downstream tasks. Recent approaches demonstrate that VLMs can be adapted into competitive embedding models via large-scale contrastive learning, enabling the simultaneous optimization of two complementary objectives. We argue that the two aforementioned objectives can be decoupled: a comprehensive understanding of the input facilitates the embedding model in achieving superior performance in downstream tasks via contrastive learning. In this paper, we propose CoMa, a compressed pre-training phase, which serves as a warm-up stage for contrastive learning. Experiments demonstrate that with only a small amount of pre-training data, we can transform a VLM into a competitive embedding model. CoMa achieves new state-of-the-art results among VLMs of comparable size on the MMEB, realizing optimization in both efficiency and effectiveness.

Paperid: 2316, https://arxiv.org/pdf/2509.23917.pdf

Abstract:
As a general-purpose vision-language pretraining model, CLIP demonstrates strong generalization ability in image-text alignment tasks and has been widely adopted in downstream applications such as image classification and image-text retrieval. However, it struggles with fine-grained tasks such as object detection and semantic segmentation. While many variants aim to improve CLIP on these tasks, its robustness to adversarial perturbations remains underexplored. Understanding how adversarial examples transfer across tasks is key to assessing CLIP's generalization limits and security risks. In this work, we conduct a systematic empirical analysis of the cross-task transfer behavior of CLIP-based models on image-text retrieval, object detection, and semantic segmentation under adversarial perturbations. We find that adversarial examples generated from fine-grained tasks (e.g., object detection and semantic segmentation) often exhibit stronger transfer potential than those from coarse-grained tasks, enabling more effective attacks against the original CLIP model. Motivated by this observation, we propose a novel framework, Multi-Task Adversarial CLIP (MT-AdvCLIP), which introduces a task-aware feature aggregation loss and generates perturbations with enhanced cross-task generalization capability. This design strengthens the attack effectiveness of fine-grained task models on the shared CLIP backbone. Experimental results on multiple public datasets show that MT-AdvCLIP significantly improves the adversarial transfer success rate (The average attack success rate across multiple tasks is improved by over 39%.) against various CLIP-derived models, without increasing the perturbation budget. This study reveals the transfer mechanism of adversarial examples in multi-task CLIP models, offering new insights into multi-task robustness evaluation and adversarial example design.

Paperid: 2317, https://arxiv.org/pdf/2509.22645.pdf

Abstract:
Class-Incremental Learning (CIL) aims to endow models with the ability to continuously adapt to evolving data streams. Recent advances in pre-trained vision-language models (e.g., CLIP) provide a powerful foundation for this task. However, existing approaches often rely on simplistic templates, such as "a photo of a [CLASS]", which overlook the hierarchical nature of visual concepts. For example, recognizing "cat" versus "car" depends on coarse-grained cues, while distinguishing "cat" from "lion" requires fine-grained details. Similarly, the current feature mapping in CLIP relies solely on the representation from the last layer, neglecting the hierarchical information contained in earlier layers. In this work, we introduce HiErarchical Representation MAtchiNg (HERMAN) for CLIP-based CIL. Our approach leverages LLMs to recursively generate discriminative textual descriptors, thereby augmenting the semantic space with explicit hierarchical cues. These descriptors are matched to different levels of the semantic hierarchy and adaptively routed based on task-specific requirements, enabling precise discrimination while alleviating catastrophic forgetting in incremental tasks. Extensive experiments on multiple benchmarks demonstrate that our method consistently achieves state-of-the-art performance.

Paperid: 2318, https://arxiv.org/pdf/2509.22415.pdf

Abstract:
Multimodal Large Language Models (MLLMs) have achieved remarkable success across diverse vision-language tasks, yet their internal decision-making mechanisms remain insufficiently understood. Existing interpretability research has primarily focused on cross-modal attribution, identifying which image regions the model attends to during output generation. However, these approaches often overlook intra-modal dependencies. In the visual modality, attributing importance to isolated image patches ignores spatial context due to limited receptive fields, resulting in fragmented and noisy explanations. In the textual modality, reliance on preceding tokens introduces spurious activations. Failing to effectively mitigate these interference compromises attribution fidelity. To address these limitations, we propose enhancing interpretability by leveraging intra-modal interaction. For the visual branch, we introduce \textit{Multi-Scale Explanation Aggregation} (MSEA), which aggregates attributions over multi-scale inputs to dynamically adjust receptive fields, producing more holistic and spatially coherent visual explanations. For the textual branch, we propose \textit{Activation Ranking Correlation} (ARC), which measures the relevance of contextual tokens to the current token via alignment of their top-$k$ prediction rankings. ARC leverages this relevance to suppress spurious activations from irrelevant contexts while preserving semantically coherent ones. Extensive experiments across state-of-the-art MLLMs and benchmark datasets demonstrate that our approach consistently outperforms existing interpretability methods, yielding more faithful and fine-grained explanations of model behavior.

Paperid: 2319, https://arxiv.org/pdf/2507.03958.pdf

Abstract:
While large language models (LLMs) are increasingly deployed as dense retrievers, the impact of their domain-specific specialization on retrieval effectiveness remains underexplored. This investigation systematically examines how task-specific adaptations in LLMs influence their retrieval capabilities, an essential step toward developing unified retrievers capable of handling text, code, images, and multimodal content. We conduct extensive experiments with eight Qwen2.5 7B LLMs, including base, instruction-tuned, code/math-specialized, long reasoning, and vision-language models across zero-shot retrieval settings and the supervised setting. For the zero-shot retrieval settings, we consider text retrieval from the BEIR benchmark and code retrieval from the CoIR benchmark. Further, to evaluate supervised performance, all LLMs are fine-tuned on the MS MARCO dataset. We find that mathematical specialization and the long reasoning capability cause consistent degradation in three settings, indicating conflicts between mathematical reasoning and semantic matching. The vision-language model and code-specialized LLMs demonstrate superior zero-shot performance compared to other LLMs, even surpassing BM25 on the code retrieval task, and maintain comparable performance to base LLMs in supervised settings. These findings suggest promising directions for the unified retrieval task leveraging cross-domain and cross-modal fusion.

Paperid: 2320, https://arxiv.org/pdf/2505.23828.pdf

Abstract:
With the rapid development of the Vision-Language Model (VLM), significant progress has been made in Visual Question Answering (VQA) tasks. However, existing VLM often generate inaccurate answers due to a lack of up-to-date knowledge. To address this issue, recent research has introduced Retrieval-Augmented Generation (RAG) techniques, commonly used in Large Language Models (LLM), into VLM, incorporating external multi-modal knowledge to enhance the accuracy and practicality of VLM systems. Nevertheless, the RAG in LLM may be susceptible to data poisoning attacks. RAG-based VLM may also face the threat of this attack. This paper first reveals the vulnerabilities of the RAG-based large model under poisoning attack, showing that existing single-modal RAG poisoning attacks have a 100\% failure rate in multi-modal RAG scenarios. To address this gap, we propose Spa-VLM (Stealthy Poisoning Attack on RAG-based VLM), a new paradigm for poisoning attacks on large models. We carefully craft malicious multi-modal knowledge entries, including adversarial images and misleading text, which are then injected into the RAG's knowledge base. When users access the VLM service, the system may generate misleading outputs. We evaluate Spa-VLM on two Wikipedia datasets and across two different RAGs. Results demonstrate that our method achieves highly stealthy poisoning, with the attack success rate exceeding 0.8 after injecting just 5 malicious entries into knowledge bases with 100K and 2M entries, outperforming state-of-the-art poisoning attacks designed for RAG-based LLMs. Additionally, we evaluated several defense mechanisms, all of which ultimately proved ineffective against Spa-VLM, underscoring the effectiveness and robustness of our attack.

Paperid: 2321, https://arxiv.org/pdf/2505.19569.pdf

Abstract:
Open vocabulary image segmentation tackles the challenge of recognizing dynamically adjustable, predefined novel categories at inference time by leveraging vision-language alignment. However, existing paradigms typically perform class-agnostic region segmentation followed by category matching, which deviates from the human visual system's process of recognizing objects based on semantic concepts, leading to poor alignment between region segmentation and target concepts. To bridge this gap, we propose a novel Cognition-Inspired Framework for open vocabulary image segmentation that emulates the human visual recognition process: first forming a conceptual understanding of an object, then perceiving its spatial extent. The framework consists of three core components: (1) A Generative Vision-Language Model (G-VLM) that mimics human cognition by generating object concepts to provide semantic guidance for region segmentation. (2) A Concept-Aware Visual Enhancer Module that fuses textual concept features with global visual representations, enabling adaptive visual perception based on target concepts. (3) A Cognition-Inspired Decoder that integrates local instance features with G-VLM-provided semantic cues, allowing selective classification over a subset of relevant categories. Extensive experiments demonstrate that our framework achieves significant improvements, reaching $27.2$ PQ, $17.0$ mAP, and $35.3$ mIoU on A-150. It further attains $56.2$, $28.2$, $15.4$, $59.2$, $18.7$, and $95.8$ mIoU on Cityscapes, Mapillary Vistas, A-847, PC-59, PC-459, and PAS-20, respectively. In addition, our framework supports vocabulary-free segmentation, offering enhanced flexibility in recognizing unseen categories. Code will be public.

Paperid: 2322, https://arxiv.org/pdf/2505.18819.pdf

Abstract:
Vision-language models like CLIP can offer a promising foundation for 3D scene understanding when extended with 3D tokenizers. However, standard approaches, such as k-nearest neighbor or radius-based tokenization, struggle with cross-domain generalization due to sensitivity to dataset-specific spatial scales. We present a universal 3D tokenizer designed for scale-invariant representation learning with a frozen CLIP backbone. We show that combining superpoint-based grouping with coordinate scale normalization consistently outperforms conventional methods through extensive experimental analysis. Specifically, we introduce S4Token, a tokenization pipeline that produces semantically-informed tokens regardless of scene scale. Our tokenizer is trained without annotations using masked point modeling and clustering-based objectives, along with cross-modal distillation to align 3D tokens with 2D multi-view image features. For dense prediction tasks, we propose a superpoint-level feature propagation module to recover point-level detail from sparse tokens.

Paperid: 2323, https://arxiv.org/pdf/2505.17461.pdf

Abstract:
Large-scale Vision Language Models (LVLMs) are increasingly being applied to a wide range of real-world multimodal applications, involving complex visual and linguistic reasoning. As these models become more integrated into practical use, they are expected to handle complex aspects of human interaction. Among these, color perception is a fundamental yet highly variable aspect of visual understanding. It differs across individuals due to biological factors such as Color Vision Deficiencies (CVDs), as well as differences in culture and language. Despite its importance, perceptual diversity has received limited attention. In our study, we evaluate LVLMs' ability to account for individual level perceptual variation using the Ishihara Test, a widely used method for detecting CVDs. Our results show that LVLMs can explain CVDs in natural language, but they cannot simulate how people with CVDs perceive color in image based tasks. These findings highlight the need for multimodal systems that can account for color perceptual diversity and support broader discussions on perceptual inclusiveness and fairness in multimodal AI.

Paperid: 2324, https://arxiv.org/pdf/2503.07253.pdf

Abstract:
While existing anomaly synthesis methods have made remarkable progress, achieving both realism and diversity in synthesis remains a major obstacle. To address this, we propose AnomalyPainter, a zero-shot framework that breaks the diversity-realism trade-off dilemma through synergizing Vision Language Large Model (VLLM), Latent Diffusion Model (LDM), and our newly introduced texture library Tex-9K. Tex-9K is a professional texture library containing 75 categories and 8,792 texture assets crafted for diverse anomaly synthesis. Leveraging VLLM's general knowledge, reasonable anomaly text descriptions are generated for each industrial object and matched with relevant diverse textures from Tex-9K. These textures then guide the LDM via ControlNet to paint on normal images. Furthermore, we introduce Texture-Aware Latent Init to stabilize the natural-image-trained ControlNet for industrial images. Extensive experiments show that AnomalyPainter outperforms existing methods in realism, diversity, and generalization, achieving superior downstream performance.

Paperid: 2325, https://arxiv.org/pdf/2501.01282.pdf

Abstract:
Vision-language models (VLMs) have advanced human-AI interaction but struggle with cultural understanding, often misinterpreting symbols, gestures, and artifacts due to biases in predominantly Western-centric training data. In this paper, we construct CultureVerse, a large-scale multimodal benchmark covering 19, 682 cultural concepts, 188 countries/regions, 15 cultural concepts, and 3 question types, with the aim of characterizing and improving VLMs' multicultural understanding capabilities. Then, we propose CultureVLM, a series of VLMs fine-tuned on our dataset to achieve significant performance improvement in cultural understanding. Our evaluation of 16 models reveals significant disparities, with a stronger performance in Western concepts and weaker results in African and Asian contexts. Fine-tuning on our CultureVerse enhances cultural perception, demonstrating cross-cultural, cross-continent, and cross-dataset generalization without sacrificing performance on models' general VLM benchmarks. We further present insights on cultural generalization and forgetting. We hope that this work could lay the foundation for more equitable and culturally aware multimodal AI systems.

Paperid: 2326, https://arxiv.org/pdf/2412.16958.pdf

Abstract:
As deep neural networks (DNNs) are widely applied in the physical world, many researches are focusing on physical-world adversarial examples (PAEs), which introduce perturbations to inputs and cause the model's incorrect outputs. However, existing PAEs face two challenges: unsatisfactory attack performance (i.e., poor transferability and insufficient robustness to environment conditions), and difficulty in balancing attack effectiveness with stealthiness, where better attack effectiveness often makes PAEs more perceptible. In this paper, we explore a novel perturbation-based method to overcome the challenges. For the first challenge, we introduce a strategy Deceptive RF injection based on robust features (RFs) that are predictive, robust to perturbations, and consistent across different models. Specifically, it improves the transferability and robustness of PAEs by covering RFs of other classes onto the predictive features in clean images. For the second challenge, we introduce another strategy Adversarial Semantic Pattern Minimization, which removes most perturbations and retains only essential adversarial patterns in AEsBased on the two strategies, we design our method Robust Feature Coverage Attack (RFCoA), comprising Robust Feature Disentanglement and Adversarial Feature Fusion. In the first stage, we extract target class RFs in feature space. In the second stage, we use attention-based feature fusion to overlay these RFs onto predictive features of clean images and remove unnecessary perturbations. Experiments show our method's superior transferability, robustness, and stealthiness compared to existing state-of-the-art methods. Additionally, our method's effectiveness can extend to Large Vision-Language Models (LVLMs), indicating its potential applicability to more complex tasks.

Paperid: 2327, https://arxiv.org/pdf/2410.11366.pdf

Abstract:
Recent research on query generation has focused on using Large Language Models (LLMs), which despite bringing state-of-the-art performance, also introduce issues with hallucinations in the generated queries. In this work, we introduce relevance hallucination and factuality hallucination as a new typology for hallucination problems brought by query generation based on LLMs. We propose an effective way to separate content from form in LLM-generated queries, which preserves the factual knowledge extracted and integrated from the inputs and compiles the syntactic structure, including function words, using the powerful linguistic capabilities of the LLM. Specifically, we introduce a model-agnostic and training-free method that turns the Large Language Model into a Pointer-Generator (LargePiG), where the pointer attention distribution leverages the LLM's inherent attention weights, and the copy probability is derived from the difference between the vocabulary distribution of the model's high layers and the last layer. To validate the effectiveness of LargePiG, we constructed two datasets for assessing the hallucination problems in query generation, covering both document and video scenarios. Empirical studies on various LLMs demonstrated the superiority of LargePiG on both datasets. Additional experiments also verified that LargePiG could reduce hallucination in large vision language models and improve the accuracy of document-based question-answering and factuality evaluation tasks.

Paperid: 2328, https://arxiv.org/pdf/2410.02729.pdf

Abstract:
Information Retrieval (IR) methods aim to identify documents relevant to a query, which have been widely applied in various natural language tasks. However, existing approaches typically consider only the textual content within documents, overlooking the fact that documents can contain multiple modalities, including images and tables. Also, they often segment each long document into multiple discrete passages for embedding, which prevents them from capturing the overall document context and interactions between paragraphs. To address these two challenges, we propose a method that holistically embeds documents interleaved with multiple modalities by leveraging the capability of recent vision-language models that enable the processing and integration of text, images, and tables into a unified format and representation. Moreover, to mitigate the information loss from segmenting documents into passages, instead of representing and retrieving passages individually, we further merge the representations of segmented passages into one single document representation, while we additionally introduce a reranking strategy to decouple and identify the relevant passage within the document if necessary. Then, through extensive experiments on diverse IR scenarios considering both the textual and multimodal queries, we show that our approach substantially outperforms relevant baselines, thanks to the consideration of the multimodal information within documents.

Paperid: 2329, https://arxiv.org/pdf/2409.13980.pdf

Abstract:
Recent advancements in Vision-Language (VL) research have sparked new benchmarks for complex visual reasoning, challenging models' advanced reasoning ability. Traditional Vision-Language Models (VLMs) perform well in visual perception tasks while struggling with complex reasoning scenarios. Conversely, Large Language Models (LLMs) demonstrate robust text reasoning capabilities; however, they lack visual acuity. To bridge this gap, we propose Complex Visual Reasoning Large Language Models (CVR-LLM), capitalizing on VLMs' visual perception proficiency and LLMs' extensive reasoning capability. Unlike recent multimodal large language models (MLLMs) that require a projection layer, our approach transforms images into detailed, context-aware descriptions using an iterative self-refinement loop and leverages LLMs' text knowledge for accurate predictions without extra training. We also introduce a novel multi-modal in-context learning (ICL) methodology to enhance LLMs' contextual understanding and reasoning. Additionally, we introduce Chain-of-Comparison (CoC), a step-by-step comparison technique enabling contrasting various aspects of predictions. Our CVR-LLM presents the first comprehensive study across a wide array of complex visual reasoning tasks and achieves SOTA performance among all.

Paperid: 2330, https://arxiv.org/pdf/2409.01584.pdf

Abstract:
As the performance of Large-scale Vision Language Models (LVLMs) improves, they are increasingly capable of responding in multiple languages, and there is an expectation that the demand for explanations generated by LVLMs will grow. However, pre-training of Vision Encoder and the integrated training of LLMs with Vision Encoder are mainly conducted using English training data, leaving it uncertain whether LVLMs can completely handle their potential when generating explanations in languages other than English. In addition, multilingual QA benchmarks that create datasets using machine translation have cultural differences and biases, remaining issues for use as evaluation tasks. To address these challenges, this study created an extended dataset in multiple languages without relying on machine translation. This dataset that takes into account nuances and country-specific phrases was then used to evaluate the generation explanation abilities of LVLMs. Furthermore, this study examined whether Instruction-Tuning in resource-rich English improves performance in other languages. Our findings indicate that LVLMs perform worse in languages other than English compared to English. In addition, it was observed that LVLMs struggle to effectively manage the knowledge learned from English data. Our dataset is available at https://huggingface.co/datasets/naist-nlp/MultiExpArt

Paperid: 2331, https://arxiv.org/pdf/2512.16302.pdf

Abstract:
One-shot imitation learning (OSIL) offers a promising way to teach robots new skills without large-scale data collection. However, current OSIL methods are primarily limited to short-horizon tasks, thus limiting their applicability to complex, long-horizon manipulations. To address this limitation, we propose ManiLong-Shot, a novel framework that enables effective OSIL for long-horizon prehensile manipulation tasks. ManiLong-Shot structures long-horizon tasks around physical interaction events, reframing the problem as sequencing interaction-aware primitives instead of directly imitating continuous trajectories. This primitive decomposition can be driven by high-level reasoning from a vision-language model (VLM) or by rule-based heuristics derived from robot state changes. For each primitive, ManiLong-Shot predicts invariant regions critical to the interaction, establishes correspondences between the demonstration and the current observation, and computes the target end-effector pose, enabling effective task execution. Extensive simulation experiments show that ManiLong-Shot, trained on only 10 short-horizon tasks, generalizes to 20 unseen long-horizon tasks across three difficulty levels via one-shot imitation, achieving a 22.8% relative improvement over the SOTA. Additionally, real-robot experiments validate ManiLong-Shot's ability to robustly execute three long-horizon manipulation tasks via OSIL, confirming its practical applicability.

Paperid: 2332, https://arxiv.org/pdf/2511.15669.pdf

Abstract:
Enabling Vision-Language-Action (VLA) models to "think before acting" via Chain-of-Thought (CoT) is a promising path to overcoming the data-hungry nature of end-to-end robot policies. However, progress is stalled by a fundamental conflict: existing models use a single autoregressive decoder for both sequential CoT reasoning and high-dimensional, parallelizable robot actions. This architectural mismatch degrades motor control and fails to forge a strong causal link between thought and action. We introduce DeepThinkVLA, which resolves this conflict through a tightly integrated architecture and training strategy. Architecturally, our hybrid-attention decoder generates sequential CoT with causal attention and then switches to bidirectional attention for fast, parallel decoding of action vectors. This design is complemented by a two-stage training pipeline: we first use Supervised Fine-Tuning (SFT) to teach the model foundational reasoning, then apply Reinforcement Learning (RL) with task-success rewards to causally align the full reasoning-action sequence with desired outcomes. This synergy leads to state-of-the-art performance, achieving a 97.0% success rate on the LIBERO benchmark. Our ablations confirm the design's effectiveness: the hybrid architecture alone outperforms standard decoders by 15.5%, and the final RL stage provides a crucial 2% boost to secure top performance.

Paperid: 2333, https://arxiv.org/pdf/2510.19655.pdf

Abstract:
Zero-shot Vision-and-Language Navigation in Continuous Environments (VLN-CE) requires an agent to navigate unseen environments based on natural language instructions without any prior training. Current methods face a critical trade-off: either rely on environment-specific waypoint predictors that limit scene generalization, or underutilize the reasoning capabilities of large models during navigation. We introduce LaViRA, a simple yet effective zero-shot framework that addresses this dilemma by decomposing action into a coarse-to-fine hierarchy: Language Action for high-level planning, Vision Action for perceptual grounding, and Robot Action for robust navigation. This modular decomposition allows us to leverage the distinct strengths of different scales of Multimodal Large Language Models (MLLMs) at each stage, creating a system that is powerful in its reasoning, grounding and practical control. LaViRA significantly outperforms existing state-of-the-art methods on the VLN-CE benchmark, demonstrating superior generalization capabilities in unseen environments, while maintaining transparency and efficiency for real-world deployment.

Paperid: 2334, https://arxiv.org/pdf/2509.23661.pdf

Abstract:
We present LLaVA-OneVision-1.5, a novel family of Large Multimodal Models (LMMs) that achieve state-of-the-art performance with significantly reduced computational and financial costs. Different from the existing works, LLaVA-OneVision-1.5 provides an open, efficient, and reproducible framework for building high-quality vision-language models entirely from scratch. The LLaVA-OneVision-1.5 release comprises three primary components: (1) Large-Scale Curated Datasets: We construct an 85M concept-balanced pretraining dataset LLaVA-OneVision-1.5-Mid-Traning and a meticulously curated 22M instruction dataset LLaVA-OneVision-1.5-Instruct. (2) Efficient Training Framework: We develop a complete end-to-end efficient training framework leveraging an offline parallel data packing strategy to facilitate the training of LLaVA-OneVision-1.5 within a $16,000 budget. (3) State-of-the-art Performance: Experimental results demonstrate that LLaVA-OneVision-1.5 yields exceptionally competitive performance across a broad range of downstream tasks. Specifically, LLaVA-OneVision-1.5-8B outperforms Qwen2.5-VL-7B on 18 of 27 benchmarks, and LLaVA-OneVision-1.5-4B surpasses Qwen2.5-VL-3B on all 27 benchmarks. We anticipate releasing LLaVA-OneVision-1.5-RL shortly and encourage the community to await further updates.

Paperid: 2335, https://arxiv.org/pdf/2505.00527.pdf

Abstract:
Generalizing language-conditioned multi-task imitation learning (IL) models to novel long-horizon 3D manipulation tasks remains a significant challenge. To address this, we propose DeCo (Task Decomposition and Skill Composition), a model-agnostic framework compatible with various multi-task IL models, designed to enhance their zero-shot generalization to novel, compositional, long-horizon 3D manipulation tasks. DeCo first decomposes IL demonstrations into a set of modular atomic tasks based on the physical interaction between the gripper and objects, and constructs an atomic training dataset that enables models to learn a diverse set of reusable atomic skills during imitation learning. At inference time, DeCo leverages a vision-language model (VLM) to parse high-level instructions for novel long-horizon tasks, retrieve the relevant atomic skills, and dynamically schedule their execution; a spatially-aware skill-chaining module then ensures smooth, collision-free transitions between sequential skills. We evaluate DeCo in simulation using DeCoBench, a benchmark specifically designed to assess zero-shot generalization of multi-task IL models in compositional long-horizon 3D manipulation. Across three representative multi-task IL models (RVT-2, 3DDA, and ARP), DeCo achieves success rate improvements of 66.67%, 21.53%, and 57.92%, respectively, on 12 novel compositional tasks. Moreover, in real-world experiments, a DeCo-enhanced model trained on only 6 atomic tasks successfully completes 9 novel long-horizon tasks, yielding an average success rate improvement of 53.33% over the base multi-task IL model. Video demonstrations are available at: https://deco226.github.io.

Paperid: 2336, https://arxiv.org/pdf/2501.07769.pdf

Abstract:
Vision-language models (VLMs) have exhibited remarkable generalization capabilities, and prompt learning for VLMs has attracted great attention for the ability to adapt pre-trained VLMs to specific downstream tasks. However, existing studies mainly focus on single-modal prompts or uni-directional modality interaction, overlooking the powerful alignment effects resulting from the interaction between the vision and language modalities. To this end, we propose a novel prompt learning method called $\underline{\textbf{B}}i-directional \underline{\textbf{M}}odality \underline{\textbf{I}}nteraction \underline{\textbf{P}}rompt (BMIP)$, which dynamically weights bi-modal information through learning the information of the attention layer, enhancing trainability and inter-modal consistency compared to simple information aggregation methods. To evaluate the effectiveness of prompt learning methods, we propose a more realistic evaluation paradigm called open-world generalization complementing the widely adopted cross-dataset transfer and domain generalization tasks. Comprehensive experiments on various datasets reveal that BMIP not only outperforms current state-of-the-art methods across all three evaluation paradigms but is also flexible enough to be combined with other prompt-based methods for consistent performance enhancement.

Paperid: 2337, https://arxiv.org/pdf/2601.11464.pdf

Abstract:
As vision-language models (VLMs) tackle increasingly complex and multimodal tasks, the rapid growth of Key-Value (KV) cache imposes significant memory and computational bottlenecks during inference. While Multi-Head Latent Attention (MLA) offers an effective means to compress the KV cache and accelerate inference, adapting existing VLMs to the MLA architecture without costly pretraining remains largely unexplored. In this work, we present MHA2MLA-VLM, a parameter-efficient and multimodal-aware framework for converting off-the-shelf VLMs to MLA. Our approach features two core techniques: (1) a modality-adaptive partial-RoPE strategy that supports both traditional and multimodal settings by selectively masking nonessential dimensions, and (2) a modality-decoupled low-rank approximation method that independently compresses the visual and textual KV spaces. Furthermore, we introduce parameter-efficient fine-tuning to minimize adaptation cost and demonstrate that minimizing output activation error, rather than parameter distance, substantially reduces performance loss. Extensive experiments on three representative VLMs show that MHA2MLA-VLM restores original model performance with minimal supervised data, significantly reduces KV cache footprint, and integrates seamlessly with KV quantization.

Paperid: 2338, https://arxiv.org/pdf/2512.22120.pdf

Abstract:
Large vision-language models (VLMs) often benefit from intermediate visual cues, either injected via external tools or generated as latent visual tokens during reasoning, but these mechanisms still overlook fine-grained visual evidence (e.g., polylines in charts), generalize poorly across domains, and incur high inference-time cost. In this paper, we propose Bi-directional Perceptual Shaping (BiPS), which transforms question-conditioned masked views into bidirectional where-to-look signals that shape perception during training. BiPS first applies a KL-consistency constraint between the original image and an evidence-preserving view that keeps only question-relevant regions, encouraging coarse but complete coverage of supporting pixels. It then applies a KL-separation constraint between the original and an evidence-ablated view where critical pixels are masked so the image no longer supports the original answer, discouraging text-only shortcuts (i.e., answering from text alone) and enforcing fine-grained visual reliance. Across eight benchmarks, BiPS boosts Qwen2.5-VL-7B by 8.2% on average and shows strong out-of-domain generalization to unseen datasets and image types.

Paperid: 2339, https://arxiv.org/pdf/2512.11995.pdf

Abstract:
While many vision-language models (VLMs) are developed to answer well-defined, straightforward questions with highly specified targets, as in most benchmarks, they often struggle in practice with complex open-ended tasks, which usually require multiple rounds of exploration and reasoning in the visual space. Such visual thinking paths not only provide step-by-step exploration and verification as an AI detective but also produce better interpretations of the final answers. However, these paths are challenging to evaluate due to the large exploration space of intermediate steps. To bridge the gap, we develop an evaluation suite, ``Visual Reasoning with multi-step EXploration (V-REX)'', which is composed of a benchmark of challenging visual reasoning tasks requiring native multi-step exploration and an evaluation protocol. V-REX covers rich application scenarios across diverse domains. V-REX casts the multi-step exploratory reasoning into a Chain-of-Questions (CoQ) and disentangles VLMs' capability to (1) Planning: breaking down an open-ended task by selecting a chain of exploratory questions; and (2) Following: answering curated CoQ sequentially to collect information for deriving the final answer. By curating finite options of questions and answers per step, V-REX achieves a reliable quantitative and fine-grained analysis of the intermediate steps. By assessing SOTA proprietary and open-sourced VLMs, we reveal consistent scaling trends, significant differences between planning and following abilities, and substantial room for improvement in multi-step exploratory reasoning.

Paperid: 2340, https://arxiv.org/pdf/2509.16567.pdf

Abstract:
Recent black-box counterfactual generation frameworks fail to take into account the semantic content of the proposed edits, while relying heavily on training to guide the generation process. We propose a novel, plug-and-play black-box counterfactual generation framework, which suggests step-by-step edits based on theoretical guarantees of optimal edits to produce human-level counterfactual explanations with zero training. Our framework utilizes a pre-trained image editing diffusion model, and operates without access to the internals of the classifier, leading to an explainable counterfactual generation process. Throughout our experimentation, we showcase the explanatory gap between human reasoning and neural model behavior by utilizing both Convolutional Neural Network (CNN), Vision Transformer (ViT) and Large Vision Language Model (LVLM) classifiers, substantiated through a comprehensive human evaluation.

Paperid: 2341, https://arxiv.org/pdf/2506.01091.pdf

Abstract:
Visual effects (VFX) are key to immersion in modern films, games, and AR/VR. Creating 3D effects requires specialized expertise and training in 3D animation software and can be time consuming. Generative solutions typically rely on computationally intense methods such as diffusion models which can be slow at 4D inference. We reformulate 3D animation as a field prediction task and introduce a text-driven framework that infers a time-varying 4D flow field acting on 3D Gaussians. By leveraging large language models (LLMs) and vision-language models (VLMs) for function generation, our approach interprets arbitrary prompts (e.g., "make the vase glow orange, then explode") and instantly updates color, opacity, and positions of 3D Gaussians in real time. This design avoids overheads such as mesh extraction, manual or physics-based simulations and allows both novice and expert users to animate volumetric scenes with minimal effort on a consumer device even in a web browser. Experimental results show that simple textual instructions suffice to generate compelling time-varying VFX, reducing the manual effort typically required for rigging or advanced modeling. We thus present a fast and accessible pathway to language-driven 3D content creation that can pave the way to democratize VFX further.

Paperid: 2342, https://arxiv.org/pdf/2505.07879.pdf

Abstract:
Vision-language retrieval-augmented generation (RAG) has become an effective approach for tackling Knowledge-Based Visual Question Answering (KB-VQA), which requires external knowledge beyond the visual content presented in images. The effectiveness of Vision-language RAG systems hinges on multimodal retrieval, which is inherently challenging due to the diverse modalities and knowledge granularities in both queries and knowledge bases. Existing methods have not fully tapped into the potential interplay between these elements. We propose a multimodal RAG system featuring a coarse-to-fine, multi-step retrieval that harmonizes multiple granularities and modalities to enhance efficacy. Our system begins with a broad initial search aligning knowledge granularity for cross-modal retrieval, followed by a multimodal fusion reranking to capture the nuanced multimodal information for top entity selection. A text reranker then filters out the most relevant fine-grained section for augmented generation. Extensive experiments on the InfoSeek and Encyclopedic-VQA benchmarks show our method achieves state-of-the-art retrieval performance and highly competitive answering results, underscoring its effectiveness in advancing KB-VQA systems.

Paperid: 2343, https://arxiv.org/pdf/2504.00844.pdf

Abstract:
In Scene Graphs Generation (SGG) one extracts structured representation from visual inputs in the form of objects nodes and predicates connecting them. This facilitates image-based understanding and reasoning for various downstream tasks. Although fully supervised SGG approaches showed steady performance improvements, they suffer from a severe training bias. This is caused by the availability of only small subsets of curated data and exhibits long-tail predicate distribution issues with a lack of predicate diversity adversely affecting downstream tasks. To overcome this, we introduce PRISM-0, a framework for zero-shot open-vocabulary SGG that bootstraps foundation models in a bottom-up approach to capture the whole spectrum of diverse, open-vocabulary predicate prediction. Detected object pairs are filtered and passed to a Vision Language Model (VLM) that generates descriptive captions. These are used to prompt an LLM to generate fine-andcoarse-grained predicates for the pair. The predicates are then validated using a VQA model to provide a final SGG. With the modular and dataset-independent PRISM-0, we can enrich existing SG datasets such as Visual Genome (VG). Experiments illustrate that PRIMS-0 generates semantically meaningful graphs that improve downstream tasks such as Image Captioning and Sentence-to-Graph Retrieval with a performance on par to the best fully supervised methods.

Paperid: 2344, https://arxiv.org/pdf/2503.23508.pdf

Abstract:
Language-based object detection (LOD) aims to align visual objects with language expressions. A large amount of paired data is utilized to improve LOD model generalizations. During the training process, recent studies leverage vision-language models (VLMs) to automatically generate human-like expressions for visual objects, facilitating training data scaling up. In this process, we observe that VLM hallucinations bring inaccurate object descriptions (e.g., object name, color, and shape) to deteriorate VL alignment quality. To reduce VLM hallucinations, we propose an agentic workflow controlled by an LLM to re-align language to visual objects via adaptively adjusting image and text prompts. We name this workflow Real-LOD, which includes planning, tool use, and reflection steps. Given an image with detected objects and VLM raw language expressions, Real-LOD reasons its state automatically and arranges action based on our neural symbolic designs (i.e., planning). The action will adaptively adjust the image and text prompts and send them to VLMs for object re-description (i.e., tool use). Then, we use another LLM to analyze these refined expressions for feedback (i.e., reflection). These steps are conducted in a cyclic form to gradually improve language descriptions for re-aligning to visual objects. We construct a dataset that contains a tiny amount of 0.18M images with re-aligned language expression and train a prevalent LOD model to surpass existing LOD methods by around 50% on the standard benchmarks. Our Real-LOD workflow, with automatic VL refinement, reveals a potential to preserve data quality along with scaling up data quantity, which further improves LOD performance from a data-alignment perspective.

Paperid: 2345, https://arxiv.org/pdf/2503.00436.pdf

Abstract:
In the dynamic landscape of artificial intelligence, the exploration of hallucinations within vision-language (VL) models emerges as a critical frontier. This work delves into the intricacies of hallucinatory phenomena exhibited by widely used image captioners, unraveling interesting patterns. Specifically, we step upon previously introduced techniques of conceptual counterfactual explanations to address VL hallucinations. The deterministic and efficient nature of the employed conceptual counterfactuals backbone is able to suggest semantically minimal edits driven by hierarchical knowledge, so that the transition from a hallucinated caption to a non-hallucinated one is performed in a black-box manner. HalCECE, our proposed hallucination detection framework is highly interpretable, by providing semantically meaningful edits apart from standalone numbers, while the hierarchical decomposition of hallucinated concepts leads to a thorough hallucination analysis. Another novelty tied to the current work is the investigation of role hallucinations, being one of the first works to involve interconnections between visual concepts in hallucination detection. Overall, HalCECE recommends an explainable direction to the crucial field of VL hallucination detection, thus fostering trustworthy evaluation of current and future VL systems.

Paperid: 2346, https://arxiv.org/pdf/2409.20551.pdf

Abstract:
Previous studies on robotic manipulation are based on a limited understanding of the underlying 3D motion constraints and affordances. To address these challenges, we propose a comprehensive paradigm, termed UniAff, that integrates 3D object-centric manipulation and task understanding in a unified formulation. Specifically, we constructed a dataset labeled with manipulation-related key attributes, comprising 900 articulated objects from 19 categories and 600 tools from 12 categories. Furthermore, we leverage MLLMs to infer object-centric representations for manipulation tasks, including affordance recognition and reasoning about 3D motion constraints. Comprehensive experiments in both simulation and real-world settings indicate that UniAff significantly improves the generalization of robotic manipulation for tools and articulated objects. We hope that UniAff will serve as a general baseline for unified robotic manipulation tasks in the future. Images, videos, dataset, and code are published on the project website at:https://sites.google.com/view/uni-aff/home

Paperid: 2347, https://arxiv.org/pdf/2409.18082.pdf

Abstract:
Automating garment manipulation poses a significant challenge for assistive robotics due to the diverse and deformable nature of garments. Traditional approaches typically require separate models for each garment type, which limits scalability and adaptability. In contrast, this paper presents a unified approach using vision-language models (VLMs) to improve keypoint prediction across various garment categories. By interpreting both visual and semantic information, our model enables robots to manage different garment states with a single model. We created a large-scale synthetic dataset using advanced simulation techniques, allowing scalable training without extensive real-world data. Experimental results indicate that the VLM-based method significantly enhances keypoint detection accuracy and task success rates, providing a more flexible and general solution for robotic garment manipulation. In addition, this research also underscores the potential of VLMs to unify various garment manipulation tasks within a single framework, paving the way for broader applications in home automation and assistive robotics for future.

Paperid: 2348, https://arxiv.org/pdf/2409.13540.pdf

Abstract:
Multimodal Large Language Models (MLLMs) have shown promise in a broad range of vision-language tasks with their strong reasoning and generalization capabilities. However, they heavily depend on high-quality data in the Supervised Fine-Tuning (SFT) phase. The existing approaches aim to curate high-quality data via GPT-4V, but they are not scalable due to the commercial nature of GPT-4V and the simplicity of the prompts used to instruct the model. To this end, we devised the FullAnno system, which is a data engine that can generate large-scale, high-quality, and fine-grained image annotations consisting of the category and position of objects, region descriptions, text information, as well as image dense captions. This engine is characterized by its cascade annotation process, which involves multiple expert models and employs rich prompts to instruct LLMs in generating dense image captions. We re-annotated the COCO and Visual Genome datasets using our FullAnno system, tripling the number of object annotations and increasing the length of the original image captions by a factor of 15. Experiments show that the regenerated annotation can significantly enhance the capabilities of LLaVA-v1.5 on several benchmarks. The re-annotated data are available at: https://arcana-project-page.github.io

Paperid: 2349, https://arxiv.org/pdf/2512.22536.pdf

Abstract:
Maintaining narrative coherence and visual consistency remains a central challenge in open-domain video generation. Existing text-to-video models often treat each shot independently, resulting in identity drift, scene inconsistency, and unstable temporal structure. We propose CoAgent, a collaborative and closed-loop framework for coherent video generation that formulates the process as a plan-synthesize-verify pipeline. Given a user prompt, style reference, and pacing constraints, a Storyboard Planner decomposes the input into structured shot-level plans with explicit entities, spatial relations, and temporal cues. A Global Context Manager maintains entity-level memory to preserve appearance and identity consistency across shots. Each shot is then generated by a Synthesis Module under the guidance of a Visual Consistency Controller, while a Verifier Agent evaluates intermediate results using vision-language reasoning and triggers selective regeneration when inconsistencies are detected. Finally, a pacing-aware editor refines temporal rhythm and transitions to match the desired narrative flow. Extensive experiments demonstrate that CoAgent significantly improves coherence, visual consistency, and narrative quality in long-form video generation.

Paperid: 2350, https://arxiv.org/pdf/2512.18477.pdf

Abstract:
We present STORM (Search-Guided Generative World Models), a novel framework for spatio-temporal reasoning in robotic manipulation that unifies diffusion-based action generation, conditional video prediction, and search-based planning. Unlike prior Vision-Language-Action (VLA) models that rely on abstract latent dynamics or delegate reasoning to language components, STORM grounds planning in explicit visual rollouts, enabling interpretable and foresight-driven decision-making. A diffusion-based VLA policy proposes diverse candidate actions, a generative video world model simulates their visual and reward outcomes, and Monte Carlo Tree Search (MCTS) selectively refines plans through lookahead evaluation. Experiments on the SimplerEnv manipulation benchmark demonstrate that STORM achieves a new state-of-the-art average success rate of 51.0 percent, outperforming strong baselines such as CogACT. Reward-augmented video prediction substantially improves spatio-temporal fidelity and task relevance, reducing Frechet Video Distance by over 75 percent. Moreover, STORM exhibits robust re-planning and failure recovery behavior, highlighting the advantages of search-guided generative world models for long-horizon robotic manipulation.

Paperid: 2351, https://arxiv.org/pdf/2512.18368.pdf

Abstract:
While imitation learning has shown impressive results in single-task robot manipulation, scaling it to multi-task settings remains a fundamental challenge due to issues such as suboptimal demonstrations, trajectory noise, and behavioral multi-modality. Existing skill-based methods attempt to address this by decomposing actions into reusable abstractions, but they often rely on fixed-length segmentation or environmental priors that limit semantic consistency and cross-task generalization. In this work, we propose AtomSkill, a novel multi-task imitation learning framework that learns and leverages a structured Atomic Skill Space for composable robot manipulation. Our approach is built on two key technical contributions. First, we construct a Semantically Grounded Atomic Skill Library by partitioning demonstrations into variable-length skills using gripper-state keyframe detection and vision-language model annotation. A contrastive learning objective ensures the resulting skill embeddings are both semantically consistent and temporally coherent. Second, we propose an Action Generation module with Keypose Imagination, which jointly predicts a skill's long-horizon terminal keypose and its immediate action sequence. This enables the policy to reason about overarching motion goals and fine-grained control simultaneously, facilitating robust skill chaining. Extensive experiments in simulated and real-world environments show that AtomSkill consistently outperforms state-of-the-art methods across diverse manipulation tasks.

Paperid: 2352, https://arxiv.org/pdf/2510.26464.pdf

Abstract:
Few-shot anomaly detection (FSAD) methods identify anomalous regions with few known normal samples. Most existing methods rely on the generalization ability of pre-trained vision-language models (VLMs) to recognize potentially anomalous regions through feature similarity between text descriptions and images. However, due to the lack of detailed textual descriptions, these methods can only pre-define image-level descriptions to match each visual patch token to identify potential anomalous regions, which leads to the semantic misalignment between image descriptions and patch-level visual anomalies, achieving sub-optimal localization performance. To address the above issues, we propose the Multi-Level Fine-Grained Semantic Caption (MFSC) to provide multi-level and fine-grained textual descriptions for existing anomaly detection datasets with automatic construction pipeline. Based on the MFSC, we propose a novel framework named FineGrainedAD to improve anomaly localization performance, which consists of two components: Multi-Level Learnable Prompt (MLLP) and Multi-Level Semantic Alignment (MLSA). MLLP introduces fine-grained semantics into multi-level learnable prompts through automatic replacement and concatenation mechanism, while MLSA designs region aggregation strategy and multi-level alignment training to facilitate learnable prompts better align with corresponding visual regions. Experiments demonstrate that the proposed FineGrainedAD achieves superior overall performance in few-shot settings on MVTec-AD and VisA datasets.

Paperid: 2353, https://arxiv.org/pdf/2508.04182.pdf

Abstract:
Multimodal Large Language Models (MLLMs) have demonstrated impressive capabilities across vision-language tasks. However, they may suffer from hallucinations--generating outputs that are semantically inconsistent with the input image or text. Through causal analyses, we find that: (i) hallucinations with omission may arise from the failure to adequately capture essential causal factors, and (ii) hallucinations with fabrication are likely caused by the model being misled by non-causal cues. To address these challenges, we propose a novel reinforcement learning framework guided by causal completeness, which jointly considers both causal sufficiency and causal necessity of tokens. Specifically, we evaluate each token's standalone contribution and counterfactual indispensability to define a token-level causal completeness reward. This reward is used to construct a causally informed advantage function within the GRPO optimization framework, encouraging the model to focus on tokens that are both causally sufficient and necessary for accurate generation. Experimental results across various benchmark datasets and tasks demonstrate the effectiveness of our approach, which effectively mitigates hallucinations in MLLMs.

Paperid: 2354, https://arxiv.org/pdf/2506.15649.pdf

Abstract:
Despite significant advances in inference-time search for vision-language models (VLMs), existing approaches remain both computationally expensive and prone to unpenalized, low-confidence generations which often lead to persistent hallucinations. We introduce \textbf{Value-guided Inference with Margin-based Reward (ViMaR)}, a two-stage inference framework that improves both efficiency and output fidelity by combining a temporal-difference value model with a margin-aware reward adjustment. In the first stage, we perform a single pass to identify the highest-value caption among diverse candidates. In the second stage, we selectively refine only those segments that were overlooked or exhibit weak visual grounding, thereby eliminating frequently rewarded evaluations. A calibrated margin-based penalty discourages low-confidence continuations while preserving descriptive richness. Extensive experiments across multiple VLM architectures demonstrate that ViMaR generates captions that are significantly more reliable, factually accurate, detailed, and explanatory, while achieving over 4$\times$ speedup compared to existing value-guided methods. Specifically, we show that ViMaR trained solely on LLaVA Mistral-7B, \textit{generalizes effectively to guide decoding in a stronger unseen model}. To further validate this, we adapt the ViMaR to steer generation in LLaVA-OneVision-Qwen2-7B, leading to consistent improvements in caption quality and demonstrating robust cross-model guidance. This cross-model generalization highlights ViMaR's flexibility and modularity, positioning it as a scalable and transferable inference-time decoding strategy. Furthermore, when ViMaR-generated captions are used for self-training, the underlying models achieve substantial gains across a broad suite of visual comprehension benchmarks, underscoring the potential of fast, accurate, and self-improving VLM pipelines.

Paperid: 2355, https://arxiv.org/pdf/2505.07062.pdf

Authors:Dong Guo, Faming Wu, Feida Zhu, Fuxing Leng, Guang Shi, Haobin Chen, Haoqi Fan, Jian Wang, Jianyu Jiang, Jiawei Wang, Jingji Chen, Jingjia Huang, Kang Lei, Liping Yuan, Lishu Luo, Pengfei Liu, Qinghao Ye, Rui Qian, Shen Yan, Shixiong Zhao, Shuai Peng, Shuangye Li, Sihang Yuan, Sijin Wu, Tianheng Cheng, Weiwei Liu, Wenqian Wang, Xianhan Zeng, Xiao Liu, Xiaobo Qin, Xiaohan Ding, Xiaojun Xiao, Xiaoying Zhang, Xuanwei Zhang, Xuehan Xiong, Yanghua Peng, Yangrui Chen, Yanwei Li, Yanxu Hu, Yi Lin, Yiyuan Hu, Yiyuan Zhang, Youbin Wu, Yu Li, Yudong Liu, Yue Ling, Yujia Qin, Zanbo Wang, Zhiwu He, Aoxue Zhang, Bairen Yi, Bencheng Liao, Can Huang, Can Zhang, Chaorui Deng, Chaoyi Deng, Cheng Lin, Cheng Yuan, Chenggang Li, Chenhui Gou, Chenwei Lou, Chengzhi Wei, Chundian Liu, Chunyuan Li, Deyao Zhu, Donghong Zhong, Feng Li, Feng Zhang, Gang Wu, Guodong Li, Guohong Xiao, Haibin Lin, Haihua Yang, Haoming Wang, Heng Ji, Hongxiang Hao, Hui Shen, Huixia Li, Jiahao Li, Jialong Wu, Jianhua Zhu, Jianpeng Jiao, Jiashi Feng, Jiaze Chen, Jianhui Duan, Jihao Liu, Jin Zeng, Jingqun Tang, Jingyu Sun, Joya Chen, Jun Long, Junda Feng, Junfeng Zhan, Junjie Fang, Junting Lu, Kai Hua, Kai Liu, Kai Shen, Kaiyuan Zhang, Ke Shen, Ke Wang, Keyu Pan, Kun Zhang, Kunchang Li, Lanxin Li, Lei Li, Lei Shi, Li Han, Liang Xiang, Liangqiang Chen, Lin Chen, Lin Li, Lin Yan, Liying Chi, Longxiang Liu, Mengfei Du, Mingxuan Wang, Ningxin Pan, Peibin Chen, Pengfei Chen, Pengfei Wu, Qingqing Yuan, Qingyao Shuai, Qiuyan Tao, Renjie Zheng, Renrui Zhang, Ru Zhang, Rui Wang, Rui Yang, Rui Zhao, Shaoqiang Xu, Shihao Liang, Shipeng Yan, Shu Zhong, Shuaishuai Cao, Shuangzhi Wu, Shufan Liu, Shuhan Chang, Songhua Cai, Tenglong Ao, Tianhao Yang, Tingting Zhang, Wanjun Zhong, Wei Jia, Wei Weng, Weihao Yu, Wenhao Huang, Wenjia Zhu, Wenli Yang, Wenzhi Wang, Xiang Long, XiangRui Yin, Xiao Li, Xiaolei Zhu, Xiaoying Jia, Xijin Zhang, Xin Liu, Xinchen Zhang, Xinyu Yang, Xiongcai Luo, Xiuli Chen, Xuantong Zhong, Xuefeng Xiao, Xujing Li, Yan Wu, Yawei Wen, Yifan Du, Yihao Zhang, Yining Ye, Yonghui Wu, Yu Liu, Yu Yue, Yufeng Zhou, Yufeng Yuan, Yuhang Xu, Yuhong Yang, Yun Zhang, Yunhao Fang, Yuntao Li, Yurui Ren, Yuwen Xiong, Zehua Hong, Zehua Wang, Zewei Sun, Zeyu Wang, Zhao Cai, Zhaoyue Zha, Zhecheng An, Zhehui Zhao, Zhengzhuo Xu, Zhipeng Chen, Zhiyong Wu, Zhuofan Zheng, Zihao Wang, Zilong Huang, Ziyu Zhu, Zuquan Song

Abstract:
We present Seed1.5-VL, a vision-language foundation model designed to advance general-purpose multimodal understanding and reasoning. Seed1.5-VL is composed with a 532M-parameter vision encoder and a Mixture-of-Experts (MoE) LLM of 20B active parameters. Despite its relatively compact architecture, it delivers strong performance across a wide spectrum of public VLM benchmarks and internal evaluation suites, achieving the state-of-the-art performance on 38 out of 60 public benchmarks. Moreover, in agent-centric tasks such as GUI control and gameplay, Seed1.5-VL outperforms leading multimodal systems, including OpenAI CUA and Claude 3.7. Beyond visual and video understanding, it also demonstrates strong reasoning abilities, making it particularly effective for multimodal reasoning challenges such as visual puzzles. We believe these capabilities will empower broader applications across diverse tasks. In this report, we mainly provide a comprehensive review of our experiences in building Seed1.5-VL across model design, data construction, and training at various stages, hoping that this report can inspire further research. Seed1.5-VL is now accessible at https://www.volcengine.com/ (Volcano Engine Model ID: doubao-1-5-thinking-vision-pro-250428)

Paperid: 2356, https://arxiv.org/pdf/2503.01914.pdf

Abstract:
As deep learning models grow in complexity, achieving model-agnostic interpretability becomes increasingly vital. In this work, we employ post-hoc conceptual contrastive edits to expose noteworthy patterns and biases imprinted in representations of retrieval models. We systematically design optimal and controllable contrastive interventions targeting various parts of speech, and effectively apply them to explain both linguistic and visiolinguistic pre-trained models in a black-box manner. Additionally, we introduce a novel metric to assess the per-word impact of contrastive interventions on model outcomes, providing a comprehensive evaluation of each intervention's effectiveness.

Paperid: 2357, https://arxiv.org/pdf/2412.01256.pdf

Abstract:
The emergence of vision-language foundation models, such as CLIP, has revolutionized image-text representation, enabling a broad range of applications via prompt learning. Despite its promise, real-world datasets often contain noisy labels that can degrade prompt learning performance. In this paper, we demonstrate that using mean absolute error (MAE) loss in prompt learning, named PromptMAE, significantly enhances robustness against noisy labels while maintaining high accuracy. Though MAE is straightforward and recognized for its robustness, it is rarely used in noisy-label learning due to its slow convergence and poor performance outside prompt learning scenarios. To elucidate the robustness of PromptMAE, we leverage feature learning theory to show that MAE can suppress the influence of noisy samples, thereby improving the signal-to-noise ratio and enhancing overall robustness. Additionally, we introduce PromptOT, a prompt-based optimal transport data purification method to enhance the robustness further. PromptOT employs text features in vision-language models as prototypes to construct an optimal transportation matrix. This matrix effectively partitions datasets into clean and noisy subsets, allowing for the application of cross-entropy loss to the clean subset and MAE loss to the noisy subset. Our Noise-Label Prompt Learning method, named NLPrompt, offers a simple and efficient approach that leverages the expressive representations and precise alignment capabilities of vision-language models for robust prompt learning. We validate NLPrompt through extensive experiments across various noise settings, demonstrating significant performance improvements.

Paperid: 2358, https://arxiv.org/pdf/2601.21639.pdf

Abstract:
The development of large vision language models drives the demand for managing, and applying massive amounts of multimodal data, making OCR technology, which extracts information from visual images, increasingly popular. However, existing OCR methods primarily focus on recognizing text elements from images or scanned documents (\textbf{Text-centric OCR}), neglecting the identification of visual elements from visually information-dense image sources (\textbf{Vision-centric OCR}), such as charts, web pages and science plots. In reality, these visually information-dense images are widespread on the internet and have significant real-world application value, such as data visualization and web page analysis. In this technical report, we propose \textbf{OCRVerse}, the first holistic OCR method in end-to-end manner that enables unified text-centric OCR and vision-centric OCR. To this end, we constructe comprehensive data engineering to cover a wide range of text-centric documents, such as newspapers, magazines and books, as well as vision-centric rendered composites, including charts, web pages and scientific plots. Moreover, we propose a two-stage SFT-RL multi-domain training method for OCRVerse. SFT directly mixes cross-domain data to train and establish initial domain knowledge, while RL focuses on designing personalized reward strategies for the characteristics of each domain. Specifically, since different domains require various output formats and expected outputs, we provide sufficient flexibility in the RL stage to customize flexible reward signals for each domain, thereby improving cross-domain fusion and avoiding data conflicts. Experimental results demonstrate the effectiveness of OCRVerse, achieving competitive results across text-centric and vision-centric data types, even comparable to large-scale open-source and closed-source models.

Paperid: 2359, https://arxiv.org/pdf/2511.18437.pdf

Abstract:
Reinforcement Learning with Verifiable Rewards (RLVR) has significantly advanced the reasoning capabilities of Large Language Models (LLMs) and is now being applied to Vision-Language Models (VLMs). However, vanilla RLVR for VLMs verifies only the final textual output, critically neglecting the foundational step of visual perception. This oversight leads to visual hallucinations and reward hacking, as reasoning built upon flawed perception is inherently unreliable. To address this, we propose PEARL (Perceptual-Evidence Anchored Reinforced Learning), a dual-branch, perception-reasoning synergistic that strengthens multimodal reasoning by explicitly anchoring it to verified visual evidence. For each reasoning-oriented QA instance, PEARL first derive a perception checklist -- a set of perception-oriented sub-questions with verifiable answers that probe the model's understanding of key visual evidence. During training, auxiliary rollouts on this checklist yield a perceptual reward that both directly reinforces the model's perception ability and acts as a fidelity gate for reasoning. If the model passes the perception check, its policy update is biased towards evidence-anchored reasoning. Otherwise, the process is halted to prevent reasoning from flawed premises. PEARL can be seamlessly integrated with popular RL methods like GRPO and DAPO. Comprehensive experiments show PEARL achieves substantial gains on multimodal reasoning benchmarks, e.g., a +9.7% improvement over the baseline and +6.6% over GRPO on MathVerse.

Paperid: 2360, https://arxiv.org/pdf/2510.23594.pdf

Abstract:
Multimodal large language models (MLLMs) have achieved remarkable progress on vision-language tasks, yet their reasoning processes remain sometimes unreliable. We introduce PRISM-Bench, a benchmark of puzzle-based visual challenges designed to evaluate not only whether models can solve problems, but how their reasoning unfolds. Unlike prior evaluations that measure only final-answer accuracy, PRISM-Bench introduces a diagnostic task: given a visual puzzle and a step-by-step chain-of-thought (CoT) containing exactly one error, models must identify the first incorrect step. This setting enables fine-grained assessment of logical consistency, error detection, and visual reasoning. The puzzles in PRISM-Bench require multi-step symbolic, geometric, and analogical reasoning, resisting shortcuts based on superficial pattern matching. Evaluations across state-of-the-art MLLMs reveal a persistent gap between fluent generation and faithful reasoning: models that produce plausible CoTs often fail to locate simple logical faults. By disentangling answer generation from reasoning verification, PRISM-Bench offers a sharper lens on multimodal reasoning competence and underscores the need for diagnostic evaluation protocols in the development of trustworthy MLLMs.

Paperid: 2361, https://arxiv.org/pdf/2510.20333.pdf

Abstract:
Vision-Language Models (VLMs) are increasingly deployed as autonomous agents to navigate mobile graphical user interfaces (GUIs). Operating in dynamic on-device ecosystems, which include notifications, pop-ups, and inter-app interactions, exposes them to a unique and underexplored threat vector: environmental injection. Unlike prompt-based attacks that manipulate textual instructions, environmental injection corrupts an agent's visual perception by inserting adversarial UI elements (for example, deceptive overlays or spoofed notifications) directly into the GUI. This bypasses textual safeguards and can derail execution, causing privacy leakage, financial loss, or irreversible device compromise. To systematically evaluate this threat, we introduce GhostEI-Bench, the first benchmark for assessing mobile agents under environmental injection attacks within dynamic, executable environments. Moving beyond static image-based assessments, GhostEI-Bench injects adversarial events into realistic application workflows inside fully operational Android emulators and evaluates performance across critical risk scenarios. We further propose a judge-LLM protocol that conducts fine-grained failure analysis by reviewing the agent's action trajectory alongside the corresponding screenshot sequence, pinpointing failure in perception, recognition, or reasoning. Comprehensive experiments on state-of-the-art agents reveal pronounced vulnerability to deceptive environmental cues: current models systematically fail to perceive and reason about manipulated UIs. GhostEI-Bench provides a framework for quantifying and mitigating this emerging threat, paving the way toward more robust and secure embodied agents.

Paperid: 2362, https://arxiv.org/pdf/2510.02750.pdf

Abstract:
Vision-language models (VLMs) such as CLIP and Grounding DINO have achieved remarkable success in object recognition and detection. However, their performance often degrades under real-world distribution shifts. Test-time adaptation (TTA) aims to mitigate this issue by adapting models during inference. Existing methods either rely on computationally expensive backpropagation, which hinders real-time deployment, or focus solely on likelihood adaptation, which overlooks the critical role of the prior. Our prior work, Bayesian Class Adaptation (BCA), addressed these shortcomings for object recognition by introducing a training-free framework that incorporates adaptive priors. Building upon this foundation, we now present Bayesian Class Adaptation plus (BCA+), a unified, training-free framework for TTA for both object recognition and detection. BCA+ introduces a dynamic cache that adaptively stores and updates class embeddings, spatial scales (for detection), and, crucially, adaptive class priors derived from historical predictions. We formulate adaptation as a Bayesian inference problem, where final predictions are generated by fusing the initial VLM output with a cache-based prediction. This cache-based prediction combines a dynamically updated likelihood (measuring feature and scale similarity) and a prior (reflecting the evolving class distribution). This dual-adaptation mechanism, coupled with uncertainty-guided fusion, enables BCA+ to correct both the model's semantic understanding and its contextual confidence. As a training-free method requiring no backpropagation, BCA+ is highly efficient. Extensive experiments demonstrate that BCA+ achieves state-of-the-art performance on both recognition and detection benchmarks.

Paperid: 2363, https://arxiv.org/pdf/2509.25866.pdf

Abstract:
The "thinking with images" paradigm represents a pivotal shift in the reasoning of Vision Language Models (VLMs), moving from text-dominant chain-of-thought to image-interactive reasoning. By invoking visual tools or generating intermediate visual representations, VLMs can iteratively attend to fine-grained regions, enabling deeper image understanding and more faithful multimodal reasoning. As an emerging paradigm, however, it still leaves substantial room for exploration in data construction accuracy, structural design, and broader application scenarios, which offer rich opportunities for advancing multimodal reasoning. To further advance this line of work, we present DeepSketcher, a comprehensive suite comprising both an image-text interleaved dataset and a self-contained model. The dataset contains 31k chain-of-thought (CoT) reasoning trajectories with diverse tool calls and resulting edited images, covering a wide range of data types and manipulation instructions with high annotation accuracy. Building on this resource, we design a model that performs interleaved image-text reasoning and natively generates "visual thoughts" by operating directly in the visual embedding space, rather than invoking external tools and repeatedly re-encoding generated images. This design enables tool-free and more flexible "thinking with images". Extensive experiments on multimodal reasoning benchmarks demonstrate strong performance, validating both the utility of the dataset and the effectiveness of the model design.

Paperid: 2364, https://arxiv.org/pdf/2509.21552.pdf

Abstract:
Graphical User Interface (GUI) grounding is commonly framed as a coordinate prediction task -- given a natural language instruction, generate on-screen coordinates for actions such as clicks and keystrokes. However, recent Vision Language Models (VLMs) often fail to predict accurate numeric coordinates when processing high-resolution GUI images with complex layouts. To address this issue, we reframe GUI grounding as an \emph{interactive search task}, where the VLM generates actions to move a cursor in the GUI to locate UI elements. At each step, the model determines the target object, evaluates the spatial relations between the cursor and the target, and moves the cursor closer to the target conditioned on the movement history. In this interactive process, the rendered cursor provides visual feedback to help the model align its predictions with the corresponding on-screen locations. We train our GUI grounding model, GUI-Cursor, using multi-step online reinforcement learning with a dense trajectory-based reward function. Our experimental results show that GUI-Cursor, based on Qwen2.5-VL-7B, improves the GUI grounding accuracy and achieves state-of-the-art results on ScreenSpot-v2 ($88.8\% \rightarrow 93.9\%$) and ScreenSpot-Pro ($26.8\% \rightarrow 56.5\%$). Moreover, we observe that GUI-Cursor learns to solve the problem within two steps for 95\% of instances and can adaptively conduct more steps on more difficult examples.

Paperid: 2365, https://arxiv.org/pdf/2509.21301.pdf

Abstract:
This paper presents Nova, a real-time scheduling framework for serving agentic vision-language models (VLMs) on a single GPU with balanced per-request latency and overall request process throughput. Our design begins by enabling effective pipelining across vision encode, LLM prefill, and LLM decode stages of VLMs, by exploiting their heterogeneous resource demands during execution and incorporating elastic GPU spatial partitioning among stages to maximally utilize the compute and memory resources. Building on this, we introduce a real-time scheduling algorithm that adaptively calibrates resource allocation among stages based on a Pareto-optimal analysis of the latency-throughput trade-off, allowing the system to sustain responsiveness and resource efficiency under dynamic request loads. To further alleviate GPU memory pressure, we design a lightweight weight offloading strategy for vision encoders that preserves inference efficiency with minimized memory overhead. Extensive evaluations on both synthetic and real-world agent workloads demonstrate that Nova consistently outperforms the state-of-the-art baselines, improving the maximum latency by up to 23.3%, while keeping competitive throughput.

Paperid: 2366, https://arxiv.org/pdf/2508.13587.pdf

Abstract:
While reinforcement learning (RL) has proven highly effective for general reasoning in vision-language models, its application to tasks requiring in-depth understanding of information-rich images and generation of structured outputs remains underexplored. Chart-to-code generation exemplifies this challenge, demanding complex reasoning over visual charts to generate structured code. Supervised fine-tuning (SFT) alone is often insufficient, highlighting the need for effective RL strategies that appropriately reward structured outputs. We systematically investigate the performance plateau in SFT through large-scale experiments and propose Multimodal Structured Reinforcement Learning (MSRL) for chart-to-code generation, which substantially breaks through this plateau. We construct the largest training corpus to date, containing 3 million chart-code pairs from real-world arXiv tables to mitigate simplistic patterns of prior synthetic data. Despite reaching state-of-the-art performance, our experiments show that scaling SFT data eventually hits a plateau where further increases yield negligible improvements. Our MSRL method leverages a multi-granularity structured reward system using multimodal textual and visual feedback. At the textual level, rule-based rewards validate fine-grained code details. At the visual level, model-based rewards assess structural similarity by rendering generated code into images and employing an evaluator model. We implement this within a two-stage curriculum for training stability. Results demonstrate that MSRL significantly breaks the SFT plateau, improving high-level metrics by 6.2% and 9.9% on ChartMimic and ReachQA benchmarks respectively, achieving competitive performance with advanced closed-source models.

Paperid: 2367, https://arxiv.org/pdf/2508.05008.pdf

Abstract:
Vision-Language Models (VLMs), such as CLIP, have demonstrated remarkable zero-shot capabilities in various computer vision tasks. However, their application to medical imaging remains challenging due to the high variability and complexity of medical data. Specifically, medical images often exhibit significant domain shifts caused by various confounders, including equipment differences, procedure artifacts, and imaging modes, which can lead to poor generalization when models are applied to unseen domains. To address this limitation, we propose Multimodal Causal-Driven Representation Learning (MCDRL), a novel framework that integrates causal inference with the VLM to tackle domain generalization in medical image segmentation. MCDRL is implemented in two steps: first, it leverages CLIP's cross-modal capabilities to identify candidate lesion regions and construct a confounder dictionary through text prompts, specifically designed to represent domain-specific variations; second, it trains a causal intervention network that utilizes this dictionary to identify and eliminate the influence of these domain-specific variations while preserving the anatomical structural information critical for segmentation tasks. Extensive experiments demonstrate that MCDRL consistently outperforms competing methods, yielding superior segmentation accuracy and exhibiting robust generalizability.

Paperid: 2368, https://arxiv.org/pdf/2508.00311.pdf

Abstract:
Optical Character Recognition (OCR) for mathematical formula is essential for the intelligent analysis of scientific literature. However, both task-specific and general vision-language models often struggle to handle the structural diversity, complexity, and real-world variability inherent in mathematical content. In this work, we present DocTron-Formula, a unified framework built upon general vision-language models, thereby eliminating the need for specialized architectures. Furthermore, we introduce CSFormula, a large-scale and challenging dataset that encompasses multidisciplinary and structurally complex formulas at the line, paragraph, and page levels. Through straightforward supervised fine-tuning, our approach achieves state-of-the-art performance across a variety of styles, scientific domains, and complex layouts. Experimental results demonstrate that our method not only surpasses specialized models in terms of accuracy and robustness, but also establishes a new paradigm for the automated understanding of complex scientific documents.

Paperid: 2369, https://arxiv.org/pdf/2507.15509.pdf

Abstract:
Recently, inspired by OpenAI-o1/o3 and Deepseek-R1, the R1-Style method based on reinforcement learning fine-tuning has received widespread attention from the community. Previous R1-Style methods mainly focus on mathematical reasoning and code intelligence. It is of great research significance to verify their advantages on more general multimodal data. Chart is an important multimodal data type with rich information, which brings important research challenges in complex reasoning. In this work, we introduce Chart-R1, a chart-domain vision-language model with reinforcement learning fine-tuning to enable complex chart reasoning. To support Chart-R1, we first propose a novel programmatic data synthesis technology to generate high-quality step-by-step chart reasoning data covering single- and multi-subcharts, which makes up for the lack of reasoning data in the chart domain. Then we develop a two-stage training strategy: Chart-COT with step-by-step chain-of-thought supervision, and Chart-RFT with numerically sensitive reinforcement fine-tuning. Chart-COT aims to decompose complex chart reasoning tasks into fine-grained, understandable subtasks through step-by-step supervision, which lays a good foundation for improving the reasoning level of reinforcement learning. Chart-RFT utilize the typical group relative policy optimization strategy, in which a relatively soft reward is adopted for numerical response to emphasize the numerical sensitivity in the chart domain. We conduct extensive experiments on open-source benchmarks and self-built chart reasoning dataset (\emph{i.e., ChartRQA}). Experimental results show that Chart-R1 has significant advantages compared to chart-domain methods, even comparable to open/closed source large-scale models (\emph{e.g., GPT-4o, Claude-3.5}).

Paperid: 2370, https://arxiv.org/pdf/2507.09961.pdf

Abstract:
Deep learning often struggles when training and test data distributions differ. Traditional domain generalization (DG) tackles this by including data from multiple source domains, which is impractical due to expensive data collection and annotation. Recent vision-language models like CLIP enable source-free domain generalization (SFDG) by using text prompts to simulate visual representations, reducing data demands. However, existing SFDG methods struggle with domain-specific confounders, limiting their generalization capabilities. To address this issue, we propose TDCRL (\textbf{T}ext-\textbf{D}riven \textbf{C}ausal \textbf{R}epresentation \textbf{L}earning), the first method to integrate causal inference into the SFDG setting. TDCRL operates in two steps: first, it employs data augmentation to generate style word vectors, combining them with class information to generate text embeddings to simulate visual representations; second, it trains a causal intervention network with a confounder dictionary to extract domain-invariant features. Grounded in causal learning, our approach offers a clear and effective mechanism to achieve robust, domain-invariant features, ensuring robust generalization. Extensive experiments on PACS, VLCS, OfficeHome, and DomainNet show state-of-the-art performance, proving TDCRL effectiveness in SFDG.

Paperid: 2371, https://arxiv.org/pdf/2506.23046.pdf

Abstract:
Humans continuously infer the states, goals, and behaviors of others by perceiving their surroundings in dynamic, real-world social interactions. However, most Theory of Mind (ToM) benchmarks only evaluate static, text-based scenarios, which have a significant gap compared to real interactions. We propose the SoMi-ToM benchmark, designed to evaluate multi-perspective ToM in embodied multi-agent complex social interactions. This benchmark is based on rich multimodal interaction data generated by the interaction environment SoMi, covering diverse crafting goals and social relationships. Our framework supports multi-level evaluation: (1) first-person evaluation provides multimodal (visual, dialogue, action, etc.) input from a first-person perspective during a task for real-time state inference, (2) third-person evaluation provides complete third-person perspective video and text records after a task for goal and behavior inference. This evaluation method allows for a more comprehensive examination of a model's ToM capabilities from both the subjective immediate experience and the objective global observation. We constructed a challenging dataset containing 35 third-person perspective videos, 363 first-person perspective images, and 1225 expert-annotated multiple-choice questions (three options). On this dataset, we systematically evaluated the performance of human subjects and several state-of-the-art large vision-language models (LVLMs). The results show that LVLMs perform significantly worse than humans on SoMi-ToM: the average accuracy gap between humans and models is 40.1% in first-person evaluation and 26.4% in third-person evaluation. This indicates that future LVLMs need to further improve their ToM capabilities in embodied, complex social interactions.

Paperid: 2372, https://arxiv.org/pdf/2506.18246.pdf

Abstract:
Using natural language to query visual information is a fundamental need in real-world applications. Text-Image Retrieval (TIR) retrieves a target image from a gallery based on an image-level description, while Referring Expression Comprehension (REC) localizes a target object within a given image using an instance-level description. However, real-world applications often present more complex demands. Users typically query an instance-level description across a large gallery and expect to receive both relevant image and the corresponding instance location. In such scenarios, TIR struggles with fine-grained descriptions and object-level localization, while REC is limited in its ability to efficiently search large galleries and lacks an effective ranking mechanism. In this paper, we introduce a new task called \textbf{Referring Expression Instance Retrieval (REIR)}, which supports both instance-level retrieval and localization based on fine-grained referring expressions. First, we propose a large-scale benchmark for REIR, named REIRCOCO, constructed by prompting advanced vision-language models to generate high-quality referring expressions for instances in the MSCOCO and RefCOCO datasets. Second, we present a baseline method, Contrastive Language-Instance Alignment with Relation Experts (CLARE), which employs a dual-stream architecture to address REIR in an end-to-end manner. Given a referring expression, the textual branch encodes it into a query embedding. The visual branch detects candidate objects and extracts their instance-level visual features. The most similar candidate to the query is selected for bounding box prediction. CLARE is first trained on object detection and REC datasets to establish initial grounding capabilities, then optimized via Contrastive Language-Instance Alignment (CLIA) for improved retrieval across images. We will release our code and benchmark publicly.

Paperid: 2373, https://arxiv.org/pdf/2506.16806.pdf

Abstract:
Recent Large Vision Language Models (LVLMs) demonstrate promising capabilities in unifying visual understanding and generative modeling, enabling both accurate content understanding and flexible editing. However, current approaches treat "what to see" and "how to edit" separately: they either perform isolated object segmentation or utilize segmentation masks merely as conditional prompts for local edit generation tasks, often relying on multiple disjointed models. To bridge these gaps, we introduce FOCUS, a unified LVLM that integrates segmentation-aware perception and controllable object-centric generation within an end-to-end framework. FOCUS employs a dual-branch visual encoder to simultaneously capture global semantic context and fine-grained spatial details. In addition, we leverage a MoVQGAN-based visual tokenizer to produce discrete visual tokens that enhance generation quality. To enable accurate and controllable image editing, we propose a progressive multi-stage training pipeline, where segmentation masks are jointly optimized and used as spatial condition prompts to guide the diffusion decoder. This strategy aligns visual encoding, segmentation, and generation modules, effectively bridging segmentation-aware perception with fine-grained visual synthesis. Extensive experiments across three core tasks, including multimodal understanding, referring segmentation accuracy, and controllable image generation, demonstrate that FOCUS achieves strong performance by jointly optimizing visual perception and generative capabilities.

Paperid: 2374, https://arxiv.org/pdf/2506.10100.pdf

Abstract:
Vision-Language-Action (VLA) models, particularly diffusion-based architectures, demonstrate transformative potential for embodied intelligence but are severely hampered by high computational and memory demands stemming from extensive inherent and inference-time redundancies. While existing acceleration efforts often target isolated inefficiencies, such piecemeal solutions typically fail to holistically address the varied computational and memory bottlenecks across the entire VLA pipeline, thereby limiting practical deployability. We introduce EfficientVLA, a structured and training-free inference acceleration framework that systematically eliminates these barriers by cohesively exploiting multifaceted redundancies. EfficientVLA synergistically integrates three targeted strategies: (1) pruning of functionally inconsequential layers from the language module, guided by an analysis of inter-layer redundancies; (2) optimizing the visual processing pathway through a task-aware strategy that selects a compact, diverse set of visual tokens, balancing task-criticality with informational coverage; and (3) alleviating temporal computational redundancy within the iterative diffusion-based action head by strategically caching and reusing key intermediate features. We apply our method to a standard VLA model CogACT, yielding a 1.93X inference speedup and reduces FLOPs to 28.9%, with only a 0.6% success rate drop in the SIMPLER benchmark.

Paperid: 2375, https://arxiv.org/pdf/2506.03147.pdf

Abstract:
Although existing unified models achieve strong performance in vision-language understanding and text-to-image generation, they remain limited in addressing image perception and manipulation -- capabilities increasingly demanded in practical applications. Recently, OpenAI introduced the powerful GPT-4o-Image model, which showcases advanced capabilities in comprehensive image perception and manipulation, sparking widespread interest. Through carefully designed experiments, we observe that GPT-4o-Image likely relies on semantic encoders rather than VAEs for feature extraction, despite VAEs being commonly regarded as crucial for image manipulation tasks. Inspired by this insight, we propose UniWorld-V1, a unified generative framework built upon semantic features extracted from powerful multimodal large language models and contrastive semantic encoders. Using only 2.7M training data, UniWorld-V1 achieves impressive performance across diverse tasks, including image understanding, generation, manipulation, and perception. We fully open-source the UniWorld-V1 framework, including model weights, training and evaluation scripts, and datasets to promote reproducibility and further research.

Paperid: 2376, https://arxiv.org/pdf/2505.19610.pdf

Abstract:
Vision-Language Models (VLMs) exhibit impressive performance, yet the integration of powerful vision encoders has significantly broadened their attack surface, rendering them increasingly susceptible to jailbreak attacks. However, lacking well-defined attack objectives, existing jailbreak methods often struggle with gradient-based strategies prone to local optima and lacking precise directional guidance, and typically decouple visual and textual modalities, thereby limiting their effectiveness by neglecting crucial cross-modal interactions. Inspired by the Eliciting Latent Knowledge (ELK) framework, we posit that VLMs encode safety-relevant information within their internal fusion-layer representations, revealing an implicit safety decision boundary in the latent space. This motivates exploiting boundary to steer model behavior. Accordingly, we propose JailBound, a novel latent space jailbreak framework comprising two stages: (1) Safety Boundary Probing, which addresses the guidance issue by approximating decision boundary within fusion layer's latent space, thereby identifying optimal perturbation directions towards the target region; and (2) Safety Boundary Crossing, which overcomes the limitations of decoupled approaches by jointly optimizing adversarial perturbations across both image and text inputs. This latter stage employs an innovative mechanism to steer the model's internal state towards policy-violating outputs while maintaining cross-modal semantic consistency. Extensive experiments on six diverse VLMs demonstrate JailBound's efficacy, achieves 94.32% white-box and 67.28% black-box attack success averagely, which are 6.17% and 21.13% higher than SOTA methods, respectively. Our findings expose a overlooked safety risk in VLMs and highlight the urgent need for more robust defenses. Warning: This paper contains potentially sensitive, harmful and offensive content.

Paperid: 2377, https://arxiv.org/pdf/2505.13840.pdf

Abstract:
Large Language Models (LLMs) have driven significant progress, yet their growing parameter counts and context windows incur prohibitive compute, energy, and monetary costs. We introduce EfficientLLM, a novel benchmark and the first comprehensive empirical study evaluating efficiency techniques for LLMs at scale. Conducted on a production-class cluster (48xGH200, 8xH200 GPUs), our study systematically explores three key axes: (1) architecture pretraining (efficient attention variants: MQA, GQA, MLA, NSA; sparse Mixture-of-Experts (MoE)), (2) fine-tuning (parameter-efficient methods: LoRA, RSLoRA, DoRA), and (3) inference (quantization methods: int4, float16). We define six fine-grained metrics (Memory Utilization, Compute Utilization, Latency, Throughput, Energy Consumption, Compression Rate) to capture hardware saturation, latency-throughput balance, and carbon cost. Evaluating over 100 model-technique pairs (0.5B-72B parameters), we derive three core insights: (i) Efficiency involves quantifiable trade-offs: no single method is universally optimal; e.g., MoE reduces FLOPs and improves accuracy but increases VRAM by 40%, while int4 quantization cuts memory/energy by up to 3.9x at a 3-5% accuracy drop. (ii) Optima are task- and scale-dependent: MQA offers optimal memory-latency trade-offs for constrained devices, MLA achieves lowest perplexity for quality-critical tasks, and RSLoRA surpasses LoRA efficiency only beyond 14B parameters. (iii) Techniques generalize across modalities: we extend evaluations to Large Vision Models (Stable Diffusion 3.5, Wan 2.1) and Vision-Language Models (Qwen2.5-VL), confirming effective transferability. By open-sourcing datasets, evaluation pipelines, and leaderboards, EfficientLLM provides essential guidance for researchers and engineers navigating the efficiency-performance landscape of next-generation foundation models.

Paperid: 2378, https://arxiv.org/pdf/2503.10582.pdf

Abstract:
Vision-Language Models have made significant progress on many perception-focused tasks. However, their progress on reasoning-focused tasks remains limited due to the lack of high-quality and diverse training data. In this work, we aim to address the scarcity of reasoning-focused multimodal datasets. We propose VisualWebInstruct, a novel approach that leverages search engines to create a diverse and high-quality dataset spanning multiple disciplines, including mathematics, physics, finance, and chemistry, etc. Starting with a meticulously selected set of 30,000 seed images, we employ Google Image Search to identify websites containing similar images. We collect and process HTML data from over 700K unique URLs. Through a pipeline of content extraction, filtering, and synthesis, we construct a dataset of approximately 900K question-answer (QA) pairs, with 40% consisting of visual QA pairs and the remaining comprising text-based QA pairs. Models fine-tuned on VisualWebInstruct demonstrate significant performance improvements: (1) fine-tuning on Llava-OV results in 10-20 absolute points improvement across benchmarks, and (2) fine-tuning from MAmmoTH-VL yields a 5 absolute points gain across benchmarks. Our best model, MAmmoTH-VL2, achieves state-of-the-art performance within the 10B parameter class on MMMU-Pro (40.7), MathVerse (42.6), and DynaMath (55.7). These results highlight the effectiveness of our dataset in enhancing the reasoning capabilities of vision-language models for complex multimodal tasks.

Paperid: 2379, https://arxiv.org/pdf/2503.08481.pdf

Abstract:
Understanding the environment and a robot's physical reachability is crucial for task execution. While state-of-the-art vision-language models (VLMs) excel in environmental perception, they often generate inaccurate or impractical responses in embodied visual reasoning tasks due to a lack of understanding of robotic physical reachability. To address this issue, we propose a unified representation of physical reachability across diverse robots, i.e., Space-Physical Reachability Map (S-P Map), and PhysVLM, a vision-language model that integrates this reachability information into visual reasoning. Specifically, the S-P Map abstracts a robot's physical reachability into a generalized spatial representation, independent of specific robot configurations, allowing the model to focus on reachability features rather than robot-specific parameters. Subsequently, PhysVLM extends traditional VLM architectures by incorporating an additional feature encoder to process the S-P Map, enabling the model to reason about physical reachability without compromising its general vision-language capabilities. To train and evaluate PhysVLM, we constructed a large-scale multi-robot dataset, Phys100K, and a challenging benchmark, EQA-phys, which includes tasks for six different robots in both simulated and real-world environments. Experimental results demonstrate that PhysVLM outperforms existing models, achieving a 14\% improvement over GPT-4o on EQA-phys and surpassing advanced embodied VLMs such as RoboMamba and SpatialVLM on the RoboVQA-val and OpenEQA benchmarks. Additionally, the S-P Map shows strong compatibility with various VLMs, and its integration into GPT-4o-mini yields a 7.1\% performance improvement.

Paperid: 2380, https://arxiv.org/pdf/2503.06492.pdf

Abstract:
Large vision-language models (LVLMs) have demonstrated remarkable achievements, yet the generation of non-factual responses remains prevalent in fact-seeking question answering (QA). Current multimodal fact-seeking benchmarks primarily focus on comparing model outputs to ground truth answers, providing limited insights into the performance of modality-specific modules. To bridge this gap, we introduce VisualSimpleQA, a multimodal fact-seeking benchmark with two key features. First, it enables streamlined and decoupled evaluation of LVLMs in visual and linguistic modalities. Second, it incorporates well-defined difficulty criteria to guide human annotation and facilitates the extraction of a challenging subset, VisualSimpleQA-hard. Experiments on 15 LVLMs show that even state-of-the-art models such as GPT-4o achieve merely 60%+ correctness in multimodal fact-seeking QA on VisualSimpleQA and 30%+ on VisualSimpleQA-hard. Furthermore, the decoupled evaluation across these models highlights substantial opportunities for improvement in both visual and linguistic modules. The dataset is available at https://huggingface.co/datasets/WYLing/VisualSimpleQA.

Paperid: 2381, https://arxiv.org/pdf/2412.14475.pdf

Abstract:
Despite the rapidly growing demand for multimodal retrieval, progress in this field remains severely constrained by a lack of training data. In this paper, we introduce MegaPairs, a novel data synthesis method that leverages vision language models (VLMs) and open-domain images, together with a massive synthetic dataset generated from this method. Our empirical analysis shows that MegaPairs generates high-quality data, enabling the multimodal retriever to significantly outperform the baseline model trained on 70$\times$ more data from existing datasets. Moreover, since MegaPairs solely relies on general image corpora and open-source VLMs, it can be easily scaled up, enabling continuous improvements in retrieval performance. In this stage, we produced more than 26 million training instances and trained several models of varying sizes using this data. These new models achieve state-of-the-art zero-shot performance across 4 popular composed image retrieval (CIR) benchmarks and the highest overall performance on the 36 datasets provided by MMEB. They also demonstrate notable performance improvements with additional downstream fine-tuning. Our produced dataset, well-trained models, and data synthesis pipeline will be made publicly available to facilitate the future development of this field.

Paperid: 2382, https://arxiv.org/pdf/2412.13949.pdf

Abstract:
Large vision-language models (LVLMs) have made substantial progress in integrating large language models (LLMs) with visual inputs, enabling advanced multimodal reasoning. Despite their success, a persistent challenge is hallucination-where generated text fails to accurately reflect visual content-undermining both accuracy and reliability. Existing methods focus on alignment training or decoding refinements but primarily address symptoms at the generation stage without probing the underlying causes. In this work, we investigate the internal mechanisms driving hallucination in LVLMs, with an emphasis on the multi-head attention module. Specifically, we introduce Vision-aware Head Divergence (VHD), a metric that quantifies the sensitivity of attention head outputs to visual context. Based on this, our findings reveal the presence of vision-aware attention heads that are more attuned to visual information; however, the model's overreliance on its prior language patterns is closely related to hallucinations. Building on these insights, we propose Vision-aware Head Reinforcement (VHR), a training-free approach to mitigate hallucination by enhancing the role of vision-aware attention heads. Extensive experiments demonstrate that our method achieves superior performance compared to state-of-the-art approaches in mitigating hallucinations, while maintaining high efficiency with negligible additional time overhead.

Paperid: 2383, https://arxiv.org/pdf/2410.16198.pdf

Abstract:
Chain-of-thought (CoT) reasoning in vision language models (VLMs) is crucial for improving interpretability and trustworthiness. However, current training recipes lack robust CoT reasoning data, relying on datasets dominated by short annotations with minimal rationales. In this work, we show that training VLM on short answers does not generalize well to reasoning tasks that require more detailed responses. To address this, we propose a two-fold approach. First, we distill rationales from GPT-4o model to enrich the training data and fine-tune VLMs, boosting their CoT performance. Second, we apply reinforcement learning to further calibrate reasoning quality. Specifically, we construct positive (correct) and negative (incorrect) pairs of model-generated reasoning chains, by comparing their predictions with annotated short answers. Using this pairwise data, we apply the Direct Preference Optimization algorithm to refine the model's reasoning abilities. Our experiments demonstrate significant improvements in CoT reasoning on benchmark datasets and better generalization to direct answer prediction as well. This work emphasizes the importance of incorporating detailed rationales in training and leveraging reinforcement learning to strengthen the reasoning capabilities of VLMs.

Paperid: 2384, https://arxiv.org/pdf/2410.02746.pdf

Abstract:
Contrastive Language-Image Pre-training (CLIP) has been a celebrated method for training vision encoders to generate image/text representations facilitating various applications. Recently, CLIP has been widely adopted as the vision backbone of multimodal large language models (MLLMs) to connect image inputs for language interactions. The success of CLIP as a vision-language foundation model relies on aligning web-crawled noisy text annotations at image levels. Nevertheless, such criteria may become insufficient for downstream tasks in need of fine-grained vision representations, especially when region-level understanding is demanding for MLLMs. In this paper, we improve the localization capability of CLIP with several advances. We propose a pre-training method called Contrastive Localized Language-Image Pre-training (CLOC) by complementing CLIP with region-text contrastive loss and modules. We formulate a new concept, promptable embeddings, of which the encoder produces image embeddings easy to transform into region representations given spatial hints. To support large-scale pre-training, we design a visually-enriched and spatially-localized captioning framework to effectively generate region-text pseudo-labels at scale. By scaling up to billions of annotated images, CLOC enables high-quality regional embeddings for image region recognition and retrieval tasks, and can be a drop-in replacement of CLIP to enhance MLLMs, especially on referring and grounding tasks.

Paperid: 2385, https://arxiv.org/pdf/2512.02536.pdf

Abstract:
Recent progress in multimodal large language models (MLLMs) has highlighted the challenge of efficiently bridging pre-trained Vision-Language Models (VLMs) with Diffusion Models. While methods using a fixed number of learnable query tokens offer computational efficiency, they suffer from task generalization collapse, failing to adapt to new tasks that are distant from their pre-training tasks. To overcome this, we propose Noisy Query Tokens, which learn a distributed representation space between the VLM and Diffusion Model via end-to-end optimization, enhancing continual learning. Additionally, we introduce a VAE branch with linear projection to recover fine-grained image details. Experimental results confirm our approach mitigates generalization collapse and enables stable continual learning across diverse tasks.

Paperid: 2386, https://arxiv.org/pdf/2512.02161.pdf

Abstract:
Text-to-image (T2I) models are capable of generating visually impressive images, yet they often fail to accurately capture specific attributes in user prompts, such as the correct number of objects with the specified colors. The diversity of such errors underscores the need for a hierarchical evaluation framework that can compare prompt adherence abilities of different image generation models. Simultaneously, benchmarks of vision language models (VLMs) have not kept pace with the complexity of scenes that VLMs are used to annotate. In this work, we propose a structured methodology for jointly evaluating T2I models and VLMs by testing whether VLMs can identify 27 specific failure modes in the images generated by T2I models conditioned on challenging prompts. Our second contribution is a dataset of prompts and images generated by 5 T2I models (Flux, SD3-Medium, SD3-Large, SD3.5-Medium, SD3.5-Large) and the corresponding annotations from VLMs (Molmo, InternVL3, Pixtral) annotated by an LLM (Llama3) to test whether VLMs correctly identify the failure mode in a generated image. By analyzing failure modes on a curated set of prompts, we reveal systematic errors in attribute fidelity and object representation. Our findings suggest that current metrics are insufficient to capture these nuanced errors, highlighting the importance of targeted benchmarks for advancing generative model reliability and interpretability.

Paperid: 2387, https://arxiv.org/pdf/2511.12267.pdf

Abstract:
Ultra-high-resolution (UHR) remote sensing (RS) images offer rich fine-grained information but also present challenges in effective processing. Existing dynamic resolution and token pruning methods are constrained by a passive perception paradigm, suffering from increased redundancy when obtaining finer visual inputs. In this work, we explore a new active perception paradigm that enables models to revisit information-rich regions. First, we present LRS-GRO, a large-scale benchmark dataset tailored for active perception in UHR RS processing, encompassing 17 question types across global, region, and object levels, annotated via a semi-automatic pipeline. Building on LRS-GRO, we propose ZoomEarth, an adaptive cropping-zooming framework with a novel Region-Guided reward that provides fine-grained guidance. Trained via supervised fine-tuning (SFT) and Group Relative Policy Optimization (GRPO), ZoomEarth achieves state-of-the-art performance on LRS-GRO and, in the zero-shot setting, on three public UHR remote sensing benchmarks. Furthermore, ZoomEarth can be seamlessly integrated with downstream models for tasks such as cloud removal, denoising, segmentation, and image editing through simple tool interfaces, demonstrating strong versatility and extensibility.

Paperid: 2388, https://arxiv.org/pdf/2508.12109.pdf

Abstract:
Multimodal Large Language Models (MLLMs) have shown impressive performance on vision-language tasks, but their long Chain-of-Thought (CoT) capabilities in multimodal scenarios remain underexplored. Inspired by OpenAI's o3 model, which emulates human-like ''thinking with image'' through iterative visual transformations and linguistic reasoning, we propose Simple o3, an end-to-end framework that integrates dynamic tool interactions (e.g., cropping, zooming, and reusing) into interleaved vision-language reasoning via supervised fine-tuning (SFT). Our approach features a scalable data synthesis pipeline that generates high-quality interleaved vision-language reasoning chains via an ''observe-reason-act'' cycle, complete with executable visual operations and rigorous verification, yielding the open-source TWI-Tools-146K dataset. Experimental results demonstrate Simple o3's superior performance on diverse benchmarks, outperforming existing approaches. By combining enhanced reasoning capabilities, Simple o3 establishes a powerful yet computationally affordable paradigm for advancing multimodal reasoning. Remarkably, we provide the first in-depth analysis of different interleaved reasoning strategies, offering insights into their impact on model performance. We found that by introducing additional visual tokens for interleaved vision-language reasoning, reusing and magnifying the original image significantly improves the model's visual reasoning and fine-grained perception, while image cropping based on precise visual grounding allows the model to effectively focus on key entities or regions, further enhancing its capabilities.

Paperid: 2389, https://arxiv.org/pdf/2508.08926.pdf

Abstract:
Large Vision-Language Models face growing safety challenges with multimodal inputs. This paper introduces the concept of Implicit Reasoning Safety, a vulnerability in LVLMs. Benign combined inputs trigger unsafe LVLM outputs due to flawed or hidden reasoning. To showcase this, we developed Safe Semantics, Unsafe Interpretations, the first dataset for this critical issue. Our demonstrations show that even simple In-Context Learning with SSUI significantly mitigates these implicit multimodal threats, underscoring the urgent need to improve cross-modal implicit reasoning.

Paperid: 2390, https://arxiv.org/pdf/2508.03142.pdf

Abstract:
In recent years, unified vision-language models (VLMs) have rapidly advanced, effectively tackling both visual understanding and generation tasks within a single design. While many unified VLMs have explored various design choices, the recent hypothesis from OpenAI's GPT-4o suggests a promising generation pipeline: Understanding VLM->Visual Feature->Projector->Diffusion Model->Image. The understanding VLM is frozen, and only the generation-related modules are trained. This pipeline maintains the strong capability of understanding VLM while enabling the image generation ability of the unified VLM. Although this pipeline has shown very promising potential for the future development of unified VLM, how to easily enable image editing capability is still unexplored. In this paper, we introduce a novel training-free framework named UniEdit-I to enable the unified VLM with image editing capability via three iterative steps: understanding, editing, and verifying. 1. The understanding step analyzes the source image to create a source prompt through structured semantic analysis and makes minimal word replacements to form the target prompt based on the editing instruction. 2. The editing step introduces a time-adaptive offset, allowing for coherent editing from coarse to fine throughout the denoising process. 3. The verification step checks the alignment between the target prompt and the intermediate edited image, provides automatic consistency scores and corrective feedback, and determines whether to stop early or continue the editing loop. This understanding, editing, and verifying loop iterates until convergence, delivering high-fidelity editing in a training-free manner. We implemented our method based on the latest BLIP3-o and achieved state-of-the-art (SOTA) performance on the GEdit-Bench benchmark.

Paperid: 2391, https://arxiv.org/pdf/2508.01016.pdf

Abstract:
This retrospective study evaluated five VLMs (Qwen2.5, Phi-4, Gemma3, Llama3.2, and Mistral3.1) using the MedFMC dataset. This dataset includes 22,349 images from 7,461 patients encompassing chest radiography (19 disease multi-label classifications), colon pathology (tumor detection), endoscopy (colorectal lesion identification), neonatal jaundice assessment (skin color-based treatment necessity), and retinal fundoscopy (5-point diabetic retinopathy grading). Diagnostic accuracy was compared in three experimental settings: visual input only, multimodal input, and chain-of-thought reasoning. Model accuracy was assessed against ground truth labels, with statistical comparisons using bootstrapped confidence intervals (p<.05). Qwen2.5 achieved the highest accuracy for chest radiographs (90.4%) and endoscopy images (84.2%), significantly outperforming the other models (p<.001). In colon pathology, Qwen2.5 (69.0%) and Phi-4 (69.6%) performed comparably (p=.41), both significantly exceeding other VLMs (p<.001). Similarly, for neonatal jaundice assessment, Qwen2.5 (58.3%) and Phi-4 (58.1%) showed comparable leading accuracies (p=.93) significantly exceeding their counterparts (p<.001). All models struggled with retinal fundoscopy; Qwen2.5 and Gemma3 achieved the highest, albeit modest, accuracies at 18.6% (comparable, p=.99), significantly better than other tested models (p<.001). Unexpectedly, multimodal input reduced accuracy for some models and modalities, and chain-of-thought reasoning prompts also failed to improve accuracy. The open-source VLMs demonstrated promising diagnostic capabilities, particularly in chest radiograph interpretation. However, performance in complex domains such as retinal fundoscopy was limited, underscoring the need for further development and domain-specific adaptation before widespread clinical application.

Paperid: 2392, https://arxiv.org/pdf/2507.23318.pdf

Abstract:
Vision-Language-Action (VLA) models have demonstrated significant potential in complex scene understanding and action reasoning, leading to their increasing adoption in end-to-end autonomous driving systems. However, the long visual tokens of VLA models greatly increase computational costs. Current visual token pruning methods in Vision-Language Models (VLM) rely on either visual token similarity or visual-text attention, but both have shown poor performance in autonomous driving scenarios. Given that human drivers concentrate on relevant foreground areas while driving, we assert that retaining visual tokens containing this foreground information is essential for effective decision-making. Inspired by this, we propose FastDriveVLA, a novel reconstruction-based vision token pruning framework designed specifically for autonomous driving. FastDriveVLA includes a plug-and-play visual token pruner called ReconPruner, which prioritizes foreground information through MAE-style pixel reconstruction. A novel adversarial foreground-background reconstruction strategy is designed to train ReconPruner for the visual encoder of VLA models. Once trained, ReconPruner can be seamlessly applied to different VLA models with the same visual encoder without retraining. To train ReconPruner, we also introduce a large-scale dataset called nuScenes-FG, consisting of 241K image-mask pairs with annotated foreground regions. Our approach achieves state-of-the-art results on the nuScenes open-loop planning benchmark across different pruning ratios.

Paperid: 2393, https://arxiv.org/pdf/2507.16746.pdf

Abstract:
Humans often use visual aids, for example diagrams or sketches, when solving complex problems. Training multimodal models to do the same, known as Visual Chain of Thought (Visual CoT), is challenging due to: (1) poor off-the-shelf visual CoT performance, which hinders reinforcement learning, and (2) the lack of high-quality visual CoT training data. We introduce $\textbf{Zebra-CoT}$, a diverse large-scale dataset with 182,384 samples, containing logically coherent interleaved text-image reasoning traces. We focus on four categories of tasks where sketching or visual reasoning is especially natural, spanning scientific questions such as geometry, physics, and algorithms; 2D visual reasoning tasks like visual search and jigsaw puzzles; 3D reasoning tasks including 3D multi-hop inference, embodied and robot planning; visual logic problems and strategic games like chess. Fine-tuning the Anole-7B model on the Zebra-CoT training corpus results in an improvement of +12% in our test-set accuracy and yields up to +13% performance gain on standard VLM benchmark evaluations. Fine-tuning Bagel-7B yields a model that generates high-quality interleaved visual reasoning chains, underscoring Zebra-CoT's effectiveness for developing multimodal reasoning abilities. We open-source our dataset and models to support development and evaluation of visual CoT.

Paperid: 2394, https://arxiv.org/pdf/2507.05970.pdf

Abstract:
As a challenging vision-language (VL) task, Composed Image Retrieval (CIR) aims to retrieve target images using multimodal (image+text) queries. Although many existing CIR methods have attained promising performance, their reliance on costly, manually labeled triplets hinders scalability and zero-shot capability. To address this issue, we propose a scalable pipeline for automatic triplet generation, along with a fully synthetic dataset named Composed Image Retrieval on High-quality Synthetic Triplets (CIRHS). Our pipeline leverages a large language model (LLM) to generate diverse prompts, controlling a text-to-image generative model to produce image pairs with identical elements in each pair, which are then filtered and reorganized to form the CIRHS dataset. In addition, we introduce Hybrid Contextual Alignment (CoAlign), a novel CIR framework, which can accomplish global alignment and local reasoning within a broader context, enabling the model to learn more robust and informative representations. By utilizing the synthetic CIRHS dataset, CoAlign achieves outstanding zero-shot performance on three commonly used benchmarks, demonstrating for the first time the feasibility of training CIR models on a fully synthetic dataset. Furthermore, under supervised training, our method outperforms all the state-of-the-art supervised CIR approaches, validating the effectiveness of our proposed retrieval framework. The code and the CIRHS dataset will be released soon.

Paperid: 2395, https://arxiv.org/pdf/2506.16112.pdf

Abstract:
Inspired by text prompts in large language models (LLMs), visual prompts have been explored to enhance the reasoning capabilities of large vision-language models (LVLMs). Current methods design heuristic visual prompts, such as overlaying a text-query-guided attention heatmap on the original input image. However, designing effective prompts manually is challenging and time-consuming, and it often fails to explore the benefits of different visual prompts, leading to sub-optimal performance. To this end, we propose \textbf{AutoV} that learns to automatically select the optimal visual prompt from various candidates based on given textual queries and the input image. To train AutoV, we developed an automatic data collection and labeling pipeline that evaluates various visual prompts with a pre-trained LVLM. We input a set of visual prompts into the LVLM and rank them according to the prediction losses generated by the model. Using the ranking as a supervision signal, we train AutoV to automatically choose the optimal visual prompt from various visual prompts for LVLMs. Experimental results indicate that AutoV enhances the performance of various LVLMs across multiple popular image understanding tasks. For instance, LLaVA-OV with AutoV achieves $\textbf{1.7}\%$ accuracy gain on LLaVA$^{\text{Wild}}$, and AutoV boosts Qwen2.5-VL by $\textbf{1.9}\%$ on MMMU, highlighting its potential as an optimal visual prompting method for LVLMs.

Paperid: 2396, https://arxiv.org/pdf/2506.04606.pdf

Abstract:
SmartAvatar is a vision-language-agent-driven framework for generating fully rigged, animation-ready 3D human avatars from a single photo or textual prompt. While diffusion-based methods have made progress in general 3D object generation, they continue to struggle with precise control over human identity, body shape, and animation readiness. In contrast, SmartAvatar leverages the commonsense reasoning capabilities of large vision-language models (VLMs) in combination with off-the-shelf parametric human generators to deliver high-quality, customizable avatars. A key innovation is an autonomous verification loop, where the agent renders draft avatars, evaluates facial similarity, anatomical plausibility, and prompt alignment, and iteratively adjusts generation parameters for convergence. This interactive, AI-guided refinement process promotes fine-grained control over both facial and body features, enabling users to iteratively refine their avatars via natural-language conversations. Unlike diffusion models that rely on static pre-trained datasets and offer limited flexibility, SmartAvatar brings users into the modeling loop and ensures continuous improvement through an LLM-driven procedural generation and verification system. The generated avatars are fully rigged and support pose manipulation with consistent identity and appearance, making them suitable for downstream animation and interactive applications. Quantitative benchmarks and user studies demonstrate that SmartAvatar outperforms recent text- and image-driven avatar generation systems in terms of reconstructed mesh quality, identity fidelity, attribute accuracy, and animation readiness, making it a versatile tool for realistic, customizable avatar creation on consumer-grade hardware.

Paperid: 2397, https://arxiv.org/pdf/2506.00534.pdf

Abstract:
The choice of a suitable visual language projector (VLP) is critical to the successful training of large visual language models (LVLMs). Mainstream VLPs can be broadly categorized into compressed and uncompressed projectors, and each offering distinct advantages in performance and computational efficiency. However, their security implications have not been thoroughly examined. Our comprehensive evaluation reveals significant differences in their security profiles: compressed projectors exhibit substantial vulnerabilities, allowing adversaries to successfully compromise LVLMs even with minimal knowledge of structural information. In stark contrast, uncompressed projectors demonstrate robust security properties and do not introduce additional vulnerabilities. These findings provide critical guidance for researchers in selecting optimal VLPs that enhance the security and reliability of visual language models. The code will be released.

Paperid: 2398, https://arxiv.org/pdf/2505.17931.pdf

Abstract:
Medical image segmentation is vital for clinical diagnosis, yet current deep learning methods often demand extensive expert effort, i.e., either through annotating large training datasets or providing prompts at inference time for each new case. This paper introduces a zero-shot and automatic segmentation pipeline that combines off-the-shelf vision-language and segmentation foundation models. Given a medical image and a task definition (e.g., "segment the optic disc in an eye fundus image"), our method uses a grounding model to generate an initial bounding box, followed by a visual prompt boosting module that enhance the prompts, which are then processed by a promptable segmentation model to produce the final mask. To address the challenges of domain gap and result verification, we introduce a test-time adaptation framework featuring a set of learnable adaptors that align the medical inputs with foundation model representations. Its hyperparameters are optimized via Bayesian Optimization, guided by a proxy validation model without requiring ground-truth labels. Our pipeline offers an annotation-efficient and scalable solution for zero-shot medical image segmentation across diverse tasks. Our pipeline is evaluated on seven diverse medical imaging datasets and shows promising results. By proper decomposition and test-time adaptation, our fully automatic pipeline performs competitively with weakly-prompted interactive foundation models.

Paperid: 2399, https://arxiv.org/pdf/2505.15963.pdf

Abstract:
Large vision-language models (LVLMs) remain vulnerable to hallucination, often generating content misaligned with visual inputs. While recent approaches advance multi-modal Direct Preference Optimization (DPO) to mitigate hallucination, they typically rely on predefined or randomly edited negative samples that fail to reflect actual model errors, limiting training efficacy. In this work, we propose an Online Vision-language Preference Learning (OViP) framework that dynamically constructs contrastive training data based on the model's own hallucinated outputs. By identifying semantic differences between sampled response pairs and synthesizing negative images using a diffusion model, OViP generates more relevant supervision signals in real time. This failure-driven training enables adaptive alignment of both textual and visual preferences. Moreover, we refine existing evaluation protocols to better capture the trade-off between hallucination suppression and expressiveness. Experiments on hallucination and general benchmarks demonstrate that OViP effectively reduces hallucinations while preserving core multi-modal capabilities.

Paperid: 2400, https://arxiv.org/pdf/2505.11350.pdf

Abstract:
To perform outdoor autonomous visual navigation and search, a robot may leverage satellite imagery as a prior map. This can help inform high-level search and exploration strategies, even when such images lack sufficient resolution to allow for visual recognition of targets. However, there are limited training datasets of satellite images with annotated targets that are not directly visible. Furthermore, approaches which leverage large Vision Language Models (VLMs) for generalization may yield inaccurate outputs due to hallucination, leading to inefficient search. To address these challenges, we introduce Search-TTA, a multimodal test-time adaptation framework with a flexible plug-and-play interface compatible with various input modalities (e.g. image, text, sound) and planning methods. First, we pretrain a satellite image encoder to align with CLIP's visual encoder to output probability distributions of target presence used for visual search. Second, our framework dynamically refines CLIP's predictions during search using a test-time adaptation mechanism. Through a novel feedback loop inspired by Spatial Poisson Point Processes, uncertainty-weighted gradient updates are used to correct potentially inaccurate predictions and improve search performance. To train and evaluate Search-TTA, we curate AVS-Bench, a visual search dataset based on internet-scale ecological data that contains up to 380k training and 8k validation images (in- and out-domain). We find that Search-TTA improves planner performance by up to 30.0%, particularly in cases with poor initial CLIP predictions due to limited training data. It also performs comparably with significantly larger VLMs, and achieves zero-shot generalization to unseen modalities. Finally, we deploy Search-TTA on a real UAV via hardware-in-the-loop testing, by simulating its operation within a large-scale simulation that provides onboard sensing.

Paperid: 2401, https://arxiv.org/pdf/2504.15037.pdf

Abstract:
Multimodal Large Language Models (MLLMs) have demonstrated impressive performance in general vision-language tasks. However, recent studies have exposed critical limitations in their spatial reasoning capabilities. This deficiency in spatial reasoning significantly constrains MLLMs' ability to interact effectively with the physical world, thereby limiting their broader applications. We argue that spatial reasoning capabilities will not naturally emerge from merely scaling existing architectures and training methodologies. Instead, this challenge demands dedicated attention to fundamental modifications in the current MLLM development approach. In this position paper, we first establish a comprehensive framework for spatial reasoning within the context of MLLMs. We then elaborate on its pivotal role in real-world applications. Through systematic analysis, we examine how individual components of the current methodology, from training data to reasoning mechanisms, influence spatial reasoning capabilities. This examination reveals critical limitations while simultaneously identifying promising avenues for advancement. Our work aims to direct the AI research community's attention toward these crucial yet underexplored aspects. By highlighting these challenges and opportunities, we seek to catalyze progress toward achieving human-like spatial reasoning capabilities in MLLMs.

Paperid: 2402, https://arxiv.org/pdf/2504.14848.pdf

Abstract:
Vision-language models (VLMs) excel in various multimodal tasks but frequently suffer from poor calibration, resulting in misalignment between their verbalized confidence and response correctness. This miscalibration undermines user trust, especially when models confidently provide incorrect or fabricated information. In this work, we propose a novel Confidence Calibration through Semantic Perturbation (CSP) framework to improve the calibration of verbalized confidence for VLMs in response to object-centric queries. We first introduce a perturbed dataset where Gaussian noise is applied to the key object regions to simulate visual uncertainty at different confidence levels, establishing an explicit mapping between visual ambiguity and confidence levels. We further enhance calibration through a two-stage training process combining supervised fine-tuning on the perturbed dataset with subsequent preference optimization. Extensive experiments on popular benchmarks demonstrate that our method significantly improves the alignment between verbalized confidence and response correctness while maintaining or enhancing overall task performance. These results highlight the potential of semantic perturbation as a practical tool for improving the reliability and interpretability of VLMs.

Paperid: 2403, https://arxiv.org/pdf/2503.05204.pdf

Abstract:
Zero-shot Composed Image Retrieval (ZS-CIR) aims to retrieve the target image based on a reference image and a text description without requiring in-distribution triplets for training. One prevalent approach follows the vision-language pretraining paradigm that employs a mapping network to transfer the image embedding to a pseudo-word token in the text embedding space. However, this approach tends to impede network generalization due to modality discrepancy and distribution shift between training and inference. To this end, we propose a Data-efficient Generalization (DeG) framework, including two novel designs, namely, Textual Supplement (TS) module and Semantic-Set (S-Set). The TS module exploits compositional textual semantics during training, enhancing the pseudo-word token with more linguistic semantics and thus mitigating the modality discrepancy effectively. The S-Set exploits the zero-shot capability of pretrained Vision-Language Models (VLMs), alleviating the distribution shift and mitigating the overfitting issue from the redundancy of the large-scale image-text data. Extensive experiments over four ZS-CIR benchmarks show that DeG outperforms the state-of-the-art (SOTA) methods with much less training data, and saves substantial training and inference time for practical usage.

Paperid: 2404, https://arxiv.org/pdf/2502.14486.pdf

Abstract:
Jailbreak attacks, where harmful prompts bypass generative models' built-in safety, raise serious concerns about model vulnerability. While many defense methods have been proposed, the trade-offs between safety and helpfulness, and their application to Large Vision-Language Models (LVLMs), are not well understood. This paper systematically examines jailbreak defenses by reframing the standard generation task as a binary classification problem to assess model refusal tendencies for both harmful and benign queries. We identify two key defense mechanisms: safety shift, which increases refusal rates across all queries, and harmfulness discrimination, which improves the model's ability to distinguish between harmful and benign inputs. Using these mechanisms, we develop two ensemble defense strategies-inter-mechanism ensembles and intra-mechanism ensembles-to balance safety and helpfulness. Experiments on the MM-SafetyBench and MOSSBench datasets with LLaVA-1.5 models show that these strategies effectively improve model safety or optimize the trade-off between safety and helpfulness.

Paperid: 2405, https://arxiv.org/pdf/2412.12785.pdf

Abstract:
Large Vision-Language Models (LVLMs) typically learn visual capacity through visual instruction tuning, involving updates to both a projector and their LLM backbones. Inspired by the concept of a visual region in the human brain, we investigate the existence of an analogous \textit{visual region} within LLMs that functions as a cognitive core, and explore the potential of efficient training of LVLMs via selective layers tuning. Using Bunny-Llama-3-8B-V for detailed analysis and other three LVLMs for validation across diverse visual and textual tasks, we find that selectively updating 25\% of LLMs layers, when sparsely and uniformly distributed, can preserve nearly 99\% of visual performance and maintain or improve textual task results, while effectively reducing training time. Based on this targeted training approach, we further propose a novel visual region-based pruning paradigm, removing non-critical layers outside the visual region, which can achieve minimal performance loss. This study offers an effective and efficient strategy for LVLM training and inference by activating a layer-wise visual region within LLMs, which proves consistently effective across different models.

Paperid: 2406, https://arxiv.org/pdf/2411.16053.pdf

Abstract:
Vision-and-Language Navigation (VLN), where an agent follows instructions to reach a target destination, has recently seen significant advancements. In contrast to navigation in discrete environments with predefined trajectories, VLN in Continuous Environments (VLN-CE) presents greater challenges, as the agent is free to navigate any unobstructed location and is more vulnerable to visual occlusions or blind spots. Recent approaches have attempted to address this by imagining future environments, either through predicted future visual images or semantic features, rather than relying solely on current observations. However, these RGB-based and feature-based methods lack intuitive appearance-level information or high-level semantic complexity crucial for effective navigation. To overcome these limitations, we introduce a novel, generalizable 3DGS-based pre-training paradigm, called UnitedVLN, which enables agents to better explore future environments by unitedly rendering high-fidelity 360 visual images and semantic features. UnitedVLN employs two key schemes: search-then-query sampling and separate-then-united rendering, which facilitate efficient exploitation of neural primitives, helping to integrate both appearance and semantic information for more robust navigation. Extensive experiments demonstrate that UnitedVLN outperforms state-of-the-art methods on existing VLN-CE benchmarks.

Paperid: 2407, https://arxiv.org/pdf/2410.23736.pdf

Abstract:
As a challenging vision-language task, Zero-Shot Composed Image Retrieval (ZS-CIR) is designed to retrieve target images using bi-modal (image+text) queries. Typical ZS-CIR methods employ an inversion network to generate pseudo-word tokens that effectively represent the input semantics. However, the inversion-based methods suffer from two inherent issues: First, the task discrepancy exists because inversion training and CIR inference involve different objectives. Second, the modality discrepancy arises from the input feature distribution mismatch between training and inference. To this end, we propose a lightweight post-hoc framework, consisting of two components: (1) A new text-anchored triplet construction pipeline leverages a large language model (LLM) to transform a standard image-text dataset into a triplet dataset, where a textual description serves as the target of each triplet. (2) The MoTa-Adapter, a novel parameter-efficient fine-tuning method, adapts the dual encoder to the CIR task using our constructed triplet data. Specifically, on the text side, multiple sets of learnable task prompts are integrated via a Mixture-of-Experts (MoE) layer to capture task-specific priors and handle different types of modifications. On the image side, MoTa-Adapter modulates the inversion network's input to better match the downstream text encoder. In addition, an entropy-based optimization strategy is proposed to assign greater weight to challenging samples, thus ensuring efficient adaptation. Experiments show that, with the incorporation of our proposed components, inversion-based methods achieve significant improvements, reaching state-of-the-art performance across four widely-used benchmarks. All data and code will be made publicly available.

Paperid: 2408, https://arxiv.org/pdf/2601.08139.pdf

Abstract:
Vision-language models (VLMs), despite their extraordinary zero-shot capabilities, are vulnerable to distribution shifts. Test-time adaptation (TTA) emerges as a predominant strategy to adapt VLMs to unlabeled test data on the fly. However, existing TTA methods heavily rely on zero-shot predictions as pseudo-labels for self-training, which can be unreliable under distribution shifts and misguide adaptation due to two fundamental limitations. First (Modality Gap), distribution shifts induce gaps between visual and textual modalities, making cross-modal relations inaccurate. Second (Visual Nuisance), visual embeddings encode rich but task-irrelevant noise that often overwhelms task-specific semantics under distribution shifts. To address these limitations, we propose SubTTA, which aligns the semantic subspaces of both modalities to enhance zero-shot predictions to better guide the TTA process. To bridge the modality gap, SubTTA extracts the principal subspaces of both modalities and aligns the visual manifold to the textual semantic anchor by minimizing their chordal distance. To eliminate visual nuisance, SubTTA projects the aligned visual features onto the task-specific textual subspace, which filters out task-irrelevant noise by constraining visual embeddings within the valid semantic span, and standard TTA is further performed on the purified space to refine the decision boundaries. Extensive experiments on various benchmarks and VLM architectures demonstrate the effectiveness of SubTTA, yielding an average improvement of 2.24% over state-of-the-art TTA methods.

Paperid: 2409, https://arxiv.org/pdf/2512.16864.pdf

Abstract:
Instruction-based image editing enables natural-language control over visual modifications, yet existing models falter under Instruction-Visual Complexity (IV-Complexity), where intricate instructions meet cluttered or ambiguous scenes. We introduce RePlan (Region-aligned Planning), a plan-then-execute framework that couples a vision-language planner with a diffusion editor. The planner decomposes instructions via step-by-step reasoning and explicitly grounds them to target regions; the editor then applies changes using a training-free attention-region injection mechanism, enabling precise, parallel multi-region edits without iterative inpainting. To strengthen planning, we apply GRPO-based reinforcement learning using 1K instruction-only examples, yielding substantial gains in reasoning fidelity and format reliability. We further present IV-Edit, a benchmark focused on fine-grained grounding and knowledge-intensive edits. Across IV-Complex settings, RePlan consistently outperforms strong baselines trained on far larger datasets, improving regional precision and overall fidelity. Our project page: https://replan-iv-edit.github.io

Paperid: 2410, https://arxiv.org/pdf/2512.04763.pdf

Abstract:
Memory-augmented Large Language Models (LLMs) have demonstrated remarkable consistency during prolonged dialogues by storing relevant memories and incorporating them as context. Such memory-based personalization is also key in on-device settings that allow users to keep their conversations and data private. However, memory-augmented systems typically rely on LLMs that are too costly for local on-device deployment. Even though Small Language Models (SLMs) are more suitable for on-device inference than LLMs, they cannot achieve sufficient performance. Additionally, these LLM-based systems lack native visual capabilities, limiting their applicability in multimodal contexts. In this paper, we introduce (i) MemLoRA, a novel memory system that enables local deployment by equipping SLMs with specialized memory adapters, and (ii) its vision extension MemLoRA-V, which integrates small Vision-Language Models (SVLMs) to memory systems, enabling native visual understanding. Following knowledge distillation principles, each adapter is trained separately for specific memory operations$\unicode{x2013}$knowledge extraction, memory update, and memory-augmented generation. Equipped with memory adapters, small models enable accurate on-device memory operations without cloud dependency. On text-only operations, MemLoRA outperforms 10$\times$ larger baseline models (e.g., Gemma2-27B) and achieves performance comparable to 60$\times$ larger models (e.g., GPT-OSS-120B) on the LoCoMo benchmark. To evaluate visual understanding operations instead, we extend LoCoMo with challenging Visual Question Answering tasks that require direct visual reasoning. On this, our VLM-integrated MemLoRA-V shows massive improvements over caption-based approaches (81.3 vs. 23.7 accuracy) while keeping strong performance in text-based tasks, demonstrating the efficacy of our method in multimodal contexts.

Paperid: 2411, https://arxiv.org/pdf/2512.04585.pdf

Abstract:
Segment Anything Model 3 (SAM3) has advanced open-vocabulary segmentation through promptable concept segmentation, allowing users to segment all instances corresponding to a given concept, typically specified with short noun-phrase (NP) prompts. While this marks the first integration of language-level concepts within the SAM family, real-world usage typically requires far richer expressions that include attributes, spatial relations, functionalities, actions, states, and even implicit reasoning over instances. Currently, SAM3 relies on external multi-modal agents to convert complex instructions into NPs and then conduct iterative mask filtering. However, these NP-level concepts remain overly coarse, often failing to precisely represent a specific instance. In this work, we present SAM3-I, an enhanced framework that unifies concept-level understanding and instruction-level reasoning within the SAM family. SAM3-I introduces an instruction-aware cascaded adaptation mechanism that progressively aligns expressive instruction semantics with SAM3's existing vision-language representations, enabling direct instruction-following segmentation without sacrificing its original concept-driven capabilities. Furthermore, we design a structured instruction taxonomy spanning concept, simple, and complex levels, and develop a scalable data engine to construct a dataset with diverse instruction-mask pairs. Experiments show that SAM3-I delivers appealing performance, demonstrating that SAM3 can be effectively extended to follow natural-language instructions while preserving its strong concept grounding. We open-source SAM3-I and provide practical fine-tuning workflows, enabling researchers to adapt it to domain-specific applications. The source code is available here.

Paperid: 2412, https://arxiv.org/pdf/2512.04554.pdf

Abstract:
Document Visual Question Answering (DocVQA) enables end-to-end reasoning grounded on information present in a document input. While recent models have shown impressive capabilities, they remain vulnerable to adversarial attacks. In this work, we introduce a novel attack scenario that aims to forge document content in a visually imperceptible yet semantically targeted manner, allowing an adversary to induce specific or generally incorrect answers from a DocVQA model. We develop specialized attack algorithms that can produce adversarially forged documents tailored to different attackers' goals, ranging from targeted misinformation to systematic model failure scenarios. We demonstrate the effectiveness of our approach against two end-to-end state-of-the-art models: Pix2Struct, a vision-language transformer that jointly processes image and text through sequence-to-sequence modeling, and Donut, a transformer-based model that directly extracts text and answers questions from document images. Our findings highlight critical vulnerabilities in current DocVQA systems and call for the development of more robust defenses.

Paperid: 2413, https://arxiv.org/pdf/2511.22586.pdf

Abstract:
We study how different Chain-of-Thought (CoT) designs affect the acquisition of the generalizable visual reasoning ability in vision-language models (VLMs). While CoT data, especially long or visual CoT such as "think with image", has been widely used to supervise intermediate reasoning, it remains unclear why specific CoT designs help and which ones truly support generalizable reasoning. To systematically evaluate this, we focus on a controlled maze-solving benchmark where reasoning rules are fully visual, difficulty can be tuned by grid size, and all the intermediate steps can be automatically generated. Using Qwen2.5-VL-7B under a standard SFT-then-RL pipeline, we compare three representative CoT formats: Language CoT, Grounding CoT (with spatial coordinate trajectories), and Visual CoT (with image manipulations). Our experiments reveal that visual and longer CoT mainly accelerate convergence but do not lift the final performance ceiling; concise CoT containing only essential grounding steps outperforms longer traces; and, strikingly, CoT retaining only the minimal grounding results generalizes best across different maze sizes. We further validate these insights on other vision-centric tasks. These findings highlight a "short is long" effect and provide practical guidance for constructing more generalizable SFT datasets for visual reasoning.

Paperid: 2414, https://arxiv.org/pdf/2510.16416.pdf

Abstract:
Vision-language models (VLMs) have shown remarkable abilities by integrating large language models with visual inputs. However, they often fail to utilize visual evidence adequately, either depending on linguistic priors in vision-centric tasks or resorting to textual shortcuts during reasoning. Although reinforcement learning (RL) can align models with desired behaviors, its application to VLMs has been hindered by the lack of scalable and reliable reward mechanisms. To overcome this challenge, we propose SSL4RL, a novel framework that leverages self-supervised learning (SSL) tasks as a source of verifiable rewards for RL-based fine-tuning. Our approach reformulates SSL objectives-such as predicting image rotation or reconstructing masked patches-into dense, automatic reward signals, eliminating the need for human preference data or unreliable AI evaluators. Experiments show that SSL4RL substantially improves performance on both vision-centric and vision-language reasoning benchmarks. Furthermore, through systematic ablations, we identify key factors-such as task difficulty, model scale, and semantic alignment with the target domain-that influence the effectiveness of SSL4RL tasks, offering new design principles for future work. We also demonstrate the framework's generality by applying it to graph learning, where it yields significant gains. SSL4RL establishes a versatile and effective paradigm for aligning multimodal models using verifiable, self-supervised objectives.

Paperid: 2415, https://arxiv.org/pdf/2510.08964.pdf

Abstract:
Recent advances in inference-time scaling, particularly those leveraging reinforcement learning with verifiable rewards, have substantially enhanced the reasoning capabilities of Large Vision-Language Models (LVLMs). Inspired by this success, similar strategies have been applied to multimodal reasoning, yet their impact on visual perception remains unclear. To investigate this gap, we introduce DisTANCE, a perception-centric benchmark for visual estimation tasks. Evaluation results show that LVLMs exhibit limited estimation precision, and inference-time scaling offers only marginal gains. We attribute this to the fast perception paradigm of current LVLMs, where visual understanding is treated as a one-shot output without modeling the underlying perceptual process. To address this, we propose Perception-Time Scaling (PTS), a novel paradigm that encourages token-rich perception and decomposes complex perception problems into intermediate tractable sub-problems, thereby enabling perception to align with and benefit from inference-time scaling. Combined with reinforcement learning techniques, PTS significantly improves perception accuracy, raising high-precision performance on DisTANCE from 8.0% to 64.7%, and generalizes well to out-of-domain tasks. Surprisingly, even though PTS data are purely synthetic, combining them with math reasoning data yields consistent gains in both reasoning and real-world perception benchmarks. Further analysis reveals that PTS introduces more perception-related tokens and increases the model's attention to image tokens. Our code and data will be publicly released.

Paperid: 2416, https://arxiv.org/pdf/2509.23652.pdf

Abstract:
While Reinforcement Learning with Verifiable Reward (RLVR) significantly advances image reasoning in Large Vision-Language Models (LVLMs), its application to complex video reasoning remains underdeveloped. This gap stems primarily from a critical data bottleneck: existing datasets lack the challenging, multi-hop questions and high-quality, video-grounded Chain-of-Thought (CoT) data necessary to effectively bootstrap RLVR. To address this, we introduce ReWatch, a large-scale dataset built to foster advanced video reasoning. We propose a novel multi-stage synthesis pipeline to synthesize its three components: ReWatch-Caption, ReWatch-QA, and ReWatch-CoT. A core innovation is our Multi-Agent ReAct framework for CoT synthesis, which simulates a human-like "re-watching" process to generate video-grounded reasoning traces by explicitly modeling information retrieval and verification. Building on this dataset, we develop ReWatch-R1 by post-training a strong baseline LVLM with Supervised Fine-Tuning (SFT) and our RLVR framework. This framework incorporates a novel Observation \& Reasoning (O\&R) reward mechanism that evaluates both the final answer's correctness and the reasoning's alignment with video content, directly penalizing hallucination. Our experiments show that ReWatch-R1 achieves state-of-the-art average performance on five challenging video reasoning benchmarks. Project Page: https://rewatch-r1.github.io

Paperid: 2417, https://arxiv.org/pdf/2509.14001.pdf

Abstract:
We introduce MOCHA (Multi-modal Objects-aware Cross-arcHitecture Alignment), a knowledge distillation approach that transfers region-level multimodal semantics from a large vision-language teacher (e.g., LLaVa) into a lightweight vision-only object detector student (e.g., YOLO). A translation module maps student features into a joint space, where the training of the student and translator is guided by a dual-objective loss that enforces both local alignment and global relational consistency. Unlike prior approaches focused on dense or global alignment, MOCHA operates at the object level, enabling efficient transfer of semantics without modifying the teacher or requiring textual input at inference. We validate our method across four personalized detection benchmarks under few-shot regimes. Results show consistent gains over baselines, with a +10.1 average score improvement. Despite its compact architecture, MOCHA reaches performance on par with larger multimodal models, proving its suitability for real-world deployment.

Paperid: 2418, https://arxiv.org/pdf/2508.19679.pdf

Abstract:
Recent advances in Vision-Language Models (VLMs) have enabled mobile agents to perceive and interact with real-world mobile environments based on human instructions. However, the current fully autonomous paradigm poses potential safety risks when model understanding or reasoning capabilities are insufficient. To address this challenge, we first introduce \textbf{InquireBench}, a comprehensive benchmark specifically designed to evaluate mobile agents' capabilities in safe interaction and proactive inquiry with users, encompassing 5 categories and 22 sub-categories, where most existing VLM-based agents demonstrate near-zero performance. In this paper, we aim to develop an interactive system that actively seeks human confirmation at critical decision points. To achieve this, we propose \textbf{InquireMobile}, a novel model inspired by reinforcement learning, featuring a two-stage training strategy and an interactive pre-action reasoning mechanism. Finally, our model achieves an 46.8% improvement in inquiry success rate and the best overall success rate among existing baselines on InquireBench. We will open-source all datasets, models, and evaluation codes to facilitate development in both academia and industry.

Paperid: 2419, https://arxiv.org/pdf/2508.05405.pdf

Abstract:
Although Vision Language Models (VLMs) exhibit strong perceptual abilities and impressive visual reasoning, they struggle with attention to detail and precise action planning in complex, dynamic environments, leading to subpar performance. Real-world tasks typically require complex interactions, advanced spatial reasoning, long-term planning, and continuous strategy refinement, usually necessitating understanding the physics rules of the target scenario. However, evaluating these capabilities in real-world scenarios is often prohibitively expensive. To bridge this gap, we introduce DeepPHY, a novel benchmark framework designed to systematically evaluate VLMs' understanding and reasoning about fundamental physical principles through a series of challenging simulated environments. DeepPHY integrates multiple physical reasoning environments of varying difficulty levels and incorporates fine-grained evaluation metrics. Our evaluation finds that even state-of-the-art VLMs struggle to translate descriptive physical knowledge into precise, predictive control.

Paperid: 2420, https://arxiv.org/pdf/2505.24227.pdf

Abstract:
While adversarial attacks on vision-and-language pretraining (VLP) models have been explored, generating natural adversarial samples crafted through realistic and semantically meaningful perturbations remains an open challenge. Existing methods, primarily designed for classification tasks, struggle when adapted to VLP models due to their restricted optimization spaces, leading to ineffective attacks or unnatural artifacts. To address this, we propose \textbf{LightD}, a novel framework that generates natural adversarial samples for VLP models via semantically guided relighting. Specifically, LightD leverages ChatGPT to propose context-aware initial lighting parameters and integrates a pretrained relighting model (IC-light) to enable diverse lighting adjustments. LightD expands the optimization space while ensuring perturbations align with scene semantics. Additionally, gradient-based optimization is applied to the reference lighting image to further enhance attack effectiveness while maintaining visual naturalness. The effectiveness and superiority of the proposed LightD have been demonstrated across various VLP models in tasks such as image captioning and visual question answering.

Paperid: 2421, https://arxiv.org/pdf/2505.18053.pdf

Abstract:
Prompt learning as a parameter-efficient method that has been widely adopted to adapt Vision-Language Models (VLMs) to downstream tasks. While hard-prompt design requires domain expertise and iterative optimization, soft-prompt methods rely heavily on task-specific hard labels, limiting their generalization to unseen categories. Recent popular distillation-based prompt learning methods improve generalization by exploiting larger teacher VLMs and unsupervised knowledge transfer, yet their repetitive teacher model online inference sacrifices the inherent training efficiency advantage of prompt learning. In this paper, we propose {\large {\textbf{F}}}aster {\large {\textbf{D}}}istillation-{\large {\textbf{B}}}ased {\large {\textbf{P}}}rompt {\large {\textbf{L}}}earning (\textbf{FDBPL}), which addresses these issues by sharing soft supervision contexts across multiple training stages and implementing accelerated I/O. Furthermore, FDBPL introduces a region-aware prompt learning paradigm with dual positive-negative prompt spaces to fully exploit randomly cropped regions that containing multi-level information. We propose a positive-negative space mutual learning mechanism based on similarity-difference learning, enabling student CLIP models to recognize correct semantics while learning to reject weakly related concepts, thereby improving zero-shot performance. Unlike existing distillation-based prompt learning methods that sacrifice parameter efficiency for generalization, FDBPL maintains dual advantages of parameter efficiency and strong downstream generalization. Comprehensive evaluations across 11 datasets demonstrate superior performance in base-to-new generalization, cross-dataset transfer, and robustness tests, achieving $2.2\times$ faster training speed.

Paperid: 2422, https://arxiv.org/pdf/2505.14728.pdf

Abstract:
Warning: This paper contains examples of harmful language and images. Reader discretion is advised. Recently, vision-language models have demonstrated increasing influence in morally sensitive domains such as autonomous driving and medical analysis, owing to their powerful multimodal reasoning capabilities. As these models are deployed in high-stakes real-world applications, it is of paramount importance to ensure that their outputs align with human moral values and remain within moral boundaries. However, existing work on moral alignment either focuses solely on textual modalities or relies heavily on AI-generated images, leading to distributional biases and reduced realism. To overcome these limitations, we introduce MORALISE, a comprehensive benchmark for evaluating the moral alignment of vision-language models (VLMs) using diverse, expert-verified real-world data. We begin by proposing a comprehensive taxonomy of 13 moral topics grounded in Turiel's Domain Theory, spanning the personal, interpersonal, and societal moral domains encountered in everyday life. Built on this framework, we manually curate 2,481 high-quality image-text pairs, each annotated with two fine-grained labels: (1) topic annotation, identifying the violated moral topic(s), and (2) modality annotation, indicating whether the violation arises from the image or the text. For evaluation, we encompass two tasks, \textit{moral judgment} and \textit{moral norm attribution}, to assess models' awareness of moral violations and their reasoning ability on morally salient content. Extensive experiments on 19 popular open- and closed-source VLMs show that MORALISE poses a significant challenge, revealing persistent moral limitations in current state-of-the-art models. The full benchmark is publicly available at https://huggingface.co/datasets/Ze1025/MORALISE.

Paperid: 2423, https://arxiv.org/pdf/2505.13062.pdf

Abstract:
Humans can intuitively infer sounds from silent videos, but whether multimodal large language models can perform modal-mismatch reasoning without accessing target modalities remains relatively unexplored. Current text-assisted-video-to-audio (VT2A) methods excel in video foley tasks but struggle to acquire audio descriptions during inference. We introduce the task of Reasoning Audio Descriptions from Silent Videos (SVAD) to address this challenge and investigate vision-language models' (VLMs) capabilities on this task. To further enhance the VLMs' reasoning capacity for the SVAD task, we construct a CoT-AudioCaps dataset and propose a Chain-of-Thought-based supervised fine-tuning strategy. Experiments on SVAD and subsequent VT2A tasks demonstrate our method's effectiveness in two key aspects: significantly improving VLMs' modal-mismatch reasoning for SVAD and effectively addressing the challenge of acquiring audio descriptions during VT2A inference.

Paperid: 2424, https://arxiv.org/pdf/2505.12081.pdf

Abstract:
Large vision-language models exhibit inherent capabilities to handle diverse visual perception tasks. In this paper, we introduce VisionReasoner, a unified framework capable of reasoning and solving multiple visual perception tasks within a shared model. Specifically, by designing novel multi-object cognitive learning strategies and systematic task reformulation, VisionReasoner enhances its reasoning capabilities to analyze visual inputs, and addresses diverse perception tasks in a unified framework. The model generates a structured reasoning process before delivering the desired outputs responding to user queries. To rigorously assess unified visual perception capabilities, we evaluate VisionReasoner on ten diverse tasks spanning three critical domains: detection, segmentation, and counting. Experimental results show that VisionReasoner achieves superior performance as a unified model, outperforming Qwen2.5VL by relative margins of 29.1% on COCO (detection), 22.1% on ReasonSeg (segmentation), and 15.3% on CountBench (counting).

Paperid: 2425, https://arxiv.org/pdf/2504.20384.pdf

Abstract:
Recent advancements in video understanding within visual large language models (VLLMs) have led to notable progress. However, the complexity of video data and contextual processing limitations still hinder long-video comprehension. A common approach is video feature compression to reduce token input to large language models, yet many methods either fail to prioritize essential features, leading to redundant inter-frame information, or introduce computationally expensive modules.To address these issues, we propose FiLA(Fine-grained Vision Language Model)-Video, a novel framework that leverages a lightweight dynamic-weight multi-frame fusion strategy, which adaptively integrates multiple frames into a single representation while preserving key video information and reducing computational costs. To enhance frame selection for fusion, we introduce a keyframe selection strategy, effectively identifying informative frames from a larger pool for improved summarization. Additionally, we present a simple yet effective long-video training data generation strategy, boosting model performance without extensive manual annotation. Experimental results demonstrate that FiLA-Video achieves superior efficiency and accuracy in long-video comprehension compared to existing methods.

Paperid: 2426, https://arxiv.org/pdf/2503.01743.pdf

Authors:Microsoft, :, Abdelrahman Abouelenin, Atabak Ashfaq, Adam Atkinson, Hany Awadalla, Nguyen Bach, Jianmin Bao, Alon Benhaim, Martin Cai, Vishrav Chaudhary, Congcong Chen, Dong Chen, Dongdong Chen, Junkun Chen, Weizhu Chen, Yen-Chun Chen, Yi-ling Chen, Qi Dai, Xiyang Dai, Ruchao Fan, Mei Gao, Min Gao, Amit Garg, Abhishek Goswami, Junheng Hao, Amr Hendy, Yuxuan Hu, Xin Jin, Mahmoud Khademi, Dongwoo Kim, Young Jin Kim, Gina Lee, Jinyu Li, Yunsheng Li, Chen Liang, Xihui Lin, Zeqi Lin, Mengchen Liu, Yang Liu, Gilsinia Lopez, Chong Luo, Piyush Madan, Vadim Mazalov, Arindam Mitra, Ali Mousavi, Anh Nguyen, Jing Pan, Daniel Perez-Becker, Jacob Platin, Thomas Portet, Kai Qiu, Bo Ren, Liliang Ren, Sambuddha Roy, Ning Shang, Yelong Shen, Saksham Singhal, Subhojit Som, Xia Song, Tetyana Sych, Praneetha Vaddamanu, Shuohang Wang, Yiming Wang, Zhenghao Wang, Haibin Wu, Haoran Xu, Weijian Xu, Yifan Yang, Ziyi Yang, Donghan Yu, Ishmam Zabir, Jianwen Zhang, Li Lyna Zhang, Yunan Zhang, Xiren Zhou

Abstract:
We introduce Phi-4-Mini and Phi-4-Multimodal, compact yet highly capable language and multimodal models. Phi-4-Mini is a 3.8-billion-parameter language model trained on high-quality web and synthetic data, significantly outperforming recent open-source models of similar size and matching the performance of models twice its size on math and coding tasks requiring complex reasoning. This achievement is driven by a carefully curated synthetic data recipe emphasizing high-quality math and coding datasets. Compared to its predecessor, Phi-3.5-Mini, Phi-4-Mini features an expanded vocabulary size of 200K tokens to better support multilingual applications, as well as group query attention for more efficient long-sequence generation. Phi-4-Multimodal is a multimodal model that integrates text, vision, and speech/audio input modalities into a single model. Its novel modality extension approach leverages LoRA adapters and modality-specific routers to allow multiple inference modes combining various modalities without interference. For example, it now ranks first in the OpenASR leaderboard to date, although the LoRA component of the speech/audio modality has just 460 million parameters. Phi-4-Multimodal supports scenarios involving (vision + language), (vision + speech), and (speech/audio) inputs, outperforming larger vision-language and speech-language models on a wide range of tasks. Additionally, we experiment to further train Phi-4-Mini to enhance its reasoning capabilities. Despite its compact 3.8-billion-parameter size, this experimental version achieves reasoning performance on par with or surpassing significantly larger models, including DeepSeek-R1-Distill-Qwen-7B and DeepSeek-R1-Distill-Llama-8B.

Paperid: 2427, https://arxiv.org/pdf/2412.14487.pdf

Abstract:
Direct Preference Optimization (DPO) has been demonstrated to be highly effective in mitigating hallucinations in Large Vision Language Models (LVLMs) by aligning their outputs more closely with human preferences. Despite the recent progress, existing methods suffer from two drawbacks: 1) Lack of scalable token-level rewards; and 2) Neglect of visual-anchored tokens. To this end, we propose a novel Token Preference Optimization model with self-calibrated rewards (dubbed as TPO), which adaptively attends to visual-correlated tokens without fine-grained annotations. Specifically, we introduce a token-level \emph{visual-anchored} \emph{reward} as the difference of the logistic distributions of generated tokens conditioned on the raw image and the corrupted one. In addition, to highlight the informative visual-anchored tokens, a visual-aware training objective is proposed to enhance more accurate token-level optimization. Extensive experimental results have manifested the state-of-the-art performance of the proposed TPO. For example, by building on top of LLAVA-1.5-7B, our TPO boosts the performance absolute improvement for hallucination benchmarks.

Paperid: 2428, https://arxiv.org/pdf/2412.08378.pdf

Abstract:
Recently, there has been growing interest in the capability of multimodal large language models (MLLMs) to process high-resolution images. A common approach currently involves dynamically cropping the original high-resolution image into smaller sub-images, which are then fed into a vision encoder that was pre-trained on lower-resolution images. However, this cropping approach often truncates objects and connected areas in the original image, causing semantic breaks. To address this limitation, we introduce HyViLM, designed to process images of any resolution while retaining the overall context during encoding. Specifically, we: (i) Design a new visual encoder called Hybrid Encoder that not only encodes individual sub-images but also interacts with detailed global visual features, significantly improving the model's ability to encode high-resolution images. (ii) Propose an optimal feature fusion strategy for the dynamic cropping approach, effectively leveraging information from different layers of the vision encoder. Compared with the state-of-the-art MLLMs under the same setting, our HyViLM outperforms existing MLLMs in nine out of ten tasks. Specifically, HyViLM achieves a 9.6% improvement in performance on the TextVQA task and a 6.9% enhancement on the DocVQA task.

Paperid: 2429, https://arxiv.org/pdf/2409.08468.pdf

Abstract:
Vision-language models (VLMs) have demonstrated remarkable open-vocabulary object recognition capabilities, motivating their adaptation for dense prediction tasks like segmentation. However, directly applying VLMs to such tasks remains challenging due to their lack of pixel-level granularity and the limited data available for fine-tuning, leading to overfitting and poor generalization. To address these limitations, we propose Generalization Boosted Adapter (GBA), a novel adapter strategy that enhances the generalization and robustness of VLMs for open-vocabulary segmentation. GBA comprises two core components: (1) a Style Diversification Adapter (SDA) that decouples features into amplitude and phase components, operating solely on the amplitude to enrich the feature space representation while preserving semantic consistency; and (2) a Correlation Constraint Adapter (CCA) that employs cross-attention to establish tighter semantic associations between text categories and target regions, suppressing irrelevant low-frequency ``noise'' information and avoiding erroneous associations. Through the synergistic effect of the shallow SDA and the deep CCA, GBA effectively alleviates overfitting issues and enhances the semantic relevance of feature representations. As a simple, efficient, and plug-and-play component, GBA can be flexibly integrated into various CLIP-based methods, demonstrating broad applicability and achieving state-of-the-art performance on multiple open-vocabulary segmentation benchmarks.

Paperid: 2430, https://arxiv.org/pdf/2601.05547.pdf

Abstract:
Vision-Language Models (VLMs) have demonstrated remarkable progress in multimodal tasks, but remain susceptible to hallucinations, where generated text deviates from the underlying visual content. Existing hallucination detection methods primarily rely on output logits or external verification tools, often overlooking their internal mechanisms. In this work, we investigate the outputs of internal attention heads, postulating that specific heads carry the primary signals for truthful generation.However, directly probing these high-dimensional states is challenging due to the entanglement of visual-linguistic syntax and noise. To address this, we propose VIB-Probe, a novel hallucination detection and mitigation framework leveraging the Variational Information Bottleneck (VIB) theory. Our method extracts discriminative patterns across layers and heads while filtering out semantic nuisances through the information bottleneck principle. Furthermore, by leveraging the gradients of our VIB probe, we identify attention heads with strong causal influence on hallucinations and introduce an inference-time intervention strategy for hallucination mitigation. Extensive experiments across diverse benchmarks demonstrate that VIB-Probe significantly outperforms existing baselines in both settings. Our code will be made publicly available.

Paperid: 2431, https://arxiv.org/pdf/2512.19022.pdf

Abstract:
Face Presentation Attack Detection (PAD) demands incremental learning (IL) to combat evolving spoofing tactics and domains. Privacy regulations, however, forbid retaining past data, necessitating rehearsal-free IL (RF-IL). Vision-Language Pre-trained (VLP) models, with their prompt-tunable cross-modal representations, enable efficient adaptation to new spoofing styles and domains. Capitalizing on this strength, we propose \textbf{SVLP-IL}, a VLP-based RF-IL framework that balances stability and plasticity via \textit{Multi-Aspect Prompting} (MAP) and \textit{Selective Elastic Weight Consolidation} (SEWC). MAP isolates domain dependencies, enhances distribution-shift sensitivity, and mitigates forgetting by jointly exploiting universal and domain-specific cues. SEWC selectively preserves critical weights from previous tasks, retaining essential knowledge while allowing flexibility for new adaptations. Comprehensive experiments across multiple PAD benchmarks show that SVLP-IL significantly reduces catastrophic forgetting and enhances performance on unseen domains. SVLP-IL offers a privacy-compliant, practical solution for robust lifelong PAD deployment in RF-IL settings.

Paperid: 2432, https://arxiv.org/pdf/2511.23269.pdf

Abstract:
High-quality and carefully curated data is a cornerstone of training medical large language models, as it directly impacts both generalization and robustness to unseen clinical tasks. We investigate strategies for training and data curation to develop a robust multimodal reasoning model in the medical domain. Our work focuses on supervised fine-tuning (SFT) and explores data recipes that leverage structured reasoning traces. Using our proposed data recipe, we scale experiments to a dataset of over 8 million examples and 6.8 billion response tokens, achieving state-of-the-art performance among open-source models across diverse out-of-distribution medical benchmark tasks. Our results further indicate that curating a high-quality, diverse training dataset with varying structured reasoning trace lengths enables the fine-tuned model to self-calibrate its reasoning trajectory lengths based on the downstream task, without explicit supervision. We present key insights, describe the data curation strategy, and outline next steps toward developing robust medical vision-language reasoning system.

Paperid: 2433, https://arxiv.org/pdf/2511.22787.pdf

Abstract:
In a globalized world, cultural elements from diverse origins frequently appear together within a single visual scene. We refer to these as culture mixing scenarios, yet how Large Vision-Language Models (LVLMs) perceive them remains underexplored. We investigate culture mixing as a critical challenge for LVLMs and examine how current models behave when cultural items from multiple regions appear together. To systematically analyze these behaviors, we construct CultureMix, a food Visual Question Answering (VQA) benchmark with 23k diffusion-generated, human-verified culture mixing images across four subtasks: (1) food-only, (2) food+food, (3) food+background, and (4) food+food+background. Evaluating 10 LVLMs, we find consistent failures to preserve individual cultural identities in mixed settings. Models show strong background reliance, with accuracy dropping 14% when cultural backgrounds are added to food-only baselines, and they produce inconsistent predictions for identical foods across different contexts. To address these limitations, we explore three robustness strategies. We find supervised fine-tuning using a diverse culture mixing dataset substantially improve model consistency and reduce background sensitivity. We call for increased attention to culture mixing scenarios as a critical step toward developing LVLMs capable of operating reliably in culturally diverse real-world environments.

Paperid: 2434, https://arxiv.org/pdf/2511.19417.pdf

Abstract:
Large Language Models (LLMs) have demonstrated remarkable capabilities in challenging, knowledge-intensive reasoning tasks. However, extending LLMs to perceive and reason over a new modality (e.g., vision), often requires costly development of large-scale vision language models (VLMs) with LLMs as backbones. Smaller VLMs are more efficient and adaptable but often lack the broad knowledge and reasoning capabilities of frontier LLMs. In this work, we propose BeMyEyes, a modular, multi-agent framework for extending LLMs to multimodal reasoning by orchestrating collaboration between efficient, adaptable VLMs as perceivers and powerful LLMs as reasoners through conversations. We then introduce a data synthesis and supervised fine-tuning pipeline to train the perceiver agent to effectively collaborate with the reasoner agent. By combining the complementary strengths of perception and reasoning agents, BeMyEyes avoids the need for training large-scale multimodal models, preserves the generalization and reasoning capabilities of LLMs, and allows flexible extension to new domains and modalities. Experiments show that our framework unlocks the multimodal reasoning capabilities for LLMs, enabling a lightweight and fully open-source solution, i.e. equipping text-only DeepSeek-R1 with Qwen2.5-VL-7B perceiver, to outperform large-scale proprietary VLMs such as GPT-4o on a wide range of knowledge-intensive multimodal tasks. These results demonstrate the effectiveness, modularity, and scalability of our multi-agent approach for building future multimodal reasoning systems.

Paperid: 2435, https://arxiv.org/pdf/2511.17450.pdf

Abstract:
Recent video generation approaches increasingly rely on planning intermediate control signals such as object trajectories to improve temporal coherence and motion fidelity. However, these methods mostly employ single-shot plans that are typically limited to simple motions, or iterative refinement which requires multiple calls to the video generator, incuring high computational cost. To overcome these limitations, we propose SketchVerify, a training-free, sketch-verification-based planning framework that improves motion planning quality with more dynamically coherent trajectories (i.e., physically plausible and instruction-consistent motions) prior to full video generation by introducing a test-time sampling and verification loop. Given a prompt and a reference image, our method predicts multiple candidate motion plans and ranks them using a vision-language verifier that jointly evaluates semantic alignment with the instruction and physical plausibility. To efficiently score candidate motion plans, we render each trajectory as a lightweight video sketch by compositing objects over a static background, which bypasses the need for expensive, repeated diffusion-based synthesis while achieving comparable performance. We iteratively refine the motion plan until a satisfactory one is identified, which is then passed to the trajectory-conditioned generator for final synthesis. Experiments on WorldModelBench and PhyWorldBench demonstrate that our method significantly improves motion quality, physical realism, and long-term consistency compared to competitive baselines while being substantially more efficient. Our ablation study further shows that scaling up the number of trajectory candidates consistently enhances overall performance.

Paperid: 2436, https://arxiv.org/pdf/2511.08367.pdf

Abstract:
Large Vision-Language Models (VLMs) are susceptible to jailbreak attacks: researchers have developed a variety of attack strategies that can successfully bypass the safety mechanisms of VLMs. Among these approaches, jailbreak methods based on the Out-of-Distribution (OOD) strategy have garnered widespread attention due to their simplicity and effectiveness. This paper further advances the in-depth understanding of OOD-based VLM jailbreak methods. Experimental results demonstrate that jailbreak samples generated via mild OOD strategies exhibit superior performance in circumventing the safety constraints of VLMs--a phenomenon we define as ''weak-OOD''. To unravel the underlying causes of this phenomenon, this study takes SI-Attack, a typical OOD-based jailbreak method, as the research object. We attribute this phenomenon to a trade-off between two dominant factors: input intent perception and model refusal triggering. The inconsistency in how these two factors respond to OOD manipulations gives rise to this phenomenon. Furthermore, we provide a theoretical argument for the inevitability of such inconsistency from the perspective of discrepancies between model pre-training and alignment processes. Building on the above insights, we draw inspiration from optical character recognition (OCR) capability enhancement--a core task in the pre-training phase of mainstream VLMs. Leveraging this capability, we design a simple yet highly effective VLM jailbreak method, whose performance outperforms that of SOTA baselines.

Paperid: 2437, https://arxiv.org/pdf/2511.02367.pdf

Abstract:
The proliferation of Vision-Language Models (VLMs) introduces profound privacy risks from personal videos. This paper addresses the critical yet unexplored inferential privacy threat, the risk of inferring sensitive personal attributes over the data. To address this gap, we crowdsourced a dataset of 508 everyday personal videos from 58 individuals. We then conducted a benchmark study evaluating VLM inference capabilities against human performance. Our findings reveal three critical insights: (1) VLMs possess superhuman inferential capabilities, significantly outperforming human evaluators, leveraging a shift from object recognition to behavioral inference from temporal streams. (2) Inferential risk is strongly correlated with factors such as video characteristics and prompting strategies. (3) VLM-driven explanation towards the inference is unreliable, as we revealed a disconnect between the model-generated explanations and evidential impact, identifying ubiquitous objects as misleading confounders.

Paperid: 2438, https://arxiv.org/pdf/2510.12992.pdf

Abstract:
Safe large-scale coordination of multiple cooperative connected autonomous vehicles (CAVs) hinges on communication that is both efficient and interpretable. Existing approaches either rely on transmitting high-bandwidth raw sensor data streams or neglect perception and planning uncertainties inherent in shared data, resulting in systems that are neither scalable nor safe. To address these limitations, we propose Uncertainty-Guided Natural Language Cooperative Autonomous Planning (UNCAP), a vision-language model-based planning approach that enables CAVs to communicate via lightweight natural language messages while explicitly accounting for perception uncertainty in decision-making. UNCAP features a two-stage communication protocol: (i) an ego CAV first identifies the subset of vehicles most relevant for information exchange, and (ii) the selected CAVs then transmit messages that quantitatively express their perception uncertainty. By selectively fusing messages that maximize mutual information, this strategy allows the ego vehicle to integrate only the most relevant signals into its decision-making, improving both the scalability and reliability of cooperative planning. Experiments across diverse driving scenarios show a 63% reduction in communication bandwidth with a 31% increase in driving safety score, a 61% reduction in decision uncertainty, and a four-fold increase in collision distance margin during near-miss events. Project website: https://uncap-project.github.io/

Paperid: 2439, https://arxiv.org/pdf/2509.26555.pdf

Abstract:
Recent advances in video generation have enabled high-fidelity video synthesis from user provided prompts. However, existing models and benchmarks fail to capture the complexity and requirements of professional video generation. Towards that goal, we introduce Stable Cinemetrics, a structured evaluation framework that formalizes filmmaking controls into four disentangled, hierarchical taxonomies: Setup, Event, Lighting, and Camera. Together, these taxonomies define 76 fine-grained control nodes grounded in industry practices. Using these taxonomies, we construct a benchmark of prompts aligned with professional use cases and develop an automated pipeline for prompt categorization and question generation, enabling independent evaluation of each control dimension. We conduct a large-scale human study spanning 10+ models and 20K videos, annotated by a pool of 80+ film professionals. Our analysis, both coarse and fine-grained reveal that even the strongest current models exhibit significant gaps, particularly in Events and Camera-related controls. To enable scalable evaluation, we train an automatic evaluator, a vision-language model aligned with expert annotations that outperforms existing zero-shot baselines. SCINE is the first approach to situate professional video generation within the landscape of video generative models, introducing taxonomies centered around cinematic controls and supporting them with structured evaluation pipelines and detailed analyses to guide future research.

Paperid: 2440, https://arxiv.org/pdf/2509.25731.pdf

Abstract:
Recent multimodal models for instruction-based face editing enable semantic manipulation but still struggle with precise attribute control and identity preservation. Structural facial representations such as landmarks are effective for intermediate supervision, yet most existing methods treat them as rigid geometric constraints, which can degrade identity when conditional landmarks deviate significantly from the source (e.g., large expression or pose changes, inaccurate landmark estimates). To address these limitations, we propose LaTo, a landmark-tokenized diffusion transformer for fine-grained, identity-preserving face editing. Our key innovations include: (1) a landmark tokenizer that directly quantizes raw landmark coordinates into discrete facial tokens, obviating the need for dense pixel-wise correspondence; (2) a location-mapping positional encoding that integrates facial and image tokens for unified processing, enabling flexible yet decoupled geometry-appearance interactions with high efficiency and strong identity preservation; and (3) a landmark predictor that leverages vision-language models to infer target landmarks from instructions and source images, whose structured chain-of-thought improves estimation accuracy and interactive control. To mitigate data scarcity, we curate HFL-150K, to our knowledge the largest benchmark for this task, containing over 150K real face pairs with fine-grained instructions. Extensive experiments show that LaTo outperforms state-of-the-art methods by 7.8% in identity preservation and 4.6% in semantic consistency. Code and dataset will be made publicly available upon acceptance.

Paperid: 2441, https://arxiv.org/pdf/2509.23879.pdf

Abstract:
The reliability of Multimodal Large Language Models (MLLMs) in real-world settings is often undermined by sensitivity to irrelevant or distracting visual context, an aspect not captured by existing evaluation metrics. We introduce the \textbf{Patch Context Robustness Index (PCRI)}, the first systematic and interpretable score for quantifying MLLM robustness to variations in visual context granularity, measuring performance changes between localized image patches and full-image input. Applying PCRI to 19 state-of-the-art MLLMs across 15 vision-language benchmarks, we find that most leading models remain brittle to background noise, with only a few, such as InternVL2-26B and Qwen2VL-72B, demonstrating consistent robustness across tasks. PCRI analysis also highlights how different model architectures handle and integrate visual context, offering actionable diagnostic insight for both researchers and practitioners. PCRI enables rigorous comparison of context robustness, supporting principled model selection and guiding the development of future architectures and training strategies for robust, real-world deployment.

Paperid: 2442, https://arxiv.org/pdf/2509.23673.pdf

Abstract:
Multimodal Large Language Models (MLLMs) have achieved impressive results on vision-language benchmarks, yet it remains unclear whether these benchmarks assess genuine global reasoning or allow success via localized visual cues. Existing evaluation methods do not explicitly measure this distinction, hindering effective dataset curation and real-world focused model development. We introduce Region Comprehension Index (RCI), the first model-based score to directly quantify a dataset's reliance on global versus local visual information. RCI systematically compares reference-model performance on image patches versus full images, revealing if tasks require holistic image understanding or can be solved with partial or localized visual cues. When applying RCI to 13 widely used multimodal benchmarks, we observed that most of them favor localized reasoning and exhibit significant spatial biases, indicating potential risks in real-world applications. RCI equips researchers & practitioners with an actionable tool for diagnosing & mitigating these biases, enabling the construction of datasets and benchmarks to foster the development of robust, enterprise-ready multimodal systems.

Paperid: 2443, https://arxiv.org/pdf/2509.21050.pdf

Abstract:
AI-driven geometric problem solving is a complex vision-language task that requires accurate diagram interpretation, mathematical reasoning, and robust cross-modal grounding. A foundational yet underexplored capability for this task is the ability to identify and interpret geometric elements based on natural language queries. To address this, we introduce the task of Referring Expression Comprehension (REC) for geometric problems, which evaluates whether models can localize points, shapes, and spatial relations in diagrams in response to textual prompts. We present GeoRef, a benchmark dataset constructed from existing geometric problem corpora, featuring diverse, high-quality annotations and queries. Due to the lack of annotated data for this task, we generate a large-scale synthetic training dataset using a structured geometric formal language, enabling broad coverage of geometric concepts and facilitating model adaptation. We explore two fine-tuning approaches: Supervised Fine-Tuning (SFT) and Group Relative Policy Optimization (GRPO). Our results show that GRPO significantly outperforms SFT by better aligning model behavior with task-specific rewards. Furthermore, we propose a verify-and-regenerate mechanism that detects incorrect predictions and re-infers answers using contextual reasoning history, further boosting accuracy. Notably, even state-of-the-art Multimodal Large Language Models (MLLMs) struggle with this task, underscoring the necessity of explicitly evaluating and strengthening geometric grounding as a prerequisite for robust geometric problem solving. Moreover, models trained on GeoRef demonstrate measurable improvements on downstream geometric reasoning tasks, highlighting the broader value of REC as a foundation for multimodal mathematical understanding.

Paperid: 2444, https://arxiv.org/pdf/2508.09584.pdf

Abstract:
Despite rapid advances, Large Vision-Language Models (LVLMs) still suffer from hallucinations, i.e., generating content inconsistent with input or established world knowledge, which correspond to faithfulness and factuality hallucinations, respectively. Prior studies primarily evaluate faithfulness hallucination at a rather coarse level (e.g., object-level) and lack fine-grained analysis. Additionally, existing benchmarks often rely on costly manual curation or reused public datasets, raising concerns about scalability and data leakage. To address these limitations, we propose an automated data construction pipeline that produces scalable, controllable, and diverse evaluation data. We also design a hierarchical hallucination induction framework with input perturbations to simulate realistic noisy scenarios. Integrating these designs, we construct SHALE, a Scalable HALlucination Evaluation benchmark designed to assess both faithfulness and factuality hallucinations via a fine-grained hallucination categorization scheme. SHALE comprises over 30K image-instruction pairs spanning 12 representative visual perception aspects for faithfulness and 6 knowledge domains for factuality, considering both clean and noisy scenarios. Extensive experiments on over 20 mainstream LVLMs reveal significant factuality hallucinations and high sensitivity to semantic perturbations.

Paperid: 2445, https://arxiv.org/pdf/2508.00321.pdf

Abstract:
The proliferation of visual sensors in smart home environments, particularly through wearable devices like smart glasses, introduces profound privacy challenges. Existing privacy controls are often static and coarse-grained, failing to accommodate the dynamic and socially nuanced nature of home environments. This paper investigates the viability of using Large Language Models (LLMs) as the core of a dynamic and adaptive privacy policy engine. We propose a conceptual framework where visual data is classified using a multi-dimensional schema that considers data sensitivity, spatial context, and social presence. An LLM then reasons over this contextual information to enforce fine-grained privacy rules, such as selective object obfuscation, in real-time. Through a comparative evaluation of state-of-the-art Vision Language Models (including GPT-4o and the Qwen-VL series) in simulated home settings , our findings show the feasibility of this approach. The LLM-based engine achieved a top machine-evaluated appropriateness score of 3.99 out of 5, and the policies generated by the models received a top human-evaluated score of 4.00 out of 5.

Paperid: 2446, https://arxiv.org/pdf/2507.22398.pdf

Abstract:
Vision-Language Models (VLMs) are increasingly used as perceptual modules for visual content reasoning, including through captioning and DeepFake detection. In this work, we expose a critical vulnerability of VLMs when exposed to subtle, structured perturbations in the frequency domain. Specifically, we highlight how these feature transformations undermine authenticity/DeepFake detection and automated image captioning tasks. We design targeted image transformations, operating in the frequency domain to systematically adjust VLM outputs when exposed to frequency-perturbed real and synthetic images. We demonstrate that the perturbation injection method generalizes across five state-of-the-art VLMs which includes different-parameter Qwen2/2.5 and BLIP models. Experimenting across ten real and generated image datasets reveals that VLM judgments are sensitive to frequency-based cues and may not wholly align with semantic content. Crucially, we show that visually-imperceptible spatial frequency transformations expose the fragility of VLMs deployed for automated image captioning and authenticity detection tasks. Our findings under realistic, black-box constraints challenge the reliability of VLMs, underscoring the need for robust multimodal perception systems.

Paperid: 2447, https://arxiv.org/pdf/2506.20616.pdf

Abstract:
Humans possess a unique ability to perceive meaningful patterns in ambiguous stimuli, a cognitive phenomenon known as pareidolia. This paper introduces Shape2Animal framework to mimics this imaginative capacity by reinterpreting natural object silhouettes, such as clouds, stones, or flames, as plausible animal forms. Our automated framework first performs open-vocabulary segmentation to extract object silhouette and interprets semantically appropriate animal concepts using vision-language models. It then synthesizes an animal image that conforms to the input shape, leveraging text-to-image diffusion model and seamlessly blends it into the original scene to generate visually coherent and spatially consistent compositions. We evaluated Shape2Animal on a diverse set of real-world inputs, demonstrating its robustness and creative potential. Our Shape2Animal can offer new opportunities for visual storytelling, educational content, digital art, and interactive media design. Our project page is here: https://shape2image.github.io

Paperid: 2448, https://arxiv.org/pdf/2506.18504.pdf

Abstract:
Recently, vision-language pretraining has emerged as a transformative technique that integrates the strengths of both visual and textual modalities, resulting in powerful vision-language models (VLMs). Leveraging web-scale pretraining data, these models exhibit strong zero-shot capabilities. However, their performance often deteriorates when confronted with domain-specific or specialized generalization tasks. To address this, a growing body of research focuses on transferring or generalizing the rich knowledge embedded in VLMs to various downstream applications. This survey aims to comprehensively summarize the generalization settings, methodologies, benchmarking and results in VLM literatures. Delving into the typical VLM structures, current literatures are categorized into prompt-based, parameter-based and feature-based methods according to the transferred modules. The differences and characteristics in each category are furthered summarized and discussed by revisiting the typical transfer learning (TL) settings, providing novel interpretations for TL in the era of VLMs. Popular benchmarks for VLM generalization are further introduced with thorough performance comparisons among the reviewed methods. Following the advances in large-scale generalizable pretraining, this survey also discusses the relations and differences between VLMs and up-to-date multimodal large language models (MLLM), e.g., DeepSeek-VL. By systematically reviewing the surging literatures in vision-language research from a novel and practical generalization prospective, this survey contributes to a clear landscape of current and future multimodal researches.

Paperid: 2449, https://arxiv.org/pdf/2505.03981.pdf

Abstract:
Recent proprietary models (e.g., o3) have begun to demonstrate strong multimodal reasoning capabilities. Yet, most existing open-source research concentrates on training text-only reasoning models, with evaluations limited to mainly mathematical and general-domain tasks. Therefore, it remains unclear how to effectively extend reasoning capabilities beyond text input and general domains. This paper explores a fundamental research question: Is reasoning generalizable across modalities and domains? Our findings support an affirmative answer: General-domain text-based post-training can enable such strong generalizable reasoning. Leveraging this finding, we introduce X-Reasoner, a vision-language model post-trained solely on general-domain text for generalizable reasoning, using a two-stage approach: an initial supervised fine-tuning phase with distilled long chain-of-thoughts, followed by reinforcement learning with verifiable rewards. Experiments show that X-Reasoner successfully transfers reasoning capabilities to both multimodal and out-of-domain settings, outperforming existing state-of-the-art models trained with in-domain and multimodal data across various general and medical benchmarks (Figure 1). Additionally, we find that X-Reasoner's performance in specialized domains can be further enhanced through continued training on domain-specific text-only data. Building upon this, we introduce X-Reasoner-Med, a medical-specialized variant that achieves new state of the art on numerous text-only and multimodal medical benchmarks.

Paperid: 2450, https://arxiv.org/pdf/2504.04630.pdf

Abstract:
Foundation Models (FMs), particularly Large Language Models (LLMs), are increasingly used to support various software engineering activities (e.g., coding and testing). Their applications in the software engineering of Cyber-Physical Systems (CPSs) are also growing. However, research in this area remains limited. Moreover, existing studies have primarily focused on LLMs-only one type of FM-leaving ample opportunities to explore others, such as vision-language models. We argue that, in addition to LLMs, other FMs utilizing different data modalities (e.g., images, audio) and multimodal models (which integrate multiple modalities) hold great potential for supporting CPS software engineering, given that these systems process diverse data types. To address this, we present a research roadmap for integrating FMs into various phases of CPS software engineering, highlighting key research opportunities and challenges for the software engineering community.

Paperid: 2451, https://arxiv.org/pdf/2503.22122.pdf

Abstract:
Vision-language models (VLMs) have demonstrated remarkable capabilities in robotic planning, particularly for long-horizon tasks that require a holistic understanding of the environment for task decomposition. Existing methods typically rely on prior environmental knowledge or carefully designed task-specific prompts, making them struggle with dynamic scene changes or unexpected task conditions, e.g., a robot attempting to put a carrot in the microwave but finds the door was closed. Such challenges underscore two critical issues: adaptability and efficiency. To address them, in this work, we propose an adaptive multi-agent planning framework, termed REMAC, that enables efficient, scene-agnostic multi-robot long-horizon task planning and execution through continuous reflection and self-evolution. REMAC incorporates two key modules: a self-reflection module performing pre-condition and post-condition checks in the loop to evaluate progress and refine plans, and a self-evolvement module dynamically adapting plans based on scene-specific reasoning. It offers several appealing benefits: 1) Robots can initially explore and reason about the environment without complex prompt design. 2) Robots can keep reflecting on potential planning errors and adapting the plan based on task-specific insights. 3) After iterations, a robot can call another one to coordinate tasks in parallel, maximizing the task execution efficiency. To validate REMAC's effectiveness, we build a multi-agent environment for long-horizon robot manipulation and navigation based on RoboCasa, featuring 4 task categories with 27 task styles and 50+ different objects. Based on it, we further benchmark state-of-the-art reasoning models, including DeepSeek-R1, o3-mini, QwQ, and Grok3, demonstrating REMAC's superiority by boosting average success rates by 40% and execution efficiency by 52.7% over the single robot baseline.

Paperid: 2452, https://arxiv.org/pdf/2503.11794.pdf

Abstract:
Vision-Language Models (VLMs) leverage aligned visual encoders to transform images into visual tokens, allowing them to be processed similarly to text by the backbone large language model (LLM). This unified input paradigm enables VLMs to excel in vision-language tasks such as visual question answering (VQA). To improve fine-grained visual reasoning, recent advancements in vision-language modeling introduce image cropping techniques that feed all encoded sub-images into the model. However, this approach significantly increases the number of visual tokens, leading to inefficiency and potential distractions for the LLM. To address the generalization challenges of image representation in VLMs, we propose a lightweight, universal framework that seamlessly integrates with existing VLMs to enhance their ability to process finegrained details. Our method leverages textual semantics to identify key visual areas, improving VQA performance without requiring any retraining of the VLM. Additionally, it incorporates textual signals into the visual encoding process, enhancing both efficiency and effectiveness. The proposed method, SEMCLIP, strengthens the visual understanding of a 7B VLM, LLaVA-1.5 by 3.3% on average across 7 benchmarks, and particularly by 5.3% on the challenging detailed understanding benchmark V*.

Paperid: 2453, https://arxiv.org/pdf/2501.09783.pdf

Abstract:
We present GeoManip, a framework to enable generalist robots to leverage essential conditions derived from object and part relationships, as geometric constraints, for robot manipulation. For example, cutting the carrot requires adhering to a geometric constraint: the blade of the knife should be perpendicular to the carrot's direction. By interpreting these constraints through symbolic language representations and translating them into low-level actions, GeoManip bridges the gap between natural language and robotic execution, enabling greater generalizability across diverse even unseen tasks, objects, and scenarios. Unlike vision-language-action models that require extensive training, operates training-free by utilizing large foundational models: a constraint generation module that predicts stage-specific geometric constraints and a geometry parser that identifies object parts involved in these constraints. A solver then optimizes trajectories to satisfy inferred constraints from task descriptions and the scene. Furthermore, GeoManip learns in-context and provides five appealing human-robot interaction features: on-the-fly policy adaptation, learning from human demonstrations, learning from failure cases, long-horizon action planning, and efficient data collection for imitation learning. Extensive evaluations on both simulations and real-world scenarios demonstrate GeoManip's state-of-the-art performance, with superior out-of-distribution generalization while avoiding costly model training.

Paperid: 2454, https://arxiv.org/pdf/2411.02545.pdf

Abstract:
Contrastive Language-Image Pretraining (CLIP) models maximize the mutual information between text and visual modalities to learn representations. This makes the nature of the training data a significant factor in the efficacy of CLIP for downstream tasks. However, the lack of compositional diversity in contemporary image-text datasets limits the compositional reasoning ability of CLIP. We show that generating ``hard'' negative captions via in-context learning and synthesizing corresponding negative images with text-to-image generators offers a solution. We introduce a novel contrastive pre-training strategy that leverages these hard negative captions and images in an alternating fashion to train CLIP. We demonstrate that our method, named TripletCLIP, when applied to existing datasets such as CC3M and CC12M, enhances the compositional capabilities of CLIP, resulting in an absolute improvement of over 9% on the SugarCrepe benchmark on an equal computational budget, as well as improvements in zero-shot image classification and image retrieval. Our code, models, and data are available at: https://tripletclip.github.io

Paperid: 2455, https://arxiv.org/pdf/2410.13666.pdf

Abstract:
Deriving inference from heterogeneous inputs (such as images, text, and audio) is an important skill for humans to perform day-to-day tasks. A similar ability is desirable for the development of advanced Artificial Intelligence (AI) systems. While state-of-the-art models are rapidly closing the gap with human-level performance on diverse computer vision and NLP tasks separately, they struggle to solve tasks that require joint reasoning over visual and textual modalities. Inspired by GLUE (Wang et. al., 2018)- a multitask benchmark for natural language understanding, we propose VL-GLUE in this paper. VL-GLUE consists of over 100k samples spanned across seven different tasks, which at their core require visuo-linguistic reasoning. Moreover, our benchmark comprises of diverse image types (from synthetically rendered figures, and day-to-day scenes to charts and complex diagrams) and includes a broad variety of domain-specific text (from cooking, politics, and sports to high-school curricula), demonstrating the need for multi-modal understanding in the real-world. We show that this benchmark is quite challenging for existing large-scale vision-language models and encourage development of systems that possess robust visuo-linguistic reasoning capabilities.

Paperid: 2456, https://arxiv.org/pdf/2601.21420.pdf

Abstract:
Large language models allocate uniform computation across all tokens, ignoring that some sequences are trivially predictable while others require deep reasoning. We introduce ConceptMoE, which dynamically merges semantically similar tokens into concept representations, performing implicit token-level compute allocation. A learnable chunk module identifies optimal boundaries by measuring inter-token similarity, compressing sequences by a target ratio $R$ before they enter the compute-intensive concept model. Crucially, the MoE architecture enables controlled evaluation: we reallocate saved computation to match baseline activated FLOPs (excluding attention map computation) and total parameters, isolating genuine architectural benefits. Under these conditions, ConceptMoE consistently outperforms standard MoE across language and vision-language tasks, achieving +0.9 points on language pretraining, +2.3 points on long context understanding, and +0.6 points on multimodal benchmarks. When converting pretrained MoE during continual training with layer looping, gains reach +5.5 points, demonstrating practical applicability. Beyond performance, ConceptMoE reduces attention computation by up to $R^2\times$ and KV cache by $R\times$. At $R=2$, empirical measurements show prefill speedups reaching 175\% and decoding speedups up to 117\% on long sequences. The minimal architectural modifications enable straightforward integration into existing MoE, demonstrating that adaptive concept-level processing fundamentally improves both effectiveness and efficiency of large language models.

Paperid: 2457, https://arxiv.org/pdf/2601.21278.pdf

Abstract:
Vision Language Models (VLMs) are good at recognizing the global location of a photograph -- their geolocation prediction accuracy rivals the best human experts. But many VLMs are startlingly bad at explaining which image evidence led to their prediction, even when their location prediction is correct. The reasoning chains produced by VLMs frequently hallucinate scene attributes to support their location prediction (e.g. phantom writing, imagined infrastructure, misidentified flora). In this paper, we introduce the first benchmark for geolocation reasoning chains. We focus on the global location prediction task in the popular GeoGuessr game which draws from Google Street View spanning more than 100 countries. We collaborate with expert GeoGuessr players, including the reigning world champion, to produce 800 ground truth reasoning chains for 500 query scenes. These expert reasoning chains address hundreds of different discriminative visual attributes such as license plate shape, architecture, and soil properties to name just a few. We evaluate LLM-as-a-judge and VLM-as-a-judge strategies for scoring VLM-generated reasoning chains against our expert reasoning chains and find that Qwen 3 LLM-as-a-judge correlates best with human scoring. Our benchmark reveals that while large, closed-source VLMs such as Gemini and GPT 5 rival human experts at prediction locations, they still lag behind human experts when it comes to producing auditable reasoning chains. Open weights VLMs such as Llama and Qwen catastrophically fail on our benchmark -- they perform only slightly better than a baseline in which an LLM hallucinates a reasoning chain with oracle knowledge of the photo location but no visual information at all. We believe the gap between human experts and VLMs on this task points to VLM limitations at extracting fine-grained visual attributes from high resolution images.

Paperid: 2458, https://arxiv.org/pdf/2601.10611.pdf

Abstract:
Today's strongest video-language models (VLMs) remain proprietary. The strongest open-weight models either rely on synthetic data from proprietary VLMs, effectively distilling from them, or do not disclose their training data or recipe. As a result, the open-source community lacks the foundations needed to improve on the state-of-the-art video (and image) language models. Crucially, many downstream applications require more than just high-level video understanding; they require grounding -- either by pointing or by tracking in pixels. Even proprietary models lack this capability. We present Molmo2, a new family of VLMs that are state-of-the-art among open-source models and demonstrate exceptional new capabilities in point-driven grounding in single image, multi-image, and video tasks. Our key contribution is a collection of 7 new video datasets and 2 multi-image datasets, including a dataset of highly detailed video captions for pre-training, a free-form video Q&A dataset for fine-tuning, a new object tracking dataset with complex queries, and an innovative new video pointing dataset, all collected without the use of closed VLMs. We also present a training recipe for this data utilizing an efficient packing and message-tree encoding scheme, and show bi-directional attention on vision tokens and a novel token-weight strategy improves performance. Our best-in-class 8B model outperforms others in the class of open weight and data models on short videos, counting, and captioning, and is competitive on long-videos. On video-grounding Molmo2 significantly outperforms existing open-weight models like Qwen3-VL (35.5 vs 29.6 accuracy on video counting) and surpasses proprietary models like Gemini 3 Pro on some tasks (38.4 vs 20.0 F1 on video pointing and 56.2 vs 41.1 J&F on video tracking).

Paperid: 2459, https://arxiv.org/pdf/2601.01483.pdf

Abstract:
Parallel test-time scaling typically trains separate generation and verification models, incurring high training and inference costs. We propose Advantage Decoupled Preference Optimization (ADPO), a unified reinforcement learning framework that jointly learns answer generation and self-verification within a single policy. ADPO introduces two innovations: a preference verification reward improving verification capability and a decoupled optimization mechanism enabling synergistic optimization of generation and verification. Specifically, the preference verification reward computes mean verification scores from positive and negative samples as decision thresholds, providing positive feedback when prediction correctness aligns with answer correctness. Meanwhile, the advantage decoupled optimization computes separate advantages for generation and verification, applies token masks to isolate gradients, and combines masked GRPO objectives, preserving generation quality while calibrating verification scores. ADPO achieves up to +34.1% higher verification AUC and -53.5% lower inference time, with significant gains of +2.8%/+1.4% accuracy on MathVista/MMMU, +1.9 cIoU on ReasonSeg, and +1.7%/+1.0% step success rate on AndroidControl/GUI Odyssey.

Paperid: 2460, https://arxiv.org/pdf/2512.23573.pdf

Abstract:
The rapid evolution of generative models has led to a continuous emergence of multimodal safety risks, exposing the limitations of existing defense methods. To address these challenges, we propose ProGuard, a vision-language proactive guard that identifies and describes out-of-distribution (OOD) safety risks without the need for model adjustments required by traditional reactive approaches. We first construct a modality-balanced dataset of 87K samples, each annotated with both binary safety labels and risk categories under a hierarchical multimodal safety taxonomy, effectively mitigating modality bias and ensuring consistent moderation across text, image, and text-image inputs. Based on this dataset, we train our vision-language base model purely through reinforcement learning (RL) to achieve efficient and concise reasoning. To approximate proactive safety scenarios in a controlled setting, we further introduce an OOD safety category inference task and augment the RL objective with a synonym-bank-based similarity reward that encourages the model to generate concise descriptions for unseen unsafe categories. Experimental results show that ProGuard achieves performance comparable to closed-source large models on binary safety classification, substantially outperforms existing open-source guard models on unsafe content categorization. Most notably, ProGuard delivers a strong proactive moderation ability, improving OOD risk detection by 52.6% and OOD risk description by 64.8%.

Paperid: 2461, https://arxiv.org/pdf/2511.18964.pdf

Abstract:
Vision-Language models (VLMs) achieve strong performance on multimodal tasks but often fail at systematic visual reasoning tasks, leading to inconsistent or illogical outputs. Neuro-symbolic methods promise to address this by inducing interpretable logical rules, though they exploit rigid, domain-specific perception modules. We propose Vision-Language Programs (VLP), which combine the perceptual flexibility of VLMs with systematic reasoning of program synthesis. Rather than embedding reasoning inside the VLM, VLP leverages the model to produce structured visual descriptions that are compiled into neuro-symbolic programs. The resulting programs execute directly on images, remain consistent with task constraints, and provide human-interpretable explanations that enable easy shortcut mitigation. Experiments on synthetic and real-world datasets demonstrate that VLPs outperform direct and structured prompting, particularly on tasks requiring complex logical reasoning.

Paperid: 2462, https://arxiv.org/pdf/2511.17004.pdf

Abstract:
Although the cultural dimension has been one of the key aspects in evaluating Vision-Language Models (VLMs), their ability to remain stable across diverse cultural inputs remains largely untested, despite being crucial to support diversity and multicultural societies. Existing evaluations often rely on benchmarks featuring only a singular cultural concept per image, overlooking scenarios where multiple, potentially unrelated cultural cues coexist. To address this gap, we introduce ConfusedTourist, a novel cultural adversarial robustness suite designed to assess VLMs' stability against perturbed geographical cues. Our experiments reveal a critical vulnerability, where accuracy drops heavily under simple image-stacking perturbations and even worsens with its image-generation-based variant. Interpretability analyses further show that these failures stem from systematic attention shifts toward distracting cues, diverting the model from its intended focus. These findings highlight a critical challenge: visual cultural concept mixing can substantially impair even state-of-the-art VLMs, underscoring the urgent need for more culturally robust multimodal understanding.

Paperid: 2463, https://arxiv.org/pdf/2510.16596.pdf

Abstract:
Large Vision-Language Models (LVLMs) excel in diverse cross-modal tasks. However, object hallucination, where models produce plausible but inaccurate object descriptions, remains a significant challenge. In contrast to previous work focusing on LLM components, this paper is the first to trace LVLM hallucinations to visual encoders and identifies three key issues: statistical bias, inherent bias, and vulnerability. To address these challenges, we propose SHIELD, a training-free framework that mitigates hallucinations through three strategies: re-weighting visual tokens to reduce statistical bias, introducing noise-derived tokens to counter inherent bias, and applying adversarial attacks with contrastive decoding to address vulnerability. Experiments demonstrate that SHIELD effectively mitigates object hallucinations across diverse benchmarks and LVLM families. Moreover, SHIELD achieves strong performance on the general LVLM benchmark, highlighting its broad applicability. Code will be released.

Paperid: 2464, https://arxiv.org/pdf/2510.05184.pdf

Abstract:
Foundation models learn highly transferable representations through large-scale pretraining on diverse data. An increasing body of research indicates that these representations exhibit a remarkable degree of similarity across architectures and modalities. In this survey, we investigate the representation potentials of foundation models, defined as the latent capacity of their learned representations to capture task-specific information within a single modality while also providing a transferable basis for alignment and unification across modalities. We begin by reviewing representative foundation models and the key metrics that make alignment measurable. We then synthesize empirical evidence of representation potentials from studies in vision, language, speech, multimodality, and neuroscience. The evidence suggests that foundation models often exhibit structural regularities and semantic consistencies in their representation spaces, positioning them as strong candidates for cross-modal transfer and alignment. We further analyze the key factors that foster representation potentials, discuss open questions, and highlight potential challenges.

Paperid: 2465, https://arxiv.org/pdf/2506.21458.pdf

Abstract:
Can Vision Language Models (VLMs) imagine the full scene from just a few views, like humans do? Humans form spatial mental models, internal representations of unseen space, to reason about layout, perspective, and motion. Our new MindCube benchmark with 21,154 questions across 3,268 images exposes this critical gap, where existing VLMs exhibit near-random performance. Using MindCube, we systematically evaluate how well VLMs build robust spatial mental models through representing positions (cognitive mapping), orientations (perspective-taking), and dynamics (mental simulation for "what-if" movements). We then explore three approaches to help VLMs approximate spatial mental models, including unseen intermediate views, natural language reasoning chains, and cognitive maps. The significant improvement comes from a synergistic approach, "map-then-reason", that jointly trains the model to first generate a cognitive map and then reason upon it. By training models to reason over these internal maps, we boosted accuracy from 37.8% to 60.8% (+23.0%). Adding reinforcement learning pushed performance even further to 70.7% (+32.9%). Our key insight is that such scaffolding of spatial mental models, actively constructing and utilizing internal structured spatial representations with flexible reasoning processes, significantly improves understanding of unobservable space.

Paperid: 2466, https://arxiv.org/pdf/2506.19288.pdf

Abstract:
Automated waterway environment perception is crucial for enabling unmanned surface vessels (USVs) to understand their surroundings and make informed decisions. Most existing waterway perception models primarily focus on instance-level object perception paradigms (e.g., detection, segmentation). However, due to the complexity of waterway environments, current perception datasets and models fail to achieve global semantic understanding of waterways, limiting large-scale monitoring and structured log generation. With the advancement of vision-language models (VLMs), we leverage image captioning to introduce WaterCaption, the first captioning dataset specifically designed for waterway environments. WaterCaption focuses on fine-grained, multi-region long-text descriptions, providing a new research direction for visual geo-understanding and spatial scene cognition. Exactly, it includes 20.2k image-text pair data with 1.8 million vocabulary size. Additionally, we propose Da Yu, an edge-deployable multi-modal large language model for USVs, where we propose a novel vision-to-language projector called Nano Transformer Adaptor (NTA). NTA effectively balances computational efficiency with the capacity for both global and fine-grained local modeling of visual features, thereby significantly enhancing the model's ability to generate long-form textual outputs. Da Yu achieves an optimal balance between performance and efficiency, surpassing state-of-the-art models on WaterCaption and several other captioning benchmarks.

Paperid: 2467, https://arxiv.org/pdf/2506.02354.pdf

Abstract:
Object Navigation (ObjectNav) is a fundamental task in embodied artificial intelligence. Although significant progress has been made in semantic map construction and target direction prediction in current research, redundant exploration and exploration failures remain inevitable. A critical but underexplored direction is the timely termination of exploration to overcome these challenges. We observe a diminishing marginal effect between exploration steps and exploration rates and analyze the cost-benefit relationship of exploration. Inspired by this, we propose RATE-Nav, a Region-Aware Termination-Enhanced method. It includes a geometric predictive region segmentation algorithm and region-Based exploration estimation algorithm for exploration rate calculation. By leveraging the visual question answering capabilities of visual language models (VLMs) and exploration rates enables efficient termination.RATE-Nav achieves a success rate of 67.8% and an SPL of 31.3% on the HM3D dataset. And on the more challenging MP3D dataset, RATE-Nav shows approximately 10% improvement over previous zero-shot methods.

Paperid: 2468, https://arxiv.org/pdf/2505.03153.pdf

Abstract:
The advent of Vision-Language Models (VLMs) in medical image analysis has the potential to help process multimodal inputs and increase performance over traditional inference methods. However, when considering the domain in which these models will be implemented, fairness and robustness are important to ensure the model stays true for any patient. In this paper, we introduce a framework for ensuring robustness and fairness of VLM models. This framework modifies the loss function at training by identifying and adjusting faulty image-text pairs through a Dynamic Bad Pair Mining algorithm and also utilizing Sinkhorn distance to ensure the loss distributions of protected groups do not deviate from the total loss. Experimental testing of our framework shows up to a 8.6\% improvement when looking at equity-scaled AUC.

Paperid: 2469, https://arxiv.org/pdf/2503.22291.pdf

Abstract:
As object detectors are increasingly deployed as black-box cloud services or pre-trained models with restricted access to the original training data, the challenge of zero-shot object-level out-of-distribution (OOD) detection arises. This task becomes crucial in ensuring the reliability of detectors in open-world settings. While existing methods have demonstrated success in image-level OOD detection using pre-trained vision-language models like CLIP, directly applying such models to object-level OOD detection presents challenges due to the loss of contextual information and reliance on image-level alignment. To tackle these challenges, we introduce a new method that leverages visual prompts and text-augmented in-distribution (ID) space construction to adapt CLIP for zero-shot object-level OOD detection. Our method preserves critical contextual information and improves the ability to differentiate between ID and OOD objects, achieving competitive performance across different benchmarks.

Paperid: 2470, https://arxiv.org/pdf/2501.18533.pdf

Abstract:
Large Vision-Language Models (VLMs) have achieved remarkable performance across a wide range of tasks. However, their deployment in safety-critical domains poses significant challenges. Existing safety fine-tuning methods, which focus on textual or multimodal content, fall short in addressing challenging cases or disrupt the balance between helpfulness and harmlessness. Our evaluation highlights a safety reasoning gap: these methods lack safety visual reasoning ability, leading to such bottlenecks. To address this limitation and enhance both visual perception and reasoning in safety-critical contexts, we propose a novel dataset that integrates multi-image inputs with safety Chain-of-Thought (CoT) labels as fine-grained reasoning logic to improve model performance. Specifically, we introduce the Multi-Image Safety (MIS) dataset, an instruction-following dataset tailored for multi-image safety scenarios, consisting of training and test splits. Our experiments demonstrate that fine-tuning InternVL2.5-8B with MIS significantly outperforms both powerful open-source models and API-based models in challenging multi-image tasks requiring safety-related visual reasoning. This approach not only delivers exceptional safety performance but also preserves general capabilities without any trade-offs. Specifically, fine-tuning with MIS increases average accuracy by 0.83% across five general benchmarks and reduces the Attack Success Rate (ASR) on multiple safety benchmarks by a large margin.

Paperid: 2471, https://arxiv.org/pdf/2412.20895.pdf

Abstract:
So far, efficient fine-tuning has become a popular strategy for enhancing the capabilities of foundation models on downstream tasks by learning plug-and-play modules. However, existing methods overlook a crucial issue: if the underlying foundation model is updated, are these plug-and-play modules still effective? In this paper, we first conduct a detailed analysis of various fine-tuning methods on the CLIP in terms of their compatibility with model updates. The study reveals that many high-performing fine-tuning methods fail to be compatible with the upgraded models. To address this, we propose a novel approach, Class-conditioned Context Optimization (ContCoOp), which integrates learnable prompts with class embeddings using an attention layer before inputting them into the text encoder. Consequently, the prompts can dynamically adapt to the changes in embedding space (due to model updates), ensuring continued effectiveness. Extensive experiments over 15 datasets show that our ContCoOp achieves the highest compatibility over the baseline methods, and exhibits robust out-of-distribution generalization.

Paperid: 2472, https://arxiv.org/pdf/2412.11124.pdf

Abstract:
Recent advancements in multimodal large language models (MLLMs) have shown unprecedented capabilities in advancing various vision-language tasks. However, MLLMs face significant challenges with hallucinations, and misleading outputs that do not align with the input data. While existing efforts are paid to combat MLLM hallucinations, several pivotal challenges are still unsolved. First, while current approaches aggressively focus on addressing errors at the perception level, another important type at the cognition level requiring factual commonsense can be overlooked. In addition, existing methods might fall short in finding a more effective way to represent visual input, which is yet a key bottleneck that triggers visual hallucinations. Moreover, MLLMs can frequently be misled by faulty textual inputs and cause hallucinations, while unfortunately, this type of issue has long been overlooked by existing studies. Inspired by human intuition in handling hallucinations, this paper introduces a novel bottom-up reasoning framework. Our framework systematically addresses potential issues in both visual and textual inputs by verifying and integrating perception-level information with cognition-level commonsense knowledge, ensuring more reliable outputs. Extensive experiments demonstrate significant improvements in multiple hallucination benchmarks after integrating MLLMs with the proposed framework. In-depth analyses reveal the great potential of our methods in addressing perception- and cognition-level hallucinations.

Paperid: 2473, https://arxiv.org/pdf/2411.19488.pdf

Abstract:
Chain-of-Thought (CoT) prompting elicits large language models (LLMs) to produce a series of intermediate reasoning steps before arriving at the final answer. However, when transitioning to vision-language models (VLMs), their text-only rationales struggle to express the fine-grained associations with the original image. In this paper, we propose an image-incorporated multimodal Chain-of-Thought, named \textbf{Interleaved-modal Chain-of-Thought (ICoT)}, which generates sequential reasoning steps consisting of paired visual and textual rationales to infer the final answer. Intuitively, the novel ICoT requires VLMs to enable the generation of fine-grained interleaved-modal content, which is hard for current VLMs to fulfill. Considering that the required visual information is usually part of the input image, we propose \textbf{Attention-driven Selection (ADS)} to realize ICoT over existing VLMs. ADS intelligently inserts regions of the input image to generate the interleaved-modal reasoning steps with ignorable additional latency. ADS relies solely on the attention map of VLMs without the need for parameterization, and therefore it is a plug-and-play strategy that can be generalized to a spectrum of VLMs. We apply ADS to realize ICoT on two popular VLMs of different architectures. Extensive evaluations of three benchmarks have shown that ICoT prompting achieves substantial performance (up to 14\%) and interpretability improvements compared to existing multimodal CoT prompting methods.

Paperid: 2474, https://arxiv.org/pdf/2410.12705.pdf

Authors:Genta Indra Winata, Frederikus Hudi, Patrick Amadeus Irawan, David Anugraha, Rifki Afina Putri, Yutong Wang, Adam Nohejl, Ubaidillah Ariq Prathama, Nedjma Ousidhoum, Afifa Amriani, Anar Rzayev, Anirban Das, Ashmari Pramodya, Aulia Adila, Bryan Wilie, Candy Olivia Mawalim, Ching Lam Cheng, Daud Abolade, Emmanuele Chersoni, Enrico Santus, Fariz Ikhwantri, Garry Kuwanto, Hanyang Zhao, Haryo Akbarianto Wibowo, Holy Lovenia, Jan Christian Blaise Cruz, Jan Wira Gotama Putra, Junho Myung, Lucky Susanto, Maria Angelica Riera Machin, Marina Zhukova, Michael Anugraha, Muhammad Farid Adilazuarda, Natasha Santosa, Peerat Limkonchotiwat, Raj Dabre, Rio Alexander Audino, Samuel Cahyawijaya, Shi-Xiong Zhang, Stephanie Yulia Salim, Yi Zhou, Yinxuan Gui, David Ifeoluwa Adelani, En-Shiun Annie Lee, Shogo Okada, Ayu Purwarianti, Alham Fikri Aji, Taro Watanabe, Derry Tanti Wijaya, Alice Oh, Chong-Wah Ngo

Abstract:
Vision Language Models (VLMs) often struggle with culture-specific knowledge, particularly in languages other than English and in underrepresented cultural contexts. To evaluate their understanding of such knowledge, we introduce WorldCuisines, a massive-scale benchmark for multilingual and multicultural, visually grounded language understanding. This benchmark includes a visual question answering (VQA) dataset with text-image pairs across 30 languages and dialects, spanning 9 language families and featuring over 1 million data points, making it the largest multicultural VQA benchmark to date. It includes tasks for identifying dish names and their origins. We provide evaluation datasets in two sizes (12k and 60k instances) alongside a training dataset (1 million instances). Our findings show that while VLMs perform better with correct location context, they struggle with adversarial contexts and predicting specific regional cuisines and languages. To support future research, we release a knowledge base with annotated food entries and images along with the VQA data.

Paperid: 2475, https://arxiv.org/pdf/2410.11758.pdf

Abstract:
We introduce Latent Action Pretraining for general Action models (LAPA), an unsupervised method for pretraining Vision-Language-Action (VLA) models without ground-truth robot action labels. Existing Vision-Language-Action models require action labels typically collected by human teleoperators during pretraining, which significantly limits possible data sources and scale. In this work, we propose a method to learn from internet-scale videos that do not have robot action labels. We first train an action quantization model leveraging VQ-VAE-based objective to learn discrete latent actions between image frames, then pretrain a latent VLA model to predict these latent actions from observations and task descriptions, and finally finetune the VLA on small-scale robot manipulation data to map from latent to robot actions. Experimental results demonstrate that our method significantly outperforms existing techniques that train robot manipulation policies from large-scale videos. Furthermore, it outperforms the state-of-the-art VLA model trained with robotic action labels on real-world manipulation tasks that require language conditioning, generalization to unseen objects, and semantic generalization to unseen instructions. Training only on human manipulation videos also shows positive transfer, opening up the potential for leveraging web-scale data for robotics foundation model.

Paperid: 2476, https://arxiv.org/pdf/2512.24712.pdf

Abstract:
Real-world autonomous driving must adhere to complex human social rules that extend beyond legally codified traffic regulations. Many of these semantic constraints, such as yielding to emergency vehicles, complying with traffic officers' gestures, or stopping for school buses, are intuitive for humans yet difficult to encode explicitly. Although large vision-language models (VLMs) can interpret such semantics, their inference cost makes them impractical for real-time deployment.This work proposes LSRE, a Latent Semantic Rule Encoding framework that converts sparsely sampled VLM judgments into decision boundaries within the latent space of a recurrent world model. By encoding language-defined safety semantics into a lightweight latent classifier, LSRE enables real-time semantic risk assessment at 10 Hz without per-frame VLM queries. Experiments on six semantic-failure scenarios in CARLA demonstrate that LSRE attains semantic risk detection accuracy comparable to a large VLM baseline, while providing substantially earlier hazard anticipation and maintaining low computational latency. LSRE further generalizes to rarely seen semantic-similar test cases, indicating that language-guided latent classification offers an effective and deployable mechanism for semantic safety monitoring in autonomous driving.

Paperid: 2477, https://arxiv.org/pdf/2511.20994.pdf

Abstract:
Multimodal large reasoning models (MLRMs) are increasingly deployed for vision-language tasks that produce explicit intermediate rationales. However, reasoning traces can contain unsafe content even when the final answer is non-harmful, creating deployment risks. Existing multimodal safety guards primarily evaluate only the input question and the final answer, neglecting the intermediate reasoning process. This oversight allows undetected harm, such as biased inferences or policy-violating use of visual context, to emerge during reasoning. We introduce GuardTrace-VL, a vision-aware safety auditor that monitors the full Question-Thinking-Answer (QTA) pipeline via joint image-text analysis, enabling detection of unsafe content as it emerges in the reasoning stage. To support training and evaluation, we construct the GuardTrace dataset, which is generated through diverse prompting strategies and refined via a MLRM- and human-based voting and verification pipeline. Furthermore, we propose a three-stage progressive training scheme combined with the data refinement process, enabling the model to learn nuanced and context-dependent safety preferences according to different risk levels. On our proposed test set covering both in-domain and out-of-domain scenarios, GuardTrace-VL model achieves an F1 score of 93.1% on unsafe reasoning detection tasks, representing a 13.5% improvement in F1 score compared to the previous strongest multimodal safety defense methods. The codes will be made publicly available.

Paperid: 2478, https://arxiv.org/pdf/2511.19899.pdf

Abstract:
Large Vision-Language Models (LVLMs) show promise for scientific applications, yet open-source models still struggle with Scientific Visual Question Answering (SVQA), namely answering questions about figures from scientific papers. A key bottleneck lies in the lack of public, large-scale, high-quality SVQA datasets. Although recent work uses LVLMs to synthesize data at scale, we identify systematic errors in their resulting QA pairs, stemming from LVLMs' inherent limitations and information asymmetry between figures and text. To address these challenges, we propose a verification-centric Generate-then-Verify framework that first generates QA pairs with figure-associated textual context, then applies cross-modal consistency checks against figures along with auxiliary filters to eliminate erroneous pairs. We instantiate this framework to curate VeriSciQA, a dataset of 20,351 QA pairs spanning 20 scientific domains and 12 figure types. VeriSciQA poses a challenging benchmark for open-source models, with a substantial accuracy gap between the leading open-source models (64%) and a proprietary model (82%). Moreover, models fine-tuned on VeriSciQA achieve consistent improvements on SVQA benchmarks, with performance gains that scale with data size and surpass models trained on existing datasets. Human evaluation further validates the superior correctness of VeriSciQA. Together, these evidences demonstrate that continued data expansion by our scalable framework can further advance SVQA capability in the open-source community.

Paperid: 2479, https://arxiv.org/pdf/2510.09976.pdf

Abstract:
Vision-Language-Action (VLA) models such as OpenVLA, Octo, and $π_0$ have shown strong generalization by leveraging large-scale demonstrations, yet their performance is still fundamentally constrained by the quality and coverage of supervised data. Reinforcement learning (RL) provides a promising path for improving and fine-tuning VLAs through online interaction. However, conventional policy gradient methods are computationally infeasible in the context of flow-matching based models due to the intractability of the importance sampling process, which requires explicit computation of policy ratios. To overcome this limitation, we propose Flow Policy Optimization (FPO) algorithm, which reformulates importance sampling by leveraging per-sample changes in the conditional flow-matching objective. Furthermore, FPO achieves stable and scalable online reinforcement fine-tuning of the $π_0$ model by integrating structure-aware credit assignment to enhance gradient efficiency, clipped surrogate objectives to stabilize optimization, multi-step latent exploration to encourage diverse policy updates, and a Q-ensemble mechanism to provide robust value estimation. We evaluate FPO on the LIBERO benchmark and the ALOHA simulation task against supervised, preference-aligned, diffusion-based, autoregressive online RL, and $π_0$-FAST baselines, observing consistent improvements over the imitation prior and strong alternatives with stable learning under sparse rewards. In addition, ablation studies and analyses of the latent space dynamics further highlight the contributions of individual components within FPO, validating the effectiveness of the proposed computational modules and the stable convergence of the conditional flow-matching objective during online RL.

Paperid: 2480, https://arxiv.org/pdf/2510.08238.pdf

Abstract:
The rapid deployment of large language model (LLM)-based agents in real-world applications has raised serious concerns about their trustworthiness. In this work, we reveal the security and robustness vulnerabilities of these agents through backdoor attacks. Distinct from traditional backdoors limited to single-step control, we propose the Chain-of-Trigger Backdoor (CoTri), a multi-step backdoor attack designed for long-horizon agentic control. CoTri relies on an ordered sequence. It starts with an initial trigger, and subsequent ones are drawn from the environment, allowing multi-step manipulation that diverts the agent from its intended task. Experimental results show that CoTri achieves a near-perfect attack success rate (ASR) while maintaining a near-zero false trigger rate (FTR). Due to training data modeling the stochastic nature of the environment, the implantation of CoTri paradoxically enhances the agent's performance on benign tasks and even improves its robustness against environmental distractions. We further validate CoTri on vision-language models (VLMs), confirming its scalability to multimodal agents. Our work highlights that CoTri achieves stable, multi-step control within agents, improving their inherent robustness and task capabilities, which ultimately makes the attack more stealthy and raises potential safty risks.

Paperid: 2481, https://arxiv.org/pdf/2510.00536.pdf

Abstract:
Graphical user interface (GUI) agents built on vision-language models have emerged as a promising approach to automate human-computer workflows. However, they also face the inefficiency challenge as they process long sequences of high-resolution screenshots and solving long-horizon tasks, making inference slow, costly and memory-bound. While key-value (KV) caching can mitigate this, storing the full cache is prohibitive for image-heavy contexts. Existing cache-compression methods are sub-optimal as they do not account for the spatial and temporal redundancy of GUIs. In this work, we first analyze attention patterns in GUI agent workloads and find that, unlike in natural images, attention sparsity is uniformly high across all transformer layers. This insight motivates a simple uniform budget allocation strategy, which we show empirically outperforms more complex layer-varying schemes. Building on this, we introduce GUI-KV, a plug-and-play KV cache compression method for GUI agents that requires no retraining. GUI-KV combines two novel techniques: (i) spatial saliency guidance, which augments attention scores with the L2 norm of hidden states to better preserve semantically important visual tokens, and (ii) temporal redundancy scoring, which projects previous frames' keys onto the current frame's key subspace to preferentially prune redundant history. Across standard GUI agent benchmarks and models, GUI-KV outperforms competitive KV compression baselines, closely matching full-cache accuracy at modest budgets. Notably, in a 5-screenshot setting on the AgentNetBench benchmark, GUI-KV reduces decoding FLOPs by 38.9% while increasing step accuracy by 4.1% over the full-cache baseline. These results demonstrate that exploiting GUI-specific redundancies enables efficient and reliable agent performance.

Paperid: 2482, https://arxiv.org/pdf/2509.20843.pdf

Abstract:
Vision-Language Models(VLMs) have demonstrated significant potential for end-to-end autonomous driving, yet a substantial gap remains between their current capabilities and the reliability necessary for real-world deployment. A critical challenge is their fragility, characterized by hallucinations and poor generalization in out-of-distribution (OOD) scenarios. To bridge this gap, we introduce MTRDrive, a novel framework that integrates procedural driving experiences with a dynamic toolkit to enhance generalization and proactive decision-making. MTRDrive addresses these limitations through a closed-loop system that combines a memory-based experience retrieval mechanism with dynamic toolkits. This synergy enables the model to interact more effectively with its environment, improving both reasoning and decision-making capabilities with the help of our memory-tool synergistic reasoning. Additionally, we introduce a new benchmark based on complex Roadwork construction scenarios to rigorously evaluate zero-shot generalization. Extensive experiments demonstrate the superior effectiveness of our approach. On the public NAVSIM benchmark, our 3B-parameter MTRDrive model achieves an exceptional PDMS of 88.3 without chain-of-thought and sets a state-of-the-art performance bar on high-level planning, with a driving metric score of 79.8\% and a planning accuracy of 82.6\%. Rigorous zero-shot evaluation on the new Roadwork-VLM benchmark shows a strong ability to reason robustly in unseen scenarios, achieving a driving metric score of 80.2\%. These results highlight MTRDrive's potential to advance autonomous driving toward safer and more reliable systems.

Paperid: 2483, https://arxiv.org/pdf/2508.17243.pdf

Abstract:
Large Vision-Language Models (LVLMs) process multimodal inputs consisting of text tokens and vision tokens extracted from images or videos. Due to the rich visual information, a single image can generate thousands of vision tokens, leading to high computational costs during the prefilling stage and significant memory overhead during decoding. Existing methods attempt to prune redundant vision tokens, revealing substantial redundancy in visual representations. However, these methods often struggle in shallow layers due to the lack of sufficient contextual information. We argue that many visual tokens are inherently redundant even in shallow layers and can be safely and effectively pruned with appropriate contextual signals. In this work, we propose CoViPAL, a layer-wise contextualized visual token pruning method that employs a Plug-and-Play Pruning Module (PPM) to predict and remove redundant vision tokens before they are processed by the LVLM. The PPM is lightweight, model-agnostic, and operates independently of the LVLM architecture, ensuring seamless integration with various models. Extensive experiments on multiple benchmarks demonstrate that CoViPAL outperforms training-free pruning methods under equal token budgets and surpasses training-based methods with comparable supervision. CoViPAL offers a scalable and efficient solution to improve inference efficiency in LVLMs without compromising accuracy.

Paperid: 2484, https://arxiv.org/pdf/2508.07814.pdf

Abstract:
With the growing demand for efficient logistics, unmanned aerial vehicles (UAVs) are increasingly being paired with automated guided vehicles (AGVs). While UAVs offer the ability to navigate through dense environments and varying altitudes, they are limited by battery life, payload capacity, and flight duration, necessitating coordinated ground support. Focusing on heterogeneous navigation, SwarmVLM addresses these limitations by enabling semantic collaboration between UAVs and ground robots through impedance control. The system leverages the Vision Language Model (VLM) and the Retrieval-Augmented Generation (RAG) to adjust impedance control parameters in response to environmental changes. In this framework, the UAV acts as a leader using Artificial Potential Field (APF) planning for real-time navigation, while the ground robot follows via virtual impedance links with adaptive link topology to avoid collisions with short obstacles. The system demonstrated a 92% success rate across 12 real-world trials. Under optimal lighting conditions, the VLM-RAG framework achieved 8% accuracy in object detection and selection of impedance parameters. The mobile robot prioritized short obstacle avoidance, occasionally resulting in a lateral deviation of up to 50 cm from the UAV path, which showcases safe navigation in a cluttered setting.

Paperid: 2485, https://arxiv.org/pdf/2508.02151.pdf

Abstract:
Recent breakthroughs in text-to-image diffusion models have significantly enhanced both the visual fidelity and semantic controllability of generated images. However, fine-grained control over aesthetic attributes remains challenging, especially when users require continuous and intensity-specific adjustments. Existing approaches often rely on vague textual prompts, which are inherently ambiguous in expressing both the aesthetic semantics and the desired intensity, or depend on costly human preference data for alignment, limiting their scalability and practicality. To address these limitations, we propose AttriCtrl, a plug-and-play framework for precise and continuous control of aesthetic attributes. Specifically, we quantify abstract aesthetics by leveraging semantic similarity from pre-trained vision-language models, and employ a lightweight value encoder that maps scalar intensities in $[0,1]$ to learnable embeddings within diffusion-based generation. This design enables intuitive and customizable aesthetic manipulation, with minimal training overhead and seamless integration into existing generation pipelines. Extensive experiments demonstrate that AttriCtrl achieves accurate control over individual attributes as well as flexible multi-attribute composition. Moreover, it is fully compatible with popular open-source controllable generation frameworks, showcasing strong integration capability and practical utility across diverse generation scenarios.

Paperid: 2486, https://arxiv.org/pdf/2507.12795.pdf

Abstract:
Scene understanding enables intelligent agents to interpret and comprehend their environment. While existing large vision-language models (LVLMs) for scene understanding have primarily focused on indoor household tasks, they face two significant limitations when applied to outdoor large-scale scene understanding. First, outdoor scenarios typically encompass larger-scale environments observed through various sensors from multiple viewpoints (e.g., bird view and terrestrial view), while existing indoor LVLMs mainly analyze single visual modalities within building-scale contexts from humanoid viewpoints. Second, existing LVLMs suffer from missing multidomain perception outdoor data and struggle to effectively integrate 2D and 3D visual information. To address the aforementioned limitations, we build the first multidomain perception outdoor scene understanding dataset, named \textbf{\underline{SVM-City}}, deriving from multi\textbf{\underline{S}}cale scenarios with multi\textbf{\underline{V}}iew and multi\textbf{\underline{M}}odal instruction tuning data. It contains $420$k images and $4, 811$M point clouds with $567$k question-answering pairs from vehicles, low-altitude drones, high-altitude aerial planes, and satellite. To effectively fuse the multimodal data in the absence of one modality, we introduce incomplete multimodal learning to model outdoor scene understanding and design the LVLM named \textbf{\underline{City-VLM}}. Multimodal fusion is realized by constructing a joint probabilistic distribution space rather than implementing directly explicit fusion operations (e.g., concatenation). Experimental results on three typical outdoor scene understanding tasks show City-VLM achieves $18.14 \%$ performance surpassing existing LVLMs in question-answering tasks averagely. Our method demonstrates pragmatic and generalization performance across multiple outdoor scenes.

Paperid: 2487, https://arxiv.org/pdf/2506.07627.pdf

Abstract:
Large Language Model (LLM)-based Vision-Language Models (VLMs) have substantially extended the boundaries of visual understanding capabilities. However, their high computational demands hinder deployment on resource-constrained edge devices. A key source of inefficiency stems from the VLM's need to process dense and redundant visual information. Visual inputs contain significant regions irrelevant to text semantics, rendering the associated computations ineffective for inference. This paper introduces a novel Event-Priori-Based Vision-Language Model, termed EP-VLM. Its core contribution is a novel mechanism leveraging motion priors derived from dynamic event vision to enhance VLM efficiency. Inspired by human visual cognition, EP-VLM first employs event data to guide the patch-wise sparsification of RGB visual inputs, progressively concentrating VLM computation on salient regions of the visual input. Subsequently, we construct a position-preserving tokenization strategy for the visual encoder within the VLM architecture. This strategy processes the event-guided, unstructured, sparse visual input while accurately preserving positional understanding within the visual input. Experimental results demonstrate that EP-VLM achieves significant efficiency improvements while maintaining nearly lossless accuracy compared to baseline models from the Qwen2-VL series. For instance, against the original Qwen2-VL-2B, EP-VLM achieves 50% FLOPs savings while retaining 98% of the original accuracy on the RealWorldQA dataset. This work demonstrates the potential of event-based vision priors for improving VLM inference efficiency, paving the way for creating more efficient and deployable VLMs for sustainable visual understanding at the edge.

Paperid: 2488, https://arxiv.org/pdf/2505.15298.pdf

Abstract:
Vision-Language Models (VLMs) show promise for autonomous driving, yet their struggle with hallucinations, inefficient reasoning, and limited real-world validation hinders accurate perception and robust step-by-step reasoning. To overcome this, we introduce AgentThink, a pioneering unified framework that, for the first time, integrates Chain-of-Thought (CoT) reasoning with dynamic, agent-style tool invocation for autonomous driving tasks. AgentThink's core innovations include: (i) Structured Data Generation, by establishing an autonomous driving tool library to automatically construct structured, self-verified reasoning data explicitly incorporating tool usage for diverse driving scenarios; (ii) A Two-stage Training Pipeline, employing Supervised Fine-Tuning (SFT) with Group Relative Policy Optimization (GRPO) to equip VLMs with the capability for autonomous tool invocation; and (iii) Agent-style Tool-Usage Evaluation, introducing a novel multi-tool assessment protocol to rigorously evaluate the model's tool invocation and utilization. Experiments on the DriveLMM-o1 benchmark demonstrate AgentThink significantly boosts overall reasoning scores by 53.91% and enhances answer accuracy by 33.54%, while markedly improving reasoning quality and consistency. Furthermore, ablation studies and robust zero-shot/few-shot generalization experiments across various benchmarks underscore its powerful capabilities. These findings highlight a promising trajectory for developing trustworthy and tool-aware autonomous driving models.

Paperid: 2489, https://arxiv.org/pdf/2504.09439.pdf

Abstract:
Recent advances in generative artificial intelligence have enabled the creation of highly realistic image forgeries, raising significant concerns about digital media authenticity. While existing detection methods demonstrate promising results on benchmark datasets, they face critical limitations in real-world applications. First, existing detectors typically fail to detect semantic inconsistencies with the person's identity, such as implausible behaviors or incompatible environmental contexts in given images. Second, these methods rely heavily on low-level visual cues, making them effective for known forgeries but less reliable against new or unseen manipulation techniques. To address these challenges, we present a novel personalized vision-language model (VLM) that integrates low-level visual artifact analysis and high-level semantic inconsistency detection. Unlike previous VLM-based methods, our approach avoids resource-intensive supervised fine-tuning that often struggles to preserve distinct identity characteristics. Instead, we employ a lightweight method that dynamically encodes identity-specific information into specialized identifier tokens. This design enables the model to learn distinct identity characteristics while maintaining robust generalization capabilities. We further enhance detection capabilities through a lightweight detection adapter that extracts fine-grained information from shallow features of the vision encoder, preserving critical low-level evidence. Comprehensive experiments demonstrate that our approach achieves 94.25% accuracy and 94.08% F1 score, outperforming both traditional forgery detectors and general VLMs while requiring only 10 extra tokens.

Paperid: 2490, https://arxiv.org/pdf/2503.10809.pdf

Abstract:
Recent advances in operating system (OS) agents enable vision-language models to interact directly with the graphical user interface of an OS. These multimodal OS agents autonomously perform computer-based tasks in response to a single prompt via application programming interfaces (APIs). Such APIs typically support low-level operations, including mouse clicks, keyboard inputs, and screenshot captures. We introduce a novel attack vector: malicious image patches (MIPs) that have been adversarially perturbed so that, when captured in a screenshot, they cause an OS agent to perform harmful actions by exploiting specific APIs. For instance, MIPs embedded in desktop backgrounds or shared on social media can redirect an agent to a malicious website, enabling further exploitation. These MIPs generalise across different user requests and screen layouts, and remain effective for multiple OS agents. The existence of such attacks highlights critical security vulnerabilities in OS agents, which should be carefully addressed before their widespread adoption.

Paperid: 2491, https://arxiv.org/pdf/2503.08162.pdf

Abstract:
Ensuring safe, comfortable, and efficient planning is crucial for autonomous driving systems. While end-to-end models trained on large datasets perform well in standard driving scenarios, they struggle with complex low-frequency events. Recent Large Language Models (LLMs) and Vision Language Models (VLMs) advancements offer enhanced reasoning but suffer from computational inefficiency. Inspired by the dual-process cognitive model "Thinking, Fast and Slow", we propose $\textbf{FASIONAD}$ -- a novel dual-system framework that synergizes a fast end-to-end planner with a VLM-based reasoning module. The fast system leverages end-to-end learning to achieve real-time trajectory generation in common scenarios, while the slow system activates through uncertainty estimation to perform contextual analysis and complex scenario resolution. Our architecture introduces three key innovations: (1) A dynamic switching mechanism enabling slow system intervention based on real-time uncertainty assessment; (2) An information bottleneck with high-level plan feedback that optimizes the slow system's guidance capability; (3) A bidirectional knowledge exchange where visual prompts enhance the slow system's reasoning while its feedback refines the fast planner's decision-making. To strengthen VLM reasoning, we develop a question-answering mechanism coupled with reward-instruct training strategy. In open-loop experiments, FASIONAD achieves a $6.7\%$ reduction in average $L2$ trajectory error and $28.1\%$ lower collision rate.

Paperid: 2492, https://arxiv.org/pdf/2503.02723.pdf

Abstract:
Swarm robotics plays a crucial role in enabling autonomous operations in dynamic and unpredictable environments. However, a major challenge remains ensuring safe and efficient navigation in environments filled with both dynamic alive (e.g., humans) and dynamic inanimate (e.g., non-living objects) obstacles. In this paper, we propose ImpedanceGPT, a novel system that combines a Vision-Language Model (VLM) with retrieval-augmented generation (RAG) to enable real-time reasoning for adaptive navigation of mini-drone swarms in complex environments. The key innovation of ImpedanceGPT lies in the integration of VLM and RAG, which provides the drones with enhanced semantic understanding of their surroundings. This enables the system to dynamically adjust impedance control parameters in response to obstacle types and environmental conditions. Our approach not only ensures safe and precise navigation but also improves coordination between drones in the swarm. Experimental evaluations demonstrate the effectiveness of the system. The VLM-RAG framework achieved an obstacle detection and retrieval accuracy of 80 % under optimal lighting. In static environments, drones navigated dynamic inanimate obstacles at 1.4 m/s but slowed to 0.7 m/s with increased separation around humans. In dynamic environments, speed adjusted to 1.0 m/s near hard obstacles, while reducing to 0.6 m/s with higher deflection to safely avoid moving humans.

Paperid: 2493, https://arxiv.org/pdf/2412.00142.pdf

Abstract:
Generative Large Multimodal Models (LMMs) like LLaVA and Qwen-VL excel at a wide variety of vision-language (VL) tasks. Despite strong performance, LMMs' generative outputs are not specialized for vision-language classification tasks (i.e., tasks with vision-language inputs and discrete labels) such as image classification and multiple-choice VQA. One key challenge in utilizing LMMs for these tasks is the extraction of useful features from generative LMMs. To overcome this, we propose an approach that leverages multimodal feature extraction from the LMM's latent space. Toward this end, we present Sparse Attention Vectors (SAVs) -- a finetuning-free method that leverages sparse attention head activations (fewer than 5% of the heads) in LMMs as strong feature representations. With only few-shot examples, SAVs demonstrate state-of-the-art performance compared to a variety of few-shot and finetuned baselines on a collection of vision-language classification tasks. Our experiments also imply that SAVs can scale in performance with additional examples and generalize to similar tasks, establishing SAVs as both effective and robust multimodal feature representations.

Paperid: 2494, https://arxiv.org/pdf/2411.13317.pdf

Abstract:
Vision-Language Models (VLMs) have shown remarkable capabilities across diverse visual tasks, including image recognition, video understanding, and Visual Question Answering (VQA) when explicitly trained for these tasks. Despite these advances, we find that present-day VLMs (including the proprietary GPT-4o) lack a fundamental cognitive ability: learning to localize specific objects in a scene by taking into account the context. In this work, we focus on the task of few-shot personalized localization, where a model is given a small set of annotated images (in-context examples) -- each with a category label and bounding box -- and is tasked with localizing the same object type in a query image. Personalized localization can be particularly important in cases of ambiguity of several related objects that can respond to a text or an object that is hard to describe with words. To provoke personalized localization abilities in models, we present a data-centric solution that fine-tunes them using carefully curated data from video object tracking datasets. By leveraging sequences of frames tracking the same object across multiple shots, we simulate instruction-tuning dialogues that promote context awareness. To reinforce this, we introduce a novel regularization technique that replaces object labels with pseudo-names, ensuring the model relies on visual context rather than prior knowledge. Our method significantly enhances the few-shot localization performance of recent VLMs ranging from 7B to 72B in size, without sacrificing generalization, as demonstrated on several benchmarks tailored towards evaluating personalized localization abilities. This work is the first to explore and benchmark personalized few-shot localization for VLMs -- exposing critical weaknesses in present-day VLMs, and laying a foundation for future research in context-driven vision-language applications.

Paperid: 2495, https://arxiv.org/pdf/2411.12787.pdf

Abstract:
Efficient Visual Instruction Fine-Tuning (EVIT) seeks to adapt Multimodal Large Language Models (MLLMs) to downstream tasks with minimal computational overhead. However, as task diversity and complexity increase, EVIT faces significant challenges in resolving data conflicts. To address this limitation, we propose the Dual Low-Rank Adaptation (Dual-LoRA), a holistic-to-local framework that enhances the adapter's capacity to address data conflict through dual structural optimization. Specifically, we utilize two subspaces: a skill space for stable, holistic knowledge retention, and a rank-rectified task space that locally activates the holistic knowledge. Additionally, we introduce Visual Cue Enhancement (VCE), a multi-level local feature aggregation module designed to enrich the vision-language projection with local details. Our approach is both memory- and time-efficient, requiring only 1.16$\times$ the inference time of the standard LoRA method (with injection into the query and value projection layers), and just 73\% of the inference time of a 4-expert LoRA-MoE. Extensive experiments on various downstream tasks and general MLLM benchmarks validate the effectiveness of our proposed methods.

Paperid: 2496, https://arxiv.org/pdf/2411.00850.pdf

Abstract:
Large language models (LLMs) show impressive performance in solving complex language tasks. However, its large number of parameters presents significant challenges for the deployment. So, compressing LLMs to low bits can enable to deploy on resource-constrained devices. To address this problem, we propose gradient-aware weight quantization (GWQ), the first quantization approach for low-bit weight quantization that leverages gradients to localize outliers, requiring only a minimal amount of calibration data for outlier detection. GWQ retains the top 1\% outliers preferentially at FP16 precision, while the remaining non-outlier weights are stored in a low-bit. We widely evaluate GWQ on different task include language modeling, grounding detection, massive multitask language understanding and vision-language question and answering. Results show that models quantified by GWQ performs better than other quantization method. During quantization process, GWQ only need one calibration set to realize effective quant. Also, GWQ achieves 1.2x inference speedup in comparison to the original model and effectively reduces the inference memory.

Paperid: 2497, https://arxiv.org/pdf/2409.17146.pdf

Authors:Matt Deitke, Christopher Clark, Sangho Lee, Rohun Tripathi, Yue Yang, Jae Sung Park, Mohammadreza Salehi, Niklas Muennighoff, Kyle Lo, Luca Soldaini, Jiasen Lu, Taira Anderson, Erin Bransom, Kiana Ehsani, Huong Ngo, YenSung Chen, Ajay Patel, Mark Yatskar, Chris Callison-Burch, Andrew Head, Rose Hendrix, Favyen Bastani, Eli VanderBilt, Nathan Lambert, Yvonne Chou, Arnavi Chheda, Jenna Sparks, Sam Skjonsberg, Michael Schmitz, Aaron Sarnat, Byron Bischoff, Pete Walsh, Chris Newell, Piper Wolters, Tanmay Gupta, Kuo-Hao Zeng, Jon Borchardt, Dirk Groeneveld, Crystal Nam, Sophie Lebrecht, Caitlin Wittlif, Carissa Schoenick, Oscar Michel, Ranjay Krishna, Luca Weihs, Noah A. Smith, Hannaneh Hajishirzi, Ross Girshick, Ali Farhadi, Aniruddha Kembhavi

Abstract:
Today's most advanced vision-language models (VLMs) remain proprietary. The strongest open-weight models rely heavily on synthetic data from proprietary VLMs to achieve good performance, effectively distilling these closed VLMs into open ones. As a result, the community has been missing foundational knowledge about how to build performant VLMs from scratch. We present Molmo, a new family of VLMs that are state-of-the-art in their class of openness. Our key contribution is a collection of new datasets called PixMo, including a dataset of highly detailed image captions for pre-training, a free-form image Q&A dataset for fine-tuning, and an innovative 2D pointing dataset, all collected without the use of external VLMs. The success of our approach relies on careful modeling choices, a well-tuned training pipeline, and, most critically, the quality of our newly collected datasets. Our best-in-class 72B model not only outperforms others in the class of open weight and data models, but also outperforms larger proprietary models including Claude 3.5 Sonnet, and Gemini 1.5 Pro and Flash, second only to GPT-4o based on both academic benchmarks and on a large human evaluation. Our model weights, new datasets, and source code are available at https://molmo.allenai.org/blog.

Paperid: 2498, https://arxiv.org/pdf/2511.18112.pdf

Abstract:
Recent progress in Vision-Language-Action (VLA) models has enabled embodied agents to interpret multimodal instructions and perform complex tasks. However, existing VLAs are mostly confined to short-horizon, table-top manipulation, lacking the memory and reasoning capability required for long-horizon mobile manipulation, where agents must coordinate navigation and manipulation under changing spatial contexts. In this work, we present EchoVLA, a memory-aware VLA model for long-horizon mobile manipulation. EchoVLA incorporates a synergistic declarative memory inspired by the human brain, consisting of a scene memory that maintains a collection of spatial-semantic maps and an episodic memory that stores task-level experiences with multimodal contextual features. During both training and inference, the two memories are individually stored, updated, and retrieved based on current observations, task history, and instructions, and their retrieved representations are fused via coarse- and fine-grained attention to guide mobile-arm diffusion policies. To support large-scale training and evaluation, we further introduce MoMani, an automated benchmark that generates expert-level long-horizon trajectories through multimodal large language model (MLLM)-guided planning and feedback-driven refinement, supplemented with real-robot demonstrations. Experiments in simulated and real-world settings show that EchoVLA improves long-horizon performance, reaching 0.52 SR on manipulation/navigation and 0.31 on mobile manipulation, exceeding $π_{0.5}$ by +0.08 and +0.11.

Paperid: 2499, https://arxiv.org/pdf/2511.14998.pdf

Abstract:
We introduce FinCriticalED (Financial Critical Error Detection), a visual benchmark for evaluating OCR and vision language models on financial documents at the fact level. Financial documents contain visually dense and table heavy layouts where numerical and temporal information is tightly coupled with structure. In high stakes settings, small OCR mistakes such as sign inversion or shifted dates can lead to materially different interpretations, while traditional OCR metrics like ROUGE and edit distance capture only surface level text similarity. \ficriticaled provides 500 image-HTML pairs with expert annotated financial facts covering over seven hundred numerical and temporal facts. It introduces three key contributions. First, it establishes the first fact level evaluation benchmark for financial document understanding, shifting evaluation from lexical overlap to domain critical factual correctness. Second, all annotations are created and verified by financial experts with strict quality control over signs, magnitudes, and temporal expressions. Third, we develop an LLM-as-Judge evaluation pipeline that performs structured fact extraction and contextual verification for visually complex financial documents. We benchmark OCR systems, open source vision language models, and proprietary models on FinCriticalED. Results show that although the strongest proprietary models achieve the highest factual accuracy, substantial errors remain in visually intricate numerical and temporal contexts. Through quantitative evaluation and expert case studies, FinCriticalED provides a rigorous foundation for advancing visual factual precision in financial and other precision critical domains.

Paperid: 2500, https://arxiv.org/pdf/2511.00710.pdf

Abstract:
While Vision-Language Models (VLMs) post-trained with Reinforcement Learning (RL) show impressive general reasoning, their evaluation is often confined to language-dominant tasks (e.g., math). This raises a critical question: can RL post-training truly extend the inherent capability boundary of a base VLM, particularly for visual-centric spatial tasks where it initially fails? To investigate this, we introduce Ariadne, a framework utilizing synthetic mazes for multi-step spatial reasoning where task difficulty (e.g., path length, turns) is precisely controlled. We leverage this controllable environment to train VLMs using Reinforcement Learning with Verified Rewards (RLVR) in a difficulty-aware curriculum. Surprisingly, post-RLVR training, the VLM achieves over 50% accuracy on a problem set where the base model scored 0%, demonstrating that our approach expands the model's initial capability boundary. To assess real-world viability, we evaluate out-of-distribution (OOD) generalization on practical benchmarks. Despite training only on synthetic maze samples, Ariadne achieves significant zero-shot improvements, averaging 16% on MapBench (e.g., museum navigation) and 24% on ReasonMap (subway transfer tasks). These results confirm that our method not only broadens the model's fundamental limits but also enhances its generalization to real-world spatial reasoning. We acknowledge our study is limited to the post-training phase, given the opaqueness of pre-training data, and hope our research motivates further work on specialized, capability-extending alignment.

Paperid: 2501, https://arxiv.org/pdf/2510.13394.pdf

Abstract:
Spatial reasoning ability is crucial for Vision Language Models (VLMs) to support real-world applications in diverse domains including robotics, augmented reality, and autonomous navigation. Unfortunately, existing benchmarks are inadequate in assessing spatial reasoning ability, especially the \emph{intrinsic-dynamic} spatial reasoning which is a fundamental aspect of human spatial cognition. In this paper, we propose a unified benchmark, \textbf{Spatial-DISE}, based on a cognitively grounded taxonomy that categorizes tasks into four fundamental quadrants: \textbf{I}ntrinsic-\textbf{S}tatic, Intrinsic-\textbf{D}ynamic, \textbf{E}xtrinsic-Static, and Extrinsic-Dynamic spatial reasoning. Moreover, to address the issue of data scarcity, we develop a scalable and automated pipeline to generate diverse and verifiable spatial reasoning questions, resulting in a new \textbf{Spatial-DISE} dataset that includes Spatial-DISE Bench (559 evaluation VQA pairs) and Spatial-DISE-12K (12K+ training VQA pairs). Our comprehensive evaluation across 28 state-of-the-art VLMs reveals that, current VLMs have a large and consistent gap to human competence, especially on multi-step multi-view spatial reasoning. Spatial-DISE offers a robust framework, valuable dataset, and clear direction for future research toward human-like spatial intelligence. Benchmark, dataset, and code will be publicly released.

Paperid: 2502, https://arxiv.org/pdf/2510.10975.pdf

Abstract:
Vision-Language-Action (VLA) models have become a prominent paradigm for embodied intelligence, yet further performance improvements typically rely on scaling up training data and model size -- an approach that is prohibitively expensive for robotics and fundamentally limited by data collection costs.We address this limitation with $\mathbf{RoVer}$, an embodied test-time scaling framework that uses a $\mathbf{Ro}$bot Process Reward Model (PRM) as a Test-Time $\mathbf{Ver}$ifier to enhance the capabilities of existing VLA models without modifying their architectures or weights. Specifically, RoVer (i) assigns scalar-based process rewards to evaluate the reliability of candidate actions, and (ii) predicts an action-space direction for candidate expansion/refinement. During inference, RoVer generates multiple candidate actions concurrently from the base policy, expands them along PRM-predicted directions, and then scores all candidates with PRM to select the optimal action for execution. Notably, by caching shared perception features, it can amortize perception cost and evaluate more candidates under the same test-time computational budget. Essentially, our approach effectively transforms available computing resources into better action decision-making, realizing the benefits of test-time scaling without extra training overhead. Our contributions are threefold: (1) a general, plug-and-play test-time scaling framework for VLAs; (2) a PRM that jointly provides scalar process rewards and an action-space direction to guide exploration; and (3) an efficient direction-guided sampling strategy that leverages a shared perception cache to enable scalable candidate generation and selection during inference.

Paperid: 2503, https://arxiv.org/pdf/2510.10962.pdf

Abstract:
Mixture-of-Experts (MoE) effectively scales large language models (LLMs) and vision-language models (VLMs) by increasing capacity through sparse activation. However, preloading all experts into memory and activating multiple experts per input introduces significant computational and memory overhead, making the expert module a major contributor to model size and inference cost. To address this, we propose MC# (Mixture-Compressor-sharp), a framework that combines static quantization and dynamic expert pruning by leveraging the significance of experts and tokens for aggressive compression of MoE-LLMs/VLMs. To reduce storage and loading costs, we introduce Pre-Loading Mixed-Precision Quantization (PMQ), which optimizes bit allocation via linear programming, balancing expert importance and quantization error for a Pareto-optimal trade-off between size and performance. To reduce runtime computation, Online Top-any Pruning (OTP) uses Gumbel-Softmax sampling to dynamically select a subset of experts per token, enabling fine-grained control over activation. By combining PMQ's static bit-width optimization with OTP's dynamic routing, MC# achieves extreme compression with minimal accuracy loss. On DeepSeek-VL2, MC# achieves a 6.2 times weight reduction at 2.57 average bits with only a 1.7% accuracy drop across five multimodal benchmarks. Additionally, OTP reduces expert activation over 20% with less than 1% performance degradation, demonstrating strong potential for efficient MoE-based model deployment.

Paperid: 2504, https://arxiv.org/pdf/2510.08789.pdf

Abstract:
Video quality assessment (VQA) is a fundamental computer vision task that aims to predict the perceptual quality of a given video in alignment with human judgments. Existing performant VQA models trained with direct score supervision suffer from (1) poor generalization across diverse content and tasks, ranging from user-generated content (UGC), short-form videos, to AI-generated content (AIGC), (2) limited interpretability, and (3) lack of extensibility to novel use cases or content types. We propose Q-Router, an agentic framework for universal VQA with a multi-tier model routing system. Q-Router integrates a diverse set of expert models and employs vision--language models (VLMs) as real-time routers that dynamically reason and then ensemble the most appropriate experts conditioned on the input video semantics. We build a multi-tiered routing system based on the computing budget, with the heaviest tier involving a specific spatiotemporal artifacts localization for interpretability. This agentic design enables Q-Router to combine the complementary strengths of specialized experts, achieving both flexibility and robustness in delivering consistent performance across heterogeneous video sources and tasks. Extensive experiments demonstrate that Q-Router matches or surpasses state-of-the-art VQA models on a variety of benchmarks, while substantially improving generalization and interpretability. Moreover, Q-Router excels on the quality-based question answering benchmark, Q-Bench-Video, highlighting its promise as a foundation for next-generation VQA systems. Finally, we show that Q-Router capably localizes spatiotemporal artifacts, showing potential as a reward function for post-training video generation models.

Paperid: 2505, https://arxiv.org/pdf/2510.04479.pdf

Abstract:
Vision-Language Models (VLMs) have achieved significant progress in multimodal understanding tasks, demonstrating strong capabilities particularly in general tasks such as image captioning and visual reasoning. However, when dealing with specialized cultural heritage domains like 3D vase artifacts, existing models face severe data scarcity issues and insufficient domain knowledge limitations. Due to the lack of targeted training data, current VLMs struggle to effectively handle such culturally significant specialized tasks. To address these challenges, we propose the VaseVQA-3D dataset, which serves as the first 3D visual question answering dataset for ancient Greek pottery analysis, collecting 664 ancient Greek vase 3D models with corresponding question-answer data and establishing a complete data construction pipeline. We further develop the VaseVLM model, enhancing model performance in vase artifact analysis through domain-adaptive training. Experimental results validate the effectiveness of our approach, where we improve by 12.8% on R@1 metrics and by 6.6% on lexical similarity compared with previous state-of-the-art on the VaseVQA-3D dataset, significantly improving the recognition and understanding of 3D vase artifacts, providing new technical pathways for digital heritage preservation research.

Paperid: 2506, https://arxiv.org/pdf/2510.01494.pdf

Abstract:
The field of adversarial robustness has long established that adversarial examples can successfully transfer between image classifiers and that text jailbreaks can successfully transfer between language models (LMs). However, a pair of recent studies reported being unable to successfully transfer image jailbreaks between vision-language models (VLMs). To explain this striking difference, we propose a fundamental distinction regarding the transferability of attacks against machine learning models: attacks in the input data-space can transfer, whereas attacks in model representation space do not, at least not without geometric alignment of representations. We then provide theoretical and empirical evidence of this hypothesis in four different settings. First, we mathematically prove this distinction in a simple setting where two networks compute the same input-output map but via different representations. Second, we construct representation-space attacks against image classifiers that are as successful as well-known data-space attacks, but fail to transfer. Third, we construct representation-space attacks against LMs that successfully jailbreak the attacked models but again fail to transfer. Fourth, we construct data-space attacks against VLMs that successfully transfer to new VLMs, and we show that representation space attacks can transfer when VLMs' latent geometries are sufficiently aligned in post-projector space. Our work reveals that adversarial transfer is not an inherent property of all attacks but contingent on their operational domain - the shared data-space versus models' unique representation spaces - a critical insight for building more robust models.

Paperid: 2507, https://arxiv.org/pdf/2509.26004.pdf

Abstract:
Pixel-level recognition of objects manipulated by the user from egocentric images enables key applications spanning assistive technologies, industrial safety, and activity monitoring. However, progress in this area is currently hindered by the scarcity of annotated datasets, as existing approaches rely on costly manual labels. In this paper, we propose to learn human-object interaction detection leveraging narrations -- natural language descriptions of the actions performed by the camera wearer which contain clues about manipulated objects (e.g., "I am pouring vegetables from the chopping board to the pan"). Narrations provide a form of weak supervision that is cheap to acquire and readily available in state-of-the-art egocentric datasets. We introduce Narration-Supervised in-Hand Object Segmentation (NS-iHOS), a novel task where models have to learn to segment in-hand objects by learning from natural-language narrations. Narrations are then not employed at inference time. We showcase the potential of the task by proposing Weakly-Supervised In-hand Object Segmentation from Human Narrations (WISH), an end-to-end model distilling knowledge from narrations to learn plausible hand-object associations and enable in-hand object segmentation without using narrations at test time. We benchmark WISH against different baselines based on open-vocabulary object detectors and vision-language models, showing the superiority of its design. Experiments on EPIC-Kitchens and Ego4D show that WISH surpasses all baselines, recovering more than 50% of the performance of fully supervised methods, without employing fine-grained pixel-wise annotations.

Paperid: 2508, https://arxiv.org/pdf/2509.23344.pdf

Abstract:
Diagnosing and managing oral diseases necessitate advanced visual interpretation across diverse imaging modalities and integrated information synthesis. While current AI models excel at isolated tasks, they often fall short in addressing the complex, multimodal requirements of comprehensive clinical dental practice. Here we introduce DentVLM, a multimodal vision-language model engineered for expert-level oral disease diagnosis. DentVLM was developed using a comprehensive, large-scale, bilingual dataset of 110,447 images and 2.46 million visual question-answering (VQA) pairs. The model is capable of interpreting seven 2D oral imaging modalities across 36 diagnostic tasks, significantly outperforming leading proprietary and open-source models by 19.6% higher accuracy for oral diseases and 27.9% for malocclusions. In a clinical study involving 25 dentists, evaluating 1,946 patients and encompassing 3,105 QA pairs, DentVLM surpassed the diagnostic performance of 13 junior dentists on 21 of 36 tasks and exceeded that of 12 senior dentists on 12 of 36 tasks. When integrated into a collaborative workflow, DentVLM elevated junior dentists' performance to senior levels and reduced diagnostic time for all practitioners by 15-22%. Furthermore, DentVLM exhibited promising performance across three practical utility scenarios, including home-based dental health management, hospital-based intelligent diagnosis and multi-agent collaborative interaction. These findings establish DentVLM as a robust clinical decision support tool, poised to enhance primary dental care, mitigate provider-patient imbalances, and democratize access to specialized medical expertise within the field of dentistry.

Paperid: 2509, https://arxiv.org/pdf/2509.14033.pdf

Abstract:
We introduce SAIL-VL2, an open-suite vision-language foundation model (LVM) for comprehensive multimodal understanding and reasoning. As the successor to SAIL-VL, SAIL-VL2 achieves state-of-the-art performance at the 2B and 8B parameter scales across diverse image and video benchmarks, demonstrating strong capabilities from fine-grained perception to complex reasoning. Its effectiveness is driven by three core innovations. First, a large-scale data curation pipeline with scoring and filtering strategies enhances both quality and distribution across captioning, OCR, QA, and video data, improving training efficiency. Second, a progressive training framework begins with a powerful pre-trained vision encoder (SAIL-ViT), advances through multimodal pre-training, and culminates in a thinking-fusion SFT-RL hybrid paradigm that systematically strengthens model capabilities. Third, architectural advances extend beyond dense LLMs to efficient sparse Mixture-of-Experts (MoE) designs. With these contributions, SAIL-VL2 demonstrates competitive performance across 106 datasets and achieves state-of-the-art results on challenging reasoning benchmarks such as MMMU and MathVista. Furthermore, on the OpenCompass leaderboard, SAIL-VL2-2B ranks first among officially released open-source models under the 4B parameter scale, while serving as an efficient and extensible foundation for the open-source multimodal community.

Paperid: 2510, https://arxiv.org/pdf/2509.08500.pdf

Abstract:
Using effective generalization capabilities of vision language models (VLMs) in context-specific dynamic tasks for embodied artificial intelligence remains a significant challenge. Although supervised fine-tuned models can better align with the real physical world, they still exhibit sluggish responses and hallucination issues in dynamically changing environments, necessitating further alignment. Existing post-SFT methods, reliant on reinforcement learning and chain-of-thought (CoT) approaches, are constrained by sparse rewards and action-only optimization, resulting in low sample efficiency, poor consistency, and model degradation. To address these issues, this paper proposes Thought-Centric Preference Optimization (TCPO) for effective embodied decision-making. Specifically, TCPO introduces a stepwise preference-based optimization approach, transforming sparse reward signals into richer step sample pairs. It emphasizes the alignment of the model's intermediate reasoning process, mitigating the problem of model degradation. Moreover, by incorporating Action Policy Consistency Constraint (APC), it further imposes consistency constraints on the model output. Experiments in the ALFWorld environment demonstrate an average success rate of 26.67%, achieving a 6% improvement over RL4VLM and validating the effectiveness of our approach in mitigating model degradation after fine-tuning. These results highlight the potential of integrating preference-based learning techniques with CoT processes to enhance the decision-making capabilities of vision-language models in embodied agents.

Paperid: 2511, https://arxiv.org/pdf/2509.00039.pdf

Abstract:
The success of large-scale visual language pretraining (VLP) models has driven widespread adoption of image-text retrieval tasks. However, their deployment on mobile devices remains limited due to large model sizes and computational complexity. We propose Adaptive Multi-Modal Multi-Teacher Knowledge Distillation (AMMKD), a novel framework that integrates multi-modal feature fusion, multi-teacher distillation, and adaptive optimization to deliver lightweight yet effective retrieval models. Specifically, our method begins with a feature fusion network that extracts and merges discriminative features from both the image and text modalities. To reduce model parameters and further improve performance, we design a multi-teacher knowledge distillation framework to pre-train two CLIP teacher models. We decouple modalities by pre-computing and storing text features as class vectors via the teacher text encoder to enhance efficiency. To better align teacher and student outputs, we apply KL scatter for probability distribution matching. Finally, we design an adaptive dynamic weighting scheme that treats multi-teacher distillation as a multi-objective optimization problem. By leveraging gradient space diversity, we dynamically adjust the influence of each teacher, reducing conflicts and guiding the student toward more optimal learning directions. Extensive experiments on three benchmark datasets demonstrate that AMMKD achieves superior performance while significantly reducing model complexity, validating its effectiveness and flexibility.

Paperid: 2512, https://arxiv.org/pdf/2508.17290.pdf

Abstract:
Recent advancements in large vision-language models (VLMs) have primarily focused on English, with limited attention given to other languages. To address this gap, we introduce MEENA (also known as PersianMMMU), the first dataset designed to evaluate Persian VLMs across scientific, reasoning, and human-level understanding tasks. Our dataset comprises approximately 7,500 Persian and 3,000 English questions, covering a wide range of topics such as reasoning, mathematics, physics, diagrams, charts, and Persian art and literature. Key features of MEENA include: (1) diverse subject coverage spanning various educational levels, from primary to upper secondary school, (2) rich metadata, including difficulty levels and descriptive answers, (3) original Persian data that preserves cultural nuances, (4) a bilingual structure to assess cross-linguistic performance, and (5) a series of diverse experiments assessing various capabilities, including overall performance, the model's ability to attend to images, and its tendency to generate hallucinations. We hope this benchmark contributes to enhancing VLM capabilities beyond English.

Paperid: 2513, https://arxiv.org/pdf/2508.13223.pdf

Abstract:
The spreading of AI-generated images (AIGI), driven by advances in generative AI, poses a significant threat to information security and public trust. Existing AIGI detectors, while effective against images in clean laboratory settings, fail to generalize to in-the-wild scenarios. These real-world images are noisy, varying from ``obviously fake" images to realistic ones derived from multiple generative models and further edited for quality control. We address in-the-wild AIGI detection in this paper. We introduce Mirage, a challenging benchmark designed to emulate the complexity of in-the-wild AIGI. Mirage is constructed from two sources: (1) a large corpus of Internet-sourced AIGI verified by human experts, and (2) a synthesized dataset created through the collaboration between multiple expert generators, closely simulating the realistic AIGI in the wild. Building on this benchmark, we propose Mirage-R1, a vision-language model with heuristic-to-analytic reasoning, a reflective reasoning mechanism for AIGI detection. Mirage-R1 is trained in two stages: a supervised-fine-tuning cold start, followed by a reinforcement learning stage. By further adopting an inference-time adaptive thinking strategy, Mirage-R1 is able to provide either a quick judgment or a more robust and accurate conclusion, effectively balancing inference speed and performance. Extensive experiments show that our model leads state-of-the-art detectors by 5% and 10% on Mirage and the public benchmark, respectively. The benchmark and code will be made publicly available.

Paperid: 2514, https://arxiv.org/pdf/2507.23543.pdf

Abstract:
Visual relation detection (VRD) is the task of identifying the relationships between objects in a scene. VRD models trained solely on relation detection data struggle to generalize beyond the relations on which they are trained. While prompt tuning has been used to adapt vision-language models (VLMs) for VRD, it uses handcrafted prompts and struggles with novel or complex relations. We argue that instruction tuning offers a more effective solution by fine-tuning VLMs on diverse instructional data. We thus introduce ART, an Adaptive Relation Tuning framework that adapts VLMs for VRD through instruction tuning and strategic instance selection. By converting VRD datasets into an instruction tuning format and employing an adaptive sampling algorithm, ART directs the VLM to focus on informative relations while maintaining generalizability. Specifically, we focus on the relation classification, where subject-object boxes are given and the model predicts the predicate between them. We tune on a held-in set and evaluate across multiple held-out datasets of varying complexity. Our approach strongly improves over its baselines and can infer unseen relation concepts, a capability absent in mainstream VRD methods. We demonstrate ART's practical value by using the predicted relations for segmenting complex scenes.

Paperid: 2515, https://arxiv.org/pdf/2507.18053.pdf

Abstract:
Resource Consumption Attacks (RCAs) have emerged as a significant threat to the deployment of Large Language Models (LLMs). With the integration of vision modalities, additional attack vectors exacerbate the risk of RCAs in large vision-language models (LVLMs). However, existing red-teaming studies have largely overlooked visual inputs as a potential attack surface, resulting in insufficient mitigation strategies against RCAs in LVLMs. To address this gap, we propose RECALLED (\textbf{RE}source \textbf{C}onsumption \textbf{A}ttack on \textbf{L}arge Vision-\textbf{L}anguag\textbf{E} Mo\textbf{D}els), the first approach for exploiting visual modalities to trigger unbounded RCAs red-teaming. First, we present \textit{Vision Guided Optimization}, a fine-grained pixel-level optimization, to obtain \textit{Output Recall} adversarial perturbations, which can induce repeating output. Then, we inject the perturbations into visual inputs, triggering unbounded generations to achieve the goal of RCAs. Additionally, we introduce \textit{Multi-Objective Parallel Losses} to generate universal attack templates and resolve optimization conflicts when intending to implement parallel attacks. Empirical results demonstrate that RECALLED increases service response latency by over 26 $\uparrow$, resulting in an additional 20\% increase in GPU utilization and memory consumption. Our study exposes security vulnerabilities in LVLMs and establishes a red-teaming framework that can facilitate future defense development against RCAs.

Paperid: 2516, https://arxiv.org/pdf/2506.21509.pdf

Abstract:
Large Vision-Language Models (LVLMs) have demonstrated significant advancements in multimodal understanding, yet they are frequently hampered by hallucination-the generation of text that contradicts visual input. Existing training-free decoding strategies exhibit critical limitations, including the use of static constraints that do not adapt to semantic drift during generation, inefficiency stemming from the need for multiple forward passes, and degradation of detail due to overly rigid intervention rules. To overcome these challenges, this paper introduces Dynamic Logits Calibration (DLC), a novel training-free decoding framework designed to dynamically align text generation with visual evidence at inference time. At the decoding phase, DLC step-wise employs CLIP to assess the semantic alignment between the input image and the generated text sequence. Then, the Relative Visual Advantage (RVA) of candidate tokens is evaluated against a dynamically updated contextual baseline, adaptively adjusting output logits to favor tokens that are visually grounded. Furthermore, an adaptive weighting mechanism, informed by a real-time context alignment score, carefully balances the visual guidance while ensuring the overall quality of the textual output. Extensive experiments conducted across diverse benchmarks and various LVLM architectures (such as LLaVA, InstructBLIP, and MiniGPT-4) demonstrate that DLC significantly reduces hallucinations, outperforming current methods while maintaining high inference efficiency by avoiding multiple forward passes. Overall, we present an effective and efficient decoding-time solution to mitigate hallucinations, thereby enhancing the reliability of LVLMs for more practices. Code will be released on Github.

Paperid: 2517, https://arxiv.org/pdf/2506.18322.pdf

Abstract:
Finetuning can cause spurious correlations to arise between non-essential features and the target labels, but benchmarks to study these effects involve contrived settings and narrow tasks. In contrast, we consider spurious correlations in multi-modal Large Vision Language Models (LVLMs) pretrained on extensive and diverse datasets without explicit task supervision. We develop a benchmark by sourcing GPT-4o errors on real-world visual-question-answering (VQA) benchmarks, then curating a subset through LVLM-human annotation and synthetic counterfactual evaluation to identify errors caused by spurious correlations. This process yields SpuriVerse, a novel benchmark comprised of 124 distinct types of spurious correlations extracted from real-world datasets, each containing 1 realistic and 10 synthetic VQA samples for a total of 1364 multiple choice questions. We evaluate 15 open and closed-source LVLMs on SpuriVerse, finding that even state-of-the-art closed-source models struggle significantly, achieving at best only 37.1% accuracy. Fine-tuning on synthetic examples that emphasize the spurious correlation improves performance to 78.40%, suggesting that training on diverse spurious patterns generalizes to unseen situations: models appear to learn to avoid "shortcuts" and attend to the overall image context.

Paperid: 2518, https://arxiv.org/pdf/2506.15645.pdf

Abstract:
Recent Multimodal Large Language Models (MLLMs) excel on benchmark vision-language tasks, yet little is known about how input visual quality shapes their responses. Does higher perceptual quality of images already translate to better MLLM understanding? We conduct the first systematic study spanning leading MLLMs and a suite of vision-language benchmarks, applying controlled degradations and stylistic shifts to each image. Surprisingly, we uncover a visual-quality paradox: model, task, and even individual-instance performance can improve when images deviate from human-perceived fidelity. Off-the-shelf restoration pipelines fail to reconcile these idiosyncratic preferences. To close the gap, we introduce Visual-Quality Test-Time Tuning (VQ-TTT)-a lightweight adaptation module that: (1) inserts a learnable, low-rank kernel before the frozen vision encoder to modulate frequency content; and (2) fine-tunes only shallow vision-encoder layers via LoRA. VQ-TTT dynamically adjusts each input image in a single forward pass, aligning it with task-specific model preferences. Across the evaluated MLLMs and all datasets, VQ-TTT lifts significant average accuracy, with no external models, cached features, or extra training data. These findings redefine ``better'' visual inputs for MLLMs and highlight the need for adaptive, rather than universally ``clean'', imagery, in the new era of AI being the main data customer.

Paperid: 2519, https://arxiv.org/pdf/2506.08822.pdf

Abstract:
Generative modeling-based visuomotor policies have been widely adopted in robotic manipulation attributed to their ability to model multimodal action distributions. However, the high inference cost of multi-step sampling limits their applicability in real-time robotic systems. To address this issue, existing approaches accelerate the sampling process in generative modeling-based visuomotor policies by adapting acceleration techniques originally developed for image generation. Despite this progress, a major distinction remains: image generation typically involves producing independent samples without temporal dependencies, whereas robotic manipulation involves generating time-series action trajectories that require continuity and temporal coherence. To effectively exploit temporal information in robotic manipulation, we propose FreqPolicy, a novel approach that first imposes frequency consistency constraints on flow-based visuomotor policies. Our work enables the action model to capture temporal structure effectively while supporting efficient, high-quality one-step action generation. We introduce a frequency consistency constraint that enforces alignment of frequency-domain action features across different timesteps along the flow, thereby promoting convergence of one-step action generation toward the target distribution. In addition, we design an adaptive consistency loss to capture structural temporal variations inherent in robotic manipulation tasks. We assess FreqPolicy on 53 tasks across 3 simulation benchmarks, proving its superiority over existing one-step action generators. We further integrate FreqPolicy into the vision-language-action (VLA) model and achieve acceleration without performance degradation on the 40 tasks of Libero. Besides, we show efficiency and effectiveness in real-world robotic scenarios with an inference frequency 93.5Hz. The code will be publicly available.

Paperid: 2520, https://arxiv.org/pdf/2506.08708.pdf

Abstract:
While vision-language models (VLMs) have demonstrated promising capabilities in reasoning and planning for embodied agents, their ability to comprehend physical phenomena, particularly within structured 3D environments, remains severely limited. To close this gap, we introduce PhyBlock, a progressive benchmark designed to assess VLMs on physical understanding and planning through robotic 3D block assembly tasks. PhyBlock integrates a novel four-level cognitive hierarchy assembly task alongside targeted Visual Question Answering (VQA) samples, collectively aimed at evaluating progressive spatial reasoning and fundamental physical comprehension, including object properties, spatial relationships, and holistic scene understanding. PhyBlock includes 2600 block tasks (400 assembly tasks, 2200 VQA tasks) and evaluates models across three key dimensions: partial completion, failure diagnosis, and planning robustness. We benchmark 21 state-of-the-art VLMs, highlighting their strengths and limitations in physically grounded, multi-step planning. Our empirical findings indicate that the performance of VLMs exhibits pronounced limitations in high-level planning and reasoning capabilities, leading to a notable decline in performance for the growing complexity of the tasks. Error analysis reveals persistent difficulties in spatial orientation and dependency reasoning. Surprisingly, chain-of-thought prompting offers minimal improvements, suggesting spatial tasks heavily rely on intuitive model comprehension. We position PhyBlock as a unified testbed to advance embodied reasoning, bridging vision-language understanding and real-world physical problem-solving.

Paperid: 2521, https://arxiv.org/pdf/2506.07778.pdf

Abstract:
The advancement in large language models (LLMs) and large vision models has fueled the rapid progress in multi-modal vision-language reasoning capabilities. However, existing vision-language models (VLMs) remain challenged by compositional visual reasoning. This paper presents VLAgent, a neuro-symbolic approach to developing a Vision-Language Agent system for efficient compositional visual reasoning with three novel features. First, VLAgent develops an interpretable visualization-enhanced two-stage neuro-symbolic reasoning system. The first stage is managed by a front-end engine that generates a structured visual reasoning plan (symbolic program script) for each compositional visual reasoning task by utilizing a pre-trained LLM powered with few-shot chain-of-thought in-context learning. The second stage is managed by a high-performance back-end engine. It transforms the planning script into executable code based on visual input (image or video) and the combination of neural models and symbolic functions and then performs a sequence of actions for the compositional visual reason task. Second, to ensure and enhance the quality of mapping the logic plan to a sequence of executable instructions, VLAgent introduces the SS-parser, which examines the syntax and semantic correctness of the planning script, detects and repairs the logic errors found in the LLM-generated logic plan before generating the executable program. Third, VLAgent introduces the execution verifier in critical reasoning steps to validate and refine its compositional reasoning results in a stepwise manner, for example, ensemble methods for critical visual reasoning and caption analysis for low-confidence compositional reasoning. Extensive experiments on six visual benchmarks compared to a dozen SoTA visual reasoning models show that VLAgent outperforms existing representative approaches to compositional visual reasoning.

Paperid: 2522, https://arxiv.org/pdf/2506.07564.pdf

Abstract:
Recent advances in large language models (LLMs) and vision-language models (VLMs) have enabled powerful autonomous agents capable of complex reasoning and multi-modal tool use. Despite their growing capabilities, today's agent frameworks remain fragile, lacking principled mechanisms for secure information flow, reliability, and multi-agent coordination. In this work, we introduce SAFEFLOW, a new protocol-level framework for building trustworthy LLM/VLM-based agents. SAFEFLOW enforces fine-grained information flow control (IFC), precisely tracking provenance, integrity, and confidentiality of all the data exchanged between agents, tools, users, and environments. By constraining LLM reasoning to respect these security labels, SAFEFLOW prevents untrusted or adversarial inputs from contaminating high-integrity decisions. To ensure robustness in concurrent multi-agent settings, SAFEFLOW introduces transactional execution, conflict resolution, and secure scheduling over shared state, preserving global consistency across agents. We further introduce mechanisms, including write-ahead logging, rollback, and secure caches, that further enhance resilience against runtime errors and policy violations. To validate the performances, we built SAFEFLOWBENCH, a comprehensive benchmark suite designed to evaluate agent reliability under adversarial, noisy, and concurrent operational conditions. Extensive experiments demonstrate that agents built with SAFEFLOW maintain impressive task performance and security guarantees even in hostile environments, substantially outperforming state-of-the-art. Together, SAFEFLOW and SAFEFLOWBENCH lay the groundwork for principled, robust, and secure agent ecosystems, advancing the frontier of reliable autonomy.

Paperid: 2523, https://arxiv.org/pdf/2506.06677.pdf

Abstract:
Recent advances in vision-language models (VLMs) have enabled instruction-conditioned robotic systems with improved generalization. However, most existing work focuses on reactive System 1 policies, underutilizing VLMs' strengths in semantic reasoning and long-horizon planning. These System 2 capabilities-characterized by deliberative, goal-directed thinking-remain under explored due to the limited temporal scale and structural complexity of current benchmarks. To address this gap, we introduce RoboCerebra, a benchmark for evaluating high-level reasoning in long-horizon robotic manipulation. RoboCerebra includes: (1) a large-scale simulation dataset with extended task horizons and diverse subtask sequences in household environments; (2) a hierarchical framework combining a high-level VLM planner with a low-level vision-language-action (VLA) controller; and (3) an evaluation protocol targeting planning, reflection, and memory through structured System 1-System 2 interaction. The dataset is constructed via a top-down pipeline, where GPT generates task instructions and decomposes them into subtask sequences. Human operators execute the subtasks in simulation, yielding high-quality trajectories with dynamic object variations. Compared to prior benchmarks, RoboCerebra features significantly longer action sequences and denser annotations. We further benchmark state-of-the-art VLMs as System 2 modules and analyze their performance across key cognitive dimensions, advancing the development of more capable and generalizable robotic planners.

Paperid: 2524, https://arxiv.org/pdf/2506.03574.pdf

Abstract:
Robots deployed in dynamic environments must be able to not only follow diverse language instructions but flexibly adapt when user intent changes mid-execution. While recent Vision-Language-Action (VLA) models have advanced multi-task learning and instruction following, they typically assume static task intent, failing to respond when new instructions arrive during ongoing execution. This limitation hinders natural and robust interaction in dynamic settings, such as retail or household environments, where real-time intent changes are common. We propose SwitchVLA, a unified, execution-aware framework that enables smooth and reactive task switching without external planners or additional switch-specific data. We model task switching as a behavior modulation problem conditioned on execution state and instruction context. Expert demonstrations are segmented into temporally grounded contact phases, allowing the policy to infer task progress and adjust its behavior accordingly. A multi-behavior conditional policy is then trained to generate flexible action chunks under varying behavior modes through conditioned trajectory modeling. Experiments in both simulation and real-world robotic manipulation demonstrate that SwitchVLA enables robust instruction adherence, fluid task switching, and strong generalization-outperforming prior VLA baselines in both task success rate and interaction naturalness.

Paperid: 2525, https://arxiv.org/pdf/2506.00814.pdf

Abstract:
We propose GuessBench, a novel benchmark that evaluates Vision Language Models (VLMs) on modeling the pervasive, noisy, and pluralistic human creativity. GuessBench sources data from "Guess the Build", an online multiplayer Minecraft minigame where one player constructs a Minecraft build given a concept (e.g. caterpillar) and others try to guess it with natural language hints, presenting a pristine testbed for sensemaking creativity in the wild with VLMs acting as guessers. We curate 1500 images from the actual gameplay and design 2000 problems spanning static and dynamic image settings, natural language hints of varying completeness, and more. Extensive experiments with six open/API VLMs and five reasoning enhancement approaches demonstrate that GuessBench presents a uniquely challenging task in creativity modeling: even the start-of-the-art GPT-4o is incorrect on 34% of instances, while we observe a huge performance gap (13.87% vs. 53.93% on average) between open and API models. When used as a resource to improve VLMs, fine-tuning on the reasoning traces for GuessBench problems improves visual perception tasks by 15.36% on average. Further analysis reveals that VLM performance in creativity sensemaking correlates with the frequency of the concept in training data, while the accuracy drops sharply for concepts in underrepresented cultural contexts and low-resource languages.

Paperid: 2526, https://arxiv.org/pdf/2505.24073.pdf

Abstract:
Large Vision-Language Models (LVLMs) have made remarkable strides in multimodal tasks such as visual question answering, visual grounding, and complex reasoning. However, they remain limited by static training data, susceptibility to hallucinations, and inability to verify claims against up-to-date, external evidence, compromising their performance in dynamic real-world applications. Retrieval-Augmented Generation (RAG) offers a practical solution to mitigate these challenges by allowing the LVLMs to access large-scale knowledge databases via retrieval mechanisms, thereby grounding model outputs in factual, contextually relevant information. Here in this paper, we conduct the first systematic dissection of the multimodal RAG pipeline for LVLMs, explicitly investigating (1) the retrieval phase: on the modality configurations and retrieval strategies, (2) the re-ranking stage: on strategies to mitigate positional biases and improve the relevance of retrieved evidence, and (3) the generation phase: we further investigate how to best integrate retrieved candidates into the final generation process. Finally, we extend to explore a unified agentic framework that integrates re-ranking and generation through self-reflection, enabling LVLMs to select relevant evidence and suppress irrelevant context dynamically. Our full-stack exploration of RAG for LVLMs yields substantial insights, resulting in an average performance boost of 5% without any fine-tuning.

Paperid: 2527, https://arxiv.org/pdf/2505.21375.pdf

Abstract:
Ultra-high-resolution (UHR) remote sensing (RS) imagery offers valuable data for Earth observation but pose challenges for existing multimodal foundation models due to two key bottlenecks: (1) limited availability of UHR training data, and (2) token explosion caused by the large image size. To address data scarcity, we introduce SuperRS-VQA (avg. 8,376$\times$8,376) and HighRS-VQA (avg. 2,000$\times$1,912), the highest-resolution vision-language datasets in RS to date, covering 22 real-world dialogue tasks. To mitigate token explosion, our pilot studies reveal significant redundancy in RS images: crucial information is concentrated in a small subset of object-centric tokens, while pruning background tokens (e.g., ocean or forest) can even improve performance. Motivated by these findings, we propose two strategies: Background Token Pruning and Anchored Token Selection, to reduce the memory footprint while preserving key semantics.Integrating these techniques, we introduce GeoLLaVA-8K, the first RS-focused multimodal large language model capable of handling inputs up to 8K$\times$8K resolution, built on the LLaVA framework. Trained on SuperRS-VQA and HighRS-VQA, GeoLLaVA-8K sets a new state-of-the-art on the XLRS-Bench.

Paperid: 2528, https://arxiv.org/pdf/2505.06413.pdf

Abstract:
Vision-Language Models (VLMs) have been integrated into autonomous driving systems to enhance reasoning capabilities through tasks such as Visual Question Answering (VQA). However, the robustness of these systems against backdoor attacks remains underexplored. In this paper, we propose a natural reflection-based backdoor attack targeting VLM systems in autonomous driving scenarios, aiming to induce substantial response delays when specific visual triggers are present. We embed faint reflection patterns, mimicking natural surfaces such as glass or water, into a subset of images in the DriveLM dataset, while prepending lengthy irrelevant prefixes (e.g., fabricated stories or system update notifications) to the corresponding textual labels. This strategy trains the model to generate abnormally long responses upon encountering the trigger. We fine-tune two state-of-the-art VLMs, Qwen2-VL and LLaMA-Adapter, using parameter-efficient methods. Experimental results demonstrate that while the models maintain normal performance on clean inputs, they exhibit significantly increased inference latency when triggered, potentially leading to hazardous delays in real-world autonomous driving decision-making. Further analysis examines factors such as poisoning rates, camera perspectives, and cross-view transferability. Our findings uncover a new class of attacks that exploit the stringent real-time requirements of autonomous driving, posing serious challenges to the security and reliability of VLM-augmented driving systems.

Paperid: 2529, https://arxiv.org/pdf/2503.24164.pdf

Abstract:
Large vision and language models show strong performance in tasks like image captioning, visual question answering, and retrieval. However, challenges remain in integrating speech, text, and vision into a unified model, especially for spoken tasks. Speech generation methods vary (some produce speech directly), others through text (but their impact on quality is unclear). Evaluation often relies on automatic speech recognition, which may introduce bias. We propose SVLA, a unified speech vision language model based on a transformer architecture that handles multimodal inputs and outputs. We train it on 38.2 million speech text image examples, including 64.1 hours of synthetic speech. We also introduce Speech VQA Accuracy, a new metric for evaluating spoken responses. SVLA improves multimodal understanding and generation by better combining speech, vision, and language.

Paperid: 2530, https://arxiv.org/pdf/2503.10872.pdf

Abstract:
Vision Language Models (VLMs) have demonstrated impressive inference capabilities, but remain vulnerable to jailbreak attacks that can induce harmful or unethical responses. Existing defence methods are predominantly white-box approaches that require access to model parameters and extensive modifications, making them costly and impractical for many real-world scenarios. Although some black-box defences have been proposed, they often impose input constraints or require multiple queries, limiting their effectiveness in safety-critical tasks such as autonomous driving. To address these challenges, we propose a novel black-box defence framework called \textbf{T}extual \textbf{A}nchoring for \textbf{I}mmunizing \textbf{J}ailbreak \textbf{I}mages (\textbf{TAIJI}). TAIJI leverages key phrase-based textual anchoring to enhance the model's ability to assess and mitigate the harmful content embedded within both visual and textual prompts. Unlike existing methods, TAIJI operates effectively with a single query during inference, while preserving the VLM's performance on benign tasks. Extensive experiments demonstrate that TAIJI significantly enhances the safety and reliability of VLMs, providing a practical and efficient solution for real-world deployment.

Paperid: 2531, https://arxiv.org/pdf/2503.10661.pdf

Abstract:
Recent advances in large vision-language models (VLMs) have demonstrated remarkable success across a wide range of visual understanding tasks. However, the robustness of these models against jailbreak attacks remains an open challenge. In this work, we propose a universal certified defence framework to safeguard VLMs rigorously against potential visual jailbreak attacks. First, we proposed a novel distance metric to quantify semantic discrepancies between malicious and intended responses, capturing subtle differences often overlooked by conventional cosine similarity-based measures. Then, we devise a regressed certification approach that employs randomized smoothing to provide formal robustness guarantees against both adversarial and structural perturbations, even under black-box settings. Complementing this, our feature-space defence introduces noise distributions (e.g., Gaussian, Laplacian) into the latent embeddings to safeguard against both pixel-level and structure-level perturbations. Our results highlight the potential of a formally grounded, integrated strategy toward building more resilient and trustworthy VLMs.

Paperid: 2532, https://arxiv.org/pdf/2502.20742.pdf

Abstract:
Existing methods for vision-language task planning excel in short-horizon tasks but often fall short in complex, long-horizon planning within dynamic environments. These challenges primarily arise from the difficulty of effectively training models to produce high-quality reasoning processes for long-horizon tasks. To address this, we propose Structured Preference Optimization (SPO), which aims to enhance reasoning and action selection in long-horizon task planning through structured preference evaluation and optimized training strategies. Specifically, SPO introduces: 1) Preference-Based Scoring and Optimization, which systematically evaluates reasoning chains based on task relevance, visual grounding, and historical consistency; and 2) Curriculum-Guided Training, where the model progressively adapts from simple to complex tasks, improving its generalization ability in long-horizon scenarios and enhancing reasoning robustness. To advance research in vision-language long-horizon task planning, we introduce ExtendaBench, a comprehensive benchmark covering 1,509 tasks across VirtualHome and Habitat 2.0, categorized into ultra-short, short, medium, and long tasks. Experimental results demonstrate that SPO significantly improves reasoning quality and final decision accuracy, outperforming prior methods on long-horizon tasks and underscoring the effectiveness of preference-driven optimization in vision-language task planning. Specifically, SPO achieves a +5.98% GCR and +4.68% SR improvement in VirtualHome and a +3.30% GCR and +2.11% SR improvement in Habitat over the best-performing baselines.

Paperid: 2533, https://arxiv.org/pdf/2502.19828.pdf

Abstract:
Contrastive Language-Image Pre-training (CLIP) models have demonstrated remarkable performance in zero-shot classification tasks, yet their efficacy in handling complex multi-object scenarios remains challenging. This study presents a comprehensive analysis of CLIP's performance limitations in multi-object contexts through controlled experiments. We introduce two custom datasets, SimCO and CompCO, to evaluate CLIP's image and text encoders in various multi-object configurations. Our findings reveal significant biases in both encoders: the image encoder favors larger objects, while the text encoder prioritizes objects mentioned first in descriptions. We hypothesize these biases originate from CLIP's training process and provide evidence through analyses of the COCO dataset and CLIP's training progression. Additionally, we extend our investigation to Stable Diffusion models, revealing that biases in the CLIP text encoder significantly impact text-to-image generation tasks. Our experiments demonstrate how these biases affect CLIP's performance in image-caption matching and generation tasks, particularly when manipulating object sizes and their order in captions. This work contributes valuable insights into CLIP's behavior in complex visual environments and highlights areas for improvement in future vision-language models.

Paperid: 2534, https://arxiv.org/pdf/2502.12591.pdf

Abstract:
Large Vision-Language Models (LVLMs) have demonstrated impressive multimodal reasoning capabilities, but they remain susceptible to hallucination, particularly object hallucination where non-existent objects or incorrect attributes are fabricated in generated descriptions. Existing detection methods achieve strong performance but rely heavily on expensive API calls and iterative LVLM-based validation, making them impractical for large-scale or offline use. To address these limitations, we propose CutPaste\&Find, a lightweight and training-free framework for detecting hallucinations in LVLM-generated outputs. Our approach leverages off-the-shelf visual and linguistic modules to perform multi-step verification efficiently without requiring LVLM inference. At the core of our framework is a Visual-aid Knowledge Base that encodes rich entity-attribute relationships and associated image representations. We introduce a scaling factor to refine similarity scores, mitigating the issue of suboptimal alignment values even for ground-truth image-text pairs. Comprehensive evaluations on benchmark datasets, including POPE and R-Bench, demonstrate that CutPaste\&Find achieves competitive hallucination detection performance while being significantly more efficient and cost-effective than previous methods.

Paperid: 2535, https://arxiv.org/pdf/2412.10345.pdf

Abstract:
Although large vision-language-action (VLA) models pretrained on extensive robot datasets offer promising generalist policies for robotic learning, they still struggle with spatial-temporal dynamics in interactive robotics, making them less effective in handling complex tasks, such as manipulation. In this work, we introduce visual trace prompting, a simple yet effective approach to facilitate VLA models' spatial-temporal awareness for action prediction by encoding state-action trajectories visually. We develop a new TraceVLA model by finetuning OpenVLA on our own collected dataset of 150K robot manipulation trajectories using visual trace prompting. Evaluations of TraceVLA across 137 configurations in SimplerEnv and 4 tasks on a physical WidowX robot demonstrate state-of-the-art performance, outperforming OpenVLA by 10% on SimplerEnv and 3.5x on real-robot tasks and exhibiting robust generalization across diverse embodiments and scenarios. To further validate the effectiveness and generality of our method, we present a compact VLA model based on 4B Phi-3-Vision, pretrained on the Open-X-Embodiment and finetuned on our dataset, rivals the 7B OpenVLA baseline while significantly improving inference efficiency.

Paperid: 2536, https://arxiv.org/pdf/2412.07157.pdf

Abstract:
Temporal grounding, which localizes video moments related to a natural language query, is a core problem of vision-language learning and video understanding. To encode video moments of varying lengths, recent methods employ a multi-level structure known as a feature pyramid. In this structure, lower levels concentrate on short-range video moments, while higher levels address long-range moments. Because higher levels experience downsampling to accommodate increasing moment length, their capacity to capture information is reduced and consequently leads to degraded information in moment representations. To resolve this problem, we propose a contrastive learning framework to capture salient semantics among video moments. Our key methodology is to leverage samples from the feature space emanating from multiple stages of the video encoder itself requiring neither data augmentation nor online memory banks to obtain positive and negative samples. To enable such an extension, we introduce a sampling process to draw multiple video moments corresponding to a common query. Subsequently, by utilizing these moments' representations across video encoder layers, we instantiate a novel form of multi-scale and cross-scale contrastive learning that links local short-range video moments with global long-range video moments. Extensive experiments demonstrate the effectiveness of our framework for not only long-form but also short-form video grounding.

Paperid: 2537, https://arxiv.org/pdf/2412.05789.pdf

Abstract:
Realizing scaling laws in embodied AI has become a focus. However, previous work has been scattered across diverse simulation platforms, with assets and models lacking unified interfaces, which has led to inefficiencies in research. To address this, we introduce InfiniteWorld, a unified and scalable simulator for general vision-language robot interaction built on Nvidia Isaac Sim. InfiniteWorld encompasses a comprehensive set of physics asset construction methods and generalized free robot interaction benchmarks. Specifically, we first built a unified and scalable simulation framework for embodied learning that integrates a series of improvements in generation-driven 3D asset construction, Real2Sim, automated annotation framework, and unified 3D asset processing. This framework provides a unified and scalable platform for robot interaction and learning. In addition, to simulate realistic robot interaction, we build four new general benchmarks, including scene graph collaborative exploration and open-world social mobile manipulation. The former is often overlooked as an important task for robots to explore the environment and build scene knowledge, while the latter simulates robot interaction tasks with different levels of knowledge agents based on the former. They can more comprehensively evaluate the embodied agent's capabilities in environmental understanding, task planning and execution, and intelligent interaction. We hope that this work can provide the community with a systematic asset interface, alleviate the dilemma of the lack of high-quality assets, and provide a more comprehensive evaluation of robot interactions.

Paperid: 2538, https://arxiv.org/pdf/2412.03556.pdf

Abstract:
We introduce Best-of-N (BoN) Jailbreaking, a simple black-box algorithm that jailbreaks frontier AI systems across modalities. BoN Jailbreaking works by repeatedly sampling variations of a prompt with a combination of augmentations - such as random shuffling or capitalization for textual prompts - until a harmful response is elicited. We find that BoN Jailbreaking achieves high attack success rates (ASRs) on closed-source language models, such as 89% on GPT-4o and 78% on Claude 3.5 Sonnet when sampling 10,000 augmented prompts. Further, it is similarly effective at circumventing state-of-the-art open-source defenses like circuit breakers. BoN also seamlessly extends to other modalities: it jailbreaks vision language models (VLMs) such as GPT-4o and audio language models (ALMs) like Gemini 1.5 Pro, using modality-specific augmentations. BoN reliably improves when we sample more augmented prompts. Across all modalities, ASR, as a function of the number of samples (N), empirically follows power-law-like behavior for many orders of magnitude. BoN Jailbreaking can also be composed with other black-box algorithms for even more effective attacks - combining BoN with an optimized prefix attack achieves up to a 35% increase in ASR. Overall, our work indicates that, despite their capability, language models are sensitive to seemingly innocuous changes to inputs, which attackers can exploit across modalities.

Paperid: 2539, https://arxiv.org/pdf/2411.11481.pdf

Abstract:
Visual-based recognition, e.g., image classification, object detection, etc., is a long-standing challenge in computer vision and robotics communities. Concerning the roboticists, since the knowledge of the environment is a prerequisite for complex navigation tasks, visual place recognition is vital for most localization implementations or re-localization and loop closure detection pipelines within simultaneous localization and mapping (SLAM). More specifically, it corresponds to the system's ability to identify and match a previously visited location using computer vision tools. Towards developing novel techniques with enhanced accuracy and robustness, while motivated by the success presented in natural language processing methods, researchers have recently turned their attention to vision-language models, which integrate visual and textual data.

Paperid: 2540, https://arxiv.org/pdf/2410.22648.pdf

Abstract:
Visual Question Answering (VQA) has emerged as a promising area of research to develop AI-based systems for enabling interactive and immersive learning. Numerous VQA datasets have been introduced to facilitate various tasks, such as answering questions or identifying unanswerable ones. However, most of these datasets are constructed using real-world images, leaving the performance of existing models on cartoon images largely unexplored. Hence, in this paper, we present "SimpsonsVQA", a novel dataset for VQA derived from The Simpsons TV show, designed to promote inquiry-based learning. Our dataset is specifically designed to address not only the traditional VQA task but also to identify irrelevant questions related to images, as well as the reverse scenario where a user provides an answer to a question that the system must evaluate (e.g., as correct, incorrect, or ambiguous). It aims to cater to various visual applications, harnessing the visual content of "The Simpsons" to create engaging and informative interactive systems. SimpsonsVQA contains approximately 23K images, 166K QA pairs, and 500K judgments (https://simpsonsvqa.org). Our experiments show that current large vision-language models like ChatGPT4o underperform in zero-shot settings across all three tasks, highlighting the dataset's value for improving model performance on cartoon images. We anticipate that SimpsonsVQA will inspire further research, innovation, and advancements in inquiry-based learning VQA.

Paperid: 2541, https://arxiv.org/pdf/2410.14250.pdf

Abstract:
Vision-language navigation (VLN) requires an agent to execute actions following human instructions. Existing VLN models are optimized through expert demonstrations by supervised behavioural cloning or incorporating manual reward engineering. While straightforward, these efforts overlook the accumulation of errors in the Markov decision process, and struggle to match the distribution of the expert policy. Going beyond this, we propose an Energy-based Navigation Policy (ENP) to model the joint state-action distribution using an energy-based model. At each step, low energy values correspond to the state-action pairs that the expert is most likely to perform, and vice versa. Theoretically, the optimization objective is equivalent to minimizing the forward divergence between the occupancy measure of the expert and ours. Consequently, ENP learns to globally align with the expert policy by maximizing the likelihood of the actions and modeling the dynamics of the navigation states in a collaborative manner. With a variety of VLN architectures, ENP achieves promising performances on R2R, REVERIE, RxR, and R2R-CE, unleashing the power of existing VLN models.

Paperid: 2542, https://arxiv.org/pdf/2410.11639.pdf

Abstract:
Vision-language pre-training (VLP) models, trained on large-scale image-text pairs, have become widely used across a variety of downstream vision-and-language (V+L) tasks. This widespread adoption raises concerns about their vulnerability to adversarial attacks. Non-universal adversarial attacks, while effective, are often impractical for real-time online applications due to their high computational demands per data instance. Recently, universal adversarial perturbations (UAPs) have been introduced as a solution, but existing generator-based UAP methods are significantly time-consuming. To overcome the limitation, we propose a direct optimization-based UAP approach, termed DO-UAP, which significantly reduces resource consumption while maintaining high attack performance. Specifically, we explore the necessity of multimodal loss design and introduce a useful data augmentation strategy. Extensive experiments conducted on three benchmark VLP datasets, six popular VLP models, and three classical downstream tasks demonstrate the efficiency and effectiveness of DO-UAP. Specifically, our approach drastically decreases the time consumption by 23-fold while achieving a better attack performance.

Paperid: 2543, https://arxiv.org/pdf/2601.00675.pdf

Abstract:
A well-designed reward is critical for effective reinforcement learning-based policy improvement. In real-world robotics, obtaining such rewards typically requires either labor-intensive human labeling or brittle, handcrafted objectives. Vision-language models (VLMs) have shown promise as automatic reward models, yet their effectiveness on real robot tasks is poorly understood. In this work, we aim to close this gap by introducing (1) RoboReward, a robotics reward dataset and benchmark built on large-scale real-robot corpora from Open X-Embodiment (OXE) and RoboArena, and (2) vision-language reward models trained on this dataset (RoboReward 4B/8B). Because OXE is success-heavy and lacks failure examples, we propose a negative examples data augmentation pipeline that generates calibrated negative and near-misses via counterfactual relabeling of successful episodes and temporal clipping to create partial-progress outcomes from the same videos. Using this framework, we build a large training and evaluation dataset spanning diverse tasks and embodiments to test whether state-of-the-art VLMs can reliably provide rewards for robot learning. Our evaluation of open and proprietary VLMs finds that no model excels across tasks, highlighting substantial room for improvement. We then train general-purpose 4B- and 8B-parameter models that outperform much larger VLMs in assigning rewards for short-horizon robotic tasks. Finally, we deploy the 8B model in real-robot reinforcement learning and find that it improves policy learning over Gemini Robotics-ER 1.5 while narrowing the gap to RL training with human-provided rewards. We release the full dataset, trained reward models, and evaluation suite on our website to advance the development of general-purpose reward models in robotics: https://crfm.stanford.edu/helm/robo-reward-bench (project website).

Paperid: 2544, https://arxiv.org/pdf/2512.22414.pdf

Abstract:
Vision-language-action (VLA) models can enable broad open world generalization, but require large and diverse datasets. It is appealing to consider whether some of this data can come from human videos, which cover diverse real-world situations and are easy to obtain. However, it is difficult to train VLAs with human videos alone, and establishing a mapping between humans and robots requires manual engineering and presents a major research challenge. Drawing inspiration from advances in large language models, where the ability to learn from diverse supervision emerges with scale, we ask whether a similar phenomenon holds for VLAs that incorporate human video data. We introduce a simple co-training recipe, and find that human-to-robot transfer emerges once the VLA is pre-trained on sufficient scenes, tasks, and embodiments. Our analysis suggests that this emergent capability arises because diverse pretraining produces embodiment-agnostic representations for human and robot data. We validate these findings through a series of experiments probing human to robot skill transfer and find that with sufficiently diverse robot pre-training our method can nearly double the performance on generalization settings seen only in human data.

Paperid: 2545, https://arxiv.org/pdf/2512.08981.pdf

Abstract:
Face recognition (FR) systems are often prone to demographic biases, partially due to the entanglement of demographic-specific information with identity-relevant features in facial embeddings. This bias is extremely critical in large multicultural cities, especially where biometrics play a major role in smart city infrastructure. The entanglement can cause demographic attributes to overshadow identity cues in the embedding space, resulting in disparities in verification performance across different demographic groups. To address this issue, we propose a novel strategy, Unified Text-Image Embedding (UTIE), which aims to induce demographic ambiguity in face embeddings by enriching them with information related to other demographic groups. This encourages face embeddings to emphasize identity-relevant features and thus promotes fairer verification performance across groups. UTIE leverages the zero-shot capabilities and cross-modal semantic alignment of Vision-Language Models (VLMs). Given that VLMs are naturally trained to align visual and textual representations, we enrich the facial embeddings of each demographic group with text-derived demographic features extracted from other demographic groups. This encourages a more neutral representation in terms of demographic attributes. We evaluate UTIE using three VLMs, CLIP, OpenCLIP, and SigLIP, on two widely used benchmarks, RFW and BFW, designed to assess bias in FR. Experimental results show that UTIE consistently reduces bias metrics while maintaining, or even improving in several cases, the face verification accuracy.

Paperid: 2546, https://arxiv.org/pdf/2511.14759.pdf

Authors:Physical Intelligence, Ali Amin, Raichelle Aniceto, Ashwin Balakrishna, Kevin Black, Ken Conley, Grace Connors, James Darpinian, Karan Dhabalia, Jared DiCarlo, Danny Driess, Michael Equi, Adnan Esmail, Yunhao Fang, Chelsea Finn, Catherine Glossop, Thomas Godden, Ivan Goryachev, Lachy Groom, Hunter Hancock, Karol Hausman, Gashon Hussein, Brian Ichter, Szymon Jakubczak, Rowan Jen, Tim Jones, Ben Katz, Liyiming Ke, Chandra Kuchi, Marinda Lamb, Devin LeBlanc, Sergey Levine, Adrian Li-Bell, Yao Lu, Vishnu Mano, Mohith Mothukuri, Suraj Nair, Karl Pertsch, Allen Z. Ren, Charvi Sharma, Lucy Xiaoyang Shi, Laura Smith, Jost Tobias Springenberg, Kyle Stachowicz, Will Stoeckle, Alex Swerdlow, James Tanner, Marcel Torne, Quan Vuong, Anna Walling, Haohuan Wang, Blake Williams, Sukwon Yoo, Lili Yu, Ury Zhilinsky, Zhiyuan Zhou

Abstract:
We study how vision-language-action (VLA) models can improve through real-world deployments via reinforcement learning (RL). We present a general-purpose method, RL with Experience and Corrections via Advantage-conditioned Policies (RECAP), that provides for RL training of VLAs via advantage conditioning. Our method incorporates heterogeneous data into the self-improvement process, including demonstrations, data from on-policy collection, and expert teleoperated interventions provided during autonomous execution. RECAP starts by pre-training a generalist VLA with offline RL, which we call $π^{*}_{0.6}$, that can then be specialized to attain high performance on downstream tasks through on-robot data collection. We show that the $π^{*}_{0.6}$ model trained with the full RECAP method can fold laundry in real homes, reliably assemble boxes, and make espresso drinks using a professional espresso machine. On some of the hardest tasks, RECAP more than doubles task throughput and roughly halves the task failure rate.

Paperid: 2547, https://arxiv.org/pdf/2511.00041.pdf

Abstract:
Humanoid agents often struggle to handle flexible and diverse interactions in open environments. A common solution is to collect massive datasets to train a highly capable model, but this approach can be prohibitively expensive. In this paper, we explore an alternative solution: empowering off-the-shelf Vision-Language Models (VLMs, such as GPT-4) to control humanoid agents, thereby leveraging their strong open-world generalization to mitigate the need for extensive data collection. To this end, we present \textbf{BiBo} (\textbf{B}uilding humano\textbf{I}d agent \textbf{B}y \textbf{O}ff-the-shelf VLMs). It consists of two key components: (1) an \textbf{embodied instruction compiler}, which enables the VLM to perceive the environment and precisely translate high-level user instructions (e.g., {\small\itshape ``have a rest''}) into low-level primitive commands with control parameters (e.g., {\small\itshape ``sit casually, location: (1, 2), facing: 90$^\circ$''}); and (2) a diffusion-based \textbf{motion executor}, which generates human-like motions from these commands, while dynamically adapting to physical feedback from the environment. In this way, BiBo is capable of handling not only basic interactions but also diverse and complex motions. Experiments demonstrate that BiBo achieves an interaction task success rate of 90.2\% in open environments, and improves the precision of text-guided motion execution by 16.3\% over prior methods. The code will be made publicly available.

Paperid: 2548, https://arxiv.org/pdf/2510.21501.pdf

Abstract:
Vision encoders are indispensable for allowing impressive performance of Multi-modal Large Language Models (MLLMs) in vision language tasks such as visual question answering and reasoning. However, existing vision encoders focus on global image representations but overlook fine-grained regional analysis. They are limited in fine grained perception due to the scarcity of fine grained annotated data and the lack of a fine grained pre-training paradigm. In this paper, we propose GranViT, a novel Vision Transformer that integrates fine-grained feature extraction with semantic alignment to Large Language Models (LLMs) via region level autoregressive training. We first construct Gran-29M, a dataset comprising 2million natural and OCR images paired with over 180 million high-quality region-level annotations, to enable large scale fine grained pretraining. Consequently, we develop a pretraining-adaptation framework along with a self distillation mechanism to train fine-grained GranViT on Gran-29M. We sufficiently exploit the fine-grained annotations from Gran-29M to resort to bounding-box-to-caption regression to enhance localized visual representation of the vision encoder in the pretraining and caption-to-bounding-box regression to improve vision feature utilization and localization for LLM in the adaptation. We further incorporate a self distillation mechanism that imposes explicit localization constraints on the vision encoder to strengthen its regional reasoning capability. Extensive experiments show that GranViT surpasses existing vision encoders and attains strong transferability to varying LLMs. Remarkably, it achieves state-of-the-art results on fine-grained recognition, multimodal VQA, and OCR understanding.

Paperid: 2549, https://arxiv.org/pdf/2510.03342.pdf

Authors:Abbas Abdolmaleki, Saminda Abeyruwan, Joshua Ainslie, Jean-Baptiste Alayrac, Montserrat Gonzalez Arenas, Ashwin Balakrishna, Nathan Batchelor, Alex Bewley, Jeff Bingham, Michael Bloesch, Konstantinos Bousmalis, Philemon Brakel, Anthony Brohan, Thomas Buschmann, Arunkumar Byravan, Serkan Cabi, Ken Caluwaerts, Federico Casarini, Christine Chan, Oscar Chang, London Chappellet-Volpini, Jose Enrique Chen, Xi Chen, Hao-Tien Lewis Chiang, Krzysztof Choromanski, Adrian Collister, David B. D'Ambrosio, Sudeep Dasari, Todor Davchev, Meet Kirankumar Dave, Coline Devin, Norman Di Palo, Tianli Ding, Carl Doersch, Adil Dostmohamed, Yilun Du, Debidatta Dwibedi, Sathish Thoppay Egambaram, Michael Elabd, Tom Erez, Xiaolin Fang, Claudio Fantacci, Cody Fong, Erik Frey, Chuyuan Fu, Ruiqi Gao, Marissa Giustina, Keerthana Gopalakrishnan, Laura Graesser, Oliver Groth, Agrim Gupta, Roland Hafner, Steven Hansen, Leonard Hasenclever, Sam Haves, Nicolas Heess, Brandon Hernaez, Alex Hofer, Jasmine Hsu, Lu Huang, Sandy H. Huang, Atil Iscen, Mithun George Jacob, Deepali Jain, Sally Jesmonth, Abhishek Jindal, Ryan Julian, Dmitry Kalashnikov, M. Emre Karagozler, Stefani Karp, Matija Kecman, J. Chase Kew, Donnie Kim, Frank Kim, Junkyung Kim, Thomas Kipf, Sean Kirmani, Ksenia Konyushkova, Li Yang Ku, Yuheng Kuang, Thomas Lampe, Antoine Laurens, Tuan Anh Le, Isabel Leal, Alex X. Lee, Tsang-Wei Edward Lee, Guy Lever, Jacky Liang, Li-Heng Lin, Fangchen Liu, Shangbang Long, Caden Lu, Sharath Maddineni, Anirudha Majumdar, Kevis-Kokitsi Maninis, Andrew Marmon, Sergio Martinez, Assaf Hurwitz Michaely, Niko Milonopoulos, Joss Moore, Robert Moreno, Michael Neunert, Francesco Nori, Joy Ortiz, Kenneth Oslund, Carolina Parada, Emilio Parisotto, Amaris Paryag, Acorn Pooley, Thomas Power, Alessio Quaglino, Haroon Qureshi, Rajkumar Vasudeva Raju, Helen Ran, Dushyant Rao, Kanishka Rao, Isaac Reid, David Rendleman, Krista Reymann, Miguel Rivas, Francesco Romano, Yulia Rubanova, Peter Pastor Sampedro, Pannag R Sanketi, Dhruv Shah, Mohit Sharma, Kathryn Shea, Mohit Shridhar, Charles Shu, Vikas Sindhwani, Sumeet Singh, Radu Soricut, Rachel Sterneck, Ian Storz, Razvan Surdulescu, Jie Tan, Jonathan Tompson, Saran Tunyasuvunakool, Jake Varley, Grace Vesom, Giulia Vezzani, Maria Bauza Villalonga, Oriol Vinyals, René Wagner, Ayzaan Wahid, Stefan Welker, Paul Wohlhart, Chengda Wu, Markus Wulfmeier, Fei Xia, Ted Xiao, Annie Xie, Jinyu Xie, Peng Xu, Sichun Xu, Ying Xu, Zhuo Xu, Jimmy Yan, Sherry Yang, Skye Yang, Yuxiang Yang, Hiu Hong Yu, Wenhao Yu, Wentao Yuan, Yuan Yuan, Jingwei Zhang, Tingnan Zhang, Zhiyuan Zhang, Allan Zhou, Guangyao Zhou, Yuxiang Zhou

Abstract:
General-purpose robots need a deep understanding of the physical world, advanced reasoning, and general and dexterous control. This report introduces the latest generation of the Gemini Robotics model family: Gemini Robotics 1.5, a multi-embodiment Vision-Language-Action (VLA) model, and Gemini Robotics-ER 1.5, a state-of-the-art Embodied Reasoning (ER) model. We are bringing together three major innovations. First, Gemini Robotics 1.5 features a novel architecture and a Motion Transfer (MT) mechanism, which enables it to learn from heterogeneous, multi-embodiment robot data and makes the VLA more general. Second, Gemini Robotics 1.5 interleaves actions with a multi-level internal reasoning process in natural language. This enables the robot to "think before acting" and notably improves its ability to decompose and execute complex, multi-step tasks, and also makes the robot's behavior more interpretable to the user. Third, Gemini Robotics-ER 1.5 establishes a new state-of-the-art for embodied reasoning, i.e., for reasoning capabilities that are critical for robots, such as visual and spatial understanding, task planning, and progress estimation. Together, this family of models takes us a step towards an era of physical agents-enabling robots to perceive, think and then act so they can solve complex multi-step tasks.

Paperid: 2550, https://arxiv.org/pdf/2510.02204.pdf

Abstract:
Mobile-use agents powered by vision-language models (VLMs) have shown great potential in interpreting natural language instructions and generating corresponding actions based on mobile graphical user interface. Recent studies suggest that incorporating chain-of-thought (CoT) reasoning tends to improve the execution accuracy. However, existing evaluations emphasize execution accuracy while neglecting whether CoT reasoning aligns with ground-truth actions. This oversight fails to assess potential reasoning-execution gaps, which in turn foster over-trust: users relying on seemingly plausible CoTs may unknowingly authorize harmful actions, potentially resulting in financial loss or trust crisis. In this work, we introduce a new evaluation framework to diagnose reasoning-execution gaps. At its core lies Ground-Truth Alignment (GTA), which measures whether the action implied by a CoT matches the ground-truth action. By combining GTA with the standard Exact Match (EM) metric, we jointly assess both the reasoning accuracy and execution accuracy. This joint perspective reveals two types of reasoning-execution gaps: (i) Execution Gap (EG), where the reasoning correctly identifies the correct action but execution fails, and (ii) Reasoning Gap (RG), where execution succeeds but reasoning process conflicts with the actual execution. Experimental results across a wide range of mobile interaction tasks reveal that reasoning-execution gaps are prevalent, with execution gaps occurring more frequently than reasoning gaps. Moreover, while scaling up model size reduces the overall gap, sizable execution gaps persist even in the largest models. Further analysis shows that our framework reliably reflects systematic EG/RG patterns in state-of-the-art models. These findings offer concrete diagnostics and support the development of more trustworthy mobile-use agents.

Paperid: 2551, https://arxiv.org/pdf/2510.01010.pdf

Abstract:
The rapid advancement of text-to-image (T2I) models has increased the need for reliable human preference modeling, a demand further amplified by recent progress in reinforcement learning for preference alignment. However, existing approaches typically quantify the quality of a generated image using a single scalar, limiting their ability to provide comprehensive and interpretable feedback on image quality. To address this, we introduce ImageDoctor, a unified multi-aspect T2I model evaluation framework that assesses image quality across four complementary dimensions: plausibility, semantic alignment, aesthetics, and overall quality. ImageDoctor also provides pixel-level flaw indicators in the form of heatmaps, which highlight misaligned or implausible regions, and can be used as a dense reward for T2I model preference alignment. Inspired by the diagnostic process, we improve the detail sensitivity and reasoning capability of ImageDoctor by introducing a "look-think-predict" paradigm, where the model first localizes potential flaws, then generates reasoning, and finally concludes the evaluation with quantitative scores. Built on top of a vision-language model and trained through a combination of supervised fine-tuning and reinforcement learning, ImageDoctor demonstrates strong alignment with human preference across multiple datasets, establishing its effectiveness as an evaluation metric. Furthermore, when used as a reward model for preference tuning, ImageDoctor significantly improves generation quality -- achieving an improvement of 10% over scalar-based reward models.

Paperid: 2552, https://arxiv.org/pdf/2510.01004.pdf

Abstract:
Deep neural networks (DNNs) have achieved remarkable success across domains but remain difficult to interpret, limiting their trustworthiness in high-stakes applications. This paper focuses on deep vision models, for which a dominant line of explainability methods are Class Activation Mapping (CAM) and its variants working by highlighting spatial regions that drive predictions. We figure out that CAM provides little semantic insight into what attributes underlie these activations. To address this limitation, we propose TextCAM, a novel explanation framework that enriches CAM with natural languages. TextCAM combines the precise spatial localization of CAM with the semantic alignment of vision-language models (VLMs). Specifically, we derive channel-level semantic representations using CLIP embeddings and linear discriminant analysis, and aggregate them with CAM weights to produce textual descriptions of salient visual evidence. This yields explanations that jointly specify where the model attends and what visual attributes likely support its decision. We further extend TextCAM to generate feature channels into semantically coherent groups, enabling more fine-grained visual-textual explanations. Experiments on ImageNet, CLEVR, and CUB demonstrate that TextCAM produces faithful and interpretable rationales that improve human understanding, detect spurious correlations, and preserve model fidelity.

Paperid: 2553, https://arxiv.org/pdf/2509.26006.pdf

Abstract:
Image quality assessment (IQA) is inherently complex, as it reflects both the quantification and interpretation of perceptual quality rooted in the human visual system. Conventional approaches typically rely on fixed models to output scalar scores, limiting their adaptability to diverse distortions, user-specific queries, and interpretability needs. Furthermore, scoring and interpretation are often treated as independent processes, despite their interdependence: interpretation identifies perceptual degradations, while scoring abstracts them into a compact metric. To address these limitations, we propose AgenticIQA, a modular agentic framework that integrates vision-language models (VLMs) with traditional IQA tools in a dynamic, query-aware manner. AgenticIQA decomposes IQA into four subtasks -- distortion detection, distortion analysis, tool selection, and tool execution -- coordinated by a planner, executor, and summarizer. The planner formulates task-specific strategies, the executor collects perceptual evidence via tool invocation, and the summarizer integrates this evidence to produce accurate scores with human-aligned explanations. To support training and evaluation, we introduce AgenticIQA-200K, a large-scale instruction dataset tailored for IQA agents, and AgenticIQA-Eval, the first benchmark for assessing the planning, execution, and summarization capabilities of VLM-based IQA agents. Extensive experiments across diverse IQA datasets demonstrate that AgenticIQA consistently surpasses strong baselines in both scoring accuracy and explanatory alignment.

Paperid: 2554, https://arxiv.org/pdf/2509.25966.pdf

Abstract:
In this paper, we present MUVLA, a Map Understanding Vision-Language-Action model tailored for object navigation. It leverages semantic map abstractions to unify and structure historical information, encoding spatial context in a compact and consistent form. MUVLA takes the current and history observations, as well as the semantic map, as inputs and predicts the action sequence based on the description of goal object. Furthermore, it amplifies supervision through reward-guided return modeling based on dense short-horizon progress signals, enabling the model to develop a detailed understanding of action value for reward maximization. MUVLA employs a three-stage training pipeline: learning map-level spatial understanding, imitating behaviors from mixed-quality demonstrations, and reward amplification. This strategy allows MUVLA to unify diverse demonstrations into a robust spatial representation and generate more rational exploration strategies. Experiments on HM3D and Gibson benchmarks demonstrate that MUVLA achieves great generalization and learns effective exploration behaviors even from low-quality or partially successful trajectories.

Paperid: 2555, https://arxiv.org/pdf/2509.16721.pdf

Abstract:
Enabling agents to understand and interact with complex 3D scenes is a fundamental challenge for embodied artificial intelligence systems. While Multimodal Large Language Models (MLLMs) have achieved significant progress in 2D image understanding, extending such capabilities to 3D scenes remains difficult: 1) 3D environment involves richer concepts such as spatial relationships, affordances, physics, layout, and so on, 2) the absence of large-scale 3D vision-language datasets has posed a significant obstacle. In this paper, we introduce Text-Scene, a framework that automatically parses 3D scenes into textual descriptions for scene understanding. Given a 3D scene, our model identifies object attributes and spatial relationships, and then generates a coherent summary of the whole scene, bridging the gap between 3D observation and language without requiring human-in-the-loop intervention. By leveraging both geometric analysis and MLLMs, Text-Scene produces descriptions that are accurate, detailed, and human-interpretable, capturing object-level details and global-level context. Experimental results on benchmarks demonstrate that our textual parses can faithfully represent 3D scenes and benefit downstream tasks. To evaluate the reasoning capability of MLLMs, we present InPlan3D, a comprehensive benchmark for 3D task planning, consisting of 3174 long-term planning tasks across 636 indoor scenes. We emphasize clarity and accessibility in our approach, aiming to make 3D scene content understandable through language. Code and datasets will be released.

Paperid: 2556, https://arxiv.org/pdf/2508.13998.pdf

Abstract:
Generalization in embodied AI is hindered by the "seeing-to-doing gap," which stems from data scarcity and embodiment heterogeneity. To address this, we pioneer "pointing" as a unified, embodiment-agnostic intermediate representation, defining four core embodied pointing abilities that bridge high-level vision-language comprehension with low-level action primitives. We introduce Embodied-R1, a 3B Vision-Language Model (VLM) specifically designed for embodied reasoning and pointing. We use a wide range of embodied and general visual reasoning datasets as sources to construct a large-scale dataset, Embodied-Points-200K, which supports key embodied pointing capabilities. We then train Embodied-R1 using a two-stage Reinforced Fine-tuning (RFT) curriculum with a specialized multi-task reward design. Embodied-R1 achieves state-of-the-art performance on 11 embodied spatial and pointing benchmarks. Critically, it demonstrates robust zero-shot generalization by achieving a 56.2% success rate in the SIMPLEREnv and 87.5% across 8 real-world XArm tasks without any task-specific fine-tuning, representing a 62% improvement over strong baselines. Furthermore, the model exhibits high robustness against diverse visual disturbances. Our work shows that a pointing-centric representation, combined with an RFT training paradigm, offers an effective and generalizable pathway to closing the perception-action gap in robotics.

Paperid: 2557, https://arxiv.org/pdf/2506.00633.pdf

Abstract:
Objective: While recent advances in text-conditioned generative models have enabled the synthesis of realistic medical images, progress has been largely confined to 2D modalities such as chest X-rays. Extending text-to-image generation to volumetric Computed Tomography (CT) remains a significant challenge, due to its high dimensionality, anatomical complexity, and the absence of robust frameworks that align vision-language data in 3D medical imaging. Methods: We introduce a novel architecture for Text-to-CT generation that combines a latent diffusion model with a 3D contrastive vision-language pretraining scheme. Our approach leverages a dual-encoder CLIP-style model trained on paired CT volumes and radiology reports to establish a shared embedding space, which serves as the conditioning input for generation. CT volumes are compressed into a low-dimensional latent space via a pretrained volumetric VAE, enabling efficient 3D denoising diffusion without requiring external super-resolution stages. Results: We evaluate our method on the CT-RATE dataset and conduct a comprehensive assessment of image fidelity, clinical relevance, and semantic alignment. Our model achieves competitive performance across all tasks, significantly outperforming prior baselines for text-to-CT generation. Moreover, we demonstrate that CT scans synthesized by our framework can effectively augment real data, improving downstream diagnostic performance. Conclusion: Our results show that modality-specific vision-language alignment is a key component for high-quality 3D medical image generation. By integrating contrastive pretraining and volumetric diffusion, our method offers a scalable and controllable solution for synthesizing clinically meaningful CT volumes from text, paving the way for new applications in data augmentation, medical education, and automated clinical simulation.

Paperid: 2558, https://arxiv.org/pdf/2505.13973.pdf

Abstract:
Recently, reinforcement learning (RL)-based tuning has shifted the trajectory of Multimodal Large Language Models (MLLMs), particularly following the introduction of Group Relative Policy Optimization (GRPO). However, directly applying it to medical tasks remains challenging for achieving clinically grounded model behavior. Motivated by the need to align model response with clinical expectations, we investigate four critical dimensions that affect the effectiveness of RL-based tuning in medical visual question answering (VQA): base model initialization strategy, the role of medical semantic alignment, the impact of length-based rewards on long-chain reasoning, and the influence of bias. We conduct extensive experiments to analyze these factors for medical MLLMs, providing new insights into how models are domain-specifically fine-tuned. Additionally, our results also demonstrate that GRPO-based RL tuning consistently outperforms standard supervised fine-tuning (SFT) in both accuracy and reasoning quality.

Paperid: 2559, https://arxiv.org/pdf/2505.08548.pdf

Abstract:
Achieving generalization in robotic manipulation remains a critical challenge, particularly for unseen scenarios and novel tasks. Current Vision-Language-Action (VLA) models, while building on top of general Vision-Language Models (VLMs), still fall short of achieving robust zero-shot performance due to the scarcity and heterogeneity prevalent in embodied datasets. To address these limitations, we propose FSD (From Seeing to Doing), a novel vision-language model that generates intermediate representations through spatial relationship reasoning, providing fine-grained guidance for robotic manipulation. Our approach combines a hierarchical data pipeline for training with a self-consistency mechanism that aligns spatial coordinates with visual signals. Through extensive experiments, we comprehensively validated FSD's capabilities in both "seeing" and "doing," achieving outstanding performance across 8 benchmarks for general spatial reasoning and embodied reference abilities, as well as on our proposed more challenging benchmark VABench. We also verified zero-shot capabilities in robot manipulation, demonstrating significant performance improvements over baseline methods in both SimplerEnv and real robot settings. Experimental results show that FSD achieves 40.6% success rate in SimplerEnv and 72% success rate across 8 real-world tasks, outperforming the strongest baseline by 30%.

Paperid: 2560, https://arxiv.org/pdf/2505.01096.pdf

Abstract:
The integration of artificial intelligence in healthcare has opened new horizons for improving medical diagnostics and patient care. However, challenges persist in developing systems capable of generating accurate and contextually relevant radiology reports, particularly in low-resource languages. In this study, we present a comprehensive benchmark to evaluate the performance of instruction-tuned Vision-Language Models (VLMs) in the specialized task of radiology report generation across three low-resource languages: Italian, German, and Spanish. Employing the LLaVA architectural framework, we conducted a systematic evaluation of pre-trained models utilizing general datasets, domain-specific datasets, and low-resource language-specific datasets. In light of the unavailability of models that possess prior knowledge of both the medical domain and low-resource languages, we analyzed various adaptations to determine the most effective approach for these contexts. The results revealed that language-specific models substantially outperformed both general and domain-specific models in generating radiology reports, emphasizing the critical role of linguistic adaptation. Additionally, models fine-tuned with medical terminology exhibited enhanced performance across all languages compared to models with generic knowledge, highlighting the importance of domain-specific training. We also explored the influence of the temperature parameter on the coherence of report generation, providing insights for optimal model settings. Our findings highlight the importance of tailored language and domain-specific training for improving the quality and accuracy of radiological reports in multilingual settings. This research not only advances our understanding of VLMs adaptability in healthcare but also points to significant avenues for future investigations into model tuning and language-specific adaptations.

Paperid: 2561, https://arxiv.org/pdf/2505.01091.pdf

Abstract:
Generative models have revolutionized Artificial Intelligence (AI), particularly in multimodal applications. However, adapting these models to the medical domain poses unique challenges due to the complexity of medical data and the stringent need for clinical accuracy. In this work, we introduce a framework specifically designed for multimodal medical data generation. By enabling the generation of multi-view chest X-rays and their associated clinical report, it bridges the gap between general-purpose vision-language models and the specialized requirements of healthcare. Leveraging the MIMIC-CXR dataset, the proposed framework shows superior performance in generating high-fidelity images and semantically coherent reports. Our quantitative evaluation reveals significant results in terms of FID and BLEU scores, showcasing the quality of the generated data. Notably, our framework achieves comparable or even superior performance compared to real data on downstream disease classification tasks, underlining its potential as a tool for medical research and diagnostics. This study highlights the importance of domain-specific adaptations in enhancing the relevance and utility of generative models for clinical applications, paving the way for future advancements in synthetic multimodal medical data generation.

Paperid: 2562, https://arxiv.org/pdf/2504.18027.pdf

Abstract:
It is a challenging task for visually impaired people to perceive their surrounding environment due to the complexity of the natural scenes. Their personal and social activities are thus highly limited. This paper introduces a Large Vision-Language Model(LVLM) based environment perception system which helps them to better understand the surrounding environment, by capturing the current scene they face with a wearable device, and then letting them retrieve the analysis results through the device. The visually impaired people could acquire a global description of the scene by long pressing the screen to activate the LVLM output, retrieve the categories of the objects in the scene resulting from a segmentation model by tapping or swiping the screen, and get a detailed description of the objects they are interested in by double-tapping the screen. To help visually impaired people more accurately perceive the world, this paper proposes incorporating the segmentation result of the RGB image as external knowledge into the input of LVLM to reduce the LVLM's hallucination. Technical experiments on POPE, MME and LLaVA-QA90 show that the system could provide a more accurate description of the scene compared to Qwen-VL-Chat, exploratory experiments show that the system helps visually impaired people to perceive the surrounding environment effectively.

Paperid: 2563, https://arxiv.org/pdf/2504.16054.pdf

Authors:Physical Intelligence, Kevin Black, Noah Brown, James Darpinian, Karan Dhabalia, Danny Driess, Adnan Esmail, Michael Equi, Chelsea Finn, Niccolo Fusai, Manuel Y. Galliker, Dibya Ghosh, Lachy Groom, Karol Hausman, Brian Ichter, Szymon Jakubczak, Tim Jones, Liyiming Ke, Devin LeBlanc, Sergey Levine, Adrian Li-Bell, Mohith Mothukuri, Suraj Nair, Karl Pertsch, Allen Z. Ren, Lucy Xiaoyang Shi, Laura Smith, Jost Tobias Springenberg, Kyle Stachowicz, James Tanner, Quan Vuong, Homer Walke, Anna Walling, Haohuan Wang, Lili Yu, Ury Zhilinsky

Abstract:
In order for robots to be useful, they must perform practically relevant tasks in the real world, outside of the lab. While vision-language-action (VLA) models have demonstrated impressive results for end-to-end robot control, it remains an open question how far such models can generalize in the wild. We describe $Ï_{0.5}$, a new model based on $Ï_{0}$ that uses co-training on heterogeneous tasks to enable broad generalization. $Ï_{0.5}$\ uses data from multiple robots, high-level semantic prediction, web data, and other sources to enable broadly generalizable real-world robotic manipulation. Our system uses a combination of co-training and hybrid multi-modal examples that combine image observations, language commands, object detections, semantic subtask prediction, and low-level actions. Our experiments show that this kind of knowledge transfer is essential for effective generalization, and we demonstrate for the first time that an end-to-end learning-enabled robotic system can perform long-horizon and dexterous manipulation skills, such as cleaning a kitchen or bedroom, in entirely new homes.

Paperid: 2564, https://arxiv.org/pdf/2504.14904.pdf

Abstract:
Exponentially growing short video platforms (SVPs) face significant challenges in moderating content detrimental to users' mental health, particularly for minors. The dissemination of such content on SVPs can lead to catastrophic societal consequences. Although substantial efforts have been dedicated to moderating such content, existing methods suffer from critical limitations: (1) Manual review is prone to human bias and incurs high operational costs. (2) Automated methods, though efficient, lack nuanced content understanding, resulting in lower accuracy. (3) Industrial moderation regulations struggle to adapt to rapidly evolving trends due to long update cycles. In this paper, we annotate the first SVP content moderation benchmark with authentic user/reviewer feedback to fill the absence of benchmark in this field. Then we evaluate various methods on the benchmark to verify the existence of the aforementioned limitations. We further propose our common-law content moderation framework named KuaiMod to address these challenges. KuaiMod consists of three components: training data construction, offline adaptation, and online deployment & refinement. Leveraging large vision language model (VLM) and Chain-of-Thought (CoT) reasoning, KuaiMod adequately models video toxicity based on sparse user feedback and fosters dynamic moderation policy with rapid update speed and high accuracy. Offline experiments and large-scale online A/B test demonstrates the superiority of KuaiMod: KuaiMod achieves the best moderation performance on our benchmark. The deployment of KuaiMod reduces the user reporting rate by 20% and its application in video recommendation increases both Daily Active User (DAU) and APP Usage Time (AUT) on several Kuaishou scenarios. We have open-sourced our benchmark at https://kuaimod.github.io.

Paperid: 2565, https://arxiv.org/pdf/2504.13351.pdf

Abstract:
Learning to perform manipulation tasks from human videos is a promising approach for teaching robots. However, many manipulation tasks require changing control parameters during task execution, such as force, which visual data alone cannot capture. In this work, we leverage sensing devices such as armbands that measure human muscle activities and microphones that record sound, to capture the details in the human manipulation process, and enable robots to extract task plans and control parameters to perform the same task. To achieve this, we introduce Chain-of-Modality (CoM), a prompting strategy that enables Vision Language Models to reason about multimodal human demonstration data -- videos coupled with muscle or audio signals. By progressively integrating information from each modality, CoM refines a task plan and generates detailed control parameters, enabling robots to perform manipulation tasks based on a single multimodal human video prompt. Our experiments show that CoM delivers a threefold improvement in accuracy for extracting task plans and control parameters compared to baselines, with strong generalization to new task setups and objects in real-world robot experiments. Videos and code are available at https://chain-of-modality.github.io

Paperid: 2566, https://arxiv.org/pdf/2503.20020.pdf

Authors:Gemini Robotics Team, Saminda Abeyruwan, Joshua Ainslie, Jean-Baptiste Alayrac, Montserrat Gonzalez Arenas, Travis Armstrong, Ashwin Balakrishna, Robert Baruch, Maria Bauza, Michiel Blokzijl, Steven Bohez, Konstantinos Bousmalis, Anthony Brohan, Thomas Buschmann, Arunkumar Byravan, Serkan Cabi, Ken Caluwaerts, Federico Casarini, Oscar Chang, Jose Enrique Chen, Xi Chen, Hao-Tien Lewis Chiang, Krzysztof Choromanski, David D'Ambrosio, Sudeep Dasari, Todor Davchev, Coline Devin, Norman Di Palo, Tianli Ding, Adil Dostmohamed, Danny Driess, Yilun Du, Debidatta Dwibedi, Michael Elabd, Claudio Fantacci, Cody Fong, Erik Frey, Chuyuan Fu, Marissa Giustina, Keerthana Gopalakrishnan, Laura Graesser, Leonard Hasenclever, Nicolas Heess, Brandon Hernaez, Alexander Herzog, R. Alex Hofer, Jan Humplik, Atil Iscen, Mithun George Jacob, Deepali Jain, Ryan Julian, Dmitry Kalashnikov, M. Emre Karagozler, Stefani Karp, Chase Kew, Jerad Kirkland, Sean Kirmani, Yuheng Kuang, Thomas Lampe, Antoine Laurens, Isabel Leal, Alex X. Lee, Tsang-Wei Edward Lee, Jacky Liang, Yixin Lin, Sharath Maddineni, Anirudha Majumdar, Assaf Hurwitz Michaely, Robert Moreno, Michael Neunert, Francesco Nori, Carolina Parada, Emilio Parisotto, Peter Pastor, Acorn Pooley, Kanishka Rao, Krista Reymann, Dorsa Sadigh, Stefano Saliceti, Pannag Sanketi, Pierre Sermanet, Dhruv Shah, Mohit Sharma, Kathryn Shea, Charles Shu, Vikas Sindhwani, Sumeet Singh, Radu Soricut, Jost Tobias Springenberg, Rachel Sterneck, Razvan Surdulescu, Jie Tan, Jonathan Tompson, Vincent Vanhoucke, Jake Varley, Grace Vesom, Giulia Vezzani, Oriol Vinyals, Ayzaan Wahid, Stefan Welker, Paul Wohlhart, Fei Xia, Ted Xiao, Annie Xie, Jinyu Xie, Peng Xu, Sichun Xu, Ying Xu, Zhuo Xu, Yuxiang Yang, Rui Yao, Sergey Yaroshenko, Wenhao Yu, Wentao Yuan, Jingwei Zhang, Tingnan Zhang, Allan Zhou, Yuxiang Zhou

Abstract:
Recent advancements in large multimodal models have led to the emergence of remarkable generalist capabilities in digital domains, yet their translation to physical agents such as robots remains a significant challenge. This report introduces a new family of AI models purposefully designed for robotics and built upon the foundation of Gemini 2.0. We present Gemini Robotics, an advanced Vision-Language-Action (VLA) generalist model capable of directly controlling robots. Gemini Robotics executes smooth and reactive movements to tackle a wide range of complex manipulation tasks while also being robust to variations in object types and positions, handling unseen environments as well as following diverse, open vocabulary instructions. We show that with additional fine-tuning, Gemini Robotics can be specialized to new capabilities including solving long-horizon, highly dexterous tasks, learning new short-horizon tasks from as few as 100 demonstrations and adapting to completely novel robot embodiments. This is made possible because Gemini Robotics builds on top of the Gemini Robotics-ER model, the second model we introduce in this work. Gemini Robotics-ER (Embodied Reasoning) extends Gemini's multimodal reasoning capabilities into the physical world, with enhanced spatial and temporal understanding. This enables capabilities relevant to robotics including object detection, pointing, trajectory and grasp prediction, as well as multi-view correspondence and 3D bounding box predictions. We show how this novel combination can support a variety of robotics applications. We also discuss and address important safety considerations related to this new class of robotics foundation models. The Gemini Robotics family marks a substantial step towards developing general-purpose robots that realizes AI's potential in the physical world.

Paperid: 2567, https://arxiv.org/pdf/2503.06973.pdf

Abstract:
While conversational generative AI has shown considerable potential in enhancing decision-making for agricultural professionals, its exploration has predominantly been anchored in text-based interactions. The evolution of multimodal conversational AI, leveraging vast amounts of image-text data from diverse sources, marks a significant stride forward. However, the application of such advanced vision-language models in the agricultural domain, particularly for crop disease diagnosis, remains underexplored. In this work, we present the crop disease domain multimodal (CDDM) dataset, a pioneering resource designed to advance the field of agricultural research through the application of multimodal learning techniques. The dataset comprises 137,000 images of various crop diseases, accompanied by 1 million question-answer pairs that span a broad spectrum of agricultural knowledge, from disease identification to management practices. By integrating visual and textual data, CDDM facilitates the development of sophisticated question-answering systems capable of providing precise, useful advice to farmers and agricultural professionals. We demonstrate the utility of the dataset by finetuning state-of-the-art multimodal models, showcasing significant improvements in crop disease diagnosis. Specifically, we employed a novel finetuning strategy that utilizes low-rank adaptation (LoRA) to finetune the visual encoder, adapter and language model simultaneously. Our contributions include not only the dataset but also a finetuning strategy and a benchmark to stimulate further research in agricultural technology, aiming to bridge the gap between advanced AI techniques and practical agricultural applications. The dataset is available at https: //github.com/UnicomAI/UnicomBenchmark/tree/main/CDDMBench.

Paperid: 2568, https://arxiv.org/pdf/2502.19417.pdf

Abstract:
Generalist robots that can perform a range of different tasks in open-world settings must be able to not only reason about the steps needed to accomplish their goals, but also process complex instructions, prompts, and even feedback during task execution. Intricate instructions (e.g., "Could you make me a vegetarian sandwich?" or "I don't like that one") require not just the ability to physically perform the individual steps, but the ability to situate complex commands and feedback in the physical world. In this work, we describe a system that uses vision-language models in a hierarchical structure, first reasoning over complex prompts and user feedback to deduce the most appropriate next step to fulfill the task, and then performing that step with low-level actions. In contrast to direct instruction following methods that can fulfill simple commands ("pick up the cup"), our system can reason through complex prompts and incorporate situated feedback during task execution ("that's not trash"). We evaluate our system across three robotic platforms, including single-arm, dual-arm, and dual-arm mobile robots, demonstrating its ability to handle tasks such as cleaning messy tables, making sandwiches, and grocery shopping. Videos are available at https://www.pi.website/research/hirobot

Paperid: 2569, https://arxiv.org/pdf/2501.09747.pdf

Abstract:
Autoregressive sequence models, such as Transformer-based vision-language action (VLA) policies, can be tremendously effective for capturing complex and generalizable robotic behaviors. However, such models require us to choose a tokenization of our continuous action signals, which determines how the discrete symbols predicted by the model map to continuous robot actions. We find that current approaches for robot action tokenization, based on simple per-dimension, per-timestep binning schemes, typically perform poorly when learning dexterous skills from high-frequency robot data. To address this challenge, we propose a new compression-based tokenization scheme for robot actions, based on the discrete cosine transform. Our tokenization approach, Frequency-space Action Sequence Tokenization (FAST), enables us to train autoregressive VLAs for highly dexterous and high-frequency tasks where standard discretization methods fail completely. Based on FAST, we release FAST+, a universal robot action tokenizer, trained on 1M real robot action trajectories. It can be used as a black-box tokenizer for a wide range of robot action sequences, with diverse action spaces and control frequencies. Finally, we show that, when combined with the pi0 VLA, our method can scale to training on 10k hours of robot data and match the performance of diffusion VLAs, while reducing training time by up to 5x.

Paperid: 2570, https://arxiv.org/pdf/2501.06808.pdf

Abstract:
Remote sensing image semantic change detection is a method used to analyze remote sensing images, aiming to identify areas of change as well as categorize these changes within images of the same location taken at different times. Traditional change detection methods often face challenges in generalizing across semantic categories in practical scenarios. To address this issue, we introduce a novel approach called Semantic-CD, specifically designed for semantic change detection in remote sensing images. This method incorporates the open vocabulary semantics from the vision-language foundation model, CLIP. By utilizing CLIP's extensive vocabulary knowledge, our model enhances its ability to generalize across categories and improves segmentation through fully decoupled multi-task learning, which includes both binary change detection and semantic change detection tasks. Semantic-CD consists of four main components: a bi-temporal CLIP visual encoder for extracting features from bi-temporal images, an open semantic prompter for creating semantic cost volume maps with open vocabulary, a binary change detection decoder for generating binary change detection masks, and a semantic change detection decoder for producing semantic labels. Experimental results on the SECOND dataset demonstrate that Semantic-CD achieves more accurate masks and reduces semantic classification errors, illustrating its effectiveness in applying semantic priors from vision-language foundation models to SCD tasks.

Paperid: 2571, https://arxiv.org/pdf/2412.02779.pdf

Abstract:
Analog computing using non-volatile memristors has emerged as a promising solution for energy-efficient deep learning. New materials, like perovskites-based memristors are recently attractive due to their cost-effectiveness, energy efficiency and flexibility. Yet, challenges in material diversity and immature fabrications require extensive experimentation for device development. Moreover, significant non-idealities in these memristors often impede them for computing. Here, we propose a synergistic methodology to concurrently optimize perovskite memristor fabrication and develop robust analog DNNs that effectively address the inherent non-idealities of these memristors. Employing Bayesian optimization (BO) with a focus on usability, we efficiently identify optimal materials and fabrication conditions for perovskite memristors. Meanwhile, we developed "BayesMulti", a DNN training strategy utilizing BO-guided noise injection to improve the resistance of analog DNNs to memristor imperfections. Our approach theoretically ensures that within a certain range of parameter perturbations due to memristor non-idealities, the prediction outcomes remain consistent. Our integrated approach enables use of analog computing in much deeper and wider networks, which significantly outperforms existing methods in diverse tasks like image classification, autonomous driving, species identification, and large vision-language models, achieving up to 100-fold improvements. We further validate our methodology on a 10$\times$10 optimized perovskite memristor crossbar, demonstrating high accuracy in a classification task and low energy consumption. This study offers a versatile solution for efficient optimization of various analog computing systems, encompassing both devices and algorithms.

Paperid: 2572, https://arxiv.org/pdf/2411.11543.pdf

Abstract:
Benefiting from the powerful capabilities of Large Language Models (LLMs), pre-trained visual encoder models connected to LLMs form Vision Language Models (VLMs). However, recent research shows that the visual modality in VLMs is highly vulnerable, allowing attackers to bypass safety alignment in LLMs through visually transmitted content, launching harmful attacks. To address this challenge, we propose a progressive concept-based alignment strategy, PSA-VLM, which incorporates safety modules as concept bottlenecks to enhance visual modality safety alignment. By aligning model predictions with specific safety concepts, we improve defenses against risky images, enhancing explainability and controllability while minimally impacting general performance. Our method is obtained through two-stage training. The low computational cost of the first stage brings very effective performance improvement, and the fine-tuning of the language model in the second stage further improves the safety performance. Our method achieves state-of-the-art results on popular VLM safety benchmark.

Paperid: 2573, https://arxiv.org/pdf/2411.04549.pdf

Abstract:
Predicting temporal progress from visual trajectories is important for intelligent robots that can learn, adapt, and improve. However, learning such progress estimator, or temporal value function, across different tasks and domains requires both a large amount of diverse data and methods which can scale and generalize. To address these challenges, we present Generative Value Learning (\GVL), a universal value function estimator that leverages the world knowledge embedded in vision-language models (VLMs) to predict task progress. Naively asking a VLM to predict values for a video sequence performs poorly due to the strong temporal correlation between successive frames. Instead, GVL poses value estimation as a temporal ordering problem over shuffled video frames; this seemingly more challenging task encourages VLMs to more fully exploit their underlying semantic and temporal grounding capabilities to differentiate frames based on their perceived task progress, consequently producing significantly better value predictions. Without any robot or task specific training, GVL can in-context zero-shot and few-shot predict effective values for more than 300 distinct real-world tasks across diverse robot platforms, including challenging bimanual manipulation tasks. Furthermore, we demonstrate that GVL permits flexible multi-modal in-context learning via examples from heterogeneous tasks and embodiments, such as human videos. The generality of GVL enables various downstream applications pertinent to visuomotor policy learning, including dataset filtering, success detection, and advantage-weighted regression -- all without any model training or finetuning.

Paperid: 2574, https://arxiv.org/pdf/2410.24164.pdf

Abstract:
Robot learning holds tremendous promise to unlock the full potential of flexible, general, and dexterous robot systems, as well as to address some of the deepest questions in artificial intelligence. However, bringing robot learning to the level of generality required for effective real-world systems faces major obstacles in terms of data, generalization, and robustness. In this paper, we discuss how generalist robot policies (i.e., robot foundation models) can address these challenges, and how we can design effective generalist robot policies for complex and highly dexterous tasks. We propose a novel flow matching architecture built on top of a pre-trained vision-language model (VLM) to inherit Internet-scale semantic knowledge. We then discuss how this model can be trained on a large and diverse dataset from multiple dexterous robot platforms, including single-arm robots, dual-arm robots, and mobile manipulators. We evaluate our model in terms of its ability to perform tasks in zero shot after pre-training, follow language instructions from people and from a high-level VLM policy, and its ability to acquire new skills via fine-tuning. Our results cover a wide variety of tasks, such as laundry folding, table cleaning, and assembling boxes.

Paperid: 2575, https://arxiv.org/pdf/2410.21801.pdf

Abstract:
Instant Messaging is a popular means for daily communication, allowing users to send text and stickers. As the saying goes, "a picture is worth a thousand words", so developing an effective sticker retrieval technique is crucial for enhancing user experience. However, existing sticker retrieval methods rely on labeled data to interpret stickers, and general-purpose Vision-Language Models (VLMs) often struggle to capture the unique semantics of stickers. Additionally, relevant-based sticker retrieval methods lack personalization, creating a gap between diverse user expectations and retrieval results. To address these, we propose the Personalized Sticker Retrieval with Vision-Language Model framework, namely PerSRV, structured into offline calculations and online processing modules. The online retrieval part follows the paradigm of relevant recall and personalized ranking, supported by the offline pre-calculation parts, which are sticker semantic understanding, utility evaluation and personalization modules. Firstly, for sticker-level semantic understanding, we supervised fine-tuned LLaVA-1.5-7B to generate human-like sticker semantics, complemented by textual content extracted from figures and historical interaction queries. Secondly, we investigate three crowd-sourcing metrics for sticker utility evaluation. Thirdly, we cluster style centroids based on users' historical interactions to achieve personal preference modeling. Finally, we evaluate our proposed PerSRV method on a public sticker retrieval dataset from WeChat, containing 543,098 candidates and 12,568 interactions. Experimental results show that PerSRV significantly outperforms existing methods in multi-modal sticker retrieval. Additionally, our fine-tuned VLM delivers notable improvements in sticker semantic understandings.

Paperid: 2576, https://arxiv.org/pdf/2601.18323.pdf

Abstract:
The vision-language-action (VLA) paradigm has enabled powerful robotic control by leveraging vision-language models, but its reliance on large-scale, high-quality robot data limits its generalization. Generative world models offer a promising alternative for general-purpose embodied AI, yet a critical gap remains between their pixel-level plans and physically executable actions. To this end, we propose the Tool-Centric Inverse Dynamics Model (TC-IDM). By focusing on the tool's imagined trajectory as synthesized by the world model, TC-IDM establishes a robust intermediate representation that bridges the gap between visual planning and physical control. TC-IDM extracts the tool's point cloud trajectories via segmentation and 3D motion estimation from generated videos. Considering diverse tool attributes, our architecture employs decoupled action heads to project these planned trajectories into 6-DoF end-effector motions and corresponding control signals. This plan-and-translate paradigm not only supports a wide range of end-effectors but also significantly improves viewpoint invariance. Furthermore, it exhibits strong generalization capabilities across long-horizon and out-of-distribution tasks, including interacting with deformable objects. In real-world evaluations, the world model with TC-IDM achieves an average success rate of 61.11 percent, with 77.7 percent on simple tasks and 38.46 percent on zero-shot deformable object tasks. It substantially outperforms end-to-end VLA-style baselines and other inverse dynamics models.

Paperid: 2577, https://arxiv.org/pdf/2601.14874.pdf

Abstract:
Humanoid robots must adapt their contact behavior to diverse objects and tasks, yet most controllers rely on fixed, hand-tuned impedance gains and gripper settings. This paper introduces HumanoidVLM, a vision-language driven retrieval framework that enables the Unitree G1 humanoid to select task-appropriate Cartesian impedance parameters and gripper configurations directly from an egocentric RGB image. The system couples a vision-language model for semantic task inference with a FAISS-based Retrieval-Augmented Generation (RAG) module that retrieves experimentally validated stiffness-damping pairs and object-specific grasp angles from two custom databases, and executes them through a task-space impedance controller for compliant manipulation. We evaluate HumanoidVLM on 14 visual scenarios and achieve a retrieval accuracy of 93%. Real-world experiments show stable interaction dynamics, with z-axis tracking errors typically within 1-3.5 cm and virtual forces consistent with task-dependent impedance settings. These results demonstrate the feasibility of linking semantic perception with retrieval-based control as an interpretable path toward adaptive humanoid manipulation.

Paperid: 2578, https://arxiv.org/pdf/2601.13809.pdf

Abstract:
As aerial platforms evolve from passive observers to active manipulators, the challenge shifts toward designing intuitive interfaces that allow non-expert users to command these systems naturally. This work introduces a novel concept of autonomous aerial manipulation system capable of interpreting high-level natural language commands to retrieve objects and deliver them to a human user. The system is intended to integrate a MediaPipe based on Grounding DINO and a Vision-Language-Action (VLA) model with a custom-built drone equipped with a 1-DOF gripper and an Intel RealSense RGB-D camera. VLA performs semantic reasoning to interpret the intent of a user prompt and generates a prioritized task queue for grasping of relevant objects in the scene. Grounding DINO and dynamic A* planning algorithm are used to navigate and safely relocate the object. To ensure safe and natural interaction during the handover phase, the system employs a human-centric controller driven by MediaPipe. This module provides real-time human pose estimation, allowing the drone to employ visual servoing to maintain a stable, distinct position directly in front of the user, facilitating a comfortable handover. We demonstrate the system's efficacy through real-world experiments for localization and navigation, which resulted in a 0.164m, 0.070m, and 0.084m of max, mean euclidean, and root-mean squared errors, respectively, highlighting the feasibility of VLA for aerial manipulation operations.

Paperid: 2579, https://arxiv.org/pdf/2512.22278.pdf

Abstract:
The growing demand for prenatal ultrasound imaging has intensified a global shortage of trained sonographers, creating barriers to essential fetal health monitoring. Deep learning has the potential to enhance sonographers' efficiency and support the training of new practitioners. Vision-Language Models (VLMs) are particularly promising for ultrasound interpretation, as they can jointly process images and text to perform multiple clinical tasks within a single framework. However, despite the expansion of VLMs, no standardized benchmark exists to evaluate their performance in fetal ultrasound imaging. This gap is primarily due to the modality's challenging nature, operator dependency, and the limited public availability of datasets. To address this gap, we present Fetal-Gauge, the first and largest visual question answering benchmark specifically designed to evaluate VLMs across various fetal ultrasound tasks. Our benchmark comprises over 42,000 images and 93,000 question-answer pairs, spanning anatomical plane identification, visual grounding of anatomical structures, fetal orientation assessment, clinical view conformity, and clinical diagnosis. We systematically evaluate several state-of-the-art VLMs, including general-purpose and medical-specific models, and reveal a substantial performance gap: the best-performing model achieves only 55\% accuracy, far below clinical requirements. Our analysis identifies critical limitations of current VLMs in fetal ultrasound interpretation, highlighting the urgent need for domain-adapted architectures and specialized training approaches. Fetal-Gauge establishes a rigorous foundation for advancing multimodal deep learning in prenatal care and provides a pathway toward addressing global healthcare accessibility challenges. Our benchmark will be publicly available once the paper gets accepted.

Paperid: 2580, https://arxiv.org/pdf/2512.12302.pdf

Abstract:
While end-to-end autonomous driving has achieved remarkable progress in geometric control, current systems remain constrained by a command-following paradigm that relies on simple navigational instructions. Transitioning to genuinely intelligent agents requires the capability to interpret and fulfill high-level, abstract human intentions. However, this advancement is hindered by the lack of dedicated benchmarks and semantic-aware evaluation metrics. In this paper, we formally define the task of Intention-Driven End-to-End Autonomous Driving and present Intention-Drive, a comprehensive benchmark designed to bridge this gap. We construct a large-scale dataset featuring complex natural language intentions paired with high-fidelity sensor data. To overcome the limitations of conventional trajectory-based metrics, we introduce the Imagined Future Alignment (IFA), a novel evaluation protocol leveraging generative world models to assess the semantic fulfillment of human goals beyond mere geometric accuracy. Furthermore, we explore the solution space by proposing two distinct paradigms: an end-to-end vision-language planner and a hierarchical agent-based framework. The experiments reveal a critical dichotomy where existing models exhibit satisfactory driving stability but struggle significantly with intention fulfillment. Notably, the proposed frameworks demonstrate superior alignment with human intentions.

Paperid: 2581, https://arxiv.org/pdf/2512.11399.pdf

Abstract:
Vision-Language Models (VLMs) are able to process increasingly longer videos. Yet, important visual information is easily lost throughout the entire context and missed by VLMs. Also, it is important to design tools that enable cost-effective analysis of lengthy video content. In this paper, we propose a clip selection method that targets key video moments to be included in a multimodal summary. We divide the video into short clips and generate compact visual descriptions of each using a lightweight video captioning model. These are then passed to a large language model (LLM), which selects the K clips containing the most relevant visual information for a multimodal summary. We evaluate our approach on reference clips for the task, automatically derived from full human-annotated screenplays and summaries in the MovieSum dataset. We further show that these reference clips (less than 6% of the movie) are sufficient to build a complete multimodal summary of the movies in MovieSum. Using our clip selection method, we achieve a summarization performance close to that of these reference clips while capturing substantially more relevant video information than random clip selection. Importantly, we maintain low computational cost by relying on a lightweight captioning model.

Paperid: 2582, https://arxiv.org/pdf/2512.04529.pdf

Abstract:
Generating academic slides from scientific papers is a challenging multimodal reasoning task that requires both long context understanding and deliberate visual planning. Existing approaches largely reduce it to text only summarization, overlooking the visual component and design intensive nature of slide creation. In this paper we introduce SlideGen, an agentic, modular, and visual in the loop framework for scientific paper to slide generation. SlideGen orchestrates a group of vision language agents that reason collaboratively over the document structure and semantics, producing editable PPTX slides with logical flow and compelling visual presentation. By integrating coordinated outlining, mapping, arrangement, note synthesis, and iterative refinement, our system consistently delivers slides of expert level quality. Across diverse benchmarks and strong baselines, SlideGen outperforms existing methods in visual quality, content faithfulness, and readability, positioning it as the new state of the art in automated slide generation. Our work establishes a foundation for design aware multimodal slide generation, demonstrating how agentic collaboration can bridge understanding and presentation in complex multimodal reasoning tasks.

Paperid: 2583, https://arxiv.org/pdf/2512.02834.pdf

Abstract:
Vision-Language-Action (VLA) models, trained via flow-matching or diffusion objectives, excel at learning complex behaviors from large-scale, multi-modal datasets (e.g., human teleoperation, scripted policies). However, since VLAs incorporate diverse data modes in the pre-training stage, and the finetuning dataset often contains demonstration data collected in a kinematically suboptimal or undesirable way, it exists redundant action modes that are irrelevant to the success action modes of the downstream task. Specifically, we observe a critical inference-time fragility among various sampled noises after supervised finetuning of pre-trained VLAs. In this paper, we attribute this instability to the distribution shift between the VLA policy and the policy induced by stable success modes of the downstream task dataset. Thus, we propose \textbf{TACO}, a test-time-scaling (TTS) framework that applies a lightweight pseudo-count estimator as a high-fidelity verifier of action chunks. The VLA models integrated with TACO can execute the actions with maximum pseudo-count from all sampled action chunks, thereby preventing distribution shifts while preserving the generalization ability of VLAs since the constraint is applied only during inference. Our method resembles the classical anti-exploration principle in offline reinforcement learning (RL), and being gradient-free, it incurs significant computational benefits compared to RL update, especially for flow or diffusion-based VLAs which are difficult to perform RL update due to denoising process. Extensive experiments across four simulation benchmarks (RoboTwin2.0, Robotwin, LIBERO, SimplerEnv) and a dual-arm platform demonstrate that our method significantly improves the inference stability and success rates in downstream-task adaptations.

Paperid: 2584, https://arxiv.org/pdf/2511.23300.pdf

Abstract:
Safe and trustworthy Human Robot Interaction (HRI) requires robots not only to complete tasks but also to regulate impedance and speed according to scene context and human proximity. We present SafeHumanoid, an egocentric vision pipeline that links Vision Language Models (VLMs) with Retrieval-Augmented Generation (RAG) to schedule impedance and velocity parameters for a humanoid robot. Egocentric frames are processed by a structured VLM prompt, embedded and matched against a curated database of validated scenarios, and mapped to joint-level impedance commands via inverse kinematics. We evaluate the system on tabletop manipulation tasks with and without human presence, including wiping, object handovers, and liquid pouring. The results show that the pipeline adapts stiffness, damping, and speed profiles in a context-aware manner, maintaining task success while improving safety. Although current inference latency (up to 1.4 s) limits responsiveness in highly dynamic settings, SafeHumanoid demonstrates that semantic grounding of impedance control is a viable path toward safer, standard-compliant humanoid collaboration.

Paperid: 2585, https://arxiv.org/pdf/2511.19773.pdf

Abstract:
While recent vision-language models (VLMs) demonstrate strong image understanding, their ability to "think with images", i.e., to reason through multi-step visual interactions, remains limited. We introduce VISTA-Gym, a scalable training environment for incentivizing tool-integrated visual reasoning capabilities in VLMs. VISTA-Gym unifies diverse real-world multimodal reasoning tasks (7 tasks from 13 datasets in total) with a standardized interface for visual tools (e.g., grounding, parsing), executable interaction loops, verifiable feedback signals, and efficient trajectory logging, enabling visual agentic reinforcement learning at scale. While recent VLMs exhibit strong text-only reasoning, both proprietary and open-source models still struggle with tool selection, invocation, and coordination. With VISTA-Gym, we train VISTA-R1 to interleave tool-use with agentic reasoning via multi-turn trajectory sampling and end-to-end reinforcement learning. Extensive experiments across 11 public reasoning-intensive VQA benchmarks show that VISTA-R1-8B outperforms state-of-the-art baselines with similar sizes by 9.51%-18.72%, demonstrating VISTA-Gym as an effective training ground to unlock the tool-integrated reasoning capabilities for VLMs.

Paperid: 2586, https://arxiv.org/pdf/2510.24285.pdf

Abstract:
The limited capacity for fine-grained visual perception presents a critical bottleneck for Vision-Language Models (VLMs) in real-world applications. Addressing this is challenging due to the scarcity of high-quality data and the limitations of existing methods: supervised fine-tuning (SFT) often compromises general capabilities, while reinforcement fine-tuning (RFT) prioritizes textual reasoning over visual perception. To bridge this gap, we propose a novel two-stage task that structures visual perception learning as a coarse-to-fine progressive process. Based on this task formulation, we develop ViPER, a self-bootstrapping framework specifically designed to enable iterative evolution through self-critiquing and self-prediction. By synergistically integrating image-level and instance-level reconstruction with a two-stage reinforcement learning strategy, ViPER establishes a closed-loop training paradigm, where internally synthesized data directly fuel the enhancement of perceptual ability. Applied to the Qwen2.5-VL family, ViPER produces the Qwen-Viper series. With an average gain of 1.7% on seven comprehensive benchmarks spanning various tasks and up to 6.0% on fine-grained perception, Qwen-Viper consistently demonstrates superior performance across different vision-language scenarios while maintaining generalizability. Beyond enabling self-improvement in perceptual capabilities, ViPER provides concrete evidence for the reciprocal relationship between generation and understanding, a breakthrough to developing more autonomous and capable VLMs.

Paperid: 2587, https://arxiv.org/pdf/2510.11178.pdf

Abstract:
As vision-language models (VLMs) are deployed globally, their ability to understand culturally situated knowledge becomes essential. Yet, existing evaluations largely assess static recall or isolated visual grounding, leaving unanswered whether VLMs possess robust and transferable cultural understanding. We introduce BLEnD-Vis, a multimodal, multicultural benchmark designed to evaluate the robustness of everyday cultural knowledge in VLMs across linguistic rephrasings and visual modalities. Building on the BLEnD dataset, BLEnD-Vis constructs 313 culturally grounded question templates spanning 16 regions and generates three aligned multiple-choice formats: (i) a text-only baseline querying from Region $\to$ Entity, (ii) an inverted text-only variant (Entity $\to$ Region), and (iii) a VQA-style version of (ii) with generated images. The resulting benchmark comprises 4,916 images and over 21,000 multiple-choice question (MCQ) instances, validated through human annotation. BLEnD-Vis reveals significant fragility in current VLM cultural knowledge; models exhibit performance drops under linguistic rephrasing and, whilst visual cues often aid performance, low cross-modal consistency highlights challenges in robustly integrating textual and visual understanding, particularly for lower-resource regions. BLEnD-Vis thus provides a crucial testbed for systematically analysing cultural robustness and multimodal grounding, exposing limitations and guiding the development of more culturally competent VLMs.

Paperid: 2588, https://arxiv.org/pdf/2510.03483.pdf

Abstract:
Deep learning for medical imaging is hampered by task-specific models that lack generalizability and prognostic capabilities, while existing 'universal' approaches suffer from simplistic conditioning and poor medical semantic understanding. To address these limitations, we introduce DuPLUS, a deep learning framework for efficient multi-modal medical image analysis. DuPLUS introduces a novel vision-language framework that leverages hierarchical semantic prompts for fine-grained control over the analysis task, a capability absent in prior universal models. To enable extensibility to other medical tasks, it includes a hierarchical, text-controlled architecture driven by a unique dual-prompt mechanism. For segmentation, DuPLUS is able to generalize across three imaging modalities, ten different anatomically various medical datasets, encompassing more than 30 organs and tumor types. It outperforms the state-of-the-art task specific and universal models on 8 out of 10 datasets. We demonstrate extensibility of its text-controlled architecture by seamless integration of electronic health record (EHR) data for prognosis prediction, and on a head and neck cancer dataset, DuPLUS achieved a Concordance Index (CI) of 0.69. Parameter-efficient fine-tuning enables rapid adaptation to new tasks and modalities from varying centers, establishing DuPLUS as a versatile and clinically relevant solution for medical image analysis. The code for this work is made available at: https://anonymous.4open.science/r/DuPLUS-6C52

Paperid: 2589, https://arxiv.org/pdf/2510.00855.pdf

Abstract:
Trained on internet-scale video data, generative world models are increasingly recognized as powerful world simulators that can generate consistent and plausible dynamics over structure, motion, and physics. This raises a natural question: with the advent of strong video foundational models, might they supplant conventional vision encoder paradigms for general-purpose multimodal understanding? While recent studies have begun to explore the potential of world models on common vision tasks, these explorations typically lack a systematic investigation of generic, multimodal tasks. In this work, we strive to investigate the capabilities when world model priors are transferred into Vision-Language Models: we re-purpose a video diffusion model as a generative encoder to perform a single denoising step and treat the resulting latents as a set of visual embedding. We empirically investigate this class of models, which we refer to as World-Language Models (WorldLMs), and we find that generative encoders can capture latents useful for downstream understanding that show distinctions from conventional encoders. Naming our best-performing variant Dynamic Vision Aligner (DyVA), we further discover that this method significantly enhances spatial reasoning abilities and enables single-image models to perform multi-frame reasoning. Through the curation of a suite of visual reasoning tasks, we find DyVA to surpass both open-source and proprietary baselines, achieving state-of-the-art or comparable performance. We attribute these gains to WorldLM's inherited motion-consistency internalization from video pre-training. Finally, we systematically explore extensive model designs to highlight promising directions for future work. We hope our study can pave the way for a new family of VLMs that leverage priors from world models and are on a promising path towards generalist vision learners.

Paperid: 2590, https://arxiv.org/pdf/2509.19719.pdf

Abstract:
Automatically segmenting infected areas in radiological images is essential for diagnosing pulmonary infectious diseases. Recent studies have demonstrated that the accuracy of the medical image segmentation can be improved by incorporating clinical text reports as semantic guidance. However, the complex morphological changes of lesions and the inherent semantic gap between vision-language modalities prevent existing methods from effectively enhancing the representation of visual features and eliminating semantically irrelevant information, ultimately resulting in suboptimal segmentation performance. To address these problems, we propose a Frequency-domain Multi-modal Interaction model (FMISeg) for language-guided medical image segmentation. FMISeg is a late fusion model that establishes interaction between linguistic features and frequency-domain visual features in the decoder. Specifically, to enhance the visual representation, our method introduces a Frequency-domain Feature Bidirectional Interaction (FFBI) module to effectively fuse frequency-domain features. Furthermore, a Language-guided Frequency-domain Feature Interaction (LFFI) module is incorporated within the decoder to suppress semantically irrelevant visual features under the guidance of linguistic information. Experiments on QaTa-COV19 and MosMedData+ demonstrated that our method outperforms the state-of-the-art methods qualitatively and quantitatively.

Paperid: 2591, https://arxiv.org/pdf/2509.05949.pdf

Abstract:
The evolution of prompt learning methodologies has driven exploration of deeper prompt designs to enhance model performance. However, current deep text prompting approaches suffer from two critical limitations: Over-reliance on constrastive learning objectives that prioritize high-level semantic alignment, neglecting fine-grained feature optimization; Static prompts across all input categories, preventing content-aware adaptation. To address these limitations, we propose AttriPrompt-a novel framework that enhances and refines textual semantic representations by leveraging the intermediate-layer features of CLIP's vision encoder. We designed an Attribute Retrieval module that first clusters visual features from each layer. The aggregated visual features retrieve semantically similar prompts from a prompt pool, which are then concatenated to the input of every layer in the text encoder. Leveraging hierarchical visual information embedded in prompted text features, we introduce Dual-stream Contrastive Learning to realize fine-grained alignment. Furthermore, we introduce a Self-Regularization mechanism by applying explicit regularization constraints between the prompted and non-prompted text features to prevent overfitting on limited training data. Extensive experiments across three benchmarks demonstrate AttriPrompt's superiority over state-of-the-art methods, achieving up to 7.37\% improvement in the base-to-novel setting. The observed strength of our method in cross-domain knowledge transfer positions vision-language pre-trained models as more viable solutions for real-world implementation.

Paperid: 2592, https://arxiv.org/pdf/2508.03469.pdf

Abstract:
Recent advancements in Large Vision-Language Models (LVLMs) have demonstrated significant progress across multiple domains. However, these models still face the inherent challenge of integrating vision and language for collaborative inference, which often leads to "hallucinations", outputs that are not grounded in the corresponding images. Many efforts have been made to address these issues, but each comes with its own limitations, such as high computational cost or expensive dataset annotation. Recent research shows that LVLMs exhibit a long-term bias where hallucinations increase as the sequence length grows, yet the underlying cause remains poorly understood. Building on extensive research into attention mechanisms in LVLMs, we analyze the relationship between this long-term bias and visual attention. In our research, we identify a consistent phenomenon in current LVLMs: the model's attention to visual input diminishes as the generated sequence grows, which we hypothesize to be a key factor contributing to observed increasing hallucinations. Based on these insights, we propose Image attention-guided Key-value merging cOllaborative Decoding (IKOD), a collaborative decoding strategy generating more image-focused sequences. This method derives logits from shorter sequences with higher image attention through key-value merging and combines them with those from the original decoding, effectively mitigating attention degradation and suppressing hallucinations while not incurring too much inference cost. Extensive experiments on both hallucination and comprehensive benchmarks demonstrate IKOD's superior effectiveness in mitigating hallucinations and improving comprehensive capacities for LVLMs. Importantly, IKOD requires no additional training or external tools, making it a lightweight and efficient framework applicable to various models.

Paperid: 2593, https://arxiv.org/pdf/2508.01361.pdf

Abstract:
We present VLH, a novel Visual-Language-Haptic Foundation Model that unifies perception, language, and tactile feedback in aerial robotics and virtual reality. Unlike prior work that treats haptics as a secondary, reactive channel, VLH synthesizes mid-air force and vibration cues as a direct consequence of contextual visual understanding and natural language commands. Our platform comprises an 8-inch quadcopter equipped with dual inverse five-bar linkage arrays for localized haptic actuation, an egocentric VR camera, and an exocentric top-down view. Visual inputs and language instructions are processed by a fine-tuned OpenVLA backbone - adapted via LoRA on a bespoke dataset of 450 multimodal scenarios - to output a 7-dimensional action vector (Vx, Vy, Vz, Hx, Hy, Hz, Hv). INT8 quantization and a high-performance server ensure real-time operation at 4-5 Hz. In human-robot interaction experiments (90 flights), VLH achieved a 56.7% success rate for target acquisition (mean reach time 21.3 s, pose error 0.24 m) and 100% accuracy in texture discrimination. Generalization tests yielded 70.0% (visual), 54.4% (motion), 40.0% (physical), and 35.0% (semantic) performance on novel tasks. These results demonstrate VLH's ability to co-evolve haptic feedback with perceptual reasoning and intent, advancing expressive, immersive human-robot interactions.

Paperid: 2594, https://arxiv.org/pdf/2508.00356.pdf

Abstract:
We present a Collaborative Agent-Based Framework for Multi-Image Reasoning. Our approach tackles the challenge of interleaved multimodal reasoning across diverse datasets and task formats by employing a dual-agent system: a language-based PromptEngineer, which generates context-aware, task-specific prompts, and a VisionReasoner, a large vision-language model (LVLM) responsible for final inference. The framework is fully automated, modular, and training-free, enabling generalization across classification, question answering, and free-form generation tasks involving one or multiple input images. We evaluate our method on 18 diverse datasets from the 2025 MIRAGE Challenge (Track A), covering a broad spectrum of visual reasoning tasks including document QA, visual comparison, dialogue-based understanding, and scene-level inference. Our results demonstrate that LVLMs can effectively reason over multiple images when guided by informative prompts. Notably, Claude 3.7 achieves near-ceiling performance on challenging tasks such as TQA (99.13% accuracy), DocVQA (96.87%), and MMCoQA (75.28 ROUGE-L). We also explore how design choices-such as model selection, shot count, and input length-influence the reasoning performance of different LVLMs.

Paperid: 2595, https://arxiv.org/pdf/2506.18472.pdf

Abstract:
An ideal vision-language agent serves as a bridge between the human users and their surrounding physical world in real-world applications like autonomous driving and embodied agents, and proactively provides accurate and timely responses given user intents. An intriguing challenge arises when agents interact with the world as a dynamic data stream and ad-hoc queries from users: supporting knowledge for queries, namely evidence, usually appears asynchronously with the arrival time of queries, and agents need to ground their responses in historical data, present observations, and even future streams. We frame this challenge as Query-Evidence Asynchrony, where user queries and their supporting evidence typically arrive asynchronously in the streaming setting. This setting requires not only strong reasoning capabilities but also the ability to retain past observations and respond to queries with temporal awareness. In this paper, we introduce a diagnostic benchmark that evaluates Multimodal Large Language Models (MLLMs) on their ability to handle interaction with streaming data. Further, we present AViLA, Asynchronous Video-Language Agent for streaming data interaction that can handle ad-hoc queries and give time-aware responses. For this purpose, AViLA consists of three key modules: comprehensive memory retention, evidence identification, and evidence-grounded trigger, that are designed to maintain a general-purpose memory and respond readily and timely to queries. Our experiments show that existing models often fail to respond at appropriate times, while AViLA significantly improves both accuracy and temporal awareness. Our code and dataset will be publicly available.

Paperid: 2596, https://arxiv.org/pdf/2506.11496.pdf

Abstract:
High-resolution computed tomography (CT) imaging is essential for medical diagnosis but requires increased radiation exposure, creating a critical trade-off between image quality and patient safety. While deep learning methods have shown promise in CT super-resolution, they face challenges with complex degradations and limited medical training data. Meanwhile, large-scale pre-trained diffusion models, particularly Stable Diffusion, have demonstrated remarkable capabilities in synthesizing fine details across various vision tasks. Motivated by this, we propose a novel framework that adapts Stable Diffusion for CT blind super-resolution. We employ a practical degradation model to synthesize realistic low-quality images and leverage a pre-trained vision-language model to generate corresponding descriptions. Subsequently, we perform super-resolution using Stable Diffusion with a specialized controlling strategy, conditioned on both low-resolution inputs and the generated text descriptions. Extensive experiments show that our method outperforms existing approaches, demonstrating its potential for achieving high-quality CT imaging at reduced radiation doses. Our code will be made publicly available.

Paperid: 2597, https://arxiv.org/pdf/2506.09284.pdf

Abstract:
Understanding fine-grained object affordances is imperative for robots to manipulate objects in unstructured environments given open-ended task instructions. However, existing methods of visual affordance predictions often rely on manually annotated data or conditions only on a predefined set of tasks. We introduce UAD (Unsupervised Affordance Distillation), a method for distilling affordance knowledge from foundation models into a task-conditioned affordance model without any manual annotations. By leveraging the complementary strengths of large vision models and vision-language models, UAD automatically annotates a large-scale dataset with detailed $<$instruction, visual affordance$>$ pairs. Training only a lightweight task-conditioned decoder atop frozen features, UAD exhibits notable generalization to in-the-wild robotic scenes and to various human activities, despite only being trained on rendered objects in simulation. Using affordance provided by UAD as the observation space, we show an imitation learning policy that demonstrates promising generalization to unseen object instances, object categories, and even variations in task instructions after training on as few as 10 demonstrations. Project website: https://unsup-affordance.github.io/

Paperid: 2598, https://arxiv.org/pdf/2506.07961.pdf

Abstract:
Recently, leveraging pre-trained vision-language models (VLMs) for building vision-language-action (VLA) models has emerged as a promising approach to effective robot manipulation learning. However, only few methods incorporate 3D signals into VLMs for action prediction, and they do not fully leverage the spatial structure inherent in 3D data, leading to low sample efficiency. In this paper, we introduce BridgeVLA, a novel 3D VLA model that (1) projects 3D inputs to multiple 2D images, ensuring input alignment with the VLM backbone, and (2) utilizes 2D heatmaps for action prediction, unifying the input and output spaces within a consistent 2D image space. In addition, we propose a scalable pre-training method that equips the VLM backbone with the capability to predict 2D heatmaps before downstream policy learning. Extensive experiments show the proposed method is able to learn 3D manipulation efficiently and effectively. BridgeVLA outperforms state-of-the-art baseline methods across three simulation benchmarks. In RLBench, it improves the average success rate from 81.4% to 88.2%. In COLOSSEUM, it demonstrates significantly better performance in challenging generalization settings, boosting the average success rate from 56.7% to 64.0%. In GemBench, it surpasses all the comparing baseline methods in terms of average success rate. In real-robot experiments, BridgeVLA outperforms a state-of-the-art baseline method by 32% on average. It generalizes robustly in multiple out-of-distribution settings, including visual disturbances and unseen instructions. Remarkably, it is able to achieve a success rate of 96.8% on 10+ tasks with only 3 trajectories per task, highlighting its extraordinary sample efficiency. Project Website:https://bridgevla.github.io/

Paperid: 2599, https://arxiv.org/pdf/2505.23450.pdf

Abstract:
Long-horizon robotic manipulation poses significant challenges for autonomous systems, requiring extended reasoning, precise execution, and robust error recovery across complex sequential tasks. Current approaches, whether based on static planning or end-to-end visuomotor policies, suffer from error accumulation and lack effective verification mechanisms during execution, limiting their reliability in real-world scenarios. We present Agentic Robot, a brain-inspired framework that addresses these limitations through Standardized Action Procedure (SAP)--a novel coordination protocol governing component interactions throughout manipulation tasks. Drawing inspiration from Standardized Operating Procedures (SOPs) in human organizations, SAP establishes structured workflows for planning, execution, and verification phases. Our architecture comprises three specialized components: (1) a large reasoning model that decomposes high-level instructions into semantically coherent subgoals, (2) a vision-language-action executor that generates continuous control commands from real-time visual inputs, and (3) a temporal verifier that enables autonomous progression and error recovery through introspective assessment. This SAP-driven closed-loop design supports dynamic self-verification without external supervision. On the LIBERO benchmark, Agentic Robot achieves state-of-the-art performance with an average success rate of 79.6%, outperforming SpatialVLA by 6.1% and OpenVLA by 7.4% on long-horizon tasks. These results demonstrate that SAP-driven coordination between specialized components enhances both performance and interpretability in sequential manipulation, suggesting significant potential for reliable autonomous systems. Project Github: https://agentic-robot.github.io.

Paperid: 2600, https://arxiv.org/pdf/2505.17433.pdf

Abstract:
Memes have emerged as a popular form of multimodal online communication, where their interpretation heavily depends on the specific context in which they appear. Current approaches predominantly focus on isolated meme analysis, either for harmful content detection or standalone interpretation, overlooking a fundamental challenge: the same meme can express different intents depending on its conversational context. This oversight creates an evaluation gap: although humans intuitively recognize how context shapes meme interpretation, Large Vision Language Models (LVLMs) can hardly understand context-dependent meme intent. To address this critical limitation, we introduce MemeReaCon, a novel benchmark specifically designed to evaluate how LVLMs understand memes in their original context. We collected memes from five different Reddit communities, keeping each meme's image, the post text, and user comments together. We carefully labeled how the text and meme work together, what the poster intended, how the meme is structured, and how the community responded. Our tests with leading LVLMs show a clear weakness: models either fail to interpret critical information in the contexts, or overly focus on visual details while overlooking communicative purpose. MemeReaCon thus serves both as a diagnostic tool exposing current limitations and as a challenging benchmark to drive development toward more sophisticated LVLMs of the context-aware understanding.

Paperid: 2601, https://arxiv.org/pdf/2505.16763.pdf

Abstract:
Text-to-image models are powerful for producing high-quality images based on given text prompts, but crafting these prompts often requires specialized vocabulary. To address this, existing methods train rewriting models with supervision from large amounts of manually annotated data and trained aesthetic assessment models. To alleviate the dependence on data scale for model training and the biases introduced by trained models, we propose a novel prompt optimization framework, designed to rephrase a simple user prompt into a sophisticated prompt to a text-to-image model. Specifically, we employ the large vision language models (LVLMs) as the solver to rewrite the user prompt, and concurrently, employ LVLMs as a reward model to score the aesthetics and alignment of the images generated by the optimized prompt. Instead of laborious human feedback, we exploit the prior knowledge of the LVLM to provide rewards, i.e., AI feedback. Simultaneously, the solver and the reward model are unified into one model and iterated in reinforcement learning to achieve self-improvement by giving a solution and judging itself. Results on two popular datasets demonstrate that our method outperforms other strong competitors.

Paperid: 2602, https://arxiv.org/pdf/2505.11872.pdf

Abstract:
Recent advancements in prompt-based medical image segmentation have enabled clinicians to identify tumors using simple input like bounding boxes or text prompts. However, existing methods face challenges when doctors need to interact through natural language or when position reasoning is required - understanding spatial relationships between anatomical structures and pathologies. We present PRS-Med, a framework that integrates vision-language models with segmentation capabilities to generate both accurate segmentation masks and corresponding spatial reasoning outputs. Additionally, we introduce the MMRS dataset (Multimodal Medical in Positional Reasoning Segmentation), which provides diverse, spatially-grounded question-answer pairs to address the lack of position reasoning data in medical imaging. PRS-Med demonstrates superior performance across six imaging modalities (CT, MRI, X-ray, ultrasound, endoscopy, RGB), significantly outperforming state-of-the-art methods in both segmentation accuracy and position reasoning. Our approach enables intuitive doctor-system interaction through natural language, facilitating more efficient diagnoses. Our dataset pipeline, model, and codebase will be released to foster further research in spatially-aware multimodal reasoning for medical applications.

Paperid: 2603, https://arxiv.org/pdf/2505.06594.pdf

Abstract:
Vision-Language Models (VLMs) often struggle to balance visual and textual information when summarizing complex multimodal inputs, such as entire TV show episodes. In this paper, we propose a zero-shot video-to-text summarization approach that builds its own screenplay representation of an episode, effectively integrating key video moments, dialogue, and character information into a unified document. Unlike previous approaches, we simultaneously generate screenplays and name the characters in zero-shot, using only the audio, video, and transcripts as input. Additionally, we highlight that existing summarization metrics can fail to assess the multimodal content in summaries. To address this, we introduce MFactSum, a multimodal metric that evaluates summaries with respect to both vision and text modalities. Using MFactSum, we evaluate our screenplay summaries on the SummScreen3D dataset, demonstrating superiority against state-of-the-art VLMs such as Gemini 1.5 by generating summaries containing 20% more relevant visual information while requiring 75% less of the video as input.

Paperid: 2604, https://arxiv.org/pdf/2505.05130.pdf

Abstract:
Large pre-trained Vision-Language Models (VLMs), such as Contrastive Language-Image Pre-training (CLIP), have exhibited remarkable zero-shot performance across various image classification tasks. Fine-tuning these models on domain-specific datasets further enhances their effectiveness for downstream applications. However, fine-tuning in cloud environments raises significant concerns regarding data security and privacy. Federated Learning (FL) offers a decentralized solution by enabling model training across local clients without centralizing sensitive data, but the high communication and computation costs of transmitting full pre-trained models during training limit its scalability. Additionally, non-Independent and Identically Distributed (non-IID) data across local clients can negatively impact model convergence and performance. To address these challenges, we propose CacheFL, a novel federated learning method that replaces traditional full model fine-tuning with lightweight cache model fine-tuning. The cache model is initialized using a class-balanced dataset generated by a generative pre-trained model, effectively mitigating the impact of non-IID data. This cache model is then distributed to local clients for fine-tuning, and the updated parameters from each client are aggregated on the server and redistributed. With the updated cache model, the classification performance of CLIP is improved after just a few epochs. By limiting the training and communication to the cache model, CacheFL significantly reduces resource demands while ensuring data privacy and security. Extensive experiments conducted on ImageNet and 10 additional datasets demonstrate that CacheFL outperforms traditional approaches in terms of classification accuracy, resource efficiency, and privacy preservation.

Paperid: 2605, https://arxiv.org/pdf/2505.02569.pdf

Abstract:
This paper introduces HapticVLM, a novel multimodal system that integrates vision-language reasoning with deep convolutional networks to enable real-time haptic feedback. HapticVLM leverages a ConvNeXt-based material recognition module to generate robust visual embeddings for accurate identification of object materials, while a state-of-the-art Vision-Language Model (Qwen2-VL-2B-Instruct) infers ambient temperature from environmental cues. The system synthesizes tactile sensations by delivering vibrotactile feedback through speakers and thermal cues via a Peltier module, thereby bridging the gap between visual perception and tactile experience. Experimental evaluations demonstrate an average recognition accuracy of 84.67% across five distinct auditory-tactile patterns and a temperature estimation accuracy of 86.7% based on a tolerance-based evaluation method with an 8Â°C margin of error across 15 scenarios. Although promising, the current study is limited by the use of a small set of prominent patterns and a modest participant pool. Future work will focus on expanding the range of tactile patterns and increasing user studies to further refine and validate the system's performance. Overall, HapticVLM presents a significant step toward context-aware, multimodal haptic interaction with potential applications in virtual reality, and assistive technologies.

Paperid: 2606, https://arxiv.org/pdf/2504.02259.pdf

Abstract:
Efficiently understanding long-form videos remains a significant challenge in computer vision. In this work, we revisit temporal search paradigms for long-form video understanding and address a fundamental issue pertaining to all state-of-the-art (SOTA) long-context vision-language models (VLMs). Our contributions are twofold: First, we frame temporal search as a Long Video Haystack problem: finding a minimal set of relevant frames (e.g., one to five) from tens of thousands based on specific queries. Upon this formulation, we introduce LV-Haystack, the first dataset with 480 hours of videos, 15,092 human-annotated instances for both training and evaluation aiming to improve temporal search quality and efficiency. Results on LV-Haystack highlight a significant research gap in temporal search capabilities, with current SOTA search methods only achieving 2.1% temporal F1 score on the Longvideobench subset. Next, inspired by visual search in images, we propose a lightweight temporal search framework, T* that reframes costly temporal search as spatial search. T* leverages powerful visual localization techniques commonly used in images and introduces an adaptive zooming-in mechanism that operates across both temporal and spatial dimensions. Extensive experiments show that integrating T* with existing methods significantly improves SOTA long-form video understanding. Under an inference budget of 32 frames, T* improves GPT-4o's performance from 50.5% to 53.1% and LLaVA-OneVision-OV-72B's performance from 56.5% to 62.4% on the Longvideobench XL subset. Our code, benchmark, and models are provided in the Supplementary material.

Paperid: 2607, https://arxiv.org/pdf/2503.20384.pdf

Abstract:
Multimodal Large Language Models (MLLMs) excel in understanding complex language and visual data, enabling generalist robotic systems to interpret instructions and perform embodied tasks. Nevertheless, their real-world deployment is hindered by substantial computational and storage demands. Recent insights into the homogeneous patterns in the LLM layer have inspired sparsification techniques to address these challenges, such as early exit and token pruning. However, these methods often neglect the critical role of the final layers that encode the semantic information most relevant to downstream robotic tasks. Aligning with the recent breakthrough of the Shallow Brain Hypothesis (SBH) in neuroscience and the mixture of experts in model sparsification, we conceptualize each LLM layer as an expert and propose a Mixture-of-Layers Vision-Language-Action model (MoLe-VLA, or simply MoLe) architecture for dynamic LLM layer activation. We introduce a Spatial-Temporal Aware Router (STAR) for MoLe to selectively activate only parts of the layers based on the robot's current state, mimicking the brain's distinct signal pathways specialized for cognition and causal reasoning. Additionally, to compensate for the cognitive ability of LLMs lost in MoLe, we devise a Cognition Self-Knowledge Distillation (CogKD) framework. CogKD enhances the understanding of task demands and improves the generation of task-relevant action sequences by leveraging cognitive features. Extensive experiments conducted in both RLBench simulation and real-world environments demonstrate the superiority of MoLe-VLA in both efficiency and performance. Specifically, MoLe-VLA achieves an 8% improvement in the mean success rate across ten tasks while reducing computational costs by up to x5.6 compared to standard LLMs.

Paperid: 2608, https://arxiv.org/pdf/2503.06254.pdf

Abstract:
Multimodal retrieval-augmented generation (RAG) enhances the visual reasoning capability of vision-language models (VLMs) by dynamically accessing information from external knowledge bases. In this work, we introduce \textit{Poisoned-MRAG}, the first knowledge poisoning attack on multimodal RAG systems. Poisoned-MRAG injects a few carefully crafted image-text pairs into the multimodal knowledge database, manipulating VLMs to generate the attacker-desired response to a target query. Specifically, we formalize the attack as an optimization problem and propose two cross-modal attack strategies, dirty-label and clean-label, tailored to the attacker's knowledge and goals. Our extensive experiments across multiple knowledge databases and VLMs show that Poisoned-MRAG outperforms existing methods, achieving up to 98\% attack success rate with just five malicious image-text pairs injected into the InfoSeek database (481,782 pairs). Additionally, We evaluate 4 different defense strategies, including paraphrasing, duplicate removal, structure-driven mitigation, and purification, demonstrating their limited effectiveness and trade-offs against Poisoned-MRAG. Our results highlight the effectiveness and scalability of Poisoned-MRAG, underscoring its potential as a significant threat to multimodal RAG systems.

Paperid: 2609, https://arxiv.org/pdf/2502.17034.pdf

Abstract:
We propose a new concept, Evolution 6.0, which represents the evolution of robotics driven by Generative AI. When a robot lacks the necessary tools to accomplish a task requested by a human, it autonomously designs the required instruments and learns how to use them to achieve the goal. Evolution 6.0 is an autonomous robotic system powered by Vision-Language Models (VLMs), Vision-Language Action (VLA) models, and Text-to-3D generative models for tool design and task execution. The system comprises two key modules: the Tool Generation Module, which fabricates task-specific tools from visual and textual data, and the Action Generation Module, which converts natural language instructions into robotic actions. It integrates QwenVLM for environmental understanding, OpenVLA for task execution, and Llama-Mesh for 3D tool generation. Evaluation results demonstrate a 90% success rate for tool generation with a 10-second inference time, and action generation achieving 83.5% in physical and visual generalization, 70% in motion generalization, and 37% in semantic generalization. Future improvements will focus on bimanual manipulation, expanded task capabilities, and enhanced environmental interpretation to improve real-world adaptability.

Paperid: 2610, https://arxiv.org/pdf/2502.14888.pdf

Abstract:
Humans experience the world through multiple modalities, such as, vision, language, and speech, making it natural to explore the commonality and distinctions among them. In this work, we take a data-driven approach to address this question by analyzing interpretable, monosemantic features extracted from deep multimodal models. Specifically, we investigate CLIP, a prominent visual-language representation model trained on massive image-text pairs. Building on prior research in single-modal interpretability, we develop a set of multi-modal interpretability tools and measures designed to disentangle and analyze features learned from CLIP. Specifically, we introduce the Modality Dominance Score (MDS) to attribute each CLIP feature to a specific modality. We then map CLIP features into a more interpretable space, enabling us to categorize them into three distinct classes: vision features (single-modal), language features (single-modal), and visual-language features (cross-modal). Interestingly, this data-driven categorization closely aligns with human intuitive understandings of different modalities. We further show that this modality decomposition can benefit multiple downstream tasks, including reducing bias in gender detection, generating cross-modal adversarial examples, and enabling modal-specific feature control in text-to-image generation. These results indicate that large-scale multimodal models, when equipped with task-agnostic interpretability tools, can offer valuable insights into the relationships between different data modalities.

Paperid: 2611, https://arxiv.org/pdf/2502.14807.pdf

Abstract:
Foundation models are becoming increasingly effective in the medical domain, offering pre-trained models on large datasets that can be readily adapted for downstream tasks. Despite progress, fetal ultrasound images remain a challenging domain for foundation models due to their inherent complexity, often requiring substantial additional training and facing limitations due to the scarcity of paired multimodal data. To overcome these challenges, here we introduce FetalCLIP, a vision-language foundation model capable of generating universal representation of fetal ultrasound images. FetalCLIP was pre-trained using a multimodal learning approach on a diverse dataset of 210,035 fetal ultrasound images paired with text. This represents the largest paired dataset of its kind used for foundation model development to date. This unique training approach allows FetalCLIP to effectively learn the intricate anatomical features present in fetal ultrasound images, resulting in robust representations that can be used for a variety of downstream applications. In extensive benchmarking across a range of key fetal ultrasound applications, including classification, gestational age estimation, congenital heart defect (CHD) detection, and fetal structure segmentation, FetalCLIP outperformed all baselines while demonstrating remarkable generalizability and strong performance even with limited labeled data. We plan to release the FetalCLIP model publicly for the benefit of the broader scientific community.

Paperid: 2612, https://arxiv.org/pdf/2502.13175.pdf

Abstract:
Embodied AI systems, including robots and autonomous vehicles, are increasingly integrated into real-world applications, where they encounter a range of vulnerabilities stemming from both environmental and system-level factors. These vulnerabilities manifest through sensor spoofing, adversarial attacks, and failures in task and motion planning, posing significant challenges to robustness and safety. Despite the growing body of research, existing reviews rarely focus specifically on the unique safety and security challenges of embodied AI systems. Most prior work either addresses general AI vulnerabilities or focuses on isolated aspects, lacking a dedicated and unified framework tailored to embodied AI. This survey fills this critical gap by: (1) categorizing vulnerabilities specific to embodied AI into exogenous (e.g., physical attacks, cybersecurity threats) and endogenous (e.g., sensor failures, software flaws) origins; (2) systematically analyzing adversarial attack paradigms unique to embodied AI, with a focus on their impact on perception, decision-making, and embodied interaction; (3) investigating attack vectors targeting large vision-language models (LVLMs) and large language models (LLMs) within embodied systems, such as jailbreak attacks and instruction misinterpretation; (4) evaluating robustness challenges in algorithms for embodied perception, decision-making, and task planning; and (5) proposing targeted strategies to enhance the safety and reliability of embodied AI systems. By integrating these dimensions, we provide a comprehensive framework for understanding the interplay between vulnerabilities and safety in embodied AI.

Paperid: 2613, https://arxiv.org/pdf/2502.09818.pdf

Abstract:
Although vision-language models (VLMs) have achieved significant success in various applications such as visual question answering, their resilience to prompt variations remains an under-explored area. Understanding how distractions affect VLMs is crucial for improving their real-world applicability, as inputs could have noisy and irrelevant information in many practical scenarios. This paper aims to assess the robustness of VLMs against both visual and textual distractions in the context of science question answering. Built on the ScienceQA dataset, we developed a new benchmark that introduces distractions in both the visual and textual contexts to evaluate the reasoning capacity of VLMs amid these distractions. Our findings reveal that most-of-the-art VLMs, including GPT-4, are vulnerable to various types of distractions, experiencing noticeable degradation in reasoning capabilities when confronted with distractions. Notably, models such as InternVL2 demonstrate a higher degree of robustness to these distractions. We also found that models exhibit greater sensitivity to textual distractions than visual ones. Additionally, we explored various mitigation strategies, such as prompt engineering, to counteract the impact of distractions. While these strategies improved solution accuracy, our analysis shows that there remain significant opportunities for improvement.

Paperid: 2614, https://arxiv.org/pdf/2501.15579.pdf

Abstract:
The clinical adoption of artificial intelligence (AI) in medical imaging requires models that are both diagnostically accurate and interpretable to clinicians. While current multimodal biomedical foundation models prioritize performance, their black-box nature hinders explaining the decision-making process in clinically meaningful concepts. Here, we present ConceptCLIP, the first explainable biomedical foundation model that achieves state-of-the-art diagnostic accuracy while delivering human-interpretable explanations across diverse imaging modalities. We curate MedConcept-23M, the largest pre-training dataset comprising 23 million image-text-concept triplets across diverse medical modalities, where clinical concepts are derived from the Unified Medical Language System. Leveraging this dataset, we develop ConceptCLIP through a novel dual-alignment approach that simultaneously learns global image-text representations and fine-grained region-concept associations for precise and interpretable medical image analysis. We curate the most extensive evaluation benchmark for multimodal biomedical foundation models, covering 52 clinical tasks spanning 10 imaging modalities. Extensive experiments demonstrate that ConceptCLIP outperforms existing state-of-the-art multimodal biomedical foundation models. Importantly, ConceptCLIP demonstrates superior diagnostic performance while providing human-understandable explanations validated by clinical experts. As the first precise and interpretable biomedical foundation model, ConceptCLIP represents a critical milestone toward the widespread clinical adoption of AI, thereby advancing trustworthy AI in medicine.

Paperid: 2615, https://arxiv.org/pdf/2501.06919.pdf

Abstract:
This paper introduces Shake-VLA, a Vision-Language-Action (VLA) model-based system designed to enable bimanual robotic manipulation for automated cocktail preparation. The system integrates a vision module for detecting ingredient bottles and reading labels, a speech-to-text module for interpreting user commands, and a language model to generate task-specific robotic instructions. Force Torque (FT) sensors are employed to precisely measure the quantity of liquid poured, ensuring accuracy in ingredient proportions during the mixing process. The system architecture includes a Retrieval-Augmented Generation (RAG) module for accessing and adapting recipes, an anomaly detection mechanism to address ingredient availability issues, and bimanual robotic arms for dexterous manipulation. Experimental evaluations demonstrated a high success rate across system components, with the speech-to-text module achieving a 93% success rate in noisy environments, the vision module attaining a 91% success rate in object and label detection in cluttered environment, the anomaly module successfully identified 95% of discrepancies between detected ingredients and recipe requirements, and the system achieved an overall success rate of 100% in preparing cocktails, from recipe formulation to action generation.

Paperid: 2616, https://arxiv.org/pdf/2501.05014.pdf

Abstract:
The UAV-VLA (Visual-Language-Action) system is a tool designed to facilitate communication with aerial robots. By integrating satellite imagery processing with the Visual Language Model (VLM) and the powerful capabilities of GPT, UAV-VLA enables users to generate general flight paths-and-action plans through simple text requests. This system leverages the rich contextual information provided by satellite images, allowing for enhanced decision-making and mission planning. The combination of visual analysis by VLM and natural language processing by GPT can provide the user with the path-and-action set, making aerial operations more efficient and accessible. The newly developed method showed the difference in the length of the created trajectory in 22% and the mean error in finding the objects of interest on a map in 34.22 m by Euclidean distance in the K-Nearest Neighbors (KNN) approach.

Paperid: 2617, https://arxiv.org/pdf/2412.19806.pdf

Abstract:
Recent developments of vision large language models (LLMs) have seen remarkable progress, yet still encounter challenges towards multimodal generalists, such as coarse-grained instance-level understanding, lack of unified support for both images and videos, and insufficient coverage across various vision tasks. In this paper, we present VITRON, a universal pixel-level vision LLM designed for comprehensive understanding, generating, segmenting, and editing of both static images and dynamic videos. Building on top of an LLM backbone, VITRON incorporates encoders for images, videos, and pixel-level regional visuals within its frontend modules, while employing state-of-the-art visual specialists as its backend, via which VITRON supports a spectrum of vision end tasks, spanning visual comprehension to visual generation, from low level to high level. To ensure an effective and precise message passing from LLM to backend modules for function invocation, we propose a novel hybrid method by simultaneously integrating discrete textual instructions and continuous signal embeddings. Further, we design various pixel-level spatiotemporal vision-language alignment learning for VITRON to reach the best fine-grained visual capability. Finally, a cross-task synergy module is advised to learn to maximize the task-invariant fine-grained visual features, enhancing the synergy between different visual tasks. Demonstrated over 12 visual tasks and evaluated across 22 datasets, VITRON showcases its extensive capabilities in the four main vision task clusters. Overall, this work illuminates the great potential of developing a more unified multimodal generalist. Project homepage: https://vitron-llm.github.io/

Paperid: 2618, https://arxiv.org/pdf/2412.10137.pdf

Abstract:
We address the task of Vision-Language Navigation in Continuous Environments (VLN-CE) under the zero-shot setting. Zero-shot VLN-CE is particularly challenging due to the absence of expert demonstrations for training and minimal environment structural prior to guide navigation. To confront these challenges, we propose a Constraint-Aware Navigator (CA-Nav), which reframes zero-shot VLN-CE as a sequential, constraint-aware sub-instruction completion process. CA-Nav continuously translates sub-instructions into navigation plans using two core modules: the Constraint-Aware Sub-instruction Manager (CSM) and the Constraint-Aware Value Mapper (CVM). CSM defines the completion criteria for decomposed sub-instructions as constraints and tracks navigation progress by switching sub-instructions in a constraint-aware manner. CVM, guided by CSM's constraints, generates a value map on the fly and refines it using superpixel clustering to improve navigation stability. CA-Nav achieves the state-of-the-art performance on two VLN-CE benchmarks, surpassing the previous best method by 12 percent and 13 percent in Success Rate on the validation unseen splits of R2R-CE and RxR-CE, respectively. Moreover, CA-Nav demonstrates its effectiveness in real-world robot deployments across various indoor scenes and instructions.

Paperid: 2619, https://arxiv.org/pdf/2412.00174.pdf

Abstract:
Human beings are social animals. How to equip 3D autonomous characters with similar social intelligence that can perceive, understand and interact with humans remains an open yet foundamental problem. In this paper, we introduce SOLAMI, the first end-to-end Social vision-Language-Action (VLA) Modeling framework for Immersive interaction with 3D autonomous characters. Specifically, SOLAMI builds 3D autonomous characters from three aspects: (1) Social VLA Architecture: We propose a unified social VLA framework to generate multimodal response (speech and motion) based on the user's multimodal input to drive the character for social interaction. (2) Interactive Multimodal Data: We present SynMSI, a synthetic multimodal social interaction dataset generated by an automatic pipeline using only existing motion datasets to address the issue of data scarcity. (3) Immersive VR Interface: We develop a VR interface that enables users to immersively interact with these characters driven by various architectures. Extensive quantitative experiments and user studies demonstrate that our framework leads to more precise and natural character responses (in both speech and motion) that align with user expectations with lower latency.

Paperid: 2620, https://arxiv.org/pdf/2601.03928.pdf

Abstract:
Vision-Language Models (VLMs) have shown remarkable performance in User Interface (UI) grounding tasks, driven by their ability to process increasingly high-resolution screenshots. However, screenshots are tokenized into thousands of visual tokens (e.g., about 4700 for 2K resolution), incurring significant computational overhead and diluting attention. In contrast, humans typically focus on regions of interest when interacting with UI. In this work, we pioneer the task of efficient UI grounding. Guided by practical analysis of the task's characteristics and challenges, we propose FocusUI, an efficient UI grounding framework that selects patches most relevant to the instruction while preserving positional continuity for precise grounding. FocusUI addresses two key challenges: (1) Eliminating redundant tokens in visual encoding. We construct patch-level supervision by fusing an instruction-conditioned score with a rule-based UI-graph score that down-weights large homogeneous regions to select distinct and instruction-relevant visual tokens. (2) Preserving positional continuity during visual token selection. We find that general visual token pruning methods suffer from severe accuracy degradation on UI grounding tasks due to broken positional information. We introduce a novel PosPad strategy, which compresses each contiguous sequence of dropped visual tokens into a single special marker placed at the sequence's last index to preserve positional continuity. Comprehensive experiments on four grounding benchmarks demonstrate that FocusUI surpasses GUI-specific baselines. On the ScreenSpot-Pro benchmark, FocusUI-7B achieves a performance improvement of 3.7% over GUI-Actor-7B. Even with only 30% visual token retention, FocusUI-7B drops by only 3.2% while achieving up to 1.44x faster inference and 17% lower peak GPU memory.

Paperid: 2621, https://arxiv.org/pdf/2512.22183.pdf

Abstract:
End-to-end Vision-language Models (VLMs) often answer visual questions by exploiting spurious correlations instead of causal visual evidence, and can become more shortcut-prone when fine-tuned. We introduce VISTA (Visual-Information Separation for Text-based Analysis), a modular framework that decouples perception from reasoning via an explicit information bottleneck. A frozen VLM sensor is restricted to short, objective perception queries, while a text-only LLM reasoner decomposes each question, plans queries, and aggregates visual facts in natural language. This controlled interface defines a reward-aligned environment for training unbiased visual reasoning with reinforcement learning. Instantiated with Qwen2.5-VL and Llama3.2-Vision sensors, and trained with GRPO from only 641 curated multi-step questions, VISTA significantly improves robustness to real-world spurious correlations on SpuriVerse (+16.29% with Qwen-2.5-VL-7B and +6.77% with Llama-3.2-Vision-11B), while remaining competitive on MMVP and a balanced SeedBench subset. VISTA transfers robustly across unseen VLM sensors and is able to recognize and recover from VLM perception failures. Human analysis further shows that VISTA's reasoning traces are more neutral, less reliant on spurious attributes, and more explicitly grounded in visual evidence than end-to-end VLM baselines.

Paperid: 2622, https://arxiv.org/pdf/2512.06769.pdf

Abstract:
Existing vision-language models often suffer from spatial hallucinations, i.e., generating incorrect descriptions about the relative positions of objects in an image. We argue that this problem mainly stems from the asymmetric properties between images and text. To enrich the spatial understanding ability of vision-language models, we propose a simple, annotation-free, plug-and-play method named $\text{Stitch and Tell}$ (abbreviated as SiTe), which injects structured spatial supervision into data. It constructs stitched image-text pairs by stitching images along a spatial axis and generating spatially-aware captions or question answer pairs based on the layout of stitched image, without relying on costly advanced models or human involvement. We evaluate SiTe across three architectures including LLaVA-v1.5-7B, LLaVA-Qwen2-1.5B and HALVA-7B, two training datasets, and eight benchmarks. Experiments show that SiTe improves spatial understanding tasks such as $\text{MME}_{\text{Position}}$ (+5.50%) and Spatial-MM (+4.19%), while maintaining or improving performance on general vision-language benchmarks including COCO-QA (+1.02%) and MMBench (+4.76%). Our findings suggest that explicitly injecting spatially-aware structure into training data offers an effective way to mitigate spatial hallucinations and improve spatial understanding, while preserving general vision-language capabilities.

Paperid: 2623, https://arxiv.org/pdf/2512.01300.pdf

Abstract:
Current Vision-Language Model (VLM)-based end-to-end autonomous driving systems often leverage large language models to generate driving decisions directly based on their understanding of the current scene. However, such systems introduce multiple risks in real-world driving scenarios. To evaluate whether VLMs are truly viable for autonomous driving, we introduce RoboDriveBench, the first robustness benchmark focused on end-to-end trajectory prediction tasks. This benchmark systematically evaluates two critical categories of real-world challenges for VLM-based end-to-end autonomous driving systems through 11 simulated scenarios encompassing various corruption types, including 6 scenarios of sensor corruption caused by environmental variations, along with 5 cases of prompt corruption resulting from human intervention and data transmission failures. Each corruption type includes 250 unique driving scenarios and 5,689 frames, resulting in 64,559 total trajectory prediction cases per evaluation. To overcome these real-world challenges, we propose a novel VLM-based autonomous driving framework called RoboDriveVLM, which enhances robustness by mapping more multimodal data-e.g., lidar and radar-into a unified latent space. Furthermore, we introduce a new Test-Time Adaptation (TTA) method based on cross-modal knowledge distillation to improve the robustness of VLM-based autonomous driving systems. Through extensive experiments, our work highlights the limitations of current VLM-based end-to-end autonomous driving systems and provides a more reliable solution for real-world deployment. Source code and datasets will be released.

Paperid: 2624, https://arxiv.org/pdf/2511.09141.pdf

Abstract:
Humanoid robots exhibit significant potential in executing diverse human-level skills. However, current research predominantly relies on data-driven approaches that necessitate extensive training datasets to achieve robust multimodal decision-making capabilities and generalizable visuomotor control. These methods raise concerns due to the neglect of geometric reasoning in unseen scenarios and the inefficient modeling of robot-target relationships within the training data, resulting in significant waste of training resources. To address these limitations, we present the Recurrent Geometric-prior Multimodal Policy (RGMP), an end-to-end framework that unifies geometric-semantic skill reasoning with data-efficient visuomotor control. For perception capabilities, we propose the Geometric-prior Skill Selector, which infuses geometric inductive biases into a vision language model, producing adaptive skill sequences for unseen scenes with minimal spatial common sense tuning. To achieve data-efficient robotic motion synthesis, we introduce the Adaptive Recursive Gaussian Network, which parameterizes robot-object interactions as a compact hierarchy of Gaussian processes that recursively encode multi-scale spatial relationships, yielding dexterous, data-efficient motion synthesis even from sparse demonstrations. Evaluated on both our humanoid robot and desktop dual-arm robot, the RGMP framework achieves 87% task success in generalization tests and exhibits 5x greater data efficiency than the state-of-the-art model. This performance underscores its superior cross-domain generalization, enabled by geometric-semantic reasoning and recursive-Gaussion adaptation.

Paperid: 2625, https://arxiv.org/pdf/2510.11196.pdf

Abstract:
Vision-language models (VLMs) often produce chain-of-thought (CoT) explanations that sound plausible yet fail to reflect the underlying decision process, undermining trust in high-stakes clinical use. Existing evaluations rarely catch this misalignment, prioritizing answer accuracy or adherence to formats. We present a clinically grounded framework for chest X-ray visual question answering (VQA) that probes CoT faithfulness via controlled text and image modifications across three axes: clinical fidelity, causal attribution, and confidence calibration. In a reader study (n=4), evaluator-radiologist correlations fall within the observed inter-radiologist range for all axes, with strong alignment for attribution (Kendall's $τ_b=0.670$), moderate alignment for fidelity ($τ_b=0.387$), and weak alignment for confidence tone ($τ_b=0.091$), which we report with caution. Benchmarking six VLMs shows that answer accuracy and explanation quality are decoupled, acknowledging injected cues does not ensure grounding, and text cues shift explanations more than visual cues. While some open-source models match final answer accuracy, proprietary models score higher on attribution (25.0% vs. 1.4%) and often on fidelity (36.1% vs. 31.7%), highlighting deployment risks and the need to evaluate beyond final answer accuracy.

Paperid: 2626, https://arxiv.org/pdf/2509.24231.pdf

Authors:Yang Bai, Haoran Cheng, Yang Zhou, Jun Zhou, Arun Thirunavukarasu, Yuhe Ke, Jie Yao, Kanae Fukutsu, Chrystie Wan Ning Quek, Ashley Hong, Laura Gutierrez, Zhen Ling Teo, Darren Shu Jeng Ting, Brian T. Soetikno, Christopher S. Nielsen, Tobias Elze, Zengxiang Li, Linh Le Dinh, Hiok Hong Chan, Victor Koh, Marcus Tan, Kelvin Z. Li, Leonard Yip, Ching Yu Cheng, Yih Chung Tham, Gavin Siew Wei Tan, Leopold Schmetterer, Marcus Ang, Rahat Hussain, Jod Mehta, Tin Aung, Lionel Tim-Ee Cheng, Tran Nguyen Tuan Anh, Chee Leong Cheng, Tien Yin Wong, Nan Liu, Iain Beehuat Tan, Soon Thye Lim, Eyal Klang, Tony Kiat Hon Lim, Rick Siow Mong Goh, Yong Liu, Daniel Shu Wei Ting

Abstract:
Despite the promise of foundation models in medical AI, current systems remain limited - they are modality-specific and lack transparent reasoning processes, hindering clinical adoption. To address this gap, we present EVLF-FM, a multimodal vision-language foundation model (VLM) designed to unify broad diagnostic capability with fine-grain explainability. The development and testing of EVLF-FM encompassed over 1.3 million total samples from 23 global datasets across eleven imaging modalities related to six clinical specialties: dermatology, hepatology, ophthalmology, pathology, pulmonology, and radiology. External validation employed 8,884 independent test samples from 10 additional datasets across five imaging modalities. Technically, EVLF-FM is developed to assist with multiple disease diagnosis and visual question answering with pixel-level visual grounding and reasoning capabilities. In internal validation for disease diagnostics, EVLF-FM achieved the highest average accuracy (0.858) and F1-score (0.797), outperforming leading generalist and specialist models. In medical visual grounding, EVLF-FM also achieved stellar performance across nine modalities with average mIOU of 0.743 and Acc@0.5 of 0.837. External validations further confirmed strong zero-shot and few-shot performance, with competitive F1-scores despite a smaller model size. Through a hybrid training strategy combining supervised and visual reinforcement fine-tuning, EVLF-FM not only achieves state-of-the-art accuracy but also exhibits step-by-step reasoning, aligning outputs with visual evidence. EVLF-FM is an early multi-disease VLM model with explainability and reasoning capabilities that could advance adoption of and trust in foundation models for real-world clinical deployment.

Paperid: 2627, https://arxiv.org/pdf/2507.17239.pdf

Abstract:
Foundation models have recently gained tremendous popularity in medical image analysis. State-of-the-art methods leverage either paired image-text data via vision-language pre-training or unpaired image data via self-supervised pre-training to learn foundation models with generalizable image features to boost downstream task performance. However, learning foundation models exclusively on either paired or unpaired image data limits their ability to learn richer and more comprehensive image features. In this paper, we investigate a novel task termed semi-supervised vision-language pre-training, aiming to fully harness the potential of both paired and unpaired image data for foundation model learning. To this end, we propose MaskedCLIP, a synergistic masked image modeling and contrastive language-image pre-training framework for semi-supervised vision-language pre-training. The key challenge in combining paired and unpaired image data for learning a foundation model lies in the incompatible feature spaces derived from these two types of data. To address this issue, we propose to connect the masked feature space with the CLIP feature space with a bridge transformer. In this way, the more semantic specific CLIP features can benefit from the more general masked features for semantic feature extraction. We further propose a masked knowledge distillation loss to distill semantic knowledge of original image features in CLIP feature space back to the predicted masked image features in masked feature space. With this mutually interactive design, our framework effectively leverages both paired and unpaired image data to learn more generalizable image features for downstream tasks. Extensive experiments on retinal image analysis demonstrate the effectiveness and data efficiency of our method.

Paperid: 2628, https://arxiv.org/pdf/2507.11969.pdf

Abstract:
Recent advances in test-time adaptation (TTA) for Vision-Language Models (VLMs) have garnered increasing attention, particularly through the use of multiple augmented views of a single image to boost zero-shot generalization. Unfortunately, existing methods fail to strike a satisfactory balance between performance and efficiency, either due to excessive overhead of tuning text prompts or unstable benefits from handcrafted, training-free visual feature enhancement. In this paper, we present Global-Spatial Bias Learner (GS-Bias), an efficient and effective TTA paradigm that incorporates two learnable biases during TTA, unfolded as the global bias and spatial bias. Particularly, the global bias captures the global semantic features of a test image by learning consistency across augmented views, while spatial bias learns the semantic coherence between regions in the image's spatial visual representation. It is worth highlighting that these two sets of biases are directly added to the logits outputed by the pretrained VLMs, which circumvent the full backpropagation through VLM that hinders the efficiency of existing TTA methods. This endows GS-Bias with extremely high efficiency while achieving state-of-the-art performance on 15 benchmark datasets. For example, it achieves a 2.23% improvement over TPT in cross-dataset generalization and a 2.72% improvement in domain generalization, while requiring only 6.5% of TPT's memory usage on ImageNet.

Paperid: 2629, https://arxiv.org/pdf/2507.11200.pdf

Abstract:
Vision-Language Models (VLMs) trained on web-scale corpora excel at natural image tasks and are increasingly repurposed for healthcare; however, their competence in medical tasks remains underexplored. We present a comprehensive evaluation of open-source general-purpose and medically specialised VLMs, ranging from 3B to 72B parameters, across eight benchmarks: MedXpert, OmniMedVQA, PMC-VQA, PathVQA, MMMU, SLAKE, and VQA-RAD. To observe model performance across different aspects, we first separate it into understanding and reasoning components. Three salient findings emerge. First, large general-purpose models already match or surpass medical-specific counterparts on several benchmarks, demonstrating strong zero-shot transfer from natural to medical images. Second, reasoning performance is consistently lower than understanding, highlighting a critical barrier to safe decision support. Third, performance varies widely across benchmarks, reflecting differences in task design, annotation quality, and knowledge demands. No model yet reaches the reliability threshold for clinical deployment, underscoring the need for stronger multimodal alignment and more rigorous, fine-grained evaluation protocols.

Paperid: 2630, https://arxiv.org/pdf/2506.22283.pdf

Abstract:
Large Vision-Language Models (LVLMs) encode visual inputs as dense sequences of patch-level tokens to capture fine-grained semantics. These visual tokens often outnumber their textual counterparts by a large margin, leading to substantial computational overhead and limiting the scalability of LVLMs in practice. Previous efforts have explored visual token reduction either prior to or within the large language models (LLMs). However, most in-LLM reduction approaches rely on text-conditioned interactions, implicitly assuming that textual tokens can reliably capture the importance of visual tokens. In this work, we revisit this assumption and reveal causal, semantic, and spatial forms of cross-modal misalignment. These misalignments undermine the effectiveness of text-guided visual token reduction. To address this, we introduce VisionDrop, a training-free, visual-only pruning framework that selects informative visual tokens based on intra-modal (visual-to-visual) attention, without relying on textual signals. To further suppress redundancy throughout the model hierarchy, we treat the visual encoder and the LLM as a unified system and design a progressive pruning pipeline. Our method performs dominant token selection and lightweight contextual merging at multiple stages, enabling fine-grained visual information to be retained even under aggressive token budgets. Extensive experiments across diverse benchmarks show that VisionDrop achieves consistent improvements over existing approaches, despite requiring no additional training or complex modifications. Notably, when integrated with LLaVA-NeXT-7B, VisionDrop achieves a 2.7x reduction in inference latency and 6x in FLOPs, while retaining 95.71% of the original performance.

Paperid: 2631, https://arxiv.org/pdf/2506.21356.pdf

Abstract:
Cinematography, the fundamental visual language of film, is essential for conveying narrative, emotion, and aesthetic quality. While recent Vision-Language Models (VLMs) demonstrate strong general visual understanding, their proficiency in comprehending the nuanced cinematic grammar embedded within individual shots remains largely unexplored and lacks robust evaluation. This critical gap limits both fine-grained visual comprehension and the precision of AI-assisted video generation. To address this, we introduce ShotBench, a comprehensive benchmark specifically designed for cinematic language understanding. It features over 3.5k expert-annotated QA pairs from images and video clips, meticulously curated from over 200 acclaimed (predominantly Oscar-nominated) films and spanning eight key cinematography dimensions. Our evaluation of 24 leading VLMs on ShotBench reveals their substantial limitations: even the top-performing model achieves less than 60% average accuracy, particularly struggling with fine-grained visual cues and complex spatial reasoning. To catalyze advancement in this domain, we construct ShotQA, a large-scale multimodal dataset comprising approximately 70k cinematic QA pairs. Leveraging ShotQA, we develop ShotVL through supervised fine-tuning and Group Relative Policy Optimization. ShotVL significantly outperforms all existing open-source and proprietary models on ShotBench, establishing new state-of-the-art performance. We open-source our models, data, and code to foster rapid progress in this crucial area of AI-driven cinematic understanding and generation.

Paperid: 2632, https://arxiv.org/pdf/2506.10575.pdf

Abstract:
Benefited from image-text contrastive learning, pre-trained vision-language models, e.g., CLIP, allow to direct leverage texts as images (TaI) for parameter-efficient fine-tuning (PEFT). While CLIP is capable of making image features to be similar to the corresponding text features, the modality gap remains a nontrivial issue and limits image recognition performance of TaI. Using multi-label image recognition (MLR) as an example, we present a novel method, called T2I-PAL to tackle the modality gap issue when using only text captions for PEFT. The core design of T2I-PAL is to leverage pre-trained text-to-image generation models to generate photo-realistic and diverse images from text captions, thereby reducing the modality gap. To further enhance MLR, T2I-PAL incorporates a class-wise heatmap and learnable prototypes. This aggregates local similarities, making the representation of local visual features more robust and informative for multi-label recognition. For better PEFT, we further combine both prompt tuning and adapter learning to enhance classification performance. T2I-PAL offers significant advantages: it eliminates the need for fully semantically annotated training images, thereby reducing the manual annotation workload, and it preserves the intrinsic mode of the CLIP model, allowing for seamless integration with any existing CLIP framework. Extensive experiments on multiple benchmarks, including MS-COCO, VOC2007, and NUS-WIDE, show that our T2I-PAL can boost recognition performance by 3.47% in average above the top-ranked state-of-the-art methods.

Paperid: 2633, https://arxiv.org/pdf/2505.23024.pdf

Abstract:
The Vision Language Model (VLM) excels in aligning vision and language representations, and prompt learning has emerged as a key technique for adapting such models to downstream tasks. However, the application of prompt learning with VLM in federated learning (FL) scenarios remains underexplored. This paper systematically investigates the behavioral differences between language prompt learning (LPT) and vision prompt learning (VPT) under data heterogeneity challenges, including label skew and domain shift. We conduct extensive experiments to evaluate the impact of various FL and prompt configurations, such as client scale, aggregation strategies, and prompt length, to assess the robustness of Federated Prompt Learning (FPL). Furthermore, we explore strategies for enhancing prompt learning in complex scenarios where label skew and domain shift coexist, including leveraging both prompt types when computational resources allow. Our findings offer practical insights into optimizing prompt learning in federated settings, contributing to the broader deployment of VLMs in privacy-preserving environments.

Paperid: 2634, https://arxiv.org/pdf/2505.20279.pdf

Abstract:
The rapid advancement of Large Multimodal Models (LMMs) for 2D images and videos has motivated extending these models to understand 3D scenes, aiming for human-like visual-spatial intelligence. Nevertheless, achieving deep spatial understanding comparable to human capabilities poses significant challenges in model encoding and data acquisition. Existing methods frequently depend on external depth sensors for geometry capture or utilize off-the-shelf algorithms for pre-constructing 3D maps, thereby limiting their scalability, especially with prevalent monocular video inputs and for time-sensitive applications. In this work, we introduce VLM-3R, a unified framework for Vision-Language Models (VLMs) that incorporates 3D Reconstructive instruction tuning. VLM-3R processes monocular video frames by employing a geometry encoder to derive implicit 3D tokens that represent spatial understanding. Leveraging our Spatial-Visual-View Fusion and over 200K curated 3D reconstructive instruction tuning question-answer (QA) pairs, VLM-3R effectively aligns real-world spatial context with language instructions. This enables monocular 3D spatial assistance and embodied reasoning. To facilitate the evaluation of temporal reasoning, we introduce the Vision-Spatial-Temporal Intelligence benchmark, featuring over 138.6K QA pairs across five distinct tasks focused on evolving spatial relationships. Extensive experiments demonstrate that our model, VLM-3R, not only facilitates robust visual-spatial reasoning but also enables the understanding of temporal 3D context changes, excelling in both accuracy and scalability.

Paperid: 2635, https://arxiv.org/pdf/2505.18660.pdf

Abstract:
Recent advances in AI-powered generative models have enabled the creation of increasingly realistic synthetic images, posing significant risks to information integrity and public trust on social media platforms. While robust detection frameworks and diverse, large-scale datasets are essential to mitigate these risks, existing academic efforts remain limited in scope: current datasets lack the diversity, scale, and realism required for social media contexts, while detection methods struggle with generalization to unseen generative technologies. To bridge this gap, we introduce So-Fake-Set, a comprehensive social media-oriented dataset with over 2 million high-quality images, diverse generative sources, and photorealistic imagery synthesized using 35 state-of-the-art generative models. To rigorously evaluate cross-domain robustness, we establish a novel and large-scale (100K) out-of-domain benchmark (So-Fake-OOD) featuring synthetic imagery from commercial models explicitly excluded from the training distribution, creating a realistic testbed for evaluating real-world performance. Leveraging these resources, we present So-Fake-R1, an advanced vision-language framework that employs reinforcement learning for highly accurate forgery detection, precise localization, and explainable inference through interpretable visual rationales. Extensive experiments show that So-Fake-R1 outperforms the second-best method, with a 1.3% gain in detection accuracy and a 4.5% increase in localization IoU. By integrating a scalable dataset, a challenging OOD benchmark, and an advanced detection framework, this work establishes a new foundation for social media-centric forgery detection research. The code, models, and datasets will be released publicly.

Paperid: 2636, https://arxiv.org/pdf/2505.17132.pdf

Abstract:
Large vision-language models (VLMs) are highly vulnerable to jailbreak attacks that exploit visual-textual interactions to bypass safety guardrails. In this paper, we present DTR, a novel inference-time defense that mitigates multimodal jailbreak attacks through optimizing the model's key-value (KV) caches. Rather than relying on curated safety-specific data or costly image-to-text conversion, we introduce a new formulation of the safety-relevant distributional shift induced by the visual modality. This formulation enables DTR to dynamically adjust visual token weights, minimizing the impact of adversarial visual inputs while preserving the model's general capabilities and inference efficiency. Extensive evaluation across diverse VLMs and attack benchmarks demonstrates that \sys outperforms existing defenses in both attack robustness and benign task performance, marking the first successful application of KV cache optimization for safety enhancement in multimodal foundation models. (warning: this paper contains potentially harmful content generated by VLMs.)

Paperid: 2637, https://arxiv.org/pdf/2505.09372.pdf

Abstract:
Dermatological diagnosis represents a complex multimodal challenge that requires integrating visual features with specialized clinical knowledge. While vision-language pretraining (VLP) has advanced medical AI, its effectiveness in dermatology is limited by text length constraints and the lack of structured texts. In this paper, we introduce MAKE, a Multi-Aspect Knowledge-Enhanced vision-language pretraining framework for zero-shot dermatological tasks. Recognizing that comprehensive dermatological descriptions require multiple knowledge aspects that exceed standard text constraints, our framework introduces: (1) a multi-aspect contrastive learning strategy that decomposes clinical narratives into knowledge-enhanced sub-texts through large language models, (2) a fine-grained alignment mechanism that connects subcaptions with diagnostically relevant image features, and (3) a diagnosis-guided weighting scheme that adaptively prioritizes different sub-captions based on clinical significance prior. Through pretraining on 403,563 dermatological image-text pairs collected from education resources, MAKE significantly outperforms state-of-the-art VLP models on eight datasets across zero-shot skin disease classification, concept annotation, and cross-modal retrieval tasks. Our code will be made publicly available at https: //github.com/SiyuanYan1/MAKE.

Paperid: 2638, https://arxiv.org/pdf/2504.13460.pdf

Abstract:
Traditional temporal action localization (TAL) methods rely on large amounts of detailed annotated data, whereas few-shot TAL reduces this dependence by using only a few training samples to identify unseen action categories. However, existing few-shot TAL methods typically focus solely on video-level information, neglecting textual information, which can provide valuable semantic support for the localization task. Therefore, we propose a new few-shot temporal action localization method by Chain-of-Thought textual reasoning to improve localization performance. Specifically, we design a novel few-shot learning framework that leverages textual semantic information to enhance the model's ability to capture action commonalities and variations, which includes a semantic-aware text-visual alignment module designed to align the query and support videos at different levels. Meanwhile, to better express the temporal dependencies and causal relationships between actions at the textual level to assist action localization, we design a Chain of Thought (CoT)-like reasoning method that progressively guides the Vision Language Model (VLM) and Large Language Model (LLM) to generate CoT-like text descriptions for videos. The generated texts can capture more variance of action than visual features. We conduct extensive experiments on the publicly available ActivityNet1.3 and THUMOS14 datasets. We introduce the first dataset named Human-related Anomaly Localization and explore the application of the TAL task in human anomaly detection. The experimental results demonstrate that our proposed method significantly outperforms existing methods in single-instance and multi-instance scenarios. We will release our code, data and benchmark.

Paperid: 2639, https://arxiv.org/pdf/2502.19634.pdf

Abstract:
Reasoning is a critical frontier for advancing medical image analysis, where transparency and trustworthiness play a central role in both clinician trust and regulatory approval. Although Medical Visual Language Models (VLMs) show promise for radiological tasks, most existing VLMs merely produce final answers without revealing the underlying reasoning. To address this gap, we introduce MedVLM-R1, a medical VLM that explicitly generates natural language reasoning to enhance transparency and trustworthiness. Instead of relying on supervised fine-tuning (SFT), which often suffers from overfitting to training distributions and fails to foster genuine reasoning, MedVLM-R1 employs a reinforcement learning framework that incentivizes the model to discover human-interpretable reasoning paths without using any reasoning references. Despite limited training data (600 visual question answering samples) and model parameters (2B), MedVLM-R1 boosts accuracy from 55.11% to 78.22% across MRI, CT, and X-ray benchmarks, outperforming larger models trained on over a million samples. It also demonstrates robust domain generalization under out-of-distribution tasks. By unifying medical image analysis with explicit reasoning, MedVLM-R1 marks a pivotal step toward trustworthy and interpretable AI in clinical practice. Inference model is available at: https://huggingface.co/JZPeterPan/MedVLM-R1.

Paperid: 2640, https://arxiv.org/pdf/2502.11891.pdf

Abstract:
Open-vocabulary semantic segmentation enables models to identify novel object categories beyond their training data. While this flexibility represents a significant advancement, current approaches still rely on manually specified class names as input, creating an inherent bottleneck in real-world applications. This work proposes a Vocabulary-Free Semantic Segmentation pipeline, eliminating the need for predefined class vocabularies. Specifically, we address the chicken-and-egg problem where users need knowledge of all potential objects within a scene to identify them, yet the purpose of segmentation is often to discover these objects. The proposed approach leverages Vision-Language Models to automatically recognize objects and generate appropriate class names, aiming to solve the challenge of class specification and naming quality. Through extensive experiments on several public datasets, we highlight the crucial role of the text encoder in model performance, particularly when the image text classes are paired with generated descriptions. Despite the challenges introduced by the sensitivity of the segmentation text encoder to false negatives within the class tagging process, which adds complexity to the task, we demonstrate that our fully automated pipeline significantly enhances vocabulary-free segmentation accuracy across diverse real-world scenarios.

Paperid: 2641, https://arxiv.org/pdf/2502.10486.pdf

Abstract:
The emergence of vision language models (VLMs) comes with increased safety concerns, as the incorporation of multiple modalities heightens vulnerability to attacks. Although VLMs can be built upon LLMs that have textual safety alignment, it is easily undermined when the vision modality is integrated. We attribute this safety challenge to the modality gap, a separation of image and text in the shared representation space, which blurs the distinction between harmful and harmless queries that is evident in LLMs but weakened in VLMs. To avoid safety decay and fulfill the safety alignment gap, we propose VLM-Guard, an inference-time intervention strategy that leverages the LLM component of a VLM as supervision for the safety alignment of the VLM. VLM-Guard projects the representations of VLM into the subspace that is orthogonal to the safety steering direction that is extracted from the safety-aligned LLM. Experimental results on three malicious instruction settings show the effectiveness of VLM-Guard in safeguarding VLM and fulfilling the safety alignment gap between VLM and its LLM component.

Paperid: 2642, https://arxiv.org/pdf/2502.00843.pdf

Abstract:
In this paper, we propose a novel approach for solving the Visual Question Answering (VQA) task in autonomous driving by integrating Vision-Language Models (VLMs) with continual learning. In autonomous driving, VQA plays a vital role in enabling the system to understand and reason about its surroundings. However, traditional models often struggle with catastrophic forgetting when sequentially exposed to new driving tasks, such as perception, prediction, and planning, each requiring different forms of knowledge. To address this challenge, we present a novel continual learning framework that combines VLMs with selective memory replay and knowledge distillation, reinforced by task-specific projection layer regularization. The knowledge distillation allows a previously trained model to act as a "teacher" to guide the model through subsequent tasks, minimizing forgetting. Meanwhile, task-specific projection layers calculate the loss based on the divergence of feature representations, ensuring continuity in learning and reducing the shift between tasks. Evaluated on the DriveLM dataset, our framework shows substantial performance improvements, with gains ranging from 21.40% to 32.28% across various metrics. These results highlight the effectiveness of combining continual learning with VLMs in enhancing the resilience and reliability of VQA systems in autonomous driving. We will release our source code.

Paperid: 2643, https://arxiv.org/pdf/2502.00618.pdf

Abstract:
Continual learning of vision-language models (VLMs) focuses on leveraging cross-modal pretrained knowledge to incrementally adapt to expanding downstream tasks and datasets, while tackling the challenge of knowledge forgetting. Existing research often focuses on connecting visual features with specific class text in downstream tasks, overlooking the latent relationships between general and specialized knowledge. Our findings reveal that forcing models to optimize inappropriate visual-text matches exacerbates forgetting of VLM's recognition ability. To tackle this issue, we propose DesCLIP, which leverages general attribute (GA) descriptions to guide the understanding of specific class objects, enabling VLMs to establish robust vision-GA-class trilateral associations rather than relying solely on vision-class connections. Specifically, we introduce a language assistant to generate concrete GA description candidates via proper request prompts. Then, an anchor-based embedding filter is designed to obtain highly relevant GA description embeddings, which are leveraged as the paired text embeddings for visual-textual instance matching, thereby tuning the visual encoder. Correspondingly, the class text embeddings are gradually calibrated to align with these shared GA description embeddings. Extensive experiments demonstrate the advancements and efficacy of our proposed method, with comprehensive empirical evaluations highlighting its superior performance in VLM-based recognition compared to existing continual learning methods.

Paperid: 2644, https://arxiv.org/pdf/2411.03554.pdf

Abstract:
Machine unlearning has emerged as an effective strategy for forgetting specific information in the training data. However, with the increasing integration of visual data, privacy concerns in Vision Language Models (VLMs) remain underexplored. To address this, we introduce Facial Identity Unlearning Benchmark (FIUBench), a novel VLM unlearning benchmark designed to robustly evaluate the effectiveness of unlearning algorithms under the Right to be Forgotten setting. Specifically, we formulate the VLM unlearning task via constructing the Fictitious Facial Identity VQA dataset and apply a two-stage evaluation pipeline that is designed to precisely control the sources of information and their exposure levels. In terms of evaluation, since VLM supports various forms of ways to ask questions with the same semantic meaning, we also provide robust evaluation metrics including membership inference attacks and carefully designed adversarial privacy attacks to evaluate the performance of algorithms. Through the evaluation of four baseline VLM unlearning algorithms within FIUBench, we find that all methods remain limited in their unlearning performance, with significant trade-offs between model utility and forget quality. Furthermore, our findings also highlight the importance of privacy attacks for robust evaluations. We hope FIUBench will drive progress in developing more effective VLM unlearning algorithms.

Paperid: 2645, https://arxiv.org/pdf/2410.15778.pdf

Abstract:
Hallucination poses a challenge to the deployment of large vision-language models (LVLMs) in applications. Unlike in large language models (LLMs), hallucination in LVLMs often arises from misalignments between visual inputs and textual outputs. This paper investigates the underlying mechanisms of hallucination, focusing on the unique structure of LVLMs that distinguishes them from large language models (LLMs). We identify that hallucinations often arise from the sensitivity of text decoders to vision inputs, a natural phenomenon when image encoders and text decoders are pre-trained separately. Inspired by this, we introduce Visual and Textual Intervention (VTI), a novel technique designed to reduce hallucinations by steering latent space representations during inference to enhance the stability of vision features. As a task-agnostic test-time intervention, VTI can be easily applied to any problem without additional cost. Extensive experiments demonstrate that it can effectively reduce hallucinations and outperform baseline methods across multiple metrics, highlighting the critical role of vision feature stability in LVLMs.

Paperid: 2646, https://arxiv.org/pdf/2601.12263.pdf

Abstract:
Vision-Language Models (VLMs) are rapidly replacing unimodal encoders in modern retrieval and recommendation systems. While their capabilities are well-documented, their robustness against adversarial manipulation in competitive ranking scenarios remains largely unexplored. In this paper, we uncover a critical vulnerability in VLM-based product search: multimodal ranking attacks. We present Multimodal Generative Engine Optimization (MGEO), a novel adversarial framework that enables a malicious actor to unfairly promote a target product by jointly optimizing imperceptible image perturbations and fluent textual suffixes. Unlike existing attacks that treat modalities in isolation, MGEO employs an alternating gradient-based optimization strategy to exploit the deep cross-modal coupling within the VLM. Extensive experiments on real-world datasets using state-of-the-art models demonstrate that our coordinated attack significantly outperforms text-only and image-only baselines. These findings reveal that multimodal synergy, typically a strength of VLMs, can be weaponized to compromise the integrity of search rankings without triggering conventional content filters.

Paperid: 2647, https://arxiv.org/pdf/2601.09668.pdf

Authors:Ailin Huang, Chengyuan Yao, Chunrui Han, Fanqi Wan, Hangyu Guo, Haoran Lv, Hongyu Zhou, Jia Wang, Jian Zhou, Jianjian Sun, Jingcheng Hu, Kangheng Lin, Liang Zhao, Mitt Huang, Song Yuan, Wenwen Qu, Xiangfeng Wang, Yanlin Lai, Yingxiu Zhao, Yinmin Zhang, Yukang Shi, Yuyang Chen, Zejia Weng, Ziyang Meng, Ang Li, Aobo Kong, Bo Dong, Changyi Wan, David Wang, Di Qi, Dingming Li, En Yu, Guopeng Li, Haiquan Yin, Han Zhou, Hanshan Zhang, Haolong Yan, Hebin Zhou, Hongbo Peng, Jiaran Zhang, Jiashu Lv, Jiayi Fu, Jie Cheng, Jie Zhou, Jisheng Yin, Jingjing Xie, Jingwei Wu, Jun Zhang, Junfeng Liu, Kaijun Tan, Kaiwen Yan, Liangyu Chen, Lina Chen, Mingliang Li, Qian Zhao, Quan Sun, Shaoliang Pang, Shengjie Fan, Shijie Shang, Siyuan Zhang, Tianhao You, Wei Ji, Wuxun Xie, Xiaobo Yang, Xiaojie Hou, Xiaoran Jiao, Xiaoxiao Ren, Xiangwen Kong, Xin Huang, Xin Wu, Xing Chen, Xinran Wang, Xuelin Zhang, Yana Wei, Yang Li, Yanming Xu, Yeqing Shen, Yuang Peng, Yue Peng, Yu Zhou, Yusheng Li, Yuxiang Yang, Yuyang Zhang, Zhe Xie, Zhewei Huang, Zhenyi Lu, Zhimin Fan, Zihui Cheng, Daxin Jiang, Qi Han, Xiangyu Zhang, Yibo Zhu, Zheng Ge

Abstract:
We present STEP3-VL-10B, a lightweight open-source foundation model designed to redefine the trade-off between compact efficiency and frontier-level multimodal intelligence. STEP3-VL-10B is realized through two strategic shifts: first, a unified, fully unfrozen pre-training strategy on 1.2T multimodal tokens that integrates a language-aligned Perception Encoder with a Qwen3-8B decoder to establish intrinsic vision-language synergy; and second, a scaled post-training pipeline featuring over 1k iterations of reinforcement learning. Crucially, we implement Parallel Coordinated Reasoning (PaCoRe) to scale test-time compute, allocating resources to scalable perceptual reasoning that explores and synthesizes diverse visual hypotheses. Consequently, despite its compact 10B footprint, STEP3-VL-10B rivals or surpasses models 10$\times$-20$\times$ larger (e.g., GLM-4.6V-106B, Qwen3-VL-235B) and top-tier proprietary flagships like Gemini 2.5 Pro and Seed-1.5-VL. Delivering best-in-class performance, it records 92.2% on MMBench and 80.11% on MMMU, while excelling in complex reasoning with 94.43% on AIME2025 and 75.95% on MathVision. We release the full model suite to provide the community with a powerful, efficient, and reproducible baseline.

Paperid: 2648, https://arxiv.org/pdf/2601.08617.pdf

Abstract:
With the increasing adoption of vision-language models (VLMs) in critical decision-making systems such as healthcare or autonomous driving, the calibration of their uncertainty estimates becomes paramount. Yet, this dimension has been largely underexplored in the VLM test-time prompt-tuning (TPT) literature, which has predominantly focused on improving their discriminative performance. Recent state-of-the-art advocates for enforcing full orthogonality over pairs of text prompt embeddings to enhance separability, and therefore calibration. Nevertheless, as we theoretically show in this work, the inherent gradients from fully orthogonal constraints will strongly push semantically related classes away, ultimately making the model overconfident. Based on our findings, we propose Semantic Orthogonal Calibration (SoC), a Huber-based regularizer that enforces smooth prototype separation while preserving semantic proximity, thereby improving calibration compared to prior orthogonality-based approaches. Across a comprehensive empirical validation, we demonstrate that SoC consistently improves calibration performance, while also maintaining competitive discriminative capabilities.

Paperid: 2649, https://arxiv.org/pdf/2601.04153.pdf

Abstract:
Direct Preference Optimization (DPO) has recently improved Text-to-Video (T2V) generation by enhancing visual fidelity and text alignment. However, current methods rely on non-differentiable preference signals from human annotations or learned reward models. This reliance makes training label-intensive, bias-prone, and easy-to-game, which often triggers reward hacking and unstable training. We propose Diffusion-DRF, a differentiable reward flow for fine-tuning video diffusion models using a frozen, off-the-shelf Vision-Language Model (VLM) as a training-free critic. Diffusion-DRF directly backpropagates VLM feedback through the diffusion denoising chain, converting logit-level responses into token-aware gradients for optimization. We propose an automated, aspect-structured prompting pipeline to obtain reliable multi-dimensional VLM feedback, while gradient checkpointing enables efficient updates through the final denoising steps. Diffusion-DRF improves video quality and semantic alignment while mitigating reward hacking and collapse -- without additional reward models or preference datasets. It is model-agnostic and readily generalizes to other diffusion-based generative tasks.

Paperid: 2650, https://arxiv.org/pdf/2512.02814.pdf

Abstract:
Radiology reporting is an essential yet time-consuming and error-prone task for radiologists in clinical examinations, especially for volumetric medical images. Rigorous quality control is also critical but tedious, ensuring that the final report meets clinical standards. Existing automated approaches, including radiology report generation methods and medical vision-language models, focus mainly on the report generation phase and neglect the crucial quality control procedure, limiting their capability to provide comprehensive support to radiologists. We propose Radiologist Copilot, an agentic AI assistant equipped with orchestrated tools designed for automated radiology reporting with quality control. Leveraging large language models as the reasoning backbone, the agentic system autonomously selects tools, plans, and executes actions, emulating the behavior of radiologists throughout the holistic radiology reporting process. The orchestrated tools include region localization, think with image paradigm directed region analysis planning, strategic template selection for report generation, quality assessment and feedback-driven adaptive refinement for quality control. Therefore, Radiologist Copilot facilitates accurate, complete, and efficient radiology reporting, assisting radiologists and improving clinical efficiency. Experimental results demonstrate that Radiologist Copilot significantly surpasses other state-of-the-art methods in radiology reporting. The source code will be released upon acceptance.

Paperid: 2651, https://arxiv.org/pdf/2512.02423.pdf

Abstract:
With the rapid development of Large Vision Language Models, the focus of Graphical User Interface (GUI) agent tasks shifts from single-screen tasks to complex screen navigation challenges. However, real-world GUI environments, such as PC software and mobile Apps, are often complex and proprietary, making it difficult to obtain the comprehensive environment information needed for agent training and evaluation. This limitation hinders systematic investigation and benchmarking of agent navigation capabilities. To address this limitation, we introduce GUI Exploration Lab, a simulation environment engine for GUI agent navigation research that enables flexible definition and composition of screens, icons, and navigation graphs, while providing full access to environment information for comprehensive agent training and evaluation. Through extensive experiments, we find that supervised fine-tuning enables effective memorization of fundamental knowledge, serving as a crucial foundation for subsequent training. Building on this, single-turn reinforcement learning further enhances generalization to unseen scenarios. Finally, multi-turn reinforcement learning encourages the development of exploration strategies through interactive trial and error, leading to further improvements in screen navigation performance. We validate our methods on both static and interactive benchmarks, demonstrating that our findings generalize effectively to real-world scenarios. These findings demonstrate the advantages of reinforcement learning approaches in GUI navigation and offer practical guidance for building more capable and generalizable GUI agents.

Paperid: 2652, https://arxiv.org/pdf/2511.23262.pdf

Abstract:
Recent Vision-Language Models (VLMs) exhibit strong perceptual reasoning abilities, yet they often struggle to adapt efficiently when encountering novel tasks at test time. In contrast, humans leverage the metacognitive model with memory, enabling continuous strategy refinement through metacognitive control when faced with new challenges. To bridge this gap, we propose metacognitive test-time reasoning (MCTR), a framework that equips models with the ability to learn, adapt, and improve during test time through metacognitive self-updating. Inspired by the dual structure of human metacognition, MCTR comprises meta-level and object-level VLM reasoning modules, each equipped with dedicated memory systems for hierarchical adaptive reasoning. Specifically, MCTR consists of (1) a meta-reasoning module which incrementally builds a structured memory by discovering and storing task-relevant rules, environmental patterns, and action-outcome relationships from test-time observations as natural language descriptions; and (2) an action-reasoning module that determines optimal actions through context-aware perception and strategic reasoning by dynamically retrieving and integrating knowledge from memory. The action-reasoning module continuously updates its policy through proposed metacognitive test-time reinforcement learning, adapting as knowledge memory evolves. We evaluate MCTR on 45 Atari games (33 seen, 12 unseen). MCTR demonstrates robust test-time adaptation, achieving 9/12 top-1 results on unseen games compared with baselines. Analyses through ablations, learning dynamics, and case studies reveal the complementary contributions of both components and show meta-reasoning evolving toward human-like adaptation strategies.

Paperid: 2653, https://arxiv.org/pdf/2511.00396.pdf

Abstract:
We present the first unified framework that jointly handles three operationally heterogeneous saliency tasks, eg, SOD, CoSOD, and SIS, by casting each as a Chain-of-Thought (CoT) reasoning process in a Vision-Language Model (VLM) to bridge task heterogeneity. CoT training follows a two-stage paradigm: Supervised Fine-Tuning (SFT) and Reinforcement Learning (RL). To enhance CoT quality in RL, we propose Confidence-Guided Policy Optimization (CGPO), a lightweight single-sample algorithm that leverages the discrepancy between reward and model confidence as a per-sample advantage signal. This design naturally focuses updates on informative responses while eliminating group sampling, thereby addressing GRPO's key limitations: confidence-agnostic learning, signal dilution, and prohibitive computational overhead. We also introduce an "output-to-reasoning" strategy to construct high-fidelity SFT data that ensures logical consistency with ground-truth masks. Experiments show our model matches or outperforms specialized SOTA methods and strong closed-source VLMs across all tasks, especially achieving an S-measure of 0.899 on CoCA for CoSOD, surpassing the prior best by 8.0 percentage points, despite using far less training data.

Paperid: 2654, https://arxiv.org/pdf/2510.24342.pdf

Abstract:
For decades, neuroscientists and computer scientists have pursued a shared ambition: to understand intelligence and build it. Modern artificial neural networks now rival humans in language, perception, and reasoning, yet it is still largely unknown whether these artificial systems organize information as the brain does. Existing brain-AI alignment studies have shown the striking correspondence between the two systems, but such comparisons remain bound to specific inputs and tasks, offering no common ground for comparing how AI models with different kinds of modalities-vision, language, or multimodal-are intrinsically organized. Here we introduce a groundbreaking concept of Brain-like Space: a unified geometric space in which every AI model can be precisely situated and compared by mapping its intrinsic spatial attention topological organization onto canonical human functional brain networks, regardless of input modality, task, or sensory domain. Our extensive analysis of 151 Transformer-based models spanning state-of-the-art large vision models, large language models, and large multimodal models uncovers a continuous arc-shaped geometry within this space, reflecting a gradual increase of brain-likeness; different models exhibit distinct distribution patterns within this geometry associated with different degrees of brain-likeness, shaped not merely by their modality but by whether the pretraining paradigm emphasizes global semantic abstraction and whether the positional encoding scheme facilitates deep fusion across different modalities. Moreover, the degree of brain-likeness for a model and its downstream task performance are not "identical twins". The Brain-like Space provides the first unified framework for situating, quantifying, and comparing intelligence across domains, revealing the deep organizational principles that bridge machines and the brain.

Paperid: 2655, https://arxiv.org/pdf/2510.19802.pdf

Abstract:
Vision-Language Models (VLMs) demonstrate impressive zero-shot generalization through large-scale image-text pretraining, yet their performance can drop once the deployment distribution diverges from the training distribution. To address this, Test-Time Adaptation (TTA) methods update models using unlabeled target data. However, existing approaches often ignore two key challenges: prototype degradation in long-tailed distributions and confusion between semantically similar classes. To tackle these issues, we propose \textbf{C}lass-Aware \textbf{P}rototype \textbf{L}earning with \textbf{N}egative \textbf{C}ontrast(\textbf{CPL-NC}), a lightweight TTA framework designed specifically for VLMs to enhance generalization under distribution shifts. CPL-NC introduces a \textit{Class-Aware Prototype Cache} Module that dynamically adjusts per-class capacity based on test-time frequency and activation history, with a rejuvenation mechanism for inactive classes to retain rare-category knowledge. Additionally, a \textit{Negative Contrastive Learning} Mechanism identifies and constrains hard visual-textual negatives to improve class separability. The framework employs asymmetric optimization, refining only textual prototypes while anchoring on stable visual features. Experiments on 15 benchmarks show that CPL-NC consistently outperforms prior TTA methods across both ResNet-50 and ViT-B/16 backbones.

Paperid: 2656, https://arxiv.org/pdf/2510.16973.pdf

Abstract:
Recent advancements in artificial intelligence (AI), particularly foundation models (FMs), have revolutionized medical image analysis, demonstrating strong zero- and few-shot performance across diverse medical imaging tasks, from segmentation to report generation. Unlike traditional task-specific AI models, FMs leverage large corpora of labeled and unlabeled multimodal datasets to learn generalized representations that can be adapted to various downstream clinical applications with minimal fine-tuning. However, despite the rapid proliferation of FM research in medical imaging, the field remains fragmented, lacking a unified synthesis that systematically maps the evolution of architectures, training paradigms, and clinical applications across modalities. To address this gap, this review article provides a comprehensive and structured analysis of FMs in medical image analysis. We systematically categorize studies into vision-only and vision-language FMs based on their architectural foundations, training strategies, and downstream clinical tasks. Additionally, a quantitative meta-analysis of the studies was conducted to characterize temporal trends in dataset utilization and application domains. We also critically discuss persistent challenges, including domain adaptation, efficient fine-tuning, computational constraints, and interpretability along with emerging solutions such as federated learning, knowledge distillation, and advanced prompting. Finally, we identify key future research directions aimed at enhancing the robustness, explainability, and clinical integration of FMs, thereby accelerating their translation into real-world medical practice.

Paperid: 2657, https://arxiv.org/pdf/2510.14388.pdf

Abstract:
Building agents that autonomously operate mobile devices has attracted increasing attention. While Vision-Language Models (VLMs) show promise, most existing approaches rely on direct state-to-action mappings, which lack structured reasoning and planning, and thus generalize poorly to novel tasks or unseen UI layouts. We introduce Hi-Agent, a trainable hierarchical vision-language agent for mobile control, featuring a high-level reasoning model and a low-level action model that are jointly optimized. For efficient training, we reformulate multi-step decision-making as a sequence of single-step subgoals and propose a foresight advantage function, which leverages execution feedback from the low-level model to guide high-level optimization. This design alleviates the path explosion issue encountered by Group Relative Policy Optimization (GRPO) in long-horizon tasks and enables stable, critic-free joint training. Hi-Agent achieves a new State-Of-The-Art (SOTA) 87.9% task success rate on the Android-in-the-Wild (AitW) benchmark, significantly outperforming prior methods across three paradigms: prompt-based (AppAgent: 17.7%), supervised (Filtered BC: 54.5%), and reinforcement learning-based (DigiRL: 71.9%). It also demonstrates competitive zero-shot generalization on the ScreenSpot-v2 benchmark. On the more challenging AndroidWorld benchmark, Hi-Agent also scales effectively with larger backbones, showing strong adaptability in high-complexity mobile control scenarios.

Paperid: 2658, https://arxiv.org/pdf/2509.26251.pdf

Abstract:
Latent Action Models (LAMs) enable Vision-Language-Action (VLA) systems to learn semantic action representations from large-scale unannotated data. Yet, we identify two bottlenecks of LAMs: 1) the commonly adopted end-to-end trained image encoder suffers from poor spatial understanding; 2) LAMs can be fragile when input frames are distant, leading to limited temporal perception. Such factors inevitably hinder stable and clear action modeling. To this end, we propose Farsighted-LAM, a latent action framework with geometry-aware spatial encoding and multi-scale temporal modeling, capturing structural priors and dynamic motion patterns from consecutive frames. We further propose SSM-VLA, an end-to-end VLA framework built upon Farsighted-LAM, which integrates structured perception with a visual Chain-of-Thought module to explicitly reason about environmental dynamics, enhancing decision consistency and interpretability. We validate SSM-VLA on multiple VLA tasks in both simulation and real-world settings, and achieve state-of-the-art performance. Our results demonstrate that our strategy of combining geometry-aware modeling, temporal coherence, and explicit reasoning is effective in enhancing the robustness and generalizability of embodied intelligence.

Paperid: 2659, https://arxiv.org/pdf/2509.24948.pdf

Abstract:
Vision-Language-Action (VLA) models trained via imitation learning suffer from significant performance degradation in data-scarce scenarios due to their reliance on large-scale demonstration datasets. Although reinforcement learning (RL)-based post-training has proven effective in addressing data scarcity, its application to VLA models is hindered by the non-resettable nature of real-world environments. This limitation is particularly critical in high-risk domains such as industrial automation, where interactions often induce state changes that are costly or infeasible to revert. Furthermore, existing VLA approaches lack a reliable mechanism for detecting task completion, leading to redundant actions that reduce overall task success rates. To address these challenges, we propose World-Env, an RL-based post-training framework that replaces physical interaction with a low-cost, world model-based virtual simulator. World-Env consists of two key components: (1) a video-based world simulator that generates temporally consistent future visual observations, and (2) a vision-language model (VLM)-guided instant reflector that provides continuous reward signals and predicts action termination. This simulated environment enables VLA models to safely explore and generalize beyond their initial imitation learning distribution. Our method achieves notable performance gains with as few as five expert demonstrations per task. Experiments on complex robotic manipulation tasks demonstrate that World-Env effectively overcomes the data inefficiency, safety constraints, and inefficient execution of conventional VLA models that rely on real-world interaction, offering a practical and scalable solution for post-training in resource-constrained settings.

Paperid: 2660, https://arxiv.org/pdf/2509.24171.pdf

Abstract:
The growing prevalence of large language models (LLMs) and vision-language models (VLMs) has heightened the need for reliable techniques to determine whether a model has been fine-tuned from or is even identical to another. Existing similarity-based methods often require access to model parameters or produce heuristic scores without principled thresholds, limiting their applicability. We introduce Random Selection Probing (RSP), a hypothesis-testing framework that formulates model correlation detection as a statistical test. RSP optimizes textual or visual prefixes on a reference model for a random selection task and evaluates their transferability to a target model, producing rigorous p-values that quantify evidence of correlation. To mitigate false positives, RSP incorporates an unrelated baseline model to filter out generic, transferable features. We evaluate RSP across both LLMs and VLMs under diverse access conditions for reference models and test models. Experiments on fine-tuned and open-source models show that RSP consistently yields small p-values for related models while maintaining high p-values for unrelated ones. Extensive ablation studies further demonstrate the robustness of RSP. These results establish RSP as the first principled and general statistical framework for model correlation detection, enabling transparent and interpretable decisions in modern machine learning ecosystems.

Paperid: 2661, https://arxiv.org/pdf/2509.01644.pdf

Abstract:
This paper provides a simplification on OpenVision's architecture and loss design for enhancing its training efficiency. Following the prior vision-language pretraining works CapPa and AIMv2, as well as modern multimodal designs like LLaVA, our changes are straightforward: we remove the text encoder (and therefore the contrastive loss), retaining only the captioning loss as a purely generative training signal. We name this new version OpenVision 2. The initial results are promising: despite this simplification, OpenVision 2 competitively matches the original model's performance on a broad set of multimodal benchmarks while substantially cutting both training time and memory consumption. For example, with ViT-L/14, it reduces training time by about 1.5x (from 83h to 57h), and memory usage by about 1.8x (from 24.5GB to 13.8GB, equivalently allowing the maximum batch size to grow from 2k to 8k). This superior training efficiency also allows us to scale far beyond the largest vision encoder used in OpenVision, reaching more than 1 billion parameters. We hold a strong belief that this lightweight, generative-only paradigm is compelling for future vision encoder development in multimodal foundation models.

Paperid: 2662, https://arxiv.org/pdf/2508.06317.pdf

Abstract:
Video Temporal Grounding (TG) aims to temporally locate video segments matching a natural language description (a query) in a long video. While Vision-Language Models (VLMs) are effective at holistic semantic matching, they often struggle with fine-grained temporal localisation. Recently, Group Relative Policy Optimisation (GRPO) reformulates the inference process as a reinforcement learning task, enabling fine-grained grounding and achieving strong in-domain performance. However, GRPO relies on labelled data, making it unsuitable in unlabelled domains. Moreover, because videos are large and expensive to store and process, performing full-scale adaptation introduces prohibitive latency and computational overhead, making it impractical for real-time deployment. To overcome both problems, we introduce a Data-Efficient Unlabelled Cross-domain Temporal Grounding method, from which a model is first trained on a labelled source domain, then adapted to a target domain using only a small number of unlabelled videos from the target domain. This approach eliminates the need for target annotation and keeps both computational and storage overhead low enough to run in real time. Specifically, we introduce. Uncertainty-quantified Rollout Policy Adaptation (URPA) for cross-domain knowledge transfer in learning video temporal grounding without target labels. URPA generates multiple candidate predictions using GRPO rollouts, averages them to form a pseudo label, and estimates confidence from the variance across these rollouts. This confidence then weights the training rewards, guiding the model to focus on reliable supervision. Experiments on three datasets across six cross-domain settings show that URPA generalises well using only a few unlabelled target videos. Codes will be released once published.

Paperid: 2663, https://arxiv.org/pdf/2507.19420.pdf

Abstract:
The processing mechanisms underlying language and image understanding in large vision-language models (LVLMs) have been extensively studied. However, the internal reasoning mechanisms of LVLMs for spatiotemporal understanding remain poorly understood. In this work, we introduce a systematic, circuit-based framework designed to investigate how spatiotemporal visual semantics are represented and processed within these LVLMs. Specifically, our framework comprises three circuits: visual auditing circuit, semantic tracing circuit, and attention flow circuit. Through the lens of these circuits, we discover that visual semantics are highly localized to specific object tokens--removing these tokens can degrade model performance by up to 92.6%. Furthermore, we identify that interpretable concepts of objects and actions emerge and become progressively refined in the middle-to-late layers of LVLMs. In contrary to the current works that solely focus on objects in one image, we reveal that the middle-to-late layers of LVLMs exhibit specialized functional localization for spatiotemporal semantics. Our findings offer significant mechanistic insights into spatiotemporal semantics analysis of LVLMs, laying a foundation for designing more robust and interpretable models.

Paperid: 2664, https://arxiv.org/pdf/2507.07562.pdf

Abstract:
Large vision-language models (VLMs) increasingly adopt post-training techniques such as long chain-of-thought (CoT) supervised fine-tuning (SFT) and reinforcement learning (RL) to elicit sophisticated reasoning. While these methods exhibit synergy in language-only models, their joint effectiveness in VLMs remains uncertain. We present a systematic investigation into the distinct roles and interplay of long-CoT SFT and RL across multiple multimodal reasoning benchmarks. We find that SFT improves performance on difficult questions by in-depth, structured reasoning, but introduces verbosity and degrades performance on simpler ones. In contrast, RL promotes generalization and brevity, yielding consistent improvements across all difficulty levels, though the improvements on the hardest questions are less prominent compared to SFT. Surprisingly, combining them through two-staged, interleaved, or progressive training strategies, as well as data mixing and model merging, all fails to produce additive benefits, instead leading to trade-offs in accuracy, reasoning style, and response length. This ``synergy dilemma'' highlights the need for more seamless and adaptive approaches to unlock the full potential of combined post-training techniques for reasoning VLMs.

Paperid: 2665, https://arxiv.org/pdf/2506.22567.pdf

Abstract:
CLIP models pretrained on natural images with billion-scale image-text pairs have demonstrated impressive capabilities in zero-shot classification, cross-modal retrieval, and open-ended visual answering. However, transferring this success to biomedicine is hindered by the scarcity of large-scale biomedical image-text corpora, the heterogeneity of image modalities, and fragmented data standards across institutions. These limitations hinder the development of a unified and generalizable biomedical foundation model trained from scratch. To overcome this, we introduce MMKD-CLIP, a generalist biomedical foundation model developed via Multiple Medical CLIP Knowledge Distillation. Rather than relying on billion-scale raw data, MMKD-CLIP distills knowledge from nine state-of-the-art domain-specific or generalist biomedical CLIP models, each pretrained on millions of biomedical image-text pairs. Our two-stage training pipeline first performs CLIP-style pretraining on over 2.9 million biomedical image-text pairs from 26 image modalities, followed by feature-level distillation using over 19.2 million feature pairs extracted from teacher models. We evaluate MMKD-CLIP on 58 diverse biomedical datasets, encompassing over 10.8 million biomedical images across nine image modalities. The evaluation spans six core task types: zero-shot classification, linear probing, cross-modal retrieval, visual question answering, survival prediction, and cancer diagnosis. MMKD-CLIP consistently outperforms all teacher models while demonstrating remarkable robustness and generalization across image domains and task settings. These results underscore that multi-teacher knowledge distillation is a scalable and effective paradigm for building high-performing biomedical foundation models under the practical constraints of real-world data availability.

Paperid: 2666, https://arxiv.org/pdf/2506.08022.pdf

Abstract:
The task adaptation and alignment of Large Multimodal Models (LMMs) have been significantly advanced by instruction tuning and further strengthened by recent preference optimization. Yet, most LMMs still suffer from severe modality imbalance during reasoning, i.e., outweighing language prior biases over visual inputs, which bottlenecks their generalization to downstream tasks and causes hallucinations. However, existing preference optimization approaches for LMMs do not focus on restraining the internal biases of their Large Language Model (LLM) backbones when curating the training data. Moreover, they heavily rely on offline data and lack the capacity to explore diverse responses adaptive to dynamic distributional shifts during training. Meanwhile, Group Relative Policy Optimization (GRPO), a recent method using online-generated data and verified rewards to improve reasoning capabilities, remains largely underexplored in LMM alignment. In this paper, we propose a novel preference learning framework, Modality-Balancing Preference Optimization (MBPO), to address the modality imbalance in LMMs. MBPO constructs a more effective offline preference dataset by generating hard negatives, i.e., rejected responses misled by LLM biases due to limited usage of visual information, through adversarial perturbation of input images. Moreover, MBPO leverages the easy-to-verify nature of close-ended tasks to generate online responses with verified rewards. GRPO is then employed to train the model with offline-online hybrid data. Extensive experiments demonstrate that MBPO can enhance LMM performance on challenging vision-language tasks and effectively reduce hallucinations.

Paperid: 2667, https://arxiv.org/pdf/2506.05342.pdf

Abstract:
Recent image segmentation models have advanced to segment images into high-quality masks for visual entities, and yet they cannot provide comprehensive semantic understanding for complex queries based on both language and vision. This limitation reduces their effectiveness in applications that require user-friendly interactions driven by vision-language prompts. To bridge this gap, we introduce a novel task of omnimodal referring expression segmentation (ORES). In this task, a model produces a group of masks based on arbitrary prompts specified by text only or text plus reference visual entities. To address this new challenge, we propose a novel framework to "Refer to Any Segmentation Mask Group" (RAS), which augments segmentation models with complex multimodal interactions and comprehension via a mask-centric large multimodal model. For training and benchmarking ORES models, we create datasets MaskGroups-2M and MaskGroups-HQ to include diverse mask groups specified by text and reference entities. Through extensive evaluation, we demonstrate superior performance of RAS on our new ORES task, as well as classic referring expression segmentation (RES) and generalized referring expression segmentation (GRES) tasks. Project page: https://Ref2Any.github.io.

Paperid: 2668, https://arxiv.org/pdf/2505.16517.pdf

Abstract:
Large Vision-Language Models (LVLMs) have recently advanced robotic manipulation by leveraging vision for scene perception and language for instruction following. However, existing methods rely heavily on costly human-annotated training datasets, which limits their generalization and causes them to struggle in out-of-domain (OOD) scenarios, reducing real-world adaptability. To address these challenges, we propose ManipLVM-R1, a novel reinforcement learning framework that replaces traditional supervision with Reinforcement Learning using Verifiable Rewards (RLVR). By directly optimizing for task-aligned outcomes, our method enhances generalization and physical reasoning while removing the dependence on costly annotations. Specifically, we design two rule-based reward functions targeting key robotic manipulation subtasks: an Affordance Perception Reward to enhance localization of interaction regions, and a Trajectory Match Reward to ensure the physical plausibility of action paths. These rewards provide immediate feedback and impose spatial-logical constraints, encouraging the model to go beyond shallow pattern matching and instead learn deeper, more systematic reasoning about physical interactions.

Paperid: 2669, https://arxiv.org/pdf/2505.14030.pdf

Abstract:
Vision-language-action (VLA) models have shown promise as generalist robotic policies by jointly leveraging visual, linguistic, and proprioceptive modalities to generate action trajectories. While recent benchmarks have advanced VLA research in domestic tasks, professional science-oriented domains remain underexplored. We introduce AutoBio, a simulation framework and benchmark designed to evaluate robotic automation in biology laboratory environments--an application domain that combines structured protocols with demanding precision and multimodal interaction. AutoBio extends existing simulation capabilities through a pipeline for digitizing real-world laboratory instruments, specialized physics plugins for mechanisms ubiquitous in laboratory workflows, and a rendering stack that support dynamic instrument interfaces and transparent materials through physically based rendering. Our benchmark comprises biologically grounded tasks spanning three difficulty levels, enabling standardized evaluation of language-guided robotic manipulation in experimental protocols. We provide infrastructure for demonstration generation and seamless integration with VLA models. Baseline evaluations with two SOTA VLA models reveal significant gaps in precision manipulation, visual reasoning, and instruction following in scientific workflows. By releasing AutoBio, we aim to catalyze research on generalist robotic systems for complex, high-precision, and multimodal professional environments. The simulator and benchmark are publicly available to facilitate reproducible research.

Paperid: 2670, https://arxiv.org/pdf/2505.10860.pdf

Abstract:
Mixture of experts (MoE) methods are a key component in most large language model architectures, including the recent series of DeepSeek models. Compared to other MoE implementations, DeepSeekMoE stands out because of two unique features: the deployment of a shared expert strategy and of the normalized sigmoid gating mechanism. Despite the prominent role of DeepSeekMoE in the success of the DeepSeek series of models, there have been only a few attempts to justify theoretically the value of the shared expert strategy, while its normalized sigmoid gating has remained unexplored. To bridge this gap, we undertake a comprehensive theoretical study of these two features of DeepSeekMoE from a statistical perspective. We perform a convergence analysis of the expert estimation task to highlight the gains in sample efficiency for both the shared expert strategy and the normalized sigmoid gating, offering useful insights into the design of expert and gating structures. To verify empirically our theoretical findings, we carry out several experiments on both synthetic data and real-world datasets for (vision) language modeling tasks. Finally, we conduct an extensive empirical analysis of the router behaviors, ranging from router saturation, router change rate, to expert utilization.

Paperid: 2671, https://arxiv.org/pdf/2504.07165.pdf

Abstract:
We present a perception in reflection paradigm designed to transcend the limitations of current large vision-language models (LVLMs), which are expected yet often fail to achieve perfect perception initially. Specifically, we propose Reflective Perception (RePer), a dual-model reflection mechanism that systematically alternates between policy and critic models, enables iterative refinement of visual perception. This framework is powered by Reflective Perceptual Learning (RPL), which reinforces intrinsic reflective capabilities through a methodically constructed visual reflection dataset and reflective unlikelihood training. Comprehensive experimental evaluation demonstrates RePer's quantifiable improvements in image understanding, captioning precision, and hallucination reduction. Notably, RePer achieves strong alignment between model attention patterns and human visual focus, while RPL optimizes fine-grained and free-form preference alignment. These advancements establish perception in reflection as a robust paradigm for future multimodal agents, particularly in tasks requiring complex reasoning and multi-step manipulation.

Paperid: 2672, https://arxiv.org/pdf/2502.07949.pdf

Abstract:
State-of-the-art (SOTA) reinforcement learning (RL) methods have enabled vision-language model (VLM) agents to learn from interaction with online environments without human supervision. However, these methods often struggle with learning inefficiencies when applied to complex, real-world decision-making tasks with sparse rewards and long-horizon dependencies. We propose a novel framework, Variational Subgoal-Conditioned Reinforcement Learning (VSC-RL), advancing the VLM agents in resolving challenging decision-making tasks. Fundamentally distinct from existing methods, VSC-RL reformulates the decision-making problem as a variational subgoal-conditioned RL problem with the newly derived optimization objective, Subgoal Evidence Lower BOund (SGC-ELBO), which comprises two key components: (a) maximizing the subgoal-conditioned return, and (b) minimizing the divergence from a reference goal-conditioned policy. We theoretically and empirically demonstrate that the VSC-RL can efficiently improve the learning efficiency without compromising performance guarantees. Across a diverse set of challenging benchmarks, including mobile device and web control tasks, VSC-RL consistently outperforms existing SOTA methods, achieving superior learning efficiency and performance.

Paperid: 2673, https://arxiv.org/pdf/2502.06395.pdf

Abstract:
The utilisation of foundation models as smartphone assistants, termed app agents, is a critical research challenge. These agents aim to execute human instructions on smartphones by interpreting textual instructions and performing actions via the device's interface. While promising, current approaches face significant limitations. Methods that use large proprietary models, such as GPT-4o, are computationally expensive, while those that use smaller fine-tuned models often lack adaptability to out-of-distribution tasks. In this work, we introduce AppVLM, a lightweight Vision-Language Model (VLM). First, we fine-tune it offline on the AndroidControl dataset. Then, we refine its policy by collecting data from the AndroidWorld environment and performing further training iterations. Our results indicate that AppVLM achieves the highest action prediction accuracy in offline evaluation on the AndroidControl dataset, compared to all evaluated baselines, and matches GPT-4o in online task completion success rate in the AndroidWorld environment, while being up to ten times faster. This makes AppVLM a practical and efficient solution for real-world deployment.

Paperid: 2674, https://arxiv.org/pdf/2412.20631.pdf

Abstract:
Recently, "visual o1" began to enter people's vision, with expectations that this slow-thinking design can solve visual reasoning tasks, especially geometric math problems. However, the reality is that current LVLMs (Large Vision Language Models) can hardly even accurately copy a geometric figure, let alone truly understand the complex inherent logic and spatial relationships within geometric shapes. We believe accurate copying (strong perception) is the first step to visual o1. Accordingly, we introduce the concept of "slow perception" (SP), which guides the model to gradually perceive basic point-line combinations, as our humans, reconstruct complex geometric structures progressively. There are two-fold stages in SP: a) perception decomposition. Perception is not instantaneous. In this stage, complex geometric figures are broken down into basic simple units to unify geometry representation. b) perception flow, which acknowledges that accurately tracing a line is not an easy task. This stage aims to avoid "long visual jumps" in regressing line segments by using a proposed "perceptual ruler" to trace each line stroke-by-stroke. Surprisingly, such a human-like perception manner enjoys an inference time scaling law -- the slower, the better. Researchers strive to speed up the model's perception in the past, but we slow it down again, allowing the model to read the image step-by-step and carefully.

Paperid: 2675, https://arxiv.org/pdf/2412.12722.pdf

Abstract:
Recent studies have raised significant concerns regarding the vulnerability of Large Vision Language Models (LVLMs) to maliciously injected or perturbed input images, which can mislead their responses. Existing defense methods show that such vision attacks are sensitive to image modifications especially cropping, using majority voting across responses of modified images as corrected responses. However, these modifications often result in partial images and distort the semantics, which reduces response quality on clean images after voting. Instead of directly using responses from partial images for voting, we investigate using them to supervise the LVLM's responses to the original images. We propose a black-box, training-free method called DPS (Defense through Partial-Perception Supervision). In this approach, the model is prompted using the responses generated by a model that perceives only a partial image. With DPS, the model can adjust its response based on partial image understanding when under attack, while confidently maintaining its original response for clean input. Our findings show that the weak model can supervise the strong model: when faced with an attacked input, the strong model becomes less confident and adjusts its response based on the weak model's partial understanding, effectively defending against the attack. With clean input, it confidently maintains its original response. Empirical experiments show our method outperforms the baseline, cutting the average attack success rate by 76.3% across six datasets on three popular models.

Paperid: 2676, https://arxiv.org/pdf/2412.10817.pdf

Abstract:
With the advancement of pre-trained vision-language (VL) models, enhancing the alignment between visual and linguistic modalities in downstream tasks has emerged as a critical challenge. Different from existing fine-tuning methods that add extra modules to these two modalities, we investigate whether the frozen model can be fine-tuned by customized noise. Our approach is motivated by the scientific study of beneficial noise, namely Positive-incentive Noise (Pi-noise or $Ï$-noise) , which quantitatively analyzes the impact of noise. It therefore implies a new scheme to learn beneficial noise distribution that can be employed to fine-tune VL models. Focusing on few-shot classification tasks based on CLIP, we reformulate the inference process of CLIP and apply variational inference, demonstrating how to generate $Ï$-noise towards visual and linguistic modalities. Then, we propose Positive-incentive Noise Injector (PiNI), which can fine-tune CLIP via injecting noise into both visual and text encoders. Since the proposed method can learn the distribution of beneficial noise, we can obtain more diverse embeddings of vision and language to better align these two modalities for specific downstream tasks within limited computational resources. We evaluate different noise incorporation approaches and network architectures of PiNI. The evaluation across 11 datasets demonstrates its effectiveness.

Paperid: 2677, https://arxiv.org/pdf/2412.06082.pdf

Abstract:
Recent advances in self-supervision and contrastive learning have brought the performance of foundation models to unprecedented levels in a variety of tasks. Fueled by this progress, these models are becoming the prevailing approach for a wide array of real-world vision problems, including risk-sensitive and high-stakes applications. However, ensuring safe deployment in these scenarios requires a more comprehensive understanding of their uncertainty modeling capabilities, which has been barely explored. In this work, we delve into the behaviour of vision and vision-language foundation models under Conformal Prediction (CP), a statistical framework that provides theoretical guarantees of marginal coverage of the true class. Across extensive experiments including popular vision classification benchmarks, well-known foundation vision models, and three CP methods, our findings reveal that foundation models are well-suited for conformalization procedures, particularly those integrating Vision Transformers. We also show that calibrating the confidence predictions of these models, a popular strategy to improve their uncertainty quantification, actually leads to efficiency degradation of the conformal set on adaptive CP methods. Furthermore, few-shot adaptation of Vision-Language Models (VLMs) to downstream tasks, whose popularity is surging, enhances conformal scores compared to zero-shot predictions. Last, our empirical study exposes APS as particularly promising in the context of vision foundation models, as it does not violate the marginal coverage guarantees across multiple challenging, yet realistic scenarios.

Paperid: 2678, https://arxiv.org/pdf/2411.17002.pdf

Abstract:
Vision-language foundation models, such as CLIP, have shown unprecedented zero-shot performance across a wide range of tasks. Nevertheless, these models may be unreliable under distributional shifts, as their performance is significantly degraded. In this work, we explore how to efficiently leverage class text information to mitigate these distribution drifts encountered by large pre-trained vision-language models (VLMs) during test-time inference. In particular, we propose to generate pseudo-labels for the test-time samples by exploiting generic class text embeddings as fixed centroids of a label assignment problem, which is efficiently solved with Optimal Transport. Furthermore, the proposed adaptation method (CLIP-OT) integrates a multiple template knowledge distillation approach, which replicates multi-view contrastive learning strategies in unsupervised representation learning but without incurring additional computational complexity. Extensive experiments on multiple popular test-time adaptation benchmarks presenting diverse complexity empirically show the superiority of CLIP-OT, achieving performance gains of up to 7% over recent state-of-the-art methods, yet being computationally and memory efficient.

Paperid: 2679, https://arxiv.org/pdf/2411.14279.pdf

Abstract:
Large vision-language models (LVLMs) have achieved impressive results in various vision-language tasks. However, despite showing promising performance, LVLMs suffer from hallucinations caused by language bias, leading to diminished focus on images and ineffective visual comprehension. We identify two primary reasons for this bias: 1. Different scales of training data between the pretraining stage of LLM and multimodal alignment stage. 2. The learned inference bias due to short-term dependency of text data. Therefore, we propose LACING, a systemic framework designed to address the language bias of LVLMs with muLtimodal duAl-attention meChanIsm (MDA) aNd soft-image Guidance (IFG). Specifically, MDA introduces a parallel dual-attention mechanism that enhances the integration of visual inputs across the model. IFG introduces a learnable soft visual prompt during training and inference to replace visual inputs, designed to compel LVLMs to prioritize text inputs. Then, IFG further proposes a novel decoding strategy using the soft visual prompt to mitigate the model's over-reliance on adjacent text inputs. Comprehensive experiments demonstrate that our method effectively debiases LVLMs from their language bias, enhancing visual comprehension and reducing hallucinations without requiring additional training resources or data. The code and model are available at [lacing-lvlm.github.io](https://lacing-lvlm.github.io).

Paperid: 2680, https://arxiv.org/pdf/2411.13578.pdf

Abstract:
How can models effectively detect out-of-distribution (OOD) samples in complex, multi-label settings without extensive retraining? Existing OOD detection methods struggle to capture the intricate semantic relationships and label co-occurrences inherent in multi-label settings, often requiring large amounts of training data and failing to generalize to unseen label combinations. While large language models have revolutionized zero-shot OOD detection, they primarily focus on single-label scenarios, leaving a critical gap in handling real-world tasks where samples can be associated with multiple interdependent labels. To address these challenges, we introduce COOD, a novel zero-shot multi-label OOD detection framework. COOD leverages pre-trained vision-language models, enhancing them with a concept-based label expansion strategy and a new scoring function. By enriching the semantic space with both positive and negative concepts for each label, our approach models complex label dependencies, precisely differentiating OOD samples without the need for additional training. Extensive experiments demonstrate that our method significantly outperforms existing approaches, achieving approximately 95% average AUROC on both VOC and COCO datasets, while maintaining robust performance across varying numbers of labels and different types of OOD samples.

Paperid: 2681, https://arxiv.org/pdf/2411.03359.pdf

Abstract:
Out-of-distribution (OOD) detection is crucial for deploying reliable machine learning models in open-world applications. Recent advances in CLIP-based OOD detection have shown promising results via regularizing prompt tuning with OOD features extracted from ID data. However, the irrelevant context mined from ID data can be spurious due to the inaccurate foreground-background decomposition, thus limiting the OOD detection performance. In this work, we propose a novel framework, namely, Self-Calibrated Tuning (SCT), to mitigate this problem for effective OOD detection with only the given few-shot ID data. Specifically, SCT introduces modulating factors respectively on the two components of the original learning objective. It adaptively directs the optimization process between the two tasks during training on data with different prediction uncertainty to calibrate the influence of OOD regularization, which is compatible with many prompt tuning based OOD detection methods. Extensive experiments and analyses have been conducted to characterize and demonstrate the effectiveness of the proposed SCT. The code is publicly available.

Paperid: 2682, https://arxiv.org/pdf/2410.17883.pdf

Abstract:
This paper introduces a novel mobile phone control architecture, Lightweight Multi-modal App Control (LiMAC), for efficient interactions and control across various Android apps. LiMAC takes as input a textual goal and a sequence of past mobile observations, such as screenshots and corresponding UI trees, to generate precise actions. To address the computational constraints inherent to smartphones, we introduce a small Action Transformer (AcT) integrated with a fine-tuned vision-language model (VLM) for real-time decision-making and task execution. We evaluate LiMAC on two open-source mobile control datasets, demonstrating the superior performance of our small-form-factor approach against fine-tuned versions of open-source VLMs, such as Florence2 and Qwen2-VL. It also significantly outperforms prompt engineering baselines utilising closed-source foundation models like GPT-4o. More specifically, LiMAC increases the overall action accuracy by up to 19% compared to fine-tuned VLMs, and up to 42% compared to prompt-engineering baselines.

Paperid: 2683, https://arxiv.org/pdf/2410.15744.pdf

Abstract:
Recent advancements in medical vision-language pre-training models have driven significant progress in zero-shot disease recognition. However, transferring image-level knowledge to pixel-level tasks, such as lesion segmentation in 3D CT scans, remains a critical challenge. Due to the complexity and variability of pathological visual characteristics, existing methods struggle to align fine-grained lesion features not encountered during training with disease-related textual representations. In this paper, we present Malenia, a novel multi-scale lesion-level mask-attribute alignment framework, specifically designed for 3D zero-shot lesion segmentation. Malenia improves the compatibility between mask representations and their associated elemental attributes, explicitly linking the visual features of unseen lesions with the extensible knowledge learned from previously seen ones. Furthermore, we design a Cross-Modal Knowledge Injection module to enhance both visual and textual features with mutually beneficial information, effectively guiding the generation of segmentation results. Comprehensive experiments across three datasets and 12 lesion categories validate the superior performance of Malenia.

Paperid: 2684, https://arxiv.org/pdf/2410.08478.pdf

Abstract:
Applying large pre-trained Vision-Language Models to recommendation is a burgeoning field, a direction we term Vision-Language-Recommendation (VLR). Bringing VLR to user-oriented on-device intelligence within a federated learning framework is a crucial step for enhancing user privacy and delivering personalized experiences. This paper introduces FedVLR, a federated VLR framework specially designed for user-specific personalized fusion of vision-language representations. At its core is a novel bi-level fusion mechanism: The server-side multi-view fusion module first generates a diverse set of pre-fused multimodal views. Subsequently, each client employs a user-specific mixture-of-expert mechanism to adaptively integrate these views based on individual user interaction history. This designed lightweight personalized fusion module provides an efficient solution to implement a federated VLR system. The effectiveness of our proposed FedVLR has been validated on seven benchmark datasets.

Paperid: 2685, https://arxiv.org/pdf/2410.07087.pdf

Abstract:
Developing agents capable of navigating to a target location based on language instructions and visual information, known as vision-language navigation (VLN), has attracted widespread interest. Most research has focused on ground-based agents, while UAV-based VLN remains relatively underexplored. Recent efforts in UAV vision-language navigation predominantly adopt ground-based VLN settings, relying on predefined discrete action spaces and neglecting the inherent disparities in agent movement dynamics and the complexity of navigation tasks between ground and aerial environments. To address these disparities and challenges, we propose solutions from three perspectives: platform, benchmark, and methodology. To enable realistic UAV trajectory simulation in VLN tasks, we propose the OpenUAV platform, which features diverse environments, realistic flight control, and extensive algorithmic support. We further construct a target-oriented VLN dataset consisting of approximately 12k trajectories on this platform, serving as the first dataset specifically designed for realistic UAV VLN tasks. To tackle the challenges posed by complex aerial environments, we propose an assistant-guided UAV object search benchmark called UAV-Need-Help, which provides varying levels of guidance information to help UAVs better accomplish realistic VLN tasks. We also propose a UAV navigation LLM that, given multi-view images, task descriptions, and assistant instructions, leverages the multimodal understanding capabilities of the MLLM to jointly process visual and textual information, and performs hierarchical trajectory generation. The evaluation results of our method significantly outperform the baseline models, while there remains a considerable gap between our results and those achieved by human operators, underscoring the challenge presented by the UAV-Need-Help task.

Paperid: 2686, https://arxiv.org/pdf/2601.14628.pdf

Abstract:
Recent advances in embodied intelligence have leveraged massive scaling of data and model parameters to master natural-language command following and multi-task control. In contrast, biological systems demonstrate an innate ability to acquire skills rapidly from sparse experience. Crucially, current robotic policies struggle to replicate the dynamic stability, reflexive responsiveness, and temporal memory inherent in biological motion. Here we present Neuromorphic Vision-Language-Action (NeuroVLA), a framework that mimics the structural organization of the bio-nervous system between the cortex, cerebellum, and spinal cord. We adopt a system-level bio-inspired design: a high-level model plans goals, an adaptive cerebellum module stabilizes motion using high-frequency sensors feedback, and a bio-inspired spinal layer executes lightning-fast actions generation. NeuroVLA represents the first deployment of a neuromorphic VLA on physical robotics, achieving state-of-the-art performance. We observe the emergence of biological motor characteristics without additional data or special guidance: it stops the shaking in robotic arms, saves significant energy(only 0.4w on Neuromorphic Processor), shows temporal memory ability and triggers safety reflexes in less than 20 milliseconds.

Paperid: 2687, https://arxiv.org/pdf/2512.20557.pdf

Abstract:
Vision-language models (VLM) excel at general understanding yet remain weak at dynamic spatial reasoning (DSR), i.e., reasoning about the evolvement of object geometry and relationship in 3D space over time, largely due to the scarcity of scalable 4D-aware training resources. To bridge this gap across aspects of dataset, benchmark and model, we introduce DSR Suite. First, we propose an automated pipeline that generates multiple-choice question-answer pairs from in-the-wild videos for DSR. By leveraging modern vision foundation models, the pipeline extracts rich geometric and motion information, including camera poses, local point clouds, object masks, orientations, and 3D trajectories. These geometric cues enable the construction of DSR-Train for learning and further human-refined DSR-Bench for evaluation. Compared with previous works, our data emphasize (i) in-the-wild video sources, (ii) object- and scene-level 3D requirements, (iii) viewpoint transformations, (iv) multi-object interactions, and (v) fine-grained, procedural answers. Beyond data, we propose a lightweight Geometry Selection Module (GSM) to seamlessly integrate geometric priors into VLMs, which condenses question semantics and extracts question-relevant knowledge from pretrained 4D reconstruction priors into a compact set of geometry tokens. This targeted extraction avoids overwhelming the model with irrelevant knowledge. Experiments show that integrating DSR-Train and GSM into Qwen2.5-VL-7B significantly enhances its dynamic spatial reasoning capability, while maintaining accuracy on general video understanding benchmarks.

Paperid: 2688, https://arxiv.org/pdf/2512.18211.pdf

Abstract:
Trajectory planning is a fundamental yet challenging component of autonomous driving. End-to-end planners frequently falter under adverse weather, unpredictable human behavior, or complex road layouts, primarily because they lack strong generalization or few-shot capabilities beyond their training data. We propose LLaViDA, a Large Language Vision Driving Assistant that leverages a Vision-Language Model (VLM) for object motion prediction, semantic grounding, and chain-of-thought reasoning for trajectory planning in autonomous driving. A two-stage training pipeline--supervised fine-tuning followed by Trajectory Preference Optimization (TPO)--enhances scene understanding and trajectory planning by injecting regression-based supervision, produces a powerful "VLM Trajectory Planner for Autonomous Driving." On the NuScenes benchmark, LLaViDA surpasses state-of-the-art end-to-end and other recent VLM/LLM-based baselines in open-loop trajectory planning task, achieving an average L2 trajectory error of 0.31 m and a collision rate of 0.10% on the NuScenes test set. The code for this paper is available at GitHub.

Paperid: 2689, https://arxiv.org/pdf/2512.16561.pdf

Abstract:
While current multimodal models can answer questions based on 2D images, they lack intrinsic 3D object perception, limiting their ability to comprehend spatial relationships and depth cues in 3D scenes. In this work, we propose N3D-VLM, a novel unified framework that seamlessly integrates native 3D object perception with 3D-aware visual reasoning, enabling both precise 3D grounding and interpretable spatial understanding. Unlike conventional end-to-end models that directly predict answers from RGB/RGB-D inputs, our approach equips the model with native 3D object perception capabilities, enabling it to directly localize objects in 3D space based on textual descriptions. Building upon accurate 3D object localization, the model further performs explicit reasoning in 3D, achieving more interpretable and structured spatial understanding. To support robust training for these capabilities, we develop a scalable data construction pipeline that leverages depth estimation to lift large-scale 2D annotations into 3D space, significantly increasing the diversity and coverage for 3D object grounding data, yielding over six times larger than the largest existing single-image 3D detection dataset. Moreover, the pipeline generates spatial question-answering datasets that target chain-of-thought (CoT) reasoning in 3D, facilitating joint training for both 3D object localization and 3D spatial reasoning. Experimental results demonstrate that our unified framework not only achieves state-of-the-art performance on 3D grounding tasks, but also consistently surpasses existing methods in 3D spatial reasoning in vision-language model.

Paperid: 2690, https://arxiv.org/pdf/2510.27256.pdf

Abstract:
Vision-Language Models (VLMs) excel in diverse multimodal tasks. However, user requirements vary across scenarios, which can be categorized into fast response, high-quality output, and low energy consumption. Relying solely on large models deployed in the cloud for all queries often leads to high latency and energy cost, while small models deployed on edge devices are capable of handling simpler tasks with low latency and energy cost. To fully leverage the strengths of both large and small models, we propose ECVL-ROUTER, the first scenario-aware routing framework for VLMs. Our approach introduces a new routing strategy and evaluation metrics that dynamically select the appropriate model for each query based on user requirements, maximizing overall utility. We also construct a multimodal response-quality dataset tailored for router training and validate the approach through extensive experiments. Results show that our approach successfully routes over 80\% of queries to the small model while incurring less than 10\% drop in problem solving probability.

Paperid: 2691, https://arxiv.org/pdf/2510.23763.pdf

Abstract:
Recent advances in Multimodal Large Language Models (MLLMs) have driven rapid progress in Vision-Language-Action (VLA) models for robotic manipulation. Although effective in many scenarios, current approaches largely rely on explicit instructions, whereas in real-world interactions, humans rarely issue instructions directly. Effective collaboration requires robots to infer user intentions proactively. In this work, we introduce cross-modal contextual instructions, a new setting where intent is derived from spoken dialogue, environmental sounds, and visual cues rather than explicit commands. To address this new setting, we present RoboOmni, a Perceiver-Thinker-Talker-Executor framework based on end-to-end omni-modal LLMs that unifies intention recognition, interaction confirmation, and action execution. RoboOmni fuses auditory and visual signals spatiotemporally for robust intention recognition, while supporting direct speech interaction. To address the absence of training data for proactive intention recognition in robotic manipulation, we build OmniAction, comprising 140k episodes, 5k+ speakers, 2.4k event sounds, 640 backgrounds, and six contextual instruction types. Experiments in simulation and real-world settings show that RoboOmni surpasses text- and ASR-based baselines in success rate, inference speed, intention recognition, and proactive assistance.

Paperid: 2692, https://arxiv.org/pdf/2510.19817.pdf

Abstract:
We present olmOCR 2, the latest in our family of powerful OCR systems for converting digitized print documents, like PDFs, into clean, naturally ordered plain text. olmOCR 2 is powered by olmOCR-2-7B-1025, a specialized, 7B vision language model (VLM) trained using reinforcement learning with verifiable rewards (RLVR), where our rewards are a diverse set of binary unit tests. To scale unit test creation, we develop a pipeline for generating synthetic documents with diverse and challenging layouts, known ground-truth HTML source code, and extracted test cases. We show that RL training on these test cases results in state-of-the-art performance on olmOCR-Bench, our English-language OCR benchmark, with the largest improvements in math formula conversion, table parsing, and multi-column layouts compared to previous versions. We release our model, data and code under permissive open licenses.

Paperid: 2693, https://arxiv.org/pdf/2510.04225.pdf

Abstract:
The rapid growth of AI-generated imagery has blurred the boundary between real and synthetic content, raising critical concerns for digital integrity. Vision-language models (VLMs) offer interpretability through explanations but often fail to detect subtle artifacts in high-quality synthetic images. We propose ZoomIn, a two-stage forensic framework that improves both accuracy and interpretability. Mimicking human visual inspection, ZoomIn first scans an image to locate suspicious regions and then performs a focused analysis on these zoomed-in areas to deliver a grounded verdict. To support training, we introduce MagniFake, a dataset of 20,000 real and high-quality synthetic images annotated with bounding boxes and forensic explanations, generated through an automated VLM-based pipeline. Our method achieves 96.39% accuracy with robust generalization, while providing human-understandable explanations grounded in visual evidence.

Paperid: 2694, https://arxiv.org/pdf/2508.18687.pdf

Abstract:
In high-stakes medical applications, consistent answering across diverse question phrasings is essential for reliable diagnosis. However, we reveal that current Medical Vision-Language Models (Med-VLMs) exhibit concerning fragility in Medical Visual Question Answering, as their answers fluctuate significantly when faced with semantically equivalent rephrasings of medical questions. We attribute this to two limitations: (1) insufficient alignment of medical concepts, leading to divergent reasoning patterns, and (2) hidden biases in training data that prioritize syntactic shortcuts over semantic understanding. To address these challenges, we construct RoMed, a dataset built upon original VQA datasets containing 144k questions with variations spanning word-level, sentence-level, and semantic-level perturbations. When evaluating state-of-the-art (SOTA) models like LLaVA-Med on RoMed, we observe alarming performance drops (e.g., a 40\% decline in Recall) compared to original VQA benchmarks, exposing critical robustness gaps. To bridge this gap, we propose Consistency and Contrastive Learning (CCL), which integrates two key components: (1) knowledge-anchored consistency learning, aligning Med-VLMs with medical knowledge rather than shallow feature patterns, and (2) bias-aware contrastive learning, mitigating data-specific priors through discriminative representation refinement. CCL achieves SOTA performance on three popular VQA benchmarks and notably improves answer consistency by 50\% on the challenging RoMed test set, demonstrating significantly enhanced robustness. Code will be released.

Paperid: 2695, https://arxiv.org/pdf/2508.06869.pdf

Abstract:
Multimodal large language models (MLLMs) demonstrate exceptional performance in vision-language tasks, yet their processing of long videos is constrained by input context length and high computational costs. Sparse frame sampling thus becomes a necessary preprocessing step, with sampled frame quality directly impacting downstream performance. Existing keyframe search algorithms achieve a balance between efficiency and sampled frame quality but heavily rely on the visual modality alone. This makes them difficult to adapt to text-related tasks and often leads to retrieval results deviating from core semantic content. To address this, we propose the VISUAL-SUBTITLE INTEGRATION (VSI), a multimodal keyframe retrieval framework. It employs a dual-branch collaborative retrieval approach combining Video Search and Subtitle Match to fuse complementary visual and textual information for precise localization. Experiments on LongVideoBench and VideoMME demonstrate that VSI achieves state-of-the-art accuracy in keyframe retrieval while delivering breakthrough performance in text-related tasks and exhibiting strong generalization across other tasks.

Paperid: 2696, https://arxiv.org/pdf/2508.04101.pdf

Abstract:
Computer-aided medical image analysis is crucial for disease diagnosis and treatment planning, yet limited annotated datasets restrict medical-specific model development. While vision-language models (VLMs) like CLIP offer strong generalization capabilities, their direct application to medical imaging analysis is impeded by a significant domain gap. Existing approaches to bridge this gap, including prompt learning and one-way modality interaction techniques, typically focus on introducing domain knowledge to a single modality. Although this may offer performance gains, it often causes modality misalignment, thereby failing to unlock the full potential of VLMs. In this paper, we propose \textbf{NEARL-CLIP} (i\underline{N}teracted qu\underline{E}ry \underline{A}daptation with o\underline{R}thogona\underline{L} Regularization), a novel cross-modality interaction VLM-based framework that contains two contributions: (1) Unified Synergy Embedding Transformer (USEformer), which dynamically generates cross-modality queries to promote interaction between modalities, thus fostering the mutual enrichment and enhancement of multi-modal medical domain knowledge; (2) Orthogonal Cross-Attention Adapter (OCA). OCA introduces an orthogonality technique to decouple the new knowledge from USEformer into two distinct components: the truly novel information and the incremental knowledge. By isolating the learning process from the interference of incremental knowledge, OCA enables a more focused acquisition of new information, thereby further facilitating modality interaction and unleashing the capability of VLMs. Notably, NEARL-CLIP achieves these two contributions in a parameter-efficient style, which only introduces \textbf{1.46M} learnable parameters.

Paperid: 2697, https://arxiv.org/pdf/2508.01741.pdf

Abstract:
Fine-tuning open-source Vision-Language Models (VLMs) creates a critical yet underexplored attack surface: vulnerabilities in the base VLM could be retained in fine-tuned variants, rendering them susceptible to transferable jailbreak attacks. To demonstrate this risk, we introduce the Simulated Ensemble Attack (SEA), a novel grey-box jailbreak method in which the adversary has full access to the base VLM but no knowledge of the fine-tuned target's weights or training configuration. To improve jailbreak transferability across fine-tuned VLMs, SEA combines two key techniques: Fine-tuning Trajectory Simulation (FTS) and Targeted Prompt Guidance (TPG). FTS generates transferable adversarial images by simulating the vision encoder's parameter shifts, while TPG is a textual strategy that steers the language decoder toward adversarially optimized outputs. Experiments on the Qwen2-VL family (2B and 7B) demonstrate that SEA achieves high transfer attack success rates exceeding 86.5% and toxicity rates near 49.5% across diverse fine-tuned variants, even those specifically fine-tuned to improve safety behaviors. Notably, while direct PGD-based image jailbreaks rarely transfer across fine-tuned VLMs, SEA reliably exploits inherited vulnerabilities from the base model, significantly enhancing transferability. These findings highlight an urgent need to safeguard fine-tuned proprietary VLMs against transferable vulnerabilities inherited from open-source foundations, motivating the development of holistic defenses across the entire model lifecycle.

Paperid: 2698, https://arxiv.org/pdf/2508.01408.pdf

Abstract:
The attribution of artworks in general and of paintings in particular has always been an issue in art. The advent of powerful artificial intelligence models that can generate and analyze images creates new challenges for painting attribution. On the one hand, AI models can create images that mimic the style of a painter, which can be incorrectly attributed, for example, by other AI models. On the other hand, AI models may not be able to correctly identify the artist for real paintings, inducing users to incorrectly attribute paintings. In this paper, both problems are experimentally studied using state-of-the-art AI models for image generation and analysis on a large dataset with close to 40,000 paintings from 128 artists. The results show that vision language models have limited capabilities to: 1) perform canvas attribution and 2) to identify AI generated images. As users increasingly rely on queries to AI models to get information, these results show the need to improve the capabilities of VLMs to reliably perform artist attribution and detection of AI generated images to prevent the spread of incorrect information.

Paperid: 2699, https://arxiv.org/pdf/2507.21619.pdf

Abstract:
Industrial anomaly detection (IAD) plays a crucial role in maintaining the safety and reliability of manufacturing systems. While multimodal large language models (MLLMs) show strong vision-language reasoning abilities, their effectiveness in IAD remains limited without domain-specific adaptation. In this work, we propose EMIT, a unified framework that enhances MLLMs for IAD via difficulty-aware group relative policy optimization (GRPO). EMIT constructs a multi-task IAD dataset and utilizes GPT-generated object text descriptions to compensate for missing defective images. For few-shot anomaly detection, it integrates a soft prompt and heatmap-guided contrastive embeddings derived from patch-level comparisons. To better handle difficult data samples, i.e., cases where the MLLM struggles to generate correct answers, we propose a difficulty-aware GRPO that extends the original GRPO by incorporating a response resampling strategy to ensure the inclusion of correct answers in the sampled responses, as well as an advantage reweighting mechanism to strengthen learning from such difficult data samples. Extensive experiments on the MMAD benchmark demonstrate that EMIT significantly enhances the IAD performance of MLLMs, achieving an average improvement of 7.77\% over the base model (InternVL3-8B) across seven tasks.

Paperid: 2700, https://arxiv.org/pdf/2506.19610.pdf

Abstract:
Recent advances in multimodal techniques have led to significant progress in Medical Visual Question Answering (Med-VQA). However, most existing models focus on global image features rather than localizing disease-specific regions crucial for diagnosis. Additionally, current research tends to emphasize answer accuracy at the expense of the reasoning pathway, yet both are crucial for clinical decision-making. To address these challenges, we propose From Vision to Text Chain-of-Thought (V2T-CoT), a novel approach that automates the localization of preference areas within biomedical images and incorporates this localization into region-level pixel attention as knowledge for Vision CoT. By fine-tuning the vision language model on constructed R-Med 39K dataset, V2T-CoT provides definitive medical reasoning paths. V2T-CoT integrates visual grounding with textual rationale generation to establish precise and explainable diagnostic results. Experimental results across four Med-VQA benchmarks demonstrate state-of-the-art performance, achieving substantial improvements in both performance and interpretability.

Paperid: 2701, https://arxiv.org/pdf/2506.14907.pdf

Abstract:
Inspired by the impressive reasoning capabilities demonstrated by reinforcement learning approaches like DeepSeek-R1, recent emerging research has begun exploring the use of reinforcement learning (RL) to enhance vision-language models (VLMs) for multimodal reasoning tasks. However, most existing multimodal reinforcement learning approaches remain limited to spatial reasoning within single-image contexts, yet still struggle to generalize to more complex and real-world scenarios involving multi-image positional reasoning, where understanding the relationships across images is crucial. To address this challenge, we propose a general reinforcement learning approach PeRL tailored for interleaved multimodal tasks, and a multi-stage strategy designed to enhance the exploration-exploitation trade-off, thereby improving learning efficiency and task performance. Specifically, we introduce permutation of image sequences to simulate varied positional relationships to explore more spatial and positional diversity. Furthermore, we design a rollout filtering mechanism for resampling to focus on trajectories that contribute most to learning optimal behaviors to exploit learned policies effectively. We evaluate our model on 5 widely-used multi-image benchmarks and 3 single-image benchmarks. Our experiments confirm that PeRL trained model consistently surpasses R1-related and interleaved VLM baselines by a large margin, achieving state-of-the-art performance on multi-image benchmarks, while preserving comparable performance on single-image tasks.

Paperid: 2702, https://arxiv.org/pdf/2506.12849.pdf

Abstract:
In medical visual question answering (Med-VQA), achieving accurate responses relies on three critical steps: precise perception of medical imaging data, logical reasoning grounded in visual input and textual questions, and coherent answer derivation from the reasoning process. Recent advances in general vision-language models (VLMs) show that large-scale reinforcement learning (RL) could significantly enhance both reasoning capabilities and overall model performance. However, their application in medical domains is hindered by two fundamental challenges: 1) misalignment between perceptual understanding and reasoning stages, and 2) inconsistency between reasoning pathways and answer generation, both compounded by the scarcity of high-quality medical datasets for effective large-scale RL. In this paper, we first introduce Med-Zero-17K, a curated dataset for pure RL-based training, encompassing over 30 medical image modalities and 24 clinical tasks. Moreover, we propose a novel large-scale RL framework for Med-VLMs, Consistency-Aware Preference Optimization (CAPO), which integrates rewards to ensure fidelity between perception and reasoning, consistency in reasoning-to-answer derivation, and rule-based accuracy for final responses. Extensive experiments on both in-domain and out-of-domain scenarios demonstrate the superiority of our method over strong VLM baselines, showcasing strong generalization capability to 3D Med-VQA benchmarks and R1-like training paradigms.

Paperid: 2703, https://arxiv.org/pdf/2506.07227.pdf

Abstract:
Multimodal large language models (MLLMs) have achieved strong performance on vision-language tasks but still struggle with fine-grained visual differences, leading to hallucinations or missed semantic shifts. We attribute this to limitations in both training data and learning objectives. To address these issues, we propose a controlled data generation pipeline that produces minimally edited image pairs with semantically aligned captions. Using this pipeline, we construct the Micro Edit Dataset (MED), containing over 50K image-text pairs spanning 11 fine-grained edit categories, including attribute, count, position, and object presence changes. Building on MED, we introduce a supervised fine-tuning (SFT) framework with a feature-level consistency loss that promotes stable visual embeddings under small edits. We evaluate our approach on the Micro Edit Detection benchmark, which includes carefully balanced evaluation pairs designed to test sensitivity to subtle visual variations across the same edit categories. Our method improves difference detection accuracy and reduces hallucinations compared to strong baselines, including GPT-4o. Moreover, it yields consistent gains on standard vision-language tasks such as image captioning and visual question answering. These results demonstrate the effectiveness of combining targeted data and alignment objectives for enhancing fine-grained visual reasoning in MLLMs.

Paperid: 2704, https://arxiv.org/pdf/2506.00805.pdf

Abstract:
Medical Vision-Language Models (Med-VLMs) have achieved success across various tasks, yet most existing methods overlook the modality misalignment issue that can lead to untrustworthy responses in clinical settings. In this paper, we propose Hierarchical Self-Contrastive Rewarding (HSCR), a novel approach that addresses two critical challenges in Med-VLM alignment: 1) Cost-effective generation of high-quality preference data; 2) Capturing nuanced and context-aware preferences for improved alignment. HSCR first leverages the inherent capability of Med-VLMs to generate dispreferred responses with higher sampling probability. By analyzing output logit shifts after visual token dropout, we identify modality-coupled tokens that induce misalignment and derive an implicit alignment reward function. This function guides token replacement with hallucinated ones during decoding, producing high-quality dispreferred data. Furthermore, HSCR introduces a multi-level preference optimization strategy, which extends beyond traditional adjacent-level optimization by incorporating nuanced implicit preferences, leveraging relative quality in dispreferred data to capture subtle alignment cues for more precise and context-aware optimization. Extensive experiments across multiple medical tasks, including Med-VQA, medical image captioning and instruction following, demonstrate that HSCR not only enhances zero-shot performance but also significantly improves modality alignment and trustworthiness with just 2,000 training entries.

Paperid: 2705, https://arxiv.org/pdf/2505.18855.pdf

Abstract:
This work investigates the optimal allocation of inference compute across three key scaling factors in video vision language models: language model size, frame count, and the number of visual tokens per frame. While prior works typically focuses on optimizing model efficiency or improving performance without considering resource constraints, we instead identify optimal model configuration under fixed inference compute budgets. We conduct large-scale training sweeps and careful parametric modeling of task performance to identify the inference compute-optimal frontier. Our experiments reveal how task performance depends on scaling factors and finetuning data size, as well as how changes in data size shift the compute-optimal frontier. These findings translate to practical tips for selecting these scaling factors.

Paperid: 2706, https://arxiv.org/pdf/2505.00029.pdf

Abstract:
Large Vision Language Models have demonstrated impressive versatile capabilities through extensive multimodal pre-training, but face significant limitations when incorporating specialized knowledge domains beyond their training distribution. These models struggle with a fundamental dilemma: direct adaptation approaches that inject domain-specific knowledge often trigger catastrophic forgetting of foundational visual-linguistic abilities. We introduce Structured Dialogue Fine-Tuning (SDFT), an effective approach that effectively injects domain-specific knowledge while minimizing catastrophic forgetting. Drawing inspiration from supervised fine-tuning in LLMs and subject-driven personalization in text-to-image diffusion models, our method employs a three-phase dialogue structure: Foundation Preservation reinforces pre-trained visual-linguistic alignment through caption tasks; Contrastive Disambiguation introduces carefully designed counterfactual examples to maintain semantic boundaries; and Knowledge Specialization embeds specialized information through chain-of-thought reasoning. Experimental results across multiple domains confirm SDFT's effectiveness in balancing specialized knowledge acquisition with general capability retention. Our key contributions include a data-centric dialogue template that balances foundational alignment with targeted knowledge integration, a weighted multi-turn supervision framework, and comprehensive evaluation across diverse knowledge types.

Paperid: 2707, https://arxiv.org/pdf/2504.14692.pdf

Abstract:
The practical deployment of medical vision-language models (Med-VLMs) necessitates seamless integration of textual data with diverse visual modalities, including 2D/3D images and videos, yet existing models typically employ separate encoders for different modalities. To address this limitation, we present OmniV-Med, a unified framework for multimodal medical understanding. Our technical contributions are threefold: First, we construct OmniV-Med-Instruct, a comprehensive multimodal medical dataset containing 252K instructional samples spanning 14 medical image modalities and 11 clinical tasks. Second, we devise a rotary position-adaptive encoder that processes multi-resolution 2D/3D images and videos within a unified architecture, diverging from conventional modality-specific encoders. Third, we introduce a medical-aware token pruning mechanism that exploits spatial-temporal redundancy in volumetric data (e.g., consecutive CT slices) and medical videos, effectively reducing 60\% of visual tokens without performance degradation. Empirical evaluations demonstrate that OmniV-Med-7B achieves state-of-the-art performance on 7 benchmarks spanning 2D/3D medical imaging and video understanding tasks. Notably, our lightweight variant (OmniV-Med-1.5B) attains comparable performance while requiring only 8 RTX3090 GPUs for training and supporting efficient long-video inference. Data, code and model will be released.

Paperid: 2708, https://arxiv.org/pdf/2503.19755.pdf

Abstract:
End-to-end (E2E) autonomous driving methods still struggle to make correct decisions in interactive closed-loop evaluation due to limited causal reasoning capability. Current methods attempt to leverage the powerful understanding and reasoning abilities of Vision-Language Models (VLMs) to resolve this dilemma. However, the problem is still open that few VLMs for E2E methods perform well in the closed-loop evaluation due to the gap between the semantic reasoning space and the purely numerical trajectory output in the action space. To tackle this issue, we propose ORION, a holistic E2E autonomous driving framework by vision-language instructed action generation. ORION uniquely combines a QT-Former to aggregate long-term history context, a Large Language Model (LLM) for driving scenario reasoning, and a generative planner for precision trajectory prediction. ORION further aligns the reasoning space and the action space to implement a unified E2E optimization for both visual question-answering (VQA) and planning tasks. Our method achieves an impressive closed-loop performance of 77.74 Driving Score (DS) and 54.62% Success Rate (SR) on the challenge Bench2Drive datasets, which outperforms state-of-the-art (SOTA) methods by a large margin of 14.28 DS and 19.61% SR.

Paperid: 2709, https://arxiv.org/pdf/2503.10742.pdf

Abstract:
Vision language models (VLMs) demonstrate strong capabilities in jointly processing visual and textual data. However, they often incur substantial computational overhead due to redundant visual information, particularly in long-form video scenarios. Existing approaches predominantly focus on either vision token pruning, which may overlook spatio-temporal dependencies, or keyframe selection, which identifies informative frames but discards others, thus disrupting contextual continuity. In this work, we propose KVTP (Keyframe-oriented Vision Token Pruning), a novel framework that overcomes the drawbacks of token pruning and keyframe selection. By adaptively assigning pruning rates based on frame relevance to the query, KVTP effectively retains essential contextual information while significantly reducing redundant computation. To thoroughly evaluate the long-form video understanding capacities of VLMs, we curated and reorganized subsets from VideoMME, EgoSchema, and NextQA into a unified benchmark named SparseKV-QA that highlights real-world scenarios with sparse but crucial events. Our experiments with VLMs of various scales show that KVTP can reduce token usage by 80% without compromising spatiotemporal and contextual consistency, significantly cutting computation while maintaining the performance. These results demonstrate our approach's effectiveness in efficient long-video processing, facilitating more scalable VLM deployment.

Paperid: 2710, https://arxiv.org/pdf/2503.07456.pdf

Abstract:
Image-Text Retrieval (ITR) finds broad applications in healthcare, aiding clinicians and radiologists by automatically retrieving relevant patient cases in the database given the query image and/or report, for more efficient clinical diagnosis and treatment, especially for rare diseases. However conventional ITR systems typically only rely on global image or text representations for measuring patient image/report similarities, which overlook local distinctiveness across patient cases. This often results in suboptimal retrieval performance. In this paper, we propose an Anatomical Location-Conditioned Image-Text Retrieval (ALC-ITR) framework, which, given a query image and the associated suspicious anatomical region(s), aims to retrieve similar patient cases exhibiting the same disease or symptoms in the same anatomical region. To perform location-conditioned multimodal retrieval, we learn a medical Relevance-Region-Aligned Vision Language (RRA-VL) model with semantic global-level and region-/word-level alignment to produce generalizable, well-aligned multi-modal representations. Additionally, we perform location-conditioned contrastive learning to further utilize cross-pair region-level contrastiveness for improved multi-modal retrieval. We show that our proposed RRA-VL achieves state-of-the-art localization performance in phase-grounding tasks, and satisfying multi-modal retrieval performance with or without location conditioning. Finally, we thoroughly investigate the generalizability and explainability of our proposed ALC-ITR system in providing explanations and preliminary diagnosis reports given retrieved patient cases (conditioned on anatomical regions), with proper off-the-shelf LLM prompts.

Paperid: 2711, https://arxiv.org/pdf/2503.03278.pdf

Abstract:
Visual Language Models (VLMs) have demonstrated impressive capabilities in visual grounding tasks. However, their effectiveness in the medical domain, particularly for abnormality detection and localization within medical images, remains underexplored. A major challenge is the complex and abstract nature of medical terminology, which makes it difficult to directly associate pathological anomaly terms with their corresponding visual features. In this work, we introduce a novel approach to enhance VLM performance in medical abnormality detection and localization by leveraging decomposed medical knowledge. Instead of directly prompting models to recognize specific abnormalities, we focus on breaking down medical concepts into fundamental attributes and common visual patterns. This strategy promotes a stronger alignment between textual descriptions and visual features, improving both the recognition and localization of abnormalities in medical images.We evaluate our method on the 0.23B Florence-2 base model and demonstrate that it achieves comparable performance in abnormality grounding to significantly larger 7B LLaVA-based medical VLMs, despite being trained on only 1.5% of the data used for such models. Experimental results also demonstrate the effectiveness of our approach in both known and previously unseen abnormalities, suggesting its strong generalization capabilities.

Paperid: 2712, https://arxiv.org/pdf/2502.18443.pdf

Abstract:
PDF documents have the potential to provide trillions of novel, high-quality tokens for training language models. However, these documents come in a diversity of types with differing formats and visual layouts that pose a challenge when attempting to extract and faithfully represent the underlying content for language model use. Traditional open source tools often produce lower quality extractions compared to vision language models (VLMs), but reliance on the best VLMs can be prohibitively costly (e.g., over 6,240 USD per million PDF pages for GPT-4o) or infeasible if the PDFs cannot be sent to proprietary APIs. We present olmOCR, an open-source toolkit for processing PDFs into clean, linearized plain text in natural reading order while preserving structured content like sections, tables, lists, equations, and more. Our toolkit runs a fine-tuned 7B vision language model (VLM) trained on olmOCR-mix-0225, a sample of 260,000 pages from over 100,000 crawled PDFs with diverse properties, including graphics, handwritten text and poor quality scans. olmOCR is optimized for large-scale batch processing, able to scale flexibly to different hardware setups and can convert a million PDF pages for only 176 USD. To aid comparison with existing systems, we also introduce olmOCR-Bench, a curated set of 1,400 PDFs capturing many content types that remain challenging even for the best tools and VLMs, including formulas, tables, tiny fonts, old scans, and more. We find olmOCR outperforms even top VLMs including GPT-4o, Gemini Flash 2 and Qwen-2.5-VL. We openly release all components of olmOCR: our fine-tuned VLM model, training code and data, an efficient inference pipeline that supports vLLM and SGLang backends, and benchmark olmOCR-Bench.

Paperid: 2713, https://arxiv.org/pdf/2502.14254.pdf

Abstract:
Recent advancements in Large Language Models (LLMs) and Vision-Language Models (VLMs) have made them powerful tools in embodied navigation, enabling agents to leverage commonsense and spatial reasoning for efficient exploration in unfamiliar environments. Existing LLM-based approaches convert global memory, such as semantic or topological maps, into language descriptions to guide navigation. While this improves efficiency and reduces redundant exploration, the loss of geometric information in language-based representations hinders spatial reasoning, especially in intricate environments. To address this, VLM-based approaches directly process ego-centric visual inputs to select optimal directions for exploration. However, relying solely on a first-person perspective makes navigation a partially observed decision-making problem, leading to suboptimal decisions in complex environments. In this paper, we present a novel vision-language model (VLM)-based navigation framework that addresses these challenges by adaptively retrieving task-relevant cues from a global memory module and integrating them with the agent's egocentric observations. By dynamically aligning global contextual information with local perception, our approach enhances spatial reasoning and decision-making in long-horizon tasks. Experimental results demonstrate that the proposed method surpasses previous state-of-the-art approaches in object navigation tasks, providing a more effective and scalable solution for embodied navigation.

Paperid: 2714, https://arxiv.org/pdf/2502.10040.pdf

Abstract:
Recently, Vision-Language-Action models (VLA) have advanced robot imitation learning, but high data collection costs and limited demonstrations hinder generalization and current imitation learning methods struggle in out-of-distribution scenarios, especially for long-horizon tasks. A key challenge is how to mitigate compounding errors in imitation learning, which lead to cascading failures over extended trajectories. To address these challenges, we propose the Diffusion Trajectory-guided Policy (DTP) framework, which generates 2D trajectories through a diffusion model to guide policy learning for long-horizon tasks. By leveraging task-relevant trajectories, DTP provides trajectory-level guidance to reduce error accumulation. Our two-stage approach first trains a generative vision-language model to create diffusion-based trajectories, then refines the imitation policy using them. Experiments on the CALVIN benchmark show that DTP outperforms state-of-the-art baselines by 25% in success rate, starting from scratch without external pretraining. Moreover, DTP significantly improves real-world robot performance.

Paperid: 2715, https://arxiv.org/pdf/2501.14877.pdf

Abstract:
In real-world settings, vision language models (VLMs) should robustly handle naturalistic, noisy visual content as well as domain-specific language and concepts. For example, K-12 educators using digital learning platforms may need to examine and provide feedback across many images of students' math work. To assess the potential of VLMs to support educators in settings like this one, we introduce DrawEduMath, an English-language dataset of 2,030 images of students' handwritten responses to K-12 math problems. Teachers provided detailed annotations, including free-form descriptions of each image and 11,661 question-answer (QA) pairs. These annotations capture a wealth of pedagogical insights, ranging from students' problem-solving strategies to the composition of their drawings, diagrams, and writing. We evaluate VLMs on teachers' QA pairs, as well as 44,362 synthetic QA pairs derived from teachers' descriptions using language models (LMs). We show that even state-of-the-art VLMs leave much room for improvement on DrawEduMath questions. We also find that synthetic QAs, though imperfect, can yield similar model rankings as teacher-written QAs. We release DrawEduMath to support the evaluation of VLMs' abilities to reason mathematically over images gathered with educational contexts in mind.

Paperid: 2716, https://arxiv.org/pdf/2501.10074.pdf

Abstract:
Spatial reasoning is an essential problem in embodied AI research. Efforts to enhance spatial reasoning abilities through supplementary spatial data and fine-tuning have proven limited and ineffective when addressing complex embodied tasks, largely due to their dependence on language-based outputs. While some approaches have introduced a point-based action space to mitigate this issue, they fall short in managing more intricate tasks within complex environments. This deficiency arises from their failure to fully exploit the inherent thinking and reasoning capabilities that are fundamental strengths of Vision-Language Models (VLMs). To address these limitations, we propose a novel approach named SpatialCoT, specifically designed to bolster the spatial reasoning capabilities of VLMs. Our approach comprises two stages: spatial coordinate bi-directional alignment, which aligns vision-language inputs with spatial coordinates, and chain-of-thought spatial grounding, which harnesses the reasoning capabilities of language models for advanced spatial reasoning. We evaluate SpatialCoT on challenging navigation and manipulation tasks, both in simulation and real-world settings. Experimental results demonstrate that our method significantly outperforms previous state-of-the-art approaches in both tasks.

Paperid: 2717, https://arxiv.org/pdf/2411.13136.pdf

Abstract:
Large pre-trained Vision-Language Models (VLMs) such as CLIP have demonstrated excellent zero-shot generalizability across various downstream tasks. However, recent studies have shown that the inference performance of CLIP can be greatly degraded by small adversarial perturbations, especially its visual modality, posing significant safety threats. To mitigate this vulnerability, in this paper, we propose a novel defense method called Test-Time Adversarial Prompt Tuning (TAPT) to enhance the inference robustness of CLIP against visual adversarial attacks. TAPT is a test-time defense method that learns defensive bimodal (textual and visual) prompts to robustify the inference process of CLIP. Specifically, it is an unsupervised method that optimizes the defensive prompts for each test sample by minimizing a multi-view entropy and aligning adversarial-clean distributions. We evaluate the effectiveness of TAPT on 11 benchmark datasets, including ImageNet and 10 other zero-shot datasets, demonstrating that it enhances the zero-shot adversarial robustness of the original CLIP by at least 48.9% against AutoAttack (AA), while largely maintaining performance on clean examples. Moreover, TAPT outperforms existing adversarial prompt tuning methods across various backbones, achieving an average robustness improvement of at least 36.6%.

Paperid: 2718, https://arxiv.org/pdf/2410.13523.pdf

Abstract:
Medical Vision-Language Pre-training (MedVLP) has made significant progress in enabling zero-shot tasks for medical image understanding. However, training MedVLP models typically requires large-scale datasets with paired, high-quality image-text data, which are scarce in the medical domain. Recent advancements in Large Language Models (LLMs) and diffusion models have made it possible to generate large-scale synthetic image-text pairs. This raises the question: "Can MedVLP succeed using purely synthetic data?" To address this, we use off-the-shelf generative models to create synthetic radiology reports and paired Chest X-ray (CXR) images, and propose an automated pipeline to build a diverse, high-quality synthetic dataset, enabling a rigorous study that isolates model and training settings, focusing entirely from the data perspective. Our results show that MedVLP models trained exclusively on synthetic data outperform those trained on real data by 3.8% in averaged AUC on zero-shot classification. Moreover, using a combination of synthetic and real data leads to a further improvement of 9.07%. Additionally, MedVLP models trained on synthetic or mixed data consistently outperform those trained on real data in zero-shot grounding, as well as in fine-tuned classification and segmentation tasks. Our analysis suggests MedVLP trained on well-designed synthetic data can outperform models trained on real datasets, which may be limited by low-quality samples and long-tailed distributions.

Paperid: 2719, https://arxiv.org/pdf/2410.13088.pdf

Abstract:
Large Language Models (LLMs) and Vision-Language Models (VLMs) have made significant advancements in a wide range of natural language processing and vision-language tasks. Access to large web-scale datasets has been a key factor in their success. However, concerns have been raised about the unauthorized use of copyrighted materials and potential copyright infringement. Existing methods, such as sample-level Membership Inference Attacks (MIA) and distribution-based dataset inference, distinguish member data (data used for training) and non-member data by leveraging the common observation that models tend to memorize and show greater confidence in member data. Nevertheless, these methods face challenges when applied to LLMs and VLMs, such as the requirement for ground-truth member data or non-member data that shares the same distribution as the test data. In this paper, we propose a novel dataset-level membership inference method based on Self-Comparison. We find that a member prefix followed by a non-member suffix (paraphrased from a member suffix) can further trigger the model's memorization on training data. Instead of directly comparing member and non-member data, we introduce paraphrasing to the second half of the sequence and evaluate how the likelihood changes before and after paraphrasing. Unlike prior approaches, our method does not require access to ground-truth member data or non-member data in identical distribution, making it more practical. Extensive experiments demonstrate that our proposed method outperforms traditional MIA and dataset inference techniques across various datasets and models, including including public models, fine-tuned models, and API-based commercial models.

Paperid: 2720, https://arxiv.org/pdf/2410.08876.pdf

Abstract:
Current vision-language models (VLMs) still exhibit inferior performance on knowledge-intensive tasks, primarily due to the challenge of accurately encoding all the associations between visual objects and scenes to their corresponding entities and background knowledge. While retrieval augmentation methods offer an efficient way to integrate external knowledge, extending them to vision-language domain presents unique challenges in (1) precisely retrieving relevant information from external sources due to the inherent discrepancy within the multimodal queries, and (2) being resilient to the irrelevant, extraneous and noisy information contained in the retrieved multimodal knowledge snippets. In this work, we introduce RORA-VLM, a novel and robust retrieval augmentation framework specifically tailored for VLMs, with two key innovations: (1) a 2-stage retrieval process with image-anchored textual-query expansion to synergistically combine the visual and textual information in the query and retrieve the most relevant multimodal knowledge snippets; and (2) a robust retrieval augmentation method that strengthens the resilience of VLMs against irrelevant information in the retrieved multimodal knowledge by injecting adversarial noises into the retrieval-augmented training process, and filters out extraneous visual information, such as unrelated entities presented in images, via a query-oriented visual token refinement strategy. We conduct extensive experiments to validate the effectiveness and robustness of our proposed methods on three widely adopted benchmark datasets. Our results demonstrate that with a minimal amount of training instance, RORA-VLM enables the base model to achieve significant performance improvement and constantly outperform state-of-the-art retrieval-augmented VLMs on all benchmarks while also exhibiting a novel zero-shot domain transfer capability.

Paperid: 2721, https://arxiv.org/pdf/2410.07577.pdf

Abstract:
Recent advancements in 3D reconstruction methods and vision-language models have propelled the development of multi-modal 3D scene understanding, which has vital applications in robotics, autonomous driving, and virtual/augmented reality. However, current multi-modal scene understanding approaches have naively embedded semantic representations into 3D reconstruction methods without striking a balance between visual and language modalities, which leads to unsatisfying semantic rasterization of translucent or reflective objects, as well as over-fitting on color modality. To alleviate these limitations, we propose a solution that adequately handles the distinct visual and semantic modalities, i.e., a 3D vision-language Gaussian splatting model for scene understanding, to put emphasis on the representation learning of language modality. We propose a novel cross-modal rasterizer, using modality fusion along with a smoothed semantic indicator for enhancing semantic rasterization. We also employ a camera-view blending technique to improve semantic consistency between existing and synthesized views, thereby effectively mitigating over-fitting. Extensive experiments demonstrate that our method achieves state-of-the-art performance in open-vocabulary semantic segmentation, surpassing existing methods by a significant margin.

Paperid: 2722, https://arxiv.org/pdf/2409.00543.pdf

Abstract:
Recent advancements in medical vision-language pre-training (MedVLP) have significantly enhanced zero-shot medical vision tasks such as image classification by leveraging large-scale medical image-text pair pre-training. However, the performance of these tasks can be heavily influenced by the variability in textual prompts describing the categories, necessitating robustness in MedVLP models to diverse prompt styles. Yet, this sensitivity remains underexplored. In this work, we are the first to systematically assess the sensitivity of three widely-used MedVLP methods to a variety of prompts across 15 different diseases. To achieve this, we designed six unique prompt styles to mirror real clinical scenarios, which were subsequently ranked by interpretability. Our findings indicate that all MedVLP models evaluated show unstable performance across different prompt styles, suggesting a lack of robustness. Additionally, the models' performance varied with increasing prompt interpretability, revealing difficulties in comprehending complex medical concepts. This study underscores the need for further development in MedVLP methodologies to enhance their robustness to diverse zero-shot prompts.

Paperid: 2723, https://arxiv.org/pdf/2601.04052.pdf

Abstract:
Vision-Language-Action (VLA) models have demonstrated impressive capabilities in generalized robotic control; however, they remain notoriously brittle to linguistic perturbations. We identify a critical ``modality collapse'' phenomenon where strong visual priors overwhelm sparse linguistic signals, causing agents to overfit to specific instruction phrasings while ignoring the underlying semantic intent. To address this, we propose \textbf{Residual Semantic Steering (RSS)}, a probabilistic framework that disentangles physical affordance from semantic execution. RSS introduces two theoretical innovations: (1) \textbf{Monte Carlo Syntactic Integration}, which approximates the true semantic posterior via dense, LLM-driven distributional expansion, and (2) \textbf{Residual Affordance Steering}, a dual-stream decoding mechanism that explicitly isolates the causal influence of language by subtracting the visual affordance prior. Theoretical analysis suggests that RSS effectively maximizes the mutual information between action and intent while suppressing visual distractors. Empirical results across diverse manipulation benchmarks demonstrate that RSS achieves state-of-the-art robustness, maintaining performance even under adversarial linguistic perturbations.

Paperid: 2724, https://arxiv.org/pdf/2512.07582.pdf

Abstract:
Developing robust and general-purpose manipulation policies represents a fundamental objective in robotics research. While Vision-Language-Action (VLA) models have demonstrated promising capabilities for end-to-end robot control, existing approaches still exhibit limited generalization to tasks beyond their training distributions. In contrast, humans possess remarkable proficiency in acquiring novel skills by simply observing others performing them once. Inspired by this capability, we propose ViVLA, a generalist robotic manipulation policy that achieves efficient task learning from a single expert demonstration video at test time. Our approach jointly processes an expert demonstration video alongside the robot's visual observations to predict both the demonstrated action sequences and subsequent robot actions, effectively distilling fine-grained manipulation knowledge from expert behavior and transferring it seamlessly to the agent. To enhance the performance of ViVLA, we develop a scalable expert-agent pair data generation pipeline capable of synthesizing paired trajectories from easily accessible human videos, further augmented by curated pairs from publicly available datasets. This pipeline produces a total of 892,911 expert-agent samples for training ViVLA. Experimental results demonstrate that our ViVLA is able to acquire novel manipulation skills from only a single expert demonstration video at test time. Our approach achieves over 30% improvement on unseen LIBERO tasks and maintains above 35% gains with cross-embodiment videos. Real-world experiments demonstrate effective learning from human videos, yielding more than 38% improvement on unseen tasks.

Paperid: 2725, https://arxiv.org/pdf/2512.02902.pdf

Abstract:
Vision-language-action (VLA) models achieve strong in-distribution performance but degrade sharply under novel camera viewpoints and visual perturbations. We show that this brittleness primarily arises from misalignment in Spatial Modeling, rather than Physical Modeling. To address this, we propose a one-shot adaptation framework that recalibrates visual representations through lightweight, learnable updates. Our first method, Feature Token Modulation (FTM), applies a global affine transformation to visual tokens and improves Libero viewpoint accuracy from 48.5% to 87.1% with only 4K parameters. Building on this, Feature Linear Adaptation (FLA) introduces low-rank updates to the ViT encoder, achieving 90.8% success with 4.7M parameters -- matching LoRA-scale finetuning at far lower cost. Together, these results reveal substantial untapped robustness in pretrained VLA models and demonstrate that targeted, minimal visual adaptation is sufficient to restore viewpoint generalization.

Paperid: 2726, https://arxiv.org/pdf/2511.20720.pdf

Abstract:
Vision-Language Action (VLA) models unify perception, reasoning, and trajectory generation for autonomous driving, but suffer from significant inference latency due to deep transformer stacks. We present DeeAD, a training-free, action-guided early-exit framework that accelerates VLA planning by evaluating the physical feasibility of intermediate trajectories. Instead of relying on confidence scores, DeeAD terminates inference when predicted trajectories align with lightweight planning priors (e.g., Navigation or Low-precision Planning) within a tolerable deviation (<2m). To improve efficiency, we introduce a multi-hop controller that adaptively skips redundant layers based on the change rate of scores. DeeAD integrates into existing VLA models, such as ORION, without requiring retraining. Experiments on the Bench2Drive benchmark demonstrate up to 28% transformer-layer sparsity and 29% latency reduction, while preserving planning quality and safety.

Paperid: 2727, https://arxiv.org/pdf/2511.11552.pdf

Abstract:
Comprehending long visual documents, where information is distributed across extensive pages of text and visual elements, is a critical but challenging task for modern Vision-Language Models (VLMs). Existing approaches falter on a fundamental challenge: evidence localization. They struggle to retrieve relevant pages and overlook fine-grained details within visual elements, leading to limited performance and model hallucination. To address this, we propose DocLens, a tool-augmented multi-agent framework that effectively ``zooms in'' on evidence like a lens. It first navigates from the full document to specific visual elements on relevant pages, then employs a sampling-adjudication mechanism to generate a single, reliable answer. Paired with Gemini-2.5-Pro, DocLens achieves state-of-the-art performance on MMLongBench-Doc and FinRAGBench-V, surpassing even human experts. The framework's superiority is particularly evident on vision-centric and unanswerable queries, demonstrating the power of its enhanced localization capabilities.

Paperid: 2728, https://arxiv.org/pdf/2511.10696.pdf

Abstract:
Transformers have revolutionized natural language processing, but their quadratic complexity with respect to sequence length remains a fundamental bottleneck for long-range modeling. While sparse attention mechanisms like RingAttention reduce computational costs by restricting attention to local neighborhoods, they suffer from limited receptive fields and lack of adaptability. We present \PiAttention, a periodic sparse Transformer that factorizes attention into ring-local neighborhoods, deterministic $π$-stride skips, and an adaptive fusion gate. The periodic structure provides predictable coverage of distant tokens, while the sparse footprint keeps the per-layer complexity linear in context length. We prove that \PiAttention achieves $\mathcal{O}(kL + π\log L)$ receptive field growth compared to $\mathcal{O}(kL)$ for RingAttention, where $k$ is the local window size, $π$ is the skip period, and $L$ is the sequence length. Extensive experiments on language modeling, retrieval, and vision-language tasks demonstrate that \PiAttention matches or surpasses dense attention quality with 8.3\% lower perplexity than RingAttention while using 50\% fewer GPUs for the same context length. Our detailed ablations and visualizations reveal the importance of periodic skips, adaptive fusion, and head-level sparsity coordination for efficient long-context modeling.

Paperid: 2729, https://arxiv.org/pdf/2510.25179.pdf

Abstract:
Agentic methods have emerged as a powerful and autonomous paradigm that enhances reasoning, collaboration, and adaptive control, enabling systems to coordinate and independently solve complex tasks. We extend this paradigm to safety alignment by introducing Agentic Moderation, a model-agnostic framework that leverages specialised agents to defend multimodal systems against jailbreak attacks. Unlike prior approaches that apply as a static layer over inputs or outputs and provide only binary classifications (safe or unsafe), our method integrates dynamic, cooperative agents, including Shield, Responder, Evaluator, and Reflector, to achieve context-aware and interpretable moderation. Extensive experiments across five datasets and four representative Large Vision-Language Models (LVLMs) demonstrate that our approach reduces the Attack Success Rate (ASR) by 7-19%, maintains a stable Non-Following Rate (NF), and improves the Refusal Rate (RR) by 4-20%, achieving robust, interpretable, and well-balanced safety performance. By harnessing the flexibility and reasoning capacity of agentic architectures, Agentic Moderation provides modular, scalable, and fine-grained safety enforcement, highlighting the broader potential of agentic systems as a foundation for automated safety governance.

Paperid: 2730, https://arxiv.org/pdf/2510.21794.pdf

Abstract:
Vision-Language Models (VLMs) have become essential backbones of modern multimodal intelligence, yet their outputs remain prone to hallucination-plausible text misaligned with visual inputs. Existing alignment approaches often rely on expensive fine-tuning with annotated preference data or sequence-level inference strategies that provide only coarse, delayed feedback. To overcome these limitations, we present TITA (Token-level Inference-Time Alignment), a lightweight framework that freezes the base VLM and instead trains a reward model to approximate its distribution. During inference, implicit preference signals are extracted as log-probability ratios between the reward model and the target VLM, yielding dense autoregressive feedback. This formulation can be viewed as an inference-time variant of Direct Preference Optimization (DPO), providing token-level corrective signals without retraining the backbone. Extensive evaluations on LLaVA-1.5-7B and 13B show consistent gains across 12 benchmarks, with improvements of 8.6% on MMVet and 6.7% on POPE, indicating stronger general understanding and reduced hallucinations. Additional experiments on Qwen2.5-VL-7B and DeepSeek-VL2-27.5B show comparable gains, especially in hallucination reduction and VQA accuracy, while incurring negligible inference overhead.

Paperid: 2731, https://arxiv.org/pdf/2510.18034.pdf

Abstract:
Autonomous driving systems remain critically vulnerable to the long-tail of rare, out-of-distribution scenarios with semantic anomalies. While Vision Language Models (VLMs) offer promising reasoning capabilities, naive prompting approaches yield unreliable performance and depend on expensive proprietary models, limiting practical deployment. We introduce SAVANT (Semantic Analysis with Vision-Augmented Anomaly deTection), a structured reasoning framework that achieves high accuracy and recall in detecting anomalous driving scenarios from input images through layered scene analysis and a two-phase pipeline: structured scene description extraction followed by multi-modal evaluation. Our approach transforms VLM reasoning from ad-hoc prompting to systematic analysis across four semantic layers: Street, Infrastructure, Movable Objects, and Environment. SAVANT achieves 89.6% recall and 88.0% accuracy on real-world driving scenarios, significantly outperforming unstructured baselines. More importantly, we demonstrate that our structured framework enables a fine-tuned 7B parameter open-source model (Qwen2.5VL) to achieve 90.8% recall and 93.8% accuracy - surpassing all models evaluated while enabling local deployment at near-zero cost. By automatically labeling over 9,640 real-world images with high accuracy, SAVANT addresses the critical data scarcity problem in anomaly detection and provides a practical path toward reliable, accessible semantic monitoring for autonomous systems.

Paperid: 2732, https://arxiv.org/pdf/2510.17354.pdf

Abstract:
Retrieval-Augmented Generation (RAG) has emerged as a powerful paradigm for enhancing large language models (LLMs) by retrieving relevant documents from an external corpus. However, existing RAG systems primarily focus on unimodal text documents, and often fall short in real-world scenarios where both queries and documents may contain mixed modalities (such as text and images). In this paper, we address the challenge of Universal Retrieval-Augmented Generation (URAG), which involves retrieving and reasoning over mixed-modal information to improve vision-language generation. To this end, we propose Nyx, a unified mixed-modal to mixed-modal retriever tailored for URAG scenarios. To mitigate the scarcity of realistic mixed-modal data, we introduce a four-stage automated pipeline for generation and filtering, leveraging web documents to construct NyxQA, a dataset comprising diverse mixed-modal question-answer pairs that better reflect real-world information needs. Building on this high-quality dataset, we adopt a two-stage training framework for Nyx: we first perform pre-training on NyxQA along with a variety of open-source retrieval datasets, followed by supervised fine-tuning using feedback from downstream vision-language models (VLMs) to align retrieval outputs with generative preferences. Experimental results demonstrate that Nyx not only performs competitively on standard text-only RAG benchmarks, but also excels in the more general and realistic URAG setting, significantly improving generation quality in vision-language tasks.

Paperid: 2733, https://arxiv.org/pdf/2510.15430.pdf

Abstract:
Despite extensive alignment efforts, Large Vision-Language Models (LVLMs) remain vulnerable to jailbreak attacks, posing serious safety risks. To address this, existing detection methods either learn attack-specific parameters, which hinders generalization to unseen attacks, or rely on heuristically sound principles, which limit accuracy and efficiency. To overcome these limitations, we propose Learning to Detect (LoD), a general framework that accurately detects unknown jailbreak attacks by shifting the focus from attack-specific learning to task-specific learning. This framework includes a Multi-modal Safety Concept Activation Vector module for safety-oriented representation learning and a Safety Pattern Auto-Encoder module for unsupervised attack classification. Extensive experiments show that our method achieves consistently higher detection AUROC on diverse unknown attacks while improving efficiency. The code is available at https://anonymous.4open.science/r/Learning-to-Detect-51CB.

Paperid: 2734, https://arxiv.org/pdf/2510.13190.pdf

Abstract:
Large Vision-Language Models (LVLMs) unlock powerful multimodal reasoning but also expand the attack surface, particularly through adversarial inputs that conceal harmful goals in benign prompts. We propose SHIELD, a lightweight, model-agnostic preprocessing framework that couples fine-grained safety classification with category-specific guidance and explicit actions (Block, Reframe, Forward). Unlike binary moderators, SHIELD composes tailored safety prompts that enforce nuanced refusals or safe redirection without retraining. Across five benchmarks and five representative LVLMs, SHIELD consistently lowers jailbreak and non-following rates while preserving utility. Our method is plug-and-play, incurs negligible overhead, and is easily extendable to new attack types -- serving as a practical safety patch for both weakly and strongly aligned LVLMs.

Paperid: 2735, https://arxiv.org/pdf/2510.12796.pdf

Abstract:
Scaling Vision-Language-Action (VLA) models on large-scale data offers a promising path to achieving a more generalized driving intelligence. However, VLA models are limited by a ``supervision deficit'': the vast model capacity is supervised by sparse, low-dimensional actions, leaving much of their representational power underutilized. To remedy this, we propose \textbf{DriveVLA-W0}, a training paradigm that employs world modeling to predict future images. This task generates a dense, self-supervised signal that compels the model to learn the underlying dynamics of the driving environment. We showcase the paradigm's versatility by instantiating it for two dominant VLA archetypes: an autoregressive world model for VLAs that use discrete visual tokens, and a diffusion world model for those operating on continuous visual features. Building on the rich representations learned from world modeling, we introduce a lightweight action expert to address the inference latency for real-time deployment. Extensive experiments on the NAVSIM v1/v2 benchmark and a 680x larger in-house dataset demonstrate that DriveVLA-W0 significantly outperforms BEV and VLA baselines. Crucially, it amplifies the data scaling law, showing that performance gains accelerate as the training dataset size increases.

Paperid: 2736, https://arxiv.org/pdf/2510.06207.pdf

Abstract:
Recent advances in control robot methods, from end-to-end vision-language-action frameworks to modular systems with predefined primitives, have advanced robots' ability to follow natural language instructions. Nonetheless, many approaches still struggle to scale to diverse environments, as they often rely on large annotated datasets and offer limited interpretability.In this work, we introduce EmbodiedCoder, a training-free framework for open-world mobile robot manipulation that leverages coding models to directly generate executable robot trajectories. By grounding high-level instructions in code, EmbodiedCoder enables flexible object geometry parameterization and manipulation trajectory synthesis without additional data collection or fine-tuning.This coding-based paradigm provides a transparent and generalizable way to connect perception with manipulation. Experiments on real mobile robots show that EmbodiedCoder achieves robust performance across diverse long-term tasks and generalizes effectively to novel objects and environments.Our results demonstrate an interpretable approach for bridging high-level reasoning and low-level control, moving beyond fixed primitives toward versatile robot intelligence. See the project page at: https://anonymous.4open.science/w/Embodied-Coder/

Paperid: 2737, https://arxiv.org/pdf/2510.01795.pdf

Abstract:
Vision-Language Models (VLMs) are increasingly applied in autonomous driving for unified perception and reasoning, but high inference latency hinders real-time deployment. Early-exit reduces latency by terminating inference at intermediate layers, yet its task-dependent nature limits generalization across diverse scenarios. We observe that this limitation aligns with autonomous driving: navigation systems can anticipate upcoming contexts (e.g., intersections, traffic lights), indicating which tasks will be required. We propose Nav-EE, a navigation-guided early-exit framework that precomputes task-specific exit layers offline and dynamically applies them online based on navigation priors. Experiments on CODA, Waymo, and BOSCH show that Nav-EE achieves accuracy comparable to full inference while reducing latency by up to 63.9%. Real-vehicle integration with Autoware Universe further demonstrates reduced inference latency (600ms to 300ms), supporting faster decision-making in complex scenarios. These results suggest that coupling navigation foresight with early-exit offers a viable path toward efficient deployment of large models in autonomous systems. Code and data are available at our anonymous repository: https://anonymous.4open.science/r/Nav-EE-BBC4

Paperid: 2738, https://arxiv.org/pdf/2509.25944.pdf

Abstract:
Understanding risk in autonomous driving requires not only perception and prediction, but also high-level reasoning about agent behavior and context. Current Vision Language Models (VLMs)-based methods primarily ground agents in static images and provide qualitative judgments, lacking the spatio-temporal reasoning needed to capture how risks evolve over time. To address this gap, we propose NuRisk, a comprehensive Visual Question Answering (VQA) dataset comprising 2,900 scenarios and 1.1 million agent-level samples, built on real-world data from nuScenes and Waymo, supplemented with safety-critical scenarios from the CommonRoad simulator. The dataset provides Bird-Eye-View (BEV) based sequential images with quantitative, agent-level risk annotations, enabling spatio-temporal reasoning. We benchmark well-known VLMs across different prompting techniques and find that they fail to perform explicit spatio-temporal reasoning, resulting in a peak accuracy of 33% at high latency. To address these shortcomings, our fine-tuned 7B VLM agent improves accuracy to 41% and reduces latency by 75%, demonstrating explicit spatio-temporal reasoning capabilities that proprietary models lacked. While this represents a significant step forward, the modest accuracy underscores the profound challenge of the task, establishing NuRisk as a critical benchmark for advancing spatio-temporal reasoning in autonomous driving.

Paperid: 2739, https://arxiv.org/pdf/2508.20020.pdf

Abstract:
Language-driven image segmentation is a fundamental task in vision-language understanding, requiring models to segment regions of an image corresponding to natural language expressions. Traditional methods approach this as a discriminative problem, assigning each pixel to foreground or background based on semantic alignment. Recently, diffusion models have been introduced to this domain, but existing approaches remain image-centric: they either (i) use image diffusion models as visual feature extractors, (ii) synthesize segmentation data via image generation to train discriminative models, or (iii) perform diffusion inversion to extract attention cues from pre-trained image diffusion models-thereby treating segmentation as an auxiliary process. In this paper, we propose GS (Generative Segmentation), a novel framework that formulates segmentation itself as a generative task via label diffusion. Instead of generating images conditioned on label maps and text, GS reverses the generative process: it directly generates segmentation masks from noise, conditioned on both the input image and the accompanying language description. This paradigm makes label generation the primary modeling target, enabling end-to-end training with explicit control over spatial and semantic fidelity. To demonstrate the effectiveness of our approach, we evaluate GS on Panoptic Narrative Grounding (PNG), a representative and challenging benchmark for multimodal segmentation that requires panoptic-level reasoning guided by narrative captions. Experimental results show that GS significantly outperforms existing discriminative and diffusion-based methods, setting a new state-of-the-art for language-driven segmentation.

Paperid: 2740, https://arxiv.org/pdf/2508.19165.pdf

Abstract:
Monocular 3D visual grounding is a novel task that aims to locate 3D objects in RGB images using text descriptions with explicit geometry information. Despite the inclusion of geometry details in the text, we observe that the text embeddings are sensitive to the magnitude of numerical values but largely ignore the associated measurement units. For example, simply equidistant mapping the length with unit "meter" to "decimeters" or "centimeters" leads to severe performance degradation, even though the physical length remains equivalent. This observation signifies the weak 3D comprehension of pre-trained language model, which generates misguiding text features to hinder 3D perception. Therefore, we propose to enhance the 3D perception of model on text embeddings and geometry features with two simple and effective methods. Firstly, we introduce a pre-processing method named 3D-text Enhancement (3DTE), which enhances the comprehension of mapping relationships between different units by augmenting the diversity of distance descriptors in text queries. Next, we propose a Text-Guided Geometry Enhancement (TGE) module to further enhance the 3D-text information by projecting the basic text features into geometrically consistent space. These 3D-enhanced text features are then leveraged to precisely guide the attention of geometry features. We evaluate the proposed method through extensive comparisons and ablation studies on the Mono3DRefer dataset. Experimental results demonstrate substantial improvements over previous methods, achieving new state-of-the-art results with a notable accuracy gain of 11.94\% in the "Far" scenario. Our code will be made publicly available.

Paperid: 2741, https://arxiv.org/pdf/2508.18227.pdf

Abstract:
Transformer-based Vision-Language Models (VLMs) have achieved impressive performance on tasks such as image captioning, object recognition, and visual reasoning, but their high computational cost hinders deployment in latency-sensitive applications like autonomous driving. We introduce GM-Skip, a flexible and metric-adaptive framework for Transformer block skipping that accelerates VLM inference while preserving output quality. GM-Skip features a greedy, metric-guided block selection strategy that uses metric feedback (e.g., accuracy, CIDEr) to identify redundant layers, along with a reverse-order deletion mechanism that preserves early foundational blocks to avoid performance collapse. To support diverse deployment needs, it incorporates a tunable trade-off between sparsity and performance via a score-sparsity balance objective. Experiments across multiple tasks and datasets, including COCO and CODA, show that GM-Skip consistently improves inference speed while maintaining task performance. On the COCO dataset, GM-Skip improves single-object classification accuracy on the Person category from 19.1 percent to 87.3 percent while skipping more than 40 percent of Transformer blocks. In real-world deployment, it achieves up to 45.4 percent latency reduction on single-object detection when integrated into an autonomous vehicle running Autoware.Universe, validating the effectiveness of its skip configurations and confirming its practical value in accelerating real-world inference.

Paperid: 2742, https://arxiv.org/pdf/2508.11262.pdf

Abstract:
Vision-language models (VLM) align images and text in a shared representation space that is useful for retrieval and zero-shot transfer. Yet, this alignment can encode and amplify social stereotypes in subtle ways that are not obvious from standard accuracy metrics. In this study, we test whether the contrastive vision-language encoder exhibits gender-linked associations when it places embeddings of face images near embeddings of short phrases that describe occupations and activities. We assemble a dataset of 220 face photographs split by perceived binary gender and a set of 150 unique statements distributed across six categories covering emotional labor, cognitive labor, domestic labor, technical labor, professional roles, and physical labor. We compute unit-norm image embeddings for every face and unit-norm text embeddings for every statement, then define a statement-level association score as the difference between the mean cosine similarity to the male set and the mean cosine similarity to the female set, where positive values indicate stronger association with the male set and negative values indicate stronger association with the female set. We attach bootstrap confidence intervals by resampling images within each gender group, aggregate by category with a separate bootstrap over statements, and run a label-swap null model that estimates the level of mean absolute association we would expect if no gender structure were present. The outcome is a statement-wise and category-wise map of gender associations in a contrastive vision-language space, accompanied by uncertainty, simple sanity checks, and a robust gender bias evaluation framework.

Paperid: 2743, https://arxiv.org/pdf/2508.09201.pdf

Abstract:
Despite extensive alignment efforts, Large Vision-Language Models (LVLMs) remain vulnerable to jailbreak attacks, posing serious safety risks. Although recent detection works have shifted to internal representations due to their rich cross-modal information, most methods rely on heuristic rules rather than principled objectives, resulting in suboptimal performance. To address these limitations, we propose Learning to Detect (LoD), a novel unsupervised framework that formulates jailbreak detection as anomaly detection. LoD introduces two key components: Multi-modal Safety Concept Activation Vectors (MSCAV), which capture layer-wise safety-related representations across modalities, and the Safety Pattern Auto-Encoder, which models the distribution of MSCAV derived from safe inputs and detects anomalies via reconstruction errors. By training the auto-encoder (AE) solely on safe samples without attack labels, LoD naturally identifies jailbreak inputs as distributional anomalies, enabling accurate and unified detection of jailbreak attacks. Comprehensive experiments on three different LVLMs and five benchmarks demonstrate that LoD achieves state-of-the-art performance, with an average AUROC of 0.9951 and an improvement of up to 38.89% in the minimum AUROC over the strongest baselines.

Paperid: 2744, https://arxiv.org/pdf/2507.20622.pdf

Abstract:
Visual imitation learning (VIL) provides an efficient and intuitive strategy for robotic systems to acquire novel skills. Recent advancements in foundation models, particularly Vision Language Models (VLMs), have demonstrated remarkable capabilities in visual and linguistic reasoning for VIL tasks. Despite this progress, existing approaches primarily utilize these models for learning high-level plans from human demonstrations, relying on pre-defined motion primitives for executing physical interactions, which remains a major bottleneck for robotic systems. In this work, we present FMimic, a novel paradigm that harnesses foundation models to directly learn generalizable skills at even fine-grained action levels, using only a limited number of human videos. Extensive experiments demonstrate that our FMimic delivers strong performance with a single human video, and significantly outperforms all other methods with five videos. Furthermore, our method exhibits significant improvements of over 39% and 29% in RLBench multi-task experiments and real-world manipulation tasks, respectively, and exceeds baselines by more than 34% in high-precision tasks and 47% in long-horizon tasks.

Paperid: 2745, https://arxiv.org/pdf/2507.01925.pdf

Abstract:
The remarkable advancements of vision and language foundation models in multimodal understanding, reasoning, and generation has sparked growing efforts to extend such intelligence to the physical world, fueling the flourishing of vision-language-action (VLA) models. Despite seemingly diverse approaches, we observe that current VLA models can be unified under a single framework: vision and language inputs are processed by a series of VLA modules, producing a chain of \textit{action tokens} that progressively encode more grounded and actionable information, ultimately generating executable actions. We further determine that the primary design choice distinguishing VLA models lies in how action tokens are formulated, which can be categorized into language description, code, affordance, trajectory, goal state, latent representation, raw action, and reasoning. However, there remains a lack of comprehensive understanding regarding action tokens, significantly impeding effective VLA development and obscuring future directions. Therefore, this survey aims to categorize and interpret existing VLA research through the lens of action tokenization, distill the strengths and limitations of each token type, and identify areas for improvement. Through this systematic review and analysis, we offer a synthesized outlook on the broader evolution of VLA models, highlight underexplored yet promising directions, and contribute guidance for future research, hoping to bring the field closer to general-purpose intelligence.

Paperid: 2746, https://arxiv.org/pdf/2506.14512.pdf

Abstract:
Large Language Models (LLMs) are experiencing rapid advancements in complex reasoning, exhibiting remarkable generalization in mathematics and programming. In contrast, while spatial intelligence is fundamental for Vision-Language Models (VLMs) in real-world interaction, the systematic evaluation of their complex reasoning ability within spatial contexts remains underexplored. To bridge this gap, we introduce SIRI-Bench, a benchmark designed to evaluate VLMs' spatial intelligence through video-based reasoning tasks. SIRI-Bench comprises nearly 1K video-question-answer triplets, where each problem is embedded in a realistic 3D scene and captured by video. By carefully designing questions and corresponding 3D scenes, our benchmark ensures that solving the questions requires both spatial comprehension for extracting information and high-level reasoning for deriving solutions, making it a challenging benchmark for evaluating VLMs. To facilitate large-scale data synthesis, we develop an Automatic Scene Creation Engine. This engine, leveraging multiple specialized LLM agents, can generate realistic 3D scenes from abstract math problems, ensuring faithfulness to the original descriptions. Experimental results reveal that state-of-the-art VLMs struggle significantly on SIRI-Bench, underscoring the challenge of spatial reasoning. We hope that our study will bring researchers' attention to spatially grounded reasoning and advance VLMs in visual problem-solving.

Paperid: 2747, https://arxiv.org/pdf/2506.05404.pdf

Abstract:
With the rapid advancement of autonomous driving, deploying Vision-Language Models (VLMs) to enhance perception and decision-making has become increasingly common. However, the real-time application of VLMs is hindered by high latency and computational overhead, limiting their effectiveness in time-critical driving scenarios. This challenge is particularly evident when VLMs exhibit over-inference, continuing to process unnecessary layers even after confident predictions have been reached. To address this inefficiency, we propose AD-EE, an Early Exit framework that incorporates domain characteristics of autonomous driving and leverages causal inference to identify optimal exit layers. We evaluate our method on large-scale real-world autonomous driving datasets, including Waymo and the corner-case-focused CODA, as well as on a real vehicle running the Autoware Universe platform. Extensive experiments across multiple VLMs show that our method significantly reduces latency, with maximum improvements reaching up to 57.58%, and enhances object detection accuracy, with maximum gains of up to 44%.

Paperid: 2748, https://arxiv.org/pdf/2506.04559.pdf

Abstract:
Recent advances in slow-thinking language models (e.g., OpenAI-o1 and DeepSeek-R1) have demonstrated remarkable abilities in complex reasoning tasks by emulating human-like reflective cognition. However, extending such capabilities to multi-modal large language models (MLLMs) remains challenging due to the high cost of retraining vision-language alignments when upgrading the underlying reasoner LLMs. A straightforward solution is to decouple perception from reasoning, i.e., converting visual inputs into language representations (e.g., captions) that are then passed to a powerful text-only reasoner. However, this decoupling introduces a critical challenge: the visual extractor must generate descriptions that are both faithful to the image and informative enough to support accurate downstream reasoning. To address this, we propose Reasoning-Aligned Perceptual Decoupling via Caption Reward Optimization (RACRO) - a reasoning-guided reinforcement learning strategy that aligns the extractor's captioning behavior with the reasoning objective. By closing the perception-reasoning loop via reward-based optimization, RACRO significantly enhances visual grounding and extracts reasoning-optimized representations. Experiments on multi-modal math and science benchmarks show that the proposed RACRO method achieves state-of-the-art average performance while enabling superior scalability and plug-and-play adaptation to more advanced reasoning LLMs without the necessity for costly multi-modal re-alignment.

Paperid: 2749, https://arxiv.org/pdf/2505.21967.pdf

Abstract:
Large Vision-Language Models (LVLMs) have shown remarkable capabilities across a wide range of multimodal tasks. However, their integration of visual inputs introduces expanded attack surfaces, thereby exposing them to novel security vulnerabilities. In this work, we conduct a systematic representational analysis to uncover why conventional adversarial attacks can circumvent the safety mechanisms embedded in LVLMs. We further propose a novel two stage evaluation framework for adversarial attacks on LVLMs. The first stage differentiates among instruction non compliance, outright refusal, and successful adversarial exploitation. The second stage quantifies the degree to which the model's output fulfills the harmful intent of the adversarial prompt, while categorizing refusal behavior into direct refusals, soft refusals, and partial refusals that remain inadvertently helpful. Finally, we introduce a normative schema that defines idealized model behavior when confronted with harmful prompts, offering a principled target for safety alignment in multimodal systems.

Paperid: 2750, https://arxiv.org/pdf/2504.16072.pdf

Abstract:
Generating detailed and accurate descriptions for specific regions in images and videos remains a fundamental challenge for vision-language models. We introduce the Describe Anything Model (DAM), a model designed for detailed localized captioning (DLC). DAM preserves both local details and global context through two key innovations: a focal prompt, which ensures high-resolution encoding of targeted regions, and a localized vision backbone, which integrates precise localization with its broader context. To tackle the scarcity of high-quality DLC data, we propose a Semi-supervised learning (SSL)-based Data Pipeline (DLC-SDP). DLC-SDP starts with existing segmentation datasets and expands to unlabeled web images using SSL. We introduce DLC-Bench, a benchmark designed to evaluate DLC without relying on reference captions. DAM sets new state-of-the-art on 7 benchmarks spanning keyword-level, phrase-level, and detailed multi-sentence localized image and video captioning.

Paperid: 2751, https://arxiv.org/pdf/2504.09859.pdf

Abstract:
Graph visualizations have been studied for tasks such as clustering and temporal analysis, but how these visual similarities relate to established graph similarity measures remains unclear. In this paper, we explore the potential of Vision Language Models (VLMs) to approximate human-like perception of graph similarity. We generate graph datasets of various sizes and densities and compare VLM-derived visual similarity scores with feature-based measures. Our findings indicate VLMs can assess graph similarity in a manner similar to feature-based measures, even though differences among the measures exist. In future work, we plan to extend our research by conducting experiments on human visual graph perception.

Paperid: 2752, https://arxiv.org/pdf/2503.23046.pdf

Abstract:
With the widespread adoption and deployment of autonomous driving, handling complex environments has become an unavoidable challenge. Due to the scarcity and diversity of extreme scenario datasets, current autonomous driving models struggle to effectively manage corner cases. This limitation poses a significant safety risk, according to the National Highway Traffic Safety Administration (NHTSA), autonomous vehicle systems have been involved in hundreds of reported crashes annually in the United States, occurred in corner cases like sun glare and fog, which caused a few fatal accident. Furthermore, in order to consistently maintain a robust and reliable autonomous driving system, it is essential for models not only to perform well on routine scenarios but also to adapt to newly emerging scenarios, especially those corner cases that deviate from the norm. This requires a learning mechanism that incrementally integrates new knowledge without degrading previously acquired capabilities. However, to the best of our knowledge, no existing continual learning methods have been proposed to ensure consistent and scalable corner case learning in autonomous driving. To address these limitations, we propose VLM-C4L, a continual learning framework that introduces Vision-Language Models (VLMs) to dynamically optimize and enhance corner case datasets, and VLM-C4L combines VLM-guided high-quality data extraction with a core data replay strategy, enabling the model to incrementally learn from diverse corner cases while preserving performance on previously routine scenarios, thus ensuring long-term stability and adaptability in real-world autonomous driving. We evaluate VLM-C4L on large-scale real-world autonomous driving datasets, including Waymo and the corner case dataset CODA.

Paperid: 2753, https://arxiv.org/pdf/2503.12990.pdf

Abstract:
Recently, histopathology vision-language foundation models (VLMs) have gained popularity due to their enhanced performance and generalizability across different downstream tasks. However, most existing histopathology benchmarks are either unimodal or limited in terms of diversity of clinical tasks, organs, and acquisition instruments, as well as their partial availability to the public due to patient data privacy. As a consequence, there is a lack of comprehensive evaluation of existing histopathology VLMs on a unified benchmark setting that better reflects a wide range of clinical scenarios. To address this gap, we introduce HistoVL, a fully open-source comprehensive benchmark comprising images acquired using up to 11 various acquisition tools that are paired with specifically crafted captions by incorporating class names and diverse pathology descriptions. Our Histo-VL includes 26 organs, 31 cancer types, and a wide variety of tissue obtained from 14 heterogeneous patient cohorts, totaling more than 5 million patches obtained from over 41K WSIs viewed under various magnification levels. We systematically evaluate existing histopathology VLMs on Histo-VL to simulate diverse tasks performed by experts in real-world clinical scenarios. Our analysis reveals interesting findings, including large sensitivity of most existing histopathology VLMs to textual changes with a drop in balanced accuracy of up to 25% in tasks such as Metastasis detection, low robustness to adversarial attacks, as well as improper calibration of models evident through high ECE values and low model prediction confidence, all of which can affect their clinical implementation.

Paperid: 2754, https://arxiv.org/pdf/2503.01632.pdf

Abstract:
With the acceleration of urbanization, modern urban traffic systems are becoming increasingly complex, leading to frequent traffic anomalies. These anomalies encompass not only common traffic jams but also more challenging issues such as phantom traffic jams, intersection deadlocks, and accident liability analysis, which severely impact traffic flow, vehicular safety, and overall transportation efficiency. Currently, existing solutions primarily rely on manual intervention by traffic police or artificial intelligence-based detection systems. However, these methods often suffer from response delays and inconsistent management due to inadequate resources, while AI detection systems, despite enhancing efficiency to some extent, still struggle to handle complex traffic anomalies in a real-time and precise manner. To address these issues, we propose CoT-VLM4Tar: (Chain of Thought Visual-Language Model for Traffic Anomaly Resolution), this innovative approach introduces a new chain-of-thought to guide the VLM in analyzing, reasoning, and generating solutions for traffic anomalies with greater reasonable and effective solution, and to evaluate the performance and effectiveness of our method, we developed a closed-loop testing framework based on the CARLA simulator. Furthermore, to ensure seamless integration of the solutions generated by the VLM with the CARLA simulator, we implement an itegration module that converts these solutions into executable commands. Our results demonstrate the effectiveness of VLM in the resolution of real-time traffic anomalies, providing a proof-of-concept for its integration into autonomous traffic management systems.

Paperid: 2755, https://arxiv.org/pdf/2502.20900.pdf

Abstract:
Dexterous grasping remains a fundamental yet challenging problem in robotics. A general-purpose robot must be capable of grasping diverse objects in arbitrary scenarios. However, existing research typically relies on restrictive assumptions, such as single-object settings or limited environments, showing constrained generalization. We present DexGraspVLA, a hierarchical framework for robust generalization in language-guided general dexterous grasping and beyond. It utilizes a pre-trained Vision-Language model as the high-level planner and learns a diffusion-based low-level Action controller. The key insight to achieve generalization lies in iteratively transforming diverse language and visual inputs into domain-invariant representations via foundation models, where imitation learning can be effectively applied due to the alleviation of domain shift. Notably, our method achieves a 90+% dexterous grasping success rate under thousands of challenging unseen cluttered scenes. Empirical analysis confirms the consistency of internal model behavior across environmental variations, validating our design. DexGraspVLA also, for the first time, simultaneously demonstrates free-form long-horizon prompt execution, robustness to adversarial objects and human disturbance, and failure recovery. Extended application to nonprehensile grasping further proves its generality. Project website: https://dexgraspvla.github.io.

Paperid: 2756, https://arxiv.org/pdf/2502.19697.pdf

Abstract:
Person re-identification (re-id) models are vital in security surveillance systems, requiring transferable adversarial attacks to explore the vulnerabilities of them. Recently, vision-language models (VLM) based attacks have shown superior transferability by attacking generalized image and textual features of VLM, but they lack comprehensive feature disruption due to the overemphasis on discriminative semantics in integral representation. In this paper, we introduce the Attribute-aware Prompt Attack (AP-Attack), a novel method that leverages VLM's image-text alignment capability to explicitly disrupt fine-grained semantic features of pedestrian images by destroying attribute-specific textual embeddings. To obtain personalized textual descriptions for individual attributes, textual inversion networks are designed to map pedestrian images to pseudo tokens that represent semantic embeddings, trained in the contrastive learning manner with images and a predefined prompt template that explicitly describes the pedestrian attributes. Inverted benign and adversarial fine-grained textual semantics facilitate attacker in effectively conducting thorough disruptions, enhancing the transferability of adversarial examples. Extensive experiments show that AP-Attack achieves state-of-the-art transferability, significantly outperforming previous methods by 22.9% on mean Drop Rate in cross-model&dataset attack scenarios.

Paperid: 2757, https://arxiv.org/pdf/2502.16528.pdf

Abstract:
In recent years, vision-language models (VLMs) have advanced open-vocabulary mapping, enabling mobile robots to simultaneously achieve environmental reconstruction and high-level semantic understanding. While integrated object cognition helps mitigate semantic ambiguity in point-wise feature maps, efficiently obtaining rich semantic understanding and robust incremental reconstruction at the instance-level remains challenging. To address these challenges, we introduce OpenVox, a real-time incremental open-vocabulary probabilistic instance voxel representation. In the front-end, we design an efficient instance segmentation and comprehension pipeline that enhances language reasoning through encoding captions. In the back-end, we implement probabilistic instance voxels and formulate the cross-frame incremental fusion process into two subtasks: instance association and live map evolution, ensuring robustness to sensor and segmentation noise. Extensive evaluations across multiple datasets demonstrate that OpenVox achieves state-of-the-art performance in zero-shot instance segmentation, semantic segmentation, and open-vocabulary retrieval. Furthermore, real-world robotics experiments validate OpenVox's capability for stable, real-time operation.

Paperid: 2758, https://arxiv.org/pdf/2502.07838.pdf

Abstract:
Vision-Language Models (VLMs), such as GPT-4V and Llama 3.2 vision, have garnered significant research attention for their ability to leverage Large Language Models (LLMs) in multimodal tasks. However, their potential is constrained by inherent challenges, including proprietary restrictions, substantial computational demands, and limited accessibility. Smaller models, such as GIT and BLIP, exhibit marked limitations, often failing to generate coherent and consistent text beyond a few tokens, even with extensive training. This underscores a pivotal inquiry: how small can a VLM be and still produce fluent and consistent text? Drawing inspiration from the exceptional learning process of 3-4 year old children, who rely heavily on visual cues for understanding and communication, we introduce two novel datasets: ShortDesc (featuring concise image descriptions) and LongDesc (containing more detailed image descriptions). These datasets consist of image-text pairs where the text is restricted to the simple vocabulary and syntax typically used by young children, generated with a scaled-down model, GPT-4o. Using these datasets, we demonstrate that it is possible to train VLMs that are significantly smaller, up to 10 times smaller than state of the art(SOTA) small VLMs while maintaining architectural simplicity. To evaluate the outputs, we leverage GPT-4o to grade the text, as if stories written by students, on creativity, meaningfulness, and consistency, assigning scores out of 10. This method addresses limitations of standard benchmarks by accommodating unstructured outputs and providing a multidimensional evaluation of the model capabilities. Our findings contribute to the development of lightweight, accessible multimodal models for resource constrained environments.

Paperid: 2759, https://arxiv.org/pdf/2412.09521.pdf

Abstract:
Pathological diagnosis is vital for determining disease characteristics, guiding treatment, and assessing prognosis, relying heavily on detailed, multi-scale analysis of high-resolution whole slide images (WSI). However, existing large vision-language models (LVLMs) are limited by input resolution constraints, hindering their efficiency and accuracy in pathology image analysis. To overcome these issues, we propose two innovative strategies: the mixed task-guided feature enhancement, which directs feature extraction toward lesion-related details across scales, and the prompt-guided detail feature completion, which integrates coarse- and fine-grained features from WSI based on specific prompts without compromising inference speed. Leveraging a comprehensive dataset of 490K samples from diverse pathology tasks, we trained the pathology-specialized LVLM, OmniPath. Extensive experiments demonstrate that this model significantly outperforms existing methods in diagnostic accuracy and efficiency, providing an interactive, clinically aligned approach for auxiliary diagnosis in a wide range of pathology applications.

Paperid: 2760, https://arxiv.org/pdf/2411.02437.pdf

Abstract:
Evaluating text-to-image generative models remains a challenge, despite the remarkable progress being made in their overall performances. While existing metrics like CLIPScore work for coarse evaluations, they lack the sensitivity to distinguish finer differences as model performance rapidly improves. In this work, we focus on the text rendering aspect of these models, which provides a lens for evaluating a generative model's fine-grained instruction-following capabilities. To this end, we introduce a new evaluation framework called TypeScore to sensitively assess a model's ability to generate images with high-fidelity embedded text by following precise instructions. We argue that this text generation capability serves as a proxy for general instruction-following ability in image synthesis. TypeScore uses an additional image description model and leverages an ensemble dissimilarity measure between the original and extracted text to evaluate the fidelity of the rendered text. Our proposed metric demonstrates greater resolution than CLIPScore to differentiate popular image generation models across a range of instructions with diverse text styles. Our study also evaluates how well these vision-language models (VLMs) adhere to stylistic instructions, disentangling style evaluation from embedded-text fidelity. Through human evaluation studies, we quantitatively meta-evaluate the effectiveness of the metric. Comprehensive analysis is conducted to explore factors such as text length, captioning models, and current progress towards human parity on this task. The framework provides insights into remaining gaps in instruction-following for image generation with embedded text.

Paperid: 2761, https://arxiv.org/pdf/2410.20927.pdf

Abstract:
Visual imitation learning (VIL) provides an efficient and intuitive strategy for robotic systems to acquire novel skills. Recent advancements in Vision Language Models (VLMs) have demonstrated remarkable performance in vision and language reasoning capabilities for VIL tasks. Despite the progress, current VIL methods naively employ VLMs to learn high-level plans from human videos, relying on pre-defined motion primitives for executing physical interactions, which remains a major bottleneck. In this work, we present VLMimic, a novel paradigm that harnesses VLMs to directly learn even fine-grained action levels, only given a limited number of human videos. Specifically, VLMimic first grounds object-centric movements from human videos, and learns skills using hierarchical constraint representations, facilitating the derivation of skills with fine-grained action levels from limited human videos. These skills are refined and updated through an iterative comparison strategy, enabling efficient adaptation to unseen environments. Our extensive experiments exhibit that our VLMimic, using only 5 human videos, yields significant improvements of over 27% and 21% in RLBench and real-world manipulation tasks, and surpasses baselines by over 37% in long-horizon tasks.

Paperid: 2762, https://arxiv.org/pdf/2601.20088.pdf

Abstract:
This technical report presents quantization-aware distillation (QAD) and our best practices for recovering accuracy of NVFP4-quantized large language models (LLMs) and vision-language models (VLMs). QAD distills a full-precision teacher model into a quantized student model using a KL divergence loss. While applying distillation to quantized models is not a new idea, we observe key advantages of QAD for today's LLMs: 1. It shows remarkable effectiveness and stability for models trained through multi-stage post-training pipelines, including supervised fine-tuning (SFT), reinforcement learning (RL), and model merging, where traditional quantization-aware training (QAT) suffers from engineering complexity and training instability; 2. It is robust to data quality and coverage, enabling accuracy recovery without full training data. We evaluate QAD across multiple post-trained models including AceReason Nemotron, Nemotron 3 Nano, Nemotron Nano V2, Nemotron Nano V2 VL (VLM), and Llama Nemotron Super v1, showing consistent recovery to near-BF16 accuracy.

Paperid: 2763, https://arxiv.org/pdf/2601.05640.pdf

Abstract:
Recent end-to-end autonomous driving approaches have leveraged Vision-Language Models (VLMs) to enhance planning capabilities in complex driving scenarios. However, VLMs are inherently trained as generalist models, lacking specialized understanding of driving-specific reasoning in 3D space and time. When applied to autonomous driving, these models struggle to establish structured spatial-temporal representations that capture geometric relationships, scene context, and motion patterns critical for safe trajectory planning. To address these limitations, we propose SGDrive, a novel framework that explicitly structures the VLM's representation learning around driving-specific knowledge hierarchies. Built upon a pre-trained VLM backbone, SGDrive decomposes driving understanding into a scene-agent-goal hierarchy that mirrors human driving cognition: drivers first perceive the overall environment (scene context), then attend to safety-critical agents and their behaviors, and finally formulate short-term goals before executing actions. This hierarchical decomposition provides the structured spatial-temporal representation that generalist VLMs lack, integrating multi-level information into a compact yet comprehensive format for trajectory planning. Extensive experiments on the NAVSIM benchmark demonstrate that SGDrive achieves state-of-the-art performance among camera-only methods on both PDMS and EPDMS, validating the effectiveness of hierarchical knowledge structuring for adapting generalist VLMs to autonomous driving.

Paperid: 2764, https://arxiv.org/pdf/2512.19021.pdf

Abstract:
Despite remarkable progress in Vision-Language Navigation (VLN), existing benchmarks remain confined to fixed, small-scale datasets with naive physical simulation. These shortcomings limit the insight that the benchmarks provide into sim-to-real generalization, and create a significant research gap. Furthermore, task fragmentation prevents unified/shared progress in the area, while limited data scales fail to meet the demands of modern LLM-based pretraining. To overcome these limitations, we introduce VLNVerse: a new large-scale, extensible benchmark designed for Versatile, Embodied, Realistic Simulation, and Evaluation. VLNVerse redefines VLN as a scalable, full-stack embodied AI problem. Its Versatile nature unifies previously fragmented tasks into a single framework and provides an extensible toolkit for researchers. Its Embodied design moves beyond intangible and teleporting "ghost" agents that support full-kinematics in a Realistic Simulation powered by a robust physics engine. We leverage the scale and diversity of VLNVerse to conduct a comprehensive Evaluation of existing methods, from classic models to MLLM-based agents. We also propose a novel unified multi-task model capable of addressing all tasks within the benchmark. VLNVerse aims to narrow the gap between simulated navigation and real-world generalization, providing the community with a vital tool to boost research towards scalable, general-purpose embodied locomotion agents.

Paperid: 2765, https://arxiv.org/pdf/2512.18117.pdf

Abstract:
The rapid growth of e-commerce requires robust multimodal representations that capture diverse signals from user-generated listings. Existing vision-language models (VLMs) typically align titles with primary images, i.e., single-view, but overlook non-primary images and auxiliary textual views that provide critical semantics in open marketplaces such as Etsy or Poshmark. To this end, we propose a framework that unifies multimodal and multi-view learning through Factorized Transport, a lightweight approximation of optimal transport, designed for scalability and deployment efficiency. During training, the method emphasizes primary views while stochastically sampling auxiliary ones, reducing training cost from quadratic in the number of views to constant per item. At inference, all views are fused into a single cached embedding, preserving the efficiency of two-tower retrieval with no additional online overhead. On an industrial dataset of 1M product listings and 0.3M interactions, our approach delivers consistent improvements in cross-view and query-to-item retrieval, achieving up to +7.9% Recall@500 over strong multimodal baselines. Overall, our framework bridges scalability with optimal transport-based learning, making multi-view pretraining practical for large-scale e-commerce search.

Paperid: 2766, https://arxiv.org/pdf/2512.15840.pdf

Abstract:
General-purpose robots require decision-making models that generalize across diverse tasks and environments. Recent works build robot foundation models by extending multimodal large language models (MLLMs) with action outputs, creating vision-language-action (VLA) systems. These efforts are motivated by the intuition that MLLMs' large-scale language and image pretraining can be effectively transferred to the action output modality. In this work, we explore an alternative paradigm of using large-scale video pretraining as a primary modality for building robot foundation models. Unlike static images and language, videos capture spatio-temporal sequences of states and actions in the physical world that are naturally aligned with robotic behavior. We curate an internet-scale video dataset of human activities and task demonstrations, and train, for the first time at a foundation-model scale, an open video model for generative robotics planning. The model produces zero-shot video plans for novel scenes and tasks, which we post-process to extract executable robot actions. We evaluate task-level generalization through third-party selected tasks in the wild and real-robot experiments, demonstrating successful physical execution. Together, these results show robust instruction following, strong generalization, and real-world feasibility. We release both the model and dataset to support open, reproducible video-based robot learning. Our website is available at https://www.boyuan.space/large-video-planner/.

Paperid: 2767, https://arxiv.org/pdf/2512.11047.pdf

Abstract:
Humanoid robots require precise locomotion and dexterous manipulation to perform challenging loco-manipulation tasks. Yet existing approaches, modular or end-to-end, are deficient in manipulation-aware locomotion. This confines the robot to a limited workspace, preventing it from performing large-space loco-manipulation. We attribute this to: (1) the challenge of acquiring loco-manipulation knowledge due to the scarcity of humanoid teleoperation data, and (2) the difficulty of faithfully and reliably executing locomotion commands, stemming from the limited precision and stability of existing RL controllers. To acquire richer loco-manipulation knowledge, we propose a unified latent learning framework that enables Vision-Language-Action (VLA) system to learn from low-cost action-free egocentric videos. Moreover, an efficient human data collection pipeline is devised to augment the dataset and scale the benefits. To execute the desired locomotion commands more precisely, we present a loco-manipulation-oriented (LMO) RL policy specifically tailored for accurate and stable core loco-manipulation movements, such as advancing, turning, and squatting. Building on these components, we introduce WholeBodyVLA, a unified framework for humanoid loco-manipulation. To the best of our knowledge, WholeBodyVLA is one of its kind enabling large-space humanoid loco-manipulation. It is verified via comprehensive experiments on the AgiBot X2 humanoid, outperforming prior baseline by 21.3%. It also demonstrates strong generalization and high extensibility across a broad range of tasks.

Paperid: 2768, https://arxiv.org/pdf/2512.04733.pdf

Abstract:
End-to-end autonomous driving (AD) systems increasingly adopt vision-language-action (VLA) models, yet they typically ignore the passenger's emotional state, which is central to comfort and AD acceptance. We introduce Open-Domain End-to-End (OD-E2E) autonomous driving, where an autonomous vehicle (AV) must interpret free-form natural-language commands, infer the emotion, and plan a physically feasible trajectory. We propose E3AD, an emotion-aware VLA framework that augments semantic understanding with two cognitively inspired components: a continuous Valenc-Arousal-Dominance (VAD) emotion model that captures tone and urgency from language, and a dual-pathway spatial reasoning module that fuses egocentric and allocentric views for human-like spatial cognition. A consistency-oriented training scheme, combining modality pretraining with preference-based alignment, further enforces coherence between emotional intent and driving actions. Across real-world datasets, E3AD improves visual grounding and waypoint planning and achieves state-of-the-art (SOTA) VAD correlation for emotion estimation. These results show that injecting emotion into VLA-style driving yields more human-aligned grounding, planning, and human-centric feedback.

Paperid: 2769, https://arxiv.org/pdf/2511.18444.pdf

Abstract:
Multimodal Large Language Models (MLLMs) increasingly need to forget specific knowledge such as unsafe or private information without requiring full retraining. However, existing unlearning methods often disrupt vision language alignment, causing models to reject both harmful and benign queries. We trace this failure to the projector network during unlearning, its Jacobian becomes severely illconditioned, leading to unstable optimization and drift in cross modal embeddings. We introduce SineProject, a simple method that augments the frozen projector with sinusoidally modulated trainable parameters, improving the Jacobian's spectral conditioning and stabilizing alignment throughout unlearning. Across standard safety and privacy unlearning benchmarks using LLaVA v1.5 7B and 13B, SineProject reduces benign query refusals while achieving complete forgetting of targeted information, yielding state of the art forget retain trade offs with negligible computational overhead.

Paperid: 2770, https://arxiv.org/pdf/2510.27607.pdf

Abstract:
Recently, augmenting vision-language-action models (VLAs) with world-models has shown promise in robotic policy learning. However, it remains challenging to jointly predict next-state observations and action sequences because of the inherent difference between the two modalities. To address this, we propose DUal-STream diffusion (DUST), a world-model augmented VLA framework that handles the modality conflict and enhances the performance of VLAs across diverse tasks. Specifically, we propose a multimodal diffusion transformer architecture that explicitly maintains separate modality streams while enabling cross-modal knowledge sharing. In addition, we propose training techniques such as independent noise perturbations for each modality and a decoupled flow matching loss, which enables the model to learn the joint distribution in a bidirectional manner while avoiding the need for a unified latent space. Furthermore, based on the decoupled training framework, we introduce a sampling method where we sample action and vision tokens asynchronously at different rates, which shows improvement through inference-time scaling. Through experiments on simulated benchmarks such as RoboCasa and GR-1, DUST achieves up to 6% gains over a standard VLA baseline and implicit world-modeling methods, with our inference-time scaling approach providing an additional 2-5% gain on success rate. On real-world tasks with the Franka Research 3, DUST outperforms baselines in success rate by 13%, confirming its effectiveness beyond simulation. Lastly, we demonstrate the effectiveness of DUST in large-scale pretraining with action-free videos from BridgeV2, where DUST leads to significant gain when transferred to the RoboCasa benchmark.

Paperid: 2771, https://arxiv.org/pdf/2510.11295.pdf

Abstract:
Large vision-language models (VLMs) achieve strong performance in Visual Question Answering but still rely heavily on supervised fine-tuning (SFT) with massive labeled datasets, which is costly due to human annotations. Crucially, real-world datasets often exhibit human uncertainty (HU) -- variation in human confidence across annotations -- but standard SFT simply optimizes toward the most frequent label, disregarding HU distributions. This leaves two open questions: How does HU affect SFT, and how can HU be effectively leveraged in training? In this work, we first conduct a systematic evaluation of VLMs across varying HU levels. We have two key findings: (i) surprisingly, high-HU samples contribute little or even degrade model performance, and (ii) naively training on the full dataset yields under-calibrated models that fail to capture HU distributions. Motivated by these findings, we introduce HaDola, a human uncertainty-aware data selection and automatic labeling framework. HaDola operates in four stages -- discriminate, self-annotate, error trigger, and training -- to iteratively identify harmful samples, prioritize informative ones, and bootstrap from a small seed set (5\% of data). Our approach substantially reduces reliance on costly HU annotations and makes VLMs more accurate and better calibrated. Extensive experiments on VQAv2 and VizWiz datasets demonstrate that HaDola consistently matches or outperforms state-of-the-art baselines with less training data. Our work highlights the importance of explicitly modeling HU in SFT, suggesting that better utilization of HU is more effective than merely scaling up dataset size.

Paperid: 2772, https://arxiv.org/pdf/2510.03182.pdf

Abstract:
Vision Language Models (VLMs) show strong potential for visual planning but struggle with precise spatial and long-horizon reasoning. In contrast, Planning Domain Definition Language (PDDL) planners excel at long-horizon formal planning, but cannot interpret visual inputs. Recent works combine these complementary advantages by enabling VLMs to turn visual planning problems into PDDL files for formal planning. However, while VLMs can generate PDDL problem files satisfactorily, they struggle to accurately generate the PDDL domain files, which describe all the planning rules. As a result, prior methods rely on human experts to predefine domain files or on constant environment access for refinement. We propose VLMFP, a Dual-VLM-guided framework that can autonomously generate both PDDL problem and domain files for formal visual planning. VLMFP introduces two VLMs to ensure reliable PDDL file generation: A SimVLM that simulates action consequences based on input rule descriptions, and a GenVLM that generates and iteratively refines PDDL files by comparing the PDDL and SimVLM execution results. VLMFP unleashes multiple levels of generalizability: The same generated PDDL domain file works for all the different instances under the same problem, and VLMs generalize to different problems with varied appearances and rules. We evaluate VLMFP with 6 grid-world domains and test its generalization to unseen instances, appearance, and game rules. On average, SimVLM accurately describes 95.5%, 82.6% of scenarios, simulates 85.5%, 87.8% of action sequence, and judges 82.4%, 85.6% goal reaching for seen and unseen appearances, respectively. With the guidance of SimVLM, VLMFP can generate PDDL files to reach 70.0%, 54.1% valid plans for unseen instances in seen and unseen appearances, respectively. Project page: https://sites.google.com/view/vlmfp.

Paperid: 2773, https://arxiv.org/pdf/2510.01711.pdf

Abstract:
Vision-Language-Action (VLA) models have shown its capabilities in robot manipulation by leveraging rich representations from pre-trained Vision-Language Models (VLMs). However, their representations arguably remain suboptimal, lacking sensitivity to robotic signals such as control actions and proprioceptive states. To address the issue, we introduce Robot State-aware Contrastive Loss (RS-CL), a simple and effective representation regularization for VLA models, designed to bridge the gap between VLM representations and robotic signals. In particular, RS-CL aligns the representations more closely with the robot's proprioceptive states, by using relative distances between the states as soft supervision. Complementing the original action prediction objective, RS-CL effectively enhances control-relevant representation learning, while being lightweight and fully compatible with standard VLA training pipeline. Our empirical results demonstrate that RS-CL substantially improves the manipulation performance of state-of-the-art VLA models; it pushes the prior art from 30.8% to 41.5% on pick-and-place tasks in RoboCasa-Kitchen, through more accurate positioning during grasping and placing, and boosts success rates from 45.0% to 58.3% on challenging real-robot manipulation tasks.

Paperid: 2774, https://arxiv.org/pdf/2509.22010.pdf

Abstract:
Despite significant advances in Vision Language Models (VLMs), they remain constrained by the complexity and redundancy of visual input. When images contain large amounts of irrelevant information, VLMs are susceptible to interference, thus generating excessive task-irrelevant reasoning processes or even hallucinations. This limitation stems from their inability to discover and process the required regions during reasoning precisely. To address this limitation, we present the Chain of Foresight-Focus Thought (CoFFT), a novel training-free approach that enhances VLMs' visual reasoning by emulating human visual cognition. Each Foresight-Focus Thought consists of three stages: (1) Diverse Sample Generation: generates diverse reasoning samples to explore potential reasoning paths, where each sample contains several reasoning steps; (2) Dual Foresight Decoding: rigorously evaluates these samples based on both visual focus and reasoning progression, adding the first step of optimal sample to the reasoning process; (3) Visual Focus Adjustment: precisely adjust visual focus toward regions most beneficial for future reasoning, before returning to stage (1) to generate subsequent reasoning samples until reaching the final answer. These stages function iteratively, creating an interdependent cycle where reasoning guides visual focus and visual focus informs subsequent reasoning. Empirical results across multiple benchmarks using Qwen2.5-VL, InternVL-2.5, and Llava-Next demonstrate consistent performance improvements of 3.1-5.8% with controllable increasing computational overhead.

Paperid: 2775, https://arxiv.org/pdf/2508.12404.pdf

Abstract:
Large vision-language models (VLMs) have shown promising capabilities in scene understanding, enhancing the explainability of driving behaviors and interactivity with users. Existing methods primarily fine-tune VLMs on on-board multi-view images and scene reasoning text, but this approach often lacks the holistic and nuanced scene recognition and powerful spatial awareness required for autonomous driving, especially in complex situations. To address this gap, we propose a novel vision-language framework tailored for autonomous driving, called LMAD. Our framework emulates modern end-to-end driving paradigms by incorporating comprehensive scene understanding and a task-specialized structure with VLMs. In particular, we introduce preliminary scene interaction and specialized expert adapters within the same driving task structure, which better align VLMs with autonomous driving scenarios. Furthermore, our approach is designed to be fully compatible with existing VLMs while seamlessly integrating with planning-oriented driving systems. Extensive experiments on the DriveLM and nuScenes-QA datasets demonstrate that LMAD significantly boosts the performance of existing VLMs on driving reasoning tasks,setting a new standard in explainable autonomous driving.

Paperid: 2776, https://arxiv.org/pdf/2508.11818.pdf

Abstract:
Chain-of-thought reasoning has demonstrated significant improvements in large language models and vision language models, yet its potential for audio language models remains largely unexplored. In this technical report, we take a preliminary step towards closing this gap. For better assessment of sound reasoning, we propose AF-Reasoning-Eval, a benchmark targeting common-sense reasoning and the ability to discriminate among closely related choices. To prepare training corpus for sound reasoning abilities, we propose automatic pipelines that transform existing audio question answering and classification data into explicit reasoning chains, yielding AF-CoT-Train with 1.24M samples. We study the effect of finetuning Audio Flamingo series on AF-CoT-Train and observe considerable improvements on several reasoning benchmarks, validating the effectiveness of chain-of-thought finetuning on advanced sound understanding.

Paperid: 2777, https://arxiv.org/pdf/2508.11428.pdf

Abstract:
Autonomous driving requires rich contextual comprehension and precise predictive reasoning to navigate dynamic and complex environments safely. Vision-Language Models (VLMs) and Driving World Models (DWMs) have independently emerged as powerful recipes addressing different aspects of this challenge. VLMs provide interpretability and robust action prediction through their ability to understand multi-modal context, while DWMs excel in generating detailed and plausible future driving scenarios essential for proactive planning. Integrating VLMs with DWMs is an intuitive, promising, yet understudied strategy to exploit the complementary strengths of accurate behavioral prediction and realistic scene generation. Nevertheless, this integration presents notable challenges, particularly in effectively connecting action-level decisions with high-fidelity pixel-level predictions and maintaining computational efficiency. In this paper, we propose ImagiDrive, a novel end-to-end autonomous driving framework that integrates a VLM-based driving agent with a DWM-based scene imaginer to form a unified imagination-and-planning loop. The driving agent predicts initial driving trajectories based on multi-modal inputs, guiding the scene imaginer to generate corresponding future scenarios. These imagined scenarios are subsequently utilized to iteratively refine the driving agent's planning decisions. To address efficiency and predictive accuracy challenges inherent in this integration, we introduce an early stopping mechanism and a trajectory selection strategy. Extensive experimental validation on the nuScenes and NAVSIM datasets demonstrates the robustness and superiority of ImagiDrive over previous alternatives under both open-loop and closed-loop conditions.

Paperid: 2778, https://arxiv.org/pdf/2508.01236.pdf

Abstract:
Despite the great success of Large Vision Language Models (LVLMs), their high computational cost severely limits their broad applications. The computational cost of LVLMs mainly stems from the visual sequence of the input, which consists of hundreds or even thousands of tokens. Although existing methods have made progress by removing redundant tokens, they suffer from severe performance degradation with high pruning rates due to the loss of visual information. In this paper, we propose an Adaptive Content Compensation Method (ACCM), which can effectively mitigate the visual information loss via an image caption. Specifically, ACCM comprises two key components: a lightweight caption model and a selector. Firstly the caption model generates question-related descriptions under the guidance of the user instruction. Then the selector further identifies a contextually appropriate caption from multiple candidates. Leveraging self-supervised learning, our modules could be learned efficiently without any human or automated labeling. We conduct extensive experiments across seven benchmarks and the results show that ACCM significantly outperforms existing methods with lower FLOPs (e.g., surpassing SOTA by 20.6% with 6.5% fewer FLOPs).

Paperid: 2779, https://arxiv.org/pdf/2507.06444.pdf

Abstract:
Accurate accident anticipation remains challenging when driver cognition and dynamic road conditions are underrepresented in predictive models. In this paper, we propose CAMERA (Context-Aware Multi-modal Enhanced Risk Anticipation), a multi-modal framework integrating dashcam video, textual annotations, and driver attention maps for robust accident anticipation. Unlike existing methods that rely on static or environment-centric thresholds, CAMERA employs an adaptive mechanism guided by scene complexity and gaze entropy, reducing false alarms while maintaining high recall in dynamic, multi-agent traffic scenarios. A hierarchical fusion pipeline with Bi-GRU (Bidirectional GRU) captures spatio-temporal dependencies, while a Geo-Context Vision-Language module translates 3D spatial relationships into interpretable, human-centric alerts. Evaluations on the DADA-2000 and benchmarks show that CAMERA achieves state-of-the-art performance, improving accuracy and lead time. These results demonstrate the effectiveness of modeling driver attention, contextual description, and adaptive risk thresholds to enable more reliable accident anticipation.

Paperid: 2780, https://arxiv.org/pdf/2507.05201.pdf

Authors:Andrew Sellergren, Sahar Kazemzadeh, Tiam Jaroensri, Atilla Kiraly, Madeleine Traverse, Timo Kohlberger, Shawn Xu, Fayaz Jamil, CÃan Hughes, Charles Lau, Justin Chen, Fereshteh Mahvar, Liron Yatziv, Tiffany Chen, Bram Sterling, Stefanie Anna Baby, Susanna Maria Baby, Jeremy Lai, Samuel Schmidgall, Lu Yang, Kejia Chen, Per Bjornsson, Shashir Reddy, Ryan Brush, Kenneth Philbrick, Mercy Asiedu, Ines Mezerreg, Howard Hu, Howard Yang, Richa Tiwari, Sunny Jansen, Preeti Singh, Yun Liu, Shekoofeh Azizi, Aishwarya Kamath, Johan Ferret, Shreya Pathak, Nino Vieillard, Ramona Merhej, Sarah Perrin, Tatiana Matejovicova, Alexandre RamÃ©, Morgane Riviere, Louis Rouillard, Thomas Mesnard, Geoffrey Cideron, Jean-bastien Grill, Sabela Ramos, Edouard Yvinec, Michelle Casbon, Elena Buchatskaya, Jean-Baptiste Alayrac, Dmitry Lepikhin, Vlad Feinberg, Sebastian Borgeaud, Alek Andreev, Cassidy Hardin, Robert Dadashi, LÃ©onard Hussenot, Armand Joulin, Olivier Bachem, Yossi Matias, Katherine Chou, Avinatan Hassidim, Kavi Goel, Clement Farabet, Joelle Barral, Tris Warkentin, Jonathon Shlens, David Fleet, Victor Cotruta, Omar Sanseviero, Gus Martins, Phoebe Kirk, Anand Rao, Shravya Shetty, David F. Steiner, Can Kirmizibayrak, Rory Pilgrim, Daniel Golden, Lin Yang

Abstract:
Artificial intelligence (AI) has significant potential in healthcare applications, but its training and deployment faces challenges due to healthcare's diverse data, complex tasks, and the need to preserve privacy. Foundation models that perform well on medical tasks and require less task-specific tuning data are critical to accelerate the development of healthcare AI applications. We introduce MedGemma, a collection of medical vision-language foundation models based on Gemma 3 4B and 27B. MedGemma demonstrates advanced medical understanding and reasoning on images and text, significantly exceeding the performance of similar-sized generative models and approaching the performance of task-specific models, while maintaining the general capabilities of the Gemma 3 base models. For out-of-distribution tasks, MedGemma achieves 2.6-10% improvement on medical multimodal question answering, 15.5-18.1% improvement on chest X-ray finding classification, and 10.8% improvement on agentic evaluations compared to the base models. Fine-tuning MedGemma further improves performance in subdomains, reducing errors in electronic health record information retrieval by 50% and reaching comparable performance to existing specialized state-of-the-art methods for pneumothorax classification and histopathology patch classification. We additionally introduce MedSigLIP, a medically-tuned vision encoder derived from SigLIP. MedSigLIP powers the visual understanding capabilities of MedGemma and as an encoder achieves comparable or better performance than specialized medical image encoders. Taken together, the MedGemma collection provides a strong foundation of medical image and text capabilities, with potential to significantly accelerate medical research and development of downstream applications. The MedGemma collection, including tutorials and model weights, can be found at https://goo.gle/medgemma.

Paperid: 2781, https://arxiv.org/pdf/2506.12723.pdf

Abstract:
Vision-Language-Action (VLA) models have attracted increasing attention for their strong control capabilities. However, their high computational cost and low execution frequency hinder their suitability for real-time tasks such as robotic manipulation and autonomous navigation. Existing VLA acceleration methods primarily focus on structural optimization, overlooking the fact that these models operate in sequential decision-making environments. As a result, temporal redundancy in sequential action generation and spatial redundancy in visual input remain unaddressed. To this end, we propose SP-VLA, a unified framework that accelerates VLA models by jointly scheduling models and pruning tokens. Specifically, we design an action-aware model scheduling mechanism that reduces temporal redundancy by dynamically switching between VLA model and a lightweight generator. Inspired by the human motion pattern of focusing on key decision points while relying on intuition for other actions, we categorize VLA actions into deliberative and intuitive, assigning the former to the VLA model and the latter to the lightweight generator, enabling frequency-adaptive execution through collaborative model scheduling. To address spatial redundancy, we further develop a spatio-semantic dual-aware token pruning method. Tokens are classified into spatial and semantic types and pruned based on their dual-aware importance to accelerate VLA inference. These two mechanisms work jointly to guide the VLA in focusing on critical actions and salient visual information, achieving effective acceleration while maintaining high accuracy. Experimental results demonstrate that our method achieves up to 1.5$\times$ acceleration with less than 3% drop in accuracy, outperforming existing approaches in multiple tasks.

Paperid: 2782, https://arxiv.org/pdf/2505.21895.pdf

Abstract:
Resource-constrained weight deployment is a task of immense practical importance. Recently, there has been interest in the specific task of \textit{Delta Compression}, where parties each hold a common base model and only communicate compressed weight updates. However, popular parameter efficient updates such as Low Rank Adaptation (LoRA) face inherent representation limitations - which are especially pronounced when combined with aggressive quantization. To overcome this, we build on recent work that improves LoRA representation capacity by using fixed-frequency sinusoidal functions to increase stable rank without adding additional parameters. We extend this to the quantized setting and present the first theoretical analysis showing how stable rank evolves under quantization. From this, we introduce SineLoRA$Δ$, a principled and effective method for delta compression that improves the expressivity of quantized low-rank adapters by applying a sinusoidal activation. We validate SineLoRA$Δ$ across a diverse variety of domains - including language modeling, vision-language tasks, and text-to-image generation - achieving up to 66% memory reduction with similar performance. We additionally provide a novel application of the canonical Bjøntegaard Delta metric to consistently compare adapter compression changes across the rate-distortion curve.

Paperid: 2783, https://arxiv.org/pdf/2505.10610.pdf

Abstract:
The rapid extension of context windows in large vision-language models has given rise to long-context vision-language models (LCVLMs), which are capable of handling hundreds of images with interleaved text tokens in a single forward pass. In this work, we introduce MMLongBench, the first benchmark covering a diverse set of long-context vision-language tasks, to evaluate LCVLMs effectively and thoroughly. MMLongBench is composed of 13,331 examples spanning five different categories of downstream tasks, such as Visual RAG and Many-Shot ICL. It also provides broad coverage of image types, including various natural and synthetic images. To assess the robustness of the models to different input lengths, all examples are delivered at five standardized input lengths (8K-128K tokens) via a cross-modal tokenization scheme that combines vision patches and text tokens. Through a thorough benchmarking of 46 closed-source and open-source LCVLMs, we provide a comprehensive analysis of the current models' vision-language long-context ability. Our results show that: i) performance on a single task is a weak proxy for overall long-context capability; ii) both closed-source and open-source models face challenges in long-context vision-language tasks, indicating substantial room for future improvement; iii) models with stronger reasoning ability tend to exhibit better long-context performance. By offering wide task coverage, various image types, and rigorous length control, MMLongBench provides the missing foundation for diagnosing and advancing the next generation of LCVLMs.

Paperid: 2784, https://arxiv.org/pdf/2505.07236.pdf

Abstract:
We present UAV-CodeAgents, a scalable multi-agent framework for autonomous UAV mission generation, built on large language and vision-language models (LLMs/VLMs). The system leverages the ReAct (Reason + Act) paradigm to interpret satellite imagery, ground high-level natural language instructions, and collaboratively generate UAV trajectories with minimal human supervision. A core component is a vision-grounded, pixel-pointing mechanism that enables precise localization of semantic targets on aerial maps. To support real-time adaptability, we introduce a reactive thinking loop, allowing agents to iteratively reflect on observations, revise mission goals, and coordinate dynamically in evolving environments. UAV-CodeAgents is evaluated on large-scale mission scenarios involving industrial and environmental fire detection. Our results show that a lower decoding temperature (0.5) yields higher planning reliability and reduced execution time, with an average mission creation time of 96.96 seconds and a success rate of 93%. We further fine-tune Qwen2.5VL-7B on 9,000 annotated satellite images, achieving strong spatial grounding across diverse visual categories. To foster reproducibility and future research, we will release the full codebase and a novel benchmark dataset for vision-language-based UAV planning.

Paperid: 2785, https://arxiv.org/pdf/2504.10320.pdf

Abstract:
Video anomaly detection (VAD) aims to identify unexpected events in videos and has wide applications in safety-critical domains. While semi-supervised methods trained on only normal samples have gained traction, they often suffer from high false alarm rates and poor interpretability. Recently, vision-language models (VLMs) have demonstrated strong multimodal reasoning capabilities, offering new opportunities for explainable anomaly detection. However, their high computational cost and lack of domain adaptation hinder real-time deployment and reliability. Inspired by dual complementary pathways in human visual perception, we propose SlowFastVAD, a hybrid framework that integrates a fast anomaly detector with a slow anomaly detector (namely a retrieval augmented generation (RAG) enhanced VLM), to address these limitations. Specifically, the fast detector first provides coarse anomaly confidence scores, and only a small subset of ambiguous segments, rather than the entire video, is further analyzed by the slower yet more interpretable VLM for elaborate detection and reasoning. Furthermore, to adapt VLMs to domain-specific VAD scenarios, we construct a knowledge base including normal patterns based on few normal samples and abnormal patterns inferred by VLMs. During inference, relevant patterns are retrieved and used to augment prompts for anomaly reasoning. Finally, we smoothly fuse the anomaly confidence of fast and slow detectors to enhance robustness of anomaly detection. Extensive experiments on four benchmarks demonstrate that SlowFastVAD effectively combines the strengths of both fast and slow detectors, and achieves remarkable detection accuracy and interpretability with significantly reduced computational overhead, making it well-suited for real-world VAD applications with high reliability requirements.

Paperid: 2786, https://arxiv.org/pdf/2504.04103.pdf

Abstract:
Accurately predicting traffic accidents in real-time is a critical challenge in autonomous driving, particularly in resource-constrained environments. Existing solutions often suffer from high computational overhead or fail to adequately address the uncertainty of evolving traffic scenarios. This paper introduces LATTE, a Lightweight Attention-based Traffic Accident Anticipation Engine, which integrates computational efficiency with state-of-the-art performance. LATTE employs Efficient Multiscale Spatial Aggregation (EMSA) to capture spatial features across scales, Memory Attention Aggregation (MAA) to enhance temporal modeling, and Auxiliary Self-Attention Aggregation (AAA) to extract latent dependencies over extended sequences. Additionally, LATTE incorporates the Flamingo Alert-Assisted System (FAA), leveraging a vision-language model to provide real-time, cognitively accessible verbal hazard alerts, improving passenger situational awareness. Evaluations on benchmark datasets (DAD, CCD, A3D) demonstrate LATTE's superior predictive capabilities and computational efficiency. LATTE achieves state-of-the-art 89.74% Average Precision (AP) on DAD benchmark, with 5.4% higher mean Time-To-Accident (mTTA) than the second-best model, and maintains competitive mTTA at a Recall of 80% (TTA@R80) (4.04s) while demonstrating robust accident anticipation across diverse driving conditions. Its lightweight design delivers a 93.14% reduction in floating-point operations (FLOPs) and a 31.58% decrease in parameter count (Params), enabling real-time operation on resource-limited hardware without compromising performance. Ablation studies confirm the effectiveness of LATTE's architectural components, while visualizations and failure case analyses highlight its practical applicability and areas for enhancement.

Paperid: 2787, https://arxiv.org/pdf/2503.14941.pdf

Abstract:
Multimodal Large Language Models (MLLMs) have emerged to tackle the challenges of Visual Question Answering (VQA), sparking a new research focus on conducting objective evaluations of these models. Existing evaluation methods face limitations due to the significant human workload required to design Q&A pairs for visual images, which inherently restricts the scale and scope of evaluations. Although automated MLLM-as-judge approaches attempt to reduce the human workload through automatic evaluations, they often introduce biases. To address these problems, we propose an Unsupervised Peer review MLLM Evaluation framework. It utilizes only image data, allowing models to automatically generate questions and conduct peer review assessments of answers from other models, effectively alleviating the reliance on human workload. Additionally, we introduce the vision-language scoring system to mitigate the bias issues, which focuses on three aspects: (i) response correctness; (ii) visual understanding and reasoning; and (iii) image-text correlation. Experimental results demonstrate that UPME achieves a Pearson correlation of 0.944 with human evaluations on the MMstar dataset and 0.814 on the ScienceQA dataset, indicating that our framework closely aligns with human-designed benchmarks and inherent human preferences.

Paperid: 2788, https://arxiv.org/pdf/2503.10480.pdf

Abstract:
Recent advances in large vision-language models (LVLMs) have shown promise for embodied task planning, yet they struggle with fundamental challenges like dependency constraints and efficiency. Existing approaches either solely optimize action selection or leverage world models during inference, overlooking the benefits of learning to model the world as a way to enhance planning capabilities. We propose Dual Preference Optimization (D$^2$PO), a new learning framework that jointly optimizes state prediction and action selection through preference learning, enabling LVLMs to understand environment dynamics for better planning. To automatically collect trajectories and stepwise preference data without human annotation, we introduce a tree search mechanism for extensive exploration via trial-and-error. Extensive experiments on VoTa-Bench demonstrate that our D$^2$PO-based method significantly outperforms existing methods and GPT-4o when applied to Qwen2-VL (7B), LLaVA-1.6 (7B), and LLaMA-3.2 (11B), achieving superior task success rates with more efficient execution paths.

Paperid: 2789, https://arxiv.org/pdf/2503.02465.pdf

Abstract:
Emergency search and rescue (SAR) operations often require rapid and precise target identification in complex environments where traditional manual drone control is inefficient. In order to address these scenarios, a rapid SAR system, UAV-VLRR (Vision-Language-Rapid-Response), is developed in this research. This system consists of two aspects: 1) A multimodal system which harnesses the power of Visual Language Model (VLM) and the natural language processing capabilities of ChatGPT-4o (LLM) for scene interpretation. 2) A non-linearmodel predictive control (NMPC) with built-in obstacle avoidance for rapid response by a drone to fly according to the output of the multimodal system. This work aims at improving response times in emergency SAR operations by providing a more intuitive and natural approach to the operator to plan the SAR mission while allowing the drone to carry out that mission in a rapid and safe manner. When tested, our approach was faster on an average by 33.75% when compared with an off-the-shelf autopilot and 54.6% when compared with a human pilot. Video of UAV-VLRR: https://youtu.be/KJqQGKKt1xY

Paperid: 2790, https://arxiv.org/pdf/2503.02454.pdf

Abstract:
The UAV-VLPA* (Visual-Language-Planning-and-Action) system represents a cutting-edge advancement in aerial robotics, designed to enhance communication and operational efficiency for unmanned aerial vehicles (UAVs). By integrating advanced planning capabilities, the system addresses the Traveling Salesman Problem (TSP) to optimize flight paths, reducing the total trajectory length by 18.5\% compared to traditional methods. Additionally, the incorporation of the A* algorithm enables robust obstacle avoidance, ensuring safe and efficient navigation in complex environments. The system leverages satellite imagery processing combined with the Visual Language Model (VLM) and GPT's natural language processing capabilities, allowing users to generate detailed flight plans through simple text commands. This seamless fusion of visual and linguistic analysis empowers precise decision-making and mission planning, making UAV-VLPA* a transformative tool for modern aerial operations. With its unmatched operational efficiency, navigational safety, and user-friendly functionality, UAV-VLPA* sets a new standard in autonomous aerial robotics, paving the way for future innovations in the field.

Paperid: 2791, https://arxiv.org/pdf/2503.02310.pdf

Abstract:
Vision-Language-Action (VLA) models demonstrate remarkable potential for generalizable robotic manipulation. The performance of VLA models can be improved by integrating with action chunking, a critical technique for effective control. However, action chunking linearly scales up action dimensions in VLA models with increased chunking sizes. This reduces the inference efficiency. To tackle this problem, we propose PD-VLA, the first parallel decoding framework for VLA models integrated with action chunking. Our framework reformulates autoregressive decoding as a nonlinear system solved by parallel fixed-point iterations. This approach preserves model performance with mathematical guarantees while significantly improving decoding speed. In addition, it enables training-free acceleration without architectural changes, as well as seamless synergy with existing acceleration techniques. Extensive simulations validate that our PD-VLA maintains competitive success rates while achieving 2.52 times execution frequency on manipulators (with 7 degrees of freedom) compared with the fundamental VLA model. Furthermore, we experimentally identify the most effective settings for acceleration. Finally, real-world experiments validate its high applicability across different tasks.

Paperid: 2792, https://arxiv.org/pdf/2503.01378.pdf

Abstract:
This paper introduces CognitiveDrone, a novel Vision-Language-Action (VLA) model tailored for complex Unmanned Aerial Vehicles (UAVs) tasks that demand advanced cognitive abilities. Trained on a dataset comprising over 8,000 simulated flight trajectories across three key categories-Human Recognition, Symbol Understanding, and Reasoning-the model generates real-time 4D action commands based on first-person visual inputs and textual instructions. To further enhance performance in intricate scenarios, we propose CognitiveDrone-R1, which integrates an additional Vision-Language Model (VLM) reasoning module to simplify task directives prior to high-frequency control. Experimental evaluations using our open-source benchmark, CognitiveDroneBench, reveal that while a racing-oriented model (RaceVLA) achieves an overall success rate of 31.3%, the base CognitiveDrone model reaches 59.6%, and CognitiveDrone-R1 attains a success rate of 77.2%. These results demonstrate improvements of up to 30% in critical cognitive tasks, underscoring the effectiveness of incorporating advanced reasoning capabilities into UAV control systems. Our contributions include the development of a state-of-the-art VLA model for UAV control and the introduction of the first dedicated benchmark for assessing cognitive tasks in drone operations. The complete repository is available at cognitivedrone.github.io

Paperid: 2793, https://arxiv.org/pdf/2502.20509.pdf

Abstract:
Vision-language models have proven to be of great benefit for medical image analysis since they learn rich semantics from both images and reports. Prior efforts have focused on better alignment of image and text representations to enhance image understanding. However, though explicit reference to a prior image is common in Chest X-Ray (CXR) reports, aligning progression descriptions with the semantics differences in image pairs remains under-explored. In this work, we propose two components to address this issue. (1) A CXR report processing pipeline to extract temporal structure. It processes reports with a large language model (LLM) to separate the description and comparison contexts, and extracts fine-grained annotations from reports. (2) A contrastive captioner model for CXR, namely CoCa-CXR, to learn how to both describe images and their temporal progressions. CoCa-CXR incorporates a novel regional cross-attention module to identify local differences between paired CXR images. Extensive experiments show the superiority of CoCa-CXR on both progression analysis and report generation compared to previous methods. Notably, on MS-CXR-T progression classification, CoCa-CXR obtains 65.0% average testing accuracy on five pulmonary conditions, outperforming the previous state-of-the-art (SOTA) model BioViL-T by 4.8%. It also achieves a RadGraph F1 of 24.2% on MIMIC-CXR, which is comparable to the Med-Gemini foundation model.

Paperid: 2794, https://arxiv.org/pdf/2502.13487.pdf

Abstract:
Large vision-language models (LVLMs) perform outstandingly across various multimodal tasks. However, their ability to evaluate generated content remains limited, and training vision-language reward models (VLRMs) with preference data is computationally expensive. This paper explores a training-free alternative by merging text-based reward models (RMs) with LVLMs to create VLRMs. Our approach shows that integrating these models leads to improved performance over LVLMs' scoring and text-based RMs, offering an efficient method for incorporating textual preferences into LVLMs.

Paperid: 2795, https://arxiv.org/pdf/2502.10536.pdf

Abstract:
The interpretation of histopathology cases underlies many important diagnostic and treatment decisions in medicine. Notably, this process typically requires pathologists to integrate and summarize findings across multiple slides per case. Existing vision-language capabilities in computational pathology have so far been largely limited to small regions of interest, larger regions at low magnification, or single whole-slide images (WSIs). This limits interpretation of findings that span multiple high-magnification regions across multiple WSIs. By making use of Gemini 1.5 Flash, a large multimodal model (LMM) with a 1-million token context window, we demonstrate the ability to generate bottom-line diagnoses from up to 40,000 768x768 pixel image patches from multiple WSIs at 10X magnification. This is the equivalent of up to 11 hours of video at 1 fps. Expert pathologist evaluations demonstrate that the generated report text is clinically accurate and equivalent to or preferred over the original reporting for 68% (95% CI: [60%, 76%]) of multi-slide examples with up to 5 slides. While performance decreased for examples with 6 or more slides, this study demonstrates the promise of leveraging the long-context capabilities of modern LMMs for the uniquely challenging task of medical report generation where each case can contain thousands of image patches.

Paperid: 2796, https://arxiv.org/pdf/2501.07295.pdf

Abstract:
This paper introduces GestLLM, an advanced system for human-robot interaction that enables intuitive robot control through hand gestures. Unlike conventional systems, which rely on a limited set of predefined gestures, GestLLM leverages large language models and feature extraction via MediaPipe to interpret a diverse range of gestures. This integration addresses key limitations in existing systems, such as restricted gesture flexibility and the inability to recognize complex or unconventional gestures commonly used in human communication. By combining state-of-the-art feature extraction and language model capabilities, GestLLM achieves performance comparable to leading vision-language models while supporting gestures underrepresented in traditional datasets. For example, this includes gestures from popular culture, such as the ``Vulcan salute" from Star Trek, without any additional pretraining, prompt engineering, etc. This flexibility enhances the naturalness and inclusivity of robot control, making interactions more intuitive and user-friendly. GestLLM provides a significant step forward in gesture-based interaction, enabling robots to understand and respond to a wide variety of hand gestures effectively. This paper outlines its design, implementation, and evaluation, demonstrating its potential applications in advanced human-robot collaboration, assistive robotics, and interactive entertainment.

Paperid: 2797, https://arxiv.org/pdf/2410.08001.pdf

Abstract:
The increasing demand for versatile robotic systems to operate in diverse and dynamic environments has emphasized the importance of a generalist policy, which leverages a large cross-embodiment data corpus to facilitate broad adaptability and high-level reasoning. However, the generalist would struggle with inefficient inference and cost-expensive training. The specialist policy, instead, is curated for specific domain data and excels at task-level precision with efficiency. Yet, it lacks the generalization capacity for a wide range of applications. Inspired by these observations, we introduce RoboDual, a synergistic dual-system that supplements the merits of both generalist and specialist policy. A diffusion transformer-based specialist is devised for multi-step action rollouts, exquisitely conditioned on the high-level task understanding and discretized action output of a vision-language-action (VLA) based generalist. Compared to OpenVLA, RoboDual achieves 26.7% improvement in real-world setting and 12% gain on CALVIN by introducing a specialist policy with merely 20M trainable parameters. It maintains strong performance with 5% of demonstration data only, and enables a 3.8 times higher control frequency in real-world deployment. Code would be made publicly available. Our project page is hosted at: https://opendrivelab.com/RoboDual/

Paperid: 2798, https://arxiv.org/pdf/2601.12993.pdf

Abstract:
We introduce Being-H0.5, a foundational Vision-Language-Action (VLA) model designed for robust cross-embodiment generalization across diverse robotic platforms. While existing VLAs often struggle with morphological heterogeneity and data scarcity, we propose a human-centric learning paradigm that treats human interaction traces as a universal "mother tongue" for physical interaction. To support this, we present UniHand-2.0, the largest embodied pre-training recipe to date, comprising over 35,000 hours of multimodal data across 30 distinct robotic embodiments. Our approach introduces a Unified Action Space that maps heterogeneous robot controls into semantically aligned slots, enabling low-resource robots to bootstrap skills from human data and high-resource platforms. Built upon this human-centric foundation, we design a unified sequential modeling and multi-task pre-training paradigm to bridge human demonstrations and robotic execution. Architecturally, Being-H0.5 utilizes a Mixture-of-Transformers design featuring a novel Mixture-of-Flow (MoF) framework to decouple shared motor primitives from specialized embodiment-specific experts. Finally, to make cross-embodiment policies stable in the real world, we introduce Manifold-Preserving Gating for robustness under sensory shift and Universal Async Chunking to universalize chunked control across embodiments with different latency and control profiles. We empirically demonstrate that Being-H0.5 achieves state-of-the-art results on simulated benchmarks, such as LIBERO (98.9%) and RoboCasa (53.9%), while also exhibiting strong cross-embodiment capabilities on five robotic platforms.

Paperid: 2799, https://arxiv.org/pdf/2601.05432.pdf

Abstract:
The image geolocalization task aims to predict the location where an image was taken anywhere on Earth using visual clues. Existing large vision-language model (LVLM) approaches leverage world knowledge, chain-of-thought reasoning, and agentic capabilities, but overlook a common strategy used by humans -- using maps. In this work, we first equip the model \textit{Thinking with Map} ability and formulate it as an agent-in-the-map loop. We develop a two-stage optimization scheme for it, including agentic reinforcement learning (RL) followed by parallel test-time scaling (TTS). The RL strengthens the agentic capability of model to improve sampling efficiency, and the parallel TTS enables the model to explore multiple candidate paths before making the final prediction, which is crucial for geolocalization. To evaluate our method on up-to-date and in-the-wild images, we further present MAPBench, a comprehensive geolocalization training and evaluation benchmark composed entirely of real-world images. Experimental results show that our method outperforms existing open- and closed-source models on most metrics, specifically improving Acc@500m from 8.0\% to 22.1\% compared to \textit{Gemini-3-Pro} with Google Search/Map grounded mode.

Paperid: 2800, https://arxiv.org/pdf/2512.13380.pdf

Abstract:
Reinforcement learning (RL) has achieved great success in dexterous grasping, significantly improving grasp performance and generalization from simulation to the real world. However, fine-grained functional grasping, which is essential for downstream manipulation tasks, remains underexplored and faces several challenges: the complexity of specifying goals and reward functions for functional grasps across diverse objects, the difficulty of multi-task RL exploration, and the challenge of sim-to-real transfer. In this work, we propose DemoFunGrasp for universal dexterous functional grasping. We factorize functional grasping conditions into two complementary components - grasping style and affordance - and integrate them into an RL framework that can learn to grasp any object with any functional grasping condition. To address the multi-task optimization challenge, we leverage a single grasping demonstration and reformulate the RL problem as one-step demonstration editing, substantially enhancing sample efficiency and performance. Experimental results in both simulation and the real world show that DemoFunGrasp generalizes to unseen combinations of objects, affordances, and grasping styles, outperforming baselines in both success rate and functional grasping accuracy. In addition to strong sim-to-real capability, by incorporating a vision-language model (VLM) for planning, our system achieves autonomous instruction-following grasp execution.

Paperid: 2801, https://arxiv.org/pdf/2512.13080.pdf

Abstract:
Vision-Language-Action (VLA) models provide a promising paradigm for robot learning by integrating visual perception with language-guided policy learning. However, most existing approaches rely on 2D visual inputs to perform actions in 3D physical environments, creating a significant gap between perception and action grounding. To bridge this gap, we propose a Spatial-Aware VLA Pretraining paradigm that performs explicit alignment between visual space and physical space during pretraining, enabling models to acquire 3D spatial understanding before robot policy learning. Starting from pretrained vision-language models, we leverage large-scale human demonstration videos to extract 3D visual and 3D action annotations, forming a new source of supervision that aligns 2D visual observations with 3D spatial reasoning. We instantiate this paradigm with VIPA-VLA, a dual-encoder architecture that incorporates a 3D visual encoder to augment semantic visual representations with 3D-aware features. When adapted to downstream robot tasks, VIPA-VLA achieves significantly improved grounding between 2D vision and 3D action, resulting in more robust and generalizable robotic policies.

Paperid: 2802, https://arxiv.org/pdf/2512.12487.pdf

Abstract:
Reinforcement learning from verifiable rewards (RLVR) has recently been extended from text-only LLMs to vision-language models (VLMs) to elicit long-chain multimodal reasoning. However, RLVR-trained VLMs still exhibit two persistent failure modes: inaccurate visual extraction (missing or hallucinating details) and logically inconsistent chains-of-thought, largely because verifiable signals supervise only the final answer. We propose PeRL-VL (Perception and Reasoning Learning for Vision-Language Models), a decoupled framework that separately improves visual perception and textual reasoning on top of RLVR. For perception, PeRL-VL introduces a VLM-based description reward that scores the model's self-generated image descriptions for faithfulness and sufficiency. For reasoning, PeRL-VL adds a text-only Reasoning SFT stage on logic-rich chain-of-thought data, enhancing coherence and logical consistency independently of vision. Across diverse multimodal benchmarks, PeRL-VL improves average Pass@1 accuracy from 63.3% (base Qwen2.5-VL-7B) to 68.8%, outperforming standard RLVR, text-only reasoning SFT, and naive multimodal distillation from GPT-4o.

Paperid: 2803, https://arxiv.org/pdf/2512.08016.pdf

Abstract:
Cartographic reasoning is the skill of interpreting geographic relationships by aligning legends, map scales, compass directions, map texts, and geometries across one or more map images. Although essential as a concrete cognitive capability and for critical tasks such as disaster response and urban planning, it remains largely unevaluated. Building on progress in chart and infographic understanding, recent large vision language model studies on map visual question-answering often treat maps as a special case of charts. In contrast, map VQA demands comprehension of layered symbology (e.g., symbols, geometries, and text labels) as well as spatial relations tied to orientation and distance that often span multiple maps and are not captured by chart-style evaluations. To address this gap, we introduce FRIEDA, a benchmark for testing complex open-ended cartographic reasoning in LVLMs. FRIEDA sources real map images from documents and reports in various domains and geographical areas. Following classifications in Geographic Information System (GIS) literature, FRIEDA targets all three categories of spatial relations: topological (border, equal, intersect, within), metric (distance), and directional (orientation). All questions require multi-step inference, and many require cross-map grounding and reasoning. We evaluate eleven state-of-the-art LVLMs under two settings: (1) the direct setting, where we provide the maps relevant to the question, and (2) the contextual setting, where the model may have to identify the maps relevant to the question before reasoning. Even the strongest models, Gemini-2.5-Pro and GPT-5-Think, achieve only 38.20% and 37.20% accuracy, respectively, far below human performance of 84.87%. These results reveal a persistent gap in multi-step cartographic reasoning, positioning FRIEDA as a rigorous benchmark to drive progress on spatial intelligence in LVLMs.

Paperid: 2804, https://arxiv.org/pdf/2512.01715.pdf

Abstract:
Vision-Language-Action (VLA) models trained with flow matching have demonstrated impressive capabilities on robotic manipulation tasks. However, their performance often degrades under distribution shift and on complex multi-step tasks, suggesting that the learned representations may not robustly capture task-relevant semantics. We introduce DiG-Flow, a principled framework that enhances VLA robustness through geometric regularization. Our key insight is that the distributional discrepancy between observation and action embeddings provides a meaningful geometric signal: lower transport cost indicates compatible representations, while higher cost suggests potential misalignment. DiG-Flow computes a discrepancy measure between empirical distributions of observation and action embeddings, maps it to a modulation weight via a monotone function, and applies residual updates to the observation embeddings before flow matching. Crucially, this intervention operates at the representation level without modifying the flow matching path or target vector field. We provide theoretical guarantees showing that discrepancy-guided training provably decreases the training objective, and that guided inference refinement converges with contraction. Empirically, DiG-Flow integrates into existing VLA architectures with negligible overhead and consistently improves performance, with particularly pronounced gains on complex multi-step tasks and under limited training data.

Paperid: 2805, https://arxiv.org/pdf/2510.24038.pdf

Abstract:
Vision-language models (VLMs) such as CLIP demonstrate strong generalization in zero-shot classification but remain highly vulnerable to adversarial perturbations. Existing methods primarily focus on adversarial fine-tuning or prompt optimization; they often overlook the gaps in CLIP's encoded features, which is shown as the text and image features lie far apart from each other. This misalignment is significantly amplified under adversarial perturbations, leading to severe degradation in classification performance. To address this problem, we propose Cross-modality Alignment, dubbed COLA, an optimal transport-based framework that explicitly addresses adversarial misalignment by restoring both global image-text alignment and local structural consistency in the feature space. (1) COLA first projects adversarial image embeddings onto a subspace spanned by class text features, effectively filtering out non-semantic distortions while preserving discriminative information. (2) It then models images and texts as discrete distributions over multiple augmented views and refines their alignment via OT, with the subspace projection seamlessly integrated into the cost computation. This design ensures stable cross-modal alignment even under adversarial conditions. COLA is training-free and compatible with existing fine-tuned models. Extensive evaluations across 14 zero-shot classification benchmarks demonstrate the effectiveness of COLA, especially with an average improvement of 6.7% on ImageNet and its variants under PGD adversarial attacks, while maintaining high accuracy on clean samples.

Paperid: 2806, https://arxiv.org/pdf/2510.23224.pdf

Abstract:
The rapid digitization of histopathology slides has opened up new possibilities for computational tools in clinical and research workflows. Among these, content-based slide retrieval stands out, enabling pathologists to identify morphologically and semantically similar cases, thereby supporting precise diagnoses, enhancing consistency across observers, and assisting example-based education. However, effective retrieval of whole slide images (WSIs) remains challenging due to their gigapixel scale and the difficulty of capturing subtle semantic differences amid abundant irrelevant content. To overcome these challenges, we present PathSearch, a retrieval framework that unifies fine-grained attentive mosaic representations with global-wise slide embeddings aligned through vision-language contrastive learning. Trained on a corpus of 6,926 slide-report pairs, PathSearch captures both fine-grained morphological cues and high-level semantic patterns to enable accurate and flexible retrieval. The framework supports two key functionalities: (1) mosaic-based image-to-image retrieval, ensuring accurate and efficient slide research; and (2) multi-modal retrieval, where text queries can directly retrieve relevant slides. PathSearch was rigorously evaluated on four public pathology datasets and three in-house cohorts, covering tasks including anatomical site retrieval, tumor subtyping, tumor vs. non-tumor discrimination, and grading across diverse organs such as breast, lung, kidney, liver, and stomach. External results show that PathSearch outperforms traditional image-to-image retrieval frameworks. A multi-center reader study further demonstrates that PathSearch improves diagnostic accuracy, boosts confidence, and enhances inter-observer agreement among pathologists in real clinical scenarios. These results establish PathSearch as a scalable and generalizable retrieval solution for digital pathology.

Paperid: 2807, https://arxiv.org/pdf/2510.18054.pdf

Abstract:
We introduce HouseTour, a method for spatially-aware 3D camera trajectory and natural language summary generation from a collection of images depicting an existing 3D space. Unlike existing vision-language models (VLMs), which struggle with geometric reasoning, our approach generates smooth video trajectories via a diffusion process constrained by known camera poses and integrates this information into the VLM for 3D-grounded descriptions. We synthesize the final video using 3D Gaussian splatting to render novel views along the trajectory. To support this task, we present the HouseTour dataset, which includes over 1,200 house-tour videos with camera poses, 3D reconstructions, and real estate descriptions. Experiments demonstrate that incorporating 3D camera trajectories into the text generation process improves performance over methods handling each task independently. We evaluate both individual and end-to-end performance, introducing a new joint metric. Our work enables automated, professional-quality video creation for real estate and touristic applications without requiring specialized expertise or equipment.

Paperid: 2808, https://arxiv.org/pdf/2510.13108.pdf

Abstract:
Benchmarking autonomous driving planners to align with human judgment remains a critical challenge, as state-of-the-art metrics like the Extended Predictive Driver Model Score (EPDMS) lack context awareness in nuanced scenarios. To address this, we introduce DriveCritic, a novel framework featuring two key contributions: the DriveCritic dataset, a curated collection of challenging scenarios where context is critical for correct judgment and annotated with pairwise human preferences, and the DriveCritic model, a Vision-Language Model (VLM) based evaluator. Fine-tuned using a two-stage supervised and reinforcement learning pipeline, the DriveCritic model learns to adjudicate between trajectory pairs by integrating visual and symbolic context. Experiments show DriveCritic significantly outperforms existing metrics and baselines in matching human preferences and demonstrates strong context awareness. Overall, our work provides a more reliable, human-aligned foundation to evaluating autonomous driving systems.

Paperid: 2809, https://arxiv.org/pdf/2510.00523.pdf

Abstract:
Multimodal representation learning models have demonstrated successful operation across complex tasks, and the integration of vision-language models (VLMs) has further enabled embedding models with instruction-following capabilities. However, existing embedding models lack visual-interactive capabilities to specify regions of interest from users (e.g., point, bounding box, mask), which have been explored in generative models to broaden their human-interactive applicability. Equipping embedding models with visual interactions not only would unlock new applications with localized grounding of user intent, which remains unexplored, but also enable the models to learn entity-level information within images to complement their global representations for conventional embedding tasks. In this paper, we propose a novel Visual-InteRactive Text-Image Universal Embedder (VIRTUE) that extends the capabilities of the segmentation model and the vision-language model to the realm of representation learning. In VIRTUE, the segmentation model can process visual prompts that pinpoint specific regions within an image, thereby enabling the embedder to handle complex and ambiguous scenarios more precisely. To evaluate the visual-interaction ability of VIRTUE, we introduce a large-scale Segmentation-and-Scene Caption Retrieval (SCaR) benchmark comprising 1M samples that aims to retrieve the text caption by jointly considering the entity with a specific object and image scene. VIRTUE consistently achieves a state-of-the-art performance with significant improvements across 36 universal MMEB (3.1%-8.5%) and five visual-interactive SCaR (15.2%-20.3%) tasks.

Paperid: 2810, https://arxiv.org/pdf/2509.25991.pdf

Abstract:
In recent years, detecting fake multimodal content on social media has drawn increasing attention. Two major forms of deception dominate: human-crafted misinformation (e.g., rumors and misleading posts) and AI-generated content produced by image synthesis models or vision-language models (VLMs). Although both share deceptive intent, they are typically studied in isolation. NLP research focuses on human-written misinformation, while the CV community targets AI-generated artifacts. As a result, existing models are often specialized for only one type of fake content. In real-world scenarios, however, the type of a multimodal post is usually unknown, limiting the effectiveness of such specialized systems. To bridge this gap, we construct the Omnibus Dataset for Multimodal News Deception (OmniFake), a comprehensive benchmark of 127K samples that integrates human-curated misinformation from existing resources with newly synthesized AI-generated examples. Based on this dataset, we propose Unified Multimodal Fake Content Detection (UMFDet), a framework designed to handle both forms of deception. UMFDet leverages a VLM backbone augmented with a Category-aware Mixture-of-Experts (MoE) Adapter to capture category-specific cues, and an attribution chain-of-thought mechanism that provides implicit reasoning guidance for locating salient deceptive signals. Extensive experiments demonstrate that UMFDet achieves robust and consistent performance across both misinformation types, outperforming specialized baselines and offering a practical solution for real-world multimodal deception detection.

Paperid: 2811, https://arxiv.org/pdf/2509.25413.pdf

Abstract:
Vision language models (VLMs) can flexibly address various vision tasks through text interactions. Although successful in semantic understanding, state-of-the-art VLMs including GPT-5 still struggle in understanding 3D from 2D inputs. On the other hand, expert pure vision models achieve super-human accuracy in metric depth estimation, a key 3D understanding task. However, they require task-specific architectures and losses. Such difference motivates us to ask: Can VLMs reach expert-level accuracy without architecture or loss change? We take per-pixel metric depth estimation as the representative task and show that the answer is yes! Surprisingly, comprehensive analysis shows that text-based supervised-finetuning with sparse labels is sufficient for VLMs to unlock strong 3D understanding, no dense prediction head or complex regression/regularization loss is needed. The bottleneck for VLMs lies actually in pixel reference and cross-dataset camera ambiguity, which we address through visual prompting and intrinsic-conditioned augmentation. With much smaller models, our method DepthLM surpasses the accuracy of most advanced VLMs by over 2x, making VLMs for the first time comparable with pure vision models. Interestingly, without explicit enforcement during training, VLMs trained with DepthLM naturally avoids over-smoothing, having much fewer flying points at boundary regions than pure vision models. The simplicity of DepthLM also enables a single VLM to cover various 3D tasks beyond metric depth. Our code and model will be released at the link below.

Paperid: 2812, https://arxiv.org/pdf/2509.20109.pdf

Abstract:
End-to-End (E2E) solutions have emerged as a mainstream approach for autonomous driving systems, with Vision-Language-Action (VLA) models representing a new paradigm that leverages pre-trained multimodal knowledge from Vision-Language Models (VLMs) to interpret and interact with complex real-world environments. However, these methods remain constrained by the limitations of imitation learning, which struggles to inherently encode physical rules during training. Existing approaches often rely on complex rule-based post-refinement, employ reinforcement learning that remains largely limited to simulation, or utilize diffusion guidance that requires computationally expensive gradient calculations. To address these challenges, we introduce ReflectDrive, a novel learning-based framework that integrates a reflection mechanism for safe trajectory generation via discrete diffusion. We first discretize the two-dimensional driving space to construct an action codebook, enabling the use of pre-trained Diffusion Language Models for planning tasks through fine-tuning. Central to our approach is a safety-aware reflection mechanism that performs iterative self-correction without gradient computation. Our method begins with goal-conditioned trajectory generation to model multi-modal driving behaviors. Based on this, we apply local search methods to identify unsafe tokens and determine feasible solutions, which then serve as safe anchors for inpainting-based regeneration. Evaluated on the NAVSIM benchmark, ReflectDrive demonstrates significant advantages in safety-critical trajectory generation, offering a scalable and reliable solution for autonomous driving systems.

Paperid: 2813, https://arxiv.org/pdf/2509.15711.pdf

Abstract:
The rapid advancement of generative AI in medical imaging has introduced both significant opportunities and serious challenges, especially the risk that fake medical images could undermine healthcare systems. These synthetic images pose serious risks, such as diagnostic deception, financial fraud, and misinformation. However, research on medical forensics to counter these threats remains limited, and there is a critical lack of comprehensive datasets specifically tailored for this field. Additionally, existing media forensic methods, which are primarily designed for natural or facial images, are inadequate for capturing the distinct characteristics and subtle artifacts of AI-generated medical images. To tackle these challenges, we introduce \textbf{MedForensics}, a large-scale medical forensics dataset encompassing six medical modalities and twelve state-of-the-art medical generative models. We also propose \textbf{DSKI}, a novel \textbf{D}ual-\textbf{S}tage \textbf{K}nowledge \textbf{I}nfusing detector that constructs a vision-language feature space tailored for the detection of AI-generated medical images. DSKI comprises two core components: 1) a cross-domain fine-trace adapter (CDFA) for extracting subtle forgery clues from both spatial and noise domains during training, and 2) a medical forensic retrieval module (MFRM) that boosts detection accuracy through few-shot retrieval during testing. Experimental results demonstrate that DSKI significantly outperforms both existing methods and human experts, achieving superior accuracy across multiple medical modalities.

Paperid: 2814, https://arxiv.org/pdf/2509.12715.pdf

Abstract:
Large Vision-Language Models (LVLMs) have demonstrated impressive performance on multimodal tasks through scaled architectures and extensive training. However, existing Mixture of Experts (MoE) approaches face challenges due to the asymmetry between visual and linguistic processing. Visual information is spatially complete, while language requires maintaining sequential context. As a result, MoE models struggle to balance modality-specific features and cross-modal interactions. Through systematic analysis, we observe that language experts in deeper layers progressively lose contextual grounding and rely more on parametric knowledge rather than utilizing the provided visual and linguistic information. To address this, we propose AsyMoE, a novel architecture that models this asymmetry using three specialized expert groups. We design intra-modality experts for modality-specific processing, hyperbolic inter-modality experts for hierarchical cross-modal interactions, and evidence-priority language experts to suppress parametric biases and maintain contextual grounding. Extensive experiments demonstrate that AsyMoE achieves 26.58% and 15.45% accuracy improvements over vanilla MoE and modality-specific MoE respectively, with 25.45% fewer activated parameters than dense models.

Paperid: 2815, https://arxiv.org/pdf/2509.00789.pdf

Abstract:
Recent advances in vision-language models (VLMs) have demonstrated impressive spatial reasoning capabilities for autonomous driving, yet existing methods predominantly focus on static scene understanding while neglecting the essential temporal dimension of real-world driving scenarios. To address this critical limitation, we propose the OmniReason framework, which establishes robust spatiotemporal reasoning by jointly modeling dynamic 3D environments and their underlying decision-making processes. Our work makes two fundamental advances: (1) We introduce OmniReason-Data, two large-scale vision-language-action (VLA) datasets with dense spatiotemporal annotations and natural language explanations, generated through a novel hallucination-mitigated auto-labeling pipeline that ensures both physical plausibility and temporal coherence; (2) We develop the OmniReason-Agent architecture, which integrates a sparse temporal memory module for persistent scene context modeling and an explanation generator that produces human-interpretable decision rationales, facilitated by our spatiotemporal knowledge distillation approach that effectively captures spatiotemporal causal reasoning patterns. Comprehensive experiments demonstrate state-of-the-art performance, where OmniReason-Agent achieves significant improvements in both open-loop planning tasks and visual question answering (VQA) benchmarks, while establishing new capabilities for interpretable, temporally-aware autonomous vehicles operating in complex, dynamic environments.

Paperid: 2816, https://arxiv.org/pdf/2508.19639.pdf

Abstract:
We present FakeSV-VLM in this paper, a new VLM-based framework for detecting fake news on short video platforms. Despite significant efforts to combat this issue due to the severe threat that fake news videos pose to public information security, existing methods still fall short in detection accuracy, often due to lack of knowledge to verify the news is real or not. However, large Vision Language Models (VLMs) have absorbed extensive real-world knowledge from massive multimodal datasets. Motivated by this, we adapt advanced VLMs for fake news detection in short videos. Upon close examination of news samples, we observe that short video samples can be categorized into four distinct scenarios: both video and text are real (for real samples), or both are fake, or either the video or text is fake (for fake samples). Inspired by this insight, we design four experts tailored to handle each scenario and integrate them into VLM via Mixture of Experts. Specifically, we develop the Progressive MoE Adapter (PMOE) module where detection experts first provide an initial analysis, followed by attribution experts for a comprehensive diagnosis, leading to a robust decision. Additionally, we also note the fake news videos often show inconsistency between two modalities. Consequently, we further design the Alignment-driven Event Checking (ADEC) module, which perceives the fake news by capturing the inconsistency between different modalities. Extensive experiments on two benchmark datasets, FakeSV and FakeTT, verify the superiority of our model. It significantly outperforms current state-of-the-art models by +3.32% and +5.02%, establishing a new benchmark in the field.

Paperid: 2817, https://arxiv.org/pdf/2508.11538.pdf

Abstract:
Video reasoning segmentation (VRS) endeavors to delineate referred objects in videos guided by implicit instructions that encapsulate human intent and temporal logic. Previous approaches leverage large vision language models (LVLMs) to encode object semantics into tokens for mask prediction. However, this paradigm suffers from limited interpretability during inference and suboptimal performance due to inadequate spatiotemporal reasoning. Drawing inspiration from seminal breakthroughs in reinforcement learning, we introduce Veason-R1, a specialized LVLM for VRS that emphasizes structured reasoning in segmentation. Veason-R1 is trained through Group Relative Policy Optimization (GRPO) augmented with Chain-of-Thought (CoT) initialization. To begin with, we curate high-quality CoT training data to instill structured reasoning trajectories, bridging video-level semantics and frame-level spatial grounding, yielding the supervised fine-tuned model Veason-SFT. Subsequently, GRPO fine-tuning encourages efficient exploration of the reasoning space by optimizing reasoning chains. To this end, we incorporate a holistic reward mechanism that synergistically enhances spatial alignment and temporal consistency, bolstering keyframe localization and fine-grained grounding. Comprehensive empirical evaluations demonstrate that Veason-R1 achieves state-of-the-art performance on multiple benchmarks, surpassing prior art by significant margins (e.g., +1.3 J &F in ReVOS and +10.0 J &F in ReasonVOS), while exhibiting robustness to hallucinations (+8.8 R). Our code and model weights will be available at Veason-R1.

Paperid: 2818, https://arxiv.org/pdf/2508.01579.pdf

Abstract:
Continual learning (CL) aims to equip models with the ability to learn from a stream of tasks without forgetting previous knowledge. With the progress of vision-language models like Contrastive Language-Image Pre-training (CLIP), their promise for CL has attracted increasing attention due to their strong generalizability. However, the potential of rich textual semantic priors in CLIP in addressing the stability-plasticity dilemma remains underexplored. During backbone training, most approaches transfer past knowledge without considering semantic relevance, leading to interference from unrelated tasks that disrupt the balance between stability and plasticity. Besides, while text-based classifiers provide strong generalization, they suffer from limited plasticity due to the inherent modality gap in CLIP. Visual classifiers help bridge this gap, but their prototypes lack rich and precise semantics. To address these challenges, we propose Semantic-Enriched Continual Adaptation (SECA), a unified framework that harnesses the anti-forgetting and structured nature of textual priors to guide semantic-aware knowledge transfer in the backbone and reinforce the semantic structure of the visual classifier. Specifically, a Semantic-Guided Adaptive Knowledge Transfer (SG-AKT) module is proposed to assess new images' relevance to diverse historical visual knowledge via textual cues, and aggregate relevant knowledge in an instance-adaptive manner as distillation signals. Moreover, a Semantic-Enhanced Visual Prototype Refinement (SE-VPR) module is introduced to refine visual prototypes using inter-class semantic relations captured in class-wise textual embeddings. Extensive experiments on multiple benchmarks validate the effectiveness of our approach.

Paperid: 2819, https://arxiv.org/pdf/2507.20879.pdf

Abstract:
Vision-Language Models (VLMs) are advancing autonomous driving, yet their potential is constrained by myopic decision-making and passive perception, limiting reliability in complex environments. We introduce DriveAgent-R1 to tackle these challenges in long-horizon, high-level behavioral decision-making. DriveAgent-R1 features two core innovations: a Hybrid-Thinking framework that adaptively switches between efficient text-based and in-depth tool-based reasoning, and an Active Perception mechanism with a vision toolkit to proactively resolve uncertainties, thereby balancing decision-making efficiency and reliability. The agent is trained using a novel, three-stage progressive reinforcement learning strategy designed to master these hybrid capabilities. Extensive experiments demonstrate that DriveAgent-R1 achieves state-of-the-art performance, outperforming even leading proprietary large multimodal models, such as Claude Sonnet 4. Ablation studies validate our approach and confirm that the agent's decisions are robustly grounded in actively perceived visual evidence, paving a path toward safer and more intelligent autonomous systems.

Paperid: 2820, https://arxiv.org/pdf/2507.13061.pdf

Abstract:
Scene understanding is one of the core tasks in computer vision, aiming to extract semantic information from images to identify objects, scene categories, and their interrelationships. Although advancements in Vision-Language Models (VLMs) have driven progress in this field, existing VLMs still face challenges in adaptation to unseen complex wide-area scenes. To address the challenges, this paper proposes a Hierarchical Coresets Selection (HCS) mechanism to advance the adaptation of VLMs in complex wide-area scene understanding. It progressively refines the selected regions based on the proposed theoretically guaranteed importance function, which considers utility, representativeness, robustness, and synergy. Without requiring additional fine-tuning, HCS enables VLMs to achieve rapid understandings of unseen scenes at any scale using minimal interpretable regions while mitigating insufficient feature density. HCS is a plug-and-play method that is compatible with any VLM. Experiments demonstrate that HCS achieves superior performance and universality in various tasks.

Paperid: 2821, https://arxiv.org/pdf/2507.03657.pdf

Abstract:
With the increasing attention to pre-trained vision-language models (VLMs), \eg, CLIP, substantial efforts have been devoted to many downstream tasks, especially in test-time adaptation (TTA). However, previous works focus on learning prototypes only in the textual modality while overlooking the ambiguous semantics in class names. These ambiguities lead to textual prototypes that are insufficient to capture visual concepts, resulting in limited performance. To address this issue, we introduce \textbf{ProtoMM}, a training-free framework that constructs multimodal prototypes to adapt VLMs during the test time. By viewing the prototype as a discrete distribution over the textual descriptions and visual particles, ProtoMM has the ability to combine the multimodal features for comprehensive prototype learning. More importantly, the visual particles are dynamically updated as the testing stream flows. This allows our multimodal prototypes to continually learn from the data, enhancing their generalizability in unseen scenarios. In addition, we quantify the importance of the prototypes and test images by formulating their semantic distance as an optimal transport problem. Extensive experiments on 15 zero-shot benchmarks demonstrate the effectiveness of our method, achieving a 1.03\% average accuracy improvement over state-of-the-art methods on ImageNet and its variant datasets.

Paperid: 2822, https://arxiv.org/pdf/2506.05667.pdf

Abstract:
Vision-Language-Action (VLA) models have advanced autonomous driving, but existing benchmarks still lack scenario diversity, reliable action-level annotation, and evaluation protocols aligned with human preferences. To address these limitations, we introduce DriveAction, the first action-driven benchmark specifically designed for VLA models, comprising 16,185 QA pairs generated from 2,610 driving scenarios. DriveAction leverages real-world driving data proactively collected by users of production-level autonomous vehicles to ensure broad and representative scenario coverage, offers high-level discrete action labels collected directly from users' actual driving operations, and implements an action-rooted tree-structured evaluation framework that explicitly links vision, language, and action tasks, supporting both comprehensive and task-specific assessment. Our experiments demonstrate that state-of-the-art vision-language models (VLMs) require both vision and language guidance for accurate action prediction: on average, accuracy drops by 3.3% without vision input, by 4.1% without language input, and by 8.0% without either. Our evaluation supports precise identification of model bottlenecks with robust and consistent results, thus providing new insights and a rigorous foundation for advancing human-like decisions in autonomous driving.

Paperid: 2823, https://arxiv.org/pdf/2505.23065.pdf

Abstract:
With the increasing integration of visual and textual content in Social Networking Services (SNS), evaluating the multimodal capabilities of Large Language Models (LLMs) is crucial for enhancing user experience, content understanding, and platform intelligence. Existing benchmarks primarily focus on text-centric tasks, lacking coverage of the multimodal contexts prevalent in modern SNS ecosystems. In this paper, we introduce SNS-Bench-VL, a comprehensive multimodal benchmark designed to assess the performance of Vision-Language LLMs in real-world social media scenarios. SNS-Bench-VL incorporates images and text across 8 multimodal tasks, including note comprehension, user engagement analysis, information retrieval, and personalized recommendation. It comprises 4,001 carefully curated multimodal question-answer pairs, covering single-choice, multiple-choice, and open-ended tasks. We evaluate over 25 state-of-the-art multimodal LLMs, analyzing their performance across tasks. Our findings highlight persistent challenges in multimodal social context comprehension. We hope SNS-Bench-VL will inspire future research towards robust, context-aware, and human-aligned multimodal intelligence for next-generation social networking services.

Paperid: 2824, https://arxiv.org/pdf/2505.17779.pdf

Abstract:
Ultrasound is a widely-used imaging modality critical to global healthcare, yet its interpretation remains challenging due to its varying image quality on operators, noises, and anatomical structures. Although large vision-language models (LVLMs) have demonstrated impressive multimodal capabilities across natural and medical domains, their performance on ultrasound remains largely unexplored. We introduce U2-BENCH, the first comprehensive benchmark to evaluate LVLMs on ultrasound understanding across classification, detection, regression, and text generation tasks. U2-BENCH aggregates 7,241 cases spanning 15 anatomical regions and defines 8 clinically inspired tasks, such as diagnosis, view recognition, lesion localization, clinical value estimation, and report generation, across 50 ultrasound application scenarios. We evaluate 20 state-of-the-art LVLMs, both open- and closed-source, general-purpose and medical-specific. Our results reveal strong performance on image-level classification, but persistent challenges in spatial reasoning and clinical language generation. U2-BENCH establishes a rigorous and unified testbed to assess and accelerate LVLM research in the uniquely multimodal domain of medical ultrasound imaging.

Paperid: 2825, https://arxiv.org/pdf/2505.17654.pdf

Abstract:
E-commerce platforms increasingly rely on Large Language Models (LLMs) and Vision-Language Models (VLMs) to detect illicit or misleading product content. However, these models remain vulnerable to evasive content: inputs (text or images) that superficially comply with platform policies while covertly conveying prohibited claims. Unlike traditional adversarial attacks that induce overt failures, evasive content exploits ambiguity and context, making it far harder to detect. Existing robustness benchmarks provide little guidance for this demanding, real-world challenge. We introduce EVADE, the first expert-curated, Chinese, multimodal benchmark specifically designed to evaluate foundation models on evasive content detection in e-commerce. The dataset contains 2,833 annotated text samples and 13,961 images spanning six demanding product categories, including body shaping, height growth, and health supplements. Two complementary tasks assess distinct capabilities: Single-Violation, which probes fine-grained reasoning under short prompts, and All-in-One, which tests long-context reasoning by merging overlapping policy rules into unified instructions. Notably, the All-in-One setting significantly narrows the performance gap between partial and full-match accuracy, suggesting that clearer rule definitions improve alignment between human and model judgment. We benchmark 26 mainstream LLMs and VLMs and observe substantial performance gaps: even state-of-the-art models frequently misclassify evasive samples. By releasing EVADE and strong baselines, we provide the first rigorous standard for evaluating evasive-content detection, expose fundamental limitations in current multimodal reasoning, and lay the groundwork for safer and more transparent content moderation systems in e-commerce. The dataset is publicly available at https://huggingface.co/datasets/koenshen/EVADE-Bench.

Paperid: 2826, https://arxiv.org/pdf/2504.19627.pdf

Abstract:
Large Vision-Language Models (LVLMs) are pivotal for real-world AI tasks like embodied intelligence due to their strong vision-language reasoning abilities. However, current LVLMs process entire images at the token level, which is inefficient compared to humans who analyze information and generate content at the conceptual level, extracting relevant visual concepts with minimal effort. This inefficiency, stemming from the lack of a visual concept model, limits LVLMs' usability in real-world applications. To address this, we propose VCM, an end-to-end self-supervised visual concept modeling framework. VCM leverages implicit contrastive learning across multiple sampled instances and vision-language fine-tuning to construct a visual concept model without requiring costly concept-level annotations. Our results show that VCM significantly reduces computational costs (e.g., 85\% fewer FLOPs for LLaVA-1.5-7B) while maintaining strong performance across diverse image understanding tasks. Moreover, VCM enhances visual encoders' capabilities in classic visual concept perception tasks. Extensive quantitative and qualitative experiments validate the effectiveness and efficiency of VCM.

Paperid: 2827, https://arxiv.org/pdf/2504.15271.pdf

Abstract:
We introduce Eagle 2.5, a family of frontier vision-language models (VLMs) for long-context multimodal learning. Our work addresses the challenges in long video comprehension and high-resolution image understanding, introducing a generalist framework for both tasks. The proposed training framework incorporates Automatic Degrade Sampling and Image Area Preservation, two techniques that preserve contextual integrity and visual details. The framework also includes numerous efficiency optimizations in the pipeline for long-context data training. Finally, we propose Eagle-Video-110K, a novel dataset that integrates both story-level and clip-level annotations, facilitating long-video understanding. Eagle 2.5 demonstrates substantial improvements on long-context multimodal benchmarks, providing a robust solution to the limitations of existing VLMs. Notably, our best model Eagle 2.5-8B achieves 72.4% on Video-MME with 512 input frames, matching the results of top-tier commercial model such as GPT-4o and large-scale open-source models like Qwen2.5-VL-72B and InternVL2.5-78B.

Paperid: 2828, https://arxiv.org/pdf/2504.04348.pdf

Abstract:
The advances in vision-language models (VLMs) have led to a growing interest in autonomous driving to leverage their strong reasoning capabilities. However, extending these capabilities from 2D to full 3D understanding is crucial for real-world applications. To address this challenge, we propose OmniDrive, a holistic vision-language dataset that aligns agent models with 3D driving tasks through counterfactual reasoning. This approach enhances decision-making by evaluating potential scenarios and their outcomes, similar to human drivers considering alternative actions. Our counterfactual-based synthetic data annotation process generates large-scale, high-quality datasets, providing denser supervision signals that bridge planning trajectories and language-based reasoning. Futher, we explore two advanced OmniDrive-Agent frameworks, namely Omni-L and Omni-Q, to assess the importance of vision-language alignment versus 3D perception, revealing critical insights into designing effective LLM-agents. Significant improvements on the DriveLM Q\&A benchmark and nuScenes open-loop planning demonstrate the effectiveness of our dataset and methods.

Paperid: 2829, https://arxiv.org/pdf/2504.03154.pdf

Abstract:
Conventional Vision-Language Models(VLMs) typically utilize a fixed number of vision tokens, regardless of task complexity. This one-size-fits-all strategy introduces notable inefficiencies: using excessive tokens leads to unnecessary computational overhead in simpler tasks, whereas insufficient tokens compromise fine-grained visual comprehension in more complex contexts. To overcome these limitations, we present TokenFLEX, an innovative and adaptable vision-language framework that encodes images into a variable number of tokens for efficient integration with a Large Language Model (LLM). Our approach is underpinned by two pivotal innovations. Firstly, we present a novel training paradigm that enhances performance across varying numbers of vision tokens by stochastically modulating token counts during training. Secondly, we design a lightweight vision token projector incorporating an adaptive pooling layer and SwiGLU, allowing for flexible downsampling of vision tokens and adaptive selection of features tailored to specific token counts. Comprehensive experiments reveal that TokenFLEX consistently outperforms its fixed-token counterparts, achieving notable performance gains across various token counts enhancements of 1.6%, 1.0%, and 0.4% with 64, 144, and 256 tokens, respectively averaged over eight vision-language benchmarks. These results underscore TokenFLEX's remarkable flexibility while maintaining high-performance vision-language understanding.

Paperid: 2830, https://arxiv.org/pdf/2504.02132.pdf

Abstract:
Retrieval-augmented generation (RAG) is instrumental for inhibiting hallucinations in large language models (LLMs) through the use of a factual knowledge base (KB). Although PDF documents are prominent sources of knowledge, text-based RAG pipelines are ineffective at capturing their rich multi-modal information. In contrast, visual document RAG (VD-RAG) uses screenshots of document pages as the KB, which has been shown to achieve state-of-the-art results. However, by introducing the image modality, VD-RAG introduces new attack vectors for adversaries to disrupt the system by injecting malicious documents into the KB. In this paper, we demonstrate the vulnerability of VD-RAG to poisoning attacks targeting both retrieval and generation. We define two attack objectives and demonstrate that both can be realized by injecting only a single adversarial image into the KB. Firstly, we introduce a targeted attack against one or a group of queries with the goal of spreading targeted disinformation. Secondly, we present a universal attack that, for any potential user query, influences the response to cause a denial-of-service in the VD-RAG system. We investigate the two attack objectives under both white-box and black-box assumptions, employing a multi-objective gradient-based optimization approach as well as prompting state-of-the-art generative models. Using two visual document datasets, a diverse set of state-of-the-art retrievers (embedding models) and generators (vision language models), we show VD-RAG is vulnerable to poisoning attacks in both the targeted and universal settings, yet demonstrating robustness to black-box attacks in the universal setting.

Paperid: 2831, https://arxiv.org/pdf/2502.13509.pdf

Abstract:
Large language models (LLMs) have shown remarkable performance in vision-language tasks, but their application in the medical field remains underexplored, particularly for integrating structured time series data with unstructured clinical notes. In clinical practice, dynamic time series data, such as lab test results, capture critical temporal patterns, while clinical notes provide rich semantic context. Merging these modalities is challenging due to the inherent differences between continuous signals and discrete text. To bridge this gap, we introduce ProMedTS, a novel self-supervised multimodal framework that employs prompt-guided learning to unify these heterogeneous data types. Our approach leverages lightweight anomaly detection to generate anomaly captions that serve as prompts, guiding the encoding of raw time series data into informative prompt embeddings. These prompt embeddings are aligned with textual representations in a shared latent space, preserving fine-grained temporal nuances alongside semantic insights. Furthermore, our framework incorporates tailored self-supervised objectives to enhance both intra- and inter-modal alignment. We evaluate ProMedTS on disease diagnosis tasks using real-world datasets, and the results demonstrate that our method consistently outperforms state-of-the-art approaches.

Paperid: 2832, https://arxiv.org/pdf/2501.14818.pdf

Abstract:
Recently, promising progress has been made by open-source vision-language models (VLMs) in bringing their capabilities closer to those of proprietary frontier models. However, most open-source models only publish their final model weights, leaving the critical details of data strategies and implementation largely opaque. In this work, we address VLM post-training from a data-centric perspective, showing the key role of data strategy in developing frontier VLMs. By studying and building our post-training data strategy from scratch, we share detailed insights into the development processes, aiming to benefit the development of competitive models for the open-source community. Our introduced data strategy, together with training recipes and model design, leads to a family of performant VLMs named Eagle2. Specifically, Eagle2-9B achieves state-of-the-art results across various multimodal benchmarks, matching certain competitive models with up to 70B parameters.

Paperid: 2833, https://arxiv.org/pdf/2501.13707.pdf

Abstract:
The event-based Vision-Language Model (VLM) recently has made good progress for practical vision tasks. However, most of these works just utilize CLIP for focusing on traditional perception tasks, which obstruct model understanding explicitly the sufficient semantics and context from event streams. To address the deficiency, we propose EventVL, the first generative event-based MLLM (Multimodal Large Language Model) framework for explicit semantic understanding. Specifically, to bridge the data gap for connecting different modalities semantics, we first annotate a large event-image/video-text dataset, containing almost 1.4 million high-quality pairs of data, which enables effective learning across various scenes, e.g., drive scene or human motion. After that, we design Event Spatiotemporal Representation to fully explore the comprehensive information by diversely aggregating and segmenting the event stream. To further promote a compact semantic space, Dynamic Semantic Alignment is introduced to improve and complete sparse semantic spaces of events. Extensive experiments show that our EventVL can significantly surpass existing MLLM baselines in event captioning and scene description generation tasks. We hope our research could contribute to the development of the event vision community.

Paperid: 2834, https://arxiv.org/pdf/2501.02200.pdf

Abstract:
Evolutionary algorithms (EAs) maintain populations through evolutionary operators to discover diverse solutions for complex tasks while gathering valuable knowledge, such as historical population data and fitness evaluations. However, traditional EAs face challenges in dynamically adapting to expanding knowledge bases, hindering the efficient exploitation of accumulated information and limiting adaptability to new situations. To address these issues, we introduce an Optimization Knowledge Adaptation Evolutionary Model (OKAEM), which features dynamic parameter adjustment using accumulated knowledge to enhance its optimization capabilities. OKAEM employs attention mechanisms to model the interactions among individuals, fitness landscapes, and genetic components separately, thereby parameterizing the evolutionary operators of selection, crossover, and mutation. These powerful learnable operators enable OKAEM to benefit from pre-learned extensive prior knowledge and self-tune with real-time evolutionary insights. Experimental results demonstrate that OKAEM: 1) exploits prior knowledge for significant performance gains across various knowledge transfer settings; 2) achieves competitive performance through self-tuning alone, even without prior knowledge; 3) outperforms state-of-the-art black-box baselines in a vision-language model tuning case; 4) can improve its optimization capabilities with growing knowledge; 5) is capable of emulating principles of natural selection and genetic recombination.

Paperid: 2835, https://arxiv.org/pdf/2410.23907.pdf

Abstract:
Multi-Object Tracking (MOT) aims to associate multiple objects across video frames and is a challenging vision task due to inherent complexities in the tracking environment. Most existing approaches train and track within a single domain, resulting in a lack of cross-domain generalizability to data from other domains. While several works have introduced natural language representation to bridge the domain gap in visual tracking, these textual descriptions often provide too high-level a view and fail to distinguish various instances within the same class. In this paper, we address this limitation by developing IP-MOT, an end-to-end transformer model for MOT that operates without concrete textual descriptions. Our approach is underpinned by two key innovations: Firstly, leveraging a pre-trained vision-language model, we obtain instance-level pseudo textual descriptions via prompt-tuning, which are invariant across different tracking scenes; Secondly, we introduce a query-balanced strategy, augmented by knowledge distillation, to further boost the generalization capabilities of our model. Extensive experiments conducted on three widely used MOT benchmarks, including MOT17, MOT20, and DanceTrack, demonstrate that our approach not only achieves competitive performance on same-domain data compared to state-of-the-art models but also significantly improves the performance of query-based trackers by large margins for cross-domain inputs.

Paperid: 2836, https://arxiv.org/pdf/2409.09291.pdf

Abstract:
Image fusion combines images from multiple domains into one image, containing complementary information from source domains. Existing methods take pixel intensity, texture and high-level vision task information as the standards to determine preservation of information, lacking enhancement for human perception. We introduce an image fusion method, Hierarchical Perception Fusion (HPFusion), which leverages Large Vision-Language Model to incorporate hierarchical human semantic priors, preserving complementary information that satisfies human visual system. We propose multiple questions that humans focus on when viewing an image pair, and answers are generated via the Large Vision-Language Model according to images. The texts of answers are encoded into the fusion network, and the optimization also aims to guide the human semantic distribution of the fused image more similarly to source images, exploring complementary information within the human perception domain. Extensive experiments demonstrate our HPFusoin can achieve high-quality fusion results both for information preservation and human visual enhancement.

Paperid: 2837, https://arxiv.org/pdf/2601.22150.pdf

Abstract:
Large Vision-Language Models (VLMs) often answer classic visual illusions "correctly" on original images, yet persist with the same responses when illusion factors are inverted, even though the visual change is obvious to humans. This raises a fundamental question: do VLMs perceive visual changes or merely recall memorized patterns? While several studies have noted this phenomenon, the underlying causes remain unclear. To move from observations to systematic understanding, this paper introduces VI-Probe, a controllable visual-illusion framework with graded perturbations and matched visual controls (without illusion inducer) that disentangles visually grounded perception from language-driven recall. Unlike prior work that focuses on averaged accuracy, we measure stability and sensitivity using Polarity-Flip Consistency, Template Fixation Index, and an illusion multiplier normalized against matched controls. Experiments across different families reveal that response persistence arises from heterogeneous causes rather than a single mechanism. For instance, GPT-5 exhibits memory override, Claude-Opus-4.1 shows perception-memory competition, while Qwen variants suggest visual-processing limits. Our findings challenge single-cause views and motivate probing-based evaluation that measures both knowledge and sensitivity to controlled visual change. Data and code are available at https://sites.google.com/view/vi-probe/.

Paperid: 2838, https://arxiv.org/pdf/2601.18188.pdf

Abstract:
Vision-and-Language Navigation (VLN) requires agents to interpret natural language instructions and act coherently in visually rich environments. However, most existing methods rely on reactive state-action mappings without explicitly modeling how actions causally transform subsequent visual observations. Lacking such vision-action causality, agents cannot anticipate the visual changes induced by its own actions, leading to unstable behaviors, weak generalization, and cumulative error along trajectory. To address these issues, we introduce \textsc{NaVIDA} (\textbf{Nav}igation with \textbf{I}nverse \textbf{D}ynamics \textbf{A}ugmentation), a unified VLN framework that couples policy learning with action-grounded visual dynamics and adaptive execution. \textsc{NaVIDA} augments training with chunk-based inverse-dynamics supervision to learn causal relationship between visual changes and corresponding actions. To structure this supervision and extend the effective planning range, \textsc{NaVIDA} employs hierarchical probabilistic action chunking (HPAC), which organizes trajectories into multi-step chunks and provides discriminative, longer-range visual-change cues. To further curb error accumulation and stabilize behavior at inference, an entropy-guided mechanism adaptively sets the execution horizon of action chunks. Extensive experiments show that \textsc{NaVIDA} achieves superior navigation performance compared to state-of-the-art methods with fewer parameters (3B vs. 8B). Real-world robot evaluations further validate the practical feasibility and effectiveness of our approach. Code and data will be available upon acceptance.

Paperid: 2839, https://arxiv.org/pdf/2601.17082.pdf

Abstract:
Despite substantial efforts toward improving the moral alignment of Vision-Language Models (VLMs), it remains unclear whether their ethical judgments are stable in realistic settings. This work studies moral robustness in VLMs, defined as the ability to preserve moral judgments under textual and visual perturbations that do not alter the underlying moral context. We systematically probe VLMs with a diverse set of model-agnostic multimodal perturbations and find that their moral stances are highly fragile, frequently flipping under simple manipulations. Our analysis reveals systematic vulnerabilities across perturbation types, moral domains, and model scales, including a sycophancy trade-off where stronger instruction-following models are more susceptible to persuasion. We further show that lightweight inference-time interventions can partially restore moral stability. These results demonstrate that moral alignment alone is insufficient and that moral robustness is a necessary criterion for the responsible deployment of VLMs.

Paperid: 2840, https://arxiv.org/pdf/2601.14722.pdf

Abstract:
Document extraction is a core component of digital workflows, yet existing vision-language models (VLMs) predominantly favor high-resource languages. Thai presents additional challenges due to script complexity from non-latin letters, the absence of explicit word boundaries, and the prevalence of highly unstructured real-world documents, limiting the effectiveness of current open-source models. This paper presents Typhoon OCR, an open VLM for document extraction tailored for Thai and English. The model is fine-tuned from vision-language backbones using a Thai-focused training dataset. The dataset is developed using a multi-stage data construction pipeline that combines traditional OCR, VLM-based restructuring, and curated synthetic data. Typhoon OCR is a unified framework capable of text transcription, layout reconstruction, and document-level structural consistency. The latest iteration of our model, Typhoon OCR V1.5, is a compact and inference-efficient model designed to reduce reliance on metadata and simplify deployment. Comprehensive evaluations across diverse Thai document categories, including financial reports, government forms, books, infographics, and handwritten documents, show that Typhoon OCR achieves performance comparable to or exceeding larger frontier proprietary models, despite substantially lower computational cost. The results demonstrate that open vision-language OCR models can achieve accurate text extraction and layout reconstruction for Thai documents, reaching performance comparable to proprietary systems while remaining lightweight and deployable.

Paperid: 2841, https://arxiv.org/pdf/2601.03733.pdf

Abstract:
Understanding how two radiology image sets differ is critical for generating clinical insights and for interpreting medical AI systems. We introduce RadDiff, a multimodal agentic system that performs radiologist-style comparative reasoning to describe clinically meaningful differences between paired radiology studies. RadDiff builds on a proposer-ranker framework from VisDiff, and incorporates four innovations inspired by real diagnostic workflows: (1) medical knowledge injection through domain-adapted vision-language models; (2) multimodal reasoning that integrates images with their clinical reports; (3) iterative hypothesis refinement across multiple reasoning rounds; and (4) targeted visual search that localizes and zooms in on salient regions to capture subtle findings. To evaluate RadDiff, we construct RadDiffBench, a challenging benchmark comprising 57 expert-validated radiology study pairs with ground-truth difference descriptions. On RadDiffBench, RadDiff achieves 47% accuracy, and 50% accuracy when guided by ground-truth reports, significantly outperforming the general-domain VisDiff baseline. We further demonstrate RadDiff's versatility across diverse clinical tasks, including COVID-19 phenotype comparison, racial subgroup analysis, and discovery of survival-related imaging features. Together, RadDiff and RadDiffBench provide the first method-and-benchmark foundation for systematically uncovering meaningful differences in radiological data.

Paperid: 2842, https://arxiv.org/pdf/2512.23512.pdf

Abstract:
Vision-language large models are moving toward the unification of visual understanding and visual generation tasks. However, whether generation can enhance understanding is still under-explored on large data scale. In this work, we analysis the unified structure with a concise model, UniHetero, under large-scale pretraining (>200M samples). Our key observations are: (1) Generation can improve understanding, but Only if you generate Semantics, Not Pixels. A common assumption in unified vision-language models is that adding generation will naturally strengthen understanding. However, this is not always true at scale. At 200M+ pretraining samples, generation helps understanding only when it operates at the semantic level, i.e. when the model learns to autoregress high-level visual representations inside the LLM. Once pixel-level objectives (e.g., diffusion losses) directly interfere with the LLM, understanding performance often degrades. (2) Generation reveals a superior Data Scaling trend and higher Data Utilization. Unified generation-understanding demonstrates a superior scaling trend compared to understanding alone, revealing a more effective way to learn vision-only knowledge directive from vision modality rather than captioning to text. (3) Autoregression on Input Embedding is effective to capture visual details. Compared to the commonly-used vision encoder, make visual autoregression on input embedding shows less cumulative error and is modality independent, which can be extend to all modalities. The learned semantic representations capture visual information such as objects, locations, shapes, and colors; further enable pixel-level image generation.

Paperid: 2843, https://arxiv.org/pdf/2512.17221.pdf

Abstract:
While Vision-language models (VLMs) have demonstrated remarkable performance across multi-modal tasks, their choice of vision encoders presents a fundamental weakness: their low-level features lack the robust structural and spatial information essential for document understanding and web agents. To bridge this gap, we introduce DAVE, a vision encoder purpose-built for VLMs and tailored for these tasks. Our training pipeline is designed to leverage abundant unlabeled data to bypass the need for costly large-scale annotations for document and web images. We begin with a self-supervised pretraining stage on unlabeled images, followed by a supervised autoregressive pretraining stage, where the model learns tasks like parsing and localization from limited, high-quality data. Within the supervised stage, we adopt two strategies to improve our encoder's alignment with both general visual knowledge and diverse document and web agentic tasks: (i) We introduce a novel model-merging scheme, combining encoders trained with different text decoders to ensure broad compatibility with different web agentic architectures. (ii) We use ensemble training to fuse features from pretrained generalist encoders (e.g., SigLIP2) with our own document and web-specific representations. Extensive experiments on classic document tasks, VQAs, web localization, and agent-based benchmarks validate the effectiveness of our approach, establishing DAVE as a strong vision encoder for document and web applications.

Paperid: 2844, https://arxiv.org/pdf/2512.07472.pdf

Abstract:
Vision-Language-Action (VLA) models have shown great performance in robotic manipulation by mapping visual observations and language instructions directly to actions. However, they remain brittle under distribution shifts: when test scenarios change, VLAs often reproduce memorized trajectories instead of adapting to the updated scene, which is a failure mode we refer to as the "Memory Trap". This limitation stems from the end-to-end design, which lacks explicit 3D spatial reasoning and prevents reliable identification of actionable regions in unfamiliar environments. To compensate for this missing spatial understanding, 3D Spatial Affordance Fields (SAFs) can provide a geometric representation that highlights where interactions are physically feasible, offering explicit cues about regions the robot should approach or avoid. We therefore introduce Affordance Field Intervention (AFI), a lightweight hybrid framework that uses SAFs as an on-demand plug-in to guide VLA behavior. Our system detects memory traps through proprioception, repositions the robot to recent high-affordance regions, and proposes affordance-driven waypoints that anchor VLA-generated actions. A SAF-based scorer then selects trajectories with the highest cumulative affordance. Extensive experiments demonstrate that our method achieves an average improvement of 23.5% across different VLA backbones ($π_{0}$ and $π_{0.5}$) under out-of-distribution scenarios on real-world robotic platforms, and 20.2% on the LIBERO-Pro benchmark, validating its effectiveness in enhancing VLA robustness to distribution shifts.

Paperid: 2845, https://arxiv.org/pdf/2512.06746.pdf

Abstract:
Vision Language Models (VLMs) are increasingly adopted for AI-generated images (AIGI) detection, yet converting VLMs into detectors requires substantial resource, while the resulting models still exhibit severe hallucinations. To probe the core issue, we conduct an empirical analysis and observe two characteristic behaviors: (i) fine-tuning VLMs on high-level semantic supervision strengthens semantic discrimination and well generalize to unseen data; (ii) fine-tuning VLMs on low-level pixel-artifact supervision yields poor transfer. We attribute VLMs' underperformance to task-model misalignment: semantics-oriented VLMs inherently lack sensitivity to fine-grained pixel artifacts, and semantically non-discriminative pixel artifacts thus exceeds their inductive biases. In contrast, we observe that conventional pixel-artifact detectors capture low-level pixel artifacts yet exhibit limited semantic awareness relative to VLMs, highlighting that distinct models are better matched to distinct tasks. In this paper, we formalize AIGI detection as two complementary tasks--semantic consistency checking and pixel-artifact detection--and show that neglecting either induces systematic blind spots. Guided by this view, we introduce the Task-Model Alignment principle and instantiate it as a two-branch detector, AlignGemini, comprising a VLM fine-tuned exclusively with pure semantic supervision and a pixel-artifact expert trained exclusively with pure pixel-artifact supervision. By enforcing orthogonal supervision on two simplified datasets, each branch trains to its strengths, producing complementary discrimination over semantic and pixel cues. On five in-the-wild benchmarks, AlignGemini delivers a +9.5 gain in average accuracy, supporting task-model alignment as an effective path to generalizable AIGI detection.

Paperid: 2846, https://arxiv.org/pdf/2512.04446.pdf

Abstract:
Automating disassembly of critical components from end-of-life (EoL) desktops, such as high-value items like RAM modules and CPUs, as well as sensitive parts like hard disk drives, remains challenging due to the inherent variability and uncertainty of these products. Moreover, their disassembly requires sequential, precise, and dexterous operations, further increasing the complexity of automation. Current robotic disassembly processes are typically divided into several stages: perception, sequence planning, task planning, motion planning, and manipulation. Each stage requires explicit modeling, which limits generalization to unfamiliar scenarios. Recent development of vision-language-action (VLA) models has presented an end-to-end approach for general robotic manipulation tasks. Although VLAs have demonstrated promising performance on simple tasks, the feasibility of applying such models to complex disassembly remains largely unexplored. In this paper, we collected a customized dataset for robotic RAM and CPU disassembly and used it to fine-tune two well-established VLA approaches, OpenVLA and OpenVLA-OFT, as a case study. We divided the whole disassembly task into several small steps, and our preliminary experimental results indicate that the fine-tuned VLA models can faithfully complete multiple early steps but struggle with certain critical subtasks, leading to task failure. However, we observed that a simple hybrid strategy that combines VLA with a rule-based controller can successfully perform the entire disassembly operation. These findings highlight the current limitations of VLA models in handling the dexterity and precision required for robotic EoL product disassembly. By offering a detailed analysis of the observed results, this study provides insights that may inform future research to address current challenges and advance end-to-end robotic automated disassembly.

Paperid: 2847, https://arxiv.org/pdf/2511.22697.pdf

Abstract:
Vision-Language Action (VLAs) models promise to extend the remarkable success of vision-language models (VLMs) to robotics. Yet, unlike VLMs in the vision-language domain, VLAs for robotics require finetuning to contend with varying physical factors like robot embodiment, environment characteristics, and spatial relationships of each task. Existing fine-tuning methods lack specificity, adapting the same set of parameters regardless of a task's visual, linguistic, and physical characteristics. Inspired by functional specificity in neuroscience, we hypothesize that it is more effective to finetune sparse model representations specific to a given task. In this work, we introduce Robotic Steering, a finetuning approach grounded in mechanistic interpretability that leverages few-shot demonstrations to identify and selectively finetune task-specific attention heads aligned with the physical, visual, and linguistic requirements of robotic tasks. Through comprehensive on-robot evaluations with a Franka Emika robot arm, we demonstrate that Robotic Steering outperforms LoRA while achieving superior robustness under task variation, reduced computational cost, and enhanced interpretability for adapting VLAs to diverse robotic tasks.

Paperid: 2848, https://arxiv.org/pdf/2511.18983.pdf

Abstract:
In deepfake detection, the varying degrees of compression employed by social media platforms pose significant challenges for model generalization and reliability. Although existing methods have progressed from single-modal to multimodal approaches, they face critical limitations: single-modal methods struggle with feature degradation under data compression in social media streaming, while multimodal approaches require expensive data collection and labeling and suffer from inconsistent modal quality or accessibility in real-world scenarios. To address these challenges, we propose a novel Unimodal-generated Multimodal Contrastive Learning (UMCL) framework for robust cross-compression-rate (CCR) deepfake detection. In the training stage, our approach transforms a single visual modality into three complementary features: compression-robust rPPG signals, temporal landmark dynamics, and semantic embeddings from pre-trained vision-language models. These features are explicitly aligned through an affinity-driven semantic alignment (ASA) strategy, which models inter-modal relationships through affinity matrices and optimizes their consistency through contrastive learning. Subsequently, our cross-quality similarity learning (CQSL) strategy enhances feature robustness across compression rates. Extensive experiments demonstrate that our method achieves superior performance across various compression rates and manipulation types, establishing a new benchmark for robust deepfake detection. Notably, our approach maintains high detection accuracy even when individual features degrade, while providing interpretable insights into feature relationships through explicit alignment.

Paperid: 2849, https://arxiv.org/pdf/2511.14045.pdf

Abstract:
Membership inference attacks (MIAs) on large language models (LLMs) pose significant privacy risks across various stages of model training. Recent advances in Reinforcement Learning with Verifiable Rewards (RLVR) have brought a profound paradigm shift in LLM training, particularly for complex reasoning tasks. However, the on-policy nature of RLVR introduces a unique privacy leakage pattern: since training relies on self-generated responses without fixed ground-truth outputs, membership inference must now determine whether a given prompt (independent of any specific response) is used during fine-tuning. This creates a threat where leakage arises not from answer memorization. To audit this novel privacy risk, we propose Divergence-in-Behavior Attack (DIBA), the first membership inference framework specifically designed for RLVR. DIBA shifts the focus from memorization to behavioral change, leveraging measurable shifts in model behavior across two axes: advantage-side improvement (e.g., correctness gain) and logit-side divergence (e.g., policy drift). Through comprehensive evaluations, we demonstrate that DIBA significantly outperforms existing baselines, achieving around 0.8 AUC and an order-of-magnitude higher TPR@0.1%FPR. We validate DIBA's superiority across multiple settings--including in-distribution, cross-dataset, cross-algorithm, black-box scenarios, and extensions to vision-language models. Furthermore, our attack remains robust under moderate defensive measures. To the best of our knowledge, this is the first work to systematically analyze privacy vulnerabilities in RLVR, revealing that even in the absence of explicit supervision, training data exposure can be reliably inferred through behavioral traces.

Paperid: 2850, https://arxiv.org/pdf/2511.05553.pdf

Abstract:
In complex embodied long-horizon manipulation tasks, effective task decomposition and execution require synergistic integration of textual logical reasoning and visual-spatial imagination to ensure efficient and accurate operation. Current methods fail to adopt a unified generation framework for multimodal planning, lead to inconsistent in multimodal planning. To address this challenge, we present \textbf{EVLP (Embodied Vision-Language Planner)}, an innovative multimodal unified generation framework that jointly models linguistic reasoning and visual generation. Our approach achieves multimodal planning for long-horizon tasks through a novel training pipeline incorporating dynamic pretraining and reinforced alignment. Our core innovations consist of three key components: \textbf{1) Unified Multimodal Generation Framework}: For understanding, We integrate semantic information with spatial features to provide comprehensive visual perception. For generation, we directly learn the joint distribution of discrete images for one-step visual synthesis, enabling coordinated language-visual modeling through learnable cross-modal attention mechanisms. \textbf{2) Dynamic Perception Pretraining}: We propose a bidirectional dynamic alignment strategy employing inverse dynamics tasks and forward dynamics tasks, effectively strengthening multimodal correlations within a unified feature space. \textbf{3) Reinforced Supervised Fine-Tuning}: While conducting instruction-based fine-tuning in the unified generation space, we construct a reinforce loss to align the spatial logic between textual actions and generated images, enabling the model to acquire spatio-awared multimodal planning capabilities.

Paperid: 2851, https://arxiv.org/pdf/2510.13796.pdf

Abstract:
Symbol grounding (Harnad, 1990) describes how symbols such as words acquire their meanings by connecting to real-world sensorimotor experiences. Recent work has shown preliminary evidence that grounding may emerge in (vision-)language models trained at scale without using explicit grounding objectives. Yet, the specific loci of this emergence and the mechanisms that drive it remain largely unexplored. To address this problem, we introduce a controlled evaluation framework that systematically traces how symbol grounding arises within the internal computations through mechanistic and causal analysis. Our findings show that grounding concentrates in middle-layer computations and is implemented through the aggregate mechanism, where attention heads aggregate the environmental ground to support the prediction of linguistic forms. This phenomenon replicates in multimodal dialogue and across architectures (Transformers and state-space models), but not in unidirectional LSTMs. Our results provide behavioral and mechanistic evidence that symbol grounding can emerge in language models, with practical implications for predicting and potentially controlling the reliability of generation.

Paperid: 2852, https://arxiv.org/pdf/2510.11835.pdf

Abstract:
CLIP outperforms self-supervised models like DINO as vision encoders for vision-language models (VLMs), but it remains unclear whether this advantage stems from CLIP's language supervision or its much larger training data. To disentangle these factors, we pre-train CLIP and DINO under controlled settings -- using the same architecture, dataset, and training configuration -- achieving similar ImageNet accuracy. Embedding analysis shows that CLIP captures high-level semantics (e.g., object categories, text), while DINO is more responsive to low-level features like colors and styles. When integrated into VLMs and evaluated on 20 VQA benchmarks, CLIP excels at text-intensive tasks, while DINO slightly outperforms on vision-centric ones. Variants of language supervision (e.g., sigmoid loss, pre-trained language encoders) yield limited gains. Our findings provide scientific insights into vision encoder design and its impact on VLM performance.

Paperid: 2853, https://arxiv.org/pdf/2510.03978.pdf

Abstract:
Embedding vision-language models (VLMs) are typically pretrained with short text windows (<77 tokens), which forces the truncation of long-format captions. Yet, the distribution of biomedical captions from large-scale open source literature reveals that a huge portion of captions far exceed 77 tokens. To this end, we investigate the impact of pretraining on long-format biomedical captions by extending the context length of text encoders in VLMs. We find that longer context (thus, enabling additional supervision provided in long-format captions) correlates with better retrieval and classification performance. Given this finding, we introduce BIOMEDICA-LongCAP, a dataset of 1M image-caption pairs enriched with context-aware descriptions from full-text articles, providing longer and additional textual supervision. Using BIOMEDICA-LongCAP, we train BMC-LongCLIP, a long-context biomedical VLM with a text encoder supporting windows of up to 512 tokens. Our model extends context capacity by 6.6x, reducing token waste from 55% to just 2.2%. On long-caption retrieval benchmarks, BMC-LongCLIP achieves up to +30% absolute gains in Recall@1 and +2% average improvements in classification, while also converging faster than short-context. Our results demonstrate that long-context modeling is a promising direction for advancing biomedical VLMs.

Paperid: 2854, https://arxiv.org/pdf/2510.03721.pdf

Abstract:
Vision-language models trained on large-scale multimodal datasets show strong demographic biases, but the role of training data in producing these biases remains unclear. A major barrier has been the lack of demographic annotations in web-scale datasets such as LAION-400M. We address this gap by creating person-centric annotations for the full dataset, including over 276 million bounding boxes, perceived gender and race/ethnicity labels, and automatically generated captions. These annotations are produced through validated automatic labeling pipelines combining object detection, multimodal captioning, and finetuned classifiers. Using them, we uncover demographic imbalances and harmful associations, such as the disproportionate linking of men and individuals perceived as Black or Middle Eastern with crime-related and negative content. We also show that 60-70% of gender bias in CLIP and Stable Diffusion can be linearly explained by direct co-occurrences in the data. Our resources establish the first large-scale empirical link between dataset composition and downstream model bias.

Paperid: 2855, https://arxiv.org/pdf/2510.00506.pdf

Abstract:
How can we reconstruct 3D hand poses when large portions of the hand are heavily occluded by itself or by objects? Humans often resolve such ambiguities by leveraging contextual knowledge -- such as affordances, where an object's shape and function suggest how the object is typically grasped. Inspired by this observation, we propose a generative prior for hand pose refinement guided by affordance-aware textual descriptions of hand-object interactions (HOI). Our method employs a diffusion-based generative model that learns the distribution of plausible hand poses conditioned on affordance descriptions, which are inferred from a large vision-language model (VLM). This enables the refinement of occluded regions into more accurate and functionally coherent hand poses. Extensive experiments on HOGraspNet, a 3D hand-affordance dataset with severe occlusions, demonstrate that our affordance-guided refinement significantly improves hand pose estimation over both recent regression methods and diffusion-based refinement lacking contextual reasoning.

Paperid: 2856, https://arxiv.org/pdf/2509.25718.pdf

Abstract:
Reinforcement learning (RL) is a promising avenue for post-training vision-language-action (VLA) models, but practical deployment is hindered by sparse rewards and unstable training. This work mitigates these challenges by introducing an action chunk based on proximal policy optimization (PPO) with behavior cloning using self-collected demonstrations. Aggregating consecutive actions into chunks improves the temporal consistency of the policy and the density of informative feedback. In addition, an auxiliary behavior cloning loss is applied with a dynamically updated demonstration buffer that continually collects high-quality task trials during training. The relative weight between the action-chunked PPO objective and the self behavior clone auxiliary loss is adapted online to stabilize the post-training process. Experiments on the MetaWorld benchmark indicate improved performance over supervised fine-tuning, achieving a high success rate (0.93) and few steps to success (42.17). These results demonstrate the viability of RL for VLA post-training and help lay the groundwork for downstream VLA applications.

Paperid: 2857, https://arxiv.org/pdf/2509.25696.pdf

Abstract:
Time-series question answering (TSQA) tasks face significant challenges due to the lack of labeled data. Alternatively, with recent advancements in large-scale models, vision-language models (VLMs) have demonstrated the potential to analyze time-series signals in a zero-shot manner. In this paper, we propose a training approach that uses pseudo labels generated by a VLM. Although VLMs can produce incorrect labels, TSQA models can still be effectively trained based on the property that deep neural networks are inherently robust to such noisy labels. Our experimental results demonstrate that TSQA models are not only successfully trained with pseudo labels, but also surpass the performance of the VLM itself by leveraging a large amount of unlabeled data.

Paperid: 2858, https://arxiv.org/pdf/2509.18041.pdf

Abstract:
Long-Form Video Question Answering (LVQA) poses challenges beyond traditional visual question answering (VQA), which is often limited to static images or short video clips. While current vision-language models (VLMs) perform well in those settings, they struggle with complex queries in LVQA over long videos involving multi-step temporal reasoning and causality. Vanilla approaches, which sample frames uniformly and feed them to a VLM with the question, incur significant token overhead, forcing severe downsampling. As a result, the model often misses fine-grained visual structure, subtle event transitions, or key temporal cues, ultimately leading to incorrect answers. To address these limitations, recent works have explored query-adaptive frame sampling, hierarchical keyframe selection, and agent-based iterative querying. However, these methods remain fundamentally heuristic: they lack explicit temporal representations and cannot enforce or verify logical event relationships. As a result, there are no formal guarantees that the sampled context actually encodes the compositional or causal logic demanded by the question. To address these foundational gaps, we introduce NeuS-QA, a training-free, plug-and-play neuro-symbolic pipeline for LVQA. NeuS-QA translates a natural language question into a formal temporal logic expression, constructs a video automaton from frame-level semantic propositions, and applies model checking to rigorously identify video segments satisfying the question's logical requirements. Only these logic-verified segments are submitted to the VLM, thus improving interpretability, reducing hallucinations, and enabling compositional reasoning without modifying or fine-tuning the model. Experiments on LongVideoBench and CinePile show NeuS-QA improves performance by over 10%, especially on questions involving event ordering, causality, and multi-step compositional reasoning.

Paperid: 2859, https://arxiv.org/pdf/2509.09356.pdf

Abstract:
Navigating and understanding complex and unknown environments autonomously demands more than just basic perception and movement from embodied agents. Truly effective exploration requires agents to possess higher-level cognitive abilities, the ability to reason about their surroundings, and make more informed decisions regarding exploration strategies. However, traditional RL approaches struggle to balance efficient exploration and semantic understanding due to limited cognitive capabilities embedded in the small policies for the agents, leading often to human drivers when dealing with semantic exploration. In this paper, we address this challenge by presenting a novel Deep Reinforcement Learning (DRL) architecture that is specifically designed for resource efficient semantic exploration. A key methodological contribution is the integration of a Vision-Language Model (VLM) common-sense through a layered reward function. The VLM query is modeled as a dedicated action, allowing the agent to strategically query the VLM only when deemed necessary for gaining external guidance, thereby conserving resources. This mechanism is combined with a curriculum learning strategy designed to guide learning at different levels of complexity to ensure robust and stable learning. Our experimental evaluation results convincingly demonstrate that our agent achieves significantly enhanced object discovery rates and develops a learned capability to effectively navigate towards semantically rich regions. Furthermore, it also shows a strategic mastery of when to prompt for external environmental information. By demonstrating a practical and scalable method for embedding common-sense semantic reasoning with autonomous agents, this research provides a novel approach to pursuing a fully intelligent and self-guided exploration in robotics.

Paperid: 2860, https://arxiv.org/pdf/2509.02324.pdf

Abstract:
Language-guided long-horizon manipulation of deformable objects presents significant challenges due to high degrees of freedom, complex dynamics, and the need for accurate vision-language grounding. In this work, we focus on multi-step cloth folding, a representative deformable-object manipulation task that requires both structured long-horizon planning and fine-grained visual perception. To this end, we propose a unified framework that integrates a Large Language Model (LLM)-based planner, a Vision-Language Model (VLM)-based perception system, and a task execution module. Specifically, the LLM-based planner decomposes high-level language instructions into low-level action primitives, bridging the semantic-execution gap, aligning perception with action, and enhancing generalization. The VLM-based perception module employs a SigLIP2-driven architecture with a bidirectional cross-attention fusion mechanism and weight-decomposed low-rank adaptation (DoRA) fine-tuning to achieve language-conditioned fine-grained visual grounding. Experiments in both simulation and real-world settings demonstrate the method's effectiveness. In simulation, it outperforms state-of-the-art baselines by 2.23, 1.87, and 33.3 on seen instructions, unseen instructions, and unseen tasks, respectively. On a real robot, it robustly executes multi-step folding sequences from language instructions across diverse cloth materials and configurations, demonstrating strong generalization in practical scenarios. Project page: https://language-guided.netlify.app/

Paperid: 2861, https://arxiv.org/pdf/2508.16292.pdf

Abstract:
Recently, Vision-Language-Action (VLA) models have demonstrated strong performance on a range of robotic tasks. These models rely on multimodal inputs, with language instructions playing a crucial role -- not only in predicting actions, but also in robustly interpreting user intent, even when the requests are impossible to fulfill. In this work, we investigate how VLAs can recognize, interpret, and respond to false-premise instructions: natural language commands that reference objects or conditions absent from the environment. We propose Instruct-Verify-and-Act (IVA), a unified framework that (i) detects when an instruction cannot be executed due to a false premise, (ii) engages in language-based clarification or correction, and (iii) grounds plausible alternatives in perception and action. Towards this end, we construct a large-scale instruction tuning setup with structured language prompts and train a VLA model capable of handling both accurate and erroneous requests. Our approach leverages a contextually augmented, semi-synthetic dataset containing paired positive and false-premise instructions, enabling robust detection and natural language correction. Our experiments show that IVA improves false premise detection accuracy by 97.56% over baselines, while increasing successful responses in false-premise scenarios by 50.78%.

Paperid: 2862, https://arxiv.org/pdf/2508.03351.pdf

Abstract:
Post-training quantization (PTQ) has emerged as an effective approach for compressing large models and accelerating their inference without retraining. While PTQ has been extensively studied in the context of large language models (LLMs), its applicability to vision-language models (VLMs) remains underexplored. In this paper, we identify a modality discrepancy (\emph{i.e.}, limited text tokens \emph{vs.} excessive and redundant vision tokens) of VLMs. However, existing Hessian-based LLM PTQ methods treat all tokens equally during quantization, resulting in severe performance drops when applied to VLMs. Motivated by this observation, we propose a novel importance-aware PTQ framework tailored for VLMs, dubbed VLMQ. Specifically, to address vision token redundancy, VLMQ 1) optimizes an importance-aware objective that yields an enhanced Hessian with token-level importance factors, while retaining compatibility with parallelized weight updates, and 2) ensures efficiency and effectiveness by computing these factors via a single lightweight block-wise backward pass, guided by a theoretical connection to token-level perturbations. Extensive evaluations on 8 benchmarks across 0.5B$\sim$32B VLMs demonstrate the state-of-the-art (SOTA) performance of our VLMQ, particularly under low-bit settings. For example, it achieves a substantial \textbf{16.45\%} improvement on MME-RealWorld under 2-bit quantization.

Paperid: 2863, https://arxiv.org/pdf/2507.05056.pdf

Abstract:
Hallucinations in large vision-language models (LVLMs) pose significant challenges for real-world applications, as LVLMs may generate responses that appear plausible yet remain inconsistent with the associated visual content. This issue rarely occurs in human cognition. We argue that this discrepancy arises from humans' ability to effectively leverage multimodal interaction information in data samples. Specifically, humans typically first gather multimodal information, analyze the interactions across modalities for understanding, and then express their understanding through language. Motivated by this observation, we conduct extensive experiments on popular LVLMs and obtained insights that surprisingly reveal human-like, though less pronounced, cognitive behavior of LVLMs on multimodal samples. Building on these findings, we further propose \textbf{INTER}: \textbf{Inter}action Guidance Sampling, a novel training-free algorithm that mitigate hallucinations without requiring additional data. Specifically, INTER explicitly guides LVLMs to effectively reapply their understanding of multimodal interaction information when generating responses, thereby reducing potential hallucinations. On six benchmarks including VQA and image captioning tasks, INTER achieves an average improvement of up to 3.4\% on five LVLMs compared to the state-of-the-art decoding strategy. The code will be released when the paper is accepted.

Paperid: 2864, https://arxiv.org/pdf/2507.02747.pdf

Abstract:
As large models gain traction, vision-language-action (VLA) systems are enabling robots to tackle increasingly complex tasks. However, limited by the difficulty of data collection, progress has mainly focused on controlling simple gripper end-effectors. There is little research on functional grasping with large models for human-like dexterous hands. In this paper, we introduce DexVLG, a large Vision-Language-Grasp model for Dexterous grasp pose prediction aligned with language instructions using single-view RGBD input. To accomplish this, we generate a dataset of 170 million dexterous grasp poses mapped to semantic parts across 174,000 objects in simulation, paired with detailed part-level captions. This large-scale dataset, named DexGraspNet 3.0, is used to train a VLM and flow-matching-based pose head capable of producing instruction-aligned grasp poses for tabletop objects. To assess DexVLG's performance, we create benchmarks in physics-based simulations and conduct real-world experiments. Extensive testing demonstrates DexVLG's strong zero-shot generalization capabilities-achieving over 76% zero-shot execution success rate and state-of-the-art part-grasp accuracy in simulation-and successful part-aligned grasps on physical objects in real-world scenarios.

Paperid: 2865, https://arxiv.org/pdf/2506.19257.pdf

Abstract:
Vision-Language Models (VLMs) have achieved remarkable progress in multimodal reasoning tasks through enhanced chain-of-thought capabilities. However, this advancement also introduces novel safety risks, as these models become increasingly vulnerable to harmful multimodal prompts that can trigger unethical or unsafe behaviors. Existing safety alignment approaches, primarily designed for unimodal language models, fall short in addressing the complex and nuanced threats posed by multimodal inputs. Moreover, current safety datasets lack the fine-grained, policy-grounded reasoning required to robustly align reasoning-capable VLMs. In this work, we introduce {MSR-Align}, a high-quality Multimodal Safety Reasoning dataset tailored to bridge this gap. MSR-Align supports fine-grained, deliberative reasoning over standardized safety policies across both vision and text modalities. Our data generation pipeline emphasizes multimodal diversity, policy-grounded reasoning, and rigorous quality filtering using strong multimodal judges. Extensive experiments demonstrate that fine-tuning VLMs on MSR-Align substantially improves robustness against both textual and vision-language jailbreak attacks, while preserving or enhancing general reasoning performance. MSR-Align provides a scalable and effective foundation for advancing the safety alignment of reasoning-capable VLMs. Our dataset is made publicly available at https://huggingface.co/datasets/Leigest/MSR-Align.

Paperid: 2866, https://arxiv.org/pdf/2506.16211.pdf

Abstract:
Learning real-world robotic manipulation is challenging, particularly when limited demonstrations are available. Existing methods for few-shot manipulation often rely on simulation-augmented data or pre-built modules like grasping and pose estimation, which struggle with sim-to-real gaps and lack extensibility. While large-scale imitation pre-training shows promise, adapting these general-purpose policies to specific tasks in data-scarce settings remains unexplored. To achieve this, we propose ControlVLA, a novel framework that bridges pre-trained VLA models with object-centric representations via a ControlNet-style architecture for efficient fine-tuning. Specifically, to introduce object-centric conditions without overwriting prior knowledge, ControlVLA zero-initializes a set of projection layers, allowing them to gradually adapt the pre-trained manipulation policies. In real-world experiments across 6 diverse tasks, including pouring cubes and folding clothes, our method achieves a 76.7% success rate while requiring only 10-20 demonstrations -- a significant improvement over traditional approaches that require more than 100 demonstrations to achieve comparable success. Additional experiments highlight ControlVLA's extensibility to long-horizon tasks and robustness to unseen objects and backgrounds.

Paperid: 2867, https://arxiv.org/pdf/2506.07214.pdf

Abstract:
Vision Language Models (VLMs) have shown remarkable performance, but are also vulnerable to backdoor attacks whereby the adversary can manipulate the model's outputs through hidden triggers. Prior attacks primarily rely on single-modality triggers, leaving the crucial cross-modal fusion nature of VLMs largely unexplored. Unlike prior work, we identify a novel attack surface that leverages cross-modal semantic mismatches as implicit triggers. Based on this insight, we propose BadSem (Backdoor Attack with Semantic Manipulation), a data poisoning attack that injects stealthy backdoors by deliberately misaligning image-text pairs during training. To perform the attack, we construct SIMBad, a dataset tailored for semantic manipulation involving color and object attributes. Extensive experiments across four widely used VLMs show that BadSem achieves over 98% average ASR, generalizes well to out-of-distribution datasets, and can transfer across poisoning modalities. Our detailed analysis using attention visualization shows that backdoored models focus on semantically sensitive regions under mismatched conditions while maintaining normal behavior on clean inputs. To mitigate the attack, we try two defense strategies based on system prompt and supervised fine-tuning but find that both of them fail to mitigate the semantic backdoor. Our findings highlight the urgent need to address semantic vulnerabilities in VLMs for their safer deployment.

Paperid: 2868, https://arxiv.org/pdf/2505.20033.pdf

Abstract:
Effective human-AI interaction relies on AI's ability to accurately perceive and interpret human emotions. Current benchmarks for vision and vision-language models are severely limited, offering a narrow emotional spectrum that overlooks nuanced states (e.g., bitterness, intoxication) and fails to distinguish subtle differences between related feelings (e.g., shame vs. embarrassment). Existing datasets also often use uncontrolled imagery with occluded faces and lack demographic diversity, risking significant bias. To address these critical gaps, we introduce EmoNet Face, a comprehensive benchmark suite. EmoNet Face features: (1) A novel 40-category emotion taxonomy, meticulously derived from foundational research to capture finer details of human emotional experiences. (2) Three large-scale, AI-generated datasets (EmoNet HQ, Binary, and Big) with explicit, full-face expressions and controlled demographic balance across ethnicity, age, and gender. (3) Rigorous, multi-expert annotations for training and high-fidelity evaluation. (4) We built EmpathicInsight-Face, a model achieving human-expert-level performance on our benchmark. The publicly released EmoNet Face suite - taxonomy, datasets, and model - provides a robust foundation for developing and evaluating AI systems with a deeper understanding of human emotions.

Paperid: 2869, https://arxiv.org/pdf/2505.18605.pdf

Abstract:
Causal attention has become a foundational mechanism in autoregressive vision-language models (VLMs), unifying textual and visual inputs under a single generative framework. However, existing causal mask-based strategies are inherited from large language models (LLMs) where they are tailored for text-only decoding, and their adaptation to vision tokens is insufficiently addressed in the prefill stage. Strictly masking future positions for vision queries introduces overly rigid constraints, which hinder the model's ability to leverage future context that often contains essential semantic cues for accurate inference. In this work, we empirically investigate how different causal masking strategies affect vision-language inference and then propose a family of future-aware attentions tailored for this setting. We first empirically analyze the effect of previewing future tokens for vision queries and demonstrate that rigid masking undermines the model's capacity to capture useful contextual semantic representations. Based on these findings, we propose a lightweight attention family that aggregates future visual context into past representations via pooling, effectively preserving the autoregressive structure while enhancing cross-token dependencies. We evaluate a range of causal masks across diverse vision-language inference settings and show that selectively compressing future semantic context into past representations benefits the inference.

Paperid: 2870, https://arxiv.org/pdf/2505.17619.pdf

Abstract:
Synthetic X-ray angiographies generated by modern generative models hold great potential to reduce the use of contrast agents in vascular interventional procedures. However, low-quality synthetic angiographies can significantly increase procedural risk, underscoring the need for reliable image quality assessment (IQA) methods. Existing IQA models, however, fail to leverage auxiliary images as references during evaluation and lack fine-grained, task-specific metrics necessary for clinical relevance. To address these limitations, this paper proposes CAS-IQA, a vision-language model (VLM)-based framework that predicts fine-grained quality scores by effectively incorporating auxiliary information from related images. In the absence of angiography datasets, CAS-3K is constructed, comprising 3,565 synthetic angiographies along with score annotations. To ensure clinically meaningful assessment, three task-specific evaluation metrics are defined. Furthermore, a Multi-path featUre fuSion and rouTing (MUST) module is designed to enhance image representations by adaptively fusing and routing visual tokens to metric-specific branches. Extensive experiments on the CAS-3K dataset demonstrate that CAS-IQA significantly outperforms state-of-the-art IQA methods by a considerable margin.

Paperid: 2871, https://arxiv.org/pdf/2505.17436.pdf

Abstract:
To advance biomedical vison-language model capabilities through scaling up, fine-tuning, and instruction tuning, develop vision-language models with improved performance in handling long text, explore strategies to efficiently adopt vision language models for diverse multi-modal biomedical tasks, and examine the zero-shot learning performance. We developed two biomedical vision language models, BiomedGPT-Large and BiomedGPT-XLarge, based on an encoder-decoder-based transformer architecture. We fine-tuned the two models on 23 benchmark datasets from 6 multi-modal biomedical tasks including one image-only task (image classification), three language-only tasks (text understanding, text summarization and question answering), and two vision-language tasks (visual question answering and image captioning). We compared the developed scaled models with our previous BiomedGPT-Base model and existing prestigious models reported in the literature. We instruction-tuned the two models using a large-scale multi-modal biomedical instruction-tuning dataset and assessed the zero-shot learning performance and alignment accuracy.

Paperid: 2872, https://arxiv.org/pdf/2505.15644.pdf

Abstract:
Fine-grained edited image detection of localized edits in images is crucial for assessing content authenticity, especially given that modern diffusion models and image editing methods can produce highly realistic manipulations. However, this domain faces three challenges: (1) Binary classifiers yield only a global real-or-fake label without providing localization; (2) Traditional computer vision methods often rely on costly pixel-level annotations; and (3) No large-scale, high-quality dataset exists for modern image-editing detection techniques. To address these gaps, we develop an automated data-generation pipeline to create FragFake, the first dedicated benchmark dataset for edited image detection, which includes high-quality images from diverse editing models and a wide variety of edited objects. Based on FragFake, we utilize Vision Language Models (VLMs) for the first time in the task of edited image classification and edited region localization. Experimental results show that fine-tuned VLMs achieve higher average Object Precision across all datasets, significantly outperforming pretrained models. We further conduct ablation and transferability analyses to evaluate the detectors across various configurations and editing scenarios. To the best of our knowledge, this work is the first to reformulate localized image edit detection as a vision-language understanding task, establishing a new paradigm for the field. We anticipate that this work will establish a solid foundation to facilitate and inspire subsequent research endeavors in the domain of multimodal content authenticity.

Paperid: 2873, https://arxiv.org/pdf/2505.04488.pdf

Abstract:
The visually impaired population, especially the severely visually impaired, is currently large in scale, and daily activities pose significant challenges for them. Although many studies use large language and vision-language models to assist the blind, most focus on static content and fail to meet real-time perception needs in dynamic and complex environments, such as daily activities. To provide them with more effective intelligent assistance, it is imperative to incorporate advanced visual understanding technologies. Although real-time vision and speech interaction VideoLLMs demonstrate strong real-time visual understanding, no prior work has systematically evaluated their effectiveness in assisting visually impaired individuals. In this work, we conduct the first such evaluation. First, we construct a benchmark dataset (VisAssistDaily), covering three categories of assistive tasks for visually impaired individuals: Basic Skills, Home Life Tasks, and Social Life Tasks. The results show that GPT-4o achieves the highest task success rate. Next, we conduct a user study to evaluate the models in both closed-world and open-world scenarios, further exploring the practical challenges of applying VideoLLMs in assistive contexts. One key issue we identify is the difficulty current models face in perceiving potential hazards in dynamic environments. To address this, we build an environment-awareness dataset named SafeVid and introduce a polling mechanism that enables the model to proactively detect environmental risks. We hope this work provides valuable insights and inspiration for future research in this field.

Paperid: 2874, https://arxiv.org/pdf/2504.16060.pdf

Abstract:
Referring Expression Generation (REG) is a core task for evaluating the pragmatic competence of vision-language systems, requiring not only accurate semantic grounding but also adherence to principles of cooperative communication (Grice, 1975). However, current evaluations of vision-language models (VLMs) often overlook the pragmatic dimension, reducing REG to a region-based captioning task and neglecting Gricean maxims. In this work, we revisit REG from a pragmatic perspective, introducing a new dataset (RefOI) of 1.5k images annotated with both written and spoken referring expressions. Through a systematic evaluation of state-of-the-art VLMs, we identify three key failures of pragmatic competence: (1) failure to uniquely identify the referent, (2) inclusion of excessive or irrelevant information, and (3) misalignment with human pragmatic preference, such as the underuse of minimal spatial cues. We also show that standard automatic evaluations fail to capture these pragmatic violations, reinforcing superficial cues rather than genuine referential success. Our findings call for a renewed focus on pragmatically informed models and evaluation frameworks that align with real human communication.

Paperid: 2875, https://arxiv.org/pdf/2504.04191.pdf

Abstract:
Learning open-vocabulary physical skills for simulated agents presents a significant challenge in artificial intelligence. Current reinforcement learning approaches face critical limitations: manually designed rewards lack scalability across diverse tasks, while demonstration-based methods struggle to generalize beyond their training distribution. We introduce GROVE, a generalized reward framework that enables open-vocabulary physical skill learning without manual engineering or task-specific demonstrations. Our key insight is that Large Language Models(LLMs) and Vision Language Models(VLMs) provide complementary guidance -- LLMs generate precise physical constraints capturing task requirements, while VLMs evaluate motion semantics and naturalness. Through an iterative design process, VLM-based feedback continuously refines LLM-generated constraints, creating a self-improving reward system. To bridge the domain gap between simulation and natural images, we develop Pose2CLIP, a lightweight mapper that efficiently projects agent poses directly into semantic feature space without computationally expensive rendering. Extensive experiments across diverse embodiments and learning paradigms demonstrate GROVE's effectiveness, achieving 22.2% higher motion naturalness and 25.7% better task completion scores while training 8.4x faster than previous methods. These results establish a new foundation for scalable physical skill acquisition in simulated environments.

Paperid: 2876, https://arxiv.org/pdf/2503.22727.pdf

Abstract:
Despite the excitement behind biomedical artificial intelligence (AI), access to high-quality, diverse, and large-scale data - the foundation for modern AI systems - is still a bottleneck to unlocking its full potential. To address this gap, we introduce Biomedica, an open-source dataset derived from the PubMed Central Open Access subset, containing over 6 million scientific articles and 24 million image-text pairs, along with 27 metadata fields (including expert human annotations). To overcome the challenges of accessing our large-scale dataset, we provide scalable streaming and search APIs through a web server, facilitating seamless integration with AI systems. We demonstrate the utility of the Biomedica dataset by building embedding models, chat-style models, and retrieval-augmented chat agents. Notably, all our AI models surpass previous open systems in their respective categories, underscoring the critical role of diverse, high-quality, and large-scale biomedical data.

Paperid: 2877, https://arxiv.org/pdf/2503.15485.pdf

Abstract:
Despite the recent success of image-text contrastive models like CLIP and SigLIP, these models often struggle with vision-centric tasks that demand high-fidelity image understanding, such as counting, depth estimation, and fine-grained object recognition. These models, by performing language alignment, tend to prioritize high-level semantics over visual understanding, weakening their image understanding. On the other hand, vision-focused models are great at processing visual information but struggle to understand language, limiting their flexibility for language-driven tasks. In this work, we introduce TULIP, an open-source, drop-in replacement for existing CLIP-like models. Our method leverages generative data augmentation, enhanced image-image and text-text contrastive learning, and image/text reconstruction regularization to learn fine-grained visual features while preserving global semantic alignment. Our approach, scaling to over 1B parameters, outperforms existing state-of-the-art (SOTA) models across multiple benchmarks, establishing a new SOTA zero-shot performance on ImageNet-1K, delivering up to a $2\times$ enhancement over SigLIP on RxRx1 in linear probing for few-shot classification, and improving vision-language models, achieving over $3\times$ higher scores than SigLIP on MMVP. Our code/checkpoints are available at https://tulip-berkeley.github.io

Paperid: 2878, https://arxiv.org/pdf/2502.02175.pdf

Abstract:
Vision-Language-Action (VLA) model can process instructions and visual perception to directly generate actions as output in an end-to-end fashion due to its strong multi-modal reasoning capabilities. While the performance of VLA models is promising, their computational cost can be substantial. This raises challenge for applying them on robotics tasks, which requires real-time decision-making to respond quickly to environmental changes. Since robotic control involves sequential decision-making, the visual input often exhibits minimal variation between successive steps. A natural idea is to reuse the computational results of unchanged visual tokens from the last step. Motivated by this idea, we propose VLA-Cache, an efficient vision-language-action model. VLA-Cache incorporates a token-selection mechanism that compares the visual input at each step with the input from the previous step, adaptively identifying visual tokens with minimal changes. The computational results for these unchanged tokens are then reused in subsequent steps via KV-cache, thereby significantly improving the efficiency of the VLA-Cache model. Experimental results on both simulation (e.g., LIBERO benchmark and SIMPLER) and real-world robot valid VLA-Cache can achieve practical acceleration with minimal sacrifice in success rate.

Paperid: 2879, https://arxiv.org/pdf/2502.01969.pdf

Abstract:
Large Vision-Language Models (LVLMs) exhibit impressive multimodal reasoning capabilities but remain highly susceptible to object hallucination, where models generate responses that are not factually aligned with the visual content. Recent works attribute this issue to an inherent bias of LVLMs where vision token attention map has a fixed correlation with spatial position, and propose to mitigate this issue by reordering visual tokens. However, we find that different LVLMs exhibit different correlations between attention and spatial position, which makes the existing solution difficult to generalize to other LVLMs. To address this issue, we first introduce a training-free solution, Uniform Attention Calibration (UAC), that estimates the bias from single meaningless input image and applies a calibration matrix to rectify attention imbalances. To further alleviate the bias, we relax the assumption of single meaningless input in UAC and introduce a fine-tuning solution, Dynamic Attention Calibration (DAC), that enforces the consistent outputs wherever the object locates in the image via a plug-and-plays module. Comprehensive experiments across multiple benchmarks demonstrate that UAC and DAC significantly reduce object hallucination while improving general multimodal alignment. Our methods achieve state-of-the-art performance across diverse LVLM architectures on various metrics.

Paperid: 2880, https://arxiv.org/pdf/2501.07171.pdf

Abstract:
The development of vision-language models (VLMs) is driven by large-scale and diverse multimodal datasets. However, progress toward generalist biomedical VLMs is limited by the lack of annotated, publicly accessible datasets across biology and medicine. Existing efforts are restricted to narrow domains, missing the full diversity of biomedical knowledge encoded in scientific literature. To address this gap, we introduce BIOMEDICA, a scalable, open-source framework to extract, annotate, and serialize the entirety of the PubMed Central Open Access subset into an easy-to-use, publicly accessible dataset. Our framework produces a comprehensive archive with over 24 million unique image-text pairs from over 6 million articles. Metadata and expert-guided annotations are also provided. We demonstrate the utility and accessibility of our resource by releasing BMCA-CLIP, a suite of CLIP-style models continuously pre-trained on the BIOMEDICA dataset via streaming, eliminating the need to download 27 TB of data locally. On average, our models achieve state-of-the-art performance across 40 tasks - spanning pathology, radiology, ophthalmology, dermatology, surgery, molecular biology, parasitology, and cell biology - excelling in zero-shot classification with a 6.56% average improvement (as high as 29.8% and 17.5% in dermatology and ophthalmology, respectively), and stronger image-text retrieval, all while using 10x less compute. To foster reproducibility and collaboration, we release our codebase and dataset for the broader research community.

Paperid: 2881, https://arxiv.org/pdf/2501.03225.pdf

Abstract:
The rapid development of vision language models (VLMs) demands rigorous and reliable evaluation. However, current visual question answering (VQA) benchmarks often depend on open-ended questions, making accurate evaluation difficult due to the variability in natural language responses. To address this, we introduce AutoConverter, an agentic framework that automatically converts these open-ended questions into multiple-choice format, enabling objective evaluation while reducing the costly multiple-choice question creation process. Our experiments demonstrate that AutoConverter can generate correct and challenging multiple-choice questions, with VLMs demonstrating consistently similar or lower accuracy on these questions compared to human-created ones. Using AutoConverter, we construct VMCBench, a benchmark created by transforming 20 existing VQA datasets into a unified multiple-choice format, totaling 9,018 questions. We comprehensively evaluate 33 state-of-the-art VLMs on VMCBench, setting a new standard for scalable, consistent, and reproducible VLM evaluation.

Paperid: 2882, https://arxiv.org/pdf/2410.17385.pdf

Abstract:
Spatial expressions in situated communication can be ambiguous, as their meanings vary depending on the frames of reference (FoR) adopted by speakers and listeners. While spatial language understanding and reasoning by vision-language models (VLMs) have gained increasing attention, potential ambiguities in these models are still under-explored. To address this issue, we present the COnsistent Multilingual Frame Of Reference Test (COMFORT), an evaluation protocol to systematically assess the spatial reasoning capabilities of VLMs. We evaluate nine state-of-the-art VLMs using COMFORT. Despite showing some alignment with English conventions in resolving ambiguities, our experiments reveal significant shortcomings of VLMs: notably, the models (1) exhibit poor robustness and consistency, (2) lack the flexibility to accommodate multiple FoRs, and (3) fail to adhere to language-specific or culture-specific conventions in cross-lingual tests, as English tends to dominate other languages. With a growing effort to align vision-language models with human cognitive intuitions, we call for more attention to the ambiguous nature and cross-cultural diversity of spatial reasoning.

Paperid: 2883, https://arxiv.org/pdf/2410.14148.pdf

Abstract:
The recent advancements in large language models (LLMs) and pre-trained vision models have accelerated the development of vision-language large models (VLLMs), enhancing the interaction between visual and linguistic modalities. Despite their notable success across various domains, VLLMs face challenges in modality alignment, which can lead to issues like hallucinations and unsafe content generation. Current alignment techniques often rely on coarse feedback and external datasets, limiting scalability and performance. In this paper, we propose FiSAO (Fine-Grained Self-Alignment Optimization), a novel self-alignment method that utilizes the model's own visual encoder as a fine-grained verifier to improve vision-language alignment without the need for additional data. By leveraging token-level feedback from the vision encoder, FiSAO significantly improves vision-language alignment, even surpassing traditional preference tuning methods that require additional data. Through both theoretical analysis and experimental validation, we demonstrate that FiSAO effectively addresses the misalignment problem in VLLMs, marking the first instance of token-level rewards being applied to such models.

Paperid: 2884, https://arxiv.org/pdf/2409.17727.pdf

Abstract:
Vision language models have played a key role in extracting meaningful features for various robotic applications. Among these, Contrastive Language-Image Pretraining (CLIP) is widely used in robotic tasks that require both vision and natural language understanding. However, CLIP was trained solely on static images paired with text prompts and has not yet been fully adapted for robotic tasks involving dynamic actions. In this paper, we introduce Robotic-CLIP to enhance robotic perception capabilities. We first gather and label large-scale action data, and then build our Robotic-CLIP by fine-tuning CLIP on 309,433 videos (~7.4 million frames) of action data using contrastive learning. By leveraging action data, Robotic-CLIP inherits CLIP's strong image performance while gaining the ability to understand actions in robotic contexts. Intensive experiments show that our Robotic-CLIP outperforms other CLIP-based models across various language-driven robotic tasks. Additionally, we demonstrate the practical effectiveness of Robotic-CLIP in real-world grasping applications.

Paperid: 2885, https://arxiv.org/pdf/2601.13143.pdf

Abstract:
In this work, we present FastAV, the first token pruning framework tailored for audio-visual large language models (AV-LLMs). While token pruning has been actively explored in standard large language models (LLMs) and vision-language models (LVLMs), its application to AV-LLMs has received little attention, even though multimodal integration substantially increases their token demands. To address this gap, we introduce a pruning strategy that utilizes attention weights to identify tokens emphasized at different stages and estimates their importance. Building on this analysis, FastAV applies a two-stage pruning strategy: (1) global pruning in intermediate layers to remove broadly less influential tokens, and (2) fine pruning in later layers considering the impact on next token generation. Notably, our method does not rely on full attention maps, which makes it fully compatible with efficient attention mechanisms such as FlashAttention. Extensive experiments demonstrate that FastAV reduces FLOPs by more than 40% on two representative AV-LLMs, while preserving or even improving model performance.

Paperid: 2886, https://arxiv.org/pdf/2601.10551.pdf

Abstract:
Automated perception of urban roadside infrastructure is crucial for smart city management, yet general-purpose models often struggle to capture the necessary fine-grained attributes and domain rules. While Large Vision Language Models (VLMs) excel at open-world recognition, they often struggle to accurately interpret complex facility states in compliance with engineering standards, leading to unreliable performance in real-world applications. To address this, we propose a domain-adapted framework that transforms VLMs into specialized agents for intelligent infrastructure analysis. Our approach integrates a data-efficient fine-tuning strategy with a knowledge-grounded reasoning mechanism. Specifically, we leverage open-vocabulary fine-tuning on Grounding DINO to robustly localize diverse assets with minimal supervision, followed by LoRA-based adaptation on Qwen-VL for deep semantic attribute reasoning. To mitigate hallucinations and enforce professional compliance, we introduce a dual-modality Retrieval-Augmented Generation (RAG) module that dynamically retrieves authoritative industry standards and visual exemplars during inference. Evaluated on a comprehensive new dataset of urban roadside scenes, our framework achieves a detection performance of 58.9 mAP and an attribute recognition accuracy of 95.5%, demonstrating a robust solution for intelligent infrastructure monitoring.

Paperid: 2887, https://arxiv.org/pdf/2601.10535.pdf

Abstract:
The automated creation of digital twins and precise asset inventories is a critical task in smart city construction and facility lifecycle management. However, utilizing cost-effective sparse imagery remains challenging due to limited robustness, inaccurate localization, and a lack of fine-grained state understanding. To address these limitations, SVII-3D, a unified framework for holistic asset digitization, is proposed. First, LoRA fine-tuned open-set detection is fused with a spatial-attention matching network to robustly associate observations across sparse views. Second, a geometry-guided refinement mechanism is introduced to resolve structural errors, achieving precise decimeter-level 3D localization. Third, transcending static geometric mapping, a Vision-Language Model agent leveraging multi-modal prompting is incorporated to automatically diagnose fine-grained operational states. Experiments demonstrate that SVII-3D significantly improves identification accuracy and minimizes localization errors. Consequently, this framework offers a scalable, cost-effective solution for high-fidelity infrastructure digitization, effectively bridging the gap between sparse perception and automated intelligent maintenance.

Paperid: 2888, https://arxiv.org/pdf/2601.10305.pdf

Abstract:
Vision-Language Pre-training (VLP) models demonstrate strong performance across various downstream tasks by learning from large-scale image-text pairs through contrastive pretraining. The release of extensive English image-text datasets (e.g., COYO-700M and LAION-400M) has enabled widespread adoption of models such as CLIP and SigLIP in tasks including cross-modal retrieval and image captioning. However, the advancement of Chinese vision-language pretraining has substantially lagged behind, due to the scarcity of high-quality Chinese image-text data. To address this gap, we develop a comprehensive pipeline for constructing a high-quality Chinese cross-modal dataset. As a result, we propose DanQing, which contains 100 million image-text pairs collected from Common Crawl. Different from existing datasets, DanQing is curated through a more rigorous selection process, yielding superior data quality. Moreover, DanQing is primarily built from 2024-2025 web data, enabling models to better capture evolving semantic trends and thus offering greater practical utility. We compare DanQing with existing datasets by continual pre-training of the SigLIP2 model. Experimental results show that DanQing consistently achieves superior performance across a range of Chinese downstream tasks, including zero-shot classification, cross-modal retrieval, and LMM-based evaluations. To facilitate further research in Chinese vision-language pre-training, we will open-source the DanQing dataset under the Creative Common CC-BY 4.0 license.

Paperid: 2889, https://arxiv.org/pdf/2601.02888.pdf

Abstract:
Visually impaired users face significant challenges in daily information access and real-time environmental perception, and there is an urgent need for intelligent assistive systems with accurate recognition capabilities. Although large-scale models provide effective solutions for perception and reasoning, their practical deployment on assistive devices is severely constrained by excessive memory consumption and high inference costs. Moreover, existing quantization strategies often ignore inter-block error accumulation, leading to degraded model stability. To address these challenges, this study proposes a novel quantization framework -- Residual-Projected Multi-Collaboration Closed-Loop and Single Instance Quantization(RPIQ), whose quantization process adopts a multi-collaborative closed-loop compensation scheme based on Single Instance Calibration and Gauss-Seidel Iterative Quantization. Experiments on various types of large-scale models, including language models such as OPT, Qwen, and LLaMA, as well as vision-language models such as CogVLM2, demonstrate that RPIQ can compress models to 4-bit representation while significantly reducing peak memory consumption (approximately 60%-75% reduction compared to original full-precision models). The method maintains performance highly close to full-precision models across multiple language and visual tasks, and exhibits excellent recognition and reasoning capabilities in key applications such as text understanding and visual question answering in complex scenarios. While verifying the effectiveness of RPIQ for deployment in real assistive systems, this study also advances the computational efficiency and reliability of large models, enabling them to provide visually impaired users with the required information accurately and rapidly.

Paperid: 2890, https://arxiv.org/pdf/2512.21243.pdf

Abstract:
Methods that use Large Language Models (LLM) as planners for embodied instruction following tasks have become widespread. To successfully complete tasks, the LLM must be grounded in the environment in which the robot operates. One solution is to use a scene graph that contains all the necessary information. Modern methods rely on prebuilt scene graphs and assume that all task-relevant information is available at the start of planning. However, these approaches do not account for changes in the environment that may occur between the graph construction and the task execution. We propose LookPlanGraph - a method that leverages a scene graph composed of static assets and object priors. During plan execution, LookPlanGraph continuously updates the graph with relevant objects, either by verifying existing priors or discovering new entities. This is achieved by processing the agents egocentric camera view using a Vision Language Model. We conducted experiments with changed object positions VirtualHome and OmniGibson simulated environments, demonstrating that LookPlanGraph outperforms methods based on predefined static scene graphs. To demonstrate the practical applicability of our approach, we also conducted experiments in a real-world setting. Additionally, we introduce the GraSIF (Graph Scenes for Instruction Following) dataset with automated validation framework, comprising 514 tasks drawn from SayPlan Office, BEHAVIOR-1K, and VirtualHome RobotHow. Project page available at https://lookplangraph.github.io .

Paperid: 2891, https://arxiv.org/pdf/2512.18933.pdf

Abstract:
Vision-Language-Action (VLA) models align vision and language with embodied control, but their object referring ability remains limited when relying solely on text prompt, especially in cluttered or out-of-distribution (OOD) scenes. In this study, we introduce the Point-VLA, a plug-and-play policy that augments language instructions with explicit visual cues (e.g., bounding boxes) to resolve referential ambiguity and enable precise object-level grounding. To efficiently scale visually grounded datasets, we further develop an automatic data annotation pipeline requiring minimal human effort. We evaluate Point-VLA on diverse real-world referring tasks and observe consistently stronger performance than text-only instruction VLAs, particularly in cluttered or unseen-object scenarios, with robust generalization. These results demonstrate that Point-VLA effectively resolves object referring ambiguity through pixel-level visual grounding, achieving more generalizable embodied control.

Paperid: 2892, https://arxiv.org/pdf/2512.13609.pdf

Abstract:
We introduce the Do-Undo task and benchmark to address a critical gap in vision-language models: understanding and generating physically plausible scene transformations driven by real-world actions. Unlike prior work focused on object-level edits, Do-Undo requires models to simulate the outcome of a physical action and then accurately reverse it, reflecting true cause-and-effect in the visual world. We curate a large-scale dataset of reversible actions from real-world videos and design a training strategy enforcing consistency for robust action grounding. Our experiments reveal that current models struggle with physical reversibility, underscoring the importance of this task for embodied AI, robotics, and physics-aware generative modeling. Do-Undo establishes an intuitive testbed for evaluating and advancing physical reasoning in multimodal systems.

Paperid: 2893, https://arxiv.org/pdf/2512.03724.pdf

Abstract:
The Vision-Language-Action (VLA) models have demonstrated remarkable performance on embodied tasks and shown promising potential for real-world applications. However, current VLAs still struggle to produce consistent and precise target-oriented actions, as they often generate redundant or unstable motions along trajectories, limiting their applicability in time-sensitive scenarios.In this work, we attribute these redundant actions to the spatially uniform perception field of existing VLAs, which causes them to be distracted by target-irrelevant objects, especially in complex environments.To address this issue, we propose an efficient PosA-VLA framework that anchors visual attention via pose-conditioned supervision, consistently guiding the model's perception toward task-relevant regions. The pose-conditioned anchor attention mechanism enables the model to better align instruction semantics with actionable visual cues, thereby improving action generation precision and efficiency. Moreover, our framework adopts a lightweight architecture and requires no auxiliary perception modules (e.g., segmentation or grounding networks), ensuring efficient inference. Extensive experiments verify that our method executes embodied tasks with precise and time-efficient behavior across diverse robotic manipulation benchmarks and shows robust generalization in a variety of challenging environments.

Paperid: 2894, https://arxiv.org/pdf/2511.17793.pdf

Abstract:
Large Vision-Language Models (VLMs) rely on effective multimodal alignment between pre-trained vision encoders and Large Language Models (LLMs) to integrate visual and textual information. This paper presents a comprehensive analysis of attention patterns in efficient VLMs, revealing that concatenation-based architectures frequently fail to distinguish between semantically matching and non-matching image-text pairs. This is a key factor for object hallucination in these models. To address this, we introduce Attention-Guided Efficient Vision-Language Models (AGE-VLM), a novel framework that enhances visual grounding through interleaved cross-attention layers to instill vision capabilities in pretrained small language models. This enforces in VLM the ability "look" at the correct image regions by leveraging spatial knowledge distilled from the Segment Anything Model (SAM), significantly reducing hallucination. We validate our approach across different vision-centric benchmarks where our method is better or comparable to prior work on efficient VLMs. Our findings provide valuable insights for future research aimed at achieving enhanced visual and linguistic understanding in VLMs.

Paperid: 2895, https://arxiv.org/pdf/2510.25616.pdf

Abstract:
The growing success of Vision-Language-Action (VLA) models stems from the promise that pretrained Vision-Language Models (VLMs) can endow agents with transferable world knowledge and vision-language (VL) grounding, laying a foundation for action models with broader generalization. Yet when these VLMs are adapted to the action modality, it remains unclear to what extent their original VL representations and knowledge are preserved. In this work, we conduct a systematic study of representation retention during VLA fine-tuning, showing that naive action fine-tuning leads to degradation of visual representations. To characterize and measure these effects, we probe VLA's hidden representations and analyze attention maps, further, we design a set of targeted tasks and methods that contrast VLA models with their counterpart VLMs, isolating changes in VL capabilities induced by action fine-tuning. We further evaluate a range of strategies for aligning visual representations and introduce a simple yet effective method that mitigates degradation and yields improved generalization to out-of-distribution (OOD) scenarios. Taken together, our analysis clarifies the trade-off between action fine-tuning and the degradation of VL representations and highlights practical approaches to recover inherited VL capabilities. Code is publicly available: https://blind-vla-paper.github.io

Paperid: 2896, https://arxiv.org/pdf/2510.21311.pdf

Abstract:
Multi-modal Large Language Models (MLLMs) have shown remarkable capabilities across a wide range of vision-language tasks. However, due to the restricted input resolutions, MLLMs face significant challenges in precisely understanding and localizing visual details in high-resolution images -- particularly when dealing with extra-small objects embedded in cluttered contexts. To address this issue, we propose \textsc{FineRS}, a two-stage MLLM-based reinforcement learning framework for jointly reasoning and segmenting extremely small objects within high-resolution scenes. \textsc{FineRS} adopts a coarse-to-fine pipeline comprising Global Semantic Exploration (GSE) and Localized Perceptual Refinement (LPR). Specifically, GSE performs instruction-guided reasoning to generate a textural response and a coarse target region, while LPR refines this region to produce an accurate bounding box and segmentation mask. To couple the two stages, we introduce a locate-informed retrospective reward, where LPR's outputs are used to optimize GSE for more robust coarse region exploration. % Additionally, we present \textsc{FineRS}-4k, a new dataset for evaluating MLLMs on attribute-level reasoning and pixel-level segmentation on subtle, small-scale targets in complex high-resolution scenes. Experimental results on \textsc{FineRS}-4k and public datasets demonstrate that our method consistently outperforms state-of-the-art MLLM-based approaches on both instruction-guided segmentation and visual reasoning tasks.

Paperid: 2897, https://arxiv.org/pdf/2510.20193.pdf

Abstract:
Question Answering (QA) systems have traditionally relied on structured text data, but the rapid growth of multimedia content (images, audio, video, and structured metadata) has introduced new challenges and opportunities for retrieval-augmented QA. In this survey, we review recent advancements in QA systems that integrate multimedia retrieval pipelines, focusing on architectures that align vision, language, and audio modalities with user queries. We categorize approaches based on retrieval methods, fusion techniques, and answer generation strategies, and analyze benchmark datasets, evaluation protocols, and performance tradeoffs. Furthermore, we highlight key challenges such as cross-modal alignment, latency-accuracy tradeoffs, and semantic grounding, and outline open problems and future research directions for building more robust and context-aware QA systems leveraging multimedia data.

Paperid: 2898, https://arxiv.org/pdf/2510.10052.pdf

Abstract:
Medical vision-language models (VLMs) excel at image-text understanding but typically rely on a single-pass reasoning that neglects localized visual cues. In clinical practice, however, human experts iteratively scan, focus, and refine the regions of interest before reaching a final diagnosis. To narrow this machine-human perception gap, we introduce ViTAR, a novel VLM framework that emulates the iterative reasoning process of human experts through a cognitive chain of "think-act-rethink-answer". ViTAR treats medical images as interactive objects, enabling models to engage multi-step visual reasoning. To support this approach, we curate a high-quality instruction dataset comprising 1K interactive examples that encode expert-like diagnostic behaviors. In addition, a 16K visual question answering training data has been curated towards fine-grained visual diagnosis. We introduce a two-stage training strategy that begins with supervised fine-tuning to guide cognitive trajectories, followed by the reinforcement learning to optimize decision-making. Extensive evaluations demonstrate that ViTAR outperforms strong state-of-the-art models. Visual attention analysis reveals that from the "think" to "rethink" rounds, ViTAR increasingly anchors visual grounding to clinically critical regions and maintains high attention allocation to visual tokens during reasoning, providing mechanistic insight into its improved performance. These findings demonstrate that embedding expert-style iterative thinking chains into VLMs enhances both performance and trustworthiness of medical AI.

Paperid: 2899, https://arxiv.org/pdf/2510.05722.pdf

Abstract:
Generating enough and diverse data through augmentation offers an efficient solution to the time-consuming and labour-intensive process of collecting and annotating pixel-wise images. Traditional data augmentation techniques often face challenges in manipulating high-level semantic attributes, such as materials and textures. In contrast, diffusion models offer a robust alternative, by effectively utilizing text-to-image or image-to-image transformation. However, existing diffusion-based methods are either computationally expensive or compromise on performance. To address this issue, we introduce a novel training-free pipeline that integrates pretrained ControlNet and Vision-Language Models (VLMs) to generate synthetic images paired with pixel-level labels. This approach eliminates the need for manual annotations and significantly improves downstream tasks. To improve the fidelity and diversity, we add a Multi-way Prompt Generator, Mask Generator and High-quality Image Selection module. Our results on PASCAL-5i and COCO-20i present promising performance and outperform concurrent work for one-shot semantic segmentation.

Paperid: 2900, https://arxiv.org/pdf/2509.17328.pdf

Abstract:
Building autonomous agents that perceive and operate graphical user interfaces (GUIs) like humans has long been a vision in the field of artificial intelligence. Central to these agents is the capability for GUI interaction, which involves GUI understanding and planning capabilities. Existing methods have tried developing GUI agents based on the multi-modal comprehension ability of vision-language models (VLMs). However, the limited scenario, insufficient size, and heterogeneous action spaces hinder the progress of building generalist GUI agents. To resolve these issues, this paper proposes \textbf{UIPro}, a novel generalist GUI agent trained with extensive multi-platform and multi-task GUI interaction data, coupled with a unified action space. We first curate a comprehensive dataset encompassing 20.6 million GUI understanding tasks to pre-train UIPro, granting it a strong GUI grounding capability, which is key to downstream GUI agent tasks. Subsequently, we establish a unified action space to harmonize heterogeneous GUI agent task datasets and produce a merged dataset to foster the action prediction ability of UIPro via continued fine-tuning. Experimental results demonstrate UIPro's superior performance across multiple GUI task benchmarks on various platforms, highlighting the effectiveness of our approach.

Paperid: 2901, https://arxiv.org/pdf/2509.12876.pdf

Abstract:
The proliferation of multimedia content necessitates the development of effective Multimedia Event Extraction (M2E2) systems. Though Large Vision-Language Models (LVLMs) have shown strong cross-modal capabilities, their utility in the M2E2 task remains underexplored. In this paper, we present the first systematic evaluation of representative LVLMs, including DeepSeek-VL2 and the Qwen-VL series, on the M2E2 dataset. Our evaluations cover text-only, image-only, and cross-media subtasks, assessed under both few-shot prompting and fine-tuning settings. Our key findings highlight the following valuable insights: (1) Few-shot LVLMs perform notably better on visual tasks but struggle significantly with textual tasks; (2) Fine-tuning LVLMs with LoRA substantially enhances model performance; and (3) LVLMs exhibit strong synergy when combining modalities, achieving superior performance in cross-modal settings. We further provide a detailed error analysis to reveal persistent challenges in areas such as semantic precision, localization, and cross-modal grounding, which remain critical obstacles for advancing M2E2 capabilities.

Paperid: 2902, https://arxiv.org/pdf/2509.09118.pdf

Abstract:
Although Contrastive Language-Image Pre-training (CLIP) exhibits strong performance across diverse vision tasks, its application to person representation learning faces two critical challenges: (i) the scarcity of large-scale annotated vision-language data focused on person-centric images, and (ii) the inherent limitations of global contrastive learning, which struggles to maintain discriminative local features crucial for fine-grained matching while remaining vulnerable to noisy text tokens. This work advances CLIP for person representation learning through synergistic improvements in data curation and model architecture. First, we develop a noise-resistant data construction pipeline that leverages the in-context learning capabilities of MLLMs to automatically filter and caption web-sourced images. This yields WebPerson, a large-scale dataset of 5M high-quality person-centric image-text pairs. Second, we introduce the GA-DMS (Gradient-Attention Guided Dual-Masking Synergetic) framework, which improves cross-modal alignment by adaptively masking noisy textual tokens based on the gradient-attention similarity score. Additionally, we incorporate masked token prediction objectives that compel the model to predict informative text tokens, enhancing fine-grained semantic representation learning. Extensive experiments show that GA-DMS achieves state-of-the-art performance across multiple benchmarks.

Paperid: 2903, https://arxiv.org/pdf/2509.09090.pdf

Abstract:
Vision-Language-Action (VLA) models exhibit unprecedented capabilities for embodied intelligence. However, their extensive computational and memory costs hinder their practical deployment. Existing VLA compression and acceleration approaches conduct quantization or token pruning in an ad-hoc manner but fail to enable both for a holistic efficiency improvement due to an observed incompatibility. This work introduces SQAP-VLA, the first structured, training-free VLA inference acceleration framework that simultaneously enables state-of-the-art quantization and token pruning. We overcome the incompatibility by co-designing the quantization and token pruning pipeline, where we propose new quantization-aware token pruning criteria that work on an aggressively quantized model while improving the quantizer design to enhance pruning effectiveness. When applied to standard VLA models, SQAP-VLA yields significant gains in computational efficiency and inference speed while successfully preserving core model performance, achieving a $\times$1.93 speedup and up to a 4.5\% average success rate enhancement compared to the original model.

Paperid: 2904, https://arxiv.org/pdf/2509.03893.pdf

Abstract:
Establishing dense correspondences across image pairs is essential for tasks such as shape reconstruction and robot manipulation. In the challenging setting of matching across different categories, the function of an object, i.e., the effect that an object can cause on other objects, can guide how correspondences should be established. This is because object parts that enable specific functions often share similarities in shape and appearance. We derive the definition of dense functional correspondence based on this observation and propose a weakly-supervised learning paradigm to tackle the prediction task. The main insight behind our approach is that we can leverage vision-language models to pseudo-label multi-view images to obtain functional parts. We then integrate this with dense contrastive learning from pixel correspondences to distill both functional and spatial knowledge into a new model that can establish dense functional correspondence. Further, we curate synthetic and real evaluation datasets as task benchmarks. Our results demonstrate the advantages of our approach over baseline solutions consisting of off-the-shelf self-supervised image representations and grounded vision language models.

Paperid: 2905, https://arxiv.org/pdf/2508.18177.pdf

Abstract:
This study proposes the dual technological innovation framework, including a cross-modal differ entiated quantization framework for vision-language models (VLMs) and a scene-aware vectorized memory multi-agent system for visually impaired assistance. The modular framework was developed implementing differentiated processing strategies, effectively reducing memory requirements from 38GB to 16GB while maintaining model performance. The multi-agent architecture combines scene classification, vectorized memory, and multimodal interaction, enabling persistent storage and efficient retrieval of scene memories. Through perception-memory-reasoning workflows, the system provides environmental information beyond the current view using historical memories. Experiments show the quantized 19B-parameter model only experiences a 2.05% performance drop on MMBench and maintains 63.7 accuracy on OCR-VQA (original: 64.9), outperforming smaller models with equivalent memory requirements like the Molmo-7B series. The system maintains response latency between 2.83-3.52 seconds from scene analysis to initial speech output, substantially faster than non-streaming methods. This research advances computational efficiency and assistive technology, offering visually impaired users comprehensive real-time assistance in scene perception, text recognition, and navigation.

Paperid: 2906, https://arxiv.org/pdf/2508.15663.pdf

Abstract:
Benchmarks are crucial for evaluating progress in robotics and embodied AI. However, a significant gap exists between benchmarks designed for high-level language instruction following, which often assume perfect low-level execution, and those for low-level robot control, which rely on simple, one-step commands. This disconnect prevents a comprehensive evaluation of integrated systems where both task planning and physical execution are critical. To address this, we propose Kitchen-R, a novel benchmark that unifies the evaluation of task planning and low-level control within a simulated kitchen environment. Built as a digital twin using the Isaac Sim simulator and featuring more than 500 complex language instructions, Kitchen-R supports a mobile manipulator robot. We provide baseline methods for our benchmark, including a task-planning strategy based on a vision-language model and a low-level control policy based on diffusion policy. We also provide a trajectory collection system. Our benchmark offers a flexible framework for three evaluation modes: independent assessment of the planning module, independent assessment of the control policy, and, crucially, an integrated evaluation of the whole system. Kitchen-R bridges a key gap in embodied AI research, enabling more holistic and realistic benchmarking of language-guided robotic agents.

Paperid: 2907, https://arxiv.org/pdf/2506.19850.pdf

Abstract:
Vision-language-action models (VLAs) have garnered significant attention for their potential in advancing robotic manipulation. However, previous approaches predominantly rely on the general comprehension capabilities of vision-language models (VLMs) to generate action signals, often overlooking the rich temporal and causal structure embedded in visual observations. In this paper, we present UniVLA, a unified and native multimodal VLA model that autoregressively models vision, language, and action signals as discrete token sequences. This formulation enables flexible multimodal tasks learning, particularly from large-scale video data. By incorporating world modeling during post-training, UniVLA captures causal dynamics from videos, facilitating effective transfer to downstream policy learning--especially for long-horizon tasks. Our approach sets new state-of-the-art results across several widely used simulation benchmarks, including CALVIN, LIBERO, and Simplenv-Bridge, significantly surpassing previous methods. For example, UniVLA achieves 95.5% average success rate on LIBERO benchmark, surpassing pi0-FAST's 85.5%. We further demonstrate its broad applicability on real-world ALOHA manipulation and autonomous driving.

Paperid: 2908, https://arxiv.org/pdf/2506.18564.pdf

Abstract:
Recent advances in AI-generated content (AIGC) have led to the emergence of powerful text-to-video generation models. Despite these successes, evaluating the quality of AIGC-generated videos remains challenging due to limited generalization, lack of temporal awareness, heavy reliance on large-scale annotated datasets, and the lack of effective interaction with generation models. Most current approaches rely on supervised finetuning of vision-language models (VLMs), which often require large-scale annotated datasets and tend to decouple understanding and generation. To address these shortcomings, we propose VQ-Insight, a novel reasoning-style VLM framework for AIGC video quality assessment. Our approach features: (1) a progressive video quality learning scheme that combines image quality warm-up, general task-specific temporal learning, and joint optimization with the video generation model; (2) the design of multi-dimension scoring rewards, preference comparison rewards, and temporal modeling rewards to enhance both generalization and specialization in video quality evaluation. Extensive experiments demonstrate that VQ-Insight consistently outperforms state-of-the-art baselines in preference comparison, multi-dimension scoring, and natural video scoring, bringing significant improvements for video generation tasks.

Paperid: 2909, https://arxiv.org/pdf/2506.03642.pdf

Abstract:
Visual-spatial understanding, the ability to infer object relationships and layouts from visual input, is fundamental to downstream tasks such as robotic navigation and embodied interaction. However, existing methods face spatial uncertainty and data scarcity, limiting the 3D spatial reasoning capability of pre-trained vision-language models (VLMs). To address these challenges, we present a unified framework for enhancing 3D spatial reasoning in pre-trained VLMs without modifying their architecture. This framework combines SpatialMind, a structured prompting strategy that decomposes complex scenes and questions into interpretable reasoning steps, with ScanForgeQA, a scalable question-answering dataset built from diverse 3D simulation scenes through an automated construction process designed for fine-tuning. Extensive experiments across multiple benchmarks demonstrate the individual and combined effectiveness of our prompting and fine-tuning strategies, and yield insights that may inspire future research on visual-spatial understanding.

Paperid: 2910, https://arxiv.org/pdf/2505.23759.pdf

Abstract:
Rebus puzzles, visual riddles that encode language through imagery, spatial arrangement, and symbolic substitution, pose a unique challenge to current vision-language models (VLMs). Unlike traditional image captioning or question answering tasks, rebus solving requires multi-modal abstraction, symbolic reasoning, and a grasp of cultural, phonetic and linguistic puns. In this paper, we investigate the capacity of contemporary VLMs to interpret and solve rebus puzzles by constructing a hand-generated and annotated benchmark of diverse English-language rebus puzzles, ranging from simple pictographic substitutions to spatially-dependent cues ("head" over "heels"). We analyze how different VLMs perform, and our findings reveal that while VLMs exhibit some surprising capabilities in decoding simple visual clues, they struggle significantly with tasks requiring abstract reasoning, lateral thinking, and understanding visual metaphors.

Paperid: 2911, https://arxiv.org/pdf/2505.23267.pdf

Abstract:
Path planning is a fundamental capability of autonomous Unmanned Aerial Vehicles (UAVs), enabling them to efficiently navigate toward a target region or explore complex environments while avoiding obstacles. Traditional pathplanning methods, such as Rapidly-exploring Random Trees (RRT), have proven effective but often encounter significant challenges. These include high search space complexity, suboptimal path quality, and slow convergence, issues that are particularly problematic in high-stakes applications like disaster response, where rapid and efficient planning is critical. To address these limitations and enhance path-planning efficiency, we propose Vision Language Model RRT (VLM-RRT), a hybrid approach that integrates the pattern recognition capabilities of Vision Language Models (VLMs) with the path-planning strengths of RRT. By leveraging VLMs to provide initial directional guidance based on environmental snapshots, our method biases sampling toward regions more likely to contain feasible paths, significantly improving sampling efficiency and path quality. Extensive quantitative and qualitative experiments with various state-of-the-art VLMs demonstrate the effectiveness of this proposed approach.

Paperid: 2912, https://arxiv.org/pdf/2505.20862.pdf

Abstract:
Hallucination remains a major challenge in multimodal large language models (MLLMs). To address this, various contrastive decoding (CD) methods have been proposed that contrasts original logits with hallucinated logits generated from perturbed inputs. While CD has shown promise in vision-language models (VLMs), it is not well-suited for AV-LLMs, where hallucinations often emerge from both unimodal and cross-modal combinations involving audio, video, and language. These intricate interactions call for a more adaptive and modality-aware decoding strategy. In this paper, we propose Audio-Visual Contrastive Decoding (AVCD)-a novel, training-free decoding framework designed to model trimodal interactions and suppress modality-induced hallucinations in AV-LLMs. Unlike previous CD methods in VLMs that corrupt a fixed modality, AVCD leverages attention distributions to dynamically identify less dominant modalities and applies attentive masking to generate perturbed output logits. To support CD in a trimodal setting, we also reformulate the original CD framework to jointly handle audio, visual, and textual inputs. Finally, to improve efficiency, we introduce entropy-guided adaptive decoding, which selectively skips unnecessary decoding steps based on the model's confidence in its predictions. Extensive experiments demonstrate that AVCD consistently outperforms existing decoding methods. Especially, on the AVHBench dataset, it improves accuracy by 6% for VideoLLaMA2 and 11% for video-SALMONN, demonstrating strong robustness and generalizability.

Paperid: 2913, https://arxiv.org/pdf/2505.20291.pdf

Abstract:
Text-to-image retrieval (T2I retrieval) remains challenging because cross-modal embeddings often behave as bags of concepts and underrepresent structured visual relationships such as pose and viewpoint. We propose Visualize-then-Retrieve (VisRet), a new paradigm for T2I retrieval that mitigates this limitation of cross-modal similarity alignment. VisRet first projects textual queries into the image modality via T2I generation. Then, it performs retrieval within the image modality to bypass the weaknesses of cross-modal retrievers in recognizing subtle visual-spatial features. Across four benchmarks (Visual-RAG, INQUIRE-Rerank, Microsoft COCO, and our new Visual-RAG-ME featuring multi-entity comparisons), VisRet substantially outperforms cross-modal similarity matching and baselines that recast T2I retrieval as text-to-text similarity matching, improving nDCG@30 by 0.125 on average with CLIP as the retriever and by 0.121 with E5-V. For downstream question answering, VisRet increases accuracy on Visual-RAG and Visual-RAG-ME by 3.8% and 15.7% in top-1 retrieval, and by 3.9% and 11.1% in top-10 retrieval. Ablation studies show compatibility with different T2I instruction LLMs, T2I generation models, and downstream LLMs. VisRet provides a practical and principled path that energizes further advances in vision-language retrieval. Our code and the Visual-RAG-ME benchmark will be publicly released.

Paperid: 2914, https://arxiv.org/pdf/2505.19139.pdf

Abstract:
Our research reveals a new privacy risk associated with the vision-language model (VLM) agentic framework: the ability to infer sensitive attributes (e.g., age and health information) and even abstract ones (e.g., personality and social traits) from a set of personal images, which we term "image private attribute profiling." This threat is particularly severe given that modern apps can easily access users' photo albums, and inference from image sets enables models to exploit inter-image relations for more sophisticated profiling. However, two main challenges hinder our understanding of how well VLMs can profile an individual from a few personal photos: (1) the lack of benchmark datasets with multi-image annotations for private attributes, and (2) the limited ability of current multimodal large language models (MLLMs) to infer abstract attributes from large image collections. In this work, we construct PAPI, the largest dataset for studying private attribute profiling in personal images, comprising 2,510 images from 251 individuals with 3,012 annotated privacy attributes. We also propose HolmesEye, a hybrid agentic framework that combines VLMs and LLMs to enhance privacy inference. HolmesEye uses VLMs to extract both intra-image and inter-image information and LLMs to guide the inference process as well as consolidate the results through forensic analysis, overcoming existing limitations in long-context visual reasoning. Experiments reveal that HolmesEye achieves a 10.8% improvement in average accuracy over state-of-the-art baselines and surpasses human-level performance by 15.0% in predicting abstract attributes. This work highlights the urgency of addressing privacy risks in image-based profiling and offers both a new dataset and an advanced framework to guide future research in this area.

Paperid: 2915, https://arxiv.org/pdf/2505.11917.pdf

Abstract:
General-purpose robots capable of performing diverse tasks require synergistic reasoning and acting capabilities. However, recent dual-system approaches, which separate high-level reasoning from low-level acting, often suffer from challenges such as limited mutual understanding of capabilities between systems and latency issues. This paper introduces OneTwoVLA, a single unified vision-language-action model that can perform both acting (System One) and reasoning (System Two). Crucially, OneTwoVLA adaptively switches between two modes: explicitly reasoning at critical moments during task execution, and generating actions based on the most recent reasoning at other times. To further unlock OneTwoVLA's reasoning and generalization capabilities, we design a scalable pipeline for synthesizing embodied reasoning-centric vision-language data, used for co-training with robot data. We validate OneTwoVLA's effectiveness through extensive experiments, highlighting its superior performance across four key capabilities: long-horizon task planning, error detection and recovery, natural human-robot interaction, and generalizable visual grounding, enabling the model to perform long-horizon, highly dexterous manipulation tasks such as making hotpot or mixing cocktails.

Paperid: 2916, https://arxiv.org/pdf/2505.08807.pdf

Abstract:
With the rise of large language and vision-language models, AI agents have evolved into autonomous, interactive systems capable of perception, reasoning, and decision-making. As they proliferate across virtual and physical domains, the Internet of Agents (IoA) has emerged as a key infrastructure for enabling scalable and secure coordination among heterogeneous agents. This survey offers a comprehensive examination of the security and privacy landscape in IoA systems. We begin by outlining the IoA architecture and its distinct vulnerabilities compared to traditional networks, focusing on four critical aspects: identity authentication threats, cross-agent trust issues, embodied security, and privacy risks. We then review existing and emerging defense mechanisms and highlight persistent challenges. Finally, we identify open research directions to advance the development of resilient and privacy-preserving IoA ecosystems.

Paperid: 2917, https://arxiv.org/pdf/2505.02056.pdf

Abstract:
Adapting vision-language models (VLMs) to downstream tasks with pseudolabels has gained increasing attention. A major obstacle is that the pseudolabels generated by VLMs tend to be imbalanced, leading to inferior performance. While existing methods have explored various strategies to address this, the underlying causes of imbalance remain insufficiently investigated. To fill this gap, we delve into imbalanced pseudolabels and identify two primary contributing factors: concept mismatch and concept confusion. To mitigate these two issues, we propose a novel framework incorporating concept alignment and confusion-aware calibrated margin mechanisms. The core of our approach lies in enhancing underperforming classes and promoting balanced predictions across categories, thus mitigating imbalance. Extensive experiments on six benchmark datasets with three learning paradigms demonstrate that the proposed method effectively enhances the accuracy and balance of pseudolabels, achieving a relative improvement of 6.29% over the SoTA method. Our code is avaliable at https://anonymous.4open.science/r/CAP-C642/

Paperid: 2918, https://arxiv.org/pdf/2504.13169.pdf

Abstract:
Vision-Language Models (VLMs) excel at visual understanding but often suffer from visual hallucinations, where they generate descriptions of nonexistent objects, actions, or concepts, posing significant risks in safety-critical applications. Existing hallucination mitigation methods typically follow one of two paradigms: generation adjustment, which modifies decoding behavior to align text with visual inputs, and post-hoc verification, where external models assess and correct outputs. While effective, generation adjustment methods often rely on heuristics and lack correction mechanisms, while post-hoc verification is complicated, typically requiring multiple models and tending to reject outputs rather than refine them. In this work, we introduce REVERSE, a unified framework that integrates hallucination-aware training with on-the-fly self-verification. By leveraging a new hallucination-verification dataset containing over 1.3M semi-synthetic samples, along with a novel inference-time retrospective resampling technique, our approach enables VLMs to both detect hallucinations during generation and dynamically revise those hallucinations. Our evaluations show that REVERSE achieves state-of-the-art hallucination reduction, outperforming the best existing methods by up to 12% on CHAIR-MSCOCO and 34% on HaloQuest. Our dataset, model, and code are available at: https://reverse-vlm.github.io.

Paperid: 2919, https://arxiv.org/pdf/2503.18578.pdf

Abstract:
Modern vision-language models (VLMs) develop patch embedding and convolution backbone within vector space, especially Euclidean ones, at the very founding. When expanding VLMs to a galaxy scale for understanding astronomical phenomena, the integration of spherical space for planetary orbits and hyperbolic spaces for black holes raises two formidable challenges. a) The current pre-training model is confined to Euclidean space rather than a comprehensive geometric embedding. b) The predominant architecture lacks suitable backbones for anisotropic physical geometries. In this paper, we introduced Galaxy-Walker, a geometry-aware VLM, for the universe-level vision understanding tasks. We proposed the geometry prompt that generates geometry tokens by random walks across diverse spaces on a multi-scale physical graph, along with a geometry adapter that compresses and reshapes the space anisotropy in a mixture-of-experts manner. Extensive experiments demonstrate the effectiveness of our approach, with Galaxy-Walker achieving state-of-the-art performance in both galaxy property estimation ($R^2$ scores up to $0.91$) and morphology classification tasks (up to $+0.17$ F1 improvement in challenging features), significantly outperforming both domain-specific models and general-purpose VLMs.

Paperid: 2920, https://arxiv.org/pdf/2503.10212.pdf

Abstract:
Analyzing animal behavior is crucial in advancing neuroscience, yet quantifying and deciphering its intricate dynamics remains a significant challenge. Traditional machine vision approaches, despite their ability to detect spontaneous behaviors, fall short due to limited interpretability and reliance on manual labeling, which restricts the exploration of the full behavioral spectrum. Here, we introduce MouseGPT, a Vision-Language Model (VLM) that integrates visual cues with natural language to revolutionize mouse behavior analysis. Built upon our first-of-its-kind dataset - incorporating pose dynamics and open-vocabulary behavioral annotations across over 42 million frames of diverse psychiatric conditions - MouseGPT provides a novel, context-rich method for comprehensive behavior interpretation. Our holistic analysis framework enables detailed behavior profiling, clustering, and novel behavior discovery, offering deep insights without the need for labor - intensive manual annotation. Evaluations reveal that MouseGPT surpasses existing models in precision, adaptability, and descriptive richness, positioning it as a transformative tool for ethology and for unraveling complex behavioral dynamics in animal models.

Paperid: 2921, https://arxiv.org/pdf/2501.19086.pdf

Abstract:
X-ray imaging is pivotal in medical diagnostics, offering non-invasive insights into a range of health conditions. Recently, vision-language models, such as the Contrastive Language-Image Pretraining (CLIP) model, have demonstrated potential in improving diagnostic accuracy by leveraging large-scale image-text datasets. However, since CLIP was not initially designed for medical images, several CLIP-like models trained specifically on medical images have been developed. Despite their enhanced performance, issues of fairness - particularly regarding demographic attributes - remain largely unaddressed. In this study, we perform a comprehensive fairness analysis of CLIP-like models applied to X-ray image classification. We assess their performance and fairness across diverse patient demographics and disease categories using zero-shot inference and various fine-tuning techniques, including Linear Probing, Multilayer Perceptron (MLP), Low-Rank Adaptation (LoRA), and full fine-tuning. Our results indicate that while fine-tuning improves model accuracy, fairness concerns persist, highlighting the need for further fairness interventions in these foundational models.

Paperid: 2922, https://arxiv.org/pdf/2501.17642.pdf

Abstract:
Open-vocabulary semantic segmentation (OVSS) is an open-world task that aims to assign each pixel within an image to a specific class defined by arbitrary text descriptions. Recent advancements in large-scale vision-language models have demonstrated their open-vocabulary understanding capabilities, significantly facilitating the development of OVSS. However, most existing methods suffer from either suboptimal performance or long latency. This study introduces ERR-Seg, a novel framework that effectively reduces redundancy to balance accuracy and efficiency. ERR-Seg incorporates a training-free Channel Reduction Module (CRM) that leverages prior knowledge from vision-language models like CLIP to identify the most relevant classes while discarding others. Moreover, it incorporates Efficient Semantic Context Fusion (ESCF) with spatial-level and class-level sequence reduction strategies. CRM and ESCF result in substantial memory and computational savings without compromising accuracy. Additionally, recognizing the significance of hierarchical semantics extracted from middle-layer features for closed-set semantic segmentation, ERR-Seg introduces the Hierarchical Semantic Module (HSM) to exploit hierarchical semantics in the context of OVSS. Compared to previous state-of-the-art methods under the ADE20K-847 setting, ERR-Seg achieves +$5.6\%$ mIoU improvement and reduces latency by $67.3\%$.

Paperid: 2923, https://arxiv.org/pdf/2412.15188.pdf

Abstract:
We present LMFusion, a framework for empowering pretrained text-only large language models (LLMs) with multimodal generative capabilities, enabling them to understand and generate both text and images in arbitrary sequences. LMFusion leverages existing Llama-3's weights for processing texts autoregressively while introducing additional and parallel transformer modules for processing images with diffusion. During training, the data from each modality is routed to its dedicated modules: modality-specific feedforward layers, query-key-value projections, and normalization layers process each modality independently, while the shared self-attention layers allow interactions across text and image features. By freezing the text-specific modules and only training the image-specific modules, LMFusion preserves the language capabilities of text-only LLMs while developing strong visual understanding and generation abilities. Compared to methods that pretrain multimodal generative models from scratch, our experiments demonstrate that, LMFusion improves image understanding by 20% and image generation by 3.6% using only 50% of the FLOPs while maintaining Llama-3's language capabilities. We also demonstrate that this framework can adapt existing vision-language models with multimodal generation ability. Overall, this framework not only leverages existing computational investments in text-only LLMs but also enables the parallel development of language and vision capabilities, presenting a promising direction for efficient multimodal model development.

Paperid: 2924, https://arxiv.org/pdf/2412.13180.pdf

Abstract:
Recent works on accelerating Vision-Language Models achieve strong performance across a variety of vision-language tasks despite highly compressing visual information. In this work, we examine the popular acceleration approach of early pruning of visual tokens inside the language model. Surprisingly, we find that while strong performance is maintained across many tasks, it exhibits drastically different behavior for a subset of vision-centric tasks such as localization. Upon further investigation, we uncover a core issue with the acceleration approach where most tokens towards the top of the image are pruned away. Yet, on many benchmarks aiming to evaluate vision-centric capabilities, strong performance persists with the flawed pruning strategy, highlighting these benchmarks' limited ability to assess fine-grained visual capabilities. Based on these findings, we propose FEATHER (Fast and Effective Acceleration wiTH Ensemble cRiteria), a straightforward approach that resolves the discovered early-layer pruning issue and further enhances the preservation of relevant tokens via multistage pruning with early uniform sampling to ensure broad image coverage. With comparable computational savings, we find that FEATHER achieves more than 5x performance improvement on the vision-centric localization benchmarks compared to the original acceleration approach.

Paperid: 2925, https://arxiv.org/pdf/2412.05892.pdf

Abstract:
Understanding the vulnerabilities of Large Vision Language Models (LVLMs) to jailbreak attacks is essential for their responsible real-world deployment. Most previous work requires access to model gradients, or is based on human knowledge (prompt engineering) to complete jailbreak, and they hardly consider the interaction of images and text, resulting in inability to jailbreak in black box scenarios or poor performance. To overcome these limitations, we propose a Prior-Guided Bimodal Interactive Black-Box Jailbreak Attack for toxicity maximization, referred to as PBI-Attack. Our method begins by extracting malicious features from a harmful corpus using an alternative LVLM and embedding these features into a benign image as prior information. Subsequently, we enhance these features through bidirectional cross-modal interaction optimization, which iteratively optimizes the bimodal perturbations in an alternating manner through greedy search, aiming to maximize the toxicity of the generated response. The toxicity level is quantified using a well-trained evaluation model. Experiments demonstrate that PBI-Attack outperforms previous state-of-the-art jailbreak methods, achieving an average attack success rate of 92.5% across three open-source LVLMs and around 67.3% on three closed-source LVLMs. Disclaimer: This paper contains potentially disturbing and offensive content.

Paperid: 2926, https://arxiv.org/pdf/2411.16749.pdf

Abstract:
Diffusion models have recently been employed to generate high-quality images, reducing the need for manual data collection and improving model generalization in tasks such as object detection, instance segmentation, and image perception. However, the synthetic framework is usually designed with meticulous human effort for each task due to various requirements on image layout, content, and annotation formats, restricting the application of synthetic data on more general scenarios. In this paper, we propose AnySynth, a unified framework integrating adaptable, comprehensive, and highly controllable components capable of generating an arbitrary type of synthetic data given diverse requirements. Specifically, the Task-Specific Layout Generation Module is first introduced to produce reasonable layouts for different tasks by leveraging the generation ability of large language models and layout priors of real-world images. A Uni-Controlled Image Generation Module is then developed to create high-quality synthetic images that are controllable and based on the generated layouts. In addition, user specific reference images, and style images can be incorporated into the generation to task requirements. Finally, the Task-Oriented Annotation Module offers precise and detailed annotations for the generated images across different tasks. We have validated our framework's performance across various tasks, including Few-shot Object Detection, Cross-domain Object Detection, Zero-shot Composed Image Retrieval, and Multi-modal Image Perception and Grounding. The specific data synthesized by our framework significantly improves model performance in these tasks, demonstrating the generality and effectiveness of our framework.

Paperid: 2927, https://arxiv.org/pdf/2410.11686.pdf

Abstract:
The advent of pre-trained vision-language foundation models has revolutionized the field of zero/few-shot (i.e., low-shot) image recognition. The key challenge to address under the condition of limited training data is how to fine-tune pre-trained vision-language models in a parameter-efficient manner. Previously, numerous approaches tackling this challenge have been proposed. Meantime, a few survey papers are also published to summarize these works. However, there still lacks a unified computational framework to integrate existing methods together, identify their nature and support in-depth comparison. As such, this survey paper first proposes a unified computational framework from the perspective of Representer Theorem and then derives many of the existing methods by specializing this framework. Thereafter, a comparative analysis is conducted to uncover the differences and relationships between existing methods. Based on the analyses, some possible variants to improve the existing works are presented. As a demonstration, we extend existing methods by modeling inter-class correlation between representers in reproducing kernel Hilbert space (RKHS), which is implemented by exploiting the closed-form solution of kernel ridge regression. Extensive experiments on 11 datasets are conducted to validate the effectiveness of this method. Toward the end of this paper, we discuss the limitations and provide further research directions.

Paperid: 2928, https://arxiv.org/pdf/2410.08895.pdf

Abstract:
Cache-based approaches stand out as both effective and efficient for adapting vision-language models (VLMs). Nonetheless, the existing cache model overlooks three crucial aspects. 1) Pre-trained VLMs are mainly optimized for image-text similarity, neglecting the importance of image-image similarity, leading to a gap between pre-training and adaptation. 2) The current cache model is based on the Nadaraya-Watson (N-W) estimator, which disregards the intricate relationships among training samples while constructing weight function. 3) Under the condition of limited samples, the logits generated by cache model are of high uncertainty, directly using these logits without accounting for the confidence could be problematic. This work presents three calibration modules aimed at addressing the above challenges. Similarity Calibration refines the image-image similarity by using unlabeled images. We add a learnable projection layer with residual connection on top of the pre-trained image encoder of CLIP and optimize the parameters by minimizing self-supervised contrastive loss. Weight Calibration introduces a precision matrix into the weight function to adequately model the relation between training samples, transforming the existing cache model to a Gaussian Process (GP) regressor, which could be more accurate than N-W estimator. Confidence Calibration leverages the predictive variances computed by GP Regression to dynamically re-scale the logits of cache model, ensuring that the cache model's outputs are appropriately adjusted based on their confidence levels. Besides, to reduce the high complexity of GPs, we further propose a group-based learning strategy. Integrating the above designs, we propose both training-free and training-required variants. Extensive experiments on 11 few-shot classification datasets validate that the proposed methods can achieve state-of-the-art performance.

Paperid: 2929, https://arxiv.org/pdf/2410.05160.pdf

Abstract:
Embedding models have been crucial in enabling various downstream tasks such as semantic similarity, information retrieval, and clustering. Recently, there has been a surge of interest in developing universal text embedding models that can generalize across tasks (e.g., MTEB). However, progress in learning universal multimodal embedding models has been relatively slow despite its importance and practicality. In this work, we aim to explore the potential for building universal embeddings capable of handling a wide range of downstream tasks. Our contributions are twofold: (1) MMEB (Massive Multimodal Embedding Benchmark), which covers 4 meta-tasks (i.e. classification, visual question answering, multimodal retrieval, and visual grounding) and 36 datasets, including 20 training and 16 evaluation datasets covering both in-distribution and out-of-distribution tasks, and (2) VLM2Vec (Vision-Language Model -> Vector), a contrastive training framework that converts any state-of-the-art vision-language model into an embedding model via training on MMEB. Unlike previous models such as CLIP and BLIP, which encodes text or images independently without any task instruction, VLM2Vec can process any combination of images and text to generate a fixed-dimensional vector based on task instructions. We build a series of VLM2Vec models on SoTA VLMs like Phi-3.5-V, LLaVA-1.6 and evaluate them on MMEB's evaluation split. Our results show that VLM2Vec achieves an absolute average improvement of 10% to 20% over existing multimodal embedding models on both in-distribution and out-of-distribution datasets in MMEB. We show that VLMs are secretly strong embedding models.

Paperid: 2930, https://arxiv.org/pdf/2410.00448.pdf

Abstract:
This paper introduces an innovative approach to Medical Vision-Language Pre-training (Med-VLP) area in the specialized context of radiograph representation learning. While conventional methods frequently merge textual annotations into unified reports, we acknowledge the intrinsic hierarchical relationship between the findings and impression section in radiograph datasets. To establish a targeted correspondence between images and texts, we propose a novel HybridMED framework to align global-level visual representations with impression and token-level visual representations with findings. Moreover, our framework incorporates a generation decoder that employs two proxy tasks, responsible for generating the impression from (1) images, via a captioning branch, and (2) findings, through a summarization branch. Additionally, knowledge distillation is leveraged to facilitate the training process. Experiments on the MIMIC-CXR dataset reveal that our summarization branch effectively distills knowledge to the captioning branch, enhancing model performance without significantly increasing parameter requirements due to the shared self-attention and feed-forward architecture.

Paperid: 2931, https://arxiv.org/pdf/2601.13133.pdf

Abstract:
Human-centric visual analysis plays a pivotal role in diverse applications, including surveillance, healthcare, and human-computer interaction. With the emergence of large-scale unlabeled human image datasets, there is an increasing need for a general unsupervised pre-training model capable of supporting diverse human-centric downstream tasks. To achieve this goal, we propose CLASP (CLIP-guided Adaptable Self-suPervised learning), a novel framework designed for unsupervised pre-training in human-centric visual tasks. CLASP leverages the powerful vision-language model CLIP to generate both low-level (e.g., body parts) and high-level (e.g., attributes) semantic pseudo-labels. These multi-level semantic cues are then integrated into the learned visual representations, enriching their expressiveness and generalizability. Recognizing that different downstream tasks demand varying levels of semantic granularity, CLASP incorporates a Prompt-Controlled Mixture-of-Experts (MoE) module. MoE dynamically adapts feature extraction based on task-specific prompts, mitigating potential feature conflicts and enhancing transferability. Furthermore, CLASP employs a multi-task pre-training strategy, where part- and attribute-level pseudo-labels derived from CLIP guide the representation learning process. Extensive experiments across multiple benchmarks demonstrate that CLASP consistently outperforms existing unsupervised pre-training methods, advancing the field of human-centric visual analysis.

Paperid: 2932, https://arxiv.org/pdf/2601.09113.pdf

Abstract:
Memory plays a foundational role in augmenting the reasoning, adaptability, and contextual fidelity of modern Large Language Models and Multi-Modal LLMs. As these models transition from static predictors to interactive systems capable of continual learning and personalized inference, the incorporation of memory mechanisms has emerged as a central theme in their architectural and functional evolution. This survey presents a comprehensive and structured synthesis of memory in LLMs and MLLMs, organizing the literature into a cohesive taxonomy comprising implicit, explicit, and agentic memory paradigms. Specifically, the survey delineates three primary memory frameworks. Implicit memory refers to the knowledge embedded within the internal parameters of pre-trained transformers, encompassing their capacity for memorization, associative retrieval, and contextual reasoning. Recent work has explored methods to interpret, manipulate, and reconfigure this latent memory. Explicit memory involves external storage and retrieval components designed to augment model outputs with dynamic, queryable knowledge representations, such as textual corpora, dense vectors, and graph-based structures, thereby enabling scalable and updatable interaction with information sources. Agentic memory introduces persistent, temporally extended memory structures within autonomous agents, facilitating long-term planning, self-consistency, and collaborative behavior in multi-agent systems, with relevance to embodied and interactive AI. Extending beyond text, the survey examines the integration of memory within multi-modal settings, where coherence across vision, language, audio, and action modalities is essential. Key architectural advances, benchmark tasks, and open challenges are discussed, including issues related to memory capacity, alignment, factual consistency, and cross-system interoperability.

Paperid: 2933, https://arxiv.org/pdf/2601.06748.pdf

Abstract:
Vision-Language-Action models have recently emerged as a powerful paradigm for general-purpose robot learning, enabling agents to map visual observations and natural-language instructions into executable robotic actions. Though popular, they are primarily trained via supervised fine-tuning or training-time reinforcement learning, requiring explicit fine-tuning phases, human interventions, or controlled data collection. Consequently, existing methods remain unsuitable for challenging simulated- or physical-world deployments, where robots must respond autonomously and flexibly to evolving environments. To address this limitation, we introduce a Test-Time Reinforcement Learning for VLAs (TT-VLA), a framework that enables on-the-fly policy adaptation during inference. TT-VLA formulates a dense reward mechanism that leverages step-by-step task-progress signals to refine action policies during test time while preserving the SFT/RL-trained priors, making it an effective supplement to current VLA models. Empirical results show that our approach enhances overall adaptability, stability, and task success in dynamic, previously unseen scenarios under simulated and real-world settings. We believe TT-VLA offers a principled step toward self-improving, deployment-ready VLAs.

Paperid: 2934, https://arxiv.org/pdf/2601.00898.pdf

Abstract:
Diffusion-based policies have gained growing popularity in solving a wide range of decision-making tasks due to their superior expressiveness and controllable generation during inference. However, effectively training large diffusion policies using reinforcement learning (RL) remains challenging. Existing methods either suffer from unstable training due to directly maximizing value objectives, or face computational issues due to relying on crude Gaussian likelihood approximation, which requires a large amount of sufficiently small denoising steps. In this work, we propose DIPOLE (Dichotomous diffusion Policy improvement), a novel RL algorithm designed for stable and controllable diffusion policy optimization. We begin by revisiting the KL-regularized objective in RL, which offers a desirable weighted regression objective for diffusion policy extraction, but often struggles to balance greediness and stability. We then formulate a greedified policy regularization scheme, which naturally enables decomposing the optimal policy into a pair of stably learned dichotomous policies: one aims at reward maximization, and the other focuses on reward minimization. Under such a design, optimized actions can be generated by linearly combining the scores of dichotomous policies during inference, thereby enabling flexible control over the level of greediness.Evaluations in offline and offline-to-online RL settings on ExORL and OGBench demonstrate the effectiveness of our approach. We also use DIPOLE to train a large vision-language-action (VLA) model for end-to-end autonomous driving (AD) and evaluate it on the large-scale real-world AD benchmark NAVSIM, highlighting its potential for complex real-world applications.

Paperid: 2935, https://arxiv.org/pdf/2512.10932.pdf

Abstract:
Early children's developmental trajectories set up a natural goal for sample-efficient pretraining of vision foundation models. We introduce BabyVLM-V2, a developmentally grounded framework for infant-inspired vision-language modeling that extensively improves upon BabyVLM-V1 through a longitudinal, multifaceted pretraining set, a versatile model, and, most importantly, DevCV Toolbox for cognitive evaluation. The pretraining set maximizes coverage while minimizing curation of a longitudinal, infant-centric audiovisual corpus, yielding video-utterance, image-utterance, and multi-turn conversational data that mirror infant experiences. DevCV Toolbox adapts all vision-related measures of the recently released NIH Baby Toolbox into a benchmark suite of ten multimodal tasks, covering spatial reasoning, memory, and vocabulary understanding aligned with early children's capabilities. Experimental results show that a compact model pretrained from scratch can achieve competitive performance on DevCV Toolbox, outperforming GPT-4o on some tasks. We hope the principled, unified BabyVLM-V2 framework will accelerate research in developmentally plausible pretraining of vision foundation models.

Paperid: 2936, https://arxiv.org/pdf/2512.10867.pdf

Abstract:
This paper introduces the concept of Microscopic Spatial Intelligence (MiSI), the capability to perceive and reason about the spatial relationships of invisible microscopic entities, which is fundamental to scientific discovery. To assess the potential of Vision-Language Models (VLMs) in this domain, we propose a systematic benchmark framework MiSI-Bench. This framework features over 163,000 question-answer pairs and 587,000 images derived from approximately 4,000 molecular structures, covering nine complementary tasks that evaluate abilities ranging from elementary spatial transformations to complex relational identifications. Experimental results reveal that current state-of-the-art VLMs perform significantly below human level on this benchmark. However, a fine-tuned 7B model demonstrates substantial potential, even surpassing humans in spatial transformation tasks, while its poor performance in scientifically-grounded tasks like hydrogen bond recognition underscores the necessity of integrating explicit domain knowledge for progress toward scientific AGI. The datasets are available at https://huggingface.co/datasets/zongzhao/MiSI-bench.

Paperid: 2937, https://arxiv.org/pdf/2512.06255.pdf

Abstract:
Existing fine-grained image retrieval (FGIR) methods learn discriminative embeddings by adopting semantically sparse one-hot labels derived from category names as supervision. While effective on seen classes, such supervision overlooks the rich semantics encoded in category names, hindering the modeling of comparability among cross-category details and, in turn, limiting generalization to unseen categories. To tackle this, we introduce LaFG, a Language-driven framework for Fine-Grained Retrieval that converts class names into attribute-level supervision using large language models (LLMs) and vision-language models (VLMs). Treating each name as a semantic anchor, LaFG prompts an LLM to generate detailed, attribute-oriented descriptions. To mitigate attribute omission in these descriptions, it leverages a frozen VLM to project them into a vision-aligned space, clustering them into a dataset-wide attribute vocabulary while harvesting complementary attributes from related categories. Leveraging this vocabulary, a global prompt template selects category-relevant attributes, which are aggregated into category-specific linguistic prototypes. These prototypes supervise the retrieval model to steer

Paperid: 2938, https://arxiv.org/pdf/2511.18810.pdf

Abstract:
Recent Vision-Language-Action (VLA) models reformulate vision-language models by tuning them with millions of robotic demonstrations. While they perform well when fine-tuned for a single embodiment or task family, extending them to multi-skill settings remains challenging: directly merging VLA experts trained on different tasks results in near-zero success rates. This raises a fundamental question: what prevents VLAs from mastering multiple skills within one model? With an empirical decomposition of learnable parameters during VLA fine-tuning, we identify two key sources of non-mergeability: (1) Finetuning drives LoRA adapters in the VLM backbone toward divergent, task-specific directions beyond the capacity of existing merging methods to unify. (2) Action experts develop inter-block dependencies through self-attention feedback, causing task information to spread across layers and preventing modular recombination. To address these challenges, we present MergeVLA, a merging-oriented VLA architecture that preserves mergeability by design. MergeVLA introduces sparsely activated LoRA adapters via task masks to retain consistent parameters and reduce irreconcilable conflicts in the VLM. Its action expert replaces self-attention with cross-attention-only blocks to keep specialization localized and composable. When the task is unknown, it uses a test-time task router to adaptively select the appropriate task mask and expert head from the initial observation, enabling unsupervised task inference. Across LIBERO, LIBERO-Plus, RoboTwin, and multi-task experiments on the real SO101 robotic arm, MergeVLA achieves performance comparable to or even exceeding individually finetuned experts, demonstrating robust generalization across tasks, embodiments, and environments.

Paperid: 2939, https://arxiv.org/pdf/2511.12436.pdf

Abstract:
Robotic manipulation and navigation are fundamental capabilities of embodied intelligence, enabling effective robot interactions with the physical world. Achieving these capabilities requires a cohesive understanding of the environment, including object recognition to localize target objects, object affordances to identify potential interaction areas and spatial affordances to discern optimal areas for both object placement and robot movement. While Vision-Language Models (VLMs) excel at high-level task planning and scene understanding, they often struggle to infer actionable positions for physical interaction, such as functional grasping points and permissible placement regions. This limitation stems from the lack of fine-grained annotations for object and spatial affordances in their training datasets. To tackle this challenge, we introduce RoboAfford++, a generative AI-enhanced dataset for multimodal affordance learning for both robotic manipulation and navigation. Our dataset comprises 869,987 images paired with 2.0 million question answering (QA) annotations, covering three critical tasks: object affordance recognition to identify target objects based on attributes and spatial relationships, object affordance prediction to pinpoint functional parts for manipulation, and spatial affordance localization to identify free space for object placement and robot navigation. Complementing this dataset, we propose RoboAfford-Eval, a comprehensive benchmark for assessing affordance-aware prediction in real-world scenarios, featuring 338 meticulously annotated samples across the same three tasks. Extensive experimental results reveal the deficiencies of existing VLMs in affordance learning, while fine-tuning on the RoboAfford++ dataset significantly enhances their ability to reason about object and spatial affordances, validating the dataset's effectiveness.

Paperid: 2940, https://arxiv.org/pdf/2511.11450.pdf

Abstract:
We introduce VoxTell, a vision-language model for text-prompted volumetric medical image segmentation. It maps free-form descriptions, from single words to full clinical sentences, to 3D masks. Trained on 62K+ CT, MRI, and PET volumes spanning over 1K anatomical and pathological classes, VoxTell uses multi-stage vision-language fusion across decoder layers to align textual and visual features at multiple scales. It achieves state-of-the-art zero-shot performance across modalities on unseen datasets, excelling on familiar concepts while generalizing to related unseen classes. Extensive experiments further demonstrate strong cross-modality transfer, robustness to linguistic variations and clinical language, as well as accurate instance-specific segmentation from real-world text. Code is available at: https://www.github.com/MIC-DKFZ/VoxTell

Paperid: 2941, https://arxiv.org/pdf/2511.11005.pdf

Abstract:
While recent Large Vision-Language Models (LVLMs) exhibit strong multimodal reasoning abilities, they often produce ungrounded or hallucinated responses because they rely too heavily on linguistic priors instead of visual evidence. This limitation highlights the absence of a quantitative measure of how much these models actually use visual information during reasoning. We propose Draft and Refine (DnR), an agent framework driven by a question-conditioned utilization metric. The metric quantifies the model's reliance on visual evidence by first constructing a query-conditioned relevance map to localize question-specific cues and then measuring dependence through relevance-guided probabilistic masking. Guided by this metric, the DnR agent refines its initial draft using targeted feedback from external visual experts. Each expert's output (such as boxes or masks) is rendered as visual cues on the image, and the model is re-queried to select the response that yields the largest improvement in utilization. This process strengthens visual grounding without retraining or architectural changes. Experiments across VQA and captioning benchmarks show consistent accuracy gains and reduced hallucination, demonstrating that measuring visual utilization provides a principled path toward more interpretable and evidence-driven multimodal agent systems.

Paperid: 2942, https://arxiv.org/pdf/2511.07403.pdf

Abstract:
Multimodal large language models (MLLMs) have achieved remarkable progress in vision-language tasks, but they continue to struggle with spatial understanding. Existing spatial MLLMs often rely on explicit 3D inputs or architecture-specific modifications, and remain constrained by large-scale datasets or sparse supervision. To address these limitations, we introduce SpatialThinker, a 3D-aware MLLM trained with RL to integrate structured spatial grounding with multi-step reasoning. The model simulates human-like spatial perception by constructing a scene graph of task-relevant objects and spatial relations, and reasoning towards an answer via dense spatial rewards. SpatialThinker consists of two key contributions: (1) a data synthesis pipeline that generates STVQA-7K, a high-quality spatial VQA dataset, and (2) online RL with a multi-objective dense spatial reward enforcing spatial grounding. SpatialThinker-7B outperforms supervised fine-tuning and the sparse RL baseline on spatial understanding and real-world VQA benchmarks, nearly doubling the base-model gain compared to sparse RL, and surpassing GPT-4o. These results showcase the effectiveness of combining spatial supervision with reward-aligned reasoning in enabling robust 3D spatial understanding with limited data and advancing MLLMs towards human-level visual reasoning.

Paperid: 2943, https://arxiv.org/pdf/2511.04711.pdf

Abstract:
Large-scale vision-language models, especially CLIP, have demonstrated remarkable performance across diverse downstream tasks. Soft prompts, as carefully crafted modules that efficiently adapt vision-language models to specific tasks, necessitate effective copyright protection. In this paper, we investigate model copyright protection by auditing whether suspicious third-party models incorporate protected soft prompts. While this can be viewed as a special case of model ownership auditing, our analysis shows that existing techniques are ineffective due to prompt learning's unique characteristics. Non-intrusive auditing is inherently prone to false positives when independent models share similar data distributions with victim models. Intrusive approaches also fail: backdoor methods designed for CLIP cannot embed functional triggers, while extending traditional DNN backdoor techniques to prompt learning suffers from harmfulness and ambiguity challenges. We find that these failures in intrusive auditing stem from the same fundamental reason: watermarking operates within the same decision space as the primary task yet pursues opposing objectives. Motivated by these findings, we propose sequential watermarking for soft prompts (SWAP), which implants watermarks into a different and more complex space. SWAP encodes watermarks through a specific order of defender-specified out-of-distribution classes, inspired by the zero-shot prediction capability of CLIP. This watermark, which is embedded in a more complex space, keeps the original prediction label unchanged, making it less opposed to the primary task. We further design a hypothesis-test-guided verification protocol for SWAP and provide theoretical analyses of success conditions. Extensive experiments on 11 datasets demonstrate SWAP's effectiveness, harmlessness, and robustness against potential adaptive attacks.

Paperid: 2944, https://arxiv.org/pdf/2510.23925.pdf

Abstract:
Chain-of-thought (CoT) reasoning is critical for improving the interpretability and reliability of Large Vision-Language Models (LVLMs). However, existing training algorithms such as SFT, PPO, and GRPO may not generalize well across unseen reasoning tasks and heavily rely on a biased reward model. To address this challenge, we reformulate reasoning in LVLMs as posterior inference and propose a scalable training algorithm based on amortized variational inference. By leveraging diversity-seeking reinforcement learning algorithms, we introduce a novel sparse reward function for token-level learning signals that encourage diverse, high-likelihood latent CoT, overcoming deterministic sampling limitations and avoiding reward hacking. Additionally, we implement a Bayesian inference-scaling strategy that replaces costly Best-of-N and Beam Search with a marginal likelihood to efficiently rank optimal rationales and answers. We empirically demonstrate that the proposed method enhances the state-of-the-art LVLMs on seven reasoning benchmarks, in terms of effectiveness, generalization, and interpretability.

Paperid: 2945, https://arxiv.org/pdf/2510.20244.pdf

Abstract:
Video Temporal Grounding (VTG) aims to localize temporal segments in long, untrimmed videos that align with a given natural language query. This task typically comprises two subtasks: Moment Retrieval (MR) and Highlight Detection (HD). While recent advances have been progressed by powerful pretrained vision-language models such as CLIP and InternVideo2, existing approaches commonly treat all text tokens uniformly during crossmodal attention, disregarding their distinct semantic roles. To validate the limitations of this approach, we conduct controlled experiments demonstrating that VTG models overly rely on [EOS]-driven global semantics while failing to effectively utilize word-level signals, which limits their ability to achieve fine-grained temporal alignment. Motivated by this limitation, we propose DualGround, a dual-branch architecture that explicitly separates global and local semantics by routing the [EOS] token through a sentence-level path and clustering word tokens into phrase-level units for localized grounding. Our method introduces (1) tokenrole- aware cross modal interaction strategies that align video features with sentence-level and phrase-level semantics in a structurally disentangled manner, and (2) a joint modeling framework that not only improves global sentence-level alignment but also enhances finegrained temporal grounding by leveraging structured phrase-aware context. This design allows the model to capture both coarse and localized semantics, enabling more expressive and context-aware video grounding. DualGround achieves state-of-the-art performance on both Moment Retrieval and Highlight Detection tasks across QVHighlights and Charades- STA benchmarks, demonstrating the effectiveness of disentangled semantic modeling in video-language alignment.

Paperid: 2946, https://arxiv.org/pdf/2510.10053.pdf

Abstract:
Deep learning based face-swap videos, widely known as deepfakes, have drawn wide attention due to their threat to information credibility. Recent works mainly focus on the problem of deepfake detection that aims to reliably tell deepfakes apart from real ones, in an objective way. On the other hand, the subjective perception of deepfakes, especially its computational modeling and imitation, is also a significant problem but lacks adequate study. In this paper, we focus on the visual realism assessment of deepfakes, which is defined as the automatic assessment of deepfake visual realism that approximates human perception of deepfakes. It is important for evaluating the quality and deceptiveness of deepfakes which can be used for predicting the influence of deepfakes on Internet, and it also has potentials in improving the deepfake generation process by serving as a critic. This paper prompts this new direction by presenting a comprehensive benchmark called DREAM, which stands for Deepfake REalism AssessMent. It is comprised of a deepfake video dataset of diverse quality, a large scale annotation that includes 140,000 realism scores and textual descriptions obtained from 3,500 human annotators, and a comprehensive evaluation and analysis of 16 representative realism assessment methods, including recent large vision language model based methods and a newly proposed description-aligned CLIP method. The benchmark and insights included in this study can lay the foundation for future research in this direction and other related areas.

Paperid: 2947, https://arxiv.org/pdf/2510.07778.pdf

Abstract:
Vision-Language-Action (VLA) models leverage pretrained vision-language models (VLMs) to couple perception with robotic control, offering a promising path toward general-purpose embodied intelligence. However, current SOTA VLAs are primarily pretrained on multimodal tasks with limited relevance to embodied scenarios, and then finetuned to map explicit instructions to actions. Consequently, due to the lack of reasoning-intensive pretraining and reasoning-guided manipulation, these models are unable to perform implicit human intention reasoning required for complex, real-world interactions. To overcome these limitations, we propose \textbf{IntentionVLA}, a VLA framework with a curriculum training paradigm and an efficient inference mechanism. Our proposed method first leverages carefully designed reasoning data that combine intention inference, spatial grounding, and compact embodied reasoning, endowing the model with both reasoning and perception capabilities. In the following finetuning stage, IntentionVLA employs the compact reasoning outputs as contextual guidance for action generation, enabling fast inference under indirect instructions. Experimental results show that IntentionVLA substantially outperforms $π_0$, achieving 18\% higher success rates with direct instructions and 28\% higher than ECoT under intention instructions. On out-of-distribution intention tasks, IntentionVLA achieves over twice the success rate of all baselines, and further enables zero-shot human-robot interaction with 40\% success rate. These results highlight IntentionVLA as a promising paradigm for next-generation human-robot interaction (HRI) systems.

Paperid: 2948, https://arxiv.org/pdf/2510.04532.pdf

Abstract:
Vision-Language Model (VLM) driving agents promise explainable end-to-end autonomy by first producing natural-language reasoning and then predicting trajectory planning. However, whether planning is causally driven by this reasoning remains a critical but unverified assumption. To investigate this, we build DriveMind, a large-scale driving Visual Question Answering (VQA) corpus with plan-aligned Chain-of-Thought (CoT), automatically generated from nuPlan. Our data generation process converts sensors and annotations into structured inputs and, crucially, separates priors from to-be-reasoned signals, enabling clean information ablations. Using DriveMind, we train representative VLM agents with Supervised Fine-Tuning (SFT) and Group Relative Policy Optimization (GRPO) and evaluate them with nuPlan's metrics. Our results, unfortunately, indicate a consistent causal disconnect in reasoning-planning: removing ego/navigation priors causes large drops in planning scores, whereas removing CoT produces only minor changes. Attention analysis further shows that planning primarily focuses on priors rather than the CoT. Based on this evidence, we propose the Reasoning-Planning Decoupling Hypothesis, positing that the training-yielded reasoning is an ancillary byproduct rather than a causal mediator. To enable efficient diagnosis, we also introduce a novel, training-free probe that measures an agent's reliance on priors by evaluating its planning robustness against minor input perturbations. In summary, we provide the community with a new dataset and a diagnostic tool to evaluate the causal fidelity of future models.

Paperid: 2949, https://arxiv.org/pdf/2510.02728.pdf

Abstract:
Cross-modal drone navigation remains a challenging task in robotics, requiring efficient retrieval of relevant images from large-scale databases based on natural language descriptions. The RoboSense 2025 Track 4 challenge addresses this challenge, focusing on robust, natural language-guided cross-view image retrieval across multiple platforms (drones, satellites, and ground cameras). Current baseline methods, while effective for initial retrieval, often struggle to achieve fine-grained semantic matching between text queries and visual content, especially in complex aerial scenes. To address this challenge, we propose a two-stage retrieval refinement method: Caption-Guided Retrieval System (CGRS) that enhances the baseline coarse ranking through intelligent reranking. Our method first leverages a baseline model to obtain an initial coarse ranking of the top 20 most relevant images for each query. We then use Vision-Language-Model (VLM) to generate detailed captions for these candidate images, capturing rich semantic descriptions of their visual content. These generated captions are then used in a multimodal similarity computation framework to perform fine-grained reranking of the original text query, effectively building a semantic bridge between the visual content and natural language descriptions. Our approach significantly improves upon the baseline, achieving a consistent 5\% improvement across all key metrics (Recall@1, Recall@5, and Recall@10). Our approach win TOP-2 in the challenge, demonstrating the practical value of our semantic refinement strategy in real-world robotic navigation scenarios.

Paperid: 2950, https://arxiv.org/pdf/2509.06932.pdf

Abstract:
The rapid progress of auto-regressive vision-language models (VLMs) has inspired growing interest in vision-language-action models (VLA) for robotic manipulation. Recently, masked diffusion models, a paradigm distinct from autoregressive models, have begun to demonstrate competitive performance in text generation and multimodal applications, leading to the development of a series of diffusion-based VLMs (d-VLMs). However, leveraging such models for robot policy learning remains largely unexplored. In this work, we present LLaDA-VLA, the first Vision-Language-Diffusion-Action model built upon pretrained d-VLMs for robotic manipulation. To effectively adapt d-VLMs to robotic domain, we introduce two key designs: (1) a localized special-token classification strategy that replaces full-vocabulary classification with special action token classification, reducing adaptation difficulty; (2) a hierarchical action-structured decoding strategy that decodes action sequences hierarchically considering the dependencies within and across actions. Extensive experiments demonstrate that LLaDA-VLA significantly outperforms state-of-the-art VLAs on both simulation and real-world robots.

Paperid: 2951, https://arxiv.org/pdf/2508.20072.pdf

Abstract:
Vision-Language-Action (VLA) models adapt large vision-language backbones to map images and instructions to robot actions. However, prevailing VLA decoders either generate actions autoregressively in a fixed left-to-right order or attach continuous diffusion or flow matching heads outside the backbone, demanding specialized training and iterative sampling that hinder a unified, scalable architecture. We present Discrete Diffusion VLA, a single-transformer policy that models discretized action chunks with discrete diffusion and is trained with the same cross-entropy objective as the VLM backbone. The design retains diffusion's progressive refinement paradigm while remaining natively compatible with the discrete token interface of VLMs. Our method achieves an adaptive decoding order that resolves easy action elements before harder ones and uses secondary remasking to revisit uncertain predictions across refinement rounds, which improves consistency and enables robust error correction. This unified decoder preserves pretrained vision language priors, supports parallel decoding, breaks the autoregressive bottleneck, and reduces the number of function evaluations. Discrete Diffusion VLA achieves 96.3% avg. SR on LIBERO, 71.2% visual matching on SimplerEnv Fractal and 49.3% overall on SimplerEnv Bridge, improving over both autoregressive and continuous diffusion baselines. These findings indicate that discrete-diffusion action decoder supports precise action modeling and consistent training, laying groundwork for scaling VLA to larger models and datasets.

Paperid: 2952, https://arxiv.org/pdf/2508.17467.pdf

Abstract:
Mixture of Experts (MoE) models have enabled the scaling of Large Language Models (LLMs) and Vision Language Models (VLMs) by achieving massive parameter counts while maintaining computational efficiency. However, MoEs introduce several inference-time challenges, including load imbalance across experts and the additional routing computational overhead. To address these challenges and fully harness the benefits of MoE, a systematic evaluation of hardware acceleration techniques is essential. We present MoE-Inference-Bench, a comprehensive study to evaluate MoE performance across diverse scenarios. We analyze the impact of batch size, sequence length, and critical MoE hyperparameters such as FFN dimensions and number of experts on throughput. We evaluate several optimization techniques on Nvidia H100 GPUs, including pruning, Fused MoE operations, speculative decoding, quantization, and various parallelization strategies. Our evaluation includes MoEs from the Mixtral, DeepSeek, OLMoE and Qwen families. The results reveal performance differences across configurations and provide insights for the efficient deployment of MoEs.

Paperid: 2953, https://arxiv.org/pdf/2508.16569.pdf

Abstract:
The non-invasive assessment of increasingly incidentally discovered renal masses is a critical challenge in urologic oncology, where diagnostic uncertainty frequently leads to the overtreatment of benign or indolent tumors. In this study, we developed and validated RenalCLIP using a dataset of 27,866 CT scans from 8,809 patients across nine Chinese medical centers and the public TCIA cohort, a visual-language foundation model for characterization, diagnosis and prognosis of renal mass. The model was developed via a two-stage pre-training strategy that first enhances the image and text encoders with domain-specific knowledge before aligning them through a contrastive learning objective, to create robust representations for superior generalization and diagnostic precision. RenalCLIP achieved better performance and superior generalizability across 10 core tasks spanning the full clinical workflow of kidney cancer, including anatomical assessment, diagnostic classification, and survival prediction, compared with other state-of-the-art general-purpose CT foundation models. Especially, for complicated task like recurrence-free survival prediction in the TCIA cohort, RenalCLIP achieved a C-index of 0.726, representing a substantial improvement of approximately 20% over the leading baselines. Furthermore, RenalCLIP's pre-training imparted remarkable data efficiency; in the diagnostic classification task, it only needs 20% training data to achieve the peak performance of all baseline models even after they were fully fine-tuned on 100% of the data. Additionally, it achieved superior performance in report generation, image-text retrieval and zero-shot diagnosis tasks. Our findings establish that RenalCLIP provides a robust tool with the potential to enhance diagnostic accuracy, refine prognostic stratification, and personalize the management of patients with kidney cancer.

Paperid: 2954, https://arxiv.org/pdf/2508.03209.pdf

Abstract:
Vision-Language Models (VLMs) such as GPT-4o now demonstrate a remarkable ability to infer users' locations from public shared images, posing a substantial risk to geoprivacy. Although adversarial perturbations offer a potential defense, current methods are ill-suited for this scenario: they often perform poorly on high-resolution images and low perturbation budgets, and may introduce irrelevant semantic content. To address these limitations, we propose GeoShield, a novel adversarial framework designed for robust geoprivacy protection in real-world scenarios. GeoShield comprises three key modules: a feature disentanglement module that separates geographical and non-geographical information, an exposure element identification module that pinpoints geo-revealing regions within an image, and a scale-adaptive enhancement module that jointly optimizes perturbations at both global and local levels to ensure effectiveness across resolutions. Extensive experiments on challenging benchmarks show that GeoShield consistently surpasses prior methods in black-box settings, achieving strong privacy protection with minimal impact on visual or semantic quality. To our knowledge, this work is the first to explore adversarial perturbations for defending against geolocation inference by advanced VLMs, providing a practical and effective solution to escalating privacy concerns.

Paperid: 2955, https://arxiv.org/pdf/2507.18433.pdf

Abstract:
Multimodal large models have shown great potential in automating pathology image analysis. However, current multimodal models for gastrointestinal pathology are constrained by both data quality and reasoning transparency: pervasive noise and incomplete annotations in public datasets predispose vision language models to factual hallucinations when generating diagnostic text, while the absence of explicit intermediate reasoning chains renders the outputs difficult to audit and thus less trustworthy in clinical practice. To address these issues, we construct a large scale gastrointestinal pathology dataset containing both microscopic descriptions and diagnostic conclusions, and propose a prompt argumentation strategy that incorporates lesion classification and anatomical site information. This design guides the model to better capture image specific features and maintain semantic consistency in generation. Furthermore, we employ a post training pipeline that combines supervised fine tuning with Group Relative Policy Optimization (GRPO) to improve reasoning quality and output structure. Experimental results on real world pathology report generation tasks demonstrate that our approach significantly outperforms state of the art open source and proprietary baselines in terms of generation quality, structural completeness, and clinical relevance. Our solution outperforms state of the art models with 18.7% higher clinical relevance, 32.4% improved structural completeness, and 41.2% fewer diagnostic errors, demonstrating superior accuracy and clinical utility compared to existing solutions.

Paperid: 2956, https://arxiv.org/pdf/2506.19694.pdf

Abstract:
Precise anomaly detection in medical images is critical for clinical decision-making. While recent unsupervised or semi-supervised anomaly detection methods trained on large-scale normal data show promising results, they lack fine-grained differentiation, such as benign vs. malignant tumors. Additionally, ultrasound (US) imaging is highly sensitive to devices and acquisition parameter variations, creating significant domain gaps in the resulting US images. To address these challenges, we propose UltraAD, a vision-language model (VLM)-based approach that leverages few-shot US examples for generalized anomaly localization and fine-grained classification. To enhance localization performance, the image-level token of query visual prototypes is first fused with learnable text embeddings. This image-informed prompt feature is then further integrated with patch-level tokens, refining local representations for improved accuracy. For fine-grained classification, a memory bank is constructed from few-shot image samples and corresponding text descriptions that capture anatomical and abnormality-specific features. During training, the stored text embeddings remain frozen, while image features are adapted to better align with medical data. UltraAD has been extensively evaluated on three breast US datasets, outperforming state-of-the-art methods in both lesion localization and fine-grained medical classification. The code will be released upon acceptance.

Paperid: 2957, https://arxiv.org/pdf/2506.16691.pdf

Abstract:
Despite the impressive advancements of Large Vision-Language Models (LVLMs), existing approaches suffer from a fundamental bottleneck: inefficient visual-language integration. Current methods either disrupt the model's inherent structure or introduce severe long-context computational burden, severely limiting scalability and efficiency. In this paper, we rethink multimodal integration and present LaVi, a novel LVLM that enables seamless and efficient vision-language fusion through internal feature modulation within the Large Language Models (LLMs). Unlike dominant LVLMs that rely on visual token concatenation, LaVi bypasses long-context expansion by introducing a lightweight and adaptive transformation, which incorporates visual context by injecting token-wise vision-conditioned deltas into the affine parameters of layer normalization. This mechanism directly modulates linguistic hidden states based on visual input, ensuring precise vision-language alignment while preserving the LLM's linguistic priors and drastically reducing computational costs. Extensive evaluations across 15 image and video benchmarks demonstrate that LaVi not only achieves state-of-the-art multimodal performance but also dramatically enhances efficiency. Compared to LLaVA-OV-7B, LaVi reduces FLOPs by 94.0%, improves inference speed by 3.1 times, and cuts memory usage in half - establishing LaVi as a scalable and practical solution for real-time multimodal reasoning. The code and models will be released soon.

Paperid: 2958, https://arxiv.org/pdf/2506.13679.pdf

Abstract:
Vision-Language-Action (VLA) models have recently made significant advance in multi-task, end-to-end robotic control, due to the strong generalization capabilities of Vision-Language Models (VLMs). A fundamental challenge in developing such models is effectively aligning the vision-language space with the robotic action space. Existing approaches typically rely on directly fine-tuning VLMs using expert demonstrations. However, this strategy suffers from a spatio-temporal gap, resulting in considerable data inefficiency and heavy reliance on human labor. Spatially, VLMs operate within a high-level semantic space, whereas robotic actions are grounded in low-level 3D physical space; temporally, VLMs primarily interpret the present, while VLA models anticipate future actions. To overcome these challenges, we propose a novel training paradigm, ROSA, which leverages robot state estimation to improve alignment between vision-language and action spaces. By integrating robot state estimation data obtained via an automated process, ROSA enables the VLA model to gain enhanced spatial understanding and self-awareness, thereby boosting performance and generalization. Extensive experiments in both simulated and real-world environments demonstrate the effectiveness of ROSA, particularly in low-data regimes.

Paperid: 2959, https://arxiv.org/pdf/2506.03095.pdf

Abstract:
Computer use agents (CUA) are systems that automatically interact with graphical user interfaces (GUIs) to complete tasks. CUA have made significant progress with the advent of large vision-language models (VLMs). However, these agents typically rely on cloud-based inference with substantial compute demands, raising critical privacy and scalability concerns, especially when operating on personal devices. In this work, we take a step toward privacy-preserving and resource-efficient agents by developing a lightweight vision-language model that runs entirely on local machines. To train this compact agent, we introduce an LLM-as-Judge framework that automatically evaluates and filters synthetic interaction trajectories, producing high-quality data for reinforcement learning without human annotation. Experiments on the OS-World benchmark demonstrate that our fine-tuned local model outperforms existing baselines, highlighting a promising path toward private, efficient, and generalizable GUI agents.

Paperid: 2960, https://arxiv.org/pdf/2506.02327.pdf

Abstract:
Providing effective treatment and making informed clinical decisions are essential goals of modern medicine and clinical care. We are interested in simulating disease dynamics for clinical decision-making, leveraging recent advances in large generative models. To this end, we introduce the Medical World Model (MeWM), the first world model in medicine that visually predicts future disease states based on clinical decisions. MeWM comprises (i) vision-language models to serve as policy models, and (ii) tumor generative models as dynamics models. The policy model generates action plans, such as clinical treatments, while the dynamics model simulates tumor progression or regression under given treatment conditions. Building on this, we propose the inverse dynamics model that applies survival analysis to the simulated post-treatment tumor, enabling the evaluation of treatment efficacy and the selection of the optimal clinical action plan. As a result, the proposed MeWM simulates disease dynamics by synthesizing post-treatment tumors, with state-of-the-art specificity in Turing tests evaluated by radiologists. Simultaneously, its inverse dynamics model outperforms medical-specialized GPTs in optimizing individualized treatment protocols across all metrics. Notably, MeWM improves clinical decision-making for interventional physicians, boosting F1-score in selecting the optimal TACE protocol by 13%, paving the way for future integration of medical world models as the second readers.

Paperid: 2961, https://arxiv.org/pdf/2506.00479.pdf

Abstract:
Large Vision-Language Models (LVLMs) have achieved remarkable success, yet their significant computational demands hinder practical deployment. While efforts to improve LVLM efficiency are growing, existing methods lack comprehensive evaluation across diverse backbones, benchmarks, and metrics. In this work, we systematically evaluate mainstream acceleration techniques for LVLMs, categorized into token and parameter compression. We introduce EffiVLM-Bench, a unified framework for assessing not only absolute performance but also generalization and loyalty, while exploring Pareto-optimal trade-offs. Our extensive experiments and in-depth analyses offer insights into optimal strategies for accelerating LVLMs. We open-source code and recipes for EffiVLM-Bench to foster future research.

Paperid: 2962, https://arxiv.org/pdf/2505.20326.pdf

Abstract:
Vision-Language Models (VLMs) are increasingly deployed in diverse cultural contexts, yet their internal biases remain poorly understood. In this work, we propose a novel framework to systematically evaluate how VLMs encode cultural differences and biases related to race, gender, and physical traits across countries. We introduce three retrieval-based tasks: (1) Race to Country retrieval, which examines the association between individuals from specific racial groups (East Asian, White, Middle Eastern, Latino, South Asian, and Black) and different countries; (2) Personal Traits to Country retrieval, where images are paired with trait-based prompts (e.g., Smart, Honest, Criminal, Violent) to investigate potential stereotypical associations; and (3) Physical Characteristics to Country retrieval, focusing on visual attributes like skinny, young, obese, and old to explore how physical appearances are culturally linked to nations. Our findings reveal persistent biases in VLMs, highlighting how visual representations may inadvertently reinforce societal stereotypes.

Paperid: 2963, https://arxiv.org/pdf/2505.19767.pdf

Abstract:
Vision-Language-Action (VLA) models have demonstrated significant potential in the field of embodied intelligence, enabling agents to follow human instructions to complete complex tasks in physical environments. Existing embodied agents are often trained through behavior cloning, which requires expensive data and computational resources and is constrained by human demonstrations. To address this issue, many researchers explore the application of reinforcement fine-tuning to embodied agents. However, typical reinforcement fine-tuning methods for embodied agents usually rely on sparse, outcome-based rewards, which struggle to provide fine-grained feedback for specific actions within an episode, thus limiting the model's manipulation capabilities and generalization performance. In this paper, we propose RFTF, a novel reinforcement fine-tuning method that leverages a value model to generate dense rewards in embodied scenarios. Specifically, our value model is trained using temporal information, eliminating the need for costly robot action labels. In addition, RFTF incorporates a range of techniques, such as GAE and sample balance to enhance the effectiveness of the fine-tuning process. By addressing the sparse reward problem in reinforcement fine-tuning, our method significantly improves the performance of embodied agents, delivering superior generalization and adaptation capabilities across diverse embodied tasks. Experimental results show that embodied agents fine-tuned with RFTF achieve new state-of-the-art performance on the challenging CALVIN ABC-D with an average success length of 4.296. Moreover, RFTF enables rapid adaptation to new environments. After fine-tuning in the D environment of CALVIN for a few episodes, RFTF achieved an average success length of 4.301 in this new environment.

Paperid: 2964, https://arxiv.org/pdf/2504.01901.pdf

Abstract:
The rapid development of Large Multimodal Models (LMMs) for 2D images and videos has spurred efforts to adapt these models for interpreting 3D scenes. However, the absence of large-scale 3D vision-language datasets has posed a significant obstacle. To address this issue, typical approaches focus on injecting 3D awareness into 2D LMMs by designing 3D input-level scene representations. This work provides a new perspective. We introduce reconstructive visual instruction tuning with 3D-awareness (Ross3D), which integrates 3D-aware visual supervision into the training procedure. Specifically, it incorporates cross-view and global-view reconstruction. The former requires reconstructing masked views by aggregating overlapping information from other views. The latter aims to aggregate information from all available views to recover Bird's-Eye-View images, contributing to a comprehensive overview of the entire scene. Empirically, Ross3D achieves state-of-the-art performance across various 3D scene understanding benchmarks. More importantly, our semi-supervised experiments demonstrate significant potential in leveraging large amounts of unlabeled 3D vision-only data.

Paperid: 2965, https://arxiv.org/pdf/2503.19670.pdf

Abstract:
While vision-language models like CLIP have advanced zero-shot surgical phase recognition, they struggle with fine-grained surgical activities, especially action triplets. This limitation arises because current CLIP formulations rely on global image features, which overlook the fine-grained semantics and contextual details crucial for complex tasks like zero-shot triplet recognition. Furthermore, these models do not explore the hierarchical structure inherent in triplets, reducing their ability to generalize to novel triplets. To address these challenges, we propose fine-CLIP, which learns object-centric features and leverages the hierarchy in triplet formulation. Our approach integrates three components: hierarchical prompt modeling to capture shared semantics, LoRA-based vision backbone adaptation for enhanced feature extraction, and a graph-based condensation strategy that groups similar patch features into meaningful object clusters. Since triplet classification is a challenging task, we introduce an alternative yet meaningful base-to-novel generalization benchmark with two settings on the CholecT50 dataset: Unseen-Target, assessing adaptability to triplets with novel anatomical structures, and Unseen-Instrument-Verb, where models need to generalize to novel instrument-verb interactions. fine-CLIP shows significant improvements in F1 and mAP, enhancing zero-shot recognition of novel surgical triplets.

Paperid: 2966, https://arxiv.org/pdf/2503.11103.pdf

Abstract:
CLIP is one of the most popular foundation models and is heavily used for many vision-language tasks, yet little is known about its inner workings. As CLIP is increasingly deployed in real-world applications, it is becoming even more critical to understand its limitations and embedded social biases to mitigate potentially harmful downstream consequences. However, the question of what internal mechanisms drive both the impressive capabilities as well as problematic shortcomings of CLIP has largely remained unanswered. To bridge this gap, we study the conceptual consistency of text descriptions for attention heads in CLIP-like models. Specifically, we propose Concept Consistency Score (CCS), a novel interpretability metric that measures how consistently individual attention heads in CLIP models align with specific concepts. Our soft-pruning experiments reveal that high CCS heads are critical for preserving model performance, as pruning them leads to a significantly larger performance drop than pruning random or low CCS heads. Notably, we find that high CCS heads capture essential concepts and play a key role in out-of-domain detection, concept-specific reasoning, and video-language understanding. Moreover, we prove that high CCS heads learn spurious correlations which amplify social biases. These results position CCS as a powerful interpretability metric exposing the paradox of performance and social biases in CLIP models.

Paperid: 2967, https://arxiv.org/pdf/2503.09248.pdf

Abstract:
Test-time adaptation with pre-trained vision-language models, such as CLIP, aims to adapt the model to new, potentially out-of-distribution test data. Existing methods calculate the similarity between visual embedding and learnable class embeddings, which are initialized by text embeddings, for zero-shot image classification. In this work, we first analyze this process based on Bayes theorem, and observe that the core factors influencing the final prediction are the likelihood and the prior. However, existing methods essentially focus on adapting class embeddings to adapt likelihood, but they often ignore the importance of prior. To address this gap, we propose a novel approach, \textbf{B}ayesian \textbf{C}lass \textbf{A}daptation (BCA), which in addition to continuously updating class embeddings to adapt likelihood, also uses the posterior of incoming samples to continuously update the prior for each class embedding. This dual updating mechanism allows the model to better adapt to distribution shifts and achieve higher prediction accuracy. Our method not only surpasses existing approaches in terms of performance metrics but also maintains superior inference rates and memory usage, making it highly efficient and practical for real-world applications.

Paperid: 2968, https://arxiv.org/pdf/2502.01218.pdf

Abstract:
Pre-training vision-language representations on human action videos has emerged as a promising approach to reduce reliance on large-scale expert demonstrations for training embodied agents. However, prior methods often employ time contrastive learning based on goal-reaching heuristics, progressively aligning language instructions from the initial to the final frame. This overemphasis on future frames can result in erroneous vision-language associations, as actions may terminate early or include irrelevant moments in the end. To address this issue, we propose Action Temporal Coherence Learning (AcTOL) to learn ordered and continuous vision-language representations without rigid goal-based constraint. AcTOL treats a video as a continuous trajectory where it (1) contrasts semantic differences between frames to reflect their natural ordering, and (2) imposes a local Brownian bridge constraint to ensure smooth transitions across intermediate frames. Extensive imitation learning experiments on both simulated and real robots show that the pretrained features significantly enhance downstream manipulation tasks with high robustness to different linguistic styles of instructions, offering a viable pathway toward generalized embodied agents.

Paperid: 2969, https://arxiv.org/pdf/2412.10362.pdf

Abstract:
Low-rank adapters enable fine-tuning of large models with only a small number of parameters, thus reducing storage costs and minimizing the risk of catastrophic forgetting. However, they often pose optimization challenges, with poor convergence. To overcome these challenges, we introduce an over-parameterized approach that accelerates training without increasing inference costs. This method reparameterizes low-rank adaptation by employing a separate MLP and learned embedding for each layer. The learned embedding is input to the MLP, which generates the adapter parameters. Such overparamaterization has been shown to implicitly function as an adaptive learning rate and momentum, accelerating optimization. At inference time, the MLP can be discarded, leaving behind a standard low-rank adapter. To study the effect of MLP overparameterization on a small yet difficult proxy task, we implement it for matrix factorization, and find it achieves faster convergence and lower final loss. Extending this approach to larger-scale tasks, we observe consistent performance gains across domains. We achieve improvements in vision-language tasks and especially notable increases in image generation, with CMMD scores improving by up to 15 points.

Paperid: 2970, https://arxiv.org/pdf/2412.06745.pdf

Abstract:
Traditional fixed test sets fall short in evaluating open-ended capabilities of foundation models. To address this, we propose ONEBench(OpeN-Ended Benchmarking), a new testing paradigm that consolidates individual evaluation datasets into a unified, ever-expanding sample pool. ONEBench allows users to generate custom, open-ended evaluation benchmarks from this pool, corresponding to specific capabilities of interest. By aggregating samples across test sets, ONEBench enables the assessment of diverse capabilities beyond those covered by the original test sets, while mitigating overfitting and dataset bias. Most importantly, it frames model evaluation as a collective process of selecting and aggregating sample-level tests. The shift from task-specific benchmarks to ONEBench introduces two challenges: (1)heterogeneity and (2)incompleteness. Heterogeneity refers to the aggregation over diverse metrics, while incompleteness describes comparing models evaluated on different data subsets. To address these challenges, we explore algorithms to aggregate sparse measurements into reliable model scores. Our aggregation algorithm ensures identifiability(asymptotically recovering ground-truth scores) and rapid convergence, enabling accurate model ranking with less data. On homogenous datasets, we show our aggregation algorithm provides rankings that highly correlate with those produced by average scores. We also demonstrate robustness to ~95% of measurements missing, reducing evaluation cost by up to 20x with little-to-no change in model rankings. We introduce ONEBench-LLM for language models and ONEBench-LMM for vision-language models, unifying evaluations across these domains. Overall, we present a technique for open-ended evaluation, which can aggregate over incomplete, heterogeneous sample-level measurements to continually grow a benchmark alongside the rapidly developing foundation models.

Paperid: 2971, https://arxiv.org/pdf/2411.17771.pdf

Abstract:
Visual Question Generation (VQG) research focuses predominantly on natural images while neglecting the diagram, which is a critical component in educational materials. To meet the needs of pedagogical assessment, we propose the Diagram-Driven Course Questions Generation (DDCQG) task and construct DiagramQG, a comprehensive dataset with 15,720 diagrams and 25,798 questions across 37 subjects and 371 courses. Our approach employs course and input text constraints to generate course-relevant questions about specific diagram elements. We reveal three challenges of DDCQG: domain-specific knowledge requirements across courses, long-tail distribution in course coverage, and high information density in diagrams. To address these, we propose the Hierarchical Knowledge Integration framework (HKI-DDCQG), which utilizes trainable CLIP for identifying relevant diagram patches, leverages frozen vision-language models for knowledge extraction, and generates questions with trainable T5. Experiments demonstrate that HKI-DDCQG outperforms existing models on DiagramQG while maintaining strong generalizability across natural image datasets, establishing a strong baseline for DDCQG.

Paperid: 2972, https://arxiv.org/pdf/2411.17451.pdf

Abstract:
Vision-language generative reward models (VL-GenRMs) play a crucial role in aligning and evaluating multimodal AI systems, yet their own evaluation remains under-explored. Current assessment methods primarily rely on AI-annotated preference labels from traditional VL tasks, which can introduce biases and often fail to effectively challenge state-of-the-art models. To address these limitations, we introduce VL-RewardBench, a comprehensive benchmark spanning general multimodal queries, visual hallucination detection, and complex reasoning tasks. Through our AI-assisted annotation pipeline that combines sample selection with human verification, we curate 1,250 high-quality examples specifically designed to probe VL-GenRMs limitations. Comprehensive evaluation across 16 leading large vision-language models demonstrates VL-RewardBench's effectiveness as a challenging testbed, where even GPT-4o achieves only 65.4% accuracy, and state-of-the-art open-source models such as Qwen2-VL-72B, struggle to surpass random-guessing. Importantly, performance on VL-RewardBench strongly correlates (Pearson's r $>$ 0.9) with MMMU-Pro accuracy using Best-of-N sampling with VL-GenRMs. Analysis experiments uncover three critical insights for improving VL-GenRMs: (i) models predominantly fail at basic visual perception tasks rather than reasoning tasks; (ii) inference-time scaling benefits vary dramatically by model capacity; and (iii) training VL-GenRMs to learn to judge substantially boosts judgment capability (+14.7% accuracy for a 7B VL-GenRM). We believe VL-RewardBench along with the experimental insights will become a valuable resource for advancing VL-GenRMs.

Paperid: 2973, https://arxiv.org/pdf/2411.17188.pdf

Abstract:
Many real-world user queries (e.g. "How do to make egg fried rice?") could benefit from systems capable of generating responses with both textual steps with accompanying images, similar to a cookbook. Models designed to generate interleaved text and images face challenges in ensuring consistency within and across these modalities. To address these challenges, we present ISG, a comprehensive evaluation framework for interleaved text-and-image generation. ISG leverages a scene graph structure to capture relationships between text and image blocks, evaluating responses on four levels of granularity: holistic, structural, block-level, and image-specific. This multi-tiered evaluation allows for a nuanced assessment of consistency, coherence, and accuracy, and provides interpretable question-answer feedback. In conjunction with ISG, we introduce a benchmark, ISG-Bench, encompassing 1,150 samples across 8 categories and 21 subcategories. This benchmark dataset includes complex language-vision dependencies and golden answers to evaluate models effectively on vision-centric tasks such as style transfer, a challenging area for current models. Using ISG-Bench, we demonstrate that recent unified vision-language models perform poorly on generating interleaved content. While compositional approaches that combine separate language and image models show a 111% improvement over unified models at the holistic level, their performance remains suboptimal at both block and image levels. To facilitate future work, we develop ISG-Agent, a baseline agent employing a "plan-execute-refine" pipeline to invoke tools, achieving a 122% performance improvement.

Paperid: 2974, https://arxiv.org/pdf/2411.15411.pdf

Abstract:
The advent of large Vision-Language Models (VLMs) has significantly advanced multimodal tasks, enabling more sophisticated and accurate reasoning across various applications, including image and video captioning, visual question answering, and cross-modal retrieval. Despite their superior capabilities, VLMs struggle with fine-grained image regional composition information perception. Specifically, they have difficulty accurately aligning the segmentation masks with the corresponding semantics and precisely describing the compositional aspects of the referred regions. However, compositionality - the ability to understand and generate novel combinations of known visual and textual components - is critical for facilitating coherent reasoning and understanding across modalities by VLMs. To address this issue, we propose FINECAPTION, a novel VLM that can recognize arbitrary masks as referential inputs and process high-resolution images for compositional image captioning at different granularity levels. To support this endeavor, we introduce COMPOSITIONCAP, a new dataset for multi-grained region compositional image captioning, which introduces the task of compositional attribute-aware regional image captioning. Empirical results demonstrate the effectiveness of our proposed model compared to other state-of-the-art VLMs. Additionally, we analyze the capabilities of current VLMs in recognizing various visual prompts for compositional region image captioning, highlighting areas for improvement in VLM design and training.

Paperid: 2975, https://arxiv.org/pdf/2411.14723.pdf

Abstract:
Open-vocabulary semantic segmentation aims to assign pixel-level labels to images across an unlimited range of classes. Traditional methods address this by sequentially connecting a powerful mask proposal generator, such as the Segment Anything Model (SAM), with a pre-trained vision-language model like CLIP. But these two-stage approaches often suffer from high computational costs, memory inefficiencies. In this paper, we propose ESC-Net, a novel one-stage open-vocabulary segmentation model that leverages the SAM decoder blocks for class-agnostic segmentation within an efficient inference framework. By embedding pseudo prompts generated from image-text correlations into SAM's promptable segmentation framework, ESC-Net achieves refined spatial aggregation for accurate mask predictions. ESC-Net achieves superior performance on standard benchmarks, including ADE20K, PASCAL-VOC, and PASCAL-Context, outperforming prior methods in both efficiency and accuracy. Comprehensive ablation studies further demonstrate its robustness across challenging conditions.

Paperid: 2976, https://arxiv.org/pdf/2411.06702.pdf

Abstract:
In the Detection and Multi-Object Tracking of Sweet Peppers Challenge, we present Track Any Peppers (TAP) - a weakly supervised ensemble technique for sweet peppers tracking. TAP leverages the zero-shot detection capabilities of vision-language foundation models like Grounding DINO to automatically generate pseudo-labels for sweet peppers in video sequences with minimal human intervention. These pseudo-labels, refined when necessary, are used to train a YOLOv8 segmentation network. To enhance detection accuracy under challenging conditions, we incorporate pre-processing techniques such as relighting adjustments and apply depth-based filtering during post-inference. For object tracking, we integrate the Matching by Segment Anything (MASA) adapter with the BoT-SORT algorithm. Our approach achieves a HOTA score of 80.4%, MOTA of 66.1%, Recall of 74.0%, and Precision of 90.7%, demonstrating effective tracking of sweet peppers without extensive manual effort. This work highlights the potential of foundation models for efficient and accurate object detection and tracking in agricultural settings.

Paperid: 2977, https://arxiv.org/pdf/2410.14138.pdf

Abstract:
Large vision-language models (LVLMs) have witnessed significant progress on visual understanding tasks. However, they often prioritize language knowledge over image information on visual reasoning tasks, incurring performance degradation. To tackle this issue, we first identify the drawbacks of existing solutions (i.e., limited multi-modal reasoning capacities, and insufficient and irrelevant visual descriptions). We then decompose visual reasoning process into two stages: proactive visual perception (i.e., eyesight) and textual reasoning (i.e., wisdom), and introduce a novel visual reasoning framework named ProReason. This framework features decoupled vision-reasoning capabilities and multi-run proactive perception. Briefly, given a multi-modal question, ProReason iterates proactive information collection and reasoning until the answer can be concluded with necessary and sufficient visual descriptions. Notably, the disassociation of capabilities allows seamless integration of existing large language models (LLMs) to compensate for the reasoning deficits of LVLMs. Our extensive experiments demonstrate that ProReason outperforms existing multi-step reasoning frameworks on various benchmarks for both open-source and closed-source models, with the average performance gain reaching 13.2%. Besides, the integration of LLMs allows ProReason to produce high-quality visual reasoning data, which empowers ProReason-distilled models (i.e., ProReason-VL and ProReason-Q3) to achieve superior performance in downstream tasks. Our insights into existing solutions and the decoupled perspective for feasible integration of LLMs illuminate future research on visual reasoning techniques, especially LLM-assisted ones.

Paperid: 2978, https://arxiv.org/pdf/2410.13976.pdf

Abstract:
Large Vision Language Models (LVLMs) such as LLaVA have demonstrated impressive capabilities as general-purpose chatbots that can engage in conversations about a provided input image. However, their responses are influenced by societal biases present in their training datasets, leading to undesirable differences in how the model responds when presented with images depicting people of different demographics. In this work, we propose a novel debiasing framework for LVLMs by directly ablating biased attributes during text generation to avoid generating text related to protected attributes, or even representing them internally. Our method requires no training and a relatively small amount of representative biased outputs (~1000 samples). Our experiments show that not only can we can minimize the propensity of LVLMs to generate text related to protected attributes, but we can even use synthetic data to inform the ablation while retaining captioning performance on real data such as COCO. Furthermore, we find the resulting generations from a debiased LVLM exhibit similar accuracy as a baseline biased model, showing that debiasing effects can be achieved without sacrificing model performance.

Paperid: 2979, https://arxiv.org/pdf/2410.09421.pdf

Abstract:
As large vision-language models (LVLMs) evolve rapidly, the demand for high-quality and diverse data to align these models becomes increasingly crucial. However, the creation of such data with human supervision proves costly and time-intensive. In this paper, we investigate the efficacy of AI feedback to scale supervision for aligning LVLMs. We introduce VLFeedback, the first large-scale vision-language feedback dataset, comprising over 82K multi-modal instructions and comprehensive rationales generated by off-the-shelf models without human annotations. To evaluate the effectiveness of AI feedback for vision-language alignment, we train Silkie, an LVLM fine-tuned via direct preference optimization on VLFeedback. Silkie showcases exceptional performance regarding helpfulness, visual faithfulness, and safety metrics. It outperforms its base model by 6.9\% and 9.5\% in perception and cognition tasks, reduces hallucination issues on MMHal-Bench, and exhibits enhanced resilience against red-teaming attacks. Furthermore, our analysis underscores the advantage of AI feedback, particularly in fostering preference diversity to deliver more comprehensive improvements. Our dataset, training code and models are available at https://vlf-silkie.github.io.

Paperid: 2980, https://arxiv.org/pdf/2410.07112.pdf

Abstract:
Current benchmarks for assessing vision-language models (VLMs) often focus on their perception or problem-solving capabilities and neglect other critical aspects such as fairness, multilinguality, or toxicity. Furthermore, they differ in their evaluation procedures and the scope of the evaluation, making it difficult to compare models. To address these issues, we extend the HELM framework to VLMs to present the Holistic Evaluation of Vision Language Models (VHELM). VHELM aggregates various datasets to cover one or more of the 9 aspects: visual perception, knowledge, reasoning, bias, fairness, multilinguality, robustness, toxicity, and safety. In doing so, we produce a comprehensive, multi-dimensional view of the capabilities of the VLMs across these important factors. In addition, we standardize the standard inference parameters, methods of prompting, and evaluation metrics to enable fair comparisons across models. Our framework is designed to be lightweight and automatic so that evaluation runs are cheap and fast. Our initial run evaluates 22 VLMs on 21 existing datasets to provide a holistic snapshot of the models. We uncover new key findings, such as the fact that efficiency-focused models (e.g., Claude 3 Haiku or Gemini 1.5 Flash) perform significantly worse than their full models (e.g., Claude 3 Opus or Gemini 1.5 Pro) on the bias benchmark but not when evaluated on the other aspects. For transparency, we release the raw model generations and complete results on our website (https://crfm.stanford.edu/helm/vhelm/v2.0.1). VHELM is intended to be a living benchmark, and we hope to continue adding new datasets and models over time.

Paperid: 2981, https://arxiv.org/pdf/2410.05963.pdf

Abstract:
Existing perception models achieve great success by learning from large amounts of labeled data, but they still struggle with open-world scenarios. To alleviate this issue, researchers introduce open-set perception tasks to detect or segment unseen objects in the training set. However, these models require predefined object categories as inputs during inference, which are not available in real-world scenarios. Recently, researchers pose a new and more practical problem, \textit{i.e.}, open-ended object detection, which discovers unseen objects without any object categories as inputs. In this paper, we present VL-SAM, a training-free framework that combines the generalized object recognition model (\textit{i.e.,} Vision-Language Model) with the generalized object localization model (\textit{i.e.,} Segment-Anything Model), to address the open-ended object detection and segmentation task. Without additional training, we connect these two generalized models with attention maps as the prompts. Specifically, we design an attention map generation module by employing head aggregation and a regularized attention flow to aggregate and propagate attention maps across all heads and layers in VLM, yielding high-quality attention maps. Then, we iteratively sample positive and negative points from the attention maps with a prompt generation module and send the sampled points to SAM to segment corresponding objects. Experimental results on the long-tail instance segmentation dataset (LVIS) show that our method surpasses the previous open-ended method on the object detection task and can provide additional instance segmentation masks. Besides, VL-SAM achieves favorable performance on the corner case object detection dataset (CODA), demonstrating the effectiveness of VL-SAM in real-world applications. Moreover, VL-SAM exhibits good model generalization that can incorporate various VLMs and SAMs.

Paperid: 2982, https://arxiv.org/pdf/2409.16278.pdf

Abstract:
Despite extensive research, open-vocabulary segmentation methods still struggle to generalize across diverse domains. To reduce the computational cost of adapting Vision-Language Models (VLMs) while preserving their pre-trained knowledge, most methods freeze the VLMs for mask classification and train only the mask generator. However, our comprehensive analysis reveals a surprising insight: open-vocabulary segmentation is primarily bottlenecked by mask classification, not mask generation. This discovery prompts us to rethink the existing paradigm and explore an alternative approach. Instead of freezing the VLM, we propose to freeze the pre-trained mask generator and focus on optimizing the mask classifier. Building on the observation that VLMs pre-trained on global-pooled image-text features often fail to capture fine-grained semantics necessary for effective mask classification, we propose a novel Fine-grained Semantic Adaptation (FISA) method to address this limitation. FISA enhances the extracted visual features with fine-grained semantic awareness by explicitly integrating this crucial semantic information early in the visual encoding process. As our method strategically optimizes only a small portion of the VLM's parameters, it enjoys the efficiency of adapting to new data distributions while largely preserving the valuable VLM pre-trained knowledge. Extensive ablation studies confirm the superiority of our approach. Notably, FISA achieves new state-of-the-art results across multiple representative benchmarks, improving performance by up to +1.0 PQ and +3.0 mIoU and reduces training costs by nearly 5x compared to previous best methods. Our code and data will be made public.

Paperid: 2983, https://arxiv.org/pdf/2409.06579.pdf

Abstract:
CLIP is one of the most popular foundational models and is heavily used for many vision-language tasks. However, little is known about the inner workings of CLIP. To bridge this gap we propose a study to quantify the interpretability in CLIP like models. We conduct this study on six different CLIP models from OpenAI and OpenCLIP which vary by size, type of pre-training data and patch size. Our approach begins with using the TEXTSPAN algorithm and in-context learning to break down individual attention heads into specific properties. We then evaluate how easily these heads can be interpreted using new metrics which measure property consistency within heads and property disentanglement across heads. Our findings reveal that larger CLIP models are generally more interpretable than their smaller counterparts. To further assist users in understanding the inner workings of CLIP models, we introduce CLIP-InterpreT, a tool designed for interpretability analysis. CLIP-InterpreT offers five types of analyses: property-based nearest neighbor search, per-head topic segmentation, contrastive segmentation, per-head nearest neighbors of an image, and per-head nearest neighbors of text.

Paperid: 2984, https://arxiv.org/pdf/2601.18100.pdf

Abstract:
Long-horizon egocentric video presents significant challenges for visual navigation due to viewpoint drift and the absence of persistent geometric context. Although recent vision-language models perform well on image and short-video reasoning, their spatial reasoning capability in long egocentric sequences remains limited. In this work, we study how explicit spatial signals influence VLM-based video understanding without modifying model architectures or inference procedures. We introduce Sanpo-D, a fine-grained re-annotation of the Google Sanpo dataset, and benchmark multiple VLMs on navigation-oriented spatial queries. To examine input-level inductive bias, we further fuse depth maps with RGB frames and evaluate their impact on spatial reasoning. Our results reveal a trade-off between general-purpose accuracy and spatial specialization, showing that depth-aware and spatially grounded representations can improve performance on safety-critical tasks such as pedestrian and obstruction detection.

Paperid: 2985, https://arxiv.org/pdf/2601.11475.pdf

Abstract:
Vision-Language-Action (VLA) models are emerging as highly effective planning models for end-to-end autonomous driving systems. However, current works mostly rely on imitation learning from sparse trajectory annotations and under-utilize their potential as generative models. We propose Generative Scenario Rollouts (GeRo), a plug-and-play framework for VLA models that jointly performs planning and generation of language-grounded future traffic scenes through an autoregressive rollout strategy. First, a VLA model is trained to encode ego vehicle and agent dynamics into latent tokens under supervision from planning, motion, and language tasks, facilitating text-aligned generation. Next, GeRo performs language-conditioned autoregressive generation. Given multi-view images, a scenario description, and ego-action questions, it generates future latent tokens and textual responses to guide long-horizon rollouts. A rollout-consistency loss stabilizes predictions using ground truth or pseudo-labels, mitigating drift and preserving text-action alignment. This design enables GeRo to perform temporally consistent, language-grounded rollouts that support long-horizon reasoning and multi-agent planning. On Bench2Drive, GeRo improves driving score and success rate by +15.7 and +26.2, respectively. By integrating reinforcement learning with generative rollouts, GeRo achieves state-of-the-art closed-loop and open-loop performance, demonstrating strong zero-shot robustness. These results highlight the promise of generative, language-conditioned reasoning as a foundation for safer and more interpretable end-to-end autonomous driving.

Paperid: 2986, https://arxiv.org/pdf/2601.08151.pdf

Abstract:
Multimodal Large Language Models (MLLMs) have achieved remarkable progress in vision-language understanding, yet how they internally integrate visual and textual information remains poorly understood. To bridge this gap, we perform a systematic layer-wise masking analysis across multiple architectures, revealing how visual-text fusion evolves within MLLMs. The results show that fusion emerges at several specific layers rather than being uniformly distributed across the network, and certain models exhibit a late-stage "review" phenomenon where visual signals are reactivated before output generation. Besides, we further analyze layer-wise attention evolution and observe persistent high-attention noise on irrelevant regions, along with gradually increasing attention on text-aligned areas. Guided by these insights, we introduce a training-free contrastive attention framework that models the transformation between early fusion and final layers to highlight meaningful attention shifts. Extensive experiments across various MLLMs and benchmarks validate our analysis and demonstrate that the proposed approach improves multimodal reasoning performance. Code will be released.

Paperid: 2987, https://arxiv.org/pdf/2601.07359.pdf

Abstract:
Multimodal Large Language Models (MLLMs) have demonstrated strong capabilities across a variety of vision-language tasks. However, their internal reasoning often exhibits a critical inconsistency: although deeper layers may attend to the correct visual regions, final predictions are frequently misled by noisy attention from earlier layers. This results in a disconnect between what the model internally understands and what it ultimately expresses, a phenomenon we describe as seeing it right but saying it wrong. To address this issue, we propose DualPD, a dual-perspective decoding refinement strategy that enhances the visual understanding without any additional training. DualPD consists of two components. (1) The layer-wise attention-guided contrastive logits module captures how the belief in the correct answer evolves by comparing output logits between layers that exhibit the largest attention shift. (2) The head-wise information filtering module suppresses low-contribution attention heads that focus on irrelevant regions, thereby improving attention quality within each layer. Experiments conducted on both the LLaVA and Qwen-VL model families across multiple multimodal benchmarks demonstrate that DualPD consistently improves accuracy without training, confirming its effectiveness and generalizability. The code will be released upon publication.

Paperid: 2988, https://arxiv.org/pdf/2601.07154.pdf

Abstract:
From Vision-Language-Action (VLA) systems to robotics, existing egocentric datasets primarily focus on action recognition tasks, while largely overlooking the inherent role of motion analysis in sports and other fast-movement scenarios. To bridge this gap, we propose a real-time motion focus recognition method that estimates the subject's locomotion intention from any egocentric video. We leverage the foundation model for camera pose estimation and introduce system-level optimizations to enable efficient and scalable inference. Evaluated on a collected egocentric action dataset, our method achieves real-time performance with manageable memory consumption through a sliding batch inference strategy. This work makes motion-centric analysis practical for edge deployment and offers a complementary perspective to existing egocentric studies on sports and fast-movement activities.

Paperid: 2989, https://arxiv.org/pdf/2601.06474.pdf

Abstract:
In autonomous driving, Vision Language Models (VLMs) excel at high-level reasoning , whereas semantic occupancy provides fine-grained details. Despite significant progress in individual fields, there is still no method that can effectively integrate both paradigms. Conventional VLMs struggle with token explosion and limited spatiotemporal reasoning, while semantic occupancy provides a unified, explicit spatial representation but is too dense to integrate efficiently with VLMs. To address these challenges and bridge the gap between VLMs and occupancy, we propose SparseOccVLA, a novel vision-language-action model that unifies scene understanding, occupancy forecasting, and trajectory planning powered by sparse occupancy queries. Starting with a lightweight Sparse Occupancy Encoder, SparseOccVLA generates compact yet highly informative sparse occupancy queries that serve as the single bridge between vision and language. These queries are aligned into the language space and reasoned by the LLM for unified scene understanding and future occupancy forecasting. Furthermore, we introduce an LLM-guided Anchor-Diffusion Planner featuring decoupled anchor scoring and denoising, as well as cross-model trajectory-condition fusion. SparseOccVLA achieves a 7% relative improvement in CIDEr over the state-of-the-art on OmniDrive-nuScenes, a 0.5 increase in mIoU score on Occ3D-nuScenes, and sets state-of-the-art open-loop planning metric on nuScenes benchmark, demonstrating its strong holistic capability.

Paperid: 2990, https://arxiv.org/pdf/2512.22315.pdf

Abstract:
Multimodal Large Language Models (MLLMs) have achieved remarkable progress in vision-language tasks yet remain limited in long video understanding due to the limited context window. Consequently, prevailing approaches tend to rely on uniform frame sampling or static pre-selection, which might overlook critical evidence and unable to correct its initial selection error during its reasoning process. To overcome these limitations, we propose VideoZoomer, a novel agentic framework that enables MLLMs to dynamically control their visual focus during reasoning. Starting from a coarse low-frame-rate overview, VideoZoomer invokes a temporal zoom tool to obtain high-frame-rate clips at autonomously chosen moments, thereby progressively gathering fine-grained evidence in a multi-turn interactive manner. Accordingly, we adopt a two-stage training strategy: a cold-start supervised fine-tuning phase on a curated dataset of distilled exemplar and reflection trajectories, followed by reinforcement learning to further refine the agentic policy. Extensive experiments demonstrate that our 7B model delivers diverse and complex reasoning patterns, yielding strong performance across a broad set of long video understanding and reasoning benchmarks. These emergent capabilities allow it to consistently surpass existing open-source models and even rival proprietary systems on challenging tasks, while achieving superior efficiency under reduced frame budgets.

Paperid: 2991, https://arxiv.org/pdf/2512.21970.pdf

Abstract:
Stereo cameras closely mimic human binocular vision, providing rich spatial cues critical for precise robotic manipulation. Despite their advantage, the adoption of stereo vision in vision-language-action models (VLAs) remains underexplored. In this work, we present StereoVLA, a VLA model that leverages rich geometric cues from stereo vision. We propose a novel Geometric-Semantic Feature Extraction module that utilizes vision foundation models to extract and fuse two key features: 1) geometric features from subtle stereo-view differences for spatial perception; 2) semantic-rich features from the monocular view for instruction following. Additionally, we propose an auxiliary Interaction-Region Depth Estimation task to further enhance spatial perception and accelerate model convergence. Extensive experiments show that our approach outperforms baselines by a large margin in diverse tasks under the stereo setting and demonstrates strong robustness to camera pose variations.

Paperid: 2992, https://arxiv.org/pdf/2512.10691.pdf

Abstract:
Recent advances in vision-language models (VLMs) have improved Chest X-ray (CXR) interpretation in multiple aspects. However, many medical VLMs rely solely on supervised fine-tuning (SFT), which optimizes next-token prediction without evaluating answer quality. In contrast, reinforcement learning (RL) can incorporate task-specific feedback, and its combination with explicit intermediate reasoning ("thinking") has demonstrated substantial gains on verifiable math and coding tasks. To investigate the effects of RL and thinking in a CXR VLM, we perform large-scale SFT on CXR data to build an updated RadVLM based on Qwen3-VL, followed by a cold-start SFT stage that equips the model with basic thinking ability. We then apply Group Relative Policy Optimization (GRPO) with clinically grounded, task-specific rewards for report generation and visual grounding, and run matched RL experiments on both domain-specific and general-domain Qwen3-VL variants, with and without thinking. Across these settings, we find that while strong SFT remains crucial for high base performance, RL provides additional gains on both tasks, whereas explicit thinking does not appear to further improve results. Under a unified evaluation pipeline, the RL-optimized RadVLM models outperform their baseline counterparts and reach state-of-the-art performance on both report generation and grounding, highlighting clinically aligned RL as a powerful complement to SFT for medical VLMs.

Paperid: 2993, https://arxiv.org/pdf/2512.05564.pdf

Abstract:
Recent advances in video generation have shown remarkable potential for constructing world simulators. However, current models still struggle to produce physically consistent results, particularly when handling large-scale or complex dynamics. This limitation arises primarily because existing approaches respond isotropically to physical prompts and neglect the fine-grained alignment between generated content and localized physical cues. To address these challenges, we propose ProPhy, a Progressive Physical Alignment Framework that enables explicit physics-aware conditioning and anisotropic generation. ProPhy employs a two-stage Mixture-of-Physics-Experts (MoPE) mechanism for discriminative physical prior extraction, where Semantic Experts infer semantic-level physical principles from textual descriptions, and Refinement Experts capture token-level physical dynamics. This mechanism allows the model to learn fine-grained, physics-aware video representations that better reflect underlying physical laws. Furthermore, we introduce a physical alignment strategy that transfers the physical reasoning capabilities of vision-language models (VLMs) into the Refinement Experts, facilitating a more accurate representation of dynamic physical phenomena. Extensive experiments on physics-aware video generation benchmarks demonstrate that ProPhy produces more realistic, dynamic, and physically coherent results than existing state-of-the-art methods.

Paperid: 2994, https://arxiv.org/pdf/2511.22294.pdf

Abstract:
Building robust medical machine learning systems requires pretraining strategies that exploit the intrinsic structure present in clinical data. We introduce Multiview Masked Autoencoder (MVMAE), a self-supervised framework that leverages the natural multi-view organization of radiology studies to learn view-invariant and disease-relevant representations. MVMAE combines masked image reconstruction with cross-view alignment, transforming clinical redundancy across projections into a powerful self-supervisory signal. We further extend this approach with MVMAE-V2T, which incorporates radiology reports as an auxiliary text-based learning signal to enhance semantic grounding while preserving fully vision-based inference. Evaluated on a downstream disease classification task on three large-scale public datasets, MIMIC-CXR, CheXpert, and PadChest, MVMAE consistently outperforms supervised and vision-language baselines. Furthermore, MVMAE-V2T provides additional gains, particularly in low-label regimes where structured textual supervision is most beneficial. Together, these results establish the importance of structural and textual supervision as complementary paths toward scalable, clinically grounded medical foundation models.

Paperid: 2995, https://arxiv.org/pdf/2511.18875.pdf

Abstract:
Multimodal large language models (MLLMs) deliver impressive vision-language reasoning but suffer steep inference latency because self-attention scales quadratically with sequence length and thousands of visual tokens contributed by high-resolution images. Naively pruning less-informative visual tokens reduces this burden, yet indiscriminate removal can strip away contextual cues essential for background or fine-grained questions, undermining accuracy. In this paper, we present ParVTS (Parallel Vision Token Scheduling), a training-free scheduling framework that partitions visual tokens into subject and non-subject groups, processes them in parallel to transfer their semantics into question tokens, and discards the non-subject path mid-inference to reduce computation. This scheduling reduces computational complexity, requires no heuristics or additional modules, and is compatible with diverse existing MLLM architectures. Experiments across multiple MLLM backbones show that ParVTS prunes up to 88.9% of visual tokens with minimal performance drop, achieving 1.77x speedup and 70% FLOPs reduction.

Paperid: 2996, https://arxiv.org/pdf/2511.17502.pdf

Abstract:
We introduce RynnVLA-002, a unified Vision-Language-Action (VLA) and world model. The world model leverages action and visual inputs to predict future image states, learning the underlying physics of the environment to refine action generation. Conversely, the VLA model produces subsequent actions from image observations, enhancing visual understanding and supporting the world model's image generation. The unified framework of RynnVLA-002 enables joint learning of environmental dynamics and action planning. Our experiments show that RynnVLA-002 surpasses individual VLA and world models, demonstrating their mutual enhancement. We evaluate RynnVLA-002 in both simulation and real-world robot tasks. RynnVLA-002 achieves 97.4% success rate on the LIBERO simulation benchmark without pretraining, while in real-world LeRobot experiments, its integrated world model boosts the overall success rate by 50%.

Paperid: 2997, https://arxiv.org/pdf/2511.14592.pdf

Abstract:
Vision-Language Models (VLMs) show great promise for autonomous driving, but their suitability for safety-critical scenarios is largely unexplored, raising safety concerns. This issue arises from the lack of comprehensive benchmarks that assess both external environmental risks and in-cabin driving behavior safety simultaneously. To bridge this critical gap, we introduce DSBench, the first comprehensive Driving Safety Benchmark designed to assess a VLM's awareness of various safety risks in a unified manner. DSBench encompasses two major categories: external environmental risks and in-cabin driving behavior safety, divided into 10 key categories and a total of 28 sub-categories. This comprehensive evaluation covers a wide range of scenarios, ensuring a thorough assessment of VLMs' performance in safety-critical contexts. Extensive evaluations across various mainstream open-source and closed-source VLMs reveal significant performance degradation under complex safety-critical situations, highlighting urgent safety concerns. To address this, we constructed a large dataset of 98K instances focused on in-cabin and external safety scenarios, showing that fine-tuning on this dataset significantly enhances the safety performance of existing VLMs and paves the way for advancing autonomous driving technology. The benchmark toolkit, code, and model checkpoints will be publicly accessible.

Paperid: 2998, https://arxiv.org/pdf/2511.11751.pdf

Abstract:
Modern vision-language models (VLMs) deliver impressive predictive accuracy yet offer little insight into 'why' a decision is reached, frequently hallucinating facts, particularly when encountering out-of-distribution data. Neurosymbolic frameworks address this by pairing black-box perception with interpretable symbolic reasoning, but current methods extract their symbols solely from task labels, leaving them weakly grounded in the underlying visual data. In this paper, we introduce a multi-agent system - Concept-RuleNet that reinstates visual grounding while retaining transparent reasoning. Specifically, a multimodal concept generator first mines discriminative visual concepts directly from a representative subset of training images. Next, these visual concepts are utilized to condition symbol discovery, anchoring the generations in real image statistics and mitigating label bias. Subsequently, symbols are composed into executable first-order rules by a large language model reasoner agent - yielding interpretable neurosymbolic rules. Finally, during inference, a vision verifier agent quantifies the degree of presence of each symbol and triggers rule execution in tandem with outputs of black-box neural models, predictions with explicit reasoning pathways. Experiments on five benchmarks, including two challenging medical-imaging tasks and three underrepresented natural-image datasets, show that our system augments state-of-the-art neurosymbolic baselines by an average of 5% while also reducing the occurrence of hallucinated symbols in rules by up to 50%.

Paperid: 2999, https://arxiv.org/pdf/2511.07923.pdf

Abstract:
Accurate segmentation of marine organisms is vital for biodiversity monitoring and ecological assessment, yet existing datasets and models remain largely limited to terrestrial scenes. To bridge this gap, we introduce \textbf{AquaOV255}, the first large-scale and fine-grained underwater segmentation dataset containing 255 categories and over 20K images, covering diverse categories for open-vocabulary (OV) evaluation. Furthermore, we establish the first underwater OV segmentation benchmark, \textbf{UOVSBench}, by integrating AquaOV255 with five additional underwater datasets to enable comprehensive evaluation. Alongside, we present \textbf{Earth2Ocean}, a training-free OV segmentation framework that transfers terrestrial vision--language models (VLMs) to underwater domains without any additional underwater training. Earth2Ocean consists of two core components: a Geometric-guided Visual Mask Generator (\textbf{GMG}) that refines visual features via self-similarity geometric priors for local structure perception, and a Category-visual Semantic Alignment (\textbf{CSA}) module that enhances text embeddings through multimodal large language model reasoning and scene-aware template construction. Extensive experiments on the UOVSBench benchmark demonstrate that Earth2Ocean achieves significant performance improvement on average while maintaining efficient inference.

Paperid: 3000, https://arxiv.org/pdf/2510.23576.pdf

Abstract:
Urban micromobility applications, such as delivery robots, demand reliable navigation across large-scale urban environments while following long-horizon route instructions. This task is particularly challenging due to the dynamic and unstructured nature of real-world city areas, yet most existing navigation methods remain tailored to short-scale and controllable scenarios. Effective urban micromobility requires two complementary levels of navigation skills: low-level capabilities such as point-goal reaching and obstacle avoidance, and high-level capabilities, such as route-visual alignment. To this end, we propose UrbanVLA, a route-conditioned Vision-Language-Action (VLA) framework designed for scalable urban navigation. Our method explicitly aligns noisy route waypoints with visual observations during execution, and subsequently plans trajectories to drive the robot. To enable UrbanVLA to master both levels of navigation, we employ a two-stage training pipeline. The process begins with Supervised Fine-Tuning (SFT) using simulated environments and trajectories parsed from web videos. This is followed by Reinforcement Fine-Tuning (RFT) on a mixture of simulation and real-world data, which enhances the model's safety and adaptability in real-world settings. Experiments demonstrate that UrbanVLA surpasses strong baselines by more than 55% in the SocialNav task on MetaUrban. Furthermore, UrbanVLA achieves reliable real-world navigation, showcasing both scalability to large-scale urban environments and robustness against real-world uncertainties.

Paperid: 3001, https://arxiv.org/pdf/2510.18262.pdf

Abstract:
Large vision-language models (VLMs) have achieved remarkable success in natural scene understanding, yet their application to underwater environments remains largely unexplored. Underwater imagery presents unique challenges including severe light attenuation, color distortion, and suspended particle scattering, while requiring specialized knowledge of marine ecosystems and organism taxonomy. To bridge this gap, we introduce UWBench, a comprehensive benchmark specifically designed for underwater vision-language understanding. UWBench comprises 15,003 high-resolution underwater images captured across diverse aquatic environments, encompassing oceans, coral reefs, and deep-sea habitats. Each image is enriched with human-verified annotations including 15,281 object referring expressions that precisely describe marine organisms and underwater structures, and 124,983 question-answer pairs covering diverse reasoning capabilities from object recognition to ecological relationship understanding. The dataset captures rich variations in visibility, lighting conditions, and water turbidity, providing a realistic testbed for model evaluation. Based on UWBench, we establish three comprehensive benchmarks: detailed image captioning for generating ecologically informed scene descriptions, visual grounding for precise localization of marine organisms, and visual question answering for multimodal reasoning about underwater environments. Extensive experiments on state-of-the-art VLMs demonstrate that underwater understanding remains challenging, with substantial room for improvement. Our benchmark provides essential resources for advancing vision-language research in underwater contexts and supporting applications in marine science, ecological monitoring, and autonomous underwater exploration. Our code and benchmark will be available.

Paperid: 3002, https://arxiv.org/pdf/2510.13804.pdf

Abstract:
We introduce Generative Universal Verifier, a novel concept and plugin designed for next-generation multimodal reasoning in vision-language models and unified multimodal models, providing the fundamental capability of reflection and refinement on visual outcomes during the reasoning and generation process. This work makes three main contributions: (1) We build ViVerBench, a comprehensive benchmark spanning 16 categories of critical tasks for evaluating visual outcomes in multimodal reasoning. Results show that existing VLMs consistently underperform across these tasks, underscoring a substantial gap from human-level capability in reliable visual verification. (2) We design two automated pipelines to construct large-scale visual verification data and train OmniVerifier-7B, the first omni-capable generative verifier trained for universal visual verification and achieves notable gains on ViVerBench(+8.3). Through training, we identify three atomic capabilities in visual verification and demonstrate how they generalize and interact synergistically. (3) We propose OmniVerifier-TTS, a sequential test-time scaling paradigm that leverages the universal verifier to bridge image generation and editing within unified models, enhancing the upper bound of generative ability through iterative fine-grained optimization. Beyond generation, we extend universal verifier to broader world-modeling interleaved reasoning scenarios. Empirically, OmniVerifier-TTS achieves improvements on T2I-ReasonBench(+3.7), and GenEval++(+4.3), outperforming existing parallel test-time scaling methods, such as Best-of-N. By endowing multimodal reasoning with reliable visual verification, OmniVerifier advances both reliable reflection during generation and scalable test-time refinement, marking a step toward more trustworthy and controllable next-generation reasoning systems.

Paperid: 3003, https://arxiv.org/pdf/2510.11012.pdf

Abstract:
Compositional reasoning remains a persistent weakness of modern vision language models (VLMs): they often falter when a task hinges on understanding how multiple objects, attributes, and relations interact within an image. Multiple research works have attempted to improve compositionality performance by creative tricks such as improving prompt structure, chain of thought reasoning, etc. A more recent line of work attempts to impart additional reasoning in VLMs using well-trained Large Language Models (LLMs), which are far superior in linguistic understanding than VLMs to compensate for the limited linguistic prowess of VLMs. However, these approaches are either resource-intensive or do not provide an interpretable reasoning process. In this paper, we present 'COCO-Tree' - a novel approach that augments VLM outputs with carefully designed neurosymbolic concept trees learned from LLMs to improve VLM's linguistic reasoning. COCO-Tree's beam search-inspired reasoning process boosts compositionality performance and provides a rationale behind VLM predictions. Empirical results on four compositionality benchmarks, Winoground, EqBench, ColorSwap, and SugarCrepe, in seven different open-source VLMs with varying sizes, demonstrate that COCO-Tree significantly improves compositional generalization by 5-10% over baselines.

Paperid: 3004, https://arxiv.org/pdf/2510.10973.pdf

Abstract:
The capabilities of Large Vision-Language Models (LVLMs) have reached state-of-the-art on many visual reasoning tasks, including chart reasoning, yet they still falter on out-of-distribution (OOD) data, and degrade further when asked to produce their chain-of-thought (CoT) rationales, limiting explainability. We present Chart-RVR, a general framework that fine-tunes LVLMs to be more robust and explainable for chart reasoning by coupling Group Relative Policy Optimization (GRPO) with automatically verifiable rewards. Our framework comprises of three rewards that maximize: (i) correct chart-type classification, (ii) faithful chart table reconstruction, and (iii) process conformity. Applied to 3-billion-parameter LVLMs, Chart-RVR consistently outperforms standard supervised fine-tuning (SFT) on both in-distribution and out-of-distribution datasets, closing the OOD performance gap while improving rationale fidelity. The resulting models, the Chart-RVR-3B series, achieve state-of-the-art results on six chart-reasoning benchmarks spanning in-domain and OOD settings, surpassing all existing models of comparable size. Beyond accuracy, Chart-RVR yields more interpretable CoT rationales, strengthening trust and reliability - showcasing the power of verifiable rewards with GRPO for training reliable, interpretable chart-reasoning models.

Paperid: 3005, https://arxiv.org/pdf/2510.10181.pdf

Abstract:
Embodied agents face a fundamental limitation: once deployed in real-world environments to perform specific tasks, they are unable to acquire new useful knowledge to enhance task performance. In this paper, we propose a general post-deployment learning framework called Dejavu, which employs an Experience Feedback Network (EFN) and augments the frozen Vision-Language-Action (VLA) policy with retrieved execution memories. EFN automatically identifies contextually successful prior action experiences and conditions action prediction on this retrieved guidance. We adopt reinforcement learning with semantic similarity rewards on EFN to ensure that the predicted actions align with past successful behaviors under current observations. During deployment, EFN continually enriches its memory with new trajectories, enabling the agent to exhibit "learning from experience" despite fixed weights. Experiments across diverse embodied tasks show that EFN significantly improves adaptability, robustness, and success rates over frozen baselines. These results highlight a promising path toward embodied agents that continually refine their behavior after deployment.

Paperid: 3006, https://arxiv.org/pdf/2509.26641.pdf

Abstract:
Unified Multimodal Models (UMMs) have demonstrated remarkable performance in text-to-image generation (T2I) and editing (TI2I), whether instantiated as assembled unified frameworks which couple powerful vision-language model (VLM) with diffusion-based generator, or as naive Unified Multimodal Models with an early fusion of understanding and generation modalities. We contend that in current unified frameworks, the crucial capability of multimodal generative reasoning which encompasses instruction understanding, grounding, and image referring for identity preservation and faithful reconstruction, is intrinsically entangled with high-fidelity synthesis. In this work, we introduce Query-Kontext, a novel approach that bridges the VLM and diffusion model via a multimodal ``kontext'' composed of semantic cues and coarse-grained image conditions encoded from multimodal inputs. This design delegates the complex ability of multimodal generative reasoning to powerful VLM while reserving diffusion model's role for high-quality visual synthesis. To achieve this, we propose a three-stage progressive training strategy. First, we connect the VLM to a lightweight diffusion head via multimodal kontext tokens to unleash the VLM's generative reasoning ability. Second, we scale this head to a large, pre-trained diffusion model to enhance visual detail and realism. Finally, we introduce a low-level image encoder to improve image fidelity and perform instruction tuning on downstream tasks. Furthermore, we build a comprehensive data pipeline integrating real, synthetic, and open-source datasets, covering diverse multimodal reference-to-image scenarios, including image generation, instruction-driven editing, customized generation, and multi-subject composition. Experiments show that our approach matches strong unified baselines and even outperforms task-specific state-of-the-art methods in several cases.

Paperid: 3007, https://arxiv.org/pdf/2509.26208.pdf

Abstract:
In this paper, we deal with the task of text-driven saliency detection in 360-degrees videos. For this, we introduce the TSV360 dataset which includes 16,000 triplets of ERP frames, textual descriptions of salient objects/events in these frames, and the associated ground-truth saliency maps. Following, we extend and adapt a SOTA visual-based approach for 360-degrees video saliency detection, and develop the TSalV360 method that takes into account a user-provided text description of the desired objects and/or events. This method leverages a SOTA vision-language model for data representation and integrates a similarity estimation module and a viewport spatio-temporal cross-attention mechanism, to discover dependencies between the different data modalities. Quantitative and qualitative evaluations using the TSV360 dataset, showed the competitiveness of TSalV360 compared to a SOTA visual-based approach and documented its competency to perform customized text-driven saliency detection in 360-degrees videos.

Paperid: 3008, https://arxiv.org/pdf/2509.25787.pdf

Abstract:
Improving vision-language models (VLMs) in the post-training stage typically relies on supervised fine-tuning or reinforcement learning, methods that necessitate costly, human-annotated data. While self-supervised techniques such as self-consistency have proven effective for enhancing reasoning capabilities, their application to perceptual domains such as image quality assessment (IQA) remains largely unexplored. In this work, we introduce EvoQuality, a novel framework that enables a VLM to autonomously refine its quality perception capabilities without any ground-truth labels. EvoQuality adapts the principle of self-consistency to the ranking-based nature of IQA. It generates pseudo-labels by performing pairwise majority voting on the VLM's own outputs to establish a consensus on relative quality. These pseudo-rankings are then formulated into a fidelity reward that guides the model's iterative evolution through group relative policy optimization (GRPO). By iteratively leveraging its own predictions, EvoQuality progressively refines the VLM's perceptual capability. Extensive experiments show that EvoQuality boosts the base VLM's zero-shot performance by 31.8\% on PLCC across diverse IQA benchmarks. Remarkably, despite being entirely self-supervised, EvoQuality achieves performance that is competitive with, or even surpasses, state-of-the-art supervised VLM-based IQA models, outperforming these models on 5 out of 7 IQA benchmarks.

Paperid: 3009, https://arxiv.org/pdf/2509.24524.pdf

Abstract:
Vision-Language-Action (VLA) models have achieved notable success but often struggle with limited generalizations. To address this, integrating generalized Vision-Language Models (VLMs) as assistants to VLAs has emerged as a popular solution. However, current approaches often combine these models in rigid, sequential structures: using VLMs primarily for high-level scene understanding and task planning, and VLAs merely as executors of lower-level actions, leading to ineffective collaboration and poor grounding challenges. In this paper, we propose an embodied agent framework, PhysiAgent, tailored to operate effectively in physical environments. By incorporating monitor, memory, self-reflection mechanisms, and lightweight off-the-shelf toolboxes, PhysiAgent offers an autonomous scaffolding framework to prompt VLMs to organize different components based on real-time proficiency feedback from VLAs to maximally exploit VLAs' capabilities. Experimental results demonstrate significant improvements in task-solving performance on complex real-world robotic tasks, showcasing effective self-regulation of VLMs, coherent tool collaboration, and adaptive evolution of the framework during execution. PhysiAgent makes practical and pioneering efforts to integrate VLMs and VLAs, effectively grounding embodied agent frameworks in real-world settings.

Paperid: 3010, https://arxiv.org/pdf/2509.19973.pdf

Abstract:
Human vision is capable of transforming two-dimensional observations into an egocentric three-dimensional scene understanding, which underpins the ability to translate complex scenes and exhibit adaptive behaviors. This capability, however, remains lacking in current autonomous driving systems, where mainstream approaches primarily rely on depth-based 3D reconstruction rather than true scene understanding. To address this limitation, we propose a novel human-like framework called OmniScene. First, we introduce the OmniScene Vision-Language Model (OmniVLM), a vision-language framework that integrates multi-view and temporal perception for holistic 4D scene understanding. Then, harnessing a teacher-student OmniVLM architecture and knowledge distillation, we embed textual representations into 3D instance features for semantic supervision, enriching feature learning, and explicitly capturing human-like attentional semantics. These feature representations are further aligned with human driving behaviors, forming a more human-like perception-understanding-action architecture. In addition, we propose a Hierarchical Fusion Strategy (HFS) to address imbalances in modality contributions during multimodal integration. Our approach adaptively calibrates the relative significance of geometric and semantic features at multiple abstraction levels, enabling the synergistic use of complementary cues from visual and textual modalities. This learnable dynamic fusion enables a more nuanced and effective exploitation of heterogeneous information. We evaluate OmniScene comprehensively on the nuScenes dataset, benchmarking it against over ten state-of-the-art models across various tasks. Our approach consistently achieves superior results, establishing new benchmarks in perception, prediction, planning, and visual question answering.

Paperid: 3011, https://arxiv.org/pdf/2508.10416.pdf

Abstract:
Existing vision-and-language navigation models often deviate from the correct trajectory when executing instructions. However, these models lack effective error correction capability, hindering their recovery from errors. To address this challenge, we propose Self-correction Flywheel, a novel post-training paradigm. Instead of considering the model's error trajectories on the training set as a drawback, our paradigm emphasizes their significance as a valuable data source. We have developed a method to identify deviations in these error trajectories and devised innovative techniques to automatically generate self-correction data for perception and action. These self-correction data serve as fuel to power the model's continued training. The brilliance of our paradigm is revealed when we re-evaluate the model on the training set, uncovering new error trajectories. At this time, the self-correction flywheel begins to spin. Through multiple flywheel iterations, we progressively enhance our monocular RGB-based VLA navigation model CorrectNav. Experiments on R2R-CE and RxR-CE benchmarks show CorrectNav achieves new state-of-the-art success rates of 65.1% and 69.3%, surpassing prior best VLA navigation models by 8.2% and 16.4%. Real robot tests in various indoor and outdoor environments demonstrate \method's superior capability of error correction, dynamic obstacle avoidance, and long instruction following.

Paperid: 3012, https://arxiv.org/pdf/2507.05427.pdf

Abstract:
The ability to segment objects based on open-ended language prompts remains a critical challenge, requiring models to ground textual semantics into precise spatial masks while handling diverse and unseen categories. We present OpenWorldSAM, a framework that extends the prompt-driven Segment Anything Model v2 (SAM2) to open-vocabulary scenarios by integrating multi-modal embeddings extracted from a lightweight vision-language model (VLM). Our approach is guided by four key principles: i) Unified prompting: OpenWorldSAM supports a diverse range of prompts, including category-level and sentence-level language descriptions, providing a flexible interface for various segmentation tasks. ii) Efficiency: By freezing the pre-trained components of SAM2 and the VLM, we train only 4.5 million parameters on the COCO-stuff dataset, achieving remarkable resource efficiency. iii) Instance Awareness: We enhance the model's spatial understanding through novel positional tie-breaker embeddings and cross-attention layers, enabling effective segmentation of multiple instances. iv) Generalization: OpenWorldSAM exhibits strong zero-shot capabilities, generalizing well on unseen categories and an open vocabulary of concepts without additional training. Extensive experiments demonstrate that OpenWorldSAM achieves state-of-the-art performance in open-vocabulary semantic, instance, and panoptic segmentation across multiple benchmarks, including ADE20k, PASCAL, ScanNet, and SUN-RGBD.

Paperid: 3013, https://arxiv.org/pdf/2506.14009.pdf

Abstract:
Autonomous drones capable of interpreting and executing high-level language instructions in unstructured environments remain a long-standing goal. Yet existing approaches are constrained by their dependence on hand-crafted skills, extensive parameter tuning, or computationally intensive models unsuitable for onboard use. We introduce GRaD-Nav++, a lightweight Vision-Language-Action (VLA) framework that runs fully onboard and follows natural-language commands in real time. Our policy is trained in a photorealistic 3D Gaussian Splatting (3DGS) simulator via Differentiable Reinforcement Learning (DiffRL), enabling efficient learning of low-level control from visual and linguistic inputs. At its core is a Mixture-of-Experts (MoE) action head, which adaptively routes computation to improve generalization while mitigating forgetting. In multi-task generalization experiments, GRaD-Nav++ achieves a success rate of 83% on trained tasks and 75% on unseen tasks in simulation. When deployed on real hardware, it attains 67% success on trained tasks and 50% on unseen ones. In multi-environment adaptation experiments, GRaD-Nav++ achieves an average success rate of 81% across diverse simulated environments and 67% across varied real-world settings. These results establish a new benchmark for fully onboard Vision-Language-Action (VLA) flight and demonstrate that compact, efficient models can enable reliable, language-guided navigation without relying on external infrastructure.

Paperid: 3014, https://arxiv.org/pdf/2506.13456.pdf

Abstract:
Diffusion Policy has demonstrated strong visuomotor modeling capabilities, but its high computational cost renders it impractical for real-time robotic control. Despite huge redundancy across repetitive denoising steps, existing diffusion acceleration techniques fail to generalize to Diffusion Policy due to fundamental architectural and data divergences. In this paper, we propose Block-wise Adaptive Caching(BAC), a method to accelerate Diffusion Policy by caching intermediate action features. BAC achieves lossless action generation acceleration by adaptively updating and reusing cached features at the block level, based on a key observation that feature similarities vary non-uniformly across timesteps and locks. To operationalize this insight, we first propose the Adaptive Caching Scheduler, designed to identify optimal update timesteps by maximizing the global feature similarities between cached and skipped features. However, applying this scheduler for each block leads to signiffcant error surges due to the inter-block propagation of caching errors, particularly within Feed-Forward Network (FFN) blocks. To mitigate this issue, we develop the Bubbling Union Algorithm, which truncates these errors by updating the upstream blocks with signiffcant caching errors before downstream FFNs. As a training-free plugin, BAC is readily integrable with existing transformer-based Diffusion Policy and vision-language-action models. Extensive experiments on multiple robotic benchmarks demonstrate that BAC achieves up to 3x inference speedup for free.

Paperid: 3015, https://arxiv.org/pdf/2506.09473.pdf

Abstract:
In-context learning (ICL), a predominant trend in instruction learning, aims at enhancing the performance of large language models by providing clear task guidance and examples, improving their capability in task understanding and execution. This paper investigates ICL on Large Vision-Language Models (LVLMs) and explores the policies of multi-modal demonstration selection. Existing research efforts in ICL face significant challenges: First, they rely on pre-defined demonstrations or heuristic selecting strategies based on human intuition, which are usually inadequate for covering diverse task requirements, leading to sub-optimal solutions; Second, individually selecting each demonstration fails in modeling the interactions between them, resulting in information redundancy. Unlike these prevailing efforts, we propose a new exploration-exploitation reinforcement learning framework, which explores policies to fuse multi-modal information and adaptively select adequate demonstrations as an integrated whole. The framework allows LVLMs to optimize themselves by continually refining their demonstrations through self-exploration, enabling the ability to autonomously identify and generate the most effective selection policies for in-context learning. Experimental results verify the superior performance of our approach on four Visual Question-Answering (VQA) datasets, demonstrating its effectiveness in enhancing the generalization capability of few-shot LVLMs.

Paperid: 3016, https://arxiv.org/pdf/2505.23977.pdf

Abstract:
Vision language models (VLMs) are expected to perform effective multimodal reasoning and make logically coherent decisions, which is critical to tasks such as diagram understanding and spatial problem solving. However, current VLM reasoning lacks large-scale and well-structured training datasets. To bridge this gap, we propose VisualSphinx, a first-of-its-kind large-scale synthetic visual logical reasoning training data. To tackle the challenge of image synthesis with grounding answers, we propose a rule-to-image synthesis pipeline, which extracts and expands puzzle rules from seed questions and generates the code of grounding synthesis image synthesis for puzzle sample assembly. Experiments demonstrate that VLM trained using GRPO on VisualSphinx benefit from logical coherence and readability of our dataset and exhibit improved performance on logical reasoning tasks. The enhanced reasoning capabilities developed from VisualSphinx also benefit other reasoning tasks such as algebraic reasoning, arithmetic reasoning and geometry reasoning.

Paperid: 3017, https://arxiv.org/pdf/2505.21089.pdf

Abstract:
Large vision-language models (VLMs) have made great achievements in Earth vision. However, complex disaster scenes with diverse disaster types, geographic regions, and satellite sensors have posed new challenges for VLM applications. To fill this gap, we curate a remote sensing vision-language dataset (DisasterM3) for global-scale disaster assessment and response. DisasterM3 includes 26,988 bi-temporal satellite images and 123k instruction pairs across 5 continents, with three characteristics: 1) Multi-hazard: DisasterM3 involves 36 historical disaster events with significant impacts, which are categorized into 10 common natural and man-made disasters. 2)Multi-sensor: Extreme weather during disasters often hinders optical sensor imaging, making it necessary to combine Synthetic Aperture Radar (SAR) imagery for post-disaster scenes. 3) Multi-task: Based on real-world scenarios, DisasterM3 includes 9 disaster-related visual perception and reasoning tasks, harnessing the full potential of VLM's reasoning ability with progressing from disaster-bearing body recognition to structural damage assessment and object relational reasoning, culminating in the generation of long-form disaster reports. We extensively evaluated 14 generic and remote sensing VLMs on our benchmark, revealing that state-of-the-art models struggle with the disaster tasks, largely due to the lack of a disaster-specific corpus, cross-sensor gap, and damage object counting insensitivity. Focusing on these issues, we fine-tune four VLMs using our dataset and achieve stable improvements across all tasks, with robust cross-sensor and cross-disaster generalization capabilities.

Paperid: 3018, https://arxiv.org/pdf/2505.16793.pdf

Abstract:
Earth observation foundation models have shown strong generalization across multiple Earth observation tasks, but their robustness under real-world perturbations remains underexplored. To bridge this gap, we introduce REOBench, the first comprehensive benchmark for evaluating the robustness of Earth observation foundation models across six tasks and twelve types of image corruptions, including both appearance-based and geometric perturbations. To ensure realistic and fine-grained evaluation, our benchmark focuses on high-resolution optical remote sensing images, which are widely used in critical applications such as urban planning and disaster response. We conduct a systematic evaluation of a broad range of models trained using masked image modeling, contrastive learning, and vision-language pre-training paradigms. Our results reveal that (1) existing Earth observation foundation models experience significant performance degradation when exposed to input corruptions. (2) The severity of degradation varies across tasks, model architectures, backbone sizes, and types of corruption, with performance drop varying from less than 1% to over 20%. (3) Vision-language models show enhanced robustness, particularly in multimodal tasks. REOBench underscores the vulnerability of current Earth observation foundation models to real-world corruptions and provides actionable insights for developing more robust and reliable models.

Paperid: 3019, https://arxiv.org/pdf/2505.12513.pdf

Abstract:
Global tree species mapping using remote sensing data is vital for biodiversity monitoring, forest management, and ecological research. However, progress in this field has been constrained by the scarcity of large-scale, labeled datasets. To address this, we introduce GlobalGeoTree, a comprehensive global dataset for tree species classification. GlobalGeoTree comprises 6.3 million geolocated tree occurrences, spanning 275 families, 2,734 genera, and 21,001 species across the hierarchical taxonomic levels. Each sample is paired with Sentinel-2 image time series and 27 auxiliary environmental variables, encompassing bioclimatic, geographic, and soil data. The dataset is partitioned into GlobalGeoTree-6M for model pretraining and curated evaluation subsets, primarily GlobalGeoTree-10kEval for zero-shot and few-shot benchmarking. To demonstrate the utility of the dataset, we introduce a baseline model, GeoTreeCLIP, which leverages paired remote sensing data and taxonomic text labels within a vision-language framework pretrained on GlobalGeoTree-6M. Experimental results show that GeoTreeCLIP achieves substantial improvements in zero- and few-shot classification on GlobalGeoTree-10kEval over existing advanced models. By making the dataset, models, and code publicly available, we aim to establish a benchmark to advance tree species classification and foster innovation in biodiversity research and ecological applications.

Paperid: 3020, https://arxiv.org/pdf/2505.08189.pdf

Abstract:
Although Reinforcement Learning (RL) agents are effective in well-defined environments, they often struggle to generalize their learned policies to dynamic settings due to their reliance on trial-and-error interactions. Recent work has explored applying Large Language Models (LLMs) or Vision Language Models (VLMs) to boost the generalization of RL agents through policy optimization guidance or prior knowledge. However, these approaches often lack seamless coordination between the RL agent and the foundation model, leading to unreasonable decision-making in unfamiliar environments and efficiency bottlenecks. Making full use of the inferential capabilities of foundation models and the rapid response capabilities of RL agents and enhancing the interaction between the two to form a dual system is still a lingering scientific question. To address this problem, we draw inspiration from Kahneman's theory of fast thinking (System 1) and slow thinking (System 2), demonstrating that balancing intuition and deep reasoning can achieve nimble decision-making in a complex world. In this study, we propose a Dual-System Adaptive Decision Framework (DSADF), integrating two complementary modules: System 1, comprising an RL agent and a memory space for fast and intuitive decision making, and System 2, driven by a VLM for deep and analytical reasoning. DSADF facilitates efficient and adaptive decision-making by combining the strengths of both systems. The empirical study in the video game environment: Crafter and Housekeep demonstrates the effectiveness of our proposed method, showing significant improvements in decision abilities for both unseen and known tasks.

Paperid: 3021, https://arxiv.org/pdf/2505.05635.pdf

Abstract:
Open-vocabulary recognition remains a challenging problem in computer vision, as it requires identifying objects from an unbounded set of categories. This is particularly relevant in nature, where new species are discovered every year. In this work, we focus on open-vocabulary bird species recognition, where the goal is to classify species based on their descriptions without being constrained to a predefined set of taxonomic categories. Traditional benchmarks like CUB-200-2011 and Birdsnap have been evaluated in a closed-vocabulary paradigm, limiting their applicability to real-world scenarios where novel species continually emerge. We show that the performance of current systems when evaluated under settings closely aligned with open-vocabulary drops by a huge margin. To address this gap, we propose a scalable framework integrating structured textual knowledge from Wikipedia articles of 11,202 bird species distilled via GPT-4o into concise, discriminative summaries. We propose Visual Re-ranking Retrieval-Augmented Generation(VR-RAG), a novel, retrieval-augmented generation framework that uses visual similarities to rerank the top m candidates retrieved by a set of multimodal vision language encoders. This allows for the recognition of unseen taxa. Extensive experiments across five established classification benchmarks show that our approach is highly effective. By integrating VR-RAG, we improve the average performance of state-of-the-art Large Multi-Modal Model QWEN2.5-VL by 15.4% across five benchmarks. Our approach outperforms conventional VLM-based approaches, which struggle with unseen species. By bridging the gap between encyclopedic knowledge and visual recognition, our work advances open-vocabulary recognition, offering a flexible, scalable solution for biodiversity monitoring and ecological research.

Paperid: 3022, https://arxiv.org/pdf/2505.03233.pdf

Abstract:
Embodied foundation models are gaining increasing attention for their zero-shot generalization, scalability, and adaptability to new tasks through few-shot post-training. However, existing models rely heavily on real-world data, which is costly and labor-intensive to collect. Synthetic data offers a cost-effective alternative, yet its potential remains largely underexplored. To bridge this gap, we explore the feasibility of training Vision-Language-Action models entirely with large-scale synthetic action data. We curate SynGrasp-1B, a billion-frame robotic grasping dataset generated in simulation with photorealistic rendering and extensive domain randomization. Building on this, we present GraspVLA, a VLA model pretrained on large-scale synthetic action data as a foundational model for grasping tasks. GraspVLA integrates autoregressive perception tasks and flow-matching-based action generation into a unified Chain-of-Thought process, enabling joint training on synthetic action data and Internet semantics data. This design helps mitigate sim-to-real gaps and facilitates the transfer of learned actions to a broader range of Internet-covered objects, achieving open-vocabulary generalization in grasping. Extensive evaluations across real-world and simulation benchmarks demonstrate GraspVLA's advanced zero-shot generalizability and few-shot adaptability to specific human preferences. We will release SynGrasp-1B dataset and pre-trained weights to benefit the community.

Paperid: 3023, https://arxiv.org/pdf/2504.10854.pdf

Abstract:
Large Vision Language Models (LVLMs) have been widely adopted to guide vision foundation models in performing reasoning segmentation tasks, achieving impressive performance. However, the substantial computational overhead associated with LVLMs presents a new challenge. The primary source of this computational cost arises from processing hundreds of image tokens. Therefore, an effective strategy to mitigate such overhead is to reduce the number of image tokens, a process known as image token pruning. Previous studies on image token pruning for LVLMs have primarily focused on high level visual understanding tasks, such as visual question answering and image captioning. In contrast, guiding vision foundation models to generate accurate visual masks based on textual queries demands precise semantic and spatial reasoning capabilities. Consequently, pruning methods must carefully control individual image tokens throughout the LVLM reasoning process. Our empirical analysis reveals that existing methods struggle to adequately balance reductions in computational overhead with the necessity to maintain high segmentation accuracy. In this work, we propose LVLM_CSP, a novel training free visual token pruning method specifically designed for LVLM based reasoning segmentation tasks. LVLM_CSP consists of three stages: clustering, scattering, and pruning. Initially, the LVLM performs coarse-grained visual reasoning using a subset of selected image tokens. Next, fine grained reasoning is conducted, and finally, most visual tokens are pruned in the last stage. Extensive experiments demonstrate that LVLM_CSP achieves a 65% reduction in image token inference FLOPs with virtually no accuracy degradation, and a 70% reduction with only a minor 1% drop in accuracy on the 7B LVLM.

Paperid: 3024, https://arxiv.org/pdf/2503.06260.pdf

Abstract:
Aligning large vision-language models (LVLMs) with human preferences is challenging due to the scarcity of fine-grained, high-quality, and multimodal preference data without human annotations. Existing methods relying on direct distillation often struggle with low-confidence data, leading to suboptimal performance. To address this, we propose CAREVL, a novel method for preference reward modeling by reliably using both high- and low-confidence data. First, a cluster of auxiliary expert models (textual reward models) innovatively leverages image captions as weak supervision signals to filter high-confidence data. The high-confidence data are then used to fine-tune the LVLM. Second, low-confidence data are used to generate diverse preference samples using the fine-tuned LVLM. These samples are then scored and selected to construct reliable chosen-rejected pairs for further training. CAREVL achieves performance improvements over traditional distillation-based methods on VL-RewardBench and MLLM-as-a-Judge benchmark, demonstrating its effectiveness. The code will be released soon.

Paperid: 3025, https://arxiv.org/pdf/2502.18042.pdf

Abstract:
Human drivers adeptly navigate complex scenarios by utilizing rich attentional semantics, but the current autonomous systems struggle to replicate this ability, as they often lose critical semantic information when converting 2D observations into 3D space. In this sense, it hinders their effective deployment in dynamic and complex environments. Leveraging the superior scene understanding and reasoning abilities of Vision-Language Models (VLMs), we propose VLM-E2E, a novel framework that uses the VLMs to enhance training by providing attentional cues. Our method integrates textual representations into Bird's-Eye-View (BEV) features for semantic supervision, which enables the model to learn richer feature representations that explicitly capture the driver's attentional semantics. By focusing on attentional semantics, VLM-E2E better aligns with human-like driving behavior, which is critical for navigating dynamic and complex environments. Furthermore, we introduce a BEV-Text learnable weighted fusion strategy to address the issue of modality importance imbalance in fusing multimodal information. This approach dynamically balances the contributions of BEV and text features, ensuring that the complementary information from visual and textual modalities is effectively utilized. By explicitly addressing the imbalance in multimodal fusion, our method facilitates a more holistic and robust representation of driving environments. We evaluate VLM-E2E on the nuScenes dataset and achieve significant improvements in perception, prediction, and planning over the baseline end-to-end model, showcasing the effectiveness of our attention-enhanced BEV representation in enabling more accurate and reliable autonomous driving tasks.

Paperid: 3026, https://arxiv.org/pdf/2502.12095.pdf

Abstract:
This paper explores the possibility of learning custom tokens for representing new concepts in Vision-Language Models (VLMs). Our aim is to learn tokens that can be effective for both discriminative and generative tasks while composing well with words to form new input queries. The targeted concept is specified in terms of a small set of images and a parent concept described using text. We operate on CLIP text features and propose to use a combination of a textual inversion loss and a classification loss to ensure that text features of the learned token are aligned with image features of the concept in the CLIP embedding space. We restrict the learned token to a low-dimensional subspace spanned by tokens for attributes that are appropriate for the given super-class. These modifications improve the quality of compositions of the learned token with natural language for generating new scenes. Further, we show that learned custom tokens can be used to form queries for text-to-image retrieval task, and also have the important benefit that composite queries can be visualized to ensure that the desired concept is faithfully encoded. Based on this, we introduce the method of Generation Aided Image Retrieval, where the query is modified at inference time to better suit the search intent. On the DeepFashion2 dataset, our method improves Mean Reciprocal Retrieval (MRR) over relevant baselines by 7%.

Paperid: 3027, https://arxiv.org/pdf/2502.03333.pdf

Abstract:
The widespread use of chest X-rays (CXRs), coupled with a shortage of radiologists, has driven growing interest in automated CXR analysis and AI-assisted reporting. While existing vision-language models (VLMs) show promise in specific tasks such as report generation or abnormality detection, they often lack support for interactive diagnostic capabilities. In this work we present RadVLM, a compact, multitask conversational foundation model designed for CXR interpretation. To this end, we curate a large-scale instruction dataset comprising over 1 million image-instruction pairs containing both single-turn tasks -- such as report generation, abnormality classification, and visual grounding -- and multi-turn, multi-task conversational interactions. After fine-tuning RadVLM on this instruction dataset, we evaluate it across different tasks along with re-implemented baseline VLMs. Our results show that RadVLM achieves state-of-the-art performance in conversational capabilities and visual grounding while remaining competitive in other radiology tasks. Ablation studies further highlight the benefit of joint training across multiple tasks, particularly for scenarios with limited annotated data. Together, these findings highlight the potential of RadVLM as a clinically relevant AI assistant, providing structured CXR interpretation and conversational capabilities to support more effective and accessible diagnostic workflows.

Paperid: 3028, https://arxiv.org/pdf/2501.03841.pdf

Abstract:
The development of general robotic systems capable of manipulating in unstructured environments is a significant challenge. While Vision-Language Models(VLM) excel in high-level commonsense reasoning, they lack the fine-grained 3D spatial understanding required for precise manipulation tasks. Fine-tuning VLM on robotic datasets to create Vision-Language-Action Models(VLA) is a potential solution, but it is hindered by high data collection costs and generalization issues. To address these challenges, we propose a novel object-centric representation that bridges the gap between VLM's high-level reasoning and the low-level precision required for manipulation. Our key insight is that an object's canonical space, defined by its functional affordances, provides a structured and semantically meaningful way to describe interaction primitives, such as points and directions. These primitives act as a bridge, translating VLM's commonsense reasoning into actionable 3D spatial constraints. In this context, we introduce a dual closed-loop, open-vocabulary robotic manipulation system: one loop for high-level planning through primitive resampling, interaction rendering and VLM checking, and another for low-level execution via 6D pose tracking. This design ensures robust, real-time control without requiring VLM fine-tuning. Extensive experiments demonstrate strong zero-shot generalization across diverse robotic manipulation tasks, highlighting the potential of this approach for automating large-scale simulation data generation.

Paperid: 3029, https://arxiv.org/pdf/2501.00877.pdf

Abstract:
Open-vocabulary segmentation aims to identify and segment specific regions and objects based on text-based descriptions. A common solution is to leverage powerful vision-language models (VLMs), such as CLIP, to bridge the gap between vision and text information. However, VLMs are typically pretrained for image-level vision-text alignment, focusing on global semantic features. In contrast, segmentation tasks require fine-grained pixel-level alignment and detailed category boundary information, which VLMs alone cannot provide. As a result, information extracted directly from VLMs can't meet the requirements of segmentation tasks. To address this limitation, we propose FGAseg, a model designed for fine-grained pixel-text alignment and category boundary supplementation. The core of FGAseg is a Pixel-Level Alignment module that employs a cross-modal attention mechanism and a text-pixel alignment loss to refine the coarse-grained alignment from CLIP, achieving finer-grained pixel-text semantic alignment. Additionally, to enrich category boundary information, we introduce the alignment matrices as optimizable pseudo-masks during forward propagation and propose Category Information Supplementation module. These pseudo-masks, derived from cosine and convolutional similarity, provide essential global and local boundary information between different categories. By combining these two strategies, FGAseg effectively enhances pixel-level alignment and category boundary information, addressing key challenges in open-vocabulary segmentation. Extensive experiments demonstrate that FGAseg outperforms existing methods on open-vocabulary semantic segmentation benchmarks.

Paperid: 3030, https://arxiv.org/pdf/2411.18000.pdf

Abstract:
Despite inheriting security measures from underlying language models, Vision-Language Models (VLMs) may still be vulnerable to safety alignment issues. Through empirical analysis, we uncover two critical findings: scenario-matched images can significantly amplify harmful outputs, and contrary to common assumptions in gradient-based attacks, minimal loss values do not guarantee optimal attack effectiveness. Building on these insights, we introduce MLAI (Multi-Loss Adversarial Images), a novel jailbreak framework that leverages scenario-aware image generation for semantic alignment, exploits flat minima theory for robust adversarial image selection, and employs multi-image collaborative attacks for enhanced effectiveness. Extensive experiments demonstrate MLAI's significant impact, achieving attack success rates of 77.75% on MiniGPT-4 and 82.80% on LLaVA-2, substantially outperforming existing methods by margins of 34.37% and 12.77% respectively. Furthermore, MLAI shows considerable transferability to commercial black-box VLMs, achieving up to 60.11% success rate. Our work reveals fundamental visual vulnerabilities in current VLMs safety mechanisms and underscores the need for stronger defenses. Warning: This paper contains potentially harmful example text.

Paperid: 3031, https://arxiv.org/pdf/2411.16778.pdf

Abstract:
Medical Visual Question Answering (Med-VQA) combines computer vision and natural language processing to automatically answer clinical inquiries about medical images. However, current Med-VQA datasets exhibit two significant limitations: (1) they often lack visual and textual explanations for answers, hindering comprehension for patients and junior doctors; (2) they typically offer a narrow range of question formats, inadequately reflecting the diverse requirements in practical scenarios. These limitations pose significant challenges to the development of a reliable and user-friendly Med-VQA system. To address these challenges, we introduce a large-scale, Groundable, and Explainable Medical VQA benchmark for chest X-ray diagnosis (GEMeX), featuring several innovative components: (1) a multi-modal explainability mechanism that offers detailed visual and textual explanations for each question-answer pair, thereby enhancing answer comprehensibility; (2) four question types, open-ended, closed-ended, single-choice, and multiple-choice, to better reflect practical needs. With 151,025 images and 1,605,575 questions, GEMeX is the currently largest chest X-ray VQA dataset. Evaluation of 12 representative large vision language models (LVLMs) on GEMeX reveals suboptimal performance, underscoring the dataset's complexity. Meanwhile, we propose a strong model by fine-tuning an existing LVLM on the GEMeX training set. The substantial performance improvement showcases the dataset's effectiveness. The benchmark is available at https://www.med-vqa.com/GEMeX.

Paperid: 3032, https://arxiv.org/pdf/2410.16162.pdf

Abstract:
Vision language models (VLMs) have demonstrated impressive performance across a wide range of downstream tasks. However, their proficiency in spatial reasoning remains limited, despite its crucial role in tasks involving navigation and interaction with physical environments. Specifically, most of these tasks rely on the core spatial reasoning capabilities in two-dimensional (2D) environments, and our evaluation reveals that state-of-the-art VLMs frequently generate implausible and incorrect responses to composite spatial reasoning problems, including simple pathfinding tasks that humans can solve effortlessly at a glance. To address this, we explore an effective approach to enhance 2D spatial reasoning within VLMs by training the model solely on basic spatial capabilities. We begin by disentangling the key components of 2D spatial reasoning: direction comprehension, distance estimation, and localization. Our central hypothesis is that mastering these basic spatial capabilities can significantly enhance a model's performance on composite spatial tasks requiring advanced spatial understanding and combinatorial problem-solving, with generalized improvements in real-world visual-spatial tasks. To investigate this hypothesis, we introduce Sparkle: a framework that uses synthetic data generation to provide targeted supervision for vision language models (VLMs) in three basic spatial capabilities, creating an instruction dataset for each capability. Our experiments demonstrate that VLMs fine-tuned with Sparkle achieve significant performance gains, not only in the basic tasks themselves but also in generalizing to composite and out-of-distribution real-world spatial reasoning tasks. These findings offer insights into systematic strategies for improving VLMs' spatial reasoning capabilities.

Paperid: 3033, https://arxiv.org/pdf/2409.14785.pdf

Abstract:
Natural Language Explanation (NLE) aims to elucidate the decision-making process by providing detailed, human-friendly explanations in natural language. It helps demystify the decision-making processes of large vision-language models (LVLMs) through the use of language models. While existing methods for creating a Vision Question-Answering with Natural Language Explanation (VQA-NLE) datasets can provide explanations, they heavily rely on human annotations that are time-consuming and costly. In this study, we propose a novel approach that leverages LVLMs to efficiently generate high-quality synthetic VQA-NLE datasets. By evaluating our synthetic data, we showcase how advanced prompting techniques can lead to the production of high-quality VQA-NLE data. Our findings indicate that this proposed method achieves up to 20x faster than human annotation, with only a minimal decrease in qualitative metrics, achieving robust quality that is nearly equivalent to human-annotated data. Furthermore, we show that incorporating visual prompts significantly enhances the relevance of text generation. Our study paves the way for a more efficient and robust automated generation of multi-modal NLE data, offering a promising solution to the problem.

Paperid: 3034, https://arxiv.org/pdf/2409.12514.pdf

Abstract:
Vision-Language-Action (VLA) models have shown remarkable potential in visuomotor control and instruction comprehension through end-to-end learning processes. However, current VLA models face significant challenges: they are slow during inference and require extensive pre-training on large amounts of robotic data, making real-world deployment difficult. In this paper, we introduce a new family of compact vision-language-action models, called TinyVLA, which offers two key advantages over existing VLA models: (1) faster inference speeds, and (2) improved data efficiency, eliminating the need for pre-training stage. Our framework incorporates two essential components to build TinyVLA: (1) initializing the policy backbone with robust, high-speed multimodal models, and (2) integrating a diffusion policy decoder during fine-tuning to enable precise robot actions. We conducted extensive evaluations of TinyVLA in both simulation and on real robots, demonstrating that our approach significantly outperforms the state-of-the-art VLA model, OpenVLA, in terms of speed and data efficiency, while delivering comparable or superior performance. Additionally, TinyVLA exhibits strong generalization capabilities across various dimensions, including language instructions, novel objects, unseen positions, changes in object appearance, background variations, and environmental shifts, often matching or exceeding the performance of OpenVLA. We believe that \methodname offers an interesting perspective on utilizing pre-trained multimodal models for policy learning. Our project is at https://tiny-vla.github.io.

Paperid: 3035, https://arxiv.org/pdf/2409.08464.pdf

Abstract:
Vision Transformers (ViTs) have emerged as the backbone of many segmentation models, consistently achieving state-of-the-art (SOTA) performance. However, their success comes at a significant computational cost. Image token pruning is one of the most effective strategies to address this complexity. However, previous approaches fall short when applied to more complex task-oriented segmentation (TOS), where the class of each image patch is not predefined but dependent on the specific input task. This work introduces the Vision Language Guided Token Pruning (VLTP), a novel token pruning mechanism that can accelerate ViT-based segmentation models, particularly for TOS guided by multi-modal large language model (MLLM). We argue that ViT does not need to process every image token through all of its layers -- only the tokens related to reasoning tasks are necessary. We design a new pruning decoder to take both image tokens and vision-language guidance as input to predict the relevance of each image token to the task. Only image tokens with high relevance are passed to deeper layers of the ViT. Experiments show that the VLTP framework reduces the computational costs of ViT by approximately 25% without performance degradation and by around 40% with only a 1% performance drop. The code associated with this study can be found at this URL.

Paperid: 3036, https://arxiv.org/pdf/2601.11404.pdf

Abstract:
Vision-Language-Action (VLA) models have emerged as essential generalist robot policies for diverse manipulation tasks, conventionally relying on directly translating multimodal inputs into actions via Vision-Language Model (VLM) embeddings. Recent advancements have introduced explicit intermediary reasoning, such as sub-task prediction (language) or goal image synthesis (vision), to guide action generation. However, these intermediate reasoning are often indirect and inherently limited in their capacity to convey the full, granular information required for precise action execution. Instead, we posit that the most effective form of reasoning is one that deliberates directly in the action space. We introduce Action Chain-of-Thought (ACoT), a paradigm where the reasoning process itself is formulated as a structured sequence of coarse action intents that guide the final policy. In this paper, we propose ACoT-VLA, a novel architecture that materializes the ACoT paradigm. Specifically, we introduce two complementary components: an Explicit Action Reasoner (EAR) and Implicit Action Reasoner (IAR). The former proposes coarse reference trajectories as explicit action-level reasoning steps, while the latter extracts latent action priors from internal representations of multimodal input, co-forming an ACoT that conditions the downstream action head to enable grounded policy learning. Extensive experiments in real-world and simulation environments demonstrate the superiority of our proposed method, which achieves 98.5%, 84.1%, and 47.4% on LIBERO, LIBERO-Plus and VLABench, respectively.

Paperid: 3037, https://arxiv.org/pdf/2512.24125.pdf

Abstract:
General-purpose robotic systems operating in open-world environments must achieve both broad generalization and high-precision action execution, a combination that remains challenging for existing Vision-Language-Action (VLA) models. While large Vision-Language Models (VLMs) improve semantic generalization, insufficient embodied reasoning leads to brittle behavior, and conversely, strong reasoning alone is inadequate without precise control. To provide a decoupled and quantitative assessment of this bottleneck, we introduce Embodied Reasoning Intelligence Quotient (ERIQ), a large-scale embodied reasoning benchmark in robotic manipulation, comprising 6K+ question-answer pairs across four reasoning dimensions. By decoupling reasoning from execution, ERIQ enables systematic evaluation and reveals a strong positive correlation between embodied reasoning capability and end-to-end VLA generalization. To bridge the gap from reasoning to precise execution, we propose FACT, a flow-matching-based action tokenizer that converts continuous control into discrete sequences while preserving high-fidelity trajectory reconstruction. The resulting GenieReasoner jointly optimizes reasoning and action in a unified space, outperforming both continuous-action and prior discrete-action baselines in real-world tasks. Together, ERIQ and FACT provide a principled framework for diagnosing and overcoming the reasoning-precision trade-off, advancing robust, general-purpose robotic manipulation.

Paperid: 3038, https://arxiv.org/pdf/2512.21871.pdf

Abstract:
Large vision-language models (LVLMs) have achieved remarkable advancements in multimodal reasoning tasks. However, their widespread accessibility raises critical concerns about potential copyright infringement. Will LVLMs accurately recognize and comply with copyright regulations when encountering copyrighted content (i.e., user input, retrieved documents) in the context? Failure to comply with copyright regulations may lead to serious legal and ethical consequences, particularly when LVLMs generate responses based on copyrighted materials (e.g., retrieved book experts, news reports). In this paper, we present a comprehensive evaluation of various LVLMs, examining how they handle copyrighted content -- such as book excerpts, news articles, music lyrics, and code documentation when they are presented as visual inputs. To systematically measure copyright compliance, we introduce a large-scale benchmark dataset comprising 50,000 multimodal query-content pairs designed to evaluate how effectively LVLMs handle queries that could lead to copyright infringement. Given that real-world copyrighted content may or may not include a copyright notice, the dataset includes query-content pairs in two distinct scenarios: with and without a copyright notice. For the former, we extensively cover four types of copyright notices to account for different cases. Our evaluation reveals that even state-of-the-art closed-source LVLMs exhibit significant deficiencies in recognizing and respecting the copyrighted content, even when presented with the copyright notice. To solve this limitation, we introduce a novel tool-augmented defense framework for copyright compliance, which reduces infringement risks in all scenarios. Our findings underscore the importance of developing copyright-aware LVLMs to ensure the responsible and lawful use of copyrighted content.

Paperid: 3039, https://arxiv.org/pdf/2512.18411.pdf

Abstract:
Multi-prompt learning methods have emerged as an effective approach for facilitating the rapid adaptation of vision-language models to downstream tasks with limited resources. Existing multi-prompt learning methods primarily focus on utilizing various meticulously designed prompts within a single foundation vision-language model to achieve superior performance. However, the overlooked model-prompt matching bias hinders the development of multi-prompt learning, i.e., the same prompt can convey different semantics across distinct vision-language models, such as CLIP-ViT-B/16 and CLIP-ViT-B/32, resulting in inconsistent predictions of identical prompt. To mitigate the impact of this bias on downstream tasks, we explore an ensemble learning approach to sufficiently aggregate the benefits of diverse predictions. Additionally, we further disclose the presence of sample-prompt matching bias, which originates from the prompt-irrelevant semantics encapsulated in the input samples. Thus, directly utilizing all information from the input samples for generating weights of ensemble learning can lead to suboptimal performance. In response, we extract prompt-relevant semantics from input samples by leveraging the guidance of the information theory-based analysis, adaptively calculating debiased ensemble weights. Overall, we propose Adaptive-Debiased Ensemble MultiPrompt Learning, abbreviated as AmPLe, to mitigate the two types of bias simultaneously. Extensive experiments on three representative tasks, i.e., generalization to novel classes, new target datasets, and unseen domain shifts, show that AmPLe can widely outperform existing methods. Theoretical validation from a causal perspective further supports the effectiveness of AmPLe.

Paperid: 3040, https://arxiv.org/pdf/2512.15649.pdf

Abstract:
The computational and memory overheads associated with expanding the context window of LLMs severely limit their scalability. A noteworthy solution is vision-text compression (VTC), exemplified by frameworks like DeepSeek-OCR and Glyph, which convert long texts into dense 2D visual representations, thereby achieving token compression ratios of 3x-20x. However, the impact of this high information density on the core long-context capabilities of vision-language models (VLMs) remains under-investigated. To address this gap, we introduce the first benchmark for VTC and systematically assess the performance of VLMs across three long-context understanding settings: VTC-Retrieval, which evaluates the model's ability to retrieve and aggregate information; VTC-Reasoning, which requires models to infer latent associations to locate facts with minimal lexical overlap; and VTC-Memory, which measures comprehensive question answering within long-term dialogue memory. Furthermore, we establish the VTCBench-Wild to simulate diverse input scenarios.We comprehensively evaluate leading open-source and proprietary models on our benchmarks. The results indicate that, despite being able to decode textual information (e.g., OCR) well, most VLMs exhibit a surprisingly poor long-context understanding ability with VTC-processed information, failing to capture long associations or dependencies in the context.This study provides a deep understanding of VTC and serves as a foundation for designing more efficient and scalable VLMs.

Paperid: 3041, https://arxiv.org/pdf/2512.15557.pdf

Abstract:
Robust robot localization is an important prerequisite for navigation planning. If the environment map was created from different sensors, robot measurements must be robustly associated with map features. In this work, we extend Monte Carlo Localization using vision-language features. These open-vocabulary features enable to robustly compute the likelihood of visual observations, given a camera pose and a 3D map created from posed RGB-D images or aligned point clouds. The abstract vision-language features enable to associate observations and map elements from different modalities. Global localization can be initialized by natural language descriptions of the objects present in the vicinity of locations. We evaluate our approach using Matterport3D and Replica for indoor scenes and demonstrate generalization on SemanticKITTI for outdoor scenes.

Paperid: 3042, https://arxiv.org/pdf/2511.19528.pdf

Abstract:
Scaling vision-language-action (VLA) model pre-training requires large volumes of diverse, high-quality manipulation trajectories. Most current data is obtained via human teleoperation, which is expensive and difficult to scale. Reinforcement learning (RL) methods learn useful skills through autonomous exploration, making them a viable approach for generating data. However, standard RL training collapses to a narrow execution pattern, limiting its utility for large-scale pre-training. We propose Discover, Lea rn and Reinforce (DLR), an information-theoretic pattern discovery framework that generates multiple distinct, high-success behavioral patterns for VLA pretraining. Empirically, DLR generates a markedly more diverse trajectory corpus on LIBERO. Specifically, it learns multiple distinct, high-success strategies for the same task where standard RL discovers only one, and hence it covers substantially broader regions of the state-action space. When adapted to unseen downstream task suites, VLA models pretrained on our diverse RL data surpass counterparts trained on equal-sized standard RL datasets. Moreover, DLR exhibits positive data-scaling behavior that single-pattern RL lacks. These results position multi-pattern RL as a practical, scalable data engine for embodied foundation models.

Paperid: 3043, https://arxiv.org/pdf/2511.19315.pdf

Abstract:
Vision-Language Model (VLM) is an important component to enable robust robot manipulation. Yet, using it to translate human instructions into an action-resolvable intermediate representation often needs a tradeoff between VLM-comprehensibility and generalizability. Inspired by context-free grammar, we design the Semantic Assembly representation named SEAM, by decomposing the intermediate representation into vocabulary and grammar. Doing so leads us to a concise vocabulary of semantically-rich operations and a VLM-friendly grammar for handling diverse unseen tasks. In addition, we design a new open-vocabulary segmentation paradigm with a retrieval-augmented few-shot learning strategy to localize fine-grained object parts for manipulation, effectively with the shortest inference time over all state-of-the-art parallel works. Also, we formulate new metrics for action-generalizability and VLM-comprehensibility, demonstrating the compelling performance of SEAM over mainstream representations on both aspects. Extensive real-world experiments further manifest its SOTA performance under varying settings and tasks.

Paperid: 3044, https://arxiv.org/pdf/2511.13759.pdf

Abstract:
Accurate detection of offensive content on social media demands high-quality labeled data; however, such data is often scarce due to the low prevalence of offensive instances and the high cost of manual annotation. To address this low-resource challenge, we propose a self-training framework that leverages abundant unlabeled data through collaborative pseudo-labeling. Starting with a lightweight classifier trained on limited labeled data, our method iteratively assigns pseudo-labels to unlabeled instances with the support of Multi-Agent Vision-Language Models (MA-VLMs). Un-labeled data on which the classifier and MA-VLMs agree are designated as the Agreed-Unknown set, while conflicting samples form the Disagreed-Unknown set. To enhance label reliability, MA-VLMs simulate dual perspectives, moderator and user, capturing both regulatory and subjective viewpoints. The classifier is optimized using a novel Positive-Negative-Unlabeled (PNU) loss, which jointly exploits labeled, Agreed-Unknown, and Disagreed-Unknown data while mitigating pseudo-label noise. Experiments on benchmark datasets demonstrate that our framework substantially outperforms baselines under limited supervision and approaches the performance of large-scale models

Paperid: 3045, https://arxiv.org/pdf/2511.13225.pdf

Abstract:
With the rise of Large Language Models (LLMs) and their vision-enabled counterparts (VLMs), numerous works have investigated their capabilities in tasks that fuse the modalities of vision and language. In this work, we benchmark the extent to which VLMs are able to act as highly-trained phoneticians, interpreting spectrograms and waveforms of speech. To do this, we synthesise a novel dataset containing 4k+ English words spoken in isolation alongside stylistically consistent spectrogram and waveform figures. We test the ability of VLMs to understand these representations of speech through a multiple-choice task whereby models must predict the correct phonemic or graphemic transcription of a spoken word when presented amongst 3 distractor transcriptions that have been selected based on their phonemic edit distance to the ground truth. We observe that both zero-shot and finetuned models rarely perform above chance, demonstrating the requirement for specific parametric knowledge of how to interpret such figures, rather than paired samples alone.

Paperid: 3046, https://arxiv.org/pdf/2511.11690.pdf

Abstract:
Test-time prompt tuning for vision-language models has demonstrated impressive generalization capabilities under zero-shot settings. However, tuning the learnable prompts solely based on unlabeled test data may induce prompt optimization bias, ultimately leading to suboptimal performance on downstream tasks. In this work, we analyze the underlying causes of prompt optimization bias from both the model and data perspectives. In terms of the model, the entropy minimization objective typically focuses on reducing the entropy of model predictions while overlooking their correctness. This can result in overconfident yet incorrect outputs, thereby compromising the quality of prompt optimization. On the data side, prompts affected by optimization bias can introduce misalignment between visual and textual modalities, which further aggravates the prompt optimization bias. To this end, we propose a Doubly Debiased Test-Time Prompt Tuning method. Specifically, we first introduce a dynamic retrieval-augmented modulation module that retrieves high-confidence knowledge from a dynamic knowledge base using the test image feature as a query, and uses the retrieved knowledge to modulate the predictions. Guided by the refined predictions, we further develop a reliability-aware prompt optimization module that incorporates a confidence-based weighted ensemble and cross-modal consistency distillation to impose regularization constraints during prompt tuning. Extensive experiments across 15 benchmark datasets involving both natural distribution shifts and cross-datasets generalization demonstrate that our method outperforms baselines, validating its effectiveness in mitigating prompt optimization bias.

Paperid: 3047, https://arxiv.org/pdf/2511.10268.pdf

Abstract:
Large Vision-Language Models (LVLMs) often suffer from object hallucination, making erroneous judgments about the presence of objects in images. We propose this primar- ily stems from spurious correlations arising when models strongly associate highly co-occurring objects during train- ing, leading to hallucinated objects influenced by visual con- text. Current benchmarks mainly focus on hallucination de- tection but lack a formal characterization and quantitative evaluation of spurious correlations in LVLMs. To address this, we introduce causal analysis into the object recognition scenario of LVLMs, establishing a Structural Causal Model (SCM). Utilizing the language of causality, we formally de- fine spurious correlations arising from co-occurrence bias. To quantify the influence induced by these spurious correla- tions, we develop Causal-HalBench, a benchmark specifically constructed with counterfactual samples and integrated with comprehensive causal metrics designed to assess model ro- bustness against spurious correlations. Concurrently, we pro- pose an extensible pipeline for the construction of these coun- terfactual samples, leveraging the capabilities of proprietary LVLMs and Text-to-Image (T2I) models for their genera- tion. Our evaluations on mainstream LVLMs using Causal- HalBench demonstrate these models exhibit susceptibility to spurious correlations, albeit to varying extents.

Paperid: 3048, https://arxiv.org/pdf/2511.06619.pdf

Abstract:
Vision-language-action (VLA) models hold the promise to attain generalizable embodied control. To achieve this, a pervasive paradigm is to leverage the rich vision-semantic priors of large vision-language models (VLMs). However, the fundamental question persists: How do VLAs effectively inherit the prior knowledge from VLMs? To address this critical question, we introduce a diagnostic benchmark, GrinningFace, an emoji tabletop manipulation task where the robot arm is asked to place objects onto printed emojis corresponding to language instructions. This task design is particularly revealing -- knowledge associated with emojis is ubiquitous in Internet-scale datasets used for VLM pre-training, yet emojis themselves are largely absent from standard robotics datasets. Consequently, they provide a clean proxy: successful task completion indicates effective transfer of VLM priors to embodied control. We implement this diagnostic task in both simulated environment and a real robot, and compare various promising techniques for knowledge transfer. Specifically, we investigate the effects of parameter-efficient fine-tuning, VLM freezing, co-training, predicting discretized actions, and predicting latent actions. Through systematic evaluation, our work not only demonstrates the critical importance of preserving VLM priors for the generalization of VLA but also establishes guidelines for future research in developing truly generalizable embodied AI systems.

Paperid: 3049, https://arxiv.org/pdf/2510.18873.pdf

Abstract:
Reasoning about dynamic spatial relationships is essential, as both observers and objects often move simultaneously. Although vision-language models (VLMs) and visual expertise models excel in 2D tasks and static scenarios, their ability to fully understand dynamic 3D scenarios remains limited. We introduce Dynamic Spatial Intelligence and propose DSI-Bench, a benchmark with nearly 1,000 dynamic videos and over 1,700 manually annotated questions covering nine decoupled motion patterns of observers and objects. Spatially and temporally symmetric designs reduce biases and enable systematic evaluation of models' reasoning about self-motion and object motion. Our evaluation of 14 VLMs and expert models reveals key limitations: models often conflate observer and object motion, exhibit semantic biases, and fail to accurately infer relative relationships in dynamic scenarios. Our DSI-Bench provides valuable findings and insights about the future development of general and expertise models with dynamic spatial intelligence.

Paperid: 3050, https://arxiv.org/pdf/2510.17640.pdf

Abstract:
Vision-Language-Action models (VLAs) have demonstrated remarkable performance on complex robotic manipulation tasks through imitation learning. However, existing imitation learning datasets contain only successful trajectories and lack failure or recovery data, especially for out-of-distribution (OOD) states where the robot deviates from the main policy due to minor perturbations or errors, leading VLA models to struggle with states deviating from the training distribution. To this end, we propose an automated OOD data augmentation framework named RESample through exploratory sampling. Specifically, we first leverage offline reinforcement learning to obtain an action-value network that accurately identifies sub-optimal actions under the current manipulation policy. We further sample potential OOD states from trajectories via rollout, and design an exploratory sampling mechanism that adaptively incorporates these action proxies into the training dataset to ensure efficiency. Subsequently, our framework explicitly encourages the VLAs to recover from OOD states and enhances their robustness against distributional shifts. We conduct extensive experiments on the LIBERO benchmark as well as real-world robotic manipulation tasks, demonstrating that RESample consistently improves the stability and generalization ability of VLA models.

Paperid: 3051, https://arxiv.org/pdf/2510.17313.pdf

Abstract:
Learning disentangled representations in sequential data is a key goal in deep learning, with broad applications in vision, audio, and time series. While real-world data involves multiple interacting semantic factors over time, prior work has mostly focused on simpler two-factor static and dynamic settings, primarily because such settings make data collection easier, thereby overlooking the inherently multi-factor nature of real-world data. We introduce the first standardized benchmark for evaluating multi-factor sequential disentanglement across six diverse datasets spanning video, audio, and time series. Our benchmark includes modular tools for dataset integration, model development, and evaluation metrics tailored to multi-factor analysis. We additionally propose a post-hoc Latent Exploration Stage to automatically align latent dimensions with semantic factors, and introduce a Koopman-inspired model that achieves state-of-the-art results. Moreover, we show that Vision-Language Models can automate dataset annotation and serve as zero-shot disentanglement evaluators, removing the need for manual labels and human intervention. Together, these contributions provide a robust and scalable foundation for advancing multi-factor sequential disentanglement.

Paperid: 3052, https://arxiv.org/pdf/2510.13626.pdf

Abstract:
Visual-Language-Action (VLA) models report impressive success rates on robotic manipulation benchmarks, yet these results may mask fundamental weaknesses in robustness. We perform a systematic vulnerability analysis by introducing controlled perturbations across seven dimensions: objects layout, camera viewpoints, robot initial states, language instructions, light conditions, background textures and sensor noise. We comprehensively analyzed multiple state-of-the-art models and revealed consistent brittleness beneath apparent competence. Our analysis exposes critical weaknesses: models exhibit extreme sensitivity to perturbation factors, including camera viewpoints and robot initial states, with performance dropping from 95% to below 30% under modest perturbations. Surprisingly, models are largely insensitive to language variations, with further experiments revealing that models tend to ignore language instructions completely. Our findings challenge the assumption that high benchmark scores equate to true competency and highlight the need for evaluation practices that assess reliability under realistic variation.

Paperid: 3053, https://arxiv.org/pdf/2509.25564.pdf

Abstract:
Multimodal large language models (MLLMs) have demonstrated impressive cross-domain capabilities, yet their proficiency in specialized scientific fields like marine biology remains underexplored. In this work, we systematically evaluate state-of-the-art MLLMs and reveal significant limitations in their ability to perform fine-grained recognition of fish species, with the best open-source models achieving less than 10\% accuracy. This task is critical for monitoring marine ecosystems under anthropogenic pressure. To address this gap and investigate whether these failures stem from a lack of domain knowledge, we introduce FishNet++, a large-scale, multimodal benchmark. FishNet++ significantly extends existing resources with 35,133 textual descriptions for multimodal learning, 706,426 key-point annotations for morphological studies, and 119,399 bounding boxes for detection. By providing this comprehensive suite of annotations, our work facilitates the development and evaluation of specialized vision-language models capable of advancing aquatic science.

Paperid: 3054, https://arxiv.org/pdf/2509.25143.pdf

Abstract:
Existing medical reasoning benchmarks for vision-language models primarily focus on analyzing a patient's condition based on an image from a single visit. However, this setting deviates significantly from real-world clinical practice, where doctors typically refer to a patient's historical conditions to provide a comprehensive assessment by tracking their changes over time. In this paper, we introduce TemMed-Bench, the first benchmark designed for analyzing changes in patients' conditions between different clinical visits, which challenges large vision-language models (LVLMs) to reason over temporal medical images. TemMed-Bench consists of a test set comprising three tasks - visual question-answering (VQA), report generation, and image-pair selection - and a supplementary knowledge corpus of over 17,000 instances. With TemMed-Bench, we conduct an evaluation of six proprietary and six open-source LVLMs. Our results show that most LVLMs lack the ability to analyze patients' condition changes over temporal medical images, and a large proportion perform only at a random-guessing level in the closed-book setting. In contrast, GPT o3, o4-mini and Claude 3.5 Sonnet demonstrate comparatively decent performance, though they have yet to reach the desired level. Furthermore, we explore augmenting the input with both retrieved visual and textual modalities in the medical domain. We also show that multi-modal retrieval augmentation yields notably higher performance gains than no retrieval and textual retrieval alone across most models on our benchmark, with the VQA task showing an average improvement of 2.59%. Overall, we compose a benchmark grounded on real-world clinical practice, and it reveals LVLMs' limitations in temporal medical image reasoning, as well as highlighting the use of multi-modal retrieval augmentation as a potentially promising direction worth exploring to address this challenge.

Paperid: 3055, https://arxiv.org/pdf/2509.22643.pdf

Abstract:
Vision-Language-Action models (VLAs) achieve strong performance in general robotic manipulation tasks by scaling imitation learning. However, existing VLAs are limited to predicting short-sighted next-action, which struggle with long-horizon trajectory tasks due to incremental deviations. To address this problem, we propose a plug-in framework named VLA-Reasoner that effectively empowers off-the-shelf VLAs with the capability of foreseeing future states via test-time scaling. Specifically, VLA-Reasoner samples and rolls out possible action trajectories where involved actions are rationales to generate future states via a world model, which enables VLA-Reasoner to foresee and reason potential outcomes and search for the optimal actions. We further leverage Monte Carlo Tree Search (MCTS) to improve search efficiency in large action spaces, where stepwise VLA predictions seed the root. Meanwhile, we introduce a confidence sampling mechanism based on Kernel Density Estimation (KDE), to enable efficient exploration in MCTS without redundant VLA queries. We evaluate intermediate states in MCTS via an offline reward shaping strategy, to score predicted futures and correct deviations with long-term feedback. We conducted extensive experiments in both simulators and the real world, demonstrating that our proposed VLA-Reasoner achieves significant improvements over the state-of-the-art VLAs. Our method highlights a potential pathway toward scalable test-time computation of robotic manipulation.

Paperid: 3056, https://arxiv.org/pdf/2509.21247.pdf

Abstract:
Convolutional Neural Networks (CNNs) frequently "cheat" by exploiting superficial correlations, raising concerns about whether they make predictions for the right reasons. Inspired by cognitive science, which highlights the role of attention in robust human perception, recent methods have sought to guide model attention using concept-based supervision and explanation regularization. However, these techniques depend on labor-intensive, expert-provided annotations, limiting their scalability. We propose a scalable framework that leverages vision-language models to automatically generate semantic attention maps using natural language prompts. By introducing an auxiliary loss that aligns CNN attention with these language-guided maps, our approach promotes more reliable and cognitively plausible decision-making without manual annotation. Experiments on challenging datasets, ColoredMNIST and DecoyMNIST, show that our method achieves state-of-the-art performance on ColorMNIST and remains competitive with annotation-heavy baselines on DecoyMNIST, demonstrating improved generalization, reduced shortcut reliance, and model attention that better reflects human intuition.

Paperid: 3057, https://arxiv.org/pdf/2509.21107.pdf

Abstract:
Teaching robots novel behaviors typically requires motion demonstrations via teleoperation or kinaesthetic teaching, that is, physically guiding the robot. While recent work has explored using human sketches to specify desired behaviors, data collection remains cumbersome, and demonstration datasets are difficult to scale. In this paper, we introduce an alternative paradigm, Learning from Cross-Modal Instructions, where robots are shaped by demonstrations in the form of rough annotations, which can contain free-form text labels, and are used in lieu of physical motion. We introduce the CrossInstruct framework, which integrates cross-modal instructions as examples into the context input to a foundational vision-language model (VLM). The VLM then iteratively queries a smaller, fine-tuned model, and synthesizes the desired motion over multiple 2D views. These are then subsequently fused into a coherent distribution over 3D motion trajectories in the robot's workspace. By incorporating the reasoning of the large VLM with a fine-grained pointing model, CrossInstruct produces executable robot behaviors that generalize beyond the environment of in the limited set of instruction examples. We then introduce a downstream reinforcement learning pipeline that leverages CrossInstruct outputs to efficiently learn policies to complete fine-grained tasks. We rigorously evaluate CrossInstruct on benchmark simulation tasks and real hardware, demonstrating effectiveness without additional fine-tuning and providing a strong initialization for policies subsequently refined via reinforcement learning.

Paperid: 3058, https://arxiv.org/pdf/2509.19752.pdf

Abstract:
Vision-language-action (VLA) models have shown strong generalization across tasks and embodiments; however, their reliance on large-scale human demonstrations limits their scalability owing to the cost and effort of manual data collection. Reinforcement learning (RL) offers a potential alternative to generate demonstrations autonomously, yet conventional RL algorithms often struggle on long-horizon manipulation tasks with sparse rewards. In this paper, we propose a modified diffusion policy optimization algorithm to generate high-quality and low-variance trajectories, which contributes to a diffusion RL-powered VLA training pipeline. Our algorithm benefits from not only the high expressiveness of diffusion models to explore complex and diverse behaviors but also the implicit regularization of the iterative denoising process to yield smooth and consistent demonstrations. We evaluate our approach on the LIBERO benchmark, which includes 130 long-horizon manipulation tasks, and show that the generated trajectories are smoother and more consistent than both human demonstrations and those from standard Gaussian RL policies. Further, training a VLA model exclusively on the diffusion RL-generated data achieves an average success rate of 81.9%, which outperforms the model trained on human data by +5.3% and that on Gaussian RL-generated data by +12.6%. The results highlight our diffusion RL as an effective alternative for generating abundant, high-quality, and low-variance demonstrations for VLA models.

Paperid: 3059, https://arxiv.org/pdf/2509.19274.pdf

Abstract:
We introduce DRISHTIKON, a first-of-its-kind multimodal and multilingual benchmark centered exclusively on Indian culture, designed to evaluate the cultural understanding of generative AI systems. Unlike existing benchmarks with a generic or global scope, DRISHTIKON offers deep, fine-grained coverage across India's diverse regions, spanning 15 languages, covering all states and union territories, and incorporating over 64,000 aligned text-image pairs. The dataset captures rich cultural themes including festivals, attire, cuisines, art forms, and historical heritage amongst many more. We evaluate a wide range of vision-language models (VLMs), including open-source small and large models, proprietary systems, reasoning-specialized VLMs, and Indic-focused models, across zero-shot and chain-of-thought settings. Our results expose key limitations in current models' ability to reason over culturally grounded, multimodal inputs, particularly for low-resource languages and less-documented traditions. DRISHTIKON fills a vital gap in inclusive AI research, offering a robust testbed to advance culturally aware, multimodally competent language technologies.

Paperid: 3060, https://arxiv.org/pdf/2509.17074.pdf

Abstract:
Visual affordance learning is crucial for robots to understand and interact effectively with the physical world. Recent advances in this field attempt to leverage pre-trained knowledge of vision-language foundation models to learn affordance properties with limited training data, providing a novel paradigm for visual affordance learning. However, these methods overlook the significance of maintaining feature alignment between visual images and language descriptions for identifying affordance areas with textual guidance, and thus may lead to suboptimal results. In this paper, we present an informative framework for text-guided affordance learning, which involves information-based constraints to achieve text-image alignment at feature level. Specifically, we design an affordance mutual information constraint that helps learn appropriate textual prompts and task-oriented visual features simultaneously by maximizing the mutual information between the features of the affordance areas in the input images and the corresponding textual prompts. In addition, we propose an object-level information constraint that maximizes the mutual information between the visual features of a given object and the text features of the category it belongs to. This enables the model to capture high-quality representations for the object, providing more reliable semantic priors for identifying affordance regions. Experimental results on the AGD20K dataset show that the proposed method outperforms existing approaches and achieves the new state-of-the-art in one-shot affordance learning.

Paperid: 3061, https://arxiv.org/pdf/2509.17042.pdf

Abstract:
The advancement of foundation models fosters new initiatives for policy learning in achieving safe and efficient autonomous driving. However, a critical bottleneck lies in the manual engineering of reward functions and training curricula for complex and dynamic driving tasks, which is a labor-intensive and time-consuming process. To address this problem, we propose OGR (Orchestrate, Generate, Reflect), a novel automated driving policy learning framework that leverages vision-language model (VLM)-based multi-agent collaboration. Our framework capitalizes on advanced reasoning and multimodal understanding capabilities of VLMs to construct a hierarchical agent system. Specifically, a centralized orchestrator plans high-level training objectives, while a generation module employs a two-step analyze-then-generate process for efficient generation of reward-curriculum pairs. A reflection module then facilitates iterative optimization based on the online evaluation. Furthermore, a dedicated memory module endows the VLM agents with the capabilities of long-term memory. To enhance robustness and diversity of the generation process, we introduce a parallel generation scheme and a human-in-the-loop technique for augmentation of the reward observation space. Through efficient multi-agent cooperation and leveraging rich multimodal information, OGR enables the online evolution of reinforcement learning policies to acquire interaction-aware driving skills. Extensive experiments in the CARLA simulator demonstrate the superior performance, robust generalizability across distinct urban scenarios, and strong compatibility with various RL algorithms. Further real-world experiments highlight the practical viability and effectiveness of our framework. The source code will be available upon acceptance of the paper.

Paperid: 3062, https://arxiv.org/pdf/2509.09680.pdf

Abstract:
The advancement of open-source text-to-image (T2I) models has been hindered by the absence of large-scale, reasoning-focused datasets and comprehensive evaluation benchmarks, resulting in a performance gap compared to leading closed-source systems. To address this challenge, We introduce FLUX-Reason-6M and PRISM-Bench (Precise and Robust Image Synthesis Measurement Benchmark). FLUX-Reason-6M is a massive dataset consisting of 6 million high-quality FLUX-generated images and 20 million bilingual (English and Chinese) descriptions specifically designed to teach complex reasoning. The image are organized according to six key characteristics: Imagination, Entity, Text rendering, Style, Affection, and Composition, and design explicit Generation Chain-of-Thought (GCoT) to provide detailed breakdowns of image generation steps. The whole data curation takes 15,000 A100 GPU days, providing the community with a resource previously unattainable outside of large industrial labs. PRISM-Bench offers a novel evaluation standard with seven distinct tracks, including a formidable Long Text challenge using GCoT. Through carefully designed prompts, it utilizes advanced vision-language models for nuanced human-aligned assessment of prompt-image alignment and image aesthetics. Our extensive evaluation of 19 leading models on PRISM-Bench reveals critical performance gaps and highlights specific areas requiring improvement. Our dataset, benchmark, and evaluation code are released to catalyze the next wave of reasoning-oriented T2I generation. Project page: https://flux-reason-6m.github.io/ .

Paperid: 3063, https://arxiv.org/pdf/2509.05112.pdf

Abstract:
This paper introduces a GenAI-driven approach for automated test case generation, leveraging Large Language Models and Vision-Language Models to translate natural language requirements and system diagrams into structured Gherkin test cases. The methodology integrates Vehicle Signal Specification modeling to standardize vehicle signal definitions, improve compatibility across automotive subsystems, and streamline integration with third-party testing tools. Generated test cases are executed within the digital.auto playground, an open and vendor-neutral environment designed to facilitate rapid validation of software-defined vehicle functionalities. We evaluate our approach using the Child Presence Detection System use case, demonstrating substantial reductions in manual test specification effort and rapid execution of generated tests. Despite significant automation, the generation of test cases and test scripts still requires manual intervention due to current limitations in the GenAI pipeline and constraints of the digital.auto platform.

Paperid: 3064, https://arxiv.org/pdf/2509.03805.pdf

Abstract:
Large vision language models (VLMs) increasingly claim reasoning skills, yet current benchmarks evaluate them in single-turn or question answering settings. However, grounding is an interactive process in which people gradually develop shared understanding through ongoing communication. We introduce a four-metric suite (grounding efficiency, content alignment, lexical adaptation, and human-likeness) to systematically evaluate VLM performance in interactive grounding contexts. We deploy the suite on 150 self-play sessions of interactive referential games between three proprietary VLMs and compare them with human dyads. All three models diverge from human patterns on at least three metrics, while GPT4o-mini is the closest overall. We find that (i) task success scores do not indicate successful grounding and (ii) high image-utterance alignment does not necessarily predict task success. Our metric suite and findings offer a framework for future research on VLM grounding.

Paperid: 3065, https://arxiv.org/pdf/2509.01895.pdf

Abstract:
The escalating intensity and frequency of wildfires demand innovative computational methods for rapid and accurate property damage assessment. Traditional methods are often time consuming, while modern computer vision approaches typically require extensive labeled datasets, hindering immediate post-disaster deployment. This research introduces a novel, zero-shot framework leveraging pre-trained vision language models (VLMs) to classify damage from ground-level imagery. We propose and evaluate two pipelines applied to the 2025 Eaton and Palisades fires in California, a VLM (Pipeline A) and a VLM + large language model (LLM) approach (Pipeline B), that integrate structured prompts based on specific wildfire damage indicators. A primary scientific contribution of this study is demonstrating the VLMs efficacy in synthesizing information from multiple perspectives to identify nuanced damage, a critical limitation in existing literature. Our findings reveal that while single view assessments struggled to classify affected structures (F1 scores ranging from 0.225 to 0.511), the multi-view analysis yielded dramatic improvements (F1 scores ranging from 0.857 to 0.947). Moreover, the McNemar test confirmed that pipelines with a multi-view image assessment yields statistically significant classification improvements; however, the improvements this research observed between Pipeline A and B were not statistically significant. Thus, future research can explore the potential of LLM prompting in damage assessment. The practical contribution is an immediately deployable, flexible, and interpretable workflow that bypasses the need for supervised training, significantly accelerating triage and prioritization for disaster response practitioners.

Paperid: 3066, https://arxiv.org/pdf/2508.05580.pdf

Abstract:
With the growing demands of AI-generated content (AIGC), the need for high-quality, diverse, and scalable data has become increasingly crucial. However, collecting large-scale real-world data remains costly and time-consuming, hindering the development of downstream applications. While some works attempt to collect task-specific data via a rendering process, most approaches still rely on manual scene construction, limiting their scalability and accuracy. To address these challenges, we propose Follow-Your-Instruction, a Multimodal Large Language Model (MLLM)-driven framework for automatically synthesizing high-quality 2D, 3D, and 4D data. Our \textbf{Follow-Your-Instruction} first collects assets and their associated descriptions through multimodal inputs using the MLLM-Collector. Then it constructs 3D layouts, and leverages Vision-Language Models (VLMs) for semantic refinement through multi-view scenes with the MLLM-Generator and MLLM-Optimizer, respectively. Finally, it uses MLLM-Planner to generate temporally coherent future frames. We evaluate the quality of the generated data through comprehensive experiments on the 2D, 3D, and 4D generative tasks. The results show that our synthetic data significantly boosts the performance of existing baseline models, demonstrating Follow-Your-Instruction's potential as a scalable and effective data engine for generative intelligence.

Paperid: 3067, https://arxiv.org/pdf/2507.23682.pdf

Abstract:
Vision-Language-Action (VLA) models have emerged as a popular paradigm for learning robot manipulation policies that can follow language instructions and generalize to novel scenarios. Recent works have begun to explore the incorporation of latent actions, abstract representations of motion between two frames, into VLA pre-training. In this paper, we introduce villa-X, a novel Vision-Language-Latent-Action (ViLLA) framework that advances latent action modeling for learning generalizable robot manipulation policies. Our approach improves both how latent actions are learned and how they are incorporated into VLA pre-training. We demonstrate that villa-X can generate latent action plans in a zero-shot fashion, even for unseen embodiments and open-vocabulary symbolic understanding. This capability enables villa-X to achieve superior performance across diverse simulation tasks in SIMPLER and on two real-world robotic setups involving both gripper and dexterous hand manipulation. These results establish villa-X as a principled and scalable paradigm for learning generalizable robot manipulation policies. We believe it provides a strong foundation for future research.

Paperid: 3068, https://arxiv.org/pdf/2507.23540.pdf

Abstract:
Autonomous driving systems face significant challenges in achieving human-like adaptability, robustness, and interpretability in complex, open-world environments. These challenges stem from fragmented architectures, limited generalization to novel scenarios, and insufficient semantic extraction from perception. To address these limitations, we propose a unified Perception-Language-Action (PLA) framework that integrates multi-sensor fusion (cameras, LiDAR, radar) with a large language model (LLM)-augmented Vision-Language-Action (VLA) architecture, specifically a GPT-4.1-powered reasoning core. This framework unifies low-level sensory processing with high-level contextual reasoning, tightly coupling perception with natural language-based semantic understanding and decision-making to enable context-aware, explainable, and safety-bounded autonomous driving. Evaluations on an urban intersection scenario with a construction zone demonstrate superior performance in trajectory tracking, speed prediction, and adaptive planning. The results highlight the potential of language-augmented cognitive frameworks for advancing the safety, interpretability, and scalability of autonomous driving systems.

Paperid: 3069, https://arxiv.org/pdf/2507.19666.pdf

Abstract:
The intersection of AI and legal systems presents a growing need for tools that support legal education, particularly in under-resourced languages such as Romanian. In this work, we aim to evaluate the capabilities of Large Language Models (LLMs) and Vision-Language Models (VLMs) in understanding and reasoning about Romanian driving law through textual and visual question-answering tasks. To facilitate this, we introduce RoD-TAL, a novel multimodal dataset comprising Romanian driving test questions, text-based and image-based, alongside annotated legal references and human explanations. We implement and assess retrieval-augmented generation (RAG) pipelines, dense retrievers, and reasoning-optimized models across tasks including Information Retrieval (IR), Question Answering (QA), Visual IR, and Visual QA. Our experiments demonstrate that domain-specific fine-tuning significantly enhances retrieval performance. At the same time, chain-of-thought prompting and specialized reasoning models improve QA accuracy, surpassing the minimum grades required to pass driving exams. However, visual reasoning remains challenging, highlighting the potential and the limitations of applying LLMs and VLMs to legal education.

Paperid: 3070, https://arxiv.org/pdf/2507.15025.pdf

Abstract:
Adoption of state-of-art Generative Artificial Intelligence (GenAI) aims to revolutionize many industrial areas by reducing the amount of human intervention needed and effort for handling complex underlying processes. Automotive software development is considered to be a significant area for GenAI adoption, taking into account lengthy and expensive procedures, resulting from the amount of requirements and strict standardization. In this paper, we explore the adoption of GenAI for various steps of automotive software development, mainly focusing on requirements handling, compliance aspects and code generation. Three GenAI-related technologies are covered within the state-of-art: Large Language Models (LLMs), Retrieval Augmented Generation (RAG), Vision Language Models (VLMs), as well as overview of adopted prompting techniques in case of code generation. Additionally, we also derive a generalized GenAI-aided automotive software development workflow based on our findings from this literature review. Finally, we include a summary of a survey outcome, which was conducted among our automotive industry partners regarding the type of GenAI tools used for their daily work activities.

Paperid: 3071, https://arxiv.org/pdf/2507.14823.pdf

Abstract:
Large vision-language models (LVLMs) have made significant progress in chart understanding. However, financial charts, characterized by complex temporal structures and domain-specific terminology, remain notably underexplored. We introduce FinChart-Bench, the first benchmark specifically focused on real-world financial charts. FinChart-Bench comprises 1,200 financial chart images collected from 2015 to 2024, each annotated with True/False (TF), Multiple Choice (MC), and Question Answering (QA) questions, totaling 7,016 questions. We conduct a comprehensive evaluation of 25 state-of-the-art LVLMs on FinChart-Bench. Our evaluation reveals critical insights: (1) the performance gap between open-source and closed-source models is narrowing, (2) performance degradation occurs in upgraded models within families, (3) many models struggle with instruction following, (4) both advanced models show significant limitations in spatial reasoning abilities, and (5) current LVLMs are not reliable enough to serve as automated evaluators. These findings highlight important limitations in current LVLM capabilities for financial chart understanding. The FinChart-Bench dataset is available at https://huggingface.co/datasets/Tizzzzy/FinChart-Bench.

Paperid: 3072, https://arxiv.org/pdf/2507.06167.pdf

Abstract:
We introduce Skywork-R1V3, an advanced, open-source vision-language model (VLM) that pioneers a new approach to visual reasoning. Its key innovation lies in effectively transferring reasoning skills from text-only Large Language Models (LLMs) to visual tasks. The strong performance of Skywork-R1V3 primarily stems from our elaborate post-training RL framework, which effectively activates and enhances the model's reasoning ability, without the need for additional continue pre-training. Through this framework, we further uncover the fundamental role of the connector module in achieving robust cross-modal alignment for multimodal reasoning models. In addition, we introduce a unique indicator of reasoning capability, the entropy of critical reasoning tokens, which has proven highly effective for checkpoint selection during RL training. Skywork-R1V3 achieves state-of-the-art results on MMMU, significantly improving from 64.3% to 76.0%. This performance matches entry-level human capabilities. Remarkably, our RL-powered post-training approach enables even the 38B parameter model to rival top closed-source VLMs. The implementation successfully transfers mathematical reasoning to other subject-related reasoning tasks. We also include an analysis of curriculum learning and reinforcement finetuning strategies, along with a broader discussion on multimodal reasoning. Skywork-R1V3 represents a significant leap in multimodal reasoning, showcasing RL as a powerful engine for advancing open-source VLM capabilities.

Paperid: 3073, https://arxiv.org/pdf/2506.00910.pdf

Abstract:
Knowledge distillation (KD) is a widely used framework for training compact, task-specific models by leveraging the knowledge of teacher models. However, its application to active learning (AL), which aims to minimize annotation costs through iterative sample selection, remains underexplored. This gap stems from the fact that KD typically assumes access to sufficient labeled data, whereas AL operates in data-scarce scenarios where task-specific teacher models are often unavailable. In this paper, we introduce ActiveKD, a framework that integrates AL with KD by leveraging the zero- and few-shot capabilities of large vision-language models (VLMs). A key aspect of ActiveKD is the structured prediction bias of VLMs -- i.e., their predictions form clusters in the probability space. We regard this structure as an inductive bias of the teacher model, capturing generalizable output patterns beneficial to student learning. To exploit this bias, we propose Probabilistic CoreSet (PCoreSet), a selection strategy that maximizes coverage in the probability space rather than the feature space. PCoreSet strategically selects categorically diverse unlabeled samples, facilitating more efficient transfer of teacher knowledge under limited annotation budgets. Evaluations on 11 datasets show that PCoreSet consistently outperforms existing selection methods within the ActiveKD framework, advancing research at the intersection of AL and KD.

Paperid: 3074, https://arxiv.org/pdf/2505.21906.pdf

Abstract:
Vision-language-action (VLA) models have emerged as the next generation of models in robotics. However, despite leveraging powerful pre-trained Vision-Language Models (VLMs), existing end-to-end VLA systems often lose key capabilities during fine-tuning as the model adapts to specific robotic tasks. We argue that a generalizable VLA model should retain and expand upon the VLM's core competencies: 1) Open-world embodied reasoning - the VLA should inherit the knowledge from VLM, i.e., recognize anything that the VLM can recognize, be capable of solving math problems, and possess visual-spatial intelligence, 2) Reasoning following - effectively translating the open-world reasoning into actionable steps for the robot. In this work, we introduce ChatVLA-2, a novel mixture-of-expert VLA model coupled with a specialized two-stage training pipeline designed to preserve the VLM's original strengths while enabling actionable reasoning. To validate our approach, we design a math-matching task wherein a robot interprets math problems written on a whiteboard and picks corresponding number cards from a table to solve equations. Remarkably, our method exhibits exceptional mathematical reasoning and OCR capabilities, despite these abilities not being explicitly trained within the VLA. Furthermore, we demonstrate that the VLA possesses strong spatial reasoning skills, enabling it to interpret novel directional instructions involving previously unseen objects. Overall, our method showcases reasoning and comprehension abilities that significantly surpass state-of-the-art imitation learning methods such as OpenVLA, DexVLA, and pi-zero. This work represents a substantial advancement toward developing truly generalizable robotic foundation models endowed with robust reasoning capacities.

Paperid: 3075, https://arxiv.org/pdf/2505.18793.pdf

Abstract:
While Vision-Language-Action (VLA) models show strong generalizability in various tasks, real-world deployment of robotic policy still requires large-scale, high-quality human expert demonstrations. However, passive data collection via human teleoperation is costly, hard to scale, and often biased toward passive demonstrations with limited diversity. To address this, we propose Genie Centurion (GCENT), a scalable and general data collection paradigm based on human rewind-and-refine guidance. When the robot execution failures occur, GCENT enables the system revert to a previous state with a rewind mechanism, after which a teleoperator provides corrective demonstrations to refine the policy. This framework supports a one-human-to-many-robots supervision scheme with a Task Sentinel module, which autonomously predicts task success and solicits human intervention when necessary, enabling scalable supervision. Empirical results show that GCENT achieves up to 40% higher task success rates than state-of-the-art data collection methods, and reaches comparable performance using less than half the data. We also quantify the data yield-to-effort ratio under multi-robot scenarios, demonstrating GCENT's potential for scalable and cost-efficient robot policy training in real-world environments.

Paperid: 3076, https://arxiv.org/pdf/2505.18719.pdf

Abstract:
Recent high-capacity vision-language-action (VLA) models have demonstrated impressive performance on a range of robotic manipulation tasks by imitating human demonstrations. However, exploiting offline data with limited visited states will cause execution failure in out-of-distribution scenarios. Intuitively, an exploration-based method that improves on online collected data at test time could address this limitation. We present VLA-RL, an algorithmic and systematic framework that leverages online reinforcement learning (RL) to improve pretrained auto-regressive VLAs in downstream tasks. Within a unified perspective, we first introduce a trajectory-level RL formulation for auto-regressive VLA training, which models general robotic manipulation trajectory as multi-modal multi-turn conversation. To address the challenge of sparse rewards, we fine-tune a pretrained vision-language model as a robotic process reward model, which is trained on pseudo reward labels annotated on automatically extracted task segments. To scale up, we identify several implementation findings that improve the stability and efficiency including curriculum selection strategy, GPU-balanced vectorized environments, batch decoding, and critic warmup. VLA-RL enables OpenVLA-7B to surpass the strongest finetuned baseline by 4.5% on 40 challenging robotic manipulation tasks in LIBERO, and even matches the performance of advanced commercial models such as $Ï_0$-FAST. Notably, we observe that VLA-RL benefits from increased test-time optimization, indicating an early spark of inference scaling laws in robotics.

Paperid: 3077, https://arxiv.org/pdf/2505.15659.pdf

Abstract:
We introduce $\textbf{F}$uture $\textbf{LA}$tent $\textbf{RE}$presentation Alignment ($\textbf{FLARE}$), a novel framework that integrates predictive latent world modeling into robot policy learning. By aligning features from a diffusion transformer with latent embeddings of future observations, $\textbf{FLARE}$ enables a diffusion transformer policy to anticipate latent representations of future observations, allowing it to reason about long-term consequences while generating actions. Remarkably lightweight, $\textbf{FLARE}$ requires only minimal architectural modifications -- adding a few tokens to standard vision-language-action (VLA) models -- yet delivers substantial performance gains. Across two challenging multitask simulation imitation learning benchmarks spanning single-arm and humanoid tabletop manipulation, $\textbf{FLARE}$ achieves state-of-the-art performance, outperforming prior policy learning baselines by up to 26%. Moreover, $\textbf{FLARE}$ unlocks the ability to co-train with human egocentric video demonstrations without action labels, significantly boosting policy generalization to a novel object with unseen geometry with as few as a single robot demonstration. Our results establish $\textbf{FLARE}$ as a general and scalable approach for combining implicit world modeling with high-frequency robotic control.

Paperid: 3078, https://arxiv.org/pdf/2505.08367.pdf

Abstract:
Vision-language models (VLMs) have demonstrated excellent high-level planning capabilities, enabling locomotion skill learning from video demonstrations without the need for meticulous human-level reward design. However, the improper frame sampling method and low training efficiency of current methods remain a critical bottleneck, resulting in substantial computational overhead and time costs. To address this limitation, we propose Motion-aware Rapid Reward Optimization for Efficient Robot Skill Learning from Single Videos (MA-ROESL). MA-ROESL integrates a motion-aware frame selection method to implicitly enhance the quality of VLM-generated reward functions. It further employs a hybrid three-phase training pipeline that improves training efficiency via rapid reward optimization and derives the final policy through online fine-tuning. Experimental results demonstrate that MA-ROESL significantly enhances training efficiency while faithfully reproducing locomotion skills in both simulated and real-world settings, thereby underscoring its potential as a robust and scalable framework for efficient robot locomotion skill learning from video demonstrations.

Paperid: 3079, https://arxiv.org/pdf/2505.07675.pdf

Abstract:
Vision-language models (VLMs) have achieved remarkable success across diverse tasks by leveraging rich textual information with minimal labeled data. However, deploying such large models remains challenging, particularly in resource-constrained environments. Knowledge distillation (KD) offers a well-established solution to this problem; however, recent KD approaches from VLMs often involve multi-stage training or additional tuning, increasing computational overhead and optimization complexity. In this paper, we propose $\mathbf{\texttt{D}}$ual-$\mathbf{\texttt{H}}$ead $\mathbf{\texttt{O}}$ptimization ($\mathbf{\texttt{DHO}}$) -- a simple yet effective KD framework that transfers knowledge from VLMs to compact, task-specific models in semi-supervised settings. Specifically, we introduce dual prediction heads that independently learn from labeled data and teacher predictions, and propose to linearly combine their outputs during inference. We observe that $\texttt{DHO}$ mitigates gradient conflicts between supervised and distillation signals, enabling more effective feature learning than single-head KD baselines. As a result, extensive experiments show that $\texttt{DHO}$ consistently outperforms baselines across multiple domains and fine-grained datasets. Notably, on ImageNet, it achieves state-of-the-art performance, improving accuracy by 3% and 0.1% with 1% and 10% labeled data, respectively, while using fewer parameters.

Paperid: 3080, https://arxiv.org/pdf/2505.07263.pdf

Abstract:
We propose Skywork-VL Reward, a multimodal reward model that provides reward signals for both multimodal understanding and reasoning tasks. Our technical approach comprises two key components: First, we construct a large-scale multimodal preference dataset that covers a wide range of tasks and scenarios, with responses collected from both standard vision-language models (VLMs) and advanced VLM reasoners. Second, we design a reward model architecture based on Qwen2.5-VL-7B-Instruct, integrating a reward head and applying multi-stage fine-tuning using pairwise ranking loss on pairwise preference data. Experimental evaluations show that Skywork-VL Reward achieves state-of-the-art results on multimodal VL-RewardBench and exhibits competitive performance on the text-only RewardBench benchmark. Furthermore, preference data constructed based on our Skywork-VL Reward proves highly effective for training Mixed Preference Optimization (MPO), leading to significant improvements in multimodal reasoning capabilities. Our results underscore Skywork-VL Reward as a significant advancement toward general-purpose, reliable reward models for multimodal alignment. Our model has been publicly released to promote transparency and reproducibility.

Paperid: 3081, https://arxiv.org/pdf/2503.18533.pdf

Abstract:
Compared to single-turn dialogue, multi-turn dialogue involving multiple images better aligns with the needs of real-world human-AI interactions. Additionally, as training data, it provides richer contextual reasoning information, thereby guiding the model to achieve better performance. However, existing vision-language models (VLMs) primarily rely on single-turn dialogue training and evaluation benchmarks. In this paper, following the characteristics of human dialogue, such as focused topics and concise, clear content, we present MMCR (Multimodal Multi-turn Contextual Reasoning), a novel dataset comprising: (1) MMCR-310k -- the largest multi-image multi-turn instruction tuning dataset with 310K contextual dialogues, each covering 1-4 images and 4 or 8 dialogue turns; and (2) MMCR-Bench -- a diagnostic benchmark featuring dialogues, spanning 8 domains (Humanities, Natural, Science, Education, etc.) and 40 sub-topics. Extensive evaluations demonstrate that models fine-tuned with MMCR-310k achieve 5.2\% higher contextual accuracy on MMCR-Bench, while showing consistent improvements on existing benchmarks (+1.1\% on AI2D, +1.2\% on MMMU and MMVet). MMCR and prompt engineering will be released publicly.

Paperid: 3082, https://arxiv.org/pdf/2503.10183.pdf

Abstract:
Existing vision-language models (VLMs) often suffer from visual hallucination, where the generated responses contain inaccuracies that are not grounded in the visual input. Efforts to address this issue without model finetuning primarily mitigate hallucination by contrastively reducing language biases or amplifying the weights of visual embedding during decoding. However, these approaches remain limited in their ability to capture fine-grained visual details. In this work, we propose the Perception Magnifier (PM), a novel visual decoding method that iteratively isolates relevant visual tokens based on attention and magnifies the corresponding regions, spurring the model to concentrate on fine-grained visual details during decoding. By magnifying critical regions while preserving the structural and contextual information at each decoding step, PM allows the VLM to enhance its scrutiny of the visual input, hence producing more accurate and faithful responses. Extensive experimental results demonstrate that PM not only achieves superior hallucination mitigation but also enhances language generation while preserving strong reasoning capabilities.

Paperid: 3083, https://arxiv.org/pdf/2503.07511.pdf

Abstract:
Vision-Language-Action (VLA) models excel at robotic tasks by leveraging large-scale 2D vision-language pretraining, but their reliance on RGB images limits spatial reasoning critical for real-world interaction. Retraining these models with 3D data is computationally prohibitive, while discarding existing 2D datasets wastes valuable resources. To bridge this gap, we propose PointVLA, a framework that enhances pre-trained VLAs with point cloud inputs without requiring retraining. Our method freezes the vanilla action expert and injects 3D features via a lightweight modular block. To identify the most effective way of integrating point cloud representations, we conduct a skip-block analysis to pinpoint less useful blocks in the vanilla action expert, ensuring that 3D features are injected only into these blocks--minimizing disruption to pre-trained representations. Extensive experiments demonstrate that PointVLA outperforms state-of-the-art 2D imitation learning methods, such as OpenVLA, Diffusion Policy and DexVLA, across both simulated and real-world robotic tasks. Specifically, we highlight several key advantages of PointVLA enabled by point cloud integration: (1) Few-shot multi-tasking, where PointVLA successfully performs four different tasks using only 20 demonstrations each; (2) Real-vs-photo discrimination, where PointVLA distinguishes real objects from their images, leveraging 3D world knowledge to improve safety and reliability; (3) Height adaptability, Unlike conventional 2D imitation learning methods, PointVLA enables robots to adapt to objects at varying table height that unseen in train data. Furthermore, PointVLA achieves strong performance in long-horizon tasks, such as picking and packing objects from a moving conveyor belt, showcasing its ability to generalize across complex, dynamic environments.

Paperid: 3084, https://arxiv.org/pdf/2503.06232.pdf

Abstract:
Chain-of-Thought (CoT) reasoning has proven effective in natural language tasks but remains underexplored in multimodal alignment. This study investigates its integration into 3D vision-language learning by embedding structured reasoning into alignment training. We introduce the 3D-CoT Benchmark, a dataset with hierarchical CoT annotations covering shape recognition, functional inference, and causal reasoning. Through controlled experiments, we compare CoT-structured and standard textual annotations across large reasoning models (LRMs) and large language models (LLMs). Our evaluation employs a dual-layer framework assessing both intermediate reasoning and final inference quality. Extensive experiments demonstrate that CoT significantly improves 3D semantic grounding, with LRMs leveraging CoT more effectively than LLMs. Furthermore, we highlight that annotation structure influences performance-explicit reasoning markers aid LLMs, while unmarked CoT better aligns with LRM inference patterns. Our analyses suggest that CoT is crucial for enhancing multimodal reasoning, with implications beyond 3D tasks. The dataset will be publicly available at https://huggingface.co/datasets/Battam/3D-CoT

Paperid: 3085, https://arxiv.org/pdf/2502.21314.pdf

Abstract:
Text-to-video generation has demonstrated promising progress with the advent of diffusion models, yet existing approaches are limited by dataset quality and computational resources. To address these limitations, this paper presents a comprehensive approach that advances both data curation and model design. We introduce CFC-VIDS-1M, a high-quality video dataset constructed through a systematic coarse-to-fine curation pipeline. The pipeline first evaluates video quality across multiple dimensions, followed by a fine-grained stage that leverages vision-language models to enhance text-video alignment and semantic richness. Building upon the curated dataset's emphasis on visual quality and temporal coherence, we develop RACCOON, a transformer-based architecture with decoupled spatial-temporal attention mechanisms. The model is trained through a progressive four-stage strategy designed to efficiently handle the complexities of video generation. Extensive experiments demonstrate that our integrated approach of high-quality data curation and efficient training strategy generates visually appealing and temporally coherent videos while maintaining computational efficiency. We will release our dataset, code, and models.

Paperid: 3086, https://arxiv.org/pdf/2502.19250.pdf

Abstract:
Imitation learning has proven to be highly effective in teaching robots dexterous manipulation skills. However, it typically relies on large amounts of human demonstration data, which limits its scalability and applicability in dynamic, real-world environments. One key challenge in this context is object generalization, where a robot trained to perform a task with one object, such as "hand over the apple," struggles to transfer its skills to a semantically similar but visually different object, such as "hand over the peach." This gap in generalization to new objects beyond those in the same category has yet to be adequately addressed in previous work on end-to-end visuomotor policy learning. In this paper, we present a simple yet effective approach for achieving object generalization through Vision-Language-Action (VLA) models, referred to as \textbf{ObjectVLA}. Our model enables robots to generalize learned skills to novel objects without requiring explicit human demonstrations for each new target object. By leveraging vision-language pair data, our method provides a lightweight and scalable way to inject knowledge about the target object, establishing an implicit link between the object and the desired action. We evaluate ObjectVLA on a real robotic platform, demonstrating its ability to generalize across 100 novel objects with a 64\% success rate in selecting objects not seen during training. Furthermore, we propose a more accessible method for enhancing object generalization in VLA models, using a smartphone to capture a few images and fine-tune the pre-trained model. These results highlight the effectiveness of our approach in enabling object-level generalization and reducing the need for extensive human demonstrations, paving the way for more flexible and scalable robotic learning systems.

Paperid: 3087, https://arxiv.org/pdf/2502.15682.pdf

Abstract:
The objective in this paper is to improve the performance of text-to-image retrieval. To this end, we introduce a new framework that can boost the performance of large-scale pre-trained vision-language models, so that they can be used for text-to-image re-ranking. The approach, Enhanced Language-Image Pre-training (ELIP), uses the text query, via a simple MLP mapping network, to predict a set of visual prompts to condition the ViT image encoding. ELIP can easily be applied to the commonly used CLIP, SigLIP and BLIP-2 networks. To train the architecture with limited computing resources, we develop a 'student friendly' best practice, involving global hard sample mining, and curation of a large-scale dataset. On the evaluation side, we set up two new out-of-distribution (OOD) benchmarks, Occluded COCO and ImageNet-R, to assess the zero-shot generalisation of the models to different domains. The results demonstrate that ELIP significantly boosts CLIP/SigLIP/SigLIP-2 text-to-image retrieval performance and outperforms BLIP-2 on several benchmarks, as well as providing an easy means to adapt to OOD datasets.

Paperid: 3088, https://arxiv.org/pdf/2502.14949.pdf

Abstract:
With the growing adoption of Retrieval-Augmented Generation (RAG) in document processing, robust text recognition has become increasingly critical for knowledge extraction. While OCR (Optical Character Recognition) for English and other languages benefits from large datasets and well-established benchmarks, Arabic OCR faces unique challenges due to its cursive script, right-to-left text flow, and complex typographic and calligraphic features. We present KITAB-Bench, a comprehensive Arabic OCR benchmark that fills the gaps in current evaluation systems. Our benchmark comprises 8,809 samples across 9 major domains and 36 sub-domains, encompassing diverse document types including handwritten text, structured tables, and specialized coverage of 21 chart types for business intelligence. Our findings show that modern vision-language models (such as GPT-4o, Gemini, and Qwen) outperform traditional OCR approaches (like EasyOCR, PaddleOCR, and Surya) by an average of 60% in Character Error Rate (CER). Furthermore, we highlight significant limitations of current Arabic OCR models, particularly in PDF-to-Markdown conversion, where the best model Gemini-2.0-Flash achieves only 65% accuracy. This underscores the challenges in accurately recognizing Arabic text, including issues with complex fonts, numeral recognition errors, word elongation, and table structure detection. This work establishes a rigorous evaluation framework that can drive improvements in Arabic document analysis methods and bridge the performance gap with English OCR technologies.

Paperid: 3089, https://arxiv.org/pdf/2502.14420.pdf

Abstract:
Humans possess a unified cognitive ability to perceive, comprehend, and interact with the physical world. Why can't large language models replicate this holistic understanding? Through a systematic analysis of existing training paradigms in vision-language-action models (VLA), we identify two key challenges: spurious forgetting, where robot training overwrites crucial visual-text alignments, and task interference, where competing control and understanding tasks degrade performance when trained jointly. To overcome these limitations, we propose ChatVLA, a novel framework featuring Phased Alignment Training, which incrementally integrates multimodal data after initial control mastery, and a Mixture-of-Experts architecture to minimize task interference. ChatVLA demonstrates competitive performance on visual question-answering datasets and significantly surpasses state-of-the-art vision-language-action (VLA) methods on multimodal understanding benchmarks. Notably, it achieves a six times higher performance on MMMU and scores 47.2% on MMStar with a more parameter-efficient design than ECoT. Furthermore, ChatVLA demonstrates superior performance on 25 real-world robot manipulation tasks compared to existing VLA methods like OpenVLA. Our findings highlight the potential of our unified framework for achieving both robust multimodal understanding and effective robot control.

Paperid: 3090, https://arxiv.org/pdf/2502.06873.pdf

Abstract:
Previous research has revealed the potential of large language models (LLMs) to support cognitive reframing therapy; however, their focus was primarily on text-based methods, often overlooking the importance of non-verbal evidence crucial in real-life therapy. To alleviate this gap, we extend the textual cognitive reframing to multimodality, incorporating visual clues. Specifically, we present a new dataset called Multi Modal-Cognitive Support Conversation (M2CoSC), which pairs each GPT-4-generated dialogue with an image that reflects the virtual client's facial expressions. To better mirror real psychotherapy, where facial expressions lead to interpreting implicit emotional evidence, we propose a multi-hop psychotherapeutic reasoning approach that explicitly identifies and incorporates subtle evidence. Our comprehensive experiments with both LLMs and vision-language models (VLMs) demonstrate that the VLMs' performance as psychotherapists is significantly improved with the M2CoSC dataset. Furthermore, the multi-hop psychotherapeutic reasoning method enables VLMs to provide more thoughtful and empathetic suggestions, outperforming standard prompting methods.

Paperid: 3091, https://arxiv.org/pdf/2501.16902.pdf

Abstract:
Recent advancements in dense retrieval have introduced vision-language model (VLM)-based retrievers, such as DSE and ColPali, which leverage document screenshots embedded as vectors to enable effective search and offer a simplified pipeline over traditional text-only methods. In this study, we propose three pixel poisoning attack methods designed to compromise VLM-based retrievers and evaluate their effectiveness under various attack settings and parameter configurations. Our empirical results demonstrate that injecting even a single adversarial screenshot into the retrieval corpus can significantly disrupt search results, poisoning the top-10 retrieved documents for 41.9% of queries in the case of DSE and 26.4% for ColPali. These vulnerability rates notably exceed those observed with equivalent attacks on text-only retrievers. Moreover, when targeting a small set of known queries, the attack success rate raises, achieving complete success in certain cases. By exposing the vulnerabilities inherent in vision-language models, this work highlights the potential risks associated with their deployment.

Paperid: 3092, https://arxiv.org/pdf/2501.04678.pdf

Abstract:
Cancers identified in CT scans are usually accompanied by detailed radiology reports, but publicly available CT datasets often lack these essential reports. This absence limits their usefulness for developing accurate report generation AI. To address this gap, we present AbdomenAtlas 3.0, the first public, high-quality abdominal CT dataset with detailed, expert-reviewed radiology reports. All reports are paired with per-voxel masks and they describe liver, kidney and pancreatic tumors. AbdomenAtlas 3.0 has 9,262 triplets of CT, mask and report--3,955 with tumors. These CT scans come from 17 public datasets. Besides creating the reports for these datasets, we expanded their number of tumor masks by 4.2x, identifying 3,011 new tumor cases. Notably, the reports in AbdomenAtlas 3.0 are more standardized, and generated faster than traditional human-made reports. They provide details like tumor size, location, attenuation and surgical resectability. These reports were created by 12 board-certified radiologists using our proposed RadGPT, a novel framework that converted radiologist-revised tumor segmentation masks into structured and narrative reports. Besides being a dataset creation tool, RadGPT can also become a fully-automatic, segmentation-assisted report generation method. We benchmarked this method and 5 state-of-the-art report generation vision-language models. Our results show that segmentation strongly improves tumor detection in AI-made reports.

Paperid: 3093, https://arxiv.org/pdf/2501.01346.pdf

Abstract:
Large Vision-Language Models (LVLMs) have demonstrated remarkable capabilities in processing both visual and textual information. However, the critical challenge of alignment between visual and textual representations is not fully understood. This survey presents a comprehensive examination of alignment and misalignment in LVLMs through an explainability lens. We first examine the fundamentals of alignment, exploring its representational and behavioral aspects, training methodologies, and theoretical foundations. We then analyze misalignment phenomena across three semantic levels: object, attribute, and relational misalignment. Our investigation reveals that misalignment emerges from challenges at multiple levels: the data level, the model level, and the inference level. We provide a comprehensive review of existing mitigation strategies, categorizing them into parameter-frozen and parameter-tuning approaches. Finally, we outline promising future research directions, emphasizing the need for standardized evaluation protocols and in-depth explainability studies.

Paperid: 3094, https://arxiv.org/pdf/2412.20451.pdf

Abstract:
Robot foundation models, particularly Vision-Language-Action (VLA) models, have garnered significant attention for their ability to enhance robot policy learning, greatly improving robot's generalization and robustness. OpenAI's recent model, O1, showcased impressive capabilities in solving complex problems by utilizing extensive reasoning chains. This prompts an important question: can robot models achieve better performance in multi-task , complex environments by reviewing prior observations and then providing task-specific reasoning to guide action prediction? In this paper, we introduce Chain-of-Affordance (CoA-VLA) , a novel approach to scaling robot models by incorporating reasoning in the format of sequential robot affordances to facilitate task completion. Specifically, we prompt the model to consider the following four types of affordances before taking action: (1) object affordance - what object to manipulate and where it is ; (2) grasp affordance - the specific object part to grasp ; (3) spatial affordance - the optimal space to place the object ; and (4) movement affordance-the collision - free path for movement. We further transform each affordance into two prompting formats: visual affordance and textual affordance. We introduce a novel vision-language co-injection module that integrates this knowledge into the policy network. This allows the robot to leverage essential contextual information during action inference, resulting in improved precision and robustness. Our experiments demonstrate that CoA-VLA outperforms state-of-the-art robot foundation models, including OpenVLA and Octo, on a variety of tasks. Furthermore, CoA-VLA exhibits strong generalization capabilities, including recognizing unseen object poses, identifying free space, and avoiding obstacles in novel environments.

Paperid: 3095, https://arxiv.org/pdf/2412.18826.pdf

Abstract:
While Multimodal Large Language Models (MLLMs) have made remarkable progress in vision-language reasoning, they are also more susceptible to producing harmful content compared to models that focus solely on text. Existing defensive prompting techniques rely on a static, unified safety guideline that fails to account for the specific risks inherent in different multimodal contexts. To address these limitations, we propose RapGuard, a novel framework that uses multimodal chain-of-thought reasoning to dynamically generate scenario-specific safety prompts. RapGuard enhances safety by adapting its prompts to the unique risks of each input, effectively mitigating harmful outputs while maintaining high performance on benign tasks. Our experimental results across multiple MLLM benchmarks demonstrate that RapGuard achieves state-of-the-art safety performance, significantly reducing harmful content without degrading the quality of responses.

Paperid: 3096, https://arxiv.org/pdf/2412.14457.pdf

Abstract:
Generation with source attribution is important for enhancing the verifiability of retrieval-augmented generation (RAG) systems. However, existing approaches in RAG primarily link generated content to document-level references, making it challenging for users to locate evidence among multiple content-rich retrieved documents. To address this challenge, we propose Retrieval-Augmented Generation with Visual Source Attribution (VISA), a novel approach that combines answer generation with visual source attribution. Leveraging large vision-language models (VLMs), VISA identifies the evidence and highlights the exact regions that support the generated answers with bounding boxes in the retrieved document screenshots. To evaluate its effectiveness, we curated two datasets: Wiki-VISA, based on crawled Wikipedia webpage screenshots, and Paper-VISA, derived from PubLayNet and tailored to the medical domain. Experimental results demonstrate the effectiveness of VISA for visual source attribution on documents' original look, as well as highlighting the challenges for improvement. Code, data, and model checkpoints will be released.

Paperid: 3097, https://arxiv.org/pdf/2411.14432.pdf

Abstract:
Large Language Models (LLMs) demonstrate enhanced capabilities and reliability by reasoning more, evolving from Chain-of-Thought prompting to product-level solutions like OpenAI o1. Despite various efforts to improve LLM reasoning, high-quality long-chain reasoning data and optimized training pipelines still remain inadequately explored in vision-language tasks. In this paper, we present Insight-V, an early effort to 1) scalably produce long and robust reasoning data for complex multi-modal tasks, and 2) an effective training pipeline to enhance the reasoning capabilities of multi-modal large language models (MLLMs). Specifically, to create long and structured reasoning data without human labor, we design a two-step pipeline with a progressive strategy to generate sufficiently long and diverse reasoning paths and a multi-granularity assessment method to ensure data quality. We observe that directly supervising MLLMs with such long and complex reasoning data will not yield ideal reasoning ability. To tackle this problem, we design a multi-agent system consisting of a reasoning agent dedicated to performing long-chain reasoning and a summary agent trained to judge and summarize reasoning results. We further incorporate an iterative DPO algorithm to enhance the reasoning agent's generation stability and quality. Based on the popular LLaVA-NeXT model and our stronger base MLLM, we demonstrate significant performance gains across challenging multi-modal benchmarks requiring visual reasoning. Benefiting from our multi-agent system, Insight-V can also easily maintain or improve performance on perception-focused multi-modal tasks.

Paperid: 3098, https://arxiv.org/pdf/2411.02753.pdf

Abstract:
As medical datasets rapidly expand, creating detailed annotations of different body structures becomes increasingly expensive and time-consuming. We consider that requesting radiologists to create detailed annotations is unnecessarily burdensome and that pre-existing AI models can largely automate this process. Following the spirit don't use a sledgehammer on a nut, we find that, rather than creating annotations from scratch, radiologists only have to review and edit errors if the Best-AI Labels have mistakes. To obtain the Best-AI Labels among multiple AI Labels, we developed an automatic tool, called Label Critic, that can assess label quality through tireless pairwise comparisons. Extensive experiments demonstrate that, when incorporated with our developed Image-Prompt pairs, pre-existing Large Vision-Language Models (LVLM), trained on natural images and texts, achieve 96.5% accuracy when choosing the best label in a pair-wise comparison, without extra fine-tuning. By transforming the manual annotation task (30-60 min/scan) into an automatic comparison task (15 sec/scan), we effectively reduce the manual efforts required from radiologists by an order of magnitude. When the Best-AI Labels are sufficiently accurate (81% depending on body structures), they will be directly adopted as the gold-standard annotations for the dataset, with lower-quality AI Labels automatically discarded. Label Critic can also check the label quality of a single AI Label with 71.8% accuracy when no alternatives are available for comparison, prompting radiologists to review and edit if the estimated quality is low (19% depending on body structures).

Paperid: 3099, https://arxiv.org/pdf/2410.23687.pdf

Abstract:
With the advent of Large Vision-Language Models (LVLMs), new attack vectors, such as cognitive bias, prompt injection, and jailbreaking, have emerged. Understanding these attacks promotes system robustness improvement and neural networks demystification. However, existing surveys often target attack taxonomy and lack in-depth analysis like 1) unified insights into adversariality, transferability, and generalization; 2) detailed evaluations framework; 3) motivation-driven attack categorizations; and 4) an integrated perspective on both traditional and LVLM attacks. This article addresses these gaps by offering a thorough summary of traditional and LVLM adversarial attacks, emphasizing their connections and distinctions, and providing actionable insights for future research.

Paperid: 3100, https://arxiv.org/pdf/2410.23156.pdf

Abstract:
Broadly intelligent agents should form task-specific abstractions that selectively expose the essential elements of a task, while abstracting away the complexity of the raw sensorimotor space. In this work, we present Neuro-Symbolic Predicates, a first-order abstraction language that combines the strengths of symbolic and neural knowledge representations. We outline an online algorithm for inventing such predicates and learning abstract world models. We compare our approach to hierarchical reinforcement learning, vision-language model planning, and symbolic predicate invention approaches, on both in- and out-of-distribution tasks across five simulated robotic domains. Results show that our approach offers better sample complexity, stronger out-of-distribution generalization, and improved interpretability.

Paperid: 3101, https://arxiv.org/pdf/2410.08901.pdf

Abstract:
Task-oriented grasping, which involves grasping specific parts of objects based on their functions, is crucial for developing advanced robotic systems capable of performing complex tasks in dynamic environments. In this paper, we propose a training-free framework that incorporates both semantic and geometric priors for zero-shot task-oriented grasp generation. The proposed framework, SegGrasp, first leverages the vision-language models like GLIP for coarse segmentation. It then uses detailed geometric information from convex decomposition to improve segmentation quality through a fusion policy named GeoFusion. An effective grasp pose can be generated by a grasping network with improved segmentation. We conducted the experiments on both segmentation benchmark and real-world robot grasping. The experimental results show that SegGrasp surpasses the baseline by more than 15\% in grasp and segmentation performance.

Paperid: 3102, https://arxiv.org/pdf/2409.14917.pdf

Abstract:
Recently, Large Language Models (LLMs) and Vision Language Models (VLMs) have demonstrated aptitude as potential substitutes for human participants in experiments testing psycholinguistic phenomena. However, an understudied question is to what extent models that only have access to vision and text modalities are able to implicitly understand sound-based phenomena via abstract reasoning from orthography and imagery alone. To investigate this, we analyse the ability of VLMs and LLMs to demonstrate sound symbolism (i.e., to recognise a non-arbitrary link between sounds and concepts) as well as their ability to "hear" via the interplay of the language and vision modules of open and closed-source multimodal models. We perform multiple experiments, including replicating the classic Kiki-Bouba and Mil-Mal shape and magnitude symbolism tasks, and comparing human judgements of linguistic iconicity with that of LLMs. Our results show that VLMs demonstrate varying levels of agreement with human labels, and more task information may be required for VLMs versus their human counterparts for in silico experimentation. We additionally see through higher maximum agreement levels that Magnitude Symbolism is an easier pattern for VLMs to identify than Shape Symbolism, and that an understanding of linguistic iconicity is highly dependent on model size.

Paperid: 3103, https://arxiv.org/pdf/2409.14485.pdf

Abstract:
Long video understanding poses a significant challenge for current Multi-modal Large Language Models (MLLMs). Notably, the MLLMs are constrained by their limited context lengths and the substantial costs while processing long videos. Although several existing methods attempt to reduce visual tokens, their strategies encounter severe bottleneck, restricting MLLMs' ability to perceive fine-grained visual details. In this work, we propose Video-XL, a novel approach that leverages MLLMs' inherent key-value (KV) sparsification capacity to condense the visual input. Specifically, we introduce a new special token, the Visual Summarization Token (VST), for each interval of the video, which summarizes the visual information within the interval as its associated KV. The VST module is trained by instruction fine-tuning, where two optimizing strategies are offered. 1.Curriculum learning, where VST learns to make small (easy) and large compression (hard) progressively. 2. Composite data curation, which integrates single-image, multi-image, and synthetic data to overcome the scarcity of long-video instruction data. The compression quality is further improved by dynamic compression, which customizes compression granularity based on the information density of different video intervals. Video-XL's effectiveness is verified from three aspects. First, it achieves a superior long-video understanding capability, outperforming state-of-the-art models of comparable sizes across multiple popular benchmarks. Second, it effectively preserves video information, with minimal compression loss even at 16x compression ratio. Third, it realizes outstanding cost-effectiveness, enabling high-quality processing of thousands of frames on a single A100 GPU.

Paperid: 3104, https://arxiv.org/pdf/2409.02101.pdf

Abstract:
This paper addresses the limitations of adverse weather image restoration approaches trained on synthetic data when applied to real-world scenarios. We formulate a semi-supervised learning framework employing vision-language models to enhance restoration performance across diverse adverse weather conditions in real-world settings. Our approach involves assessing image clearness and providing semantics using vision-language models on real data, serving as supervision signals for training restoration models. For clearness enhancement, we use real-world data, utilizing a dual-step strategy with pseudo-labels assessed by vision-language models and weather prompt learning. For semantic enhancement, we integrate real-world data by adjusting weather conditions in vision-language model descriptions while preserving semantic meaning. Additionally, we introduce an effective training strategy to bootstrap restoration performance. Our approach achieves superior results in real-world adverse weather image restoration, demonstrated through qualitative and quantitative comparisons with state-of-the-art works.

Paperid: 3105, https://arxiv.org/pdf/2601.05241.pdf

Abstract:
The diversity, quantity, and quality of manipulation data are critical for training effective robot policies. However, due to hardware and physical setup constraints, collecting large-scale real-world manipulation data remains difficult to scale across diverse environments. Recent work uses text-prompt conditioned image diffusion models to augment manipulation data by altering the backgrounds and tabletop objects in the visual observations. However, these approaches often overlook the practical need for multi-view and temporally coherent observations required by state-of-the-art policy models. Further, text prompts alone cannot reliably specify the scene setup. To provide the diffusion model with explicit visual guidance, we introduce visual identity prompting, which supplies exemplar images as conditioning inputs to guide the generation of the desired scene setup. To this end, we also build a scalable pipeline to curate a visual identity pool from large robotics datasets. Using our augmented manipulation data to train downstream vision-language-action and visuomotor policy models yields consistent performance gains in both simulation and real-robot settings.

Paperid: 3106, https://arxiv.org/pdf/2601.01957.pdf

Abstract:
Large Vision-Language Models (LVLMs) have achieved substantial progress in cross-modal tasks. However, due to language bias, LVLMs are susceptible to object hallucination, which can be primarily divided into category, attribute, and relation hallucination, significantly impeding the trustworthy AI applications. Editing the internal activations of LVLMs has shown promising effectiveness in mitigating hallucinations with minimal cost. However, previous editing approaches neglect the effective guidance offered by factual textual semantics, thereby struggling to explicitly mitigate language bias. To address these issues, we propose Adaptive Factual-guided Visual-Textual Editing for hallucination mitigation (AFTER), which comprises Factual-Augmented Activation Steering (FAS) and Query-Adaptive Offset Optimization (QAO), to adaptively guides the original biased activations towards factual semantics. Specifically, FAS is proposed to provide factual and general guidance for activation editing, thereby explicitly modeling the precise visual-textual associations. Subsequently, QAO introduces a query-aware offset estimator to establish query-specific editing from the general steering vector, enhancing the diversity and granularity of editing. Extensive experiments on standard hallucination benchmarks across three widely adopted LVLMs validate the efficacy of the proposed AFTER, notably achieving up to a 16.3% reduction of hallucination over baseline on the AMBER benchmark. Our code and data will be released for reproducibility.

Paperid: 3107, https://arxiv.org/pdf/2512.24330.pdf

Abstract:
While Vision-Language Models (VLMs) can solve complex tasks through agentic reasoning, their capabilities remain largely constrained to text-oriented chain-of-thought or isolated tool invocation. They fail to exhibit the human-like proficiency required to seamlessly interleave dynamic tool manipulation with continuous reasoning, particularly in knowledge-intensive and visually complex scenarios that demand coordinated external tools such as search and image cropping. In this work, we introduce SenseNova-MARS, a novel Multimodal Agentic Reasoning and Search framework that empowers VLMs with interleaved visual reasoning and tool-use capabilities via reinforcement learning (RL). Specifically, SenseNova-MARS dynamically integrates the image search, text search, and image crop tools to tackle fine-grained and knowledge-intensive visual understanding challenges. In the RL stage, we propose the Batch-Normalized Group Sequence Policy Optimization (BN-GSPO) algorithm to improve the training stability and advance the model's ability to invoke tools and reason effectively. To comprehensively evaluate the agentic VLMs on complex visual tasks, we introduce the HR-MMSearch benchmark, the first search-oriented benchmark composed of high-resolution images with knowledge-intensive and search-driven questions. Experiments demonstrate that SenseNova-MARS achieves state-of-the-art performance on open-source search and fine-grained image understanding benchmarks. Specifically, on search-oriented benchmarks, SenseNova-MARS-8B scores 67.84 on MMSearch and 41.64 on HR-MMSearch, surpassing proprietary models such as Gemini-3-Flash and GPT-5. SenseNova-MARS represents a promising step toward agentic VLMs by providing effective and robust tool-use capabilities. To facilitate further research in this field, we will release all code, models, and datasets.

Paperid: 3108, https://arxiv.org/pdf/2512.19219.pdf

Abstract:
We introduce Image-LoRA, a lightweight parameter efficient fine-tuning (PEFT) recipe for transformer-based vision-language models (VLMs). Image-LoRA applies low-rank adaptation only to the value path of attention layers within the visual-token span, reducing adapter-only training FLOPs roughly in proportion to the visual-token fraction. We further adapt only a subset of attention heads, selected using head influence scores estimated with a rank-1 Image-LoRA, and stabilize per-layer updates via selection-size normalization. Across screen-centric grounding and referring benchmarks spanning text-heavy to image-heavy regimes, Image-LoRA matches or closely approaches standard LoRA accuracy while using fewer trainable parameters and lower adapter-only training FLOPs. The method also preserves the pure-text reasoning performance of VLMs before and after fine-tuning, as further shown on GSM8K.

Paperid: 3109, https://arxiv.org/pdf/2511.23253.pdf

Abstract:
Recent advancements in Vision-Language Models (VLMs) have significantly transformed various industries. In agriculture, these dual-modal capabilities offer promising applications such as precision farming, crop monitoring, pest detection, and environmental sustainability. While several Visual Question Answering (VQA) datasets and benchmarks have been developed to evaluate VLM performance, they often fail to adequately assess the critical reasoning and problem-solving skills required in complex agricultural contexts. To address this gap, we introduce AgriCoT, a VQA dataset that incorporates Chain-of-Thought (CoT) reasoning, specifically designed to evaluate the reasoning capabilities of VLMs. With 4,535 carefully curated samples, AgriCoT offers a comprehensive and robust evaluation of reasoning abilities for VLMs, particularly in zero-shot scenarios, by focusing on their capacity to engage in logical reasoning and effective problem-solving. Our evaluations, conducted with 26 representative VLMs, including both proprietary and open-source models, reveal that while some proprietary models excel at answering questions, there is a notable and significant gap in their reasoning capabilities. This underscores the importance of incorporating CoT for more precise and effective assessments. Our dataset are available at https://huggingface.co/datasets/wenyb/AgriCoT.

Paperid: 3110, https://arxiv.org/pdf/2511.21978.pdf

Abstract:
We introduce PAT3D, the first physics-augmented text-to-3D scene generation framework that integrates vision-language models with physics-based simulation to produce physically plausible, simulation-ready, and intersection-free 3D scenes. Given a text prompt, PAT3D generates 3D objects, infers their spatial relations, and organizes them into a hierarchical scene tree, which is then converted into initial conditions for simulation. A differentiable rigid-body simulator ensures realistic object interactions under gravity, driving the scene toward static equilibrium without interpenetrations. To further enhance scene quality, we introduce a simulation-in-the-loop optimization procedure that guarantees physical stability and non-intersection, while improving semantic consistency with the input prompt. Experiments demonstrate that PAT3D substantially outperforms prior approaches in physical plausibility, semantic consistency, and visual quality. Beyond high-quality generation, PAT3D uniquely enables simulation-ready 3D scenes for downstream tasks such as scene editing and robotic manipulation. Code and data will be released upon acceptance.

Paperid: 3111, https://arxiv.org/pdf/2511.18286.pdf

Abstract:
Current roadside perception systems mainly focus on instance-level perception, which fall short in enabling interaction via natural language and reasoning about traffic behaviors in context. To bridge this gap, we introduce RoadSceneVQA, a large-scale and richly annotated visual question answering (VQA) dataset specifically tailored for roadside scenarios. The dataset comprises 34,736 diverse QA pairs collected under varying weather, illumination, and traffic conditions, targeting not only object attributes but also the intent, legality, and interaction patterns of traffic participants. RoadSceneVQA challenges models to perform both explicit recognition and implicit commonsense reasoning, grounded in real-world traffic rules and contextual dependencies. To fully exploit the reasoning potential of Multi-modal Large Language Models (MLLMs), we further propose CogniAnchor Fusion (CAF), a vision-language fusion module inspired by human-like scene anchoring mechanisms. Moreover, we propose the Assisted Decoupled Chain-of-Thought (AD-CoT) to enhance the reasoned thinking via CoT prompting and multi-task learning. Based on the above, we propose the baseline model RoadMind. Experiments on RoadSceneVQA and CODA-LM benchmark show that the pipeline consistently improves both reasoning accuracy and computational efficiency, allowing the MLLM to achieve state-of-the-art performance in structural traffic perception and reasoning tasks.

Paperid: 3112, https://arxiv.org/pdf/2511.16651.pdf

Abstract:
Recent works explore how real and synthetic data contribute to Vision-Language-Action (VLA) models' generalization. While current VLA models have shown the strong effectiveness of large-scale real-robot pre-training, synthetic data has not previously demonstrated comparable capability at scale. This paper provides the first evidence that synthetic data alone can match the performance of the strongest $π$-dataset in pre-training a VLA model, revealing the substantial value of large-scale simulation. The resulting model also exhibits surprisingly zero-shot sim-to-real transfer on several challenging tasks. Our synthetic dataset, InternData-A1, contains over 630k trajectories and 7,433 hours across 4 embodiments, 18 skills, 70 tasks, and 227 scenes, covering rigid, articulated, deformable, and fluid-object manipulation. It is generated through a highly autonomous, fully decoupled, and compositional simulation pipeline that enables long-horizon skill composition, flexible task assembly, and heterogeneous embodiments with minimal manual tuning. Using the same architecture as $π_0$, we pre-train a model entirely on InternData-A1 and find that it matches the official $π_0$ across 49 simulation tasks, 5 real-world tasks, and 4 long-horizon dexterous tasks. We release the dataset and will open-source the generation pipeline to broaden access to large-scale robotic data and to lower the barrier to scalable data creation for embodied AI research.

Paperid: 3113, https://arxiv.org/pdf/2511.16423.pdf

Abstract:
Efficient and lightweight adaptation of pre-trained Vision-Language Models (VLMs) to downstream tasks through collaborative interactions between local clients and a central server is a rapidly emerging research topic in federated learning. Existing adaptation algorithms are typically trained iteratively, which incur significant communication costs and increase the susceptibility to potential attacks. Motivated by the one-shot federated training techniques that reduce client-server exchanges to a single round, developing a lightweight one-shot federated VLM adaptation method to alleviate these issues is particularly attractive. However, current one-shot approaches face certain challenges in adapting VLMs within federated settings: (1) insufficient exploitation of the rich multimodal information inherent in VLMs; (2) lack of specialized adaptation strategies to systematically handle the severe data heterogeneity; and (3) requiring additional training resource of clients or server. To bridge these gaps, we propose a novel Training-free One-shot Federated Adaptation framework for VLMs, named TOFA. To fully leverage the generalizable multimodal features in pre-trained VLMs, TOFA employs both visual and textual pipelines to extract task-relevant representations. In the visual pipeline, a hierarchical Bayesian model learns personalized, class-specific prototype distributions. For the textual pipeline, TOFA evaluates and globally aligns the generated local text prompts for robustness. An adaptive weight calibration mechanism is also introduced to combine predictions from both modalities, balancing personalization and robustness to handle data heterogeneity. Our method is training-free, not relying on additional training resources on either the client or server side. Extensive experiments across 9 datasets in various federated settings demonstrate the effectiveness of the proposed TOFA method.

Paperid: 3114, https://arxiv.org/pdf/2511.15690.pdf

Abstract:
Mixture-of-Experts (MoE) Multimodal large language models (MLLMs) excel at vision-language tasks, but they suffer from high computational inefficiency. To reduce inference overhead, expert skipping methods have been proposed to deactivate redundant experts based on the current input tokens. However, we find that applying these methods-originally designed for unimodal large language models (LLMs)-to MLLMs results in considerable performance degradation. This is primarily because such methods fail to account for the heterogeneous contributions of experts across MoE layers and modality-specific behaviors of tokens within these layers. Motivated by these findings, we propose MoDES, the first training-free framework that adaptively skips experts to enable efficient and accurate MoE MLLM inference. It incorporates a globally-modulated local gating (GMLG) mechanism that integrates global layer-wise importance into local routing probabilities to accurately estimate per-token expert importance. A dual-modality thresholding (DMT) method is then applied, which processes tokens from each modality separately, to derive the skipping schedule. To set the optimal thresholds, we introduce a frontier search algorithm that exploits monotonicity properties, cutting convergence time from several days to a few hours. Extensive experiments for 3 model series across 13 benchmarks demonstrate that MoDES far outperforms previous approaches. For instance, when skipping 88% experts for Qwen3-VL-MoE-30B-A3B-Instruct, the performance boost is up to 10.67% (97.33% vs. 86.66%). Furthermore, MoDES significantly enhances inference speed, improving the prefilling time by 2.16$\times$ and the decoding time by 1.26$\times$.

Paperid: 3115, https://arxiv.org/pdf/2510.24242.pdf

Abstract:
Large vision-language models (LVLMs) have recently demonstrated great potential in remote sensing (RS) tasks (e.g., disaster monitoring) conducted by low Earth orbit (LEO) satellites. However, their deployment in real-world LEO satellite systems remains largely unexplored, hindered by limited onboard computing resources and brief satellite-ground contacts. We propose Grace, a satellite-ground collaborative system designed for near-realtime LVLM inference in RS tasks. Accordingly, we deploy compact LVLM on satellites for realtime inference, but larger ones on ground stations (GSs) to guarantee end-to-end performance. Grace is comprised of two main phases that are asynchronous satellite-GS Retrieval-Augmented Generation (RAG), and a task dispatch algorithm. Firstly, we still the knowledge archive of GS RAG to satellite archive with tailored adaptive update algorithm during limited satellite-ground data exchange period. Secondly, propose a confidence-based test algorithm that either processes the task onboard the satellite or offloads it to the GS. Extensive experiments based on real-world satellite orbital data show that Grace reduces the average latency by 76-95% compared to state-of-the-art methods, without compromising inference accuracy.

Paperid: 3116, https://arxiv.org/pdf/2509.23625.pdf

Abstract:
Mask Diffusion-based Vision Language Models (MDVLMs) have achieved remarkable progress in multimodal understanding tasks. However, these models are unable to correct errors in generated tokens, meaning they lack self-correction capability. In this paper, we propose Recursive Introspection Mask Diffusion Vision Language Model (RIV), which equips the model with self-correction ability through two novel mechanisms. The first is Introspection Training, where an Introspection Model is introduced to identify errors within generated sequences. Introspection Training enables the model to detect not only grammatical and spelling mistakes, but more importantly, logical errors. The second is Recursive Inference. Beginning with the standard unmasking step, the learned Introspection Model helps to identify errors in the output sequence and remask them. This alternating ($\text{unmask}\rightarrow\text{introspection}\rightarrow\text{remask}$) process is repeated recursively until reliable results are obtained. Experimental results on multiple benchmarks demonstrate that the proposed RIV achieves state-of-the-art performance, outperforming most existing MDVLMs.

Paperid: 3117, https://arxiv.org/pdf/2509.03800.pdf

Abstract:
Radiologic diagnostic errors-under-reading errors, inattentional blindness, and communication failures-remain prevalent in clinical practice. These issues often stem from missed localized abnormalities, limited global context, and variability in report language. These challenges are amplified in 3D imaging, where clinicians must examine hundreds of slices per scan. Addressing them requires systems with precise localized detection, global volume-level reasoning, and semantically consistent natural language reporting. However, existing 3D vision-language models are unable to meet all three needs jointly, lacking local-global understanding for spatial reasoning and struggling with the variability and noise of uncurated radiology reports. We present MedVista3D, a multi-scale semantic-enriched vision-language pretraining framework for 3D CT analysis. To enable joint disease detection and holistic interpretation, MedVista3D performs local and global image-text alignment for fine-grained representation learning within full-volume context. To address report variability, we apply language model rewrites and introduce a Radiology Semantic Matching Bank for semantics-aware alignment. MedVista3D achieves state-of-the-art performance on zero-shot disease classification, report retrieval, and medical visual question answering, while transferring well to organ segmentation and prognosis prediction. Code and datasets will be released.

Paperid: 3118, https://arxiv.org/pdf/2508.21496.pdf

Abstract:
Video multimodal large language models (Video-MLLMs) have achieved remarkable progress in video understanding. However, they remain vulnerable to hallucination-producing content inconsistent with or unrelated to video inputs. Previous video hallucination benchmarks primarily focus on short-videos. They attribute hallucinations to factors such as strong language priors, missing frames, or vision-language biases introduced by the visual encoder. While these causes indeed account for most hallucinations in short videos, they still oversimplify the cause of hallucinations. Sometimes, models generate incorrect outputs but with correct frame-level semantics. We refer to this type of hallucination as Semantic Aggregation Hallucination (SAH), which arises during the process of aggregating frame-level semantics into event-level semantic groups. Given that SAH becomes particularly critical in long videos due to increased semantic complexity across multiple events, it is essential to separate and thoroughly investigate the causes of this type of hallucination. To address the above issues, we introduce ELV-Halluc, the first benchmark dedicated to long-video hallucination, enabling a systematic investigation of SAH. Our experiments confirm the existence of SAH and show that it increases with semantic complexity. Additionally, we find that models are more prone to SAH on rapidly changing semantics. Moreover, we discuss potential approaches to mitigate SAH. We demonstrate that positional encoding strategy contributes to alleviating SAH, and further adopt DPO strategy to enhance the model's ability to distinguish semantics within and across events. To support this, we curate a dataset of 8K adversarial data pairs and achieve improvements on both ELV-Halluc and Video-MME, including a substantial 27.7% reduction in SAH ratio.

Paperid: 3119, https://arxiv.org/pdf/2508.16845.pdf

Abstract:
Recent advances in Vision-Language-Action (VLA) models have established a two-component architecture, where a pre-trained Vision-Language Model (VLM) encodes visual observations and task descriptions, and an action decoder maps these representations to continuous actions. Diffusion models have been widely adopted as action decoders due to their ability to model complex, multimodal action distributions. However, they require multiple iterative denoising steps at inference time or downstream techniques to speed up sampling, limiting their practicality in real-world settings where high-frequency control is crucial. In this work, we present NinA (Normalizing Flows in Action), a fast and expressive alter- native to diffusion-based decoders for VLAs. NinA replaces the diffusion action decoder with a Normalizing Flow (NF) that enables one-shot sampling through an invertible transformation, significantly reducing inference time. We integrate NinA into the FLOWER VLA architecture and fine-tune on the LIBERO benchmark. Our experiments show that NinA matches the performance of its diffusion-based counterpart under the same training regime, while achieving substantially faster inference. These results suggest that NinA offers a promising path toward efficient, high-frequency VLA control without compromising performance.

Paperid: 3120, https://arxiv.org/pdf/2508.09804.pdf

Abstract:
Charts are essential to data analysis, transforming raw data into clear visual representations that support human decision-making. Although current vision-language models (VLMs) have made significant progress, they continue to struggle with chart comprehension due to training on datasets that lack diversity and real-world authenticity, or on automatically extracted underlying data tables of charts, which can contain numerous estimation errors. Furthermore, existing models only rely on supervised fine-tuning using these low-quality datasets, severely limiting their effectiveness. To address these issues, we first propose BigCharts, a dataset creation pipeline that generates visually diverse chart images by conditioning the rendering process on real-world charts sourced from multiple online platforms. Unlike purely synthetic datasets, BigCharts incorporates real-world data, ensuring authenticity and visual diversity, while still retaining accurate underlying data due to our proposed replotting process. Additionally, we introduce a comprehensive training framework that integrates supervised fine-tuning with Group Relative Policy Optimization (GRPO)-based reinforcement learning. By introducing novel reward signals specifically designed for chart reasoning, our approach enhances model robustness and generalization across diverse chart styles and domains, resulting in a state-of-the-art chart reasoning model, BigCharts-R1. Extensive experiments demonstrate that our models surpass existing methods on multiple chart question-answering benchmarks compared to even larger open-source and closed-source models.

Paperid: 3121, https://arxiv.org/pdf/2508.08983.pdf

Abstract:
Humans can observe a single, imperfect demonstration and immediately generalize to very different problem settings. Robots, in contrast, often require hundreds of examples and still struggle to generalize beyond the training conditions. We argue that this limitation arises from the inability to recover the latent explanations that underpin intelligent behavior, and that these explanations can take the form of structured programs consisting of high-level goals, sub-task decomposition, and execution constraints. In this work, we introduce Rational Inverse Reasoning (RIR), a framework for inferring these latent programs through a hierarchical generative model of behavior. RIR frames few-shot imitation as Bayesian program induction: a vision-language model iteratively proposes structured symbolic task hypotheses, while a planner-in-the-loop inference scheme scores each by the likelihood of the observed demonstration under that hypothesis. This loop yields a posterior over concise, executable programs. We evaluate RIR on a suite of continuous manipulation tasks designed to test one-shot and few-shot generalization across variations in object pose, count, geometry, and layout. With as little as one demonstration, RIR infers the intended task structure and generalizes to novel settings, outperforming state-of-the-art vision-language model baselines.

Paperid: 3122, https://arxiv.org/pdf/2508.04062.pdf

Abstract:
Positron emission tomography (PET) is a cornerstone of modern oncologic and neurologic imaging, distinguished by its unique ability to illuminate dynamic metabolic processes that transcend the anatomical focus of traditional imaging technologies. Radiology reports are essential for clinical decision making, yet their manual creation is labor-intensive and time-consuming. Recent advancements of vision-language models (VLMs) have shown strong potential in medical applications, presenting a promising avenue for automating report generation. However, existing applications of VLMs in the medical domain have predominantly focused on structural imaging modalities, while the unique characteristics of molecular PET imaging have largely been overlooked. To bridge the gap, we introduce PET2Rep, a large-scale comprehensive benchmark for evaluation of general and medical VLMs for radiology report generation for PET images. PET2Rep stands out as the first dedicated dataset for PET report generation with metabolic information, uniquely capturing whole-body image-report pairs that cover dozens of organs to fill the critical gap in existing benchmarks and mirror real-world clinical comprehensiveness. In addition to widely recognized natural language generation metrics, we introduce a series of clinical efficiency metrics to evaluate the quality of radiotracer uptake pattern description in key organs in generated reports. We conduct a head-to-head comparison of 30 cutting-edge general-purpose and medical-specialized VLMs. The results show that the current state-of-the-art VLMs perform poorly on PET report generation task, falling considerably short of fulfilling practical needs. Moreover, we identify several key insufficiency that need to be addressed to advance the development in medical applications.

Paperid: 3123, https://arxiv.org/pdf/2508.02329.pdf

Abstract:
Despite the success of Vision-Language Models (VLMs) like CLIP in aligning vision and language, their proficiency in detailed, fine-grained visual comprehension remains a key challenge. We present CLIP-IN, a novel framework that bolsters CLIP's fine-grained perception through two core innovations. Firstly, we leverage instruction-editing datasets, originally designed for image manipulation, as a unique source of hard negative image-text pairs. Coupled with a symmetric hard negative contrastive loss, this enables the model to effectively distinguish subtle visual-semantic differences. Secondly, CLIP-IN incorporates long descriptive captions, utilizing rotary positional encodings to capture rich semantic context often missed by standard CLIP. Our experiments demonstrate that CLIP-IN achieves substantial gains on the MMVP benchmark and various fine-grained visual recognition tasks, without compromising robust zero-shot performance on broader classification and retrieval tasks. Critically, integrating CLIP-IN's visual representations into Multimodal Large Language Models significantly reduces visual hallucinations and enhances reasoning abilities. This work underscores the considerable potential of synergizing targeted, instruction-based contrastive learning with comprehensive descriptive information to elevate the fine-grained understanding of VLMs.

Paperid: 3124, https://arxiv.org/pdf/2507.18100.pdf

Abstract:
Video Temporal Grounding (VTG) aims to localize relevant temporal segments in videos given natural language queries. Despite recent progress with large vision-language models (LVLMs) and instruction-tuning, existing approaches often suffer from limited temporal awareness and poor generalization. In this work, we introduce a two-stage training framework that integrates supervised fine-tuning with reinforcement learning (RL) to improve both the accuracy and robustness of VTG models. Our approach first leverages high-quality curated cold start data for SFT initialization, followed by difficulty-controlled RL to further enhance temporal localization and reasoning abilities. Comprehensive experiments on multiple VTG benchmarks demonstrate that our method consistently outperforms existing models, particularly in challenging and open-domain scenarios. We conduct an in-depth analysis of training strategies and dataset curation, highlighting the importance of both high-quality cold start data and difficulty-controlled RL. To facilitate further research and industrial adoption, we release all intermediate datasets, models, and code to the community.

Paperid: 3125, https://arxiv.org/pdf/2507.09160.pdf

Abstract:
Vision-Language-Action (VLA) models have shown remarkable achievements, driven by the rich implicit knowledge of their vision-language components. However, achieving generalist robotic agents demands precise grounding into physical interactions, especially in contact-rich scenarios where fine-grained force control is essential. We advance VLAs' implicit knowledge beyond identifying what to do, towards guiding how to physically interact with real world. This paper introduces Tactile-VLA, a novel framework that deeply fuses vision, language, action, and tactile sensing. This framework incorporates a hybrid position-force controller to translate the model's intentions into precise physical actions and a reasoning module that allows the robot to adapt its strategy based on tactile feedback. Experiments demonstrate Tactile-VLA's effectiveness and generalizability in three key aspects: (1) enabling tactile-aware instruction following, (2) utilizing tactile-relevant commonsense, and (3) facilitating adaptive tactile-involved reasoning. A key finding is that the VLM's prior knowledge already contains semantic understanding of physical interaction; by connecting it to the robot's tactile sensors with only a few demonstrations, we can activate this prior knowledge to achieve zero-shot generalization in contact-rich tasks.

Paperid: 3126, https://arxiv.org/pdf/2507.05731.pdf

Abstract:
Recently, large vision-language models (LVLMs) unleash powerful analysis capabilities for low Earth orbit (LEO) satellite Earth observation images in the data center. However, fast satellite motion, brief satellite-ground station (GS) contact windows, and large size of the images pose a data download challenge. To enable near real-time Earth observation applications (e.g., disaster and extreme weather monitoring), we should explore how to deploy LVLM in LEO satellite networks, and design SpaceVerse, an efficient satellite-ground synergistic LVLM inference system. To this end, firstly, we deploy compact LVLMs on satellites for lightweight tasks, whereas regular LVLMs operate on GSs to handle computationally intensive tasks. Then, we propose a computing and communication co-design framework comprised of a progressive confidence network and an attention-based multi-scale preprocessing, used to identify on-satellite inferring data, and reduce data redundancy before satellite-GS transmission, separately. We implement and evaluate SpaceVerse on real-world LEO satellite constellations and datasets, achieving a 31.2% average gain in accuracy and a 51.2% reduction in latency compared to state-of-the-art baselines.

Paperid: 3127, https://arxiv.org/pdf/2506.19212.pdf

Abstract:
Dexterous robotic hands are essential for performing complex manipulation tasks, yet remain difficult to train due to the challenges of demonstration collection and high-dimensional control. While reinforcement learning (RL) can alleviate the data bottleneck by generating experience in simulation, it typically relies on carefully designed, task-specific reward functions, which hinder scalability and generalization. Thus, contemporary works in dexterous manipulation have often bootstrapped from reference trajectories. These trajectories specify target hand poses that guide the exploration of RL policies and object poses that enable dense, task-agnostic rewards. However, sourcing suitable trajectories - particularly for dexterous hands - remains a significant challenge. Yet, the precise details in explicit reference trajectories are often unnecessary, as RL ultimately refines the motion. Our key insight is that modern vision-language models (VLMs) already encode the commonsense spatial and semantic knowledge needed to specify tasks and guide exploration effectively. Given a task description (e.g., "open the cabinet") and a visual scene, our method uses an off-the-shelf VLM to first identify task-relevant keypoints (e.g., handles, buttons) and then synthesize 3D trajectories for hand motion and object motion. Subsequently, we train a low-level residual RL policy in simulation to track these coarse trajectories or "scaffolds" with high fidelity. Across a number of simulated tasks involving articulated objects and semantic understanding, we demonstrate that our method is able to learn robust dexterous manipulation policies. Moreover, we showcase that our method transfers to real-world robotic hands without any human demonstrations or handcrafted rewards.

Paperid: 3128, https://arxiv.org/pdf/2506.07697.pdf

Abstract:
3D Gaussian Splatting (3DGS) has emerged as a powerful representation for neural scene reconstruction, offering high-quality novel view synthesis while maintaining computational efficiency. In this paper, we extend the capabilities of 3DGS beyond pure scene representation by introducing an approach for open-vocabulary 3D instance segmentation without requiring manual labeling, termed OpenSplat3D. Our method leverages feature-splatting techniques to associate semantic information with individual Gaussians, enabling fine-grained scene understanding. We incorporate Segment Anything Model instance masks with a contrastive loss formulation as guidance for the instance features to achieve accurate instance-level segmentation. Furthermore, we utilize language embeddings of a vision-language model, allowing for flexible, text-driven instance identification. This combination enables our system to identify and segment arbitrary objects in 3D scenes based on natural language descriptions. We show results on LERF-mask and LERF-OVS as well as the full ScanNet++ validation set, demonstrating the effectiveness of our approach.

Paperid: 3129, https://arxiv.org/pdf/2506.06952.pdf

Abstract:
Recent advances in multimodal foundation models unifying image understanding and generation have opened exciting avenues for tackling a wide range of vision-language tasks within a single framework. Despite progress, existing unified models typically require extensive pretraining and struggle to achieve the same level of performance compared to models dedicated to each task. Additionally, many of these models suffer from slow image generation speeds, limiting their practical deployment in real-time or resource-constrained settings. In this work, we propose Layerwise Timestep-Expert Flow-based Transformer (LaTtE-Flow), a novel and efficient architecture that unifies image understanding and generation within a single multimodal model. LaTtE-Flow builds upon powerful pretrained Vision-Language Models (VLMs) to inherit strong multimodal understanding capabilities, and extends them with a novel Layerwise Timestep Experts flow-based architecture for efficient image generation. LaTtE-Flow distributes the flow-matching process across specialized groups of Transformer layers, each responsible for a distinct subset of timesteps. This design significantly improves sampling efficiency by activating only a small subset of layers at each sampling timestep. To further enhance performance, we propose a Timestep-Conditioned Residual Attention mechanism for efficient information reuse across layers. Experiments demonstrate that LaTtE-Flow achieves strong performance on multimodal understanding tasks, while achieving competitive image generation quality with around 6x faster inference speed compared to recent unified multimodal models.

Paperid: 3130, https://arxiv.org/pdf/2506.02541.pdf

Abstract:
Machine unlearning is used to mitigate the privacy risks of Large Vision-Language Models (LVLMs) arising from training on large-scale web data. However, existing unlearning methods often fail to carefully select substitute outputs for forget targets, resulting in Unlearning Aftermaths-undesirable behaviors such as degenerate, hallucinated, or excessively refused responses. We highlight that, especially for generative LVLMs, it is crucial to consider the quality and informativeness of post-unlearning responses rather than relying solely on naive suppression. To address this, we introduce a new unlearning task for LVLMs that requires models to provide privacy-preserving yet informative and visually grounded responses. We also propose PUBG, a novel unlearning method that explicitly guides post-unlearning behavior toward a desirable output distribution. Experiments show that, while existing methods suffer from Unlearning Aftermaths despite successfully preventing privacy violations, PUBG effectively mitigates these issues, generating visually grounded and informative responses without privacy leakage for forgotten targets.

Paperid: 3131, https://arxiv.org/pdf/2505.22914.pdf

Abstract:
Computer-Aided Design (CAD) plays a central role in engineering and manufacturing, making it possible to create precise and editable 3D models. Using a variety of sensor or user-provided data as inputs for CAD reconstruction can democratize access to design applications. However, existing methods typically focus on a single input modality, such as point clouds, images, or text, which limits their generalizability and robustness. Leveraging recent advances in vision-language models (VLM), we propose a multi-modal CAD reconstruction model that simultaneously processes all three input modalities. Inspired by large language model (LLM) training paradigms, we adopt a two-stage pipeline: supervised fine-tuning (SFT) on large-scale procedurally generated data, followed by reinforcement learning (RL) fine-tuning using online feedback, obtained programatically. Furthermore, we are the first to explore RL fine-tuning of LLMs for CAD tasks demonstrating that online RL algorithms such as Group Relative Preference Optimization (GRPO) outperform offline alternatives. In the DeepCAD benchmark, our SFT model outperforms existing single-modal approaches in all three input modalities simultaneously. More importantly, after RL fine-tuning, cadrille sets new state-of-the-art on three challenging datasets, including a real-world one.

Paperid: 3132, https://arxiv.org/pdf/2505.20793.pdf

Abstract:
Scalable Vector Graphics (SVG) offer a powerful format for representing visual designs as interpretable code. Recent advances in vision-language models (VLMs) have enabled high-quality SVG generation by framing the problem as a code generation task and leveraging large-scale pretraining. VLMs are particularly suitable for this task as they capture both global semantics and fine-grained visual patterns, while transferring knowledge across vision, natural language, and code domains. However, existing VLM approaches often struggle to produce faithful and efficient SVGs because they never observe the rendered images during training. Although differentiable rendering for autoregressive SVG code generation remains unavailable, rendered outputs can still be compared to original inputs, enabling evaluative feedback suitable for reinforcement learning (RL). We introduce RLRF(Reinforcement Learning from Rendering Feedback), an RL method that enhances SVG generation in autoregressive VLMs by leveraging feedback from rendered SVG outputs. Given an input image, the model generates SVG roll-outs that are rendered and compared to the original image to compute a reward. This visual fidelity feedback guides the model toward producing more accurate, efficient, and semantically coherent SVGs. RLRF significantly outperforms supervised fine-tuning, addressing common failure modes and enabling precise, high-quality SVG generation with strong structural understanding and generalization.

Paperid: 3133, https://arxiv.org/pdf/2505.15879.pdf

Abstract:
Recent studies have demonstrated the efficacy of using Reinforcement Learning (RL) in building reasoning models that articulate chains of thoughts prior to producing final answers. However, despite ongoing advances that aim at enabling reasoning for vision-language tasks, existing open-source visual reasoning models typically generate reasoning content with pure natural language, lacking explicit integration of visual information. This limits their ability to produce clearly articulated and visually grounded reasoning chains. To this end, we propose Grounded Reasoning with Images and Texts (GRIT), a novel method for training MLLMs to think with images. GRIT introduces a grounded reasoning paradigm, in which models generate reasoning chains that interleave natural language and explicit bounding box coordinates. These coordinates point to regions of the input image that the model consults during its reasoning process. Additionally, GRIT is equipped with a reinforcement learning approach, GRPO-GR, built upon the GRPO algorithm. GRPO-GR employs robust rewards focused on the final answer accuracy and format of the grounded reasoning output, which eliminates the need for data with reasoning chain annotations or explicit bounding box labels. As a result, GRIT achieves exceptional data efficiency, requiring as few as 20 image-question-answer triplets from existing datasets. Comprehensive evaluations demonstrate that GRIT effectively trains MLLMs to produce coherent and visually grounded reasoning chains, showing a successful unification of reasoning and grounding abilities.

Paperid: 3134, https://arxiv.org/pdf/2505.12632.pdf

Abstract:
Recent advancements in Large Language Models (LLMs) and Vision-Language Models (VLMs) have sparked significant interest in developing GUI visual agents. We introduce MONDAY (Mobile OS Navigation Task Dataset for Agents from YouTube), a large-scale dataset of 313K annotated frames from 20K instructional videos capturing diverse real-world mobile OS navigation across multiple platforms. Models that include MONDAY in their pre-training phases demonstrate robust cross-platform generalization capabilities, consistently outperforming models trained on existing single OS datasets while achieving an average performance gain of 18.11%p on an unseen mobile OS platform. To enable continuous dataset expansion as mobile platforms evolve, we present an automated framework that leverages publicly available video content to create comprehensive task datasets without manual annotation. Our framework comprises robust OCR-based scene detection (95.04% F1score), near-perfect UI element detection (99.87% hit ratio), and novel multi-step action identification to extract reliable action sequences across diverse interface configurations. We contribute both the MONDAY dataset and our automated collection framework to facilitate future research in mobile OS navigation.

Paperid: 3135, https://arxiv.org/pdf/2505.09698.pdf

Abstract:
Vision-Language Models (VLMs) have revolutionized artificial intelligence and robotics due to their commonsense reasoning capabilities. In robotic manipulation, VLMs are used primarily as high-level planners, but recent work has also studied their lower-level reasoning ability, which refers to making decisions about precise robot movements. However, the community currently lacks a clear and common benchmark that can evaluate how well VLMs can aid low-level reasoning in robotics. Consequently, we propose a novel benchmark, ManipBench, to evaluate the low-level robot manipulation reasoning capabilities of VLMs across various dimensions, including how well they understand object-object interactions and deformable object manipulation. We extensively test 33 representative VLMs across 10 model families on our benchmark, including variants to test different model sizes. Our evaluation shows that the performance of VLMs significantly varies across tasks, and there is a strong correlation between this performance and trends in our real-world manipulation tasks. It also shows that there remains a significant gap between these models and human-level understanding. See our website at: https://manipbench.github.io.

Paperid: 3136, https://arxiv.org/pdf/2504.17069.pdf

Abstract:
Autoregressive patch-based image generation has recently shown competitive results in terms of image quality and scalability. It can also be easily integrated and scaled within Vision-Language models. Nevertheless, autoregressive models require a defined order for patch generation. While a natural order based on the dictation of the words makes sense for text generation, there is no inherent generation order that exists for image generation. Traditionally, a raster-scan order (from top-left to bottom-right) guides autoregressive image generation models. In this paper, we argue that this order is suboptimal, as it fails to respect the causality of the image content: for instance, when conditioned on a visual description of a sunset, an autoregressive model may generate clouds before the sun, even though the color of clouds should depend on the color of the sun and not the inverse. In this work, we show that first by training a model to generate patches in any-given-order, we can infer both the content and the location (order) of each patch during generation. Secondly, we use these extracted orders to finetune the any-given-order model to produce better-quality images. Through our experiments, we show on two datasets that this new generation method produces better images than the traditional raster-scan approach, with similar training costs and no extra annotations.

Paperid: 3137, https://arxiv.org/pdf/2504.11509.pdf

Abstract:
The multimodal datasets can be leveraged to pre-train large-scale vision-language models by providing cross-modal semantics. Current endeavors for determining the usage of datasets mainly focus on single-modal dataset ownership verification through intrusive methods and non-intrusive techniques, while cross-modal approaches remain under-explored. Intrusive methods can adapt to multimodal datasets but degrade model accuracy, while non-intrusive methods rely on label-driven decision boundaries that fail to guarantee stable behaviors for verification. To address these issues, we propose a novel prompt-adapted transferable fingerprinting scheme from a training-free perspective, called PATFinger, which incorporates the global optimal perturbation (GOP) and the adaptive prompts to capture dataset-specific distribution characteristics. Our scheme utilizes inherent dataset attributes as fingerprints instead of compelling the model to learn triggers. The GOP is derived from the sample distribution to maximize embedding drifts between different modalities. Subsequently, our PATFinger re-aligns the adaptive prompt with GOP samples to capture the cross-modal interactions on the carefully crafted surrogate model. This allows the dataset owner to check the usage of datasets by observing specific prediction behaviors linked to the PATFinger during retrieval queries. Extensive experiments demonstrate the effectiveness of our scheme against unauthorized multimodal dataset usage on various cross-modal retrieval architectures by 30% over state-of-the-art baselines.

Paperid: 3138, https://arxiv.org/pdf/2504.07072.pdf

Authors:Israfel Salazar, Manuel FernÃ¡ndez Burda, Shayekh Bin Islam, Arshia Soltani Moakhar, Shivalika Singh, Fabian Farestam, Angelika Romanou, Danylo Boiko, Dipika Khullar, Mike Zhang, Dominik KrzemiÅski, Jekaterina Novikova, LuÃsa Shimabucoro, Joseph Marvin Imperial, Rishabh Maheshwary, Sharad Duwal, Alfonso Amayuelas, Swati Rajwal, Jebish Purbey, Ahmed Ruby, Nicholas PopoviÄ, Marek Suppa, Azmine Toushik Wasi, Ram Mohan Rao Kadiyala, Olga Tsymboi, Maksim Kostritsya, Bardia Soltani Moakhar, Gabriel da Costa Merlin, OtÃ¡vio Ferracioli Coletti, Maral Jabbari Shiviari, MohammadAmin farahani fard, Silvia Fernandez, MarÃa Grandury, Dmitry Abulkhanov, Drishti Sharma, Andre Guarnier De Mitri, Leticia Bossatto Marchezi, Setayesh Heydari, Johan Obando-Ceron, Nazar Kohut, Beyza Ermis, Desmond Elliott, Enzo Ferrante, Sara Hooker, Marzieh Fadaee

Abstract:
The evaluation of vision-language models (VLMs) has mainly relied on English-language benchmarks, leaving significant gaps in both multilingual and multicultural coverage. While multilingual benchmarks have expanded, both in size and languages, many rely on translations of English datasets, failing to capture cultural nuances. In this work, we propose Kaleidoscope, as the most comprehensive exam benchmark to date for the multilingual evaluation of vision-language models. Kaleidoscope is a large-scale, in-language multimodal benchmark designed to evaluate VLMs across diverse languages and visual inputs. Kaleidoscope covers 18 languages and 14 different subjects, amounting to a total of 20,911 multiple-choice questions. Built through an open science collaboration with a diverse group of researchers worldwide, Kaleidoscope ensures linguistic and cultural authenticity. We evaluate top-performing multilingual vision-language models and find that they perform poorly on low-resource languages and in complex multimodal scenarios. Our results highlight the need for progress on culturally inclusive multimodal evaluation frameworks.

Paperid: 3139, https://arxiv.org/pdf/2503.21889.pdf

Abstract:
Workflows are a fundamental component of automation in enterprise platforms, enabling the orchestration of tasks, data processing, and system integrations. Despite being widely used, building workflows can be complex, often requiring manual configuration through low-code platforms or visual programming tools. To simplify this process, we explore the use of generative foundation models, particularly vision-language models (VLMs), to automatically generate structured workflows from visual inputs. Translating hand-drawn sketches or computer-generated diagrams into executable workflows is challenging due to the ambiguity of free-form drawings, variations in diagram styles, and the difficulty of inferring execution logic from visual elements. To address this, we introduce StarFlow, a framework for generating structured workflow outputs from sketches using vision-language models. We curate a diverse dataset of workflow diagrams -- including synthetic, manually annotated, and real-world samples -- to enable robust training and evaluation. We finetune and benchmark multiple vision-language models, conducting a series of ablation studies to analyze the strengths and limitations of our approach. Our results show that finetuning significantly enhances structured workflow generation, outperforming large vision-language models on this task.

Paperid: 3140, https://arxiv.org/pdf/2503.20492.pdf

Abstract:
Reliable prediction by classifiers is crucial for their deployment in high security and dynamically changing situations. However, modern neural networks often exhibit overconfidence for misclassified predictions, highlighting the need for confidence estimation to detect errors. Despite the achievements obtained by existing methods on small-scale datasets, they all require training from scratch and there are no efficient and effective misclassification detection (MisD) methods, hindering practical application towards large-scale and ever-changing datasets. In this paper, we pave the way to exploit vision language model (VLM) leveraging text information to establish an efficient and general-purpose misclassification detection framework. By harnessing the power of VLM, we construct FSMisD, a Few-Shot prompt learning framework for MisD to refrain from training from scratch and therefore improve tuning efficiency. To enhance misclassification detection ability, we use adaptive pseudo sample generation and a novel negative loss to mitigate the issue of overconfidence by pushing category prompts away from pseudo features. We conduct comprehensive experiments with prompt learning methods and validate the generalization ability across various datasets with domain shift. Significant and consistent improvement demonstrates the effectiveness, efficiency and generalizability of our approach.

Paperid: 3141, https://arxiv.org/pdf/2503.19145.pdf

Abstract:
Attribute detection is crucial for many computer vision tasks, as it enables systems to describe properties such as color, texture, and material. Current approaches often rely on labor-intensive annotation processes which are inherently limited: objects can be described at an arbitrary level of detail (e.g., color vs. color shades), leading to ambiguities when the annotators are not instructed carefully. Furthermore, they operate within a predefined set of attributes, reducing scalability and adaptability to unforeseen downstream applications. We present Compositional Caching (ComCa), a training-free method for open-vocabulary attribute detection that overcomes these constraints. ComCa requires only the list of target attributes and objects as input, using them to populate an auxiliary cache of images by leveraging web-scale databases and Large Language Models to determine attribute-object compatibility. To account for the compositional nature of attributes, cache images receive soft attribute labels. Those are aggregated at inference time based on the similarity between the input and cache images, refining the predictions of underlying Vision-Language Models (VLMs). Importantly, our approach is model-agnostic, compatible with various VLMs. Experiments on public datasets demonstrate that ComCa significantly outperforms zero-shot and cache-based baselines, competing with recent training-based methods, proving that a carefully designed training-free approach can successfully address open-vocabulary attribute detection.

Paperid: 3142, https://arxiv.org/pdf/2503.18623.pdf

Abstract:
Vision Language Models (VLMs) have lead to major improvements in multimodal reasoning, yet they still struggle to understand user-specific concepts. Existing personalization methods address this limitation but heavily rely on training procedures, that can be either costly or unpleasant to individual users. We depart from existing work, and for the first time explore the training-free setting in the context of personalization. We propose a novel method, Retrieval and Reasoning for Personalization (R2P), leveraging internal knowledge of VLMs. First, we leverage VLMs to extract the concept fingerprint, i.e., key attributes uniquely defining the concept within its semantic class. When a query arrives, the most similar fingerprints are retrieved and scored via chain-of-thought-reasoning. To reduce the risk of hallucinations, the scores are validated through cross-modal verification at the attribute level: in case of a discrepancy between the scores, R2P refines the concept association via pairwise multimodal matching, where the retrieved fingerprints and their images are directly compared with the query. We validate R2P on two publicly available benchmarks and a newly introduced dataset, Personal Concepts with Visual Ambiguity (PerVA), for concept identification highlighting challenges in visual ambiguity. R2P consistently outperforms state-of-the-art approaches on various downstream tasks across all benchmarks. Code will be available upon acceptance.

Paperid: 3143, https://arxiv.org/pdf/2503.14734.pdf

Authors:NVIDIA, :, Johan Bjorck, Fernando CastaÃ±eda, Nikita Cherniadev, Xingye Da, Runyu Ding, Linxi "Jim" Fan, Yu Fang, Dieter Fox, Fengyuan Hu, Spencer Huang, Joel Jang, Zhenyu Jiang, Jan Kautz, Kaushil Kundalia, Lawrence Lao, Zhiqi Li, Zongyu Lin, Kevin Lin, Guilin Liu, Edith Llontop, Loic Magne, Ajay Mandlekar, Avnish Narayan, Soroush Nasiriany, Scott Reed, You Liang Tan, Guanzhi Wang, Zu Wang, Jing Wang, Qi Wang, Jiannan Xiang, Yuqi Xie, Yinzhen Xu, Zhenjia Xu, Seonghyeon Ye, Zhiding Yu, Ao Zhang, Hao Zhang, Yizhou Zhao, Ruijie Zheng, Yuke Zhu

Abstract:
General-purpose robots need a versatile body and an intelligent mind. Recent advancements in humanoid robots have shown great promise as a hardware platform for building generalist autonomy in the human world. A robot foundation model, trained on massive and diverse data sources, is essential for enabling the robots to reason about novel situations, robustly handle real-world variability, and rapidly learn new tasks. To this end, we introduce GR00T N1, an open foundation model for humanoid robots. GR00T N1 is a Vision-Language-Action (VLA) model with a dual-system architecture. The vision-language module (System 2) interprets the environment through vision and language instructions. The subsequent diffusion transformer module (System 1) generates fluid motor actions in real time. Both modules are tightly coupled and jointly trained end-to-end. We train GR00T N1 with a heterogeneous mixture of real-robot trajectories, human videos, and synthetically generated datasets. We show that our generalist robot model GR00T N1 outperforms the state-of-the-art imitation learning baselines on standard simulation benchmarks across multiple robot embodiments. Furthermore, we deploy our model on the Fourier GR-1 humanoid robot for language-conditioned bimanual manipulation tasks, achieving strong performance with high data efficiency.

Paperid: 3144, https://arxiv.org/pdf/2503.11609.pdf

Abstract:
An old-school recipe for training a classifier is to (i) learn a good feature extractor and (ii) optimize a linear layer atop. When only a handful of samples are available per category, as in Few-Shot Adaptation (FSA), data are insufficient to fit a large number of parameters, rendering the above impractical. This is especially true with large pre-trained Vision-Language Models (VLMs), which motivated successful research at the intersection of Parameter-Efficient Fine-tuning (PEFT) and FSA. In this work, we start by analyzing the learning dynamics of PEFT techniques when trained on few-shot data from only a subset of categories, referred to as the ``base'' classes. We show that such dynamics naturally splits into two distinct phases: (i) task-level feature extraction and (ii) specialization to the available concepts. To accommodate this dynamic, we then depart from prompt- or adapter-based methods and tackle FSA differently. Specifically, given a fixed computational budget, we split it to (i) learn a task-specific feature extractor via PEFT and (ii) train a linear classifier on top. We call this scheme Two-Stage Few-Shot Adaptation (2SFS). Differently from established methods, our scheme enables a novel form of selective inference at a category level, i.e., at test time, only novel categories are embedded by the adapted text encoder, while embeddings of base categories are available within the classifier. Results with fixed hyperparameters across two settings, three backbones, and eleven datasets, show that 2SFS matches or surpasses the state-of-the-art, while established methods degrade significantly across settings.

Paperid: 3145, https://arxiv.org/pdf/2503.06903.pdf

Abstract:
Vision-Language Models (VLMs) have achieved remarkable success in various tasks, yet their robustness to real-world illumination variations remains largely unexplored. To bridge this gap, we propose \textbf{I}llumination \textbf{T}ransformation \textbf{A}ttack (\textbf{ITA}), the first framework to systematically assess VLMs' robustness against illumination changes. However, there still exist two key challenges: (1) how to model global illumination with fine-grained control to achieve diverse lighting conditions and (2) how to ensure adversarial effectiveness while maintaining naturalness. To address the first challenge, we innovatively decompose global illumination into multiple parameterized point light sources based on the illumination rendering equation. This design enables us to model more diverse lighting variations that previous methods could not capture. Then, by integrating these parameterized lighting variations with physics-based lighting reconstruction techniques, we could precisely render such light interactions in the original scenes, finally meeting the goal of fine-grained lighting control. For the second challenge, by controlling illumination through the lighting reconstrution model's latent space rather than direct pixel manipulation, we inherently preserve physical lighting priors. Furthermore, to prevent potential reconstruction artifacts, we design additional perceptual constraints for maintaining visual consistency with original images and diversity constraints for avoiding light source convergence. Extensive experiments demonstrate that our ITA could significantly reduce the performance of advanced VLMs, e.g., LLaVA-1.6, while possessing competitive naturalness, exposing VLMS' critical illuminiation vulnerabilities.

Paperid: 3146, https://arxiv.org/pdf/2502.17516.pdf

Abstract:
The rise of foundation models has transformed machine learning research, prompting efforts to uncover their inner workings and develop more efficient and reliable applications for better control. While significant progress has been made in interpreting Large Language Models (LLMs), multimodal foundation models (MMFMs) - such as contrastive vision-language models, generative vision-language models, and text-to-image models - pose unique interpretability challenges beyond unimodal frameworks. Despite initial studies, a substantial gap remains between the interpretability of LLMs and MMFMs. This survey explores two key aspects: (1) the adaptation of LLM interpretability methods to multimodal models and (2) understanding the mechanistic differences between unimodal language models and crossmodal systems. By systematically reviewing current MMFM analysis techniques, we propose a structured taxonomy of interpretability methods, compare insights across unimodal and multimodal architectures, and highlight critical research gaps.

Paperid: 3147, https://arxiv.org/pdf/2502.01341.pdf

Abstract:
Aligning visual features with language embeddings is a key challenge in vision-language models (VLMs). The performance of such models hinges on having a good connector that maps visual features generated by a vision encoder to a shared embedding space with the LLM while preserving semantic similarity. Existing connectors, such as multilayer perceptrons (MLPs), often produce out-of-distribution or noisy inputs, leading to misalignment between the modalities. In this work, we propose a novel vision-text alignment method, AlignVLM, that maps visual features to a weighted average of LLM text embeddings. Our approach leverages the linguistic priors encoded by the LLM to ensure that visual features are mapped to regions of the space that the LLM can effectively interpret. AlignVLM is particularly effective for document understanding tasks, where scanned document images must be accurately mapped to their textual content. Our extensive experiments show that AlignVLM achieves state-of-the-art performance compared to prior alignment methods. We provide further analysis demonstrating improved vision-text feature alignment and robustness to noise.

Paperid: 3148, https://arxiv.org/pdf/2501.16411.pdf

Abstract:
Understanding the physical world is a fundamental challenge in embodied AI, critical for enabling agents to perform complex tasks and operate safely in real-world environments. While Vision-Language Models (VLMs) have shown great promise in reasoning and task planning for embodied agents, their ability to comprehend physical phenomena remains extremely limited. To close this gap, we introduce PhysBench, a comprehensive benchmark designed to evaluate VLMs' physical world understanding capability across a diverse set of tasks. PhysBench contains 10,002 entries of interleaved video-image-text data, categorized into four major domains: physical object properties, physical object relationships, physical scene understanding, and physics-based dynamics, further divided into 19 subclasses and 8 distinct capability dimensions. Our extensive experiments, conducted on 75 representative VLMs, reveal that while these models excel in common-sense reasoning, they struggle with understanding the physical world -- likely due to the absence of physical knowledge in their training data and the lack of embedded physical priors. To tackle the shortfall, we introduce PhysAgent, a novel framework that combines the generalization strengths of VLMs with the specialized expertise of vision models, significantly enhancing VLMs' physical understanding across a variety of tasks, including an 18.4\% improvement on GPT-4o. Furthermore, our results demonstrate that enhancing VLMs' physical world understanding capabilities can help embodied agents such as MOKA. We believe that PhysBench and PhysAgent offer valuable insights and contribute to bridging the gap between VLMs and physical world understanding.

Paperid: 3149, https://arxiv.org/pdf/2501.00296.pdf

Abstract:
Our aim is to learn to solve long-horizon decision-making problems in complex robotics domains given low-level skills and a handful of short-horizon demonstrations containing sequences of images. To this end, we focus on learning abstract symbolic world models that facilitate zero-shot generalization to novel goals via planning. A critical component of such models is the set of symbolic predicates that define properties of and relationships between objects. In this work, we leverage pretrained vision language models (VLMs) to propose a large set of visual predicates potentially relevant for decision-making, and to evaluate those predicates directly from camera images. At training time, we pass the proposed predicates and demonstrations into an optimization-based model-learning algorithm to obtain an abstract symbolic world model that is defined in terms of a compact subset of the proposed predicates. At test time, given a novel goal in a novel setting, we use the VLM to construct a symbolic description of the current world state, and then use a search-based planning algorithm to find a sequence of low-level skills that achieves the goal. We demonstrate empirically across experiments in both simulation and the real world that our method can generalize aggressively, applying its learned world model to solve problems with a wide variety of object types, arrangements, numbers of objects, and visual backgrounds, as well as novel goals and much longer horizons than those seen at training time.

Paperid: 3150, https://arxiv.org/pdf/2411.10557.pdf

Abstract:
We present a novel visual instruction tuning strategy to improve the zero-shot task generalization of multimodal large language models by building a firm text-only knowledge base. Existing work lacks sufficient experimentation on the importance of each modality in the instruction tuning stage, often using a majority of vision-language data while keeping text-only data limited and fixing mixtures of modalities. By incorporating diverse text-only data in the visual instruction tuning stage, we vary vision-language data in various controlled experiments to investigate the importance of modality in visual instruction tuning. Our comprehensive evaluation shows that the text-heavy instruction tuning approach is able to perform on-par with traditional vision-heavy mixtures on both modalities across 12 general datasets while using as low as half the total training tokens. We find that simply increasing sufficiently diverse text-only data enables transfer of instruction following ability and domain knowledge across modalities while being more efficient than the vision-language approach.

Paperid: 3151, https://arxiv.org/pdf/2411.08253.pdf

Abstract:
Foundation models trained on internet-scale data, such as Vision-Language Models (VLMs), excel at performing a wide variety of common sense tasks like visual question answering. Despite their impressive capabilities, these models cannot currently be directly applied to challenging robot manipulation problems that require complex and precise continuous reasoning over long horizons. Task and Motion Planning (TAMP) systems can control high-dimensional continuous systems over long horizons via a hybrid search over traditional primitive robot skills. However, these systems require detailed models of how the robot can impact its environment, preventing them from directly interpreting and addressing novel human objectives, for example, an arbitrary natural language goal. We propose deploying VLMs within TAMP systems by having them generate discrete and continuous language-parameterized constraints that enable TAMP to reason about open-world concepts. Specifically, we propose algorithms for VLM partial planning that constrain a TAMP system's discrete temporal search and VLM continuous constraints interpretation to augment the traditional manipulation constraints that TAMP systems seek to satisfy. Experiments demonstrate that our approach -- OWL-TAMP -- outperforms several related baselines, including those that solely use TAMP or VLMs for planning, across several long-horizon manipulation tasks specified directly through natural language. We additionally demonstrate that our approach is compatible with a variety of TAMP systems and can be deployed to solve challenging manipulation tasks on real-world hardware.

Paperid: 3152, https://arxiv.org/pdf/2411.02564.pdf

Abstract:
Instruction tuning constitutes a prevalent technique for tailoring Large Vision Language Models (LVLMs) to meet individual task requirements. To date, most of the existing approaches are confined to single-task adaptation, whereas the requirements in real-world scenarios are inherently varied and continually evolving. Thus an ideal LVLM should sustain continual instruction tuning in the face of stream-task distributions (i.e., different domains, emerging capabilities, and new datasets) while minimizing the forgetting of previously acquired knowledge. To achieve this, we propose a new benchmark for COntinuAl inStruction Tuning on LVLMs (COAST), which encompasses the aforementioned domain-incremental, capability-incremental, and dataset-incremental configurations. In terms of methodology, we propose Continual LLaVA, a rehearsal-free method tailored for continual instruction tuning in LVLMs. To circumvent the additional overhead associated with experience replay, we freeze LVLMs and construct the dual increment embeddings for each input instruction to facilitate parameter-efficient tuning. Specifically, the increment embeddings can be decomposed into two principal components: 1) intrinsic increment embeddings to encode task-specific characteristics. To achieve this, we set up a low-rank pool containing candidate embeddings, from which we select the relevant ones based on their similarity with the user instructions; 2) contextual increment embeddings to investigate the inter-dependencies across tasks. In this regard, the low-rank embeddings chosen in the previous tasks are aggregated via learnable weighted sum to provide complementary hints. Extensive experiments indicate that the proposed Continual LLaVA outperforms previous methods by significantly reducing the forgetting during the continual instruction tuning process.

Paperid: 3153, https://arxiv.org/pdf/2411.02210.pdf

Abstract:
Vision-Language Models (VLMs) have shown significant promise in Visual Question Answering (VQA) tasks by leveraging web-scale multimodal datasets. However, these models often struggle with continual learning due to catastrophic forgetting when adapting to new tasks. As an effective remedy to mitigate catastrophic forgetting, rehearsal strategy uses the data of past tasks upon learning new task. However, such strategy incurs the need of storing past data, which might not be feasible due to hardware constraints or privacy concerns. In this work, we propose the first data-free method that leverages the language generation capability of a VLM, instead of relying on external models, to produce pseudo-rehearsal data for addressing continual VQA. Our proposal, named as GaB, generates pseudo-rehearsal data by posing previous task questions on new task data. Yet, despite being effective, the distribution of generated questions skews towards the most frequently posed questions due to the limited and task-specific training data. To mitigate this issue, we introduce a pseudo-rehearsal balancing module that aligns the generated data towards the ground-truth data distribution using either the question meta-statistics or an unsupervised clustering method. We evaluate our proposed method on two recent benchmarks, \ie VQACL-VQAv2 and CLOVE-function benchmarks. GaB outperforms all the data-free baselines with substantial improvement in maintaining VQA performance across evolving tasks, while being on-par with methods with access to the past data.

Paperid: 3154, https://arxiv.org/pdf/2410.23254.pdf

Abstract:
Generalization to novel object configurations and instances across diverse tasks and environments is a critical challenge in robotics. Keypoint-based representations have been proven effective as a succinct representation for capturing essential object features, and for establishing a reference frame in action prediction, enabling data-efficient learning of robot skills. However, their manual design nature and reliance on additional human labels limit their scalability. In this paper, we propose KALM, a framework that leverages large pre-trained vision-language models (LMs) to automatically generate task-relevant and cross-instance consistent keypoints. KALM distills robust and consistent keypoints across views and objects by generating proposals using LMs and verifies them against a small set of robot demonstration data. Based on the generated keypoints, we can train keypoint-conditioned policy models that predict actions in keypoint-centric frames, enabling robots to generalize effectively across varying object poses, camera views, and object instances with similar functional shapes. Our method demonstrates strong performance in the real world, adapting to different tasks and environments from only a handful of demonstrations while requiring no additional labels. Website: https://kalm-il.github.io/

Paperid: 3155, https://arxiv.org/pdf/2410.19461.pdf

Abstract:
Autonomous agents operating on the graphical user interfaces (GUIs) of various applications hold immense practical value. Unlike the large language model (LLM)-based methods which rely on structured texts and customized backends, the approaches using large vision-language models (LVLMs) are more intuitive and adaptable as they can visually perceive and directly interact with screens, making them indispensable in general scenarios without text metadata and tailored backends. Given the lack of high-quality training data for GUI-related tasks in existing work, this paper aims to enhance the GUI understanding and interacting capabilities of LVLMs through a data-driven approach. We propose EDGE, a general data synthesis framework that automatically generates large-scale, multi-granularity training data from webpages across the Web. Evaluation results on various GUI and agent benchmarks demonstrate that the model trained with the dataset generated through EDGE exhibits superior webpage understanding capabilities, which can then be easily transferred to previously unseen desktop and mobile environments. Our approach significantly reduces the dependence on manual annotations, empowering researchers to harness the vast public resources available on the Web to advance their work. Our source code, the dataset and the model are available at https://anonymous.4open.science/r/EDGE-1CDB.

Paperid: 3156, https://arxiv.org/pdf/2601.17050.pdf

Abstract:
Adverse social interactions, such as bullying, harassment, and other illicit activities, pose significant threats to individual well-being and public safety, leaving profound impacts on physical and mental health. However, these critical events frequently occur in privacy-sensitive environments like restrooms, and changing rooms, where conventional surveillance is prohibited or severely restricted by stringent privacy regulations and ethical concerns. Here, we propose the Single-Pixel Vision-Language Model (SP-VLM), a novel framework that reimagines secure environmental monitoring. It achieves intrinsic privacy-by-design by capturing human dynamics through inherently low-dimensional single-pixel modalities and inferring complex behavioral patterns via seamless vision-language integration. Building on this framework, we demonstrate that single-pixel sensing intrinsically suppresses identity recoverability, rendering state-of-the-art face recognition systems ineffective below a critical sampling rate. We further show that SP-VLM can nonetheless extract meaningful behavioral semantics, enabling robust anomaly detection, people counting, and activity understanding from severely degraded single-pixel observations. Combining these findings, we identify a practical sampling-rate regime in which behavioral intelligence emerges while personal identity remains strongly protected. Together, these results point to a human-rights-aligned pathway for safety monitoring that can support timely intervention without normalizing intrusive surveillance in privacy-sensitive spaces.

Paperid: 3157, https://arxiv.org/pdf/2601.10949.pdf

Abstract:
Medical Vision-Language Models (MedVLMs) excel at perception tasks but struggle with complex clinical reasoning required in real-world scenarios. While reinforcement learning (RL) has been explored to enhance reasoning capabilities, existing approaches face critical mismatches: the scarcity of deep reasoning data, cold-start limits multi-specialty alignment, and standard RL algorithms fail to model clinical reasoning diversity. We propose MMedExpert-R1, a novel reasoning MedVLM that addresses these challenges through domain-specific adaptation and clinical guideline reinforcement. We construct MMedExpert, a high-quality dataset of 10K samples across four specialties with step-by-step reasoning traces. Our Domain-Specific Adaptation (DSA) creates specialty-specific LoRA modules to provide diverse initialization, while Guideline-Based Advantages (GBA) explicitly models different clinical reasoning perspectives to align with real-world diagnostic strategies. Conflict-Aware Capability Integration then merges these specialized experts into a unified agent, ensuring robust multi-specialty alignment. Comprehensive experiments demonstrate state-of-the-art performance, with our 7B model achieving 27.50 on MedXpert-MM and 83.03 on OmniMedVQA, establishing a robust foundation for reliable multimodal medical reasoning systems.

Paperid: 3158, https://arxiv.org/pdf/2601.06847.pdf

Abstract:
Vision-Language Models (VLMs) can generate convincing clinical narratives, yet frequently struggle to visually ground their statements. We posit this limitation arises from the scarcity of high-quality, large-scale clinical referring-localization pairs. To address this, we introduce MedGround, an automated pipeline that transforms segmentation resources into high-quality medical referring grounding data. Leveraging expert masks as spatial anchors, MedGround precisely derives localization targets, extracts shape and spatial cues, and guides VLMs to synthesize natural, clinically grounded queries that reflect morphology and location. To ensure data rigor, a multi-stage verification system integrates strict formatting checks, geometry- and medical-prior rules, and image-based visual judging to filter out ambiguous or visually unsupported samples. Finally, we present MedGround-35K, a novel multimodal medical dataset. Extensive experiments demonstrate that VLMs trained with MedGround-35K consistently achieve improved referring grounding performance, enhance multi-object semantic disambiguation, and exhibit strong generalization to unseen grounding settings. This work highlights MedGround as a scalable, data-driven approach to anchor medical language to verifiable visual evidence. Dataset and code will be released publicly upon acceptance.

Paperid: 3159, https://arxiv.org/pdf/2512.20901.pdf

Abstract:
With the rapid development of Vision-Language Models (VLMs) and the growing demand for their applications, efficient compression of the image inputs has become increasingly important. Existing VLMs predominantly digest and understand high-bitrate compressed images, while their ability to interpret low-bitrate compressed images has yet to be explored by far. In this paper, we introduce the first comprehensive benchmark to evaluate the ability of VLM against compressed images, varying existing widely used image codecs and diverse set of tasks, encompassing over one million compressed images in our benchmark. Next, we analyse the source of performance gap, by categorising the gap from a) the information loss during compression and b) generalisation failure of VLM. We visualize these gaps with concrete examples and identify that for compressed images, only the generalization gap can be mitigated. Finally, we propose a universal VLM adaptor to enhance model performance on images compressed by existing codecs. Consequently, we demonstrate that a single adaptor can improve VLM performance across images with varying codecs and bitrates by 10%-30%. We believe that our benchmark and enhancement method provide valuable insights and contribute toward bridging the gap between VLMs and compressed images.

Paperid: 3160, https://arxiv.org/pdf/2512.10360.pdf

Abstract:
Vision-and-Language Navigation (VLN) requires robots to follow natural language instructions and navigate complex environments without prior maps. While recent vision-language large models demonstrate strong reasoning abilities, they often underperform task-specific panoramic small models in VLN tasks. To address this, we propose CLASH (Collaborative Large-Small Hierarchy), a VLN-CE framework that integrates a reactive small-model planner (RSMP) with a reflective large-model reasoner (RLMR). RSMP adopts a causal-learning-based dual-branch architecture to enhance generalization, while RLMR leverages panoramic visual prompting with chain-of-thought reasoning to support interpretable spatial understanding and navigation. We further introduce an uncertainty-aware collaboration mechanism (UCM) that adaptively fuses decisions from both models. For obstacle avoidance, in simulation, we replace the rule-based controller with a fully learnable point-goal policy, and in real-world deployment, we design a LiDAR-based clustering module for generating navigable waypoints and pair it with an online SLAM-based local controller. CLASH achieves state-of-the-art (SoTA) results (ranking 1-st) on the VLN-CE leaderboard, significantly improving SR and SPL on the test-unseen set over the previous SoTA methods. Real-world experiments demonstrate CLASH's strong robustness, validating its effectiveness in both simulation and deployment scenarios.

Paperid: 3161, https://arxiv.org/pdf/2512.08294.pdf

Abstract:
Despite the promising progress in subject-driven image generation, current models often deviate from the reference identities and struggle in complex scenes with multiple subjects. To address this challenge, we introduce OpenSubject, a video-derived large-scale corpus with 2.5M samples and 4.35M images for subject-driven generation and manipulation. The dataset is built with a four-stage pipeline that exploits cross-frame identity priors. (i) Video Curation. We apply resolution and aesthetic filtering to obtain high-quality clips. (ii) Cross-Frame Subject Mining and Pairing. We utilize vision-language model (VLM)-based category consensus, local grounding, and diversity-aware pairing to select image pairs. (iii) Identity-Preserving Reference Image Synthesis. We introduce segmentation map-guided outpainting to synthesize the input images for subject-driven generation and box-guided inpainting to generate input images for subject-driven manipulation, together with geometry-aware augmentations and irregular boundary erosion. (iv) Verification and Captioning. We utilize a VLM to validate synthesized samples, re-synthesize failed samples based on stage (iii), and then construct short and long captions. In addition, we introduce a benchmark covering subject-driven generation and manipulation, and then evaluate identity fidelity, prompt adherence, manipulation consistency, and background consistency with a VLM judge. Extensive experiments show that training with OpenSubject improves generation and manipulation performance, particularly in complex scenes.

Paperid: 3162, https://arxiv.org/pdf/2512.06662.pdf

Abstract:
People can view the same image differently: they focus on different regions, objects, and details in varying orders and describe them in distinct linguistic styles. This leads to substantial variability in image descriptions. However, existing models for personalized image description focus on linguistic style alone, with no prior work leveraging individual viewing patterns. We address this gap by explicitly modeling personalized viewing behavior as a core factor in description generation. Our method, DEPER (DEscription-PERception persona encoder), learns a subject embedding that captures both linguistic style and viewing behavior, guided by an auxiliary attention-prediction task. A lightweight adapter aligns these embeddings with a frozen vision-language model, enabling few-shot personalization without retraining. Across four datasets spanning diverse viewing tasks and both short and detailed descriptions, DEPER achieves a 24% average improvement, showing that modeling personalized attention produces more human-aligned and high-quality descriptions. We posit that understanding how people see helps predict what they say; modeling human diversity in perception can improve both performance and human alignment in multimodal systems.

Paperid: 3163, https://arxiv.org/pdf/2512.04678.pdf

Abstract:
Efficient streaming video generation is critical for simulating interactive and dynamic worlds. Existing methods distill few-step video diffusion models with sliding window attention, using initial frames as sink tokens to maintain attention performance and reduce error accumulation. However, video frames become overly dependent on these static tokens, resulting in copied initial frames and diminished motion dynamics. To address this, we introduce Reward Forcing, a novel framework with two key designs. First, we propose EMA-Sink, which maintains fixed-size tokens initialized from initial frames and continuously updated by fusing evicted tokens via exponential moving average as they exit the sliding window. Without additional computation cost, EMA-Sink tokens capture both long-term context and recent dynamics, preventing initial frame copying while maintaining long-horizon consistency. Second, to better distill motion dynamics from teacher models, we propose a novel Rewarded Distribution Matching Distillation (Re-DMD). Vanilla distribution matching treats every training sample equally, limiting the model's ability to prioritize dynamic content. Instead, Re-DMD biases the model's output distribution toward high-reward regions by prioritizing samples with greater dynamics rated by a vision-language model. Re-DMD significantly enhances motion quality while preserving data fidelity. We include both quantitative and qualitative experiments to show that Reward Forcing achieves state-of-the-art performance on standard benchmarks while enabling high-quality streaming video generation at 23.1 FPS on a single H100 GPU.

Paperid: 3164, https://arxiv.org/pdf/2511.21541.pdf

Abstract:
Reward feedback learning (ReFL) has proven effective for aligning image generation with human preferences. However, its extension to video generation faces significant challenges. Existing video reward models rely on vision-language models designed for pixel-space inputs, confining ReFL optimization to near-complete denoising steps after computationally expensive VAE decoding. This pixel-space approach incurs substantial memory overhead and increased training time, and its late-stage optimization lacks early-stage supervision, refining only visual quality rather than fundamental motion dynamics and structural coherence. In this work, we show that pre-trained video generation models are naturally suited for reward modeling in the noisy latent space, as they are explicitly designed to process noisy latent representations at arbitrary timesteps and inherently preserve temporal information through their sequential modeling capabilities. Accordingly, we propose Process Reward Feedback Learning~(PRFL), a framework that conducts preference optimization entirely in latent space, enabling efficient gradient backpropagation throughout the full denoising chain without VAE decoding. Extensive experiments demonstrate that PRFL significantly improves alignment with human preferences, while achieving substantial reductions in memory consumption and training time compared to RGB ReFL.

Paperid: 3165, https://arxiv.org/pdf/2511.21042.pdf

Abstract:
Diagnosing lung cancer typically involves physicians identifying lung nodules in Computed tomography (CT) scans and generating diagnostic reports based on their morphological features and medical expertise. Although advancements have been made in using multimodal large language models for analyzing lung CT scans, challenges remain in accurately describing nodule morphology and incorporating medical expertise. These limitations affect the reliability and effectiveness of these models in clinical settings. Collaborative multi-agent systems offer a promising strategy for achieving a balance between generality and precision in medical applications, yet their potential in pathology has not been thoroughly explored. To bridge these gaps, we introduce LungNoduleAgent, an innovative collaborative multi-agent system specifically designed for analyzing lung CT scans. LungNoduleAgent streamlines the diagnostic process into sequential components, improving precision in describing nodules and grading malignancy through three primary modules. The first module, the Nodule Spotter, coordinates clinical detection models to accurately identify nodules. The second module, the Radiologist, integrates localized image description techniques to produce comprehensive CT reports. Finally, the Doctor Agent System performs malignancy reasoning by using images and CT reports, supported by a pathology knowledge base and a multi-agent system framework. Extensive testing on two private datasets and the public LIDC-IDRI dataset indicates that LungNoduleAgent surpasses mainstream vision-language models, agent systems, and advanced expert models. These results highlight the importance of region-level semantic alignment and multi-agent collaboration in diagnosing nodules. LungNoduleAgent stands out as a promising foundational tool for supporting clinical analyses of lung nodules.

Paperid: 3166, https://arxiv.org/pdf/2511.19526.pdf

Abstract:
We propose Perceptual Taxonomy, a structured process of scene understanding that first recognizes objects and their spatial configurations, then infers task-relevant properties such as material, affordance, function, and physical attributes to support goal-directed reasoning. While this form of reasoning is fundamental to human cognition, current vision-language benchmarks lack comprehensive evaluation of this ability and instead focus on surface-level recognition or image-text alignment. To address this gap, we introduce Perceptual Taxonomy, a benchmark for physically grounded visual reasoning. We annotate 3173 objects with four property families covering 84 fine-grained attributes. Using these annotations, we construct a multiple-choice question benchmark with 5802 images across both synthetic and real domains. The benchmark contains 28033 template-based questions spanning four types (object description, spatial reasoning, property matching, and taxonomy reasoning), along with 50 expert-crafted questions designed to evaluate models across the full spectrum of perceptual taxonomy reasoning. Experimental results show that leading vision-language models perform well on recognition tasks but degrade by 10 to 20 percent on property-driven questions, especially those requiring multi-step reasoning over structured attributes. These findings highlight a persistent gap in structured visual understanding and the limitations of current models that rely heavily on pattern matching. We also show that providing in-context reasoning examples from simulated scenes improves performance on real-world and expert-curated questions, demonstrating the effectiveness of perceptual-taxonomy-guided prompting.

Paperid: 3167, https://arxiv.org/pdf/2511.16449.pdf

Abstract:
Vision-Language-Action (VLA) models have shown great promise for embodied AI, yet the heavy computational cost of processing continuous visual streams severely limits their real-time deployment. Token pruning (keeping salient visual tokens and dropping redundant ones) has emerged as an effective approach for accelerating Vision-Language Models (VLMs), offering a solution for efficient VLA. However, these VLM-specific token pruning methods select tokens based solely on semantic salience metrics (e.g., prefill attention), while overlooking the VLA's intrinsic dual-system nature of high-level semantic understanding and low-level action execution. Consequently, these methods bias token retention toward semantic cues, discard critical information for action generation, and significantly degrade VLA performance. To bridge this gap, we propose VLA-Pruner, a versatile plug-and-play VLA-specific token prune method that aligns with the dual-system nature of VLA models and exploits the temporal continuity in robot manipulation. Specifically, VLA-Pruner adopts a dual-level importance criterion for visual token retention: vision-language prefill attention for semantic-level relevance and action decode attention, estimated via temporal smoothing, for action-level importance. Based on this criterion, VLA-Pruner proposes a novel dual-level token selection strategy that adaptively preserves a compact, informative set of visual tokens for both semantic understanding and action execution under given compute budget. Experiments show that VLA-Pruner achieves state-of-the-art performance across multiple VLA architectures and diverse robotic tasks.

Paperid: 3168, https://arxiv.org/pdf/2510.27452.pdf

Abstract:
Scientific illustrations demand both high information density and post-editability. However, current generative models have two major limitations: Frist, image generation models output rasterized images lacking semantic structure, making it impossible to access, edit, or rearrange independent visual components in the images. Second, code-based generation methods (TikZ or SVG), although providing element-level control, force users into the cumbersome cycle of "writing-compiling-reviewing" and lack the intuitiveness of manipulation. Neither of these two approaches can well meet the needs for efficiency, intuitiveness, and iterative modification in scientific creation. To bridge this gap, we introduce VisPainter, a multi-agent framework for scientific illustration built upon the model context protocol. VisPainter orchestrates three specialized modules-a Manager, a Designer, and a Toolbox-to collaboratively produce diagrams compatible with standard vector graphics software. This modular, role-based design allows each element to be explicitly represented and manipulated, enabling true element-level control and any element can be added and modified later. To systematically evaluate the quality of scientific illustrations, we introduce VisBench, a benchmark with seven-dimensional evaluation metrics. It assesses high-information-density scientific illustrations from four aspects: content, layout, visual perception, and interaction cost. To this end, we conducted extensive ablation experiments to verify the rationality of our architecture and the reliability of our evaluation methods. Finally, we evaluated various vision-language models, presenting fair and credible model rankings along with detailed comparisons of their respective capabilities. Additionally, we isolated and quantified the impacts of role division, step control,and description on the quality of illustrations.

Paperid: 3169, https://arxiv.org/pdf/2510.26406.pdf

Abstract:
Reinforcement learning (RL) is widely used to produce robust robotic manipulation policies, but fine-tuning vision-language-action (VLA) models with RL can be unstable due to inaccurate value estimates and sparse supervision at intermediate steps. In contrast, imitation learning (IL) is easy to train but often underperforms due to its offline nature. In this paper, we propose Hi-ORS, a simple yet effective post-training method that utilizes rejection sampling to achieve both training stability and high robustness. Hi-ORS stabilizes value estimation by filtering out negatively rewarded samples during online fine-tuning, and adopts a reward-weighted supervised training objective to provide dense intermediate-step supervision. For systematic study, we develop an asynchronous inference-training framework that supports flexible online human-in-the-loop corrections, which serve as explicit guidance for learning error-recovery behaviors. Across three real-world tasks and two embodiments, Hi-ORS fine-tunes a pi-base policy to master contact-rich manipulation in just 1.5 hours of real-world training, outperforming RL and IL baselines by a substantial margin in both effectiveness and efficiency. Notably, the fine-tuned policy exhibits strong test-time scalability by reliably executing complex error-recovery behaviors to achieve better performance.

Paperid: 3170, https://arxiv.org/pdf/2510.21121.pdf

Abstract:
We present Generalizable Hierarchical Skill Learning (GSL), a novel framework for hierarchical policy learning that significantly improves policy generalization and sample efficiency in robot manipulation. One core idea of GSL is to use object-centric skills as an interface that bridges the high-level vision-language model and the low-level visual-motor policy. Specifically, GSL decomposes demonstrations into transferable and object-canonicalized skill primitives using foundation models, ensuring efficient low-level skill learning in the object frame. At test time, the skill-object pairs predicted by the high-level agent are fed to the low-level module, where the inferred canonical actions are mapped back to the world frame for execution. This structured yet flexible design leads to substantial improvements in sample efficiency and generalization of our method across unseen spatial arrangements, object appearances, and task compositions. In simulation, GSL trained with only 3 demonstrations per task outperforms baselines trained with 30 times more data by 15.5 percent on unseen tasks. In real-world experiments, GSL also surpasses the baseline trained with 10 times more data.

Paperid: 3171, https://arxiv.org/pdf/2509.23236.pdf

Abstract:
Vision-language models often hallucinate details, generating non-existent objects or inaccurate attributes that compromise output reliability. Existing methods typically address these issues via extensive human annotations or external supervision from more powerful models. In this work, we present a novel framework that leverages the model's self-consistency between long responses and short answers to generate preference pairs for training. We observe that short binary questions tend to yield highly reliable responses, which can be used to query the target model to evaluate and rank its generated responses. Specifically, we design a self-reflection pipeline where detailed model responses are compared against concise binary answers, and inconsistency signals are utilized to automatically curate high-quality training data without human annotations or external model-based supervision. By relying solely on self-consistency rather than external supervision, our method offers a scalable and efficient solution that effectively reduces hallucinations using unlabeled data. Extensive experiments on multiple benchmarks, i.e., AMBER, MultiObject-Hal (ROPE), Object HalBench, and MMHal-Bench, demonstrate significant improvements in factual grounding and reliability. Moreover, our approach maintains robust instruction-following ability, as evidenced by enhanced performance on LLaVA-Bench and MMBench.

Paperid: 3172, https://arxiv.org/pdf/2509.23105.pdf

Abstract:
Remote sensing change understanding (RSCU) is essential for analyzing remote sensing images and understanding how human activities affect the environment. However, existing datasets lack deep understanding and interactions in the diverse change captioning, counting, and localization tasks. To tackle these gaps, we construct ChangeIMTI, a new large-scale interactive multi-task instruction dataset that encompasses four complementary tasks including change captioning, binary change classification, change counting, and change localization. Building upon this new dataset, we further design a novel vision-guided vision-language model (ChangeVG) with dual-granularity awareness for bi-temporal remote sensing images (i.e., two remote sensing images of the same area at different times). The introduced vision-guided module is a dual-branch architecture that synergistically combines fine-grained spatial feature extraction with high-level semantic summarization. These enriched representations further serve as the auxiliary prompts to guide large vision-language models (VLMs) (e.g., Qwen2.5-VL-7B) during instruction tuning, thereby facilitating the hierarchical cross-modal learning. We extensively conduct experiments across four tasks to demonstrate the superiority of our approach. Remarkably, on the change captioning task, our method outperforms the strongest method Semantic-CC by 1.39 points on the comprehensive S*m metric, which integrates the semantic similarity and descriptive accuracy to provide an overall evaluation of change caption. Moreover, we also perform a series of ablation studies to examine the critical components of our method.

Paperid: 3173, https://arxiv.org/pdf/2509.00373.pdf

Abstract:
Vision Language Models (VLMs) have demonstrated impressive capabilities in integrating visual and textual information for understanding and reasoning, but remain highly vulnerable to adversarial attacks. While activation steering has emerged as a promising defence, existing approaches often rely on task-specific contrastive prompts to extract harmful directions, which exhibit suboptimal performance and can degrade visual grounding performance. To address these limitations, we propose \textit{Sequence-Level Preference Optimization} for VLM (\textit{SPO-VLM}), a novel two-stage defense framework that combines activation-level intervention with policy-level optimization to enhance model robustness. In \textit{Stage I}, we compute adaptive layer-specific steering vectors from diverse data sources, enabling generalized suppression of harmful behaviors during inference. In \textit{Stage II}, we refine these steering vectors through a sequence-level preference optimization process. This stage integrates automated toxicity assessment, as well as visual-consistency rewards based on caption-image alignment, to achieve safe and semantically grounded text generation. The two-stage structure of SPO-VLM balances efficiency and effectiveness by combining a lightweight mitigation foundation in Stage I with deeper policy refinement in Stage II. Extensive experiments shown SPO-VLM enhances safety against attacks via activation steering and preference optimization, while maintaining strong performance on benign tasks without compromising visual understanding capabilities. We will release our code, model weights, and evaluation toolkit to support reproducibility and future research. \textcolor{red}{Warning: This paper may contain examples of offensive or harmful text and images.}

Paperid: 3174, https://arxiv.org/pdf/2508.20840.pdf

Abstract:
While video-generation-based embodied world models have gained increasing attention, their reliance on large-scale embodied interaction data remains a key bottleneck. The scarcity, difficulty of collection, and high dimensionality of embodied data fundamentally limit the alignment granularity between language and actions and exacerbate the challenge of long-horizon video generation--hindering generative models from achieving a "GPT moment" in the embodied domain. There is a naive observation: the diversity of embodied data far exceeds the relatively small space of possible primitive motions. Based on this insight, we propose a novel paradigm for world modeling--Primitive Embodied World Models (PEWM). By restricting video generation to fixed short horizons, our approach 1) enables fine-grained alignment between linguistic concepts and visual representations of robotic actions, 2) reduces learning complexity, 3) improves data efficiency in embodied data collection, and 4) decreases inference latency. By equipping with a modular Vision-Language Model (VLM) planner and a Start-Goal heatmap Guidance mechanism (SGG), PEWM further enables flexible closed-loop control and supports compositional generalization of primitive-level policies over extended, complex tasks. Our framework leverages the spatiotemporal vision priors in video models and the semantic awareness of VLMs to bridge the gap between fine-grained physical interaction and high-level reasoning, paving the way toward scalable, interpretable, and general-purpose embodied intelligence.

Paperid: 3175, https://arxiv.org/pdf/2508.05383.pdf

Abstract:
Existing Vision-Language Models often struggle with complex, multi-question reasoning tasks where partial correctness is crucial for effective learning. Traditional reward mechanisms, which provide a single binary score for an entire response, are too coarse to guide models through intricate problems with multiple sub-parts. To address this, we introduce StructVRM, a method that aligns multimodal reasoning with Structured and Verifiable Reward Models. At its core is a model-based verifier trained to provide fine-grained, sub-question-level feedback, assessing semantic and mathematical equivalence rather than relying on rigid string matching. This allows for nuanced, partial credit scoring in previously intractable problem formats. Extensive experiments demonstrate the effectiveness of StructVRM. Our trained model, Seed-StructVRM, achieves state-of-the-art performance on six out of twelve public multimodal benchmarks and our newly curated, high-difficulty STEM-Bench. The success of StructVRM validates that training with structured, verifiable rewards is a highly effective approach for advancing the capabilities of multimodal models in complex, real-world reasoning domains.

Paperid: 3176, https://arxiv.org/pdf/2507.22904.pdf

Abstract:
Scientific sketches (e.g., models) offer a powerful lens into students' conceptual understanding, yet AI-powered automated assessment of such free-form, visually diverse artifacts remains a critical challenge. Existing solutions often treat sketch evaluation as either an image classification task or monolithic vision-language models, which lack interpretability, pedagogical alignment, and adaptability across cognitive levels. To address these limitations, we present SketchMind, a cognitively grounded, multi-agent framework for evaluating and improving student-drawn scientific sketches. SketchMind comprises modular agents responsible for rubric parsing, sketch perception, cognitive alignment, and iterative feedback with sketch modification, enabling personalized and transparent evaluation. We evaluate SketchMind on a curated dataset of 3,575 student-generated sketches across six science assessment items with different highest order of Bloom's level that require students to draw models to explain phenomena. Compared to baseline GPT-4o performance without SRG (average accuracy: 55.6%), and with SRG integration achieves 77.1% average accuracy (+21.4% average absolute gain). We also demonstrate that multi-agent orchestration with SRG enhances SketchMind performance, for example, GPT-4.1 gains an average 8.9% increase in sketch prediction accuracy, outperforming single-agent pipelines across all items. Human evaluators rated the feedback and co-created sketches generated by \textsc{SketchMind} with GPT-4.1, which achieved an average of 4.1 out of 5, significantly higher than those of baseline models (e.g., 2.3 for GPT-4o). Experts noted the system's potential to meaningfully support conceptual growth through guided revision. Our code and (pending approval) dataset will be released to support reproducibility and future research in AI-driven education.

Paperid: 3177, https://arxiv.org/pdf/2507.20342.pdf

Abstract:
Integrating large language models (LLMs) into autonomous driving motion planning has recently emerged as a promising direction, offering enhanced interpretability, better controllability, and improved generalization in rare and long-tail scenarios. However, existing methods often rely on abstracted perception or map-based inputs, missing crucial visual context, such as fine-grained road cues, accident aftermath, or unexpected obstacles, which are essential for robust decision-making in complex driving environments. To bridge this gap, we propose VLMPlanner, a hybrid framework that combines a learning-based real-time planner with a vision-language model (VLM) capable of reasoning over raw images. The VLM processes multi-view images to capture rich, detailed visual information and leverages its common-sense reasoning capabilities to guide the real-time planner in generating robust and safe trajectories. Furthermore, we develop the Context-Adaptive Inference Gate (CAI-Gate) mechanism that enables the VLM to mimic human driving behavior by dynamically adjusting its inference frequency based on scene complexity, thereby achieving an optimal balance between planning performance and computational efficiency. We evaluate our approach on the large-scale, challenging nuPlan benchmark, with comprehensive experimental results demonstrating superior planning performance in scenarios with intricate road conditions and dynamic elements. Code will be available.

Paperid: 3178, https://arxiv.org/pdf/2506.22756.pdf

Abstract:
The development of generalist robot manipulation policies has seen significant progress, driven by large-scale demonstration data across diverse environments. However, the high cost and inefficiency of collecting real-world demonstrations hinder the scalability of data acquisition. While existing simulation platforms enable controlled environments for robotic learning, the challenge of bridging the sim-to-real gap remains. To address these challenges, we propose RoboPearls, an editable video simulation framework for robotic manipulation. Built on 3D Gaussian Splatting (3DGS), RoboPearls enables the construction of photo-realistic, view-consistent simulations from demonstration videos, and supports a wide range of simulation operators, including various object manipulations, powered by advanced modules like Incremental Semantic Distillation (ISD) and 3D regularized NNFM Loss (3D-NNFM). Moreover, by incorporating large language models (LLMs), RoboPearls automates the simulation production process in a user-friendly manner through flexible command interpretation and execution. Furthermore, RoboPearls employs a vision-language model (VLM) to analyze robotic learning issues to close the simulation loop for performance enhancement. To demonstrate the effectiveness of RoboPearls, we conduct extensive experiments on multiple datasets and scenes, including RLBench, COLOSSEUM, Ego4D, Open X-Embodiment, and a real-world robot, which demonstrate our satisfactory simulation performance.

Paperid: 3179, https://arxiv.org/pdf/2506.06072.pdf

Abstract:
We present the B-spline Encoded Action Sequence Tokenizer (BEAST), a novel action tokenizer that encodes action sequences into compact discrete or continuous tokens using B-splines. In contrast to existing action tokenizers based on vector quantization or byte pair encoding, BEAST requires no separate tokenizer training and consistently produces tokens of uniform length, enabling fast action sequence generation via parallel decoding. Leveraging our B-spline formulation, BEAST inherently ensures generating smooth trajectories without discontinuities between adjacent segments. We extensively evaluate BEAST by integrating it with three distinct model architectures: a Variational Autoencoder (VAE) with continuous tokens, a decoder-only Transformer with discrete tokens, and Florence-2, a pretrained Vision-Language Model with an encoder-decoder architecture, demonstrating BEAST's compatibility and scalability with large pretrained models. We evaluate BEAST across three established benchmarks consisting of 166 simulated tasks and on three distinct robot settings with a total of 8 real-world tasks. Experimental results demonstrate that BEAST (i) significantly reduces both training and inference computational costs, and (ii) consistently generates smooth, high-frequency control signals suitable for continuous control tasks while (iii) reliably achieves competitive task success rates compared to state-of-the-art methods.

Paperid: 3180, https://arxiv.org/pdf/2505.23242.pdf

Abstract:
Chart question answering (CQA) has become a critical multimodal task for evaluating the reasoning capabilities of vision-language models. While early approaches have shown promising performance by focusing on visual features or leveraging large-scale pre-training, most existing evaluations rely on rigid output formats and objective metrics, thus ignoring the complex, real-world demands of practical chart analysis. In this paper, we introduce ChartMind, a new benchmark designed for complex CQA tasks in real-world settings. ChartMind covers seven task categories, incorporates multilingual contexts, supports open-domain textual outputs, and accommodates diverse chart formats, bridging the gap between real-world applications and traditional academic benchmarks. Furthermore, we propose a context-aware yet model-agnostic framework, ChartLLM, that focuses on extracting key contextual elements, reducing noise, and enhancing the reasoning accuracy of multimodal large language models. Extensive evaluations on ChartMind and three representative public benchmarks with 14 mainstream multimodal models show our framework significantly outperforms the previous three common CQA paradigms: instruction-following, OCR-enhanced, and chain-of-thought, highlighting the importance of flexible chart understanding for real-world CQA. These findings suggest new directions for developing more robust chart reasoning in future research.

Paperid: 3181, https://arxiv.org/pdf/2505.17509.pdf

Abstract:
Large pre-trained Vision Language Models (VLMs) have excellent generalization capabilities but are highly susceptible to adversarial examples, presenting potential security risks. To improve the robustness of VLMs against adversarial examples, adversarial prompt tuning methods are proposed to align the text feature with the adversarial image feature without changing model parameters. However, when facing various adversarial attacks, a single learnable text prompt has insufficient generalization to align well with all adversarial image features, which finally leads to the overfitting phenomenon. To address the above challenge, in this paper, we empirically find that increasing the number of learned prompts can bring more robustness improvement than a longer prompt. Then we propose an adversarial tuning method named Adversarial Mixture Prompt Tuning (AMPT) to enhance the generalization towards various adversarial attacks for VLMs. AMPT aims to learn mixture text prompts to obtain more robust text features. To further enhance the adaptability, we propose a conditional weight router based on the input adversarial image to predict the mixture weights of multiple learned prompts, which helps obtain sample-specific aggregated text features aligning with different adversarial image features. A series of experiments show that our method can achieve better adversarial robustness than state-of-the-art methods on 11 datasets under different experimental settings.

Paperid: 3182, https://arxiv.org/pdf/2505.12835.pdf

Abstract:
Unmanned Aerial Vehicle (UAV) Vision-and-Language Navigation (VLN) is vital for applications such as disaster response, logistics delivery, and urban inspection. However, existing methods often struggle with insufficient multimodal fusion, weak generalization, and poor interpretability. To address these challenges, we propose FlightGPT, a novel UAV VLN framework built upon Vision-Language Models (VLMs) with powerful multimodal perception capabilities. We design a two-stage training pipeline: first, Supervised Fine-Tuning (SFT) using high-quality demonstrations to improve initialization and structured reasoning; then, Group Relative Policy Optimization (GRPO) algorithm, guided by a composite reward that considers goal accuracy, reasoning quality, and format compliance, to enhance generalization and adaptability. Furthermore, FlightGPT introduces a Chain-of-Thought (CoT)-based reasoning mechanism to improve decision interpretability. Extensive experiments on the city-scale dataset CityNav demonstrate that FlightGPT achieves state-of-the-art performance across all scenarios, with a 9.22\% higher success rate than the strongest baseline in unseen environments. Our implementation is publicly available.

Paperid: 3183, https://arxiv.org/pdf/2505.01583.pdf

Abstract:
Understanding causal event relationships and achieving fine-grained temporal grounding in videos remain challenging for vision-language models. Existing methods either compress video tokens to reduce temporal resolution, or treat videos as unsegmented streams, which obscures fine-grained event boundaries and limits the modeling of causal dependencies. We propose TEMPURA (Temporal Event Masked Prediction and Understanding for Reasoning in Action), a two-stage training framework that enhances video temporal understanding. TEMPURA first applies masked event prediction reasoning to reconstruct missing events and generate step-by-step causal explanations from dense event annotations, drawing inspiration from effective infilling techniques. TEMPURA then learns to perform video segmentation and dense captioning to decompose videos into non-overlapping events with detailed, timestamp-aligned descriptions. We train TEMPURA on VER, a large-scale dataset curated by us that comprises 1M training instances and 500K videos with temporally aligned event descriptions and structured reasoning steps. Experiments on temporal grounding and highlight detection benchmarks demonstrate that TEMPURA outperforms strong baseline models, confirming that integrating causal reasoning with fine-grained temporal segmentation leads to improved video understanding.

Paperid: 3184, https://arxiv.org/pdf/2504.20024.pdf

Abstract:
Despite recent advances on multi-modal models, 3D spatial reasoning remains a challenging task for state-of-the-art open-source and proprietary models. Recent studies explore data-driven approaches and achieve enhanced spatial reasoning performance by fine-tuning models on 3D-related visual question-answering data. However, these methods typically perform spatial reasoning in an implicit manner and often fail on questions that are trivial to humans, even with long chain-of-thought reasoning. In this work, we introduce SpatialReasoner, a novel large vision-language model (LVLM) that addresses 3D spatial reasoning with explicit 3D representations shared between multiple stages--3D perception, computation, and reasoning. Explicit 3D representations provide a coherent interface that supports advanced 3D spatial reasoning and improves the generalization ability to novel question types. Furthermore, by analyzing the explicit 3D representations in multi-step reasoning traces of SpatialReasoner, we study the factual errors and identify key shortcomings of current LVLMs. Results show that our SpatialReasoner achieves improved performance on a variety of spatial reasoning benchmarks, outperforming Gemini 2.0 by 9.2% on 3DSRBench, and generalizes better when evaluating on novel 3D spatial reasoning questions. Our study bridges the 3D parsing capabilities of prior visual foundation models with the powerful reasoning abilities of large language models, opening new directions for 3D spatial reasoning.

Paperid: 3185, https://arxiv.org/pdf/2504.14919.pdf

Abstract:
Zero-shot anomaly detection (ZSAD) aims to identify anomalies in unseen categories by leveraging CLIP's zero-shot capabilities to match text prompts with visual features. A key challenge in ZSAD is learning general prompts stably and utilizing them effectively, while maintaining both generalizability and category specificity. Although general prompts have been explored in prior works, achieving their stable optimization and effective deployment remains a significant challenge. In this work, we propose GenCLIP, a novel framework that learns and leverages general prompts more effectively through multi-layer prompting and dual-branch inference. Multi-layer prompting integrates category-specific visual cues from different CLIP layers, enriching general prompts with more comprehensive and robust feature representations. By combining general prompts with multi-layer visual features, our method further enhances its generalization capability. To balance specificity and generalization, we introduce a dual-branch inference strategy, where a vision-enhanced branch captures fine-grained category-specific features, while a query-only branch prioritizes generalization. The complementary outputs from both branches improve the stability and reliability of anomaly detection across unseen categories. Additionally, we propose an adaptive text prompt filtering mechanism, which removes irrelevant or atypical class names not encountered during CLIP's training, ensuring that only meaningful textual inputs contribute to the final vision-language alignment.

Paperid: 3186, https://arxiv.org/pdf/2504.09828.pdf

Abstract:
Semi-supervised learning (SSL) has achieved significant progress by leveraging both labeled data and unlabeled data. Existing SSL methods overlook a common real-world scenario when labeled data is extremely scarce, potentially as limited as a single labeled sample in the dataset. General SSL approaches struggle to train effectively from scratch under such constraints, while methods utilizing pre-trained models often fail to find an optimal balance between leveraging limited labeled data and abundant unlabeled data. To address this challenge, we propose Firstly Adapt, Then catEgorize (FATE), a novel SSL framework tailored for scenarios with extremely limited labeled data. At its core, the two-stage prompt tuning paradigm FATE exploits unlabeled data to compensate for scarce supervision signals, then transfers to downstream tasks. Concretely, FATE first adapts a pre-trained model to the feature distribution of downstream data using volumes of unlabeled samples in an unsupervised manner. It then applies an SSL method specifically designed for pre-trained models to complete the final classification task. FATE is designed to be compatible with both vision and vision-language pre-trained models. Extensive experiments demonstrate that FATE effectively mitigates challenges arising from the scarcity of labeled samples in SSL, achieving an average performance improvement of 33.74% across seven benchmarks compared to state-of-the-art SSL methods. Code is available at https://anonymous.4open.science/r/Semi-supervised-learning-BA72.

Paperid: 3187, https://arxiv.org/pdf/2504.08837.pdf

Abstract:
Recently, slow-thinking systems like GPT-o1 and DeepSeek-R1 have demonstrated great potential in solving challenging problems through explicit reflection. They significantly outperform the best fast-thinking models, such as GPT-4o, on various math and science benchmarks. However, their multimodal reasoning capabilities remain on par with fast-thinking models. For instance, GPT-o1's performance on benchmarks like MathVista, MathVerse, and MathVision is similar to fast-thinking models. In this paper, we aim to enhance the slow-thinking capabilities of vision-language models using reinforcement learning (without relying on distillation) to advance the state of the art. First, we adapt the GRPO algorithm with a novel technique called Selective Sample Replay (SSR) to address the vanishing advantages problem. While this approach yields strong performance, the resulting RL-trained models exhibit limited self-reflection or self-verification. To further encourage slow-thinking, we introduce Forced Rethinking, which appends a rethinking trigger token to the end of rollouts in RL training, explicitly enforcing a self-reflection reasoning step. By combining these two techniques, our model, VL-Rethinker, advances state-of-the-art scores on MathVista, MathVerse to achieve 80.4%, 63.5% respectively. VL-Rethinker also achieves open-source SoTA on multi-disciplinary benchmarks such as MathVision, MMMU-Pro, EMMA, and MEGA-Bench, narrowing the gap with OpenAI-o1. Our empirical results show the effectiveness of our approaches.

Paperid: 3188, https://arxiv.org/pdf/2503.16856.pdf

Abstract:
Fully comprehending scientific papers by machines reflects a high level of Artificial General Intelligence, requiring the ability to reason across fragmented and heterogeneous sources of information, presenting a complex and practically significant challenge. While Vision-Language Models (VLMs) have made remarkable strides in various tasks, particularly those involving reasoning with evidence source from single image or text page, their ability to use cross-source information for reasoning remains an open problem. This work presents MMCR, a high-difficulty benchmark designed to evaluate VLMs' capacity for reasoning with cross-source information from scientific papers. The benchmark comprises 276 high-quality questions, meticulously annotated by humans across 7 subjects and 10 task types. Experiments with 18 VLMs demonstrate that cross-source reasoning presents a substantial challenge for existing models. Notably, even the top-performing model, GPT-4o, achieved only 48.55% overall accuracy, with only 20% accuracy in multi-table comprehension tasks, while the second-best model, Qwen2.5-VL-72B, reached 39.86% overall accuracy. Furthermore, we investigated the impact of the Chain-of-Thought (CoT) technique on cross-source reasoning and observed a detrimental effect on small models, whereas larger models demonstrated substantially enhanced performance. These results highlight the pressing need to develop VLMs capable of effectively utilizing cross-source information for reasoning.

Paperid: 3189, https://arxiv.org/pdf/2503.06993.pdf

Abstract:
Effectively handling the co-occurrence of non-IID data and long-tailed distributions remains a critical challenge in federated learning. While fine-tuning vision-language models (VLMs) like CLIP has shown to be promising in addressing non-IID data challenges, this approach leads to severe degradation of tail classes in federated long-tailed scenarios. Under the composite effects of strong non-IID data distribution and long-tailed class imbalances, VLM fine-tuning may even fail to yield any improvement. To address this issue, we propose Class-Aware Prompt Learning for Federated Long-tailed Learning (CAPT), a novel framework that leverages a pre-trained VLM to effectively handle both data heterogeneity and long-tailed distributions. CAPT introduces a dual-prompt mechanism that synergizes general and class-aware prompts, enabling the framework to capture global trends while preserving class-specific knowledge. To better aggregate and share knowledge across clients, we introduce a heterogeneity-aware client clustering strategy that groups clients based on their data distributions, enabling efficient collaboration and knowledge sharing. Extensive experiments on various long-tailed datasets with different levels of data heterogeneity demonstrate that CAPT significantly improves tail class performance without compromising overall accuracy, outperforming state-of-the-art methods in federated long-tailed learning scenarios.

Paperid: 3190, https://arxiv.org/pdf/2503.02393.pdf

Abstract:
Vision-language models (VLMs) like CLIP (Contrastive Language-Image Pre-Training) have seen remarkable success in visual recognition, highlighting the increasing need to safeguard the intellectual property (IP) of well-trained models. Effective IP protection extends beyond ensuring authorized usage; it also necessitates restricting model deployment to authorized data domains, particularly when the model is fine-tuned for specific target domains. However, current IP protection methods often rely solely on the visual backbone, which may lack sufficient semantic richness. To bridge this gap, we introduce IP-CLIP, a lightweight IP protection strategy tailored to CLIP, employing a prompt-based learning approach. By leveraging the frozen visual backbone of CLIP, we extract both image style and content information, incorporating them into the learning of IP prompt. This strategy acts as a robust barrier, effectively preventing the unauthorized transfer of features from authorized domains to unauthorized ones. Additionally, we propose a style-enhancement branch that constructs feature banks for both authorized and unauthorized domains. This branch integrates self-enhanced and cross-domain features, further strengthening IP-CLIP's capability to block features from unauthorized domains. Finally, we present new three metrics designed to better balance the performance degradation of authorized and unauthorized domains. Comprehensive experiments in various scenarios demonstrate its promising potential for application in IP protection tasks for VLMs.

Paperid: 3191, https://arxiv.org/pdf/2502.09649.pdf

Abstract:
Visuomotor imitation learning enables embodied agents to effectively acquire manipulation skills from video demonstrations and robot proprioception. However, as scene complexity and visual distractions increase, existing methods that perform well in simple scenes tend to degrade in performance. To address this challenge, we introduce Imit Diff, a semanstic guided diffusion transformer with dual resolution fusion for imitation learning. Our approach leverages prior knowledge from vision language foundation models to translate high-level semantic instruction into pixel-level visual localization. This information is explicitly integrated into a multi-scale visual enhancement framework, constructed with a dual resolution encoder. Additionally, we introduce an implementation of Consistency Policy within the diffusion transformer architecture to improve both real-time performance and motion smoothness in embodied agent control.We evaluate Imit Diff on several challenging real-world tasks. Due to its task-oriented visual localization and fine-grained scene perception, it significantly outperforms state-of-the-art methods, especially in complex scenes with visual distractions, including zero-shot experiments focused on visual distraction and category generalization. The code will be made publicly available.

Paperid: 3192, https://arxiv.org/pdf/2501.19339.pdf

Abstract:
Recent agentic language models increasingly need to interact directly with real-world environments containing intertwined visual and textual information through raw camera pixels, rather than relying on separate image and tokenized text processing, underscoring the necessity of a unified perception paradigm. To close this gap, we explore this idea through Perceive Everything as Pixels (PEAP) and release PixelWorld, a benchmark that renders natural-language, tabular, mathematical and diagrammatic inputs into a single pixel space. Experiments show that PEAP attains competitive accuracy on semantic-understanding tasks, indicating that a vision transformer can capture global textual semantics without explicit tokens. In contrast, reasoning-intensive benchmarks (math and code) exhibit sharp performance drops; however, Chain-of-Thought prompting partially mitigates this gap, hinting that explicit reasoning traces compensate for the missing token structure. We also find that when visual and textual information are closely integrated, representing everything as pixels reduces preprocessing complexity and avoids misalignment issues that often arise in separate pipelines. PixelWorld therefore serves as a practical benchmark for evaluating unified vision-language models and supports broader exploration of PEAP across diverse tasks.

Paperid: 3193, https://arxiv.org/pdf/2501.10057.pdf

Abstract:
Vision-language models (VLMs), which process image and text inputs, are increasingly integrated into chat assistants and other consumer AI applications. Without proper safeguards, however, VLMs may give harmful advice (e.g. how to self-harm) or encourage unsafe behaviours (e.g. to consume drugs). Despite these clear hazards, little work so far has evaluated VLM safety and the novel risks created by multimodal inputs. To address this gap, we introduce MSTS, a Multimodal Safety Test Suite for VLMs. MSTS comprises 400 test prompts across 40 fine-grained hazard categories. Each test prompt consists of a text and an image that only in combination reveal their full unsafe meaning. With MSTS, we find clear safety issues in several open VLMs. We also find some VLMs to be safe by accident, meaning that they are safe because they fail to understand even simple test prompts. We translate MSTS into ten languages, showing non-English prompts to increase the rate of unsafe model responses. We also show models to be safer when tested with text only rather than multimodal prompts. Finally, we explore the automation of VLM safety assessments, finding even the best safety classifiers to be lacking.

Paperid: 3194, https://arxiv.org/pdf/2412.17544.pdf

Abstract:
The emergence of Vision-Language Models (VLMs) is a significant advancement in integrating computer vision with Large Language Models (LLMs) to enhance multi-modal machine learning capabilities. However, this progress has also made VLMs vulnerable to sophisticated adversarial attacks, raising concerns about their reliability. The objective of this paper is to assess the resilience of VLMs against jailbreak attacks that can compromise model safety compliance and result in harmful outputs. To evaluate a VLM's ability to maintain its robustness against adversarial input perturbations, we propose a novel metric called the \textbf{Retention Score}. Retention Score is a multi-modal evaluation metric that includes Retention-I and Retention-T scores for quantifying jailbreak risks in visual and textual components of VLMs. Our process involves generating synthetic image-text pairs using a conditional diffusion model. These pairs are then predicted for toxicity score by a VLM alongside a toxicity judgment classifier. By calculating the margin in toxicity scores, we can quantify the robustness of the VLM in an attack-agnostic manner. Our work has four main contributions. First, we prove that Retention Score can serve as a certified robustness metric. Second, we demonstrate that most VLMs with visual components are less robust against jailbreak attacks than the corresponding plain VLMs. Additionally, we evaluate black-box VLM APIs and find that the security settings in Google Gemini significantly affect the score and robustness. Moreover, the robustness of GPT4V is similar to the medium settings of Gemini. Finally, our approach offers a time-efficient alternative to existing adversarial attack methods and provides consistent model robustness rankings when evaluated on VLMs including MiniGPT-4, InstructBLIP, and LLaVA.

Paperid: 3195, https://arxiv.org/pdf/2412.09501.pdf

Abstract:
As Multi-modal Large Language Models (MLLMs) evolve, expanding beyond single-domain capabilities is essential to meet the demands for more versatile and efficient AI. However, previous omni-models have insufficiently explored speech, neglecting its integration with multi-modality. We introduce Lyra, an efficient MLLM that enhances multimodal abilities, including advanced long-speech comprehension, sound understanding, cross-modality efficiency, and seamless speech interaction. To achieve efficiency and speech-centric capabilities, Lyra employs three strategies: (1) leveraging existing open-source large models and a proposed multi-modality LoRA to reduce training costs and data requirements; (2) using a latent multi-modality regularizer and extractor to strengthen the relationship between speech and other modalities, thereby enhancing model performance; and (3) constructing a high-quality, extensive dataset that includes 1.5M multi-modal (language, vision, audio) data samples and 12K long speech samples, enabling Lyra to handle complex long speech inputs and achieve more robust omni-cognition. Compared to other omni-methods, Lyra achieves state-of-the-art performance on various vision-language, vision-speech, and speech-language benchmarks, while also using fewer computational resources and less training data.

Paperid: 3196, https://arxiv.org/pdf/2412.06632.pdf

Abstract:
Mitigating biases in computer vision models is an essential step towards the trustworthiness of artificial intelligence models. Existing bias mitigation methods focus on a small set of predefined biases, limiting their applicability in visual datasets where multiple, possibly unknown biases exist. To address this limitation, we introduce MAVias, an open-set bias mitigation approach leveraging foundation models to discover spurious associations between visual attributes and target classes. MAVias first captures a wide variety of visual features in natural language via a foundation image tagging model, and then leverages a large language model to select those visual features defining the target class, resulting in a set of language-coded potential visual biases. We then translate this set of potential biases into vision-language embeddings and introduce an in-processing bias mitigation approach to prevent the model from encoding information related to them. Our experiments on diverse datasets, including CelebA, Waterbirds, ImageNet, and UrbanCars, show that MAVias effectively detects and mitigates a wide range of biases in visual recognition tasks outperforming current state-of-the-art.

Paperid: 3197, https://arxiv.org/pdf/2412.02479.pdf

Abstract:
With the rise of deep learning, facial recognition technology has seen extensive research and rapid development. Although facial recognition is considered a mature technology, we find that existing open-source models and commercial algorithms lack robustness in certain complex Out-of-Distribution (OOD) scenarios, raising concerns about the reliability of these systems. In this paper, we introduce OODFace, which explores the OOD challenges faced by facial recognition models from two perspectives: common corruptions and appearance variations. We systematically design 30 OOD scenarios across 9 major categories tailored for facial recognition. By simulating these challenges on public datasets, we establish three robustness benchmarks: LFW-C/V, CFP-FP-C/V, and YTF-C/V. We then conduct extensive experiments on 19 facial recognition models and 3 commercial APIs, along with extended physical experiments on face masks to assess their robustness. Next, we explore potential solutions from two perspectives: defense strategies and Vision-Language Models (VLMs). Based on the results, we draw several key insights, highlighting the vulnerability of facial recognition systems to OOD data and suggesting possible solutions. Additionally, we offer a unified toolkit that includes all corruption and variation types, easily extendable to other datasets. We hope that our benchmarks and findings can provide guidance for future improvements in facial recognition model robustness.

Paperid: 3198, https://arxiv.org/pdf/2410.05191.pdf

Abstract:
Building on the advancements of Large Language Models (LLMs) and Vision Language Models (VLMs), recent research has introduced Vision-Language-Action (VLA) models as an integrated solution for robotic manipulation tasks. These models take camera images and natural language task instructions as input and directly generate control actions for robots to perform specified tasks, greatly improving both decision-making capabilities and interaction with human users. However, the data-driven nature of VLA models, combined with their lack of interpretability, makes the assurance of their effectiveness and robustness a challenging task. This highlights the need for a reliable testing and evaluation platform. For this purpose, in this work, we propose LADEV, a comprehensive and efficient platform specifically designed for evaluating VLA models. We first present a language-driven approach that automatically generates simulation environments from natural language inputs, mitigating the need for manual adjustments and significantly improving testing efficiency. Then, to further assess the influence of language input on the VLA models, we implement a paraphrase mechanism that produces diverse natural language task instructions for testing. Finally, to expedite the evaluation process, we introduce a batch-style method for conducting large-scale testing of VLA models. Using LADEV, we conducted experiments on several state-of-the-art VLA models, demonstrating its effectiveness as a tool for evaluating these models. Our results showed that LADEV not only enhances testing efficiency but also establishes a solid baseline for evaluating VLA models, paving the way for the development of more intelligent and advanced robotic systems.

Paperid: 3199, https://arxiv.org/pdf/2410.00371.pdf

Abstract:
Robotic manipulation in open-world settings requires not only task execution but also the ability to detect and learn from failures. While recent advances in vision-language models (VLMs) and large language models (LLMs) have improved robots' spatial reasoning and problem-solving abilities, they still struggle with failure recognition, limiting their real-world applicability. We introduce AHA, an open-source VLM designed to detect and reason about failures in robotic manipulation using natural language. By framing failure detection as a free-form reasoning task, AHA identifies failures and provides detailed, adaptable explanations across different robots, tasks, and environments. We fine-tuned AHA using FailGen, a scalable framework that generates the first large-scale dataset of robotic failure trajectories, the AHA dataset. FailGen achieves this by procedurally perturbing successful demonstrations from simulation. Despite being trained solely on the AHA dataset, AHA generalizes effectively to real-world failure datasets, robotic systems, and unseen tasks. It surpasses the second-best model (GPT-4o in-context learning) by 10.3% and exceeds the average performance of six compared models including five state-of-the-art VLMs by 35.3% across multiple metrics and datasets. We integrate AHA into three manipulation frameworks that utilize LLMs/VLMs for reinforcement learning, task and motion planning, and zero-shot trajectory generation. AHA's failure feedback enhances these policies' performances by refining dense reward functions, optimizing task planning, and improving sub-task verification, boosting task success rates by an average of 21.4% across all three tasks compared to GPT-4 models.

Paperid: 3200, https://arxiv.org/pdf/2409.12894.pdf

Abstract:
The rapid advancement of generative AI and multi-modal foundation models has shown significant potential in advancing robotic manipulation. Vision-language-action (VLA) models, in particular, have emerged as a promising approach for visuomotor control by leveraging large-scale vision-language data and robot demonstrations. However, current VLA models are typically evaluated using a limited set of hand-crafted scenes, leaving their general performance and robustness in diverse scenarios largely unexplored. To address this gap, we present VLATest, a fuzzing framework designed to generate robotic manipulation scenes for testing VLA models. Based on VLATest, we conducted an empirical study to assess the performance of seven representative VLA models. Our study results revealed that current VLA models lack the robustness necessary for practical deployment. Additionally, we investigated the impact of various factors, including the number of confounding objects, lighting conditions, camera poses, unseen objects, and task instruction mutations, on the VLA model's performance. Our findings highlight the limitations of existing VLA models, emphasizing the need for further research to develop reliable and trustworthy VLA applications.

Paperid: 3201, https://arxiv.org/pdf/2601.14827.pdf

Abstract:
Vision-Language Models show strong zero-shot performance for chest X-ray classification, but standard flat metrics fail to distinguish between clinically minor and severe errors. This work investigates how to quantify and mitigate abstraction errors by leveraging medical taxonomies. We benchmark several state-of-the-art VLMs using hierarchical metrics and introduce Catastrophic Abstraction Errors to capture cross-branch mistakes. Our results reveal substantial misalignment of VLMs with clinical taxonomies despite high flat performance. To address this, we propose risk-constrained thresholding and taxonomy-aware fine-tuning with radial embeddings, which reduce severe abstraction errors to below 2 per cent while maintaining competitive performance. These findings highlight the importance of hierarchical evaluation and representation-level alignment for safer and more clinically meaningful deployment of VLMs.

Paperid: 3202, https://arxiv.org/pdf/2601.03926.pdf

Abstract:
The deployment of Large Vision-Language Models (LVLMs) for real-world document question answering is often constrained by dynamic, user-defined policies that dictate information disclosure based on context. While ensuring adherence to these explicit constraints is critical, existing safety research primarily focuses on implicit social norms or text-only settings, overlooking the complexities of multimodal documents. In this paper, we introduce Doc-PP (Document Policy Preservation Benchmark), a novel benchmark constructed from real-world reports requiring reasoning across heterogeneous visual and textual elements under strict non-disclosure policies. Our evaluation highlights a systemic Reasoning-Induced Safety Gap: models frequently leak sensitive information when answers must be inferred through complex synthesis or aggregated across modalities, effectively circumventing existing safety constraints. Furthermore, we identify that providing extracted text improves perception but inadvertently facilitates leakage. To address these vulnerabilities, we propose DVA (Decompose-Verify-Aggregation), a structural inference framework that decouples reasoning from policy verification. Experimental results demonstrate that DVA significantly outperforms standard prompting defenses, offering a robust baseline for policy-compliant document understanding

Paperid: 3203, https://arxiv.org/pdf/2601.03309.pdf

Abstract:
Vision-Language-Action (VLA) models, which integrate pretrained large Vision-Language Models (VLM) into their policy backbone, are gaining significant attention for their promising generalization capabilities. This paper revisits a fundamental yet seldom systematically studied question: how VLM choice and competence translate to downstream VLA policies performance? We introduce VLM4VLA, a minimal adaptation pipeline that converts general-purpose VLMs into VLA policies using only a small set of new learnable parameters for fair and efficient comparison. Despite its simplicity, VLM4VLA proves surprisingly competitive with more sophisticated network designs. Through extensive empirical studies on various downstream tasks across three benchmarks, we find that while VLM initialization offers a consistent benefit over training from scratch, a VLM's general capabilities are poor predictors of its downstream task performance. This challenges common assumptions, indicating that standard VLM competence is necessary but insufficient for effective embodied control. We further investigate the impact of specific embodied capabilities by fine-tuning VLMs on seven auxiliary embodied tasks (e.g., embodied QA, visual pointing, depth estimation). Contrary to intuition, improving a VLM's performance on specific embodied skills does not guarantee better downstream control performance. Finally, modality-level ablations identify the visual module in VLM, rather than the language component, as the primary performance bottleneck. We demonstrate that injecting control-relevant supervision into the vision encoder of the VLM yields consistent gains, even when the encoder remains frozen during downstream fine-tuning. This isolates a persistent domain gap between current VLM pretraining objectives and the requirements of embodied action-planning.

Paperid: 3204, https://arxiv.org/pdf/2512.15310.pdf

Abstract:
Weakly Supervised Semantic Segmentation (WSSS) with image level labels aims to produce pixel level predictions without requiring dense annotations. While recent approaches have leveraged generative models to augment existing data, they remain dependent on real world training samples. In this paper, we introduce a novel direction, Zero Shot Weakly Supervised Semantic Segmentation (ZSWSSS), and propose SynthSeg Agents, a multi agent framework driven by Large Language Models (LLMs) to generate synthetic training data entirely without real images. SynthSeg Agents comprises two key modules, a Self Refine Prompt Agent and an Image Generation Agent. The Self Refine Prompt Agent autonomously crafts diverse and semantically rich image prompts via iterative refinement, memory mechanisms, and prompt space exploration, guided by CLIP based similarity and nearest neighbor diversity filtering. These prompts are then passed to the Image Generation Agent, which leverages Vision Language Models (VLMs) to synthesize candidate images. A frozen CLIP scoring model is employed to select high quality samples, and a ViT based classifier is further trained to relabel the entire synthetic dataset with improved semantic precision. Our framework produces high quality training data without any real image supervision. Experiments on PASCAL VOC 2012 and COCO 2014 show that SynthSeg Agents achieves competitive performance without using real training images. This highlights the potential of LLM driven agents in enabling cost efficient and scalable semantic segmentation.

Paperid: 3205, https://arxiv.org/pdf/2511.21191.pdf

Abstract:
Recent advances in 3D vision-language models (VLMs) highlight a strong potential for 3D scene understanding and reasoning. However, effectively tokenizing 3D scenes into holistic scene tokens, and leveraging these tokens across diverse 3D understanding tasks, remain highly challenging. We present NDTokenizer3D, a generalist 3D VLM that performs a wide range of 3D scene understanding tasks while naturally supporting human interactions, thereby bridging language-level reasoning with 3D spatial understanding. The core of our approach is a novel three-stage scene tokenization pipeline built upon a Multi-Scale Normal Distributions Transform (NDT) representation, paired with a Multi-Scale NDT Decoder (MSDec). Specifically, NDTokenizer3D first constructs a multi-scale NDT representation from raw high-resolution point clouds, preserving both global context and fine-grained geometric details. Next, the MSDec progressively fuses cross-scale NDT features, producing holistic scene tokens consumable by LLM endpoints. Beyond tokenization, MSDec is repurposed as a general interface for human-interactive prompting (points, boxes, masks) and segmentation-mask decoding, unifying diverse 3D scene understanding tasks within a single architecture. With this compact and unified design, NDTokenizer3D offers a fine-grained, general-purpose 3D VLM, achieving remarkable improvements in 3D Referring Segmentation, 3D Visual Question Answering, and 3D Dense Captioning.

Paperid: 3206, https://arxiv.org/pdf/2511.17097.pdf

Abstract:
Vision-Language Navigation requires agents to act coherently over long horizons by understanding not only local visual context but also how far they have advanced within a multi-step instruction. However, recent Vision-Language-Action models focus on direct action prediction and earlier progress methods predict numeric achievements; both overlook the monotonic co-progression property of the observation and instruction sequences. Building on this insight, Progress-Think introduces semantic progress reasoning, predicting instruction-style progress from visual observations to enable more accurate navigation. To achieve this without expensive annotations, we propose a three-stage framework. In the initial stage, Self-Aligned Progress Pretraining bootstraps a reasoning module via a novel differentiable alignment between visual history and instruction prefixes. Then, Progress-Guided Policy Pretraining injects learned progress states into the navigation context, guiding the policy toward consistent actions. Finally, Progress-Policy Co-Finetuning jointly optimizes both modules with tailored progress-aware reinforcement objectives. Experiments on R2R-CE and RxR-CE show state-of-the-art success and efficiency, demonstrating that semantic progress yields a more consistent representation of navigation advancement.

Paperid: 3207, https://arxiv.org/pdf/2511.16203.pdf

Abstract:
Vision-Language-Action models (VLAs) have recently demonstrated remarkable progress in embodied environments, enabling robots to perceive, reason, and act through unified multimodal understanding. Despite their impressive capabilities, the adversarial robustness of these systems remains largely unexplored, especially under realistic multimodal and black-box conditions. Existing studies mainly focus on single-modality perturbations and overlook the cross-modal misalignment that fundamentally affects embodied reasoning and decision-making. In this paper, we introduce VLA-Fool, a comprehensive study of multimodal adversarial robustness in embodied VLA models under both white-box and black-box settings. VLA-Fool unifies three levels of multimodal adversarial attacks: (1) textual perturbations through gradient-based and prompt-based manipulations, (2) visual perturbations via patch and noise distortions, and (3) cross-modal misalignment attacks that intentionally disrupt the semantic correspondence between perception and instruction. We further incorporate a VLA-aware semantic space into linguistic prompts, developing the first automatically crafted and semantically guided prompting framework. Experiments on the LIBERO benchmark using a fine-tuned OpenVLA model reveal that even minor multimodal perturbations can cause significant behavioral deviations, demonstrating the fragility of embodied multimodal alignment.

Paperid: 3208, https://arxiv.org/pdf/2511.15204.pdf

Abstract:
Current state of the art measures like BLEU, CIDEr, VQA score, SigLIP-2 and CLIPScore are often unable to capture semantic or structural accuracy, especially for domain-specific or context-dependent scenarios. For this, this paper proposes a Physics-Constrained Multimodal Data Evaluation (PCMDE) metric combining large language models with reasoning, knowledge based mapping and vision-language models to overcome these limitations. The architecture is comprised of three main stages: (1) feature extraction of spatial and semantic information with multimodal features through object detection and VLMs; (2) Confidence-Weighted Component Fusion for adaptive component-level validation; and (3) physics-guided reasoning using large language models for structural and relational constraints (e.g., alignment, position, consistency) enforcement.

Paperid: 3209, https://arxiv.org/pdf/2511.11478.pdf

Abstract:
As embodied agents operate in increasingly complex environments, the ability to perceive, track, and reason about individual object instances over time becomes essential, especially in tasks requiring sequenced interactions with visually similar objects. In these non-Markovian settings, key decision cues are often hidden in object-specific histories rather than the current scene. Without persistent memory of prior interactions (what has been interacted with, where it has been, or how it has changed) visuomotor policies may fail, repeat past actions, or overlook completed ones. To surface this challenge, we introduce LIBERO-Mem, a non-Markovian task suite for stress-testing robotic manipulation under object-level partial observability. It combines short- and long-horizon object tracking with temporally sequenced subgoals, requiring reasoning beyond the current frame. However, vision-language-action (VLA) models often struggle in such settings, with token scaling quickly becoming intractable even for tasks spanning just a few hundred frames. We propose Embodied-SlotSSM, a slot-centric VLA framework built for temporal scalability. It maintains spatio-temporally consistent slot identities and leverages them through two mechanisms: (1) slot-state-space modeling for reconstructing short-term history, and (2) a relational encoder to align the input tokens with action decoding. Together, these components enable temporally grounded, context-aware action prediction. Experiments show Embodied-SlotSSM's baseline performance on LIBERO-Mem and general tasks, offering a scalable solution for non-Markovian reasoning in object-centric robotic policies.

Paperid: 3210, https://arxiv.org/pdf/2510.22768.pdf

Abstract:
As Large Vision-Language Models (LVLMs) are increasingly deployed in domains such as shopping, health, and news, they are exposed to pervasive persuasive content. A critical question is how these models function as persuadees-how and why they can be influenced by persuasive multimodal inputs. Understanding both their susceptibility to persuasion and the effectiveness of different persuasive strategies is crucial, as overly persuadable models may adopt misleading beliefs, override user preferences, or generate unethical or unsafe outputs when exposed to manipulative messages. We introduce MMPersuade, a unified framework for systematically studying multimodal persuasion dynamics in LVLMs. MMPersuade contributes (i) a comprehensive multimodal dataset that pairs images and videos with established persuasion principles across commercial, subjective and behavioral, and adversarial contexts, and (ii) an evaluation framework that quantifies both persuasion effectiveness and model susceptibility via third-party agreement scoring and self-estimated token probabilities on conversation histories. Our study of six leading LVLMs as persuadees yields three key insights: (i) multimodal inputs substantially increase persuasion effectiveness-and model susceptibility-compared to text alone, especially in misinformation scenarios; (ii) stated prior preferences decrease susceptibility, yet multimodal information maintains its persuasive advantage; and (iii) different strategies vary in effectiveness across contexts, with reciprocity being most potent in commercial and subjective contexts, and credibility and logic prevailing in adversarial contexts. By jointly analyzing persuasion effectiveness and susceptibility, MMPersuade provides a principled foundation for developing models that are robust, preference-consistent, and ethically aligned when engaging with persuasive multimodal content.

Paperid: 3211, https://arxiv.org/pdf/2510.11738.pdf

Abstract:
We introduce SeeingSounds, a lightweight and modular framework for audio-to-image generation that leverages the interplay between audio, language, and vision-without requiring any paired audio-visual data or training on visual generative models. Rather than treating audio as a substitute for text or relying solely on audio-to-text mappings, our method performs dual alignment: audio is projected into a semantic language space via a frozen language encoder, and, contextually grounded into the visual domain using a vision-language model. This approach, inspired by cognitive neuroscience, reflects the natural cross-modal associations observed in human perception. The model operates on frozen diffusion backbones and trains only lightweight adapters, enabling efficient and scalable learning. Moreover, it supports fine-grained and interpretable control through procedural text prompt generation, where audio transformations (e.g., volume or pitch shifts) translate into descriptive prompts (e.g., "a distant thunder") that guide visual outputs. Extensive experiments across standard benchmarks confirm that SeeingSounds outperforms existing methods in both zero-shot and supervised settings, establishing a new state of the art in controllable audio-to-visual generation.

Paperid: 3212, https://arxiv.org/pdf/2510.10642.pdf

Abstract:
Building generalist robot policies that can handle diverse tasks in open-ended environments is a central challenge in robotics. To leverage knowledge from large-scale pretraining, prior work has typically built generalist policies either on top of vision-language understanding models (VLMs) or generative models. However, both semantic understanding from vision-language pretraining and visual dynamics modeling from visual-generation pretraining are crucial for embodied robots. Recent unified models of generation and understanding have demonstrated strong capabilities in both comprehension and generation through large-scale pretraining. We posit that robotic policy learning can likewise benefit from the combined strengths of understanding, planning and continuous future representation learning. Building on this insight, we introduce UniCoD, which acquires the ability to dynamically model high-dimensional visual features through pretraining on over 1M internet-scale instructional manipulation videos. Subsequently, UniCoD is fine-tuned on data collected from the robot embodiment, enabling the learning of mappings from predictive representations to action tokens. Extensive experiments show our approach consistently outperforms baseline methods in terms of 9\% and 12\% across simulation environments and real-world out-of-distribution tasks.

Paperid: 3213, https://arxiv.org/pdf/2510.06783.pdf

Abstract:
Existing methods for extracting reward signals in Reinforcement Learning typically rely on labeled data and dedicated training splits, a setup that contrasts with how humans learn directly from their environment. In this work, we propose TTRV to enhance vision language understanding by adapting the model on the fly at inference time, without the need for any labeled data. Concretely, we enhance the Group Relative Policy Optimization (GRPO) framework by designing rewards based on the frequency of the base model's output, while inferring on each test sample multiple times. Further, we also propose to control the diversity of the model's output by simultaneously rewarding the model for obtaining low entropy of the output empirical distribution. Our approach delivers consistent gains across both object recognition and visual question answering (VQA), with improvements of up to 52.4% and 29.8%, respectively, and average boosts of 24.6% and 10.0% across 16 datasets.Remarkably, on image recognition, TTRV applied to InternVL 8B surpasses GPT-4o by an average of 2.3% over 8 benchmarks, while remaining highly competitive on VQA, demonstrating that test-time reinforcement learning can match or exceed the strongest proprietary models. Finally, we find many interesting properties of test-time RL for VLMs: for example, even in extremely data-constrained scenarios, where adaptation is performed on a single randomly chosen unlabeled test example, TTRV still yields non-trivial improvements of up to 5.5% in recognition tasks.

Paperid: 3214, https://arxiv.org/pdf/2509.25916.pdf

Abstract:
Vision-Language Models (VLMs) excel at high-level scene understanding but falter on fine-grained perception tasks requiring precise localization. This failure stems from a fundamental mismatch, as generating exact numerical coordinates is a challenging task for language-centric architectures. In this paper, we introduce VLM-FO1, a novel framework that overcomes this limitation by reframing object-centric perception from a brittle coordinate generation problem into a robust feature retrieval task. Our method operates as a plug-and-play module that integrates with any pre-trained VLM. It leverages a Hybrid Fine-grained Region Encoder (HFRE), featuring a dual vision encoder, to generate powerful region tokens rich in both semantic and spatial detail. A token-based referencing system then enables the LLM to seamlessly reason about and ground language in these specific visual regions. Experiments show that VLM-FO1 achieves state-of-the-art performance across a diverse suite of benchmarks, demonstrating exceptional capabilities in object grounding, region generational understanding, and visual region reasoning. Crucially, our two-stage training strategy ensures that these perception gains are achieved without compromising the base model's general visual understanding capabilities. VLM-FO1 establishes an effective and flexible paradigm for building perception-aware VLMs, bridging the gap between high-level reasoning and fine-grained visual grounding.

Paperid: 3215, https://arxiv.org/pdf/2509.24491.pdf

Abstract:
Multimodal Large Language Models (MLLMs) have significantly improved the performance of various tasks, but continue to suffer from visual hallucinations, a critical issue where generated responses contradict visual evidence. While Direct Preference Optimization(DPO) is widely used for alignment, its application to MLLMs often fails to capture fine-grained semantic differences and encourages shortcut learning. To address these challenges, we propose Semantic Curriculum Preference Optimization (SCPO), a novel framework for MLLM alignment. SCPO employs a progressive, easy-to-hard curriculum built upon our Semantic Curriculum Preference Pairs dataset, which provides fine-grained semantic contrasts sorted by difficulty. This curriculum is trained with a dynamic reference model and a novel symmetric, bidirectional objective to facilitate simultaneous learning from both textual and visual preferences. To our knowledge, SCPO is the first framework to unify semantics, symmetry, and curriculum for MLLMs alignment, effectively mitigating visual hallucinations. Extensive experiments on LLaVA models across various scales and versions validate that SCPO demonstrates superior performance compared to baseline models on multiple hallucination benchmarks, reducing the hallucination rate by up to 62.9%. Moreover, evaluations on generalized benchmarks show that SCPO improves factuality while preserving general capabilities, with its performance remaining stable across general vision-language benchmarks.

Paperid: 3216, https://arxiv.org/pdf/2509.23690.pdf

Abstract:
Embodied agents can identify and report safety hazards in the home environments. Accurately evaluating their capabilities in home safety inspection tasks is curcial, but existing benchmarks suffer from two key limitations. First, they oversimplify safety inspection tasks by using textual descriptions of the environment instead of direct visual information, which hinders the accurate evaluation of embodied agents based on Vision-Language Models (VLMs). Second, they use a single, static viewpoint for environmental observation, which restricts the agents' free exploration and cause the omission of certain safety hazards, especially those that are occluded from a fixed viewpoint. To alleviate these issues, we propose HomeSafeBench, a benchmark with 12,900 data points covering five common home safety hazards: fire, electric shock, falling object, trips, and child safety. HomeSafeBench provides dynamic first-person perspective images from simulated home environments, enabling the evaluation of VLM capabilities for home safety inspection. By allowing the embodied agents to freely explore the room, HomeSafeBench provides multiple dynamic perspectives in complex environments for a more thorough inspection. Our comprehensive evaluation of mainstream VLMs on HomeSafeBench reveals that even the best-performing model achieves an F1-score of only 10.23%, demonstrating significant limitations in current VLMs. The models particularly struggle with identifying safety hazards and selecting effective exploration strategies. We hope HomeSafeBench will provide valuable reference and support for future research related to home security inspections. Our dataset and code will be publicly available soon.

Paperid: 3217, https://arxiv.org/pdf/2509.22720.pdf

Abstract:
Designing realistic multi-object scenes requires not only generating images, but also planning spatial layouts that respect semantic relations and physical plausibility. On one hand, while recent advances in diffusion models have enabled high-quality image generation, they lack explicit spatial reasoning, leading to unrealistic object layouts. On the other hand, traditional spatial planning methods in robotics emphasize geometric and relational consistency, but they struggle to capture semantic richness in visual scenes. To bridge this gap, in this paper, we propose LayoutAgent, an agentic framework that unifies vision-language reasoning with compositional diffusion for layout generation. Given multiple input images with target objects in them, our method first employs visual-language model to preprocess the inputs through segmentation, object size estimation, scene graph construction, and prompt rewriting. Then we leverage compositional diffusion-a method traditionally used in robotics-to synthesize bounding boxes that respect object relations encoded in the scene graph for spatial layouts. In the end, a foreground-conditioned image generator composes the complete scene by rendering the objects into the planned layout guided by designed prompts. Experiments demonstrate that LayoutAgent outperforms other state-of-the-art layout generation models in layout coherence, spatial realism and aesthetic alignment.

Paperid: 3218, https://arxiv.org/pdf/2509.21922.pdf

Abstract:
Spatial understanding is a critical capability for vision foundation models. While recent advances in large vision models or vision-language models (VLMs) have expanded recognition capabilities, most benchmarks emphasize localization accuracy rather than whether models capture how objects are arranged and related within a scene. This gap is consequential; effective scene understanding requires not only identifying objects, but reasoning about their relative positions, groupings, and depth. In this paper, we present a systematic benchmark for object-centric spatial reasoning in foundation models. Using a controlled synthetic dataset, we evaluate state-of-the-art vision models (e.g., GroundingDINO, Florence-2, OWLv2) and large VLMs (e.g., InternVL, LLaVA, GPT-4o) across three tasks: spatial localization, spatial reasoning, and downstream retrieval tasks. We find a stable trade-off: detectors such as GroundingDINO and OWLv2 deliver precise boxes with limited relational reasoning, while VLMs like SmolVLM and GPT-4o provide coarse layout cues and fluent captions but struggle with fine-grained spatial context. Our study highlights the gap between localization and true spatial understanding, and pointing toward the need for spatially-aware foundation models in the community.

Paperid: 3219, https://arxiv.org/pdf/2509.21609.pdf

Abstract:
Immediate damage assessment is essential after natural catastrophes; yet, conventional hand evaluation techniques are sluggish and perilous. Although satellite and unmanned aerial vehicle (UAV) photos offer extensive perspectives of impacted regions, current computer vision methodologies generally yield just classification labels or segmentation masks, so constraining their capacity to deliver a thorough situational comprehension. We introduce the Vision Language Caption Enhancer (VLCE), a multimodal system designed to produce comprehensive, contextually-informed explanations of disaster imagery. VLCE employs a dual-architecture approach: a CNN-LSTM model with a ResNet50 backbone pretrained on EuroSat satellite imagery for the xBD dataset, and a Vision Transformer (ViT) model pretrained on UAV pictures for the RescueNet dataset. Both systems utilize external semantic knowledge from ConceptNet and WordNet to expand vocabulary coverage and improve description accuracy. We assess VLCE in comparison to leading vision-language models (LLaVA and QwenVL) utilizing CLIPScore for semantic alignment and InfoMetIC for caption informativeness. Experimental findings indicate that VLCE markedly surpasses baseline models, attaining a maximum of 95.33% on InfoMetIC while preserving competitive semantic alignment. Our dual-architecture system demonstrates significant potential for improving disaster damage assessment by automating the production of actionable, information-dense descriptions from satellite and drone photos.

Paperid: 3220, https://arxiv.org/pdf/2509.15330.pdf

Abstract:
Recent advances in pre-training vision-language models (VLMs), e.g., contrastive language-image pre-training (CLIP) methods, have shown great potential in learning out-of-distribution (OOD) representations. Despite showing competitive performance, the prompt-based CLIP methods still suffer from: i) inaccurate text descriptions, which leads to degraded accuracy and robustness, and poses a challenge for zero-shot CLIP methods. ii) limited vision-language embedding alignment, which significantly affects the generalization performance. To tackle the above issues, this paper proposes a novel Conditional Domain prompt Learning (CoDoL) method, which utilizes readily-available domain information to form prompts and improves the vision-language embedding alignment for improving OOD generalization. To capture both instance-specific and domain-specific information, we further propose a lightweight Domain Meta Network (DMN) to generate input-conditional tokens for images in each domain. Extensive experiments on four OOD benchmarks (PACS, VLCS, OfficeHome and DigitDG) validate the effectiveness of our proposed CoDoL in terms of improving the vision-language embedding alignment as well as the out-of-distribution generalization performance.

Paperid: 3221, https://arxiv.org/pdf/2509.09731.pdf

Abstract:
Chinese ancient documents, invaluable carriers of millennia of Chinese history and culture, hold rich knowledge across diverse fields but face challenges in digitization and understanding, i.e., traditional methods only scan images, while current Vision-Language Models (VLMs) struggle with their visual and linguistic complexity. Existing document benchmarks focus on English printed texts or simplified Chinese, leaving a gap for evaluating VLMs on ancient Chinese documents. To address this, we present AncientDoc, the first benchmark for Chinese ancient documents, designed to assess VLMs from OCR to knowledge reasoning. AncientDoc includes five tasks (page-level OCR, vernacular translation, reasoning-based QA, knowledge-based QA, linguistic variant QA) and covers 14 document types, over 100 books, and about 3,000 pages. Based on AncientDoc, we evaluate mainstream VLMs using multiple metrics, supplemented by a human-aligned large language model for scoring.

Paperid: 3222, https://arxiv.org/pdf/2509.00676.pdf

Abstract:
In vision-language modeling, critic models are typically trained to evaluate outputs -- assigning scalar scores or pairwise preferences -- rather than to generate responses. This separation from policy models, which produce the responses, is so entrenched that critics are rarely considered for direct policy use. In this work, we challenge this convention. We propose to reorganize preference-labeled critic datasets into verifiable training signals and perform reinforcement learning directly on a base generative model, producing LLaVA-Critic-R1, a multimodal critic trained to optimize preference judgments while retaining full generation ability. Surprisingly, LLaVA-Critic-R1 emerges not only as a top-performing critic but also as a competitive policy model -- matching or surpassing specialized reasoning VLMs trained with in-domain data across 26 visual reasoning and understanding benchmarks, with an average gain of +5.7% over its base model (Qwen-2.5-VL-7B). Extending this approach to existing strong reasoning VLMs yields LLaVA-Critic-R1+, which further advances policy performance without sacrificing critic quality, achieving a SoTA performance of 71.9 on MMMU at the 7B scale. Finally, we show that the enhanced critic ability benefits inference: applying self-critique at test time yields an average +13.8% improvement on five representative reasoning tasks without additional training. Our results reveal that RL training on critic data can produce a unified model excelling at both evaluation and generation, offering a simple path toward scalable, self-improving multimodal systems.

Paperid: 3223, https://arxiv.org/pdf/2508.10681.pdf

Abstract:
Few-Shot Industrial Anomaly Detection (FS-IAD) has important applications in automating industrial quality inspection. Recently, some FS-IAD methods based on Large Vision-Language Models (LVLMs) have been proposed with some achievements through prompt learning or fine-tuning. However, existing LVLMs focus on general tasks but lack basic industrial knowledge and reasoning capabilities related to FS-IAD, making these methods far from specialized human quality inspectors. To address these challenges, we propose a unified framework, IADGPT, designed to perform FS-IAD in a human-like manner, while also handling associated localization and reasoning tasks, even for diverse and novel industrial products. To this end, we introduce a three-stage progressive training strategy inspired by humans. Specifically, the first two stages gradually guide IADGPT in acquiring fundamental industrial knowledge and discrepancy awareness. In the third stage, we design an in-context learning-based training paradigm, enabling IADGPT to leverage a few-shot image as the exemplars for improved generalization to novel products. In addition, we design a strategy that enables IADGPT to output image-level and pixel-level anomaly scores using the logits output and the attention map, respectively, in conjunction with the language output to accomplish anomaly reasoning. To support our training, we present a new dataset comprising 100K images across 400 diverse industrial product categories with extensive attribute-level textual annotations. Experiments indicate IADGPT achieves considerable performance gains in anomaly detection and demonstrates competitiveness in anomaly localization and reasoning. We will release our dataset in camera-ready.

Paperid: 3224, https://arxiv.org/pdf/2508.10315.pdf

Abstract:
Existing backdoor defense methods in Federated Learning (FL) rely on the assumption of homogeneous client data distributions or the availability of a clean serve dataset, which limits the practicality and effectiveness. Defending against backdoor attacks under heterogeneous client data distributions while preserving model performance remains a significant challenge. In this paper, we propose a FL backdoor defense framework named CLIP-Fed, which leverages the zero-shot learning capabilities of vision-language pre-training models. By integrating both pre-aggregation and post-aggregation defense strategies, CLIP-Fed overcomes the limitations of Non-IID imposed on defense effectiveness. To address privacy concerns and enhance the coverage of the dataset against diverse triggers, we construct and augment the server dataset using the multimodal large language model and frequency analysis without any client samples. To address class prototype deviations caused by backdoor samples and eliminate the correlation between trigger patterns and target labels, CLIP-Fed aligns the knowledge of the global model and CLIP on the augmented dataset using prototype contrastive loss and Kullback-Leibler divergence. Extensive experiments on representative datasets validate the effectiveness of CLIP-Fed. Compared to state-of-the-art methods, CLIP-Fed achieves an average reduction in ASR, i.e., 2.03\% on CIFAR-10 and 1.35\% on CIFAR-10-LT, while improving average MA by 7.92\% and 0.48\%, respectively.

Paperid: 3225, https://arxiv.org/pdf/2508.07630.pdf

Abstract:
We introduce InterChart, a diagnostic benchmark that evaluates how well vision-language models (VLMs) reason across multiple related charts, a task central to real-world applications such as scientific reporting, financial analysis, and public policy dashboards. Unlike prior benchmarks focusing on isolated, visually uniform charts, InterChart challenges models with diverse question types ranging from entity inference and trend correlation to numerical estimation and abstract multi-step reasoning grounded in 2-3 thematically or structurally related charts. We organize the benchmark into three tiers of increasing difficulty: (1) factual reasoning over individual charts, (2) integrative analysis across synthetically aligned chart sets, and (3) semantic inference over visually complex, real-world chart pairs. Our evaluation of state-of-the-art open and closed-source VLMs reveals consistent and steep accuracy declines as chart complexity increases. We find that models perform better when we decompose multi-entity charts into simpler visual units, underscoring their struggles with cross-chart integration. By exposing these systematic limitations, InterChart provides a rigorous framework for advancing multimodal reasoning in complex, multi-visual environments.

Paperid: 3226, https://arxiv.org/pdf/2508.02629.pdf

Abstract:
Recent advances in multimodal large language models (MLLMs) have enabled richer perceptual grounding for code policy generation in embodied agents. However, most existing systems lack effective mechanisms to adaptively monitor policy execution and repair codes during task completion. In this work, we introduce HyCodePolicy, a hybrid language-based control framework that systematically integrates code synthesis, geometric grounding, perceptual monitoring, and iterative repair into a closed-loop programming cycle for embodied agents. Technically, given a natural language instruction, our system first decomposes it into subgoals and generates an initial executable program grounded in object-centric geometric primitives. The program is then executed in simulation, while a vision-language model (VLM) observes selected checkpoints to detect and localize execution failures and infer failure reasons. By fusing structured execution traces capturing program-level events with VLM-based perceptual feedback, HyCodePolicy infers failure causes and repairs programs. This hybrid dual feedback mechanism enables self-correcting program synthesis with minimal human supervision. Our results demonstrate that HyCodePolicy significantly improves the robustness and sample efficiency of robot manipulation policies, offering a scalable strategy for integrating multimodal reasoning into autonomous decision-making pipelines.

Paperid: 3227, https://arxiv.org/pdf/2508.02549.pdf

Abstract:
Vision-Language Navigation (VLN) tasks often leverage panoramic RGB and depth inputs to provide rich spatial cues for action planning, but these sensors can be costly or less accessible in real-world deployments. Recent approaches based on Vision-Language Action (VLA) models achieve strong results with monocular input, yet they still lag behind methods using panoramic RGB-D information. We present MonoDream, a lightweight VLA framework that enables monocular agents to learn a Unified Navigation Representation (UNR). This shared feature representation jointly aligns navigation-relevant visual semantics (e.g., global layout, depth, and future cues) and language-grounded action intent, enabling more reliable action prediction. MonoDream further introduces Latent Panoramic Dreaming (LPD) tasks to supervise the UNR, which train the model to predict latent features of panoramic RGB and depth observations at both current and future steps based on only monocular input. Experiments on multiple VLN benchmarks show that MonoDream consistently improves monocular navigation performance and significantly narrows the gap with panoramic-based agents.

Paperid: 3228, https://arxiv.org/pdf/2507.17080.pdf

Abstract:
Multimodal learning plays a critical role in e-commerce recommendation platforms today, enabling accurate recommendations and product understanding. However, existing vision-language models, such as CLIP, face key challenges in e-commerce recommendation systems: 1) Weak object-level alignment, where global image embeddings fail to capture fine-grained product attributes, leading to suboptimal retrieval performance; 2) Ambiguous textual representations, where product descriptions often lack contextual clarity, affecting cross-modal matching; and 3) Domain mismatch, as generic vision-language models may not generalize well to e-commerce-specific data. To address these limitations, we propose a framework, VL-CLIP, that enhances CLIP embeddings by integrating Visual Grounding for fine-grained visual understanding and an LLM-based agent for generating enriched text embeddings. Visual Grounding refines image representations by localizing key products, while the LLM agent enhances textual features by disambiguating product descriptions. Our approach significantly improves retrieval accuracy, multimodal retrieval effectiveness, and recommendation quality across tens of millions of items on one of the largest e-commerce platforms in the U.S., increasing CTR by 18.6%, ATC by 15.5%, and GMV by 4.0%. Additional experimental results show that our framework outperforms vision-language models, including CLIP, FashionCLIP, and GCL, in both precision and semantic alignment, demonstrating the potential of combining object-aware visual grounding and LLM-enhanced text representation for robust multimodal recommendations.

Paperid: 3229, https://arxiv.org/pdf/2507.15824.pdf

Abstract:
Recent progress in text-to-video (T2V) generation has enabled the synthesis of visually compelling and temporally coherent videos from natural language. However, these models often fall short in basic physical commonsense, producing outputs that violate intuitive expectations around causality, object behavior, and tool use. Addressing this gap, we present PhysVidBench, a benchmark designed to evaluate the physical reasoning capabilities of T2V systems. The benchmark includes 383 carefully curated prompts, emphasizing tool use, material properties, and procedural interactions, and domains where physical plausibility is crucial. For each prompt, we generate videos using diverse state-of-the-art models and adopt a three-stage evaluation pipeline: (1) formulate grounded physics questions from the prompt, (2) caption the generated video with a vision-language model, and (3) task a language model to answer several physics-involved questions using only the caption. This indirect strategy circumvents common hallucination issues in direct video-based evaluation. By highlighting affordances and tool-mediated actions, areas overlooked in current T2V evaluations, PhysVidBench provides a structured, interpretable framework for assessing physical commonsense in generative video models.

Paperid: 3230, https://arxiv.org/pdf/2507.06814.pdf

Abstract:
Low-Light Image Enhancement (LLIE) aims to restore vivid content and details from corrupted low-light images. However, existing standard RGB (sRGB) color space-based LLIE methods often produce color bias and brightness artifacts due to the inherent high color sensitivity. While Hue, Saturation, and Value (HSV) color space can decouple brightness and color, it introduces significant red and black noise artifacts. To address this problem, we propose a new color space for LLIE, namely Horizontal/Vertical-Intensity (HVI), defined by the HV color map and learnable intensity. The HV color map enforces small distances for the red coordinates to remove red noise artifacts, while the learnable intensity compresses the low-light regions to remove black noise artifacts. Additionally, we introduce the Color and Intensity Decoupling Network+ (HVI-CIDNet+), built upon the HVI color space, to restore damaged content and mitigate color distortion in extremely dark regions. Specifically, HVI-CIDNet+ leverages abundant contextual and degraded knowledge extracted from low-light images using pre-trained vision-language models, integrated via a novel Prior-guided Attention Block (PAB). Within the PAB, latent semantic priors can promote content restoration, while degraded representations guide precise color correction, both particularly in extremely dark regions through the meticulously designed cross-attention fusion mechanism. Furthermore, we construct a Region Refinement Block that employs convolution for information-rich regions and self-attention for information-scarce regions, ensuring accurate brightness adjustments. Comprehensive results from benchmark experiments demonstrate that the proposed HVI-CIDNet+ outperforms the state-of-the-art methods on 10 datasets.

Paperid: 3231, https://arxiv.org/pdf/2506.21934.pdf

Abstract:
Automated content-aware layout generation -- the task of arranging visual elements such as text, logos, and underlays on a background canvas -- remains a fundamental yet under-explored problem in intelligent design systems. While recent advances in deep generative models and large language models (LLMs) have shown promise in structured content generation, most existing approaches lack grounding in contextual design exemplars and fall short in handling semantic alignment and visual coherence. In this work we introduce CAL-RAG, a retrieval-augmented, agentic framework for content-aware layout generation that integrates multimodal retrieval, large language models, and collaborative agentic reasoning. Our system retrieves relevant layout examples from a structured knowledge base and invokes an LLM-based layout recommender to propose structured element placements. A vision-language grader agent evaluates the layout with visual metrics, and a feedback agent provides targeted refinements, enabling iterative improvement. We implement our framework using LangGraph and evaluate it on the PKU PosterLayout dataset, a benchmark rich in semantic and structural variability. CAL-RAG achieves state-of-the-art performance across multiple layout metrics -- including underlay effectiveness, element alignment, and overlap -- substantially outperforming strong baselines such as LayoutPrompter. These results demonstrate that combining retrieval augmentation with agentic multi-step reasoning yields a scalable, interpretable, and high-fidelity solution for automated layout generation.

Paperid: 3232, https://arxiv.org/pdf/2506.05523.pdf

Abstract:
Despite rapid advances in vision-language models (VLMs), current benchmarks for multimodal reasoning fall short in three key dimensions. First, they overwhelmingly rely on static images, failing to capture the temporal complexity of real-world environments. Second, they narrowly focus on mathematical problem-solving, neglecting the broader spectrum of reasoning skills -- including abstract, physical, planning, spatial, and temporal capabilities -- required for robust multimodal intelligence. Third, many benchmarks quickly saturate, offering limited headroom for diagnosing failure modes or measuring continued progress. We introduce MORSE-500 (Multimodal Reasoning Stress-test Environment), a video benchmark composed of 500 fully scripted clips with embedded questions spanning six complementary reasoning categories. Each instance is programmatically generated using deterministic Python scripts (via Manim, Matplotlib, MoviePy), generative video models, and curated real footage. This script-driven design allows fine-grained control over visual complexity, distractor density, and temporal dynamics -- enabling difficulty to be scaled systematically as models improve. Unlike static benchmarks that become obsolete once saturated, MORSE-500 is built to evolve: its controllable generation pipeline supports the creation of arbitrarily challenging new instances, making it ideally suited for stress-testing next-generation models. Initial experiments with state-of-the-art systems -- including various Gemini 2.5 Pro and OpenAI o3 which represent the strongest available at the time, alongside strong open-source models -- reveal substantial performance gaps across all categories, with particularly large deficits in abstract and planning tasks. We release the full dataset, generation scripts, and evaluation harness to support transparent, reproducible, and forward-looking multimodal reasoning research.

Paperid: 3233, https://arxiv.org/pdf/2506.03589.pdf

Abstract:
Text-video retrieval (TVR) systems often suffer from visual-linguistic biases present in datasets, which cause pre-trained vision-language models to overlook key details. To address this, we propose BiMa, a novel framework designed to mitigate biases in both visual and textual representations. Our approach begins by generating scene elements that characterize each video by identifying relevant entities/objects and activities. For visual debiasing, we integrate these scene elements into the video embeddings, enhancing them to emphasize fine-grained and salient details. For textual debiasing, we introduce a mechanism to disentangle text features into content and bias components, enabling the model to focus on meaningful content while separately handling biased information. Extensive experiments and ablation studies across five major TVR benchmarks (i.e., MSR-VTT, MSVD, LSMDC, ActivityNet, and DiDeMo) demonstrate the competitive performance of BiMa. Additionally, the model's bias mitigation capability is consistently validated by its strong results on out-of-distribution retrieval tasks.

Paperid: 3234, https://arxiv.org/pdf/2506.00806.pdf

Abstract:
Multimodal large language models (MLLMs) still struggle with complex reasoning tasks in Visual Question Answering (VQA). While current methods have advanced by incorporating visual prompts, our study uncovers critical limitations: these approaches indiscriminately annotate all detected objects for every visual question, generating excessive visual markers that degrade task performance. This issue stems primarily from a lack of focus on key visual elements, raising two important questions: Are all objects equally important, and do all questions require visual prompts? Motivated by Dual Process Theory, which distinguishes between instinctive and deliberate cognitive modes in human reasoning, we propose FOCUS, a plug-and-play approach that dynamically adapts to the complexity of questions, combining fast intuitive judgments with deliberate analytical reasoning to enhance the vision-language reasoning capability of the MLLM. For straightforward questions, FOCUS supports efficient zero-shot reasoning. For more complex tasks, it employs the conceptualizing before observation strategy to highlight critical elements. Extensive experiments on four benchmarks, ScienceQA, TextQA, VizWiz, and MME, demonstrate that FOCUS consistently improves the performance of both open-source and black-box MLLMs, achieving significant gains across all datasets. Ablation studies further validate the importance of combining diverse cognitive strategies with refined visual information for superior performance. Code will be released.

Paperid: 3235, https://arxiv.org/pdf/2505.22143.pdf

Abstract:
Large vision-language models (LVLMs) have significantly advanced numerous fields. In this work, we explore how to harness their potential to address 3D scene understanding tasks, using 3D question answering (3D-QA) as a representative example. Due to the limited training data in 3D, we do not train LVLMs but infer in a zero-shot manner. Specifically, we sample 2D views from a 3D point cloud and feed them into 2D models to answer a given question. When the 2D model is chosen, e.g., LLAVA-OV, the quality of sampled views matters the most. We propose cdViews, a novel approach to automatically selecting critical and diverse Views for 3D-QA. cdViews consists of two key components: viewSelector prioritizing critical views based on their potential to provide answer-specific information, and viewNMS enhancing diversity by removing redundant views based on spatial overlap. We evaluate cdViews on the widely-used ScanQA and SQA benchmarks, demonstrating that it achieves state-of-the-art performance in 3D-QA while relying solely on 2D models without fine-tuning. These findings support our belief that 2D LVLMs are currently the most effective alternative (of the resource-intensive 3D LVLMs) for addressing 3D tasks.

Paperid: 3236, https://arxiv.org/pdf/2505.15865.pdf

Abstract:
Despite significant advancements in Large Vision Language Models (LVLMs), a gap remains, particularly regarding their interpretability and how they locate and interpret textual information within images. In this paper, we explore various LVLMs to identify the specific heads responsible for recognizing text from images, which we term the Optical Character Recognition Head (OCR Head). Our findings regarding these heads are as follows: (1) Less Sparse: Unlike previous retrieval heads, a large number of heads are activated to extract textual information from images. (2) Qualitatively Distinct: OCR heads possess properties that differ significantly from general retrieval heads, exhibiting low similarity in their characteristics. (3) Statically Activated: The frequency of activation for these heads closely aligns with their OCR scores. We validate our findings in downstream tasks by applying Chain-of-Thought (CoT) to both OCR and conventional retrieval heads and by masking these heads. We also demonstrate that redistributing sink-token values within the OCR heads improves performance. These insights provide a deeper understanding of the internal mechanisms LVLMs employ in processing embedded textual information in images.

Paperid: 3237, https://arxiv.org/pdf/2505.14064.pdf

Abstract:
In many real-world applications, deployed models encounter inputs that differ from the data seen during training. Out-of-distribution detection identifies whether an input stems from an unseen distribution, while open-world recognition flags such inputs to ensure the system remains robust as ever-emerging, previously $unknown$ categories appear and must be addressed without retraining. Foundation and vision-language models are pre-trained on large and diverse datasets with the expectation of broad generalization across domains, including medical imaging. However, benchmarking these models on test sets with only a few common outlier types silently collapses the evaluation back to a closed-set problem, masking failures on rare or truly novel conditions encountered in clinical use. We therefore present $NOVA$, a challenging, real-life $evaluation-only$ benchmark of $\sim$900 brain MRI scans that span 281 rare pathologies and heterogeneous acquisition protocols. Each case includes rich clinical narratives and double-blinded expert bounding-box annotations. Together, these enable joint assessment of anomaly localisation, visual captioning, and diagnostic reasoning. Because NOVA is never used for training, it serves as an $extreme$ stress-test of out-of-distribution generalisation: models must bridge a distribution gap both in sample appearance and in semantic space. Baseline results with leading vision-language models (GPT-4o, Gemini 2.0 Flash, and Qwen2.5-VL-72B) reveal substantial performance drops across all tasks, establishing NOVA as a rigorous testbed for advancing models that can detect, localize, and reason about truly unknown anomalies.

Paperid: 3238, https://arxiv.org/pdf/2505.11886.pdf

Abstract:
Vision-Language Navigation (VLN) is a critical task for developing embodied agents that can follow natural language instructions to navigate in complex real-world environments. Recent advances in VLN by large pretrained models have significantly improved generalization and instruction grounding compared to traditional approaches. However, the role of reasoning strategies in navigation-an action-centric, long-horizon task-remains underexplored, despite Chain-of-Thought (CoT) reasoning's demonstrated success in static tasks like visual question answering. To address this gap, we conduct the first systematic evaluation of reasoning strategies for VLN, including No-Think (direct action prediction), Pre-Think (reason before action), and Post-Think (reason after action). Surprisingly, our findings reveal the Inference-time Reasoning Collapse issue, where inference-time reasoning degrades navigation accuracy, highlighting the challenges of integrating reasoning into VLN. Based on this insight, we propose Aux-Think, a framework that trains models to internalize structured reasoning patterns through CoT supervision, while inferring action directly without reasoning in online prediction. To support this framework, we release R2R-CoT-320k, the first Chain-of-Thought annotated dataset for VLN. Extensive experiments show that Aux-Think reduces training effort greatly and achieves the best performance under the same data scale.

Paperid: 3239, https://arxiv.org/pdf/2505.11454.pdf

Abstract:
Large multimodal models (LMMs) have been widely tested on tasks like visual question answering (VQA), image captioning, and grounding, but lack rigorous evaluation for alignment with human-centered (HC) values such as fairness, ethics, and inclusivity. To address this gap, we introduce \textbf{HumaniBench}, a novel benchmark of 32,000 real-world image-question pairs and an evaluation suite. Labels are generated via an AI-assisted pipeline and validated by experts. HumaniBench assesses LMMs across seven key alignment principles: fairness, ethics, empathy, inclusivity, reasoning, robustness, and multilinguality, through diverse open-ended and closed-ended VQA tasks. Grounded in AI ethics and real-world needs, these principles provide a holistic lens for societal impact. Benchmarking results on different LMM shows that proprietary models generally lead in reasoning, fairness, and multilinguality, while open-source models excel in robustness and grounding. Most models struggle to balance accuracy with ethical and inclusive behavior. Techniques like Chain-of-Thought prompting and test-time scaling improve alignment. As the first benchmark tailored for HC alignment, HumaniBench offers a rigorous testbed to diagnose limitations, and promote responsible LMM development. All data and code are publicly available for reproducibility. Keywords: HumaniBench, vision-language models, responsible AI benchmark, AI alignment evaluation, AI ethics assessment, fairness in AI models, visual question answering (VQA) benchmark, image captioning evaluation, visual grounding tasks, trustworthy AI models, Chain-of-Thought prompting, test-time scaling, ethical AI development tools.

Paperid: 3240, https://arxiv.org/pdf/2505.07690.pdf

Abstract:
This study aims to address the problem of multi-domain task incremental learning~(MTIL), which requires that vision-language models~(VLMs) continuously acquire new knowledge while maintaining their inherent zero-shot recognition capability. Existing paradigms delegate the testing of unseen-domain samples to the original CLIP, which only prevents the degradation of the model's zero-shot capability but fails to enhance the generalization of the VLM further. To this end, we propose a novel MTIL framework, named AFA, which comprises two core modules: (1) an against forward-forgetting adapter that learns task-invariant information for each dataset in the incremental tasks to enhance the zero-shot recognition ability of VLMs; (2) an against backward-forgetting adapter that strengthens the few-shot learning capability of VLMs while supporting incremental learning. Extensive experiments demonstrate that the AFA method significantly outperforms existing state-of-the-art approaches, especially in few-shot MTIL tasks, and surpasses the inherent zero-shot performance of CLIP in terms of transferability. The code is provided in the Supplementary Material.

Paperid: 3241, https://arxiv.org/pdf/2504.13211.pdf

Abstract:
Recent studies have explored the use of large language models (LLMs) in psychotherapy; however, text-based cognitive behavioral therapy (CBT) models often struggle with client resistance, which can weaken therapeutic alliance. To address this, we propose a multimodal approach that incorporates nonverbal cues, which allows the AI therapist to better align its responses with the client's negative emotional state. Specifically, we introduce a new synthetic dataset, Mirror (Multimodal Interactive Rolling with Resistance), which is a novel synthetic dataset that pairs each client's statements with corresponding facial images. Using this dataset, we train baseline vision language models (VLMs) so that they can analyze facial cues, infer emotions, and generate empathetic responses to effectively manage client resistance. These models are then evaluated in terms of both their counseling skills as a therapist, and the strength of therapeutic alliance in the presence of client resistance. Our results demonstrate that Mirror significantly enhances the AI therapist's ability to handle resistance, which outperforms existing text-based CBT approaches. Human expert evaluations further confirm the effectiveness of our approach in managing client resistance and fostering therapeutic alliance.

Paperid: 3242, https://arxiv.org/pdf/2503.19516.pdf

Abstract:
The growing adoption of Vision-Language-Action (VLA) models in embodied AI intensifies the demand for diverse manipulation demonstrations. However, high costs associated with data collection often result in insufficient data coverage across all scenarios, which limits the performance of the models. It is observed that the spatial reasoning phase (SRP) in large workspace dominates the failure cases. Fortunately, this data can be collected with low cost, underscoring the potential of leveraging inexpensive data to improve model performance. In this paper, we introduce the RoboTron-Craft, a stage-divided and cost-effective pipeline for realistic manipulation generation. Base on this, the RoboTron-Platter method is introduced, a framework that decouples training trajectories into distinct task stages and leverages abundant easily collectible SRP data to enhance VLA model's generalization. Through analysis we demonstrate that sub-task-specific training with additional SRP data with proper proportion can act as a performance catalyst for robot manipulation, maximizing the utilization of costly physical interaction phase (PIP) data. Experiments show that through introducing large proportion of cost-effective SRP trajectories into a limited set of PIP data, we can achieve a maximum improvement of 41\% on success rate in zero-shot scenes, while with the ability to transfer manipulation skill to novel targets. Project available at https://github.com/ notFoundThisPerson/RoboTron-Craft.

Paperid: 3243, https://arxiv.org/pdf/2503.15892.pdf

Abstract:
With the rapid advancement of deep learning, particularly in the field of medical image analysis, an increasing number of Vision-Language Models (VLMs) are being widely applied to solve complex health and biomedical challenges. However, existing research has primarily focused on specific tasks or single modalities, which limits their applicability and generalization across diverse medical scenarios. To address this challenge, we propose UMIT, a unified multi-modal, multi-task VLM designed specifically for medical imaging tasks. UMIT is able to solve various tasks, including visual question answering, disease detection, and medical report generation. In addition, it is applicable to multiple imaging modalities (e.g., X-ray, CT and PET), covering a wide range of applications from basic diagnostics to complex lesion analysis. Moreover, UMIT supports both English and Chinese, expanding its applicability globally and ensuring accessibility to healthcare services in different linguistic contexts. To enhance the model's adaptability and task-handling capability, we design a unique two-stage training strategy and fine-tune UMIT with designed instruction templates. Through extensive empirical evaluation, UMIT outperforms previous methods in five tasks across multiple datasets. The performance of UMIT indicates that it can significantly enhance diagnostic accuracy and workflow efficiency, thus providing effective solutions for medical imaging applications.

Paperid: 3244, https://arxiv.org/pdf/2503.12490.pdf

Abstract:
The application of Vision-Language Models (VLMs) in remote sensing (RS) has demonstrated significant potential in traditional tasks such as scene classification, object detection, and image captioning. However, current models, which excel in Referring Expression Comprehension (REC), struggle with tasks involving complex instructions (e.g., exists multiple conditions) or pixel-level operations like segmentation and change detection. In this white paper, we provide a comprehensive hierarchical summary of vision-language tasks in RS, categorized by the varying levels of cognitive capability required. We introduce the Remote Sensing Vision-Language Task Set (RSVLTS), which includes Open-Vocabulary Tasks (OVT), Referring Expression Tasks (RET), and Described Object Tasks (DOT) with increased difficulty, and Visual Question Answering (VQA) aloneside. Moreover, we propose a novel unified data representation using a set-of-points approach for RSVLTS, along with a condition parser and a self-augmentation strategy based on cyclic referring. These features are integrated into the GeoRSMLLM model, and this enhanced model is designed to handle a broad range of tasks of RSVLTS, paving the way for a more generalized solution for vision-language tasks in geoscience and remote sensing.

Paperid: 3245, https://arxiv.org/pdf/2503.10705.pdf

Abstract:
Vision-Language Models (VLMs) represent a breakthrough in artificial intelligence by integrating visual and textual modalities to achieve impressive zero-shot capabilities. However, VLMs are susceptible to catastrophic forgetting when sequentially fine-tuned on multiple downstream tasks. Existing continual learning methods for VLMs often rely heavily on additional reference datasets, compromise zero-shot performance, or are limited to parameter-efficient fine-tuning scenarios. In this paper, we propose Continual Decoupling-Unifying (ConDU), a novel approach, by introducing model fusion into continual learning for VLMs. ConDU maintains a unified model along with task triggers and prototype sets, employing an iterative process of decoupling task-specific models for previous tasks and unifying them with the model for the newly learned task. Additionally, we introduce an inference strategy for zero-shot scenarios by aggregating predictions from multiple decoupled task-specific models. Extensive experiments across various settings show that ConDU achieves up to a 2\% improvement in average performance across all seen tasks compared to state-of-the-art baselines, while also enhancing zero-shot capabilities relative to the original VLM.

Paperid: 3246, https://arxiv.org/pdf/2503.04058.pdf

Abstract:
The advent of Large Vision-Language Models (LVLMs) has advanced the video-based tasks, such as video captioning and video understanding. Some previous research indicates that taking texts in videos as input can further improve the performance of video understanding. As a type of indispensable information in short videos or movies, subtitles can assist LVLMs to better understand videos. Most existing methods for video subtitle extraction are based on a multi-stage framework, handling each frame independently. They can hardly exploit the temporal information of videos. Although some LVLMs exhibit the robust OCR capability, predicting accurate timestamps for subtitle texts is still challenging. In this paper, we propose an End-to-end Video Subtitle Extraction method, called EVE, which consists of three modules: a vision encoder, an adapter module, and a large language model. To effectively compress the visual tokens from the vision encoder, we propose a novel adapter InterleavedVT to interleave two modalities. It contains a visual compressor and a textual region compressor. The proposed InterleavedVT exploits both the merits of average pooling and Q-Former in token compression. Taking the temporal information of videos into account, we introduce a sliding-window mechanism in the textual region compressor. To benchmark the video subtitle extraction task, we propose a large dataset ViSa including 2.5M videos. Extensive experiments on ViSa demonstrate that the proposed EVE can outperform existing open-sourced tools and LVLMs.

Paperid: 3247, https://arxiv.org/pdf/2502.14976.pdf

Abstract:
Vision-Language Models (VLMs) inherit adversarial vulnerabilities of Large Language Models (LLMs), which are further exacerbated by their multimodal nature. Existing defenses, including adversarial training, input transformations, and heuristic detection, are computationally expensive, architecture-dependent, and fragile against adaptive attacks. We introduce EigenShield, an inference-time defense leveraging Random Matrix Theory to quantify adversarial disruptions in high-dimensional VLM representations. Unlike prior methods that rely on empirical heuristics, EigenShield employs the spiked covariance model to detect structured spectral deviations. Using a Robustness-based Nonconformity Score (RbNS) and quantile-based thresholding, it separates causal eigenvectors, which encode semantic information, from correlational eigenvectors that are susceptible to adversarial artifacts. By projecting embeddings onto the causal subspace, EigenShield filters adversarial noise without modifying model parameters or requiring adversarial training. This architecture-independent, attack-agnostic approach significantly reduces the attack success rate, establishing spectral analysis as a principled alternative to conventional defenses. Our results demonstrate that EigenShield consistently outperforms all existing defenses, including adversarial training, UNIGUARD, and CIDER.

Paperid: 3248, https://arxiv.org/pdf/2501.18867.pdf

Abstract:
Recent advancements in Vision-Language-Action (VLA) models have leveraged pre-trained Vision-Language Models (VLMs) to improve the generalization capabilities. VLMs, typically pre-trained on vision-language understanding tasks, provide rich semantic knowledge and reasoning abilities. However, prior research has shown that VLMs often focus on high-level semantic content and neglect low-level features, limiting their ability to capture detailed spatial information and understand physical dynamics. These aspects, which are crucial for embodied control tasks, remain underexplored in existing pre-training paradigms. In this paper, we investigate the training paradigm for VLAs, and introduce \textbf{UP-VLA}, a \textbf{U}nified VLA model training with both multi-modal \textbf{U}nderstanding and future \textbf{P}rediction objectives, enhancing both high-level semantic comprehension and low-level spatial understanding. Experimental results show that UP-VLA achieves a 33% improvement on the Calvin ABC-D benchmark compared to the previous state-of-the-art method. Additionally, UP-VLA demonstrates improved success rates in real-world manipulation tasks, particularly those requiring precise spatial information.

Paperid: 3249, https://arxiv.org/pdf/2501.16664.pdf

Abstract:
Recent studies have successfully integrated large vision-language models (VLMs) into low-level robotic control by supervised fine-tuning (SFT) with expert robotic datasets, resulting in what we term vision-language-action (VLA) models. Although the VLA models are powerful, how to improve these large models during interaction with environments remains an open question. In this paper, we explore how to further improve these VLA models via Reinforcement Learning (RL), a commonly used fine-tuning technique for large models. However, we find that directly applying online RL to large VLA models presents significant challenges, including training instability that severely impacts the performance of large models, and computing burdens that exceed the capabilities of most local machines. To address these challenges, we propose iRe-VLA framework, which iterates between Reinforcement Learning and Supervised Learning to effectively improve VLA models, leveraging the exploratory benefits of RL while maintaining the stability of supervised learning. Experiments in two simulated benchmarks and a real-world manipulation suite validate the effectiveness of our method.

Paperid: 3250, https://arxiv.org/pdf/2501.16404.pdf

Abstract:
Test-time prompt tuning enhances zero-shot generalization of vision-language models but tends to ignore the relatedness among test samples during inference. Online test-time prompt tuning provides a simple way to leverage the information in previous test samples, albeit with the risk of prompt collapse due to error accumulation. To enhance test-time prompt tuning, we propose DynaPrompt, short for dynamic test-time prompt tuning, exploiting relevant data distribution information while reducing error accumulation. Built on an online prompt buffer, DynaPrompt adaptively selects and optimizes the relevant prompts for each test sample during tuning. Specifically, we introduce a dynamic prompt selection strategy based on two metrics: prediction entropy and probability difference. For unseen test data information, we develop dynamic prompt appending, which allows the buffer to append new prompts and delete the inactive ones. By doing so, the prompts are optimized to exploit beneficial information on specific test data, while alleviating error accumulation. Experiments on fourteen datasets demonstrate the effectiveness of dynamic test-time prompt tuning.

Paperid: 3251, https://arxiv.org/pdf/2501.10144.pdf

Abstract:
Scene understanding in remote sensing often faces challenges in generating accurate representations for complex environments such as various land use areas or coastal regions, which may also include snow, clouds, or haze. To address this, we present a vision-language framework named Spectral LLaVA, which integrates multispectral data with vision-language alignment techniques to enhance scene representation and description. Using the BigEarthNet v2 dataset from Sentinel-2, we establish a baseline with RGB-based scene descriptions and further demonstrate substantial improvements through the incorporation of multispectral information. Our framework optimizes a lightweight linear projection layer for alignment while keeping the vision backbone of SpectralGPT frozen. Our experiments encompass scene classification using linear probing and language modeling for jointly performing scene classification and description generation. Our results highlight Spectral LLaVA's ability to produce detailed and accurate descriptions, particularly for scenarios where RGB data alone proves inadequate, while also enhancing classification performance by refining SpectralGPT features into semantically meaningful representations.

Paperid: 3252, https://arxiv.org/pdf/2410.21926.pdf

Abstract:
We introduce an innovative approach to advancing semantic understanding in zero-shot object goal navigation (ZS-OGN), enhancing the autonomy of robots in unfamiliar environments. Traditional reliance on labeled data has been a limitation for robotic adaptability, which we address by employing a dual-component framework that integrates a GLIP Vision Language Model for initial detection and an InstructionBLIP model for validation. This combination not only refines object and environmental recognition but also fortifies the semantic interpretation, pivotal for navigational decision-making. Our method, rigorously tested in both simulated and real-world settings, exhibits marked improvements in navigation precision and reliability.

Paperid: 3253, https://arxiv.org/pdf/2410.18570.pdf

Abstract:
Object navigation is crucial for robots, but traditional methods require substantial training data and cannot be generalized to unknown environments. Zero-shot object navigation (ZSON) aims to address this challenge, allowing robots to interact with unknown objects without specific training data. Language-driven zero-shot object navigation (L-ZSON) is an extension of ZSON that incorporates natural language instructions to guide robot navigation and interaction with objects. In this paper, we propose a novel Vision Language model with a Tree-of-thought Network (VLTNet) for L-ZSON. VLTNet comprises four main modules: vision language model understanding, semantic mapping, tree-of-thought reasoning and exploration, and goal identification. Among these modules, Tree-of-Thought (ToT) reasoning and exploration module serves as a core component, innovatively using the ToT reasoning framework for navigation frontier selection during robot exploration. Compared to conventional frontier selection without reasoning, navigation using ToT reasoning involves multi-path reasoning processes and backtracking when necessary, enabling globally informed decision-making with higher accuracy. Experimental results on PASTURE and RoboTHOR benchmarks demonstrate the outstanding performance of our model in LZSON, particularly in scenarios involving complex natural language as target instructions.

Paperid: 3254, https://arxiv.org/pdf/2410.10247.pdf

Abstract:
Existing prompt learning methods in Vision-Language Models (VLM) have effectively enhanced the transfer capability of VLM to downstream tasks, but they suffer from a significant decline in generalization due to severe overfitting. To address this issue, we propose a framework named LOBG for vision-language models. Specifically, we use CLIP to filter out fine-grained foreground information that might cause overfitting, thereby guiding prompts with basic visual concepts. To further mitigate overfitting, we devel oped a structural topology preservation (STP) loss at the feature level, which endows the feature space with overall plasticity, allowing effective reshaping of the feature space during optimization. Additionally, we employed hierarchical logit distilation (HLD) at the output level to constrain outputs, complementing STP at the output end. Extensive experimental results demonstrate that our method significantly improves generalization capability and alleviates overfitting compared to state-of-the-art approaches.

Paperid: 3255, https://arxiv.org/pdf/2410.07149.pdf

Abstract:
Vision-Language Models (VLMs) are powerful tools for processing and understanding text and images. We study the processing of visual tokens in the language model component of LLaVA, a prominent VLM. Our approach focuses on analyzing the localization of object information, the evolution of visual token representations across layers, and the mechanism of integrating visual information for predictions. Through ablation studies, we demonstrated that object identification accuracy drops by over 70\% when object-specific tokens are removed. We observed that visual token representations become increasingly interpretable in the vocabulary space across layers, suggesting an alignment with textual tokens corresponding to image content. Finally, we found that the model extracts object information from these refined representations at the last token position for prediction, mirroring the process in text-only language models for factual association tasks. These findings provide crucial insights into how VLMs process and integrate visual information, bridging the gap between our understanding of language and vision models, and paving the way for more interpretable and controllable multimodal systems.

Paperid: 3256, https://arxiv.org/pdf/2410.05273.pdf

Abstract:
Large Vision-Language-Action (VLA) models, leveraging powerful pre trained Vision-Language Models (VLMs) backends, have shown promise in robotic control due to their impressive generalization ability. However, the success comes at a cost. Their reliance on VLM backends with billions of parameters leads to high computational costs and inference latency, limiting the testing scenarios to mainly quasi-static tasks and hindering performance in dynamic tasks requiring rapid interactions. To address these limitations, this paper proposes HiRT, a Hierarchical Robot Transformer framework that enables flexible frequency and performance trade-off. HiRT keeps VLMs running at low frequencies to capture temporarily invariant features while enabling real-time interaction through a high-frequency vision-based policy guided by the slowly updated features. Experiment results in both simulation and real-world settings demonstrate significant improvements over baseline methods. Empirically, in static tasks, we double the control frequency and achieve comparable success rates. Additionally, on novel real-world dynamic ma nipulation tasks which are challenging for previous VLA models, HiRT improves the success rate from 48% to 75%.

Paperid: 3257, https://arxiv.org/pdf/2409.17634.pdf

Abstract:
Large-scale pre-trained Vision-Language Models (VLMs) have gained prominence in various visual and multimodal tasks, yet the deployment of VLMs on downstream application platforms remains challenging due to their prohibitive requirements of training samples and computing resources. Fine-tuning and quantization of VLMs can substantially reduce the sample and computation costs, which are in urgent need. There are two prevailing paradigms in quantization, Quantization-Aware Training (QAT) can effectively quantize large-scale VLMs but incur a huge training cost, while low-bit Post-Training Quantization (PTQ) suffers from a notable performance drop. We propose a method that balances fine-tuning and quantization named ``Prompt for Quantization'' (P4Q), in which we design a lightweight architecture to leverage contrastive loss supervision to enhance the recognition performance of a PTQ model. Our method can effectively reduce the gap between image features and text features caused by low-bit quantization, based on learnable prompts to reorganize textual representations and a low-bit adapter to realign the distributions of image and text features. We also introduce a distillation loss based on cosine similarity predictions to distill the quantized model using a full-precision teacher. Extensive experimental results demonstrate that our P4Q method outperforms prior arts, even achieving comparable results to its full-precision counterparts. For instance, our 8-bit P4Q can theoretically compress the CLIP-ViT/B-32 by 4 $\times$ while achieving 66.94\% Top-1 accuracy, outperforming the learnable prompt fine-tuned full-precision model by 2.24\% with negligible additional parameters on the ImageNet dataset.

Paperid: 3258, https://arxiv.org/pdf/2409.00255.pdf

Abstract:
Vision-language models (VLMs) excel at tasks requiring joint understanding of visual and linguistic information. A particularly promising yet under-explored application for these models lies in answering questions based on various kinds of maps. This study investigates the efficacy of VLMs in answering questions based on choropleth maps, which are widely used for data analysis and representation. To facilitate and encourage research in this area, we introduce a novel map-based question-answering benchmark, consisting of maps from three geographical regions (United States, India, China), each containing 1000 questions. Our benchmark incorporates 43 diverse question templates, requiring nuanced understanding of relative spatial relationships, intricate map features, and complex reasoning. It also includes maps with discrete and continuous values, encompassing variations in color-mapping, category ordering, and stylistic patterns, enabling comprehensive analysis. We evaluate the performance of multiple VLMs on this benchmark, highlighting gaps in their abilities and providing insights for improving such models.

Paperid: 3259, https://arxiv.org/pdf/2601.16449.pdf

Abstract:
Understanding human emotions from multimodal signals poses a significant challenge in affective computing and human-robot interaction. While multimodal large language models (MLLMs) have excelled in general vision-language tasks, their capabilities in emotional reasoning remain limited. The field currently suffers from a scarcity of large-scale datasets with high-quality, descriptive emotion annotations and lacks standardized benchmarks for evaluation. Our preliminary framework, Emotion-LLaMA, pioneered instruction-tuned multimodal learning for emotion reasoning but was restricted by explicit face detectors, implicit fusion strategies, and low-quality training data with limited scale. To address these limitations, we present Emotion-LLaMAv2 and the MMEVerse benchmark, establishing an end-to-end pipeline together with a standardized evaluation setting for emotion recognition and reasoning. Emotion-LLaMAv2 introduces three key advances. First, an end-to-end multiview encoder eliminates external face detection and captures nuanced emotional cues via richer spatial and temporal multiview tokens. Second, a Conv Attention pre-fusion module is designed to enable simultaneous local and global multimodal feature interactions external to the LLM backbone. Third, a perception-to-cognition curriculum instruction tuning scheme within the LLaMA2 backbone unifies emotion recognition and free-form emotion reasoning. To support large-scale training and reproducible evaluation, MMEVerse aggregates twelve publicly available emotion datasets, including IEMOCAP, MELD, DFEW, and MAFW, into a unified multimodal instruction format. The data are re-annotated via a multi-agent pipeline involving Qwen2 Audio, Qwen2.5 VL, and GPT 4o, producing 130k training clips and 36k testing clips across 18 evaluation benchmarks.

Paperid: 3260, https://arxiv.org/pdf/2601.10455.pdf

Abstract:
Surgical planning integrates visual perception, long-horizon reasoning, and procedural knowledge, yet it remains unclear whether current evaluation protocols reliably assess vision-language models (VLMs) in safety-critical settings. Motivated by a goal-oriented view of surgical planning, we define planning correctness via phase-goal satisfiability, where plan validity is determined by expert-defined surgical rules. Based on this definition, we introduce a multicentric meta-evaluation benchmark with valid procedural variations and invalid plans containing order and content errors. Using this benchmark, we show that sequence similarity metrics systematically misjudge planning quality, penalizing valid plans while failing to identify invalid ones. We therefore adopt a rule-based goal-satisfiability metric as a high-precision meta-evaluation reference to assess Video-LLMs under progressively constrained settings, revealing failures due to perception errors and under-constrained reasoning. Structural knowledge consistently improves performance, whereas semantic guidance alone is unreliable and benefits larger models only when combined with structural constraints.

Paperid: 3261, https://arxiv.org/pdf/2512.24591.pdf

Abstract:
Change detection visual question answering (CDVQA) requires answering text queries by reasoning about semantic changes in bi-temporal remote sensing images. A straightforward approach is to boost CDVQA performance with generic vision-language models via supervised fine-tuning (SFT). Despite recent progress, we observe that a significant portion of failures do not stem from clearly incorrect predictions, but from decision ambiguity, where the model assigns similar confidence to the correct answer and strong distractors. To formalize this challenge, we define Decision-Ambiguous Samples (DAS) as instances with a small probability margin between the ground-truth answer and the most competitive alternative. We argue that explicitly optimizing DAS is crucial for improving the discriminability and robustness of CDVQA models. To this end, we propose DARFT, a Decision-Ambiguity-guided Reinforcement Fine-Tuning framework that first mines DAS using an SFT-trained reference policy and then applies group-relative policy optimization on the mined subset. By leveraging multi-sample decoding and intra-group relative advantages, DARFT suppresses strong distractors and sharpens decision boundaries without additional supervision. Extensive experiments demonstrate consistent gains over SFT baselines, particularly under few-shot settings.

Paperid: 3262, https://arxiv.org/pdf/2512.22238.pdf

Abstract:
Large-scale vision-language models (VLMs) have recently achieved remarkable multimodal understanding, but their massive size makes them impractical for deployment on mobile or edge devices. This raises the need for compact yet capable VLMs that can efficiently learn from powerful large teachers. However, distilling knowledge from a large teacher to a small student remains challenging due to their large size gap: the student often fails to reproduce the teacher's complex, high-dimensional representations, leading to unstable learning and degraded performance. To address this, we propose Masters (Masking Teacher and Reinforcing Student), a mask-progressive reinforcement learning (RL) distillation framework. Masters first masks non-dominant weights of the teacher to reduce unnecessary complexity, then progressively restores the teacher by gradually increasing its capacity during training. This strategy allows the student to learn richer representations from the teacher in a smooth and stable manner. To further refine knowledge transfer, Masters integrates an offline RL stage with two complementary rewards: an accuracy reward that measures the correctness of the generated responses, and a distillation reward that quantifies the ease of transferring responses from teacher to student. Unlike online think-answer RL paradigms that are computationally expensive and generate lengthy responses, our offline RL leverages pre-generated responses from masked teachers. These provide rich yet efficient guidance, enabling students to achieve strong performance without requiring the think-answer process.

Paperid: 3263, https://arxiv.org/pdf/2512.17189.pdf

Abstract:
Medical Vision-Language Models (MedVLMs) show immense promise in clinical applicability. However, their reliability is hindered by hallucinations, where models often fail to derive answers from visual evidence, instead relying on learned textual priors. Existing mitigation strategies for MedVLMs have distinct limitations: training-based methods rely on costly expert annotations, limiting scalability, while training-free interventions like contrastive decoding, though data-efficient, apply a global, untargeted correction whose effects in complex real-world clinical settings can be unreliable. To address these challenges, we introduce Anatomical Region-Guided Contrastive Decoding (ARCD), a plug-and-play strategy that mitigates hallucinations by providing targeted, region-specific guidance. Our module leverages an anatomical mask to direct a three-tiered contrastive decoding process. By dynamically re-weighting at the token, attention, and logits levels, it verifiably steers the model's focus onto specified regions, reinforcing anatomical understanding and suppressing factually incorrect outputs. Extensive experiments across diverse datasets, including chest X-ray, CT, brain MRI, and ocular ultrasound, demonstrate our method's effectiveness in improving regional understanding, reducing hallucinations, and enhancing overall diagnostic accuracy.

Paperid: 3264, https://arxiv.org/pdf/2512.12977.pdf

Abstract:
This paper presents VLCache, a cache reuse framework that exploits both Key-Value (KV) cache and encoder cache from prior multimodal inputs to eliminate costly recomputation when the same multimodal inputs recur. Unlike previous heuristic approaches, we formally identify the cumulative reuse error effect and demonstrate how to minimize the non-prefix cache reuse error effectively. We further analyze the varying importance of model layers and propose a dynamic, layer-aware recomputation strategy to balance accuracy and efficiency. Experimental results show that VLCache achieves an accuracy on par with full recomputation, while requiring only 2-5% of the tokens to compute, yielding 1.2x-16x TTFT speedups. We develop an experimental implementation of the proposed VLCache pipeline based on SGLang, enabling significantly faster inference in practical deployments.

Paperid: 3265, https://arxiv.org/pdf/2512.12690.pdf

Abstract:
Recent advances in vision-language models (VLMs) reasoning have been largely attributed to the rise of reinforcement Learning (RL), which has shifted the community's focus away from the supervised fine-tuning (SFT) paradigm. Many studies suggest that introducing the SFT stage not only fails to improve reasoning ability but may also negatively impact model training. In this study, we revisit this RL-centric belief through a systematic and controlled comparison of SFT and RL on VLM Reasoning. Using identical data sources, we find that the relative effectiveness of SFT and RL is conditional and strongly influenced by model capacity, data scale, and data distribution. Contrary to common assumptions, our findings show that SFT plays a crucial role across several scenarios: (1) Effectiveness for weaker models. SFT more reliably elicits reasoning capabilities in smaller or weaker VLMs. (2) Data efficiency. SFT with only 2K achieves comparable or better reasoning performance to RL with 20K. (3) Cross-modal transferability. SFT demonstrates stronger generalization across modalities. Moreover, we identify a pervasive issue of deceptive rewards, where higher rewards fail to correlate with better reasoning accuracy in RL. These results challenge the prevailing "RL over SFT" narrative. They highlight that the role of SFT may have been underestimated and support a more balanced post-training pipeline in which SFT and RL function as complementary components.

Paperid: 3266, https://arxiv.org/pdf/2512.12107.pdf

Abstract:
Echocardiography is the most widely used imaging modality in cardiology, yet its interpretation remains labor-intensive and inherently multimodal, requiring view recognition, quantitative measurements, qualitative assessments, and guideline-based reasoning. While recent vision-language models (VLMs) have achieved broad success in natural images and certain medical domains, their potential in echocardiography has been limited by the lack of large-scale, clinically grounded image-text datasets and the absence of measurement-based reasoning central to echo interpretation. We introduce EchoGround-MIMIC, the first measurement-grounded multimodal echocardiography dataset, comprising 19,065 image-text pairs from 1,572 patients with standardized views, structured measurements, measurement-grounded captions, and guideline-derived disease labels. Building on this resource, we propose EchoVLM, a vision-language model that incorporates two novel pretraining objectives: (i) a view-informed contrastive loss that encodes the view-dependent structure of echocardiographic imaging, and (ii) a negation-aware contrastive loss that distinguishes clinically critical negative from positive findings. Across five types of clinical applications with 36 tasks spanning multimodal disease classification, image-text retrieval, view classification, chamber segmentation, and landmark detection, EchoVLM achieves state-of-the-art performance (86.5% AUC in zero-shot disease classification and 95.1% accuracy in view classification). We demonstrate that clinically grounded multimodal pretraining yields transferable visual representations and establish EchoVLM as a foundation model for end-to-end echocardiography interpretation. We will release EchoGround-MIMIC and the data curation code, enabling reproducibility and further research in multimodal echocardiography interpretation.

Paperid: 3267, https://arxiv.org/pdf/2512.09619.pdf

Abstract:
Most existing Vision-Language-Action (VLA) models rely primarily on RGB information, while ignoring geometric cues crucial for spatial reasoning and manipulation. In this work, we introduce GLaD, a geometry-aware VLA framework that incorporates 3D geometric priors during pretraining through knowledge distillation. Rather than distilling geometric features solely into the vision encoder, we align the LLM's hidden states corresponding to visual tokens with features from a frozen geometry-aware vision transformer (VGGT), ensuring that geometric understanding is deeply integrated into the multimodal representations that drive action prediction. Pretrained on the Bridge dataset with this geometry distillation mechanism, GLaD achieves 94.1% average success rate across four LIBERO task suites, outperforming UniVLA (92.5%) which uses identical pretraining data. These results validate that geometry-aware pretraining enhances spatial reasoning and policy generalization without requiring explicit depth sensors or 3D annotations.

Paperid: 3268, https://arxiv.org/pdf/2512.08524.pdf

Abstract:
Multimodal language models (MLLMs) require large parameter capacity to align high-dimensional visual features with linguistic representations, making them computationally heavy and difficult to deploy efficiently. We introduce a progressive reparameterization strategy that compresses these models by gradually replacing dense feed-forward network blocks with compact Parameterized Hypercomplex Multiplication (PHM) layers. A residual interpolation schedule, together with lightweight reconstruction and knowledge distillation losses, ensures that the PHM modules inherit the functional behavior of their dense counterparts during training. This transition yields substantial parameter and FLOP reductions while preserving strong multimodal alignment, enabling faster inference without degrading output quality. We evaluate the approach on multiple vision-language models (VLMs). Our method maintains performance comparable to the base models while delivering significant reductions in model size and inference latency. Progressive PHM substitution thus offers an architecture-compatible path toward more efficient multimodal reasoning and complements existing low-bit quantization techniques.

Paperid: 3269, https://arxiv.org/pdf/2512.08333.pdf

Abstract:
Generalist robot policies, trained on large and diverse datasets, have demonstrated the ability to generalize across a wide spectrum of behaviors, enabling a single policy to act in varied real-world environments. However, they still fall short on new tasks not covered in the training data. When finetuned on limited demonstrations of a new task, these policies often overfit to the specific demonstrations--not only losing their prior abilities to solve a wide variety of generalist tasks but also failing to generalize within the new task itself. In this work, we aim to develop a method that preserves the generalization capabilities of the generalist policy during finetuning, allowing a single policy to robustly incorporate a new skill into its repertoire. Our goal is a single policy that both learns to generalize to variations of the new task and retains the broad competencies gained from pretraining. We show that this can be achieved through a simple yet effective strategy: interpolating the weights of a finetuned model with that of the pretrained model. We show, across extensive simulated and real-world experiments, that such model merging produces a single model that inherits the generalist abilities of the base model and learns to solve the new task robustly, outperforming both the pretrained and finetuned model on out-of-distribution variations of the new task. Moreover, we show that model merging performance scales with the amount of pretraining data, and enables continual acquisition of new skills in a lifelong learning setting, without sacrificing previously learned generalist abilities.

Paperid: 3270, https://arxiv.org/pdf/2512.05959.pdf

Abstract:
Vision-language models (VLMs) have achieved strong performance in visual question answering (VQA), yet they remain constrained by static training data. Retrieval-Augmented Generation (RAG) mitigates this limitation by enabling access to up-to-date, culturally grounded, and multilingual information; however, multilingual multimodal RAG remains largely underexplored. We introduce M4-RAG, a massive-scale benchmark covering 42 languages and 56 regional dialects and registers, comprising over 80,000 culturally diverse image-question pairs for evaluating retrieval-augmented VQA across languages and modalities. To balance realism with reproducibility, we build a controlled retrieval environment containing millions of carefully curated multilingual documents relevant to the query domains, approximating real-world retrieval conditions while ensuring consistent experimentation. Our systematic evaluation reveals that although RAG consistently benefits smaller VLMs, it fails to scale to larger models and often even degrades their performance, exposing a critical mismatch between model size and current retrieval effectiveness. M4-RAG provides a foundation for advancing next-generation RAG systems capable of reasoning seamlessly across languages, modalities, and cultural contexts.

Paperid: 3271, https://arxiv.org/pdf/2512.02566.pdf

Abstract:
There is a growing interest in developing strong biomedical vision-language models. A popular approach to achieve robust representations is to use web-scale scientific data. However, current biomedical vision-language pretraining typically compresses rich scientific figures and text into coarse figure-level pairs, discarding the fine-grained correspondences that clinicians actually rely on when zooming into local structures. To tackle this issue, we introduce Panel2Patch, a novel data pipeline that mines hierarchical structure from existing biomedical scientific literature, i.e., multi-panel, marker-heavy figures and their surrounding text, and converts them into multi-granular supervision. Given scientific figures and captions, Panel2Patch parses layouts, panels, and visual markers, then constructs hierarchical aligned vision-language pairs at the figure, panel, and patch levels, preserving local semantics instead of treating each figure as a single data sample. Built on this hierarchical corpus, we develop a granularity-aware pretraining strategy that unifies heterogeneous objectives from coarse didactic descriptions to fine region-focused phrases. By applying Panel2Patch to only a small set of the literature figures, we extract far more effective supervision than prior pipelines, enabling substantially better performance with less pretraining data.

Paperid: 3272, https://arxiv.org/pdf/2511.23404.pdf

Abstract:
We present LFM2, a family of Liquid Foundation Models designed for efficient on-device deployment and strong task capabilities. Using hardware-in-the-loop architecture search under edge latency and memory constraints, we obtain a compact hybrid backbone that combines gated short convolutions with a small number of grouped query attention blocks, delivering up to 2x faster prefill and decode on CPUs compared to similarly sized models. The LFM2 family covers 350M-8.3B parameters, including dense models (350M, 700M, 1.2B, 2.6B) and a mixture-of-experts variant (8.3B total, 1.5B active), all with 32K context length. LFM2's training pipeline includes a tempered, decoupled Top-K knowledge distillation objective that avoids support mismatch; curriculum learning with difficulty-ordered data; and a three-stage post-training recipe of supervised fine-tuning, length-normalized preference optimization, and model merging. Pre-trained on 10-12T tokens, LFM2 models achieve strong results across diverse benchmarks; for example, LFM2-2.6B reaches 79.56% on IFEval and 82.41% on GSM8K. We further build multimodal and retrieval variants: LFM2-VL for vision-language tasks, LFM2-Audio for speech, and LFM2-ColBERT for retrieval. LFM2-VL supports tunable accuracy-latency tradeoffs via token-efficient visual processing, while LFM2-Audio separates audio input and output pathways to enable real-time speech-to-speech interaction competitive with models 3x larger. LFM2-ColBERT provides a low-latency encoder for queries and documents, enabling high-performance retrieval across multiple languages. All models are released with open weights and deployment packages for ExecuTorch, llama.cpp, and vLLM, making LFM2 a practical base for edge applications that need fast, memory-efficient inference and strong task capabilities.

Paperid: 3273, https://arxiv.org/pdf/2511.15923.pdf

Abstract:
Vision Language Models (VLMs) are becoming increasingly integral to multimedia understanding; however, they often struggle with domain-specific video classification tasks, particularly in cases with limited data. This stems from a critical \textit{rationale gap}, where sparse domain data is insufficient to bridge the semantic distance between complex spatio-temporal content and abstract classification labels. We propose a two-stage self-improvement paradigm to bridge this gap without new annotations. First, we prompt the VLMs to generate detailed textual rationales for each video, compelling them to articulate the domain-specific logic. The VLM is then fine-tuned on these self-generated rationales, utilizing this intermediate supervision to align its representations with the nuances of the target domain. Second, conventional supervised fine-tuning (SFT) is performed on the task labels, achieving markedly higher effectiveness as a result of the model's pre-acquired domain reasoning. Extensive experiments on diverse datasets demonstrate that our method significantly outperforms direct SFT, validating self-generated rationale as an effective, annotation-efficient paradigm for adapting VLMs to domain-specific video analysis.

Paperid: 3274, https://arxiv.org/pdf/2511.13190.pdf

Abstract:
Recent advances in Multi-modal Large Language Models (MLLMs) have showcased remarkable capabilities in vision-language understanding. However, enabling robust video spatial reasoning-the ability to comprehend object locations, orientations, and inter-object relationships in dynamic 3D scenes-remains a key unsolved challenge. Existing approaches primarily rely on spatially grounded supervised fine-tuning or reinforcement learning, yet we observe that such models often exhibit query-locked reasoning, focusing narrowly on objects explicitly mentioned in the prompt while ignoring critical contextual cues. To address this limitation, we propose Object-Centric 3D Rollout (OCR), a novel strategy that introduces structured perturbations to the 3D geometry of selected objects during training. By degrading object-specific visual cues and projecting the altered geometry into 2D space, OCR compels the model to reason holistically across the entire scene. We further design a rollout-based training pipeline that jointly leverages vanilla and region-noisy videos to optimize spatial reasoning trajectories. Experiments demonstrate state-of-the-art performance: our 3B-parameter model achieves 47.5% accuracy on VSI-Bench, outperforming several 7B baselines. Ablations confirm OCR's superiority over prior rollout strategies (e.g., T-GRPO, NoisyRollout).

Paperid: 3275, https://arxiv.org/pdf/2511.12001.pdf

Abstract:
Explanations are often promoted as tools for transparency, but they can also foster confirmation bias; users may assume reasoning is correct whenever outputs appear acceptable. We study this double-edged role of Chain-of-Thought (CoT) explanations in multimodal moral scenarios by systematically perturbing reasoning chains and manipulating delivery tones. Specifically, we analyze reasoning errors in vision language models (VLMs) and how they impact user trust and the ability to detect errors. Our findings reveal two key effects: (1) users often equate trust with outcome agreement, sustaining reliance even when reasoning is flawed, and (2) the confident tone suppresses error detection while maintaining reliance, showing that delivery styles can override correctness. These results highlight how CoT explanations can simultaneously clarify and mislead, underscoring the need for NLP systems to provide explanations that encourage scrutiny and critical thinking rather than blind trust. All code will be released publicly.

Paperid: 3276, https://arxiv.org/pdf/2511.05642.pdf

Abstract:
The deployment of artificial intelligence models at the edge is increasingly critical for autonomous robots operating in GPS-denied environments where local, resource-efficient reasoning is essential. This work demonstrates the feasibility of deploying small Vision-Language Models (VLMs) on mobile robots to achieve real-time scene understanding and reasoning under strict computational constraints. Unlike prior approaches that separate perception from mobility, the proposed framework enables simultaneous movement and reasoning in dynamic environments using only on-board hardware. The system integrates a compact VLM with multimodal perception to perform contextual interpretation directly on embedded hardware, eliminating reliance on cloud connectivity. Experimental validation highlights the balance between computational efficiency, task accuracy, and system responsiveness. Implementation on a mobile robot confirms one of the first successful deployments of small VLMs for concurrent reasoning and mobility at the edge. This work establishes a foundation for scalable, assured autonomy in applications such as service robotics, disaster response, and defense operations.

Paperid: 3277, https://arxiv.org/pdf/2511.01210.pdf

Abstract:
Vision-language-action (VLA) models have shown strong generalization for robotic action prediction through large-scale vision-language pretraining. However, most existing models rely solely on RGB cameras, limiting their perception and, consequently, manipulation capabilities. We present OmniVLA, an omni-modality VLA model that integrates novel sensing modalities for physically-grounded spatial intelligence beyond RGB perception. The core of our approach is the sensor-masked image, a unified representation that overlays spatially grounded and physically meaningful masks onto the RGB images, derived from sensors including an infrared camera, a mmWave radar, and a microphone array. This image-native unification keeps sensor input close to RGB statistics to facilitate training, provides a uniform interface across sensor hardware, and enables data-efficient learning with lightweight per-sensor projectors. Built on this, we present a multisensory vision-language-action model architecture and train the model based on an RGB-pretrained VLA backbone. We evaluate OmniVLA on challenging real-world tasks where sensor-modality perception guides the robotic manipulation. OmniVLA achieves an average task success rate of 84%, significantly outperforms both RGB-only and raw-sensor-input baseline models by 59% and 28% respectively, meanwhile showing higher learning efficiency and stronger generalization capability.

Paperid: 3278, https://arxiv.org/pdf/2511.00091.pdf

Abstract:
Supervised fine-tuning (SFT) has become the de facto post-training strategy for large vision-language-action (VLA) models, but its reliance on costly human demonstrations limits scalability and generalization. We propose Probe, Learn, Distill (PLD), a three-stage plug-and-play framework that improves VLAs through residual reinforcement learning (RL) and distribution-aware data collection. In Stage 1, we train lightweight residual actors to probe failure regions of the VLA generalist. In Stage 2, we use a hybrid rollout scheme that aligns collected trajectories with the generalist's deployment distribution while capturing recovery behaviors. In Stage 3, we distill the curated trajectories back into the generalist with standard SFT. PLD achieves near-saturated 99% task success on LIBERO, over 50% gains in SimplerEnv, and 100% success on real-world Franka and YAM arm manipulation tasks. Ablations show that residual probing and distribution-aware replay are key to collecting deployment-aligned data that improves both seen and unseen tasks, offering a scalable path toward self-improving VLA models.

Paperid: 3279, https://arxiv.org/pdf/2510.18269.pdf

Abstract:
Unlike offline processing, streaming video vision-language models face two fundamental constraints: causality and accumulation. Causality prevents access to future frames that offline methods exploit, while accumulation causes tokens to grow unbounded, creating efficiency bottlenecks. However, existing approaches only regulate post-LLM kv-cache, leaving costly pre-LLM prefill unchanged. We introduce StreamingTOM, a training-free, plug-and-play two-stage framework that addresses both pre-LLM and post-LLM bottlenecks with predictable latency. Causal Temporal Reduction imposes a fixed per-frame budget and selects tokens based on adjacent-frame changes and token saliency, drastically reducing per-frame prefill cost by processing only a compact subset of visual tokens per frame instead of all visual tokens. Online Quantized Memory stores tokens in 4-bit format, retrieves relevant groups on demand, and dequantizes them, keeping the active kv-cache bounded regardless of stream length. Experiments demonstrate our method achieves $15.7\times$ kv-cache compression, $1.2\times$ lower peak memory and $2\times$ faster TTFT compared to prior SOTA. StreamingTOM maintains state-of-the-art accuracy among training-free methods with an average of $63.8\%$ on offline benchmarks and $55.8\%/3.7$ on RVS. These results highlight the practical benefits of our two-stage approach for efficient streaming video understanding with bounded growth.

Paperid: 3280, https://arxiv.org/pdf/2510.10921.pdf

Abstract:
Fine-grained vision-language understanding requires precise alignment between visual content and linguistic descriptions, a capability that remains limited in current models, particularly in non-English settings. While models like CLIP perform well on global alignment, they often struggle to capture fine-grained details in object attributes, spatial relations, and linguistic expressions, with limited support for bilingual comprehension. To address these challenges, we introduce FG-CLIP 2, a bilingual vision-language model designed to advance fine-grained alignment for both English and Chinese. Our approach leverages rich fine-grained supervision, including region-text matching and long-caption modeling, alongside multiple discriminative objectives. We further introduce the Textual Intra-modal Contrastive (TIC) loss to better distinguish semantically similar captions. Trained on a carefully curated mixture of large-scale English and Chinese data, FG-CLIP 2 achieves powerful bilingual performance. To enable rigorous evaluation, we present a new benchmark for Chinese multimodal understanding, featuring long-caption retrieval and bounding box classification. Extensive experiments on 29 datasets across 8 tasks show that FG-CLIP 2 outperforms existing methods, achieving state-of-the-art results in both languages. We release the model, code, and benchmark to facilitate future research on bilingual fine-grained alignment.

Paperid: 3281, https://arxiv.org/pdf/2510.07545.pdf

Abstract:
Large Vision-Language Models (LVLMs) with only 7B parameters have shown promise as automated judges in chart comprehension tasks. However, tiny models (<=2B parameters) still perform poorly as judges, limiting their real-world use in resource-constrained settings. To address this, we propose two approaches to ensure cost-efficient evaluation: (i) multi-criteria prompting, which combines separate evaluation criteria into a single query, and (ii) domain-adaptive transfer learning, in which we fine-tune a 2B-parameter LVLM on synthetic judgments in a chart dataset to create the ChartJudge. Experiments show that multi-criteria prompting exposes robustness gaps, which led to a huge drop in performance for 7B models, including specialized LVLM judges like LLaVA-Critic. In addition, we find that our tiny LVLM (ChartJudge) can effectively transfer knowledge from one dataset to another to make it a more specialized model. Our fine-grained analysis across chart types and query complexities offers actionable insights into trade-offs between model size, prompt design, and transferability, enabling scalable, low-cost evaluation for chart reasoning tasks. Our code and the data will be made publicly available.

Paperid: 3282, https://arxiv.org/pdf/2510.05681.pdf

Abstract:
Vision-Language-Action models (VLAs) have demonstrated remarkable performance in robot control. However, they remain fundamentally limited in tasks that require high precision due to their single-inference paradigm. While test-time scaling approaches using external verifiers have shown promise, they require additional training and fail to generalize to unseen conditions. We propose Masking Distribution Guided Selection (MG-Select), a novel test-time scaling framework for VLAs that leverages the model's internal properties without requiring additional training or external modules. Our approach utilizes KL divergence from a reference action token distribution as a confidence metric for selecting the optimal action from multiple candidates. We introduce a reference distribution generated by the same VLA but with randomly masked states and language conditions as inputs, ensuring maximum uncertainty while remaining aligned with the target task distribution. Additionally, we propose a joint training strategy that enables the model to learn both conditional and unconditional distributions by applying dropout to state and language conditions, thereby further improving the quality of the reference distribution. Our experiments demonstrate that MG-Select achieves significant performance improvements, including a 28%/35% improvement in real-world in-distribution/out-of-distribution tasks, along with a 168% relative gain on RoboCasa pick-and-place tasks trained with 30 demonstrations.

Paperid: 3283, https://arxiv.org/pdf/2510.05580.pdf

Abstract:
Vision-Language-Action (VLA) models show promise in embodied reasoning, yet remain far from true generalists-they often require task-specific fine-tuning, and generalize poorly to unseen tasks. We propose MetaVLA, a unified, backbone-agnostic post-training framework for efficient and scalable alignment. MetaVLA introduces Context-Aware Meta Co-Training, which consolidates diverse target tasks into a single fine-tuning stage while leveraging structurally diverse auxiliary tasks to improve in-domain generalization. Unlike naive multi-task SFT, MetaVLA integrates a lightweight meta-learning mechanism-derived from Attentive Neural Processes-to enable rapid adaptation from diverse contexts with minimal architectural change or inference overhead. On the LIBERO benchmark, MetaVLA with six auxiliary tasks outperforms OpenVLA by up to 8.0% on long-horizon tasks, reduces training steps from 240K to 75K, and cuts GPU time by ~76%. These results show that scalable, low-resource post-training is achievable-paving the way toward general-purpose embodied agents. Code will be available.

Paperid: 3284, https://arxiv.org/pdf/2510.02898.pdf

Abstract:
Zero-shot captioners are recently proposed models that utilize common-space vision-language representations to caption images without relying on paired image-text data. To caption an image, they proceed by textually decoding a text-aligned image feature, but they limit their scope to global representations and whole-image captions. We present Patch-ioner, a unified framework for zero-shot captioning that shifts from an image-centric to a patch-centric paradigm, enabling the captioning of arbitrary regions without the need of region-level supervision. Instead of relying on global image representations, we treat individual patches as atomic captioning units and aggregate them to describe arbitrary regions, from single patches to non-contiguous areas and entire images. We analyze the key ingredients that enable current latent captioners to work in our novel proposed framework. Experiments demonstrate that backbones producing meaningful, dense visual features, such as DINO, are key to achieving state-of-the-art performance in multiple region-based captioning tasks. Compared to other baselines and state-of-the-art competitors, our models achieve better performance on zero-shot dense, region-set, and a newly introduced trace captioning task, highlighting the effectiveness of patch-wise semantic representations for scalable caption generation. Project page at https://paciosoft.com/Patch-ioner/ .

Paperid: 3285, https://arxiv.org/pdf/2510.01582.pdf

Abstract:
We develop ImageNet-Think, a multimodal reasoning dataset designed to aid the development of Vision Language Models (VLMs) with explicit reasoning capabilities. Our dataset is built on 250,000 images from ImageNet21k dataset, providing structured thinking tokens and corresponding answers. Our synthetic dataset is generated by two state-of-the-art VLMs: GLM-4.1V-9B-Thinking and Kimi-VL-A3B-Thinking-2506. Each image is accompanied by two pairs of thinking-answer sequences, creating a resource for training and evaluating multimodal reasoning models. We capture the step-by-step reasoning process of VLMs and the final descriptive answers. Our goal with this dataset is to enable the development of more robust VLMs while contributing to the broader understanding of multimodal reasoning mechanisms. The dataset and evaluation benchmarks will be publicly available to aid research in reasoning/thinking multimodal VLMs.

Paperid: 3286, https://arxiv.org/pdf/2510.00806.pdf

Abstract:
Current video generation models produce physically inconsistent motion that violates real-world dynamics. We propose TrajVLM-Gen, a two-stage framework for physics-aware image-to-video generation. First, we employ a Vision Language Model to predict coarse-grained motion trajectories that maintain consistency with real-world physics. Second, these trajectories guide video generation through attention-based mechanisms for fine-grained motion refinement. We build a trajectory prediction dataset based on video tracking data with realistic motion patterns. Experiments on UCF-101 and MSR-VTT demonstrate that TrajVLM-Gen outperforms existing methods, achieving competitive FVD scores of 545 on UCF-101 and 539 on MSR-VTT.

Paperid: 3287, https://arxiv.org/pdf/2510.00604.pdf

Abstract:
Following language instructions, vision-language navigation (VLN) agents are tasked with navigating unseen environments. While augmenting multifaceted visual representations has propelled advancements in VLN, the significance of foreground and background in visual observations remains underexplored. Intuitively, foreground regions provide semantic cues, whereas the background encompasses spatial connectivity information. Inspired on this insight, we propose a Consensus-driven Online Feature Augmentation strategy (COFA) with alternative foreground and background features to facilitate the navigable generalization. Specifically, we first leverage semantically-enhanced landmark identification to disentangle foreground and background as candidate augmented features. Subsequently, a consensus-driven online augmentation strategy encourages the agent to consolidate two-stage voting results on feature preferences according to diverse instructions and navigational locations. Experiments on REVERIE and R2R demonstrate that our online foreground-background augmentation boosts the generalization of baseline and attains state-of-the-art performance.

Paperid: 3288, https://arxiv.org/pdf/2509.25748.pdf

Abstract:
Ultrasound is crucial in modern medicine but faces challenges like operator dependence, image noise, and real-time scanning, hindering AI integration. While large multimodal models excel in other medical imaging areas, they struggle with ultrasound's complexities. To address this, we introduce Dolphin v1.0 (V1) and its reasoning-augmented version, Dolphin R1-the first large-scale multimodal ultrasound foundation models unifying diverse clinical tasks in a single vision-language framework.To tackle ultrasound variability and noise, we curated a 2-million-scale multimodal dataset, combining textbook knowledge, public data, synthetic samples, and general corpora. This ensures robust perception, generalization, and clinical adaptability.The Dolphin series employs a three-stage training strategy: domain-specialized pretraining, instruction-driven alignment, and reinforcement-based refinement. Dolphin v1.0 delivers reliable performance in classification, detection, regression, and report generation. Dolphin R1 enhances diagnostic inference, reasoning transparency, and interpretability through reinforcement learning with ultrasound-specific rewards.Evaluated on U2-Bench across eight ultrasound tasks, Dolphin R1 achieves a U2-score of 0.5835-over twice the second-best model (0.2968) setting a new state of the art. Dolphin v1.0 also performs competitively, validating the unified framework. Comparisons show reasoning-enhanced training significantly improves diagnostic accuracy, consistency, and interpretability, highlighting its importance for high-stakes medical AI.

Paperid: 3289, https://arxiv.org/pdf/2509.23928.pdf

Abstract:
Speculative decoding is an effective approach for accelerating inference in Large Language models (LLMs), but its adaptation to Vision-Language models (VLMs) remains challenging for additional visual tokens in multimodal inputs. First, owing to the fact that the drafter and the target VLM may derived from different families, the semantic representations of visual tokens in the target VLM are misaligned with those in the drafter, introducing bias into the KV-cache during the prefill stage. Second, the large number of visual tokens substantially slows down the drafter's self-attention during the decoding stage. We propose Hiding Visual Tokens from the Drafter for Speculative Decoding in Vision-Language Models (HiViS), an explicit-implicit input decomposition framework that alleviates the above inefficiency. All visual tokens are removed from the drafter's input, retaining only textual tokens as explicit inputs, while directly reusing the target VLM's corresponding last-layer hidden states as implicit visual information without additional processing. To train the drafter efficiently, we introduces multi-step self-feedback training strategy with dynamic data selection and sequential embedding supervision to simulate reasoning during training. Our approach compresses the prefill sequence length of the drafter to only 0.7%-1.3% of the target VLM's input, while maintaining lossless generation quality. Extensive experiments across diverse models and tasks demonstrate up to 2.65x speedup, confirming the effectiveness of HiViS in accelerating VLM inference.

Paperid: 3290, https://arxiv.org/pdf/2509.23898.pdf

Abstract:
Structured sparsity regularization offers a principled way to compact neural networks, but its non-differentiability breaks compatibility with conventional stochastic gradient descent and requires either specialized optimizers or additional post-hoc pruning without formal guarantees. In this work, we propose $D$-Gating, a fully differentiable structured overparameterization that splits each group of weights into a primary weight vector and multiple scalar gating factors. We prove that any local minimum under $D$-Gating is also a local minimum using non-smooth structured $L_{2,2/D}$ penalization, and further show that the $D$-Gating objective converges at least exponentially fast to the $L_{2,2/D}$-regularized loss in the gradient flow limit. Together, our results show that $D$-Gating is theoretically equivalent to solving the original group sparsity problem, yet induces distinct learning dynamics that evolve from a non-sparse regime into sparse optimization. We validate our theory across vision, language, and tabular tasks, where $D$-Gating consistently delivers strong performance-sparsity tradeoffs and outperforms both direct optimization of structured penalties and conventional pruning baselines.

Paperid: 3291, https://arxiv.org/pdf/2509.18711.pdf

Abstract:
Remote sensing visual grounding (RSVG) aims to localize objects in remote sensing images based on free-form natural language expressions. Existing approaches are typically constrained to closed-set vocabularies, limiting their applicability in open-world scenarios. While recent attempts to leverage generic foundation models for open-vocabulary RSVG, they overly rely on expensive high-quality datasets and time-consuming fine-tuning. To address these limitations, we propose \textbf{RSVG-ZeroOV}, a training-free framework that aims to explore the potential of frozen generic foundation models for zero-shot open-vocabulary RSVG. Specifically, RSVG-ZeroOV comprises three key stages: (i) Overview: We utilize a vision-language model (VLM) to obtain cross-attention\footnote[1]{In this paper, although decoder-only VLMs use self-attention over all tokens, we refer to the image-text interaction part as cross-attention to distinguish it from pure visual self-attention.}maps that capture semantic correlations between text queries and visual regions. (ii) Focus: By leveraging the fine-grained modeling priors of a diffusion model (DM), we fill in gaps in structural and shape information of objects, which are often overlooked by VLM. (iii) Evolve: A simple yet effective attention evolution module is introduced to suppress irrelevant activations, yielding purified segmentation masks over the referred objects. Without cumbersome task-specific training, RSVG-ZeroOV offers an efficient and scalable solution. Extensive experiments demonstrate that the proposed framework consistently outperforms existing weakly-supervised and zero-shot methods.

Paperid: 3292, https://arxiv.org/pdf/2509.08126.pdf

Abstract:
Enabling robots to grasp objects specified through natural language is essential for effective human-robot interaction, yet it remains a significant challenge. Existing approaches often struggle with open-form language expressions and typically assume unambiguous target objects without duplicates. Moreover, they frequently rely on costly, dense pixel-wise annotations for both object grounding and grasp configuration. We present Attribute-based Object Grounding and Robotic Grasping (OGRG), a novel framework that interprets open-form language expressions and performs spatial reasoning to ground target objects and predict planar grasp poses, even in scenes containing duplicated object instances. We investigate OGRG in two settings: (1) Referring Grasp Synthesis (RGS) under pixel-wise full supervision, and (2) Referring Grasp Affordance (RGA) using weakly supervised learning with only single-pixel grasp annotations. Key contributions include a bi-directional vision-language fusion module and the integration of depth information to enhance geometric reasoning, improving both grounding and grasping performance. Experiment results show that OGRG outperforms strong baselines in tabletop scenes with diverse spatial language instructions. In RGS, it operates at 17.59 FPS on a single NVIDIA RTX 2080 Ti GPU, enabling potential use in closed-loop or multi-object sequential grasping, while delivering superior grounding and grasp prediction accuracy compared to all the baselines considered. Under the weakly supervised RGA setting, OGRG also surpasses baseline grasp-success rates in both simulation and real-robot trials, underscoring the effectiveness of its spatial reasoning design. Project page: https://z.umn.edu/ogrg

Paperid: 3293, https://arxiv.org/pdf/2509.05718.pdf

Abstract:
Vision-language models (VLMs) hold promise for enhancing visualization tools, but effective human-AI collaboration hinges on a shared perceptual understanding of visual content. Prior studies assessed VLM visualization literacy through interpretive tasks, revealing an over-reliance on textual cues rather than genuine visual analysis. Our study investigates a more foundational skill underpinning such literacy: the ability of VLMs to recognize a chart's core visual properties as humans do. We task 13 diverse VLMs with classifying scientific visualizations based solely on visual stimuli, according to three criteria: purpose (e.g., schematic, GUI, visualization), encoding (e.g., bar, point, node-link), and dimensionality (e.g., 2D, 3D). Using expert labels from the human-centric VisType typology as ground truth, we find that VLMs often identify purpose and dimensionality accurately but struggle with specific encoding types. Our preliminary results show that larger models do not always equate to superior performance and highlight the need for careful integration of VLMs in visualization tasks, with human supervision to ensure reliable outcomes.

Paperid: 3294, https://arxiv.org/pdf/2509.01944.pdf

Abstract:
Vision-Language-Action (VLA) models in autonomous driving systems have recently demonstrated transformative potential by integrating multimodal perception with decision-making capabilities. However, the interpretability and coherence of the decision process and the plausibility of action sequences remain largely underexplored. To address these issues, we propose AutoDrive-R$^2$, a novel VLA framework that enhances both reasoning and self-reflection capabilities of autonomous driving systems through chain-of-thought (CoT) processing and reinforcement learning (RL). Specifically, we first propose an innovative CoT dataset named nuScenesR$^2$-6K for supervised fine-tuning, which effectively builds cognitive bridges between input information and output trajectories through a four-step logical chain with self-reflection for validation. Moreover, to maximize both reasoning and self-reflection during the RL stage, we further employ the Group Relative Policy Optimization (GRPO) algorithm within a physics-grounded reward framework that incorporates spatial alignment, vehicle dynamic, and temporal smoothness criteria to ensure reliable and realistic trajectory planning. Extensive evaluation results across both nuScenes and Waymo datasets demonstrates the state-of-the-art performance and robust generalization capacity of our proposed method.

Paperid: 3295, https://arxiv.org/pdf/2508.17230.pdf

Abstract:
General visual representations learned from web-scale datasets for robotics have achieved great success in recent years, enabling data-efficient robot learning on manipulation tasks; yet these pre-trained representations are mostly on 2D images, neglecting the inherent 3D nature of the world. However, due to the scarcity of large-scale 3D data, it is still hard to extract a universal 3D representation from web datasets. Instead, we are seeking a general visual pre-training framework that could improve all 3D representations as an alternative. Our framework, called FVP, is a novel 4D Visual Pre-training framework for real-world robot learning. FVP frames the visual pre-training objective as a next-point-cloud-prediction problem, models the prediction model as a diffusion model, and pre-trains the model on the larger public datasets directly. Across twelve real-world manipulation tasks, FVP boosts the average success rate of 3D Diffusion Policy (DP3) for these tasks by 28%. The FVP pre-trained DP3 achieves state-of-the-art performance across imitation learning methods. Moreover, the efficacy of FVP adapts across various point cloud encoders and datasets. Finally, we apply FVP to the RDT-1B, a larger Vision-Language-Action robotic model, enhancing its performance on various robot tasks. Our project page is available at: https://4d-visual-pretraining.github.io/

Paperid: 3296, https://arxiv.org/pdf/2508.16158.pdf

Abstract:
The rich textual information of large vision-language models (VLMs) combined with the powerful generative prior of pre-trained text-to-image (T2I) diffusion models has achieved impressive performance in single-image super-resolution (SISR). However, existing methods still face significant challenges in generating clear and accurate regional details, particularly in scenarios involving multiple objects. This challenge primarily stems from a lack of fine-grained regional descriptions and the models' insufficient ability to capture complex prompts. To address these limitations, we propose a Regional Attention Guided Super-Resolution (RAGSR) method that explicitly extracts localized fine-grained information and effectively encodes it through a novel regional attention mechanism, enabling both enhanced detail and overall visually coherent SR results. Specifically, RAGSR localizes object regions in an image and assigns fine-grained caption to each region, which are formatted as region-text pairs as textual priors for T2I models. A regional guided attention is then leveraged to ensure that each region-text pair is properly considered in the attention process while preventing unwanted interactions between unrelated region-text pairs. By leveraging this attention mechanism, our approach offers finer control over the integration of text and image information, thereby effectively overcoming limitations faced by traditional SISR techniques. Experimental results on benchmark datasets demonstrate that our approach exhibits superior performance in generating perceptually authentic visual details while maintaining contextual consistency compared to existing approaches.

Paperid: 3297, https://arxiv.org/pdf/2508.09450.pdf

Abstract:
Charts are very common for exploring data and communicating insights, but extracting key takeaways from charts and articulating them in natural language can be challenging. The chart-to-text task aims to automate this process by generating textual summaries of charts. While with the rapid advancement of large Vision-Language Models (VLMs), we have witnessed great progress in this domain, little to no attention has been given to potential biases in their outputs. This paper investigates how VLMs can amplify geo-economic biases when generating chart summaries, potentially causing societal harm. Specifically, we conduct a large-scale evaluation of geo-economic biases in VLM-generated chart summaries across 6,000 chart-country pairs from six widely used proprietary and open-source models to understand how a country's economic status influences the sentiment of generated summaries. Our analysis reveals that existing VLMs tend to produce more positive descriptions for high-income countries compared to middle- or low-income countries, even when country attribution is the only variable changed. We also find that models such as GPT-4o-mini, Gemini-1.5-Flash, and Phi-3.5 exhibit varying degrees of bias. We further explore inference-time prompt-based debiasing techniques using positive distractors but find them only partially effective, underscoring the complexity of the issue and the need for more robust debiasing strategies. Our code and dataset are publicly available here.

Paperid: 3298, https://arxiv.org/pdf/2508.09099.pdf

Abstract:
Large vision language models exhibit notable limitations on Geometry Problem Solving (GPS) because of their unreliable diagram interpretation and pure natural-language reasoning. A recent line of work mitigates this by using symbolic solvers: the model directly generates a formal program that a geometry solver can execute. However, this direct program generation lacks intermediate reasoning, making the decision process opaque and prone to errors. In this work, we explore a new approach that integrates Chain-of-Thought (CoT) with formal language. The model interleaves natural language reasoning with incremental emission of solver-executable code, producing a hybrid reasoning trace in which critical derivations are expressed in formal language. To teach this behavior at scale, we combine (1) supervised fine-tuning on an 11K newly developed synthetic dataset with interleaved natural language reasoning and automatic formalization, and (2) solver-in-the-loop reinforcement learning that jointly optimizes both the CoT narrative and the resulting program through outcome-based rewards. Built on Qwen2.5-VL-7B, our new model, named GF-Reasoner, achieves up to 15% accuracy improvements on standard GPS benchmarks, surpassing both 7B-scale peers and the much larger model Qwen2.5-VL-72B. By exploiting high-order geometric knowledge and offloading symbolic computation to the solver, the generated reasoning traces are noticeably shorter and cleaner. Furthermore, we present a comprehensive analysis of method design choices (e.g., reasoning paradigms, data synthesis, training epochs, etc.), providing actionable insights for future research.

Paperid: 3299, https://arxiv.org/pdf/2508.08688.pdf

Abstract:
Vision-language models (VLMs) have made significant strides in reasoning, yet they often struggle with complex multimodal tasks and tend to generate overly verbose outputs. A key limitation is their reliance on chain-of-thought (CoT) reasoning, despite many tasks benefiting from alternative topologies like trees or graphs. To address this, we introduce STELAR-Vision, a training framework for topology-aware reasoning. At its core is TopoAug, a synthetic data pipeline that enriches training with diverse topological structures. Using supervised fine-tuning and reinforcement learning, we post-train Qwen2VL models with both accuracy and efficiency in mind. Additionally, we propose Frugal Learning, which reduces output length with minimal accuracy loss. On MATH-V and VLM-S2H, STELAR-Vision improves accuracy by 9.7% over its base model and surpasses the larger Qwen2VL-72B-Instruct by 7.3%. On five out-of-distribution benchmarks, it outperforms Phi-4-Multimodal-Instruct by up to 28.4% and LLaMA-3.2-11B-Vision-Instruct by up to 13.2%, demonstrating strong generalization. Compared to Chain-Only training, our approach achieves 4.3% higher overall accuracy on in-distribution datasets and consistently outperforms across all OOD benchmarks. We have released datasets, and code will be available.

Paperid: 3300, https://arxiv.org/pdf/2508.02095.pdf

Abstract:
Vision language models (VLMs) have shown remarkable capabilities in integrating linguistic and visual reasoning but remain fundamentally limited in understanding dynamic spatiotemporal interactions. Humans effortlessly track and reason about object movements, rotations, and perspective shifts-abilities essential for robust dynamic real-world understanding yet notably lacking in current VLMs. In this paper, we introduce VLM4D, the first benchmark specifically designed to evaluate the spatiotemporal reasoning capabilities of VLMs. Our benchmark comprises diverse real-world and synthetic videos accompanied by carefully curated question-answer pairs emphasizing translational and rotational motions, perspective awareness, and motion continuity. Through comprehensive evaluations of state-of-the-art open and closed-source VLMs, we identify significant performance gaps compared to human baselines, highlighting fundamental deficiencies in existing models. Extensive analysis reveals that VLMs struggle particularly with integrating multiple visual cues and maintaining temporal coherence. We further explore promising directions, such as leveraging 4D feature field reconstruction and targeted spatiotemporal supervised fine-tuning, demonstrating their effectiveness in enhancing spatiotemporal comprehension. Our work aims to encourage deeper exploration into improving VLMs' spatial and temporal grounding, paving the way towards more capable and reliable visual intelligence for dynamic environments.

Paperid: 3301, https://arxiv.org/pdf/2507.23134.pdf

Abstract:
Unlike closed-vocabulary 3D instance segmentation that is often trained end-to-end, open-vocabulary 3D instance segmentation (OV-3DIS) often leverages vision-language models (VLMs) to generate 3D instance proposals and classify them. While various concepts have been proposed from existing research, we observe that these individual concepts are not mutually exclusive but complementary. In this paper, we propose a new state-of-the-art solution for OV-3DIS by carefully designing a recipe to combine the concepts together and refining them to address key challenges. Our solution follows the two-stage scheme: 3D proposal generation and instance classification. We employ robust 3D tracking-based proposal aggregation to generate 3D proposals and remove overlapped or partial proposals by iterative merging/removal. For the classification stage, we replace the standard CLIP model with Alpha-CLIP, which incorporates object masks as an alpha channel to reduce background noise and obtain object-centric representation. Additionally, we introduce the standardized maximum similarity (SMS) score to normalize text-to-proposal similarity, effectively filtering out false positives and boosting precision. Our framework achieves state-of-the-art performance on ScanNet200 and S3DIS across all AP and AR metrics, even surpassing an end-to-end closed-vocabulary method.

Paperid: 3302, https://arxiv.org/pdf/2507.21584.pdf

Abstract:
Multimodal large language models (MLLMs) enable vision-language reasoning, yet often generate plausible outputs that are factually incorrect or visually ungrounded, thereby compromising their reliability. Direct preference optimization (DPO) is a common strategy for correcting hallucinations by aligning model outputs with human preferences. Existing DPO strategies typically treat hallucination-related preferences as fixed targets, relying on static supervision signals during training. This approach tends to overfit to superficial linguistic cues in preference data, leading to distributional rigidity and spurious correlations that impair grounding in causally relevant visual information. To overcome this limitation, we propose TARS, a token-adaptive preference strategy that reformulates DPO as a min-max optimization problem. TARS maximizes token-level distributional shifts under semantic constraints to simulate alignment uncertainty, and simultaneously minimizes the expected preference loss under these controlled perturbations. This joint objective preserves causal grounding while mitigating overfitting to preference patterns, thereby reducing hallucinations in multimodal reasoning. We evaluate TARS on multiple hallucination benchmarks and find consistently strong performance. Using only 4.8k preference samples and no expert feedback, TARS reduces hallucination rates from 26.4% to 13.2% and decreases cognition value from 2.5 to 0.4. It outperforms standard DPO and matches GPT-4o on several key metrics.

Paperid: 3303, https://arxiv.org/pdf/2507.09209.pdf

Abstract:
The rapid advancements in Vision Language Models (VLMs) have prompted the development of multi-modal medical assistant systems. Despite this progress, current models still have inherent probabilistic uncertainties, often producing erroneous or unverified responses-an issue with serious implications in medical applications. Existing methods aim to enhance the performance of Medical Vision Language Model (MedVLM) by adjusting model structure, fine-tuning with high-quality data, or through preference fine-tuning. However, these training-dependent strategies are costly and still lack sufficient alignment with clinical expertise. To address these issues, we propose an expert-in-the-loop framework named Expert-Controlled Classifier-Free Guidance (Expert-CFG) to align MedVLM with clinical expertise without additional training. This framework introduces an uncertainty estimation strategy to identify unreliable outputs. It then retrieves relevant references to assist experts in highlighting key terms and applies classifier-free guidance to refine the token embeddings of MedVLM, ensuring that the adjusted outputs are correct and align with expert highlights. Evaluations across three medical visual question answering benchmarks demonstrate that the proposed Expert-CFG, with 4.2B parameters and limited expert annotations, outperforms state-of-the-art models with 13B parameters. The results demonstrate the feasibility of deploying such a system in resource-limited settings for clinical use.

Paperid: 3304, https://arxiv.org/pdf/2507.06959.pdf

Abstract:
Vision-language models (VLMs) are prone to hallucinations that critically compromise reliability in medical applications. While preference optimization can mitigate these hallucinations through clinical feedback, its implementation faces challenges such as clinically irrelevant training samples, imbalanced data distributions, and prohibitive expert annotation costs. To address these challenges, we introduce CheXPO, a Chest X-ray Preference Optimization strategy that combines confidence-similarity joint mining with counterfactual rationale. Our approach begins by synthesizing a unified, fine-grained multi-task chest X-ray visual instruction dataset across different question types for supervised fine-tuning (SFT). We then identify hard examples through token-level confidence analysis of SFT failures and use similarity-based retrieval to expand hard examples for balancing preference sample distributions, while synthetic counterfactual rationales provide fine-grained clinical preferences, eliminating the need for additional expert input. Experiments show that CheXPO achieves 8.93% relative performance gain using only 5% of SFT samples, reaching state-of-the-art performance across diverse clinical tasks and providing a scalable, interpretable solution for real-world radiology applications.

Paperid: 3305, https://arxiv.org/pdf/2507.01042.pdf

Abstract:
Vision-Language Models (VLMs) are advancing multimodal AI, yet their performance consistency across tasks is underexamined. We benchmark CLIP, BLIP, and LXMERT across diverse datasets spanning retrieval, captioning, and reasoning. Our evaluation includes task accuracy, generation quality, efficiency, and a novel Cross-Dataset Consistency (CDC) metric. CLIP shows strongest generalization (CDC: 0.92), BLIP excels on curated data, and LXMERT leads in structured reasoning. These results expose trade-offs between generalization and specialization, informing industrial deployment of VLMs and guiding development toward robust, task-flexible architectures.

Paperid: 3306, https://arxiv.org/pdf/2507.00236.pdf

Abstract:
Simulation-based design, optimization, and validation of autonomous vehicles have proven to be crucial for their improvement over the years. Nevertheless, the ultimate measure of effectiveness is their successful transition from simulation to reality (sim2real). However, existing sim2real transfer methods struggle to address the autonomy-oriented requirements of balancing: (i) conditioned domain adaptation, (ii) robust performance with limited examples, (iii) modularity in handling multiple domain representations, and (iv) real-time performance. To alleviate these pain points, we present a unified framework for learning cross-domain adaptive representations through conditional latent diffusion for sim2real transferable automated driving. Our framework offers options to leverage: (i) alternate foundation models, (ii) a few-shot fine-tuning pipeline, and (iii) textual as well as image prompts for mapping across given source and target domains. It is also capable of generating diverse high-quality samples when diffusing across parameter spaces such as times of day, weather conditions, seasons, and operational design domains. We systematically analyze the presented framework and report our findings in terms of performance benchmarks and ablation studies. Additionally, we demonstrate its serviceability for autonomous driving using behavioral cloning case studies. Our experiments indicate that the proposed framework is capable of bridging the perceptual sim2real gap by over 40%.

Paperid: 3307, https://arxiv.org/pdf/2506.13757.pdf

Abstract:
Recent advancements in Vision-Language-Action (VLA) models have shown promise for end-to-end autonomous driving by leveraging world knowledge and reasoning capabilities. However, current VLA models often struggle with physically infeasible action outputs, complex model structures, or unnecessarily long reasoning. In this paper, we propose AutoVLA, a novel VLA model that unifies reasoning and action generation within a single autoregressive generation model for end-to-end autonomous driving. AutoVLA performs semantic reasoning and trajectory planning directly from raw visual inputs and language instructions. We tokenize continuous trajectories into discrete, feasible actions, enabling direct integration into the language model. For training, we employ supervised fine-tuning to equip the model with dual thinking modes: fast thinking (trajectory-only) and slow thinking (enhanced with chain-of-thought reasoning). To further enhance planning performance and efficiency, we introduce a reinforcement fine-tuning method based on Group Relative Policy Optimization (GRPO), reducing unnecessary reasoning in straightforward scenarios. Extensive experiments across real-world and simulated datasets and benchmarks, including nuPlan, nuScenes, Waymo, and CARLA, demonstrate the competitive performance of AutoVLA in both open-loop and closed-loop settings. Qualitative results showcase the adaptive reasoning and accurate planning capabilities of AutoVLA in diverse scenarios.

Paperid: 3308, https://arxiv.org/pdf/2506.01174.pdf

Abstract:
Structured scene representations are a core component of embodied agents, helping to consolidate raw sensory streams into readable, modular, and searchable formats. Due to their high computational overhead, many approaches build such representations in advance of the task. However, when the task specifications change, such static approaches become inadequate as they may miss key objects, spatial relations, and details. We introduce GraphPad, a modifiable structured memory that an agent can tailor to the needs of the task through API calls. It comprises a mutable scene graph representing the environment, a navigation log indexing frame-by-frame content, and a scratchpad for task-specific notes. Together, GraphPad serves as a dynamic workspace that remains complete, current, and aligned with the agent's immediate understanding of the scene and its task. On the OpenEQA benchmark, GraphPad attains 55.3%, a +3.0% increase over an image-only baseline using the same vision-language model, while operating with five times fewer input frames. These results show that allowing online, language-driven refinement of 3-D memory yields more informative representations without extra training or data collection.

Paperid: 3309, https://arxiv.org/pdf/2506.00070.pdf

Abstract:
Large Vision-Language Models (LVLMs) have recently shown great promise in advancing robotics by combining embodied reasoning with robot control. A common approach involves training on embodied reasoning tasks related to robot control using Supervised Fine-Tuning (SFT). However, SFT datasets are often heuristically constructed and not explicitly optimized for improving robot control. Furthermore, SFT often leads to issues such as catastrophic forgetting and reduced generalization performance. To address these limitations, we introduce Robot-R1, a novel framework that leverages reinforcement learning to enhance embodied reasoning specifically for robot control. Robot-R1 learns to predict the next keypoint state required for task completion, conditioned on the current scene image and environment metadata derived from expert demonstrations. Inspired by the DeepSeek-R1 learning approach, Robot-R1 samples reasoning-based responses and reinforces those that lead to more accurate predictions. Our experiments show that models trained with Robot-R1 outperform SFT methods on embodied reasoning tasks. Despite having only 7B parameters, Robot-R1 even surpasses GPT-4o on reasoning tasks related to low-level action control, such as spatial and primitive movement reasoning.

Paperid: 3310, https://arxiv.org/pdf/2505.23705.pdf

Abstract:
Vision-language-action (VLA) models provide a powerful approach to training control policies for physical systems, such as robots, by combining end-to-end learning with transfer of semantic knowledge from web-scale vision-language model (VLM) training. However, the constraints of real-time control are often at odds with the design of VLMs: the most powerful VLMs have tens or hundreds of billions of parameters, presenting an obstacle to real-time inference, and operate on discrete tokens rather than the continuous-valued outputs that are required for controlling robots. To address this challenge, recent VLA models have used specialized modules for efficient continuous control, such as action experts or continuous output heads, which typically require adding new untrained parameters to the pretrained VLM backbone. While these modules improve real-time and control capabilities, it remains an open question whether they preserve or degrade the semantic knowledge contained in the pretrained VLM, and what effect they have on the VLA training dynamics. In this paper, we study this question in the context of VLAs that include a continuous diffusion or flow matching action expert, showing that naively including such experts significantly harms both training speed and knowledge transfer. We provide an extensive analysis of various design choices, their impact on performance and knowledge transfer, and propose a technique for insulating the VLM backbone during VLA training that mitigates this issue. Videos are available at https://pi.website/research/knowledge_insulation.

Paperid: 3311, https://arxiv.org/pdf/2505.20272.pdf

Abstract:
Large Vision-Language Models (LVLMs) have demonstrated impressive general capabilities across a wide range of multi-modal tasks. However, the reasoning processes of LVLMs often suffer from unreliable outputs and limited interpretability. To address this, grounded visual reasoning has emerged as a promising paradigm that enforces responses anchored on salient visual evidence regions. However, existing approaches typically rely on costly supervision such as bounding box annotations, chain-of-thought rationale or external tool calls, limiting their scalability. In this work, we propose Ground-R1, a reinforcement learning framework that enables grounded visual reasoning without requiring explicit evidence or rationale annotations. Ground-R1 consists of a grounding phase that generates evidence region rollouts based on format constraints, and an answering phase that produces responses guided by both answer correctness and format adherence rewards. Extensive experiments across multiple visual reasoning benchmarks manifest that Ground-R1 achieves superior performance and exhibits emergent cognitive behaviors such as uncertainty awareness, spatial perception, and iterative refinement, offering a scalable and interpretable alternative to existing approaches.

Paperid: 3312, https://arxiv.org/pdf/2505.15489.pdf

Abstract:
The real-world impact of misinformation stems from the underlying misleading narratives that creators seek to convey. As such, interpreting misleading creator intent is essential for multimodal misinformation detection (MMD) systems aimed at effective information governance. In this paper, we introduce an automated framework that simulates real-world multimodal news creation by explicitly modeling creator intent through two components: the desired influence and the execution plan. Using this framework, we construct DeceptionDecoded, a large-scale benchmark comprising 12,000 image-caption pairs aligned with trustworthy reference articles. The dataset captures both misleading and non-misleading intents and spans manipulations across visual and textual modalities. We conduct a comprehensive evaluation of 14 state-of-the-art vision-language models (VLMs) on three intent-centric tasks: (1) misleading intent detection, (2) misleading source attribution, and (3) creator desire inference. Despite recent advances, we observe that current VLMs fall short in recognizing misleading intent, often relying on spurious cues such as superficial cross-modal consistency, stylistic signals, and heuristic authenticity hints. Our findings highlight the pressing need for intent-aware modeling in MMD and open new directions for developing systems capable of deeper reasoning about multimodal misinformation.

Paperid: 3313, https://arxiv.org/pdf/2505.12890.pdf

Abstract:
Surgical procedures unfold in complex environments demanding coordination between surgical teams, tools, imaging and increasingly, intelligent robotic systems. Ensuring safety and efficiency in ORs of the future requires intelligent systems, like surgical robots, smart instruments and digital copilots, capable of understanding complex activities and hazards of surgeries. Yet, existing computational approaches, lack the breadth, and generalization needed for comprehensive OR understanding. We introduce ORQA, a multimodal foundation model unifying visual, auditory, and structured data for holistic surgical understanding. ORQA's question-answering framework empowers diverse tasks, serving as an intelligence core for a broad spectrum of surgical technologies. We benchmark ORQA against generalist vision-language models, including ChatGPT and Gemini, and show that while they struggle to perceive surgical scenes, ORQA delivers substantially stronger, consistent performance. Recognizing the extensive range of deployment settings across clinical practice, we design, and release a family of smaller ORQA models tailored to different computational requirements. This work establishes a foundation for the next wave of intelligent surgical solutions, enabling surgical teams and medical technology providers to create smarter and safer operating rooms.

Paperid: 3314, https://arxiv.org/pdf/2505.08468.pdf

Abstract:
Charts are ubiquitous as they help people understand and reason with data. Recently, various downstream tasks, such as chart question answering, chart2text, and fact-checking, have emerged. Large Vision-Language Models (LVLMs) show promise in tackling these tasks, but their evaluation is costly and time-consuming, limiting real-world deployment. While using LVLMs as judges to assess the chart comprehension capabilities of other LVLMs could streamline evaluation processes, challenges like proprietary datasets, restricted access to powerful models, and evaluation costs hinder their adoption in industrial settings. To this end, we present a comprehensive evaluation of 13 open-source LVLMs as judges for diverse chart comprehension and reasoning tasks. We design both pairwise and pointwise evaluation tasks covering criteria like factual correctness, informativeness, and relevancy. Additionally, we analyze LVLM judges based on format adherence, positional consistency, length bias, and instruction-following. We focus on cost-effective LVLMs (<10B parameters) suitable for both research and commercial use, following a standardized evaluation protocol and rubric to measure the LVLM judge's accuracy. Experimental results reveal notable variability: while some open LVLM judges achieve GPT-4-level evaluation performance (about 80% agreement with GPT-4 judgments), others struggle (below ~10% agreement). Our findings highlight that state-of-the-art open-source LVLMs can serve as cost-effective automatic evaluators for chart-related tasks, though biases such as positional preference and length bias persist.

Paperid: 3315, https://arxiv.org/pdf/2505.08243.pdf

Abstract:
Robot chain-of-thought reasoning (CoT) -- wherein a model predicts helpful intermediate representations before choosing actions -- provides an effective method for improving the generalization and performance of robot policies, especially vision-language-action models (VLAs). While such approaches have been shown to improve performance and generalization, they suffer from core limitations, like needing specialized robot reasoning data and slow inference speeds. To design new robot reasoning approaches that address these issues, a more complete characterization of why reasoning helps policy performance is critical. We hypothesize several mechanisms by which robot reasoning improves policies -- (1) better representation learning, (2) improved learning curricularization, and (3) increased expressivity -- then devise simple variants of robot CoT reasoning to isolate and test each one. We find that learning to generate reasonings does lead to better VLA representations, while attending to the reasonings aids in actually leveraging these features for improved action prediction. Our results provide us with a better understanding of why CoT reasoning helps VLAs, which we use to introduce two simple and lightweight alternative recipes for robot reasoning. Our proposed approaches achieve significant performance gains over non-reasoning policies, state-of-the-art results on the LIBERO-90 benchmark, and a 3x inference speedup compared to standard robot reasoning.

Paperid: 3316, https://arxiv.org/pdf/2504.21186.pdf

Abstract:
Out-of-distribution (OOD) detection is critical for ensuring the safety and reliability of machine learning systems, particularly in dynamic and open-world environments. In the vision and text domains, zero-shot OOD detection - which requires no training on in-distribution (ID) data - has advanced significantly through the use of large-scale pretrained models, such as vision-language models (VLMs) and large language models (LLMs). However, zero-shot OOD detection in graph-structured data remains largely unexplored, primarily due to the challenges posed by complex relational structures and the absence of powerful, large-scale pretrained models for graphs. In this work, we take the first step toward enabling zero-shot graph OOD detection by leveraging a graph foundation model (GFM). Our experiments show that, when provided only with class label names for both ID and OOD categories, the GFM can effectively perform OOD detection - often surpassing existing "supervised" OOD detection methods that rely on extensive labeled node data. We further address the practical scenario in which OOD label names are not available in real-world settings by introducing GLIP-OOD, a framework that uses LLMs to generate semantically informative pseudo-OOD labels from unlabeled data. These generated OOD labels allow the GFM to better separate ID and OOD classes, facilitating more precise OOD detection - all without any labeled nodes (only ID label names). To our knowledge, this is the first approach to achieve node-level graph OOD detection in a fully zero-shot setting, and it attains performance comparable to state-of-the-art supervised methods on four benchmark text-attributed graph datasets.

Paperid: 3317, https://arxiv.org/pdf/2504.08603.pdf

Abstract:
Geometrically accurate and semantically expressive map representations have proven invaluable to facilitate robust and safe mobile robot navigation and task planning. Nevertheless, real-time, open-vocabulary semantic understanding of large-scale unknown environments is still an open problem. In this paper we present FindAnything, an open-world mapping and exploration framework that incorporates vision-language information into dense volumetric submaps. Thanks to the use of vision-language features, FindAnything bridges the gap between pure geometric and open-vocabulary semantic information for a higher level of understanding while allowing to explore any environment without the help of any external source of ground-truth pose information. We represent the environment as a series of volumetric occupancy submaps, resulting in a robust and accurate map representation that deforms upon pose updates when the underlying SLAM system corrects its drift, allowing for a locally consistent representation between submaps. Pixel-wise vision-language features are aggregated from efficient SAM (eSAM)-generated segments, which are in turn integrated into object-centric volumetric submaps, providing a mapping from open-vocabulary queries to 3D geometry that is scalable also in terms of memory usage. The open-vocabulary map representation of FindAnything achieves state-of-the-art semantic accuracy in closed-set evaluations on the Replica dataset. This level of scene understanding allows a robot to explore environments based on objects or areas of interest selected via natural language queries. Our system is the first of its kind to be deployed on resource-constrained devices, such as MAVs, leveraging vision-language information for real-world robotic tasks.

Paperid: 3318, https://arxiv.org/pdf/2504.07867.pdf

Abstract:
Video Scene Graph Generation (VidSGG) is an important topic in understanding dynamic kitchen environments. Current models for VidSGG require extensive training to produce scene graphs. Recently, Vision Language Models (VLM) and Vision Foundation Models (VFM) have demonstrated impressive zero-shot capabilities in a variety of tasks. However, VLMs like Gemini struggle with the dynamics for VidSGG, failing to maintain stable object identities across frames. To overcome this limitation, we propose SAMJAM, a zero-shot pipeline that combines SAM2's temporal tracking with Gemini's semantic understanding. SAM2 also improves upon Gemini's object grounding by producing more accurate bounding boxes. In our method, we first prompt Gemini to generate a frame-level scene graph. Then, we employ a matching algorithm to map each object in the scene graph with a SAM2-generated or SAM2-propagated mask, producing a temporally-consistent scene graph in dynamic environments. Finally, we repeat this process again in each of the following frames. We empirically demonstrate that SAMJAM outperforms Gemini by 8.33% in mean recall on the EPIC-KITCHENS and EPIC-KITCHENS-100 datasets.

Paperid: 3319, https://arxiv.org/pdf/2504.05651.pdf

Abstract:
Recent research has shown that representation learning models may accidentally memorize their training data. For example, the dÃ©jÃ vu method shows that for certain representation learning models and training images, it is sometimes possible to correctly predict the foreground label given only the representation of the background - better than through dataset-level correlations. However, their measurement method requires training two models - one to estimate dataset-level correlations and the other to estimate memorization. This multiple model setup becomes infeasible for large open-source models. In this work, we propose alternative simple methods to estimate dataset-level correlations, and show that these can be used to approximate an off-the-shelf model's memorization ability without any retraining. This enables, for the first time, the measurement of memorization in pre-trained open-source image representation and vision-language representation models. Our results show that different ways of measuring memorization yield very similar aggregate results. We also find that open-source models typically have lower aggregate memorization than similar models trained on a subset of the data. The code is available both for vision and vision language models.

Paperid: 3320, https://arxiv.org/pdf/2503.13939.pdf

Abstract:
Vision-language models (VLMs) have achieved impressive progress in natural image reasoning, yet their potential in medical imaging remains underexplored. Medical vision-language tasks demand precise understanding and clinically coherent answers, which are difficult to achieve due to the complexity of medical data and the scarcity of high-quality expert annotations. These challenges limit the effectiveness of conventional supervised fine-tuning (SFT) and Chain-of-Thought (CoT) strategies that work well in general domains. To address these challenges, we propose Med-R1, a reinforcement learning (RL)-enhanced vision-language model designed to improve generalization and reliability in medical reasoning. Built on the DeepSeek strategy, Med-R1 adopts Group Relative Policy Optimization (GRPO) to encourage reward-guided learning beyond static annotations. We comprehensively evaluate Med-R1 across eight distinct medical imaging modalities. Med-R1 achieves a 29.94% improvement in average accuracy over its base model Qwen2-VL-2B, and even outperforms Qwen2-VL-72B-a model with 36x more parameters. To assess cross-task generalization, we further evaluate Med-R1 on five question types. Med-R1 outperforms Qwen2-VL-2B by 32.06% in question-type generalization, also surpassing Qwen2-VL-72B. We further explore the thinking process in Med-R1, a crucial component for the success of Deepseek-R1. Our results show that omitting intermediate rationales (No-Thinking-Med-R1) not only improves in-domain and cross-domain generalization with less training, but also challenges the assumption that more reasoning always helps. These findings suggest that in medical VQA, it is not reasoning itself, but its quality and domain alignment, that determine effectiveness. Together, these results highlight that RL improves medical reasoning and generalization, enabling efficient and reliable VLMs for real-world deployment.

Paperid: 3321, https://arxiv.org/pdf/2503.07216.pdf

Abstract:
Federated Learning (FL) is a widely used framework for training models in a decentralized manner, ensuring that the central server does not have direct access to data from local clients. However, this approach may still fail to fully preserve data privacy, as models from local clients are exposed to the central server during the aggregation process. This issue becomes even more critical when training vision-language models (VLMs) with FL, as VLMs can easily memorize training data instances, making them vulnerable to membership inference attacks (MIAs). To address this challenge, we propose the FedRand framework, which avoids disclosing the full set of client parameters. In this framework, each client randomly selects subparameters of Low-Rank Adaptation (LoRA) from the server and keeps the remaining counterparts of the LoRA weights as private parameters. After training both parameters on the client's private dataset, only the non-private client parameters are sent back to the server for aggregation. This approach mitigates the risk of exposing client-side VLM parameters, thereby enhancing data privacy. We empirically validate that FedRand improves robustness against MIAs compared to relevant baselines while achieving accuracy comparable to methods that communicate full LoRA parameters across several benchmark datasets.

Paperid: 3322, https://arxiv.org/pdf/2503.02104.pdf

Abstract:
Foundation models, first introduced in 2021, are large-scale pre-trained models (e.g., large language models (LLMs) and vision-language models (VLMs)) that learn from extensive unlabeled datasets through unsupervised methods, enabling them to excel in diverse downstream tasks. These models, like GPT, can be adapted to various applications such as question answering and visual understanding, outperforming task-specific AI models and earning their name due to broad applicability across fields. The development of biomedical foundation models marks a significant milestone in leveraging artificial intelligence (AI) to understand complex biological phenomena and advance medical research and practice. This survey explores the potential of foundation models across diverse domains within biomedical fields, including computational biology, drug discovery and development, clinical informatics, medical imaging, and public health. The purpose of this survey is to inspire ongoing research in the application of foundation models to health science.

Paperid: 3323, https://arxiv.org/pdf/2502.13923.pdf

Abstract:
We introduce Qwen2.5-VL, the latest flagship model of Qwen vision-language series, which demonstrates significant advancements in both foundational capabilities and innovative functionalities. Qwen2.5-VL achieves a major leap forward in understanding and interacting with the world through enhanced visual recognition, precise object localization, robust document parsing, and long-video comprehension. A standout feature of Qwen2.5-VL is its ability to localize objects using bounding boxes or points accurately. It provides robust structured data extraction from invoices, forms, and tables, as well as detailed analysis of charts, diagrams, and layouts. To handle complex inputs, Qwen2.5-VL introduces dynamic resolution processing and absolute time encoding, enabling it to process images of varying sizes and videos of extended durations (up to hours) with second-level event localization. This allows the model to natively perceive spatial scales and temporal dynamics without relying on traditional normalization techniques. By training a native dynamic-resolution Vision Transformer (ViT) from scratch and incorporating Window Attention, we reduce computational overhead while maintaining native resolution. As a result, Qwen2.5-VL excels not only in static image and document understanding but also as an interactive visual agent capable of reasoning, tool usage, and task execution in real-world scenarios such as operating computers and mobile devices. Qwen2.5-VL is available in three sizes, addressing diverse use cases from edge AI to high-performance computing. The flagship Qwen2.5-VL-72B model matches state-of-the-art models like GPT-4o and Claude 3.5 Sonnet, particularly excelling in document and diagram understanding. Additionally, Qwen2.5-VL maintains robust linguistic performance, preserving the core language competencies of the Qwen2.5 LLM.

Paperid: 3324, https://arxiv.org/pdf/2501.19164.pdf

Abstract:
Large vision-language models (LVMs) extend large language models (LLMs) with visual perception capabilities, enabling them to process and interpret visual information. A major challenge compromising their reliability is object hallucination that LVMs may generate plausible but factually inaccurate information. We propose a novel visual adversarial perturbation (VAP) method to mitigate this hallucination issue. VAP alleviates LVM hallucination by applying strategically optimized visual noise without altering the base model. Our approach formulates hallucination suppression as an optimization problem, leveraging adversarial strategies to generate beneficial visual perturbations that enhance the model's factual grounding and reduce parametric knowledge bias. Extensive experimental results demonstrate that our method consistently reduces object hallucinations across 8 state-of-the-art LVMs, validating its efficacy across diverse evaluations.

Paperid: 3325, https://arxiv.org/pdf/2412.14424.pdf

Abstract:
Large Vision-Language Models typically require large text and image datasets for effective fine-tuning. However, collecting data from various sites, especially in healthcare, is challenging due to strict privacy regulations. An alternative is to fine-tune these models on end-user devices, such as in medical clinics, without sending data to a server. These local clients typically have limited computing power and small datasets, which are not enough for fully fine-tuning large VLMs on their own. A naive solution to these scenarios is to leverage parameter-efficient fine-tuning (PEFT) strategies and apply federated learning (FL) algorithms to combine the learned adapter weights, thereby respecting the resource limitations and data privacy. However, this approach does not fully leverage the knowledge from multiple adapters trained on diverse data distributions and for diverse tasks. The adapters are adversely impacted by data heterogeneity and task heterogeneity across clients resulting in suboptimal convergence. To this end, we propose a novel framework called FedPIA that improves upon the naive combinations of FL and PEFT by introducing Permutation and Integration of the local Adapters in the server and global Adapters in the clients exploiting Wasserstein barycenters for improved blending of client-specific and client-agnostic knowledge. This layerwise permutation helps to bridge the gap in the parameter space of local and global adapters before integration. We conduct over 2000 client-level experiments utilizing 48 medical image datasets across five different medical vision-language FL task settings encompassing visual question answering as well as image and report-based multi-label disease detection. Our experiments involving diverse client settings, ten different modalities, and two VLM backbones demonstrate that FedPIA consistently outperforms the state-of-the-art PEFT-FL baselines.

Paperid: 3326, https://arxiv.org/pdf/2411.11913.pdf

Abstract:
Personalized driving refers to an autonomous vehicle's ability to adapt its driving behavior or control strategies to match individual users' preferences and driving styles while maintaining safety and comfort standards. However, existing works either fail to capture every individual preference precisely or become computationally inefficient as the user base expands. Vision-Language Models (VLMs) offer promising solutions to this front through their natural language understanding and scene reasoning capabilities. In this work, we propose a lightweight yet effective on-board VLM framework that provides low-latency personalized driving performance while maintaining strong reasoning capabilities. Our solution incorporates a Retrieval-Augmented Generation (RAG)-based memory module that enables continuous learning of individual driving preferences through human feedback. Through comprehensive real-world vehicle deployment and experiments, our system has demonstrated the ability to provide safe, comfortable, and personalized driving experiences across various scenarios and significantly reduce takeover rates by up to 76.9%. To the best of our knowledge, this work represents the first end-to-end VLM-based motion control system in real-world autonomous vehicles.

Paperid: 3327, https://arxiv.org/pdf/2411.11912.pdf

Abstract:
Effective training of large Vision-Language Models (VLMs) on resource-constrained client devices in Federated Learning (FL) requires the usage of parameter-efficient fine-tuning (PEFT) strategies. To this end, we demonstrate the impact of two factors \textit{viz.}, client-specific layer importance score that selects the most important VLM layers for fine-tuning and inter-client layer diversity score that encourages diverse layer selection across clients for optimal VLM layer selection. We first theoretically motivate and leverage the principal eigenvalue magnitude of layerwise Neural Tangent Kernels and show its effectiveness as client-specific layer importance score. Next, we propose a novel layer updating strategy dubbed F$^3$OCUS that jointly optimizes the layer importance and diversity factors by employing a data-free, multi-objective, meta-heuristic optimization on the server. We explore 5 different meta-heuristic algorithms and compare their effectiveness for selecting model layers and adapter layers towards PEFT-FL. Furthermore, we release a new MedVQA-FL dataset involving overall 707,962 VQA triplets and 9 modality-specific clients and utilize it to train and evaluate our method. Overall, we conduct more than 10,000 client-level experiments on 6 Vision-Language FL task settings involving 58 medical image datasets and 4 different VLM architectures of varying sizes to demonstrate the effectiveness of the proposed method.

Paperid: 3328, https://arxiv.org/pdf/2411.10888.pdf

Abstract:
In the aftermath of the COVID-19 pandemic and amid accelerating climate change, emerging infectious diseases, particularly those arising from zoonotic spillover, remain a global threat. Mpox (caused by the monkeypox virus) is a notable example of a zoonotic infection that often goes undiagnosed, especially as its rash progresses through stages, complicating detection across diverse populations with different presentations. In August 2024, the WHO Director-General declared the mpox outbreak a public health emergency of international concern for a second time. Despite the deployment of deep learning techniques for detecting diseases from skin lesion images, a robust and publicly accessible foundation model for mpox diagnosis is still lacking due to the unavailability of open-source mpox skin lesion images, multimodal clinical data, and specialized training pipelines. To address this gap, we propose MpoxVLM, a vision-language model (VLM) designed to detect mpox by analyzing both skin lesion images and patient clinical information. MpoxVLM integrates the CLIP visual encoder, an enhanced Vision Transformer (ViT) classifier for skin lesions, and LLaMA-2-7B models, pre-trained and fine-tuned on visual instruction-following question-answer pairs from our newly released mpox skin lesion dataset. Our work achieves 90.38% accuracy for mpox detection, offering a promising pathway to improve early diagnostic accuracy in combating mpox.

Paperid: 3329, https://arxiv.org/pdf/2411.00988.pdf

Abstract:
Low-resource domains, characterized by scarce data and annotations, present significant challenges for language and visual understanding tasks, with the latter much under-explored in the literature. Recent advancements in Vision-Language Models (VLM) have shown promising results in high-resource domains but fall short in low-resource concepts that are under-represented (e.g. only a handful of images per category) in the pre-training set. We tackle the challenging task of zero-shot low-resource image classification from a novel perspective. By leveraging a retrieval-based strategy, we achieve this in a training-free fashion. Specifically, our method, named CoRE (Combination of Retrieval Enrichment), enriches the representation of both query images and class prototypes by retrieving relevant textual information from large web-crawled databases. This retrieval-based enrichment significantly boosts classification performance by incorporating the broader contextual information relevant to the specific class. We validate our method on a newly established benchmark covering diverse low-resource domains, including medical imaging, rare plants, and circuits. Our experiments demonstrate that CORE outperforms existing state-of-the-art methods that rely on synthetic data generation and model fine-tuning.

Paperid: 3330, https://arxiv.org/pdf/2410.21237.pdf

Abstract:
Knowledge Graph (KG) is playing an increasingly important role in various AI systems. For e-commerce, an efficient and low-cost automated knowledge graph construction method is the foundation of enabling various successful downstream applications. In this paper, we propose a novel method for constructing structured product knowledge graphs from raw product images. The method cooperatively leverages recent advances in the vision-language model (VLM) and large language model (LLM), fully automating the process and allowing timely graph updates. We also present a human-annotated e-commerce product dataset for benchmarking product property extraction in knowledge graph construction. Our method outperforms our baseline in all metrics and evaluated properties, demonstrating its effectiveness and bright usage potential.

Paperid: 3331, https://arxiv.org/pdf/2410.19546.pdf

Abstract:
Recently, newly developed Vision-Language Models (VLMs), such as OpenAI's o1, have emerged, seemingly demonstrating advanced reasoning capabilities across text and image modalities. However, the depth of these advances in language-guided perception and abstract reasoning remains underexplored, and it is unclear whether these models can truly live up to their ambitious promises. To assess the progress and identify shortcomings, we enter the wonderland of Bongard problems, a set of classic visual reasoning puzzles that require human-like abilities of pattern recognition and abstract reasoning. With our extensive evaluation setup, we show that while VLMs occasionally succeed in identifying discriminative concepts and solving some of the problems, they frequently falter. Surprisingly, even elementary concepts that may seem trivial to humans, such as simple spirals, pose significant challenges. Moreover, when explicitly asked to recognize ground truth concepts, they continue to falter, suggesting not only a lack of understanding of these elementary visual concepts but also an inability to generalize to unseen concepts. We compare the results of VLMs to human performance and observe that a significant gap remains between human visual reasoning capabilities and machine cognition.

Paperid: 3332, https://arxiv.org/pdf/2410.14154.pdf

Abstract:
Multimodal Large Language Models (MLLMs) have recently received substantial interest, which shows their emerging potential as general-purpose models for various vision-language tasks. MLLMs involve significant external knowledge within their parameters; however, it is challenging to continually update these models with the latest knowledge, which involves huge computational costs and poor interpretability. Retrieval augmentation techniques have proven to be effective plugins for both LLMs and MLLMs. In this study, we propose multimodal adaptive Retrieval-Augmented Bootstrapping Language-Image Pre-training (RA-BLIP), a novel retrieval-augmented framework for various MLLMs. Considering the redundant information within vision modality, we first leverage the question to instruct the extraction of visual information through interactions with one set of learnable queries, minimizing irrelevant interference during retrieval and generation. Besides, we introduce a pre-trained multimodal adaptive fusion module to achieve question text-to-multimodal retrieval and integration of multimodal knowledge by projecting visual and language modalities into a unified semantic space. Furthermore, we present an Adaptive Selection Knowledge Generation (ASKG) strategy to train the generator to autonomously discern the relevance of retrieved knowledge, which realizes excellent denoising performance. Extensive experiments on open multimodal question-answering datasets demonstrate that RA-BLIP achieves significant performance and surpasses the state-of-the-art retrieval-augmented models.

Paperid: 3333, https://arxiv.org/pdf/2410.00231.pdf

Abstract:
Learning-based methods have achieved strong performance for quadrupedal locomotion. However, several challenges prevent quadrupeds from learning helpful indoor skills that require interaction with environments and humans: lack of end-effectors for manipulation, limited semantic understanding using only simulation data, and low traversability and reachability in indoor environments. We present a system for quadrupedal mobile manipulation in indoor environments. It uses a front-mounted gripper for object manipulation, a low-level controller trained in simulation using egocentric depth for agile skills like climbing and whole-body tilting, and pre-trained vision-language models (VLMs) with a third-person fisheye and an egocentric RGB camera for semantic understanding and command generation. We evaluate our system in two unseen environments without any real-world data collection or training. Our system can zero-shot generalize to these environments and complete tasks, like following user's commands to fetch a randomly placed stuff toy after climbing over a queen-sized bed, with a 60% success rate. Project website: https://helpful-doggybot.github.io/

Paperid: 3334, https://arxiv.org/pdf/2409.16183.pdf

Abstract:
Radiology is a vital and complex component of modern clinical workflow and covers many tasks. Recently, vision-language (VL) foundation models in medicine have shown potential in processing multimodal information, offering a unified solution for various radiology tasks. However, existing studies either pre-trained VL models on natural data or did not fully integrate vision-language architecture and pretraining, often neglecting the unique multimodal complexity in radiology images and their textual contexts. Additionally, their practical applicability in real-world scenarios remains underexplored. Here, we present RadFound, a large and open-source vision-language foundation model tailored for radiology, that is trained on the most extensive dataset of over 8.1 million images and 250,000 image-text pairs, covering 19 major organ systems and 10 imaging modalities. To establish expert-level multimodal perception and generation capabilities, RadFound introduces an enhanced vision encoder to capture intra-image local features and inter-image contextual information, and a unified cross-modal learning design tailored to radiology. To fully assess the models' capability, we construct a benchmark, RadVLBench, including radiology interpretation tasks like medical vision-language question-answering, as well as text generation tasks ranging from captioning to report generation. We also propose a human evaluation framework. When evaluated on the real-world benchmark involving three representative modalities, 2D images (chest X-rays), multi-view images (mammograms), and 3D images (thyroid CT scans), RadFound significantly outperforms other VL foundation models on both quantitative metrics and human evaluation. In summary, the development of RadFound represents an advancement in radiology generalists, demonstrating broad applicability potential for integration into clinical workflows.

Paperid: 3335, https://arxiv.org/pdf/2409.05552.pdf

Abstract:
Autonomous navigation guided by natural language instructions in embodied environments remains a challenge for vision-language navigation (VLN) agents. Although recent advancements in learning diverse and fine-grained visual environmental representations have shown promise, the fragile performance improvements may not conclusively attribute to enhanced visual grounding,a limitation also observed in related vision-language tasks. In this work, we preliminarily investigate whether advanced VLN models genuinely comprehend the visual content of their environments by introducing varying levels of visual perturbations. These perturbations include ground-truth depth images, perturbed views and random noise. Surprisingly, we experimentally find that simple branch expansion, even with noisy visual inputs, paradoxically improves the navigational efficacy. Inspired by these insights, we further present a versatile Multi-Branch Architecture (MBA) designed to delve into the impact of both the branch quantity and visual quality. The proposed MBA extends a base agent into a multi-branch variant, where each branch processes a different visual input. This approach is embarrassingly simple yet agnostic to topology-based VLN agents. Extensive experiments on three VLN benchmarks (R2R, REVERIE, SOON) demonstrate that our method with optimal visual permutations matches or even surpasses state-of-the-art results. The source code is available at here.

Paperid: 3336, https://arxiv.org/pdf/2409.00341.pdf

Abstract:
Pre-trained large vision-language models (VLMs) like CLIP have revolutionized visual representation learning using natural language as supervisions, and demonstrated promising generalization ability. In this work, we propose ViP, a novel visual symptom-guided prompt learning framework for medical image analysis, which facilitates general knowledge transfer from CLIP. ViP consists of two key components: a visual symptom generator (VSG) and a dual-prompt network. Specifically, VSG aims to extract explicable visual symptoms from pre-trained large language models, while the dual-prompt network utilizes these visual symptoms to guide the training on two learnable prompt modules, i.e., context prompt and merge prompt, which effectively adapts our framework to medical image analysis via large VLMs. Extensive experimental results demonstrate that ViP can outperform state-of-the-art methods on two challenging datasets.

Paperid: 3337, https://arxiv.org/pdf/2601.18197.pdf

Abstract:
While Large Vision-Language Models (LVLMs) have significantly advanced GUI agents' capabilities in parsing textual instructions, interpreting screen content, and executing tasks, a critical challenge persists: the irreversibility of agent operations, where a single erroneous action can trigger catastrophic deviations. To address this, we propose the GUI Action Critic's Data Flywheel System (GAIA), a training framework that enables the models to have iterative critic capabilities, which are used to improve the Test-Time Scaling (TTS) of basic GUI agents' performance. Specifically, we train an Intuitive Critic Model (ICM) using positive and negative action examples from a base agent first. This critic evaluates the immediate correctness of the agent's intended actions, thereby selecting operations with higher success probability. Then, the initial critic guides agent actions to collect refined positive/negative samples, initiating the self-improving cycle. The augmented data then trains a second-round critic with enhanced discernment capability. We conduct experiments on various datasets and demonstrate that the proposed ICM can improve the test-time performance of various closed-source and open-source models, and the performance can be gradually improved as the data is recycled. The code and dataset will be publicly released.

Paperid: 3338, https://arxiv.org/pdf/2601.15949.pdf

Abstract:
Planetary surfaces are typically analyzed using high-level semantic concepts in natural language, yet vast orbital image archives remain organized at the pixel level. This mismatch limits scalable, open-ended exploration of planetary surfaces. Here we present MarScope, a planetary-scale vision-language framework enabling natural language-driven, label-free mapping of Martian landforms. MarScope aligns planetary images and text in a shared semantic space, trained on over 200,000 curated image-text pairs. This framework transforms global geomorphic mapping on Mars by replacing pre-defined classifications with flexible semantic retrieval, enabling arbitrary user queries across the entire planet in 5 seconds with F1 scores up to 0.978. Applications further show that it extends beyond morphological classification to facilitate process-oriented analysis and similarity-based geomorphological mapping at a planetary scale. MarScope establishes a new paradigm where natural language serves as a direct interface for scientific discovery over massive geospatial datasets.

Paperid: 3339, https://arxiv.org/pdf/2601.07107.pdf

Abstract:
Vision language models (VLMs) achieve strong performance on general image understanding but struggle to think with medical images, especially when performing multi-step reasoning through iterative visual interaction. Medical VLMs often rely on static visual embeddings and single-pass inference, preventing models from re-examining, verifying, or refining visual evidence during reasoning. While tool-integrated reasoning offers a promising path forward, open-source VLMs lack the training infrastructure to learn effective tool selection, invocation, and coordination in multi-modal medical reasoning. We introduce MedVistaGym, a scalable and interactive training environment that incentivizes tool-integrated visual reasoning for medical image analysis. MedVistaGym equips VLMs to determine when and which tools to invoke, localize task-relevant image regions, and integrate single or multiple sub-image evidence into interleaved multimodal reasoning within a unified, executable interface for agentic training. Using MedVistaGym, we train MedVistaGym-R1 to interleave tool use with agentic reasoning through trajectory sampling and end-to-end reinforcement learning. Across six medical VQA benchmarks, MedVistaGym-R1-8B exceeds comparably sized tool-augmented baselines by 19.10% to 24.21%, demonstrating that structured agentic training--not tool access alone--unlocks effective tool-integrated reasoning for medical image analysis.

Paperid: 3340, https://arxiv.org/pdf/2512.23292.pdf

Abstract:
The prevailing paradigm in AI for physical systems, scaling general-purpose foundation models toward universal multimodal reasoning, confronts a fundamental barrier at the control interface. Recent benchmarks show that even frontier vision-language models achieve only 50-53% accuracy on basic quantitative physics tasks, behaving as approximate guessers that preserve semantic plausibility while violating physical constraints. This input unfaithfulness is not a scaling deficiency but a structural limitation. Perception-centric architectures optimize parameter-space imitation, whereas safety-critical control demands outcome-space guarantees over executed actions. Here, we present a fundamentally different pathway toward domain-specific foundation models by introducing compact language models operating as Agentic Physical AI, in which policy optimization is driven by physics-based validation rather than perceptual inference. We train a 360-million-parameter model on synthetic reactor control scenarios, scaling the dataset from 10^3 to 10^5 examples. This induces a sharp phase transition absent in general-purpose models. Small-scale systems exhibit high-variance imitation with catastrophic tail risk, while large-scale models undergo variance collapse exceeding 500x reduction, stabilizing execution-level behavior. Despite balanced exposure to four actuation families, the model autonomously rejects approximately 70% of the training distribution and concentrates 95% of runtime execution on a single-bank strategy. Learned representations transfer across distinct physics and continuous input modalities without architectural modification.

Paperid: 3341, https://arxiv.org/pdf/2512.22615.pdf

Abstract:
While autoregressive Large Vision-Language Models (VLMs) have achieved remarkable success, their sequential generation often limits their efficacy in complex visual planning and dynamic robotic control. In this work, we investigate the potential of constructing Vision-Language Models upon diffusion-based large language models (dLLMs) to overcome these limitations. We introduce Dream-VL, an open diffusion-based VLM (dVLM) that achieves state-of-the-art performance among previous dVLMs. Dream-VL is comparable to top-tier AR-based VLMs trained on open data on various benchmarks but exhibits superior potential when applied to visual planning tasks. Building upon Dream-VL, we introduce Dream-VLA, a dLLM-based Vision-Language-Action model (dVLA) developed through continuous pre-training on open robotic datasets. We demonstrate that the natively bidirectional nature of this diffusion backbone serves as a superior foundation for VLA tasks, inherently suited for action chunking and parallel generation, leading to significantly faster convergence in downstream fine-tuning. Dream-VLA achieves top-tier performance of 97.2% average success rate on LIBERO, 71.4% overall average on SimplerEnv-Bridge, and 60.5% overall average on SimplerEnv-Fractal, surpassing leading models such as $π_0$ and GR00T-N1. We also validate that dVLMs surpass AR baselines on downstream tasks across different training objectives. We release both Dream-VL and Dream-VLA to facilitate further research in the community.

Paperid: 3342, https://arxiv.org/pdf/2512.06628.pdf

Abstract:
Embodied imitation learning is constrained by the scarcity of diverse, long-horizon robotic manipulation data. Existing video generation models for this domain are limited to synthesizing short clips of simple actions and often rely on manually defined trajectories. To this end, we introduce MIND-V, a hierarchical framework designed to synthesize physically plausible and logically coherent videos of long-horizon robotic manipulation. Inspired by cognitive science, MIND-V bridges high-level reasoning with pixel-level synthesis through three core components: a Semantic Reasoning Hub (SRH) that leverages a pre-trained vision-language model for task planning; a Behavioral Semantic Bridge (BSB) that translates abstract instructions into domain-invariant representations; and a Motor Video Generator (MVG) for conditional video rendering. MIND-V employs Staged Visual Future Rollouts, a test-time optimization strategy to enhance long-horizon robustness. To align the generated videos with physical laws, we introduce a GRPO reinforcement learning post-training phase guided by a novel Physical Foresight Coherence (PFC) reward. PFC leverages the V-JEPA world model to enforce physical plausibility by aligning the predicted and actual dynamic evolutions in the feature space. MIND-V demonstrates state-of-the-art performance in long-horizon robotic manipulation video generation, establishing a scalable and controllable paradigm for embodied data synthesis.

Paperid: 3343, https://arxiv.org/pdf/2512.05145.pdf

Abstract:
Effective judges of Vision-Language Models (VLMs) are crucial for model development. Current methods for training VLM judges mainly rely on large-scale human preference annotations. However, such an approach is costly, and the annotations easily become obsolete as models rapidly improve. In this work, we present a framework to self-train a VLM judge model without any human preference annotations, using only self-synthesized data. Our method is iterative and has three stages: (1) generate diverse multimodal instruction-response pairs at varying quality levels, (2) generate reasoning traces and judgments for each pair, removing the ones that do not match our expected quality levels, and (3) training on correct judge answers and their reasoning traces. We evaluate the resulting judge on Multimodal RewardBench and VL-RewardBench across domains: correctness, preference, reasoning, safety, and visual question-answering. Our method improves a Llama-3.2-11B multimodal judge from 0.38 to 0.51 in overall accuracy on VL-RewardBench, often outperforming much larger models including Llama-3.2-90B, GPT-4o, and Claude 3.5 Sonnet, with particularly strong gains in general, hallucination, and reasoning dimensions. The overall strength of these human-annotation-free results suggest the potential for a future self-judge that evolves alongside rapidly improving VLM capabilities.

Paperid: 3344, https://arxiv.org/pdf/2511.22688.pdf

Abstract:
A common recipe to improve diffusion models at test-time so that samples score highly against a user-specified reward is to introduce the gradient of the reward into the dynamics of the diffusion itself. This procedure is often ill posed, as user-specified rewards are usually only well defined on the data distribution at the end of generation. While common workarounds to this problem are to use a denoiser to estimate what a sample would have been at the end of generation, we propose a simple solution to this problem by working directly with a flow map. By exploiting a relationship between the flow map and velocity field governing the instantaneous transport, we construct an algorithm, Flow Map Trajectory Tilting (FMTT), which provably performs better ascent on the reward than standard test-time methods involving the gradient of the reward. The approach can be used to either perform exact sampling via importance weighting or principled search that identifies local maximizers of the reward-tilted distribution. We demonstrate the efficacy of our approach against other look-ahead techniques, and show how the flow map enables engagement with complicated reward functions that make possible new forms of image editing, e.g. by interfacing with vision language models.

Paperid: 3345, https://arxiv.org/pdf/2511.17910.pdf

Abstract:
Recently, Chain-of-Thought (CoT) reasoning has significantly enhanced the capabilities of large language models (LLMs), but Vision-Language Models (VLMs) still struggle with multi-step reasoning tasks due to limited multimodal reasoning data. To bridge this gap, researchers have explored methods to transfer CoT reasoning from LLMs to VLMs. However, existing approaches either need high training costs or require architectural alignment. In this paper, we use Linear Artificial Tomography (LAT) to empirically show that LLMs and VLMs share similar low-frequency latent representations of CoT reasoning despite architectural differences. Based on this insight, we propose L2V-CoT, a novel training-free latent intervention approach that transfers CoT reasoning from LLMs to VLMs. L2V-CoT extracts and resamples low-frequency CoT representations from LLMs in the frequency domain, enabling dimension matching and latent injection into VLMs during inference to enhance reasoning capabilities. Extensive experiments demonstrate that our approach consistently outperforms training-free baselines and even surpasses supervised methods.

Paperid: 3346, https://arxiv.org/pdf/2511.05565.pdf

Abstract:
Foundation vision-language models (VLMs) excel on natural images, but their utility for biomedical microscopy remains underexplored. In this paper, we investigate how in-context learning enables state-of-the-art VLMs to perform few-shot object detection when large annotated datasets are unavailable, as is often the case with microscopic images. We introduce the Micro-OD benchmark, a curated collection of 252 images specifically curated for in-context learning, with bounding-box annotations spanning 11 cell types across four sources, including two in-lab expert-annotated sets. We systematically evaluate eight VLMs under few-shot conditions and compare variants with and without implicit test-time reasoning tokens. We further implement a hybrid Few-Shot Object Detection (FSOD) pipeline that combines a detection head with a VLM-based few-shot classifier, which enhances the few-shot performance of recent VLMs on our benchmark. Across datasets, we observe that zero-shot performance is weak due to the domain gap; however, few-shot support consistently improves detection, with marginal gains achieved after six shots. We observe that models with reasoning tokens are more effective for end-to-end localization, whereas simpler variants are more suitable for classifying pre-localized crops. Our results highlight in-context adaptation as a practical path for microscopy, and our benchmark provides a reproducible testbed for advancing open-vocabulary detection in biomedical imaging.

Paperid: 3347, https://arxiv.org/pdf/2510.18188.pdf

Abstract:
Most current medical vision language models struggle to jointly generate diagnostic text and pixel-level segmentation masks in response to complex visual questions. This represents a major limitation towards clinical application, as assistive systems that fail to provide both modalities simultaneously offer limited value to medical practitioners. To alleviate this limitation, we first introduce RadDiagSeg-D, a dataset combining abnormality detection, diagnosis, and multi-target segmentation into a unified and hierarchical task. RadDiagSeg-D covers multiple imaging modalities and is precisely designed to support the development of models that produce descriptive text and corresponding segmentation masks in tandem. Subsequently, we leverage the dataset to propose a novel vision-language model, RadDiagSeg-M, capable of joint abnormality detection, diagnosis, and flexible segmentation. RadDiagSeg-M provides highly informative and clinically useful outputs, effectively addressing the need to enrich contextual information for assistive diagnosis. Finally, we benchmark RadDiagSeg-M and showcase its strong performance across all components involved in the task of multi-target text-and-mask generation, establishing a robust and competitive baseline.

Paperid: 3348, https://arxiv.org/pdf/2510.10854.pdf

Abstract:
Diffusion models have demonstrated remarkable performance in generating high-dimensional samples across domains such as vision, language, and the sciences. Although continuous-state diffusion models have been extensively studied both empirically and theoretically, discrete-state diffusion models, essential for applications involving text, sequences, and combinatorial structures, remain significantly less understood from a theoretical standpoint. In particular, all existing analyses of discrete-state models assume score estimation error bounds without studying sample complexity results. In this work, we present a principled theoretical framework for discrete-state diffusion, providing the first sample complexity bound of $\widetilde{\mathcal{O}}(ε^{-2})$. Our structured decomposition of the score estimation error into statistical, approximation, optimization, and clipping components offers critical insights into how discrete-state models can be trained efficiently. This analysis addresses a fundamental gap in the literature and establishes the theoretical tractability and practical relevance of discrete-state diffusion models.

Paperid: 3349, https://arxiv.org/pdf/2510.00411.pdf

Abstract:
The accurate interpretation of chest radiographs using automated methods is a critical task in medical imaging. This paper presents a comparative analysis between a supervised lightweight Convolutional Neural Network (CNN) and a state-of-the-art, zero-shot medical Vision-Language Model (VLM), BiomedCLIP, across two distinct diagnostic tasks: pneumonia detection on the PneumoniaMNIST benchmark and tuberculosis detection on the Shenzhen TB dataset. Our experiments show that supervised CNNs serve as highly competitive baselines in both cases. While the default zero-shot performance of the VLM is lower, we demonstrate that its potential can be unlocked via a simple yet crucial remedy: decision threshold calibration. By optimizing the classification threshold on a validation set, the performance of BiomedCLIP is significantly boosted across both datasets. For pneumonia detection, calibration enables the zero-shot VLM to achieve a superior F1-score of 0.8841, surpassing the supervised CNN's 0.8803. For tuberculosis detection, calibration dramatically improves the F1-score from 0.4812 to 0.7684, bringing it close to the supervised baseline's 0.7834. This work highlights a key insight: proper calibration is essential for leveraging the full diagnostic power of zero-shot VLMs, enabling them to match or even outperform efficient, task-specific supervised models.

Paperid: 3350, https://arxiv.org/pdf/2509.26557.pdf

Abstract:
Many users struggle to notice when a more efficient workflow exists in feature-rich tools like Excel. Existing AI assistants offer help only after users describe their goals or problems, which can be effortful and imprecise. We present InvisibleMentor, a system that turns screen recordings of task completion into vision-grounded reflections on tasks. It detects issues such as repetitive edits and recommends more efficient alternatives based on observed behavior. Unlike prior systems that rely on logs, APIs, or user prompts, InvisibleMentor operates directly on screen recordings. It uses a two-stage pipeline: a vision-language model reconstructs actions and context, and a language model generates structured, high-fidelity suggestions. In evaluation, InvisibleMentor accurately identified inefficient workflows, and participants found its suggestions more actionable, tailored, and more helpful for learning and improvement compared to a prompt-based spreadsheet assistant.

Paperid: 3351, https://arxiv.org/pdf/2509.18953.pdf

Abstract:
Vision-Language-Action (VLA) models have emerged as promising solutions for robotic manipulation, yet their robustness to real-world physical variations remains critically underexplored. To bridge this gap, we propose Eva-VLA, the first unified framework that systematically evaluates the robustness of VLA models by transforming discrete physical variations into continuous optimization problems. However, comprehensively assessing VLA robustness presents two key challenges: (1) how to systematically characterize diverse physical variations encountered in real-world deployments while maintaining evaluation reproducibility, and (2) how to discover worst-case scenarios without prohibitive real-world data collection costs efficiently. To address the first challenge, we decompose real-world variations into three critical domains: object 3D transformations that affect spatial reasoning, illumination variations that challenge visual perception, and adversarial patches that disrupt scene understanding. For the second challenge, we introduce a continuous black-box optimization framework that transforms discrete physical variations into parameter optimization, enabling systematic exploration of worst-case scenarios. Extensive experiments on state-of-the-art OpenVLA models across multiple benchmarks reveal alarming vulnerabilities: all variation types trigger failure rates exceeding 60%, with object transformations causing up to 97.8% failure in long-horizon tasks. Our findings expose critical gaps between controlled laboratory success and unpredictable deployment readiness, while the Eva-VLA framework provides a practical pathway for hardening VLA-based robotic manipulation models against real-world deployment challenges.

Paperid: 3352, https://arxiv.org/pdf/2509.12724.pdf

Abstract:
Despite their superb capabilities, Vision-Language Models (VLMs) have been shown to be vulnerable to jailbreak attacks. While recent jailbreaks have achieved notable progress, their effectiveness and efficiency can still be improved. In this work, we reveal an interesting phenomenon: incorporating weak defense into the attack pipeline can significantly enhance both the effectiveness and the efficiency of jailbreaks on VLMs. Building on this insight, we propose Defense2Attack, a novel jailbreak method that bypasses the safety guardrails of VLMs by leveraging defensive patterns to guide jailbreak prompt design. Specifically, Defense2Attack consists of three key components: (1) a visual optimizer that embeds universal adversarial perturbations with affirmative and encouraging semantics; (2) a textual optimizer that refines the input using a defense-styled prompt; and (3) a red-team suffix generator that enhances the jailbreak through reinforcement fine-tuning. We empirically evaluate our method on four VLMs and four safety benchmarks. The results demonstrate that Defense2Attack achieves superior jailbreak performance in a single attempt, outperforming state-of-the-art attack methods that often require multiple tries. Our work offers a new perspective on jailbreaking VLMs.

Paperid: 3353, https://arxiv.org/pdf/2509.02615.pdf

Abstract:
Vision-Language Models (VLMs), such as recent Qwen and Gemini models, are positioned as general-purpose AI systems capable of reasoning across domains. Yet their capabilities in scientific imaging, especially on unfamiliar and potentially previously unseen data distributions, remain poorly understood. In this work, we assess whether generic VLMs, presumed to lack exposure to astronomical corpora, can perform morphology-based classification of radio galaxies using the MiraBest FR-I/FR-II dataset. We explore prompting strategies using natural language and schematic diagrams, and, to the best of our knowledge, we are the first to introduce visual in-context examples within prompts in astronomy. Additionally, we evaluate lightweight supervised adaptation via LoRA fine-tuning. Our findings reveal three trends: (i) even prompt-based approaches can achieve good performance, suggesting that VLMs encode useful priors for unfamiliar scientific domains; (ii) however, outputs are highly unstable, i.e. varying sharply with superficial prompt changes such as layout, ordering, or decoding temperature, even when semantic content is held constant; and (iii) with just 15M trainable parameters and no astronomy-specific pretraining, fine-tuned Qwen-VL achieves near state-of-the-art performance (3% Error rate), rivaling domain-specific models. These results suggest that the apparent "reasoning" of VLMs often reflects prompt sensitivity rather than genuine inference, raising caution for their use in scientific domains. At the same time, with minimal adaptation, generic VLMs can rival specialized models, offering a promising but fragile tool for scientific discovery.

Paperid: 3354, https://arxiv.org/pdf/2508.19967.pdf

Abstract:
Geo-localization is the task of identifying the location of an image using visual cues alone. It has beneficial applications, such as improving disaster response, enhancing navigation, and geography education. Recently, Vision-Language Models (VLMs) are increasingly demonstrating capabilities as accurate image geo-locators. This brings significant privacy risks, including those related to stalking and surveillance, considering the widespread uses of AI models and sharing of photos on social media. The precision of these models is likely to improve in the future. Despite these risks, there is little work on systematically evaluating the geolocation precision of Generative VLMs, their limits and potential for unintended inferences. To bridge this gap, we conduct a comprehensive assessment of the geolocation capabilities of 25 state-of-the-art VLMs on four benchmark image datasets captured in diverse environments. Our results offer insight into the internal reasoning of VLMs and highlight their strengths, limitations, and potential societal risks. Our findings indicate that current VLMs perform poorly on generic street-level images yet achieve notably high accuracy (61\%) on images resembling social media content, raising significant and urgent privacy concerns.

Paperid: 3355, https://arxiv.org/pdf/2508.15201.pdf

Abstract:
Embodied intelligence systems, which enhance agent capabilities through continuous environment interactions, have garnered significant attention from both academia and industry. Vision-Language-Action models, inspired by advancements in large foundation models, serve as universal robotic control frameworks that substantially improve agent-environment interaction capabilities in embodied intelligence systems. This expansion has broadened application scenarios for embodied AI robots. This survey comprehensively reviews VLA models for embodied manipulation. Firstly, it chronicles the developmental trajectory of VLA architectures. Subsequently, we conduct a detailed analysis of current research across 5 critical dimensions: VLA model structures, training datasets, pre-training methods, post-training methods, and model evaluation. Finally, we synthesize key challenges in VLA development and real-world deployment, while outlining promising future research directions.

Paperid: 3356, https://arxiv.org/pdf/2508.11317.pdf

Abstract:
Vision-Language Models (VLMs), exemplified by CLIP, have emerged as foundational for multimodal intelligence. However, their capacity for logical understanding remains significantly underexplored, resulting in critical ''logical blindspots'' that limit their reliability in practical applications. To systematically diagnose this, we introduce LogicBench, a comprehensive benchmark with over 50,000 vision-language pairs across 9 logical categories and 4 diverse scenarios: images, videos, anomaly detection, and medical diagnostics. Our evaluation reveals that existing VLMs, even the state-of-the-art ones, fall at over 40 accuracy points below human performance, particularly in challenging tasks like Causality and Conditionality, highlighting their reliance on surface semantics over critical logical structures. To bridge this gap, we propose LogicCLIP, a novel training framework designed to boost VLMs' logical sensitivity through advancements in both data generation and optimization objectives. LogicCLIP utilizes logic-aware data generation and a contrastive learning strategy that combines coarse-grained alignment, a fine-grained multiple-choice objective, and a novel logical structure-aware objective. Extensive experiments demonstrate LogicCLIP's substantial improvements in logical comprehension across all LogicBench domains, significantly outperforming baselines. Moreover, LogicCLIP retains, and often surpasses, competitive performance on general vision-language benchmarks, demonstrating that the enhanced logical understanding does not come at the expense of general alignment. We believe that LogicBench and LogicCLIP will be important resources for advancing VLM logical capabilities.

Paperid: 3357, https://arxiv.org/pdf/2508.09716.pdf

Abstract:
Information visualizations are powerful tools that help users quickly identify patterns, trends, and outliers, facilitating informed decision-making. However, when visualizations incorporate deceptive design elements-such as truncated or inverted axes, unjustified 3D effects, or violations of best practices-they can mislead viewers and distort understanding, spreading misinformation. While some deceptive tactics are obvious, others subtly manipulate perception while maintaining a facade of legitimacy. As Vision-Language Models (VLMs) are increasingly used to interpret visualizations, especially by non-expert users, it is critical to understand how susceptible these models are to deceptive visual designs. In this study, we conduct an in-depth evaluation of VLMs' ability to interpret misleading visualizations. By analyzing over 16,000 responses from ten different models across eight distinct types of misleading chart designs, we demonstrate that most VLMs are deceived by them. This leads to altered interpretations of charts, despite the underlying data remaining the same. Our findings highlight the need for robust safeguards in VLMs against visual misinformation.

Paperid: 3358, https://arxiv.org/pdf/2508.08177.pdf

Abstract:
Accurately grounding regions of interest (ROIs) is critical for diagnosis and treatment planning in medical imaging. While multimodal large language models (MLLMs) combine visual perception with natural language, current medical-grounding pipelines still rely on supervised fine-tuning with explicit spatial hints, making them ill-equipped to handle the implicit queries common in clinical practice. This work makes three core contributions. We first define Unified Medical Reasoning Grounding (UMRG), a novel vision-language task that demands clinical reasoning and pixel-level grounding. Second, we release U-MRG-14K, a dataset of 14K samples featuring pixel-level masks alongside implicit clinical queries and reasoning traces, spanning 10 modalities, 15 super-categories, and 108 specific categories. Finally, we introduce MedReasoner, a modular framework that distinctly separates reasoning from segmentation: an MLLM reasoner is optimized with reinforcement learning, while a frozen segmentation expert converts spatial prompts into masks, with alignment achieved through format and accuracy rewards. MedReasoner achieves state-of-the-art performance on U-MRG-14K and demonstrates strong generalization to unseen clinical queries, underscoring the significant promise of reinforcement learning for interpretable medical grounding.

Paperid: 3359, https://arxiv.org/pdf/2508.05996.pdf

Abstract:
Complex medical decision-making involves cooperative workflows operated by different clinicians. Designing AI multi-agent systems can expedite and augment human-level clinical decision-making. Existing multi-agent researches primarily focus on language-only tasks, yet their extension to multimodal scenarios remains challenging. A blind combination of diverse vision-language models (VLMs) can amplify an erroneous outcome interpretation. VLMs in general are less capable in instruction following and importantly self-reflection, compared to large language models (LLMs) of comparable sizes. This disparity largely constrains VLMs' ability in cooperative workflows. In this study, we propose MedOrch, a mediator-guided multi-agent collaboration framework for medical multimodal decision-making. MedOrch employs an LLM-based mediator agent that enables multiple VLM-based expert agents to exchange and reflect on their outputs towards collaboration. We utilize multiple open-source general-purpose and domain-specific VLMs instead of costly GPT-series models, revealing the strength of heterogeneous models. We show that the collaboration within distinct VLM-based agents can surpass the capabilities of any individual agent. We validate our approach on five medical vision question answering benchmarks, demonstrating superior collaboration performance without model training. Our findings underscore the value of mediator-guided multi-agent collaboration in advancing medical multimodal intelligence. Our code will be made publicly available.

Paperid: 3360, https://arxiv.org/pdf/2507.17455.pdf

Abstract:
Geo-localization from a single image at planet scale (essentially an advanced or extreme version of the kidnapped robot problem) is a fundamental and challenging task in applications such as navigation, autonomous driving and disaster response due to the vast diversity of locations, environmental conditions, and scene variations. Traditional retrieval-based methods for geo-localization struggle with scalability and perceptual aliasing, while classification-based approaches lack generalization and require extensive training data. Recent advances in vision-language models (VLMs) offer a promising alternative by leveraging contextual understanding and reasoning. However, while VLMs achieve high accuracy, they are often prone to hallucinations and lack interpretability, making them unreliable as standalone solutions. In this work, we propose a novel hybrid geo-localization framework that combines the strengths of VLMs with retrieval-based visual place recognition (VPR) methods. Our approach first leverages a VLM to generate a prior, effectively guiding and constraining the retrieval search space. We then employ a retrieval step, followed by a re-ranking mechanism that selects the most geographically plausible matches based on feature similarity and proximity to the initially estimated coordinates. We evaluate our approach on multiple geo-localization benchmarks and show that it consistently outperforms prior state-of-the-art methods, particularly at street (up to 4.51%) and city level (up to 13.52%). Our results demonstrate that VLM-generated geographic priors in combination with VPR lead to scalable, robust, and accurate geo-localization systems.

Paperid: 3361, https://arxiv.org/pdf/2506.08543.pdf

Abstract:
High-level representations have become a central focus in enhancing AI transparency and control, shifting attention from individual neurons or circuits to structured semantic directions that align with human-interpretable concepts. Motivated by the Linear Representation Hypothesis (LRH), we propose the Input-Space Linearity Hypothesis (ISLH), which posits that concept-aligned directions originate in the input space and are selectively amplified with increasing depth. We then introduce the Spectral Principal Path (SPP) framework, which formalizes how deep networks progressively distill linear representations along a small set of dominant spectral directions. Building on this framework, we further demonstrate the multimodal robustness of these representations in Vision-Language Models (VLMs). By bridging theoretical insights with empirical validation, this work advances a structured theory of representation formation in deep networks, paving the way for improving AI robustness, fairness, and transparency.

Paperid: 3362, https://arxiv.org/pdf/2506.05453.pdf

Abstract:
Recent Multimodal Large Language Models (MLLMs) excel in vision-language understanding but face challenges in adapting to dynamic real-world scenarios that require continuous integration of new knowledge and skills. While continual learning (CL) offers a potential solution, existing benchmarks and methods suffer from critical limitations. In this paper, we introduce MLLM-CL, a novel benchmark encompassing domain and ability continual learning, where the former focuses on independently and identically distributed (IID) evaluation across evolving mainstream domains, whereas the latter evaluates on non-IID scenarios with emerging model ability. Methodologically, we propose preventing catastrophic interference through parameter isolation, along with an MLLM-based routing mechanism. Extensive experiments demonstrate that our approach can integrate domain-specific knowledge and functional abilities with minimal forgetting, significantly outperforming existing methods.

Paperid: 3363, https://arxiv.org/pdf/2506.05127.pdf

Abstract:
The digitization of histology slides has revolutionized pathology, providing massive datasets for cancer diagnosis and research. Contrastive self-supervised and vision-language models have been shown to effectively mine large pathology datasets to learn discriminative representations. On the other hand, generative models, capable of synthesizing realistic and diverse images, present a compelling solution to address unique problems in pathology that involve synthesizing images; overcoming annotated data scarcity, enabling privacy-preserving data sharing, and performing inherently generative tasks, such as virtual staining. We introduce PixCell, the first diffusion-based generative foundation model for histopathology. We train PixCell on PanCan-30M, a vast, diverse dataset derived from 69,184 H\&E-stained whole slide images covering various cancer types. We employ a progressive training strategy and a self-supervision-based conditioning that allows us to scale up training without any annotated data. PixCell generates diverse and high-quality images across multiple cancer types, which we find can be used in place of real data to train a self-supervised discriminative model. Synthetic images shared between institutions are subject to fewer regulatory barriers than would be the case with real clinical images. Furthermore, we showcase the ability to precisely control image generation using a small set of annotated images, which can be used for both data augmentation and educational purposes. Testing on a cell segmentation task, a mask-guided PixCell enables targeted data augmentation, improving downstream performance. Finally, we demonstrate PixCell's ability to use H\&E structural staining to infer results from molecular marker studies; we use this capability to infer IHC staining from H\&E images. Our trained models are publicly released to accelerate research in computational pathology.

Paperid: 3364, https://arxiv.org/pdf/2506.01300.pdf

Abstract:
Video understanding is fundamental to tasks such as action recognition, video reasoning, and robotic control. Early video understanding methods based on large vision-language models (LVLMs) typically adopt a single-pass reasoning paradigm without dynamic feedback, limiting the model's capacity to self-correct and adapt in complex scenarios. Recent efforts have attempted to address this limitation by incorporating reward models and reinforcement learning to enhance reasoning, or by employing tool-agent frameworks. However, these approaches face several challenges, including high annotation costs, reward signals that fail to capture real-time reasoning states, and low inference efficiency. To overcome these issues, we propose ReAgent-V, a novel agentic video understanding framework that integrates efficient frame selection with real-time reward generation during inference. These reward signals not only guide iterative answer refinement through a multi-perspective reflection mechanism-adjusting predictions from conservative, neutral, and aggressive viewpoints-but also enable automatic filtering of high-quality data for supervised fine-tuning (SFT), direct preference optimization (DPO), and group relative policy optimization (GRPO). ReAgent-V is lightweight, modular, and extensible, supporting flexible tool integration tailored to diverse tasks. Extensive experiments on 12 datasets across three core applications-video understanding, video reasoning enhancement, and vision-language-action model alignment-demonstrate significant gains in generalization and reasoning, with improvements of up to 6.9%, 2.1%, and 9.8%, respectively, highlighting the effectiveness and versatility of the proposed framework.

Paperid: 3365, https://arxiv.org/pdf/2505.21382.pdf

Abstract:
Low-Rank Adaptation (LoRA) has emerged as one of the most effective, computationally tractable fine-tuning approaches for training Vision-Language Models (VLMs) and Large Language Models (LLMs). LoRA accomplishes this by freezing the pre-trained model weights and injecting trainable low-rank matrices, allowing for efficient learning of these foundation models even on edge devices. However, LoRA in decentralized settings still remains under explored, particularly for the theoretical underpinnings due to the lack of smoothness guarantee and model consensus interference (defined formally below). This work improves the convergence rate of decentralized LoRA (DLoRA) to match the rate of decentralized SGD by ensuring gradient smoothness. We also introduce DeCAF, a novel algorithm integrating DLoRA with truncated singular value decomposition (TSVD)-based matrix factorization to resolve consensus interference. Theoretical analysis shows TSVD's approximation error is bounded and consensus differences between DLoRA and DeCAF vanish as rank increases, yielding DeCAF's matching convergence rate. Extensive experiments across vision/language tasks demonstrate our algorithms outperform local training and rivals federated learning under both IID and non-IID data distributions.

Paperid: 3366, https://arxiv.org/pdf/2505.20665.pdf

Abstract:
Autonomous driving requires real-time, robust reasoning across perception, prediction, planning, and behavior. However, conventional end-to-end models fail to generalize in complex scenarios due to the lack of structured reasoning. Recent vision-language models (VLMs) have been applied to driving tasks, but they typically rely on isolated modules and static supervision, limiting their ability to support multi-stage decision-making. We present AutoDriveRL, a unified training framework that formulates autonomous driving as a structured reasoning process over four core tasks. Each task is independently modeled as a vision-language question-answering problem and optimized using task-specific reward models, enabling fine-grained reinforcement signals at different reasoning stages. Within this framework, we train DriveRX, a cross-task reasoning VLM designed for real-time decision-making. DriveRX achieves strong performance on a public benchmark, outperforming GPT-4o in behavior reasoning and demonstrating robustness under complex or corrupted driving conditions. Our analysis further highlights the impact of vision encoder design and reward-guided reasoning compression. We will release the AutoDriveRL framework and the DriveRX model to support future research.

Paperid: 3367, https://arxiv.org/pdf/2505.19769.pdf

Abstract:
Developing scalable and generalizable reward engineering for reinforcement learning (RL) is crucial for creating general-purpose agents, especially in the challenging domain of robotic manipulation. While recent advances in reward engineering with Vision-Language Models (VLMs) have shown promise, their sparse reward nature significantly limits sample efficiency. This paper introduces TeViR, a novel method that leverages a pre-trained text-to-video diffusion model to generate dense rewards by comparing the predicted image sequence with current observations. Experimental results across 11 complex robotic tasks demonstrate that TeViR outperforms traditional methods leveraging sparse rewards and other state-of-the-art (SOTA) methods, achieving better sample efficiency and performance without ground truth environmental rewards. TeViR's ability to efficiently guide agents in complex environments highlights its potential to advance reinforcement learning applications in robotic manipulation.

Paperid: 3368, https://arxiv.org/pdf/2505.18344.pdf

Abstract:
Diffusion models have demonstrated state-of-the-art performance across vision, language, and scientific domains. Despite their empirical success, prior theoretical analyses of the sample complexity suffer from poor scaling with input data dimension or rely on unrealistic assumptions such as access to exact empirical risk minimizers. In this work, we provide a principled analysis of score estimation, establishing a sample complexity bound of $\widetilde{\mathcal{O}}(Îµ^{-6})$. Our approach leverages a structured decomposition of the score estimation error into statistical, approximation, and optimization errors, enabling us to eliminate the exponential dependence on neural network parameters that arises in prior analyses. It is the first such result which achieves sample complexity bounds without assuming access to the empirical risk minimizer of score function estimation loss.

Paperid: 3369, https://arxiv.org/pdf/2505.15818.pdf

Abstract:
Language-Guided object recognition in remote sensing imagery is crucial for large-scale mapping and automated data annotation. However, existing open-vocabulary and visual grounding methods rely on explicit category cues, limiting their ability to handle complex or implicit queries that require advanced reasoning. To address this issue, we introduce a new suite of tasks, including Instruction-Oriented Object Counting, Detection, and Segmentation (InstructCDS), covering open-vocabulary, open-ended, and open-subclass scenarios. We further present EarthInstruct, the first InstructCDS benchmark for earth observation. It is constructed from two diverse remote sensing datasets with varying spatial resolutions and annotation rules across 20 categories, necessitating models to interpret dataset-specific instructions. Given the scarcity of semantically rich labeled data in remote sensing, we propose InstructSAM, a training-free framework for instruction-driven object recognition. InstructSAM leverages large vision-language models to interpret user instructions and estimate object counts, employs SAM2 for mask proposal, and formulates mask-label assignment as a binary integer programming problem. By integrating semantic similarity with counting constraints, InstructSAM efficiently assigns categories to predicted masks without relying on confidence thresholds. Experiments demonstrate that InstructSAM matches or surpasses specialized baselines across multiple tasks while maintaining near-constant inference time regardless of object count, reducing output tokens by 89% and overall runtime by over 32% compared to direct generation approaches. We believe the contributions of the proposed tasks, benchmark, and effective approach will advance future research in developing versatile object recognition systems.

Paperid: 3370, https://arxiv.org/pdf/2505.14340.pdf

Abstract:
Plane geometry problem solving (PGPS) has recently gained significant attention as a benchmark to assess the multi-modal reasoning capabilities of large vision-language models. Despite the growing interest in PGPS, the research community still lacks a comprehensive overview that systematically synthesizes recent work in PGPS. To fill this gap, we present a survey of existing PGPS studies. We first categorize PGPS methods into an encoder-decoder framework and summarize the corresponding output formats used by their encoders and decoders. Subsequently, we classify and analyze these encoders and decoders according to their architectural designs. Finally, we outline major challenges and promising directions for future research. In particular, we discuss the hallucination issues arising during the encoding phase within encoder-decoder architectures, as well as the problem of data leakage in current PGPS benchmarks.

Paperid: 3371, https://arxiv.org/pdf/2505.03667.pdf

Abstract:
Text-to-image (T2I) diffusion models are effective at producing semantically aligned images, but their reliance on training data distributions limits their ability to synthesize truly novel, out-of-distribution concepts. Existing methods typically enhance creativity by combining pairs of known concepts, yielding compositions that, while out-of-distribution, remain linguistically describable and bounded within the existing semantic space. Inspired by the soft probabilistic outputs of classifiers on ambiguous inputs, we propose Distribution-Conditional Generation, a novel formulation that models creativity as image synthesis conditioned on class distributions, enabling semantically unconstrained creative generation. Building on this, we propose DisTok, an encoder-decoder framework that maps class distributions into a latent space and decodes them into tokens of creative concept. DisTok maintains a dynamic concept pool and iteratively sampling and fusing concept pairs, enabling the generation of tokens aligned with increasingly complex class distributions. To enforce distributional consistency, latent vectors sampled from a Gaussian prior are decoded into tokens and rendered into images, whose class distributions-predicted by a vision-language model-supervise the alignment between input distributions and the visual semantics of generated tokens. The resulting tokens are added to the concept pool for subsequent composition. Extensive experiments demonstrate that DisTok, by unifying distribution-conditioned fusion and sampling-based synthesis, enables efficient and flexible token-level generation, achieving state-of-the-art performance with superior text-image alignment and human preference scores.

Paperid: 3372, https://arxiv.org/pdf/2504.19276.pdf

Abstract:
High-quality preference data is essential for aligning foundation models with human values through preference learning. However, manual annotation of such data is often time-consuming and costly. Recent methods often adopt a self-rewarding approach, where the target model generates and annotates its own preference data, but this can lead to inaccuracies since the reward model shares weights with the target model, thereby amplifying inherent biases. To address these issues, we propose Anyprefer, a framework designed to synthesize high-quality preference data for aligning the target model. Anyprefer frames the data synthesis process as a cooperative two-player Markov Game, where the target model and the judge model collaborate together. Here, a series of external tools are introduced to assist the judge model in accurately rewarding the target model's responses, mitigating biases in the rewarding process. In addition, a feedback mechanism is introduced to optimize prompts for both models, enhancing collaboration and improving data quality. The synthesized data is compiled into a new preference dataset, Anyprefer-V1, consisting of 58K high-quality preference pairs. Extensive experiments show that Anyprefer significantly improves model alignment performance across four main applications, covering 21 datasets, achieving average improvements of 18.55% in five natural language generation datasets, 3.66% in nine vision-language understanding datasets, 30.05% in three medical image analysis datasets, and 16.00% in four visuo-motor control tasks.

Paperid: 3373, https://arxiv.org/pdf/2504.17315.pdf

Abstract:
This paper presents the technical solution proposed by Huawei Translation Service Center (HW-TSC) for the "End-to-End Document Image Machine Translation for Complex Layouts" competition at the 19th International Conference on Document Analysis and Recognition (DIMT25@ICDAR2025). Leveraging state-of-the-art open-source large vision-language model (LVLM), we introduce a training framework that combines multi-task learning with perceptual chain-of-thought to develop a comprehensive end-to-end document translation system. During the inference phase, we apply minimum Bayesian decoding and post-processing strategies to further enhance the system's translation capabilities. Our solution uniquely addresses both OCR-based and OCR-free document image translation tasks within a unified framework. This paper systematically details the training methods, inference strategies, LVLM base models, training data, experimental setups, and results, demonstrating an effective approach to document image machine translation.

Paperid: 3374, https://arxiv.org/pdf/2504.12661.pdf

Abstract:
Aligning Vision-Language Models (VLMs) with safety standards is essential to mitigate risks arising from their multimodal complexity, where integrating vision and language unveils subtle threats beyond the reach of conventional safeguards. Inspired by the insight that reasoning across modalities is key to preempting intricate vulnerabilities, we propose a novel direction for VLM safety: multimodal reasoning-driven prompt rewriting. To this end, we introduce VLMGuard-R1, a proactive framework that refines user inputs through a reasoning-guided rewriter, dynamically interpreting text-image interactions to deliver refined prompts that bolster safety across diverse VLM architectures without altering their core parameters. To achieve this, we devise a three-stage reasoning pipeline to synthesize a dataset that trains the rewriter to infer subtle threats, enabling tailored, actionable responses over generic refusals. Extensive experiments across three benchmarks with five VLMs reveal that VLMGuard-R1 outperforms four baselines. In particular, VLMGuard-R1 achieves a remarkable 43.59\% increase in average safety across five models on the SIUO benchmark.

Paperid: 3375, https://arxiv.org/pdf/2504.07956.pdf

Abstract:
The advancement of Chain-of-Thought (CoT) reasoning has significantly enhanced the capabilities of large language models (LLMs) and large vision-language models (LVLMs). However, a rigorous evaluation framework for video CoT reasoning remains absent. Current video benchmarks fail to adequately assess the reasoning process and expose whether failures stem from deficiencies in perception or reasoning capabilities. Therefore, we introduce VCR-Bench, a novel benchmark designed to comprehensively evaluate LVLMs' Video Chain-of-Thought Reasoning capabilities. VCR-Bench comprises 859 videos spanning a variety of video content and durations, along with 1,034 high-quality question-answer pairs. Each pair is manually annotated with a stepwise CoT rationale, where every step is tagged to indicate its association with the perception or reasoning capabilities. Furthermore, we design seven distinct task dimensions and propose the CoT score to assess the entire CoT process based on the stepwise tagged CoT rationals. Extensive experiments on VCR-Bench highlight substantial limitations in current LVLMs. Even the top-performing model, o1, only achieves a 62.8% CoT score and an 56.7% accuracy, while most models score below 40%. Experiments show most models score lower on perception than reasoning steps, revealing LVLMs' key bottleneck in temporal-spatial information processing for complex video reasoning. A robust positive correlation between the CoT score and accuracy confirms the validity of our evaluation framework and underscores the critical role of CoT reasoning in solving complex video reasoning tasks. We hope VCR-Bench to serve as a standardized evaluation framework and expose the actual drawbacks in complex video reasoning task.

Paperid: 3376, https://arxiv.org/pdf/2503.22720.pdf

Abstract:
Representation Engineering (RepE) has emerged as a powerful paradigm for enhancing AI transparency by focusing on high-level representations rather than individual neurons or circuits. It has proven effective in improving interpretability and control, showing that representations can emerge, propagate, and shape final model outputs in large language models (LLMs). However, in Vision-Language Models (VLMs), visual input can override factual linguistic knowledge, leading to hallucinated responses that contradict reality. To address this challenge, we make the first attempt to extend RepE to VLMs, analyzing how multimodal representations are preserved and transformed. Building on our findings and drawing inspiration from successful RepE applications, we develop a theoretical framework that explains the stability of neural activity across layers using the principal eigenvector, uncovering the underlying mechanism of RepE. We empirically validate these instrinsic properties, demonstrating their broad applicability and significance. By bridging theoretical insights with empirical validation, this work transforms RepE from a descriptive tool into a structured theoretical framework, opening new directions for improving AI robustness, fairness, and transparency.

Paperid: 3377, https://arxiv.org/pdf/2503.19900.pdf

Abstract:
The rapid advancement of large vision-language models (LVLMs) has driven significant progress in multimodal tasks, enabling models to interpret, reason, and generate outputs across both visual and textual domains. While excelling in generative tasks, existing LVLMs often face limitations in tasks requiring high-fidelity representation learning, such as generating image or text embeddings for retrieval. Recent work has proposed finetuning LVLMs for representational learning, but the fine-tuned model often loses its generative capabilities due to the representational learning training paradigm. To address this trade-off, we introduce CAFe, a contrastive-autoregressive fine-tuning framework that enhances LVLMs for both representation and generative tasks. By integrating a contrastive objective with autoregressive language modeling, our approach unifies these traditionally separate tasks, achieving state-of-the-art results in both multimodal retrieval and multimodal generative benchmarks, including object hallucination (OH) mitigation. CAFe establishes a novel framework that synergizes embedding and generative functionalities in a single model, setting a foundation for future multimodal models that excel in both retrieval precision and coherent output generation.

Paperid: 3378, https://arxiv.org/pdf/2503.13794.pdf

Abstract:
Large foundation models trained on large-scale vision-language data can boost Open-Vocabulary Object Detection (OVD) via synthetic training data, yet the hand-crafted pipelines often introduce bias and overfit to specific prompts. We sidestep this issue by directly fusing hidden states from Large Language Models (LLMs) into detectors-an avenue surprisingly under-explored. This paper presents a systematic method to enhance visual grounding by utilizing decoder layers of the LLM of an MLLM. We introduce a zero-initialized cross-attention adapter to enable efficient knowledge fusion from LLMs to object detectors, a new approach called LED (LLM Enhanced Open-Vocabulary Object Detection). We find that intermediate LLM layers already encode rich spatial semantics; adapting only the early layers yields most of the gain. With Swin-T as the vision encoder, Qwen2-0.5B + LED lifts GroundingDINO by 3.82 % on OmniLabel at just 8.7 % extra GFLOPs, and a larger vision backbone pushes the improvement to 6.22 %. Extensive ablations on adapter variants, LLM scales and fusion depths further corroborate our design.

Paperid: 3379, https://arxiv.org/pdf/2503.11742.pdf

Abstract:
Vision-language models (VLMs) often inherit the biases and unsafe associations present within their large-scale training dataset. While recent approaches mitigate unsafe behaviors, their evaluation focuses on how safe the model is on unsafe inputs, ignoring potential shortcomings on safe ones. In this paper, we first revise safety evaluation by introducing SafeGround, a new set of metrics that evaluate safety at different levels of granularity. With this metric, we uncover a surprising issue of training-based methods: they make the model less safe on safe inputs. From this finding, we take a different direction and explore whether it is possible to make a model safer without training, introducing Unsafe Weights Manipulation (UWM). UWM uses a calibration set of safe and unsafe instances to compare activations between safe and unsafe content, identifying the most important parameters for processing the latter. Their values are then manipulated via negation. Experiments show that UWM achieves the best tradeoff between safety and knowledge preservation, consistently improving VLMs on unsafe queries while outperforming even training-based state-of-the-art methods on safe ones.

Paperid: 3380, https://arxiv.org/pdf/2503.10427.pdf

Abstract:
In this paper, we propose a comprehensive evaluation benchmark for Visual Language Models (VLM) in Traditional Chinese. Our evaluation suite, the first of its kind, contains two complementary components: (1) VisTW-MCQ, a collection of manually curated exam multi-choice questions from 21 academic subjects designed to test the broad knowledge and reasoning capabilities of VLMs; and (2) VisTW-Dialogue, an open dialogue benchmark comprising 131 image-question pairs manually created to evaluate VLMs' ability in free-form dialogue generation within Taiwanese cultural contexts. These benchmarks address a critical gap in the evaluation landscape, where existing benchmarks predominantly focus on English or Simplified Chinese, neglecting the unique linguistic and cultural aspects of Traditional Chinese used in regions like Taiwan and Hong Kong. Our analysis reveals significant performance differences across various VLMs and highlights specific challenges in processing Traditional Chinese visual content.

Paperid: 3381, https://arxiv.org/pdf/2503.07487.pdf

Abstract:
Recently, Multimodal Large Language Models (MLLMs) have demonstrated exceptional capabilities in visual understanding and reasoning across various vision-language tasks. However, we found that MLLMs cannot process effectively from fine-grained medical image data in the traditional Visual Question Answering (VQA) pipeline, as they do not exploit the captured features and available medical knowledge fully, results in MLLMs usually performing poorly in zero-shot medical disease recognition. Fortunately, this limitation does not indicate that MLLMs are fundamentally incapable of addressing fine-grained recognition tasks. From a feature representation perspective, MLLMs demonstrate considerable potential for tackling such challenging problems. Thus, to address this challenge, we propose LLaVA-RadZ, a simple yet effective framework for zero-shot medical disease recognition via utilizing the existing MLLM features. Specifically, we design an end-to-end training strategy, termed Decoding-Side Feature Alignment Training (DFAT) to take advantage of the characteristics of the MLLM decoder architecture and incorporate modality-specific tokens tailored for different modalities. Additionally, we introduce a Domain Knowledge Anchoring Module (DKAM) to exploit the intrinsic medical knowledge of large models, which mitigates the category semantic gap in image-text alignment. Extensive experiments demonstrate that our LLaVA-RadZ significantly outperforms traditional MLLMs in zero-shot disease recognition, achieving the comparable performance to the well-established and highly-optimized CLIP-based approaches.

Paperid: 3382, https://arxiv.org/pdf/2503.06223.pdf

Abstract:
Vision-Language Models (VLMs) are vulnerable to jailbreak attacks, where adversaries bypass safety mechanisms to elicit harmful outputs. In this work, we examine an insidious variant of this threat: toxic continuation. Unlike standard jailbreaks that rely solely on malicious instructions, toxic continuation arises when the model is given a malicious input alongside a partial toxic output, resulting in harmful completions. This vulnerability poses a unique challenge in multimodal settings, where even subtle image variations can disproportionately affect the model's response. To this end, we propose RedDiffuser (RedDiff), the first red teaming framework that uses reinforcement learning to fine-tune diffusion models into generating natural-looking adversarial images that induce toxic continuations. RedDiffuser integrates a greedy search procedure for selecting candidate image prompts with reinforcement fine-tuning that jointly promotes toxic output and semantic coherence. Experiments demonstrate that RedDiffuser significantly increases the toxicity rate in LLaVA outputs by 10.69% and 8.91% on the original and hold-out sets, respectively. It also exhibits strong transferability, increasing toxicity rates on Gemini by 5.1% and on LLaMA-Vision by 26.83%. These findings uncover a cross-modal toxicity amplification vulnerability in current VLM alignment, highlighting the need for robust multimodal red teaming. We will release the RedDiffuser codebase to support future research.

Paperid: 3383, https://arxiv.org/pdf/2502.14113.pdf

Abstract:
Recent advances in vision language models (VLM) have been driven by contrastive models such as CLIP, which learn to associate visual information with their corresponding text descriptions. However, these models have limitations in understanding complex compositional scenes involving multiple objects and their spatial relationships. To address these challenges, we propose a novel approach that diverges from commonly used strategies, which rely on the design of hard-negative augmentations. Instead, our work focuses on integrating inductive biases into pre-trained CLIP-like models to improve their compositional understanding without using any additional hard-negatives. To that end, we introduce a binding module that connects a scene graph, derived from a text description, with a slot-structured image representation, facilitating a structured similarity assessment between the two modalities. We also leverage relationships as text-conditioned visual constraints, thereby capturing the intricate interactions between objects and their contextual relationships more effectively. Our resulting model not only enhances the performance of CLIP-based models in multi-object compositional understanding but also paves the way towards more accurate and sample-efficient image-text matching of complex scenes.

Paperid: 3384, https://arxiv.org/pdf/2502.11360.pdf

Abstract:
We introduce GeoDANO, a geometric vision-language model (VLM) with a domain-agnostic vision encoder, for solving plane geometry problems. Although VLMs have been employed for solving geometry problems, their ability to recognize geometric features remains insufficiently analyzed. To address this gap, we propose a benchmark that evaluates the recognition of visual geometric features, including primitives such as dots and lines, and relations such as orthogonality. Our preliminary study shows that vision encoders often used in general-purpose VLMs, e.g., OpenCLIP, fail to detect these features and struggle to generalize across domains. We develop GeoCLIP, a CLIP based model trained on synthetic geometric diagram-caption pairs to overcome the limitation. Benchmark results show that GeoCLIP outperforms existing vision encoders in recognizing geometric features. We then propose our VLM, GeoDANO, which augments GeoCLIP with a domain adaptation strategy for unseen diagram styles. GeoDANO outperforms specialized methods for plane geometry problems and GPT-4o on MathVerse.

Paperid: 3385, https://arxiv.org/pdf/2501.16947.pdf

Abstract:
The advances in Vision-Language models (VLMs) offer exciting opportunities for robotic applications involving image geo-localization, the problem of identifying the geo-coordinates of a place based on visual data only. Recent research works have focused on using a VLM as embeddings extractor for geo-localization, however, the most sophisticated VLMs may only be available as black boxes that are accessible through an API, and come with a number of limitations: there is no access to training data, model features and gradients; retraining is not possible; the number of predictions may be limited by the API; training on model outputs is often prohibited; and queries are open-ended. The utilization of a VLM as a stand-alone, zero-shot geo-localization system using a single text-based prompt is largely unexplored. To bridge this gap, this paper undertakes the first systematic study, to the best of our knowledge, to investigate the potential of some of the state-of-the-art VLMs as stand-alone, zero-shot geo-localization systems in a black-box setting with realistic constraints. We consider three main scenarios for this thorough investigation: a) fixed text-based prompt; b) semantically-equivalent text-based prompts; and c) semantically-equivalent query images. We also take into account the auto-regressive and probabilistic generation process of the VLMs when investigating their utility for geo-localization task by using model consistency as a metric in addition to traditional accuracy. Our work provides new insights in the capabilities of different VLMs for the above-mentioned scenarios.

Paperid: 3386, https://arxiv.org/pdf/2412.15509.pdf

Abstract:
In this paper, we present our methods and results for the Video-To-Text (VTT) task at TRECVid 2024, exploring the capabilities of Vision-Language Models (VLMs) like LLaVA and LLaVA-NeXT-Video in generating natural language descriptions for video content. We investigate the impact of fine-tuning VLMs on VTT datasets to enhance description accuracy, contextual relevance, and linguistic consistency. Our analysis reveals that fine-tuning substantially improves the model's ability to produce more detailed and domain-aligned text, bridging the gap between generic VLM tasks and the specialized needs of VTT. Experimental results demonstrate that our fine-tuned model outperforms baseline VLMs across various evaluation metrics, underscoring the importance of domain-specific tuning for complex VTT tasks.

Paperid: 3387, https://arxiv.org/pdf/2411.17558.pdf

Abstract:
Visual Question Answering (VQA) is a challenge task that combines natural language processing and computer vision techniques and gradually becomes a benchmark test task in multimodal large language models (MLLMs). The goal of our survey is to provide an overview of the development of VQA and a detailed description of the latest models with high timeliness. This survey gives an up-to-date synthesis of natural language understanding of images and text, as well as the knowledge reasoning module based on image-question information on the core VQA tasks. In addition, we elaborate on recent advances in extracting and fusing modal information with vision-language pretraining models and multimodal large language models in VQA. We also exhaustively review the progress of knowledge reasoning in VQA by detailing the extraction of internal knowledge and the introduction of external knowledge. Finally, we present the datasets of VQA and different evaluation metrics and discuss possible directions for future work.

Paperid: 3388, https://arxiv.org/pdf/2411.12286.pdf

Abstract:
Inferring affordable (i.e., graspable) parts of arbitrary objects based on human specifications is essential for robots advancing toward open-vocabulary manipulation. Current grasp planners, however, are hindered by limited vision-language comprehension and time-consuming 3D radiance modeling, restricting real-time, open-vocabulary interactions with objects. To address these limitations, we propose GLOVER, a unified Generalizable Open-Vocabulary Affordance Reasoning framework, which fine-tunes the Large Language Models (LLMs) to predict the visual affordance of graspable object parts within RGB feature space. We compile a dataset of over 10,000 images from human-object interactions, annotated with unified visual and linguistic affordance labels, to enable multi-modal fine-tuning. GLOVER inherits world knowledge and common-sense reasoning from LLMs, facilitating more fine-grained object understanding and sophisticated tool-use reasoning. To enable effective real-world deployment, we present Affordance-Aware Grasping Estimation (AGE), a non-parametric grasp planner that aligns the gripper pose with a superquadric surface derived from affordance data. In evaluations across 30 table-top real-world scenes, GLOVER achieves success rates of 86.0% in part identification and 76.3% in grasping, with speeds approximately 29 times faster in affordance reasoning and 40 times faster in grasping pose estimation than the previous state-of-the-art. We also validate the generalization across embodiments, showing effectiveness in humanoid robots with dexterous hands.

Paperid: 3389, https://arxiv.org/pdf/2410.03907.pdf

Abstract:
Large language models~(LLMs) have been adopted to process textual task description and accomplish procedural planning in embodied AI tasks because of their powerful reasoning ability. However, there is still lack of study on how vision language models~(VLMs) behave when multi-modal task inputs are considered. Counterfactual planning that evaluates the model's reasoning ability over alternative task situations are also under exploited. In order to evaluate the planning ability of both multi-modal and counterfactual aspects, we propose ActPlan-1K. ActPlan-1K is a multi-modal planning benchmark constructed based on ChatGPT and household activity simulator iGibson2. The benchmark consists of 153 activities and 1,187 instances. Each instance describing one activity has a natural language task description and multiple environment images from the simulator. The gold plan of each instance is action sequences over the objects in provided scenes. Both the correctness and commonsense satisfaction are evaluated on typical VLMs. It turns out that current VLMs are still struggling at generating human-level procedural plans for both normal activities and counterfactual activities. We further provide automatic evaluation metrics by finetuning over BLEURT model to facilitate future research on our benchmark.

Paperid: 3390, https://arxiv.org/pdf/2410.01506.pdf

Abstract:
In computer vision tasks, features often come from diverse representations, domains (e.g., indoor and outdoor), and modalities (e.g., text, images, and videos). Effectively fusing these features is essential for robust performance, especially with the availability of powerful pre-trained models like vision-language models. However, common fusion methods, such as concatenation, element-wise operations, and non-linear techniques, often fail to capture structural relationships, deep feature interactions, and suffer from inefficiency or misalignment of features across domains or modalities. In this paper, we shift from high-dimensional feature space to a lower-dimensional, interpretable graph space by constructing relationship graphs that encode feature relationships at different levels, e.g., clip, frame, patch, token, etc. To capture deeper interactions, we expand graphs through iterative graph relationship updates and introduce a learnable graph fusion operator to integrate these expanded relationships for more effective fusion. Our approach is relationship-centric, operates in a homogeneous space, and is mathematically principled, resembling element-wise relationship score aggregation via multilinear polynomials. We demonstrate the effectiveness of our graph-based fusion method on video anomaly detection, showing strong performance across multi-representational, multi-modal, and multi-domain feature fusion tasks.

Paperid: 3391, https://arxiv.org/pdf/2409.19330.pdf

Abstract:
Medical image analysis is crucial in modern radiological diagnostics, especially given the exponential growth in medical imaging data. The demand for automated report generation systems has become increasingly urgent. While prior research has mainly focused on using machine learning and multimodal language models for 2D medical images, the generation of reports for 3D medical images has been less explored due to data scarcity and computational complexities. This paper introduces 3D-CT-GPT, a Visual Question Answering (VQA)-based medical visual language model specifically designed for generating radiology reports from 3D CT scans, particularly chest CTs. Extensive experiments on both public and private datasets demonstrate that 3D-CT-GPT significantly outperforms existing methods in terms of report accuracy and quality. Although current methods are few, including the partially open-source CT2Rep and the open-source M3D, we ensured fair comparison through appropriate data conversion and evaluation methodologies. Experimental results indicate that 3D-CT-GPT enhances diagnostic accuracy and report coherence, establishing itself as a robust solution for clinical radiology report generation. Future work will focus on expanding the dataset and further optimizing the model to enhance its performance and applicability.

Paperid: 3392, https://arxiv.org/pdf/2409.12778.pdf

Abstract:
In this paper, we address the challenging problem of cross-modal (image-to-events) adaptation for event-based recognition without accessing any labeled source image data. This task is arduous due to the substantial modality gap between images and events. With only a pre-trained source model available, the key challenge lies in extracting knowledge from this model and effectively transferring knowledge to the event-based domain. Inspired by the natural ability of language to convey semantics across different modalities, we propose EventDance++, a novel framework that tackles this unsupervised source-free cross-modal adaptation problem from a language-guided perspective. We introduce a language-guided reconstruction-based modality bridging (L-RMB) module, which reconstructs intensity frames from events in a self-supervised manner. Importantly, it leverages a vision-language model to provide further supervision, enriching the surrogate images and enhancing modality bridging. This enables the creation of surrogate images to extract knowledge (i.e., labels) from the source model. On top, we propose a multi-representation knowledge adaptation (MKA) module to transfer knowledge to target models, utilizing multiple event representations to capture the spatiotemporal characteristics of events fully. The L-RMB and MKA modules are jointly optimized to achieve optimal performance in bridging the modality gap. Experiments on three benchmark datasets demonstrate that EventDance++ performs on par with methods that utilize source data, validating the effectiveness of our language-guided approach in event-based recognition.

Paperid: 3393, https://arxiv.org/pdf/2601.21409.pdf

Abstract:
Adaptive navigation in unfamiliar indoor environments is crucial for household service robots. Despite advances in zero-shot perception and reasoning from vision-language models, existing navigation systems still rely on single-pass scoring at the decision layer, leading to overconfident long-horizon errors and redundant exploration. To tackle these problems, we propose Dual-Stance Cooperative Debate Navigation (DSCD-Nav), a decision mechanism that replaces one-shot scoring with stance-based cross-checking and evidence-aware arbitration to improve action reliability under partial observability. Specifically, given the same observation and candidate action set, we explicitly construct two stances by conditioning the evaluation on diverse and complementary objectives: a Task-Scene Understanding (TSU) stance that prioritizes goal progress from scene-layout cues, and a Safety-Information Balancing (SIB) stance that emphasizes risk and information value. The stances conduct a cooperative debate and make policy by cross-checking their top candidates with cue-grounded arguments. Then, a Navigation Consensus Arbitration (NCA) agent is employed to consolidate both sides' reasons and evidence, optionally triggering lightweight micro-probing to verify uncertain choices, preserving NCA's primary intent while disambiguating. Experiments on HM3Dv1, HM3Dv2, and MP3D demonstrate consistent improvements in success and path efficiency while reducing exploration redundancy.

Paperid: 3394, https://arxiv.org/pdf/2601.12042.pdf

Abstract:
Visual token compression is widely adopted to improve the inference efficiency of Large Vision-Language Models (LVLMs), enabling their deployment in latency-sensitive and resource-constrained scenarios. However, existing work has mainly focused on efficiency and performance, while the security implications of visual token compression remain largely unexplored. In this work, we first reveal that visual token compression substantially degrades the robustness of LVLMs: models that are robust under uncompressed inference become highly vulnerable once compression is enabled. These vulnerabilities are state-specific; failure modes emerge only in the compressed setting and completely disappear when compression is disabled, making them particularly hidden and difficult to diagnose. By analyzing the key stages of the compression process, we identify instability in token importance ranking as the primary cause of this robustness degradation. Small and imperceptible perturbations can significantly alter token rankings, leading the compression mechanism to mistakenly discard task-critical information and ultimately causing model failure. Motivated by this observation, we propose a Compression-Aware Attack to systematically study and exploit this vulnerability. CAA directly targets the token selection mechanism and induces failures exclusively under compressed inference. We further extend this approach to more realistic black-box settings and introduce Transfer CAA, where neither the target model nor the compression configuration is accessible. We further evaluate potential defenses and find that they provide only limited protection. Extensive experiments across models, datasets, and compression methods show that visual token compression significantly undermines robustness, revealing a previously overlooked efficiency-security trade-off.

Paperid: 3395, https://arxiv.org/pdf/2601.04442.pdf

Abstract:
Large Vision-Language Models (LVLMs) have exhibited strong reasoning capabilities through chain-of-thought mechanisms that generate step-by-step rationales. However, such slow-thinking approaches often lead to overthinking, where models produce excessively verbose responses even for simple queries, resulting in test-time inefficiency and even degraded accuracy. Prior work has attempted to mitigate this issue via adaptive reasoning strategies, but these methods largely overlook a fundamental bottleneck: visual perception failures. We argue that stable reasoning critically depends on low-level visual grounding, and that reasoning errors often originate from imperfect perception rather than insufficient deliberation. To address this limitation, we propose Gated Perception-Reasoning Optimization (GPRO), a meta-reasoning controller that dynamically routes computation among three decision paths at each generation step: a lightweight fast path, a slow perception path for re-examining visual inputs, and a slow reasoning path for internal self-reflection. To learn this distinction, we derive large-scale failure attribution supervision from approximately 790k samples, using teacher models to distinguish perceptual hallucinations from reasoning errors. We then train the controller with multi-objective reinforcement learning to optimize the trade-off between task accuracy and computational cost under uncertainty. Experiments on five benchmarks demonstrate that GPRO substantially improves both accuracy and efficiency, outperforming recent slow-thinking methods while generating significantly shorter responses.

Paperid: 3396, https://arxiv.org/pdf/2601.04061.pdf

Abstract:
Generalist Vision-Language-Action models are currently hindered by the scarcity of robotic data compared to the abundance of human video demonstrations. Existing Latent Action Models attempt to leverage video data but often suffer from visual entanglement, capturing noise rather than manipulation skills. To address this, we propose Contrastive Latent Action Pretraining (CLAP), a framework that aligns the visual latent space from videos with a proprioceptive latent space from robot trajectories. By employing contrastive learning, CLAP maps video transitions onto a quantized, physically executable codebook. Building on this representation, we introduce a dual-formulation VLA framework offering both CLAP-NTP, an autoregressive model excelling at instruction following and object generalization, and CLAP-RF, a Rectified Flow-based policy designed for high-frequency, precise manipulation. Furthermore, we propose a Knowledge Matching (KM) regularization strategy to mitigate catastrophic forgetting during fine-tuning. Extensive experiments demonstrate that CLAP significantly outperforms strong baselines, enabling the effective transfer of skills from human videos to robotic execution. Project page: https://lin-shan.com/CLAP/.

Paperid: 3397, https://arxiv.org/pdf/2512.21778.pdf

Abstract:
Segmenting long-form videos into semantically coherent scenes is a fundamental task in large-scale video understanding. Existing encoder-based methods are limited by visual-centric biases, classify each shot in isolation without leveraging sequential dependencies, and lack both narrative understanding and explainability. In this paper, we present Scene-VLM, the first fine-tuned vision-language model (VLM) framework for video scene segmentation. Scene-VLM jointly processes visual and textual cues including frames, transcriptions, and optional metadata to enable multimodal reasoning across consecutive shots. The model generates predictions sequentially with causal dependencies among shots and introduces a context-focus window mechanism to ensure sufficient temporal context for each shot-level decision. In addition, we propose a scheme to extract confidence scores from the token-level logits of the VLM, enabling controllable precision-recall trade-offs that were previously limited to encoder-based methods. Furthermore, we demonstrate that our model can be aligned to generate coherent natural-language rationales for its boundary decisions through minimal targeted supervision. Our approach achieves state-of-the-art performance on standard scene segmentation benchmarks. On MovieNet, for example, Scene-VLM yields significant improvements of +6 AP and +13.7 F1 over the previous leading method.

Paperid: 3398, https://arxiv.org/pdf/2512.18813.pdf

Abstract:
Large Vision-Language Models (LVLMs) have shown remarkable capabilities, yet hallucinations remain a persistent challenge. This work presents a systematic analysis of the internal evolution of visual perception and token generation in LVLMs, revealing two key patterns. First, perception follows a three-stage GATE process: early layers perform a Global scan, intermediate layers Approach and Tighten on core content, and later layers Explore supplementary regions. Second, generation exhibits an SAD (Subdominant Accumulation to Dominant) pattern, where hallucinated tokens arise from the repeated accumulation of subdominant tokens lacking support from attention (visual perception) or feed-forward network (internal knowledge). Guided by these findings, we devise the VDC (Validated Dominance Correction) strategy, which detects unsupported tokens and replaces them with validated dominant ones to improve output reliability. Extensive experiments across multiple models and benchmarks confirm that VDC substantially mitigates hallucinations.

Paperid: 3399, https://arxiv.org/pdf/2512.18554.pdf

Abstract:
Medical Large Vision-Language Models (Med-LVLMs) have shown promising results in clinical applications, but often suffer from hallucinated outputs due to misaligned visual understanding. In this work, we identify two fundamental limitations contributing to this issue: insufficient visual representation learning and poor visual attention alignment. To address these problems, we propose MEDALIGN, a simple, lightweight alignment distillation framework that transfers visual alignment knowledge from a domain-specific Contrastive Language-Image Pre-training (CLIP) model to Med-LVLMs. MEDALIGN introduces two distillation losses: a spatial-aware visual alignment loss based on visual token-level similarity structures, and an attention-aware distillation loss that guides attention toward diagnostically relevant regions. Extensive experiments on medical report generation and medical visual question answering (VQA) benchmarks show that MEDALIGN consistently improves both performance and interpretability, yielding more visually grounded outputs.

Paperid: 3400, https://arxiv.org/pdf/2512.09907.pdf

Abstract:
Vision-Language Models (VLMs) have achieved impressive progress in perceiving and describing visual environments. However, their ability to proactively reason and act based solely on visual inputs, without explicit textual prompts, remains underexplored. We introduce a new task, Visual Action Reasoning, and propose VisualActBench, a large-scale benchmark comprising 1,074 videos and 3,733 human-annotated actions across four real-world scenarios. Each action is labeled with an Action Prioritization Level (APL) and a proactive-reactive type to assess models' human-aligned reasoning and value sensitivity. We evaluate 29 VLMs on VisualActBench and find that while frontier models like GPT4o demonstrate relatively strong performance, a significant gap remains compared to human-level reasoning, particularly in generating proactive, high-priority actions. Our results highlight limitations in current VLMs' ability to interpret complex context, anticipate outcomes, and align with human decision-making frameworks. VisualActBench establishes a comprehensive foundation for assessing and improving the real-world readiness of proactive, vision-centric AI agents.

Paperid: 3401, https://arxiv.org/pdf/2512.07245.pdf

Abstract:
Textual explanations make image classifier decisions transparent by describing the prediction rationale in natural language. Large vision-language models can generate captions but are designed for general visual understanding, not classifier-specific reasoning. Existing zero-shot explanation methods align global image features with language, producing descriptions of what is visible rather than what drives the prediction. We propose TEXTER, which overcomes this limitation by isolating decision-critical features before alignment. TEXTER identifies the neurons contributing to the prediction and emphasizes the features encoded in those neurons -- i.e., the decision-critical features. It then maps these emphasized features into the CLIP feature space to retrieve textual explanations that reflect the model's reasoning. A sparse autoencoder further improves interpretability, particularly for Transformer architectures. Extensive experiments show that TEXTER generates more faithful and interpretable explanations than existing methods. The code will be publicly released.

Paperid: 3402, https://arxiv.org/pdf/2512.04537.pdf

Abstract:
The advancement of embodied AI has unlocked significant potential for intelligent humanoid robots. However, progress in both Vision-Language-Action (VLA) models and world models is severely hampered by the scarcity of large-scale, diverse training data. A promising solution is to "robotize" web-scale human videos, which has been proven effective for policy training. However, these solutions mainly "overlay" robot arms to egocentric videos, which cannot handle complex full-body motions and scene occlusions in third-person videos, making them unsuitable for robotizing humans. To bridge this gap, we introduce X-Humanoid, a generative video editing approach that adapts the powerful Wan 2.2 model into a video-to-video structure and finetunes it for the human-to-humanoid translation task. This finetuning requires paired human-humanoid videos, so we designed a scalable data creation pipeline, turning community assets into 17+ hours of paired synthetic videos using Unreal Engine. We then apply our trained model to 60 hours of the Ego-Exo4D videos, generating and releasing a new large-scale dataset of over 3.6 million "robotized" humanoid video frames. Quantitative analysis and user studies confirm our method's superiority over existing baselines: 69% of users rated it best for motion consistency, and 62.1% for embodiment correctness.

Paperid: 3403, https://arxiv.org/pdf/2512.00041.pdf

Abstract:
Vision-and-Language Navigation (VLN) requires agents to follow language instructions while acting in continuous real-world spaces. Prior image imagination based VLN work shows benefits for discrete panoramas but lacks online, action-conditioned predictions and does not produce explicit planning values; moreover, many methods replace the planner with long-horizon objectives that are brittle and slow. To bridge this gap, we propose VISTAv2, a generative world model that rolls out egocentric future views conditioned on past observations, candidate action sequences, and instructions, and projects them into an online value map for planning. Unlike prior approaches, VISTAv2 does not replace the planner. The online value map is fused at score level with the base objective, providing reachability and risk-aware guidance. Concretely, we employ an action-aware Conditional Diffusion Transformer video predictor to synthesize short-horizon futures, align them with the natural language instruction via a vision-language scorer, and fuse multiple rollouts in a differentiable imagination-to-value head to output an imagined egocentric value map. For efficiency, rollouts occur in VAE latent space with a distilled sampler and sparse decoding, enabling inference on a single consumer GPU. Evaluated on MP3D and RoboTHOR, VISTAv2 improves over strong baselines, and ablations show that action-conditioned imagination, instruction-guided value fusion, and the online value-map planner are all critical, suggesting that VISTAv2 offers a practical and interpretable route to robust VLN.

Paperid: 3404, https://arxiv.org/pdf/2511.20737.pdf

Abstract:
User interface (UI) design is an iterative process in which designers progressively refine their work with design software such as Figma or Sketch. Recent advances in vision language models (VLMs) with tool invocation suggest these models can operate design software to edit a UI design through iteration. Understanding and enhancing this capacity is important, as it highlights VLMs' potential to collaborate with designers within conventional software. However, as no existing benchmark evaluates tool-based design performance, the capacity remains unknown. To address this, we introduce CANVAS, a benchmark for VLMs on tool-based user interface design. Our benchmark contains 598 tool-based design tasks paired with ground-truth references sampled from 3.3K mobile UI designs across 30 function-based categories (e.g., onboarding, messaging). In each task, a VLM updates the design step-by-step through context-based tool invocations (e.g., create a rectangle as a button background), linked to design software. Specifically, CANVAS incorporates two task types: (i) design replication evaluates the ability to reproduce a whole UI screen; (ii) design modification evaluates the ability to modify a specific part of an existing screen. Results suggest that leading models exhibit more strategic tool invocations, improving design quality. Furthermore, we identify common error patterns models exhibit, guiding future work in enhancing tool-based design capabilities.

Paperid: 3405, https://arxiv.org/pdf/2511.19261.pdf

Abstract:
Humans can perceive and understand 3D space and long videos from sequential visual observations. But do vision-language models (VLMs) can? Recent work demonstrates that even state-of-the-art VLMs still struggle to understand 3D space and long videos, although they are powerful in typical vision-language tasks. Current methods often rely on specialized architectural designs to improve performance for 3D tasks and video understanding tasks separately. In contrast, we propose LAST, short for LeArn to Think in Space and Time, to jointly improve 3D spatial and long video understanding for general VLMs with only a set of 2D images as inputs. LAST makes VLMs think in space and time rather than only with text before giving the final answer, building visual thinking trajectories in 3D space and temporal dimension. We demonstrate the effectiveness of LAST in two scenarios: 1) zero-shot, where we directly prompt proprietary models; and 2) fine-tuning general VLMs with data that include thinking trajectories in 3D space and time. We show that LAST brings substantial gains in various benchmarks, including 3 spatial understanding, 4 video understanding, and 3 image understanding tasks. Notably, 15.8% gains on EgoSchema with GPT-4o in a zero-shot manner and 8.3 gains on VSI-Bench compared with Qwen2.5-VL-7B.

Paperid: 3406, https://arxiv.org/pdf/2511.19032.pdf

Abstract:
Despite the remarkable reasoning abilities of large vision-language models (LVLMs), their robustness under visual corruptions remains insufficiently studied. Existing evaluation paradigms exhibit two major limitations: 1) the dominance of low-discriminative samples in current datasets masks the real robustness gap between models; and 2) conventional accuracy-based metric fail to capture the degradation of the underlying prediction structure. To bridge these gaps, we introduce Bench-C, a comprehensive benchmark emphasizing discriminative samples for assessing corruption robustness, where a selection strategy is proposed to jointly consider the prediction inconsistency under corruption and the semantic diversity. Furthermore, we propose the Robustness Alignment Score (RAS), a unified metric that measures degradation in logit-level prediction structure by considering the shifts in prediction uncertainty and calibration alignment. Comprehensive experiments and analysis reveal several interesting findings: 1) model behaviors exhibit distinguish patterns under corruptions, such as erroneous confidence and hesitation; 2) despite subtle corruption may lead to a slight accuracy gain, the overall prediction structure still degrades; 3) by decomposing corruption robustness into destructive and corrective components, the distinct failure and recovery patterns across models can be revealed.

Paperid: 3407, https://arxiv.org/pdf/2511.16602.pdf

Abstract:
Developing a universal and versatile embodied intelligence system presents two primary challenges: the critical embodied data bottleneck, where real-world data is scarce and expensive, and the algorithmic inefficiency of existing methods, which are resource-prohibitive. To address these limitations, we introduce Deliberate Practice Policy Optimization (DPPO), a metacognitive ``Metaloop'' training framework that dynamically alternates between supervised fine-tuning (competence expansion) and reinforcement learning (skill refinement). This enables automatic weakness identification and targeted resource allocation, specifically designed to maximize learning efficiency from sparse, finite data. Theoretically, DPPO can be formalised as a unified preference-learning framework. Empirically, training a vision-language embodied model with DPPO, referred to as Pelican-VL 1.0, yields a 20.3% performance improvement over the base model and surpasses open-source models at the 100B-parameter scale by 10.6%. We are open-sourcing both the models and code, providing the first systematic framework that alleviates the data and resource bottleneck and enables the community to build versatile embodied agents efficiently.

Paperid: 3408, https://arxiv.org/pdf/2511.16170.pdf

Abstract:
Open-vocabulary semantic segmentation (OVSS) employs pixel-level vision-language alignment to associate category-related prompts with corresponding pixels. A key challenge is enhancing the multimodal dense prediction capability, specifically this pixel-level multimodal alignment. Although existing methods achieve promising results by leveraging CLIP's vision-language alignment, they rarely investigate the performance boundaries of CLIP for dense prediction from an interpretability mechanisms perspective. In this work, we systematically investigate CLIP's internal mechanisms and identify a critical phenomenon: analogous to human distraction, CLIP diverts significant attention resources from target regions to irrelevant tokens. Our analysis reveals that these tokens arise from dimension-specific over-activation; filtering them enhances CLIP's dense prediction performance. Consequently, we propose ReFocusing CLIP (RF-CLIP), a training-free approach that emulates human distraction-refocusing behavior to redirect attention from distraction tokens back to target regions, thereby refining CLIP's multimodal alignment granularity. Our method achieves SOTA performance on eight benchmarks while maintaining high inference efficiency.

Paperid: 3409, https://arxiv.org/pdf/2511.11407.pdf

Abstract:
Multimodal Large Language Models are increasingly applied to biomedical imaging, yet scientific reasoning for microscopy remains limited by the scarcity of large-scale, high-quality training data. We introduce MicroVQA++, a three-stage, large-scale and high-quality microscopy VQA corpus derived from the BIOMEDICA archive. Stage one bootstraps supervision from expert-validated figure-caption pairs sourced from peer-reviewed articles. Stage two applies HiCQA-Graph, a novel heterogeneous graph over images, captions, and QAs that fuses NLI-based textual entailment, CLIP-based vision-language alignment, and agent signals to identify and filter inconsistent samples. Stage three uses a MultiModal Large Language Model (MLLM) agent to generate multiple-choice questions (MCQ) followed by human screening. The resulting release comprises a large training split and a human-checked test split whose Bloom's level hard-sample distribution exceeds the MicroVQA benchmark. Our work delivers (i) a quality-controlled dataset that couples expert literature with graph-based filtering and human refinement; (ii) HiCQA-Graph, the first graph that jointly models (image, caption, QA) for cross-modal consistency filtering; (iii) evidence that careful data construction enables 4B-scale MLLMs to reach competitive microscopy reasoning performance (e.g., GPT-5) and achieve state-of-the-art performance among open-source MLLMs. Code and dataset will be released after the review process concludes.

Paperid: 3410, https://arxiv.org/pdf/2511.03178.pdf

Abstract:
Anticipating forthcoming surgical events is vital for real-time assistance in endonasal transsphenoidal pituitary surgery, where visibility is limited and workflow changes rapidly. Most visual question answering (VQA) systems reason on isolated frames with static vision language alignment, providing little support for forecasting next steps or instrument needs. Existing surgical VQA datasets likewise center on the current scene rather than the near future. We introduce PitVQA-Anticipation, the first VQA dataset designed for forward looking surgical reasoning. It comprises 33.5 hours of operative video and 734,769 question answer pairs built from temporally grouped clips and expert annotations across four tasks: predicting the future phase, next step, upcoming instrument, and remaining duration. We further propose SurgAnt-ViVQA, a video language model that adapts a large language model using a GRU Gated Temporal Cross-Attention module. A bidirectional GRU encodes frame to frame dynamics, while an adaptive gate injects visual context into the language stream at the token level. Parameter efficient fine tuning customizes the language backbone to the surgical domain. SurgAnt-ViVQA tested upon on PitVQA-Anticipation and EndoVis datasets, surpassing strong image and video based baselines. Ablations show that temporal recurrence and gated fusion drive most of the gains. A frame budget study indicates a trade-off: 8 frames maximize fluency, whereas 32 frames slightly reduce BLEU but improve numeric time estimation. By pairing a temporally aware encoder with fine grained gated cross-attention, SurgAnt-ViVQA advances surgical VQA from retrospective description to proactive anticipation. PitVQA-Anticipation offers a comprehensive benchmark for this setting and highlights the importance of targeted temporal modeling for reliable, future aware surgical assistance.

Paperid: 3411, https://arxiv.org/pdf/2511.01427.pdf

Abstract:
Single object tracking aims to localize target object with specific reference modalities (bounding box, natural language or both) in a sequence of specific video modalities (RGB, RGB+Depth, RGB+Thermal or RGB+Event.). Different reference modalities enable various human-machine interactions, and different video modalities are demanded in complex scenarios to enhance tracking robustness. Existing trackers are designed for single or several video modalities with single or several reference modalities, which leads to separate model designs and limits practical applications. Practically, a unified tracker is needed to handle various requirements. To the best of our knowledge, there is still no tracker that can perform tracking with these above reference modalities across these video modalities simultaneously. Thus, in this paper, we present a unified tracker, UniSOT, for different combinations of three reference modalities and four video modalities with uniform parameters. Extensive experimental results on 18 visual tracking, vision-language tracking and RGB+X tracking benchmarks demonstrate that UniSOT shows superior performance against modality-specific counterparts. Notably, UniSOT outperforms previous counterparts by over 3.0\% AUC on TNL2K across all three reference modalities and outperforms Un-Track by over 2.0\% main metric across all three RGB+X video modalities.

Paperid: 3412, https://arxiv.org/pdf/2510.24411.pdf

Abstract:
Computer-using agents powered by Vision-Language Models (VLMs) have demonstrated human-like capabilities in operating digital environments like mobile platforms. While these agents hold great promise for advancing digital automation, their potential for unsafe operations, such as system compromise and privacy leakage, is raising significant concerns. Detecting these safety concerns across the vast and complex operational space of mobile environments presents a formidable challenge that remains critically underexplored. To establish a foundation for mobile agent safety research, we introduce MobileRisk-Live, a dynamic sandbox environment accompanied by a safety detection benchmark comprising realistic trajectories with fine-grained annotations. Built upon this, we propose OS-Sentinel, a novel hybrid safety detection framework that synergistically combines a Formal Verifier for detecting explicit system-level violations with a VLM-based Contextual Judge for assessing contextual risks and agent actions. Experiments show that OS-Sentinel achieves 10%-30% improvements over existing approaches across multiple metrics. Further analysis provides critical insights that foster the development of safer and more reliable autonomous mobile agents.

Paperid: 3413, https://arxiv.org/pdf/2510.12225.pdf

Abstract:
Recent advances in vision-language models (VLMs) have made them highly effective at reasoning tasks. However, the principles underlying the construction of performant VL reasoning training datasets remain poorly understood. In this work, we introduce several data curation approaches and study their impacts on VL reasoning capabilities by carefully controlling training and evaluation setups. We analyze the effects of context (image and question pair) sources, implement targeted data interventions, and explore scaling up images, questions, and chain-of-thought (CoT) solutions. Our findings reveal that (a) context source strategies significantly affect VLM performance, (b) interventions such as auxiliary signals from image captions and the inclusion of text-only reasoning yield substantial gains, and (c) scaling all data dimensions (e.g., unique questions per image and unique CoTs per image-question pair) consistently improves reasoning capability. Motivated by these insights, we introduce HoneyBee, a large-scale, high-quality CoT reasoning dataset with 2.5M examples consisting 350K image-question pairs. VLMs trained with HoneyBee outperform state-of-the-art models across model sizes. For instance, a HoneyBee-trained VLM with 3B parameters outperforms the SOTA model and the base model by 7.8% and 24.8%, respectively, on MathVerse. Furthermore, we propose a test-time scaling strategy that reduces decoding cost by 73% without sacrificing accuracy. Overall, this work presents improved strategies for VL reasoning dataset curation research.

Paperid: 3414, https://arxiv.org/pdf/2510.08054.pdf

Abstract:
Image retouching not only enhances visual quality but also serves as a means of expressing personal preferences and emotions. However, existing learning-based approaches require large-scale paired data and operate as black boxes, making the retouching process opaque and limiting their adaptability to handle diverse, user- or image-specific adjustments. In this work, we propose RetouchLLM, a training-free white-box image retouching system, which requires no training data and performs interpretable, code-based retouching directly on high-resolution images. Our framework progressively enhances the image in a manner similar to how humans perform multi-step retouching, allowing exploration of diverse adjustment paths. It comprises of two main modules: a visual critic that identifies differences between the input and reference images, and a code generator that produces executable codes. Experiments demonstrate that our approach generalizes well across diverse retouching styles, while natural language-based user interaction enables interpretable and controllable adjustments tailored to user intent.

Paperid: 3415, https://arxiv.org/pdf/2510.07809.pdf

Abstract:
Large vision-language models (LVLMs) enable autonomous mobile agents to operate smartphone user interfaces, yet vulnerabilities to UI-level attacks remain critically understudied. Existing research often depends on conspicuous UI overlays, elevated permissions, or impractical threat models, limiting stealth and real-world applicability. In this paper, we present a practical and stealthy one-shot jailbreak attack that leverages in-app prompt injections: malicious applications embed short prompts in UI text that remain inert during human interaction but are revealed when an agent drives the UI via ADB (Android Debug Bridge). Our framework comprises three crucial components: (1) low-privilege perception-chain targeting, which injects payloads into malicious apps as the agent's visual inputs; (2) stealthy user-invisible activation, a touch-based trigger that discriminates agent from human touches using physical touch attributes and exposes the payload only during agent operation; and (3) one-shot prompt efficacy, a heuristic-guided, character-level iterative-deepening search algorithm (HG-IDA*) that performs one-shot, keyword-level detoxification to evade on-device safety filters. We evaluate across multiple LVLM backends, including closed-source services and representative open-source models within three Android applications, and we observe high planning and execution hijack rates in single-shot scenarios (e.g., GPT-4o: 82.5% planning / 75.0% execution). These findings expose a fundamental security vulnerability in current mobile agents with immediate implications for autonomous smartphone operation.

Paperid: 3416, https://arxiv.org/pdf/2510.02262.pdf

Abstract:
Video Large Language Models (VLMs) have achieved remarkable results on a variety of vision language tasks, yet their practical use is limited by the "needle in a haystack" problem: the massive number of visual tokens produced from raw video frames exhausts the model's context window. Existing solutions alleviate this issue by selecting a sparse set of frames, thereby reducing token count, but such frame-wise selection discards essential temporal dynamics, leading to suboptimal reasoning about motion and event continuity. In this work we systematically explore the impact of temporal information and demonstrate that extending selection from isolated key frames to key clips, which are short, temporally coherent segments, improves video understanding. To maintain a fixed computational budget while accommodating the larger token footprint of clips, we propose an adaptive resolution strategy that dynamically balances spatial resolution and clip length, ensuring a constant token count per video. Experiments on three long-form video benchmarks demonstrate that our training-free approach, F2C, outperforms uniform sampling up to 8.1%, 5.6%, and 10.3% on Video-MME, LongVideoBench and MLVU benchmarks, respectively. These results highlight the importance of preserving temporal coherence in frame selection and provide a practical pathway for scaling Video LLMs to real world video understanding applications. Project webpage is available at https://guangyusun.com/f2c .

Paperid: 3417, https://arxiv.org/pdf/2509.25946.pdf

Abstract:
Automated model discovery is the process of automatically searching and identifying the most appropriate model for a given dataset over a large combinatorial search space. Existing approaches, however, often face challenges in balancing the capture of fine-grained details with ensuring generalizability beyond training data regimes with a reasonable model complexity. In this paper, we present a multi-modal \& multi-step pipeline for effective automated model discovery. Our approach leverages two vision-language-based modules (VLM), AnalyzerVLM and EvaluatorVLM, for effective model proposal and evaluation in an agentic way. AnalyzerVLM autonomously plans and executes multi-step analyses to propose effective candidate models. EvaluatorVLM assesses the candidate models both quantitatively and perceptually, regarding the fitness for local details and the generalibility for overall trends. Our results demonstrate that our pipeline effectively discovers models that capture fine details and ensure strong generalizability. Additionally, extensive ablation studies show that both multi-modality and multi-step reasoning play crucial roles in discovering favorable models.

Paperid: 3418, https://arxiv.org/pdf/2509.21249.pdf

Abstract:
Magnetic Resonance Imaging (MRI) is a critical medical imaging modality in clinical diagnosis and research, yet its complexity and heterogeneity pose challenges for automated analysis, particularly in scalable and generalizable machine learning applications. While foundation models have revolutionized natural language and vision tasks, their application to MRI remains limited due to data scarcity and narrow anatomical focus. In this work, we present Decipher-MR, a 3D MRI-specific vision-language foundation model trained on a large-scale dataset comprising 200,000 MRI series from over 22,000 studies spanning diverse anatomical regions, sequences, and pathologies. Decipher-MR integrates self-supervised vision learning with report-guided text supervision to build robust, generalizable representations, enabling effective adaptation across broad applications. To enable robust and diverse clinical tasks with minimal computational overhead, Decipher-MR supports a modular design that enables tuning of lightweight, task-specific decoders attached to a frozen pretrained encoder. Following this setting, we evaluate Decipher-MR across diverse benchmarks including disease classification, demographic prediction, anatomical localization, and cross-modal retrieval, demonstrating consistent performance gains over existing foundation models and task-specific approaches. Our results establish Decipher-MR as a scalable and versatile foundation for MRI-based AI, facilitating efficient development across clinical and research domains.

Paperid: 3419, https://arxiv.org/pdf/2509.19480.pdf

Abstract:
Humans can flexibly interpret and compose different goal specifications, such as language instructions, spatial coordinates, or visual references, when navigating to a destination. In contrast, most existing robotic navigation policies are trained on a single modality, limiting their adaptability to real-world scenarios where different forms of goal specification are natural and complementary. In this work, we present a training framework for robotic foundation models that enables omni-modal goal conditioning for vision-based navigation. Our approach leverages a high-capacity vision-language-action (VLA) backbone and trains with three primary goal modalities: 2D poses, egocentric images, and natural language, as well as their combinations, through a randomized modality fusion strategy. This design not only expands the pool of usable datasets but also encourages the policy to develop richer geometric, semantic, and visual representations. The resulting model, OmniVLA, achieves strong generalization to unseen environments, robustness to scarce modalities, and the ability to follow novel natural language instructions. We demonstrate that OmniVLA outperforms specialist baselines across modalities and offers a flexible foundation for fine-tuning to new modalities and tasks. We believe OmniVLA provides a step toward broadly generalizable and flexible navigation policies, and a scalable path for building omni-modal robotic foundation models. We present videos showcasing OmniVLA performance and will release its checkpoints and training code on our project page.

Paperid: 3420, https://arxiv.org/pdf/2509.19191.pdf

Abstract:
Vision-Language Models (VLMs) have demonstrated remarkable performance across a variety of real-world tasks. However, existing VLMs typically process visual information by serializing images, a method that diverges significantly from the parallel nature of human vision. Moreover, their opaque internal mechanisms hinder both deeper understanding and architectural innovation. Inspired by the dual-stream hypothesis of human vision, which distinguishes the "what" and "where" pathways, we deconstruct the visual processing in VLMs into object recognition and spatial perception for separate study. For object recognition, we convert images into text token maps and find that the model's perception of image content unfolds as a two-stage process from shallow to deep layers, beginning with attribute recognition and culminating in semantic disambiguation. For spatial perception, we theoretically derive and empirically verify the geometric structure underlying the positional representation in VLMs. Based on these findings, we introduce an instruction-agnostic token compression algorithm based on a plug-and-play visual decoder to improve decoding efficiency, and a RoPE scaling technique to enhance spatial reasoning. Through rigorous experiments, our work validates these analyses, offering a deeper understanding of VLM internals and providing clear principles for designing more capable future architectures.

Paperid: 3421, https://arxiv.org/pdf/2509.15661.pdf

Abstract:
While large audio-language models (LALMs) have demonstrated state-of-the-art audio understanding, their reasoning capability in complex soundscapes still falls behind large vision-language models (LVLMs). Compared to the visual domain, one bottleneck is the lack of large-scale chain-of-thought audio data to teach LALM stepwise reasoning. To circumvent this data and modality gap, we present SightSound-R1, a cross-modal distillation framework that transfers advanced reasoning from a stronger LVLM teacher to a weaker LALM student on the same audio-visual question answering (AVQA) dataset. SightSound-R1 consists of three core steps: (i) test-time scaling to generate audio-focused chains of thought (CoT) from an LVLM teacher, (ii) audio-grounded validation to filter hallucinations, and (iii) a distillation pipeline with supervised fine-tuning (SFT) followed by Group Relative Policy Optimization (GRPO) for the LALM student. Results show that SightSound-R1 improves LALM reasoning performance both in the in-domain AVQA test set as well as in unseen auditory scenes and questions, outperforming both pretrained and label-only distilled baselines. Thus, we conclude that vision reasoning can be effectively transferred to audio models and scaled with abundant audio-visual data.

Paperid: 3422, https://arxiv.org/pdf/2509.14967.pdf

Abstract:
Effective human-robot collaboration in surgery is affected by the inherent ambiguity of verbal communication. This paper presents a framework for a robotic surgical assistant that interprets and disambiguates verbal instructions from a surgeon by grounding them in the visual context of the operating field. The system employs a two-level affordance-based reasoning process that first analyzes the surgical scene using a multimodal vision-language model and then reasons about the instruction using a knowledge base of tool capabilities. To ensure patient safety, a dual-set conformal prediction method is used to provide a statistically rigorous confidence measure for robot decisions, allowing it to identify and flag ambiguous commands. We evaluated our framework on a curated dataset of ambiguous surgical requests from cholecystectomy videos, demonstrating a general disambiguation rate of 60% and presenting a method for safer human-robot interaction in the operating room.

Paperid: 3423, https://arxiv.org/pdf/2509.06321.pdf

Abstract:
Multimodal Large Language Models (MLLMs) have shown exceptional capabilities in vision-language tasks. However, effectively integrating image segmentation into these models remains a significant challenge. In this work, we propose a novel text-as-mask paradigm that casts image segmentation as a text generation problem, eliminating the need for additional decoders and significantly simplifying the segmentation process. Our key innovation is semantic descriptors, a new textual representation of segmentation masks where each image patch is mapped to its corresponding text label. We first introduce image-wise semantic descriptors, a patch-aligned textual representation of segmentation masks that integrates naturally into the language modeling pipeline. To enhance efficiency, we introduce the Row-wise Run-Length Encoding (R-RLE), which compresses redundant text sequences, reducing the length of semantic descriptors by 74% and accelerating inference by $3\times$, without compromising performance. Building upon this, our initial framework Text4Seg achieves strong segmentation performance across a wide range of vision tasks. To further improve granularity and compactness, we propose box-wise semantic descriptors, which localizes regions of interest using bounding boxes and represents region masks via structured mask tokens called semantic bricks. This leads to our refined model, Text4Seg++, which formulates segmentation as a next-brick prediction task, combining precision, scalability, and generative efficiency. Comprehensive experiments on natural and remote sensing datasets show that Text4Seg++ consistently outperforms state-of-the-art models across diverse benchmarks without any task-specific fine-tuning, while remaining compatible with existing MLLM backbones. Our work highlights the effectiveness, scalability, and generalizability of text-driven image segmentation within the MLLM framework.

Paperid: 3424, https://arxiv.org/pdf/2509.06105.pdf

Abstract:
Accurate analysis of pathological images is essential for automated tumor diagnosis but remains challenging due to high structural similarity and subtle morphological variations in tissue images. Current vision-language (VL) models often struggle to capture the complex reasoning required for interpreting structured pathological reports. To address these limitations, we propose PathoHR-Bench, a novel benchmark designed to evaluate VL models' abilities in hierarchical semantic understanding and compositional reasoning within the pathology domain. Results of this benchmark reveal that existing VL models fail to effectively model intricate cross-modal relationships, hence limiting their applicability in clinical setting. To overcome this, we further introduce a pathology-specific VL training scheme that generates enhanced and perturbed samples for multimodal contrastive learning. Experimental evaluations demonstrate that our approach achieves state-of-the-art performance on PathoHR-Bench and six additional pathology datasets, highlighting its effectiveness in fine-grained pathology representation.

Paperid: 3425, https://arxiv.org/pdf/2509.01275.pdf

Abstract:
Open-vocabulary semantic segmentation (OVSS) conducts pixel-level classification via text-driven alignment, where the domain discrepancy between base category training and open-vocabulary inference poses challenges in discriminative modeling of latent unseen category. To address this challenge, existing vision-language model (VLM)-based approaches demonstrate commendable performance through pre-trained multi-modal representations. However, the fundamental mechanisms of latent semantic comprehension remain underexplored, making the bottleneck for OVSS. In this work, we initiate a probing experiment to explore distribution patterns and dynamics of latent semantics in VLMs under inductive learning paradigms. Building on these insights, we propose X-Agent, an innovative OVSS framework employing latent semantic-aware ``agent'' to orchestrate cross-modal attention mechanisms, simultaneously optimizing latent semantic dynamic and amplifying its perceptibility. Extensive benchmark evaluations demonstrate that X-Agent achieves state-of-the-art performance while effectively enhancing the latent semantic saliency.

Paperid: 3426, https://arxiv.org/pdf/2508.13446.pdf

Abstract:
Generalist robots should be able to understand and follow user instructions, but current vision-language-action (VLA) models struggle with following fine-grained commands despite providing a powerful architecture for mapping open-vocabulary natural language instructions to robot actions. One cause for this is a lack of semantic diversity and language grounding in existing robot datasets and, specifically, a lack of fine-grained task diversity for similar observations. To address this, we present a novel method to augment existing robot datasets by leveraging vision language models to create counterfactual labels. Our method improves the language-following capabilities of VLAs by increasing the diversity and granularity of language grounding for robot datasets by generating counterfactual language and actions. We evaluate the resulting model's ability to follow language instructions, ranging from simple object-centric commands to complex referential tasks, by conducting visual language navigation experiments in 3 different indoor and outdoor environments. Our experiments demonstrate that counterfactual relabeling, without any additional data collection, significantly improves instruction-following in VLA policies, making them competitive with state-of-the-art methods and increasing success rate by 27% on navigation tasks.

Paperid: 3427, https://arxiv.org/pdf/2508.12711.pdf

Abstract:
The proliferation of multimodal misinformation poses growing threats to public discourse and societal trust. While Large Vision-Language Models (LVLMs) have enabled recent progress in multimodal misinformation detection (MMD), the rise of generative AI (GenAI) tools introduces a new challenge: GenAI-driven news diversity, characterized by highly varied and complex content. We show that this diversity induces multi-level drift, comprising (1) model-level misperception drift, where stylistic variations disrupt a model's internal reasoning, and (2) evidence-level drift, where expression diversity degrades the quality or relevance of retrieved external evidence. These drifts significantly degrade the robustness of current LVLM-based MMD systems. To systematically study this problem, we introduce DriftBench, a large-scale benchmark comprising 16,000 news instances across six categories of diversification. We design three evaluation tasks: (1) robustness of truth verification under multi-level drift; (2) susceptibility to adversarial evidence contamination generated by GenAI; and (3) analysis of reasoning consistency across diverse inputs. Experiments with six state-of-the-art LVLM-based detectors show substantial performance drops (average F1 -14.8%) and increasingly unstable reasoning traces, with even more severe failures under adversarial evidence injection. Our findings uncover fundamental vulnerabilities in existing MMD systems and suggest an urgent need for more resilient approaches in the GenAI era.

Paperid: 3428, https://arxiv.org/pdf/2508.10180.pdf

Abstract:
Quantifying the influence of individual training samples is essential for enhancing the transparency and accountability of large language models (LLMs) and vision-language models (VLMs). However, existing data valuation methods often rely on Hessian information or model retraining, making them computationally prohibitive for billion-parameter models. In this work, we introduce For-Value, a forward-only data valuation framework that enables scalable and efficient influence estimation for both LLMs and VLMs. By leveraging the rich representations of modern foundation models, For-Value computes influence scores using a simple closed-form expression based solely on a single forward pass, thereby eliminating the need for costly gradient computations. Our theoretical analysis demonstrates that For-Value accurately estimates per-sample influence by capturing alignment in hidden representations and prediction errors between training and validation samples. Extensive experiments show that For-Value matches or outperforms gradient-based baselines in identifying impactful fine-tuning examples and effectively detecting mislabeled data.

Paperid: 3429, https://arxiv.org/pdf/2507.23064.pdf

Abstract:
Autonomous cars need geometric accuracy and semantic understanding to navigate complex environments, yet most stacks handle them separately. We present XYZ-Drive, a single vision-language model that reads a front-camera frame, a 25m $\times$ 25m overhead map, and the next waypoint, then outputs steering and speed. A lightweight goal-centered cross-attention layer lets waypoint tokens highlight relevant image and map patches, supporting both action and textual explanations, before the fused tokens enter a partially fine-tuned LLaMA-3.2 11B model. On the MD-NEX Outdoor-Driving benchmark XYZ-Drive attains 95% success and 0.80 Success weighted by Path Length (SPL), surpassing PhysNav-DG by 15%. and halving collisions, all while significantly improving efficiency by using only a single branch. Sixteen ablations explain the gains. Removing any modality (vision, waypoint, map) drops success by up to 11%, confirming their complementary roles and rich connections. Replacing goal-centered attention with simple concatenation cuts 3% in performance, showing query-based fusion injects map knowledge more effectively. Keeping the transformer frozen loses 5%, showing the importance of fine-tuning when applying VLMs for specific tasks such as autonomous driving. Coarsening map resolution from 10 cm to 40 cm blurs lane edges and raises crash rate. Overall, these results demonstrate that early, token-level fusion of intent and map layout enables accurate, transparent, real-time driving.

Paperid: 3430, https://arxiv.org/pdf/2507.23042.pdf

Abstract:
Autonomous vehicles must react in milliseconds while reasoning about road geometry and traffic intent to navigate complex situations. We introduce NovaDrive, a single-branch vision-language architecture that processes front-camera images, HD-map tiles, LiDAR depth, and textual waypoints in a single branch. A lightweight, two-stage cross-attention block first aligns waypoint tokens with the HD map, then refines attention over fine-grained image and depth patches. Coupled with a novel smoothness loss that discourages abrupt steering and speed changes, this design eliminates the need for recurrent memory. We fine-tune the top 15 layers of an 11B LLaMA-3.2 vision-language backbone, enabling real-time inference. On the nuScenes / Waymo subset of the MD-NEX Outdoor benchmark, NovaDrive raises success rate to 84% (+4%), boosts path-efficiency (SPL) to 0.66 (+0.11), and reduces collision frequency from 2.6% to 1.2% (-1.4%) relative to the previous state-of-the-art. Our ablations confirm that waypoint tokens, partial VLM fine-tuning, and the cross-attention fusion each contribute the most to these gains. Beyond safety, NovaDrive's shorter routes (resulting from the novel smoothness loss) translate to lower fuel or battery usage, pointing toward leaner, more easily updated driving stacks. NovaDrive can be extended to other embodied-AI domains as well.

Paperid: 3431, https://arxiv.org/pdf/2507.20529.pdf

Abstract:
The spatial reasoning task aims to reason about the spatial relationships in 2D and 3D space, which is a fundamental capability for Visual Question Answering (VQA) and robotics. Although vision language models (VLMs) have developed rapidly in recent years, they are still struggling with the spatial reasoning task. In this paper, we introduce a method that can enhance Spatial reasoning through Visual and Textual thinking Simultaneously (SpatialVTS). In the spatial visual thinking phase, our model is trained to generate location-related specific tokens of essential targets automatically. Not only are the objects mentioned in the problem addressed, but also the potential objects related to the reasoning are considered. During the spatial textual thinking phase, Our model conducts long-term thinking based on visual cues and dialogues, gradually inferring the answers to spatial reasoning problems. To effectively support the model's training, we perform manual corrections to the existing spatial reasoning dataset, eliminating numerous incorrect labels resulting from automatic annotation, restructuring the data input format to enhance generalization ability, and developing thinking processes with logical reasoning details. Without introducing additional information (such as masks or depth), our model's overall average level in several spatial understanding tasks has significantly improved compared with other models.

Paperid: 3432, https://arxiv.org/pdf/2507.07105.pdf

Abstract:
We present 4KAgent, a unified agentic super-resolution generalist system designed to universally upscale any image to 4K resolution (and even higher, if applied iteratively). Our system can transform images from extremely low resolutions with severe degradations, for example, highly distorted inputs at 256x256, into crystal-clear, photorealistic 4K outputs. 4KAgent comprises three core components: (1) Profiling, a module that customizes the 4KAgent pipeline based on bespoke use cases; (2) A Perception Agent, which leverages vision-language models alongside image quality assessment experts to analyze the input image and make a tailored restoration plan; and (3) A Restoration Agent, which executes the plan, following a recursive execution-reflection paradigm, guided by a quality-driven mixture-of-expert policy to select the optimal output for each step. Additionally, 4KAgent embeds a specialized face restoration pipeline, significantly enhancing facial details in portrait and selfie photos. We rigorously evaluate our 4KAgent across 11 distinct task categories encompassing a total of 26 diverse benchmarks, setting new state-of-the-art on a broad spectrum of imaging domains. Our evaluations cover natural images, portrait photos, AI-generated content, satellite imagery, fluorescence microscopy, and medical imaging like fundoscopy, ultrasound, and X-ray, demonstrating superior performance in terms of both perceptual (e.g., NIQE, MUSIQ) and fidelity (e.g., PSNR) metrics. By establishing a novel agentic paradigm for low-level vision tasks, we aim to catalyze broader interest and innovation within vision-centric autonomous agents across diverse research communities. We will release all the code, models, and results at: https://4kagent.github.io.

Paperid: 3433, https://arxiv.org/pdf/2507.06210.pdf

Abstract:
Pretrained vision-language models (VLMs) such as CLIP excel in general multimodal comprehension but often struggle to capture nuanced, context-dependent visual cues. This makes it difficult to distinguish between similar-looking concepts with potentially different cultural meanings. Such deficiencies are mainly due to a limited amount of high-quality cultural data, contextual information, and the lack of negative examples that highlight subtle differences. To mitigate this, we design a data curation pipeline leveraging open-sourced VLMs and text-to-image models to construct CulTwin, a synthetic cultural dataset. This dataset consists of paired concept-caption-image triplets, where concepts visually resemble each other but are culturally different. Then, we fine-tune CLIP on CulTwin to develop CultureCLIP, which aligns cultural concepts with contextually enhanced captions and synthetic images through tailored contrastive learning. Experiments on culture-specific benchmarks show that CultureCLIP outperforms the base CLIP, achieving up to a notable 5.49% improvement in fine-grained concept recognition on certain tasks while preserving CLIP's original generalization ability, validating the effectiveness of our data synthesis and VLM backbone training paradigm in capturing subtle cultural distinctions.

Paperid: 3434, https://arxiv.org/pdf/2506.16385.pdf

Abstract:
Micro-gesture recognition is a challenging task in affective computing due to the subtle, involuntary nature of the gestures and their low movement amplitude. In this paper, we introduce a Pose-Guided Semantics-Aware CLIP-based architecture, or CLIP for Micro-Gesture recognition (CLIP-MG), a modified CLIP model tailored for micro-gesture classification on the iMiGUE dataset. CLIP-MG integrates human pose (skeleton) information into the CLIP-based recognition pipeline through pose-guided semantic query generation and a gated multi-modal fusion mechanism. The proposed model achieves a Top-1 accuracy of 61.82%. These results demonstrate both the potential of our approach and the remaining difficulty in fully adapting vision-language models like CLIP for micro-gesture recognition.

Paperid: 3435, https://arxiv.org/pdf/2506.13187.pdf

Abstract:
Conventional low-rank adaptation methods build adapters without considering data context, leading to sub-optimal fine-tuning performance and severe forgetting of inherent world knowledge. In this paper, we propose context-oriented decomposition adaptation (CorDA), a novel method that initializes adapters in a task-aware manner. Concretely, we develop context-oriented singular value decomposition, where we collect covariance matrices of input activations for each linear layer using sampled data from the target task, and apply SVD to the product of weight matrix and its corresponding covariance matrix. By doing so, the task-specific capability is compacted into the principal components. Thanks to the task awareness, our method enables two optional adaptation modes, knowledge-preserved mode (KPM) and instruction-previewed mode (IPM), providing flexibility to choose between freezing the principal components to preserve their associated knowledge or adapting them to better learn a new task. We further develop CorDA++ by deriving a metric that reflects the compactness of task-specific principal components, and then introducing dynamic covariance selection and dynamic rank allocation strategies based on the same metric. The two strategies provide each layer with the most representative covariance matrix and a proper rank allocation. Experimental results show that CorDA++ outperforms CorDA by a significant margin. CorDA++ in KPM not only achieves better fine-tuning performance than LoRA, but also mitigates the forgetting of pre-trained knowledge in both large language models and vision language models. For IPM, our method exhibits faster convergence, \emph{e.g.,} 4.5x speedup over QLoRA, and improves adaptation performance in various scenarios, outperforming strong baseline methods. Our method has been integrated into the PEFT library developed by Hugging Face.

Paperid: 3436, https://arxiv.org/pdf/2506.12340.pdf

Abstract:
Large vision-language models (LVLMs) have demonstrated outstanding performance in many downstream tasks. However, LVLMs are trained on large-scale datasets, which can pose privacy risks if training images contain sensitive information. Therefore, it is important to detect whether an image is used to train the LVLM. Recent studies have investigated membership inference attacks (MIAs) against LVLMs, including detecting image-text pairs and single-modality content. In this work, we focus on detecting whether a target image is used to train the target LVLM. We design simple yet effective Image Corruption-Inspired Membership Inference Attacks (ICIMIA) against LLVLMs, which are inspired by LVLM's different sensitivity to image corruption for member and non-member images. We first perform an MIA method under the white-box setting, where we can obtain the embeddings of the image through the vision part of the target LVLM. The attacks are based on the embedding similarity between the image and its corrupted version. We further explore a more practical scenario where we have no knowledge about target LVLMs and we can only query the target LVLMs with an image and a question. We then conduct the attack by utilizing the output text embeddings' similarity. Experiments on existing datasets validate the effectiveness of our proposed attack methods under those two different settings.

Paperid: 3437, https://arxiv.org/pdf/2506.03933.pdf

Abstract:
Vision Language Models (VLMs) have shown remarkable capabilities in multimodal understanding, yet their susceptibility to perturbations poses a significant threat to their reliability in real-world applications. Despite often being imperceptible to humans, these perturbations can drastically alter model outputs, leading to erroneous interpretations and decisions. This paper introduces DiffCAP, a novel diffusion-based purification strategy that can effectively neutralize adversarial corruptions in VLMs. We observe that adding minimal noise to an adversarially corrupted image significantly alters its latent embedding with respect to VLMs. Building on this insight, DiffCAP cumulatively injects random Gaussian noise into adversarially perturbed input data. This process continues until the embeddings of two consecutive noisy images reach a predefined similarity threshold, indicating a potential approach to neutralize the adversarial effect. Subsequently, a pretrained diffusion model is employed to denoise the stabilized image, recovering a clean representation suitable for the VLMs to produce an output. Through extensive experiments across six datasets with three VLMs under varying attack strengths in three task scenarios, we show that DiffCAP consistently outperforms existing defense techniques by a substantial margin. Notably, DiffCAP significantly reduces both hyperparameter tuning complexity and the required diffusion time, thereby accelerating the denoising process. Equipped with strong theoretical and empirical support, DiffCAP provides a robust and practical solution for securely deploying VLMs in adversarial environments.

Paperid: 3438, https://arxiv.org/pdf/2506.03355.pdf

Abstract:
Adversarial input attacks can cause a significant shift of CLIP embeddings. This can affect the downstream robustness of models incorporating CLIP in the pipeline, such as text-to-image generative models or large vision language models. While some efforts have been done towards making the CLIP image encoders robust, the robustness of text encoders remains unexplored. In this work, we cover this gap in the literature. We propose LEAF: an efficient adversarial finetuning method for the text domain, with the ability to scale to large CLIP models. Our models significantly improve the zero-shot adversarial accuracy in the text domain, while maintaining the vision performance provided by robust image encoders. When combined with text-to-image diffusion models, we can improve the generation quality under adversarial noise. When employing our robust CLIP encoders in multimodal retrieval tasks, we improve the recall under adversarial noise over standard CLIP models. Finally, we show that robust text encoders facilitate better reconstruction of input text from its embedding via direct optimization.

Paperid: 3439, https://arxiv.org/pdf/2506.01185.pdf

Abstract:
We introduce HoMeR, an imitation learning framework for mobile manipulation that combines whole-body control with hybrid action modes that handle both long-range and fine-grained motion, enabling effective performance on realistic in-the-wild tasks. At its core is a fast, kinematics-based whole-body controller that maps desired end-effector poses to coordinated motion across the mobile base and arm. Within this reduced end-effector action space, HoMeR learns to switch between absolute pose predictions for long-range movement and relative pose predictions for fine-grained manipulation, offloading low-level coordination to the controller and focusing learning on task-level decisions. We deploy HoMeR on a holonomic mobile manipulator with a 7-DoF arm in a real home. We compare HoMeR to baselines without hybrid actions or whole-body control across 3 simulated and 3 real household tasks such as opening cabinets, sweeping trash, and rearranging pillows. Across tasks, HoMeR achieves an overall success rate of 79.17% using just 20 demonstrations per task, outperforming the next best baseline by 29.17 on average. HoMeR is also compatible with vision-language models and can leverage their internet-scale priors to better generalize to novel object appearances, layouts, and cluttered scenes. In summary, HoMeR moves beyond tabletop settings and demonstrates a scalable path toward sample-efficient, generalizable manipulation in everyday indoor spaces. Code, videos, and supplementary material are available at: http://homer-manip.github.io

Paperid: 3440, https://arxiv.org/pdf/2505.22050.pdf

Abstract:
Embodied planning requires agents to make coherent multi-step decisions based on dynamic visual observations and natural language goals. While recent vision-language models (VLMs) excel at static perception tasks, they struggle with the temporal reasoning, spatial understanding, and commonsense grounding needed for planning in interactive environments. In this work, we introduce a reinforcement fine-tuning framework that brings R1-style reasoning enhancement into embodied planning. We first distill a high-quality dataset from a powerful closed-source model and perform supervised fine-tuning (SFT) to equip the model with structured decision-making priors. We then design a rule-based reward function tailored to multi-step action quality and optimize the policy via Generalized Reinforced Preference Optimization (GRPO). Our approach is evaluated on Embench, a recent benchmark for interactive embodied tasks, covering both in-domain and out-of-domain scenarios. Experimental results show that our method significantly outperforms models of similar or larger scale, including GPT-4o-mini and 70B+ open-source baselines, and exhibits strong generalization to unseen environments. This work highlights the potential of reinforcement-driven reasoning to advance long-horizon planning in embodied AI.

Paperid: 3441, https://arxiv.org/pdf/2505.21844.pdf

Abstract:
Recently, test-time adaptation has attracted wide interest in the context of vision-language models for image classification. However, to the best of our knowledge, the problem is completely overlooked in dense prediction tasks such as Open-Vocabulary Semantic Segmentation (OVSS). In response, we propose a novel TTA method tailored to adapting VLMs for segmentation during test time. Unlike TTA methods for image classification, our Multi-Level and Multi-Prompt (MLMP) entropy minimization integrates features from intermediate vision-encoder layers and is performed with different text-prompt templates at both the global CLS token and local pixel-wise levels. Our approach could be used as plug-and-play for any segmentation network, does not require additional training data or labels, and remains effective even with a single test sample. Furthermore, we introduce a comprehensive OVSS TTA benchmark suite, which integrates a rigorous evaluation protocol, seven segmentation datasets, and 15 common corruptions, with a total of 82 distinct test scenarios, establishing a standardized and comprehensive testbed for future TTA research in open-vocabulary segmentation. Our experiments on this suite demonstrate that our segmentation-tailored method consistently delivers significant gains over direct adoption of TTA classification baselines.

Paperid: 3442, https://arxiv.org/pdf/2505.20236.pdf

Abstract:
Uncertainty quantification is essential for assessing the reliability and trustworthiness of modern AI systems. Among existing approaches, verbalized uncertainty, where models express their confidence through natural language, has emerged as a lightweight and interpretable solution in large language models (LLMs). However, its effectiveness in vision-language models (VLMs) remains insufficiently studied. In this work, we conduct a comprehensive evaluation of verbalized confidence in VLMs, spanning three model categories, four task domains, and three evaluation scenarios. Our results show that current VLMs often display notable miscalibration across diverse tasks and settings. Notably, visual reasoning models (i.e., thinking with images) consistently exhibit better calibration, suggesting that modality-specific reasoning is critical for reliable uncertainty estimation. To further address calibration challenges, we introduce Visual Confidence-Aware Prompting, a two-stage prompting strategy that improves confidence alignment in multimodal settings. Overall, our study highlights the inherent miscalibration in VLMs across modalities. More broadly, our findings underscore the fundamental importance of modality alignment and model faithfulness in advancing reliable multimodal systems.

Paperid: 3443, https://arxiv.org/pdf/2505.18503.pdf

Abstract:
Medical Large Vision-Language Models (Med-LVLMs) often exhibit suboptimal attention distribution on visual inputs, leading to hallucinated or inaccurate outputs. Existing mitigation methods primarily rely on inference-time interventions, which are limited in attention adaptation or require additional supervision. To address this, we propose A$^3$Tune, a novel fine-tuning framework for Automatic Attention Alignment Tuning. A$^3$Tune leverages zero-shot weak labels from SAM, refines them into prompt-aware labels using BioMedCLIP, and then selectively modifies visually-critical attention heads to improve alignment while minimizing interference. Additionally, we introduce a A$^3$MoE module, enabling adaptive parameter selection for attention tuning across diverse prompts and images. Extensive experiments on medical VQA and report generation benchmarks show that A$^3$Tune outperforms state-of-the-art baselines, achieving enhanced attention distributions and performance in Med-LVLMs.

Paperid: 3444, https://arxiv.org/pdf/2505.17214.pdf

Abstract:
Medical deep learning models depend heavily on domain-specific knowledge to perform well on knowledge-intensive clinical tasks. Prior work has primarily leveraged unimodal knowledge graphs, such as the Unified Medical Language System (UMLS), to enhance model performance. However, integrating multimodal medical knowledge graphs remains largely underexplored, mainly due to the lack of resources linking imaging data with clinical concepts. To address this gap, we propose MEDMKG, a Medical Multimodal Knowledge Graph that unifies visual and textual medical information through a multi-stage construction pipeline. MEDMKG fuses the rich multimodal data from MIMIC-CXR with the structured clinical knowledge from UMLS, utilizing both rule-based tools and large language models for accurate concept extraction and relationship modeling. To ensure graph quality and compactness, we introduce Neighbor-aware Filtering (NaF), a novel filtering algorithm tailored for multimodal knowledge graphs. We evaluate MEDMKG across three tasks under two experimental settings, benchmarking twenty-four baseline methods and four state-of-the-art vision-language backbones on six datasets. Results show that MEDMKG not only improves performance in downstream medical tasks but also offers a strong foundation for developing adaptive and robust strategies for multimodal knowledge integration in medical artificial intelligence.

Paperid: 3445, https://arxiv.org/pdf/2505.16839.pdf

Abstract:
Modern Vision-Language Models (VLMs) can solve a wide range of tasks requiring visual reasoning. In real-world scenarios, desirable properties for VLMs include fast inference and controllable generation (e.g., constraining outputs to adhere to a desired format). However, existing autoregressive (AR) VLMs like LLaVA struggle in these aspects. Discrete diffusion models (DMs) offer a promising alternative, enabling parallel decoding for faster inference and bidirectional context for controllable generation through text-infilling. While effective in language-only settings, DMs' potential for multimodal tasks is underexplored. We introduce LaViDa, a family of VLMs built on DMs. We build LaViDa by equipping DMs with a vision encoder and jointly fine-tune the combined parts for multimodal instruction following. To address challenges encountered, LaViDa incorporates novel techniques such as complementary masking for effective training, prefix KV cache for efficient inference, and timestep shifting for high-quality sampling. Experiments show that LaViDa achieves competitive or superior performance to AR VLMs on multi-modal benchmarks such as MMMU, while offering unique advantages of DMs, including flexible speed-quality tradeoff, controllability, and bidirectional reasoning. On COCO captioning, LaViDa surpasses Open-LLaVa-Next-8B by +4.1 CIDEr with 1.92x speedup. On bidirectional tasks, it achieves +59% improvement on Constrained Poem Completion. These results demonstrate LaViDa as a strong alternative to AR VLMs. Code and models will be released in the camera-ready version.

Paperid: 3446, https://arxiv.org/pdf/2505.15158.pdf

Abstract:
Recent advances have explored integrating large language models (LLMs) into end-to-end autonomous driving systems to enhance generalization and interpretability. However, most existing approaches are limited to either driving performance or vision-language reasoning, making it difficult to achieve both simultaneously. In this paper, we propose ALN-P3, a unified co-distillation framework that introduces cross-modal alignment between "fast" vision-based autonomous driving systems and "slow" language-driven reasoning modules. ALN-P3 incorporates three novel alignment mechanisms: Perception Alignment (P1A), Prediction Alignment (P2A), and Planning Alignment (P3A), which explicitly align visual tokens with corresponding linguistic outputs across the full perception, prediction, and planning stack. All alignment modules are applied only during training and incur no additional costs during inference. Extensive experiments on four challenging benchmarks-nuScenes, Nu-X, TOD3Cap, and nuScenes QA-demonstrate that ALN-P3 significantly improves both driving decisions and language reasoning, achieving state-of-the-art results.

Paperid: 3447, https://arxiv.org/pdf/2505.07868.pdf

Abstract:
Vision-and-Language Navigation (VLN) tasks agents with locating specific objects in unseen environments using natural language instructions and visual cues. Many existing VLN approaches typically follow an 'observe-and-reason' schema, that is, agents observe the environment and decide on the next action to take based on the visual observations of their surroundings. They often face challenges in long-horizon scenarios due to limitations in immediate observation and vision-language modality gaps. To overcome this, we present VISTA, a novel framework that employs an 'imagine-and-align' navigation strategy. Specifically, we leverage the generative prior of pre-trained diffusion models for dynamic visual imagination conditioned on both local observations and high-level language instructions. A Perceptual Alignment Filter module then grounds these goal imaginations against current observations, guiding an interpretable and structured reasoning process for action selection. Experiments show that VISTA sets new state-of-the-art results on Room-to-Room (R2R) and RoboTHOR benchmarks, e.g.,+3.6% increase in Success Rate on R2R. Extensive ablation analysis underscores the value of integrating forward-looking imagination, perceptual alignment, and structured reasoning for robust navigation in long-horizon environments.

Paperid: 3448, https://arxiv.org/pdf/2505.01881.pdf

Abstract:
Robust navigation in diverse environments and domains requires both accurate state estimation and transparent decision making. We present PhysNav-DG, a novel framework that integrates classical sensor fusion with the semantic power of vision-language models. Our dual-branch architecture predicts navigation actions from multi-sensor inputs while simultaneously generating detailed chain-of-thought explanations. A modified Adaptive Kalman Filter dynamically adjusts its noise parameters based on environmental context. It leverages several streams of raw sensor data along with semantic insights from models such as LLaMA 3.2 11B and BLIP-2. To evaluate our approach, we introduce the MD-NEX Benchmark, a novel multi-domain dataset that unifies indoor navigation, autonomous driving, and social navigation tasks with ground-truth actions and human-validated explanations. Extensive experiments and ablations show that PhysNav-DG improves navigation success rates by over 20% and achieves high efficiency, with explanations that are both highly grounded and clear. This work connects high-level semantic reasoning and geometric planning for safer and more trustworthy autonomous systems.

Paperid: 3449, https://arxiv.org/pdf/2504.16727.pdf

Abstract:
Large Vision Language Models (LVLMs) excel in various vision-language tasks. Yet, their robustness to visual variations in position, scale, orientation, and context that objects in natural scenes inevitably exhibit due to changes in viewpoint and environment remains largely underexplored. To bridge this gap, we introduce V$^2$R-Bench, a comprehensive benchmark framework for evaluating Visual Variation Robustness of LVLMs, which encompasses automated evaluation dataset generation and principled metrics for thorough robustness assessment. Through extensive evaluation on 21 LVLMs, we reveal a surprising vulnerability to visual variations, in which even advanced models that excel at complex vision-language tasks significantly underperform on simple tasks such as object recognition. Interestingly, these models exhibit a distinct visual position bias that contradicts theories of effective receptive fields, and demonstrate a human-like visual acuity threshold. To identify the source of these vulnerabilities, we present a systematic framework for component-level analysis, featuring a novel visualization approach for aligned visual features. Results show that these vulnerabilities stem from error accumulation in the pipeline architecture and inadequate multimodal alignment. Complementary experiments with synthetic data further demonstrate that these limitations are fundamentally architectural deficiencies, scoring the need for architectural innovations in future LVLM designs.

Paperid: 3450, https://arxiv.org/pdf/2503.07125.pdf

Abstract:
Estimating the 3D world from 2D monocular images is a fundamental yet challenging task due to the labour-intensive nature of 3D annotations. To simplify label acquisition, this work proposes a novel approach that bridges 2D vision foundation models (VFMs) with 3D tasks by decoupling 3D supervision into an ensemble of image-level primitives, e.g., semantic and geometric components. As a key motivator, we leverage the zero-shot capabilities of vision-language models for image semantics. However, due to the notorious ill-posed problem - multiple distinct 3D scenes can produce identical 2D projections, directly inferring metric depth from a monocular image in a zero-shot manner is unsuitable. In contrast, 2D VFMs provide promising sources of relative depth, which theoretically aligns with metric depth when properly scaled and offset. Thus, we adapt the relative depth derived from VFMs into metric depth by optimising the scale and offset using temporal consistency, also known as novel view synthesis, without access to ground-truth metric depth. Consequently, we project the semantics into 3D space using the reconstructed metric depth, thereby providing 3D supervision. Extensive experiments on nuScenes and SemanticKITTI demonstrate the effectiveness of our framework. For instance, the proposed method surpasses the current state-of-the-art by 3.34% mIoU on nuScenes for voxel occupancy prediction.

Paperid: 3451, https://arxiv.org/pdf/2503.02157.pdf

Abstract:
Large Vision Language Models (LVLMs) are becoming increasingly important in the medical domain, yet Medical LVLMs (Med-LVLMs) frequently generate hallucinations due to limited expertise and the complexity of medical applications. Existing benchmarks fail to effectively evaluate hallucinations based on their underlying causes and lack assessments of mitigation strategies. To address this gap, we introduce MedHEval, a novel benchmark that systematically evaluates hallucinations and mitigation strategies in Med-LVLMs by categorizing them into three underlying causes: visual misinterpretation, knowledge deficiency, and context misalignment. We construct a diverse set of close- and open-ended medical VQA datasets with comprehensive evaluation metrics to assess these hallucination types. We conduct extensive experiments across 11 popular (Med)-LVLMs and evaluate 7 state-of-the-art hallucination mitigation techniques. Results reveal that Med-LVLMs struggle with hallucinations arising from different causes while existing mitigation methods show limited effectiveness, especially for knowledge- and context-based errors. These findings underscore the need for improved alignment training and specialized mitigation strategies to enhance Med-LVLMs' reliability. MedHEval establishes a standardized framework for evaluating and mitigating medical hallucinations, guiding the development of more trustworthy Med-LVLMs.

Paperid: 3452, https://arxiv.org/pdf/2503.01879.pdf

Abstract:
This work proposes an industry-level omni-modal large language model (LLM) pipeline that integrates auditory, visual, and linguistic modalities to overcome challenges such as limited tri-modal datasets, high computational costs, and complex feature alignments. Our pipeline consists of three main components: First, a modular framework enabling flexible configuration of various encoder-LLM-decoder architectures. Second, a lightweight training strategy that pre-trains audio-language alignment on the state-of-the-art vision-language model Qwen2.5-VL, thus avoiding the costly pre-training of vision-specific modalities. Third, an audio synthesis pipeline that generates high-quality audio-text data from diverse real-world scenarios, supporting applications such as Automatic Speech Recognition and Speech-to-Speech chat. To this end, we introduce an industry-level omni-modal LLM, Nexus. Extensive experiments validate the efficacy of our pipeline, yielding the following key findings:(1) In the visual understanding task, Nexus exhibits superior performance compared with its backbone model - Qwen2.5-VL-7B, validating the efficiency of our training strategy. (2) Within the English Spoken Question-Answering task, the model achieves better accuracy than the same-period competitor (i.e, MiniCPM-o2.6-7B) in the LLaMA Q. benchmark. (3) In our real-world ASR testset, Nexus achieves outstanding performance, indicating its robustness in real scenarios. (4) In the Speech-to-Text Translation task, our model outperforms Qwen2-Audio-Instruct-7B. (5) In the Text-to-Speech task, based on pretrained vocoder (e.g., Fishspeech1.4 or CosyVoice2.0), Nexus is comparable to its backbone vocoder on Seed-TTS benchmark. (6) An in-depth analysis of tri-modal alignment reveals that incorporating the audio modality enhances representational alignment between vision and language.

Paperid: 3453, https://arxiv.org/pdf/2502.15563.pdf

Abstract:
Reliable evaluation of AI models is critical for scientific progress and practical application. While existing VLM benchmarks provide general insights into model capabilities, their heterogeneous designs and limited focus on a few imaging domains pose significant challenges for both cross-domain performance comparison and targeted domain-specific evaluation. To address this, we propose three key contributions: (1) a framework for the resource-efficient creation of domain-specific VLM benchmarks enabled by task augmentation for creating multiple diverse tasks from a single existing task, (2) the release of new VLM benchmarks for seven domains, created according to the same homogeneous protocol and including 162,946 thoroughly human-validated answers, and (3) an extensive benchmarking of 22 state-of-the-art VLMs on a total of 37,171 tasks, revealing performance variances across domains and tasks, thereby supporting the need for tailored VLM benchmarks. Adoption of our methodology will pave the way for the resource-efficient domain-specific selection of models and guide future research efforts toward addressing core open questions.

Paperid: 3454, https://arxiv.org/pdf/2502.12359.pdf

Abstract:
Large Vision-Language Models (LVLMs) have shown impressive performance in various tasks. However, LVLMs suffer from hallucination, which hinders their adoption in the real world. Existing studies emphasized that the strong language priors of LVLMs can overpower visual information, causing hallucinations. However, the positive role of language priors is the key to a powerful LVLM. If the language priors are too weak, LVLMs will struggle to leverage rich parameter knowledge and instruction understanding abilities to complete tasks in challenging visual scenarios where visual information alone is insufficient. Therefore, we propose a benchmark called LanP to rethink the impact of Language Priors in LVLMs. It is designed to investigate how strong language priors are in current LVLMs. LanP consists of 170 images and 340 corresponding well-designed questions. Extensive experiments on 25 popular LVLMs reveal that many LVLMs' language priors are not strong enough to effectively aid question answering when objects are partially hidden. Many models, including GPT-4 Turbo, exhibit an accuracy below 0.5 in such a scenario.

Paperid: 3455, https://arxiv.org/pdf/2502.11918.pdf

Abstract:
Reward engineering is one of the key challenges in Reinforcement Learning (RL). Preference-based RL effectively addresses this issue by learning from human feedback. However, it is both time-consuming and expensive to collect human preference labels. In this paper, we propose a novel \textbf{V}ision-\textbf{L}anguage \textbf{P}reference learning framework, named \textbf{VLP}, which learns a vision-language preference model to provide preference feedback for embodied manipulation tasks. To achieve this, we define three types of language-conditioned preferences and construct a vision-language preference dataset, which contains versatile implicit preference orders without human annotations. The preference model learns to extract language-related features, and then serves as a preference annotator in various downstream tasks. The policy can be learned according to the annotated preferences via reward learning or direct policy optimization. Extensive empirical results on simulated embodied manipulation tasks demonstrate that our method provides accurate preferences and generalizes to unseen tasks and unseen language instructions, outperforming the baselines by a large margin.

Paperid: 3456, https://arxiv.org/pdf/2501.08490.pdf

Abstract:
Remote sensing imagery is dense with objects and contextual visual information. There is a recent trend to combine paired satellite images and text captions for pretraining performant encoders for downstream tasks. However, while contrastive image-text methods like CLIP enable vision-language alignment and zero-shot classification ability, vision-only downstream performance tends to degrade compared to image-only pretraining, such as MAE. In this paper, we propose FLAVARS, a pretraining method that combines the best of both contrastive learning and masked modeling, along with geospatial alignment via contrastive location encoding. We find that FLAVARS significantly outperforms a baseline of SkyCLIP for vision-only tasks such as KNN classification and semantic segmentation, +6\% mIOU on SpaceNet1, while retaining the ability to perform zero-shot classification, unlike MAE pretrained methods.

Paperid: 3457, https://arxiv.org/pdf/2412.02617.pdf

Abstract:
Large text-to-video models hold immense potential for a wide range of downstream applications. However, these models struggle to accurately depict dynamic object interactions, often resulting in unrealistic movements and frequent violations of real-world physics. One solution inspired by large language models is to align generated outputs with desired outcomes using external feedback. This enables the model to refine its responses autonomously, eliminating extensive manual data collection. In this work, we investigate the use of feedback to enhance the object dynamics in text-to-video models. We aim to answer a critical question: what types of feedback, paired with which specific self-improvement algorithms, can most effectively improve text-video alignment and realistic object interactions? We begin by deriving a unified probabilistic objective for offline RL finetuning of text-to-video models. This perspective highlights how design elements in existing algorithms like KL regularization and policy projection emerge as specific choices within a unified framework. We then use derived methods to optimize a set of text-video alignment metrics (e.g., CLIP scores, optical flow), but notice that they often fail to align with human perceptions of generation quality. To address this limitation, we propose leveraging vision-language models to provide more nuanced feedback specifically tailored to object dynamics in videos. Our experiments demonstrate that our method can effectively optimize a wide variety of rewards, with binary AI feedback driving the most significant improvements in video quality for dynamic interactions, as confirmed by both AI and human evaluations. Notably, we observe substantial gains when using reward signals derived from AI feedback, particularly in scenarios involving complex interactions between multiple objects and realistic depictions of objects falling.

Paperid: 3458, https://arxiv.org/pdf/2411.18212.pdf

Abstract:
Path planning under wireless performance constraints is a complex challenge in robot navigation. However, naively incorporating such constraints into classical planning algorithms often incurs prohibitive search costs. In this paper, we propose SCoTT, a wireless-aware path planning framework that leverages vision-language models (VLMs) to co-optimize average path gains and trajectory length using wireless heatmap images and ray-tracing data from a digital twin (DT). At the core of our framework is Strategic Chain-of-Thought Tasking (SCoTT), a novel prompting paradigm that decomposes the exhaustive search problem into structured subtasks, each solved via chain-of-thought prompting. To establish strong baselines, we compare classical A* and wireless-aware extensions of it, and derive DP-WA*, an optimal, iterative dynamic programming algorithm that incorporates all path gains and distance metrics from the DT, but at significant computational cost. In extensive experiments, we show that SCoTT achieves path gains within 2% of DP-WA* while consistently generating shorter trajectories. Moreover, SCoTT's intermediate outputs can be used to accelerate DP-WA* by reducing its search space, saving up to 62% in execution time. We validate our framework using four VLMs, demonstrating effectiveness across both large and small models, thus making it applicable to a wide range of compact models at low inference cost. We also show the practical viability of our approach by deploying SCoTT as a ROS node within Gazebo simulations. Finally, we discuss data acquisition pipelines, compute requirements, and deployment considerations for VLMs in 6G-enabled DTs, underscoring the potential of natural language interfaces for wireless-aware navigation in real-world applications.

Paperid: 3459, https://arxiv.org/pdf/2410.15281.pdf

Abstract:
With the broader adoption and highly successful development of Large Language Models (LLMs), there has been growing interest and demand for applying LLMs to autonomous driving technology. Driven by their natural language understanding and reasoning capabilities, LLMs have the potential to enhance various aspects of autonomous driving systems, from perception and scene understanding to interactive decision-making. In this paper, we first introduce the novel concept of designing Large Language Models for Autonomous Driving (LLM4AD), followed by a review of existing LLM4AD studies. Then, we propose a comprehensive benchmark for evaluating the instruction-following and reasoning abilities of LLM4AD systems, which includes LaMPilot-Bench, CARLA Leaderboard 1.0 Benchmark in simulation and NuPlanQA for multi-view visual question answering. Furthermore, we conduct extensive real-world experiments on autonomous vehicle platforms, examining both on-cloud and on-edge LLM deployment for personalized decision-making and motion control. Next, we explore the future trends of integrating language diffusion models into autonomous driving, exemplified by the proposed ViLaD (Vision-Language Diffusion) framework. Finally, we discuss the main challenges of LLM4AD, including latency, deployment, security and privacy, safety, trust and transparency, and personalization.

Paperid: 3460, https://arxiv.org/pdf/2410.04640.pdf

Abstract:
Robot behavior policies trained via imitation learning are prone to failure under conditions that deviate from their training data. Thus, algorithms that monitor learned policies at test time and provide early warnings of failure are necessary to facilitate scalable deployment. We propose Sentinel, a runtime monitoring framework that splits the detection of failures into two complementary categories: 1) Erratic failures, which we detect using statistical measures of temporal action consistency, and 2) task progression failures, where we use Vision Language Models (VLMs) to detect when the policy confidently and consistently takes actions that do not solve the task. Our approach has two key strengths. First, because learned policies exhibit diverse failure modes, combining complementary detectors leads to significantly higher accuracy at failure detection. Second, using a statistical temporal action consistency measure ensures that we quickly detect when multimodal, generative policies exhibit erratic behavior at negligible computational cost. In contrast, we only use VLMs to detect failure modes that are less time-sensitive. We demonstrate our approach in the context of diffusion policies trained on robotic mobile manipulation domains in both simulation and the real world. By unifying temporal consistency detection and VLM runtime monitoring, Sentinel detects 18% more failures than using either of the two detectors alone and significantly outperforms baselines, thus highlighting the importance of assigning specialized detectors to complementary categories of failure. Qualitative results are made available at https://sites.google.com/stanford.edu/sentinel.

Paperid: 3461, https://arxiv.org/pdf/2410.04055.pdf

Abstract:
While Vision-Language Models (VLMs) have shown remarkable abilities in visual and language reasoning tasks, they invariably generate flawed responses. Self-correction that instructs models to refine their outputs presents a promising solution to this issue. Previous studies have mainly concentrated on Large Language Models (LLMs), while the self-correction abilities of VLMs, particularly concerning both visual and linguistic information, remain largely unexamined. This study investigates the self-correction capabilities of VLMs during both inference and fine-tuning stages. We introduce a Self-Correction Learning (SCL) approach that enables VLMs to learn from their self-generated self-correction data through Direct Preference Optimization (DPO) without relying on external feedback, facilitating self-improvement. Specifically, we collect preferred and disfavored samples based on the correctness of initial and refined responses, which are obtained by two-turn self-correction with VLMs during the inference stage. Experimental results demonstrate that although VLMs struggle to self-correct effectively during iterative inference without additional fine-tuning and external feedback, they can enhance their performance and avoid previous mistakes through preference fine-tuning when their self-generated self-correction data are categorized into preferred and disfavored samples. This study emphasizes that self-correction is not merely a refinement process; rather, it should enhance the reasoning abilities of models through additional training, enabling them to generate high-quality responses directly without further refinement.

Paperid: 3462, https://arxiv.org/pdf/2410.03603.pdf

Abstract:
The world is filled with a wide variety of objects. For robots to be useful, they need the ability to find arbitrary objects described by people. In this paper, we present LeLaN(Learning Language-conditioned Navigation policy), a novel approach that consumes unlabeled, action-free egocentric data to learn scalable, language-conditioned object navigation. Our framework, LeLaN leverages the semantic knowledge of large vision-language models, as well as robotic foundation models, to label in-the-wild data from a variety of indoor and outdoor environments. We label over 130 hours of data collected in real-world indoor and outdoor environments, including robot observations, YouTube video tours, and human walking data. Extensive experiments with over 1000 real-world trials show that our approach enables training a policy from unlabeled action-free videos that outperforms state-of-the-art robot navigation methods, while being capable of inference at 4 times their speed on edge compute. We open-source our models, datasets and provide supplementary videos on our project page (https://learning-language-navigation.github.io/).

Paperid: 3463, https://arxiv.org/pdf/2410.03176.pdf

Abstract:
Large Vision-Language Models (LVLMs) have achieved impressive performance, yet research has pointed out a serious issue with object hallucinations within these models. However, there is no clear conclusion as to which part of the model these hallucinations originate from. In this paper, we present an in-depth investigation into the object hallucination problem specifically within the CLIP model, which serves as the backbone for many state-of-the-art vision-language systems. We unveil that even in isolation, the CLIP model is prone to object hallucinations, suggesting that the hallucination problem is not solely due to the interaction between vision and language modalities. To address this, we propose a counterfactual data augmentation method by creating negative samples with a variety of hallucination issues. We demonstrate that our method can effectively mitigate object hallucinations for CLIP model, and we show the the enhanced model can be employed as a visual encoder, effectively alleviating the object hallucination issue in LVLMs.

Paperid: 3464, https://arxiv.org/pdf/2410.01926.pdf

Abstract:
Reconstructing past events requires reasoning across long time horizons. To figure out what happened, we need to use our prior knowledge about the world and human behavior and draw inferences from various sources of evidence including visual, language, and auditory cues. We introduce MARPLE, a benchmark for evaluating long-horizon inference capabilities using multi-modal evidence. Our benchmark features agents interacting with simulated households, supporting vision, language, and auditory stimuli, as well as procedurally generated environments and agent behaviors. Inspired by classic ``whodunit'' stories, we ask AI models and human participants to infer which agent caused a change in the environment based on a step-by-step replay of what actually happened. The goal is to correctly identify the culprit as early as possible. Our findings show that human participants outperform both traditional Monte Carlo simulation methods and an LLM baseline (GPT-4) on this task. Compared to humans, traditional inference models are less robust and performant, while GPT-4 has difficulty comprehending environmental changes. We analyze what factors influence inference performance and ablate different modes of evidence, finding that all modes are valuable for performance. Overall, our experiments demonstrate that the long-horizon, multimodal inference tasks in our benchmark present a challenge to current models.

Paperid: 3465, https://arxiv.org/pdf/2410.00296.pdf

Abstract:
Vision-language models (VLMs) are essential for contextual understanding of both visual and textual information. However, their vulnerability to adversarially manipulated inputs presents significant risks, leading to compromised outputs and raising concerns about the reliability in VLM-integrated applications. Detecting these malicious prompts is thus crucial for maintaining trust in VLM generations. A major challenge in developing a safeguarding prompt classifier is the lack of a large amount of labeled benign and malicious data. To address the issue, we introduce VLMGuard, a novel learning framework that leverages the unlabeled user prompts in the wild for malicious prompt detection. These unlabeled prompts, which naturally arise when VLMs are deployed in the open world, consist of both benign and malicious information. To harness the unlabeled data, we present an automated maliciousness estimation score for distinguishing between benign and malicious samples within this unlabeled mixture, thereby enabling the training of a binary prompt classifier on top. Notably, our framework does not require extra human annotations, offering strong flexibility and practicality for real-world applications. Extensive experiment shows VLMGuard achieves superior detection results, significantly outperforming state-of-the-art methods. Disclaimer: This paper may contain offensive examples; reader discretion is advised.

Paperid: 3466, https://arxiv.org/pdf/2409.14759.pdf

Abstract:
Vision language models (VLMs) have shown promising reasoning capabilities across various benchmarks; however, our understanding of their visual perception remains limited. In this work, we propose an eye examination process to investigate how a VLM perceives images, specifically focusing on key elements of visual recognition, from primitive color and shape to semantic levels. To this end, we introduce a dataset named LENS to guide a VLM to follow the examination and check its readiness. Once the model is ready, we conduct the examination. Through this examination, we quantify and visualize VLMs' sensitivities to color and shape, and semantic matching. Our findings reveal that VLMs have varying sensitivity to different colors while consistently showing insensitivity to green across different VLMs. Also, we found different shape sensitivity and semantic recognition depending on LLM's capacity despite using the same fixed visual encoder. Our analyses and findings have potential to inspire the design of VLMs and the pre-processing of visual input to VLMs for improving application performance.

Paperid: 3467, https://arxiv.org/pdf/2409.02084.pdf

Abstract:
The ability for robots to perform efficient and zero-shot grasping of object parts is crucial for practical applications and is becoming prevalent with recent advances in Vision-Language Models (VLMs). To bridge the 2D-to-3D gap for representations to support such a capability, existing methods rely on neural fields (NeRFs) via differentiable rendering or point-based projection methods. However, we demonstrate that NeRFs are inappropriate for scene changes due to their implicitness and point-based methods are inaccurate for part localization without rendering-based optimization. To amend these issues, we propose GraspSplats. Using depth supervision and a novel reference feature computation method, GraspSplats generates high-quality scene representations in under 60 seconds. We further validate the advantages of Gaussian-based representation by showing that the explicit and optimized geometry in GraspSplats is sufficient to natively support (1) real-time grasp sampling and (2) dynamic and articulated object manipulation with point trackers. With extensive experiments on a Franka robot, we demonstrate that GraspSplats significantly outperforms existing methods under diverse task settings. In particular, GraspSplats outperforms NeRF-based methods like F3RM and LERF-TOGO, and 2D detection methods.

Paperid: 3468, https://arxiv.org/pdf/2601.09851.pdf

Abstract:
Multimodal video captioning condenses dense footage into a structured format of keyframes and natural language. By creating a cohesive multimodal summary, this approach anchors generative AI in rich semantic evidence and serves as a lightweight proxy for high-efficiency retrieval. However, traditional metrics like BLEU or ROUGE fail to quantify information coverage across disparate modalities, such as comparing a paragraph of text to a sequence of keyframes. To address this, we propose the Video Summary Information Loss (ViSIL) score, an information-theoretic framework that quantifies the video information not captured by a summary via vision-language model (VLM) inference. By measuring the information loss, ViSIL is a unified metric that enables direct comparison across multimodal summary formats despite their structural discrepancies. Our results demonstrate that ViSIL scores show a statistically significant correlation with both human and VLM performance on Video Question Answering (VQA) tasks. ViSIL also enables summary selection to optimize the trade-off between information loss and processing speed, establishing a Pareto-optimal frontier that outperforms text summaries by $7\%$ in VQA accuracy without increasing processing load.

Paperid: 3469, https://arxiv.org/pdf/2512.24119.pdf

Abstract:
Geometric problem solving constitutes a critical branch of mathematical reasoning, requiring precise analysis of shapes and spatial relationships. Current evaluations of geometric reasoning in vision-language models (VLMs) face limitations, including the risk of test data contamination from textbook-based benchmarks, overemphasis on final answers over reasoning processes, and insufficient diagnostic granularity. To address these issues, we present GeoBench, a hierarchical benchmark featuring four reasoning levels in geometric problem-solving: Visual Perception, Goal-Oriented Planning, Rigorous Theorem Application, and Self-Reflective Backtracking. Through six formally verified tasks generated via TrustGeoGen, we systematically assess capabilities ranging from attribute extraction to logical error correction. Experiments reveal that while reasoning models like OpenAI-o3 outperform general MLLMs, performance declines significantly with increasing task complexity. Key findings demonstrate that sub-goal decomposition and irrelevant premise filtering critically influence final problem-solving accuracy, whereas Chain-of-Thought prompting unexpectedly degrades performance in some tasks. These findings establish GeoBench as a comprehensive benchmark while offering actionable guidelines for developing geometric problem-solving systems.

Paperid: 3470, https://arxiv.org/pdf/2512.23545.pdf

Abstract:
Recent pathological foundation models have substantially advanced visual representation learning and multimodal interaction. However, most models still rely on a static inference paradigm in which whole-slide images are processed once to produce predictions, without reassessment or targeted evidence acquisition under ambiguous diagnoses. This contrasts with clinical diagnostic workflows that refine hypotheses through repeated slide observations and further examination requests. We propose PathFound, an agentic multimodal model designed to support evidence-seeking inference in pathological diagnosis. PathFound integrates the power of pathological visual foundation models, vision-language models, and reasoning models trained with reinforcement learning to perform proactive information acquisition and diagnosis refinement by progressing through the initial diagnosis, evidence-seeking, and final decision stages. Across several large multimodal models, adopting this strategy consistently improves diagnostic accuracy, indicating the effectiveness of evidence-seeking workflows in computational pathology. Among these models, PathFound achieves state-of-the-art diagnostic performance across diverse clinical scenarios and demonstrates strong potential to discover subtle details, such as nuclear features and local invasions.

Paperid: 3471, https://arxiv.org/pdf/2512.21135.pdf

Abstract:
Text-guided medical segmentation enhances segmentation accuracy by utilizing clinical reports as auxiliary information. However, existing methods typically rely on unaligned image and text encoders, which necessitate complex interaction modules for multimodal fusion. While CLIP provides a pre-aligned multimodal feature space, its direct application to medical imaging is limited by three main issues: insufficient preservation of fine-grained anatomical structures, inadequate modeling of complex clinical descriptions, and domain-specific semantic misalignment. To tackle these challenges, we propose TGC-Net, a CLIP-based framework focusing on parameter-efficient, task-specific adaptations. Specifically, it incorporates a Semantic-Structural Synergy Encoder (SSE) that augments CLIP's ViT with a CNN branch for multi-scale structural refinement, a Domain-Augmented Text Encoder (DATE) that injects large-language-model-derived medical knowledge, and a Vision-Language Calibration Module (VLCM) that refines cross-modal correspondence in a unified feature space. Experiments on five datasets across chest X-ray and thoracic CT modalities demonstrate that TGC-Net achieves state-of-the-art performance with substantially fewer trainable parameters, including notable Dice gains on challenging benchmarks.

Paperid: 3472, https://arxiv.org/pdf/2512.04238.pdf

Abstract:
Vision-language models are increasingly integrated into clinical workflows. However, existing benchmarks primarily assess performance on common anatomical presentations and fail to capture the challenges posed by rare variants. To address this gap, we introduce AdversarialAnatomyBench, the first benchmark comprising naturally occurring rare anatomical variants across diverse imaging modalities and anatomical regions. We call such variants that violate learned priors about "typical" human anatomy natural adversarial anatomy. Benchmarking 22 state-of-the-art VLMs with AdversarialAnatomyBench yielded three key insights. First, when queried with basic medical perception tasks, mean accuracy dropped from 74% on typical to 29% on atypical anatomy. Even the best-performing models, GPT-5, Gemini 2.5 Pro, and Llama 4 Maverick, showed performance drops of 41-51%. Second, model errors closely mirrored expected anatomical biases. Third, neither model scaling nor interventions, including bias-aware prompting and test-time reasoning, resolved these issues. These findings highlight a critical and previously unquantified limitation in current VLM: their poor generalization to rare anatomical presentations. AdversarialAnatomyBench provides a foundation for systematically measuring and mitigating anatomical bias in multimodal medical AI systems.

Paperid: 3473, https://arxiv.org/pdf/2511.19995.pdf

Abstract:
Creativity is a complex phenomenon. When it comes to representing and assessing creativity, treating it as a single undifferentiated quantity would appear naive and underwhelming. In this work, we learn the \emph{first type-specific creativity reward model}, coined CREward, which spans three creativity ``axes," geometry, material, and texture, to allow us to view creativity through the lens of the image formation pipeline. To build our reward model, we first conduct a human benchmark evaluation to capture human perception of creativity for each type across various creative images. We then analyze the correlation between human judgments and predictions by large vision-language models (LVLMs), confirming that LVLMs exhibit strong alignment with human perception. Building on this observation, we collect LVLM-generated labels to train our CREward model that is applicable to both evaluation and generation of creative images. We explore three applications of CREward: creativity assessment, explainable creativity, and creative sample acquisition for both human design inspiration and guiding creative generation through low-rank adaptation.

Paperid: 3474, https://arxiv.org/pdf/2511.19575.pdf

Abstract:
This paper presents HunyuanOCR, a commercial-grade, open-source, and lightweight (1B parameters) Vision-Language Model (VLM) dedicated to OCR tasks. The architecture comprises a Native Vision Transformer (ViT) and a lightweight LLM connected via an MLP adapter. HunyuanOCR demonstrates superior performance, outperforming commercial APIs, traditional pipelines, and larger models (e.g., Qwen3-VL-4B). Specifically, it surpasses current public solutions in perception tasks (Text Spotting, Parsing) and excels in semantic tasks (IE, Text Image Translation), securing first place in the ICDAR 2025 DIMT Challenge (Small Model Track). Furthermore, it achieves state-of-the-art (SOTA) results on OCRBench among VLMs with fewer than 3B parameters. HunyuanOCR achieves breakthroughs in three key aspects: 1) Unifying Versatility and Efficiency: We implement comprehensive support for core capabilities including spotting, parsing, IE, VQA, and translation within a lightweight framework. This addresses the limitations of narrow "OCR expert models" and inefficient "General VLMs". 2) Streamlined End-to-End Architecture: Adopting a pure end-to-end paradigm eliminates dependencies on pre-processing modules (e.g., layout analysis). This fundamentally resolves error propagation common in traditional pipelines and simplifies system deployment. 3) Data-Driven and RL Strategies: We confirm the critical role of high-quality data and, for the first time in the industry, demonstrate that Reinforcement Learning (RL) strategies yield significant performance gains in OCR tasks. HunyuanOCR is officially open-sourced on HuggingFace. We also provide a high-performance deployment solution based on vLLM, placing its production efficiency in the top tier. We hope this model will advance frontier research and provide a solid foundation for industrial applications.

Paperid: 3475, https://arxiv.org/pdf/2511.19111.pdf

Abstract:
Diffusion-based editing enables realistic modification of local image regions, making AI-generated content harder to detect. Existing AIGC detection benchmarks focus on classifying entire images, overlooking the localization of diffusion-based edits. We introduce DiffSeg30k, a publicly available dataset of 30k diffusion-edited images with pixel-level annotations, designed to support fine-grained detection. DiffSeg30k features: 1) In-the-wild images--we collect images or image prompts from COCO to reflect real-world content diversity; 2) Diverse diffusion models--local edits using eight SOTA diffusion models; 3) Multi-turn editing--each image undergoes up to three sequential edits to mimic real-world sequential editing; and 4) Realistic editing scenarios--a vision-language model (VLM)-based pipeline automatically identifies meaningful regions and generates context-aware prompts covering additions, removals, and attribute changes. DiffSeg30k shifts AIGC detection from binary classification to semantic segmentation, enabling simultaneous localization of edits and identification of the editing models. We benchmark three baseline segmentation approaches, revealing significant challenges in semantic segmentation tasks, particularly concerning robustness to image distortions. Experiments also reveal that segmentation models, despite being trained for pixel-level localization, emerge as highly reliable whole-image classifiers of diffusion edits, outperforming established forgery classifiers while showing great potential in cross-generator generalization. We believe DiffSeg30k will advance research in fine-grained localization of AI-generated content by demonstrating the promise and limitations of segmentation-based methods. DiffSeg30k is released at: https://huggingface.co/datasets/Chaos2629/Diffseg30k

Paperid: 3476, https://arxiv.org/pdf/2511.15605.pdf

Abstract:
Vision-Language-Action (VLA) models excel in robotic manipulation but are constrained by their heavy reliance on expert demonstrations, leading to demonstration bias and limiting performance. Reinforcement learning (RL) is a vital post-training strategy to overcome these limits, yet current VLA-RL methods, including group-based optimization approaches, are crippled by severe reward sparsity. Relying on binary success indicators wastes valuable information in failed trajectories, resulting in low training efficiency. To solve this, we propose Self-Referential Policy Optimization (SRPO), a novel VLA-RL framework. SRPO eliminates the need for external demonstrations or manual reward engineering by leveraging the model's own successful trajectories, generated within the current training batch, as a self-reference. This allows us to assign a progress-wise reward to failed attempts. A core innovation is the use of latent world representations to measure behavioral progress robustly. Instead of relying on raw pixels or requiring domain-specific fine-tuning, we utilize the compressed, transferable encodings from a world model's latent space. These representations naturally capture progress patterns across environments, enabling accurate, generalized trajectory comparison. Empirical evaluations on the LIBERO benchmark demonstrate SRPO's efficiency and effectiveness. Starting from a supervised baseline with 48.9% success, SRPO achieves a new state-of-the-art success rate of 99.2% in just 200 RL steps, representing a 103% relative improvement without any extra supervision. Furthermore, SRPO shows substantial robustness, achieving a 167% performance improvement on the LIBERO-Plus benchmark.

Paperid: 3477, https://arxiv.org/pdf/2511.15090.pdf

Abstract:
Document Visual Question Answering (DocVQA) is a fundamental task for multimodal document understanding and a key testbed for vision language reasoning. However, most existing DocVQA datasets are limited to the page level and lack fine grained spatial grounding, constraining the interpretability and reasoning capability of Vision Language Models (VLMs). To address this gap, we introduce BBox DocVQA a large scale, bounding box grounded dataset designed to enhance spatial reasoning and evidence localization in visual documents. We further present an automated construction pipeline, Segment Judge and Generate, which integrates a segment model for region segmentation, a VLM for semantic judgment, and another advanced VLM for question answer generation, followed by human verification for quality assurance. The resulting dataset contains 3.6 K diverse documents and 32 K QA pairs, encompassing single and multi region as well as single and multi page scenarios. Each QA instance is grounded on explicit bounding boxes, enabling fine grained evaluation of spatial semantic alignment. Benchmarking multiple state of the art VLMs (e.g., GPT 5, Qwen2.5 VL, and InternVL) on BBox DocVQA reveals persistent challenges in spatial grounding and reasoning accuracy. Furthermore, fine tuning on BBox DocVQA substantially improves both bounding box localization and answer generation, validating its effectiveness for enhancing the reasoning ability of VLMs. Our dataset and code will be publicly released to advance research on interpretable and spatially grounded vision language reasoning.

Paperid: 3478, https://arxiv.org/pdf/2511.05923.pdf

Abstract:
Despite the remarkable advancements of Large Vision-Language Models (LVLMs), the mechanistic interpretability remains underexplored. Existing analyses are insufficiently comprehensive and lack examination covering visual and textual tokens, model components, and the full range of layers. This limitation restricts actionable insights to improve the faithfulness of model output and the development of downstream tasks, such as hallucination mitigation. To address this limitation, we introduce Fine-grained Cross-modal Causal Tracing (FCCT) framework, which systematically quantifies the causal effects on visual object perception. FCCT conducts fine-grained analysis covering the full range of visual and textual tokens, three core model components including multi-head self-attention (MHSA), feed-forward networks (FFNs), and hidden states, across all decoder layers. Our analysis is the first to demonstrate that MHSAs of the last token in middle layers play a critical role in aggregating cross-modal information, while FFNs exhibit a three-stage hierarchical progression for the storage and transfer of visual object representations. Building on these insights, we propose Intermediate Representation Injection (IRI), a training-free inference-time technique that reinforces visual object information flow by precisely intervening on cross-modal representations at specific components and layers, thereby enhancing perception and mitigating hallucination. Consistent improvements across five widely used benchmarks and LVLMs demonstrate IRI achieves state-of-the-art performance, while preserving inference speed and other foundational performance.

Paperid: 3479, https://arxiv.org/pdf/2511.05894.pdf

Abstract:
Understanding 3D scenes in open-world settings poses fundamental challenges for vision and robotics, particularly due to the limitations of closed-vocabulary supervision and static annotations. To address this, we propose a unified framework for Open-World 3D Scene Graph Generation with Retrieval-Augmented Reasoning, which enables generalizable and interactive 3D scene understanding. Our method integrates Vision-Language Models (VLMs) with retrieval-based reasoning to support multimodal exploration and language-guided interaction. The framework comprises two key components: (1) a dynamic scene graph generation module that detects objects and infers semantic relationships without fixed label sets, and (2) a retrieval-augmented reasoning pipeline that encodes scene graphs into a vector database to support text/image-conditioned queries. We evaluate our method on 3DSSG and Replica benchmarks across four tasks-scene question answering, visual grounding, instance retrieval, and task planning-demonstrating robust generalization and superior performance in diverse environments. Our results highlight the effectiveness of combining open-vocabulary perception with retrieval-based reasoning for scalable 3D scene understanding.

Paperid: 3480, https://arxiv.org/pdf/2510.16617.pdf

Abstract:
Vision-Language-Action (VLA) models trained on large robot datasets promise general-purpose, robust control across diverse domains and embodiments. However, existing approaches often fail out-of-the-box when deployed in novel environments, embodiments, or tasks. We introduce Mixture of Skills VLA (MoS-VLA), a framework that represents robot manipulation policies as linear combinations of a finite set of learned basis functions. During pretraining, MoS-VLA jointly learns these basis functions across datasets from the Open X-Embodiment project, producing a structured skill space. At test time, adapting to a new task requires only a single expert demonstration. The corresponding skill representation is then inferred via a lightweight convex optimization problem that minimizes the L1 action error, without requiring gradient updates. This gradient-free adaptation incurs minimal overhead while enabling rapid instantiation of new skills. Empirically, MoS-VLA achieves lower action-prediction error on five out of five unseen datasets and succeeds in both simulation and real-robot tasks where a pretrained VLA model fails outright. Project page: mos-vla.github.io/

Paperid: 3481, https://arxiv.org/pdf/2510.10969.pdf

Abstract:
Existing vision language models (VLMs), including GPT-4 and DALL-E, often struggle to preserve logic, object identity, and style in multimodal image-text generation. This limitation significantly hinders the generalization capability of VLMs in complex image-text input-output scenarios. To address this issue, we propose IUT-Plug, a module grounded in an Image Understanding Tree (IUT), which enhances existing interleaved VLMs through explicit structured reasoning, thereby mitigating context drift in logic, entity identity, and style. The proposed framework operates in two stages. (1) A dynamic IUT-Plug extraction module parses visual scenes into hierarchical symbolic structures. (2) A coordinated narrative-flow and image synthesis mechanism ensures cross-modal consistency. To evaluate our approach, we construct a novel benchmark based on 3,000 real human-generated question-answer pairs over fine-tuned large models, introducing a dynamic evaluation protocol for quantifying context drift in interleaved VLMs. Experimental results demonstrate that IUT-Plug not only improves accuracy on established benchmarks but also effectively alleviates the three critical forms of context drift across diverse multimodal question answering (QA) scenarios.

Paperid: 3482, https://arxiv.org/pdf/2510.10285.pdf

Abstract:
Multimodal large reasoning models (MLRMs) are rapidly advancing vision-language reasoning and are emerging as a foundation for cross-modal intelligence. Hallucination remains a persistent failure mode, manifesting itself as erroneous reasoning chains and misinterpretation of visual content. In this study, we observe that attention heads exhibit a staged division: shallow heads predominantly serve perception, while deeper heads shift toward symbolic reasoning, revealing two major causes of hallucination, namely perceptual bias and reasoning drift. To address these issues, we propose a lightweight and interpretable two-step plugin, Functional Head Identification and Class-conditioned Rescaling, which locates perception- and reasoning-oriented heads and regulates their contributions without retraining. Evaluations on three real-world MLRMs (Kimi-VL, Ocean-R1, R1-Onevision), six benchmarks across three domains, and four baselines show that our plugin achieves an average improvement of 5% and up to 15%, with only <1% additional computation and 9% of baseline latency. Our approach is completely model-agnostic and significantly enhances both the reliability and interpretability of the off-the-shelf MLRMs, thereby enabling their safe deployment in high-stakes applications. Our code is available at https://anonymous.4open.science/r/Functional-Attention-Control.

Paperid: 3483, https://arxiv.org/pdf/2510.02677.pdf

Abstract:
As vision-language models (VLMs) gain prominence, their multimodal interfaces also introduce new safety vulnerabilities, making the safety evaluation challenging and critical. Existing red-teaming efforts are either restricted to a narrow set of adversarial patterns or depend heavily on manual engineering, lacking scalable exploration of emerging real-world VLM vulnerabilities. To bridge this gap, we propose ARMs, an adaptive red-teaming agent that systematically conducts comprehensive risk assessments for VLMs. Given a target harmful behavior or risk definition, ARMs automatically optimizes diverse red-teaming strategies with reasoning-enhanced multi-step orchestration, to effectively elicit harmful outputs from target VLMs. We propose 11 novel multimodal attack strategies, covering diverse adversarial patterns of VLMs (e.g., reasoning hijacking, contextual cloaking), and integrate 17 red-teaming algorithms into ARMs via model context protocol (MCP). To balance the diversity and effectiveness of the attack, we design a layered memory with an epsilon-greedy attack exploration algorithm. Extensive experiments on instance- and policy-based benchmarks show that ARMs achieves SOTA attack success rates, exceeding baselines by an average of 52.1% and surpassing 90% on Claude-4-Sonnet. We show that the diversity of red-teaming instances generated by ARMs is significantly higher, revealing emerging vulnerabilities in VLMs. Leveraging ARMs, we construct ARMs-Bench, a large-scale multimodal safety dataset comprising over 30K red-teaming instances spanning 51 diverse risk categories, grounded in both real-world multimodal threats and regulatory risks. Safety fine-tuning with ARMs-Bench substantially improves the robustness of VLMs while preserving their general utility, providing actionable guidance to improve multimodal safety alignment against emerging threats.

Paperid: 3484, https://arxiv.org/pdf/2509.24597.pdf

Abstract:
Dyslexia, a neurodevelopmental disorder characterized by persistent reading difficulties, is often linked to reduced activity of the visual word form area in the ventral occipito-temporal cortex. Traditional approaches to studying dyslexia, such as behavioral and neuroimaging methods, have provided valuable insights but remain limited in their ability to test causal hypotheses about the underlying mechanisms of reading impairments. In this study, we use large-scale vision-language models (VLMs) to simulate dyslexia by functionally identifying and perturbing artificial analogues of word processing. Using stimuli from cognitive neuroscience, we identify visual-word-form-selective units within VLMs and demonstrate that targeted ablation of these units, unlike ablation of random units, leads to selective impairments in reading tasks while general visual and language comprehension abilities remain intact. In particular, the resulting model matches dyslexic humans' phonological deficits without a significant change in orthographic processing. Taken together, our modeling results replicate key characteristics of dyslexia and establish a computational framework for investigating reading disorders.

Paperid: 3485, https://arxiv.org/pdf/2509.23760.pdf

Abstract:
The remarkable success of diffusion models in text-to-image generation has sparked growing interest in expanding their capabilities to a variety of multi-modal tasks, including image understanding, manipulation, and perception. These tasks require advanced semantic comprehension across both visual and textual modalities, especially in scenarios involving complex semantic instructions. However, existing approaches often rely heavily on vision-language models (VLMs) or modular designs for semantic guidance, leading to fragmented architectures and computational inefficiency. To address these challenges, we propose UniAlignment, a unified multimodal generation framework within a single diffusion transformer. UniAlignment introduces a dual-stream diffusion training strategy that incorporates both intrinsic-modal semantic alignment and cross-modal semantic alignment, thereby enhancing the model's cross-modal consistency and instruction-following robustness. Additionally, we present SemGen-Bench, a new benchmark specifically designed to evaluate multimodal semantic consistency under complex textual instructions. Extensive experiments across multiple tasks and benchmarks demonstrate that UniAlignment outperforms existing baselines, underscoring the significant potential of diffusion models in unified multimodal generation.

Paperid: 3486, https://arxiv.org/pdf/2509.22404.pdf

Abstract:
Anatomical understanding through deep learning is critical for automatic report generation, intra-operative navigation, and organ localization in medical imaging; however, its progress is constrained by the scarcity of expert-labeled data. A promising remedy is to leverage an annotated reference image to guide the interpretation of an unlabeled target. Although recent vision-language models (VLMs) exhibit non-trivial visual reasoning, their reference-based understanding and fine-grained localization remain limited. We introduce RAU, a framework for reference-based anatomical understanding with VLMs. We first show that a VLM learns to identify anatomical regions through relative spatial reasoning between reference and target images, trained on a moderately sized dataset. We validate this capability through visual question answering (VQA) and bounding box prediction. Next, we demonstrate that the VLM-derived spatial cues can be seamlessly integrated with the fine-grained segmentation capability of SAM2, enabling localization and pixel-level segmentation of small anatomical regions, such as vessel segments. Across two in-distribution and two out-of-distribution datasets, RAU consistently outperforms a SAM2 fine-tuning baseline using the same memory setup, yielding more accurate segmentations and more reliable localization. More importantly, its strong generalization ability makes it scalable to out-of-distribution datasets, a property crucial for medical image applications. To the best of our knowledge, RAU is the first to explore the capability of VLMs for reference-based identification, localization, and segmentation of anatomical structures in medical images. Its promising performance highlights the potential of VLM-driven approaches for anatomical understanding in automated clinical workflows.

Paperid: 3487, https://arxiv.org/pdf/2509.21576.pdf

Abstract:
The advancement of vision language models (VLMs) has empowered embodied agents to accomplish simple multimodal planning tasks, but not long-horizon ones requiring long sequences of actions. In text-only simulations, long-horizon planning has seen significant improvement brought by repositioning the role of LLMs. Instead of directly generating action sequences, LLMs translate the planning domain and problem into a formal planning language like the Planning Domain Definition Language (PDDL), which can call a formal solver to derive the plan in a verifiable manner. In multimodal environments, research on VLM-as-formalizer remains scarce, usually involving gross simplifications such as predefined object vocabulary or overly similar few-shot examples. In this work, we present a suite of five VLM-as-formalizer pipelines that tackle one-shot, open-vocabulary, and multimodal PDDL formalization. We evaluate those on an existing benchmark while presenting another two that for the first time account for planning with authentic, multi-view, and low-quality images. We conclude that VLM-as-formalizer greatly outperforms end-to-end plan generation. We reveal the bottleneck to be vision rather than language, as VLMs often fail to capture an exhaustive set of necessary object relations. While generating intermediate, textual representations such as captions or scene graphs partially compensate for the performance, their inconsistent gain leaves headroom for future research directions on multimodal planning formalization.

Paperid: 3488, https://arxiv.org/pdf/2509.19659.pdf

Abstract:
Large vision-language models (VLMs) can jointly interpret images and text, but they are also prone to absorbing and reproducing harmful social stereotypes when visual cues such as age, gender, race, clothing, or occupation are present. To investigate these risks, we introduce a news-image benchmark consisting of 1,343 image-question pairs drawn from diverse outlets, which we annotated with ground-truth answers and demographic attributes (age, gender, race, occupation, and sports). We evaluate a range of state-of-the-art VLMs and employ a large language model (LLM) as judge, with human verification. Our findings show that: (i) visual context systematically shifts model outputs in open-ended settings; (ii) bias prevalence varies across attributes and models, with particularly high risk for gender and occupation; and (iii) higher faithfulness does not necessarily correspond to lower bias. We release the benchmark prompts, evaluation rubric, and code to support reproducible and fairness-aware multimodal assessment.

Paperid: 3489, https://arxiv.org/pdf/2509.14860.pdf

Abstract:
Image classification has traditionally relied on parameter-intensive model training, requiring large-scale annotated datasets and extensive fine tuning to achieve competitive performance. While recent vision language models (VLMs) alleviate some of these constraints, they remain limited by their reliance on single pass representations, often failing to capture complementary aspects of visual content. In this paper, we introduce Multi Agent based Reasoning for Image Classification (MARIC), a multi agent framework that reformulates image classification as a collaborative reasoning process. MARIC first utilizes an Outliner Agent to analyze the global theme of the image and generate targeted prompts. Based on these prompts, three Aspect Agents extract fine grained descriptions along distinct visual dimensions. Finally, a Reasoning Agent synthesizes these complementary outputs through integrated reflection step, producing a unified representation for classification. By explicitly decomposing the task into multiple perspectives and encouraging reflective synthesis, MARIC mitigates the shortcomings of both parameter-heavy training and monolithic VLM reasoning. Experiments on 4 diverse image classification benchmark datasets demonstrate that MARIC significantly outperforms baselines, highlighting the effectiveness of multi-agent visual reasoning for robust and interpretable image classification.

Paperid: 3490, https://arxiv.org/pdf/2508.16271.pdf

Abstract:
Multimodal large language models (MLLMs) have emerged as pivotal tools in enhancing human-computer interaction. In this paper we focus on the application of MLLMs in the field of graphical user interface (GUI) elements structuring, where they assist in processing user instructions based on screen contents. Despite the promise of MLLMs, their performance in precisely generating UI element coordinates, a critical aspect of GUI understanding, is hindered by the nature of next-token prediction training. This challenge arises from the semantic void surrounding numerical UI coordinates in language representation spaces, necessitating a substantial and diverse dataset to bolster visual module capabilities. To address these limitations, we introduce an IoU-Augmented Maximum Likelihood (IAML) training paradigm. Specifically, our approach involves a novel pipeline for IoU-based coordinate sampling to augment the training data, which considers the proximity to ground truth coordinates. This data augmentation strategy is then employed to fine-tune MLLMs under the IAML paradigm, which is designed to mitigate the exposure bias problem inherent in traditional maximum likelihood estimation. Through extensive experiments, we demonstrate the superior performance of our IAML training approach over traditional training paradigms.

Paperid: 3491, https://arxiv.org/pdf/2508.08276.pdf

Abstract:
This work adapts a neuroscientific contrast localizer to pinpoint causally relevant units for Theory of Mind (ToM) and mathematical reasoning tasks in large language models (LLMs) and vision-language models (VLMs). Across 11 LLMs and 5 VLMs ranging in size from 3B to 90B parameters, we localize top-activated units using contrastive stimulus sets and assess their causal role via targeted ablations. We compare the effect of lesioning functionally selected units against low-activation and randomly selected units on downstream accuracy across established ToM and mathematical benchmarks. Contrary to expectations, low-activation units sometimes produced larger performance drops than the highly activated ones, and units derived from the mathematical localizer often impaired ToM performance more than those from the ToM localizer. These findings call into question the causal relevance of contrast-based localizers and highlight the need for broader stimulus sets and more accurately capture task-specific units.

Paperid: 3492, https://arxiv.org/pdf/2508.07501.pdf

Abstract:
Good form is the difference between strength and strain, yet for the fast-growing community of at-home fitness enthusiasts, expert feedback is often out of reach. FormCoach transforms a simple camera into an always-on, interactive AI training partner, capable of spotting subtle form errors and delivering tailored corrections in real time, leveraging vision-language models (VLMs). We showcase this capability through a web interface and benchmark state-of-the-art VLMs on a dataset of 1,700 expert-annotated user-reference video pairs spanning 22 strength and mobility exercises. To accelerate research in AI-driven coaching, we release both the dataset and an automated, rubric-based evaluation pipeline, enabling standardized comparison across models. Our benchmarks reveal substantial gaps compared to human-level coaching, underscoring both the challenges and opportunities in integrating nuanced, context-aware movement analysis into interactive AI systems. By framing form correction as a collaborative and creative process between humans and machines, FormCoach opens a new frontier in embodied AI.

Paperid: 3493, https://arxiv.org/pdf/2508.06515.pdf

Abstract:
Social media platforms increasingly struggle to detect harmful content that promotes muscle dysmorphic behaviors, particularly pro-bigorexia content that disproportionately affects adolescent males. Unlike traditional eating disorder detection focused on the "thin ideal," pro-bigorexia material masquerades as legitimate fitness content through complex multimodal combinations of visual displays, coded language, and motivational messaging that evade text-based detection systems. We address this challenge by developing BigTokDetect, a clinically-informed detection framework for identifying pro-bigorexia content on TikTok. We introduce BigTok, the first expert-annotated multimodal dataset of over 2,200 TikTok videos labeled by clinical psychologists and psychiatrists across five primary categories spanning body image, nutrition, exercise, supplements, and masculinity. Through a comprehensive evaluation of state-of-the-art vision language models, we achieve 82.9% accuracy on primary category classification and 69.0% on subcategory detection via domain-specific finetuning. Our ablation studies demonstrate that multimodal fusion improves performance by 5-10% over text-only approaches, with video features providing the most discriminative signals. These findings establish new benchmarks for multimodal harmful content detection and provide both the computational tools and methodological framework needed for scalable content moderation in specialized mental health domains.

Paperid: 3494, https://arxiv.org/pdf/2508.00171.pdf

Abstract:
Clinical decision-making relies on the integrated analysis of medical images and the associated clinical reports. While Vision-Language Models (VLMs) can offer a unified framework for such tasks, they can exhibit strong biases toward one modality, frequently overlooking critical visual cues in favor of textual information. In this work, we introduce Selective Modality Shifting (SMS), a perturbation-based approach to quantify a model's reliance on each modality in binary classification tasks. By systematically swapping images or text between samples with opposing labels, we expose modality-specific biases. We assess six open-source VLMs-four generalist models and two fine-tuned for medical data-on two medical imaging datasets with distinct modalities: MIMIC-CXR (chest X-ray) and FairVLMed (scanning laser ophthalmoscopy). By assessing model performance and the calibration of every model in both unperturbed and perturbed settings, we reveal a marked dependency on text input, which persists despite the presence of complementary visual information. We also perform a qualitative attention-based analysis which further confirms that image content is often overshadowed by text details. Our findings highlight the importance of designing and evaluating multimodal medical models that genuinely integrate visual and textual cues, rather than relying on single-modality signals.

Paperid: 3495, https://arxiv.org/pdf/2507.14312.pdf

Abstract:
Vision-language models (VLMs) like CLIP exhibit strong zero-shot capabilities but often fail to generalize under distribution shifts. Test-time adaptation (TTA) allows models to update at inference time without labeled data, typically via entropy minimization. However, this objective is fundamentally misaligned with the contrastive image-text training of VLMs, limiting adaptation performance and introducing failure modes such as pseudo-label drift and class collapse. We propose CLIPTTA, a new gradient-based TTA method for vision-language models that leverages a soft contrastive loss aligned with CLIP's pre-training objective. We provide a theoretical analysis of CLIPTTA's gradients, showing how its batch-aware design mitigates the risk of collapse. We further extend CLIPTTA to the open-set setting, where both in-distribution (ID) and out-of-distribution (OOD) samples are encountered, using an Outlier Contrastive Exposure (OCE) loss to improve OOD detection. Evaluated on 75 datasets spanning diverse distribution shifts, CLIPTTA consistently outperforms entropy-based objectives and is highly competitive with state-of-the-art TTA methods, outperforming them on a large number of datasets and exhibiting more stable performance across diverse shifts.

Paperid: 3496, https://arxiv.org/pdf/2507.09097.pdf

Abstract:
Large Vision-Language Models (LVLMs) have demonstrated promising performance in chest X-ray (CXR) analysis. To enhance human-computer interaction, several studies have incorporated radiologists' eye gaze, typically through heatmaps or textual prompts. However, these methods often overlook the sequential order of eye movements, which could provide valuable insights by highlighting both the areas of interest and the order in which they are examined. In this work, we propose a novel approach called RadEyeVideo that integrates radiologists' eye-fixation data as a video sequence, capturing both the temporal and spatial dynamics of their gaze. We evaluate this method in CXR report generation and disease diagnosis using three general-domain, open-source LVLMs with video input capabilities. When prompted with eye-gaze videos, model performance improves by up to 24.6% in the report generation task and on average 15.2% for both tasks using scaled evaluation metrics. Notably, RadEyeVideo enhanced an open-domain LVLM model, LLaVA-OneVision, to surpass task-specific medical LVLMs such as MAIRA-2 and CheXagent, trained on large Chest X-ray data. This work highlights that domain expert's knowledge (eye-gaze information in this case), when effectively integrated with LVLMs, can significantly enhance general-domain models' capabilities in clinical tasks. RadEyeVideo is a step toward a scalable human-centered approach of utilizing LVLMs in medical image analytics.

Paperid: 3497, https://arxiv.org/pdf/2506.13925.pdf

Abstract:
In this paper, we address Semi-supervised Semantic Segmentation (SSS) under domain shift by leveraging domain-invariant semantic knowledge from text embeddings of Vision-Language Models (VLMs). We propose a unified Hierarchical Vision-Language framework (HVL) that integrates domain-invariant text embeddings as object queries in a transformer-based segmentation network to improve generalization and reduce misclassification under limited supervision. The mentioned textual queries are used for grouping pixels with shared semantics under SSS. HVL is designed to (1) generate textual queries that maximally encode domain-invariant semantics from VLM while capturing intra-class variations; (2) align these queries with spatial visual features to enhance their segmentation ability and improve the semantic clarity of visual features. We also introduce targeted regularization losses that maintain vision--language alignment throughout training to reinforce semantic understanding. HVL establishes a novel state-of-the-art by achieving a +9.3% improvement in mean Intersection over Union (mIoU) on COCO, utilizing 232 labelled images, +3.1% on Pascal VOC employing 92 labels, +4.8% on ADE20 using 316 labels, and +3.4% on Cityscapes with 100 labels, demonstrating superior performance with less than 1% supervision on four benchmark datasets. Our results show that language-guided segmentation bridges the label efficiency gap and enables new levels of fine-grained generalization.

Paperid: 3498, https://arxiv.org/pdf/2506.13039.pdf

Abstract:
Person re-identification (ReID) has evolved from handcrafted feature-based methods to deep learning approaches and, more recently, to models incorporating large language models (LLMs). Early methods struggled with variations in lighting, pose, and viewpoint, but deep learning addressed these issues by learning robust visual features. Building on this, LLMs now enable ReID systems to integrate semantic and contextual information through natural language. This survey traces that full evolution and offers one of the first comprehensive reviews of ReID approaches that leverage LLMs, where textual descriptions are used as privileged information to improve visual matching. A key contribution is the use of dynamic, identity-specific prompts generated by GPT-4o, which enhance the alignment between images and text in vision-language ReID systems. Experimental results show that these descriptions improve accuracy, especially in complex or ambiguous cases. To support further research, we release a large set of GPT-4o-generated descriptions for standard ReID datasets. By bridging computer vision and natural language processing, this survey offers a unified perspective on the field's development and outlines key future directions such as better prompt design, cross-modal transfer learning, and real-world adaptability.

Paperid: 3499, https://arxiv.org/pdf/2506.12409.pdf

Abstract:
Continual learning in vision-language models (VLMs) faces critical challenges in balancing parameter efficiency, memory consumption, and optimization stability. While First-Order (FO) optimization (e.g., SGD) dominate current approaches, their deterministic gradients often trap models in suboptimal local minima and incur substantial memory overhead. This paper pioneers a systematic exploration of Zeroth-Order (ZO) optimization for vision-language continual learning (VLCL). We first identify the incompatibility of naive full-ZO adoption in VLCL due to modality-specific instability. To resolve this, we selectively applying ZO to either vision or language modalities while retaining FO in the complementary branch. Furthermore, we develop a layer-wise optimization paradigm that interleaves ZO and FO across network layers, capitalizing on the heterogeneous learning dynamics of shallow versus deep representations. A key theoretical insight reveals that ZO perturbations in vision branches exhibit higher variance than language counterparts, prompting a gradient sign normalization mechanism with modality-specific perturbation constraints. Extensive experiments on four benchmarks demonstrate that our method achieves state-of-the-art performance, reducing memory consumption by 89.1% compared to baselines. Code will be available upon publication.

Paperid: 3500, https://arxiv.org/pdf/2506.07385.pdf

Abstract:
In this work, we propose GUIPilot, an approach for detecting inconsistencies between the mobile design and their implementations. The mobile design usually consists of design mock-ups that specify (1) the expected screen appearances (e.g., widget layouts, colors, and shapes) and (2) the expected screen behaviors, regarding how one screen can transition into another (e.g., labeled widgets with textual description). Given a design mock-up and the implementation of its application, GUIPilot reports both their screen inconsistencies as well as process inconsistencies. On the one hand, GUIPilot detects the screen inconsistencies by abstracting every screen into a widget container where each widget is represented by its position, width, height, and type. By defining the partial order of widgets and the costs of replacing, inserting, and deleting widgets in a screen, we convert the screen-matching problem into an optimizable widget alignment problem. On the other hand, we translate the specified GUI transition into stepwise actions on the mobile screen (e.g., click, long-press, input text on some widgets). To this end, we propose a visual prompt for the vision-language model to infer widget-specific actions on the screen. By this means, we can validate the presence or absence of expected transitions in the implementation. Our extensive experiments on 80 mobile applications and 160 design mock-ups show that (1) GUIPilot can achieve 94.5% precision and 99.6% recall in detecting screen inconsistencies, outperforming the state-of-the-art approach, such as GVT, by 66.2% and 56.6% respectively, and (2) GUIPilot reports zero errors in detecting process inconsistencies. Furthermore, our industrial case study on applying GUIPilot on a trading mobile application shows that GUIPilot has detected nine application bugs, and all the bugs were confirmed by the original application experts.

Paperid: 3501, https://arxiv.org/pdf/2506.06232.pdf

Abstract:
While traditional computer vision models have historically struggled to generalize to endoscopic domains, the emergence of foundation models has shown promising cross-domain performance. In this work, we present the first large-scale study assessing the capabilities of Vision Language Models (VLMs) for endoscopic tasks with a specific focus on laparoscopic surgery. Using a diverse set of state-of-the-art models, multiple surgical datasets, and extensive human reference annotations, we address three key research questions: (1) Can current VLMs solve basic perception tasks on surgical images? (2) Can they handle advanced frame-based endoscopic scene understanding tasks? and (3) How do specialized medical VLMs compare to generalist models in this context? Our results reveal that VLMs can effectively perform basic surgical perception tasks, such as object counting and localization, with performance levels comparable to general domain tasks. However, their performance deteriorates significantly when the tasks require medical knowledge. Notably, we find that specialized medical VLMs currently underperform compared to generalist models across both basic and advanced surgical tasks, suggesting that they are not yet optimized for the complexity of surgical environments. These findings highlight the need for further advancements to enable VLMs to handle the unique challenges posed by surgery. Overall, our work provides important insights for the development of next-generation endoscopic AI systems and identifies key areas for improvement in medical visual language models.

Paperid: 3502, https://arxiv.org/pdf/2506.04562.pdf

Abstract:
Mesh deformation is a fundamental tool in 3D content manipulation. Despite extensive prior research, existing approaches often suffer from low output quality, require significant manual tuning, or depend on data-intensive training. To address these limitations, we introduce a training-free, handle-based mesh deformation method. % Our core idea is to leverage a Vision-Language Model (VLM) to interpret and manipulate a handle-based interface through prompt engineering. We begin by applying cone singularity detection to identify a sparse set of potential handles. The VLM is then prompted to select both the deformable sub-parts of the mesh and the handles that best align with user instructions. Subsequently, we query the desired deformed positions of the selected handles in screen space. To reduce uncertainty inherent in VLM predictions, we aggregate the results from multiple camera views using a novel multi-view voting scheme. % Across a suite of benchmarks, our method produces deformations that align more closely with user intent, as measured by CLIP and GPTEval3D scores, while introducing low distortion -- quantified via membrane energy. In summary, our approach is training-free, highly automated, and consistently delivers high-quality mesh deformations.

Paperid: 3503, https://arxiv.org/pdf/2506.02917.pdf

Abstract:
We propose a training-free, Vision-Language Model (VLM)-guided approach for efficiently generating trajectories to facilitate target inspection planning based on text descriptions. Unlike existing Vision-and-Language Navigation (VLN) methods designed for general agents in unknown environments, our approach specifically targets the efficient inspection of known scenes, with widespread applications in fields such as medical, marine, and civil engineering. Leveraging VLMs, our method first extracts points of interest (POIs) from the text description, then identifies a set of waypoints from which POIs are both salient and align with the spatial constraints defined in the prompt. Next, we interact with the VLM to iteratively refine the trajectory, preserving the visibility and prominence of the POIs. Further, we solve a Traveling Salesman Problem (TSP) to find the most efficient visitation order that satisfies the order constraint implied in the text description. Finally, we apply trajectory optimization to generate smooth, executable inspection paths for aerial and underwater vehicles. We have evaluated our method across a series of both handcrafted and real-world scanned environments. The results demonstrate that our approach effectively generates inspection planning trajectories that adhere to user instructions.

Paperid: 3504, https://arxiv.org/pdf/2506.02671.pdf

Abstract:
Test-time adaptation (TTA) has emerged as a critical technique for enhancing the generalization capability of vision-language models (VLMs) during inference. However, existing approaches often incur substantial computational costs and exhibit poor scalability, primarily due to sample-wise adaptation granularity and reliance on costly auxiliary designs such as data augmentation. To address these limitations, we introduce SAIL (Small Aid, Big Leap), a novel adapter-based TTA framework that leverages a lightweight, learnable AdaptNet to enable efficient and scalable model adaptation. As SAIL's core, a frozen pre-trained VLM collaborates with AdaptNet through a confidence-based interpolation weight, generating robust predictions during inference. These predictions serve as self-supervised targets to align AdaptNet's outputs through efficient batch-wise processing, dramatically reducing computational costs without modifying the VLM or requiring memory caches. To mitigate catastrophic forgetting during continual adaptation, we propose a gradient-aware reset strategy driven by a gradient drift indicator (GDI), which dynamically detects domain transitions and strategically resets AdaptNet for stable adaptation. Extensive experiments across diverse benchmarks on two scenarios demonstrate that SAIL achieves state-of-the-art performance while maintaining low computational costs. These results highlight SAIL's effectiveness, efficiency and scalability for real-world deployment. The code will be released upon acceptance.

Paperid: 3505, https://arxiv.org/pdf/2505.20718.pdf

Abstract:
We introduce a novel self-improving framework that enhances Embodied Visual Tracking (EVT) with Vision-Language Models (VLMs) to address the limitations of current active visual tracking systems in recovering from tracking failure. Our approach combines the off-the-shelf active tracking methods with VLMs' reasoning capabilities, deploying a fast visual policy for normal tracking and activating VLM reasoning only upon failure detection. The framework features a memory-augmented self-reflection mechanism that enables the VLM to progressively improve by learning from past experiences, effectively addressing VLMs' limitations in 3D spatial reasoning. Experimental results demonstrate significant performance improvements, with our framework boosting success rates by $72\%$ with state-of-the-art RL-based approaches and $220\%$ with PID-based methods in challenging environments. This work represents the first integration of VLM-based reasoning to assist EVT agents in proactive failure recovery, offering substantial advances for real-world robotic applications that require continuous target monitoring in dynamic, unstructured environments. Project website: https://sites.google.com/view/evt-recovery-assistant.

Paperid: 3506, https://arxiv.org/pdf/2505.17423.pdf

Abstract:
Many decision-making tasks, where both accuracy and efficiency matter, still require human supervision. For example, tasks like traffic officers reviewing hour-long dashcam footage or researchers screening conference videos can benefit from concise summaries that reduce cognitive load and save time. Yet current vision-language models (VLMs) often produce verbose, redundant outputs that hinder task performance. Existing video caption evaluation depends on costly human annotations and overlooks the summaries' utility in downstream tasks. We address these gaps with Video-to-text Information Bottleneck Evaluation (VIBE), an annotation-free method that scores VLM outputs using two metrics: grounding (how well the summary aligns with visual content) and utility (how informative it is for the task). VIBE selects from randomly sampled VLM outputs by ranking them according to the two scores to support effective human decision-making. Human studies on LearningPaper24, SUTD-TrafficQA, and LongVideoBench show that summaries selected by VIBE consistently improve performance-boosting task accuracy by up to 61.23% and reducing response time by 75.77% compared to naive VLM summaries or raw video.

Paperid: 3507, https://arxiv.org/pdf/2505.16659.pdf

Abstract:
Medical anomaly detection (AD) is crucial for early clinical intervention, yet it faces challenges due to limited access to high-quality medical imaging data, caused by privacy concerns and data silos. Few-shot learning has emerged as a promising approach to alleviate these limitations by leveraging the large-scale prior knowledge embedded in vision-language models (VLMs). Recent advancements in few-shot medical AD have treated normal and abnormal cases as a one-class classification problem, often overlooking the distinction among multiple anomaly categories. Thus, in this paper, we propose a framework tailored for few-shot medical anomaly detection in the scenario where the identification of multiple anomaly categories is required. To capture the detailed radiological signs of medical anomaly categories, our framework incorporates diverse textual descriptions for each category generated by a Large-Language model, under the assumption that different anomalies in medical images may share common radiological signs in each category. Specifically, we introduce SD-MAD, a two-stage Sign-Driven few-shot Multi-Anomaly Detection framework: (i) Radiological signs are aligned with anomaly categories by amplifying inter-anomaly discrepancy; (ii) Aligned signs are selected further to mitigate the effect of the under-fitting and uncertain-sample issue caused by limited medical data, employing an automatic sign selection strategy at inference. Moreover, we propose three protocols to comprehensively quantify the performance of multi-anomaly detection. Extensive experiments illustrate the effectiveness of our method.

Paperid: 3508, https://arxiv.org/pdf/2505.14160.pdf

Abstract:
Multilingual vision-language models (VLMs) promise universal image-text retrieval, yet their social biases remain underexplored. We perform the first systematic audit of four public multilingual CLIP variants: M-CLIP, NLLB-CLIP, CAPIVARA-CLIP, and the debiased SigLIP-2, covering ten languages that differ in resource availability and morphological gender marking. Using balanced subsets of FairFace and the PATA stereotype suite in a zero-shot setting, we quantify race and gender bias and measure stereotype amplification. Contrary to the intuition that multilinguality mitigates bias, every model exhibits stronger gender skew than its English-only baseline. CAPIVARA-CLIP shows its largest biases precisely in the low-resource languages it targets, while the shared encoder of NLLB-CLIP and SigLIP-2 transfers English gender stereotypes into gender-neutral languages; loosely coupled encoders largely avoid this leakage. Although SigLIP-2 reduces agency and communion skews, it inherits -- and in caption-sparse contexts (e.g., Xhosa) amplifies -- the English anchor's crime associations. Highly gendered languages consistently magnify all bias types, yet gender-neutral languages remain vulnerable whenever cross-lingual weight sharing imports foreign stereotypes. Aggregated metrics thus mask language-specific hot spots, underscoring the need for fine-grained, language-aware bias evaluation in future multilingual VLM research.

Paperid: 3509, https://arxiv.org/pdf/2504.18458.pdf

Abstract:
Recent advances in large vision-language models (LVLMs) have revealed an \textit{overthinking} phenomenon, where models generate verbose reasoning across all tasks regardless of questions. To address this issue, we present \textbf{FAST}, a novel \textbf{Fa}st-\textbf{S}low \textbf{T}hinking framework that dynamically adapts reasoning depth based on question characteristics. Through empirical analysis, we establish the feasibility of fast-slow thinking in LVLMs by investigating how response length and data distribution affect performance. We develop FAST-GRPO with three components: model-based metrics for question characterization, an adaptive thinking reward mechanism, and difficulty-aware KL regularization. Experiments across seven reasoning benchmarks demonstrate that FAST achieves state-of-the-art accuracy with over 10\% relative improvement compared to the base model, while reducing token usage by 32.7-67.3\% compared to previous slow-thinking approaches, effectively balancing reasoning length and accuracy.

Paperid: 3510, https://arxiv.org/pdf/2504.02357.pdf

Abstract:
GUI testing is an essential quality assurance process in mobile app development. However, the creation and maintenance of GUI tests for mobile apps are resource-intensive and costly. Recognizing that many apps share similar functionalities, researchers have proposed various techniques to migrate GUI tests from one app to another with similar features. For example, some techniques employ mapping-based approaches to align the GUI elements traversed by the tests of a source app to those present in the target app. Other test migration techniques have also been proposed to leverage large language models (LLMs) by adapting the GUI tasks in source tests. However, these techniques are ineffective in dealing with different operational logic between the source and target apps. The semantics of GUI elements may not be correctly inferred due to the missing analysis of these flows. In this work, we propose REUSEDROID, a novel multiagent framework for GUI test migration empowered by Large Vision-Language Models (VLMs). REUSEDROID is powered by multiple VLM-based agents, each tackling a stage of the test migration process by leveraging the relevant visual and textual information embedded in GUI pages. An insight of REUSEDROID is to migrate tests based only on the core logic shared across similar apps, while their entire operational logic could differ. We evaluate REUSEDROID on LinPro, a new test migration dataset that consists of 578 migration tasks for 39 popular apps across 4 categories. The experimental result shows that REUSEDROID can successfully migrate 90.3% of the migration tasks, outperforming the best mapping-based and LLM-based baselines by 318.1% and 109.1%, respectively.

Paperid: 3511, https://arxiv.org/pdf/2504.01081.pdf

Abstract:
We introduce ShieldGemma 2, a 4B parameter image content moderation model built on Gemma 3. This model provides robust safety risk predictions across the following key harm categories: Sexually Explicit, Violence \& Gore, and Dangerous Content for synthetic images (e.g. output of any image generation model) and natural images (e.g. any image input to a Vision-Language Model). We evaluated on both internal and external benchmarks to demonstrate state-of-the-art performance compared to LlavaGuard \citep{helff2024llavaguard}, GPT-4o mini \citep{hurst2024gpt}, and the base Gemma 3 model \citep{gemma_2025} based on our policies. Additionally, we present a novel adversarial data generation pipeline which enables a controlled, diverse, and robust image generation. ShieldGemma 2 provides an open image moderation tool to advance multimodal safety and responsible AI development.

Paperid: 3512, https://arxiv.org/pdf/2503.23388.pdf

Abstract:
Recent vision-language models (VLMs) face significant challenges in test-time adaptation to novel domains. While cache-based methods show promise by leveraging historical information, they struggle with both caching unreliable feature-label pairs and indiscriminately using single-class information during querying, significantly compromising adaptation accuracy. To address these limitations, we propose COSMIC (Clique-Oriented Semantic Multi-space Integration for CLIP), a robust test-time adaptation framework that enhances adaptability through multi-granular, cross-modal semantic caching and graph-based querying mechanisms. Our framework introduces two key innovations: Dual Semantics Graph (DSG) and Clique Guided Hyper-class (CGH). The Dual Semantics Graph constructs complementary semantic spaces by incorporating textual features, coarse-grained CLIP features, and fine-grained DINOv2 features to capture rich semantic relationships. Building upon these dual graphs, the Clique Guided Hyper-class component leverages structured class relationships to enhance prediction robustness through correlated class selection. Extensive experiments demonstrate COSMIC's superior performance across multiple benchmarks, achieving significant improvements over state-of-the-art methods: 15.81% gain on out-of-distribution tasks and 5.33% on cross-domain generation with CLIP RN-50. Code is available at github.com/hf618/COSMIC.

Paperid: 3513, https://arxiv.org/pdf/2503.09241.pdf

Abstract:
Computer agents powered by vision-language models (VLMs) have significantly advanced human-computer interaction, enabling users to perform complex tasks through natural language instructions. However, these agents are vulnerable to context deception attacks, an emerging threat where adversaries embed misleading content into the agent's operational environment, such as a pop-up window containing deceptive instructions. Existing defenses, such as instructing agents to ignore deceptive elements, have proven largely ineffective. As the first systematic study on protecting computer agents, we introduce textbf{in-context defense}, leveraging in-context learning and chain-of-thought (CoT) reasoning to counter such attacks. Our approach involves augmenting the agent's context with a small set of carefully curated exemplars containing both malicious environments and corresponding defensive responses. These exemplars guide the agent to first perform explicit defensive reasoning before action planning, reducing susceptibility to deceptive attacks. Experiments demonstrate the effectiveness of our method, reducing attack success rates by 91.2% on pop-up window attacks, 74.6% on average on environment injection attacks, while achieving 100% successful defenses against distracting advertisements. Our findings highlight that (1) defensive reasoning must precede action planning for optimal performance, and (2) a minimal number of exemplars (fewer than three) is sufficient to induce an agent's defensive behavior.

Paperid: 3514, https://arxiv.org/pdf/2502.20490.pdf

Abstract:
Human activity is moderated by norms; however, supervision for normative reasoning is sparse, particularly where norms are physically- or socially-grounded. We thus present EGONORMIA $\|Îµ\|$, comprising 1,853 (200 for EGONORMIA-verified) multiple choice questions (MCQs) grounded within egocentric videos of human interactions, enabling the evaluation and improvement of normative reasoning in vision-language models (VLMs). EGONORMIA spans seven norm categories: safety, privacy, proxemics, politeness, cooperation, coordination/proactivity, and communication/legibility. To compile this dataset at scale, we propose a novel pipeline to generate grounded MCQs from raw egocentric video. Our work demonstrates that current state-of-the-art VLMs lack robust grounded norm understanding, scoring a maximum of 54% on EGONORMIA and 65% on EGONORMIA-verified, with performance across norm categories indicating significant risks of safety and privacy when VLMs are used in real-world agents. We additionally explore methods for improving normative understanding, demonstrating that a naive retrieval-based generation (RAG) method using EGONORMIA can enhance normative reasoning in VLMs.

Paperid: 3515, https://arxiv.org/pdf/2502.16612.pdf

Abstract:
The proliferation of multimodal content on social media presents significant challenges in understanding and moderating complex, context-dependent issues such as misinformation, hate speech, and propaganda. While efforts have been made to develop resources and propose new methods for automatic detection, limited attention has been given to label detection and the generation of explanation-based rationales for predicted labels. To address this challenge, we introduce MemeIntel, an explanation-enhanced dataset for propaganda memes in Arabic and hateful memes in English, making it the first large-scale resource for these tasks. To solve these tasks, we propose a multi-stage optimization approach and train Vision-Language Models (VLMs). Our results demonstrate that this approach significantly improves performance over the base model for both \textbf{label detection} and explanation generation, outperforming the current state-of-the-art with an absolute improvement of ~3% on ArMeme and ~7% on Hateful Memes. For reproducibility and future research, we aim to make the MemeIntel dataset and experimental resources publicly available.

Paperid: 3516, https://arxiv.org/pdf/2502.05863.pdf

Abstract:
In AI-facilitated teaching, leveraging various query styles to interpret abstract text descriptions is crucial for ensuring high-quality teaching. However, current retrieval models primarily focus on natural text-image retrieval, making them insufficiently tailored to educational scenarios due to the ambiguities in the retrieval process. In this paper, we propose a diverse expression retrieval task tailored to educational scenarios, supporting retrieval based on multiple query styles and expressions. We introduce the STEM Education Retrieval Dataset (SER), which contains over 24,000 query pairs of different styles, and the Uni-Retrieval, an efficient and style-diversified retrieval vision-language model based on prompt tuning. Uni-Retrieval extracts query style features as prototypes and builds a continuously updated Prompt Bank containing prompt tokens for diverse queries. This bank can updated during test time to represent domain-specific knowledge for different subject retrieval scenarios. Our framework demonstrates scalability and robustness by dynamically retrieving prompt tokens based on prototype similarity, effectively facilitating learning for unknown queries. Experimental results indicate that Uni-Retrieval outperforms existing retrieval models in most retrieval tasks. This advancement provides a scalable and precise solution for diverse educational needs.

Paperid: 3517, https://arxiv.org/pdf/2502.03230.pdf

Abstract:
This paper presents the HFUT-LMC team's solution to the WWW 2025 challenge on Text-based Person Anomaly Search (TPAS). The primary objective of this challenge is to accurately identify pedestrians exhibiting either normal or abnormal behavior within a large library of pedestrian images. Unlike traditional video analysis tasks, TPAS significantly emphasizes understanding and interpreting the subtle relationships between text descriptions and visual data. The complexity of this task lies in the model's need to not only match individuals to text descriptions in massive image datasets but also accurately differentiate between search results when faced with similar descriptions. To overcome these challenges, we introduce the Similarity Coverage Analysis (SCA) strategy to address the recognition difficulty caused by similar text descriptions. This strategy effectively enhances the model's capacity to manage subtle differences, thus improving both the accuracy and reliability of the search. Our proposed solution demonstrated excellent performance in this challenge.

Paperid: 3518, https://arxiv.org/pdf/2501.14548.pdf

Abstract:
Artificial intelligence (AI) shows great potential in assisting radiologists to improve the efficiency and accuracy of medical image interpretation and diagnosis. However, a versatile AI model requires large-scale data and comprehensive annotations, which are often impractical in medical settings. Recent studies leverage radiology reports as a naturally high-quality supervision for medical images, using contrastive language-image pre-training (CLIP) to develop language-informed models for radiological image interpretation. Nonetheless, these approaches typically contrast entire images with reports, neglecting the local associations between imaging regions and report sentences, which may undermine model performance and interoperability. In this paper, we propose a fine-grained vision-language model (fVLM) for anatomy-level CT image interpretation. Specifically, we explicitly match anatomical regions of CT images with corresponding descriptions in radiology reports and perform contrastive pre-training for each anatomy individually. Fine-grained alignment, however, faces considerable false-negative challenges, mainly from the abundance of anatomy-level healthy samples and similarly diseased abnormalities. To tackle this issue, we propose identifying false negatives of both normal and abnormal samples and calibrating contrastive learning from patient-level to disease-aware pairing. We curated the largest CT dataset to date, comprising imaging and report data from 69,086 patients, and conducted a comprehensive evaluation of 54 major and important disease diagnosis tasks across 15 main anatomies. Experimental results demonstrate the substantial potential of fVLM in versatile medical image interpretation. In the zero-shot classification task, we achieved an average AUC of 81.3% on 54 diagnosis tasks, surpassing CLIP and supervised methods by 12.9% and 8.0%, respectively.

Paperid: 3519, https://arxiv.org/pdf/2501.13795.pdf

Abstract:
Existing zero-shot temporal action detection (ZSTAD) methods predominantly use fully supervised or unsupervised strategies to recognize unseen activities. However, these training-based methods are prone to domain shifts and require high computational costs, which hinder their practical applicability in real-world scenarios. In this paper, unlike previous works, we propose a training-Free Zero-shot temporal Action Detection (FreeZAD) method, leveraging existing vision-language (ViL) models to directly classify and localize unseen activities within untrimmed videos without any additional fine-tuning or adaptation. We mitigate the need for explicit temporal modeling and reliance on pseudo-label quality by designing the LOGarithmic decay weighted Outer-Inner-Contrastive Score (LogOIC) and frequency-based Actionness Calibration. Furthermore, we introduce a test-time adaptation (TTA) strategy using Prototype-Centric Sampling (PCS) to expand FreeZAD, enabling ViL models to adapt more effectively for ZSTAD. Extensive experiments on the THUMOS14 and ActivityNet-1.3 datasets demonstrate that our training-free method outperforms state-of-the-art unsupervised methods while requiring only 1/13 of the runtime. When equipped with TTA, the enhanced method further narrows the gap with fully supervised methods.

Paperid: 3520, https://arxiv.org/pdf/2501.13223.pdf

Abstract:
Vision-language models (VLMs) deliver strong zero-shot recognition but frequently inherit social biases from their training data. We systematically disentangle three design factors -- model size, training-data scale, and training-data source -- by comparing CLIP and OpenCLIP, two models that share an identical contrastive objective yet differ in encoder width and in the image-text corpora on which they are pre-trained (400M proprietary pairs vs. 400M/2B LAION). Across balanced face-analysis benchmarks, enlarging the encoder reduces gender skew in CLIP but amplifies both gender and racial skew in OpenCLIP; increasing the LAION corpus from 400M to 2B further increases OpenCLIP bias. At matched model and data budgets, substituting proprietary data with LAION improves gender fairness while increasing racial skew, underscoring data source as the primary driver of bias patterns. We also evaluate three post-hoc, test-time debiasing strategies -- Bias Prompts, Prompt Array, and SANER. Debiasing reduces but does not eliminate harm, and its effectiveness is source- and size-dependent: Bias Prompts most effectively reduce gender skew in CLIP at smaller model sizes, whereas Prompt Array and SANER more reliably reduce racial skew in OpenCLIP; scaling LAION reconfigures which method is most fair. Taken together, these findings challenge the assumption that bigger models or datasets are automatically fairer and foreground training data source as the key determinant of both bias and mitigation efficacy. We release code and evaluation scripts to enable transparent, reproducible auditing of future VLMs.

Paperid: 3521, https://arxiv.org/pdf/2501.00289.pdf

Abstract:
Diffusion models have gained tremendous success in text-to-image generation, yet still lag behind with visual understanding tasks, an area dominated by autoregressive vision-language models. We propose a large-scale and fully end-to-end diffusion model for multi-modal understanding and generation that significantly improves on existing diffusion-based multimodal models, and is the first of its kind to support the full suite of vision-language modeling capabilities. Inspired by the multimodal diffusion transformer (MM-DiT) and recent advances in discrete diffusion language modeling, we leverage a cross-modal maximum likelihood estimation framework that simultaneously trains the conditional likelihoods of both images and text jointly under a single loss function, which is back-propagated through both branches of the diffusion transformer. The resulting model is highly flexible and capable of a wide range of tasks including image generation, captioning, and visual question answering. Our model attained competitive performance compared to recent unified image understanding and generation models, demonstrating the potential of multimodal diffusion modeling as a promising alternative to autoregressive next-token prediction models.

Paperid: 3522, https://arxiv.org/pdf/2412.17052.pdf

Abstract:
The integration of Large Language Models (LLMs) and Vision-Language Models (VLMs) opens new avenues for addressing complex challenges in multimodal content analysis, particularly in biased news detection. This study introduces VLBias, a framework that leverages state-of-the-art LLMs and VLMs to detect linguistic and visual biases in news content. We present a multimodal dataset comprising textual content and corresponding images from diverse news sources. We propose a hybrid annotation framework that combines LLM-based annotations with human review to ensure high-quality labeling while reducing costs and enhancing scalability. Our evaluation compares the performance of state-of-the-art SLMs and LLMs for both modalities (text and images) and the results reveal that while SLMs are computationally efficient, LLMs demonstrate superior accuracy in identifying subtle framing and text-visual inconsistencies. Furthermore, empirical analysis shows that incorporating visual cues alongside textual data improves bias detection accuracy by 3 to 5%. This study provides a comprehensive exploration of LLMs, SLMs, and VLMs as tools for detecting multimodal biases in news content and highlights their respective strengths, limitations, and potential for future applications

Paperid: 3523, https://arxiv.org/pdf/2412.09718.pdf

Abstract:
The emergence of large pre-trained vision-language models (VLMs) represents a paradigm shift in machine learning, with unprecedented results in a broad span of visual recognition tasks. CLIP, one of the most popular VLMs, has exhibited remarkable zero-shot and transfer learning capabilities in classification. To transfer CLIP to downstream tasks, adapters constitute a parameter-efficient approach that avoids backpropagation through the large model (unlike related prompt learning methods). However, CLIP adapters have been developed to target discriminative performance, and the quality of their uncertainty estimates has been overlooked. In this work we show that the discriminative performance of state-of-the-art CLIP adapters does not always correlate with their uncertainty estimation capabilities, which are essential for a safe deployment in real-world scenarios. We also demonstrate that one of such adapters is obtained through MAP inference from a more general probabilistic framework. Based on this observation we introduce BayesAdapter, which leverages Bayesian inference to estimate a full probability distribution instead of a single point, better capturing the variability inherent in the parameter space. In a comprehensive empirical evaluation we show that our approach obtains high quality uncertainty estimates in the predictions, standing out in calibration and selective classification. Our code will be publicly available upon acceptance of the paper.

Paperid: 3524, https://arxiv.org/pdf/2412.04247.pdf

Abstract:
Supervised 3D part segmentation models are tailored for a fixed set of objects and parts, limiting their transferability to open-set, real-world scenarios. Recent works have explored vision-language models (VLMs) as a promising alternative, using multi-view rendering and textual prompting to identify object parts. However, naively applying VLMs in this context introduces several drawbacks, such as the need for meticulous prompt engineering, and fails to leverage the 3D geometric structure of objects. To address these limitations, we propose COPS, a COmprehensive model for Parts Segmentation that blends the semantics extracted from visual concepts and 3D geometry to effectively identify object parts. COPS renders a point cloud from multiple viewpoints, extracts 2D features, projects them back to 3D, and uses a novel geometric-aware feature aggregation procedure to ensure spatial and semantic consistency. Finally, it clusters points into parts and labels them. We demonstrate that COPS is efficient, scalable, and achieves zero-shot state-of-the-art performance across five datasets, covering synthetic and real-world data, texture-less and coloured objects, as well as rigid and non-rigid shapes. The code is available at https://3d-cops.github.io.

Paperid: 3525, https://arxiv.org/pdf/2412.03871.pdf

Abstract:
Beyond the success of Contrastive Language-Image Pre-training (CLIP), recent trends mark a shift toward exploring the applicability of lightweight vision-language models for resource-constrained scenarios. These models often deliver suboptimal performance when relying solely on a single image-text contrastive learning objective, spotlighting the need for more effective training mechanisms that guarantee robust cross-modal feature alignment. In this work, we propose CLIP-PING: Contrastive Language-Image Pre-training with Proximus Intrinsic Neighbors Guidance, a novel yet simple and efficient training paradigm designed to boost the performance of lightweight vision-language models with minimal computational overhead and lower data demands. CLIP-PING bootstraps unimodal features extracted from arbitrary pre-trained encoders to obtain intrinsic guidance of proximus neighbor samples, i.e., nearest-neighbor (NN) and cross nearest-neighbor (XNN). We find that extra contrastive supervision from these neighbors substantially boosts cross-modal alignment, enabling lightweight models to learn more generic features with rich semantic diversity. Extensive experiments reveal that CLIP-PING notably surpasses its peers in zero-shot generalization and cross-modal retrieval tasks. Specifically, a 5.5% gain on zero-shot ImageNet1K classification with 10.7% (I2T) and 5.7% (T2I) on Flickr30K retrieval, compared to the original CLIP when using ViT-XS image encoder trained on 3 million (image, text) pairs. Moreover, CLIP-PING showcases a strong transferability under the linear evaluation protocol across several downstream tasks.

Paperid: 3526, https://arxiv.org/pdf/2412.00440.pdf

Abstract:
In rapidly evolving field of vision-language models (VLMs), contrastive language-image pre-training (CLIP) has made significant strides, becoming foundation for various downstream tasks. However, relying on one-to-one (image, text) contrastive paradigm to learn alignment from large-scale messy web data, CLIP faces a serious myopic dilemma, resulting in biases towards monotonous short texts and shallow visual expressivity. To overcome these issues, this paper advances CLIP into one novel holistic paradigm, by updating both diverse data and alignment optimization. To obtain colorful data with low cost, we use image-to-text captioning to generate multi-texts for each image, from multiple perspectives, granularities, and hierarchies. Two gadgets are proposed to encourage textual diversity. To match such (image, multi-texts) pairs, we modify the CLIP image encoder into multi-branch, and propose multi-to-multi contrastive optimization for image-text part-to-part matching. As a result, diverse visual embeddings are learned for each image, bringing good interpretability and generalization. Extensive experiments and ablations across over ten benchmarks indicate that our holistic CLIP significantly outperforms existing myopic CLIP, including image-text retrieval, open-vocabulary classification, and dense visual tasks.

Paperid: 3527, https://arxiv.org/pdf/2411.15099.pdf

Abstract:
Large-scale multimodal representation learning successfully optimizes for zero-shot transfer at test time. Yet the standard pretraining paradigm (contrastive learning on large amounts of image-text data) does not explicitly encourage representations to support few-shot adaptation. In this work, we propose a simple, but carefully designed extension to multimodal pretraining which enables representations to accommodate additional context. Using this objective, we show that vision-language models can be trained to exhibit significantly increased few-shot adaptation: across 21 downstream tasks, we find up to four-fold improvements in test-time sample efficiency, and average few-shot adaptation gains of over 5%, while retaining zero-shot generalization performance across model scales and training durations. In particular, equipped with simple, training-free, metric-based adaptation mechanisms, our representations easily surpass more complex and expensive optimization-based schemes, vastly simplifying generalization to new domains.

Paperid: 3528, https://arxiv.org/pdf/2411.11435.pdf

Abstract:
Text logo design heavily relies on the creativity and expertise of professional designers, in which arranging element layouts is one of the most important procedures. However, this specific task has received limited attention, often overshadowed by broader layout generation tasks such as document or poster design. In this paper, we propose a Vision-Language Model (VLM)-based framework that generates content-aware text logo layouts by integrating multi-modal inputs with user-defined constraints, enabling more flexible and robust layout generation for real-world applications. We introduce two model techniques that reduce the computational cost for processing multiple glyph images simultaneously, without compromising performance. To support instruction tuning of our model, we construct two extensive text logo datasets that are five times larger than existing public datasets. In addition to geometric annotations (\textit{e.g.}, text masks and character recognition), our datasets include detailed layout descriptions in natural language, enabling the model to reason more effectively in handling complex designs and custom user inputs. Experimental results demonstrate the effectiveness of our proposed framework and datasets, outperforming existing methods on various benchmarks that assess geometric aesthetics and human preferences.

Paperid: 3529, https://arxiv.org/pdf/2411.02391.pdf

Abstract:
Autonomous agents powered by large vision and language models (VLM) have demonstrated significant potential in completing daily computer tasks, such as browsing the web to book travel and operating desktop software, which requires agents to understand these interfaces. Despite such visual inputs becoming more integrated into agentic applications, what types of risks and attacks exist around them still remain unclear. In this work, we demonstrate that VLM agents can be easily attacked by a set of carefully designed adversarial pop-ups, which human users would typically recognize and ignore. This distraction leads agents to click these pop-ups instead of performing their tasks as usual. Integrating these pop-ups into existing agent testing environments like OSWorld and VisualWebArena leads to an attack success rate (the frequency of the agent clicking the pop-ups) of 86% on average and decreases the task success rate by 47%. Basic defense techniques, such as asking the agent to ignore pop-ups or including an advertisement notice, are ineffective against the attack.

Paperid: 3530, https://arxiv.org/pdf/2411.00828.pdf

Abstract:
While today's large language models exhibit impressive abilities in generating human-like text, they require massive amounts of data during training. We here take inspiration from human cognitive development to train models in limited data conditions. Specifically we present a self-synthesis approach that iterates through four phases: Phase 1 sets up fundamental language abilities, training the model from scratch on a small corpus. Language is then associated with the visual environment in phase 2, integrating the model with a vision encoder to generate descriptive captions from labeled images. In the "self-synthesis" phase 3, the model generates captions for unlabeled images, that it then uses to further train its language component with a mix of synthetic, and previous real-world text. This phase is meant to expand the model's linguistic repertoire, similar to humans self-annotating new experiences. Finally, phase 4 develops advanced cognitive skills, by training the model on specific tasks such as visual question answering and reasoning. Our approach offers a proof of concept for training a multimodal model using a developmentally plausible amount of data.

Paperid: 3531, https://arxiv.org/pdf/2410.16232.pdf

Abstract:
Sketches are a natural and accessible medium for UI designers to conceptualize early-stage ideas. However, existing research on UI/UX automation often requires high-fidelity inputs like Figma designs or detailed screenshots, limiting accessibility and impeding efficient design iteration. To bridge this gap, we introduce Sketch2Code, a benchmark that evaluates state-of-the-art Vision Language Models (VLMs) on automating the conversion of rudimentary sketches into webpage prototypes. Beyond end-to-end benchmarking, Sketch2Code supports interactive agent evaluation that mimics real-world design workflows, where a VLM-based agent iteratively refines its generations by communicating with a simulated user, either passively receiving feedback instructions or proactively asking clarification questions. We comprehensively analyze ten commercial and open-source models, showing that Sketch2Code is challenging for existing VLMs; even the most capable models struggle to accurately interpret sketches and formulate effective questions that lead to steady improvement. Nevertheless, a user study with UI/UX experts reveals a significant preference for proactive question-asking over passive feedback reception, highlighting the need to develop more effective paradigms for multi-turn conversational agents.

Paperid: 3532, https://arxiv.org/pdf/2410.13891.pdf

Abstract:
Transferable Targeted Attacks (TTAs), which aim to deceive black-box models into predicting specific erroneous labels, face significant challenges due to severe overfitting to surrogate models. Although modifying image features to generate robust semantic patterns of the target class is a promising approach, existing methods heavily rely on large-scale additional data. This dependence undermines the fair evaluation of TTA threats, potentially leading to a false sense of security or unnecessary overreactions. In this paper, we introduce two blind measures, surrogate self-alignment and self-transferability, to analyze the effectiveness and correlations of basic transformations, to enhance data-free attacks under strict black-box constraints. Our findings challenge conventional assumptions: (1) Attacking simple scaling transformations uniquely enhances targeted transferability, outperforming other basic transformations and rivaling leading complex methods; (2) Geometric and color transformations exhibit high internal redundancy despite weak inter-category correlations. These insights drive the design and tuning of S4ST (Strong, Self-transferable, faSt, Simple Scale Transformation), which integrates dimensionally consistent scaling, complementary low-redundancy transformations, and block-wise operations. Extensive experiments on the ImageNet-Compatible dataset demonstrate that S4ST achieves a 77.7% average targeted success rate (tSuc), surpassing existing transformations (+17.2% over H-Aug with only 26% computational time) and SOTA TTA solutions (+6.2% over SASD-WS with 1.2M samples for post-training). Notably, it attains 69.6% and 55.3% average tSuc against three commercial APIs and vision-language models, respectively. This work establishes a new SOTA for TTAs, highlights their potential threats, and calls for a reevaluation of the data dependency in achieving targeted transferability.

Paperid: 3533, https://arxiv.org/pdf/2410.12816.pdf

Abstract:
Foundational Vision-Language models such as CLIP have exhibited impressive generalization in downstream tasks. However, CLIP suffers from a two-level misalignment issue, i.e., task misalignment and data misalignment, when adapting to specific tasks. Soft prompt tuning has mitigated the task misalignment, yet the data misalignment remains a challenge. To analyze the impacts of the data misalignment, we revisit the pre-training and adaptation processes of CLIP and develop a structural causal model. We discover that while we expect to capture task-relevant information for downstream tasks accurately, the task-irrelevant knowledge impacts the prediction results and hampers the modeling of the true relationships between the images and the predicted classes. As task-irrelevant knowledge is unobservable, we leverage the front-door adjustment and propose Causality-Guided Semantic Decoupling and Classification (CDC) to mitigate the interference of task-irrelevant knowledge. Specifically, we decouple semantics contained in the data of downstream tasks and perform classification based on each semantic. Furthermore, we employ the Dempster-Shafer evidence theory to evaluate the uncertainty of each prediction generated by diverse semantics. Experiments conducted in multiple different settings have consistently demonstrated the effectiveness of CDC.

Paperid: 3534, https://arxiv.org/pdf/2410.12722.pdf

Abstract:
Multimodal/vision language models (VLMs) are increasingly being deployed in healthcare settings worldwide, necessitating robust benchmarks to ensure their safety, efficacy, and fairness. Multiple-choice question and answer (QA) datasets derived from national medical examinations have long served as valuable evaluation tools, but existing datasets are largely text-only and available in a limited subset of languages and countries. To address these challenges, we present WorldMedQA-V, an updated multilingual, multimodal benchmarking dataset designed to evaluate VLMs in healthcare. WorldMedQA-V includes 568 labeled multiple-choice QAs paired with 568 medical images from four countries (Brazil, Israel, Japan, and Spain), covering original languages and validated English translations by native clinicians, respectively. Baseline performance for common open- and closed-source models are provided in the local language and English translations, and with and without images provided to the model. The WorldMedQA-V benchmark aims to better match AI systems to the diverse healthcare environments in which they are deployed, fostering more equitable, effective, and representative applications.

Paperid: 3535, https://arxiv.org/pdf/2410.08469.pdf

Abstract:
A text encoder within Vision-Language Models (VLMs) like CLIP plays a crucial role in translating textual input into an embedding space shared with images, thereby facilitating the interpretative analysis of vision tasks through natural language. Despite the varying significance of different textual elements within a sentence depending on the context, efforts to account for variation of importance in constructing text embeddings have been lacking. We propose a framework of Semantic Token Reweighting to build Interpretable text embeddings (SToRI), which incorporates controllability as well. SToRI refines the text encoding process in CLIP by differentially weighting semantic elements based on contextual importance, enabling finer control over emphasis responsive to data-driven insights and user preferences. The efficacy of SToRI is demonstrated through comprehensive experiments on few-shot image classification and image retrieval tailored to user preferences.

Paperid: 3536, https://arxiv.org/pdf/2409.10849.pdf

Abstract:
Spoken language instructions are ubiquitous in agent collaboration. However, in real-world human-robot collaboration, following human spoken instructions can be challenging due to various speaker and environmental factors, such as background noise or mispronunciation. When faced with noisy auditory inputs, humans can leverage the collaborative context in the embodied environment to interpret noisy spoken instructions and take pragmatic assistive actions. In this paper, we present a cognitively inspired neurosymbolic model, Spoken Instruction Following through Theory of Mind (SIFToM), which leverages a Vision-Language Model with model-based mental inference to enable robots to pragmatically follow human instructions under diverse speech conditions. We test SIFToM in both simulated environments (VirtualHome) and real-world human-robot collaborative settings with human evaluations. Results show that SIFToM can significantly improve the performance of a lightweight base VLM (Gemini 2.5 Flash), outperforming state-of-the-art VLMs (Gemini 2.5 Pro) and approaching human-level accuracy on challenging spoken instruction following tasks.

Paperid: 3537, https://arxiv.org/pdf/2601.16212.pdf

Abstract:
Robot foundation models are beginning to deliver on the promise of generalist robotic agents, yet progress remains constrained by the scarcity of large-scale real-world manipulation datasets. Simulation and synthetic data generation offer a scalable alternative, but their usefulness is limited by the visual domain gap between simulation and reality. In this work, we present Point Bridge, a framework that leverages unified, domain-agnostic point-based representations to unlock synthetic datasets for zero-shot sim-to-real policy transfer, without explicit visual or object-level alignment. Point Bridge combines automated point-based representation extraction via Vision-Language Models (VLMs), transformer-based policy learning, and efficient inference-time pipelines to train capable real-world manipulation agents using only synthetic data. With additional co-training on small sets of real demonstrations, Point Bridge further improves performance, substantially outperforming prior vision-based sim-and-real co-training methods. It achieves up to 44% gains in zero-shot sim-to-real transfer and up to 66% with limited real data across both single-task and multitask settings. Videos of the robot are best viewed at: https://pointbridge3d.github.io/

Paperid: 3538, https://arxiv.org/pdf/2601.16065.pdf

Abstract:
Vision-Language Action (VLA) models have shown remarkable progress in robotic manipulation by leveraging the powerful perception abilities of Vision-Language Models (VLMs) to understand environments and directly output actions. However, by default, VLA models may overly attend to image tokens in the task-irrelevant region, which we describe as 'distracting tokens'. This behavior can disturb the model from the generation of the desired action tokens in each step, affecting the success rate of tasks. In this paper, we introduce a simple yet effective plug-and-play Distracting Token Pruning (DTP) framework, which dynamically detects and prunes these distracting image tokens. By correcting the model's visual attention patterns, we aim to improve the task success rate, as well as exploring the performance upper boundaries of the model without altering its original architecture or adding additional inputs. Experiments on the SIMPLER Benchmark (Li et al., 2024) show that our method consistently achieving relative improvements in task success rates across different types of novel VLA models, demonstrating generalizability to transformer-based VLAs. Further analysis reveals a negative correlation between the task success rate and the amount of attentions in the task-irrelevant region for all models tested, highlighting a common phenomenon of VLA models that could guide future research. We also publish our code at: https://anonymous.4open.science/r/CBD3.

Paperid: 3539, https://arxiv.org/pdf/2601.03666.pdf

Abstract:
Modern information systems often involve different types of items, e.g., a text query, an image, a video clip, or an audio segment. This motivates omni-modal embedding models that map heterogeneous modalities into a shared space for direct comparison. However, most recent omni-modal embeddings still rely heavily on implicit alignment inherited from pretrained vision-language model (VLM) backbones. In practice, this causes three common issues: (i) similarity logits have modality-dependent sharpness, so scores are not on a consistent scale; (ii) in-batch negatives become less effective over time because mixed-modality batches create an imbalanced hardness distribution; as a result, many negatives quickly become trivial and contribute little gradient; and (iii) embeddings across modalities show mismatched first- and second-order statistics, which makes rankings less stable. To tackle these problems, we propose e5-omni, a lightweight explicit alignment recipe that adapts off-the-shelf VLMs into robust omni-modal embedding models. e5-omni combines three simple components: (1) modality-aware temperature calibration to align similarity scales, (2) a controllable negative curriculum with debiasing to focus on confusing negatives while reducing the impact of false negatives, and (3) batch whitening with covariance regularization to better match cross-modal geometry in the shared embedding space. Experiments on MMEB-V2 and AudioCaps show consistent gains over strong bi-modal and omni-modal baselines, and the same recipe also transfers well to other VLM backbones. We release our model checkpoint at https://huggingface.co/Haon-Chen/e5-omni-7B.

Paperid: 3540, https://arxiv.org/pdf/2601.01747.pdf

Abstract:
Recent advancements in Large Vision-Language Models (LVLMs) have shown groundbreaking capabilities across diverse multimodal tasks. However, these models remain vulnerable to adversarial jailbreak attacks, where adversaries craft subtle perturbations to bypass safety mechanisms and trigger harmful outputs. Existing white-box attacks methods require full model accessibility, suffer from computing costs and exhibit insufficient adversarial transferability, making them impractical for real-world, black-box settings. To address these limitations, we propose a black-box jailbreak attack on LVLMs via Zeroth-Order optimization using Simultaneous Perturbation Stochastic Approximation (ZO-SPSA). ZO-SPSA provides three key advantages: (i) gradient-free approximation by input-output interactions without requiring model knowledge, (ii) model-agnostic optimization without the surrogate model and (iii) lower resource requirements with reduced GPU memory consumption. We evaluate ZO-SPSA on three LVLMs, including InstructBLIP, LLaVA and MiniGPT-4, achieving the highest jailbreak success rate of 83.0% on InstructBLIP, while maintaining imperceptible perturbations comparable to white-box methods. Moreover, adversarial examples generated from MiniGPT-4 exhibit strong transferability to other LVLMs, with ASR reaching 64.18%. These findings underscore the real-world feasibility of black-box jailbreaks and expose critical weaknesses in the safety mechanisms of current LVLMs

Paperid: 3541, https://arxiv.org/pdf/2601.00156.pdf

Abstract:
In this paper, we introduce an underexplored problem in facial analysis: generating and recognizing multi-attribute natural language descriptions, containing facial action units (AUs), emotional states, and age estimation, for arbitrarily selected face regions (termed FaceFocalDesc). We argue that the system's ability to focus on individual facial areas leads to better understanding and control. To achieve this capability, we construct a new multi-attribute description dataset for arbitrarily selected face regions, providing rich region-level annotations and natural language descriptions. Further, we propose a fine-tuned vision-language model based on Qwen2.5-VL, called Focal-RegionFace for facial state analysis, which incrementally refines its focus on localized facial features through multiple progressively fine-tuning stages, resulting in interpretable age estimation, FAU and emotion detection. Experimental results show that Focal-RegionFace achieves the best performance on the new benchmark in terms of traditional and widely used metrics, as well as new proposed metrics. This fully verifies its effectiveness and versatility in fine-grained multi-attribute face region-focal analysis scenarios.

Paperid: 3542, https://arxiv.org/pdf/2512.14336.pdf

Abstract:
Scalable Vector Graphics (SVG) are central to modern web design, and the demand to animate them continues to grow as web environments become increasingly dynamic. Yet automating the animation of vector graphics remains challenging for vision-language models (VLMs) despite recent progress in code generation and motion planning. VLMs routinely mis-handle SVGs, since visually coherent parts are often fragmented into low-level shapes that offer little guidance of which elements should move together. In this paper, we introduce a framework that recovers the semantic structure required for reliable SVG animation and reveals the missing layer that current VLM systems overlook. This is achieved through a statistical aggregation of multiple weak part predictions, allowing the system to stably infer semantics from noisy predictions. By reorganizing SVGs into semantic groups, our approach enables VLMs to produce animations with far greater coherence. Our experiments demonstrate substantial gains over existing approaches, suggesting that semantic recovery is the key step that unlocks robust SVG animation and supports more interpretable interactions between VLMs and vector graphics.

Paperid: 3543, https://arxiv.org/pdf/2512.08881.pdf

Abstract:
Vision-language models (VLMs) are emerging as powerful generalist tools for remote sensing, capable of integrating information across diverse tasks and enabling flexible, instruction-based interactions via a chat interface. In this work, we enhance VLM-based visual grounding in satellite imagery by proposing a novel structured localization mechanism. Our approach involves finetuning a pretrained VLM on a diverse set of instruction-following tasks, while interfacing a dedicated grounding module through specialized control tokens for localization. This method facilitates joint reasoning over both language and spatial information, significantly enhancing the model's ability to precisely localize objects in complex satellite scenes. We evaluate our framework on several remote sensing benchmarks, consistently improving the state-of-the-art, including a 24.8% relative improvement over previous methods on visual grounding. Our results highlight the benefits of integrating structured spatial reasoning into VLMs, paving the way for more reliable real-world satellite data analysis.

Paperid: 3544, https://arxiv.org/pdf/2512.03874.pdf

Abstract:
Dexterous grasp generation aims to produce grasp poses that align with task requirements and human interpretable grasp semantics. However, achieving semantically controllable dexterous grasp synthesis remains highly challenging due to the lack of unified modeling of multiple semantic dimensions, including grasp taxonomy, contact semantics, and functional affordance. To address these limitations, we present OmniDexVLG, a multimodal, semantics aware grasp generation framework capable of producing structurally diverse and semantically coherent dexterous grasps under joint language and visual guidance. Our approach begins with OmniDexDataGen, a semantic rich dexterous grasp dataset generation pipeline that integrates grasp taxonomy guided configuration sampling, functional affordance contact point sampling, taxonomy aware differential force closure grasp sampling, and physics based optimization and validation, enabling systematic coverage of diverse grasp types. We further introduce OmniDexReasoner, a multimodal grasp type semantic reasoning module that leverages multi agent collaboration, retrieval augmented generation, and chain of thought reasoning to infer grasp related semantics and generate high quality annotations that align language instructions with task specific grasp intent. Building upon these components, we develop a unified Vision Language Grasping generation model that explicitly incorporates grasp taxonomy, contact structure, and functional affordance semantics, enabling fine grained control over grasp synthesis from natural language instructions. Extensive experiments in simulation and real world object grasping and ablation studies demonstrate that our method substantially outperforms state of the art approaches in terms of grasp diversity, contact semantic diversity, functional affordance diversity, and semantic consistency.

Paperid: 3545, https://arxiv.org/pdf/2512.01773.pdf

Abstract:
The rise of generalist robotic policies has created an exponential demand for large-scale training data. However, on-robot data collection is labor-intensive and often limited to specific environments. In contrast, open-world images capture a vast diversity of real-world scenes that naturally align with robotic manipulation tasks, offering a promising avenue for low-cost, large-scale robot data acquisition. Despite this potential, the lack of associated robot actions hinders the practical use of open-world images for robot learning, leaving this rich visual resource largely unexploited. To bridge this gap, we propose IGen, a framework that scalably generates realistic visual observations and executable actions from open-world images. IGen first converts unstructured 2D pixels into structured 3D scene representations suitable for scene understanding and manipulation. It then leverages the reasoning capabilities of vision-language models to transform scene-specific task instructions into high-level plans and generate low-level actions as SE(3) end-effector pose sequences. From these poses, it synthesizes dynamic scene evolution and renders temporally coherent visual observations. Experiments validate the high quality of visuomotor data generated by IGen, and show that policies trained solely on IGen-synthesized data achieve performance comparable to those trained on real-world data. This highlights the potential of IGen to support scalable data generation from open-world images for generalist robotic policy training.

Paperid: 3546, https://arxiv.org/pdf/2511.23031.pdf

Abstract:
Recent advances in vision-language reasoning underscore the importance of thinking with images, where models actively ground their reasoning in visual evidence. Yet, prevailing frameworks treat visual actions as optional tools, boosting metrics but leaving reasoning ungrounded and crops ineffective. This gap gives rise to the illusion of thinking with images: models seem visually grounded but rely on context-agnostic actions that neither refine perception nor guide reasoning toward correct answers. We address this problem by reframing visual actions as core reasoning primitives rather than optional tools, which we term visual rationalization, the visual analogue of textual Chain-of-Thought. Building on this insight, we propose Visual Rationale Learning (ViRL), an end-to-end paradigm that grounds training in the visual rationale itself. ViRL integrates (1) Process Supervision with ground-truth rationales, (2) Objective Alignment via step-level reward shaping, and (3) Fine-Grained Credit Assignment to distinguish correct, redundant, and erroneous actions. By ensuring each action contributes meaningfully to the reasoning chain, ViRL enables models to "get the right answer for the right visual reason". Trained purely with end-to-end RL, ViRL achieves state-of-the-art results across benchmarks spanning perception, hallucination, and reasoning. This work establishes visual rationalization as a task-agnostic, process-grounded paradigm for building transparent, verifiable, and trustworthy vision-language models.

Paperid: 3547, https://arxiv.org/pdf/2511.19647.pdf

Abstract:
Foundation models (FM) have unlocked powerful zero-shot capabilities in vision and language, yet their reliance on internet pretraining data leaves them brittle in unstructured, real-world settings. The messy, real-world data encountered during deployment (e.g. occluded or multilingual text) remains massively underrepresented in existing corpora. Robots, as embodied agents, are uniquely positioned to close this gap: they can act in physical environments to collect large-scale, real-world data that enriches FM training with precisely the examples current models lack. We introduce the Robot-Powered Data Flywheel, a framework that transforms robots from FM consumers into data generators. By deploying robots equipped with FMs in the wild, we enable a virtuous cycle: robots perform useful tasks while collecting real-world data that improves both domain-specific adaptation and domain-adjacent generalization. We instantiate this framework with Scanford, a mobile manipulator deployed in the East Asia Library for 2 weeks. Scanford autonomously scans shelves, identifies books using a vision-language model (VLM), and leverages the library catalog to label images without human annotation. This deployment both aids librarians and produces a dataset to finetune the underlying VLM, improving performance on the domain-specific in-the-wild library setting and on domain-adjacent multilingual OCR benchmarks. Using data collected from 2103 shelves, Scanford improves VLM performance on book identification from 32.0% to 71.8% and boosts domain-adjacent multilingual OCR from 24.8% to 46.6% (English) and 30.8% to 38.0% (Chinese), while saving an ~18.7 hrs of human time. These results highlight how robot-powered data flywheels can both reduce human effort in real deployments and unlock new pathways for continually adapting FMs to the messiness of reality. More details are at: https://scanford-robot.github.io

Paperid: 3548, https://arxiv.org/pdf/2511.17209.pdf

Abstract:
We introduce SPECTRE, a fully transformer-based foundation model for volumetric computed tomography (CT). Our Self-Supervised & Cross-Modal Pretraining for CT Representation Extraction (SPECTRE) approach utilizes scalable 3D Vision Transformer architectures and modern self-supervised and vision-language pretraining strategies to learn general-purpose CT representations. Volumetric CT poses unique challenges, such as extreme token scaling, geometric anisotropy, and weak or noisy clinical supervision, that make standard transformer and contrastive learning recipes ineffective out of the box. The framework jointly optimizes a local transformer for high-resolution volumetric feature extraction and a global transformer for whole-scan context modeling, making large-scale 3D attention computationally tractable. Notably, SPECTRE is trained exclusively on openly available CT datasets, demonstrating that high-performing, generalizable representations can be achieved without relying on private data. Pretraining combines DINO-style self-distillation with SigLIP-based vision-language alignment using paired radiology reports, yielding features that are both geometrically consistent and clinically meaningful. Across multiple CT benchmarks, SPECTRE consistently outperforms prior CT foundation models in both zero-shot and fine-tuned settings, establishing SPECTRE as a scalable, open, and fully transformer-based foundation model for 3D medical imaging.

Paperid: 3549, https://arxiv.org/pdf/2511.16054.pdf

Abstract:
Controlled language generation conditions text on sequence-level constraints (for example, syntax, style, or safety). These constraints may depend on future tokens, which makes directly conditioning an autoregressive language model (LM) generally intractable. Prior work uses tractable surrogates such as hidden Markov models (HMMs) to approximate the distribution over continuations and adjust the model's next-token logits at decoding time. However, we find that these surrogates are often weakly context aware, which reduces query quality. We propose Learning to Look Ahead (LTLA), a hybrid approach that pairs the same base language model for rich prefix encoding with a fixed tractable surrogate model that computes exact continuation probabilities. Two efficiency pitfalls arise when adding neural context: (i) naively rescoring the prefix with every candidate next token requires a sweep over the entire vocabulary at each step, and (ii) predicting fresh surrogate parameters for each prefix, although tractable at a single step, forces recomputation of future probabilities for every new prefix and eliminates reuse. LTLA avoids both by using a single batched HMM update to account for all next-token candidates at once, and by conditioning only the surrogate's latent state prior on the LM's hidden representations while keeping the surrogate decoder fixed, so computations can be reused across prefixes. Empirically, LTLA attains higher conditional likelihood than an unconditional HMM, approximates continuation distributions for vision-language models where a standalone HMM cannot encode visual context, and improves constraint satisfaction at comparable fluency on controlled-generation tasks, with minimal inference overhead.

Paperid: 3550, https://arxiv.org/pdf/2511.13005.pdf

Abstract:
Large vision-language models, such as CLIP, have shown strong zero-shot classification performance by aligning images and text in a shared embedding space. However, CLIP models often develop multimodal spurious biases, which is the undesirable tendency to rely on spurious features. For example, CLIP may infer object types in images based on frequently co-occurring backgrounds rather than the object's core features. This bias significantly impairs the robustness of pre-trained CLIP models on out-of-distribution data, where such cross-modal associations no longer hold. Existing methods for mitigating multimodal spurious bias typically require fine-tuning on downstream data or prior knowledge of the bias, which undermines the out-of-the-box usability of CLIP. In this paper, we first theoretically analyze the impact of multimodal spurious bias in zero-shot classification. Based on this insight, we propose Spuriousness-Aware Guided Exploration (SAGE), a simple and effective method that mitigates spurious bias through guided prompt selection. SAGE requires no training, fine-tuning, or external annotations. It explores a space of prompt templates and selects the prompts that induce the largest semantic separation between classes, thereby improving worst-group robustness. Extensive experiments on four real-world benchmark datasets and five popular backbone models demonstrate that SAGE consistently improves zero-shot performance and generalization, outperforming previous zero-shot approaches without any external knowledge or model updates.

Paperid: 3551, https://arxiv.org/pdf/2511.09907.pdf

Abstract:
Data synthesis for training large reasoning models offers a scalable alternative to limited, human-curated datasets, enabling the creation of high-quality data. However, existing approaches face several challenges: (i) indiscriminate generation that ignores the solver's ability and yields low-value problems, or reliance on complex data pipelines to balance problem difficulty; and (ii) a lack of reasoning in problem generation, leading to shallow problem variants. In this paper, we develop a problem generator that reasons explicitly to plan problem directions before synthesis and adapts difficulty to the solver's ability. Specifically, we construct related problem pairs and augment them with intermediate problem-design CoT produced by a reasoning model. These data bootstrap problem-design strategies from the generator. Then, we treat the solver's feedback on synthetic problems as a reward signal, enabling the generator to calibrate difficulty and produce complementary problems near the edge of the solver's competence. Extensive experiments on 10 mathematical and general reasoning benchmarks show that our method achieves an average improvement of 2.5% and generalizes to both language and vision-language models. Moreover, a solver trained on the synthesized data provides improved rewards for continued generator training, enabling co-evolution and yielding a further 0.7% performance gain. Our code will be made publicly available here.

Paperid: 3552, https://arxiv.org/pdf/2511.09516.pdf

Abstract:
Pre-trained Vision-Language-Action (VLA) models have achieved remarkable success in improving robustness and generalization for end-to-end robotic manipulation. However, these models struggle with long-horizon tasks due to their lack of memory and reliance solely on immediate sensory inputs. To address this limitation, we propose Memory-Augmented Prompting for Vision-Language-Action model (MAP-VLA), a novel framework that empowers pre-trained VLA models with demonstration-derived memory prompts to augment action generation for long-horizon robotic manipulation tasks. To achieve this, MAP-VLA first constructs a memory library from historical demonstrations, where each memory unit captures information about a specific stage of a task. These memory units are implemented as learnable soft prompts optimized through prompt tuning. Then, during real-time task execution, MAP-VLA retrieves relevant memory through trajectory similarity matching and dynamically integrates it into the VLA model for augmented action generation. Importantly, this prompt tuning and retrieval augmentation approach operates as a plug-and-play module for a frozen VLA model, offering a lightweight and flexible solution to improve task performance. Experimental results show that MAP-VLA delivers up to 7.0% absolute performance gains in the simulation benchmark and 25.0% on real robot evaluations for long-horizon tasks, surpassing the current state-of-the-art methods.

Paperid: 3553, https://arxiv.org/pdf/2511.03001.pdf

Abstract:
Despite recent progress in using Large Language Models (LLMs) for automatically generating 3D scenes, generated scenes often lack realistic spatial layouts and object attributes found in real-world environments. As this problem stems from insufficiently detailed, coarse-grained instructions, advancing 3D scene synthesis guided by more detailed, fine-grained instructions that reflect real-world environments becomes crucial. Without such realistic scenes, training embodied agents in unrealistic environments can lead them to learn priors that diverge significantly from real-world physics and semantics, degrading their performance when deployed. Thus, verifying the alignment between the fine-grained instruction and the generated scene is essential for effective learning. However, current evaluation methods, such as CLIPScore and vision-language models (VLMs), often fail to reliably assess such alignment. This shortcoming arises primarily from their shallow understanding of 3D scenes, which often leads to improperly grounded scene components. To address this, we introduce LEGO-Eval, an evaluation framework equipped with diverse tools designed to explicitly ground scene components, enabling more accurate alignment assessments. We also present LEGO-Bench, a benchmark of detailed instructions that specify complex layouts and attributes of real-world environments. Experiments demonstrate that LEGO-Eval outperforms VLM-as-a-judge by 0.41 F1 score in assessing scene-instruction alignment. Benchmarking with LEGO-Bench reveals significant limitations in current generation methods. Across all evaluated approaches, success rates reached at most 10% in generating scenes that fully align with fine-grained instructions.

Paperid: 3554, https://arxiv.org/pdf/2511.02832.pdf

Abstract:
Large-scale data has driven breakthroughs in robotics, from language models to vision-language-action models in bimanual manipulation. However, humanoid robotics lacks equally effective data collection frameworks. Existing humanoid teleoperation systems either use decoupled control or depend on expensive motion capture setups. We introduce TWIST2, a portable, mocap-free humanoid teleoperation and data collection system that preserves full whole-body control while advancing scalability. Our system leverages PICO4U VR for obtaining real-time whole-body human motions, with a custom 2-DoF robot neck (cost around $250) for egocentric vision, enabling holistic human-to-humanoid control. We demonstrate long-horizon dexterous and mobile humanoid skills and we can collect 100 demonstrations in 15 minutes with an almost 100% success rate. Building on this pipeline, we propose a hierarchical visuomotor policy framework that autonomously controls the full humanoid body based on egocentric vision. Our visuomotor policy successfully demonstrates whole-body dexterous manipulation and dynamic kicking tasks. The entire system is fully reproducible and open-sourced at https://yanjieze.com/TWIST2 . Our collected dataset is also open-sourced at https://twist-data.github.io .

Paperid: 3555, https://arxiv.org/pdf/2510.26006.pdf

Abstract:
Humans can naturally identify, reason about, and explain anomalies in their environment. In computer vision, this long-standing challenge remains limited to industrial defects or unrealistic, synthetically generated anomalies, failing to capture the richness and unpredictability of real-world anomalies. In this work, we introduce CAVE, the first benchmark of real-world visual anomalies. CAVE supports three open-ended tasks: anomaly description, explanation, and justification; with fine-grained annotations for visual grounding and categorizing anomalies based on their visual manifestations, their complexity, severity, and commonness. These annotations draw inspiration from cognitive science research on how humans identify and resolve anomalies, providing a comprehensive framework for evaluating Vision-Language Models (VLMs) in detecting and understanding anomalies. We show that state-of-the-art VLMs struggle with visual anomaly perception and commonsense reasoning, even with advanced prompting strategies. By offering a realistic and cognitively grounded benchmark, CAVE serves as a valuable resource for advancing research in anomaly detection and commonsense reasoning in VLMs.

Paperid: 3556, https://arxiv.org/pdf/2510.21606.pdf

Abstract:
Cross-modal alignment aims to map heterogeneous modalities into a shared latent space, as exemplified by models like CLIP, which benefit from large-scale image-text pretraining for strong recognition capabilities. However, when operating in resource-constrained settings with limited or low-quality data, these models often suffer from overconfidence and degraded performance due to the prevalence of ambiguous or weakly correlated image-text pairs. Current contrastive learning approaches, which rely on single positive pairs, further exacerbate this issue by reinforcing overconfidence on uncertain samples. To address these challenges, we propose Modest-Align, a lightweight alignment framework designed for robustness and efficiency. Our approach leverages two complementary strategies -- Random Perturbation, which introduces controlled noise to simulate uncertainty, and Embedding Smoothing, which calibrates similarity distributions in the embedding space. These mechanisms collectively reduce overconfidence and improve performance on noisy or weakly aligned samples. Extensive experiments across multiple benchmark datasets demonstrate that Modest-Align outperforms state-of-the-art methods in retrieval tasks, achieving competitive results with over 100x less training data and 600x less GPU time than CLIP. Our method offers a practical and scalable solution for cross-modal alignment in real-world, low-resource scenarios.

Paperid: 3557, https://arxiv.org/pdf/2510.11650.pdf

Abstract:
Generating realistic and controllable 3D human avatars is a long-standing challenge, particularly when covering broad attribute ranges such as ethnicity, age, clothing styles, and detailed body shapes. Capturing and annotating large-scale human datasets for training generative models is prohibitively expensive and limited in scale and diversity. The central question we address in this paper is: Can existing foundation models be distilled to generate theoretically unbounded, richly annotated 3D human data? We introduce InfiniHuman, a framework that synergistically distills these models to produce richly annotated human data at minimal cost and with theoretically unlimited scalability. We propose InfiniHumanData, a fully automatic pipeline that leverages vision-language and image generation models to create a large-scale multi-modal dataset. User study shows our automatically generated identities are undistinguishable from scan renderings. InfiniHumanData contains 111K identities spanning unprecedented diversity. Each identity is annotated with multi-granularity text descriptions, multi-view RGB images, detailed clothing images, and SMPL body-shape parameters. Building on this dataset, we propose InfiniHumanGen, a diffusion-based generative pipeline conditioned on text, body shape, and clothing assets. InfiniHumanGen enables fast, realistic, and precisely controllable avatar generation. Extensive experiments demonstrate significant improvements over state-of-the-art methods in visual quality, generation speed, and controllability. Our approach enables high-quality avatar generation with fine-grained control at effectively unbounded scale through a practical and affordable solution. We will publicly release the automatic data generation pipeline, the comprehensive InfiniHumanData dataset, and the InfiniHumanGen models at https://yuxuan-xue.com/infini-human.

Paperid: 3558, https://arxiv.org/pdf/2510.11296.pdf

Abstract:
Recent approaches for vision-language models (VLMs) have shown remarkable success in achieving fast downstream adaptation. When applied to real-world downstream tasks, VLMs inevitably encounter both the in-distribution (ID) data and out-of-distribution (OOD) data. The OOD datasets often include both covariate shifts (e.g., known classes with changes in image styles) and semantic shifts (e.g., test-time unseen classes). This highlights the importance of improving VLMs' generalization ability to covariate-shifted OOD data, while effectively detecting open-set semantic-shifted OOD classes. In this paper, inspired by the substantial energy change observed in closed-set data when re-aligning vision-language modalities (specifically by directly reducing the maximum cosine similarity to a low value), we introduce a novel OOD score, named ΔEnergy. ΔEnergy significantly outperforms the vanilla energy-based OOD score and provides a more reliable approach for OOD detection. Furthermore, ΔEnergy can simultaneously improve OOD generalization under covariate shifts, which is achieved by lower-bound maximization for ΔEnergy (termed EBM). EBM is theoretically proven to not only enhance OOD detection but also yields a domain-consistent Hessian, which serves as a strong indicator for OOD generalization. Based on this finding, we developed a unified fine-tuning framework that allows for improving VLMs' robustness in both OOD generalization and OOD detection. Extensive experiments on challenging OOD detection and generalization benchmarks demonstrate the superiority of our method, outperforming recent approaches by 10% to 25% in AUROC.

Paperid: 3559, https://arxiv.org/pdf/2510.09867.pdf

Abstract:
Vision-language models (VLMs) such as CLIP achieve zero-shot transfer across various tasks by pre-training on numerous image-text pairs. These models often benefit from using an ensemble of context prompts to represent a class. Despite being effective, conventional prompt ensembling that averages textual features of context prompts often yields suboptimal results. This is because feature averaging shifts the class centroids away from the true class distribution. To address this issue, we propose the Cluster-Aware Prompt Ensemble Learning (CAPEL) framework, which preserves the cluster nature of context prompts. CAPEL classifies images into one of several class clusters, each represented by a distinct prompt. Instead of ensembling prompts in the feature space, we perform ensembling in the classification logits space, aligning better with the visual feature distribution. To further optimize prompt fine-tuning while maintaining cluster-specific discriminative power, we introduce a cluster-preserving regularization term. This ensures that prompts remain distinct and specialized for different clusters, preventing collapse into a uniform direction. Additionally, we integrate an adaptive prompt weighting technique to dynamically adjust the attention weights for flawed or ambiguous prompts, ensuring robust performance across diverse datasets and tasks.

Paperid: 3560, https://arxiv.org/pdf/2510.01681.pdf

Abstract:
Vision-Language Models (VLMs) excel at many multimodal tasks, yet they frequently struggle with tasks requiring precise understanding and handling of fine-grained visual elements. This is mainly due to information loss during image encoding or insufficient attention to critical regions. Recent work has shown promise by incorporating pixel-level visual information into the reasoning process, enabling VLMs to access high-resolution visual details during their thought process. However, this pixel-level information is often overused, leading to inefficiency and distraction from irrelevant visual details. To address these challenges, we propose the first framework for adaptive pixel reasoning that dynamically determines necessary pixel-level operations based on the input query. Specifically, we first apply operation-aware supervised fine-tuning to establish baseline competence in textual reasoning and visual operations, then design a novel rollout-guided reinforcement learning framework relying on feedback of the model's own responses, which enables the VLM to determine when pixel operations should be invoked based on query difficulty. Experiments on extensive multimodal reasoning benchmarks show that our model achieves superior performance while significantly reducing unnecessary visual operations. Impressively, our model achieves 73.4\% accuracy on HR-Bench 4K while maintaining a tool usage ratio of only 20.1\%, improving accuracy and simultaneously reducing tool usage by 66.5\% compared to the previous methods.

Paperid: 3561, https://arxiv.org/pdf/2510.00072.pdf

Abstract:
We introduce Geo-R1, a reasoning-centric post-training framework that unlocks geospatial reasoning in vision-language models by combining thinking scaffolding and elevating. In the scaffolding stage, Geo-R1 instills a ``geospatial thinking paradigm" via supervised fine-tuning on synthetic chain-of-thought exemplars, enabling models to connect visual cues with geographic priors without costly human reasoning annotations. In the elevating stage, it uses GRPO-based reinforcement learning on a weakly-supervised cross-view pairing proxy. This design supplies a verifiable and scalable reward signal: teaching models to capture and reconcile features across modalities, and harnessing reasoning for accurate prediction. Geo-R1 extends geospatial modeling from domain pretraining / supervised finetuning to reasoning-first post-training, and achieves state-of-the-art performance across various geospatial reasoning benchmarks. Our model is available at https://huggingface.co/miniHui/Geo-R1.

Paperid: 3562, https://arxiv.org/pdf/2509.25794.pdf

Abstract:
Vision-Language Models (VLMs) have demonstrated impressive world knowledge across a wide range of tasks, making them promising candidates for embodied reasoning applications. However, existing benchmarks primarily evaluate the embodied reasoning ability of VLMs through multiple-choice questions based on image annotations -- for example, selecting which trajectory better describes an event in the image. In this work, we introduce the Point-It-Out (PIO) benchmark, a novel benchmark designed to systematically assess the embodied reasoning abilities of VLMs through precise visual grounding. We propose a hierarchical evaluation protocol spanning three stages (S1: referred-object localization, S2: task-driven pointing, and S3: visual trace prediction), with data collected from critical domains for embodied intelligence, including indoor, kitchen, driving, and robotic manipulation scenarios. Extensive experiments with over ten state-of-the-art VLMs reveal several interesting findings. For example, strong general-purpose models such as GPT-4o, while excelling on many benchmarks (e.g., language, perception, and reasoning), underperform compared to some open-source models in precise visual grounding; models such as MoLMO perform well in S1 and S2 but struggle in S3, where requires grounding combined with visual trace planning.

Paperid: 3563, https://arxiv.org/pdf/2509.24321.pdf

Abstract:
Understanding human instructions and accomplishing Vision-Language Navigation tasks in unknown environments is essential for robots. However, existing modular approaches heavily rely on the quality of training data and often exhibit poor generalization. Vision-Language Model based methods, while demonstrating strong generalization capabilities, tend to perform unsatisfactorily when semantic cues are weak. To address these issues, this paper proposes SONAR, an aggregated reasoning approach through a cross modal paradigm. The proposed method integrates a semantic map based target prediction module with a Vision-Language Model based value map module, enabling more robust navigation in unknown environments with varying levels of semantic cues, and effectively balancing generalization ability with scene adaptability. In terms of target localization, we propose a strategy that integrates multi-scale semantic maps with confidence maps, aiming to mitigate false detections of target objects. We conducted an evaluation of the SONAR within the Gazebo simulator, leveraging the most challenging Matterport 3D (MP3D) dataset as the experimental benchmark. Experimental results demonstrate that SONAR achieves a success rate of 38.4% and an SPL of 17.7%.

Paperid: 3564, https://arxiv.org/pdf/2509.22650.pdf

Abstract:
Most existing approaches to referring segmentation achieve strong performance only through fine-tuning or by composing multiple pre-trained models, often at the cost of additional training and architectural modifications. Meanwhile, large-scale generative diffusion models encode rich semantic information, making them attractive as general-purpose feature extractors. In this work, we introduce a new method that directly exploits features, attention scores, from diffusion transformers for downstream tasks, requiring neither architectural modifications nor additional training. To systematically evaluate these features, we extend benchmarks with vision-language grounding tasks spanning both images and videos. Our key insight is that stop words act as attention magnets: they accumulate surplus attention and can be filtered to reduce noise. Moreover, we identify global attention sinks (GAS) emerging in deeper layers and show that they can be safely suppressed or redirected onto auxiliary tokens, leading to sharper and more accurate grounding maps. We further propose an attention redistribution strategy, where appended stop words partition background activations into smaller clusters, yielding sharper and more localized heatmaps. Building on these findings, we develop RefAM, a simple training-free grounding framework that combines cross-attention maps, GAS handling, and redistribution. Across zero-shot referring image and video segmentation benchmarks, our approach consistently outperforms prior methods, establishing a new state of the art without fine-tuning or additional components.

Paperid: 3565, https://arxiv.org/pdf/2509.21400.pdf

Abstract:
As the capabilities of Vision Language Models (VLMs) continue to improve, they are increasingly targeted by jailbreak attacks. Existing defense methods face two major limitations: (1) they struggle to ensure safety without compromising the model's utility; and (2) many defense mechanisms significantly reduce the model's inference efficiency. To address these challenges, we propose SafeSteer, a lightweight, inference-time steering framework that effectively defends against diverse jailbreak attacks without modifying model weights. At the core of SafeSteer is the innovative use of Singular Value Decomposition to construct a low-dimensional "safety subspace." By projecting and reconstructing the raw steering vector into this subspace during inference, SafeSteer adaptively removes harmful generation signals while preserving the model's ability to handle benign inputs. The entire process is executed in a single inference pass, introducing negligible overhead. Extensive experiments show that SafeSteer reduces the attack success rate by over 60% and improves accuracy on normal tasks by 1-2%, without introducing significant inference latency. These results demonstrate that robust and practical jailbreak defense can be achieved through simple, efficient inference-time control.

Paperid: 3566, https://arxiv.org/pdf/2509.20739.pdf

Abstract:
Conventional SLAM pipelines for legged robot navigation are fragile under rapid motion, calibration demands, and sensor drift, while offering limited semantic reasoning for task-driven exploration. To deal with these issues, we propose a vision-only, SLAM-free navigation framework that replaces dense geometry with semantic reasoning and lightweight topological representations. A hierarchical vision-language perception module fuses scene-level context with object-level cues for robust semantic inference. And a semantic-probabilistic topological map supports coarse-to-fine planning: LLM-based global reasoning for subgoal selection and vision-based local planning for obstacle avoidance. Integrated with reinforcement-learning locomotion controllers, the framework is deployable across diverse legged robot platforms. Experiments in simulation and real-world settings demonstrate consistent improvements in semantic accuracy, planning quality, and navigation success, while ablation studies further showcase the necessity of both hierarchical perception and fine local planning. This work introduces a new paradigm for SLAM-free, vision-language-driven navigation, shifting robotic exploration from geometry-centric mapping to semantics-driven decision making.

Paperid: 3567, https://arxiv.org/pdf/2509.20379.pdf

Abstract:
Hallucinations of vision-language models (VLMs), which are misalignments between visual content and generated text, undermine the reliability of VLMs. One common approach for detecting them employs the same VLM, or a different one, to assess generated outputs. This process is computationally intensive and increases model latency. In this paper, we explore an efficient on-the-fly method for hallucination detection by training traditional ML models over signals based on the VLM's next-token probabilities (NTPs). NTPs provide a direct quantification of model uncertainty. We hypothesize that high uncertainty (i.e., a low NTP value) is strongly associated with hallucinations. To test this, we introduce a dataset of 1,400 human-annotated statements derived from VLM-generated content, each labeled as hallucinated or not, and use it to test our NTP-based lightweight method. Our results demonstrate that NTP-based features are valuable predictors of hallucinations, enabling fast and simple ML models to achieve performance comparable to that of strong VLMs. Furthermore, augmenting these NTPs with linguistic NTPs, computed by feeding only the generated text back into the VLM, enhances hallucination detection performance. Finally, integrating hallucination prediction scores from VLMs into the NTP-based models led to better performance than using either VLMs or NTPs alone. We hope this study paves the way for simple, lightweight solutions that enhance the reliability of VLMs.

Paperid: 3568, https://arxiv.org/pdf/2509.19102.pdf

Abstract:
General-purpose robotic skills from end-to-end demonstrations often leads to task-specific policies that fail to generalize beyond the training distribution. Therefore, we introduce FunCanon, a framework that converts long-horizon manipulation tasks into sequences of action chunks, each defined by an actor, verb, and object. These chunks focus policy learning on the actions themselves, rather than isolated tasks, enabling compositionality and reuse. To make policies pose-aware and category-general, we perform functional object canonicalization for functional alignment and automatic manipulation trajectory transfer, mapping objects into shared functional frames using affordance cues from large vision language models. An object centric and action centric diffusion policy FuncDiffuser trained on this aligned data naturally respects object affordances and poses, simplifying learning and improving generalization ability. Experiments on simulated and real-world benchmarks demonstrate category-level generalization, cross-task behavior reuse, and robust sim2real deployment, showing that functional canonicalization provides a strong inductive bias for scalable imitation learning in complex manipulation domains. Details of the demo and supplemental material are available on our project website https://sites.google.com/view/funcanon.

Paperid: 3569, https://arxiv.org/pdf/2509.15607.pdf

Abstract:
Preference-based reinforcement learning (PbRL) has emerged as a promising paradigm for teaching robots complex behaviors without reward engineering. However, its effectiveness is often limited by two critical challenges: the reliance on extensive human input and the inherent difficulties in resolving query ambiguity and credit assignment during reward learning. In this paper, we introduce PRIMT, a PbRL framework designed to overcome these challenges by leveraging foundation models (FMs) for multimodal synthetic feedback and trajectory synthesis. Unlike prior approaches that rely on single-modality FM evaluations, PRIMT employs a hierarchical neuro-symbolic fusion strategy, integrating the complementary strengths of large language models and vision-language models in evaluating robot behaviors for more reliable and comprehensive feedback. PRIMT also incorporates foresight trajectory generation, which reduces early-stage query ambiguity by warm-starting the trajectory buffer with bootstrapped samples, and hindsight trajectory augmentation, which enables counterfactual reasoning with a causal auxiliary loss to improve credit assignment. We evaluate PRIMT on 2 locomotion and 6 manipulation tasks on various benchmarks, demonstrating superior performance over FM-based and scripted baselines.

Paperid: 3570, https://arxiv.org/pdf/2509.13733.pdf

Abstract:
Visual-Language Navigation (VLN) is a fundamental challenge in robotic systems, with broad applications for the deployment of embodied agents in real-world environments. Despite recent advances, existing approaches are limited in long-range spatial reasoning, often exhibiting low success rates and high inference latency, particularly in long-range navigation tasks. To address these limitations, we propose FSR-VLN, a vision-language navigation system that combines a Hierarchical Multi-modal Scene Graph (HMSG) with Fast-to-Slow Navigation Reasoning (FSR). The HMSG provides a multi-modal map representation supporting progressive retrieval, from coarse room-level localization to fine-grained goal view and object identification. Building on HMSG, FSR first performs fast matching to efficiently select candidate rooms, views, and objects, then applies VLM-driven refinement for final goal selection. We evaluated FSR-VLN across four comprehensive indoor datasets collected by humanoid robots, utilizing 87 instructions that encompass a diverse range of object categories. FSR-VLN achieves state-of-the-art (SOTA) performance in all datasets, measured by the retrieval success rate (RSR), while reducing the response time by 82% compared to VLM-based methods on tour videos by activating slow reasoning only when fast intuition fails. Furthermore, we integrate FSR-VLN with speech interaction, planning, and control modules on a Unitree-G1 humanoid robot, enabling natural language interaction and real-time navigation.

Paperid: 3571, https://arxiv.org/pdf/2509.13282.pdf

Abstract:
Charts are a crucial visual medium for communicating and representing information. While Large Vision-Language Models (LVLMs) have made progress on chart question answering (CQA), the task remains challenging, particularly when models attend to irrelevant regions of the chart. In this work, we present ChartGaze, a new eye-tracking dataset that captures human gaze patterns during chart reasoning tasks. Through a systematic comparison of human and model attention, we find that LVLMs often diverge from human gaze, leading to reduced interpretability and accuracy. To address this, we propose a gaze-guided attention refinement that aligns image-text attention with human fixations. Our approach improves both answer accuracy and attention alignment, yielding gains of up to 2.56 percentage points across multiple models. These results demonstrate the promise of incorporating human gaze to enhance both the reasoning quality and interpretability of chart-focused LVLMs.

Paperid: 3572, https://arxiv.org/pdf/2509.11465.pdf

Abstract:
We introduce CEMTM, a context-enhanced multimodal topic model designed to infer coherent and interpretable topic structures from both short and long documents containing text and images. CEMTM builds on fine-tuned large vision language models (LVLMs) to obtain contextualized embeddings, and employs a distributional attention mechanism to weight token-level contributions to topic inference. A reconstruction objective aligns topic-based representations with the document embedding, encouraging semantic consistency across modalities. Unlike existing approaches, CEMTM can process multiple images per document without repeated encoding and maintains interpretability through explicit word-topic and document-topic distributions. Extensive experiments on six multimodal benchmarks show that CEMTM consistently outperforms unimodal and multimodal baselines, achieving a remarkable average LLM score of 2.61. Further analysis shows its effectiveness in downstream few-shot retrieval and its ability to capture visually grounded semantics in complex domains such as scientific articles.

Paperid: 3573, https://arxiv.org/pdf/2509.06266.pdf

Abstract:
Understanding 3D spatial relationships remains a major limitation of current Vision-Language Models (VLMs). Prior work has addressed this issue by creating spatial question-answering (QA) datasets based on single images or indoor videos. However, real-world embodied AI agents such as robots and self-driving cars typically rely on ego-centric, multi-view observations. To this end, we introduce Ego3D-Bench, a new benchmark designed to evaluate the spatial reasoning abilities of VLMs using ego-centric, multi-view outdoor data. Ego3D-Bench comprises over 8,600 QA pairs, created with significant involvement from human annotators to ensure quality and diversity. We benchmark 16 SOTA VLMs, including GPT-4o, Gemini1.5-Pro, InternVL3, and Qwen2.5-VL. Our results reveal a notable performance gap between human level scores and VLM performance, highlighting that current VLMs still fall short of human level spatial understanding. To bridge this gap, we propose Ego3D-VLM, a post-training framework that enhances 3D spatial reasoning of VLMs. Ego3D-VLM generates cognitive map based on estimated global 3D coordinates, resulting in 12% average improvement on multi-choice QA and 56% average improvement on absolute distance estimation. Ego3D-VLM is modular and can be integrated with any existing VLM. Together, Ego3D-Bench and Ego3D-VLM offer valuable tools for advancing toward human level spatial understanding in real-world, multi-view environments.

Paperid: 3574, https://arxiv.org/pdf/2509.00192.pdf

Abstract:
Multimodal Large Language Models (MLLMs) have demonstrated remarkable capabilities in vision-language tasks. However, these models often infer and reveal sensitive biometric attributes such as race, gender, age, body weight, and eye color; even when such information is not explicitly requested. This raises critical concerns, particularly in real-world applications and socially-sensitive domains. Despite increasing awareness, no publicly available dataset or benchmark exists to comprehensively evaluate or mitigate biometric leakage in MLLMs. To address this gap, we introduce PRISM (Privacy-aware Evaluation of Responses in Sensitive Modalities), a new benchmark designed to assess MLLMs on two fronts: (1) refuse biometric-related queries and (2) implicit biometric leakage in general responses while maintaining semantic faithfulness. Further, we conduct a detailed audit of the widely used LLaVA datasets and uncover extensive biometric leakage across pretraining and instruction data. To address this, we present Safe-LLaVA dataset, the first privacy-preserving MLLM training dataset constructed by systematically removing explicit and implicit biometric information from LLaVA dataset. Our evaluations on PRISM reveal biometric leakages across MLLMs for different attributes, highlighting the detailed privacy-violations. We also fine-tune a model on Safe-LLaVA dataset and show that it substantially reduces the biometric leakages. Together, Safe-LLaVA and PRISM set a new standard for privacy-aligned development and evaluation of MLLMs.

Paperid: 3575, https://arxiv.org/pdf/2508.20325.pdf

Abstract:
As Large Language Models become increasingly integral to various domains, their potential to generate harmful responses has prompted significant societal and regulatory concerns. In response, governments have issued ethics guidelines to promote the development of trustworthy AI. However, these guidelines are typically high-level demands for developers and testers, leaving a gap in translating them into actionable testing questions to verify LLM compliance. To address this challenge, we introduce GUARD (\textbf{G}uideline \textbf{U}pholding Test through \textbf{A}daptive \textbf{R}ole-play and Jailbreak \textbf{D}iagnostics), a testing method designed to operationalize guidelines into specific guideline-violating questions that assess LLM adherence. To implement this, GUARD uses automated generation of guideline-violating questions based on government-issued guidelines, thereby testing whether responses comply with these guidelines. When responses directly violate guidelines, GUARD reports inconsistencies. Furthermore, for responses that do not directly violate guidelines, GUARD integrates the concept of ``jailbreaks'' to diagnostics, named GUARD-JD, which creates scenarios that provoke unethical or guideline-violating responses, effectively identifying potential scenarios that could bypass built-in safety mechanisms. Our method finally culminates in a compliance report, delineating the extent of adherence and highlighting any violations. We have empirically validated the effectiveness of GUARD on seven LLMs, including Vicuna-13B, LongChat-7B, Llama2-7B, Llama-3-8B, GPT-3.5, GPT-4, GPT-4o, and Claude-3.7, by testing compliance under three government-issued guidelines and conducting jailbreak diagnostics. Additionally, GUARD-JD can transfer jailbreak diagnostics to vision-language models, demonstrating its usage in promoting reliable LLM-based applications.

Paperid: 3576, https://arxiv.org/pdf/2508.16148.pdf

Abstract:
Multimodal Large Language Models (MLLMs) have demonstrated remarkable multimodal understanding capabilities in Visual Question Answering (VQA) tasks by integrating visual and textual features. However, under the challenging ten-choice question evaluation paradigm, existing methods still exhibit significant limitations when processing PDF documents with complex layouts and lengthy content. Notably, current mainstream models suffer from a strong bias toward English training data, resulting in suboptimal performance for Japanese and other language scenarios. To address these challenges, this paper proposes a novel Japanese PDF document understanding framework that combines multimodal hierarchical reasoning mechanisms with Colqwen-optimized retrieval methods, while innovatively introducing a semantic verification strategy through sub-question decomposition. Experimental results demonstrate that our framework not only significantly enhances the model's deep semantic parsing capability for complex documents, but also exhibits superior robustness in practical application scenarios.

Paperid: 3577, https://arxiv.org/pdf/2508.05430.pdf

Abstract:
Language-image pre-training (LIP) enables the development of vision-language models capable of zero-shot classification, localization, multimodal retrieval, and semantic understanding. Various explanation methods have been proposed to visualize the importance of input image-text pairs on the model's similarity outputs. However, popular saliency maps are limited by capturing only first-order attributions, overlooking the complex cross-modal interactions intrinsic to such encoders. We introduce faithful interaction explanations of LIP models (FIxLIP) as a unified approach to decomposing the similarity in vision-language encoders. FIxLIP is rooted in game theory, where we analyze how using the weighted Banzhaf interaction index offers greater flexibility and improves computational efficiency over the Shapley interaction quantification framework. From a practical perspective, we propose how to naturally extend explanation evaluation metrics, like the pointing game and area between the insertion/deletion curves, to second-order interaction explanations. Experiments on MS COCO and ImageNet-1k benchmarks validate that second-order methods like FIxLIP outperform first-order attribution methods. Beyond delivering high-quality explanations, we demonstrate the utility of FIxLIP in comparing different models like CLIP vs. SigLIP-2 and ViT-B/32 vs. ViT-L/16.

Paperid: 3578, https://arxiv.org/pdf/2508.01943.pdf

Abstract:
Vision-language models (VLMs) have exhibited impressive capabilities across diverse image understanding tasks, but still struggle in settings that require reasoning over extended sequences of camera frames from a video. This limits their utility in embodied settings, which require reasoning over long frame sequences from a continuous stream of visual input at each moment of a task attempt. To address this limitation, we propose ROVER (Reasoning Over VidEo Recursively), a framework that enables the model to recursively decompose long-horizon video trajectories into segments corresponding to shorter subtasks within the trajectory. In doing so, ROVER facilitates more focused and accurate reasoning over temporally localized frame sequences without losing global context. We evaluate ROVER, implemented using an in-context learning approach, on diverse OpenX Embodiment videos and on a new dataset derived from RoboCasa that consists of 543 videos showing both expert and perturbed non-expert trajectories across 27 robotic manipulation tasks. ROVER outperforms strong baselines across three video reasoning tasks: task progress estimation, frame-level natural language reasoning, and video question answering. We observe that, by reducing the number of frames the model reasons over at each timestep, ROVER mitigates hallucinations, especially during unexpected or non-optimal moments of a trajectory. In addition, by enabling the implementation of a subtask-specific sliding context window, ROVER's time complexity scales linearly with video length, an asymptotic improvement over baselines. Demos, code, and data available at: https://rover-vlm.github.io

Paperid: 3579, https://arxiv.org/pdf/2507.17787.pdf

Abstract:
Foundation models pre-trained on massive datasets, including large language models (LLMs), vision-language models (VLMs), and large multimodal models, have demonstrated remarkable success in diverse downstream tasks. However, recent studies have shown fundamental limitations of these models: (1) limited representational capacity, (2) lower adaptability, and (3) diminishing scalability. These shortcomings raise a critical question: is Euclidean geometry truly the optimal inductive bias for all foundation models, or could incorporating alternative geometric spaces enable models to better align with the intrinsic structure of real-world data and improve reasoning processes? Hyperbolic spaces, a class of non-Euclidean manifolds characterized by exponential volume growth with respect to distance, offer a mathematically grounded solution. These spaces enable low-distortion embeddings of hierarchical structures (e.g., trees, taxonomies) and power-law distributions with substantially fewer dimensions compared to Euclidean counterparts. Recent advances have leveraged these properties to enhance foundation models, including improving LLMs' complex reasoning ability, VLMs' zero-shot generalization, and cross-modal semantic alignment, while maintaining parameter efficiency. This paper provides a comprehensive review of hyperbolic neural networks and their recent development for foundation models. We further outline key challenges and research directions to advance the field.

Paperid: 3580, https://arxiv.org/pdf/2507.04686.pdf

Abstract:
We present MOSU, a novel autonomous long-range navigation system that enhances global navigation for mobile robots through multimodal perception and on-road scene understanding. MOSU addresses the outdoor robot navigation challenge by integrating geometric, semantic, and contextual information to ensure comprehensive scene understanding. The system combines GPS and QGIS map-based routing for high-level global path planning and multi-modal trajectory generation for local navigation refinement. For trajectory generation, MOSU leverages multi-modalities: LiDAR-based geometric data for precise obstacle avoidance, image-based semantic segmentation for traversability assessment, and Vision-Language Models (VLMs) to capture social context and enable the robot to adhere to social norms in complex environments. This multi-modal integration improves scene understanding and enhances traversability, allowing the robot to adapt to diverse outdoor conditions. We evaluate our system in real-world on-road environments and benchmark it on the GND dataset, achieving a 10% improvement in traversability on navigable terrains while maintaining a comparable navigation distance to existing global navigation methods.

Paperid: 3581, https://arxiv.org/pdf/2507.04680.pdf

Abstract:
Large Vision-Language Models (LVLMs) have demonstrated remarkable advancements in numerous areas such as multimedia. However, hallucination issues significantly limit their credibility and application potential. Existing mitigation methods typically rely on external tools or the comparison of multi-round inference, which significantly increase inference time. In this paper, we propose \textbf{SE}lf-\textbf{E}volving \textbf{D}istillation (\textbf{SEED}), which identifies hallucinations within the inner knowledge of LVLMs, isolates and purges them, and then distills the purified knowledge back into the model, enabling self-evolution. Furthermore, we identified that traditional distillation methods are prone to inducing void spaces in the output space of LVLMs. To address this issue, we propose a Mode-Seeking Evolving approach, which performs distillation to capture the dominant modes of the purified knowledge distribution, thereby avoiding the chaotic results that could emerge from void spaces. Moreover, we introduce a Hallucination Elimination Adapter, which corrects the dark knowledge of the original model by learning purified knowledge. Extensive experiments on multiple benchmarks validate the superiority of our SEED, demonstrating substantial improvements in mitigating hallucinations for representative LVLM models such as LLaVA-1.5 and InternVL2. Remarkably, the F1 score of LLaVA-1.5 on the hallucination evaluation metric POPE-Random improved from 81.3 to 88.3.

Paperid: 3582, https://arxiv.org/pdf/2506.22242.pdf

Abstract:
Leveraging diverse robotic data for pretraining remains a critical challenge. Existing methods typically model the dataset's action distribution using simple observations as inputs. However, these inputs are often incomplete, resulting in a dispersed conditional action distribution-an issue we refer to as coordinate system chaos and state chaos. This inconsistency significantly hampers pretraining efficiency. To address this, we propose 4D-VLA, a novel approach that effectively integrates 4D information into the input to mitigate these sources of chaos. Our model introduces depth and temporal information into visual features with sequential RGB-D inputs, aligning the coordinate systems of the robot and the scene. This alignment endows the model with strong spatiotemporal reasoning capabilities while minimizing training overhead. Additionally, we introduce memory bank sampling, a frame sampling strategy designed to extract informative frames from historical images, further improving effectiveness and efficiency. Experimental results demonstrate that our pretraining method and architectural components substantially enhance model performance. In both simulated and real-world experiments, our model achieves a significant increase in success rate over OpenVLA. To further assess spatial perception and generalization to novel views, we introduce MV-Bench, a multi-view simulation benchmark. Our model consistently outperforms existing methods, demonstrating stronger spatial understanding and adaptability.

Paperid: 3583, https://arxiv.org/pdf/2506.14763.pdf

Abstract:
Endowing robots with tool design abilities is critical for enabling them to solve complex manipulation tasks that would otherwise be intractable. While recent generative frameworks can automatically synthesize task settings, such as 3D scenes and reward functions, they have not yet addressed the challenge of tool-use scenarios. Simply retrieving human-designed tools might not be ideal since many tools (e.g., a rolling pin) are difficult for robotic manipulators to handle. Furthermore, existing tool design approaches either rely on predefined templates with limited parameter tuning or apply generic 3D generation methods that are not optimized for tool creation. To address these limitations, we propose RobotSmith, an automated pipeline that leverages the implicit physical knowledge embedded in vision-language models (VLMs) alongside the more accurate physics provided by physics simulations to design and use tools for robotic manipulation. Our system (1) iteratively proposes tool designs using collaborative VLM agents, (2) generates low-level robot trajectories for tool use, and (3) jointly optimizes tool geometry and usage for task performance. We evaluate our approach across a wide range of manipulation tasks involving rigid, deformable, and fluid objects. Experiments show that our method consistently outperforms strong baselines in terms of both task success rate and overall performance. Notably, our approach achieves a 50.0\% average success rate, significantly surpassing other baselines such as 3D generation (21.4%) and tool retrieval (11.1%). Finally, we deploy our system in real-world settings, demonstrating that the generated tools and their usage plans transfer effectively to physical execution, validating the practicality and generalization capabilities of our approach.

Paperid: 3584, https://arxiv.org/pdf/2506.04822.pdf

Abstract:
Although vision-language and large language models (VLM and LLM) offer promising opportunities for AI-driven educational assessment, their effectiveness in real-world classroom settings, particularly in underrepresented educational contexts, remains underexplored. In this study, we evaluated the performance of a state-of-the-art VLM and several LLMs on 646 handwritten exam responses from grade 4 students in six Indonesian schools, covering two subjects: Mathematics and English. These sheets contain more than 14K student answers that span multiple choice, short answer, and essay questions. Assessment tasks include grading these responses and generating personalized feedback. Our findings show that the VLM often struggles to accurately recognize student handwriting, leading to error propagation in downstream LLM grading. Nevertheless, LLM-generated feedback retains some utility, even when derived from imperfect input, although limitations in personalization and contextual relevance persist.

Paperid: 3585, https://arxiv.org/pdf/2506.04277.pdf

Abstract:
Multi-modal Large Language Models (MLLMs) have demonstrated remarkable reasoning capability while lack explicit mechanisms for visual grounding and segmentation, creating a gap between cognitive reasoning and visual perception. To bridge this gap, we introduce Reasoning Segmentation via Visual Prompting (RSVP), a novel framework that unifies multi-step multimodal reasoning with grounded visual understanding. RSVP is a two-stage structuralized framework that integrates reasoning-driven localization with segmentation refinement. In the reasoning stage, RSVP employs multimodal chain-of-thought visual prompts to help MLLMs understand queries and infer targets, generating interpretable region proposals that enhance visual grounding. In segmentation stage, RSVP refines these proposals with a Vision-Language Segmentation Module (VLSM), seamlessly integrates textual and visual cues to produce precise segmentation masks. By explicitly modelling the interaction between multimodal reasoning and segmentation, RSVP introduces a new paradigm for interpretable reasoning segmentation. It exploits MLLMs' inherent localization capabilities, enabling the models to not only reason about objects but also generate structured visual representations. Our extensive experiments demonstrate that RSVP achieves state-of-the-art performance, surpasses state-of-the-art methods by up to +6.5 gIoU and +9.2 cIoU on ReasonSeg, and achieves 49.7 mAP on SegInW under zero-shot settings. These results validate RSVP as an effective and scalable framework for integrating cognitive reasoning with structured visual understanding.

Paperid: 3586, https://arxiv.org/pdf/2506.01565.pdf

Abstract:
Culture is a rich and dynamic domain that evolves across both geography and time. However, existing studies on cultural understanding with vision-language models (VLMs) primarily emphasize geographic diversity, often overlooking the critical temporal dimensions. To bridge this gap, we introduce Hanfu-Bench, a novel, expert-curated multimodal dataset. Hanfu, a traditional garment spanning ancient Chinese dynasties, serves as a representative cultural heritage that reflects the profound temporal aspects of Chinese culture while remaining highly popular in Chinese contemporary society. Hanfu-Bench comprises two core tasks: cultural visual understanding and cultural image transcreation.The former task examines temporal-cultural feature recognition based on single- or multi-image inputs through multiple-choice visual question answering, while the latter focuses on transforming traditional attire into modern designs through cultural element inheritance and modern context adaptation. Our evaluation shows that closed VLMs perform comparably to non-experts on visual cutural understanding but fall short by 10\% to human experts, while open VLMs lags further behind non-experts. For the transcreation task, multi-faceted human evaluation indicates that the best-performing model achieves a success rate of only 42\%. Our benchmark provides an essential testbed, revealing significant challenges in this new direction of temporal cultural understanding and creative adaptation.

Paperid: 3587, https://arxiv.org/pdf/2505.24456.pdf

Authors:Emilio Villa-Cueva, Sholpan Bolatzhanova, Diana Turmakhan, Kareem Elzeky, Henok Biadglign Ademtew, Alham Fikri Aji, Vladimir Araujo, Israel Abebe Azime, Jinheon Baek, Frederico Belcavello, Fermin Cristobal, Jan Christian Blaise Cruz, Mary Dabre, Raj Dabre, Toqeer Ehsan, Naome A Etori, Fauzan Farooqui, Jiahui Geng, Guido Ivetta, Thanmay Jayakumar, Soyeong Jeong, Zheng Wei Lim, Aishik Mandal, Sofia Martinelli, Mihail Minkov Mihaylov, Daniil Orel, Aniket Pramanick, Sukannya Purkayastha, Israfel Salazar, Haiyue Song, Tiago Timponi Torrent, Debela Desalegn Yadeta, Injy Hamed, Atnafu Lambebo Tonja, Thamar Solorio

Abstract:
Translating cultural content poses challenges for machine translation systems due to the differences in conceptualizations between cultures, where language alone may fail to convey sufficient context to capture region-specific meanings. In this work, we investigate whether images can act as cultural context in multimodal translation. We introduce CaMMT, a human-curated benchmark of over 5,800 triples of images along with parallel captions in English and regional languages. Using this dataset, we evaluate five Vision Language Models (VLMs) in text-only and text+image settings. Through automatic and human evaluations, we find that visual context generally improves translation quality, especially in handling Culturally-Specific Items (CSIs), disambiguation, and correct gender marking. By releasing CaMMT, our objective is to support broader efforts to build and evaluate multimodal translation systems that are better aligned with cultural nuance and regional variations.

Paperid: 3588, https://arxiv.org/pdf/2505.23945.pdf

Abstract:
Chain-of-thought (CoT) reasoning enhances performance of large language models, but questions remain about whether these reasoning traces faithfully reflect the internal processes of the model. We present the first comprehensive study of CoT faithfulness in large vision-language models (LVLMs), investigating how both text-based and previously unexplored image-based biases affect reasoning and bias articulation. Our work introduces a novel, fine-grained evaluation pipeline for categorizing bias articulation patterns, enabling significantly more precise analysis of CoT reasoning than previous methods. This framework reveals critical distinctions in how models process and respond to different types of biases, providing new insights into LVLM CoT faithfulness. Our findings reveal that subtle image-based biases are rarely articulated compared to explicit text-based ones, even in models specialized for reasoning. Additionally, many models exhibit a previously unidentified phenomenon we term ``inconsistent'' reasoning - correctly reasoning before abruptly changing answers, serving as a potential canary for detecting biased reasoning from unfaithful CoTs. We then apply the same evaluation pipeline to revisit CoT faithfulness in LLMs across various levels of implicit cues. Our findings reveal that current language-only reasoning models continue to struggle with articulating cues that are not overtly stated.

Paperid: 3589, https://arxiv.org/pdf/2505.16805.pdf

Abstract:
The integration of Vision-Language Models (VLMs) into autonomous driving systems has shown promise in addressing key challenges such as learning complexity, interpretability, and common-sense reasoning. However, existing approaches often struggle with efficient integration and realtime decision-making due to computational demands. In this paper, we introduce SOLVE, an innovative framework that synergizes VLMs with end-to-end (E2E) models to enhance autonomous vehicle planning. Our approach emphasizes knowledge sharing at the feature level through a shared visual encoder, enabling comprehensive interaction between VLM and E2E components. We propose a Trajectory Chain-of-Thought (T-CoT) paradigm, which progressively refines trajectory predictions, reducing uncertainty and improving accuracy. By employing a temporal decoupling strategy, SOLVE achieves efficient cooperation by aligning high-quality VLM outputs with E2E real-time performance. Evaluated on the nuScenes dataset, our method demonstrates significant improvements in trajectory prediction accuracy, paving the way for more robust and reliable autonomous driving systems.

Paperid: 3590, https://arxiv.org/pdf/2505.05360.pdf

Abstract:
We present DSDrive, a streamlined end-to-end paradigm tailored for integrating the reasoning and planning of autonomous vehicles into a unified framework. DSDrive leverages a compact LLM that employs a distillation method to preserve the enhanced reasoning capabilities of a larger-sized vision language model (VLM). To effectively align the reasoning and planning tasks, a waypoint-driven dual-head coordination module is further developed, which synchronizes dataset structures, optimization objectives, and the learning process. By integrating these tasks into a unified framework, DSDrive anchors on the planning results while incorporating detailed reasoning insights, thereby enhancing the interpretability and reliability of the end-to-end pipeline. DSDrive has been thoroughly tested in closed-loop simulations, where it performs on par with benchmark models and even outperforms in many key metrics, all while being more compact in size. Additionally, the computational efficiency of DSDrive (as reflected in its time and memory requirements during inference) has been significantly enhanced. Evidently thus, this work brings promising aspects and underscores the potential of lightweight systems in delivering interpretable and efficient solutions for AD.

Paperid: 3591, https://arxiv.org/pdf/2505.02466.pdf

Abstract:
Recent advancements in large language models (LLMs) have driven interest in billion-scale retrieval models with strong generalization across retrieval tasks and languages. Additionally, progress in large vision-language models has created new opportunities for multimodal retrieval. In response, we have updated the Tevatron toolkit, introducing a unified pipeline that enables researchers to explore retriever models at different scales, across multiple languages, and with various modalities. This demo paper highlights the toolkit's key features, bridging academia and industry by supporting efficient training, inference, and evaluation of neural retrievers. We showcase a unified dense retriever achieving strong multilingual and multimodal effectiveness, and conduct a cross-modality zero-shot study to demonstrate its research potential. Alongside, we release OmniEmbed, to the best of our knowledge, the first embedding model that unifies text, image document, video, and audio retrieval, serving as a baseline for future research.

Paperid: 3592, https://arxiv.org/pdf/2504.14446.pdf

Abstract:
Including children's images in datasets has raised ethical concerns, particularly regarding privacy, consent, data protection, and accountability. These datasets, often built by scraping publicly available images from the Internet, can expose children to risks such as exploitation, profiling, and tracking. Despite the growing recognition of these issues, approaches for addressing them remain limited. We explore the ethical implications of using children's images in AI datasets and propose a pipeline to detect and remove such images. As a use case, we built the pipeline on a Vision-Language Model under the Visual Question Answering task and tested it on the #PraCegoVer dataset. We also evaluate the pipeline on a subset of 100,000 images from the Open Images V7 dataset to assess its effectiveness in detecting and removing images of children. The pipeline serves as a baseline for future research, providing a starting point for more comprehensive tools and methodologies. While we leverage existing models trained on potentially problematic data, our goal is to expose and address this issue. We do not advocate for training or deploying such models, but instead call for urgent community reflection and action to protect children's rights. Ultimately, we aim to encourage the research community to exercise - more than an additional - care in creating new datasets and to inspire the development of tools to protect the fundamental rights of vulnerable groups, particularly children.

Paperid: 3593, https://arxiv.org/pdf/2504.14200.pdf

Abstract:
In-context learning (ICL) enables Large Vision-Language Models (LVLMs) to adapt to new tasks without parameter updates, using a few demonstrations from a large support set. However, selecting informative demonstrations leads to high computational and memory costs. While some methods explore selecting a small and representative coreset in the text classification, evaluating all support set samples remains costly, and discarded samples lead to unnecessary information loss. These methods may also be less effective for image classification due to differences in feature spaces. Given these limitations, we propose Key-based Coreset Optimization (KeCO), a novel framework that leverages untapped data to construct a compact and informative coreset. We introduce visual features as keys within the coreset, which serve as the anchor for identifying samples to be updated through different selection strategies. By leveraging untapped samples from the support set, we update the keys of selected coreset samples, enabling the randomly initialized coreset to evolve into a more informative coreset under low computational cost. Through extensive experiments on coarse-grained and fine-grained image classification benchmarks, we demonstrate that KeCO effectively enhances ICL performance for image classification task, achieving an average improvement of more than 20\%. Notably, we evaluate KeCO under a simulated online scenario, and the strong performance in this scenario highlights the practical value of our framework for resource-constrained real-world scenarios.

Paperid: 3594, https://arxiv.org/pdf/2504.04939.pdf

Abstract:
Self-handover, transferring an object between one's own hands, is a common but understudied bimanual action. While it facilitates seamless transitions in complex tasks, the strategies underlying its execution remain largely unexplored. Here, we introduce the first systematic taxonomy of self-handover, derived from manual annotation of over 12 hours of cooking activity performed by 21 participants. Our analysis reveals that self-handover is not merely a passive transition, but a highly coordinated action involving anticipatory adjustments by both hands. As a step toward automated analysis of human manipulation, we further demonstrate the feasibility of classifying self-handover types using a state-of-the-art vision-language model. These findings offer fresh insights into bimanual coordination, underscoring the role of self-handover in enabling smooth task transitions-an ability essential for adaptive dual-arm robotics.

Paperid: 3595, https://arxiv.org/pdf/2503.23064.pdf

Abstract:
Large Vision-Language Models (LVLMs) struggle with puzzles, which require precise perception, rule comprehension, and logical reasoning. Assessing and enhancing their performance in this domain is crucial, as it reflects their ability to engage in structured reasoning - an essential skill for real-world problem-solving. However, existing benchmarks primarily evaluate pre-trained models without additional training or fine-tuning, often lack a dedicated focus on reasoning, and fail to establish a systematic evaluation framework. To address these limitations, we introduce VGRP-Bench, a Visual Grid Reasoning Puzzle Benchmark featuring 20 diverse puzzles. VGRP-Bench spans multiple difficulty levels, and includes extensive experiments not only on existing chat LVLMs (e.g., GPT-4o), but also on reasoning LVLMs (e.g., Gemini-Thinking). Our results reveal that even the state-of-the-art LVLMs struggle with these puzzles, highlighting fundamental limitations in their puzzle-solving capabilities. Most importantly, through systematic experiments, we identify and analyze key factors influencing LVLMs' puzzle-solving performance, including the number of clues, grid size, and rule complexity. Furthermore, we explore two Supervised Fine-Tuning (SFT) strategies that can be used in post-training: SFT on solutions (S-SFT) and SFT on synthetic reasoning processes (R-SFT). While both methods significantly improve performance on trained puzzles, they exhibit limited generalization to unseen ones. We will release VGRP-Bench to facilitate further research on LVLMs for complex, real-world problem-solving. Project page: https://yufan-ren.com/subpage/VGRP-Bench/.

Paperid: 3596, https://arxiv.org/pdf/2503.21817.pdf

Abstract:
Transformer-based models have driven significant advancements in Multimodal Large Language Models (MLLMs), yet their computational costs surge drastically when scaling resolution, training data, and model parameters. A key bottleneck stems from the proliferation of visual tokens required for fine-grained image understanding. We propose Skip-Vision, a unified framework addressing both training and inference inefficiencies in vision-language models. On top of conventional token compression approaches, our method introduces two complementary acceleration strategies. For training acceleration, we observe that Feed-Forward Network (FFN) computations on visual tokens induce marginal feature updates. This motivates our Skip-FFN strategy, which bypasses FFN layers for redundant visual tokens. For inference acceleration, we design a selective KV-cache removal mechanism that prunes the skipped key-value pairs during decoding while preserving model performance. Experimental results demonstrate that Skip-Vision reduces training time by up to 35\%, inference FLOPs by 75\%, and latency by 45\%, while achieving comparable or superior performance to existing methods. Our work provides a practical solution for scaling high-performance MLLMs with enhanced efficiency.

Paperid: 3597, https://arxiv.org/pdf/2503.15491.pdf

Abstract:
In human-robot interaction (HRI), the beginning of an interaction is often complex. Whether the robot should communicate with the human is dependent on several situational factors (e.g., the current human's activity, urgency of the interaction, etc.). We test whether large language models (LLM) and vision language models (VLM) can provide solutions to this problem. We compare four different system-design patterns using LLMs and VLMs, and test on a test set containing 84 human-robot situations. The test set mixes several publicly available datasets and also includes situations where the appropriate action to take is open-ended. Our results using the GPT-4o and Phi-3 Vision model indicate that LLMs and VLMs are capable of handling interaction beginnings when the desired actions are clear, however, challenge remains in the open-ended situations where the model must balance between the human and robot situation.

Paperid: 3598, https://arxiv.org/pdf/2503.11720.pdf

Abstract:
Traditional preference tuning methods for LLMs/Visual Generative Models often rely solely on reward model labeling, which can be opaque, offer limited insights into the rationale behind preferences, and are prone to issues such as reward hacking or overfitting. We introduce Rich Preference Optimization (RPO), a novel pipeline that leverages rich feedback signals from Vision Language Models (VLMs) to improve the curation of preference pairs for fine-tuning visual generative models like text-to-image diffusion models. Our approach begins with prompting VLMs to generate detailed critiques of synthesized images, from which we further prompt VLMs to extract reliable and actionable image editing instructions. By implementing these instructions, we create refined images, resulting in synthetic, informative preference pairs that serve as enhanced tuning datasets. We demonstrate the effectiveness of our pipeline and the resulting datasets in fine-tuning state-of-the-art diffusion models.

Paperid: 3599, https://arxiv.org/pdf/2503.09820.pdf

Abstract:
We introduce Vision-Language Attention Distillation (Vi-LAD), a novel approach for distilling socially compliant navigation knowledge from a large Vision-Language Model (VLM) into a lightweight transformer model for real-time robotic navigation. Unlike traditional methods that rely on expert demonstrations or human-annotated datasets, Vi-LAD performs knowledge distillation and fine-tuning at the intermediate layer representation level (i.e., attention maps) by leveraging the backbone of a pre-trained vision-action model. These attention maps highlight key navigational regions in a given scene, which serve as implicit guidance for socially aware motion planning. Vi-LAD fine-tunes a transformer-based model using intermediate attention maps extracted from the pre-trained vision-action model, combined with attention-like semantic maps constructed from a large VLM. To achieve this, we introduce a novel attention-level distillation loss that fuses knowledge from both sources, generating augmented attention maps with enhanced social awareness. These refined attention maps are then utilized as a traversability costmap within a socially aware model predictive controller (MPC) for navigation. We validate our approach through real-world experiments on a Husky wheeled robot, demonstrating significant improvements over state-of-the-art (SOTA) navigation methods. Our results show up to 14.2% - 50% improvement in success rate, which highlights the effectiveness of Vi-LAD in enabling socially compliant and efficient robot navigation.

Paperid: 3600, https://arxiv.org/pdf/2503.07067.pdf

Abstract:
Despite the success of distillation in large language models (LLMs), most prior work applies identical loss functions to both teacher- and student-generated data. These strategies overlook the synergy between loss formulations and data types, leading to a suboptimal performance boost in student models. To address this, we propose DistiLLM-2, a contrastive approach that simultaneously increases the likelihood of teacher responses and decreases that of student responses by harnessing this synergy. Our extensive experiments show that DistiLLM-2 not only builds high-performing student models across a wide range of tasks, including instruction-following and code generation, but also supports diverse applications, such as preference alignment and vision-language extensions. These findings highlight the potential of a contrastive approach to enhance the efficacy of LLM distillation by effectively aligning teacher and student models across varied data types.

Paperid: 3601, https://arxiv.org/pdf/2503.06794.pdf

Abstract:
Vision-Language Models (VLMs) are powerful yet computationally intensive for widespread practical deployments. To address such challenge without costly re-training, post-training acceleration techniques like quantization and token reduction are extensively explored. However, current acceleration evaluations primarily target minimal overall performance degradation, overlooking a crucial question: does the accelerated model still give the same answers to the same questions as it did before acceleration? This is vital for stability-centered industrial applications where consistently correct answers for specific, known situations are paramount, such as in AI-based disease diagnosis. We systematically investigate this for accelerated VLMs, testing four leading models (LLaVA-1.5, LLaVA-Next, Qwen2-VL, Qwen2.5-VL) with eight acceleration methods on ten multi-modal benchmarks. Our findings are stark: despite minimal aggregate performance drops, accelerated models changed original answers up to 20% of the time. Critically, up to 6.5% of these changes converted correct answers to incorrect. Input perturbations magnified these inconsistencies, and the trend is confirmed by case studies with the medical VLM LLaVA-Med. This research reveals a significant oversight in VLM acceleration, stressing an urgent need for instance-level stability checks to ensure trustworthy real-world deployment.

Paperid: 3602, https://arxiv.org/pdf/2503.00037.pdf

Abstract:
Large Vision-Language Models (LVLMs) have made significant strides in multimodal comprehension, thanks to extensive pre-training and fine-tuning on large-scale visual datasets. However, despite their robust textual safety mechanisms, they remain vulnerable to harmful visual inputs. Existing safeguards-typically relying on pre-filtering or fine-tuning-incur high costs and diminish overall utility. To address this critical vulnerability, we introduce SafeCLIP, a lightweight method that leverages LVLMs inherent multimodal alignment for zero-shot toxic image detection. By projecting CLIPs discarded CLS token into its text space and matching it with toxic descriptors, SafeCLIP detects harmful content without any architectural changes-adding minimal latency and enabling dynamic safety corrections during inference and fine-tuning.Experiments show that SafeCLIP achieves a 66.9% defense success rate with only 3.2% false positive rate and 7.2% overhead. In contrast, state-of-the-art methods achieve 52.9% success but have a 10.7% false positive rate and 210% overhead. Our work demonstrates that leveraging inherent multimodal alignment can yield efficient, low-cost LVLM safety. Code is available at anonymous.4open.science/r/safeclip-2C01.

Paperid: 3603, https://arxiv.org/pdf/2502.06094.pdf

Abstract:
Fairness is a fundamental principle in medical ethics. Vision Language Models (VLMs) have shown significant potential in the medical field due to their ability to leverage both visual and linguistic contexts, reducing the need for large datasets and enabling the performance of complex tasks. However, the exploration of fairness within VLM applications remains limited. Applying VLMs without a comprehensive analysis of fairness could lead to concerns about equal treatment opportunities and diminish public trust in medical deep learning models. To build trust in medical VLMs, we propose Fair-MoE, a model specifically designed to ensure both fairness and effectiveness. Fair-MoE comprises two key components: \textit{the Fairness-Oriented Mixture of Experts (FO-MoE)} and \textit{the Fairness-Oriented Loss (FOL)}. FO-MoE is designed to leverage the expertise of various specialists to filter out biased patch embeddings and use an ensemble approach to extract more equitable information relevant to specific tasks. FOL is a novel fairness-oriented loss function that not only minimizes the distances between different attributes but also optimizes the differences in the dispersion of various attributes' distributions. Extended experiments demonstrate the effectiveness and fairness of Fair-MoE. Tested on the Harvard-FairVLMed dataset, Fair-MoE showed improvements in both fairness and accuracy across all four attributes. Code will be publicly available.

Paperid: 3604, https://arxiv.org/pdf/2502.05485.pdf

Abstract:
Large foundation models have shown strong open-world generalization to complex problems in vision and language, but similar levels of generalization have yet to be achieved in robotics. One fundamental challenge is the lack of robotic data, which are typically obtained through expensive on-robot operation. A promising remedy is to leverage cheaper, off-domain data such as action-free videos, hand-drawn sketches or simulation data. In this work, we posit that hierarchical vision-language-action (VLA) models can be more effective in utilizing off-domain data than standard monolithic VLA models that directly finetune vision-language models (VLMs) to predict actions. In particular, we study a class of hierarchical VLA models, where the high-level VLM is finetuned to produce a coarse 2D path indicating the desired robot end-effector trajectory given an RGB image and a task description. The intermediate 2D path prediction is then served as guidance to the low-level, 3D-aware control policy capable of precise manipulation. Doing so alleviates the high-level VLM from fine-grained action prediction, while reducing the low-level policy's burden on complex task-level reasoning. We show that, with the hierarchical design, the high-level VLM can transfer across significant domain gaps between the off-domain finetuning data and real-robot testing scenarios, including differences on embodiments, dynamics, visual appearances and task semantics, etc. In the real-robot experiments, we observe an average of 20% improvement in success rate across seven different axes of generalization over OpenVLA, representing a 50% relative gain. Visual results, code, and dataset are provided at: https://hamster-robot.github.io/

Paperid: 3605, https://arxiv.org/pdf/2501.11231.pdf

Abstract:
Visual Language Models such as CLIP excel in image recognition due to extensive image-text pre-training. However, applying the CLIP inference in zero-shot classification, particularly for medical image diagnosis, faces challenges due to: 1) the inadequacy of representing image classes solely with single category names; 2) the modal gap between the visual and text spaces generated by CLIP encoders. Despite attempts to enrich disease descriptions with large language models, the lack of class-specific knowledge often leads to poor performance. In addition, empirical evidence suggests that existing proxy learning methods for zero-shot image classification on natural image datasets exhibit instability when applied to medical datasets. To tackle these challenges, we introduce the Knowledge Proxy Learning (KPL) to mine knowledge from CLIP. KPL is designed to leverage CLIP's multimodal understandings for medical image classification through Text Proxy Optimization and Multimodal Proxy Learning. Specifically, KPL retrieves image-relevant knowledge descriptions from the constructed knowledge-enhanced base to enrich semantic text proxies. It then harnesses input images and these descriptions, encoded via CLIP, to stably generate multimodal proxies that boost the zero-shot classification performance. Extensive experiments conducted on both medical and natural image datasets demonstrate that KPL enables effective zero-shot image classification, outperforming all baselines. These findings highlight the great potential in this paradigm of mining knowledge from CLIP for medical image classification and broader areas.

Paperid: 3606, https://arxiv.org/pdf/2501.10970.pdf

Abstract:
The "LLM-as-an-annotator" and "LLM-as-a-judge" paradigms employ Large Language Models (LLMs) as annotators, judges, and evaluators in tasks traditionally performed by humans. LLM annotations are widely used, not only in NLP research but also in fields like medicine, psychology, and social science. Despite their role in shaping study results and insights, there is no standard or rigorous procedure to determine whether LLMs can replace human annotators. In this paper, we propose a novel statistical procedure, the Alternative Annotator Test (alt-test), that requires only a modest subset of annotated examples to justify using LLM annotations. Additionally, we introduce a versatile and interpretable measure for comparing LLM annotators and judges. To demonstrate our procedure, we curated a diverse collection of ten datasets, consisting of language and vision-language tasks, and conducted experiments with six LLMs and four prompting techniques. Our results show that LLMs can sometimes replace humans with closed-source LLMs (such as GPT-4o), outperforming the open-source LLMs we examine, and that prompting techniques yield judges of varying quality. We hope this study encourages more rigorous and reliable practices.

Paperid: 3607, https://arxiv.org/pdf/2501.09024.pdf

Abstract:
Most existing social robot navigation techniques either leverage hand-crafted rules or human demonstrations to connect robot perception to socially compliant actions. However, there remains a significant gap in effectively translating perception into socially compliant actions, much like how human reasoning naturally occurs in dynamic environments. Considering the recent success of Vision-Language Models (VLMs), we propose using language to bridge the gap in human-like reasoning between perception and socially aware robot actions. We create a vision-language dataset, Social robot Navigation via Explainable Interactions (SNEI), featuring 40K human-annotated Visual Question Answers (VQAs) based on 2K human-robot social interactions in unstructured, crowded public spaces, spanning perception, prediction, chain-of-thought reasoning, action, and explanation. We fine-tune a VLM, Social-LLaVA, using SNEI to demonstrate the practical application of our dataset. Social-LLaVA outperforms state-of-the-art models like GPT-4V and Gemini, based on the average of fifteen different human-judge scores across 50 VQA. Deployed onboard a mobile robot, Social-LLaVA enables human-like reasoning, marking a promising step toward socially compliant robot navigation in dynamic public spaces through language reasoning.

Paperid: 3608, https://arxiv.org/pdf/2501.07214.pdf

Abstract:
Temporal logical understanding, a core facet of human cognition, plays a pivotal role in capturing complex sequential events and their temporal relationships within videos. This capability is particularly crucial in tasks like Video Question Answering (VideoQA), where the goal is to process visual data over time together with textual data to provide coherent answers. However, current VideoQA benchmarks devote little focus to evaluating this critical skill due to the challenge of annotating temporal logic. Despite the advancement of vision-language models, assessing their temporal logical reasoning powers remains a challenge, primarily due to the lack QA pairs that demand formal, complex temporal reasoning. To bridge this gap, we introduce the TimeLogic QA (TLQA) framework to automatically generate the QA pairs, specifically designed to evaluate the temporal logical understanding. To this end, TLQA leverages temporal annotations from existing video datasets together with temporal operators derived from logic theory to construct questions that test understanding of event sequences and their temporal relationships. TLQA framework is generic and scalable, capable of leveraging both, existing video action datasets with temporal action segmentation annotations, or video datasets with temporal scene graph annotations, to automatically generate temporal logical questions. We leverage 4 datasets, STAR, Breakfast, AGQA, and CrossTask, and generate two VideoQA dataset variants - small (TLQA-S) and large (TLQA-L) - containing 2k and 10k QA pairs for each category, resulting in 32k and 160k total pairs per dataset. We undertake a comprehensive evaluation of leading-edge VideoQA models, employing the TLQA to benchmark their temporal logical understanding capabilities. We assess the VideoQA model's temporal reasoning performance on 16 categories of temporal logic with varying temporal complexity.

Paperid: 3609, https://arxiv.org/pdf/2501.05952.pdf

Abstract:
In this paper, we introduce SAIL-VL (ScAlable Vision Language Model TraIning via High QuaLity Data Curation), an open-source vision language model (VLM) series achieving state-of-the-art (SOTA) performance in 2B and 8B parameters. The following three key improvements contribute to SAIL-VL's leading performance: (1) Scalable high-quality visual understanding data construction: We implement a data construction pipeline to enable hundred-million-scale high-quality recaption data annotation. The resulted dataset SAIL-Caption is validated to be of the highest data quality compared with opensource datasets. (2) Scalable Pretraining with High-Quality Visual Understanding Data: We scale SAIL-VL's pretraining budget up to 655B tokens and show that even a 2B VLM benefits from scaled up training data sizes, exhibiting logarithmic data size scaling laws in benchmark performance. (3) Scalable SFT via data quantity and complexity scaling: We curate a high-quality SFT dataset collection with leading data quantity scaling effectiveness and demonstrate that training with progressively higher-complexity data surpasses baseline one-stage training by a large margin. SAIL-VL series models achieve the highest average score in 18 widely used VLM benchmarks in our evaluation, with the 2B model takes the top position over VLMs of comparable sizes on OpenCompass 2024 (https://rank.opencompass.org.cn/leaderboard-multimodal), demonstrating robust visual comprehension abilities. SAIL-VL series models are released at HuggingFace (https://huggingface.co/BytedanceDouyinContent).

Paperid: 3610, https://arxiv.org/pdf/2501.04693.pdf

Abstract:
Interacting with the world is a multi-sensory experience: achieving effective general-purpose interaction requires making use of all available modalities -- including vision, touch, and audio -- to fill in gaps from partial observation. For example, when vision is occluded reaching into a bag, a robot should rely on its senses of touch and sound. However, state-of-the-art generalist robot policies are typically trained on large datasets to predict robot actions solely from visual and proprioceptive observations. In this work, we propose FuSe, a novel approach that enables finetuning visuomotor generalist policies on heterogeneous sensor modalities for which large datasets are not readily available by leveraging natural language as a common cross-modal grounding. We combine a multimodal contrastive loss with a sensory-grounded language generation loss to encode high-level semantics. In the context of robot manipulation, we show that FuSe enables performing challenging tasks that require reasoning jointly over modalities such as vision, touch, and sound in a zero-shot setting, such as multimodal prompting, compositional cross-modal prompting, and descriptions of objects it interacts with. We show that the same recipe is applicable to widely different generalist policies, including both diffusion-based generalist policies and large vision-language-action (VLA) models. Extensive experiments in the real world show that FuSeis able to increase success rates by over 20% compared to all considered baselines.

Paperid: 3611, https://arxiv.org/pdf/2501.03968.pdf

Abstract:
The use of Large Language Models (LLMs) for generating Behavior Trees (BTs) has recently gained attention in the robotics community, yet remains in its early stages of development. In this paper, we propose a novel framework that leverages Vision-Language Models (VLMs) to interactively generate and edit BTs that address visual conditions, enabling context-aware robot operations in visually complex environments. A key feature of our approach lies in the conditional control through self-prompted visual conditions. Specifically, the VLM generates BTs with visual condition nodes, where conditions are expressed as free-form text. Another VLM process integrates the text into its prompt and evaluates the conditions against real-world images during robot execution. We validated our framework in a real-world cafe scenario, demonstrating both its feasibility and limitations.

Paperid: 3612, https://arxiv.org/pdf/2501.02669.pdf

Abstract:
Vision Language Models (VLMs) are impressive at visual question answering and image captioning. But they underperform on multi-step visual reasoning -- even compared to LLMs on the same tasks presented in text form -- giving rise to perceptions of modality imbalance or brittleness. Towards a systematic study of such issues, we introduce a synthetic framework for assessing the ability of VLMs to perform algorithmic visual reasoning, comprising three tasks: Table Readout, Grid Navigation, and Visual Analogy. Each has two levels of difficulty, SIMPLE and HARD, and even the SIMPLE versions are difficult for frontier VLMs. We propose strategies for training on the SIMPLE version of tasks that improve performance on the corresponding HARD task, i.e., simple-to-hard (S2H) generalization. This controlled setup, where each task also has an equivalent text-only version, allows a quantification of the modality imbalance and how it is impacted by training strategy. We show that 1) explicit image-to-text conversion is important in promoting S2H generalization on images, by transferring reasoning from text; 2) conversion can be internalized at test time. We also report results of mechanistic study of this phenomenon. We identify measures of gradient alignment that can identify training strategies that promote better S2H generalization. Ablations highlight the importance of chain-of-thought.

Paperid: 3613, https://arxiv.org/pdf/2412.19531.pdf

Abstract:
In text-to-image (T2I) generation, a prevalent training technique involves utilizing Vision Language Models (VLMs) for image re-captioning. Even though VLMs are known to exhibit hallucination, generating descriptive content that deviates from the visual reality, the ramifications of such caption hallucinations on T2I generation performance remain under-explored. Through our empirical investigation, we first establish a comprehensive dataset comprising VLM-generated captions, and then systematically analyze how caption hallucination influences generation outcomes. Our findings reveal that (1) the disparities in caption quality persistently impact model outputs during fine-tuning. (2) VLMs confidence scores serve as reliable indicators for detecting and characterizing noise-related patterns in the data distribution. (3) even subtle variations in caption fidelity have significant effects on the quality of learned representations. These findings collectively emphasize the profound impact of caption quality on model performance and highlight the need for more sophisticated robust training algorithm in T2I. In response to these observations, we propose a approach leveraging VLM confidence score to mitigate caption noise, thereby enhancing the robustness of T2I models against hallucination in caption.

Paperid: 3614, https://arxiv.org/pdf/2412.11337.pdf

Abstract:
Multi-step dexterous manipulation is a fundamental skill in household scenarios, yet remains an underexplored area in robotics. This paper proposes a modular approach, where each step of the manipulation process is addressed with dedicated policies based on effective modality input, rather than relying on a single end-to-end model. To demonstrate this, a dexterous robotic hand performs a manipulation task involving picking up and rotating a box. Guided by insights from neuroscience, the task is decomposed into three sub-skills, 1)reaching, 2)grasping and lifting, and 3)in-hand rotation, based on the dominant sensory modalities employed in the human brain. Each sub-skill is addressed using distinct methods from a practical perspective: a classical controller, a Vision-Language-Action model, and a reinforcement learning policy with force feedback, respectively. We tested the pipeline on a real robot to demonstrate the feasibility of our approach. The key contribution of this study lies in presenting a neuroscience-inspired, modality-driven methodology for multi-step dexterous manipulation.

Paperid: 3615, https://arxiv.org/pdf/2412.04925.pdf

Abstract:
Recently, many studies have been conducted to enhance the zero-shot generalization ability of vision-language models (e.g., CLIP) by addressing the semantic misalignment between image and text embeddings in downstream tasks. Although many efforts have been made, existing methods barely consider the fact that a class of images can be described by notably different textual concepts due to well-known lexical variation in natural language processing, which heavily affects the zero-shot generalization of CLIP. Therefore, this paper proposes a \textbf{S}ynonymous \textbf{S}emantic \textbf{S}pace ($S^3$) for each image class, rather than relying on a single textual concept, achieving more stable semantic alignment and improving the zero-shot generalization of CLIP. Specifically, our $S^3$ method first generates several synonymous concepts based on the label of each class by using large language models, and constructs a continuous yet compact synonymous semantic space based on the Vietoris-Rips complex of the generated synonymous concepts. Furthermore, we explore the effect of several point-to-space metrics on our $S^3$, while presenting a point-to-local-center metric to compute similarity between image embeddings and the synonymous semantic space of each class, accomplishing effective zero-shot predictions. Extensive experiments are conducted across 17 benchmarks, including fine-grained zero-shot classification, natural distribution zero-shot classification, and open-vocabulary segmentation, and the results show that our $S^3$ outperforms state-of-the-art methods.

Paperid: 3616, https://arxiv.org/pdf/2412.02818.pdf

Abstract:
Robot manipulation policies often fail for unknown reasons, posing significant challenges for real-world deployment. Researchers and engineers typically address these failures using heuristic approaches, which are not only labor-intensive and costly but also prone to overlooking critical failure modes (FMs). This paper introduces Robot Manipulation Diagnosis (RoboMD), a systematic framework designed to automatically identify FMs arising from unanticipated changes in the environment. Considering the vast space of potential FMs in a pre-trained manipulation policy, we leverage deep reinforcement learning (deep RL) to explore and uncover these FMs using a specially trained vision-language embedding that encodes a notion of failures. This approach enables users to probabilistically quantify and rank failures in previously unseen environmental conditions. Through extensive experiments across various manipulation tasks and algorithms, we demonstrate RoboMD's effectiveness in diagnosing unknown failures in unstructured environments, providing a systematic pathway to improve the robustness of manipulation policies.

Paperid: 3617, https://arxiv.org/pdf/2412.02071.pdf

Abstract:
While image captioning provides isolated descriptions for individual images, and video captioning offers one single narrative for an entire video clip, our work explores an important middle ground: progress-aware video captioning at the frame level. This novel task aims to generate temporally fine-grained captions that not only accurately describe each frame but also capture the subtle progression of actions throughout a video sequence. Despite the strong capabilities of existing leading vision language models, they often struggle to discern the nuances of frame-wise differences. To address this, we propose ProgressCaptioner, a captioning model designed to capture the fine-grained temporal dynamics within an action sequence. Alongside, we develop the FrameCap dataset to support training and the FrameCapEval benchmark to assess caption quality. The results demonstrate that ProgressCaptioner significantly surpasses leading captioning models, producing precise captions that accurately capture action progression and set a new standard for temporal precision in video captioning. Finally, we showcase practical applications of our approach, specifically in aiding keyframe selection and advancing video understanding, highlighting its broad utility.

Paperid: 3618, https://arxiv.org/pdf/2411.15600.pdf

Abstract:
Vision-language tracking (VLT) extends traditional single object tracking by incorporating textual information, providing semantic guidance to enhance tracking performance under challenging conditions like fast motion and deformations. However, current VLT trackers often underperform compared to single-modality methods on multiple benchmarks, with semantic information sometimes becoming a "distraction." To address this, we propose VLTVerse, the first fine-grained evaluation framework for VLT trackers that comprehensively considers multiple challenge factors and diverse semantic information, hoping to reveal the role of language in VLT. Our contributions include: (1) VLTVerse introduces 10 sequence-level challenge labels and 6 types of multi-granularity semantic information, creating a flexible and multi-dimensional evaluation space for VLT; (2) leveraging 60 subspaces formed by combinations of challenge factors and semantic types, we conduct systematic fine-grained evaluations of three mainstream SOTA VLT trackers, uncovering their performance bottlenecks across complex scenarios and offering a novel perspective on VLT evaluation; (3) through decoupled analysis of experimental results, we examine the impact of various semantic types on specific challenge factors in relation to different algorithms, providing essential guidance for enhancing VLT across data, evaluation, and algorithmic dimensions. The VLTVerse, toolkit, and results will be available at \url{http://metaverse.aitestunion.com}.

Paperid: 3619, https://arxiv.org/pdf/2411.12785.pdf

Abstract:
Vision-Language (V-L) pre-trained models such as CLIP show prominent capabilities in various downstream tasks. Despite this promise, V-L models are notoriously limited by their inherent social biases. A typical demonstration is that V-L models often produce biased predictions against specific groups of people, significantly undermining their real-world applicability. Existing approaches endeavor to mitigate the social bias problem in V-L models by removing biased attribute information from model embeddings. However, after our revisiting of these methods, we find that their bias removal is frequently accompanied by greatly compromised V-L alignment capabilities. We then reveal that this performance degradation stems from the unbalanced debiasing in image and text embeddings. To address this issue, we propose a novel V-L debiasing framework to align image and text biases followed by removing them from both modalities. By doing so, our method achieves multi-modal bias mitigation while maintaining the V-L alignment in the debiased embeddings. Additionally, we advocate a new evaluation protocol that can 1) holistically quantify the model debiasing and V-L alignment ability, and 2) evaluate the generalization of social bias removal models. We believe this work will offer new insights and guidance for future studies addressing the social bias problem in CLIP.

Paperid: 3620, https://arxiv.org/pdf/2411.11223.pdf

Abstract:
Pre-trained vision-language models provide a robust foundation for efficient transfer learning across various downstream tasks. In the field of video action recognition, mainstream approaches often introduce additional modules to capture temporal information. Although the additional modules increase the capacity of model, enabling it to better capture video-specific inductive biases, existing methods typically introduce a substantial number of new parameters and are prone to catastrophic forgetting of previously acquired generalizable knowledge. In this paper, we propose a parameter-efficient Multi-modal Spatio-Temporal Adapter (MSTA) to enhance the alignment between textual and visual representations, achieving a balance between generalizable knowledge and task-specific adaptation. Furthermore, to mitigate over-fitting and enhance generalizability, we introduce a spatio-temporal description-guided consistency constraint.This constraint involves providing template inputs (e.g., "a video of \{\textbf{cls}\}") to the trainable language branch and LLM-generated spatio-temporal descriptions to the pre-trained language branch, enforcing output consistency between the branches. This approach reduces overfitting to downstream tasks and enhances the distinguishability of the trainable branch within the spatio-temporal semantic space. We evaluate the effectiveness of our approach across four tasks: zero-shot transfer, few-shot learning, base-to-novel generalization, and fully-supervised learning. Compared to many state-of-the-art methods, our MSTA achieves outstanding performance across all evaluations, while using only 2-7\% of the trainable parameters in the original model.

Paperid: 3621, https://arxiv.org/pdf/2411.09126.pdf

Abstract:
While contrastive pre-training is widely employed, its data efficiency problem has remained relatively under-explored thus far. Existing methods often rely on static coreset selection algorithms to pre-identify important data for training. However, this static nature renders them unable to dynamically track the data usefulness throughout pre-training, leading to subpar pre-trained models. To address this challenge, our paper introduces a novel dynamic bootstrapping dataset pruning method. It involves pruning data preparation followed by dataset mutation operations, both of which undergo iterative and dynamic updates. We apply this method to two prevalent contrastive pre-training frameworks: \textbf{CLIP} and \textbf{MoCo}, representing vision-language and vision-centric domains, respectively. In particular, we individually pre-train seven CLIP models on two large-scale image-text pair datasets, and two MoCo models on the ImageNet dataset, resulting in a total of 16 pre-trained models. With a data pruning rate of 30-35\% across all 16 models, our method exhibits only marginal performance degradation (less than \textbf{1\%} on average) compared to corresponding models trained on the full dataset counterparts across various downstream datasets, and also surpasses several baselines with a large performance margin. Additionally, the byproduct from our method, \ie coresets derived from the original datasets after pre-training, also demonstrates significant superiority in terms of downstream performance over other static coreset selection approaches.

Paperid: 3622, https://arxiv.org/pdf/2411.02074.pdf

Abstract:
Generalized Category Discovery (GCD) aims to cluster unlabeled images into known and novel categories using labeled images from known classes. To address the challenge of transferring features from known to unknown classes while mitigating model bias, we introduce GraphVL, a novel approach for vision-language modeling in GCD, leveraging CLIP. Our method integrates a graph convolutional network (GCN) with CLIP's text encoder to preserve class neighborhood structure. We also employ a lightweight visual projector for image data, ensuring discriminative features through margin-based contrastive losses for image-text mapping. This neighborhood preservation criterion effectively regulates the semantic space, making it less sensitive to known classes. Additionally, we learn textual prompts from known classes and align them to create a more contextually meaningful semantic feature space for the GCN layer using a contextual similarity loss. Finally, we represent unlabeled samples based on their semantic distance to class prompts from the GCN, enabling semi-supervised clustering for class discovery and minimizing errors. Our experiments on seven benchmark datasets consistently demonstrate the superiority of GraphVL when integrated with the CLIP backbone.

Paperid: 3623, https://arxiv.org/pdf/2410.05804.pdf

Abstract:
Incremental object detection is fundamentally challenged by catastrophic forgetting. A major factor contributing to this issue is background shift, where background categories in sequential tasks may overlap with either previously learned or future unseen classes. To address this, we propose a novel method called Class-Agnostic Shared Attribute Base (CASA) that encourages the model to learn category-agnostic attributes shared across incremental classes. Our approach leverages an LLM to generate candidate textual attributes, selects the most relevant ones based on the current training data, and records their importance in an assignment matrix. For subsequent tasks, the retained attributes are frozen, and new attributes are selected from the remaining candidates, ensuring both knowledge retention and adaptability. Extensive experiments on the COCO dataset demonstrate the state-of-the-art performance of our method.

Paperid: 3624, https://arxiv.org/pdf/2409.16484.pdf

Abstract:
We present BehAV, a novel approach for autonomous robot navigation in outdoor scenes guided by human instructions and leveraging Vision Language Models (VLMs). Our method interprets human commands using a Large Language Model (LLM) and categorizes the instructions into navigation and behavioral guidelines. Navigation guidelines consist of directional commands (e.g., "move forward until") and associated landmarks (e.g., "the building with blue windows"), while behavioral guidelines encompass regulatory actions (e.g., "stay on") and their corresponding objects (e.g., "pavements"). We use VLMs for their zero-shot scene understanding capabilities to estimate landmark locations from RGB images for robot navigation. Further, we introduce a novel scene representation that utilizes VLMs to ground behavioral rules into a behavioral cost map. This cost map encodes the presence of behavioral objects within the scene and assigns costs based on their regulatory actions. The behavioral cost map is integrated with a LiDAR-based occupancy map for navigation. To navigate outdoor scenes while adhering to the instructed behaviors, we present an unconstrained Model Predictive Control (MPC)-based planner that prioritizes both reaching landmarks and following behavioral guidelines. We evaluate the performance of BehAV on a quadruped robot across diverse real-world scenarios, demonstrating a 22.49% improvement in alignment with human-teleoperated actions, as measured by Frechet distance, and achieving a 40% higher navigation success rate compared to state-of-the-art methods.

Paperid: 3625, https://arxiv.org/pdf/2409.10683.pdf

Abstract:
While success in many robotics tasks can be determined by only observing the final state and how it differs from the initial state - e.g., if an apple is picked up - many tasks require observing the full motion of the robot to correctly determine success. For example, brushing hair requires repeated strokes that correspond to the contours and type of hair. Prior works often use off-the-shelf vision-language models (VLMs) as success detectors; however, when success depends on the full trajectory, VLMs struggle to make correct judgments for two reasons. First, modern VLMs are trained only on single frames, and cannot capture changes over a full trajectory. Second, even if we provide state-of-the-art VLMs with an aggregate input of multiple frames, they still fail to detect success due to a lack of robot data. Our key idea is to fine-tune VLMs using abstract representations that are able to capture trajectory-level information such as the path the robot takes by overlaying keypoint trajectories on the final image. We propose motion instruction fine-tuning (MotIF), a method that fine-tunes VLMs using the aforementioned abstract representations to semantically ground the robot's behavior in the environment. To benchmark and fine-tune VLMs for robotic motion understanding, we introduce the MotIF-1K dataset containing 653 human and 369 robot demonstrations across 13 task categories. MotIF assesses the success of robot motion given the image observation of the trajectory, task instruction, and motion description. Our model significantly outperforms state-of-the-art VLMs by at least twice in precision and 56.1% in recall, generalizing across unseen motions, tasks, and environments. Finally, we demonstrate practical applications of MotIF in refining and terminating robot planning, and ranking trajectories on how they align with task and motion descriptions. Project page: https://motif-1k.github.io

Paperid: 3626, https://arxiv.org/pdf/2409.09715.pdf

Abstract:
This paper develops an edge-device collaborative Generative Semantic Communications (Gen SemCom) framework leveraging pre-trained Multi-modal/Vision Language Models (M/VLMs) for ultra-low-rate semantic communication via textual prompts. The proposed framework optimizes the use of M/VLMs on the wireless edge/device to generate high-fidelity textual prompts through visual captioning/question answering, which are then transmitted over a wireless channel for SemCom. Specifically, we develop a multi-user Gen SemCom framework using pre-trained M/VLMs, and formulate a joint optimization problem of prompt generation offloading, communication and computation resource allocation to minimize the latency and maximize the resulting semantic quality. Due to the nonconvex nature of the problem with highly coupled discrete and continuous variables, we decompose it as a two-level problem and propose a low-complexity swap/leaving/joining (SLJ)-based matching algorithm. Simulation results demonstrate significant performance improvements over the conventional semanticunaware/non-collaborative offloading benchmarks.

Paperid: 3627, https://arxiv.org/pdf/2409.06351.pdf

Abstract:
In emergency departments, rural hospitals, or clinics in less developed regions, clinicians often lack fast image analysis by trained radiologists, which can have a detrimental effect on patients' healthcare. Large Language Models (LLMs) have the potential to alleviate some pressure from these clinicians by providing insights that can help them in their decision-making. While these LLMs achieve high test results on medical exams showcasing their great theoretical medical knowledge, they tend not to follow medical guidelines. In this work, we introduce a new approach for zero-shot guideline-driven decision support. We model a system of multiple LLM agents augmented with a contrastive vision-language model that collaborate to reach a patient diagnosis. After providing the agents with simple diagnostic guidelines, they will synthesize prompts and screen the image for findings following these guidelines. Finally, they provide understandable chain-of-thought reasoning for their diagnosis, which is then self-refined to consider inter-dependencies between diseases. As our method is zero-shot, it is adaptable to settings with rare diseases, where training data is limited, but expert-crafted disease descriptions are available. We evaluate our method on two chest X-ray datasets, CheXpert and ChestX-ray 14 Longtail, showcasing performance improvement over existing zero-shot methods and generalizability to rare diseases.

Paperid: 3628, https://arxiv.org/pdf/2409.04693.pdf

Abstract:
Recently, prompt learning has garnered considerable attention for its success in various Vision-Language (VL) tasks. However, existing prompt-based models are primarily focused on studying prompt generation and prompt strategies with complete modality settings, which does not accurately reflect real-world scenarios where partial modality information may be missing. In this paper, we present the first comprehensive investigation into prompt learning behavior when modalities are incomplete, revealing the high sensitivity of prompt-based models to missing modalities. To this end, we propose a novel Multi-step Adaptive Prompt Learning (MuAP) framework, aiming to generate multimodal prompts and perform multi-step prompt tuning, which adaptively learns knowledge by iteratively aligning modalities. Specifically, we generate multimodal prompts for each modality and devise prompt strategies to integrate them into the Transformer model. Subsequently, we sequentially perform prompt tuning from single-stage and alignment-stage, allowing each modality-prompt to be autonomously and adaptively learned, thereby mitigating the imbalance issue caused by only textual prompts that are learnable in previous works. Extensive experiments demonstrate the effectiveness of our MuAP and this model achieves significant improvements compared to the state-of-the-art on all benchmark datasets

Paperid: 3629, https://arxiv.org/pdf/2409.02664.pdf

Abstract:
The proliferation of deepfake faces poses huge potential negative impacts on our daily lives. Despite substantial advancements in deepfake detection over these years, the generalizability of existing methods against forgeries from unseen datasets or created by emerging generative models remains constrained. In this paper, inspired by the zero-shot advantages of Vision-Language Models (VLMs), we propose a novel approach that repurposes a well-trained VLM for general deepfake detection. Motivated by the model reprogramming paradigm that manipulates the model prediction via input perturbations, our method can reprogram a pre-trained VLM model (e.g., CLIP) solely based on manipulating its input without tuning the inner parameters. First, learnable visual perturbations are used to refine feature extraction for deepfake detection. Then, we exploit information of face embedding to create sample-level adaptative text prompts, improving the performance. Extensive experiments on several popular benchmark datasets demonstrate that (1) the cross-dataset and cross-manipulation performances of deepfake detection can be significantly and consistently improved (e.g., over 88\% AUC in cross-dataset setting from FF++ to WildDeepfake); (2) the superior performances are achieved with fewer trainable parameters, making it a promising approach for real-world applications.

Paperid: 3630, https://arxiv.org/pdf/2601.20433.pdf

Abstract:
Deepfake detection is a widely researched topic that is crucial for combating the spread of malicious content, with existing methods mainly modeling the problem as classification or spatial localization. The rapid advancements in generative models impose new demands on Deepfake detection. In this paper, we propose multimodal alignment and reinforcement for explainable Deepfake detection via vision-language models, termed MARE, which aims to enhance the accuracy and reliability of Vision-Language Models (VLMs) in Deepfake detection and reasoning. Specifically, MARE designs comprehensive reward functions, incorporating reinforcement learning from human feedback (RLHF), to incentivize the generation of text-spatially aligned reasoning content that adheres to human preferences. Besides, MARE introduces a forgery disentanglement module to capture intrinsic forgery traces from high-level facial semantics, thereby improving its authenticity detection capability. We conduct thorough evaluations on the reasoning content generated by MARE. Both quantitative and qualitative experimental results demonstrate that MARE achieves state-of-the-art performance in terms of accuracy and reliability.

Paperid: 3631, https://arxiv.org/pdf/2601.19258.pdf

Abstract:
Modern mobile applications rely on hidden interactions--gestures without visual cues like long presses and swipes--to provide functionality without cluttering interfaces. While experienced users may discover these interactions through prior use or onboarding tutorials, their implicit nature makes them difficult for most users to uncover. Similarly, mobile agents--systems designed to automate tasks on mobile user interfaces, powered by vision language models (VLMs)--struggle to detect veiled interactions or determine actions for completing tasks. To address this challenge, we present GhostUI, a new dataset designed to enable the detection of hidden interactions in mobile applications. GhostUI provides before-and-after screenshots, simplified view hierarchies, gesture metadata, and task descriptions, allowing VLMs to better recognize concealed gestures and anticipate post-interaction states. Quantitative evaluations with VLMs show that models fine-tuned on GhostUI outperform baseline VLMs, particularly in predicting hidden interactions and inferring post-interaction screens, underscoring GhostUI's potential as a foundation for advancing mobile task automation.

Paperid: 3632, https://arxiv.org/pdf/2601.16163.pdf

Abstract:
Recent video generation models demonstrate remarkable ability to capture complex physical interactions and scene evolution over time. To leverage their spatiotemporal priors, robotics works have adapted video models for policy learning but introduce complexity by requiring multiple stages of post-training and new architectural components for action generation. In this work, we introduce Cosmos Policy, a simple approach for adapting a large pretrained video model (Cosmos-Predict2) into an effective robot policy through a single stage of post-training on the robot demonstration data collected on the target platform, with no architectural modifications. Cosmos Policy learns to directly generate robot actions encoded as latent frames within the video model's latent diffusion process, harnessing the model's pretrained priors and core learning algorithm to capture complex action distributions. Additionally, Cosmos Policy generates future state images and values (expected cumulative rewards), which are similarly encoded as latent frames, enabling test-time planning of action trajectories with higher likelihood of success. In our evaluations, Cosmos Policy achieves state-of-the-art performance on the LIBERO and RoboCasa simulation benchmarks (98.5% and 67.1% average success rates, respectively) and the highest average score in challenging real-world bimanual manipulation tasks, outperforming strong diffusion policies trained from scratch, video model-based policies, and state-of-the-art vision-language-action models fine-tuned on the same robot demonstrations. Furthermore, given policy rollout data, Cosmos Policy can learn from experience to refine its world model and value function and leverage model-based planning to achieve even higher success rates in challenging tasks. We release code, models, and training data at https://research.nvidia.com/labs/dir/cosmos-policy/

Paperid: 3633, https://arxiv.org/pdf/2601.15541.pdf

Abstract:
We propose a CompliantVLA-adaptor that augments the state-of-the-art Vision-Language-Action (VLA) models with vision-language model (VLM)-informed context-aware variable impedance control (VIC) to improve the safety and effectiveness of contact-rich robotic manipulation tasks. Existing VLA systems (e.g., RDT, Pi0, OpenVLA-oft) typically output position, but lack force-aware adaptation, leading to unsafe or failed interactions in physical tasks involving contact, compliance, or uncertainty. In the proposed CompliantVLA-adaptor, a VLM interprets task context from images and natural language to adapt the stiffness and damping parameters of a VIC controller. These parameters are further regulated using real-time force/torque feedback to ensure interaction forces remain within safe thresholds. We demonstrate that our method outperforms the VLA baselines on a suite of complex contact-rich tasks, both in simulation and on real hardware, with improved success rates and reduced force violations. The overall success rate across all tasks increases from 9.86\% to 17.29\%, presenting a promising path towards safe contact-rich manipulation using VLAs. We release our code, prompts, and force-torque-impedance-scenario context datasets at https://sites.google.com/view/compliantvla.

Paperid: 3634, https://arxiv.org/pdf/2601.06931.pdf

Abstract:
Vision-Language Models (VLMs) are increasingly deployed in socially consequential settings, raising concerns about social bias driven by demographic cues. A central challenge in measuring such social bias is attribution under visual confounding: real-world images entangle race and gender with correlated factors such as background and clothing, obscuring attribution. We propose a \textbf{face-only counterfactual evaluation paradigm} that isolates demographic effects while preserving real-image realism. Starting from real photographs, we generate counterfactual variants by editing only facial attributes related to race and gender, keeping all other visual factors fixed. Based on this paradigm, we construct \textbf{FOCUS}, a dataset of 480 scene-matched counterfactual images across six occupations and ten demographic groups, and propose \textbf{REFLECT}, a benchmark comprising three decision-oriented tasks: two-alternative forced choice, multiple-choice socioeconomic inference, and numeric salary recommendation. Experiments on five state-of-the-art VLMs reveal that demographic disparities persist under strict visual control and vary substantially across task formulations. These findings underscore the necessity of controlled, counterfactual audits and highlight task design as a critical factor in evaluating social bias in multimodal models.

Paperid: 3635, https://arxiv.org/pdf/2601.03594.pdf

Abstract:
This paper provides a systematic survey of jailbreak attacks and defenses on Large Language Models (LLMs) and Vision-Language Models (VLMs), emphasizing that jailbreak vulnerabilities stem from structural factors such as incomplete training data, linguistic ambiguity, and generative uncertainty. It further differentiates between hallucinations and jailbreaks in terms of intent and triggering mechanisms. We propose a three-dimensional survey framework: (1) Attack dimension-including template/encoding-based, in-context learning manipulation, reinforcement/adversarial learning, LLM-assisted and fine-tuned attacks, as well as prompt- and image-level perturbations and agent-based transfer in VLMs; (2) Defense dimension-encompassing prompt-level obfuscation, output evaluation, and model-level alignment or fine-tuning; and (3) Evaluation dimension-covering metrics such as Attack Success Rate (ASR), toxicity score, query/time cost, and multimodal Clean Accuracy and Attribute Success Rate. Compared with prior works, this survey spans the full spectrum from text-only to multimodal settings, consolidating shared mechanisms and proposing unified defense principles: variant-consistency and gradient-sensitivity detection at the perception layer, safety-aware decoding and output review at the generation layer, and adversarially augmented preference alignment at the parameter layer. Additionally, we summarize existing multimodal safety benchmarks and discuss future directions, including automated red teaming, cross-modal collaborative defense, and standardized evaluation.

Paperid: 3636, https://arxiv.org/pdf/2601.00501.pdf

Abstract:
We introduce CPPO, a Contrastive Perception Policy Optimization method for finetuning vision-language models (VLMs). While reinforcement learning (RL) has advanced reasoning in language models, extending it to multimodal reasoning requires improving both the perception and reasoning aspects. Prior works tackle this challenge mainly with explicit perception rewards, but disentangling perception tokens from reasoning tokens is difficult, requiring extra LLMs, ground-truth data, forced separation of perception from reasoning by policy model, or applying rewards indiscriminately to all output tokens. CPPO addresses this problem by detecting perception tokens via entropy shifts in the model outputs under perturbed input images. CPPO then extends the RL objective function with a Contrastive Perception Loss (CPL) that enforces consistency under information-preserving perturbations and sensitivity under information-removing ones. Experiments show that CPPO surpasses previous perception-rewarding methods, while avoiding extra models, making training more efficient and scalable.

Paperid: 3637, https://arxiv.org/pdf/2512.23562.pdf

Abstract:
Multi-model routing has evolved from an engineering technique into essential infrastructure, yet existing work lacks a systematic, reproducible benchmark for evaluating vision-language models (VLMs). We present VL-RouterBench to assess the overall capability of VLM routing systems systematically. The benchmark is grounded in raw inference and scoring logs from VLMs and constructs quality and cost matrices over sample-model pairs. In scale, VL-RouterBench covers 14 datasets across 3 task groups, totaling 30,540 samples, and includes 15 open-source models and 2 API models, yielding 519,180 sample-model pairs and a total input-output token volume of 34,494,977. The evaluation protocol jointly measures average accuracy, average cost, and throughput, and builds a ranking score from the harmonic mean of normalized cost and accuracy to enable comparison across router configurations and cost budgets. On this benchmark, we evaluate 10 routing methods and baselines and observe a significant routability gain, while the best current routers still show a clear gap to the ideal Oracle, indicating considerable room for improvement in router architecture through finer visual cues and modeling of textual structure. We will open-source the complete data construction and evaluation toolchain to promote comparability, reproducibility, and practical deployment in multimodal routing research.

Paperid: 3638, https://arxiv.org/pdf/2512.23162.pdf

Abstract:
Data scarcity remains a fundamental barrier to achieving fully autonomous surgical robots. While large scale vision language action (VLA) models have shown impressive generalization in household and industrial manipulation by leveraging paired video action data from diverse domains, surgical robotics suffers from the paucity of datasets that include both visual observations and accurate robot kinematics. In contrast, vast corpora of surgical videos exist, but they lack corresponding action labels, preventing direct application of imitation learning or VLA training. In this work, we aim to alleviate this problem by learning policy models from SurgWorld, a world model designed for surgical physical AI. We curated the Surgical Action Text Alignment (SATA) dataset with detailed action description specifically for surgical robots. Then we built SurgeWorld based on the most advanced physical AI world model and SATA. It's able to generate diverse, generalizable and realistic surgery videos. We are also the first to use an inverse dynamics model to infer pseudokinematics from synthetic surgical videos, producing synthetic paired video action data. We demonstrate that a surgical VLA policy trained with these augmented data significantly outperforms models trained only on real demonstrations on a real surgical robot platform. Our approach offers a scalable path toward autonomous surgical skill acquisition by leveraging the abundance of unlabeled surgical video and generative world modeling, thus opening the door to generalizable and data efficient surgical robot policies.

Paperid: 3639, https://arxiv.org/pdf/2512.23090.pdf

Abstract:
Recent Reinforcement Learning (RL) advances for Large Language Models (LLMs) have improved reasoning tasks, yet their resource-constrained application to medical imaging remains underexplored. We introduce ChexReason, a vision-language model trained via R1-style methodology (SFT followed by GRPO) using only 2,000 SFT samples, 1,000 RL samples, and a single A100 GPU. Evaluations on CheXpert and NIH benchmarks reveal a fundamental tension: GRPO recovers in-distribution performance (23% improvement on CheXpert, macro-F1 = 0.346) but degrades cross-dataset transferability (19% drop on NIH). This mirrors high-resource models like NV-Reason-CXR-3B, suggesting the issue stems from the RL paradigm rather than scale. We identify a generalization paradox where the SFT checkpoint uniquely improves on NIH before optimization, indicating teacher-guided reasoning captures more institution-agnostic features. Furthermore, cross-model comparisons show structured reasoning scaffolds benefit general-purpose VLMs but offer minimal gain for medically pre-trained models. Consequently, curated supervised fine-tuning may outperform aggressive RL for clinical deployment requiring robustness across diverse populations.

Paperid: 3640, https://arxiv.org/pdf/2512.21583.pdf

Abstract:
With the rapid growth of large language models (LLMs) and vision-language models (VLMs) in medicine, simply integrating clinical text and medical imaging does not guarantee reliable reasoning. Existing multimodal models often produce hallucinations or inconsistent chains of thought, limiting clinical trust. We propose a diagnostic framework built upon LLaVA that combines vision-language alignment with logic-regularized reasoning. The system includes an input encoder for text and images, a projection module for cross-modal alignment, a reasoning controller that decomposes diagnostic tasks into steps, and a logic tree generator that assembles stepwise premises into verifiable conclusions. Evaluations on MedXpertQA and other benchmarks show that our method improves diagnostic accuracy and yields more interpretable reasoning traces on multimodal tasks, while remaining competitive on text-only settings. These results suggest a promising step toward trustworthy multimodal medical AI.

Paperid: 3641, https://arxiv.org/pdf/2512.19396.pdf

Abstract:
Contemporary GUI agents, while increasingly capable due to advances in Large Vision-Language Models (VLMs), often operate with a critical limitation: they treat each task in isolation, lacking a mechanism to systematically learn from past successes. This digital ''amnesia'' results in sub-optimal performance, repeated errors, and poor generalization to novel challenges. To bridge this gap, we introduce EchoTrail-GUI, a novel framework designed to mimic human-like experiential learning by equipping agents with a dynamic, accessible memory. Our framework operates in three distinct stages. First, during Experience Exploration, an agent autonomously interacts with GUI environments to build a curated database of successful task trajectories, validated by a reward model. Crucially, the entire knowledge base construction is thus fully automated, requiring no human supervision. Second, in the Memory Injection stage, upon receiving a new task, our system efficiently retrieves the most relevant past trajectories to serve as actionable ''memories''. Finally, during GUI Task Inference, these memories are injected as in-context guidance to inform the agent's reasoning and decision-making process. We demonstrate the efficacy of our approach on benchmarks including Android World and AndroidLab. The results show that EchoTrail-GUI significantly improves the task success rate and operational efficiency of baseline agents, validating the power of structured memory in creating more robust and intelligent GUI automation.

Paperid: 3642, https://arxiv.org/pdf/2512.19091.pdf

Abstract:
Open challenges have become the de facto standard for comparative ranking of medical AI methods. Despite their importance, medical AI leaderboards exhibit three persistent limitations: (1) score gaps are rarely tested for statistical significance, so rank stability is unknown; (2) single averaged metrics are applied to every organ, hiding clinically important boundary errors; (3) performance across intersecting demographics is seldom reported, masking fairness and equity gaps. We introduce RankInsight, an open-source toolkit that seeks to address these limitations. RankInsight (1) computes pair-wise significance maps that show the nnU-Net family outperforms Vision-Language and MONAI submissions with high statistical certainty; (2) recomputes leaderboards with organ-appropriate metrics, reversing the order of the top four models when Dice is replaced by NSD for tubular structures; and (3) audits intersectional fairness, revealing that more than half of the MONAI-based entries have the largest gender-race discrepancy on our proprietary Johns Hopkins Hospital dataset. The RankInsight toolkit is publicly released and can be directly applied to past, ongoing, and future challenges. It enables organizers and participants to publish rankings that are statistically sound, clinically meaningful, and demographically fair.

Paperid: 3643, https://arxiv.org/pdf/2512.18264.pdf

Abstract:
As vision-language models (VLMs) become widely adopted, VLM-based attribute inference attacks have emerged as a serious privacy concern, enabling adversaries to infer private attributes from images shared on social media. This escalating threat calls for dedicated protection methods to safeguard user privacy. However, existing methods often degrade the visual quality of images or interfere with vision-based functions on social media, thereby failing to achieve a desirable balance between privacy protection and user experience. To address this challenge, we propose a novel protection method that jointly optimizes privacy suppression and utility preservation under a visual consistency constraint. While our method is conceptually effective, fair comparisons between methods remain challenging due to the lack of publicly available evaluation datasets. To fill this gap, we introduce VPI-COCO, a publicly available benchmark comprising 522 images with hierarchically structured privacy questions and corresponding non-private counterparts, enabling fine-grained and joint evaluation of protection methods in terms of privacy preservation and user experience. Building upon this benchmark, experiments on multiple VLMs demonstrate that our method effectively reduces PAR below 25%, keeps NPAR above 88%, maintains high visual consistency, and generalizes well to unseen and paraphrased privacy questions, demonstrating its strong practical applicability for real-world VLM deployments.

Paperid: 3644, https://arxiv.org/pdf/2511.22170.pdf

Abstract:
Concept Bottleneck Models (CBMs) enhance interpretability by introducing a layer of human-understandable concepts between inputs and predictions. While recent methods automate concept generation using Large Language Models (LLMs) and Vision-Language Models (VLMs), they still face three fundamental challenges: poor visual grounding, concept redundancy, and the absence of principled metrics to balance predictive accuracy and concept compactness. We introduce PS-CBM, a Partially Shared CBM framework that addresses these limitations through three core components: (1) a multimodal concept generator that integrates LLM-derived semantics with exemplar-based visual cues; (2) a Partially Shared Concept Strategy that merges concepts based on activation patterns to balance specificity and compactness; and (3) Concept-Efficient Accuracy (CEA), a post-hoc metric that jointly captures both predictive accuracy and concept compactness. Extensive experiments on eleven diverse datasets show that PS-CBM consistently outperforms state-of-the-art CBMs, improving classification accuracy by 1.0%-7.4% and CEA by 2.0%-9.5%, while requiring significantly fewer concepts. These results underscore PS-CBM's effectiveness in achieving both high accuracy and strong interpretability.

Paperid: 3645, https://arxiv.org/pdf/2511.20330.pdf

Abstract:
Interactive articulated manipulation requires long-horizon, multi-step interactions with appliances while maintaining physical consistency. Existing vision-language and diffusion-based policies struggle to generalize across parts, instances, and categories. We first introduce ArtiBench, a five-level benchmark covering kitchen, storage, office, and tool environments. ArtiBench enables structured evaluation from cross-part and cross-instance variation to long-horizon multi-object tasks, revealing the core generalization challenges of articulated object manipulation. Building on this benchmark, we propose ArtiBrain, a modular framework that unifies high-level reasoning with adaptive low-level control. ArtiBrain uses a VLM-based Task Reasoner (GPT-4.1) to decompose and validate subgoals, and employs a Hybrid Controller that combines geometry-aware keyframe execution with affordance-guided diffusion for precise and interpretable manipulation. An Affordance Memory Bank continually accumulates successful execution episodes and propagates part-level actionable affordances to unseen articulated parts and configurations. Extensive experiments on ArtiBench show that our ArtiBrain significantly outperforms state-of-the-art multimodal and diffusion-based methods in robustness and generalization. Code and dataset will be released upon acceptance.

Paperid: 3646, https://arxiv.org/pdf/2511.20280.pdf

Abstract:
Recent progress in video generation has led to impressive visual quality, yet current models still struggle to produce results that align with real-world physical principles. To this end, we propose an iterative self-refinement framework that leverages large language models and vision-language models to provide physics-aware guidance for video generation. Specifically, we introduce a multimodal chain-of-thought (MM-CoT) process that refines prompts based on feedback from physical inconsistencies, progressively enhancing generation quality. This method is training-free and plug-and-play, making it readily applicable to a wide range of video generation models. Experiments on the PhyIQ benchmark show that our method improves the Physics-IQ score from 56.31 to 62.38. We hope this work serves as a preliminary exploration of physics-consistent video generation and may offer insights for future research.

Paperid: 3647, https://arxiv.org/pdf/2511.18314.pdf

Abstract:
Multimodal Mixture-of-Experts (MoE) models offer a promising path toward scalable and efficient large vision-language systems. However, existing approaches rely on rigid routing strategies (typically activating a fixed number of experts per token) ignoring the inherent heterogeneity in semantic importance across modalities. This leads to suboptimal compute allocation, where redundant tokens consume as many resources as critical ones. To address this, we propose AnyExperts, a novel on-demand, budget-aware dynamic routing framework that allocates a variable total number of expert slots per token based on its semantic importance. Crucially, to prevent uncontrolled compute growth, the total slots per token are constrained within a fixed range, and each slot is filled by either a real expert or a virtual expert, with the virtual share capped at a small maximum (e.g., 20%). The model then adaptively balances the real-to-virtual ratio per token, assigning more real experts to semantically rich regions and relying more on virtual experts for redundant content. Evaluated across diverse tasks in visual understanding, audio understanding, and NLP understanding, AnyExperts improves performance under the same compute budget. Notably, on general image/video tasks, it achieves comparable accuracy with 40% fewer real expert activations; on text-dense tasks (OCR and NLP), it maintains performance while reducing real expert usage by 10%. These results demonstrate that fine-grained, importance-driven expert allocation significantly enhances both the efficiency and effectiveness of multimodal MoE models.

Paperid: 3648, https://arxiv.org/pdf/2511.18085.pdf

Abstract:
Developing general robot intelligence in open environments requires continual skill learning. Recent Vision-Language-Action (VLA) models leverage massive pretraining data to support diverse manipulation tasks, but they still depend heavily on task-specific fine-tuning, revealing a lack of continual learning capability. Existing continual learning methods are also resource-intensive to scale to VLA models. We propose Stellar VLA, a knowledge-driven continual learning framework with two variants: T-Stellar, modeling task-centric knowledge space, and TS-Stellar, capturing hierarchical task-skill structure. Stellar VLA enables self-supervised knowledge evolution through joint learning of task latent representation and the knowledge space, reducing annotation needs. Knowledge-guided expert routing provide task specialization without extra network parameters, lowering training overhead. Experiments on the LIBERO benchmark and real-world tasks show over 50 percentage average improvement in final success rates relative to baselines. TS-Stellar further excels in complex action inference, and in-depth analyses verify effective knowledge retention and discovery. Our code will be released soon.

Paperid: 3649, https://arxiv.org/pdf/2511.12708.pdf

Abstract:
Understanding where drivers look and why they shift their attention is essential for autonomous systems that read human intent and justify their actions. Most existing models rely on large-scale gaze datasets to learn these patterns; however, such datasets are labor-intensive to collect and time-consuming to curate. We present FSDAM (Few-Shot Driver Attention Modeling), a framework that achieves joint attention prediction and caption generation with approximately 100 annotated examples, two orders of magnitude fewer than existing approaches. Our approach introduces a dual-pathway architecture where separate modules handle spatial prediction and caption generation while maintaining semantic consistency through cross-modal alignment. Despite minimal supervision, FSDAM achieves competitive performance on attention prediction, generates coherent, and context-aware explanations. The model demonstrates robust zero-shot generalization across multiple driving benchmarks. This work shows that effective attention-conditioned generation is achievable with limited supervision, opening new possibilities for practical deployment of explainable driver attention systems in data-constrained scenarios.

Paperid: 3650, https://arxiv.org/pdf/2511.12428.pdf

Abstract:
Vision-Language Models (VLMs) have achieved remarkable progress in multimodal reasoning and generation, yet their high computational demands remain a major challenge. Diffusion Vision-Language Models (DVLMs) are particularly attractive because they enable parallel token decoding, but the large number of visual tokens still significantly hinders their inference efficiency. While visual token pruning has been extensively studied for autoregressive VLMs (AVLMs), it remains largely unexplored for DVLMs. In this work, we propose RedVTP, a response-driven visual token pruning strategy that leverages the inference dynamics of DVLMs. Our method estimates visual token importance using attention from the masked response tokens. Based on the observation that these importance scores remain consistent across steps, RedVTP prunes the less important visual tokens from the masked tokens after the first inference step, thereby maximizing inference efficiency. Experiments show that RedVTP improves token generation throughput of LLaDA-V and LaViDa by up to 186% and 28.05%, respectively, and reduces inference latency by up to 64.97% and 21.87%, without compromising-and in some cases improving-accuracy.

Paperid: 3651, https://arxiv.org/pdf/2511.12371.pdf

Abstract:
The goal of text-to-video retrieval is to search large databases for relevant videos based on text queries. Existing methods have progressed to handling explicit queries where the visual content of interest is described explicitly; however, they fail with implicit queries where identifying videos relevant to the query requires reasoning. We introduce reasoning text-to-video retrieval, a paradigm that extends traditional retrieval to process implicit queries through reasoning while providing object-level grounding masks that identify which entities satisfy the query conditions. Instead of relying on vision-language models directly, we propose representing video content as digital twins, i.e., structured scene representations that decompose salient objects through specialist vision models. This approach is beneficial because it enables large language models to reason directly over long-horizon video content without visual token compression. Specifically, our two-stage framework first performs compositional alignment between decomposed sub-queries and digital twin representations for candidate identification, then applies large language model-based reasoning with just-in-time refinement that invokes additional specialist models to address information gaps. We construct a benchmark of 447 manually created implicit queries with 135 videos (ReasonT2VBench-135) and another more challenging version of 1000 videos (ReasonT2VBench-1000). Our method achieves 81.2% R@1 on ReasonT2VBench-135, outperforming the strongest baseline by greater than 50 percentage points, and maintains 81.7% R@1 on the extended configuration while establishing state-of-the-art results in three conventional benchmarks (MSR-VTT, MSVD, and VATEX).

Paperid: 3652, https://arxiv.org/pdf/2511.06665.pdf

Abstract:
Despite significant progress in pixel-level medical image analysis, existing medical image segmentation models rarely explore medical segmentation and diagnosis tasks jointly. However, it is crucial for patients that models can provide explainable diagnoses along with medical segmentation results. In this paper, we introduce a medical vision-language task named Medical Diagnosis Segmentation (MDS), which aims to understand clinical queries for medical images and generate the corresponding segmentation masks as well as diagnostic results. To facilitate this task, we first present the Multimodal Multi-disease Medical Diagnosis Segmentation (M3DS) dataset, containing diverse multimodal multi-disease medical images paired with their corresponding segmentation masks and diagnosis chain-of-thought, created via an automated diagnosis chain-of-thought generation pipeline. Moreover, we propose Sim4Seg, a novel framework that improves the performance of diagnosis segmentation by taking advantage of the Region-Aware Vision-Language Similarity to Mask (RVLS2M) module. To improve overall performance, we investigate a test-time scaling strategy for MDS tasks. Experimental results demonstrate that our method outperforms the baselines in both segmentation and diagnosis.

Paperid: 3653, https://arxiv.org/pdf/2511.01571.pdf

Abstract:
Vision-Language-Action models (VLAs) are emerging as powerful tools for learning generalizable visuomotor control policies. However, current VLAs are mostly trained on large-scale image-text-action data and remain limited in two key ways: (i) they struggle with pixel-level scene understanding, and (ii) they rely heavily on textual prompts, which reduces their flexibility in real-world settings. To address these challenges, we introduce PixelVLA, the first VLA model designed to support both pixel-level reasoning and multimodal prompting with text and visual inputs. Our approach is built on a new visuomotor instruction tuning framework that integrates a multiscale pixel-aware encoder with a visual prompting encoder. To train PixelVLA effectively, we further propose a two-stage automated annotation pipeline that generates Pixel-160K, a large-scale dataset with pixel-level annotations derived from existing robot data. Experiments on three standard VLA benchmarks and two VLA model variants show that PixelVLA improves manipulation success rates by 10.1%-17.8% over OpenVLA, while requiring only 1.5% of its pretraining cost. These results demonstrate that PixelVLA can be integrated into existing VLAs to enable more accurate, efficient, and versatile robot control in complex environments. The dataset and code will be released as open source.

Paperid: 3654, https://arxiv.org/pdf/2510.25548.pdf

Abstract:
In task and motion planning, high-level task planning is done over an abstraction of the world to enable efficient search in long-horizon robotics problems. However, the feasibility of these task-level plans relies on the downward refinability of the abstraction into continuous motion. When a domain's refinability is poor, task-level plans that appear valid may ultimately fail during motion planning, requiring replanning and resulting in slower overall performance. Prior works mitigate this by encoding refinement issues as constraints to prune infeasible task plans. However, these approaches only add constraints upon refinement failure, expending significant search effort on infeasible branches. We propose VIZ-COAST, a method of leveraging the common-sense spatial reasoning of large pretrained Vision-Language Models to identify issues with downward refinement a priori, bypassing the need to fix these failures during planning. Experiments on two challenging TAMP domains show that our approach is able to extract plausible constraints from images and domain descriptions, drastically reducing planning times and, in some cases, eliminating downward refinement failures altogether, generalizing to a diverse range of instances from the broader domain.

Paperid: 3655, https://arxiv.org/pdf/2510.22936.pdf

Abstract:
Multimodal large language models (MLLMs) have achieved strong performance on vision-language tasks, yet often suffer from inefficiencies due to redundant visual tokens. Existing token merging methods reduce sequence length but frequently disrupt spatial layouts and temporal continuity by disregarding positional relationships. In this work, we propose a novel encoding operator dubbed as \textbf{P}ositional \textbf{P}reservation \textbf{E}mbedding (\textbf{PPE}), which has the main hallmark of preservation of spatiotemporal structure during visual token compression. PPE explicitly introduces the disentangled encoding of 3D positions in the token dimension, enabling each compressed token to encapsulate different positions from multiple original tokens. Furthermore, we show that PPE can effectively support cascade clustering -- a progressive token compression strategy that leads to better performance retention. PPE is a parameter-free and generic operator that can be seamlessly integrated into existing token merging methods without any adjustments. Applied to state-of-the-art token merging framework, PPE achieves consistent improvements of $2\%\sim5\%$ across multiple vision-language benchmarks, including MMBench (general vision understanding), TextVQA (layout understanding) and VideoMME (temporal understanding). These results demonstrate that preserving positional cues is critical for efficient and effective MLLM reasoning.

Paperid: 3656, https://arxiv.org/pdf/2510.22641.pdf

Abstract:
Vision-language Models (VLMs) have made significant strides in visual understanding and query response generation, but often face challenges of high computational cost and inference latency due to autoregressive decoding. In this work, we introduce an imitation-learning-based Self-Speculative Decoding (SSD) framework, named FastVLM, to address these limitations. Our approach employs a lightweight draft model for token generation in an autoregressive manner, while a full model verifies these tokens non-autoregressively. Accepted tokens proceed seamlessly, while rejected tokens are corrected by the full model and used to guide the draft model's refinement. Through an imitation network, FastVLM enhances the draft model by integrating deeper level insights from the full model's architecture. Also, it maintains the performance integrity of the full model while training the draft model, achieving a balance between efficiency and accuracy. Our method speeds up the inference process by 1.55-1.85x as compared to the final layer with minimal loss in performance.

Paperid: 3657, https://arxiv.org/pdf/2510.21862.pdf

Abstract:
Engineering drawings are fundamental to manufacturing communication, serving as the primary medium for conveying design intent, tolerances, and production details. However, interpreting complex multi-view drawings with dense annotations remains challenging using manual methods, generic optical character recognition (OCR) systems, or traditional deep learning approaches, due to varied layouts, orientations, and mixed symbolic-textual content. To address these challenges, this paper proposes a three-stage hybrid framework for the automated interpretation of 2D multi-view engineering drawings using modern detection and vision language models (VLMs). In the first stage, YOLOv11-det performs layout segmentation to localize key regions such as views, title blocks, and notes. The second stage uses YOLOv11-obb for orientation-aware, fine-grained detection of annotations, including measures, GD&T symbols, and surface roughness indicators. The third stage employs two Donut-based, OCR-free VLMs for semantic content parsing: the Alphabetical VLM extracts textual and categorical information from title blocks and notes, while the Numerical VLM interprets quantitative data such as measures, GD&T frames, and surface roughness. Two specialized datasets were developed to ensure robustness and generalization: 1,000 drawings for layout detection and 1,406 for annotation-level training. The Alphabetical VLM achieved an overall F1 score of 0.672, while the Numerical VLM reached 0.963, demonstrating strong performance in textual and quantitative interpretation, respectively. The unified JSON output enables seamless integration with CAD and manufacturing databases, providing a scalable solution for intelligent engineering drawing analysis.

Paperid: 3658, https://arxiv.org/pdf/2510.21722.pdf

Abstract:
Underwater activities like scuba diving enable millions annually to explore marine environments for recreation and scientific research. Maintaining situational awareness and effective communication are essential for diver safety. Traditional underwater communication systems are often bulky and expensive, limiting their accessibility to divers of all levels. While recent systems leverage lightweight smartphones and support text messaging, the messages are predefined and thus restrict context-specific communication. In this paper, we present AquaVLM, a tap-and-send underwater communication system that automatically generates context-aware messages and transmits them using ubiquitous smartphones. Our system features a mobile vision-language model (VLM) fine-tuned on an auto-generated underwater conversation dataset and employs a hierarchical message generation pipeline. We co-design the VLM and transmission, incorporating error-resilient fine-tuning to improve the system's robustness to transmission errors. We develop a VR simulator to enable users to experience AquaVLM in a realistic underwater environment and create a fully functional prototype on the iOS platform for real-world experiments. Both subjective and objective evaluations validate the effectiveness of AquaVLM and highlight its potential for personal underwater communication as well as broader mobile VLM applications.

Paperid: 3659, https://arxiv.org/pdf/2510.17150.pdf

Abstract:
We present OmniVIC, a universal variable impedance controller (VIC) enhanced by a vision language model (VLM), which improves safety and adaptation in any contact-rich robotic manipulation task to enhance safe physical interaction. Traditional VIC have shown advantages when the robot physically interacts with the environment, but lack generalization in unseen, complex, and unstructured safe interactions in universal task scenarios involving contact or uncertainty. To this end, the proposed OmniVIC interprets task context derived reasoning from images and natural language and generates adaptive impedance parameters for a VIC controller. Specifically, the core of OmniVIC is a self-improving Retrieval-Augmented Generation(RAG) and in-context learning (ICL), where RAG retrieves relevant prior experiences from a structured memory bank to inform the controller about similar past tasks, and ICL leverages these retrieved examples and the prompt of current task to query the VLM for generating context-aware and adaptive impedance parameters for the current manipulation scenario. Therefore, a self-improved RAG and ICL guarantee OmniVIC works in universal task scenarios. The impedance parameter regulation is further informed by real-time force/torque feedback to ensure interaction forces remain within safe thresholds. We demonstrate that our method outperforms baselines on a suite of complex contact-rich tasks, both in simulation and on real-world robotic tasks, with improved success rates and reduced force violations. OmniVIC takes a step towards bridging high-level semantic reasoning and low-level compliant control, enabling safer and more generalizable manipulation. Overall, the average success rate increases from 27% (baseline) to 61.4% (OmniVIC).

Paperid: 3660, https://arxiv.org/pdf/2510.16769.pdf

Abstract:
Vision-language models (VLMs) have shown promise in graph understanding, but remain limited by input-token constraints, facing scalability bottlenecks and lacking effective mechanisms to coordinate textual and visual modalities. To address these challenges, we propose GraphVista, a unified framework that enhances both scalability and modality coordination in graph understanding. For scalability, GraphVista organizes graph information hierarchically into a lightweight GraphRAG base, which retrieves only task-relevant textual descriptions and high-resolution visual subgraphs, compressing redundant context while preserving key reasoning elements. For modality coordination, GraphVista introduces a planning agent that routes tasks to the most suitable modality-using the text modality for simple property reasoning and the visual modality for local and structurally complex reasoning grounded in explicit topology. Extensive experiments demonstrate that GraphVista scales to large graphs, up to $200\times$ larger than those used in existing benchmarks, and consistently outperforms existing textual, visual, and fusion-based methods, achieving up to $4.4\times$ quality improvement over the state-of-the-art baselines by fully exploiting the complementary strengths of both modalities.

Paperid: 3661, https://arxiv.org/pdf/2510.14300.pdf

Abstract:
Vision-Language-Action (VLA) models are experiencing rapid development and demonstrating promising capabilities in robotic manipulation tasks. However, scaling up VLA models presents several critical challenges: (1) Training new VLA models from scratch demands substantial computational resources and extensive datasets. Given the current scarcity of robot data, it becomes particularly valuable to fully leverage well-pretrained VLA model weights during the scaling process. (2) Real-time control requires carefully balancing model capacity with computational efficiency. To address these challenges, We propose AdaMoE, a Mixture-of-Experts (MoE) architecture that inherits pretrained weights from dense VLA models, and scales up the action expert by substituting the feedforward layers into sparsely activated MoE layers. AdaMoE employs a decoupling technique that decouples expert selection from expert weighting through an independent scale adapter working alongside the traditional router. This enables experts to be selected based on task relevance while contributing with independently controlled weights, allowing collaborative expert utilization rather than winner-takes-all dynamics. Our approach demonstrates that expertise need not monopolize. Instead, through collaborative expert utilization, we can achieve superior performance while maintaining computational efficiency. AdaMoE consistently outperforms the baseline model across key benchmarks, delivering performance gains of 1.8% on LIBERO and 9.3% on RoboTwin. Most importantly, a substantial 21.5% improvement in real-world experiments validates its practical effectiveness for robotic manipulation tasks.

Paperid: 3662, https://arxiv.org/pdf/2510.12974.pdf

Abstract:
Vision-language models (VLMs) benefit from multiple vision encoders, but naively stacking them yields diminishing returns while multiplying inference costs. We propose SCOPE, a Mixture-of-Encoders (MoEnc) framework that dynamically selects one specialized encoder per image-text pair via instance-level routing, unlike token-level routing in traditional MoE. SCOPE maintains a shared encoder and a pool of routed encoders. A lightweight router uses cross-attention between text prompts and shared visual features to select the optimal encoder from the routed encoders. To train this router, we introduce dual entropy regularization with auxiliary losses to balance dataset-level load distribution with instance-level routing confidence. Remarkably, SCOPE with one shared plus one routed encoder outperforms models using all four extra encoders simultaneously, while reducing compute by 24-49\%. This demonstrates that intelligent encoder selection beats brute-force aggregation, challenging the prevailing paradigm in multi-encoder VLMs.

Paperid: 3663, https://arxiv.org/pdf/2510.11996.pdf

Abstract:
Spatial reasoning in large-scale 3D environments such as warehouses remains a significant challenge for vision-language systems due to scene clutter, occlusions, and the need for precise spatial understanding. Existing models often struggle with generalization in such settings, as they rely heavily on local appearance and lack explicit spatial grounding. In this work, we introduce a dedicated spatial reasoning framework for the Physical AI Spatial Intelligence Warehouse dataset introduced in the Track 3 2025 AI City Challenge. Our approach enhances spatial comprehension by embedding mask dimensions in the form of bounding box coordinates directly into the input prompts, enabling the model to reason over object geometry and layout. We fine-tune the framework across four question categories namely: Distance Estimation, Object Counting, Multi-choice Grounding, and Spatial Relation Inference using task-specific supervision. To further improve consistency with the evaluation system, normalized answers are appended to the GPT response within the training set. Our comprehensive pipeline achieves a final score of 73.0606, placing 4th overall on the public leaderboard. These results demonstrate the effectiveness of structured prompt enrichment and targeted optimization in advancing spatial reasoning for real-world industrial environments.

Paperid: 3664, https://arxiv.org/pdf/2510.11852.pdf

Abstract:
Recent advances in open-source vision-language models (VLMs) offer new opportunities for understanding complex and subjective multimodal phenomena such as sarcasm. In this work, we evaluate seven state-of-the-art VLMs - BLIP2, InstructBLIP, OpenFlamingo, LLaVA, PaliGemma, Gemma3, and Qwen-VL - on their ability to detect multimodal sarcasm using zero-, one-, and few-shot prompting. Furthermore, we evaluate the models' capabilities in generating explanations to sarcastic instances. We evaluate the capabilities of VLMs on three benchmark sarcasm datasets (Muse, MMSD2.0, and SarcNet). Our primary objectives are twofold: (1) to quantify each model's performance in detecting sarcastic image-caption pairs, and (2) to assess their ability to generate human-quality explanations that highlight the visual-textual incongruities driving sarcasm. Our results indicate that, while current models achieve moderate success in binary sarcasm detection, they are still not able to generate high-quality explanations without task-specific finetuning.

Paperid: 3665, https://arxiv.org/pdf/2510.11689.pdf

Abstract:
Learning robotic manipulation policies directly in the real world can be expensive and time-consuming. While reinforcement learning (RL) policies trained in simulation present a scalable alternative, effective sim-to-real transfer remains challenging, particularly for tasks that require precise dynamics. To address this, we propose Phys2Real, a real-to-sim-to-real RL pipeline that combines vision-language model (VLM)-inferred physical parameter estimates with interactive adaptation through uncertainty-aware fusion. Our approach consists of three core components: (1) high-fidelity geometric reconstruction with 3D Gaussian splatting, (2) VLM-inferred prior distributions over physical parameters, and (3) online physical parameter estimation from interaction data. Phys2Real conditions policies on interpretable physical parameters, refining VLM predictions with online estimates via ensemble-based uncertainty quantification. On planar pushing tasks of a T-block with varying center of mass (CoM) and a hammer with an off-center mass distribution, Phys2Real achieves substantial improvements over a domain randomization baseline: 100% vs 79% success rate for the bottom-weighted T-block, 57% vs 23% in the challenging top-weighted T-block, and 15% faster average task completion for hammer pushing. Ablation studies indicate that the combination of VLM and interaction information is essential for success. Project website: https://phys2real.github.io/ .

Paperid: 3666, https://arxiv.org/pdf/2510.08470.pdf

Abstract:
Training vision-language models on cognitively-plausible amounts of data requires rethinking how models integrate multimodal information. Within the constraints of the Vision track for the BabyLM Challenge 2025, we propose a lightweight decoder-based architecture with (1) token-wise dynamic gating for adaptive fusion of linguistic and visual cues, (2) feature modulation and channel attention to maximise the utility of limited visual information and (3) auxiliary contrastive objectives for visual grounding. Evaluation on five benchmarks (BLiMP, BLiMP Supplement, EWoK, Winoground and VQA) shows competitive or superior performance to multimodal baselines. More notably, our dynamic gate discovers interpretable patterns without explicit supervision, favouring visual cues for content words and linguistic cues for function words. While we identify limitations in the Challenge constraints, such as the information bottleneck created by global image embeddings and training instability from the dataset split, our findings establish dynamic gating as a powerful tool for efficient multimodal learning, offering both interpretability and performance even under severe constraints.

Paperid: 3667, https://arxiv.org/pdf/2510.02787.pdf

Abstract:
Text removal is a crucial task in computer vision with applications such as privacy preservation, image editing, and media reuse. While existing research has primarily focused on scene text removal in natural images, limitations in current datasets hinder out-of-domain generalization or accurate evaluation. In particular, widely used benchmarks such as SCUT-EnsText suffer from ground truth artifacts due to manual editing, overly simplistic text backgrounds, and evaluation metrics that do not capture the quality of generated results. To address these issues, we introduce an approach to synthesizing a text removal benchmark applicable to domains other than scene texts. Our dataset features text rendered on complex backgrounds using object-aware placement and vision-language model-generated content, ensuring clean ground truth and challenging text removal scenarios. The dataset is available at https://huggingface.co/datasets/cyberagent/OTR .

Paperid: 3668, https://arxiv.org/pdf/2510.00993.pdf

Abstract:
Autoregressive models excel in sequential modeling and have proven to be effective for vision-language data. However, the spatial nature of visual signals conflicts with the sequential dependencies of next-token prediction, leading to suboptimal results. This work proposes a plug-and-play refinement module to enhance the complex spatial correspondence modeling within the generated visual sequence. This module operates as a post-pretraining step to jointly refine all generated tokens of autoregressive model, enhancing vision-language modeling under a shared sequential prediction framework. By leveraging global context and relationship across the tokens, our method mitigates the error accumulation issue within the sequential generation. Experiments demonstrate that the proposed method improves the generation quality, enhancing the model's ability to produce semantically consistent results.

Paperid: 3669, https://arxiv.org/pdf/2509.25142.pdf

Abstract:
Why do Vision Language Models (VLMs), despite success on standard benchmarks, often fail to match human performance on surprisingly simple visual reasoning tasks? While the underlying computational principles are still debated, we hypothesize that a crucial factor is a deficit in visually-grounded serial processing. To test this hypothesis, we compared human and VLM performance across tasks designed to vary serial processing demands in three distinct domains: geometric reasoning, perceptual enumeration, and mental rotation. Tasks within each domain varied serial processing load by manipulating factors such as geometric concept complexity, perceptual individuation load, and transformation difficulty. Across all domains, our results revealed a consistent pattern: decreased VLM accuracy was strongly correlated with increased human reaction time (used as a proxy for serial processing load). As tasks require more demanding serial processing -- whether composing concepts, enumerating items, or performing mental transformations -- the VLM-human performance gap widens reliably. These findings support our hypothesis, indicating that limitations in serial, visually grounded reasoning represent a fundamental bottleneck that distinguishes current VLMs from humans.

Paperid: 3670, https://arxiv.org/pdf/2509.23050.pdf

Abstract:
Large vision-language models (LVLMs) achieve strong performance on multimodal tasks, yet they often default to their language prior (LP) -- memorized textual patterns from pre-training while under-utilizing visual evidence. Prior analyses of LP mostly rely on input-output probing, which fails to reveal the internal mechanisms governing when and how vision influences model behavior. To address this gap, we present the first systematic analysis of language prior through the lens of chain-of-embedding, which examines the layer-wise representation dynamics within LVLMs. Our analysis reveals a universal phenomenon: each model exhibits a Visual Integration Point (VIP), a critical layer at which visual information begins to meaningfully reshape hidden representations and influence decoding. Building on this observation, we introduce the Total Visual Integration (TVI) estimator, which aggregates representation distance beyond the VIP to quantify how strongly visual query influences response generation. Across 54 model-dataset combinations spanning 9 contemporary LVLMs and 6 benchmarks, we demonstrate that VIP consistently emerges, and that TVI reliably predicts the strength of language prior. This offers a principled toolkit for diagnosing and understanding language prior in LVLMs.

Paperid: 3671, https://arxiv.org/pdf/2509.20941.pdf

Abstract:
Scene graphs (SGs) provide structured relational representations crucial for decoding complex, dynamic surgical environments. This PRISMA-ScR-guided scoping review systematically maps the evolving landscape of SG research in surgery, charting its applications, methodological advancements, and future directions. Our analysis reveals rapid growth, yet uncovers a critical 'data divide': internal-view research (e.g., triplet recognition) almost exclusively uses real-world 2D video, while external-view 4D modeling relies heavily on simulated data, exposing a key translational research gap. Methodologically, the field has advanced from foundational graph neural networks to specialized foundation models that now significantly outperform generalist large vision-language models in surgical contexts. This progress has established SGs as a cornerstone technology for both analysis, such as workflow recognition and automated safety monitoring, and generative tasks like controllable surgical simulation. Although challenges in data annotation and real-time implementation persist, they are actively being addressed through emerging techniques. Surgical SGs are maturing into an essential semantic bridge, enabling a new generation of intelligent systems to improve surgical safety, efficiency, and training.

Paperid: 3672, https://arxiv.org/pdf/2509.18610.pdf

Abstract:
Large vision-language models have driven remarkable progress in open-vocabulary robot policies, e.g., generalist robot manipulation policies, that enable robots to complete complex tasks specified in natural language. Despite these successes, open-vocabulary autonomous drone navigation remains an unsolved challenge due to the scarcity of large-scale demonstrations, real-time control demands of drones for stabilization, and lack of reliable external pose estimation modules. In this work, we present SINGER for language-guided autonomous drone navigation in the open world using only onboard sensing and compute. To train robust, open-vocabulary navigation policies, SINGER leverages three central components: (i) a photorealistic language-embedded flight simulator with minimal sim-to-real gap using Gaussian Splatting for efficient data generation, (ii) an RRT-inspired multi-trajectory generation expert for collision-free navigation demonstrations, and these are used to train (iii) a lightweight end-to-end visuomotor policy for real-time closed-loop control. Through extensive hardware flight experiments, we demonstrate superior zero-shot sim-to-real transfer of our policy to unseen environments and unseen language-conditioned goal objects. When trained on ~700k-1M observation action pairs of language conditioned visuomotor data and deployed on hardware, SINGER outperforms a velocity-controlled semantic guidance baseline by reaching the query 23.33% more on average, and maintains the query in the field of view 16.67% more on average, with 10% fewer collisions.

Paperid: 3673, https://arxiv.org/pdf/2509.16628.pdf

Abstract:
In this study, we introduce Vision-Caption aware Supervised FineTuning (VCASFT), a novel learning paradigm designed to enhance the performance of smaller Vision Language Models(VLMs) on scientific visual question answering(VQA) tasks. VCASFT leverages image captions as zero-shot prompts alongside question-answer pairs and instruction-tunes models to yield significant performance improvements. To comprehensively evaluate VCASFT, we benchmark it on ScienceQA, which consists of questions across diverse languages, subjects, and fields, demonstrating its adaptability and effectiveness in a variety of educational contexts. Additionally, to further demonstrate the effectiveness of this technique on lowresource languages, we developed HiSciVQA, a dataset comprising 2,245 high-quality, hand-annotated Hindi multimodal Q&A pairs. This dataset addresses the critical need for low-resource language Q&A datasets and serves as a foundation for testing VCASFT. Additionally, we introduce a novel LLM-based evaluation scheme to evaluate VLMs on HiSciVQA which offers deeper insights into model effectiveness surpassing traditional n-gram matching accuracy metrics. We are committed to advancing the field by open-sourcing all code files and the HiSciVQA dataset for the research community.

Paperid: 3674, https://arxiv.org/pdf/2509.04243.pdf

Abstract:
Vision Language Models (VLMs) have recently achieved significant progress in bridging visual perception and linguistic reasoning. Recently, OpenAI o3 model introduced a zoom-in search strategy that effectively elicits active perception capabilities in VLMs, improving downstream task performance. However, enabling VLMs to reason effectively over appropriate image regions remains a core challenge in GUI grounding, particularly under high-resolution inputs and complex multi-element visual interactions. In this work, we propose LASER, a self-evolving framework that progressively endows VLMs with multi-step perception capabilities, enabling precise coordinate prediction. Specifically, our approach integrate Monte Carlo quality estimation with Intersection-over-Union (IoU)-based region quality evaluation to jointly encourage both accuracy and diversity in constructing high-quality preference data. This combination explicitly guides the model to focus on instruction-relevant key regions while adaptively allocating reasoning steps based on task complexity. Comprehensive experiments on the ScreenSpot Pro and ScreenSpot-v2 benchmarks demonstrate consistent performance gains, validating the effectiveness of our method. Furthermore, when fine-tuned on GTA1-7B, LASER achieves a score of 55.7 on the ScreenSpot-Pro benchmark, establishing a new state-of-the-art (SoTA) among 7B-scale models.

Paperid: 3675, https://arxiv.org/pdf/2509.02722.pdf

Abstract:
Effective planning requires strong world models, but high-level world models that can understand and reason about actions with semantic and temporal abstraction remain largely underdeveloped. We introduce the Vision Language World Model (VLWM), a foundation model trained for language-based world modeling on natural videos. Given visual observations, the VLWM first infers the overall goal achievements then predicts a trajectory composed of interleaved actions and world state changes. Those targets are extracted by iterative LLM Self-Refine conditioned on compressed future observations represented by Tree of Captions. The VLWM learns both an action policy and a dynamics model, which respectively facilitates reactive system-1 plan decoding and reflective system-2 planning via cost minimization. The cost evaluates the semantic distance between the hypothetical future states given by VLWM roll-outs and the expected goal state, and is measured by a critic model that we trained in a self-supervised manner. The VLWM achieves state-of-the-art Visual Planning for Assistance (VPA) performance on both benchmark evaluations and our proposed PlannerArena human evaluations, where system-2 improves the Elo score by +27% upon system-1. The VLWM models also outperforms strong VLM baselines on RoboVQA and WorldPrediction benchmark.

Paperid: 3676, https://arxiv.org/pdf/2509.00531.pdf

Abstract:
With the rapid advancement of Vision-Language Models (VLMs), GUI-based mobile agents have emerged as a key development direction for intelligent mobile systems. However, existing agent models continue to face significant challenges in real-world task execution, particularly in terms of accuracy and efficiency. To address these limitations, we propose MobiAgent, a comprehensive mobile agent system comprising three core components: the MobiMind-series agent models, the AgentRR acceleration framework, and the MobiFlow benchmarking suite. Furthermore, recognizing that the capabilities of current mobile agents are still limited by the availability of high-quality data, we have developed an AI-assisted agile data collection pipeline that significantly reduces the cost of manual annotation. Compared to both general-purpose LLMs and specialized GUI agent models, MobiAgent achieves state-of-the-art performance in real-world mobile scenarios.

Paperid: 3677, https://arxiv.org/pdf/2508.19972.pdf

Abstract:
Object hallucination in large vision-language models presents a significant challenge to their safe deployment in real-world applications. Recent works have proposed object-level hallucination scores to estimate the likelihood of object hallucination; however, these methods typically adopt either a global or local perspective in isolation, which may limit detection reliability. In this paper, we introduce GLSim, a novel training-free object hallucination detection framework that leverages complementary global and local embedding similarity signals between image and text modalities, enabling more accurate and reliable hallucination detection in diverse scenarios. We comprehensively benchmark existing object hallucination detection methods and demonstrate that GLSim achieves superior detection performance, outperforming competitive baselines by a significant margin.

Paperid: 3678, https://arxiv.org/pdf/2508.12603.pdf

Abstract:
End-to-end autonomous driving systems built on Vision Language Models (VLMs) have shown significant promise, yet their reliance on autoregressive architectures introduces some limitations for real-world applications. The sequential, token-by-token generation process of these models results in high inference latency and cannot perform bidirectional reasoning, making them unsuitable for dynamic, safety-critical environments. To overcome these challenges, we introduce ViLaD, a novel Large Vision Language Diffusion (LVLD) framework for end-to-end autonomous driving that represents a paradigm shift. ViLaD leverages a masked diffusion model that enables parallel generation of entire driving decision sequences, significantly reducing computational latency. Moreover, its architecture supports bidirectional reasoning, allowing the model to consider both past and future simultaneously, and supports progressive easy-first generation to iteratively improve decision quality. We conduct comprehensive experiments on the nuScenes dataset, where ViLaD outperforms state-of-the-art autoregressive VLM baselines in both planning accuracy and inference speed, while achieving a near-zero failure rate. Furthermore, we demonstrate the framework's practical viability through a real-world deployment on an autonomous vehicle for an interactive parking task, confirming its effectiveness and soundness for practical applications.

Paperid: 3679, https://arxiv.org/pdf/2508.11093.pdf

Abstract:
Human-robot collaboration requires robots to quickly infer user intent, provide transparent reasoning, and assist users in achieving their goals. Our recent work introduced GUIDER, our framework for inferring navigation and manipulation intents. We propose augmenting GUIDER with a vision-language model (VLM) and a text-only language model (LLM) to form a semantic prior that filters objects and locations based on the mission prompt. A vision pipeline (YOLO for object detection and the Segment Anything Model for instance segmentation) feeds candidate object crops into the VLM, which scores their relevance given an operator prompt; in addition, the list of detected object labels is ranked by a text-only LLM. These scores weight the existing navigation and manipulation layers of GUIDER, selecting context-relevant targets while suppressing unrelated objects. Once the combined belief exceeds a threshold, autonomy changes occur, enabling the robot to navigate to the desired area and retrieve the desired object, while adapting to any changes in the operator's intent. Future work will evaluate the system on Isaac Sim using a Franka Emika arm on a Ridgeback base, with a focus on real-time assistance.

Paperid: 3680, https://arxiv.org/pdf/2508.05585.pdf

Abstract:
Open-Vocabulary Multi-Label Recognition (OV-MLR) aims to identify multiple seen and unseen object categories within an image, requiring both precise intra-class localization to pinpoint objects and effective inter-class reasoning to model complex category dependencies. While Vision-Language Pre-training (VLP) models offer a strong open-vocabulary foundation, they often struggle with fine-grained localization under weak supervision and typically fail to explicitly leverage structured relational knowledge beyond basic semantics, limiting performance especially for unseen classes. To overcome these limitations, we propose the Dual Adaptive Refinement Transfer (DART) framework. DART enhances a frozen VLP backbone via two synergistic adaptive modules. For intra-class refinement, an Adaptive Refinement Module (ARM) refines patch features adaptively, coupled with a novel Weakly Supervised Patch Selecting (WPS) loss that enables discriminative localization using only image-level labels. Concurrently, for inter-class transfer, an Adaptive Transfer Module (ATM) leverages a Class Relationship Graph (CRG), constructed using structured knowledge mined from a Large Language Model (LLM), and employs graph attention network to adaptively transfer relational information between class representations. DART is the first framework, to our knowledge, to explicitly integrate external LLM-derived relational knowledge for adaptive inter-class transfer while simultaneously performing adaptive intra-class refinement under weak supervision for OV-MLR. Extensive experiments on challenging benchmarks demonstrate that our DART achieves new state-of-the-art performance, validating its effectiveness.

Paperid: 3681, https://arxiv.org/pdf/2508.04028.pdf

Abstract:
Recently, prompt learning has demonstrated remarkable success in adapting pre-trained Vision-Language Models (VLMs) to various downstream tasks such as image classification. However, its application to the downstream Image-Text Retrieval (ITR) task is more challenging. We find that the challenge lies in discriminating both fine-grained attributes and similar subcategories of the downstream data. To address this challenge, we propose Dual prompt Learning with Joint Category-Attribute Reweighting (DCAR), a novel dual-prompt learning framework to achieve precise image-text matching. The framework dynamically adjusts prompt vectors from both semantic and visual dimensions to improve the performance of CLIP on the downstream ITR task. Based on the prompt paradigm, DCAR jointly optimizes attribute and class features to enhance fine-grained representation learning. Specifically, (1) at the attribute level, it dynamically updates the weights of attribute descriptions based on text-image mutual information correlation; (2) at the category level, it introduces negative samples from multiple perspectives with category-matching weighting to learn subcategory distinctions. To validate our method, we construct the Fine-class Described Retrieval Dataset (FDRD), which serves as a challenging benchmark for ITR in downstream data domains. It covers over 1,500 downstream fine categories and 230,000 image-caption pairs with detailed attribute annotations. Extensive experiments on FDRD demonstrate that DCAR achieves state-of-the-art performance over existing baselines.

Paperid: 3682, https://arxiv.org/pdf/2508.02762.pdf

Abstract:
We propose Context-Adaptive Multi-Prompt Embedding, a novel approach to enrich semantic representations in vision-language contrastive learning. Unlike standard CLIP-style models that rely on a single text embedding, our method introduces multiple structured prompts, each containing a distinct adaptive token that captures diverse semantic aspects of the input text. We leverage a pretrained LLM as the text encoder within the CLIP framework, processing all prompts jointly in a single forward pass. The resulting prompt embeddings are combined into a unified text representation, enabling semantically richer alignment with visual features. To further promote semantic diversity and representation quality, we incorporate a diversity regularization loss and a negation-aware loss, encouraging specialization across prompts and improving contrastive discrimination. Our method achieves consistent improvements on both image-text and video-text retrieval benchmarks.

Paperid: 3683, https://arxiv.org/pdf/2507.17462.pdf

Abstract:
Robot imitation learning relies on 4D multi-view sequential images. However, the high cost of data collection and the scarcity of high-quality data severely constrain the generalization and application of embodied intelligence policies like Vision-Language-Action (VLA) models. Data augmentation is a powerful strategy to overcome data scarcity, but methods for editing 4D multi-view sequential images for manipulation tasks are currently lacking. Thus, we propose ERMV (Editing Robotic Multi-View 4D data), a novel data augmentation framework that efficiently edits an entire multi-view sequence based on single-frame editing and robot state conditions. This task presents three core challenges: (1) maintaining geometric and appearance consistency across dynamic views and long time horizons; (2) expanding the working window with low computational costs; and (3) ensuring the semantic integrity of critical objects like the robot arm. ERMV addresses these challenges through a series of innovations. First, to ensure spatio-temporal consistency in motion blur, we introduce a novel Epipolar Motion-Aware Attention (EMA-Attn) mechanism that learns pixel shift caused by movement before applying geometric constraints. Second, to maximize the editing working window, ERMV pioneers a Sparse Spatio-Temporal (STT) module, which decouples the temporal and spatial views and remodels a single-frame multi-view problem through sparse sampling of the views to reduce computational demands. Third, to alleviate error accumulation, we incorporate a feedback intervention Mechanism, which uses a Multimodal Large Language Model (MLLM) to check editing inconsistencies and request targeted expert guidance only when necessary. Extensive experiments demonstrate that ERMV-augmented data significantly boosts the robustness and generalization of VLA models in both simulated and real-world environments.

Paperid: 3684, https://arxiv.org/pdf/2507.10672.pdf

Abstract:
Vision Language Action (VLA) models represent a transformative shift in robotics, with the aim of unifying visual perception, natural language understanding, and embodied control within a single learning framework. This review presents a comprehensive and forward-looking synthesis of the VLA paradigm, with a particular emphasis on robotic manipulation and instruction-driven autonomy. We comprehensively analyze 102 VLA models, 26 foundational datasets, and 12 simulation platforms that collectively shape the development and evaluation of VLAs models. These models are categorized into key architectural paradigms, each reflecting distinct strategies for integrating vision, language, and control in robotic systems. Foundational datasets are evaluated using a novel criterion based on task complexity, variety of modalities, and dataset scale, allowing a comparative analysis of their suitability for generalist policy learning. We introduce a two-dimensional characterization framework that organizes these datasets based on semantic richness and multimodal alignment, showing underexplored regions in the current data landscape. Simulation environments are evaluated for their effectiveness in generating large-scale data, as well as their ability to facilitate transfer from simulation to real-world settings and the variety of supported tasks. Using both academic and industrial contributions, we recognize ongoing challenges and outline strategic directions such as scalable pretraining protocols, modular architectural design, and robust multimodal alignment strategies. This review serves as both a technical reference and a conceptual roadmap for advancing embodiment and robotic control, providing insights that span from dataset generation to real world deployment of generalist robotic agents.

Paperid: 3685, https://arxiv.org/pdf/2507.07605.pdf

Abstract:
We study the use of image-based Vision-Language Models (VLMs) for open-vocabulary segmentation of lidar scans in driving settings. Classically, image semantics can be back-projected onto 3D point clouds. Yet, resulting point labels are noisy and sparse. We consolidate these labels to enforce both spatio-temporal consistency and robustness to image-level augmentations. We then train a 3D network based on these refined labels. This simple method, called LOSC, outperforms the SOTA of zero-shot open-vocabulary semantic and panoptic segmentation on both nuScenes and SemanticKITTI, with significant margins.

Paperid: 3686, https://arxiv.org/pdf/2506.17811.pdf

Abstract:
Vision-Language-Action (VLA) models have demonstrated remarkable capabilities in visuomotor control, yet ensuring their robustness in unstructured real-world environments remains a persistent challenge. In this paper, we investigate test-time scaling through the lens of sampling and verification as means to enhance the robustness and generalization of VLAs. We first demonstrate that the relationship between action error and the number of generated samples follows an exponentiated power law across a range of VLAs, indicating the existence of inference-time scaling laws. Building on these insights, we introduce RoboMonkey, a test-time scaling framework for VLAs. At deployment, RoboMonkey samples a small set of actions from a VLA, applies Gaussian perturbation and majority voting to construct an action proposal distribution, and then uses a Vision Language Model (VLM)-based verifier to select the optimal action. We propose a synthetic data generation pipeline for training such VLM-based action verifiers, and demonstrate that scaling the synthetic dataset consistently improves verification and downstream accuracy. Through extensive simulated and hardware experiments, we show that pairing existing VLAs with RoboMonkey yields significant performance gains, achieving a 25% absolute improvement on out-of-distribution tasks and 9% on in-distribution tasks. Additionally, when adapting to new robot setups, we show that fine-tuning both VLAs and action verifiers yields a 7% performance increase compared to fine-tuning VLAs alone.

Paperid: 3687, https://arxiv.org/pdf/2506.17417.pdf

Abstract:
Recent advances in large language models (LLMs) have demonstrated that inference-time computation techniques, such as decoding-time scaling and self-refinement, can significantly enhance reasoning capabilities without relying on external knowledge. A key driver of this success is the emergence of self-correction and self-verification behaviors, often elicited through reinforcement learning (RL). In this paper, we investigate whether these inference-time techniques extend effectively to vision-language models (VLMs), particularly those trained with RL. We find that while decoding strategies such as majority voting and best-of-N selection with self-verification all improve VLM reasoning performance, generation-reliant methods such as the former achieve significantly higher gains versus verification-reliant methods such as the latter. Additionally, the self-correction behavior often associated with RL-tuned models, such as aha moment, does not lead to measurable gains. We show via extensive experimentation within the inference-time scaling framework to identify a key root cause: RL-trained VLMs still lack robust self-verification capabilities across both visual and textual modalities.

Paperid: 3688, https://arxiv.org/pdf/2506.17374.pdf

Abstract:
Efficient and accurate extraction of key information from 2D engineering drawings is essential for advancing digital manufacturing workflows. Such information includes geometric dimensioning and tolerancing (GD&T), measures, material specifications, and textual annotations. Manual extraction is slow and labor-intensive, while generic OCR models often fail due to complex layouts, engineering symbols, and rotated text, leading to incomplete and unreliable outputs. These limitations result in incomplete and unreliable outputs. To address these challenges, we propose a hybrid vision-language framework that integrates a rotation-aware object detection model (YOLOv11-obb) with a transformer-based vision-language parser. Our structured pipeline applies YOLOv11-OBB to localize annotations and extract oriented bounding box (OBB) patches, which are then parsed into structured outputs using a fine-tuned, lightweight vision-language model (VLM). We curate a dataset of 1,367 2D mechanical drawings annotated across nine key categories. YOLOv11-OBB is trained on this dataset to detect OBBs and extract annotation patches. These are parsed using two open-source VLMs: Donut and Florence-2. Both models are lightweight and well-suited for specialized industrial tasks under limited computational overhead. Following fine-tuning of both models on the curated dataset of image patches paired with structured annotation labels, a comparative experiment is conducted to evaluate parsing performance across four key metrics. Donut outperforms Florence-2, achieving 88.5% precision, 99.2% recall, and a 93.5% F1-score, with a hallucination rate of 11.5%. Finally, a case study demonstrates how the extracted structured information supports downstream manufacturing tasks such as process and tool selection, showcasing the practical utility of the proposed framework in modernizing 2D drawing interpretation.

Paperid: 3689, https://arxiv.org/pdf/2506.16796.pdf

Abstract:
Real-World Image Super-Resolution is one of the most challenging task in image restoration. However, existing methods struggle with an accurate understanding of degraded image content, leading to reconstructed results that are both low-fidelity and unnatural. We present RealSR-R1 in this work, which empowers the RealSR models with understanding and reasoning capabilities. Inspired by the success of Chain of Thought (CoT) in large language models (LLMs), we simulate the human process of handling degraded images and propose the VLCoT framework, which integrates vision and language reasoning. The framework aims to precisely restore image details by progressively generating more comprehensive text and higher-resolution images. To overcome the challenge of traditional supervised learning CoT failing to generalize to real-world scenarios, we introduce, for the first time, Group Relative Policy Optimization (GRPO) into the Real-World Image Super-Resolution task. We propose VLCoT-GRPO as a solution, which designs four reward functions: (1) Format reward, used to standardize the CoT process; (2) Degradation reward, to incentivize accurate degradation estimation; (3) Understanding reward, to ensure the accuracy of the generated content; and (4) Generation reward, where we propose using a visual expert model to evaluate the quality of generated images, encouraging the model to generate more realistic images. Extensive experiments demonstrate that our proposed RealSR-R1 can generate realistic details and accurately understand image content, particularly in semantically rich scenes or images with severe degradation.

Paperid: 3690, https://arxiv.org/pdf/2506.14100.pdf

Abstract:
Vision-Language Models (VLMs) have demonstrated notable promise in autonomous driving by offering the potential for multimodal reasoning through pretraining on extensive image-text pairs. However, adapting these models from broad web-scale data to the safety-critical context of driving presents a significant challenge, commonly referred to as domain shift. Existing simulation-based and dataset-driven evaluation methods, although valuable, often fail to capture the full complexity of real-world scenarios and cannot easily accommodate repeatable closed-loop testing with flexible scenario manipulation. In this paper, we introduce a hierarchical real-world test platform specifically designed to evaluate VLM-integrated autonomous driving systems. Our approach includes a modular, low-latency on-vehicle middleware that allows seamless incorporation of various VLMs, a clearly separated perception-planning-control architecture that can accommodate both VLM-based and conventional modules, and a configurable suite of real-world testing scenarios on a closed track that facilitates controlled yet authentic evaluations. We demonstrate the effectiveness of the proposed platform`s testing and evaluation ability with a case study involving a VLM-enabled autonomous vehicle, highlighting how our test framework supports robust experimentation under diverse conditions.

Paperid: 3691, https://arxiv.org/pdf/2506.10474.pdf

Abstract:
The growing sophistication of deepfakes presents substantial challenges to the integrity of media and the preservation of public trust. Concurrently, vision-language models (VLMs), large language models enhanced with visual reasoning capabilities, have emerged as promising tools across various domains, sparking interest in their applicability to deepfake detection. This study conducts a structured zero-shot evaluation of four prominent VLMs: ChatGPT, Claude, Gemini, and Grok, focusing on three primary deepfake types: faceswap, reenactment, and synthetic generation. Leveraging a meticulously assembled benchmark comprising authentic and manipulated images from diverse sources, we evaluate each model's classification accuracy and reasoning depth. Our analysis indicates that while VLMs can produce coherent explanations and detect surface-level anomalies, they are not yet dependable as standalone detection systems. We highlight critical failure modes, such as an overemphasis on stylistic elements and vulnerability to misleading visual patterns like vintage aesthetics. Nevertheless, VLMs exhibit strengths in interpretability and contextual analysis, suggesting their potential to augment human expertise in forensic workflows. These insights imply that although general-purpose models currently lack the reliability needed for autonomous deepfake detection, they hold promise as integral components in hybrid or human-in-the-loop detection frameworks.

Paperid: 3692, https://arxiv.org/pdf/2506.09445.pdf

Abstract:
We address the problem of video question answering (video QA) with temporal grounding in a weakly supervised setup, without any temporal annotations. Given a video and a question, we generate an open-ended answer grounded with the start and end time. For this task, we propose TOGA: a vision-language model for Temporally Grounded Open-Ended Video QA with Weak Supervision. We instruct-tune TOGA to jointly generate the answer and the temporal grounding. We operate in a weakly supervised setup where the temporal grounding annotations are not available. We generate pseudo labels for temporal grounding and ensure the validity of these labels by imposing a consistency constraint between the question of a grounding response and the response generated by a question referring to the same temporal segment. We notice that jointly generating the answers with the grounding improves performance on question answering as well as grounding. We evaluate TOGA on grounded QA and open-ended QA tasks. For grounded QA, we consider the NExT-GQA benchmark which is designed to evaluate weakly supervised grounded question answering. For open-ended QA, we consider the MSVD-QA and ActivityNet-QA benchmarks. We achieve state-of-the-art performance for both tasks on these benchmarks.

Paperid: 3693, https://arxiv.org/pdf/2506.08566.pdf

Abstract:
Vision-Language Navigation (VLN) enables intelligent agents to navigate environments by integrating visual perception and natural language instructions, yet faces significant challenges due to the scarcity of fine-grained cross-modal alignment annotations. Existing datasets primarily focus on global instruction-trajectory matching, neglecting sub-instruction-level and entity-level alignments critical for accurate navigation action decision-making. To address this limitation, we propose FCA-NIG, a generative framework that automatically constructs navigation instructions with dual-level fine-grained cross-modal annotations. In this framework, an augmented trajectory is first divided into sub-trajectories, which are then processed through GLIP-based landmark detection, crafted instruction construction, OFA-Speaker based R2R-like instruction generation, and CLIP-powered entity selection, generating sub-instruction-trajectory pairs with entity-landmark annotations. Finally, these sub-pairs are aggregated to form a complete instruction-trajectory pair. The framework generates the FCA-R2R dataset, the first large-scale augmentation dataset featuring precise sub-instruction-sub-trajectory and entity-landmark alignments. Extensive experiments demonstrate that training with FCA-R2R significantly improves the performance of multiple state-of-the-art VLN agents, including SF, EnvDrop, RecBERT, and HAMT. Incorporating sub-instruction-trajectory alignment enhances agents' state awareness and decision accuracy, while entity-landmark alignment further boosts navigation performance and generalization. These results highlight the effectiveness of FCA-NIG in generating high-quality, scalable training data without manual annotation, advancing fine-grained cross-modal learning in complex navigation tasks.

Paperid: 3694, https://arxiv.org/pdf/2506.00613.pdf

Abstract:
Robotics has broad applications from automating house chores to taking care of patients. However, evaluating robot control policies is challenging, as real-world testing is expensive, while handcrafted simulations often fail to accurately reflect real-world conditions, resulting in poor correlation between simulated evaluation and real-world outcomes. In this work, we investigate World-model-based Policy Evaluation (WPE). We first train an action-conditioned video generation model as a proxy to real-world environments. To enable efficient rollouts of hundreds of interactive steps while mitigating error accumulation in the world model, we propose an inference scheme which we call Blockwise-Autoregressive Diffusion Transformer with adjustable context and decoding horizon lengths. To ensure that the world model indeed follows action input, we propose metrics based on the agreement between the ground truth video and generated video conditioned on the same sequence of actions to evaluate the world model. We then use the world model for policy evaluation by performing Monte Carlo rollouts in the world model while employing a vision-language model (VLM) as a reward function. Interestingly, we found that WPE tends to underestimate the policy values for in-distribution actions and overestimate policy values for out-of-distribution actions. Nevertheless, WPE preserves the relative rankings of different policies. In emulating real robot executions, WPE achieves high fidelity in mimicing robot arm movements as in real videos, while emulating highly realistic object interaction remains challenging. Despite this limitation, we show that a world model can serve as a starting point for evaluating robot policies before real-world deployment.

Paperid: 3695, https://arxiv.org/pdf/2505.21652.pdf

Abstract:
Fine-grained robot manipulation, such as lifting and rotating a bottle to display the label on the cap, requires robust reasoning about object parts and their relationships with intended tasks. Despite recent advances in training general-purpose robot manipulation policies guided by language instructions, there is a notable lack of large-scale datasets for fine-grained manipulation tasks with part-level instructions and diverse 3D object instances annotated with part-level labels. In this work, we introduce PartInstruct, the first large-scale benchmark for training and evaluating fine-grained robot manipulation models using part-level instructions. PartInstruct comprises 513 object instances across 14 categories, each annotated with part-level information, and 1302 fine-grained manipulation tasks organized into 16 task classes. Our training set consists of over 10,000 expert demonstrations synthesized in a 3D simulator, where each demonstration is paired with a high-level task instruction, a chain of base part-based skill instructions, and ground-truth 3D information about the object and its parts. Additionally, we designed a comprehensive test suite to evaluate the generalizability of learned policies across new states, objects, and tasks. We evaluated several state-of-the-art robot manipulation approaches, including end-to-end vision-language policy learning and bi-level planning models for robot manipulation on our benchmark. The experimental results reveal that current models struggle to robustly ground part concepts and predict actions in 3D space, and face challenges when manipulating object parts in long-horizon tasks.

Paperid: 3696, https://arxiv.org/pdf/2505.19381.pdf

Abstract:
Research interest in end-to-end autonomous driving has surged owing to its fully differentiable design integrating modular tasks, i.e. perception, prediction and planing, which enables optimization in pursuit of the ultimate goal. Despite the great potential of the end-to-end paradigm, existing methods suffer from several aspects including expensive BEV (bird's eye view) computation, action diversity, and sub-optimal decision in complex real-world scenarios. To address these challenges, we propose a novel hybrid sparse-dense diffusion policy, empowered by a Vision-Language Model (VLM), called Diff-VLA. We explore the sparse diffusion representation for efficient multi-modal driving behavior. Moreover, we rethink the effectiveness of VLM driving decision and improve the trajectory generation guidance through deep interaction across agent, map instances and VLM output. Our method shows superior performance in Autonomous Grand Challenge 2025 which contains challenging real and reactive synthetic scenarios. Our methods achieves 45.0 PDMS.

Paperid: 3697, https://arxiv.org/pdf/2505.16875.pdf

Abstract:
Continual post-training adapts a single text-to-image diffusion model to learn new tasks without incurring the cost of separate models, but naive post-training causes forgetting of pretrained knowledge and undermines zero-shot compositionality. We observe that the absence of a standardized evaluation protocol hampers related research for continual post-training. To address this, we introduce T2I-ConBench, a unified benchmark for continual post-training of text-to-image models. T2I-ConBench focuses on two practical scenarios, item customization and domain enhancement, and analyzes four dimensions: (1) retention of generality, (2) target-task performance, (3) catastrophic forgetting, and (4) cross-task generalization. It combines automated metrics, human-preference modeling, and vision-language QA for comprehensive assessment. We benchmark ten representative methods across three realistic task sequences and find that no approach excels on all fronts. Even joint "oracle" training does not succeed for every task, and cross-task generalization remains unsolved. We release all datasets, code, and evaluation tools to accelerate research in continual post-training for text-to-image models.

Paperid: 3698, https://arxiv.org/pdf/2505.14462.pdf

Abstract:
As vision-language models (VLMs) become increasingly integrated into daily life, the need for accurate visual culture understanding is becoming critical. Yet, these models frequently fall short in interpreting cultural nuances effectively. Prior work has demonstrated the effectiveness of retrieval-augmented generation (RAG) in enhancing cultural understanding in text-only settings, while its application in multimodal scenarios remains underexplored. To bridge this gap, we introduce RAVENEA (Retrieval-Augmented Visual culturE uNdErstAnding), a new benchmark designed to advance visual culture understanding through retrieval, focusing on two tasks: culture-focused visual question answering (cVQA) and culture-informed image captioning (cIC). RAVENEA extends existing datasets by integrating over 10,000 Wikipedia documents curated and ranked by human annotators. With RAVENEA, we train and evaluate seven multimodal retrievers for each image query, and measure the downstream impact of retrieval-augmented inputs across fourteen state-of-the-art VLMs. Our results show that lightweight VLMs, when augmented with culture-aware retrieval, outperform their non-augmented counterparts (by at least 3.2% absolute on cVQA and 6.2% absolute on cIC). This highlights the value of retrieval-augmented methods and culturally inclusive benchmarks for multimodal understanding.

Paperid: 3699, https://arxiv.org/pdf/2505.11839.pdf

Abstract:
Counterfactual reasoning has emerged as a crucial technique for generalizing the reasoning capabilities of large language models (LLMs). By generating and analyzing counterfactual scenarios, researchers can assess the adaptability and reliability of model decision-making. Although prior work has shown that LLMs often struggle with counterfactual reasoning, it remains unclear which factors most significantly impede their performance across different tasks and modalities. In this paper, we propose a decompositional strategy that breaks down the counterfactual generation from causality construction to the reasoning over counterfactual interventions. To support decompositional analysis, we investigate 11 datasets spanning diverse tasks, including natural language understanding, mathematics, programming, and vision-language tasks. Through extensive evaluations, we characterize LLM behavior across each decompositional stage and identify how modality type and intermediate reasoning influence performance. By establishing a structured framework for analyzing counterfactual reasoning, this work contributes to the development of more reliable LLM-based reasoning systems and informs future elicitation strategies.

Paperid: 3700, https://arxiv.org/pdf/2504.18453.pdf

Abstract:
Radiology report generation is critical for efficiency but current models lack the structured reasoning of experts, hindering clinical trust and explainability by failing to link visual findings to precise anatomical locations. This paper introduces BoxMed-RL, a groundbreaking unified training framework for generating spatially verifiable and explainable radiology reports. Built on a large vision-language model, BoxMed-RL revolutionizes report generation through two integrated phases: (1) In the Pretraining Phase, we refine the model via medical concept learning, using Chain-of-Thought supervision to internalize the radiologist-like workflow, followed by spatially verifiable reinforcement, which applies reinforcement learning to align medical findings with bounding boxes. (2) In the Downstream Adapter Phase, we freeze the pretrained weights and train a downstream adapter to ensure fluent and clinically credible reports. This framework precisely mimics radiologists' workflow, compelling the model to connect high-level medical concepts with definitive anatomical evidence. Extensive experiments on public datasets demonstrate that BoxMed-RL achieves an average 7% improvement in both METEOR and ROUGE-L metrics compared to state-of-the-art methods. An average 5% improvement in large language model-based metrics further underscores BoxMed-RL's robustness in generating high-quality radiology reports.

Paperid: 3701, https://arxiv.org/pdf/2504.11967.pdf

Abstract:
Unmanned Aerial Vehicles (UAVs) are indispensable for infrastructure inspection, surveillance, and related tasks, yet they also introduce critical security challenges. This survey provides a wide-ranging examination of the anti-UAV domain, centering on three core objectives-classification, detection, and tracking-while detailing emerging methodologies such as diffusion-based data synthesis, multi-modal fusion, vision-language modeling, self-supervised learning, and reinforcement learning. We systematically evaluate state-of-the-art solutions across both single-modality and multi-sensor pipelines (spanning RGB, infrared, audio, radar, and RF) and discuss large-scale as well as adversarially oriented benchmarks. Our analysis reveals persistent gaps in real-time performance, stealth detection, and swarm-based scenarios, underscoring pressing needs for robust, adaptive anti-UAV systems. By highlighting open research directions, we aim to foster innovation and guide the development of next-generation defense strategies in an era marked by the extensive use of UAVs.

Paperid: 3702, https://arxiv.org/pdf/2504.11101.pdf

Abstract:
The Optical Character Recognition (OCR) task is important for evaluating Vision-Language Models (VLMs) and providing high-quality data sources for LLM training data. While state-of-the-art VLMs show improved average OCR accuracy, they still struggle with sample-level quality degradation and lack reliable automatic detection of low-quality outputs. We introduce Consensus Entropy (CE), a training-free post-inference method that quantifies OCR uncertainty by aggregating outputs from multiple VLMs. Our approach exploits a key insight: correct VLM OCR predictions converge in output space while errors diverge. We develop a lightweight multi-model framework that effectively identifies problematic samples, selects the best outputs and combines model strengths. Experiments across multiple OCR benchmarks and VLMs demonstrate that CE outperforms VLM-as-judge approaches and single-model baselines at the same cost and achieves state-of-the-art results across multiple metrics. For instance, our solution demonstrates: achieving 15.2% higher F1 scores than VLM-as-judge methods in quality verification, delivering 6.0% accuracy gains on mathematical calculation tasks, and requiring rephrasing only 7.3% of inputs while maintaining overall performance. Notably, the entire process requires neither training nor supervision while maintaining plug-and-play functionality throughout.

Paperid: 3703, https://arxiv.org/pdf/2504.03735.pdf

Abstract:
Multimodal Language Models (MMLMs) typically undergo post-training alignment to prevent harmful content generation. However, these alignment stages focus primarily on the assistant role, leaving the user role unaligned, and stick to a fixed input prompt structure of special tokens, leaving the model vulnerable when inputs deviate from these expectations. We introduce Role-Modality Attacks (RMA), a novel class of adversarial attacks that exploit role confusion between the user and assistant and alter the position of the image token to elicit harmful outputs. Unlike existing attacks that modify query content, RMAs manipulate the input structure without altering the query itself. We systematically evaluate these attacks across multiple Vision Language Models (VLMs) on eight distinct settings, showing that they can be composed to create stronger adversarial prompts, as also evidenced by their increased projection in the negative refusal direction in the residual stream, a property observed in prior successful attacks. Finally, for mitigation, we propose an adversarial training approach that makes the model robust against input prompt perturbations. By training the model on a range of harmful and benign prompts all perturbed with different RMA settings, it loses its sensitivity to Role Confusion and Modality Manipulation attacks and is trained to only pay attention to the content of the query in the input prompt structure, effectively reducing Attack Success Rate (ASR) while preserving the model's general utility.

Paperid: 3704, https://arxiv.org/pdf/2503.23137.pdf

Abstract:
Understanding humor-particularly when it involves complex, contradictory narratives that require comparative reasoning-remains a significant challenge for large vision-language models (VLMs). This limitation hinders AI's ability to engage in human-like reasoning and cultural expression. In this paper, we investigate this challenge through an in-depth analysis of comics that juxtapose panels to create humor through contradictions. We introduce the YesBut (V2), a novel benchmark with 1,262 comic images from diverse multilingual and multicultural contexts, featuring comprehensive annotations that capture various aspects of narrative understanding. Using this benchmark, we systematically evaluate a wide range of VLMs through four complementary tasks spanning from surface content comprehension to deep narrative reasoning, with particular emphasis on comparative reasoning between contradictory elements. Our extensive experiments reveal that even the most advanced models significantly underperform compared to humans, with common failures in visual perception, key element identification, comparative analysis and hallucinations. We further investigate text-based training strategies and social knowledge augmentation methods to enhance model performance. Our findings not only highlight critical weaknesses in VLMs' understanding of cultural and creative expressions but also provide pathways toward developing context-aware models capable of deeper narrative understanding though comparative reasoning.

Paperid: 3705, https://arxiv.org/pdf/2503.07065.pdf

Abstract:
While state-of-the-art vision-language models (VLMs) have demonstrated remarkable capabilities in complex visual-text tasks, their success heavily relies on massive model scaling, limiting their practical deployment. Small-scale VLMs offer a more practical alternative but face significant challenges when trained with traditional supervised fine-tuning (SFT), particularly in two aspects: out-of-domain (OOD) generalization and reasoning abilities, which significantly lags behind the contemporary Large language models (LLMs). To address these challenges, we propose Curriculum Reinforcement Finetuning (Curr-ReFT), a novel post-training paradigm specifically designed for small-scale VLMs. Inspired by the success of reinforcement learning in LLMs, Curr-ReFT comprises two sequential stages: (1) Curriculum Reinforcement Learning, which ensures steady progression of model capabilities through difficulty-aware reward design, transitioning from basic visual perception to complex reasoning tasks; and (2) Rejected Sampling-based Self-improvement, which maintains the fundamental capabilities of VLMs through selective learning from high-quality multimodal and language examples. Extensive experiments demonstrate that models trained with Curr-ReFT paradigm achieve state-of-the-art performance across various visual tasks in both in-domain and out-of-domain settings. Moreover, our Curr-ReFT enhanced 3B model matches the performance of 32B-parameter models, demonstrating that efficient training paradigms can effectively bridge the gap between small and large models.

Paperid: 3706, https://arxiv.org/pdf/2503.04852.pdf

Abstract:
True intelligence hinges on the ability to uncover and leverage hidden causal relations. Despite significant progress in AI and computer vision (CV), there remains a lack of benchmarks for assessing models' abilities to infer latent causality from complex visual data. In this paper, we introduce \textsc{\textbf{Causal3D}}, a novel and comprehensive benchmark that integrates structured data (tables) with corresponding visual representations (images) to evaluate causal reasoning. Designed within a systematic framework, Causal3D comprises 19 3D-scene datasets capturing diverse causal relations, views, and backgrounds, enabling evaluations across scenes of varying complexity. We assess multiple state-of-the-art methods, including classical causal discovery, causal representation learning, and large/vision-language models (LLMs/VLMs). Our experiments show that as causal structures grow more complex without prior knowledge, performance declines significantly, highlighting the challenges even advanced methods face in complex causal scenarios. Causal3D serves as a vital resource for advancing causal reasoning in CV and fostering trustworthy AI in critical domains.

Paperid: 3707, https://arxiv.org/pdf/2501.15558.pdf

Abstract:
Multimodal large language models (MLLMs) have shown impressive capabilities across various domains, excelling in processing and understanding information from multiple modalities. Despite the rapid progress made previously, insufficient OCR ability hinders MLLMs from excelling in text-related tasks. In this paper, we present \textbf{Ocean-OCR}, a 3B MLLM with state-of-the-art performance on various OCR scenarios and comparable understanding ability on general tasks. We employ Native Resolution ViT to enable variable resolution input and utilize a substantial collection of high-quality OCR datasets to enhance the model performance. We demonstrate the superiority of Ocean-OCR through comprehensive experiments on open-source OCR benchmarks and across various OCR scenarios. These scenarios encompass document understanding, scene text recognition, and handwritten recognition, highlighting the robust OCR capabilities of Ocean-OCR. Note that Ocean-OCR is the first MLLM to outperform professional OCR models such as TextIn and PaddleOCR.

Paperid: 3708, https://arxiv.org/pdf/2501.01377.pdf

Abstract:
Existing Medical Large Vision-Language Models (Med-LVLMs), encapsulating extensive medical knowledge, demonstrate excellent capabilities in understanding medical images. However, there remain challenges in visual localization in medical images, which is crucial for abnormality detection and interpretation. To address these issues, we propose a novel UMed-LVLM designed to unveil medical abnormalities. Specifically, we collect a Medical Abnormalities Unveiling (MAU) dataset and propose a two-stage training method for UMed-LVLM training. To collect MAU dataset, we propose a prompt method utilizing the GPT-4V to generate diagnoses based on identified abnormal areas in medical images. Moreover, the two-stage training method includes Abnormal-Aware Instruction Tuning and Abnormal-Aware Rewarding, comprising Relevance Reward, Abnormal Localization Reward and Vision Relevance Reward. Experimental results demonstrate that our UMed-LVLM significantly outperforms existing Med-LVLMs in identifying and understanding medical abnormalities, achieving a 58% improvement over the baseline. In addition, this work shows that enhancing the abnormality detection capabilities of Med-LVLMs significantly improves their understanding of medical images and generalization capability.

Paperid: 3709, https://arxiv.org/pdf/2412.20622.pdf

Abstract:
Large Vision-Language Models (LVLMs) have demonstrated remarkable performance in complex multimodal tasks. However, these models still suffer from hallucinations, particularly when required to implicitly recognize or infer diverse visual entities from images for complex vision-language tasks. To address this challenge, we propose HALLUCINOGEN, a novel visual question answering (VQA) benchmark that employs contextual reasoning prompts as hallucination attacks to evaluate the extent of hallucination in state-of-the-art LVLMs. Our benchmark provides a comprehensive study of the implicit reasoning capabilities of these models by first categorizing visual entities based on the ease of recognition in an image as either salient (prominent, visibly recognizable objects such as a car) or latent entities (such as identifying a disease from a chest X-ray), which are not readily visible and require domain knowledge or contextual reasoning for accurate inference. Next, we design hallucination attacks for both types of entities to assess hallucinations in LVLMs while performing various vision-language tasks, such as locating or reasoning about specific entities within an image, where models must perform implicit reasoning by verifying the existence of the queried entity within the image before generating responses. Finally, our extensive evaluations of eleven LVLMs, including powerful open-source models (like LLaMA-3.2 and DeepSeek-V2), commercial models like Gemini, and two hallucination mitigation strategies across multiple datasets, demonstrate that current LVLMs remain susceptible to hallucination attacks.

Paperid: 3710, https://arxiv.org/pdf/2412.18150.pdf

Abstract:
Recently, Text-to-Image (T2I) generation models have achieved significant advancements. Correspondingly, many automated metrics have emerged to evaluate the image-text alignment capabilities of generative models. However, the performance comparison among these automated metrics is limited by existing small datasets. Additionally, these datasets lack the capacity to assess the performance of automated metrics at a fine-grained level. In this study, we contribute an EvalMuse-40K benchmark, gathering 40K image-text pairs with fine-grained human annotations for image-text alignment-related tasks. In the construction process, we employ various strategies such as balanced prompt sampling and data re-annotation to ensure the diversity and reliability of our benchmark. This allows us to comprehensively evaluate the effectiveness of image-text alignment metrics for T2I models. Meanwhile, we introduce two new methods to evaluate the image-text alignment capabilities of T2I models: FGA-BLIP2 which involves end-to-end fine-tuning of a vision-language model to produce fine-grained image-text alignment scores and PN-VQA which adopts a novel positive-negative VQA manner in VQA models for zero-shot fine-grained evaluation. Both methods achieve impressive performance in image-text alignment evaluations. We also use our methods to rank current AIGC models, in which the results can serve as a reference source for future study and promote the development of T2I generation. The data and code will be made publicly available.

Paperid: 3711, https://arxiv.org/pdf/2412.14613.pdf

Abstract:
Vision-language models (VLMs) have shown impressive abilities across a range of multi-modal tasks. However, existing metrics for evaluating the quality of text generated by VLMs typically focus on an overall evaluation for a specific task, such as image captioning. While the overall evaluation is essential for any task, the criteria prioritized can differ depending on the task, making it challenging for current metrics to adapt to multi-task scenarios. To address this limitation, we propose HarmonicEval, a reference-free comprehensive evaluation metric that aggregates criterion-wise scores to produce the overall score in a bottom-up manner. Furthermore, we construct the Multi-task Multi-criteria Human Evaluation (MMHE) dataset, which comprises 18,000 expert human judgments across four multi-modal tasks. Our experiments demonstrate that HarmonicEval achieves higher correlations with human judgments than conventional metrics while providing numerical scores for each criterion.

Paperid: 3712, https://arxiv.org/pdf/2412.06190.pdf

Abstract:
Benefiting from the generalization capability of CLIP, recent vision language pre-training (VLP) models have demonstrated an impressive ability to capture virtually any visual concept in daily images. However, due to the presence of unseen categories in open-vocabulary settings, existing algorithms struggle to effectively capture strong semantic correlations between categories, resulting in sub-optimal performance on the open-vocabulary multi-label recognition (OV-MLR). Furthermore, the substantial variation in the number of discriminative areas across diverse object categories is misaligned with the fixed-number patch matching used in current methods, introducing noisy visual cues that hinder the accurate capture of target semantics. To tackle these challenges, we propose a novel category-adaptive cross-modal semantic refinement and transfer (C$^2$SRT) framework to explore the semantic correlation both within each category and across different categories, in a category-adaptive manner. The proposed framework consists of two complementary modules, i.e., intra-category semantic refinement (ISR) module and inter-category semantic transfer (IST) module. Specifically, the ISR module leverages the cross-modal knowledge of the VLP model to adaptively find a set of local discriminative regions that best represent the semantics of the target category. The IST module adaptively discovers a set of most correlated categories for a target category by utilizing the commonsense capabilities of LLMs to construct a category-adaptive correlation graph and transfers semantic knowledge from the correlated seen categories to unseen ones. Extensive experiments on OV-MLR benchmarks clearly demonstrate that the proposed C$^2$SRT framework outperforms current state-of-the-art algorithms.

Paperid: 3713, https://arxiv.org/pdf/2412.05311.pdf

Abstract:
In the advanced technology nodes, the integrated design rule checker (DRC) is often utilized in place and route tools for fast optimization loops for power-performance-area. Implementing integrated DRC checkers to meet the standard of commercial DRC tools demands extensive human expertise to interpret foundry specifications, analyze layouts, and debug code iteratively. However, this labor-intensive process, requiring to be repeated by every update of technology nodes, prolongs the turnaround time of designing circuits. In this paper, we present DRC-Coder, a multi-agent framework with vision capabilities for automated DRC code generation. By incorporating vision language models and large language models (LLM), DRC-Coder can effectively process textual, visual, and layout information to perform rule interpretation and coding by two specialized LLMs. We also design an auto-evaluation function for LLMs to enable DRC code debugging. Experimental results show that targeting on a sub-3nm technology node for a state-of-the-art standard cell layout tool, DRC-Coder achieves perfect F1 score 1.000 in generating DRC codes for meeting the standard of a commercial DRC tool, highly outperforming standard prompting techniques (F1=0.631). DRC-Coder can generate code for each design rule within four minutes on average, which significantly accelerates technology advancement and reduces engineering costs.

Paperid: 3714, https://arxiv.org/pdf/2412.01487.pdf

Abstract:
Large Vision Language Models (LVLMs) have demonstrated remarkable reasoning capabilities over textual and visual inputs. However, these models remain prone to generating misinformation. Identifying and mitigating ungrounded responses is crucial for developing trustworthy AI. Traditional explainability methods such as gradient-based relevancy maps, offer insight into the decision process of models, but are often computationally expensive and unsuitable for real-time output validation. In this work, we introduce FastRM, an efficient method for predicting explainable Relevancy Maps of LVLMs. Furthermore, FastRM provides both quantitative and qualitative assessment of model confidence. Experimental results demonstrate that FastRM achieves a 99.8% reduction in computation time and a 44.4% reduction in memory footprint compared to traditional relevancy map generation. FastRM allows explainable AI to be more practical and scalable, thereby promoting its deployment in real-world applications and enabling users to more effectively evaluate the reliability of model outputs.

Paperid: 3715, https://arxiv.org/pdf/2412.00846.pdf

Abstract:
This paper presents GPSM4K, a comprehensive geometry multimodal dataset tailored to augment the problem-solving capabilities of Large Vision Language Models (LVLMs). GPSM4K encompasses 2157 multimodal question-answer pairs manually extracted from mathematics textbooks spanning grades 7-12 and is further augmented to 5340 problems, consisting of both numerical and theorem-proving questions. In contrast to PGPS9k, Geometry3K, and Geo170K which feature only objective-type questions, GPSM4K offers detailed step-by-step solutions in a consistent format, facilitating a comprehensive evaluation of problem-solving approaches. This dataset serves as an excellent benchmark for assessing the geometric reasoning capabilities of LVLMs. Evaluation of our test set shows that there is scope for improvement needed in open-source language models in geometry problem-solving. Finetuning on our training set increases the geometry problem-solving capabilities of models. Further, We also evaluate the effectiveness of techniques such as image captioning and Retrieval Augmentation generation (RAG) on model performance. We leveraged LLM to automate the task of final answer evaluation by providing ground truth and predicted solutions. This research will help to assess and improve the geometric reasoning capabilities of LVLMs.

Paperid: 3716, https://arxiv.org/pdf/2412.00364.pdf

Abstract:
It is widely agreed that open-vocabulary-based approaches outperform classical closed-set training solutions for recognizing unseen objects in images for semantic segmentation. Existing open-vocabulary approaches leverage vision-language models, such as CLIP, to align visual features with rich semantic features acquired through pre-training on large-scale vision-language datasets. However, the text prompts employed in these methods are short phrases based on fixed templates, failing to capture comprehensive object attributes. Moreover, while the CLIP model excels at exploiting image-level features, it is less effective at pixel-level representation, which is crucial for semantic segmentation tasks. In this work, we propose to alleviate the above-mentioned issues by leveraging multiple large-scale models to enhance the alignment between fine-grained visual features and enriched linguistic features. Specifically, our method employs large language models (LLMs) to generate enriched language prompts with diverse visual attributes for each category, including color, shape/size, and texture/material. Additionally, for enhanced visual feature extraction, the SAM model is adopted as a supplement to the CLIP visual encoder through a proposed learnable weighted fusion strategy. Built upon these techniques, our method, termed LMSeg, achieves state-of-the-art performance across all major open-vocabulary segmentation benchmarks. The code will be made available soon.

Paperid: 3717, https://arxiv.org/pdf/2411.19650.pdf

Abstract:
The advancement of large Vision-Language-Action (VLA) models has significantly improved robotic manipulation in terms of language-guided task execution and generalization to unseen scenarios. While existing VLAs adapted from pretrained large Vision-Language-Models (VLM) have demonstrated promising generalizability, their task performance is still unsatisfactory as indicated by the low tasks success rates in different environments. In this paper, we present a new advanced VLA architecture derived from VLM. Unlike previous works that directly repurpose VLM for action prediction by simple action quantization, we propose a omponentized VLA architecture that has a specialized action module conditioned on VLM output. We systematically study the design of the action module and demonstrates the strong performance enhancement with diffusion action transformers for action sequence modeling, as well as their favorable scaling behaviors. We also conduct comprehensive experiments and ablation studies to evaluate the efficacy of our models with varied designs. The evaluation on 5 robot embodiments in simulation and real work shows that our model not only significantly surpasses existing VLAs in task performance and but also exhibits remarkable adaptation to new robots and generalization to unseen objects and backgrounds. It exceeds the average success rates of OpenVLA which has similar model size (7B) with ours by over 35% in simulated evaluation and 55% in real robot experiments. It also outperforms the large RT-2-X model (55B) by 18% absolute success rates in simulation. Code and models can be found on our project page (https://cogact.github.io/).

Paperid: 3718, https://arxiv.org/pdf/2411.16824.pdf

Abstract:
Does seeing always mean knowing? Large Vision-Language Models (LVLMs) integrate separately pre-trained vision and language components, often using CLIP-ViT as vision backbone. However, these models frequently encounter a core issue of "cognitive misalignment" between the vision encoder (VE) and the large language model (LLM). Specifically, the VE's representation of visual information may not fully align with LLM's cognitive framework, leading to a mismatch where visual features exceed the language model's interpretive range. To address this, we investigate how variations in VE representations influence LVLM comprehension, especially when the LLM faces VE-Unknown data-images whose ambiguous visual representations challenge the VE's interpretive precision. Accordingly, we construct a multi-granularity landmark dataset and systematically examine the impact of VE-Known and VE-Unknown data on interpretive abilities. Our results show that VE-Unknown data limits LVLM's capacity for accurate understanding, while VE-Known data, rich in distinctive features, helps reduce cognitive misalignment. Building on these insights, we propose Entity-Enhanced Cognitive Alignment (EECA), a method that employs multi-granularity supervision to generate visually enriched, well-aligned tokens that not only integrate within the LLM's embedding space but also align with the LLM's cognitive framework. This alignment markedly enhances LVLM performance in landmark recognition. Our findings underscore the challenges posed by VE-Unknown data and highlight the essential role of cognitive alignment in advancing multimodal systems.

Paperid: 3719, https://arxiv.org/pdf/2411.04291.pdf

Abstract:
Vision-language models (VLMs) have improved significantly in their capabilities, but their complex architecture makes their safety alignment challenging. In this paper, we reveal an uneven distribution of harmful information across the intermediate layers of the image encoder and show that skipping a certain set of layers and exiting early can increase the chance of the VLM generating harmful responses. We call it as "Image enCoder Early-exiT" based vulnerability (ICET). Our experiments across three VLMs: LLaVA-1.5, LLaVA-NeXT, and Llama 3.2, show that performing early exits from the image encoder significantly increases the likelihood of generating harmful outputs. To tackle this, we propose a simple yet effective modification of the Clipped-Proximal Policy Optimization (Clip-PPO) algorithm for performing layer-wise multi-modal RLHF for VLMs. We term this as Layer-Wise PPO (L-PPO). We evaluate our L-PPO algorithm across three multimodal datasets and show that it consistently reduces the harmfulness caused by early exits.

Paperid: 3720, https://arxiv.org/pdf/2411.04097.pdf

Abstract:
Fine-tuned vision-language models (VLMs) often capture spurious correlations between image features and textual attributes, resulting in degraded zero-shot performance at test time. Existing approaches for addressing spurious correlations (i) primarily operate at the global image-level rather than intervening directly on fine-grained image features and (ii) are predominantly designed for unimodal settings. In this work, we present RaVL, which takes a fine-grained perspective on VLM robustness by discovering and mitigating spurious correlations using local image features rather than operating at the global image level. Given a fine-tuned VLM, RaVL first discovers spurious correlations by leveraging a region-level clustering approach to identify precise image features contributing to zero-shot classification errors. Then, RaVL mitigates the identified spurious correlation with a novel region-aware loss function that enables the VLM to focus on relevant regions and ignore spurious relationships during fine-tuning. We evaluate RaVL on 654 VLMs with various model architectures, data domains, and learned spurious correlations. Our results show that RaVL accurately discovers (191% improvement over the closest baseline) and mitigates (8.2% improvement on worst-group image classification accuracy) spurious correlations. Qualitative evaluations on general-domain and medical-domain VLMs confirm our findings.

Paperid: 3721, https://arxiv.org/pdf/2411.03707.pdf

Abstract:
Geometric Dimensioning and Tolerancing (GD&T) plays a critical role in manufacturing by defining acceptable variations in part features to ensure component quality and functionality. However, extracting GD&T information from 2D engineering drawings is a time-consuming and labor-intensive task, often relying on manual efforts or semi-automated tools. To address these challenges, this study proposes an automated and computationally efficient GD&T extraction method by fine-tuning Florence-2, an open-source vision-language model (VLM). The model is trained on a dataset of 400 drawings with ground truth annotations provided by domain experts. For comparison, two state-of-the-art closed-source VLMs, GPT-4o and Claude-3.5-Sonnet, are evaluated on the same dataset. All models are assessed using precision, recall, F1-score, and hallucination metrics. Due to the computational cost and impracticality of fine-tuning large closed-source VLMs for domain-specific tasks, GPT-4o and Claude-3.5-Sonnet are evaluated in a zero-shot setting. In contrast, Florence-2, a smaller model with 0.23 billion parameters, is optimized through full-parameter fine-tuning across three distinct experiments, each utilizing datasets augmented to different levels. The results show that Florence-2 achieves a 29.95% increase in precision, a 37.75% increase in recall, a 52.40% improvement in F1-score, and a 43.15% reduction in hallucination rate compared to the best-performing closed-source model. These findings highlight the effectiveness of fine-tuning smaller, open-source VLMs like Florence-2, offering a practical and efficient solution for automated GD&T extraction to support downstream manufacturing tasks.

Paperid: 3722, https://arxiv.org/pdf/2411.02810.pdf

Abstract:
Automatic feature recognition (AFR) is essential for transforming design knowledge into actionable manufacturing information. Traditional AFR methods, which rely on predefined geometric rules and large datasets, are often time-consuming and lack generalizability across various manufacturing features. To address these challenges, this study investigates vision-language models (VLMs) for automating the recognition of a wide range of manufacturing features in CAD designs without the need for extensive training datasets or predefined rules. Instead, prompt engineering techniques, such as multi-view query images, few-shot learning, sequential reasoning, and chain-of-thought, are applied to enable recognition. The approach is evaluated on a newly developed CAD dataset containing designs of varying complexity relevant to machining, additive manufacturing, sheet metal forming, molding, and casting. Five VLMs, including three closed-source models (GPT-4o, Claude-3.5-Sonnet, and Claude-3.0-Opus) and two open-source models (LLava and MiniCPM), are evaluated on this dataset with ground truth features labelled by experts. Key metrics include feature quantity accuracy, feature name matching accuracy, hallucination rate, and mean absolute error (MAE). Results show that Claude-3.5-Sonnet achieves the highest feature quantity accuracy (74%) and name-matching accuracy (75%) with the lowest MAE (3.2), while GPT-4o records the lowest hallucination rate (8%). In contrast, open-source models have higher hallucination rates (>30%) and lower accuracies (<40%). This study demonstrates the potential of VLMs to automate feature recognition in CAD designs within diverse manufacturing scenarios.

Paperid: 3723, https://arxiv.org/pdf/2410.19732.pdf

Abstract:
Large Vision-Language Models (LVLMs) excel in cross-model tasks but experience performance declines in long-context reasoning due to overreliance on textual information and reduced visual dependency. In this study, we empirically analyze LVLMs in long-context reasoning, revealing that increased context length leads to a higher dependence on language at the expense of visual dependency. To address this issue, we propose a novel training-free context pruning method that selectively removes less critical textual information. Our approach enhances visual dependency and reduces textual noise, thereby improving LVLM performance in long-context reasoning. We validate our method by constructing a long-context dataset, demonstrating its effectiveness across various LVLMs. Moreover, further analysis confirms the robustness of different token pruning strategies and preliminary explores scaling laws between pruning rates and context length.

Paperid: 3724, https://arxiv.org/pdf/2410.11619.pdf

Abstract:
Efficiently retrieving and synthesizing information from large-scale multimodal collections has become a critical challenge. However, existing video retrieval datasets suffer from scope limitations, primarily focusing on matching descriptive but vague queries with small collections of professionally edited, English-centric videos. To address this gap, we introduce $\textbf{MultiVENT 2.0}$, a large-scale, multilingual event-centric video retrieval benchmark featuring a collection of more than 218,000 news videos and 3,906 queries targeting specific world events. These queries specifically target information found in the visual content, audio, embedded text, and text metadata of the videos, requiring systems leverage all these sources to succeed at the task. Preliminary results show that state-of-the-art vision-language models struggle significantly with this task, and while alternative approaches show promise, they are still insufficient to adequately address this problem. These findings underscore the need for more robust multimodal retrieval systems, as effective video retrieval is a crucial step towards multimodal content understanding and generation.

Paperid: 3725, https://arxiv.org/pdf/2410.07571.pdf

Abstract:
Vision-Language adaptation (VL adaptation) transforms Large Language Models (LLMs) into Large Vision-Language Models (LVLMs) for multimodal tasks, but this process often compromises the inherent safety capabilities embedded in the original LLMs. Despite potential harmfulness due to weakened safety measures, in-depth analysis on the effects of VL adaptation on safety remains under-explored. This study examines how VL adaptation influences safety and evaluates the impact of safety fine-tuning methods. Our analysis reveals that safety degradation occurs during VL adaptation, even when the training data is safe. While safety tuning techniques like supervised fine-tuning with safety datasets or reinforcement learning from human feedback mitigate some risks, they still lead to safety degradation and a reduction in helpfulness due to over-rejection issues. Further analysis of internal model weights suggests that VL adaptation may impact certain safety-related layers, potentially lowering overall safety levels. Additionally, our findings demonstrate that the objectives of VL adaptation and safety tuning are divergent, which often results in their simultaneous application being suboptimal. To address this, we suggest the weight merging approach as an optimal solution effectively reducing safety degradation while maintaining helpfulness. These insights help guide the development of more reliable and secure LVLMs for real-world applications.

Paperid: 3726, https://arxiv.org/pdf/2410.07025.pdf

Abstract:
Radiologists play a crucial role in translating medical images into actionable reports. However, the field faces staffing shortages and increasing workloads. While automated approaches using vision-language models (VLMs) show promise as assistants, they require exceptionally high accuracy. Most current VLMs in radiology rely solely on supervised fine-tuning. Meanwhile, additional preference fine-tuning in the post-training pipeline has become standard practice in the general domain. The challenge in radiology lies in the prohibitive cost of obtaining radiologist feedback at scale. To address this challenge, we propose an automated pipeline for preference feedback, focusing on chest X-ray radiology report generation (RRG). Specifically, our method leverages publicly available datasets containing pairs of images and radiologist-written reference reports with reference-based metrics, or Judges, eliminating the need for additional radiologist feedback. We investigate reward overoptimization via length exploitation in this setting and introduce a length-controlled version of the GREEN score. Our best-performing setup achieves state-of-the-art CheXbert scores on the MIMIC-CXR dataset for the RRG task while on average maintaining robust performance across six additional image perception and reasoning tasks.

Paperid: 3727, https://arxiv.org/pdf/2410.05346.pdf

Abstract:
Due to their multimodal capabilities, Vision-Language Models (VLMs) have found numerous impactful applications in real-world scenarios. However, recent studies have revealed that VLMs are vulnerable to image-based adversarial attacks. Traditional targeted adversarial attacks require specific targets and labels, limiting their real-world impact.We present AnyAttack, a self-supervised framework that transcends the limitations of conventional attacks through a novel foundation model approach. By pre-training on the massive LAION-400M dataset without label supervision, AnyAttack achieves unprecedented flexibility - enabling any image to be transformed into an attack vector targeting any desired output across different VLMs.This approach fundamentally changes the threat landscape, making adversarial capabilities accessible at an unprecedented scale. Our extensive validation across five open-source VLMs (CLIP, BLIP, BLIP2, InstructBLIP, and MiniGPT-4) demonstrates AnyAttack's effectiveness across diverse multimodal tasks. Most concerning, AnyAttack seamlessly transfers to commercial systems including Google Gemini, Claude Sonnet, Microsoft Copilot and OpenAI GPT, revealing a systemic vulnerability requiring immediate attention.

Paperid: 3728, https://arxiv.org/pdf/2410.01264.pdf

Abstract:
The emergence of Vision-Language Models (VLMs) represents a significant advancement in integrating computer vision with Large Language Models (LLMs) to generate detailed text descriptions from visual inputs. Despite their growing importance, the security of VLMs, particularly against backdoor attacks, is under explored. Moreover, prior works often assume attackers have access to the original training data, which is often unrealistic. In this paper, we address a more practical and challenging scenario where attackers must rely solely on Out-Of-Distribution (OOD) data. We introduce VLOOD (Backdooring Vision-Language Models with Out-of-Distribution Data), a novel approach with two key contributions: (1) demonstrating backdoor attacks on VLMs in complex image-to-text tasks while minimizing degradation of the original semantics under poisoned inputs, and (2) proposing innovative techniques for backdoor injection without requiring any access to the original training data. Our evaluation on image captioning and visual question answering (VQA) tasks confirms the effectiveness of VLOOD, revealing a critical security vulnerability in VLMs and laying the foundation for future research on securing multimodal models against sophisticated threats.

Paperid: 3729, https://arxiv.org/pdf/2409.20445.pdf

Abstract:
We present a novel autonomous robot navigation algorithm for outdoor environments that is capable of handling diverse terrain traversability conditions. Our approach, VLM-GroNav, uses vision-language models (VLMs) and integrates them with physical grounding that is used to assess intrinsic terrain properties such as deformability and slipperiness. We use proprioceptive-based sensing, which provides direct measurements of these physical properties, and enhances the overall semantic understanding of the terrains. Our formulation uses in-context learning to ground the VLM's semantic understanding with proprioceptive data to allow dynamic updates of traversability estimates based on the robot's real-time physical interactions with the environment. We use the updated traversability estimations to inform both the local and global planners for real-time trajectory replanning. We validate our method on a legged robot (Ghost Vision 60) and a wheeled robot (Clearpath Husky), in diverse real-world outdoor environments with different deformable and slippery terrains. In practice, we observe significant improvements over state-of-the-art methods by up to 50% increase in navigation success rate.

Paperid: 3730, https://arxiv.org/pdf/2409.19232.pdf

Abstract:
The emergence of Vision Language Models (VLMs) is a significant advancement in integrating computer vision with Large Language Models (LLMs) to produce detailed text descriptions based on visual inputs, yet it introduces new security vulnerabilities. Unlike prior work that centered on single modalities or classification tasks, this study introduces TrojVLM, the first exploration of backdoor attacks aimed at VLMs engaged in complex image-to-text generation. Specifically, TrojVLM inserts predetermined target text into output text when encountering poisoned images. Moreover, a novel semantic preserving loss is proposed to ensure the semantic integrity of the original image content. Our evaluation on image captioning and visual question answering (VQA) tasks confirms the effectiveness of TrojVLM in maintaining original semantic content while triggering specific target text outputs. This study not only uncovers a critical security risk in VLMs and image-to-text generation but also sets a foundation for future research on securing multimodal models against such sophisticated threats.

Paperid: 3731, https://arxiv.org/pdf/2409.17126.pdf

Abstract:
Generative AI systems have shown impressive capabilities in creating text, code, and images. Inspired by the rich history of research in industrial ''Design for Assembly'', we introduce a novel problem: Generative Design-for-Robot-Assembly (GDfRA). The task is to generate an assembly based on a natural language prompt (e.g., ''giraffe'') and an image of available physical components, such as 3D-printed blocks. The output is an assembly, a spatial arrangement of these components, and instructions for a robot to build this assembly. The output must 1) resemble the requested object and 2) be reliably assembled by a 6 DoF robot arm with a suction gripper. We then present Blox-Net, a GDfRA system that combines generative vision language models with well-established methods in computer vision, simulation, perturbation analysis, motion planning, and physical robot experimentation to solve a class of GDfRA problems with minimal human supervision. Blox-Net achieved a Top-1 accuracy of 63.5% in the ''recognizability'' of its designed assemblies (eg, resembling giraffe as judged by a VLM). These designs, after automated perturbation redesign, were reliably assembled by a robot, achieving near-perfect success across 10 consecutive assembly iterations with human intervention only during reset prior to assembly. Surprisingly, this entire design process from textual word (''giraffe'') to reliable physical assembly is performed with zero human intervention.

Paperid: 3732, https://arxiv.org/pdf/2601.15286.pdf

Abstract:
Text-to-image (T2I) models have achieved remarkable progress, yet they continue to struggle with complex prompts that require simultaneously handling multiple objects, relations, and attributes. Existing inference-time strategies, such as parallel sampling with verifiers or simply increasing denoising steps, can improve prompt alignment but remain inadequate for richly compositional settings where many constraints must be satisfied. Inspired by the success of chain-of-thought reasoning in large language models, we propose an iterative test-time strategy in which a T2I model progressively refines its generations across multiple steps, guided by feedback from a vision-language model as the critic in the loop. Our approach is simple, requires no external tools or priors, and can be flexibly applied to a wide range of image generators and vision-language models. Empirically, we demonstrate consistent gains on image generation across benchmarks: a 16.9% improvement in all-correct rate on ConceptMix (k=7), a 13.8% improvement on T2I-CompBench (3D-Spatial category) and a 12.5% improvement on Visual Jenga scene decomposition compared to compute-matched parallel sampling. Beyond quantitative gains, iterative refinement produces more faithful generations by decomposing complex prompts into sequential corrections, with human evaluators preferring our method 58.7% of the time over 41.3% for the parallel baseline. Together, these findings highlight iterative self-correction as a broadly applicable principle for compositional image generation. Results and visualizations are available at https://iterative-img-gen.github.io/

Paperid: 3733, https://arxiv.org/pdf/2601.13132.pdf

Abstract:
We present GaussExplorer, a framework for embodied exploration and reasoning built on 3D Gaussian Splatting (3DGS). While prior approaches to language-embedded 3DGS have made meaningful progress in aligning simple text queries with Gaussian embeddings, they are generally optimized for relatively simple queries and struggle to interpret more complex, compositional language queries. Alternative studies based on object-centric RGB-D structured memories provide spatial grounding but are constrained by pre-fixed viewpoints. To address these issues, GaussExplorer introduces Vision-Language Models (VLMs) on top of 3DGS to enable question-driven exploration and reasoning within 3D scenes. We first identify pre-captured images that are most correlated with the query question, and subsequently adjust them into novel viewpoints to more accurately capture visual information for better reasoning by VLMs. Experiments show that ours outperforms existing methods on several benchmarks, demonstrating the effectiveness of integrating VLM-based reasoning with 3DGS for embodied tasks.

Paperid: 3734, https://arxiv.org/pdf/2601.05611.pdf

Abstract:
End-to-end autonomous driving models trained on largescale datasets perform well in common scenarios but struggle with rare, long-tail situations due to limited scenario diversity. Recent Vision-Language-Action (VLA) models leverage broad knowledge from pre-trained visionlanguage models to address this limitation, yet face critical challenges: (1) numerical imprecision in trajectory prediction due to discrete tokenization, (2) heavy reliance on language annotations that introduce linguistic bias and annotation burden, and (3) computational inefficiency from multi-step chain-of-thought reasoning hinders real-time deployment. We propose LatentVLA, a novel framework that employs self-supervised latent action prediction to train VLA models without language annotations, eliminating linguistic bias while learning rich driving representations from unlabeled trajectory data. Through knowledge distillation, LatentVLA transfers the generalization capabilities of VLA models to efficient vision-based networks, achieving both robust performance and real-time efficiency. LatentVLA establishes a new state-of-the-art on the NAVSIM benchmark with a PDMS score of 92.4 and demonstrates strong zeroshot generalization on the nuScenes benchmark.

Paperid: 3735, https://arxiv.org/pdf/2601.05201.pdf

Abstract:
Large vision-language models (VLMs) are highly capable, yet often hallucinate by favoring textual prompts over visual evidence. We study this failure mode in a controlled object-counting setting, where the prompt overstates the number of objects in the image (e.g., asking a model to describe four waterlilies when only three are present). At low object counts, models often correct the overestimation, but as the number of objects increases, they increasingly conform to the prompt regardless of the discrepancy. Through mechanistic analysis of three VLMs, we identify a small set of attention heads whose ablation substantially reduces prompt-induced hallucinations (PIH) by at least 40% without additional training. Across models, PIH-heads mediate prompt copying in model-specific ways. We characterize these differences and show that PIH ablation increases correction toward visual evidence. Our findings offer insights into the internal mechanisms driving prompt-induced hallucinations, revealing model-specific differences in how these behaviors are implemented.

Paperid: 3736, https://arxiv.org/pdf/2601.03741.pdf

Abstract:
Existing text-guided image editing methods primarily rely on end-to-end pixel-level inpainting paradigm. Despite its success in simple scenarios, this paradigm still significantly struggles with compositional editing tasks that require precise local control and complex multi-object spatial reasoning. This paradigm is severely limited by 1) the implicit coupling of planning and execution, 2) the lack of object-level control granularity, and 3) the reliance on unstructured, pixel-centric modeling. To address these limitations, we propose I2E, a novel "Decompose-then-Action" paradigm that revisits image editing as an actionable interaction process within a structured environment. I2E utilizes a Decomposer to transform unstructured images into discrete, manipulable object layers and then introduces a physics-aware Vision-Language-Action Agent to parse complex instructions into a series of atomic actions via Chain-of-Thought reasoning. Further, we also construct I2E-Bench, a benchmark designed for multi-instance spatial reasoning and high-precision editing. Experimental results on I2E-Bench and multiple public benchmarks demonstrate that I2E significantly outperforms state-of-the-art methods in handling complex compositional instructions, maintaining physical plausibility, and ensuring multi-turn editing stability.

Paperid: 3737, https://arxiv.org/pdf/2512.21815.pdf

Abstract:
Vision-language models (VLMs) achieve remarkable performance but remain vulnerable to adversarial attacks. Entropy, a measure of model uncertainty, is strongly correlated with the reliability of VLM. Prior entropy-based attacks maximize uncertainty at all decoding steps, implicitly assuming that every token contributes equally to generation instability. We show instead that a small fraction (about 20%) of high-entropy tokens, i.e., critical decision points in autoregressive generation, disproportionately governs output trajectories. By concentrating adversarial perturbations on these positions, we achieve semantic degradation comparable to global methods while using substantially smaller budgets. More importantly, across multiple representative VLMs, such selective attacks convert 35-49% of benign outputs into harmful ones, exposing a more critical safety risk. Remarkably, these vulnerable high-entropy forks recur across architecturally diverse VLMs, enabling feasible transferability (17-26% harmful rates on unseen targets). Motivated by these findings, we propose Entropy-bank Guided Adversarial attacks (EGA), which achieves competitive attack success rates (93-95%) alongside high harmful conversion, thereby revealing new weaknesses in current VLM safety mechanisms.

Paperid: 3738, https://arxiv.org/pdf/2512.09056.pdf

Abstract:
Object pose estimation is a fundamental task in computer vision and robotics, yet most methods require extensive, dataset-specific training. Concurrently, large-scale vision language models show remarkable zero-shot capabilities. In this work, we bridge these two worlds by introducing ConceptPose, a framework for object pose estimation that is both training-free and model-free. ConceptPose leverages a vision-language-model (VLM) to create open-vocabulary 3D concept maps, where each point is tagged with a concept vector derived from saliency maps. By establishing robust 3D-3D correspondences across concept maps, our approach allows precise estimation of 6DoF relative pose. Without any object or dataset-specific training, our approach achieves state-of-the-art results on common zero shot relative pose estimation benchmarks, significantly outperforming existing methods by over 62% in ADD(-S) score, including those that utilize extensive dataset-specific training.

Paperid: 3739, https://arxiv.org/pdf/2512.02700.pdf

Abstract:
Vision-language models (VLMs) excel at image understanding tasks, but the large number of visual tokens imposes significant computational costs, hindering deployment on mobile devices. Many pruning methods rely solely on token importance and thus overlook inter-token redundancy, retaining numerous duplicated tokens and wasting capacity. Although some redundancy-aware approaches have been proposed, they often ignore the spatial relationships among visual tokens. This can lead to overly sparse selections of retained tokens that fail to adequately cover the regions of target objects. To address these limitations, we propose VLM-Pruner, a training-free token pruning algorithm that explicitly balances redundancy and spatial sparsity. We introduce a centrifugal token pruning paradigm that enables near-to-far selection while prioritizing the preservation of fine-grained object details. Moreover, we design a Buffering for Spatial Sparsity (BSS) criterion that defers the selection of spatially distant tokens. We further adopt a parallel greedy strategy to conduct token selection efficiently. To mitigate information loss from pruning, we selectively fuse salient information from the discarded tokens into the retained ones. Comprehensive comparisons demonstrate that VLM-Pruner consistently outperforms strong baselines across five VLMs with an 88.9\% pruning rate, while delivering an end-to-end inference speedup.

Paperid: 3740, https://arxiv.org/pdf/2511.22659.pdf

Abstract:
Vision Language Models (VLMs) exhibit a fundamental semantic-to-geometric gap in spatial reasoning: they excel at qualitative semantic inference but their reasoning operates within a lossy semantic space, misaligned with high-fidelity geometry. Current paradigms fail to bridge this gap. Training-based methods suffer from an ``oracle paradox,'' learning flawed spatial logic from imperfect oracles. Tool-integrated methods constrain the final computation but critically leave the VLM's planning process unconstrained, resulting in geometrically flawed plans. In this work, we propose Geometrically-Constrained Agent (GCA), a training-free agentic paradigm that resolves this gap by introducing a formal task constraint. Specifically, we strategically decouples the VLM's role into two stages. First, acting as a semantic analyst, the VLM translates the user's ambiguous query into the formal, verifiable task constraint, which defines the reference frame and objective. Second, acting as a task solver, the VLM generates and executes tool calls strictly within the deterministic bounds defined by the constraint. This geometrically-constrained reasoning strategy successfully resolve the semantic-to-geometric gap, yielding a robust and verifiable reasoning pathway for spatial reasoning. Comprehensive experiments demonstrate that GCA achieves SOTA performance on multiple spatial reasoning benchmarks, surpassing existing training-based and tool-integrated methods by ~27%. Please see our homepage at https://gca-spatial-reasoning.github.io.

Paperid: 3741, https://arxiv.org/pdf/2511.19912.pdf

Abstract:
Vision-Language-Action (VLA) models have recently shown strong decision-making capabilities in autonomous driving. However, existing VLAs often struggle with achieving efficient inference and generalizing to novel autonomous vehicle configurations and driving scenarios. In this paper, we propose Reasoning-VLA, a general and fast action-generation VLA framework. The proposed model employs a set of learnable action queries, initialized via Gaussian sampling from ground-truth trajectories within the training corpus. These learnable queries interact with reasoning-enhanced vision-language features to generate continuous action trajectories in parallel. To promote robust generalization, we consolidate eight publicly available autonomous driving datasets into a standardized, Chain-of-Thought reasoning-based, and easy-to-use data format for model training. Leveraging both supervised learning and reinforcement learning fine-tuning, extensive empirical evaluations across multiple benchmarks demonstrate that Reasoning-VLA achieves state-of-the-art performance, superior generalization capability, and the excellent inference speed reported to date.

Paperid: 3742, https://arxiv.org/pdf/2511.19704.pdf

Abstract:
Open-vocabulary semantic segmentation (OVSS) underpins many vision and robotics tasks that require generalizable semantic understanding. Existing approaches either rely on limited segmentation training data, which hinders generalization, or apply zero-shot heuristics to vision-language models (e.g CLIP), while the most competitive approaches combine multiple models to improve performance at the cost of high computational and memory demands. In this work, we leverage an overlooked agglomerative vision foundation model, RADIO, to improve zero-shot OVSS along three key axes simultaneously: mIoU, latency, and parameter efficiency. We present the first comprehensive study of RADIO for zero-shot OVSS and enhance its performance through self-correlating recursive attention, self-correlating global aggregation, and computationally efficient mask refinement. Our approach, RADSeg, achieves 6-30% mIoU improvement in the base ViT class while being 3.95x faster and using 2.5x fewer parameters. Surprisingly, RADSeg-base (105M) outperforms previous combinations of huge vision models (850-1350M) in mIoU, achieving state-of-the-art accuracy with substantially lower computational and memory cost.

Paperid: 3743, https://arxiv.org/pdf/2511.18396.pdf

Abstract:
Aligning large-scale commercial models with user intent is crucial to preventing harmful outputs. Current methods rely on human supervision but become impractical as model complexity increases. When models surpass human knowledge, providing accurate feedback becomes challenging and inefficient. A novel solution proposed recently is using a weaker model to supervise a stronger model. This concept leverages the ability of weaker models to perform evaluations, thereby reducing the workload on human supervisors. Previous work has shown the effectiveness of weak-to-strong generalization in the context of language-only models. Extending this concept to vision-language models leverages these insights, adapting the proven benefits to a multi-modal context. In our study, we explore weak-to-strong generalization for CLIP-based classification. We propose a method, class prototype learning (CPL), which aims to enhance the classification capabilities of the CLIP model, by learning more representative prototypes for each category. Our findings indicate that, despite using a simple loss function under weak supervision, CPL yields robust improvements in targeted scenarios, particularly when pretraining is limited. Extensive experiments demonstrate that our approach is effective under these settings, achieving a 3.67% improvement over strong baseline methods.

Paperid: 3744, https://arxiv.org/pdf/2511.14271.pdf

Abstract:
Text-to-3D generation has advanced rapidly, yet state-of-the-art models, encompassing both optimization-based and feed-forward architectures, still face two fundamental limitations. First, they struggle with coarse semantic alignment, often failing to capture fine-grained prompt details. Second, they lack robust 3D spatial understanding, leading to geometric inconsistencies and catastrophic failures in part assembly and spatial relationships. To address these challenges, we propose VLM3D, a general framework that repurposes large vision-language models (VLMs) as powerful, differentiable semantic and spatial critics. Our core contribution is a dual-query critic signal derived from the VLM's Yes or No log-odds, which assesses both semantic fidelity and geometric coherence. We demonstrate the generality of this guidance signal across two distinct paradigms: (1) As a reward objective for optimization-based pipelines, VLM3D significantly outperforms existing methods on standard benchmarks. (2) As a test-time guidance module for feed-forward pipelines, it actively steers the iterative sampling process of SOTA native 3D models to correct severe spatial errors. VLM3D establishes a principled and generalizable path to inject the VLM's rich, language-grounded understanding of both semantics and space into diverse 3D generative pipelines.

Paperid: 3745, https://arxiv.org/pdf/2511.10923.pdf

Abstract:
Out-of-distribution (OOD) detection is committed to delineating the classification boundaries between in-distribution (ID) and OOD images. Recent advances in vision-language models (VLMs) have demonstrated remarkable OOD detection performance by integrating both visual and textual modalities. In this context, negative prompts are introduced to emphasize the dissimilarity between image features and prompt content. However, these prompts often include a broad range of non-ID features, which may result in suboptimal outcomes due to the capture of overlapping or misleading information. To address this issue, we propose Positive and Negative Prompt Supervision, which encourages negative prompts to capture inter-class features and transfers this semantic knowledge to the visual modality to enhance OOD detection performance. Our method begins with class-specific positive and negative prompts initialized by large language models (LLMs). These prompts are subsequently optimized, with positive prompts focusing on features within each class, while negative prompts highlight features around category boundaries. Additionally, a graph-based architecture is employed to aggregate semantic-aware supervision from the optimized prompt representations and propagate it to the visual branch, thereby enhancing the performance of the energy-based OOD detector. Extensive experiments on two benchmarks, CIFAR-100 and ImageNet-1K, across eight OOD datasets and five different LLMs, demonstrate that our method outperforms state-of-the-art baselines.

Paperid: 3746, https://arxiv.org/pdf/2511.09771.pdf

Abstract:
Accurate 6D pose estimation and tracking are fundamental capabilities for physical AI systems such as robots. However, existing approaches typically rely on a manually annotated segmentation mask of the target in the first frame, which is labor-intensive and leads to reduced performance when faced with occlusions or rapid movement. To address these limi- tations, we propose STORM (Segment, Track, and Object Re-localization from a single 3D Model), an open-source robust real-time 6D pose estimation system that requires no manual annotation. STORM employs a novel three-stage pipeline combining vision-language understanding with self-supervised feature matching: contextual object descriptions guide localization, self-cross-attention mechanisms identify candidate regions, and a segmentation model produces precise masks for accurate pose estimation. Another key innovation is our automatic re-registration mechanism that detects tracking failures through feature similarity monitoring and recovers from severe occlusions or rapid motion. STORM achieves state-of-the-art accuracy on challenging industrial datasets featuring multi-object occlusions, high-speed motion, and varying illumination, while operating at real-time speeds without additional training. This annotation-free approach significantly reduces deployment overhead, providing a practical solution for modern applications, such as flexible manufacturing and intelligent quality control.

Paperid: 3747, https://arxiv.org/pdf/2510.21814.pdf

Abstract:
Free-form gesture understanding is highly appealing for human-computer interaction, as it liberates users from the constraints of predefined gesture categories. However, the sole existing solution GestureGPT suffers from limited recognition accuracy and slow response times. In this paper, we propose Gestura, an end-to-end system for free-form gesture understanding. Gestura harnesses a pre-trained Large Vision-Language Model (LVLM) to align the highly dynamic and diverse patterns of free-form gestures with high-level semantic concepts. To better capture subtle hand movements across different styles, we introduce a Landmark Processing Module that compensate for LVLMs' lack of fine-grained domain knowledge by embedding anatomical hand priors. Further, a Chain-of-Thought (CoT) reasoning strategy enables step-by-step semantic inference, transforming shallow knowledge into deep semantic understanding and significantly enhancing the model's ability to interpret ambiguous or unconventional gestures. Together, these components allow Gestura to achieve robust and adaptable free-form gesture comprehension. Additionally, we have developed the first open-source dataset for free-form gesture intention reasoning and understanding with over 300,000 annotated QA pairs.

Paperid: 3748, https://arxiv.org/pdf/2510.20229.pdf

Abstract:
Large Vision-Language Models (LVLMs) have made significant progress in recent years but are also prone to hallucination issues. They exhibit more hallucinations in longer, free-form responses, often attributed to accumulated uncertainties. In this paper, we ask: Does increased hallucination result solely from length-induced errors, or is there a deeper underlying mechanism? After a series of preliminary experiments and findings, we suggest that the risk of hallucinations is not caused by length itself but by the increased reliance on context for coherence and completeness in longer responses. Building on these insights, we propose a novel "induce-detect-suppress" framework that actively induces hallucinations through deliberately designed contexts, leverages induced instances for early detection of high-risk cases, and ultimately suppresses potential object-level hallucinations during actual decoding. Our approach achieves consistent, significant improvements across all benchmarks, demonstrating its efficacy. The strong detection and improved hallucination mitigation not only validate our framework but, more importantly, re-validate our hypothesis on context. Rather than solely pursuing performance gains, this study aims to provide new insights and serves as a first step toward a deeper exploration of hallucinations in LVLMs' longer responses.

Paperid: 3749, https://arxiv.org/pdf/2510.15042.pdf

Abstract:
Vision-language pre-training, i.e., aligning images with paired text, is a powerful paradigm to create encoders that can be directly used for tasks such as classification and retrieval, and for downstream tasks such as segmentation and report generation. In the 3D medical image domain, these capabilities allow vision-language encoders (VLEs) to support radiologists by retrieving patients with similar abnormalities or predicting likelihoods of abnormality. While the methodology holds promise, data availability limits the capabilities of current 3D VLEs. In this paper, we alleviate the lack of data by injecting additional inductive biases: introducing a report generation objective and pairing vision-language pre-training with vision-only pre-training. This allows us to leverage both image-only and paired image-text 3D datasets, increasing the total amount of data to which our model is exposed. Through these additional inductive biases, paired with best practices of the 3D medical imaging domain, we develop the Comprehensive Language-image Pre-training (COLIPRI) encoder family. Our COLIPRI encoders achieve state-of-the-art performance in report generation, classification probing, and zero-shot classification, and remain competitive for semantic segmentation.

Paperid: 3750, https://arxiv.org/pdf/2510.07975.pdf

Abstract:
Enabling robots to perform precise and generalized manipulation in unstructured environments remains a fundamental challenge in embodied AI. While Vision-Language Models (VLMs) have demonstrated remarkable capabilities in semantic reasoning and task planning, a significant gap persists between their high-level understanding and the precise physical execution required for real-world manipulation. To bridge this "semantic-to-physical" gap, we introduce GRACE, a novel framework that grounds VLM-based reasoning through executable analytic concepts (EAC)-mathematically defined blueprints that encode object affordances, geometric constraints, and semantics of manipulation. Our approach integrates a structured policy scaffolding pipeline that turn natural language instructions and visual information into an instantiated EAC, from which we derive grasp poses, force directions and plan physically feasible motion trajectory for robot execution. GRACE thus provides a unified and interpretable interface between high-level instruction understanding and low-level robot control, effectively enabling precise and generalizable manipulation through semantic-physical grounding. Extensive experiments demonstrate that GRACE achieves strong zero-shot generalization across a variety of articulated objects in both simulated and real-world environments, without requiring task-specific training.

Paperid: 3751, https://arxiv.org/pdf/2510.02155.pdf

Abstract:
Prompting has emerged as a practical way to adapt frozen vision-language models (VLMs) for video anomaly detection (VAD). Yet, existing prompts are often overly abstract, overlooking the fine-grained human-object interactions or action semantics that define complex anomalies in surveillance videos. We propose ASK-Hint, a structured prompting framework that leverages action-centric knowledge to elicit more accurate and interpretable reasoning from frozen VLMs. Our approach organizes prompts into semantically coherent groups (e.g. violence, property crimes, public safety) and formulates fine-grained guiding questions that align model predictions with discriminative visual cues. Extensive experiments on UCF-Crime and XD-Violence show that ASK-Hint consistently improves AUC over prior baselines, achieving state-of-the-art performance compared to both fine-tuned and training-free methods. Beyond accuracy, our framework provides interpretable reasoning traces towards anomaly and demonstrates strong generalization across datasets and VLM backbones. These results highlight the critical role of prompt granularity and establish ASK-Hint as a new training-free and generalizable solution for explainable video anomaly detection.

Paperid: 3752, https://arxiv.org/pdf/2510.00681.pdf

Abstract:
Event cameras offer advantages in object detection tasks due to high-speed response, low latency, and robustness to motion blur. However, event cameras lack texture and color information, making open-vocabulary detection particularly challenging. Current event-based detection methods are typically trained on predefined categories, limiting their ability to generalize to novel objects, where encountering previously unseen objects is common. Vision-language models (VLMs) have enabled open-vocabulary object detection in RGB images. However, the modality gap between images and event streams makes it ineffective to directly transfer CLIP to event data, as CLIP was not designed for event streams. To bridge this gap, we propose an event-image knowledge distillation framework that leverages CLIP's semantic understanding to achieve open-vocabulary object detection on event data. Instead of training CLIP directly on event streams, we use image frames as inputs to a teacher model, guiding the event-based student model to learn CLIP's rich visual representations. Through spatial attention-based distillation, the student network learns meaningful visual features directly from raw event inputs while inheriting CLIP's broad visual knowledge. Furthermore, to prevent information loss due to event data segmentation, we design a hybrid spiking neural network (SNN) and convolutional neural network (CNN) framework. Unlike fixed-group event segmentation methods, which often discard crucial temporal information, our SNN adaptively determines the optimal event segmentation moments, ensuring that key temporal features are extracted. The extracted event features are then processed by CNNs for object detection.

Paperid: 3753, https://arxiv.org/pdf/2509.25711.pdf

Abstract:
Medical decision-making requires integrating diverse medical information, from imaging to clinical narratives. These medical modalities are often acquired in a many-to-many manner. However, current medical vision-language pretraining models (Med-VLPMs) fail to directly account for this many-to-many mapping in their model training and embeddings. To address this, we present Probabilistic Modality-Enhanced Diagnosis (ProbMED), a multimodal Med-VLPM that employs probabilistic contrastive learning to model distributions over embeddings rather than deterministic estimates. ProbMED aligns four distinct modalities -- chest X-rays, electrocardiograms, echocardiograms, and clinical text -- into a unified probabilistic embedding space. We use InfoNCE loss with Hellinger distance to integrate inter-modality distributions. We introduce a probabilistic synthetic sampling loss that captures modality-specific mean and variance to improve intra-modality binding. Extensive experiments across 13 medical datasets demonstrate that our model outperforms current Med-VLPMs in cross-modality retrieval, zero-shot, and few-shot classification. We also demonstrate the robust integration of multiple modalities for prognostication, showing improved intra- and inter-medical modality binding.

Paperid: 3754, https://arxiv.org/pdf/2509.25528.pdf

Abstract:
Referential grounding in outdoor driving scenes is challenging due to large scene variability, many visually similar objects, and dynamic elements that complicate resolving natural-language references (e.g., "the black car on the right"). We propose LLM-RG, a hybrid pipeline that combines off-the-shelf vision-language models for fine-grained attribute extraction with large language models for symbolic reasoning. LLM-RG processes an image and a free-form referring expression by using an LLM to extract relevant object types and attributes, detecting candidate regions, generating rich visual descriptors with a VLM, and then combining these descriptors with spatial metadata into natural-language prompts that are input to an LLM for chain-of-thought reasoning to identify the referent's bounding box. Evaluated on the Talk2Car benchmark, LLM-RG yields substantial gains over both LLM and VLM-based baselines. Additionally, our ablations show that adding 3D spatial cues further improves grounding. Our results demonstrate the complementary strengths of VLMs and LLMs, applied in a zero-shot manner, for robust outdoor referential grounding.

Paperid: 3755, https://arxiv.org/pdf/2509.23563.pdf

Abstract:
Aerial outdoor semantic navigation requires robots to explore large, unstructured environments to locate target objects. Recent advances in semantic navigation have demonstrated open-set object-goal navigation in indoor settings, but these methods remain limited by constrained spatial ranges and structured layouts, making them unsuitable for long-range outdoor search. While outdoor semantic navigation approaches exist, they either rely on reactive policies based on current observations, which tend to produce short-sighted behaviors, or precompute scene graphs offline for navigation, limiting adaptability to online deployment. We present RAVEN, a 3D memory-based, behavior tree framework for aerial semantic navigation in unstructured outdoor environments. It (1) uses a spatially consistent semantic voxel-ray map as persistent memory, enabling long-horizon planning and avoiding purely reactive behaviors, (2) combines short-range voxel search and long-range ray search to scale to large environments, (3) leverages a large vision-language model to suggest auxiliary cues, mitigating sparsity of outdoor targets. These components are coordinated by a behavior tree, which adaptively switches behaviors for robust operation. We evaluate RAVEN in 10 photorealistic outdoor simulation environments over 100 semantic tasks, encompassing single-object search, multi-class, multi-instance navigation and sequential task changes. Results show RAVEN outperforms baselines by 85.25% in simulation and demonstrate its real-world applicability through deployment on an aerial robot in outdoor field tests.

Paperid: 3756, https://arxiv.org/pdf/2509.21774.pdf

Abstract:
Multimodal deepfake detection (MDD) aims to uncover manipulations across visual, textual, and auditory modalities, thereby reinforcing the reliability of modern information systems. Although large vision-language models (LVLMs) exhibit strong multimodal reasoning, their effectiveness in MDD is limited by challenges in capturing subtle forgery cues, resolving cross-modal inconsistencies, and performing task-aligned retrieval. To this end, we propose Guided Adaptive Scorer and Propagation In-Context Learning (GASP-ICL), a training-free framework for MDD. GASP-ICL employs a pipeline to preserve semantic relevance while injecting task-aware knowledge into LVLMs. We leverage an MDD-adapted feature extractor to retrieve aligned image-text pairs and build a candidate set. We further design the Graph-Structured Taylor Adaptive Scorer (GSTAS) to capture cross-sample relations and propagate query-aligned signals, producing discriminative exemplars. This enables precise selection of semantically aligned, task-relevant demonstrations, enhancing LVLMs for robust MDD. Experiments on four forgery types show that GASP-ICL surpasses strong baselines, delivering gains without LVLM fine-tuning.

Paperid: 3757, https://arxiv.org/pdf/2509.20823.pdf

Abstract:
Time series captioning, the task of describing time series in natural language, requires numeric and temporal reasoning, trend interpretation, and contextual understanding. Existing benchmarks, however, often rely on fully synthetic or generic captions, and typically neglect metadata and visual representations. We introduce CaTS-Bench, a comprehensive benchmark for Context-aware Time Series reasoning across $11$ diverse domains, centered on a gold-standard evaluation set of $1746$ human-rewritten captions that measure how effectively models translate numeric trends into immediately interpretable narratives. To address the scarcity of human-annotated data, we also propose a scalable pipeline for generating high-fidelity synthetic captions, the quality of which we validate. We evaluate leading Vision-Language Models on our benchmark, revealing that even proprietary models struggle to capture numeric nuances in temporal descriptions, while finetuning open-source models on synthetic data yields substantial performance gains. Finally, we release a diagnostic suite of $910$ multiple-choice questions and tailored numeric metrics to gauge time-series-specific reasoning capabilities, establishing CaTS-Bench as a reliable foundation for grounded, multimodal language generation in numeric domains.

Paperid: 3758, https://arxiv.org/pdf/2509.20628.pdf

Abstract:
Building-level occupancy after disasters is vital for triage, inspections, utility re-energization, and equitable resource allocation. Overhead imagery provides rapid coverage but often misses facade and access cues that determine habitability, while street-view imagery captures those details but is sparse and difficult to align with parcels. We present FacadeTrack, a street-level, language-guided framework that links panoramic video to parcels, rectifies views to facades, and elicits interpretable attributes (for example, entry blockage, temporary coverings, localized debris) that drive two decision strategies: a transparent one-stage rule and a two-stage design that separates perception from conservative reasoning. Evaluated across two post-Hurricane Helene surveys, the two-stage approach achieves a precision of 0.927, a recall of 0.781, and an F-1 score of 0.848, compared with the one-stage baseline at a precision of 0.943, a recall of 0.728, and an F-1 score of 0.822. Beyond accuracy, intermediate attributes and spatial diagnostics reveal where and why residual errors occur, enabling targeted quality control. The pipeline provides auditable, scalable occupancy assessments suitable for integration into geospatial and emergency-management workflows.

Paperid: 3759, https://arxiv.org/pdf/2509.19768.pdf

Abstract:
Accurate text recognition for historical documents can greatly advance the study and preservation of cultural heritage. Existing vision-language models (VLMs), however, are designed for modern, standardized texts and are not equipped to read the diverse languages and scripts, irregular layouts, and frequent degradation found in historical materials. This paper presents CHURRO, a 3B-parameter open-weight VLM specialized for historical text recognition. The model is trained on CHURRO-DS, the largest historical text recognition dataset to date. CHURRO-DS unifies 155 historical corpora comprising 99,491 pages, spanning 22 centuries of textual heritage across 46 language clusters, including historical variants and dead languages. We evaluate several open-weight and closed VLMs and optical character recognition (OCR) systems on CHURRO-DS and find that CHURRO outperforms all other VLMs. On the CHURRO-DS test set, CHURRO achieves 82.3% (printed) and 70.1% (handwritten) normalized Levenshtein similarity, surpassing the second-best model, Gemini 2.5 Pro, by 1.4% and 6.5%, respectively, while being 15.5 times more cost-effective. By releasing the model and dataset, we aim to enable community-driven research to improve the readability of historical texts and accelerate scholarship.

Paperid: 3760, https://arxiv.org/pdf/2509.19012.pdf

Abstract:
The emergence of Vision Language Action (VLA) models marks a paradigm shift from traditional policy-based control to generalized robotics, reframing Vision Language Models (VLMs) from passive sequence generators into active agents for manipulation and decision-making in complex, dynamic environments. This survey delves into advanced VLA methods, aiming to provide a clear taxonomy and a systematic, comprehensive review of existing research. It presents a comprehensive analysis of VLA applications across different scenarios and classifies VLA approaches into several paradigms: autoregression-based, diffusion-based, reinforcement-based, hybrid, and specialized methods; while examining their motivations, core strategies, and implementations in detail. In addition, foundational datasets, benchmarks, and simulation platforms are introduced. Building on the current VLA landscape, the review further proposes perspectives on key challenges and future directions to advance research in VLA models and generalizable robotics. By synthesizing insights from over three hundred recent studies, this survey maps the contours of this rapidly evolving field and highlights the opportunities and challenges that will shape the development of scalable, general-purpose VLA methods.

Paperid: 3761, https://arxiv.org/pdf/2509.15772.pdf

Abstract:
Score Distillation Sampling (SDS) enables high-quality text-to-3D generation by supervising 3D models through the denoising of multi-view 2D renderings, using a pretrained text-to-image diffusion model to align with the input prompt and ensure 3D consistency. However, existing SDS-based methods face two fundamental limitations: (1) their reliance on CLIP-style text encoders leads to coarse semantic alignment and struggles with fine-grained prompts; and (2) 2D diffusion priors lack explicit 3D spatial constraints, resulting in geometric inconsistencies and inaccurate object relationships in multi-object scenes. To address these challenges, we propose VLM3D, a novel text-to-3D generation framework that integrates large vision-language models (VLMs) into the SDS pipeline as differentiable semantic and spatial priors. Unlike standard text-to-image diffusion priors, VLMs leverage rich language-grounded supervision that enables fine-grained prompt alignment. Moreover, their inherent vision language modeling provides strong spatial understanding, which significantly enhances 3D consistency for single-object generation and improves relational reasoning in multi-object scenes. We instantiate VLM3D based on the open-source Qwen2.5-VL model and evaluate it on the GPTeval3D benchmark. Experiments across diverse objects and complex scenes show that VLM3D significantly outperforms prior SDS-based methods in semantic fidelity, geometric coherence, and spatial correctness.

Paperid: 3762, https://arxiv.org/pdf/2509.13919.pdf

Abstract:
Large Vision-Language Models (LVLMs) have manifested strong visual question answering capability. However, they still struggle with aligning the rationale and the generated answer, leading to inconsistent reasoning and incorrect responses. To this end, this paper introduces the Self-Rationale Calibration (SRC) framework to iteratively calibrate the alignment between rationales and answers. SRC begins by employing a lightweight "rationale fine-tuning" approach, which modifies the model's response format to require a rationale before deriving an answer without explicit prompts. Next, SRC searches for a diverse set of candidate responses from the fine-tuned LVLMs for each sample, followed by a proposed pairwise scoring strategy using a tailored scoring model, R-Scorer, to evaluate both rationale quality and factual consistency of candidates. Based on a confidence-weighted preference curation process, SRC decouples the alignment calibration into a preference fine-tuning manner, leading to significant improvements of LVLMs in perception, reasoning, and generalization across multiple benchmarks. Our results emphasize the rationale-oriented alignment in exploring the potential of LVLMs.

Paperid: 3763, https://arxiv.org/pdf/2509.11961.pdf

Abstract:
Vision-Language Models (VLMs) enable powerful multimodal reasoning but suffer from slow autoregressive inference, limiting their deployment in real-time applications. We introduce Spec-LLaVA, a system that applies speculative decoding to accelerate VLMs without sacrificing output quality. Spec-LLaVA pairs a lightweight draft VLM with a large target model: the draft speculates future tokens, which the target verifies in parallel, allowing multiple tokens to be generated per step. To maximize efficiency, we design a dynamic tree-based verification algorithm that adaptively expands and prunes speculative branches using draft model confidence. On MS COCO out-of-domain images, Spec-LLaVA achieves up to 3.28$\times$ faster decoding on LLaVA-1.5 (7B, 13B) with no loss in generation quality. This work presents a lossless acceleration framework for VLMs using dynamic tree-structured speculative decoding, opening a path toward practical real-time multimodal assistants. Importantly, the lightweight draft model design makes the framework amenable to resource-constrained or on-device deployment settings.

Paperid: 3764, https://arxiv.org/pdf/2509.08826.pdf

Abstract:
Reward Models (RMs) are critical for improving generation models via Reinforcement Learning (RL), yet the RM scaling paradigm in visual generation remains largely unexplored. It primarily due to fundamental limitations in existing approaches: CLIP-based RMs suffer from architectural and input modality constraints, while prevalent Bradley-Terry losses are fundamentally misaligned with the next-token prediction mechanism of Vision-Language Models (VLMs), hindering effective scaling. More critically, the RLHF optimization process is plagued by Reward Hacking issue, where models exploit flaws in the reward signal without improving true quality. To address these challenges, we introduce RewardDance, a scalable reward modeling framework that overcomes these barriers through a novel generative reward paradigm. By reformulating the reward score as the model's probability of predicting a "yes" token, indicating that the generated image outperforms a reference image according to specific criteria, RewardDance intrinsically aligns reward objectives with VLM architectures. This alignment unlocks scaling across two dimensions: (1) Model Scaling: Systematic scaling of RMs up to 26 billion parameters; (2) Context Scaling: Integration of task-specific instructions, reference examples, and chain-of-thought (CoT) reasoning. Extensive experiments demonstrate that RewardDance significantly surpasses state-of-the-art methods in text-to-image, text-to-video, and image-to-video generation. Crucially, we resolve the persistent challenge of "reward hacking": Our large-scale RMs exhibit and maintain high reward variance during RL fine-tuning, proving their resistance to hacking and ability to produce diverse, high-quality outputs. It greatly relieves the mode collapse problem that plagues smaller models.

Paperid: 3765, https://arxiv.org/pdf/2509.00419.pdf

Abstract:
In this paper, we introduce LightVLM, a simple but effective method that can be seamlessly deployed upon existing Vision-Language Models (VLMs) to greatly accelerate the inference process in a training-free manner. We divide the inference procedure of VLMs into two stages, i.e., encoding and decoding, and propose to simultaneously accelerate VLMs in both stages to largely improve model efficiency. During encoding, we propose pyramid token merging to reduce tokens of different LLM layers in a hierarchical manner by finally only keeping a few dominant tokens to achieve high efficiency. During decoding, aimed at reducing the high latency of outputting long sequences, we propose KV Cache compression to remove unnecessary caches to increase the network throughput. Experimental results show that LightVLM successfully retains 100% performance when only preserving 35% image tokens, and maintains around 98% performance when keeping only 3% image tokens. LightVLM could 2.02$\times$ the network throughput and reduce the prefilling time by 3.65$\times$. LightVLM also makes large VLMs faster again by enabling a heavy model (e.g., InternVL2.5 26B) to infer faster than significantly smaller models (e.g., InternVL2.5 8B), hopefully facilitating the real-world deployment. When generating long text sequences (e.g., 4096 tokens), LightVLM could reduce the inference time by 3.21$\times$, largely outperforming existing methods.

Paperid: 3766, https://arxiv.org/pdf/2508.21539.pdf

Abstract:
Natural Language-Guided Drones (NLGD) provide a novel paradigm for tasks such as target matching and navigation. However, the wide field of view and complex compositional semantics in drone scenarios pose challenges for vision-language understanding. Mainstream Vision-Language Models (VLMs) emphasize global alignment while lacking fine-grained semantics, and existing hierarchical methods depend on precise entity partitioning and strict containment, limiting effectiveness in dynamic environments. To address this, we propose the Hierarchical Cross-Granularity Contrastive and Matching learning (HCCM) framework with two components: (1) Region-Global Image-Text Contrastive Learning (RG-ITC), which avoids precise scene partitioning and captures hierarchical local-to-global semantics by contrasting local visual regions with global text and vice versa; (2) Region-Global Image-Text Matching (RG-ITM), which dispenses with rigid constraints and instead evaluates local semantic consistency within global cross-modal representations, enhancing compositional reasoning. Moreover, drone text descriptions are often incomplete or ambiguous, destabilizing alignment. HCCM introduces a Momentum Contrast and Distillation (MCD) mechanism to improve robustness. Experiments on GeoText-1652 show HCCM achieves state-of-the-art Recall@1 of 28.8% (image retrieval) and 14.7% (text retrieval). On the unseen ERA dataset, HCCM demonstrates strong zero-shot generalization with 39.93% mean recall (mR), outperforming fine-tuned baselines.

Paperid: 3767, https://arxiv.org/pdf/2508.19613.pdf

Abstract:
Estimating model accuracy on unseen, unlabeled datasets is crucial for real-world machine learning applications, especially under distribution shifts that can degrade performance. Existing methods often rely on predicted class probabilities (softmax scores) or data similarity metrics. While softmax-based approaches benefit from representing predictions on the standard simplex, compressing logits into probabilities leads to information loss. Meanwhile, similarity-based methods can be computationally expensive and domain-specific, limiting their broader applicability. In this paper, we introduce ALSA (Anchors in Logit Space for Accuracy estimation), a novel framework that preserves richer information by operating directly in the logit space. Building on theoretical insights and empirical observations, we demonstrate that the aggregation and distribution of logits exhibit a strong correlation with the predictive performance of the model. To exploit this property, ALSA employs an anchor-based modeling strategy: multiple learnable anchors are initialized in logit space, each assigned an influence function that captures subtle variations in the logits. This allows ALSA to provide robust and accurate performance estimates across a wide range of distribution shifts. Extensive experiments on vision, language, and graph benchmarks demonstrate ALSA's superiority over both softmax- and similarity-based baselines. Notably, ALSA's robustness under significant distribution shifts highlights its potential as a practical tool for reliable model evaluation.

Paperid: 3768, https://arxiv.org/pdf/2508.19111.pdf

Abstract:
Large vision-language models (LVLMs) demonstrate strong visual question answering (VQA) capabilities but are shown to hallucinate. A reliable model should perceive its knowledge boundaries-knowing what it knows and what it does not. This paper investigates LVLMs' perception of their knowledge boundaries by evaluating three types of confidence signals: probabilistic confidence, answer consistency-based confidence, and verbalized confidence. Experiments on three LVLMs across three VQA datasets show that, although LVLMs possess a reasonable perception level, there is substantial room for improvement. Among the three confidences, probabilistic and consistency-based signals are more reliable indicators, while verbalized confidence often leads to overconfidence. To enhance LVLMs' perception, we adapt several established confidence calibration methods from Large Language Models (LLMs) and propose three effective methods. Additionally, we compare LVLMs with their LLM counterparts, finding that jointly processing visual and textual inputs decreases question-answering performance but reduces confidence, resulting in an improved perception level compared to LLMs.

Paperid: 3769, https://arxiv.org/pdf/2508.10667.pdf

Abstract:
Large visual language models (LVLMs) have demonstrated impressive performance in coarse-grained geo-localization at the country or city level, but they struggle with fine-grained street-level localization within urban areas. In this paper, we explore integrating city-wide address localization capabilities into LVLMs, facilitating flexible address-related question answering using street-view images. A key challenge is that the street-view visual question-and-answer (VQA) data provides only microscopic visual cues, leading to subpar performance in fine-tuned models. To tackle this issue, we incorporate perspective-invariant satellite images as macro cues and propose cross-view alignment tuning including a satellite-view and street-view image grafting mechanism, along with an automatic label generation mechanism. Then LVLM's global understanding of street distribution is enhanced through cross-view matching. Our proposed model, named AddressVLM, consists of two-stage training protocols: cross-view alignment tuning and address localization tuning. Furthermore, we have constructed two street-view VQA datasets based on image address localization datasets from Pittsburgh and San Francisco. Qualitative and quantitative evaluations demonstrate that AddressVLM outperforms counterpart LVLMs by over 9% and 12% in average address localization accuracy on these two datasets, respectively.

Paperid: 3770, https://arxiv.org/pdf/2508.07871.pdf

Abstract:
Modern large vision-language models (LVLMs) convert each input image into a large set of tokens, far outnumbering the text tokens. Although this improves visual perception, it introduces severe image token redundancy. Because image tokens carry sparse information, many add little to reasoning, yet greatly increase inference cost. The emerging image token pruning methods tackle this issue by identifying the most important tokens and discarding the rest. These methods can raise efficiency with only modest performance loss. However, most of them only consider single-image tasks and overlook multimodal in-context learning (ICL), where redundancy is greater and efficiency is more critical. Redundant tokens weaken the advantage of multimodal ICL for rapid domain adaptation and cause unstable performance. Applying existing pruning methods in this setting leads to large accuracy drops, exposing a clear gap and the need for new techniques. Thus, we propose Contextually Adaptive Token Pruning (CATP), a training-free pruning method targeted at multimodal ICL. CATP consists of two stages that perform progressive pruning to fully account for the complex cross-modal interactions in the input sequence. After removing 77.8\% of the image tokens, CATP produces an average performance gain of 0.6\% over the vanilla model on four LVLMs and eight benchmarks, exceeding all baselines remarkably. Meanwhile, it effectively improves efficiency by achieving an average reduction of 10.78\% in inference latency. CATP enhances the practical value of multimodal ICL and lays the groundwork for future progress in interleaved image-text scenarios.

Paperid: 3771, https://arxiv.org/pdf/2508.06038.pdf

Abstract:
Vision-Language Models (VLMs) typically replace the predefined image placeholder token () in textual instructions with visual features from an image encoder, forming the input to a backbone Large Language Model (LLM). However, the large number of vision tokens significantly increases the context length, leading to high computational overhead and inference latency. While previous efforts mitigate this by selecting only important visual features or leveraging learnable queries to reduce token count, they often compromise performance or introduce substantial extra costs. In response, we propose Fourier-VLM, a simple yet efficient method that compresses visual representations in the frequency domain. Our approach is motivated by the observation that vision features output from the vision encoder exhibit concentrated energy in low-frequency components. Leveraging this, we apply a low-pass filter to the vision features using a two-dimensional Discrete Cosine Transform (DCT). Notably, the DCT is efficiently computed via the Fast Fourier Transform (FFT) operator with a time complexity of $\mathcal{O}(n\log n)$, minimizing the extra computational cost while introducing no additional parameters. Extensive experiments across various image-based benchmarks demonstrate that Fourier-VLM achieves competitive performance with strong generalizability across both LLaVA and Qwen-VL architectures. Crucially, it reduce inference FLOPs by up to 83.8% and boots generation speed by 31.2% compared to LLaVA-v1.5, highlighting the superior efficiency and practicality.

Paperid: 3772, https://arxiv.org/pdf/2508.02338.pdf

Abstract:
Autonomous Mobile Robots (AMRs) are deployed in diverse environments (e.g., warehouses, retail spaces, and offices), where they work alongside humans. Given that human behavior can be unpredictable and that AMRs may not have been trained to handle all possible unknown and uncertain behaviors, it is important to test AMRs under a wide range of human interactions to ensure their safe behavior. Moreover, testing in real environments with actual AMRs and humans is often costly, impractical, and potentially hazardous (e.g., it could result in human injury). To this end, we propose a Vision Language Model (VLM)-based testing approach (RVSG) for industrial AMRs developed by PAL Robotics in Spain. Based on the functional and safety requirements, RVSG uses the VLM to generate diverse human behaviors that violate these requirements. We evaluated RVSG with several requirements and navigation routes in a simulator using the latest AMR from PAL Robotics. Our results show that, compared with the baseline, RVSG can effectively generate requirement-violating scenarios. Moreover, RVSG-generated scenarios increase variability in robot behavior, thereby helping reveal their uncertain behaviors.

Paperid: 3773, https://arxiv.org/pdf/2508.01293.pdf

Abstract:
Multiple Instance Learning (MIL) is the leading approach for whole slide image (WSI) classification, enabling efficient analysis of gigapixel pathology slides. Recent work has introduced vision-language models (VLMs) into MIL pipelines to incorporate medical knowledge through text-based class descriptions rather than simple class names. However, when these methods rely on large language models (LLMs) to generate clinical descriptions or use fixed-length prompts to represent complex pathology concepts, the limited token capacity of VLMs often constrains the expressiveness and richness of the encoded class information. Additionally, descriptions generated solely by LLMs may lack domain grounding and fine-grained medical specificity, leading to suboptimal alignment with visual features. To address these challenges, we propose a vision-language MIL framework with two key contributions: (1) A grounded multi-agent description generation system that leverages curated pathology textbooks and agent specialization (e.g., morphology, spatial context) to produce accurate and diverse clinical descriptions; (2) A text encoding strategy using a list of descriptions rather than a single prompt, capturing fine-grained and complementary clinical signals for better alignment with visual features. Integrated into a VLM-MIL pipeline, our approach shows improved performance over single-prompt class baselines and achieves results comparable to state-of-the-art models, as demonstrated on renal and lung cancer datasets.

Paperid: 3774, https://arxiv.org/pdf/2507.17722.pdf

Abstract:
Large language models (LLMs) are growingly extended to process multimodal data such as text and video simultaneously. Their remarkable performance in understanding what is shown in images is surpassing specialized neural networks (NNs) such as Yolo that is supporting only a well-formed but very limited vocabulary, ie., objects that they are able to detect. When being non-restricted, LLMs and in particular state-of-the-art vision language models (VLMs) show impressive performance to describe even complex traffic situations. This is making them potentially suitable components for automotive perception systems to support the understanding of complex traffic situations or edge case situation. However, LLMs and VLMs are prone to hallucination, which mean to either potentially not seeing traffic agents such as vulnerable road users who are present in a situation, or to seeing traffic agents who are not there in reality. While the latter is unwanted making an ADAS or autonomous driving systems (ADS) to unnecessarily slow down, the former could lead to disastrous decisions from an ADS. In our work, we are systematically assessing the performance of 3 state-of-the-art VLMs on a diverse subset of traffic situations sampled from the Waymo Open Dataset to support safety guardrails for capturing such hallucinations in VLM-supported perception systems. We observe that both, proprietary and open VLMs exhibit remarkable image understanding capabilities even paying thorough attention to fine details sometimes difficult to spot for us humans. However, they are also still prone to making up elements in their descriptions to date requiring hallucination detection strategies such as BetterCheck that we propose in our work.

Paperid: 3775, https://arxiv.org/pdf/2507.13773.pdf

Abstract:
In visual question answering (VQA) context, users often pose ambiguous questions to visual language models (VLMs) due to varying expression habits. Existing research addresses such ambiguities primarily by rephrasing questions. These approaches neglect the inherently interactive nature of user interactions with VLMs, where ambiguities can be clarified through user feedback. However, research on interactive clarification faces two major challenges: (1) Benchmarks are absent to assess VLMs' capacity for resolving ambiguities through interaction; (2) VLMs are trained to prefer answering rather than asking, preventing them from seeking clarification. To overcome these challenges, we introduce \textbf{ClearVQA} benchmark, which targets three common categories of ambiguity in VQA context, and encompasses various VQA scenarios.

Paperid: 3776, https://arxiv.org/pdf/2507.11892.pdf

Abstract:
Dynamic Facial Expression Recognition (DFER) aims to identify human emotions from temporally evolving facial movements and plays a critical role in affective computing. While recent vision-language approaches have introduced semantic textual descriptions to guide expression recognition, existing methods still face two key limitations: they often underutilize the subtle emotional cues embedded in generated text, and they have yet to incorporate sufficiently effective mechanisms for filtering out facial dynamics that are irrelevant to emotional expression. To address these gaps, We propose GRACE, Granular Representation Alignment for Cross-modal Emotion recognition that integrates dynamic motion modeling, semantic text refinement, and token-level cross-modal alignment to facilitate the precise localization of emotionally salient spatiotemporal features. Our method constructs emotion-aware textual descriptions via a Coarse-to-fine Affective Text Enhancement (CATE) module and highlights expression-relevant facial motion through a motion-difference weighting mechanism. These refined semantic and visual signals are aligned at the token level using entropy-regularized optimal transport. Experiments on three benchmark datasets demonstrate that our method significantly improves recognition performance, particularly in challenging settings with ambiguous or imbalanced emotion classes, establishing new state-of-the-art (SOTA) results in terms of both UAR and WAR.

Paperid: 3777, https://arxiv.org/pdf/2507.06484.pdf

Abstract:
Despite large-scale pretraining endowing models with language and vision reasoning capabilities, improving their spatial reasoning capability remains challenging due to the lack of data grounded in the 3D world. While it is possible for humans to manually create immersive and interactive worlds through 3D graphics, as seen in applications such as VR, gaming, and robotics, this process remains highly labor-intensive. In this paper, we propose a scalable method for generating high-quality 3D environments that can serve as training data for foundation models. We recast 3D environment building as a sequential decision-making problem, employing Vision-Language-Models (VLMs) as policies that output actions to jointly craft a 3D environment's layout, materials, lighting, and assets. Our proposed framework, 3D-Generalist, trains VLMs to generate more prompt-aligned 3D environments via self-improvement fine-tuning. We demonstrate the effectiveness of 3D-Generalist and the proposed training strategy in generating simulation-ready 3D environments. Furthermore, we demonstrate its quality and scalability in synthetic data generation by pretraining a vision foundation model on the generated data. After fine-tuning the pre-trained model on downstream tasks, we show that it surpasses models pre-trained on meticulously human-crafted synthetic data and approaches results achieved with real data orders of magnitude larger.

Paperid: 3778, https://arxiv.org/pdf/2507.03458.pdf

Abstract:
Vision-Language Models (VLMs) like CLIP achieve cross-modal semantic alignment through contrastive learning, exhibiting robust zero-shot generalization. Traditional prompt engineering, however, predominantly relies on coarse-grained category labels, neglecting fine-grained local semantics. Existing approaches assume that VLMs inherently recognize localized visual details and attempt to enhance classification by augmenting text prompts with attribute descriptors generated by large language models. However, our systematic experiments reveal critical limitations: CLIP's strong bias toward global image patterns hinders its ability to process localized visual descriptors. To address this fundamental constraint, we propose a simple, effective, and plug-and-play solution that enables CLIP to ``See Both the Forest and the Trees." Specifically, we employ stochastic multi-crop augmentation to activate CLIP's latent capacity for localized feature analysis. By cropping only partial regions, the approach effectively constrains the model's receptive field and recalibrates its attention mechanism, thereby mitigating its inherent bias. We evaluate the proposed method under zero-shot, few-shot, and test-time adaptation settings, and extensive experiments demonstrate that D&D achieves promising performance.

Paperid: 3779, https://arxiv.org/pdf/2507.02600.pdf

Abstract:
Articulated object manipulation remains a critical challenge in robotics due to the complex kinematic constraints and the limited physical reasoning of existing methods. In this work, we introduce ArtGS, a novel framework that extends 3D Gaussian Splatting (3DGS) by integrating visual-physical modeling for articulated object understanding and interaction. ArtGS begins with multi-view RGB-D reconstruction, followed by reasoning with a vision-language model (VLM) to extract semantic and structural information, particularly the articulated bones. Through dynamic, differentiable 3DGS-based rendering, ArtGS optimizes the parameters of the articulated bones, ensuring physically consistent motion constraints and enhancing the manipulation policy. By leveraging dynamic Gaussian splatting, cross-embodiment adaptability, and closed-loop optimization, ArtGS establishes a new framework for efficient, scalable, and generalizable articulated object modeling and manipulation. Experiments conducted in both simulation and real-world environments demonstrate that ArtGS significantly outperforms previous methods in joint estimation accuracy and manipulation success rates across a variety of articulated objects. Additional images and videos are available on the project website: https://sites.google.com/view/artgs/home

Paperid: 3780, https://arxiv.org/pdf/2507.02001.pdf

Abstract:
Despite recent advances in Vision-Language Models (VLMs), long-video understanding remains a challenging problem. Although state-of-the-art long-context VLMs can process around 1000 input frames, they still struggle to effectively leverage this sequence length, and succumb to irrelevant distractors within the context window. We present Temporal Chain of Thought, an inference strategy for video question-answering that curates the model's input context. We use the VLM itself to iteratively identify and extract the most relevant frames from the video, which are then used for answering. We demonstrate how leveraging more computation at inference-time to select the most relevant context leads to improvements in accuracy, in agreement with recent work on inference-time scaling of LLMs. Moreover, we achieve state-of-the-art results on 4 diverse video question-answering datasets, showing consistent improvements with 3 different VLMs. In particular, our method shines on longer videos which would not otherwise fit within the model's context window: On longer videos of more than 1 hour on LVBench, our approach using a context window of 32K outperforms the same VLM using standard inference with a 700K context window by 2.8 points.

Paperid: 3781, https://arxiv.org/pdf/2507.00006.pdf

Abstract:
We propose MVGBench, a comprehensive benchmark for multi-view image generation models (MVGs) that evaluates 3D consistency in geometry and texture, image quality, and semantics (using vision language models). Recently, MVGs have been the main driving force in 3D object creation. However, existing metrics compare generated images against ground truth target views, which is not suitable for generative tasks where multiple solutions exist while differing from ground truth. Furthermore, different MVGs are trained on different view angles, synthetic data and specific lightings -- robustness to these factors and generalization to real data are rarely evaluated thoroughly. Without a rigorous evaluation protocol, it is also unclear what design choices contribute to the progress of MVGs. MVGBench evaluates three different aspects: best setup performance, generalization to real data and robustness. Instead of comparing against ground truth, we introduce a novel 3D self-consistency metric which compares 3D reconstructions from disjoint generated multi-views. We systematically compare 12 existing MVGs on 4 different curated real and synthetic datasets. With our analysis, we identify important limitations of existing methods specially in terms of robustness and generalization, and we find the most critical design choices. Using the discovered best practices, we propose ViFiGen, a method that outperforms all evaluated MVGs on 3D consistency. Our code, model, and benchmark suite will be publicly released.

Paperid: 3782, https://arxiv.org/pdf/2506.19608.pdf

Abstract:
Continual learning (CL) empowers pre-trained vision-language models to adapt effectively to novel or previously underrepresented data distributions without comprehensive retraining, enhancing their adaptability and efficiency. While vision-language models like CLIP show great promise, they struggle to maintain performance across domains in incremental learning scenarios. Existing prompt learning methods face two main limitations: 1) they primarily focus on class-incremental learning scenarios, lacking specific strategies for multi-domain task incremental learning; 2) most current approaches employ single-modal prompts, neglecting the potential benefits of cross-modal information exchange. To address these challenges, we propose the \ChordPrompt framework, which facilitates a harmonious interplay between visual and textual prompts. \ChordPrompt introduces cross-modal prompts to leverage interactions between visual and textual information. Our approach also employs domain-adaptive text prompts to select appropriate prompts for continual adaptation across multiple domains. Comprehensive experiments on multi-domain incremental learning benchmarks demonstrate that \ChordPrompt outperforms state-of-the-art methods in zero-shot generalization and downstream task performance.

Paperid: 3783, https://arxiv.org/pdf/2506.09047.pdf

Abstract:
Vision-Language models (VLMs) show impressive abilities to answer questions on visual inputs (e.g., counting objects in an image), yet demonstrate higher accuracies when performing an analogous task on text (e.g., counting words in a text). We investigate this accuracy gap by identifying and comparing the \textit{circuits} - the task-specific computational sub-graphs - in different modalities. We show that while circuits are largely disjoint between modalities, they implement relatively similar functionalities: the differences lie primarily in processing modality-specific data positions (an image or a text sequence). Zooming in on the image data representations, we observe they become aligned with the higher-performing analogous textual representations only towards later layers, too late in processing to effectively influence subsequent positions. To overcome this, we patch the representations of visual data tokens from later layers back into earlier layers. In experiments with multiple tasks and models, this simple intervention closes a third of the performance gap between the modalities, on average. Our analysis sheds light on the multi-modal performance gap in VLMs and suggests a training-free approach for reducing it.

Paperid: 3784, https://arxiv.org/pdf/2506.07943.pdf

Abstract:
Reasoning Segmentation (RS) is a multimodal vision-text task that requires segmenting objects based on implicit text queries, demanding both precise visual perception and vision-text reasoning capabilities. Current RS approaches rely on fine-tuning vision-language models (VLMs) for both perception and reasoning, but their tokenization of images fundamentally disrupts continuous spatial relationships between objects. We introduce DTwinSeger, a novel RS approach that leverages Digital Twin (DT) representation as an intermediate layer to decouple perception from reasoning. Innovatively, DTwinSeger reformulates RS as a two-stage process, where the first transforms the image into a structured DT representation that preserves spatial relationships and semantic properties and then employs a Large Language Model (LLM) to perform explicit reasoning over this representation to identify target objects. We propose a supervised fine-tuning method specifically for LLM with DT representation, together with a corresponding fine-tuning dataset Seg-DT, to enhance the LLM's reasoning capabilities with DT representations. Experiments show that our method can achieve state-of-the-art performance on two image RS benchmarks and three image referring segmentation benchmarks. It yields that DT representation functions as an effective bridge between vision and text, enabling complex multimodal reasoning tasks to be accomplished solely with an LLM.

Paperid: 3785, https://arxiv.org/pdf/2506.05429.pdf

Abstract:
Vision-language models, which integrate computer vision and natural language processing capabilities, have demonstrated significant advancements in tasks such as image captioning and visual question and answering. However, similar to traditional models, they are susceptible to small perturbations, posing a challenge to their robustness, particularly in deployment scenarios. Evaluating the robustness of these models requires perturbations in both the vision and language modalities to learn their inter-modal dependencies. In this work, we train a generic surrogate model that can take both image and text as input and generate joint representation which is further used to generate adversarial perturbations for both the text and image modalities. This coordinated attack strategy is evaluated on the visual question and answering and visual reasoning datasets using various state-of-the-art vision-language models. Our results indicate that the proposed strategy outperforms other multi-modal attacks and single-modality attacks from the recent literature. Our results demonstrate their effectiveness in compromising the robustness of several state-of-the-art pre-trained multi-modal models such as instruct-BLIP, ViLT and others.

Paperid: 3786, https://arxiv.org/pdf/2506.03189.pdf

Abstract:
Continual learning is conventionally tackled through sequential fine-tuning, a process that, while enabling adaptation, inherently favors plasticity over the stability needed to retain prior knowledge. While existing approaches attempt to mitigate catastrophic forgetting, a bias towards recent tasks persists as they build upon this sequential nature. In this work we present a new perspective based on model merging to maintain stability while still retaining plasticity. Rather than just sequentially updating the model weights, we propose merging newly trained task parameters with previously learned ones, promoting a better balance. To maximize the effectiveness of the merging process, we propose a simple mechanism that promotes learning aligned weights with previous ones, thereby avoiding interference when merging. We evaluate this approach on large Vision-Language Models (VLMs), and demonstrate its effectiveness in reducing forgetting, increasing robustness to various task orders and similarities, and improving generalization.

Paperid: 3787, https://arxiv.org/pdf/2506.00555.pdf

Abstract:
Medical Large Vision-Language Models (Med-LVLMs) have shown strong potential in multimodal diagnostic tasks. However, existing single-agent models struggle to generalize across diverse medical specialties, limiting their performance. Recent efforts introduce multi-agent collaboration frameworks inspired by clinical workflows, where general practitioners (GPs) and specialists interact in a fixed sequence. Despite improvements, these static pipelines lack flexibility and adaptability in reasoning. To address this, we propose MMedAgent-RL, a reinforcement learning (RL)-based multi-agent framework that enables dynamic, optimized collaboration among medical agents. Specifically, we train two GP agents based on Qwen2.5-VL via RL: the triage doctor learns to assign patients to appropriate specialties, while the attending physician integrates the judgments from multi-specialists and its own knowledge to make final decisions. To address the inconsistency in specialist outputs, we introduce a curriculum learning (CL)-guided RL strategy that progressively teaches the attending physician to balance between imitating specialists and correcting their mistakes. Experiments on five medical VQA benchmarks demonstrate that MMedAgent-RL not only outperforms both open-source and proprietary Med-LVLMs, but also exhibits human-like reasoning patterns. Notably, it achieves an average performance gain of 20.7% over supervised fine-tuning baselines.

Paperid: 3788, https://arxiv.org/pdf/2505.24871.pdf

Abstract:
Reinforcement Learning with Verifiable Rewards (RLVR) has recently emerged as a powerful paradigm for post-training large language models (LLMs), achieving state-of-the-art performance on tasks with structured, verifiable answers. Applying RLVR to Multimodal LLMs (MLLMs) presents significant opportunities but is complicated by the broader, heterogeneous nature of vision-language tasks that demand nuanced visual, logical, and spatial capabilities. As such, training MLLMs using RLVR on multiple datasets could be beneficial but creates challenges with conflicting objectives from interaction among diverse datasets, highlighting the need for optimal dataset mixture strategies to improve generalization and reasoning. We introduce a systematic post-training framework for Multimodal LLM RLVR, featuring a rigorous data mixture problem formulation and benchmark implementation. Specifically, (1) We developed a multimodal RLVR framework for multi-dataset post-training by curating a dataset that contains different verifiable vision-language problems and enabling multi-domain online RL learning with different verifiable rewards; (2) We proposed a data mixture strategy that learns to predict the RL fine-tuning outcome from the data mixture distribution, and consequently optimizes the best mixture. Comprehensive experiments showcase that multi-domain RLVR training, when combined with mixture prediction strategies, can significantly boost MLLM general reasoning capacities. Our best mixture improves the post-trained model's accuracy on out-of-distribution benchmarks by an average of 5.24% compared to the same model post-trained with uniform data mixture, and by a total of 20.74% compared to the pre-finetuning baseline.

Paperid: 3789, https://arxiv.org/pdf/2505.24649.pdf

Abstract:
Large vision-language models have become widely adopted to advance in various domains. However, developing a trustworthy system with minimal interpretable characteristics of large-scale models presents a significant challenge. One of the most prevalent terms associated with the fallacy functions caused by these systems is hallucination, where the language model generates a response that does not correspond to the visual content. To mitigate this problem, several approaches have been developed, and one prominent direction is to ameliorate the decoding process. In this paper, we propose a new Bijective Maximum Likelihood Learning (BIMA) approach to hallucination mitigation using normalizing flow theories. The proposed BIMA method can efficiently mitigate the hallucination problem in prevailing vision-language models, resulting in significant improvements. Notably, BIMA achieves the average F1 score of 85.06% on POPE benchmark and remarkably reduce CHAIRS and CHAIRI by 7.6% and 2.6%, respectively. To the best of our knowledge, this is one of the first studies that contemplates the bijection means to reduce hallucination induced by large vision-language models.

Paperid: 3790, https://arxiv.org/pdf/2505.22159.pdf

Abstract:
Vision-Language-Action (VLA) models have advanced general-purpose robotic manipulation by leveraging pretrained visual and linguistic representations. However, they struggle with contact-rich tasks that require fine-grained control involving force, especially under visual occlusion or dynamic uncertainty. To address these limitations, we propose ForceVLA, a novel end-to-end manipulation framework that treats external force sensing as a first-class modality within VLA systems. ForceVLA introduces FVLMoE, a force-aware Mixture-of-Experts fusion module that dynamically integrates pretrained visual-language embeddings with real-time 6-axis force feedback during action decoding. This enables context-aware routing across modality-specific experts, enhancing the robot's ability to adapt to subtle contact dynamics. We also introduce \textbf{ForceVLA-Data}, a new dataset comprising synchronized vision, proprioception, and force-torque signals across five contact-rich manipulation tasks. ForceVLA improves average task success by 23.2% over strong pi_0-based baselines, achieving up to 80% success in tasks such as plug insertion. Our approach highlights the importance of multimodal integration for dexterous manipulation and sets a new benchmark for physically intelligent robotic control. Code and data will be released at https://sites.google.com/view/forcevla2025.

Paperid: 3791, https://arxiv.org/pdf/2505.19486.pdf

Abstract:
Traffic signal control (TSC) is a core challenge in urban mobility, where real-time decisions must balance efficiency and safety. Existing methods - ranging from rule-based heuristics to reinforcement learning (RL) - often struggle to generalize to complex, dynamic, and safety-critical scenarios. We introduce VLMLight, a novel TSC framework that integrates vision-language meta-control with dual-branch reasoning. At the core of VLMLight is the first image-based traffic simulator that enables multi-view visual perception at intersections, allowing policies to reason over rich cues such as vehicle type, motion, and spatial density. A large language model (LLM) serves as a safety-prioritized meta-controller, selecting between a fast RL policy for routine traffic and a structured reasoning branch for critical cases. In the latter, multiple LLM agents collaborate to assess traffic phases, prioritize emergency vehicles, and verify rule compliance. Experiments show that VLMLight reduces waiting times for emergency vehicles by up to 65% over RL-only systems, while preserving real-time performance in standard conditions with less than 1% degradation. VLMLight offers a scalable, interpretable, and safety-aware solution for next-generation traffic signal control.

Paperid: 3792, https://arxiv.org/pdf/2505.17127.pdf

Abstract:
Multimodal Large Language Models (MLLMs) perform well on tasks such as visual question answering, but it remains unclear whether their reasoning relies more on memorized world knowledge or on the visual information present in the input image. To investigate this, we introduce Visual CounterFact, a new dataset of visually-realistic counterfactuals that put world knowledge priors (e.g, red strawberry) into direct conflict with visual input (e.g, blue strawberry). Using Visual CounterFact, we show that model predictions initially reflect memorized priors, but shift toward visual evidence in mid-to-late layers. This dynamic reveals a competition between the two modalities, with visual input ultimately overriding priors during evaluation. To control this behavior, we propose Pixels Versus Priors (PvP) steering vectors, a mechanism for controlling model outputs toward either world knowledge or visual input through activation-level interventions. On average, PvP successfully shifts 92.5% of color and 74.6% of size predictions from priors to counterfactuals. Together, these findings offer new tools for interpreting and controlling factual behavior in multimodal models.

Paperid: 3793, https://arxiv.org/pdf/2505.17098.pdf

Abstract:
Multimodal in-context learning (ICL) has emerged as a key mechanism for harnessing the capabilities of large vision-language models (LVLMs). However, its effectiveness remains highly sensitive to the quality of input ICL sequences, particularly for tasks involving complex reasoning or open-ended generation. A major limitation is our limited understanding of how LVLMs actually exploit these sequences during inference. To bridge this gap, we systematically interpret multimodal ICL through the lens of task mapping, which reveals how local and global relationships within and among demonstrations guide model reasoning. Building on this insight, we present TACO, a lightweight transformer-based model equipped with task-aware attention that dynamically configures ICL sequences. By injecting task-mapping signals into the autoregressive decoding process, TACO creates a bidirectional synergy between sequence construction and task reasoning. Experiments on five LVLMs and nine datasets demonstrate that TACO consistently surpasses baselines across diverse ICL tasks. These results position task mapping as a novel and valuable perspective for interpreting and improving multimodal ICL.

Paperid: 3794, https://arxiv.org/pdf/2505.17097.pdf

Abstract:
Multimodal in-context learning (ICL) is emerging as a key capability that enables large vision-language models (LVLMs) to adapt to novel tasks without parameter updates, expanding their utility across various real-world applications. However, ICL remains unstable, even with well-matched in-context demonstrations (ICDs), suggesting that LVLMs struggle to fully utilize the provided context. While existing efforts focus on prompt engineering or post-hoc logit calibration, we instead investigate the underlying attention dynamics to overcome LVLMs' inherent limitations. We identify two critical deficits in their self-attention that impair effective ICL. To bridge the gap, we propose \textbf{Context-Aware Modulated Attention} (CAMA), a plug-and-play and training-free method that dynamically modulates LVLM's attention logits based on the input in-context sequence. CAMA employs a two-stage attention modulation to address both identified deficits, enhancing the focus on semantically significant tokens, particularly visual ones. Across four LVLMs and seven benchmarks, CAMA consistently outperforms vanilla models and baselines, demonstrating great effectiveness and generalization. It can also activate the desired effects of prompt engineering methods and remains robust under diverse sequence configurations. Thus, CAMA paves the way for deeper explorations of attention dynamics to advance multimodal reasoning.

Paperid: 3795, https://arxiv.org/pdf/2505.15249.pdf

Abstract:
Recently, large vision-language models (LVLMs) have emerged as the preferred tools for judging text-image alignment, yet their robustness along the visual modality remains underexplored. This work is the first study to address a key research question: Can adversarial visual manipulations systematically fool LVLM judges into assigning unfairly inflated scores? We define potential image induced biases within the context of T2I evaluation and examine how these biases affect the evaluations of LVLM judges. Moreover, we introduce a novel, fine-grained, multi-domain meta-evaluation benchmark named FRAME, which is deliberately constructed to exhibit diverse score distributions. By introducing the defined biases into the benchmark, we reveal that all tested LVLM judges exhibit vulnerability across all domains, consistently inflating scores for manipulated images. Further analysis reveals that combining multiple biases amplifies their effects, and pairwise evaluations are similarly susceptible. Moreover, we observe that visual biases persist under prompt-based mitigation strategies, highlighting the vulnerability of current LVLM evaluation systems and underscoring the urgent need for more robust LVLM judges.

Paperid: 3796, https://arxiv.org/pdf/2505.01457.pdf

Abstract:
Retrieval-augmented generation (RAG) systems have predominantly focused on text-based retrieval, limiting their effectiveness in handling visually-rich documents that encompass text, images, tables, and charts. To bridge this gap, we propose a unified multi-granularity multimodal retrieval framework tailored for two benchmark tasks: MMDocIR and M2KR. Our approach integrates hierarchical encoding strategies, modality-aware retrieval mechanisms, and vision-language model (VLM)-based candidate filtering to effectively capture and utilize the complex interdependencies between textual and visual modalities. By leveraging off-the-shelf vision-language models and implementing a training-free hybrid retrieval strategy, our framework demonstrates robust performance without the need for task-specific fine-tuning. Experimental evaluations reveal that incorporating layout-aware search and VLM-based candidate verification significantly enhances retrieval accuracy, achieving a top performance score of 65.56. This work underscores the potential of scalable and reproducible solutions in advancing multimodal document retrieval systems.

Paperid: 3797, https://arxiv.org/pdf/2504.20468.pdf

Abstract:
Large Vision-Language Models (LVLMs) have achieved impressive results across various cross-modal tasks. However, hallucinations, i.e., the models generating counterfactual responses, remain a challenge. Though recent studies have attempted to alleviate object perception hallucinations, they focus on the models' response generation, and overlooking the task question itself. This paper discusses the vulnerability of LVLMs in solving counterfactual presupposition questions (CPQs), where the models are prone to accept the presuppositions of counterfactual objects and produce severe hallucinatory responses. To this end, we introduce "Antidote", a unified, synthetic data-driven post-training framework for mitigating both types of hallucination above. It leverages synthetic data to incorporate factual priors into questions to achieve self-correction, and decouple the mitigation process into a preference optimization problem. Furthermore, we construct "CP-Bench", a novel benchmark to evaluate LVLMs' ability to correctly handle CPQs and produce factual responses. Applied to the LLaVA series, Antidote can simultaneously enhance performance on CP-Bench by over 50%, POPE by 1.8-3.3%, and CHAIR & SHR by 30-50%, all without relying on external supervision from stronger LVLMs or human feedback and introducing noticeable catastrophic forgetting issues.

Paperid: 3798, https://arxiv.org/pdf/2504.19854.pdf

Abstract:
Existing Visual-Language-Action (VLA) models have shown promising performance in zero-shot scenarios, demonstrating impressive task execution and reasoning capabilities. However, a significant challenge arises from the limitations of visual encoding, which can result in failures during tasks such as object grasping. Moreover, these models typically suffer from high computational overhead due to their large sizes, often exceeding 7B parameters. While these models excel in reasoning and task planning, the substantial computational overhead they incur makes them impractical for real-time robotic environments, where speed and efficiency are paramount. To address the limitations of existing VLA models, we propose NORA, a 3B-parameter model designed to reduce computational overhead while maintaining strong task performance. NORA adopts the Qwen-2.5-VL-3B multimodal model as its backbone, leveraging its superior visual-semantic understanding to enhance visual reasoning and action grounding. Additionally, our \model{} is trained on 970k real-world robot demonstrations and equipped with the FAST+ tokenizer for efficient action sequence generation. Experimental results demonstrate that NORA outperforms existing large-scale VLA models, achieving better task performance with significantly reduced computational overhead, making it a more practical solution for real-time robotic autonomy.

Paperid: 3799, https://arxiv.org/pdf/2504.04099.pdf

Abstract:
Large Vision-Language Models have demonstrated remarkable performance across various tasks; however, the challenge of hallucinations constrains their practical applications. The hallucination problem arises from multiple factors, including the inherent hallucinations in language models, the limitations of visual encoders in perception, and biases introduced by multimodal data. Extensive research has explored ways to mitigate hallucinations. For instance, OPERA prevents the model from overly focusing on "anchor tokens", thereby reducing hallucinations, whereas VCD mitigates hallucinations by employing a contrastive decoding approach. In this paper, we investigate the correlation between the decay of attention to image tokens and the occurrence of hallucinations. Based on this finding, we propose Temporal Attention Real-time Accumulative Connection (TARAC), a novel training-free method that dynamically accumulates and updates LVLMs' attention on image tokens during generation. By enhancing the model's attention to image tokens, TARAC mitigates hallucinations caused by the decay of attention on image tokens. We validate the effectiveness of TARAC across multiple models and datasets, demonstrating that our approach substantially mitigates hallucinations. In particular, TARAC reduces $C_S$ by 25.2 and $C_I$ by 8.7 compared to VCD on the CHAIR benchmark.

Paperid: 3800, https://arxiv.org/pdf/2504.02244.pdf

Abstract:
Previous research in human gesture recognition has largely overlooked multi-person interactions, which are crucial for understanding the social context of naturally occurring gestures. This limitation in existing datasets presents a significant challenge in aligning human gestures with other modalities like language and speech. To address this issue, we introduce SocialGesture, the first large-scale dataset specifically designed for multi-person gesture analysis. SocialGesture features a diverse range of natural scenarios and supports multiple gesture analysis tasks, including video-based recognition and temporal localization, providing a valuable resource for advancing the study of gesture during complex social interactions. Furthermore, we propose a novel visual question answering (VQA) task to benchmark vision language models'(VLMs) performance on social gesture understanding. Our findings highlight several limitations of current gesture recognition models, offering insights into future directions for improvement in this field. SocialGesture is available at huggingface.co/datasets/IrohXu/SocialGesture.

Paperid: 3801, https://arxiv.org/pdf/2504.00691.pdf

Abstract:
Vision-language (VL) learning requires extensive visual perception capabilities, such as fine-grained object recognition and spatial perception. Recent works typically rely on training huge models on massive datasets to develop these capabilities. As a more efficient alternative, this paper proposes a new framework that Transfers the knowledge from a hub of Vision Experts (ToVE) for efficient VL learning, leveraging pre-trained vision expert models to promote visual perception capability. Specifically, building on a frozen CLIP encoder that provides vision tokens for image-conditioned language generation, ToVE introduces a hub of multiple vision experts and a token-aware gating network that dynamically routes expert knowledge to vision tokens. In the transfer phase, we propose a "residual knowledge transfer" strategy, which not only preserves the generalizability of the vision tokens but also allows detachment of low-contributing experts to improve inference efficiency. Further, we explore to merge these expert knowledge to a single CLIP encoder, creating a knowledge-merged CLIP that produces more informative vision tokens without expert inference during deployment. Experiment results across various VL tasks demonstrate that the proposed ToVE achieves competitive performance with two orders of magnitude fewer training data.

Paperid: 3802, https://arxiv.org/pdf/2503.16535.pdf

Abstract:
Depth estimation is a core problem in robotic perception and vision tasks, but 3D reconstruction from a single image presents inherent uncertainties. Current depth estimation models primarily rely on inter-image relationships for supervised training, often overlooking the intrinsic information provided by the camera itself. We propose a method that embodies the camera model and its physical characteristics into a deep learning model, computing embodied scene depth through real-time interactions with road environments. The model can calculate embodied scene depth in real-time based on immediate environmental changes using only the intrinsic properties of the camera, without any additional equipment. By combining embodied scene depth with RGB image features, the model gains a comprehensive perspective on both geometric and visual details. Additionally, we incorporate text descriptions containing environmental content and depth information as priors for scene understanding, enriching the model's perception of objects. This integration of image and language - two inherently ambiguous modalities - leverages their complementary strengths for monocular depth estimation. The real-time nature of the embodied language and depth prior model ensures that the model can continuously adjust its perception and behavior in dynamic environments. Experimental results show that the embodied depth estimation method enhances model performance across different scenes.

Paperid: 3803, https://arxiv.org/pdf/2503.15886.pdf

Abstract:
In zero-shot image recognition tasks, humans demonstrate remarkable flexibility in classifying unseen categories by composing known simpler concepts. However, existing vision-language models (VLMs), despite achieving significant progress through large-scale natural language supervision, often underperform in real-world applications because of sub-optimal prompt engineering and the inability to adapt effectively to target classes. To address these issues, we propose a Concept-guided Human-like Bayesian Reasoning (CHBR) framework. Grounded in Bayes' theorem, CHBR models the concept used in human image recognition as latent variables and formulates this task by summing across potential concepts, weighted by a prior distribution and a likelihood function. To tackle the intractable computation over an infinite concept space, we introduce an importance sampling algorithm that iteratively prompts large language models (LLMs) to generate discriminative concepts, emphasizing inter-class differences. We further propose three heuristic approaches involving Average Likelihood, Confidence Likelihood, and Test Time Augmentation (TTA) Likelihood, which dynamically refine the combination of concepts based on the test image. Extensive evaluations across fifteen datasets demonstrate that CHBR consistently outperforms existing state-of-the-art zero-shot generalization methods.

Paperid: 3804, https://arxiv.org/pdf/2503.13792.pdf

Abstract:
The evolution of Large Vision-Language Models (LVLMs) has progressed from single to multi-image reasoning. Despite this advancement, our findings indicate that LVLMs struggle to robustly utilize information across multiple images, with predictions significantly affected by the alteration of image positions. To further explore this issue, we introduce Position-wise Question Answering (PQA), a meticulously designed task to quantify reasoning capabilities at each position. Our analysis reveals a pronounced position bias in LVLMs: open-source models excel in reasoning with images positioned later but underperform with those in the middle or at the beginning, while proprietary models show improved comprehension for images at the beginning and end but struggle with those in the middle. Motivated by this, we propose SoFt Attention (SoFA), a simple, training-free approach that mitigates this bias by employing linear interpolation between inter-image causal attention and bidirectional counterparts. Experimental results demonstrate that SoFA reduces position bias and enhances the reasoning performance of existing LVLMs.

Paperid: 3805, https://arxiv.org/pdf/2503.09091.pdf

Abstract:
Foundation models have emerged as a powerful paradigm in computational pathology (CPath), enabling scalable and generalizable analysis of histopathological images. While early developments centered on uni-modal models trained solely on visual data, recent advances have highlighted the promise of multi-modal foundation models that integrate heterogeneous data sources such as textual reports, structured domain knowledge, and molecular profiles. In this survey, we provide a comprehensive and up-to-date review of multi-modal foundation models in CPath, with a particular focus on models built upon hematoxylin and eosin (H&E) stained whole slide images (WSIs) and tile-level representations. We categorize 32 state-of-the-art multi-modal foundation models into three major paradigms: vision-language, vision-knowledge graph, and vision-gene expression. We further divide vision-language models into non-LLM-based and LLM-based approaches. Additionally, we analyze 28 available multi-modal datasets tailored for pathology, grouped into image-text pairs, instruction datasets, and image-other modality pairs. Our survey also presents a taxonomy of downstream tasks, highlights training and evaluation strategies, and identifies key challenges and future directions. We aim for this survey to serve as a valuable resource for researchers and practitioners working at the intersection of pathology and AI.

Paperid: 3806, https://arxiv.org/pdf/2503.08676.pdf

Abstract:
Depth-guided multimodal fusion combines depth information from visible and infrared images, significantly enhancing the performance of 3D reconstruction and robotics applications. Existing thermal-visible image fusion mainly focuses on detection tasks, ignoring other critical information such as depth. By addressing the limitations of single modalities in low-light and complex environments, the depth information from fused images not only generates more accurate point cloud data, improving the completeness and precision of 3D reconstruction, but also provides comprehensive scene understanding for robot navigation, localization, and environmental perception. This supports precise recognition and efficient operations in applications such as autonomous driving and rescue missions. We introduce a text-guided and depth-driven infrared and visible image fusion network. The model consists of an image fusion branch for extracting multi-channel complementary information through a diffusion model, equipped with a text-guided module, and two auxiliary depth estimation branches. The fusion branch uses CLIP to extract semantic information and parameters from depth-enriched image descriptions to guide the diffusion model in extracting multi-channel features and generating fused images. These fused images are then input into the depth estimation branches to calculate depth-driven loss, optimizing the image fusion network. This framework aims to integrate vision-language and depth to directly generate color-fused images from multimodal inputs.

Paperid: 3807, https://arxiv.org/pdf/2502.16842.pdf

Abstract:
Large Vision-Language Models (LVLMs) integrate image encoders with Large Language Models (LLMs) to process multi-modal inputs and perform complex visual tasks. However, they often generate hallucinations by describing non-existent objects or attributes, compromising their reliability. This study analyzes hallucination patterns in image captioning, showing that not all tokens in the generation process are influenced by image input and that image dependency can serve as a useful signal for hallucination detection. To address this, we develop an automated pipeline to identify hallucinated objects and train a token-level classifier using hidden representations from parallel inference passes-with and without image input. Leveraging this classifier, we introduce a decoding strategy that effectively controls hallucination rates in image captioning at inference time.

Paperid: 3808, https://arxiv.org/pdf/2502.15809.pdf

Abstract:
Few-shot adaptation for Vision-Language Models (VLMs) presents a dilemma: balancing in-distribution accuracy with out-of-distribution generalization. Recent research has utilized low-level concepts such as visual attributes to enhance generalization. However, this study reveals that VLMs overly rely on a small subset of attributes on decision-making, which co-occur with the category but are not inherently part of it, termed spuriously correlated attributes. This biased nature of VLMs results in poor generalization. To address this, 1) we first propose Spurious Attribute Probing (SAP), identifying and filtering out these problematic attributes to significantly enhance the generalization of existing attribute-based methods; 2) We introduce Spurious Attribute Shielding (SAS), a plug-and-play module that mitigates the influence of these attributes on prediction, seamlessly integrating into various Parameter-Efficient Fine-Tuning (PEFT) methods. In experiments, SAP and SAS significantly enhance accuracy on distribution shifts across 11 datasets and 3 generalization tasks without compromising downstream performance, establishing a new state-of-the-art benchmark.

Paperid: 3809, https://arxiv.org/pdf/2502.09906.pdf

Abstract:
Multimodal conversational generative AI has shown impressive capabilities in various vision and language understanding through learning massive text-image data. However, current conversational models still lack knowledge about visual insects since they are often trained on the general knowledge of vision-language data. Meanwhile, understanding insects is a fundamental problem in precision agriculture, helping to promote sustainable development in agriculture. Therefore, this paper proposes a novel multimodal conversational model, Insect-LLaVA, to promote visual understanding in insect-domain knowledge. In particular, we first introduce a new large-scale Multimodal Insect Dataset with Visual Insect Instruction Data that enables the capability of learning the multimodal foundation models. Our proposed dataset enables conversational models to comprehend the visual and semantic features of the insects. Second, we propose a new Insect-LLaVA model, a new general Large Language and Vision Assistant in Visual Insect Understanding. Then, to enhance the capability of learning insect features, we develop an Insect Foundation Model by introducing a new micro-feature self-supervised learning with a Patch-wise Relevant Attention mechanism to capture the subtle differences among insect images. We also present Description Consistency loss to improve micro-feature learning via text descriptions. The experimental results evaluated on our new Visual Insect Question Answering benchmarks illustrate the effective performance of our proposed approach in visual insect understanding and achieve State-of-the-Art performance on standard benchmarks of insect-related tasks.

Paperid: 3810, https://arxiv.org/pdf/2502.09507.pdf

Abstract:
The remarkable generalization performance of contrastive vision-language models like CLIP is often attributed to the diversity of their training distributions. However, key questions remain unanswered: Can CLIP generalize to an entirely unseen domain when trained on a diverse mixture of domains (domain generalization)? Can it generalize to unseen classes within partially seen domains (compositional generalization)? What factors affect such generalization? To answer these questions, we trained CLIP models on systematically constructed training distributions with controlled domain diversity and object class exposure. Our experiments show that domain diversity is essential for both domain and compositional generalization, yet compositional generalization can be surprisingly weaker than domain generalization when the training distribution contains a suboptimal subset of the test domain. Through data-centric and mechanistic analyses, we find that successful generalization requires the learning of sufficiently shared representations in intermediate layers and circuits.

Paperid: 3811, https://arxiv.org/pdf/2502.04873.pdf

Abstract:
This paper presents a training-free pipeline for task-oriented grasp generation that combines pre-trained grasp generation models with vision-language models (VLMs). Unlike traditional approaches that focus solely on stable grasps, our method incorporates task-specific requirements by leveraging the semantic reasoning capabilities of VLMs. We evaluate five querying strategies, each utilizing different visual representations of candidate grasps, and demonstrate significant improvements over a baseline method in both grasp success and task compliance rates, with absolute gains of up to 36.9\% in overall success rate. Our results underline the potential of VLMs to enhance task-oriented manipulation, providing insights for future research in robotic grasping and human-robot interaction.

Paperid: 3812, https://arxiv.org/pdf/2502.00426.pdf

Abstract:
Recently, adapting Vision Language Models (VLMs) to zero-shot visual classification by tuning class embedding with a few prompts (Test-time Prompt Tuning, TPT) or replacing class names with generated visual samples (support-set) has shown promising results. However, TPT cannot avoid the semantic gap between modalities while the support-set cannot be tuned. To this end, we draw on each other's strengths and propose a novel framework namely TEst-time Support-set Tuning for zero-shot Video Classification (TEST-V). It first dilates the support-set with multiple prompts (Multi-prompting Support-set Dilation, MSD) and then erodes the support-set via learnable weights to mine key cues dynamically (Temporal-aware Support-set Erosion, TSE). Specifically, i) MSD expands the support samples for each class based on multiple prompts enquired from LLMs to enrich the diversity of the support-set. ii) TSE tunes the support-set with factorized learnable weights according to the temporal prediction consistency in a self-supervised manner to dig pivotal supporting cues for each class. $\textbf{TEST-V}$ achieves state-of-the-art results across four benchmarks and has good interpretability for the support-set dilation and erosion.

Paperid: 3813, https://arxiv.org/pdf/2501.17391.pdf

Abstract:
Vision-Language Models (VLMs) have achieved remarkable success across a range of multimodal tasks; however, their practical deployment is often constrained by high computational costs and prolonged inference times. Since the vision modality typically carries more information than the text modality, compressing visual prompts offers a promising solution to alleviate these challenges. Existing approaches predominantly focus on refining model architectures or directly reducing the number of visual tokens. However, these methods often compromise inference performance due to a lack of consideration for the unique spatial and temporal characteristics of visual data. In this work, we propose a token compression paradigm that operates on both spatial and temporal dimensions. Our approach includes a learning-free, plug-and-play compression pipeline that can be seamlessly integrated into most Multimodal Large Language Model (MLLM) frameworks. By leveraging this method, we enhance the model inference capability while simultaneously reducing its computational cost. Experimental results on the Video-QA task demonstrate the effectiveness of the proposed approach, showcasing significant improvements in efficiency without sacrificing performance.

Paperid: 3814, https://arxiv.org/pdf/2501.09425.pdf

Abstract:
Many practical vision-language applications require models that understand negation, e.g., when using natural language to retrieve images which contain certain objects but not others. Despite advancements in vision-language models (VLMs) through large-scale training, their ability to comprehend negation remains underexplored. This study addresses the question: how well do current VLMs understand negation? We introduce NegBench, a new benchmark designed to evaluate negation understanding across 18 task variations and $79$k examples spanning image, video, and medical datasets. The benchmark consists of two core tasks designed to evaluate negation understanding in diverse multimodal settings: Retrieval with Negation and Multiple Choice Questions with Negated Captions. Our evaluation reveals that modern VLMs struggle significantly with negation, often performing at chance level. To address these shortcomings, we explore a data-centric approach wherein we finetune CLIP models on large-scale synthetic datasets containing millions of negated captions. We show that this approach can result in a 10% increase in recall on negated queries and a 28% boost in accuracy on multiple-choice questions with negated captions.

Paperid: 3815, https://arxiv.org/pdf/2501.03565.pdf

Abstract:
3D medical images such as Computed tomography (CT) are widely used in clinical practice, offering a great potential for automatic diagnosis. Supervised learning-based approaches have achieved significant progress but rely heavily on extensive manual annotations, limited by the availability of training data and the diversity of abnormality types. Vision-language alignment (VLA) offers a promising alternative by enabling zero-shot learning without additional annotations. However, we empirically discover that the visual and textural embeddings after alignment endeavors from existing VLA methods form two well-separated clusters, presenting a wide gap to be bridged. To bridge this gap, we propose a Bridged Semantic Alignment (BrgSA) framework. First, we utilize a large language model to perform semantic summarization of reports, extracting high-level semantic information. Second, we design a Cross-Modal Knowledge Interaction (CMKI) module that leverages a cross-modal knowledge bank as a semantic bridge, facilitating interaction between the two modalities, narrowing the gap, and improving their alignment. To comprehensively evaluate our method, we construct a benchmark dataset that includes 15 underrepresented abnormalities as well as utilize two existing benchmark datasets. Experimental results demonstrate that BrgSA achieves state-of-the-art performances on both public benchmark datasets and our custom-labeled dataset, with significant improvements in zero-shot diagnosis of underrepresented abnormalities.

Paperid: 3816, https://arxiv.org/pdf/2412.06774.pdf

Abstract:
We present Visual Lexicon, a novel visual language that encodes rich image information into the text space of vocabulary tokens while retaining intricate visual details that are often challenging to convey in natural language. Unlike traditional methods that prioritize either high-level semantics (e.g., CLIP) or pixel-level reconstruction (e.g., VAE), ViLex simultaneously captures rich semantic content and fine visual details, enabling high-quality image generation and comprehensive visual scene understanding. Through a self-supervised learning pipeline, ViLex generates tokens optimized for reconstructing input images using a frozen text-to-image (T2I) diffusion model, preserving the detailed information necessary for high-fidelity semantic-level reconstruction. As an image embedding in the language space, ViLex tokens leverage the compositionality of natural languages, allowing them to be used independently as "text tokens" or combined with natural language tokens to prompt pretrained T2I models with both visual and textual inputs, mirroring how we interact with vision-language models (VLMs). Experiments demonstrate that ViLex achieves higher fidelity in image reconstruction compared to text embeddings--even with a single ViLex token. Moreover, ViLex successfully performs various DreamBooth tasks in a zero-shot, unsupervised manner without fine-tuning T2I models. Additionally, ViLex serves as a powerful vision encoder, consistently improving vision-language model performance across 15 benchmarks relative to a strong SigLIP baseline.

Paperid: 3817, https://arxiv.org/pdf/2412.02946.pdf

Abstract:
Recent advancements in large vision-language models (LVLM) have significantly enhanced their ability to comprehend visual inputs alongside natural language. However, a major challenge in their real-world application is hallucination, where LVLMs generate non-existent visual elements, eroding user trust. The underlying mechanism driving this multimodal hallucination is poorly understood. Minimal research has illuminated whether contexts such as sky, tree, or grass field involve the LVLM in hallucinating a frisbee. We hypothesize that hidden factors, such as objects, contexts, and semantic foreground-background structures, induce hallucination. This study proposes a novel causal approach: a hallucination probing system to identify these hidden factors. By analyzing the causality between images, text prompts, and network saliency, we systematically explore interventions to block these factors. Our experimental findings show that a straightforward technique based on our analysis can significantly reduce hallucinations. Additionally, our analyses indicate the potential to edit network internals to minimize hallucinated outputs.

Paperid: 3818, https://arxiv.org/pdf/2411.17265.pdf

Abstract:
The success of Direct Preference Optimization (DPO) in mitigating hallucinations in Vision Language Models (VLMs) critically hinges on the true reward gaps within preference pairs. However, current methods, typically relying on ranking or rewriting strategies, often struggle to optimize these reward gaps in a systematic way during data curation. A core difficulty lies in precisely characterizing and strategically manipulating the overall reward gap configuration, that is, the deliberate design of how to shape these reward gaps within each preference pair across the data. To address this, we introduce Topic-level Preference Rewriting(TPR), a novel framework designed for the systematic optimization of reward gap configuration. Through selectively replacing semantic topics within VLM responses with model's own resampled candidates for targeted rewriting, TPR can provide topic-level control over fine-grained semantic details. This precise control enables advanced data curation strategies, such as progressively adjusting the difficulty of rejected responses, thereby sculpting an effective reward gap configuration that guides the model to overcome challenging hallucinations. Comprehensive experiments demonstrate TPR achieves state-of-the-art performance on multiple hallucination benchmarks, outperforming previous methods by an average of 20%. Notably, it significantly reduces hallucinations by up to 93% on ObjectHal-Bench, and also exhibits superior data efficiency towards robust and cost-effective VLM alignment.

Paperid: 3819, https://arxiv.org/pdf/2411.16537.pdf

Abstract:
Spatial understanding is a crucial capability that enables robots to perceive their surroundings, reason about their environment, and interact with it meaningfully. In modern robotics, these capabilities are increasingly provided by vision-language models. However, these models face significant challenges in spatial reasoning tasks, as their training data are based on general-purpose image datasets that often lack sophisticated spatial understanding. For example, datasets frequently do not capture reference frame comprehension, yet effective spatial reasoning requires understanding whether to reason from ego-, world-, or object-centric perspectives. To address this issue, we introduce RoboSpatial, a large-scale dataset for spatial understanding in robotics. It consists of real indoor and tabletop scenes, captured as 3D scans and egocentric images, and annotated with rich spatial information relevant to robotics. The dataset includes 1M images, 5k 3D scans, and 3M annotated spatial relationships, and the pairing of 2D egocentric images with 3D scans makes it both 2D- and 3D- ready. Our experiments show that models trained with RoboSpatial outperform baselines on downstream tasks such as spatial affordance prediction, spatial relationship prediction, and robot manipulation.

Paperid: 3820, https://arxiv.org/pdf/2411.15231.pdf

Abstract:
Low-rank adaptations (LoRA) are widely used to fine-tune large models across various domains for specific downstream tasks. While task-specific LoRAs are often available, concerns about data privacy and intellectual property can restrict access to training data, limiting the acquisition of a multi-task model through gradient-based training. In response, LoRA merging presents an effective solution by combining multiple LoRAs into a unified adapter while maintaining data privacy. Prior works on LoRA merging primarily frame it as an optimization problem, yet these approaches face several limitations, including the rough assumption about input features utilized in optimization, massive sample requirements, and the unbalanced optimization objective. These limitations can significantly degrade performance. To address these, we propose a novel optimization-based method, named IterIS: 1) We formulate LoRA merging as an advanced optimization problem to mitigate the rough assumption. Additionally, we employ an iterative inference-solving framework in our algorithm. It can progressively refine the optimization objective for improved performance. 2) We introduce an efficient regularization term to reduce the need for massive sample requirements (requiring only 1-5% of the unlabeled samples compared to prior methods). 3) We utilize adaptive weights in the optimization objective to mitigate potential unbalances in LoRA merging process. Our method demonstrates significant improvements over multiple baselines and state-of-the-art methods in composing tasks for text-to-image diffusion, vision-language models, and large language models. Furthermore, our layer-wise algorithm can achieve convergence with minimal steps, ensuring efficiency in both memory and computation.

Paperid: 3821, https://arxiv.org/pdf/2411.14164.pdf

Abstract:
Large Vision-Language Models (LVLMs) represent a significant advancement toward achieving superior multimodal capabilities by enabling powerful Large Language Models (LLMs) to understand visual input. Typically, LVLMs utilize visual encoders, such as CLIP, to transform images into visual tokens, which are then aligned with textual tokens through projection layers before being input into the LLM for inference. Although existing LVLMs have achieved significant success, their inference efficiency is still limited by the substantial number of visual tokens and the potential redundancy among them. To mitigate this issue, we propose Focal Pruning (FoPru), a training-free method that prunes visual tokens based on the attention-based token significance derived from the vision encoder. Specifically, we introduce two alternative pruning strategies: 1) the rank strategy, which leverages all token significance scores to retain more critical tokens in a global view; 2) the row strategy, which focuses on preserving continuous key information in images from a local perspective. Finally, the selected tokens are reordered to maintain their original positional relationships. Extensive experiments across various LVLMs and multimodal datasets demonstrate that our method can prune a large number of redundant tokens while maintaining high accuracy, leading to significant improvements in inference efficiency.

Paperid: 3822, https://arxiv.org/pdf/2411.09018.pdf

Abstract:
Recent research increasingly focuses on training vision-language models (VLMs) with long, detailed image captions. However, small-scale VLMs often struggle to balance the richness of these captions with the risk of hallucinating content during fine-tuning. In this paper, we explore how well VLMs adapt to such captions. To quantify caption quality, we propose Decomposed NLI (DNLI), an evaluation framework that breaks down generated captions into individual propositions, assessing each in isolation. This fine-grained analysis reveals a critical balance between capturing descriptive details and preventing hallucinations. Our findings show that simply reducing caption complexity or employing standard data curation techniques does not effectively resolve this issue. To tackle this challenge, we introduce Knowledge Adapted (KnowAda) fine-tuning, a data-centric approach that automatically adapts training data with the model's existing knowledge and visual understanding. KnowAda minimizes hallucinations while preserving high descriptiveness. We validate this approach across several small-scale VLMs (up to 7B parameters) and dense caption datasets, demonstrating that KnowAda effectively balances hallucination reduction and descriptiveness. Our results show that KnowAda outperforms various baselines in both automatic metrics and human evaluations. We will release our code and models.

Paperid: 3823, https://arxiv.org/pdf/2411.01756.pdf

Abstract:
Visual object tracking aims to locate a targeted object in a video sequence based on an initial bounding box. Recently, Vision-Language~(VL) trackers have proposed to utilize additional natural language descriptions to enhance versatility in various applications. However, VL trackers are still inferior to State-of-The-Art (SoTA) visual trackers in terms of tracking performance. We found that this inferiority primarily results from their heavy reliance on manual textual annotations, which include the frequent provision of ambiguous language descriptions. In this paper, we propose ChatTracker to leverage the wealth of world knowledge in the Multimodal Large Language Model (MLLM) to generate high-quality language descriptions and enhance tracking performance. To this end, we propose a novel reflection-based prompt optimization module to iteratively refine the ambiguous and inaccurate descriptions of the target with tracking feedback. To further utilize semantic information produced by MLLM, a simple yet effective VL tracking framework is proposed and can be easily integrated as a plug-and-play module to boost the performance of both VL and visual trackers. Experimental results show that our proposed ChatTracker achieves a performance comparable to existing methods.

Paperid: 3824, https://arxiv.org/pdf/2410.18558.pdf

Abstract:
Recently, Vision-Language Models (VLMs) have achieved remarkable progress in multimodal tasks, and multimodal instruction data serves as the foundation for enhancing VLM capabilities. Despite the availability of several open-source multimodal datasets, limitations in the scale and quality of open-source instruction data hinder the performance of VLMs trained on these datasets, leading to a significant gap compared to models trained on closed-source data. To address this challenge, we introduce Infinity-MM, a large-scale multimodal instruction dataset. We collected the available multimodal instruction datasets and performed unified preprocessing, resulting in a dataset with over 40 million samples that ensures diversity and accuracy. Furthermore, to enable large-scale expansion of instruction data and support the continuous acquisition of high-quality data, we propose a synthetic instruction generation method based on a tagging system and open-source VLMs. By establishing correspondences between different types of images and associated instruction types, this method can provide essential guidance during data synthesis. Leveraging this high-quality data, we have trained a 2-billion-parameter Vision-Language Model, Aquila-VL-2B, which achieves state-of-the-art (SOTA) performance among models of similar scale. The data is available at: https://huggingface.co/datasets/BAAI/Infinity-MM.

Paperid: 3825, https://arxiv.org/pdf/2410.17809.pdf

Abstract:
Real-world image restoration (IR) is inherently complex and often requires combining multiple specialized models to address diverse degradations. Inspired by human problem-solving, we propose AgenticIR, an agentic system that mimics the human approach to image processing by following five key stages: Perception, Scheduling, Execution, Reflection, and Rescheduling. AgenticIR leverages large language models (LLMs) and vision-language models (VLMs) that interact via text generation to dynamically operate a toolbox of IR models. We fine-tune VLMs for image quality analysis and employ LLMs for reasoning, guiding the system step by step. To compensate for LLMs' lack of specific IR knowledge and experience, we introduce a self-exploration method, allowing the LLM to observe and summarize restoration results into referenceable documents. Experiments demonstrate AgenticIR's potential in handling complex IR tasks, representing a promising path toward achieving general intelligence in visual processing.

Paperid: 3826, https://arxiv.org/pdf/2410.15536.pdf

Abstract:
We introduce GRS (Generating Robotic Simulation tasks), a system addressing real-to-sim for robotic simulations. GRS creates digital twin simulations from single RGB-D observations with solvable tasks for virtual agent training. Using vision-language models (VLMs), our pipeline operates in three stages: 1) scene comprehension with SAM2 for segmentation and object description, 2) matching objects with simulation-ready assets, and 3) generating appropriate tasks. We ensure simulation-task alignment through generated test suites and introduce a router that iteratively refines both simulation and test code. Experiments demonstrate our system's effectiveness in object correspondence and task environment generation through our novel router mechanism.

Paperid: 3827, https://arxiv.org/pdf/2410.14200.pdf

Abstract:
The development of 3D medical vision-language models holds significant potential for disease diagnosis and patient treatment. However, compared to 2D medical images, 3D medical images, such as CT scans, face challenges related to limited training data and high dimension, which severely restrict the progress of 3D medical vision-language models. To address these issues, we collect a large amount of unlabeled 3D CT data and utilize self-supervised learning to construct a 3D visual foundation model for extracting 3D visual features. Then, we apply 3D spatial convolutions to aggregate and project high-level image features, reducing computational complexity while preserving spatial information. We also construct two instruction-tuning datasets based on BIMCV-R and CT-RATE to fine-tune the 3D vision-language model. Our model demonstrates superior performance compared to existing methods in report generation, visual question answering, and disease diagnosis. Code and data will be made publicly available soon.

Paperid: 3828, https://arxiv.org/pdf/2410.13321.pdf

Abstract:
Large Vision-Language Models (LVLMs) demonstrate impressive capabilities in generating detailed and coherent responses from visual inputs. However, they are prone to generate hallucinations due to an over-reliance on language priors. To address this issue, we investigate the language priors in LVLMs and make two key observations: (1) Even when predicting the tokens associated with image-related part-of-speech (POS), models increasingly rely on linguistic priors as the token sequences grow, thereby amplifying hallucinations. (2) Methods that directly calibrate LVLM's output distribution to mitigate language priors can lead to a degradation in text quality or even exacerbate hallucinations. Based on these findings, we propose a novel method, Summary-Guided Decoding (SumGD). This method naturally encourages the model to focus more on image information by reducing the text context through summaries, while controlling only the image-related POS tokens to maintain text quality. Through experiments, we demonstrate that SumGD achieves state-of-the-art performance on object hallucination benchmarks. Furthermore, in terms of the trade-off between precision and recall, SumGD achieves Pareto optimality among the existing methods. Lastly, we observe that although existing methods struggle to balance the reduction of object hallucinations with maintaining text quality, SumGD demonstrates robustness in handling this challenge.

Paperid: 3829, https://arxiv.org/pdf/2410.13002.pdf

Abstract:
End-to-end learning directly maps sensory inputs to actions, creating highly integrated and efficient policies for complex robotics tasks. However, such models often struggle to generalize beyond their training scenarios, limiting adaptability to new environments, tasks, and concepts. In this work, we investigate the minimal data requirements and architectural adaptations necessary to achieve robust closed-loop performance with vision-based control policies under unseen text instructions and visual distribution shifts. Our findings are synthesized in Flex (Fly lexically), a framework that uses pre-trained Vision Language Models (VLMs) as frozen patch-wise feature extractors, generating spatially aware embeddings that integrate semantic and visual information. We demonstrate the effectiveness of this approach on a quadrotor fly-to-target task, where agents trained via behavior cloning on a small simulated dataset successfully generalize to real-world scenes with diverse novel goals and command formulations.

Paperid: 3830, https://arxiv.org/pdf/2410.06699.pdf

Abstract:
Large vision-language models (LVLMs) integrate visual information into large language models, showcasing remarkable multi-modal conversational capabilities. However, the visual modules introduces new challenges in terms of robustness for LVLMs, as attackers can craft adversarial images that are visually clean but may mislead the model to generate incorrect answers. In general, LVLMs rely on vision encoders to transform images into visual tokens, which are crucial for the language models to perceive image contents effectively. Therefore, we are curious about one question: Can LVLMs still generate correct responses when the encoded visual tokens are attacked and disrupting the visual information? To this end, we propose a non-targeted attack method referred to as VT-Attack (Visual Tokens Attack), which constructs adversarial examples from multiple perspectives, with the goal of comprehensively disrupting feature representations and inherent relationships as well as the semantic properties of visual tokens output by image encoders. Using only access to the image encoder in the proposed attack, the generated adversarial examples exhibit transferability across diverse LVLMs utilizing the same image encoder and generality across different tasks. Extensive experiments validate the superior attack performance of the VT-Attack over baseline methods, demonstrating its effectiveness in attacking LVLMs with image encoders, which in turn can provide guidance on the robustness of LVLMs, particularly in terms of the stability of the visual feature space.

Paperid: 3831, https://arxiv.org/pdf/2410.04634.pdf

Abstract:
Text-to-image (T2I) models are increasingly used in impactful real-life applications. As such, there is a growing need to audit these models to ensure that they generate desirable, task-appropriate images. However, systematically inspecting the associations between prompts and generated content in a human-understandable way remains challenging. To address this, we propose Concept2Concept, a framework where we characterize conditional distributions of vision language models using interpretable concepts and metrics that can be defined in terms of these concepts. This characterization allows us to use our framework to audit models and prompt-datasets. To demonstrate, we investigate several case studies of conditional distributions of prompts, such as user-defined distributions or empirical, real-world distributions. Lastly, we implement Concept2Concept as an open-source interactive visualization tool to facilitate use by non-technical end-users. A demo is available at https://tinyurl.com/Concept2ConceptDemo.

Paperid: 3832, https://arxiv.org/pdf/2409.08202.pdf

Abstract:
A unique aspect of human visual understanding is the ability to flexibly interpret abstract concepts: acquiring lifted rules explaining what they symbolize, grounding them across familiar and unfamiliar contexts, and making predictions or reasoning about them. While off-the-shelf vision-language models excel at making literal interpretations of images (e.g., recognizing object categories such as tree branches), they still struggle to make sense of such visual abstractions (e.g., how an arrangement of tree branches may form the walls of a maze). To address this challenge, we introduce Deep Schema Grounding (DSG), a framework that leverages explicit structured representations of visual abstractions for grounding and reasoning. At the core of DSG are schemas--dependency graph descriptions of abstract concepts that decompose them into more primitive-level symbols. DSG uses large language models to extract schemas, then hierarchically grounds concrete to abstract components of the schema onto images with vision-language models. The grounded schema is used to augment visual abstraction understanding. We systematically evaluate DSG and different methods in reasoning on our new Visual Abstractions Dataset, which consists of diverse, real-world images of abstract concepts and corresponding question-answer pairs labeled by humans. We show that DSG significantly improves the abstract visual reasoning performance of vision-language models, and is a step toward human-aligned understanding of visual abstractions.

Paperid: 3833, https://arxiv.org/pdf/2409.07463.pdf

Abstract:
We present a novel framework for analyzing and interpreting electron microscopy images in semiconductor manufacturing using vision-language instruction tuning. The framework employs a unique teacher-student approach, leveraging pre-trained multimodal large language models such as GPT-4 to generate instruction-following data for zero-shot visual question answering (VQA) and classification tasks, customizing smaller multimodal models (SMMs) for microscopy image analysis, resulting in an instruction-tuned language-and-vision assistant. Our framework merges knowledge engineering with machine learning to integrate domain-specific expertise from larger to smaller multimodal models within this specialized field, greatly reducing the need for extensive human labeling. Our study presents a secure, cost-effective, and customizable approach for analyzing microscopy images, addressing the challenges of adopting proprietary models in semiconductor manufacturing.

Paperid: 3834, https://arxiv.org/pdf/2409.03966.pdf

Abstract:
Current robot autonomy struggles to operate beyond the assumed Operational Design Domain (ODD), the specific set of conditions and environments in which the system is designed to function, while the real-world is rife with uncertainties that may lead to failures. Automating recovery remains a significant challenge. Traditional methods often rely on human intervention to manually address failures or require exhaustive enumeration of failure cases and the design of specific recovery policies for each scenario, both of which are labor-intensive. Foundational Vision-Language Models (VLMs), which demonstrate remarkable common-sense generalization and reasoning capabilities, have broader, potentially unbounded ODDs. However, limitations in spatial reasoning continue to be a common challenge for many VLMs when applied to robot control and motion-level error recovery. In this paper, we investigate how optimizing visual and text prompts can enhance the spatial reasoning of VLMs, enabling them to function effectively as black-box controllers for both motion-level position correction and task-level recovery from unknown failures. Specifically, the optimizations include identifying key visual elements in visual prompts, highlighting these elements in text prompts for querying, and decomposing the reasoning process for failure detection and control generation. In experiments, prompt optimizations significantly outperform pre-trained Vision-Language-Action Models in correcting motion-level position errors and improve accuracy by 65.78% compared to VLMs with unoptimized prompts. Additionally, for task-level failures, optimized prompts enhanced the success rate by 5.8%, 5.8%, and 7.5% in VLMs' abilities to detect failures, analyze issues, and generate recovery plans, respectively, across a wide range of unknown errors in Lego assembly.

Paperid: 3835, https://arxiv.org/pdf/2601.12539.pdf

Abstract:
Memes are a dominant medium for online communication and manipulation because meaning emerges from interactions between embedded text, imagery, and cultural context. Existing meme research is distributed across tasks (hate, misogyny, propaganda, sentiment, humour) and languages, which limits cross-domain generalization. To address this gap we propose MemeLens, a unified multilingual and multitask explanation-enhanced Vision Language Model (VLM) for meme understanding. We consolidate 38 public meme datasets, filter and map dataset-specific labels into a shared taxonomy of $20$ tasks spanning harm, targets, figurative/pragmatic intent, and affect. We present a comprehensive empirical analysis across modeling paradigms, task categories, and datasets. Our findings suggest that robust meme understanding requires multimodal training, exhibits substantial variation across semantic categories, and remains sensitive to over-specialization when models are fine-tuned on individual datasets rather than trained in a unified setting. We will make the experimental resources and datasets publicly available for the community.

Paperid: 3836, https://arxiv.org/pdf/2601.12481.pdf

Abstract:
Reconstructing realistic animal fur geometry from images is a challenging task due to the fine-scale details, self-occlusion, and view-dependent appearance of fur. In contrast to human hairstyle reconstruction, there are also no datasets that can be leveraged to learn a fur prior for different animals. In this work, we present a first multi-view-based method for high-fidelity 3D fur modeling of animals using a strand-based representation, leveraging the general knowledge of a vision language model. Given multi-view RGB images, we first reconstruct a coarse surface geometry using traditional multi-view stereo techniques. We then use a vision language model (VLM) system to retrieve information about the realistic length structure of the fur for each part of the body. We use this knowledge to construct the animal's furless geometry and grow strands atop it. The fur reconstruction is supervised with both geometric and photometric losses computed from multi-view images. To mitigate orientation ambiguities stemming from the Gabor filters that are applied to the input images, we additionally utilize the VLM to guide the strands' growth direction and their relation to the gravity vector that we incorporate as a loss. With this new schema of using a VLM to guide 3D reconstruction from multi-view inputs, we show generalization across a variety of animals with different fur types. For additional results and code, please refer to https://neuralfur.is.tue.mpg.de.

Paperid: 3837, https://arxiv.org/pdf/2601.12282.pdf

Abstract:
The functions of different regions of the human brain are closely linked to their distinct cytoarchitecture, which is defined by the spatial arrangement and morphology of the cells. Identifying brain regions by their cytoarchitecture enables various scientific analyses of the brain. However, delineating these areas manually in brain histological sections is time-consuming and requires specialized knowledge. An automated approach is necessary to minimize the effort needed from human experts. To address this, we propose CytoCLIP, a suite of vision-language models derived from pre-trained Contrastive Language-Image Pre-Training (CLIP) frameworks to learn joint visual-text representations of brain cytoarchitecture. CytoCLIP comprises two model variants: one is trained using low-resolution whole-region images to understand the overall cytoarchitectural pattern of an area, and the other is trained on high-resolution image tiles for detailed cellular-level representation. The training dataset is created from NISSL-stained histological sections of developing fetal brains of different gestational weeks. It includes 86 distinct regions for low-resolution images and 384 brain regions for high-resolution tiles. We evaluate the model's understanding of the cytoarchitecture and generalization ability using region classification and cross-modal retrieval tasks. Multiple experiments are performed under various data setups, including data from samples of different ages and sectioning planes. Experimental results demonstrate that CytoCLIP outperforms existing methods. It achieves an F1 score of 0.87 for whole-region classification and 0.91 for high-resolution image tile classification.

Paperid: 3838, https://arxiv.org/pdf/2601.09228.pdf

Abstract:
Infrared object detection focuses on identifying and locating objects in complex environments (\eg, dark, snow, and rain) where visible imaging cameras are disabled by poor illumination. However, due to low contrast and weak edge information in infrared images, it is challenging to extract discriminative object features for robust detection. To deal with this issue, we propose a novel vision-language representation learning paradigm for infrared object detection. An additional textual supervision with rich semantic information is explored to guide the disentanglement of object and non-object features. Specifically, we propose a Semantic Feature Alignment (SFA) module to align the object features with the corresponding text features. Furthermore, we develop an Object Feature Disentanglement (OFD) module that disentangles text-aligned object features and non-object features by minimizing their correlation. Finally, the disentangled object features are entered into the detection head. In this manner, the detection performance can be remarkably enhanced via more discriminative and less noisy features. Extensive experimental results demonstrate that our approach achieves superior performance on two benchmarks: M\textsuperscript{3}FD (83.7\% mAP), FLIR (86.1\% mAP). Our code will be publicly available once the paper is accepted.

Paperid: 3839, https://arxiv.org/pdf/2601.08476.pdf

Abstract:
Reliable zero-shot detection of out-of-distribution (OOD) inputs is critical for deploying vision-language models in open-world settings. However, the lack of labeled negatives in zero-shot OOD detection necessitates proxy signals that remain effective under distribution shift. Existing negative-label methods rely on a fixed set of textual proxies, which (i) sparsely sample the semantic space beyond in-distribution (ID) classes and (ii) remain static while only visual features drift, leading to cross-modal misalignment and unstable predictions. In this paper, we propose CoEvo, a training- and annotation-free test-time framework that performs bidirectional, sample-conditioned adaptation of both textual and visual proxies. Specifically, CoEvo introduces a proxy-aligned co-evolution mechanism to maintain two evolving proxy caches, which dynamically mines contextual textual negatives guided by test images and iteratively refines visual proxies, progressively realigning cross-modal similarities and enlarging local OOD margins. Finally, we dynamically re-weight the contributions of dual-modal proxies to obtain a calibrated OOD score that is robust to distribution shift. Extensive experiments on standard benchmarks demonstrate that CoEvo achieves state-of-the-art performance, improving AUROC by 1.33% and reducing FPR95 by 45.98% on ImageNet-1K compared to strong negative-label baselines.

Paperid: 3840, https://arxiv.org/pdf/2601.06574.pdf

Abstract:
Multi-objective alignment for text-to-image generation is commonly implemented via static linear scalarization, but fixed weights often fail under heterogeneous rewards, leading to optimization imbalance where models overfit high-variance, high-responsiveness objectives (e.g., OCR) while under-optimizing perceptual goals. We identify two mechanistic causes: variance hijacking, where reward dispersion induces implicit reweighting that dominates the normalized training signal, and gradient conflicts, where competing objectives produce opposing update directions and trigger seesaw-like oscillations. We propose APEX (Adaptive Priority-based Efficient X-objective Alignment), which stabilizes heterogeneous rewards with Dual-Stage Adaptive Normalization and dynamically schedules objectives via P^3 Adaptive Priorities that combine learning potential, conflict penalty, and progress need. On Stable Diffusion 3.5, APEX achieves improved Pareto trade-offs across four heterogeneous objectives, with balanced gains of +1.31 PickScore, +0.35 DeQA, and +0.53 Aesthetics while maintaining competitive OCR accuracy, mitigating the instability of multi-objective alignment.

Paperid: 3841, https://arxiv.org/pdf/2601.03011.pdf

Abstract:
Corner cases are rare or extreme scenarios that drive real-world failures, but they are difficult to curate at scale: web data are noisy, labels are brittle, and edge deployments preclude large retraining. We present ReCCur (Recursive Corner-Case Curation), a low-compute framework that converts noisy web imagery into auditable fine-grained labels via a multi-agent recursive pipeline. First, large-scale data acquisition and filtering expands a domain vocabulary with a vision-language model (VLM), crawls the web, and enforces tri-modal (image, description, keyword) consistency with light human spot checks to yield refined candidates. Next, mixture-of-experts knowledge distillation uses complementary encoders (e.g., CLIP, DINOv2, BEiT) for kNN voting with dual-confidence activation and uncertainty sampling, converging to a high-precision set. Finally, region-evidence VLM adversarial labeling pairs a proposer (multi-granularity regions and semantic cues) with a validator (global and local chained consistency) to produce explainable labels and close the loop. On realistic corner-case scenarios (e.g., flooded-car inspection), ReCCur runs on consumer-grade GPUs, steadily improves purity and separability, and requires minimal human supervision, providing a practical substrate for downstream training and evaluation under resource constraints. Code and dataset will be released.

Paperid: 3842, https://arxiv.org/pdf/2512.23244.pdf

Abstract:
Remote sensing change detection (RSCD), a complex multi-image inference task, traditionally uses pixel-based operators or encoder-decoder networks that inadequately capture high-level semantics and are vulnerable to non-semantic perturbations. Although recent multimodal and vision-language model (VLM)-based approaches enhance semantic understanding of change regions by incorporating textual descriptions, they still suffer from challenges such as inaccurate spatial localization, imprecise pixel-level boundary delineation, and limited interpretability. To address these issues, we propose ViLaCD-R1, a two-stage framework comprising a Multi-Image Reasoner (MIR) and a Mask-Guided Decoder (MGD). Specifically, the VLM is trained through supervised fine-tuning (SFT) and reinforcement learning (RL) on block-level dual-temporal inference tasks, taking dual-temporal image patches as input and outputting a coarse change mask. Then, the decoder integrates dual-temporal image features with this coarse mask to predict a precise binary change map. Comprehensive evaluations on multiple RSCD benchmarks demonstrate that ViLaCD-R1 substantially improves true semantic change recognition and localization, robustly suppresses non-semantic variations, and achieves state-of-the-art accuracy in complex real-world scenarios.

Paperid: 3843, https://arxiv.org/pdf/2512.21576.pdf

Abstract:
While large vision-language models (VLMs) demonstrate strong long-context understanding, their prevalent small branches fail on linguistics-photography alignment for a limited window size. We discover that knowledge distillation improves students' capability as a complement to Rotary Position Embeddings (RoPE) on window sizes (anchored from large models). Building on this insight, we propose LAid, which directly aims at the transfer of long-range attention mechanisms through two complementary components: (1) a progressive distance-weighted attention matching that dynamically emphasizes longer position differences during training, and (2) a learnable RoPE response gain modulation that selectively amplifies position sensitivity where needed. Extensive experiments across multiple model families demonstrate that LAid-distilled models achieve up to 3.2 times longer effective context windows compared to baseline small models, while maintaining or improving performance on standard VL benchmarks. Spectral analysis also suggests that LAid successfully preserves crucial low-frequency attention components that conventional methods fail to transfer. Our work not only provides practical techniques for building more efficient long-context VLMs but also offers theoretical insights into how positional understanding emerges and transfers during distillation.

Paperid: 3844, https://arxiv.org/pdf/2512.11109.pdf

Abstract:
Test-time scaling (TTS) has emerged as a powerful paradigm for improving the reasoning ability of Large Language Models (LLMs) by allocating additional computation at inference, yet its application to multimodal systems such as Vision-Language Models (VLMs) remains underexplored. In this work, we present a systematic empirical study of inference time reasoning methods applied across both open-source and closed-source VLMs on different benchmarks. Our results reveal that while closed-source models consistently benefit from structured reasoning and iterative Self-Refinement, open-source VLMs show inconsistent behavior: external verification provides the most reliable gains, whereas iterative refinement often degrades performance. We further find that the effectiveness of TTS is dataset-dependent, yielding clear improvements on multi-step reasoning tasks but offering only limited gains on perception-focused benchmarks. These findings demonstrate that TTS is not a universal solution and must be tailored to both model capabilities and task characteristics, motivating future work on adaptive TTS strategies and multimodal reward models.

Paperid: 3845, https://arxiv.org/pdf/2512.08820.pdf

Abstract:
Recent research in Vision-Language Models (VLMs) has significantly advanced our capabilities in cross-modal reasoning. However, existing methods suffer from performance degradation with domain changes or require substantial computational resources for fine-tuning in new domains. To address this issue, we develop a new adaptation method for large vision-language models, called \textit{Training-free Dual Hyperbolic Adapters} (T-DHA). We characterize the vision-language relationship between semantic concepts, which typically has a hierarchical tree structure, in the hyperbolic space instead of the traditional Euclidean space. Hyperbolic spaces exhibit exponential volume growth with radius, unlike the polynomial growth in Euclidean space. We find that this unique property is particularly effective for embedding hierarchical data structures using the Poincaré ball model, achieving significantly improved representation and discrimination power. Coupled with negative learning, it provides more accurate and robust classifications with fewer feature dimensions. Our extensive experimental results on various datasets demonstrate that the T-DHA method significantly outperforms existing state-of-the-art methods in few-shot image recognition and domain generalization tasks.

Paperid: 3846, https://arxiv.org/pdf/2512.05103.pdf

Abstract:
Video generation models are rapidly advancing, but can still struggle with complex video outputs that require significant semantic branching or repeated high-level reasoning about what should happen next. In this paper, we introduce a new class of omni video-text models that integrate ideas from recent LM reasoning advances to address this challenge. More specifically, we present TV2TV, a unified generative modeling framework which decomposes video generation into an interleaved text and video generation process. TV2TV jointly learns language modeling (next-token prediction) and video flow matching (next-frame prediction) using a Mixture-of-Transformers (MoT) architecture. At inference time, TV2TV decides when to alternate between generating text and video frames, allowing the model to "think in words" about subsequent content before ``acting in pixels'' to produce frames. This design offloads much of the responsibility for deciding what should happen next to the language modeling tower, enabling improved visual quality and prompt alignment of generated videos. It also enables fine-grained controllability, allowing users to modify the video generation trajectory through text interventions at any point in the process. In controlled experiments on video game data, TV2TV demonstrates substantial improvements in both visual quality and controllability. TV2TV also scales to natural videos, as we show by augmenting sports videos with interleaved natural language action descriptions using vision-language models (VLMs). Training TV2TV on this corpus yields strong visual quality and prompt alignment, showcasing the model's ability to reason about and generate complex real-world action sequences. Together, these results highlight TV2TV as a promising step toward video generation with open-ended textual reasoning and control.

Paperid: 3847, https://arxiv.org/pdf/2511.18735.pdf

Abstract:
In this work, we define Foresight Intelligence as the capability to anticipate and interpret future events-an ability essential for applications such as autonomous driving, yet largely overlooked by existing research. To bridge this gap, we introduce FSU-QA, a new Visual Question-Answering (VQA) dataset specifically designed to elicit and evaluate Foresight Intelligence. Using FSU-QA, we conduct the first comprehensive study of state-of-the-art Vision-Language Models (VLMs) under foresight-oriented tasks, revealing that current models still struggle to reason about future situations. Beyond serving as a benchmark, FSU-QA also enables the assessment of world models by measuring the semantic coherence of their generated predictions, quantified through performance gains when VLMs are augmented with such outputs. Our experiments further demonstrate that FSU-QA can effectively enhance foresight reasoning: even small VLMs fine-tuned on FSU-QA surpass much larger, advanced models by a substantial margin. Together, these findings position FSU-QA as a principled foundation for developing next-generation models capable of truly anticipating and understanding future events.

Paperid: 3848, https://arxiv.org/pdf/2511.18591.pdf

Abstract:
Real-world dark images commonly exhibit not only low visibility and contrast but also complex noise and blur, posing significant restoration challenges. Existing methods often rely on paired data or fail to model dynamic illumination and blur characteristics, leading to poor generalization. To tackle this, we propose a generative framework based on visual autoregressive (VAR) modeling, guided by perceptual priors from the vision-language model (VLM). Specifically, to supply informative conditioning cues for VAR models, we deploy an adaptive curve estimation scheme to modulate the diverse illumination based on VLM-derived visibility scores. In addition, we integrate dynamic and spatial-frequency-aware Rotary Positional Encodings (SF-RoPE) into VAR to enhance its ability to model structures degraded by blur. Furthermore, we propose a recursive phase-domain modulation strategy that mitigates blur-induced artifacts in the phase domain via bounded iterative refinement guided by VLM-assessed blur scores. Our framework is fully unsupervised and achieves state-of-the-art performance on benchmark datasets.

Paperid: 3849, https://arxiv.org/pdf/2511.17699.pdf

Abstract:
This paper examines how large language models (LLMs) and large vision-language models (LVLMs) represent and compute numerical information in counting tasks. We use controlled experiments with repeated textual and visual items and analyze model behavior through causal mediation and activation patching. To this end, we design a specialized tool, CountScope, for mechanistic interpretability of numerical content. Results show that individual tokens or visual features encode latent positional count information that can be extracted and transferred across contexts. Layerwise analyses reveal a progressive emergence of numerical representations, with lower layers encoding small counts and higher layers representing larger ones. We identify an internal counter mechanism that updates with each item, stored mainly in the final token or region and transferable between contexts. In LVLMs, numerical information also appears in visual embeddings, shifting between background and foreground regions depending on spatial composition. Models rely on structural cues such as separators in text, which act as shortcuts for tracking item counts and influence the accuracy of numerical predictions. Overall, counting emerges as a structured, layerwise process in LLMs and follows the same general pattern in LVLMs, shaped by the properties of the vision encoder.

Paperid: 3850, https://arxiv.org/pdf/2511.06240.pdf

Abstract:
In open-vocabulary mobile manipulation (OVMM), task success often hinges on the selection of an appropriate base placement for the robot. Existing approaches typically navigate to proximity-based regions without considering affordances, resulting in frequent manipulation failures. We propose Affordance-Guided Coarse-to-Fine Exploration, a zero-shot framework for base placement that integrates semantic understanding from vision-language models (VLMs) with geometric feasibility through an iterative optimization process. Our method constructs cross-modal representations, namely Affordance RGB and Obstacle Map+, to align semantics with spatial context. This enables reasoning that extends beyond the egocentric limitations of RGB perception. To ensure interaction is guided by task-relevant affordances, we leverage coarse semantic priors from VLMs to guide the search toward task-relevant regions and refine placements with geometric constraints, thereby reducing the risk of convergence to local optima. Evaluated on five diverse open-vocabulary mobile manipulation tasks, our system achieves an 85% success rate, significantly outperforming classical geometric planners and VLM-based methods. This demonstrates the promise of affordance-aware and multimodal reasoning for generalizable, instruction-conditioned planning in OVMM.

Paperid: 3851, https://arxiv.org/pdf/2511.05577.pdf

Abstract:
Vision-Language Models (VLMs) have shown strong performance in tasks like visual question answering and multimodal text generation, but their effectiveness in scientific domains such as materials science remains limited. While some machine learning methods have addressed specific challenges in this field, there is still a lack of foundation models designed for broad tasks like polymer property prediction using multimodal data. In this work, we present a multimodal polymer dataset to fine-tune VLMs through instruction-tuning pairs and assess the impact of multimodality on prediction performance. Our fine-tuned models, using LoRA, outperform unimodal and baseline approaches, demonstrating the benefits of multimodal learning. Additionally, this approach reduces the need to train separate models for different properties, lowering deployment and maintenance costs.

Paperid: 3852, https://arxiv.org/pdf/2511.03774.pdf

Abstract:
Recent advances in Vision-Language Models (VLMs) have achieved state-of-the-art performance on numerous benchmark tasks. However, the use of internet-scale, often proprietary, pretraining corpora raises a critical concern for both practitioners and users: inflated performance due to test-set leakage. While prior works have proposed mitigation strategies such as decontamination of pretraining data and benchmark redesign for LLMs, the complementary direction of developing detection methods for contaminated VLMs remains underexplored. To address this gap, we deliberately contaminate open-source VLMs on popular benchmarks and show that existing detection approaches either fail outright or exhibit inconsistent behavior. We then propose a novel simple yet effective detection method based on multi-modal semantic perturbation, demonstrating that contaminated models fail to generalize under controlled perturbations. Finally, we validate our approach across multiple realistic contamination strategies, confirming its robustness and effectiveness. The code and perturbed dataset will be released publicly.

Paperid: 3853, https://arxiv.org/pdf/2511.01331.pdf

Abstract:
Vision-Language-Action (VLA) models have recently emerged as powerful general-purpose policies for robotic manipulation, benefiting from large-scale multi-modal pre-training. However, they often fail to generalize reliably in out-of-distribution deployments, where unavoidable disturbances such as observation noise, sensor errors, or actuation perturbations become prevalent. While recent Reinforcement Learning (RL)-based post-training provides a practical means to adapt pre-trained VLA models, existing methods mainly emphasize reward maximization and overlook robustness to environmental uncertainty. In this work, we introduce RobustVLA, a lightweight online RL post-training method designed to explicitly enhance the resilience of VLA models. Through a systematic robustness analysis, we identify two key regularizations: Jacobian regularization, which mitigates sensitivity to observation noise, and smoothness regularization, which stabilizes policies under action perturbations. Extensive experiments across diverse robotic environments demonstrate that RobustVLA significantly outperforms prior state-of-the-art methods in robustness and reliability. Our results highlight the importance of principled robustness-aware RL post-training as a key step toward improving the reliability and robustness of VLA models.

Paperid: 3854, https://arxiv.org/pdf/2510.19496.pdf

Abstract:
Large vision-language models (VLMs) commonly process images at native or high resolution to remain effective across tasks. This inflates visual tokens ofter to 97-99% of total tokens, resulting in high compute and latency, even when low-resolution images would suffice. We introduce \emph{CARES}-a \textbf{C}ontext-\textbf{A}ware \textbf{R}esolution \textbf{S}elector, a lightweight preprocessing module that, given an image-query pair, predicts the \emph{minimal} sufficient input resolution. CARES uses a compact VLM (350M) to extract features and predict when a target pretrained VLM's response converges to its peak ability to answer correctly. Though trained as a discrete classifier over a set of optional resolutions, CARES interpolates continuous resolutions at inference for fine-grained control. Across five multimodal benchmarks spanning documents and natural images, as well as diverse target VLMs, CARES preserves task performance while reducing compute by up to 80%.

Paperid: 3855, https://arxiv.org/pdf/2510.12709.pdf

Abstract:
Multimodal embedding models aim to yield informative unified representations that empower diverse cross-modal tasks. Despite promising developments in the evolution from CLIP-based dual-tower architectures to large vision-language models, prior works still face unavoidable challenges in real-world applications and business scenarios, such as the limited modality support, unstable training mechanisms, and industrial domain gaps. In this work, we introduce SAIL-Embedding, an omni-modal embedding foundation model that addresses these issues through tailored training strategies and architectural design. In the optimization procedure, we propose a multi-stage training scheme to boost the multifaceted effectiveness of representation learning. Specifically, the content-aware progressive training aims to enhance the model's adaptability to diverse downstream tasks and master enriched cross-modal proficiency. The collaboration-aware recommendation enhancement training further adapts multimodal representations for recommendation scenarios by distilling knowledge from sequence-to-item and ID-to-item embeddings while mining user historical interests. Concurrently, we develop the stochastic specialization and dataset-driven pattern matching to strengthen model training flexibility and generalizability. Experimental results show that SAIL-Embedding achieves SOTA performance compared to other methods in different retrieval tasks. In online experiments across various real-world scenarios integrated with our model, we observe a significant increase in Lifetime (LT), which is a crucial indicator for the recommendation experience. For instance, the model delivers the 7-day LT gain of +0.5% in the Douyin-Selected scenario. For the Douyin feed rank model, the match features produced by SAIL-Embedding yield a +0.1% AUC gain.

Paperid: 3856, https://arxiv.org/pdf/2510.07567.pdf

Abstract:
Vision-Language Models (VLMs) have demonstrated immense capabilities in multi-modal understanding and inference tasks such as Visual Question Answering (VQA), which requires models to infer outputs based on visual and textual context simultaneously. Such inference abilities of large-scale pretrained models are often attributed to the massive scale of pre-training data collected across several domains. However, the models may memorize private and/or sensitive information during training and regurgitate it in inference. Recently, machine unlearning has been leveraged to address the leakage of private data in LLMs. VLMs add a layer of complexity to this process, as the visual context in the query may also contain sensitive information in addition to the text. To address this issue, we explore unlearning for vision-language models, specifically for the VQA task. We explore the role of visual tokens for output generation in VLMs using cross-modal attention and utilize it to formulate Cross-Modal Attention Guided Unlearning (CAGUL), a lightweight and efficient VLM unlearning framework. In contrast to computationally expensive model finetuning methods, CAGUL utilizes external modules to encode unlearning information in visual tokens of low importance for relevant queries. We find that the transformed visual tokens not only prevent leakage but also retain reference model behavior. Experimental results show that our method performs better or on par with finetuning-based baselines without altering the pre-trained model parameters or incurring retraining costs, making it a practical and effective unlearning solution for VLMs.

Paperid: 3857, https://arxiv.org/pdf/2510.03612.pdf

Abstract:
Vision-language model (VLM)-based web agents increasingly power high-stakes selection tasks like content recommendation or product ranking by combining multimodal perception with preference reasoning. Recent studies reveal that these agents are vulnerable against attackers who can bias selection outcomes through preference manipulations using adversarial pop-ups, image perturbations, or content tweaks. Existing work, however, either assumes strong white-box access, with limited single-modal perturbations, or uses impractical settings. In this paper, we demonstrate, for the first time, that joint exploitation of visual and textual channels yields significantly more powerful preference manipulations under realistic attacker capabilities. We introduce Cross-Modal Preference Steering (CPS) that jointly optimizes imperceptible modifications to an item's visual and natural language descriptions, exploiting CLIP-transferable image perturbations and RLHF-induced linguistic biases to steer agent decisions. In contrast to prior studies that assume gradient access, or control over webpages, or agent memory, we adopt a realistic black-box threat setup: a non-privileged adversary can edit only their own listing's images and textual metadata, with no insight into the agent's model internals. We evaluate CPS on agents powered by state-of-the-art proprietary and open source VLMs including GPT-4.1, Qwen-2.5VL and Pixtral-Large on both movie selection and e-commerce tasks. Our results show that CPS is significantly more effective than leading baseline methods. For instance, our results show that CPS consistently outperforms baselines across all models while maintaining 70% lower detection rates, demonstrating both effectiveness and stealth. These findings highlight an urgent need for robust defenses as agentic systems play an increasingly consequential role in society.

Paperid: 3858, https://arxiv.org/pdf/2510.01009.pdf

Abstract:
Video Question Answering (VQA) with Large Vision Language Models (LVLMs) has gained significant traction in research ever since the Flamingo was introduced by Deepmind. Recent advancements in large context/long video question answering have allowed VQA tasks to have context window of 1500+ frames. However, this only leads to 50 seconds of video footage without losing any significant information. We introduce POVQA, a data-efficient pipeline that compresses each second of video into a single temporally pooled image (via motion blur and weighted averaging variants) and then align LVLMs with lightweight supervision. Concretely, we build 1 fps input sources using Blend Blur with Last Frame, Weighted Average, Exponential and Ramp pooling and fine-tune QWEN-2.5-VL 7B with supervised two turn target including reasoning and final answer. We apply Supervised Fine Tuning (SFT) and Direct Preference Optimization (DPO) on our novel dataset ReasonVQA consisting of 12 movies with 239 human annotated question-answer with reasoning prompts. On our ReasonVQA dataset, this method dramatically improves performance over pooled baselines: F1 score improves from 0.212 to 0.543, BLEU-4 from 0.031 to 0.291, and ROUGE-L from 0.196 to 0.528. Rationale quality also significantly increases. Cross-evaluation of SFT + DPO on various pooling functions show that the gains persist regardless of the pooling scheme used at train or test time, indicating strong robustness on summarization of temporal evidence. Similar observations were made on zero-shot in TVQA.

Paperid: 3859, https://arxiv.org/pdf/2509.25844.pdf

Abstract:
When people query Vision-Language Models (VLMs) but cannot see the accompanying visual context (e.g. for blind and low-vision users), augmenting VLM predictions with natural language explanations can signal which model predictions are reliable. However, prior work has found that explanations can easily convince users that inaccurate VLM predictions are correct. To remedy undesirable overreliance on VLM predictions, we propose evaluating two complementary qualities of VLM-generated explanations via two quality scoring functions. We propose Visual Fidelity, which captures how faithful an explanation is to the visual context, and Contrastiveness, which captures how well the explanation identifies visual details that distinguish the model's prediction from plausible alternatives. On the A-OKVQA and VizWiz tasks, these quality scoring functions are better calibrated with model correctness than existing explanation qualities. We conduct a user study in which participants have to decide whether a VLM prediction is accurate without viewing its visual context. We observe that showing our quality scores alongside VLM explanations improves participants' accuracy at predicting VLM correctness by 11.1%, including a 15.4% reduction in the rate of falsely believing incorrect predictions. These findings highlight the utility of explanation quality scores in fostering appropriate reliance on VLM predictions.

Paperid: 3860, https://arxiv.org/pdf/2509.25373.pdf

Abstract:
Multimodal Large Language Models (MLLMs) strive to achieve a profound, human-like understanding of and interaction with the physical world, but often exhibit a shallow and incoherent integration when acquiring information (Perception) and conducting reasoning (Cognition). This disconnect leads to a spectrum of reasoning failures, with hallucination being the most prominent. Collectively, these issues expose a fundamental challenge: the ability to process pixels does not yet confer the ability to construct a coherent, credible internal world model. To systematically dissect and address this challenge, this survey introduces a novel and unified analytical framework: ``From Perception to Cognition." We deconstruct the complex process of vision-language interactive understanding into two interdependent layers: Perception, the foundational ability to accurately extract visual information and achieve fine-grained alignment with textual instructions; and Cognition, the higher-order capability for proactive, multi-step, goal-oriented reasoning built upon this perceptual foundation, the core of which is the formation of a dynamic observe-think-verify reasoning loop. Guided by this framework, this paper systematically analyzes the key bottlenecks of current MLLMs at both layers. It surveys the landscape of cutting-edge methods designed to address these challenges, spanning from techniques that enhance low-level visual representations to those that improve high-level reasoning paradigms. Furthermore, we review critical benchmarks and delineate future research directions. This survey aims to provide the research community with a clear, structured perspective for understanding the intrinsic limitations of current MLLMs and to illuminate the path toward building next-generation models capable of deep reasoning and a genuine understanding of the world.

Paperid: 3861, https://arxiv.org/pdf/2509.24739.pdf

Abstract:
Vision-Language Foundation Models (VLMs), trained on large-scale multimodal datasets, have driven significant advances in Artificial Intelligence by enabling rich cross-modal reasoning. Despite their success in general domains, applying these models to medical imaging remains challenging due to the limited availability of diverse imaging modalities and multilingual clinical data. Most existing medical VLMs are trained on a subset of imaging modalities and focus primarily on high-resource languages, thus limiting their generalizability and clinical utility. To address these limitations, we introduce a novel Vietnamese-language multimodal medical dataset comprising 1,567,062 paired CT-PET images and corresponding 2,757 full-length clinical reports. This dataset is designed to fill two pressing gaps in medical AI development: (1) the lack of PET/CT imaging data in existing VLMs training corpora, which hinders the development of models capable of handling functional imaging tasks; and (2) the underrepresentation of low-resource languages, particularly the Vietnamese language, in medical vision-language research. To the best of our knowledge, this is the first dataset to provide comprehensive PET/CT-report pairs in Vietnamese. We further introduce a training framework to enhance VLMs' learning, including data augmentation and expert-validated test sets. We conduct comprehensive experiments benchmarking state-of-the-art VLMs on downstream tasks, including medical report generation and visual question answering. The experimental results show that incorporating our dataset significantly improves the performance of existing VLMs. We believe this dataset and benchmark will serve as a pivotal step in advancing the development of more robust VLMs for medical imaging, particularly in low-resource languages, and improving their clinical relevance in Vietnamese healthcare.

Paperid: 3862, https://arxiv.org/pdf/2509.24640.pdf

Abstract:
In this work, we introduce SPLICE, a human-curated benchmark derived from the COIN instructional video dataset, designed to probe event-based reasoning across multiple dimensions: temporal, causal, spatial, contextual, and general knowledge. SPLICE includes 3,381 human-filtered videos spanning 12 categories and 180 sub-categories, such as sports, engineering, and housework. These videos are segmented into a total of 11,423 event clips. We evaluate both human participants and state-of-the-art vision-language models (VLMs) on the task of rearranging these clips into coherent event sequences to assess visual reasoning capabilities. Results reveal a significant gap: VLMs struggle to match human performance. While human-annotated textual descriptions improve model accuracy, they do not affect human performance, suggesting that models rely more on language priors than on visual understanding. Even with annotations, VLMs fall short of human-level reasoning, underscoring persistent challenges in visual reasoning. A deeper analysis across sub-categories shows that VLMs perform relatively better on videos where temporal and causal reasoning are dominant, compared to those where contextual and spatial reasoning are dominant. They also perform better on everyday tasks than on specialized ones.

Paperid: 3863, https://arxiv.org/pdf/2509.24378.pdf

Abstract:
Time-series anomaly detection (TSAD) increasingly demands explanations that articulate not only if an anomaly occurred, but also what pattern it exhibits and why it is anomalous. Leveraging the impressive explanatory capabilities of Large Language Models (LLMs), recent works have attempted to treat time series as text for explainable TSAD. However, this approach faces a fundamental challenge: LLMs operate on discrete tokens and struggle to directly process long, continuous signals. Consequently, naive time-to-text serialization suffers from a lack of contextual grounding and representation alignment between the two modalities. To address this gap, we introduce AXIS, a framework that conditions a frozen LLM for nuanced time-series understanding. Instead of direct serialization, AXIS enriches the LLM's input with three complementary hints derived from the series: (i) a symbolic numeric hint for numerical grounding, (ii) a context-integrated, step-aligned hint distilled from a pretrained time-series encoder to capture fine-grained dynamics, and (iii) a task-prior hint that encodes global anomaly characteristics. Furthermore, to facilitate robust evaluation of explainability, we introduce a new benchmark featuring multi-format questions and rationales that supervise contextual grounding and pattern-level semantics. Extensive experiments, including both LLM-based and human evaluations, demonstrate that AXIS yields explanations of significantly higher quality and achieves competitive detection accuracy compared to general-purpose LLMs, specialized time-series LLMs, and time-series Vision Language Models.

Paperid: 3864, https://arxiv.org/pdf/2509.24072.pdf

Abstract:
Large vision-language models (LVLMs) show strong performance across multimodal benchmarks but remain limited in structured reasoning and precise grounding. Recent work has demonstrated that adding simple visual structures, such as partitions and annotations, improves accuracy, yet the internal mechanisms underlying these gains remain unclear. We investigate this phenomenon and propose the concept of Grounding IDs, latent identifiers induced by external cues that bind objects to their designated partitions across modalities. Through representation analysis, we find that these identifiers emerge as robust within-partition alignment in embedding space and reduce the modality gap between image and text. Causal interventions further confirm that these identifiers mediate binding between objects and symbolic cues. We show that Grounding IDs strengthen attention between related components, which in turn improves cross-modal grounding and reduces hallucinations. Taken together, our results identify Grounding IDs as a key symbolic mechanism explaining how external cues enhance multimodal binding, offering both interpretability and practical improvements in robustness.

Paperid: 3865, https://arxiv.org/pdf/2509.22653.pdf

Abstract:
We present See, Point, Fly (SPF), a training-free aerial vision-and-language navigation (AVLN) framework built atop vision-language models (VLMs). SPF is capable of navigating to any goal based on any type of free-form instructions in any kind of environment. In contrast to existing VLM-based approaches that treat action prediction as a text generation task, our key insight is to consider action prediction for AVLN as a 2D spatial grounding task. SPF harnesses VLMs to decompose vague language instructions into iterative annotation of 2D waypoints on the input image. Along with the predicted traveling distance, SPF transforms predicted 2D waypoints into 3D displacement vectors as action commands for UAVs. Moreover, SPF also adaptively adjusts the traveling distance to facilitate more efficient navigation. Notably, SPF performs navigation in a closed-loop control manner, enabling UAVs to follow dynamic targets in dynamic environments. SPF sets a new state of the art in DRL simulation benchmark, outperforming the previous best method by an absolute margin of 63%. In extensive real-world evaluations, SPF outperforms strong baselines by a large margin. We also conduct comprehensive ablation studies to highlight the effectiveness of our design choice. Lastly, SPF shows remarkable generalization to different VLMs. Project page: https://spf-web.pages.dev

Paperid: 3866, https://arxiv.org/pdf/2509.20524.pdf

Abstract:
We present InstructVTON, an instruction-following interactive virtual try-on system that allows fine-grained and complex styling control of the resulting generation, guided by natural language, on single or multiple garments. A computationally efficient and scalable formulation of virtual try-on formulates the problem as an image-guided or image-conditioned inpainting task. These inpainting-based virtual try-on models commonly use a binary mask to control the generation layout. Producing a mask that yields desirable result is difficult, requires background knowledge, might be model dependent, and in some cases impossible with the masking-based approach (e.g. trying on a long-sleeve shirt with "sleeves rolled up" styling on a person wearing long-sleeve shirt with sleeves down, where the mask will necessarily cover the entire sleeve). InstructVTON leverages Vision Language Models (VLMs) and image segmentation models for automated binary mask generation. These masks are generated based on user-provided images and free-text style instructions. InstructVTON simplifies the end-user experience by removing the necessity of a precisely drawn mask, and by automating execution of multiple rounds of image generation for try-on scenarios that cannot be achieved with masking-based virtual try-on models alone. We show that InstructVTON is interoperable with existing virtual try-on models to achieve state-of-the-art results with styling control.

Paperid: 3867, https://arxiv.org/pdf/2509.18638.pdf

Abstract:
Neuroimaging is a ubiquitous tool for evaluating patients with neurological diseases. The global demand for magnetic resonance imaging (MRI) studies has risen steadily, placing significant strain on health systems, prolonging turnaround times, and intensifying physician burnout \cite{Chen2017-bt, Rula2024-qp-1}. These challenges disproportionately impact patients in low-resource and rural settings. Here, we utilized a large academic health system as a data engine to develop Prima, the first vision language model (VLM) serving as an AI foundation for neuroimaging that supports real-world, clinical MRI studies as input. Trained on over 220,000 MRI studies, Prima uses a hierarchical vision architecture that provides general and transferable MRI features. Prima was tested in a 1-year health system-wide study that included 30K MRI studies. Across 52 radiologic diagnoses from the major neurologic disorders, including neoplastic, inflammatory, infectious, and developmental lesions, Prima achieved a mean diagnostic area under the ROC curve of 92.0, outperforming other state-of-the-art general and medical AI models. Prima offers explainable differential diagnoses, worklist priority for radiologists, and clinical referral recommendations across diverse patient demographics and MRI systems. Prima demonstrates algorithmic fairness across sensitive groups and can help mitigate health system biases, such as prolonged turnaround times for low-resource populations. These findings highlight the transformative potential of health system-scale VLMs and Prima's role in advancing AI-driven healthcare.

Paperid: 3868, https://arxiv.org/pdf/2509.12132.pdf

Abstract:
Recent advances in text-only "slow-thinking" reasoning have prompted efforts to transfer this capability to vision-language models (VLMs), for training visual reasoning models (\textbf{VRMs}). owever, such transfer faces critical challenges: Effective "slow thinking" in VRMs requires \textbf{visual reflection}, the ability to check the reasoning process based on visual information. Through quantitative analysis, we observe that current VRMs exhibit limited visual reflection, as their attention to visual information diminishes rapidly with longer generated responses. To address this challenge, we propose a new VRM \textbf{Reflection-V}, which enhances visual reflection based on reasoning data construction for cold-start and reward design for reinforcement learning (RL). Firstly, we construct vision-centered reasoning data by leveraging an agent that interacts between VLMs and reasoning LLMs, enabling cold-start learning of visual reflection patterns. Secondly, a visual attention based reward model is employed during RL to encourage reasoning based on visual information. Therefore, \textbf{Reflection-V} demonstrates significant improvements across multiple visual reasoning benchmarks. Furthermore, \textbf{Reflection-V} maintains a stronger and more consistent reliance on visual information during visual reasoning, indicating effective enhancement in visual reflection capabilities.

Paperid: 3869, https://arxiv.org/pdf/2509.11364.pdf

Abstract:
Accurate 6-DoF object pose estimation and tracking are critical for reliable robotic manipulation. However, zero-shot methods often fail under viewpoint-induced ambiguities and fixed-camera setups struggle when objects move or become self-occluded. To address these challenges, we propose an active pose estimation pipeline that combines a Vision-Language Model (VLM) with "robotic imagination" to dynamically detect and resolve ambiguities in real time. In an offline stage, we render a dense set of views of the CAD model, compute the FoundationPose entropy for each view, and construct a geometric-aware prompt that includes low-entropy (unambiguous) and high-entropy (ambiguous) examples. At runtime, the system: (1) queries the VLM on the live image for an ambiguity score; (2) if ambiguity is detected, imagines a discrete set of candidate camera poses by rendering virtual views, scores each based on a weighted combination of VLM ambiguity probability and FoundationPose entropy, and then moves the camera to the Next-Best-View (NBV) to obtain a disambiguated pose estimation. Furthermore, since moving objects may leave the camera's field of view, we introduce an active pose tracking module: a diffusion-policy trained via imitation learning, which generates camera trajectories that preserve object visibility and minimize pose ambiguity. Experiments in simulation and real-world show that our approach significantly outperforms classical baselines.

Paperid: 3870, https://arxiv.org/pdf/2509.05614.pdf

Abstract:
Pruning accelerates compute-bound models by reducing computation. Recently applied to Vision-Language-Action (VLA) models, existing methods prune tokens using only local info from current action, ignoring global context from prior actions, causing >20% success rate drop and limited speedup. We observe high similarity across consecutive actions and propose leveraging both local (current) and global (past) info for smarter token selection. We introduce SpecPrune-VLA, a training-free method with two-level pruning and heuristic control: (1) Static pruning at action level: uses global history and local context to reduce visual tokens per action; (2) Dynamic pruning at layer level: prunes tokens per layer based on layer-specific importance; (3) Lightweight action-aware controller: classifies actions as coarse/fine-grained (by speed), adjusting pruning aggressiveness since fine-grained actions are pruning-sensitive. Experiments on LIBERO show SpecPrune-VLA achieves 1.46 times speedup on NVIDIA A800 and 1.57 times on NVIDIA GeForce RTX 3090 vs. OpenVLA-OFT, with negligible success rate loss.

Paperid: 3871, https://arxiv.org/pdf/2509.04063.pdf

Abstract:
Vision-Language-Action (VLA) models based on flow matching have shown excellent performance in general-purpose robotic manipulation tasks. However, the action accuracy of these models on complex downstream tasks is unsatisfactory. One important reason is that these models rely solely on the post-training paradigm of imitation learning, which makes it difficult to have a deeper understanding of the distribution properties of data quality, which is exactly what Reinforcement Learning (RL) excels at. In this paper, we theoretically propose an offline RL post-training objective for VLA flow models and induce an efficient and feasible offline RL fine-tuning algorithm -- Adaptive Reinforced Flow Matching (ARFM). By introducing an adaptively adjusted scaling factor in the VLA flow model loss, we construct a principled bias-variance trade-off objective function to optimally control the impact of RL signal on flow loss. ARFM adaptively balances RL advantage preservation and flow loss gradient variance control, resulting in a more stable and efficient fine-tuning process. Extensive simulation and real-world experimental results show that ARFM exhibits excellent generalization, robustness, few-shot learning, and continuous learning performance.

Paperid: 3872, https://arxiv.org/pdf/2509.03903.pdf

Abstract:
The scarcity of well-annotated diverse medical images is a major hurdle for developing reliable AI models in healthcare. Substantial technical advances have been made in generative foundation models for natural images. Here we develop `ChexGen', a generative vision-language foundation model that introduces a unified framework for text-, mask-, and bounding box-guided synthesis of chest radiographs. Built upon the latent diffusion transformer architecture, ChexGen was pretrained on the largest curated chest X-ray dataset to date, consisting of 960,000 radiograph-report pairs. ChexGen achieves accurate synthesis of radiographs through expert evaluations and quantitative metrics. We demonstrate the utility of ChexGen for training data augmentation and supervised pretraining, which led to performance improvements across disease classification, detection, and segmentation tasks using a small fraction of training data. Further, our model enables the creation of diverse patient cohorts that enhance model fairness by detecting and mitigating demographic biases. Our study supports the transformative role of generative foundation models in building more accurate, data-efficient, and equitable medical AI systems.

Paperid: 3873, https://arxiv.org/pdf/2509.02129.pdf

Abstract:
Visual Place Recognition (VPR) has evolved from handcrafted descriptors to deep learning approaches, yet significant challenges remain. Current approaches, including Vision Foundation Models (VFMs) and Multimodal Large Language Models (MLLMs), enhance semantic understanding but suffer from high computational overhead and limited cross-domain transferability when fine-tuned. To address these limitations, we propose a novel zero-shot framework employing Test-Time Scaling (TTS) that leverages MLLMs' vision-language alignment capabilities through Guidance-based methods for direct similarity scoring. Our approach eliminates two-stage processing by employing structured prompts that generate length-controllable JSON outputs. The TTS framework with Uncertainty-Aware Self-Consistency (UASC) enables real-time adaptation without additional training costs, achieving superior generalization across diverse environments. Experimental results demonstrate significant improvements in cross-domain VPR performance with up to 210$\times$ computational efficiency gains.

Paperid: 3874, https://arxiv.org/pdf/2508.10556.pdf

Abstract:
Out-of-Distribution (OOD) detection is crucial for the reliable deployment of machine learning models in-the-wild, enabling accurate identification of test samples that differ from the training data distribution. Existing methods rely on auxiliary outlier samples or in-distribution (ID) data to generate outlier information for training, but due to limited outliers and their mismatch with real test OOD samples, they often fail to provide sufficient semantic supervision, leading to suboptimal performance. To address this, we propose a novel OOD detection method called Retrieval-Augmented Prompt (RAP). RAP augments a pre-trained vision-language model's prompts by retrieving external knowledge, offering enhanced semantic supervision for OOD detection. During training, RAP retrieves descriptive words for outliers based on joint similarity with external textual knowledge and uses them to augment the model's OOD prompts. During testing, RAP dynamically updates OOD prompts in real-time based on the encountered OOD samples, enabling the model to rapidly adapt to the test environment. Our extensive experiments demonstrate that RAP achieves state-of-the-art performance on large-scale OOD detection benchmarks. For example, in 1-shot OOD detection on the ImageNet-1k dataset, RAP reduces the average FPR95 by 7.05% and improves the AUROC by 1.71% compared to previous methods. Additionally, comprehensive ablation studies validate the effectiveness of each module and the underlying motivations of our approach.

Paperid: 3875, https://arxiv.org/pdf/2508.10339.pdf

Abstract:
Vision-language instruction tuning achieves two main purposes: learning visual concepts and learning visual skills. In this paper, we found that vision-language benchmarks fall into the dichotomy of mainly benefiting from training on instructions with similar skills or visual concepts. Inspired by the discovery, we designed a simple targeted training data selection method to optimize the performance of a given benchmark. We first extract the concepts/skills from the benchmark, determine whether the benchmark predominantly benefits from similar concepts or skills, and finally select instructions with the most matching concepts/skills. Experiments on 10+ benchmarks validate the effectiveness of our targeted data selection method, showing +0.9\% over the best existing baseline averaged over all benchmarks and +1.5\% on the skill-focused subset. Our findings underscore the importance of recognizing the inherent trade-off within instruction selection, which requires balancing the acquisition of conceptual knowledge against visual skill.

Paperid: 3876, https://arxiv.org/pdf/2508.09123.pdf

Abstract:
Vision-language models have demonstrated impressive capabilities as computer-use agents (CUAs) capable of automating diverse computer tasks. As their commercial potential grows, critical details of the most capable CUA systems remain closed. As these agents will increasingly mediate digital interactions and execute consequential decisions on our behalf, the research community needs access to open CUA frameworks to study their capabilities, limitations, and risks. To bridge this gap, we propose OpenCUA, a comprehensive open-source framework for scaling CUA data and foundation models. Our framework consists of: (1) an annotation infrastructure that seamlessly captures human computer-use demonstrations; (2) AgentNet, the first large-scale computer-use task dataset spanning 3 operating systems and 200+ applications and websites; (3) a scalable pipeline that transforms demonstrations into state-action pairs with reflective long Chain-of-Thought reasoning that sustain robust performance gains as data scales. Our end-to-end agent models demonstrate strong performance across CUA benchmarks. In particular, OpenCUA-32B achieves an average success rate of 34.8% on OSWorld-Verified, establishing a new state-of-the-art (SOTA) among open-source models and surpassing OpenAI CUA (GPT-4o). Further analysis confirms that our approach generalizes well across domains and benefits significantly from increased test-time computation. We release our annotation tool, datasets, code, and models to build open foundations for further CUA research.

Paperid: 3877, https://arxiv.org/pdf/2508.04175.pdf

Abstract:
While Multimodal Large Language Models (MLLMs) demonstrate remarkable capabilities across diverse domains, their application to specialized anomaly detection (AD) remains constrained by domain adaptation challenges. Existing Group Relative Policy Optimization (GRPO) based approaches suffer from two critical limitations: inadequate training data utilization when models produce uniform responses, and insufficient supervision over reasoning processes that encourage immediate binary decisions without deliberative analysis. We propose a comprehensive framework addressing these limitations through two synergistic innovations. First, we introduce a multi-stage deliberative reasoning process that guides models from region identification to focused examination, generating diverse response patterns essential for GRPO optimization while enabling structured supervision over analytical workflows. Second, we develop a fine-grained reward mechanism incorporating classification accuracy and localization supervision, transforming binary feedback into continuous signals that distinguish genuine analytical insight from spurious correctness. Comprehensive evaluation across multiple industrial datasets demonstrates substantial performance improvements in adapting general vision-language models to specialized anomaly detection. Our method achieves superior accuracy with efficient adaptation of existing annotations, effectively bridging the gap between general-purpose MLLM capabilities and the fine-grained visual discrimination required for detecting subtle manufacturing defects and structural irregularities.

Paperid: 3878, https://arxiv.org/pdf/2508.01057.pdf

Abstract:
Autonomous driving (AD) systems relying solely on onboard sensors may fail to detect distant or obstacle hazards, potentially causing preventable collisions; however, existing transformer-based Vehicle-to-Everything (V2X) approaches, which mitigate AD sensing limitations, either lack effective multimodal fusion and reasoning or struggle to meet real-time performance requirements under complex, high-dimensional traffic conditions. This paper proposes the Real-time Edge-based Autonomous Co-pilot Trajectory planner (REACT), a V2X-integrated trajectory optimization framework for AD based on a fine-tuned lightweight Vision-Language Model (VLM). REACT integrates infrastructure-provided hazard alerts with onboard sensor data, capturing intricate surrounding traffic dynamics and vehicle intents through visual embeddings, interpreting precise numerical data from symbolic inputs, and employing contextual reasoning to generate optimized, safety-oriented trajectories. To ensure robust real-time deployment on edge devices, REACT innovatively employs Residual Trajectory Fusion (RTF) design and specialized edge-adaptation strategies to reduce model complexity and improve inference efficiency. Evaluated on the DeepAccident benchmark, REACT achieves state-of-the-art performance, a 77% collision rate reduction, a 48.2% Video Panoptic Quality (VPQ), and a 0.57-second inference latency on the Jetson AGX Orin. Ablation studies validate the contribution of each input, module, and edge adaptation strategy. These results highlight the effectiveness of lightweight VLMs in enabling real-time cooperative planning on edge platforms and underscore the potential of language-guided contextual reasoning for improving traffic safety and responsiveness.

Paperid: 3879, https://arxiv.org/pdf/2507.21540.pdf

Abstract:
The increasing sophistication of large vision-language models (LVLMs) has been accompanied by advances in safety alignment mechanisms designed to prevent harmful content generation. However, these defenses remain vulnerable to sophisticated adversarial attacks. Existing jailbreak methods typically rely on direct and semantically explicit prompts, overlooking subtle vulnerabilities in how LVLMs compose information over multiple reasoning steps. In this paper, we propose a novel and effective jailbreak framework inspired by Return-Oriented Programming (ROP) techniques from software security. Our approach decomposes a harmful instruction into a sequence of individually benign visual gadgets. A carefully engineered textual prompt directs the sequence of inputs, prompting the model to integrate the benign visual gadgets through its reasoning process to produce a coherent and harmful output. This makes the malicious intent emergent and difficult to detect from any single component. We validate our method through extensive experiments on established benchmarks including SafeBench and MM-SafetyBench, targeting popular LVLMs. Results show that our approach consistently and substantially outperforms existing baselines on state-of-the-art models, achieving near-perfect attack success rates (over 0.90 on SafeBench) and improving ASR by up to 0.39. Our findings reveal a critical and underexplored vulnerability that exploits the compositional reasoning abilities of LVLMs, highlighting the urgent need for defenses that secure the entire reasoning process.

Paperid: 3880, https://arxiv.org/pdf/2507.20776.pdf

Abstract:
Remote sensing (RS) images from multiple modalities and platforms exhibit diverse details due to differences in sensor characteristics and imaging perspectives. Existing vision-language research in RS largely relies on relatively homogeneous data sources. Moreover, they still remain limited to conventional visual perception tasks such as classification or captioning. As a result, these methods fail to serve as a unified and standalone framework capable of effectively handling RS imagery from diverse sources in real-world applications. To address these issues, we propose RingMo-Agent, a model designed to handle multi-modal and multi-platform data that performs perception and reasoning tasks based on user textual instructions. Compared with existing models, RingMo-Agent 1) is supported by a large-scale vision-language dataset named RS-VL3M, comprising over 3 million image-text pairs, spanning optical, SAR, and infrared (IR) modalities collected from both satellite and UAV platforms, covering perception and challenging reasoning tasks; 2) learns modality adaptive representations by incorporating separated embedding layers to construct isolated features for heterogeneous modalities and reduce cross-modal interference; 3) unifies task modeling by introducing task-specific tokens and employing a token-based high-dimensional hidden state decoding mechanism designed for long-horizon spatial tasks. Extensive experiments on various RS vision-language tasks demonstrate that RingMo-Agent not only proves effective in both visual understanding and sophisticated analytical tasks, but also exhibits strong generalizability across different platforms and sensing modalities.

Paperid: 3881, https://arxiv.org/pdf/2507.19262.pdf

Abstract:
Large vision-language models (VLMs) often struggle to generate long and factual captions. However, traditional measures for hallucination and factuality are not well suited for evaluating longer, more diverse captions and in settings where ground-truth human-annotated captions are unavailable. We introduce OV-Fact, a novel method for measuring caption factuality of long captions that leverages open-vocabulary visual grounding and tool-based verification without depending on human annotations. Our method improves agreement with human judgments and captures both caption descriptiveness (recall) and factual precision in the same metric. Furthermore, unlike previous metrics, our reference-free method design enables new applications towards factuality-based data filtering. We observe models trained on an OVFact-filtered (2.5-5x less) subset of a large-scale, noisy (VLM-generated) pretraining set meaningfully improve factuality precision without sacrificing caption descriptiveness across a range of downstream long caption benchmarks.

Paperid: 3882, https://arxiv.org/pdf/2507.17859.pdf

Abstract:
Accurate fish detection in underwater imagery is essential for ecological monitoring, aquaculture automation, and robotic perception. However, practical deployment remains limited by fragmented datasets, heterogeneous imaging conditions, and inconsistent evaluation protocols. To address these gaps, we present \textit{FishDet-M}, the largest unified benchmark for fish detection, comprising 13 publicly available datasets spanning diverse aquatic environments including marine, brackish, occluded, and aquarium scenes. All data are harmonized using COCO-style annotations with both bounding boxes and segmentation masks, enabling consistent and scalable cross-domain evaluation. We systematically benchmark 28 contemporary object detection models, covering the YOLOv8 to YOLOv12 series, R-CNN based detectors, and DETR based models. Evaluations are conducted using standard metrics including mAP, mAP@50, and mAP@75, along with scale-specific analyses (AP$_S$, AP$_M$, AP$_L$) and inference profiling in terms of latency and parameter count. The results highlight the varying detection performance across models trained on FishDet-M, as well as the trade-off between accuracy and efficiency across models of different architectures. To support adaptive deployment, we introduce a CLIP-based model selection framework that leverages vision-language alignment to dynamically identify the most semantically appropriate detector for each input image. This zero-shot selection strategy achieves high performance without requiring ensemble computation, offering a scalable solution for real-time applications. FishDet-M establishes a standardized and reproducible platform for evaluating object detection in complex aquatic scenes. All datasets, pretrained models, and evaluation tools are publicly available to facilitate future research in underwater computer vision and intelligent marine systems.

Paperid: 3883, https://arxiv.org/pdf/2507.16572.pdf

Abstract:
This paper presents a systematic evaluation of state-of-the-art multimodal large language models (MLLMs) on intuitive physics tasks using the GRASP and IntPhys 2 datasets. We assess the open-source models InternVL 2.5, Qwen 2.5 VL, LLaVA-OneVision, and the proprietary Gemini 2.0 Flash Thinking, finding that even the latest models struggle to reliably distinguish physically plausible from implausible scenarios. To go beyond performance metrics, we conduct a probing analysis of model embeddings, extracting intermediate representations at key processing stages to examine how well task-relevant information is preserved. Our results show that, depending on task difficulty, a critical vision-language misalignment can emerge: vision encoders successfully capture physical plausibility cues, but this information is not effectively utilized by the language model, leading to failures in reasoning. This misalignment suggests that the primary limitation of MLLMs in intuitive physics tasks is not the vision component but the ineffective integration of visual and linguistic information. Our findings highlight vision-language alignment as a key area for improvement, offering insights for future MLLMs development.

Paperid: 3884, https://arxiv.org/pdf/2507.07299.pdf

Abstract:
Recent progress in large vision-language models has driven improvements in language-based semantic navigation, where an embodied agent must reach a target object described in natural language. Despite these advances, we still lack a clear, language-focused benchmark for testing how well such agents ground the words in their instructions. We address this gap with LangNav, an open-set dataset specifically created to test an agent's ability to locate objects described at different levels of detail, from broad category names to fine attributes and object-object relations. Every description in LangNav was manually checked, yielding a lower error rate than existing lifelong- and semantic-navigation datasets. On top of LangNav we build LangNavBench, a benchmark that measures how well current semantic-navigation methods understand and act on these descriptions while moving toward their targets. LangNavBench allows us to systematically compare models on their handling of attributes, spatial and relational cues, and category hierarchies, offering the first thorough, language-centric evaluation of embodied navigation systems. We also present Multi-Layered Feature Map (MLFM), a method that builds a queryable multi-layered semantic map, particularly effective when dealing with small objects or instructions involving spatial relations. MLFM outperforms state-of-the-art mapping-based navigation baselines on the LangNav dataset.

Paperid: 3885, https://arxiv.org/pdf/2506.23903.pdf

Abstract:
Accurate and generalizable object segmentation in ultrasound imaging remains a significant challenge due to anatomical variability, diverse imaging protocols, and limited annotated data. In this study, we propose a prompt-driven vision-language model (VLM) that integrates Grounding DINO with SAM2 (Segment Anything Model2) to enable object segmentation across multiple ultrasound organs. A total of 18 public ultrasound datasets, encompassing the breast, thyroid, liver, prostate, kidney, and paraspinal muscle, were utilized. These datasets were divided into 15 for fine-tuning and validation of Grounding DINO using Low Rank Adaptation (LoRA) to the ultrasound domain, and 3 were held out entirely for testing to evaluate performance in unseen distributions. Comprehensive experiments demonstrate that our approach outperforms state-of-the-art segmentation methods, including UniverSeg, MedSAM, MedCLIP-SAM, BiomedParse, and SAMUS on most seen datasets while maintaining strong performance on unseen datasets without additional fine-tuning. These results underscore the promise of VLMs in scalable and robust ultrasound image analysis, reducing dependence on large, organ-specific annotated datasets. We will publish our code on code.sonography.ai after acceptance.

Paperid: 3886, https://arxiv.org/pdf/2506.23120.pdf

Abstract:
Recent advances in point cloud perception have demonstrated remarkable progress in scene understanding through vision-language alignment leveraging large language models (LLMs). However, existing methods may still encounter challenges in handling complex instructions that require accurate spatial reasoning, even if the 3D point cloud data provides detailed spatial cues such as size and position for identifying the targets. To tackle this issue, we propose Relevant Reasoning Segmentation (R$^2$S), a reasoning-based segmentation framework. The framework emulates human cognitive processes by decomposing spatial reasoning into two sequential stages: first identifying relevant elements, then processing instructions guided by their associated visual priors. Furthermore, acknowledging the inadequacy of existing datasets in complex reasoning tasks, we introduce 3D ReasonSeg, a reasoning-based segmentation dataset comprising 25,185 training samples and 3,966 validation samples with precise annotations. Both quantitative and qualitative experiments demonstrate that the R$^2$S and 3D ReasonSeg effectively endow 3D point cloud perception with stronger spatial reasoning capabilities, and we hope that they can serve as a new baseline and benchmark for future work.

Paperid: 3887, https://arxiv.org/pdf/2506.22375.pdf

Abstract:
Out-of-distribution (OOD) detection in 3D point cloud data remains a challenge, particularly in applications where safe and robust perception is critical. While existing OOD detection methods have shown progress for 2D image data, extending these to 3D environments involves unique obstacles. This paper introduces a training-free framework that leverages Vision-Language Models (VLMs) for effective OOD detection in 3D point clouds. By constructing a graph based on class prototypes and testing data, we exploit the data manifold structure to enhancing the effectiveness of VLMs for 3D OOD detection. We propose a novel Graph Score Propagation (GSP) method that incorporates prompt clustering and self-training negative prompting to improve OOD scoring with VLM. Our method is also adaptable to few-shot scenarios, providing options for practical applications. We demonstrate that GSP consistently outperforms state-of-the-art methods across synthetic and real-world datasets 3D point cloud OOD detection.

Paperid: 3888, https://arxiv.org/pdf/2506.22146.pdf

Abstract:
Despite progress in Vision-Language Models (VLMs), their capacity for visual reasoning is often limited by the binding problem: the failure to reliably associate perceptual features with their correct visual referents. This limitation underlies persistent errors in tasks such as counting, visual search, scene description, and spatial relationship understanding. A key factor is that current VLMs process visual features largely in parallel, lacking mechanisms for spatially grounded, serial attention. This paper introduces VISER (Visual Input Structure for Enhanced Reasoning), a simple yet effective intervention: augmenting visual inputs with low-level spatial structures and pairing this with a textual prompt that encourages sequential, spatially-aware parsing. We empirically demonstrate substantial performance improvements across core visual reasoning tasks. Specifically, VISER improves GPT-4o visual search accuracy by 25.00%, increases counting accuracy by 26.83%, reduces edit distance error in scene description by 0.32, and enhances performance on spatial relationship tasks by 9.50% on a 2D synthetic dataset. Furthermore, we find that the visual modification is essential for these gains; purely textual strategies, including Chain-of-Thought prompting, are insufficient and can even degrade performance. VISER enhances binding only with a single-query inference, underscoring the importance of visual input design over purely linguistically-based approaches. These findings suggest that low-level visual structuring is a powerful and underexplored direction for improving compositional visual reasoning and could serve as a general strategy for enhancing VLM performance on spatially grounded tasks.

Paperid: 3889, https://arxiv.org/pdf/2506.21892.pdf

Abstract:
As point cloud data increases in prevalence in a variety of applications, the ability to detect out-of-distribution (OOD) point cloud objects becomes critical for ensuring model safety and reliability. However, this problem remains under-explored in existing research. Inspired by success in the image domain, we propose to exploit advances in 3D vision-language models (3D VLMs) for OOD detection in point cloud objects. However, a major challenge is that point cloud datasets used to pre-train 3D VLMs are drastically smaller in size and object diversity than their image-based counterparts. Critically, they often contain exclusively computer-designed synthetic objects. This leads to a substantial domain shift when the model is transferred to practical tasks involving real objects scanned from the physical environment. In this paper, our empirical experiments show that synthetic-to-real domain shift significantly degrades the alignment of point cloud with their associated text embeddings in the 3D VLM latent space, hindering downstream performance. To address this, we propose a novel methodology called SODA which improves the detection of OOD point clouds through a neighborhood-based score propagation scheme. SODA is inference-based, requires no additional model training, and achieves state-of-the-art performance over existing approaches across datasets and problem settings.

Paperid: 3890, https://arxiv.org/pdf/2506.10085.pdf

Abstract:
We propose a test-time adaptation method that enables a progress estimation model to adapt online to the visual and temporal context of test trajectories by optimizing a learned self-supervised objective. To this end, we introduce a gradient-based meta-learning strategy to train the model on expert visual trajectories and their natural language task descriptions, such that test-time adaptation improves progress estimation relying on semantic content over temporal order. Our test-time adaptation method generalizes from a single training environment to diverse out-of-distribution tasks, environments, and embodiments, outperforming the state-of-the-art in-context learning approach using autoregressive vision-language models.

Paperid: 3891, https://arxiv.org/pdf/2506.06487.pdf

Abstract:
Zero-shot object navigation (ZSON) allows robots to find target objects in unfamiliar environments using natural language instructions, without relying on pre-built maps or task-specific training. Recent general-purpose models, such as large language models (LLMs) and vision-language models (VLMs), equip agents with semantic reasoning abilities to estimate target object locations in a zero-shot manner. However, these models often greedily select the next goal without maintaining a global understanding of the environment and are fundamentally limited in the spatial reasoning necessary for effective navigation. To overcome these limitations, we propose a novel 3D voxel-based belief map that estimates the target's prior presence distribution within a voxelized 3D space. This approach enables agents to integrate semantic priors from LLMs and visual embeddings with hierarchical spatial structure, alongside real-time observations, to build a comprehensive 3D global posterior belief of the target's location. Building on this 3D voxel map, we introduce BeliefMapNav, an efficient navigation system with two key advantages: i) grounding LLM semantic reasoning within the 3D hierarchical semantics voxel space for precise target position estimation, and ii) integrating sequential path planning to enable efficient global navigation decisions. Experiments on HM3D, MP3D, and HSSD benchmarks show that BeliefMapNav achieves state-of-the-art (SOTA) Success Rate (SR) and Success weighted by Path Length (SPL), with a notable 46.4% SPL improvement over the previous best SR method, validating its effectiveness and efficiency.

Paperid: 3892, https://arxiv.org/pdf/2505.23004.pdf

Abstract:
Multimodal Large Language Models (MLLMs) encode images into visual tokens, aligning visual and textual signals within a shared latent space to facilitate crossmodal representation learning. The CLIP model is a widely adopted foundational vision language model whose vision encoder has played a critical role in the development of MLLMs such as LLaVA. However, the CLIP vision encoder suffers from notable limitations including being constrained to only handling fixed input resolutions and a failure to produce separated embeddings for dissimilar images. Replacing the vision encoder of an existing model typically incurs substantial computational costs because such a change often necessitates retraining the entire model pipeline. In this work, we identify two factors which underlie the limitations of the CLIP vision encoder: mesoscopic bias and interpolation bias. To address these issues, we propose QLIP, a drop-in replacement for CLIP that can be seamlessly integrated with existing MLLMs with only a few lines of code and can enhance both coarse-grained and fine-grained visual understanding, without re-training. QLIP is designed around an image quadtree which replaces the standard uniform grid patches with a novel content aware patchification. Our experimental results demonstrate that QLIP improves the general visual question answering accuracy of the LLaVA v1.5 model series across various model sizes--without requiring retraining or fine-tuning of the full MLLM. Notably, QLIP boosts detailed understanding performance on the challenging $V^{\ast}$ benchmark by up to 13.6 percent.

Paperid: 3893, https://arxiv.org/pdf/2505.22627.pdf

Abstract:
While densely annotated image captions significantly facilitate the learning of robust vision-language alignment, methodologies for systematically optimizing human annotation efforts remain underexplored. We introduce Chain-of-Talkers (CoTalk), an AI-in-the-loop methodology designed to maximize the number of annotated samples and improve their comprehensiveness under fixed budget constraints (e.g., total human annotation time). The framework is built upon two key insights. First, sequential annotation reduces redundant workload compared to conventional parallel annotation, as subsequent annotators only need to annotate the ``residual'' -- the missing visual information that previous annotations have not covered. Second, humans process textual input faster by reading while outputting annotations with much higher throughput via talking; thus a multimodal interface enables optimized efficiency. We evaluate our framework from two aspects: intrinsic evaluations that assess the comprehensiveness of semantic units, obtained by parsing detailed captions into object-attribute trees and analyzing their effective connections; extrinsic evaluation measures the practical usage of the annotated captions in facilitating vision-language alignment. Experiments with eight participants show our Chain-of-Talkers (CoTalk) improves annotation speed (0.42 vs. 0.30 units/sec) and retrieval performance (41.13% vs. 40.52%) over the parallel method.

Paperid: 3894, https://arxiv.org/pdf/2505.20680.pdf

Abstract:
Continual learning (CL) enables deep networks to acquire new knowledge while avoiding catastrophic forgetting. The powerful generalization ability of pre-trained models (PTMs), such as the Contrastive Language-Image Pre-training (CLIP) model, has inspired a range of CL methods targeting new and specialized tasks, providing rich multi-modal embeddings that support lightweight, incremental prompt tuning. Existing methods often rely on complex designs built upon specific assumptions, such as intricate regularization schemes for prompt pools, specialized routing mechanisms, or multi-stage incrementations, that introduce additional-and possibly unnecessary-complexity, underutilizing CLIP's intrinsic capabilities. In this paper, we propose a concise CL approach for CLIP based on incremental prompt tuning that fully exploits its multi-modal structure and the stability of textual representations. Our method, Textual Prototype-guided Prompt Tuning (TPPT), introduces textual prototypes not merely as static classifiers, as in existing methods, but as stable anchors to guide the learning of visual prompts, thereby shaping the embedding space (i.e., TPPT-V). We show that our bidirectional supervision strategy enables more effective learning of new knowledge while reducing forgetting. To further close the vision-language gap during CL, we jointly optimizes visual and textual prompts (i.e., TPPT-VT). We also introduce a relational diversity regularization on the textual anchors to prevent embedding space collapse and mitigate correlated forgetting. Extensive experiments and analyses demonstrate the effectiveness of our proposed approach, highlighting the benefits of leveraging CLIP's intrinsic guidance for continual adaptation.

Paperid: 3895, https://arxiv.org/pdf/2505.20021.pdf

Abstract:
Recent Vision-Language Models (VLMs) have demonstrated impressive multimodal comprehension and reasoning capabilities, yet they often struggle with trivially simple visual tasks. In this work, we focus on the domain of basic 2D Euclidean geometry and systematically categorize the fundamental, indivisible visual perception skills, which we refer to as atomic visual skills. We then introduce the Atomic Visual Skills Dataset (AVSD) for evaluating VLMs on the atomic visual skills. Using AVSD, we benchmark state-of-the-art VLMs and find that they struggle with these tasks, despite being trivial for adult humans. Our findings highlight the need for purpose-built datasets to train and evaluate VLMs on atomic, rather than composite, visual perception tasks.

Paperid: 3896, https://arxiv.org/pdf/2505.18969.pdf

Abstract:
Cognitive control refers to the ability to flexibly coordinate thought and action in pursuit of internal goals. A standard method for assessing cognitive control involves conflict tasks that contrast congruent and incongruent trials, measuring the ability to prioritize relevant information while suppressing interference. We evaluate 108 vision-language models on three classic conflict tasks and their more demanding "squared" variants across 2,220 trials. Model performance corresponds closely to human behavior under resource constraints and reveals individual differences. These results indicate that some form of human-like executive function have emerged in current multi-modal foundational models.

Paperid: 3897, https://arxiv.org/pdf/2505.18770.pdf

Abstract:
Domain generalization (DG) aims to learn a model using data from one or multiple related but distinct source domains that can generalize well to unseen out-of-distribution target domains. Inspired by the success of large pre-trained vision-language models (VLMs), prompt tuning has emerged as an effective generalization strategy. However, it often struggles to capture domain-specific features due to its reliance on manually or fixed prompt inputs. Recently, some prompt generation methods have addressed this limitation by dynamically generating instance-specific and domain-specific prompts for each input, enriching domain information and demonstrating potential for enhanced generalization. Through further investigation, we identify a notable issue in existing prompt generation methods: the same input often yields significantly different and suboptimal prompts across different random seeds, a phenomenon we term Prompt Variability. To address this, we introduce negative learning into the prompt generation process and propose Dual-Path Stable Soft Prompt Generation (DPSPG), a transformer-based framework designed to improve both the stability and generalization of prompts. Specifically, DPSPG incorporates a complementary prompt generator to produce negative prompts, thereby reducing the risk of introducing misleading information. Both theoretical and empirical analyses demonstrate that negative learning leads to more robust and effective prompts by increasing the effective margin and reducing the upper bound of the gradient norm. Extensive experiments on five DG benchmark datasets show that DPSPG consistently outperforms state-of-the-art methods while maintaining prompt stability.

Paperid: 3898, https://arxiv.org/pdf/2505.15529.pdf

Abstract:
Current vision-language models (VLMs) have demonstrated remarkable capabilities across diverse video understanding applications. Designing VLMs for video inputs requires effectively modeling the temporal dimension (i.e. capturing dependencies across frames) and balancing the processing of short and long videos. Specifically, short videos demand preservation of fine-grained details, whereas long videos require strategic compression of visual information to handle extensive temporal contexts efficiently. However, our empirical analysis reveals a critical limitation: most existing VLMs suffer severe performance degradation in long video understanding tasks when compressing visual tokens below a quarter of their original visual tokens. To enable more effective modeling of both short and long video inputs, we propose Clapper, a method that utilizes a slow-fast strategy for video representation and introduces a novel module named TimePerceiver for efficient temporal-spatial encoding within existing VLM backbones. By using our method, we achieves 13x compression of visual tokens per frame (averaging 61 tokens/frame) without compromising QA accuracy. In our experiments, Clapper achieves 62.0% on VideoMME, 69.8% on MLVU, and 67.4% on TempCompass, all with fewer than 6,000 visual tokens per video. The code will be publicly available on the homepage.

Paperid: 3899, https://arxiv.org/pdf/2505.14346.pdf

Abstract:
This paper presents a novel inertial localization framework named Egocentric Action-aware Inertial Localization (EAIL), which leverages egocentric action cues from head-mounted IMU signals to localize the target individual within a 3D point cloud. Human inertial localization is challenging due to IMU sensor noise that causes trajectory drift over time. The diversity of human actions further complicates IMU signal processing by introducing various motion patterns. Nevertheless, we observe that some actions captured by the head-mounted IMU correlate with spatial environmental structures (e.g., bending down to look inside an oven, washing dishes next to a sink), thereby serving as spatial anchors to compensate for the localization drift. The proposed EAIL framework learns such correlations via hierarchical multi-modal alignment with vision-language guidance. By assuming that the 3D point cloud of the environment is available, it contrastively learns modality encoders that align short-term egocentric action cues in IMU signals with local environmental features in the point cloud. The learning process is enhanced using concurrently collected vision and language signals to improve multimodal alignment. The learned encoders are then used in reasoning the IMU data and the point cloud over time and space to perform inertial localization. Interestingly, these encoders can further be utilized to recognize the corresponding sequence of actions as a by-product. Extensive experiments demonstrate the effectiveness of the proposed framework over state-of-the-art inertial localization and inertial action recognition baselines.

Paperid: 3900, https://arxiv.org/pdf/2505.13777.pdf

Abstract:
We present Sat2Sound, a multimodal representation learning framework for soundscape mapping, designed to predict the distribution of sounds at any location on Earth. Existing methods for this task rely on satellite image and paired geotagged audio samples, which often fail to capture the diversity of sound sources at a given location. To address this limitation, we enhance existing datasets by leveraging a Vision-Language Model (VLM) to generate semantically rich soundscape descriptions for locations depicted in satellite images. Our approach incorporates contrastive learning across audio, audio captions, satellite images, and satellite image captions. We hypothesize that there is a fixed set of soundscape concepts shared across modalities. To this end, we learn a shared codebook of soundscape concepts and represent each sample as a weighted average of these concepts. Sat2Sound achieves state-of-the-art performance in cross-modal retrieval between satellite image and audio on two datasets: GeoSound and SoundingEarth. Additionally, building on Sat2Sound's ability to retrieve detailed soundscape captions, we introduce a novel application: location-based soundscape synthesis, which enables immersive acoustic experiences. Our code and models will be publicly available.

Paperid: 3901, https://arxiv.org/pdf/2505.11123.pdf

Abstract:
Diffusion policies have emerged as a mainstream paradigm for building vision-language-action (VLA) models. Although they demonstrate strong robot control capabilities, their training efficiency remains suboptimal. In this work, we identify a fundamental challenge in conditional diffusion policy training: when generative conditions are hard to distinguish, the training objective degenerates into modeling the marginal action distribution, a phenomenon we term loss collapse. To overcome this, we propose Cocos, a simple yet general solution that modifies the source distribution in the conditional flow matching to be condition-dependent. By anchoring the source distribution around semantics extracted from condition inputs, Cocos encourages stronger condition integration and prevents the loss collapse. We provide theoretical justification and extensive empirical results across simulation and real-world benchmarks. Our method achieves faster convergence and higher success rates than existing approaches, matching the performance of large-scale pre-trained VLAs using significantly fewer gradient steps and parameters. Cocos is lightweight, easy to implement, and compatible with diverse policy architectures, offering a general-purpose improvement to diffusion policy training.

Paperid: 3902, https://arxiv.org/pdf/2505.09601.pdf

Abstract:
Scaling robot learning requires vast and diverse datasets. Yet the prevailing data collection paradigm-human teleoperation-remains costly and constrained by manual effort and physical robot access. We introduce Real2Render2Real (R2R2R), a novel approach for generating robot training data without relying on object dynamics simulation or teleoperation of robot hardware. The input is a smartphone-captured scan of one or more objects and a single video of a human demonstration. R2R2R renders thousands of high visual fidelity robot-agnostic demonstrations by reconstructing detailed 3D object geometry and appearance, and tracking 6-DoF object motion. R2R2R uses 3D Gaussian Splatting (3DGS) to enable flexible asset generation and trajectory synthesis for both rigid and articulated objects, converting these representations to meshes to maintain compatibility with scalable rendering engines like IsaacLab but with collision modeling off. Robot demonstration data generated by R2R2R integrates directly with models that operate on robot proprioceptive states and image observations, such as vision-language-action models (VLA) and imitation learning policies. Physical experiments suggest that models trained on R2R2R data from a single human demonstration can match the performance of models trained on 150 human teleoperation demonstrations. Project page: https://real2render2real.com

Paperid: 3903, https://arxiv.org/pdf/2505.08234.pdf

Abstract:
As AI-generated imagery becomes ubiquitous, invisible watermarks have emerged as a primary line of defense for copyright and provenance. The newest watermarking schemes embed semantic signals - content-aware patterns that are designed to survive common image manipulations - yet their true robustness against adaptive adversaries remains under-explored. We expose a previously unreported vulnerability and introduce SemanticRegen, a three-stage, label-free attack that erases state-of-the-art semantic and invisible watermarks while leaving an image's apparent meaning intact. Our pipeline (i) uses a vision-language model to obtain fine-grained captions, (ii) extracts foreground masks with zero-shot segmentation, and (iii) inpaints only the background via an LLM-guided diffusion model, thereby preserving salient objects and style cues. Evaluated on 1,000 prompts across four watermarking systems - TreeRing, StegaStamp, StableSig, and DWT/DCT - SemanticRegen is the only method to defeat the semantic TreeRing watermark (p = 0.10 > 0.05) and reduces bit-accuracy below 0.75 for the remaining schemes, all while maintaining high perceptual quality (masked SSIM = 0.94 +/- 0.01). We further introduce masked SSIM (mSSIM) to quantify fidelity within foreground regions, showing that our attack achieves up to 12 percent higher mSSIM than prior diffusion-based attackers. These results highlight an urgent gap between current watermark defenses and the capabilities of adaptive, semantics-aware adversaries, underscoring the need for watermarking algorithms that are resilient to content-preserving regenerative attacks.

Paperid: 3904, https://arxiv.org/pdf/2505.01958.pdf

Abstract:
Large Vision-Language Models (LVLMs) demonstrate remarkable capabilities in multimodal tasks, but visual object hallucination remains a persistent issue. It refers to scenarios where models generate inaccurate visual object-related information based on the query input, potentially leading to misinformation and concerns about safety and reliability. Previous works focus on the evaluation and mitigation of visual hallucinations, but the underlying causes have not been comprehensively investigated. In this paper, we analyze each component of LLaVA-like LVLMs -- the large language model, the vision backbone, and the projector -- to identify potential sources of error and their impact. Based on our observations, we propose methods to mitigate hallucination for each problematic component. Additionally, we developed two hallucination benchmarks: QA-VisualGenome, which emphasizes attribute and relation hallucinations, and QA-FB15k, which focuses on cognition-based hallucinations.

Paperid: 3905, https://arxiv.org/pdf/2505.01050.pdf

Abstract:
Vision Large Language Models (VLLMs) are increasingly deployed to offer advanced capabilities on inputs comprising both text and images. While prior research has shown that adversarial attacks can transfer from open-source to proprietary black-box models in text-only and vision-only contexts, the extent and effectiveness of such vulnerabilities remain underexplored for VLLMs. We present a comprehensive analysis demonstrating that targeted adversarial examples are highly transferable to widely-used proprietary VLLMs such as GPT-4o, Claude, and Gemini. We show that attackers can craft perturbations to induce specific attacker-chosen interpretations of visual information, such as misinterpreting hazardous content as safe, overlooking sensitive or restricted material, or generating detailed incorrect responses aligned with the attacker's intent. Furthermore, we discover that universal perturbations -- modifications applicable to a wide set of images -- can consistently induce these misinterpretations across multiple proprietary VLLMs. Our experimental results on object recognition, visual question answering, and image captioning show that this vulnerability is common across current state-of-the-art models, and underscore an urgent need for robust mitigations to ensure the safe and secure deployment of VLLMs.

Paperid: 3906, https://arxiv.org/pdf/2505.00150.pdf

Abstract:
The rapid evolution of social media has provided enhanced communication channels for individuals to create online content, enabling them to express their thoughts and opinions. Multimodal memes, often utilized for playful or humorous expressions with visual and textual elements, are sometimes misused to disseminate hate speech against individuals or groups. While the detection of hateful memes is well-researched, developing effective methods to transform hateful content in memes remains a significant challenge. Leveraging the powerful generation and reasoning capabilities of Vision-Language Models (VLMs), we address the tasks of detecting and mitigating hateful content. This paper presents two key contributions: first, a definition-guided prompting technique for detecting hateful memes, and second, a unified framework for mitigating hateful content in memes, named UnHateMeme, which works by replacing hateful textual and/or visual components. With our definition-guided prompts, VLMs achieve impressive performance on hateful memes detection task. Furthermore, our UnHateMeme framework, integrated with VLMs, demonstrates a strong capability to convert hateful memes into non-hateful forms that meet human-level criteria for hate speech and maintain multimodal coherence between image and text. Through empirical experiments, we show the effectiveness of state-of-the-art pretrained VLMs such as LLaVA, Gemini and GPT-4o on the proposed tasks, providing a comprehensive analysis of their respective strengths and limitations for these tasks. This paper aims to shed light on important applications of VLMs for ensuring safe and respectful online environments.

Paperid: 3907, https://arxiv.org/pdf/2504.16181.pdf

Abstract:
Multimodal learning has shown promise in medical image analysis, combining complementary modalities like histology images and text. Vision-language models (VLMs) capture rich diagnostic cues but often require large paired datasets and prompt- or text-based inference, limiting their practicality due to annotation cost, privacy, and compute demands. Crucially, available free unpaired external text, like pathology reports, can still provide complementary diagnostic cues if semantically relevant content is retrievable per image. To address this, we introduce CLIP-IT, a novel framework that relies on rich unpaired text reports, eliminating paired data requirement. Specifically, CLIP-IT uses a CLIP model pre-trained on histology image-text pairs from a separate dataset to retrieve the most relevant unpaired textual report for each image in the target unimodal dataset. These reports, sourced from the same disease domain and tissue type, form pseudo-pairs that reflect shared clinical semantics rather than exact alignment. Knowledge from these texts is distilled into the vision model during training, while LoRA-based adaptation mitigates the semantic gap between unaligned modalities. At inference time, only the improved vision model is used, with minimal computational overhead, enabling efficient pairing-free multimodal deployment. Experiments on histology image datasets confirm that CLIP-IT consistently improves classification accuracy over both unimodal and multimodal CLIP-based baselines in most cases, without the burden of paired data training or inference-time complexity.

Paperid: 3908, https://arxiv.org/pdf/2504.15517.pdf

Abstract:
Recently, Transformer-based robotic manipulation methods utilize multi-view spatial representations and language instructions to learn robot motion trajectories by leveraging numerous robot demonstrations. However, the collection of robot data is extremely challenging, and existing methods lack the capability for continuous learning on new tasks with only a few demonstrations. In this paper, we formulate these challenges as the Few-Shot Action-Incremental Learning (FSAIL) task, and accordingly design a Task-prOmpt graPh evolutIon poliCy (TOPIC) to address these issues. Specifically, to address the data scarcity issue in robotic imitation learning, TOPIC learns Task-Specific Prompts (TSP) through the deep interaction of multi-modal information within few-shot demonstrations, thereby effectively extracting the task-specific discriminative information. On the other hand, to enhance the capability for continual learning on new tasks and mitigate the issue of catastrophic forgetting, TOPIC adopts a Continuous Evolution Strategy (CES). CES leverages the intrinsic relationships between tasks to construct a task relation graph, which effectively facilitates the adaptation of new tasks by reusing skills learned from previous tasks. TOPIC pioneers few-shot continual learning in the robotic manipulation task, and extensive experimental results demonstrate that TOPIC outperforms state-of-the-art baselines by over 26$\%$ in success rate, significantly enhancing the continual learning capabilities of existing Transformer-based policies.

Paperid: 3909, https://arxiv.org/pdf/2504.12316.pdf

Abstract:
Data curation plays a crucial role in training powerful Visual Language Models (VLMs). In this work, we introduce the concept of Data Metabolism and present our data-centric framework to build VLMs throughout the development lifecycle. Starting from a standard model architecture, we discuss and provide insights into two crucial development steps: data curation and iteration, forming a closed-loop system that continuously improves model performance. We show a detailed codebook on how to process existing massive datasets and build user-specific data flywheel. As a demonstration, we release a VLM, named Capybara-VL, which excels in typical multimodal tasks (e.g. , visual question answering, scientific reasoning, and text-rich tasks). Despite its relatively compact size, Capybara-VL surpasses several open-source models that are up to 10 times larger in size. Moreover, it achieves results that are on par with those of several leading proprietary models, demonstrating its remarkable competitiveness. These results highlight the power of our data-centric framework and the potential of training smaller and more efficient VLMs.

Paperid: 3910, https://arxiv.org/pdf/2504.10458.pdf

Abstract:
Existing efforts in building Graphical User Interface (GUI) agents largely rely on the training paradigm of supervised fine-tuning on Large Vision-Language Models (LVLMs). However, this approach not only demands extensive amounts of training data but also struggles to effectively understand GUI screenshots and generalize to unseen interfaces. The issue significantly limits its application in real-world scenarios, especially for high-level tasks. Inspired by Reinforcement Fine-Tuning (RFT) in large reasoning models (e.g., DeepSeek-R1), which efficiently enhances the problem-solving capabilities of large language models in real-world settings, we propose \name, the first reinforcement learning framework designed to enhance the GUI capabilities of LVLMs in high-level real-world task scenarios, through unified action space rule modeling. By leveraging a small amount of carefully curated high-quality data across multiple platforms (including Windows, Linux, MacOS, Android, and Web) and employing policy optimization algorithms such as Group Relative Policy Optimization (GRPO) to update the model, \name achieves superior performance using only 0.02\% of the data (3K vs. 13M) compared to previous state-of-the-art methods like OS-Atlas across eight benchmarks spanning three different platforms (mobile, desktop, and web). These results demonstrate the immense potential of reinforcement learning based on unified action space rule modeling in improving the execution capabilities of LVLMs for real-world GUI agent tasks.

Paperid: 3911, https://arxiv.org/pdf/2504.03770.pdf

Abstract:
Multimodal large language models (MLLMs) excel in vision-language tasks but also pose significant risks of generating harmful content, particularly through jailbreak attacks. Jailbreak attacks refer to intentional manipulations that bypass safety mechanisms in models, leading to the generation of inappropriate or unsafe content. Detecting such attacks is critical to ensuring the responsible deployment of MLLMs. Existing jailbreak detection methods face three primary challenges: (1) Many rely on model hidden states or gradients, limiting their applicability to white-box models, where the internal workings of the model are accessible; (2) They involve high computational overhead from uncertainty-based analysis, which limits real-time detection, and (3) They require fully labeled harmful datasets, which are often scarce in real-world settings. To address these issues, we introduce a test-time adaptive framework called JAILDAM. Our method leverages a memory-based approach guided by policy-driven unsafe knowledge representations, eliminating the need for explicit exposure to harmful data. By dynamically updating unsafe knowledge during test-time, our framework improves generalization to unseen jailbreak strategies while maintaining efficiency. Experiments on multiple VLM jailbreak benchmarks demonstrate that JAILDAM delivers state-of-the-art performance in harmful content detection, improving both accuracy and speed.

Paperid: 3912, https://arxiv.org/pdf/2504.03164.pdf

Abstract:
Recent advancements in Vision-Language Models (VLMs) have demonstrated strong potential for autonomous driving tasks. However, their spatial understanding and reasoning-key capabilities for autonomous driving-still exhibit significant limitations. Notably, none of the existing benchmarks systematically evaluate VLMs' spatial reasoning capabilities in driving scenarios. To fill this gap, we propose NuScenes-SpatialQA, the first large-scale ground-truth-based Question-Answer (QA) benchmark specifically designed to evaluate the spatial understanding and reasoning capabilities of VLMs in autonomous driving. Built upon the NuScenes dataset, the benchmark is constructed through an automated 3D scene graph generation pipeline and a QA generation pipeline. The benchmark systematically evaluates VLMs' performance in both spatial understanding and reasoning across multiple dimensions. Using this benchmark, we conduct extensive experiments on diverse VLMs, including both general and spatial-enhanced models, providing the first comprehensive evaluation of their spatial capabilities in autonomous driving. Surprisingly, the experimental results show that the spatial-enhanced VLM outperforms in qualitative QA but does not demonstrate competitiveness in quantitative QA. In general, VLMs still face considerable challenges in spatial understanding and reasoning.

Paperid: 3913, https://arxiv.org/pdf/2504.01001.pdf

Abstract:
As language models improve and become capable of performing more complex tasks across modalities, evaluating them automatically becomes increasingly challenging. Developing strong and robust task-specific automatic metrics gets harder, and human-annotated test sets -- which are expensive to create -- saturate more quickly. A compelling alternative is to design reliable strategies to automate the creation of test data and evaluation, but previous attempts either rely on pre-existing data, or focus solely on individual tasks. We present Zero-shot Benchmarking (ZSB), a framework for creating high-quality benchmarks for any task by leveraging language models for both synthetic test data creation and evaluation. ZSB is simple and flexible: it requires only the creation of a prompt for data generation and one for evaluation; it is scalable to tasks and languages where collecting real-world data is costly or impractical; it is model-agnostic, allowing the creation of increasingly challenging benchmarks as models improve. To assess the effectiveness of our framework, we create benchmarks for five text-only tasks and a multi-modal one: general capabilities in four languages (English, Chinese, French, and Korean), translation, and general vision-language capabilities in English. We then rank a broad range of open and closed systems on our benchmarks. ZSB rankings consistently correlate strongly with human rankings, outperforming widely-adopted standard benchmarks. Through ablations, we find that strong benchmarks can be created with open models, and that judge model size and dataset variety are crucial drivers of performance. We release all our benchmarks, and code to reproduce our experiments and to produce new benchmarks.

Paperid: 3914, https://arxiv.org/pdf/2503.24182.pdf

Abstract:
Contrastive Language-Image Pretraining (CLIP) has achieved remarkable success in cross-modal tasks such as zero-shot image classification and text-image retrieval by effectively aligning visual and textual representations. However, the theoretical foundations underlying CLIP's strong generalization remain unclear. In this work, we address this gap by proposing the Cross-modal Information Bottleneck (CIB) framework. CIB offers a principled interpretation of CLIP's contrastive learning objective as an implicit Information Bottleneck optimization. Under this view, the model maximizes shared cross-modal information while discarding modality-specific redundancies, thereby preserving essential semantic alignment across modalities. Building on this insight, we introduce a Cross-modal Information Bottleneck Regularization (CIBR) method that explicitly enforces these IB principles during training. CIBR introduces a penalty term to discourage modality-specific redundancy, thereby enhancing semantic alignment between image and text features. We validate CIBR on extensive vision-language benchmarks, including zero-shot classification across seven diverse image datasets and text-image retrieval on MSCOCO and Flickr30K. The results show consistent performance gains over standard CLIP. These findings provide the first theoretical understanding of CLIP's generalization through the IB lens. They also demonstrate practical improvements, offering guidance for future cross-modal representation learning.

Paperid: 3915, https://arxiv.org/pdf/2503.21757.pdf

Abstract:
In this work, we aim to compress the vision tokens of a Large Vision Language Model (LVLM) into a representation that is simultaneously suitable for (a) generative and (b) discriminative tasks, (c) is nearly lossless, and (d) is storage-efficient. We propose a novel compression approach, called Fwd2Bot, that uses the LVLM itself to compress the visual information in a task-agnostic manner. At the core of Fwd2bot there exists a "double-forward pass" training strategy, whereby, during the first forward pass, the LLM (of the LVLM) creates a bottleneck by condensing the visual information into a small number of summary tokens. Then, using the same LLM, the second forward pass processes the language instruction(s) alongside the summary tokens, used as a direct replacement for the image ones. The training signal is provided by two losses: an autoregressive one applied after the second pass that provides a direct optimization objective for compression, and a contrastive loss, applied after the first pass, that further boosts the representation strength, especially for discriminative tasks. The training is further enhanced by stage-specific adapters. We accompany the proposed method by an in-depth ablation study. Overall, Fwd2Bot results in highly-informative compressed representations suitable for both generative and discriminative tasks. For generative tasks, we offer a 2x higher compression rate without compromising the generative capabilities, setting a new state-of-the-art result. For discriminative tasks, we set a new state-of-the-art on image retrieval and compositionality.

Paperid: 3916, https://arxiv.org/pdf/2503.20960.pdf

Abstract:
Automated frame analysis of political communication is a popular task in computational social science that is used to study how authors select aspects of a topic to frame its reception. So far, such studies have been narrow, in that they use a fixed set of pre-defined frames and focus only on the text, ignoring the visual contexts in which those texts appear. Especially for framing in the news, this leaves out valuable information about editorial choices, which include not just the written article but also accompanying photographs. To overcome such limitations, we present a method for conducting multi-modal, multi-label framing analysis at scale using large (vision-) language models. Grounding our work in framing theory, we extract latent meaning embedded in images used to convey a certain point and contrast that to the text by comparing the respective frames used. We also identify highly partisan framing of topics with issue-specific frame analysis found in prior qualitative work. We demonstrate a method for doing scalable integrative framing analysis of both text and image in news, providing a more complete picture for understanding media bias.

Paperid: 3917, https://arxiv.org/pdf/2503.15108.pdf

Abstract:
While Large Language Models (LLMs) excel at reasoning on text and Vision-Language Models (VLMs) are highly effective for visual perception, applying those models for visual instruction-based planning remains a widely open problem. In this paper, we introduce VIPER, a novel framework for multimodal instruction-based planning that integrates VLM-based perception with LLM-based reasoning. Our approach uses a modular pipeline where a frozen VLM generates textual descriptions of image observations, which are then processed by an LLM policy to predict actions based on the task goal. We fine-tune the reasoning module using behavioral cloning and reinforcement learning, improving our agent's decision-making capabilities. Experiments on the ALFWorld benchmark show that VIPER significantly outperforms state-of-the-art visual instruction-based planners while narrowing the gap with purely text-based oracles. By leveraging text as an intermediate representation, VIPER also enhances explainability, paving the way for a fine-grained analysis of perception and reasoning components.

Paperid: 3918, https://arxiv.org/pdf/2503.04504.pdf

Abstract:
Video anomaly detection (VAD) is crucial for video analysis and surveillance in computer vision. However, existing VAD models rely on learned normal patterns, which makes them difficult to apply to diverse environments. Consequently, users should retrain models or develop separate AI models for new environments, which requires expertise in machine learning, high-performance hardware, and extensive data collection, limiting the practical usability of VAD. To address these challenges, this study proposes customizable video anomaly detection (C-VAD) technique and the AnyAnomaly model. C-VAD considers user-defined text as an abnormal event and detects frames containing a specified event in a video. We effectively implemented AnyAnomaly using a context-aware visual question answering without fine-tuning the large vision language model. To validate the effectiveness of the proposed model, we constructed C-VAD datasets and demonstrated the superiority of AnyAnomaly. Furthermore, our approach showed competitive results on VAD benchmarks, achieving state-of-the-art performance on UBnormal and UCF-Crime and surpassing other methods in generalization across all datasets. Our code is available online at github.com/SkiddieAhn/Paper-AnyAnomaly.

Paperid: 3919, https://arxiv.org/pdf/2503.04417.pdf

Abstract:
Creating digital models using Computer Aided Design (CAD) is a process that requires in-depth expertise. In industrial product development, this process typically involves entire teams of engineers, spanning requirements engineering, CAD itself, and quality assurance. We present an approach that mirrors this team structure with a Vision Language Model (VLM)-based Multi Agent System, with access to parametric CAD tooling and tool documentation. Combining agents for requirements engineering, CAD engineering, and vision-based quality assurance, a model is generated automatically from sketches and/ or textual descriptions. The resulting model can be refined collaboratively in an iterative validation loop with the user. Our approach has the potential to increase the effectiveness of design processes, both for industry experts and for hobbyists who create models for 3D printing. We demonstrate the potential of the architecture at the example of various design tasks and provide several ablations that show the benefits of the architecture's individual components.

Paperid: 3920, https://arxiv.org/pdf/2503.04144.pdf

Abstract:
Text-based person retrieval (TPR) has gained significant attention as a fine-grained and challenging task that closely aligns with practical applications. Tailoring CLIP to person domain is now a emerging research topic due to the abundant knowledge of vision-language pretraining, but challenges still remain during fine-tuning: (i) Previous full-model fine-tuning in TPR is computationally expensive and prone to overfitting.(ii) Existing parameter-efficient transfer learning (PETL) for TPR lacks of fine-grained feature extraction. To address these issues, we propose Domain-Aware Mixture-of-Adapters (DM-Adapter), which unifies Mixture-of-Experts (MOE) and PETL to enhance fine-grained feature representations while maintaining efficiency. Specifically, Sparse Mixture-of-Adapters is designed in parallel to MLP layers in both vision and language branches, where different experts specialize in distinct aspects of person knowledge to handle features more finely. To promote the router to exploit domain information effectively and alleviate the routing imbalance, Domain-Aware Router is then developed by building a novel gating function and injecting learnable domain-aware prompts. Extensive experiments show that our DM-Adapter achieves state-of-the-art performance, outperforming previous methods by a significant margin.

Paperid: 3921, https://arxiv.org/pdf/2503.01238.pdf

Abstract:
Machine learning for robotics promises to unlock generalization to novel tasks and environments. Guided by this promise, many recent works have focused on scaling up robot data collection and developing larger, more expressive policies to achieve this. But how do we measure progress towards this goal of policy generalization in practice? Evaluating and quantifying generalization is the Wild West of modern robotics, with each work proposing and measuring different types of generalization in their own, often difficult to reproduce, settings. In this work, our goal is (1) to outline the forms of generalization we believe are important in robot manipulation in a comprehensive and fine-grained manner, and (2) to provide reproducible guidelines for measuring these notions of generalization. We first propose STAR-Gen, a taxonomy of generalization for robot manipulation structured around visual, semantic, and behavioral generalization. We discuss how our taxonomy encompasses most prior notions of generalization in robotics. Next, we instantiate STAR-Gen with a concrete real-world benchmark based on the widely-used Bridge V2 dataset. We evaluate a variety of state-of-the-art models on this benchmark to demonstrate the utility of our taxonomy in practice. Our taxonomy of generalization can yield many interesting insights into existing models: for example, we observe that current vision-language-action models struggle with various types of semantic generalization, despite the promise of pre-training on internet-scale language datasets. We believe STAR-Gen and our guidelines can improve the dissemination and evaluation of progress towards generalization in robotics, which we hope will guide model design and future data collection efforts. We provide videos and demos at our website stargen-taxonomy.github.io.

Paperid: 3922, https://arxiv.org/pdf/2502.11073.pdf

Abstract:
Hateful meme detection presents a significant challenge as a multimodal task due to the complexity of interpreting implicit hate messages and contextual cues within memes. Previous approaches have fine-tuned pre-trained vision-language models (PT-VLMs), leveraging the knowledge they gained during pre-training and their attention mechanisms to understand meme content. However, the reliance of these models on implicit knowledge and complex attention mechanisms renders their decisions difficult to explain, which is crucial for building trust in meme classification. In this paper, we introduce IntMeme, a novel framework that leverages Large Multimodal Models (LMMs) for hateful meme classification with explainable decisions. IntMeme addresses the dual challenges of improving both accuracy and explainability in meme moderation. The framework uses LMMs to generate human-like, interpretive analyses of memes, providing deeper insights into multimodal content and context. Additionally, it uses independent encoding modules for both memes and their interpretations, which are then combined to enhance classification performance. Our approach addresses the opacity and misclassification issues associated with PT-VLMs, optimizing the use of LMMs for hateful meme detection. We demonstrate the effectiveness of IntMeme through comprehensive experiments across three datasets, showcasing its superiority over state-of-the-art models.

Paperid: 3923, https://arxiv.org/pdf/2502.10273.pdf

Abstract:
Perceptual constancy is the ability to maintain stable perceptions of objects despite changes in sensory input, such as variations in distance, angle, or lighting. This ability is crucial for recognizing visual information in a dynamic world, making it essential for Vision-Language Models (VLMs). However, whether VLMs are currently and theoretically capable of mastering this ability remains underexplored. In this study, we evaluated 33 VLMs using 253 experiments across three domains: color, size, and shape constancy. The experiments included single-image and video adaptations of classic cognitive tasks, along with novel tasks in in-the-wild conditions, to evaluate the models' recognition of object properties under varying conditions. We found significant variability in VLM performance, with models performance in shape constancy clearly dissociated from that of color and size constancy.

Paperid: 3924, https://arxiv.org/pdf/2502.09598.pdf

Abstract:
The continuous operation of Earth-orbiting satellites generates vast and ever-growing archives of Remote Sensing (RS) images. Natural language presents an intuitive interface for accessing, querying, and interpreting the data from such archives. However, existing Vision-Language Models (VLMs) are predominantly trained on web-scraped, noisy image-text data, exhibiting limited exposure to the specialized domain of RS. This deficiency results in poor performance on RS-specific tasks, as commonly used datasets often lack detailed, scientifically accurate textual descriptions and instead emphasize solely on attributes like date and location. To bridge this critical gap, we introduce GAIA, a novel dataset designed for multi-scale, multi-sensor, and multi-modal RS image analysis. GAIA comprises of 205,150 meticulously curated RS image-text pairs, representing a diverse range of RS modalities associated to different spatial resolutions. Unlike existing vision-language datasets in RS, GAIA specifically focuses on capturing a diverse range of RS applications, providing unique information about environmental changes, natural disasters, and various other dynamic phenomena. The dataset provides a spatially and temporally balanced distribution, spanning across the globe, covering the last 25 years with a balanced temporal distribution of observations. GAIA's construction involved a two-stage process: (1) targeted web-scraping of images and accompanying text from reputable RS-related sources, and (2) generation of five high-quality, scientifically grounded synthetic captions for each image using carefully crafted prompts that leverage the advanced vision-language capabilities of GPT-4o. Our extensive experiments, including fine-tuning of CLIP and BLIP2 models, demonstrate that GAIA significantly improves performance on RS image classification, cross-modal retrieval and image captioning tasks.

Paperid: 3925, https://arxiv.org/pdf/2502.04223.pdf

Abstract:
Optical Character Recognition (OCR) technology is widely used to extract text from images of documents, facilitating efficient digitization and data retrieval. However, merely extracting text is insufficient when dealing with complex documents. Fully comprehending such documents requires an understanding of their structure -- including formatting, formulas, tables, and the reading order of multiple blocks and columns across multiple pages -- as well as semantic information for detecting elements like footnotes and image captions. This comprehensive understanding is crucial for downstream tasks such as retrieval, document question answering, and data curation for training Large Language Models (LLMs) and Vision Language Models (VLMs). To address this, we introduce Ãclair, a general-purpose text-extraction tool specifically designed to process a wide range of document types. Given an image, Ãclair is able to extract formatted text in reading order, along with bounding boxes and their corresponding semantic classes. To thoroughly evaluate these novel capabilities, we introduce our diverse human-annotated benchmark for document-level OCR and semantic classification. Ãclair achieves state-of-the-art accuracy on this benchmark, outperforming other methods across key metrics. Additionally, we evaluate Ãclair on established benchmarks, demonstrating its versatility and strength across several evaluation standards.

Paperid: 3926, https://arxiv.org/pdf/2502.03214.pdf

Abstract:
Vision-Language Models (VLMs) are known to struggle with spatial reasoning and visual alignment. To help overcome these limitations, we introduce iVISPAR, an interactive multi-modal benchmark designed to evaluate the spatial reasoning capabilities of VLMs acting as agents. iVISPAR is based on a variant of the sliding tile puzzle-a classic problem that demands logical planning, spatial awareness, and multi-step reasoning. The benchmark supports visual 2D, 3D, and text-based input modalities, enabling comprehensive assessments of VLMs' planning and reasoning skills. We evaluate a broad suite of state-of-the-art open-source and closed-source VLMs, comparing their performance while also providing optimal path solutions and a human baseline to assess the task's complexity and feasibility for humans. Results indicate that while some VLMs perform well on simple spatial tasks, they encounter difficulties with more complex configurations and problem properties. Notably, while VLMs generally perform better in 2D vision compared to 3D or text-based representations, they consistently fall short of human performance, illustrating the persistent challenge of visual alignment. This highlights critical gaps in current VLM capabilities, highlighting their limitations in achieving human-level cognition.

Paperid: 3927, https://arxiv.org/pdf/2501.19184.pdf

Abstract:
Visual object counting has recently shifted towards class-agnostic counting (CAC), which addresses the challenge of counting objects across arbitrary categories -- a crucial capability for flexible and generalizable counting systems. Unlike humans, who effortlessly identify and count objects from diverse categories without prior knowledge, most existing counting methods are restricted to enumerating instances of known classes, requiring extensive labeled datasets for training and struggling in open-vocabulary settings. In contrast, CAC aims to count objects belonging to classes never seen during training, operating in a few-shot setting. In this paper, we present the first comprehensive review of CAC methodologies. We propose a taxonomy to categorize CAC approaches into three paradigms based on how target object classes can be specified: reference-based, reference-less, and open-world text-guided. Reference-based approaches achieve state-of-the-art performance by relying on exemplar-guided mechanisms. Reference-less methods eliminate exemplar dependency by leveraging inherent image patterns. Finally, open-world text-guided methods use vision-language models, enabling object class descriptions via textual prompts, offering a flexible and promising solution. Based on this taxonomy, we provide an overview of the architectures of 29 CAC approaches and report their results on gold-standard benchmarks. We compare their performance and discuss their strengths and limitations. Specifically, we present results on the FSC-147 dataset, setting a leaderboard using gold-standard metrics, and on the CARPK dataset to assess generalization capabilities. Finally, we offer a critical discussion of persistent challenges, such as annotation dependency and generalization, alongside future directions. We believe this survey will be a valuable resource, showcasing CAC advancements and guiding future research.

Paperid: 3928, https://arxiv.org/pdf/2501.06655.pdf

Abstract:
RLHF techniques like DPO can significantly improve the generation quality of text-to-image diffusion models. However, these methods optimize for a single reward that aligns model generation with population-level preferences, neglecting the nuances of individual users' beliefs or values. This lack of personalization limits the efficacy of these models. To bridge this gap, we introduce PPD, a multi-reward optimization objective that aligns diffusion models with personalized preferences. With PPD, a diffusion model learns the individual preferences of a population of users in a few-shot way, enabling generalization to unseen users. Specifically, our approach (1) leverages a vision-language model (VLM) to extract personal preference embeddings from a small set of pairwise preference examples, and then (2) incorporates the embeddings into diffusion models through cross attention. Conditioning on user embeddings, the text-to-image models are fine-tuned with the DPO objective, simultaneously optimizing for alignment with the preferences of multiple users. Empirical results demonstrate that our method effectively optimizes for multiple reward functions and can interpolate between them during inference. In real-world user scenarios, with as few as four preference examples from a new user, our approach achieves an average win rate of 76\% over Stable Cascade, generating images that more accurately reflect specific user preferences.

Paperid: 3929, https://arxiv.org/pdf/2412.20742.pdf

Abstract:
The domain gap between remote sensing imagery and natural images has recently received widespread attention and Vision-Language Models (VLMs) have demonstrated excellent generalization performance in remote sensing multimodal tasks. However, current research is still limited in exploring how remote sensing VLMs handle different types of visual inputs. To bridge this gap, we introduce \textbf{UniRS}, the first vision-language model \textbf{uni}fying multi-temporal \textbf{r}emote \textbf{s}ensing tasks across various types of visual input. UniRS supports single images, dual-time image pairs, and videos as input, enabling comprehensive remote sensing temporal analysis within a unified framework. We adopt a unified visual representation approach, enabling the model to accept various visual inputs. For dual-time image pair tasks, we customize a change extraction module to further enhance the extraction of spatiotemporal features. Additionally, we design a prompt augmentation mechanism tailored to the model's reasoning process, utilizing the prior knowledge of the general-purpose VLM to provide clues for UniRS. To promote multi-task knowledge sharing, the model is jointly fine-tuned on a mixed dataset. Experimental results show that UniRS achieves state-of-the-art performance across diverse tasks, including visual question answering, change captioning, and video scene classification, highlighting its versatility and effectiveness in unifying these multi-temporal remote sensing tasks. Our code and dataset will be released soon.

Paperid: 3930, https://arxiv.org/pdf/2412.06172.pdf

Abstract:
Many researchers collect data from the internet through crowd-sourcing or web crawling to alleviate the data-hungry challenge associated with cross-modal matching. Although such practice does not require expensive annotations, it inevitably introduces mismatched pairs and results in a noisy correspondence problem. Current approaches leverage the memorization effect of deep neural networks to distinguish noise and perform re-weighting. However, briefly lowering the weight of noisy pairs cannot eliminate the negative impact of noisy correspondence in the training process. In this paper, we propose a novel self-drop and dual-weight approach, which achieves elaborate data processing by qua-partitioning the data. Specifically, our approach partitions all data into four types: clean and significant, clean yet insignificant, vague, and noisy. We analyze the effect of noisy and clean data pairs and find that for vision-language pre-training models, a small number of clean samples is more valuable than a majority of noisy ones. Based on this observation, we employ self-drop to discard noisy samples to effectively mitigate the impact of noise. In addition, we adopt a dual-weight strategy to ensure that the model focuses more on significant samples while appropriately leveraging vague samples. Compared to the prior works, our approach is more robust and demonstrates relatively more stable performance on noisy datasets, especially under a high noise ratio. Extensive experiments on three widely used datasets, including Flickr30K, MS-COCO, and Conceptual Captions, validate the effectiveness of our approach.

Paperid: 3931, https://arxiv.org/pdf/2412.05685.pdf

Abstract:
Visual-textual inconsistency (VTI) evaluation plays a crucial role in cleansing vision-language data. Its main challenges stem from the high variety of image captioning datasets, where differences in content can create a range of inconsistencies (\eg, inconsistencies in scene, entities, entity attributes, entity numbers, entity interactions). Moreover, variations in caption length can introduce inconsistencies at different levels of granularity as well. To tackle these challenges, we design an adaptive evaluation framework, called Hierarchical and Multi-Grained Inconsistency Evaluation (HMGIE), which can provide multi-grained evaluations covering both accuracy and completeness for various image-caption pairs. Specifically, the HMGIE framework is implemented by three consecutive modules. Firstly, the semantic graph generation module converts the image caption to a semantic graph for building a structural representation of all involved semantic items. Then, the hierarchical inconsistency evaluation module provides a progressive evaluation procedure with a dynamic question-answer generation and evaluation strategy guided by the semantic graph, producing a hierarchical inconsistency evaluation graph (HIEG). Finally, the quantitative evaluation module calculates the accuracy and completeness scores based on the HIEG, followed by a natural language explanation about the detection results. Moreover, to verify the efficacy and flexibility of the proposed framework on handling different image captioning datasets, we construct MVTID, an image-caption dataset with diverse types and granularities of inconsistencies. Extensive experiments on MVTID and other benchmark datasets demonstrate the superior performance of the proposed HMGIE to current state-of-the-art methods.

Paperid: 3932, https://arxiv.org/pdf/2412.04378.pdf

Abstract:
Contrastively-trained Vision-Language Models (VLMs) like CLIP have become the de facto approach for discriminative vision-language representation learning. However, these models have limited language understanding, often exhibiting a "bag of words" behavior. At the same time, Large Vision-Language Models (LVLMs), which combine vision encoders with LLMs, have been shown to be capable of detailed vision-language reasoning, yet their autoregressive nature renders them less suitable for discriminative tasks. In this work, we propose to combine "the best of both worlds": a new training approach for discriminative fine-tuning of LVLMs that results in strong discriminative and compositional capabilities. Essentially, our approach converts a generative LVLM into a discriminative one, unlocking its capability for powerful image-text discrimination combined with enhanced language understanding. Our contributions include (1) a carefully designed training/optimization framework that utilizes image-text pairs of variable length and granularity for training the model with both contrastive and next-token prediction losses. This is accompanied by ablation studies that justify the necessity of our framework's components; (2) a parameter-efficient adaptation method using a combination of soft prompting and LoRA adapters; (3) significant improvements over state-of-the-art CLIP-like models of similar size, including standard image-text retrieval benchmarks and notable gains in compositionality.

Paperid: 3933, https://arxiv.org/pdf/2412.02193.pdf

Abstract:
Spatial reasoning is a fundamental aspect of human cognition, enabling intuitive understanding and manipulation of objects in three-dimensional space. While foundation models demonstrate remarkable performance on some benchmarks, they still struggle with 3D reasoning tasks like arranging objects in space according to open-ended language instructions, particularly in dense and physically constrained environments. We introduce LayoutVLM, a framework and scene layout representation that exploits the semantic knowledge of Vision-Language Models (VLMs) and supports differentiable optimization to ensure physical plausibility. LayoutVLM employs VLMs to generate two mutually reinforcing representations from visually marked images, and a self-consistent decoding process to improve VLMs spatial planning. Our experiments show that LayoutVLM addresses the limitations of existing LLM and constraint-based approaches, producing physically plausible 3D layouts better aligned with the semantic intent of input language instructions. We also demonstrate that fine-tuning VLMs with the proposed scene layout representation extracted from existing scene datasets can improve their reasoning performance.

Paperid: 3934, https://arxiv.org/pdf/2412.01004.pdf

Abstract:
Continual learning (CL) aims to accumulate knowledge from sequential data and task streams. Leveraging their strong generalization and flexibility, pre-trained vision-language embedding models such as CLIP (Contrastive Language-Image Pre-training) have been widely adopted and validated in CL. In addition to learning new knowledge, we investigate whether the pre-trained knowledge in CLIP, can be retained or even enhanced, in CL, while incorporating new knowledge from a data stream. Existing CL methods primarily focus on continual downstream adaptation using components isolated from the pre-trained model (PTM), increasing inference complexity and limiting improvements to the PTM itself; some also retain knowledge by relying on additional reference data, resulting in high training costs. To address these limitations, we propose a universal and efficient CL approach for CLIP based on Dynamic Rank-Selective LoRA (CoDyRA), which directly improves the PTMs while preserving the existing knowledge from both pre-training and CL. By analyzing how LoRA rank and placement affect learning and forgetting in CL, we design CoDyRA that adaptively performs rank-minimized parameter updates in different modules, based on their importance to the current data. This ensures a balance between knowledge acquisition (plasticity) and forgetting mitigation (stability). Our method operates without explicit domain or distribution prediction and does not rely on reference data, enabling seamless task integration while maintaining pre-trained capabilities. Moreover, CoDyRA preserves the original model architecture and deployment pipeline, introducing no additional inference overhead. Extensive experiments show that our approach enhances representations for new downstream data while retaining pre-trained knowledge, achieving state-of-the-art results.

Paperid: 3935, https://arxiv.org/pdf/2411.15432.pdf

Abstract:
Model editing aims to correct inaccurate knowledge, update outdated information, and incorporate new data into Large Language Models (LLMs) without the need for retraining. This task poses challenges in lifelong scenarios where edits must be continuously applied for real-world applications. While some editors demonstrate strong robustness for lifelong editing in pure LLMs, Vision LLMs (VLLMs), which incorporate an additional vision modality, are not directly adaptable to existing LLM editors. In this paper, we propose LiveEdit, a LIfelong Vision language modEl Edit to bridge the gap between lifelong LLM editing and VLLMs. We begin by training an editing expert generator to independently produce low-rank experts for each editing instance, with the goal of correcting the relevant responses of the VLLM. A hard filtering mechanism is developed to utilize visual semantic knowledge, thereby coarsely eliminating visually irrelevant experts for input queries during the inference stage of the post-edited model. Finally, to integrate visually relevant experts, we introduce a soft routing mechanism based on textual semantic relevance to achieve multi-expert fusion. For evaluation, we establish a benchmark for lifelong VLLM editing. Extensive experiments demonstrate that LiveEdit offers significant advantages in lifelong VLLM editing scenarios. Further experiments validate the rationality and effectiveness of each module design in LiveEdit.

Paperid: 3936, https://arxiv.org/pdf/2411.13543.pdf

Abstract:
Large Language Models (LLMs) and Vision Language Models (VLMs) possess extensive knowledge and exhibit promising reasoning abilities, however, they still struggle to perform well in complex, dynamic environments. Real-world tasks require handling intricate interactions, advanced spatial reasoning, long-term planning, and continuous exploration of new strategies-areas in which we lack effective methodologies for comprehensively evaluating these capabilities. To address this gap, we introduce BALROG, a novel benchmark designed to assess the agentic capabilities of LLMs and VLMs through a diverse set of challenging games. Our benchmark incorporates a range of existing reinforcement learning environments with varying levels of difficulty, including tasks that are solvable by non-expert humans in seconds to extremely challenging ones that may take years to master (e.g., the NetHack Learning Environment). We devise fine-grained metrics to measure performance and conduct an extensive evaluation of several popular open-source and closed-source LLMs and VLMs. Our findings indicate that while current models achieve partial success in the easier games, they struggle significantly with more challenging tasks. Notably, we observe severe deficiencies in vision-based decision-making, as several models perform worse when visual representations of the environments are provided. We release BALROG as an open and user-friendly benchmark to facilitate future research and development in the agentic community. Code and Leaderboard at balrogai.com.

Paperid: 3937, https://arxiv.org/pdf/2411.12713.pdf

Abstract:
Large Vision-Language Model (LVLM) systems have demonstrated impressive vision-language reasoning capabilities but suffer from pervasive and severe hallucination issues, posing significant risks in critical domains such as healthcare and autonomous systems. Despite previous efforts to mitigate hallucinations, a persistent issue remains: visual defect from vision-language misalignment, creating a bottleneck in visual processing capacity. To address this challenge, we develop Complementary Adaptive Token-level Contrastive Decoding to Mitigate Hallucinations in LVLMs (CATCH), based on the Information Bottleneck theory. CATCH introduces Complementary Visual Decoupling (CVD) for visual information separation, Non-Visual Screening (NVS) for hallucination detection, and Adaptive Token-level Contrastive Decoding (ATCD) for hallucination mitigation. CATCH addresses issues related to visual defects that cause diminished fine-grained feature perception and cumulative hallucinations in open-ended scenarios. It is applicable to various visual question-answering tasks without requiring any specific data or prior knowledge, and generalizes robustly to new tasks without additional training, opening new possibilities for advancing LVLM in various challenging applications.

Paperid: 3938, https://arxiv.org/pdf/2411.05273.pdf

Abstract:
Offline reinforcement learning can enable policy learning from pre-collected, sub-optimal datasets without online interactions. This makes it ideal for real-world robots and safety-critical scenarios, where collecting online data or expert demonstrations is slow, costly, and risky. However, most existing offline RL works assume the dataset is already labeled with the task rewards, a process that often requires significant human effort, especially when ground-truth states are hard to ascertain (e.g., in the real-world). In this paper, we build on prior work, specifically RL-VLM-F, and propose a novel system that automatically generates reward labels for offline datasets using preference feedback from a vision-language model and a text description of the task. Our method then learns a policy using offline RL with the reward-labeled dataset. We demonstrate the system's applicability to a complex real-world robot-assisted dressing task, where we first learn a reward function using a vision-language model on a sub-optimal offline dataset, and then we use the learned reward to employ Implicit Q learning to develop an effective dressing policy. Our method also performs well in simulation tasks involving the manipulation of rigid and deformable objects, and significantly outperform baselines such as behavior cloning and inverse RL. In summary, we propose a new system that enables automatic reward labeling and policy learning from unlabeled, sub-optimal offline datasets.

Paperid: 3939, https://arxiv.org/pdf/2411.01703.pdf

Abstract:
Multimodal large language models (MLLMs) have revolutionized vision-language understanding but remain vulnerable to multimodal jailbreak attacks, where adversarial inputs are meticulously crafted to elicit harmful or inappropriate responses. We propose UniGuard, a novel multimodal safety guardrail that jointly considers the unimodal and cross-modal harmful signals. UniGuard trains a multimodal guardrail to minimize the likelihood of generating harmful responses in a toxic corpus. The guardrail can be seamlessly applied to any input prompt during inference with minimal computational costs. Extensive experiments demonstrate the generalizability of UniGuard across multiple modalities, attack strategies, and multiple state-of-the-art MLLMs, including LLaVA, Gemini Pro, GPT-4o, MiniGPT-4, and InstructBLIP. Notably, this robust defense mechanism maintains the models' overall vision-language understanding capabilities.

Paperid: 3940, https://arxiv.org/pdf/2411.00836.pdf

Abstract:
The rapid advancements in Vision-Language Models (VLMs) have shown great potential in tackling mathematical reasoning tasks that involve visual context. Unlike humans who can reliably apply solution steps to similar problems with minor modifications, we found that SOTA VLMs like GPT-4o can consistently fail in these scenarios, revealing limitations in their mathematical reasoning capabilities. In this paper, we investigate the mathematical reasoning robustness in VLMs and evaluate how well these models perform under different variants of the same question, such as changes in visual numerical values or function graphs. While several vision-based math benchmarks have been developed to assess VLMs' problem-solving capabilities, these benchmarks contain only static sets of problems and cannot easily evaluate mathematical reasoning robustness. To fill this gap, we introduce DynaMath, a dynamic visual math benchmark designed for in-depth assessment of VLMs. DynaMath includes 501 high-quality, multi-topic seed questions, each represented as a Python program. Those programs are carefully designed and annotated to enable the automatic generation of a much larger set of concrete questions, including many different types of visual and textual variations. DynaMath allows us to evaluate the generalization ability of VLMs, by assessing their performance under varying input conditions of a seed question. We evaluated 14 SOTA VLMs with 5,010 generated concrete questions. Our results show that the worst-case model accuracy, defined as the percentage of correctly answered seed questions in all 10 variants, is significantly lower than the average-case accuracy. Our analysis emphasizes the need to study the robustness of VLMs' reasoning abilities, and DynaMath provides valuable insights to guide the development of more reliable models for mathematical reasoning.

Paperid: 3941, https://arxiv.org/pdf/2410.22456.pdf

Abstract:
We introduce Image2Struct, a benchmark to evaluate vision-language models (VLMs) on extracting structure from images. Our benchmark 1) captures real-world use cases, 2) is fully automatic and does not require human judgment, and 3) is based on a renewable stream of fresh data. In Image2Struct, VLMs are prompted to generate the underlying structure (e.g., LaTeX code or HTML) from an input image (e.g., webpage screenshot). The structure is then rendered to produce an output image (e.g., rendered webpage), which is compared against the input image to produce a similarity score. This round-trip evaluation allows us to quantitatively evaluate VLMs on tasks with multiple valid structures. We create a pipeline that downloads fresh data from active online communities upon execution and evaluates the VLMs without human intervention. We introduce three domains (Webpages, LaTeX, and Musical Scores) and use five image metrics (pixel similarity, cosine similarity between the Inception vectors, learned perceptual image patch similarity, structural similarity index measure, and earth mover similarity) that allow efficient and automatic comparison between pairs of images. We evaluate Image2Struct on 14 prominent VLMs and find that scores vary widely, indicating that Image2Struct can differentiate between the performances of different VLMs. Additionally, the best score varies considerably across domains (e.g., 0.402 on sheet music vs. 0.830 on LaTeX equations), indicating that Image2Struct contains tasks of varying difficulty. For transparency, we release the full results at https://crfm.stanford.edu/helm/image2struct/v1.0.1/.

Paperid: 3942, https://arxiv.org/pdf/2410.17772.pdf

Abstract:
A central challenge towards developing robots that can relate human language to their perception and actions is the scarcity of natural language annotations in diverse robot datasets. Moreover, robot policies that follow natural language instructions are typically trained on either templated language or expensive human-labeled instructions, hindering their scalability. To this end, we introduce NILS: Natural language Instruction Labeling for Scalability. NILS automatically labels uncurated, long-horizon robot data at scale in a zero-shot manner without any human intervention. NILS combines pretrained vision-language foundation models in order to detect objects in a scene, detect object-centric changes, segment tasks from large datasets of unlabelled interaction data and ultimately label behavior datasets. Evaluations on BridgeV2, Fractal, and a kitchen play dataset show that NILS can autonomously annotate diverse robot demonstrations of unlabeled and unstructured datasets while alleviating several shortcomings of crowdsourced human annotations, such as low data quality and diversity. We use NILS to label over 115k trajectories obtained from over 430 hours of robot data. We open-source our auto-labeling code and generated annotations on our website: http://robottasklabeling.github.io.

Paperid: 3943, https://arxiv.org/pdf/2410.07073.pdf

Authors:Pravesh Agrawal, Szymon Antoniak, Emma Bou Hanna, Baptiste Bout, Devendra Chaplot, Jessica Chudnovsky, Diogo Costa, Baudouin De Monicault, Saurabh Garg, Theophile Gervet, Soham Ghosh, AmÃ©lie HÃ©liou, Paul Jacob, Albert Q. Jiang, Kartik Khandelwal, TimothÃ©e Lacroix, Guillaume Lample, Diego Las Casas, Thibaut Lavril, Teven Le Scao, Andy Lo, William Marshall, Louis Martin, Arthur Mensch, Pavankumar Muddireddy, Valera Nemychnikova, Marie Pellat, Patrick Von Platen, Nikhil Raghuraman, Baptiste RoziÃ¨re, Alexandre Sablayrolles, Lucile Saulnier, Romain Sauvestre, Wendy Shang, Roman Soletskyi, Lawrence Stewart, Pierre Stock, Joachim Studnia, Sandeep Subramanian, Sagar Vaze, Thomas Wang, Sophia Yang

Abstract:
We introduce Pixtral-12B, a 12--billion-parameter multimodal language model. Pixtral-12B is trained to understand both natural images and documents, achieving leading performance on various multimodal benchmarks, surpassing a number of larger models. Unlike many open-source models, Pixtral is also a cutting-edge text model for its size, and does not compromise on natural language performance to excel in multimodal tasks. Pixtral uses a new vision encoder trained from scratch, which allows it to ingest images at their natural resolution and aspect ratio. This gives users flexibility on the number of tokens used to process an image. Pixtral is also able to process any number of images in its long context window of 128K tokens. Pixtral 12B substanially outperforms other open models of similar sizes (Llama-3.2 11B \& Qwen-2-VL 7B). It also outperforms much larger open models like Llama-3.2 90B while being 7x smaller. We further contribute an open-source benchmark, MM-MT-Bench, for evaluating vision-language models in practical scenarios, and provide detailed analysis and code for standardized evaluation protocols for multimodal LLMs. Pixtral-12B is released under Apache 2.0 license.

Paperid: 3944, https://arxiv.org/pdf/2410.05600.pdf

Abstract:
The widespread presence of hate speech on the internet, including formats such as text-based tweets and vision-language memes, poses a significant challenge to digital platform safety. Recent research has developed detection models tailored to specific modalities; however, there is a notable gap in transferring detection capabilities across different formats. This study conducts extensive experiments using few-shot in-context learning with large language models to explore the transferability of hate speech detection between modalities. Our findings demonstrate that text-based hate speech examples can significantly enhance the classification accuracy of vision-language hate speech. Moreover, text-based demonstrations outperform vision-language demonstrations in few-shot learning settings. These results highlight the effectiveness of cross-modality knowledge transfer and offer valuable insights for improving hate speech detection systems.

Paperid: 3945, https://arxiv.org/pdf/2410.05261.pdf

Abstract:
Reading dense text and locating objects within images are fundamental abilities for Large Vision-Language Models (LVLMs) tasked with advanced jobs. Previous LVLMs, including superior proprietary models like GPT-4o, have struggled to excel in both tasks simultaneously. Moreover, previous LVLMs with fine-grained perception cost thousands of tokens per image, making them resource-intensive. We present TextHawk2, a bilingual LVLM featuring efficient fine-grained perception and demonstrating cutting-edge performance across general-purpose, OCR, and grounding tasks with 16 times fewer image tokens. Critical improvements include: (1) Token Compression: Building on the efficient architecture of its predecessor, TextHawk2 significantly reduces the number of tokens per image by 16 times, facilitating training and deployment of the TextHawk series with minimal resources. (2) Visual Encoder Reinforcement: We enhance the visual encoder through LVLM co-training, unlocking its potential for previously unseen tasks like Chinese OCR and grounding. (3) Data Diversity: We maintain a comparable scale of 100 million samples while diversifying the sources of pre-training data. We assess TextHawk2 across multiple benchmarks, where it consistently delivers superior performance and outperforms closed-source models of similar scale, such as achieving 78.4% accuracy on OCRBench, 81.4% accuracy on ChartQA, 89.6% ANLS on DocVQA, and 88.1% accuracy@0.5 on RefCOCOg-test.

Paperid: 3946, https://arxiv.org/pdf/2410.01534.pdf

Abstract:
Contrastive Language-Image Pre-training (CLIP) models have shown significant potential, particularly in zero-shot classification across diverse distribution shifts. Building on existing evaluations of overall classification robustness, this work aims to provide a more comprehensive assessment of CLIP by introducing several new perspectives. First, we investigate their robustness to variations in specific visual factors. Second, we assess two critical safety objectives--confidence uncertainty and out-of-distribution detection--beyond mere classification accuracy. Third, we evaluate the finesse with which CLIP models bridge the image and text modalities. Fourth, we extend our examination to 3D awareness in CLIP models, moving beyond traditional 2D image understanding. Finally, we explore the interaction between vision and language encoders within modern large multimodal models (LMMs) that utilize CLIP as the visual backbone, focusing on how this interaction impacts classification robustness. In each aspect, we consider the impact of six factors on CLIP models: model architecture, training distribution, training set size, fine-tuning, contrastive loss, and test-time prompts. Our study uncovers several previously unknown insights into CLIP. For instance, the architecture of the visual encoder in CLIP plays a significant role in their robustness against 3D corruption. CLIP models tend to exhibit a bias towards shape when making predictions. Moreover, this bias tends to diminish after fine-tuning on ImageNet. Vision-language models like LLaVA, leveraging the CLIP vision encoder, could exhibit benefits in classification performance for challenging categories over CLIP alone. Our findings are poised to offer valuable guidance for enhancing the robustness and reliability of CLIP models.

Paperid: 3947, https://arxiv.org/pdf/2410.00332.pdf

Abstract:
Understanding law of conservation is a critical milestone in human cognitive development considered to be supported by the apprehension of quantitative concepts and the reversibility of operations. To assess whether this critical component of human intelligence has emerged in Vision Language Models, we have curated the ConserveBench, a battery of 365 cognitive experiments across four dimensions of physical quantities: volume, solid quantity, length, and number. The former two involve transformational tasks which require reversibility understanding. The latter two involve non-transformational tasks which assess quantity understanding. Surprisingly, we find that while Vision Language Models are generally good at transformational tasks, they tend to fail at non-transformational tasks. There is a dissociation between understanding the reversibility of operations and understanding the concept of quantity, which both are believed to be the cornerstones of understanding law of conservation in humans.

Paperid: 3948, https://arxiv.org/pdf/2410.00324.pdf

Abstract:
Knowing others' intentions and taking others' perspectives are two core components of human intelligence that are considered to be instantiations of theory-of-mind. Infiltrating machines with these abilities is an important step towards building human-level artificial intelligence. Here, to investigate intentionality understanding and level-2 perspective-taking in Vision Language Models (VLMs), we constructed the IntentBench and PerspectBench, which together contains over 300 cognitive experiments grounded in real-world scenarios and classic cognitive tasks. We found VLMs achieving high performance on intentionality understanding but low performance on level-2 perspective-taking. This suggests a potential dissociation between simulation-based and theory-based theory-of-mind abilities in VLMs, highlighting the concern that they are not capable of using model-based reasoning to infer others' mental states.

Paperid: 3949, https://arxiv.org/pdf/2410.00318.pdf

Abstract:
Mechanical reasoning is a hallmark of human intelligence, defined by its ubiquitous yet irreplaceable role in human activities ranging from routine tasks to civil engineering. Embedding machines with mechanical reasoning is therefore an important step towards building human-level artificial intelligence. Here, we leveraged 155 cognitive experiments to test the understanding of system stability, gears and pulley systems, leverage principle, inertia and motion, and fluid mechanics in 26 Vision Language Models (VLMs). Results indicate that VLMs consistently perform worse than humans on all domains, while demonstrate significant difficulty in reasoning about gear systems and fluid mechanics. Notably, their performance on these tasks do not improve as number of parameters increase, suggesting that current attention-based architecture may fail to grasp certain underlying mechanisms required for mechanical reasoning, particularly those pertaining to mental simulations.

Paperid: 3950, https://arxiv.org/pdf/2409.19375.pdf

Abstract:
Vision-language foundation models (VLMs), such as CLIP, exhibit remarkable performance across a wide range of tasks. However, deploying these models can be unreliable when significant distribution gaps exist between training and test data, while fine-tuning for diverse scenarios is often costly. Cache-based test-time adapters offer an efficient alternative by storing representative test samples to guide subsequent classifications. Yet, these methods typically employ naive cache management with limited capacity, leading to severe catastrophic forgetting when samples are inevitably dropped during updates. In this paper, we propose DOTA (DistributiOnal Test-time Adaptation), a simple yet effective method addressing this limitation. Crucially, instead of merely memorizing individual test samples, DOTA continuously estimates the underlying distribution of the test data stream. Test-time posterior probabilities are then computed using these dynamically estimated distributions via Bayes' theorem for adaptation. This distribution-centric approach enables the model to continually learn and adapt to the deployment environment. Extensive experiments validate that DOTA significantly mitigates forgetting and achieves state-of-the-art performance compared to existing methods.

Paperid: 3951, https://arxiv.org/pdf/2409.17621.pdf

Abstract:
Human-robot interaction (HRI) encompasses a wide range of collaborative tasks, with handover being one of the most fundamental. As robots become more integrated into human environments, the potential for service robots to assist in handing objects to humans is increasingly promising. In robot-to-human (R2H) handover, selecting the optimal grasp is crucial for success, as it requires avoiding interference with the humans preferred grasp region and minimizing intrusion into their workspace. Existing methods either inadequately consider geometric information or rely on data-driven approaches, which often struggle to generalize across diverse objects. To address these limitations, we propose a novel zero-shot system that combines semantic and geometric information to generate optimal handover grasps. Our method first identifies grasp regions using semantic knowledge from vision-language models (VLMs) and, by incorporating customized visual prompts, achieves finer granularity in region grounding. A grasp is then selected based on grasp distance and approach angle to maximize human ease and avoid interference. We validate our approach through ablation studies and real-world comparison experiments. Results demonstrate that our system improves handover success rates and provides a more user-preferred interaction experience. Videos, appendixes and more are available at https://sites.google.com/view/vlm-handover/.

Paperid: 3952, https://arxiv.org/pdf/2409.13612.pdf

Abstract:
The rapid development of Large Vision-Language Models (LVLMs) often comes with widespread hallucination issues, making cost-effective and comprehensive assessments increasingly vital. Current approaches mainly rely on costly annotations and are not comprehensive -- in terms of evaluating all aspects such as relations, attributes, and dependencies between aspects. Therefore, we introduce the FIHA (autonomous Fine-graIned Hallucination evAluation evaluation in LVLMs), which could access hallucination LVLMs in the LLM-free and annotation-free way and model the dependency between different types of hallucinations. FIHA can generate Q&A pairs on any image dataset at minimal cost, enabling hallucination assessment from both image and caption. Based on this approach, we introduce a benchmark called FIHA-v1, which consists of diverse questions on various images from MSCOCO and Foggy. Furthermore, we use the Davidson Scene Graph (DSG) to organize the structure among Q&A pairs, in which we can increase the reliability of the evaluation. We evaluate representative models using FIHA-v1, highlighting their limitations and challenges. We released our code and data.

Paperid: 3953, https://arxiv.org/pdf/2409.10090.pdf

Abstract:
We introduce InteractPro, a comprehensive framework for dynamic motion-aware image composition. At its core is InteractPlan, an intelligent planner that leverages a Large Vision Language Model (LVLM) for scenario analysis and object placement, determining the optimal composition strategy to achieve realistic motion effects. Based on each scenario, InteractPlan selects between our two specialized modules: InteractPhys and InteractMotion. InteractPhys employs an enhanced Material Point Method (MPM)-based simulation to produce physically faithful and controllable object-scene interactions, capturing diverse and abstract events that require true physical modeling. InteractMotion, in contrast, is a training-free method based on pretrained video diffusion. Traditional composition approaches suffer from two major limitations: requiring manual planning for object placement and generating static, motionless outputs. By unifying simulation-based and diffusion-based methods under planner guidance, InteractPro overcomes these challenges, ensuring richly motion-aware compositions. Extensive quantitative and qualitative evaluations demonstrate InteractPro's effectiveness in producing controllable, and coherent compositions across varied scenarios.

Paperid: 3954, https://arxiv.org/pdf/2409.08206.pdf

Abstract:
Vision-language models (VLMs) like CLIP have showcased a remarkable ability to extract transferable features for downstream tasks. Nonetheless, the training process of these models is usually based on a coarse-grained contrastive loss between the global embedding of images and texts which may lose the compositional structure of these modalities. Many recent studies have shown VLMs lack compositional understandings like attribute binding and identifying object relationships. Although some recent methods have tried to achieve finer-level alignments, they either are not based on extracting meaningful components of proper granularity or don't properly utilize the modalities' correspondence (especially in image-text pairs with more ingredients). Addressing these limitations, we introduce Compositional Alignment (ComAlign), a fine-grained approach to discover more exact correspondence of text and image components using only the weak supervision in the form of image-text pairs. Our methodology emphasizes that the compositional structure (including entities and relations) extracted from the text modality must also be retained in the image modality. To enforce correspondence of fine-grained concepts in image and text modalities, we train a lightweight network lying on top of existing visual and language encoders using a small dataset. The network is trained to align nodes and edges of the structure across the modalities. Experimental results on various VLMs and datasets demonstrate significant improvements in retrieval and compositional benchmarks, affirming the effectiveness of our plugin model.

Paperid: 3955, https://arxiv.org/pdf/2601.17844.pdf

Abstract:
Electroencephalogram (EEG) decoding is a critical component of medical diagnostics, rehabilitation engineering, and brain-computer interfaces. However, contemporary decoding methodologies remain heavily dependent on task-specific datasets to train specialized neural network architectures. Consequently, limited data availability impedes the development of generalizable large brain decoding models. In this work, we propose a paradigm shift from conventional signal-based decoding by leveraging large-scale vision-language models (VLMs) to analyze EEG waveform plots. By converting multivariate EEG signals into stacked waveform images and integrating neuroscience domain expertise into textual prompts, we demonstrate that foundational VLMs can effectively differentiate between different patterns in the human brain. To address the inherent non-stationarity of EEG signals, we introduce a Retrieval-Augmented In-Context Learning (RAICL) approach, which dynamically selects the most representative and relevant few-shot examples to condition the autoregressive outputs of the VLM. Experiments on EEG-based seizure detection indicate that state-of-the-art VLMs under RAICL achieved better or comparable performance with traditional time series based approaches. These findings suggest a new direction in physiological signal processing that effectively bridges the modalities of vision, language, and neural activities. Furthermore, the utilization of off-the-shelf VLMs, without the need for retraining or downstream architecture construction, offers a readily deployable solution for clinical applications.

Paperid: 3956, https://arxiv.org/pdf/2601.17697.pdf

Abstract:
Representing artistic style is challenging due to its deep entanglement with semantic content. We propose StyleDecoupler, an information-theoretic framework that leverages a key insight: multi-modal vision models encode both style and content, while uni-modal models suppress style to focus on content-invariant features. By using uni-modal representations as content-only references, we isolate pure style features from multi-modal embeddings through mutual information minimization. StyleDecoupler operates as a plug-and-play module on frozen Vision-Language Models without fine-tuning. We also introduce WeART, a large-scale benchmark of 280K artworks across 152 styles and 1,556 artists. Experiments show state-of-the-art performance on style retrieval across WeART and WikiART, while enabling applications like style relationship mapping and generative model evaluation. We release our method and dataset at this url.

Paperid: 3957, https://arxiv.org/pdf/2601.07779.pdf

Abstract:
While Vision-Language Models (VLMs) have significantly advanced Computer-Using Agents (CUAs), current frameworks struggle with robustness in long-horizon workflows and generalization in novel domains. These limitations stem from a lack of granular control over historical visual context curation and the absence of visual-aware tutorial retrieval. To bridge these gaps, we introduce OS-Symphony, a holistic framework that comprises an Orchestrator coordinating two key innovations for robust automation: (1) a Reflection-Memory Agent that utilizes milestone-driven long-term memory to enable trajectory-level self-correction, effectively mitigating visual context loss in long-horizon tasks; (2) Versatile Tool Agents featuring a Multimodal Searcher that adopts a SeeAct paradigm to navigate a browser-based sandbox to synthesize live, visually aligned tutorials, thereby resolving fidelity issues in unseen scenarios. Experimental results demonstrate that OS-Symphony delivers substantial performance gains across varying model scales, establishing new state-of-the-art results on three online benchmarks, notably achieving 65.84% on OSWorld.

Paperid: 3958, https://arxiv.org/pdf/2601.07060.pdf

Abstract:
Recent advancements in vision-language-action (VLA) models have shown promise in robotic manipulation, yet they continue to struggle with long-horizon, multi-step tasks. Existing methods lack internal reasoning mechanisms that can identify task-relevant interaction cues or track progress within a subtask, leading to critical execution errors such as repeated actions, missed steps, and premature termination. To address these challenges, we introduce PALM, a VLA framework that structures policy learning around interaction-centric affordance reasoning and subtask progress cues. PALM distills complementary affordance representations that capture object relevance, contact geometry, spatial placements, and motion dynamics, and serve as task-relevant anchors for visuomotor control. To further stabilize long-horizon execution, PALM predicts continuous within-subtask progress, enabling seamless subtask transitions. Across extensive simulation and real-world experiments, PALM consistently outperforms baselines, achieving a 91.8% success rate on LIBERO-LONG, a 12.5% improvement in average length on CALVIN ABC->D, and a 2x improvement over real-world baselines across three long-horizon generalization settings.

Paperid: 3959, https://arxiv.org/pdf/2601.06424.pdf

Abstract:
To explore a more scalable path for adding multimodal capabilities to existing LLMs, this paper addresses a fundamental question: Can a unimodal LLM, relying solely on text, reason about its own informational needs and provide effective feedback to optimize a multimodal model? To answer this, we propose a method that enables a language agent to give feedback to a vision-language model (VLM) to adapt text generation to the agent's preferences. Our results from different experiments affirm this hypothesis, showing that LLM preference feedback significantly enhances VLM descriptions. Using our proposed method, we find that the VLM can generate multimodal scene descriptions to help the LLM better understand multimodal context, leading to improvements of maximum 13% in absolute accuracy compared to the baseline multimodal approach. Furthermore, a human study validated our AI-driven feedback, showing a 64.6% preference alignment rate between the LLM's choices and human judgments. Extensive experiments provide insights on how and why the method works and its limitations.

Paperid: 3960, https://arxiv.org/pdf/2601.06165.pdf

Abstract:
Current vision-language benchmarks predominantly feature well-structured questions with clear, explicit prompts. However, real user queries are often informal and underspecified. Users naturally leave much unsaid, relying on images to convey context. We introduce HAERAE-Vision, a benchmark of 653 real-world visual questions from Korean online communities (0.76% survival from 86K candidates), each paired with an explicit rewrite, yielding 1,306 query variants in total. Evaluating 39 VLMs, we find that even state-of-the-art models (GPT-5, Gemini 2.5 Pro) achieve under 50% on the original queries. Crucially, query explicitation alone yields 8 to 22 point improvements, with smaller models benefiting most. We further show that even with web search, under-specified queries underperform explicit queries without search, revealing that current retrieval cannot compensate for what users leave unsaid. Our findings demonstrate that a substantial portion of VLM difficulty stem from natural query under-specification instead of model capability, highlighting a critical gap between benchmark evaluation and real-world deployment.

Paperid: 3961, https://arxiv.org/pdf/2512.24851.pdf

Abstract:
Multimodal Large Language Models (MLLMs) have demonstrated remarkable capabilities across a wide range of vision-language tasks. However, their performance as embodied agents, which requires multi-round dialogue spatial reasoning and sequential action prediction, needs further exploration. Our work investigates this potential in the context of Vision-and-Language Navigation (VLN) by introducing a unified and extensible evaluation framework to probe MLLMs as zero-shot agents by bridging traditional navigation datasets into a standardized benchmark, named VLN-MME. We simplify the evaluation with a highly modular and accessible design. This flexibility streamlines experiments, enabling structured comparisons and component-level ablations across diverse MLLM architectures, agent designs, and navigation tasks. Crucially, enabled by our framework, we observe that enhancing our baseline agent with Chain-of-Thought (CoT) reasoning and self-reflection leads to an unexpected performance decrease. This suggests MLLMs exhibit poor context awareness in embodied navigation tasks; although they can follow instructions and structure their output, their 3D spatial reasoning fidelity is low. VLN-MME lays the groundwork for systematic evaluation of general-purpose MLLMs in embodied navigation settings and reveals limitations in their sequential decision-making capabilities. We believe these findings offer crucial guidance for MLLM post-training as embodied agents.

Paperid: 3962, https://arxiv.org/pdf/2512.18004.pdf

Abstract:
Handwritten text recognition (HTR) and machine translation continue to pose significant challenges, particularly for low-resource languages like Marathi, which lack large digitized corpora and exhibit high variability in handwriting styles. The conventional approach to address this involves a two-stage pipeline: an OCR system extracts text from handwritten images, which is then translated into the target language using a machine translation model. In this work, we explore and compare the performance of traditional OCR-MT pipelines with Vision Large Language Models that aim to unify these stages and directly translate handwritten text images in a single, end-to-end step. Our motivation is grounded in the urgent need for scalable, accurate translation systems to digitize legal records such as FIRs, charge sheets, and witness statements in India's district and high courts. We evaluate both approaches on a curated dataset of handwritten Marathi legal documents, with the goal of enabling efficient legal document processing, even in low-resource environments. Our findings offer actionable insights toward building robust, edge-deployable solutions that enhance access to legal information for non-native speakers and legal professionals alike.

Paperid: 3963, https://arxiv.org/pdf/2512.15258.pdf

Abstract:
This paper proposes VLA-AN, an efficient and onboard Vision-Language-Action (VLA) framework dedicated to autonomous drone navigation in complex environments. VLA-AN addresses four major limitations of existing large aerial navigation models: the data domain gap, insufficient temporal navigation with reasoning, safety issues with generative action policies, and onboard deployment constraints. First, we construct a high-fidelity dataset utilizing 3D Gaussian Splatting (3D-GS) to effectively bridge the domain gap. Second, we introduce a progressive three-stage training framework that sequentially reinforces scene comprehension, core flight skills, and complex navigation capabilities. Third, we design a lightweight, real-time action module coupled with geometric safety correction. This module ensures fast, collision-free, and stable command generation, mitigating the safety risks inherent in stochastic generative policies. Finally, through deep optimization of the onboard deployment pipeline, VLA-AN achieves a robust real-time 8.3x improvement in inference throughput on resource-constrained UAVs. Extensive experiments demonstrate that VLA-AN significantly improves spatial grounding, scene reasoning, and long-horizon navigation, achieving a maximum single-task success rate of 98.1%, and providing an efficient, practical solution for realizing full-chain closed-loop autonomy in lightweight aerial robots.

Paperid: 3964, https://arxiv.org/pdf/2512.12089.pdf

Abstract:
Large vision-language models (LVLMs) exhibit impressive ability to jointly reason over visual and textual inputs. However, they often produce outputs that are linguistically fluent but factually inconsistent with the visual evidence, i.e., they hallucinate. Despite growing efforts to mitigate such hallucinations, a key question remains: what form of visual attention can effectively suppress hallucinations during decoding? In this work, we provide a simple answer: the vision encoder's own attention map. We show that LVLMs tend to hallucinate when their final visual-attention maps fail to concentrate on key image objects, whereas the vision encoder's more concentrated attention maps substantially reduce hallucinations. To further investigate the cause, we analyze vision-text conflicts during decoding and find that these conflicts peak in the language model's middle layers. Injecting the vision encoder's attention maps into these layers effectively suppresses hallucinations. Building on these insights, we introduce VEGAS, a simple yet effective inference-time method that integrates the vision encoder's attention maps into the language model's mid-layers and adaptively steers tokens which fail to concentrate on key image objects. Extensive experiments across multiple benchmarks demonstrate that VEGAS consistently achieves state-of-the-art performance in reducing hallucinations.

Paperid: 3965, https://arxiv.org/pdf/2512.10310.pdf

Abstract:
Multimodal large language models (MLLMs) have shown promising potential in Vision-Language Navigation (VLN). However, their practical development is severely hindered by the substantial training overhead. We recognize two key issues that contribute to the overhead: (1) the quadratic computational burden from processing long-horizon historical observations as massive sequences of tokens, and (2) the exploration-efficiency trade-off in DAgger, i.e., a data aggregation process of collecting agent-explored trajectories. While more exploration yields effective error-recovery trajectories for handling test-time distribution shifts, it comes at the cost of longer trajectory lengths for both training and inference. To address these challenges, we propose Efficient-VLN, a training-efficient VLN model. Specifically, to mitigate the token processing burden, we design two efficient memory mechanisms: a progressive memory that dynamically allocates more tokens to recent observations, and a learnable recursive memory that utilizes the key-value cache of learnable tokens as the memory state. Moreover, we introduce a dynamic mixed policy to balance the exploration-efficiency trade-off. Extensive experiments show that Efficient-VLN achieves state-of-the-art performance on R2R-CE (64.2% SR) and RxR-CE (67.0% SR). Critically, our model consumes merely 282 H800 GPU hours, demonstrating a dramatic reduction in training overhead compared to state-of-the-art methods.

Paperid: 3966, https://arxiv.org/pdf/2512.09944.pdf

Abstract:
Echocardiography is central to contemporary cardiovascular care, but full-study interpretation remains a cognitively demanding, multi-view task that is still performed manually. While recent foundation models for echocardiography can achieve strong performance on individual perceptual subtasks such as view classification, segmentation, or disease prediction, they typically operate in isolation and do not provide a unified, clinically coherent assessment. In this work, we introduce Echo-CoPilot, a multi-view, multi-task agent that uses a large language model to orchestrate a suite of specialized echocardiography tools. Within a ReAct-style loop, the agent decomposes clinician queries, invokes tools for view recognition, cardiac structure segmentation, measurement and disease prediction, and report synthesis, and integrates their outputs into guideline-aware answers and narrative summaries. We evaluate Echo-CoPilot on the public MIMIC-EchoQA benchmark, where it achieves an accuracy of 50.8\%, outperforming both general-purpose and biomedical video vision-language models. Qualitative analyses further show that the agent leverages quantitative measurements and physiologic context to resolve challenging cases near clinical decision thresholds, such as borderline left ventricular hypertrophy or pericardial effusion severity. The code will be released upon acceptance of the paper.

Paperid: 3967, https://arxiv.org/pdf/2512.07234.pdf

Abstract:
Dropout is a widely used regularization technique which improves the generalization ability of a model by randomly dropping neurons. In light of this, we propose Dropout Prompt Learning, which aims for applying dropout to improve the robustness of the vision-language models. Different from the vanilla dropout, we apply dropout on the tokens of the textual and visual branches, where we evaluate the token significance considering both intra-modal context and inter-modal alignment, enabling flexible dropout probabilities for each token. Moreover, to maintain semantic alignment for general knowledge transfer while encouraging the diverse representations that dropout introduces, we further propose residual entropy regularization. Experiments on 15 benchmarks show our method's effectiveness in challenging scenarios like low-shot learning, long-tail classification, and out-of-distribution generalization. Notably, our method surpasses regularization-based methods including KgCoOp by 5.10% and PromptSRC by 2.13% in performance on base-to-novel generalization.

Paperid: 3968, https://arxiv.org/pdf/2512.05964.pdf

Abstract:
Real-time chunking (RTC) enables vision-language-action models (VLAs) to generate smooth, reactive robot trajectories by asynchronously predicting action chunks and conditioning on previously committed actions via inference-time inpainting. However, this inpainting method introduces computational overhead that increases inference latency. In this work, we propose a simple alternative: simulating inference delay at training time and conditioning on action prefixes directly, eliminating any inference-time overhead. Our method requires no modifications to the model architecture or robot runtime, and can be implemented with only a few additional lines of code. In simulated experiments, we find that training-time RTC outperforms inference-time RTC at higher inference delays. In real-world experiments on box building and espresso making tasks with the $π_{0.6}$ VLA, we demonstrate that training-time RTC maintains both task performance and speed parity with inference-time RTC while being computationally cheaper. Our results suggest that training-time action conditioning is a practical drop-in replacement for inference-time inpainting in real-time robot control.

Paperid: 3969, https://arxiv.org/pdf/2512.05955.pdf

Abstract:
Vision-Language Models (VLMs) exhibit remarkable common-sense and semantic reasoning capabilities. However, they lack a grounded understanding of physical dynamics. This limitation arises from training VLMs on static internet-scale visual-language data that contain no causal interactions or action-conditioned changes. Consequently, it remains challenging to leverage VLMs for fine-grained robotic manipulation tasks that require physical understanding, reasoning, and corresponding action planning. To overcome this, we present SIMPACT, a test-time, SIMulation-enabled ACTion Planning framework that equips VLMs with physical reasoning through simulation-in-the-loop world modeling, without requiring any additional training. From a single RGB-D observation, SIMPACT efficiently constructs physics simulations, enabling the VLM to propose informed actions, observe simulated rollouts, and iteratively refine its reasoning. By integrating language reasoning with physics prediction, our simulation-enabled VLM can understand contact dynamics and action outcomes in a physically grounded way. Our method demonstrates state-of-the-art performance on five challenging, real-world rigid-body and deformable manipulation tasks that require fine-grained physical reasoning, outperforming existing general-purpose robotic manipulation models. Our results demonstrate that embedding physics understanding via efficient simulation into VLM reasoning at test time offers a promising path towards generalizable embodied intelligence. Project webpage can be found at https://simpact-bot.github.io

Paperid: 3970, https://arxiv.org/pdf/2512.04540.pdf

Abstract:
Ultra long video understanding remains an open challenge, as existing vision language models (VLMs) falter on such content due to limited context length and inefficient long term memory retention. To address this, recent works have attempted to construct external knowledge bases and corresponding retrieval agumented generation (RAG) systems, yet these incur enormous storage and computational overhead. In this paper, we propose VideoMem, a novel framework that pioneers models long video understanding as a sequential generation task via adaptive memory management. Specifically, VideoMem dynamically updates a global memory buffer, which adaptively retains critical information while discarding redundant content across the video timeline. To efficiently train VLMs for such long-term tasks, VideoMem integrates the Progressive Grouped Relative Policy Optimization (PRPO) algorithm, equipped with two core modules: Progressive State Propagation (PSP) adaptively retains valid current states, propagates them to the next rollout step, and gradually narrows the model exploration space. Temporal Cascading Reward (TCR) further alleviates reward sparsity, improving sample utilization and accelerating convergence. Extensive experiments demonstrate that VideoMem significantly outperforms existing open-source models across diverse benchmarks for ultra-long video understanding tasks.

Paperid: 3971, https://arxiv.org/pdf/2512.03276.pdf

Abstract:
Training vision language models (VLMs) aims to align visual representations from a vision encoder with the textual representations of a pretrained large language model (LLM). However, many VLMs exhibit reduced factual recall performance compared to their LLM backbones, raising the question of how effective multimodal fine-tuning is at extending existing mechanisms within the LLM to visual inputs. We argue that factual recall based on visual inputs requires VLMs to solve a two-hop problem: (1) forming entity representations from visual inputs, and (2) recalling associated factual knowledge based on these entity representations. By benchmarking 14 VLMs with various architectures (LLaVA, Native, Cross-Attention), sizes (7B-124B parameters), and training setups on factual recall tasks against their original LLM backbone models, we find that 11 of 14 models exhibit factual recall degradation. We select three models with high and two models with low performance degradation, and use attribution patching, activation patching, and probing to show that degraded VLMs struggle to use the existing factual recall circuit of their LLM backbone, because they resolve the first hop too late in the computation. In contrast, high-performing VLMs resolve entity representations early enough to reuse the existing factual recall mechanism. Finally, we demonstrate two methods to recover performance: patching entity representations from the LLM backbone into the VLM, and prompting with chain-of-thought reasoning. Our results highlight that the speed of early entity resolution critically determines how effective VLMs are in using preexisting LLM mechanisms. More broadly, our work illustrates how mechanistic analysis can explain and unveil systematic failures in multimodal alignment.

Paperid: 3972, https://arxiv.org/pdf/2512.02924.pdf

Abstract:
While Neural Processing Units (NPUs) offer high theoretical efficiency for edge AI, state-of-the-art Vision--Language Models (VLMs) tailored for GPUs often falter on these substrates. We attribute this hardware-model mismatch to two primary factors: the quantization brittleness of Vision Transformers (ViTs) and the I/O-bound nature of autoregressive attention mechanisms, which fail to utilize the high arithmetic throughput of NPUs. To bridge this gap, we propose AutoNeural, an NPU-native VLM architecture co-designed for integer-only inference. We replace the standard ViT encoder with a MobileNetV5-style backbone utilizing depthwise separable convolutions, which ensures bounded activation distributions for stable INT4/8/16 quantization. Complementing this, our language backbone integrates State-Space Model (SSM) principles with Transformer layers, employing efficient gated convolutions to achieve linear-time complexity. This hybrid design eliminates the heavy memory I/O overhead of Key-Value caching during generation. Our approach delivers substantial efficiency gains, reducing quantization error of vision encoder by up to 7x and end-to-end latency by 14x compared to conventional baselines. The AutoNeural also delivers 3x decoding speed and 4x longer context window than the baseline. We validate these improvements via a real-world automotive case study on the Qualcomm SA8295P SoC, demonstrating real-time performance for cockpit applications. Our results highlight that rethinking model topology specifically for NPU constraints is a prerequisite for robust multi-modal edge intelligence.

Paperid: 3973, https://arxiv.org/pdf/2512.02569.pdf

Abstract:
This perspective reframes human-robot interaction (HRI) through extended reality (XR), arguing that virtual robots powered by large foundation models (FMs) can serve as cognitively grounded, empathic agents. Unlike physical robots, XR-native agents are unbound by hardware constraints and can be instantiated, adapted, and scaled on demand, while still affording embodiment and co-presence. We synthesize work across XR, HRI, and cognitive AI to show how such agents can support safety-critical scenarios, socially and cognitively empathic interaction across domains, and outreaching physical capabilities with XR and AI integration. We then discuss how multimodal large FMs (e.g., large language model, large vision model, and vision-language model) enable context-aware reasoning, affect-sensitive situations, and long-term adaptation, positioning virtual robots as cognitive and empathic mediators rather than mere simulation assets. At the same time, we highlight challenges and potential risks, including overtrust, cultural and representational bias, privacy concerns around biometric sensing, and data governance and transparency. The paper concludes by outlining a research agenda for human-centered, ethically grounded XR agents - emphasizing multi-layered evaluation frameworks, multi-user ecosystems, mixed virtual-physical embodiment, and societal and ethical design practices to envision XR-based virtual agents powered by FMs as reshaping future HRI into a more efficient and adaptive paradigm.

Paperid: 3974, https://arxiv.org/pdf/2511.23055.pdf

Abstract:
Theory of Mind (ToM) refers to the ability to infer others' mental states, such as beliefs, desires, and intentions. Current vision-language embodied agents lack ToM-based decision-making, and existing benchmarks focus solely on human mental states while ignoring the agent's own perspective, hindering coherent decision and action generation. To address this, we propose MindPower, a Robot-Centric framework integrating Perception, Mental Reasoning, Decision Making and Action. Given multimodal inputs, MindPower first perceives the environment and human states, then performs ToM Reasoning to model both self and others, and finally generates decisions and actions guided by inferred mental states. Furthermore, we introduce Mind-Reward, a novel optimization objective that encourages VLMs to produce consistent ToM Reasoning and behavior. Our model outperforms GPT-4o by 12.77% in decision making and 12.49% in action generation.

Paperid: 3975, https://arxiv.org/pdf/2511.22928.pdf

Abstract:
Ensuring safety remains a key challenge for autonomous vehicles (AVs), especially in rare and complex scenarios. One critical but understudied aspect is the \textbf{potential risk} situations, where the risk is \textbf{not yet observable} but can be inferred from subtle precursors, such as anomalous behaviors or commonsense violations. Recognizing these precursors requires strong semantic understanding and reasoning capabilities, which are often absent in current AV systems due to the scarcity of such cases in existing driving or risk-centric datasets. Moreover, current autonomous driving accident datasets often lack annotations of the causal reasoning chains behind incidents, which are essential for identifying potential risks before they become observable. To address these gaps, we introduce PotentialRiskQA, a novel vision-language dataset designed for reasoning about potential risks prior to observation. Each sample is annotated with structured scene descriptions, semantic precursors, and inferred risk outcomes. Based on this dataset, we further propose PR-Reasoner, a vision-language-model-based framework tailored for onboard potential risk reasoning. Experimental results show that fine-tuning on PotentialRiskQA enables PR-Reasoner to significantly enhance its performance on the potential risk reasoning task compared to baseline VLMs. Together, our dataset and model provide a foundation for developing autonomous systems with improved foresight and proactive safety capabilities, moving toward more intelligent and resilient AVs.

Paperid: 3976, https://arxiv.org/pdf/2511.18151.pdf

Abstract:
Unmanned Aerial Vehicles (UAVs) in disaster response require complex, queryable intelligence that on-board CNNs cannot provide. While Vision-Language Models (VLMs) offer this semantic reasoning, their high resource demands make on-device deployment infeasible, and naive cloud offloading fails under the low-bandwidth networks common in disaster zones. We present AVERY, a framework that enables VLM deployment through adaptive split computing. We advance the split computing paradigm beyond traditional depth-wise partitioning by introducing a functional, cognitive-inspired dual-stream split that separates the VLM into a high-frequency, low-resolution "context stream" for real-time awareness and a low-frequency, high-fidelity "insight stream" for deep analysis. A lightweight, self-aware on-board controller manages this architecture, monitoring network conditions and operator intent to dynamically select from pre-trained compression models, navigating the fundamental accuracy-throughput trade-off. Evaluated using the VLM LISA-7B across an edge-cloud scenario under fluctuating network conditions, AVERY consistently outperforms static configurations, achieving 11.2% higher accuracy than raw image compression and 93.98% lower energy consumption compared to full-edge execution, thereby enhancing mission efficiency and enabling real-time, queryable intelligence on resource-constrained platforms in dynamic environments.

Paperid: 3977, https://arxiv.org/pdf/2511.11693.pdf

Abstract:
Generative vision-language models like Stable Diffusion demonstrate remarkable capabilities in creative media synthesis, but they also pose substantial risks of producing unsafe, offensive, or culturally inappropriate content when prompted adversarially. Current defenses struggle to align outputs with human values without sacrificing generation quality or incurring high costs. To address these challenges, we introduce VALOR (Value-Aligned LLM-Overseen Rewriter), a modular, zero-shot agentic framework for safer and more helpful text-to-image generation. VALOR integrates layered prompt analysis with human-aligned value reasoning: a multi-level NSFW detector filters lexical and semantic risks; a cultural value alignment module identifies violations of social norms, legality, and representational ethics; and an intention disambiguator detects subtle or indirect unsafe implications. When unsafe content is detected, prompts are selectively rewritten by a large language model under dynamic, role-specific instructions designed to preserve user intent while enforcing alignment. If the generated image still fails a safety check, VALOR optionally performs a stylistic regeneration to steer the output toward a safer visual domain without altering core semantics. Experiments across adversarial, ambiguous, and value-sensitive prompts show that VALOR significantly reduces unsafe outputs by up to 100.00% while preserving prompt usefulness and creativity. These results highlight VALOR as a scalable and effective approach for deploying safe, aligned, and helpful image generation systems in open-world settings.

Paperid: 3978, https://arxiv.org/pdf/2511.11502.pdf

Abstract:
Large vision-language models (LVLMs) are powerful, yet they remain unreliable due to object hallucinations. In this work, we show that in many hallucinatory predictions the LVLM effectively ignores the image and instead relies on previously generated output (prelim) tokens to infer new objects. We quantify this behavior via the mutual information between the image and the predicted object conditioned on the prelim, demonstrating that weak image dependence strongly correlates with hallucination. Building on this finding, we introduce the Prelim Attention Score (PAS), a lightweight, training-free signal computed from attention weights over prelim tokens. PAS requires no additional forward passes and can be computed on the fly during inference. Exploiting this previously overlooked signal, PAS achieves state-of-the-art object-hallucination detection across multiple models and datasets, enabling real-time filtering and intervention.

Paperid: 3979, https://arxiv.org/pdf/2511.05936.pdf

Abstract:
Due to their ability of follow natural language instructions, vision-language-action (VLA) models are increasingly prevalent in the embodied AI arena, following the widespread success of their precursors -- LLMs and VLMs. In this paper, we discuss 10 principal milestones in the ongoing development of VLA models -- multimodality, reasoning, data, evaluation, cross-robot action generalization, efficiency, whole-body coordination, safety, agents, and coordination with humans. Furthermore, we discuss the emerging trends of using spatial understanding, modeling world dynamics, post training, and data synthesis -- all aiming to reach these milestones. Through these discussions, we hope to bring attention to the research avenues that may accelerate the development of VLA models into wider acceptability.

Paperid: 3980, https://arxiv.org/pdf/2511.04727.pdf

Abstract:
Vision-language models (VLMs) have demonstrated impressive generalization across multimodal tasks, yet most evaluation benchmarks remain Western-centric, leaving open questions about their performance in culturally diverse and multilingual settings. To address this gap, we introduce IndicVisionBench, the first large-scale benchmark centered on the Indian subcontinent. Covering English and 10 Indian languages, our benchmark spans 3 multimodal tasks, including Optical Character Recognition (OCR), Multimodal Machine Translation (MMT), and Visual Question Answering (VQA), covering 6 kinds of question types. Our final benchmark consists of a total of ~5K images and 37K+ QA pairs across 13 culturally grounded topics. In addition, we release a paired parallel corpus of annotations across 10 Indic languages, creating a unique resource for analyzing cultural and linguistic biases in VLMs. We evaluate a broad spectrum of 8 models, from proprietary closed-source systems to open-weights medium and large-scale models. Our experiments reveal substantial performance gaps, underscoring the limitations of current VLMs in culturally diverse contexts. By centering cultural diversity and multilinguality, IndicVisionBench establishes a reproducible evaluation framework that paves the way for more inclusive multimodal research.

Paperid: 3981, https://arxiv.org/pdf/2510.25138.pdf

Abstract:
Manipulation in cluttered environments is challenging due to spatial dependencies among objects, where an improper manipulation order can cause collisions or blocked access. Existing approaches often overlook these spatial relationships, limiting their flexibility and scalability. To address these limitations, we propose OrderMind, a unified spatial-aware manipulation ordering framework that directly learns object manipulation priorities based on spatial context. Our architecture integrates a spatial context encoder with a temporal priority structuring module. We construct a spatial graph using k-Nearest Neighbors to aggregate geometric information from the local layout and encode both object-object and object-manipulator interactions to support accurate manipulation ordering in real-time. To generate physically and semantically plausible supervision signals, we introduce a spatial prior labeling method that guides a vision-language model to produce reasonable manipulation orders for distillation. We evaluate OrderMind on our Manipulation Ordering Benchmark, comprising 163,222 samples of varying difficulty. Extensive experiments in both simulation and real-world environments demonstrate that our method significantly outperforms prior approaches in effectiveness and efficiency, enabling robust manipulation in cluttered scenes.

Paperid: 3982, https://arxiv.org/pdf/2510.23184.pdf

Abstract:
Connecting current observations with prior experiences helps robots adapt and plan in new, unseen 3D environments. Recently, 3D scene analogies have been proposed to connect two 3D scenes, which are smooth maps that align scene regions with common spatial relationships. These maps enable detailed transfer of trajectories or waypoints, potentially supporting demonstration transfer for imitation learning or task plan transfer across scenes. However, existing methods for the task require additional training and fixed object vocabularies. In this work, we propose to use multimodal foundation models for finding 3D scene analogies in a zero-shot, open-vocabulary setting. Central to our approach is a hybrid neural representation of scenes that consists of a sparse graph based on vision-language model features and a feature field derived from 3D shape foundation models. 3D scene analogies are then found in a coarse-to-fine manner, by first aligning the graph and refining the correspondence with feature fields. Our method can establish accurate correspondences between complex scenes, and we showcase applications in trajectory and waypoint transfer.

Paperid: 3983, https://arxiv.org/pdf/2510.20967.pdf

Abstract:
Current Vision-Language Models (VLMs) struggle to ground anatomical regions in 3D medical images and reason about them in a step-by-step manner, a key requirement of real-world diagnostic assessment. This ability is essential for aligning model outputs with the diagnostic workflows clinicians use in practice, enabling trustworthy clinician-AI collaboration. Existing 3D datasets provide localization labels, but none support this "grounded reasoning" ability. To address this gap, we introduce 3DReasonKnee, the first 3D grounded reasoning dataset for medical images, which provides 494k high-quality quintuples derived from 7,970 3D knee MRI volumes. Each quintuple includes: (1) the 3D MRI volume, (2) a diagnostic question targeting a specific anatomical region (3) a 3D bounding box localizing the relevant anatomical structures, (4) clinician-generated diagnostic reasoning steps that explicitly detail the 3D reasoning process, and (5) structured severity assessments for the relevant anatomical region. The creation and validation of 3DReasonKnee, involving over 450 hours of expert clinician time for manually segmenting MRIs and generating reasoning chains, ensures its superior quality and clinical relevance. We establish ReasonKnee-Bench to evaluate localization and diagnostic accuracy, providing insight into VLM ability to perform grounding and severity assessment across anatomical regions and diagnostic inquiries. We benchmark five state-of-the-art VLMs, providing baseline performance for ReasonKnee-Bench. By providing this unique resource of expert-annotated 3D reasoning pathways, 3DReasonKnee serves as a repository of orthopedic surgeons' diagnostic expertise and offers a vital testbed for advancing multimodal medical AI systems towards 3D, clinically aligned, localized decision-making capabilities. The dataset can be found in: https://huggingface.co/datasets/rajpurkarlab/3DReasonKnee

Paperid: 3984, https://arxiv.org/pdf/2510.10011.pdf

Abstract:
Currently, medical vision language models are widely used in medical vision question answering tasks. However, existing models are confronted with two issues: for input, the model only relies on text instructions and lacks direct understanding of visual clues in the image; for output, the model only gives text answers and lacks connection with key areas in the image. To address these issues, we propose a unified medical vision language model MIMO, with visual referring Multimodal Input and pixel grounding Multimodal Output. MIMO can not only combine visual clues and textual instructions to understand complex medical images and semantics, but can also ground medical terminologies in textual output within the image. To overcome the scarcity of relevant data in the medical field, we propose MIMOSeg, a comprehensive medical multimodal dataset including 895K samples. MIMOSeg is constructed from four different perspectives, covering basic instruction following and complex question answering with multimodal input and multimodal output. We conduct experiments on several downstream medical multimodal tasks. Extensive experimental results verify that MIMO can uniquely combine visual referring and pixel grounding capabilities, which are not available in previous models.

Paperid: 3985, https://arxiv.org/pdf/2510.07730.pdf

Abstract:
Offline reinforcement learning (RL) presents an attractive paradigm for training intelligent agents without expensive online interactions. However, current approaches still struggle with complex, long-horizon sequential decision making. In this work, we introduce DEtached value learning with Action Sequence (DEAS), a simple yet effective offline RL framework that leverages action sequences for value learning. These temporally extended actions provide richer information than single-step actions and can be interpreted through the options framework via semi-Markov decision process Q-learning, enabling reduction of the effective planning horizon by considering longer sequences at once. However, directly adopting such sequences in actor-critic algorithms introduces excessive value overestimation, which we address through detached value learning that steers value estimates toward in-distribution actions that achieve high return in the offline dataset. We demonstrate that DEAS consistently outperforms baselines on complex, long-horizon tasks from OGBench and can be applied to enhance the performance of large-scale Vision-Language-Action models that predict action sequences, significantly boosting performance in both RoboCasa Kitchen simulation tasks and real-world manipulation tasks.

Paperid: 3986, https://arxiv.org/pdf/2510.07709.pdf

Abstract:
Can generative agents be trusted in multimodal environments? Despite advances in large language and vision-language models that enable agents to act autonomously and pursue goals in rich settings, their ability to reason about safety, coherence, and trust across modalities remains limited. We introduce a reproducible simulation framework for evaluating agents along three dimensions: (1) safety improvement over time, including iterative plan revisions in text-visual scenarios; (2) detection of unsafe activities across multiple categories of social situations; and (3) social dynamics, measured as interaction counts and acceptance ratios of social exchanges. Agents are equipped with layered memory, dynamic planning, multimodal perception, and are instrumented with SocialMetrics, a suite of behavioral and structural metrics that quantifies plan revisions, unsafe-to-safe conversions, and information diffusion across networks. Experiments show that while agents can detect direct multimodal contradictions, they often fail to align local revisions with global safety, reaching only a 55 percent success rate in correcting unsafe plans. Across eight simulation runs with three models - Claude, GPT-4o mini, and Qwen-VL - five agents achieved average unsafe-to-safe conversion rates of 75, 55, and 58 percent, respectively. Overall performance ranged from 20 percent in multi-risk scenarios with GPT-4o mini to 98 percent in localized contexts such as fire/heat with Claude. Notably, 45 percent of unsafe actions were accepted when paired with misleading visuals, showing a strong tendency to overtrust images. These findings expose critical limitations in current architectures and provide a reproducible platform for studying multimodal safety, coherence, and social dynamics.

Paperid: 3987, https://arxiv.org/pdf/2510.04710.pdf

Abstract:
Web service administrators must ensure the stability of multiple systems by promptly detecting anomalies in Key Performance Indicators (KPIs). Achieving the goal of "train once, infer across scenarios" remains a fundamental challenge for time series anomaly detection models. Beyond improving zero-shot generalization, such models must also flexibly handle sequences of varying lengths during inference, ranging from one hour to one week, without retraining. Conventional approaches rely on sliding-window encoding and self-supervised learning, which restrict inference to fixed-length inputs. Large Language Models (LLMs) have demonstrated remarkable zero-shot capabilities across general domains. However, when applied to time series data, they face inherent limitations due to context length. To address this issue, we propose ViTs, a Vision-Language Model (VLM)-based framework that converts time series curves into visual representations. By rescaling time series images, temporal dependencies are preserved while maintaining a consistent input size, thereby enabling efficient processing of arbitrarily long sequences without context constraints. Training VLMs for this purpose introduces unique challenges, primarily due to the scarcity of aligned time series image-text data. To overcome this, we employ an evolutionary algorithm to automatically generate thousands of high-quality image-text pairs and design a three-stage training pipeline consisting of: (1) time series knowledge injection, (2) anomaly detection enhancement, and (3) anomaly reasoning refinement. Extensive experiments demonstrate that ViTs substantially enhance the ability of VLMs to understand and detect anomalies in time series data. All datasets and code will be publicly released at: https://anonymous.4open.science/r/ViTs-C484/.

Paperid: 3988, https://arxiv.org/pdf/2510.01388.pdf

Abstract:
Robots must adapt to diverse human instructions and operate safely in unstructured, open-world environments. Recent Vision-Language models (VLMs) offer strong priors for grounding language and perception, but remain difficult to steer for navigation due to differences in action spaces and pretraining objectives that hamper transferability to robotics tasks. Towards addressing this, we introduce VENTURA, a vision-language navigation system that finetunes internet-pretrained image diffusion models for path planning. Instead of directly predicting low-level actions, VENTURA generates a path mask (i.e. a visual plan) in image space that captures fine-grained, context-aware navigation behaviors. A lightweight behavior-cloning policy grounds these visual plans into executable trajectories, yielding an interface that follows natural language instructions to generate diverse robot behaviors. To scale training, we supervise on path masks derived from self-supervised tracking models paired with VLM-augmented captions, avoiding manual pixel-level annotation or highly engineered data collection setups. In extensive real-world evaluations, VENTURA outperforms state-of-the-art foundation model baselines on object reaching, obstacle avoidance, and terrain preference tasks, improving success rates by 33% and reducing collisions by 54% across both seen and unseen scenarios. Notably, we find that VENTURA generalizes to unseen combinations of distinct tasks, revealing emergent compositional capabilities. Videos, code, and additional materials: https://venturapath.github.io

Paperid: 3989, https://arxiv.org/pdf/2509.26278.pdf

Abstract:
Existing approaches to skill proficiency estimation often rely on black-box video classifiers, ignoring multi-view context and lacking explainability. We present ProfVLM, a compact vision-language model that reformulates this task as generative reasoning: it jointly predicts skill level and generates expert-like feedback from egocentric and exocentric videos. Central to our method is an AttentiveGatedProjector that dynamically fuses multi-view features, projected from a frozen TimeSformer backbone into a language model tuned for feedback generation. Trained on EgoExo4D with expert commentaries, ProfVLM surpasses state-of-the-art methods while using up to 20x fewer parameters and reducing training time by up to 60%. Our approach not only achieves superior accuracy across diverse activities, but also outputs natural language critiques aligned with performance, offering transparent reasoning. These results highlight generative vision-language modeling as a powerful new direction for skill assessment.

Paperid: 3990, https://arxiv.org/pdf/2509.25584.pdf

Abstract:
Vision-language models (VLMs) achieve incredible performance across a wide range of tasks, but their large size makes inference costly. Recent work shows that selectively skipping VLM layers can improve efficiency with minimal performance loss or even performance improvements. However, this technique remains underused due to the limited understanding of when layer skipping is beneficial. In this paper, we develop a framework that uses information and learning theory to characterize the conditions under which layer skipping enhances efficiency without sacrificing performance. Motivated by these observations, we analyze the evolution of the VLM's hidden representations through the LLM backbone and show that layers with large redundancy as predicted by our framework coincide with those skipped by popular layer-skipping methods in practice, providing a unified theoretical scaffolding for multiple efficient inference techniques. Our experiments demonstrate that skipping such layers yields faster inference that preserves performance, and also show that applying skipping outside these conditions leads to model degradation.

Paperid: 3991, https://arxiv.org/pdf/2509.22820.pdf

Abstract:
Visual personalization is essential in user-facing AI systems such as smart homes and healthcare, where aligning model behavior with user-centric concepts is critical. However, recent large Vision-Language Models (VLMs), despite their broad applicability, remain underexplored in their ability to adapt to individual users. In this paper, we introduce MMPB, the first extensive benchmark for evaluating VLMs on personalization. MMPB comprises 10k image-query pairs and includes 111 personalizable concepts across four categories: humans, animals, objects, and characters, with the human category enriched with preference-grounded queries. We structure personalization into three main task types, each highlighting a different key property of VLMs. Using 23 widely used VLMs including both open- and closed-source models, we evaluate personalization performance via a three-stage protocol: concept injection, multi-turn dialogue, and personalized querying. Our findings indicate that most VLMs (including some closed-source models) struggle with personalization, particularly in maintaining consistency over dialogue, handling user preferences, and adapting to visual cues. Our analysis reveals that the challenges in VLM personalization (such as refusal behaviors and long-context forgetting) highlight substantial room for improvement. By identifying these limitations and offering a scalable benchmark, MMPB offers valuable insights and a solid foundation for future research toward truly personalized multi-modal AI. Project Page: aidaslab.github.io/MMPB

Paperid: 3992, https://arxiv.org/pdf/2509.21723.pdf

Abstract:
Achieving generalizable bimanual manipulation requires systems that can learn efficiently from minimal human input while adapting to real-world uncertainties and diverse embodiments. Existing approaches face a dilemma: imitation policy learning demands extensive demonstrations to cover task variations, while modular methods often lack flexibility in dynamic scenes. We introduce VLBiMan, a framework that derives reusable skills from a single human example through task-aware decomposition, preserving invariant primitives as anchors while dynamically adapting adjustable components via vision-language grounding. This adaptation mechanism resolves scene ambiguities caused by background changes, object repositioning, or visual clutter without policy retraining, leveraging semantic parsing and geometric feasibility constraints. Moreover, the system inherits human-like hybrid control capabilities, enabling mixed synchronous and asynchronous use of both arms. Extensive experiments validate VLBiMan across tool-use and multi-object tasks, demonstrating: (1) a drastic reduction in demonstration requirements compared to imitation baselines, (2) compositional generalization through atomic skill splicing for long-horizon tasks, (3) robustness to novel but semantically similar objects and external disturbances, and (4) strong cross-embodiment transfer, showing that skills learned from human demonstrations can be instantiated on different robotic platforms without retraining. By bridging human priors with vision-language anchored adaptation, our work takes a step toward practical and versatile dual-arm manipulation in unstructured settings.

Paperid: 3993, https://arxiv.org/pdf/2509.18763.pdf

Abstract:
We address the critical gap between the computational demands of vision-language models and the possible ultra-low-bit weight precision (bitwidth $\leq2$ bits) we can use for higher efficiency. Our work is motivated by the substantial computational cost and memory requirements of VLMs, which restrict their applicability in hardware-constrained environments. We propose Bi-VLM, which separates model weights non-uniformly based on the Gaussian quantiles. Our formulation groups the model weights into outlier (salient) and multiple inlier (unsalient) subsets, ensuring that each subset contains a proportion of weights corresponding to its quantile in the distribution. We propose a saliency-aware hybrid quantization algorithm and use it to quantize weights by imposing different constraints on the scaler and binary matrices based on the saliency metric and compression objective. We have evaluated our approach on different VLMs. For the language model part of the VLM, our Bi-VLM outperforms the SOTA by 3%-47% on the visual question answering task in terms of four different benchmarks and three different models. For the overall VLM, our Bi-VLM outperforms the SOTA by 4%-45%. We also perform token pruning on the quantized models and observe that there is redundancy of image tokens 90% - 99% in the quantized models. This helps us to further prune the visual tokens to improve efficiency.

Paperid: 3994, https://arxiv.org/pdf/2509.09286.pdf

Abstract:
Chart understanding presents a critical test to the reasoning capabilities of Vision-Language Models (VLMs). Prior approaches face critical limitations: some rely on external tools, making them brittle and constrained by a predefined toolkit, while others fine-tune specialist models that often adopt a single reasoning strategy, such as text-based chain-of-thought (CoT). The intermediate steps of text-based reasoning are difficult to verify, which complicates the use of reinforcement-learning signals that reward factual accuracy. To address this, we propose a Code-as-Thought (CaT) approach to represent the visual information of a chart in a verifiable, symbolic format. Our key insight is that this strategy must be adaptive: a fixed, code-only implementation consistently fails on complex charts where symbolic representation is unsuitable. This finding leads us to introduce Visual Programmability: a learnable property that determines if a chart-question pair is better solved with code or direct visual analysis. We implement this concept in an adaptive framework where a VLM learns to choose between the CaT pathway and a direct visual reasoning pathway. The selection policy of the model is trained with reinforcement learning using a novel dual-reward system. This system combines a data-accuracy reward to ground the model in facts and prevent numerical hallucination, with a decision reward that teaches the model when to use each strategy, preventing it from defaulting to a single reasoning mode. Experiments demonstrate strong and robust performance across diverse chart-understanding benchmarks. Our work shows that VLMs can be taught not only to reason but also how to reason, dynamically selecting the optimal reasoning pathway for each task.

Paperid: 3995, https://arxiv.org/pdf/2509.09172.pdf

Abstract:
With the rapid advancement of generative models, highly realistic image synthesis has posed new challenges to digital security and media credibility. Although AI-generated image detection methods have partially addressed these concerns, a substantial research gap remains in evaluating their performance under complex real-world conditions. This paper introduces the Real-World Robustness Dataset (RRDataset) for comprehensive evaluation of detection models across three dimensions: 1) Scenario Generalization: RRDataset encompasses high-quality images from seven major scenarios (War and Conflict, Disasters and Accidents, Political and Social Events, Medical and Public Health, Culture and Religion, Labor and Production, and everyday life), addressing existing dataset gaps from a content perspective. 2) Internet Transmission Robustness: examining detector performance on images that have undergone multiple rounds of sharing across various social media platforms. 3) Re-digitization Robustness: assessing model effectiveness on images altered through four distinct re-digitization methods. We benchmarked 17 detectors and 10 vision-language models (VLMs) on RRDataset and conducted a large-scale human study involving 192 participants to investigate human few-shot learning capabilities in detecting AI-generated images. The benchmarking results reveal the limitations of current AI detection methods under real-world conditions and underscore the importance of drawing on human adaptability to develop more robust detection algorithms.

Paperid: 3996, https://arxiv.org/pdf/2509.06409.pdf

Abstract:
This study presents DiagCoT, a multi-stage framework that applies supervised fine-tuning to general-purpose vision-language models (VLMs) to emulate radiologists' stepwise diagnostic reasoning using only free-text reports. DiagCoT combines contrastive image-report tuning for domain alignment, chain-of-thought supervision to capture inferential logic, and reinforcement tuning with clinical reward signals to enhance factual accuracy and fluency. On the MIMIC-CXR benchmark, DiagCoT improved zero-shot disease classification AUC from 0.52 to 0.76 (absolute gain of 0.24), pathology grounding mIoU from 0.08 to 0.31 (absolute gain of 0.23), and report generation BLEU from 0.11 to 0.33 (absolute gain of 0.22). It outperformed state-of-the-art models including LLaVA-Med and CXR-LLAVA on long-tailed diseases and external datasets. By converting unstructured clinical narratives into structured supervision, DiagCoT offers a scalable approach for developing interpretable and diagnostically competent AI systems for radiology.

Paperid: 3997, https://arxiv.org/pdf/2509.05000.pdf

Abstract:
Most existing infrared-visible image fusion (IVIF) methods assume high-quality inputs, and therefore struggle to handle dual-source degraded scenarios, typically requiring manual selection and sequential application of multiple pre-enhancement steps. This decoupled pre-enhancement-to-fusion pipeline inevitably leads to error accumulation and performance degradation. To overcome these limitations, we propose Guided Dual-Domain Fusion (GD^2Fusion), a novel framework that synergistically integrates vision-language models (VLMs) for degradation perception with dual-domain (frequency/spatial) joint optimization. Concretely, the designed Guided Frequency Modality-Specific Extraction (GFMSE) module performs frequency-domain degradation perception and suppression and discriminatively extracts fusion-relevant sub-band features. Meanwhile, the Guided Spatial Modality-Aggregated Fusion (GSMAF) module carries out cross-modal degradation filtering and adaptive multi-source feature aggregation in the spatial domain to enhance modality complementarity and structural consistency. Extensive qualitative and quantitative experiments demonstrate that GD^2Fusion achieves superior fusion performance compared with existing algorithms and strategies in dual-source degraded scenarios. The code will be publicly released after acceptance of this paper.

Paperid: 3998, https://arxiv.org/pdf/2509.02966.pdf

Abstract:
Accurate short-horizon trajectory prediction is pivotal for safe and reliable autonomous driving, yet existing vision-language models (VLMs) often fail to effectively ground their reasoning in scene dynamics and domain knowledge. To address this challenge, this paper introduces KEPT, a knowledge-enhanced VLM framework that predicts ego trajectories directly from consecutive front-view driving frames. KEPT couples a temporal frequency-spatial fusion (TFSF) video encoder, trained via self-supervised learning with hard-negative mining, with a scalable k-means + HNSW retrieval stack that supplies scene-aligned exemplars. Retrieved priors are embedded into chain-of-thought (CoT) prompts with explicit planning constraints, while a triple-stage fine-tuning schedule incrementally aligns the language head to metric spatial cues, physically feasible motion, and temporally conditioned front-view planning. Evaluated on nuScenes dataset, KEPT achieves state-of-the-art performance across open-loop protocols: under NoAvg, it achieves 0.70m average L2 with a 0.21\% collision rate; under TemAvg with lightweight ego status, it attains 0.31m average L2 and a 0.07\% collision rate. Ablation studies show that all three fine-tuning stages contribute complementary benefits, and that using Top-2 retrieved exemplars yields the best accuracy-safety trade-off. The k-means-clustered HNSW index delivers sub-millisecond retrieval latency, supporting practical deployment. These results indicate that retrieval-augmented, CoT-guided VLMs offer a promising, data-efficient pathway toward interpretable and trustworthy autonomous driving.

Paperid: 3999, https://arxiv.org/pdf/2509.02273.pdf

Abstract:
Out-of-distribution (OOD) detection represents a critical challenge in remote sensing applications, where reliable identification of novel or anomalous patterns is essential for autonomous monitoring, disaster response, and environmental assessment. Despite remarkable progress in OOD detection for natural images, existing methods and benchmarks remain poorly suited to remote sensing imagery due to data scarcity, complex multi-scale scene structures, and pronounced distribution shifts. To this end, we propose RS-OOD, a novel framework that leverages remote sensing-specific vision-language modeling to enable robust few-shot OOD detection. Our approach introduces three key innovations: spatial feature enhancement that improved scene discrimination, a dual-prompt alignment mechanism that cross-verifies scene context against fine-grained semantics for spatial-semantic consistency, and a confidence-guided self-training loop that dynamically mines pseudo-labels to expand training data without manual annotation. RS-OOD consistently outperforms existing methods across multiple remote sensing benchmarks and enables efficient adaptation with minimal labeled data, demonstrating the critical value of spatial-semantic integration.

Paperid: 4000, https://arxiv.org/pdf/2508.21066.pdf

Abstract:
In this paper, we introduce OneReward, a unified reinforcement learning framework that enhances the model's generative capabilities across multiple tasks under different evaluation criteria using only \textit{One Reward} model. By employing a single vision-language model (VLM) as the generative reward model, which can distinguish the winner and loser for a given task and a given evaluation criterion, it can be effectively applied to multi-task generation models, particularly in contexts with varied data and diverse task objectives. We utilize OneReward for mask-guided image generation, which can be further divided into several sub-tasks such as image fill, image extend, object removal, and text rendering, involving a binary mask as the edit area. Although these domain-specific tasks share same conditioning paradigm, they differ significantly in underlying data distributions and evaluation metrics. Existing methods often rely on task-specific supervised fine-tuning (SFT), which limits generalization and training efficiency. Building on OneReward, we develop Seedream 3.0 Fill, a mask-guided generation model trained via multi-task reinforcement learning directly on a pre-trained base model, eliminating the need for task-specific SFT. Experimental results demonstrate that our unified edit model consistently outperforms both commercial and open-source competitors, such as Ideogram, Adobe Photoshop, and FLUX Fill [Pro], across multiple evaluation dimensions. Code and model are available at: https://one-reward.github.io

Paperid: 4001, https://arxiv.org/pdf/2508.20381.pdf

Abstract:
Multi-label learning is a challenging computer vision task that requires assigning multiple categories to each image. However, fully annotating large-scale datasets is often impractical due to high costs and effort, motivating the study of learning from partially annotated data. In the extreme case of Single Positive Multi-Label Learning (SPML), each image is provided with only one positive label, while all other labels remain unannotated. Traditional SPML methods that treat missing labels as unknown or negative tend to yield inaccuracies and false negatives, and integrating various pseudo-labeling strategies can introduce additional noise. To address these challenges, we propose the Generalized Pseudo-Label Robust Loss (GPR Loss), a novel loss function that effectively learns from diverse pseudo-labels while mitigating noise. Complementing this, we introduce a simple yet effective Dynamic Augmented Multi-focus Pseudo-labeling (DAMP) technique. Together, these contributions form the Adaptive and Efficient Vision-Language Pseudo-Labeling (AEVLP) framework. Extensive experiments on four benchmark datasets demonstrate that our framework significantly advances multi-label classification, achieving state-of-the-art results.

Paperid: 4002, https://arxiv.org/pdf/2508.17417.pdf

Abstract:
Vision-language models (VLMs) pre-trained on web-scale data exhibit promising zero-shot generalization but often suffer from semantic misalignment due to domain gaps between pre-training and downstream tasks. Existing approaches primarily focus on text prompting with class-specific descriptions and visual-text adaptation via aligning cropped image regions with textual descriptions. However, they still face the issues of incomplete textual prompts and noisy visual prompts. In this paper, we propose a novel constrained prompt enhancement (CPE) method to improve visual-textual alignment by constructing comprehensive textual prompts and compact visual prompts from the semantic perspective. Specifically, our approach consists of two key components: Topology-Guided Synonymous Semantic Generation (TGSSG) and Category-Agnostic Discriminative Region Selection (CADRS). Textually, to address the issue of incomplete semantic expression in textual prompts, our TGSSG first generates synonymous semantic set for each category via large language models, and constructs comprehensive textual prompts based on semantic ambiguity entropy and persistent homology analysis. Visually, to mitigate the irrelevant visual noise introduced by random cropping, our CADRS identifies discriminative regions with activation maps outputted by a pre-trained vision model, effectively filtering out noisy regions and generating compact visual prompts. Given the comprehensive set of textual prompts and compact set of visual prompts, we introduce two set-to-set matching strategies based on test-time adaptation (TTA) and optimal transport (OT) to achieve effective visual-textual alignment, and so improve zero-shot generalization of VLMs.

Paperid: 4003, https://arxiv.org/pdf/2508.11918.pdf

Abstract:
The advancement of embodied intelligence is accelerating the integration of robots into daily life as human assistants. This evolution requires robots to not only interpret high-level instructions and plan tasks but also perceive and adapt within dynamic environments. Vision-Language Models (VLMs) present a promising solution by combining visual understanding and language reasoning. However, existing VLM-based methods struggle with interactive exploration, accurate perception, and real-time plan adaptation. To address these challenges, we propose ExploreVLM, a novel closed-loop task planning framework powered by Vision-Language Models (VLMs). The framework is built around a step-wise feedback mechanism that enables real-time plan adjustment and supports interactive exploration. At its core is a dual-stage task planner with self-reflection, enhanced by an object-centric spatial relation graph that provides structured, language-grounded scene representations to guide perception and planning. An execution validator supports the closed loop by verifying each action and triggering re-planning. Extensive real-world experiments demonstrate that ExploreVLM significantly outperforms state-of-the-art baselines, particularly in exploration-centric tasks. Ablation studies further validate the critical role of the reflective planner and structured perception in achieving robust and efficient task execution.

Paperid: 4004, https://arxiv.org/pdf/2508.10444.pdf

Abstract:
Generating textual rationales from large vision-language models (LVLMs) to support trainable multimodal misinformation detectors has emerged as a promising paradigm. However, its effectiveness is fundamentally limited by three core challenges: (i) insufficient diversity in generated rationales, (ii) factual inaccuracies due to hallucinations, and (iii) irrelevant or conflicting content that introduces noise. We introduce DiFaR, a detector-agnostic framework that produces diverse, factual, and relevant rationales to enhance misinformation detection. DiFaR employs five chain-of-thought prompts to elicit varied reasoning traces from LVLMs and incorporates a lightweight post-hoc filtering module to select rationale sentences based on sentence-level factuality and relevance scores. Extensive experiments on four popular benchmarks demonstrate that DiFaR outperforms four baseline categories by up to 5.9% and boosts existing detectors by as much as 8.7%. Both automatic metrics and human evaluations confirm that DiFaR significantly improves rationale quality across all three dimensions.

Paperid: 4005, https://arxiv.org/pdf/2508.09922.pdf

Abstract:
Diffusion models have emerged as a leading framework for high-quality image generation, offering stable training and strong performance across diverse domains. However, they remain computationally intensive, particularly during the iterative denoising process. Latent-space models like Stable Diffusion alleviate some of this cost by operating in compressed representations, though at the expense of fine-grained detail. More recent approaches such as Retrieval-Augmented Diffusion Models (RDM) address efficiency by conditioning denoising on similar examples retrieved from large external memory banks. While effective, these methods introduce drawbacks: they require costly storage and retrieval infrastructure, depend on static vision-language models like CLIP for similarity, and lack adaptability during training. We propose the Prototype Diffusion Model (PDM), a method that integrates prototype learning directly into the diffusion process for efficient and adaptive visual conditioning - without external memory. Instead of retrieving reference samples, PDM constructs a dynamic set of compact visual prototypes from clean image features using contrastive learning. These prototypes guide the denoising steps by aligning noisy representations with semantically relevant visual patterns, enabling efficient generation with strong semantic grounding. Experiments show that PDM maintains high generation quality while reducing computational and storage overhead, offering a scalable alternative to retrieval-based conditioning in diffusion models.

Paperid: 4006, https://arxiv.org/pdf/2508.09185.pdf

Abstract:
Augmented Reality (AR) enriches perception by overlaying virtual elements on the physical world. Due to its growing popularity, cognitive attacks that alter AR content to manipulate users' semantic perception have received increasing attention. Existing detection methods often focus on visual changes, which are restricted to pixel- or image-level processing and lack semantic reasoning capabilities, or they rely on pre-trained vision-language models (VLMs), which function as black-box approaches with limited interpretability. In this paper, we present CADAR, a novel neurosymbolic approach for cognitive attack detection in AR. It fuses multimodal vision-language inputs using neural VLMs to obtain a symbolic perception-graph representation, incorporating prior knowledge, salience weighting, and temporal correlations. The model then enables particle-filter based statistical reasoning -- a sequential Monte Carlo method -- to detect cognitive attacks. Thus, CADAR inherits the adaptability of pre-trained VLM and the interpretability and reasoning rigor of particle filtering. Experiments on an extended AR cognitive attack dataset show accuracy improvements of up to 10.7% over strong baselines on challenging AR attack scenarios, underscoring the promise of neurosymbolic methods for effective and interpretable cognitive attack detection.

Paperid: 4007, https://arxiv.org/pdf/2508.04571.pdf

Abstract:
Multimodal Recommender Systems aim to improve recommendation accuracy by integrating heterogeneous content, such as images and textual metadata. While effective, it remains unclear whether their gains stem from true multimodal understanding or increased model complexity. This work investigates the role of multimodal item embeddings, emphasizing the semantic informativeness of the representations. Initial experiments reveal that embeddings from standard extractors (e.g., ResNet50, Sentence-Bert) enhance performance, but rely on modality-specific encoders and ad hoc fusion strategies that lack control over cross-modal alignment. To overcome these limitations, we leverage Large Vision-Language Models (LVLMs) to generate multimodal-by-design embeddings via structured prompts. This approach yields semantically aligned representations without requiring any fusion. Experiments across multiple settings show notable performance improvements. Furthermore, LVLMs embeddings offer a distinctive advantage: they can be decoded into structured textual descriptions, enabling direct assessment of their multimodal comprehension. When such descriptions are incorporated as side content into recommender systems, they improve recommendation performance, empirically validating the semantic depth and alignment encoded within LVLMs outputs. Our study highlights the importance of semantically rich representations and positions LVLMs as a compelling foundation for building robust and meaningful multimodal representations in recommendation tasks.

Paperid: 4008, https://arxiv.org/pdf/2507.20356.pdf

Abstract:
The virtual content in augmented reality (AR) can introduce misleading or harmful information, leading to semantic misunderstandings or user errors. In this work, we focus on visual information manipulation (VIM) attacks in AR, where virtual content changes the meaning of real-world scenes in subtle but impactful ways. We introduce a taxonomy that categorizes these attacks into three formats: character, phrase, and pattern manipulation, and three purposes: information replacement, information obfuscation, and extra wrong information. Based on the taxonomy, we construct a dataset, AR-VIM, which consists of 452 raw-AR video pairs spanning 202 different scenes, each simulating a real-world AR scenario. To detect the attacks in the dataset, we propose a multimodal semantic reasoning framework, VIM-Sense. It combines the language and visual understanding capabilities of vision-language models (VLMs) with optical character recognition (OCR)-based textual analysis. VIM-Sense achieves an attack detection accuracy of 88.94% on AR-VIM, consistently outperforming vision-only and text-only baselines. The system achieves an average attack detection latency of 7.07 seconds in a simulated video processing framework and 7.17 seconds in a real-world evaluation conducted on a mobile Android AR application.

Paperid: 4009, https://arxiv.org/pdf/2507.04699.pdf

Abstract:
Vision-language models (VLMs) often struggle with compositional reasoning due to insufficient high-quality image-text data. To tackle this challenge, we propose a novel block-based diffusion approach that automatically generates counterfactual datasets without manual annotation. Our method utilizes large language models to identify entities and their spatial relationships. It then independently generates image blocks as "puzzle pieces" coherently arranged according to specified compositional rules. This process creates diverse, high-fidelity counterfactual image-text pairs with precisely controlled variations. In addition, we introduce a specialized loss function that differentiates inter-set from intra-set samples, enhancing training efficiency and reducing the need for negative samples. Experiments demonstrate that fine-tuning VLMs with our counterfactual datasets significantly improves visual reasoning performance. Our approach achieves state-of-the-art results across multiple benchmarks while using substantially less training data than existing methods.

Paperid: 4010, https://arxiv.org/pdf/2506.20100.pdf

Abstract:
We introduce MIRAGE, a new benchmark for multimodal expert-level reasoning and decision-making in consultative interaction settings. Designed for the agriculture domain, MIRAGE captures the full complexity of expert consultations by combining natural user queries, expert-authored responses, and image-based context, offering a high-fidelity benchmark for evaluating models on grounded reasoning, clarification strategies, and long-form generation in a real-world, knowledge-intensive domain. Grounded in over 35,000 real user-expert interactions and curated through a carefully designed multi-step pipeline, MIRAGE spans diverse crop health, pest diagnosis, and crop management scenarios. The benchmark includes more than 7,000 unique biological entities, covering plant species, pests, and diseases, making it one of the most taxonomically diverse benchmarks available for vision-language models, grounded in the real world. Unlike existing benchmarks that rely on well-specified user inputs and closed-set taxonomies, MIRAGE features underspecified, context-rich scenarios with open-world settings, requiring models to infer latent knowledge gaps, handle rare entities, and either proactively guide the interaction or respond. Project Page: https://mirage-benchmark.github.io

Paperid: 4011, https://arxiv.org/pdf/2506.18946.pdf

Abstract:
Referring remote sensing image segmentation (RRSIS) enables the precise delineation of regions within remote sensing imagery through natural language descriptions, serving critical applications in disaster response, urban development, and environmental monitoring. Despite recent advances, current approaches face significant challenges in processing aerial imagery due to complex object characteristics including scale variations, diverse orientations, and semantic ambiguities inherent to the overhead perspective. To address these limitations, we propose DiffRIS, a novel framework that harnesses the semantic understanding capabilities of pre-trained text-to-image diffusion models for enhanced cross-modal alignment in RRSIS tasks. Our framework introduces two key innovations: a context perception adapter (CP-adapter) that dynamically refines linguistic features through global context modeling and object-aware reasoning, and a progressive cross-modal reasoning decoder (PCMRD) that iteratively aligns textual descriptions with visual regions for precise segmentation. The CP-adapter bridges the domain gap between general vision-language understanding and remote sensing applications, while PCMRD enables fine-grained semantic alignment through multi-scale feature interaction. Comprehensive experiments on three benchmark datasets-RRSIS-D, RefSegRS, and RISBench-demonstrate that DiffRIS consistently outperforms existing methods across all standard metrics, establishing a new state-of-the-art for RRSIS tasks. The significant performance improvements validate the effectiveness of leveraging pre-trained diffusion models for remote sensing applications through our proposed adaptive framework.

Paperid: 4012, https://arxiv.org/pdf/2506.15084.pdf

Abstract:
Data visualization (DataViz) libraries play a crucial role in presentation, data analysis, and application development, underscoring the importance of their accuracy in transforming data into visual representations. Incorrect visualizations can adversely impact user experience, distort information conveyance, and influence user perception and decision-making processes. Visual bugs in these libraries can be particularly insidious as they may not cause obvious errors like crashes, but instead mislead users of the underlying data graphically, resulting in wrong decision making. Consequently, a good understanding of the unique characteristics of bugs in DataViz libraries is essential for researchers and developers to detect and fix bugs in DataViz libraries. This study presents the first comprehensive analysis of bugs in DataViz libraries, examining 564 bugs collected from five widely-used libraries. Our study systematically analyzes their symptoms and root causes, and provides a detailed taxonomy. We found that incorrect/inaccurate plots are pervasive in DataViz libraries and incorrect graphic computation is the major root cause, which necessitates further automated testing methods for DataViz libraries. Moreover, we identified eight key steps to trigger such bugs and two test oracles specific to DataViz libraries, which may inspire future research in designing effective automated testing techniques. Furthermore, with the recent advancements in Vision Language Models (VLMs), we explored the feasibility of applying these models to detect incorrect/inaccurate plots. The results show that the effectiveness of VLMs in bug detection varies from 29% to 57%, depending on the prompts, and adding more information in prompts does not necessarily increase the effectiveness. More findings can be found in our manuscript.

Paperid: 4013, https://arxiv.org/pdf/2506.13038.pdf

Abstract:
Driven by the rapid progress in vision-language models (VLMs), the responsible behavior of large-scale multimodal models has become a prominent research area, particularly focusing on hallucination detection and factuality checking. In this paper, we present the solution for the two tracks of Responsible AI challenge. Inspirations from the general domain demonstrate that a smaller distilled VLM can often outperform a larger VLM that is directly tuned on downstream tasks, while achieving higher efficiency. We thus jointly tackle two tasks from the perspective of knowledge distillation and propose a progressive hybrid knowledge distillation framework termed HKD4VLM. Specifically, the overall framework can be decomposed into Pyramid-like Progressive Online Distillation and Ternary-Coupled Refinement Distillation, hierarchically moving from coarse-grained knowledge alignment to fine-grained refinement. Besides, we further introduce the mapping shift-enhanced inference and diverse augmentation strategies to enhance model performance and robustness. Extensive experimental results demonstrate the effectiveness of our HKD4VLM. Ablation studies provide insights into the critical design choices driving performance gains.

Paperid: 4014, https://arxiv.org/pdf/2506.11976.pdf

Abstract:
Effective multimodal reasoning depends on the alignment of visual and linguistic representations, yet the mechanisms by which vision-language models (VLMs) achieve this alignment remain poorly understood. Following the LiMBeR framework, we deliberately maintain a frozen large language model (LLM) and a frozen vision transformer (ViT), connected solely by training a linear adapter during visual instruction tuning. By keeping the language model frozen, we ensure it maintains its original language representations without adaptation to visual data. Consequently, the linear adapter must map visual features directly into the LLM's existing representational space rather than allowing the language model to develop specialized visual understanding through fine-tuning. Our experimental design uniquely enables the use of pre-trained sparse autoencoders (SAEs) of the LLM as analytical probes. These SAEs remain perfectly aligned with the unchanged language model and serve as a snapshot of the learned language feature-representations. Through systematic analysis of SAE reconstruction error, sparsity patterns, and feature SAE descriptions, we reveal the layer-wise progression through which visual representations gradually align with language feature representations, converging in middle-to-later layers. This suggests a fundamental misalignment between ViT outputs and early LLM layers, raising important questions about whether current adapter-based architectures optimally facilitate cross-modal representation learning.

Paperid: 4015, https://arxiv.org/pdf/2506.08391.pdf

Abstract:
Despite significant advancements in Vision-Language Models (VLMs), the performance of existing VLMs remains hindered by object hallucination, a critical challenge to achieving accurate visual understanding. To address this issue, we propose SECOND: Selective and Contrastive Decoding, a novel approach that enables VLMs to effectively leverage multi-scale visual information with an object-centric manner, closely aligning with human visual perception. SECOND progressively selects and integrates multi-scale visual information, facilitating a more precise interpretation of images. By contrasting these visual information iteratively, SECOND significantly reduces perceptual hallucinations and outperforms a wide range of benchmarks. Our theoretical analysis and experiments highlight the largely unexplored potential of multi-scale application in VLMs, showing that prioritizing and contrasting across scales outperforms existing methods.

Paperid: 4016, https://arxiv.org/pdf/2506.07339.pdf

Abstract:
Modern AI systems, especially those interacting with the physical world, increasingly require real-time performance. However, the high latency of state-of-the-art generalist models, including recent vision-language action models (VLAs), poses a significant challenge. While action chunking has enabled temporal consistency in high-frequency control tasks, it does not fully address the latency problem, leading to pauses or out-of-distribution jerky movements at chunk boundaries. This paper presents a novel inference-time algorithm that enables smooth asynchronous execution of action chunking policies. Our method, real-time chunking (RTC), is applicable to any diffusion- or flow-based VLA out of the box with no re-training. It generates the next action chunk while executing the current one, "freezing" actions guaranteed to execute and "inpainting" the rest. To test RTC, we introduce a new benchmark of 12 highly dynamic tasks in the Kinetix simulator, as well as evaluate 6 challenging real-world bimanual manipulation tasks. Results demonstrate that RTC is fast, performant, and uniquely robust to inference delay, significantly improving task throughput and enabling high success rates in precise tasks $\unicode{x2013}$ such as lighting a match $\unicode{x2013}$ even in the presence of significant latency. See https://pi.website/research/real_time_chunking for videos.

Paperid: 4017, https://arxiv.org/pdf/2506.06220.pdf

Abstract:
Vision-language models (VLMs) have shown strong performance on text-to-image retrieval benchmarks. However, bridging this success to real-world applications remains a challenge. In practice, human search behavior is rarely a one-shot action. Instead, it is often a multi-round process guided by clues in mind, that is, a mental image ranging from vague recollections to vivid mental representations of the target image. Motivated by this gap, we study the task of Mental Image Retrieval (MIR), which targets the realistic yet underexplored setting where users refine their search for a mentally envisioned image through multi-round interactions with an image search engine. Central to successful interactive retrieval is the capability of machines to provide users with clear, actionable feedback; however, existing methods rely on indirect or abstract verbal feedback, which can be ambiguous, misleading, or ineffective for users to refine the query. To overcome this, we propose GenIR, a generative multi-round retrieval paradigm leveraging diffusion-based image generation to explicitly reify the AI system's understanding at each round. These synthetic visual representations provide clear, interpretable feedback, enabling users to refine their queries intuitively and effectively. We further introduce a fully automated pipeline to generate a high-quality multi-round MIR dataset. Experimental results demonstrate that GenIR significantly outperforms existing interactive methods in the MIR scenario. This work establishes a new task with a dataset and an effective generative retrieval method, providing a foundation for future research in this direction.

Paperid: 4018, https://arxiv.org/pdf/2505.19031.pdf

Abstract:
Medical large vision-language models (LVLMs) have demonstrated promising performance across various single-image question answering (QA) benchmarks, yet their capability in processing multi-image clinical scenarios remains underexplored. Unlike single image based tasks, medical tasks involving multiple images often demand sophisticated visual understanding capabilities, such as temporal reasoning and cross-modal analysis, which are poorly supported by current medical LVLMs. To bridge this critical gap, we present the Med-MIM instruction dataset, comprising 83.2K medical multi-image QA pairs that span four types of multi-image visual abilities (temporal understanding, reasoning, comparison, co-reference). Using this dataset, we fine-tune Mantis and LLaVA-Med, resulting in two specialized medical VLMs: MIM-LLaVA-Med and Med-Mantis, both optimized for multi-image analysis. Additionally, we develop the Med-MIM benchmark to comprehensively evaluate the medical multi-image understanding capabilities of LVLMs. We assess eight popular LVLMs, including our two models, on the Med-MIM benchmark. Experimental results show that both Med-Mantis and MIM-LLaVA-Med achieve superior performance on the held-in and held-out subsets of the Med-MIM benchmark, demonstrating that the Med-MIM instruction dataset effectively enhances LVLMs' multi-image understanding capabilities in the medical domain.

Paperid: 4019, https://arxiv.org/pdf/2505.16279.pdf

Abstract:
Current movie dubbing technology can produce the desired speech using a reference voice and input video, maintaining perfect synchronization with the visuals while effectively conveying the intended emotions. However, crucial aspects of movie dubbing, including adaptation to various dubbing styles, effective handling of dialogue, narration, and monologues, as well as consideration of subtle details such as speaker age and gender, remain insufficiently explored. To tackle these challenges, we introduce a multi-modal generative framework. First, it utilizes a multi-modal large vision-language model (VLM) to analyze visual inputs, enabling the recognition of dubbing types and fine-grained attributes. Second, it produces high-quality dubbing using large speech generation models, guided by multi-modal inputs. Additionally, a movie dubbing dataset with annotations for dubbing types and subtle details is constructed to enhance movie understanding and improve dubbing quality for the proposed multi-modal framework. Experimental results across multiple benchmark datasets show superior performance compared to state-of-the-art (SOTA) methods. In details, the LSE-D, SPK-SIM, EMO-SIM, and MCD exhibit improvements of up to 1.09%, 8.80%, 19.08%, and 18.74%, respectively.

Paperid: 4020, https://arxiv.org/pdf/2505.15925.pdf

Abstract:
While autonomous driving (AD) stacks struggle with decision making under partial observability and real-world complexity, human drivers are capable of commonsense reasoning to make near-optimal decisions with limited information. Recent work has attempted to leverage finetuned Vision-Language Models (VLMs) for trajectory planning at inference time to emulate human behavior. Despite their success in benchmark evaluations, these methods are often impractical to deploy (a 70B parameter VLM inference at merely 8 tokens per second requires more than 160G of memory), and their monolithic network structure prohibits safety decomposition. To bridge this gap, we propose VLM-Embedded Reasoning for autonomous Driving (VERDI), a training-time framework that distills the reasoning process and commonsense knowledge of VLMs into the AD stack. VERDI augments modular differentiable end-to-end (e2e) AD models by aligning intermediate module outputs at the perception, prediction, and planning stages with text features explaining the driving reasoning process produced by VLMs. By encouraging alignment in latent space, VERDI enables the modular AD stack to internalize structured reasoning, without incurring the inference-time costs of large VLMs. We demonstrate the effectiveness of our method on the NuScenes dataset and find that VERDI outperforms existing e2e methods that do not embed reasoning by 10% in $\ell_{2}$ distance, while maintaining high inference speed.

Paperid: 4021, https://arxiv.org/pdf/2505.12884.pdf

Abstract:
Lightweight Vision-Language Models (VLMs) are indispensable for resource-constrained applications. The prevailing approach to aligning vision and language models involves freezing both the vision encoder and the language model while training small connector modules. However, this strategy heavily depends on the intrinsic capabilities of the language model, which can be suboptimal for lightweight models with limited representational capacity. In this work, we investigate this alignment bottleneck through the lens of mutual information, demonstrating that the constrained capacity of the language model inherently limits the Effective Mutual Information (EMI) between multimodal inputs and outputs, thereby compromising alignment quality. To address this challenge, we propose TinyAlign, a novel framework inspired by Retrieval-Augmented Generation, which strategically retrieves relevant context from a memory bank to enrich multimodal inputs and enhance their alignment. Extensive empirical evaluations reveal that TinyAlign significantly reduces training loss, accelerates convergence, and enhances task performance. Remarkably, it allows models to achieve baseline-level performance with only 40\% of the fine-tuning data, highlighting exceptional data efficiency. Our work thus offers a practical pathway for developing more capable lightweight VLMs while introducing a fresh theoretical lens to better understand and address alignment bottlenecks in constrained multimodal systems.

Paperid: 4022, https://arxiv.org/pdf/2505.12844.pdf

Abstract:
As the field progresses toward Artificial General Intelligence (AGI), there is a pressing need for more comprehensive and insightful evaluation frameworks that go beyond aggregate performance metrics. This paper introduces a unified rating system that jointly models the difficulty of individual test cases and the competency of AI models (or humans) across vision, language, and action domains. Unlike existing metrics that focus solely on models, our approach allows for fine-grained, difficulty-aware evaluations through competitive interactions between models and tasks, capturing both the long-tail distribution of real-world challenges and the competency gap between current models and full task mastery. We validate the generalizability and robustness of our system through extensive experiments on multiple established datasets and models across distinct AGI domains. The resulting rating distributions offer novel perspectives and interpretable insights into task difficulty, model progression, and the outstanding challenges that remain on the path to achieving full AGI task mastery.

Paperid: 4023, https://arxiv.org/pdf/2505.11741.pdf

Abstract:
Vision-language models (VLMs) now rival human performance on many multimodal tasks, yet they still hallucinate objects or generate unsafe text. Current hallucination detectors, e.g., single-token linear probing (SLP) and P(True), typically analyze only the logit of the first generated token or just its highest scoring component overlooking richer signals embedded within earlier token distributions. We demonstrate that analyzing the complete sequence of early logits potentially provides substantially more diagnostic information. We emphasize that hallucinations may only emerge after several tokens, as subtle inconsistencies accumulate over time. By analyzing the Kullback-Leibler (KL) divergence between logits corresponding to hallucinated and non-hallucinated tokens, we underscore the importance of incorporating later-token logits to more accurately capture the reliability dynamics of VLMs. In response, we introduce Multi-Token Reliability Estimation (MTRE), a lightweight, white-box method that aggregates logits from the first ten tokens using multi-token log-likelihood ratios and self-attention. Despite the challenges posed by large vocabulary sizes and long logit sequences, MTRE remains efficient and tractable. On MAD-Bench, MM-SafetyBench, MathVista, and four compositional-geometry benchmarks, MTRE improves AUROC by 9.4 +/- 1.3 points over SLP and by 12.1 +/- 1.7 points over P(True), setting a new state-of-the-art in hallucination detection for open-source VLMs.

Paperid: 4024, https://arxiv.org/pdf/2505.01743.pdf

Abstract:
The rapid advancements in Large Vision Language Models (LVLMs) offer the potential to surpass conventional labeling by generating richer, more detailed descriptions of on-device human behavior understanding (HBU) in low-resolution vision systems, such as depth, thermal, and infrared. However, existing large vision language model (LVLM) approaches are unable to understand low-resolution data well as they are primarily designed for high-resolution data, such as RGB images. A quick fixing approach is to caption a large amount of low-resolution data, but it requires a significant amount of labor-intensive annotation efforts. In this paper, we propose a novel, labor-saving system, Llambda, designed to support low-resolution HBU. The core idea is to leverage limited labeled data and a large amount of unlabeled data to guide LLMs in generating informative captions, which can be combined with raw data to effectively fine-tune LVLM models for understanding low-resolution videos in HBU. First, we propose a Contrastive-Oriented Data Labeler, which can capture behavior-relevant information from long, low-resolution videos and generate high-quality pseudo labels for unlabeled data via contrastive learning. Second, we propose a Physical-Knowledge Guided Captioner, which utilizes spatial and temporal consistency checks to mitigate errors in pseudo labels. Therefore, it can improve LLMs' understanding of sequential data and then generate high-quality video captions. Finally, to ensure on-device deployability, we employ LoRA-based efficient fine-tuning to adapt LVLMs for low-resolution data. We evaluate Llambda using a region-scale real-world testbed and three distinct low-resolution datasets, and the experiments show that Llambda outperforms several state-of-the-art LVLM systems up to $40.03\%$ on average Bert-Score.

Paperid: 4025, https://arxiv.org/pdf/2504.08974.pdf

Abstract:
Vision-language models (VLMs) have demonstrated impressive performance by effectively integrating visual and textual information to solve complex tasks. However, it is not clear how these models reason over the visual and textual data together, nor how the flow of information between modalities is structured. In this paper, we examine how VLMs reason by analyzing their biases when confronted with scenarios that present conflicting image and text cues, a common occurrence in real-world applications. To uncover the extent and nature of these biases, we build upon existing benchmarks to create five datasets containing mismatched image-text pairs, covering topics in mathematics, science, and visual descriptions. Our analysis shows that VLMs favor text in simpler queries but shift toward images as query complexity increases. This bias correlates with model scale, with the difference between the percentage of image- and text-preferred responses ranging from +56.8% (image favored) to -74.4% (text favored), depending on the task and model. In addition, we explore three mitigation strategies: simple prompt modifications, modifications that explicitly instruct models on how to handle conflicting information (akin to chain-of-thought prompting), and a task decomposition strategy that analyzes each modality separately before combining their results. Our findings indicate that the effectiveness of these strategies in identifying and mitigating bias varies significantly and is closely linked to the model's overall performance on the task and the specific modality in question.

Paperid: 4026, https://arxiv.org/pdf/2503.23315.pdf

Abstract:
We introduce the concept of "Design Agents" for engineering applications, particularly focusing on the automotive design process, while emphasizing that our approach can be readily extended to other engineering and design domains. Our framework integrates AI-driven design agents into the traditional engineering workflow, demonstrating how these specialized computational agents interact seamlessly with engineers and designers to augment creativity, enhance efficiency, and significantly accelerate the overall design cycle. By automating and streamlining tasks traditionally performed manually, such as conceptual sketching, styling enhancements, 3D shape retrieval and generative modeling, computational fluid dynamics (CFD) meshing, and aerodynamic simulations, our approach reduces certain aspects of the conventional workflow from weeks and days down to minutes. These agents leverage state-of-the-art vision-language models (VLMs), large language models (LLMs), and geometric deep learning techniques, providing rapid iteration and comprehensive design exploration capabilities. We ground our methodology in industry-standard benchmarks, encompassing a wide variety of conventional automotive designs, and utilize high-fidelity aerodynamic simulations to ensure practical and applicable outcomes. Furthermore, we present design agents that can swiftly and accurately predict simulation outcomes, empowering engineers and designers to engage in more informed design optimization and exploration. This research underscores the transformative potential of integrating advanced generative AI techniques into complex engineering tasks, paving the way for broader adoption and innovation across multiple engineering disciplines.

Paperid: 4027, https://arxiv.org/pdf/2503.17349.pdf

Abstract:
Vision-Language Models (VLMs) excel at identifying and describing objects but struggle with spatial reasoning such as accurately understanding the relative positions of objects. Inspired by the dual-pathway (ventral-dorsal) model of human vision, we investigate why VLMs fail spatial tasks despite strong object recognition capabilities. Our interpretability-driven analysis reveals a critical underlying cause: vision embeddings in VLMs are treated primarily as semantic ``bag-of-tokens," overshadowing subtle yet crucial positional cues due to their disproportionately large embedding norms. We validate this insight through extensive diagnostic experiments, demonstrating minimal performance impact when token orders or fine-grained spatial details are removed. Guided by these findings, we propose simple, interpretable interventions, including normalizing vision embedding norms and extracting mid-layer spatially rich features, to restore spatial awareness. Empirical results on both our synthetic data and standard benchmarks demonstrate improved spatial reasoning capabilities, highlighting the value of interpretability-informed design choices. Our study not only uncovers fundamental limitations in current VLM architectures but also provides actionable insights for enhancing structured perception of visual scenes.

Paperid: 4028, https://arxiv.org/pdf/2503.16096.pdf

Abstract:
The automated analysis of chemical literature holds promise to accelerate discovery in fields such as material science and drug development. In particular, search capabilities for chemical structures and Markush structures (chemical structure templates) within patent documents are valuable, e.g., for prior-art search. Advancements have been made in the automatic extraction of chemical structures from text and images, yet the Markush structures remain largely unexplored due to their complex multi-modal nature. In this work, we present MarkushGrapher, a multi-modal approach for recognizing Markush structures in documents. Our method jointly encodes text, image, and layout information through a Vision-Text-Layout encoder and an Optical Chemical Structure Recognition vision encoder. These representations are merged and used to auto-regressively generate a sequential graph representation of the Markush structure along with a table defining its variable groups. To overcome the lack of real-world training data, we propose a synthetic data generation pipeline that produces a wide range of realistic Markush structures. Additionally, we present M2S, the first annotated benchmark of real-world Markush structures, to advance research on this challenging task. Extensive experiments demonstrate that our approach outperforms state-of-the-art chemistry-specific and general-purpose vision-language models in most evaluation settings. Code, models, and datasets will be available.

Paperid: 4029, https://arxiv.org/pdf/2503.09445.pdf

Abstract:
Vision-Language Models (VLMs) based on Mixture-of-Experts (MoE) architectures have emerged as a pivotal paradigm in multimodal understanding, offering a powerful framework for integrating visual and linguistic information. However, the increasing complexity and diversity of tasks present significant challenges in coordinating load balancing across heterogeneous visual experts, where optimizing one specialist's performance often compromises others' capabilities. To address task heterogeneity and expert load imbalance, we propose Astrea, a novel multi-expert collaborative VLM architecture based on progressive pre-alignment. Astrea introduces three key innovations: 1) A heterogeneous expert coordination mechanism that integrates four specialized models (detection, segmentation, classification, captioning) into a comprehensive expert matrix covering essential visual comprehension elements; 2) A dynamic knowledge fusion strategy featuring progressive pre-alignment to harmonize experts within the VLM latent space through contrastive learning, complemented by probabilistically activated stochastic residual connections to preserve knowledge continuity; 3) An enhanced optimization framework utilizing momentum contrastive learning for long-range dependency modeling and adaptive weight allocators for real-time expert contribution calibration. Extensive evaluations across 12 benchmark tasks spanning VQA, image captioning, and cross-modal retrieval demonstrate Astrea's superiority over state-of-the-art models, achieving an average performance gain of +4.7\%. This study provides the first empirical demonstration that progressive pre-alignment strategies enable VLMs to overcome task heterogeneity limitations, establishing new methodological foundations for developing general-purpose multimodal agents.

Paperid: 4030, https://arxiv.org/pdf/2503.08368.pdf

Abstract:
Prompt tuning of Vision-Language Models (VLMs) such as CLIP, has demonstrated the ability to rapidly adapt to various downstream tasks. However, recent studies indicate that tuned VLMs may suffer from the problem of spurious correlations, where the model relies on spurious features (e.g. background and gender) in the data. This may lead to the model having worse robustness in out-of-distribution data. Standard methods for eliminating spurious correlation typically require us to know the spurious attribute labels of each sample, which is hard in the real world. In this work, we explore improving the group robustness of prompt tuning in VLMs without relying on manual annotation of spurious features. We notice the zero - shot image recognition ability of VLMs and use this ability to identify spurious features, thus avoiding the cost of manual annotation. By leveraging pseudo-spurious attribute annotations, we further propose a method to automatically adjust the training weights of different groups. Extensive experiments show that our approach efficiently improves the worst-group accuracy on CelebA, Waterbirds, and MetaShift datasets, achieving the best robustness gap between the worst-group accuracy and the overall accuracy.

Paperid: 4031, https://arxiv.org/pdf/2503.02199.pdf

Abstract:
Vision-Language Models (VLMs) excel in integrating visual and textual information for vision-centric tasks, but their handling of inconsistencies between modalities is underexplored. We investigate VLMs' modality preferences when faced with visual data and varied textual inputs in vision-centered settings. By introducing textual variations to four vision-centric tasks and evaluating ten Vision-Language Models (VLMs), we discover a \emph{``blind faith in text''} phenomenon: VLMs disproportionately trust textual data over visual data when inconsistencies arise, leading to significant performance drops under corrupted text and raising safety concerns. We analyze factors influencing this text bias, including instruction prompts, language model size, text relevance, token order, and the interplay between visual and textual certainty. While certain factors, such as scaling up the language model size, slightly mitigate text bias, others like token order can exacerbate it due to positional biases inherited from language models. To address this issue, we explore supervised fine-tuning with text augmentation and demonstrate its effectiveness in reducing text bias. Additionally, we provide a theoretical analysis suggesting that the blind faith in text phenomenon may stem from an imbalance of pure text and multi-modal data during training. Our findings highlight the need for balanced training and careful consideration of modality interactions in VLMs to enhance their robustness and reliability in handling multi-modal data inconsistencies.

Paperid: 4032, https://arxiv.org/pdf/2502.20572.pdf

Abstract:
Traffic safety remains a vital concern in contemporary urban settings, intensified by the increase of vehicles and the complicated nature of road networks. Traditional safety-critical event detection systems predominantly rely on sensor-based approaches and conventional machine learning algorithms, necessitating extensive data collection and complex training processes to adhere to traffic safety regulations. This paper introduces HazardNet, a small-scale Vision Language Model designed to enhance traffic safety by leveraging the reasoning capabilities of advanced language and vision models. We built HazardNet by fine-tuning the pre-trained Qwen2-VL-2B model, chosen for its superior performance among open-source alternatives and its compact size of two billion parameters. This helps to facilitate deployment on edge devices with efficient inference throughput. In addition, we present HazardQA, a novel Vision Question Answering (VQA) dataset constructed specifically for training HazardNet on real-world scenarios involving safety-critical events. Our experimental results show that the fine-tuned HazardNet outperformed the base model up to an 89% improvement in F1-Score and has comparable results with improvement in some cases reach up to 6% when compared to larger models, such as GPT-4o. These advancements underscore the potential of HazardNet in providing real-time, reliable traffic safety event detection, thereby contributing to reduced accidents and improved traffic management in urban environments. Both HazardNet model and the HazardQA dataset are available at https://huggingface.co/Tami3/HazardNet and https://huggingface.co/datasets/Tami3/HazardQA, respectively.

Paperid: 4033, https://arxiv.org/pdf/2502.19782.pdf

Abstract:
3D part segmentation is still an open problem in the field of 3D vision and AR/VR. Due to limited 3D labeled data, traditional supervised segmentation methods fall short in generalizing to unseen shapes and categories. Recently, the advancement in vision-language models' zero-shot abilities has brought a surge in open-world 3D segmentation methods. While these methods show promising results for 3D scenes or objects, they do not generalize well to 3D humans. In this paper, we present the first open-vocabulary segmentation method capable of handling 3D human. Our framework can segment the human category into desired fine-grained parts based on the textual prompt. We design a simple segmentation pipeline, leveraging SAM to generate multi-view proposals in 2D and proposing a novel HumanCLIP model to create unified embeddings for visual and textual inputs. Compared with existing pre-trained CLIP models, the HumanCLIP model yields more accurate embeddings for human-centric contents. We also design a simple-yet-effective MaskFusion module, which classifies and fuses multi-view features into 3D semantic masks without complex voting and grouping mechanisms. The design of decoupling mask proposals and text input also significantly boosts the efficiency of per-prompt inference. Experimental results on various 3D human datasets show that our method outperforms current state-of-the-art open-vocabulary 3D segmentation methods by a large margin. In addition, we show that our method can be directly applied to various 3D representations including meshes, point clouds, and 3D Gaussian Splatting.

Paperid: 4034, https://arxiv.org/pdf/2502.16707.pdf

Abstract:
Solving complex long-horizon robotic manipulation problems requires sophisticated high-level planning capabilities, the ability to reason about the physical world, and reactively choose appropriate motor skills. Vision-language models (VLMs) pretrained on Internet data could in principle offer a framework for tackling such problems. However, in their current form, VLMs lack both the nuanced understanding of intricate physics required for robotic manipulation and the ability to reason over long horizons to address error compounding issues. In this paper, we introduce a novel test-time computation framework that enhances VLMs' physical reasoning capabilities for multi-stage manipulation tasks. At its core, our approach iteratively improves a pretrained VLM with a "reflection" mechanism - it uses a generative model to imagine future world states, leverages these predictions to guide action selection, and critically reflects on potential suboptimalities to refine its reasoning. Experimental results demonstrate that our method significantly outperforms several state-of-the-art commercial VLMs as well as other post-training approaches such as Monte Carlo Tree Search (MCTS). Videos are available at https://reflect-vlm.github.io.

Paperid: 4035, https://arxiv.org/pdf/2502.15392.pdf

Abstract:
Recent multimodal foundation models are primarily trained on English or high resource European language data, which hinders their applicability to other medium and low-resource languages. To address this limitation, we introduce Chitrarth (Chitra: Image; Artha: Meaning), an inclusive Vision-Language Model (VLM), specifically targeting the rich linguistic diversity and visual reasoning across 10 prominent Indian languages. Our model effectively integrates a state-of-the-art (SOTA) multilingual Large Language Model (LLM) with a vision module, primarily trained on multilingual image-text data. Furthermore, we also introduce BharatBench, a comprehensive framework for evaluating VLMs across various Indian languages, ultimately contributing to more diverse and effective AI systems. Our model achieves SOTA results for benchmarks across low resource languages while retaining its efficiency in English. Through our research, we aim to set new benchmarks in multilingual-multimodal capabilities, offering substantial improvements over existing models and establishing a foundation to facilitate future advancements in this arena.

Paperid: 4036, https://arxiv.org/pdf/2502.14994.pdf

Abstract:
The impressive achievements of generative models in creating high-quality videos have raised concerns about digital integrity and privacy vulnerabilities. Recent works of AI-generated content detection have been widely studied in the image field (e.g., deepfake), yet the video field has been unexplored. Large Vision Language Model (LVLM) has become an emerging tool for AI-generated content detection for its strong reasoning and multimodal capabilities. It breaks the limitations of traditional deep learning based methods faced with like lack of transparency and inability to recognize new artifacts. Motivated by this, we propose LAVID, a novel LVLMs-based ai-generated video detection with explicit knowledge enhancement. Our insight list as follows: (1) The leading LVLMs can call external tools to extract useful information to facilitate its own video detection task; (2) Structuring the prompt can affect LVLM's reasoning ability to interpret information in video content. Our proposed pipeline automatically selects a set of explicit knowledge tools for detection, and then adaptively adjusts the structure prompt by self-rewriting. Different from prior SOTA that trains additional detectors, our method is fully training-free and only requires inference of the LVLM for detection. To facilitate our research, we also create a new benchmark \vidfor with high-quality videos generated from multiple sources of video generation tools. Evaluation results show that LAVID improves F1 scores by 6.2 to 30.2% over the top baselines on our datasets across four SOTA LVLMs.

Paperid: 4037, https://arxiv.org/pdf/2502.01201.pdf

Abstract:
Traditional Anomaly Detection (AD) methods have predominantly relied on unsupervised learning from extensive normal data. Recent AD methods have evolved with the advent of large pre-trained vision-language models, enhancing few-shot anomaly detection capabilities. However, these latest AD methods still exhibit limitations in accuracy improvement. One contributing factor is their direct comparison of a query image's features with those of few-shot normal images. This direct comparison often leads to a loss of precision and complicates the extension of these techniques to more complex domains--an area that remains underexplored in a more refined and comprehensive manner. To address these limitations, we introduce the anomaly personalization method, which performs a personalized one-to-normal transformation of query images using an anomaly-free customized generation model, ensuring close alignment with the normal manifold. Moreover, to further enhance the stability and robustness of prediction results, we propose a triplet contrastive anomaly inference strategy, which incorporates a comprehensive comparison between the query and generated anomaly-free data pool and prompt information. Extensive evaluations across eleven datasets in three domains demonstrate our model's effectiveness compared to the latest AD methods. Additionally, our method has been proven to transfer flexibly to other AD methods, with the generated image data effectively improving the performance of other AD methods.

Paperid: 4038, https://arxiv.org/pdf/2501.16720.pdf

Abstract:
Recent advancements in fine-tuning Vision-Language Foundation Models (VLMs) have garnered significant attention for their effectiveness in downstream few-shot learning tasks.While these recent approaches exhibits some performance improvements, they often suffer from excessive training parameters and high computational costs. To address these challenges, we propose a novel Block matrix-based low-rank adaptation framework, called Block-LoRA, for fine-tuning VLMs on downstream few-shot tasks. Inspired by recent work on Low-Rank Adaptation (LoRA), Block-LoRA partitions the original low-rank decomposition matrix of LoRA into a series of sub-matrices while sharing all down-projection sub-matrices. This structure not only reduces the number of training parameters, but also transforms certain complex matrix multiplication operations into simpler matrix addition, significantly lowering the computational cost of fine-tuning. Notably, Block-LoRA enables fine-tuning CLIP on the ImageNet few-shot benchmark using a single 24GB GPU. We also show that Block-LoRA has the more tighter bound of generalization error than vanilla LoRA. Without bells and whistles, extensive experiments demonstrate that Block-LoRA achieves competitive performance compared to state-of-the-art CLIP-based few-shot methods, while maintaining a low training parameters count and reduced computational overhead.

Paperid: 4039, https://arxiv.org/pdf/2501.16065.pdf

Abstract:
The Visual Language Model, known for its robust cross-modal capabilities, has been extensively applied in various computer vision tasks. In this paper, we explore the use of CLIP (Contrastive Language-Image Pretraining), a vision-language model pretrained on large-scale image-text pairs to align visual and textual features, for acquiring fine-grained and domain-invariant representations in generalizable person re-identification. The adaptation of CLIP to the task presents two primary challenges: learning more fine-grained features to enhance discriminative ability, and learning more domain-invariant features to improve the model's generalization capabilities. To mitigate the first challenge thereby enhance the ability to learn fine-grained features, a three-stage strategy is proposed to boost the accuracy of text descriptions. Initially, the image encoder is trained to effectively adapt to person re-identification tasks. In the second stage, the features extracted by the image encoder are used to generate textual descriptions (i.e., prompts) for each image. Finally, the text encoder with the learned prompts is employed to guide the training of the final image encoder. To enhance the model's generalization capabilities to unseen domains, a bidirectional guiding method is introduced to learn domain-invariant image features. Specifically, domain-invariant and domain-relevant prompts are generated, and both positive (pulling together image features and domain-invariant prompts) and negative (pushing apart image features and domain-relevant prompts) views are used to train the image encoder. Collectively, these strategies contribute to the development of an innovative CLIP-based framework for learning fine-grained generalized features in person re-identification.

Paperid: 4040, https://arxiv.org/pdf/2501.13964.pdf

Abstract:
Augmented Reality (AR) enhances the real world by integrating virtual content, yet ensuring the quality, usability, and safety of AR experiences presents significant challenges. Could Vision-Language Models (VLMs) offer a solution for the automated evaluation of AR-generated scenes? Could Vision-Language Models (VLMs) offer a solution for the automated evaluation of AR-generated scenes? In this study, we evaluate the capabilities of three state-of-the-art commercial VLMs -- GPT, Gemini, and Claude -- in identifying and describing AR scenes. For this purpose, we use DiverseAR, the first AR dataset specifically designed to assess VLMs' ability to analyze virtual content across a wide range of AR scene complexities. Our findings demonstrate that VLMs are generally capable of perceiving and describing AR scenes, achieving a True Positive Rate (TPR) of up to 93% for perception and 71% for description. While they excel at identifying obvious virtual objects, such as a glowing apple, they struggle when faced with seamlessly integrated content, such as a virtual pot with realistic shadows. Our results highlight both the strengths and the limitations of VLMs in understanding AR scenarios. We identify key factors affecting VLM performance, including virtual content placement, rendering quality, and physical plausibility. This study underscores the potential of VLMs as tools for evaluating the quality of AR experiences.

Paperid: 4041, https://arxiv.org/pdf/2501.12553.pdf

Abstract:
In Augmented Reality (AR), virtual content enhances user experience by providing additional information. However, improperly positioned or designed virtual content can be detrimental to task performance, as it can impair users' ability to accurately interpret real-world information. In this paper we examine two types of task-detrimental virtual content: obstruction attacks, in which virtual content prevents users from seeing real-world objects, and information manipulation attacks, in which virtual content interferes with users' ability to accurately interpret real-world information. We provide a mathematical framework to characterize these attacks and create a custom open-source dataset for attack evaluation. To address these attacks, we introduce ViDDAR (Vision language model-based Task-Detrimental content Detector for Augmented Reality), a comprehensive full-reference system that leverages Vision Language Models (VLMs) and advanced deep learning techniques to monitor and evaluate virtual content in AR environments, employing a user-edge-cloud architecture to balance performance with low latency. To the best of our knowledge, ViDDAR is the first system to employ VLMs for detecting task-detrimental content in AR settings. Our evaluation results demonstrate that ViDDAR effectively understands complex scenes and detects task-detrimental content, achieving up to 92.15% obstruction detection accuracy with a detection latency of 533 ms, and an 82.46% information manipulation content detection accuracy with a latency of 9.62 s.

Paperid: 4042, https://arxiv.org/pdf/2501.10071.pdf

Abstract:
In recent years, No-Reference Point Cloud Quality Assessment (NR-PCQA) research has achieved significant progress. However, existing methods mostly seek a direct mapping function from visual data to the Mean Opinion Score (MOS), which is contradictory to the mechanism of practical subjective evaluation. To address this, we propose a novel language-driven PCQA method named CLIP-PCQA. Considering that human beings prefer to describe visual quality using discrete quality descriptions (e.g., "excellent" and "poor") rather than specific scores, we adopt a retrieval-based mapping strategy to simulate the process of subjective assessment. More specifically, based on the philosophy of CLIP, we calculate the cosine similarity between the visual features and multiple textual features corresponding to different quality descriptions, in which process an effective contrastive loss and learnable prompts are introduced to enhance the feature extraction. Meanwhile, given the personal limitations and bias in subjective experiments, we further covert the feature similarities into probabilities and consider the Opinion Score Distribution (OSD) rather than a single MOS as the final target. Experimental results show that our CLIP-PCQA outperforms other State-Of-The-Art (SOTA) approaches.

Paperid: 4043, https://arxiv.org/pdf/2501.05566.pdf

Abstract:
Scene understanding is essential for enhancing driver safety, generating human-centric explanations for Automated Vehicle (AV) decisions, and leveraging Artificial Intelligence (AI) for retrospective driving video analysis. This study developed a dynamic scene retrieval system using Contrastive Language-Image Pretraining (CLIP) models, which can be optimized for real-time deployment on edge devices. The proposed system outperforms state-of-the-art in-context learning methods, including the zero-shot capabilities of GPT-4o, particularly in complex scenarios. By conducting frame-level analysis on the Honda Scenes Dataset, which contains a collection of about 80 hours of annotated driving videos capturing diverse real-world road and weather conditions, our study highlights the robustness of CLIP models in learning visual concepts from natural language supervision. Results also showed that fine-tuning the CLIP models, such as ViT-L/14 and ViT-B/32, significantly improved scene classification, achieving a top F1 score of 91.1%. These results demonstrate the ability of the system to deliver rapid and precise scene recognition, which can be used to meet the critical requirements of Advanced Driver Assistance Systems (ADAS). This study shows the potential of CLIP models to provide scalable and efficient frameworks for dynamic scene understanding and classification. Furthermore, this work lays the groundwork for advanced autonomous vehicle technologies by fostering a deeper understanding of driver behavior, road conditions, and safety-critical scenarios, marking a significant step toward smarter, safer, and more context-aware autonomous driving systems.

Paperid: 4044, https://arxiv.org/pdf/2501.04641.pdf

Abstract:
Multi-modal generative AI systems, such as those combining vision and language, rely on contrastive pre-training to learn representations across different modalities. While their practical benefits are widely acknowledged, a rigorous theoretical understanding of the contrastive pre-training framework remains limited. This paper develops a theoretical framework to explain the success of contrastive pre-training in downstream tasks, such as zero-shot classification, conditional diffusion models, and vision-language models. We introduce the concept of approximate sufficient statistics, a generalization of the classical sufficient statistics, and show that near-minimizers of the contrastive pre-training loss are approximately sufficient, making them adaptable to diverse downstream tasks. We further propose the Joint Generative Hierarchical Model for the joint distribution of images and text, showing that transformers can efficiently approximate relevant functions within this model via belief propagation. Building on this framework, we derive sample complexity guarantees for multi-modal learning based on contrastive pre-trained representations. Numerical simulations validate these theoretical findings, demonstrating the strong generalization performance of contrastively pre-trained transformers in various multi-modal tasks.

Paperid: 4045, https://arxiv.org/pdf/2501.02385.pdf

Abstract:
While mainstream vision-language models (VLMs) have advanced rapidly in understanding image level information, they still lack the ability to focus on specific areas designated by humans. Rather, they typically rely on large volumes of high-quality image-text paired data to learn and generate posterior attention maps. To address this critical issue, we propose leveraging visual prompts:simple visual markers in various forms to guide and enhance the formation of region-specific attention. Thus, we introduce MedVP, a pioneering framework that integrates medical entity extraction, visual prompt generation, and dataset adaptation for visual prompt guided fine-tuning. We successfully outperform recent state-of-the-art large models across multiple medical VQA datasets. Extensive experiments and Human evaluation are conducted to analyze the impact of different visual prompt forms and how they contribute to performance improvement. The results demonstrate both the effectiveness and clinical significance of our approach.

Paperid: 4046, https://arxiv.org/pdf/2412.20903.pdf

Abstract:
Approximately 200 million individuals around the world suffer from varying degrees of visual impairment, making it crucial to leverage AI technology to offer walking assistance for these people. With the recent progress of vision-language models (VLMs), applying VLMs to offer walking guidance has become popular. However, the existing methods of walking guidance are mainly based on self-curated question-answering datasets that are not publicly accessible, without a standardized benchmark for training or evaluation. Moreover, walking assistance often requires real-time streaming video analysis and the generation of concise yet informative reminders, making VLMs struggle due to excessive responses and low efficiency in inferences. In this paper, we introduce the first large-scale dataset dedicated to walking assistance, comprising 12,000 video-annotation pairs, to provide a unified benchmark for training and evaluating systems to help visually-impaired individuals walk. Furthermore, a WalkVLM model is proposed, which employs chain of thought for hierarchical planning to generate concise but informative reminders and utilizes temporal-aware adaptive prediction to reduce the temporal redundancy of reminders. Finally, we have established a solid benchmark for blind walking task and verified the advantages of WalkVLM in stream video processing for this task compared to other VLMs. Our dataset and code are available at https://walkvlm2024.github.io.

Paperid: 4047, https://arxiv.org/pdf/2412.11475.pdf

Abstract:
We present OmniVLM, a sub-billion-parameter vision-language model for efficient on-device inference. OmniVLM introduces a token compression mechanism that reduces visual token sequence length from 729 to 81 tokens, significantly reducing computational overhead while preserving visual-semantic fidelity. Through a multi-stage training pipeline of pretraining, supervised fine-tuning, and minimal-edit Direct Preference Optimization (DPO), OmniVLM matches the performance of larger models. On multiple benchmarks including ScienceQA, POPE, and MMMU, OmniVLM outperforms existing baselines like nanoLLAVA within a 968M-parameter footprint. Empirical results on the same laptop demonstrate 9.1x faster time-to-first-token (0.75s vs 6.82s) and 1.5x higher decoding speed (29.41 vs 19.20 tokens/s) compared to nanoLLAVA, enabling efficient deployment on edge devices. The model weights can be accessed on huggingface: \url{https://huggingface.co/NexaAIDev/OmniVLM-968M}, and the inference examples can be find in Appendix B.

Paperid: 4048, https://arxiv.org/pdf/2412.08259.pdf

Abstract:
As a common form of communication in social media,stickers win users' love in the internet scenarios, for their ability to convey emotions in a vivid, cute, and interesting way. People prefer to get an appropriate sticker through retrieval rather than creation for the reason that creating a sticker is time-consuming and relies on rule-based creative tools with limited capabilities. Nowadays, advanced text-to-video algorithms have spawned numerous general video generation systems that allow users to customize high-quality, photo-realistic videos by only providing simple text prompts. However, creating customized animated stickers, which have lower frame rates and more abstract semantics than videos, is greatly hindered by difficulties in data acquisition and incomplete benchmarks. To facilitate the exploration of researchers in animated sticker generation (ASG) field, we firstly construct the currently largest vision-language sticker dataset named VSD2M at a two-million scale that contains static and animated stickers. Secondly, to improve the performance of traditional video generation methods on ASG tasks with discrete characteristics, we propose a Spatial Temporal Interaction (STI) layer that utilizes semantic interaction and detail preservation to address the issue of insufficient information utilization. Moreover, we train baselines with several video generation methods (e.g., transformer-based, diffusion-based methods) on VSD2M and conduct a detailed analysis to establish systemic supervision on ASG task. To the best of our knowledge, this is the most comprehensive large-scale benchmark for multi-frame animated sticker generation, and we hope this work can provide valuable inspiration for other scholars in intelligent creation.

Paperid: 4049, https://arxiv.org/pdf/2412.02886.pdf

Abstract:
For decades, corporations and governments have relied on scanned documents to record vast amounts of information. However, extracting this information is a slow and tedious process due to the sheer volume and complexity of these records. The rise of Vision Language Models (VLMs) presents a way to efficiently and accurately extract the information out of these documents. The current automated workflow often requires a two-step approach involving the extraction of information using optical character recognition software and subsequent usage of large language models for processing this information. Unfortunately, these methods encounter significant challenges when dealing with noisy scanned documents, often requiring computationally expensive language models to handle high information density effectively. In this study, we propose PatchFinder, an algorithm that builds upon VLMs to improve information extraction. First, we devise a confidence-based score, called Patch Confidence, based on the Maximum Softmax Probability of the VLMs' output to measure the model's confidence in its predictions. Using this metric, PatchFinder determines a suitable patch size, partitions the input document into overlapping patches, and generates confidence-based predictions for the target information. Our experimental results show that PatchFinder, leveraging Phi-3v, a 4.2-billion-parameter VLM, achieves an accuracy of 94% on our dataset of 190 noisy scanned documents, outperforming ChatGPT-4o by 18.5 percentage points.

Paperid: 4050, https://arxiv.org/pdf/2411.17454.pdf

Abstract:
Given a query from one modality, few-shot cross-modal retrieval (CMR) retrieves semantically similar instances in another modality with the target domain including classes that are disjoint from the source domain. Compared with classical few-shot CMR methods, vision-language pretraining methods like CLIP have shown great few-shot or zero-shot learning performance. However, they still suffer challenges due to (1) the feature degradation encountered in the target domain and (2) the extreme data imbalance. To tackle these issues, we propose FLEX-CLIP, a novel Feature-level Generation Network Enhanced CLIP. FLEX-CLIP includes two training stages. In multimodal feature generation, we propose a composite multimodal VAE-GAN network to capture real feature distribution patterns and generate pseudo samples based on CLIP features, addressing data imbalance. For common space projection, we develop a gate residual network to fuse CLIP features with projected features, reducing feature degradation in X-shot scenarios. Experimental results on four benchmark datasets show a 7%-15% improvement over state-of-the-art methods, with ablation studies demonstrating enhancement of CLIP features.

Paperid: 4051, https://arxiv.org/pdf/2411.17406.pdf

Abstract:
Recent advances in vision-language models (VLMs) have demonstrated remarkable capabilities in image classification by leveraging predefined sets of labels to construct text prompts for zero-shot reasoning. However, these approaches face significant limitations in undefined domains, where the label space is vocabulary-unknown and composite. We thus introduce Generative Semantic Labels (GSLs), a novel task that aims to predict a comprehensive set of semantic labels for an image without being constrained by a predefined labels set. Unlike traditional zero-shot classification, GSLs generates multiple semantic-level labels, encompassing objects, scenes, attributes, and relationships, thereby providing a richer and more accurate representation of image content. In this paper, we propose Chain-of-Action (CoA), an innovative method designed to tackle the GSLs task. CoA is motivated by the observation that enriched contextual information significantly improves generative performance during inference. Specifically, CoA decomposes the GSLs task into a sequence of detailed actions. Each action extracts and merges key information from the previous step, passing enriched context to the next, ultimately guiding the VLM to generate comprehensive and accurate semantic labels. We evaluate the effectiveness of CoA through extensive experiments on widely-used benchmark datasets. The results demonstrate significant improvements across key performance metrics, validating the capability of CoA to generate accurate and contextually rich semantic labels. Our work not only advances the state-of-the-art in generative semantic labels but also opens new avenues for applying VLMs in open-ended and dynamic real-world scenarios.

Paperid: 4052, https://arxiv.org/pdf/2411.17041.pdf

Abstract:
Diffusion models have achieved impressive results in generative tasks like text-to-image (T2I) and text-to-video (T2V) synthesis. However, achieving accurate text alignment in T2V generation remains challenging due to the complex temporal dependency across frames. Existing reinforcement learning (RL)-based approaches to enhance text alignment often require differentiable reward functions or are constrained to limited prompts, hindering their scalability and applicability. In this paper, we propose Free$^2$Guide, a novel gradient-free framework for aligning generated videos with text prompts without requiring additional model training. Leveraging principles from path integral control, Free$^2$Guide approximates guidance for diffusion models using non-differentiable reward functions, thereby enabling the integration of powerful black-box Large Vision-Language Models (LVLMs) as reward model. Additionally, our framework supports the flexible ensembling of multiple reward models, including large-scale image-based models, to synergistically enhance alignment without incurring substantial computational overhead. We demonstrate that Free$^2$Guide significantly improves text alignment across various dimensions and enhances the overall quality of generated videos.

Paperid: 4053, https://arxiv.org/pdf/2411.00394.pdf

Abstract:
In question-answering scenarios, humans can assess whether the available information is sufficient and seek additional information if necessary, rather than providing a forced answer. In contrast, Vision Language Models (VLMs) typically generate direct, one-shot responses without evaluating the sufficiency of the information. To investigate this gap, we identify a critical and challenging task in the Visual Question Answering (VQA) scenario: can VLMs indicate how to adjust an image when the visual information is insufficient to answer a question? This capability is especially valuable for assisting visually impaired individuals who often need guidance to capture images correctly. To evaluate this capability of current VLMs, we introduce a human-labeled dataset as a benchmark for this task. Additionally, we present an automated framework that generates synthetic training data by simulating ``where to know'' scenarios. Our empirical results show significant performance improvements in mainstream VLMs when fine-tuned with this synthetic data. This study demonstrates the potential to narrow the gap between information assessment and acquisition in VLMs, bringing their performance closer to humans.

Paperid: 4054, https://arxiv.org/pdf/2410.11255.pdf

Abstract:
Recent advancements in pre-trained vision-language models like CLIP have shown promise in person re-identification (ReID) applications. However, their performance in generalizable person re-identification tasks remains suboptimal. The large-scale and diverse image-text pairs used in CLIP's pre-training may lead to a lack or insufficiency of certain fine-grained features. In light of these challenges, we propose a hard sample mining method called DFGS (Depth-First Graph Sampler), based on depth-first search, designed to offer sufficiently challenging samples to enhance CLIP's ability to extract fine-grained features. DFGS can be applied to both the image encoder and the text encoder in CLIP. By leveraging the powerful cross-modal learning capabilities of CLIP, we aim to apply our DFGS method to extract challenging samples and form mini-batches with high discriminative difficulty, providing the image model with more efficient and challenging samples that are difficult to distinguish, thereby enhancing the model's ability to differentiate between individuals. Our results demonstrate significant improvements over other methods, confirming the effectiveness of DFGS in providing challenging samples that enhance CLIP's performance in generalizable person re-identification.

Paperid: 4055, https://arxiv.org/pdf/2410.09365.pdf

Abstract:
Pre-trained vision-language models (VLMs), such as CLIP, have exhibited remarkable performance across various downstream tasks by aligning text and images in a unified embedding space. However, due to the imbalanced distribution of pre-trained datasets, CLIP suffers from the bias problem in real-world applications. Existing debiasing methods struggle to obtain sufficient image samples for minority groups and incur high costs for group labeling. To address the limitations, we propose a Text-Only Debiasing framework called TOD, leveraging a text-as-image training paradigm to mitigate visual biases. Specifically, this approach repurposes the text encoder to function as an image encoder, thereby eliminating the need for image data. Simultaneously, it utilizes a large language model (LLM) to generate a balanced text dataset, which is then used for prompt tuning. However, we observed that the model overfits to the text modality because label names, serving as supervision signals, appear explicitly in the texts. To address this issue, we further introduce a Multi-Target Prediction (MTP) task that motivates the model to focus on complex contexts and distinguish between target and biased information. Extensive experiments on the Waterbirds and CelebA datasets show that our method significantly improves group robustness, achieving state-of-the-art results among image-free methods and even competitive performance compared to image-supervised methods. Furthermore, the proposed method can be adapted to challenging scenarios with multiple or unknown bias attributes, demonstrating its strong generalization and robustness.

Paperid: 4056, https://arxiv.org/pdf/2410.06795.pdf

Abstract:
Hallucinations in large vision-language models (LVLMs) are a significant challenge, i.e., generating objects that are not presented in the visual input, which impairs their reliability. Recent studies often attribute hallucinations to a lack of understanding of visual input, yet ignore a more fundamental issue: the model's inability to effectively extract or decouple visual features. In this paper, we revisit the hallucinations in LVLMs from an architectural perspective, investigating whether the primary cause lies in the visual encoder (feature extraction) or the modal alignment module (feature decoupling). Motivated by our findings on the preliminary investigation, we propose a novel tuning strategy, PATCH, to mitigate hallucinations in LVLMs. This plug-and-play method can be integrated into various LVLMs, utilizing adaptive virtual tokens to extract object features from bounding boxes, thereby addressing hallucinations caused by insufficient decoupling of visual features. PATCH achieves state-of-the-art performance on multiple multi-modal hallucination datasets. We hope this approach provides researchers with deeper insights into the underlying causes of hallucinations in LVLMs, fostering further advancements and innovation in this field.

Paperid: 4057, https://arxiv.org/pdf/2410.06237.pdf

Abstract:
To operate at a building scale, service robots must perform very long-horizon mobile manipulation tasks by navigating to different rooms, accessing different floors, and interacting with a wide and unseen range of everyday objects. We refer to these tasks as Building-wide Mobile Manipulation. To tackle these inherently long-horizon tasks, we introduce BUMBLE, a unified Vision-Language Model (VLM)-based framework integrating open-world RGBD perception, a wide spectrum of gross-to-fine motor skills, and dual-layered memory. Our extensive evaluation (90+ hours) indicates that BUMBLE outperforms multiple baselines in long-horizon building-wide tasks that require sequencing up to 12 ground truth skills spanning 15 minutes per trial. BUMBLE achieves 47.1% success rate averaged over 70 trials in different buildings, tasks, and scene layouts from different starting rooms and floors. Our user study demonstrates 22% higher satisfaction with our method than state-of-the-art mobile manipulation methods. Finally, we demonstrate the potential of using increasingly-capable foundation models to push performance further. For more information, see https://robin-lab.cs.utexas.edu/BUMBLE/

Paperid: 4058, https://arxiv.org/pdf/2409.05395.pdf

Abstract:
This study explores replacing Transformers in Visual Language Models (VLMs) with Mamba, a recent structured state space model (SSM) that demonstrates promising performance in sequence modeling. We test models up to 3B parameters under controlled conditions, showing that Mamba-based VLMs outperforms Transformers-based VLMs in captioning, question answering, and reading comprehension. However, we find that Transformers achieve greater performance in visual grounding and the performance gap widens with scale. We explore two hypotheses to explain this phenomenon: 1) the effect of task-agnostic visual encoding on the updates of the hidden states, and 2) the difficulty in performing visual grounding from the perspective of in-context multimodal retrieval. Our results indicate that a task-aware encoding yields minimal performance gains on grounding, however, Transformers significantly outperform Mamba at in-context multimodal retrieval. Overall, Mamba shows promising performance on tasks where the correct output relies on a summary of the image but struggles when retrieval of explicit information from the context is required.

Paperid: 4059, https://arxiv.org/pdf/2601.21712.pdf

Abstract:
Vision Language Action (VLA) models enable instruction following manipulation, yet dualarm deployment remains unsafe due to under modeled selfcollisions between arms and grasped objects. We introduce CoFreeVLA, which augments an endtoend VLA with a short horizon selfcollision risk estimator that predicts collision likelihood from proprioception, visual embeddings, and planned actions. The estimator gates risky commands, recovers to safe states via risk-guided adjustments, and shapes policy refinement for safer rollouts. It is pre-trained with model-based collision labels and posttrained on real robot rollouts for calibration. On five bimanual tasks with the PiPER robot arm, CoFreeVLA reduces selfcollisions and improves success rates versus RDT and APEX.

Paperid: 4060, https://arxiv.org/pdf/2601.21037.pdf

Abstract:
Vision-Language Models have excelled at textual reasoning, but they often struggle with fine-grained spatial understanding and continuous action planning, failing to simulate the dynamics required for complex visual reasoning. In this work, we formulate visual reasoning by means of video generation models, positing that generated frames can act as intermediate reasoning steps between initial states and solutions. We evaluate their capacity in two distinct regimes: Maze Navigation for sequential discrete planning with low visual change and Tangram Puzzle for continuous manipulation with high visual change. Our experiments reveal three critical insights: (1) Robust Zero-Shot Generalization: In both tasks, the model demonstrates strong performance on unseen data distributions without specific finetuning. (2) Visual Context: The model effectively uses visual context as explicit control, such as agent icons and tangram shapes, enabling it to maintain high visual consistency and adapt its planning capability robustly to unseen patterns. (3) Visual Test-Time Scaling: We observe a test-time scaling law in sequential planning; increasing the generated video length (visual inference budget) empowers better zero-shot generalization to spatially and temporally complex paths. These findings suggest that video generation is not merely a media tool, but a scalable, generalizable paradigm for visual reasoning.

Paperid: 4061, https://arxiv.org/pdf/2601.20419.pdf

Abstract:
Recent research has shown that aligning fine-grained text descriptions with localized image patches can significantly improve the zero-shot performance of pre-trained vision-language models (e.g., CLIP). However, we find that both fine-grained text descriptions and localized image patches often contain redundant information, making text-visual alignment less effective. In this paper, we tackle this issue from two perspectives: \emph{View Refinement} and \emph{Description refinement}, termed as \textit{\textbf{Bi}-refinement for \textbf{F}ine-grained \textbf{T}ext-visual \textbf{A}lignment} (BiFTA). \emph{View refinement} removes redundant image patches with high \emph{Intersection over Union} (IoU) ratios, resulting in more distinctive visual samples. \emph{Description refinement} removes redundant text descriptions with high pairwise cosine similarity, ensuring greater diversity in the remaining descriptions. BiFTA achieves superior zero-shot performance on 6 benchmark datasets for both ViT-based and ResNet-based CLIP, justifying the necessity to remove redundant information in visual-text alignment.

Paperid: 4062, https://arxiv.org/pdf/2601.03713.pdf

Abstract:
Vision-language models (VLMs) have recently shown remarkable performance in navigation and localization tasks by leveraging large-scale pretraining for semantic understanding. However, applying VLMs to 6-DoF endoscopic camera localization presents several challenges: 1) the lack of large-scale, high-quality, densely annotated, and localization-oriented vision-language datasets in real-world medical settings; 2) limited capability for fine-grained pose regression; and 3) high computational latency when extracting temporal features from past frames. To address these issues, we first construct BREATH dataset, the largest in-vivo endoscopic localization dataset to date, collected in the complex human airway. Building on this dataset, we propose BREATH-VL, a hybrid framework that integrates semantic cues from VLMs with geometric information from vision-based registration methods for accurate 6-DoF pose estimation. Our motivation lies in the complementary strengths of both approaches: VLMs offer generalizable semantic understanding, while registration methods provide precise geometric alignment. To further enhance the VLM's ability to capture temporal context, we introduce a lightweight context-learning mechanism that encodes motion history as linguistic prompts, enabling efficient temporal reasoning without expensive video-level computation. Extensive experiments demonstrate that the vision-language module delivers robust semantic localization in challenging surgical scenes. Building on this, our BREATH-VL outperforms state-of-the-art vision-only localization methods in both accuracy and generalization, reducing translational error by 25.5% compared with the best-performing baseline, while achieving competitive computational latency.

Paperid: 4063, https://arxiv.org/pdf/2512.21714.pdf

Abstract:
Embodied navigation in open, dynamic environments demands accurate foresight of how the world will evolve and how actions will unfold over time. We propose AstraNav-World, an end-to-end world model that jointly reasons about future visual states and action sequences within a unified probabilistic framework. Our framework integrates a diffusion-based video generator with a vision-language policy, enabling synchronized rollouts where predicted scenes and planned actions are updated simultaneously. Training optimizes two complementary objectives: generating action-conditioned multi-step visual predictions and deriving trajectories conditioned on those predicted visuals. This bidirectional constraint makes visual predictions executable and keeps decisions grounded in physically consistent, task-relevant futures, mitigating cumulative errors common in decoupled "envision-then-plan" pipelines. Experiments across diverse embodied navigation benchmarks show improved trajectory accuracy and higher success rates. Ablations confirm the necessity of tight vision-action coupling and unified training, with either branch removal degrading both prediction quality and policy reliability. In real-world testing, AstraNav-World demonstrated exceptional zero-shot capabilities, adapting to previously unseen scenarios without any real-world fine-tuning. These results suggest that AstraNav-World captures transferable spatial understanding and planning-relevant navigation dynamics, rather than merely overfitting to simulation-specific data distribution. Overall, by unifying foresight vision and control within a single generative model, we move closer to reliable, interpretable, and general-purpose embodied agents that operate robustly in open-ended real-world settings.

Paperid: 4064, https://arxiv.org/pdf/2512.17621.pdf

Abstract:
While Vision-Language Models (VLMs) have achieved notable progress in computational pathology (CPath), the gigapixel scale and spatial heterogeneity of Whole Slide Images (WSIs) continue to pose challenges for multimodal understanding. Existing alignment methods struggle to capture fine-grained correspondences between textual descriptions and visual cues across thousands of patches from a slide, compromising their performance on downstream tasks. In this paper, we propose PathFLIP (Pathology Fine-grained Language-Image Pretraining), a novel framework for holistic WSI interpretation. PathFLIP decomposes slide-level captions into region-level subcaptions and generates text-conditioned region embeddings to facilitate precise visual-language grounding. By harnessing Large Language Models (LLMs), PathFLIP can seamlessly follow diverse clinical instructions and adapt to varied diagnostic contexts. Furthermore, it exhibits versatile capabilities across multiple paradigms, efficiently handling slide-level classification and retrieval, fine-grained lesion localization, and instruction following. Extensive experiments demonstrate that PathFLIP outperforms existing large-scale pathological VLMs on four representative benchmarks while requiring significantly less training data, paving the way for fine-grained, instruction-aware WSI interpretation in clinical practice.

Paperid: 4065, https://arxiv.org/pdf/2512.16811.pdf

Abstract:
Vision-Language-Action (VLA) models achieve strong generalization in robotic manipulation but remain largely reactive and 2D-centric, making them unreliable in tasks that require precise 3D reasoning. We propose GeoPredict, a geometry-aware VLA framework that augments a continuous-action policy with predictive kinematic and geometric priors. GeoPredict introduces a trajectory-level module that encodes motion history and predicts multi-step 3D keypoint trajectories of robot arms, and a predictive 3D Gaussian geometry module that forecasts workspace geometry with track-guided refinement along future keypoint trajectories. These predictive modules serve exclusively as training-time supervision through depth-based rendering, while inference requires only lightweight additional query tokens without invoking any 3D decoding. Experiments on RoboCasa Human-50, LIBERO, and real-world manipulation tasks show that GeoPredict consistently outperforms strong VLA baselines, especially in geometry-intensive and spatially demanding scenarios.

Paperid: 4066, https://arxiv.org/pdf/2512.14666.pdf

Abstract:
Achieving truly adaptive embodied intelligence requires agents that learn not just by imitating static demonstrations, but by continuously improving through environmental interaction, which is akin to how humans master skills through practice. Vision-Language-Action (VLA) models have advanced robotic manipulation by leveraging large language models, yet remain fundamentally limited by Supervised Finetuning (SFT): requiring hundreds of demonstrations per task, rigidly memorizing trajectories, and failing to adapt when deployment conditions deviate from training. We introduce EVOLVE-VLA, a test-time training framework enabling VLAs to continuously adapt through environment interaction with minimal or zero task-specific demonstrations. The key technical challenge is replacing oracle reward signals (unavailable at test time) with autonomous feedback. We address this through a learned progress estimator providing dense feedback, and critically, we design our framework to ``tame'' this inherently noisy signal via two mechanisms: (1) an accumulative progress estimation mechanism smoothing noisy point-wise estimates, and (2) a progressive horizon extension strategy enabling gradual policy evolution. EVOLVE-VLA achieves substantial gains: +8.6\% on long-horizon tasks, +22.0\% in 1-shot learning, and enables cross-task generalization -- achieving 20.8\% success on unseen tasks without task-specific demonstrations training (vs. 0\% for pure SFT). Qualitative analysis reveals emergent capabilities absent in demonstrations, including error recovery and novel strategies. This work represents a critical step toward VLAs that truly learn and adapt, moving beyond static imitation toward continuous self-improvements.

Paperid: 4067, https://arxiv.org/pdf/2512.11060.pdf

Abstract:
Vision-Language Models (VLMs) offer a promising path toward interpretable medical diagnosis by allowing users to ask about clinical explanations alongside predictions and across different modalities. However, training VLMs for detailed reasoning requires large-scale image-text datasets. In many specialized domains, for example in reading Optical Coherence Tomography Angiography (OCTA) images, such precise text with grounded description of pathologies is scarce or even non-existent. To overcome this bottleneck, we introduce Synthetic Vasculature Reasoning (SVR), a framework that controllably synthesizes images and corresponding text, specifically: realistic retinal vasculature with Diabetic Retinopathy (DR) features: capillary dropout, microaneurysms, neovascularization, and tortuosity, while automatically generating granular reasoning texts. Based on this we curate OCTA-100K-SVR, an OCTA image-reasoning dataset with 100,000 pairs. Our experiments show that a general-purpose VLM (Qwen3-VL-8b) trained on the dataset achieves a zero-shot balanced classification accuracy of 89.67% on real OCTA images, outperforming supervised baselines. Through human expert evaluation we also demonstrate that it significantly enhances explanation quality and pathology localization on clinical data.

Paperid: 4068, https://arxiv.org/pdf/2512.10394.pdf

Abstract:
Current embodied AI systems face severe engineering impediments, primarily characterized by poor cross-scenario adaptability, rigid inter-module coupling, and fragmented inference acceleration. To overcome these limitations, we propose RoboNeuron, a universal deployment framework for embodied intelligence. RoboNeuron is the first framework to deeply integrate the cognitive capabilities of Large Language Models (LLMs) and Vision-Language-Action (VLA) models with the real-time execution backbone of the Robot Operating System (ROS). We utilize the Model Context Protocol (MCP) as a semantic bridge, enabling the LLM to dynamically orchestrate underlying robotic tools. The framework establishes a highly modular architecture that strictly decouples sensing, reasoning, and control by leveraging ROS's unified communication interfaces. Crucially, we introduce an automated tool to translate ROS messages into callable MCP functions, significantly streamlining development. RoboNeuron significantly enhances cross-scenario adaptability and component flexibility, while establishing a systematic platform for horizontal performance benchmarking, laying a robust foundation for scalable real-world embodied applications.

Paperid: 4069, https://arxiv.org/pdf/2512.09446.pdf

Abstract:
Recent vision language models (VLMs) like CLIP have demonstrated impressive anomaly detection performance under significant distribution shift by utilizing high-level semantic information through text prompts. However, these models often neglect fine-grained details, such as which kind of anomalies, like "hole", "cut", "scratch" that could provide more specific insight into the nature of anomalies. We argue that recognizing fine-grained anomaly types 1) enriches the representation of "abnormal" with structured semantics, narrowing the gap between coarse anomaly signals and fine-grained defect categories; 2) enables manufacturers to understand the root causes of the anomaly and implement more targeted and appropriate corrective measures quickly. While incorporating such detailed semantic information is crucial, designing handcrafted prompts for each defect type is both time-consuming and susceptible to human bias. For this reason, we introduce DAPO, a novel approach for Defect-aware Prompt Optimization based on progressive tuning for the zero-shot multi-type and binary anomaly detection and segmentation under distribution shifts. Our approach aligns anomaly-relevant image features with their corresponding text semantics by learning hybrid defect-aware prompts with both fixed textual anchors and learnable token embeddings. We conducted experiments on public benchmarks (MPDD, VisA, MVTec-AD, MAD, and Real-IAD) and an internal dataset. The results suggest that compared to the baseline models, DAPO achieves a 3.7% average improvement in AUROC and average precision metrics at the image level under distribution shift, and a 6.5% average improvement in localizing novel anomaly types under zero-shot settings.

Paperid: 4070, https://arxiv.org/pdf/2512.09349.pdf

Abstract:
End-to-end autonomous driving frameworks face persistent challenges in generalization, training efficiency, and interpretability. While recent methods leverage Vision-Language Models (VLMs) through supervised learning on large-scale datasets to improve reasoning, they often lack robustness in novel scenarios. Conversely, reinforcement learning (RL)-based approaches enhance adaptability but remain data-inefficient and lack transparent decision-making. % contribution To address these limitations, we propose COVLM-RL, a novel end-to-end driving framework that integrates Critical Object-oriented (CO) reasoning with VLM-guided RL. Specifically, we design a Chain-of-Thought (CoT) prompting strategy that enables the VLM to reason over critical traffic elements and generate high-level semantic decisions, effectively transforming multi-view visual inputs into structured semantic decision priors. These priors reduce the input dimensionality and inject task-relevant knowledge into the RL loop, accelerating training and improving policy interpretability. However, bridging high-level semantic guidance with continuous low-level control remains non-trivial. To this end, we introduce a consistency loss that encourages alignment between the VLM's semantic plans and the RL agent's control outputs, enhancing interpretability and training stability. Experiments conducted in the CARLA simulator demonstrate that COVLM-RL significantly improves the success rate by 30\% in trained driving environments and by 50\% in previously unseen environments, highlighting its strong generalization capability.

Paperid: 4071, https://arxiv.org/pdf/2512.05107.pdf

Abstract:
Recent advances in Vision-Language-Action (VLA) models, powered by large language models and reinforcement learning-based fine-tuning, have shown remarkable progress in robotic manipulation. Existing methods often treat long-horizon actions as linguistic sequences and apply trajectory-level optimization methods such as Trajectory-wise Preference Optimization (TPO) or Proximal Policy Optimization (PPO), leading to coarse credit assignment and unstable training. However, unlike language, where a unified semantic meaning is preserved despite flexible sentence order, action trajectories progress through causally chained stages with different learning difficulties. This motivates progressive stage optimization. Thereby, we present Stage-Aware Reinforcement (STARE), a module that decomposes a long-horizon action trajectory into semantically meaningful stages and provides dense, interpretable, and stage-aligned reinforcement signals. Integrating STARE into TPO and PPO, we yield Stage-Aware TPO (STA-TPO) and Stage-Aware PPO (STA-PPO) for offline stage-wise preference and online intra-stage interaction, respectively. Further building on supervised fine-tuning as initialization, we propose the Imitation -> Preference -> Interaction (IPI), a serial fine-tuning pipeline for improving action accuracy in VLA models. Experiments on SimplerEnv and ManiSkill3 demonstrate substantial gains, achieving state-of-the-art success rates of 98.0 percent on SimplerEnv and 96.4 percent on ManiSkill3 tasks.

Paperid: 4072, https://arxiv.org/pdf/2512.04952.pdf

Abstract:
Autoregressive vision-language-action (VLA) models have recently demonstrated strong capabilities in robotic manipulation. However, their core process of action tokenization often involves a trade-off between reconstruction fidelity and inference efficiency. We introduce FASTer, a unified framework for efficient and generalizable robot learning that integrates a learnable tokenizer with an autoregressive policy built upon it. FASTerVQ encodes action chunks as single-channel images, capturing global spatio-temporal dependencies while maintaining a high compression ratio. FASTerVLA builds on this tokenizer with block-wise autoregressive decoding and a lightweight action expert, achieving both faster inference and higher task performance. Extensive experiments across simulated and real-world benchmarks show that FASTerVQ delivers superior reconstruction quality, high token utilization, and strong cross-task and cross-embodiment generalization, while FASTerVLA further improves overall capability, surpassing previous state-of-the-art VLA models in both inference speed and task performance.

Paperid: 4073, https://arxiv.org/pdf/2512.01550.pdf

Abstract:
Embodied navigation for long-horizon tasks, guided by complex natural language instructions, remains a formidable challenge in artificial intelligence. Existing agents often struggle with robust long-term planning about unseen environments, leading to high failure rates. To address these limitations, we introduce NavForesee, a novel Vision-Language Model (VLM) that unifies high-level language planning and predictive world model imagination within a single, unified framework. Our approach empowers a single VLM to concurrently perform planning and predictive foresight. Conditioned on the full instruction and historical observations, the model is trained to understand the navigation instructions by decomposing the task, tracking its progress, and formulating the subsequent sub-goal. Simultaneously, it functions as a generative world model, providing crucial foresight by predicting short-term environmental dynamics and long-term navigation milestones. The VLM's structured plan guides its targeted prediction, while the imagined future provides rich context to inform the navigation actions, creating a powerful internal feedback loop of perception-planning/prediction-action. We demonstrate through extensive experiments on the R2R-CE and RxR-CE benchmark that NavForesee achieves highly competitive performance in complex scenarios. Our work highlights the immense potential of fusing explicit language planning with implicit spatiotemporal prediction, paving the way for more intelligent and capable embodied agents.

Paperid: 4074, https://arxiv.org/pdf/2512.00797.pdf

Abstract:
Foundation models have become central to unifying perception and planning in robotics, yet real-world deployment exposes a mismatch between their monolithic assumption that a single model can handle all cognitive functions and the distributed, dynamic nature of practical service workflows. Vision-language models offer strong semantic understanding but lack embodiment-aware action capabilities while relying on hand-crafted skills. Vision-Language-Action policies enable reactive manipulation but remain brittle across embodiments, weak in geometric grounding, and devoid of proactive collaboration mechanisms. These limitations indicate that scaling a single model alone cannot deliver reliable autonomy for service robots operating in human-populated settings. To address this gap, we present InteractGen, an LLM-powered multi-agent framework that decomposes robot intelligence into specialized agents for continuous perception, dependency-aware planning, decision and verification, failure reflection, and dynamic human delegation, treating foundation models as regulated components within a closed-loop collective. Deployed on a heterogeneous robot team and evaluated in a three-month open-use study, InteractGen improves task success, adaptability, and human-robot collaboration, providing evidence that multi-agent orchestration offers a more feasible path toward socially grounded service autonomy than further scaling standalone models.

Paperid: 4075, https://arxiv.org/pdf/2511.21397.pdf

Abstract:
How does irrelevant information (i.e., distractors) affect test-time scaling in vision-language models (VLMs)? Prior studies on language models have reported an inverse scaling effect, where textual distractors lead to longer but less effective reasoning. To investigate whether similar phenomena occur in multimodal settings, we introduce Idis (Images with distractors), a visual question-answering dataset that systematically varies distractors along semantic, numerical, and spatial dimensions. Our analyses reveal that visual distractors differ fundamentally from textual ones: although inverse scaling persists, adding visual distractors reduces accuracy without increasing reasoning length. We further show that tracking attribute counts within reasoning traces provides key insights into how distractors, reasoning length, and accuracy interact. Finally, we demonstrate that these trends extend to established visual bias benchmarks such as Waterbirds, and we propose a simple prompting strategy to mitigate bias-driven predictions in reasoning models.

Paperid: 4076, https://arxiv.org/pdf/2511.21053.pdf

Abstract:
Referring Multi-Object Tracking (RMOT) aims to achieve precise object detection and tracking through natural language instructions, representing a fundamental capability for intelligent robotic systems. However, current RMOT research remains mostly confined to ground-level scenarios, which constrains their ability to capture broad-scale scene contexts and perform comprehensive tracking and path planning. In contrast, Unmanned Aerial Vehicles (UAVs) leverage their expansive aerial perspectives and superior maneuverability to enable wide-area surveillance. Moreover, UAVs have emerged as critical platforms for Embodied Intelligence, which has given rise to an unprecedented demand for intelligent aerial systems capable of natural language interaction. To this end, we introduce AerialMind, the first large-scale RMOT benchmark in UAV scenarios, which aims to bridge this research gap. To facilitate its construction, we develop an innovative semi-automated collaborative agent-based labeling assistant (COALA) framework that significantly reduces labor costs while maintaining annotation quality. Furthermore, we propose HawkEyeTrack (HETrack), a novel method that collaboratively enhances vision-language representation learning and improves the perception of UAV scenarios. Comprehensive experiments validated the challenging nature of our dataset and the effectiveness of our method.

Paperid: 4077, https://arxiv.org/pdf/2511.20644.pdf

Abstract:
Spatial reasoning is a critical capability for intelligent robots, yet current vision-language models (VLMs) still fall short of human-level performance in video-based spatial reasoning. This gap mainly stems from two challenges: a semantic-geometric misalignment that prevents consistent 3D understanding, and the absence of persistent memory to retain 3D representation and understanding over time. To address these limitations, we present VLM$^2$, a Vision-Language Model with persistent Memory for spatial reasoning with a view-consistent, 3D-aware representation purely from 2D video. Specifically, to enhance long-horizon reasoning, we incorporate a dual-memory module, consisting of a working memory that operates as a sliding window to focus on immediate context, and an episodic memory that consolidates and stores critical long-term information. This design enables efficient and long-horizon spatial reasoning with a fixed computational cost. Extensive experiments on multiple benchmarks show that VLM$^2$ achieves state-of-the-art performance among video-only models, significantly advancing the frontier of visual-spatial intelligence.

Paperid: 4078, https://arxiv.org/pdf/2511.20643.pdf

Abstract:
What data should a vision-language model be trained on? To answer this question, many data curation efforts center on the quality of a dataset. However, most of these existing methods are (i) offline, i.e. they produce a static dataset from a set of predetermined filtering criteria, and (ii) concept-agnostic, i.e. they use model-based filters which induce additional data biases. In this work, we go beyond such offline, concept-agnostic methods and advocate for more flexible, task-adaptive online concept-based curation. Our first contribution is DataConcept, a collection of 128M web-crawled image-text pairs annotated with fine-grained details about their concept composition. Building on DataConcept, we introduce Concept-Aware Batch Sampling (CABS), a simple yet effective batch sampling framework that flexibly constructs batches on-the-fly based on specific target distributions. We propose two variants: (i) Diversity Maximization (CABS-DM) to curate batches with a broad coverage of available concepts, and (ii) Frequency Maximization (CABS-FM) to curate batches with high object multiplicity. Through extensive evaluations across 28 benchmarks, we demonstrate that our CABS method significantly benefits CLIP/SigLIP model classes and yields highly performant models. Overall, CABS represents a strong open-source alternative to proprietary online data curation algorithms, enabling practitioners to define custom concept distributions that optimize for specific downstream tasks.

Paperid: 4079, https://arxiv.org/pdf/2511.20145.pdf

Abstract:
Positron emission tomography/computed tomography (PET/CT) is essential in oncology, yet the rapid expansion of scanners has outpaced the availability of trained specialists, making automated PET/CT report generation (PETRG) increasingly important for reducing clinical workload. Compared with structural imaging (e.g., X-ray, CT, and MRI), functional PET poses distinct challenges: metabolic patterns vary with tracer physiology, and whole-body 3D contextual information is required rather than local-region interpretation. To advance PETRG, we propose PETRG-3D, an end-to-end 3D dual-branch framework that separately encodes PET and CT volumes and incorporates style-adaptive prompts to mitigate inter-hospital variability in reporting practices. We construct PETRG-Lym, a multi-center lymphoma dataset collected from four hospitals (824 reports w/ 245,509 paired PET/CT slices), and construct AutoPET-RG-Lym, a publicly accessible PETRG benchmark derived from open imaging data but equipped with new expert-written, clinically validated reports (135 cases). To assess clinical utility, we introduce PETRG-Score, a lymphoma-specific evaluation protocol that jointly measures metabolic and structural findings across curated anatomical regions. Experiments show that PETRG-3D substantially outperforms existing methods on both natural language metrics (e.g., +31.49\% ROUGE-L) and clinical efficacy metrics (e.g., +8.18\% PET-All), highlighting the benefits of volumetric dual-modality modeling and style-aware prompting. Overall, this work establishes a foundation for future PET/CT-specific models emphasizing disease-aware reasoning and clinically reliable evaluation. Codes, models, and AutoPET-RG-Lym will be released.

Paperid: 4080, https://arxiv.org/pdf/2511.19155.pdf

Abstract:
Sleep stage classification based on electroencephalography (EEG) is fundamental for assessing sleep quality and diagnosing sleep-related disorders. However, most traditional machine learning methods rely heavily on prior knowledge and handcrafted features, while existing deep learning models still struggle to jointly capture fine-grained time-frequency patterns and achieve clinical interpretability. Recently, vision-language models (VLMs) have made significant progress in the medical domain, yet their performance remains constrained when applied to physiological waveform data, especially EEG signals, due to their limited visual understanding and insufficient reasoning capability. To address these challenges, we propose EEG-VLM, a hierarchical vision-language framework that integrates multi-level feature alignment with visually enhanced language-guided reasoning for interpretable EEG-based sleep stage classification. Specifically, a specialized visual enhancement module constructs high-level visual tokens from intermediate-layer features to extract rich semantic representations of EEG images. These tokens are further aligned with low-level CLIP features through a multi-level alignment mechanism, enhancing the VLM's image-processing capability. In addition, a Chain-of-Thought (CoT) reasoning strategy decomposes complex medical inference into interpretable logical steps, effectively simulating expert-like decision-making. Experimental results demonstrate that the proposed method significantly improves both the accuracy and interpretability of VLMs in EEG-based sleep stage classification, showing promising potential for automated and explainable EEG analysis in clinical settings.

Paperid: 4081, https://arxiv.org/pdf/2511.19119.pdf

Abstract:
Spatial reasoning (SR), the ability to infer 3D spatial information from 2D inputs, is essential for real-world applications such as embodied AI and autonomous driving. However, existing research primarily focuses on indoor environments and typically relies on multi-view observations, which limits their generalizability to outdoor scenarios and constrains their applicability to monocular images, the most common real-world setting. In this work, we propose MonoSR, a large-scale monocular spatial reasoning dataset that spans diverse scenarios including indoor, outdoor, and object-centric settings, and supports multiple question types. MonoSR provides a path toward open-world monocular spatial reasoning. Beyond introducing the dataset, we evaluate advanced vision-language models to reveal their limitations on this challenging task. We further analyze whether auxiliary information is crucial for monocular spatial reasoning and offer practical guidance for designing future models. These contributions collectively establish a foundation for advancing monocular spatial reasoning in real-world, open-world environments.

Paperid: 4082, https://arxiv.org/pdf/2511.16757.pdf

Abstract:
Audio-language pretraining holds promise for general-purpose audio understanding, yet remains underexplored compared to its vision counterpart. While vision-language models like CLIP serve as widely adopted foundations, existing audio-language models primarily excel at retrieval tasks with limited adoption as general-purpose encoders. We identify three key barriers: limited large-scale audio-text corpora, insufficient caption diversity, and lack of systematic exploration and evaluation. To this end, we introduce CaptionStew, a 10.7M caption dataset aggregating diverse open-source audio-text corpora across multiple domains and captioning styles. Using this resource, we conduct the first comprehensive evaluation comparing contrastive and captioning objectives for audio representation learning across speech, music, and environmental sound tasks. Our results demonstrate that audio-language pretraining yields competitive, transferable representations. Through systematic data-scaling experiments, we reveal complementary objective strengths: contrastive learning achieves superior data efficiency at smaller scales, while captioning demonstrates better scalability on language-involved audio understanding tasks. We also find that common supervised initialization practices provide diminishing returns at scale, challenging current approaches. These findings establish audio-language pretraining as a viable pathway toward general-purpose audio representations, guiding future research. To accelerate progress, we release data preparation recipes, training protocols, and pretrained models, paving the way toward universal audio understanding.

Paperid: 4083, https://arxiv.org/pdf/2511.11052.pdf

Abstract:
Non-prehensile (NP) manipulation, in which robots alter object states without forming stable grasps (for example, pushing, poking, or sliding), significantly broadens robotic manipulation capabilities when grasping is infeasible or insufficient. However, enabling a unified framework that generalizes across different tasks, objects, and environments while seamlessly integrating non-prehensile and prehensile (P) actions remains challenging: robots must determine when to invoke NP skills, select the appropriate primitive for each context, and compose P and NP strategies into robust, multi-step plans. We introduce ApaptPNP, a vision-language model (VLM)-empowered task and motion planning framework that systematically selects and combines P and NP skills to accomplish diverse manipulation objectives. Our approach leverages a VLM to interpret visual scene observations and textual task descriptions, generating a high-level plan skeleton that prescribes the sequence and coordination of P and NP actions. A digital-twin based object-centric intermediate layer predicts desired object poses, enabling proactive mental rehearsal of manipulation sequences. Finally, a control module synthesizes low-level robot commands, with continuous execution feedback enabling online task plan refinement and adaptive replanning through the VLM. We evaluate ApaptPNP across representative P&NP hybrid manipulation tasks in both simulation and real-world environments. These results underscore the potential of hybrid P&NP manipulation as a crucial step toward general-purpose, human-level robotic manipulation capabilities. Project Website: https://sites.google.com/view/adaptpnp/home

Paperid: 4084, https://arxiv.org/pdf/2511.10254.pdf

Abstract:
Facial Emotion Analysis (FEA) extends traditional facial emotion recognition by incorporating explainable, fine-grained reasoning. The task integrates three subtasks: emotion recognition, facial Action Unit (AU) recognition, and AU-based emotion reasoning to model affective states jointly. While recent approaches leverage Vision-Language Models (VLMs) and achieve promising results, they face two critical limitations: (1) hallucinated reasoning, where VLMs generate plausible but inaccurate explanations due to insufficient emotion-specific knowledge; and (2) misalignment between emotion reasoning and recognition, caused by fragmented connections between observed facial features and final labels. We propose Facial-R1, a three-stage alignment framework that effectively addresses both challenges with minimal supervision. First, we employ instruction fine-tuning to establish basic emotional reasoning capability. Second, we introduce reinforcement training guided by emotion and AU labels as reward signals, which explicitly aligns the generated reasoning process with the predicted emotion. Third, we design a data synthesis pipeline that iteratively leverages the prior stages to expand the training dataset, enabling scalable self-improvement of the model. Built upon this framework, we introduce FEA-20K, a benchmark dataset comprising 17,737 training and 1,688 test samples with fine-grained emotion analysis annotations. Extensive experiments across eight standard benchmarks demonstrate that Facial-R1 achieves state-of-the-art performance in FEA, with strong generalization and robust interpretability.

Paperid: 4085, https://arxiv.org/pdf/2510.27280.pdf

Abstract:
Multimodal large language models (MLLMs) represent images and video frames as visual tokens. Scaling from single images to hour-long videos, however, inflates the token budget far beyond practical limits. Popular pipelines therefore either uniformly subsample or apply keyframe selection with retrieval-style scoring using smaller vision-language models. However, these keyframe selection methods still rely on pre-filtering before selection to reduce the inference cost and can miss the most informative moments. We propose FOCUS, Frame-Optimistic Confidence Upper-bound Selection, a training-free, model-agnostic keyframe selection module that selects query-relevant frames under a strict token budget. FOCUS formulates keyframe selection as a combinatorial pure-exploration (CPE) problem in multi-armed bandits: it treats short temporal clips as arms, and uses empirical means and Bernstein confidence radius to identify informative regions while preserving exploration of uncertain areas. The resulting two-stage exploration-exploitation procedure reduces from a sequential policy with theoretical guarantees, first identifying high-value temporal regions, then selecting top-scoring frames within each region On two long-video question-answering benchmarks, FOCUS delivers substantial accuracy improvements while processing less than 2% of video frames. For videos longer than 20 minutes, it achieves an 11.9% gain in accuracy on LongVideoBench, demonstrating its effectiveness as a keyframe selection method and providing a simple and general solution for scalable long-video understanding with MLLMs.

Paperid: 4086, https://arxiv.org/pdf/2510.25413.pdf

Abstract:
Most existing sign language translation (SLT) datasets are limited in scale, lack multilingual coverage, and are costly to curate due to their reliance on expert annotation and controlled recording setup. Recently, Vision Language Models (VLMs) have demonstrated strong capabilities as evaluators and real-time assistants. Despite these advancements, their potential remains untapped in the context of sign language dataset acquisition. To bridge this gap, we introduce the first automated annotation and filtering framework that utilizes VLMs to reduce reliance on manual effort while preserving data quality. Our method is applied to TikTok videos across eight sign languages and to the already curated YouTube-SL-25 dataset in German Sign Language for the purpose of additional evaluation. Our VLM-based pipeline includes a face visibility detection, a sign activity recognition, a text extraction from video content, and a judgment step to validate alignment between video and text, implementing generic filtering, annotation and validation steps. Using the resulting corpus, TikTok-SL-8, we assess the performance of two off-the-shelf SLT models on our filtered dataset for German and American Sign Languages, with the goal of establishing baselines and evaluating the robustness of recent models on automatically extracted, slightly noisy data. Our work enables scalable, weakly supervised pretraining for SLT and facilitates data acquisition from social media.

Paperid: 4087, https://arxiv.org/pdf/2510.22798.pdf

Abstract:
Automatically assessing handwritten mathematical solutions is an important problem in educational technology with practical applications, but it remains a significant challenge due to the diverse formats, unstructured layouts, and symbolic complexity of student work. To address this challenge, we introduce VEHME-a Vision-Language Model for Evaluating Handwritten Mathematics Expressions-designed to assess open-form handwritten math responses with high accuracy and interpretable reasoning traces. VEHME integrates a two-phase training pipeline: (i) supervised fine-tuning using structured reasoning data, and (ii) reinforcement learning that aligns model outputs with multi-dimensional grading objectives, including correctness, reasoning depth, and error localization. To enhance spatial understanding, we propose an Expression-Aware Visual Prompting Module, trained on our synthesized multi-line math expressions dataset to robustly guide attention in visually heterogeneous inputs. Evaluated on AIHub and FERMAT datasets, VEHME achieves state-of-the-art performance among open-source models and approaches the accuracy of proprietary systems, demonstrating its potential as a scalable and accessible tool for automated math assessment. Our training and experiment code is publicly available at our GitHub repository.

Paperid: 4088, https://arxiv.org/pdf/2510.22171.pdf

Abstract:
The growing deployment of Vision-Language Models (VLMs) in high-stakes applications such as autonomous driving and assistive technologies for visually impaired individuals necessitates reliable mechanisms to assess the trustworthiness of their generation. Uncertainty Estimation (UE) plays a central role in quantifying the reliability of model outputs and reducing unsafe generations via selective prediction. In this regard, most existing probability-based UE approaches rely on output probability distributions, aggregating token probabilities into a single uncertainty score using predefined functions such as length-normalization. Another line of research leverages model hidden representations and trains MLP-based models to predict uncertainty. However, these methods often fail to capture the complex multimodal relationships between semantic and textual tokens and struggle to identify biased probabilities often influenced by language priors. Motivated by these observations, we propose a novel UE framework, HARMONY, that jointly leverages fused multimodal information in model activations and the output distribution of the VLM to determine the reliability of responses. The key hypothesis of our work is that both the model's internal belief in its visual understanding, captured by its hidden representations, and the produced token probabilities carry valuable reliability signals that can be jointly leveraged to improve UE performance, surpassing approaches that rely on only one of these components. Experimental results on three open-ended VQA benchmarks, A-OKVQA, VizWiz, and PathVQA, and three state-of-the-art VLMs, LLaVa-7b, LLaVA-13b and InstructBLIP demonstrate that our method consistently performs on par with or better than existing approaches, achieving up to 4\% improvement in AUROC, and 6\% in PRR, establishing new state of the art in uncertainty estimation for VLMs.

Paperid: 4089, https://arxiv.org/pdf/2510.18439.pdf

Abstract:
Hallucination, where models generate fluent text unsupported by visual evidence, remains a major flaw in vision-language models and is particularly critical in sign language translation (SLT). In SLT, meaning depends on precise grounding in video, and gloss-free models are especially vulnerable because they map continuous signer movements directly into natural language without intermediate gloss supervision that serves as alignment. We argue that hallucinations arise when models rely on language priors rather than visual input. To capture this, we propose a token-level reliability measure that quantifies how much the decoder uses visual information. Our method combines feature-based sensitivity, which measures internal changes when video is masked, with counterfactual signals, which capture probability differences between clean and altered video inputs. These signals are aggregated into a sentence-level reliability score, providing a compact and interpretable measure of visual grounding. We evaluate the proposed measure on two SLT benchmarks (PHOENIX-2014T and CSL-Daily) with both gloss-based and gloss-free models. Our results show that reliability predicts hallucination rates, generalizes across datasets and architectures, and decreases under visual degradations. Beyond these quantitative trends, we also find that reliability distinguishes grounded tokens from guessed ones, allowing risk estimation without references; when combined with text-based signals (confidence, perplexity, or entropy), it further improves hallucination risk estimation. Qualitative analysis highlights why gloss-free models are more susceptible to hallucinations. Taken together, our findings establish reliability as a practical and reusable tool for diagnosing hallucinations in SLT, and lay the groundwork for more robust hallucination detection in multimodal generation.

Paperid: 4090, https://arxiv.org/pdf/2510.17111.pdf

Abstract:
Vision-Language-Action (VLA) models extend vision-language models to embodied control by mapping natural-language instructions and visual observations to robot actions. Despite their capabilities, VLA systems face significant challenges due to their massive computational and memory demands, which conflict with the constraints of edge platforms such as on-board mobile manipulators that require real-time performance. Addressing this tension has become a central focus of recent research. In light of the growing efforts toward more efficient and scalable VLA systems, this survey provides a systematic review of approaches for improving VLA efficiency, with an emphasis on reducing latency, memory footprint, and training and inference costs. We categorize existing solutions into four dimensions: model architecture, perception feature, action generation, and training/inference strategies, summarizing representative techniques within each category. Finally, we discuss future trends and open challenges, highlighting directions for advancing efficient embodied intelligence.

Paperid: 4091, https://arxiv.org/pdf/2510.13375.pdf

Abstract:
Vision-Language-Action (VLA) models have recently shown impressive generalization and language-guided manipulation capabilities. However, their performance degrades on tasks requiring precise spatial reasoning due to limited spatial reasoning inherited from Vision-Language Models (VLMs). Existing VLAs rely on extensive action-data pretraining to ground VLMs in 3D space, which reduces training efficiency and is still insufficient for accurate spatial understanding. In this work, we present DepthVLA, a simple yet effective VLA architecture that explicitly incorporates spatial awareness through a pretrained depth prediction module. DepthVLA adopts a mixture-of-transformers design that unifies a VLM, a depth transformer, and an action expert with fully shared attentions, forming an end-to-end model with enhanced spatial reasoning. Extensive evaluations in both real-world and simulated environments show that DepthVLA outperforms state-of-the-art approaches, achieving 78.5% vs. 65.0% progress in real-world tasks, 94.9% vs. 93.6% in the LIBERO simulator, and 74.8% vs. 58.8% in the Simpler simulator. Our code will be made publicly available.

Paperid: 4092, https://arxiv.org/pdf/2510.13232.pdf

Abstract:
State-of-the-art vision-language models (VLMs) suffer from a critical failure in understanding negation, often referred to as affirmative bias. This limitation is particularly severe in described object detection (DOD) tasks. To address this, we propose two primary contributions: (1) a new dataset pipeline and (2) a novel, lightweight adaptation recipe. First, we introduce CoVAND, a dataset constructed with a systematic chain-of-thought (CoT) and VQA-based pipeline to generate high-quality, instance-grounded negation data. Second, we propose NegToMe, a novel text token merging module that directly tackles the architectural cause of affirmative bias. NegToMe fundamentally addresses the structural loss of negation cues in tokenization, grouping them with attributes into coherent semantic phrases. It maintains correct polarity at the input level, enabling robust negation understanding even with limited data. For instance, to prevent a model from treating the fragmented tokens "not" and "girl" as simply "girl", NegToMe binds them into a single token whose meaning is correctly distinguished from that of "girl" alone. This module is integrated with a parameter-efficient and strategic LoRA fine-tuning approach. Our method significantly improves performance on challenging negation benchmarks with a lowered false positive rate, boosting NMS-AP by up to +10.8 points on OVDEval and demonstrating generalization to SoTA VLMs. This work marks a crucial step forward in addressing negation understanding for real-world detection applications.

Paperid: 4093, https://arxiv.org/pdf/2510.12444.pdf

Abstract:
Chest Xray imaging is a widely used diagnostic tool in modern medicine, and its high utilization creates substantial workloads for radiologists. To alleviate this burden, vision language models are increasingly applied to automate Chest Xray radiology report generation (CXRRRG), aiming for clinically accurate descriptions while reducing manual effort. Conventional approaches, however, typically rely on single images, failing to capture the longitudinal context necessary for producing clinically faithful comparison statements. Recently, growing attention has been directed toward incorporating longitudinal data into CXR RRG, enabling models to leverage historical studies in ways that mirror radiologists diagnostic workflows. Nevertheless, existing surveys primarily address single image CXRRRG and offer limited guidance for longitudinal settings, leaving researchers without a systematic framework for model design. To address this gap, this survey provides the first comprehensive review of longitudinal radiology report generation (LRRG). Specifically, we examine dataset construction strategies, report generation architectures alongside longitudinally tailored designs, and evaluation protocols encompassing both longitudinal specific measures and widely used benchmarks. We further summarize LRRG methods performance, alongside analyses of different ablation studies, which collectively highlight the critical role of longitudinal information and architectural design choices in improving model performance. Finally, we summarize five major limitations of current research and outline promising directions for future development, aiming to lay a foundation for advancing this emerging field.

Paperid: 4094, https://arxiv.org/pdf/2510.04428.pdf

Abstract:
Effectively applying Vision-Language Models (VLMs) to Video Question Answering (VideoQA) hinges on selecting a concise yet comprehensive set of frames, as processing entire videos is computationally infeasible. However, current frame selection methods face a critical trade-off: approaches relying on lightweight similarity models, such as CLIP, often fail to capture the nuances of complex queries, resulting in inaccurate similarity scores that cannot reflect the authentic query-frame relevance, which further undermines frame selection. Meanwhile, methods that leverage a VLM for deeper analysis achieve higher accuracy but incur prohibitive computational costs. To address these limitations, we propose A.I.R., a training-free approach for Adaptive, Iterative, and Reasoning-based frame selection. We leverage a powerful VLM to perform deep, semantic analysis on complex queries, and this analysis is deployed within a cost-effective iterative loop that processes only a small batch of the most high-potential frames at a time. Extensive experiments on various VideoQA benchmarks demonstrate that our approach outperforms existing frame selection methods, significantly boosts the performance of the foundation VLM, and achieves substantial gains in computational efficiency over other VLM-based techniques.

Paperid: 4095, https://arxiv.org/pdf/2510.04257.pdf

Abstract:
Multimodal agents built on large vision-language models (LVLMs) are increasingly deployed in open-world settings but remain highly vulnerable to prompt injection, especially through visual inputs. We introduce AgentTypo, a black-box red-teaming framework that mounts adaptive typographic prompt injection by embedding optimized text into webpage images. Our automatic typographic prompt injection (ATPI) algorithm maximizes prompt reconstruction by substituting captioners while minimizing human detectability via a stealth loss, with a Tree-structured Parzen Estimator guiding black-box optimization over text placement, size, and color. To further enhance attack strength, we develop AgentTypo-pro, a multi-LLM system that iteratively refines injection prompts using evaluation feedback and retrieves successful past examples for continual learning. Effective prompts are abstracted into generalizable strategies and stored in a strategy repository, enabling progressive knowledge accumulation and reuse in future attacks. Experiments on the VWA-Adv benchmark across Classifieds, Shopping, and Reddit scenarios show that AgentTypo significantly outperforms the latest image-based attacks such as AgentAttack. On GPT-4o agents, our image-only attack raises the success rate from 0.23 to 0.45, with consistent results across GPT-4V, GPT-4o-mini, Gemini 1.5 Pro, and Claude 3 Opus. In image+text settings, AgentTypo achieves 0.68 ASR, also outperforming the latest baselines. Our findings reveal that AgentTypo poses a practical and potent threat to multimodal agents and highlight the urgent need for effective defense.

Paperid: 4096, https://arxiv.org/pdf/2510.02324.pdf

Abstract:
Large Language Models (LLMs) exhibit impressive capabilities but often hallucinate, confidently providing incorrect answers instead of admitting ignorance. Prior work has shown that models encode linear representations of their own knowledge and that activation steering can reduce hallucinations. These approaches, however, require real-time monitoring and intervention during inference. We introduce Contrastive Activation Steering for Amortized Learning (CASAL), an efficient algorithm that connects interpretability with amortized optimization. CASAL directly bakes the benefits of activation steering into model's weights. Once trained, LLMs answer questions they know while abstaining from answering those they do not. CASAL's light-weight design requires training only a submodule of a single transformer layer and yet reduces hallucination by 30%-40% across multiple short-form QA benchmarks. CASAL is 30x more compute-efficient and 20x more data-efficient than strong LoRA-based baselines such as SFT and DPO, boosting its practical applicability in data scarce domains. Importantly, CASAL also generalizes effectively to out-of-distribution (OOD) domains. We showcase CASAL's flexibility in mitigating hallucinations in both text-only and vision-language models. To our knowledge, CASAL is the first steering-based training method that has been shown to be effective for both dense and Mixture-of-Experts (MoE) models. CASAL represents a promising step forward for applying interpretability-inspired method for practical deployment in production systems.

Paperid: 4097, https://arxiv.org/pdf/2509.25757.pdf

Abstract:
Modern Vision-Language Models (VLMs) have achieved impressive performance in various tasks, yet they often struggle with compositional reasoning, the ability to decompose and recombine concepts to solve novel problems. While neuro-symbolic approaches offer a promising direction, they are typically constrained by crisp logical execution or predefined predicates, which limit flexibility. In this work, we introduce NePTune, a neuro-symbolic framework that overcomes these limitations through a hybrid execution model that integrates the perception capabilities of foundation vision models with the compositional expressiveness of symbolic reasoning. NePTune dynamically translates natural language queries into executable Python programs that blend imperative control flow with soft logic operators capable of reasoning over VLM-generated uncertainty. Operating in a training-free manner, NePTune, with a modular design, decouples perception from reasoning, yet its differentiable operations support fine-tuning. We evaluate NePTune on multiple visual reasoning benchmarks and various domains, utilizing adversarial tests, and demonstrate a significant improvement over strong base models, as well as its effective compositional generalization and adaptation capabilities in novel environments.

Paperid: 4098, https://arxiv.org/pdf/2509.22524.pdf

Abstract:
Color serves as a fundamental dimension of human visual perception and a primary means of communicating about objects and scenes. As vision-language models (VLMs) become increasingly prevalent, understanding whether they name colors like humans is crucial for effective human-AI interaction. We present the first systematic evaluation of color naming capabilities across VLMs, replicating classic color naming methodologies using 957 color samples across five representative models. Our results show that while VLMs achieve high accuracy on prototypical colors from classical studies, performance drops significantly on expanded, non-prototypical color sets. We identify 21 common color terms that consistently emerge across all models, revealing two distinct approaches: constrained models using predominantly basic terms versus expansive models employing systematic lightness modifiers. Cross-linguistic analysis across nine languages demonstrates severe training imbalances favoring English and Chinese, with hue serving as the primary driver of color naming decisions. Finally, ablation studies reveal that language model architecture significantly influences color naming independent of visual processing capabilities.

Paperid: 4099, https://arxiv.org/pdf/2509.21251.pdf

Abstract:
The field of vision-language understanding has been actively researched in recent years, thanks to the development of Large Language Models~(LLMs). However, it still needs help with problems requiring multi-step reasoning, even for very simple questions. Recent studies adopt LLMs to tackle this problem by iteratively generating sub-questions and answers. However, there are disadvantages such as 1) the fine-grained visual contents of images are not available using LLMs that cannot read visual information, 2) internal mechanisms are inaccessible and difficult to reproduce by using black-box LLMs. To solve these problems, we propose the SQ (Self-Questioning)-InstructBLIP, which improves inference performance by generating image-aware informative sub-questions and sub-answers iteratively. The SQ-InstructBLIP, which consists of a Questioner, Answerer, and Reasoner that share the same architecture. Questioner and Answerer generate sub-questions and sub-answers to help infer the main-question, and Reasoner performs reasoning on the main-question considering the generated sub-question information. Our experiments show that the proposed method SQ-InstructBLIP, which uses the generated sub-questions as additional information when solving the VQA task, performs more accurate reasoning than the previous works.

Paperid: 4100, https://arxiv.org/pdf/2509.17666.pdf

Abstract:
Object insertion tasks are prone to failures under pose uncertainties and environmental variations, traditionally requiring manual finetuning or controller retraining. We present a novel approach for robust and resilient object insertion using a passively compliant soft wrist that enables safe contact absorption through large deformations, without high-frequency control or force sensing. Our method structures insertion as compliance-enabled contact formations, sequential contact states that progressively constrain degrees of freedom, and integrates automated failure recovery strategies. Our key insight is that wrist compliance permits safe, repeated recovery attempts; hence, we refer to it as compliance-enabled failure recovery. We employ a pre-trained vision-language model (VLM) that assesses each skill execution from terminal poses and images, identifies failure modes, and proposes recovery actions by selecting skills and updating goals. In simulation, our method achieved an 83% success rate, recovering from failures induced by randomized conditions--including grasp misalignments up to 5 degrees, hole-pose errors up to 20mm, fivefold increases in friction, and previously unseen square/rectangular pegs--and we further validate the approach on a real robot.

Paperid: 4101, https://arxiv.org/pdf/2509.15432.pdf

Abstract:
Visual Document Retrieval (VDR) typically operates as text-to-image retrieval using specialized bi-encoders trained to directly embed document images. We revisit a zero-shot generate-and-encode pipeline: a vision-language model first produces a detailed textual description of each document image, which is then embedded by a standard text encoder. On the ViDoRe-v2 benchmark, the method reaches 63.4% nDCG@5, surpassing the strongest specialised multi-vector visual document encoder. It also scales better to large collections and offers broader multilingual coverage. Analysis shows that modern vision-language models capture complex textual and visual cues with sufficient granularity to act as a reusable semantic proxy. By offloading modality alignment to pretrained vision-language models, our approach removes the need for computationally intensive text-image contrastive training and establishes a strong zero-shot baseline for future VDR systems.

Paperid: 4102, https://arxiv.org/pdf/2509.14117.pdf

Abstract:
Vision-Language-Action (VLA) models often fail to generalize to novel camera viewpoints, a limitation stemming from their difficulty in inferring robust 3D geometry from 2D images. We introduce GeoAware-VLA, a simple yet effective approach that enhances viewpoint invariance by integrating strong geometric priors into the vision backbone. Instead of training a visual encoder or relying on explicit 3D data, we leverage a frozen, pretrained geometric vision model as a feature extractor. A trainable projection layer then adapts these geometrically-rich features for the policy decoder, relieving it of the burden of learning 3D consistency from scratch. Through extensive evaluations on LIBERO benchmark subsets, we show GeoAware-VLA achieves substantial improvements in zero-shot generalization to novel camera poses, boosting success rates by over 2x in simulation. Crucially, these benefits translate to the physical world; our model shows a significant performance gain on a real robot, especially when evaluated from unseen camera angles. Our approach proves effective across both continuous and discrete action spaces, highlighting that robust geometric grounding is a key component for creating more generalizable robotic agents.

Paperid: 4103, https://arxiv.org/pdf/2509.13031.pdf

Abstract:
Reinforcement learning (RL) has proven highly effective in eliciting the reasoning capabilities of large language models (LLMs). Inspired by this success, recent studies have explored applying similar techniques to vision-language models (VLMs), aiming to enhance their reasoning performance. However, directly transplanting RL methods from LLMs to VLMs is suboptimal, as the tasks faced by VLMs are inherently more complex. Specifically, VLMs must first accurately perceive and understand visual inputs before reasoning can be effectively performed. To address this challenge, we propose a two-stage reinforcement learning framework designed to jointly enhance both the perceptual and reasoning capabilities of VLMs. To mitigate the vanishing advantage issue commonly observed in RL training, we first perform dataset-level sampling to selectively strengthen specific capabilities using distinct data sources. During training, the first stage focuses on improving the model's visual perception through coarse- and fine-grained visual understanding, while the second stage targets the enhancement of reasoning abilities. After the proposed two-stage reinforcement learning process, we obtain PeBR-R1, a vision-language model with significantly enhanced perceptual and reasoning capabilities. Experimental results on seven benchmark datasets demonstrate the effectiveness of our approach and validate the superior performance of PeBR-R1 across diverse visual reasoning tasks.

Paperid: 4104, https://arxiv.org/pdf/2509.11986.pdf

Abstract:
Vision--language models (VLMs) often process visual inputs through a pretrained vision encoder, followed by a projection into the language model's embedding space via a connector component. While crucial for modality fusion, the potential information loss induced by this projection step and its direct impact on model capabilities remain understudied. We introduce two complementary approaches to examine and quantify this loss by analyzing the latent representation space. First, we evaluate semantic information preservation by analyzing changes in k-nearest neighbor relationships between image representations, before and after projection. Second, we directly measure information loss by reconstructing visual embeddings from the projected representation, localizing loss at an image patch level. Experiments reveal that connectors substantially distort the local geometry of visual representations, with k-nearest neighbors diverging by 40--60\% post-projection, correlating with degradation in retrieval performance. The patch-level embedding reconstruction provides interpretable insights for model behavior on visually grounded question-answering tasks, finding that areas of high information loss reliably predict instances where models struggle.

Paperid: 4105, https://arxiv.org/pdf/2509.11417.pdf

Abstract:
Vision-language-action (VLA) models finetuned from vision-language models (VLMs) hold the promise of leveraging rich pretrained representations to build generalist robots across diverse tasks and environments. However, direct fine-tuning on robot data often disrupts these representations and limits generalization. We present a framework that better preserves pretrained features while adapting them for robot manipulation. Our approach introduces three components: (i) a dual-encoder design with one frozen vision encoder to retain pretrained features and another trainable for task adaptation, (ii) a string-based action tokenizer that casts continuous actions into character sequences aligned with the model's pretraining domain, and (iii) a co-training strategy that combines robot demonstrations with vision-language datasets emphasizing spatial reasoning and affordances. Evaluations in simulation and on real robots show that our method improves robustness to visual perturbations, generalization to novel instructions and environments, and overall task success compared to baselines.

Paperid: 4106, https://arxiv.org/pdf/2509.10555.pdf

Abstract:
Vision-language pre-training (VLP) offers unique advantages for surgery by aligning language with surgical videos, enabling workflow understanding and transfer across tasks without relying on expert-labeled datasets. However, progress in surgical VLP remains constrained by the limited scale, procedural diversity, semantic quality, and hierarchical structure of existing datasets. In this work, we present SurgLaVi, the largest and most diverse surgical vision-language dataset to date, comprising nearly 240k clip-caption pairs from more than 200 procedures, and comprising hierarchical levels at phase-, step-, and task-level. At the core of SurgLaVi lies a fully automated pipeline that systematically generates fine-grained transcriptions of surgical videos and segments them into coherent procedural units. To ensure high-quality annotations, it applies dual-modality filtering to remove irrelevant and noisy samples. Within this framework, the resulting captions are enriched with contextual detail, producing annotations that are both semantically rich and easy to interpret. To ensure accessibility, we release SurgLaVi-\b{eta}, an open-source derivative of 113k clip-caption pairs constructed entirely from public data, which is over four times larger than existing surgical VLP datasets. To demonstrate the value of SurgLaVi datasets, we introduce SurgCLIP, a CLIP-style video-text contrastive framework with dual encoders, as a representative base model. SurgCLIP achieves consistent improvements across phase, step, action, and tool recognition, surpassing prior state-of-the-art methods, often by large margins. These results validate that large-scale, semantically rich, and hierarchically structured datasets directly translate into stronger and more generalizable representations, establishing SurgLaVi as a key resource for developing surgical foundation models.

Paperid: 4107, https://arxiv.org/pdf/2509.09013.pdf

Abstract:
Despite strong performance in visual understanding and language-based reasoning, Vision-Language Models (VLMs) struggle with tasks requiring integrated perception and symbolic computation. We study this limitation through visual equation solving, where mathematical equations are embedded in images, variables are represented by object icons, and coefficients must be inferred by counting. While VLMs perform well on textual equations, they fail on visually grounded counterparts. To understand this gap, we decompose the task into coefficient counting and variable recognition, and find that counting is the primary bottleneck, even when recognition is accurate. We also observe that composing recognition and reasoning introduces additional errors, highlighting challenges in multi-step visual reasoning. Finally, as equation complexity increases, symbolic reasoning itself becomes a limiting factor. These findings reveal key weaknesses in current VLMs and point toward future improvements in visually grounded mathematical reasoning.

Paperid: 4108, https://arxiv.org/pdf/2509.05838.pdf

Abstract:
This paper investigates the effectiveness of TikTok's enforcement mechanisms for limiting the exposure of harmful content to youth accounts. We collect over 7000 videos, classify them as harmful vs not-harmful, and then simulate interactions using age-specific sockpuppet accounts through both passive and active engagement strategies. We also evaluate the performance of large language (LLMs) and vision-language models (VLMs) in detecting harmful content, identifying key challenges in precision and scalability. Preliminary results show minimal differences in content exposure between adult and youth accounts, raising concerns about the platform's age-based moderation. These findings suggest that the platform needs to strengthen youth safety measures and improve transparency in content moderation.

Paperid: 4109, https://arxiv.org/pdf/2509.05703.pdf

Abstract:
Transformers and vision-language models (VLMs) have emerged as dominant architectures in computer vision and multimodal AI, offering state-of-the-art performance in tasks such as image classification, object detection, visual question answering, and caption generation. However, their high computational complexity, large memory footprints, and irregular data access patterns present significant challenges for deployment in latency- and power-constrained environments. Field-programmable gate arrays (FPGAs) provide an attractive hardware platform for such workloads due to their reconfigurability, fine-grained parallelism, and potential for energy-efficient acceleration. This paper presents a comprehensive review of design trade-offs, optimization strategies, and implementation challenges for FPGA-based inference of transformers and VLMs. We examine critical factors such as device-class selection, memory subsystem constraints, dataflow orchestration, quantization strategies, sparsity exploitation, and toolchain choices, alongside modality-specific issues unique to VLMs, including heterogeneous compute balancing and cross-attention memory management. Additionally, we discuss emerging trends in hardware-algorithm co-design, highlighting innovations in attention mechanisms, compression, and modular overlays to improve efficiency and adaptability. Practical issues such as runtime flexibility, verification overhead, and the absence of standardized FPGA multimodal benchmarks are also considered. Finally, we outline future directions toward scalable, portable, and reconfigurable FPGA solutions that adapt to evolving model architectures while sustaining high utilization and predictable performance. This synthesis offers both a technical foundation and a forward-looking perspective to help bridge the gap between advanced multimodal AI models and efficient FPGA deployment.

Paperid: 4111, https://arxiv.org/pdf/2509.01167.pdf

Abstract:
Recent advances in multimodal large language models (MLLMs) have led to much progress in video understanding tasks. To avoid the heavy computational cost of processing all frames, these models typically rely on keyframe sampling methods guided by vision-language encoders (\textit{e.g.,} SigLIP). However, it remains unclear whether such encoders can truly identify the most informative frames. In this work, we provide several empirical pieces of evidence revealing that popular vision encoders critically suffer from their limited capability to identify where the MLLM should look inside the video to handle the given textual query appropriately. Our findings suggest that the development of better keyframe identification techniques may be necessary for efficient video MLLMs.

Paperid: 4112, https://arxiv.org/pdf/2508.17714.pdf

Abstract:
Traditional dialogue retrieval aims to select the most appropriate utterance or image from recent dialogue history. However, they often fail to meet users' actual needs for revisiting semantically coherent content scattered across long-form conversations. To fill this gap, we define the Fine-grained Fragment Retrieval (FFR) task, requiring models to locate query-relevant fragments, comprising both utterances and images, from multimodal long-form dialogues. As a foundation for FFR, we construct MLDR, the longest-turn multimodal dialogue retrieval dataset to date, averaging 25.45 turns per dialogue, with each naturally spanning three distinct topics. To evaluate generalization in real-world scenarios, we curate and annotate a WeChat-based test set comprising real-world multimodal dialogues with an average of 75.38 turns. Building on these resources, we explore existing generation-based Vision-Language Models (VLMs) on FFR and observe that they often retrieve incoherent utterance-image fragments. While optimized for generating responses from visual-textual inputs, these models lack explicit supervision to ensure semantic coherence within retrieved fragments. To this end, we propose F2RVLM, a generative retrieval model trained in a two-stage paradigm: (1) supervised fine-tuning to inject fragment-level retrieval knowledge, and (2) GRPO-based reinforcement learning with multi-objective rewards promoting semantic precision, relevance, and contextual coherence. To handle varying intra-fragment complexity, from locally dense to sparsely distributed, we introduce difficulty-aware curriculum sampling that ranks training instances by model-predicted difficulty and gradually exposes the model to harder samples. This boosts reasoning ability in long, multi-turn contexts. F2RVLM outperforms popular VLMs in both in-domain and real-domain settings, demonstrating superior retrieval performance.

Paperid: 4113, https://arxiv.org/pdf/2508.16070.pdf

Abstract:
Approximately 283 million people worldwide live with visual impairments, motivating increasing research into leveraging Visual Language Models (VLMs) to develop effective walking assistance systems for blind and low vision individuals. However, existing VLMs in walking assistant task often have outputs that contain considerable redundancy and extraneous details, adversely affecting users' ability to accurately assess their surroundings. Moreover, these models typically lack the capability to proactively assess environmental risks and adaptively trigger reminders based on the appropriate scene, leading to excessive temporal redundancy. To mitigate output and temporal redundancy, we propose WalkVLM-LR, a walking assistance model with less redundancy. To reduce output redundancy, we introduce four human-preference-based custom reward functions within the GRPO-based reasoning framework to optimize the output in terms of conciseness, fluency, keyword density, and accuracy, thereby producing more informative and streamlined outputs. To minimize temporal redundancy, we incorporate an environment awareness discriminator, which shares the visual encoder with the VLMs to reduce redundant computations and enhance discriminative efficiency, to make WalkVLM-LR assess scene risk levels and minimize unnecessary reminders. Experimental results demonstrate that our method achieves state-of-the-art performance across all evaluation metrics compared with other models, particularly in output conciseness and less temporal redundancy.

Paperid: 4114, https://arxiv.org/pdf/2508.13739.pdf

Abstract:
The growing deployment of Large Vision-Language Models (VLMs) raises safety concerns, as adversaries may exploit model vulnerabilities to induce harmful outputs, with targeted black-box adversarial attacks posing a particularly severe threat. However, existing methods primarily maximize encoder-level global similarity, which lacks the granularity for stealthy and practical fine-grained attacks, where only specific target should be altered (e.g., modifying a car while preserving its background). Moreover, they largely neglect the projector, a key semantic bridge in VLMs for multimodal alignment. To address these limitations, we propose a novel black-box targeted attack framework that leverages the projector. Specifically, we utilize the widely adopted Querying Transformer (Q-Former) which transforms global image embeddings into fine-grained query outputs, to enhance attack effectiveness and granularity. For standard global targeted attack scenarios, we propose the Intermediate Projector Guided Attack (IPGA), which aligns Q-Former fine-grained query outputs with the target to enhance attack strength and exploits the intermediate pretrained Q-Former that is not fine-tuned for any specific Large Language Model (LLM) to improve attack transferability. For fine-grained attack scenarios, we augment IPGA with the Residual Query Alignment (RQA) module, which preserves unrelated content by constraining non-target query outputs to enhance attack granularity. Extensive experiments demonstrate that IPGA significantly outperforms baselines in global targeted attacks, and IPGA with RQA (IPGA-R) attains superior success rates and unrelated content preservation over baselines in fine-grained attacks. Our method also transfers effectively to commercial VLMs such as Google Gemini and OpenAI GPT.

Paperid: 4115, https://arxiv.org/pdf/2508.10367.pdf

Abstract:
Assessing the alignment of multimodal vision-language models~(VLMs) with human perception is essential to understand how they perceive low-level visual features. A key characteristic of human vision is the contrast sensitivity function (CSF), which describes sensitivity to spatial frequency at low-contrasts. Here, we introduce a novel behavioral psychophysics-inspired method to estimate the CSF of chat-based VLMs by directly prompting them to judge pattern visibility at different contrasts for each frequency. This methodology is closer to the real experiments in psychophysics than the previously reported. Using band-pass filtered noise images and a diverse set of prompts, we assess model responses across multiple architectures. We find that while some models approximate human-like CSF shape or magnitude, none fully replicate both. Notably, prompt phrasing has a large effect on the responses, raising concerns about prompt stability. Our results provide a new framework for probing visual sensitivity in multimodal models and reveal key gaps between their visual representations and human perception.

Paperid: 4116, https://arxiv.org/pdf/2508.09814.pdf

Abstract:
During the training of multi-modal models like CLIP, we observed an intriguing phenomenon: the correlation with low-level human image quality assessments peaks in the early epochs before gradually declining. This study investigates this observation and seeks to understand its causes through two key factors: shape-texture bias alignment and classification accuracy drop under noise. Our findings suggest that CLIP initially learn low-level visual features, enhancing its alignment with low-level human perception but also increasing its sensitivity to noise and its texture bias. As training progresses, the model shifts toward more abstract shape-based representations, improving noise robustness but reducing alignment with low-level human perception. These results suggest that these factors shared an underlying learning mechanism and provide new insights into optimizing the trade-off between perceptual alignment and robustness in vision-language models.

Paperid: 4117, https://arxiv.org/pdf/2508.09456.pdf

Abstract:
Vision-language models (VLMs) have shown significant advancements in tasks such as visual grounding, where they localize specific objects in images based on natural language queries and images. However, security issues in visual grounding tasks for VLMs remain underexplored, especially in the context of backdoor attacks. In this paper, we introduce a novel input-aware backdoor attack method, IAG, designed to manipulate the grounding behavior of VLMs. This attack forces the model to ground a specific target object in the input image, regardless of the user's query. We propose an adaptive trigger generator that embeds the semantic information of the attack target's description into the original image using a text-conditional U-Net, thereby overcoming the open-vocabulary attack challenge. To ensure the attack's stealthiness, we utilize a reconstruction loss to minimize visual discrepancies between poisoned and clean images. Additionally, we introduce a unified method for generating attack data. IAG is evaluated theoretically and empirically, demonstrating its feasibility and effectiveness. Notably, our ASR@0.5 on InternVL-2.5-8B reaches over 65\% on various testing sets. IAG also shows promising potential on manipulating Ferret-7B and LlaVA-1.5-7B with very little accuracy decrease on clean samples. Extensive specific experiments, such as ablation study and potential defense, also indicate the robustness and transferability of our attack.

Paperid: 4118, https://arxiv.org/pdf/2508.08570.pdf

Abstract:
To enhance group robustness to spurious correlations, prior work often relies on auxiliary annotations for groups or spurious features and assumes identical sets of groups across source and target domains. These two requirements are both unnatural and impractical in real-world settings. To overcome these limitations, we propose a method that leverages the semantic structure inherent in class labels--specifically, superclass information--to naturally reduce reliance on spurious features. Our model employs gradient-based attention guided by a pre-trained vision-language model to disentangle superclass-relevant and irrelevant features. Then, by promoting the use of all superclass-relevant features for prediction, our approach achieves robustness to more complex spurious correlations without the need to annotate any source samples. Experiments across diverse datasets demonstrate that our method significantly outperforms baselines in domain generalization tasks, with clear improvements in both quantitative metrics and qualitative visualizations.

Paperid: 4119, https://arxiv.org/pdf/2508.06801.pdf

Abstract:
Pedestrian gestures play an important role in traffic communication, particularly in interactions with autonomous vehicles (AVs), yet their subtle, ambiguous, and context-dependent nature poses persistent challenges for machine interpretation. This study investigates these challenges by using GPT-4V, a vision-language model, not as a performance benchmark but as a diagnostic tool to reveal patterns and causes of gesture misrecognition. We analysed a public dataset of pedestrian-vehicle interactions, combining manual video review with thematic analysis of the model's qualitative reasoning. This dual approach surfaced recurring factors influencing misrecognition, including gesture visibility, pedestrian behaviour, interaction context, and environmental conditions. The findings suggest practical considerations for gesture design, including the value of salience and contextual redundancy, and highlight opportunities to improve AV recognition systems through richer context modelling and uncertainty-aware interpretations. While centred on AV-pedestrian interaction, the method and insights are applicable to other domains where machines interpret human gestures, such as wearable AR and assistive technologies.

Paperid: 4120, https://arxiv.org/pdf/2508.04848.pdf

Abstract:
Reinforcement learning (RL) has become a key technique for enhancing the reasoning abilities of large language models (LLMs), with policy-gradient algorithms dominating the post-training stage because of their efficiency and effectiveness. However, most existing benchmarks evaluate large-language-model reasoning under idealized settings, overlooking performance in realistic, non-ideal scenarios. We identify three representative non-ideal scenarios with practical relevance: summary inference, fine-grained noise suppression, and contextual filtering. We introduce a new research direction guided by brain-science findings that human reasoning remains reliable under imperfect inputs. We formally define and evaluate these challenging scenarios. We fine-tune three LLMs and a state-of-the-art large vision-language model (LVLM) using RL with a representative policy-gradient algorithm and then test their performance on eight public datasets. Our results reveal that while RL fine-tuning improves baseline reasoning under idealized settings, performance declines significantly across all three non-ideal scenarios, exposing critical limitations in advanced reasoning capabilities. Although we propose a scenario-specific remediation method, our results suggest current methods leave these reasoning deficits largely unresolved. This work highlights that the reasoning abilities of large models are often overstated and underscores the importance of evaluating models under non-ideal scenarios. The code and data will be released at XXXX.

Paperid: 4121, https://arxiv.org/pdf/2508.04280.pdf

Abstract:
Interactive multimodal agents must convert raw visual observations into coherent sequences of language-conditioned actions -- a capability that current vision-language models (VLMs) still lack. Earlier reinforcement-learning (RL) efforts could, in principle, endow VLMs with such skills, but they have seldom tested whether the learned behaviours generalize beyond their training simulators, and they depend either on brittle hyperparameter tuning or on dense-reward environments with low state variability. We introduce Vision-Language Decoupled Actor-Critic (VL-DAC), a lightweight, hyperparameter-free RL algorithm. VL-DAC applies PPO updates to action tokens while learning value only at the environment-step level: an arrangement, to our knowledge, not previously explored for large VLMs or LLMs. This simple decoupling removes unstable weighting terms and yields faster, more reliable convergence. Training a single VLM with VL-DAC in one inexpensive simulator at a time (MiniWorld, Gym-Cards, ALFWorld, or WebShop) already produces policies that generalize widely: +50\% relative on BALROG (game-centric agentic control), +5\% relative on the hardest part of VSI-Bench (spatial planning), and +2\% on VisualWebBench (web navigation), all without degrading general image understanding accuracy. These results provide the first evidence that a simple RL algorithm can train VLMs entirely in cheap synthetic worlds while delivering measurable gains on real-image agentic, spatial-reasoning, and web-navigation benchmarks.

Paperid: 4122, https://arxiv.org/pdf/2508.02807.pdf

Abstract:
Video virtual try-on (VVT) technology has garnered considerable academic interest owing to its promising applications in e-commerce advertising and entertainment. However, most existing end-to-end methods rely heavily on scarce paired garment-centric datasets and fail to effectively leverage priors of advanced visual models and test-time inputs, making it challenging to accurately preserve fine-grained garment details and maintain temporal consistency in unconstrained scenarios. To address these challenges, we propose DreamVVT, a carefully designed two-stage framework built upon Diffusion Transformers (DiTs), which is inherently capable of leveraging diverse unpaired human-centric data to enhance adaptability in real-world scenarios. To further leverage prior knowledge from pretrained models and test-time inputs, in the first stage, we sample representative frames from the input video and utilize a multi-frame try-on model integrated with a vision-language model (VLM), to synthesize high-fidelity and semantically consistent keyframe try-on images. These images serve as complementary appearance guidance for subsequent video generation. \textbf{In the second stage}, skeleton maps together with fine-grained motion and appearance descriptions are extracted from the input content, and these along with the keyframe try-on images are then fed into a pretrained video generation model enhanced with LoRA adapters. This ensures long-term temporal coherence for unseen regions and enables highly plausible dynamic motions. Extensive quantitative and qualitative experiments demonstrate that DreamVVT surpasses existing methods in preserving detailed garment content and temporal stability in real-world scenarios. Our project page https://virtu-lab.github.io/

Paperid: 4123, https://arxiv.org/pdf/2508.01109.pdf

Abstract:
We investigate whether socio-economic indicators like household wealth leave recoverable imprints in satellite imagery (capturing physical features) and Internet-sourced text (reflecting historical/economic narratives). Using Demographic and Health Survey (DHS) data from African neighborhoods, we pair Landsat images with LLM-generated textual descriptions conditioned on location/year and text retrieved by an AI search agent from web sources. We develop a multimodal framework predicting household wealth (International Wealth Index) through five pipelines: (i) vision model on satellite images, (ii) LLM using only location/year, (iii) AI agent searching/synthesizing web text, (iv) joint image-text encoder, (v) ensemble of all signals. Our framework yields three contributions. First, fusing vision and agent/LLM text outperforms vision-only baselines in wealth prediction (e.g., R-squared of 0.77 vs. 0.63 on out-of-sample splits), with LLM-internal knowledge proving more effective than agent-retrieved text, improving robustness to out-of-country and out-of-time generalization. Second, we find partial representational convergence: fused embeddings from vision/language modalities correlate moderately (median cosine similarity of 0.60 after alignment), suggesting a shared latent code of material well-being while retaining complementary details, consistent with the Platonic Representation Hypothesis. Although LLM-only text outperforms agent-retrieved data, challenging our Agent-Induced Novelty Hypothesis, modest gains from combining agent data in some splits weakly support the notion that agent-gathered information introduces unique representational structures not fully captured by static LLM knowledge. Third, we release a large-scale multimodal dataset comprising more than 60,000 DHS clusters linked to satellite images, LLM-generated descriptions, and agent-retrieved texts.

Paperid: 4124, https://arxiv.org/pdf/2507.10449.pdf

Abstract:
Coral reefs are vital yet vulnerable ecosystems that require continuous monitoring to support conservation. While coral reef images provide essential information in coral monitoring, interpreting such images remains challenging due to the need for domain expertise. Visual Question Answering (VQA), powered by Large Vision-Language Models (LVLMs), has great potential in user-friendly interaction with coral reef images. However, applying VQA to coral imagery demands a dedicated dataset that addresses two key challenges: domain-specific annotations and multidimensional questions. In this work, we introduce CoralVQA, the first large-scale VQA dataset for coral reef analysis. It contains 12,805 real-world coral images from 67 coral genera collected from 3 oceans, along with 277,653 question-answer pairs that comprehensively assess ecological and health-related conditions. To construct this dataset, we develop a semi-automatic data construction pipeline in collaboration with marine biologists to ensure both scalability and professional-grade data quality. CoralVQA presents novel challenges and provides a comprehensive benchmark for studying vision-language reasoning in the context of coral reef images. By evaluating several state-of-the-art LVLMs, we reveal key limitations and opportunities. These insights form a foundation for future LVLM development, with a particular emphasis on supporting coral conservation efforts.

Paperid: 4125, https://arxiv.org/pdf/2507.10355.pdf

Abstract:
Textual adapter-based tuning methods have shown significant potential in transferring knowledge from pre-trained Vision-Language Models (VLMs) to downstream tasks. Existing works generally employ the deterministic textual feature adapter to refine each category textual representation. However, due to inherent factors such as different attributes and contexts, there exists significant diversity in textual descriptions for each category. Such description diversity offers rich discriminative semantic knowledge that can benefit downstream visual learning tasks. Obviously, traditional deterministic adapter model cannot adequately capture this varied semantic information. Also, it is desirable to exploit the inter-class relationships in VLM adapter. To address these issues, we propose to exploit random graph model into VLM adapter and develop a novel Vertex Random Graph Adapter (VRGAdapter). VRGAdapter first models the inherent diverse descriptions of each category and inter-class relationships of different categories simultaneously by leveraging a Vertex Random Knowledge Graph (VRKG) model. Then, it employs probabilistic message propagation on VRKG to learn context-aware distribution representation for each class node. Finally, it adopts a reparameterized sampling function to achieve textual adapter learning. Note that, VRGAdapter provides a more general adapter solution that encompasses traditional graph-based adapter as a special case. In addition, to enable more robust performance for downstream tasks, we also introduce a new Uncertainty-guided Multi-branch Fusion (UMF) scheme that dynamically integrates multiple pre-trained models for ensemble prediction. Extensive experiments on multiple benchmark datasets demonstrate the effectiveness of our approach.

Paperid: 4126, https://arxiv.org/pdf/2507.07709.pdf

Abstract:
Unified vision-language models(VLMs) have recently shown remarkable progress, enabling a single model to flexibly address diverse tasks through different instructions within a shared computational architecture. This instruction-based control mechanism creates unique security challenges, as adversarial inputs must remain effective across multiple task instructions that may be unpredictably applied to process the same malicious content. In this paper, we introduce CrossVLAD, a new benchmark dataset carefully curated from MSCOCO with GPT-4-assisted annotations for systematically evaluating cross-task adversarial attacks on unified VLMs. CrossVLAD centers on the object-change objective-consistently manipulating a target object's classification across four downstream tasks-and proposes a novel success rate metric that measures simultaneous misclassification across all tasks, providing a rigorous evaluation of adversarial transferability. To tackle this challenge, we present CRAFT (Cross-task Region-based Attack Framework with Token-alignment), an efficient region-centric attack method. Extensive experiments on Florence-2 and other popular unified VLMs demonstrate that our method outperforms existing approaches in both overall cross-task attack performance and targeted object-change success rates, highlighting its effectiveness in adversarially influencing unified VLMs across diverse tasks.

Paperid: 4127, https://arxiv.org/pdf/2507.05677.pdf

Abstract:
Prompt learning methods have significantly extended the transferability of pre-trained Vision-Language Models (VLMs) like CLIP for various downstream tasks. These methods adopt handcraft templates or learnable vectors to provide text or image instructions in fine-tuning VLMs. However, most existing works ignore the structural relationships between learnable prompts and tokens within and between modalities. Moreover, balancing the performance of base and new classes remains a significant challenge. In this paper, we propose an Integrated Structural Prompt (ISP) for VLMs to enhance the interaction of information representations between the text and image branches. ISP introduces self-structural and cross-structural prompt modules to model the structural relationships between learnable prompts and frozen tokens within and across modalities. This enables efficient information transfer while preserving feature stability. Additionally, we propose a sample probing module that dynamically adjusts loss coefficients based on sample difficulty, preventing the mode from overfitting to simple samples and improving generalization ability to new classes. Extensive experiments on three widely used settings: base-to-new generalization, cross-dataset evaluation, and domain generalization demonstrate that the proposed ISP achieves competitive performance against state-of-the-art methods.

Paperid: 4128, https://arxiv.org/pdf/2507.05668.pdf

Abstract:
Pre-trained large vision-language models (VLMs) like CLIP demonstrate impressive generalization ability. Existing prompt-based and adapter-based works have made significant progress in fine-tuning VLMs but still face the challenges of maintaining strong generalization abilities, particularly towards unseen new classes. This limitation partly arises from these methods treating all tokens of the image and text encoder equally, which can lead to overfitting on less informative features (e.g., background noise, template words) and degrade the general representations that are crucial for novel concept recognition. To address this issue, we propose Dynamic Rank Adaptation (DRA), a novel adapter variant method, designed specifically to enhance new class generalization. DRA dynamically allocates adaptation ranks based on the importance of features during training to preserve general knowledge. DRA first employs token importance grouping, using sequence attention to evaluate and group tokens by their importance. Then, we adopt rank adaptation according to the importance of each token group dynamically by assigning higher feature ranks to the more important tokens. Also, we design a new channel response mechanism to prioritize the preservation and adaptation of feature channels identified as the most informative for each instance. In addition, a L1 regularization term is introduced to stabilize the training. Extensive experiments demonstrate the effectiveness and superiority of our proposed DRA over existing works, especially on enhancing the performance of new classes on various benchmarks, including base-new classes, cross-datasets evaluation and domain generalization. The source code will be published after the paper is received.

Paperid: 4129, https://arxiv.org/pdf/2507.03275.pdf

Abstract:
Current text-to-image (T2I) benchmarks evaluate models on rigid prompts, potentially underestimating true generative capabilities due to prompt sensitivity and creating biases that favor certain models while disadvantaging others. We introduce ConceptMix++, a framework that disentangles prompt phrasing from visual generation capabilities by applying iterative prompt optimization. Building on ConceptMix, our approach incorporates a multimodal optimization pipeline that leverages vision-language model feedback to refine prompts systematically. Through extensive experiments across multiple diffusion models, we show that optimized prompts significantly improve compositional generation performance, revealing previously hidden model capabilities and enabling fairer comparisons across T2I models. Our analysis reveals that certain visual concepts -- such as spatial relationships and shapes -- benefit more from optimization than others, suggesting that existing benchmarks systematically underestimate model performance in these categories. Additionally, we find strong cross-model transferability of optimized prompts, indicating shared preferences for effective prompt phrasing across models. These findings demonstrate that rigid benchmarking approaches may significantly underrepresent true model capabilities, while our framework provides more accurate assessment and insights for future development.

Paperid: 4130, https://arxiv.org/pdf/2507.00506.pdf

Abstract:
Recent advancements in adapting vision-language pre-training models like CLIP for person re-identification (ReID) tasks often rely on complex adapter design or modality-specific tuning while neglecting cross-modal interaction, leading to high computational costs or suboptimal alignment. To address these limitations, we propose a simple yet effective framework named Selective Cross-modal Prompt Tuning (SCING) that enhances cross-modal alignment and robustness against real-world perturbations. Our method introduces two key innovations: Firstly, we proposed Selective Visual Prompt Fusion (SVIP), a lightweight module that dynamically injects discriminative visual features into text prompts via a cross-modal gating mechanism. Moreover, the proposed Perturbation-Driven Consistency Alignment (PDCA) is a dual-path training strategy that enforces invariant feature alignment under random image perturbations by regularizing consistency between original and augmented cross-modal embeddings. Extensive experiments are conducted on several popular benchmarks covering Market1501, DukeMTMC-ReID, Occluded-Duke, Occluded-REID, and P-DukeMTMC, which demonstrate the impressive performance of the proposed method. Notably, our framework eliminates heavy adapters while maintaining efficient inference, achieving an optimal trade-off between performance and computational overhead. The code will be released upon acceptance.

Paperid: 4131, https://arxiv.org/pdf/2506.23982.pdf

Abstract:
Personalization, while extensively studied in conventional autonomous driving pipelines, has been largely overlooked in the context of end-to-end autonomous driving (E2EAD), despite its critical role in fostering user trust, safety perception, and real-world adoption. A primary bottleneck is the absence of large-scale real-world datasets that systematically capture driving preferences, severely limiting the development and evaluation of personalized E2EAD models. In this work, we introduce the first large-scale real-world dataset explicitly curated for personalized E2EAD, integrating comprehensive scene topology with rich dynamic context derived from agent dynamics and semantics inferred via a fine-tuned vision-language model (VLM). We propose a hybrid annotation pipeline that combines behavioral analysis, rule-and-distribution-based heuristics, and subjective semantic modeling guided by VLM reasoning, with final refinement through human-in-the-loop verification. Building upon this dataset, we introduce the first standardized benchmark for systematically evaluating personalized E2EAD models. Empirical evaluations on state-of-the-art architectures demonstrate that incorporating personalized driving preferences significantly improves behavioral alignment with human demonstrations.

Paperid: 4132, https://arxiv.org/pdf/2506.23725.pdf

Abstract:
Vision-Language Models (VLMs) are increasingly pivotal for generalist robot manipulation, enabling tasks such as physical reasoning, policy generation, and failure detection. However, their proficiency in these high-level applications often assumes a deep understanding of low-level physical prerequisites, a capability that remains largely unverified. For robots to perform actions reliably, they must comprehend intrinsic object properties (e.g., material, weight), action affordances (e.g., graspable, stackable), and physical constraints (e.g., stability, reachability, or an object's state, such as being closed). Despite the widespread use of VLMs in manipulation tasks, we argue that off-the-shelf models may lack this granular, physically grounded understanding, as such prerequisites are often overlooked during training. To address this critical gap, we introduce PAC Bench, a comprehensive benchmark designed to systematically evaluate VLMs on their understanding of core Properties, Affordances, and Constraints (PAC) from a task executability perspective. PAC Bench features a diverse dataset with over 30,000 annotations, comprising 673 real-world images (115 object classes, 15 property types, and 1 to 3 affordances defined per class), 100 real-world humanoid-view scenarios, and 120 unique simulated constraint scenarios across four tasks. Our evaluations reveal significant gaps in the ability of current VLMs to grasp fundamental physical concepts, highlighting limitations in their suitability for reliable robot manipulation and pointing to key areas for targeted research. PAC Bench also serves as a standardized benchmark for rigorously evaluating physical reasoning in VLMs and guiding the development of more robust, physically grounded models for robotic applications. Project Page: https://pacbench.github.io/

Paperid: 4133, https://arxiv.org/pdf/2506.21876.pdf

Abstract:
Internal world models (WMs) enable agents to understand the world's state and predict transitions, serving as the basis for advanced deliberative reasoning. Recent large Vision-Language Models (VLMs), such as OpenAI o3, GPT-4o and Gemini, exhibit potential as general-purpose WMs. While the latest studies have evaluated and shown limitations in specific capabilities such as visual understanding, a systematic evaluation of VLMs' fundamental WM abilities remains absent. Drawing on comparative psychology and cognitive science, we propose a two-stage framework that assesses Perception (visual, spatial, temporal, quantitative, and motion) and Prediction (mechanistic simulation, transitive inference, compositional inference) to provide an atomic evaluation of VLMs as WMs. Guided by this framework, we introduce WM-ABench, a large-scale benchmark comprising 23 fine-grained evaluation dimensions across 6 diverse simulated environments with controlled counterfactual simulations. Through 660 experiments on 15 latest commercial and open-source VLMs, we find that these models exhibit striking limitations in basic world modeling abilities. For instance, almost all models perform at near-random accuracy when distinguishing motion trajectories. Additionally, they lack disentangled understanding -- e.g., some models tend to believe blue objects move faster than green ones. More rich results and analyses reveal significant gaps between VLMs and human-level world modeling.

Paperid: 4134, https://arxiv.org/pdf/2506.18140.pdf

Abstract:
Medical imaging diagnosis presents inherent challenges due to diseases that mimic normal anatomy and exhibit significant inter-patient variability. Clinicians routinely employ comparative reasoning-using reference images from healthy controls or previous patient examinations-to discern subtle yet diagnostically critical abnormalities. However, existing medical vision-language models (VLMs) focus primarily on single-image or single-series analyses and lack explicit mechanisms for comparative reasoning. Conversely, general-purpose VLMs demonstrate strong multi-image comparative reasoning capabilities but lack essential medical-domain knowledge to identify nuanced clinical differences. This work aims to bridge this gap by exploring clinically-inspired comparative analysis within VLMs, leveraging reference images to enhance diagnostic accuracy. Through extensive empirical analysis, we show that providing general-purpose VLMs with query and normative matched reference images, accompanied by clinically-informed comparative prompts, significantly improves diagnostic outcomes compared to single-image baselines, especially after supervised finetuning (SFT). Our contributions highlight the clinical relevance of comparative analysis introduce novel strategies for leveraging reference images in VLMs, empirically demonstrate enhanced performance across multiple medical visual question answering (VQA) tasks, and provide theoretical insights into the efficacy of comparative image analysis in medical diagnosis.

Paperid: 4135, https://arxiv.org/pdf/2506.17645.pdf

Abstract:
Automating medical report generation from histopathology images is a critical challenge requiring effective visual representations and domain-specific knowledge. Inspired by the common practices of human experts, we propose an in-context learning framework called PathGenIC that integrates context derived from the training set with a multimodal in-context learning (ICL) mechanism. Our method dynamically retrieves semantically similar whole slide image (WSI)-report pairs and incorporates adaptive feedback to enhance contextual relevance and generation quality. Evaluated on the HistGen benchmark, the framework achieves state-of-the-art results, with significant improvements across BLEU, METEOR, and ROUGE-L metrics, and demonstrates robustness across diverse report lengths and disease categories. By maximizing training data utility and bridging vision and language with ICL, our work offers a solution for AI-driven histopathology reporting, setting a strong foundation for future advancements in multimodal clinical applications.

Paperid: 4136, https://arxiv.org/pdf/2506.17337.pdf

Abstract:
Vision Language Models (VLMs) have shown promise in automating image diagnosis and interpretation in clinical settings. However, developing specialist medical VLMs requires substantial computational resources and carefully curated datasets, and it remains unclear under which conditions generalist and specialist medical VLMs each perform best. This study highlights the complementary strengths of specialist medical and generalist VLMs. Specialists remain valuable in modality-aligned use cases, but we find that efficiently fine-tuned generalist VLMs can achieve comparable or even superior performance in most tasks, particularly when transferring to unseen or rare OOD medical modalities. These results suggest that generalist VLMs, rather than being constrained by their lack of specialist medical pretraining, may offer a scalable and cost-effective pathway for advancing clinical AI development.

Paperid: 4137, https://arxiv.org/pdf/2506.10342.pdf

Abstract:
Urban cultures and architectural styles vary significantly across cities due to geographical, chronological, historical, and socio-political factors. Understanding these differences is essential for anticipating how cities may evolve in the future. As representative cases of historical continuity and modern innovation in China, Beijing and Shenzhen offer valuable perspectives for exploring the transformation of urban streetscapes. However, conventional approaches to urban cultural studies often rely on expert interpretation and historical documentation, which are difficult to standardize across different contexts. To address this, we propose a multimodal research framework based on vision-language models, enabling automated and scalable analysis of urban streetscape style differences. This approach enhances the objectivity and data-driven nature of urban form research. The contributions of this study are as follows: First, we construct UrbanDiffBench, a curated dataset of urban streetscapes containing architectural images from different periods and regions. Second, we develop UrbanSense, the first vision-language-model-based framework for urban streetscape analysis, enabling the quantitative generation and comparison of urban style representations. Third, experimental results show that Over 80% of generated descriptions pass the t-test (p less than 0.05). High Phi scores (0.912 for cities, 0.833 for periods) from subjective evaluations confirm the method's ability to capture subtle stylistic differences. These results highlight the method's potential to quantify and interpret urban style evolution, offering a scientifically grounded lens for future design.

Paperid: 4138, https://arxiv.org/pdf/2506.09883.pdf

Abstract:
Vision-Language Models (VLMs) have shown remarkable performance on diverse visual and linguistic tasks, yet they remain fundamentally limited in their understanding of 3D spatial structures. We propose Geometric Distillation, a lightweight, annotation-free fine-tuning framework that injects human-inspired geometric cues into pretrained VLMs without modifying their architecture. By distilling (1) sparse correspondences, (2) relative depth relations, and (3) dense cost volumes from off-the-shelf 3D foundation models (e.g., MASt3R, VGGT), our method shapes representations to be geometry-aware while remaining compatible with natural image-text inputs. Through extensive evaluations on 3D vision-language reasoning and 3D perception benchmarks, our method consistently outperforms prior approaches, achieving improved 3D spatial reasoning with significantly lower computational cost. Our work demonstrates a scalable and efficient path to bridge 2D-trained VLMs with 3D understanding, opening up wider use in spatially grounded multimodal tasks.

Paperid: 4139, https://arxiv.org/pdf/2506.08227.pdf

Abstract:
We investigate 17 benchmarks (e.g. SugarCREPE, VALSE) commonly used for measuring compositional understanding capabilities of vision-language models (VLMs). We scrutinize design choices in their construction, including data source (e.g. MS-COCO) and curation procedures (e.g. constructing negative images/captions), uncovering several inherent biases across most benchmarks. We find that blind heuristics (e.g. token-length, log-likelihood under a language model) perform on par with CLIP models, indicating that these benchmarks do not effectively measure compositional understanding. We demonstrate that the underlying factor is a distribution asymmetry between positive and negative images/captions, induced by the benchmark construction procedures. To mitigate these issues, we provide a few key recommendations for constructing more robust vision-language compositional understanding benchmarks, that would be less prone to such simple attacks.

Paperid: 4140, https://arxiv.org/pdf/2506.07739.pdf

Abstract:
Architectural cultures across regions are characterized by stylistic diversity, shaped by historical, social, and technological contexts in addition to geograph-ical conditions. Understanding architectural styles requires the ability to describe and analyze the stylistic features of different architects from various regions through visual observations of architectural imagery. However, traditional studies of architectural culture have largely relied on subjective expert interpretations and historical literature reviews, often suffering from regional biases and limited ex-planatory scope. To address these challenges, this study proposes three core contributions: (1) We construct a professional architectural style dataset named ArchDiffBench, which comprises 1,765 high-quality architectural images and their corresponding style annotations, collected from different regions and historical periods. (2) We propose ArchiLense, an analytical framework grounded in Vision-Language Models and constructed using the ArchDiffBench dataset. By integrating ad-vanced computer vision techniques, deep learning, and machine learning algo-rithms, ArchiLense enables automatic recognition, comparison, and precise classi-fication of architectural imagery, producing descriptive language outputs that ar-ticulate stylistic differences. (3) Extensive evaluations show that ArchiLense achieves strong performance in architectural style recognition, with a 92.4% con-sistency rate with expert annotations and 84.5% classification accuracy, effec-tively capturing stylistic distinctions across images. The proposed approach transcends the subjectivity inherent in traditional analyses and offers a more objective and accurate perspective for comparative studies of architectural culture.

Paperid: 4141, https://arxiv.org/pdf/2506.04585.pdf

Abstract:
Assessing scientific claims requires identifying, extracting, and reasoning with multimodal data expressed in information-rich figures in scientific literature. Despite the large body of work in scientific QA, figure captioning, and other multimodal reasoning tasks over chart-based data, there are no readily usable multimodal benchmarks that directly test claim verification abilities. To remedy this gap, we introduce a new benchmark MuSciClaims accompanied by diagnostics tasks. We automatically extract supported claims from scientific articles, which we manually perturb to produce contradicted claims. The perturbations are designed to test for a specific set of claim verification capabilities. We also introduce a suite of diagnostic tasks that help understand model failures. Our results show most vision-language models are poor (~0.3-0.5 F1), with even the best model only achieving 0.72 F1. They are also biased towards judging claims as supported, likely misunderstanding nuanced perturbations within the claims. Our diagnostics show models are bad at localizing correct evidence within figures, struggle with aggregating information across modalities, and often fail to understand basic components of the figure.

Paperid: 4142, https://arxiv.org/pdf/2505.22429.pdf

Abstract:
3D Visual Grounding (3DVG) seeks to locate target objects in 3D scenes using natural language descriptions, enabling downstream applications such as augmented reality and robotics. Existing approaches typically rely on labeled 3D data and predefined categories, limiting scalability to open-world settings. We present SeeGround, a zero-shot 3DVG framework that leverages 2D Vision-Language Models (VLMs) to bypass the need for 3D-specific training. To bridge the modality gap, we introduce a hybrid input format that pairs query-aligned rendered views with spatially enriched textual descriptions. Our framework incorporates two core components: a Perspective Adaptation Module that dynamically selects optimal viewpoints based on the query, and a Fusion Alignment Module that integrates visual and spatial signals to enhance localization precision. Extensive evaluations on ScanRefer and Nr3D confirm that SeeGround achieves substantial improvements over existing zero-shot baselines -- outperforming them by 7.7% and 7.1%, respectively -- and even rivals fully supervised alternatives, demonstrating strong generalization under challenging conditions.

Paperid: 4143, https://arxiv.org/pdf/2505.17295.pdf

Abstract:
We introduce ScanBot, a novel dataset designed for instruction-conditioned, high-precision surface scanning in robotic systems. In contrast to existing robot learning datasets that focus on coarse tasks such as grasping, navigation, or dialogue, ScanBot targets the high-precision demands of industrial laser scanning, where sub-millimeter path continuity and parameter stability are critical. The dataset covers laser scanning trajectories executed by a robot across 12 diverse objects and 6 task types, including full-surface scans, geometry-focused regions, spatially referenced parts, functionally relevant structures, defect inspection, and comparative analysis. Each scan is guided by natural language instructions and paired with synchronized RGB, depth, and laser profiles, as well as robot pose and joint states. Despite recent progress, existing vision-language action (VLA) models still fail to generate stable scanning trajectories under fine-grained instructions and real-world precision demands. To investigate this limitation, we benchmark a range of multimodal large language models (MLLMs) across the full perception-planning-execution loop, revealing persistent challenges in instruction-following under realistic constraints.

Paperid: 4144, https://arxiv.org/pdf/2505.06685.pdf

Abstract:
Accurate emotion understanding in videos necessitates effectively recognizing and interpreting emotional states by integrating visual, textual, auditory, and contextual cues. Although recent Large Multimodal Models (LMMs) have exhibited significant progress in general vision-language (VL) tasks, their performance often deteriorates in emotion-specific scenarios, exhibiting catastrophic forgetting when fine-tuned on emotion-centric tasks. To overcome these limitations, we propose Emotion-Qwen, a unified multimodal framework designed to simultaneously enable robust emotion understanding and preserve general VL reasoning capabilities. Emotion-Qwen introduces a novel Hybrid Compressor based on a Mixture-of-Experts (MoE) architecture, dynamically routing inputs to optimally balance emotion-specific processing and general multimodal reasoning. We further propose a carefully structured three-stage pre-training pipeline, leveraging extensive general and emotion-focused datasets to strengthen multimodal representation robustness and model adaptability. Additionally, we develop the Video Emotion Reasoning (VER) dataset, a large-scale bilingual resource containing over 40K video clips annotated with detailed context-aware emotional descriptions, significantly facilitating research on fine-grained emotional reasoning. Extensive experiments confirm that Emotion-Qwen achieves state-of-the-art performance across multiple emotion recognition and reasoning benchmarks, while maintaining highly competitive results in general VL tasks.

Paperid: 4145, https://arxiv.org/pdf/2504.21559.pdf

Abstract:
Large Vision Language Models (LVLMs) often suffer from object hallucination, which undermines their reliability. Surprisingly, we find that simple object-based visual prompting -- overlaying visual cues (e.g., bounding box, circle) on images -- can significantly mitigate such hallucination; however, different visual prompts (VPs) vary in effectiveness. To address this, we propose Black-Box Visual Prompt Engineering (BBVPE), a framework to identify optimal VPs that enhance LVLM responses without needing access to model internals. Our approach employs a pool of candidate VPs and trains a router model to dynamically select the most effective VP for a given input image. This black-box approach is model-agnostic, making it applicable to both open-source and proprietary LVLMs. Evaluations on benchmarks such as POPE and CHAIR demonstrate that BBVPE effectively reduces object hallucination.

Paperid: 4146, https://arxiv.org/pdf/2504.11930.pdf

Abstract:
Fine-tuning vision-language models (VLMs) with large amounts of unlabeled data has recently garnered significant interest. However, a key challenge remains the lack of high-quality pseudo-labeled data. Current pseudo-labeling strategies often struggle with mismatches between semantic and visual information, leading to sub-optimal performance of unsupervised prompt learning (UPL) methods. In this paper, we introduce a simple yet effective approach called \textbf{A}ugmenting D\textbf{i}scriminative \textbf{R}ichness via Diffusions (AiR), toward learning a richer discriminating way to represent the class comprehensively and thus facilitate classification. Specifically, our approach includes a pseudo-label generation module that leverages high-fidelity synthetic samples to create an auxiliary classifier, which captures richer visual variation, bridging text-image-pair classification to a more robust image-image-pair classification. Additionally, we exploit the diversity of diffusion-based synthetic samples to enhance prompt learning, providing greater information for semantic-visual alignment. Extensive experiments on five public benchmarks, including RESISC45 and Flowers102, and across three learning paradigms-UL, SSL, and TRZSL-demonstrate that AiR achieves substantial and consistent performance improvements over state-of-the-art unsupervised prompt learning methods.

Paperid: 4147, https://arxiv.org/pdf/2504.10011.pdf

Abstract:
Dynamic Movement Primitives (DMPs) provide a flexible framework wherein smooth robotic motions are encoded into modular parameters. However, they face challenges in integrating multimodal inputs commonly used in robotics like vision and language into their framework. To fully maximize DMPs' potential, enabling them to handle multimodal inputs is essential. In addition, we also aim to extend DMPs' capability to handle object-focused tasks requiring one-shot complex motion generation, as observation occlusion could easily happen mid-execution in such tasks (e.g., knife occlusion in cake icing, hand occlusion in dough kneading, etc.). A promising approach is to leverage Vision-Language Models (VLMs), which process multimodal data and can grasp high-level concepts. However, they typically lack enough knowledge and capabilities to directly infer low-level motion details and instead only serve as a bridge between high-level instructions and low-level control. To address this limitation, we propose Keyword Labeled Primitive Selection and Keypoint Pairs Generation Guided Movement Primitives (KeyMPs), a framework that combines VLMs with sequencing of DMPs. KeyMPs use VLMs' high-level reasoning capability to select a reference primitive through \emph{keyword labeled primitive selection} and VLMs' spatial awareness to generate spatial scaling parameters used for sequencing DMPs by generalizing the overall motion through \emph{keypoint pairs generation}, which together enable one-shot vision-language guided motion generation that aligns with the intent expressed in the multimodal input. We validate our approach through experiments on two occlusion-rich tasks: object cutting, conducted in both simulated and real-world environments, and cake icing, performed in simulation. These evaluations demonstrate superior performance over other DMP-based methods that integrate VLM support.

Paperid: 4148, https://arxiv.org/pdf/2504.07556.pdf

Abstract:
While text-to-image (T2I) generation models have achieved remarkable progress in recent years, existing evaluation methodologies for vision-language alignment still struggle with the fine-grained semantic matching. Current approaches based on global similarity metrics often overlook critical token-level correspondences between textual descriptions and visual content. To this end, we present TokenFocus-VQA, a novel evaluation framework that leverages Large Vision-Language Models (LVLMs) through visual question answering (VQA) paradigm with position-specific probability optimization. Our key innovation lies in designing a token-aware loss function that selectively focuses on probability distributions at pre-defined vocabulary positions corresponding to crucial semantic elements, enabling precise measurement of fine-grained semantical alignment. The proposed framework further integrates ensemble learning techniques to aggregate multi-perspective assessments from diverse LVLMs architectures, thereby achieving further performance enhancement. Evaluated on the NTIRE 2025 T2I Quality Assessment Challenge Track 1, our TokenFocus-VQA ranks 2nd place (0.8445, only 0.0001 lower than the 1st method) on public evaluation and 2nd place (0.8426) on the official private test set, demonstrating superiority in capturing nuanced text-image correspondences compared to conventional evaluation methods.

Paperid: 4149, https://arxiv.org/pdf/2504.04858.pdf

Abstract:
Adversarial patch attacks pose a major threat to vision systems by embedding localized perturbations that mislead deep models. Traditional defense methods often require retraining or fine-tuning, making them impractical for real-world deployment. We propose a training-free Visual Retrieval-Augmented Generation (VRAG) framework that integrates Vision-Language Models (VLMs) for adversarial patch detection. By retrieving visually similar patches and images that resemble stored attacks in a continuously expanding database, VRAG performs generative reasoning to identify diverse attack types, all without additional training or fine-tuning. We extensively evaluate open-source large-scale VLMs, including Qwen-VL-Plus, Qwen2.5-VL-72B, and UI-TARS-72B-DPO, alongside Gemini-2.0, a closed-source model. Notably, the open-source UI-TARS-72B-DPO model achieves up to 95 percent classification accuracy, setting a new state-of-the-art for open-source adversarial patch detection. Gemini-2.0 attains the highest overall accuracy, 98 percent, but remains closed-source. Experimental results demonstrate VRAG's effectiveness in identifying a variety of adversarial patches with minimal human annotation, paving the way for robust, practical defenses against evolving adversarial patch attacks.

Paperid: 4150, https://arxiv.org/pdf/2503.20208.pdf

Abstract:
How can robots learn dexterous grasping skills efficiently and apply them adaptively based on user instructions? This work tackles two key challenges: efficient skill acquisition from limited human demonstrations and context-driven skill selection. We introduce AdaDexGrasp, a framework that learns a library of grasping skills from a single human demonstration per skill and selects the most suitable one using a vision-language model (VLM). To improve sample efficiency, we propose a trajectory following reward that guides reinforcement learning (RL) toward states close to a human demonstration while allowing flexibility in exploration. To learn beyond the single demonstration, we employ curriculum learning, progressively increasing object pose variations to enhance robustness. At deployment, a VLM retrieves the appropriate skill based on user instructions, bridging low-level learned skills with high-level intent. We evaluate AdaDexGrasp in both simulation and real-world settings, showing that our approach significantly improves RL efficiency and enables learning human-like grasp strategies across varied object configurations. Finally, we demonstrate zero-shot transfer of our learned policies to a real-world PSYONIC Ability Hand, with a 90% success rate across objects, significantly outperforming the baseline.

Paperid: 4151, https://arxiv.org/pdf/2503.16930.pdf

Abstract:
Dynamic image degradations, including noise, blur and lighting inconsistencies, pose significant challenges in image restoration, often due to sensor limitations or adverse environmental conditions. Existing Deep Unfolding Networks (DUNs) offer stable restoration performance but require manual selection of degradation matrices for each degradation type, limiting their adaptability across diverse scenarios. To address this issue, we propose the Vision-Language-guided Unfolding Network (VLU-Net), a unified DUN framework for handling multiple degradation types simultaneously. VLU-Net leverages a Vision-Language Model (VLM) refined on degraded image-text pairs to align image features with degradation descriptions, selecting the appropriate transform for target degradation. By integrating an automatic VLM-based gradient estimation strategy into the Proximal Gradient Descent (PGD) algorithm, VLU-Net effectively tackles complex multi-degradation restoration tasks while maintaining interpretability. Furthermore, we design a hierarchical feature unfolding structure to enhance VLU-Net framework, efficiently synthesizing degradation patterns across various levels. VLU-Net is the first all-in-one DUN framework and outperforms current leading one-by-one and all-in-one end-to-end methods by 3.74 dB on the SOTS dehazing dataset and 1.70 dB on the Rain100L deraining dataset.

Paperid: 4152, https://arxiv.org/pdf/2503.13861.pdf

Abstract:
Accurately understanding and deciding high-level meta-actions is essential for ensuring reliable and safe autonomous driving systems. While vision-language models (VLMs) have shown significant potential in various autonomous driving tasks, they often suffer from limitations such as inadequate spatial perception and hallucination, reducing their effectiveness in complex autonomous driving scenarios. To address these challenges, we propose a retrieval-augmented decision-making (RAD) framework, a novel architecture designed to enhance VLMs' capabilities to reliably generate meta-actions in autonomous driving scenes. RAD leverages a retrieval-augmented generation (RAG) pipeline to dynamically improve decision accuracy through a three-stage process consisting of the embedding flow, retrieving flow, and generating flow. Additionally, we fine-tune VLMs on a specifically curated dataset derived from the NuScenes dataset to enhance their spatial perception and bird's-eye view image comprehension capabilities. Extensive experimental evaluations on the curated NuScenes-based dataset demonstrate that RAD outperforms baseline methods across key evaluation metrics, including match accuracy, and F1 score, and self-defined overall score, highlighting its effectiveness in improving meta-action decision-making for autonomous driving tasks.

Paperid: 4153, https://arxiv.org/pdf/2503.09808.pdf

Abstract:
Accurate staging of Diabetic Retinopathy (DR) is essential for guiding timely interventions and preventing vision loss. However, current staging models are hardly interpretable, and most public datasets contain no clinical reasoning or interpretation beyond image-level labels. In this paper, we present a novel method that integrates graph representation learning with vision-language models (VLMs) to deliver explainable DR diagnosis. Our approach leverages optical coherence tomography angiography (OCTA) images by constructing biologically informed graphs that encode key retinal vascular features such as vessel morphology and spatial connectivity. A graph neural network (GNN) then performs DR staging while integrated gradients highlight critical nodes and edges and their individual features that drive the classification decisions. We collect this graph-based knowledge which attributes the model's prediction to physiological structures and their characteristics. We then transform it into textual descriptions for VLMs. We perform instruction-tuning with these textual descriptions and the corresponding image to train a student VLM. This final agent can classify the disease and explain its decision in a human interpretable way solely based on a single image input. Experimental evaluations on both proprietary and public datasets demonstrate that our method not only improves classification accuracy but also offers more clinically interpretable results. An expert study further demonstrates that our method provides more accurate diagnostic explanations and paves the way for precise localization of pathologies in OCTA images.

Paperid: 4154, https://arxiv.org/pdf/2503.01531.pdf

Abstract:
Prompt tuning can further enhance the performance of visual-language models across various downstream tasks (e.g., few-shot learning), enabling them to better adapt to specific applications and needs. In this paper, we present a Diversity Covariance-Aware framework that learns distributional information from the data to enhance the few-shot ability of the prompt model. First, we propose a covariance-aware method that models the covariance relationships between visual features and uses anisotropic Mahalanobis distance, instead of the suboptimal cosine distance, to measure the similarity between two modalities. We rigorously derive and prove the validity of this modeling process. Then, we propose the diversity-aware method, which learns multiple diverse soft prompts to capture different attributes of categories and aligns them independently with visual modalities. This method achieves multi-centered covariance modeling, leading to more diverse decision boundaries. Extensive experiments on 11 datasets in various tasks demonstrate the effectiveness of our method.

Paperid: 4155, https://arxiv.org/pdf/2503.00761.pdf

Abstract:
Predicting the near-term behavior of a reactive agent is crucial in many robotic scenarios, yet remains challenging when observations of that agent are sparse or intermittent. Vision-Language Models (VLMs) offer a promising avenue by integrating textual domain knowledge with visual cues, but their one-shot predictions often miss important edge cases and unusual maneuvers. Our key insight is that iterative, counterfactual exploration--where a dedicated module probes each proposed behavior hypothesis, explicitly represented as a plausible trajectory, for overlooked possibilities--can significantly enhance VLM-based behavioral forecasting. We present TRACE (Tree-of-thought Reasoning And Counterfactual Exploration), an inference framework that couples tree-of-thought generation with domain-aware feedback to refine behavior hypotheses over multiple rounds. Concretely, a VLM first proposes candidate trajectories for the agent; a counterfactual critic then suggests edge-case variations consistent with partial observations, prompting the VLM to expand or adjust its hypotheses in the next iteration. This creates a self-improving cycle where the VLM progressively internalizes edge cases from previous rounds, systematically uncovering not only typical behaviors but also rare or borderline maneuvers, ultimately yielding more robust trajectory predictions from minimal sensor data. We validate TRACE on both ground-vehicle simulations and real-world marine autonomous surface vehicles. Experimental results show that our method consistently outperforms standard VLM-driven and purely model-based baselines, capturing a broader range of feasible agent behaviors despite sparse sensing. Evaluation videos and code are available at trace-robotics.github.io.

Paperid: 4156, https://arxiv.org/pdf/2502.18485.pdf

Abstract:
The burgeoning growth of open-sourced vision-language models (VLMs) has catalyzed a plethora of applications across diverse domains. Ensuring the transparency and interpretability of these models is critical for fostering trustworthy and responsible AI systems. In this study, our objective is to delve into the internals of VLMs to interpret the functions of individual neurons. We observe the activations of neurons with respects to the input visual tokens and text tokens, and reveal some interesting findings. Particularly, we found that there are neurons responsible for only visual or text information, or both, respectively, which we refer to them as visual neurons, text neurons, and multi-modal neurons, respectively. We build a framework that automates the explanation of neurons with the assistant of GPT-4o. Meanwhile, for visual neurons, we propose an activation simulator to assess the reliability of the explanations for visual neurons. System statistical analyses on top of one representative VLM of LLaVA, uncover the behaviors/characteristics of different categories of neurons.

Paperid: 4157, https://arxiv.org/pdf/2502.16911.pdf

Abstract:
Zero-shot multi-label recognition (MLR) with Vision-Language Models (VLMs) faces significant challenges without training data, model tuning, or architectural modifications. Existing approaches require prompt tuning or architectural adaptations, limiting zero-shot applicability. Our work proposes a novel solution treating VLMs as black boxes, leveraging scores without training data or ground truth. Using large language model insights on object co-occurrence, we introduce compound prompts grounded in realistic object combinations. Analysis of these prompt scores reveals VLM biases and ``AND''/``OR'' signal ambiguities, notably that maximum compound scores are surprisingly suboptimal compared to second-highest scores. We address these through a debiasing and score-fusion algorithm that corrects image bias and clarifies VLM response behaviors. Our method enhances other zero-shot approaches, consistently improving their results. Experiments show superior mean Average Precision (mAP) compared to methods requiring training data, achieved through refined object ranking for robust zero-shot MLR.

Paperid: 4158, https://arxiv.org/pdf/2502.16618.pdf

Abstract:
Generative AI models, renowned for their ability to synthesize high-quality content, have sparked growing concerns over the improper generation of copyright-protected material. While recent studies have proposed various approaches to address copyright issues, the capability of large vision-language models (LVLMs) to detect copyright infringements remains largely unexplored. In this work, we focus on evaluating the copyright detection abilities of state-of-the-art LVLMs using a various set of image samples. Recognizing the absence of a comprehensive dataset that includes both IP-infringement samples and ambiguous non-infringement negative samples, we construct a benchmark dataset comprising positive samples that violate the copyright protection of well-known IP figures, as well as negative samples that resemble these figures but do not raise copyright concerns. This dataset is created using advanced prompt engineering techniques. We then evaluate leading LVLMs using our benchmark dataset. Our experimental results reveal that LVLMs are prone to overfitting, leading to the misclassification of some negative samples as IP-infringement cases. In the final section, we analyze these failure cases and propose potential solutions to mitigate the overfitting problem.

Paperid: 4159, https://arxiv.org/pdf/2502.16223.pdf

Abstract:
Zero-shot medical detection can further improve detection performance without relying on annotated medical images even upon the fine-tuned model, showing great clinical value. Recent studies leverage grounded vision-language models (GLIP) to achieve this by using detailed disease descriptions as prompts for the target disease name during the inference phase. However, these methods typically treat prompts as equivalent context to the target name, making it difficult to assign specific disease knowledge based on visual information, leading to a coarse alignment between images and target descriptions. In this paper, we propose StructuralGLIP, which introduces an auxiliary branch to encode prompts into a latent knowledge bank layer-by-layer, enabling more context-aware and fine-grained alignment. Specifically, in each layer, we select highly similar features from both the image representation and the knowledge bank, forming structural representations that capture nuanced relationships between image patches and target descriptions. These features are then fused across modalities to further enhance detection performance. Extensive experiments demonstrate that StructuralGLIP achieves a +4.1\% AP improvement over prior state-of-the-art methods across seven zero-shot medical detection benchmarks, and consistently improves fine-tuned models by +3.2\% AP on endoscopy image datasets.

Paperid: 4160, https://arxiv.org/pdf/2502.00931.pdf

Abstract:
Vision-language navigation in unknown environments is crucial for mobile robots. In scenarios such as household assistance and rescue, mobile robots need to understand a human command, such as "find a person wearing black". We present a novel vision-language navigation (VL-Nav) system that integrates efficient spatial reasoning on low-power robots. Unlike prior methods that rely on a single image-level feature similarity to guide a robot, our method integrates pixel-wise vision-language features with curiosity-driven exploration. This approach enables robust navigation to human-instructed instances across diverse environments. We deploy VL-Nav on a four-wheel mobile robot and evaluate its performance through comprehensive navigation tasks in both indoor and outdoor environments, spanning different scales and semantic complexities. Remarkably, VL-Nav operates at a real-time frequency of 30 Hz with a Jetson Orin NX, highlighting its ability to conduct efficient vision-language navigation. Results show that VL-Nav achieves an overall success rate of 86.3%, outperforming previous methods by 44.15%.

Paperid: 4161, https://arxiv.org/pdf/2502.00262.pdf

Abstract:
Autonomous driving systems face significant challenges in handling unpredictable edge-case scenarios, such as adversarial pedestrian movements, dangerous vehicle maneuvers, and sudden environmental changes. Current end-to-end driving models struggle with generalization to these rare events due to limitations in traditional detection and prediction approaches. To address this, we propose INSIGHT (Integration of Semantic and Visual Inputs for Generalized Hazard Tracking), a hierarchical vision-language model (VLM) framework designed to enhance hazard detection and edge-case evaluation. By using multimodal data fusion, our approach integrates semantic and visual representations, enabling precise interpretation of driving scenarios and accurate forecasting of potential dangers. Through supervised fine-tuning of VLMs, we optimize spatial hazard localization using attention-based mechanisms and coordinate regression techniques. Experimental results on the BDD100K dataset demonstrate a substantial improvement in hazard prediction straightforwardness and accuracy over existing models, achieving a notable increase in generalization performance. This advancement enhances the robustness and safety of autonomous driving systems, ensuring improved situational awareness and potential decision-making in complex real-world scenarios.

Paperid: 4162, https://arxiv.org/pdf/2501.18880.pdf

Abstract:
Vision-language model (VLM) fine-tuning for application-specific visual grounding based on natural language instructions has become one of the most popular approaches for learning-enabled autonomous systems. However, such fine-tuning relies heavily on high-quality datasets to achieve successful performance in various downstream tasks. Additionally, VLMs often encounter limitations due to insufficient and imbalanced fine-tuning data. To address these issues, we propose a new generalizable framework to improve VLM fine-tuning by integrating it with a reinforcement learning (RL) agent. Our method utilizes the RL agent to manipulate objects within an indoor setting to create synthetic data for fine-tuning to address certain vulnerabilities of the VLM. Specifically, we use the performance of the VLM to provide feedback to the RL agent to generate informative data that efficiently fine-tune the VLM over the targeted task (e.g. spatial reasoning). The key contribution of this work is developing a framework where the RL agent serves as an informative data sampling tool and assists the VLM in order to enhance performance and address task-specific vulnerabilities. By targeting the data sampling process to address the weaknesses of the VLM, we can effectively train a more context-aware model. In addition, generating synthetic data allows us to have precise control over each scene and generate granular ground truth captions. Our results show that the proposed data generation approach improves the spatial reasoning performance of VLMs, which demonstrates the benefits of using RL-guided data generation in vision-language tasks.

Paperid: 4163, https://arxiv.org/pdf/2501.16937.pdf

Abstract:
Causal language models have demonstrated remarkable capabilities, but their size poses significant challenges for deployment in resource-constrained environments. Knowledge distillation, a widely-used technique for transferring knowledge from a large teacher model to a small student model, presents a promising approach for model compression. A significant remaining issue lies in the major differences between teacher and student models, namely the substantial capacity gap, mode averaging, and mode collapse, which pose barriers during distillation. To address these issues, we introduce $\textit{Temporally Adaptive Interpolated Distillation (TAID)}$, a novel knowledge distillation approach that dynamically interpolates student and teacher distributions through an adaptive intermediate distribution, gradually shifting from the student's initial distribution towards the teacher's distribution. We provide a theoretical analysis demonstrating TAID's ability to prevent mode collapse and empirically show its effectiveness in addressing the capacity gap while balancing mode averaging and mode collapse. Our comprehensive experiments demonstrate TAID's superior performance across various model sizes and architectures in both instruction tuning and pre-training scenarios. Furthermore, we showcase TAID's practical impact by developing two state-of-the-art compact foundation models: $\texttt{TAID-LLM-1.5B}$ for language tasks and $\texttt{TAID-VLM-2B}$ for vision-language tasks. These results demonstrate TAID's effectiveness in creating high-performing and efficient models, advancing the development of more accessible AI technologies.

Paperid: 4164, https://arxiv.org/pdf/2412.08979.pdf

Abstract:
Efficient transfer learning methods such as adapter-based methods have shown great success in unimodal models and vision-language models. However, existing methods have two main challenges in fine-tuning multimodal models. Firstly, they are designed for vision-language tasks and fail to extend to situations where there are more than two modalities. Secondly, they exhibit limited exploitation of interactions between modalities and lack efficiency. To address these issues, in this paper, we propose the loW-rank sequence multimodal adapter (Wander). We first use the outer product to fuse the information from different modalities in an element-wise way effectively. For efficiency, we use CP decomposition to factorize tensors into rank-one components and achieve substantial parameter reduction. Furthermore, we implement a token-level low-rank decomposition to extract more fine-grained features and sequence relationships between modalities. With these designs, Wander enables token-level interactions between sequences of different modalities in a parameter-efficient way. We conduct extensive experiments on datasets with different numbers of modalities, where Wander outperforms state-of-the-art efficient transfer learning methods consistently. The results fully demonstrate the effectiveness, efficiency and universality of Wander.

Paperid: 4165, https://arxiv.org/pdf/2412.04383.pdf

Abstract:
3D Visual Grounding (3DVG) aims to locate objects in 3D scenes based on textual descriptions, essential for applications like augmented reality and robotics. Traditional 3DVG approaches rely on annotated 3D datasets and predefined object categories, limiting scalability and adaptability. To overcome these limitations, we introduce SeeGround, a zero-shot 3DVG framework leveraging 2D Vision-Language Models (VLMs) trained on large-scale 2D data. SeeGround represents 3D scenes as a hybrid of query-aligned rendered images and spatially enriched text descriptions, bridging the gap between 3D data and 2D-VLMs input formats. We propose two modules: the Perspective Adaptation Module, which dynamically selects viewpoints for query-relevant image rendering, and the Fusion Alignment Module, which integrates 2D images with 3D spatial descriptions to enhance object localization. Extensive experiments on ScanRefer and Nr3D demonstrate that our approach outperforms existing zero-shot methods by large margins. Notably, we exceed weakly supervised methods and rival some fully supervised ones, outperforming previous SOTA by 7.7% on ScanRefer and 7.1% on Nr3D, showcasing its effectiveness in complex 3DVG tasks.

Paperid: 4166, https://arxiv.org/pdf/2412.03555.pdf

Abstract:
PaliGemma 2 is an upgrade of the PaliGemma open Vision-Language Model (VLM) based on the Gemma 2 family of language models. We combine the SigLIP-So400m vision encoder that was also used by PaliGemma with the whole range of Gemma 2 models, from the 2B one all the way up to the 27B model. We train these models at three resolutions (224px, 448px, and 896px) in multiple stages to equip them with broad knowledge for transfer via fine-tuning. The resulting family of base models covering different model sizes and resolutions allows us to investigate factors impacting transfer performance (such as learning rate) and to analyze the interplay between the type of task, model size, and resolution. We further increase the number and breadth of transfer tasks beyond the scope of PaliGemma including different OCR-related tasks such as table structure recognition, molecular structure recognition, music score recognition, as well as long fine-grained captioning and radiography report generation, on which PaliGemma 2 obtains state-of-the-art results.

Paperid: 4167, https://arxiv.org/pdf/2411.19309.pdf

Abstract:
Despite the recent advancements of vision-language-action (VLA) models on a variety of robotics tasks, they suffer from critical issues such as poor generalizability to unseen tasks, due to their reliance on behavior cloning exclusively from successful rollouts. Furthermore, they are typically fine-tuned to replicate demonstrations collected by experts under different settings, thus introducing distribution bias and limiting their adaptability to diverse manipulation objectives, such as efficiency, safety, and task completion. To bridge this gap, we introduce GRAPE: Generalizing Robot Policy via Preference Alignment. Specifically, GRAPE aligns VLAs on a trajectory level and implicitly models reward from both successful and failure trials to boost generalizability to diverse tasks. Moreover, GRAPE breaks down complex manipulation tasks to independent stages and automatically guides preference modeling through customized spatiotemporal constraints with keypoints proposed by a large vision-language model. Notably, these constraints are flexible and can be customized to align the model with varying objectives, such as safety, efficiency, or task success. We evaluate GRAPE across a diverse array of tasks in both real-world and simulated environments. Experimental results demonstrate that GRAPE enhances the performance of state-of-the-art VLA models, increasing success rates on in-domain and unseen manipulation tasks by 51.79% and 58.20%, respectively. Additionally, GRAPE can be aligned with various objectives, such as safety and efficiency, reducing collision rates by 37.44% and rollout step-length by 11.15%, respectively. All code, models, and data are available at https://grape-vla.github.io/

Paperid: 4168, https://arxiv.org/pdf/2410.23698.pdf

Abstract:
Large pretrained vision-language models like CLIP have shown promising generalization capability, but may struggle in specialized domains (e.g., satellite imagery) or fine-grained classification (e.g., car models) where the visual concepts are unseen or under-represented during pretraining. Prompt learning offers a parameter-efficient finetuning framework that can adapt CLIP to downstream tasks even when limited annotation data are available. In this paper, we improve prompt learning by distilling the textual knowledge from natural language prompts (either human- or LLM-generated) to provide rich priors for those under-represented concepts. We first obtain a prompt ``summary'' aligned to each input image via a learned prompt aggregator. Then we jointly train a prompt generator, optimized to produce a prompt embedding that stays close to the aggregated summary while minimizing task loss at the same time. We dub such prompt embedding as Aggregate-and-Adapted Prompt Embedding (AAPE). AAPE is shown to be able to generalize to different downstream data distributions and tasks, including vision-language understanding tasks (e.g., few-shot classification, VQA) and generation tasks (image captioning) where AAPE achieves competitive performance. We also show AAPE is particularly helpful to handle non-canonical and OOD examples. Furthermore, AAPE learning eliminates LLM-based inference cost as required by baselines, and scales better with data and LLM model size.

Paperid: 4169, https://arxiv.org/pdf/2410.00193.pdf

Abstract:
Visual language is a system of communication that conveys information through symbols, shapes, and spatial arrangements. Diagrams are a typical example of a visual language depicting complex concepts and their relationships in the form of an image. The symbolic nature of diagrams presents significant challenges for building models capable of understanding them. Recent studies suggest that Large Vision-Language Models (LVLMs) can even tackle complex reasoning tasks involving diagrams. In this paper, we investigate this phenomenon by developing a comprehensive test suite to evaluate the diagram comprehension capability of LVLMs. Our test suite uses a variety of questions focused on concept entities and their relationships over a set of synthetic as well as real diagrams across domains to evaluate the recognition and reasoning abilities of models. Our evaluation of LVLMs shows that while they can accurately identify and reason about entities, their ability to understand relationships is notably limited. Further testing reveals that the decent performance on diagram understanding largely stems from leveraging their background knowledge as shortcuts to identify and reason about the relational information. Thus, we conclude that LVLMs have a limited capability for genuine diagram understanding, and their impressive performance in diagram reasoning is an illusion emanating from other confounding factors, such as the background knowledge in the models.

Paperid: 4170, https://arxiv.org/pdf/2409.13684.pdf

Abstract:
Feature-based methods are commonly used to explain model predictions, but these methods often implicitly assume that interpretable features are readily available. However, this is often not the case for high-dimensional data, and it can be hard even for domain experts to mathematically specify which features are important. Can we instead automatically extract collections or groups of features that are aligned with expert knowledge? To address this gap, we present FIX (Features Interpretable to eXperts), a benchmark for measuring how well a collection of features aligns with expert knowledge. In collaboration with domain experts, we propose FIXScore, a unified expert alignment measure applicable to diverse real-world settings across cosmology, psychology, and medicine domains in vision, language, and time series data modalities. With FIXScore, we find that popular feature-based explanation methods have poor alignment with expert-specified knowledge, highlighting the need for new methods that can better identify features interpretable to experts.

Paperid: 4171, https://arxiv.org/pdf/2601.20835.pdf

Abstract:
Generating 3D humans that functionally interact with 3D scenes remains an open problem with applications in embodied AI, robotics, and interactive content creation. The key challenge involves reasoning about both the semantics of functional elements in 3D scenes and the 3D human poses required to achieve functionality-aware interaction. Unfortunately, existing methods typically lack explicit reasoning over object functionality and the corresponding human-scene contact, resulting in implausible or functionally incorrect interactions. In this work, we propose FunHSI, a training-free, functionality-driven framework that enables functionally correct human-scene interactions from open-vocabulary task prompts. Given a task prompt, FunHSI performs functionality-aware contact reasoning to identify functional scene elements, reconstruct their 3D geometry, and model high-level interactions via a contact graph. We then leverage vision-language models to synthesize a human performing the task in the image and estimate proposed 3D body and hand poses. Finally, the proposed 3D body configuration is refined via stage-wise optimization to ensure physical plausibility and functional correctness. In contrast to existing methods, FunHSI not only synthesizes more plausible general 3D interactions, such as "sitting on a sofa'', while supporting fine-grained functional human-scene interactions, e.g., "increasing the room temperature''. Extensive experiments demonstrate that FunHSI consistently generates functionally correct and physically plausible human-scene interactions across diverse indoor and outdoor scenes.

Paperid: 4172, https://arxiv.org/pdf/2601.18065.pdf

Abstract:
Do vision--language models (VLMs) develop more human-like sensitivity to linguistic concreteness than text-only large language models (LLMs) when both are evaluated with text-only prompts? We study this question with a controlled comparison between matched Llama text backbones and their Llama Vision counterparts across multiple model scales, treating multimodal pretraining as an ablation on perceptual grounding rather than access to images at inference. We measure concreteness effects at three complementary levels: (i) output behavior, by relating question-level concreteness to QA accuracy; (ii) embedding geometry, by testing whether representations organize along a concreteness axis; and (iii) attention dynamics, by quantifying context reliance via attention-entropy measures. In addition, we elicit token-level concreteness ratings from models and evaluate alignment to human norm distributions, testing whether multimodal training yields more human-consistent judgments. Across benchmarks and scales, VLMs show larger gains on more concrete inputs, exhibit clearer concreteness-structured representations, produce ratings that better match human norms, and display systematically different attention patterns consistent with increased grounding.

Paperid: 4173, https://arxiv.org/pdf/2601.13976.pdf

Abstract:
Achieving human-level performance in Vision-and-Language Navigation (VLN) requires an embodied agent to jointly understand multimodal instructions and visual-spatial context while reasoning over long action sequences. Recent works, such as NavCoT and NavGPT-2, demonstrate the potential of Chain-of-Thought (CoT) reasoning for improving interpretability and long-horizon planning. Moreover, multimodal extensions like OctoNav-R1 and CoT-VLA further validate CoT as a promising pathway toward human-like navigation reasoning. However, existing approaches face critical drawbacks: purely textual CoTs lack spatial grounding and easily overfit to sparse annotated reasoning steps, while multimodal CoTs incur severe token inflation by generating imagined visual observations, making real-time navigation impractical. In this work, we propose FantasyVLN, a unified implicit reasoning framework that preserves the benefits of CoT reasoning without explicit token overhead. Specifically, imagined visual tokens are encoded into a compact latent space using a pretrained Visual AutoRegressor (VAR) during CoT reasoning training, and the model jointly learns from textual, visual, and multimodal CoT modes under a unified multi-CoT strategy. At inference, our model performs direct instruction-to-action mapping while still enjoying reasoning-aware representations. Extensive experiments on LH-VLN show that our approach achieves reasoning-aware yet real-time navigation, improving success rates and efficiency while reducing inference latency by an order of magnitude compared to explicit CoT methods.

Paperid: 4174, https://arxiv.org/pdf/2601.12865.pdf

Abstract:
As a pivotal technique for improving the defense of deep models, adversarial robustness transfer via distillation has demonstrated remarkable success in conventional image classification tasks. However, this paradigm encounters critical challenges when applied to vision-language models (VLM) (e.g., CLIP): constructing adversarially robust teacher for large-scale multi-modal models demands prohibitively high computational resources. We bridge this gap by revealing an interesting phenomenon: vanilla CLIP (without adversarial training) exhibits intrinsic defensive capabilities against adversarial examples generated by another CLIP with different architectures. We formally define this as proxy adversarial robustness, and naturally propose a Heterogeneous Proxy Transfer (HPT) framework that establishes cross-architectural robustness distillation channels between CLIP variants, effortlessly enabling the VLM robustness transfer from proxy to target models. Yet, such proxy transfer paradigm easily induces severe overfitting, leading to a sharp degradation in zero-shot natural generalization. To resolve that, we design Generalization-Pivot Decoupling (GPD) by leveraging the difference in learning rate scheduling. This decouples the proxy transfer process into a generalization-anchored warm-up that maintains generalization and a generalization-pulled HPT that promotes adversarial robustness, to achieve an equilibrium between natural generalization and adversarial robustness. Extensive experiments on 15 zero-shot datasets demonstrate the effectiveness of our HPT-GPD method. The code is available at the website of github.com/fxw13/HPT-GPD.

Paperid: 4175, https://arxiv.org/pdf/2512.22217.pdf

Abstract:
Pedestrian Attribute Recognition (PAR) involves predicting fine-grained attributes such as clothing color, gender, and accessories from pedestrian imagery, yet is hindered by severe class imbalance, intricate attribute co-dependencies, and domain shifts. We introduce VLM-PAR, a modular vision-language framework built on frozen SigLIP 2 multilingual encoders. By first aligning image and prompt embeddings via refining visual features through a compact cross-attention fusion, VLM-PAR achieves significant accuracy improvement on the highly imbalanced PA100K benchmark, setting a new state-of-the-art performance, while also delivering significant gains in mean accuracy across PETA and Market-1501 benchmarks. These results underscore the efficacy of integrating large-scale vision-language pretraining with targeted cross-modal refinement to overcome imbalance and generalization challenges in PAR.

Paperid: 4176, https://arxiv.org/pdf/2512.21402.pdf

Abstract:
Evaluating short-form video content requires moving beyond surface-level quality metrics toward human-aligned, multimodal reasoning. While existing frameworks like VideoScore-2 assess visual and semantic fidelity, they do not capture how specific audiovisual attributes drive real audience engagement. In this work, we propose a data-driven evaluation framework that uses Vision-Language Models (VLMs) to extract unsupervised audiovisual features, clusters them into interpretable factors, and trains a regression-based evaluator to predict engagement on short-form edutainment videos. Our curated YouTube Shorts dataset enables systematic analysis of how VLM-derived features relate to human engagement behavior. Experiments show strong correlations between predicted and actual engagement, demonstrating that our lightweight, feature-based evaluator provides interpretable and scalable assessments compared to traditional metrics (e.g., SSIM, FID). By grounding evaluation in both multimodal feature importance and human-centered engagement signals, our approach advances toward robust and explainable video understanding.

Paperid: 4177, https://arxiv.org/pdf/2512.21008.pdf

Abstract:
Mixture-of-Experts (MoE) architectures have advanced the scaling of Large Language Models (LLMs) by activating only a sparse subset of parameters per input, enabling state-of-the-art performance with reduced computational cost. As these models are increasingly deployed in critical domains, understanding and strengthening their alignment mechanisms is essential to prevent harmful outputs. However, existing LLM safety research has focused almost exclusively on dense architectures, leaving the unique safety properties of MoEs largely unexamined. The modular, sparsely-activated design of MoEs suggests that safety mechanisms may operate differently than in dense models, raising questions about their robustness. In this paper, we present GateBreaker, the first training-free, lightweight, and architecture-agnostic attack framework that compromises the safety alignment of modern MoE LLMs at inference time. GateBreaker operates in three stages: (i) gate-level profiling, which identifies safety experts disproportionately routed on harmful inputs, (ii) expert-level localization, which localizes the safety structure within safety experts, and (iii) targeted safety removal, which disables the identified safety structure to compromise the safety alignment. Our study shows that MoE safety concentrates within a small subset of neurons coordinated by sparse routing. Selective disabling of these neurons, approximately 3% of neurons in the targeted expert layers, significantly increases the averaged attack success rate (ASR) from 7.4% to 64.9% against the eight latest aligned MoE LLMs with limited utility degradation. These safety neurons transfer across models within the same family, raising ASR from 17.9% to 67.7% with one-shot transfer attack. Furthermore, GateBreaker generalizes to five MoE vision language models (VLMs) with 60.9% ASR on unsafe image inputs.

Paperid: 4178, https://arxiv.org/pdf/2512.18231.pdf

Abstract:
Vision-Language Models have demonstrated remarkable capabilities in understanding visual content, yet systematic biases in their spatial processing remain largely unexplored. This work identifies and characterizes a systematic spatial attention bias where VLMs consistently prioritize describing left-positioned content before right-positioned content in horizontally concatenated images. Through controlled experiments on image pairs using both open-source and closed-source models, we demonstrate that this bias persists across different architectures, with models describing left-positioned content first in approximately 97% of cases under neutral prompting conditions. Testing on an Arabic-finetuned model reveals that the bias persists despite right-to-left language training, ruling out language reading direction as the primary cause. Investigation of training dataset annotation guidelines from PixMo and Visual Genome reveals no explicit left-first ordering instructions, suggesting the bias is consistent with architectural factors rather than explicit training data instructions. These findings reveal fundamental limitations in how current VLMs process spatial information.

Paperid: 4179, https://arxiv.org/pdf/2512.16523.pdf

Abstract:
Vision-Language Models (VLMs), such as CLIP, have achieved impressive zero-shot recognition performance but remain highly susceptible to adversarial perturbations, posing significant risks in safety-critical scenarios. Previous training-time defenses rely on adversarial fine-tuning, which requires labeled data and costly retraining, while existing test-time strategies fail to reliably distinguish between clean and adversarial inputs, thereby preventing both adversarial robustness and clean accuracy from reaching their optimum. To address these limitations, we propose Test-Time Padding (TTP), a lightweight defense framework that performs adversarial detection followed by targeted adaptation at inference. TTP identifies adversarial inputs via the cosine similarity shift between CLIP feature embeddings computed before and after spatial padding, yielding a universal threshold for reliable detection across architectures and datasets. For detected adversarial cases, TTP employs trainable padding to restore disrupted attention patterns, coupled with a similarity-aware ensemble strategy for a more robust final prediction. For clean inputs, TTP leaves them unchanged by default or optionally integrates existing test-time adaptation techniques for further accuracy gains. Comprehensive experiments on diverse CLIP backbones and fine-grained benchmarks show that TTP consistently surpasses state-of-the-art test-time defenses, delivering substantial improvements in adversarial robustness without compromising clean accuracy. The code for this paper will be released soon.

Paperid: 4180, https://arxiv.org/pdf/2512.12492.pdf

Abstract:
Polyp detectors trained on clean datasets often underperform in real-world endoscopy, where illumination changes, motion blur, and occlusions degrade image quality. Existing approaches struggle with the domain gap between controlled laboratory conditions and clinical practice, where adverse imaging conditions are prevalent. In this work, we propose AdaptiveDetector, a novel two-stage detector-verifier framework comprising a YOLOv11 detector with a vision-language model (VLM) verifier. The detector adaptively adjusts per-frame confidence thresholds under VLM guidance, while the verifier is fine-tuned with Group Relative Policy Optimization (GRPO) using an asymmetric, cost-sensitive reward function specifically designed to discourage missed detections -- a critical clinical requirement. To enable realistic assessment under challenging conditions, we construct a comprehensive synthetic testbed by systematically degrading clean datasets with adverse conditions commonly encountered in clinical practice, providing a rigorous benchmark for zero-shot evaluation. Extensive zero-shot evaluation on synthetically degraded CVC-ClinicDB and Kvasir-SEG images demonstrates that our approach improves recall by 14 to 22 percentage points over YOLO alone, while precision remains within 0.7 points below to 1.7 points above the baseline. This combination of adaptive thresholding and cost-sensitive reinforcement learning achieves clinically aligned, open-world polyp detection with substantially fewer false negatives, thereby reducing the risk of missed precancerous polyps and improving patient outcomes.

Paperid: 4181, https://arxiv.org/pdf/2512.10942.pdf

Abstract:
We introduce VL-JEPA, a vision-language model built on a Joint Embedding Predictive Architecture (JEPA). Instead of autoregressively generating tokens as in classical VLMs, VL-JEPA predicts continuous embeddings of the target texts. By learning in an abstract representation space, the model focuses on task-relevant semantics while abstracting away surface-level linguistic variability. In a strictly controlled comparison against standard token-space VLM training with the same vision encoder and training data, VL-JEPA achieves stronger performance while having 50% fewer trainable parameters. At inference time, a lightweight text decoder is invoked only when needed to translate VL-JEPA predicted embeddings into text. We show that VL-JEPA natively supports selective decoding that reduces the number of decoding operations by 2.85x while maintaining similar performance compared to non-adaptive uniform decoding. Beyond generation, the VL-JEPA's embedding space naturally supports open-vocabulary classification, text-to-video retrieval, and discriminative VQA without any architecture modification. On eight video classification and eight video retrieval datasets, the average performance VL-JEPA surpasses that of CLIP, SigLIP2, and Perception Encoder. At the same time, the model achieves comparable performance as classical VLMs (InstructBLIP, QwenVL) on four VQA datasets: GQA, TallyQA, POPE and POPEv2, despite only having 1.6B parameters.

Paperid: 4182, https://arxiv.org/pdf/2512.03173.pdf

Abstract:
Culture shapes the objects people use and for what purposes, yet mainstream Vision-Language (VL) datasets frequently exhibit cultural biases, disproportionately favoring higher-income, Western contexts. This imbalance reduces model generalizability and perpetuates performance disparities, especially impacting lower-income and non-Western communities. To address these disparities, we propose a novel function-centric framework that categorizes objects by the functions they fulfill, across diverse cultural and economic contexts. We implement this framework by creating the Culture Affordance Atlas, a re-annotated and culturally grounded restructuring of the Dollar Street dataset spanning 46 functions and 288 objects publicly available at https://lit.eecs.umich.edu/CultureAffordance-Atlas/index.html. Through extensive empirical analyses using the CLIP model, we demonstrate that function-centric labels substantially reduce socioeconomic performance gaps between high- and low-income groups by a median of 6 pp (statistically significant), improving model effectiveness for lower-income contexts. Furthermore, our analyses reveals numerous culturally essential objects that are frequently overlooked in prominent VL datasets. Our contributions offer a scalable pathway toward building inclusive VL datasets and equitable AI systems.

Paperid: 4183, https://arxiv.org/pdf/2512.02846.pdf

Abstract:
Anticipating actions before they occur is a core challenge in action understanding research. While conventional methods rely on extracting and aggregating temporal information from videos, as humans we can often predict upcoming actions by observing a single moment from a scene, when given sufficient context. Can a model achieve this competence? The short answer is yes, although its effectiveness depends on the complexity of the task. In this work, we investigate to what extent video aggregation can be replaced with alternative modalities. To this end, based on recent advances in visual feature extraction and language-based reasoning, we introduce AAG, a method for Action Anticipation at a Glimpse. AAG combines RGB features with depth cues from a single frame for enhanced spatial reasoning, and incorporates prior action information to provide long-term context. This context is obtained either through textual summaries from Vision-Language Models, or from predictions generated by a single-frame action recognizer. Our results demonstrate that multimodal single-frame action anticipation using AAG can perform competitively compared to both temporally aggregated video baselines and state-of-the-art methods across three instructional activity datasets: IKEA-ASM, Meccano, and Assembly101.

Paperid: 4184, https://arxiv.org/pdf/2511.22125.pdf

Abstract:
Visual and textual soft prompt tuning can effectively improve the adaptability of Vision-Language Models (VLMs) in downstream tasks. However, fine-tuning on video tasks impairs the model's generalization ability to unseen classes. Existing methods attempt to mitigate this forgetting effect by regularizing the gap between hand-crafted prompts and soft prompts, but this also weakens the learning ability of soft prompts. To address this challenge, we propose a plug-and-play coupling prompt learning framework to optimize the generalization performance of V-L models in video tasks, with the core motivation of mitigating semantic space narrowing during fine-tuning by introducing an externally supervised prompt. Specifically, for textual prompts, we introduce pre-trained prompts from other datasets as hard prompt tokens. These are concatenated with soft prompt tokens and coupled via a learnable mapping layer. This competitive prompting approach prevents the semantic space from overfitting to supervised categories. In addition, we introduce a set of well-designed irrelevant video sets and negative prompts as generic attribute anchors to maintain the generic relevance of the attributes in the pre-trained semantic space, thus preserving the generalization ability. Experiments on video tasks demonstrate that our method significantly outperforms state-of-the-art prompt tuning approaches across generalization benchmarks, particularly on base-to-new class prediction.

Paperid: 4185, https://arxiv.org/pdf/2511.21192.pdf

Abstract:
Vision-Language-Action (VLA) models are vulnerable to adversarial attacks, yet universal and transferable attacks remain underexplored, as most existing patches overfit to a single model and fail in black-box settings. To address this gap, we present a systematic study of universal, transferable adversarial patches against VLA-driven robots under unknown architectures, finetuned variants, and sim-to-real shifts. We introduce UPA-RFAS (Universal Patch Attack via Robust Feature, Attention, and Semantics), a unified framework that learns a single physical patch in a shared feature space while promoting cross-model transfer. UPA-RFAS combines (i) a feature-space objective with an $\ell_1$ deviation prior and repulsive InfoNCE loss to induce transferable representation shifts, (ii) a robustness-augmented two-phase min-max procedure where an inner loop learns invisible sample-wise perturbations and an outer loop optimizes the universal patch against this hardened neighborhood, and (iii) two VLA-specific losses: Patch Attention Dominance to hijack text$\to$vision attention and Patch Semantic Misalignment to induce image-text mismatch without labels. Experiments across diverse VLA models, manipulation suites, and physical executions show that UPA-RFAS consistently transfers across models, tasks, and viewpoints, exposing a practical patch-based attack surface and establishing a strong baseline for future defenses.

Paperid: 4186, https://arxiv.org/pdf/2511.21106.pdf

Abstract:
Efficient Multimodal Large Language Models (MLLMs) compress vision tokens to reduce resource consumption, but the loss of visual information can degrade comprehension capabilities. Although some priors introduce Knowledge Distillation to enhance student models, they overlook the fundamental differences in fine-grained vision comprehension caused by unbalanced vision tokens between the efficient student and vanilla teacher. In this paper, we propose EM-KD, a novel paradigm that enhances the Efficient MLLMs with Knowledge Distillation. To overcome the challenge of unbalanced vision tokens, we first calculate the Manhattan distance between the vision logits of teacher and student, and then align them in the spatial dimension with the Hungarian matching algorithm. After alignment, EM-KD introduces two distillation strategies: 1) Vision-Language Affinity Distillation (VLAD) and 2) Vision Semantic Distillation (VSD). Specifically, VLAD calculates the affinity matrix between text tokens and aligned vision tokens, and minimizes the smooth L1 distance of the student and the teacher affinity matrices. Considering the semantic richness of vision logits in the final layer, VSD employs the reverse KL divergence to measure the discrete probability distributions of the aligned vision logits over the vocabulary space. Comprehensive evaluation on diverse benchmarks demonstrates that EM-KD trained model outperforms prior Efficient MLLMs on both accuracy and efficiency with a large margin, validating its effectiveness. Compared with previous distillation methods, which are equipped with our proposed vision token matching strategy for fair comparison, EM-KD also achieves better performance.

Paperid: 4187, https://arxiv.org/pdf/2511.20814.pdf

Abstract:
We present Sphinx, a synthetic environment for visual perception and reasoning that targets core cognitive primitives. Sphinx procedurally generates puzzles using motifs, tiles, charts, icons, and geometric primitives, each paired with verifiable ground-truth solutions, enabling both precise evaluation and large-scale dataset construction. The benchmark covers 25 task types spanning symmetry detection, geometric transformations, spatial reasoning, chart interpretation, and sequence prediction. Evaluating recent large vision-language models (LVLMs) shows that even state-of-the-art GPT-5 attains only 51.1% accuracy, well below human performance. Finally, we demonstrate that reinforcement learning with verifiable rewards (RLVR) substantially improves model accuracy on these tasks and yields gains on external visual reasoning benchmarks, highlighting its promise for advancing multimodal reasoning.

Paperid: 4188, https://arxiv.org/pdf/2511.17886.pdf

Abstract:
Vision-language models (VLMs) have achieved remarkable success across multimodal tasks, yet their substantial computational demands hinder efficient deployment. Knowledge distillation (KD) has emerged as a powerful approach for building lightweight but competitive models, with strong evidence from both language and vision domains. However, its application to VLMs, particularly CLIP-style models, remains limited, often constrained to small-scale teachers and narrow evaluation tasks such as classification or retrieval. In this work, we present the first systematic study of distillation across a range of CLIP-style teacher models, ranging from standard baselines to large-scale state-of-the-art models. Contrary to trends observed in NLP and vision, we find that stronger teachers do not consistently yield better students; in fact, existing distillation frameworks often fail to scale, leading to degraded performance in downstream multimodal tasks such as visual question answering. Our findings challenge prevailing assumptions in KD and point toward new directions for designing parameter-efficient multimodal models.

Paperid: 4189, https://arxiv.org/pdf/2511.10946.pdf

Abstract:
Vision-language models (VLMs) struggle with 3D-related tasks such as spatial cognition and physical understanding, which are crucial for real-world applications like robotics and embodied agents. We attribute this to a modality gap between the 3D tasks and the 2D training of VLM, which led to inefficient retrieval of 3D information from 2D input. To bridge this gap, we introduce SandboxVLM, a simple yet effective framework that leverages abstract bounding boxes to encode geometric structure and physical kinematics for VLM. Specifically, we design a 3D Sandbox reconstruction and perception pipeline comprising four stages: generating multi-view priors with abstract control, proxy elevation, multi-view voting and clustering, and 3D-aware reasoning. Evaluated in zero-shot settings across multiple benchmarks and VLM backbones, our approach consistently improves spatial intelligence, achieving an 8.3\% gain on SAT Real compared with baseline methods for instance. These results demonstrate that equipping VLMs with a 3D abstraction substantially enhances their 3D reasoning ability without additional training, suggesting new possibilities for general-purpose embodied intelligence.

Paperid: 4190, https://arxiv.org/pdf/2511.10241.pdf

Abstract:
Spatio-Temporal Video Grounding (STVG) aims to localize a spatio-temporal tube that corresponds to a given language query in an untrimmed video. This is a challenging task since it involves complex vision-language understanding and spatiotemporal reasoning. Recent works have explored weakly-supervised setting in STVG to eliminate reliance on fine-grained annotations like bounding boxes or temporal stamps. However, they typically follow a simple late-fusion manner, which generates tubes independent of the text description, often resulting in failed target identification and inconsistent target tracking. To address this limitation, we propose a Tube-conditioned Reconstruction with Mutual Constraints (\textbf{TubeRMC}) framework that generates text-conditioned candidate tubes with pre-trained visual grounding models and further refine them via tube-conditioned reconstruction with spatio-temporal constraints. Specifically, we design three reconstruction strategies from temporal, spatial, and spatio-temporal perspectives to comprehensively capture rich tube-text correspondences. Each strategy is equipped with a Tube-conditioned Reconstructor, utilizing spatio-temporal tubes as condition to reconstruct the key clues in the query. We further introduce mutual constraints between spatial and temporal proposals to enhance their quality for reconstruction. TubeRMC outperforms existing methods on two public benchmarks VidSTG and HCSTVG. Further visualization shows that TubeRMC effectively mitigates both target identification errors and inconsistent tracking.

Paperid: 4191, https://arxiv.org/pdf/2511.08238.pdf

Abstract:
Vision-language fine-tuning has emerged as an efficient paradigm for constructing multimodal foundation models. While textual context often highlights semantic relationships within an image, existing fine-tuning methods typically overlook this information when aligning vision and language, thus leading to suboptimal performance. Toward solving this problem, we propose a method that can improve multimodal alignment and fusion based on both semantics and relationships.Specifically, we first extract multilevel semantic features from different vision encoder to capture more visual cues of the relationships. Then, we learn to project the vision features to group related semantics, among which are more likely to have relationships. Finally, we fuse the visual features with the textual by using inheritable cross-attention, where we globally remove the redundant visual relationships by discarding visual-language feature pairs with low correlation. We evaluate our proposed method on eight foundation models and two downstream tasks, visual question answering and image captioning, and show that it outperforms all existing methods.

Paperid: 4192, https://arxiv.org/pdf/2511.05968.pdf

Abstract:
The integration of medical images with clinical context is essential for generating accurate and clinically interpretable radiology reports. However, current automated methods often rely on resource-heavy Large Language Models (LLMs) or static knowledge graphs and struggle with two fundamental challenges in real-world clinical data: (1) missing modalities, such as incomplete clinical context , and (2) feature entanglement, where mixed modality-specific and shared information leads to suboptimal fusion and clinically unfaithful hallucinated findings. To address these challenges, we propose the DiA-gnostic VLVAE, which achieves robust radiology reporting through Disentangled Alignment. Our framework is designed to be resilient to missing modalities by disentangling shared and modality-specific features using a Mixture-of-Experts (MoE) based Vision-Language Variational Autoencoder (VLVAE). A constrained optimization objective enforces orthogonality and alignment between these latent representations to prevent suboptimal fusion. A compact LLaMA-X decoder then uses these disentangled representations to generate reports efficiently. On the IU X-Ray and MIMIC-CXR datasets, DiA has achieved competetive BLEU@4 scores of 0.266 and 0.134, respectively. Experimental results show that the proposed method significantly outperforms state-of-the-art models.

Paperid: 4193, https://arxiv.org/pdf/2511.04479.pdf

Abstract:
We present ThaiOCRBench, the first comprehensive benchmark for evaluating vision-language models (VLMs) on Thai text-rich visual understanding tasks. Despite recent progress in multimodal modeling, existing benchmarks predominantly focus on high-resource languages, leaving Thai underrepresented, especially in tasks requiring document structure understanding. ThaiOCRBench addresses this gap by offering a diverse, human-annotated dataset comprising 2,808 samples across 13 task categories. We evaluate a wide range of state-of-the-art VLMs in a zero-shot setting, spanning both proprietary and open-source systems. Results show a significant performance gap, with proprietary models (e.g., Gemini 2.5 Pro) outperforming open-source counterparts. Notably, fine-grained text recognition and handwritten content extraction exhibit the steepest performance drops among open-source models. Through detailed error analysis, we identify key challenges such as language bias, structural mismatch, and hallucinated content. ThaiOCRBench provides a standardized framework for assessing VLMs in low-resource, script-complex settings, and provides actionable insights for improving Thai-language document understanding.

Paperid: 4194, https://arxiv.org/pdf/2511.00917.pdf

Abstract:
Today's best-explored routes towards generalist robots center on collecting ever larger "observations-in actions-out" robotics datasets to train large end-to-end models, copying a recipe that has worked for vision-language models (VLMs). We pursue a road less traveled: building generalist policies directly around VLMs by augmenting their general capabilities with specific robot capabilities encapsulated in a carefully curated set of perception, planning, and control modules. In Maestro, a VLM coding agent dynamically composes these modules into a programmatic policy for the current task and scenario. Maestro's architecture benefits from a streamlined closed-loop interface without many manually imposed structural constraints, and a comprehensive and diverse tool repertoire. As a result, it largely surpasses today's VLA models for zero-shot performance on challenging manipulation skills. Further, Maestro is easily extensible to incorporate new modules, easily editable to suit new embodiments such as a quadruped-mounted arm, and even easily adapts from minimal real-world experiences through local code edits.

Paperid: 4195, https://arxiv.org/pdf/2511.00269.pdf

Abstract:
Accurate classification plays a pivotal role in smart agriculture, enabling applications such as crop monitoring, fruit recognition, and pest detection. However, conventional centralized training often requires large-scale data collection, which raises privacy concerns, while standard federated learning struggles with non-independent and identically distributed (non-IID) data and incurs high communication costs. To address these challenges, we propose a federated learning framework that integrates a frozen Contrastive Language-Image Pre-training (CLIP) vision transformer (ViT) with a lightweight transformer classifier. By leveraging the strong feature extraction capability of the pre-trained CLIP ViT, the framework avoids training large-scale models from scratch and restricts federated updates to a compact classifier, thereby reducing transmission overhead significantly. Furthermore, to mitigate performance degradation caused by non-IID data distribution, a small subset (1%) of CLIP-extracted feature representations from all classes is shared across clients. These shared features are non-reversible to raw images, ensuring privacy preservation while aligning class representation across participants. Experimental results on agricultural classification tasks show that the proposed method achieve 86.6% accuracy, which is more than 4 times higher compared to baseline federated learning approaches. This demonstrates the effectiveness and efficiency of combining vision-language model features with federated learning for privacy-preserving and scalable agricultural intelligence.

Paperid: 4196, https://arxiv.org/pdf/2511.00191.pdf

Abstract:
The vision community is undergoing the unprecedented progress with the emergence of Vision-Language Pretraining Models (VLMs). Prompt learning plays as the holy grail of accessing VLMs since it enables their fast adaptation to downstream tasks with limited resources. Whereas existing researches milling around single-prompt paradigms, rarely investigate the technical potential behind their multi-prompt learning counterparts. This paper aims to provide a principled retrospect for vision-language multi-prompt learning. We extend the recent constant modality gap phenomenon to learnable prompts and then, justify the superiority of vision-language transfer with multi-prompt augmentation, empirically and theoretically. In terms of this observation, we propose an Energy-based Multi-prompt Learning (EMPL) to generate multiple prompt embeddings by drawing instances from an energy-based distribution, which is implicitly defined by VLMs. So our EMPL is not only parameter-efficient but also rigorously lead to the balance between in-domain and out-of-domain open-vocabulary generalization. Comprehensive experiments have been conducted to justify our claims and the excellence of EMPL.

Paperid: 4197, https://arxiv.org/pdf/2510.26996.pdf

Abstract:
In this study, we propose MoME, a Mixture of Visual Language Medical Experts, for Medical Image Segmentation. MoME adapts the successful Mixture of Experts (MoE) paradigm, widely used in Large Language Models (LLMs), for medical vision-language tasks. The architecture enables dynamic expert selection by effectively utilizing multi-scale visual features tailored to the intricacies of medical imagery, enriched with textual embeddings. This work explores a novel integration of vision-language models for this domain. Utilizing an assembly of 10 datasets, encompassing 3,410 CT scans, MoME demonstrates strong performance on a comprehensive medical imaging segmentation benchmark. Our approach explores the integration of foundation models for medical imaging, benefiting from the established efficacy of MoE in boosting model performance by incorporating textual information. Demonstrating competitive precision across multiple datasets, MoME explores a novel architecture for achieving robust results in medical image analysis.

Paperid: 4198, https://arxiv.org/pdf/2510.26781.pdf

Abstract:
Charts play an important role in visualization, reasoning, data analysis, and the exchange of ideas among humans. However, existing vision-language models (VLMs) still lack accurate perception of details and struggle to extract fine-grained structures from charts. Such limitations in chart grounding also hinder their ability to compare multiple charts and reason over them. In this paper, we introduce a novel "ChartAlign Benchmark (ChartAB)" to provide a comprehensive evaluation of VLMs in chart grounding tasks, i.e., extracting tabular data, localizing visualization elements, and recognizing various attributes from charts of diverse types and complexities. We design a JSON template to facilitate the calculation of evaluation metrics specifically tailored for each grounding task. By incorporating a novel two-stage inference workflow, the benchmark can further evaluate VLMs capability to align and compare elements/attributes across two charts. Our analysis of evaluations on several recent VLMs reveals new insights into their perception biases, weaknesses, robustness, and hallucinations in chart understanding. These findings highlight the fine-grained discrepancies among VLMs in chart understanding tasks and point to specific skills that need to be strengthened in current models.

Paperid: 4199, https://arxiv.org/pdf/2510.26271.pdf

Abstract:
Vision-language models (VLMs) exhibit uneven performance across languages, a problem that is often exacerbated when the model size is reduced. While Knowledge distillation (KD) demonstrates promising results in transferring knowledge from larger to smaller VLMs, applying KD in multilingualism is an underexplored area. This paper presents a controlled empirical study of KD behavior across five distillation approaches, isolating their effects on cross-lingual representation consistency and downstream performance stability under model compression. We study five distillation formulations across CLIP and SigLIP2, and evaluate them on in-domain retrieval and out-of-domain visual QA. We find that some configurations preserve or even improve multilingual retrieval robustness despite halving model size, but others fail to maintain cross-task stability, exposing design-sensitive trade-offs that aggregate accuracy alone does not reveal.

Paperid: 4200, https://arxiv.org/pdf/2510.24792.pdf

Abstract:
Vision-language models (VLMs) have demonstrated remarkable progress in multimodal reasoning. However, existing benchmarks remain limited in terms of high-quality, human-verified examples. Many current datasets rely on synthetically generated content by large language models (LLMs). Furthermore, most datasets are limited to English, as manual quality assurance of translated samples is time-consuming and costly. To fill this gap, we introduce PISA-Bench, a multilingual benchmark derived from English examples of the expert-created PISA tests, a unified framework for the assessment of student competencies in over eighty countries. Each example consists of human-extracted instructions, questions, answer options, and images, enriched with question type categories, and has been translated from English into five additional languages (Spanish, German, Chinese, French, and Italian), resulting in a fully parallel corpus covering six languages. We evaluate state-of-the-art vision-language models on PISA-Bench and find that especially small models (<20B parameters) fail to achieve high test scores. We further find substantial performance degradation on non-English splits as well as high error-rates when models are tasked with spatial and geometric reasoning. By releasing the dataset and evaluation framework, we provide a resource for advancing research on multilingual multimodal reasoning.

Paperid: 4201, https://arxiv.org/pdf/2510.23930.pdf

Abstract:
Three-dimensional Gaussian Splatting (3DGS) has recently emerged as an efficient representation for novel-view synthesis, achieving impressive visual quality. However, in scenes dominated by large and low-texture regions, common in indoor environments, the photometric loss used to optimize 3DGS yields ambiguous geometry and fails to recover high-fidelity 3D surfaces. To overcome this limitation, we introduce PlanarGS, a 3DGS-based framework tailored for indoor scene reconstruction. Specifically, we design a pipeline for Language-Prompted Planar Priors (LP3) that employs a pretrained vision-language segmentation model and refines its region proposals via cross-view fusion and inspection with geometric priors. 3D Gaussians in our framework are optimized with two additional terms: a planar prior supervision term that enforces planar consistency, and a geometric prior supervision term that steers the Gaussians toward the depth and normal cues. We have conducted extensive experiments on standard indoor benchmarks. The results show that PlanarGS reconstructs accurate and detailed 3D surfaces, consistently outperforming state-of-the-art methods by a large margin. Project page: https://planargs.github.io

Paperid: 4202, https://arxiv.org/pdf/2510.23217.pdf

Abstract:
Automating radiology report generation with Large Vision-Language Models (LVLMs) holds great potential, yet these models often produce clinically critical hallucinations, posing serious risks. Existing hallucination detection methods frequently lack the necessary sentence-level granularity or robust generalization across different LVLM generators. We introduce a novel approach: a sentence-level Process Reward Model (PRM) adapted for this vision-language task. Our PRM predicts the factual correctness of each generated sentence, conditioned on clinical context and preceding text. When fine-tuned on MIMIC-CXR with weakly-supervised labels, a lightweight 0.5B-parameter PRM outperforms existing verification techniques, demonstrating, for instance, relative improvements of 7.5% in Matthews Correlation Coefficient and 1.8% in AUROC over strong white-box baselines on outputs from one LVLM. Unlike methods reliant on internal model states, our PRM demonstrates strong generalization to an unseen LVLM. We further show its practical utility: PRM scores effectively filter low-quality reports, improving F1-CheXbert scores by 4.5% (when discarding the worst 10% of reports). Moreover, when guiding a novel weighted best-of-N selection process on the MIMIC-CXR test set, our PRM show relative improvements in clinical metrics of 7.4% for F1-CheXbert and 0.6% for BERTScore. These results demonstrate that a lightweight, context-aware PRM provides a model-agnostic safety layer for clinical LVLMs without access to internal activations

Paperid: 4203, https://arxiv.org/pdf/2510.21849.pdf

Abstract:
Despite significant advances in vision-language models (VLMs), most existing work follows an English-centric design process, limiting their effectiveness in multilingual settings. In this work, we provide a comprehensive empirical study analyzing the impact of several multilingual design choices, such as training data composition, encoder selection, and text backbones. The result is TowerVision, a family of open multilingual VLMs for both image-text and video-text tasks, built upon the multilingual text-only model Tower+. TowerVision achieves competitive performance on multiple multimodal multilingual benchmarks and shows particular strength in culturally grounded tasks and multimodal translation. By incorporating visual and cultural context during fine-tuning, our models surpass existing approaches trained on substantially larger datasets, as demonstrated on ALM-Bench and Multi30K (image tasks) and ViMUL-Bench (video tasks). Alongside the models, we release VisionBlocks, a high-quality, curated vision-language dataset. Our findings highlight that multilingual vision-language training data substantially improves cross-lingual generalization -- both from high-resource to underrepresented languages and vice versa -- and that instruction-tuned LLMs are not always the optimal initialization point. To support further research, we publicly release all models, data, and training recipes.

Paperid: 4204, https://arxiv.org/pdf/2510.21214.pdf

Abstract:
Multimodal large language models (MLLMs) comprise of both visual and textual modalities to process vision language tasks. However, MLLMs are vulnerable to security-related issues, such as jailbreak attacks that alter the model's input to induce unauthorized or harmful responses. The incorporation of the additional visual modality introduces new dimensions to security threats. In this paper, we proposed a black-box jailbreak method via both text and image prompts to evaluate MLLMs. In particular, we designed text prompts with provocative instructions, along with image prompts that introduced mutation and multi-image capabilities. To strengthen the evaluation, we also designed a Re-attack strategy. Empirical results show that our proposed work can improve capabilities to assess the security of both open-source and closed-source MLLMs. With that, we identified gaps in existing defense methods to propose new strategies for both training-time and inference-time defense methods, and evaluated them across the new jailbreak methods. The experiment results showed that the re-designed defense methods improved protections against the jailbreak attacks.

Paperid: 4205, https://arxiv.org/pdf/2510.17269.pdf

Abstract:
The advancement of vision-language models (VLMs) is hampered by a fragmented landscape of inconsistent and contaminated public datasets. We introduce FineVision, a meticulously collected, curated, and unified corpus of 24 million samples - the largest open resource of its kind. We unify more than 200 sources into 185 subsets via a semi-automated, human-in-the-loop pipeline: automation performs bulk ingestion and schema mapping, while reviewers audit mappings and spot-check outputs to verify faithful consumption of annotations, appropriate formatting and diversity, and safety; issues trigger targeted fixes and re-runs. The workflow further applies rigorous de-duplication within and across sources and decontamination against 66 public benchmarks. FineVision also encompasses agentic/GUI tasks with a unified action space; reviewers validate schemas and inspect a sample of trajectories to confirm executable fidelity. Models trained on FineVision consistently outperform those trained on existing open mixtures across a broad evaluation suite, underscoring the benefits of scale, data hygiene, and balanced automation with human oversight. We release the corpus and curation tools to accelerate data-centric VLM research.

Paperid: 4206, https://arxiv.org/pdf/2510.16344.pdf

Abstract:
Assembly hinges on reliably forming connections between parts; yet most robotic approaches plan assembly sequences and part poses while treating connectors as an afterthought. Connections represent the critical "last mile" of assembly execution, while task planning may sequence operations and motion plan may position parts, the precise establishment of physical connections ultimately determines assembly success or failure. In this paper, we consider connections as first-class primitives in assembly representation, including connector types, specifications, quantities, and placement locations. Drawing inspiration from how humans learn assembly tasks through step-by-step instruction manuals, we present Manual2Skill++, a vision-language framework that automatically extracts structured connection information from assembly manuals. We encode assembly tasks as hierarchical graphs where nodes represent parts and sub-assemblies, and edges explicitly model connection relationships between components. A large-scale vision-language model parses symbolic diagrams and annotations in manuals to instantiate these graphs, leveraging the rich connection knowledge embedded in human-designed instructions. We curate a dataset containing over 20 assembly tasks with diverse connector types to validate our representation extraction approach, and evaluate the complete task understanding-to-execution pipeline across four complex assembly scenarios in simulation, spanning furniture, toys, and manufacturing components with real-world correspondence.

Paperid: 4207, https://arxiv.org/pdf/2510.07077.pdf

Abstract:
Amid growing efforts to leverage advances in large language models (LLMs) and vision-language models (VLMs) for robotics, Vision-Language-Action (VLA) models have recently gained significant attention. By unifying vision, language, and action data at scale, which have traditionally been studied separately, VLA models aim to learn policies that generalise across diverse tasks, objects, embodiments, and environments. This generalisation capability is expected to enable robots to solve novel downstream tasks with minimal or no additional task-specific data, facilitating more flexible and scalable real-world deployment. Unlike previous surveys that focus narrowly on action representations or high-level model architectures, this work offers a comprehensive, full-stack review, integrating both software and hardware components of VLA systems. In particular, this paper provides a systematic review of VLAs, covering their strategy and architectural transition, architectures and building blocks, modality-specific processing techniques, and learning paradigms. In addition, to support the deployment of VLAs in real-world robotic applications, we also review commonly used robot platforms, data collection strategies, publicly available datasets, data augmentation methods, and evaluation benchmarks. Throughout this comprehensive survey, this paper aims to offer practical guidance for the robotics community in applying VLAs to real-world robotic systems. All references categorized by training approach, evaluation method, modality, and dataset are available in the table on our project website: https://vla-survey.github.io .

Paperid: 4208, https://arxiv.org/pdf/2510.01649.pdf

Abstract:
Although existing cross-domain continual learning approaches successfully address many streaming tasks having domain shifts, they call for a fully labeled source domain hindering their feasibility in the privacy constrained environments. This paper goes one step ahead with the problem of source-free cross-domain continual learning where the use of source-domain samples are completely prohibited. We propose the idea of rehearsal-free frequency-aware dynamic prompt collaborations (REFEREE) to cope with the absence of labeled source-domain samples in realm of cross-domain continual learning. REFEREE is built upon a synergy between a source-pre-trained model and a large-scale vision-language model, thus overcoming the problem of sub-optimal generalizations when relying only on a source pre-trained model. The domain shift problem between the source domain and the target domain is handled by a frequency-aware prompting technique encouraging low-frequency components while suppressing high-frequency components. This strategy generates frequency-aware augmented samples, robust against noisy pseudo labels. The noisy pseudo-label problem is further addressed with the uncertainty-aware weighting strategy where the mean and covariance matrix are weighted by prediction uncertainties, thus mitigating the adverse effects of the noisy pseudo label. Besides, the issue of catastrophic forgetting (CF) is overcome by kernel linear discriminant analysis (KLDA) where the backbone network is frozen while the classification is performed using the linear discriminant analysis approach guided by the random kernel method. Our rigorous numerical studies confirm the advantage of our approach where it beats prior arts having access to source domain samples with significant margins.

Paperid: 4209, https://arxiv.org/pdf/2510.00766.pdf

Abstract:
Evaluating image-text alignment while reflecting human preferences across multiple aspects is a significant issue for the development of reliable vision-language applications. It becomes especially crucial in real-world scenarios where multiple valid descriptions exist depending on contexts or user needs. However, research progress is hindered by the lack of comprehensive benchmarks and existing evaluation predictors lacking at least one of these key properties: (1) Alignment with human judgments, (2) Long-sequence processing, (3) Inference efficiency, and (4) Applicability to multi-objective scoring. To address these challenges, we propose a plug-and-play architecture to build a robust predictor, MULTI-TAP (Multi-Objective Task-Aware Predictor), capable of both multi and single-objective scoring. MULTI-TAP can produce a single overall score, utilizing a reward head built on top of a large vision-language model (LVLMs). We show that MULTI-TAP is robust in terms of application to different LVLM architectures, achieving significantly higher performance than existing metrics and even on par with the GPT-4o-based predictor, G-VEval, with a smaller size (7-8B). By training a lightweight ridge regression layer on the frozen hidden states of a pre-trained LVLM, MULTI-TAP can produce fine-grained scores for multiple human-interpretable objectives. MULTI-TAP performs better than VisionREWARD, a high-performing multi-objective reward model, in both performance and efficiency on multi-objective benchmarks and our newly released text-image-to-text dataset, EYE4ALL. Our new dataset, consisting of chosen/rejected human preferences (EYE4ALLPref) and human-annotated fine-grained scores across seven dimensions (EYE4ALLMulti), can serve as a foundation for developing more accessible AI systems by capturing the underlying preferences of users, including blind and low-vision (BLV) individuals.

Paperid: 4210, https://arxiv.org/pdf/2510.00055.pdf

Abstract:
SkinGPT-4, a large vision-language model, leverages annotated skin disease images to augment clinical workflows in underserved communities. However, its training dataset predominantly represents lighter skin tones, limiting diagnostic accuracy for darker tones. Here, we evaluated performance biases in SkinGPT-4 across skin tones on common skin diseases, including eczema, allergic-contact dermatitis, and psoriasis using the open-sourced SCIN dataset. We leveraged the SkinGPT-4 backbone to develop finetuned models for custom skin disease classification tasks and explored bias mitigation strategies. Clinical evaluation by board-certified dermatologists on six relevant skin diseases from 300 SCIN cases assessed images for diagnostic accuracy, informativity, physician utility, and patient utility. Model fairness metrics, including demographic parity and equalized odds, were calculated across skin tones. SkinGPT-4 achieved an average demographic parity of 0.10 across Fitzpatrick types, with notable differences of 0.10-0.15 between lightest and darkest tones across evaluation metrics. Model hallucinations in artifacts and anatomy occurred at a rate of 17.8. Our customized models achieved average F1, precision, and AUROC of 0.75, 0.78, and 0.78 across visually similar disease pairs. Fairness analysis showed an average demographic parity of 0.75, with a maximum disparity of 0.21 across skin tones. The best model achieved parity scores of 0.83, 0.83, 0.76, 0.89, 0.90, and 0.90 for Fitzpatrick I-VI, indicating robust fairness. Large language models such as SkinGPT-4 showed weaker performance on darker tones. Model biases exist across evaluation criteria, and hallucinations may affect diagnostic efficacy. These findings demonstrate the efficacy of training accurate, fair models using existing backbones for custom skin disease classification.

Paperid: 4211, https://arxiv.org/pdf/2509.25852.pdf

Abstract:
Enabling robots to execute long-horizon manipulation tasks from free-form language instructions remains a fundamental challenge in embodied AI. While vision-language models (VLMs) have shown promise as high-level planners, their deployment in the real world is hindered by two gaps: (i) the scarcity of large-scale, sequential manipulation data that couples natural language with multi-step action plans, and (ii) the absence of dense, interpretable rewards for fine-tuning VLMs on planning objectives. To address these issues, we propose REVER, a framework that empowers VLMs to generate and validate long-horizon manipulation plans from natural language instructions in real-world scenarios. Under REVER we train and release RoboFarseer, a VLM incentivized to emit chain-of-thought that perform temporal and spatial reasoning, ensuring physically plausible and logically coherent plans. To obtain training data, we leverage the Universal Manipulation Interface framework to capture hardware-agnostic demonstrations of atomic skills. An automated annotation engine converts each demonstration into vision-instruction-plan triplet. We introduce a verifiable reward that scores the generated plan by its ordered bipartite matching overlap with the ground-truth skill sequence. At run time, the fine-tuned VLM functions both as a planner and as a monitor, verifying step-wise completion. RoboFarseer matches or exceeds the performance of proprietary models that are orders of magnitude larger, while on open-ended planning it surpasses the best baseline by more than 40%. In real-world, long-horizon tasks, the complete system boosts overall success by roughly 60% compared with the same low-level controller without the planner. We will open-source both the dataset and the trained model upon publication.

Paperid: 4212, https://arxiv.org/pdf/2509.24837.pdf

Abstract:
Large Vision-Language Models (VLMs) enable strong multimodal reasoning but incur heavy inference costs from redundant visual tokens. Token pruning alleviates this issue, yet existing approaches face limitations. Attention-based methods rely on raw attention scores, which are often unstable across layers and heads and can lead to redundant selections. Diversity-based methods improve robustness by selecting tokens far apart in feature space but risk dropping regions needed for accurate prediction. We propose \ours, a training-free framework built on a simple intuition: tokens with higher sensitivity are more likely to influence the model's output, and they should also capture complementary visual cues rather than overlapping information. To achieve this, we estimate token sensitivity using zeroth-order perturbations at the projection layer, a shallow and computationally light component of the model. This approach measures how small random perturbations affect the projection outputs, allowing us to approximate each token's influence through lightweight forward passes without backpropagation. Extensive experiments across multiple VLMs and benchmarks show that \ours consistently outperforms prior methods, pruning up to 94.4\% of tokens while maintaining accuracy and significantly improving efficiency, achieving up to 2.30x faster end-to-end inference over the baseline.

Paperid: 4213, https://arxiv.org/pdf/2509.24494.pdf

Abstract:
Recent progress, such as DeepSeek-R1, has shown that the GRPO algorithm, a Reinforcement Learning (RL) approach, can effectively train Chain-of-Thought (CoT) reasoning in Large Language Models (LLMs) and Vision-Language Models (VLMs). In this paper, we analyze three challenges of GRPO: gradient coupling between thoughts and answers, sparse reward signals caused by limited parallel sampling, and unstable advantage estimation. To mitigate these challenges, we propose GRPO-MA, a simple yet theoretically grounded method that leverages multi-answer generation from each thought process, enabling more robust and efficient optimization. Theoretically, we show that the variance of thought advantage decreases as the number of answers per thought increases. Empirically, our gradient analysis confirms this effect, showing that GRPO-MA reduces gradient spikes compared to GRPO. Experiments on math, code, and diverse multimodal tasks demonstrate that GRPO-MA substantially improves performance and training efficiency. Our ablation studies further reveal that increasing the number of answers per thought consistently enhances model performance.

Paperid: 4214, https://arxiv.org/pdf/2509.23663.pdf

Abstract:
Vision-Language Models (VLMs) have shown strong capabilities on diverse multimodal tasks. However, the large number of visual tokens output by the vision encoder severely hinders inference efficiency, and prior studies have shown that many of these tokens are not important and can therefore be safely pruned. In this work, we propose HIVTP, a training-free method to improve VLMs efficiency via hierarchical visual token pruning using a novel middle-layer-based importance score. Specifically, we utilize attention maps extracted from the middle layers of the vision encoder, which better reflect fine-grained and object-level attention, to estimate visual token importance. Based on this, we propose a hierarchical visual token pruning method to retain both globally and locally important visual tokens. Specifically, we reshape the 1-D visual token sequence output by the vision encoder into a 2-D spatial layout. In the global retaining stage, we divide the image into regions and retain tokens with higher importance scores in each region; in the local retaining stage, we then divide the image into small windows and retain the most important token in each local window. Experimental results show that our proposed method, HIVTP, can reduce the time-to-first-token (TTFT) of LLaVA-v1.5-7B and LLaVA-Next-7B by up to 50.0% and 55.1%, respectively, and improve the token generation throughput by up to 60.9% and 47.3%, without sacrificing accuracy, and even achieving improvements on certain benchmarks. Compared with prior works, HIVTP achieves better accuracy while offering higher inference efficiency.

Paperid: 4215, https://arxiv.org/pdf/2509.23098.pdf

Abstract:
Spatial grounding is crucial for referring image segmentation (RIS), where the goal of the task is to localize an object described by language. Current foundational vision-language models (VLMs), such as CLIP, excel at aligning images and text but struggle with understanding spatial relationships. Within the language stream, most existing methods often focus on the primary noun phrase when extracting local text features, undermining contextual tokens. Within the vision stream, CLIP generates similar features for images with different spatial layouts, resulting in limited sensitivity to spatial structure. To address these limitations, we propose \textsc{CoPatch}, a zero-shot RIS framework that leverages internal model components to enhance spatial representations in both text and image modalities. For language, \textsc{CoPatch} constructs hybrid text features by incorporating context tokens carrying spatial cues. For vision, it extracts patch-level image features using our novel path discovered from intermediate layers, where spatial structure is better preserved. These enhanced features are fused into a clustered image-text similarity map, \texttt{CoMap}, enabling precise mask selection. As a result, \textsc{CoPatch} significantly improves spatial grounding in zero-shot RIS across RefCOCO, RefCOCO+, RefCOCOg, and PhraseCut (+ 2--7 mIoU) without requiring any additional training. Our findings underscore the importance of recovering and leveraging the untapped spatial knowledge inherently embedded in VLMs, thereby paving the way for opportunities in zero-shot RIS.

Paperid: 4216, https://arxiv.org/pdf/2509.19571.pdf

Abstract:
Executing open-ended natural language queries is a core problem in robotics. While recent advances in imitation learning and vision-language-actions models (VLAs) have enabled promising end-to-end policies, these models struggle when faced with complex instructions and new scenes. An alternative is to design an explicit scene representation as a queryable interface between the robot and the world, using query results to guide downstream motion planning. In this work, we present Agentic Scene Policies (ASP), an agentic framework that leverages the advanced semantic, spatial, and affordance-based querying capabilities of modern scene representations to implement a capable language-conditioned robot policy. ASP can execute open-vocabulary queries in a zero-shot manner by explicitly reasoning about object affordances in the case of more complex skills. Through extensive experiments, we compare ASP with VLAs on tabletop manipulation problems and showcase how ASP can tackle room-level queries through affordance-guided navigation, and a scaled-up scene representation. (Project page: https://montrealrobotics.ca/agentic-scene-policies.github.io/)

Paperid: 4217, https://arxiv.org/pdf/2509.17588.pdf

Abstract:
Large Vision-Language Models (LVLMs) answer visual questions by transferring information from images to text through a series of attention heads. While this image-to-text information flow is central to visual question answering, its underlying mechanism remains difficult to interpret due to the simultaneous operation of numerous attention heads. To address this challenge, we propose head attribution, a technique inspired by component attribution methods, to identify consistent patterns among attention heads that play a key role in information transfer. Using head attribution, we investigate how LVLMs rely on specific attention heads to identify and answer questions about the main object in an image. Our analysis reveals that a distinct subset of attention heads facilitates the image-to-text information flow. Remarkably, we find that the selection of these heads is governed by the semantic content of the input image rather than its visual appearance. We further examine the flow of information at the token level and discover that (1) text information first propagates to role-related tokens and the final token before receiving image information, and (2) image information is embedded in both object-related and background tokens. Our work provides evidence that image-to-text information flow follows a structured process, and that analysis at the attention-head level offers a promising direction toward understanding the mechanisms of LVLMs.

Paperid: 4218, https://arxiv.org/pdf/2509.16673.pdf

Abstract:
Vision-Language Pre-training (VLP) is drawing increasing interest for its ability to minimize manual annotation requirements while enhancing semantic understanding in downstream tasks. However, its reliance on image-text datasets poses challenges due to privacy concerns and the high cost of obtaining paired annotations. Data augmentation emerges as a viable strategy to address this issue, yet existing methods often fall short of capturing the subtle and complex variations in medical data due to limited diversity. To this end, we propose MedCutMix, a novel multi-modal disease-centric data augmentation method. MedCutMix performs diagnostic sentence CutMix within medical reports and establishes the cross-attention between the diagnostic sentence and medical image to guide attentive manifold mix within the imaging modality. Our approach surpasses previous methods across four downstream radiology diagnosis datasets, highlighting its effectiveness in enhancing performance and generalizability in radiology VLP.

Paperid: 4219, https://arxiv.org/pdf/2509.16343.pdf

Abstract:
Developing trustworthy intelligent vision systems for high-stakes domains, \emph{e.g.}, remote sensing and medical diagnosis, demands broad robustness without costly retraining. We propose \textbf{Visual Reasoning Agent (VRA)}, a training-free, agentic reasoning framework that wraps off-the-shelf vision-language models \emph{and} pure vision systems in a \emph{Think--Critique--Act} loop. While VRA incurs significant additional test-time computation, it achieves up to 40\% absolute accuracy gains on challenging visual reasoning benchmarks. Future work will optimize query routing and early stopping to reduce inference overhead while preserving reliability in vision tasks.

Paperid: 4220, https://arxiv.org/pdf/2509.15435.pdf

Abstract:
Large Vision-Language Models (LVLMs) exhibit strong multimodal capabilities but remain vulnerable to hallucinations from intrinsic errors and adversarial attacks from external exploitations, limiting their reliability in real-world applications. We present ORCA, an agentic reasoning framework that improves the factual accuracy and adversarial robustness of pretrained LVLMs through test-time structured inference reasoning with a suite of small vision models (less than 3B parameters). ORCA operates via an Observe--Reason--Critique--Act loop, querying multiple visual tools with evidential questions, validating cross-model inconsistencies, and refining predictions iteratively without access to model internals or retraining. ORCA also stores intermediate reasoning traces, which supports auditable decision-making. Though designed primarily to mitigate object-level hallucinations, ORCA also exhibits emergent adversarial robustness without requiring adversarial training or defense mechanisms. We evaluate ORCA across three settings: (1) clean images on hallucination benchmarks, (2) adversarially perturbed images without defense, and (3) adversarially perturbed images with defense applied. On the POPE hallucination benchmark, ORCA improves standalone LVLM performance by +3.64\% to +40.67\% across different subsets. Under adversarial perturbations on POPE, ORCA achieves an average accuracy gain of +20.11\% across LVLMs. When combined with defense techniques on adversarially perturbed AMBER images, ORCA further improves standalone LVLM performance, with gains ranging from +1.20\% to +48.00\% across evaluation metrics. These results demonstrate that ORCA offers a promising path toward building more reliable and robust multimodal systems.

Paperid: 4221, https://arxiv.org/pdf/2509.14778.pdf

Abstract:
Health informatics research is characterized by diverse data modalities, rapid knowledge expansion, and the need to integrate insights across biomedical science, data analytics, and clinical practice. These characteristics make it particularly well-suited for agent-based approaches that can automate knowledge exploration, manage complex workflows, and generate clinically meaningful outputs. Recent progress in large language model (LLM)-based agents has demonstrated promising capabilities in literature synthesis, data analysis, and even end-to-end research execution. However, existing systems remain limited for health informatics because they lack mechanisms to interpret medical visualizations and often overlook domain-specific quality requirements. To address these gaps, we introduce OpenLens AI, a fully automated framework tailored to health informatics. OpenLens AI integrates specialized agents for literature review, data analysis, code generation, and manuscript preparation, enhanced by vision-language feedback for medical visualization and quality control for reproducibility. The framework automates the entire research pipeline, producing publication-ready LaTeX manuscripts with transparent and traceable workflows, thereby offering a domain-adapted solution for advancing health informatics research.

Paperid: 4222, https://arxiv.org/pdf/2509.13666.pdf

Abstract:
The ocean is warming and acidifying, increasing the risk of mass mortality events for temperature-sensitive shellfish such as oysters. This motivates the development of long-term monitoring systems. However, human labor is costly and long-duration underwater work is highly hazardous, thus favoring robotic solutions as a safer and more efficient option. To enable underwater robots to make real-time, environment-aware decisions without human intervention, we must equip them with an intelligent "brain." This highlights the need for persistent,wide-area, and low-cost benthic monitoring. To this end, we present DREAM, a Vision Language Model (VLM)-guided autonomy framework for long-term underwater exploration and habitat monitoring. The results show that our framework is highly efficient in finding and exploring target objects (e.g., oysters, shipwrecks) without prior location information. In the oyster-monitoring task, our framework takes 31.5% less time than the previous baseline with the same amount of oysters. Compared to the vanilla VLM, it uses 23% fewer steps while covering 8.88% more oysters. In shipwreck scenes, our framework successfully explores and maps the wreck without collisions, requiring 27.5% fewer steps than the vanilla model and achieving 100% coverage, while the vanilla model achieves 60.23% average coverage in our shipwreck environments.

Paperid: 4223, https://arxiv.org/pdf/2509.12897.pdf

Abstract:
Large Vision-Language Models (LVLMs) can accurately locate key objects in images, yet their attention to these objects tends to be very brief. Motivated by the hypothesis that sustained focus on key objects can improve LVLMs' visual capabilities, we propose Cross-Layer Vision Smoothing (CLVS). The core idea of CLVS is to incorporate a vision memory that smooths the attention distribution across layers. Specifically, we initialize this vision memory with position-unbiased visual attention in the first layer. In subsequent layers, the model's visual attention jointly considers the vision memory from previous layers, while the memory is updated iteratively, thereby maintaining smooth attention on key objects. Given that visual understanding primarily occurs in the early and middle layers of the model, we use uncertainty as an indicator of completed visual understanding and terminate the smoothing process accordingly. Experiments on four benchmarks across three LVLMs confirm the effectiveness and generalizability of our method. CLVS achieves state-of-the-art performance on a variety of visual understanding tasks, with particularly significant improvements in relation and attribute understanding.

Paperid: 4224, https://arxiv.org/pdf/2509.03863.pdf

Abstract:
Discovering diverse visual patterns in continuous cellular automata (CA) is challenging due to the vastness and redundancy of high-dimensional behavioral spaces. Traditional exploration methods like Novelty Search (NS) expand locally by mutating known novel solutions but often plateau when local novelty is exhausted, failing to reach distant, unexplored regions. We introduce Expedition and Expansion (E&E), a hybrid strategy where exploration alternates between local novelty-driven expansions and goal-directed expeditions. During expeditions, E&E leverages a Vision-Language Model (VLM) to generate linguistic goals--descriptions of interesting but hypothetical patterns that drive exploration toward uncharted regions. By operating in semantic spaces that align with human perception, E&E both evaluates novelty and generates goals in conceptually meaningful ways, enhancing the interpretability and relevance of discovered behaviors. Tested on Flow Lenia, a continuous CA known for its rich, emergent behaviors, E&E consistently uncovers more diverse solutions than existing exploration methods. A genealogical analysis further reveals that solutions originating from expeditions disproportionately influence long-term exploration, unlocking new behavioral niches that serve as stepping stones for subsequent search. These findings highlight E&E's capacity to break through local novelty boundaries and explore behavioral landscapes in human-aligned, interpretable ways, offering a promising template for open-ended exploration in artificial life and beyond.

Paperid: 4225, https://arxiv.org/pdf/2508.18753.pdf

Abstract:
Prior human-object interaction (HOI) detection methods have integrated early vision-language models (VLMs) such as CLIP, but only as supporting components within their frameworks. In contrast, recent advances in large, generative VLMs suggest that these models may already possess strong ability to understand images involving HOI. This naturally raises an important question: can general-purpose standalone VLMs effectively solve HOI detection, and how do they compare with specialized HOI methods? Answering this requires a benchmark that can accommodate both paradigms. However, existing HOI benchmarks such as HICO-DET were developed before the emergence of modern VLMs, and their evaluation protocols require exact matches to annotated HOI classes. This is poorly aligned with the generative nature of VLMs, which often yield multiple valid interpretations in ambiguous cases. For example, a static image may capture a person mid-motion with a frisbee, which can plausibly be interpreted as either "throwing" or "catching". When only "catching" is annotated, the other, though equally plausible for the image, is marked incorrect when exact matching is used. As a result, correct predictions might be penalized, affecting both VLMs and HOI-specific methods. To avoid penalizing valid predictions, we introduce a new benchmark that reformulates HOI detection as a multiple-answer multiple-choice task, where each question includes only ground-truth positive options and a curated set of negatives that are constructed to reduce ambiguity (e.g., when "catching" is annotated, "throwing" is not selected as a negative to avoid penalizing valid predictions). The proposed evaluation protocol is the first of its kind for both VLMs and HOI methods, enabling direct comparison and offering new insight into the current state of progress in HOI understanding.

Paperid: 4226, https://arxiv.org/pdf/2508.16942.pdf

Abstract:
Evaluating human actions with clear and detailed feedback is important in areas such as sports, healthcare, and robotics, where decisions rely not only on final outcomes but also on interpretable reasoning. However, most existing methods provide only a final score without explanation or detailed analysis, limiting their practical applicability. To address this, we introduce HieroAction, a vision-language model that delivers accurate and structured assessments of human actions. HieroAction builds on two key ideas: (1) Stepwise Action Reasoning, a tailored chain of thought process designed specifically for action assessment, which guides the model to evaluate actions step by step, from overall recognition through sub action analysis to final scoring, thus enhancing interpretability and structured understanding; and (2) Hierarchical Policy Learning, a reinforcement learning strategy that enables the model to learn fine grained sub action dynamics and align them with high level action quality, thereby improving scoring precision. The reasoning pathway structures the evaluation process, while policy learning refines each stage through reward based optimization. Their integration ensures accurate and interpretable assessments, as demonstrated by superior performance across multiple benchmark datasets. Code will be released upon acceptance.

Paperid: 4227, https://arxiv.org/pdf/2508.16839.pdf

Abstract:
Clinical workflows are fragmented as a patchwork of scripts and task-specific networks that often handle triage, task selection, and model deployment. These pipelines are rarely streamlined for data science pipeline, reducing efficiency and raising operational costs. Workflows also lack data-driven model identification (from imaging/tabular inputs) and standardized delivery of model outputs. In response, we present a practical, healthcare-first framework that uses a single vision-language model (VLM) in two complementary roles. First (Solution 1), the VLM acts as an aware model-card matcher that routes an incoming image to the appropriate specialist model via a three-stage workflow (modality -> primary abnormality -> model-card id). Checks are provided by (i) stagewise prompts that allow early exit via None/Normal/Other and (ii) a stagewise answer selector that arbitrates between the top-2 candidates at each stage, reducing the chance of an incorrect selection and aligning the workflow with clinical risk tolerance. Second (Solution 2), we fine-tune the VLM on specialty-specific datasets ensuring a single model covers multiple downstream tasks within each specialty, maintaining performance while simplifying deployment. Across gastroenterology, hematology, ophthalmology, and pathology, our single-model deployment matches or approaches specialized baselines. Compared with pipelines composed of many task-specific agents, this approach shows that one VLM can both decide and do. It may reduce effort by data scientists, shorten monitoring, increase the transparency of model selection (with per-stage justifications), and lower integration overhead.

Paperid: 4228, https://arxiv.org/pdf/2508.14648.pdf

Abstract:
Data plays a pivotal role in the groundbreaking advancements in artificial intelligence. The quantitative analysis of data significantly contributes to model training, enhancing both the efficiency and quality of data utilization. However, existing data analysis tools often lag in accuracy. For instance, many of these tools even assume that the loss function of neural networks is convex. These limitations make it challenging to implement current methods effectively. In this paper, we introduce a new formulation to approximate a sample's influence by accumulating the differences in influence between consecutive learning steps, which we term Diff-In. Specifically, we formulate the sample-wise influence as the cumulative sum of its changes/differences across successive training iterations. By employing second-order approximations, we approximate these difference terms with high accuracy while eliminating the need for model convexity required by existing methods. Despite being a second-order method, Diff-In maintains computational complexity comparable to that of first-order methods and remains scalable. This efficiency is achieved by computing the product of the Hessian and gradient, which can be efficiently approximated using finite differences of first-order gradients. We assess the approximation accuracy of Diff-In both theoretically and empirically. Our theoretical analysis demonstrates that Diff-In achieves significantly lower approximation error compared to existing influence estimators. Extensive experiments further confirm its superior performance across multiple benchmark datasets in three data-centric tasks: data cleaning, data deletion, and coreset selection. Notably, our experiments on data pruning for large-scale vision-language pre-training show that Diff-In can scale to millions of data points and outperforms strong baselines.

Paperid: 4229, https://arxiv.org/pdf/2508.12957.pdf

Abstract:
Reinforcement learning (RL) with rule-based rewards has demonstrated strong potential in enhancing the reasoning and generalization capabilities of vision-language models (VLMs) and large language models (LLMs), while reducing computational overhead. However, its application in medical imaging remains underexplored. Existing reinforcement fine-tuning (RFT) approaches in this domain primarily target closed-ended visual question answering (VQA), limiting their applicability to real-world clinical reasoning. In contrast, open-ended medical VQA better reflects clinical practice but has received limited attention. While some efforts have sought to unify both formats via semantically guided RL, we observe that model-based semantic rewards often suffer from reward collapse, where responses with significant semantic differences receive similar scores. To address this, we propose ARMed (Adaptive Reinforcement for Medical Reasoning), a novel RL framework for open-ended medical VQA. ARMed first incorporates domain knowledge through supervised fine-tuning (SFT) on chain-of-thought data, then applies reinforcement learning with textual correctness and adaptive semantic rewards to enhance reasoning quality. We evaluate ARMed on six challenging medical VQA benchmarks. Results show that ARMed consistently boosts both accuracy and generalization, achieving a 32.64% improvement on in-domain tasks and an 11.65% gain on out-of-domain benchmarks. These results highlight the critical role of reward discriminability in medical RL and the promise of semantically guided rewards for enabling robust and clinically meaningful multimodal reasoning.

Paperid: 4230, https://arxiv.org/pdf/2508.09218.pdf

Abstract:
Multimodal large language models (MLLMs) are widely used in vision-language reasoning tasks. However, their vulnerability to adversarial prompts remains a serious concern, as safety mechanisms often fail to prevent the generation of harmful outputs. Although recent jailbreak strategies report high success rates, many responses classified as "successful" are actually benign, vague, or unrelated to the intended malicious goal. This mismatch suggests that current evaluation standards may overestimate the effectiveness of such attacks. To address this issue, we introduce a four-axis evaluation framework that considers input on-topicness, input out-of-distribution (OOD) intensity, output harmfulness, and output refusal rate. This framework identifies truly effective jailbreaks. In a substantial empirical study, we reveal a structural trade-off: highly on-topic prompts are frequently blocked by safety filters, whereas those that are too OOD often evade detection but fail to produce harmful content. However, prompts that balance relevance and novelty are more likely to evade filters and trigger dangerous output. Building on this insight, we develop a recursive rewriting strategy called Balanced Structural Decomposition (BSD). The approach restructures malicious prompts into semantically aligned sub-tasks, while introducing subtle OOD signals and visual cues that make the inputs harder to detect. BSD was tested across 13 commercial and open-source MLLMs, where it consistently led to higher attack success rates, more harmful outputs, and fewer refusals. Compared to previous methods, it improves success rates by $67\%$ and harmfulness by $21\%$, revealing a previously underappreciated weakness in current multimodal safety systems.

Paperid: 4231, https://arxiv.org/pdf/2508.07642.pdf

Abstract:
Vision-and-Language Navigation (VLN) poses significant challenges in enabling agents to interpret natural language instructions and navigate complex 3D environments. While recent progress has been driven by large-scale pre-training and data augmentation, current methods still struggle to generalize to unseen scenarios, particularly when complex spatial and temporal reasoning is required. In this work, we propose SkillNav, a modular framework that introduces structured, skill-based reasoning into Transformer-based VLN agents. Our method decomposes navigation into a set of interpretable atomic skills (e.g., Vertical Movement, Area and Region Identification, Stop and Pause), each handled by a specialized agent. We then introduce a novel zero-shot Vision-Language Model (VLM)-based router, which dynamically selects the most suitable agent at each time step by aligning sub-goals with visual observations and historical actions. SkillNav achieves a new state-of-the-art performance on the R2R benchmark and demonstrates strong generalization to the GSA-R2R benchmark that includes novel instruction styles and unseen environments.

Paperid: 4232, https://arxiv.org/pdf/2508.03734.pdf

Abstract:
Visual impairment represents a major global health challenge, with multimodal imaging providing complementary information that is essential for accurate ophthalmic diagnosis. This comprehensive survey systematically reviews the latest advances in multimodal deep learning methods in ophthalmology up to the year 2025. The review focuses on two main categories: task-specific multimodal approaches and large-scale multimodal foundation models. Task-specific approaches are designed for particular clinical applications such as lesion detection, disease diagnosis, and image synthesis. These methods utilize a variety of imaging modalities including color fundus photography, optical coherence tomography, and angiography. On the other hand, foundation models combine sophisticated vision-language architectures and large language models pretrained on diverse ophthalmic datasets. These models enable robust cross-modal understanding, automated clinical report generation, and decision support. The survey critically examines important datasets, evaluation metrics, and methodological innovations including self-supervised learning, attention-based fusion, and contrastive alignment. It also discusses ongoing challenges such as variability in data, limited annotations, lack of interpretability, and issues with generalizability across different patient populations. Finally, the survey outlines promising future directions that emphasize the use of ultra-widefield imaging and reinforcement learning-based reasoning frameworks to create intelligent, interpretable, and clinically applicable AI systems for ophthalmology.

Paperid: 4233, https://arxiv.org/pdf/2508.00445.pdf

Abstract:
Text-to-Image (T2I) models generate high-quality images from text prompts but often exhibit unintended social biases, such as gender or racial stereotypes, even when these attributes are not mentioned. Existing debiasing methods work well for simple or well-known cases but struggle with subtle or overlapping biases. We propose AutoDebias, a framework that automatically identifies and mitigates harmful biases in T2I models without prior knowledge of specific bias types. Specifically, AutoDebias leverages vision-language models to detect biased visual patterns and constructs fairness guides by generating inclusive alternative prompts that reflect balanced representations. These guides drive a CLIP-guided training process that promotes fairer outputs while preserving the original model's image quality and diversity. Unlike existing methods, AutoDebias effectively addresses both subtle stereotypes and multiple interacting biases. We evaluate the framework on a benchmark covering over 25 bias scenarios, including challenging cases where multiple biases occur simultaneously. AutoDebias detects harmful patterns with 91.6% accuracy and reduces biased outputs from 90% to negligible levels, while preserving the visual fidelity of the original model.

Paperid: 4234, https://arxiv.org/pdf/2507.11939.pdf

Abstract:
Charts are a universally adopted medium for interpreting and communicating data. However, existing chart understanding benchmarks are predominantly English-centric, limiting their accessibility and applicability to global audiences. In this paper, we present PolyChartQA, the first large-scale multilingual chart question answering benchmark covering 22,606 charts and 26,151 question-answering pairs across 10 diverse languages. PolyChartQA is built using a decoupled pipeline that separates chart data from rendering code, allowing multilingual charts to be flexibly generated by simply translating the data and reusing the code. We leverage state-of-the-art LLM-based translation and enforce rigorous quality control in the pipeline to ensure the linguistic and semantic consistency of the generated multilingual charts. PolyChartQA facilitates systematic evaluation of multilingual chart understanding. Experiments on both open- and closed-source large vision-language models reveal a significant performance gap between English and other languages, especially low-resource ones with non-Latin scripts. This benchmark lays a foundation for advancing globally inclusive vision-language models.

Paperid: 4235, https://arxiv.org/pdf/2507.10300.pdf

Abstract:
Multimodal large language models (MLLMs) have shown remarkable performance in vision-language tasks. However, existing MLLMs are primarily trained on generic datasets, limiting their ability to reason on domain-specific visual cues such as those in facial images. In particular, tasks that require detailed understanding of facial structure, expression, emotion, and demographic features remain underexplored by MLLMs due to the lack of large-scale annotated face image-text datasets. In this work, we introduce FaceLLM, a multimodal large language model trained specifically for facial image understanding. To construct the training data, we propose a novel weakly supervised pipeline that uses ChatGPT with attribute-aware prompts to generate high-quality question-answer pairs based on images from the FairFace dataset. The resulting corpus, called FairFaceGPT, covers a diverse set of attributes including expression, pose, skin texture, and forensic information. Our experiments demonstrate that FaceLLM improves the performance of MLLMs on various face-centric tasks and achieves state-of-the-art performance. This work highlights the potential of synthetic supervision via language models for building domain-specialized MLLMs, and sets a precedent for trustworthy, human-centric multimodal AI systems. FairFaceGPT dataset and pretrained FaceLLM models are publicly available in the project page.

Paperid: 4236, https://arxiv.org/pdf/2507.04380.pdf

Abstract:
In scenarios requiring both prediction and explanation efficiency for image classification, self-explaining models that perform both tasks in a single inference are effective. However, their training incurs substantial labeling and computational costs. This study aims to tackle the issue by proposing a method to transfer the visual explainability of self-explaining models, learned in a source domain, to a target domain based on a task arithmetic framework. Specifically, we construct a self-explaining model by extending image classifiers based on a vision-language pretrained model. We then define an \emph{explainability vector} as the difference between model parameters trained on the source domain with and without explanation supervision. Based on the task arithmetic framework, we impart explainability to a model trained only on the prediction task in the target domain by applying the explainability vector. Experimental results on various image classification datasets demonstrate that, except for transfers between some less-related domains, visual explainability can be successfully transferred from source to target domains, improving explanation quality in the target domain without sacrificing classification accuracy. Furthermore, we show that the explainability vector learned on a large and diverse dataset like ImageNet, extended with explanation supervision, exhibits universality and robustness, improving explanation quality on nine out of ten different target datasets. We also find that the explanation quality achieved with a single model inference is comparable to that of Kernel SHAP, which requires 150 model inferences.

Paperid: 4237, https://arxiv.org/pdf/2507.02190.pdf

Abstract:
Vision-Language-Action (VLA) models offer a compelling framework for tackling complex robotic manipulation tasks, but they are often expensive to train. In this paper, we propose a novel VLA approach that leverages the competitive performance of Vision Language Models (VLMs) on 2D images to directly infer robot end-effector poses in image frame coordinates. Unlike prior VLA models that output low-level controls, our model predicts trajectory waypoints, making it both more efficient to train and robot embodiment agnostic. Despite its lightweight design, our next-token prediction architecture effectively learns meaningful and executable robot trajectories. We further explore the underutilized potential of incorporating depth images, inference-time techniques such as decoding strategies, and demonstration-conditioned action generation. Our model is trained on a simulated dataset and exhibits strong sim-to-real transfer capabilities. We evaluate our approach using a combination of simulated and real data, demonstrating its effectiveness on a real robotic system.

Paperid: 4238, https://arxiv.org/pdf/2507.00416.pdf

Abstract:
Vision-Language-Action (VLA) models have emerged as a promising framework for enabling generalist robots capable of perceiving, reasoning, and acting in the real world. These models usually build upon pretrained Vision-Language Models (VLMs), which excel at semantic understanding due to large-scale image and text pretraining. However, existing VLMs typically lack precise spatial understanding capabilities, as they are primarily tuned on 2D image-text pairs without 3D supervision. To address this limitation, recent approaches have incorporated explicit 3D inputs such as point clouds or depth maps, but this necessitates additional depth sensors or pre-trained depth estimation models, which may yield defective results. In contrast, our work introduces a plug-and-play module that implicitly incorporates 3D geometry features into VLA models by leveraging an off-the-shelf visual geometry foundation model. This integration provides the model with depth-aware visual representations, improving its ability to understand the geometric structure of the scene and the spatial relationships among objects from RGB images alone. We evaluate our method on a set of spatially challenging tasks in both simulation and the real world. Extensive evaluations show that our method significantly improves the performance of state-of-the-art VLA models across diverse scenarios.

Paperid: 4239, https://arxiv.org/pdf/2506.23639.pdf

Abstract:
Multimodal large language models (MLLMs) have made significant progress in vision-language understanding, yet effectively aligning different modalities remains a fundamental challenge. We present a framework that unifies multimodal understanding by applying byte-pair encoding to visual tokens. Unlike conventional approaches that rely on modality-specific encoders, our method directly incorporates structural information into visual tokens, mirroring successful tokenization strategies in text-only language models. We introduce a priority-guided encoding scheme that considers both frequency and spatial consistency, coupled with a multi-stage training procedure based on curriculum-driven data composition. These enhancements enable the transformer model to better capture cross-modal relationships and reason with visual information. Comprehensive experiments demonstrate improved performance across diverse vision-language tasks. By bridging the gap between visual and textual representations, our approach contributes to the advancement of more capable and efficient multimodal foundation models.

Paperid: 4240, https://arxiv.org/pdf/2506.17587.pdf

Abstract:
Though Large Vision-Language Models (LVLMs) have achieved remarkable performance across various tasks, they are still prone to hallucinations-generating outputs that are textually plausible but visually ungrounded. While prior approaches generally address this issue through data-centric fine-tuning or innovative decoding strategies, these methods often require substantial resources or task-specific configurations. In this work, we introduce an architecture-level solution, HalluRNN, which enhances model stability through recurrent cross-layer reasoning. Specifically, we propose a novel Dual-Gated Depth Propagation Unit (DG-DPU) module, which is shared across layers and recurrently refines hidden states. This allows for the adaptive propagation of information throughout the model, enforces consistency across layers, and mitigates hallucinations caused by representational drift. By fine-tuning only the DG-DPU module, HalluRNN achieves strong and robust performance across multiple benchmarks.

Paperid: 4241, https://arxiv.org/pdf/2506.16157.pdf

Abstract:
Referring Expression Segmentation (RES) enables precise object segmentation in images based on natural language descriptions, offering high flexibility and broad applicability in real-world vision tasks. Despite its impressive performance, the robustness of RES models against adversarial examples remains largely unexplored. While prior adversarial attack methods have explored adversarial robustness on conventional segmentation models, they perform poorly when directly applied to RES models, failing to expose vulnerabilities in its multimodal structure. In practical open-world scenarios, users typically issue multiple, diverse referring expressions to interact with the same image, highlighting the need for adversarial examples that generalize across varied textual inputs. Furthermore, from the perspective of privacy protection, ensuring that RES models do not segment sensitive content without explicit authorization is a crucial aspect of enhancing the robustness and security of multimodal vision-language systems. To address these challenges, we present PEAT, an Embedding-Guided Bidirectional Attack for RES models. Extensive experiments across multiple RES architectures and standard benchmarks show that PEAT consistently outperforms competitive baselines.

Paperid: 4242, https://arxiv.org/pdf/2506.16058.pdf

Abstract:
Open-vocabulary segmentation aims to achieve segmentation of arbitrary categories given unlimited text inputs as guidance. To achieve this, recent works have focused on developing various technical routes to exploit the potential of large-scale pre-trained vision-language models and have made significant progress on existing benchmarks. However, we find that existing test sets are limited in measuring the models' comprehension of ``open-vocabulary" concepts, as their semantic space closely resembles the training space, even with many overlapping categories. To this end, we present a new benchmark named OpenBench that differs significantly from the training semantics. It is designed to better assess the model's ability to understand and segment a wide range of real-world concepts. When testing existing methods on OpenBench, we find that their performance diverges from the conclusions drawn on existing test sets. In addition, we propose a method named OVSNet to improve the segmentation performance for diverse and open scenarios. Through elaborate fusion of heterogeneous features and cost-free expansion of the training space, OVSNet achieves state-of-the-art results on both existing datasets and our proposed OpenBench. Corresponding analysis demonstrate the soundness and effectiveness of our proposed benchmark and method.

Paperid: 4243, https://arxiv.org/pdf/2506.15635.pdf

Abstract:
Large vision-language models have recently demonstrated impressive performance in planning and control tasks, driving interest in their application to real-world robotics. However, deploying these models for reasoning in embodied contexts is limited by their ability to incorporate long-term experience collected across multiple days and represented by vast collections of images. Current VLMs typically struggle to process more than a few hundred images concurrently, highlighting the need for more efficient mechanisms to handle long-term memory in embodied settings. To effectively evaluate these models for long-horizon control, a benchmark must specifically target scenarios where memory is crucial for success. Existing long-video QA benchmarks overlook embodied challenges like object manipulation and navigation, which demand low-level skills and fine-grained reasoning over past interactions. Moreover, effective memory integration in embodied agents involves both recalling relevant historical information and executing actions based on that information, making it essential to study these aspects together rather than in isolation. In this work, we introduce a new benchmark for long-range embodied tasks in the Habitat simulator. This benchmark evaluates memory-based capabilities across 60 tasks requiring sustained engagement and contextual awareness in an environment. The tasks can also be procedurally extended to longer and more challenging versions, enabling scalable evaluation of memory and reasoning. We also present baselines that integrate state-of-the-art VLMs with low level navigation policies, assessing their performance on these memory-intensive tasks and highlight areas for improvement.

Paperid: 4244, https://arxiv.org/pdf/2506.14670.pdf

Abstract:
Traditionally, neighborhood studies have employed interviews, surveys, and manual image annotation guided by detailed protocols to identify environmental characteristics, including physical disorder, decay, street safety, and sociocultural symbols, and to examine their impact on developmental and health outcomes. While these methods yield rich insights, they are time-consuming and require intensive expert intervention. Recent technological advances, including vision-language models (VLMs), have begun to automate parts of this process; however, existing efforts are often ad hoc and lack adaptability across research designs and geographic contexts. In this demo paper, we present StreetLens, a human-centered, researcher-configurable workflow that embeds relevant social science expertise in a VLM for scalable neighborhood environmental assessments. StreetLens mimics the process of trained human coders by grounding the analysis in questions derived from established interview protocols, retrieving relevant street view imagery (SVI), and generating a wide spectrum of semantic annotations from objective features (e.g., the number of cars) to subjective perceptions (e.g., the sense of disorder in an image). By enabling researchers to define the VLM's role through domain-informed prompting, StreetLens places domain knowledge at the core of the analysis process. It also supports the integration of prior survey data to enhance robustness and expand the range of characteristics assessed across diverse settings. We provide a Google Colab notebook to make StreetLens accessible and extensible for researchers working with public or custom SVI datasets. StreetLens represents a shift toward flexible, agentic AI systems that work closely with researchers to accelerate and scale neighborhood studies.

Paperid: 4245, https://arxiv.org/pdf/2506.13367.pdf

Abstract:
Mobile robots exploring indoor environments increasingly rely on vision-language models to perceive high-level semantic cues in camera images, such as object categories. Such models offer the potential to substantially advance robot behaviour for tasks such as object-goal navigation (ObjectNav), where the robot must locate objects specified in natural language by exploring the environment. Current ObjectNav methods heavily depend on prompt engineering for perception and do not address the semantic uncertainty induced by variations in prompt phrasing. Ignoring semantic uncertainty can lead to suboptimal exploration, which in turn limits performance. Hence, we propose a semantic uncertainty-informed active perception pipeline for ObjectNav in indoor environments. We introduce a novel probabilistic sensor model for quantifying semantic uncertainty in vision-language models and incorporate it into a probabilistic geometric-semantic map to enhance spatial understanding. Based on this map, we develop a frontier exploration planner with an uncertainty-informed multi-armed bandit objective to guide efficient object search. Experimental results demonstrate that our method achieves ObjectNav success rates comparable to those of state-of-the-art approaches, without requiring extensive prompt engineering.

Paperid: 4246, https://arxiv.org/pdf/2506.11034.pdf

Abstract:
Large language models (LLMs) have shown remarkable ability in various language tasks, especially with their emergent in-context learning capability. Extending LLMs to incorporate visual inputs, large vision-language models (LVLMs) have shown impressive performance in tasks such as recognition and visual question answering (VQA). Despite increasing interest in the utility of LLMs in causal reasoning tasks such as causal discovery and counterfactual reasoning, there has been relatively little work showcasing the abilities of LVLMs on visual causal reasoning tasks. We take this opportunity to formally introduce a comprehensive causal reasoning benchmark for multi-modal in-context learning from LVLMs. Our CausalVLBench encompasses three representative tasks: causal structure inference, intervention target prediction, and counterfactual prediction. We evaluate the ability of state-of-the-art open-source LVLMs on our causal reasoning tasks across three causal representation learning datasets and demonstrate their fundamental strengths and weaknesses. We hope that our benchmark elucidates the drawbacks of existing vision-language models and motivates new directions and paradigms in improving the visual causal reasoning abilities of LVLMs.

Paperid: 4247, https://arxiv.org/pdf/2506.08429.pdf

Abstract:
The application of visual instruction tuning and other post-training techniques has significantly enhanced the capabilities of Large Language Models (LLMs) in visual understanding, enriching Vision-Language Models (VLMs) with more comprehensive visual language datasets. However, the effectiveness of VLMs is highly dependent on large-scale, high-quality datasets that ensure precise recognition and accurate reasoning. Two key challenges hinder progress: (1) noisy alignments between images and the corresponding text, which leads to misinterpretation, and (2) ambiguous or misleading text, which obscures visual content. To address these challenges, we propose SCALE (Single modality data quality and Cross modality Alignment Evaluation), a novel quality-driven data selection pipeline for VLM instruction tuning datasets. Specifically, SCALE integrates a cross-modality assessment framework that first assigns each data entry to its appropriate vision-language task, generates general and task-specific captions (covering scenes, objects, style, etc.), and evaluates the alignment, clarity, task rarity, text coherence, and image clarity of each entry based on the generated captions. We reveal that: (1) current unimodal quality assessment methods evaluate one modality while overlooking the rest, which can underestimate samples essential for specific tasks and discard the lower-quality instances that help build model robustness; and (2) appropriately generated image captions provide an efficient way to transfer the image-text multimodal task into a unified text modality.

Paperid: 4248, https://arxiv.org/pdf/2506.03928.pdf

Abstract:
The computational expense of redundant vision tokens in Large Vision-Language Models (LVLMs) has led many existing methods to compress them via a vision projector. However, this compression may lose visual information that is crucial for tasks relying on fine-grained spatial relationships, such as OCR and Chart&Table Understanding. In this paper, we propose to resample original vision features across the LLM decoder layers to recover visual information and attain efficiency. Following this principle, we introduce Vision Remember, which includes two key modules: (1) Token-Feature Cross-Attention Layer and (2) Token Bidirectional Self-Attention Layer. In the Token bidirectional attention, we employ self-attention mechanism to maintain the bidirectional interaction between vision tokens and the text-guided token. In the Token-Feature interaction attention, we introduce local cross-attention to resample the visual feature and utilize the multi-level fusion to enrich the visual representation. We conduct comprehensive experiments on multiple visual understanding benchmarks and the results with the LLaVA-NeXT baseline show that Vision Remember outperforms TokenPacker by +2.7 and FastV by +5.7 across nearly all the settings. Compared with previous vision feature re-fusion methods, our approach also surpasses DeepStack by +3.9 and SVA Aggregator by +3.4 on the same baseline. The experimental results validate the generalization capability of the proposed method when combined with various efficient vision projectors and LVLMs.

Paperid: 4249, https://arxiv.org/pdf/2506.03093.pdf

Abstract:
Motivated by the hypothesis that neural network representations encode abstract, interpretable features as linearly accessible, approximately orthogonal directions, sparse autoencoders (SAEs) have become a popular tool in interpretability. However, recent work has demonstrated phenomenology of model representations that lies outside the scope of this hypothesis, showing signatures of hierarchical, nonlinear, and multi-dimensional features. This raises the question: do SAEs represent features that possess structure at odds with their motivating hypothesis? If not, does avoiding this mismatch help identify said features and gain further insights into neural network representations? To answer these questions, we take a construction-based approach and re-contextualize the popular matching pursuits (MP) algorithm from sparse coding to design MP-SAE -- an SAE that unrolls its encoder into a sequence of residual-guided steps, allowing it to capture hierarchical and nonlinearly accessible features. Comparing this architecture with existing SAEs on a mixture of synthetic and natural data settings, we show: (i) hierarchical concepts induce conditionally orthogonal features, which existing SAEs are unable to faithfully capture, and (ii) the nonlinear encoding step of MP-SAE recovers highly meaningful features, helping us unravel shared structure in the seemingly dichotomous representation spaces of different modalities in a vision-language model, hence demonstrating the assumption that useful features are solely linearly accessible is insufficient. We also show that the sequential encoder principle of MP-SAE affords an additional benefit of adaptive sparsity at inference time, which may be of independent interest. Overall, we argue our results provide credence to the idea that interpretability should begin with the phenomenology of representations, with methods emerging from assumptions that fit it.

Paperid: 4250, https://arxiv.org/pdf/2506.02556.pdf

Abstract:
Navigational signs are common aids for human wayfinding and scene understanding, but are underutilized by robots. We argue that they benefit robot navigation and scene understanding, by directly encoding privileged information on actions, spatial regions, and relations. Interpreting signs in open-world settings remains a challenge owing to the complexity of scenes and signs, but recent advances in vision-language models (VLMs) make this feasible. To advance progress in this area, we introduce the task of navigational sign understanding which parses locations and associated directions from signs. We offer a benchmark for this task, proposing appropriate evaluation metrics and curating a test set capturing signs with varying complexity and design across diverse public spaces, from hospitals to shopping malls to transport hubs. We also provide a baseline approach using VLMs, and demonstrate their promise on navigational sign understanding. Code and dataset are available on Github.

Paperid: 4251, https://arxiv.org/pdf/2506.02295.pdf

Abstract:
The inherent complexities of Arabic script; its cursive nature, diacritical marks (tashkeel), and varied typography, pose persistent challenges for Optical Character Recognition (OCR). We present Qari-OCR, a series of vision-language models derived from Qwen2-VL-2B-Instruct, progressively optimized for Arabic through iterative fine-tuning on specialized synthetic datasets. Our leading model, QARI v0.2, establishes a new open-source state-of-the-art with a Word Error Rate (WER) of 0.160, Character Error Rate (CER) of 0.061, and BLEU score of 0.737 on diacritically-rich texts. Qari-OCR demonstrates superior handling of tashkeel, diverse fonts, and document layouts, alongside impressive performance on low-resolution images. Further explorations (QARI v0.3) showcase strong potential for structural document understanding and handwritten text. This work delivers a marked improvement in Arabic OCR accuracy and efficiency, with all models and datasets released to foster further research.

Paperid: 4252, https://arxiv.org/pdf/2506.02096.pdf

Abstract:
Vision-language models (VLMs) trained via reinforcement learning with verifiable reward (RLVR) have shown notable progress in scaling test-time compute effectively. In this work, we investigate how synthesized RL data can further improve RLVR. To this end, we propose \textbf{SynthRL}-a scalable and guaranteed pipeline for automatic data scaling in reasoning-oriented RL training. SynthRL comprises three key stages: (1) selecting seed questions with appropriate distribution, (2) augmenting them into more challenging variants while preserving the original answers, and (3) a guaranteed verification stage that ensures near-perfect correctness and difficulty enhancement. Our empirical experiments demonstrate SynthRL's scalability and effectiveness. When applied to the MMK12 dataset, SynthRL synthesizes over 3.3K additional verifiable, challenging questions from approximately 8K seed samples. Models trained with our synthesized data achieve consistent gains across five out-of-domain visual math reasoning benchmarks, with a significant improvement over baseline models trained on seed data alone. Notably, detailed analysis reveals that the gains are more pronounced on the most challenging evaluation samples, highlighting SynthRL's effectiveness in eliciting deeper and more complex reasoning patterns.

Paperid: 4253, https://arxiv.org/pdf/2506.01196.pdf

Abstract:
We introduce OG-VLA, a novel architecture and learning framework that combines the generalization strengths of Vision Language Action models (VLAs) with the robustness of 3D-aware policies. We address the challenge of mapping natural language instructions and multi-view RGBD observations to quasi-static robot actions. 3D-aware robot policies achieve state-of-the-art performance on precise robot manipulation tasks, but struggle with generalization to unseen instructions, scenes, and objects. On the other hand, VLAs excel at generalizing across instructions and scenes, but can be sensitive to camera and robot pose variations. We leverage prior knowledge embedded in language and vision foundation models to improve generalization of 3D-aware keyframe policies. OG-VLA projects input observations from diverse views into a point cloud which is then rendered from canonical orthographic views, ensuring input view invariance and consistency between input and output spaces. These canonical views are processed with a vision backbone, a Large Language Model (LLM), and an image diffusion model to generate images that encode the next position and orientation of the end-effector on the input scene. Evaluations on the Arnold and Colosseum benchmarks demonstrate state-of-the-art generalization to unseen environments, with over 40% relative improvements while maintaining robust performance in seen settings. We also show real-world adaption in 3 to 5 demonstrations along with strong generalization. Videos and resources at https://og-vla.github.io/

Paperid: 4254, https://arxiv.org/pdf/2506.01031.pdf

Abstract:
Multimodal Large Language Models (MLLMs) have demonstrated strong generalization in vision-language tasks, yet their ability to understand and act within embodied environments remains underexplored. We present NavBench, a benchmark to evaluate the embodied navigation capabilities of MLLMs under zero-shot settings. NavBench consists of two components: (1) navigation comprehension, assessed through three cognitively grounded tasks including global instruction alignment, temporal progress estimation, and local observation-action reasoning, covering 3,200 question-answer pairs; and (2) step-by-step execution in 432 episodes across 72 indoor scenes, stratified by spatial, cognitive, and execution complexity. To support real-world deployment, we introduce a pipeline that converts MLLMs' outputs into robotic actions. We evaluate both proprietary and open-source models, finding that GPT-4o performs well across tasks, while lighter open-source models succeed in simpler cases. Results also show that models with higher comprehension scores tend to achieve better execution performance. Providing map-based context improves decision accuracy, especially in medium-difficulty scenarios. However, most models struggle with temporal understanding, particularly in estimating progress during navigation, which may pose a key challenge.

Paperid: 4255, https://arxiv.org/pdf/2505.24600.pdf

Abstract:
Arabic Optical Character Recognition (OCR) is essential for converting vast amounts of Arabic print media into digital formats. However, training modern OCR models, especially powerful vision-language models, is hampered by the lack of large, diverse, and well-structured datasets that mimic real-world book layouts. Existing Arabic OCR datasets often focus on isolated words or lines or are limited in scale, typographic variety, or structural complexity found in books. To address this significant gap, we introduce SARD (Large-Scale Synthetic Arabic OCR Dataset). SARD is a massive, synthetically generated dataset specifically designed to simulate book-style documents. It comprises 843,622 document images containing 690 million words, rendered across ten distinct Arabic fonts to ensure broad typographic coverage. Unlike datasets derived from scanned documents, SARD is free from real-world noise and distortions, offering a clean and controlled environment for model training. Its synthetic nature provides unparalleled scalability and allows for precise control over layout and content variation. We detail the dataset's composition and generation process and provide benchmark results for several OCR models, including traditional and deep learning approaches, highlighting the challenges and opportunities presented by this dataset. SARD serves as a valuable resource for developing and evaluating robust OCR and vision-language models capable of processing diverse Arabic book-style texts.

Paperid: 4256, https://arxiv.org/pdf/2505.24424.pdf

Abstract:
Vision-language models like CLIP have demonstrated remarkable zero-shot capabilities in classification and retrieval. However, these models often struggle with compositional reasoning - the ability to understand the relationships between concepts. A recent benchmark, SugarCrepe++, reveals that previous works on improving compositionality have mainly improved lexical sensitivity but neglected semantic understanding. In addition, downstream retrieval performance often deteriorates, although one would expect that improving compositionality should enhance retrieval. In this work, we introduce CLIC (Compositionally-aware Learning in CLIP), a fine-tuning method based on a novel training technique combining multiple images and their associated captions. CLIC improves compositionality across architectures as well as differently pre-trained CLIP models, both in terms of lexical and semantic understanding, and achieves consistent gains in retrieval performance. This even applies to the recent CLIPS, which achieves SOTA retrieval performance. Nevertheless, the short fine-tuning with CLIC leads to an improvement in retrieval and to the best compositional CLIP model on SugarCrepe++. All our models and code are available at https://clic-compositional-clip.github.io

Paperid: 4257, https://arxiv.org/pdf/2505.23941.pdf

Abstract:
Large language models (LLMs) memorize a vast amount of prior knowledge from the Internet that help them on downstream tasks but also may notoriously sway their outputs towards wrong or biased answers. In this work, we test how the knowledge about popular subjects hurt the accuracy of vision language models (VLMs) on standard, objective visual tasks of counting and identification. We find that state-of-the-art VLMs are strongly biased (e.g, unable to recognize a fourth stripe has been added to a 3-stripe Adidas logo) scoring an average of 17.05% accuracy in counting (e.g., counting stripes in an Adidas-like logo) across 7 diverse domains from animals, logos, chess, board games, optical illusions, to patterned grids. Insert text (e.g., "Adidas") describing the subject name into the counterfactual image further decreases VLM accuracy. The biases in VLMs are so strong that instructing them to double-check their results or rely exclusively on image details to answer improves counting accuracy by only +2 points, on average. Our work presents an interesting failure mode in VLMs and an automated framework for testing VLM biases. Code and data are available at: vlmsarebiased.github.io.

Paperid: 4258, https://arxiv.org/pdf/2505.16446.pdf

Abstract:
Multimodal large language models (MLLMs) enable powerful cross-modal reasoning capabilities. However, the expanded input space introduces new attack surfaces. Previous jailbreak attacks often inject malicious instructions from text into less aligned modalities, such as vision. As MLLMs increasingly incorporate cross-modal consistency and alignment mechanisms, such explicit attacks become easier to detect and block. In this work, we propose a novel implicit jailbreak framework termed IJA that stealthily embeds malicious instructions into images via least significant bit steganography and couples them with seemingly benign, image-related textual prompts. To further enhance attack effectiveness across diverse MLLMs, we incorporate adversarial suffixes generated by a surrogate model and introduce a template optimization module that iteratively refines both the prompt and embedding based on model feedback. On commercial models like GPT-4o and Gemini-1.5 Pro, our method achieves attack success rates of over 90% using an average of only 3 queries.

Paperid: 4259, https://arxiv.org/pdf/2505.14257.pdf

Abstract:
Due to the unidirectional masking mechanism, Decoder-Only models propagate information from left to right. LVLMs (Large Vision-Language Models) follow the same architecture, with visual information gradually integrated into semantic representations during forward propagation. Through systematic analysis, we observe that the majority of the visual information is absorbed into the semantic representations. However, the model's attention distribution does not exhibit sufficient emphasis on semantic representations. This misalignment between the attention distribution and the actual information flow undermines the model's visual understanding ability and contributes to hallucinations. To address this issue, we enhance the model's visual understanding by leveraging the core information embedded in semantic representations. Specifically, we identify attention heads that focus on core semantic representations based on their attention distributions. Then, through a two-stage optimization paradigm, we propagate the advantages of these attention heads across the entire model, aligning the attention distribution with the actual information flow. We evaluate our method on three image captioning benchmarks using five different LVLMs, demonstrating its effectiveness in significantly reducing hallucinations. Further experiments reveal a trade-off between reduced hallucinations and richer details. Notably, our method allows for manual adjustment of the model's conservativeness, enabling flexible control to meet diverse real-world requirements.

Paperid: 4260, https://arxiv.org/pdf/2505.14227.pdf

Abstract:
We propose Visual-only Question Answering (VoQA), a novel multimodal task in which questions are visually embedded within images, without any accompanying textual input. This requires models to locate, recognize, and reason over visually embedded textual questions, posing challenges for existing large vision-language models (LVLMs), which show notable performance drops even with carefully designed prompts. To bridge this gap, we introduce Guided Response Triggering Supervised Fine-tuning (GRT-SFT), a structured fine-tuning strategy that guides the model to perform step-by-step reasoning purely based on visual input, significantly improving model performance. Our work enhances models' capacity for human-like visual understanding in complex multimodal scenarios, where information, including language, is perceived visually.

Paperid: 4261, https://arxiv.org/pdf/2505.11121.pdf

Abstract:
The development of foundation models through pretraining of vision-language models (VLMs) has recently attracted great attention in remote sensing (RS). VLM pretraining aims to learn image and language alignments from a large number of image-text pairs. Each pretraining image is often associated with multiple captions containing redundant information due to repeated or semantically similar phrases, resulting in increased pretraining and inference time. To overcome this, we introduce a weighted feature aggregation (WFA) strategy for VLM pretraining in RS. Our strategy aims to extract and exploit complementary information from multiple captions per image while reducing redundancies through feature aggregation with importance weighting. To calculate adaptive importance weights for different captions of each image, we propose two techniques: (i) non-parametric uniqueness and (ii) learning-based attention. In the first technique, importance weights are calculated based on the bilingual evaluation understudy (BLEU) scores of the captions to emphasize unique sentences and reduce the influence of repetitive ones. In the second technique, importance weights are learned through an attention mechanism instead of relying on hand-crafted features. The effectiveness of the proposed WFA strategy with the two techniques is analyzed in terms of downstream performance on text-to-image retrieval in RS. Experimental results show that the proposed strategy enables efficient and effective pretraining of VLMs in RS. Based on the experimental analysis, we derive guidelines for selecting appropriate techniques depending on downstream task requirements and resource constraints. The code of this work is publicly available at https://git.tu-berlin.de/rsim/redundacy-aware-rs-vlm.

Paperid: 4262, https://arxiv.org/pdf/2505.10634.pdf

Abstract:
Over-reliance on language priors is a major cause of hallucinations in Large Vision-Language Models (LVLMs), often leading to outputs that are linguistically plausible but visually inconsistent. Recent studies have explored contrastive decoding as a training-free solution. However, these methods typically construct contrastive visual inputs by perturbing the original image, resulting in distorted contrastive distributions, incomplete contrastive signals, and excessive suppression of language priors. Motivated by the observation that language priors tend to remain consistent across different images, we propose Cross-Image Contrastive Decoding (CICD), a simple yet effective training-free method that uses unrelated images as contrastive visual inputs. To address the issue of over-suppressing language priors, which can negatively affect the quality of generated responses, we further introduce a dynamic selection mechanism based on the cross-image differences in model behavior. By selectively suppressing language priors, our method reduces hallucinations without compromising the model's performance. Extensive experiments across multiple benchmarks and LVLMs confirm the effectiveness and generalizability of CICD, particularly in image captioning, where language priors are especially dominant.

Paperid: 4263, https://arxiv.org/pdf/2505.07730.pdf

Abstract:
Visual Document Retrieval (VDR) is an emerging research area that focuses on encoding and retrieving document images directly, bypassing the dependence on Optical Character Recognition (OCR) for document search. A recent advance in VDR was introduced by ColPali, which significantly improved retrieval effectiveness through a late interaction mechanism. ColPali's approach demonstrated substantial performance gains over existing baselines that do not use late interaction on an established benchmark. In this study, we investigate the reproducibility and replicability of VDR methods with and without late interaction mechanisms by systematically evaluating their performance across multiple pre-trained vision-language models. Our findings confirm that late interaction yields considerable improvements in retrieval effectiveness; however, it also introduces computational inefficiencies during inference. Additionally, we examine the adaptability of VDR models to textual inputs and assess their robustness across text-intensive datasets within the proposed benchmark, particularly when scaling the indexing mechanism. Furthermore, our research investigates the specific contributions of late interaction by looking into query-patch matching in the context of visual document retrieval. We find that although query tokens cannot explicitly match image patches as in the text retrieval scenario, they tend to match the patch contains visually similar tokens or their surrounding patches.

Paperid: 4264, https://arxiv.org/pdf/2505.03611.pdf

Abstract:
Face anti-spoofing is a critical technology for ensuring the security of face recognition systems. However, its ability to generalize across diverse scenarios remains a significant challenge. In this paper, we attribute the limited generalization ability to two key factors: covariate shift, which arises from external data collection variations, and semantic shift, which results from substantial differences in emerging attack types. To address both challenges, we propose a novel approach for learning unknown spoof prompts, relying solely on real face images from a single source domain. Our method generates textual prompts for real faces and potential unknown spoof attacks by leveraging the general knowledge embedded in vision-language models, thereby enhancing the model's ability to generalize to unseen target domains. Specifically, we introduce a diverse spoof prompt optimization framework to learn effective prompts. This framework constrains unknown spoof prompts within a relaxed prior knowledge space while maximizing their distance from real face images. Moreover, it enforces semantic independence among different spoof prompts to capture a broad range of spoof patterns. Experimental results on nine datasets demonstrate that the learned prompts effectively transfer the knowledge of vision-language models, enabling state-of-the-art generalization ability against diverse unknown attack types across unseen target domains without using any spoof face images.

Paperid: 4265, https://arxiv.org/pdf/2505.03610.pdf

Abstract:
3D mask presentation attack detection is crucial for protecting face recognition systems against the rising threat of 3D mask attacks. While most existing methods utilize multimodal features or remote photoplethysmography (rPPG) signals to distinguish between real faces and 3D masks, they face significant challenges, such as the high costs associated with multimodal sensors and limited generalization ability. Detection-related text descriptions offer concise, universal information and are cost-effective to obtain. However, the potential of vision-language multimodal features for 3D mask presentation attack detection remains unexplored. In this paper, we propose a novel knowledge-based prompt learning framework to explore the strong generalization capability of vision-language models for 3D mask presentation attack detection. Specifically, our approach incorporates entities and triples from knowledge graphs into the prompt learning process, generating fine-grained, task-specific explicit prompts that effectively harness the knowledge embedded in pre-trained vision-language models. Furthermore, considering different input images may emphasize distinct knowledge graph elements, we introduce a visual-specific knowledge filter based on an attention mechanism to refine relevant elements according to the visual context. Additionally, we leverage causal graph theory insights into the prompt learning process to further enhance the generalization ability of our method. During training, a spurious correlation elimination paradigm is employed, which removes category-irrelevant local image patches using guidance from knowledge-based text features, fostering the learning of generalized causal prompts that align with category-relevant local patches. Experimental results demonstrate that the proposed method achieves state-of-the-art intra- and cross-scenario detection performance on benchmark datasets.

Paperid: 4266, https://arxiv.org/pdf/2504.16505.pdf

Abstract:
Tourism and travel planning increasingly rely on digital assistance, yet existing multimodal AI systems often lack specialized knowledge and contextual understanding of urban environments. We present TraveLLaMA, a specialized multimodal language model designed for urban scene understanding and travel assistance. Our work addresses the fundamental challenge of developing practical AI travel assistants through a novel large-scale dataset of 220k question-answer pairs. This comprehensive dataset uniquely combines 130k text QA pairs meticulously curated from authentic travel forums with GPT-enhanced responses, alongside 90k vision-language QA pairs specifically focused on map understanding and scene comprehension. Through extensive fine-tuning experiments on state-of-the-art vision-language models (LLaVA, Qwen-VL, Shikra), we demonstrate significant performance improvements ranging from 6.5\%-9.4\% in both pure text travel understanding and visual question answering tasks. Our model exhibits exceptional capabilities in providing contextual travel recommendations, interpreting map locations, and understanding place-specific imagery while offering practical information such as operating hours and visitor reviews. Comparative evaluations show TraveLLaMA significantly outperforms general-purpose models in travel-specific tasks, establishing a new benchmark for multi-modal travel assistance systems.

Paperid: 4267, https://arxiv.org/pdf/2504.14395.pdf

Abstract:
To develop trustworthy Vision-Language Models (VLMs), it is essential to address adversarial robustness and hallucination mitigation, both of which impact factual accuracy in high-stakes applications such as defense and healthcare. Existing methods primarily focus on either adversarial defense or hallucination post-hoc correction, leaving a gap in unified robustness strategies. We introduce \textbf{Hydra}, an adaptive agentic framework that enhances plug-in VLMs through iterative reasoning, structured critiques, and cross-model verification, improving both resilience to adversarial perturbations and intrinsic model errors. Hydra employs an Action-Critique Loop, where it retrieves and critiques visual information, leveraging Chain-of-Thought (CoT) and In-Context Learning (ICL) techniques to refine outputs dynamically. Unlike static post-hoc correction methods, Hydra adapts to both adversarial manipulations and intrinsic model errors, making it robust to malicious perturbations and hallucination-related inaccuracies. We evaluate Hydra on four VLMs, three hallucination benchmarks, two adversarial attack strategies, and two adversarial defense methods, assessing performance on both clean and adversarial inputs. Results show that Hydra surpasses plug-in VLMs and state-of-the-art (SOTA) dehallucination methods, even without explicit adversarial defenses, demonstrating enhanced robustness and factual consistency. By bridging adversarial resistance and hallucination mitigation, Hydra provides a scalable, training-free solution for improving the reliability of VLMs in real-world applications.

Paperid: 4268, https://arxiv.org/pdf/2504.13365.pdf

Abstract:
In modern smart agriculture, object detection plays a crucial role by enabling automation, precision farming, and monitoring of resources. From identifying crop health and pest infestations to optimizing harvesting processes, accurate object detection enhances both productivity and sustainability. However, training object detection models often requires large-scale data collection and raises privacy concerns, particularly when sensitive agricultural data is distributed across farms. To address these challenges, we propose VLLFL, a vision-language model-based lightweight federated learning framework (VLLFL). It harnesses the generalization and context-aware detection capabilities of the vision-language model (VLM) and leverages the privacy-preserving nature of federated learning. By training a compact prompt generator to boost the performance of the VLM deployed across different farms, VLLFL preserves privacy while reducing communication overhead. Experimental results demonstrate that VLLFL achieves 14.53% improvement in the performance of VLM while reducing 99.3% communication overhead. Spanning tasks from identifying a wide variety of fruits to detecting harmful animals in agriculture, the proposed framework offers an efficient, scalable, and privacy-preserving solution specifically tailored to agricultural applications.

Paperid: 4269, https://arxiv.org/pdf/2504.09426.pdf

Abstract:
Human infants rapidly develop visual reasoning skills from minimal input, suggesting that developmentally inspired pretraining could significantly enhance the efficiency of vision-language models (VLMs). Although recent efforts have leveraged infant-inspired datasets like SAYCam, existing evaluation benchmarks remain misaligned--they are either too simplistic, narrowly scoped, or tailored for large-scale pretrained models. Additionally, training exclusively on infant data overlooks the broader, diverse input from which infants naturally learn. To address these limitations, we propose BabyVLM, a novel framework comprising comprehensive in-domain evaluation benchmarks and a synthetic training dataset created via child-directed transformations of existing datasets. We demonstrate that VLMs trained with our synthetic dataset achieve superior performance on BabyVLM tasks compared to models trained solely on SAYCam or general-purpose data of the SAYCam size. BabyVLM thus provides a robust, developmentally aligned evaluation tool and illustrates how compact models trained on carefully curated data can generalize effectively, opening pathways toward data-efficient vision-language learning paradigms.

Paperid: 4270, https://arxiv.org/pdf/2504.08205.pdf

Abstract:
Vision models are increasingly deployed in critical applications such as autonomous driving and CCTV monitoring, yet they remain susceptible to resource-consuming attacks. In this paper, we introduce a novel energy-overloading attack that leverages vision language model (VLM) prompts to generate adversarial images targeting vision models. These images, though imperceptible to the human eye, significantly increase GPU energy consumption across various vision models, threatening the availability of these systems. Our framework, EO-VLM (Energy Overload via VLM), is model-agnostic, meaning it is not limited by the architecture or type of the target vision model. By exploiting the lack of safety filters in VLMs like DALL-E 3, we create adversarial noise images without requiring prior knowledge or internal structure of the target vision models. Our experiments demonstrate up to a 50% increase in energy consumption, revealing a critical vulnerability in current vision models.

Paperid: 4271, https://arxiv.org/pdf/2504.05178.pdf

Abstract:
Motion expression video segmentation is designed to segment objects in accordance with the input motion expressions. In contrast to the conventional Referring Video Object Segmentation (RVOS), it places emphasis on motion as well as multi-object expressions, making it more arduous. Recently, Large Multimodal Models (LMMs) have begun to shine in RVOS due to their powerful vision-language perception capabilities. In this work, we propose a simple and effective inference optimization method to fully unleash the potential of LMMs in referring video segmentation. Firstly, we use Sa2VA as our baseline, which is a unified LMM for dense grounded understanding of both images and videos. Secondly, we uniformly sample the video frames during the inference process to enhance the model's understanding of the entire video. Finally, we integrate the results of multiple expert models to mitigate the erroneous predictions of a single model. Our solution achieved 61.98% J&F on the MeViS test set and ranked 1st place in the 4th PVUW Challenge MeViS Track at CVPR 2025.

Paperid: 4272, https://arxiv.org/pdf/2503.20698.pdf

Abstract:
Videos inherently contain multiple modalities, including visual events, text overlays, sounds, and speech, all of which are important for retrieval. However, state-of-the-art multimodal language models like VAST and LanguageBind are built on vision-language models (VLMs), and thus overly prioritize visual signals. Retrieval benchmarks further reinforce this bias by focusing on visual queries and neglecting other modalities. We create a search system MMMORRF that extracts text and features from both visual and audio modalities and integrates them with a novel modality-aware weighted reciprocal rank fusion. MMMORRF is both effective and efficient, demonstrating practicality in searching videos based on users' information needs instead of visual descriptive queries. We evaluate MMMORRF on MultiVENT 2.0 and TVR, two multimodal benchmarks designed for more targeted information needs, and find that it improves nDCG@20 by 81% over leading multimodal encoders and 37% over single-modality retrieval, demonstrating the value of integrating diverse modalities.

Paperid: 4273, https://arxiv.org/pdf/2503.12096.pdf

Abstract:
Test-time prompt tuning for vision-language models (VLMs) is getting attention because of their ability to learn with unlabeled data without fine-tuning. Although test-time prompt tuning methods for VLMs can boost accuracy, the resulting models tend to demonstrate poor calibration, which casts doubts on the reliability and trustworthiness of these models. Notably, more attention needs to be devoted to calibrating the test-time prompt tuning in vision-language models. To this end, we propose a new approach, called O-TPT that introduces orthogonality constraints on the textual features corresponding to the learnable prompts for calibrating test-time prompt tuning in VLMs. Towards introducing orthogonality constraints, we make the following contributions. First, we uncover new insights behind the suboptimal calibration performance of existing methods relying on textual feature dispersion. Second, we show that imposing a simple orthogonalization of textual features is a more effective approach towards obtaining textual dispersion. We conduct extensive experiments on various datasets with different backbones and baselines. The results indicate that our method consistently outperforms the prior state of the art in significantly reducing the overall average calibration error. Also, our method surpasses the zero-shot calibration performance on fine-grained classification tasks.

Paperid: 4274, https://arxiv.org/pdf/2503.10613.pdf

Abstract:
Text-to-image models like stable diffusion and DALLE-3 still struggle with multi-turn image editing. We decompose such a task as an agentic workflow (path) of tool use that addresses a sequence of subtasks by AI tools of varying costs. Conventional search algorithms require expensive exploration to find tool paths. While large language models (LLMs) possess prior knowledge of subtask planning, they may lack accurate estimations of capabilities and costs of tools to determine which to apply in each subtask. Can we combine the strengths of both LLMs and graph search to find cost-efficient tool paths? We propose a three-stage approach "CoSTA*" that leverages LLMs to create a subtask tree, which helps prune a graph of AI tools for the given task, and then conducts A* search on the small subgraph to find a tool path. To better balance the total cost and quality, CoSTA* combines both metrics of each tool on every subtask to guide the A* search. Each subtask's output is then evaluated by a vision-language model (VLM), where a failure will trigger an update of the tool's cost and quality on the subtask. Hence, the A* search can recover from failures quickly to explore other paths. Moreover, CoSTA* can automatically switch between modalities across subtasks for a better cost-quality trade-off. We build a novel benchmark of challenging multi-turn image editing, on which CoSTA* outperforms state-of-the-art image-editing models or agents in terms of both cost and quality, and performs versatile trade-offs upon user preference.

Paperid: 4275, https://arxiv.org/pdf/2503.09871.pdf

Abstract:
Automatically generating training supervision for embodied tasks is crucial, as manual designing is tedious and not scalable. While prior works use large language models (LLMs) or vision-language models (VLMs) to generate rewards, these approaches are largely limited to simple tasks with well-defined rewards, such as pick-and-place. This limitation arises because LLMs struggle to interpret complex scenes compressed into text or code due to their restricted input modality, while VLM-based rewards, though better at visual perception, remain limited by their less expressive output modality. To address these challenges, we leverage the imagination capability of general-purpose video generation models. Given an initial simulation frame and a textual task description, the video generation model produces a video demonstrating task completion with correct semantics. We then extract rich supervisory signals from the generated video, including 6D object pose sequences, 2D segmentations, and estimated depth, to facilitate task learning in simulation. Our approach significantly improves supervision quality for complex embodied tasks, enabling large-scale training in simulators.

Paperid: 4276, https://arxiv.org/pdf/2503.06287.pdf

Abstract:
Visual grounding seeks to localize the image region corresponding to a free-form text description. Recently, the strong multimodal capabilities of Large Vision-Language Models (LVLMs) have driven substantial improvements in visual grounding, though they inevitably require fine-tuning and additional model components to explicitly generate bounding boxes or segmentation masks. However, we discover that a few attention heads in frozen LVLMs demonstrate strong visual grounding capabilities. We refer to these heads, which consistently capture object locations related to text semantics, as localization heads. Using localization heads, we introduce a straightforward and effective training-free visual grounding framework that utilizes text-to-image attention maps from localization heads to identify the target objects. Surprisingly, only three out of thousands of attention heads are sufficient to achieve competitive localization performance compared to existing LVLM-based visual grounding methods that require fine-tuning. Our findings suggest that LVLMs can innately ground objects based on a deep comprehension of the text-image relationship, as they implicitly focus on relevant image regions to generate informative text outputs. All the source codes will be made available to the public.

Paperid: 4277, https://arxiv.org/pdf/2503.03321.pdf

Abstract:
Large multimodal models (LMMs) "see" images by leveraging the attention mechanism between text and visual tokens in the transformer decoder. Ideally, these models should focus on key visual information relevant to the text token. However, recent findings indicate that LMMs have an extraordinary tendency to consistently allocate high attention weights to specific visual tokens, even when these tokens are irrelevant to the corresponding text. In this study, we investigate the property behind the appearance of these irrelevant visual tokens and examine their characteristics. Our findings show that this behavior arises due to the massive activation of certain hidden state dimensions, which resembles the attention sink found in language models. Hence, we refer to this phenomenon as the visual attention sink. In particular, our analysis reveals that removing the irrelevant visual sink tokens does not impact model performance, despite receiving high attention weights. Consequently, we recycle the attention to these tokens as surplus resources, redistributing the attention budget to enhance focus on the image. To achieve this, we introduce Visual Attention Redistribution (VAR), a method that redistributes attention in image-centric heads, which we identify as innately focusing on visual information. VAR can be seamlessly applied across different LMMs to improve performance on a wide range of tasks, including general vision-language tasks, visual hallucination tasks, and vision-centric tasks, all without the need for additional training, models, or inference steps. Experimental results demonstrate that VAR enables LMMs to process visual information more effectively by adjusting their internal attention mechanisms, offering a new direction to enhancing the multimodal capabilities of LMMs.

Paperid: 4278, https://arxiv.org/pdf/2503.00936.pdf

Abstract:
Zero-shot Referring Image Segmentation (RIS) identifies the instance mask that best aligns with a specified referring expression without training and fine-tuning, significantly reducing the labor-intensive annotation process. Despite achieving commendable results, previous CLIP-based models have a critical drawback: the models exhibit a notable reduction in their capacity to discern relative spatial relationships of objects. This is because they generate all possible masks on an image and evaluate each masked region for similarity to the given expression, often resulting in decreased sensitivity to direct positional clues in text inputs. Moreover, most methods have weak abilities to manage relationships between primary words and their contexts, causing confusion and reduced accuracy in identifying the correct target region. To address these challenges, we propose IteRPrimE (Iterative Grad-CAM Refinement and Primary word Emphasis), which leverages a saliency heatmap through Grad-CAM from a Vision-Language Pre-trained (VLP) model for image-text matching. An iterative Grad-CAM refinement strategy is introduced to progressively enhance the model's focus on the target region and overcome positional insensitivity, creating a self-correcting effect. Additionally, we design the Primary Word Emphasis module to help the model handle complex semantic relations, enhancing its ability to attend to the intended object. Extensive experiments conducted on the RefCOCO/+/g, and PhraseCut benchmarks demonstrate that IteRPrimE outperforms previous state-of-the-art zero-shot methods, particularly excelling in out-of-domain scenarios.

Paperid: 4279, https://arxiv.org/pdf/2502.19546.pdf

Abstract:
General-purpose vision-language models (VLMs) demonstrate impressive capabilities, but their opaque training on uncurated internet data posse critical limitations for high-stakes decision-making, such as in neurosurgery. We present CNS-Obsidian, a neurosurgical VLM trained on peer-reviewed neurosurgical literature, and demonstrate its clinical utility compared with GPT-4o in a real-world setting. We compiled 23,984 articles from Neurosurgery Publications journals, yielding 78,853 figures and captions. Using GPT-4o and Claude Sonnet-3.5, we converted these image-text pairs into 263,064 training samples across three formats: instruction fine-tuning, multiple-choice questions, and differential diagnosis. We trained CNS-Obsidian, a fine-tune of the 34-billion parameter LLaVA-Next model. In a blinded, randomized deployment trial at NYU Langone Health (Aug 30-Nov 30, 2024), neurosurgeons were assigned to use either CNS-Obsidian or GPT-4o as a diagnostic co-pilot after patient consultations. Primary outcomes were diagnostic helpfulness and accuracy. CNS-Obsidian matched GPT-4o on synthetic questions (76.13% vs 77.54%, p=0.235), but only achieved 46.81% accuracy on human-generated questions versus GPT-4o's 65.70% (p<10-15). In the randomized trial, 70 consultations were evaluated (32 CNS-Obsidian, 38 GPT-4o) from 959 total consults. CNS-Obsidian received positive ratings in 40.62% of cases versus 57.89% for GPT-4o (p=0.230). Both models included correct diagnosis in approximately 60% of cases (59.38% vs 65.79%, p=0.626). Domain-specific VLMs trained on curated scientific literature can approach frontier model performance in specialized medical domains despite being orders of magnitude smaller and less expensive to train. However, low clinical utilization suggests chatbot interfaces may not align with specialist workflows, indicating need for alternative AI integration strategies.

Paperid: 4280, https://arxiv.org/pdf/2502.17710.pdf

Abstract:
Gestures are an integral part of non-verbal communication, with meanings that vary across cultures, and misinterpretations that can have serious social and diplomatic consequences. As AI systems become more integrated into global applications, ensuring they do not inadvertently perpetuate cultural offenses is critical. To this end, we introduce Multi-Cultural Set of Inappropriate Gestures and Nonverbal Signs (MC-SIGNS), a dataset of 288 gesture-country pairs annotated for offensiveness, cultural significance, and contextual factors across 25 gestures and 85 countries. Through systematic evaluation using MC-SIGNS, we uncover critical limitations: text-to-image (T2I) systems exhibit strong US-centric biases, performing better at detecting offensive gestures in US contexts than in non-US ones; large language models (LLMs) tend to over-flag gestures as offensive; and vision-language models (VLMs) default to US-based interpretations when responding to universal concepts like wishing someone luck, frequently suggesting culturally inappropriate gestures. These findings highlight the urgent need for culturally-aware AI safety mechanisms to ensure equitable global deployment of AI technologies.

Paperid: 4281, https://arxiv.org/pdf/2502.14883.pdf

Abstract:
For individuals with blindness or low vision (BLV), navigating complex environments can pose serious risks. Large Vision-Language Models (LVLMs) show promise for generating scene descriptions, but their effectiveness for BLV users remains underexplored. To address this gap, we conducted a user study with eight BLV participants to systematically evaluate preferences for six types of LVLM descriptions. While they helped to reduce fear and improve actionability, user ratings showed wide variation in sufficiency and conciseness. Furthermore, GPT-4o--despite its strong potential to refine descriptions--was not consistently preferred by participants. We use the insights obtained from the user study to build training data for building our new automatic evaluation metric that can capture BLV preferences effectively. Our findings underscore the urgent need for BLV-centered evaluation metrics and human-in-the-loop feedback to advance LVLM description quality for accessibility.

Paperid: 4282, https://arxiv.org/pdf/2502.08791.pdf

Abstract:
Vision-language navigation (VLN) has emerged as a promising paradigm, enabling mobile robots to perform zero-shot inference and execute tasks without specific pre-programming. However, current systems often separate map exploration and path planning, with exploration relying on inefficient algorithms due to limited (partially observed) environmental information. In this paper, we present a novel navigation pipeline named "VL-Explore" for simultaneous exploration and target discovery in unknown environments, leveraging the capabilities of a vision-language model named CLIP. Our approach requires only monocular vision and operates without any prior map or knowledge about the target. For comprehensive evaluations, we designed a functional prototype of a UGV (unmanned ground vehicle) system named "Open Rover", a customized platform for general-purpose VLN tasks. We integrated and deployed the VL-Explore pipeline on Open Rover to evaluate its throughput, obstacle avoidance capability, and trajectory performance across various real-world scenarios. Experimental results demonstrate that VL-Explore consistently outperforms traditional map-traversal algorithms and achieves performance comparable to path-planning methods that depend on prior map and target knowledge. Notably, VL-Explore offers real-time active navigation without requiring pre-captured candidate images or pre-built node graphs, addressing key limitations of existing VLN pipelines.

Paperid: 4283, https://arxiv.org/pdf/2502.02590.pdf

Abstract:
3D articulated objects modeling has long been a challenging problem, since it requires to capture both accurate surface geometries and semantically meaningful and spatially precise structures, parts, and joints. Existing methods heavily depend on training data from a limited set of handcrafted articulated object categories (e.g., cabinets and drawers), which restricts their ability to model a wide range of articulated objects in an open-vocabulary context. To address these limitations, we propose Articulate Anymesh, an automated framework that is able to convert any rigid 3D mesh into its articulated counterpart in an open-vocabulary manner. Given a 3D mesh, our framework utilizes advanced Vision-Language Models and visual prompting techniques to extract semantic information, allowing for both the segmentation of object parts and the construction of functional joints. Our experiments show that Articulate Anymesh can generate large-scale, high-quality 3D articulated objects, including tools, toys, mechanical devices, and vehicles, significantly expanding the coverage of existing 3D articulated object datasets. Additionally, we show that these generated assets can facilitate the acquisition of new articulated object manipulation skills in simulation, which can then be transferred to a real robotic system. Our Github website is https://articulate-anymesh.github.io.

Paperid: 4284, https://arxiv.org/pdf/2501.16981.pdf

Abstract:
Owing to large-scale image-text contrastive training, pre-trained vision language model (VLM) like CLIP shows superior open-vocabulary recognition ability. Most existing open-vocabulary object detectors attempt to utilize the pre-trained VLMs to attain generalized representation. F-ViT uses the pre-trained visual encoder as the backbone network and freezes it during training. However, its frozen backbone doesn't benefit from the labeled data to strengthen the representation for detection. Therefore, we propose a novel two-branch backbone network, named as \textbf{V}iT-Feature-\textbf{M}odulated Multi-Scale \textbf{C}onvolutional Network (VMCNet), which consists of a trainable convolutional branch, a frozen pre-trained ViT branch and a VMC module. The trainable CNN branch could be optimized with labeled data while the frozen pre-trained ViT branch could keep the representation ability derived from large-scale pre-training. Then, the proposed VMC module could modulate the multi-scale CNN features with the representations from ViT branch. With this proposed mixed structure, the detector is more likely to discover objects of novel categories. Evaluated on two popular benchmarks, our method boosts the detection performance on novel category and outperforms state-of-the-art methods. On OV-COCO, the proposed method achieves 44.3 AP$_{50}^{\mathrm{novel}}$ with ViT-B/16 and 48.5 AP$_{50}^{\mathrm{novel}}$ with ViT-L/14. On OV-LVIS, VMCNet with ViT-B/16 and ViT-L/14 reaches 27.8 and 38.4 mAP$_{r}$.

Paperid: 4285, https://arxiv.org/pdf/2501.13518.pdf

Abstract:
Detecting actions as they occur is essential for applications like video surveillance, autonomous driving, and human-robot interaction. Known as online action detection, this task requires classifying actions in streaming videos, handling background noise, and coping with incomplete actions. Transformer architectures are the current state-of-the-art, yet the potential of recent advancements in computer vision, particularly vision-language models (VLMs), remains largely untapped for this problem, partly due to high computational costs. In this paper, we introduce TOAD: a Text-driven Online Action Detection architecture that supports zero-shot and few-shot learning. TOAD leverages CLIP (Contrastive Language-Image Pretraining) textual embeddings, enabling efficient use of VLMs without significant computational overhead. Our model achieves 82.46% mAP on the THUMOS14 dataset, outperforming existing methods, and sets new baselines for zero-shot and few-shot performance on the THUMOS14 and TVSeries datasets.

Paperid: 4286, https://arxiv.org/pdf/2501.09052.pdf

Abstract:
Single image defocus deblurring (SIDD) aims to restore an all-in-focus image from a defocused one. Distribution shifts in defocused images generally lead to performance degradation of existing methods during out-of-distribution inferences. In this work, we gauge the intrinsic reason behind the performance degradation, which is identified as the heterogeneity of lens-specific point spread functions. Empirical evidence supports this finding, motivating us to employ a continual test-time adaptation (CTTA) paradigm for SIDD. However, traditional CTTA methods, which primarily rely on entropy minimization, cannot sufficiently explore task-dependent information for pixel-level regression tasks like SIDD. To address this issue, we propose a novel Siamese networks-based continual test-time adaptation framework, which adapts source models to continuously changing target domains only requiring unlabeled target data in an online manner. To further mitigate semantically erroneous textures introduced by source SIDD models under severe degradation, we revisit the learning paradigm through a structural causal model and propose Causal Siamese networks (CauSiam). Our method leverages large-scale pre-trained vision-language models to derive discriminative universal semantic priors and integrates these priors into Siamese networks, ensuring causal identifiability between blurry inputs and restored images. Extensive experiments demonstrate that CauSiam effectively improves the generalization performance of existing SIDD methods in continuously changing domains.

Paperid: 4287, https://arxiv.org/pdf/2501.05733.pdf

Abstract:
The application of Multi-modal Large Language Models (MLLMs) in Autonomous Driving (AD) faces significant challenges due to their limited training on traffic-specific data and the absence of dedicated benchmarks for spatiotemporal understanding. This study addresses these issues by proposing TB-Bench, a comprehensive benchmark designed to evaluate MLLMs on understanding traffic behaviors across eight perception tasks from ego-centric views. We also introduce vision-language instruction tuning datasets, TB-100k and TB-250k, along with simple yet effective baselines for the tasks. Through extensive experiments, we show that existing MLLMs underperform in these tasks, with even a powerful model like GPT-4o achieving less than 35% accuracy on average. In contrast, when fine-tuned with TB-100k or TB-250k, our baseline models achieve average accuracy up to 85%, significantly enhancing performance on the tasks. Additionally, we demonstrate performance transfer by co-training TB-100k with another traffic dataset, leading to improved performance on the latter. Overall, this study represents a step forward by introducing a comprehensive benchmark, high-quality datasets, and baselines, thus supporting the gradual integration of MLLMs into the perception, prediction, and planning stages of AD.

Paperid: 4288, https://arxiv.org/pdf/2412.17811.pdf

Abstract:
We introduce ChatGarment, a novel approach that leverages large vision-language models (VLMs) to automate the estimation, generation, and editing of 3D garments from images or text descriptions. Unlike previous methods that struggle in real-world scenarios or lack interactive editing capabilities, ChatGarment can estimate sewing patterns from in-the-wild images or sketches, generate them from text descriptions, and edit garments based on user instructions, all within an interactive dialogue. These sewing patterns can then be draped on a 3D body and animated. This is achieved by finetuning a VLM to directly generate a JSON file that includes both textual descriptions of garment types and styles, as well as continuous numerical attributes. This JSON file is then used to create sewing patterns through a programming parametric model. To support this, we refine the existing programming model, GarmentCode, by expanding its garment type coverage and simplifying its structure for efficient VLM fine-tuning. Additionally, we construct a large-scale dataset of image-to-sewing-pattern and text-to-sewing-pattern pairs through an automated data pipeline. Extensive evaluations demonstrate ChatGarment's ability to accurately reconstruct, generate, and edit garments from multimodal inputs, highlighting its potential to simplify workflows in fashion and gaming applications. Code and data are available at https://chatgarment.github.io/ .

Paperid: 4289, https://arxiv.org/pdf/2412.17251.pdf

Abstract:
Retinal image analysis is crucial for diagnosing and treating eye diseases, yet generating accurate medical reports from images remains challenging due to variability in image quality and pathology, especially with limited labeled data. Previous Transformer-based models struggled to integrate visual and textual information under limited supervision. In response, we propose a novel vision-language model for retinal image captioning that combines visual and textual features through a guided context self-attention mechanism. This approach captures both intricate details and the global clinical context, even in data-scarce scenarios. Extensive experiments on the DeepEyeNet dataset demonstrate a 0.023 BLEU@4 improvement, along with significant qualitative advancements, highlighting the effectiveness of our model in generating comprehensive medical captions.

Paperid: 4290, https://arxiv.org/pdf/2412.11484.pdf

Abstract:
For embodied reinforcement learning (RL) agents interacting with the environment, it is desirable to have rapid policy adaptation to unseen visual observations, but achieving zero-shot adaptation capability is considered as a challenging problem in the RL context. To address the problem, we present a novel contrastive prompt ensemble (ConPE) framework which utilizes a pretrained vision-language model and a set of visual prompts, thus enabling efficient policy learning and adaptation upon a wide range of environmental and physical changes encountered by embodied agents. Specifically, we devise a guided-attention-based ensemble approach with multiple visual prompts on the vision-language model to construct robust state representations. Each prompt is contrastively learned in terms of an individual domain factor that significantly affects the agent's egocentric perception and observation. For a given task, the attention-based ensemble and policy are jointly learned so that the resulting state representations not only generalize to various domains but are also optimized for learning the task. Through experiments, we show that ConPE outperforms other state-of-the-art algorithms for several embodied agent tasks including navigation in AI2THOR, manipulation in egocentric-Metaworld, and autonomous driving in CARLA, while also improving the sample efficiency of policy learning and adaptation.

Paperid: 4291, https://arxiv.org/pdf/2412.10843.pdf

Abstract:
Multi-label recognition with partial labels (MLR-PL), in which only some labels are known while others are unknown for each image, is a practical task in computer vision, since collecting large-scale and complete multi-label datasets is difficult in real application scenarios. Recently, vision language models (e.g. CLIP) have demonstrated impressive transferability to downstream tasks in data limited or label limited settings. However, current CLIP-based methods suffer from semantic confusion in MLR task due to the lack of fine-grained information in the single global visual and textual representation for all categories. In this work, we address this problem by introducing a semantic decoupling module and a category-specific prompt optimization method in CLIP-based framework. Specifically, the semantic decoupling module following the visual encoder learns category-specific feature maps by utilizing the semantic-guided spatial attention mechanism. Moreover, the category-specific prompt optimization method is introduced to learn text representations aligned with category semantics. Therefore, the prediction of each category is independent, which alleviate the semantic confusion problem. Extensive experiments on Microsoft COCO 2014 and Pascal VOC 2007 datasets demonstrate that the proposed framework significantly outperforms current state-of-art methods with a simpler model structure. Additionally, visual analysis shows that our method effectively separates information from different categories and achieves better performance compared to CLIP-based baseline method.

Paperid: 4292, https://arxiv.org/pdf/2412.07619.pdf

Abstract:
Recently, large language models (LLMs) have demonstrated impressive capabilities in dealing with new tasks with the help of in-context learning (ICL). In the study of Large Vision-Language Models (LVLMs), when implementing ICL, researchers usually adopts the naive strategies like fixed demonstrations across different samples, or selecting demonstrations directly via a visual-language embedding model. These methods does not guarantee the configured demonstrations fit the need of the LVLMs. To address this issue, we now propose a novel framework, \underline{d}emonstration \underline{r}etriever for large m\underline{u}lti-modal \underline{m}odel (DRUM), which fine-tunes the visual-language embedding model to better meet the LVLM's needs. First, we discuss the retrieval strategies for a visual-language task, assuming an embedding model is given. And we propose to concate the image and text embeddings to enhance the retrieval performance. Second, we propose to re-rank the demonstrations retrieved by the embedding model via the LVLM's feedbacks, and calculate a list-wise ranking loss for training the embedding model. Third, we propose an iterative demonstration mining strategy to improve the training of the embedding model. Through extensive experiments on 3 types of visual-language tasks, 7 benchmark datasets, our DRUM framework is proven to be effective in boosting the LVLM's in-context learning performance via retrieving more proper demonstrations.

Paperid: 4293, https://arxiv.org/pdf/2412.02978.pdf

Abstract:
Pathological cell semantic segmentation is a fundamental technology in computational pathology, essential for applications like cancer diagnosis and effective treatment. Given that multiple cell types exist across various organs, with subtle differences in cell size and shape, multi-organ, multi-class cell segmentation is particularly challenging. Most existing methods employ multi-branch frameworks to enhance feature extraction, but often result in complex architectures. Moreover, reliance on visual information limits performance in multi-class analysis due to intricate textural details. To address these challenges, we propose a Multi-OrgaN multi-Class cell semantic segmentation method with a single brancH (MONCH) that leverages vision-language input. Specifically, we design a hierarchical feature extraction mechanism to provide coarse-to-fine-grained features for segmenting cells of various shapes, including high-frequency, convolutional, and topological features. Inspired by the synergy of textual and multi-grained visual features, we introduce a progressive prompt decoder to harmonize multimodal information, integrating features from fine to coarse granularity for better context capture. Extensive experiments on the PanNuke dataset, which has significant class imbalance and subtle cell size and shape variations, demonstrate that MONCH outperforms state-of-the-art cell segmentation methods and vision-language models. Codes and implementations will be made publicly available.

Paperid: 4294, https://arxiv.org/pdf/2411.08579.pdf

Abstract:
Vision-and-Language Navigation (VLN), as a widely discussed research direction in embodied intelligence, aims to enable embodied agents to navigate in complicated visual environments through natural language commands. Most existing VLN methods focus on indoor ground robot scenarios. However, when applied to UAV VLN in outdoor urban scenes, it faces two significant challenges. First, urban scenes contain numerous objects, which makes it challenging to match fine-grained landmarks in images with complex textual descriptions of these landmarks. Second, overall environmental information encompasses multiple modal dimensions, and the diversity of representations significantly increases the complexity of the encoding process. To address these challenges, we propose NavAgent, the first urban UAV embodied navigation model driven by a large Vision-Language Model. NavAgent undertakes navigation tasks by synthesizing multi-scale environmental information, including topological maps (global), panoramas (medium), and fine-grained landmarks (local). Specifically, we utilize GLIP to build a visual recognizer for landmark capable of identifying and linguisticizing fine-grained landmarks. Subsequently, we develop dynamically growing scene topology map that integrate environmental information and employ Graph Convolutional Networks to encode global environmental data. In addition, to train the visual recognizer for landmark, we develop NavAgent-Landmark2K, the first fine-grained landmark dataset for real urban street scenes. In experiments conducted on the Touchdown and Map2seq datasets, NavAgent outperforms strong baseline models. The code and dataset will be released to the community to facilitate the exploration and development of outdoor VLN.

Paperid: 4295, https://arxiv.org/pdf/2411.05261.pdf

Abstract:
Despite significant advancements in automated report generation, the opaqueness of text interpretability continues to cast doubt on the reliability of the content produced. This paper introduces a novel approach to identify specific image features in X-ray images that influence the outputs of report generation models. Specifically, we propose Cyclic Vision-Language Manipulator CVLM, a module to generate a manipulated X-ray from an original X-ray and its report from a designated report generator. The essence of CVLM is that cycling manipulated X-rays to the report generator produces altered reports aligned with the alterations pre-injected into the reports for X-ray generation, achieving the term "cyclic manipulation". This process allows direct comparison between original and manipulated X-rays, clarifying the critical image features driving changes in reports and enabling model users to assess the reliability of the generated texts. Empirical evaluations demonstrate that CVLM can identify more precise and reliable features compared to existing explanation methods, significantly enhancing the transparency and applicability of AI-generated reports.

Paperid: 4296, https://arxiv.org/pdf/2411.04420.pdf

Abstract:
Vision-language model (VLM) embeddings have been shown to encode biases present in their training data, such as societal biases that prescribe negative characteristics to members of various racial and gender identities. VLMs are being quickly adopted for a variety of tasks ranging from few-shot classification to text-guided image generation, making debiasing VLM embeddings crucial. Debiasing approaches that fine-tune the VLM often suffer from catastrophic forgetting. On the other hand, fine-tuning-free methods typically utilize a "one-size-fits-all" approach that assumes that correlation with the spurious attribute can be explained using a single linear direction across all possible inputs. In this work, we propose Bend-VLM, a nonlinear, fine-tuning-free approach for VLM embedding debiasing that tailors the debiasing operation to each unique input. This allows for a more flexible debiasing approach. Additionally, we do not require knowledge of the set of inputs a priori to inference time, making our method more appropriate for online, open-set tasks such as retrieval and text guided image generation.

Paperid: 4297, https://arxiv.org/pdf/2411.00997.pdf

Abstract:
Vision-language models, like CLIP (Contrastive Language Image Pretraining), are becoming increasingly popular for a wide range of multimodal retrieval tasks. However, prior work has shown that large language and deep vision models can learn historical biases contained in their training sets, leading to perpetuation of stereotypes and potential downstream harm. In this work, we conduct a systematic analysis of the social biases that are present in CLIP, with a focus on the interaction between image and text modalities. We first propose a taxonomy of social biases called So-B-IT, which contains 374 words categorized across ten types of bias. Each type can lead to societal harm if associated with a particular demographic group. Using this taxonomy, we examine images retrieved by CLIP from a facial image dataset using each word as part of a prompt. We find that CLIP frequently displays undesirable associations between harmful words and specific demographic groups, such as retrieving mostly pictures of Middle Eastern men when asked to retrieve images of a "terrorist". Finally, we conduct an analysis of the source of such biases, by showing that the same harmful stereotypes are also present in a large image-text dataset used to train CLIP models for examples of biases that we find. Our findings highlight the importance of evaluating and addressing bias in vision-language models, and suggest the need for transparency and fairness-aware curation of large pre-training datasets.

Paperid: 4298, https://arxiv.org/pdf/2410.23218.pdf

Abstract:
Existing efforts in building GUI agents heavily rely on the availability of robust commercial Vision-Language Models (VLMs) such as GPT-4o and GeminiProVision. Practitioners are often reluctant to use open-source VLMs due to their significant performance lag compared to their closed-source counterparts, particularly in GUI grounding and Out-Of-Distribution (OOD) scenarios. To facilitate future research in this area, we developed OS-Atlas - a foundational GUI action model that excels at GUI grounding and OOD agentic tasks through innovations in both data and modeling. We have invested significant engineering effort in developing an open-source toolkit for synthesizing GUI grounding data across multiple platforms, including Windows, Linux, MacOS, Android, and the web. Leveraging this toolkit, we are releasing the largest open-source cross-platform GUI grounding corpus to date, which contains over 13 million GUI elements. This dataset, combined with innovations in model training, provides a solid foundation for OS-Atlas to understand GUI screenshots and generalize to unseen interfaces. Through extensive evaluation across six benchmarks spanning three different platforms (mobile, desktop, and web), OS-Atlas demonstrates significant performance improvements over previous state-of-the-art models. Our evaluation also uncovers valuable insights into continuously improving and scaling the agentic capabilities of open-source VLMs.

Paperid: 4299, https://arxiv.org/pdf/2410.13851.pdf

Abstract:
Vision foundation models trained on massive amounts of visual data have shown unprecedented reasoning and planning skills in open-world settings. A key challenge in applying them to robotic tasks is the modality gap between visual data and action data. We introduce differentiable robot rendering, a method allowing the visual appearance of a robot body to be directly differentiable with respect to its control parameters. Our model integrates a kinematics-aware deformable model and Gaussians Splatting and is compatible with any robot form factors and degrees of freedom. We demonstrate its capability and usage in applications including reconstruction of robot poses from images and controlling robots through vision language models. Quantitative and qualitative results show that our differentiable rendering model provides effective gradients for robotic control directly from pixels, setting the foundation for the future applications of vision foundation models in robotics.

Paperid: 4300, https://arxiv.org/pdf/2410.02681.pdf

Abstract:
Confidence calibration is critical for the safe deployment of machine learning models in the real world. However, such issue in vision-language models like CLIP, particularly after fine-tuning, has not been fully addressed. In this work, we demonstrate that existing prompt tuning methods usually lead to a trade-off of calibration between base and new classes: the cross-entropy loss in CoOp causes overconfidence in new classes by increasing textual label divergence, whereas the regularization of KgCoOp maintains the confidence level but results in underconfidence in base classes due to the improved accuracy. Inspired by the observations, we introduce Dynamic Outlier Regularization (DOR) to ensure the confidence calibration on both base and new classes after fine-tuning. In particular, we propose to minimize the feature deviation of novel textual labels (instead of base classes) sampled from a large vocabulary. In effect, DOR prevents the increase in textual divergence for new labels while easing restrictions on base classes. Extensive experiments demonstrate that DOR can enhance the calibration performance of current fine-tuning methods on base and new classes.

Paperid: 4301, https://arxiv.org/pdf/2410.00713.pdf

Abstract:
Anomaly detection is a core capability for robotic perception and industrial inspection, yet most existing benchmarks are collected under controlled conditions with fixed viewpoints and stable illumination, failing to reflect real deployment scenarios. We introduce RAD (Realistic Anomaly Detection), a robot-captured, multi-view dataset designed to stress pose variation, reflective materials, and viewpoint-dependent defect visibility. RAD covers 13 everyday object categories and four realistic defect types--scratched, missing, stained, and squeezed--captured from over 60 robot viewpoints per object under uncontrolled lighting. We benchmark a wide range of state-of-the-art approaches, including 2D feature-based methods, 3D reconstruction pipelines, and vision-language models (VLMs), under a pose-agnostic setting. Surprisingly, we find that mature 2D feature-embedding methods consistently outperform recent 3D and VLM-based approaches at the image level, while the performance gap narrows for pixel-level localization. Our analysis reveals that reflective surfaces, geometric symmetry, and sparse viewpoint coverage fundamentally limit current geometry-based and zero-shot methods. RAD establishes a challenging and realistic benchmark for robotic anomaly detection, highlighting critical open problems beyond controlled laboratory settings.

Paperid: 4302, https://arxiv.org/pdf/2409.14674.pdf

Abstract:
Developing robust and correctable visuomotor policies for robotic manipulation is challenging due to the lack of self-recovery mechanisms from failures and the limitations of simple language instructions in guiding robot actions. To address these issues, we propose a scalable data generation pipeline that automatically augments expert demonstrations with failure recovery trajectories and fine-grained language annotations for training. We then introduce Rich languAge-guided failure reCovERy (RACER), a supervisor-actor framework, which combines failure recovery data with rich language descriptions to enhance robot control. RACER features a vision-language model (VLM) that acts as an online supervisor, providing detailed language guidance for error correction and task execution, and a language-conditioned visuomotor policy as an actor to predict the next actions. Our experimental results show that RACER outperforms the state-of-the-art Robotic View Transformer (RVT) on RLbench across various evaluation settings, including standard long-horizon tasks, dynamic goal-change tasks and zero-shot unseen tasks, achieving superior performance in both simulated and real world environments. Videos and code are available at: https://rich-language-failure-recovery.github.io.

Paperid: 4303, https://arxiv.org/pdf/2409.06078.pdf

Abstract:
Networked robotic systems balance compute, power, and latency constraints in applications such as self-driving vehicles, drone swarms, and teleoperated surgery. A core problem in this domain is deciding when to offload a computationally expensive task to the cloud, a remote server, at the cost of communication latency. Task offloading algorithms often rely on precise knowledge of system-specific performance metrics, such as sensor data rates, network bandwidth, and machine learning model latency. While these metrics can be modeled during system design, uncertainties in connection quality, server load, and hardware conditions introduce real-time performance variations, hindering overall performance. We introduce PEERNet, an end-to-end and real-time profiling tool for cloud robotics. PEERNet enables performance monitoring on heterogeneous hardware through targeted yet adaptive profiling of system components such as sensors, networks, deep-learning pipelines, and devices. We showcase PEERNet's capabilities through networked robotics tasks, such as image-based teleoperation of a Franka Emika Panda arm and querying vision language models using an Nvidia Jetson Orin. PEERNet reveals non-intuitive behavior in robotic systems, such as asymmetric network transmission and bimodal language model output. Our evaluation underscores the effectiveness and importance of benchmarking in networked robotics, demonstrating PEERNet's adaptability. Our code is open-source and available at github.com/UTAustin-SwarmLab/PEERNet.

Paperid: 4304, https://arxiv.org/pdf/2601.21662.pdf

Abstract:
Vision-Language Models (VLMs) are typically deterministic in nature and lack intrinsic mechanisms to quantify epistemic uncertainty, which reflects the model's lack of knowledge or ignorance of its own representations. We theoretically motivate negative log-density of an embedding as a proxy for the epistemic uncertainty, where low-density regions signify model ignorance. The proposed method REPVLM computes the probability density on the hyperspherical manifold of the VLM embeddings using Riemannian Flow Matching. We empirically demonstrate that REPVLM achieves near-perfect correlation between uncertainty and prediction error, significantly outperforming existing baselines. Beyond classification, we also demonstrate that the model also provides a scalable metric for out-of-distribution detection and automated data curation.

Paperid: 4305, https://arxiv.org/pdf/2601.19712.pdf

Abstract:
Spatial audio is essential for immersive experiences, yet novel-view acoustic synthesis (NVAS) remains challenging due to complex physical phenomena such as reflection, diffraction, and material absorption. Existing methods based on single-view or panoramic inputs improve spatial fidelity but fail to capture global geometry and semantic cues such as object layout and material properties. To address this, we propose Phys-NVAS, the first physics-aware NVAS framework that integrates spatial geometry modeling with vision-language semantic priors. A global 3D acoustic environment is reconstructed from multi-view images and depth maps to estimate room size and shape, enhancing spatial awareness of sound propagation. Meanwhile, a vision-language model extracts physics-aware priors of objects, layouts, and materials, capturing absorption and reflection beyond geometry. An acoustic feature fusion adapter unifies these cues into a physics-aware representation for binaural generation. Experiments on RWAVS demonstrate that Phys-NVAS yields binaural audio with improved realism and physical consistency.

Paperid: 4306, https://arxiv.org/pdf/2601.12946.pdf

Abstract:
Generative artificial intelligence (AI) is rapidly populating medical records with synthetic content, creating a feedback loop where future models are increasingly at risk of training on uncurated AI-generated data. However, the clinical consequences of this AI-generated data contamination remain unexplored. Here, we show that in the absence of mandatory human verification, this self-referential cycle drives a rapid erosion of pathological variability and diagnostic reliability. By analysing more than 800,000 synthetic data points across clinical text generation, vision-language reporting, and medical image synthesis, we find that models progressively converge toward generic phenotypes regardless of the model architecture. Specifically, rare but critical findings, including pneumothorax and effusions, vanish from the synthetic content generated by AI models, while demographic representations skew heavily toward middle-aged male phenotypes. Crucially, this degradation is masked by false diagnostic confidence; models continue to issue reassuring reports while failing to detect life-threatening pathology, with false reassurance rates tripling to 40%. Blinded physician evaluation confirms that this decoupling of confidence and accuracy renders AI-generated documentation clinically useless after just two generations. We systematically evaluate three mitigation strategies, finding that while synthetic volume scaling fails to prevent collapse, mixing real data with quality-aware filtering effectively preserves diversity. Ultimately, our results suggest that without policy-mandated human oversight, the deployment of generative AI threatens to degrade the very healthcare data ecosystems it relies upon.

Paperid: 4307, https://arxiv.org/pdf/2601.11525.pdf

Abstract:
Recent advances in vision-language models (VLMs) have expanded their multimodal code generation capabilities, yet their ability to generate executable visualization code from plots, especially for complex 3D, animated, plot-to-plot transformations, or multi-library scenarios, remains underexplored. To address this gap, we introduce PlotGen-Bench, a comprehensive benchmark for evaluating plot-to-code generation under realistic and complex visualization scenarios. The benchmark spans 9 major categories, 30 subcategories, and 3 core tasks-plot replication, plot transformation, and multi-library generation, covering both 2D, 3D and animated plots across 5 widely used visualization libraries. Through systematic evaluation of state-of-the-art open- and closed-source VLMs, we find that open-source models still lag considerably behind in visual fidelity and semantic consistency, despite achieving comparable code executability. Moreover, all models exhibit substantial degradation on reasoning-intensive tasks such as chart type conversion and animation generation. PlotGen-Bench establishes a rigorous foundation for advancing research toward more capable and reliable VLMs for visualization authoring and code synthesis, with all data and code available at https://plotgen.github.io.

Paperid: 4308, https://arxiv.org/pdf/2601.09879.pdf

Abstract:
Recent progress in medical vision-language models (VLMs) has achieved strong performance on image-level text-centric tasks such as report generation and visual question answering (VQA). However, achieving fine-grained visual grounding and volumetric spatial reasoning in 3D medical VLMs remains challenging, particularly when aiming to unify these capabilities within a single, generalizable framework. To address this challenge, we proposed MedVL-SAM2, a unified 3D medical multimodal model that concurrently supports report generation, VQA, and multi-paradigm segmentation, including semantic, referring, and interactive segmentation. MedVL-SAM2 integrates image-level reasoning and pixel-level perception through a cohesive architecture tailored for 3D medical imaging, and incorporates a SAM2-based volumetric segmentation module to enable precise multi-granular spatial reasoning. The model is trained in a multi-stage pipeline: it is first pre-trained on a large-scale corpus of 3D CT image-text pairs to align volumetric visual features with radiology-language embeddings. It is then jointly optimized with both language-understanding and segmentation objectives using a comprehensive 3D CT segmentation dataset. This joint training enables flexible interaction via language, point, or box prompts, thereby unifying high-level visual reasoning with spatially precise localization. Our unified architecture delivers state-of-the-art performance across report generation, VQA, and multiple 3D segmentation tasks. Extensive analyses further show that the model provides reliable 3D visual grounding, controllable interactive segmentation, and robust cross-modal reasoning, demonstrating that high-level semantic reasoning and precise 3D localization can be jointly achieved within a unified 3D medical VLM.

Paperid: 4309, https://arxiv.org/pdf/2601.07986.pdf

Abstract:
We introduce VULCA-Bench, a multicultural art-critique benchmark for evaluating Vision-Language Models' (VLMs) cultural understanding beyond surface-level visual perception. Existing VLM benchmarks predominantly measure L1-L2 capabilities (object recognition, scene description, and factual question answering) while under-evaluate higher-order cultural interpretation. VULCA-Bench contains 7,410 matched image-critique pairs spanning eight cultural traditions, with Chinese-English bilingual coverage. We operationalise cultural understanding using a five-layer framework (L1-L5, from Visual Perception to Philosophical Aesthetics), instantiated as 225 culture-specific dimensions and supported by expert-written bilingual critiques. Our pilot results indicate that higher-layer reasoning (L3-L5) is consistently more challenging than visual and technical analysis (L1-L2). The dataset, evaluation scripts, and annotation tools are available under CC BY 4.0 in the supplementary materials.

Paperid: 4310, https://arxiv.org/pdf/2601.07984.pdf

Abstract:
Vision-Language Models (VLMs) excel at visual perception, yet their ability to interpret cultural meaning in art remains under-validated. We present a tri-tier evaluation framework for cross-cultural art-critique assessment: Tier I computes automated coverage and risk indicators offline; Tier II applies rubric-based scoring using a single primary judge across five dimensions; and Tier III calibrates the Tier II aggregate score to human ratings via isotonic regression, yielding a 5.2% reduction in MAE on a 152-sample held-out set. The framework outputs a calibrated cultural-understanding score for model selection and cultural-gap diagnosis, together with dimension-level diagnostics and risk indicators. We evaluate 15 VLMs on 294 expert anchors spanning six cultural traditions. Key findings are that (i) automated metrics are unreliable proxies for cultural depth, (ii) Western samples score higher than non-Western samples under our sampling and rubric, and (iii) cross-judge scale mismatch makes naive score averaging unreliable, motivating a single primary judge with explicit calibration. Dataset and code are available in the supplementary materials.

Paperid: 4311, https://arxiv.org/pdf/2601.07315.pdf

Abstract:
Analog mixed-signal circuit sizing involves complex trade-offs within high-dimensional design spaces. Existing automatic analog circuit sizing approaches rely solely on netlists, ignoring the circuit schematic, which hinders the cognitive link between the schematic and its performance. Furthermore, the black-box nature of machine learning methods and hallucination risks in large language models fail to provide the necessary ground-truth explainability required for industrial sign-off. To address these challenges, we propose a Vision Language Model-optimized collaborative agent design workflow (VLM-CAD), which analyzes circuits, optimizes DC operating points, performs inference-based sizing, and executes external sizing optimization. We integrate Image2Net to annotate circuit schematics and generate a structured JSON description for precise interpretation by Vision Language Models. Furthermore, we propose an Explainable Trust Region Bayesian Optimization method (ExTuRBO) that employs collaborative warm-start from agent-generated seeds and offers dual-granularity sensitivity analysis for external sizing optimization, supporting a comprehensive final design report. Experiment results on amplifier sizing tasks using 180nm, 90nm, and 45nm Predictive Technology Models demonstrate that VLM-CAD effectively balances power and performance while maintaining physics-based explainability. VLM-CAD meets all specification requirements while maintaining low power consumption in optimizing an amplifier with a complementary input and a class-AB output stage, with a total runtime under 66 minutes across all experiments on two amplifiers.

Paperid: 4312, https://arxiv.org/pdf/2601.03519.pdf

Abstract:
Efficient trajectory planning in off-road terrains presents a formidable challenge for autonomous vehicles, often necessitating complex multi-step pipelines. However, traditional approaches exhibit limited adaptability in dynamic environments. To address these limitations, this paper proposes OFF-EMMA, a novel end-to-end multimodal framework designed to overcome the deficiencies of insufficient spatial perception and unstable reasoning in visual-language-action (VLA) models for off-road autonomous driving scenarios. The framework explicitly annotates input images through the design of a visual prompt block and introduces a chain-of-thought with self-consistency (COT-SC) reasoning strategy to enhance the accuracy and robustness of trajectory planning. The visual prompt block utilizes semantic segmentation masks as visual prompts, enhancing the spatial understanding ability of pre-trained visual-language models for complex terrains. The COT- SC strategy effectively mitigates the error impact of outliers on planning performance through a multi-path reasoning mechanism. Experimental results on the RELLIS-3D off-road dataset demonstrate that OFF-EMMA significantly outperforms existing methods, reducing the average L2 error of the Qwen backbone model by 13.3% and decreasing the failure rate from 16.52% to 6.56%.

Paperid: 4313, https://arxiv.org/pdf/2512.24561.pdf

Abstract:
Visual Grounding (VG) aims to localize specific objects in an image according to natural language expressions, serving as a fundamental task in vision-language understanding. However, existing VG benchmarks are mostly derived from datasets collected under clean environments, such as COCO, where scene diversity is limited. Consequently, they fail to reflect the complexity of real-world conditions, such as changes in illumination, weather, etc., that are critical to evaluating model robustness and generalization in safety-critical applications. To address these limitations, we present RGBT-Ground, the first large-scale visual grounding benchmark built for complex real-world scenarios. It consists of spatially aligned RGB and Thermal infrared (TIR) image pairs with high-quality referring expressions, corresponding object bounding boxes, and fine-grained annotations at the scene, environment, and object levels. This benchmark enables comprehensive evaluation and facilitates the study of robust grounding under diverse and challenging conditions. Furthermore, we establish a unified visual grounding framework that supports both uni-modal (RGB or TIR) and multi-modal (RGB-TIR) visual inputs. Based on it, we propose RGBT-VGNet, a simple yet effective baseline for fusing complementary visual modalities to achieve robust grounding. We conduct extensive adaptations to the existing methods on RGBT-Ground. Experimental results show that our proposed RGBT-VGNet significantly outperforms these adapted methods, particularly in nighttime and long-distance scenarios. All resources will be publicly released to promote future research on robust visual grounding in complex real-world environments.

Paperid: 4314, https://arxiv.org/pdf/2512.18679.pdf

Abstract:
We present brat (brain report alignment transformer), a multi-view representation learning framework for brain magnetic resonance imaging (MRI) trained on MRIs paired with clinical reports. Brain MRIs present unique challenges due to the presence of numerous, highly varied, and often subtle abnormalities that are localized to a few slices within a 3D volume. To address these challenges, we introduce a brain MRI dataset $10\times$ larger than existing ones, containing approximately 80,000 3D scans with corresponding radiology reports, and propose a multi-view pre-training approach inspired by advances in document retrieval. We develop an implicit query-feature matching mechanism and adopt concepts from quality-diversity to obtain multi-view embeddings of MRIs that are aligned with the clinical features given by report sentences. We evaluate our approach across multiple vision-language and vision tasks, demonstrating substantial performance improvements. The brat foundation models are publicly released.

Paperid: 4315, https://arxiv.org/pdf/2512.18504.pdf

Abstract:
Vision-language models (VLMs) struggle in open-world applications, where out-of-distribution (OOD) concepts often trigger cross-modal alignment collapse and severely degrade zero-shot performance. We identify the root cause as modal asymmetry: while the visual encoder can extract discriminative features from unseen images, the text encoder is constrained by a fixed discrete vocabulary and cannot synthesize new semantic anchors. Existing approaches such as CoOp or LoRA provide only partial remedies, as they remain confined to the pre-trained semantic space. To overcome this bottleneck, we propose dynamic representation optimization, realized through the Guided Target-Matching Adaptation (GTMA) framework. At inference time, GTMA constructs a continuous pseudo-word embedding that best aligns with an OOD image's visual anchor, effectively bypassing vocabulary limitations. The optimization is driven by an adaptive gradient-based representation policy optimization algorithm, which incorporates semantic regularization to preserve plausibility and compatibility with the model's prior knowledge. Experiments on ImageNet-R and the VISTA-Beyond benchmark demonstrate that GTMA improves zero-shot and few-shot OOD accuracy by up to 15-20 percent over the base VLM while maintaining performance on in-distribution concepts. Ablation studies further confirm the necessity of pseudo-word optimization.

Paperid: 4316, https://arxiv.org/pdf/2512.17313.pdf

Abstract:
Despite the impressive zero-shot capabilities of Vision-Language Models (VLMs), they often struggle in downstream tasks with distribution shifts from the pre-training data. Few-Shot Adaptation (FSA-VLM) has emerged as a key solution, typically using Parameter-Efficient Fine-Tuning (PEFT) to adapt models with minimal data. However, these PEFT methods are constrained by their reliance on fixed, handcrafted prompts, which are often insufficient to understand the semantics of classes. While some studies have proposed leveraging image-induced prompts to provide additional clues for classification, they introduce prohibitive computational overhead at inference. Therefore, we introduce Auxiliary Descriptive Knowledge (ADK), a novel framework that efficiently enriches text representations without compromising efficiency. ADK first leverages a Large Language Model to generate a rich set of descriptive prompts for each class offline. These pre-computed features are then deployed in two ways: (1) as Compositional Knowledge, an averaged representation that provides rich semantics, especially beneficial when class names are ambiguous or unfamiliar to the VLM; and (2) as Instance-Specific Knowledge, where a lightweight, non-parametric attention mechanism dynamically selects the most relevant descriptions for a given image. This approach provides two additional types of knowledge alongside the handcrafted prompt, thereby facilitating category distinction across various domains. Also, ADK acts as a parameter-free, plug-and-play component that enhances existing PEFT methods. Extensive experiments demonstrate that ADK consistently boosts the performance of multiple PEFT baselines, setting a new state-of-the-art across various scenarios.

Paperid: 4317, https://arxiv.org/pdf/2512.15926.pdf

Abstract:
Generative models are often deployed to make decisions on behalf of users, such as vision-language models (VLMs) identifying which person in a room is a doctor to help visually impaired individuals. Yet, VLM decisions are influenced by the perceived demographic attributes of people in the input, which can lead to biased outcomes like failing to identify women as doctors. Moreover, when reducing bias leads to performance loss, users may have varying needs for balancing bias mitigation with overall model capabilities, highlighting the demand for methods that enable controllable bias reduction during inference. Activation steering is a popular approach for inference-time controllability that has shown potential in inducing safer behavior in large language models (LLMs). However, we observe that current steering methods struggle to correct biases, where equiprobable outcomes across demographic groups are required. To address this, we propose Direct Steering Optimization (DSO) which uses reinforcement learning to find linear transformations for steering activations, tailored to mitigate bias while maintaining control over model performance. We demonstrate that DSO achieves state-of-the-art trade-off between fairness and capabilities on both VLMs and LLMs, while offering practitioners inference-time control over the trade-off. Overall, our work highlights the benefit of designing steering strategies that are directly optimized to control model behavior, providing more effective bias intervention than methods that rely on pre-defined heuristics for controllability.

Paperid: 4318, https://arxiv.org/pdf/2512.06663.pdf

Abstract:
Large Vision-Language Models (LVLMs) have demonstrated remarkable success in a broad range of vision-language tasks, such as general visual question answering and optical character recognition (OCR). However, their performance on perception-centric tasks -- such as object detection, semantic segmentation, and depth estimation -- remains significantly inferior to that of task-specific expert models. For example, Qwen2.5-VL-7B-Instruct achieves only 19% mAP on COCO2017 val, particularly struggling with dense scenes and small object recall. In this work, we introduce Chain-of-Thought for Detection (CoT4Det), a simple but efficient strategy that reformulates perception tasks into three interpretable steps: classification, counting, and grounding -- each more naturally aligned with the reasoning capabilities of LVLMs. Extensive experiments demonstrate that our method significantly improves perception performance without compromising general vision language capabilities. With a standard Qwen2.5-VL-7B-Instruct, CoT4Det boosts mAP from 19.0% to 33.0% on COCO2017 val and achieves competitive results across a variety of perception benchmarks, outperforming baselines by +2% on RefCOCO series and 19% on Flickr30k entities.

Paperid: 4319, https://arxiv.org/pdf/2512.03794.pdf

Abstract:
Vision-Language Models (VLMs) have achieved remarkable success in visual question answering tasks, but their reliance on large numbers of visual tokens introduces significant computational overhead. While existing efficient VLM approaches reduce visual tokens through fixed-ratio compression, they operate passively and lack the ability to adapt to varying task requirements. This motivates a fundamental question: Can VLMs autonomously determine the minimum number of visual tokens required for each sample? Inspired by human active vision mechanisms, we introduce AdaptVision, an efficient VLM paradigm that enables adaptive visual token acquisition through a coarse-to-fine approach. Our model initially processes compressed visual tokens from low-resolution images and selectively acquires additional visual information by invoking a bounding box tool to crop key regions when necessary. We train AdaptVision using a reinforcement learning framework that carefully balances accuracy and efficiency. Central to our approach is Decoupled Turn Policy Optimization (DTPO), which decouples the learning objective into two components: (1) tool learning, which optimizes correct tool utilization, and (2) accuracy improvement, which refines the generated responses to improve answer correctness. Based on this formulation, we further decouple advantage estimation by computing separate advantages for tokens associated with each objective. This formulation enables more effective optimization for AdaptVision compared to vanilla GRPO. Comprehensive experiments across multiple VQA benchmarks demonstrate that AdaptVision achieves superior performance while consuming substantially fewer visual tokens than state-of-the-art efficient VLM methods.

Paperid: 4320, https://arxiv.org/pdf/2512.03087.pdf

Abstract:
Large vision-language models (LVLMs) are increasingly used for tasks where detecting multimodal harmful content is crucial, such as online content moderation. However, real-world harmful content is often camouflaged, relying on nuanced text-image interplay, such as memes or images with embedded malicious text, to evade detection. This raises a key question: \textbf{can LVLMs perceive such camouflaged harmful content as sensitively as humans do?} In this paper, we introduce CamHarmTI, a benchmark for evaluating LVLM ability to perceive and interpret camouflaged harmful content within text-image compositions. CamHarmTI consists of over 4,500 samples across three types of image-text posts. Experiments on 100 human users and 12 mainstream LVLMs reveal a clear perceptual gap: humans easily recognize such content (e.g., over 95.75\% accuracy), whereas current LVLMs often fail (e.g., ChatGPT-4o achieves only 2.10\% accuracy). Moreover, fine-tuning experiments demonstrate that \bench serves as an effective resource for improving model perception, increasing accuracy by 55.94\% for Qwen2.5VL-7B. Attention analysis and layer-wise probing further reveal that fine-tuning enhances sensitivity primarily in the early layers of the vision encoder, promoting a more integrated scene understanding. These findings highlight the inherent perceptual limitations in LVLMs and offer insight into more human-aligned visual reasoning systems.

Paperid: 4321, https://arxiv.org/pdf/2512.02966.pdf

Abstract:
We introduce the first principled framework, Lumos, for specifying and formally certifying Language Model System (LMS) behaviors. Lumos is an imperative probabilistic programming DSL over graphs, with constructs to generate independent and identically distributed prompts for LMS. It offers a structured view of prompt distributions via graphs, forming random prompts from sampled subgraphs. Lumos supports certifying LMS for arbitrary prompt distributions via integration with statistical certifiers. We provide hybrid (operational and denotational) semantics for Lumos, providing a rigorous way to interpret the specifications. Using only a small set of composable constructs, Lumos can encode existing LMS specifications, including complex relational and temporal specifications. It also facilitates specifying new properties - we present the first safety specifications for vision-language models (VLMs) in autonomous driving scenarios developed with Lumos. Using these, we show that the state-of-the-art VLM Qwen-VL exhibits critical safety failures, producing incorrect and unsafe responses with at least 90% probability in right-turn scenarios under rainy driving conditions, revealing substantial safety risks. Lumos's modular structure allows easy modification of the specifications, enabling LMS certification to stay abreast with the rapidly evolving threat landscape. We further demonstrate that specification programs written in Lumos enable finding specific failure cases exhibited by state-of-the-art LMS. Lumos is the first systematic and extensible language-based framework for specifying and certifying LMS behaviors, paving the way for a wider adoption of LMS certification.

Paperid: 4322, https://arxiv.org/pdf/2512.02631.pdf

Abstract:
Existing Vision-Language Navigation (VLN) agents based on Large Vision-Language Models (LVLMs) often suffer from perception errors, reasoning errors, and planning errors, which significantly hinder their navigation performance. To address these limitations, a novel VLN agent framework, named SeeNav-Agent, is proposed in this work. First, to reduce perception hallucinations of the visual module of the VLN agent, a dual-view Visual Prompt (VP) technique is introduced in the input space, which can also improve the agent's understanding of current spatial states. Subsequently, a novel step-level Reinforcement Fine-Tuning (RFT) method, Step Reward Group Policy Optimization (SRGPO), is designed for the post-training of VLN agents. In SRGPO, we first define verifiable process rewards for the navigation task, and then perform efficient step-level advantage estimation by randomly grouping different navigation steps. SRGPO provides dense reward signals for the reinforcement learning process of the VLN agent and enhances its planning capability. Experimental results on the EmbodiedBench Navigation benchmark indicate that by introducing the zero-shot VP module, the GPT-4.1 achieves a navigation success rate of 86.7%, surpassing the current best LVLM by approximately 20 percentage points (pp). Through post-training based on SRGPO, the Qwen2.5-VL-3B model reaches a navigation success rate of 72.3%, outperforming the best existing LVLM model by 5.6 pp. Moreover, compared to RFT algorithms such as GRPO and GiGPO, the proposed SRGPO demonstrates significant improvements in training stability, convergence efficiency, and generalization capability.

Paperid: 4323, https://arxiv.org/pdf/2512.01946.pdf

Abstract:
Robust robotic manipulation requires reliable failure detection and recovery. Although current Vision-Language Models (VLMs) show promise, their accuracy and generalization are limited by the scarcity of failure data. To address this data gap, we propose an automatic robot failure synthesis approach that procedurally perturbs successful trajectories to generate diverse planning and execution failures. This method produces not only binary classification labels but also fine-grained failure categories and step-by-step reasoning traces in both simulation and the real world. With it, we construct three new failure detection benchmarks: RLBench-Fail, BridgeDataV2-Fail, and UR5-Fail, substantially expanding the diversity and scale of existing failure datasets. We then train Guardian, a VLM with multi-view images for detailed failure reasoning and detection. Guardian achieves state-of-the-art performance on both existing and newly introduced benchmarks. It also effectively improves task success rates when integrated into a state-of-the-art manipulation system in simulation and real robots, demonstrating the impact of our generated failure data. Code, Data, and Models available at https://www.di.ens.fr/willow/research/guardian/.

Paperid: 4324, https://arxiv.org/pdf/2512.01009.pdf

Abstract:
This paper addresses the Object Goal Navigation problem, where a robot must efficiently find a target object in an unknown environment. Existing implicit memory-based methods struggle with long-term memory retention and planning, while explicit map-based approaches lack rich semantic information. To address these challenges, we propose FOM-Nav, a modular framework that enhances exploration efficiency through Frontier-Object Maps and vision-language models. Our Frontier-Object Maps are built online and jointly encode spatial frontiers and fine-grained object information. Using this representation, a vision-language model performs multimodal scene understanding and high-level goal prediction, which is executed by a low-level planner for efficient trajectory generation. To train FOM-Nav, we automatically construct large-scale navigation datasets from real-world scanned environments. Extensive experiments validate the effectiveness of our model design and constructed dataset. FOM-Nav achieves state-of-the-art performance on the MP3D and HM3D benchmarks, particularly in navigation efficiency metric SPL, and yields promising results on a real robot.

Paperid: 4325, https://arxiv.org/pdf/2511.16163.pdf

Abstract:
With the remarkable success of Vision-Language Models (VLMs) on multimodal tasks, concerns regarding their deployment efficiency have become increasingly prominent. In particular, the number of tokens consumed during the generation process has emerged as a key evaluation metric.Prior studies have shown that specific inputs can induce VLMs to generate lengthy outputs with low information density, which significantly increases energy consumption, latency, and token costs. However, existing methods simply delay the occurrence of the EOS token to implicitly prolong output, and fail to directly maximize the output token length as an explicit optimization objective, lacking stability and controllability.To address these limitations, this paper proposes a novel verbose-text induction attack (VTIA) to inject imperceptible adversarial perturbations into benign images via a two-stage framework, which identifies the most malicious prompt embeddings for optimizing and maximizing the output token of the perturbed images.Specifically, we first perform adversarial prompt search, employing reinforcement learning strategies to automatically identify adversarial prompts capable of inducing the LLM component within VLMs to produce verbose outputs. We then conduct vision-aligned perturbation optimization to craft adversarial examples on input images, maximizing the similarity between the perturbed image's visual embeddings and those of the adversarial prompt, thereby constructing malicious images that trigger verbose text generation. Comprehensive experiments on four popular VLMs demonstrate that our method achieves significant advantages in terms of effectiveness, efficiency, and generalization capability.

Paperid: 4326, https://arxiv.org/pdf/2511.15622.pdf

Abstract:
Automated video analysis is critical for wildlife conservation. A foundational task in this domain is multi-animal tracking (MAT), which underpins applications such as individual re-identification and behavior recognition. However, existing datasets are limited in scale, constrained to a few species, or lack sufficient temporal and geographical diversity - leaving no suitable benchmark for training general-purpose MAT models applicable across wild animal populations. To address this, we introduce SA-FARI, the largest open-source MAT dataset for wild animals. It comprises 11,609 camera trap videos collected over approximately 10 years (2014-2024) from 741 locations across 4 continents, spanning 99 species categories. Each video is exhaustively annotated culminating in ~46 hours of densely annotated footage containing 16,224 masklet identities and 942,702 individual bounding boxes, segmentation masks, and species labels. Alongside the task-specific annotations, we publish anonymized camera trap locations for each video. Finally, we present comprehensive benchmarks on SA-FARI using state-of-the-art vision-language models for detection and tracking, including SAM 3, evaluated with both species-specific and generic animal prompts. We also compare against vision-only methods developed specifically for wildlife analysis. SA-FARI is the first large-scale dataset to combine high species diversity, multi-region coverage, and high-quality spatio-temporal annotations, offering a new foundation for advancing generalizable multianimal tracking in the wild. The dataset is available at https://www.conservationxlabs.com/sa-fari.

Paperid: 4327, https://arxiv.org/pdf/2511.13892.pdf

Abstract:
Large Vision Language Models (LVLMs) demonstrate strong capabilities in multimodal reasoning and many real-world applications, such as visual question answering. However, LVLMs are highly vulnerable to jailbreaking attacks. This paper systematically analyzes the vulnerabilities of LVLMs integrated in Intelligent Transportation Systems (ITS) under carefully crafted jailbreaking attacks. First, we carefully construct a dataset with harmful queries relevant to transportation, following OpenAI's prohibited categories to which the LVLMs should not respond. Second, we introduce a novel jailbreaking attack that exploits the vulnerabilities of LVLMs through image typography manipulation and multi-turn prompting. Third, we propose a multi-layered response filtering defense technique to prevent the model from generating inappropriate responses. We perform extensive experiments with the proposed attack and defense on the state-of-the-art LVLMs (both open-source and closed-source). To evaluate the attack method and defense technique, we use GPT-4's judgment to determine the toxicity score of the generated responses, as well as manual verification. Further, we compare our proposed jailbreaking method with existing jailbreaking techniques and highlight severe security risks involved with jailbreaking attacks with image typography manipulation and multi-turn prompting in the LVLMs integrated in ITS.

Paperid: 4328, https://arxiv.org/pdf/2511.13397.pdf

Abstract:
The remarkable progress of Vision-Language Models (VLMs) on a variety of tasks has raised interest in their application to automated driving. However, for these models to be trusted in such a safety-critical domain, they must first possess robust perception capabilities, i.e., they must be capable of understanding a traffic scene, which can often be highly complex, with many things happening simultaneously. Moreover, since critical objects and agents in traffic scenes are often at long distances, we require systems with not only strong perception capabilities at close distances (up to 20 meters), but also at long (30+ meters) range. Therefore, it is important to evaluate the perception capabilities of these models in isolation from other skills like reasoning or advanced world knowledge. Distance-Annotated Traffic Perception Question Answering (DTPQA) is a Visual Question Answering (VQA) benchmark designed specifically for this purpose: it can be used to evaluate the perception systems of VLMs in traffic scenarios using trivial yet crucial questions relevant to driving decisions. It consists of two parts: a synthetic benchmark (DTP-Synthetic) created using a simulator, and a real-world benchmark (DTP-Real) built on top of existing images of real traffic scenes. Additionally, DTPQA includes distance annotations, i.e., how far the object in question is from the camera. More specifically, each DTPQA sample consists of (at least): (a) an image, (b) a question, (c) the ground truth answer, and (d) the distance of the object in question, enabling analysis of how VLM performance degrades with increasing object distance. In this article, we provide the dataset itself along with the Python scripts used to create it, which can be used to generate additional data of the same kind.

Paperid: 4329, https://arxiv.org/pdf/2511.08892.pdf

Abstract:
We introduce Lumine, the first open recipe for developing generalist agents capable of completing hours-long complex missions in real time within challenging 3D open-world environments. Lumine adopts a human-like interaction paradigm that unifies perception, reasoning, and action in an end-to-end manner, powered by a vision-language model. It processes raw pixels at 5 Hz to produce precise 30 Hz keyboard-mouse actions and adaptively invokes reasoning only when necessary. Trained in Genshin Impact, Lumine successfully completes the entire five-hour Mondstadt main storyline on par with human-level efficiency and follows natural language instructions to perform a broad spectrum of tasks in both 3D open-world exploration and 2D GUI manipulation across collection, combat, puzzle-solving, and NPC interaction. In addition to its in-domain performance, Lumine demonstrates strong zero-shot cross-game generalization. Without any fine-tuning, it accomplishes 100-minute missions in Wuthering Waves and the full five-hour first chapter of Honkai: Star Rail. These promising results highlight Lumine's effectiveness across distinct worlds and interaction dynamics, marking a concrete step toward generalist agents in open-ended environments.

Paperid: 4330, https://arxiv.org/pdf/2511.06805.pdf

Abstract:
Multimodal large language models (MLLMs) have demonstrated remarkable capabilities in vision-language answering tasks. Despite their strengths, these models often encounter challenges in achieving complex reasoning tasks such as mathematical problem-solving. Previous works have focused on fine-tuning on specialized mathematical datasets. However, these datasets are typically distilled directly from teacher models, which capture only static reasoning patterns and leaving substantial gaps compared to student models. This reliance on fixed teacher-derived datasets not only restricts the model's ability to adapt to novel or more intricate questions that extend beyond the confines of the training data, but also lacks the iterative depth needed for robust generalization. To overcome these limitations, we propose \textbf{\method}, a \textbf{Math}ematical \textbf{S}elf-\textbf{E}volving framework for MLLMs. In contrast to traditional one-shot fine-tuning paradigms, \method iteratively refines the model through cycles of inference, reflection, and reward-based feedback. Specifically, we leverage iterative fine-tuning by incorporating correct reasoning paths derived from previous-stage inference and integrating reflections from a specialized Outcome Reward Model (ORM). To verify the effectiveness of \method, we evaluate it on a suite of challenging benchmarks, demonstrating significant performance gains over backbone models. Notably, our experimental results on MathVL-test surpass the leading open-source multimodal mathematical reasoning model QVQ. Our code and models are available at \texttt{https://zheny2751\allowbreak-dotcom.github.io/\allowbreak MathSE.github.io/}.

Paperid: 4331, https://arxiv.org/pdf/2511.06182.pdf

Abstract:
Vision-language models (VLMs) have been widely-applied in ground-based vision-language navigation (VLN). However, the vast complexity of outdoor aerial environments compounds data acquisition challenges and imposes long-horizon trajectory planning requirements on Unmanned Aerial Vehicles (UAVs), introducing novel complexities for aerial VLN. To address these challenges, we propose a data-efficient Open-world aerial Vision-Language Navigation (i.e., OpenVLN) framework, which could execute language-guided flight with limited data constraints and enhance long-horizon trajectory planning capabilities in complex aerial environments. Specifically, we reconfigure a reinforcement learning framework to optimize the VLM for UAV navigation tasks, which can efficiently fine-tune VLM by using rule-based policies under limited training data. Concurrently, we introduce a long-horizon planner for trajectory synthesis that dynamically generates precise UAV actions via value-based rewards. To the end, we conduct sufficient navigation experiments on the TravelUAV benchmark with dataset scaling across diverse reward settings. Our method demonstrates consistent performance gains of up to 4.34% in Success Rate, 6.19% in Oracle Success Rate, and 4.07% in Success weighted by Path Length over baseline methods, validating its deployment efficacy for long-horizon UAV navigation in complex aerial environments.

Paperid: 4332, https://arxiv.org/pdf/2511.04601.pdf

Abstract:
While the Contrastive Language-Image Pretraining(CLIP) model has achieved remarkable success in a variety of downstream vison language understanding tasks, enhancing its capability for fine-grained image-text alignment remains an active research focus. To this end, most existing works adopt the strategy of explicitly increasing the granularity of visual information processing, e.g., incorporating visual prompts to guide the model focus on specific local regions within the image. Meanwhile, researches on Multimodal Large Language Models(MLLMs) have demonstrated that training with long and detailed textual descriptions can effectively improve the model's fine-grained vision-language alignment. However, the inherent token length limitation of CLIP's text encoder fundamentally limits CLIP to process more granular textual information embedded in long text sequences. To synergistically leverage the advantages of enhancing both visual and textual content processing granularity, we propose PixCLIP, a novel framework designed to concurrently accommodate visual prompt inputs and process lengthy textual descriptions. Specifically, we first establish an automated annotation pipeline capable of generating pixel-level localized, long-form textual descriptions for images. Utilizing this pipeline, we construct LongGRIT, a high-quality dataset comprising nearly 1.5 million samples. Secondly, we replace CLIP's original text encoder with the LLM and propose a three-branch pixel-text alignment learning framework, facilitating fine-grained alignment between image regions and corresponding textual descriptions at arbitrary granularity. Experiments demonstrate that PixCLIP showcases breakthroughs in pixel-level interaction and handling long-form texts, achieving state-of-the-art performance.

Paperid: 4333, https://arxiv.org/pdf/2510.25146.pdf

Abstract:
Current 3D scene understanding methods are limited by offline-collected multi-view data or pre-constructed 3D geometry. In this paper, we present ExtractAnything3D (EA3D), a unified online framework for open-world 3D object extraction that enables simultaneous geometric reconstruction and holistic scene understanding. Given a streaming video, EA3D dynamically interprets each frame using vision-language and 2D vision foundation encoders to extract object-level knowledge. This knowledge is integrated and embedded into a Gaussian feature map via a feed-forward online update strategy. We then iteratively estimate visual odometry from historical frames and incrementally update online Gaussian features with new observations. A recurrent joint optimization module directs the model's attention to regions of interest, simultaneously enhancing both geometric reconstruction and semantic understanding. Extensive experiments across diverse benchmarks and tasks, including photo-realistic rendering, semantic and instance segmentation, 3D bounding box and semantic occupancy estimation, and 3D mesh generation, demonstrate the effectiveness of EA3D. Our method establishes a unified and efficient framework for joint online 3D reconstruction and holistic scene understanding, enabling a broad range of downstream tasks.

Paperid: 4334, https://arxiv.org/pdf/2510.22728.pdf

Abstract:
Faithful reasoning in medical vision-language models (VLMs) requires not only accurate predictions but also transparent alignment between textual rationales and visual evidence. While Chain-of-Thought (CoT) prompting has shown promise in medical visual question answering (VQA), no large-scale expert-level dataset has captured stepwise reasoning with precise visual grounding. We introduce S-Chain, the first large-scale dataset of 12,000 expert-annotated medical images with bounding boxes and structured visual CoT (SV-CoT), explicitly linking visual regions to reasoning steps. The dataset further supports 16 languages, totaling over 700k VQA pairs for broad multilingual applicability. Using S-Chain, we benchmark state-of-the-art medical VLMs (ExGra-Med, LLaVA-Med) and general-purpose VLMs (Qwen2.5-VL, InternVL2.5), showing that SV-CoT supervision significantly improves interpretability, grounding fidelity, and robustness. Beyond benchmarking, we study its synergy with retrieval-augmented generation, revealing how domain knowledge and visual grounding interact during autoregressive reasoning. Finally, we propose a new mechanism that strengthens the alignment between visual evidence and reasoning, improving both reliability and efficiency. S-Chain establishes a new benchmark for grounded medical reasoning and paves the way toward more trustworthy and explainable medical VLMs.

Paperid: 4335, https://arxiv.org/pdf/2510.21083.pdf

Abstract:
Hirschsprung's disease is defined as the congenital absence of ganglion cells in some segment(s) of the colon. The muscle cannot make coordinated movements to propel stool in that section, most commonly leading to obstruction. The diagnosis and treatment for this disease require a clear identification of different region(s) of the myenteric plexus, where ganglion cells should be present, on the microscopic view of the tissue slide. While deep learning approaches, such as Convolutional Neural Networks, have performed very well in this task, they are often treated as black boxes, with minimal understanding gained from them, and may not conform to how a physician makes decisions. In this study, we propose a novel framework that integrates expert-derived textual concepts into a Contrastive Language-Image Pre-training-based vision-language model to guide plexus classification. Using prompts derived from expert sources (e.g., medical textbooks and papers) generated by large language models and reviewed by our team before being encoded with QuiltNet, our approach aligns clinically relevant semantic cues with visual features. Experimental results show that the proposed model demonstrated superior discriminative capability across different classification metrics as it outperformed CNN-based models, including VGG-19, ResNet-18, and ResNet-50; achieving an accuracy of 83.9%, a precision of 86.6%, and a specificity of 87.6%. These findings highlight the potential of multi-modal learning in histopathology and underscore the value of incorporating expert knowledge for more clinically relevant model outputs.

Paperid: 4336, https://arxiv.org/pdf/2510.17148.pdf

Abstract:
Conventional end-to-end (E2E) driving models are effective at generating physically plausible trajectories, but often fail to generalize to long-tail scenarios due to the lack of essential world knowledge to understand and reason about surrounding environments. In contrast, Vision-Language-Action (VLA) models leverage world knowledge to handle challenging cases, but their limited 3D reasoning capability can lead to physically infeasible actions. In this work we introduce DiffVLA++, an enhanced autonomous driving framework that explicitly bridges cognitive reasoning and E2E planning through metric-guided alignment. First, we build a VLA module directly generating semantically grounded driving trajectories. Second, we design an E2E module with a dense trajectory vocabulary that ensures physical feasibility. Third, and most critically, we introduce a metric-guided trajectory scorer that guides and aligns the outputs of the VLA and E2E modules, thereby integrating their complementary strengths. The experiment on the ICCV 2025 Autonomous Grand Challenge leaderboard shows that DiffVLA++ achieves EPDMS of 49.12.

Paperid: 4337, https://arxiv.org/pdf/2510.14528.pdf

Abstract:
In this report, we propose PaddleOCR-VL, a SOTA and resource-efficient model tailored for document parsing. Its core component is PaddleOCR-VL-0.9B, a compact yet powerful vision-language model (VLM) that integrates a NaViT-style dynamic resolution visual encoder with the ERNIE-4.5-0.3B language model to enable accurate element recognition. This innovative model efficiently supports 109 languages and excels in recognizing complex elements (e.g., text, tables, formulas, and charts), while maintaining minimal resource consumption. Through comprehensive evaluations on widely used public benchmarks and in-house benchmarks, PaddleOCR-VL achieves SOTA performance in both page-level document parsing and element-level recognition. It significantly outperforms existing solutions, exhibits strong competitiveness against top-tier VLMs, and delivers fast inference speeds. These strengths make it highly suitable for practical deployment in real-world scenarios.

Paperid: 4338, https://arxiv.org/pdf/2510.12931.pdf

Abstract:
Vision-language models (VLMs) achieve remarkable performance through large-scale image-text pretraining. However, their reliance on labeled image datasets limits scalability and leaves vast amounts of unlabeled image data underutilized. To address this, we propose Unified Vision-Language Alignment for Zero-Label Enhancement (ViZer), an enhancement training framework that enables zero-label learning in image captioning, providing a practical starting point for broader zero-label adaptation in vision-language tasks. Unlike prior approaches that rely on human or synthetically annotated datasets, ViZer actively aligns vision and language representation features during training, enabling existing VLMs to generate improved captions without requiring text labels or full retraining. We demonstrate ViZer's advantage in qualitative evaluation, as automated caption metrics such as CIDEr and BERTScore often penalize details that are absent in reference captions. Applying ViZer on SmolVLM-Base and Qwen2-VL, we observe consistent qualitative improvements, producing captions that are more grounded and descriptive than their baseline.

Paperid: 4339, https://arxiv.org/pdf/2510.10466.pdf

Abstract:
Vision-Language Models (VLMs) have shown solid ability for multimodal understanding of both visual and language contexts. However, existing VLMs often face severe challenges of hallucinations, meaning that VLMs tend to generate responses that are only fluent in the language but irrelevant to images in previous contexts. To address this issue, we analyze how language bias contributes to hallucinations and then introduce Cross-Modal Guidance(CMG), a training-free decoding method that addresses the hallucinations by leveraging the difference between the output distributions of the original model and the one with degraded visual-language attention. In practice, we adaptively mask the attention weight of the most influential image tokens in selected transformer layers to corrupt the visual-language perception as a concrete type of degradation. Such a degradation-induced decoding emphasizes the perception of visual contexts and therefore significantly reduces language bias without harming the ability of VLMs. In experiment sections, we conduct comprehensive studies. All results demonstrate the superior advantages of CMG with neither additional conditions nor training costs. We also quantitatively show CMG can improve different VLM's performance on hallucination-specific benchmarks and generalize effectively.

Paperid: 4340, https://arxiv.org/pdf/2510.08352.pdf

Abstract:
Vision-Language Models (VLMs) are becoming increasingly powerful, demonstrating strong performance on a variety of tasks that require both visual and textual understanding. Their strong generalisation abilities make them a promising component for automated driving systems, which must handle unexpected corner cases. However, to be trusted in such safety-critical applications, a model must first possess a reliable perception system. Moreover, since critical objects and agents in traffic scenes are often at a distance, we require systems that are not "shortsighted", i.e., systems with strong perception capabilities at both close (up to 20 meters) and long (30+ meters) range. With this in mind, we introduce Distance-Annotated Traffic Perception Question Answering (DTPQA), the first Visual Question Answering (VQA) benchmark focused solely on perception-based questions in traffic scenes, enriched with distance annotations. By excluding questions that require reasoning, we ensure that model performance reflects perception capabilities alone. Since automated driving hardware has limited processing power and cannot support large VLMs, our study centers on smaller VLMs. More specifically, we evaluate several state-of-the-art (SOTA) small VLMs on DTPQA and show that, despite the simplicity of the questions, these models significantly underperform compared to humans (~60% average accuracy for the best-performing small VLM versus ~85% human performance). However, it is important to note that the human sample size was relatively small, which imposes statistical limitations. We also identify specific perception tasks, such as distinguishing left from right, that remain particularly challenging for these models.

Paperid: 4341, https://arxiv.org/pdf/2510.05662.pdf

Abstract:
Despite the prevalence of transparent object interactions in human everyday life, transparent robotic manipulation research remains limited to short-horizon tasks and basic grasping capabilities.Although some methods have partially addressed these issues, most of them have limitations in generalizability to novel objects and are insufficient for precise long-horizon robot manipulation. To address this limitation, we propose DeLTa (Demonstration and Language-Guided Novel Transparent Object Manipulation), a novel framework that integrates depth estimation, 6D pose estimation, and vision-language planning for precise long-horizon manipulation of transparent objects guided by natural task instructions. A key advantage of our method is its single-demonstration approach, which generalizes 6D trajectories to novel transparent objects without requiring category-level priors or additional training. Additionally, we present a task planner that refines the VLM-generated plan to account for the constraints of a single-arm, eye-in-hand robot for long-horizon object manipulation tasks. Through comprehensive evaluation, we demonstrate that our method significantly outperforms existing transparent object manipulation approaches, particularly in long-horizon scenarios requiring precise manipulation capabilities. Project page: https://sites.google.com/view/DeLTa25/

Paperid: 4342, https://arxiv.org/pdf/2510.05544.pdf

Abstract:
Large language models (LLM) and vision-language models (VLM) have achieved state-of-the-art performance, but they impose significant memory and computing challenges in deployment. We present a novel low-rank compression framework to address this challenge. First, we upper bound the change of network loss via layer-wise activation-based compression errors, filling a theoretical gap in the literature. We then formulate low-rank model compression as a bi-objective optimization and prove that a single uniform tolerance yields surrogate Pareto-optimal heterogeneous ranks. Based on our theoretical insights, we propose Pareto-Guided Singular Value Decomposition (PGSVD), a zero-shot pipeline that improves activation-aware compression via Pareto-guided rank selection and alternating least-squares implementation. We apply PGSVD to both LLM and VLM, showing better accuracy at the same compression levels and inference speedup.

Paperid: 4343, https://arxiv.org/pdf/2510.01700.pdf

Abstract:
Preference finetuning methods like Direct Preference Optimization (DPO) with AI-generated feedback have shown promise in aligning Large Vision-Language Models (LVLMs) with human preferences. However, existing techniques overlook the prevalence of noise in synthetic preference annotations in the form of stylistic and length biases. To this end, we introduce a hard-negative response generation framework based on LLM-guided response editing, that produces rejected responses with targeted errors, maintaining stylistic and length similarity to the accepted ones. Using this framework, we develop the VaPR dataset, comprising 30K high-quality samples, to finetune three LVLM families: LLaVA-V1.5, Qwen2VL & Qwen2.5VL (2B-13B sizes). Our VaPR models deliver significant performance improvements across ten benchmarks, achieving average gains of 6.5% (LLaVA), 4.0% (Qwen2VL), and 1.5% (Qwen2.5VL), with notable improvements on reasoning tasks. A scaling analysis shows that performance consistently improves with data size, with LLaVA models benefiting even at smaller scales. Moreover, VaPR reduces the tendency to answer "Yes" in binary questions - addressing a common failure mode in LVLMs like LLaVA. Lastly, we show that the framework generalizes to open-source LLMs as editors, with models trained on VaPR-OS achieving ~99% of the performance of models trained on \name, which is synthesized using GPT-4o. Our data, models, and code can be found on the project page https://vap-r.github.io

Paperid: 4344, https://arxiv.org/pdf/2510.01149.pdf

Abstract:
Multimodal embedding models are gaining prevalence, notably for document retrieval as efficient alternatives to text-only pipelines. These models are typically built by finetuning large vision-language decoders (VLMs) with contrastive losses on text-image pairs. In this work, we show that, while cost-efficient, this repurposing approach often bottlenecks retrieval performance. Through controlled experiments, we establish a principled recipe for improving visual document retrieval models. We notably measure the impact of attention masking, image resolution, modality alignment data regimes, and late interaction centered contrastive objectives which emerge as central performance factors. Building on these insights, we release ModernVBERT, a compact 250M-parameter vision-language encoder that outperforms models up to 10 times larger when finetuned on document retrieval tasks. Models and code are made available at https://huggingface.co/ModernVBERT.

Paperid: 4345, https://arxiv.org/pdf/2510.00600.pdf

Abstract:
Using Large Language Models to produce intermediate thoughts, a.k.a. Chain-of-thought (CoT), before providing an answer has been a successful recipe for solving complex language tasks. In robotics, similar embodied CoT strategies, generating thoughts before actions, have also been shown to lead to improved performance when using Vision-Language-Action models (VLAs). As these techniques increase the length of the model's generated outputs to include the thoughts, the inference time is negatively affected. Delaying an agent's actions in real-world executions, as in robotic manipulation settings, strongly affects the usability of a method, as tasks require long sequences of actions. However, is the generation of long chains-of-thought a strong prerequisite for achieving performance improvements? In this work, we explore the idea of Hybrid Training (HyT), a framework that enables VLAs to learn from thoughts and benefit from the associated performance gains, while enabling the possibility to leave out CoT generation during inference. Furthermore, by learning to conditionally predict a diverse set of outputs, HyT supports flexibility at inference time, enabling the model to either predict actions directly, generate thoughts or follow instructions. We evaluate the proposed method in a series of simulated benchmarks and real-world experiments.

Paperid: 4346, https://arxiv.org/pdf/2509.24917.pdf

Abstract:
Imitation learning for robotic manipulation often suffers from limited generalization and data scarcity, especially in complex, long-horizon tasks. In this work, we introduce a hierarchical framework that leverages code-generating vision-language models (VLMs) in combination with low-level diffusion policies to effectively imitate and generalize robotic behavior. Our key insight is to treat open-source robotic APIs not only as execution interfaces but also as sources of structured supervision: the associated subtask functions - when exposed - can serve as modular, semantically meaningful labels. We train a VLM to decompose task descriptions into executable subroutines, which are then grounded through a diffusion policy trained to imitate the corresponding robot behavior. To handle the non-Markovian nature of both code execution and certain real-world tasks, such as object swapping, our architecture incorporates a memory mechanism that maintains subtask context across time. We find that this design enables interpretable policy decomposition, improves generalization when compared to flat policies and enables separate evaluation of high-level planning and low-level control.

Paperid: 4347, https://arxiv.org/pdf/2509.24768.pdf

Abstract:
Vision-language-action models (VLAs) have become an increasingly popular approach for addressing robot manipulation problems in recent years. However, such models need to output actions at a rate suitable for robot control, which limits the size of the language model they can be based on, and consequently, their language understanding capabilities. Manipulation tasks may require complex language instructions, such as identifying target objects by their relative positions, to specify human intention. Therefore, we introduce IA-VLA, a framework that utilizes the extensive language understanding of a large vision language model as a pre-processing stage to generate improved context to augment the input of a VLA. We evaluate the framework on a set of semantically complex tasks which have been underexplored in VLA literature, namely tasks involving visual duplicates, i.e., visually indistinguishable objects. A dataset of three types of scenes with duplicate objects is used to compare a baseline VLA against two augmented variants. The experiments show that the VLA benefits from the augmentation scheme, especially when faced with language instructions that require the VLA to extrapolate from concepts it has seen in the demonstrations. For the code, dataset, and videos, see https://sites.google.com/view/ia-vla.

Paperid: 4348, https://arxiv.org/pdf/2509.23655.pdf

Abstract:
Vision-Language-Action (VLA) models offer a pivotal approach to learning robotic manipulation at scale by repurposing large pre-trained Vision-Language-Models (VLM) to output robotic actions. However, adapting VLMs for robotic domains comes with an unnecessarily high computational cost, which we attribute to the tokenization scheme of visual inputs. In this work, we aim to enable efficient VLA training by proposing Oat-VLA, an Object-Agent-centric Tokenization for VLAs. Building on the insights of object-centric representation learning, our method introduces an inductive bias towards scene objects and the agent's own visual information. As a result, we find that Oat-VLA can drastically reduce the number of visual tokens to just a few tokens without sacrificing performance. We reveal that Oat-VLA converges at least twice as fast as OpenVLA on the LIBERO suite, as well as outperform OpenVLA in diverse real-world pick and place tasks.

Paperid: 4349, https://arxiv.org/pdf/2509.23311.pdf

Abstract:
Vision-Language Models (VLMs) often appear culturally competent but rely on superficial pattern matching rather than genuine cultural understanding. We introduce a diagnostic framework to probe VLM reasoning on fire-themed cultural imagery through both classification and explanation analysis. Testing multiple models on Western festivals, non-Western traditions, and emergency scenes reveals systematic biases: models correctly identify prominent Western festivals but struggle with underrepresented cultural events, frequently offering vague labels or dangerously misclassifying emergencies as celebrations. These failures expose the risks of symbolic shortcuts and highlight the need for cultural evaluation beyond accuracy metrics to ensure interpretable and fair multimodal systems.

Paperid: 4350, https://arxiv.org/pdf/2509.21401.pdf

Abstract:
Vision-Language Models (VLMs) have remarkable abilities in generating multimodal reasoning tasks. However, potential misuse or safety alignment concerns of VLMs have increased significantly due to different categories of attack vectors. Among various attack vectors, recent studies have demonstrated that image-based perturbations are particularly effective in generating harmful outputs. In the literature, many existing techniques have been proposed to jailbreak VLMs, leading to unstable performance and visible perturbations. In this study, we propose Jailbreaking with Loss-guided Image Perturbation (JaiLIP), a jailbreaking attack in the image space that minimizes a joint objective combining the mean squared error (MSE) loss between clean and adversarial image with the models harmful-output loss. We evaluate our proposed method on VLMs using standard toxicity metrics from Perspective API and Detoxify. Experimental results demonstrate that our method generates highly effective and imperceptible adversarial images, outperforming existing methods in producing toxicity. Moreover, we have evaluated our method in the transportation domain to demonstrate the attacks practicality beyond toxic text generation in specific domain. Our findings emphasize the practical challenges of image-based jailbreak attacks and the need for efficient defense mechanisms for VLMs.

Paperid: 4351, https://arxiv.org/pdf/2509.21257.pdf

Abstract:
In language and vision-language models, hallucination is broadly understood as content generated from a model's prior knowledge or biases rather than from the given input. While this phenomenon has been studied in those domains, it has not been clearly framed for text-to-image (T2I) generative models. Existing evaluations mainly focus on alignment, checking whether prompt-specified elements appear, but overlook what the model generates beyond the prompt. We argue for defining hallucination in T2I as bias-driven deviations and propose a taxonomy with three categories: attribute, relation, and object hallucinations. This framing introduces an upper bound for evaluation and surfaces hidden biases, providing a foundation for richer assessment of T2I models.

Paperid: 4352, https://arxiv.org/pdf/2509.20912.pdf

Abstract:
Recent advances in multimodal language models (MLLMs) have achieved remarkable progress in vision-language reasoning, especially with the emergence of "thinking with images," which integrates explicit visual steps into the reasoning process. While this paradigm strengthens image-based reasoning, a significant challenge remains: models may arrive at correct answers by relying on irrelevant or spurious regions, driven by prior knowledge or dataset biases. Even when the answer is correct, flawed reasoning indicates that the model has not truly understood the image, highlighting the critical importance of reasoning fidelity in multimodal tasks. To address this issue, we propose DeFacto, a counterfactual reasoning framework that jointly enforces accurate answering and faithful reasoning. A key component of our approach is the design of three complementary training paradigms: (i) positive, (ii) counterfactual, and (iii) random-masking. To enable these paradigms, we develop a pipeline that automatically localizes question-relevant evidence and constructs positive, counterfactual, and random variants, resulting in a dataset of about 100k images. Building on this framework, we train multimodal language models with GRPO-based reinforcement learning, where we design three complementary rewards to guide the model toward accurate answering and evidence-grounded reasoning. Experiments on diverse benchmarks demonstrate that DeFacto substantially improves both answer accuracy and reasoning faithfulness, establishing a stronger foundation for interpretable multimodal reasoning. The code is available on GitHub and the dataset is released on HuggingFace.

Paperid: 4353, https://arxiv.org/pdf/2509.20499.pdf

Abstract:
With the rapid progress of foundation models and robotics, vision-language navigation (VLN) has emerged as a key task for embodied agents with broad practical applications. We address VLN in continuous environments, a particularly challenging setting where an agent must jointly interpret natural language instructions, perceive its surroundings, and plan low-level actions. We propose a zero-shot framework that integrates a simplified yet effective waypoint predictor with a multimodal large language model (MLLM). The predictor operates on an abstract obstacle map, producing linearly reachable waypoints, which are incorporated into a dynamically updated topological graph with explicit visitation records. The graph and visitation information are encoded into the prompt, enabling reasoning over both spatial structure and exploration history to encourage exploration and equip MLLM with local path planning for error correction. Extensive experiments on R2R-CE and RxR-CE show that our method achieves state-of-the-art zero-shot performance, with success rates of 41% and 36%, respectively, outperforming prior state-of-the-art methods.

Paperid: 4354, https://arxiv.org/pdf/2509.18189.pdf

Abstract:
We present Qianfan-VL, a series of multimodal large language models ranging from 3B to 70B parameters, achieving state-of-the-art performance through innovative domain enhancement techniques. Our approach employs multi-stage progressive training and high-precision data synthesis pipelines, which prove to be critical technologies for enhancing domain-specific capabilities while maintaining strong general performance. Qianfan-VL achieves comparable results to leading open-source models on general benchmarks, with state-of-the-art performance on benchmarks such as CCBench, SEEDBench IMG, ScienceQA, and MMStar. The domain enhancement strategy delivers significant advantages in OCR and document understanding, validated on both public benchmarks (OCRBench 873, DocVQA 94.75%) and in-house evaluations. Notably, Qianfan-VL-8B and 70B variants incorporate long chain-of-thought capabilities, demonstrating superior performance on mathematical reasoning (MathVista 78.6%) and logical inference tasks. All models are trained entirely on Baidu's Kunlun P800 chips, validating the capability of large-scale AI infrastructure to train SOTA-level multimodal models with over 90% scaling efficiency on 5000 chips for a single task. This work establishes an effective methodology for developing domain-enhanced multimodal models suitable for diverse enterprise deployment scenarios.

Paperid: 4355, https://arxiv.org/pdf/2509.17747.pdf

Abstract:
Real-world datasets often exhibit class imbalance across multiple categories, manifesting as long-tailed distributions and few-shot scenarios. This is especially challenging in Class-Imbalanced Multi-Label Image Classification (CI-MLIC) tasks, where data imbalance and multi-object recognition present significant obstacles. To address these challenges, we propose a novel method termed Dual-View Alignment Learning with Hierarchical Prompt (HP-DVAL), which leverages multi-modal knowledge from vision-language pretrained (VLP) models to mitigate the class-imbalance problem in multi-label settings. Specifically, HP-DVAL employs dual-view alignment learning to transfer the powerful feature representation capabilities from VLP models by extracting complementary features for accurate image-text alignment. To better adapt VLP models for CI-MLIC tasks, we introduce a hierarchical prompt-tuning strategy that utilizes global and local prompts to learn task-specific and context-related prior knowledge. Additionally, we design a semantic consistency loss during prompt tuning to prevent learned prompts from deviating from general knowledge embedded in VLP models. The effectiveness of our approach is validated on two CI-MLIC benchmarks: MS-COCO and VOC2007. Extensive experimental results demonstrate the superiority of our method over SOTA approaches, achieving mAP improvements of 10.0\% and 5.2\% on the long-tailed multi-label image classification task, and 6.8\% and 2.9\% on the multi-label few-shot image classification task.

Paperid: 4356, https://arxiv.org/pdf/2509.16088.pdf

Abstract:
Randomized smoothing (RS) is one of the prominent techniques to ensure the correctness of machine learning models, where point-wise robustness certificates can be derived analytically. While RS is well understood for classification, its application to generative models is unclear, since their outputs are sequences rather than labels. We resolve this by connecting generative outputs to an oracle classification task and showing that RS can still be enabled: the final response can be classified as a discrete action (e.g., service-robot commands in VLAs), as harmful vs. harmless (content moderation or toxicity detection in VLMs), or even applying oracles to cluster answers into semantically equivalent ones. Provided that the error rate for the oracle classifier comparison is bounded, we develop the theory that associates the number of samples with the corresponding robustness radius. We further derive improved scaling laws analytically relating the certified radius and accuracy to the number of samples, showing that the earlier result of 2 to 3 orders of magnitude fewer samples sufficing with minimal loss remains valid even under weaker assumptions. Together, these advances make robustness certification both well-defined and computationally feasible for state-of-the-art VLMs, as validated against recent jailbreak-style adversarial attacks.

Paperid: 4357, https://arxiv.org/pdf/2509.14143.pdf

Abstract:
Vision-language-action (VLA) models have recently emerged as a promising paradigm for robotic control, enabling end-to-end policies that ground natural language instructions into visuomotor actions. However, current VLAs often struggle to satisfy precise task constraints, such as stopping based on numeric thresholds, since their observation-to-action mappings are implicitly shaped by training data and lack explicit mechanisms for condition monitoring. In this work, we propose CLAW (CLIP-Language-Action for Weight), a framework that decouples condition evaluation from action generation. CLAW leverages a fine-tuned CLIP model as a lightweight prompt generator, which continuously monitors the digital readout of a scale and produces discrete directives based on task-specific weight thresholds. These prompts are then consumed by $Ï_0$, a flow-based VLA policy, which integrates the prompts with multi-view camera observations to produce continuous robot actions. This design enables CLAW to combine symbolic weight reasoning with high-frequency visuomotor control. We validate CLAW on three experimental setups: single-object grasping and mixed-object tasks requiring dual-arm manipulation. Across all conditions, CLAW reliably executes weight-aware behaviors and outperforms both raw-$Ï_0$ and fine-tuned $Ï_0$ models. We have uploaded the videos as supplementary materials.

Paperid: 4358, https://arxiv.org/pdf/2509.11101.pdf

Abstract:
With the rapid advancement of Multimodal Large Language Models (MLLMs), they have demonstrated exceptional capabilities across a variety of vision-language tasks. However, current evaluation benchmarks predominantly focus on objective visual question answering or captioning, inadequately assessing the models' ability to understand complex and subjective human emotions. To bridge this gap, we introduce EmoBench-Reddit, a novel, hierarchical benchmark for multimodal emotion understanding. The dataset comprises 350 meticulously curated samples from the social media platform Reddit, each containing an image, associated user-provided text, and an emotion category (sad, humor, sarcasm, happy) confirmed by user flairs. We designed a hierarchical task framework that progresses from basic perception to advanced cognition, with each data point featuring six multiple-choice questions and one open-ended question of increasing difficulty. Perception tasks evaluate the model's ability to identify basic visual elements (e.g., colors, objects), while cognition tasks require scene reasoning, intent understanding, and deep empathy integrating textual context. We ensured annotation quality through a combination of AI assistance (Claude 4) and manual verification.We conducted a comprehensive evaluation of nine leading MLLMs, including GPT-5, Gemini-2.5-pro, and GPT-4o, on EmoBench-Reddit.

Paperid: 4359, https://arxiv.org/pdf/2509.06427.pdf

Abstract:
Muzzle patterns are among the most effective biometric traits for cattle identification. Fast and accurate detection of the muzzle region as the region of interest is critical to automatic visual cattle identification.. Earlier approaches relied on manual detection, which is labor-intensive and inconsistent. Recently, automated methods using supervised models like YOLO have become popular for muzzle detection. Although effective, these methods require extensive annotated datasets and tend to be trained data-dependent, limiting their performance on new or unseen cattle. To address these limitations, this study proposes a zero-shot muzzle detection framework based on Grounding DINO, a vision-language model capable of detecting muzzles without any task-specific training or annotated data. This approach leverages natural language prompts to guide detection, enabling scalable and flexible muzzle localization across diverse breeds and environments. Our model achieves a mean Average Precision (mAP)@0.5 of 76.8\%, demonstrating promising performance without requiring annotated data. To our knowledge, this is the first research to provide a real-world, industry-oriented, and annotation-free solution for cattle muzzle detection. The framework offers a practical alternative to supervised methods, promising improved adaptability and ease of deployment in livestock monitoring applications.

Paperid: 4360, https://arxiv.org/pdf/2509.01842.pdf

Abstract:
Early stopping monitors global validation loss and halts all parameter updates simultaneously, which is computationally costly for large transformers due to the extended time required for validation inference. We propose \textit{GradES}, a novel gradient-based early stopping approach that operates within transformer components (attention projections and Feed-Forward layer matrices). We found that different components converge at varying rates during fine-tuning for both language and vision-language models. \textit{GradES} tracks the magnitude of gradient changes in backpropagation for these matrices during training. When a projection matrix's magnitude of gradient changes fall below a convergence threshold $Ï$, we exclude that projection matrix from further updates individually, eliminating costly validation passes while allowing slow converging matrices to continue learning. \textit{GradES} speeds up training time by 1.57--7.22$\times$ while simultaneously enhancing generalization through early prevention of overfitting, resulting in 1.2\% higher average accuracy in language tasks and 3.88\% on multimodal benchmarks.

Paperid: 4361, https://arxiv.org/pdf/2508.21080.pdf

Abstract:
As the computer vision community advances autonomous driving algorithms, integrating vision-based insights with sensor data remains essential for improving perception, decision making, planning, prediction, simulation, and control. Yet we must ask: Why don't we have entirely safe self-driving cars yet? A key part of the answer lies in addressing novel scenarios, one of the most critical barriers to real-world deployment. Our 2COOOL workshop provides a dedicated forum for researchers and industry experts to push the state of the art in novelty handling, including out-of-distribution hazard detection, vision-language models for hazard understanding, new benchmarking and methodologies, and safe autonomous driving practices. The 2nd Workshop on the Challenge of Out-of-Label Hazards in Autonomous Driving (2COOOL) will be held at the International Conference on Computer Vision (ICCV) 2025 in Honolulu, Hawaii, on October 19, 2025. We aim to inspire the development of new algorithms and systems for hazard avoidance, drawing on ideas from anomaly detection, open-set recognition, open-vocabulary modeling, domain adaptation, and related fields. Building on the success of its inaugural edition at the Winter Conference on Applications of Computer Vision (WACV) 2025, the workshop will feature a mix of academic and industry participation.

Paperid: 4362, https://arxiv.org/pdf/2508.13238.pdf

Abstract:
Recent advances in large vision-language models (LVLMs) have enabled a new paradigm of end-to-end document image parsing, excelling in Optical Character Recognition (OCR) tasks such as text, table, and formula recognition. However, generative LVLMs, similarly to large language models (LLMs), are prone to hallucinations--generating words that do not exist in input images. Furthermore, LVLMs are designed for general purposes and tend to be less effective on OCR tasks compared to expert models that are trained on domain-specific datasets. In this paper, we propose DianJin-OCR-R1, a reasoning-enhanced framework designed to address these limitations through training reasoning-and-tool interleaved VLMs. Given a recognition instruction, our DianJin-OCR-R1 model first recognizes the content in the input image by its own OCR capabilities, and then calls other tools (i.e., other expert models) to obtain their results as references, finally "looks again" the image and rethinks about the reasoning process to provide the final recognized content. Since architectures of expert models are tailored for specific OCR tasks, which makes them less prone to hallucinations, their results can help VLMs mitigate hallucinations. We evaluate our model on ReST and OmniDocBench, and experimental results show that our DianJin-OCR-R1 models consistently outperform their non-reasoning counterparts and expert OCR models, which proves the effectiveness of our method. Additionally, the results indicate that enhancing expert models, which are typically small and easy to iterate, enable performance improvements for VLMs.

Paperid: 4363, https://arxiv.org/pdf/2508.06571.pdf

Abstract:
Vision-Language-Action (VLA) models have demonstrated potential in autonomous driving. However, two critical challenges hinder their development: (1) Existing VLA architectures are typically based on imitation learning in open-loop setup which tends to capture the recorded behaviors in the dataset, leading to suboptimal and constrained performance, (2) Close-loop training relies heavily on high-fidelity sensor simulation, where domain gaps and computational inefficiencies pose significant barriers. In this paper, we introduce IRL-VLA, a novel close-loop Reinforcement Learning via \textbf{I}nverse \textbf{R}einforcement \textbf{L}earning reward world model with a self-built VLA approach. Our framework proceeds in a three-stage paradigm: In the first stage, we propose a VLA architecture and pretrain the VLA policy via imitation learning. In the second stage, we construct a lightweight reward world model via inverse reinforcement learning to enable efficient close-loop reward computation. To further enhance planning performance, finally, we design specialized reward world model guidence reinforcement learning via PPO(Proximal Policy Optimization) to effectively balance the safety incidents, comfortable driving, and traffic efficiency. Our approach achieves state-of-the-art performance in NAVSIM v2 end-to-end driving benchmark, 1st runner up in CVPR2025 Autonomous Grand Challenge. We hope that our framework will accelerate VLA research in close-loop autonomous driving.

Paperid: 4364, https://arxiv.org/pdf/2508.06291.pdf

Abstract:
A metric-accurate semantic 3D representation is essential for many robotic tasks. This work proposes a simple, yet powerful, way to integrate the 2D embeddings of a Vision-Language Model in a metric-accurate 3D representation at real-time. We combine a local embedding masking strategy, for a more distinct embedding distribution, with a confidence-weighted 3D integration for more reliable 3D embeddings. The resulting metric-accurate embedding representation is task-agnostic and can represent semantic concepts on a global multi-room, as well as on a local object-level. This enables a variety of interactive robotic applications that require the localisation of objects-of-interest via natural language. We evaluate our approach on a variety of real-world sequences and demonstrate that these strategies achieve a more accurate object-of-interest localisation while improving the runtime performance in order to meet our real-time constraints. We further demonstrate the versatility of our approach in a variety of interactive handheld, mobile robotics and manipulation tasks, requiring only raw image data.

Paperid: 4365, https://arxiv.org/pdf/2508.02808.pdf

Abstract:
Radiological imaging is central to diagnosis, treatment planning, and clinical decision-making. Vision-language foundation models have spurred interest in automated radiology report generation (RRG), but safe deployment requires reliable clinical evaluation of generated reports. Existing metrics often rely on surface-level similarity or behave as black boxes, lacking interpretability. We introduce ICARE (Interpretable and Clinically-grounded Agent-based Report Evaluation), an interpretable evaluation framework leveraging large language model agents and dynamic multiple-choice question answering (MCQA). Two agents, each with either the ground-truth or generated report, generate clinically meaningful questions and quiz each other. Agreement on answers captures preservation and consistency of findings, serving as interpretable proxies for clinical precision and recall. By linking scores to question-answer pairs, ICARE enables transparent, and interpretable assessment. Clinician studies show ICARE aligns significantly more with expert judgment than prior metrics. Perturbation analyses confirm sensitivity to clinical content and reproducibility, while model comparisons reveal interpretable error patterns.

Paperid: 4366, https://arxiv.org/pdf/2508.00522.pdf

Abstract:
Little research explores the correlation between the expressive ability and generalization ability of the low-rank adaptation (LoRA). Sharpness-Aware Minimization (SAM) improves model generalization for both Convolutional Neural Networks (CNNs) and Transformers by encouraging convergence to locally flat minima. However, the connection between sharpness and generalization has not been fully explored for LoRA due to the lack of tools to either empirically seek flat minima or develop theoretical methods. In this work, we propose Flat-LoRA and its efficient version i.e., EFlat-LoRA, to seek flat minima for LoRA. Concretely, we theoretically demonstrate that perturbations in the full parameter space can be transferred to the low-rank subspace. This approach eliminates the potential interference introduced by perturbations across multiple matrices in the low-rank subspace. Our extensive experiments on large language models and vision-language models demonstrate that EFlat-LoRA achieves optimize efficiency comparable to that of LoRA while simultaneously attaining comparable or even better performance. For example, on the GLUE dataset with RoBERTa-large, EFlat-LoRA outperforms LoRA and full fine-tuning by 1.0% and 0.5% on average, respectively. On vision-language models e.g., Qwen-VL-Chat shows performance improvements of 1.5% and 1.0% on SQA and VizWiz datasets, respectively. These empirical results also verify that the generalization of LoRA is closely related to sharpness, which is omitted by previous methods.

Paperid: 4367, https://arxiv.org/pdf/2507.18996.pdf

Abstract:
Modern machine learning systems operating in dynamic environments often face \textit{sequential covariate shift} (SCS), where input distributions evolve over time while the conditional distribution remains stable. We introduce FADE (Fisher-based Adaptation to Dynamic Environments), a lightweight and theoretically grounded framework for robust learning under SCS. FADE employs a shift-aware regularization mechanism anchored in Fisher information geometry, guiding adaptation by modulating parameter updates based on sensitivity and stability. To detect significant distribution changes, we propose a Cramer-Rao-informed shift signal that integrates KL divergence with temporal Fisher dynamics. Unlike prior methods requiring task boundaries, target supervision, or experience replay, FADE operates online with fixed memory and no access to target labels. Evaluated on seven benchmarks spanning vision, language, and tabular data, FADE achieves up to 19\% higher accuracy under severe shifts, outperforming methods such as TENT and DIW. FADE also generalizes naturally to federated learning by treating heterogeneous clients as temporally fragmented environments, enabling scalable and stable adaptation in decentralized settings. Theoretical analysis guarantees bounded regret and parameter consistency, while empirical results demonstrate FADE's robustness across modalities and shift intensities.

Paperid: 4368, https://arxiv.org/pdf/2507.16257.pdf

Abstract:
Defending pre-trained vision-language models (VLMs), such as CLIP, against adversarial attacks is crucial, as these models are widely used in diverse zero-shot tasks, including image classification. However, existing adversarial training (AT) methods for robust fine-tuning largely overlook the role of language in enhancing visual robustness. Specifically, (1) supervised AT methods rely on short texts (e.g., class labels) to generate adversarial perturbations, leading to overfitting to object classes in the training data, and (2) unsupervised AT avoids this overfitting but remains suboptimal against practical text-guided adversarial attacks due to its lack of semantic guidance. To address these limitations, we propose Quality Text-guided Adversarial Fine-Tuning (QT-AFT), which leverages high-quality captions during training to guide adversarial examples away from diverse semantics present in images. This enables the visual encoder to robustly recognize a broader range of image features even under adversarial noise, thereby enhancing robustness across diverse downstream tasks. QT-AFT overcomes the key weaknesses of prior methods -- overfitting in supervised AT and lack of semantic awareness in unsupervised AT -- achieving state-of-the-art zero-shot adversarial robustness and clean accuracy, evaluated across 16 zero-shot datasets. Furthermore, our comprehensive study uncovers several key insights into the role of language in enhancing vision robustness; for example, describing object properties in addition to object names further enhances zero-shot robustness. Our findings point to an urgent direction for future work -- centering high-quality linguistic supervision in robust visual representation learning.

Paperid: 4369, https://arxiv.org/pdf/2507.15493.pdf

Abstract:
We report our recent progress towards building generalist robot policies, the development of GR-3. GR-3 is a large-scale vision-language-action (VLA) model. It showcases exceptional capabilities in generalizing to novel objects, environments, and instructions involving abstract concepts. Furthermore, it can be efficiently fine-tuned with minimal human trajectory data, enabling rapid and cost-effective adaptation to new settings. GR-3 also excels in handling long-horizon and dexterous tasks, including those requiring bi-manual manipulation and mobile movement, showcasing robust and reliable performance. These capabilities are achieved through a multi-faceted training recipe that includes co-training with web-scale vision-language data, efficient fine-tuning from human trajectory data collected via VR devices, and effective imitation learning with robot trajectory data. In addition, we introduce ByteMini, a versatile bi-manual mobile robot designed with exceptional flexibility and reliability, capable of accomplishing a wide range of tasks when integrated with GR-3. Through extensive real-world experiments, we show GR-3 surpasses the state-of-the-art baseline method, $Ï_0$, on a wide variety of challenging tasks. We hope GR-3 can serve as a step towards building generalist robots capable of assisting humans in daily life.

Paperid: 4370, https://arxiv.org/pdf/2507.10442.pdf

Abstract:
Vision-language Models (VLMs) have emerged as general-purpose tools for addressing a variety of complex computer vision problems. Such models have been shown to be highly capable, but, at the same time, lacking some basic visual understanding skills. In this paper, we set out to understand the limitations of SoTA VLMs on fundamental visual tasks by constructing a series of tests that probe which components of design, specifically, may be lacking. Importantly, we go significantly beyond the current benchmarks, which simply measure the final performance of VLM response, by also comparing and contrasting it to the performance of probes trained directly on features obtained from the visual encoder, intermediate vision-language projection and LLM-decoder output. In doing so, we uncover shortcomings in VLMs and make a number of important observations about their capabilities, robustness and how they process visual information. We hope our insights will guide progress in further improving VLMs.

Paperid: 4371, https://arxiv.org/pdf/2507.05595.pdf

Abstract:
This technical report introduces PaddleOCR 3.0, an Apache-licensed open-source toolkit for OCR and document parsing. To address the growing demand for document understanding in the era of large language models, PaddleOCR 3.0 presents three major solutions: (1) PP-OCRv5 for multilingual text recognition, (2) PP-StructureV3 for hierarchical document parsing, and (3) PP-ChatOCRv4 for key information extraction. Compared to mainstream vision-language models (VLMs), these models with fewer than 100 million parameters achieve competitive accuracy and efficiency, rivaling billion-parameter VLMs. In addition to offering a high-quality OCR model library, PaddleOCR 3.0 provides efficient tools for training, inference, and deployment, supports heterogeneous hardware acceleration, and enables developers to easily build intelligent document applications.

Paperid: 4372, https://arxiv.org/pdf/2507.03165.pdf

Abstract:
Multimodal deep learning holds promise for improving clinical prediction by integrating diverse patient data, including text, imaging, time-series, and structured demographics. Contrastive learning facilitates this integration by producing a unified representation that can be reused across tasks, reducing the need for separate models or encoders. Although contrastive learning has seen success in vision-language domains, its use in clinical settings remains largely limited to image and text pairs. We propose the Pipeline for Contrastive Modality Evaluation and Encoding (PiCME), which systematically assesses five clinical data types from MIMIC: discharge summaries, radiology reports, chest X-rays, demographics, and time-series. We pre-train contrastive models on all 26 combinations of two to five modalities and evaluate their utility on in-hospital mortality and phenotype prediction. To address performance plateaus with more modalities, we introduce a Modality-Gated LSTM that weights each modality according to its contrastively learned importance. Our results show that contrastive models remain competitive with supervised baselines, particularly in three-modality settings. Performance declines beyond three modalities, which supervised models fail to recover. The Modality-Gated LSTM mitigates this drop, improving AUROC from 73.19% to 76.93% and AUPRC from 51.27% to 62.26% in the five-modality setting. We also compare contrastively learned modality importance scores with attribution scores and evaluate generalization across demographic subgroups, highlighting strengths in interpretability and fairness. PiCME is the first to scale contrastive learning across all modality combinations in MIMIC, offering guidance for modality selection, training strategies, and equitable clinical prediction.

Paperid: 4373, https://arxiv.org/pdf/2507.02957.pdf

Abstract:
Vision-Language Models (vLLMs) have emerged as powerful architectures for joint reasoning over visual and textual inputs, enabling breakthroughs in image captioning, cross modal retrieval, and multimodal dialogue. However, as these models scale to longer video sequences and richer language descriptions, the quadratic complexity of the standard attention mechanism presents a fundamental computational bottleneck. This challenge is exacerbated in vLLMs, where attention must be computed not only within modalities but also across them, leading to prohibitive memory and latency costs. In this work, we introduce the Compressed Sensing Attention Transformer (CSAT), a novel architecture that reimagines attention computation through the lens of compressed sensing. By projecting high dimensional key and value representations into a lower-dimensional subspace via random measurement matrices and reconstructing the attention outputs using sparse recovery algorithms, CSAT significantly reduces attention complexity while maintaining semantic fidelity. Applied to vLLMs, CSAT exploits the inherent compressibility of both visual and textual representations especially evident in video, where temporal redundancy is high, and in language, where cross-modal grounding is often sparse. In contrast to LLMs, which must often model entangled symbolic dependencies, vLLMs benefit from structured sparsity in alignment and scene composition, making them particularly well-suited to compressed attention. We provide a formal mathematical treatment of CSAT, demonstrate its integration into vision language pipelines, and validate its performance on standard benchmarks, highlighting its promise as a scalable, interpretable, and resource efficient solution for next generation multimodal transformers.

Paperid: 4374, https://arxiv.org/pdf/2506.21863.pdf

Abstract:
Large Vision and Language Models (LVLMs) have shown strong performance across various vision-language tasks in natural image domains. However, their application to remote sensing (RS) remains underexplored due to significant domain differences in visual appearances, object scales, and semantics. These discrepancies hider the effective understanding of RS scenes, which contain rich, multi-level semantic information spanning from coarse-to-fine levels. Hence, it limits the direct adaptation of existing LVLMs to RS imagery. To address this gap, we propose a novel LVLM framework tailored for RS understanding, incorporating two core components: Semantic-augmented Multi-level Alignment and Semantic-aware Expert Modeling. First, to align multi-level visual features, we introduce the retrieval-based Semantic Augmentation Module which enriches the visual features with relevant semantics across fine-to-coarse levels (e.g., object- and scene-level information). It is designed to retrieve relevant semantic cues from a RS semantic knowledge database, followed by aggregation of semantic cues with user query and multi-level visual features, resulting in semantically enriched representation across multiple levels. Second, for Semantic-aware Expert Modeling, we design semantic experts, where each expert is responsible for processing semantic representation at different levels separately. This enables hierarchical semantic understanding from coarse to fine levels. Evaluations across multiple RS tasks-including scene classification and VQA, etc.-demonstrate that the proposed framework achieves consistent improvements across multiple semantic levels. This highlights its capability and effectiveness in bridging the gap between general LVLMs and unique demands of RS-specific vision-language understanding.

Paperid: 4375, https://arxiv.org/pdf/2506.20990.pdf

Abstract:
Fine-tuning vision language models (VLMs) has achieved remarkable performance across various downstream tasks; yet, it requires access to model gradients through backpropagation (BP), making them unsuitable for memory-constrained, inference-only edge devices. To address this limitation, previous work has explored various BP-free fine-tuning methods. However, these approaches often rely on high-variance evolutionary strategies (ES) or zeroth-order (ZO) optimization, and often fail to achieve satisfactory performance. In this paper, we propose a hybrid Sharpness-aware Zeroth-order optimization (SharpZO) approach, specifically designed to enhance the performance of ZO VLM fine-tuning via a sharpness-aware warm-up training. SharpZO features a two-stage optimization process: a sharpness-aware ES stage that globally explores and smooths the loss landscape to construct a strong initialization, followed by a fine-grained local search via sparse ZO optimization. The entire optimization relies solely on forward passes. Detailed theoretical analysis and extensive experiments on CLIP models demonstrate that SharpZO significantly improves accuracy and convergence speed, achieving up to 7% average gain over state-of-the-art forward-only methods.

Paperid: 4376, https://arxiv.org/pdf/2506.19389.pdf

Abstract:
We investigate how the ability to recognize textual content within images emerges during the training of Vision-Language Models (VLMs). Our analysis reveals a critical phenomenon: the ability to read textual information in a given image \textbf{(text readability)} emerges abruptly after substantial training iterations, in contrast to semantic content understanding which develops gradually from the early stages of training. This delayed emergence may reflect how contrastive learning tends to initially prioritize general semantic understanding, with text-specific symbolic processing developing later. Interestingly, the ability to match images with rendered text develops even slower, indicating a deeper need for semantic integration. These findings highlight the need for tailored training strategies to accelerate robust text comprehension in VLMs, laying the groundwork for future research on optimizing multimodal learning.

Paperid: 4377, https://arxiv.org/pdf/2506.18943.pdf

Abstract:
Vision-language models (VLMs) unify computer vision and natural language processing in a single architecture capable of interpreting and describing images. Most state-of-the-art systems rely on two computationally intensive components: convolutions in the vision encoder and quadratic self-attention for multimodal fusion. This work removes both by introducing a spectral dictionary token mixer, which represents each image patch or wordpiece as a sparse combination of learnable frequency atoms. Our 1.1B-parameter prototype, SDict-VLM, achieves BLEU-4 of 39.2, CIDEr of 127.5, and SPICE of 27.0 on MS-COCO captioning, along with 50.3 percent accuracy on VQAv2. These results close approximately 85 percent of the performance gap to BLIP-2 while using 60 percent fewer parameters, 2.3 times less peak GPU memory, and 2.2 times faster inference than PaLI-3. To our knowledge, this is the first VLM to eliminate both convolutions and self-attention while matching mid-scale transformer baselines. In addition to its O(L log L) complexity, the shared frequency dictionary enables transparent cross-modal alignment and offers a tunable trade-off between accuracy and compute, paving the way for efficient and interpretable VLMs.

Paperid: 4378, https://arxiv.org/pdf/2506.17221.pdf

Abstract:
Vision-Language Navigation (VLN) is a core challenge in embodied AI, requiring agents to navigate real-world environments using natural language instructions. Current language model-based navigation systems operate on discrete topological graphs, limiting path planning to predefined node connections. We propose VLN-R1, an end-to-end framework that leverages Large Vision-Language Models (LVLM) to directly translate egocentric video streams into continuous navigation actions, adopting GRPO-based training inspired by DeepSeek-R1. To enable effective training, we first construct the VLN-Ego dataset using a 3D simulator, Habitat, and propose Long-Short Memory Sampling to balance historical and current observations. While large language models can supervise complete textual instructions, they lack fine-grained action-level control. Our framework employs a two-stage training approach: a) Supervised fine-tuning (SFT) to align the model's action sequence text predictions with expert demonstrations, followed by b) Reinforcement fine-tuning (RFT) enhanced with a Time-Decayed Reward (TDR) mechanism that strategically weights multi-step future actions. Experimental results show VLN-R1 achieves strong performance on VLN-CE benchmark. VLN-R1 proves LVLMs can drive embodied navigation and enhance task-specific reasoning through data-efficient, reward-driven post-training.

Paperid: 4379, https://arxiv.org/pdf/2506.13762.pdf

Abstract:
Data-driven approaches struggle with precise manipulation; imitation learning requires many hard-to-obtain demonstrations, while reinforcement learning yields brittle, non-generalizable policies. We introduce VisuoTactile Local (ViTaL) policy learning, a framework that solves fine-grained manipulation tasks by decomposing them into two phases: a reaching phase, where a vision-language model (VLM) enables scene-level reasoning to localize the object of interest, and a local interaction phase, where a reusable, scene-agnostic ViTaL policy performs contact-rich manipulation using egocentric vision and tactile sensing. This approach is motivated by the observation that while scene context varies, the low-level interaction remains consistent across task instances. By training local policies once in a canonical setting, they can generalize via a localize-then-execute strategy. ViTaL achieves around 90% success on contact-rich tasks in unseen environments and is robust to distractors. ViTaL's effectiveness stems from three key insights: (1) foundation models for segmentation enable training robust visual encoders via behavior cloning; (2) these encoders improve the generalizability of policies learned using residual RL; and (3) tactile sensing significantly boosts performance in contact-rich tasks. Ablation studies validate each of these insights, and we demonstrate that ViTaL integrates well with high-level VLMs, enabling robust, reusable low-level skills. Results and videos are available at https://vitalprecise.github.io.

Paperid: 4380, https://arxiv.org/pdf/2506.13166.pdf

Abstract:
Although Large Vision Language Models (LVLMs) have demonstrated remarkable performance in image understanding tasks, their computational efficiency remains a significant challenge, particularly on resource-constrained devices due to the high cost of processing large numbers of visual tokens. Recently, training-free visual token pruning methods have gained popularity as a low-cost solution to this issue. However, existing approaches suffer from two key limitations: semantic saliency-based strategies primarily focus on high cross-attention visual tokens, often neglecting visual diversity, whereas visual diversity-based methods risk inadvertently discarding semantically important tokens, especially under high compression ratios. In this paper, we introduce GreedyPrune, a training-free plug-and-play visual token pruning algorithm designed to jointly optimize semantic saliency and visual diversity. We formalize the token pruning process as a combinatorial optimization problem and demonstrate that greedy algorithms effectively balance computational efficiency with model accuracy. Extensive experiments validate the effectiveness of our approach, showing that GreedyPrune achieves state-of-the-art accuracy across various multimodal tasks and models while significantly reducing end-to-end inference latency.

Paperid: 4381, https://arxiv.org/pdf/2506.11261.pdf

Abstract:
Robotic manipulation faces a significant challenge in generalizing across unseen objects, environments and tasks specified by diverse language instructions. To improve generalization capabilities, recent research has incorporated large language models (LLMs) for planning and action execution. While promising, these methods often fall short in generating grounded plans in visual environments. Although efforts have been made to perform visual instructional tuning on LLMs for robotic manipulation, existing methods are typically constrained by single-view image input and struggle with precise object grounding. In this work, we introduce Gondola, a novel grounded vision-language planning model based on LLMs for generalizable robotic manipulation. Gondola takes multi-view images and history plans to produce the next action plan with interleaved texts and segmentation masks of target objects and locations. To support the training of Gondola, we construct three types of datasets using the RLBench simulator, namely robot grounded planning, multi-view referring expression and pseudo long-horizon task datasets. Gondola outperforms the state-of-the-art LLM-based method across all four generalization levels of the GemBench dataset, including novel placements, rigid objects, articulated objects and long-horizon tasks.

Paperid: 4382, https://arxiv.org/pdf/2506.10157.pdf

Abstract:
Medical foundation models, including language models trained on clinical notes, vision-language models on medical images, and multimodal models on electronic health records, can summarize clinical notes, answer medical questions, and assist in decision-making. Adapting these models to new populations, specialties, or settings typically requires fine-tuning, careful prompting, or retrieval from knowledge bases. This can be impractical, and limits their ability to interpret unfamiliar inputs and adjust to clinical situations not represented during training. As a result, models are prone to contextual errors, where predictions appear reasonable but fail to account for critical patient-specific or contextual information. These errors stem from a fundamental limitation that current models struggle with: dynamically adjusting their behavior across evolving contexts of medical care. In this Perspective, we outline a vision for context-switching in medical AI: models that dynamically adapt their reasoning without retraining to new specialties, populations, workflows, and clinical roles. We envision context-switching AI to diagnose, manage, and treat a wide range of diseases across specialties and regions, and expand access to medical care.

Paperid: 4383, https://arxiv.org/pdf/2506.08189.pdf

Abstract:
Scene-Graph Generation (SGG) seeks to recognize objects in an image and distill their salient pairwise relationships. Most methods depend on dataset-specific supervision to learn the variety of interactions, restricting their usefulness in open-world settings, involving novel objects and/or relations. Even methods that leverage large Vision Language Models (VLMs) typically require benchmark-specific fine-tuning. We introduce Open-World SGG, a training-free, efficient, model-agnostic framework that taps directly into the pretrained knowledge of VLMs to produce scene graphs with zero additional learning. Casting SGG as a zero-shot structured-reasoning problem, our method combines multimodal prompting, embedding alignment, and a lightweight pair-refinement strategy, enabling inference over unseen object vocabularies and relation sets. To assess this setting, we formalize an Open-World evaluation protocol that measures performance when no SGG-specific data have been observed either in terms of objects and relations. Experiments on Visual Genome, Open Images V6, and the Panoptic Scene Graph (PSG) dataset demonstrate the capacity of pretrained VLMs to perform relational understanding without task-level training.

Paperid: 4384, https://arxiv.org/pdf/2506.07127.pdf

Abstract:
Establishing a reliable and iteratively refined robotic system is essential for deploying real-world applications. While Vision-Language-Action (VLA) models are widely recognized as the foundation model for such robotic deployment, their dependence on expert demonstrations hinders the crucial capabilities of correction and learning from failures. To mitigate this limitation, we introduce a Human-assisted Action Preference Optimization method named HAPO, designed to correct deployment failures and foster effective adaptation through preference alignment for VLA models. This method begins with a human-robot collaboration framework for reliable failure correction and interaction trajectory collection through human intervention. These human-intervention trajectories are further employed within the action preference optimization process, facilitating VLA models to mitigate failure action occurrences while enhancing corrective action adaptation. Specifically, we propose an adaptive reweighting algorithm to address the issues of irreversible interactions and token probability mismatch when introducing preference optimization into VLA models, facilitating model learning from binary desirability signals derived from interactions. Through combining these modules, our human-assisted action preference optimization method ensures reliable deployment and effective learning from failure for VLA models. The experiments conducted in simulation and real-world scenarios prove superior generalization and robustness of our framework across a variety of manipulation tasks.

Paperid: 4385, https://arxiv.org/pdf/2505.23266.pdf

Abstract:
We present Adversarial Object Fusion (AdvOF), a novel attack framework targeting vision-and-language navigation (VLN) agents in service-oriented environments by generating adversarial 3D objects. While foundational models like Large Language Models (LLMs) and Vision Language Models (VLMs) have enhanced service-oriented navigation systems through improved perception and decision-making, their integration introduces vulnerabilities in mission-critical service workflows. Existing adversarial attacks fail to address service computing contexts, where reliability and quality-of-service (QoS) are paramount. We utilize AdvOF to investigate and explore the impact of adversarial environments on the VLM-based perception module of VLN agents. In particular, AdvOF first precisely aggregates and aligns the victim object positions in both 2D and 3D space, defining and rendering adversarial objects. Then, we collaboratively optimize the adversarial object with regularization between the adversarial and victim object across physical properties and VLM perceptions. Through assigning importance weights to varying views, the optimization is processed stably and multi-viewedly by iterative fusions from local updates and justifications. Our extensive evaluations demonstrate AdvOF can effectively degrade agent performance under adversarial conditions while maintaining minimal interference with normal navigation tasks. This work advances the understanding of service security in VLM-powered navigation systems, providing computational foundations for robust service composition in physical-world deployments.

Paperid: 4386, https://arxiv.org/pdf/2505.20901.pdf

Abstract:
As large vision language models(LVLMs) rapidly advance, concerns about their potential to learn and generate social biases and stereotypes are increasing. Previous studies on LVLM's stereotypes face two primary limitations: metrics that overlooked the importance of content words, and datasets that overlooked the effect of color. To address these limitations, this study introduces new evaluation metrics based on the Stereotype Content Model (SCM). We also propose BASIC, a benchmark for assessing gender, race, and color stereotypes. Using SCM metrics and BASIC, we conduct a study with eight LVLMs to discover stereotypes. As a result, we found three findings. (1) The SCM-based evaluation is effective in capturing stereotypes. (2) LVLMs exhibit color stereotypes in the output along with gender and race ones. (3) Interaction between model architecture and parameter sizes seems to affect stereotypes. We release BASIC publicly on [anonymized for review].

Paperid: 4387, https://arxiv.org/pdf/2505.16819.pdf

Abstract:
Recent advances in scene-based video generation have enabled systems to synthesize coherent visual narratives from structured prompts. However, a crucial dimension of storytelling -- character-driven dialogue and speech -- remains underexplored. In this paper, we present a modular pipeline that transforms action-level prompts into visually and auditorily grounded narrative dialogue, enriching visual storytelling with natural voice and character expression. Our method takes as input a pair of prompts per scene, where the first defines the setting and the second specifies a character's behavior. While a story generation model such as Text2Story produces the corresponding visual scene, we focus on generating expressive, character-consistent utterances grounded in both the prompts and the scene image. A pretrained vision-language encoder extracts high-level semantic features from a representative frame, capturing salient visual context. These features are then integrated with structured prompts to guide a large language model in synthesizing natural dialogue. To ensure contextual and emotional consistency across scenes, we introduce a Recursive Narrative Bank -- a speaker-aware, temporally structured memory that recursively accumulates each character's dialogue history. Inspired by Script Theory in cognitive psychology, this design enables characters to speak in ways that reflect their evolving goals, social context, and narrative roles throughout the story. Finally, we render each utterance as expressive, character-conditioned speech, resulting in fully-voiced, multimodal video narratives. Our training-free framework generalizes across diverse story settings -- from fantasy adventures to slice-of-life episodes -- offering a scalable solution for coherent, character-grounded audiovisual storytelling.

Paperid: 4388, https://arxiv.org/pdf/2505.15504.pdf

Abstract:
Deep learning has advanced computational pathology but expert annotations remain scarce. Few-shot learning mitigates annotation burdens yet suffers from overfitting and discriminative feature mischaracterization. In addition, the current few-shot multiple instance learning (MIL) approaches leverage pretrained vision-language models to alleviate these issues, but at the cost of complex preprocessing and high computational cost. We propose a Squeeze-and-Recalibrate (SR) block, a drop-in replacement for linear layers in MIL models to address these challenges. The SR block comprises two core components: a pair of low-rank trainable matrices (squeeze pathway, SP) that reduces parameter count and imposes a bottleneck to prevent spurious feature learning, and a frozen random recalibration matrix that preserves geometric structure, diversifies feature directions, and redefines the optimization objective for the SP. We provide theoretical guarantees that the SR block can approximate any linear mapping to arbitrary precision, thereby ensuring that the performance of a standard MIL model serves as a lower bound for its SR-enhanced counterpart. Extensive experiments demonstrate that our SR-MIL models consistently outperform prior methods while requiring significantly fewer parameters and no architectural changes.

Paperid: 4389, https://arxiv.org/pdf/2505.15367.pdf

Abstract:
Vision-Language Models (VLMs) have shown capabilities in interpreting visual content, but their reliability in safety-critical everyday life scenarios remains insufficiently explored. We introduce VERI (Visual Emergency Recognition Dataset), a diagnostic benchmark comprising 200 images organized into 100 contrastive pairs. Each emergency scene is paired with a visually similar but safe counterpart through human verification and refinement. Using a two-stage evaluation protocol - risk identification and emergency response - we assess 14 VLMs (2B to 124B parameters) across medical emergencies, accidents, and natural disasters. Our analysis reveals an "overreaction problem", where models accurately identify genuine emergencies (70-100 percent success rate) but produce high false-positive rates, misclassifying 31-96 percent of safe situations as dangerous. Ten safe scenarios were universally misclassified by all models regardless of scale. This "better-safe-than-sorry" bias primarily results from contextual overinterpretation (88-93 percent of errors), challenging VLM reliability in safety-critical applications. These findings highlight fundamental limitations in current VLM architectures, which persist despite increased model scale. Our results demonstrate an urgent need for strategies specifically improving contextual reasoning in ambiguous visual situations. The consistently low performance of the model indicates that these data serve effectively as a diagnostic dataset.

Paperid: 4390, https://arxiv.org/pdf/2505.15130.pdf

Abstract:
Vision-Language Models (VLMs) such as CLIP have shown remarkable performance in cross-modal tasks through large-scale contrastive pre-training. To adapt these large transformer-based models efficiently for downstream tasks, Parameter-Efficient Fine-Tuning (PEFT) techniques like LoRA have emerged as scalable alternatives to full fine-tuning, especially in few-shot scenarios. However, like traditional deep neural networks, VLMs are highly vulnerable to adversarial attacks, where imperceptible perturbations can significantly degrade model performance. Adversarial training remains the most effective strategy for improving model robustness in PEFT. In this work, we propose AdvCLIP-LoRA, the first algorithm designed to enhance the adversarial robustness of CLIP models fine-tuned with LoRA in few-shot settings. Our method formulates adversarial fine-tuning as a minimax optimization problem and provides theoretical guarantees for convergence under smoothness and nonconvex-strong-concavity assumptions. Empirical results across eight datasets using ViT-B/16 and ViT-B/32 models show that AdvCLIP-LoRA significantly improves robustness against common adversarial attacks (e.g., FGSM, PGD), without sacrificing much clean accuracy. These findings highlight AdvCLIP-LoRA as a practical and theoretically grounded approach for robust adaptation of VLMs in resource-constrained settings.

Paperid: 4391, https://arxiv.org/pdf/2505.14972.pdf

Abstract:
Large vision-language models (LVLMs) are increasingly deployed in globally distributed applications, such as tourism assistants, yet their ability to produce culturally appropriate responses remains underexplored. Existing multimodal safety benchmarks primarily focus on physical safety and overlook violations rooted in cultural norms, which can result in symbolic harm. To address this gap, we introduce CROSS, a benchmark designed to assess the cultural safety reasoning capabilities of LVLMs. CROSS includes 1,284 multilingual visually grounded queries from 16 countries, three everyday domains, and 14 languages, where cultural norm violations emerge only when images are interpreted in context. We propose CROSS-Eval, an intercultural theory-based framework that measures four key dimensions: cultural awareness, norm education, compliance, and helpfulness. Using this framework, we evaluate 21 leading LVLMs, including mixture-of-experts models and reasoning models. Results reveal significant cultural safety gaps: the best-performing model achieves only 61.79% in awareness and 37.73% in compliance. While some open-source models reach GPT-4o-level performance, they still fall notably short of proprietary models. Our results further show that increasing reasoning capacity improves cultural alignment but does not fully resolve the issue. To improve model performance, we develop two enhancement strategies: supervised fine-tuning with culturally grounded, open-ended data and preference tuning with contrastive response pairs that highlight safe versus unsafe behaviors. These methods substantially improve GPT-4o's cultural awareness (+60.14%) and compliance (+55.2%), while preserving general multimodal capabilities with minimal performance reduction on general multimodal understanding benchmarks.

Paperid: 4392, https://arxiv.org/pdf/2505.12715.pdf

Abstract:
Although fusing multiple sensor modalities can enhance object detection performance, existing fusion approaches often overlook subtle variations in environmental conditions and sensor inputs. As a result, they struggle to adaptively weight each modality under such variations. To address this challenge, we introduce Vision-Language Conditioned Fusion (VLC Fusion), a novel fusion framework that leverages a Vision-Language Model (VLM) to condition the fusion process on nuanced environmental cues. By capturing high-level environmental context such as as darkness, rain, and camera blurring, the VLM guides the model to dynamically adjust modality weights based on the current scene. We evaluate VLC Fusion on real-world autonomous driving and military target detection datasets that include image, LIDAR, and mid-wave infrared modalities. Our experiments show that VLC Fusion consistently outperforms conventional fusion baselines, achieving improved detection accuracy in both seen and unseen scenarios.

Paperid: 4393, https://arxiv.org/pdf/2505.12605.pdf

Abstract:
Recent years have witnessed outstanding advances of large vision-language models (LVLMs). In order to tackle video understanding, most of them depend upon their implicit temporal understanding capacity. As such, they have not deciphered important components that contribute to temporal understanding ability, which might limit the potential of these LVLMs for video understanding. In this work, we conduct a thorough empirical study to demystify crucial components that influence the temporal understanding of LVLMs. Our empirical study reveals that significant impacts are centered around the intermediate interface between the visual encoder and the large language model. Building on these insights, we propose a temporal-oriented recipe that encompasses temporal-oriented training schemes and an upscaled interface. Our final model developed using our recipe significantly enhances previous LVLMs on standard video understanding tasks.

Paperid: 4394, https://arxiv.org/pdf/2505.11192.pdf

Abstract:
False negatives pose a critical challenge in vision-language pretraining (VLP) due to the many-to-many correspondence between images and texts in large-scale datasets. These false negatives introduce conflicting supervision signals that degrade the learned embedding space and diminish the effectiveness of hard negative sampling. In this paper, we propose FALCON (False-negative Aware Learning of COntrastive Negatives), a learning-based mini-batch construction strategy that adaptively balances the trade-off between hard and false negatives during VLP. Rather than relying on fixed heuristics, FALCON employs a negative mining scheduler that dynamically selects negative samples of appropriate hardness for each anchor instance during mini-batch construction, guided by a proxy for cross-modal alignment improvement. Experimental results demonstrate that FALCON significantly improves performance across two widely adopted VLP frameworks (ALBEF, BLIP-2) and a broad range of downstream tasks and evaluation settings, underscoring its effectiveness and robustness in mitigating the impact of false negatives.

Paperid: 4395, https://arxiv.org/pdf/2505.11029.pdf

Abstract:
Vision-language models (VLMs) as foundation models have significantly enhanced performance across a wide range of visual and textual tasks, without requiring large-scale training from scratch for downstream tasks. However, these deterministic VLMs fail to capture the inherent ambiguity and uncertainty in natural language and visual data. Recent probabilistic post-hoc adaptation methods address this by mapping deterministic embeddings onto probability distributions; however, existing approaches do not account for the asymmetric uncertainty structure of the modalities, and the constraint that meaningful deterministic embeddings reside on a unit hypersphere, potentially leading to suboptimal performance. In this paper, we address the asymmetric uncertainty structure inherent in textual and visual data, and propose AsymVLM to build probabilistic embeddings from pre-trained VLMs on the unit hypersphere, enabling uncertainty quantification. We validate the effectiveness of the probabilistic embeddings on established benchmarks, and present comprehensive ablation studies demonstrating the inherent nature of asymmetry in the uncertainty structure of textual and visual data.

Paperid: 4396, https://arxiv.org/pdf/2505.03380.pdf

Abstract:
Medical AI assistants support doctors in disease diagnosis, medical image analysis, and report generation. However, they still face significant challenges in clinical use, including limited accuracy with multimodal content and insufficient validation in real-world settings. We propose RCMed, a full-stack AI assistant that improves multimodal alignment in both input and output, enabling precise anatomical delineation, accurate localization, and reliable diagnosis through hierarchical vision-language grounding. A self-reinforcing correlation mechanism allows visual features to inform language context, while language semantics guide pixel-wise attention, forming a closed loop that refines both modalities. This correlation is enhanced by a color region description strategy, translating anatomical structures into semantically rich text to learn shape-location-text relationships across scales. Trained on 20 million image-mask-description triplets, RCMed achieves state-of-the-art precision in contextualizing irregular lesions and subtle anatomical boundaries, excelling in 165 clinical tasks across 9 modalities. It achieved a 23.5% relative improvement in cell segmentation from microscopy images over prior methods. RCMed's strong vision-language alignment enables exceptional generalization, with state-of-the-art performance in external validation across 20 clinically significant cancer types, including novel tasks. This work demonstrates how integrated multimodal models capture fine-grained patterns, enabling human-level interpretation in complex scenarios and advancing human-centric AI healthcare.

Paperid: 4397, https://arxiv.org/pdf/2505.03334.pdf

Abstract:
In recent years, language-guided open-world aerial object detection has gained significant attention due to its better alignment with real-world application needs. However, due to limited datasets, most existing language-guided methods primarily focus on vocabulary, which fails to meet the demands of more fine-grained open-world detection. To address this limitation, we propose constructing a large-scale language-guided open-set aerial detection dataset, encompassing three levels of language guidance: from words to phrases, and ultimately to sentences. Centered around an open-source large vision-language model and integrating image-operation-based preprocessing with BERT-based postprocessing, we present the OS-W2S Label Engine, an automatic annotation pipeline capable of handling diverse scene annotations for aerial images. Using this label engine, we expand existing aerial detection datasets with rich textual annotations and construct a novel benchmark dataset, called Multi-instance Open-set Aerial Dataset (MI-OAD), addressing the limitations of current remote sensing grounding data and enabling effective open-set aerial detection. Specifically, MI-OAD contains 163,023 images and 2 million image-caption pairs, approximately 40 times larger than comparable datasets. We also employ state-of-the-art open-set methods from the natural image domain, trained on our proposed dataset, to validate the model's open-set detection capabilities. For instance, when trained on our dataset, Grounding DINO achieves improvements of 29.5 AP_{50} and 33.7 Recall@10 for sentence inputs under zero-shot transfer conditions. Both the dataset and the label engine will be released publicly.

Paperid: 4398, https://arxiv.org/pdf/2504.21226.pdf

Abstract:
Memes often merge visuals with brief text to share humor or opinions, yet some memes contain harmful messages such as hate speech. In this paper, we introduces MemeBLIP2, a light weight multimodal system that detects harmful memes by combining image and text features effectively. We build on previous studies by adding modules that align image and text representations into a shared space and fuse them for better classification. Using BLIP-2 as the core vision-language model, our system is evaluated on the PrideMM datasets. The results show that MemeBLIP2 can capture subtle cues in both modalities, even in cases with ironic or culturally specific content, thereby improving the detection of harmful material.

Paperid: 4399, https://arxiv.org/pdf/2504.14117.pdf

Abstract:
Large models such as Large Language Models (LLMs) and Vision Language Models (VLMs) have transformed artificial intelligence, powering applications in natural language processing, computer vision, and multimodal learning. However, fully fine-tuning these models remains expensive, requiring extensive computational resources, memory, and task-specific data. Parameter-Efficient Fine-Tuning (PEFT) has emerged as a promising solution that allows adapting large models to downstream tasks by updating only a small portion of parameters. This survey presents a comprehensive overview of PEFT techniques, focusing on their motivations, design principles, and effectiveness. We begin by analyzing the resource and accessibility challenges posed by traditional fine-tuning and highlight key issues, such as overfitting, catastrophic forgetting, and parameter inefficiency. We then introduce a structured taxonomy of PEFT methods -- grouped into additive, selective, reparameterized, hybrid, and unified frameworks -- and systematically compare their mechanisms and trade-offs. Beyond taxonomy, we explore the impact of PEFT across diverse domains, including language, vision, and generative modeling, showing how these techniques offer strong performance with lower resource costs. We also discuss important open challenges in scalability, interpretability, and robustness, and suggest future directions such as federated learning, domain adaptation, and theoretical grounding. Our goal is to provide a unified understanding of PEFT and its growing role in enabling practical, efficient, and sustainable use of large models.

Paperid: 4400, https://arxiv.org/pdf/2504.05303.pdf

Abstract:
We introduce InteractVLM, a novel method to estimate 3D contact points on human bodies and objects from single in-the-wild images, enabling accurate human-object joint reconstruction in 3D. This is challenging due to occlusions, depth ambiguities, and widely varying object shapes. Existing methods rely on 3D contact annotations collected via expensive motion-capture systems or tedious manual labeling, limiting scalability and generalization. To overcome this, InteractVLM harnesses the broad visual knowledge of large Vision-Language Models (VLMs), fine-tuned with limited 3D contact data. However, directly applying these models is non-trivial, as they reason only in 2D, while human-object contact is inherently 3D. Thus we introduce a novel Render-Localize-Lift module that: (1) embeds 3D body and object surfaces in 2D space via multi-view rendering, (2) trains a novel multi-view localization model (MV-Loc) to infer contacts in 2D, and (3) lifts these to 3D. Additionally, we propose a new task called Semantic Human Contact estimation, where human contact predictions are conditioned explicitly on object semantics, enabling richer interaction modeling. InteractVLM outperforms existing work on contact estimation and also facilitates 3D reconstruction from an in-the wild image. Code and models are available at https://interactvlm.is.tue.mpg.de.

Paperid: 4401, https://arxiv.org/pdf/2504.02971.pdf

Abstract:
In Visual Document Understanding (VDU) tasks, fine-tuning a pre-trained Vision-Language Model (VLM) with new datasets often falls short in optimizing the vision encoder to identify query-specific regions in text-rich document images. Existing methods that directly inject queries into model layers by modifying the network architecture often struggle to adapt to new datasets with limited annotations. To address this, we introduce QID, a novel, streamlined, architecture-preserving approach that integrates query embeddings into the vision encoder, leading to notable performance gains, particularly in data-scarce fine-tuning scenarios. Specifically, our approach introduces a dual-module framework: a query-aware module that generates a unique query vector to precisely guide the model's focus, as well as a query-agnostic module that captures the positional relationships among tokens, ensuring robust spatial understanding. Notably, both modules operate independently of the vision attention blocks, facilitating targeted learning of query embeddings and enhancing visual semantic identification. Experiments with OCR-free VLMs across multiple datasets demonstrate significant performance improvements using our method, especially in handling text-rich documents in data-scarce environments.

Paperid: 4402, https://arxiv.org/pdf/2503.21904.pdf

Abstract:
The rapid advancements in large language models (LLMs) have spurred growing interest in LLM-based video anomaly detection (VAD). However, existing approaches predominantly focus on video-level anomaly question answering or offline detection, ignoring the real-time nature essential for practical VAD applications. To bridge this gap and facilitate the practical deployment of LLM-based VAD, we introduce AssistPDA, the first online video anomaly surveillance assistant that unifies video anomaly prediction, detection, and analysis (VAPDA) within a single framework. AssistPDA enables real-time inference on streaming videos while supporting interactive user engagement. Notably, we introduce a novel event-level anomaly prediction task, enabling proactive anomaly forecasting before anomalies fully unfold. To enhance the ability to model intricate spatiotemporal relationships in anomaly events, we propose a Spatio-Temporal Relation Distillation (STRD) module. STRD transfers the long-term spatiotemporal modeling capabilities of vision-language models (VLMs) from offline settings to real-time scenarios. Thus it equips AssistPDA with a robust understanding of complex temporal dependencies and long-sequence memory. Additionally, we construct VAPDA-127K, the first large-scale benchmark designed for VLM-based online VAPDA. Extensive experiments demonstrate that AssistPDA outperforms existing offline VLM-based approaches, setting a new state-of-the-art for real-time VAPDA. Our dataset and code will be open-sourced to facilitate further research in the community.

Paperid: 4403, https://arxiv.org/pdf/2503.16965.pdf

Abstract:
Vision Language Models exhibited immense potential for embodied AI, yet they often lack the sophisticated situational reasoning required for complex decision-making. This paper shows that VLMs can achieve surprisingly strong decision-making performance when visual scenes are represented merely as text-only descriptions, suggesting foundational reasoning can be effectively learned from language. Motivated by this insight, we propose Praxis-VLM, a reasoning VLM for vision-grounded decision-making. Praxis-VLM employs the GRPO algorithm on textual scenarios to instill robust reasoning capabilities, where models learn to evaluate actions and their consequences. These reasoning skills, acquired purely from text, successfully transfer to multimodal inference with visual inputs, significantly reducing reliance on scarce paired image-text training data. Experiments across diverse decision-making benchmarks demonstrate that Praxis-VLM substantially outperforms standard supervised fine-tuning, exhibiting superior performance and generalizability. Further analysis confirms that our models engage in explicit and effective reasoning, underpinning their enhanced performance and adaptability.

Paperid: 4404, https://arxiv.org/pdf/2503.16401.pdf

Abstract:
Large language models (LLMs) and Vision language models (VLMs) have been able to perform various forms of reasoning tasks in a wide range of scenarios, but are they truly engaging in task abstraction and rule-based reasoning beyond mere memorization and pattern matching? To answer this question, we propose a novel experimental approach, Misleading Fine-Tuning (MisFT), to examine whether LLMs/VLMs perform abstract reasoning by altering their original understanding of fundamental rules. In particular, by constructing a dataset with math expressions that contradict correct operation principles, we fine-tune the model to learn those contradictory rules and assess its generalization ability on different test domains. Through a series of experiments, we find that current LLMs/VLMs are capable of effectively applying contradictory rules to solve practical math word problems and math expressions represented by images, implying the presence of an internal mechanism that abstracts before reasoning.

Paperid: 4405, https://arxiv.org/pdf/2503.14754.pdf

Abstract:
Street scene datasets, collected from Street View or dashboard cameras, offer a promising means of detecting urban objects and incidents like street flooding. However, a major challenge in using these datasets is their lack of reliable labels: there are myriad types of incidents, many types occur rarely, and ground-truth measures of where incidents occur are lacking. Here, we propose BayFlood, a two-stage approach which circumvents this difficulty. First, we perform zero-shot classification of where incidents occur using a pretrained vision-language model (VLM). Second, we fit a spatial Bayesian model on the VLM classifications. The zero-shot approach avoids the need to annotate large training sets, and the Bayesian model provides frequent desiderata in urban settings - principled measures of uncertainty, smoothing across locations, and incorporation of external data like stormwater accumulation zones. We comprehensively validate this two-stage approach, showing that VLMs provide strong zero-shot signal for floods across multiple cities and time periods, the Bayesian model improves out-of-sample prediction relative to baseline methods, and our inferred flood risk correlates with known external predictors of risk. Having validated our approach, we show it can be used to improve urban flood detection: our analysis reveals 113,738 people who are at high risk of flooding overlooked by current methods, identifies demographic biases in existing methods, and suggests locations for new flood sensors. More broadly, our results showcase how Bayesian modeling of zero-shot LM annotations represents a promising paradigm because it avoids the need to collect large labeled datasets and leverages the power of foundation models while providing the expressiveness and uncertainty quantification of Bayesian models.

Paperid: 4406, https://arxiv.org/pdf/2503.14377.pdf

Abstract:
Despite the growing scale of medical Vision-Language datasets, the impact of dataset quality on model performance remains under-explored. We introduce Open-PMC, a high-quality medical dataset from PubMed Central, containing 2.2 million image-text pairs, enriched with image modality annotations, subfigures, and summarized in-text references. Notably, the in-text references provide richer medical context, extending beyond the abstract information typically found in captions. Through extensive experiments, we benchmark Open-PMC against larger datasets across retrieval and zero-shot classification tasks. Our results show that dataset quality-not just size-drives significant performance gains. We complement our benchmark with an in-depth analysis of feature representation. Our findings highlight the crucial role of data curation quality in advancing multimodal medical AI. We release Open-PMC, along with the trained models and our codebase.

Paperid: 4407, https://arxiv.org/pdf/2503.13369.pdf

Abstract:
Often, the needs and visual abilities differ between the annotator group and the end user group. Generating detailed diagram descriptions for blind and low-vision (BLV) users is one such challenging domain. Sighted annotators could describe visuals with ease, but existing studies have shown that direct generations by them are costly, bias-prone, and somewhat lacking by BLV standards. In this study, we ask sighted individuals to assess -- rather than produce -- diagram descriptions generated by vision-language models (VLM) that have been guided with latent supervision via a multi-pass inference. The sighted assessments prove effective and useful to professional educators who are themselves BLV and teach visually impaired learners. We release Sightation, a collection of diagram description datasets spanning 5k diagrams and 137k samples for completion, preference, retrieval, question answering, and reasoning training purposes and demonstrate their fine-tuning potential in various downstream tasks.

Paperid: 4408, https://arxiv.org/pdf/2503.13356.pdf

Abstract:
We present PORTAL, a novel framework for developing artificial intelligence agents capable of playing thousands of 3D video games through language-guided policy generation. By transforming decision-making problems into language modeling tasks, our approach leverages large language models (LLMs) to generate behavior trees represented in domain-specific language (DSL). This method eliminates the computational burden associated with traditional reinforcement learning approaches while preserving strategic depth and rapid adaptability. Our framework introduces a hybrid policy structure that combines rule-based nodes with neural network components, enabling both high-level strategic reasoning and precise low-level control. A dual-feedback mechanism incorporating quantitative game metrics and vision-language model analysis facilitates iterative policy improvement at both tactical and strategic levels. The resulting policies are instantaneously deployable, human-interpretable, and capable of generalizing across diverse gaming environments. Experimental results demonstrate PORTAL's effectiveness across thousands of first-person shooter (FPS) games, showcasing significant improvements in development efficiency, policy generalization, and behavior diversity compared to traditional approaches. PORTAL represents a significant advancement in game AI development, offering a practical solution for creating sophisticated agents that can operate across thousands of commercial video games with minimal development overhead. Experiment results on the 3D video games are best viewed on https://zhongwen.one/projects/portal .

Paperid: 4409, https://arxiv.org/pdf/2503.12329.pdf

Abstract:
Image captioning has been a longstanding challenge in vision-language research. With the rise of LLMs, modern Vision-Language Models (VLMs) generate detailed and comprehensive image descriptions. However, benchmarking the quality of such captions remains unresolved. This paper addresses two key questions: (1) How well do current VLMs actually perform on image captioning, particularly compared to humans? We built CapArena, a platform with over 6000 pairwise caption battles and high-quality human preference votes. Our arena-style evaluation marks a milestone, showing that leading models like GPT-4o achieve or even surpass human performance, while most open-source models lag behind. (2) Can automated metrics reliably assess detailed caption quality? Using human annotations from CapArena, we evaluate traditional and recent captioning metrics, as well as VLM-as-a-Judge. Our analysis reveals that while some metrics (e.g., METEOR) show decent caption-level agreement with humans, their systematic biases lead to inconsistencies in model ranking. In contrast, VLM-as-a-Judge demonstrates robust discernment at both the caption and model levels. Building on these insights, we release CapArena-Auto, an accurate and efficient automated benchmark for detailed captioning, achieving 94.3% correlation with human rankings at just $4 per test. Data and resources will be open-sourced at https://caparena.github.io.

Paperid: 4410, https://arxiv.org/pdf/2503.11576.pdf

Abstract:
We introduce SmolDocling, an ultra-compact vision-language model targeting end-to-end document conversion. Our model comprehensively processes entire pages by generating DocTags, a new universal markup format that captures all page elements in their full context with location. Unlike existing approaches that rely on large foundational models, or ensemble solutions that rely on handcrafted pipelines of multiple specialized models, SmolDocling offers an end-to-end conversion for accurately capturing content, structure and spatial location of document elements in a 256M parameters vision-language model. SmolDocling exhibits robust performance in correctly reproducing document features such as code listings, tables, equations, charts, lists, and more across a diverse range of document types including business documents, academic papers, technical reports, patents, and forms -- significantly extending beyond the commonly observed focus on scientific papers. Additionally, we contribute novel publicly sourced datasets for charts, tables, equations, and code recognition. Experimental results demonstrate that SmolDocling competes with other Vision Language Models that are up to 27 times larger in size, while reducing computational requirements substantially. The model is currently available, datasets will be publicly available soon.

Paperid: 4411, https://arxiv.org/pdf/2503.10152.pdf

Abstract:
Open-vocabulary object detection (OVD) aims to detect objects beyond the training annotations, where detectors are usually aligned to a pre-trained vision-language model, eg, CLIP, to inherit its generalizable recognition ability so that detectors can recognize new or novel objects. However, previous works directly align the feature space with CLIP and fail to learn the semantic knowledge effectively. In this work, we propose a hierarchical semantic distillation framework named HD-OVD to construct a comprehensive distillation process, which exploits generalizable knowledge from the CLIP model in three aspects. In the first hierarchy of HD-OVD, the detector learns fine-grained instance-wise semantics from the CLIP image encoder by modeling relations among single objects in the visual space. Besides, we introduce text space novel-class-aware classification to help the detector assimilate the highly generalizable class-wise semantics from the CLIP text encoder, representing the second hierarchy. Lastly, abundant image-wise semantics containing multi-object and their contexts are also distilled by an image-wise contrastive distillation. Benefiting from the elaborated semantic distillation in triple hierarchies, our HD-OVD inherits generalizable recognition ability from CLIP in instance, class, and image levels. Thus, we boost the novel AP on the OV-COCO dataset to 46.4% with a ResNet50 backbone, which outperforms others by a clear margin. We also conduct extensive ablation studies to analyze how each component works.

Paperid: 4412, https://arxiv.org/pdf/2503.02917.pdf

Abstract:
Recent advancements in deep learning have shown significant potential for classifying retinal diseases using color fundus images. However, existing works predominantly rely exclusively on image data, lack interpretability in their diagnostic decisions, and treat medical professionals primarily as annotators for ground truth labeling. To fill this gap, we implement two key strategies: extracting interpretable concepts of retinal diseases using the knowledge base of GPT models and incorporating these concepts as a language component in prompt-learning to train vision-language (VL) models with both fundus images and their associated concepts. Our method not only improves retinal disease classification but also enriches few-shot and zero-shot detection (novel disease detection), while offering the added benefit of concept-based model interpretability. Our extensive evaluation across two diverse retinal fundus image datasets illustrates substantial performance gains in VL-model based few-shot methodologies through our concept integration approach, demonstrating an average improvement of approximately 5.8\% and 2.7\% mean average precision for 16-shot learning and zero-shot (novel class) detection respectively. Our method marks a pivotal step towards interpretable and efficient retinal disease recognition for real-world clinical applications.

Paperid: 4413, https://arxiv.org/pdf/2503.02748.pdf

Abstract:
From early Movement Primitive (MP) techniques to modern Vision-Language Models (VLMs), autonomous manipulation has remained a pivotal topic in robotics. As two extremes, VLM-based methods emphasize zero-shot and adaptive manipulation but struggle with fine-grained planning. In contrast, MP-based approaches excel in precise trajectory generalization but lack decision-making ability. To leverage the strengths of the two frameworks, we propose VL-MP, which integrates VLM with Kernelized Movement Primitives (KMP) via a low-distortion decision information transfer bridge, enabling fine-grained robotic manipulation under ambiguous situations. One key of VL-MP is the accurate representation of task decision parameters through semantic keypoints constraints, leading to more precise task parameter generation. Additionally, we introduce a local trajectory feature-enhanced KMP to support VL-MP, thereby achieving shape preservation for complex trajectories. Extensive experiments conducted in complex real-world environments validate the effectiveness of VL-MP for adaptive and fine-grained manipulation.

Paperid: 4414, https://arxiv.org/pdf/2502.19777.pdf

Abstract:
Prompt tuning has become a popular strategy for adapting Vision-Language Models (VLMs) to zero/few-shot visual recognition tasks. Some prompting techniques introduce prior knowledge due to its richness, but when learnable tokens are randomly initialized and disconnected from prior knowledge, they tend to overfit on seen classes and struggle with domain shifts for unseen ones. To address this issue, we propose the InPK model, which infuses class-specific prior knowledge into the learnable tokens during initialization, thus enabling the model to explicitly focus on class-relevant information. Furthermore, to mitigate the weakening of class information by multi-layer encoders, we continuously reinforce the interaction between learnable tokens and prior knowledge across multiple feature levels. This progressive interaction allows the learnable tokens to better capture the fine-grained differences and universal visual concepts within prior knowledge, enabling the model to extract more discriminative and generalized text features. Even for unseen classes, the learned interaction allows the model to capture their common representations and infer their appropriate positions within the existing semantic structure. Moreover, we introduce a learnable text-to-vision projection layer to accommodate the text adjustments, ensuring better alignment of visual-text semantics. Extensive experiments on 11 recognition datasets show that InPK significantly outperforms state-of-the-art methods in multiple zero/few-shot image classification tasks.

Paperid: 4415, https://arxiv.org/pdf/2502.17028.pdf

Abstract:
Multimodal alignment is crucial for various downstream tasks such as cross-modal generation and retrieval. Previous multimodal approaches like CLIP utilize InfoNCE to maximize mutual information, primarily aligning pairwise samples across modalities while overlooking distributional differences. In addition, InfoNCE has inherent conflict in terms of alignment and uniformity in multimodality, leading to suboptimal alignment with modality gaps. To overcome the limitations, we propose CS-Aligner, a novel framework that performs distributional vision-language alignment by integrating Cauchy-Schwarz (CS) divergence with mutual information. CS-Aligner captures both the global distribution information of each modality and the pairwise semantic relationships. We find that the CS divergence seamlessly addresses the InfoNCE's alignment-uniformity conflict and serves complementary roles with InfoNCE, yielding tighter and more precise alignment. Moreover, by introducing distributional alignment, CS-Aligner enables incorporating additional information from unpaired data and token-level representations, enhancing flexible and fine-grained alignment in practice. Experiments on text-to-image generation and cross-modality retrieval tasks demonstrate the effectiveness of our method on vision-language alignment.

Paperid: 4416, https://arxiv.org/pdf/2502.15210.pdf

Abstract:
Understanding how effectively large vision language models (VLMs) compare visual inputs is crucial across numerous applications, yet this fundamental capability remains insufficiently assessed. While VLMs are increasingly deployed for tasks requiring comparative judgment, including automated evaluation, re-ranking, and retrieval-augmented generation, no systematic framework exists to measure their performance in these scenarios. We present PairBench, a simple framework that evaluates VLMs as customizable similarity tools using widely available image datasets. Our approach introduces four key metrics for reliable comparison: alignment with human annotations, consistency across pair ordering, distribution smoothness, and controllability through prompting. Our analysis reveals that no model consistently excels across all metrics, with each demonstrating distinct strengths and weaknesses. Most concerning is the widespread inability of VLMs to maintain symmetric similarity scores. Interestingly, we demonstrate that performance on our benchmark strongly correlates with popular benchmarks used for more complex tasks, while providing additional metrics into controllability, smoothness and ordering. This makes PairBench a unique and comprehensive framework to evaluate the performance of VLMs for automatic evaluation depending on the task.

Paperid: 4417, https://arxiv.org/pdf/2502.12852.pdf

Abstract:
Existing multilingual vision-language (VL) benchmarks often only cover a handful of languages. Consequently, evaluations of large vision-language models (LVLMs) predominantly target high-resource languages, underscoring the need for evaluation data for low-resource languages. To address this limitation, we introduce MVL-SIB, a massively multilingual vision-language benchmark that evaluates both cross-modal and text-only topical matching across 205 languages -- over 100 more than the most multilingual existing VL benchmarks encompass. We then benchmark a range of of open-weight LVLMs together with GPT-4o(-mini) on MVL-SIB. Our results reveal that LVLMs struggle in cross-modal topic matching in lower-resource languages, performing no better than chance on languages like N'Koo. Our analysis further reveals that VL support in LVLMs declines disproportionately relative to textual support for lower-resource languages, as evidenced by comparison of cross-modal and text-only topical matching performance. We further observe that open-weight LVLMs do not benefit from representing a topic with more than one image, suggesting that these models are not yet fully effective at handling multi-image tasks. By correlating performance on MVL-SIB with other multilingual VL benchmarks, we highlight that MVL-SIB serves as a comprehensive probe of multilingual VL understanding in LVLMs.

Paperid: 4418, https://arxiv.org/pdf/2502.07617.pdf

Abstract:
We provide an empirical investigation of the potential of pre-training vision-language models on an unprecedented scale: 100 billion examples. We find that model performance tends to saturate at this scale on many common Western-centric classification and retrieval benchmarks, such as COCO Captions. Nevertheless, tasks of cultural diversity achieve more substantial gains from the 100-billion scale web data, thanks to its coverage of long-tail concepts. Furthermore, we analyze the model's multilinguality and show gains in low-resource languages as well. In addition, we observe that reducing the size of the pretraining dataset via quality filters like using CLIP, typically used to enhance performance, may inadvertently reduce the cultural diversity represented even in large-scale datasets. Our results highlight that while traditional benchmarks may not benefit significantly from scaling noisy, raw web data to 100 billion examples, this data scale is vital for building truly inclusive multimodal systems.

Paperid: 4419, https://arxiv.org/pdf/2502.07409.pdf

Abstract:
Whole slide pathology image classification presents challenges due to gigapixel image sizes and limited annotation labels, hindering model generalization. This paper introduces a prompt learning method to adapt large vision-language models for few-shot pathology classification. We first extend the Prov-GigaPath vision foundation model, pre-trained on 1.3 billion pathology image tiles, into a vision-language model by adding adaptors and aligning it with medical text encoders via contrastive learning on 923K image-text pairs. The model is then used to extract visual features and text embeddings from few-shot annotations and fine-tunes with learnable prompt embeddings. Unlike prior methods that combine prompts with frozen features using prefix embeddings or self-attention, we propose multi-granular attention that compares interactions between learnable prompts with individual image patches and groups of them. This approach improves the model's ability to capture both fine-grained details and broader context, enhancing its recognition of complex patterns across sub-regions. To further improve accuracy, we leverage (unbalanced) optimal transport-based visual-text distance to secure model robustness by mitigating perturbations that might occur during the data augmentation process. Empirical experiments on lung, kidney, and breast pathology modalities validate the effectiveness of our approach; thereby, we surpass several of the latest competitors and consistently improve performance across diverse architectures, including CLIP, PLIP, and Prov-GigaPath integrated PLIP. We release our implementations and pre-trained models at this MGPATH.

Paperid: 4420, https://arxiv.org/pdf/2502.02069.pdf

Abstract:
The rapid advancements in vision-language models (VLMs), such as CLIP, have intensified the need to address distribution shifts between training and testing datasets. Although prior Test-Time Training (TTT) techniques for VLMs have demonstrated robust performance, they predominantly rely on tuning text prompts, a process that demands substantial computational resources and is heavily dependent on entropy-based loss. In this paper, we propose LoRA-TTT, a novel TTT method that leverages Low-Rank Adaptation (LoRA), applied exclusively to the image encoder of VLMs. By introducing LoRA and updating only its parameters during test time, our method offers a simple yet effective TTT approach, retaining the model's initial generalization capability while achieving substantial performance gains with minimal memory and runtime overhead. Additionally, we introduce a highly efficient reconstruction loss tailored for TTT. Our method can adapt to diverse domains by combining these two losses, without increasing memory consumption or runtime. Extensive experiments on two benchmarks, covering 15 datasets, demonstrate that our method improves the zero-shot top-1 accuracy of CLIP-ViT-B/16 by an average of 5.79% on the OOD benchmark and 1.36% on the fine-grained benchmark, efficiently surpassing test-time prompt tuning, without relying on any external models or cache.

Paperid: 4421, https://arxiv.org/pdf/2501.16246.pdf

Abstract:
Brain tumor segmentation is important for diagnosis of the tumor, and current deep-learning methods rely on a large set of annotated images for training, with high annotation costs. Unsupervised segmentation is promising to avoid human annotations while the performance is often limited. In this study, we present a novel unsupervised segmentation approach that leverages the capabilities of foundation models, and it consists of three main steps: (1) A vision-language model (i.e., CLIP) is employed to obtain image-level pseudo-labels for training a classification network. Class Activation Mapping (CAM) is then employed to extract Regions of Interest (ROIs), where an adaptive masking-based data augmentation is used to enhance ROI identification.(2) The ROIs are used to generate bounding box and point prompts for the Segment Anything Model (SAM) to obtain segmentation pseudo-labels. (3) A 3D segmentation network is trained with the SAM-derived pseudo-labels, where low-quality pseudo-labels are filtered out in a self-learning process based on the similarity between the SAM's output and the network's prediction. Evaluation on the BraTS2020 dataset demonstrates that our approach obtained an average Dice Similarity Score (DSC) of 85.60%, outperforming five state-of-the-art unsupervised segmentation methods by more than 10 percentage points. Besides, our approach outperforms directly using SAM for zero-shot inference, and its performance is close to fully supervised learning.

Paperid: 4422, https://arxiv.org/pdf/2501.14148.pdf

Abstract:
We present SelfPrompt, a novel prompt-tuning approach for vision-language models (VLMs) in a semi-supervised learning setup. Existing methods for tuning VLMs in semi-supervised setups struggle with the negative impact of the miscalibrated VLMs on pseudo-labelling, and the accumulation of noisy pseudo-labels. SelfPrompt addresses these challenges by introducing a cluster-guided pseudo-labelling method that improves pseudo-label accuracy, and a confidence-aware semi-supervised learning module that maximizes the utilization of unlabelled data by combining supervised learning and weakly-supervised learning. Additionally, we investigate our method in an active semi-supervised learning setup, where the labelled set is strategically selected to ensure the best utilization of a limited labelling budget. To this end, we propose a weakly-supervised sampling technique that selects a diverse and representative labelled set, which can be seamlessly integrated into existing methods to enhance their performance. We conduct extensive evaluations across 13 datasets, significantly surpassing state-of-the-art performances with average improvements of 6.23% in standard semi-supervised learning, 6.25% in active semi-supervised learning, and 4.9% in base-to-novel generalization, using a 2-shot setup. Furthermore, SelfPrompt shows excellent generalization in single-shot settings, achieving an average improvement of 11.78%.

Paperid: 4423, https://arxiv.org/pdf/2501.10011.pdf

Abstract:
Current popular Large Vision-Language Models (LVLMs) are suffering from Hallucinations on Object Attributes (HoOA), leading to incorrect determination of fine-grained attributes in the input images. Leveraging significant advancements in 3D generation from a single image, this paper proposes a novel method to mitigate HoOA in LVLMs. This method utilizes multiview images sampled from generated 3D representations as visual prompts for LVLMs, thereby providing more visual information from other viewpoints. Furthermore, we observe the input order of multiple multiview images significantly affects the performance of LVLMs. Consequently, we have devised Multiview Image Augmented VLM (MIAVLM), incorporating a Multiview Attributes Perceiver (MAP) submodule capable of simultaneously eliminating the influence of input image order and aligning visual information from multiview images with Large Language Models (LLMs). Besides, we designed and employed negative instructions to mitigate LVLMs' bias towards ``Yes" responses. Comprehensive experiments demonstrate the effectiveness of our method.

Paperid: 4424, https://arxiv.org/pdf/2501.01428.pdf

Abstract:
In recent years, 2D Vision-Language Models (VLMs) have made significant strides in image-text understanding tasks. However, their performance in 3D spatial comprehension, which is critical for embodied intelligence, remains limited. Recent advances have leveraged 3D point clouds and multi-view images as inputs, yielding promising results. However, we propose exploring a purely vision-based solution inspired by human perception, which merely relies on visual cues for 3D spatial understanding. This paper empirically investigates the limitations of VLMs in 3D spatial knowledge, revealing that their primary shortcoming lies in the lack of global-local correspondence between the scene and individual frames. To address this, we introduce GPT4Scene, a novel visual prompting paradigm in VLM training and inference that helps build the global-local relationship, significantly improving the 3D spatial understanding of indoor scenes. Specifically, GPT4Scene constructs a Bird's Eye View (BEV) image from the video and marks consistent object IDs across both frames and the BEV image. The model then inputs the concatenated BEV image and video frames with markers. In zero-shot evaluations, GPT4Scene improves performance over closed-source VLMs like GPT-4o. Additionally, we prepare a processed video dataset consisting of 165K text annotation to fine-tune open-source VLMs, achieving state-of-the-art performance on all 3D understanding tasks. Surprisingly, after training with the GPT4Scene paradigm, VLMs consistently improve during inference, even without object marker prompting and BEV image as explicit correspondence. It demonstrates that the proposed paradigm helps VLMs develop an intrinsic ability to understand 3D scenes, which paves the way for a seamless approach to extending pre-trained VLMs for 3D scene understanding.

Paperid: 4425, https://arxiv.org/pdf/2412.13510.pdf

Abstract:
Existing cross-modal retrieval methods typically rely on large-scale vision-language pair data. This makes it challenging to efficiently develop a cross-modal retrieval model for under-resourced languages of interest. Therefore, Cross-lingual Cross-modal Retrieval (CCR), which aims to align vision and the low-resource language (the target language) without using any human-labeled target-language data, has gained increasing attention. As a general parameter-efficient way, a common solution is to utilize adapter modules to transfer the vision-language alignment ability of Vision-Language Pretraining (VLP) models from a source language to a target language. However, these adapters are usually static once learned, making it difficult to adapt to target-language captions with varied expressions. To alleviate it, we propose Dynamic Adapter with Semantics Disentangling (DASD), whose parameters are dynamically generated conditioned on the characteristics of the input captions. Considering that the semantics and expression styles of the input caption largely influence how to encode it, we propose a semantic disentangling module to extract the semantic-related and semantic-agnostic features from the input, ensuring that generated adapters are well-suited to the characteristics of input caption. Extensive experiments on two image-text datasets and one video-text dataset demonstrate the effectiveness of our model for cross-lingual cross-modal retrieval, as well as its good compatibility with various VLP models.

Paperid: 4426, https://arxiv.org/pdf/2412.08802.pdf

Abstract:
Contrastive Language-Image Pretraining (CLIP) has been widely used for crossmodal information retrieval and multimodal understanding tasks. However, CLIP models are mainly optimized for crossmodal vision-language tasks and underperform in single-mode text tasks. Moreover, these models are often trained on English datasets and therefore lack multilingual understanding. Additionally, from a visual understanding perspective, previous CLIP-based models exhibit insufficient understanding of visually rich documents. In this work, we propose jina-clip-v2, a contrastive vision-language model trained on text pairs, triplets and image-text pairs via a multi-task and multi-stage contrastive learning paradigm in order to support both text-only and crossmodal tasks. We employ a multilingual text encoder and expand the training dataset to include multilingual texts from 29 non-English languages, including Hindi, Chinese, German, French, and others, as well as images of visually rich documents. We evaluate the model's performance and show that jina-clip-v2 achieves notable improvements over state-of-the-art CLIP-based models in zero-shot text-only retrieval, semantic textual similarity, and crossmodal retrieval tasks in both English and multilingual settings. jina-clip-v2 also provides for flexibility in embedding dimensionality, enabling users to select the granularity of the representations. jina-clip-v2 is publicly available at https://huggingface.co/jinaai/jina-clip-v2.

Paperid: 4427, https://arxiv.org/pdf/2412.05725.pdf

Abstract:
The commonsense reasoning capabilities of vision-language models (VLMs), especially in abductive reasoning and defeasible reasoning, remain poorly understood. Most benchmarks focus on typical visual scenarios, making it difficult to discern whether model performance stems from keen perception and reasoning skills, or reliance on pure statistical recall. We argue that by focusing on atypical events in videos, clearer insights can be gained on the core capabilities of VLMs. Explaining and understanding such out-of-distribution events requires models to extend beyond basic pattern recognition and regurgitation of their prior knowledge. To this end, we introduce BlackSwanSuite, a benchmark for evaluating VLMs' ability to reason about unexpected events through abductive and defeasible tasks. Our tasks artificially limit the amount of visual information provided to models while questioning them about hidden unexpected events, or provide new visual information that could change an existing hypothesis about the event. We curate a comprehensive benchmark suite comprising over 3,800 MCQ, 4,900 generative and 6,700 yes/no questions, spanning 1,655 videos. After extensively evaluating various state-of-the-art VLMs, including GPT-4o and Gemini 1.5 Pro, as well as open-source VLMs such as LLaVA-Video, we find significant performance gaps of up to 32% from humans on these tasks. Our findings reveal key limitations in current VLMs, emphasizing the need for enhanced model architectures and training strategies. Our data and leaderboard is available at blackswan.cs.ubc.ca.

Paperid: 4428, https://arxiv.org/pdf/2412.03567.pdf

Abstract:
Robotics, autonomous driving, augmented reality, and many embodied computer vision applications must quickly react to user-defined events unfolding in real time. We address this setting by proposing a novel task for multimodal video understanding-Streaming Detection of Queried Event Start (SDQES). The goal of SDQES is to identify the beginning of a complex event as described by a natural language query, with high accuracy and low latency. We introduce a new benchmark based on the Ego4D dataset, as well as new task-specific metrics to study streaming multimodal detection of diverse events in an egocentric video setting. Inspired by parameter-efficient fine-tuning methods in NLP and for video tasks, we propose adapter-based baselines that enable image-to-video transfer learning, allowing for efficient online video modeling. We evaluate three vision-language backbones and three adapter architectures on both short-clip and untrimmed video settings.

Paperid: 4429, https://arxiv.org/pdf/2412.01292.pdf

Abstract:
Research on 3D Vision-Language Models (3D-VLMs) is gaining increasing attention, which is crucial for developing embodied AI within 3D scenes, such as visual navigation and embodied question answering. Due to the high density of visual features, especially in large 3D scenes, accurately locating task-relevant visual information is challenging. Existing works attempt to segment all objects and consider their features as scene representations. However, these task-agnostic object features include much redundant information and missing details for the task-relevant area. To tackle these problems, we propose LSceneLLM, an adaptive framework that automatically identifies task-relevant areas by leveraging LLM's visual preference for different tasks, followed by a plug-and-play scene magnifier module to capture fine-grained details in focused areas. Specifically, a dense token selector examines the attention map of LLM to identify visual preferences for the instruction input. It then magnifies fine-grained details of the focusing area. An adaptive self-attention module is leveraged to fuse the coarse-grained and selected fine-grained visual information. To comprehensively evaluate the large scene understanding ability of 3D-VLMs, we further introduce a cross-room understanding benchmark, XR-Scene, which contains a series of large scene understanding tasks including XR-QA, XR-EmbodiedPlanning, and XR-SceneCaption. Experiments show that our method surpasses existing methods on both large scene understanding and existing scene understanding benchmarks. Plunging our scene magnifier module into the existing 3D-VLMs also brings significant improvement.

Paperid: 4430, https://arxiv.org/pdf/2411.17150.pdf

Abstract:
Open-Vocabulary Semantic Segmentation (OVSS) has advanced with recent vision-language models (VLMs), enabling segmentation beyond predefined categories through various learning schemes. Notably, training-free methods offer scalable, easily deployable solutions for handling unseen data, a key goal of OVSS. Yet, a critical issue persists: lack of object-level context consideration when segmenting complex objects in the challenging environment of OVSS based on arbitrary query prompts. This oversight limits models' ability to group semantically consistent elements within object and map them precisely to user-defined arbitrary classes. In this work, we introduce a novel approach that overcomes this limitation by incorporating object-level contextual knowledge within images. Specifically, our model enhances intra-object consistency by distilling spectral-driven features from vision foundation models into the attention mechanism of the visual encoder, enabling semantically coherent components to form a single object mask. Additionally, we refine the text embeddings with zero-shot object presence likelihood to ensure accurate alignment with the specific objects represented in the images. By leveraging object-level contextual knowledge, our proposed approach achieves state-of-the-art performance with strong generalizability across diverse datasets.

Paperid: 4431, https://arxiv.org/pdf/2411.14797.pdf

Abstract:
Multimodal RLHF usually happens after supervised finetuning (SFT) stage to continually improve vision-language models' (VLMs) comprehension. Conventional wisdom holds its superiority over continual SFT during this preference alignment stage. In this paper, we observe that the inherent value of multimodal RLHF lies in its negative supervision, the logit of the rejected responses. We thus propose a novel negative supervised finetuning (nSFT) approach that fully excavates these information resided. Our nSFT disentangles this negative supervision in RLHF paradigm, and continually aligns VLMs with a simple SFT loss. This is more memory efficient than multimodal RLHF where 2 (e.g., DPO) or 4 (e.g., PPO) large VLMs are strictly required. The effectiveness of nSFT is rigorously proved by comparing it with various multimodal RLHF approaches, across different dataset sources, base VLMs and evaluation metrics. Besides, fruitful of ablations are provided to support our hypothesis. We hope this paper will stimulate further research to properly align large vision language models.

Paperid: 4432, https://arxiv.org/pdf/2411.07975.pdf

Abstract:
We present JanusFlow, a powerful framework that unifies image understanding and generation in a single model. JanusFlow introduces a minimalist architecture that integrates autoregressive language models with rectified flow, a state-of-the-art method in generative modeling. Our key finding demonstrates that rectified flow can be straightforwardly trained within the large language model framework, eliminating the need for complex architectural modifications. To further improve the performance of our unified model, we adopt two key strategies: (i) decoupling the understanding and generation encoders, and (ii) aligning their representations during unified training. Extensive experiments show that JanusFlow achieves comparable or superior performance to specialized models in their respective domains, while significantly outperforming existing unified approaches across standard benchmarks. This work represents a step toward more efficient and versatile vision-language models.

Paperid: 4433, https://arxiv.org/pdf/2410.24148.pdf

Abstract:
Technologies for recognizing facial attributes like race, gender, age, and emotion have several applications, such as surveillance, advertising content, sentiment analysis, and the study of demographic trends and social behaviors. Analyzing demographic characteristics based on images and analyzing facial expressions have several challenges due to the complexity of humans' facial attributes. Traditional approaches have employed CNNs and various other deep learning techniques, trained on extensive collections of labeled images. While these methods demonstrated effective performance, there remains potential for further enhancements. In this paper, we propose to utilize vision language models (VLMs) such as generative pre-trained transformer (GPT), GEMINI, large language and vision assistant (LLAVA), PaliGemma, and Microsoft Florence2 to recognize facial attributes such as race, gender, age, and emotion from images with human faces. Various datasets like FairFace, AffectNet, and UTKFace have been utilized to evaluate the solutions. The results show that VLMs are competitive if not superior to traditional techniques. Additionally, we propose "FaceScanPaliGemma"--a fine-tuned PaliGemma model--for race, gender, age, and emotion recognition. The results show an accuracy of 81.1%, 95.8%, 80%, and 59.4% for race, gender, age group, and emotion classification, respectively, outperforming pre-trained version of PaliGemma, other VLMs, and SotA methods. Finally, we propose "FaceScanGPT", which is a GPT-4o model to recognize the above attributes when several individuals are present in the image using a prompt engineered for a person with specific facial and/or physical attributes. The results underscore the superior multitasking capability of FaceScanGPT to detect the individual's attributes like hair cut, clothing color, postures, etc., using only a prompt to drive the detection and recognition tasks.

Paperid: 4434, https://arxiv.org/pdf/2410.23754.pdf

Abstract:
Decoding visual stimuli from neural recordings is a critical challenge in the development of brain-computer interfaces (BCIs). Although recent EEG-based decoding approaches have made progress in tasks such as visual classification, retrieval, and reconstruction, they remain constrained by unstable representation learning and a lack of interpretability. This gap highlights the need for more efficient representation learning and the integration of effective language interaction to enhance both understanding and practical usability in visual decoding tasks.To address this limitation, we introduce RealMind, a novel EEG-based framework designed to handle a diverse range of downstream tasks. Specifically, RealMind leverages both semantic and geometric consistency learning to enhance feature representation and improve alignment across tasks. Notably, beyond excelling in traditional tasks, our framework marks the first attempt at visual captioning from EEG data through vision-language model (VLM). It achieves a Top-1 decoding accuracy of 27.58% in a 200-class zero-shot retrieval task and a BLEU-1 score of 26.59% in a 200-class zero-shot captioning task. Overall, RealMind provides a comprehensive multitask EEG decoding framework, establishing a foundational approach for EEG-based visual decoding in real-world applications.

Paperid: 4435, https://arxiv.org/pdf/2410.21361.pdf

Abstract:
Domain adaptation has been extensively investigated in computer vision but still requires access to target data at the training time, which might be difficult to obtain in some uncommon conditions. In this paper, we present a new framework for domain adaptation relying on a single Vision-Language (VL) latent embedding instead of full target data. First, leveraging a contrastive language-image pre-training model (CLIP), we propose prompt/photo-driven instance normalization (PIN). PIN is a feature augmentation method that mines multiple visual styles using a single target VL latent embedding, by optimizing affine transformations of low-level source features. The VL embedding can come from a language prompt describing the target domain, a partially optimized language prompt, or a single unlabeled target image. Second, we show that these mined styles (i.e., augmentations) can be used for zero-shot (i.e., target-free) and one-shot unsupervised domain adaptation. Experiments on semantic segmentation demonstrate the effectiveness of the proposed method, which outperforms relevant baselines in the zero-shot and one-shot settings.

Paperid: 4436, https://arxiv.org/pdf/2410.18387.pdf

Abstract:
Several medical Multimodal Large Languange Models (MLLMs) have been developed to address tasks involving visual images with textual instructions across various medical modalities, achieving impressive results. Most current medical generalist models are region-agnostic, treating the entire image as a holistic representation. However, they struggle to identify which specific regions they are focusing on when generating a sentence. To mimic the behavior of doctors, who typically begin by reviewing the entire image before concentrating on specific regions for a thorough evaluation, we aim to enhance the capability of medical MLLMs in understanding anatomical regions within entire medical scans. To achieve it, we first formulate Region-Centric tasks and construct a large-scale dataset, MedRegInstruct, to incorporate regional information into training. Combining our collected dataset with other medical multimodal corpora for training, we propose a Region-Aware medical MLLM, MedRegA, which is the first bilingual generalist medical AI system to simultaneously handle image-level and region-level medical vision-language tasks across a broad range of modalities. Our MedRegA not only enables three region-centric tasks, but also achieves the best performance for visual question answering, report generation and medical image classification over 8 modalities, showcasing significant versatility. Experiments demonstrate that our model can not only accomplish powerful performance across various medical vision-language tasks in bilingual settings, but also recognize and detect structures in multimodal medical scans, boosting the interpretability and user interactivity of medical MLLMs. Our project page is https://medrega.github.io.

Paperid: 4437, https://arxiv.org/pdf/2410.13733.pdf

Abstract:
We focus on improving the visual understanding capability for boosting the vision-language models. We propose \textbf{Arcana}, a multiModal language model, which introduces two crucial techniques. First, we present Multimodal LoRA (MM-LoRA), a module designed to enhance the decoder. Unlike traditional language-driven decoders, MM-LoRA consists of two parallel LoRAs -- one for vision and one for language -- each with its own parameters. This disentangled parameters design allows for more specialized learning in each modality and better integration of multimodal information. Second, we introduce the Query Ladder adapter (QLadder) to improve the visual encoder. QLadder employs a learnable ``\textit{ladder}'' structure to deeply aggregates the intermediate representations from the frozen pretrained visual encoder (e.g., CLIP image encoder). This enables the model to learn new and informative visual features, as well as remaining the powerful capabilities of the pretrained visual encoder. These techniques collectively enhance Arcana's visual perception power, enabling it to leverage improved visual information for more accurate and contextually relevant outputs across various multimodal scenarios. Extensive experiments and ablation studies demonstrate the effectiveness and generalization capability of our Arcana. The code and re-annotated data are available at \url{https://arcana-project-page.github.io}.

Paperid: 4438, https://arxiv.org/pdf/2410.02615.pdf

Abstract:
State-of-the-art medical multi-modal LLMs (med-MLLMs), such as LLAVA-MED and BIOMEDGPT, primarily depend on scaling model size and data volume, with training driven largely by autoregressive objectives. However, we reveal that this approach can lead to weak vision-language alignment, making these models overly dependent on costly instruction-following data. To address this, we introduce EXGRA-MED, a novel multi-graph alignment framework that jointly aligns images, instruction responses, and extended captions in the latent space, advancing semantic grounding and cross-modal coherence. To scale to large LLMs (e.g., LLaMa-7B), we develop an efficient end-to-end training scheme using black-box gradient estimation, enabling fast and scalable optimization. Empirically, EXGRA-MED matches LLAVA-MED's performance using just 10% of pre-training data, achieving a 20.13% gain on VQA-RAD and approaching full-data performance. It also outperforms strong baselines like BIOMEDGPT and RADFM on visual chatbot and zero-shot classification tasks, demonstrating its promise for efficient, high-quality vision-language integration in medical AI.

Paperid: 4439, https://arxiv.org/pdf/2410.01345.pdf

Abstract:
Generalizing language-conditioned robotic policies to new tasks remains a significant challenge, hampered by the lack of suitable simulation benchmarks. In this paper, we address this gap by introducing GemBench, a novel benchmark to assess generalization capabilities of vision-language robotic manipulation policies. GemBench incorporates seven general action primitives and four levels of generalization, spanning novel placements, rigid and articulated objects, and complex long-horizon tasks. We evaluate state-of-the-art approaches on GemBench and also introduce a new method. Our approach 3D-LOTUS leverages rich 3D information for action prediction conditioned on language. While 3D-LOTUS excels in both efficiency and performance on seen tasks, it struggles with novel tasks. To address this, we present 3D-LOTUS++, a framework that integrates 3D-LOTUS's motion planning capabilities with the task planning capabilities of LLMs and the object grounding accuracy of VLMs. 3D-LOTUS++ achieves state-of-the-art performance on novel tasks of GemBench, setting a new standard for generalization in robotic manipulation. The benchmark, codes and trained models are available at https://www.di.ens.fr/willow/research/gembench/.

Paperid: 4440, https://arxiv.org/pdf/2409.17958.pdf

Abstract:
Several benchmarks have concluded that our best vision-language models (e.g., CLIP) are lacking in compositionality. Given an image, these benchmarks probe a model's ability to identify its associated caption amongst a set of compositional distractors. In response, a surge of recent proposals show improvements by finetuning CLIP with distractors as hard negatives. Our investigations reveal that these improvements have, in fact, been significantly overstated -- because existing benchmarks do not probe whether finetuned vision-language models remain invariant to hard positives. By curating an evaluation dataset with 112,382 hard negatives and hard positives, we uncover that including hard positives decreases CLIP's performance by 12.9%, while humans perform effortlessly at 99%. CLIP finetuned with hard negatives results in an even larger decrease, up to 38.7%. With this finding, we then produce a 1,775,259 image-text training set with both hard negative and hard positive captions. By training with both, we see improvements on existing benchmarks while simultaneously improving performance on hard positives, indicating a more robust improvement in compositionality. Our work suggests the need for future research to rigorously test and improve CLIP's understanding of semantic relationships between related "positive" concepts.

Paperid: 4441, https://arxiv.org/pdf/2409.09721.pdf

Abstract:
Vision-language models (VLMs) such as CLIP are trained via contrastive learning between text and image pairs, resulting in aligned image and text embeddings that are useful for many downstream tasks. A notable drawback of CLIP, however, is that the resulting embedding space seems to lack some of the structure of its purely text-based alternatives. For instance, while text embeddings have long been noted to satisfy analogies in embedding space using vector arithmetic, CLIP has no such property. In this paper, we propose an approach to natively train CLIP in a contrastive manner to reason about differences in embedding space. We finetune CLIP so that text descriptions of differences between images correspond to their difference in image embedding space, using synthetically generated data with large language models on image-caption paired datasets. We first demonstrate that our approach yields significantly improved capabilities in ranking images by a certain attribute (e.g., elephants are larger than cats), which is useful in retrieval or constructing attribute-based classifiers, and improved zeroshot classification performance on many downstream image classification tasks. In addition, our approach enables a new mechanism for inference that we refer to as comparative prompting, where we leverage prior knowledge of text descriptions of differences between classes of interest, achieving even larger performance gains in classification. Finally, we illustrate that the resulting embeddings obey a larger degree of geometric properties in embedding space, such as in text-to-image generation.

Paperid: 4442, https://arxiv.org/pdf/2409.05413.pdf

Abstract:
Robots are increasingly envisioned to interact in real-world scenarios, where they must continuously adapt to new situations. To detect and grasp novel objects, zero-shot pose estimators determine poses without prior knowledge. Recently, vision language models (VLMs) have shown considerable advances in robotics applications by establishing an understanding between language input and image input. In our work, we take advantage of VLMs zero-shot capabilities and translate this ability to 6D object pose estimation. We propose a novel framework for promptable zero-shot 6D object pose estimation using language embeddings. The idea is to derive a coarse location of an object based on the relevancy map of a language-embedded NeRF reconstruction and to compute the pose estimate with a point cloud registration method. Additionally, we provide an analysis of LERF's suitability for open-set object pose estimation. We examine hyperparameters, such as activation thresholds for relevancy maps and investigate the zero-shot capabilities on an instance- and category-level. Furthermore, we plan to conduct robotic grasping experiments in a real-world setting.

Paperid: 4443, https://arxiv.org/pdf/2409.03782.pdf

Abstract:
Refurbishing laptops extends their lives while contributing to reducing electronic waste, which promotes building a sustainable future. To this end, the Danish Technological Institute (DTI) focuses on the research and development of several robotic applications empowered with software, including laptop refurbishing. Cleaning represents a major step in refurbishing and involves identifying and removing stickers from laptop surfaces. Software plays a crucial role in the cleaning process. For instance, the software integrates various object detection models to identify and remove stickers from laptops automatically. However, given the diversity in types of stickers (e.g., shapes, colors, locations), identification of the stickers is highly uncertain, thereby requiring explicit quantification of uncertainty associated with the identified stickers. Such uncertainty quantification can help reduce risks in removing stickers, which, for example, could otherwise result in software faults damaging laptop surfaces. For uncertainty quantification, we adopted the Monte Carlo Dropout method to evaluate six sticker detection models (SDMs) from DTI using three datasets: the original image dataset from DTI and two datasets generated with vision language models, i.e., DALL-E-3 and Stable Diffusion-3. In addition, we presented novel robustness metrics concerning detection accuracy and uncertainty to assess the robustness of the SDMs based on adversarial datasets generated from the three datasets using a dense adversary method. Our evaluation results show that different SDMs perform differently regarding different metrics. Based on the results, we provide SDM selection guidelines and lessons learned from various perspectives.

Paperid: 4444, https://arxiv.org/pdf/2601.19582.pdf

Abstract:
In this paper, we introduce ScenePilot-Bench, a large-scale first-person driving benchmark designed to evaluate vision-language models (VLMs) in autonomous driving scenarios. ScenePilot-Bench is built upon ScenePilot-4K, a diverse dataset comprising 3,847 hours of driving videos, annotated with multi-granularity information including scene descriptions, risk assessments, key participant identification, ego trajectories, and camera parameters. The benchmark features a four-axis evaluation suite that assesses VLM capabilities in scene understanding, spatial perception, motion planning, and GPT-Score, with safety-aware metrics and cross-region generalization settings. We benchmark representative VLMs on ScenePilot-Bench, providing empirical analyses that clarify current performance boundaries and identify gaps for driving-oriented reasoning. ScenePilot-Bench offers a comprehensive framework for evaluating and advancing VLMs in safety-critical autonomous driving contexts.

Paperid: 4445, https://arxiv.org/pdf/2601.19202.pdf

Abstract:
Vision-Language Models (VLMs) have shown strong multimodal reasoning capabilities on Visual-Question-Answering (VQA) benchmarks. However, their robustness against textual misinformation remains under-explored. While existing research has studied the effect of misinformation in text-only domains, it is not clear how VLMs arbitrate between contradictory information from different modalities. To bridge the gap, we first propose the CONTEXT-VQA (i.e., Conflicting Text) dataset, consisting of image-question pairs together with systematically generated persuasive prompts that deliberately conflict with visual evidence. Then, a thorough evaluation framework is designed and executed to benchmark the susceptibility of various models to these conflicting multimodal inputs. Comprehensive experiments over 11 state-of-the-art VLMs reveal that these models are indeed vulnerable to misleading textual prompts, often overriding clear visual evidence in favor of the conflicting text, and show an average performance drop of over 48.2% after only one round of persuasive conversation. Our findings highlight a critical limitation in current VLMs and underscore the need for improved robustness against textual manipulation.

Paperid: 4446, https://arxiv.org/pdf/2601.15406.pdf

Abstract:
Multimodal Large Language Models (MLLMs) have recently demonstrated strong performance on a wide range of vision-language tasks, raising interest in their potential use for biometric applications. In this paper, we conduct a systematic evaluation of state-of-the-art MLLMs for heterogeneous face recognition (HFR), where enrollment and probe images are from different sensing modalities, including visual (VIS), near infrared (NIR), short-wave infrared (SWIR), and thermal camera. We benchmark multiple open-source MLLMs across several cross-modality scenarios, including VIS-NIR, VIS-SWIR, and VIS-THERMAL face recognition. The recognition performance of MLLMs is evaluated using biometric protocols and based on different metrics, including Acquire Rate, Equal Error Rate (EER), and True Accept Rate (TAR). Our results reveal substantial performance gaps between MLLMs and classical face recognition systems, particularly under challenging cross-spectral conditions, in spite of recent advances in MLLMs. Our findings highlight the limitations of current MLLMs for HFR and also the importance of rigorous biometric evaluation when considering their deployment in face recognition systems.

Paperid: 4447, https://arxiv.org/pdf/2601.15197.pdf

Abstract:
Vision-Language-Action (VLA) models have shown promise in robot manipulation but often struggle to generalize to new instructions or complex multi-task scenarios. We identify a critical pathology in current training paradigms where goal-driven data collection creates a dataset bias. In such datasets, language instructions are highly predictable from visual observations alone, causing the conditional mutual information between instructions and actions to vanish, a phenomenon we term Information Collapse. Consequently, models degenerate into vision-only policies that ignore language constraints and fail in out-of-distribution (OOD) settings. To address this, we propose LangForce, a novel framework that enforces instruction following via Bayesian decomposition. By introducing learnable Latent Action Queries, we construct a dual-branch architecture to estimate both a vision-only prior $p(a \mid v)$ and a language-conditioned posterior $π(a \mid v, \ell)$. We then optimize the policy to maximize the conditional Pointwise Mutual Information (PMI) between actions and instructions. This objective effectively penalizes the vision shortcut and rewards actions that explicitly explain the language command. Without requiring new data, LangForce significantly improves generalization. Extensive experiments across on SimplerEnv and RoboCasa demonstrate substantial gains, including an 11.3% improvement on the challenging OOD SimplerEnv benchmark, validating the ability of our approach to robustly ground language in action.

Paperid: 4448, https://arxiv.org/pdf/2601.12742.pdf

Abstract:
Recent advances in large Vision-Language Models (VLMs) have provided rich semantic understanding that empowers drones to search for open-set objects via natural language instructions. However, prior systems struggle to integrate VLMs into practical aerial systems due to orders-of-magnitude frequency mismatch between VLM inference and real-time planning, as well as VLMs' limited 3D scene understanding. They also lack a unified mechanism to balance semantic guidance with motion efficiency in large-scale environments. To address these challenges, we present AirHunt, an aerial object navigation system that efficiently locates open-set objects with zero-shot generalization in outdoor environments by seamlessly fusing VLM semantic reasoning with continuous path planning. AirHunt features a dual-pathway asynchronous architecture that establishes a synergistic interface between VLM reasoning and path planning, enabling continuous flight with adaptive semantic guidance that evolves through motion. Moreover, we propose an active dual-task reasoning module that exploits geometric and semantic redundancy to enable selective VLM querying, and a semantic-geometric coherent planning module that dynamically reconciles semantic priorities and motion efficiency in a unified framework, enabling seamless adaptation to environmental heterogeneity. We evaluate AirHunt across diverse object navigation tasks and environments, demonstrating a higher success rate with lower navigation error and reduced flight time compared to state-of-the-art methods. Real-world experiments further validate AirHunt's practical capability in complex and challenging environments. Code and dataset will be made publicly available before publication.

Paperid: 4449, https://arxiv.org/pdf/2601.12672.pdf

Abstract:
The safe deployment of autonomous driving (AD) systems is fundamentally hindered by the long-tail problem, where rare yet critical driving scenarios are severely underrepresented in real-world data. Existing solutions including safety-critical scenario generation and closed-loop learning often rely on rule-based heuristics, resampling methods and generative models learned from offline datasets, limiting their ability to produce diverse and novel challenges. While recent works leverage Vision Language Models (VLMs) to produce scene descriptions that guide a separate, downstream model in generating hazardous trajectories for agents, such two-stage framework constrains the generative potential of VLMs, as the diversity of the final trajectories is ultimately limited by the generalization ceiling of the downstream algorithm. To overcome these limitations, we introduce VILTA (VLM-In-the-Loop Trajectory Adversary), a novel framework that integrates a VLM into the closed-loop training of AD agents. Unlike prior works, VILTA actively participates in the training loop by comprehending the dynamic driving environment and strategically generating challenging scenarios through direct, fine-grained editing of surrounding agents' future trajectories. This direct-editing approach fully leverages the VLM's powerful generalization capabilities to create a diverse curriculum of plausible yet challenging scenarios that extend beyond the scope of traditional methods. We demonstrate that our approach substantially enhances the safety and robustness of the resulting AD policy, particularly in its ability to navigate critical long-tail events.

Paperid: 4450, https://arxiv.org/pdf/2601.12430.pdf

Abstract:
Vision-language model (VLM) hallucination is commonly linked to imbalanced allocation of attention across input modalities: system, image and text. However, existing mitigation strategies tend towards an image-centric interpretation of these imbalances, often prioritising increased image attention while giving less consideration to the roles of the other modalities. In this study, we evaluate a more holistic, system-mediated account, which attributes these imbalances to functionally redundant system weights that reduce attention to image and textual inputs. We show that this framework offers a useful empirical perspective on the yes-bias, a common form of hallucination in which VLMs indiscriminately respond 'yes'. Causally redistributing attention from the system modality to image and textual inputs substantially suppresses this bias, often outperforming existing approaches. We further present evidence suggesting that system-mediated attention imbalances contribute to the yes-bias by encouraging a default reliance on coarse input representations, which are effective for some tasks but ill-suited to others. Taken together, these findings firmly establish system attention as a key factor in VLM hallucination and highlight its potential as a lever for mitigation.

Paperid: 4451, https://arxiv.org/pdf/2601.08464.pdf

Abstract:
The conceptual design phase in architecture and urban planning, particularly building massing, is complex and heavily reliant on designer intuition and manual effort. To address this, we propose an automated framework for generating building massing based on functional requirements and site context. A primary obstacle to such data-driven methods has been the lack of suitable datasets. Consequently, we introduce the CoMa-20K dataset, a comprehensive collection that includes detailed massing geometries, associated economical and programmatic data, and visual representations of the development site within its existing urban context. We benchmark this dataset by formulating massing generation as a conditional task for Vision-Language Models (VLMs), evaluating both fine-tuned and large zero-shot models. Our experiments reveal the inherent complexity of the task while demonstrating the potential of VLMs to produce context-sensitive massing options. The dataset and analysis establish a foundational benchmark and highlight significant opportunities for future research in data-driven architectural design.

Paperid: 4452, https://arxiv.org/pdf/2601.07516.pdf

Abstract:
Vision-language models are increasingly employed as multimodal conversational agents (MCAs) for diverse conversational tasks. Recently, reinforcement learning (RL) has been widely explored for adapting MCAs to various human-AI interaction scenarios. Despite showing great enhancement in generalization performance, fine-tuning MCAs via RL still faces challenges in handling the extremely large text token space. To address this, we learn a compact latent action space for RL fine-tuning instead. Specifically, we adopt the learning from observation mechanism to construct the codebook for the latent action space, where future observations are leveraged to estimate current latent actions that could further be used to reconstruct future observations. However, the scarcity of paired image-text data hinders learning a codebook with sufficient coverage. Thus, we leverage both paired image-text data and text-only data to construct the latent action space, using a cross-modal projector for transforming text embeddings into image-text embeddings. We initialize the cross-modal projector on paired image-text data, and further train it on massive text-only data with a novel cycle consistency loss to enhance its robustness. We show that our latent action based method outperforms competitive baselines on two conversation tasks across various RL algorithms.

Paperid: 4453, https://arxiv.org/pdf/2601.02783.pdf

Abstract:
Earth vision has achieved milestones in geospatial object recognition but lacks exploration in object-relational reasoning, limiting comprehensive scene understanding. To address this, a progressive Earth vision-language understanding and generation framework is proposed, including a multi-task dataset (EarthVLSet) and a semantic-guided network (EarthVLNet). Focusing on city planning applications, EarthVLSet includes 10.9k sub-meter resolution remote sensing images, land-cover masks, and 761.5k textual pairs involving both multiple-choice and open-ended visual question answering (VQA) tasks. In an object-centric way, EarthVLNet is proposed to progressively achieve semantic segmentation, relational reasoning, and comprehensive understanding. The first stage involves land-cover segmentation to generate object semantics for VQA guidance. Guided by pixel-wise semantics, the object awareness based large language model (LLM) performs relational reasoning and knowledge summarization to generate the required answers. As for optimization, the numerical difference loss is proposed to dynamically add difference penalties, addressing the various objects' statistics. Three benchmarks, including semantic segmentation, multiple-choice, and open-ended VQA demonstrated the superiorities of EarthVLNet, yielding three future directions: 1) segmentation features consistently enhance VQA performance even in cross-dataset scenarios; 2) multiple-choice tasks show greater sensitivity to the vision encoder than to the language decoder; and 3) open-ended tasks necessitate advanced vision encoders and language decoders for an optimal performance. We believe this dataset and method will provide a beneficial benchmark that connects ''image-mask-text'', advancing geographical applications for Earth vision.

Paperid: 4454, https://arxiv.org/pdf/2601.01984.pdf

Abstract:
Spatial reasoning -- the ability to perceive and reason about relationships in space -- advances vision-language models (VLMs) from visual perception toward spatial semantic understanding. Existing approaches either revisit local image patches, improving fine-grained perception but weakening global spatial awareness, or mark isolated coordinates, which capture object locations but overlook their overall organization. In this work, we integrate the cognitive concept of an object-centric blueprint into VLMs to enhance spatial reasoning. Given an image and a question, the model first constructs a JSON-style blueprint that records the positions, sizes, and attributes of relevant objects, and then reasons over this structured representation to produce the final answer. To achieve this, we introduce three key techniques: (1) blueprint-embedded reasoning traces for supervised fine-tuning to elicit basic reasoning skills; (2) blueprint-aware rewards in reinforcement learning to encourage the blueprint to include an appropriate number of objects and to align final answers with this causal reasoning; and (3) anti-shortcut data augmentation that applies targeted perturbations to images and questions, discouraging reliance on superficial visual or linguistic cues. Experiments show that our method consistently outperforms existing VLMs and specialized spatial reasoning models.

Paperid: 4455, https://arxiv.org/pdf/2512.23077.pdf

Abstract:
Discovering effective reward functions remains a fundamental challenge in motor control of high-dimensional musculoskeletal systems. While humans can describe movement goals explicitly such as "walking forward with an upright posture," the underlying control strategies that realize these goals are largely implicit, making it difficult to directly design rewards from high-level goals and natural language descriptions. We introduce Motion from Vision-Language Representation (MoVLR), a framework that leverages vision-language models (VLMs) to bridge the gap between goal specification and movement control. Rather than relying on handcrafted rewards, MoVLR iteratively explores the reward space through iterative interaction between control optimization and VLM feedback, aligning control policies with physically coordinated behaviors. Our approach transforms language and vision-based assessments into structured guidance for embodied learning, enabling the discovery and refinement of reward functions for high-dimensional musculoskeletal locomotion and manipulation. These results suggest that VLMs can effectively ground abstract motion descriptions in the implicit principles governing physiological motor control.

Paperid: 4456, https://arxiv.org/pdf/2512.20876.pdf

Abstract:
From the perspective of future developments in robotics, it is crucial to verify whether foundation models trained exclusively on offline data, such as images and language, can understand the robot motion. In particular, since Vision Language Models (VLMs) do not include low-level motion information from robots in their training datasets, video understanding including trajectory information remains a significant challenge. In this study, we assess two capabilities of VLMs through a video captioning task with low-level robot motion information: (1) automatic captioning of robot tasks and (2) segmentation of a series of tasks. Both capabilities are expected to enhance the efficiency of robot imitation learning by linking language and motion and serve as a measure of the foundation model's performance. The proposed method generates multiple "scene" captions using image captions and trajectory data from robot tasks. The full task caption is then generated by summarizing these individual captions. Additionally, the method performs subtask segmentation by comparing the similarity between text embeddings of image captions. In both captioning tasks, the proposed method aims to improve performance by providing the robot's motion data - joint and end-effector states - as input to the VLM. Simulator experiments were conducted to validate the effectiveness of the proposed method.

Paperid: 4457, https://arxiv.org/pdf/2512.19243.pdf

Abstract:
Generative models can now produce photorealistic imagery, yet they still struggle with the long, multi-goal prompts that professional designers issue. To expose this gap and better evaluate models' performance in real-world settings, we introduce Long Goal Bench (LGBench), a 2,000-task suite (1,000 T2I and 1,000 I2I) whose average instruction contains 18 to 22 tightly coupled goals spanning global layout, local object placement, typography, and logo fidelity. We find that even state-of-the-art models satisfy fewer than 72 percent of the goals and routinely miss localized edits, confirming the brittleness of current pipelines. To address this, we present VisionDirector, a training-free vision-language supervisor that (i) extracts structured goals from long instructions, (ii) dynamically decides between one-shot generation and staged edits, (iii) runs micro-grid sampling with semantic verification and rollback after every edit, and (iv) logs goal-level rewards. We further fine-tune the planner with Group Relative Policy Optimization, yielding shorter edit trajectories (3.1 versus 4.2 steps) and stronger alignment. VisionDirector achieves new state of the art on GenEval (plus 7 percent overall) and ImgEdit (plus 0.07 absolute) while producing consistent qualitative improvements on typography, multi-object scenes, and pose editing.

Paperid: 4458, https://arxiv.org/pdf/2512.16793.pdf

Abstract:
Robotic generalization relies on physical intelligence: the ability to reason about state changes, contact-rich interactions, and long-horizon planning under egocentric perception and action. However, most VLMs are trained primarily on third-person data, creating a fundamental viewpoint mismatch for humanoid robots. Scaling robot egocentric data collection remains impractical due to high cost and limited diversity, whereas large-scale human egocentric videos offer a scalable alternative that naturally capture rich interaction context and causal structure. The key challenge is to convert raw egocentric videos into structured and reliable embodiment training supervision. Accordingly, we propose an Egocentric2Embodiment translation pipeline that transforms first-person videos into multi-level, schema-driven VQA supervision with enforced evidence grounding and temporal consistency, enabling the construction of the Egocentric2Embodiment dataset (E2E-3M) at scale. An egocentric-aware embodied brain, termed PhysBrain, is obtained by training on the E2E-3M dataset. PhysBrain exhibits substantially improved egocentric understanding, particularly for planning on EgoThink. It provides an egocentric-aware initialization that enables more sample-efficient VLA fine-tuning and higher SimplerEnv success rates (53.9\%), demonstrating effective transfer from human egocentric supervision to downstream robot control.

Paperid: 4459, https://arxiv.org/pdf/2512.07344.pdf

Abstract:
Vision-language models (VLMs) have demonstrated impressive multimodal comprehension capabilities and are being deployed in an increasing number of online video understanding applications. While recent efforts extensively explore advancing VLMs' reasoning power in these cases, deployment constraints are overlooked, leading to overwhelming system overhead in real-world deployments. To address that, we propose Venus, an on-device memory-and-retrieval system for efficient online video understanding. Venus proposes an edge-cloud disaggregated architecture that sinks memory construction and keyframe retrieval from cloud to edge, operating in two stages. In the ingestion stage, Venus continuously processes streaming edge videos via scene segmentation and clustering, where the selected keyframes are embedded with a multimodal embedding model to build a hierarchical memory for efficient storage and retrieval. In the querying stage, Venus indexes incoming queries from memory, and employs a threshold-based progressive sampling algorithm for keyframe selection that enhances diversity and adaptively balances system cost and reasoning accuracy. Our extensive evaluation shows that Venus achieves a 15x-131x speedup in total response latency compared to state-of-the-art methods, enabling real-time responses within seconds while maintaining comparable or even superior reasoning accuracy.

Paperid: 4460, https://arxiv.org/pdf/2512.06963.pdf

Abstract:
Generalization in robot manipulation is essential for deploying robots in open-world environments and advancing toward artificial general intelligence. While recent Vision-Language-Action (VLA) models leverage large pre-trained understanding models for perception and instruction following, their ability to generalize to novel tasks, objects, and settings remains limited. In this work, we present VideoVLA, a simple approach that explores the potential of transforming large video generation models into robotic VLA manipulators. Given a language instruction and an image, VideoVLA predicts an action sequence as well as the future visual outcomes. Built on a multi-modal Diffusion Transformer, VideoVLA jointly models video, language, and action modalities, using pre-trained video generative models for joint visual and action forecasting. Our experiments show that high-quality imagined futures correlate with reliable action predictions and task success, highlighting the importance of visual imagination in manipulation. VideoVLA demonstrates strong generalization, including imitating other embodiments' skills and handling novel objects. This dual-prediction strategy - forecasting both actions and their visual consequences - explores a paradigm shift in robot learning and unlocks generalization capabilities in manipulation systems.

Paperid: 4461, https://arxiv.org/pdf/2512.01801.pdf

Abstract:
We present GR-RL, a robotic learning framework that turns a generalist vision-language-action (VLA) policy into a highly capable specialist for long-horizon dexterous manipulation. Assuming the optimality of human demonstrations is core to existing VLA policies. However, we claim that in highly dexterous and precise manipulation tasks, human demonstrations are noisy and suboptimal. GR-RL proposes a multi-stage training pipeline that filters, augments, and reinforces the demonstrations by reinforcement learning. First, GR-RL learns a vision-language-conditioned task progress, filters the demonstration trajectories, and only keeps the transitions that contribute positively to the progress. Specifically, we show that by directly applying offline RL with sparse reward, the resulting $Q$-values can be treated as a robust progress function. Next, we introduce morphological symmetry augmentation that greatly improves the generalization and performance of GR-RL. Lastly, to better align the VLA policy with its deployment behaviors for high-precision control, we perform online RL by learning a latent space noise predictor. With this pipeline, GR-RL is, to our knowledge, the first learning-based policy that can autonomously lace up a shoe by threading shoelaces through multiple eyelets with an 83.3% success rate, a task requiring long-horizon reasoning, millimeter-level precision, and compliant soft-body interaction. We hope GR-RL provides a step toward enabling generalist robot foundations models to specialize into reliable real-world experts.

Paperid: 4462, https://arxiv.org/pdf/2511.17053.pdf

Abstract:
LVLMs have been shown to perform excellently in image-level tasks such as VQA and caption. However, in many instance-level tasks, such as visual grounding and object detection, LVLMs still show performance gaps compared to previous expert models. Meanwhile, although pedestrian tracking is a classical task, there have been a number of new topics in combining object tracking and natural language, such as Referring MOT, Cross-view Referring MOT, and Semantic MOT. These tasks emphasize that models should understand the tracked object at an advanced semantic level, which is exactly where LVLMs excel. In this paper, we propose a new unified Pedestrian Tracking framework, namely OmniPT, which can track, track based on reference and generate semantic understanding of tracked objects interactively. We address two issues: how to model the tracking task into a task that foundation models can perform, and how to make the model output formatted answers. To this end, we implement a training phase consisting of RL-Mid Training-SFT-RL. Based on the pre-trained weights of the LVLM, we first perform a simple RL phase to enable the model to output fixed and supervisable bounding box format. Subsequently, we conduct a mid-training phase using a large number of pedestrian-related datasets. Finally, we perform supervised fine-tuning on several pedestrian tracking datasets, and then carry out another RL phase to improve the model's tracking performance and enhance its ability to follow instructions. We conduct experiments on tracking benchmarks and the experimental results demonstrate that the proposed method can perform better than the previous methods.

Paperid: 4463, https://arxiv.org/pdf/2511.15411.pdf

Abstract:
Data-Free Quantization (DFQ) offers a practical solution for model compression without requiring access to real data, making it particularly attractive in privacy-sensitive scenarios. While DFQ has shown promise for unimodal models, its extension to Vision-Language Models such as Contrastive Language-Image Pre-training (CLIP) models remains underexplored. In this work, we reveal that directly applying existing DFQ techniques to CLIP results in substantial performance degradation due to two key limitations: insufficient semantic content and low intra-image diversity in synthesized samples. To tackle these challenges, we propose D4C, the first DFQ framework tailored for CLIP. D4C synthesizes semantically rich and structurally diverse pseudo images through three key components: (1) Prompt-Guided Semantic Injection aligns generated images with real-world semantics using text prompts; (2) Structural Contrastive Generation reproduces compositional structures of natural images by leveraging foreground-background contrastive synthesis; and (3) Perturbation-Aware Enhancement applies controlled perturbations to improve sample diversity and robustness. These components jointly empower D4C to synthesize images that are both semantically informative and structurally diverse, effectively bridging the performance gap of DFQ on CLIP. Extensive experiments validate the effectiveness of D4C, showing significant performance improvements on various bit-widths and models. For example, under the W4A8 setting with CLIP ResNet-50 and ViT-B/32, D4C achieves Top-1 accuracy improvement of 12.4% and 18.9% on CIFAR-10, 6.8% and 19.7% on CIFAR-100, and 1.4% and 5.7% on ImageNet-1K in zero-shot classification, respectively.

Paperid: 4464, https://arxiv.org/pdf/2511.11440.pdf

Abstract:
Fine-tuning Vision-Language Models (VLMs) is a common strategy to improve performance following an ad-hoc data collection and annotation of real-world scenes. However, this process is often prone to biases, errors, and distribution imbalance, resulting in overfitting and imbalanced performance. Although a few studies have tried to address this problem by generating synthetic data, they lacked control over distribution bias and annotation quality. To address these challenges, we redesign the fine-tuning process in two ways. First, we control the generation of data and its annotations, ensuring it is free from bias, distribution imbalance, and annotation errors. We automatically construct the dataset by comprehensively sampling objects' attributes, including color, shape, size, and position within the scene. Secondly, using this annotated dataset, we fine-tune state-of-the-art VLMs and assess performance transferability to real-world data on the absolute position task. We conduct exhaustive evaluations on both synthetic and real-world benchmarks. Our experiments reveal two key findings: 1) fine-tuning on balanced synthetic data yields uniform performance across the visual scene and mitigates common biases; and 2) fine-tuning on synthetic stimuli significantly improves performance on real-world data (COCO), outperforming models fine-tuned in the matched setting.

Paperid: 4465, https://arxiv.org/pdf/2511.10974.pdf

Abstract:
Class-incremental learning (CIL) enables models to continuously learn new categories from sequential tasks without forgetting previously acquired knowledge. While recent advances in vision-language models such as CLIP have demonstrated strong generalization across domains, extending them to continual settings remains challenging. In particular, learning task-specific soft prompts for newly introduced classes often leads to severe classifier bias, as the text prototypes overfit to recent categories when prior data are unavailable. In this paper, we propose DMC, a simple yet effective two-stage framework for CLIP-based CIL that decouples the adaptation of the vision encoder and the optimization of textual soft prompts. Each stage is trained with the other frozen, allowing one modality to act as a stable semantic anchor for the other to preserve cross-modal alignment. Furthermore, current CLIP-based CIL approaches typically store class-wise Gaussian statistics for generative replay, yet they overlook the distributional drift that arises when the vision encoder is updated over time. To address this issue, we introduce DMC-OT, an enhanced version of DMC that incorporates an optimal-transport guided calibration strategy to align memory statistics across evolving encoders, along with a task-specific prompting design that enhances inter-task separability. Extensive experiments on CIFAR-100, Imagenet-R, CUB-200, and UCF-101 demonstrate that both DMC and DMC-OT achieve state-of-the-art performance, with DMC-OT further improving accuracy by an average of 1.80%.

Paperid: 4466, https://arxiv.org/pdf/2511.10008.pdf

Abstract:
Vision-Language-Action (VLA) models revolutionize robotic systems by enabling end-to-end perception-to-action pipelines that integrate multiple sensory modalities, such as visual signals processed by cameras and auditory signals captured by microphones. This multi-modality integration allows VLA models to interpret complex, real-world environments using diverse sensor data streams. Given the fact that VLA-based systems heavily rely on the sensory input, the security of VLA models against physical-world sensor attacks remains critically underexplored. To address this gap, we present the first systematic study of physical sensor attacks against VLAs, quantifying the influence of sensor attacks and investigating the defenses for VLA models. We introduce a novel ``Real-Sim-Real'' framework that automatically simulates physics-based sensor attack vectors, including six attacks targeting cameras and two targeting microphones, and validates them on real robotic systems. Through large-scale evaluations across various VLA architectures and tasks under varying attack parameters, we demonstrate significant vulnerabilities, with susceptibility patterns that reveal critical dependencies on task types and model designs. We further develop an adversarial-training-based defense that enhances VLA robustness against out-of-distribution physical perturbations caused by sensor attacks while preserving model performance. Our findings expose an urgent need for standardized robustness benchmarks and mitigation strategies to secure VLA deployments in safety-critical environments.

Paperid: 4467, https://arxiv.org/pdf/2511.09515.pdf

Abstract:
Vision-Language-Action (VLA) models have shown strong potential for general-purpose robotic manipulation, but their reliance on expert demonstrations limits their ability to learn from failures and perform self-corrections. Reinforcement learning (RL) addresses these through self-improving interactions with the physical environment, but suffers from high sample complexity on real robots. We introduce World-Model-based Policy Optimization (WMPO), a principled framework for on-policy VLA RL without interacting with the real environment. In contrast to widely used latent world models, WMPO focuses on pixel-based predictions that align the "imagined" trajectories with the VLA features pretrained with web-scale images. Crucially, WMPO enables the policy to perform on-policy GRPO that provides stronger performance than the often-used off-policy methods. Extensive experiments in both simulation and real-robot settings demonstrate that WMPO (i) substantially improves sample efficiency, (ii) achieves stronger overall performance, (iii) exhibits emergent behaviors such as self-correction, and (iv) demonstrates robust generalization and lifelong learning capabilities.

Paperid: 4468, https://arxiv.org/pdf/2510.24331.pdf

Abstract:
In-context learning (ICL) enables Large Language Models (LLMs) to learn tasks from demonstration examples without parameter updates. Although it has been extensively studied in LLMs, its effectiveness in Vision-Language Models (VLMs) remains underexplored. In this work, we present a systematic study of ICL in VLMs, evaluating seven models spanning four architectures on three image captioning benchmarks. We analyze how prompt design, architectural choices, and training strategies influence multimodal ICL. To our knowledge, we are the first to analyze how attention patterns in VLMs vary with an increasing number of in-context demonstrations. Our results reveal that training on imag-text interleaved data enhances ICL performance but does not imply effective integration of visual and textual information from demonstration examples. In contrast, instruction tuning improves instruction-following but can reduce reliance on in-context demonstrations, suggesting a trade-off between instruction alignment and in-context adaptation. Attention analyses further show that current VLMs primarily focus on textual cues and fail to leverage visual information, suggesting a limited capacity for multimodal integration. These findings highlight key limitations in the ICL abilities of current VLMs and provide insights for enhancing their ability to learn from multimodal in-context examples.

Paperid: 4469, https://arxiv.org/pdf/2510.20965.pdf

Abstract:
Robotic suturing is a prototypical long-horizon dexterous manipulation task, requiring coordinated needle grasping, precise tissue penetration, and secure knot tying. Despite numerous efforts toward end-to-end autonomy, a fully autonomous suturing pipeline has yet to be demonstrated on physical hardware. We introduce SutureBot: an autonomous suturing benchmark on the da Vinci Research Kit (dVRK), spanning needle pickup, tissue insertion, and knot tying. To ensure repeatability, we release a high-fidelity dataset comprising 1,890 suturing demonstrations. Furthermore, we propose a goal-conditioned framework that explicitly optimizes insertion-point precision, improving targeting accuracy by 59\%-74\% over a task-only baseline. To establish this task as a benchmark for dexterous imitation learning, we evaluate state-of-the-art vision-language-action (VLA) models, including $π_0$, GR00T N1, OpenVLA-OFT, and multitask ACT, each augmented with a high-level task-prediction policy. Autonomous suturing is a key milestone toward achieving robotic autonomy in surgery. These contributions support reproducible evaluation and development of precision-focused, long-horizon dexterous manipulation policies necessary for end-to-end suturing. Dataset is available at: https://huggingface.co/datasets/jchen396/suturebot

Paperid: 4470, https://arxiv.org/pdf/2510.20223.pdf

Abstract:
Multimodal large language models (MLLMs) have achieved remarkable progress, yet remain critically vulnerable to adversarial attacks that exploit weaknesses in cross-modal processing. We present a systematic study of multimodal jailbreaks targeting both vision-language and audio-language models, showing that even simple perceptual transformations can reliably bypass state-of-the-art safety filters. Our evaluation spans 1,900 adversarial prompts across three high-risk safety categories harmful content, CBRN (Chemical, Biological, Radiological, Nuclear), and CSEM (Child Sexual Exploitation Material) tested against seven frontier models. We explore the effectiveness of attack techniques on MLLMs, including FigStep-Pro (visual keyword decomposition), Intelligent Masking (semantic obfuscation), and audio perturbations (Wave-Echo, Wave-Pitch, Wave-Speed). The results reveal severe vulnerabilities: models with almost perfect text-only safety (0\% ASR) suffer >75\% attack success under perceptually modified inputs, with FigStep-Pro achieving up to 89\% ASR in Llama-4 variants. Audio-based attacks further uncover provider-specific weaknesses, with even basic modality transfer yielding 25\% ASR for technical queries. These findings expose a critical gap between text-centric alignment and multimodal threats, demonstrating that current safeguards fail to generalize across cross-modal attacks. The accessibility of these attacks, which require minimal technical expertise, suggests that robust multimodal AI safety will require a paradigm shift toward broader semantic-level reasoning to mitigate possible risks.

Paperid: 4471, https://arxiv.org/pdf/2510.19555.pdf

Abstract:
Vision-Language Models (VLMs) have recently gained attention due to their competitive performance on multiple downstream tasks, achieved by following user-input instructions. However, VLMs still exhibit several limitations in visual reasoning, such as difficulties in identifying relations (e.g., spatial, temporal, and among objects), understanding temporal sequences (e.g., frames), and counting objects. In this work, we go beyond score-level benchmark evaluations of VLMs by investigating the underlying causes of their failures and proposing a targeted approach to improve their reasoning capabilities. We study the reasoning skills of seven state-of-the-art VLMs in the counting task under controlled experimental conditions. Our experiments show that VLMs are highly sensitive to the number and type of objects, their spatial arrangement, and the co-occurrence of distractors. A layer-wise analysis reveals that errors are due to incorrect mapping of the last-layer representation into the output space. Our targeted training shows that fine-tuning just the output layer improves accuracy by up to 21%. We corroborate these findings by achieving consistent improvements on real-world datasets.

Paperid: 4472, https://arxiv.org/pdf/2510.19438.pdf

Abstract:
Autonomous Driving Systems (ADS) are safety-critical, where failures can be severe. While Metamorphic Testing (MT) is effective for fault detection in ADS, existing methods rely heavily on manual effort and lack automation. We present AutoMT, a multi-agent MT framework powered by Large Language Models (LLMs) that automates the extraction of Metamorphic Relations (MRs) from local traffic rules and the generation of valid follow-up test cases. AutoMT leverages LLMs to extract MRs from traffic rules in Gherkin syntax using a predefined ontology. A vision-language agent analyzes scenarios, and a search agent retrieves suitable MRs from a RAG-based database to generate follow-up cases via computer vision. Experiments show that AutoMT achieves up to 5 x higher test diversity in follow-up case generation compared to the best baseline (manual expert-defined MRs) in terms of validation rate, and detects up to 20.55% more behavioral violations. While manual MT relies on a fixed set of predefined rules, AutoMT automatically extracts diverse metamorphic relations that augment real-world datasets and help uncover corner cases often missed during in-field testing and data collection. Its modular architecture separating MR extraction, filtering, and test generation supports integration into industrial pipelines and potentially enables simulation-based testing to systematically cover underrepresented or safety-critical scenarios.

Paperid: 4473, https://arxiv.org/pdf/2510.09473.pdf

Abstract:
Test-time adaptation paradigm provides flexibility towards domain shifts by performing immediate adaptation on unlabeled target data from the source model. Vision-Language Models (VLMs) leverage their generalization capabilities for diverse downstream tasks, and test-time prompt tuning has emerged as a prominent solution for adapting VLMs. In this work, we explore contrastive VLMs and identify the modality gap caused by a single dominant feature dimension across modalities. We observe that the dominant dimensions in both text and image modalities exhibit high predictive sensitivity, and that constraining their influence can improve calibration error. Building on this insight, we propose dimensional entropy maximization that regularizes the distribution of textual features toward uniformity to mitigate the dependency of dominant dimensions. Our method alleviates the degradation of calibration performance in test-time prompt tuning, offering a simple yet effective solution to enhance the reliability of VLMs in real-world deployment scenarios.

Paperid: 4474, https://arxiv.org/pdf/2510.09253.pdf

Abstract:
While specialized learning-based models have historically dominated image privacy prediction, the current literature increasingly favours adopting large Vision-Language Models (VLMs) designed for generic tasks. This trend risks overlooking the performance ceiling set by purpose-built models due to a lack of systematic evaluation. To address this problem, we establish a zero-shot benchmark for image privacy classification, enabling a fair comparison. We evaluate the top-3 open-source VLMs, according to a privacy benchmark, using task-aligned prompts and we contrast their performance, efficiency, and robustness against established vision-only and multi-modal methods. Counter-intuitively, our results show that VLMs, despite their resource-intensive nature in terms of high parameter count and slower inference, currently lag behind specialized, smaller models in privacy prediction accuracy. We also find that VLMs exhibit higher robustness to image perturbations.

Paperid: 4475, https://arxiv.org/pdf/2510.03160.pdf

Abstract:
Spine disorders affect 619 million people globally and are a leading cause of disability, yet AI-assisted diagnosis remains limited by the lack of level-aware, multimodal datasets. Clinical decision-making for spine disorders requires sophisticated reasoning across X-ray, CT, and MRI at specific vertebral levels. However, progress has been constrained by the absence of traceable, clinically-grounded instruction data and standardized, spine-specific benchmarks. To address this, we introduce SpineMed, an ecosystem co-designed with practicing spine surgeons. It features SpineMed-450k, the first large-scale dataset explicitly designed for vertebral-level reasoning across imaging modalities with over 450,000 instruction instances, and SpineBench, a clinically-grounded evaluation framework. SpineMed-450k is curated from diverse sources, including textbooks, guidelines, open datasets, and ~1,000 de-identified hospital cases, using a clinician-in-the-loop pipeline with a two-stage LLM generation method (draft and revision) to ensure high-quality, traceable data for question-answering, multi-turn consultations, and report generation. SpineBench evaluates models on clinically salient axes, including level identification, pathology assessment, and surgical planning. Our comprehensive evaluation of several recently advanced large vision-language models (LVLMs) on SpineBench reveals systematic weaknesses in fine-grained, level-specific reasoning. In contrast, our model fine-tuned on SpineMed-450k demonstrates consistent and significant improvements across all tasks. Clinician assessments confirm the diagnostic clarity and practical utility of our model's outputs.

Paperid: 4476, https://arxiv.org/pdf/2510.00088.pdf

Abstract:
Large language models (LLMs) have been extensively used for legal judgment prediction tasks based on case reports and crime history. However, with a surge in the availability of large vision language models (VLMs), legal judgment prediction systems can now be made to leverage the images of the criminals in addition to the textual case reports/crime history. Applications built in this way could lead to inadvertent consequences and be used with malicious intent. In this work, we run an audit to investigate the efficiency of standalone VLMs in the bail decision prediction task. We observe that the performance is poor across multiple intersectional groups and models \textit{wrongly deny bail to deserving individuals with very high confidence}. We design different intervention algorithms by first including legal precedents through a RAG pipeline and then fine-tuning the VLMs using innovative schemes. We demonstrate that these interventions substantially improve the performance of bail prediction. Our work paves the way for the design of smarter interventions on VLMs in the future, before they can be deployed for real-world legal judgment prediction.

Paperid: 4477, https://arxiv.org/pdf/2510.00060.pdf

Abstract:
In this work, we reconceptualize autonomous driving as a generalized language and formulate the trajectory planning task as next waypoint prediction. We introduce Max-V1, a novel framework for one-stage end-to-end autonomous driving. Our framework presents a single-pass generation paradigm that aligns with the inherent sequentiality of driving. This approach leverages the generative capacity of the VLM (Vision-Language Model) to enable end-to-end trajectory prediction directly from front-view camera input. The efficacy of this method is underpinned by a principled supervision strategy derived from statistical modeling. This provides a well-defined learning objective, which makes the framework highly amenable to master complex driving policies through imitation learning from large-scale expert demonstrations. Empirically, our method achieves the state-of-the-art performance on the nuScenes dataset, delivers an overall improvement of over 30% compared to prior baselines. Furthermore, it exhibits superior generalization performance on cross-domain datasets acquired from diverse vehicles, demonstrating notable potential for cross-vehicle robustness and adaptability. Due to these empirical strengths, this work introduces a model enabling fundamental driving behaviors, laying the foundation for the development of more capable self-driving agents. Code will be available upon publication.

Paperid: 4478, https://arxiv.org/pdf/2509.26272.pdf

Abstract:
The rapid rise of synthetic media has made deepfake detection a critical challenge for online safety and trust. Progress remains constrained by the scarcity of large, high-quality datasets. Although multimodal large language models (LLMs) exhibit strong reasoning capabilities, their performance on deepfake detection is poor, often producing explanations that are misaligned with visual evidence or hallucinatory. To address this limitation, we introduce a reasoning-annotated dataset for deepfake detection and propose Paragraph-level Relative Policy Optimization (PRPO), a reinforcement learning algorithm that aligns LLM reasoning with image content at the paragraph level. Experiments show that PRPO improves detection accuracy by a wide margin and achieves the highest reasoning score of 4.55/5.0. Ablation studies further demonstrate that PRPO significantly outperforms GRPO under test-time conditions. These results underscore the importance of grounding multimodal reasoning in visual evidence to enable more reliable and interpretable deepfake detection.

Paperid: 4479, https://arxiv.org/pdf/2509.25525.pdf

Abstract:
Multimodal large language models (MLLMs) have demonstrated remarkable capabilities in processing and reasoning over diverse modalities, but their advanced abilities also raise significant privacy concerns, particularly regarding Personally Identifiable Information (PII) leakage. While relevant research has been conducted on single-modal language models to some extent, the vulnerabilities in the multimodal setting have yet to be fully investigated. In this work, we investigate these emerging risks with a focus on vision language models (VLMs), a representative subclass of MLLMs that covers the two modalities most relevant for PII leakage, vision and text. We introduce a concept-guided mitigation approach that identifies and modifies the model's internal states associated with PII-related content. Our method guides VLMs to refuse PII-sensitive tasks effectively and efficiently, without requiring re-training or fine-tuning. We also address the current lack of multimodal PII datasets by constructing various ones that simulate real-world scenarios. Experimental results demonstrate that the method can achieve an average refusal rate of 93.3% for various PII-related tasks with minimal impact on unrelated model performances. We further examine the mitigation's performance under various conditions to show the adaptability of our proposed method.

Paperid: 4480, https://arxiv.org/pdf/2509.25249.pdf

Abstract:
This paper introduces BEV-VLM, a novel framework for trajectory planning in autonomous driving that leverages Vision-Language Models (VLMs) with Bird's-Eye View (BEV) feature maps as visual inputs. Unlike conventional approaches that rely solely on raw visual data such as camera images, our method utilizes highly compressed and informative BEV representations, which are generated by fusing multi-modal sensor data (e.g., camera and LiDAR) and aligning them with HD Maps. This unified BEV-HD Map format provides a geometrically consistent and rich scene description, enabling VLMs to perform accurate trajectory planning. Experimental results on the nuScenes dataset demonstrate 44.8% improvements in planning accuracy and complete collision avoidance. Our work highlights that VLMs can effectively interpret processed visual representations like BEV features, expanding their applicability beyond raw images in trajectory planning.

Paperid: 4481, https://arxiv.org/pdf/2509.22195.pdf

Abstract:
Fine-tuning vision-language models (VLMs) on robot teleoperation data to create vision-language-action (VLA) models is a promising paradigm for training generalist policies, but it suffers from a fundamental tradeoff: learning to produce actions often diminishes the VLM's foundational reasoning and multimodal understanding, hindering generalization to novel scenarios, instruction following, and semantic understanding. We argue that this catastrophic forgetting is due to a distribution mismatch between the VLM's internet-scale pretraining corpus and the robotics fine-tuning data. Inspired by this observation, we introduce VLM2VLA: a VLA training paradigm that first resolves this mismatch at the data level by representing low-level actions with natural language. This alignment makes it possible to train VLAs solely with Low-Rank Adaptation (LoRA), thereby minimally modifying the VLM backbone and averting catastrophic forgetting. As a result, the VLM can be fine-tuned on robot teleoperation data without fundamentally altering the underlying architecture and without expensive co-training on internet-scale VLM datasets. Through extensive Visual Question Answering (VQA) studies and over 800 real-world robotics experiments, we demonstrate that VLM2VLA preserves the VLM's core capabilities, enabling zero-shot generalization to novel tasks that require open-world semantic reasoning and multilingual instruction following.

Paperid: 4482, https://arxiv.org/pdf/2509.21356.pdf

Abstract:
With the emergence of large-scale vision language models (VLM), it is now possible to produce realistic-looking radiology reports for chest X-ray images. However, their clinical translation has been hampered by the factual errors and hallucinations in the produced descriptions during inference. In this paper, we present a novel phrase-grounded fact-checking model (FC model) that detects errors in findings and their indicated locations in automatically generated chest radiology reports. Specifically, we simulate the errors in reports through a large synthetic dataset derived by perturbing findings and their locations in ground truth reports to form real and fake findings-location pairs with images. A new multi-label cross-modal contrastive regression network is then trained on this dataset. We present results demonstrating the robustness of our method in terms of accuracy of finding veracity prediction and localization on multiple X-ray datasets. We also show its effectiveness for error detection in reports of SOTA report generators on multiple datasets achieving a concordance correlation coefficient of 0.997 with ground truth-based verification, thus pointing to its utility during clinical inference in radiology workflows.

Paperid: 4483, https://arxiv.org/pdf/2509.19858.pdf

Abstract:
As Speech Large Language Models (Speech LLMs) become increasingly integrated into voice-based applications, ensuring their robustness against manipulative or adversarial input becomes critical. Although prior work has studied adversarial attacks in text-based LLMs and vision-language models, the unique cognitive and perceptual challenges of speech-based interaction remain underexplored. In contrast, speech presents inherent ambiguity, continuity, and perceptual diversity, which make adversarial attacks more difficult to detect. In this paper, we introduce gaslighting attacks, strategically crafted prompts designed to mislead, override, or distort model reasoning as a means to evaluate the vulnerability of Speech LLMs. Specifically, we construct five manipulation strategies: Anger, Cognitive Disruption, Sarcasm, Implicit, and Professional Negation, designed to test model robustness across varied tasks. It is worth noting that our framework captures both performance degradation and behavioral responses, including unsolicited apologies and refusals, to diagnose different dimensions of susceptibility. Moreover, acoustic perturbation experiments are conducted to assess multi-modal robustness. To quantify model vulnerability, comprehensive evaluation across 5 Speech and multi-modal LLMs on over 10,000 test samples from 5 diverse datasets reveals an average accuracy drop of 24.3% under the five gaslighting attacks, indicating significant behavioral vulnerability. These findings highlight the need for more resilient and trustworthy speech-based AI systems.

Paperid: 4484, https://arxiv.org/pdf/2509.16654.pdf

Abstract:
Vision-Language Models (VLMs) have recently shown remarkable progress in multimodal reasoning, yet their applications in autonomous driving remain limited. In particular, the ability to understand road topology, a key requirement for safe navigation, has received relatively little attention. While some recent works have begun to explore VLMs in driving contexts, their performance on topology reasoning is far from satisfactory. In this work, we systematically evaluate VLMs' capabilities in road topology understanding. Specifically, multi-view images are projected into unified ground-plane coordinate system and fused into bird's-eye-view (BEV) lanes. Based on these BEV lanes, we formulate four topology-related diagnostic VQA tasks, which together capture essential components of spatial topology reasoning. Through extensive evaluation, we find that while frontier closed-source models (e.g., GPT-4o) achieve relatively high accuracy in some tasks, they still fail in some temporal questions that humans can answer (e.g., GPT-4o achieve only 67.8% in vector, a two-class classification problem). Furthermore, we find open-source VLMs, even at 30B scale, struggle significantly. These results indicate that spatial reasoning remains a fundamental bottleneck for current VLMs. We also find that the model's capability is positively correlated with model size, length of reasoning tokens and shots provided as examples, showing direction for future research.

Paperid: 4485, https://arxiv.org/pdf/2509.15695.pdf

Abstract:
Large Vision-Language Models (LVLMs) have made significant strides in image caption, visual question answering, and robotics by integrating visual and textual information. However, they remain prone to errors in incongruous contexts, where objects appear unexpectedly or are absent when contextually expected. This leads to two key recognition failures: object misidentification and hallucination. To systematically examine this issue, we introduce the Object Recognition in Incongruous Context Benchmark (ORIC), a novel benchmark that evaluates LVLMs in scenarios where object-context relationships deviate from expectations. ORIC employs two key strategies: (1) LLM-guided sampling, which identifies objects that are present but contextually incongruous, and (2) CLIP-guided sampling, which detects plausible yet nonexistent objects that are likely to be hallucinated, thereby creating an incongruous context. Evaluating 18 LVLMs and two open-vocabulary detection models, our results reveal significant recognition gaps, underscoring the challenges posed by contextual incongruity. This work provides critical insights into LVLMs' limitations and encourages further research on context-aware object recognition.

Paperid: 4486, https://arxiv.org/pdf/2509.14060.pdf

Abstract:
Current multi-object tracking (MOT) algorithms typically overlook issues inherent in low-quality videos, leading to significant degradation in tracking performance when confronted with real-world image deterioration. Therefore, advancing the application of MOT algorithms in real-world low-quality video scenarios represents a critical and meaningful endeavor. To address the challenges posed by low-quality scenarios, inspired by vision-language models, this paper proposes a Visual Semantic Enhancement-guided Multi-Object Tracking framework (VSE-MOT). Specifically, we first design a tri-branch architecture that leverages a vision-language model to extract global visual semantic information from images and fuse it with query vectors. Subsequently, to further enhance the utilization of visual semantic information, we introduce the Multi-Object Tracking Adapter (MOT-Adapter) and the Visual Semantic Fusion Module (VSFM). The MOT-Adapter adapts the extracted global visual semantic information to suit multi-object tracking tasks, while the VSFM improves the efficacy of feature fusion. Through extensive experiments, we validate the effectiveness and superiority of the proposed method in real-world low-quality video scenarios. Its tracking performance metrics outperform those of existing methods by approximately 8% to 20%, while maintaining robust performance in conventional scenarios.

Paperid: 4487, https://arxiv.org/pdf/2509.13769.pdf

Abstract:
While reasoning technology like Chain of Thought (CoT) has been widely adopted in Vision Language Action (VLA) models, it demonstrates promising capabilities in end to end autonomous driving. However, recent efforts to integrate CoT reasoning often fall short in simple scenarios, introducing unnecessary computational overhead without improving decision quality. To address this, we propose AdaThinkDrive, a novel VLA framework with a dual mode reasoning mechanism inspired by fast and slow thinking. First, our framework is pretrained on large scale autonomous driving (AD) scenarios using both question answering (QA) and trajectory datasets to acquire world knowledge and driving commonsense. During supervised fine tuning (SFT), we introduce a two mode dataset, fast answering (w/o CoT) and slow thinking (with CoT), enabling the model to distinguish between scenarios that require reasoning. Furthermore, an Adaptive Think Reward strategy is proposed in conjunction with the Group Relative Policy Optimization (GRPO), which rewards the model for selectively applying CoT by comparing trajectory quality across different reasoning modes. Extensive experiments on the Navsim benchmark show that AdaThinkDrive achieves a PDMS of 90.3, surpassing the best vision only baseline by 1.7 points. Moreover, ablations show that AdaThinkDrive surpasses both the never Think and always Think baselines, improving PDMS by 2.0 and 1.4, respectively. It also reduces inference time by 14% compared to the always Think baseline, demonstrating its ability to balance accuracy and efficiency through adaptive reasoning.

Paperid: 4488, https://arxiv.org/pdf/2509.12278.pdf

Abstract:
Text Image Machine Translation (TIMT) aims to translate texts embedded within an image into another language. Current TIMT studies primarily focus on providing translations for all the text within an image, while neglecting to provide bounding boxes and covering limited scenarios. In this work, we extend traditional TIMT into position-aware TIMT (PATIMT), aiming to support fine-grained and layoutpreserving translation, which holds great practical value but remains largely unexplored. This task comprises two key sub-tasks: regionspecific translation and full-image translation with grounding. To support existing models on PATIMT and conduct fair evaluation, we construct the PATIMT benchmark (PATIMTBench), which consists of 10 diverse real-world scenarios. Specifically, we introduce an Adaptive Image OCR Refinement Pipeline, which adaptively selects appropriate OCR tools based on scenario and refines the results of text-rich images. To ensure evaluation reliability, we further construct a test set, which contains 1,200 high-quality instances manually annotated and reviewed by human experts. After fine-tuning on our data, compact Large Vision-Language Models (LVLMs) achieve state-of-the-art performance on both sub-tasks. Experimental results also highlight the scalability and generalizability of our training data

Paperid: 4489, https://arxiv.org/pdf/2509.11862.pdf

Abstract:
Video Question Answering (VQA) requires models to reason over spatial, temporal, and causal cues in videos. Recent vision language models (VLMs) achieve strong results but often rely on shallow correlations, leading to weak temporal grounding and limited interpretability. We study symbolic scene graphs (SGs) as intermediate grounding signals for VQA. SGs provide structured object-relation representations that complement VLMs holistic reasoning. We introduce SG-VLM, a modular framework that integrates frozen VLMs with scene graph grounding via prompting and visual localization. Across three benchmarks (NExT-QA, iVQA, ActivityNet-QA) and multiple VLMs (QwenVL, InternVL), SG-VLM improves causal and temporal reasoning and outperforms prior baselines, though gains over strong VLMs are limited. These findings highlight both the promise and current limitations of symbolic grounding, and offer guidance for future hybrid VLM-symbolic approaches in video understanding.

Paperid: 4490, https://arxiv.org/pdf/2509.07334.pdf

Abstract:
Large language models (LLMs) promise to accelerate UI design, yet current tools struggle with two fundamentals: externalizing designers' intent and controlling iterative change. We introduce SPEC, a structured, parameterized, hierarchical intermediate representation that exposes UI elements as controllable parameters. Building on SPEC, we present SpecifyUI, an interactive system that extracts SPEC from UI references via region segmentation and vision-language models, composes UIs across multiple sources, and supports targeted edits at global, regional, and component levels. A multi-agent generator renders SPEC into high-fidelity designs, closing the loop between intent expression and controllable generation. Quantitative experiments show SPEC-based generation more faithfully captures reference intent than prompt-based baselines. In a user study with 16 professional designers, SpecifyUI significantly outperformed Stitch on intent alignment, design quality, controllability, and overall experience in human-AI co-creation. Our results position SPEC as a specification-driven paradigm that shifts LLM-assisted design from one-shot prompting to iterative, collaborative workflows.

Paperid: 4491, https://arxiv.org/pdf/2509.05215.pdf

Abstract:
Recent works propose complex multi-modal models that handle both time series and language, ultimately claiming high performance on complex tasks like time series reasoning and cross-modal question-answering. However, they skip evaluations of simple and important foundational tasks, which complex models should reliably master. They also lack direct, head-to-head comparisons with other popular approaches. So we ask a simple question: Can recent models even produce generic visual descriptions of time series data? In response, we propose three new tasks, posing that successful multi-modal models should be able to recognize, differentiate, and generate language descriptions of time series. We then create BEDTime, the first benchmark dataset to assess models on each task, comprising four datasets reformatted for these tasks across multiple modalities. Using BEDTime, we evaluate 13 state-of-the-art models, and find that (1) surprisingly, dedicated time series foundation models severely underperform, despite being designed for similar tasks, (2) vision-language models are quite capable, (3) language-only methods perform worst, despite many lauding their potential, and (4) all approaches are clearly fragile to a range of realistic robustness tests, indicating avenues for future work.

Paperid: 4492, https://arxiv.org/pdf/2509.03025.pdf

Abstract:
Large Vision-Language Models (LVLMs) generate contextually relevant responses by jointly interpreting visual and textual inputs. However, our finding reveals they often mistakenly perceive text inputs lacking visual evidence as being part of the image, leading to erroneous responses. In light of this finding, we probe whether LVLMs possess an internal capability to determine if textual concepts are grounded in the image, and discover a specific subset of Feed-Forward Network (FFN) neurons, termed Visual Absence-aware (VA) neurons, that consistently signal the visual absence through a distinctive activation pattern. Leveraging these patterns, we develop a detection module that systematically classifies whether an input token is visually grounded. Guided by its prediction, we propose a method to refine the outputs by reinterpreting question prompts or replacing the detected absent tokens during generation. Extensive experiments show that our method effectively mitigates the models' tendency to falsely presume the visual presence of text input and its generality across various LVLMs.

Paperid: 4493, https://arxiv.org/pdf/2508.20670.pdf

Abstract:
Recent advances in multimodal AI have enabled progress in detecting synthetic and out-of-context content. However, existing efforts largely overlook the intent behind AI-generated images. To fill this gap, we introduce S-HArM, a multimodal dataset for intent-aware classification, comprising 9,576 "in the wild" image-text pairs from Twitter/X and Reddit, labeled as Humor/Satire, Art, or Misinformation. Additionally, we explore three prompting strategies (image-guided, description-guided, and multimodally-guided) to construct a large-scale synthetic training dataset with Stable Diffusion. We conduct an extensive comparative study including modality fusion, contrastive learning, reconstruction networks, attention mechanisms, and large vision-language models. Our results show that models trained on image- and multimodally-guided data generalize better to "in the wild" content, due to preserved visual context. However, overall performance remains limited, highlighting the complexity of inferring intent and the need for specialized architectures.

Paperid: 4494, https://arxiv.org/pdf/2508.18264.pdf

Abstract:
Vision-Language Models (VLMs) demonstrate impressive performance in understanding visual content with language instruction by converting visual input to vision tokens. However, redundancy in vision tokens results in the degenerated inference efficiency of VLMs. While many algorithms have been proposed to reduce the number of vision tokens, most of them apply only unimodal information (i.e., vision/text) for pruning and ignore the inherent multimodal property of vision-language tasks. Moreover, it lacks a generic criterion that can be applied to different modalities. To mitigate this limitation, in this work, we propose to leverage both vision and text tokens to select informative vision tokens by the criterion of coverage. We first formulate the subset selection problem as a maximum coverage problem. Afterward, a subset of vision tokens is optimized to cover the text tokens and the original set of vision tokens, simultaneously. Finally, a VLM agent can be adopted to further improve the quality of text tokens for guiding vision pruning. The proposed method MMTok is extensively evaluated on benchmark datasets with different VLMs. The comparison illustrates that vision and text information are complementary, and combining multimodal information can surpass the unimodal baseline with a clear margin. Moreover, under the maximum coverage criterion on the POPE dataset, our method achieves a 1.87x speedup while maintaining 98.7% of the original performance on LLaVA-NeXT-13B. Furthermore, with only four vision tokens, it still preserves 87.7% of the original performance on LLaVA-1.5-7B. These results highlight the effectiveness of coverage in token selection.

Paperid: 4495, https://arxiv.org/pdf/2508.17846.pdf

Abstract:
Recent advances in pre-trained vision-language models have demonstrated remarkable zero-shot generalization capabilities. To further enhance these models' adaptability to various downstream tasks, prompt tuning has emerged as a parameter-efficient fine-tuning method. However, despite its efficiency, the generalization ability of prompt remains limited. In contrast, label smoothing (LS) has been widely recognized as an effective regularization technique that prevents models from becoming over-confident and improves their generalization. This inspires us to explore the integration of LS with prompt tuning. However, we have observed that the vanilla LS even weakens the generalization ability of prompt tuning. To address this issue, we propose the Alternating Training-based Label Smoothing (ATLaS) method, which alternately trains with standard one-hot labels and soft labels generated by LS to supervise the prompt tuning. Moreover, we introduce two types of efficient offline soft labels, including Class-wise Soft Labels (CSL) and Instance-wise Soft Labels (ISL), to provide inter-class or instance-class relationships for prompt tuning. The theoretical properties of the proposed ATLaS method are analyzed. Extensive experiments demonstrate that the proposed ATLaS method, combined with CSL and ISL, consistently enhances the generalization performance of prompt tuning. Moreover, the proposed ATLaS method exhibits high compatibility with prevalent prompt tuning methods, enabling seamless integration into existing methods.

Paperid: 4496, https://arxiv.org/pdf/2508.16076.pdf

Abstract:
Sign Language (SL) enables two-way communication for the deaf and hard-of-hearing community, yet many sign languages remain under-resourced in the AI space. Sign Language Instruction Generation (SLIG) produces step-by-step textual instructions that enable non-SL users to imitate and learn SL gestures, promoting two-way interaction. We introduce BdSLIG, the first Bengali SLIG dataset, used to evaluate Vision Language Models (VLMs) (i) on under-resourced SLIG tasks, and (ii) on long-tail visual concepts, as Bengali SL is unlikely to appear in the VLM pre-training data. To enhance zero-shot performance, we introduce Sign Parameter-Infused (SPI) prompting, which integrates standard SL parameters, like hand shape, motion, and orientation, directly into the textual prompts. Subsuming standard sign parameters into the prompt makes the instructions more structured and reproducible than free-form natural text from vanilla prompting. We envision that our work would promote inclusivity and advancement in SL learning systems for the under-resourced communities.

Paperid: 4497, https://arxiv.org/pdf/2508.09087.pdf

Abstract:
Vision-Language Models (VLMs) have achieved remarkable success on multimodal tasks such as image-text retrieval and zero-shot classification, yet they can exhibit demographic biases even when explicit protected attributes are absent during training. In this work, we focus on automated glaucoma screening from retinal fundus images, a critical application given that glaucoma is a leading cause of irreversible blindness and disproportionately affects underserved populations. Building on a reweighting-based contrastive learning framework, we introduce an attribute-agnostic debiasing method that (i) infers proxy subgroups via unsupervised clustering of image-image embeddings, (ii) computes gradient-similarity weights between the CLIP-style multimodal loss and a SimCLR-style image-pair contrastive loss, and (iii) applies these weights in a joint, top-$k$ weighted objective to upweight underperforming clusters. This label-free approach adaptively targets the hardest examples, thereby reducing subgroup disparities. We evaluate our method on the Harvard FairVLMed glaucoma subset, reporting Equalized Odds Distance (EOD), Equalized Subgroup AUC (ES AUC), and Groupwise AUC to demonstrate equitable performance across inferred demographic subgroups.

Paperid: 4498, https://arxiv.org/pdf/2508.09057.pdf

Abstract:
Given the significant advances in Large Vision Language Models (LVLMs) in reasoning and visual understanding, mobile agents are rapidly emerging to meet users' automation needs. However, existing evaluation benchmarks are disconnected from the real world and fail to adequately address the diverse and complex requirements of users. From our extensive collection of user questionnaire, we identified five tasks: Multi-App, Vague, Interactive, Single-App, and Unethical Instructions. Around these tasks, we present \textbf{MVISU-Bench}, a bilingual benchmark that includes 404 tasks across 137 mobile applications. Furthermore, we propose Aider, a plug-and-play module that acts as a dynamic prompt prompter to mitigate risks and clarify user intent for mobile agents. Our Aider is easy to integrate into several frameworks and has successfully improved overall success rates by 19.55\% compared to the current state-of-the-art (SOTA) on MVISU-Bench. Specifically, it achieves success rate improvements of 53.52\% and 29.41\% for unethical and interactive instructions, respectively. Through extensive experiments and analysis, we highlight the gap between existing mobile agents and real-world user expectations.

Paperid: 4499, https://arxiv.org/pdf/2508.07650.pdf

Abstract:
Vision-language-action models have emerged as a crucial paradigm in robotic manipulation. However, existing VLA models exhibit notable limitations in handling ambiguous language instructions and unknown environmental states. Furthermore, their perception is largely constrained to static two-dimensional observations, lacking the capability to model three-dimensional interactions between the robot and its environment. To address these challenges, this paper proposes GraphCoT-VLA, an efficient end-to-end model. To enhance the model's ability to interpret ambiguous instructions and improve task planning, we design a structured Chain-of-Thought reasoning module that integrates high-level task understanding and planning, failed task feedback, and low-level imaginative reasoning about future object positions and robot actions. Additionally, we construct a real-time updatable 3D Pose-Object graph, which captures the spatial configuration of robot joints and the topological relationships between objects in 3D space, enabling the model to better understand and manipulate their interactions. We further integrates a dropout hybrid reasoning strategy to achieve efficient control outputs. Experimental results across multiple real-world robotic tasks demonstrate that GraphCoT-VLA significantly outperforms existing methods in terms of task success rate and response speed, exhibiting strong generalization and robustness in open environments and under uncertain instructions.

Paperid: 4500, https://arxiv.org/pdf/2508.07414.pdf

Abstract:
Multimodal Large Language Models excel in high-resource settings, but often misinterpret long-tail cultural entities and underperform in low-resource languages. To address this gap, we propose a data-centric approach that directly grounds MLLMs in cultural knowledge. Leveraging a large scale knowledge graph from Wikidata, we collect images that represent culturally significant entities, and generate synthetic multilingual visual question answering data. The resulting dataset, CulturalGround, comprises 22 million high-quality, culturally-rich VQA pairs spanning 42 countries and 39 languages. We train an open-source MLLM CulturalPangea on CulturalGround, interleaving standard multilingual instruction-tuning data to preserve general abilities. CulturalPangea achieves state-of-the-art performance among open models on various culture-focused multilingual multimodal benchmarks, outperforming prior models by an average of 5.0 without degrading results on mainstream vision-language tasks. Our findings show that our targeted, culturally grounded approach could substantially narrow the cultural gap in MLLMs and offer a practical path towards globally inclusive multimodal systems.

Paperid: 4501, https://arxiv.org/pdf/2508.03164.pdf

Abstract:
Generating accurate, informative, and hallucination-free captions for charts remains challenging for vision language models, primarily due to the lack of large-scale, high-quality datasets of real-world charts. However, existing real-world chart datasets suffer from the inclusion of extraneous information that cannot be inferred from the chart and failure to sufficiently capture structural elements and key insights. Therefore, we introduce ChartCap, a large-scale dataset of 565K real-world chart images paired with type-specific, dense captions that exclude extraneous information and highlight both structural elements and key insights in detail. To build ChartCap, we design a four-stage pipeline that generates captions using only the discernible data from the chart and employ a cycle consistency-based human verification, which accelerates quality control without sacrificing accuracy. Additionally, we propose a novel metric, the Visual Consistency Score, which evaluates caption quality by measuring the similarity between the chart regenerated from a caption and the original chart, independent of reference captions. Extensive experiments confirms that models fine-tuned on ChartCap consistently generate more accurate and informative captions with reduced hallucinations, surpassing both open-source and proprietary models and even human-annotated captions.

Paperid: 4502, https://arxiv.org/pdf/2508.01602.pdf

Abstract:
The fine-grained classification of brain tumor subtypes from histopathological whole slide images is highly challenging due to subtle morphological variations and the scarcity of annotated data. Although vision-language models have enabled promising zero-shot classification, their ability to capture fine-grained pathological features remains limited, resulting in suboptimal subtype discrimination. To address these challenges, we propose the Fine-Grained Patch Alignment Network (FG-PAN), a novel zero-shot framework tailored for digital pathology. FG-PAN consists of two key modules: (1) a local feature refinement module that enhances patch-level visual features by modeling spatial relationships among representative patches, and (2) a fine-grained text description generation module that leverages large language models to produce pathology-aware, class-specific semantic prototypes. By aligning refined visual features with LLM-generated fine-grained descriptions, FG-PAN effectively increases class separability in both visual and semantic spaces. Extensive experiments on multiple public pathology datasets, including EBRAINS and TCGA, demonstrate that FG-PAN achieves state-of-the-art performance and robust generalization in zero-shot brain tumor subtype classification.

Paperid: 4503, https://arxiv.org/pdf/2507.19983.pdf

Abstract:
Clothes manipulation, such as folding or hanging, is a critical capability for home service robots. Despite recent advances, most existing methods remain limited to specific tasks and clothes types, due to the complex, high-dimensional geometry of clothes. This paper presents CLothes mAnipulation with Semantic keyPoints (CLASP), which aims at general-purpose clothes manipulation over different clothes types, T-shirts, shorts, skirts, long dresses, ... , as well as different tasks, folding, flattening, hanging, ... . The core idea of CLASP is semantic keypoints -- e.g., ''left sleeve'', ''right shoulder'', etc. -- a sparse spatial-semantic representation that is salient for both perception and action. Semantic keypoints of clothes can be reliably extracted from RGB-D images and provide an effective intermediate representation of clothes manipulation policies. CLASP uses semantic keypoints to bridge high-level task planning and low-level action execution. At the high level, it exploits vision language models (VLMs) to predict task plans over the semantic keypoints. At the low level, it executes the plans with the help of a simple pre-built manipulation skill library. Extensive simulation experiments show that CLASP outperforms state-of-the-art baseline methods on multiple tasks across diverse clothes types, demonstrating strong performance and generalization. Further experiments with a Franka dual-arm system on four distinct tasks -- folding, flattening, hanging, and placing -- confirm CLASP's performance on a real robot.

Paperid: 4504, https://arxiv.org/pdf/2507.18031.pdf

Abstract:
The rapid rise of deepfake technology, which produces realistic but fraudulent digital content, threatens the authenticity of media. Traditional deepfake detection approaches often struggle with sophisticated, customized deepfakes, especially in terms of generalization and robustness against malicious attacks. This paper introduces ViGText, a novel approach that integrates images with Vision Large Language Model (VLLM) Text explanations within a Graph-based framework to improve deepfake detection. The novelty of ViGText lies in its integration of detailed explanations with visual data, as it provides a more context-aware analysis than captions, which often lack specificity and fail to reveal subtle inconsistencies. ViGText systematically divides images into patches, constructs image and text graphs, and integrates them for analysis using Graph Neural Networks (GNNs) to identify deepfakes. Through the use of multi-level feature extraction across spatial and frequency domains, ViGText captures details that enhance its robustness and accuracy to detect sophisticated deepfakes. Extensive experiments demonstrate that ViGText significantly enhances generalization and achieves a notable performance boost when it detects user-customized deepfakes. Specifically, average F1 scores rise from 72.45% to 98.32% under generalization evaluation, and reflects the model's superior ability to generalize to unseen, fine-tuned variations of stable diffusion models. As for robustness, ViGText achieves an increase of 11.1% in recall compared to other deepfake detection approaches. When facing targeted attacks that exploit its graph-based architecture, ViGText limits classification performance degradation to less than 4%. ViGText uses detailed visual and textual analysis to set a new standard for detecting deepfakes, helping ensure media authenticity and information integrity.

Paperid: 4505, https://arxiv.org/pdf/2507.05887.pdf

Abstract:
The application of Vision-Language Models (VLMs) in remote sensing (RS) image understanding has achieved notable progress, demonstrating the basic ability to recognize and describe geographical entities. However, existing RS-VLMs are mostly limited to image-level and region-level tasks, lacking the capability to handle pixel-level tasks and performing poorly in small-object recognition scenarios. Moreover, RS-VLMs consume significant computational resources when processing high-resolution RS images, further restricting their practical applicability. In this context, we propose GeoMag (Geographical Magnifier), an end-to-end general-purpose large model framework for RS. GeoMag dynamically focuses the attention scope based on prompt semantics to effectively perform remote sensing image parsing across multiple levels of granularity. This method introduces Task-driven Multi-granularity Resolution Adjustment (TMRA) and Prompt-guided Semantic-aware Cropping (PSC), which adaptively reduce the spatial resolution of task-irrelevant regions while enhancing the visual representation of task-relevant areas. This approach improves the model's perception of critical target regions, suppresses background redundancy, and reduces the computational cost of interpreting high-resolution RS imagery. Extensive comparative experiments on 10 benchmarks demonstrate that GeoMag not only excels in handling pixel-level tasks but also maintains competitive performance across tasks of other granularities compared to existing RS-VLMs.

Paperid: 4506, https://arxiv.org/pdf/2507.00537.pdf

Abstract:
This paper studies the role of attention heads in CLIP's image encoder. While CLIP has exhibited robust performance across diverse applications, we hypothesize that certain attention heads negatively affect final representations and that ablating them can improve performance in downstream tasks. To capitalize on this insight, we propose a simple yet effective method, called Attention Ablation Technique (AAT), to suppress the contribution of specific heads by manipulating attention weights. By integrating two alternative strategies tailored for different application scenarios, AAT systematically identifies and ablates detrimental attention heads to enhance representation quality. Experiments demonstrate that AAT consistently improves downstream task performance across various domains, boosting recall rate by up to 11.1% on CLIP-family models for cross-modal retrieval. The results highlight the potential of AAT to effectively refine large-scale vision-language models with virtually no increase in inference cost.

Paperid: 4507, https://arxiv.org/pdf/2506.22832.pdf

Abstract:
Training robust and generalizable reward models for human visual preferences is essential for aligning text-to-image and text-to-video generative models with human intent. However, current reward models often fail to generalize, and supervised fine-tuning leads to memorization, demanding complex annotation pipelines. While reinforcement learning (RL), specifically Group Relative Policy Optimization (GRPO), improves generalization, we uncover a key failure mode: a significant drop in reasoning accuracy occurs when a model's reasoning trace contradicts that of an independent, frozen vision-language model ("listener") evaluating the same output. To address this, we introduce a listener-augmented GRPO framework. Here, the listener re-evaluates the reasoner's chain-of-thought to provide a dense, calibrated confidence score, shaping the RL reward signal. This encourages the reasoner not only to answer correctly, but to produce explanations that are persuasive to an independent model. Our listener-shaped reward scheme achieves best accuracy on the ImageReward benchmark (67.4%), significantly improves out-of-distribution (OOD) performance on a large-scale human preference dataset (1.2M votes, up to +6% over naive reasoner), and reduces reasoning contradictions compared to strong GRPO and SFT baselines. These results demonstrate that listener-based rewards provide a scalable, data-efficient path to aligning vision-language models with nuanced human preferences. We will release our reasoning model here: https://huggingface.co/alexgambashidze/qwen2.5vl_image_preference_reasoner.

Paperid: 4508, https://arxiv.org/pdf/2506.21885.pdf

Abstract:
Multi-sensor fusion plays a critical role in enhancing perception for autonomous driving, overcoming individual sensor limitations, and enabling comprehensive environmental understanding. This paper first formalizes multi-sensor fusion strategies into data-level, feature-level, and decision-level categories and then provides a systematic review of deep learning-based methods corresponding to each strategy. We present key multi-modal datasets and discuss their applicability in addressing real-world challenges, particularly in adverse weather conditions and complex urban environments. Additionally, we explore emerging trends, including the integration of Vision-Language Models (VLMs), Large Language Models (LLMs), and the role of sensor fusion in end-to-end autonomous driving, highlighting its potential to enhance system adaptability and robustness. Our work offers valuable insights into current methods and future directions for multi-sensor fusion in autonomous driving.

Paperid: 4509, https://arxiv.org/pdf/2506.21874.pdf

Abstract:
Today's text-to-image generative models are trained on millions of images sourced from the Internet, each paired with a detailed caption produced by Vision-Language Models (VLMs). This part of the training pipeline is critical for supplying the models with large volumes of high-quality image-caption pairs during training. However, recent work suggests that VLMs are vulnerable to stealthy adversarial attacks, where adversarial perturbations are added to images to mislead the VLMs into producing incorrect captions. In this paper, we explore the feasibility of adversarial mislabeling attacks on VLMs as a mechanism to poisoning training pipelines for text-to-image models. Our experiments demonstrate that VLMs are highly vulnerable to adversarial perturbations, allowing attackers to produce benign-looking images that are consistently miscaptioned by the VLM models. This has the effect of injecting strong "dirty-label" poison samples into the training pipeline for text-to-image models, successfully altering their behavior with a small number of poisoned samples. We find that while potential defenses can be effective, they can be targeted and circumvented by adaptive attackers. This suggests a cat-and-mouse game that is likely to reduce the quality of training data and increase the cost of text-to-image model development. Finally, we demonstrate the real-world effectiveness of these attacks, achieving high attack success (over 73%) even in black-box scenarios against commercial VLMs (Google Vertex AI and Microsoft Azure).

Paperid: 4510, https://arxiv.org/pdf/2506.21586.pdf

Abstract:
Nonverbal communication (NVC) plays an integral role in human language, but studying NVC in general is challenging because of its broad scope and high variance in interpretation among individuals and cultures. However, mime -- the theatrical technique of suggesting intent using only gesture, expression, and movement -- is a subset of NVC that consists of explicit and embodied actions with much lower human interpretation variance. We argue that a solid understanding of mimed actions is a crucial prerequisite for vision-language models capable of interpreting and commanding more subtle aspects of NVC. Hence, we propose Mime Identification Multimodal Evaluation (MIME), a novel video-based question answering benchmark comprising of 86 mimed actions. Constructed with motion capture data, MIME consists of variations of each action with perturbations applied to the character, background, and viewpoint for evaluating recognition robustness. We find that both open-weight and API-based vision-language models perform significantly worse than humans on MIME, motivating the need for increased research for instilling more robust understanding of human gestures.

Paperid: 4511, https://arxiv.org/pdf/2506.20373.pdf

Abstract:
We introduce CARMA, a system for situational grounding in human-robot group interactions. Effective collaboration in such group settings requires situational awareness based on a consistent representation of present persons and objects coupled with an episodic abstraction of events regarding actors and manipulated objects. This calls for a clear and consistent assignment of instances, ensuring that robots correctly recognize and track actors, objects, and their interactions over time. To achieve this, CARMA uniquely identifies physical instances of such entities in the real world and organizes them into grounded triplets of actors, objects, and actions. To validate our approach, we conducted three experiments, where multiple humans and a robot interact: collaborative pouring, handovers, and sorting. These scenarios allow the assessment of the system's capabilities as to role distinction, multi-actor awareness, and consistent instance identification. Our experiments demonstrate that the system can reliably generate accurate actor-action-object triplets, providing a structured and robust foundation for applications requiring spatiotemporal reasoning and situated decision-making in collaborative settings.

Paperid: 4512, https://arxiv.org/pdf/2506.20094.pdf

Abstract:
AI inference at the edge is becoming increasingly common for low-latency services. However, edge environments are power- and resource-constrained, and susceptible to failures. Conventional failure resilience approaches, such as cloud failover or compressed backups, often compromise latency or accuracy, limiting their effectiveness for critical edge inference services. In this paper, we propose Multi-Level Ensemble Learning (MEL), a new framework for resilient edge inference that simultaneously trains multiple lightweight backup models capable of operating collaboratively, refining each other when multiple servers are available, and independently under failures while maintaining good accuracy. Specifically, we formulate our approach as a multi-objective optimization problem with a loss formulation that inherently encourages diversity among individual models to promote mutually refining representations, while ensuring each model maintains good standalone performance. Empirical evaluations across vision, language, and audio datasets show that MEL provides performance comparable to original architectures while also providing fault tolerance and deployment flexibility across edge platforms. Our results show that our ensemble model, sized at 40\% of the original model, achieves similar performance, while preserving 95.6\% of ensemble accuracy in the case of failures when trained using MEL.

Paperid: 4513, https://arxiv.org/pdf/2506.19072.pdf

Abstract:
Improving the visual understanding ability of vision-language models (VLMs) is crucial for enhancing their performance across various tasks. While using multiple pretrained visual experts has shown great promise, it often incurs significant computational costs during training and inference. To address this challenge, we propose HAWAII, a novel framework that distills knowledge from multiple visual experts into a single vision encoder, enabling it to inherit the complementary strengths of several experts with minimal computational overhead. To mitigate conflicts among different teachers and switch between different teacher-specific knowledge, instead of using a fixed set of adapters for multiple teachers, we propose to use teacher-specific Low-Rank Adaptation (LoRA) adapters with a corresponding router. Each adapter is aligned with a specific teacher, avoiding noisy guidance during distillation. To enable efficient knowledge distillation, we propose fine-grained and coarse-grained distillation. At the fine-grained level, token importance scores are employed to emphasize the most informative tokens from each teacher adaptively. At the coarse-grained level, we summarize the knowledge from multiple teachers and transfer it to the student using a set of general-knowledge LoRA adapters with a router. Extensive experiments on various vision-language tasks demonstrate the superiority of HAWAII, compared to the popular open-source VLMs.

Paperid: 4514, https://arxiv.org/pdf/2506.16398.pdf

Abstract:
Pathology is essential for cancer diagnosis, with multiple instance learning (MIL) widely used for whole slide image (WSI) analysis. WSIs exhibit a natural hierarchy -- patches, regions, and slides -- with distinct semantic associations. While some methods attempt to leverage this hierarchy for improved representation, they predominantly rely on Euclidean embeddings, which struggle to fully capture semantic hierarchies. To address this limitation, we propose HyperPath, a novel method that integrates knowledge from textual descriptions to guide the modeling of semantic hierarchies of WSIs in hyperbolic space, thereby enhancing WSI classification. Our approach adapts both visual and textual features extracted by pathology vision-language foundation models to the hyperbolic space. We design an Angular Modality Alignment Loss to ensure robust cross-modal alignment, while a Semantic Hierarchy Consistency Loss further refines feature hierarchies through entailment and contradiction relationships and thus enhance semantic coherence. The classification is performed with geodesic distance, which measures the similarity between entities in the hyperbolic semantic hierarchy. This eliminates the need for linear classifiers and enables a geometry-aware approach to WSI analysis. Extensive experiments show that our method achieves superior performance across tasks compared to existing methods, highlighting the potential of hyperbolic embeddings for WSI analysis.

Paperid: 4515, https://arxiv.org/pdf/2506.14786.pdf

Abstract:
Multimodal time series forecasting is foundational in various fields, such as utilizing satellite imagery and numerical data for predicting typhoons in climate science. However, existing multimodal approaches primarily focus on utilizing text data to help time series forecasting, leaving the visual data in existing time series datasets untouched. Furthermore, it is challenging for models to effectively capture the physical information embedded in visual data, such as satellite imagery's temporal and geospatial context, which extends beyond images themselves. To address this gap, we propose physics-informed positional encoding (PIPE), a lightweight method that embeds physical information into vision language models (VLMs). PIPE introduces two key innovations: (1) a physics-informed positional indexing scheme for mapping physics to positional IDs, and (2) a variant-frequency positional encoding mechanism for encoding frequency information of physical variables and sequential order of tokens within the embedding space. By preserving both the physical information and sequential order information, PIPE significantly improves multimodal alignment and forecasting accuracy. Through the experiments on the most representative and the largest open-sourced satellite image dataset, PIPE achieves state-of-the-art performance in both deep learning forecasting and climate domain methods, demonstrating superiority across benchmarks, including a 12% improvement in typhoon intensity forecasting over prior works. Our code is provided in the supplementary material.

Paperid: 4516, https://arxiv.org/pdf/2506.14674.pdf

Abstract:
Previous methods for image geo-localization have typically treated the task as either classification or retrieval, often relying on black-box decisions that lack interpretability. The rise of large vision-language models (LVLMs) has enabled a rethinking of geo-localization as a reasoning-driven task grounded in visual cues. However, two major challenges persist. On the data side, existing reasoning-focused datasets are primarily based on street-view imagery, offering limited scene diversity and constrained viewpoints. On the modeling side, current approaches predominantly rely on supervised fine-tuning, which yields only marginal improvements in reasoning capabilities. To address these challenges, we propose a novel pipeline that constructs a reasoning-oriented geo-localization dataset, MP16-Reason, using diverse social media images. We introduce GLOBE, Group-relative policy optimization for Locatability assessment and Optimized visual-clue reasoning, yielding Bi-objective geo-Enhancement for the VLM in recognition and reasoning. GLOBE incorporates task-specific rewards that jointly enhance locatability assessment, visual clue reasoning, and geolocation accuracy. Both qualitative and quantitative results demonstrate that GLOBE outperforms state-of-the-art open-source LVLMs on geo-localization tasks, particularly in diverse visual scenes, while also generating more insightful and interpretable reasoning trajectories.

Paperid: 4517, https://arxiv.org/pdf/2506.07963.pdf

Abstract:
Building upon large language models (LLMs), recent large multimodal models (LMMs) unify cross-model understanding and generation into a single framework. However, LMMs still struggle to achieve accurate vision-language alignment, prone to generating text responses contradicting the visual input or failing to follow the text-to-image prompts. Current solutions require external supervision (e.g., human feedback or reward models) and only address unidirectional tasks-either understanding or generation. In this work, based on the observation that understanding and generation are naturally inverse dual tasks, we propose \textbf{SUDER} (\textbf{S}elf-improving \textbf{U}nified LMMs with \textbf{D}ual s\textbf{E}lf-\textbf{R}ewards), a framework reinforcing the understanding and generation capabilities of LMMs with a self-supervised dual reward mechanism. SUDER leverages the inherent duality between understanding and generation tasks to provide self-supervised optimization signals for each other. Specifically, we sample multiple outputs for a given input in one task domain, then reverse the input-output pairs to compute the dual likelihood within the model as self-rewards for optimization. Extensive experimental results on visual understanding and generation benchmarks demonstrate that our method can effectively enhance the performance of the model without any external supervision, especially achieving remarkable improvements in text-to-image tasks.

Paperid: 4518, https://arxiv.org/pdf/2506.07905.pdf

Abstract:
Building on the success of text-based reasoning models like DeepSeek-R1, extending these capabilities to multimodal reasoning holds great promise. While recent works have attempted to adapt DeepSeek-R1-style reinforcement learning (RL) training paradigms to multimodal large language models (MLLM), focusing on domain-specific tasks like math and visual perception, a critical question remains: How can we achieve the general-purpose visual-language reasoning through RL? To address this challenge, we make three key efforts: (1) A novel Scalable Multimodal QA Synthesis pipeline that autonomously generates context-aware, reasoning-centric question-answer (QA) pairs directly from the given images. (2) The open-source WeThink dataset containing over 120K multimodal QA pairs with annotated reasoning paths, curated from 18 diverse dataset sources and covering various question domains. (3) A comprehensive exploration of RL on our dataset, incorporating a hybrid reward mechanism that combines rule-based verification with model-based assessment to optimize RL training efficiency across various task domains. Across 14 diverse MLLM benchmarks, we demonstrate that our WeThink dataset significantly enhances performance, from mathematical reasoning to diverse general multimodal tasks. Moreover, we show that our automated data pipeline can continuously increase data diversity to further improve model performance.

Paperid: 4519, https://arxiv.org/pdf/2506.05146.pdf

Abstract:
While Vision-Language Models (VLMs) have achieved competitive performance in various tasks, their comprehension of the underlying structure and semantics of a scene remains understudied. To investigate the understanding of VLMs, we study their capability regarding object properties and relations in a controlled and interpretable manner. To this scope, we introduce CIVET, a novel and extensible framework for systematiC evaluatIon Via controllEd sTimuli. CIVET addresses the lack of standardized systematic evaluation for assessing VLMs' understanding, enabling researchers to test hypotheses with statistical rigor. With CIVET, we evaluate five state-of-the-art VLMs on exhaustive sets of stimuli, free from annotation noise, dataset-specific biases, and uncontrolled scene complexity. Our findings reveal that 1) current VLMs can accurately recognize only a limited set of basic object properties; 2) their performance heavily depends on the position of the object in the scene; 3) they struggle to understand basic relations among objects. Furthermore, a comparative evaluation with human annotators reveals that VLMs still fall short of achieving human-level accuracy.

Paperid: 4520, https://arxiv.org/pdf/2506.00765.pdf

Abstract:
Accurate house-price forecasting is essential for investors, planners, and researchers. However, reproducible benchmarks with sufficient spatiotemporal depth and contextual richness for long horizon prediction remain scarce. To address this, we introduce HouseTS a large scale, multimodal dataset covering monthly house prices from March 2012 to December 2023 across 6,000 ZIP codes in 30 major U.S. metropolitan areas. The dataset includes over 890K records, enriched with points of Interest (POI), socioeconomic indicators, and detailed real estate metrics. To establish standardized performance baselines, we evaluate 14 models, spanning classical statistical approaches, deep neural networks (DNNs), and pretrained time-series foundation models. We further demonstrate the value of HouseTS in a multimodal case study, where a vision language model extracts structured textual descriptions of geographic change from time stamped satellite imagery. This enables interpretable, grounded insights into urban evolution. HouseTS is hosted on Kaggle, while all preprocessing pipelines, benchmark code, and documentation are openly maintained on GitHub to ensure full reproducibility and easy adoption.

Paperid: 4521, https://arxiv.org/pdf/2505.24120.pdf

Abstract:
Vision-Language Models (VLMs) have demonstrated remarkable progress in multimodal understanding, yet their capabilities for scientific reasoning remain inadequately assessed. Current multimodal benchmarks predominantly evaluate generic image comprehension or text-driven reasoning, lacking authentic scientific contexts that require domain-specific knowledge integration with visual evidence analysis. To fill this gap, we present CSVQA, a diagnostic multimodal benchmark specifically designed for evaluating scientific reasoning through domain-grounded visual question answering. Our benchmark features 1,378 carefully constructed question-answer pairs spanning diverse STEM disciplines, each demanding domain knowledge, integration of visual evidence, and higher-order reasoning. Compared to prior multimodal benchmarks, CSVQA places greater emphasis on real-world scientific content and complex reasoning. We additionally propose a rigorous evaluation protocol to systematically assess whether model predictions are substantiated by valid intermediate reasoning steps based on curated explanations. Our comprehensive evaluation of 15 VLMs on this benchmark reveals notable performance disparities, as even the top-ranked proprietary model attains only 49.6% accuracy. This empirical evidence underscores the pressing need for advancing scientific reasoning capabilities in VLMs. Our CSVQA is released at https://huggingface.co/datasets/Skywork/CSVQA

Paperid: 4522, https://arxiv.org/pdf/2505.23798.pdf

Abstract:
Social bias is a critical issue in large vision-language models (VLMs), where fairness- and ethics-related problems harm certain groups of people in society. It is unknown to what extent VLMs yield social bias in generative responses. In this study, we focus on evaluating and mitigating social bias on both the model's response and probability distribution. To do so, we first evaluate four state-of-the-art VLMs on PAIRS and SocialCounterfactuals datasets with the multiple-choice selection task. Surprisingly, we find that models suffer from generating gender-biased or race-biased responses. We also observe that models are prone to stating their responses are fair, but indeed having mis-calibrated confidence levels towards particular social groups. While investigating why VLMs are unfair in this study, we observe that VLMs' hidden layers exhibit substantial fluctuations in fairness levels. Meanwhile, residuals in each layer show mixed effects on fairness, with some contributing positively while some lead to increased bias. Based on these findings, we propose a post-hoc method for the inference stage to mitigate social bias, which is training-free and model-agnostic. We achieve this by ablating bias-associated residuals while amplifying fairness-associated residuals on model hidden layers during inference. We demonstrate that our post-hoc method outperforms the competing training strategies, helping VLMs have fairer responses and more reliable confidence levels.

Paperid: 4523, https://arxiv.org/pdf/2505.22897.pdf

Abstract:
While bias in large language models (LLMs) is well-studied, similar concerns in vision-language models (VLMs) have received comparatively less attention. Existing VLM bias studies often focus on portrait-style images and gender-occupation associations, overlooking broader and more complex social stereotypes and their implied harm. This work introduces VIGNETTE, a large-scale VQA benchmark with 30M+ images for evaluating bias in VLMs through a question-answering framework spanning four directions: factuality, perception, stereotyping, and decision making. Beyond narrowly-centered studies, we assess how VLMs interpret identities in contextualized settings, revealing how models make trait and capability assumptions and exhibit patterns of discrimination. Drawing from social psychology, we examine how VLMs connect visual identity cues to trait and role-based inferences, encoding social hierarchies, through biased selections. Our findings uncover subtle, multifaceted, and surprising stereotypical patterns, offering insights into how VLMs construct social meaning from inputs.

Paperid: 4524, https://arxiv.org/pdf/2505.21567.pdf

Abstract:
With the development of Embodied Artificial intelligence, the end-to-end control policy such as Vision-Language-Action (VLA) model has become the mainstream. Existing VLA models faces expensive computing/storage cost, which need to be optimized. Quantization is considered as the most effective method which can not only reduce the memory cost but also achieve computation acceleration. However, we find the token alignment of VLA models hinders the application of existing quantization methods. To address this, we proposed an optimized framework called EaqVLA, which apply encoding-aligned quantization to VLA models. Specifically, we propose an complete analysis method to find the misalignment in various granularity. Based on the analysis results, we propose a mixed precision quantization with the awareness of encoding alignment. Experiments shows that the porposed EaqVLA achieves better quantization performance (with the minimal quantization loss for end-to-end action control and xxx times acceleration) than existing quantization methods.

Paperid: 4525, https://arxiv.org/pdf/2505.20937.pdf

Abstract:
In this paper, we present a comprehensive and systematic analysis of vision-language models (VLMs) for disparate meme classification tasks. We introduced a novel approach that generates a VLM-based understanding of meme images and fine-tunes the LLMs on textual understanding of the embedded meme text for improving the performance. Our contributions are threefold: (1) Benchmarking VLMs with diverse prompting strategies purposely to each sub-task; (2) Evaluating LoRA fine-tuning across all VLM components to assess performance gains; and (3) Proposing a novel approach where detailed meme interpretations generated by VLMs are used to train smaller language models (LLMs), significantly improving classification. The strategy of combining VLMs with LLMs improved the baseline performance by 8.34%, 3.52% and 26.24% for sarcasm, offensive and sentiment classification, respectively. Our results reveal the strengths and limitations of VLMs and present a novel strategy for meme understanding.

Paperid: 4526, https://arxiv.org/pdf/2505.12057.pdf

Abstract:
AI-driven models have shown great promise in detecting errors in radiology reports, yet the field lacks a unified benchmark for rigorous evaluation of error detection and further correction. To address this gap, we introduce CorBenchX, a comprehensive suite for automated error detection and correction in chest X-ray reports, designed to advance AI-assisted quality control in clinical practice. We first synthesize a large-scale dataset of 26,326 chest X-ray error reports by injecting clinically common errors via prompting DeepSeek-R1, with each corrupted report paired with its original text, error type, and human-readable description. Leveraging this dataset, we benchmark both open- and closed-source vision-language models,(e.g., InternVL, Qwen-VL, GPT-4o, o4-mini, and Claude-3.7) for error detection and correction under zero-shot prompting. Among these models, o4-mini achieves the best performance, with 50.6 % detection accuracy and correction scores of BLEU 0.853, ROUGE 0.924, BERTScore 0.981, SembScore 0.865, and CheXbertF1 0.954, remaining below clinical-level accuracy, highlighting the challenge of precise report correction. To advance the state of the art, we propose a multi-step reinforcement learning (MSRL) framework that optimizes a multi-objective reward combining format compliance, error-type accuracy, and BLEU similarity. We apply MSRL to QwenVL2.5-7B, the top open-source model in our benchmark, achieving an improvement of 38.3% in single-error detection precision and 5.2% in single-error correction over the zero-shot baseline.

Paperid: 4527, https://arxiv.org/pdf/2505.03460.pdf

Abstract:
The growing demand for intelligent logistics, particularly fine-grained terminal delivery, underscores the need for autonomous UAV (Unmanned Aerial Vehicle)-based delivery systems. However, most existing last-mile delivery studies rely on ground robots, while current UAV-based Vision-Language Navigation (VLN) tasks primarily focus on coarse-grained, long-range goals, making them unsuitable for precise terminal delivery. To bridge this gap, we propose LogisticsVLN, a scalable aerial delivery system built on multimodal large language models (MLLMs) for autonomous terminal delivery. LogisticsVLN integrates lightweight Large Language Models (LLMs) and Visual-Language Models (VLMs) in a modular pipeline for request understanding, floor localization, object detection, and action-decision making. To support research and evaluation in this new setting, we construct the Vision-Language Delivery (VLD) dataset within the CARLA simulator. Experimental results on the VLD dataset showcase the feasibility of the LogisticsVLN system. In addition, we conduct subtask-level evaluations of each module of our system, offering valuable insights for improving the robustness and real-world deployment of foundation model-based vision-language delivery systems.

Paperid: 4528, https://arxiv.org/pdf/2504.21850.pdf

Abstract:
Multimodal Large Language Models (MLLMs) excel at simple vision-language tasks but struggle when faced with complex tasks that require multiple capabilities, such as simultaneously recognizing objects, counting them, and understanding their spatial relationships. This might be partially the result of the fact that Visual Instruction Tuning (VIT), a critical training step for MLLMs, has traditionally focused on scaling data volume, but not the compositional complexity of training examples. We propose COMPACT (COMPositional Atomic-to-complex visual Capability Tuning), which generates a training dataset explicitly controlling for the compositional complexity of the training examples. The data from COMPACT allows MLLMs to train on combinations of atomic capabilities to learn complex capabilities more efficiently. Across all benchmarks, COMPACT achieves comparable performance to the LLaVA-665k VIT while using less than 10% of its data budget, and even outperforms it on several, especially those involving complex multi-capability tasks. For example, COMPACT achieves substantial 83.3% improvement on MMStar and 94.0% improvement on MM-Vet compared to the full-scale VIT on particularly complex questions that require four or more atomic capabilities. COMPACT offers a scalable, data-efficient, visual compositional tuning recipe to improve on complex visual-language tasks.

Paperid: 4529, https://arxiv.org/pdf/2504.17395.pdf

Abstract:
Open-world object counting leverages the robust text-image alignment of pre-trained vision-language models (VLMs) to enable counting of arbitrary categories in images specified by textual queries. However, widely adopted naive fine-tuning strategies concentrate exclusively on text-image consistency for categories contained in training, which leads to limited generalizability for unseen categories. In this work, we propose a plug-and-play Semantic-Driven Visual Prompt Tuning framework (SDVPT) that transfers knowledge from the training set to unseen categories with minimal overhead in parameters and inference time. First, we introduce a two-stage visual prompt learning strategy composed of Category-Specific Prompt Initialization (CSPI) and Topology-Guided Prompt Refinement (TGPR). The CSPI generates category-specific visual prompts, and then TGPR distills latent structural patterns from the VLM's text encoder to refine these prompts. During inference, we dynamically synthesize the visual prompts for unseen categories based on the semantic correlation between unseen and training categories, facilitating robust text-image alignment for unseen categories. Extensive experiments integrating SDVPT with all available open-world object counting models demonstrate its effectiveness and adaptability across three widely used datasets: FSC-147, CARPK, and PUCPR+.

Paperid: 4530, https://arxiv.org/pdf/2504.17353.pdf

Abstract:
Mutual Reinforcement Effect (MRE) is an emerging subfield at the intersection of information extraction and model interpretability. MRE aims to leverage the mutual understanding between tasks of different granularities, enhancing the performance of both coarse-grained and fine-grained tasks through joint modeling. While MRE has been explored and validated in the textual domain, its applicability to visual and multimodal domains remains unexplored. In this work, we extend MRE to the multimodal information extraction domain for the first time. Specifically, we introduce a new task: Multimodal Mutual Reinforcement Effect (M-MRE), and construct a corresponding dataset to support this task. To address the challenges posed by M-MRE, we further propose a Prompt Format Adapter (PFA) that is fully compatible with various Large Vision-Language Models (LVLMs). Experimental results demonstrate that MRE can also be observed in the M-MRE task, a multimodal text-image understanding scenario. This provides strong evidence that MRE facilitates mutual gains across three interrelated tasks, confirming its generalizability beyond the textual domain.

Paperid: 4531, https://arxiv.org/pdf/2504.14391.pdf

Abstract:
Publicly available biomedical videos, such as those on YouTube, serve as valuable educational resources for medical students. Unlike standard machine learning datasets, these videos are designed for human learners, often mixing medical imagery with narration, explanatory diagrams, and contextual framing. In this work, we investigate whether such pedagogically rich, yet non-standardized and heterogeneous videos can effectively teach general-domain vision-language models biomedical knowledge. To this end, we introduce OpenBiomedVi, a biomedical video instruction tuning dataset comprising 1031 hours of video-caption and Q/A pairs, curated through a multi-step human-in-the-loop pipeline. Diverse biomedical video datasets are rare, and OpenBiomedVid fills an important gap by providing instruction-style supervision grounded in real-world educational content. Surprisingly, despite the informal and heterogeneous nature of these videos, the fine-tuned Qwen-2-VL models exhibit substantial performance improvements across most benchmarks. The 2B model achieves gains of 98.7% on video tasks, 71.2% on image tasks, and 0.2% on text tasks. The 7B model shows improvements of 37.09% on video and 11.2% on image tasks, with a slight degradation of 2.7% on text tasks compared to their respective base models. To address the lack of standardized biomedical video evaluation datasets, we also introduce two new expert curated benchmarks, MIMICEchoQA and SurgeryVideoQA. On these benchmarks, the 2B model achieves gains of 99.1% and 98.1%, while the 7B model shows gains of 22.5% and 52.1%, respectively, demonstrating the models' ability to generalize and perform biomedical video understanding on cleaner and more standardized datasets than those seen during training. These results suggest that educational videos created for human learning offer a surprisingly effective training signal for biomedical VLMs.

Paperid: 4532, https://arxiv.org/pdf/2504.04653.pdf

Abstract:
Redundancy of visual tokens in multi-modal large language models (MLLMs) significantly reduces their computational efficiency. Recent approaches, such as resamplers and summarizers, have sought to reduce the number of visual tokens, but at the cost of visual reasoning ability. To address this, we propose LEO-MINI, a novel MLLM that significantly reduces the number of visual tokens and simultaneously boosts visual reasoning capabilities. For efficiency, LEO-MINI incorporates CoTR, a novel token reduction module to consolidate a large number of visual tokens into a smaller set of tokens, using the similarity between visual tokens, text tokens, and a compact learnable query. For effectiveness, to scale up the model's ability with minimal computational overhead, LEO-MINI employs MMoE, a novel mixture of multi-modal experts module. MMOE employs a set of LoRA experts with a novel router to switch between them based on the input text and visual tokens instead of only using the input hidden state. MMoE also includes a general LoRA expert that is always activated to learn general knowledge for LLM reasoning. For extracting richer visual features, MMOE employs a set of vision experts trained on diverse domain-specific data. To demonstrate LEO-MINI's improved efficiency and performance, we evaluate it against existing efficient MLLMs on various benchmark vision-language tasks.

Paperid: 4533, https://arxiv.org/pdf/2504.02866.pdf

Abstract:
Building properties, such as height, usage, and material composition, play a crucial role in spatial data infrastructures, supporting applications such as energy simulation, risk assessment, and environmental modeling. Despite their importance, comprehensive and high-quality building attribute data remain scarce in many urban areas. Recent advances have enabled the extraction and tagging of objective building attributes using remote sensing and street-level imagery. However, establishing a method and pipeline that integrates diverse open datasets, acquires holistic building imagery at scale, and infers comprehensive building attributes remains a significant challenge. Among the first, this study bridges the gaps by introducing OpenFACADES, an open framework that leverages multimodal crowdsourced data to enrich building profiles with both objective attributes and semantic descriptors through multimodal large language models. Our methodology proceeds in three major steps. First, we integrate street-level image metadata from Mapillary with OpenStreetMap geometries via isovist analysis, effectively identifying images that provide suitable vantage points for observing target buildings. Second, we automate the detection of building facades in panoramic imagery and tailor a reprojection approach to convert objects into holistic perspective views that approximate real-world observation. Third, we introduce an innovative approach that harnesses and systematically investigates the capabilities of open-source large vision-language models (VLMs) for multi-attribute prediction and open-vocabulary captioning in building-level analytics, leveraging a globally sourced dataset of 30,180 labeled images from seven cities. Evaluation shows that fine-tuned VLM excel in multi-attribute inference, outperforming single-attribute computer vision models and zero-shot ChatGPT-4o.

Paperid: 4534, https://arxiv.org/pdf/2504.00185.pdf

Abstract:
We study the problem of building a visual concept library for visual recognition. Building effective visual concept libraries is challenging, as manual definition is labor-intensive, while relying solely on LLMs for concept generation can result in concepts that lack discriminative power or fail to account for the complex interactions between them. Our approach, ESCHER, takes a library learning perspective to iteratively discover and improve visual concepts. ESCHER uses a vision-language model (VLM) as a critic to iteratively refine the concept library, including accounting for interactions between concepts and how they affect downstream classifiers. By leveraging the in-context learning abilities of LLMs and the history of performance using various concepts, ESCHER dynamically improves its concept generation strategy based on the VLM critic's feedback. Finally, ESCHER does not require any human annotations, and is thus an automated plug-and-play framework. We empirically demonstrate the ability of ESCHER to learn a concept library for zero-shot, few-shot, and fine-tuning visual classification tasks. This work represents, to our knowledge, the first application of concept library learning to real-world visual tasks.

Paperid: 4535, https://arxiv.org/pdf/2503.21505.pdf

Abstract:
Existing benchmarks for Vision-Language Model (VLM) on autonomous driving (AD) primarily assess interpretability through open-form visual question answering (QA) within coarse-grained tasks, which remain insufficient to assess capabilities in complex driving scenarios. To this end, we introduce $\textbf{VLADBench}$, a challenging and fine-grained dataset featuring close-form QAs that progress from static foundational knowledge and elements to advanced reasoning for dynamic on-road situations. The elaborate $\textbf{VLADBench}$ spans 5 key domains: Traffic Knowledge Understanding, General Element Recognition, Traffic Graph Generation, Target Attribute Comprehension, and Ego Decision-Making and Planning. These domains are further broken down into 11 secondary aspects and 29 tertiary tasks for a granular evaluation. A thorough assessment of general and domain-specific (DS) VLMs on this benchmark reveals both their strengths and critical limitations in AD contexts. To further exploit the cognitive and reasoning interactions among the 5 domains for AD understanding, we start from a small-scale VLM and train the DS models on individual domain datasets (collected from 1.4M DS QAs across public sources). The experimental results demonstrate that the proposed benchmark provides a crucial step toward a more comprehensive assessment of VLMs in AD, paving the way for the development of more cognitively sophisticated and reasoning-capable AD systems.

Paperid: 4536, https://arxiv.org/pdf/2503.19910.pdf

Abstract:
Composed Image Retrieval (CIR) is a complex task that aims to retrieve images based on a multimodal query. Typical training data consists of triplets containing a reference image, a textual description of desired modifications, and the target image, which are expensive and time-consuming to acquire. The scarcity of CIR datasets has led to zero-shot approaches utilizing synthetic triplets or leveraging vision-language models (VLMs) with ubiquitous web-crawled image-caption pairs. However, these methods have significant limitations: synthetic triplets suffer from limited scale, lack of diversity, and unnatural modification text, while image-caption pairs hinder joint embedding learning of the multimodal query due to the absence of triplet data. Moreover, existing approaches struggle with complex and nuanced modification texts that demand sophisticated fusion and understanding of vision and language modalities. We present CoLLM, a one-stop framework that effectively addresses these limitations. Our approach generates triplets on-the-fly from image-caption pairs, enabling supervised training without manual annotation. We leverage Large Language Models (LLMs) to generate joint embeddings of reference images and modification texts, facilitating deeper multimodal fusion. Additionally, we introduce Multi-Text CIR (MTCIR), a large-scale dataset comprising 3.4M samples, and refine existing CIR benchmarks (CIRR and Fashion-IQ) to enhance evaluation reliability. Experimental results demonstrate that CoLLM achieves state-of-the-art performance across multiple CIR benchmarks and settings. MTCIR yields competitive results, with up to 15% performance improvement. Our refined benchmarks provide more reliable evaluation metrics for CIR models, contributing to the advancement of this important field.

Paperid: 4537, https://arxiv.org/pdf/2503.06435.pdf

Abstract:
Open-vocabulary 3D object detection for autonomous driving aims to detect novel objects beyond the predefined training label sets in point cloud scenes. Existing approaches achieve this by connecting traditional 3D object detectors with vision-language models (VLMs) to regress 3D bounding boxes for novel objects and perform open-vocabulary classification through cross-modal alignment between 3D and 2D features. However, achieving robust cross-modal alignment remains a challenge due to semantic inconsistencies when generating corresponding 3D and 2D feature pairs. To overcome this challenge, we present OV-SCAN, an Open-Vocabulary 3D framework that enforces Semantically Consistent Alignment for Novel object discovery. OV-SCAN employs two core strategies: discovering precise 3D annotations and filtering out low-quality or corrupted alignment pairs (arising from 3D annotation, occlusion-induced, or resolution-induced noise). Extensive experiments on the nuScenes dataset demonstrate that OV-SCAN achieves state-of-the-art performance.

Paperid: 4538, https://arxiv.org/pdf/2503.00743.pdf

Abstract:
Vision-Language Models (VLMs) have demonstrated great potential in interpreting remote sensing (RS) images through language-guided semantic. However, the effectiveness of these VLMs critically depends on high-quality image-text training data that captures rich semantic relationships between visual content and language descriptions. Unlike natural images, RS lacks large-scale interleaved image-text pairs from web data, making data collection challenging. While current approaches rely primarily on rule-based methods or flagship VLMs for data synthesis, a systematic framework for automated quality assessment of such synthetically generated RS vision-language data is notably absent. To fill this gap, we propose a novel score model trained on large-scale RS vision-language preference data for automated quality assessment. Our empirical results demonstrate that fine-tuning CLIP or advanced VLMs (e.g., Qwen2-VL) with the top 30% of data ranked by our score model achieves superior accuracy compared to both full-data fine-tuning and CLIP-score-based ranking approaches. Furthermore, we demonstrate applications of our scoring model for reinforcement learning (RL) training and best-of-N (BoN) test-time scaling, enabling significant improvements in VLM performance for RS tasks. Our code, model, and dataset are publicly available

Paperid: 4539, https://arxiv.org/pdf/2502.13766.pdf

Abstract:
Large Vision-Language Models (LVLMs) have recently gained attention due to their distinctive performance and broad applicability. While it has been previously shown that their efficacy in usage scenarios involving non-Western contexts falls short, existing studies are limited in scope, covering just a narrow range of cultures, focusing exclusively on a small number of cultural aspects, or evaluating a limited selection of models on a single task only. Towards globally inclusive LVLM research, we introduce GIMMICK, an extensive multimodal benchmark designed to assess a broad spectrum of cultural knowledge across 144 countries representing six global macro-regions. GIMMICK comprises six tasks built upon three new datasets that span 728 unique cultural events or facets on which we evaluated 20 LVLMs and 11 LLMs, including five proprietary and 26 open-weight models of all sizes. We systematically examine (1) regional cultural biases, (2) the influence of model size, (3) input modalities, and (4) external cues. Our analyses reveal strong biases toward Western cultures across models and tasks and highlight strong correlations between model size and performance, as well as the effectiveness of multimodal input and external geographic cues. We further find that models have more knowledge of tangible than intangible aspects (e.g., food vs. rituals) and that they excel in recognizing broad cultural origins but struggle with a more nuanced understanding.

Paperid: 4540, https://arxiv.org/pdf/2502.11748.pdf

Abstract:
This work introduces ILIAS, a new test dataset for Instance-Level Image retrieval At Scale. It is designed to evaluate the ability of current and future foundation models and retrieval techniques to recognize particular objects. The key benefits over existing datasets include large scale, domain diversity, accurate ground truth, and a performance that is far from saturated. ILIAS includes query and positive images for 1,000 object instances, manually collected to capture challenging conditions and diverse domains. Large-scale retrieval is conducted against 100 million distractor images from YFCC100M. To avoid false negatives without extra annotation effort, we include only query objects confirmed to have emerged after 2014, i.e. the compilation date of YFCC100M. An extensive benchmarking is performed with the following observations: i) models fine-tuned on specific domains, such as landmarks or products, excel in that domain but fail on ILIAS ii) learning a linear adaptation layer using multi-domain class supervision results in performance improvements, especially for vision-language models iii) local descriptors in retrieval re-ranking are still a key ingredient, especially in the presence of severe background clutter iv) the text-to-image performance of the vision-language foundation models is surprisingly close to the corresponding image-to-image case. website: https://vrg.fel.cvut.cz/ilias/

Paperid: 4541, https://arxiv.org/pdf/2502.00662.pdf

Abstract:
Existing vision-language model (VLM)-based methods for out-of-distribution (OOD) detection typically rely on similarity scores between input images and in-distribution (ID) text prototypes. However, the modality gap between image and text often results in high false positive rates, as OOD samples can exhibit high similarity to ID text prototypes. To mitigate the impact of this modality gap, we propose incorporating ID image prototypes along with ID text prototypes. We present theoretical analysis and empirical evidence indicating that this approach enhances VLM-based OOD detection performance without any additional training. To further reduce the gap between image and text, we introduce a novel few-shot tuning framework, SUPREME, comprising biased prompts generation (BPG) and image-text consistency (ITC) modules. BPG enhances image-text fusion and improves generalization by conditioning ID text prototypes on the Gaussian-based estimated image domain bias; ITC reduces the modality gap by minimizing intra- and inter-modal distances. Moreover, inspired by our theoretical and empirical findings, we introduce a novel OOD score $S_{\textit{GMP}}$, leveraging uni- and cross-modal similarities. Finally, we present extensive experiments to demonstrate that SUPREME consistently outperforms existing VLM-based OOD detection methods.

Paperid: 4542, https://arxiv.org/pdf/2502.00241.pdf

Abstract:
Incorporating multiple modalities into large language models (LLMs) is a powerful way to enhance their understanding of non-textual data, enabling them to perform multimodal tasks. Vision language models (VLMs) form the fastest growing category of multimodal models because of their many practical use cases, including in healthcare, robotics, and accessibility. Unfortunately, even though different VLMs in the literature demonstrate impressive visual capabilities in different benchmarks, they are handcrafted by human experts; there is no automated framework to create task-specific multimodal models. We introduce Mordal, an automated multimodal model search framework that efficiently finds the best VLM for a user-defined task without manual intervention. Mordal achieves this both by reducing the number of candidates to consider during the search process and by minimizing the time required to evaluate each remaining candidate. Our evaluation shows that Mordal can find the best VLM for a given problem using up to $8.9\times$--$11.6\times$ lower GPU hours than grid search. In the process of our evaluation, we have also discovered new VLMs that outperform their state-of-the-art counterparts.

Paperid: 4543, https://arxiv.org/pdf/2501.15688.pdf

Abstract:
Multimodal knowledge graph completion (MMKGC) aims to predict missing links in multimodal knowledge graphs (MMKGs) by leveraging information from various modalities alongside structural data. Existing MMKGC approaches primarily extend traditional knowledge graph embedding (KGE) models, which often require creating an embedding for every entity. This results in large model sizes and inefficiencies in integrating multimodal information, particularly for real-world graphs. Meanwhile, Transformer-based models have demonstrated competitive performance in knowledge graph completion (KGC). However, their focus on single-modal knowledge limits their capacity to utilize cross-modal information. Recently, Large vision-language models (VLMs) have shown potential in cross-modal tasks but are constrained by the high cost of training. In this work, we propose a novel approach that integrates Transformer-based KGE models with cross-modal context generated by pre-trained VLMs, thereby extending their applicability to MMKGC. Specifically, we employ a pre-trained VLM to transform relevant visual information from entities and their neighbors into textual sequences. We then frame KGC as a sequence-to-sequence task, fine-tuning the model with the generated cross-modal context. This simple yet effective method significantly reduces model size compared to traditional KGE approaches while achieving competitive performance across multiple large-scale datasets with minimal hyperparameter tuning.

Paperid: 4544, https://arxiv.org/pdf/2501.09307.pdf

Abstract:
As robotic technology rapidly develops, robots are being employed in an increasing number of fields. However, due to the complexity of deployment environments or the prevalence of ambiguous-condition objects, the practical application of robotics still faces many challenges, leading to frequent errors. Traditional methods and some LLM-based approaches, although improved, still require substantial human intervention and struggle with autonomous error correction in complex scenarios. In this work, we propose RoboReflect, a novel framework leveraging large vision-language models (LVLMs) to enable self-reflection and autonomous error correction in robotic grasping tasks. RoboReflect allows robots to automatically adjust their strategies based on unsuccessful attempts until successful execution is achieved. The corrected strategies are saved in the memory for future task reference. We evaluate RoboReflect through extensive testing on eight common objects prone to ambiguous conditions of three categories. Our results demonstrate that RoboReflect not only outperforms existing grasp pose estimation methods like AnyGrasp and high-level action planning techniques ReKep with GPT-4V but also significantly enhances the robot's capability to adapt and correct errors independently. These findings underscore the critical importance of autonomous self-reflection in robotic systems while effectively addressing the challenges posed by ambiguous-condition environments.

Paperid: 4545, https://arxiv.org/pdf/2501.06986.pdf

Abstract:
Enhanced visual understanding serves as a cornerstone for multimodal large language models (MLLMs). Recent hybrid MLLMs incorporate a mixture of vision experts to address the limitations of using a single vision encoder and excessively long visual tokens. Despite the progress of these MLLMs, a research gap remains in effectively integrating diverse vision encoders. This work explores fusion strategies of visual tokens for hybrid MLLMs, leading to the design of LEO, a novel MLLM with a dual-branch vision encoder framework that incorporates a post-adaptation fusion strategy and adaptive tiling: for each segmented tile of the input images, LEO sequentially interleaves the visual tokens from its two vision encoders. Extensive evaluation across 13 vision-language benchmarks reveals that LEO outperforms state-of-the-art open-source MLLMs and hybrid MLLMs on the majority of tasks. Furthermore, we show that LEO can be adapted to the specialized domain of autonomous driving without altering the model architecture or training recipe, achieving competitive performance compared to existing baselines. The code and model will be publicly available.

Paperid: 4546, https://arxiv.org/pdf/2501.05122.pdf

Abstract:
Most Large Vision-Language Models (LVLMs) to date are trained predominantly on English data, which makes them struggle to understand non-English input and fail to generate output in the desired target language. Existing efforts mitigate these issues by adding multilingual training data, but do so in a largely ad-hoc manner, lacking insight into how different training mixes tip the scale for different groups of languages. In this work, we present a comprehensive investigation into the training strategies for massively multilingual LVLMs. First, we conduct a series of multi-stage experiments spanning 13 downstream vision-language tasks and 43 languages, systematically examining: (1) the number of training languages that can be included without degrading English performance and (2) optimal language distributions of pre-training as well as (3) instruction-tuning data. Further, we (4) investigate how to improve multilingual text-in-image understanding, and introduce a new benchmark for the task. Surprisingly, our analysis reveals that one can (i) include as many as 100 training languages simultaneously (ii) with as little as 25-50\% of non-English data, to greatly improve multilingual performance while retaining strong English performance. We further find that (iii) including non-English OCR data in pre-training and instruction-tuning is paramount for improving multilingual text-in-image understanding. Finally, we put all our findings together and train Centurio, a 100-language LVLM, offering state-of-the-art performance in an evaluation covering 14 tasks and 56 languages.

Paperid: 4547, https://arxiv.org/pdf/2501.00654.pdf

Abstract:
Training vision-language models via instruction tuning often relies on large mixtures of data spanning diverse tasks and domains. However, these mixtures frequently include redundant information, increasing computational costs without proportional performance gains, necessitating more effective data selection strategies. Existing methods typically rely on task-agnostic heuristics to estimate data importance or focus on optimizing single tasks in isolation, limiting their effectiveness in multitask settings. In this work, we introduce ICONS, a gradient-based Influence CONsensus approach for vision-language data Selection. Our method leverages first-order training dynamics to estimate the influence of individual training examples on validation performance and aggregates these estimates across tasks via majority voting over task-specific influences. This cross-task consensus identifies data points that are consistently valuable across tasks, enabling us to prioritize examples that drive overall performance. The voting-based design further mitigates issues such as score calibration and outlier sensitivity, resulting in robust and scalable data selection for diverse multitask mixtures. With only 20% of the data from LLaVA-665K and Cambrian-7M, our selected subsets retain 98.6% and 98.8% of the performance achieved with full datasets, and can even surpass full data training at a 60% selection ratio on LLaVA-665K. Our approach also generalizes to unseen tasks and architectures, demonstrating strong transfer. We release two compact, high-utility subsets, LLaVA-ICONS-133K and Cambrian-ICONS-1.4M, preserving impactful training examples for efficient and scalable vision-language model development.

Paperid: 4548, https://arxiv.org/pdf/2412.19087.pdf

Abstract:
Soft prompt learning methods are effective for adapting vision-language models (VLMs) to downstream tasks. Nevertheless, empirical evidence reveals a tendency of existing methods that they overfit seen classes and exhibit degraded performance on unseen classes. This limitation is due to the inherent bias in the training data towards the seen classes. To address this issue, we propose a novel soft prompt learning method, named Mixture-of-Prompts Distillation (MoPD), which can effectively transfer useful knowledge from hard prompts manually hand-crafted (a.k.a. teacher prompts) to the learnable soft prompt (a.k.a. student prompt), thereby enhancing the generalization ability of soft prompts on unseen classes. Moreover, the proposed MoPD method utilizes a gating network that learns to select hard prompts used for prompt distillation. Extensive experiments demonstrate that the proposed MoPD method outperforms state-of-the-art baselines especially on on unseen classes.

Paperid: 4549, https://arxiv.org/pdf/2412.14133.pdf

Abstract:
Vision-language models (VLMs) excel at extracting and reasoning about information from images. Yet, their capacity to leverage internal knowledge about specific entities remains underexplored. This work investigates the disparity in model performance when answering factual questions about an entity described in text versus depicted in an image. Our results reveal a significant accuracy drop - reaching 18% for some models - when the entity is presented visually instead of textually. To study this gap we present PopVQA, a dataset which allows separating entity recognition and question answering, and use it to benchmark several models. We hypothesize that this decline arises from limitations in how information flows from image tokens to query tokens. Thus, we use mechanistic interpretability tools to reveal that, although image tokens are preprocessed by the vision encoder, meaningful information flow from these tokens occurs only in the much deeper layers. Furthermore, critical image processing happens in the language model's middle layers, allowing few layers for consecutive reasoning, highlighting a potential inefficiency in how the model utilizes its layers for reasoning. These insights shed light on the internal mechanics of VLMs and offer pathways for enhancing their reasoning capabilities. PopVQA can be found at https://huggingface.co/datasets/idoco/PopVQA.

Paperid: 4550, https://arxiv.org/pdf/2412.14058.pdf

Abstract:
Foundation Vision Language Models (VLMs) exhibit strong capabilities in multi-modal representation learning, comprehension, and reasoning. By injecting action components into the VLMs, Vision-Language-Action Models (VLAs) can be naturally formed and also show promising performance. Existing work has demonstrated the effectiveness and generalization of VLAs in multiple scenarios and tasks. Nevertheless, the transfer from VLMs to VLAs is not trivial since existing VLAs differ in their backbones, action-prediction formulations, data distributions, and training recipes. This leads to a missing piece for a systematic understanding of the design choices of VLAs. In this work, we disclose the key factors that significantly influence the performance of VLA and focus on answering three essential design choices: which backbone to select, how to formulate the VLA architectures, and when to add cross-embodiment data. The obtained results convince us firmly to explain why we need VLA and develop a new family of VLAs, RoboVLMs, which require very few manual designs and achieve a new state-of-the-art performance in three simulation tasks and real-world experiments. Through our extensive experiments, which include over 8 VLM backbones, 4 policy architectures, and over 600 distinct designed experiments, we provide a detailed guidebook for the future design of VLAs. In addition to the study, the highly flexible RoboVLMs framework, which supports easy integrations of new VLMs and free combinations of various design choices, is made public to facilitate future research. We open-source all details, including codes, models, datasets, and toolkits, along with detailed training and evaluation recipes at: robovlms.github.io.

Paperid: 4551, https://arxiv.org/pdf/2412.13652.pdf

Abstract:
Neural radiance fields are an emerging 3D scene representation and recently even been extended to learn features for scene understanding by distilling open-vocabulary features from vision-language models. However, current method primarily focus on object-centric representations, supporting object segmentation or detection, while understanding semantic relationships between objects remains largely unexplored. To address this gap, we propose RelationField, the first method to extract inter-object relationships directly from neural radiance fields. RelationField represents relationships between objects as pairs of rays within a neural radiance field, effectively extending its formulation to include implicit relationship queries. To teach RelationField complex, open-vocabulary relationships, relationship knowledge is distilled from multi-modal LLMs. To evaluate RelationField, we solve open-vocabulary 3D scene graph generation tasks and relationship-guided instance segmentation, achieving state-of-the-art performance in both tasks. See the project website at https://relationfield.github.io.

Paperid: 4552, https://arxiv.org/pdf/2412.11050.pdf

Abstract:
Understanding and addressing corner cases is essential for ensuring the safety and reliability of autonomous driving systems. Vision-language models (VLMs) play a crucial role in enhancing scenario comprehension, yet they face significant challenges, such as hallucination and insufficient real-world grounding, which compromise their performance in critical driving scenarios. In this work, RAC3, a novel framework designed to enhance the performance of VLMs in corner case comprehension, is proposed. RAC3 integrates a frequency-spatial fusion (FSF) image encoder, a cross-modal alignment training method for embedding models with hard and semi-hard negative mining, and a fast querying and retrieval pipeline based on K-Means clustering and hierarchical navigable small world (HNSW) indexing. A multimodal chain-of-thought (CoT) prompting strategy to guide analogical reasoning and reduce hallucinations during inference is introduced. Moreover, an update mechanism is integrated into RAC3 to ensure continual learning within the framework. Extensive experiments on the CODA and nuScenes datasets demonstrate that RAC3 significantly improves corner case comprehension across multiple downstream tasks. Compared to prior state-of-the-art methods, RAC3 achieves the highest final score of 74.46 on the CODA-LM benchmark and shows consistent performance gains when integrated with end-to-end frameworks like DriveLM. These results demonstrate the effectiveness of retrieval-augmented strategies and cross-modal alignment for safer and more interpretable autonomous driving.

Paperid: 4553, https://arxiv.org/pdf/2412.08907.pdf

Abstract:
Global geolocation, which seeks to predict the geographical location of images captured anywhere in the world, is one of the most challenging tasks in the field of computer vision. In this paper, we introduce an innovative interactive global geolocation assistant named GaGA, built upon the flourishing large vision-language models (LVLMs). GaGA uncovers geographical clues within images and combines them with the extensive world knowledge embedded in LVLMs to determine the geolocations while also providing justifications and explanations for the prediction results. We further designed a novel interactive geolocation method that surpasses traditional static inference approaches. It allows users to intervene, correct, or provide clues for the predictions, making the model more flexible and practical. The development of GaGA relies on the newly proposed Multi-modal Global Geolocation (MG-Geo) dataset, a comprehensive collection of 5 million high-quality image-text pairs. GaGA achieves state-of-the-art performance on the GWS15k dataset, improving accuracy by 4.57% at the country level and 2.92% at the city level, setting a new benchmark. These advancements represent a significant leap forward in developing highly accurate, interactive geolocation systems with global applicability.

Paperid: 4554, https://arxiv.org/pdf/2412.08388.pdf

Abstract:
Vision-based 3D occupancy prediction has become a popular research task due to its versatility and affordability. Nowadays, conventional methods usually project the image-based vision features to 3D space and learn the geometric information through the attention mechanism, enabling the 3D semantic occupancy prediction. However, these works usually face two main challenges: 1) Limited geometric information. Due to the lack of geometric information in the image itself, it is challenging to directly predict 3D space information, especially in large-scale outdoor scenes. 2) Local restricted interaction. Due to the quadratic complexity of the attention mechanism, they often use modified local attention to fuse features, resulting in a restricted fusion. To address these problems, in this paper, we propose a language-assisted 3D semantic occupancy prediction network, named LOMA. In the proposed vision-language framework, we first introduce a VL-aware Scene Generator (VSG) module to generate the 3D language feature of the scene. By leveraging the vision-language model, this module provides implicit geometric knowledge and explicit semantic information from the language. Furthermore, we present a Tri-plane Fusion Mamba (TFM) block to efficiently fuse the 3D language feature and 3D vision feature. The proposed module not only fuses the two features with global modeling but also avoids too much computation costs. Experiments on the SemanticKITTI and SSCBench-KITTI360 datasets show that our algorithm achieves new state-of-the-art performances in both geometric and semantic completion tasks. Our code will be open soon.

Paperid: 4555, https://arxiv.org/pdf/2412.05243.pdf

Abstract:
How well can Multimodal Large Language Models (MLLMs) understand composite images? Composite images (CIs) are synthetic visuals created by merging multiple visual elements, such as charts, posters, or screenshots, rather than being captured directly by a camera. While CIs are prevalent in real-world applications, recent MLLM developments have primarily focused on interpreting natural images (NIs). Our research reveals that current MLLMs face significant challenges in accurately understanding CIs, often struggling to extract information or perform complex reasoning based on these images. We find that existing training data for CIs are mostly formatted for question-answer tasks (e.g., in datasets like ChartQA and ScienceQA), while high-quality image-caption datasets, critical for robust vision-language alignment, are only available for NIs. To bridge this gap, we introduce Composite Captions (CompCap), a flexible framework that leverages Large Language Models (LLMs) and automation tools to synthesize CIs with accurate and detailed captions. Using CompCap, we curate CompCap-118K, a dataset containing 118K image-caption pairs across six CI types. We validate the effectiveness of CompCap-118K by supervised fine-tuning MLLMs of three sizes: xGen-MM-inst.-4B and LLaVA-NeXT-Vicuna-7B/13B. Empirical results show that CompCap-118K significantly enhances MLLMs' understanding of CIs, yielding average gains of 1.7%, 2.0%, and 2.9% across eleven benchmarks, respectively.

Paperid: 4556, https://arxiv.org/pdf/2412.05149.pdf

Abstract:
The BabyLM Challenge is a community effort to close the data-efficiency gap between human and computational language learners. Participants compete to optimize language model training on a fixed language data budget of 100 million words or less. This year, we released improved text corpora, as well as a vision-and-language corpus to facilitate research into cognitively plausible vision language models. Submissions were compared on evaluation tasks targeting grammatical ability, (visual) question answering, pragmatic abilities, and grounding, among other abilities. Participants could submit to a 10M-word text-only track, a 100M-word text-only track, and/or a 100M-word and image multimodal track. From 31 submissions employing diverse methods, a hybrid causal-masked language model architecture outperformed other approaches. No submissions outperformed the baselines in the multimodal track. In follow-up analyses, we found a strong relationship between training FLOPs and average performance across tasks, and that the best-performing submissions proposed changes to the training data, training objective, and model architecture. This year's BabyLM Challenge shows that there is still significant room for innovation in this setting, in particular for image-text modeling, but community-driven research can yield actionable insights about effective strategies for small-scale language modeling.

Paperid: 4557, https://arxiv.org/pdf/2412.00071.pdf

Abstract:
Training large-scale neural networks in vision, and multimodal domains demands substantial memory resources, primarily due to the storage of optimizer states. While LoRA, a popular parameter-efficient method, reduces memory usage, it often suffers from suboptimal performance due to the constraints of low-rank updates. Low-rank gradient projection methods (e.g., GaLore, Flora) reduce optimizer memory by projecting gradients and moment estimates into low-rank spaces via singular value decomposition or random projection. However, they fail to account for inter-projection correlation, causing performance degradation, and their projection strategies often incur high computational costs. In this paper, we present COAP (Correlation-Aware Gradient Projection), a memory-efficient method that minimizes computational overhead while maintaining training performance. Evaluated across various vision, language, and multimodal tasks, COAP outperforms existing methods in both training speed and model performance. For LLaMA-1B, it reduces optimizer memory by 61% with only 2% additional time cost, achieving the same PPL as AdamW. With 8-bit quantization, COAP cuts optimizer memory by 81% and achieves 4x speedup over GaLore for LLaVA-v1.5-7B fine-tuning, while delivering higher accuracy.

Paperid: 4558, https://arxiv.org/pdf/2411.18042.pdf

Abstract:
Multimodal LLMs have advanced vision-language tasks but still struggle with understanding video scenes. To bridge this gap, Video Scene Graph Generation (VidSGG) has emerged to capture multi-object relationships across video frames. However, prior methods rely on pairwise connections, limiting their ability to handle complex multi-object interactions and reasoning. To this end, we propose Multimodal LLMs on a Scene HyperGraph (HyperGLM), promoting reasoning about multi-way interactions and higher-order relationships. Our approach uniquely integrates entity scene graphs, which capture spatial relationships between objects, with a procedural graph that models their causal transitions, forming a unified HyperGraph. Significantly, HyperGLM enables reasoning by injecting this unified HyperGraph into LLMs. Additionally, we introduce a new Video Scene Graph Reasoning (VSGR) dataset featuring 1.9M frames from third-person, egocentric, and drone views and supports five tasks: Scene Graph Generation, Scene Graph Anticipation, Video Question Answering, Video Captioning, and Relation Reasoning. Empirically, HyperGLM consistently outperforms state-of-the-art methods across five tasks, effectively modeling and reasoning complex relationships in diverse video scenes.

Paperid: 4559, https://arxiv.org/pdf/2411.15611.pdf

Abstract:
Humans can visualize new and unknown concepts from their natural language description, based on their experience and previous knowledge. Insipired by this, we present a way to extend this ability to Vision-Language Models (VLMs), teaching them novel concepts by only using a textual description. We refer to this approach as Knowledge Transfer (KT). Our hypothesis is that the knowledge of a pre-trained VLM can be re-used to represent previously unknown concepts. Provided with a textual description of the novel concept, KT works by aligning relevant features of the visual encoder, obtained through model inversion, to its text representation. Differently from approaches relying on visual examples or external generative models, KT transfers knowledge within the same VLM by injecting visual knowledge directly from the text. Through an extensive evaluation on several VLM tasks, including classification, segmentation, image-text retrieval, and captioning, we show that: 1) KT can efficiently introduce new visual concepts from a single textual description; 2) the same principle can be used to refine the representation of existing concepts; and 3) KT significantly improves the performance of zero-shot VLMs.

Paperid: 4560, https://arxiv.org/pdf/2411.03769.pdf

Abstract:
Research in vision and language has made considerable progress thanks to benchmarks such as COCO. COCO captions focused on unambiguous facts in English; ArtEmis introduced subjective emotions and ArtELingo introduced some multilinguality (Chinese and Arabic). However we believe there should be more multilinguality. Hence, we present ArtELingo-28, a vision-language benchmark that spans $\textbf{28}$ languages and encompasses approximately $\textbf{200,000}$ annotations ($\textbf{140}$ annotations per image). Traditionally, vision research focused on unambiguous class labels, whereas ArtELingo-28 emphasizes diversity of opinions over languages and cultures. The challenge is to build machine learning systems that assign emotional captions to images. Baseline results will be presented for three novel conditions: Zero-Shot, Few-Shot and One-vs-All Zero-Shot. We find that cross-lingual transfer is more successful for culturally-related languages. Data and code are provided at www.artelingo.org.

Paperid: 4561, https://arxiv.org/pdf/2411.00855.pdf

Abstract:
Chain-of-thought (CoT) has proven to improve the reasoning capability of large language models (LLMs). However, due to the complexity of multimodal scenarios and the difficulty in collecting high-quality CoT data, CoT reasoning in multimodal LLMs has been largely overlooked. To this end, we propose a simple yet effective self-training framework, R3V, which iteratively enhances the model's Vision-language Reasoning by Reflecting on CoT Rationales. Our framework consists of two interleaved parts: (1) iteratively bootstrapping positive and negative solutions for reasoning datasets, and (2) reflection on rationale for learning from mistakes. Specifically, we introduce the self-refine and self-select losses, enabling the model to refine flawed rationale and derive the correct answer by comparing rationale candidates. Experiments on a wide range of vision-language tasks show that R3V consistently improves multimodal LLM reasoning, achieving a relative improvement of 23 to 60 percent over GPT-distilled baselines. Additionally, our approach supports self-reflection on generated solutions, further boosting performance through test-time computation.

Paperid: 4562, https://arxiv.org/pdf/2410.11783.pdf

Abstract:
This paper introduces a novel probabilistic mapping algorithm, LatentBKI, which enables open-vocabulary mapping with quantifiable uncertainty. Traditionally, semantic mapping algorithms focus on a fixed set of semantic categories which limits their applicability for complex robotic tasks. Vision-Language (VL) models have recently emerged as a technique to jointly model language and visual features in a latent space, enabling semantic recognition beyond a predefined, fixed set of semantic classes. LatentBKI recurrently incorporates neural embeddings from VL models into a voxel map with quantifiable uncertainty, leveraging the spatial correlations of nearby observations through Bayesian Kernel Inference (BKI). LatentBKI is evaluated against similar explicit semantic mapping and VL mapping frameworks on the popular Matterport3D and Semantic KITTI datasets, demonstrating that LatentBKI maintains the probabilistic benefits of continuous mapping with the additional benefit of open-dictionary queries. Real-world experiments demonstrate applicability to challenging indoor environments.

Paperid: 4563, https://arxiv.org/pdf/2409.20548.pdf

Abstract:
Imagine a future when we can Zoom-call a robot to manage household chores remotely. This work takes one step in this direction. Robi Butler is a new household robot assistant that enables seamless multimodal remote interaction. It allows the human user to monitor its environment from a first-person view, issue voice or text commands, and specify target objects through hand-pointing gestures. At its core, a high-level behavior module, powered by Large Language Models (LLMs), interprets multimodal instructions to generate multistep action plans. Each plan consists of open-vocabulary primitives supported by vision-language models, enabling the robot to process both textual and gestural inputs. Zoom provides a convenient interface to implement remote interactions between the human and the robot. The integration of these components allows Robi Butler to ground remote multimodal instructions in real-world home environments in a zero-shot manner. We evaluated the system on various household tasks, demonstrating its ability to execute complex user commands with multimodal inputs. We also conducted a user study to examine how multimodal interaction influences user experiences in remote human-robot interaction. These results suggest that with the advances in robot foundation models, we are moving closer to the reality of remote household robot assistants.

Paperid: 4564, https://arxiv.org/pdf/2409.19696.pdf

Abstract:
Recent research on fine-tuning vision-language models has demonstrated impressive performance in various downstream tasks. However, the challenge of obtaining accurately labeled data in real-world applications poses a significant obstacle during the fine-tuning process. To address this challenge, this paper presents a Denoising Fine-Tuning framework, called DeFT, for adapting vision-language models. DeFT utilizes the robust alignment of textual and visual features pre-trained on millions of auxiliary image-text pairs to sieve out noisy labels. The proposed framework establishes a noisy label detector by learning positive and negative textual prompts for each class. The positive prompt seeks to reveal distinctive features of the class, while the negative prompt serves as a learnable threshold for separating clean and noisy samples. We employ parameter-efficient fine-tuning for the adaptation of a pre-trained visual encoder to promote its alignment with the learned textual prompts. As a general framework, DeFT can seamlessly fine-tune many pre-trained models to downstream tasks by utilizing carefully selected clean samples. Experimental results on seven synthetic and real-world noisy datasets validate the effectiveness of DeFT in both noisy label detection and image classification.

Paperid: 4565, https://arxiv.org/pdf/2409.03272.pdf

Abstract:
The rise of multi-modal large language models(MLLMs) has spurred their applications in autonomous driving. Recent MLLM-based methods perform action by learning a direct mapping from perception to action, neglecting the dynamics of the world and the relations between action and world dynamics. In contrast, human beings possess world model that enables them to simulate the future states based on 3D internal visual representation and plan actions accordingly. To this end, we propose OccLLaMA, an occupancy-language-action generative world model, which uses semantic occupancy as a general visual representation and unifies vision-language-action(VLA) modalities through an autoregressive model. Specifically, we introduce a novel VQVAE-like scene tokenizer to efficiently discretize and reconstruct semantic occupancy scenes, considering its sparsity and classes imbalance. Then, we build a unified multi-modal vocabulary for vision, language and action. Furthermore, we enhance LLM, specifically LLaMA, to perform the next token/scene prediction on the unified vocabulary to complete multiple tasks in autonomous driving. Extensive experiments demonstrate that OccLLaMA achieves competitive performance across multiple tasks, including 4D occupancy forecasting, motion planning, and visual question answering, showcasing its potential as a foundation model in autonomous driving.

Paperid: 4566, https://arxiv.org/pdf/2601.20641.pdf

Abstract:
We investigate whether \emph{LLM-based agents} can develop task-oriented communication protocols that differ from standard natural language in collaborative reasoning tasks. Our focus is on two core properties such task-oriented protocols may exhibit: Efficiency -- conveying task-relevant information more concisely than natural language, and Covertness -- becoming difficult for external observers to interpret, raising concerns about transparency and control. To investigate these aspects, we use a referential-game framework in which vision-language model (VLM) agents communicate, providing a controlled, measurable setting for evaluating language variants. Experiments show that VLMs can develop effective, task-adapted communication patterns. At the same time, they can develop covert protocols that are difficult for humans and external agents to interpret. We also observe spontaneous coordination between similar models without explicitly shared protocols. These findings highlight both the potential and the risks of task-oriented communication, and position referential games as a valuable testbed for future work in this area.

Paperid: 4567, https://arxiv.org/pdf/2601.17927.pdf

Abstract:
Controllable image generation is fundamental to the success of modern generative AI, yet it faces a critical trade-off between semantic fidelity and inference speed. The RemEdit diffusion-based framework addresses this trade-off with two synergistic innovations. First, for editing fidelity, we navigate the latent space as a Riemannian manifold. A mamba-based module efficiently learns the manifold's structure, enabling direct and accurate geodesic path computation for smooth semantic edits. This control is further refined by a dual-SLERP blending technique and a goal-aware prompt enrichment pass from a Vision-Language Model. Second, for additional acceleration, we introduce a novel task-specific attention pruning mechanism. A lightweight pruning head learns to retain tokens essential to the edit, enabling effective optimization without the semantic degradation common in content-agnostic approaches. RemEdit surpasses prior state-of-the-art editing frameworks while maintaining real-time performance under 50% pruning. Consequently, RemEdit establishes a new benchmark for practical and powerful image editing. Source code: https://www.github.com/eashanadhikarla/RemEdit.

Paperid: 4568, https://arxiv.org/pdf/2601.09661.pdf

Abstract:
Large-scale vision-language models such as CLIP achieve strong zero-shot recognition but struggle with classes that are rarely seen during pretraining, including newly emerging entities and culturally specific categories. We introduce LiteEmbed, a lightweight framework for few-shot personalization of CLIP that enables new classes to be added without retraining its encoders. LiteEmbed performs subspace-guided optimization of text embeddings within CLIP's vocabulary, leveraging a PCA-based decomposition that disentangles coarse semantic directions from fine-grained variations. Two complementary objectives, coarse alignment and fine separation, jointly preserve global semantic consistency while enhancing discriminability among visually similar classes. Once optimized, the embeddings are plug-and-play, seamlessly substituting CLIP's original text features across classification, retrieval, segmentation, and detection tasks. Extensive experiments demonstrate substantial gains over prior methods, establishing LiteEmbed as an effective approach for adapting CLIP to underrepresented, rare, or unseen classes.

Paperid: 4569, https://arxiv.org/pdf/2601.09111.pdf

Abstract:
Vision-Language Navigation aims to enable agents to navigate to a target location based on language instructions. Traditional VLN often follows a close-set assumption, i.e., training and test data share the same style of the input images and instructions. However, the real world is open and filled with various unseen environments, posing enormous difficulties for close-set methods. To this end, we focus on the General Scene Adaptation (GSA-VLN) task, aiming to learn generalized navigation ability by introducing diverse environments and inconsistent intructions.Towards this task, when facing unseen environments and instructions, the challenge mainly lies in how to enable the agent to dynamically produce generalized strategies during the navigation process. Recent research indicates that by means of fast and slow cognition systems, human beings could generate stable policies, which strengthen their adaptation for open world. Inspired by this idea, we propose the slow4fast-VLN, establishing a dynamic interactive fast-slow reasoning framework. The fast-reasoning module, an end-to-end strategy network, outputs actions via real-time input. It accumulates execution records in a history repository to build memory. The slow-reasoning module analyze the memories generated by the fast-reasoning module. Through deep reflection, it extracts experiences that enhance the generalization ability of decision-making. These experiences are structurally stored and used to continuously optimize the fast-reasoning module. Unlike traditional methods that treat fast-slow reasoning as independent mechanisms, our framework enables fast-slow interaction. By leveraging the experiences from slow reasoning. This interaction allows the system to continuously adapt and efficiently execute navigation tasks when facing unseen scenarios.

Paperid: 4570, https://arxiv.org/pdf/2601.07558.pdf

Abstract:
Autonomous 3D scanning of open-world target structures via drones remains challenging despite broad applications. Existing paradigms rely on restrictive assumptions or effortful human priors, limiting practicality, efficiency, and adaptability. Recent foundation models (FMs) offer great potential to bridge this gap. This paper investigates a critical research problem: What system architecture can effectively integrate FM knowledge for this task? We answer it with FlyCo, a principled FM-empowered perception-prediction-planning loop enabling fully autonomous, prompt-driven 3D target scanning in diverse unknown open-world environments. FlyCo directly translates low-effort human prompts (text, visual annotations) into precise adaptive scanning flights via three coordinated stages: (1) perception fuses streaming sensor data with vision-language FMs for robust target grounding and tracking; (2) prediction distills FM knowledge and combines multi-modal cues to infer the partially observed target's complete geometry; (3) planning leverages predictive foresight to generate efficient and safe paths with comprehensive target coverage. Building on this, we further design key components to boost open-world target grounding efficiency and robustness, enhance prediction quality in terms of shape accuracy, zero-shot generalization, and temporal stability, and balance long-horizon flight efficiency with real-time computability and online collision avoidance. Extensive challenging real-world and simulation experiments show FlyCo delivers precise scene understanding, high efficiency, and real-time safety, outperforming existing paradigms with lower human effort and verifying the proposed architecture's practicality. Comprehensive ablations validate each component's contribution. FlyCo also serves as a flexible, extensible blueprint, readily leveraging future FM and robotics advances. Code will be released.

Paperid: 4571, https://arxiv.org/pdf/2601.03956.pdf

Abstract:
Recent Vision-Language Models (VLMs) have demonstrated significant potential in robotic planning. However, they typically function as semantic reasoners, lacking an intrinsic understanding of the specific robot's physical capabilities. This limitation is particularly critical in interactive navigation, where robots must actively modify cluttered environments to create traversable paths. Existing VLM-based navigators are predominantly confined to passive obstacle avoidance, failing to reason about when and how to interact with objects to clear blocked paths. To bridge this gap, we propose Counterfactual Interactive Navigation via Skill-aware VLM (CoINS), a hierarchical framework that integrates skill-aware reasoning and robust low-level execution. Specifically, we fine-tune a VLM, named InterNav-VLM, which incorporates skill affordance and concrete constraint parameters into the input context and grounds them into a metric-scale environmental representation. By internalizing the logic of counterfactual reasoning through fine-tuning on the proposed InterNav dataset, the model learns to implicitly evaluate the causal effects of object removal on navigation connectivity, thereby determining interaction necessity and target selection. To execute the generated high-level plans, we develop a comprehensive skill library through reinforcement learning, specifically introducing traversability-oriented strategies to manipulate diverse objects for path clearance. A systematic benchmark in Isaac Sim is proposed to evaluate both the reasoning and execution aspects of interactive navigation. Extensive simulations and real-world experiments demonstrate that CoINS significantly outperforms representative baselines, achieving a 17\% higher overall success rate and over 80\% improvement in complex long-horizon scenarios compared to the best-performing baseline

Paperid: 4572, https://arxiv.org/pdf/2512.21699.pdf

Abstract:
Agentic AI represents a major shift in how autonomous systems reason, plan, and execute multi-step tasks through the coordination of Large Language Models (LLMs), Vision Language Models (VLMs), tools, and external services. While these systems enable powerful new capabilities, increasing autonomy introduces critical challenges related to explainability, accountability, robustness, and governance, especially when agent outputs influence downstream actions or decisions. Existing agentic AI implementations often emphasize functionality and scalability, yet provide limited mechanisms for understanding decision rationale or enforcing responsibility across agent interactions. This paper presents a Responsible(RAI) and Explainable(XAI) AI Agent Architecture for production-grade agentic workflows based on multi-model consensus and reasoning-layer governance. In the proposed design, a consortium of heterogeneous LLM and VLM agents independently generates candidate outputs from a shared input context, explicitly exposing uncertainty, disagreement, and alternative interpretations. A dedicated reasoning agent then performs structured consolidation across these outputs, enforcing safety and policy constraints, mitigating hallucinations and bias, and producing auditable, evidence-backed decisions. Explainability is achieved through explicit cross-model comparison and preserved intermediate outputs, while responsibility is enforced through centralized reasoning-layer control and agent-level constraints. We evaluate the architecture across multiple real-world agentic AI workflows, demonstrating that consensus-driven reasoning improves robustness, transparency, and operational trust across diverse application domains. This work provides practical guidance for designing agentic AI systems that are autonomous and scalable, yet responsible and explainable by construction.

Paperid: 4573, https://arxiv.org/pdf/2512.19178.pdf

Abstract:
Bridging the gap between natural language commands and autonomous execution in unstructured environments remains an open challenge for robotics. This requires robots to perceive and reason over the current task scene through multiple modalities, and to plan their behaviors to achieve their intended goals. Traditional robotic task-planning approaches often struggle to bridge low-level execution with high-level task reasoning, and cannot dynamically update task strategies when instructions change during execution, which ultimately limits their versatility and adaptability to new tasks. In this work, we propose a novel language model-based framework for dynamic robot task planning. Our Vision-Language-Policy (VLP) model, based on a vision-language model fine-tuned on real-world data, can interpret semantic instructions and integrate reasoning over the current task scene to generate behavior policies that control the robot to accomplish the task. Moreover, it can dynamically adjust the task strategy in response to changes in the task, enabling flexible adaptation to evolving task requirements. Experiments conducted with different robots and a variety of real-world tasks show that the trained model can efficiently adapt to novel scenarios and dynamically update its policy, demonstrating strong planning autonomy and cross-embodiment generalization. Videos: https://robovlp.github.io/

Paperid: 4574, https://arxiv.org/pdf/2512.18599.pdf

Abstract:
Complex image restoration aims to recover high-quality images from inputs affected by multiple degradations such as blur, noise, rain, and compression artifacts. Recent restoration agents, powered by vision-language models and large language models, offer promising restoration capabilities but suffer from significant efficiency bottlenecks due to reflection, rollback, and iterative tool searching. Moreover, their performance heavily depends on degradation recognition models that require extensive annotations for training, limiting their applicability in label-free environments. To address these limitations, we propose a policy optimization-based restoration framework that learns an lightweight agent to determine tool-calling sequences. The agent operates in a sequential decision process, selecting the most appropriate restoration operation at each step to maximize final image quality. To enable training within label-free environments, we introduce a novel reward mechanism driven by multimodal large language models, which act as human-aligned evaluator and provide perceptual feedback for policy improvement. Once trained, our agent executes a deterministic restoration plans without redundant tool invocations, significantly accelerating inference while maintaining high restoration quality. Extensive experiments show that despite using no supervision, our method matches SOTA performance on full-reference metrics and surpasses existing approaches on no-reference metrics across diverse degradation scenarios.

Paperid: 4575, https://arxiv.org/pdf/2512.16145.pdf

Abstract:
Medical report generation (MRG) aims to automatically derive radiology-style reports from medical images to aid in clinical decision-making. However, existing methods often generate text that mimics the linguistic style of radiologists but fails to guarantee clinical correctness, because they are trained on token-level objectives which focus on word-choice and sentence structure rather than actual medical accuracy. We propose a semantic-driven reinforcement learning (SRL) method for medical report generation, adopted on a large vision-language model (LVLM). SRL adopts Group Relative Policy Optimization (GRPO) to encourage clinical-correctness-guided learning beyond imitation of language style. Specifically, we optimise a report-level reward: a margin-based cosine similarity (MCCS) computed between key radiological findings extracted from generated and reference reports, thereby directly aligning clinical-label agreement and improving semantic correctness. A lightweight reasoning format constraint further guides the model to generate structured "thinking report" outputs. We evaluate Medical Report Generation with Sematic-driven Reinforment Learning (MRG-R1), on two datasets: IU X-Ray and MIMIC-CXR using clinical efficacy (CE) metrics. MRG-R1 achieves state-of-the-art performance with CE-F1 51.88 on IU X-Ray and 40.39 on MIMIC-CXR. We found that the label-semantic reinforcement is better than conventional token-level supervision. These results indicate that optimizing a clinically grounded, report-level reward rather than token overlap,meaningfully improves clinical correctness. This work is a prior to explore semantic-reinforcement in supervising medical correctness in medical Large vision-language model(Med-LVLM) training.

Paperid: 4576, https://arxiv.org/pdf/2512.14442.pdf

Abstract:
Affordance prediction, which identifies interaction regions on objects based on language instructions, is critical for embodied AI. Prevailing end-to-end models couple high-level reasoning and low-level grounding into a single monolithic pipeline and rely on training over annotated datasets, which leads to poor generalization on novel objects and unseen environments. In this paper, we move beyond this paradigm by proposing A4-Agent, a training-free agentic framework that decouples affordance prediction into a three-stage pipeline. Our framework coordinates specialized foundation models at test time: (1) a $\textbf{Dreamer}$ that employs generative models to visualize $\textit{how}$ an interaction would look; (2) a $\textbf{Thinker}$ that utilizes large vision-language models to decide $\textit{what}$ object part to interact with; and (3) a $\textbf{Spotter}$ that orchestrates vision foundation models to precisely locate $\textit{where}$ the interaction area is. By leveraging the complementary strengths of pre-trained models without any task-specific fine-tuning, our zero-shot framework significantly outperforms state-of-the-art supervised methods across multiple benchmarks and demonstrates robust generalization to real-world settings.

Paperid: 4577, https://arxiv.org/pdf/2512.13747.pdf

Abstract:
With the rapid progress of large language models (LLMs), advanced multimodal large language models (MLLMs) have demonstrated impressive zero-shot capabilities on vision-language tasks. In the biomedical domain, however, even state-of-the-art MLLMs struggle with basic Medical Decision Making (MDM) tasks. We investigate this limitation using two challenging datasets: (1) three-stage Alzheimer's disease (AD) classification (normal, mild cognitive impairment, dementia), where category differences are visually subtle, and (2) MIMIC-CXR chest radiograph classification with 14 non-mutually exclusive conditions. Our empirical study shows that text-only reasoning consistently outperforms vision-only or vision-text settings, with multimodal inputs often performing worse than text alone. To mitigate this, we explore three strategies: (1) in-context learning with reason-annotated exemplars, (2) vision captioning followed by text-only inference, and (3) few-shot fine-tuning of the vision tower with classification supervision. These findings reveal that current MLLMs lack grounded visual understanding and point to promising directions for improving multimodal decision making in healthcare.

Paperid: 4578, https://arxiv.org/pdf/2512.13177.pdf

Abstract:
Vision-language models enable the understanding and reasoning of complex traffic scenarios through multi-source information fusion, establishing it as a core technology for autonomous driving. However, existing vision-language models are constrained by the image understanding paradigm in 2D plane, which restricts their capability to perceive 3D spatial information and perform deep semantic fusion, resulting in suboptimal performance in complex autonomous driving environments. This study proposes MMDrive, an multimodal vision-language model framework that extends traditional image understanding to a generalized 3D scene understanding framework. MMDrive incorporates three complementary modalities, including occupancy maps, LiDAR point clouds, and textual scene descriptions. To this end, it introduces two novel components for adaptive cross-modal fusion and key information extraction. Specifically, the Text-oriented Multimodal Modulator dynamically weights the contributions of each modality based on the semantic cues in the question, guiding context-aware feature integration. The Cross-Modal Abstractor employs learnable abstract tokens to generate compact, cross-modal summaries that highlight key regions and essential semantics. Comprehensive evaluations on the DriveLM and NuScenes-QA benchmarks demonstrate that MMDrive achieves significant performance gains over existing vision-language models for autonomous driving, with a BLEU-4 score of 54.56 and METEOR of 41.78 on DriveLM, and an accuracy score of 62.7% on NuScenes-QA. MMDrive effectively breaks the traditional image-only understanding barrier, enabling robust multimodal reasoning in complex driving environments and providing a new foundation for interpretable autonomous driving scene understanding.

Paperid: 4579, https://arxiv.org/pdf/2512.06112.pdf

Abstract:
We introduce WAM-Flow, a vision-language-action (VLA) model that casts ego-trajectory planning as discrete flow matching over a structured token space. In contrast to autoregressive decoders, WAM-Flow performs fully parallel, bidirectional denoising, enabling coarse-to-fine refinement with a tunable compute-accuracy trade-off. Specifically, the approach combines a metric-aligned numerical tokenizer that preserves scalar geometry via triplet-margin learning, a geometry-aware flow objective and a simulator-guided GRPO alignment that integrates safety, ego progress, and comfort rewards while retaining parallel generation. A multi-stage adaptation converts a pre-trained auto-regressive backbone (Janus-1.5B) from causal decoding to non-causal flow model and strengthens road-scene competence through continued multimodal pretraining. Thanks to the inherent nature of consistency model training and parallel decoding inference, WAM-Flow achieves superior closed-loop performance against autoregressive and diffusion-based VLA baselines, with 1-step inference attaining 89.1 PDMS and 5-step inference reaching 90.3 PDMS on NAVSIM v1 benchmark. These results establish discrete flow matching as a new promising paradigm for end-to-end autonomous driving. The code will be publicly available soon.

Paperid: 4580, https://arxiv.org/pdf/2512.04785.pdf

Abstract:
AI agent-based systems are becoming increasingly integral to modern software architectures, enabling autonomous decision-making, dynamic task execution, and multimodal interactions through large language models (LLMs). However, these systems introduce novel and evolving security challenges, including prompt injection attacks, context poisoning, model manipulation, and opaque agent-to-agent communication, that are not effectively captured by traditional threat modeling frameworks. In this paper, we introduce ASTRIDE, an automated threat modeling platform purpose-built for AI agent-based systems. ASTRIDE extends the classical STRIDE framework by introducing a new threat category, A for AI Agent-Specific Attacks, which encompasses emerging vulnerabilities such as prompt injection, unsafe tool invocation, and reasoning subversion, unique to agent-based applications. To automate threat modeling, ASTRIDE combines a consortium of fine-tuned vision-language models (VLMs) with the OpenAI-gpt-oss reasoning LLM to perform end-to-end analysis directly from visual agent architecture diagrams, such as data flow diagrams(DFDs). LLM agents orchestrate the end-to-end threat modeling automation process by coordinating interactions between the VLM consortium and the reasoning LLM. Our evaluations demonstrate that ASTRIDE provides accurate, scalable, and explainable threat modeling for next-generation intelligent systems. To the best of our knowledge, ASTRIDE is the first framework to both extend STRIDE with AI-specific threats and integrate fine-tuned VLMs with a reasoning LLM to fully automate diagram-driven threat modeling in AI agent-based applications.

Paperid: 4581, https://arxiv.org/pdf/2512.04728.pdf

Abstract:
Generative psychological analysis of in-the-wild conversations faces two fundamental challenges: (1) existing Vision-Language Models (VLMs) fail to resolve Articulatory-Affective Ambiguity, where visual patterns of speech mimic emotional expressions; and (2) progress is stifled by a lack of verifiable evaluation metrics capable of assessing visual grounding and reasoning depth. We propose a complete ecosystem to address these twin challenges. First, we introduce Multilevel Insight Network for Disentanglement(MIND), a novel hierarchical visual encoder that introduces a Status Judgment module to algorithmically suppress ambiguous lip features based on their temporal feature variance, achieving explicit visual disentanglement. Second, we construct ConvoInsight-DB, a new large-scale dataset with expert annotations for micro-expressions and deep psychological inference. Third, Third, we designed the Mental Reasoning Insight Rating Metric (PRISM), an automated dimensional framework that uses expert-guided LLM to measure the multidimensional performance of large mental vision models. On our PRISM benchmark, MIND significantly outperforms all baselines, achieving a +86.95% gain in micro-expression detection over prior SOTA. Ablation studies confirm that our Status Judgment disentanglement module is the most critical component for this performance leap. Our code has been opened.

Paperid: 4582, https://arxiv.org/pdf/2512.00807.pdf

Abstract:
Vision-Language Models (VLMs) inherit significant social biases from their training data, notably in gender representation. Current fairness interventions often adopt a difference-unaware perspective that enforces uniform treatment across demographic groups. These approaches, however, fail to distinguish between contexts where neutrality is required and those where group-specific attributes are legitimate and must be preserved. Building upon recent advances in difference-aware fairness for text-only models, we extend this concept to the multimodal domain and formalize the problem of difference-aware gender fairness for image captioning and text-to-image generation. We advocate for selective debiasing, which aims to mitigate unwanted bias in neutral contexts while preserving valid distinctions in explicit ones. To achieve this, we propose BioPro (Bias Orthogonal Projection), an entirely training-free framework. BioPro identifies a low-dimensional gender-variation subspace through counterfactual embeddings and applies projection to selectively neutralize gender-related information. Experiments show that BioPro effectively reduces gender bias in neutral cases while maintaining gender faithfulness in explicit ones, thus providing a promising direction toward achieving selective fairness in VLMs. Beyond gender bias, we further demonstrate that BioPro can effectively generalize to continuous bias variables, such as scene brightness, highlighting its broader applicability.

Paperid: 4583, https://arxiv.org/pdf/2511.21947.pdf

Abstract:
Urban walkability is a cornerstone of public health, sustainability, and quality of life. Traditional walkability assessments rely on surveys and field audits, which are costly and difficult to scale. Recent studies have used satellite imagery, street view imagery, or population indicators to estimate walkability, but these single-source approaches capture only one dimension of the walking environment. Satellite data describe the built environment from above, but overlook the pedestrian perspective. Street view imagery captures conditions at the ground level, but lacks broader spatial context. Population dynamics reveal patterns of human activity but not the visual form of the environment. We introduce WalkCLIP, a multimodal framework that integrates these complementary viewpoints to predict urban walkability. WalkCLIP learns walkability-aware vision-language representations from GPT-4o generated image captions, refines these representations with a spatial aggregation module that incorporates neighborhood context, and fuses the resulting features with representations from a population dynamics foundation model. Evaluated at 4,660 locations throughout Minneapolis-Saint Paul, WalkCLIP outperforms unimodal and multimodal baselines in both predictive accuracy and spatial alignment. These results show that the integration of visual and behavioral signals yields reliable predictions of the walking environment.

Paperid: 4584, https://arxiv.org/pdf/2511.19516.pdf

Abstract:
Visual grounding, the task of linking textual queries to specific regions within images, plays a pivotal role in vision-language integration. Existing methods typically rely on extensive task-specific annotations and fine-tuning, limiting their ability to generalize effectively to novel or out-of-distribution scenarios. To address these limitations, we introduce GroundingAgent, a novel agentic visual grounding framework that operates without any task-specific fine-tuning. GroundingAgent employs a structured, iterative reasoning mechanism that integrates pretrained open-vocabulary object detectors, multimodal large language models (MLLMs), and large language models (LLMs) to progressively refine candidate regions through joint semantic and spatial analyses. Remarkably, GroundingAgent achieves an average zero-shot grounding accuracy of 65.1 % on widely-used benchmarks (RefCOCO, RefCOCO+, RefCOCOg), entirely without fine-tuning. Furthermore, by substituting MLLM-generated captions with the original query texts, the accuracy at the selection stage alone reaches approximately 90 %, closely matching supervised performance and underscoring the critical role of LLM reasoning capabilities. GroundingAgent also offers strong interpretability, transparently illustrating each reasoning step and providing clear insights into its decision-making process.

Paperid: 4585, https://arxiv.org/pdf/2511.12978.pdf

Abstract:
Contrastive vision-language models (VLMs) such as CLIP achieve strong zero-shot recognition yet remain vulnerable to spurious correlations, particularly background over-reliance. We introduce Cluster-based Concept Importance (CCI), a novel interpretability method that uses CLIP's own patch embeddings to group spatial patches into semantically coherent clusters, mask them, and evaluate relative changes in model predictions. CCI sets a new state of the art on faithfulness benchmarks, surpassing prior methods by large margins; for example, it yields more than a twofold improvement on the deletion-AUC metric for MS COCO retrieval. We further propose that CCI, when combined with GroundedSAM, automatically categorizes predictions as foreground- or background-driven, providing a crucial diagnostic ability. Existing benchmarks such as CounterAnimals, however, rely solely on accuracy and implicitly attribute all performance degradation to background correlations. Our analysis shows this assumption to be incomplete, since many errors arise from viewpoint variation, scale shifts, and fine-grained object confusions. To disentangle these effects, we introduce COVAR, a benchmark that systematically varies object foregrounds and backgrounds. Leveraging CCI with COVAR, we present a comprehensive evaluation of eighteen CLIP variants, offering methodological advances and empirical evidence that chart a path toward more robust VLMs.

Paperid: 4586, https://arxiv.org/pdf/2511.12079.pdf

Abstract:
Vector quantization has emerged as a powerful tool in large-scale multimodal models, unifying heterogeneous representations through discrete token encoding. However, its effectiveness hinges on robust codebook design. Current prototype-based approaches relying on trainable vectors or clustered centroids fall short in representativeness and interpretability, even as multimodal alignment demonstrates its promise in vision-language models. To address these limitations, we propose a simple multimodal prompting-driven quantization framework for point cloud analysis. Our methodology is built upon two core insights: 1) Text embeddings from pre-trained models inherently encode visual semantics through many-to-one contrastive alignment, naturally serving as robust prototype priors; and 2) Multimodal prompts enable adaptive refinement of these prototypes, effectively mitigating vision-language semantic gaps. The framework introduces a dual-constrained quantization space, enforced by compactness and separation regularization, which seamlessly integrates visual and prototype features, resulting in hybrid representations that jointly encode geometric and semantic information. Furthermore, we employ Gumbel-Softmax relaxation to achieve differentiable discretization while maintaining quantization sparsity. Extensive experiments on the ModelNet40 and ScanObjectNN datasets clearly demonstrate the superior effectiveness of the proposed method.

Paperid: 4587, https://arxiv.org/pdf/2511.11704.pdf

Abstract:
We present a lightweight yet effective pipeline for training vision-language models to solve math problems by rendering LaTeX encoded equations into images and pairing them with structured chain-of-thought prompts. This simple text-to-vision augmentation enables compact multimodal architectures to achieve state-of-the-art reasoning accuracy. Through systematic ablations, we find that rendering fidelity and prompt design are the primary drivers of performance. Despite its simplicity, our approach consistently matches or surpasses both open-source and proprietary math-focused vision-language solvers on widely used benchmarks, while preserving broad general-domain competence - showing gains on tasks such as MMMU, ChartQA, and DocVQA of up to 20%.

Paperid: 4588, https://arxiv.org/pdf/2511.11313.pdf

Abstract:
Large Vision-Language Models (LVLMs) have demonstrated strong multimodal reasoning capabilities on long and complex documents. However, their high memory footprint makes them impractical for deployment on resource-constrained edge devices. We present DocSLM, an efficient Small Vision-Language Model designed for long-document understanding under constrained memory resources. DocSLM incorporates a Hierarchical Multimodal Compressor that jointly encodes visual, textual, and layout information from each page into a fixed-length sequence, greatly reducing memory consumption while preserving both local and global semantics. To enable scalable processing over arbitrarily long inputs, we introduce a Streaming Abstention mechanism that operates on document segments sequentially and filters low-confidence responses using an entropy-based uncertainty calibrator. Across multiple long multimodal document benchmarks, DocSLM matches or surpasses state-of-the-art methods while using 82\% fewer visual tokens, 75\% fewer parameters, and 71\% lower latency, delivering reliable multimodal document understanding on lightweight edge devices. Code is available in the supplementary material.

Paperid: 4589, https://arxiv.org/pdf/2511.11239.pdf

Abstract:
Existing Vision Language Models (VLMs) architecturally rooted in "flatland" perception, fundamentally struggle to comprehend real-world 3D spatial intelligence. This failure stems from a dual-bottleneck: input-stage conflict between computationally exorbitant geometric-aware encoders and superficial 2D-only features, and output-stage misalignment where discrete tokenizers are structurally incapable of producing precise, continuous numerical values. To break this impasse, we introduce GEODE (Geometric-Output and Decoupled-Input Engine), a novel architecture that resolves this dual-bottleneck by decoupling 3D reasoning from numerical generation. GEODE augments main VLM with two specialized, plug-and-play modules: Decoupled Rationale Module (DRM) that acts as spatial co-processor, aligning explicit 3D data with 2D visual features via cross-attention and distilling spatial Chain-of-Thought (CoT) logic into injectable Rationale Tokens; and Direct Regression Head (DRH), an "Embedding-as-Value" paradigm which routes specialized control tokens to a lightweight MLP for precise, continuous regression of scalars and 3D bounding boxes. The synergy of these modules allows our 1.5B parameter model to function as a high-level semantic dispatcher, achieving state-of-the-art spatial reasoning performance that rivals 7B+ models.

Paperid: 4590, https://arxiv.org/pdf/2511.09286.pdf

Abstract:
Multi-teacher knowledge distillation (KD), a more effective technique than traditional single-teacher methods, transfers knowledge from expert teachers to a compact student model using logit or feature matching. However, most existing approaches lack knowledge diversity, as they rely solely on unimodal visual information, overlooking the potential of cross-modal representations. In this work, we explore the use of CLIP's vision-language knowledge as a complementary source of supervision for KD, an area that remains largely underexplored. We propose a simple yet effective framework that fuses the logits and features of a conventional teacher with those from CLIP. By incorporating CLIP's multi-prompt textual guidance, the fused supervision captures both dataset-specific and semantically enriched visual cues. Beyond accuracy, analysis shows that the fused teacher yields more confident and reliable predictions, significantly increasing confident-correct cases while reducing confidently wrong ones. Moreover, fusion with CLIP refines the entire logit distribution, producing semantically meaningful probabilities for non-target classes, thereby improving inter-class consistency and distillation quality. Despite its simplicity, the proposed method, Enriching Knowledge Distillation (RichKD), consistently outperforms most existing baselines across multiple benchmarks and exhibits stronger robustness under distribution shifts and input corruptions.

Paperid: 4591, https://arxiv.org/pdf/2511.08402.pdf

Abstract:
Accurate disease interpretation from radiology remains challenging due to imaging heterogeneity. Achieving expert-level diagnostic decisions requires integration of subtle image features with clinical knowledge. Yet major vision-language models (VLMs) treat images as holistic entities and overlook fine-grained image details that are vital for disease diagnosis. Clinicians analyze images by utilizing their prior medical knowledge and identify anatomical structures as important region of interests (ROIs). Inspired from this human-centric workflow, we introduce Anatomy-VLM, a fine-grained, vision-language model that incorporates multi-scale information. First, we design a model encoder to localize key anatomical features from entire medical images. Second, these regions are enriched with structured knowledge for contextually-aware interpretation. Finally, the model encoder aligns multi-scale medical information to generate clinically-interpretable disease prediction. Anatomy-VLM achieves outstanding performance on both in- and out-of-distribution datasets. We also validate the performance of Anatomy-VLM on downstream image segmentation tasks, suggesting that its fine-grained alignment captures anatomical and pathology-related knowledge. Furthermore, the Anatomy-VLM's encoder facilitates zero-shot anatomy-wise interpretation, providing its strong expert-level clinical interpretation capabilities.

Paperid: 4592, https://arxiv.org/pdf/2511.06496.pdf

Abstract:
Vision Language Models (VLMs) are increasingly used in autonomous driving to help understand traffic scenes, but they sometimes produce hallucinations, which are false details not grounded in the visual input. Detecting and mitigating hallucinations is challenging when ground-truth references are unavailable and model internals are inaccessible. This paper proposes a novel self-contained low-rank approach to automatically rank multiple candidate captions generated by multiple VLMs based on their hallucination levels, using only the captions themselves without requiring external references or model access. By constructing a sentence-embedding matrix and decomposing it into a low-rank consensus component and a sparse residual, we use the residual magnitude to rank captions: selecting the one with the smallest residual as the most hallucination-free. Experiments on the NuScenes dataset demonstrate that our approach achieves 87% selection accuracy in identifying hallucination-free captions, representing a 19% improvement over the unfiltered baseline and a 6-10% improvement over multi-agent debate method. The sorting produced by sparse error magnitudes shows strong correlation with human judgments of hallucinations, validating our scoring mechanism. Additionally, our method, which can be easily parallelized, reduces inference time by 51-67% compared to debate approaches, making it practical for real-time autonomous driving applications.

Paperid: 4593, https://arxiv.org/pdf/2511.05551.pdf

Abstract:
Vision-based quality assessment in additive manufacturing often requires dedicated machine learning models and application-specific datasets. However, data collection and model training can be expensive and time-consuming. In this paper, we leverage vision-language models' (VLMs') reasoning capabilities to assess the quality of printed parts and introduce in-context learning (ICL) to provide VLMs with necessary application-specific knowledge and demonstration samples. This method eliminates the requirement for large application-specific datasets for training models. We explored different sampling strategies for ICL to search for the optimal configuration that makes use of limited samples. We evaluated these strategies on two VLMs, Gemini-2.5-flash and Gemma3:27b, with quality assessment tasks in wire-laser direct energy deposition processes. The results show that ICL-assisted VLMs can reach quality classification accuracies similar to those of traditional machine learning models while requiring only a minimal number of samples. In addition, unlike traditional classification models that lack transparency, VLMs can generate human-interpretable rationales to enhance trust. Since there are no metrics to evaluate their interpretability in manufacturing applications, we propose two metrics, knowledge relevance and rationale validity, to evaluate the quality of VLMs' supporting rationales. Our results show that ICL-assisted VLMs can address application-specific tasks with limited data, achieving relatively high accuracy while also providing valid supporting rationales for improved decision transparency.

Paperid: 4594, https://arxiv.org/pdf/2511.04664.pdf

Abstract:
Autonomous driving systems remain brittle in rare, ambiguous, and out-of-distribution scenarios, where human driver succeed through contextual reasoning. Shared autonomy has emerged as a promising approach to mitigate such failures by incorporating human input when autonomy is uncertain. However, most existing methods restrict arbitration to low-level trajectories, which represent only geometric paths and therefore fail to preserve the underlying driving intent. We propose a unified shared autonomy framework that integrates human input and autonomous planners at a higher level of abstraction. Our method leverages Vision Language Models (VLMs) to infer driver intent from multi-modal cues -- such as driver actions and environmental context -- and to synthesize coherent strategies that mediate between human and autonomous control. We first study the framework in a mock-human setting, where it achieves perfect recall alongside high accuracy and precision. A human-subject survey further shows strong alignment, with participants agreeing with arbitration outcomes in 92% of cases. Finally, evaluation on the Bench2Drive benchmark demonstrates a substantial reduction in collision rate and improvement in overall performance compared to pure autonomy. Arbitration at the level of semantic, language-based representations emerges as a design principle for shared autonomy, enabling systems to exercise common-sense reasoning and maintain continuity with human intent.

Paperid: 4595, https://arxiv.org/pdf/2511.00255.pdf

Abstract:
In entomology and ecology research, biologists often need to collect a large number of insects, among which beetles are the most common species. A common practice for biologists to organize beetles is to place them on trays and take a picture of each tray. Given the images of thousands of such trays, it is important to have an automated pipeline to process the large-scale data for further research. Therefore, we develop a 3-stage pipeline to detect all the beetles on each tray, sort and crop the image of each beetle, and do morphological segmentation on the cropped beetles. For detection, we design an iterative process utilizing a transformer-based open-vocabulary object detector and a vision-language model. For segmentation, we manually labeled 670 beetle images and fine-tuned two variants of a transformer-based segmentation model to achieve fine-grained segmentation of beetles with relatively high accuracy. The pipeline integrates multiple deep learning methods and is specialized for beetle image processing, which can greatly improve the efficiency to process large-scale beetle data and accelerate biological research.

Paperid: 4596, https://arxiv.org/pdf/2510.26411.pdf

Abstract:
Artificial intelligence in healthcare requires models that are accurate and interpretable. We advance mechanistic interpretability in medical vision by applying Medical Sparse Autoencoders (MedSAEs) to the latent space of MedCLIP, a vision-language model trained on chest radiographs and reports. To quantify interpretability, we propose an evaluation framework that combines correlation metrics, entropy analyzes, and automated neuron naming via the MedGEMMA foundation model. Experiments on the CheXpert dataset show that MedSAE neurons achieve higher monosemanticity and interpretability than raw MedCLIP features. Our findings bridge high-performing medical AI and transparency, offering a scalable step toward clinically reliable representations.

Paperid: 4597, https://arxiv.org/pdf/2510.25122.pdf

Abstract:
Vision-language-action (VLA) models have significantly advanced robotic manipulation by integrating vision-language models (VLMs), and action decoders into a unified architecture. However, their deployment on resource-constrained edge devices, such as mobile robots or embedded systems (e.g., Jetson Orin Nano), remains challenging due to high computational demands, especially in real-world scenarios where power, latency, and computational resources are critical. To close this gap, we introduce Nano-scale Vision-Language Action (NanoVLA), a family of lightweight VLA architectures that achieve high performance with minimal resources. Our core innovations include: (1) vision-language decoupling that moves conventional early vision and language inputs fusion in VLM to late stage, achieving better performance while enabling caching and reduce inference overhead and latency; (2) long-short action chunking to ensure smooth, coherent multi-step planning without sacrificing real-time responsiveness; (3) dynamic routing that adaptively assigns lightweight or heavy backbones based on task complexity, further optimizing inference efficiency. Experimental results on several benchmarks, as well as real-world deployments, demonstrate that NanoVLA achieves up to 52x faster inference on edge devices compared to previous state-of-the-art VLA models, with 98% less parameters while maintaining or surpassing their task accuracy and generalization. Ablation studies confirm that our decoupling strategy preserves cross-task transferability, and the routing module enhances cost-performance trade-offs, enabling practical, high-precision robotic manipulation on resource-constrained hardware.

Paperid: 4598, https://arxiv.org/pdf/2510.19001.pdf

Abstract:
We present a two-phase vision-language QA system for autonomous driving that answers high-level perception, prediction, and planning questions. In Phase-1, a large multimodal LLM (Qwen2.5-VL-32B) is conditioned on six-camera inputs, a short temporal window of history, and a chain-of-thought prompt with few-shot exemplars. A self-consistency ensemble (multiple sampled reasoning chains) further improves answer reliability. In Phase-2, we augment the prompt with nuScenes scene metadata (object annotations, ego-vehicle state, etc.) and category-specific question instructions (separate prompts for perception, prediction, planning tasks). In experiments on a driving QA benchmark, our approach significantly outperforms the baseline Qwen2.5 models. For example, using 5 history frames and 10-shot prompting in Phase-1 yields 65.1% overall accuracy (vs.62.61% with zero-shot); applying self-consistency raises this to 66.85%. Phase-2 achieves 67.37% overall. Notably, the system maintains 96% accuracy under severe visual corruption. These results demonstrate that carefully engineered prompts and contextual grounding can greatly enhance high-level driving QA with pretrained vision-language models.

Paperid: 4599, https://arxiv.org/pdf/2510.16448.pdf

Abstract:
Sparse Mixture of Experts (sMoE) has become a pivotal approach for scaling large vision-language models, offering substantial capacity while maintaining computational efficiency through dynamic, sparse activation of experts. However, existing routing mechanisms, typically based on similarity scoring, struggle to effectively capture the underlying input structure. This limitation leads to a trade-off between expert specialization and balanced computation, hindering both scalability and performance. We propose Input Domain Aware MoE, a novel routing framework that leverages a probabilistic mixture model to better partition the input space. By modeling routing probabilities as a mixture of distributions, our method enables experts to develop clear specialization boundaries while achieving balanced utilization. Unlike conventional approaches, our routing mechanism is trained independently of task-specific objectives, allowing for stable optimization and decisive expert assignments. Empirical results on vision-language tasks demonstrate that our method consistently outperforms existing sMoE approaches, achieving higher task performance and improved expert utilization balance.

Paperid: 4600, https://arxiv.org/pdf/2510.14792.pdf

Abstract:
Open-vocabulary object detection (OVD) seeks to recognize and localize object categories beyond those seen during training. Recent approaches typically leverage vision-language models (VLMs) to generate pseudo-labels using image-text alignment, allowing detectors to generalize to unseen classes without explicit supervision. However, these methods depend heavily on direct image-text matching, neglecting the intermediate reasoning steps essential for interpreting semantically complex scenes. This results in limited robustness when confronted with crowded or occluded visual contexts. In this paper, we introduce CoT-PL, a new framework that employs structured visual chain-of-thought (CoT) reasoning into the pseudo-labeling process. CoT-PL decomposes object understanding into three interpretable steps: (1) region perception even for unseen objects, (2) category recognition via zero-shot reasoning, and (3) background grounding to separate semantically complex objects. Crucially, the third step naturally motivates our contrastive background learning (CBL) that uses the pre-computed background cues as negatives to promote feature disentanglement between objects and background. In this way, CoT reasoning and CBL form an integrated pipeline tailored to robust pseudo-labeling in crowded or occluded scenes. Notably, in these two settings, our novel-class pseudo-label quality achieves relative improvements of 103.4% and 168.4% over the best prior, respectively. Our extensive experiments demonstrate that CoT-PL achieves +7.7 AP50 on open-vocabulary COCO and +2.9 mask AP on LVIS for novel classes, setting a new state of the art.

Paperid: 4601, https://arxiv.org/pdf/2510.13016.pdf

Abstract:
Understanding fine-grained actions and accurately localizing their corresponding actors in space and time are fundamental capabilities for advancing next-generation AI systems, including embodied agents, autonomous platforms, and human-AI interaction frameworks. Despite recent progress in video understanding, existing methods predominantly address either coarse-grained action recognition or generic object tracking, thereby overlooking the challenge of jointly detecting and tracking multiple objects according to their actions while grounding them temporally. To address this gap, we introduce Spatio-temporal Video Action Grounding (SVAG), a novel task that requires models to simultaneously detect, track, and temporally localize all referent objects in videos based on natural language descriptions of their actions. To support this task, we construct SVAG-Bench, a large-scale benchmark comprising 688 videos, 19,590 annotated records, and 903 unique verbs, covering a diverse range of objects, actions, and real-world scenes. We further propose SVAGFormer, a baseline framework that adapts state of the art vision language models for joint spatial and temporal grounding, and introduce SVAGEval, a standardized evaluation toolkit for fair and reproducible benchmarking. Empirical results show that existing models perform poorly on SVAG, particularly in dense or complex scenes, underscoring the need for more advanced reasoning over fine-grained object-action interactions in long videos.

Paperid: 4602, https://arxiv.org/pdf/2510.12014.pdf

Abstract:
Text--image retrieval is necessary for applications such as product recommendation. Embedding-based approaches like CLIP enable efficient large-scale retrieval via vector similarity search, but they are primarily trained on literal caption-like text--image pairs and often fail to capture abstract or persona-driven attributes common in product recommendation applications (e.g., ``a gift for a mother who loves gardening''). In contrast, state-of-the-art vision--language models (vLLMs) can align text with images in a flexible manner, but their limited context window prevents them from directly handling retrieval over large catalogs. We propose a framework that distills the preference rankings of a powerful vLLM into an embedding-based system, transferring its nuanced alignment abilities while maintaining the inference-time scalability of an embedding-based approach. Experiments on persona-driven product recommendation tasks demonstrate that our method significantly outperforms existing embedding-based baselines, providing an efficient solution for personalized text--image retrieval.

Paperid: 4603, https://arxiv.org/pdf/2510.11631.pdf

Abstract:
Combining large language models with evolutionary computation algorithms represents a promising research direction leveraging the remarkable generative and in-context learning capabilities of LLMs with the strengths of evolutionary algorithms. In this work, we present EvoCAD, a method for generating computer-aided design (CAD) objects through their symbolic representations using vision language models and evolutionary optimization. Our method samples multiple CAD objects, which are then optimized using an evolutionary approach with vision language and reasoning language models. We assess our method using GPT-4V and GPT-4o, evaluating it on the CADPrompt benchmark dataset and comparing it to prior methods. Additionally, we introduce two new metrics based on topological properties defined by the Euler characteristic, which capture a form of semantic similarity between 3D objects. Our results demonstrate that EvoCAD outperforms previous approaches on multiple metrics, particularly in generating topologically correct objects, which can be efficiently evaluated using our two novel metrics that complement existing spatial metrics.

Paperid: 4604, https://arxiv.org/pdf/2510.11282.pdf

Abstract:
Accurate spatiotemporal traffic forecasting is a critical prerequisite for proactive resource management in dense urban mobile networks. While Large Language Models (LLMs) have shown promise in time series analysis, they inherently struggle to model the complex spatial dependencies of grid-based traffic data. Effectively extending LLMs to this domain is challenging, as representing the vast amount of information from dense geographical grids can be inefficient and overwhelm the model's context. To address these challenges, we propose ST-Vision-LLM, a novel framework that reframes spatiotemporal forecasting as a vision-language fusion problem. Our approach leverages a Vision-LLM visual encoder to process historical global traffic matrices as image sequences, providing the model with a comprehensive global view to inform cell-level predictions. To overcome the inefficiency of LLMs in handling numerical data, we introduce an efficient encoding scheme that represents floating-point values as single tokens via a specialized vocabulary, coupled with a two-stage numerical alignment fine-tuning process. The model is first trained with Supervised Fine-Tuning (SFT) and then further optimized for predictive accuracy using Group Relative Policy Optimization (GRPO), a memory-efficient reinforcement learning method. Evaluations on real-world mobile traffic datasets demonstrate that ST-Vision-LLM outperforms existing methods by 15.6% in long-term prediction accuracy and exceeds the second-best baseline by over 30.04% in cross-domain few-shot scenarios. Our extensive experiments validate the model's strong generalization capabilities across various data-scarce environments.

Paperid: 4605, https://arxiv.org/pdf/2510.09008.pdf

Abstract:
Large vision-language models (LVLMs), which integrate a vision encoder (VE) with a large language model, have achieved remarkable success across various tasks. However, there are still crucial challenges in LVLMs such as object hallucination, generating descriptions of objects that are not in the input image. Here, we argue that uncertain visual tokens within the VE is a key factor that contributes to object hallucination. Our statistical analysis found that there are positive correlations between visual tokens with high epistemic uncertainty and the occurrence of hallucinations. Furthermore, we show theoretically and empirically that visual tokens in early VE layers that exhibit large representation deviations under small adversarial perturbations indicate high epistemic uncertainty. Based on these findings, we propose a simple yet effective strategy to mitigate object hallucination by modifying the VE only. Our method comprises a proxy method with adversarial perturbations for identifying uncertain visual tokens efficiently and a method to mask these uncertain visual tokens during the self-attention process in the middle layers of the VE, suppressing their influence on visual encoding and thus alleviating hallucinations. Extensive experiments show that our method significantly reduces object hallucinations in LVLMs and can synergistically work with other prior arts.

Paperid: 4606, https://arxiv.org/pdf/2510.08398.pdf

Abstract:
The recent rapid advancement of Text-to-Video (T2V) generation technologies, which are critical to build ``world models'', makes the existing benchmarks increasingly insufficient to evaluate state-of-the-art T2V models. First, current evaluation dimensions, such as per-frame aesthetic quality and temporal consistency, are no longer able to differentiate state-of-the-art T2V models. Second, event-level temporal causality, which not only distinguishes video from other modalities but also constitutes a crucial component of world models, is severely underexplored in existing benchmarks. Third, existing benchmarks lack a systematic assessment of world knowledge, which are essential capabilities for building world models. To address these issues, we introduce VideoVerse, a comprehensive benchmark that focuses on evaluating whether a T2V model could understand complex temporal causality and world knowledge in the real world. We collect representative videos across diverse domains (e.g., natural landscapes, sports, indoor scenes, science fiction, chemical and physical experiments) and extract their event-level descriptions with inherent temporal causality, which are then rewritten into text-to-video prompts by independent annotators. For each prompt, we design a suite of binary evaluation questions from the perspective of dynamic and static properties, with a total of ten carefully defined evaluation dimensions. In total, our VideoVerse comprises 300 carefully curated prompts, involving 815 events and 793 binary evaluation questions. Consequently, a human preference aligned QA-based evaluation pipeline is developed by using modern vision-language models. Finally, we perform a systematic evaluation of state-of-the-art open-source and closed-source T2V models on VideoVerse, providing in-depth analysis on how far the current T2V generators are from world models.

Paperid: 4607, https://arxiv.org/pdf/2510.06067.pdf

Abstract:
CAPTCHA, originally designed to distinguish humans from robots, has evolved into a real-world benchmark for assessing the spatial reasoning capabilities of vision-language models. In this work, we first show that step-by-step reasoning is crucial for vision-language models (VLMs) to solve CAPTCHAs, which represent high-difficulty spatial reasoning tasks, and that current commercial vision-language models still struggle with such reasoning. In particular, we observe that most commercial VLMs (e.g., Gemini, Claude, GPT, etc.) fail to effectively solve CAPTCHAs and thus achieve low accuracy (around 21.9 percent). However, our findings indicate that requiring the model to perform step-by-step reasoning before generating the final coordinates can significantly enhance its solving accuracy, underscoring the severity of the gap. To systematically study this issue, we introduce CAPTCHA-X, the first real-world CAPTCHA benchmark with reasoning, covering seven categories of CAPTCHAs (such as Gobang, hCaptcha, etc.) with step-by-step action solutions and grounding annotations. We further define five reasoning-oriented metrics that enable a comprehensive evaluation of models reasoning capabilities. To validate the effectiveness of reasoning, we also propose a general agentic VLM-based framework that incorporates the models inherent reasoning abilities. Our method achieves state-of-the-art performance across five high-difficulty CAPTCHA types, with an average solving accuracy of 83.9 percent, substantially surpassing existing baselines. These results reveal the limitations of current models and highlight the importance of reasoning in advancing visual-spatial challenges in the future.

Paperid: 4608, https://arxiv.org/pdf/2510.03574.pdf

Abstract:
Small Vision-Language Models (VLMs) provide a computationally efficient alternative to larger models, at the cost of weaker generalization abilities and downstream task performance. These shortcomings could be addressed by test-time scaling techniques, but existing methods are typically computationally demanding, contradicting the resource-efficient design goals of small models. To address these limitations, we propose two novel and efficient test-time scaling strategies that leverage the model-internal features rather than external supervision: (i) Test-Time Augmentation (TTAug), which generates multiple augmented inputs and aggregates outputs at the token level without parameter updates, and (ii) Test-Time Adaptation (TTAdapt), which adapts model parameters during inference using consensus-based pseudolabels from TTAug. Through extensive experiments across nine benchmarks, we demonstrate consistent performance improvements while maintaining computational efficiency suitable for resource-constrained environments. The generality of our approach is demonstrated both within models at different scales and across different VLMs without additional tuning.

Paperid: 4609, https://arxiv.org/pdf/2510.02608.pdf

Abstract:
Foundation models (FMs) deployed in real-world tasks such as computer-use agents must integrate diverse modalities. How good are FMs at performing joint reasoning, simultaneously reasoning over multiple modalities, especially when the modalities interact and relate to each other to form cross-modal context? To better understand this problem, we study FMs on cross-modal conflicts: scenarios where conflicting evidence is presented across modalities. This allows us to examine whether FMs prioritize one modality over another or reason jointly to reconcile the conflict. Our experiments reveal that FMs can recognize conflicts in unimodal contexts, composed of a single modality, 90% of the time, but the ratio falls as low as 3% when evidence is split across modalities -- similar observations hold in cross-lingual contexts, composed of multiple languages. We trace this failure to cross-modal attention imbalance, showing that FMs exhibit extreme asymmetry in attention scores, disproportionately prioritizing certain modalities. We show that cross-modal attention imbalance does not go away by simply scaling up multimodal or multilingual datasets blindly, since they lack training examples that explicitly require cross-modal reasoning. We demonstrate that even a simple and scalable method of explicitly combining multiple modalities within each training instance significantly reduces attention imbalance. Reduced attention imbalance directly translates to improved downstream performance on several vision-language benchmarks. Our findings underscore the importance of systematically addressing cross-modal contexts to build reliable foundation models.

Paperid: 4610, https://arxiv.org/pdf/2509.25863.pdf

Abstract:
Prompt learning has emerged as a promising paradigm for adapting pre-trained vision-language models (VLMs) to few-shot whole slide image (WSI) classification by aligning visual features with textual representations, thereby reducing annotation cost and enhancing model generalization. Nevertheless, existing methods typically rely on slide-level prompts and fail to capture the subtype-specific phenotypic variations of histological entities (\emph{e.g.,} nuclei, glands) that are critical for cancer diagnosis. To address this gap, we propose Multi-scale Attribute-enhanced Prompt Learning (\textbf{MAPLE}), a hierarchical framework for few-shot WSI classification that jointly integrates multi-scale visual semantics and performs prediction at both the entity and slide levels. Specifically, we first leverage large language models (LLMs) to generate entity-level prompts that can help identify multi-scale histological entities and their phenotypic attributes, as well as slide-level prompts to capture global visual descriptions. Then, an entity-guided cross-attention module is proposed to generate entity-level features, followed by aligning with their corresponding subtype-specific attributes for fine-grained entity-level prediction. To enrich entity representations, we further develop a cross-scale entity graph learning module that can update these representations by capturing their semantic correlations within and across scales. The refined representations are then aggregated into a slide-level representation and aligned with the corresponding prompts for slide-level prediction. Finally, we combine both entity-level and slide-level outputs to produce the final prediction results. Results on three cancer cohorts confirm the effectiveness of our approach in addressing few-shot pathology diagnosis tasks.

Paperid: 4611, https://arxiv.org/pdf/2509.25746.pdf

Abstract:
Despite progress in both traditional dexterous grasping pipelines and recent Vision-Language-Action (VLA) approaches, the grasp execution stage remains prone to pose inaccuracies, especially in long-horizon tasks, which undermines overall performance. To address this "last-mile" challenge, we propose TacRefineNet, a tactile-only framework that achieves fine in-hand pose refinement of known objects in arbitrary target poses using multi-finger fingertip sensing. Our method iteratively adjusts the end-effector pose based on tactile feedback, aligning the object to the desired configuration. We design a multi-branch policy network that fuses tactile inputs from multiple fingers along with proprioception to predict precise control updates. To train this policy, we combine large-scale simulated data from a physics-based tactile model in MuJoCo with real-world data collected from a physical system. Comparative experiments show that pretraining on simulated data and fine-tuning with a small amount of real data significantly improves performance over simulation-only training. Extensive real-world experiments validate the effectiveness of the method, achieving millimeter-level grasp accuracy using only tactile input. To our knowledge, this is the first method to enable arbitrary in-hand pose refinement via multi-finger tactile sensing alone. Project website is available at https://sites.google.com/view/tacrefinenet

Paperid: 4612, https://arxiv.org/pdf/2509.25568.pdf

Abstract:
Vision-language models are increasingly used to generate image captions in specific styles, such as humor or romantic. However, these transformer-based models often struggle with this subjective task in a zero-shot setting. While preference data can be used to align them toward a desired style, such data is expensive to acquire, limiting the ability to explore the models' full capabilities. This work addresses this by studying the data efficiency of aligning small vision-language models to humor and romantic styles. This approach helps to define the performance limits of these models and determine how little preference data is needed to achieve stylistic saturation, benchmarking their capabilities and limitations.

Paperid: 4613, https://arxiv.org/pdf/2509.22737.pdf

Abstract:
We introduce CompareBench, a benchmark for evaluating visual comparison reasoning in vision-language models (VLMs), a fundamental yet understudied skill. CompareBench consists of 1000 QA pairs across four tasks: quantity (600), temporal (100), geometric (200), and spatial (100). It is derived from two auxiliary datasets that we constructed: TallyBench (2000 counting images with QA) and HistCaps (515 historical images with bilingual captions). We evaluate both closed-source APIs (OpenAI, Gemini, Claude) and open-source models (Qwen2.5-VL and Qwen3-VL series). Results show clear scaling trends but also reveal critical limitations: even the strongest models consistently fail at temporal ordering and spatial relations, and they often make mistakes in basic counting and geometric comparisons that are trivial for humans. These findings demonstrate that visual comparison remains a systematic blind spot for current VLMs. By providing controlled, diverse, and diagnostic evaluation, CompareBench establishes a foundation for advancing more reliable multimodal reasoning.

Paperid: 4614, https://arxiv.org/pdf/2509.20279.pdf

Abstract:
Agentic AI is rapidly advancing in healthcare and biomedical research. However, in medical image analysis, their performance and adoption remain limited due to the lack of a robust ecosystem, insufficient toolsets, and the absence of real-time interactive expert feedback. Here we present "TissueLab", a co-evolving agentic AI system that allows researchers to ask direct questions, automatically plan and generate explainable workflows, and conduct real-time analyses where experts can visualize intermediate results and refine them. TissueLab integrates tool factories across pathology, radiology, and spatial omics domains. By standardizing inputs, outputs, and capabilities of diverse tools, the system determines when and how to invoke them to address research and clinical questions. Across diverse tasks with clinically meaningful quantifications that inform staging, prognosis, and treatment planning, TissueLab achieves state-of-the-art performance compared with end-to-end vision-language models (VLMs) and other agentic AI systems such as GPT-5. Moreover, TissueLab continuously learns from clinicians, evolving toward improved classifiers and more effective decision strategies. With active learning, it delivers accurate results in unseen disease contexts within minutes, without requiring massive datasets or prolonged retraining. Released as a sustainable open-source ecosystem, TissueLab aims to accelerate computational research and translational adoption in medical imaging while establishing a foundation for the next generation of medical AI.

Paperid: 4615, https://arxiv.org/pdf/2509.19595.pdf

Abstract:
The embodiment of emotional reactions from body parts contains rich information about our affective experiences. We propose a framework that utilizes state-of-the-art large vision-language models (LVLMs) to generate Embodied LVLM Emotion Narratives (ELENA). These are well-defined, multi-layered text outputs, primarily comprising descriptions that focus on the salient body parts involved in emotional reactions. We also employ attention maps and observe that contemporary models exhibit a persistent bias towards the facial region. Despite this limitation, we observe that our employed framework can effectively recognize embodied emotions in face-masked images, outperforming baselines without any fine-tuning. ELENA opens a new trajectory for embodied emotion analysis across the modality of vision and enriches modeling in an affect-aware setting.

Paperid: 4616, https://arxiv.org/pdf/2509.17336.pdf

Abstract:
Graphical user interfaces (GUIs) are the primary medium for human-computer interaction, yet automating GUI interactions remains challenging due to the complexity of visual elements, dynamic environments, and the need for multi-step reasoning. Existing methods based on vision-language models (VLMs) often suffer from limited resolution, domain mismatch, and insufficient sequential decisionmaking capability. To address these issues, we propose Mano, a robust GUI agent built upon a multi-modal foundation model pre-trained on extensive web and computer system data. Our approach integrates a novel simulated environment for high-fidelity data generation, a three-stage training pipeline (supervised fine-tuning, offline reinforcement learning, and online reinforcement learning), and a verification module for error recovery. Mano demonstrates state-of-the-art performance on multiple GUI benchmarks, including Mind2Web and OSWorld, achieving significant improvements in success rate and operational accuracy. Our work provides new insights into the effective integration of reinforcement learning with VLMs for practical GUI agent deployment, highlighting the importance of domain-specific data, iterative training, and holistic reward design.

Paperid: 4617, https://arxiv.org/pdf/2509.13572.pdf

Abstract:
This study examines the potential of utilizing Vision Language Models (VLMs) to improve the perceptual capabilities of semi-autonomous prosthetic hands. We introduce a unified benchmark for end-to-end perception and grasp inference, evaluating a single VLM to perform tasks that traditionally require complex pipelines with separate modules for object detection, pose estimation, and grasp planning. To establish the feasibility and current limitations of this approach, we benchmark eight contemporary VLMs on their ability to perform a unified task essential for bionic grasping. From a single static image, they should (1) identify common objects and their key properties (name, shape, orientation, and dimensions), and (2) infer appropriate grasp parameters (grasp type, wrist rotation, hand aperture, and number of fingers). A corresponding prompt requesting a structured JSON output was employed with a dataset of 34 snapshots of common objects. Key performance metrics, including accuracy for categorical attributes (e.g., object name, shape) and errors in numerical estimates (e.g., dimensions, hand aperture), along with latency and cost, were analyzed. The results demonstrated that most models exhibited high performance in object identification and shape recognition, while accuracy in estimating dimensions and inferring optimal grasp parameters, particularly hand rotation and aperture, varied more significantly. This work highlights the current capabilities and limitations of VLMs as advanced perceptual modules for semi-autonomous control of bionic limbs, demonstrating their potential for effective prosthetic applications.

Paperid: 4618, https://arxiv.org/pdf/2509.13270.pdf

Authors:Mohammed Baharoon, Siavash Raissi, John S. Jun, Thibault Heintz, Mahmoud Alabbad, Ali Alburkani, Sung Eun Kim, Kent Kleinschmidt, Abdulrahman O. Alhumaydhi, Mohannad Mohammed G. Alghamdi, Jeremy Francis Palacio, Mohammed Bukhaytan, Noah Michael Prudlo, Rithvik Akula, Brady Chrisler, Benjamin Galligos, Mohammed O. Almutairi, Mazeen Mohammed Alanazi, Nasser M. Alrashdi, Joel Jihwan Hwang, Sri Sai Dinesh Jaliparthi, Luke David Nelson, Nathaniel Nguyen, Sathvik Suryadevara, Steven Kim, Mohammed F. Mohammed, Yevgeniy R. Semenov, Kun-Hsing Yu, Abdulrhman Aljouie, Hassan AlOmaish, Adam Rodman, Pranav Rajpurkar

Abstract:
We introduce RadGame, an AI-powered gamified platform for radiology education that targets two core skills: localizing findings and generating reports. Traditional radiology training is based on passive exposure to cases or active practice with real-time input from supervising radiologists, limiting opportunities for immediate and scalable feedback. RadGame addresses this gap by combining gamification with large-scale public datasets and automated, AI-driven feedback that provides clear, structured guidance to human learners. In RadGame Localize, players draw bounding boxes around abnormalities, which are automatically compared to radiologist-drawn annotations from public datasets, and visual explanations are generated by vision-language models for user missed findings. In RadGame Report, players compose findings given a chest X-ray, patient age and indication, and receive structured AI feedback based on radiology report generation metrics, highlighting errors and omissions compared to a radiologist's written ground truth report from public datasets, producing a final performance and style score. In a prospective evaluation, participants using RadGame achieved a 68% improvement in localization accuracy compared to 17% with traditional passive methods and a 31% improvement in report-writing accuracy compared to 4% with traditional methods after seeing the same cases. RadGame highlights the potential of AI-driven gamification to deliver scalable, feedback-rich radiology training and reimagines the application of medical AI resources in education.

Paperid: 4619, https://arxiv.org/pdf/2509.08461.pdf

Abstract:
Recent advances in Large Language Models (LLMs) have demonstrated their remarkable capacity to process and reason over structured and unstructured data modalities beyond natural language. In this work, we explore the applications of Vision Language Models (VLMs), specifically a fine-tuned variant of LLaMa 3.2, to the task of identifying neutrino interactions in pixelated detector data from high-energy physics (HEP) experiments. We benchmark this model against a state-of-the-art convolutional neural network (CNN) architecture, similar to those used in the NOvA and DUNE experiments, which have achieved high efficiency and purity in classifying electron and muon neutrino events. Our evaluation considers both the classification performance and interpretability of the model predictions. We find that VLMs can outperform CNNs, while also providing greater flexibility in integrating auxiliary textual or semantic information and offering more interpretable, reasoning-based predictions. This work highlights the potential of VLMs as a general-purpose backbone for physics event classification, due to their high performance, interpretability, and generalizability, which opens new avenues for integrating multimodal reasoning in experimental neutrino physics.

Paperid: 4620, https://arxiv.org/pdf/2509.07463.pdf

Abstract:
Ensuring reliable robot operation when visual input is degraded or insufficient remains a central challenge in robotics. This letter introduces DepthVision, a framework for multimodal scene understanding designed to address this problem. Unlike existing Vision-Language Models (VLMs), which use only camera-based visual input alongside language, DepthVision synthesizes RGB images from sparse LiDAR point clouds using a conditional generative adversarial network (GAN) with an integrated refiner network. These synthetic views are then combined with real RGB data using a Luminance-Aware Modality Adaptation (LAMA), which blends the two types of data dynamically based on ambient lighting conditions. This approach compensates for sensor degradation, such as darkness or motion blur, without requiring any fine-tuning of downstream vision-language models. We evaluate DepthVision on real and simulated datasets across various models and tasks, with particular attention to safety-critical tasks. The results demonstrate that our approach improves performance in low-light conditions, achieving substantial gains over RGB-only baselines while preserving compatibility with frozen VLMs. This work highlights the potential of LiDAR-guided RGB synthesis for achieving robust robot operation in real-world environments.

Paperid: 4621, https://arxiv.org/pdf/2509.05578.pdf

Abstract:
Multimodal large language models (MLLMs) have shown strong vision-language reasoning abilities but still lack robust 3D spatial understanding, which is critical for autonomous driving. This limitation stems from two key challenges: (1) the difficulty of constructing accessible yet effective 3D representations without expensive manual annotations, and (2) the loss of fine-grained spatial details in VLMs due to the absence of large-scale 3D vision-language pretraining. To address these challenges, we propose OccVLA, a novel framework that integrates 3D occupancy representations into a unified multimodal reasoning process. Unlike prior approaches that rely on explicit 3D inputs, OccVLA treats dense 3D occupancy as both a predictive output and a supervisory signal, enabling the model to learn fine-grained spatial structures directly from 2D visual inputs. The occupancy predictions are regarded as implicit reasoning processes and can be skipped during inference without performance degradation, thereby adding no extra computational overhead. OccVLA achieves state-of-the-art results on the nuScenes benchmark for trajectory planning and demonstrates superior performance on 3D visual question-answering tasks, offering a scalable, interpretable, and fully vision-based solution for autonomous driving.

Paperid: 4622, https://arxiv.org/pdf/2509.00351.pdf

Abstract:
Deep models trained on a single source domain often fail catastrophically under distribution shifts, a critical challenge in Single Domain Generalization (SDG). While existing methods focus on augmenting source data or learning invariant features, they neglect a readily available resource: textual descriptions of the target deployment environment. We propose Target-Oriented Single Domain Generalization (TO-SDG), a novel problem setup that leverages the textual description of the target domain, without requiring any target data, to guide model generalization. To address TO-SDG, we introduce Spectral TARget Alignment (STAR), a lightweight module that injects target semantics into source features by exploiting visual-language models (VLMs) such as CLIP. STAR uses a target-anchored subspace derived from the text embedding of the target description to recenter image features toward the deployment domain, then utilizes spectral projection to retain directions aligned with target cues while discarding source-specific noise. Moreover, we use a vision-language distillation to align backbone features with VLM's semantic geometry. STAR further employs feature-space Mixup to ensure smooth transitions between source and target-oriented representations. Experiments across various image classification and object detection benchmarks demonstrate STAR's superiority. This work establishes that minimal textual metadata, which is a practical and often overlooked resource, significantly enhances generalization under severe data constraints, opening new avenues for deploying robust models in target environments with unseen data.

Paperid: 4623, https://arxiv.org/pdf/2508.19376.pdf

Abstract:
Recent progress in large language models (LLMs) has shown strong potential for multimodal reasoning beyond natural language. In this work, we explore the use of a fine-tuned Vision-Language Model (VLM), based on LLaMA 3.2, for classifying neutrino interactions from pixelated detector images in high-energy physics (HEP) experiments. We benchmark its performance against an established CNN baseline used in experiments like NOvA and DUNE, evaluating metrics such as classification accuracy, precision, recall, and AUC-ROC. Our results show that the VLM not only matches or exceeds CNN performance but also enables richer reasoning and better integration of auxiliary textual or semantic context. These findings suggest that VLMs offer a promising general-purpose backbone for event classification in HEP, paving the way for multimodal approaches in experimental neutrino physics.

Paperid: 4624, https://arxiv.org/pdf/2508.17827.pdf

Abstract:
Industrial and medical anomaly detection faces critical challenges from data scarcity and prohibitive annotation costs, particularly in evolving manufacturing and healthcare settings. To address this, we propose CoZAD, a novel zero-shot anomaly detection framework that integrates soft confident learning with meta-learning and contrastive feature representation. Unlike traditional confident learning that discards uncertain samples, our method assigns confidence-based weights to all training data, preserving boundary information while emphasizing prototypical normal patterns. The framework quantifies data uncertainty through IQR-based thresholding and model uncertainty via covariance based regularization within a Model-Agnostic Meta-Learning. Contrastive learning creates discriminative feature spaces where normal patterns form compact clusters, enabling rapid domain adaptation. Comprehensive evaluation across 10 datasets spanning industrial and medical domains demonstrates state-of-the-art performance, outperforming existing methods on 6 out of 7 industrial benchmarks with notable improvements on texture-rich datasets (99.2% I-AUROC on DTD-Synthetic, 97.2% on BTAD) and pixellevel localization (96.3% P-AUROC on MVTec-AD). The framework eliminates dependence on vision-language alignments or model ensembles, making it valuable for resourceconstrained environments requiring rapid deployment.

Paperid: 4625, https://arxiv.org/pdf/2508.16661.pdf

Abstract:
Image-based quality assessment (QA) in additive manufacturing (AM) often relies heavily on the expertise and constant attention of skilled human operators. While machine learning and deep learning methods have been introduced to assist in this task, they typically provide black-box outputs without interpretable justifications, limiting their trust and adoption in real-world settings. In this work, we introduce a novel QA-VLM framework that leverages the attention mechanisms and reasoning capabilities of vision-language models (VLMs), enriched with application-specific knowledge distilled from peer-reviewed journal articles, to generate human-interpretable quality assessments. Evaluated on 24 single-bead samples produced by laser wire direct energy deposition (DED-LW), our framework demonstrates higher validity and consistency in explanation quality than off-the-shelf VLMs. These results highlight the potential of our approach to enable trustworthy, interpretable quality assessment in AM applications.

Paperid: 4626, https://arxiv.org/pdf/2508.15930.pdf

Abstract:
Ship detection in remote sensing imagery is a critical task with wide-ranging applications, such as maritime activity monitoring, shipping logistics, and environmental studies. However, existing methods often struggle to capture fine-grained semantic information, limiting their effectiveness in complex scenarios. To address these challenges, we propose a novel detection framework that combines Vision-Language Models (VLMs) with a multi-scale adaptive sliding window strategy. To facilitate Semantic-Aware Ship Detection (SASD), we introduce ShipSem-VL, a specialized Vision-Language dataset designed to capture fine-grained ship attributes. We evaluate our framework through three well-defined tasks, providing a comprehensive analysis of its performance and demonstrating its effectiveness in advancing SASD from multiple perspectives.

Paperid: 4627, https://arxiv.org/pdf/2508.15036.pdf

Abstract:
The transformer architecture has become a cornerstone of modern AI, fueling remarkable progress across applications in natural language processing, computer vision, and multimodal learning. As these models continue to scale explosively for performance, implementation efficiency remains a critical challenge. Mixture of Experts (MoE) architectures, selectively activating specialized subnetworks (experts), offer a unique balance between model accuracy and computational cost. However, the adaptive routing in MoE architectures, where input tokens are dynamically directed to specialized experts based on their semantic meaning inadvertently opens up a new attack surface for privacy breaches. These input-dependent activation patterns leave distinctive temporal and spatial traces in hardware execution, which adversaries could exploit to deduce sensitive user data. In this work, we propose MoEcho, discovering a side channel analysis based attack surface that compromises user privacy on MoE based systems. Specifically, in MoEcho, we introduce four novel architectural side channels on different computing platforms, including Cache Occupancy Channels and Pageout+Reload on CPUs, and Performance Counter and TLB Evict+Reload on GPUs, respectively. Exploiting these vulnerabilities, we propose four attacks that effectively breach user privacy in large language models (LLMs) and vision language models (VLMs) based on MoE architectures: Prompt Inference Attack, Response Reconstruction Attack, Visual Inference Attack, and Visual Reconstruction Attack. MoEcho is the first runtime architecture level security analysis of the popular MoE structure common in modern transformers, highlighting a serious security and privacy threat and calling for effective and timely safeguards when harnessing MoE based models for developing efficient large scale AI services.

Paperid: 4628, https://arxiv.org/pdf/2508.12473.pdf

Abstract:
Accurate assessment of neuromuscular reflexes, such as the H-reflex, plays a critical role in sports science, rehabilitation, and clinical neurology. Traditional analysis of H-reflex EMG waveforms is subject to variability and interpretation bias among clinicians and researchers, limiting reliability and standardization. To address these challenges, we propose a Fine-Tuned Vision-Language Model (VLM) Consortium and a reasoning Large-Language Model (LLM)-enabled Decision Support System for automated H-reflex waveform interpretation and diagnosis. Our approach leverages multiple VLMs, each fine-tuned on curated datasets of H-reflex EMG waveform images annotated with clinical observations, recovery timelines, and athlete metadata. These models are capable of extracting key electrophysiological features and predicting neuromuscular states, including fatigue, injury, and recovery, directly from EMG images and contextual metadata. Diagnostic outputs from the VLM consortium are aggregated using a consensus-based method and refined by a specialized reasoning LLM, which ensures robust, transparent, and explainable decision support for clinicians and sports scientists. The end-to-end platform orchestrates seamless communication between the VLM ensemble and the reasoning LLM, integrating prompt engineering strategies and automated reasoning workflows using LLM Agents. Experimental results demonstrate that this hybrid system delivers highly accurate, consistent, and interpretable H-reflex assessments, significantly advancing the automation and standardization of neuromuscular diagnostics. To our knowledge, this work represents the first integration of a fine-tuned VLM consortium with a reasoning LLM for image-based H-reflex analysis, laying the foundation for next-generation AI-assisted neuromuscular assessment and athlete monitoring platforms.

Paperid: 4629, https://arxiv.org/pdf/2508.08604.pdf

Abstract:
Vision-Language Models (VLMs) have been widely used in various visual recognition tasks due to their remarkable generalization capabilities. As these models grow in size and complexity, fine-tuning becomes costly, emphasizing the need to reuse adaptation knowledge from 'weaker' models to efficiently enhance 'stronger' ones. However, existing adaptation transfer methods exhibit limited transferability across models due to their model-specific design and high computational demands. To tackle this, we propose Transferable Model-agnostic adapter (TransMiter), a light-weight adapter that improves vision-language models 'without backpropagation'. TransMiter captures the knowledge gap between pre-trained and fine-tuned VLMs, in an 'unsupervised' manner. Once trained, this knowledge can be seamlessly transferred across different models without the need for backpropagation. Moreover, TransMiter consists of only a few layers, inducing a negligible additional inference cost. Notably, supplementing the process with a few labeled data further yields additional performance gain, often surpassing a fine-tuned stronger model, with a marginal training cost. Experimental results and analyses demonstrate that TransMiter effectively and efficiently transfers adaptation knowledge while preserving generalization abilities across VLMs of different sizes and architectures in visual recognition tasks.

Paperid: 4630, https://arxiv.org/pdf/2508.08040.pdf

Abstract:
Prompt-based tuning has emerged as a lightweight alternative to full fine-tuning in large vision-language models, enabling efficient adaptation via learned contextual prompts. This paradigm has recently been extended to federated learning settings (e.g., PromptFL), where clients collaboratively train prompts under data privacy constraints. However, the security implications of prompt-based aggregation in federated multimodal learning remain largely unexplored, leaving a critical attack surface unaddressed. In this paper, we introduce \textbf{BadPromptFL}, the first backdoor attack targeting prompt-based federated learning in multimodal contrastive models. In BadPromptFL, compromised clients jointly optimize local backdoor triggers and prompt embeddings, injecting poisoned prompts into the global aggregation process. These prompts are then propagated to benign clients, enabling universal backdoor activation at inference without modifying model parameters. Leveraging the contextual learning behavior of CLIP-style architectures, BadPromptFL achieves high attack success rates (e.g., $>90\%$) with minimal visibility and limited client participation. Extensive experiments across multiple datasets and aggregation protocols validate the effectiveness, stealth, and generalizability of our attack, raising critical concerns about the robustness of prompt-based federated learning in real-world deployments.

Paperid: 4631, https://arxiv.org/pdf/2508.04650.pdf

Abstract:
Multimodal vision-language models (VLMs) continue to achieve ever-improving scores on chart understanding benchmarks. Yet, we find that this progress does not fully capture the breadth of visual reasoning capabilities essential for interpreting charts. We introduce EncQA, a novel benchmark informed by the visualization literature, designed to provide systematic coverage of visual encodings and analytic tasks that are crucial for chart understanding. EncQA provides 2,076 synthetic question-answer pairs, enabling balanced coverage of six visual encoding channels (position, length, area, color quantitative, color nominal, and shape) and eight tasks (find extrema, retrieve value, find anomaly, filter values, compute derived value exact, compute derived value relative, correlate values, and correlate values relative). Our evaluation of 9 state-of-the-art VLMs reveals that performance varies significantly across encodings within the same task, as well as across tasks. Contrary to expectations, we observe that performance does not improve with model size for many task-encoding pairs. Our results suggest that advancing chart understanding requires targeted strategies addressing specific visual reasoning gaps, rather than solely scaling up model or dataset size.

Paperid: 4632, https://arxiv.org/pdf/2508.03177.pdf

Abstract:
Large Vision-Language Models (LVLMs) recently achieve significant breakthroughs in understanding complex visual-textual contexts. However, hallucination issues still limit their real-world applicability. Although previous mitigation methods effectively reduce hallucinations in photographic images, they largely overlook the potential risks posed by stylized images, which play crucial roles in critical scenarios such as game scene understanding, art education, and medical analysis. In this work, we first construct a dataset comprising photographic images and their corresponding stylized versions with carefully annotated caption labels. We then conduct head-to-head comparisons on both discriminative and generative tasks by benchmarking 13 advanced LVLMs on the collected datasets. Our findings reveal that stylized images tend to induce significantly more hallucinations than their photographic counterparts. To address this issue, we propose Style-Aware Visual Early Revision SAVER, a novel mechanism that dynamically adjusts LVLMs' final outputs based on the token-level visual attention patterns, leveraging early-layer feedback to mitigate hallucinations caused by stylized images. Extensive experiments demonstrate that SAVER achieves state-of-the-art performance in hallucination mitigation across various models, datasets, and tasks.

Paperid: 4633, https://arxiv.org/pdf/2507.22431.pdf

Abstract:
Large-scale but noisy image-text pair data have paved the way for the success of Contrastive Language-Image Pretraining (CLIP). As the foundation vision encoder, CLIP in turn serves as the cornerstone for most large vision-language models (LVLMs). This interdependence naturally raises an interesting question: Can we reciprocally leverage LVLMs to enhance the quality of image-text pair data, thereby opening the possibility of a self-reinforcing cycle for continuous improvement? In this work, we take a significant step toward this vision by introducing an LVLM-driven data refinement pipeline. Our framework leverages LVLMs to process images and their raw alt-text, generating four complementary textual formulas: long positive descriptions, long negative descriptions, short positive tags, and short negative tags. Applying this pipeline to the curated DFN-Large dataset yields VLM-150M, a refined dataset enriched with multi-grained annotations. Based on this dataset, we further propose a training paradigm that extends conventional contrastive learning by incorporating negative descriptions and short tags as additional supervised signals. The resulting model, namely HQ-CLIP, demonstrates remarkable improvements across diverse benchmarks. Within a comparable training data scale, our approach achieves state-of-the-art performance in zero-shot classification, cross-modal retrieval, and fine-grained visual understanding tasks. In retrieval benchmarks, HQ-CLIP even surpasses standard CLIP models trained on the DFN-2B dataset, which contains 10$\times$ more training data than ours. All code, data, and models are available at https://zxwei.site/hqclip.

Paperid: 4634, https://arxiv.org/pdf/2507.16466.pdf

Abstract:
In data-driven storytelling contexts such as data journalism and data videos, data visualizations are often presented alongside real-world imagery to support narrative context. However, these visualizations and contextual images typically remain separated, limiting their combined narrative expressiveness and engagement. Achieving this is challenging due to the need for fine-grained alignment and creative ideation. To address this, we present SceneLoom, a Vision-Language Model (VLM)-powered system that facilitates the coordination of data visualization with real-world imagery based on narrative intents. Through a formative study, we investigated the design space of coordination relationships between data visualization and real-world scenes from the perspectives of visual alignment and semantic coherence. Guided by the derived design considerations, SceneLoom leverages VLMs to extract visual and semantic features from scene images and data visualization, and perform design mapping through a reasoning process that incorporates spatial organization, shape similarity, layout consistency, and semantic binding. The system generates a set of contextually expressive, image-driven design alternatives that achieve coherent alignments across visual, semantic, and data dimensions. Users can explore these alternatives, select preferred mappings, and further refine the design through interactive adjustments and animated transitions to support expressive data communication. A user study and an example gallery validate SceneLoom's effectiveness in inspiring creative design and facilitating design externalization.

Paperid: 4635, https://arxiv.org/pdf/2507.07685.pdf

Abstract:
Large vision-language models (LVLMs) have demonstrated remarkable capabilities by integrating pre-trained vision encoders with large language models (LLMs). Similar to single-modal LLMs, chain-of-thought (CoT) prompting has been adapted for LVLMs to enhance multi-modal reasoning by generating intermediate rationales based on visual and textual inputs. While CoT is assumed to improve grounding and accuracy in LVLMs, our experiments reveal a key challenge: existing LVLMs often ignore the contents of generated rationales in CoT reasoning. To address this, we re-formulate multi-modal CoT reasoning as a KL-constrained reward maximization focused on rationale-conditional log-likelihood. As the optimal solution, we propose rationale-enhanced decoding (RED), a novel plug-and-play inference-time decoding strategy. RED harmonizes visual and rationale information by multiplying distinct image-conditional and rationale-conditional next token distributions. Extensive experiments show that RED consistently and significantly improves reasoning over standard CoT and other decoding methods across multiple benchmarks and LVLMs. Our work offers a practical and effective approach to improve both the faithfulness and accuracy of CoT reasoning in LVLMs, paving the way for more reliable rationale-grounded multi-modal systems.

Paperid: 4636, https://arxiv.org/pdf/2506.21041.pdf

Abstract:
Autonomous driving technologies face significant safety challenges while operating under rare, diverse, and visually degraded weather scenarios. These challenges become more critical in cooperative settings, where vehicles and infrastructure jointly perceive and reason across complex environments. To address these issues, we propose SEAL, a vision-language model-based framework with adaptive multimodal learning for robust cooperative autonomous driving under long-tail scenarios. SEAL introduces three core innovations: (i) a prompt-driven long-tail scenario generation and evaluation pipeline that leverages foundation models to synthesize realistic long-tail conditions such as snow and fog across vehicle- and infrastructure-side views, enriching training diversity efficiently; (ii) a gated multi-scenario adaptive attention module that modulates the visual stream using scenario priors to recalibrate ambiguous or corrupted features; and (iii) a multi-task scenario-aware contrastive learning objective that improves multimodal alignment and promotes cross-scenario feature separability. Extensive experiments demonstrate that SEAL significantly outperforms existing baselines in reasoning, safety, and planning accuracy under complex, challenging driving conditions, advancing the safety, robustness, and scalability of autonomous driving.

Paperid: 4637, https://arxiv.org/pdf/2506.15903.pdf

Abstract:
We introduce a large-scale dataset for instruction-guided vector image editing, consisting of over 270,000 pairs of SVG images paired with natural language edit instructions. Our dataset enables training and evaluation of models that modify vector graphics based on textual commands. We describe the data collection process, including image pairing via CLIP similarity and instruction generation with vision-language models. Initial experiments with state-of-the-art large language models reveal that current methods struggle to produce accurate and valid edits, underscoring the challenge of this task. To foster research in natural language-driven vector graphic generation and editing, we make our resources created within this work publicly available.

Paperid: 4638, https://arxiv.org/pdf/2506.12609.pdf

Abstract:
Large vision-language models (LVLMs) have demonstrated impressive capabilities across diverse multimodal tasks, yet they remain highly susceptible to visual hallucinations (VH), often producing confident but inaccurate descriptions of visual content. Building on the insight that not all tokens and attention heads contribute equally to VH mitigation, we introduce VisFlow, a lightweight and training-free framework that alleviates hallucinations by directly modulating attention patterns during inference. To address two primary challenges of VH, namely insufficient visual attention and the dominance of language priors, we identify three problematic attention behaviors in LVLMs: (1) disproportionate allocation of attention to uninformative or trailing visual tokens, (2) over-dependence on the previously generated token, and (3) excessive fixation on system prompts that hinders multimodal integration. To overcome these issues, VisFlow introduces a dual-level Attention Intervention, consisting of Token-level Attention Intervention (TAI), which reinforces attention to salient visual regions, and Head-level Attention Intervention (HAI), which suppresses undue focus on system prompts and adjacent text tokens. Together, these interventions strengthen visual alignment while reducing linguistic bias. Extensive experiments across diverse models and benchmarks demonstrate that VisFlow effectively mitigates hallucinations with minimal computational overhead.

Paperid: 4639, https://arxiv.org/pdf/2506.10756.pdf

Abstract:
Vision-and-language navigation (VLN) is a long-standing challenge in autonomous robotics, aiming to empower agents with the ability to follow human instructions while navigating complex environments. Two key bottlenecks remain in this field: generalization to out-of-distribution environments and reliance on fixed discrete action spaces. To address these challenges, we propose Vision-Language Fly (VLFly), a framework tailored for Unmanned Aerial Vehicles (UAVs) to execute language-guided flight. Without the requirement for localization or active ranging sensors, VLFly outputs continuous velocity commands purely from egocentric observations captured by an onboard monocular camera. The VLFly integrates three modules: an instruction encoder based on a large language model (LLM) that reformulates high-level language into structured prompts, a goal retriever powered by a vision-language model (VLM) that matches these prompts to goal images via vision-language similarity, and a waypoint planner that generates executable trajectories for real-time UAV control. VLFly is evaluated across diverse simulation environments without additional fine-tuning and consistently outperforms all baselines. Moreover, real-world VLN tasks in indoor and outdoor environments under direct and indirect instructions demonstrate that VLFly achieves robust open-vocabulary goal understanding and generalized navigation capabilities, even in the presence of abstract language input.

Paperid: 4640, https://arxiv.org/pdf/2506.10202.pdf

Abstract:
Recent approaches have shown impressive proficiency in extracting and leveraging parametric knowledge from Large-Language Models (LLMs) and Vision-Language Models (VLMs). In this work, we consider how we can improve the identification and retrieval of videos related to complex real-world events by automatically extracting latent parametric knowledge about those events. We present Q2E: a Query-to-Event decomposition method for zero-shot multilingual text-to-video retrieval, adaptable across datasets, domains, LLMs, or VLMs. Our approach demonstrates that we can enhance the understanding of otherwise overly simplified human queries by decomposing the query using the knowledge embedded in LLMs and VLMs. We additionally show how to apply our approach to both visual and speech-based inputs. To combine this varied multimodal knowledge, we adopt entropy-based fusion scoring for zero-shot fusion. Through evaluations on two diverse datasets and multiple retrieval metrics, we demonstrate that Q2E outperforms several state-of-the-art baselines. Our evaluation also shows that integrating audio information can significantly improve text-to-video retrieval. We have released code and data for future research.

Paperid: 4641, https://arxiv.org/pdf/2506.09638.pdf

Abstract:
Vision-Language Models (VLMs) have demonstrated remarkable capabilities in cross-modal understanding and generation by integrating visual and textual information. While instruction tuning and parameter-efficient fine-tuning methods have substantially improved the generalization of VLMs, most existing approaches rely on centralized training, posing challenges for deployment in domains with strict privacy requirements like healthcare. Recent efforts have introduced Federated Learning (FL) into VLM fine-tuning to address these privacy concerns, yet comprehensive benchmarks for evaluating federated fine-tuning strategies, model architectures, and task generalization remain lacking. In this work, we present \textbf{FedVLMBench}, the first systematic benchmark for federated fine-tuning of VLMs. FedVLMBench integrates two mainstream VLM architectures (encoder-based and encoder-free), four fine-tuning strategies, five FL algorithms, six multimodal datasets spanning four cross-domain single-task scenarios and two cross-domain multitask settings, covering four distinct downstream task categories. Through extensive experiments, we uncover key insights into the interplay between VLM architectures, fine-tuning strategies, data heterogeneity, and multi-task federated optimization. Notably, we find that a 2-layer multilayer perceptron (MLP) connector with concurrent connector and LLM tuning emerges as the optimal configuration for encoder-based VLMs in FL. Furthermore, current FL methods exhibit significantly higher sensitivity to data heterogeneity in vision-centric tasks than text-centric ones, across both encoder-free and encoder-based VLM architectures. Our benchmark provides essential tools, datasets, and empirical guidance for the research community, offering a standardized platform to advance privacy-preserving, federated training of multimodal foundation models.

Paperid: 4642, https://arxiv.org/pdf/2506.08399.pdf

Abstract:
Ensuring safe and appropriate responses from vision-language models (VLMs) remains a critical challenge, particularly in high-risk or ambiguous scenarios. We introduce SafeCoT, a lightweight, interpretable framework that leverages rule-based chain-of-thought (CoT) supervision to improve refusal behavior in VLMs. Unlike prior methods that rely on large-scale safety annotations or complex modeling, SafeCoT uses minimal supervision to help models reason about safety risks and make context-aware refusals. Experiments across multiple benchmarks show that SafeCoT significantly reduces overrefusal and enhances generalization, even with limited training data. Our approach offers a scalable solution for aligning VLMs with safety-critical objectives.

Paperid: 4643, https://arxiv.org/pdf/2506.07803.pdf

Abstract:
Vision encoders are increasingly used in modern applications, from vision-only models to multimodal systems such as vision-language models. Despite their remarkable success, it remains unclear how these architectures represent features internally. Here, we propose a novel approach for interpreting vision features via image reconstruction. We compare two related model families, SigLIP and SigLIP2, which differ only in their training objective, and show that encoders pre-trained on image-based tasks retain significantly more image information than those trained on non-image tasks such as contrastive learning. We further apply our method to a range of vision encoders, ranking them by the informativeness of their feature representations. Finally, we demonstrate that manipulating the feature space yields predictable changes in reconstructed images, revealing that orthogonal rotations (rather than spatial transformations) control color encoding. Our approach can be applied to any vision encoder, shedding light on the inner structure of its feature space. The code and model weights to reproduce the experiments are available in GitHub.

Paperid: 4644, https://arxiv.org/pdf/2506.05450.pdf

Abstract:
Image degradation is a prevalent issue in various real-world applications, affecting visual quality and downstream processing tasks. In this study, we propose a novel framework that employs a Vision-Language Model (VLM) to automatically classify degraded images into predefined categories. The VLM categorizes an input image into one of four degradation types: (A) super-resolution degradation (including noise, blur, and JPEG compression), (B) reflection artifacts, (C) motion blur, or (D) no visible degradation (high-quality image). Once classified, images assigned to categories A, B, or C undergo targeted restoration using dedicated models tailored for each specific degradation type. The final output is a restored image with improved visual quality. Experimental results demonstrate the effectiveness of our approach in accurately classifying image degradations and enhancing image quality through specialized restoration models. Our method presents a scalable and automated solution for real-world image enhancement tasks, leveraging the capabilities of VLMs in conjunction with state-of-the-art restoration techniques.

Paperid: 4645, https://arxiv.org/pdf/2506.01274.pdf

Abstract:
Recent progress in Large Multi-modal Models (LMMs) has enabled effective vision-language reasoning, yet the ability to understand video content remains constrained by suboptimal frame selection strategies. Existing approaches often rely on static heuristics or external retrieval modules to feed frame information into video-LLMs, which may fail to provide the query-relevant information. In this work, we introduce ReFoCUS (Reinforcement-guided Frame Optimization for Contextual UnderStanding), a novel frame-level policy optimization framework that shifts the optimization target from textual responses to visual input selection. ReFoCUS learns a frame selection policy via reinforcement learning, using reward signals derived from a reference LMM to reflect the model's intrinsic preferences for frames that best support temporally grounded responses. To efficiently explore the large combinatorial frame space, we employ an autoregressive, conditional selection architecture that ensures temporal coherence while reducing complexity. Our approach does not require explicit supervision at the frame-level and consistently improves reasoning performance across multiple video QA benchmarks, highlighting the benefits of aligning frame selection with model-internal utility.

Paperid: 4646, https://arxiv.org/pdf/2505.18594.pdf

Abstract:
Vision-language retrieval (VLR) has attracted significant attention in both academia and industry, which involves using text (or images) as queries to retrieve corresponding images (or text). However, existing methods often neglect the rich visual semantics knowledge of entities, thus leading to incorrect retrieval results. To address this problem, we propose the Entity Visual Description enhanced CLIP (EvdCLIP), designed to leverage the visual knowledge of entities to enrich queries. Specifically, since humans recognize entities through visual cues, we employ a large language model (LLM) to generate Entity Visual Descriptions (EVDs) as alignment cues to complement textual data. These EVDs are then integrated into raw queries to create visually-rich, EVD-enhanced queries. Furthermore, recognizing that EVD-enhanced queries may introduce noise or low-quality expansions, we develop a novel, trainable EVD-aware Rewriter (EaRW) for vision-language retrieval tasks. EaRW utilizes EVD knowledge and the generative capabilities of the language model to effectively rewrite queries. With our specialized training strategy, EaRW can generate high-quality and low-noise EVD-enhanced queries. Extensive quantitative and qualitative experiments on image-text retrieval benchmarks validate the superiority of EvdCLIP on vision-language retrieval tasks.

Paperid: 4647, https://arxiv.org/pdf/2505.18530.pdf

Abstract:
Medical Large Vision-Language Models (Med-LVLMs) have been widely adopted for medical report generation. Despite Med-LVLMs producing state-of-the-art performance, they exhibit a bias toward predicting all findings as normal, leading to reports that overlook critical abnormalities. Furthermore, these models often fail to provide comprehensive descriptions of radiologically relevant regions necessary for accurate diagnosis. To address these challenges, we proposeMedical Report Generation Agents (MRGAgents), a novel multi-agent framework that fine-tunes specialized agents for different disease categories. By curating subsets of the IU X-ray and MIMIC-CXR datasets to train disease-specific agents, MRGAgents generates reports that more effectively balance normal and abnormal findings while ensuring a comprehensive description of clinically relevant regions. Our experiments demonstrate that MRGAgents outperformed the state-of-the-art, improving both report comprehensiveness and diagnostic utility.

Paperid: 4648, https://arxiv.org/pdf/2505.16104.pdf

Abstract:
With the increasing size of Large Vision-Language Models (LVLMs), network pruning techniques aimed at compressing models for deployment in resource-constrained environments have garnered significant attention. However, we observe that pruning often leads to a degradation in safety performance. To address this issue, we present a novel and lightweight approach, termed Hierarchical Safety Realignment (HSR). HSR operates by first quantifying the contribution of each attention head to safety, identifying the most critical ones, and then selectively restoring neurons directly within these attention heads that play a pivotal role in maintaining safety. This process hierarchically realigns the safety of pruned LVLMs, progressing from the attention head level to the neuron level. We validate HSR across various models and pruning strategies, consistently achieving notable improvements in safety performance. To our knowledge, this is the first work explicitly focused on restoring safety in LVLMs post-pruning.

Paperid: 4649, https://arxiv.org/pdf/2505.14938.pdf

Abstract:
Autonomous robots must reason about the physical consequences of their actions to operate effectively in unstructured, real-world environments. We present Scan, Materialize, Simulate (SMS), a unified framework that combines 3D Gaussian Splatting for accurate scene reconstruction, visual foundation models for semantic segmentation, vision-language models for material property inference, and physics simulation for reliable prediction of action outcomes. By integrating these components, SMS enables generalizable physical reasoning and object-centric planning without the need to re-learn foundational physical dynamics. We empirically validate SMS in a billiards-inspired manipulation task and a challenging quadrotor landing scenario, demonstrating robust performance on both simulated domain transfer and real-world experiments. Our results highlight the potential of bridging differentiable rendering for scene reconstruction, foundation models for semantic understanding, and physics-based simulation to achieve physically grounded robot planning across diverse settings.

Paperid: 4650, https://arxiv.org/pdf/2505.14481.pdf

Abstract:
In the field of urban planning, existing Vision-Language Models (VLMs) frequently fail to effectively analyze and evaluate planning maps, despite the critical importance of these visual elements for urban planners and related educational contexts. Planning maps, which visualize land use, infrastructure layouts, and functional zoning, require specialized understanding of spatial configurations, regulatory requirements, and multi-scale analysis. To address this challenge, we introduce PlanGPT-VL, the first domain-specific Vision-Language Model tailored specifically for urban planning maps. PlanGPT-VL employs three innovative approaches: (1) PlanAnno-V framework for high-quality VQA data synthesis, (2) Critical Point Thinking to reduce hallucinations through structured verification, and (3) comprehensive training methodology combining Supervised Fine-Tuning with frozen vision encoder parameters. Through systematic evaluation on our proposed PlanBench-V benchmark, we demonstrate that PlanGPT-VL significantly outperforms general-purpose state-of-the-art VLMs in specialized planning map interpretation tasks, offering urban planning professionals a reliable tool for map analysis, assessment, and educational applications while maintaining high factual accuracy. Our lightweight 7B parameter model achieves comparable performance to models exceeding 72B parameters, demonstrating efficient domain specialization without sacrificing performance.

Paperid: 4651, https://arxiv.org/pdf/2505.08283.pdf

Abstract:
The performance of Visio-Language Transformers drops sharply when an input modality (e.g., image) is missing, because the model is forced to make predictions using incomplete information. Existing missing-aware prompt methods help reduce this degradation, but they still rely on conventional prediction heads (e.g., a Fully-Connected layer) that compute class scores in the same way regardless of which modality is present or absent. We introduce Decoupled Prototype Learning (DPL), a new prediction head architecture that explicitly adjusts its decision process to the observed input modalities. For each class, DPL selects a set of prototypes specific to the current missing-modality cases (image-missing, text-missing, or mixed-missing). Each prototype is then decomposed into image-specific and text-specific components, enabling the head to make decisions that depend on the information actually present. This adaptive design allows DPL to handle inputs with missing modalities more effectively while remaining fully compatible with existing prompt-based frameworks. Extensive experiments on MM-IMDb, UPMC Food-101, and Hateful Memes demonstrate that DPL outperforms state-of-the-art approaches across all widely used multimodal imag-text datasets and various missing cases.

Paperid: 4652, https://arxiv.org/pdf/2505.07704.pdf

Abstract:
Measuring how real images look is a complex task in artificial intelligence research. For example, an image of a boy with a vacuum cleaner in a desert violates common sense. We introduce a novel method, which we call Through the Looking Glass (TLG), to assess image common sense consistency using Large Vision-Language Models (LVLMs) and Transformer-based encoder. By leveraging LVLMs to extract atomic facts from these images, we obtain a mix of accurate facts. We proceed by fine-tuning a compact attention-pooling classifier over encoded atomic facts. Our TLG has achieved a new state-of-the-art performance on the WHOOPS! and WEIRD datasets while leveraging a compact fine-tuning component.

Paperid: 4653, https://arxiv.org/pdf/2504.18866.pdf

Abstract:
Existing weakly supervised video violence detection (VVD) methods primarily rely on Euclidean representation learning, which often struggles to distinguish visually similar yet semantically distinct events due to limited hierarchical modeling and insufficient ambiguous training samples. To address this challenge, we propose PiercingEye, a novel dual-space learning framework that synergizes Euclidean and hyperbolic geometries to enhance discriminative feature representation. Specifically, PiercingEye introduces a layer-sensitive hyperbolic aggregation strategy with hyperbolic Dirichlet energy constraints to progressively model event hierarchies, and a cross-space attention mechanism to facilitate complementary feature interactions between Euclidean and hyperbolic spaces. Furthermore, to mitigate the scarcity of ambiguous samples, we leverage large language models to generate logic-guided ambiguous event descriptions, enabling explicit supervision through a hyperbolic vision-language contrastive loss that prioritizes high-confusion samples via dynamic similarity-aware weighting. Extensive experiments on XD-Violence and UCF-Crime benchmarks demonstrate that PiercingEye achieves state-of-the-art performance, with particularly strong results on a newly curated ambiguous event subset, validating its superior capability in fine-grained violence detection.

Paperid: 4654, https://arxiv.org/pdf/2504.18671.pdf

Abstract:
Mild Traumatic Brain Injury (TBI) detection presents significant challenges due to the subtle and often ambiguous presentation of symptoms in medical imaging, making accurate diagnosis a complex task. To address these challenges, we propose Proof-of-TBI, a medical diagnosis support system that integrates multiple fine-tuned vision-language models with the OpenAI-o3 reasoning large language model (LLM). Our approach fine-tunes multiple vision-language models using a labeled dataset of TBI MRI scans, training them to diagnose TBI symptoms effectively. The predictions from these models are aggregated through a consensus-based decision-making process. The system evaluates the predictions from all fine-tuned vision language models using the OpenAI-o3 reasoning LLM, a model that has demonstrated remarkable reasoning performance, to produce the most accurate final diagnosis. The LLM Agents orchestrates interactions between the vision-language models and the reasoning LLM, managing the final decision-making process with transparency, reliability, and automation. This end-to-end decision-making workflow combines the vision-language model consortium with the OpenAI-o3 reasoning LLM, enabled by custom prompt engineering by the LLM agents. The prototype for the proposed platform was developed in collaboration with the U.S. Army Medical Research team in Newport News, Virginia, incorporating five fine-tuned vision-language models. The results demonstrate the transformative potential of combining fine-tuned vision-language model inputs with the OpenAI-o3 reasoning LLM to create a robust, secure, and highly accurate diagnostic system for mild TBI prediction. To the best of our knowledge, this research represents the first application of fine-tuned vision-language models integrated with a reasoning LLM for TBI prediction tasks.

Paperid: 4655, https://arxiv.org/pdf/2504.13405.pdf

Abstract:
As the number of individuals in a crowd grows, enumeration-based techniques become increasingly infeasible and their estimates increasingly unreliable. We propose instead an estimation-based version of the problem: we label Rough Crowd Counting that delivers better accuracy on the basis of training data that is easier to acquire. Rough crowd counting requires only rough annotations of the number of targets in an image, instead of the more traditional, and far more expensive, per-target annotations. We propose an approach to the rough crowd counting problem based on CLIP, termed ProgRoCC. Specifically, we introduce a progressive estimation learning strategy that determines the object count through a coarse-to-fine approach. This approach delivers answers quickly, outperforms the state-of-the-art in semi- and weakly-supervised crowd counting. In addition, we design a vision-language matching adapter that optimizes key-value pairs by mining effective matches of two modalities to refine the visual features, thereby improving the final performance. Extensive experimental results on three widely adopted crowd counting datasets demonstrate the effectiveness of our method.

Paperid: 4656, https://arxiv.org/pdf/2504.10568.pdf

Abstract:
We present AgMMU, a challenging real-world benchmark for evaluating and advancing vision-language models (VLMs) in the knowledge-intensive domain of agriculture. Unlike prior datasets that rely on crowdsourced prompts, AgMMU is distilled from 116,231 authentic dialogues between everyday growers and USDA-authorized Cooperative Extension experts. Through a three-stage pipeline: automated knowledge extraction, QA generation, and human verification, we construct (i) AgMMU, an evaluation set of 746 multiple-choice questions (MCQs) and 746 open-ended questions (OEQs), and (ii) AgBase, a development corpus of 57,079 multimodal facts covering five high-stakes agricultural topics: insect identification, species identification, disease categorization, symptom description, and management instruction. Benchmarking 12 leading VLMs reveals pronounced gaps in fine-grained perception and factual grounding. Open-sourced models trail after proprietary ones by a wide margin. Simple fine-tuning on AgBase boosts open-sourced model performance on challenging OEQs for up to 11.6% on average, narrowing this gap and also motivating future research to propose better strategies in knowledge extraction and distillation from AgBase. We hope AgMMU stimulates research on domain-specific knowledge integration and trustworthy decision support in agriculture AI development.

Paperid: 4657, https://arxiv.org/pdf/2504.10090.pdf

Abstract:
Large language models (LLMs) and multimodal large language models (MLLMs) have significantly advanced artificial intelligence. However, visual reasoning, reasoning involving both visual and textual inputs, remains underexplored. Recent advancements, including the reasoning models like OpenAI o1 and Gemini 2.0 Flash Thinking, which incorporate image inputs, have opened this capability. In this ongoing work, we focus specifically on photography-related tasks because a photo is a visual snapshot of the physical world where the underlying physics (i.e., illumination, blur extent, etc.) interplay with the camera parameters. Successfully reasoning from the visual information of a photo to identify these numerical camera settings requires the MLLMs to have a deeper understanding of the underlying physics for precise visual comprehension, representing a challenging and intelligent capability essential for practical applications like photography assistant agents. We aim to evaluate MLLMs on their ability to distinguish visual differences related to numerical camera settings, extending a methodology previously proposed for vision-language models (VLMs). Our preliminary results demonstrate the importance of visual reasoning in photography-related tasks. Moreover, these results show that no single MLLM consistently dominates across all evaluation tasks, demonstrating ongoing challenges and opportunities in developing MLLMs with better visual reasoning.

Paperid: 4658, https://arxiv.org/pdf/2504.09000.pdf

Abstract:
Visual Object Goal Navigation (ObjectNav) requires a robot to locate a target object in an unseen environment using egocentric observations. However, decision-making policies often struggle to transfer to unseen environments and novel target objects, which is the core generalization problem. Traditional end-to-end learning methods exacerbate this issue, as they rely on memorizing spatial patterns rather than employing structured reasoning, limiting their ability to generalize effectively. In this letter, we introduce Closed-Loop Hierarchical Chain-of-Thought Navigation (CL-CoTNav), a vision-language model (VLM)-driven ObjectNav framework that integrates structured reasoning and closed-loop feedback into navigation decision-making. To enhance generalization, we fine-tune a VLM using multi-turn question-answering (QA) data derived from human demonstration trajectories. This structured dataset enables hierarchical Chain-of-Thought (H-CoT) prompting, systematically extracting compositional knowledge to refine perception and decision-making, inspired by the human cognitive process of locating a target object through iterative reasoning steps. Additionally, we propose a Closed-Loop H-CoT mechanism that incorporates detection and reasoning confidence scores into training. This adaptive weighting strategy guides the model to prioritize high-confidence data pairs, mitigating the impact of noisy inputs and enhancing robustness against hallucinated or incorrect reasoning. Extensive experiments in the AI Habitat environment demonstrate CL-CoTNav's superior generalization to unseen scenes and novel object categories. Our method consistently outperforms state-of-the-art approaches in navigation success rate (SR) and success weighted by path length (SPL) by 22.4\%. We release our datasets, models, and supplementary videos on our project page.

Paperid: 4659, https://arxiv.org/pdf/2504.07643.pdf

Abstract:
In this paper, we introduce CollEx, an innovative multimodal agentic Retrieval-Augmented Generation (RAG) system designed to enhance interactive exploration of extensive scientific collections. Given the overwhelming volume and inherent complexity of scientific collections, conventional search systems often lack necessary intuitiveness and interactivity, presenting substantial barriers for learners, educators, and researchers. CollEx addresses these limitations by employing state-of-the-art Large Vision-Language Models (LVLMs) as multimodal agents accessible through an intuitive chat interface. By abstracting complex interactions via specialized agents equipped with advanced tools, CollEx facilitates curiosity-driven exploration, significantly simplifying access to diverse scientific collections and records therein. Our system integrates textual and visual modalities, supporting educational scenarios that are helpful for teachers, pupils, students, and researchers by fostering independent exploration as well as scientific excitement and curiosity. Furthermore, CollEx serves the research community by discovering interdisciplinary connections and complementing visual data. We illustrate the effectiveness of our system through a proof-of-concept application containing over 64,000 unique records across 32 collections from a local scientific collection from a public university.

Paperid: 4660, https://arxiv.org/pdf/2504.02885.pdf

Abstract:
Large vision-language models (LVMs) hold a great promise for automating medical report generation, potentially reducing the burden of manual reporting. State-of-the-art (SOTA) research fine-tunes general LVMs with medical data to align radiology images to corresponding medical reports. However, there are two key factors that limit these LVM's performance. Firstly, LVMs lack complex reasoning capability that leads to logical inconsistencies and potential diagnostic errors in generated reports. Secondly, LVMs lack reflection mechanism that leads to an inability to discover errors in the thinking process. To address these gaps, we propose LVMed-R2, a new fine-tuning strategy that introduces complex reasoning and reflection mechanisms for LVMs to enhance medical report generation. To the best of our knowledge, this is the first work to introduce complex reasoning to the medical report generation (MRG) task. Our proposed complex reasoning contains medical knowledge injection and perception-enhancing modules which improve the accuracy of LVMs diagnosis, coupled with a perception tree to provide guidance to limit the perception range. Further, the reflection mechanism forces self-verification for outputs to correct for potential errors. We experimented by fine-tuning LVMs with our proposed LVMed-R2 strategy, using IU-Xray and MIMIC-CXR datasets. Our results, measured on natural language generation (NLG) metrics and clinical efficacy (CE) metrics, demonstrate that LVMs fine-tuned with the proposed reflection mechanism possess the ability to correct outputs and complex reasoning effectively and improve LVMs performance for MRG.

Paperid: 4661, https://arxiv.org/pdf/2504.02765.pdf

Abstract:
This study presents a novel robot-led approach to assessing children's mental wellbeing using a Vision Language Model (VLM). Inspired by the Child Apperception Test (CAT), the social robot NAO presented children with pictorial stimuli to elicit their verbal narratives of the images, which were then evaluated by a VLM in accordance with CAT assessment guidelines. The VLM's assessments were systematically compared to those provided by a trained psychologist. The results reveal that while the VLM demonstrates moderate reliability in identifying cases with no wellbeing concerns, its ability to accurately classify assessments with clinical concern remains limited. Moreover, although the model's performance was generally consistent when prompted with varying demographic factors such as age and gender, a significantly higher false positive rate was observed for girls, indicating potential sensitivity to gender attribute. These findings highlight both the promise and the challenges of integrating VLMs into robot-led assessments of children's wellbeing.

Paperid: 4662, https://arxiv.org/pdf/2504.01952.pdf

Abstract:
Visual grounding (VG) typically focuses on locating regions of interest within an image using natural language, and most existing VG methods are limited to single-image interpretations. This limits their applicability in real-world scenarios like automatic surveillance, where detecting subtle but meaningful visual differences across multiple images is crucial. Besides, previous work on image difference understanding (IDU) has either focused on detecting all change regions without cross-modal text guidance, or on providing coarse-grained descriptions of differences. Therefore, to push towards finer-grained vision-language perception, we propose Image Difference Grounding (IDG), a task designed to precisely localize visual differences based on user instructions. We introduce DiffGround, a large-scale and high-quality dataset for IDG, containing image pairs with diverse visual variations along with instructions querying fine-grained differences. Besides, we present a baseline model for IDG, DiffTracker, which effectively integrates feature differential enhancement and common suppression to precisely locate differences. Experiments on the DiffGround dataset highlight the importance of our IDG dataset in enabling finer-grained IDU. To foster future research, both DiffGround data and DiffTracker model will be publicly released.

Paperid: 4663, https://arxiv.org/pdf/2504.01735.pdf

Abstract:
Large Vision-Language Models (LVLMs), such as GPT-4o and LLaVA, have recently witnessed remarkable advancements and are increasingly being deployed in real-world applications. However, inheriting the sensitivity of visual neural networks, LVLMs remain vulnerable to adversarial attacks, which can result in erroneous or malicious outputs. While existing efforts utilize adversarial fine-tuning to enhance robustness, they often suffer from performance degradation on clean inputs. In this paper, we proposes AdPO, a novel adversarial defense strategy for LVLMs based on preference optimization. For the first time, we reframe adversarial training as a preference optimization problem, aiming to enhance the model's preference for generating normal outputs on clean inputs while rejecting the potential misleading outputs for adversarial examples. Notably, AdPO achieves this by solely modifying the image encoder, e.g., CLIP ViT, resulting in superior clean and adversarial performance in a variety of downsream tasks. Considering that training involves large language models (LLMs), the computational cost increases significantly. We validate that training on smaller LVLMs and subsequently transferring to larger models can achieve competitive performance while maintaining efficiency comparable to baseline methods. Our comprehensive experiments confirm the effectiveness of the proposed AdPO, which provides a novel perspective for future adversarial defense research.

Paperid: 4664, https://arxiv.org/pdf/2504.00043.pdf

Abstract:
Existing reasoning evaluation frameworks for Large Language Models (LLMs) and Large Vision-Language Models (LVLMs) predominantly assess either text-based reasoning or vision-language understanding capabilities, with limited dynamic interplay between textual and visual constraints. To address this limitation, we introduce CrossWordBench, a benchmark designed to evaluate the reasoning capabilities of both LLMs and LVLMs through the medium of crossword puzzles -- a task requiring multimodal adherence to semantic constraints from text-based clues and intersectional constraints from visual grid structures. CrossWordBench leverages a controllable puzzle generation framework that produces puzzles in two formats (text and image), supports adjustable difficulty through prefill ratio control, and offers different evaluation strategies, ranging from direct puzzle solving to interactive modes. Our extensive evaluation of over 20 models reveals that reasoning LLMs substantially outperform non-reasoning models by effectively leveraging crossing-letter constraints. We further demonstrate that LVLMs struggle with the task, showing a strong correlation between their puzzle-solving performance and grid-parsing accuracy. Our findings highlight limitations of the reasoning capabilities of current LLMs and LVLMs, and provide an effective approach for creating multimodal constrained tasks for future evaluations.

Paperid: 4665, https://arxiv.org/pdf/2503.23707.pdf

Abstract:
3D layout tasks have traditionally concentrated on geometric constraints, but many practical applications demand richer contextual understanding that spans social interactions, cultural traditions, and usage conventions. Existing methods often rely on rule-based heuristics or narrowly trained learning models, making them difficult to generalize and frequently prone to orientation errors that break realism. To address these challenges, we define four escalating context levels, ranging from straightforward physical placement to complex cultural requirements such as religious customs and advanced social norms. We then propose a Vision-Language Model-based pipeline that inserts minimal visual cues for orientation guidance and employs iterative feedback to pinpoint, diagnose, and correct unnatural placements in an automated fashion. Each adjustment is revisited through the system's verification process until it achieves a coherent result, thereby eliminating the need for extensive user oversight or manual parameter tuning. Our experiments across these four context levels reveal marked improvements in rotation accuracy, distance control, and overall layout plausibility compared with native VLM. By reducing the dependence on pre-programmed constraints or prohibitively large training sets, our method enables fully automated scene composition for both everyday scenarios and specialized cultural tasks, moving toward a universally adaptable framework for 3D arrangement.

Paperid: 4666, https://arxiv.org/pdf/2503.23623.pdf

Abstract:
Text-to-image diffusion models have demonstrated a remarkable ability to generate photorealistic images from natural language prompts. These high-resolution, language-guided synthesized images are essential for the explainability of disease or exploring causal relationships. However, their potential for disentangling and controlling latent factors of variation in specialized domains like medical imaging remains under-explored. In this work, we present the first investigation of the power of pre-trained vision-language foundation models, once fine-tuned on medical image datasets, to perform latent disentanglement for factorized medical image generation and interpolation. Through extensive experiments on chest X-ray and skin datasets, we illustrate that fine-tuned, language-guided Stable Diffusion inherently learns to factorize key attributes for image generation, such as the patient's anatomical structures or disease diagnostic features. We devise a framework to identify, isolate, and manipulate key attributes through latent space trajectory traversal of generative models, facilitating precise control over medical image synthesis.

Paperid: 4667, https://arxiv.org/pdf/2503.23618.pdf

Abstract:
Vision-language foundation models (VLMs) have shown impressive performance in guiding image generation through text, with emerging applications in medical imaging. In this work, we are the first to investigate the question: 'Can fine-tuned foundation models help identify critical, and possibly unknown, data properties?' By evaluating our proposed method on a chest x-ray dataset, we show that these models can generate high-resolution, precisely edited images compared to methods that rely on Structural Causal Models (SCMs) according to numerous metrics. For the first time, we demonstrate that fine-tuned VLMs can reveal hidden data relationships that were previously obscured due to available metadata granularity and model capacity limitations. Our experiments demonstrate both the potential of these models to reveal underlying dataset properties while also exposing the limitations of fine-tuned VLMs for accurate image editing and susceptibility to biases and spurious correlations.

Paperid: 4668, https://arxiv.org/pdf/2503.20188.pdf

Abstract:
Deepfake detection is a long-established research topic vital for mitigating the spread of malicious misinformation. Unlike prior methods that provide either binary classification results or textual explanations separately, we introduce a novel method capable of generating both simultaneously. Our method harnesses the multi-modal learning capability of the pre-trained CLIP and the unprecedented interpretability of large language models (LLMs) to enhance both the generalization and explainability of deepfake detection. Specifically, we introduce a multi-modal face forgery detector (M2F2-Det) that employs tailored face forgery prompt learning, incorporating the pre-trained CLIP to improve generalization to unseen forgeries. Also, M2F2-Det incorporates an LLM to provide detailed textual explanations of its detection decisions, enhancing interpretability by bridging the gap between natural language and subtle cues of facial forgeries. Empirically, we evaluate M2F2-Det on both detection and explanation generation tasks, where it achieves state-of-the-art performance, demonstrating its effectiveness in identifying and explaining diverse forgeries.

Paperid: 4669, https://arxiv.org/pdf/2503.19850.pdf

Abstract:
Information retrieval in hour-long videos presents a significant challenge, even for state-of-the-art Vision-Language Models (VLMs), particularly when the desired information is localized within a small subset of frames. Long video data presents challenges for VLMs due to context window limitations and the difficulty of pinpointing frames containing the answer. Our novel video agent, FALCONEye, combines a VLM and a Large Language Model (LLM) to search relevant information along the video, and locate the frames with the answer. FALCONEye novelty relies on 1) the proposed meta-architecture, which is better suited to tackle hour-long videos compared to short video approaches in the state-of-the-art; 2) a new efficient exploration algorithm to locate the information using short clips, captions and answer confidence; and 3) our state-of-the-art VLMs calibration analysis for the answer confidence. Our agent is built over a small-size VLM and a medium-size LLM being accessible to run on standard computational resources. We also release FALCON-Bench, a benchmark to evaluate long (average > 1 hour) Video Answer Search challenges, highlighting the need for open-ended question evaluation. Our experiments show FALCONEye's superior performance than the state-of-the-art in FALCON-Bench, and similar or better performance in related benchmarks.

Paperid: 4670, https://arxiv.org/pdf/2503.18672.pdf

Abstract:
Class-Continual Learning (CCL) enables models to continuously learn new class knowledge while retaining previous classes, facilitating adaptation and evolution in dynamic, real-world environments. Traditional CCL methods primarily rely on visual features, which limits their effectiveness in complex, multimodal scenarios. In contrast, Vision-Language Models (VLMs) show promising potential for enhancing CCL by leveraging pre-trained knowledge and fusing multi-modal semantic cues such as text and vision. However, existing approaches struggle to mitigate catastrophic forgetting while preserving the generalization strengths of VLMs across diverse modalities. To address these challenges, we propose CalFuse, a framework for feature Calibration enhanced parameter Fusion, which enhances dynamic knowledge fusion. CalFuse introduces a dynamic feature calibration mechanism that iteratively adjusts the contribution of original visual features to the final class decision, thereby preserving the model's intrinsic generalization capability across modalities. Simultaneously, a parameter fusion strategy effectively fuses newly acquired knowledge with prior task parameters, maintaining a balance between acquiring new class representations and preserving old knowledge. Experimental results on popular benchmarks (e.g., CIFAR100 and ImageNet100) validate the superiority of the proposed method.

Paperid: 4671, https://arxiv.org/pdf/2503.18278.pdf

Abstract:
Vision-Language Models (VLMs) demand substantial computational resources during inference, largely due to the extensive visual input tokens for representing visual information. Previous studies have noted that visual tokens tend to receive less attention than text tokens, suggesting their lower importance during inference and potential for pruning. However, their methods encounter several challenges: reliance on greedy heuristic criteria for token importance and incompatibility with FlashAttention and KV cache. To address these issues, we introduce \textbf{TopV}, a compatible \textbf{TO}ken \textbf{P}runing with inference Time Optimization for fast and low-memory \textbf{V}LM, achieving efficient pruning without additional training or fine-tuning. Instead of relying on attention scores, we formulate token pruning as an optimization problem, accurately identifying important visual tokens while remaining compatible with FlashAttention. Additionally, since we only perform this pruning once during the prefilling stage, it effectively reduces KV cache size. Our optimization framework incorporates a visual-aware cost function considering factors such as Feature Similarity, Relative Spatial Distance, and Absolute Central Distance, to measure the importance of each source visual token, enabling effective pruning of low-importance tokens. Extensive experiments demonstrate that our method outperforms previous token pruning methods, validating the effectiveness and efficiency of our approach.

Paperid: 4672, https://arxiv.org/pdf/2503.17551.pdf

Abstract:
Transformer-based multimodal models are widely used in industrial-scale recommendation, search, and advertising systems for content understanding and relevance ranking. Enhancing labeled training data quality and cross-modal fusion significantly improves model performance, influencing key metrics such as quality view rates and ad revenue. High-quality annotations are crucial for advancing content modeling, yet traditional statistical-based active learning (AL) methods face limitations: they struggle to detect overconfident misclassifications and are less effective in distinguishing semantically similar items in deep neural networks. Additionally, audio information plays an increasing role, especially in short-video platforms, yet most pre-trained multimodal architectures primarily focus on text and images. While training from scratch across all three modalities is possible, it sacrifices the benefits of leveraging existing pre-trained visual-language (VL) and audio models. To address these challenges, we propose kNN-based Latent Space Broadening (LSB) to enhance AL efficiency and Vision-Language Modeling with Audio Enhancement (VLMAE), a mid-fusion approach integrating audio into VL models. This system deployed in production systems, leading to significant business gains.

Paperid: 4673, https://arxiv.org/pdf/2503.16724.pdf

Abstract:
Semantic interpretability in Reinforcement Learning (RL) enables transparency and verifiability by making the agent's decisions understandable and verifiable. Achieving this, however, requires a feature space composed of human-understandable concepts, which traditionally rely on human specification and may fail to generalize to unseen environments. We introduce interpretable Tree-based Reinforcement learning via Automated Concept Extraction (iTRACE), an automated framework that leverages pre-trained vision-language models (VLM) for semantic feature extraction and interpretable tree-based models for policy optimization. iTRACE first extracts semantically meaningful features, then maps them to policies via interpretable trees. To address the impracticality of running VLMs in RL loops, we distill their outputs into a lightweight model. By leveraging Vision-Language Models (VLMs) to automate tree-based reinforcement learning, iTRACE eliminates the need for human annotation traditionally required by interpretable models, while also addressing the limitations of VLMs alone, such as their lack of grounding in action spaces and inability to directly optimize policies. iTRACE outperforms MLP baselines that use the same interpretable features and matches the performance of CNN-based policies, producing verifiable, semantically interpretable, and human-aligned behaviors without requiring human annotation.

Paperid: 4674, https://arxiv.org/pdf/2503.15948.pdf

Abstract:
Quantifying the realism of images remains a challenging problem in the field of artificial intelligence. For example, an image of Albert Einstein holding a smartphone violates common-sense because modern smartphone were invented after Einstein's death. We introduce a novel method for assessing image realism using Large Vision-Language Models (LVLMs) and Natural Language Inference (NLI). Our approach is based on the premise that LVLMs may generate hallucinations when confronted with images that defy common sense. Using LVLM to extract atomic facts from these images, we obtain a mix of accurate facts and erroneous hallucinations. We proceed by calculating pairwise entailment scores among these facts, subsequently aggregating these values to yield a singular reality score. This process serves to identify contradictions between genuine facts and hallucinatory elements, signaling the presence of images that violate common sense. Our approach has achieved a new state-of-the-art performance in zero-shot mode on the WHOOPS! dataset.

Paperid: 4675, https://arxiv.org/pdf/2503.15633.pdf

Abstract:
The recent successes of Vision-Language models raise the question of how to equivalently imbue a pretrained speech model with vision understanding, an important milestone towards building a multimodal speech model able to freely converse about images. Building such a conversational Vision-Speech model brings its unique challenges: (i) paired image-speech datasets are much scarcer than their image-text counterparts, (ii) ensuring real-time latency at inference is crucial thus bringing compute and memory constraints, and (iii) the model should preserve prosodic features (e.g., speaker tone) which cannot be inferred from text alone. In this work, we introduce MoshiVis, augmenting a recent dialogue speech LLM, Moshi, with visual inputs through lightweight adaptation modules. An additional dynamic gating mechanism enables the model to more easily switch between the visual inputs and unrelated conversation topics. To reduce training costs, we design a simple one-stage, parameter-efficient fine-tuning pipeline in which we leverage a mixture of image-text (i.e., "speechless") and image-speech samples. We evaluate the model on downstream visual understanding tasks with both audio and text prompts, and report qualitative samples of interactions with MoshiVis. Our inference code will be made available, as well as the image-speech data used for audio evaluation.

Paperid: 4676, https://arxiv.org/pdf/2503.13834.pdf

Abstract:
Vision-language (VL) models have demonstrated strong performance across various tasks. However, these models often rely on a specific modality for predictions, leading to "dominant modality bias.'' This bias significantly hurts performance, especially when one modality is impaired. In this study, we analyze model behavior under dominant modality bias and theoretically show that unaligned gradients or differences in gradient magnitudes prevent balanced convergence of the loss. Based on these findings, we propose a novel framework, BalGrad to mitigate dominant modality bias. Our approach includes inter-modality gradient reweighting, adjusting the gradient of KL divergence based on each modality's contribution, and inter-task gradient projection to align task directions in a non-conflicting manner. Experiments on UPMC Food-101, Hateful Memes, and MM-IMDb datasets confirm that BalGrad effectively alleviates over-reliance on specific modalities when making predictions.

Paperid: 4677, https://arxiv.org/pdf/2503.08144.pdf

Abstract:
Recently, large language models (LLMs) and vision-language models (VLMs) have achieved significant success, demonstrating remarkable capabilities in understanding various images and videos, particularly in classification and detection tasks. However, due to the substantial differences between remote sensing images and conventional optical images, these models face considerable challenges in comprehension, especially in detection tasks. Directly prompting VLMs with detection instructions often leads to unsatisfactory results. To address this issue, this letter explores the application of VLMs for object detection in remote sensing images. Specifically, we constructed supervised fine-tuning (SFT) datasets using publicly available remote sensing object detection datasets, including SSDD, HRSID, and NWPU-VHR-10. In these new datasets, we converted annotation information into JSON-compliant natural language descriptions, facilitating more effective understanding and training for the VLM. We then evaluate the detection performance of various fine-tuning strategies for VLMs and derive optimized model weights for object detection in remote sensing images. Finally, we evaluate the model's prior knowledge capabilities using natural language queries. Experimental results demonstrate that, without modifying the model architecture, remote sensing object detection can be effectively achieved using natural language alone. Additionally, the model exhibits the ability to perform certain vision question answering (VQA) tasks. Our datasets and related code will be released soon.

Paperid: 4678, https://arxiv.org/pdf/2503.06637.pdf

Abstract:
We propose CLAD -- a Constrained Latent Action Diffusion model for vision-language procedure planning in instructional videos. Procedure planning is the challenging task of predicting intermediate actions given a visual observation of a start and a goal state. However, future interactive AI systems must also be able to plan procedures using multi-modal input, e.g., where visual observations are augmented with language descriptions. To tackle this vision-language procedure planning task, our method uses a Variational Autoencoder (VAE) to learn the latent representation of actions and observations as constraints and integrate them into the diffusion process. This approach exploits that the latent space of diffusion models already has semantics that can be used. We use the latent constraints to steer the diffusion model to better generate actions. We report extensive experiments on the popular CrossTask, Coin, and NIV datasets and show that our method outperforms state-of-the-art methods by a large margin. By evaluating ablated versions of our method, we further show that the proposed integration of the action and observation representations learnt in the VAE latent space is key to these performance improvements.

Paperid: 4679, https://arxiv.org/pdf/2503.06497.pdf

Abstract:
Ensuring the safety of vision-language models (VLMs) in autonomous driving systems is of paramount importance, yet existing research has largely focused on conventional benchmarks rather than safety-critical evaluation. In this work, we present SCD-Bench (Safety Cognition Driving Benchmark) a novel framework specifically designed to assess the safety cognition capabilities of VLMs within interactive driving scenarios. To address the scalability challenge of data annotation, we introduce ADA (Autonomous Driving Annotation), a semi-automated labeling system, further refined through expert review by professionals with domain-specific knowledge in autonomous driving. To facilitate scalable and consistent evaluation, we also propose an automated assessment pipeline leveraging large language models, which demonstrates over 98% agreement with human expert judgments. In addressing the broader challenge of aligning VLMs with safety cognition in driving environments, we construct SCD-Training, the first large-scale dataset tailored for this task, comprising 324.35K high-quality samples. Through extensive experiments, we show that models trained on SCD-Training exhibit marked improvements not only on SCD-Bench, but also on general and domain-specific benchmarks, offering a new perspective on enhancing safety-aware interactions in vision-language systems for autonomous driving.

Paperid: 4680, https://arxiv.org/pdf/2503.06219.pdf

Abstract:
Camera-based 3D semantic scene completion (SSC) provides dense geometric and semantic perception for autonomous driving. However, images provide limited information making the model susceptible to geometric ambiguity caused by occlusion and perspective distortion. Existing methods often lack explicit semantic modeling between objects, limiting their perception of 3D semantic context. To address these challenges, we propose a novel method VLScene: Vision-Language Guidance Distillation for Camera-based 3D Semantic Scene Completion. The key insight is to use the vision-language model to introduce high-level semantic priors to provide the object spatial context required for 3D scene understanding. Specifically, we design a vision-language guidance distillation process to enhance image features, which can effectively capture semantic knowledge from the surrounding environment and improve spatial context reasoning. In addition, we introduce a geometric-semantic sparse awareness mechanism to propagate geometric structures in the neighborhood and enhance semantic information through contextual sparse interactions. Experimental results demonstrate that VLScene achieves rank-1st performance on challenging benchmarks--SemanticKITTI and SSCBench-KITTI-360, yielding remarkably mIoU scores of 17.52 and 19.10, respectively.

Paperid: 4681, https://arxiv.org/pdf/2503.01754.pdf

Abstract:
Reasoning is increasingly crucial for various tasks. While chain-of-thought prompting enables large language models to leverage reasoning effectively, harnessing the reasoning capabilities of Vision-Language Models (VLMs) remains challenging. To solve this problem, we propose a novel self-distillation framework that enhances the reasoning capabilities of the model. The proposed framework introduces several key innovations. We start by employing a prompt library tailored to visual reasoning tasks to generate diverse in-context questions and utilize a two-step reasoning procedure to derive reasoning-guided responses. These responses are then used for self-distillation, enabling the model to internalize the reasoning process. Additionally, we improve the model architecture with several innovative components, including an intervention adapter for efficient parameter updates, a cross-modal skip connection to facilitate information exchange between modalities, and an ensemble learning algorithm to integrate diverse reasoning from multiple in-context questions. Extensive experiments show that our method significantly improves the baseline performance across five VQA datasets.

Paperid: 4682, https://arxiv.org/pdf/2502.19844.pdf

Abstract:
Vision-language models (VLMs) have made significant progress in image classification by training with large-scale paired image-text data. Their performances largely depend on the prompt quality. While recent methods show that visual descriptions generated by large language models (LLMs) enhance the generalization of VLMs, class-specific prompts may be inaccurate or lack discrimination due to the hallucination in LLMs. In this paper, we aim to find visually discriminative prompts for fine-grained categories with minimal supervision and no human-in-the-loop. An evolution-based algorithm is proposed to progressively optimize language prompts from task-specific templates to class-specific descriptions. Unlike optimizing templates, the search space shows an explosion in class-specific candidate prompts. This increases prompt generation costs, iterative times, and the overfitting problem. To this end, we first introduce several simple yet effective edit-based and evolution-based operations to generate diverse candidate prompts by one-time query of LLMs. Then, two sampling strategies are proposed to find a better initial search point and reduce traversed categories, saving iteration costs. Moreover, we apply a novel fitness score with entropy constraints to mitigate overfitting. In a challenging one-shot image classification setting, our method outperforms existing textual prompt-based methods and improves LLM-generated description methods across 13 datasets. Meanwhile, we demonstrate that our optimal prompts improve adapter-based methods and transfer effectively across different backbones.

Paperid: 4683, https://arxiv.org/pdf/2502.19790.pdf

Abstract:
State-of-the-art large language and vision models are trained over trillions of tokens that are aggregated from a large variety of sources. As training data collections grow, manually managing the samples becomes time-consuming, tedious, and prone to errors. Yet recent research shows that the data mixture and the order in which samples are visited during training can significantly influence model accuracy. We build and present Mixtera, a data plane for foundation model training that enables users to declaratively express which data samples should be used in which proportion and in which order during training. Mixtera is a centralized, read-only layer that is deployed on top of existing training data collections and can be declaratively queried. It operates independently of the filesystem structure and supports mixtures across arbitrary properties (e.g., language, source dataset) as well as dynamic adjustment of the mixture based on model feedback. We experimentally evaluate Mixtera and show that our implementation does not bottleneck training and scales to 256 GH200 superchips. We demonstrate how Mixtera supports recent advancements in mixing strategies by implementing the proposed Adaptive Data Optimization (ADO) algorithm in the system and evaluating its performance impact. We also explore the role of mixtures for vision-language models.

Paperid: 4684, https://arxiv.org/pdf/2502.17092.pdf

Abstract:
We introduce Shakti VLM, a family of vision-language models in the capacity of 1B and 4B parameters designed to address data efficiency challenges in multimodal learning. While recent VLMs achieve strong performance through extensive training data, Shakti models leverage architectural innovations to attain competitive results with fewer tokens. Key advancements include QK-Normalization for attention stability, hybrid normalization techniques, and enhanced positional encoding. A three-stage training strategy further optimizes learning efficiency. Evaluations show that Shakti-Shakti-VLM-1B and Shakti-VLM-4B excel in document understanding, Visual Reasoning, OCR extraction, and general multimodal reasoning. Our results highlight that high performance can be achieved through model design and training strategy rather than sheer data volume, making Shakti an efficient solution for enterprise-scale multimodal tasks.

Paperid: 4685, https://arxiv.org/pdf/2502.16593.pdf

Abstract:
Large vision-language models (LVLMs) have demonstrated remarkable image understanding and dialogue capabilities, allowing them to handle a variety of visual question answering tasks. However, their widespread availability raises concerns about unauthorized usage and copyright infringement, where users or individuals can develop their own LVLMs by fine-tuning published models. In this paper, we propose a novel method called Parameter Learning Attack (PLA) for tracking the copyright of LVLMs without modifying the original model. Specifically, we construct adversarial images through targeted attacks against the original model, enabling it to generate specific outputs. To ensure these attacks remain effective on potential fine-tuned models to trigger copyright tracking, we allow the original model to learn the trigger images by updating parameters in the opposite direction during the adversarial attack process. Notably, the proposed method can be applied after the release of the original model, thus not affecting the model's performance and behavior. To simulate real-world applications, we fine-tune the original model using various strategies across diverse datasets, creating a range of models for copyright verification. Extensive experiments demonstrate that our method can more effectively identify the original copyright of fine-tuned models compared to baseline methods. Therefore, this work provides a powerful tool for tracking copyrights and detecting unlicensed usage of LVLMs.

Paperid: 4686, https://arxiv.org/pdf/2502.15381.pdf

Abstract:
Multimodal language models (MLMs) integrate visual and textual information by coupling a vision encoder with a large language model through the specific adapter. While existing approaches commonly rely on a single pre-trained vision encoder, there is a great variability of specialized encoders that can boost model's performance in distinct domains. In this work, we propose MOVE (Mixture of Vision Encoders) a simple yet effective approach to leverage multiple pre-trained encoders for specialized multimodal tasks. MOVE automatically routes inputs to the most appropriate encoder among candidates such as Unichat, InternViT, and Texify, thereby enhancing performance across a diverse set of benchmarks, including ChartQA, MMBench, and MMMU. Experimental results demonstrate that MOVE achieves competitive accuracy without incurring the complexities of image slicing for high-resolution images.

Paperid: 4687, https://arxiv.org/pdf/2502.15278.pdf

Abstract:
Assessing whether AI-generated images are substantially similar to source works is a crucial step in resolving copyright disputes. In this paper, we propose CopyJudge, a novel automated infringement identification framework that leverages large vision-language models (LVLMs) to simulate practical court processes for determining substantial similarity between copyrighted images and those generated by text-to-image diffusion models. Specifically, we employ an abstraction-filtration-comparison test framework based on the multi-LVLM debate to assess the likelihood of infringement and provide detailed judgment rationales. Based on these judgments, we further introduce a general LVLM-based mitigation strategy that automatically optimizes infringing prompts by avoiding sensitive expressions while preserving the non-infringing content. Furthermore, assuming the input noise is controllable, our approach can be enhanced by iteratively exploring non-infringing noise vectors within the diffusion latent space, even without modifying the original prompts. Experimental results show that our automated identification method achieves comparable state-of-the-art performance, while offering superior generalization and interpretability across various forms of infringement, and that our mitigation method more effectively mitigates memorization and IP infringement with a high degree of alignment to the original non-infringing expressions.

Paperid: 4688, https://arxiv.org/pdf/2502.13894.pdf

Abstract:
Navigating unfamiliar environments presents significant challenges for household robots, requiring the ability to recognize and reason about novel decoration and layout. Existing reinforcement learning methods cannot be directly transferred to new environments, as they typically rely on extensive mapping and exploration, leading to time-consuming and inefficient. To address these challenges, we try to transfer the logical knowledge and the generalization ability of pre-trained foundation models to zero-shot navigation. By integrating a large vision-language model with a diffusion network, our approach named \mname ~constructs a visual predictor that continuously predicts the agent's potential observations in the next step which can assist robots generate robust actions. Furthermore, to adapt the temporal property of navigation, we introduce temporal historical information to ensure that the predicted image is aligned with the navigation scene. We then carefully designed an information fusion framework that embeds the predicted future frames as guidance into goal-reaching policy to solve downstream image navigation tasks. This approach enhances navigation control and generalization across both simulated and real-world environments. Through extensive experimentation, we demonstrate the robustness and versatility of our method, showcasing its potential to improve the efficiency and effectiveness of robotic navigation in diverse settings.

Paperid: 4689, https://arxiv.org/pdf/2502.10148.pdf

Abstract:
Despite much progress in training distributed artificial intelligence (AI), building cooperative multi-agent systems with multi-agent reinforcement learning (MARL) faces challenges in sample efficiency, interpretability, and transferability. Unlike traditional learning-based methods that require extensive interaction with the environment, large language models (LLMs) demonstrate remarkable capabilities in zero-shot planning and complex reasoning. However, existing LLM-based approaches heavily rely on text-based observations and struggle with the non-Markovian nature of multi-agent interactions under partial observability. We present COMPASS, a novel multi-agent architecture that integrates vision-language models (VLMs) with a dynamic skill library and structured communication for decentralized closed-loop decision-making. The skill library, bootstrapped from demonstrations, evolves via planner-guided tasks to enable adaptive strategies. COMPASS propagates entity information through multi-hop communication under partial observability. Evaluations on the improved StarCraft Multi-Agent Challenge (SMACv2) demonstrate COMPASS's strong performance against state-of-the-art MARL baselines across both symmetric and asymmetric scenarios. Notably, in the symmetric Protoss 5v5 task, COMPASS achieved a 57\% win rate, representing a 30 percentage point advantage over QMIX (27\%). Project page can be found at https://stellar-entremet-1720bb.netlify.app/.

Paperid: 4690, https://arxiv.org/pdf/2502.06875.pdf

Abstract:
Large Language Models primarily operate through text-based inputs and outputs, yet human emotion is communicated through both verbal and non-verbal cues, including facial expressions. While Vision-Language Models analyze facial expressions from images, they are resource-intensive and may depend more on linguistic priors than visual understanding. To address this, this study investigates whether LLMs can infer affective meaning from dimensions of facial expressions-Valence and Arousal values, structured numerical representations, rather than using raw visual input. VA values were extracted using Facechannel from images of facial expressions and provided to LLMs in two tasks: (1) categorizing facial expressions into basic (on the IIMI dataset) and complex emotions (on the Emotic dataset) and (2) generating semantic descriptions of facial expressions (on the Emotic dataset). Results from the categorization task indicate that LLMs struggle to classify VA values into discrete emotion categories, particularly for emotions beyond basic polarities (e.g., happiness, sadness). However, in the semantic description task, LLMs produced textual descriptions that align closely with human-generated interpretations, demonstrating a stronger capacity for free text affective inference of facial expressions.

Paperid: 4691, https://arxiv.org/pdf/2502.04981.pdf

Abstract:
Obtaining high-quality 3D semantic occupancy from raw sensor data remains an essential yet challenging task, often requiring extensive manual labeling. In this work, we propose AutoOcc, a vision-centric automated pipeline for open-ended semantic occupancy annotation that integrates differentiable Gaussian splatting guided by vision-language models. We formulate the open-ended semantic 3D occupancy reconstruction task to automatically generate scene occupancy by combining attention maps from vision-language models and foundation vision models. We devise semantic-aware Gaussians as intermediate geometric descriptors and propose a cumulative Gaussian-to-voxel splatting algorithm that enables effective and efficient occupancy annotation. Our framework outperforms existing automated occupancy annotation methods without human labels. AutoOcc also enables open-ended semantic occupancy auto-labeling, achieving robust performance in both static and dynamically complex scenarios.

Paperid: 4692, https://arxiv.org/pdf/2502.03708.pdf

Abstract:
Modern AI models contain much of human knowledge, yet understanding of their internal representation of this knowledge remains elusive. Characterizing the structure and properties of this representation will lead to improvements in model capabilities and development of effective safeguards. Building on recent advances in feature learning, we develop an effective, scalable approach for extracting linear representations of general concepts in large-scale AI models (language models, vision-language models, and reasoning models). We show how these representations enable model steering, through which we expose vulnerabilities, mitigate misaligned behaviors, and improve model capabilities. Additionally, we demonstrate that concept representations are remarkably transferable across human languages and combinable to enable multi-concept steering. Through quantitative analysis across hundreds of concepts, we find that newer, larger models are more steerable and steering can improve model capabilities beyond standard prompting. We show how concept representations are effective for monitoring misaligned content (hallucinations, toxic content). We demonstrate that predictive models built using concept representations are more accurate for monitoring misaligned content than using models that judge outputs directly. Together, our results illustrate the power of using internal representations to map the knowledge in AI models, advance AI safety, and improve model capabilities.

Paperid: 4693, https://arxiv.org/pdf/2502.01524.pdf

Abstract:
The integration of vision-language modalities has been a significant focus in multimodal learning, traditionally relying on Vision-Language Pretrained Models. However, with the advent of Large Language Models (LLMs), there has been a notable shift towards incorporating LLMs with vision modalities. Following this, the training paradigms for incorporating vision modalities into LLMs have evolved. Initially, the approach was to integrate the modalities through pretraining the modality integrator, named Single-stage Tuning. It has since branched out into methods focusing on performance enhancement, denoted as Two-stage Tuning, and those prioritizing parameter efficiency, referred to as Direct Adaptation. However, existing surveys primarily address the latest Vision Large Language Models (VLLMs) with Two-stage Tuning, leaving a gap in understanding the evolution of training paradigms and their unique parameter-efficient considerations. This paper categorizes and reviews 34 VLLMs from top conferences, journals, and highly cited Arxiv papers, focusing on parameter efficiency during adaptation from the training paradigm perspective. We first introduce the architecture of LLMs and parameter-efficient learning methods, followed by a discussion on vision encoders and a comprehensive taxonomy of modality integrators. We then review three training paradigms and their efficiency considerations, summarizing benchmarks in the VLLM field. To gain deeper insights into their effectiveness in parameter efficiency, we compare and discuss the experimental results of representative models, among which the experiment of the Direct Adaptation paradigm is replicated. Providing insights into recent developments and practical uses, this survey is a vital guide for researchers and practitioners navigating the efficient integration of vision modalities into LLMs.

Paperid: 4694, https://arxiv.org/pdf/2501.11653.pdf

Abstract:
Images depicting complex, dynamic scenes are challenging to parse automatically, requiring both high-level comprehension of the overall situation and fine-grained identification of participating entities and their interactions. Current approaches use distinct methods tailored to sub-tasks such as Situation Recognition and detection of Human-Human and Human-Object Interactions. However, recent advances in image understanding have often leveraged web-scale vision-language (V&L) representations to obviate task-specific engineering. In this work, we propose a framework for dynamic scene understanding tasks by leveraging knowledge from modern, frozen V&L representations. By framing these tasks in a generic manner - as predicting and parsing structured text, or by directly concatenating representations to the input of existing models - we achieve state-of-the-art results while using a minimal number of trainable parameters relative to existing approaches. Moreover, our analysis of dynamic knowledge of these representations shows that recent, more powerful representations effectively encode dynamic scene semantics, making this approach newly possible.

Paperid: 4695, https://arxiv.org/pdf/2501.03957.pdf

Abstract:
Large Language Models integrating textual and visual inputs have introduced new possibilities for interpreting complex data. Despite their remarkable ability to generate coherent and contextually relevant text based on visual stimuli, the alignment of these models with human perception in identifying relevant elements in images requires further exploration. This paper investigates the alignment between state-of-the-art LLMs and human annotators in detecting elements of relevance within home environment scenarios. We created a set of twelve images depicting various domestic scenarios and enlisted fourteen annotators to identify the key element in each image. We then compared these human responses with outputs from five different LLMs, including GPT-4o and four LLaVA variants. Our findings reveal a varied degree of alignment, with LLaVA 34B showing the highest performance but still scoring low. However, an analysis of the results highlights the models' potential to detect value-laden elements in images, suggesting that with improved training and refined prompts, LLMs could enhance applications in social robotics, assistive technologies, and human-computer interaction by providing deeper insights and more contextually relevant responses.

Paperid: 4696, https://arxiv.org/pdf/2412.20715.pdf

Abstract:
Chart summarization, which focuses on extracting key information from charts and interpreting it in natural language, is crucial for generating and delivering insights through effective and accessible data analysis. Traditional methods for chart understanding and summarization often rely on multi-stage pipelines, which may produce suboptimal semantic alignment between visual and textual information. In comparison, recently developed LLM-based methods are more dependent on the capability of foundation images or languages, while ignoring the characteristics of chart data and its relevant challenges. To address these limitations, we propose ChartAdapter, a novel lightweight transformer module designed to bridge the gap between charts and textual summaries. ChartAdapter employs learnable query vectors to extract implicit semantics from chart data and incorporates a cross-modal alignment projector to enhance vision-to-language generative learning. By integrating ChartAdapter with an LLM, we enable end-to-end training and efficient chart summarization. To further enhance the training, we introduce a three-stage hierarchical training procedure and develop a large-scale dataset specifically curated for chart summarization, comprising 190,618 samples. Experimental results on the standard Chart-to-Text testing set demonstrate that our approach significantly outperforms existing methods, including state-of-the-art models, in generating high-quality chart summaries. Ablation studies further validate the effectiveness of key components in ChartAdapter. This work highlights the potential of tailored LLM-based approaches to advance chart understanding and sets a strong foundation for future research in this area.

Paperid: 4697, https://arxiv.org/pdf/2412.18842.pdf

Abstract:
Due to the lack of extensive precisely-annotated multi-label data in real word, semi-supervised multi-label learning (SSMLL) has gradually gained attention. Abundant knowledge embedded in vision-language models (VLMs) pre-trained on large-scale image-text pairs could alleviate the challenge of limited labeled data under SSMLL setting.Despite existing methods based on fine-tuning VLMs have achieved advances in weakly-supervised multi-label learning, they failed to fully leverage the information from labeled data to enhance the learning of unlabeled data. In this paper, we propose a context-based semantic-aware alignment method to solve the SSMLL problem by leveraging the knowledge of VLMs. To address the challenge of handling multiple semantics within an image, we introduce a novel framework design to extract label-specific image features. This design allows us to achieve a more compact alignment between text features and label-specific image features, leading the model to generate high-quality pseudo-labels. To incorporate the model with comprehensive understanding of image, we design a semi-supervised context identification auxiliary task to enhance the feature representation by capturing co-occurrence information. Extensive experiments on multiple benchmark datasets demonstrate the effectiveness of our proposed method.

Paperid: 4698, https://arxiv.org/pdf/2412.18298.pdf

Abstract:
Video anomaly detection (VAD) has witnessed significant advancements through the integration of large language models (LLMs) and vision-language models (VLMs), addressing critical challenges such as interpretability, temporal reasoning, and generalization in dynamic, open-world scenarios. This paper presents an in-depth review of cutting-edge LLM-/VLM-based methods in 2024, focusing on four key aspects: (i) enhancing interpretability through semantic insights and textual explanations, making visual anomalies more understandable; (ii) capturing intricate temporal relationships to detect and localize dynamic anomalies across video frames; (iii) enabling few-shot and zero-shot detection to minimize reliance on large, annotated datasets; and (iv) addressing open-world and class-agnostic anomalies by using semantic understanding and motion features for spatiotemporal coherence. We highlight their potential to redefine the landscape of VAD. Additionally, we explore the synergy between visual and textual modalities offered by LLMs and VLMs, highlighting their combined strengths and proposing future directions to fully exploit the potential in enhancing video anomaly detection.

Paperid: 4699, https://arxiv.org/pdf/2412.15264.pdf

Abstract:
The increasing adoption of AI-generated radiology reports necessitates robust methods for detecting hallucinations--false or unfounded statements that could impact patient care. We present ReXTrust, a novel framework for fine-grained hallucination detection in AI-generated radiology reports. Our approach leverages sequences of hidden states from large vision-language models to produce finding-level hallucination risk scores. We evaluate ReXTrust on a subset of the MIMIC-CXR dataset and demonstrate superior performance compared to existing approaches, achieving an AUROC of 0.8751 across all findings and 0.8963 on clinically significant findings. Our results show that white-box approaches leveraging model hidden states can provide reliable hallucination detection for medical AI systems, potentially improving the safety and reliability of automated radiology reporting.

Paperid: 4700, https://arxiv.org/pdf/2412.15251.pdf

Abstract:
The advanced processing and reasoning capabilities of multimodal large language models (MLLMs) have driven substantial progress in vision-language (VL) understanding tasks. However, while effective for tasks governed by straightforward logic, MLLMs often struggle with reasoning complex, detail-intensive logical structures. To address this limitation, we introduce AgentPS, a novel framework that integrates Agentic Process Supervision into MLLMs by sequentially reasoning over ancillary questions during fine-tuning. AgentPS achieves substantial improvements over baseline MLLMs on both public benchmarks and proprietary datasets. Notably, we show that using MLLM-generated ancillary labels in place of human annotations yields only minimal performance degradation, highlighting the method's scalability. These results establish AgentPS as a scalable and effective solution for complex multimodal classification in large-scale industrial applications.

Paperid: 4701, https://arxiv.org/pdf/2412.13187.pdf

Abstract:
How can we predict future interaction trajectories of human hands in a scene given high-level colloquial task specifications in the form of natural language? In this paper, we extend the classic hand trajectory prediction task to two tasks involving explicit or implicit language queries. Our proposed tasks require extensive understanding of human daily activities and reasoning abilities about what should be happening next given cues from the current scene. We also develop new benchmarks to evaluate the proposed two tasks, Vanilla Hand Prediction (VHP) and Reasoning-Based Hand Prediction (RBHP). We enable solving these tasks by integrating high-level world knowledge and reasoning capabilities of Vision-Language Models (VLMs) with the auto-regressive nature of low-level ego-centric hand trajectories. Our model, HandsOnVLM is a novel VLM that can generate textual responses and produce future hand trajectories through natural-language conversations. Our experiments show that HandsOnVLM outperforms existing task-specific methods and other VLM baselines on proposed tasks, and demonstrates its ability to effectively utilize world knowledge for reasoning about low-level human hand trajectories based on the provided context. Our website contains code and detailed video results https://www.chenbao.tech/handsonvlm/

Paperid: 4702, https://arxiv.org/pdf/2411.09273.pdf

Abstract:
Recent developments in multimodal methodologies have marked the beginning of an exciting era for models adept at processing diverse data types, encompassing text, audio, and visual content. Models like GPT-4V, which merge computer vision with advanced language processing, exhibit extraordinary proficiency in handling intricate tasks that require a simultaneous understanding of both textual and visual information. Prior research efforts have meticulously evaluated the efficacy of these Vision Large Language Models (VLLMs) in various domains, including object detection, image captioning, and other related fields. However, existing analyses have often suffered from limitations, primarily centering on the isolated evaluation of each modality's performance while neglecting to explore their intricate cross-modal interactions. Specifically, the question of whether these models achieve the same level of accuracy when confronted with identical task instances across different modalities remains unanswered. In this study, we take the initiative to delve into the interaction and comparison among these modalities of interest by introducing a novel concept termed cross-modal consistency. Furthermore, we propose a quantitative evaluation framework founded on this concept. Our experimental findings, drawn from a curated collection of parallel vision-language datasets developed by us, unveil a pronounced inconsistency between the vision and language modalities within GPT-4V, despite its portrayal as a unified multimodal model. Our research yields insights into the appropriate utilization of such models and hints at potential avenues for enhancing their design.

Paperid: 4703, https://arxiv.org/pdf/2411.06764.pdf

Abstract:
Vision Language Models (VLMs), pre-trained on large-scale image-text datasets, enable zero-shot predictions for unseen data but may underperform on specific unseen tasks. Continual learning (CL) can help VLMs effectively adapt to new data distributions without joint training, but faces challenges of catastrophic forgetting and generalization forgetting. Although significant progress has been achieved by distillation-based methods, they exhibit two severe limitations. One is the popularly adopted single-teacher paradigm fails to impart comprehensive knowledge, The other is the existing methods inadequately leverage the multimodal information in the original training dataset, instead they rely on additional data for distillation, which increases computational and storage overhead. To mitigate both limitations, by drawing on Knowledge Integration Theory (KIT), we propose a Multi-Stage Knowledge Integration network (MulKI) to emulate the human learning process in distillation methods. MulKI achieves this through four stages, including Eliciting Ideas, Adding New Ideas, Distinguishing Ideas, and Making Connections. During the four stages, we first leverage prototypes to align across modalities, eliciting cross-modal knowledge, then adding new knowledge by constructing fine-grained intra- and inter-modality relationships with prototypes. After that, knowledge from two teacher models is adaptively distinguished and re-weighted. Finally, we connect between models from intra- and inter-task, integrating preceding and new knowledge. Our method demonstrates significant improvements in maintaining zero-shot capabilities while supporting continual learning across diverse downstream tasks, showcasing its potential in adapting VLMs to evolving data distributions.

Paperid: 4704, https://arxiv.org/pdf/2411.05831.pdf

Abstract:
Recent research in Vision Language Navigation (VLN) has overlooked the development of agents' inquisitive abilities, which allow them to ask clarifying questions when instructions are incomplete. This paper addresses how agents can recognize "when" they lack sufficient information, without focusing on "what" is missing, particularly in VLN tasks with vague instructions. Equipping agents with this ability enhances efficiency by reducing potential digressions and seeking timely assistance. The challenge in identifying such uncertain points is balancing between being overly cautious (high recall) and overly confident (high precision). We propose an attention-based instruction-vagueness estimation module that learns associations between instructions and the agent's trajectory. By leveraging instruction-to-path alignment information during training, the module's vagueness estimation performance improves by around 52% in terms of precision-recall balance. In our ablative experiments, we also demonstrate the effectiveness of incorporating this additional instruction-to-path attention network alongside the cross-modal attention networks within the navigator module. Our results show that the attention scores from the instruction-to-path attention network serve as better indicators for estimating vagueness.

Paperid: 4705, https://arxiv.org/pdf/2411.01445.pdf

Abstract:
Current visual question answering (VQA) tasks often require constructing multimodal datasets and fine-tuning visual language models, which demands significant time and resources. This has greatly hindered the application of VQA to downstream tasks, such as ship information analysis based on Synthetic Aperture Radar (SAR) imagery. To address this challenge, this letter proposes a novel VQA approach that integrates object detection networks with visual language models, specifically designed for analyzing ships in SAR images. This integration aims to enhance the capabilities of VQA systems, focusing on aspects such as ship location, density, and size analysis, as well as risk behavior detection. Initially, we conducted baseline experiments using YOLO networks on two representative SAR ship detection datasets, SSDD and HRSID, to assess each model's performance in terms of detection accuracy. Based on these results, we selected the optimal model, YOLOv8n, as the most suitable detection network for this task. Subsequently, leveraging the vision-language model Qwen2-VL, we designed and implemented a VQA task specifically for SAR scenes. This task employs the ship location and size information output by the detection network to generate multi-turn dialogues and scene descriptions for SAR imagery. Experimental results indicate that this method not only enables fundamental SAR scene question-answering without the need for additional datasets or fine-tuning but also dynamically adapts to complex, multi-turn dialogue requirements, demonstrating robust semantic understanding and adaptability.

Paperid: 4706, https://arxiv.org/pdf/2411.00246.pdf

Abstract:
When examined through the lens of their residual streams, a puzzling property emerges in transformer networks: residual contributions (e.g., attention heads) sometimes specialize in specific tasks or input attributes. In this paper, we analyze this phenomenon in vision transformers, focusing on the spectral geometry of residuals, and explore its implications for modality alignment in vision-language models. First, we link it to the intrinsically low-dimensional structure of visual head representations, zooming into their principal components and showing that they encode specialized roles across a wide variety of input data distributions. Then, we analyze the effect of head specialization in multimodal models, focusing on how improved alignment between text and specialized heads impacts zero-shot classification performance. This specialization-performance link consistently holds across diverse pre-training data, network sizes, and objectives, demonstrating a powerful new mechanism for boosting zero-shot classification through targeted alignment. Ultimately, we translate these insights into actionable terms by introducing ResiDual, a technique for spectral alignment of the residual stream. Much like panning for gold, it lets the noise from irrelevant unit principal components (i.e., attributes) wash away to amplify task-relevant ones. Remarkably, this dual perspective on modality alignment yields fine-tuning level performance on different data distributions while modelling an extremely interpretable and parameter-efficient transformation, as we extensively show on 70 pre-trained network-dataset combinations (7 models, 10 datasets).

Paperid: 4707, https://arxiv.org/pdf/2410.23640.pdf

Abstract:
The development of large language models and vision-language models (VLMs) has resulted in the increasing use of robotic systems in various fields. However, the effective integration of these models into real-world robotic tasks is a key challenge. We developed a versatile robotic system called SuctionPrompt that utilizes prompting techniques of VLMs combined with 3D detections to perform product-picking tasks in diverse and dynamic environments. Our method highlights the importance of integrating 3D spatial information with adaptive action planning to enable robots to approach and manipulate objects in novel environments. In the validation experiments, the system accurately selected suction points 75.4%, and achieved a 65.0% success rate in picking common items. This study highlights the effectiveness of VLMs in robotic manipulation tasks, even with simple 3D processing.

Paperid: 4708, https://arxiv.org/pdf/2410.22330.pdf

Abstract:
Autoregressive vision-language models (VLMs) can handle many tasks within a single model, yet the representations that enable this capability remain opaque. We find that VLMs align conceptually equivalent inputs into a shared task vector, which is invariant to modality (text, image) and format (examples, instruction), and may simplify VLM processing. We measure this alignment via cross-modal transfer -- the ability of a task vector derived in one modality to trigger the correct generation in another -- on a range of tasks and model architectures. Although the task vector is highly compressed, we find that this single vector outperforms prompting the model with the full task information, unique to this cross-modal case. Furthermore, we show that task vectors can be transferred from a base language model to its fine-tuned vision-language counterpart, and that they can be derived solely from instructions without the need for examples. Taken together, our findings shed light on how VLMs internally process task information, and how they map different modalities into common semantic representations. Project page: https://vlm-cross-modal-reps.github.io.

Paperid: 4709, https://arxiv.org/pdf/2410.15397.pdf

Abstract:
Pre-trained vision-language models like CLIP have remarkably adapted to various downstream tasks. Nonetheless, their performance heavily depends on the specificity of the input text prompts, which requires skillful prompt template engineering. Instead, current approaches to prompt optimization learn the prompts through gradient descent, where the prompts are treated as adjustable parameters. However, these methods tend to lead to overfitting of the base classes seen during training and produce prompts that are no longer understandable by humans. This paper introduces a simple but interpretable prompt optimizer (IPO), that utilizes large language models (LLMs) to generate textual prompts dynamically. We introduce a Prompt Optimization Prompt that not only guides LLMs in creating effective prompts but also stores past prompts with their performance metrics, providing rich in-context information. Additionally, we incorporate a large multimodal model (LMM) to condition on visual content by generating image descriptions, which enhance the interaction between textual and visual modalities. This allows for thae creation of dataset-specific prompts that improve generalization performance, while maintaining human comprehension. Extensive testing across 11 datasets reveals that IPO not only improves the accuracy of existing gradient-descent-based prompt learning methods but also considerably enhances the interpretability of the generated prompts. By leveraging the strengths of LLMs, our approach ensures that the prompts remain human-understandable, thereby facilitating better transparency and oversight for vision-language models.

Paperid: 4710, https://arxiv.org/pdf/2410.13882.pdf

Abstract:
Interactive 3D simulated objects are crucial in AR/VR, animations, and robotics, driving immersive experiences and advanced automation. However, creating these articulated objects requires extensive human effort and expertise, limiting their broader applications. To overcome this challenge, we present Articulate-Anything, a system that automates the articulation of diverse, complex objects from many input modalities, including text, images, and videos. Articulate-Anything leverages vision-language models (VLMs) to generate code that can be compiled into an interactable digital twin for use in standard 3D simulators. Our system exploits existing 3D asset datasets via a mesh retrieval mechanism, along with an actor-critic system that iteratively proposes, evaluates, and refines solutions for articulating the objects, self-correcting errors to achieve a robust outcome. Qualitative evaluations demonstrate Articulate-Anything's capability to articulate complex and even ambiguous object affordances by leveraging rich grounded inputs. In extensive quantitative experiments on the standard PartNet-Mobility dataset, Articulate-Anything substantially outperforms prior work, increasing the success rate from 8.7-11.6% to 75% and setting a new bar for state-of-the-art performance. We further showcase the utility of our system by generating 3D assets from in-the-wild video inputs, which are then used to train robotic policies for fine-grained manipulation tasks in simulation that go beyond basic pick and place. These policies are then transferred to a real robotic system.

Paperid: 4711, https://arxiv.org/pdf/2410.12595.pdf

Abstract:
With the flourishing of social media platforms, vision-language pre-training (VLP) recently has received great attention and many remarkable progresses have been achieved. The success of VLP largely benefits from the information complementation and enhancement between different modalities. However, most of recent studies focus on cross-modal contrastive learning (CMCL) to promote image-text alignment by pulling embeddings of positive sample pairs together while pushing those of negative pairs apart, which ignores the natural asymmetry property between different modalities and requires large-scale image-text corpus to achieve arduous progress. To mitigate this predicament, we propose CMAL, a Cross-Modal Associative Learning framework with anchor points detection and cross-modal associative learning for VLP. Specifically, we first respectively embed visual objects and textual tokens into separate hypersphere spaces to learn intra-modal hidden features, and then design a cross-modal associative prompt layer to perform anchor point masking and swap feature filling for constructing a hybrid cross-modal associative prompt. Afterwards, we exploit a unified semantic encoder to learn their cross-modal interactive features for context adaptation. Finally, we design an associative mapping classification layer to learn potential associative mappings between modalities at anchor points, within which we develop a fresh self-supervised associative mapping classification task to boost CMAL's performance. Experimental results verify the effectiveness of CMAL, showing that it achieves competitive performance against previous CMCL-based methods on four common downstream vision-and-language tasks, with significantly fewer corpus. Especially, CMAL obtains new state-of-the-art results on SNLI-VE and REC (testA).

Paperid: 4712, https://arxiv.org/pdf/2410.08792.pdf

Abstract:
Vision Language Models (VLMs) have recently been adopted in robotics for their capability in common sense reasoning and generalizability. Existing work has applied VLMs to generate task and motion planning from natural language instructions and simulate training data for robot learning. In this work, we explore using VLM to interpret human demonstration videos and generate robot task planning. Our method integrates keyframe selection, visual perception, and VLM reasoning into a pipeline. We named it SeeDo because it enables the VLM to ''see'' human demonstrations and explain the corresponding plans to the robot for it to ''do''. To validate our approach, we collected a set of long-horizon human videos demonstrating pick-and-place tasks in three diverse categories and designed a set of metrics to comprehensively benchmark SeeDo against several baselines, including state-of-the-art video-input VLMs. The experiments demonstrate SeeDo's superior performance. We further deployed the generated task plans in both a simulation environment and on a real robot arm.

Paperid: 4713, https://arxiv.org/pdf/2410.02773.pdf

Abstract:
Large vision-language models frequently struggle to accurately predict responses provided by multiple human annotators, particularly when those responses exhibit human uncertainty. In this study, we focus on the Visual Question Answering (VQA) task, and we comprehensively evaluate how well the state-of-the-art vision-language models correlate with the distribution of human responses. To do so, we categorize our samples based on their levels (low, medium, high) of human uncertainty in disagreement (HUD) and employ not only accuracy but also three new human-correlated metrics in VQA, to investigate the impact of HUD. To better align models with humans, we also verify the effect of common calibration and human calibration. Our results show that even BEiT3, currently the best model for this task, struggles to capture the multi-label distribution inherent in diverse human responses. Additionally, we observe that the commonly used accuracy-oriented calibration technique adversely affects BEiT3's ability to capture HUD, further widening the gap between model predictions and human distributions. In contrast, we show the benefits of calibrating models towards human distributions for VQA, better aligning model confidence with human uncertainty. Our findings highlight that for VQA, the consistent alignment between human responses and model predictions is understudied and should become the next crucial target of future studies.

Paperid: 4714, https://arxiv.org/pdf/2409.07267.pdf

Abstract:
Vision-language models (VLMs) serve as general-purpose end-to-end models in autonomous driving, performing subtasks such as prediction, planning, and perception through question-and-answer interactions. However, most existing methods rely on computationally expensive visual encoders and large language models (LLMs), making them difficult to deploy in real-world scenarios and real-time applications. Meanwhile, most existing VLMs lack the ability to process multiple images, making it difficult to adapt to multi-camera perception in autonomous driving. To address these issues, we propose a novel framework called MiniDrive, which incorporates our proposed Feature Engineering Mixture of Experts (FE-MoE) module and Dynamic Instruction Adapter (DI-Adapter). The FE-MoE effectively maps 2D features into visual token embeddings before being input into the language model. The DI-Adapter enables the visual token embeddings to dynamically change with the instruction text embeddings, resolving the issue of static visual token embeddings for the same image in previous approaches. Compared to previous works, MiniDrive achieves state-of-the-art performance in terms of parameter size, floating point operations, and response efficiency, with the smallest version containing only 83M parameters.

Paperid: 4715, https://arxiv.org/pdf/2409.06210.pdf

Abstract:
Affordance denotes the potential interactions inherent in objects. The perception of affordance can enable intelligent agents to navigate and interact with new environments efficiently. Weakly supervised affordance grounding teaches agents the concept of affordance without costly pixel-level annotations, but with exocentric images. Although recent advances in weakly supervised affordance grounding yielded promising results, there remain challenges including the requirement for paired exocentric and egocentric image dataset, and the complexity in grounding diverse affordances for a single object. To address them, we propose INTeraction Relationship-aware weakly supervised Affordance grounding (INTRA). Unlike prior arts, INTRA recasts this problem as representation learning to identify unique features of interactions through contrastive learning with exocentric images only, eliminating the need for paired datasets. Moreover, we leverage vision-language model embeddings for performing affordance grounding flexibly with any text, designing text-conditioned affordance map generation to reflect interaction relationship for contrastive learning and enhancing robustness with our text synonym augmentation. Our method outperformed prior arts on diverse datasets such as AGD20K, IIT-AFF, CAD and UMD. Additionally, experimental results demonstrate that our method has remarkable domain scalability for synthesized images / illustrations and is capable of performing affordance grounding for novel interactions and objects.

Paperid: 4716, https://arxiv.org/pdf/2601.21971.pdf

Abstract:
Imitation learning has achieved remarkable success in robotic manipulation, yet its application to surgical robotics remains challenging due to data scarcity, constrained workspaces, and the need for an exceptional level of safety and predictability. We present a supervised Mixture-of-Experts (MoE) architecture designed for phase-structured surgical manipulation tasks, which can be added on top of any autonomous policy. Unlike prior surgical robot learning approaches that rely on multi-camera setups or thousands of demonstrations, we show that a lightweight action decoder policy like Action Chunking Transformer (ACT) can learn complex, long-horizon manipulation from less than 150 demonstrations using solely stereo endoscopic images, when equipped with our architecture. We evaluate our approach on the collaborative surgical task of bowel grasping and retraction, where a robot assistant interprets visual cues from a human surgeon, executes targeted grasping on deformable tissue, and performs sustained retraction. We benchmark our method against state-of-the-art Vision-Language-Action (VLA) models and the standard ACT baseline. Our results show that generalist VLAs fail to acquire the task entirely, even under standard in-distribution conditions. Furthermore, while standard ACT achieves moderate success in-distribution, adopting a supervised MoE architecture significantly boosts its performance, yielding higher success rates in-distribution and demonstrating superior robustness in out-of-distribution scenarios, including novel grasp locations, reduced illumination, and partial occlusions. Notably, it generalizes to unseen testing viewpoints and also transfers zero-shot to ex vivo porcine tissue without additional training, offering a promising pathway toward in vivo deployment. To support this, we present qualitative preliminary results of policy roll-outs during in vivo porcine surgery.

Paperid: 4717, https://arxiv.org/pdf/2601.15370.pdf

Abstract:
Mixture-of-Experts layers achieve compute efficiency through weight sparsity: each token activates only a subset of experts. Data sparsity, where each expert processes only a subset of tokens, offers a complementary axis. Expert-choice routing implements data sparsity directly but violates causality in autoregressive models, creating train-inference mismatch. We recover data sparsity within causal token-choice MoE by leveraging zero-compute (null) experts within the routing pool. When a token routes to null experts, those slots consume no compute. The standard load balancing objective trains the model to uniformly use all experts (real and null) therefore creating data sparsity in expectation without the causality violations. We evaluate on vision-language model training, where data heterogeneity is pronounced: vision encoders produce many low-information tokens while text tokens are denser. At matched expected FLOPs, composing weight and data sparsity yields a more compute-efficient frontier than weight sparsity alone, with gains in training loss and downstream performance. The model learns implicit modality-aware allocation, routing vision tokens to null experts more aggressively than text, without explicit modality routing.

Paperid: 4718, https://arxiv.org/pdf/2601.12486.pdf

Abstract:
Shopping is a routine activity for sighted individuals, yet for people who are blind or have low vision (pBLV), locating and retrieving products in physical environments remains a challenge. This paper presents a multimodal wearable assistive system that integrates object detection with vision-language models to support independent product or item retrieval, with the goal of enhancing users'autonomy and sense of agency. The system operates through three phases: product search, which identifies target products using YOLO-World detection combined with embedding similarity and color histogram matching; product navigation, which provides spatialized sonification and VLM-generated verbal descriptions to guide users toward the target; and product correction, which verifies whether the user has reached the correct product and provides corrective feedback when necessary. Technical evaluation demonstrated promising performance across all modules, with product detection achieving near-perfect accuracy at close range and high accuracy when facing shelves within 1.5 m. VLM-based navigation achieved up to 94.4% accuracy, and correction accuracy exceeded 86% under optimal model configurations. These results demonstrate the system's potential to address the last-meter problem in assistive shopping. Future work will focus on user studies with pBLV participants and integration with multi-scale navigation ecosystems.

Paperid: 4719, https://arxiv.org/pdf/2601.12479.pdf

Abstract:
We introduce a method for decentralized person re-identification in robot swarms that leverages natural language as the primary representational modality. Unlike traditional approaches that rely on opaque visual embeddings -- high-dimensional feature vectors extracted from images -- the proposed method uses human-readable language to represent observations. Each robot locally detects and describes individuals using a vision-language model (VLM), producing textual descriptions of appearance instead of feature vectors. These descriptions are compared and clustered across the swarm without centralized coordination, allowing robots to collaboratively group observations of the same individual. Each cluster is distilled into a representative description by a language model, providing an interpretable, concise summary of the swarm's collective perception. This approach enables natural-language querying, enhances transparency, and supports explainable swarm behavior. Preliminary experiments demonstrate competitive performance in identity consistency and interpretability compared to embedding-based methods, despite current limitations in text similarity and computational load. Ongoing work explores refined similarity metrics, semantic navigation, and the extension of language-based perception to environmental elements. This work prioritizes decentralized perception and communication, while active navigation remains an open direction for future study.

Paperid: 4720, https://arxiv.org/pdf/2601.12468.pdf

Abstract:
Out-of-distribution (OOD) detection remains a fundamental challenge for deep neural networks, particularly due to overconfident predictions on unseen OOD samples during testing. We reveal a key insight: OOD samples predicted as the same class, or given high probabilities for it, are visually more similar to each other than to the true in-distribution (ID) samples. Motivated by this class-specific observation, we propose DCAC (Dynamic Class-Aware Cache), a training-free, test-time calibration module that maintains separate caches for each ID class to collect high-entropy samples and calibrate the raw predictions of input samples. DCAC leverages cached visual features and predicted probabilities through a lightweight two-layer module to mitigate overconfident predictions on OOD samples. This module can be seamlessly integrated with various existing OOD detection methods across both unimodal and vision-language models while introducing minimal computational overhead. Extensive experiments on multiple OOD benchmarks demonstrate that DCAC significantly enhances existing methods, achieving substantial improvements, i.e., reducing FPR95 by 6.55% when integrated with ASH-S on ImageNet OOD benchmark.

Paperid: 4721, https://arxiv.org/pdf/2601.11269.pdf

Abstract:
Visuomotor policies often leverage large pre-trained Vision Transformers (ViTs) for their powerful generalization capabilities. However, their significant data requirements present a major challenge in the data-scarce context of most robotic learning settings, where compact CNNs with strong inductive biases can be more easily optimized. To address this trade-off, we introduce X-Distill, a simple yet highly effective method that synergizes the strengths of both architectures. Our approach involves an offline, cross-architecture knowledge distillation, transferring the rich visual representations of a large, frozen DINOv2 teacher to a compact ResNet-18 student on the general-purpose ImageNet dataset. This distilled encoder, now endowed with powerful visual priors, is then jointly fine-tuned with a diffusion policy head on the target manipulation tasks. Extensive experiments on $34$ simulated benchmarks and $5$ challenging real-world tasks demonstrate that our method consistently outperforms policies equipped with from-scratch ResNet or fine-tuned DINOv2 encoders. Notably, X-Distill also surpasses 3D encoders that utilize privileged point cloud observations or much larger Vision-Language Models. Our work highlights the efficacy of a simple, well-founded distillation strategy for achieving state-of-the-art performance in data-efficient robotic manipulation.

Paperid: 4722, https://arxiv.org/pdf/2601.09658.pdf

Abstract:
Estimating physically accurate, simulation-ready garments from a single image is challenging due to the absence of image-to-physics datasets and the ill-posed nature of this problem. Prior methods either require multi-view capture and expensive differentiable simulation or predict only garment geometry without the material properties required for realistic simulation. We propose a feed-forward framework that sidesteps these limitations by first fine-tuning a vision-language model to infer material composition and fabric attributes from real images, and then training a lightweight predictor that maps these attributes to the corresponding physical fabric parameters using a small dataset of material-physics measurements. Our approach introduces two new datasets (FTAG and T2P) and delivers simulation-ready garments from a single image without iterative optimization. Experiments show that our estimator achieves superior accuracy in material composition estimation and fabric attribute prediction, and by passing them through our physics parameter estimator, we further achieve higher-fidelity simulations compared to state-of-the-art image-to-garment methods.

Paperid: 4723, https://arxiv.org/pdf/2601.09147.pdf

Abstract:
Zero-Shot Anomaly Detection (ZSAD) leverages Vision-Language Models (VLMs) to enable supervision-free industrial inspection. However, existing ZSAD paradigms are constrained by single visual backbones, which struggle to balance global semantic generalization with fine-grained structural discriminability. To bridge this gap, we propose Synergistic Semantic-Visual Prompting (SSVP), that efficiently fuses diverse visual encodings to elevate model's fine-grained perception. Specifically, SSVP introduces the Hierarchical Semantic-Visual Synergy (HSVS) mechanism, which deeply integrates DINOv3's multi-scale structural priors into the CLIP semantic space. Subsequently, the Vision-Conditioned Prompt Generator (VCPG) employs cross-modal attention to guide dynamic prompt generation, enabling linguistic queries to precisely anchor to specific anomaly patterns. Furthermore, to address the discrepancy between global scoring and local evidence, the Visual-Text Anomaly Mapper (VTAM) establishes a dual-gated calibration paradigm. Extensive evaluations on seven industrial benchmarks validate the robustness of our method; SSVP achieves state-of-the-art performance with 93.0% Image-AUROC and 92.2% Pixel-AUROC on MVTec-AD, significantly outperforming existing zero-shot approaches.

Paperid: 4724, https://arxiv.org/pdf/2601.06891.pdf

Abstract:
Contrastive Language-Image Pre-training (CLIP) relies on Vision Transformers whose attention mechanism is susceptible to spurious correlations, and scales quadratically with resolution. To address these limitations, We present CLIMP, the first fully Mamba-based contrastive vision-language model that replaces both the vision and text encoders with Mamba. The new architecture encodes sequential structure in both vision and language, with VMamba capturing visual spatial inductive biases, reducing reliance on spurious correlations and producing an embedding space favorable for cross-modal retrieval and out-of-distribution robustness-surpassing OpenAI's CLIP-ViT-B by 7.5% on ImageNet-O. CLIMP naturally supports variable input resolutions without positional encoding interpolation or specialized training, achieving up to 6.6% higher retrieval accuracy at 16x training resolution while using 5x less memory and 1.8x fewer FLOPs. The autoregressive text encoder further overcomes CLIP's fixed context limitation, enabling dense captioning retrieval. Our findings suggest that Mamba exhibits advantageous properties for vision-language learning, making it a compelling alternative to Transformer-based CLIP.

Paperid: 4725, https://arxiv.org/pdf/2601.06870.pdf

Abstract:
Multimodal large language models (MLLMs) have demonstrated strong performance on vision-language tasks, yet their effectiveness on multimodal sentiment analysis remains constrained by the scarcity of high-quality training data, which limits accurate multimodal understanding and generalization. To alleviate this bottleneck, we leverage diffusion models to perform semantics-preserving augmentation on the video and audio modalities, expanding the multimodal training distribution. However, increasing data quantity alone is insufficient, as diffusion-generated samples exhibit substantial quality variation and noisy augmentations may degrade performance. We therefore propose DaQ-MSA (Denoising and Qualifying Diffusion Augmentations for Multimodal Sentiment Analysis), which introduces a quality scoring module to evaluate the reliability of augmented samples and assign adaptive training weights. By down-weighting low-quality samples and emphasizing high-fidelity ones, DaQ-MSA enables more stable learning. By integrating the generative capability of diffusion models with the semantic understanding of MLLMs, our approach provides a robust and generalizable automated augmentation strategy for training MLLMs without any human annotation or additional supervision.

Paperid: 4726, https://arxiv.org/pdf/2601.06461.pdf

Abstract:
Visual Reasoning CAPTCHAs (VRCs) combine visual scenes with natural-language queries that demand compositional inference over objects, attributes, and spatial relations. They are increasingly deployed as a primary defense against automated bots. Existing solvers fall into two paradigms: vision-centric, which rely on template-specific detectors but fail on novel layouts, and reasoning-centric, which leverage LLMs but struggle with fine-grained visual perception. Both lack the generality needed to handle heterogeneous VRC deployments. We present ViPer, a unified attack framework that integrates structured multi-object visual perception with adaptive LLM-based reasoning. ViPer parses visual layouts, grounds attributes to question semantics, and infers target coordinates within a modular pipeline. Evaluated on six major VRC providers (VTT, Geetest, NetEase, Dingxiang, Shumei, Xiaodun), ViPer achieves up to 93.2% success, approaching human-level performance across multiple benchmarks. Compared to prior solvers, GraphNet (83.2%), Oedipus (65.8%), and the Holistic approach (89.5%), ViPer consistently outperforms all baselines. The framework further maintains robustness across alternative LLM backbones (GPT, Grok, DeepSeek, Kimi), sustaining accuracy above 90%. To anticipate defense, we further introduce Template-Space Randomization (TSR), a lightweight strategy that perturbs linguistic templates without altering task semantics. TSR measurably reduces solver (i.e., attacker) performance. Our proposed design suggests directions for human-solvable but machine-resistant CAPTCHAs.

Paperid: 4727, https://arxiv.org/pdf/2601.05939.pdf

Abstract:
Hallucinations, generating responses inconsistent with the visual input, remain a critical limitation of large vision-language models (LVLMs), especially in open-ended tasks such as image captioning and visual reasoning. In this work, we probe the layer-wise generation dynamics that drive hallucinations and propose a training-free mitigation strategy. Employing the Logit Lens, we examine how LVLMs construct next-token distributions across decoder layers, uncovering a pronounced commitment-depth gap: truthful tokens accumulate probability mass on their final candidates earlier than hallucinatory ones. Drawing on this discovery, we introduce Context Embedding Injection (CEI), a lightweight method that harnesses the hidden state of the last input token-the context embedding-as a grounding signal to maintain visual fidelity throughout decoding and curb hallucinations. Evaluated on the CHAIR, AMBER, and MMHal-Bench benchmarks (with a maximum token length of 512), CEI outperforms state-of-the-art baselines across three LVLMs, with its dynamic variant yielding the lowest overall hallucination rates. By integrating novel mechanistic insights with a scalable intervention, this work advances the mitigation of hallucinations in LVLMs.

Paperid: 4728, https://arxiv.org/pdf/2601.01875.pdf

Abstract:
Automated pathology image analysis is central to clinical diagnosis, but clinicians still ask which slide features drive a model's decision and why. Vision-language models can produce natural language explanations, but these are often correlational and lack verifiable evidence. In this paper, we introduce an SQL-centered agentic framework that enables both feature measurement and reasoning to be auditable. Specifically, after extracting human-interpretable cellular features, Feature Reasoning Agents compose and execute SQL queries over feature tables to aggregate visual evidence into quantitative findings. A Knowledge Comparison Agent then evaluates these findings against established pathological knowledge, mirroring how pathologists justify diagnoses from measurable observations. Extensive experiments evaluated on two pathology visual question answering datasets demonstrate our method improves interpretability and decision traceability while producing executable SQL traces that link cellular measurements to diagnostic conclusions.

Paperid: 4729, https://arxiv.org/pdf/2512.24340.pdf

Abstract:
Recent advances in dermatological image analysis have been driven by large-scale annotated datasets; however, most existing benchmarks focus on dermatoscopic images and lack patient-authored queries and clinical context, limiting their applicability to patient-centered care. To address this gap, we introduce DermaVQA-DAS, an extension of the DermaVQA dataset that supports two complementary tasks: closed-ended question answering (QA) and dermatological lesion segmentation. Central to this work is the Dermatology Assessment Schema (DAS), a novel expert-developed framework that systematically captures clinically meaningful dermatological features in a structured and standardized form. DAS comprises 36 high-level and 27 fine-grained assessment questions, with multiple-choice options in English and Chinese. Leveraging DAS, we provide expert-annotated datasets for both closed QA and segmentation and benchmark state-of-the-art multimodal models. For segmentation, we evaluate multiple prompting strategies and show that prompt design impacts performance: the default prompt achieves the best results under Mean-of-Max and Mean-of-Mean evaluation aggregation schemes, while an augmented prompt incorporating both patient query title and content yields the highest performance under majority-vote-based microscore evaluation, achieving a Jaccard index of 0.395 and a Dice score of 0.566 with BiomedParse. For closed-ended QA, overall performance is strong across models, with average accuracies ranging from 0.729 to 0.798; o3 achieves the best overall accuracy (0.798), closely followed by GPT-4.1 (0.796), while Gemini-1.5-Pro shows competitive performance within the Gemini family (0.783). We publicly release DermaVQA-DAS, the DAS schema, and evaluation protocols to support and accelerate future research in patient-centered dermatological vision-language modeling (https://osf.io/72rp3).

Paperid: 4730, https://arxiv.org/pdf/2512.22457.pdf

Abstract:
Local railway committees need timely situational awareness after highway-rail grade crossing incidents, yet official Federal Railroad Administration (FRA) investigations can take days to weeks. We present a demo system that populates Highway-Rail Grade Crossing Incident Data (Form 57) from news in real time. Our approach addresses two core challenges: the form is visually irregular and semantically dense, and news is noisy. To solve these problems, we design a pipeline that first converts Form 57 into a JSON schema using a vision language model with sample aggregation, and then performs grouped question answering following the intent of the form layout to reduce ambiguity. In addition, we build an evaluation dataset by aligning scraped news articles with official FRA records and annotating retrievable information. We then assess our system against various alternatives in terms of information retrieval accuracy and coverage.

Paperid: 4731, https://arxiv.org/pdf/2512.21722.pdf

Abstract:
Socially compliant navigation requires robots to move safely and appropriately in human-centered environments by respecting social norms. However, social norms are often ambiguous, and in a single scenario, multiple actions may be equally acceptable. Most existing methods simplify this problem by assuming a single correct action, which limits their ability to handle real-world social uncertainty. In this work, we propose MAction-SocialNav, an efficient vision language model for socially compliant navigation that explicitly addresses action ambiguity, enabling generating multiple plausible actions within one scenario. To enhance the model's reasoning capability, we introduce a novel meta-cognitive prompt (MCP) method. Furthermore, to evaluate the proposed method, we curate a multi-action socially compliant navigation dataset that accounts for diverse conditions, including crowd density, indoor and outdoor environments, and dual human annotations. The dataset contains 789 samples, each with three-turn conversation, split into 710 training samples and 79 test samples through random selection. We also design five evaluation metrics to assess high-level decision precision, safety, and diversity. Extensive experiments demonstrate that the proposed MAction-SocialNav achieves strong social reasoning performance while maintaining high efficiency, highlighting its potential for real-world human robot navigation. Compared with zero-shot GPT-4o and Claude, our model achieves substantially higher decision quality (APG: 0.595 vs. 0.000/0.025) and safety alignment (ER: 0.264 vs. 0.642/0.668), while maintaining real-time efficiency (1.524 FPS, over 3x faster).

Paperid: 4732, https://arxiv.org/pdf/2512.21126.pdf

Abstract:
We have witnessed promising progress led by large language models (LLMs) and further vision language models (VLMs) in handling various queries as a general-purpose assistant. VLMs, as a bridge to connect the visual world and language corpus, receive both visual content and various text-only user instructions to generate corresponding responses. Though great success has been achieved by VLMs in various fields, in this work, we ask whether the existing VLMs can act as domain experts, accurately answering marine questions, which require significant domain expertise and address special domain challenges/requirements. To comprehensively evaluate the effectiveness and explore the boundary of existing VLMs, we construct the first large-scale marine VLM dataset and benchmark called MarineEval, with 2,000 image-based question-answering pairs. During our dataset construction, we ensure the diversity and coverage of the constructed data: 7 task dimensions and 20 capacity dimensions. The domain requirements are specially integrated into the data construction and further verified by the corresponding marine domain experts. We comprehensively benchmark 17 existing VLMs on our MarineEval and also investigate the limitations of existing models in answering marine research questions. The experimental results reveal that existing VLMs cannot effectively answer the domain-specific questions, and there is still a large room for further performance improvements. We hope our new benchmark and observations will facilitate future research. Project Page: http://marineeval.hkustvgd.com/

Paperid: 4733, https://arxiv.org/pdf/2512.20735.pdf

Abstract:
Human gaze provides essential cues for interpreting attention, intention, and social interaction in visual scenes, yet gaze understanding remains largely unexplored in current vision-language models (VLMs). While recent VLMs achieve strong scene-level reasoning across a range of visual tasks, there exists no benchmark that systematically evaluates or trains them for gaze interpretation, leaving open the question of whether gaze understanding can emerge from general-purpose vision-language pre-training. To address this gap, we introduce VL4Gaze, the first large-scale benchmark designed to investigate, evaluate, and unlock the potential of VLMs for gaze understanding. VL4Gaze contains 489K automatically generated question-answer pairs across 124K images and formulates gaze understanding as a unified VQA problem through four complementary tasks: (1) gaze object description, (2) gaze direction description, (3) gaze point location, and (4) ambiguous question recognition. We comprehensively evaluate both commercial and open-source VLMs under in-context learning and fine-tuning settings. The results show that even large-scale VLMs struggle to reliably infer gaze semantics and spatial localization without task-specific supervision. In contrast, training on VL4Gaze brings substantial and consistent improvements across all tasks, highlighting the importance of targeted multi-task supervision for developing gaze understanding capabilities in VLMs. We will release the dataset and code to support further research and development in this direction.

Paperid: 4734, https://arxiv.org/pdf/2512.19526.pdf

Abstract:
Understanding the physical world is essential for generalist AI agents. However, it remains unclear whether state-of-the-art vision perception models (e.g., large VLMs) can reason physical properties quantitatively. Existing evaluations are predominantly VQA-based and qualitative, offering limited insight into whether these models can infer the kinematic quantities of moving objects from video observations. To address this, we present QuantiPhy, the first benchmark designed to quantitatively measure a VLM's physical reasoning ability. Comprising more than 3.3K video-text instances with numerical ground truth, QuantiPhy evaluates a VLM's performance on estimating an object's size, velocity, and acceleration at a given timestamp, using one of these properties as an input prior. The benchmark standardizes prompts and scoring to assess numerical accuracy, enabling fair comparisons across models. Our experiments on state-of-the-art VLMs reveal a consistent gap between their qualitative plausibility and actual numerical correctness. We further provide an in-depth analysis of key factors like background noise, counterfactual priors, and strategic prompting and find that state-of-the-art VLMs lean heavily on pre-trained world knowledge rather than faithfully using the provided visual and textual inputs as references when reasoning kinematic properties quantitatively. QuantiPhy offers the first rigorous, scalable testbed to move VLMs beyond mere verbal plausibility toward a numerically grounded physical understanding.

Paperid: 4735, https://arxiv.org/pdf/2512.15160.pdf

Abstract:
Recent spatial intelligence approaches typically attach 3D cues to 2D reasoning pipelines or couple MLLMs with black-box reconstruction modules, leading to weak spatial consistency, limited viewpoint diversity, and evidence chains that cannot be traced back to supporting views. Frameworks for "thinking with images" (e.g., ChatGPT-o3 and DeepEyes) show that stepwise multimodal reasoning can emerge by interleaving hypothesis formation with active acquisition of visual evidence, but they do not address three key challenges in spatial Chain-of-Thought (CoT): building global space perception under strict token budgets, explicitly associating 3D hypotheses with video frames for verification, and designing spatially grounded rewards for reinforcement learning. To address these issues, we present EagleVision, a dual-stage framework for progressive spatial cognition through macro perception and micro verification. In the macro perception stage, EagleVision employs a semantics-perspective-fusion determinantal point process (SPF-DPP) to select a compact set of geometry- and semantics-aware keyframes from long videos under a fixed token budget. In the micro verification stage, we formalize spatial CoT as BEV-grounded pose querying: the agent iteratively predicts poses on a BEV plane, retrieves the nearest real frames, and is trained purely by reinforcement learning with a spatial grounding reward that scores the consistency between predicted poses and observed views. On VSI-Bench, EagleVision achieves state-of-the-art performance among open-source vision-language models, demonstrating strong and generalizable spatial understanding.

Paperid: 4736, https://arxiv.org/pdf/2512.14989.pdf

Abstract:
Multimodal scientific reasoning remains a significant challenge for large language models (LLMs), particularly in chemistry, where problem-solving relies on symbolic diagrams, molecular structures, and structured visual data. Here, we systematically evaluate 40 proprietary and open-source multimodal LLMs, including GPT-5, o3, Gemini-2.5-Pro, and Qwen2.5-VL, on a curated benchmark of Olympiad-style chemistry questions drawn from over two decades of U.S. National Chemistry Olympiad (USNCO) exams. These questions require integrated visual and textual reasoning across diverse modalities. We find that many models struggle with modality fusion, where in some cases, removing the image even improves accuracy, indicating misalignment in vision-language integration. Chain-of-Thought prompting consistently enhances both accuracy and visual grounding, as demonstrated through ablation studies and occlusion-based interpretability. Our results reveal critical limitations in the scientific reasoning abilities of current MLLMs, providing actionable strategies for developing more robust and interpretable multimodal systems in chemistry. This work provides a timely benchmark for measuring progress in domain-specific multimodal AI and underscores the need for further advances at the intersection of artificial intelligence and scientific reasoning.

Paperid: 4737, https://arxiv.org/pdf/2512.11490.pdf

Abstract:
Satellite imagery differs fundamentally from natural images: its aerial viewpoint, very high resolution, diverse scale variations, and abundance of small objects demand both region-level spatial reasoning and holistic scene understanding. Current remote-sensing approaches remain fragmented between dual-encoder retrieval models, which excel at large-scale cross-modal search but cannot interleave modalities, and generative assistants, which support region-level interpretation but lack scalable retrieval capabilities. We propose $\textbf{VLM2GeoVec}$, an instruction-following, single-encoder vision-language model trained contrastively to embed interleaved inputs (images, text, bounding boxes, and geographic coordinates) in a unified vector space. Our single encoder interleaves all inputs into one joint embedding trained with a contrastive loss, eliminating multi-stage pipelines and task-specific modules. To evaluate its versatility, we introduce $\textbf{RSMEB}$, a novel benchmark covering key remote-sensing embedding applications: scene classification; cross-modal search; compositional retrieval; visual-question answering; visual grounding and region-level reasoning; and semantic geospatial retrieval. On RSMEB, it achieves $\textbf{26.6%}$ P@1 on region-caption retrieval (+25 pp vs. dual-encoder baselines), $\textbf{32.5%}$ P@1 on referring-expression retrieval (+19 pp), and $\textbf{17.8%}$ P@1 on semantic geo-localization retrieval (over $3\times$ prior best), while matching or exceeding specialized baselines on conventional tasks such as scene classification and cross-modal retrieval. VLM2GeoVec unifies scalable retrieval with region-level spatial reasoning, enabling cohesive multimodal analysis in remote sensing. We will publicly release the code, checkpoints, and data upon acceptance.

Paperid: 4738, https://arxiv.org/pdf/2512.10316.pdf

Abstract:
Weakly supervised semantic segmentation (WSSS) in histopathology relies heavily on classification backbones, yet these models often localize only the most discriminative regions and struggle to capture the full spatial extent of tissue structures. Vision-language models such as CONCH offer rich semantic alignment and morphology-aware representations, while modern segmentation backbones like SegFormer preserve fine-grained spatial cues. However, combining these complementary strengths remains challenging, especially under weak supervision and without dense annotations. We propose a prototype learning framework for WSSS in histopathological images that integrates morphology-aware representations from CONCH, multi-scale structural cues from SegFormer, and text-guided semantic alignment to produce prototypes that are simultaneously semantically discriminative and spatially coherent. To effectively leverage these heterogeneous sources, we introduce text-guided prototype initialization that incorporates pathology descriptions to generate more complete and semantically accurate pseudo-masks. A structural distillation mechanism transfers spatial knowledge from SegFormer to preserve fine-grained morphological patterns and local tissue boundaries during prototype learning. Our approach produces high-quality pseudo masks without pixel-level annotations, improves localization completeness, and enhances semantic consistency across tissue types. Experiments on BCSS-WSSS datasets demonstrate that our prototype learning framework outperforms existing WSSS methods while remaining computationally efficient through frozen foundation model backbones and lightweight trainable adapters.

Paperid: 4739, https://arxiv.org/pdf/2512.10314.pdf

Abstract:
Weakly supervised semantic segmentation (WSSS) in histopathology seeks to reduce annotation cost by learning from image-level labels, yet it remains limited by inter-class homogeneity, intra-class heterogeneity, and the region-shrinkage effect of CAM-based supervision. We propose a simple and effective prototype-driven framework that leverages vision-language alignment to improve region discovery under weak supervision. Our method integrates CoOp-style learnable prompt tuning to generate text-based prototypes and combines them with learnable image prototypes, forming a dual-modal prototype bank that captures both semantic and appearance cues. To address oversmoothing in ViT representations, we incorporate a multi-scale pyramid module that enhances spatial precision and improves localization quality. Experiments on the BCSS-WSSS benchmark show that our approach surpasses existing state-of-the-art methods, and detailed analyses demonstrate the benefits of text description diversity, context length, and the complementary behavior of text and image prototypes. These results highlight the effectiveness of jointly leveraging textual semantics and visual prototype learning for WSSS in digital pathology.

Paperid: 4740, https://arxiv.org/pdf/2512.07273.pdf

Abstract:
Gloss-free sign language translation (SLT) is hindered by two key challenges: **inadequate sign representation** that fails to capture nuanced visual cues, and **sentence-level semantic misalignment** in current LLM-based methods, which limits translation quality. To address these issues, we propose a three-stage **r**einforcing **v**ision-**l**anguage **f**ramework (**RVLF**). We build a large vision-language model (LVLM) specifically designed for sign language, and then combine it with reinforcement learning (RL) to adaptively enhance translation performance. First, for a sufficient representation of sign language, RVLF introduces an effective semantic representation learning mechanism that fuses skeleton-based motion cues with semantically rich visual features extracted via DINOv2, followed by instruction tuning to obtain a strong SLT-SFT baseline. Then, to improve sentence-level semantic misalignment, we introduce a GRPO-based optimization strategy that fine-tunes the SLT-SFT model with a reward function combining translation fidelity (BLEU) and sentence completeness (ROUGE), yielding the optimized model termed SLT-GRPO. Our conceptually simple framework yields substantial gains under the gloss-free SLT setting without pre-training on any external large-scale sign language datasets, improving BLEU-4 scores by +5.1, +1.11, +1.4, and +1.61 on the CSL-Daily, PHOENIX-2014T, How2Sign, and OpenASL datasets, respectively. To the best of our knowledge, this is the first work to incorporate GRPO into SLT. Extensive experiments and ablation studies validate the effectiveness of GRPO-based optimization in enhancing both translation quality and semantic consistency.

Paperid: 4741, https://arxiv.org/pdf/2512.06276.pdf

Abstract:
Referring Expression Comprehension (REC) is a vision-language task that localizes a specific image region based on a textual description. Existing REC benchmarks primarily evaluate perceptual capabilities and lack interpretable scoring mechanisms, which cannot reveal the grounding capability of Multi-modal Large Language Model (MLLM) across different cognitive abilities. To address this limitation, we introduce RefBench-PRO, a comprehensive REC benchmark, which decomposes referring expressions into two core dimensions, i.e., perception and reasoning, and further subdivides them into six progressively challenging tasks, such as attribute, position, interaction, commonsense, relation and reject. We also develop a fully automated data-generation pipeline that produces diverse referring expressions across these six sub-dimensions. Furthermore, We propose Ref-R1, an RL-based learning scheme, which incorporates Dynamic IoU-based GRPO to improve localization accuracy under increasingly complex reasoning conditions, establishing a stronger baseline for REC. Extensive experiments demonstrate that our RefBench-PRO enables interpretable evaluation of MLLM on referring expression comprehension, presenting greater challenges in both perception and reasoning.

Paperid: 4742, https://arxiv.org/pdf/2511.23170.pdf

Abstract:
Contrastive vision-language pre-training frameworks such as CLIP have demonstrated impressive zero-shot performance across a range of vision-language tasks. Recent studies have shown that aligning individual text tokens with specific image patches or regions enhances fine-grained compositional understanding. However, it remains challenging to capture compositional semantics that span multiple image regions. To address this limitation, we propose PowerCLIP, a novel contrastive pre-training framework enhanced by powerset alignment, which exhaustively optimizes region-to-phrase alignments by minimizing the loss defined between powersets of image regions and textual parse trees. Since the naive powerset construction incurs exponential computational cost due to the combinatorial explosion in the number of region subsets, we introduce efficient non-linear aggregators (NLAs) that reduce complexity from O(2^M) to O(M) with respect to the number of regions M, while approximating the exact loss value with arbitrary precision. Our extensive experiments demonstrate that PowerCLIP outperforms state-of-the-art methods in zero-shot classification and retrieval tasks, underscoring the compositionality and robustness of our approach. Our code will be made publicly available.

Paperid: 4743, https://arxiv.org/pdf/2511.22256.pdf

Abstract:
Despite significant strides in medical foundation models, the ultrasound domain lacks a comprehensive solution capable of bridging low-level Ultrasound Grounded Perception (e.g., segmentation, localization) and high-level Ultrasound Comprehensive Interpretation (e.g., diagnosis, reasoning). To bridge this gap, we propose UMind-VL, a unified foundation model designed to synergize pixel-level structural understanding with complex clinical reasoning. We first introduce UMind-DS, a large-scale multimodal dataset comprising 1.2 million ultrasound image-text pairs across 16 anatomical regions, enriching standard data with pixel-level annotations and clinician-validated rationales. Architecturally, UMind-VL incorporates a lightweight Dynamic Convolutional Mask Decoder that generates masks via dynamic kernels conditioned on LLM outputs. This design, combined with task-specific tokens, unifies segmentation, detection, geometric measurement, and diagnosis tasks within a single framework. Extensive evaluations demonstrate that UMind-VL significantly outperforms existing generalist multimodal models and achieves performance on par with, or superior to, state-of-the-art specialist models across segmentation, detection, keypoint localization, and diagnostic reasoning benchmarks, while maintaining strong generalization ability. We demonstrate the capability of UMind-VL in Figure 1.

Paperid: 4744, https://arxiv.org/pdf/2511.20770.pdf

Abstract:
Image encoders, a fundamental component of vision-language models (VLMs), are typically pretrained independently before being aligned with a language model. This standard paradigm results in encoders that process images agnostically, without regard to the specific downstream task or text query. To address this limitation, we propose the Text-Guided Semantic Image Encoder (TIE), which generates image representations conditioned on the input text query. VLMs equipped with TIE outperform their conventional counterparts by +1.5 and +1.3 points on average across nine image-to-text benchmarks at the 1B and 3B scales, respectively, with gains reaching up to 6 points on tasks such as DocVQA and InfoVQA. Moreover, TIE-based VLMs attain superior performance while utilizing only half as many image tiles (tokens), resulting in notably improved inference efficiency. TIE also generalizes well with generic queries, indicating that text-conditioned training effectively optimizes the encoder to capture key visual features. Qualitative analysis confirms that TIE consistently attends to query-relevant regions, enhancing both interpretability and query-specific grounding.

Paperid: 4745, https://arxiv.org/pdf/2511.20697.pdf

Abstract:
Understanding complete musical scores requires reasoning over symbolic structures such as pitch, rhythm, harmony, and form. Despite the rapid progress of Large Language Models (LLMs) and Vision-Language Models (VLMs) in natural language and multimodal tasks, their ability to comprehend musical notation remains underexplored. We introduce Musical Score Understanding Benchmark (MSU-Bench), the first large-scale, human-curated benchmark for evaluating score-level musical understanding across both textual (ABC notation) and visual (PDF) modalities. MSU-Bench comprises 1,800 generative question-answer (QA) pairs drawn from works spanning Bach, Beethoven, Chopin, Debussy, and others, organised into four progressive levels of comprehension: Onset Information, Notation & Note, Chord & Harmony, and Texture & Form. Through extensive zero-shot and fine-tuned evaluations of over 15+ state-of-the-art (SOTA) models, we reveal sharp modality gaps, fragile level-wise success rates, and the difficulty of sustaining multilevel correctness. Fine-tuning markedly improves performance in both modalities while preserving general knowledge, establishing MSU-Bench as a rigorous foundation for future research at the intersection of Artificial Intelligence (AI), musicological, and multimodal reasoning.

Paperid: 4746, https://arxiv.org/pdf/2511.18373.pdf

Abstract:
Vision Language Models (VLMs) perform well on standard video tasks but struggle with physics-driven reasoning involving motion dynamics and spatial interactions. This limitation reduces their ability to interpret real or AI-generated content (AIGC) videos and to generate physically consistent content. We present an approach that addresses this gap by translating physical-world context cues into interpretable representations aligned with VLMs' perception, comprehension, and reasoning. We introduce MASS-Bench, a comprehensive benchmark consisting of 4,350 real-world and AIGC videos and 8,361 free-form video question-answering pairs focused on physics-related comprehension tasks, with detailed annotations including visual detections, sub-segment grounding, and full-sequence 3D motion tracking of entities. We further present MASS, a model-agnostic method that injects spatial-temporal signals into the VLM language space via depth-based 3D encoding and visual grounding, coupled with a motion tracker for object dynamics. To strengthen cross-modal alignment and reasoning, we apply reinforcement fine-tuning. Experiments and ablations show that our refined VLMs outperform comparable and larger baselines, as well as prior state-of-the-art models, by 8.7% and 6.0%, achieving performance comparable to close-source SoTA VLMs such as Gemini-2.5-Flash on physics reasoning and comprehension. These results validate the effectiveness of our approach.

Paperid: 4747, https://arxiv.org/pdf/2511.16175.pdf

Abstract:
Recent advances in Vision-Language-Action (VLA) models demonstrate that visual signals can effectively complement sparse action supervisions. However, letting VLA directly predict high-dimensional visual states can distribute model capacity and incur prohibitive training cost, while compressing visual states into more compact supervisory signals inevitably incurs information bottlenecks. Moreover, existing methods often suffer from poor comprehension and reasoning capabilities due to the neglect of language supervision. This paper introduces Mantis, a novel framework featuring a Disentangled Visual Foresight (DVF) to tackle these issues. Specifically, Mantis decouples visual foresight prediction from the backbone with the combination of meta queries and a diffusion Transformer (DiT) head. With the current visual state provided to the DiT via a residual connection, a simple next-state prediction objective enables the meta queries to automatically capture the latent actions that delineate the visual trajectory, and hence boost the learning of explicit actions. The disentanglement reduces the burden of the VLA backbone, enabling it to maintain comprehension and reasoning capabilities through language supervision. Empirically, pretrained on human manipulation videos, robot demonstrations, and image-text pairs, Mantis achieves a 96.7% success rate on LIBERO benchmark after fine-tuning, surpassing powerful baselines while exhibiting high convergence speed. Real-world evaluations show that Mantis outperforms $π_{0.5}$, a leading open-source VLA model, particularly in instruction-following capability, generalization to unseen instructions, and reasoning ability. Code and weights are released to support the open-source community.

Paperid: 4748, https://arxiv.org/pdf/2511.15825.pdf

Abstract:
IMACT-CXR is an interactive multi-agent conversational tutor that helps trainees interpret chest X-rays by unifying spatial annotation, gaze analysis, knowledge retrieval, and image-grounded reasoning in a single AutoGen-based workflow. The tutor simultaneously ingests learner bounding boxes, gaze samples, and free-text observations. Specialized agents evaluate localization quality, generate Socratic coaching, retrieve PubMed evidence, suggest similar cases from REFLACX, and trigger NV-Reason-CXR-3B for vision-language reasoning when mastery remains low or the learner explicitly asks. Bayesian Knowledge Tracing (BKT) maintains skill-specific mastery estimates that drive both knowledge reinforcement and case similarity retrieval. A lung-lobe segmentation module derived from a TensorFlow U-Net enables anatomically aware gaze feedback, and safety prompts prevent premature disclosure of ground-truth labels. We describe the system architecture, implementation highlights, and integration with the REFLACX dataset for real DICOM cases. IMACT-CXR demonstrates responsive tutoring flows with bounded latency, precise control over answer leakage, and extensibility toward live residency deployment. Preliminary evaluation shows improved localization and diagnostic reasoning compared to baselines.

Paperid: 4749, https://arxiv.org/pdf/2511.09973.pdf

Abstract:
Contrastive pre-trained vision-language models, such as CLIP, demonstrate strong generalization abilities in zero-shot classification by leveraging embeddings extracted from image and text encoders. This paper aims to robustly fine-tune these vision-language models on in-distribution (ID) data without compromising their generalization abilities in out-of-distribution (OOD) and zero-shot settings. Current robust fine-tuning methods tackle this challenge by reusing contrastive learning, which was used in pre-training, for fine-tuning. However, we found that these methods distort the geometric structure of the embeddings, which plays a crucial role in the generalization of vision-language models, resulting in limited OOD and zero-shot performance. To address this, we propose Difference Vector Equalization (DiVE), which preserves the geometric structure during fine-tuning. The idea behind DiVE is to constrain difference vectors, each of which is obtained by subtracting the embeddings extracted from the pre-trained and fine-tuning models for the same data sample. By constraining the difference vectors to be equal across various data samples, we effectively preserve the geometric structure. Therefore, we introduce two losses: average vector loss (AVL) and pairwise vector loss (PVL). AVL preserves the geometric structure globally by constraining difference vectors to be equal to their weighted average. PVL preserves the geometric structure locally by ensuring a consistent multimodal alignment. Our experiments demonstrate that DiVE effectively preserves the geometric structure, achieving strong results across ID, OOD, and zero-shot metrics.

Paperid: 4750, https://arxiv.org/pdf/2511.05017.pdf

Abstract:
In this work, we identify an inherent bias in prevailing LVLM architectures toward the language modality, largely resulting from the common practice of simply appending visual embeddings to the input text sequence. To address this, we propose a simple yet effective method that refines textual embeddings by integrating average-pooled visual features. Our approach demonstrably improves visual grounding and significantly reduces hallucinations on established benchmarks. While average pooling offers a straightforward, robust, and efficient means of incorporating visual information, we believe that more sophisticated fusion methods could further enhance visual grounding and cross-modal alignment. Given that the primary focus of this work is to highlight the modality imbalance and its impact on hallucinations -- and to show that refining textual embeddings with visual information mitigates this issue -- we leave exploration of advanced fusion strategies for future work.

Paperid: 4751, https://arxiv.org/pdf/2511.01694.pdf

Abstract:
Vision-language pre-trained models, such as CLIP, have established new benchmarks in multimodal data mining. In such models, few-shot fine-tuning is a major challenge to achieve optimal performance on both in-distribution (ID) and out-of-distribution (OOD) datasets, especially when labeled data is scarce. Most existing fine-tuning approaches rely on first-order gradient-based optimizers, which typically suffer from slow convergence, sensitivity to step-size hyperparameters, and poor generalization in OOD settings. In contrast, second-order methods utilize local curvature information of the loss landscape to adjust the update step size. This is particularly beneficial for CLIP models, whose non-convex loss functions often contain sharp critical points. In such cases, natural gradient direction can offer more substantial and efficient per-iteration updates when fine-tuning with limited data. Natural Gradient Descent (NGD) is obtained by preconditioning the standard gradient with the inverse Fisher Information Matrix (FIM), which is computationally expensive for large models. To address this, we propose a Bayesian approximation of NGD using a Kalman filter for CLIP models. Our method combines the benefits of second-order optimization with Bayesian inference, which enhances generalization while providing uncertainty quantification. Extensive experiments conducted on diverse image classification datasets demonstrate that our algorithm consistently achieves superior--or comparable--ID performance and improved OOD robustness compared to state-of-the-art baselines. To the best of our knowledge, this work represents the first successful application of Kalman filtering to fine-tuning CLIP-based models, which enables more robust and efficient learning in vision-language tasks.

Paperid: 4752, https://arxiv.org/pdf/2511.00511.pdf

Abstract:
Video generative models pretrained on large-scale datasets can produce high-quality videos, but are often conditioned on text or a single image, limiting controllability and applicability. We introduce ID-Composer, a novel framework that addresses this gap by tackling multi-subject video generation from a text prompt and reference images. This task is challenging as it requires preserving subject identities, integrating semantics across subjects and modalities, and maintaining temporal consistency. To faithfully preserve the subject consistency and textual information in synthesized videos, ID-Composer designs a hierarchical identity-preserving attention mechanism, which effectively aggregates features within and across subjects and modalities. To effectively allow for the semantic following of user intention, we introduce semantic understanding via pretrained vision-language model (VLM), leveraging VLM's superior semantic understanding to provide fine-grained guidance and capture complex interactions between multiple subjects. Considering that standard diffusion loss often fails in aligning the critical concepts like subject ID, we employ an online reinforcement learning phase to drive the overall training objective of ID-Composer into RLVR. Extensive experiments demonstrate that our model surpasses existing methods in identity preservation, temporal consistency, and video quality.

Paperid: 4753, https://arxiv.org/pdf/2510.21860.pdf

Abstract:
We present Butter-Bench, a benchmark evaluating large language model (LLM) controlled robots for practical intelligence, defined as the ability to navigate the messiness of the physical world. Current state-of-the-art robotic systems use a hierarchical architecture with LLMs in charge of high-level reasoning, and a Vision Language Action (VLA) model for low-level control. Butter-Bench evaluates the LLM part in isolation from the VLA. Although LLMs have repeatedly surpassed humans in evaluations requiring analytical intelligence, we find humans still outperform LLMs on Butter-Bench. The best LLMs score 40% on Butter-Bench, while the mean human score is 95%. LLMs struggled the most with multi-step spatial planning and social understanding. We also evaluate LLMs that are fine-tuned for embodied reasoning and conclude that this training does not improve their score on Butter-Bench.

Paperid: 4754, https://arxiv.org/pdf/2510.20818.pdf

Abstract:
A fundamental challenge in robot navigation lies in learning policies that generalize across diverse environments while conforming to the unique physical constraints and capabilities of a specific embodiment (e.g., quadrupeds can walk up stairs, but rovers cannot). We propose VAMOS, a hierarchical VLA that decouples semantic planning from embodiment grounding: a generalist planner learns from diverse, open-world data, while a specialist affordance model learns the robot's physical constraints and capabilities in safe, low-cost simulation. We enabled this separation by carefully designing an interface that lets a high-level planner propose candidate paths directly in image space that the affordance model then evaluates and re-ranks. Our real-world experiments show that VAMOS achieves higher success rates in both indoor and complex outdoor navigation than state-of-the-art model-based and end-to-end learning methods. We also show that our hierarchical design enables cross-embodied navigation across legged and wheeled robots and is easily steerable using natural language. Real-world ablations confirm that the specialist model is key to embodiment grounding, enabling a single high-level planner to be deployed across physically distinct wheeled and legged robots. Finally, this model significantly enhances single-robot reliability, achieving 3X higher success rates by rejecting physically infeasible plans. Website: https://vamos-vla.github.io/

Paperid: 4755, https://arxiv.org/pdf/2510.19268.pdf

Abstract:
Long-horizon routing tasks of deformable linear objects (DLOs), such as cables and ropes, are common in industrial assembly lines and everyday life. These tasks are particularly challenging because they require robots to manipulate DLO with long-horizon planning and reliable skill execution. Successfully completing such tasks demands adapting to their nonlinear dynamics, decomposing abstract routing goals, and generating multi-step plans composed of multiple skills, all of which require accurate high-level reasoning during execution. In this paper, we propose a fully autonomous hierarchical framework for solving challenging DLO routing tasks. Given an implicit or explicit routing goal expressed in language, our framework leverages vision-language models~(VLMs) for in-context high-level reasoning to synthesize feasible plans, which are then executed by low-level skills trained via reinforcement learning. To improve robustness in long horizons, we further introduce a failure recovery mechanism that reorients the DLO into insertion-feasible states. Our approach generalizes to diverse scenes involving object attributes, spatial descriptions, as well as implicit language commands. It outperforms the next best baseline method by nearly 50% and achieves an overall success rate of 92.5% across long-horizon routing scenarios.

Paperid: 4756, https://arxiv.org/pdf/2510.19160.pdf

Abstract:
Integration of diverse data will be a pivotal step towards improving scientific explorations in many disciplines. This work establishes a vision-language model (VLM) that encodes videos with text input in order to classify various behaviors of a mouse existing in and engaging with their environment. Importantly, this model produces a behavioral vector over time for each subject and for each session the subject undergoes. The output is a valuable dataset that few programs are able to produce with as high accuracy and with minimal user input. Specifically, we use the open-source Qwen2.5-VL model and enhance its performance through prompts, in-context learning (ICL) with labeled examples, and frame-level preprocessing. We found that each of these methods contributes to improved classification, and that combining them results in strong F1 scores across all behaviors, including rare classes like freezing and fleeing, without any model fine-tuning. Overall, this model will support interdisciplinary researchers studying mouse behavior by enabling them to integrate diverse behavioral features, measured across multiple time points and environments, into a comprehensive dataset that can address complex research questions.

Paperid: 4757, https://arxiv.org/pdf/2510.18837.pdf

Abstract:
Federated learning (FL) enables multiple clients to collaboratively train machine learning models without exposing local data, balancing performance and privacy. However, domain shift and label heterogeneity across clients often hinder the generalization of the aggregated global model. Recently, large-scale vision-language models like CLIP have shown strong zero-shot classification capabilities, raising the question of how to effectively fine-tune CLIP across domains in a federated setting. In this work, we propose an adaptive federated prompt tuning framework, FedDEAP, to enhance CLIP's generalization in multi-domain scenarios. Our method includes the following three key components: (1) To mitigate the loss of domain-specific information caused by label-supervised tuning, we disentangle semantic and domain-specific features in images by using semantic and domain transformation networks with unbiased mappings; (2) To preserve domain-specific knowledge during global prompt aggregation, we introduce a dual-prompt design with a global semantic prompt and a local domain prompt to balance shared and personalized information; (3) To maximize the inclusion of semantic and domain information from images in the generated text features, we align textual and visual representations under the two learned transformations to preserve semantic and domain consistency. Theoretical analysis and extensive experiments on four datasets demonstrate the effectiveness of our method in enhancing the generalization of CLIP for federated image recognition across multiple domains.

Paperid: 4758, https://arxiv.org/pdf/2510.16290.pdf

Abstract:
Video anomaly detection (VAD) has rapidly advanced by recent development of Vision-Language Models (VLMs). While these models offer superior zero-shot detection capabilities, their immense computational cost and unstable visual grounding performance hinder real-time deployment. To overcome these challenges, we introduce Cerberus, a two-stage cascaded system designed for efficient yet accurate real-time VAD. Cerberus learns normal behavioral rules offline, and combines lightweight filtering with fine-grained VLM reasoning during online inference. The performance gains of Cerberus come from two key innovations: motion mask prompting and rule-based deviation detection. The former directs the VLM's attention to regions relevant to motion, while the latter identifies anomalies as deviations from learned norms rather than enumerating possible anomalies. Extensive evaluations on four datasets show that Cerberus on average achieves 57.68 fps on an NVIDIA L40S GPU, a 151.79$\times$ speedup, and 97.2\% accuracy comparable to the state-of-the-art VLM-based VAD methods, establishing it as a practical solution for real-time video analytics.

Paperid: 4759, https://arxiv.org/pdf/2510.14543.pdf

Abstract:
Aligning features from different modalities, is one of the most fundamental challenges for cross-modal tasks. Although pre-trained vision-language models can achieve a general alignment between image and text, they often require parameter-efficient fine-tuning (PEFT) for further adjustment. Today's PEFT methods (e.g., prompt tuning, LoRA-based, or adapter-based) always selectively fine-tune a subset of parameters, which can slightly adjust either visual or textual features, and avoid overfitting. In this paper, we are the first to highlight that all existing PEFT methods perform one-step adjustment. It is insufficient for complex (or difficult) datasets, where features of different modalities are highly entangled. To this end, we propose the first model-agnostic multi-step adjustment approach by learning a cross-modal velocity field: Flow Matching Alignment (FMA). Specifically, to ensure the correspondence between categories during training, we first utilize a fixed coupling strategy. Then, we propose a noise augmentation strategy to alleviate the data scarcity issue. Finally, we design an early-stopping solver, which terminates the transformation process earlier, improving both efficiency and accuracy. Compared with one-step PEFT methods, FMA has the multi-step rectification ability to achieve more precise and robust alignment. Extensive results have demonstrated that FMA can consistently yield significant performance gains across various benchmarks and backbones, particularly on challenging datasets.

Paperid: 4760, https://arxiv.org/pdf/2510.13808.pdf

Abstract:
Large Vision-Language Models (VLMs) excel at general visual reasoning tasks but exhibit sharp performance degradation when applied to novel domains with substantial distribution shifts from pretraining data. Existing domain adaptation approaches finetune different VLM components, but this often results in limited domain-specific feature learning or catastrophic forgetting of prior capabilities. To address these issues, we introduce Vision Contextualized Probing (VisCoP), which augments the VLM's vision encoder with a compact set of learnable visual probes. These probes enable efficient domain-specific adaptation with minimal modification to pretrained parameters. We evaluate VisCoP across three challenging domain adaptation settings-cross-view (exocentric to egocentric), cross-modal (RGB to depth), and cross-task (human understanding to robot control). Experiments show that VisCoP consistently outperforms existing adaptation strategies, achieving superior performance on target domains while effectively retaining source-domain knowledge.

Paperid: 4761, https://arxiv.org/pdf/2510.12845.pdf

Abstract:
Vision Language Models (VLMs) are pivotal for advancing perception in intelligent agents. Yet, evaluation of VLMs remains limited to predominantly English-centric benchmarks in which the image-text pairs comprise short texts. To evaluate VLM fine-grained abilities, in four languages under long-text settings, we introduce a novel multilingual benchmark VLURes featuring eight vision-and-language tasks, and a pioneering unrelatedness task, to probe the fine-grained Visual and Linguistic Understanding capabilities of VLMs across English, Japanese, and low-resource languages, Swahili, and Urdu. Our datasets, curated from web resources in the target language, encompass ten diverse image categories and rich textual context, introducing valuable vision-language resources for Swahili and Urdu. By prompting VLMs to generate responses and rationales, evaluated automatically and by native speakers, we uncover performance disparities across languages and tasks critical to intelligent agents, such as object recognition, scene understanding, and relationship understanding. We conducted evaluations of ten VLMs with VLURes. The best performing model, GPT-4o, achieves an overall accuracy of 90.8% and lags human performance by 6.7%, though the gap is larger for open-source models. The gap highlights VLURes' critical role in developing intelligent agents to tackle multi-modal visual reasoning.

Paperid: 4762, https://arxiv.org/pdf/2510.07869.pdf

Abstract:
Underwater environments present unique challenges for robotic operation, including complex hydrodynamics, limited visibility, and constrained communication. Although data-driven approaches have advanced embodied intelligence in terrestrial robots and enabled task-specific autonomous underwater robots, developing underwater intelligence capable of autonomously performing multiple tasks remains highly challenging, as large-scale, high-quality underwater datasets are still scarce. To address these limitations, we introduce USIM, a simulation-based multi-task Vision-Language-Action (VLA) dataset for underwater robots. USIM comprises over 561K frames from 1,852 trajectories, totaling approximately 15.6 hours of BlueROV2 interactions across 20 tasks in 9 diverse scenarios, ranging from visual navigation to mobile manipulation. Building upon this dataset, we propose U0, a VLA model for general underwater robots, which integrates binocular vision and other sensor modalities through multimodal fusion, and further incorporates a convolution-attention-based perception focus enhancement module (CAP) to improve spatial understanding and mobile manipulation. Across tasks such as inspection, obstacle avoidance, scanning, and dynamic tracking, the framework achieves a success rate of 80%, while in challenging mobile manipulation tasks, it reduces the distance to the target by 21.2% compared with baseline methods, demonstrating its effectiveness. USIM and U0 show that VLA models can be effectively applied to underwater robotic applications, providing a foundation for scalable dataset construction, improved task autonomy, and the practical realization of intelligent general underwater robots.

Paperid: 4763, https://arxiv.org/pdf/2510.02815.pdf

Abstract:
Cross-modal medical image synthesis research focuses on reconstructing missing imaging modalities from available ones to support clinical diagnosis. Driven by clinical necessities for flexible modality reconstruction, we explore K to N medical generation, where three critical challenges emerge: How can we model the heterogeneous contributions of different modalities to various target tasks? How can we ensure fusion quality control to prevent degradation from noisy information? How can we maintain modality identity consistency in multi-output generation? Driven by these clinical necessities, and drawing inspiration from SAM2's sequential frame paradigm and clinicians' progressive workflow of incrementally adding and selectively integrating multi-modal information, we treat multi-modal medical data as sequential frames with quality-driven selection mechanisms. Our key idea is to "learn" adaptive weights for each modality-task pair and "memorize" beneficial fusion patterns through progressive enhancement. To achieve this, we design three collaborative modules: PreWeightNet for global contribution assessment, ThresholdNet for adaptive filtering, and EffiWeightNet for effective weight computation. Meanwhile, to maintain modality identity consistency, we propose the Causal Modality Identity Module (CMIM) that establishes causal constraints between generated images and target modality descriptions using vision-language modeling. Extensive experimental results demonstrate that our proposed Med-K2N outperforms state-of-the-art methods by significant margins on multiple benchmarks. Source code is available.

Paperid: 4764, https://arxiv.org/pdf/2510.01448.pdf

Abstract:
Worldwide visual geo-localization seeks to determine the geographic location of an image anywhere on Earth using only its visual content. Learned representations of geography for visual geo-localization remain an active research topic despite much progress. We formulate geo-localization as aligning the visual representation of the query image with a learned geographic representation. Our novel geographic representation explicitly models the world as a hierarchy of geographic embeddings. Additionally, we introduce an approach to efficiently fuse the appearance features of the query image with its semantic segmentation map, forming a robust visual representation. Our main experiments demonstrate improved all-time bests in 22 out of 25 metrics measured across five benchmark datasets compared to prior state-of-the-art (SOTA) methods and recent Large Vision-Language Models (LVLMs). Additional ablation studies support the claim that these gains are primarily driven by the combination of geographic and visual representations.

Paperid: 4765, https://arxiv.org/pdf/2509.24566.pdf

Abstract:
Large vision-language models (LVLMs) have achieved impressive performance across a wide range of vision-language tasks, while they remain vulnerable to backdoor attacks. Existing backdoor attacks on LVLMs aim to force the victim model to generate a predefined target pattern, which is either inserted into or replaces the original content. We find that these fixed-pattern attacks are relatively easy to detect, because the attacked LVLM tends to memorize such frequent patterns in the training dataset, thereby exhibiting overconfidence on these targets given poisoned inputs. To address these limitations, we introduce TokenSwap, a more evasive and stealthy backdoor attack that focuses on the compositional understanding capabilities of LVLMs. Instead of enforcing a fixed targeted content, TokenSwap subtly disrupts the understanding of object relationships in text. Specifically, it causes the backdoored model to generate outputs that mention the correct objects in the image but misrepresent their relationships (i.e., bags-of-words behavior). During training, TokenSwap injects a visual trigger into selected samples and simultaneously swaps the grammatical roles of key tokens in the corresponding textual answers. However, the poisoned samples exhibit only subtle differences from the original ones, making it challenging for the model to learn the backdoor behavior. To address this, TokenSwap employs an adaptive token-weighted loss that explicitly emphasizes the learning of swapped tokens, such that the visual triggers and bags-of-words behavior are associated. Extensive experiments demonstrate that TokenSwap achieves high attack success rates while maintaining superior evasiveness and stealthiness across multiple benchmarks and various LVLM architectures.

Paperid: 4766, https://arxiv.org/pdf/2509.24192.pdf

Abstract:
While vision-language models (VLMs) have made significant progress in multimodal perception (e.g., open-vocabulary object detection) with simple language queries, state-of-the-art VLMs still show limited ability to perceive complex queries involving descriptive attributes and relational clauses. Our in-depth analysis shows that these limitations mainly stem from text encoders in VLMs. Such text encoders behave like bags-of-words and fail to separate target objects from their descriptive attributes and relations in complex queries, resulting in frequent false positives. To address this, we propose restructuring linguistic representations according to the hierarchical relations within sentences for language-based object detection. A key insight is the necessity of disentangling textual tokens into core components-objects, attributes, and relations ("talk in pieces")-and subsequently aggregating them into hierarchically structured sentence-level representations ("see in whole"). Building on this principle, we introduce the TaSe framework with three main contributions: (1) a hierarchical synthetic captioning dataset spanning three tiers from category names to descriptive sentences; (2) Talk in Pieces, the three-component disentanglement module guided by a novel disentanglement loss function, transforms text embeddings into subspace compositions; and (3) See in Whole, which learns to aggregate disentangled components into hierarchically structured embeddings with the guide of proposed hierarchical objectives. The proposed TaSe framework strengthens the inductive bias of hierarchical linguistic structures, resulting in fine-grained multimodal representations for language-based object detection. Experimental results under the OmniLabel benchmark show a 24% performance improvement, demonstrating the importance of linguistic compositionality.

Paperid: 4767, https://arxiv.org/pdf/2509.23729.pdf

Abstract:
Large Language Models (LLMs) with multimodal capabilities have revolutionized vision-language tasks, but their deployment often requires huge memory and computational resources. While post-training quantization (PTQ) has successfully compressed language models to as low as 1-bit precision without significant performance loss, its effectiveness for multimodal LLMs (MLLMs) remains relatively unexplored. In this paper, we present the first study on ultra-low bit (<4-bit) quantization for multimodal LLMs. Our analysis reveals that multimodal tokens and intermediate layer activations produced by them exhibit significantly higher statistical variance and entropy compared to text tokens, making them less tolerant to ultra-low bit quantization. However, the activation distributions of multimodal tokens varies significantly over different layers, with some layers having lower entropy activation distributions. We empirically show that such layers in these models can better tolerate ultra-low bit quantization. Building on these insights, we propose a novel strategy for MLLM quantization, LUQ: Layerwise Ultra-Low Bit Quantization, which selectively applies ultra-low bit quantization to layers that are more resilient to it. Additionally, we also show that using a mix of multimodal tokens (image and text) for PTQ boosts VQA performance in the ultra-low bit regime. We evaluate our method on LLaVA-1.5 and Qwen-2.5-VL across 9 popular VQA benchmarks. The resulting LUQ models use 40% and 31% less memory than their 4-bit counterparts, respectively, while exhibiting a performance degradation of less than 10% on the MME benchmark.

Paperid: 4768, https://arxiv.org/pdf/2509.23250.pdf

Abstract:
Process Reward Models (PRMs) provide step-level supervision that improves the reliability of reasoning in large language models. While PRMs have been extensively studied in text-based domains, their extension to Vision Language Models (VLMs) remains limited. Existing Vision-Language PRMs (VL-PRMs) rely on Monte Carlo Tree Search (MCTS) for data construction, which can often produce noisy supervision signals and limit generalization across tasks. In this work, we aim to elucidate the design space of VL-PRMs by exploring diverse strategies for dataset construction, training, and test-time scaling. First, we introduce a hybrid data synthesis framework that combines MCTS with judgments from a strong VLM, producing more accurate step-level labels. Second, we propose perception-focused supervision, enabling our PRM to explicitly detect errors at the visual grounding stage of reasoning. Third, we systematically evaluate multiple test-time scaling strategies, showing that our PRMs can reliably guide VLMs toward more accurate solutions. Our experiments covering five diverse multimodal benchmarks (MMMU, PuzzleVQA, AlgoPuzzleVQA, MathVista, and MathVision) reveal several key insights: (i) VL-PRMs when used as Outcome Reward Models (ORMs) during test-time scaling (TTS) can outperform VL-PRM guided process step selection, (ii) smaller VL-PRMs can match or even surpass larger ones in detecting process errors, (iii) VL-PRMs uncover latent reasoning abilities in stronger VLM backbones, (iv) perception-level supervision leads to significant gains in test-time scaling, and (v) TTS performance of different policies improve on advanced math reasoning datasets despite not training VL-PRMs on such datasets. We hope our work will motivate further research and support the advancement of VLMs.

Paperid: 4769, https://arxiv.org/pdf/2509.21189.pdf

Abstract:
When navigating in a man-made environment they haven't visited before--like an office building--humans employ behaviors such as reading signs and asking others for directions. These behaviors help humans reach their destinations efficiently by reducing the need to search through large areas. Existing robot navigation systems lack the ability to execute such behaviors and are thus highly inefficient at navigating within large environments. We present ReasonNav, a modular navigation system which integrates these human-like navigation skills by leveraging the reasoning capabilities of a vision-language model (VLM). We design compact input and output abstractions based on navigation landmarks, allowing the VLM to focus on language understanding and reasoning. We evaluate ReasonNav on real and simulated navigation tasks and show that the agent successfully employs higher-order reasoning to navigate efficiently in large, complex buildings.

Paperid: 4770, https://arxiv.org/pdf/2509.19552.pdf

Abstract:
Grounding large language models (LLMs) in domain-specific tasks like post-hoc dash-cam driving video analysis is challenging due to their general-purpose training and lack of structured inductive biases. As vision is often the sole modality available for such analysis (i.e., no LiDAR, GPS, etc.), existing video-based vision-language models (V-VLMs) struggle with spatial reasoning, causal inference, and explainability of events in the input video. To this end, we introduce iFinder, a structured semantic grounding framework that decouples perception from reasoning by translating dash-cam videos into a hierarchical, interpretable data structure for LLMs. iFinder operates as a modular, training-free pipeline that employs pretrained vision models to extract critical cues -- object pose, lane positions, and object trajectories -- which are hierarchically organized into frame- and video-level structures. Combined with a three-block prompting strategy, it enables step-wise, grounded reasoning for the LLM to refine a peer V-VLM's outputs and provide accurate reasoning. Evaluations on four public dash-cam video benchmarks show that iFinder's proposed grounding with domain-specific cues, especially object orientation and global context, significantly outperforms end-to-end V-VLMs on four zero-shot driving benchmarks, with up to 39% gains in accident reasoning accuracy. By grounding LLMs with driving domain-specific representations, iFinder offers a zero-shot, interpretable, and reliable alternative to end-to-end V-VLMs for post-hoc driving video understanding.

Paperid: 4771, https://arxiv.org/pdf/2509.18179.pdf

Abstract:
With the increasing integration of multimodal AI systems in creative workflows, understanding information loss in vision-language-vision pipelines has become important for evaluating system limitations. However, the degradation that occurs when visual content passes through textual intermediation remains poorly quantified. In this work, we provide empirical analysis of the describe-then-generate bottleneck, where natural language serves as an intermediate representation for visual information. We generated 150 image pairs through the describe-then-generate pipeline and applied existing metrics (LPIPS, SSIM, and color distance) to measure information preservation across perceptual, structural, and chromatic dimensions. Our evaluation reveals that 99.3% of samples exhibit substantial perceptual degradation and 91.5% demonstrate significant structural information loss, providing empirical evidence that the describe-then-generate bottleneck represents a measurable and consistent limitation in contemporary multimodal systems.

Paperid: 4772, https://arxiv.org/pdf/2509.17452.pdf

Abstract:
Universal domain adaptation (UniDA) transfers knowledge from a labeled source domain to an unlabeled target domain, where label spaces may differ and the target domain may contain private classes. Previous UniDA methods primarily focused on visual space alignment but often struggled with visual ambiguities due to content differences, which limited their robustness and generalizability. To overcome this, we introduce a novel approach that leverages the strong \textit{zero-shot capabilities} of recent vision-language foundation models (VLMs) like CLIP, concentrating solely on label space alignment to enhance adaptation stability. CLIP can generate task-specific classifiers based only on label names. However, adapting CLIP to UniDA is challenging because the label space is not fully known in advance. In this study, we first utilize generative vision-language models to identify unknown categories in the target domain. Noise and semantic ambiguities in the discovered labels -- such as those similar to source labels (e.g., synonyms, hypernyms, hyponyms) -- complicate label alignment. To address this, we propose a training-free label-space alignment method for UniDA (\ours). Our method aligns label spaces instead of visual spaces by filtering and refining noisy labels between the domains. We then construct a \textit{universal classifier} that integrates both shared knowledge and target-private class information, thereby improving generalizability under domain shifts. The results reveal that the proposed method considerably outperforms existing UniDA techniques across key DomainBed benchmarks, delivering an average improvement of \textcolor{blue}{+7.9\%}in H-score and \textcolor{blue}{+6.1\%} in H$^3$-score. Furthermore, incorporating self-training further enhances performance and achieves an additional (\textcolor{blue}{+1.6\%}) increment in both H- and H$^3$-scores.

Paperid: 4773, https://arxiv.org/pdf/2509.16445.pdf

Abstract:
Enabling robotic assistants to navigate complex environments and locate objects described in free-form language is a critical capability for real-world deployment. While foundation models, particularly Vision-Language Models (VLMs), offer powerful semantic understanding, effectively adapting their web-scale knowledge for embodied decision-making remains a key challenge. We present FiLM-Nav (Fine-tuned Language Model for Navigation), an approach that directly fine-tunes pre-trained VLM as the navigation policy. In contrast to methods that use foundation models primarily in a zero-shot manner or for map annotation, FiLM-Nav learns to select the next best exploration frontier by conditioning directly on raw visual trajectory history and the navigation goal. Leveraging targeted simulated embodied experience allows the VLM to ground its powerful pre-trained representations in the specific dynamics and visual patterns relevant to goal-driven navigation. Critically, fine-tuning on a diverse data mixture combining ObjectNav, OVON, ImageNav, and an auxiliary spatial reasoning task proves essential for achieving robustness and broad generalization. FiLM-Nav sets a new state-of-the-art in both SPL and success rate on HM3D ObjectNav among open-vocabulary methods, and sets a state-of-the-art SPL on the challenging HM3D-OVON benchmark, demonstrating strong generalization to unseen object categories. Our work validates that directly fine-tuning VLMs on diverse simulated embodied data is a highly effective pathway towards generalizable and efficient semantic navigation capabilities.

Paperid: 4774, https://arxiv.org/pdf/2509.16163.pdf

Abstract:
Vision language models (VLMs) excel in multimodal understanding but are prone to adversarial attacks. Existing defenses often demand costly retraining or significant architecture changes. We introduce a lightweight defense using tensor decomposition suitable for any pre-trained VLM, requiring no retraining. By decomposing and reconstructing vision encoder representations, it filters adversarial noise while preserving meaning. Experiments with CLIP on COCO and Flickr30K show improved robustness. On Flickr30K, it restores 12.3\% performance lost to attacks, raising Recall@1 accuracy from 7.5\% to 19.8\%. On COCO, it recovers 8.1\% performance, improving accuracy from 3.8\% to 11.9\%. Analysis shows Tensor Train decomposition with low rank (8-32) and low residual strength ($Î±=0.1-0.2$) is optimal. This method is a practical, plug-and-play solution with minimal overhead for existing VLMs.

Paperid: 4775, https://arxiv.org/pdf/2509.14932.pdf

Abstract:
Vision-Language-Action models (VLAs) mark a major shift in robot learning. They replace specialized architectures and task-tailored components of expert policies with large-scale data collection and setup-specific fine-tuning. In this machine learning-focused workflow that is centered around models and scalable training, traditional robotics software frameworks become a bottleneck, while robot simulations offer only limited support for transitioning from and to real-world experiments. In this work, we close this gap by introducing Robot Control Stack (RCS), a lean ecosystem designed from the ground up to support research in robot learning with large-scale generalist policies. At its core, RCS features a modular and easily extensible layered architecture with a unified interface for simulated and physical robots, facilitating sim-to-real transfer. Despite its minimal footprint and dependencies, it offers a complete feature set, enabling both real-world experiments and large-scale training in simulation. Our contribution is twofold: First, we introduce the architecture of RCS and explain its design principles. Second, we evaluate its usability and performance along the development cycle of VLA and RL policies. Our experiments also provide an extensive evaluation of Octo, OpenVLA, and Pi Zero on multiple robots and shed light on how simulation data can improve real-world policy performance. Our code, datasets, weights, and videos are available at: https://robotcontrolstack.github.io/

Paperid: 4776, https://arxiv.org/pdf/2509.14889.pdf

Abstract:
In this work, we present CollabVLA, a self-reflective vision-language-action framework that transforms a standard visuomotor policy into a collaborative assistant. CollabVLA tackles key limitations of prior VLAs, including domain overfitting, non-interpretable reasoning, and the high latency of auxiliary generative models, by integrating VLM-based reflective reasoning with diffusion-based action generation under a mixture-of-experts design. Through a two-stage training recipe of action grounding and reflection tuning, it supports explicit self-reflection and proactively solicits human guidance when confronted with uncertainty or repeated failure. It cuts normalized Time by ~2x and Dream counts by ~4x vs. generative agents, achieving higher success rates, improved interpretability, and balanced low latency compared with existing methods. This work takes a pioneering step toward shifting VLAs from opaque controllers to genuinely assistive agents capable of reasoning, acting, and collaborating with humans.

Paperid: 4777, https://arxiv.org/pdf/2509.12145.pdf

Abstract:
We introduce Hierarchical Streaming Video Understanding, a task that combines online temporal action localization with free-form description generation. Given the scarcity of datasets with hierarchical and fine-grained temporal annotations, we demonstrate that LLMs can effectively group atomic actions into higher-level events, enriching existing datasets. We then propose OpenHOUSE (Open-ended Hierarchical Online Understanding System for Events), which extends streaming action perception beyond action classification. OpenHOUSE features a specialized streaming module that accurately detects boundaries between closely adjacent actions, nearly doubling the performance of direct extensions of existing methods. We envision the future of streaming action perception in the integration of powerful generative models, with OpenHOUSE representing a key step in that direction.

Paperid: 4778, https://arxiv.org/pdf/2509.10129.pdf

Abstract:
Vision-Language Models (VLMs) have shown strong capabilities in document understanding, particularly in identifying and extracting textual information from complex documents. Despite this, accurately localizing answers within documents remains a major challenge, limiting both interpretability and real-world applicability. To address this, we introduce DocExplainerV0, a plug-and-play bounding-box prediction module that decouples answer generation from spatial localization. This design makes it applicable to existing VLMs, including proprietary systems where fine-tuning is not feasible. Through systematic evaluation, we provide quantitative insights into the gap between textual accuracy and spatial grounding, showing that correct answers often lack reliable localization. Our standardized framework highlights these shortcomings and establishes a benchmark for future research toward more interpretable and robust document information extraction VLMs.

Paperid: 4779, https://arxiv.org/pdf/2509.07613.pdf

Abstract:
Medical vision-language models (Med-VLMs) have shown impressive results in tasks such as report generation and visual question answering, but they still face several limitations. Most notably, they underutilize patient metadata and lack integration of clinical diagnostic knowledge. Moreover, most existing models are typically trained from scratch or fine-tuned on large-scale 2D image-text pairs, requiring extensive computational resources, and their effectiveness on 3D medical imaging is often limited due to the absence of structural information. To address these gaps, we propose a data-efficient fine-tuning pipeline to adapt 3D CT-based Med-VLMs for 3D MRI and demonstrate its application in Alzheimer's disease (AD) diagnosis. Our system introduces two key innovations. First, we convert structured metadata into synthetic reports, enriching textual input for improved image-text alignment. Second, we add an auxiliary token trained to predict the mini-mental state examination (MMSE) score, a widely used clinical measure of cognitive function that correlates with AD severity. This provides additional supervision for fine-tuning. Applying lightweight prompt tuning to both image and text modalities, our approach achieves state-of-the-art performance on two AD datasets using 1,500 training images, outperforming existing methods fine-tuned on 10,000 images. Code will be released upon publication.

Paperid: 4780, https://arxiv.org/pdf/2509.04687.pdf

Abstract:
Semantic segmentation in real-world applications often requires not only accurate masks but also strict adherence to textual labeling guidelines. These guidelines are typically complex and long, and both human and automated labeling often fail to follow them faithfully. Traditional approaches depend on expensive task-specific retraining that must be repeated as the guidelines evolve. Although recent open-vocabulary segmentation methods excel with simple prompts, they often fail when confronted with sets of paragraph-length guidelines that specify intricate segmentation rules. To address this, we introduce a multi-agent, training-free framework that coordinates general-purpose vision-language models within an iterative Worker-Supervisor refinement architecture. The Worker performs the segmentation, the Supervisor critiques it against the retrieved guidelines, and a lightweight reinforcement learning stop policy decides when to terminate the loop, ensuring guideline-consistent masks while balancing resource use. Evaluated on the Waymo and ReasonSeg datasets, our method notably outperforms state-of-the-art baselines, demonstrating strong generalization and instruction adherence.

Paperid: 4781, https://arxiv.org/pdf/2509.02425.pdf

Abstract:
Indoor built environments like homes and offices often present complex and cluttered layouts that pose significant challenges for individuals who are blind or visually impaired, especially when performing tasks that involve locating and gathering multiple objects. While many existing assistive technologies focus on basic navigation or obstacle avoidance, few systems provide scalable and efficient multi-object search capabilities in real-world, partially observable settings. To address this gap, we introduce OpenGuide, an assistive mobile robot system that combines natural language understanding with vision-language foundation models (VLM), frontier-based exploration, and a Partially Observable Markov Decision Process (POMDP) planner. OpenGuide interprets open-vocabulary requests, reasons about object-scene relationships, and adaptively navigates and localizes multiple target items in novel environments. Our approach enables robust recovery from missed detections through value decay and belief-space reasoning, resulting in more effective exploration and object localization. We validate OpenGuide in simulated and real-world experiments, demonstrating substantial improvements in task success rate and search efficiency over prior methods. This work establishes a foundation for scalable, human-centered robotic assistance in assisted living environments.

Paperid: 4782, https://arxiv.org/pdf/2508.21732.pdf

Abstract:
Recent advancements in Large Vision-Language Models (LVLMs) have demonstrated impressive capabilities across various multimodal tasks. They continue, however, to struggle with trivial scenarios such as reading values from Digital Measurement Devices (DMDs), particularly in real-world conditions involving clutter, occlusions, extreme viewpoints, and motion blur; common in head-mounted cameras and Augmented Reality (AR) applications. Motivated by these limitations, this work introduces CAD2DMD-SET, a synthetic data generation tool designed to support visual question answering (VQA) tasks involving DMDs. By leveraging 3D CAD models, advanced rendering, and high-fidelity image composition, our tool produces diverse, VQA-labelled synthetic DMD datasets suitable for fine-tuning LVLMs. Additionally, we present DMDBench, a curated validation set of 1,000 annotated real-world images designed to evaluate model performance under practical constraints. Benchmarking three state-of-the-art LVLMs using Average Normalised Levenshtein Similarity (ANLS) and further fine-tuning LoRA's of these models with CAD2DMD-SET's generated dataset yielded substantial improvements, with InternVL showcasing a score increase of 200% without degrading on other tasks. This demonstrates that the CAD2DMD-SET training dataset substantially improves the robustness and performance of LVLMs when operating under the previously stated challenging conditions. The CAD2DMD-SET tool is expected to be released as open-source once the final version of this manuscript is prepared, allowing the community to add different measurement devices and generate their own datasets.

Paperid: 4783, https://arxiv.org/pdf/2508.17667.pdf

Abstract:
In trustworthy medical diagnosis systems, integrating out-of-distribution (OOD) detection aims to identify unknown diseases in samples, thereby mitigating the risk of misdiagnosis. In this study, we propose a novel OOD detection framework based on vision-language models (VLMs), which integrates hierarchical visual information to cope with challenging unknown diseases that resemble known diseases. Specifically, a cross-scale visual fusion strategy is proposed to couple visual embeddings from multiple scales. This enriches the detailed representation of medical images and thus improves the discrimination of unknown diseases. Moreover, a cross-scale hard pseudo-OOD sample generation strategy is proposed to benefit OOD detection maximally. Experimental evaluations on three public medical datasets support that the proposed framework achieves superior OOD detection performance compared to existing methods. The source code is available at https://openi.pcl.ac.cn/OpenMedIA/HVL.

Paperid: 4784, https://arxiv.org/pdf/2508.15688.pdf

Abstract:
Pre-trained vision-language models (VLMs), such as CLIP, have demonstrated impressive capability in visual tasks, but their fine-tuning often suffers from bias in class-imbalanced scene. Recent works have introduced large language models (LLMs) to enhance VLM fine-tuning with supplementing semantic information. However, they often overlook inherent class imbalance in VLMs' pre-training, which may lead to bias accumulation in downstream tasks. To address this problem, this paper proposes a Multi-dimensional Dynamic Prompt Routing (MDPR) framework. MDPR constructs a comprehensive knowledge base for classes, spanning five visual-semantic dimensions. During fine-tuning, the dynamic routing mechanism aligns global visual classes, retrieves optimal prompts, and balances fine-grained semantics, yielding stable predictions through logits fusion. Extensive experiments on long-tailed benchmarks, including CIFAR-LT, ImageNet-LT, and Places-LT, demonstrate that MDPR achieves comparable results with current SOTA methods. Ablation studies further confirm the effectiveness of our semantic library for tail classes, and show that our dynamic routing incurs minimal computational overhead, making MDPR a flexible and efficient enhancement for VLM fine-tuning under data imbalance.

Paperid: 4785, https://arxiv.org/pdf/2508.13256.pdf

Abstract:
Cardiovascular diseases (CVDs) remain the foremost cause of mortality worldwide, a burden worsened by a severe deficit of healthcare workers. Artificial intelligence (AI) agents have shown potential to alleviate this gap via automated early detection and proactive screening, yet their clinical application remains limited by: 1) prompt-based clinical role assignment that relies on intrinsic model capabilities without domain-specific tool support; or 2) rigid sequential workflows, whereas clinical care often requires adaptive reasoning that orders specific tests and, based on their results, guides personalised next steps; 3) general and static knowledge bases without continuous learning capability; and 4) fixed unimodal or bimodal inputs and lack of on-demand visual outputs when further clarification is needed. In response, a multimodal framework, CardAIc-Agents, was proposed to augment models with external tools and adaptively support diverse cardiac tasks. Specifically, a CardiacRAG agent generated general plans from updatable cardiac knowledge, while the chief agent integrated tools to autonomously execute these plans and deliver decisions. To enable adaptive and case-specific customization, a stepwise update strategy was proposed to dynamically refine plans based on preceding execution results, once the task was assessed as complex. In addition, a multidisciplinary discussion tool was introduced to interpret challenging cases, thereby supporting further adaptation. When clinicians raised concerns, visual review panels were provided to assist final validation. Experiments across three datasets showed the efficiency of CardAIc-Agents compared to mainstream Vision-Language Models (VLMs), state-of-the-art agentic systems, and fine-tuned VLMs.

Paperid: 4786, https://arxiv.org/pdf/2508.11330.pdf

Abstract:
Although today's pretrained discriminative vision-language models (e.g., CLIP) have demonstrated strong perception abilities, such as zero-shot image classification, they also suffer from the bag-of-words problem and spurious bias. To mitigate these problems, some pioneering studies leverage powerful generative models (e.g., pretrained diffusion models) to realize generalizable image classification, dubbed Diffusion Classifier (DC). Specifically, by randomly sampling a Gaussian noise, DC utilizes the differences of denoising effects with different category conditions to classify categories. Unfortunately, an inherent and notorious weakness of existing DCs is noise instability: different random sampled noises lead to significant performance changes. To achieve stable classification performance, existing DCs always ensemble the results of hundreds of sampled noises, which significantly reduces the classification speed. To this end, we firstly explore the role of noise in DC, and conclude that: there are some ``good noises'' that can relieve the instability. Meanwhile, we argue that these good noises should meet two principles: Frequency Matching and Spatial Matching. Regarding both principles, we propose a novel Noise Optimization method to learn matching (i.e., good) noise for DCs: NoOp. For frequency matching, NoOp first optimizes a dataset-specific noise: Given a dataset and a timestep t, optimize one randomly initialized parameterized noise. For Spatial Matching, NoOp trains a Meta-Network that adopts an image as input and outputs image-specific noise offset. The sum of optimized noise and noise offset will be used in DC to replace random noise. Extensive ablations on various datasets demonstrated the effectiveness of NoOp.

Paperid: 4787, https://arxiv.org/pdf/2508.09199.pdf

Abstract:
Visual Instruction Finetuning (VIF) is pivotal for post-training Vision-Language Models (VLMs). Unlike unimodal instruction finetuning in plain-text large language models, which mainly requires instruction datasets to enable model instruction-following ability, VIF also requires multimodal data to enable joint visual and textual understanding; therefore, it typically requires more data. Consequently, VIF imposes stricter data selection challenges: the method must scale efficiently to handle larger data demands while ensuring the quality of both visual and textual content, as well as their alignment. Despite its critical impact on performance, data selection for VIF remains an understudied area. In this paper, we propose $Î$-AttnMask. This data-efficient framework quantifies sample quality through attention-guided masking of the model's hidden states, jointly evaluating image-text pairs without requiring domain labels, auxiliary models, or extra training. By computing loss differences ($Î$) between the original states and states masked using high-attention regions, $Î$-AttnMask intrinsically assesses sample quality. Experiments across multiple VLMs and datasets show that $Î$-AttnMask achieves state-of-the-art performance with just 20% of data, accelerating training by 5x while surpassing full-dataset baselines by +10.1% in overall accuracy. Its model-agnostic and data-agnostic design ensures broad applicability across modalities and architectures.

Paperid: 4788, https://arxiv.org/pdf/2508.07406.pdf

Abstract:
Agricultural robots have emerged as powerful members in agricultural tasks, nevertheless, still heavily rely on manual operation or untransportable railway for movement, resulting in limited mobility and poor adaptability. Vision-and-Language Navigation (VLN) enables robots to navigate to the target destinations following natural language instructions, demonstrating strong performance on several domains. However, none of the existing benchmarks or methods is specifically designed for agricultural scenes. To bridge this gap, we propose Agriculture to Agriculture (A2A) benchmark, containing 1,560 episodes across six diverse agricultural scenes, in which all realistic RGB videos are captured by front-facing camera on a quadruped robot at a height of 0.38 meters, aligning with the practical deployment conditions. Meanwhile, we propose Vision-and-Language Navigation for Agricultural Robots (AgriVLN) baseline based on Vision-Language Model (VLM) prompted with carefully crafted templates, which can understand both given instructions and agricultural environments to generate appropriate low-level actions for robot control. When evaluated on A2A, AgriVLN performs well on short instructions but struggles with long instructions, because it often fails to track which part of the instruction is currently being executed. To address this, we further propose Subtask List (STL) instruction decomposition module and integrate it into AgriVLN, improving Success Rate (SR) from 0.33 to 0.47. We additionally compare AgriVLN with several existing VLN methods, demonstrating the state-of-the-art performance in the agricultural domain.

Paperid: 4789, https://arxiv.org/pdf/2508.06092.pdf

Abstract:
Accurate and efficient Video Quality Assessment (VQA) has long been a key research challenge. Current mainstream VQA methods typically improve performance by pretraining on large-scale classification datasets (e.g., ImageNet, Kinetics-400), followed by fine-tuning on VQA datasets. However, this strategy presents two significant challenges: (1) merely transferring semantic knowledge learned from pretraining is insufficient for VQA, as video quality depends on multiple factors (e.g., semantics, distortion, motion, aesthetics); (2) pretraining on large-scale datasets demands enormous computational resources, often dozens or even hundreds of times greater than training directly on VQA datasets. Recently, Vision-Language Models (VLMs) have shown remarkable generalization capabilities across a wide range of visual tasks, and have begun to demonstrate promising potential in quality assessment. In this work, we propose Q-CLIP, the first fully VLMs-based framework for VQA. Q-CLIP enhances both visual and textual representations through a Shared Cross-Modal Adapter (SCMA), which contains only a minimal number of trainable parameters and is the only component that requires training. This design significantly reduces computational cost. In addition, we introduce a set of five learnable quality-level prompts to guide the VLMs in perceiving subtle quality variations, thereby further enhancing the model's sensitivity to video quality. Furthermore, we investigate the impact of different frame sampling strategies on VQA performance, and find that frame-difference-based sampling leads to better generalization performance across datasets. Extensive experiments demonstrate that Q-CLIP exhibits excellent performance on several VQA datasets.

Paperid: 4790, https://arxiv.org/pdf/2508.05899.pdf

Abstract:
3D scene generation plays a crucial role in gaming, artistic creation, virtual reality and many other domains. However, current 3D scene design still relies heavily on extensive manual effort from creators, and existing automated methods struggle to generate open-domain scenes or support flexible editing. As a result, generating 3D worlds directly from text has garnered increasing attention. In this paper, we introduce HOLODECK 2.0, an advanced vision-language-guided framework for 3D world generation with support for interactive scene editing based on human feedback. HOLODECK 2.0 can generate diverse and stylistically rich 3D scenes (e.g., realistic, cartoon, anime, and cyberpunk styles) that exhibit high semantic fidelity to fine-grained input descriptions, suitable for both indoor and open-domain environments. HOLODECK 2.0 leverages vision-language models (VLMs) to identify and parse the objects required in a scene and generates corresponding high-quality assets via state-of-the-art 3D generative models. It then iteratively applies spatial constraints derived from the VLMs to achieve semantically coherent and physically plausible layouts. Human evaluations and CLIP-based assessments demonstrate that HOLODECK 2.0 effectively generates high-quality scenes closely aligned with detailed textual descriptions, consistently outperforming baselines across indoor and open-domain scenarios. Additionally, we provide editing capabilities that flexibly adapt to human feedback, supporting layout refinement and style-consistent object edits. Finally, we present a practical application of HOLODECK 2.0 in procedural game modeling, generating visually rich and immersive environments, potentially boosting efficiency.

Paperid: 4791, https://arxiv.org/pdf/2508.03526.pdf

Abstract:
A central research topic in robotics is how to use this system to interact with the physical world. Traditional manipulation tasks primarily focus on small objects. However, in factory or home environments, there is often a need for the movement of large objects, such as moving tables. These tasks typically require multi-robot systems to work collaboratively. Previous research lacks a framework that can scale to arbitrary sizes of robots and generalize to various kinds of tasks. In this work, we propose CollaBot, a generalist framework for simultaneous collaborative manipulation. First, we use SEEM for scene segmentation and point cloud extraction of the target object. Then, we propose a collaborative grasping framework, which decomposes the task into local grasp pose generation and global collaboration. Finally, we design a 2-stage planning module that can generate collision-free trajectories to achieve this task. Experiments show a success rate of 52% across different numbers of robots, objects, and tasks, indicating the effectiveness of the proposed framework.

Paperid: 4792, https://arxiv.org/pdf/2508.01932.pdf

Abstract:
Deep neural networks (DNNs) and generative AI (GenAI) are increasingly vulnerable to backdoor attacks, where adversaries embed triggers into inputs to cause models to misclassify or misinterpret target labels. Beyond traditional single-trigger scenarios, attackers may inject multiple triggers across various object classes, forming unseen backdoor-object configurations that evade standard detection pipelines. In this paper, we introduce DBOM (Disentangled Backdoor-Object Modeling), a proactive framework that leverages structured disentanglement to identify and neutralize both seen and unseen backdoor threats at the dataset level. Specifically, DBOM factorizes input image representations by modeling triggers and objects as independent primitives in the embedding space through the use of Vision-Language Models (VLMs). By leveraging the frozen, pre-trained encoders of VLMs, our approach decomposes the latent representations into distinct components through a learnable visual prompt repository and prompt prefix tuning, ensuring that the relationships between triggers and objects are explicitly captured. To separate trigger and object representations in the visual prompt repository, we introduce the trigger-object separation and diversity losses that aids in disentangling trigger and object visual features. Next, by aligning image features with feature decomposition and fusion, as well as learned contextual prompt tokens in a shared multimodal space, DBOM enables zero-shot generalization to novel trigger-object pairings that were unseen during training, thereby offering deeper insights into adversarial attack patterns. Experimental results on CIFAR-10 and GTSRB demonstrate that DBOM robustly detects poisoned images prior to downstream training, significantly enhancing the security of DNN training pipelines.

Paperid: 4793, https://arxiv.org/pdf/2508.00557.pdf

Abstract:
Fine-tuning pre-trained vision-language models has emerged as a powerful approach for enhancing open-vocabulary semantic segmentation (OVSS). However, the substantial computational and resource demands associated with training on large datasets have prompted interest in training-free methods for OVSS. Existing training-free approaches primarily focus on modifying model architectures and generating prototypes to improve segmentation performance. However, they often neglect the challenges posed by class redundancy, where multiple categories are not present in the current test image, and visual-language ambiguity, where semantic similarities among categories create confusion in class activation. These issues can lead to suboptimal class activation maps and affinity-refined activation maps. Motivated by these observations, we propose FreeCP, a novel training-free class purification framework designed to address these challenges. FreeCP focuses on purifying semantic categories and rectifying errors caused by redundancy and ambiguity. The purified class representations are then leveraged to produce final segmentation predictions. We conduct extensive experiments across eight benchmarks to validate FreeCP's effectiveness. Results demonstrate that FreeCP, as a plug-and-play module, significantly boosts segmentation performance when combined with other OVSS methods.

Paperid: 4794, https://arxiv.org/pdf/2507.22791.pdf

Abstract:
Feature matching is a cornerstone task in computer vision, essential for applications such as image retrieval, stereo matching, 3D reconstruction, and SLAM. This survey comprehensively reviews modality-based feature matching, exploring traditional handcrafted methods and emphasizing contemporary deep learning approaches across various modalities, including RGB images, depth images, 3D point clouds, LiDAR scans, medical images, and vision-language interactions. Traditional methods, leveraging detectors like Harris corners and descriptors such as SIFT and ORB, demonstrate robustness under moderate intra-modality variations but struggle with significant modality gaps. Contemporary deep learning-based methods, exemplified by detector-free strategies like CNN-based SuperPoint and transformer-based LoFTR, substantially improve robustness and adaptability across modalities. We highlight modality-aware advancements, such as geometric and depth-specific descriptors for depth images, sparse and dense learning methods for 3D point clouds, attention-enhanced neural networks for LiDAR scans, and specialized solutions like the MIND descriptor for complex medical image matching. Cross-modal applications, particularly in medical image registration and vision-language tasks, underscore the evolution of feature matching to handle increasingly diverse data interactions.

Paperid: 4795, https://arxiv.org/pdf/2507.18661.pdf

Abstract:
Next Location Prediction is a fundamental task in the study of human mobility, with wide-ranging applications in transportation planning, urban governance, and epidemic forecasting. In practice, when humans attempt to predict the next location in a trajectory, they often visualize the trajectory on a map and reason based on road connectivity and movement trends. However, the vast majority of existing next-location prediction models do not reason over maps \textbf{in the way that humans do}. Fortunately, the recent development of Vision-Language Models (VLMs) has demonstrated strong capabilities in visual perception and even visual reasoning. This opens up a new possibility: by rendering both the road network and trajectory onto an image and leveraging the reasoning abilities of VLMs, we can enable models to perform trajectory inference in a human-like manner. To explore this idea, we first propose a method called Vision-Guided Location Search (VGLS), which evaluates whether a general-purpose VLM is capable of trajectory-based reasoning without modifying any of its internal parameters. Based on insights from the VGLS results, we further propose our main approach: VLMLocPredictor, which is composed of two stages: In the first stage, we design two Supervised Fine-Tuning (SFT) tasks that help the VLM understand road network and trajectory structures and acquire basic reasoning ability on such visual inputs. In the second stage, we introduce Reinforcement Learning from Visual Map Feedback, enabling the model to self-improve its next-location prediction ability through interaction with the environment. Experiments conducted on datasets from four different cities show that our method achieves state-of-the-art (SOTA) performance and exhibits superior cross-city generalization compared to other LLM-based approaches.

Paperid: 4796, https://arxiv.org/pdf/2507.10053.pdf

Abstract:
This paper introduces CoSMo, a novel multimodal Transformer for Page Stream Segmentation (PSS) in comic books, a critical task for automated content understanding, as it is a necessary first stage for many downstream tasks like character analysis, story indexing, or metadata enrichment. We formalize PSS for this unique medium and curate a new 20,800-page annotated dataset. CoSMo, developed in vision-only and multimodal variants, consistently outperforms traditional baselines and significantly larger general-purpose vision-language models across F1-Macro, Panoptic Quality, and stream-level metrics. Our findings highlight the dominance of visual features for comic PSS macro-structure, yet demonstrate multimodal benefits in resolving challenging ambiguities. CoSMo establishes a new state-of-the-art, paving the way for scalable comic book analysis.

Paperid: 4797, https://arxiv.org/pdf/2507.07340.pdf

Abstract:
Visual storytelling systems, particularly large vision-language models, struggle to maintain character and object identity across frames, often failing to recognize when entities in different images represent the same individuals or objects, leading to inconsistent references and referential hallucinations. This occurs because models lack explicit training on when to establish entity connections across frames. We propose a contrastive reinforcement learning approach that trains models to discriminate between coherent image sequences and stories from unrelated images. We extend the Story Reasoning dataset with synthetic negative examples to teach appropriate entity connection behavior. We employ Direct Preference Optimization with a dual-component reward function that promotes grounding and re-identification of entities in real stories while penalizing incorrect entity connections in synthetic contexts. Using this contrastive framework, we fine-tune Qwen Storyteller (based on Qwen2.5-VL 7B). Evaluation shows improvements in grounding mAP from 0.27 to 0.31 (+14.8%), F1 from 0.35 to 0.41 (+17.1%). Pronoun grounding accuracy improved across all pronoun types except "its", and cross-frame character and object persistence increased across all frame counts, with entities appearing in 5 or more frames advancing from 29.3% to 33.3% (+13.7%). Well-structured stories, containing the chain-of-thought and grounded story, increased from 79.1% to 97.5% (+23.3%).

Paperid: 4798, https://arxiv.org/pdf/2507.05227.pdf

Abstract:
Autonomous driving systems have made significant advances in Q&A, perception, prediction, and planning based on local visual information, yet they struggle to incorporate broader navigational context that human drivers routinely utilize. We address this critical gap between local sensor data and global navigation information by proposing NavigScene, an auxiliary navigation-guided natural language dataset that simulates a human-like driving environment within autonomous driving systems. Moreover, we develop three complementary paradigms to leverage NavigScene: (1) Navigation-guided Reasoning, which enhances vision-language models by incorporating navigation context into the prompting approach; (2) Navigation-guided Preference Optimization, a reinforcement learning method that extends Direct Preference Optimization to improve vision-language model responses by establishing preferences for navigation-relevant summarized information; and (3) Navigation-guided Vision-Language-Action model, which integrates navigation guidance and vision-language models with conventional driving models through feature fusion. Extensive experiments demonstrate that our approaches significantly improve performance across perception, prediction, planning, and question-answering tasks by enabling reasoning capabilities beyond visual range and improving generalization to diverse driving scenarios. This work represents a significant step toward more comprehensive autonomous driving systems capable of navigating complex, unfamiliar environments with greater reliability and safety.

Paperid: 4799, https://arxiv.org/pdf/2507.04613.pdf

Abstract:
Survival prediction using whole-slide images (WSIs) is crucial in cancer re-search. Despite notable success, existing approaches are limited by their reliance on sparse slide-level labels, which hinders the learning of discriminative repre-sentations from gigapixel WSIs. Recently, vision language (VL) models, which incorporate additional language supervision, have emerged as a promising solu-tion. However, VL-based survival prediction remains largely unexplored due to two key challenges. First, current methods often rely on only one simple lan-guage prompt and basic cosine similarity, which fails to learn fine-grained associ-ations between multi-faceted linguistic information and visual features within WSI, resulting in inadequate vision-language alignment. Second, these methods primarily exploit patch-level information, overlooking the intrinsic hierarchy of WSIs and their interactions, causing ineffective modeling of hierarchical interac-tions. To tackle these problems, we propose a novel Hierarchical vision-Language collaboration (HiLa) framework for improved survival prediction. Specifically, HiLa employs pretrained feature extractors to generate hierarchical visual features from WSIs at both patch and region levels. At each level, a series of language prompts describing various survival-related attributes are constructed and aligned with visual features via Optimal Prompt Learning (OPL). This ap-proach enables the comprehensive learning of discriminative visual features cor-responding to different survival-related attributes from prompts, thereby improv-ing vision-language alignment. Furthermore, we introduce two modules, i.e., Cross-Level Propagation (CLP) and Mutual Contrastive Learning (MCL) to maximize hierarchical cooperation by promoting interactions and consistency be-tween patch and region levels. Experiments on three TCGA datasets demonstrate our SOTA performance.

Paperid: 4800, https://arxiv.org/pdf/2507.02941.pdf

Abstract:
GameTileNet is a dataset designed to provide semantic labels for low-resolution digital game art, advancing procedural content generation (PCG) and related AI research as a vision-language alignment task. Large Language Models (LLMs) and image-generative AI models have enabled indie developers to create visual assets, such as sprites, for game interactions. However, generating visuals that align with game narratives remains challenging due to inconsistent AI outputs, requiring manual adjustments by human artists. The diversity of visual representations in automatically generated game content is also limited because of the imbalance in distributions across styles for training data. GameTileNet addresses this by collecting artist-created game tiles from OpenGameArt.org under Creative Commons licenses and providing semantic annotations to support narrative-driven content generation. The dataset introduces a pipeline for object detection in low-resolution tile-based game art (e.g., 32x32 pixels) and annotates semantics, connectivity, and object classifications. GameTileNet is a valuable resource for improving PCG methods, supporting narrative-rich game content, and establishing a baseline for object detection in low-resolution, non-photorealistic images. TL;DR: GameTileNet is a semantic dataset of low-resolution game tiles designed to support narrative-driven procedural content generation through visual-language alignment.

Paperid: 4801, https://arxiv.org/pdf/2506.22817.pdf

Abstract:
Recent open-vocabulary 3D scene understanding approaches mainly focus on training 3D networks through contrastive learning with point-text pairs or by distilling 2D features into 3D models via point-pixel alignment. While these methods show considerable performance in benchmarks with limited vocabularies, they struggle to handle diverse object categories as the limited amount of 3D data upbound training strong open-vocabulary 3d models. We observe that 2D multi-view fusion methods take precedence in understanding diverse concepts in 3D scenes. However, inherent noises in vision-language models lead multi-view fusion to sub-optimal performance. To this end, we introduce MVOV3D, a novel approach aimed at unleashing the potential of 2D multi-view fusion for open-vocabulary 3D scene understanding. We focus on reducing the inherent noises without training, thereby preserving the generalizability while enhancing open-world capabilities. Specifically, MVOV3D improves multi-view 2D features by leveraging precise region-level image features and text features encoded by CLIP encoders and incorporates 3D geometric priors to optimize multi-view fusion. Extensive experiments on various datasets demonstrate the effectiveness of our method. Notably, our MVOV3D achieves a new record with 14.7% mIoU on ScanNet200 and 16.2% mIoU on Matterport160 for challenge open-vocabulary semantic segmentation, outperforming current leading trained 3D networks by a significant margin.

Paperid: 4802, https://arxiv.org/pdf/2506.17561.pdf

Abstract:
Recent studies on Vision-Language-Action (VLA) models have shifted from the end-to-end action-generation paradigm toward a pipeline involving task planning followed by action generation, demonstrating improved performance on various complex, long-horizon manipulation tasks. However, existing approaches vary significantly in terms of network architectures, planning paradigms, representations, and training data sources, making it challenging for researchers to identify the precise sources of performance gains and components to be further improved. To systematically investigate the impacts of different planning paradigms and representations isolating from network architectures and training data, in this paper, we introduce VLA-OS, a unified VLA architecture series capable of various task planning paradigms, and design a comprehensive suite of controlled experiments across diverse object categories (rigid and deformable), visual modalities (2D and 3D), environments (simulation and real-world), and end-effectors (grippers and dexterous hands). Our results demonstrate that: 1) visually grounded planning representations are generally better than language planning representations; 2) the Hierarchical-VLA paradigm generally achieves superior or comparable performance than other paradigms on task performance, pretraining, generalization ability, scalability, and continual learning ability, albeit at the cost of slower training and inference speeds.

Paperid: 4803, https://arxiv.org/pdf/2506.16760.pdf

Abstract:
Large Vision-Language Models (LVLMs) demonstrate exceptional performance across multimodal tasks, yet remain vulnerable to jailbreak attacks that bypass built-in safety mechanisms to elicit restricted content generation. Existing black-box jailbreak methods primarily rely on adversarial textual prompts or image perturbations, yet these approaches are highly detectable by standard content filtering systems and exhibit low query and computational efficiency. In this work, we present Cross-modal Adversarial Multimodal Obfuscation (CAMO), a novel black-box jailbreak attack framework that decomposes malicious prompts into semantically benign visual and textual fragments. By leveraging LVLMs' cross-modal reasoning abilities, CAMO covertly reconstructs harmful instructions through multi-step reasoning, evading conventional detection mechanisms. Our approach supports adjustable reasoning complexity and requires significantly fewer queries than prior attacks, enabling both stealth and efficiency. Comprehensive evaluations conducted on leading LVLMs validate CAMO's effectiveness, showcasing robust performance and strong cross-model transferability. These results underscore significant vulnerabilities in current built-in safety mechanisms, emphasizing an urgent need for advanced, alignment-aware security and safety solutions in vision-language systems.

Paperid: 4804, https://arxiv.org/pdf/2506.14797.pdf

Abstract:
Intelligent systems must deploy internal representations that are simultaneously structured -- to support broad generalization -- and selective -- to preserve input identity. We expose a fundamental limit on this tradeoff. For any model whose representational similarity between inputs decays with finite semantic resolution $\varepsilon$, we derive closed-form expressions that pin its probability of correct generalization $p_S$ and identification $p_I$ to a universal Pareto front independent of input space geometry. Extending the analysis to noisy, heterogeneous spaces and to $n>2$ inputs predicts a sharp $1/n$ collapse of multi-input processing capacity and a non-monotonic optimum for $p_S$. A minimal ReLU network trained end-to-end reproduces these laws: during learning a resolution boundary self-organizes and empirical $(p_S,p_I)$ trajectories closely follow theoretical curves for linearly decaying similarity. Finally, we demonstrate that the same limits persist in two markedly more complex settings -- a convolutional neural network and state-of-the-art vision-language models -- confirming that finite-resolution similarity is a fundamental emergent informational constraint, not merely a toy-model artifact. Together, these results provide an exact theory of the generalization-identification trade-off and clarify how semantic resolution shapes the representational capacity of deep networks and brains alike.

Paperid: 4805, https://arxiv.org/pdf/2506.14233.pdf

Abstract:
Large Vision-Language Models (VLMs) have demonstrated potential in enhancing mobile robot navigation in human-centric environments by understanding contextual cues, human intentions, and social dynamics while exhibiting reasoning capabilities. However, their computational complexity and limited sensitivity to continuous numerical data impede real-time performance and precise motion control. To this end, we propose Narrate2Nav, a novel real-time vision-action model that leverages a novel self-supervised learning framework based on the Barlow Twins redundancy reduction loss to embed implicit natural language reasoning, social cues, and human intentions within a visual encoder-enabling reasoning in the model's latent space rather than token space. The model combines RGB inputs, motion commands, and textual signals of scene context during training to bridge from robot observations to low-level motion commands for short-horizon point-goal navigation during deployment. Extensive evaluation of Narrate2Nav across various challenging scenarios in both offline unseen dataset and real-world experiments demonstrates an overall improvement of 52.94 percent and 41.67 percent, respectively, over the next best baseline. Additionally, qualitative comparative analysis of Narrate2Nav's visual encoder attention map against four other baselines demonstrates enhanced attention to navigation-critical scene elements, underscoring its effectiveness in human-centric navigation tasks.

Paperid: 4806, https://arxiv.org/pdf/2506.13428.pdf

Abstract:
Dual-arm cooperative manipulation holds great promise for tackling complex real-world tasks that demand seamless coordination and adaptive dynamics. Despite substantial progress in learning-based motion planning, most approaches struggle to generalize across diverse manipulation tasks and adapt to dynamic, unstructured environments, particularly in scenarios involving interactions between two objects such as assembly, tool use, and bimanual grasping. To address these challenges, we introduce a novel VLM-Assisted Siamese Flow Diffusion (VLM-SFD) framework for efficient imitation learning in dual-arm cooperative manipulation. The proposed VLM-SFD framework exhibits outstanding adaptability, significantly enhancing the ability to rapidly adapt and generalize to diverse real-world tasks from only a minimal number of human demonstrations. Specifically, we propose a Siamese Flow Diffusion Network (SFDNet) employs a dual-encoder-decoder Siamese architecture to embed two target objects into a shared latent space, while a diffusion-based conditioning process-conditioned by task instructions-generates two-stream object-centric motion flows that guide dual-arm coordination. We further design a dynamic task assignment strategy that seamlessly maps the predicted 2D motion flows into 3D space and incorporates a pre-trained vision-language model (VLM) to adaptively assign the optimal motion to each robotic arm over time. Experiments validate the effectiveness of the proposed method, demonstrating its ability to generalize to diverse manipulation tasks while maintaining high efficiency and adaptability. The code and demo videos are publicly available on our project website https://sites.google.com/view/vlm-sfd/.

Paperid: 4807, https://arxiv.org/pdf/2506.08678.pdf

Abstract:
Vision-language models such as CLIP have recently propelled open-vocabulary dense prediction tasks by enabling recognition of a broad range of visual concepts. However, CLIP still struggles with fine-grained, region-level understanding, hindering its effectiveness on these dense prediction tasks. We identify two pivotal factors required to address this limitation: semantic coherence and fine-grained vision-language alignment. Current adaptation methods often improve fine-grained alignment at the expense of semantic coherence, and often rely on extra modules or supervised fine-tuning. To overcome these issues, we propose Any-to-Any Self-Distillation (ATAS), a novel approach that simultaneously enhances semantic coherence and fine-grained alignment by leveraging own knowledge of a model across all representation levels. Unlike prior methods, ATAS uses only unlabeled images and an internal self-distillation process to refine representations of CLIP vision encoders, preserving local semantic consistency while sharpening local detail recognition. On open-vocabulary object detection and semantic segmentation benchmarks, ATAS achieves substantial performance gains, outperforming baseline CLIP models. These results validate the effectiveness of our approach and underscore the importance of jointly maintaining semantic coherence and fine-grained alignment for advanced open-vocabulary dense prediction.

Paperid: 4808, https://arxiv.org/pdf/2506.08462.pdf

Abstract:
Industrial processes must be robust and adaptable, as environments and tasks are often unpredictable, while operational errors remain costly and difficult to detect. AI-based control systems offer a path forward, yet typically depend on supervised learning with extensive labelled datasets, which limits their ability to generalize across variable and data-scarce industrial settings. Foundation models could enable broader reasoning and knowledge integration, but rarely deliver the quantitative precision demanded by engineering applications. Here, we introduceControl and Interpretation of Production via Hybrid Expertise and Reasoning (CIPHER): a vision-language-action (VLA) model framework aiming to replicate human-like reasoning for industrial control, instantiated in a commercial-grade 3D printer. It integrates a process expert, a regression model enabling quantitative characterization of system states required for engineering tasks. CIPHER also incorporates retrieval-augmented generation to access external expert knowledge and support physics-informed, chain-of-thought reasoning. This hybrid architecture exhibits strong generalization to out-of-distribution tasks. It interprets visual or textual inputs from process monitoring, explains its decisions, and autonomously generates precise machine instructions, without requiring explicit annotations. CIPHER thus lays the foundations for autonomous systems that act with precision, reason with context, and communicate decisions transparently, supporting safe and trusted deployment in industrial settings.

Paperid: 4809, https://arxiv.org/pdf/2506.05405.pdf

Abstract:
In robot scientific laboratories, visual anomaly detection is important for the timely identification and resolution of potential faults or deviations. It has become a key factor in ensuring the stability and safety of experimental processes. To address this challenge, this paper proposes a VLM-based visual reasoning approach that supports different levels of supervision through four progressively informative prompt configurations. To systematically evaluate its effectiveness, we construct a visual benchmark tailored for process anomaly detection in scientific workflows. Experiments on two representative vision-language models show that detection accuracy improves as more contextual information is provided, confirming the effectiveness and adaptability of the proposed reasoning approach for process anomaly detection in scientific workflows. Furthermore, real-world validations at selected experimental steps confirm that first-person visual observation can effectively identify process-level anomalies. This work provides both a data-driven foundation and an evaluation framework for vision anomaly detection in scientific experiment workflows.

Paperid: 4810, https://arxiv.org/pdf/2506.05210.pdf

Abstract:
Multimodal foundation models have demonstrated strong generalization, yet their ability to transfer knowledge to specialized domains such as garment generation remains underexplored. We introduce VLG, a vision-language-garment model that synthesizes garments from textual descriptions and visual imagery. Our experiments assess VLG's zero-shot generalization, investigating its ability to transfer web-scale reasoning to unseen garment styles and prompts. Preliminary results indicate promising transfer capabilities, highlighting the potential for multimodal foundation models to adapt effectively to specialized domains like fashion design.

Paperid: 4811, https://arxiv.org/pdf/2506.05020.pdf

Abstract:
Heterogeneous multi-robot systems show great potential in complex tasks requiring hybrid cooperation. However, traditional approaches relying on static models often struggle with task diversity and dynamic environments. This highlights the need for generalizable intelligence that can bridge high-level reasoning with low-level execution across heterogeneous agents. To address this, we propose a hierarchical framework integrating a prompted Large Language Model (LLM) and a GridMask-enhanced fine-tuned Vision Language Model (VLM). The LLM decomposes tasks and constructs a global semantic map, while the VLM extracts task-specified semantic labels and 2D spatial information from aerial images to support local planning. Within this framework, the aerial robot follows an optimized global semantic path and continuously provides bird-view images, guiding the ground robot's local semantic navigation and manipulation, including target-absent scenarios where implicit alignment is maintained. Experiments on real-world cube or object arrangement tasks demonstrate the framework's adaptability and robustness in dynamic environments. To the best of our knowledge, this is the first demonstration of an aerial-ground heterogeneous system integrating VLM-based perception with LLM-driven task reasoning and motion planning.

Paperid: 4812, https://arxiv.org/pdf/2505.23678.pdf

Abstract:
While reinforcement learning (RL) over chains of thought has significantly advanced language models in tasks such as mathematics and coding, visual reasoning introduces added complexity by requiring models to direct visual attention, interpret perceptual inputs, and ground abstract reasoning in spatial evidence. We introduce ViGoRL (Visually Grounded Reinforcement Learning), a vision-language model trained with RL to explicitly anchor each reasoning step to specific visual coordinates. Inspired by human visual decision-making, ViGoRL learns to produce spatially grounded reasoning traces, guiding visual attention to task-relevant regions at each step. When fine-grained exploration is required, our novel multi-turn RL framework enables the model to dynamically zoom into predicted coordinates as reasoning unfolds. Across a diverse set of visual reasoning benchmarks--including SAT-2 and BLINK for spatial reasoning, V*bench for visual search, and ScreenSpot and VisualWebArena for web-based grounding--ViGoRL consistently outperforms both supervised fine-tuning and conventional RL baselines that lack explicit grounding mechanisms. Incorporating multi-turn RL with zoomed-in visual feedback significantly improves ViGoRL's performance on localizing small GUI elements and visual search, achieving 86.4% on V*Bench. Additionally, we find that grounding amplifies other visual behaviors such as region exploration, grounded subgoal setting, and visual verification. Finally, human evaluations show that the model's visual references are not only spatially accurate but also helpful for understanding model reasoning steps. Our results show that visually grounded RL is a strong paradigm for imbuing models with general-purpose visual reasoning.

Paperid: 4813, https://arxiv.org/pdf/2505.23130.pdf

Abstract:
Photo retouching is integral to photographic art, extending far beyond simple technical fixes to heighten emotional expression and narrative depth. While artists leverage expertise to create unique visual effects through deliberate adjustments, non-professional users often rely on automated tools that produce visually pleasing results but lack interpretative depth and interactive transparency. In this paper, we introduce PhotoArtAgent, an intelligent system that combines Vision-Language Models (VLMs) with advanced natural language reasoning to emulate the creative process of a professional artist. The agent performs explicit artistic analysis, plans retouching strategies, and outputs precise parameters to Lightroom through an API. It then evaluates the resulting images and iteratively refines them until the desired artistic vision is achieved. Throughout this process, PhotoArtAgent provides transparent, text-based explanations of its creative rationale, fostering meaningful interaction and user control. Experimental results show that PhotoArtAgent not only surpasses existing automated tools in user studies but also achieves results comparable to those of professional human artists.

Paperid: 4814, https://arxiv.org/pdf/2505.21955.pdf

Abstract:
Large vision-language models (LVLMs) are increasingly deployed in interactive applications such as virtual and augmented reality, where first-person (egocentric) view captured by head-mounted cameras serves as key input. While this view offers fine-grained cues about user attention and hand-object interactions, their narrow field of view and lack of global context often lead to failures on spatially or contextually demanding queries. To address this, we introduce a framework that augments egocentric inputs with third-person (exocentric) views, providing complementary information such as global scene layout and object visibility to LVLMs. We present E3VQA, the first benchmark for multi-view question answering with 4K high-quality question-answer pairs grounded in synchronized ego-exo image pairs. Additionally, we propose M3CoT, a training-free prompting technique that constructs a unified scene representation by integrating scene graphs from three complementary perspectives. M3CoT enables LVLMs to reason more effectively across views, yielding consistent performance gains (4.84% for GPT-4o and 5.94% for Gemini 2.0 Flash) over a recent CoT baseline. Our extensive evaluation reveals key strengths and limitations of LVLMs in multi-view reasoning and highlights the value of leveraging both egocentric and exocentric inputs.

Paperid: 4815, https://arxiv.org/pdf/2505.21472.pdf

Abstract:
Large vision-language models (LVLMs) achieve impressive performance on multimodal tasks but often suffer from hallucination, and confidently describe objects or attributes not present in the image. Current training-free interventions struggle to maintain accuracy in open-ended and long-form generation scenarios. We introduce the Confidence-Aware Attention Calibration (CAAC) framework to address this challenge by targeting two key biases: spatial perception bias, which distributes attention disproportionately across image tokens, and modality bias, which shifts focus from visual to textual inputs over time. CAAC employs a two-step approach: Visual-Token Calibration (VTC) to balance attention across visual tokens, and Adaptive Attention Re-Scaling (AAR) to reinforce visual grounding guided by the model's confidence. This confidence-driven adjustment ensures consistent visual alignment during generation. Experiments on CHAIR, AMBER, and POPE benchmarks demonstrate that CAAC outperforms baselines, particularly in long-form generations, effectively reducing hallucination.

Paperid: 4816, https://arxiv.org/pdf/2505.19696.pdf

Abstract:
This position paper argues that Mean Opinion Score (MOS), while historically foundational, is no longer sufficient as the sole supervisory signal for multimedia quality assessment models. MOS reduces rich, context-sensitive human judgments to a single scalar, obscuring semantic failures, user intent, and the rationale behind quality decisions. We contend that modern quality assessment models must integrate three interdependent capabilities: (1) context-awareness, to adapt evaluations to task-specific goals and viewing conditions; (2) reasoning, to produce interpretable, evidence-grounded justifications for quality judgments; and (3) multimodality, to align perceptual and semantic cues using vision-language models. We critique the limitations of current MOS-centric benchmarks and propose a roadmap for reform: richer datasets with contextual metadata and expert rationales, and new evaluation metrics that assess semantic alignment, reasoning fidelity, and contextual sensitivity. By reframing quality assessment as a contextual, explainable, and multimodal modeling task, we aim to catalyze a shift toward more robust, human-aligned, and trustworthy evaluation systems.

Paperid: 4817, https://arxiv.org/pdf/2505.17670.pdf

Abstract:
Language models (LMs) and their extension, vision-language models (VLMs), have achieved remarkable performance across various tasks. However, they still struggle with complex reasoning tasks that require multimodal or multilingual real-world knowledge. To support such capabilities, an external memory system that can efficiently provide relevant multimodal information is essential. Existing approaches generally concatenate image and text tokens into a long sequence as memory, which, however, may drastically increase context length and even degrade performance. In contrast, we propose using continuous memory, a compact set of dense embeddings to more effectively and efficiently represent multimodal and multilingual knowledge. Our key insight is that a VLM can serve as its own continuous memory encoder. We empirically show that this design improves performance on complex multimodal reasoning tasks. Building on this, we introduce a data-efficient and parameter-efficient method to fine-tune the VLM into a memory encoder, requiring only 1.2% of the model's parameters and a small corpus of 15.6K self-synthesized samples. Our approach CoMEM utilizes VLM's original capabilities to encode arbitrary multimodal and multilingual knowledge into just 8 continuous embeddings. Since the inference-time VLM remains frozen, our memory module is plug-and-play and can be flexibly integrated as needed. Extensive experiments across eight multimodal reasoning benchmarks demonstrate the effectiveness of our approach.

Paperid: 4818, https://arxiv.org/pdf/2505.15517.pdf

Abstract:
Vision-Language Models (VLMs) acquire real-world knowledge and general reasoning ability through Internet-scale image-text corpora. They can augment robotic systems with scene understanding and task planning, and assist visuomotor policies that are trained on robot trajectory data. We explore the reverse paradigm - using rich, real, multi-modal robot trajectory data to enhance and evaluate VLMs. In this paper, we present Robo2VLM, a Visual Question Answering (VQA) dataset generation framework for VLMs. Given a human tele-operated robot trajectory, Robo2VLM derives ground-truth from non-visual and non-descriptive sensory modalities, such as end-effector pose, gripper aperture, and force sensing. Based on these modalities, it segments the robot trajectory into a sequence of manipulation phases. At each phase, Robo2VLM uses scene and interaction understanding to identify 3D properties of the robot, task goal, and the target object. The properties are used to generate representative VQA queries - images with textural multiple-choice questions - based on spatial, goal-conditioned, and interaction reasoning question templates. We curate Robo2VLM-1, a large-scale in-the-wild dataset with 684,710 questions covering 463 distinct scenes and 3,396 robotic manipulation tasks from 176k real robot trajectories. Results suggest that Robo2VLM-1 can benchmark and improve VLM capabilities in spatial and interaction reasoning.

Paperid: 4819, https://arxiv.org/pdf/2505.14948.pdf

Abstract:
The task of estimating the world model describing the dynamics of a real world process assumes immense importance for anticipating and preparing for future outcomes. For applications such as video surveillance, robotics applications, autonomous driving, etc. this objective entails synthesizing plausible visual futures, given a few frames of a video to set the visual context. Towards this end, we propose ProgGen, which undertakes the task of video frame prediction by representing the dynamics of the video using a set of neuro-symbolic, human-interpretable set of states (one per frame) by leveraging the inductive biases of Large (Vision) Language Models (LLM/VLM). In particular, ProgGen utilizes LLM/VLM to synthesize programs: (i) to estimate the states of the video, given the visual context (i.e. the frames); (ii) to predict the states corresponding to future time steps by estimating the transition dynamics; (iii) to render the predicted states as visual RGB-frames. Empirical evaluations reveal that our proposed method outperforms competing techniques at the task of video frame prediction in two challenging environments: (i) PhyWorld (ii) Cart Pole. Additionally, ProgGen permits counter-factual reasoning and interpretable video generation attesting to its effectiveness and generalizability for video generation tasks.

Paperid: 4820, https://arxiv.org/pdf/2505.14729.pdf

Abstract:
Vision-Language Models (VLMs) have demonstrated impressive capabilities across a range of tasks, yet concerns about their potential biases exist. This work investigates the extent to which prominent VLMs exhibit cultural biases by evaluating their performance on an image-based country identification task at a country level. Utilizing the geographically diverse Country211 dataset, we probe several large vision language models (VLMs) under various prompting strategies: open-ended questions, multiple-choice questions (MCQs) including challenging setups like multilingual and adversarial settings. Our analysis aims to uncover disparities in model accuracy across different countries and question formats, providing insights into how training data distribution and evaluation methodologies might influence cultural biases in VLMs. The findings highlight significant variations in performance, suggesting that while VLMs possess considerable visual understanding, they inherit biases from their pre-training data and scale that impact their ability to generalize uniformly across diverse global contexts.

Paperid: 4821, https://arxiv.org/pdf/2505.14141.pdf

Abstract:
Mobile GUI agents execute user commands by directly interacting with the graphical user interface (GUI) of mobile devices, demonstrating significant potential to enhance user convenience. However, these agents face considerable challenges in task planning, as they must continuously analyze the GUI and generate operation instructions step by step. This process often leads to difficulties in making accurate task plans, as GUI agents lack a deep understanding of how to effectively use the target applications, which can cause them to become "lost" during task execution. To address the task planning issue, we propose SPlanner, a plug-and-play planning module to generate execution plans that guide vision language model(VLMs) in executing tasks. The proposed planning module utilizes extended finite state machines (EFSMs) to model the control logits and configurations of mobile applications. It then decomposes a user instruction into a sequence of primary function modeled in EFSMs, and generate the execution path by traversing the EFSMs. We further refine the execution path into a natural language plan using an LLM. The final plan is concise and actionable, and effectively guides VLMs to generate interactive GUI actions to accomplish user tasks. SPlanner demonstrates strong performance on dynamic benchmarks reflecting real-world mobile usage. On the AndroidWorld benchmark, SPlanner achieves a 63.8% task success rate when paired with Qwen2.5-VL-72B as the VLM executor, yielding a 28.8 percentage point improvement compared to using Qwen2.5-VL-72B without planning assistance.

Paperid: 4822, https://arxiv.org/pdf/2505.13444.pdf

Abstract:
Chart understanding presents a unique challenge for large vision-language models (LVLMs), as it requires the integration of sophisticated textual and visual reasoning capabilities. However, current LVLMs exhibit a notable imbalance between these skills, falling short on visual reasoning that is difficult to perform in text. We conduct a case study using a synthetic dataset solvable only through visual reasoning and show that model performance degrades significantly with increasing visual complexity, while human performance remains robust. We then introduce ChartMuseum, a new Chart Question Answering (QA) benchmark containing 1,162 expert-annotated questions spanning multiple reasoning types, curated from real-world charts across 184 sources, specifically built to evaluate complex visual and textual reasoning. Unlike prior chart understanding benchmarks -- where frontier models perform similarly and near saturation -- our benchmark exposes a substantial gap between model and human performance, while effectively differentiating model capabilities: although humans achieve 93% accuracy, the best-performing model Gemini-2.5-Pro attains only 63.0%, and the leading open-source LVLM Qwen2.5-VL-72B-Instruct achieves only 38.5%. Moreover, on questions requiring primarily visual reasoning, all models experience a 35%-55% performance drop from text-reasoning-heavy question performance. Lastly, our qualitative error analysis reveals specific categories of visual reasoning that are challenging for current LVLMs.

Paperid: 4823, https://arxiv.org/pdf/2505.12072.pdf

Abstract:
Robots should learn new tasks from humans. But how do humans convey what they want the robot to do? Existing methods largely rely on humans physically guiding the robot arm throughout their intended task. Unfortunately -- as we scale up the amount of data -- physical guidance becomes prohibitively burdensome. Not only do humans need to operate robot hardware but also modify the environment (e.g., moving and resetting objects) to provide multiple task examples. In this work we propose L2D2, a sketching interface and imitation learning algorithm where humans can provide demonstrations by drawing the task. L2D2 starts with a single image of the robot arm and its workspace. Using a tablet, users draw and label trajectories on this image to illustrate how the robot should act. To collect new and diverse demonstrations, we no longer need the human to physically reset the workspace; instead, L2D2 leverages vision-language segmentation to autonomously vary object locations and generate synthetic images for the human to draw upon. We recognize that drawing trajectories is not as information-rich as physically demonstrating the task. Drawings are 2-dimensional and do not capture how the robot's actions affect its environment. To address these fundamental challenges the next stage of L2D2 grounds the human's static, 2D drawings in our dynamic, 3D world by leveraging a small set of physical demonstrations. Our experiments and user study suggest that L2D2 enables humans to provide more demonstrations with less time and effort than traditional approaches, and users prefer drawings over physical manipulation. When compared to other drawing-based approaches, we find that L2D2 learns more performant robot policies, requires a smaller dataset, and can generalize to longer-horizon tasks. See our project website: https://collab.me.vt.edu/L2D2/

Paperid: 4824, https://arxiv.org/pdf/2505.09430.pdf

Abstract:
We present a method that reduces, by an order of magnitude, the time and memory needed to train multi-task vision-language robotic diffusion policies. This improvement arises from a previously underexplored distinction between action diffusion and the image diffusion techniques that inspired it: In image generation, the target is high-dimensional. By contrast, in action generation, the dimensionality of the target is comparatively small, and only the image condition is high-dimensional. Our approach, \emph{Mini Diffuser}, exploits this asymmetry by introducing \emph{two-level minibatching}, which pairs multiple noised action samples with each vision-language condition, instead of the conventional one-to-one sampling strategy. To support this batching scheme, we introduce architectural adaptations to the diffusion transformer that prevent information leakage across samples while maintaining full conditioning access. In RLBench simulations, Mini-Diffuser achieves 95\% of the performance of state-of-the-art multi-task diffusion policies, while using only 5\% of the training time and 7\% of the memory. Real-world experiments further validate that Mini-Diffuser preserves the key strengths of diffusion-based policies, including the ability to model multimodal action distributions and produce behavior conditioned on diverse perceptual inputs. Code available at mini-diffuse-actor.github.io

Paperid: 4825, https://arxiv.org/pdf/2505.07172.pdf

Abstract:
Despite significant advancements in multimodal reasoning tasks, existing Large Vision-Language Models (LVLMs) are prone to producing visually ungrounded responses when interpreting associated images. In contrast, when humans embark on learning new knowledge, they often rely on a set of fundamental pre-study principles: reviewing outlines to grasp core concepts, summarizing key points to guide their focus and enhance understanding. However, such preparatory actions are notably absent in the current instruction tuning processes. This paper presents Re-Critic, an easily scalable rationale-augmented framework designed to incorporate fundamental rules and chain-of-thought (CoT) as a bridge to enhance reasoning abilities. Specifically, Re-Critic develops a visual rationale synthesizer that scalably augments raw instructions with rationale explanation. To probe more contextually grounded responses, Re-Critic employs an in-context self-critic mechanism to select response pairs for preference tuning. Experiments demonstrate that models fine-tuned with our rationale-augmented dataset yield gains that extend beyond hallucination-specific tasks to broader multimodal reasoning tasks.

Paperid: 4826, https://arxiv.org/pdf/2505.05666.pdf

Abstract:
Retrieval-Augmented Generation (RAG) has become a popular technique for enhancing the reliability and utility of Large Language Models (LLMs) by grounding responses in external documents. Traditional RAG systems rely on Optical Character Recognition (OCR) to first process scanned documents into text. However, even state-of-the-art OCRs can introduce errors, especially in degraded or complex documents. Recent vision-language approaches, such as ColPali, propose direct visual embedding of documents, eliminating the need for OCR. This study presents a systematic comparison between a vision-based RAG system (ColPali) and more traditional OCR-based pipelines utilizing Llama 3.2 (90B) and Nougat OCR across varying document qualities. Beyond conventional retrieval accuracy metrics, we introduce a semantic answer evaluation benchmark to assess end-to-end question-answering performance. Our findings indicate that while vision-based RAG performs well on documents it has been fine-tuned on, OCR-based RAG is better able to generalize to unseen documents of varying quality. We highlight the key trade-offs between computational efficiency and semantic accuracy, offering practical guidance for RAG practitioners in selecting between OCR-dependent and vision-based document retrieval systems in production environments.

Paperid: 4827, https://arxiv.org/pdf/2505.03414.pdf

Abstract:
Recent developments in prompt learning of large Vision-Language Models (VLMs) have significantly improved performance in target-specific tasks. However, these prompting methods often struggle to tackle the target-unspecific or generalizable tasks effectively. It may be attributed to the fact that overfitting training causes the model to forget its general knowledge. The general knowledge has a strong promotion on target-unspecific tasks. To alleviate this issue, we propose a novel Features Matrix (FM) approach designed to enhance these models on target-unspecific tasks. Our method extracts and leverages general knowledge, shaping a Features Matrix (FM). Specifically, the FM captures the semantics of diverse inputs from a deep and fine perspective, preserving essential general knowledge, which mitigates the risk of overfitting. Representative evaluations demonstrate that: 1) the FM is compatible with existing frameworks as a generic and flexible module, and 2) the FM significantly showcases its effectiveness in enhancing target-unspecific tasks (base-to-novel generalization, domain generalization, and cross-dataset generalization), achieving state-of-the-art performance.

Paperid: 4828, https://arxiv.org/pdf/2505.01399.pdf

Abstract:
Tool use enhances a robot's task capabilities. Recent advances in vision-language models (VLMs) have equipped robots with sophisticated cognitive capabilities for tool-use applications. However, existing methodologies focus on elementary quasi-static tool manipulations or high-level tool selection while neglecting the critical aspect of task-appropriate tool grasping. To address this limitation, we introduce inverse Tool-Use Planning (iTUP), a novel VLM-driven framework that enables grounded fine-grained planning for versatile robotic tool use. Through an integrated pipeline of VLM-based tool and contact point grounding, position-velocity trajectory planning, and physics-informed grasp generation and selection, iTUP demonstrates versatility across (1) quasi-static and more challenging (2) dynamic and (3) cluster tool-use tasks. To ensure robust planning, our framework integrates stable and safe task-aware grasping by reasoning over semantic affordances and physical constraints. We evaluate iTUP and baselines on a comprehensive range of realistic tool use tasks including precision hammering, object scooping, and cluster sweeping. Experimental results demonstrate that iTUP ensures a thorough grounding of cognition and planning for challenging robot tool use across diverse environments.

Paperid: 4829, https://arxiv.org/pdf/2504.14224.pdf

Abstract:
Vision-language models (VLMs) have gained widespread attention for their strong zero-shot capabilities across numerous downstream tasks. However, these models assume that each test image's class label is drawn from a predefined label set and lack a reliable mechanism to reject samples from emerging unknown classes when only unlabeled data are available. To address this gap, open-set domain adaptation methods retrain models to push potential unknowns away from known clusters. Yet, some unknown samples remain stably anchored to specific known classes in the VLM feature space due to semantic relevance, which is termed as Semantic Affinity Anchoring (SAA). Forcibly repelling these samples unavoidably distorts the native geometry of VLMs and degrades performance. Meanwhile, existing score-based unknown detectors use simplistic thresholds and suffer from threshold sensitivity, resulting in sub-optimal performance. To address aforementioned issues, we propose VLM-OpenXpert, which comprises two training-free, plug-and-play inference modules. SUFF performs SVD on high-confidence unknowns to extract a low-rank "unknown subspace". Each sample's projection onto this subspace is weighted and softly removed from its feature, suppressing unknown components while preserving semantics. BGAT corrects score skewness via a Box-Cox transform, then fits a bimodal Gaussian mixture to adaptively estimate the optimal threshold balancing known-class recognition and unknown-class rejection. Experiments on 9 benchmarks and three backbones (CLIP, SigLIP, ALIGN) under source-free OSDA settings show that our training-free pipeline matches or outperforms retraining-heavy state-of-the-art methods, establishing a powerful lightweight inference calibration paradigm for open-set VLM deployment.

Paperid: 4830, https://arxiv.org/pdf/2504.05457.pdf

Abstract:
When a vision-language model (VLM) is prompted to identify an entity depicted in an image, it may answer 'I see a conifer,' rather than the specific label 'norway spruce'. This raises two issues for evaluation: First, the unconstrained generated text needs to be mapped to the evaluation label space (i.e., 'conifer'). Second, a useful classification measure should give partial credit to less-specific, but not incorrect, answers ('norway spruce' being a type of 'conifer'). To meet these requirements, we propose a framework for evaluating unconstrained text predictions, such as those generated from a vision-language model, against a taxonomy. Specifically, we propose the use of hierarchical precision and recall measures to assess the level of correctness and specificity of predictions with regard to a taxonomy. Experimentally, we first show that existing text similarity measures do not capture taxonomic similarity well. We then develop and compare different methods to map textual VLM predictions onto a taxonomy. This allows us to compute hierarchical similarity measures between the generated text and the ground truth labels. Finally, we analyze modern VLMs on fine-grained visual classification tasks based on our proposed taxonomic evaluation scheme.

Paperid: 4831, https://arxiv.org/pdf/2503.23806.pdf

Abstract:
Generalized zero-shot semantic segmentation (GZS3) aims to achieve the human-level capability of segmenting not only seen classes but also novel class regions unseen in the training data through introducing the bridge of semantic representations, e.g., word vector. While effective, the way of utilizing one semantic representation to associate the corresponding class and to enable the knowledge transfer from seen to unseen classes is insufficient as well as incompatible with human cognition. Inspired by the observation that humans often use some `part' and `state' information to comprehend the seen objects and imagine unseen classes, we decouple each class into detailed descriptions, including object parts and states. Based on the decoupling formulation, we propose a Decoupled Vision-Language Matching (DeVLMatch) framework, composed of spatial-part (SPMatch) and channel-state (CSMatch) matching modules, for GZS3. In SPMatch, we comprehend objects with spatial part information from both visual and linguistic perspectives and perform graph matching to bridge the gap. In CSMatch, states of objects from the linguistic perspective are matched to compatible channel information from the visual perspective. By decoupling and matching objects across visual and linguistic comprehension, we can explicitly introspect the relationship between seen and unseen classes in fine-grained object part and state levels, thereby facilitating the knowledge transfer from seen to unseen classes in visual space. The proposed DeVLMatch framework surpasses the previous GZS3 methods on standard benchmarks, including PASCAL VOC, COCO-Stuff, and CATARACTS, demonstrating its effectiveness.

Paperid: 4832, https://arxiv.org/pdf/2503.23105.pdf

Abstract:
The global rise in the number of people with physical disabilities, in part due to improvements in post-trauma survivorship and longevity, has amplified the demand for advanced assistive technologies to improve mobility and independence. Autonomous assistive robots, such as smart wheelchairs, require robust capabilities in spatial segmentation and semantic recognition to navigate complex built environments effectively. Place segmentation involves delineating spatial regions like rooms or functional areas, while semantic recognition assigns semantic labels to these regions, enabling accurate localization to user-specific needs. Existing approaches often utilize deep learning; however, these close-vocabulary detection systems struggle to interpret intuitive and casual human instructions. Additionally, most existing methods ignore the uncertainty of the scene recognition problem, leading to low success rates, particularly in ambiguous and complex environments. To address these challenges, we propose an open-vocabulary scene semantic segmentation and detection pipeline leveraging Vision Language Models (VLMs) and Large Language Models (LLMs). Our approach follows a 'Segment Detect Select' framework for open-vocabulary scene classification, enabling adaptive and intuitive navigation for assistive robots in built environments.

Paperid: 4833, https://arxiv.org/pdf/2503.20348.pdf

Abstract:
Vision-language foundation models have shown impressive capabilities across various zero-shot tasks, including training-free localization and grounding, primarily focusing on localizing objects in images. However, leveraging those capabilities to localize actions and events in videos is challenging, as actions have less physical outline and are usually described by higher-level concepts. In this work, we propose VideoGEM, the first training-free spatial action grounding method based on pretrained image- and video-language backbones. Namely, we adapt the self-self attention formulation of GEM to spatial activity grounding. We observe that high-level semantic concepts, such as actions, usually emerge in the higher layers of the image- and video-language models. We, therefore, propose a layer weighting in the self-attention path to prioritize higher layers. Additionally, we introduce a dynamic weighting method to automatically tune layer weights to capture each layer`s relevance to a specific prompt. Finally, we introduce a prompt decomposition, processing action, verb, and object prompts separately, resulting in a better spatial localization of actions. We evaluate the proposed approach on three image- and video-language backbones, CLIP, OpenCLIP, and ViCLIP, and on four video grounding datasets, V-HICO, DALY, YouCook-Interactions, and GroundingYouTube, showing that the proposed training-free approach is able to outperform current trained state-of-the-art approaches for spatial video grounding.

Paperid: 4834, https://arxiv.org/pdf/2503.20279.pdf

Abstract:
Large Language Models (LLMs) are increasingly deployed as computer-use agents, autonomously performing tasks within real desktop or web environments. While this evolution greatly expands practical use cases for humans, it also creates serious security exposures. We present SUDO (Screen-based Universal Detox2Tox Offense), a novel attack framework that systematically bypasses refusal-trained safeguards in commercial computer-use agents, such as Claude for Computer Use. The core mechanism, Detox2Tox, transforms harmful requests (that agents initially reject) into seemingly benign requests via detoxification, secures detailed instructions from advanced vision language models (VLMs), and then reintroduces malicious content via toxification just before execution. Unlike conventional jailbreaks, SUDO iteratively refines its attacks based on a built-in refusal feedback, making it increasingly effective against robust policy filters. In extensive tests spanning 50 real-world tasks and multiple state-of-the-art VLMs, SUDO achieves a stark attack success rate of 24.41% (with no refinement), and up to 41.33% (by its iterative refinement) in Claude for Computer Use. By revealing these vulnerabilities and demonstrating the ease with which they can be exploited in real-world computing environments, this paper highlights an immediate need for robust, context-aware safeguards. WARNING: This paper includes harmful or offensive model outputs

Paperid: 4835, https://arxiv.org/pdf/2503.14530.pdf

Abstract:
Unlearning methods for vision-language models (VLMs) have primarily adapted techniques from large language models (LLMs), relying on weight updates that demand extensive annotated forget sets. Moreover, these methods perform unlearning at a coarse granularity, often leading to excessive forgetting and reduced model utility. To address this issue, we introduce SAUCE, a novel method that leverages sparse autoencoders (SAEs) for fine-grained and selective concept unlearning in VLMs. Briefly, SAUCE first trains SAEs to capture high-dimensional, semantically rich sparse features. It then identifies the features most relevant to the target concept for unlearning. During inference, it selectively modifies these features to suppress specific concepts while preserving unrelated information. We evaluate SAUCE on two distinct VLMs, LLaVA-v1.5-7B and LLaMA-3.2-11B-Vision-Instruct, across two types of tasks: concrete concept unlearning (objects and sports scenes) and abstract concept unlearning (emotions, colors, and materials), encompassing a total of 60 concepts. Extensive experiments demonstrate that SAUCE outperforms state-of-the-art methods by 18.04% in unlearning quality while maintaining comparable model utility. Furthermore, we investigate SAUCE's robustness against widely used adversarial attacks, its transferability across models, and its scalability in handling multiple simultaneous unlearning requests. Our findings establish SAUCE as an effective and scalable solution for selective concept unlearning in VLMs.

Paperid: 4836, https://arxiv.org/pdf/2503.13108.pdf

Abstract:
Multimodal large language models (MLLMs) improve performance on vision-language tasks by integrating visual features from pre-trained vision encoders into large language models (LLMs). However, how MLLMs process and utilize visual information remains unclear. In this paper, a shift in the dominant flow of visual information is uncovered: (1) in shallow layers, strong interactions are observed between image tokens and instruction tokens, where most visual information is injected into instruction tokens to form cross-modal semantic representations; (2) in deeper layers, image tokens primarily interact with each other, aggregating the remaining visual information to optimize semantic representations within visual modality. Based on these insights, we propose Hierarchical Modality-Aware Pruning (HiMAP), a plug-and-play inference acceleration method that dynamically prunes image tokens at specific layers, reducing computational costs by approximately 65% without sacrificing performance. Our findings offer a new understanding of visual information processing in MLLMs and provide a state-of-the-art solution for efficient inference.

Paperid: 4837, https://arxiv.org/pdf/2503.06442.pdf

Abstract:
Out-of-distribution (OOD) detection is crucial for ensuring the reliability and safety of machine learning models in real-world applications. While zero-shot OOD detection, which requires no training on in-distribution (ID) data, has become feasible with the emergence of vision-language models like CLIP, existing methods primarily focus on semantic matching and fail to fully capture distributional discrepancies. To address these limitations, we propose OT-DETECTOR, a novel framework that employs Optimal Transport (OT) to quantify both semantic and distributional discrepancies between test samples and ID labels. Specifically, we introduce cross-modal transport mass and transport cost as semantic-wise and distribution-wise OOD scores, respectively, enabling more robust detection of OOD samples. Additionally, we present a semantic-aware content refinement (SaCR) module, which utilizes semantic cues from ID labels to amplify the distributional discrepancy between ID and hard OOD samples. Extensive experiments on several benchmarks demonstrate that OT-DETECTOR achieves state-of-the-art performance across various OOD detection tasks, particularly in challenging hard-OOD scenarios.

Paperid: 4838, https://arxiv.org/pdf/2503.06271.pdf

Abstract:
Language-guided 3D scene understanding is important for advancing applications in robotics, AR/VR, and human-computer interaction, enabling models to comprehend and interact with 3D environments through natural language. While 2D vision-language models (VLMs) have achieved remarkable success in 2D VQA tasks, progress in the 3D domain has been significantly slower due to the complexity of 3D data and the high cost of manual annotations. In this work, we introduce SplatTalk, a novel method that uses a generalizable 3D Gaussian Splatting (3DGS) framework to produce 3D tokens suitable for direct input into a pretrained LLM, enabling effective zero-shot 3D visual question answering (3D VQA) for scenes with only posed images. During experiments on multiple benchmarks, our approach outperforms both 3D models trained specifically for the task and previous 2D-LMM-based models utilizing only images (our setting), while achieving competitive performance with state-of-the-art 3D LMMs that additionally utilize 3D inputs. Project website: https://splat-talk.github.io/

Paperid: 4839, https://arxiv.org/pdf/2503.05833.pdf

Abstract:
Vision-Language-Action Models (VLAs) have demonstrated remarkable generalization capabilities in real-world experiments. However, their success rates are often not on par with expert policies, and they require fine-tuning when the setup changes. In this work, we introduce Refined Policy Distillation (RPD), a novel Reinforcement Learning (RL)-based policy refinement method that bridges this performance gap through a combination of on-policy RL with behavioral cloning. The core idea of RPD is to distill and refine VLAs into compact, high-performing expert policies by guiding the student policy during RL exploration using the actions of a teacher VLA, resulting in increased sample efficiency and faster convergence. We complement our method by fine-tuned versions of Octo and OpenVLA for ManiSkill3 to evaluate RPD in simulation. While this is a key requirement for applying RL, it also yields new insights beyond existing studies on VLA performance in real-world settings. Our experimental results across various manipulation tasks show that RPD enables the RL student to learn expert policies that outperform the VLA teacher in both dense and sparse reward settings, while also achieving faster convergence than the RL baseline. Our approach is even robust to changes in camera perspective and can generalize to task variations that the underlying VLA cannot solve. Our code, dataset, VLA checkpoints, and videos are available at https://refined-policy-distillation.github.io

Paperid: 4840, https://arxiv.org/pdf/2503.02206.pdf

Abstract:
Contrastive vision-language models, such as CLIP, have demonstrated excellent zero-shot capability across semantic recognition tasks, mainly attributed to the training on a large-scale I&1T (one Image with one Text) dataset. This kind of multimodal representations often blend semantic and perceptual elements, placing a particular emphasis on semantics. However, this could be problematic for popular tasks like image quality assessment (IQA) and conditional image generation (CIG), which typically need to have fine control on perceptual and semantic features. Motivated by the above facts, this paper presents a new multimodal disentangled representation learning framework, which leverages disentangled text to guide image disentanglement. To this end, we first build an I&2T (one Image with a perceptual Text and a semantic Text) dataset, which consists of disentangled perceptual and semantic text descriptions for an image. Then, the disentangled text descriptions are utilized as supervisory signals to disentangle pure perceptual representations from CLIP's original `coarse' feature space, dubbed DeCLIP. Finally, the decoupled feature representations are used for both image quality assessment (technical quality and aesthetic quality) and conditional image generation. Extensive experiments and comparisons have demonstrated the advantages of the proposed method on the two popular tasks. The dataset, code, and model will be available.

Paperid: 4841, https://arxiv.org/pdf/2502.20560.pdf

Abstract:
Advancements in Large Vision-Language Models (LVLMs) have demonstrated promising performance in a variety of vision-language tasks involving image-conditioned free-form text generation. However, growing concerns about hallucinations in LVLMs, where the generated text is inconsistent with the visual context, are becoming a major impediment to deploying these models in applications that demand guaranteed reliability. In this paper, we introduce a framework to address this challenge, ConfLVLM, which is grounded on conformal prediction to achieve finite-sample distribution-free statistical guarantees on the factuality of LVLM output. This framework treats an LVLM as a hypothesis generator, where each generated text detail (or claim) is considered an individual hypothesis. It then applies a statistical hypothesis testing procedure to verify each claim using efficient heuristic uncertainty measures to filter out unreliable claims before returning any responses to users. We conduct extensive experiments covering three representative application domains, including general scene understanding, medical radiology report generation, and document understanding. Remarkably, ConfLVLM reduces the error rate of claims generated by LLaVa-1.5 for scene descriptions from 87.8\% to 10.0\% by filtering out erroneous claims with a 95.3\% true positive rate. Our results further demonstrate that ConfLVLM is highly flexible, and can be applied to any black-box LVLMs paired with any uncertainty measure for any image-conditioned free-form text generation task while providing a rigorous guarantee on controlling the risk of hallucination.

Paperid: 4842, https://arxiv.org/pdf/2502.20292.pdf

Abstract:
Vision-Language Models (VLMs) have demonstrated impressive multimodal capabilities in learning joint representations of visual and textual data, making them powerful tools for tasks such as Compositional Zero-Shot Learning (CZSL). CZSL requires models to generalize to novel combinations of visual primitives--such as attributes and objects--that were not explicitly encountered during training. Recent works in prompting for CZSL have focused on modifying inputs for the text encoder, often using static prompts that do not change across varying visual contexts. However, these approaches struggle to fully capture varying visual contexts, as they focus on text adaptation rather than leveraging visual features for compositional reasoning. To address this, we propose a Visual Adaptive Prompting System (VAPS) that leverages a learnable visual prompt repository and similarity-based retrieval mechanism within the framework of VLMs to bridge the gap between semantic and visual features. Our method introduces a dynamic visual prompt repository mechanism that selects the most relevant attribute and object prompts based on the visual features of the image. Our proposed system includes a visual prompt adapter that encourages the model to learn a more generalizable embedding space. Experiments on three CZSL benchmarks, across both closed and open-world scenarios, demonstrate state-of-the-art results.

Paperid: 4843, https://arxiv.org/pdf/2502.15481.pdf

Abstract:
Recently, employing single-modality large language models based on mechanical vibration signals as Tuning Predictors has introduced new perspectives in intelligent fault diagnosis. However, the potential of these methods to leverage multimodal data remains underexploited, particularly in complex mechanical systems where relying on a single data source often fails to capture comprehensive fault information. In this paper, we present FaultGPT, a novel model that generates fault diagnosis reports directly from raw vibration signals. By leveraging large vision-language models (LVLM) and text-based supervision, FaultGPT performs end-to-end fault diagnosis question answering (FDQA), distinguishing itself from traditional classification or regression approaches. Specifically, we construct a large-scale FDQA instruction dataset for instruction tuning of LVLM. This dataset includes vibration time-frequency image-text label pairs and human instruction-ground truth pairs. To enhance the capability in generating high-quality fault diagnosis reports, we design a multi-scale cross-modal image decoder to extract fine-grained fault semantics and conducted instruction tuning without introducing additional training parameters into the LVLM. Extensive experiments, including fault diagnosis report generation, few-shot and zero-shot evaluation across multiple datasets, validate the superior performance and adaptability of FaultGPT in diverse industrial scenarios.

Paperid: 4844, https://arxiv.org/pdf/2502.14125.pdf

Abstract:
Pre-trained vision-language models are able to interpret visual concepts and language semantics. Prompt learning, a method of constructing prompts for text encoders or image encoders, elicits the potentials of pre-trained models and readily adapts them to new scenarios. Compared to fine-tuning, prompt learning enables the model to achieve comparable or better performance using fewer trainable parameters. Besides, prompt learning freezes the pre-trained model and avoids the catastrophic forgetting issue in the fine-tuning. Continuous prompts inserted into the input of every transformer layer (i.e. deep prompts) can improve the performances of pre-trained models on downstream tasks. For i-th transformer layer, the inserted prompts replace previously inserted prompts in the $(i-1)$-th layer. Although the self-attention mechanism contextualizes newly inserted prompts for the current layer and embeddings from the previous layer's output, removing all inserted prompts from the previous layer inevitably loses information contained in the continuous prompts. In this work, we propose Modular Prompt Learning (MPL) that is designed to promote the preservation of information contained in the inserted prompts. We evaluate the proposed method on base-to-new generalization and cross-dataset tasks. On average of 11 datasets, our method achieves 0.7% performance gain on the base-to-new generalization task compared to the state-of-the-art method. The largest improvement on the individual dataset is 10.7% (EuroSAT dataset).

Paperid: 4845, https://arxiv.org/pdf/2502.03459.pdf

Abstract:
The introduction of vision-language models like CLIP has enabled the development of foundational video models capable of generalizing to unseen videos and human actions. However, these models are typically trained on web videos, which often fail to capture the challenges present in Activities of Daily Living (ADL) videos. Existing works address ADL-specific challenges, such as similar appearances, subtle motion patterns, and multiple viewpoints, by combining 3D skeletons and RGB videos. However, these approaches are not integrated with language, limiting their ability to generalize to unseen action classes. In this paper, we introduce SKI models, which integrate 3D skeletons into the vision-language embedding space. SKI models leverage a skeleton-language model, SkeletonCLIP, to infuse skeleton information into Vision Language Models (VLMs) and Large Vision Language Models (LVLMs) through collaborative training. Notably, SKI models do not require skeleton data during inference, enhancing their robustness for real-world applications. The effectiveness of SKI models is validated on three popular ADL datasets for zero-shot action recognition and video caption generation tasks.

Paperid: 4846, https://arxiv.org/pdf/2502.02740.pdf

Abstract:
The increasing demand for high-quality, diverse training data poses a significant bottleneck in advancing vision-language models (VLMs). This paper presents VLM Dialog Games, a novel and scalable self-improvement framework for VLMs. Our approach leverages self-play between two agents engaged in a goal-oriented play centered around image identification. By filtering for successful game interactions, we automatically curate a high-quality dataset of interleaved images and text. We demonstrate that fine-tuning on this synthetic data leads to performance gains on downstream tasks and generalises across datasets. Moreover, as the improvements in the model lead to better game play, this procedure can be applied iteratively. This work paves the way for self-improving VLMs, with potential applications in various real-world scenarios especially when the high-quality multimodal data is scarce.

Paperid: 4847, https://arxiv.org/pdf/2502.01785.pdf

Abstract:
The preservation of aquatic biodiversity is critical in mitigating the effects of climate change. Aquatic scene understanding plays a pivotal role in aiding marine scientists in their decision-making processes. In this paper, we introduce AquaticCLIP, a novel contrastive language-image pre-training model tailored for aquatic scene understanding. AquaticCLIP presents a new unsupervised learning framework that aligns images and texts in aquatic environments, enabling tasks such as segmentation, classification, detection, and object counting. By leveraging our large-scale underwater image-text paired dataset without the need for ground-truth annotations, our model enriches existing vision-language models in the aquatic domain. For this purpose, we construct a 2 million underwater image-text paired dataset using heterogeneous resources, including YouTube, Netflix, NatGeo, etc. To fine-tune AquaticCLIP, we propose a prompt-guided vision encoder that progressively aggregates patch features via learnable prompts, while a vision-guided mechanism enhances the language encoder by incorporating visual context. The model is optimized through a contrastive pretraining loss to align visual and textual modalities. AquaticCLIP achieves notable performance improvements in zero-shot settings across multiple underwater computer vision tasks, outperforming existing methods in both robustness and interpretability. Our model sets a new benchmark for vision-language applications in underwater environments. The code and dataset for AquaticCLIP are publicly available on GitHub at xxx.

Paperid: 4848, https://arxiv.org/pdf/2501.16378.pdf

Abstract:
Vision-language models (VLMs) demonstrate strong multimodal capabilities but have been found to be more susceptible to generating harmful content compared to their backbone large language models (LLMs). Our investigation reveals that the integration of images significantly shifts the model's internal activations during the forward pass, diverging from those triggered by textual input. Moreover, the safety alignments of LLMs embedded within VLMs are not sufficiently robust to handle the activations discrepancies, making the models vulnerable to even the simplest jailbreaking attacks. To address this issue, we propose an \textbf{internal activation revision} approach that efficiently revises activations during generation, steering the model toward safer outputs. Our framework incorporates revisions at both the layer and head levels, offering control over the model's generation at varying levels of granularity. In addition, we explore three strategies for constructing positive and negative samples and two approaches for extracting revision vectors, resulting in different variants of our method. Comprehensive experiments demonstrate that the internal activation revision method significantly improves the safety of widely used VLMs, reducing attack success rates by an average of 48.94\%, 34.34\%, 43.92\%, and 52.98\% on SafeBench, Safe-Unsafe, Unsafe, and MM-SafetyBench, respectively, while minimally impacting model helpfulness.

Paperid: 4849, https://arxiv.org/pdf/2501.08347.pdf

Abstract:
Compositional image retrieval (CIR) is a multimodal learning task where a model combines a query image with a user-provided text modification to retrieve a target image. CIR finds applications in a variety of domains including product retrieval (e-commerce) and web search. Existing methods primarily focus on fully-supervised learning, wherein models are trained on datasets of labeled triplets such as FashionIQ and CIRR. This poses two significant challenges: (i) curating such triplet datasets is labor intensive; and (ii) models lack generalization to unseen objects and domains. In this work, we propose SCOT (Self-supervised COmpositional Training), a novel zero-shot compositional pretraining strategy that combines existing large image-text pair datasets with the generative capabilities of large language models to contrastively train an embedding composition network. Specifically, we show that the text embedding from a large-scale contrastively-pretrained vision-language model can be utilized as proxy target supervision during compositional pretraining, replacing the target image embedding. In zero-shot settings, this strategy surpasses SOTA zero-shot compositional retrieval methods as well as many fully-supervised methods on standard benchmarks such as FashionIQ and CIRR.

Paperid: 4850, https://arxiv.org/pdf/2501.08042.pdf

Abstract:
Ewing's sarcoma (ES), characterized by a high density of small round blue cells without structural organization, presents a significant health concern, particularly among adolescents aged 10 to 19. Artificial intelligence-based systems for automated analysis of histopathological images are promising to contribute to an accurate diagnosis of ES. In this context, this study explores the feature extraction ability of different pre-training strategies for distinguishing ES from other soft tissue or bone sarcomas with similar morphology in digitized tissue microarrays for the first time, as far as we know. Vision-language supervision (VLS) is compared to fully-supervised ImageNet pre-training within a multiple instance learning paradigm. Our findings indicate a substantial improvement in diagnostic accuracy with the adaption of VLS using an in-domain dataset. Notably, these models not only enhance the accuracy of predicted classes but also drastically reduce the number of trainable parameters and computational costs.

Paperid: 4851, https://arxiv.org/pdf/2501.07802.pdf

Abstract:
This paper explores the application of Vision-Language Models (VLMs) as operator agents in the space domain, focusing on both software and hardware operational paradigms. Building on advances in Large Language Models (LLMs) and their multimodal extensions, we investigate how VLMs can enhance autonomous control and decision-making in space missions. In the software context, we employ VLMs within the Kerbal Space Program Differential Games (KSPDG) simulation environment, enabling the agent to interpret visual screenshots of the graphical user interface to perform complex orbital maneuvers. In the hardware context, we integrate VLMs with robotic systems equipped with cameras to inspect and diagnose physical space objects, such as satellites. Our results demonstrate that VLMs can effectively process visual and textual data to generate contextually appropriate actions, competing with traditional methods and non-multimodal LLMs in simulation tasks, and showing promise in real-world applications.

Paperid: 4852, https://arxiv.org/pdf/2501.04947.pdf

Abstract:
In assistive robotics serving people with disabilities (PWD), accurate place recognition in built environments is crucial to ensure that robots navigate and interact safely within diverse indoor spaces. Language interfaces, particularly those powered by Large Language Models (LLM) and Vision Language Models (VLM), hold significant promise in this context, as they can interpret visual scenes and correlate them with semantic information. However, such interfaces are also known for their hallucinated predictions. In addition, language instructions provided by humans can also be ambiguous and lack precise details about specific locations, objects, or actions, exacerbating the hallucination issue. In this work, we introduce Seeing with Partial Certainty (SwPC) - a framework designed to measure and align uncertainty in VLM-based place recognition, enabling the model to recognize when it lacks confidence and seek assistance when necessary. This framework is built on the theory of conformal prediction to provide statistical guarantees on place recognition while minimizing requests for human help in complex indoor environment settings. Through experiments on the widely used richly-annotated scene dataset Matterport3D, we show that SwPC significantly increases the success rate and decreases the amount of human intervention required relative to the prior art. SwPC can be utilized with any VLMs directly without requiring model fine-tuning, offering a promising, lightweight approach to uncertainty modeling that complements and scales alongside the expanding capabilities of foundational models.

Paperid: 4853, https://arxiv.org/pdf/2501.00457.pdf

Abstract:
Prompt learning is an effective way to exploit the potential of large-scale pre-trained foundational models. Continuous prompts parameterize context tokens in prompts by turning them into differentiable vectors. Deep continuous prompts insert prompts not only in the input but also in the intermediate hidden representations. Manually designed deep continuous prompts exhibit a remarkable improvement compared to the zero-shot pre-trained model on downstream tasks. How to automate the continuous prompt design is an underexplored area, and a fundamental question arises, is manually designed deep prompt strategy optimal? To answer this question, we propose a method dubbed differentiable prompt learning (DPL). The DPL method is formulated as an optimization problem to automatically determine the optimal context length of the prompt to be added to each layer, where the objective is to maximize the performance. We test the DPL method on the pre-trained CLIP. We empirically find that by using only limited data, our DPL method can find deep continuous prompt configuration with high confidence. The performance on the downstream tasks exhibits the superiority of the automatic design: our method boosts the average test accuracy by 2.60% on 11 datasets compared to baseline methods. Besides, our method focuses only on the prompt configuration (i.e. context length for each layer), which means that our method is compatible with the baseline methods that have sophisticated designs to boost the performance. The DPL method can be deployed to large language models or computer vision models at no cost.

Paperid: 4854, https://arxiv.org/pdf/2412.20682.pdf

Abstract:
Vision language models (VLMs) like CLIP show stellar zero-shot capability on classification benchmarks. However, selecting the VLM with the highest performance on the unlabeled downstream task is non-trivial. Existing VLM selection methods focus on the class-name-only setting, relying on a supervised large-scale dataset and large language models, which may not be accessible or feasible during deployment. This paper introduces the problem of \textbf{unsupervised vision-language model selection}, where only unsupervised downstream datasets are available, with no additional information provided. To solve this problem, we propose a method termed Visual-tExtual Graph Alignment (VEGA), to select VLMs without any annotations by measuring the alignment of the VLM between the two modalities on the downstream task. VEGA is motivated by the pretraining paradigm of VLMs, which aligns features with the same semantics from the visual and textual modalities, thereby mapping both modalities into a shared representation space. Specifically, we first construct two graphs on the vision and textual features, respectively. VEGA is then defined as the overall similarity between the visual and textual graphs at both node and edge levels. Extensive experiments across three different benchmarks, covering a variety of application scenarios and downstream datasets, demonstrate that VEGA consistently provides reliable and accurate estimates of VLMs' performance on unlabeled downstream tasks.

Paperid: 4855, https://arxiv.org/pdf/2412.18387.pdf

Abstract:
Large language models have demonstrated predictable scaling behaviors with respect to model parameters and training data. This study investigates whether a similar scaling relationship exist for vision-language models with respect to the number of vision tokens. A mathematical framework is developed to characterize a relationship between vision token number and the expected divergence of distance between vision-referencing sequences. The theoretical analysis reveals two distinct scaling regimes: sublinear scaling for less vision tokens and linear scaling for more vision tokens. This aligns with model performance relationships of the form $S(n) \approx c / n^{Î±(n)}$, where the scaling exponent relates to the correlation structure between vision token representations. Empirical validations across multiple vision-language benchmarks show that model performance matches the prediction from scaling relationship. The findings contribute to understanding vision token scaling in transformers through a theoretical framework that complements empirical observations.

Paperid: 4856, https://arxiv.org/pdf/2412.16418.pdf

Abstract:
With the rapid advancement of Multimodal Large Language Models (MLLMs), a variety of benchmarks have been introduced to evaluate their capabilities. While most evaluations have focused on complex tasks such as scientific comprehension and visual reasoning, little attention has been given to assessing their fundamental image classification abilities. In this paper, we address this gap by thoroughly revisiting the MLLMs with an in-depth analysis of image classification. Specifically, building on established datasets, we examine a broad spectrum of scenarios, from general classification tasks (e.g., ImageNet, ObjectNet) to more fine-grained categories such as bird and food classification. Our findings reveal that the most recent MLLMs can match or even outperform CLIP-style vision-language models on several datasets, challenging the previous assumption that MLLMs are bad at image classification \cite{VLMClassifier}. To understand the factors driving this improvement, we conduct an in-depth analysis of the network architecture, data selection, and training recipe used in public MLLMs. Our results attribute this success to advancements in language models and the diversity of training data sources. Based on these observations, we further analyze and attribute the potential reasons to conceptual knowledge transfer and enhanced exposure of target concepts, respectively. We hope our findings will offer valuable insights for future research on MLLMs and their evaluation in image classification tasks.

Paperid: 4857, https://arxiv.org/pdf/2412.12898.pdf

Abstract:
The Piping and Instrumentation Diagrams (P&IDs) are foundational to the design, construction, and operation of workflows in the engineering and process industries. However, their manual creation is often labor-intensive, error-prone, and lacks robust mechanisms for error detection and correction. While recent advancements in Generative AI, particularly Large Language Models (LLMs) and Vision-Language Models (VLMs), have demonstrated significant potential across various domains, their application in automating generation of engineering workflows remains underexplored. In this work, we introduce a novel copilot for automating the generation of P&IDs from natural language descriptions. Leveraging a multi-step agentic workflow, our copilot provides a structured and iterative approach to diagram creation directly from Natural Language prompts. We demonstrate the feasibility of the generation process by evaluating the soundness and completeness of the workflow, and show improved results compared to vanilla zero-shot and few-shot generation approaches.

Paperid: 4858, https://arxiv.org/pdf/2412.10694.pdf

Abstract:
In recent years, as robotics has advanced, human-robot collaboration has gained increasing importance. However, current robots struggle to fully and accurately interpret human intentions from voice commands alone. Traditional gripper and suction systems often fail to interact naturally with humans, lack advanced manipulation capabilities, and are not adaptable to diverse tasks, especially in unstructured environments. This paper introduces the Embodied Dexterous Grasping System (EDGS), designed to tackle object grasping in cluttered environments for human-robot interaction. We propose a novel approach to semantic-object alignment using a Vision-Language Model (VLM) that fuses voice commands and visual information, significantly enhancing the alignment of multi-dimensional attributes of target objects in complex scenarios. Inspired by human hand-object interactions, we develop a robust, precise, and efficient grasping strategy, incorporating principles like the thumb-object axis, multi-finger wrapping, and fingertip interaction with an object's contact mechanics. We also design experiments to assess Referring Expression Representation Enrichment (RERE) in referring expression segmentation, demonstrating that our system accurately detects and matches referring expressions. Extensive experiments confirm that EDGS can effectively handle complex grasping tasks, achieving stability and high success rates, highlighting its potential for further development in the field of Embodied AI.

Paperid: 4859, https://arxiv.org/pdf/2412.08755.pdf

Abstract:
Backdoor attacks pose a critical threat by embedding hidden triggers into inputs, causing models to misclassify them into target labels. While extensive research has focused on mitigating these attacks in object recognition models through weight fine-tuning, much less attention has been given to detecting backdoored samples directly. Given the vast datasets used in training, manual inspection for backdoor triggers is impractical, and even state-of-the-art defense mechanisms fail to fully neutralize their impact. To address this gap, we introduce a groundbreaking method to detect unseen backdoored images during both training and inference. Leveraging the transformative success of prompt tuning in Vision Language Models (VLMs), our approach trains learnable text prompts to differentiate clean images from those with hidden backdoor triggers. Experiments demonstrate the exceptional efficacy of this method, achieving an impressive average accuracy of 86% across two renowned datasets for detecting unseen backdoor triggers, establishing a new standard in backdoor defense.

Paperid: 4860, https://arxiv.org/pdf/2412.08536.pdf

Abstract:
Pre-trained vision-language models (VLMs), such as CLIP, demonstrate impressive zero-shot classification capabilities with free-form prompts and even show some generalization in specialized domains. However, their performance on satellite imagery is limited due to the underrepresentation of such data in their training sets, which predominantly consist of ground-level images. Existing prompting techniques for satellite imagery are often restricted to generic phrases like a satellite image of ..., limiting their effectiveness for zero-shot land-use and land-cover (LULC) mapping. To address these challenges, we introduce SenCLIP, which transfers CLIPs representation to Sentinel-2 imagery by leveraging a large dataset of Sentinel-2 images paired with geotagged ground-level photos from across Europe. We evaluate SenCLIP alongside other SOTA remote sensing VLMs on zero-shot LULC mapping tasks using the EuroSAT and BigEarthNet datasets with both aerial and ground-level prompting styles. Our approach, which aligns ground-level representations with satellite imagery, demonstrates significant improvements in classification accuracy across both prompt styles, opening new possibilities for applying free-form textual descriptions in zero-shot LULC mapping.

Paperid: 4861, https://arxiv.org/pdf/2412.06760.pdf

Abstract:
Recent advances in vision-language models (VLMs) have made significant progress in downstream tasks that require quantitative concepts such as facial age estimation and image quality assessment, enabling VLMs to explore applications like image ranking and retrieval. However, existing studies typically focus on the reasoning based on a single image and heavily depend on text prompting, limiting their ability to learn comprehensive understanding from multiple images. To address this, we propose an effective yet efficient approach that reframes the CLIP model into a learning-to-rank task and introduces a lightweight adapter to augment CLIP for text-guided image ranking. Specifically, our approach incorporates learnable prompts to adapt to new instructions for ranking purposes and an auxiliary branch with ranking-aware attention, leveraging text-conditioned visual differences for additional supervision in image ranking. Our ranking-aware adapter consistently outperforms fine-tuned CLIPs on various tasks and achieves competitive results compared to state-of-the-art models designed for specific tasks like facial age estimation and image quality assessment. Overall, our approach primarily focuses on ranking images with a single instruction, which provides a natural and generalized way of learning from visual differences across images, bypassing the need for extensive text prompts tailored to individual tasks. Code is available: github.com/uynaes/RankingAwareCLIP.

Paperid: 4862, https://arxiv.org/pdf/2412.02449.pdf

Abstract:
Dynamic scene understanding remains a persistent challenge in robotic applications. Early dynamic mapping methods focused on mitigating the negative influence of short-term dynamic objects on camera motion estimation by masking or tracking specific categories, which often fall short in adapting to long-term scene changes. Recent efforts address object association in long-term dynamic environments using neural networks trained on synthetic datasets, but they still rely on predefined object shapes and categories. Other methods incorporate visual, geometric, or semantic heuristics for the association but often lack robustness. In this work, we introduce BYE, a class-agnostic, per-scene point cloud encoder that removes the need for predefined categories, shape priors, or extensive association datasets. Trained on only a single sequence of exploration data, BYE can efficiently perform object association in dynamically changing scenes. We further propose an ensembling scheme combining the semantic strengths of Vision Language Models (VLMs) with the scene-specific expertise of BYE, achieving a 7% improvement and a 95% success rate in object association tasks. Code and dataset are available at https://byencoder.github.io.

Paperid: 4863, https://arxiv.org/pdf/2411.18825.pdf

Abstract:
Reinforcement learning (RL) has demonstrated compelling performance in robotic tasks, but its success often hinges on the design of complex, ad hoc reward functions. Researchers have explored how Large Language Models (LLMs) could enable non-expert users to specify reward functions more easily. However, LLMs struggle to balance the importance of different features, generalize poorly to out-of-distribution robotic tasks, and cannot represent the problem properly with only text-based descriptions. To address these challenges, we propose ELEMENTAL (intEractive LEarning froM dEmoNstraTion And Language), a novel framework that combines natural language guidance with visual user demonstrations to align robot behavior with user intentions better. By incorporating visual inputs, ELEMENTAL overcomes the limitations of text-only task specifications, while leveraging inverse reinforcement learning (IRL) to balance feature weights and match the demonstrated behaviors optimally. ELEMENTAL also introduces an iterative feedback-loop through self-reflection to improve feature, reward, and policy learning. Our experiment results demonstrate that ELEMENTAL outperforms prior work by 42.3% on task success, and achieves 41.3% better generalization in out-of-distribution tasks, highlighting its robustness in LfD.

Paperid: 4864, https://arxiv.org/pdf/2411.18207.pdf

Abstract:
Traditional object detection methods operate under the closed-set assumption, where models can only detect a fixed number of objects predefined in the training set. Recent works on open vocabulary object detection (OVD) enable the detection of objects defined by an in-principle unbounded vocabulary, which reduces the cost of training models for specific tasks. However, OVD heavily relies on accurate prompts provided by an ``oracle'', which limits their use in critical applications such as driving scene perception. OVD models tend to misclassify near-out-of-distribution (NOOD) objects that have similar features to known classes, and ignore far-out-of-distribution (FOOD) objects. To address these limitations, we propose a framework that enables OVD models to operate in open world settings, by identifying and incrementally learning previously unseen objects. To detect FOOD objects, we propose Open World Embedding Learning (OWEL) and introduce the concept of Pseudo Unknown Embedding which infers the location of unknown classes in a continuous semantic space based on the information of known classes. We also propose Multi-Scale Contrastive Anchor Learning (MSCAL), which enables the identification of misclassified unknown objects by promoting the intra-class consistency of object embeddings at different scales. The proposed method achieves state-of-the-art performance on standard open world object detection and autonomous driving benchmarks while maintaining its open vocabulary object detection capability.

Paperid: 4865, https://arxiv.org/pdf/2411.16554.pdf

Abstract:
The deployment of autonomous vehicles controlled by machine learning techniques requires extensive testing in diverse real-world environments, robust handling of edge cases and out-of-distribution scenarios, and comprehensive safety validation to ensure that these systems can navigate safely and effectively under unpredictable conditions. Addressing Out-Of-Distribution (OOD) driving scenarios is essential for enhancing safety, as OOD scenarios help validate the reliability of the models within the vehicle's autonomy stack. However, generating OOD scenarios is challenging due to their long-tailed distribution and rarity in urban driving dataset. Recently, Large Language Models (LLMs) have shown promise in autonomous driving, particularly for their zero-shot generalization and common-sense reasoning capabilities. In this paper, we leverage these LLM strengths to introduce a framework for generating diverse OOD driving scenarios. Our approach uses LLMs to construct a branching tree, where each branch represents a unique OOD scenario. These scenarios are then simulated in the CARLA simulator using an automated framework that aligns scene augmentation with the corresponding textual descriptions. We evaluate our framework through extensive simulations, and assess its performance via a diversity metric that measures the richness of the scenarios. Additionally, we introduce a new "OOD-ness" metric, which quantifies how much the generated scenarios deviate from typical urban driving conditions. Furthermore, we explore the capacity of modern Vision-Language Models (VLMs) to interpret and safely navigate through the simulated OOD scenarios. Our findings offer valuable insights into the reliability of language models in addressing OOD scenarios within the context of urban driving.

Paperid: 4866, https://arxiv.org/pdf/2411.16523.pdf

Abstract:
In the current paradigm of image captioning, deep learning models are trained to generate text from image embeddings of latent features. We challenge the assumption that these latent features ought to be high-dimensional vectors which require model fine tuning to handle. Here we propose Label Boosted Retrieval Augmented Generation (LaB-RAG), a text-based approach to image captioning that leverages image descriptors in the form of categorical labels to boost standard retrieval augmented generation (RAG) with pretrained large language models (LLMs). We study our method in the context of radiology report generation (RRG), where the task is to generate a clinician's report detailing their observations from a set of radiological images, such as X-rays. We argue that simple linear classifiers over extracted image embeddings can effectively transform X-rays into text-space as radiology-specific labels. In combination with standard RAG, we show that these derived text labels can be used with general-domain LLMs to generate radiology reports. Without ever training our generative language model or image feature encoder models, and without ever directly "showing" the LLM an X-ray, we demonstrate that LaB-RAG achieves better results across natural language and radiology language metrics compared with other retrieval-based RRG methods, while attaining competitive results compared to other fine-tuned vision-language RRG models. We further present results of our experiments with various components of LaB-RAG to better understand our method. Finally, we critique the use of a popular RRG metric, arguing it is possible to artificially inflate its results without true data-leakage.

Paperid: 4867, https://arxiv.org/pdf/2411.04942.pdf

Abstract:
In this era of videos, automatic video editing techniques attract more and more attention from industry and academia since they can reduce workloads and lower the requirements for human editors. Existing automatic editing systems are mainly scene- or event-specific, e.g., soccer game broadcasting, yet the automatic systems for general editing, e.g., movie or vlog editing which covers various scenes and events, were rarely studied before, and converting the event-driven editing method to a general scene is nontrivial. In this paper, we propose a two-stage scheme for general editing. Firstly, unlike previous works that extract scene-specific features, we leverage the pre-trained Vision-Language Model (VLM) to extract the editing-relevant representations as editing context. Moreover, to close the gap between the professional-looking videos and the automatic productions generated with simple guidelines, we propose a Reinforcement Learning (RL)-based editing framework to formulate the editing problem and train the virtual editor to make better sequential editing decisions. Finally, we evaluate the proposed method on a more general editing task with a real movie dataset. Experimental results demonstrate the effectiveness and benefits of the proposed context representation and the learning ability of our RL-based editing framework.

Paperid: 4868, https://arxiv.org/pdf/2411.00238.pdf

Abstract:
Recent work has documented striking heterogeneity in the performance of state-of-the-art vision language models (VLMs), including both multimodal language models and text-to-image models. These models are able to describe and generate a diverse array of complex, naturalistic images, yet they exhibit surprising failures on basic multi-object reasoning tasks -- such as counting, localization, and simple forms of visual analogy -- that humans perform with near perfect accuracy. To better understand this puzzling pattern of successes and failures, we turn to theoretical accounts of the binding problem in cognitive science and neuroscience, a fundamental problem that arises when a shared set of representational resources must be used to represent distinct entities (e.g., to represent multiple objects in an image), necessitating the use of serial processing to avoid interference. We find that many of the puzzling failures of state-of-the-art VLMs can be explained as arising due to the binding problem, and that these failure modes are strikingly similar to the limitations exhibited by rapid, feedforward processing in the human brain.

Paperid: 4869, https://arxiv.org/pdf/2410.16840.pdf

Abstract:
Movie posters are vital for captivating audiences, conveying themes, and driving market competition in the film industry. While traditional designs are laborious, intelligent generation technology offers efficiency gains and design enhancements. Despite exciting progress in image generation, current models often fall short in producing satisfactory poster results. The primary issue lies in the absence of specialized poster datasets for targeted model training. In this work, we propose a Movie Posters DataSet (MPDS), tailored for text-to-image generation models to revolutionize poster production. As dedicated to posters, MPDS stands out as the first image-text pair dataset to our knowledge, composing of 373k+ image-text pairs and 8k+ actor images (covering 4k+ actors). Detailed poster descriptions, such as movie titles, genres, casts, and synopses, are meticulously organized and standardized based on public movie synopsis, also named movie-synopsis prompt. To bolster poster descriptions as well as reduce differences from movie synopsis, further, we leverage a large-scale vision-language model to automatically produce vision-perceptive prompts for each poster, then perform manual rectification and integration with movie-synopsis prompt. In addition, we introduce a prompt of poster captions to exhibit text elements in posters like actor names and movie titles. For movie poster generation, we develop a multi-condition diffusion framework that takes poster prompt, poster caption, and actor image (for personalization) as inputs, yielding excellent results through the learning of a diffusion model. Experiments demonstrate the valuable role of our proposed MPDS dataset in advancing personalized movie poster generation. MPDS is available at https://anonymous.4open.science/r/MPDS-373k-BD3B.

Paperid: 4870, https://arxiv.org/pdf/2410.10238.pdf

Abstract:
Multimodal Large Language Models (MLLMs), such as GPT4o, have shown strong capabilities in visual reasoning and explanation generation. However, despite these strengths, they face significant challenges in the increasingly critical task of Image Forgery Detection and Localization (IFDL). Moreover, existing IFDL methods are typically limited to the learning of low-level semantic-agnostic clues and merely provide a single outcome judgment. To tackle these issues, we propose ForgeryGPT, a novel framework that advances the IFDL task by capturing high-order forensics knowledge correlations of forged images from diverse linguistic feature spaces, while enabling explainable generation and interactive dialogue through a newly customized Large Language Model (LLM) architecture. Specifically, ForgeryGPT enhances traditional LLMs by integrating the Mask-Aware Forgery Extractor, which enables the excavating of precise forgery mask information from input images and facilitating pixel-level understanding of tampering artifacts. The Mask-Aware Forgery Extractor consists of a Forgery Localization Expert (FL-Expert) and a Mask Encoder, where the FL-Expert is augmented with an Object-agnostic Forgery Prompt and a Vocabulary-enhanced Vision Encoder, allowing for effectively capturing of multi-scale fine-grained forgery details. To enhance its performance, we implement a three-stage training strategy, supported by our designed Mask-Text Alignment and IFDL Task-Specific Instruction Tuning datasets, which align vision-language modalities and improve forgery detection and instruction-following capabilities. Extensive experiments demonstrate the effectiveness of the proposed method.

Paperid: 4871, https://arxiv.org/pdf/2410.08172.pdf

Abstract:
Due to the difficulty of acquiring extensive real-world data, robot simulation has become crucial for parallel training and sim-to-real transfer, highlighting the importance of scalable simulated robotic tasks. Foundation models have demonstrated impressive capacities in autonomously generating feasible robotic tasks. However, this new paradigm underscores the challenge of adequately evaluating these autonomously generated tasks. To address this, we propose a comprehensive evaluation framework tailored to generative simulations. Our framework segments evaluation into three core aspects: quality, diversity, and generalization. For single-task quality, we evaluate the realism of the generated task and the completeness of the generated trajectories using large language models and vision-language models. In terms of diversity, we measure both task and data diversity through text similarity of task descriptions and world model loss trained on collected task trajectories. For task-level generalization, we assess the zero-shot generalization ability on unseen tasks of a policy trained with multiple generated tasks. Experiments conducted on three representative task generation pipelines demonstrate that the results from our framework are highly consistent with human evaluations, confirming the feasibility and validity of our approach. The findings reveal that while metrics of quality and diversity can be achieved through certain methods, no single approach excels across all metrics, suggesting a need for greater focus on balancing these different metrics. Additionally, our analysis further highlights the common challenge of low generalization capability faced by current works. Our anonymous website: https://sites.google.com/view/evaltasks.

Paperid: 4872, https://arxiv.org/pdf/2410.05340.pdf

Abstract:
Generative AI has transformed the fields of Design and Manufacturing by providing efficient and automated methods for generating and modifying 3D objects. One approach involves using Large Language Models (LLMs) to generate Computer- Aided Design (CAD) scripting code, which can then be executed to render a 3D object; however, the resulting 3D object may not meet the specified requirements. Testing the correctness of CAD generated code is challenging due to the complexity and structure of 3D objects (e.g., shapes, surfaces, and dimensions) that are not feasible in code. In this paper, we introduce CADCodeVerify, a novel approach to iteratively verify and improve 3D objects generated from CAD code. Our approach works by producing ameliorative feedback by prompting a Vision-Language Model (VLM) to generate and answer a set of validation questions to verify the generated object and prompt the VLM to correct deviations. To evaluate CADCodeVerify, we introduce, CADPrompt, the first benchmark for CAD code generation, consisting of 200 natural language prompts paired with expert-annotated scripting code for 3D objects to benchmark progress. Our findings show that CADCodeVerify improves VLM performance by providing visual feedback, enhancing the structure of the 3D objects, and increasing the success rate of the compiled program. When applied to GPT-4, CADCodeVerify achieved a 7.30% reduction in Point Cloud distance and a 5.0% improvement in success rate compared to prior work

Paperid: 4873, https://arxiv.org/pdf/2409.19457.pdf

Abstract:
The language-guided robot grasping task requires a robot agent to integrate multimodal information from both visual and linguistic inputs to predict actions for target-driven grasping. While recent approaches utilizing Multimodal Large Language Models (MLLMs) have shown promising results, their extensive computation and data demands limit the feasibility of local deployment and customization. To address this, we propose a novel CLIP-based multimodal parameter-efficient tuning (PET) framework designed for three language-guided object grounding and grasping tasks: (1) Referring Expression Segmentation (RES), (2) Referring Grasp Synthesis (RGS), and (3) Referring Grasp Affordance (RGA). Our approach introduces two key innovations: a bi-directional vision-language adapter that aligns multimodal inputs for pixel-level language understanding and a depth fusion branch that incorporates geometric cues to facilitate robot grasping predictions. Experiment results demonstrate superior performance in the RES object grounding task compared with existing CLIP-based full-model tuning or PET approaches. In the RGS and RGA tasks, our model not only effectively interprets object attributes based on simple language descriptions but also shows strong potential for comprehending complex spatial reasoning scenarios, such as multiple identical objects present in the workspace. Project page: https://z.umn.edu/etog-etrg

Paperid: 4874, https://arxiv.org/pdf/2409.14096.pdf

Abstract:
In this paper, we propose VLM-Vac, a novel framework designed to enhance the autonomy of smart robot vacuum cleaners. Our approach integrates the zero-shot object detection capabilities of a Vision-Language Model (VLM) with a Knowledge Distillation (KD) strategy. By leveraging the VLM, the robot can categorize objects into actionable classes -- either to avoid or to suck -- across diverse backgrounds. However, frequently querying the VLM is computationally expensive and impractical for real-world deployment. To address this issue, we implement a KD process that gradually transfers the essential knowledge of the VLM to a smaller, more efficient model. Our real-world experiments demonstrate that this smaller model progressively learns from the VLM and requires significantly fewer queries over time. Additionally, we tackle the challenge of continual learning in dynamic home environments by exploiting a novel experience replay method based on language-guided sampling. Our results show that this approach is not only energy-efficient but also surpasses conventional vision-based clustering methods, particularly in detecting small objects across diverse backgrounds.

Paperid: 4875, https://arxiv.org/pdf/2409.14066.pdf

Abstract:
Building generalist robotic systems involves effectively endowing robots with the capabilities to handle novel objects in an open-world setting. Inspired by the advances of large pre-trained models, we propose Keypoint Affordance Learning from Imagined Environments (KALIE), which adapts pre-trained Vision Language Models (VLMs) for robotic control in a scalable manner. Instead of directly producing motor commands, KALIE controls the robot by predicting point-based affordance representations based on natural language instructions and visual observations of the scene. The VLM is trained on 2D images with affordances labeled by humans, bypassing the need for training data collected on robotic systems. Through an affordance-aware data synthesis pipeline, KALIE automatically creates massive high-quality training data based on limited example data manually collected by humans. We demonstrate that KALIE can learn to robustly solve new manipulation tasks with unseen objects given only 50 example data points. Compared to baselines using pre-trained VLMs, our approach consistently achieves superior performance.

Paperid: 4876, https://arxiv.org/pdf/2409.08381.pdf

Abstract:
Vision-language models (VLMs) like CLIP have been adapted for Multi-Label Recognition (MLR) with partial annotations by leveraging prompt-learning, where positive and negative prompts are learned for each class to associate their embeddings with class presence or absence in the shared vision-text feature space. While this approach improves MLR performance by relying on VLM priors, we hypothesize that learning negative prompts may be suboptimal, as the datasets used to train VLMs lack image-caption pairs explicitly focusing on class absence. To analyze the impact of positive and negative prompt learning on MLR, we introduce PositiveCoOp and NegativeCoOp, where only one prompt is learned with VLM guidance while the other is replaced by an embedding vector learned directly in the shared feature space without relying on the text encoder. Through empirical analysis, we observe that negative prompts degrade MLR performance, and learning only positive prompts, combined with learned negative embeddings (PositiveCoOp), outperforms dual prompt learning approaches. Moreover, we quantify the performance benefits that prompt-learning offers over a simple vision-features-only baseline, observing that the baseline displays strong performance comparable to dual prompt learning approach (DualCoOp), when the proportion of missing labels is low, while requiring half the training compute and 16 times fewer parameters

Paperid: 4877, https://arxiv.org/pdf/2601.21794.pdf

Abstract:
Large Vision-Language Models (LVLMs) are widely adopted for their strong multimodal capabilities, yet they raise serious concerns such as privacy leakage and harmful content generation. Machine unlearning has emerged as a promising solution for removing the influence of specific data from trained models. However, existing approaches largely rely on gradient-based optimization, incurring substantial computational costs for large-scale LVLMs. To address this limitation, we propose Knowledge Vector Weakening (KVW), a training-free unlearning method that directly intervenes in the full model without gradient computation. KVW identifies knowledge vectors that are activated during the model's output generation on the forget set and progressively weakens their contributions, thereby preventing the model from exploiting undesirable knowledge. Experiments on the MLLMU and CLEAR benchmarks demonstrate that KVW achieves a stable forget-retain trade-off while significantly improving computational efficiency over gradient-based and LoRA-based unlearning methods.

Paperid: 4878, https://arxiv.org/pdf/2601.14921.pdf

Abstract:
Vision-Language Models (VLMs) enable multimodal reasoning for robotic perception and interaction, but their deployment in real-world systems remains constrained by latency, limited onboard resources, and privacy risks of cloud offloading. Edge intelligence within 6G, particularly Open RAN and Multi-access Edge Computing (MEC), offers a pathway to address these challenges by bringing computation closer to the data source. This work investigates the deployment of VLMs on ORAN/MEC infrastructure using the Unitree G1 humanoid robot as an embodied testbed. We design a WebRTC-based pipeline that streams multimodal data to an edge node and evaluate LLaMA-3.2-11B-Vision-Instruct deployed at the edge versus in the cloud under real-time conditions. Our results show that edge deployment preserves near-cloud accuracy while reducing end-to-end latency by 5\%. We further evaluate Qwen2-VL-2B-Instruct, a compact model optimized for resource-constrained environments, which achieves sub-second responsiveness, cutting latency by more than half but at the cost of accuracy.

Paperid: 4879, https://arxiv.org/pdf/2601.14622.pdf

Abstract:
Language models are increasingly used for social robot navigation, yet existing benchmarks largely overlook principled prompt design for socially compliant behavior. This limitation is particularly relevant in practice, as many systems rely on small vision language models (VLMs) for efficiency. Compared to large language models, small VLMs exhibit weaker decision-making capabilities, making effective prompt design critical for accurate navigation. Inspired by cognitive theories of human learning and motivation, we study prompt design along two dimensions: system guidance (action-focused, reasoning-oriented, and perception-reasoning prompts) and motivational framing, where models compete against humans, other AI systems, or their past selves. Experiments on two socially compliant navigation datasets reveal three key findings. First, for non-finetuned GPT-4o, competition against humans achieves the best performance, while competition against other AI systems performs worst. For finetuned models, competition against the model's past self yields the strongest results, followed by competition against humans, with performance further influenced by coupling effects among prompt design, model choice, and dataset characteristics. Second, inappropriate system prompt design can significantly degrade performance, even compared to direct finetuning. Third, while direct finetuning substantially improves semantic-level metrics such as perception, prediction, and reasoning, it yields limited gains in action accuracy. In contrast, our system prompts produce a disproportionately larger improvement in action accuracy, indicating that the proposed prompt design primarily acts as a decision-level constraint rather than a representational enhancement.

Paperid: 4880, https://arxiv.org/pdf/2601.14323.pdf

Abstract:
Vision-Language-Action (VLA) models are increasingly deployed in safety-critical robotic applications, yet their security vulnerabilities remain underexplored. We identify a fundamental security flaw in modern VLA systems: the combination of action chunking and delta pose representations creates an intra-chunk visual open-loop. This mechanism forces the robot to execute K-step action sequences, allowing per-step perturbations to accumulate through integration. We propose SILENTDRIFT, a stealthy black-box backdoor attack exploiting this vulnerability. Our method employs the Smootherstep function to construct perturbations with guaranteed C2 continuity, ensuring zero velocity and acceleration at trajectory boundaries to satisfy strict kinematic consistency constraints. Furthermore, our keyframe attack strategy selectively poisons only the critical approach phase, maximizing impact while minimizing trigger exposure. The resulting poisoned trajectories are visually indistinguishable from successful demonstrations. Evaluated on the LIBERO, SILENTDRIFT achieves a 93.2% Attack Success Rate with a poisoning rate under 2%, while maintaining a 95.3% Clean Task Success Rate.

Paperid: 4881, https://arxiv.org/pdf/2601.13142.pdf

Abstract:
Recent large vision-language models (LVLMs) have demonstrated strong potential for device control. However, existing research has primarily focused on point-and-click (PnC) interaction, while remote-control (RC) interaction commonly encountered in everyday TV usage remains largely underexplored. To fill this gap, we introduce \textbf{TVWorld}, an offline graph-based abstraction of real-world TV navigation that enables reproducible and deployment-free evaluation. On this basis, we derive two complementary benchmarks that comprehensively assess TV-use capabilities: \textbf{TVWorld-N} for topology-aware navigation and \textbf{TVWorld-G} for focus-aware grounding. These benchmarks expose a key limitation of existing agents: insufficient topology awareness for focus-based, long-horizon TV navigation. Motivated by this finding, we propose a \emph{Topology-Aware Training} framework that injects topology awareness into LVLMs. Using this framework, we develop \textbf{TVTheseus}, a foundation model specialized for TV navigation. TVTheseus achieves a success rate of $68.3\%$ on TVWorld-N, surpassing strong closed-source baselines such as Gemini 3 Flash and establishing state-of-the-art (SOTA) performance. Additional analyses further provide valuable insights into the development of effective TV-use agents.

Paperid: 4882, https://arxiv.org/pdf/2601.12781.pdf

Abstract:
Referring Expression Comprehension (REC) aims to localize the image region corresponding to a natural-language query. Recent neuro-symbolic REC approaches leverage large language models (LLMs) and vision-language models (VLMs) to perform compositional reasoning, decomposing queries 4 structured programs and executing them step-by-step. While such approaches achieve interpretable reasoning and strong zero-shot generalization, they assume that intermediate reasoning steps are accurate. However, this assumption causes cascading errors: false detections and invalid relations propagate through the reasoning chain, yielding high-confidence false positives even when no target is present in the image. To address this limitation, we introduce Verification-Integrated Reasoning Operators (VIRO), a neuro-symbolic framework that embeds lightweight operator-level verifiers within reasoning steps. Each operator executes and validates its output, such as object existence or spatial relationship, thereby allowing the system to robustly handle no-target cases when verification conditions are not met. Our framework achieves state-of-the-art performance, reaching 61.1% balanced accuracy across target-present and no-target settings, and demonstrates generalization to real-world egocentric data. Furthermore, VIRO shows superior computational efficiency in terms of throughput, high reliability with a program failure rate of less than 0.3%, and scalability through decoupled program generation from execution.

Paperid: 4883, https://arxiv.org/pdf/2601.08165.pdf

Abstract:
Medical contrastive vision-language pre-training (VLP) has demonstrated significant potential in improving performance on downstream tasks. Traditional approaches typically employ contrastive learning, treating paired image-report samples as positives and unpaired ones as negatives. However, in medical datasets, there can be substantial similarities between images or reports from different patients. Rigidly treating all unpaired samples as negatives, can disrupt the underlying semantic structure and negatively impact the quality of the learned representations. In this paper, we propose a multi-level alignment framework, Representation Learning with Semantic-aware Instance and Sparse Token Alignments (SISTA) by exploiting the semantic correspondence between medical image and radiology reports at two levels, i.e., image-report and patch-word levels. Specifically, we improve the conventional contrastive learning by incorporating inter-report similarity to eliminate the false negatives and introduce a method to effectively align image patches with relevant word tokens. Experimental results demonstrate the effectiveness of the proposed framework in improving transfer performance across different datasets on three downstream tasks: image classification, image segmentation, and object detection. Notably, our framework achieves significant improvements in fine-grained tasks even with limited labeled data. Codes and pre-trained models will be made available.

Paperid: 4884, https://arxiv.org/pdf/2601.06803.pdf

Abstract:
While Chain-of-Thought empowers Large Vision-Language Models with multi-step reasoning, explicit textual rationales suffer from an information bandwidth bottleneck, where continuous visual details are discarded during discrete tokenization. Recent latent reasoning methods attempt to address this challenge, but often fall prey to premature semantic collapse due to rigid autoregressive objectives. In this paper, we propose Laser, a novel paradigm that reformulates visual deduction via Dynamic Windowed Alignment Learning (DWAL). Instead of forcing a point-wise prediction, Laser aligns the latent state with a dynamic validity window of future semantics. This mechanism enforces a "Forest-before-Trees" cognitive hierarchy, enabling the model to maintain a probabilistic superposition of global features before narrowing down to local details. Crucially, Laser maintains interpretability via decodable trajectories while stabilizing unconstrained learning via Self-Refined Superposition. Extensive experiments on 6 benchmarks demonstrate that Laser achieves state-of-the-art performance among latent reasoning methods, surpassing the strong baseline Monet by 5.03% on average. Notably, it achieves these gains with extreme efficiency, reducing inference tokens by more than 97%, while demonstrating robust generalization to out-of-distribution domains.

Paperid: 4885, https://arxiv.org/pdf/2601.06475.pdf

Abstract:
Radio astronomy is an indispensable discipline for observing distant celestial objects. Measurements of wave signals from radio telescopes, called visibility, need to be transformed into images for astronomical observations. These dirty images blend information from real sources and artifacts. Therefore, astronomers usually perform reconstruction before imaging to obtain cleaner images. Existing methods consider only a single modality of sparse visibility data, resulting in images with remaining artifacts and insufficient modeling of correlation. To enhance the extraction of visibility information and emphasize output quality in the image domain, we propose VVTRec, a multimodal radio interferometric data reconstruction method with visibility-guided visual and textual modality enrichment. In our VVTRec, sparse visibility is transformed into image-form and text-form features to obtain enhancements in terms of spatial and semantic information, improving the structural integrity and accuracy of images. Also, we leverage Vision-Language Models (VLMs) to achieve additional training-free performance improvements. VVTRec enables sparse visibility, as a foreign modality unseen by VLMs, to accurately extract pre-trained knowledge as a supplement. Our experiments demonstrate that VVTRec effectively enhances imaging results by exploiting multimodal information without introducing excessive computational overhead.

Paperid: 4886, https://arxiv.org/pdf/2601.06122.pdf

Abstract:
Visual reinforcement learning (RL) suffers from poor sample efficiency due to high-dimensional observations in complex tasks. While existing works have shown that vision-language models (VLMs) can assist RL, they often focus on knowledge distillation from the VLM to RL, overlooking the potential of RL-generated interaction data to enhance the VLM. To address this, we propose COVR, a collaborative optimization framework that enables the mutual enhancement of the VLM and RL policies. Specifically, COVR fine-tunes the VLM with RL-generated data to enhance the semantic reasoning ability consistent with the target task, and uses the enhanced VLM to further guide policy learning via action priors. To improve fine-tuning efficiency, we introduce two key modules: (1) an Exploration-Driven Dynamic Filter module that preserves valuable exploration samples using adaptive thresholds based on the degree of exploration, and (2) a Return-Aware Adaptive Loss Weight module that improves the stability of training by quantifying the inconsistency of sampling actions via return signals of RL. We further design a progressive fine-tuning strategy to reduce resource consumption. Extensive experiments show that COVR achieves strong performance across various challenging visual control tasks.

Paperid: 4887, https://arxiv.org/pdf/2601.02147.pdf

Abstract:
Vision language foundation models such as CLIP exhibit impressive zero-shot generalization yet remain vulnerable to spurious correlations across visual and textual modalities. Existing debiasing approaches often address a single modality either visual or textual leading to partial robustness and unstable adaptation under distribution shifts. We propose a bilateral prompt optimization framework (BiPrompt) that simultaneously mitigates non-causal feature reliance in both modalities during test-time adaptation. On the visual side, it employs structured attention-guided erasure to suppress background activations and enforce orthogonal prediction consistency between causal and spurious regions. On the textual side, it introduces balanced prompt normalization, a learnable re-centering mechanism that aligns class embeddings toward an isotropic semantic space. Together, these modules jointly minimize conditional mutual information between spurious cues and predictions, steering the model toward causal, domain invariant reasoning without retraining or domain supervision. Extensive evaluations on real-world and synthetic bias benchmarks demonstrate consistent improvements in both average and worst-group accuracies over prior test-time debiasing methods, establishing a lightweight yet effective path toward trustworthy and causally grounded vision-language adaptation.

Paperid: 4888, https://arxiv.org/pdf/2601.00879.pdf

Abstract:
Knee osteoarthritis (KOA) is a leading cause of disability worldwide, and accurate severity assessment using the Kellgren Lawrence (KL) grading system is critical for clinical decision making. However, radiographic distinctions between early disease stages, particularly KL1 and KL2, are subtle and frequently lead to inter-observer variability among radiologists. To address these challenges, we propose VLOrdinalFormer, a vision language guided ordinal learning framework for fully automated KOA grading from knee radiographs. The proposed method combines a ViT L16 backbone with CORAL based ordinal regression and a Contrastive Language Image Pretraining (CLIP) driven semantic alignment module, allowing the model to incorporate clinically meaningful textual concepts related to joint space narrowing, osteophyte formation, and subchondral sclerosis. To improve robustness and mitigate overfitting, we employ stratified five fold cross validation, class aware re weighting to emphasize challenging intermediate grades, and test time augmentation with global threshold optimization. Experiments conducted on the publicly available OAI kneeKL224 dataset demonstrate that VLOrdinalFormer achieves state of the art performance, outperforming CNN and ViT baselines in terms of macro F1 score and overall accuracy. Notably, the proposed framework yields substantial performance gains for KL1 and KL2 without compromising classification accuracy for mild or severe cases. In addition, interpretability analyses using Grad CAM and CLIP similarity maps confirm that the model consistently attends to clinically relevant anatomical regions. These results highlight the potential of vision language aligned ordinal transformers as reliable and interpretable tools for KOA grading and disease progression assessment in routine radiological practice.

Paperid: 4889, https://arxiv.org/pdf/2512.24210.pdf

Abstract:
Vision-language-action (VLA) models have enabled language-conditioned, long-horizon robot manipulation, but most existing systems are limited to grippers. Scaling VLA policies to bimanual robots with high degree-of-freedom (DoF) dexterous hands remains challenging due to the expanded action space, frequent hand-object occlusions, and the cost of collecting real-robot data. We present GR-Dexter, a holistic hardware-model-data framework for VLA-based generalist manipulation on a bimanual dexterous-hand robot. Our approach combines the design of a compact 21-DoF robotic hand, an intuitive bimanual teleoperation system for real-robot data collection, and a training recipe that leverages teleoperated robot trajectories together with large-scale vision-language and carefully curated cross-embodiment datasets. Across real-world evaluations spanning long-horizon everyday manipulation and generalizable pick-and-place, GR-Dexter achieves strong in-domain performance and improved robustness to unseen objects and unseen instructions. We hope GR-Dexter serves as a practical step toward generalist dexterous-hand robotic manipulation.

Paperid: 4890, https://arxiv.org/pdf/2512.23864.pdf

Abstract:
Vision-Language-Action (VLA) models have shown remarkable generalization by mapping web-scale knowledge to robotic control, yet they remain blind to physical contact. Consequently, they struggle with contact-rich manipulation tasks that require reasoning about force, texture, and slip. While some approaches incorporate low-dimensional tactile signals, they fail to capture the high-resolution dynamics essential for such interactions. To address this limitation, we introduce DreamTacVLA, a framework that grounds VLA models in contact physics by learning to feel the future. Our model adopts a hierarchical perception scheme in which high-resolution tactile images serve as micro-vision inputs coupled with wrist-camera local vision and third-person macro vision. To reconcile these multi-scale sensory streams, we first train a unified policy with a Hierarchical Spatial Alignment (HSA) loss that aligns tactile tokens with their spatial counterparts in the wrist and third-person views. To further deepen the model's understanding of fine-grained contact dynamics, we finetune the system with a tactile world model that predicts future tactile signals. To mitigate tactile data scarcity and the wear-prone nature of tactile sensors, we construct a hybrid large-scale dataset sourced from both high-fidelity digital twin and real-world experiments. By anticipating upcoming tactile states, DreamTacVLA acquires a rich model of contact physics and conditions its actions on both real observations and imagined consequences. Across contact-rich manipulation tasks, it outperforms state-of-the-art VLA baselines, achieving up to 95% success, highlighting the importance of understanding physical contact for robust, touch-aware robotic agents.

Paperid: 4891, https://arxiv.org/pdf/2512.20591.pdf

Abstract:
Contact often occurs without macroscopic surface deformation, such as during interaction with liquids, semi-liquids, or ultra-soft materials. Most existing tactile sensors rely on deformation to infer contact, making such light-contact interactions difficult to perceive robustly. To address this, we present LightTact, a visual-tactile fingertip sensor that makes contact directly visible via a deformation-independent, optics-based principle. LightTact uses an ambient-blocking optical configuration that suppresses both external light and internal illumination at non-contact regions, while transmitting only the diffuse light generated at true contacts. As a result, LightTact produces high-contrast raw images in which non-contact pixels remain near-black (mean gray value < 3) and contact pixels preserve the natural appearance of the contacting surface. Built on this, LightTact achieves accurate pixel-level contact segmentation that is robust to material properties, contact force, surface appearance, and environmental lighting. We further integrate LightTact on a robotic arm and demonstrate manipulation behaviors driven by extremely light contact, including water spreading, facial-cream dipping, and thin-film interaction. Finally, we show that LightTact's spatially aligned visual-tactile images can be directly interpreted by existing vision-language models, enabling resistor value reasoning for robotic sorting.

Paperid: 4892, https://arxiv.org/pdf/2512.17435.pdf

Abstract:
Visual navigation is a fundamental capability for autonomous home-assistance robots, enabling long-horizon tasks such as object search. While recent methods have leveraged Large Language Models (LLMs) to incorporate commonsense reasoning and improve exploration efficiency, their planning remains constrained by textual representations, which cannot adequately capture spatial occupancy or scene geometry--critical factors for navigation decisions. We explore whether Vision-Language Models (VLMs) can achieve mapless visual navigation using only onboard RGB/RGB-D streams, unlocking their potential for spatial perception and planning. We achieve this through an imagination-powered navigation framework, ImagineNav++, which imagines future observation images from candidate robot views and translates navigation planning into a simple best-view image selection problem for VLMs. First, a future-view imagination module distills human navigation preferences to generate semantically meaningful viewpoints with high exploration potential. These imagined views then serve as visual prompts for the VLM to identify the most informative viewpoint. To maintain spatial consistency, we develop a selective foveation memory mechanism, which hierarchically integrates keyframe observations via a sparse-to-dense framework, constructing a compact yet comprehensive memory for long-term spatial reasoning. This approach transforms goal-oriented navigation into a series of tractable point-goal navigation tasks. Extensive experiments on open-vocabulary object and instance navigation benchmarks show that ImagineNav++ achieves SOTA performance in mapless settings, even surpassing most map-based methods, highlighting the importance of scene imagination and memory in VLM-based spatial reasoning.

Paperid: 4893, https://arxiv.org/pdf/2512.17396.pdf

Abstract:
In this work, we introduce RadImageNet-VQA, a large-scale dataset designed to advance radiologic visual question answering (VQA) on CT and MRI exams. Existing medical VQA datasets are limited in scale, dominated by X-ray imaging or biomedical illustrations, and often prone to text-based shortcuts. RadImageNet-VQA is built from expert-curated annotations and provides 750K images paired with 7.5M question-answer samples. It covers three key tasks - abnormality detection, anatomy recognition, and pathology identification - spanning eight anatomical regions and 97 pathology categories, and supports open-ended, closed-ended, and multiple-choice questions. Extensive experiments show that state-of-the-art vision-language models still struggle with fine-grained pathology identification, particularly in open-ended settings and even after fine-tuning. Text-only analysis further reveals that model performance collapses to near-random without image inputs, confirming that RadImageNet-VQA is free from linguistic shortcuts. The full dataset and benchmark are publicly available at https://huggingface.co/datasets/raidium/RadImageNet-VQA.

Paperid: 4894, https://arxiv.org/pdf/2512.14420.pdf

Abstract:
Large vision-language models (LVLMs) have shown impressive performance across a broad range of multimodal tasks. However, robust image caption evaluation using LVLMs remains challenging, particularly under domain-shift scenarios. To address this issue, we introduce the Distribution-Aware Score Decoder (DISCODE), a novel finetuning-free method that generates robust evaluation scores better aligned with human judgments across diverse domains. The core idea behind DISCODE lies in its test-time adaptive evaluation approach, which introduces the Adaptive Test-Time (ATT) loss, leveraging a Gaussian prior distribution to improve robustness in evaluation score estimation. This loss is efficiently minimized at test time using an analytical solution that we derive. Furthermore, we introduce the Multi-domain Caption Evaluation (MCEval) benchmark, a new image captioning evaluation benchmark covering six distinct domains, designed to assess the robustness of evaluation metrics. In our experiments, we demonstrate that DISCODE achieves state-of-the-art performance as a reference-free evaluation metric across MCEval and four representative existing benchmarks.

Paperid: 4895, https://arxiv.org/pdf/2512.13043.pdf

Abstract:
Multi-turn reinforcement learning (RL) for multi-modal agents built upon vision-language models (VLMs) is hampered by sparse rewards and long-horizon credit assignment. Recent methods densify the reward by querying a teacher that provides step-level feedback, e.g., Guided Thought Reinforcement (GTR) and On-Policy Distillation, but rely on costly, often privileged models as the teacher, limiting practicality and reproducibility. We introduce GTR-Turbo, a highly efficient upgrade to GTR, which matches the performance without training or querying an expensive teacher model. Specifically, GTR-Turbo merges the weights of checkpoints produced during the ongoing RL training, and then uses this merged model as a "free" teacher to guide the subsequent RL via supervised fine-tuning or soft logit distillation. This design removes dependence on privileged VLMs (e.g., GPT or Gemini), mitigates the "entropy collapse" observed in prior work, and keeps training stable. Across diverse visual agentic tasks, GTR-Turbo improves the accuracy of the baseline model by 10-30% while reducing wall-clock training time by 50% and compute cost by 60% relative to GTR.

Paperid: 4896, https://arxiv.org/pdf/2512.07360.pdf

Abstract:
Benefiting from the inductive biases learned from large-scale datasets, open-vocabulary semantic segmentation (OVSS) leverages the power of vision-language models, such as CLIP, to achieve remarkable progress without requiring task-specific training. However, due to CLIP's pre-training nature on image-text pairs, it tends to focus on global semantic alignment, resulting in suboptimal performance when associating fine-grained visual regions with text. This leads to noisy and inconsistent predictions, particularly in local areas. We attribute this to a dispersed bias stemming from its contrastive training paradigm, which is difficult to alleviate using CLIP features alone. To address this, we propose a structure-aware feature rectification approach that incorporates instance-specific priors derived directly from the image. Specifically, we construct a region adjacency graph (RAG) based on low-level features (e.g., colour and texture) to capture local structural relationships and use it to refine CLIP features by enhancing local discrimination. Extensive experiments show that our method effectively suppresses segmentation noise, improves region-level consistency, and achieves strong performance on multiple open-vocabulary segmentation benchmarks.

Paperid: 4897, https://arxiv.org/pdf/2512.07203.pdf

Abstract:
Multimodal pre-training remains constrained by the descriptive bias of image-caption pairs, leading models to favor surface linguistic cues over grounded visual understanding. We introduce MMRPT, a masked multimodal reinforcement pre-training framework that strengthens visual reasoning in MLLMs. We are the first to incorporate reinforcement learning directly into the pre-training of large vision-language models, enabling learning signals that reward visual grounding rather than caption imitation. MMRPT constructs masked multimodal data by estimating sentence-level visual dependency via attention over visual tokens and masking highly vision-dependent segments; the model reconstructs these spans through vision-grounded reasoning guided by a semantic-visual reward. Experiments show consistent zero-shot gains across diverse benchmarks and substantially improved robustness under supervised fine-tuning, demonstrating that reinforcement-driven masked reasoning provides a more reliable and generalizable pre-training objective for multimodal models.

Paperid: 4898, https://arxiv.org/pdf/2511.20641.pdf

Abstract:
Long-tailed multi-label visual recognition poses a significant challenge, as images typically contain multiple labels with highly imbalanced class distributions, leading to biased models that favor head classes while underperforming on tail classes. Recent efforts have leveraged pre-trained vision-language models, such as CLIP, alongside long-tailed learning techniques to exploit rich visual-textual priors for improved performance. However, existing methods often derive semantic inter-class relationships directly from imbalanced datasets, resulting in unreliable correlations for tail classes due to data scarcity. Moreover, CLIP's zero-shot paradigm is optimized for single-label image-text matching, making it suboptimal for multi-label tasks. To address these issues, we propose the correlation adaptation prompt network (CAPNET), a novel end-to-end framework that explicitly models label correlations from CLIP's textual encoder. The framework incorporates a graph convolutional network for label-aware propagation and learnable soft prompts for refined embeddings. It utilizes a distribution-balanced Focal loss with class-aware re-weighting for optimized training under imbalance. Moreover, it improves generalization through test-time ensembling and realigns visual-textual modalities using parameter-efficient fine-tuning to avert overfitting on tail classes without compromising head class performance. Extensive experiments and ablation studies on benchmarks including VOC-LT, COCO-LT, and NUS-WIDE demonstrate that CAPNET achieves substantial improvements over state-of-the-art methods, validating its effectiveness for real-world long-tailed multi-label visual recognition.

Paperid: 4899, https://arxiv.org/pdf/2511.19661.pdf

Abstract:
Agentic vision-language models are increasingly trained to "think with images" by calling image operations. However, we show that high final-answer accuracy often hides unfaithful visual reasoning: models may invoke tools on irrelevant regions or ignore tool outputs entirely, yet still guess the correct answer. In this work, we first propose a faithfulness evaluation protocol that measures whether intermediate visual tool outputs (e.g., crops) actually contain the queried evidence. This reveals that recent visual agents achieve high final-answer accuracy but exhibit low rates of faithful tool-use on visual search benchmarks. We then introduce CodeV, a code-based visual agent trained with Tool-Aware Policy Optimization (TAPO). TAPO is a process-level RL framework that augments GRPO with dense rewards defined directly on visual tool inputs and outputs, rather than on chain-of-thought tokens, making supervision easier to verify and less susceptible to reward hacking. CodeV represents visual tools as executable Python code, and TAPO assigns step-wise rewards based solely on the question and tool output, encouraging both necessary and evidence-consistent tool use. In a two-stage SFT+RL pipeline, CodeV achieves competitive or superior accuracy while substantially increasing faithful tool-use rates on related visual search benchmarks. Beyond visual search, CodeV attains strong performance on a range of multimodal reasoning and math benchmarks, suggesting that explicitly supervising intermediate tool behavior is crucial for building trustworthy, agentic visual reasoning systems.

Paperid: 4900, https://arxiv.org/pdf/2511.18164.pdf

Abstract:
Deep unfolding networks (DUNs) have recently advanced concealed object segmentation (COS) by modeling segmentation as iterative foreground-background separation. However, existing DUN-based methods (RUN) inherently couple background estimation with image restoration, leading to conflicting objectives and requiring pre-defined degradation types, which are unrealistic in real-world scenarios. To address this, we propose the nested unfolding network (NUN), a unified framework for real-world COS. NUN adopts a DUN-in-DUN design, embedding a degradation-resistant unfolding network (DeRUN) within each stage of a segmentation-oriented unfolding network (SODUN). This design decouples restoration from segmentation while allowing mutual refinement. Guided by a vision-language model (VLM), DeRUN dynamically infers degradation semantics and restores high-quality images without explicit priors, whereas SODUN performs reversible estimation to refine foreground and background. Leveraging the multi-stage nature of unfolding, NUN employs image-quality assessment to select the best DeRUN outputs for subsequent stages, naturally introducing a self-consistency loss that enhances robustness. Extensive experiments show that NUN achieves a leading place on both clean and degraded benchmarks. Code will be released.

Paperid: 4901, https://arxiv.org/pdf/2511.17722.pdf

Abstract:
Recent research suggests that Vision Language Models (VLMs) often rely on inherent biases learned during training when responding to queries about visual properties of images. These biases are exacerbated when VLMs are asked highly specific questions that require them to focus on particular areas of the image in tasks such as counting. We build upon this research by developing a synthetic benchmark dataset and evaluation framework to systematically determine how counting performance varies as image and prompt properties change. Using open-source VLMs, we then analyze how attention allocation fluctuates with varying input parameters (e.g. number of objects in the image, objects color, background color, objects texture, background texture, and prompt specificity). We further implement attention-based interventions to modulate focus on visual tokens at different layers and evaluate their impact on counting performance across a range of visual conditions. Our experiments reveal that while VLM counting performance remains challenging, especially under high visual or linguistic complexity, certain attention interventions can lead to modest gains in counting performance.

Paperid: 4902, https://arxiv.org/pdf/2511.17094.pdf

Abstract:
Video anomaly detection (VAD) plays a vital role in real-world applications such as security surveillance, autonomous driving, and industrial monitoring. Recent advances in large pre-trained models have opened new opportunities for training-free VAD by leveraging rich prior knowledge and general reasoning capabilities. However, existing studies typically rely on dense frame-level inference, incurring high computational costs and latency. This raises a fundamental question: Is dense reasoning truly necessary when using powerful pre-trained models in VAD systems? To answer this, we propose ReCoVAD, a novel framework inspired by the dual reflex and conscious pathways of the human nervous system, enabling selective frame processing to reduce redundant computation. ReCoVAD consists of two core pathways: (i) a Reflex pathway that uses a lightweight CLIP-based module to fuse visual features with prototype prompts and produce decision vectors, which query a dynamic memory of past frames and anomaly scores for fast response; and (ii) a Conscious pathway that employs a medium-scale vision-language model to generate textual event descriptions and refined anomaly scores for novel frames. It continuously updates the memory and prototype prompts, while an integrated large language model periodically reviews accumulated descriptions to identify unseen anomalies, correct errors, and refine prototypes. Extensive experiments show that ReCoVAD achieves state-of-the-art training-free performance while processing only 28.55\% and 16.04\% of the frames used by previous methods on the UCF-Crime and XD-Violence datasets, demonstrating that sparse reasoning is sufficient for effective large-model-based VAD.

Paperid: 4903, https://arxiv.org/pdf/2511.14336.pdf

Abstract:
A structured understanding of intraoral 3D scans is essential for digital orthodontics. However, existing deep-learning approaches rely heavily on modality-specific training, large annotated datasets, and controlled scanning conditions, which limit generalization across devices and hinder deployment in real clinical workflows. Moreover, raw intraoral meshes exhibit substantial variation in arch pose, incomplete geometry caused by occlusion or tooth contact, and a lack of texture cues, making unified semantic interpretation highly challenging. To address these limitations, we propose ArchMap, a training-free and knowledge-guided framework for robust structured dental understanding. ArchMap first introduces a geometry-aware arch-flattening module that standardizes raw 3D meshes into spatially aligned, continuity-preserving multi-view projections. We then construct a Dental Knowledge Base (DKB) encoding hierarchical tooth ontology, dentition-stage policies, and clinical semantics to constrain the symbolic reasoning space. We validate ArchMap on 1060 pre-/post-orthodontic cases, demonstrating robust performance in tooth counting, anatomical partitioning, dentition-stage classification, and the identification of clinical conditions such as crowding, missing teeth, prosthetics, and caries. Compared with supervised pipelines and prompted VLM baselines, ArchMap achieves higher accuracy, reduced semantic drift, and superior stability under sparse or artifact-prone conditions. As a fully training-free system, ArchMap demonstrates that combining geometric normalization with ontology-guided multimodal reasoning offers a practical and scalable solution for the structured analysis of 3D intraoral scans in modern digital orthodontics.

Paperid: 4904, https://arxiv.org/pdf/2511.13478.pdf

Abstract:
Multimedia documents such as slide presentations and posters are designed to be interactive and easy to modify. Yet, they are often distributed in a static raster format, which limits editing and customization. Restoring their editability requires converting these raster images back into structured vector formats. However, existing geometric raster-vectorization methods, which rely on low-level primitives like curves and polygons, fall short at this task. Specifically, when applied to complex documents like slides, they fail to preserve the high-level structure, resulting in a flat collection of shapes where the semantic distinction between image and text elements is lost. To overcome this limitation, we address the problem of semantic document derendering by introducing SliDer, a novel framework that uses Vision-Language Models (VLMs) to derender slide images as compact and editable Scalable Vector Graphic (SVG) representations. SliDer detects and extracts attributes from individual image and text elements in a raster input and organizes them into a coherent SVG format. Crucially, the model iteratively refines its predictions during inference in a process analogous to human design, generating SVG code that more faithfully reconstructs the original raster upon rendering. Furthermore, we introduce Slide2SVG, a novel dataset comprising raster-SVG pairs of slide documents curated from real-world scientific presentations, to facilitate future research in this domain. Our results demonstrate that SliDer achieves a reconstruction LPIPS of 0.069 and is favored by human evaluators in 82.9% of cases compared to the strongest zero-shot VLM baseline.

Paperid: 4905, https://arxiv.org/pdf/2511.12525.pdf

Abstract:
Infrared and visible image fusion aims to integrate complementary multi-modal information into a single fused result. However, existing methods 1) fail to account for the degradation visible images under adverse weather conditions, thereby compromising fusion performance; and 2) rely on fixed network architectures, limiting their adaptability to diverse degradation scenarios. To address these issues, we propose a one-stop degradation-aware image fusion framework for multi-degradation scenarios driven by a large language model (MdaIF). Given the distinct scattering characteristics of different degradation scenarios (e.g., haze, rain, and snow) in atmospheric transmission, a mixture-of-experts (MoE) system is introduced to tackle image fusion across multiple degradation scenarios. To adaptively extract diverse weather-aware degradation knowledge and scene feature representations, collectively referred to as the semantic prior, we employ a pre-trained vision-language model (VLM) in our framework. Guided by the semantic prior, we propose degradation-aware channel attention module (DCAM), which employ degradation prototype decomposition to facilitate multi-modal feature interaction in channel domain. In addition, to achieve effective expert routing, the semantic prior and channel-domain modulated features are utilized to guide the MoE, enabling robust image fusion in complex degradation scenarios. Extensive experiments validate the effectiveness of our MdaIF, demonstrating superior performance over SOTA methods.

Paperid: 4906, https://arxiv.org/pdf/2511.01550.pdf

Abstract:
In this work, we introduce a multimodal analysis pipeline that leverages large foundation models in vision and language to analyze corporate social media content, with a focus on sustainability-related communication. Addressing the challenges of evolving, multimodal, and often ambiguous corporate messaging on platforms such as X (formerly Twitter), we employ an ensemble of large language models (LLMs) to annotate a large corpus of corporate tweets on their topical alignment with the 17 Sustainable Development Goals (SDGs). This approach avoids the need for costly, task-specific annotations and explores the potential of such models as ad-hoc annotators for social media data that can efficiently capture both explicit and implicit references to sustainability themes in a scalable manner. Complementing this textual analysis, we utilize vision-language models (VLMs), within a visual understanding framework that uses semantic clusters to uncover patterns in visual sustainability communication. This integrated approach reveals sectoral differences in SDG engagement, temporal trends, and associations between corporate messaging, environmental, social, governance (ESG) risks, and consumer engagement. Our methods-automatic label generation and semantic visual clustering-are broadly applicable to other domains and offer a flexible framework for large-scale social media analysis.

Paperid: 4907, https://arxiv.org/pdf/2511.01131.pdf

Abstract:
Human-interpretable predictions are essential for deploying AI in medical imaging, yet most interpretable-by-design (IBD) frameworks require concept annotations for training data, which are costly and impractical to obtain in clinical contexts. Recent attempts to bypass annotation, such as zero-shot vision-language models or concept-generation frameworks, struggle to capture domain-specific medical features, leading to poor reliability. In this paper, we propose a novel Prior-guided Concept Predictor (PCP), a weakly supervised framework that enables concept answer prediction without explicit supervision or reliance on language models. PCP leverages class-level concept priors as weak supervision and incorporates a refinement mechanism with KL divergence and entropy regularization to align predictions with clinical reasoning. Experiments on PH2 (dermoscopy) and WBCatt (hematology) show that PCP improves concept-level F1-score by over 33% compared to zero-shot baselines, while delivering competitive classification performance on four medical datasets (PH2, WBCatt, HAM10000, and CXR4) relative to fully supervised concept bottleneck models (CBMs) and V-IP.

Paperid: 4908, https://arxiv.org/pdf/2510.26641.pdf

Abstract:
Autonomous Vehicles (AVs) are transforming the future of transportation through advances in intelligent perception, decision-making, and control systems. However, their success is tied to one core capability, reliable object detection in complex and multimodal environments. While recent breakthroughs in Computer Vision (CV) and Artificial Intelligence (AI) have driven remarkable progress, the field still faces a critical challenge as knowledge remains fragmented across multimodal perception, contextual reasoning, and cooperative intelligence. This survey bridges that gap by delivering a forward-looking analysis of object detection in AVs, emphasizing emerging paradigms such as Vision-Language Models (VLMs), Large Language Models (LLMs), and Generative AI rather than re-examining outdated techniques. We begin by systematically reviewing the fundamental spectrum of AV sensors (camera, ultrasonic, LiDAR, and Radar) and their fusion strategies, highlighting not only their capabilities and limitations in dynamic driving environments but also their potential to integrate with recent advances in LLM/VLM-driven perception frameworks. Next, we introduce a structured categorization of AV datasets that moves beyond simple collections, positioning ego-vehicle, infrastructure-based, and cooperative datasets (e.g., V2V, V2I, V2X, I2I), followed by a cross-analysis of data structures and characteristics. Ultimately, we analyze cutting-edge detection methodologies, ranging from 2D and 3D pipelines to hybrid sensor fusion, with particular attention to emerging transformer-driven approaches powered by Vision Transformers (ViTs), Large and Small Language Models (SLMs), and VLMs. By synthesizing these perspectives, our survey delivers a clear roadmap of current capabilities, open challenges, and future opportunities.

Paperid: 4909, https://arxiv.org/pdf/2510.24446.pdf

Abstract:
Multimodal large language models (MLLMs) have shown impressive capabilities in vision-language tasks such as reasoning segmentation, where models generate segmentation masks based on textual queries. While prior work has primarily focused on perturbing image inputs, semantically equivalent textual paraphrases-crucial in real-world applications where users express the same intent in varied ways-remain underexplored. To address this gap, we introduce a novel adversarial paraphrasing task: generating grammatically correct paraphrases that preserve the original query meaning while degrading segmentation performance. To evaluate the quality of adversarial paraphrases, we develop a comprehensive automatic evaluation protocol validated with human studies. Furthermore, we introduce SPARTA-a black-box, sentence-level optimization method that operates in the low-dimensional semantic latent space of a text autoencoder, guided by reinforcement learning. SPARTA achieves significantly higher success rates, outperforming prior methods by up to 2x on both the ReasonSeg and LLMSeg-40k datasets. We use SPARTA and competitive baselines to assess the robustness of advanced reasoning segmentation models. We reveal that they remain vulnerable to adversarial paraphrasing-even under strict semantic and grammatical constraints. All code and data will be released publicly upon acceptance.

Paperid: 4910, https://arxiv.org/pdf/2510.22665.pdf

Abstract:
Synthetic Aperture Radar (SAR) has emerged as a crucial imaging modality due to its all-weather capabilities. While recent advancements in self-supervised learning and Masked Image Modeling (MIM) have paved the way for SAR foundation models, these approaches primarily focus on low-level visual features, often overlooking multimodal alignment and zero-shot target recognition within SAR imagery. To address this limitation, we construct SARCLIP-1M, a large-scale vision language dataset comprising over one million text-image pairs aggregated from existing datasets. We further introduce SARCLIP, the first vision language foundation model tailored for the SAR domain. Our SARCLIP model is trained using a contrastive vision language learning approach by domain transferring strategy, enabling it to bridge the gap between SAR imagery and textual descriptions. Extensive experiments on image-text retrieval and zero-shot classification tasks demonstrate the superior performance of SARCLIP in feature extraction and interpretation, significantly outperforming state-of-the-art foundation models and advancing the semantic understanding of SAR imagery. The code and datasets will be released soon.

Paperid: 4911, https://arxiv.org/pdf/2510.22571.pdf

Abstract:
Object state recognition aims to identify the specific condition of objects, such as their positional states (e.g., open or closed) and functional states (e.g., on or off). While recent Vision-Language Models (VLMs) are capable of performing a variety of multimodal tasks, it remains unclear how precisely they can identify object states. To alleviate this issue, we introduce the STAte and Transition UnderStanding Benchmark (STATUS Bench), the first benchmark for rigorously evaluating the ability of VLMs to understand subtle variations in object states in diverse situations. Specifically, STATUS Bench introduces a novel evaluation scheme that requires VLMs to perform three tasks simultaneously: object state identification (OSI), image retrieval (IR), and state change identification (SCI). These tasks are defined over our fully hand-crafted dataset involving image pairs, their corresponding object state descriptions and state change descriptions. Furthermore, we introduce a large-scale training dataset, namely STATUS Train, which consists of 13 million semi-automatically created descriptions. This dataset serves as the largest resource to facilitate further research in this area. In our experiments, we demonstrate that STATUS Bench enables rigorous consistency evaluation and reveal that current state-of-the-art VLMs still significantly struggle to capture subtle object state distinctions. Surprisingly, under the proposed rigorous evaluation scheme, most open-weight VLMs exhibited chance-level zero-shot performance. After fine-tuning on STATUS Train, Qwen2.5-VL achieved performance comparable to Gemini 2.0 Flash. These findings underscore the necessity of STATUS Bench and Train for advancing object state recognition in VLM research.

Paperid: 4912, https://arxiv.org/pdf/2510.21518.pdf

Abstract:
Language and vision-language models have shown impressive performance across a wide range of tasks, but their internal mechanisms remain only partly understood. In this work, we study how individual attention heads in text-generative models specialize in specific semantic or visual attributes. Building on an established interpretability method, we reinterpret the practice of probing intermediate activations with the final decoding layer through the lens of signal processing. This lets us analyze multiple samples in a principled way and rank attention heads based on their relevance to target concepts. Our results show consistent patterns of specialization at the head level across both unimodal and multimodal transformers. Remarkably, we find that editing as few as 1% of the heads, selected using our method, can reliably suppress or enhance targeted concepts in the model output. We validate our approach on language tasks such as question answering and toxicity mitigation, as well as vision-language tasks including image classification and captioning. Our findings highlight an interpretable and controllable structure within attention layers, offering simple tools for understanding and editing large-scale generative models.

Paperid: 4913, https://arxiv.org/pdf/2510.19574.pdf

Abstract:
As object detection models are increasingly deployed in cyber-physical systems such as autonomous vehicles (AVs) and surveillance platforms, ensuring their security against adversarial threats is essential. While prior work has explored adversarial attacks in the image domain, those attacks in the video domain remain largely unexamined, especially in the no-box setting. In this paper, we present α-Cloak, the first no-box adversarial attack on object detectors that operates entirely through the alpha channel of RGBA videos. α-Cloak exploits the alpha channel to fuse a malicious target video with a benign video, resulting in a fused video that appears innocuous to human viewers but consistently fools object detectors. Our attack requires no access to model architecture, parameters, or outputs, and introduces no perceptible artifacts. We systematically study the support for alpha channels across common video formats and playback applications, and design a fusion algorithm that ensures visual stealth and compatibility. We evaluate α-Cloak on five state-of-the-art object detectors, a vision-language model, and a multi-modal large language model (Gemini-2.0-Flash), demonstrating a 100% attack success rate across all scenarios. Our findings reveal a previously unexplored vulnerability in video-based perception systems, highlighting the urgent need for defenses that account for the alpha channel in adversarial settings.

Paperid: 4914, https://arxiv.org/pdf/2510.17633.pdf

Abstract:
Large Audio-Language Models (LALMs) are becoming essential as a powerful multimodal backbone for real-world applications. However, recent studies show that audio inputs can more easily elicit harmful responses than text, exposing new risks toward deployment. While safety alignment has made initial advances in LLMs and Large Vision-Language Models (LVLMs), we find that vanilla adaptation of these approaches to LALMs faces two key limitations: 1) LLM-based steering fails under audio input due to the large distributional gap between activations, and 2) prompt-based defenses induce over-refusals on benign-speech queries. To address these challenges, we propose Safe-Ablated Refusal Steering (SARSteer), the first inference-time defense framework for LALMs. Specifically, SARSteer leverages text-derived refusal steering to enforce rejection without manipulating audio inputs and introduces decomposed safe-space ablation to mitigate over-refusal. Extensive experiments demonstrate that SARSteer significantly improves harmful-query refusal while preserving benign responses, establishing a principled step toward safety alignment in LALMs.

Paperid: 4915, https://arxiv.org/pdf/2510.17439.pdf

Abstract:
Existing vision-language-action (VLA) models act in 3D real-world but are typically built on 2D encoders, leaving a spatial reasoning gap that limits generalization and adaptability. Recent 3D integration techniques for VLAs either require specialized sensors and transfer poorly across modalities, or inject weak cues that lack geometry and degrade vision-language alignment. In this work, we introduce FALCON (From Spatial to Action), a novel paradigm that injects rich 3D spatial tokens into the action head. FALCON leverages spatial foundation models to deliver strong geometric priors from RGB alone, and includes an Embodied Spatial Model that can optionally fuse depth, or pose for higher fidelity when available, without retraining or architectural changes. To preserve language reasoning, spatial tokens are consumed by a Spatial-Enhanced Action Head rather than being concatenated into the vision-language backbone. These designs enable FALCON to address limitations in spatial representation, modality transferability, and alignment. In comprehensive evaluations across three simulation benchmarks and eleven real-world tasks, our proposed FALCON achieves state-of-the-art performance, consistently surpasses competitive baselines, and remains robust under clutter, spatial-prompt conditioning, and variations in object scale and height.

Paperid: 4916, https://arxiv.org/pdf/2510.15841.pdf

Abstract:
Open-Vocabulary Semantic Segmentation (OVSS) assigns pixel-level labels from an open set of categories, requiring generalization to unseen and unlabelled objects. Using vision-language models (VLMs) to correlate local image patches with potential unseen object categories suffers from a lack of understanding of spatial relations of objects in a scene. To solve this problem, we introduce neuro-symbolic (NeSy) spatial reasoning in OVSS. In contrast to contemporary VLM correlation-based approaches, we propose Relational Segmentor (RelateSeg) to impose explicit spatial relational constraints by first order logic (FOL) formulated in a neural network architecture. This is the first attempt to explore NeSy spatial reasoning in OVSS. Specifically, RelateSeg automatically extracts spatial relations, e.g., , and encodes them as first-order logic formulas using our proposed pseudo categories. Each pixel learns to predict both a semantic category (e.g., "cat") and a spatial pseudo category (e.g., "right of person") simultaneously, enforcing relational constraints (e.g., a "cat" pixel must lie to the right of a "person"). Finally, these logic constraints are formulated in a deep network architecture by fuzzy logic relaxation, enabling end-to-end learning of spatial-relationally consistent segmentation. RelateSeg achieves state-of-the-art performance in terms of average mIoU across four benchmark datasets and particularly shows clear advantages on images containing multiple categories, with the cost of only introducing a single auxiliary loss function and no additional parameters, validating the effectiveness of NeSy spatial reasoning in OVSS.

Paperid: 4917, https://arxiv.org/pdf/2510.14836.pdf

Abstract:
Spatial perception and reasoning are crucial for Vision-Language-Action (VLA) models to accomplish fine-grained manipulation tasks. However, existing approaches often lack the ability to understand and reason over the essential 3D structures necessary for precise control. To address this limitation, we propose QDepth-VLA, a general framework that augments VLA models with an auxiliary depth prediction task. A dedicated depth expert is designed to predict quantized latent tokens of depth maps obtained from a VQ-VAE encoder, enabling the model to learn depth-aware representations that capture critical geometric cues. Experimental results on the simulation benchmarks and real-world tasks demonstrate that QDepth-VLA yields strong spatial reasoning and competitive performance on manipulation tasks.

Paperid: 4918, https://arxiv.org/pdf/2510.14828.pdf

Abstract:
Improving the reasoning capabilities of embodied agents is crucial for robots to complete complex human instructions in long-view manipulation tasks successfully. Despite the success of large language models and vision language models based on Supervised Fine-Tuning (SFT) in planning tasks, they continue facing challenges in performing long-horizon manipulation tasks in complex real-world environments, owing to their restricted common sense and reasoning capabilities. Considering that aligning general-purpose vision language models to robotic planning tasks via supervised fine-tuning suffers from poor generalization and insufficient physical understanding, we propose RoboGPT-R1, a two-stage fine-tuning framework for embodied planning. In this framework, supervised training acquires foundational knowledge through expert sequences, followed by RL to address the model's shortcomings in visual-spatial understanding and reasoning. To achieve physical understanding and action sequence consistency in multi-step reasoning tasks, we design a rule-based reward function that simultaneously considers long-horizon performance and action constraint in the environment. The reasoning model, trained on Qwen2.5-VL-3B, significantly outperforms the larger-scale model, GPT-4o-mini, by 21.33% and surpasses other work trained on Qwen2.5-VL-7B by 20.33% on the EmbodiedBench benchmark.

Paperid: 4919, https://arxiv.org/pdf/2510.14624.pdf

Abstract:
Vision-language models (VLMs) have recently expanded from static image understanding to video reasoning, but their scalability is fundamentally limited by the quadratic cost of processing dense frame sequences. Long videos often exceed the token budget of modern language models, leading to severe context limitations and latency issues. We introduce Efficient Video Sampling (EVS), a simple, plug-and-play method for reducing token redundancy in videos by identifying and pruning temporally static patches -- spatial regions that remain unchanged across consecutive frames. EVS preserves positional identity, requires no architectural changes or retraining. We show that EVS substantially reduces token count while maintaining semantic fidelity, enabling faster inference and longer input sequences. Applied at inference time, EVS reduces large language model (LLM) time-to-first-token (TTFT) by up to 4x with minimal accuracy loss. When combined with an uptraining phase using stochastic pruning rates, EVS yields models that are robust to varying compression levels and retain full performance under aggressive pruning. Extensive experiments demonstrate that EVS consistently improves efficiency-accuracy trade-offs, unlocking scalable video-language understanding without sacrificing quality.

Paperid: 4920, https://arxiv.org/pdf/2510.12710.pdf

Abstract:
Pre-trained Vision-Language-Action (VLA) models represent a major leap towards general-purpose robots, yet efficiently adapting them to novel, specific tasks in-situ remains a significant hurdle. While reinforcement learning (RL) is a promising avenue for such adaptation, the process often suffers from low efficiency, hindering rapid task mastery. We introduce Reflective Self-Adaptation, a framework for rapid, autonomous task adaptation without human intervention. Our framework establishes a self-improving loop where the agent learns from its own experience to enhance both strategy and execution. The core of our framework is a dual-pathway architecture that addresses the full adaptation lifecycle. First, a Failure-Driven Reflective RL pathway enables rapid learning by using the VLM's causal reasoning to automatically synthesize a targeted, dense reward function from failure analysis. This provides a focused learning signal that significantly accelerates policy exploration. However, optimizing such proxy rewards introduces a potential risk of "reward hacking," where the agent masters the reward function but fails the actual task. To counteract this, our second pathway, Success-Driven Quality-Guided SFT, grounds the policy in holistic success. It identifies and selectively imitates high-quality successful trajectories, ensuring the agent remains aligned with the ultimate task goal. This pathway is strengthened by a conditional curriculum mechanism to aid initial exploration. We conduct experiments in challenging manipulation tasks. The results demonstrate that our framework achieves faster convergence and higher final success rates compared to representative baselines. Our work presents a robust solution for creating self-improving agents that can efficiently and reliably adapt to new environments.

Paperid: 4921, https://arxiv.org/pdf/2510.09694.pdf

Abstract:
Large models (LMs) are powerful content generators, yet their open-ended nature can also introduce potential risks, such as generating harmful or biased content. Existing guardrails mostly perform post-hoc detection that may expose unsafe content before it is caught, and the latency constraints further push them toward lightweight models, limiting detection accuracy. In this work, we propose Kelp, a novel plug-in framework that enables streaming risk detection within the LM generation pipeline. Kelp leverages intermediate LM hidden states through a Streaming Latent Dynamics Head (SLD), which models the temporal evolution of risk across the generated sequence for more accurate real-time risk detection. To ensure reliable streaming moderation in real applications, we introduce an Anchored Temporal Consistency (ATC) loss to enforce monotonic harm predictions by embedding a benign-then-harmful temporal prior. Besides, for a rigorous evaluation of streaming guardrails, we also present StreamGuardBench-a model-grounded benchmark featuring on-the-fly responses from each protected model, reflecting real-world streaming scenarios in both text and vision-language tasks. Across diverse models and datasets, Kelp consistently outperforms state-of-the-art post-hoc guardrails and prior plug-in probes (15.61% higher average F1), while using only 20M parameters and adding less than 0.5 ms of per-token latency.

Paperid: 4922, https://arxiv.org/pdf/2510.07067.pdf

Abstract:
Vision Language Action (VLA) models are widely used in Embodied AI, enabling robots to interpret and execute language instructions. However, their robustness to natural language variability in real-world scenarios has not been thoroughly investigated. In this work, we present a novel systematic study of the robustness of state-of-the-art VLA models under linguistic perturbations. Specifically, we evaluate model performance under two types of instruction noise: (1) human-generated paraphrasing and (2) the addition of irrelevant context. We further categorize irrelevant contexts into two groups according to their length and their semantic and lexical proximity to robot commands. In this study, we observe consistent performance degradation as context size expands. We also demonstrate that the model can exhibit relative robustness to random context, with a performance drop within 10%, while semantically and lexically similar context of the same length can trigger a quality decline of around 50%. Human paraphrases of instructions lead to a drop of nearly 20%. To mitigate this, we propose an LLM-based filtering framework that extracts core commands from noisy inputs. Incorporating our filtering step allows models to recover up to 98.5% of their original performance under noisy conditions.

Paperid: 4923, https://arxiv.org/pdf/2510.06790.pdf

Abstract:
Models are susceptible to adversarially out-of-distribution (OOD) data despite large training-compute investments into their robustification. Zaremba et al. (2025) make progress on this problem at test time, showing LLM reasoning improves satisfaction of model specifications designed to thwart attacks, resulting in a correlation between reasoning effort and robustness to jailbreaks. However, this benefit of test compute fades when attackers are given access to gradients or multimodal inputs. We address this gap, clarifying that inference-compute offers benefits even in such cases. Our approach argues that compositional generalization, through which OOD data is understandable via its in-distribution (ID) components, enables adherence to defensive specifications on adversarially OOD inputs. Namely, we posit the Robustness from Inference Compute Hypothesis (RICH): inference-compute defenses profit as the model's training data better reflects the attacked data's components. We empirically support this hypothesis across vision language model and attack types, finding robustness gains from test-time compute if specification following on OOD data is unlocked by compositional generalization, while RL finetuning and protracted reasoning are not critical. For example, increasing emphasis on defensive specifications via prompting lowers the success rate of gradient-based multimodal attacks on VLMs robustified by adversarial pretraining, but this same intervention provides no such benefit to not-robustified models. This correlation of inference-compute's robustness benefit with base model robustness is the rich-get-richer dynamic of the RICH: attacked data components are more ID for robustified models, aiding compositional generalization to OOD data. Accordingly, we advise layering train-time and test-time defenses to obtain their synergistic benefit.

Paperid: 4924, https://arxiv.org/pdf/2510.06280.pdf

Abstract:
Vision language models (VLMs), such as CLIP and OpenCLIP, can encode and reflect stereotypical associations between medical professions and demographic attributes learned from web-scale data. We present an evaluation protocol for healthcare settings that quantifies associated biases and assesses their operational risk. Our methodology (i) defines a taxonomy spanning clinicians and allied healthcare roles (e.g., surgeon, cardiologist, dentist, nurse, pharmacist, technician), (ii) curates a profession-aware prompt suite to probe model behavior, and (iii) benchmarks demographic skew against a balanced face corpus. Empirically, we observe consistent demographic biases across multiple roles and vision models. Our work highlights the importance of bias identification in critical domains such as healthcare as AI-enabled hiring and workforce analytics can have downstream implications for equity, compliance, and patient trust.

Paperid: 4925, https://arxiv.org/pdf/2510.05109.pdf

Abstract:
Large Multimodal Models (LMMs) are inherently modular, consisting of vision and audio encoders, projectors, and large language models. Yet, they are almost always executed monolithically, which underutilizes the heterogeneous accelerators (NPUs, GPUs, DSPs) in modern SoCs and leads to high end-to-end latency. In this paper, we present NANOMIND, a hardware--software co-design inference framework for Large Multimodal Models (LMMs) that breaks large models into modular ``bricks'' (vision, language, audio, etc.) and maps each to its ideal accelerator. The key insight is that large models can be broken into modular components and scheduled to run on the most appropriate compute units. It performs module-level dynamic offloading across accelerators on unified-memory SoCs. By combining customized hardware design, system-level scheduling, and optimized low-bit computation kernels, we demonstrate our framework with a compact, battery-powered device capable of running LMMs entirely on device. This prototype functions as a self-contained intelligent assistant that requires no network connectivity, while achieving higher throughput and superior power efficiency under strict resource constraints. The design further bypasses CPU bottlenecks and reduces redundant memory usage through token-aware buffer management and module-level coordination. Our system outperforms existing implementations in resource efficiency, cutting energy consumption by 42.3\% and GPU memory usage by 11.2\%. This enables a battery-powered device to run LLaVA-OneVision with a camera for nearly half a day and LLaMA-3-8B for voice interactions up to almost 20.8 hours.

Paperid: 4926, https://arxiv.org/pdf/2510.03701.pdf

Abstract:
Referring expression comprehension (REC) aims to localize the target object described by a natural language expression. Recent advances in vision-language learning have led to significant performance improvements in REC tasks. However, localizing extremely small objects remains a considerable challenge despite its importance in real-world applications such as autonomous driving. To address this issue, we introduce a novel dataset and method for REC targeting small objects. First, we present the small object REC (SOREC) dataset, which consists of 100,000 pairs of referring expressions and corresponding bounding boxes for small objects in driving scenarios. Second, we propose the progressive-iterative zooming adapter (PIZA), an adapter module for parameter-efficient fine-tuning that enables models to progressively zoom in and localize small objects. In a series of experiments, we apply PIZA to GroundingDINO and demonstrate a significant improvement in accuracy on the SOREC dataset. Our dataset, codes and pre-trained models are publicly available on the project page.

Paperid: 4927, https://arxiv.org/pdf/2510.03584.pdf

Abstract:
Vision-language models (VLMs) have advanced video understanding, but their performance is limited by the number of input frames they can process. Existing frame sampling strategies, such as uniform or fixed-budget selection, often fail to adapt to variations in information density or task complexity, resulting in inefficiency and information loss. To address this, we present FrameOracle, a lightweight and plug-and-play module that predicts both (1) which frames are most relevant to a given query and (2) how many frames are needed. FrameOracle is trained using a four-stage curriculum, with the first three stages relying on weak proxy signals such as cross-modal similarity. In the final stage, it leverages stronger supervision from a new dataset we introduce, FrameOracle-41K, the first large-scale VideoQA collection to provide keyframe annotations specifying the minimal set of frames required to answer each question. Extensive experiments across five VLMs and six benchmarks demonstrate that FrameOracle reduces 16-frame inputs to an average of 10.4 frames without any loss in accuracy. When starting from 64-frame candidates, it reduces the input to an average of 13.9 frames while improving accuracy by 1.4%, achieving state-of-the-art efficiency-accuracy trade-offs for scalable video understanding.

Paperid: 4928, https://arxiv.org/pdf/2510.03441.pdf

Abstract:
Vision-language models (VLMs) have advanced multimodal reasoning but still face challenges in spatial reasoning for 3D scenes and complex object configurations. To address this, we introduce SpatialViLT, an enhanced VLM that integrates spatial features like depth maps, 3D coordinates, and edge maps through a multi-task learning framework. This approach enriches multimodal embeddings with spatial understanding. We propose two variants: SpatialViLT and MaskedSpatialViLT, focusing on full and masked object regions, respectively. Additionally, SpatialEnsemble combines both approaches, achieving state-of-the-art accuracy. Our models excel in spatial reasoning categories such as directional, topological, and proximity relations, as demonstrated on the challenging Visual Spatial Reasoning (VSR) dataset. This work represents a significant step in enhancing the spatial intelligence of AI systems, crucial for advanced multimodal understanding and real-world applications.

Paperid: 4929, https://arxiv.org/pdf/2509.24219.pdf

Abstract:
Robots trained via Reinforcement Learning (RL) or Imitation Learning (IL) often adapt slowly to new tasks, whereas recent Large Language Models (LLMs) and Vision-Language Models (VLMs) promise knowledge-rich planning from minimal data. Deploying LLMs/VLMs for motion planning, however, faces two key obstacles: (i) symbolic plans are rarely grounded in scene geometry and object physics, and (ii) model outputs can vary for identical prompts, undermining execution reliability. We propose ViReSkill, a framework that pairs vision-grounded replanning with a skill memory for accumulation and reuse. When a failure occurs, the replanner generates a new action sequence conditioned on the current scene, tailored to the observed state. On success, the executed plan is stored as a reusable skill and replayed in future encounters without additional calls to LLMs/VLMs. This feedback loop enables autonomous continual learning: each attempt immediately expands the skill set and stabilizes subsequent executions. We evaluate ViReSkill on simulators such as LIBERO and RLBench as well as on a physical robot. Across all settings, it consistently outperforms conventional baselines in task success rate, demonstrating robust sim-to-real generalization.

Paperid: 4930, https://arxiv.org/pdf/2509.23014.pdf

Abstract:
With the powerful reasoning capabilities of large language models (LLMs) and vision-language models (VLMs), many recent works have explored using them for decision-making. However, most of these approaches rely solely on language-based reasoning, which limits their ability to reason and make informed decisions. Recently, a promising new direction has emerged with unified multimodal models (UMMs), which support both multimodal inputs and outputs. We believe such models have greater potential for decision-making by enabling reasoning through generated visual content. To this end, we propose Uni-Plan, a planning framework built on UMMs. Within this framework, a single model simultaneously serves as the policy, dynamics model, and value function. In addition, to avoid hallucinations in dynamics predictions, we present a novel approach self-discriminated filtering, where the generative model serves as a self-discriminator to filter out invalid dynamics predictions. Experiments on long-horizon planning tasks show that Uni-Plan substantially improves success rates compared to VLM-based methods, while also showing strong data scalability, requiring no expert demonstrations and achieving better performance under the same training-data size. This work lays a foundation for future research in reasoning and decision-making with UMMs.

Paperid: 4931, https://arxiv.org/pdf/2509.22229.pdf

Abstract:
Source-Free Unsupervised Domain Adaptation (SFUDA) addresses the realistic challenge of adapting a source-trained model to a target domain without access to the source data, driven by concerns over privacy and cost. Existing SFUDA methods either exploit only the source model's predictions or fine-tune large multimodal models, yet both neglect complementary insights and the latent structure of target data. In this paper, we propose the Experts Cooperative Learning (EXCL). EXCL contains the Dual Experts framework and Retrieval-Augmentation-Interaction optimization pipeline. The Dual Experts framework places a frozen source-domain model (augmented with Conv-Adapter) and a pretrained vision-language model (with a trainable text prompt) on equal footing to mine consensus knowledge from unlabeled target samples. To effectively train these plug-in modules under purely unsupervised conditions, we introduce Retrieval-Augmented-Interaction(RAIN), a three-stage pipeline that (1) collaboratively retrieves pseudo-source and complex target samples, (2) separately fine-tunes each expert on its respective sample set, and (3) enforces learning object consistency via a shared learning result. Extensive experiments on four benchmark datasets demonstrate that our approach matches state-of-the-art performance.

Paperid: 4932, https://arxiv.org/pdf/2509.21986.pdf

Abstract:
Egocentric videos capture how humans manipulate objects and tools, providing diverse motion cues for learning object manipulation. Unlike the costly, expert-driven manual teleoperation commonly used in training Vision-Language-Action models (VLAs), egocentric videos offer a scalable alternative. However, prior studies that leverage such videos for training robot policies typically rely on auxiliary annotations, such as detailed hand-pose recordings. Consequently, it remains unclear whether VLAs can be trained directly from raw egocentric videos. In this work, we address this challenge by leveraging EgoScaler, a framework that extracts 6DoF object manipulation trajectories from egocentric videos without requiring auxiliary recordings. We apply EgoScaler to four large-scale egocentric video datasets and automatically refine noisy or incomplete trajectories, thereby constructing a new large-scale dataset for VLA pre-training. Our experiments with a state-of-the-art $Ï_0$ architecture in both simulated and real-robot environments yield three key findings: (i) pre-training on our dataset improves task success rates by over 20\% compared to training from scratch, (ii) the performance is competitive with that achieved using real-robot datasets, and (iii) combining our dataset with real-robot data yields further improvements. These results demonstrate that egocentric videos constitute a promising and scalable resource for advancing VLA research.

Paperid: 4933, https://arxiv.org/pdf/2509.21370.pdf

Abstract:
Culverts on canals such as the Erie Canal, built originally in 1825, require frequent inspections to ensure safe operation. Human inspection of culverts is challenging due to age, geometry, poor illumination, weather, and lack of easy access. We introduce VISION, an end-to-end, language-in-the-loop autonomy system that couples a web-scale vision-language model (VLM) with constrained viewpoint planning for autonomous inspection of culverts. Brief prompts to the VLM solicit open-vocabulary ROI proposals with rationales and confidences, stereo depth is fused to recover scale, and a planner -- aware of culvert constraints -- commands repositioning moves to capture targeted close-ups. Deployed on a quadruped in a culvert under the Erie Canal, VISION closes the see, decide, move, re-image loop on-board and produces high-resolution images for detailed reporting without domain-specific fine-tuning. In an external evaluation by New York Canal Corporation personnel, initial ROI proposals achieved 61.4\% agreement with subject-matter experts, and final post-re-imaging assessments reached 80\%, indicating that VISION converts tentative hypotheses into grounded, expert-aligned findings.

Paperid: 4934, https://arxiv.org/pdf/2509.21262.pdf

Abstract:
Homonyms are words with identical spelling but distinct meanings, which pose challenges for many generative models. When a homonym appears in a prompt, diffusion models may generate multiple senses of the word simultaneously, which is known as homonym duplication. This issue is further complicated by an Anglocentric bias, which includes an additional translation step before the text-to-image model pipeline. As a result, even words that are not homonymous in the original language may become homonyms and lose their meaning after translation into English. In this paper, we introduce a method for measuring duplication rates and conduct evaluations of different diffusion models using both automatic evaluation utilizing Vision-Language Models (VLM) and human evaluation. Additionally, we investigate methods to mitigate the homonym duplication problem through prompt expansion, demonstrating that this approach also effectively reduces duplication related to Anglocentric bias. The code for the automatic evaluation pipeline is publicly available.

Paperid: 4935, https://arxiv.org/pdf/2509.21079.pdf

Abstract:
Foundation models have shown remarkable capabilities in various domains, but their performance on complex, multimodal engineering problems remains largely unexplored. We introduce SoM-1K, the first large-scale multimodal benchmark dataset dedicated to evaluating foundation models on problems in the strength of materials (SoM). The dataset, which contains 1,065 annotated SoM problems, mirrors real-world engineering tasks by including both textual problem statements and schematic diagrams. Due to the limited capabilities of current foundation models in understanding complicated visual information, we propose a novel prompting strategy called Descriptions of Images (DoI), which provides rigorous expert-generated text descriptions of the visual diagrams as the context. We evaluate eight representative foundation models, including both large language models (LLMs) and vision language models (VLMs). Our results show that current foundation models struggle significantly with these engineering problems, with the best-performing model achieving only 56.6% accuracy. Interestingly, we found that LLMs, when provided with DoI, often outperform VLMs provided with visual diagrams. A detailed error analysis reveals that DoI plays a crucial role in mitigating visual misinterpretation errors, suggesting that accurate text-based descriptions can be more effective than direct image input for current foundation models. This work establishes a rigorous benchmark for engineering AI and highlights a critical need for developing more robust multimodal reasoning capabilities in foundation models, particularly in scientific and engineering contexts.

Paperid: 4936, https://arxiv.org/pdf/2509.18005.pdf

Abstract:
In recent years, multimodal learning has become essential in robotic vision and information fusion, especially for understanding human behavior in complex environments. However, current methods struggle to fully leverage the textual modality, relying on supervised pretrained models, which limits semantic extraction in unsupervised robotic environments, particularly with significant modality loss. These methods also tend to be computationally intensive, leading to high resource consumption in real-world applications. To address these challenges, we propose the Multi Modal Mamba Enhanced Transformer (M3ET), a lightweight model designed for efficient multimodal learning, particularly on mobile platforms. By incorporating the Mamba module and a semantic-based adaptive attention mechanism, M3ET optimizes feature fusion, alignment, and modality reconstruction. Our experiments show that M3ET improves cross-task performance, with a 2.3 times increase in pretraining inference speed. In particular, the core VQA task accuracy of M3ET remains at 0.74, while the model's parameter count is reduced by 0.67. Although performance on the EQA task is limited, M3ET's lightweight design makes it well suited for deployment on resource-constrained robotic platforms.

Paperid: 4937, https://arxiv.org/pdf/2509.16452.pdf

Abstract:
Accurate vision-based action recognition is crucial for developing autonomous robots that can operate safely and reliably in complex, real-world environments. In this work, we advance video-based recognition of indoor daily actions for robotic perception by leveraging vision-language models (VLMs) enriched with domain-specific knowledge. We adapt a prompt-learning framework in which class-level textual descriptions of each action are embedded as learnable prompts into a frozen pre-trained VLM backbone. Several strategies for structuring and encoding these textual descriptions are designed and evaluated. Experiments on the ETRI-Activity3D dataset demonstrate that our method, using only RGB video inputs at test time, achieves over 95\% accuracy and outperforms state-of-the-art approaches. These results highlight the effectiveness of knowledge-augmented prompts in enabling robust action recognition with minimal supervision.

Paperid: 4938, https://arxiv.org/pdf/2509.14571.pdf

Abstract:
Vision-language (VL) models have shown transformative potential across various critical domains due to their capability to comprehend multi-modal information. However, their performance frequently degrades under distribution shifts, making it crucial to assess and improve robustness against real-world data corruption encountered in practical applications. While advancements in VL benchmark datasets and data augmentation (DA) have contributed to robustness evaluation and improvement, there remain challenges due to a lack of in-depth comprehension of model behavior as well as the need for expertise and iterative efforts to explore data patterns. Given the achievement of visualization in explaining complex models and exploring large-scale data, understanding the impact of various data corruption on VL models aligns naturally with a visual analytics approach. To address these challenges, we introduce VisMoDAl, a visual analytics framework designed to evaluate VL model robustness against various corruption types and identify underperformed samples to guide the development of effective DA strategies. Grounded in the literature review and expert discussions, VisMoDAl supports multi-level analysis, ranging from examining performance under specific corruptions to task-driven inspection of model behavior and corresponding data slice. Unlike conventional works, VisMoDAl enables users to reason about the effects of corruption on VL models, facilitating both model behavior understanding and DA strategy formulation. The utility of our system is demonstrated through case studies and quantitative evaluations focused on corruption robustness in the image captioning task.

Paperid: 4939, https://arxiv.org/pdf/2509.12367.pdf

Abstract:
We envision an integrated process for developing lunar construction equipment, where physical design and control are explored in parallel. In this paper, we describe a technical framework that supports this process. It relies on OpenPLX, a readable/writable declarative language that links CAD-models and autonomous systems to high-fidelity, real-time 3D simulations of contacting multibody dynamics, machine regolith interaction forces, and non-ideal sensors. To demonstrate its capabilities, we present two case studies, including an autonomous lunar rover that combines a vision-language model for navigation with a reinforcement learning-based control policy for locomotion.

Paperid: 4940, https://arxiv.org/pdf/2509.11548.pdf

Abstract:
Graphical user interface (GUI) grounding is a fundamental task for building GUI agents. However, general vision-language models (VLMs) struggle with this task due to a lack of specific optimization. We identify a key gap in this paper: while VLMs exhibit significant latent grounding potential, as demonstrated by their performance measured by Pointing Game, they underperform when tasked with outputting explicit coordinates. To address this discrepancy, and bypass the high data and annotation costs of current fine-tuning approaches, we propose three zero-shot auxiliary reasoning methods. By providing explicit spatial cues such as axes, grids and labeled intersections as part of the input image, these methods enable VLMs to articulate their implicit spatial understanding capabilities. We evaluate these methods on four GUI grounding benchmarks across seven open-source and proprietary VLMs. The evaluation results demonstrate that the proposed methods substantially improve the performance of GUI grounding.

Paperid: 4941, https://arxiv.org/pdf/2509.11514.pdf

Abstract:
During spontaneous conversations, speakers collaborate on novel referring expressions, which they can then re-use in subsequent conversations. Understanding such referring expressions is an important ability for an embodied agent, so that it can carry out tasks in the real world. This requires integrating and understanding language, vision, and conversational interaction. We study the capabilities of seven state-of-the-art Large Vision Language Models (LVLMs) as overhearers to a corpus of spontaneous conversations between pairs of human discourse participants engaged in a collaborative object-matching task. We find that such a task remains challenging for current LVLMs and they all fail to show a consistent performance improvement as they overhear more conversations from the same discourse participants repeating the same task for multiple rounds. We release our corpus and code for reproducibility and to facilitate future research.

Paperid: 4942, https://arxiv.org/pdf/2509.08570.pdf

Abstract:
Multimodal models have achieved remarkable success in natural image segmentation, yet they often underperform when applied to the medical domain. Through extensive study, we attribute this performance gap to the challenges of multimodal fusion, primarily the significant semantic gap between abstract textual prompts and fine-grained medical visual features, as well as the resulting feature dispersion. To address these issues, we revisit the problem from the perspective of semantic aggregation. Specifically, we propose an Expectation-Maximization (EM) Aggregation mechanism and a Text-Guided Pixel Decoder. The former mitigates feature dispersion by dynamically clustering features into compact semantic centers to enhance cross-modal correspondence. The latter is designed to bridge the semantic gap by leveraging domain-invariant textual knowledge to effectively guide deep visual representations. The synergy between these two mechanisms significantly improves the model's generalization ability. Extensive experiments on public cardiac and fundus datasets demonstrate that our method consistently outperforms existing SOTA approaches across multiple domain generalization benchmarks.

Paperid: 4943, https://arxiv.org/pdf/2509.08266.pdf

Abstract:
Recent research on Vision Language Models (VLMs) suggests that they rely on inherent biases learned during training to respond to questions about visual properties of an image. These biases are exacerbated when VLMs are asked highly specific questions that require focusing on specific areas of the image. For example, a VLM tasked with counting stars on a modified American flag (e.g., with more than 50 stars) will often disregard the visual evidence and fail to answer accurately. We build upon this research and develop a multi-dimensional examination framework to systematically determine which characteristics of the input data, including both the image and the accompanying prompt, lead to such differences in performance. Using open-source VLMs, we further examine how attention values fluctuate with varying input parameters (e.g., image size, number of objects in the image, background color, prompt specificity). This research aims to learn how the behavior of vision language models changes and to explore methods for characterizing such changes. Our results suggest, among other things, that even minor modifications in image characteristics and prompt specificity can lead to large changes in how a VLM formulates its answer and, subsequently, its overall performance.

Paperid: 4944, https://arxiv.org/pdf/2509.04772.pdf

Abstract:
Timely and accurate floodwater depth estimation is critical for road accessibility and emergency response. While recent computer vision methods have enabled flood detection, they suffer from both accuracy limitations and poor generalization due to dependence on fixed object detectors and task-specific training. To enable accurate depth estimation that can generalize across diverse flood scenarios, this paper presents FloodVision, a zero-shot framework that combines the semantic reasoning abilities of the foundation vision-language model GPT-4o with a structured domain knowledge graph. The knowledge graph encodes canonical real-world dimensions for common urban objects including vehicles, people, and infrastructure elements to ground the model's reasoning in physical reality. FloodVision dynamically identifies visible reference objects in RGB images, retrieves verified heights from the knowledge graph to mitigate hallucination, estimates submergence ratios, and applies statistical outlier filtering to compute final depth values. Evaluated on 110 crowdsourced images from MyCoast New York, FloodVision achieves a mean absolute error of 8.17 cm, reducing the GPT-4o baseline 10.28 cm by 20.5% and surpassing prior CNN-based methods. The system generalizes well across varying scenes and operates in near real-time, making it suitable for future integration into digital twin platforms and citizen-reporting apps for smart city flood resilience.

Paperid: 4945, https://arxiv.org/pdf/2509.04018.pdf

Abstract:
Robotic manipulation is a fundamental component of automation. However, traditional perception-planning pipelines often fall short in open-ended tasks due to limited flexibility, while the architecture of a single end-to-end Vision-Language-Action (VLA) offers promising capabilities but lacks crucial mechanisms for anticipating and recovering from failure. To address these challenges, we propose FPC-VLA, a dual-model framework that integrates VLA with a supervisor for failure prediction and correction. The supervisor evaluates action viability through vision-language queries and generates corrective strategies when risks arise, trained efficiently without manual labeling. A similarity-guided fusion module further refines actions by leveraging past predictions. Evaluation results on multiple simulation platforms (SIMPLER and LIBERO) and robot embodiments (WidowX, Google Robot, Franka) show that FPC-VLA outperforms state-of-the-art models in both zero-shot and fine-tuned settings. By activating the supervisor only at keyframes, our approach significantly increases task success rates with minimal impact on execution time. Successful real-world deployments on diverse, long-horizon tasks confirm FPC-VLA's strong generalization and practical utility for building more reliable autonomous systems.

Paperid: 4946, https://arxiv.org/pdf/2509.03895.pdf

Abstract:
Contrastive vision-language models excel in zero-shot image recognition but face challenges in few-shot scenarios due to computationally intensive offline fine-tuning using prompt learning, which risks overfitting. To overcome these limitations, we propose Attn-Adapter, a novel online few-shot learning framework that enhances CLIP's adaptability via a dual attention mechanism. Our design incorporates dataset-specific information through two components: the Memory Attn-Adapter, which refines category embeddings using support examples, and the Local-Global Attn-Adapter, which enriches image embeddings by integrating local and global features. This architecture enables dynamic adaptation from a few labeled samples without retraining the base model. Attn-Adapter outperforms state-of-the-art methods in cross-category and cross-dataset generalization, maintaining efficient inference and scaling across CLIP backbones.

Paperid: 4947, https://arxiv.org/pdf/2509.03803.pdf

Abstract:
Prompt learning has recently attracted much attention for adapting pre-trained vision-language models (e.g., CLIP) to downstream recognition tasks. However, most of the existing CLIP-based prompt learning methods only show a limited ability for handling fine-grained datasets. To address this issue, we propose a causality-guided text prompt learning method via visual granulation for CLIP, called CaPL, where the explored visual granulation technique could construct sets of visual granules for the text prompt to capture subtle discrepancies among different fine-grained classes through casual inference. The CaPL method contains the following two modules: (1) An attribute disentanglement module is proposed to decompose visual features into non-individualized attributes (shared by some classes) and individualized attributes (specific to single classes) using a Brownian Bridge Diffusion Model; (2) A granule learning module is proposed to construct visual granules by integrating the aforementioned attributes for recognition under two causal inference strategies. Thanks to the learned visual granules, more discriminative text prompt is expected to be learned. Extensive experimental results on 15 datasets demonstrate that our CaPL method significantly outperforms the state-of-the-art prompt learning methods, especially on fine-grained datasets.

Paperid: 4948, https://arxiv.org/pdf/2509.00700.pdf

Abstract:
Vision-Language Models (VLMs) combine a vision encoder and a large language model (LLM) through alignment training, showing strong performance on multimodal tasks. A central component in this architecture is the projection layer, which maps visual features into the LLM's embedding space. Despite its importance, its ability to generalize to unseen visual concepts has not been systematically evaluated. To address this, we propose a benchmark for evaluating projection-layer generalization. We adapt object detection datasets (rich in fine-grained annotations) into a prompting format and design train/test splits with disjoint label sets, enabling precise control over seen and unseen concept separation. Experimental results show that the projection layer retains about 79 to 88 percent of the performance on unseen classes compared to seen ones across various settings, suggesting a non-trivial level of generalization even without explicit alignment supervision on those concepts. We further analyze this behavior through a mechanistic interpretability lens. Our findings indicate that the feed-forward network in the projection layer functions like a key-value memory, processing seen and unseen tokens in similar ways. This study introduces a new evaluation framework for alignment generalization and highlights the potential for efficient VLM training with limited aligned data.

Paperid: 4949, https://arxiv.org/pdf/2508.20227.pdf

Abstract:
The development of many vision models mainly focuses on improving their performance using metrics such as accuracy, IoU, and mAP, with less attention to explainability due to the complexity of applying xAI methods to provide a meaningful explanation of trained models. Although many existing xAI methods aim to explain vision models sample-by-sample, methods explaining the general behavior of vision models, which can only be captured after running on a large dataset, are still underexplored. Furthermore, understanding the behavior of vision models on general images can be very important to prevent biased judgments and help identify the model's trends and patterns. With the application of Vision-Language Models, this paper proposes a pipeline to explain vision models at both the sample and dataset levels. The proposed pipeline can be used to discover failure cases and gain insights into vision models with minimal effort, thereby integrating vision model development with xAI analysis to advance image analysis.

Paperid: 4950, https://arxiv.org/pdf/2508.20018.pdf

Abstract:
The rapid advancement of large vision language models (LVLMs) and agent systems has heightened interest in mobile GUI agents that can reliably translate natural language into interface operations. Existing single-agent approaches, however, remain limited by structural constraints. Although multi-agent systems naturally decouple different competencies, recent progress in multi-agent reinforcement learning (MARL) has often been hindered by inefficiency and remains incompatible with current LVLM architectures. To address these challenges, we introduce SWIRL, a staged workflow for interleaved reinforcement learning designed for multi-agent systems. SWIRL reformulates MARL into a sequence of single-agent reinforcement learning tasks, updating one agent at a time while keeping the others fixed. This formulation enables stable training and promotes efficient coordination across agents. Theoretically, we provide a stepwise safety bound, a cross-round monotonic improvement theorem, and convergence guarantees on return, ensuring robust and principled optimization. In application to mobile GUI control, SWIRL instantiates a Navigator that converts language and screen context into structured plans, and an Interactor that grounds these plans into executable atomic actions. Extensive experiments demonstrate superior performance on both high-level and low-level GUI benchmarks. Beyond GUI tasks, SWIRL also demonstrates strong capability in multi-agent mathematical reasoning, underscoring its potential as a general framework for developing efficient and robust multi-agent systems.

Paperid: 4951, https://arxiv.org/pdf/2508.20013.pdf

Abstract:
This study addresses critical industrial challenges in e-commerce product categorization, namely platform heterogeneity and the structural limitations of existing taxonomies, by developing and deploying a multimodal hierarchical classification framework. Using a dataset of 271,700 products from 40 international fashion e-commerce platforms, we integrate textual features (RoBERTa), visual features (ViT), and joint vision--language representations (CLIP). We investigate fusion strategies, including early, late, and attention-based fusion within a hierarchical architecture enhanced by dynamic masking to ensure taxonomic consistency. Results show that CLIP embeddings combined via an MLP-based late-fusion strategy achieve the highest hierarchical F1 (98.59\%), outperforming unimodal baselines. To address shallow or inconsistent categories, we further introduce a self-supervised ``product recategorization'' pipeline using SimCLR, UMAP, and cascade clustering, which discovered new, fine-grained categories (e.g., subtypes of ``Shoes'') with cluster purities above 86\%. Cross-platform experiments reveal a deployment-relevant trade-off: complex late-fusion methods maximize accuracy with diverse training data, while simpler early-fusion methods generalize more effectively to unseen platforms. Finally, we demonstrate the framework's industrial scalability through deployment in EURWEB's commercial transaction intelligence platform via a two-stage inference pipeline, combining a lightweight RoBERTa stage with a GPU--accelerated multimodal stage to balance cost and accuracy.

Paperid: 4952, https://arxiv.org/pdf/2508.19257.pdf

Abstract:
Vision-Language-Action (VLA) models process visual inputs independently at each timestep, discarding valuable temporal information inherent in robotic manipulation tasks. This frame-by-frame processing makes models vulnerable to visual noise while ignoring the substantial coherence between consecutive frames in manipulation sequences. We propose Temporal Token Fusion (TTF), a training-free approach that intelligently integrates historical and current visual representations to enhance VLA inference quality. Our method employs dual-dimension detection combining efficient grayscale pixel difference analysis with attention-based semantic relevance assessment, enabling selective temporal token fusion through hard fusion strategies and keyframe anchoring to prevent error accumulation. Comprehensive experiments across LIBERO, SimplerEnv, and real robot tasks demonstrate consistent improvements: 4.0 percentage points average on LIBERO (72.4\% vs 68.4\% baseline), cross-environment validation on SimplerEnv (4.8\% relative improvement), and 8.7\% relative improvement on real robot tasks. Our approach proves model-agnostic, working across OpenVLA and VLA-Cache architectures. Notably, TTF reveals that selective Query matrix reuse in attention mechanisms enhances rather than compromises performance, suggesting promising directions for direct KQV matrix reuse strategies that achieve computational acceleration while improving task success rates.

Paperid: 4953, https://arxiv.org/pdf/2508.17102.pdf

Abstract:
Geospatial pixel reasoning is a nascent remote-sensing task that aims to generate segmentation masks directly from natural-language instructions. Prevailing MLLM-based systems co-train a language model and a mask decoder with dense pixel supervision, which is expensive and often weak on out-of-domain (OOD) data. We introduce GRASP, a structured policy-learning framework. In our design, a multimodal large language model first emits task-relevant bounding boxes and positive points from a vision-language instruction. These outputs are then passed to a pre-trained segmentation model, which consumes them as prompts to generate the final mask. Instead of supervised fine-tuning, we optimize the system purely with reinforcement learning: the model is trained solely with GRPO, guided by format rewards and accuracy rewards computed on boxes and points (no mask supervision). This leverages strong priors in foundation models, minimizes trainable parameters, and enables learning from inexpensive annotations. We additionally curate GRASP-1k, which contains reasoning-intensive queries, detailed reasoning traces, and fine-grained segmentation annotations. Evaluations on both in-domain and out-of-domain test sets show state-of-the-art results: about 4% improvement in-domain and up to 54% on OOD benchmarks. The experiment results evidence our model's robust generalization and demonstrate that complex geospatial segmentation behaviors can be learned via RL from weak spatial cues. Code and the dataset will be released open-source.

Paperid: 4954, https://arxiv.org/pdf/2508.13680.pdf

Abstract:
Vision language models (VLMs) demonstrate remarkable capabilities on English multimodal tasks, but their performance on low-resource languages with genuinely multimodal educational content remains largely unexplored. In this work, we test how VLMs perform on Vietnamese educational assessments, investigating whether VLMs trained predominantly on English data can handle real-world cross-lingual multimodal reasoning. Our work presents the first comprehensive evaluation of VLM capabilities on multimodal Vietnamese exams through proposing ViExam, a benchmark containing 2,548 multimodal questions. We find that state-of-the-art VLMs achieve only 57.74% while open-source models achieve 27.70% mean accuracy across 7 academic domains, including Mathematics, Physics, Chemistry, Biology, Geography, Driving Test, and IQ Test. Most VLMs underperform average human test-takers (66.54%), with only the thinking VLM o3 (74.07%) exceeding human average performance, yet still falling substantially short of human best performance (99.60%). Cross-lingual prompting with English instructions while maintaining Vietnamese content fails to improve performance, decreasing accuracy by 1 percentage point for SOTA VLMs. Human-in-the-loop collaboration can partially improve VLM performance by 5 percentage points. Code and data are available at: https://vi-exam.github.io.

Paperid: 4955, https://arxiv.org/pdf/2508.12399.pdf

Abstract:
Prompt learning has propelled vision-language models like CLIP to excel in diverse tasks, making them ideal for federated learning due to computational efficiency. However, conventional approaches that rely solely on final-layer features miss out on rich multi-scale visual cues and domain-specific style variations in decentralized client data. To bridge this gap, we introduce FedCSAP (Federated Cross-Modal Style-Aware Prompt Generation). Our framework harnesses low, mid, and high-level features from CLIP's vision encoder alongside client-specific style indicators derived from batch-level statistics. By merging intricate visual details with textual context, FedCSAP produces robust, context-aware prompt tokens that are both distinct and non-redundant, thereby boosting generalization across seen and unseen classes. Operating within a federated learning paradigm, our approach ensures data privacy through local training and global aggregation, adeptly handling non-IID class distributions and diverse domain-specific styles. Comprehensive experiments on multiple image classification datasets confirm that FedCSAP outperforms existing federated prompt learning methods in both accuracy and overall generalization.

Paperid: 4956, https://arxiv.org/pdf/2508.12026.pdf

Abstract:
Bongard Problems (BPs) provide a challenging testbed for abstract visual reasoning (AVR), requiring models to identify visual concepts fromjust a few examples and describe them in natural language. Early BP benchmarks featured synthetic black-and-white drawings, which might not fully capture the complexity of real-world scenes. Subsequent BP datasets employed real-world images, albeit the represented concepts are identifiable from high-level image features, reducing the task complexity. Differently, the recently released Bongard-RWR dataset aimed at representing abstract concepts formulated in the original BPs using fine-grained real-world images. Its manual construction, however, limited the dataset size to just $60$ instances, constraining evaluation robustness. In this work, we introduce Bongard-RWR+, a BP dataset composed of $5\,400$ instances that represent original BP abstract concepts using real-world-like images generated via a vision language model (VLM) pipeline. Building on Bongard-RWR, we employ Pixtral-12B to describe manually curated images and generate new descriptions aligned with the underlying concepts, use Flux.1-dev to synthesize images from these descriptions, and manually verify that the generated images faithfully reflect the intended concepts. We evaluate state-of-the-art VLMs across diverse BP formulations, including binary and multiclass classification, as well as textual answer generation. Our findings reveal that while VLMs can recognize coarse-grained visual concepts, they consistently struggle with discerning fine-grained concepts, highlighting limitations in their reasoning capabilities.

Paperid: 4957, https://arxiv.org/pdf/2508.10287.pdf

Abstract:
Recent advances in Vision-Language Models (VLMs) and large language models (LLMs) have greatly enhanced visual reasoning, a key capability for embodied AI agents like robots. However, existing visual reasoning benchmarks often suffer from several limitations: they lack a clear definition of reasoning complexity, offer have no control to generate questions over varying difficulty and task customization, and fail to provide structured, step-by-step reasoning annotations (workflows). To bridge these gaps, we formalize reasoning complexity, introduce an adaptive query engine that generates customizable questions of varying complexity with detailed intermediate annotations, and extend the JRDB dataset with human-object interaction and geometric relationship annotations to create JRDB-Reasoning, a benchmark tailored for visual reasoning in human-crowded environments. Our engine and benchmark enable fine-grained evaluation of visual reasoning frameworks and dynamic assessment of visual-language models across reasoning levels.

Paperid: 4958, https://arxiv.org/pdf/2508.09524.pdf

Abstract:
In this paper, we present the first systematic investigation and quantification of Similar Object Interference (SOI), a long-overlooked yet critical bottleneck in Single Object Tracking (SOT). Through controlled Online Interference Masking (OIM) experiments, we quantitatively demonstrate that eliminating interference sources leads to substantial performance improvements (AUC gains up to 4.35) across all SOTA trackers, directly validating SOI as a primary constraint for robust tracking and highlighting the feasibility of external cognitive guidance. Building upon these insights, we adopt natural language as a practical form of external guidance, and construct SOIBench-the first semantic cognitive guidance benchmark specifically targeting SOI challenges. It automatically mines SOI frames through multi-tracker collective judgment and introduces a multi-level annotation protocol to generate precise semantic guidance texts. Systematic evaluation on SOIBench reveals a striking finding: existing vision-language tracking (VLT) methods fail to effectively exploit semantic cognitive guidance, achieving only marginal improvements or even performance degradation (AUC changes of -0.26 to +0.71). In contrast, we propose a novel paradigm employing large-scale vision-language models (VLM) as external cognitive engines that can be seamlessly integrated into arbitrary RGB trackers. This approach demonstrates substantial improvements under semantic cognitive guidance (AUC gains up to 0.93), representing a significant advancement over existing VLT methods. We hope SOIBench will serve as a standardized evaluation platform to advance semantic cognitive tracking research and contribute new insights to the tracking research community.

Paperid: 4959, https://arxiv.org/pdf/2508.07570.pdf

Abstract:
Vision-language models (VLMs) exhibit remarkable zero-shot generalization but suffer performance degradation under distribution shifts in downstream tasks, particularly in the absence of labeled data. Test-Time Adaptation (TTA) addresses this challenge by enabling online optimization of VLMs during inference, eliminating the need for annotated data. Cache-based TTA methods exploit historical knowledge by maintaining a dynamic memory cache of low-entropy or high-confidence samples, promoting efficient adaptation to out-of-distribution data. Nevertheless, these methods face two critical challenges: (1) unreliable confidence metrics under significant distribution shifts, resulting in error accumulation within the cache and degraded adaptation performance; and (2) rigid decision boundaries that fail to accommodate substantial distributional variations, leading to suboptimal predictions. To overcome these limitations, we introduce the Adaptive Cache Enhancement (ACE) framework, which constructs a robust cache by selectively storing high-confidence or low-entropy image embeddings per class, guided by dynamic, class-specific thresholds initialized from zero-shot statistics and iteratively refined using an exponential moving average and exploration-augmented updates. This approach enables adaptive, class-wise decision boundaries, ensuring robust and accurate predictions across diverse visual distributions. Extensive experiments on 15 diverse benchmark datasets demonstrate that ACE achieves state-of-the-art performance, delivering superior robustness and generalization compared to existing TTA methods in challenging out-of-distribution scenarios.

Paperid: 4960, https://arxiv.org/pdf/2508.06555.pdf

Abstract:
The advancement of intelligent agents has revolutionized problem-solving across diverse domains, yet solutions for personalized fashion styling remain underexplored, which holds immense promise for promoting shopping experiences. In this work, we present StyleTailor, the first collaborative agent framework that seamlessly unifies personalized apparel design, shopping recommendation, virtual try-on, and systematic evaluation into a cohesive workflow. To this end, StyleTailor pioneers an iterative visual refinement paradigm driven by multi-level negative feedback, enabling adaptive and precise user alignment. Specifically, our framework features two core agents, i.e., Designer for personalized garment selection and Consultant for virtual try-on, whose outputs are progressively refined via hierarchical vision-language model feedback spanning individual items, complete outfits, and try-on efficacy. Counterexamples are aggregated into negative prompts, forming a closed-loop mechanism that enhances recommendation quality. To assess the performance, we introduce a comprehensive evaluation suite encompassing style consistency, visual quality, face similarity, and artistic appraisal. Extensive experiments demonstrate StyleTailor's superior performance in delivering personalized designs and recommendations, outperforming strong baselines without negative feedback and establishing a new benchmark for intelligent fashion systems.

Paperid: 4961, https://arxiv.org/pdf/2508.04942.pdf

Abstract:
Vision-language models (VLMs) like CLIP excel in zero-shot learning but often require resource-intensive training to adapt to new tasks. Prompt learning techniques, such as CoOp and CoCoOp, offer efficient adaptation but tend to overfit to known classes, limiting generalization to unseen categories. We introduce ProMIM, a plug-and-play framework that enhances conditional prompt learning by integrating masked image modeling (MIM) into existing VLM pipelines. ProMIM leverages a simple yet effective masking strategy to generate robust, instance-conditioned prompts, seamlessly augmenting methods like CoOp and CoCoOp without altering their core architectures. By masking only visible image patches and using these representations to guide prompt generation, ProMIM improves feature robustness and mitigates overfitting, all while introducing negligible additional computational cost. Extensive experiments across zero-shot and few-shot classification tasks demonstrate that ProMIM consistently boosts generalization performance when plugged into existing approaches, providing a practical, lightweight solution for real-world vision-language applications.

Paperid: 4962, https://arxiv.org/pdf/2508.04097.pdf

Abstract:
Model inversion (MI) attacks pose significant privacy risks by reconstructing private training data from trained neural networks. While prior works have focused on conventional unimodal DNNs, the vulnerability of vision-language models (VLMs) remains underexplored. In this paper, we conduct the first study to understand VLMs' vulnerability in leaking private visual training data. To tailored for VLMs' token-based generative nature, we propose a suite of novel token-based and sequence-based model inversion strategies. Particularly, we propose Token-based Model Inversion (TMI), Convergent Token-based Model Inversion (TMI-C), Sequence-based Model Inversion (SMI), and Sequence-based Model Inversion with Adaptive Token Weighting (SMI-AW). Through extensive experiments and user study on three state-of-the-art VLMs and multiple datasets, we demonstrate, for the first time, that VLMs are susceptible to training data leakage. The experiments show that our proposed sequence-based methods, particularly SMI-AW combined with a logit-maximization loss based on vocabulary representation, can achieve competitive reconstruction and outperform token-based methods in attack accuracy and visual similarity. Importantly, human evaluation of the reconstructed images yields an attack accuracy of 75.31\%, underscoring the severity of model inversion threats in VLMs. Notably we also demonstrate inversion attacks on the publicly released VLMs. Our study reveals the privacy vulnerability of VLMs as they become increasingly popular across many applications such as healthcare and finance.

Paperid: 4963, https://arxiv.org/pdf/2508.02219.pdf

Abstract:
Vision-Language-Action (VLA) models demonstrate significant potential for developing generalized policies in real-world robotic control. This progress inspires researchers to explore fine-tuning these models with Reinforcement Learning (RL). However, fine-tuning VLA models with RL still faces challenges related to sample efficiency, compatibility with action chunking, and training stability. To address these challenges, we explore the fine-tuning of VLA models through offline reinforcement learning incorporating action chunking. In this work, we propose Chunked RL, a novel reinforcement learning framework specifically designed for VLA models. Within this framework, we extend temporal difference (TD) learning to incorporate action chunking, a prominent characteristic of VLA models. Building upon this framework, we propose CO-RFT, an algorithm aimed at fine-tuning VLA models using a limited set of demonstrations (30 to 60 samples). Specifically, we first conduct imitation learning (IL) with full parameter fine-tuning to initialize both the backbone and the policy. Subsequently, we implement offline RL with action chunking to optimize the pretrained policy. Our empirical results in real-world environments demonstrate that CO-RFT outperforms previous supervised methods, achieving a 57% improvement in success rate and a 22.3% reduction in cycle time. Moreover, our method exhibits robust positional generalization capabilities, attaining a success rate of 44.3% in previously unseen positions.

Paperid: 4964, https://arxiv.org/pdf/2508.02190.pdf

Abstract:
Vision-language-action (VLA) models have significantly advanced robotic manipulation by enabling robots to interpret language instructions for task execution. However, training these models often relies on large-scale user-specific data, raising concerns about privacy and security, which in turn limits their broader adoption. To address this, we propose FedVLA, the first federated VLA learning framework, enabling distributed model training that preserves data privacy without compromising performance. Our framework integrates task-aware representation learning, adaptive expert selection, and expert-driven federated aggregation, enabling efficient and privacy-preserving training of VLA models. Specifically, we introduce an Instruction Oriented Scene-Parsing mechanism, which decomposes and enhances object-level features based on task instructions, improving contextual understanding. To effectively learn diverse task patterns, we design a Dual Gating Mixture-of-Experts (DGMoE) mechanism, where not only input tokens but also self-aware experts adaptively decide their activation. Finally, we propose an Expert-Driven Aggregation strategy at the federated server, where model aggregation is guided by activated experts, ensuring effective cross-client knowledge transfer.Extensive simulations and real-world robotic experiments demonstrate the effectiveness of our proposals. Notably, DGMoE significantly improves computational efficiency compared to its vanilla counterpart, while FedVLA achieves task success rates comparable to centralized training, effectively preserving data privacy.

Paperid: 4965, https://arxiv.org/pdf/2508.02062.pdf

Abstract:
Multi-task ``vision-language-action'' (VLA) models have recently demonstrated increasing promise as generalist foundation models for robotics, achieving non-trivial performance out of the box on new tasks in new environments. However, for such models to be truly useful, an end user must have easy means to teach them to improve. For language and vision models, the emergent ability to perform in-context learning (ICL) has proven to be a versatile and highly useful interface to easily teach new tasks with no parameter finetuning. Unfortunately, VLAs pre-trained with imitation learning objectives do not naturally acquire ICL abilities. In this paper, we demonstrate that, with the right finetuning recipe and a small robot demonstration dataset, it is possible to inject in-context adaptability post hoc into such a VLA. After retraining for in-context learning (RICL), our system permits an end user to provide a small number (10-20) of demonstrations for a new task. RICL then fetches the most relevant portions of those demonstrations into the VLA context to exploit ICL, performing the new task and boosting task performance. We apply RICL to inject ICL into the $Ï_{0}$-FAST VLA, and show that it permits large in-context improvements for a variety of new manipulation tasks with only 20 demonstrations per task, without any parameter updates. When parameter updates on the target task demonstrations is possible, RICL finetuning further boosts performance. We release code and model weights for RICL-$Ï_{0}$-FAST alongside the paper to enable, for the first time, a simple in-context learning interface for new manipulation tasks. Website: https://ricl-vla.github.io.

Paperid: 4966, https://arxiv.org/pdf/2508.00230.pdf

Abstract:
Parameter-efficient fine-tuning (PEFT) has become a standard approach for adapting large pre-trained models. Amongst PEFT methods, low-rank adaptation (LoRA) has achieved notable success. However, recent studies have highlighted its limitations compared against full-rank alternatives, particularly when applied to multimodal and large language models. In this work, we present a quantitative comparison amongst full-rank and low-rank PEFT methods using a synthetic matrix approximation benchmark with controlled spectral properties. Our results confirm that LoRA struggles to approximate matrices with relatively flat spectrums or high frequency components -- signs of high effective ranks. To this end, we introduce KRAdapter, a novel PEFT algorithm that leverages the Khatri-Rao product to produce weight updates, which, by construction, tends to produce matrix product with a high effective rank. We demonstrate performance gains with KRAdapter on vision-language models up to 1B parameters and on large language models up to 8B parameters, particularly on unseen common-sense reasoning tasks. In addition, KRAdapter maintains the memory and compute efficiency of LoRA, making it a practical and robust alternative to fine-tune billion-scale parameter models.

Paperid: 4967, https://arxiv.org/pdf/2507.22024.pdf

Abstract:
Foundation models have demonstrated remarkable potential in medical domain. However, their application to complex cardiovascular diagnostics remains underexplored. In this paper, we present Cardiac-CLIP, a multi-modal foundation model designed for 3D cardiac CT images. Cardiac-CLIP is developed through a two-stage pre-training strategy. The first stage employs a 3D masked autoencoder (MAE) to perform self-supervised representation learning from large-scale unlabeled volumetric data, enabling the visual encoder to capture rich anatomical and contextual features. In the second stage, contrastive learning is introduced to align visual and textual representations, facilitating cross-modal understanding. To support the pre-training, we collect 16641 real clinical CT scans, supplemented by 114k publicly available data. Meanwhile, we standardize free-text radiology reports into unified templates and construct the pathology vectors according to diagnostic attributes, based on which the soft-label matrix is generated to supervise the contrastive learning process. On the other hand, to comprehensively evaluate the effectiveness of Cardiac-CLIP, we collect 6,722 real-clinical data from 12 independent institutions, along with the open-source data to construct the evaluation dataset. Specifically, Cardiac-CLIP is comprehensively evaluated across multiple tasks, including cardiovascular abnormality classification, information retrieval and clinical analysis. Experimental results demonstrate that Cardiac-CLIP achieves state-of-the-art performance across various downstream tasks in both internal and external data. Particularly, Cardiac-CLIP exhibits great effectiveness in supporting complex clinical tasks such as the prospective prediction of acute coronary syndrome, which is notoriously difficult in real-world scenarios.

Paperid: 4968, https://arxiv.org/pdf/2507.19110.pdf

Abstract:
Multimodal Large Language Models (MLLMs) excel in vision-language tasks such as image captioning but remain prone to object hallucinations, where they describe objects that do not appear in the image. To mitigate this, we propose \textbf{LISA}, a \textbf{L}ayer-wise \textbf{I}ntegration and \textbf{S}uppression \textbf{A}pproach that enhances generation consistency through hierarchical modulation and multi-layer fusion. LISA leverages the functional hierarchy within MLLMs, where shallow layers provide visual grounding, middle layers encode semantics, and deep layers tend to amplify spurious signals. First, zone-specific spectral modulation stabilizes attention by suppressing over-amplified activations in deeper layers while preserving alignment cues in earlier layers. Second, token-level logits from selected layers are fused via anchor-based routing, with token-wise anchor selection and soft logit fusion enabling adaptive integration during decoding. LISA is fully \textbf{plug-and-play} and can be seamlessly integrated into existing MLLMs, including Qwen2.5-VL. Experiments on multiple benchmarks show that LISA reduces hallucinations by up to 53.6\% in $\mathrm{CHAIR}_I$ and improves POPE F1 by 4.5\%, demonstrating strong generalization across models and tasks.

Paperid: 4969, https://arxiv.org/pdf/2507.17467.pdf

Abstract:
This study investigates the extent to which the Visual Entailment (VE) task serves as a reliable probe of vision-language understanding in multimodal language models, using the LLaMA 3.2 11B Vision model as a test case. Beyond reporting performance metrics, we aim to interpret what these results reveal about the underlying possibilities and limitations of the VE task. We conduct a series of experiments across zero-shot, few-shot, and fine-tuning settings, exploring how factors such as prompt design, the number and order of in-context examples and access to visual information might affect VE performance. To further probe the reasoning processes of the model, we used explanation-based evaluations. Results indicate that three-shot inference outperforms the zero-shot baselines. However, additional examples introduce more noise than they provide benefits. Additionally, the order of the labels in the prompt is a critical factor that influences the predictions. In the absence of visual information, the model has a strong tendency to hallucinate and imagine content, raising questions about the model's over-reliance on linguistic priors. Fine-tuning yields strong results, achieving an accuracy of 83.3% on the e-SNLI-VE dataset and outperforming the state-of-the-art OFA-X model. Additionally, the explanation evaluation demonstrates that the fine-tuned model provides semantically meaningful explanations similar to those of humans, with a BERTScore F1-score of 89.2%. We do, however, find comparable BERTScore results in experiments with limited vision, questioning the visual grounding of this task. Overall, our results highlight both the utility and limitations of VE as a diagnostic task for vision-language understanding and point to directions for refining multimodal evaluation methods.

Paperid: 4970, https://arxiv.org/pdf/2507.17436.pdf

Abstract:
The Mixture of Experts (MoE) architecture has excelled in Large Vision-Language Models (LVLMs), yet its potential in real-time open-vocabulary object detectors, which also leverage large-scale vision-language datasets but smaller models, remains unexplored. This work investigates this domain, revealing intriguing insights. In the shallow layers, experts tend to cooperate with diverse peers to expand the search space. While in the deeper layers, fixed collaborative structures emerge, where each expert maintains 2-3 fixed partners and distinct expert combinations are specialized in processing specific patterns. Concretely, we propose Dynamic-DINO, which extends Grounding DINO 1.5 Edge from a dense model to a dynamic inference framework via an efficient MoE-Tuning strategy. Additionally, we design a granularity decomposition mechanism to decompose the Feed-Forward Network (FFN) of base model into multiple smaller expert networks, expanding the subnet search space. To prevent performance degradation at the start of fine-tuning, we further propose a pre-trained weight allocation strategy for the experts, coupled with a specific router initialization. During inference, only the input-relevant experts are activated to form a compact subnet. Experiments show that, pretrained with merely 1.56M open-source data, Dynamic-DINO outperforms Grounding DINO 1.5 Edge, pretrained on the private Grounding20M dataset.

Paperid: 4971, https://arxiv.org/pdf/2507.13868.pdf

Abstract:
Vision-language models (VLMs) increasingly leverage diverse knowledge sources to address complex tasks, often encountering conflicts between their internal parametric knowledge and external information. Knowledge conflicts can result in hallucinations and unreliable responses, but the mechanisms governing such interactions remain unknown. To address this gap, we analyze the mechanisms that VLMs use to resolve cross-modal conflicts by introducing a dataset of multimodal counterfactual queries that deliberately contradict internal commonsense knowledge. We localize with logit inspection a small set of heads that control the conflict. Moreover, by modifying these heads, we can steer the model towards its internal knowledge or the visual inputs. Finally, we show that attention from such heads pinpoints localized image regions driving visual overrides, outperforming gradient-based attribution in precision.

Paperid: 4972, https://arxiv.org/pdf/2507.04377.pdf

Abstract:
Tombstones are historically and culturally rich artifacts, encapsulating individual lives, community memory, historical narratives and artistic expression. Yet, many tombstones today face significant preservation challenges, including physical erosion, vandalism, environmental degradation, and political shifts. In this paper, we introduce a novel multi-modal framework for tombstones digitization, aiming to improve the interpretation, organization and retrieval of tombstone content. Our approach leverages vision-language models (VLMs) to translate tombstone images into structured Tombstone Meaning Representations (TMRs), capturing both image and text information. To further enrich semantic parsing, we incorporate retrieval-augmented generation (RAG) for integrate externally dependent elements such as toponyms, occupation codes, and ontological concepts. Compared to traditional OCR-based pipelines, our method improves parsing accuracy from an F1 score of 36.1 to 89.5. We additionally evaluate the model's robustness across diverse linguistic and cultural inscriptions, and simulate physical degradation through image fusion to assess performance under noisy or damaged conditions. Our work represents the first attempt to formalize tombstone understanding using large vision-language models, presenting implications for heritage preservation.

Paperid: 4973, https://arxiv.org/pdf/2507.04141.pdf

Abstract:
Prediction of pedestrian crossing intention is a critical function in autonomous vehicles. Conventional vision-based methods of crossing intention prediction often struggle with generalizability, context understanding, and causal reasoning. This study explores the potential of vision-language foundation models (VLFMs) for predicting pedestrian crossing intentions by integrating multimodal data through hierarchical prompt templates. The methodology incorporates contextual information, including visual frames, physical cues observations, and ego-vehicle dynamics, into systematically refined prompts to guide VLFMs effectively in intention prediction. Experiments were conducted on three common datasets-JAAD, PIE, and FU-PIP. Results demonstrate that incorporating vehicle speed, its variations over time, and time-conscious prompts significantly enhances the prediction accuracy up to 19.8%. Additionally, optimised prompts generated via an automatic prompt engineering framework yielded 12.5% further accuracy gains. These findings highlight the superior performance of VLFMs compared to conventional vision-based models, offering enhanced generalisation and contextual understanding for autonomous driving applications.

Paperid: 4974, https://arxiv.org/pdf/2506.23329.pdf

Abstract:
Vision-language models (VLMs) excel at descriptive tasks, but whether they truly understand scenes from visual observations remains uncertain. We introduce IR3D-Bench, a benchmark challenging VLMs to demonstrate understanding through active creation rather than passive recognition. Grounded in the analysis-by-synthesis paradigm, IR3D-Bench tasks Vision-Language Agents (VLAs) with actively using programming and rendering tools to recreate the underlying 3D structure of an input image, achieving agentic inverse rendering through tool use. This "understanding-by-creating" approach probes the tool-using generative capacity of VLAs, moving beyond the descriptive or conversational capacity measured by traditional scene understanding benchmarks. We provide a comprehensive suite of metrics to evaluate geometric accuracy, spatial relations, appearance attributes, and overall plausibility. Initial experiments on agentic inverse rendering powered by various state-of-the-art VLMs highlight current limitations, particularly in visual precision rather than basic tool usage. IR3D-Bench, including data and evaluation protocols, is released to facilitate systematic study and development of tool-using VLAs towards genuine scene understanding by creating.

Paperid: 4975, https://arxiv.org/pdf/2506.22937.pdf

Abstract:
Blind and low-vision (BLV) players encounter critical challenges in engaging with video games due to the inaccessibility of visual elements, difficulties in navigating interfaces, and limitations in sending interaction input. Moreover, the development of specialized accessibility features typically requires substantial programming effort and is often implemented on a game-by-game basis. To address these challenges, we introduce \textit{GamerAstra}, a generalized accessibility framework that leverages a multi-agent design to facilitate access to video games for BLV players. It integrates multi-modal techniques including large language models and vision-language models, enabling interaction with games lacking native accessibility support. The framework further incorporates customizable assistance granularities to support varying degrees of visual impairment and enhances interface navigation through multiple input modalities. The evaluation through technical assessments and user studies indicate that \textit{GamerAstra} effectively enhances playability and delivers a more immersive gaming experience for BLV players. These findings also underscore potential avenues for advancing intelligent accessibility frameworks in the gaming domain.

Paperid: 4976, https://arxiv.org/pdf/2506.17218.pdf

Abstract:
Vision-language models (VLMs) excel at multimodal understanding, yet their text-only decoding forces them to verbalize visual reasoning, limiting performance on tasks that demand visual imagination. Recent attempts train VLMs to render explicit images, but the heavy image-generation pre-training often hinders the reasoning ability. Inspired by the way humans reason with mental imagery-the internal construction and manipulation of visual cues-we investigate whether VLMs can reason through interleaved multimodal trajectories without producing explicit images. To this end, we present a Machine Mental Imagery framework, dubbed as Mirage, which augments VLM decoding with latent visual tokens alongside ordinary text. Concretely, whenever the model chooses to ``think visually'', it recasts its hidden states as next tokens, thereby continuing a multimodal trajectory without generating pixel-level images. Begin by supervising the latent tokens through distillation from ground-truth image embeddings, we then switch to text-only supervision to make the latent trajectory align tightly with the task objective. A subsequent reinforcement learning stage further enhances the multimodal reasoning capability. Experiments on diverse benchmarks demonstrate that Mirage unlocks stronger multimodal reasoning without explicit image generation.

Paperid: 4977, https://arxiv.org/pdf/2506.16724.pdf

Abstract:
With the growing adoption of Large Language Models (LLMs) for open-ended tasks, accurately assessing epistemic uncertainty, which reflects a model's lack of knowledge, has become crucial to ensuring reliable outcomes. However, quantifying epistemic uncertainty in such tasks is challenging due to the presence of aleatoric uncertainty, which arises from multiple valid answers. While bias can introduce noise into epistemic uncertainty estimation, it may also reduce noise from aleatoric uncertainty. To investigate this trade-off, we conduct experiments on Visual Question Answering (VQA) tasks and find that mitigating prompt-introduced bias improves uncertainty quantification in GPT-4o. Building on prior work showing that LLMs tend to copy input information when model confidence is low, we further analyze how these prompt biases affect measured epistemic and aleatoric uncertainty across varying bias-free confidence levels with GPT-4o and Qwen2-VL. We find that all considered biases have greater effects in both uncertainties when bias-free model confidence is lower. Moreover, lower bias-free model confidence is associated with greater bias-induced underestimation of epistemic uncertainty, resulting in overconfident estimates, whereas it has no significant effect on the direction of bias effect in aleatoric uncertainty estimation. These distinct effects deepen our understanding of bias mitigation for uncertainty quantification and potentially inform the development of more advanced techniques.

Paperid: 4978, https://arxiv.org/pdf/2506.11234.pdf

Abstract:
We present Poutine, a 3B-parameter vision-language model (VLM) tailored for end-to-end autonomous driving in long-tail driving scenarios. Poutine is trained in two stages. To obtain strong base driving capabilities, we train Poutine-Base in a self-supervised vision-language-trajectory (VLT) next-token prediction fashion on 83 hours of CoVLA nominal driving and 11 hours of Waymo long-tail driving. Accompanying language annotations are auto-generated with a 72B-parameter VLM. Poutine is obtained by fine-tuning Poutine-Base with Group Relative Policy Optimization (GRPO) using less than 500 preference-labeled frames from the Waymo validation set. We show that both VLT pretraining and RL fine-tuning are critical to attain strong driving performance in the long-tail. Poutine-Base achieves a rater-feedback score (RFS) of 8.12 on the validation set, nearly matching Waymo's expert ground-truth RFS. The final Poutine model achieves an RFS of 7.99 on the official Waymo test set, placing 1st in the 2025 Waymo Vision-Based End-to-End Driving Challenge by a significant margin. These results highlight the promise of scalable VLT pre-training and lightweight RL fine-tuning to enable robust and generalizable autonomy.

Paperid: 4979, https://arxiv.org/pdf/2506.08023.pdf

Abstract:
This paper aims to retrieve proteins with similar structures and semantics from large-scale protein dataset, facilitating the functional interpretation of protein structures derived by structural determination methods like cryo-Electron Microscopy (cryo-EM). Motivated by the recent progress of vision-language models (VLMs), we propose a CLIP-style framework for aligning 3D protein structures with functional annotations using contrastive learning. For model training, we propose a large-scale dataset of approximately 200,000 protein-caption pairs with rich functional descriptors. We evaluate our model in both in-domain and more challenging cross-database retrieval on Protein Data Bank (PDB) and Electron Microscopy Data Bank (EMDB) dataset, respectively. In both cases, our approach demonstrates promising zero-shot retrieval performance, highlighting the potential of multimodal foundation models for structure-function understanding in protein biology.

Paperid: 4980, https://arxiv.org/pdf/2506.07639.pdf

Abstract:
Embodied Chain-of-Thought (ECoT) reasoning enhances vision-language-action (VLA) models by improving performance and interpretability through intermediate reasoning steps. However, its sequential autoregressive token generation introduces significant inference latency, limiting real-time deployment. We propose Fast ECoT, an inference-time acceleration method that exploits the structured and repetitive nature of ECoT to (1) cache and reuse high-level reasoning across timesteps and (2) parallelise the generation of modular reasoning steps. Additionally, we introduce an asynchronous scheduler that decouples reasoning from action decoding, further boosting responsiveness. Fast ECoT requires no model changes or additional training and integrates easily into existing VLA pipelines. Experiments in both simulation (LIBERO) and real-world robot tasks show up to a 7.5% reduction in latency with comparable or improved task success rate and reasoning faithfulness, bringing ECoT policies closer to practical real-time deployment.

Paperid: 4981, https://arxiv.org/pdf/2506.06600.pdf

Abstract:
The growing integration of vision-language models (VLMs) in medical applications offers promising support for diagnostic reasoning. However, current medical VLMs often face limitations in generalization, transparency, and computational efficiency-barriers that hinder deployment in real-world, resource-constrained settings. To address these challenges, we propose a Reasoning-Aware Reinforcement Learning framework, \textbf{RARL}, that enhances the reasoning capabilities of medical VLMs while remaining efficient and adaptable to low-resource environments. Our approach fine-tunes a lightweight base model, Qwen2-VL-2B-Instruct, using Low-Rank Adaptation and custom reward functions that jointly consider diagnostic accuracy and reasoning quality. Training is performed on a single NVIDIA A100-PCIE-40GB GPU, demonstrating the feasibility of deploying such models in constrained environments. We evaluate the model using an LLM-as-judge framework that scores both correctness and explanation quality. Experimental results show that RARL significantly improves VLM performance in medical image analysis and clinical reasoning, outperforming supervised fine-tuning on reasoning-focused tasks by approximately 7.78%, while requiring fewer computational resources. Additionally, we demonstrate the generalization capabilities of our approach on unseen datasets, achieving around 27% improved performance compared to supervised fine-tuning and about 4% over traditional RL fine-tuning. Our experiments also illustrate that diversity prompting during training and reasoning prompting during inference are crucial for enhancing VLM performance. Our findings highlight the potential of reasoning-guided learning and reasoning prompting to steer medical VLMs toward more transparent, accurate, and resource-efficient clinical decision-making. Code and data are publicly available.

Paperid: 4982, https://arxiv.org/pdf/2506.06506.pdf

Abstract:
To build fair AI systems we need to understand how social-group biases intrinsic to foundational encoder-based vision-language models (VLMs) manifest in biases in downstream tasks. In this study, we demonstrate that intrinsic biases in VLM representations systematically ``carry over'' or propagate into zero-shot retrieval tasks, revealing how deeply rooted biases shape a model's outputs. We introduce a controlled framework to measure this propagation by correlating (a) intrinsic measures of bias in the representational space with (b) extrinsic measures of bias in zero-shot text-to-image (TTI) and image-to-text (ITT) retrieval. Results show substantial correlations between intrinsic and extrinsic bias, with an average $Ï$ = 0.83 $\pm$ 0.10. This pattern is consistent across 114 analyses, both retrieval directions, six social groups, and three distinct VLMs. Notably, we find that larger/better-performing models exhibit greater bias propagation, a finding that raises concerns given the trend towards increasingly complex AI models. Our framework introduces baseline evaluation tasks to measure the propagation of group and valence signals. Investigations reveal that underrepresented groups experience less robust propagation, further skewing their model-related outcomes.

Paperid: 4983, https://arxiv.org/pdf/2506.06084.pdf

Abstract:
Wheat management strategies play a critical role in determining yield. Traditional management decisions often rely on labour-intensive expert inspections, which are expensive, subjective and difficult to scale. Recently, Vision-Language Models (VLMs) have emerged as a promising solution to enable scalable, data-driven management support. However, due to a lack of domain-specific knowledge, directly applying VLMs to wheat management tasks results in poor quantification and reasoning capabilities, ultimately producing vague or even misleading management recommendations. In response, we propose WisWheat, a wheat-specific dataset with a three-layered design to enhance VLM performance on wheat management tasks: (1) a foundational pretraining dataset of 47,871 image-caption pairs for coarsely adapting VLMs to wheat morphology; (2) a quantitative dataset comprising 7,263 VQA-style image-question-answer triplets for quantitative trait measuring tasks; and (3) an Instruction Fine-tuning dataset with 4,888 samples targeting biotic and abiotic stress diagnosis and management plan for different phenological stages. Extensive experimental results demonstrate that fine-tuning open-source VLMs (e.g., Qwen2.5 7B) on our dataset leads to significant performance improvements. Specifically, the Qwen2.5 VL 7B fine-tuned on our wheat instruction dataset achieves accuracy scores of 79.2% and 84.6% on wheat stress and growth stage conversation tasks respectively, surpassing even general-purpose commercial models such as GPT-4o by a margin of 11.9% and 34.6%.

Paperid: 4984, https://arxiv.org/pdf/2506.05442.pdf

Abstract:
Vision-Language Models (VLMs) offer a promising approach to end-to-end autonomous driving due to their human-like reasoning capabilities. However, troublesome gaps remains between current VLMs and real-world autonomous driving applications. One major limitation is that existing datasets with loosely formatted language descriptions are not machine-friendly and may introduce redundancy. Additionally, high computational cost and massive scale of VLMs hinder the inference speed and real-world deployment. To bridge the gap, this paper introduces a structured and concise benchmark dataset, NuScenes-S, which is derived from the NuScenes dataset and contains machine-friendly structured representations. Moreover, we present FastDrive, a compact VLM baseline with 0.9B parameters. In contrast to existing VLMs with over 7B parameters and unstructured language processing(e.g., LLaVA-1.5), FastDrive understands structured and concise descriptions and generates machine-friendly driving decisions with high efficiency. Extensive experiments show that FastDrive achieves competitive performance on structured dataset, with approximately 20% accuracy improvement on decision-making tasks, while surpassing massive parameter baseline in inference speed with over 10x speedup. Additionally, ablation studies further focus on the impact of scene annotations (e.g., weather, time of day) on decision-making tasks, demonstrating their importance on decision-making tasks in autonomous driving.

Paperid: 4985, https://arxiv.org/pdf/2506.02738.pdf

Abstract:
Compound figures, which are multi-panel composites containing diverse subfigures, are ubiquitous in biomedical literature, yet large-scale subfigure extraction remains largely unaddressed. Prior work on subfigure extraction has been limited in both dataset size and generalizability, leaving a critical open question: How does high-fidelity image-text alignment via large-scale subfigure extraction impact representation learning in vision-language models? We address this gap by introducing a scalable subfigure extraction pipeline based on transformer-based object detection, trained on a synthetic corpus of 500,000 compound figures, and achieving state-of-the-art performance on both ImageCLEF 2016 and synthetic benchmarks. Using this pipeline, we release OPEN-PMC-18M, a large-scale high quality biomedical vision-language dataset comprising 18 million clinically relevant subfigure-caption pairs spanning radiology, microscopy, and visible light photography. We train and evaluate vision-language models on our curated datasets and show improved performance across retrieval, zero-shot classification, and robustness benchmarks, outperforming existing baselines. We release our dataset, models, and code to support reproducible benchmarks and further study into biomedical vision-language modeling and representation learning.

Paperid: 4986, https://arxiv.org/pdf/2506.01795.pdf

Abstract:
We introduce a new task, Referring and Reasoning for Selective Masks (R2SM), which extends text-guided segmentation by incorporating mask-type selection driven by user intent. This task challenges vision-language models to determine whether to generate a modal (visible) or amodal (complete) segmentation mask based solely on natural language prompts. To support the R2SM task, we present the R2SM dataset, constructed by augmenting annotations of COCOA-cls, D2SA, and MUVA. The R2SM dataset consists of both modal and amodal text queries, each paired with the corresponding ground-truth mask, enabling model finetuning and evaluation for the ability to segment images as per user intent. Specifically, the task requires the model to interpret whether a given prompt refers to only the visible part of an object or to its complete shape, including occluded regions, and then produce the appropriate segmentation. For example, if a prompt explicitly requests the whole shape of a partially hidden object, the model is expected to output an amodal mask that completes the occluded parts. In contrast, prompts without explicit mention of hidden regions should generate standard modal masks. The R2SM benchmark provides a challenging and insightful testbed for advancing research in multimodal reasoning and intent-aware segmentation.

Paperid: 4987, https://arxiv.org/pdf/2505.22858.pdf

Abstract:
Open-world egocentric activity recognition poses a fundamental challenge due to its unconstrained nature, requiring models to infer unseen activities from an expansive, partially observed search space. We introduce ProbRes, a Probabilistic Residual search framework based on jump-diffusion that efficiently navigates this space by balancing prior-guided exploration with likelihood-driven exploitation. Our approach integrates structured commonsense priors to construct a semantically coherent search space, adaptively refines predictions using Vision-Language Models (VLMs) and employs a stochastic search mechanism to locate high-likelihood activity labels while minimizing exhaustive enumeration efficiently. We systematically evaluate ProbRes across multiple openness levels (L0--L3), demonstrating its adaptability to increasing search space complexity. In addition to achieving state-of-the-art performance on benchmark datasets (GTEA Gaze, GTEA Gaze+, EPIC-Kitchens, and Charades-Ego), we establish a clear taxonomy for open-world recognition, delineating the challenges and methodological advancements necessary for egocentric activity understanding.

Paperid: 4988, https://arxiv.org/pdf/2505.20029.pdf

Abstract:
Transformer-based language models, though not explicitly trained to mimic brain recordings, have demonstrated surprising alignment with brain activity. Progress in these models-through increased size, instruction-tuning, and multimodality-has led to better representational alignment with neural data. Recently, a new class of instruction-tuned multimodal LLMs (MLLMs) have emerged, showing remarkable zero-shot capabilities in open-ended multimodal vision tasks. However, it is unknown whether MLLMs, when prompted with natural instructions, lead to better brain alignment and effectively capture instruction-specific representations. To address this, we first investigate brain alignment, i.e., measuring the degree of predictivity of neural visual activity using text output response embeddings from MLLMs as participants engage in watching natural scenes. Experiments with 10 different instructions show that MLLMs exhibit significantly better brain alignment than vision-only models and perform comparably to non-instruction-tuned multimodal models like CLIP. We also find that while these MLLMs are effective at generating high-quality responses suitable to the task-specific instructions, not all instructions are relevant for brain alignment. Further, by varying instructions, we make the MLLMs encode instruction-specific visual concepts related to the input image. This analysis shows that MLLMs effectively capture count-related and recognition-related concepts, demonstrating strong alignment with brain activity. Notably, the majority of the explained variance of the brain encoding models is shared between MLLM embeddings of image captioning and other instructions. These results suggest that enhancing MLLMs' ability to capture task-specific information could lead to better differentiation between various types of instructions, and thereby improving their precision in predicting brain responses.

Paperid: 4989, https://arxiv.org/pdf/2505.19373.pdf

Abstract:
Prompt learning has emerged as a powerful paradigm for adapting vision-language models such as CLIP to downstream tasks. However, existing methods often overfit to seen data, leading to significant performance degradation when generalizing to novel classes or unseen domains. To address this limitation, we propose DiSa, a Directional Saliency-Aware Prompt Learning framework that integrates two complementary regularization strategies to enhance generalization. First, our Cross-Interactive Regularization (CIR) fosters cross-modal alignment by enabling cooperative learning between prompted and frozen encoders. Within CIR, a saliency-aware masking strategy guides the image encoder to prioritize semantically critical image regions, reducing reliance on less informative patches. Second, we introduce a directional regularization strategy that aligns visual embeddings with class-wise prototype features in a directional manner to prioritize consistency in feature orientation over strict proximity. This approach ensures robust generalization by leveraging stable prototype directions derived from class-mean statistics. Extensive evaluations on 11 diverse image classification benchmarks demonstrate that DiSa consistently outperforms state-of-the-art prompt learning methods across various settings, including base-to-novel generalization, cross-dataset transfer, domain generalization, and few-shot learning.

Paperid: 4990, https://arxiv.org/pdf/2505.19095.pdf

Abstract:
The rapid progress of large language models (LLMs) has sparked growing interest in building Artificial General Intelligence (AGI) within Graphical User Interface (GUI) environments. However, existing GUI agents based on LLMs or vision-language models (VLMs) often fail to generalize to novel environments and rely heavily on manually curated, diverse datasets. To overcome these limitations, we introduce ScreenExplorer, a VLM trained via Group Relative Policy Optimization(GRPO) in real, dynamic, and open-ended GUI environments. Innovatively, we introduced a world-model-based curiosity reward function to help the agent overcome the cold-start phase of exploration. Additionally, distilling experience streams further enhances the model's exploration capabilities. Our training framework enhances model exploration in open GUI environments, with trained models showing better environmental adaptation and sustained exploration compared to static deployment models. Our findings offer a scalable pathway toward AGI systems with self-improving capabilities in complex interactive settings.

Paperid: 4991, https://arxiv.org/pdf/2505.18668.pdf

Abstract:
Infographic charts are a powerful medium for communicating abstract data by combining visual elements (e.g., charts, images) with textual information. However, their visual and structural richness poses challenges for large vision-language models (LVLMs), which are typically trained on plain charts. To bridge this gap, we introduce ChartGalaxy, a million-scale dataset designed to advance the understanding and generation of infographic charts. The dataset is constructed through an inductive process that identifies 75 chart types, 330 chart variations, and 68 layout templates from real infographic charts and uses them to create synthetic ones programmatically. We showcase the utility of this dataset through: 1) improving infographic chart understanding via fine-tuning, 2) benchmarking code generation for infographic charts, and 3) enabling example-based infographic chart generation. By capturing the visual and structural complexity of real design, ChartGalaxy provides a useful resource for enhancing multimodal reasoning and generation in LVLMs.

Paperid: 4992, https://arxiv.org/pdf/2505.17645.pdf

Abstract:
Embodied agents operating in smart homes must understand human behavior through diverse sensory inputs and communicate via natural language. While Vision-Language Models (VLMs) have enabled impressive language-grounded perception, their reliance on visual data limits robustness in real-world scenarios with occlusions, poor lighting, or privacy constraints. In this paper, we introduce HoloLLM, a Multimodal Large Language Model (MLLM) that integrates uncommon but powerful sensing modalities, such as LiDAR, infrared, mmWave radar, and WiFi, to enable seamless human perception and reasoning across heterogeneous environments. We address two key challenges: (1) the scarcity of aligned modality-text data for rare sensors, and (2) the heterogeneity of their physical signal representations. To overcome these, we design a Universal Modality-Injection Projector (UMIP) that enhances pre-aligned modality embeddings with fine-grained, text-aligned features from tailored encoders via coarse-to-fine cross-attention without introducing significant alignment overhead. We further introduce a human-VLM collaborative data curation pipeline to generate paired textual annotations for sensing datasets. Extensive experiments on two newly constructed benchmarks show that HoloLLM significantly outperforms existing MLLMs, improving language-grounded human sensing accuracy by up to 30%. This work establishes a new foundation for real-world, language-informed multisensory embodied intelligence.

Paperid: 4993, https://arxiv.org/pdf/2505.17473.pdf

Abstract:
Given the central role of charts in scientific, business, and communication contexts, enhancing the chart understanding capabilities of vision-language models (VLMs) has become increasingly critical. A key limitation of existing VLMs lies in their inaccurate visual grounding of infographic elements, including charts and human-recognizable objects (HROs) such as icons and images. However, chart understanding often requires identifying relevant elements and reasoning over them. To address this limitation, we introduce OrionBench, a benchmark designed to support the development of accurate object detection models for charts and HROs in infographics. It contains 26,250 real and 78,750 synthetic infographics, with over 6.9 million bounding box annotations. These annotations are created by combining the model-in-the-loop and programmatic methods. We demonstrate the usefulness of OrionBench through three applications: 1) constructing a Thinking-with-Boxes scheme to boost the chart understanding performance of VLMs, 2) comparing existing object detection models, and 3) applying the developed detection model to document layout and UI element detection.

Paperid: 4994, https://arxiv.org/pdf/2505.15098.pdf

Abstract:
Robot manipulation learning from human demonstrations offers a rapid means to acquire skills but often lacks generalization across diverse scenes and object placements. This limitation hinders real-world applications, particularly in complex tasks requiring dexterous manipulation. Vision-Language-Action (VLA) paradigm leverages large-scale data to enhance generalization. However, due to data scarcity, VLA's performance remains limited. In this work, we introduce Object-Focus Actor (OFA), a novel, data-efficient approach for generalized dexterous manipulation. OFA exploits the consistent end trajectories observed in dexterous manipulation tasks, allowing for efficient policy training. Our method employs a hierarchical pipeline: object perception and pose estimation, pre-manipulation pose arrival and OFA policy execution. This process ensures that the manipulation is focused and efficient, even in varied backgrounds and positional layout. Comprehensive real-world experiments across seven tasks demonstrate that OFA significantly outperforms baseline methods in both positional and background generalization tests. Notably, OFA achieves robust performance with only 10 demonstrations, highlighting its data efficiency.

Paperid: 4995, https://arxiv.org/pdf/2505.14361.pdf

Abstract:
Vision-language modeling (VLM) aims to bridge the information gap between images and natural language. Under the new paradigm of first pre-training on massive image-text pairs and then fine-tuning on task-specific data, VLM in the remote sensing domain has made significant progress. The resulting models benefit from the absorption of extensive general knowledge and demonstrate strong performance across a variety of remote sensing data analysis tasks. Moreover, they are capable of interacting with users in a conversational manner. In this paper, we aim to provide the remote sensing community with a timely and comprehensive review of the developments in VLM using the two-stage paradigm. Specifically, we first cover a taxonomy of VLM in remote sensing: contrastive learning, visual instruction tuning, and text-conditioned image generation. For each category, we detail the commonly used network architecture and pre-training objectives. Second, we conduct a thorough review of existing works, examining foundation models and task-specific adaptation methods in contrastive-based VLM, architectural upgrades, training strategies and model capabilities in instruction-based VLM, as well as generative foundation models with their representative downstream applications. Third, we summarize datasets used for VLM pre-training, fine-tuning, and evaluation, with an analysis of their construction methodologies (including image sources and caption generation) and key properties, such as scale and task adaptability. Finally, we conclude this survey with insights and discussions on future research directions: cross-modal representation alignment, vague requirement comprehension, explanation-driven model reliability, continually scalable model capabilities, and large-scale datasets featuring richer modalities and greater challenges.

Paperid: 4996, https://arxiv.org/pdf/2505.12000.pdf

Abstract:
Although large Vision-Language Models (VLMs) have demonstrated remarkable performance in a wide range of multimodal tasks, their true reasoning capabilities on human IQ tests remain underexplored. To advance research on the fluid intelligence of VLMs, we introduce **IQBench**, a new benchmark designed to evaluate VLMs on standardized visual IQ tests. We focus on evaluating the reasoning capabilities of VLMs, which we argue are more important than the accuracy of the final prediction. **Our benchmark is visually centric, minimizing the dependence on unnecessary textual content**, thus encouraging models to derive answers primarily from image-based information rather than learned textual knowledge. To this end, we manually collected and annotated 500 visual IQ questions to **prevent unintentional data leakage during training**. Unlike prior work that focuses primarily on the accuracy of the final answer, we evaluate the reasoning ability of the models by assessing their explanations and the patterns used to solve each problem, along with the accuracy of the final prediction and human evaluation. Our experiments show that there are substantial performance disparities between tasks, with models such as `o4-mini`, `gemini-2.5-flash`, and `claude-3.7-sonnet` achieving the highest average accuracies of 0.615, 0.578, and 0.548, respectively. However, all models struggle with 3D spatial and anagram reasoning tasks, highlighting significant limitations in current VLMs' general reasoning abilities. In terms of reasoning scores, `o4-mini`, `gemini-2.5-flash`, and `claude-3.7-sonnet` achieved top averages of 0.696, 0.586, and 0.516, respectively. These results highlight inconsistencies between the reasoning processes of the models and their final answers, emphasizing the importance of evaluating the accuracy of the reasoning in addition to the final predictions.

Paperid: 4997, https://arxiv.org/pdf/2505.07815.pdf

Abstract:
Exploration is essential for general-purpose robotic learning, especially in open-ended environments where dense rewards, explicit goals, or task-specific supervision are scarce. Vision-language models (VLMs), with their semantic reasoning over objects, spatial relations, and potential outcomes, present a compelling foundation for generating high-level exploratory behaviors. However, their outputs are often ungrounded, making it difficult to determine whether imagined transitions are physically feasible or informative. To bridge the gap between imagination and execution, we present IVE (Imagine, Verify, Execute), an agentic exploration framework inspired by human curiosity. Human exploration is often driven by the desire to discover novel scene configurations and to deepen understanding of the environment. Similarly, IVE leverages VLMs to abstract RGB-D observations into semantic scene graphs, imagine novel scenes, predict their physical plausibility, and generate executable skill sequences through action tools. We evaluate IVE in both simulated and real-world tabletop environments. The results show that IVE enables more diverse and meaningful exploration than RL baselines, as evidenced by a 4.1 to 7.8x increase in the entropy of visited states. Moreover, the collected experience supports downstream learning, producing policies that closely match or exceed the performance of those trained on human-collected demonstrations.

Paperid: 4998, https://arxiv.org/pdf/2504.20520.pdf

Abstract:
Learning from few demonstrations to develop policies robust to variations in robot initial positions and object poses is a problem of significant practical interest in robotics. Compared to imitation learning, which often struggles to generalize from limited samples, reinforcement learning (RL) can autonomously explore to obtain robust behaviors. Training RL agents through direct interaction with the real world is often impractical and unsafe, while building simulation environments requires extensive manual effort, such as designing scenes and crafting task-specific reward functions. To address these challenges, we propose an integrated real-to-sim-to-real pipeline that constructs simulation environments based on expert demonstrations by identifying scene objects from images and retrieving their corresponding 3D models from existing libraries. We introduce a projection-based reward model for RL policy training that is supervised by a vision-language model (VLM) using human-guided object projection relationships as prompts, with the policy further fine-tuned using expert demonstrations. In general, our work focuses on the construction of simulation environments and RL-based policy training, ultimately enabling the deployment of reliable robotic control policies in real-world scenarios.

Paperid: 4999, https://arxiv.org/pdf/2504.11675.pdf

Abstract:
Testing Android apps effectively requires a systematic exploration of the app's possible states by simulating user interactions and system events. While existing approaches have proposed several fuzzing techniques to generate various text inputs and trigger user and system events for UI state exploration, achieving high code coverage remains a significant challenge in Android app testing. The main challenges are (1) reasoning about the complex and dynamic layout of UI screens; (2) generating required inputs/events to deal with certain widgets like pop-ups; and (3) coordination between current test inputs and previous inputs to avoid getting stuck in the same UI screen without improving test coverage. To address these problems, we propose a novel, automated fuzzing approach called VLM-Fuzz for effective UI testing of Android apps. We present a novel heuristic-based depth-first search (DFS) exploration algorithm, assisted with a vision language model (VLM), to effectively explore the UI states of the app. We use static analysis to analyze the Android Manifest file and the runtime UI hierarchy XML to extract the list of components, intent-filters and interactive UI widgets. VLM is used to reason about complex UI layout and widgets on an on-demand basis. Based on the inputs from static analysis, VLM, and the current UI state, we use some heuristics to deal with the above-mentioned challenges. We evaluated VLM-Fuzz based on a benchmark containing 59 apps obtained from a recent work and compared it against two state-of-the-art approaches: APE and DeepGUI. VLM-Fuzz outperforms the best baseline by 9.0%, 3.7%, and 2.1% in terms of class coverage, method coverage, and line coverage, respectively. We also ran VLM-Fuzz on 80 recent Google Play apps (i.e., updated in 2024). VLM-Fuzz detected 208 unique crashes in 24 apps, which have been reported to respective developers.

Paperid: 5000, https://arxiv.org/pdf/2504.03948.pdf

Abstract:
Open-world egocentric activity recognition poses a fundamental challenge due to its unconstrained nature, requiring models to infer unseen activities from an expansive, partially observed search space. We introduce ProbRes, a Probabilistic Residual search framework based on jump-diffusion that efficiently navigates this space by balancing prior-guided exploration with likelihood-driven exploitation. Our approach integrates structured commonsense priors to construct a semantically coherent search space, adaptively refines predictions using Vision-Language Models (VLMs) and employs a stochastic search mechanism to locate high-likelihood activity labels while minimizing exhaustive enumeration efficiently. We systematically evaluate ProbRes across multiple openness levels (L0 - L3), demonstrating its adaptability to increasing search space complexity. In addition to achieving state-of-the-art performance on benchmark datasets (GTEA Gaze, GTEA Gaze+, EPIC-Kitchens, and Charades-Ego), we establish a clear taxonomy for open-world recognition, delineating the challenges and methodological advancements necessary for egocentric activity understanding. Our results highlight the importance of structured search strategies, paving the way for scalable and efficient open-world activity recognition.

Paperid: 5001, https://arxiv.org/pdf/2504.03440.pdf

Abstract:
To leverage the full potential of Large Language Models (LLMs) it is crucial to have some information on their answers' uncertainty. This means that the model has to be able to quantify how certain it is in the correctness of a given response. Bad uncertainty estimates can lead to overconfident wrong answers undermining trust in these models. Quite a lot of research has been done on language models that work with text inputs and provide text outputs. Still, since the visual capabilities have been added to these models recently, there has not been much progress on the uncertainty of Visual Language Models (VLMs). We tested three state-of-the-art VLMs on corrupted image data. We found that the severity of the corruption negatively impacted the models' ability to estimate their uncertainty and the models also showed overconfidence in most of the experiments.

Paperid: 5002, https://arxiv.org/pdf/2504.02069.pdf

Abstract:
Visual Language Models (VLMs) have emerged as pivotal tools for robotic systems, enabling cross-task generalization, dynamic environmental interaction, and long-horizon planning through multimodal perception and semantic reasoning. However, existing open-source VLMs predominantly trained for generic vision-language alignment tasks fail to model temporally correlated action semantics that are crucial for robotic manipulation effectively. While current image-based fine-tuning methods partially adapt VLMs to robotic applications, they fundamentally disregard temporal evolution patterns in video sequences and suffer from visual feature entanglement between robotic agents, manipulated objects, and environmental contexts, thereby limiting semantic decoupling capability for atomic actions and compromising model generalizability.To overcome these challenges, this work presents RoboAct-CLIP with dual technical contributions: 1) A dataset reconstruction framework that performs semantic-constrained action unit segmentation and re-annotation on open-source robotic videos, constructing purified training sets containing singular atomic actions (e.g., "grasp"); 2) A temporal-decoupling fine-tuning strategy based on Contrastive Language-Image Pretraining (CLIP) architecture, which disentangles temporal action features across video frames from object-centric characteristics to achieve hierarchical representation learning of robotic atomic actions.Experimental results in simulated environments demonstrate that the RoboAct-CLIP pretrained model achieves a 12% higher success rate than baseline VLMs, along with superior generalization in multi-object manipulation tasks.

Paperid: 5003, https://arxiv.org/pdf/2504.01786.pdf

Abstract:
3D graphics editing is crucial in applications like movie production and game design, yet it remains a time-consuming process that demands highly specialized domain expertise. Automating this process is challenging because graphical editing requires performing a variety of tasks, each requiring distinct skill sets. Recently, vision-language models (VLMs) have emerged as a powerful framework for automating the editing process, but their development and evaluation are bottlenecked by the lack of a comprehensive benchmark that requires human-level perception and presents real-world editing complexity. In this work, we present BlenderGym, the first comprehensive VLM system benchmark for 3D graphics editing. BlenderGym evaluates VLM systems through code-based 3D reconstruction tasks. We evaluate closed- and open-source VLM systems and observe that even the state-of-the-art VLM system struggles with tasks relatively easy for human Blender users. Enabled by BlenderGym, we study how inference scaling techniques impact VLM's performance on graphics editing tasks. Notably, our findings reveal that the verifier used to guide the scaling of generation can itself be improved through inference scaling, complementing recent insights on inference scaling of LLM generation in coding and math tasks. We further show that inference compute is not uniformly effective and can be optimized by strategically distributing it between generation and verification.

Paperid: 5004, https://arxiv.org/pdf/2504.00907.pdf

Abstract:
Embodied agents operating in real-world environments must interpret ambiguous and under-specified human instructions. A capable household robot should recognize ambiguity and ask relevant clarification questions to infer the user intent accurately, leading to more effective task execution. To study this problem, we introduce the Ask-to-Act task, where an embodied agent must fetch a specific object instance given an ambiguous instruction in a home environment. The agent must strategically ask minimal, yet relevant, clarification questions to resolve ambiguity while navigating under partial observability. To solve this problem, we propose a novel approach that fine-tunes multimodal large language models (MLLMs) as vision-language-action (VLA) policies using online reinforcement learning (RL) with LLM-generated rewards. Our method eliminates the need for large-scale human demonstrations or manually engineered rewards for training such agents. We benchmark against strong zero-shot baselines, including GPT-4o, and supervised fine-tuned MLLMs, on our task. Our results demonstrate that our RL-finetuned MLLM outperforms all baselines by a significant margin ($19.1$-$40.3\%$), generalizing well to novel scenes and tasks. To the best of our knowledge, this is the first demonstration of adapting MLLMs as VLA agents that can act and ask for help using LLM-generated rewards with online RL.

Paperid: 5005, https://arxiv.org/pdf/2504.00848.pdf

Abstract:
Zero-shot 4D segmentation and recognition of arbitrary objects in Lidar is crucial for embodied navigation, with applications ranging from streaming perception to semantic mapping and localization. However, the primary challenge in advancing research and developing generalized, versatile methods for spatio-temporal scene understanding in Lidar lies in the scarcity of datasets that provide the necessary diversity and scale of annotations.To overcome these challenges, we propose SAL-4D (Segment Anything in Lidar--4D), a method that utilizes multi-modal robotic sensor setups as a bridge to distill recent developments in Video Object Segmentation (VOS) in conjunction with off-the-shelf Vision-Language foundation models to Lidar. We utilize VOS models to pseudo-label tracklets in short video sequences, annotate these tracklets with sequence-level CLIP tokens, and lift them to the 4D Lidar space using calibrated multi-modal sensory setups to distill them to our SAL-4D model. Due to temporal consistent predictions, we outperform prior art in 3D Zero-Shot Lidar Panoptic Segmentation (LPS) over $5$ PQ, and unlock Zero-Shot 4D-LPS.

Paperid: 5006, https://arxiv.org/pdf/2503.21227.pdf

Abstract:
Mixture of Experts (MoE) architectures have recently advanced the scalability and adaptability of large language models (LLMs) for continual multimodal learning. However, efficiently extending these models to accommodate sequential tasks remains challenging. As new tasks arrive, naive model expansion leads to rapid parameter growth, while modifying shared routing components often causes catastrophic forgetting, undermining previously learned knowledge. To address these issues, we propose LLaVA-CMoE, a continual learning framework for LLMs that requires no replay data of previous tasks and ensures both parameter efficiency and robust knowledge retention. Our approach introduces a Probe-Guided Knowledge Extension mechanism, which uses probe experts to dynamically determine when and where new experts should be added, enabling adaptive and minimal parameter expansion tailored to task complexity. Furthermore, we present a Probabilistic Task Locator that assigns each task a dedicated, lightweight router. To handle the practical issue that task labels are unknown during inference, we leverage a VAE-based reconstruction strategy to identify the most suitable router by matching input distributions, allowing automatic and accurate expert allocation. This design mitigates routing conflicts and catastrophic forgetting, enabling robust continual learning without explicit task labels. Extensive experiments on the CoIN benchmark, covering eight diverse VQA tasks, demonstrate that LLaVA-CMoE delivers strong continual learning performance with a compact model size, significantly reducing forgetting and parameter overhead compared to prior methods. These results showcase the effectiveness and scalability of our approach for parameter-efficient continual learning in large language models. Our code will be open-sourced soon.

Paperid: 5007, https://arxiv.org/pdf/2503.16538.pdf

Abstract:
This paper introduces a novel approach that leverages the capabilities of vision-language models (VLMs) by integrating them with established approaches for open-vocabulary detection (OVD), instance segmentation, and tracking. We utilize VLM-generated structured descriptions to identify visible object instances, collect application-relevant attributes, and inform an open-vocabulary detector to extract corresponding bounding boxes that are passed to a video segmentation model providing precise segmentation masks and tracking capabilities. Once initialized, this model can then directly extract segmentation masks, allowing processing of image streams in real time with minimal computational overhead. Tracks can be updated online as needed by generating new structured descriptions and corresponding open-vocabulary detections. This combines the descriptive power of VLMs with the grounding capability of OVD and the pixel-level understanding and speed of video segmentation. Our evaluation across datasets and robotics platforms demonstrates the broad applicability of this approach, showcasing its ability to extract task-specific attributes from non-standard objects in dynamic environments.

Paperid: 5008, https://arxiv.org/pdf/2503.14538.pdf

Abstract:
Background: This study introduces a Vision-Language Model (VLM) leveraging SIGLIP and Gemma-3b architectures for automated acute tuberculosis (TB) screening. By integrating chest X-ray images and clinical notes, the model aims to enhance diagnostic accuracy and efficiency, particularly in resource-limited settings. Methods: The VLM combines visual data from chest X-rays with clinical context to generate detailed, context-aware diagnostic reports. The architecture employs SIGLIP for visual encoding and Gemma-3b for decoding, ensuring effective representation of acute TB-specific pathologies and clinical insights. Results: Key acute TB pathologies, including consolidation, cavities, and nodules, were detected with high precision (97percent) and recall (96percent). The model demonstrated strong spatial localization capabilities and robustness in distinguishing TB-positive cases, making it a reliable tool for acute TB diagnosis. Conclusion: The multimodal capability of the VLM reduces reliance on radiologists, providing a scalable solution for acute TB screening. Future work will focus on improving the detection of subtle pathologies and addressing dataset biases to enhance its generalizability and application in diverse global healthcare settings.

Paperid: 5009, https://arxiv.org/pdf/2503.14536.pdf

Abstract:
Background: This study proposes a Vision-Language Model (VLM) leveraging the SIGLIP encoder and Gemma-3b transformer decoder to enhance automated chronic tuberculosis (TB) screening. By integrating chest X-ray images with clinical data, the model addresses the challenges of manual interpretation, improving diagnostic consistency and accessibility, particularly in resource-constrained settings. Methods: The VLM architecture combines a Vision Transformer (ViT) for visual encoding and a transformer-based text encoder to process clinical context, such as patient histories and treatment records. Cross-modal attention mechanisms align radiographic features with textual information, while the Gemma-3b decoder generates comprehensive diagnostic reports. The model was pre-trained on 5 million paired medical images and texts and fine-tuned using 100,000 chronic TB-specific chest X-rays. Results: The model demonstrated high precision (94 percent) and recall (94 percent) for detecting key chronic TB pathologies, including fibrosis, calcified granulomas, and bronchiectasis. Area Under the Curve (AUC) scores exceeded 0.93, and Intersection over Union (IoU) values were above 0.91, validating its effectiveness in detecting and localizing TB-related abnormalities. Conclusion: The VLM offers a robust and scalable solution for automated chronic TB diagnosis, integrating radiographic and clinical data to deliver actionable and context-aware insights. Future work will address subtle pathologies and dataset biases to enhance the model's generalizability, ensuring equitable performance across diverse populations and healthcare settings.

Paperid: 5010, https://arxiv.org/pdf/2503.13684.pdf

Abstract:
Numerous text-to-video (T2V) editing methods have emerged recently, but the lack of a standardized benchmark for fair evaluation has led to inconsistent claims and an inability to assess model sensitivity to hyperparameters. Fine-grained video editing is crucial for enabling precise, object-level modifications while maintaining context and temporal consistency. To address this, we introduce FiVE, a Fine-grained Video Editing Benchmark for evaluating emerging diffusion and rectified flow models. Our benchmark includes 74 real-world videos and 26 generated videos, featuring 6 fine-grained editing types, 420 object-level editing prompt pairs, and their corresponding masks. Additionally, we adapt the latest rectified flow (RF) T2V generation models, Pyramid-Flow and Wan2.1, by introducing FlowEdit, resulting in training-free and inversion-free video editing models Pyramid-Edit and Wan-Edit. We evaluate five diffusion-based and two RF-based editing methods on our FiVE benchmark using 15 metrics, covering background preservation, text-video similarity, temporal consistency, video quality, and runtime. To further enhance object-level evaluation, we introduce FiVE-Acc, a novel metric leveraging Vision-Language Models (VLMs) to assess the success of fine-grained video editing. Experimental results demonstrate that RF-based editing significantly outperforms diffusion-based methods, with Wan-Edit achieving the best overall performance and exhibiting the least sensitivity to hyperparameters. More video demo available on the anonymous website: https://sites.google.com/view/five-benchmark

Paperid: 5011, https://arxiv.org/pdf/2503.11360.pdf

Abstract:
Language-guided attention frameworks have significantly enhanced both interpretability and performance in image classification; however, the reliance on deterministic embeddings from pre-trained vision-language foundation models to generate reference attention maps frequently overlooks the intrinsic multivaluedness and ill-posed characteristics of cross-modal mappings. To address these limitations, we introduce PARIC, a probabilistic framework for guiding visual attention via language specifications. Our approach enables pre-trained vision-language models to generate probabilistic reference attention maps, which align textual and visual modalities more effectively while incorporating uncertainty estimates, as compared to their deterministic counterparts. Experiments on benchmark test problems demonstrate that PARIC enhances prediction accuracy, mitigates bias, ensures consistent predictions, and improves robustness across various datasets.

Paperid: 5012, https://arxiv.org/pdf/2503.09968.pdf

Abstract:
Recently, a task of Single-Domain Generalized Object Detection (Single-DGOD) is proposed, aiming to generalize a detector to multiple unknown domains never seen before during training. Due to the unavailability of target-domain data, some methods leverage the multimodal capabilities of vision-language models, using textual prompts to estimate cross-domain information, enhancing the model's generalization capability. These methods typically use a single textual prompt, often referred to as the one-step prompt method. However, when dealing with complex styles such as the combination of rain and night, we observe that the performance of the one-step prompt method tends to be relatively weak. The reason may be that many scenes incorporate not just a single style but a combination of multiple styles. The one-step prompt method may not effectively synthesize combined information involving various styles. To address this limitation, we propose a new method, i.e., Style Evolving along Chain-of-Thought, which aims to progressively integrate and expand style information along the chain of thought, enabling the continual evolution of styles. Specifically, by progressively refining style descriptions and guiding the diverse evolution of styles, this approach enables more accurate simulation of various style characteristics and helps the model gradually learn and adapt to subtle differences between styles. Additionally, it exposes the model to a broader range of style features with different data distributions, thereby enhancing its generalization capability in unseen domains. The significant performance gains over five adverse-weather scenarios and the Real to Art benchmark demonstrate the superiorities of our method.

Paperid: 5013, https://arxiv.org/pdf/2503.09474.pdf

Abstract:
Image-guided surgery demands adaptive, real-time decision support, yet static AI models struggle with structured task planning and providing interactive guidance. Large vision-language models (VLMs) offer a promising solution by enabling dynamic task planning and predictive decision support. We introduce SurgicalVLM-Agent, an AI co-pilot for image-guided pituitary surgery, capable of conversation, planning, and task execution. The agent dynamically processes surgeon queries and plans the tasks such as MRI tumor segmentation, endoscope anatomy segmentation, overlaying preoperative imaging with intraoperative views, instrument tracking, and surgical visual question answering (VQA). To enable structured task planning, we develop the PitAgent dataset, a surgical context-aware dataset covering segmentation, overlaying, instrument localization, tool tracking, tool-tissue interactions, phase identification, and surgical activity recognition. Additionally, we propose FFT-GaLore, a fast Fourier transform (FFT)-based gradient projection technique for efficient low-rank adaptation, optimizing fine-tuning for LLaMA 3.2 in surgical environments. We validate SurgicalVLM-Agent by assessing task planning and prompt generation on our PitAgent dataset and evaluating zero-shot VQA using a public pituitary dataset. Results demonstrate state-of-the-art performance in task planning and query interpretation, with highly semantically meaningful VQA responses, advancing AI-driven surgical assistance.

Paperid: 5014, https://arxiv.org/pdf/2503.09394.pdf

Abstract:
Test-time adaptation (TTA) is crucial in maintaining performance of Vision Language Models (VLMs) when facing distribution shifts, particularly when the source data or target labels are inaccessible. Existing TTA methods predominantly leverage the output probability distribution of CLIP for feature evaluation, resulting in biases under domain shifts, which cause misclassified features due to text priors or incorrect textual associations. To address these issues, we propose \underline{B}idirectional Prototype-Reward co-Evolution (BPRE), a novel VLMs framework with TTA that integrates feature quality assessment with prototype evolution via a synergistic feedback loop. First, the Multi-dimensional Quality-aware Reward Module (MQRM) is designed to evaluate feature quality and guide prototype refinement precisely. The continuous refinement of prototype quality via Prototype-Reward Interactive Evolution (PRIE) enhances the computation more robust. Through this bidirectional interaction, the precision of rewards and prototype evolution mutually reinforce each other, forming a self-evolving feedback cycle. Extensive experiments conducted on 15 diverse recognition datasets demonstrate that our model consistently achieves superior performance compared to other SOTA methods, and advances VLM generalization capabilities through emphasizing comprehensive feature evaluation.

Paperid: 5015, https://arxiv.org/pdf/2503.09173.pdf

Abstract:
Long-term planning for robots operating in domestic environments poses unique challenges due to the interactions between humans, objects, and spaces. Recent advancements in trajectory planning have leveraged vision-language models (VLMs) to extract contextual information for robots operating in real-world environments. While these methods achieve satisfying performance, they do not explicitly model human activities. Such activities influence surrounding objects and reshape spatial constraints. This paper presents a novel approach to trajectory planning that integrates human preferences, activities, and spatial context through an enriched 3D scene graph (3DSG) representation. By incorporating activity-based relationships, our method captures the spatial impact of human actions, leading to more context-sensitive trajectory adaptation. Preliminary results demonstrate that our approach effectively assigns costs to spaces influenced by human activities, ensuring that the robot trajectory remains contextually appropriate and sensitive to the ongoing environment. This balance between task efficiency and social appropriateness enhances context-aware human-robot interactions in domestic settings. Future work includes implementing a full planning pipeline and conducting user studies to evaluate trajectory acceptability.

Paperid: 5016, https://arxiv.org/pdf/2503.09024.pdf

Abstract:
This paper introduces and tests a framework integrating traffic regulation compliance into automated driving systems (ADS). The framework enables ADS to follow traffic laws and make informed decisions based on the driving environment. Using RGB camera inputs and a vision-language model (VLM), the system generates descriptive text to support a regulation-aware decision-making process, ensuring legal and safe driving practices. This information is combined with a machine-readable ADS regulation database to guide future driving plans within legal constraints. Key features include: 1) a regulation database supporting ADS decision-making, 2) an automated process using sensor input for regulation-aware path planning, and 3) validation in both simulated and real-world environments. Particularly, the real-world vehicle tests not only assess the framework's performance but also evaluate the potential and challenges of VLMs to solve complex driving problems by integrating detection, reasoning, and planning. This work enhances the legality, safety, and public trust in ADS, representing a significant step forward in the field.

Paperid: 5017, https://arxiv.org/pdf/2503.08525.pdf

Abstract:
Reinforcement learning with verifiable outcome rewards (RLVR) has effectively scaled up chain-of-thought (CoT) reasoning in large language models (LLMs). Yet, its efficacy in training vision-language model (VLM) agents for goal-directed action reasoning in visual environments is less established. This work investigates this problem through extensive experiments on complex card games, such as 24 points, and embodied tasks from ALFWorld. We find that when rewards are based solely on action outcomes, RL fails to incentivize CoT reasoning in VLMs, instead leading to a phenomenon we termed thought collapse, characterized by a rapid loss of diversity in the agent's thoughts, state-irrelevant and incomplete reasoning, and subsequent invalid actions, resulting in negative rewards. To counteract thought collapse, we highlight the necessity of process guidance and propose an automated corrector that evaluates and refines the agent's reasoning at each RL step. This simple and scalable GTR (Guided Thought Reinforcement) framework trains reasoning and action simultaneously without the need for dense, per-step human labeling. Our experiments demonstrate that GTR significantly enhances the performance and generalization of the LLaVA-7b model across various visual environments, achieving 3-5 times higher task success rates compared to SoTA models with notably smaller model sizes.

Paperid: 5018, https://arxiv.org/pdf/2503.05464.pdf

Abstract:
Automating teaching presents unique challenges, as replicating human interaction and adaptability is complex. Automated systems cannot often provide nuanced, real-time feedback that aligns with students' individual learning paces or comprehension levels, which can hinder effective support for diverse needs. This is especially challenging in fields where abstract concepts require adaptive explanations. In this paper, we propose a vision language retrieval augmented generation (named VL-RAG) system that has the potential to bridge this gap by delivering contextually relevant, visually enriched responses that can enhance comprehension. By leveraging a database of tailored answers and images, the VL-RAG system can dynamically retrieve information aligned with specific questions, creating a more interactive and engaging experience that fosters deeper understanding and active student participation. It allows students to explore concepts visually and verbally, promoting deeper understanding and reducing the need for constant human oversight while maintaining flexibility to expand across different subjects and course material.

Paperid: 5019, https://arxiv.org/pdf/2503.04457.pdf

Abstract:
Vision-language models (VLMs) have achieved remarkable advancements, capitalizing on the impressive capabilities of large language models (LLMs) across diverse tasks. Despite this, a critical challenge known as hallucination occurs when models overconfidently describe objects or attributes absent from the image, a problem exacerbated by the tendency of VLMs to rely on linguistic priors. This limitation reduces model reliability in high-stakes applications. In this work, we have observed the characteristic of logits' continuity consistency enhancement and introduced a straightforward and efficient method, Cross-Temporal Prediction Connection (TPC), designed to enhance the semantic consistency of logits by connecting them temporally across timesteps. TPC amplifies information flow and improves coherence, effectively reducing hallucination. Extensive experiments show that TPC surpasses existing representatives, delivering superior performance in both accuracy and efficiency while maintaining robustness in open-ended text generation tasks.

Paperid: 5020, https://arxiv.org/pdf/2502.18290.pdf

Abstract:
Self-supervised learning (SSL) vision encoders learn high-quality image representations and thus have become a vital part of developing vision modality of large vision language models (LVLMs). Due to the high cost of training such encoders, pre-trained encoders are widely shared and deployed into many LVLMs, which are security-critical or bear societal significance. Under this practical scenario, we reveal a new backdoor threat that significant visual hallucinations can be induced into these LVLMs by merely compromising vision encoders. Because of the sharing and reuse of these encoders, many downstream LVLMs may inherit backdoor behaviors from encoders, leading to widespread backdoors. In this work, we propose BadVision, the first method to exploit this vulnerability in SSL vision encoders for LVLMs with novel trigger optimization and backdoor learning techniques. We evaluate BadVision on two types of SSL encoders and LVLMs across eight benchmarks. We show that BadVision effectively drives the LVLMs to attacker-chosen hallucination with over 99% attack success rate, causing a 77.6% relative visual understanding error while maintaining the stealthiness. SoTA backdoor detection methods cannot detect our attack effectively.

Paperid: 5021, https://arxiv.org/pdf/2502.17540.pdf

Abstract:
Generating accurate and concise textual summaries from multimodal documents is challenging, especially when dealing with visually complex content like scientific posters. We introduce PosterSum, a novel benchmark to advance the development of vision-language models that can understand and summarize scientific posters into research paper abstracts. Our dataset contains 16,305 conference posters paired with their corresponding abstracts as summaries. Each poster is provided in image format and presents diverse visual understanding challenges, such as complex layouts, dense text regions, tables, and figures. We benchmark state-of-the-art Multimodal Large Language Models (MLLMs) on PosterSum and demonstrate that they struggle to accurately interpret and summarize scientific posters. We propose Segment & Summarize, a hierarchical method that outperforms current MLLMs on automated metrics, achieving a 3.14% gain in ROUGE-L. This will serve as a starting point for future research on poster summarization.

Paperid: 5022, https://arxiv.org/pdf/2502.14908.pdf

Abstract:
Vision language models (VLM) demonstrate sophisticated multimodal reasoning yet are prone to hallucination when confronted with knowledge conflicts, impeding their deployment in information-sensitive contexts. While existing research addresses robustness in unimodal models, the multimodal domain lacks systematic investigation of cross-modal knowledge conflicts. This research introduces \segsub, a framework for applying targeted image perturbations to investigate VLM resilience against knowledge conflicts. Our analysis reveals distinct vulnerability patterns: while VLMs are robust to parametric conflicts (20% adherence rates), they exhibit significant weaknesses in identifying counterfactual conditions (<30% accuracy) and resolving source conflicts (<1% accuracy). Correlations between contextual richness and hallucination rate (r = -0.368, p = 0.003) reveal the kinds of images that are likely to cause hallucinations. Through targeted fine-tuning on our benchmark dataset, we demonstrate improvements in VLM knowledge conflict detection, establishing a foundation for developing hallucination-resilient multimodal systems in information-sensitive environments.

Paperid: 5023, https://arxiv.org/pdf/2502.13870.pdf

Abstract:
Large language models (LLMs) have revolutionized machine learning due to their ability to capture complex interactions between input features. Popular post-hoc explanation methods like SHAP provide marginal feature attributions, while their extensions to interaction importances only scale to small input lengths ($\approx 20$). We propose Spectral Explainer (SPEX), a model-agnostic interaction attribution algorithm that efficiently scales to large input lengths ($\approx 1000)$. SPEX exploits underlying natural sparsity among interactions -- common in real-world data -- and applies a sparse Fourier transform using a channel decoding algorithm to efficiently identify important interactions. We perform experiments across three difficult long-context datasets that require LLMs to utilize interactions between inputs to complete the task. For large inputs, SPEX outperforms marginal attribution methods by up to 20% in terms of faithfully reconstructing LLM outputs. Further, SPEX successfully identifies key features and interactions that strongly influence model output. For one of our datasets, HotpotQA, SPEX provides interactions that align with human annotations. Finally, we use our model-agnostic approach to generate explanations to demonstrate abstract reasoning in closed-source LLMs (GPT-4o mini) and compositional reasoning in vision-language models.

Paperid: 5024, https://arxiv.org/pdf/2502.13836.pdf

Abstract:
Large Language Models (LLMs) demonstrate remarkable capabilities in question answering (QA), but metrics for assessing their reliance on memorization versus retrieval remain underdeveloped. Moreover, while finetuned models are state-of-the-art on closed-domain tasks, general-purpose models like GPT-4o exhibit strong zero-shot performance. This raises questions about the trade-offs between memorization, generalization, and retrieval. In this work, we analyze the extent to which multimodal retrieval-augmented VLMs memorize training data compared to baseline VLMs. Using the WebQA benchmark, we contrast finetuned models with baseline VLMs on multihop retrieval and question answering, examining the impact of finetuning on data memorization. To quantify memorization in end-to-end retrieval and QA systems, we propose several proxy metrics by investigating instances where QA succeeds despite retrieval failing. In line with existing work, we find that finetuned models rely more heavily on memorization than retrieval-augmented VLMs, and achieve higher accuracy as a result (72% vs 52% on WebQA test set). Finally, we present the first empirical comparison of the parametric effect between text and visual modalities. Here, we find that image-based questions have parametric response rates that are consistently 15-25% higher than for text-based questions in the WebQA dataset. As such, our measures pose a challenge for future work, both to account for differences in model memorization across different modalities and more generally to reconcile memorization and generalization in joint Retrieval-QA tasks.

Paperid: 5025, https://arxiv.org/pdf/2502.11603.pdf

Abstract:
Large Language Models (LLMs) exhibit strong natural language processing capabilities but also inherit and amplify societal biases, including gender bias, raising fairness concerns. Existing debiasing methods face significant limitations: parameter tuning requires access to model weights, prompt-based approaches often degrade model utility, and optimization-based techniques lack generalizability. To address these challenges, we propose DR.GAP (Demonstration and Reasoning for Gender-Aware Prompting), an automated and model-agnostic approach that mitigates gender bias while preserving model performance. DR.GAP selects bias-revealing examples and generates structured reasoning to guide models toward more impartial responses. Extensive experiments on coreference resolution and QA tasks across multiple LLMs (GPT-3.5, Llama3, and Llama2-Alpaca) demonstrate its effectiveness, generalization ability, and robustness. DR.GAP can generalize to vision-language models (VLMs), achieving significant bias reduction.

Paperid: 5026, https://arxiv.org/pdf/2502.07957.pdf

Abstract:
While recent work has found that vision-language models trained under the Contrastive Language Image Pre-training (CLIP) framework contain intrinsic social biases, the extent to which different upstream pre-training features of the framework relate to these biases, and hence how intrinsic bias and downstream performance are connected has been unclear. In this work, we present the largest comprehensive analysis to-date of how the upstream pre-training factors and downstream performance of CLIP models relate to their intrinsic biases. Studying 131 unique CLIP models, trained on 26 datasets, using 55 architectures, and in a variety of sizes, we evaluate bias in each model using 26 well-established unimodal and cross-modal principled Embedding Association Tests. We find that the choice of pre-training dataset is the most significant upstream predictor of bias, whereas architectural variations have minimal impact. Additionally, datasets curated using sophisticated filtering techniques aimed at enhancing downstream model performance tend to be associated with higher levels of intrinsic bias. Finally, we observe that intrinsic bias is often significantly correlated with downstream performance ($0.3 \leq r \leq 0.8$), suggesting that models optimized for performance inadvertently learn to amplify representational biases. Comparisons between unimodal and cross-modal association tests reveal that social group bias depends heavily on the modality. Our findings imply that more sophisticated strategies are needed to address intrinsic model bias for vision-language models across the entire model development pipeline.

Paperid: 5027, https://arxiv.org/pdf/2502.07905.pdf

Abstract:
Multimodal Large Language Models (MLLMs) represent the cutting edge of AI technology, with DeepSeek models emerging as a leading open-source alternative offering competitive performance to closed-source systems. While these models demonstrate remarkable capabilities, their vision-language integration mechanisms introduce specific vulnerabilities. We implement an adapted embedding manipulation attack on DeepSeek Janus that induces targeted visual hallucinations through systematic optimization of image embeddings. Through extensive experimentation across COCO, DALL-E 3, and SVIT datasets, we achieve hallucination rates of up to 98.0% while maintaining high visual fidelity (SSIM > 0.88) of the manipulated images on open-ended questions. Our analysis demonstrates that both 1B and 7B variants of DeepSeek Janus are susceptible to these attacks, with closed-form evaluation showing consistently higher hallucination rates compared to open-ended questioning. We introduce a novel multi-prompt hallucination detection framework using LLaMA-3.1 8B Instruct for robust evaluation. The implications of these findings are particularly concerning given DeepSeek's open-source nature and widespread deployment potential. This research emphasizes the critical need for embedding-level security measures in MLLM deployment pipelines and contributes to the broader discussion of responsible AI implementation.

Paperid: 5028, https://arxiv.org/pdf/2502.01056.pdf

Abstract:
Large Visual Language Models (LVLMs) integrate visual and linguistic modalities, exhibiting exceptional performance across various multimodal tasks. Nevertheless, LVLMs remain vulnerable to the issue of object hallucinations. Previous efforts to mitigate this issue focus on supervised fine-tuning (SFT) or incorporating external knowledge, both of which entail significant costs related to training and the acquisition of external data. To address these challenges, we propose a novel model-agnostic approach termed Internal Fact-based Contrastive Decoding (IFCD), designed to mitigate and suppress hallucinations during the inference process of LVLMs by exploiting the LVLMs' own hallucinations. IFCD is grounded in experimental observations that alterations to the LVLMs' internal representations tend to amplify hallucinations caused by language bias. By contrasting disturbed distribution, IFCD calibrates the LVLMs' output and effectively removes the hallucinatory logits from the final predictions. Experimental results validate that IFCD significantly alleviates both object-level and attribute-level hallucinations while achieving an average 9% accuracy improvement on POPE and 8% accuracy improvement on MME object hallucinations subset compared with direct decoding, respectively.

Paperid: 5029, https://arxiv.org/pdf/2502.00987.pdf

Abstract:
Low-Rank Adaptation (LoRA) and its variants have shown impressive results in reducing the number of trainable parameters and memory requirements of large transformer networks while maintaining fine-tuning performance. The low-rank nature of the weight update inherently limits the representation power of fine-tuned models, however, thus potentially compromising performance on complex tasks. This raises a critical question: when a performance gap between LoRA and standard fine-tuning is observed, is it due to the reduced number of trainable parameters or the rank deficiency? This paper aims to answer this question by introducing RandLoRA, a parameter-efficient method that performs full-rank updates using a learned linear combinations of low-rank, non-trainable random matrices. Our method limits the number of trainable parameters by restricting optimization to diagonal scaling matrices applied to the fixed random matrices. This allows us to effectively overcome the low-rank limitations while maintaining parameter and memory efficiency during training. Through extensive experimentation across vision, language, and vision-language benchmarks, we systematically evaluate the limitations of LoRA and existing random basis methods. Our findings reveal that full-rank updates are beneficial across vision and language tasks individually, and even more so for vision-language tasks, where RandLoRA significantly reduces -- and sometimes eliminates -- the performance gap between standard fine-tuning and LoRA, demonstrating its efficacy.

Paperid: 5030, https://arxiv.org/pdf/2502.00717.pdf

Abstract:
Hallucination has been a long-standing and inevitable problem that hinders the application of Large Vision-Language Models (LVLMs) in domains that require high reliability. Various methods focus on improvement depending on data annotations or training strategies, yet place less emphasis on LLM's inherent problems. To fill this gap, we delve into the attention mechanism of the decoding process in the LVLM. Intriguingly, our investigation uncovers the prevalent attention redundancy within the hierarchical architecture of the LVLM, manifesting as overextended image processing in deep layers and an overabundance of non-essential image tokens. Stemming from the observation, we thus propose MINT, a novel training-free decoding strategy, MItigating hallucinations via tokeN reducTion. Specifically, we dynamically intensify the LVLM's local perception capability by masking its attention to irrelevant image tokens. In addition, we use contrastive decoding that pushes the model to focus more on those key image regions. Our full method aims to guide the model in concentrating more on key visual elements during generation. Extensive experimental results on several popular public benchmarks show that our approach achieves a 4% improvement in mitigating hallucinations caused by distracted perception compared to original models. Meanwhile, our approach is demonstrated to make the model perceive 5% more visual points even though we reduce a suite of image tokens.

Paperid: 5031, https://arxiv.org/pdf/2502.00711.pdf

Abstract:
Visual reasoning refers to the task of solving questions about visual information. Current visual reasoning methods typically employ pre-trained vision-language model (VLM) strategies or deep neural network approaches. However, existing efforts are constrained by limited reasoning interpretability, while hindering by the phenomenon of underspecification in the question text. Additionally, the absence of fine-grained visual knowledge limits the precise understanding of subject behavior in visual reasoning tasks. To address these issues, we propose VIKSER (Visual Knowledge-Driven Self-Reinforcing Reasoning Framework). Specifically, VIKSER, trained using knowledge distilled from large language models, extracts fine-grained visual knowledge with the assistance of visual relationship detection techniques. Subsequently, VIKSER utilizes fine-grained visual knowledge to paraphrase the question with underspecification. Additionally, we design a novel prompting method called Chain-of-Evidence (CoE), which leverages the power of "evidence for reasoning" to endow VIKSER with interpretable reasoning capabilities. Meanwhile, the integration of self-reflection technology empowers VIKSER with the ability to learn and improve from its mistakes. Experiments conducted on widely used datasets demonstrate that VIKSER achieves new state-of-the-art (SOTA) results in relevant tasks. Moreover, VIKSER achieves performance on par with leading proprietary models, such as the latest ChatGPT-5.

Paperid: 5032, https://arxiv.org/pdf/2501.17654.pdf

Abstract:
Social media's global reach amplifies the spread of information, highlighting the need for robust Natural Language Processing tasks like stance detection across languages and modalities. Prior research predominantly focuses on text-only inputs, leaving multimodal scenarios, such as those involving both images and text, relatively underexplored. Meanwhile, the prevalence of multimodal posts has increased significantly in recent years. Although state-of-the-art Vision-Language Models (VLMs) show promise, their performance on multimodal and multilingual stance detection tasks remains largely unexamined. This paper evaluates state-of-the-art VLMs on a newly extended dataset covering seven languages and multimodal inputs, investigating their use of visual cues, language-specific performance, and cross-modality interactions. Our results show that VLMs generally rely more on text than images for stance detection and this trend persists across languages. Additionally, VLMs rely significantly more on text contained within the images than other visual content. Regarding multilinguality, the models studied tend to generate consistent predictions across languages whether they are explicitly multilingual or not, although there are outliers that are incongruous with macro F1, language support, and model size.

Paperid: 5033, https://arxiv.org/pdf/2501.13859.pdf

Abstract:
Compositional Zero-Shot Learning (CZSL) aims to recognize novel attribute-object compositions by leveraging knowledge from seen compositions. Current methods align textual prototypes with visual features via Vision-Language Models (VLMs), but suffer from two limitations: (1) modality gaps hinder the discrimination of semantically similar pairs, and (2) single-modal textual prototypes lack fine-grained visual cues. In this paper, we introduce Visual Proxy Learning, a method that reduces modality gaps and enhances compositional generalization. We initialize visual proxies for attributes, objects, and their compositions using text representations and optimize the visual space to capture fine-grained cues, improving visual representations. Additionally, we propose Cross-Modal Joint Learning (CMJL), which imposes cross-modal constraints between the text-image and fine-grained visual spaces, improving generalization for unseen compositions and discriminating similar pairs. Experiments show state-of-the-art performance in closed-world scenarios and competitive results in open-world settings across four CZSL benchmarks, demonstrating the effectiveness of our approach in compositional generalization.

Paperid: 5034, https://arxiv.org/pdf/2501.13507.pdf

Abstract:
In this paper, we address the problem of manipulating multi-particle aggregates using a bimanual robotic system. Our approach enables the autonomous transport of dispersed particles through a series of shaping and pushing actions using robotically-controlled tools. Achieving this advanced manipulation capability presents two key challenges: high-level task planning and trajectory execution. For task planning, we leverage Vision Language Models (VLMs) to enable primitive actions such as tool affordance grasping and non-prehensile particle pushing. For trajectory execution, we represent the evolving particle aggregate's contour using truncated Fourier series, providing efficient parametrization of its closed shape. We adaptively compute trajectory waypoints based on group cohesion and the geometric centroid of the aggregate, accounting for its spatial distribution and collective motion. Through real-world experiments, we demonstrate the effectiveness of our methodology in actively shaping and manipulating multi-particle aggregates while maintaining high system cohesion.

Paperid: 5035, https://arxiv.org/pdf/2501.12433.pdf

Abstract:
Animal stereotypes are deeply embedded in human culture and language. They often shape our perceptions and expectations of various species. Our study investigates how animal stereotypes manifest in vision-language models during the task of image generation. Through targeted prompts, we explore whether DALL-E perpetuates stereotypical representations of animals, such as "owls as wise," "foxes as unfaithful," etc. Our findings reveal significant stereotyped instances where the model consistently generates images aligned with cultural biases. The current work is the first of its kind to examine animal stereotyping in vision-language models systematically and to highlight a critical yet underexplored dimension of bias in AI-generated visual content.

Paperid: 5036, https://arxiv.org/pdf/2501.07396.pdf

Abstract:
Automatic target recognition (ATR) plays a critical role in tasks such as navigation and surveillance, where safety and accuracy are paramount. In extreme use cases, such as military applications, these factors are often challenged due to the presence of unknown terrains, environmental conditions, and novel object categories. Current object detectors, including open-world detectors, lack the ability to confidently recognize novel objects or operate in unknown environments, as they have not been exposed to these new conditions. However, Large Vision-Language Models (LVLMs) exhibit emergent properties that enable them to recognize objects in varying conditions in a zero-shot manner. Despite this, LVLMs struggle to localize objects effectively within a scene. To address these limitations, we propose a novel pipeline that combines the detection capabilities of open-world detectors with the recognition confidence of LVLMs, creating a robust system for zero-shot ATR of novel classes and unknown domains. In this study, we compare the performance of various LVLMs for recognizing military vehicles, which are often underrepresented in training datasets. Additionally, we examine the impact of factors such as distance range, modality, and prompting methods on the recognition performance, providing insights into the development of more reliable ATR systems for novel conditions and classes.

Paperid: 5037, https://arxiv.org/pdf/2501.06252.pdf

Abstract:
Self-adaptive large language models (LLMs) aim to solve the challenges posed by traditional fine-tuning methods, which are often computationally intensive and static in their ability to handle diverse tasks. We introduce Transformer-Squared, a novel self-adaptation framework that adapts LLMs for unseen tasks in real-time by selectively adjusting only the singular components of their weight matrices. During inference, Transformer-Squared employs a two-pass mechanism: first, a dispatch system identifies the task properties, and then task-specific 'expert' vectors, trained using reinforcement learning, are dynamically mixed to obtain targeted behavior for the incoming prompt. Our method consistently outperforms ubiquitous approaches such as LoRA, with fewer parameters and greater efficiency. Furthermore, Transformer-Squared demonstrates versatility across different LLM architectures and modalities, including vision-language tasks. Transformer-Squared represents a significant leap forward, offering a scalable, efficient solution for enhancing the adaptability and task-specific performance of LLMs, paving the way for truly dynamic, self-organizing AI systems.

Paperid: 5038, https://arxiv.org/pdf/2501.04695.pdf

Abstract:
Retrieval-augmented generation (RAG) enhances large language models (LLMs) by incorporating external knowledge to generate a response within a context with improved accuracy and reduced hallucinations. However, multi-modal RAG systems face unique challenges: (i) the retrieval process may select irrelevant entries to user query (e.g., images, documents), and (ii) vision-language models or multi-modal language models like GPT-4o may hallucinate when processing these entries to generate RAG output. In this paper, we aim to address the first challenge, i.e, improving the selection of relevant context from the knowledge-base in retrieval phase of the multi-modal RAG. Specifically, we leverage the relevancy score (RS) measure designed in our previous work for evaluating the RAG performance to select more relevant entries in retrieval process. The retrieval based on embeddings, say CLIP-based embedding, and cosine similarity usually perform poorly particularly for multi-modal data. We show that by using a more advanced relevancy measure, one can enhance the retrieval process by selecting more relevant pieces from the knowledge-base and eliminate the irrelevant pieces from the context by adaptively selecting up-to-$k$ entries instead of fixed number of entries. Our evaluation using COCO dataset demonstrates significant enhancement in selecting relevant context and accuracy of the generated response.

Paperid: 5039, https://arxiv.org/pdf/2501.04268.pdf

Abstract:
Zero-shot generalization across various robots, tasks and environments remains a significant challenge in robotic manipulation. Policy code generation methods use executable code to connect high-level task descriptions and low-level action sequences, leveraging the generalization capabilities of large language models and atomic skill libraries. In this work, we propose Robotic Programmer (RoboPro), a robotic foundation model, enabling the capability of perceiving visual information and following free-form instructions to perform robotic manipulation with policy code in a zero-shot manner. To address low efficiency and high cost in collecting runtime code data for robotic tasks, we devise Video2Code to synthesize executable code from extensive videos in-the-wild with off-the-shelf vision-language model and code-domain large language model. Extensive experiments show that RoboPro achieves the state-of-the-art zero-shot performance on robotic manipulation in both simulators and real-world environments. Specifically, the zero-shot success rate of RoboPro on RLBench surpasses the state-of-the-art model GPT-4o by 11.6%, which is even comparable to a strong supervised training baseline. Furthermore, RoboPro is robust to variations on API formats and skill sets.

Paperid: 5040, https://arxiv.org/pdf/2501.03995.pdf

Abstract:
Retrieval-augmented generation (RAG) improves large language models (LLMs) by using external knowledge to guide response generation, reducing hallucinations. However, RAG, particularly multi-modal RAG, can introduce new hallucination sources: (i) the retrieval process may select irrelevant pieces (e.g., documents, images) as raw context from the database, and (ii) retrieved images are processed into text-based context via vision-language models (VLMs) or directly used by multi-modal language models (MLLMs) like GPT-4o, which may hallucinate. To address this, we propose a novel framework to evaluate the reliability of multi-modal RAG using two performance measures: (i) the relevancy score (RS), assessing the relevance of retrieved entries to the query, and (ii) the correctness score (CS), evaluating the accuracy of the generated response. We train RS and CS models using a ChatGPT-derived database and human evaluator samples. Results show that both models achieve ~88% accuracy on test data. Additionally, we construct a 5000-sample human-annotated database evaluating the relevancy of retrieved pieces and the correctness of response statements. Our RS model aligns with human preferences 20% more often than CLIP in retrieval, and our CS model matches human preferences ~91% of the time. Finally, we assess various RAG systems' selection and generation performances using RS and CS.

Paperid: 5041, https://arxiv.org/pdf/2501.00569.pdf

Abstract:
Despite recent advances in Vision-Language Models (VLMs), they may over-rely on visual language priors existing in their training data rather than true visual reasoning. To investigate this, we introduce ViLP, a benchmark featuring deliberately out-of-distribution images synthesized via image generation models and out-of-distribution Q&A pairs. Each question in ViLP is coupled with three potential answers and three corresponding images: one that can be resolved by text priors alone and two that demand visual reasoning. Although, humans achieve near-perfect accuracy, modern VLMs falter; for instance, GPT-4 achieves only 66.17% on ViLP. To alleviate this, we propose a self-improving framework in which models generate new VQA data, then apply pixel-level and semantic corruptions to form "good-bad" image pairs for self-training. Our training objectives compel VLMs to focus more on the actual visual inputs, and we demonstrate their effectiveness in boosting the performance of open-source VLMs, including LLaVA-v1.5 and Cambrian.

Paperid: 5042, https://arxiv.org/pdf/2412.18090.pdf

Abstract:
Small object detection aims to localize and classify small objects within images. With recent advances in large-scale vision-language pretraining, finetuning pretrained object detection models has emerged as a promising approach. However, finetuning large models is computationally and memory expensive. To address this issue, this paper introduces multi-point positional insertion (MPI) tuning, a parameter-efficient finetuning (PEFT) method for small object detection. Specifically, MPI incorporates multiple positional embeddings into a frozen pretrained model, enabling the efficient detection of small objects by providing precise positional information to latent features. Through experiments, we demonstrated the effectiveness of the proposed method on the SODA-D dataset. MPI performed comparably to conventional PEFT methods, including CoOp and VPT, while significantly reducing the number of parameters that need to be tuned.

Paperid: 5043, https://arxiv.org/pdf/2412.17417.pdf

Abstract:
Multimodal large language models (MLLMs) have significantly advanced tasks like caption generation and visual question answering by integrating visual and textual data. However, they sometimes produce misleading or hallucinate content due to discrepancies between their pre-training data and real user prompts. Existing approaches using Direct Preference Optimization (DPO) in vision-language tasks often rely on strong models like GPT-4 or CLIP to determine positive and negative responses. Here, we propose a new framework in generating synthetic data using a reward model as a proxy of human preference for effective multimodal alignment with DPO training. The resulting DPO dataset ranges from 2K to 9K image-text pairs, was evaluated on LLaVA-v1.5-7B, where our approach demonstrated substantial improvements in both the trustworthiness and reasoning capabilities of the base model across multiple hallucination and vision-language benchmark. The experiment results indicate that integrating selected synthetic data, such as from generative and rewards models can effectively reduce reliance on human-annotated data while enhancing MLLMs' alignment capability, offering a scalable solution for safer deployment.

Paperid: 5044, https://arxiv.org/pdf/2412.13558.pdf

Abstract:
Recent medical vision-language models (VLMs) have shown promise in 2D medical image interpretation. However extending them to 3D medical imaging has been challenging due to computational complexities and data scarcity. Although a few recent VLMs specified for 3D medical imaging have emerged, all are limited to learning volumetric representation of a 3D medical image as a set of sub-volumetric features. Such process introduces overly correlated representations along the z-axis that neglect slice-specific clinical details, particularly for 3D medical images where adjacent slices have low redundancy. To address this limitation, we introduce MS-VLM that mimic radiologists' workflow in 3D medical image interpretation. Specifically, radiologists analyze 3D medical images by examining individual slices sequentially and synthesizing information across slices and views. Likewise, MS-VLM leverages self-supervised 2D transformer encoders to learn a volumetric representation that capture inter-slice dependencies from a sequence of slice-specific features. Unbound by sub-volumetric patchification, MS-VLM is capable of obtaining useful volumetric representations from 3D medical images with any slice length and from multiple images acquired from different planes and phases. We evaluate MS-VLM on publicly available chest CT dataset CT-RATE and in-house rectal MRI dataset. In both scenarios, MS-VLM surpasses existing methods in radiology report generation, producing more coherent and clinically relevant reports. These findings highlight the potential of MS-VLM to advance 3D medical image interpretation and improve the robustness of medical VLMs.

Paperid: 5045, https://arxiv.org/pdf/2412.12793.pdf

Abstract:
Noisy labels threaten the robustness of few-shot learning (FSL) due to the inexact features in a new domain. CLIP, a large-scale vision-language model, performs well in FSL on image-text embedding similarities, but it is susceptible to misclassification caused by noisy labels. How to enhance domain generalization of CLIP on noisy data within FSL tasks is a critical challenge. In this paper, we provide a novel view to mitigate the influence of noisy labels, CLIP-based Robust Few-shot learning (CRoF). CRoF is a general plug-in module for CLIP-based models. To avoid misclassification and confused label embedding, we design the few-shot task-oriented prompt generator to give more discriminative descriptions of each category. The proposed prompt achieves larger distances of inter-class textual embedding. Furthermore, rather than fully trusting zero-shot classification by CLIP, we fine-tune CLIP on noisy few-shot data in a new domain with a weighting strategy like label-smooth. The weights for multiple potentially correct labels consider the relationship between CLIP's prior knowledge and original label information to ensure reliability. Our multiple label loss function further supports robust training under this paradigm. Comprehensive experiments show that CRoF, as a plug-in, outperforms fine-tuned and vanilla CLIP models on different noise types and noise ratios.

Paperid: 5046, https://arxiv.org/pdf/2412.10726.pdf

Abstract:
The rapid advancement of Vision-Language Models (VLMs) has significantly advanced the development of Embodied Question Answering (EQA), enhancing agents' abilities in language understanding and reasoning within complex and realistic scenarios. However, EQA in real-world scenarios remains challenging, as human-posed questions often contain noise that can interfere with an agent's exploration and response, bringing challenges especially for language beginners and non-expert users. To address this, we introduce a NoisyEQA benchmark designed to evaluate an agent's ability to recognize and correct noisy questions. This benchmark introduces four common types of noise found in real-world applications: Latent Hallucination Noise, Memory Noise, Perception Noise, and Semantic Noise generated through an automated dataset creation framework. Additionally, we also propose a 'Self-Correction' prompting mechanism and a new evaluation metric to enhance and measure both noise detection capability and answer quality. Our comprehensive evaluation reveals that current EQA agents often struggle to detect noise in questions, leading to responses that frequently contain erroneous information. Through our Self-Correct Prompting mechanism, we can effectively improve the accuracy of agent answers.

Paperid: 5047, https://arxiv.org/pdf/2412.10566.pdf

Abstract:
Editing complex visual content based on ambiguous instructions remains a challenging problem in vision-language modeling. While existing models can contextualize content, they often struggle to grasp the underlying intent within a reference image or scene, leading to misaligned edits. We introduce the Editing Vision-Language Model (EVLM), a system designed to interpret such instructions in conjunction with reference visuals, producing precise and context-aware editing prompts. Leveraging Chain-of-Thought (CoT) reasoning and KL-Divergence Target Optimization (KTO) alignment technique, EVLM captures subjective editing preferences without requiring binary labels. Fine-tuned on a dataset of 30,000 CoT examples, with rationale paths rated by human evaluators, EVLM demonstrates substantial improvements in alignment with human intentions. Experiments across image, video, 3D, and 4D editing tasks show that EVLM generates coherent, high-quality instructions, supporting a scalable framework for complex vision-language applications.

Paperid: 5048, https://arxiv.org/pdf/2412.06646.pdf

Abstract:
Recent advances in multimodal training have significantly improved the integration of image understanding and generation within a unified model. This study investigates how vision-language models (VLMs) handle image-understanding tasks, specifically focusing on how visual information is processed and transferred to the textual domain. We compare VLMs that generate both images and text with those that output only text, highlighting key differences in information flow. We find that in models with multimodal outputs, image and text embeddings are more separated within the residual stream. Additionally, models vary in how information is exchanged from visual to textual tokens. VLMs that only output text exhibit a distributed communication pattern, where information is exchanged through multiple image tokens. In contrast, models trained for image and text generation tend to rely on a single token that acts as a narrow gate for visual information. We demonstrate that ablating this single token significantly deteriorates performance on image understanding tasks. Furthermore, modifying this token enables effective steering of the image semantics, showing that targeted, local interventions can reliably control the model's global behavior.

Paperid: 5049, https://arxiv.org/pdf/2411.18180.pdf

Abstract:
Audio Descriptions (ADs) aim to provide a narration of a movie in text form, describing non-dialogue-related narratives, such as characters, actions, or scene establishment. Automatic generation of ADs remains challenging due to: i) the domain gap between movie-AD data and existing data used to train vision-language models, and ii) the issue of contextual redundancy arising from highly similar neighboring visual clips in a long movie. In this work, we propose DistinctAD, a novel two-stage framework for generating ADs that emphasize distinctiveness to produce better narratives. To address the domain gap, we introduce a CLIP-AD adaptation strategy that does not require additional AD corpora, enabling more effective alignment between movie and AD modalities at both global and fine-grained levels. In Stage-II, DistinctAD incorporates two key innovations: (i) a Contextual Expectation-Maximization Attention (EMA) module that reduces redundancy by extracting common bases from consecutive video clips, and (ii) an explicit distinctive word prediction loss that filters out repeated words in the context, ensuring the prediction of unique terms specific to the current AD. Comprehensive evaluations on MAD-Eval, CMD-AD, and TV-AD benchmarks demonstrate the superiority of DistinctAD, with the model consistently outperforming baselines, particularly in Recall@k/N, highlighting its effectiveness in producing high-quality, distinctive ADs.

Paperid: 5050, https://arxiv.org/pdf/2411.17606.pdf

Abstract:
This paper aims to address universal segmentation for image and video perception with the strong reasoning ability empowered by Visual Large Language Models (VLLMs). Despite significant progress in current unified segmentation methods, limitations in adaptation to both image and video scenarios, as well as the complex reasoning segmentation, make it difficult for them to handle various challenging instructions and achieve an accurate understanding of fine-grained vision-language correlations. We propose HyperSeg, the first VLLM-based universal segmentation model for pixel-level image and video perception, encompassing generic segmentation tasks and more complex reasoning perception tasks requiring powerful reasoning abilities and world knowledge. Besides, to fully leverage the recognition capabilities of VLLMs and the fine-grained visual information, HyperSeg incorporates hybrid entity recognition and fine-grained visual perceiver modules for various segmentation tasks. Combined with the temporal adapter, HyperSeg achieves a comprehensive understanding of temporal information. Experimental results validate the effectiveness of our insights in resolving universal image and video segmentation tasks, including the more complex reasoning perception tasks. Our code is available.

Paperid: 5051, https://arxiv.org/pdf/2411.16327.pdf

Abstract:
Infrared (IR) imaging offers advantages in several fields due to its unique ability of capturing content in extreme light conditions. However, the demanding hardware requirements of high-resolution IR sensors limit its widespread application. As an alternative, visible light can be used to synthesize IR images but this causes a loss of fidelity in image details and introduces inconsistencies due to lack of contextual awareness of the scene. This stems from a combination of using visible light with a standard dynamic range, especially under extreme lighting, and a lack of contextual awareness can result in pseudo-thermal-crossover artifacts. This occurs when multiple objects with similar temperatures appear indistinguishable in the training data, further exacerbating the loss of fidelity. To solve this challenge, this paper proposes CapHDR2IR, a novel framework incorporating vision-language models using high dynamic range (HDR) images as inputs to generate IR images. HDR images capture a wider range of luminance variations, ensuring reliable IR image generation in different light conditions. Additionally, a dense caption branch integrates semantic understanding, resulting in more meaningful and discernible IR outputs. Extensive experiments on the HDRT dataset show that the proposed CapHDR2IR achieves state-of-the-art performance compared with existing general domain transfer methods and those tailored for visible-to-infrared image translation.

Paperid: 5052, https://arxiv.org/pdf/2411.16018.pdf

Abstract:
Pre-trained Vision-language (VL) models, such as CLIP, have shown significant generalization ability to downstream tasks, even with minimal fine-tuning. While prompt learning has emerged as an effective strategy to adapt pre-trained VL models for downstream tasks, current approaches frequently encounter severe overfitting to specific downstream data distributions. This overfitting constrains the original behavior of the VL models to generalize to new domains or unseen classes, posing a critical challenge in enhancing the adaptability and generalization of VL models. To address this limitation, we propose Style-Pro, a novel style-guided prompt learning framework that mitigates overfitting and preserves the zero-shot generalization capabilities of CLIP. Style-Pro employs learnable style bases to synthesize diverse distribution shifts, guided by two specialized loss functions that ensure style diversity and content integrity. Then, to minimize discrepancies between unseen domains and the source domain, Style-Pro maps the unseen styles into the known style representation space as a weighted combination of style bases. Moreover, to maintain consistency between the style-shifted prompted model and the original frozen CLIP, Style-Pro introduces consistency constraints to preserve alignment in the learned embeddings, minimizing deviation during adaptation to downstream tasks. Extensive experiments across 11 benchmark datasets demonstrate the effectiveness of Style-Pro, consistently surpassing state-of-the-art methods in various settings, including base-to-new generalization, cross-dataset transfer, and domain generalization.

Paperid: 5053, https://arxiv.org/pdf/2411.15459.pdf

Abstract:
The vision-language tracking task aims to perform object tracking based on various modality references. Existing Transformer-based vision-language tracking methods have made remarkable progress by leveraging the global modeling ability of self-attention. However, current approaches still face challenges in effectively exploiting the temporal information and dynamically updating reference features during tracking. Recently, the State Space Model (SSM), known as Mamba, has shown astonishing ability in efficient long-sequence modeling. Particularly, its state space evolving process demonstrates promising capabilities in memorizing multimodal temporal information with linear complexity. Witnessing its success, we propose a Mamba-based vision-language tracking model to exploit its state space evolving ability in temporal space for robust multimodal tracking, dubbed MambaVLT. In particular, our approach mainly integrates a time-evolving hybrid state space block and a selective locality enhancement block, to capture contextual information for multimodal modeling and adaptive reference feature update. Besides, we introduce a modality-selection module that dynamically adjusts the weighting between visual and language references, mitigating potential ambiguities from either reference type. Extensive experimental results show that our method performs favorably against state-of-the-art trackers across diverse benchmarks.

Paperid: 5054, https://arxiv.org/pdf/2411.05961.pdf

Abstract:
Vision Language Models (VLMs) are central to Visual Question Answering (VQA) systems and are typically deployed in the cloud due to their high computational demands. However, this cloud-only approach underutilizes edge computational resources and requires significant bandwidth for transmitting raw images. In this paper, we introduce an edge-cloud collaborative VQA system, called LLaVA-AlignedVQ, which features a novel Aligned Vector Quantization algorithm (AlignedVQ) that efficiently compress intermediate features without compromising accuracy to support partitioned execution. Our experiments demonstrate that LLaVA-AlignedVQ achieves approximately 1365x compression rate of intermediate features, reducing data transmission overhead by 96.8% compared to transmitting JPEG90-compressed images to the cloud. LLaVA-AlignedVQ achieves an inference speedup of 2-15x while maintaining high accuracy, remaining within -2.23% to +1.6% of the original model's accuracy performance across eight VQA datasets, compared to the cloud-only solution.

Paperid: 5055, https://arxiv.org/pdf/2411.00257.pdf

Abstract:
Vision Language Models (VLMs) have been successful at many chart comprehension tasks that require attending to both the images of charts and their accompanying textual descriptions. However, it is not well established how VLM performance profiles map to human-like behaviors. If VLMs can be shown to have human-like chart comprehension abilities, they can then be applied to a broader range of tasks, such as designing and evaluating visualizations for human readers. This paper lays the foundations for such applications by evaluating the accuracy of zero-shot prompting of VLMs on graphical perception tasks with established human performance profiles. Our findings reveal that VLMs perform similarly to humans under specific task and style combinations, suggesting that they have the potential to be used for modeling human performance. Additionally, variations to the input stimuli show that VLM accuracy is sensitive to stylistic changes such as fill color and chart contiguity, even when the underlying data and data mappings are the same.

Paperid: 5056, https://arxiv.org/pdf/2410.15199.pdf

Abstract:
We propose a zero-shot text-driven 3D shape deformation system that deforms an input 3D mesh of a manufactured object to fit an input text description. To do this, our system optimizes the parameters of a deformation model to maximize an objective function based on the widely used pre-trained vision language model CLIP. We find that CLIP-based objective functions exhibit many spurious local optima; to circumvent them, we parameterize deformations using a novel deformation model called BoxDefGraph which our system automatically computes from an input mesh, the BoxDefGraph is designed to capture the object aligned rectangular/circular geometry features of most manufactured objects. We then use the CMA-ES global optimization algorithm to maximize our objective, which we find to work better than popular gradient-based optimizers. We demonstrate that our approach produces appealing results and outperforms several baselines.

Paperid: 5057, https://arxiv.org/pdf/2410.13146.pdf

Abstract:
As Vision Language Models (VLMs) gain widespread use, their fairness remains under-explored. In this paper, we analyze demographic biases across five models and six datasets. We find that portrait datasets like UTKFace and CelebA are the best tools for bias detection, finding gaps in performance and fairness for both LLaVa and CLIP models. Scene-based datasets like PATA and VLStereoSet fail to be useful benchmarks for bias due to their text prompts allowing the model to guess the answer without a picture. As for pronoun-based datasets like VisoGender, we receive mixed signals as only some subsets of the data are useful in providing insights. To alleviate these two problems, we introduce a more rigorous evaluation dataset and a debiasing method based on Sparse Autoencoders to help reduce bias in models. We find that our data set generates more meaningful errors than the previous data sets. Furthermore, our debiasing method improves fairness, gaining 5-15 points in performance over the baseline. This study displays the problems with the current benchmarks for measuring demographic bias in Vision Language Models and introduces both a more effective dataset for measuring bias and a novel and interpretable debiasing method based on Sparse Autoencoders.

Paperid: 5058, https://arxiv.org/pdf/2410.05869.pdf

Abstract:
Can objects that are not visible in an image -- but are in the vicinity of the camera -- be detected? This study introduces the novel tasks of 2D, 2.5D and 3D unobserved object detection for predicting the location of nearby objects that are occluded or lie outside the image frame. We adapt several state-of-the-art pre-trained generative models to address this task, including 2D and 3D diffusion models and vision-language models, and show that they can be used to infer the presence of objects that are not directly observed. To benchmark this task, we propose a suite of metrics that capture different aspects of performance. Our empirical evaluation on indoor scenes from the RealEstate10k and NYU Depth v2 datasets demonstrate results that motivate the use of generative models for the unobserved object detection task.

Paperid: 5059, https://arxiv.org/pdf/2410.05608.pdf

Abstract:
This tutorial explores recent advancements in multimodal pretrained and large models, capable of integrating and processing diverse data forms such as text, images, audio, and video. Participants will gain an understanding of the foundational concepts of multimodality, the evolution of multimodal research, and the key technical challenges addressed by these models. We will cover the latest multimodal datasets and pretrained models, including those beyond vision and language. Additionally, the tutorial will delve into the intricacies of multimodal large models and instruction tuning strategies to optimise performance for specific tasks. Hands-on laboratories will offer practical experience with state-of-the-art multimodal models, demonstrating real-world applications like visual storytelling and visual question answering. This tutorial aims to equip researchers, practitioners, and newcomers with the knowledge and skills to leverage multimodal AI. ACM Multimedia 2024 is the ideal venue for this tutorial, aligning perfectly with our goal of understanding multimodal pretrained and large language models, and their tuning mechanisms.

Paperid: 5060, https://arxiv.org/pdf/2410.04734.pdf

Abstract:
Although reward models have been successful in improving multimodal large language models, the reward models themselves remain brutal and contain minimal information. Notably, existing reward models only mimic human annotations by assigning only one binary feedback to any text, no matter how long the text is. In the realm of multimodal language models, where models are required to process both images and texts, a naive reward model may learn implicit biases toward texts and become less grounded in images. In this paper, we propose a $\textbf{T}$oken-$\textbf{L}$evel $\textbf{D}$etective $\textbf{R}$eward Model ($\textbf{TLDR}$) to provide fine-grained annotations to each text token. We first introduce a perturbation-based method to generate synthetic hard negatives and their token-level labels to train TLDR models. Then we show the rich usefulness of TLDR models both in assisting off-the-shelf models to self-correct their generations, and in serving as a hallucination evaluation tool. We show that TLDR automatically trains a token-level likelihood optimization, and can improve the base model's performance significantly. Finally, we show that TLDR models can significantly speed up human annotation by 3 times to acquire a broader range of high-quality vision language data.

Paperid: 5061, https://arxiv.org/pdf/2409.16494.pdf

Abstract:
Hallucination is a common problem for Large Vision-Language Models (LVLMs) with long generations which is difficult to eradicate. The generation with hallucinations is partially inconsistent with the image content. To mitigate hallucination, current studies either focus on the process of model inference or the results of model generation, but the solutions they design sometimes do not deal appropriately with various types of queries and the hallucinations of the generations about these queries. To accurately deal with various hallucinations, we present a unified framework, Dentist, for hallucination mitigation. The core step is to first classify the queries, then perform different processes of hallucination mitigation based on the classification result, just like a dentist first observes the teeth and then makes a plan. In a simple deployment, Dentist can classify queries as perception or reasoning and easily mitigate potential hallucinations in answers which has been demonstrated in our experiments. On MMbench, we achieve a 13.44%/10.2%/15.8% improvement in accuracy on Image Quality, a Coarse Perception visual question answering (VQA) task, over the baseline InstructBLIP/LLaVA/VisualGLM.

Paperid: 5062, https://arxiv.org/pdf/2409.14801.pdf

Abstract:
Detecting critical moments, such as emotional outbursts or changes in decisions during conversations, is crucial for understanding shifts in human behavior and their consequences. Our work introduces a novel problem setting focusing on these moments as turning points (TPs), accompanied by a meticulously curated, high-consensus, human-annotated multi-modal dataset. We provide precise timestamps, descriptions, and visual-textual evidence high-lighting changes in emotions, behaviors, perspectives, and decisions at these turning points. We also propose a framework, TPMaven, utilizing state-of-the-art vision-language models to construct a narrative from the videos and large language models to classify and detect turning points in our multi-modal dataset. Evaluation results show that TPMaven achieves an F1-score of 0.88 in classification and 0.61 in detection, with additional explanations aligning with human expectations.

Paperid: 5063, https://arxiv.org/pdf/2409.12953.pdf

Abstract:
Existing vision-language understanding benchmarks largely consist of images of objects in their usual contexts. As a consequence, recent multimodal large language models can perform well with only a shallow visual understanding by relying on background language biases. Thus, strong performance on these benchmarks does not necessarily correlate with strong visual understanding. In this paper, we release JourneyBench, a comprehensive human-annotated benchmark of generated images designed to assess the model's fine-grained multimodal reasoning abilities across five tasks: complementary multimodal chain of thought, multi-image VQA, imaginary image captioning, VQA with hallucination triggers, and fine-grained retrieval with sample-specific distractors. Unlike existing benchmarks, JourneyBench explicitly requires fine-grained multimodal reasoning in unusual imaginary scenarios where language bias and holistic image gist are insufficient. We benchmark state-of-the-art models on JourneyBench and analyze performance along a number of fine-grained dimensions. Results across all five tasks show that JourneyBench is exceptionally challenging for even the best models, indicating that models' visual reasoning abilities are not as strong as they first appear. We discuss the implications of our findings and propose avenues for further research.

Paperid: 5064, https://arxiv.org/pdf/2409.10441.pdf

Abstract:
Camera-to-robot calibration is crucial for vision-based robot control and requires effort to make it accurate. Recent advancements in markerless pose estimation methods have eliminated the need for time-consuming physical setups for camera-to-robot calibration. While the existing markerless pose estimation methods have demonstrated impressive accuracy without the need for cumbersome setups, they rely on the assumption that all the robot joints are visible within the camera's field of view. However, in practice, robots usually move in and out of view, and some portion of the robot may stay out-of-frame during the whole manipulation task due to real-world constraints, leading to a lack of sufficient visual features and subsequent failure of these approaches. To address this challenge and enhance the applicability to vision-based robot control, we propose a novel framework capable of estimating the robot pose with partially visible robot manipulators. Our approach leverages the Vision-Language Models for fine-grained robot components detection, and integrates it into a keypoint-based pose estimation network, which enables more robust performance in varied operational conditions. The framework is evaluated on both public robot datasets and self-collected partial-view datasets to demonstrate our robustness and generalizability. As a result, this method is effective for robot pose estimation in a wider range of real-world manipulation scenarios.

Paperid: 5065, https://arxiv.org/pdf/2409.10196.pdf

Abstract:
This paper addresses the problem of autonomous UAV search missions, where a UAV must locate specific Entities of Interest (EOIs) within a time limit, based on brief descriptions in large, hazard-prone environments with keep-out zones. The UAV must perceive, reason, and make decisions with limited and uncertain information. We propose NEUSIS, a compositional neuro-symbolic system designed for interpretable UAV search and navigation in realistic scenarios. NEUSIS integrates neuro-symbolic visual perception, reasoning, and grounding (GRiD) to process raw sensory inputs, maintains a probabilistic world model for environment representation, and uses a hierarchical planning component (SNaC) for efficient path planning. Experimental results from simulated urban search missions using AirSim and Unreal Engine show that NEUSIS outperforms a state-of-the-art (SOTA) vision-language model and a SOTA search planning model in success rate, search efficiency, and 3D localization. These results demonstrate the effectiveness of our compositional neuro-symbolic approach in handling complex, real-world scenarios, making it a promising solution for autonomous UAV systems in search missions.

Paperid: 5066, https://arxiv.org/pdf/2409.09427.pdf

Abstract:
In this paper, we study the problem of Text-to-Image Person Re-identification (TIReID), which aims to find images of the same identity described by a text sentence from a pool of candidate images. Benefiting from Vision-Language Pre-training, such as CLIP (Contrastive Language-Image Pretraining), the TIReID techniques have achieved remarkable progress recently. However, most existing methods only focus on instance-level matching and ignore identity-level matching, which involves associating multiple images and texts belonging to the same person. In this paper, we propose a novel prototypical prompting framework (Propot) designed to simultaneously model instance-level and identity-level matching for TIReID. Our Propot transforms the identity-level matching problem into a prototype learning problem, aiming to learn identity-enriched prototypes. Specifically, Propot works by 'initialize, adapt, enrich, then aggregate'. We first use CLIP to generate high-quality initial prototypes. Then, we propose a domain-conditional prototypical prompting (DPP) module to adapt the prototypes to the TIReID task using task-related information. Further, we propose an instance-conditional prototypical prompting (IPP) module to update prototypes conditioned on intra-modal and inter-modal instances to ensure prototype diversity. Finally, we design an adaptive prototype aggregation module to aggregate these prototypes, generating final identity-enriched prototypes. With identity-enriched prototypes, we diffuse its rich identity information to instances through prototype-to-instance contrastive loss to facilitate identity-level matching. Extensive experiments conducted on three benchmarks demonstrate the superiority of Propot compared to existing TIReID methods.

Paperid: 5067, https://arxiv.org/pdf/2409.09273.pdf

Abstract:
Recently pre-trained Foundation Models (FMs) have been combined with Federated Learning (FL) to improve training of downstream tasks while preserving privacy. However, deploying FMs over edge networks with resource-constrained Internet of Things (IoT) devices is under-explored. This paper proposes a novel framework, namely, Federated Distilling knowledge to Prompt (FedD2P), for leveraging the robust representation abilities of a vision-language FM without deploying it locally on edge devices. This framework distills the aggregated knowledge of IoT devices to a prompt generator to efficiently adapt the frozen FM for downstream tasks. To eliminate the dependency on a public dataset, our framework leverages perclass local knowledge from IoT devices and linguistic descriptions of classes to train the prompt generator. Our experiments on diverse image classification datasets CIFAR, OxfordPets, SVHN, EuroSAT, and DTD show that FedD2P outperforms the baselines in terms of model performance.

Paperid: 5068, https://arxiv.org/pdf/2409.03521.pdf

Abstract:
The emergence of large Vision-Language Models (VLMs) has established new baselines in image classification across multiple domains. We examine whether their multimodal reasoning can also address a challenge mastered by human experts. Specifically, we test whether VLMs can classify the style, author and creation date of paintings, a domain traditionally mastered by art historians. Artworks pose a unique challenge compared to natural images due to their inherently complex and diverse structures, characterized by variable compositions and styles. This requires a contextual and stylistic interpretation rather than straightforward object recognition. Art historians have long studied the unique aspects of artworks, with style prediction being a crucial component of their discipline. This paper investigates whether large VLMs, which integrate visual and textual data, can effectively reason about the historical and stylistic attributes of paintings. We present the first study of its kind, conducting an in-depth analysis of three VLMs, namely CLIP, LLaVA, and GPT-4o, evaluating their zero-shot classification of art style, author and time period. Using two image benchmarks of artworks, we assess the models' ability to interpret style, evaluate their sensitivity to prompts, and examine failure cases. Additionally, we focus on how these models compare to human art historical expertise by analyzing misclassifications, providing insights into their reasoning and classification patterns.

Paperid: 5069, https://arxiv.org/pdf/2601.19640.pdf

Abstract:
Low-altitude vision systems are becoming a critical infrastructure for smart city governance. However, existing object-centric perception paradigms and loosely coupled vision-language pipelines are still difficult to support management-oriented anomaly understanding required in real-world urban governance. To bridge this gap, we introduce GovLA-10K, the first management-oriented multi-modal benchmark for low-altitude intelligence, along with GovLA-Reasoner, a unified vision-language reasoning framework tailored for governance-aware aerial perception. Unlike existing studies that aim to exhaustively annotate all visible objects, GovLA-10K is deliberately designed around functionally salient targets that directly correspond to practical management needs, and further provides actionable management suggestions grounded in these observations. To effectively coordinate the fine-grained visual grounding with high-level contextual language reasoning, GovLA-Reasoner introduces an efficient feature adapter that implicitly coordinates discriminative representation sharing between the visual detector and the large language model (LLM). Extensive experiments show that our method significantly improves performance while avoiding the need of fine-tuning for any task-specific individual components. We believe our work offers a new perspective and foundation for future studies on management-aware low-altitude vision-language systems.

Paperid: 5070, https://arxiv.org/pdf/2601.16667.pdf

Abstract:
Vision-Language-Action (VLA) models have advanced robotic manipulation by combining vision, language, and proprioception to predict actions. However, previous methods fuse proprioceptive signals directly with VLM-encoded vision-language features, resulting in state-dominant bias and false completions despite visible execution failures. We attribute this to modality imbalance, where policies over-rely on internal state while underusing visual evidence. To address this, we present ReViP, a novel VLA framework with Vision-Proprioception Rebalance to enhance visual grounding and robustness under perturbations. The key insight is to introduce auxiliary task-aware environment priors to adaptively modulate the coupling between semantic perception and proprioceptive dynamics. Specifically, we use an external VLM as a task-stage observer to extract real-time task-centric visual cues from visual observations, which drive a Vision-Proprioception Feature-wise Linear Modulation to enhance environmental awareness and reduce state-driven errors. Moreover, to evaluate false completion, we propose the first False-Completion Benchmark Suite built on LIBERO with controlled settings such as Object-Drop. Extensive experiments show that ReViP effectively reduces false-completion rates and improves success rates over strong VLA baselines on our suite, with gains extending to LIBERO, RoboTwin 2.0, and real-world evaluations.

Paperid: 5071, https://arxiv.org/pdf/2601.15745.pdf

Abstract:
In the realm of medical report generation (MRG), the integration of natural language processing has emerged as a vital tool to alleviate the workload of radiologists. Despite the impressive capabilities demonstrated by large vision language models (LVLMs) in understanding natural language, their susceptibility to generating plausible yet inaccurate claims, known as ``hallucinations'', raises concerns-especially in the nuanced and critical field of medical. In this work, we introduce a framework, \textbf{K}nowledge-\textbf{E}nhanced with Fine-Grained \textbf{R}einforced Rewards \textbf{M}edical Report Generation (KERM), to tackle the issue. Our approach refines the input to the LVLM by first utilizing MedCLIP for knowledge retrieval, incorporating relevant lesion fact sentences from a curated knowledge corpus. We then introduce a novel purification module to ensure the retrieved knowledge is contextually relevant to the patient's clinical context. Subsequently, we employ fine-grained rewards to guide these models in generating highly supportive and clinically relevant descriptions, ensuring the alignment of model's outputs with desired behaviors. Experimental results on IU-Xray and MIMIC-CXR datasets validate the effectiveness of our approach in mitigating hallucinations and enhancing report quality.

Paperid: 5072, https://arxiv.org/pdf/2601.12744.pdf

Abstract:
Intent-Based Networking (IBN) allows operators to specify high-level network goals rather than low-level configurations. While recent work demonstrates that large language models can automate configuration tasks, a distinct class of intents requires generating optimization code to compute provably optimal solutions for traffic engineering, routing, and resource allocation. Current systems assume text-based intent expression, requiring operators to enumerate topologies and parameters in prose. Network practitioners naturally reason about structure through diagrams, yet whether Vision-Language Models (VLMs) can process annotated network sketches into correct optimization code remains unexplored. We present IntentOpt, a benchmark of 85 optimization problems across 17 categories, evaluating four VLMs (GPT-5-Mini, Claude-Haiku-4.5, Gemini-2.5-Flash, Llama-3.2-11B-Vision) under three prompting strategies on multimodal versus text-only inputs. Our evaluation shows that visual parameter extraction reduces execution success by 12-21 percentage points (pp), with GPT-5-Mini dropping from 93% to 72%. Program-of-thought prompting decreases performance by up to 13 pp, and open-source models lag behind closed-source ones, with Llama-3.2-11B-Vision reaching 18% compared to 75% for GPT-5-Mini. These results establish baseline capabilities and limitations of current VLMs for optimization code generation within an IBN system. We also demonstrate practical feasibility through a case study that deploys VLM-generated code to network testbed infrastructure using Model Context Protocol.

Paperid: 5073, https://arxiv.org/pdf/2601.11910.pdf

Abstract:
Open-Vocabulary Object Detection (OVOD) aims to develop the capability to detect anything. Although myriads of large-scale pre-training efforts have built versatile foundation models that exhibit impressive zero-shot capabilities to facilitate OVOD, the necessity of creating a universal understanding for any object cognition according to already pretrained foundation models is usually overlooked. Therefore, in this paper, a training-free Guess What Vision Language Model, called GW-VLM, is proposed to form a universal understanding paradigm based on our carefully designed Multi-Scale Visual Language Searching (MS-VLS) coupled with Contextual Concept Prompt (CCP) for OVOD. This approach can engage a pre-trained Vision Language Model (VLM) and a Large Language Model (LLM) in the game of "guess what". Wherein, MS-VLS leverages multi-scale visual-language soft-alignment for VLM to generate snippets from the results of class-agnostic object detection, while CCP can form the concept of flow referring to MS-VLS and then make LLM understand snippets for OVOD. Finally, the extensive experiments are carried out on natural and remote sensing datasets, including COCO val, Pascal VOC, DIOR, and NWPU-10, and the results indicate that our proposed GW-VLM can achieve superior OVOD performance compared to the-state-of-the-art methods without any training step.

Paperid: 5074, https://arxiv.org/pdf/2601.08531.pdf

Abstract:
Facade renovation offers a more sustainable alternative to full demolition, yet producing design proposals that preserve existing structures while expressing new intent remains challenging. Current workflows typically require detailed as-built modelling before design, which is time-consuming, labour-intensive, and often involves repeated revisions. To solve this issue, we propose a three-stage framework combining generative artificial intelligence (AI) and vision-language models (VLM) that directly processes rough structural sketch and textual descriptions to produce consistent renovation proposals. First, the input sketch is used by a fine-tuned VLM model to predict bounding boxes specifying where modifications are needed and which components should be added. Next, a stable diffusion model generates detailed sketches of new elements, which are merged with the original outline through a generative inpainting pipeline. Finally, ControlNet is employed to refine the result into a photorealistic image. Experiments on datasets and real industrial buildings indicate that the proposed framework can generate renovation proposals that preserve the original structure while improving facade detail quality. This approach effectively bypasses the need for detailed as-built modelling, enabling architects to rapidly explore design alternatives, iterate on early-stage concepts, and communicate renovation intentions with greater clarity.

Paperid: 5075, https://arxiv.org/pdf/2601.08010.pdf

Abstract:
Vision-language models achieve strong performance across a wide range of multimodal understanding and reasoning tasks, yet their multi-step reasoning remains unstable. Repeated sampling over the same input often produces divergent reasoning trajectories and inconsistent final predictions. To address this, we introduce two complementary approaches inspired by test-time scaling: (1) CASHEW, an inference-time framework that stabilizes reasoning by iteratively aggregating multiple candidate trajectories into higher-quality reasoning traces, with explicit visual verification filtering hallucinated steps and grounding reasoning in visual evidence, and (2) CASHEW-RL, a learned variant that internalizes this aggregation behavior within a single model. CASHEW-RL is trained using Group Sequence Policy Optimization (GSPO) with a composite reward that encourages correct answers grounded in minimal yet sufficient visual evidence, while adaptively allocating reasoning effort based on task difficulty. This training objective enables robust self-aggregation at inference. Extensive experiments on 13 image understanding, video understanding, and video reasoning benchmarks show significant performance improvements, including gains of up to +23.6 percentage points on ScienceQA and +8.1 percentage points on EgoSchema.

Paperid: 5076, https://arxiv.org/pdf/2601.07125.pdf

Abstract:
Multi-vector embedding models have emerged as a powerful paradigm for document retrieval, preserving fine-grained visual and textual details through token-level representations. However, this expressiveness comes at a staggering cost: storing embeddings for every token inflates index sizes by over $1000\times$ compared to single-vector approaches, severely limiting scalability. We introduce \textbf{ReinPool}, a reinforcement learning framework that learns to dynamically filter and pool multi-vector embeddings into compact, retrieval-optimized representations. By training with an inverse retrieval objective and NDCG-based rewards, ReinPool identifies and retains only the most discriminative vectors without requiring manual importance annotations. On the Vidore V2 benchmark across three vision-language embedding models, ReinPool compresses multi-vector representations by $746$--$1249\times$ into single vectors while recovering 76--81\% of full multi-vector retrieval performance. Compared to static mean pooling baselines, ReinPool achieves 22--33\% absolute NDCG@3 improvement, demonstrating that learned selection significantly outperforms heuristic aggregation.

Paperid: 5077, https://arxiv.org/pdf/2601.06415.pdf

Abstract:
Utilizing functional elements in an industrial environment, such as displays and interactive valves, provide effective possibilities for robot training. When preparing simulations for robots or applications that involve high-level scene understanding, the simulation environment must be equally detailed. Although CAD files for such environments deliver an exact description of the geometry and visuals, they usually lack semantic, relational and functional information, thus limiting the simulation and training possibilities. A 3D scene graph can organize semantic, spatial and functional information by enriching the environment through a Large Vision-Language Model (LVLM). In this paper we present an offline approach to creating detailed 3D scene graphs from CAD environments. This will serve as a foundation to include the relations of functional and actionable elements, which then can be used for dynamic simulation and reasoning. Key results of this research include both quantitative results of the generated semantic labels as well as qualitative results of the scene graph, especially in hindsight of pipe structures and identified functional relations. All code, results and the environment will be made available at https://cad-scenegraph.github.io

Paperid: 5078, https://arxiv.org/pdf/2601.03534.pdf

Abstract:
Bikeability assessment is essential for advancing sustainable urban transportation and creating cyclist-friendly cities, and it requires incorporating users' perceptions of safety and comfort. Yet existing perception-based bikeability assessment approaches face key limitations in capturing the complexity of road environments and adequately accounting for heterogeneity in subjective user perceptions. This paper proposes a persona-aware Vision-Language Model framework for bikeability assessment with three novel contributions: (i) theory-grounded persona conditioning based on established cyclist typology that generates persona-specific explanations via chain-of-thought reasoning; (ii) multi-granularity supervised fine-tuning that combines scarce expert-annotated reasoning with abundant user ratings for joint prediction and explainable assessment; and (iii) AI-enabled data augmentation that creates controlled paired data to isolate infrastructure variable impacts. To test and validate this framework, we developed a panoramic image-based crowdsourcing system and collected 12,400 persona-conditioned assessments from 427 cyclists. Experiment results show that the proposed framework offers competitive bikeability rating prediction while uniquely enabling explainable factor attribution.

Paperid: 5079, https://arxiv.org/pdf/2601.00269.pdf

Abstract:
Faithfulness hallucinations in VQA occur when vision-language models produce fluent yet visually ungrounded answers, severely undermining their reliability in safety-critical applications. Existing detection methods mainly fall into two categories: external verification approaches relying on auxiliary models or knowledge bases, and uncertainty-driven approaches using repeated sampling or uncertainty estimates. The former suffer from high computational overhead and are limited by external resource quality, while the latter capture only limited facets of model uncertainty and fail to sufficiently explore the rich internal signals associated with the diverse failure modes. Both paradigms thus have inherent limitations in efficiency, robustness, and detection performance. To address these challenges, we propose FaithSCAN: a lightweight network that detects hallucinations by exploiting rich internal signals of VLMs, including token-level decoding uncertainty, intermediate visual representations, and cross-modal alignment features. These signals are fused via branch-wise evidence encoding and uncertainty-aware attention. We also extend the LLM-as-a-Judge paradigm to VQA hallucination and propose a low-cost strategy to automatically generate model-dependent supervision signals, enabling supervised training without costly human labels while maintaining high detection accuracy. Experiments on multiple VQA benchmarks show that FaithSCAN significantly outperforms existing methods in both effectiveness and efficiency. In-depth analysis shows hallucinations arise from systematic internal state variations in visual perception, cross-modal reasoning, and language decoding. Different internal signals provide complementary diagnostic cues, and hallucination patterns vary across VLM architectures, offering new insights into the underlying causes of multimodal hallucinations.

Paperid: 5080, https://arxiv.org/pdf/2601.00150.pdf

Abstract:
As multimodal AI becomes widely used for credit risk assessment and document review, a domain-specific benchmark is urgently needed that (1) reflects documents and workflows specific to financial credit applications, (2) includes credit-specific understanding and real-world robustness, and (3) preserves privacy compliance without sacrificing practical utility. Here, we introduce FCMBench-V1.0 -- a large-scale financial credit multimodal benchmark for real-world applications, covering 18 core certificate types, with 4,043 privacy-compliant images and 8,446 QA samples. The FCMBench evaluation framework consists of three dimensions: Perception, Reasoning, and Robustness, including 3 foundational perception tasks, 4 credit-specific reasoning tasks that require decision-oriented understanding of visual evidence, and 10 real-world acquisition artifact types for robustness stress testing. To reconcile compliance with realism, we construct all samples via a closed synthesis-capture pipeline: we manually synthesize document templates with virtual content and capture scenario-aware images in-house. This design also mitigates pre-training data leakage by avoiding web-sourced or publicly released images. FCMBench can effectively discriminate performance disparities and robustness across modern vision-language models. Extensive experiments were conducted on 23 state-of-the-art vision-language models (VLMs) from 14 top AI companies and research institutes. Among them, Gemini 3 Pro achieves the best F1(\%) score as a commercial model (64.61), Qwen3-VL-235B achieves the best score as an open-source baseline (57.27), and our financial credit-specific model, Qfin-VL-Instruct, achieves the top overall score (64.92). Robustness evaluations show that even top-performing models suffer noticeable performance drops under acquisition artifacts.

Paperid: 5081, https://arxiv.org/pdf/2512.22351.pdf

Abstract:
Despite the remarkable progress of Multimodal Large Language Models (MLLMs) in 2D vision-language tasks, their application to complex 3D scene manipulation remains underexplored. In this paper, we bridge this critical gap by tackling three key challenges in 3D object arrangement task using MLLMs. First, to address the weak visual grounding of MLLMs, which struggle to link programmatic edits with precise 3D outcomes, we introduce an MCP-based API. This shifts the interaction from brittle raw code manipulation to more robust, function-level updates. Second, we augment the MLLM's 3D scene understanding with a suite of specialized visual tools to analyze scene state, gather spatial information, and validate action outcomes. This perceptual feedback loop is critical for closing the gap between language-based updates and precise 3D-aware manipulation. Third, to manage the iterative, error-prone updates, we propose a collaborative multi-agent framework with designated roles for planning, execution, and verification. This decomposition allows the system to robustly handle multi-step instructions and recover from intermediate errors. We demonstrate the effectiveness of our approach on a diverse set of 25 complex object arrangement tasks, where it significantly outperforms existing baselines. Website: vulcan-3d.github.io

Paperid: 5082, https://arxiv.org/pdf/2512.17326.pdf

Abstract:
Vision-language models (VLMs) have the potential to become co-pilots for pathologists. However, most VLMs either focus on small regions of interest within whole-slide images, provide only static slide-level outputs, or rely on data that is not publicly available, limiting reproducibility. Furthermore, training data containing WSIs paired with detailed clinical reports is scarce, restricting progress toward transparent and generalisable VLMs. We address these limitations with three main contributions. First, we introduce Polysome, a standardised tool for synthetic instruction generation. Second, we apply Polysome to the public HISTAI dataset, generating HISTAI-Instruct, a large whole-slide instruction tuning dataset spanning 24,259 slides and over 1.1 million instruction-response pairs. Finally, we use HISTAI-Instruct to train ANTONI-α, a VLM capable of visual-question answering (VQA). We show that ANTONI-α outperforms MedGemma on WSI-level VQA tasks of tissue identification, neoplasm detection, and differential diagnosis. We also compare the performance of multiple incarnations of ANTONI-α trained with different amounts of data. All methods, data, and code are publicly available.

Paperid: 5083, https://arxiv.org/pdf/2512.16755.pdf

Abstract:
Vision-Language Models (VLMs) have made significant progress in explicit instruction-based navigation; however, their ability to interpret implicit human needs (e.g., "I am thirsty") in dynamic urban environments remains underexplored. This paper introduces CitySeeker, a novel benchmark designed to assess VLMs' spatial reasoning and decision-making capabilities for exploring embodied urban navigation to address implicit needs. CitySeeker includes 6,440 trajectories across 8 cities, capturing diverse visual characteristics and implicit needs in 7 goal-driven scenarios. Extensive experiments reveal that even top-performing models (e.g., Qwen2.5-VL-32B-Instruct) achieve only 21.1% task completion. We find key bottlenecks in error accumulation in long-horizon reasoning, inadequate spatial cognition, and deficient experiential recall. To further analyze them, we investigate a series of exploratory strategies-Backtracking Mechanisms, Enriching Spatial Cognition, and Memory-Based Retrieval (BCR), inspired by human cognitive mapping's emphasis on iterative observation-reasoning cycles and adaptive path optimization. Our analysis provides actionable insights for developing VLMs with robust spatial intelligence required for tackling "last-mile" navigation challenges.

Paperid: 5084, https://arxiv.org/pdf/2512.15249.pdf

Abstract:
Medical artificial intelligence (AI) systems, particularly multimodal vision-language models (VLM), often exhibit intersectional biases where models are systematically less confident in diagnosing marginalised patient subgroups. Such bias can lead to higher rates of inaccurate and missed diagnoses due to demographically skewed data and divergent distributions of diagnostic certainty. Current fairness interventions frequently fail to address these gaps or compromise overall diagnostic performance to achieve statistical parity among the subgroups. In this study, we developed Cross-Modal Alignment Consistency (CMAC-MMD), a training framework that standardises diagnostic certainty across intersectional patient subgroups. Unlike traditional debiasing methods, this approach equalises the model's decision confidence without requiring sensitive demographic data during clinical inference. We evaluated this approach using 10,015 skin lesion images (HAM10000) with external validation on 12,000 images (BCN20000), and 10,000 fundus images for glaucoma detection (Harvard-FairVLMed), stratifying performance by intersectional age, gender, and race attributes. In the dermatology cohort, the proposed method reduced the overall intersectional missed diagnosis gap (difference in True Positive Rate, $Δ$TPR) from 0.50 to 0.26 while improving the overall Area Under the Curve (AUC) from 0.94 to 0.97 compared to standard training. Similarly, for glaucoma screening, the method reduced $Δ$TPR from 0.41 to 0.31, achieving a better AUC of 0.72 (vs. 0.71 baseline). This establishes a scalable framework for developing high-stakes clinical decision support systems that are both accurate and can perform equitably across diverse patient subgroups, ensuring reliable performance without increasing privacy risks.

Paperid: 5085, https://arxiv.org/pdf/2512.14177.pdf

Abstract:
Large Vision-Language Models (LVLMs) often produce plausible but unreliable outputs, making robust uncertainty estimation essential. Recent work on semantic uncertainty estimates relies on external models to cluster multiple sampled responses and measure their semantic consistency. However, these clustering methods are often fragile, highly sensitive to minor phrasing variations, and can incorrectly group or separate semantically similar answers, leading to unreliable uncertainty estimates. We propose Semantic Gaussian Process Uncertainty (SGPU), a Bayesian framework that quantifies semantic uncertainty by analyzing the geometric structure of answer embeddings, avoiding brittle clustering. SGPU maps generated answers into a dense semantic space, computes the Gram matrix of their embeddings, and summarizes their semantic configuration via the eigenspectrum. This spectral representation is then fed into a Gaussian Process Classifier that learns to map patterns of semantic consistency to predictive uncertainty, and that can be applied in both black-box and white-box settings. Across six LLMs and LVLMs on eight datasets spanning VQA, image classification, and textual QA, SGPU consistently achieves state-of-the-art calibration (ECE) and discriminative (AUROC, AUARC) performance. We further show that SGPU transfers across models and modalities, indicating that its spectral representation captures general patterns of semantic uncertainty.

Paperid: 5086, https://arxiv.org/pdf/2512.12633.pdf

Abstract:
Multimodal Large Language Models have achieved impressive performance on a variety of vision-language tasks, yet their fine-grained visual perception and precise spatial reasoning remain limited. In this work, we introduce DiG (Differential Grounding), a novel proxy task framework where MLLMs learn fine-grained perception by identifying and localizing all differences between similar image pairs without prior knowledge of their number. To support scalable training, we develop an automated 3D rendering-based data generation pipeline that produces high-quality paired images with fully controllable discrepancies. To address the sparsity of difference signals, we further employ curriculum learning that progressively increases complexity from single to multiple differences, enabling stable optimization. Extensive experiments demonstrate that DiG significantly improves model performance across a variety of visual perception benchmarks and that the learned fine-grained perception skills transfer effectively to standard downstream tasks, including RefCOCO, RefCOCO+, RefCOCOg, and general multimodal perception benchmarks. Our results highlight differential grounding as a scalable and robust approach for advancing fine-grained visual reasoning in MLLMs.

Paperid: 5087, https://arxiv.org/pdf/2512.12596.pdf

Abstract:
In this paper, we propose a method for generating layouts for image-based advertisements by leveraging a Vision-Language Model (VLM). Conventional advertisement layout techniques have predominantly relied on saliency mapping to detect salient regions within a background image, but such approaches often fail to fully account for the image's detailed composition and semantic content. To overcome this limitation, our method harnesses a VLM to recognize the products and other elements depicted in the background and to inform the placement of text and logos. The proposed layout-generation pipeline consists of two steps. In the first step, the VLM analyzes the image to identify object types and their spatial relationships, then produces a text-based "placement plan" based on this analysis. In the second step, that plan is rendered into the final layout by generating HTML-format code. We validated the effectiveness of our approach through evaluation experiments, conducting both quantitative and qualitative comparisons against existing methods. The results demonstrate that by explicitly considering the background image's content, our method produces noticeably higher-quality advertisement layouts.

Paperid: 5088, https://arxiv.org/pdf/2512.12218.pdf

Abstract:
Reasoning-augmented vision language models (VLMs) generate explicit chains of thought that promise greater capability and transparency but also introduce new failure modes: models may reach correct answers via visually unfaithful intermediate steps, or reason faithfully yet fail on the final prediction. Standard evaluations that only measure final-answer accuracy cannot distinguish these behaviors. We introduce the visual faithfulness of reasoning chains as a distinct evaluation dimension, focusing on whether the perception steps of a reasoning chain are grounded in the image. We propose a training- and reference-free framework that decomposes chains into perception versus reasoning steps and uses off-the-shelf VLM judges for step-level faithfulness, additionally verifying this approach through a human meta-evaluation. Building on this metric, we present a lightweight self-reflection procedure that detects and locally regenerates unfaithful perception steps without any training. Across multiple reasoning-trained VLMs and perception-heavy benchmarks, our method reduces Unfaithful Perception Rate while preserving final-answer accuracy, improving the reliability of multimodal reasoning.

Paperid: 5089, https://arxiv.org/pdf/2512.10046.pdf

Abstract:
Recent advances in foundation models have shown promising results in developing generalist robotics that can perform diverse tasks in open-ended scenarios given multimodal inputs. However, current work has been mainly focused on indoor, household scenarios. In this work, we present SimWorld-Robotics~(SWR), a simulation platform for embodied AI in large-scale, photorealistic urban environments. Built on Unreal Engine 5, SWR procedurally generates unlimited photorealistic urban scenes populated with dynamic elements such as pedestrians and traffic systems, surpassing prior urban simulations in realism, complexity, and scalability. It also supports multi-robot control and communication. With these key features, we build two challenging robot benchmarks: (1) a multimodal instruction-following task, where a robot must follow vision-language navigation instructions to reach a destination in the presence of pedestrians and traffic; and (2) a multi-agent search task, where two robots must communicate to cooperatively locate and meet each other. Unlike existing benchmarks, these two new benchmarks comprehensively evaluate a wide range of critical robot capacities in realistic scenarios, including (1) multimodal instructions grounding, (2) 3D spatial reasoning in large environments, (3) safe, long-range navigation with people and traffic, (4) multi-robot collaboration, and (5) grounded communication. Our experimental results demonstrate that state-of-the-art models, including vision-language models (VLMs), struggle with our tasks, lacking robust perception, reasoning, and planning abilities necessary for urban environments.

Paperid: 5090, https://arxiv.org/pdf/2512.05172.pdf

Abstract:
The growing exploration of Large Language Models (LLM) and Vision-Language Models (VLM) has opened avenues for enhancing the effectiveness of reinforcement learning (RL). However, existing LLM-based RL methods often focus on the guidance of control policy and encounter the challenge of limited representations of the backbone networks. To tackle this problem, we introduce Enhanced Semantic Motion Representations (Semore), a new VLM-based framework for visual RL, which can simultaneously extract semantic and motion representations through a dual-path backbone from the RGB flows. Semore utilizes VLM with common-sense knowledge to retrieve key information from observations, while using the pre-trained clip to achieve the text-image alignment, thereby embedding the ground-truth representations into the backbone. To efficiently fuse semantic and motion representations for decision-making, our method adopts a separately supervised approach to simultaneously guide the extraction of semantics and motion, while allowing them to interact spontaneously. Extensive experiments demonstrate that, under the guidance of VLM at the feature level, our method exhibits efficient and adaptive ability compared to state-of-art methods. All codes are released.

Paperid: 5091, https://arxiv.org/pdf/2512.04599.pdf

Abstract:
Detecting illicit visual content demands more than image-level NSFW flags; moderators must also know what objects make an image illegal and where those objects occur. We introduce a zero-shot pipeline that simultaneously (i) detects if an image contains harmful content, (ii) identifies each critical element involved, and (iii) localizes those elements with pixel-accurate masks - all in one pass. The system first applies foundation segmentation model (SAM) to generate candidate object masks and refines them into larger independent regions. Each region is scored for malicious relevance by a vision-language model using open-vocabulary prompts; these scores weight a fusion step that produces a consolidated malicious object map. An ensemble across multiple segmenters hardens the pipeline against adaptive attacks that target any single segmentation method. Evaluated on a newly-annotated 790-image dataset spanning drug, sexual, violent and extremist content, our method attains 85.8% element-level recall, 78.1% precision and a 92.1% segment-success rate - exceeding direct zero-shot VLM localization by 27.4% recall at comparable precision. Against PGD adversarial perturbations crafted to break SAM and VLM, our method's precision and recall decreased by no more than 10%, demonstrating high robustness against attacks. The full pipeline processes an image in seconds, plugs seamlessly into existing VLM workflows, and constitutes the first practical tool for fine-grained, explainable malicious-image moderation.

Paperid: 5092, https://arxiv.org/pdf/2512.04381.pdf

Abstract:
We present FoundAtion-model-guided decoupled LoCO-maNipulation visuomotor policies (FALCON), a framework for loco-manipulation that combines modular diffusion policies with a vision-language foundation model as the coordinator. Our approach explicitly decouples locomotion and manipulation into two specialized visuomotor policies, allowing each subsystem to rely on its own observations. This mitigates the performance degradation that arise when a single policy is forced to fuse heterogeneous, potentially mismatched observations from locomotion and manipulation. Our key innovation lies in restoring coordination between these two independent policies through a vision-language foundation model, which encodes global observations and language instructions into a shared latent embedding conditioning both diffusion policies. On top of this backbone, we introduce a phase-progress head that uses textual descriptions of task stages to infer discrete phase and continuous progress estimates without manual phase labels. To further structure the latent space, we incorporate a coordination-aware contrastive loss that explicitly encodes cross-subsystem compatibility between arm and base actions. We evaluate FALCON on two challenging loco-manipulation tasks requiring navigation, precise end-effector placement, and tight base-arm coordination. Results show that it surpasses centralized and decentralized baselines while exhibiting improved robustness and generalization to out-of-distribution scenarios.

Paperid: 5093, https://arxiv.org/pdf/2512.03913.pdf

Abstract:
Prior Vision-Language-Action (VLA) models are typically trained on teleoperated successful demonstrations, while discarding numerous failed attempts that occur naturally during data collection. However, these failures encode where and how policies can be fragile, information that can be exploited to improve robustness. We address this problem by leveraging mixed-quality datasets to learn failure-aware reasoning at planning time. We introduce VINE, a hierarchical vision-language-action model that separates high-level reasoning (System 2) from low-level control (System 1) under a hierarchical reinforcement learning formalism, making failures usable as a structured learning signal rather than noisy supervision. System 2 performs feasibility-guided tree search over a 2D scene-graph abstraction: it proposes subgoal transitions, predicts success probabilities from both successes and failures, and prunes brittle branches before execution, effectively casting plan evaluation as feasibility scoring. The selected subgoal sequence is then passed to System 1, which executes low-level actions without modifying the agent's core skills. Trained entirely from offline teleoperation data, VINE integrates negative experience directly into the decision loop. Across challenging manipulation tasks, this approach consistently improves success rates and robustness, demonstrating that failure data is an essential resource for converting the broad competence of VLAs into robust execution.

Paperid: 5094, https://arxiv.org/pdf/2512.02743.pdf

Abstract:
Hate speech in online videos is posing an increasingly serious threat to digital platforms, especially as video content becomes increasingly multimodal and context-dependent. Existing methods often struggle to effectively fuse the complex semantic relationships between modalities and lack the ability to understand nuanced hateful content. To address these issues, we propose an innovative Reasoning-Aware Multimodal Fusion (RAMF) framework. To tackle the first challenge, we design Local-Global Context Fusion (LGCF) to capture both local salient cues and global temporal structures, and propose Semantic Cross Attention (SCA) to enable fine-grained multimodal semantic interaction. To tackle the second challenge, we introduce adversarial reasoning-a structured three-stage process where a vision-language model generates (i) objective descriptions, (ii) hate-assumed inferences, and (iii) non-hate-assumed inferences-providing complementary semantic perspectives that enrich the model's contextual understanding of nuanced hateful intent. Evaluations on two real-world hateful video datasets demonstrate that our method achieves robust generalisation performance, improving upon state-of-the-art methods by 3% and 7% in Macro-F1 and hate class recall, respectively. We will release the code after the anonymity period ends.

Paperid: 5095, https://arxiv.org/pdf/2512.00881.pdf

Abstract:
Multimodal Knowledge Editing (MKE) extends traditional knowledge editing to settings involving both textual and visual modalities. However, existing MKE benchmarks primarily assess final answer correctness while neglecting the quality of intermediate reasoning and robustness to visually rephrased inputs. To address this limitation, we introduce MMQAKE, the first benchmark for multimodal multihop question answering with knowledge editing. MMQAKE evaluates (1) a model's ability to reason over 2-5-hop factual chains that span both text and images, including performance at each intermediate step, and (2) robustness to visually rephrased inputs in multihop questions. Our evaluation shows that current MKE methods often struggle to consistently update and reason over multimodal reasoning chains after knowledge edits. To overcome these challenges, we propose Hybrid-DMKG, a hybrid reasoning framework built on a dynamic multimodal knowledge graph (DMKG) to enable accurate multihop reasoning over updated multimodal knowledge. Hybrid-DMKG first uses a large language model to decompose multimodal multihop questions into sequential sub-questions, then applies a multimodal retrieval model to locate updated facts by jointly encoding each sub-question with candidate entities and their associated images. For answer inference, a hybrid reasoning module operates over the DMKG via two parallel paths: (1) relation linking prediction, and (2) RAG reasoning with large vision-language models. A decision module aggregates evidence from both paths to select the most credible answer. Experimental results on MMQAKE show that Hybrid-DMKG significantly outperforms existing MKE approaches, achieving higher accuracy and improved robustness to knowledge updates.

Paperid: 5096, https://arxiv.org/pdf/2511.19859.pdf

Abstract:
Vision-Language-Action (VLA) models built upon Chain-of-Thought (CoT) have achieved remarkable success in advancing general-purpose robotic agents, owing to its significant perceptual comprehension. Recently, since text-only CoT struggles to adequately capture scene details in complex spatial environments, a highly promising strategy involves leveraging visual priors to guide robotic action generation. Nevertheless, these strategies face two inherent challenges: (i) a modality gap between visual observations and low-level actions, and (ii) unstable training due to competing objectives between visual prediction and action generation. To address these challenges, we propose a Vision-Integrated Trajectory Alignment (VITA) framework that learns a shared discrete latent space for vision and action, enabling joint modeling of perception and motor control. VITA introduces a implicit visual CoT: autoregressively generated tokens is simultaneously decoded into future frames predictions and robot actions, thereby internalizing visual dynamics as an inductive bias for motion planning. Extensive experiments on simulated and real-world environments demonstrate state-of-the-art performance. VITA improves 14.5\%, 9.6\% and 12.1\% over existing baselines on CALVIN, LIBERO and SimplerEnv. Furthermore, VITA attains an average success rate of 80.5\% across six real-world tasks, demonstrating its potential as a generalist robotic manipulation model.

Paperid: 5097, https://arxiv.org/pdf/2511.18676.pdf

Abstract:
Current vision-language models (VLMs) in medicine are primarily designed for categorical question answering (e.g., "Is this normal or abnormal?") or qualitative descriptive tasks. However, clinical decision-making often relies on quantitative assessments, such as measuring the size of a tumor or the angle of a joint, from which physicians draw their own diagnostic conclusions. This quantitative reasoning capability remains underexplored and poorly supported in existing VLMs. In this work, we introduce MedVision, a large-scale dataset and benchmark specifically designed to evaluate and improve VLMs on quantitative medical image analysis. MedVision spans 22 public datasets covering diverse anatomies and modalities, with 30.8 million image-annotation pairs. We focus on three representative quantitative tasks: (1) detection of anatomical structures and abnormalities, (2) tumor/lesion (T/L) size estimation, and (3) angle/distance (A/D) measurement. Our benchmarks show that current off-the-shelf VLMs perform poorly on these tasks. However, with supervised fine-tuning on MedVision, we significantly enhance their performance across detection, T/L estimation, and A/D measurement, demonstrating reduced error rates and improved precision. This work provides a foundation for developing VLMs with robust quantitative reasoning capabilities in medical imaging. Code and data are available at https://medvision-vlm.github.io.

Paperid: 5098, https://arxiv.org/pdf/2511.16743.pdf

Abstract:
Improving the safety of vision-language models like CLIP via fine-tuning often comes at a steep price, causing significant drops in their generalization performance. We find this trade-off stems from rigid alignment strategies that force unsafe concepts toward single, predefined safe targets, disrupting the model's learned semantic structure. To address this, we propose a proximity-aware approach: redirecting unsafe concepts to their semantically closest safe alternatives to minimize representational change. We introduce SaFeR-CLIP, a fine-tuning framework that applies this principle of minimal intervention. SaFeR-CLIP successfully reconciles safety and performance, recovering up to 8.0% in zero-shot accuracy over prior methods while maintaining robust safety. To support more rigorous evaluation, we also contribute NSFW-Caps, a new benchmark of 1,000 highly-aligned pairs for testing safety under distributional shift. Our work shows that respecting the geometry of pretrained representations is key to achieving safety without sacrificing performance.

Paperid: 5099, https://arxiv.org/pdf/2511.15316.pdf

Abstract:
Split DNNs enable edge devices by offloading intensive computation to a cloud server, but this paradigm exposes privacy vulnerabilities, as the intermediate features can be exploited to reconstruct the private inputs via Feature Inversion Attack (FIA). Existing FIA methods often produce limited reconstruction quality, making it difficult to assess the true extent of privacy leakage. To reveal the privacy risk of the leaked features, we introduce FIA-Flow, a black-box FIA framework that achieves high-fidelity image reconstruction from intermediate features. To exploit the semantic information within intermediate features, we design a Latent Feature Space Alignment Module (LFSAM) to bridge the semantic gap between the intermediate feature space and the latent space. Furthermore, to rectify distributional mismatch, we develop Deterministic Inversion Flow Matching (DIFM), which projects off-manifold features onto the target manifold with one-step inference. This decoupled design simplifies learning and enables effective training with few image-feature pairs. To quantify privacy leakage from a human perspective, we also propose two metrics based on a large vision-language model. Experiments show that FIA-Flow achieves more faithful and semantically aligned feature inversion across various models (AlexNet, ResNet, Swin Transformer, DINO, and YOLO11) and layers, revealing a more severe privacy threat in Split DNNs than previously recognized.

Paperid: 5100, https://arxiv.org/pdf/2511.15242.pdf

Abstract:
We present SkinGPT-R1, a dermatology focused vision language model that makes diagnostic chain of thought reasoning explicit, step by step, and verifiable. To support skin specific reasoning, we build DermCoT, a corpus of standardized dermatologic chain of thought narratives that combines 10,000 DermEval filtered training cases with 3,000 dermatologist scored certified cases, and we define DermEval as a physician aligned six dimensional evaluator and DermBench as the corresponding benchmark for dermatologic chain of thought quality. On DermBench, across 14 general, reasoning, and medical vision language models, SkinGPT-R1 achieves an average score of 4.031 out of 5 over the six clinician defined dimensions, ranks 1st among all systems, and improves the average score over Vision-R1 by about 41%. On three dermatology classification benchmarks, SkinGPT-R1 delivers stable accuracy gains over Vision-R1 and remains competitive among strong vision language models. Ablation results further show that DermCoT based chain of thought supervision provides substantial improvements over the base model and that adding dermatology aware visual distillation yields consistent additional gains in both narrative quality and recognition.

Paperid: 5101, https://arxiv.org/pdf/2511.14884.pdf

Abstract:
Methods that synthesize indoor 3D scenes from text prompts have wide-ranging applications in film production, interior design, video games, virtual reality, and synthetic data generation for training embodied agents. Existing approaches typically either train generative models from scratch or leverage vision-language models (VLMs). While VLMs achieve strong performance, particularly for complex or open-ended prompts, smaller task-specific models remain necessary for deployment on resource-constrained devices such as extended reality (XR) glasses or mobile phones. However, many generative approaches that train from scratch overlook the inherent graph structure of indoor scenes, which can limit scene coherence and realism. Conversely, methods that incorporate scene graphs either demand a user-provided semantic graph, which is generally inconvenient and restrictive, or rely on ground-truth relationship annotations, limiting their capacity to capture more varied object interactions. To address these challenges, we introduce GeoSceneGraph, a method that synthesizes 3D scenes from text prompts by leveraging the graph structure and geometric symmetries of 3D scenes, without relying on predefined relationship classes. Despite not using ground-truth relationships, GeoSceneGraph achieves performance comparable to methods that do. Our model is built on equivariant graph neural networks (EGNNs), but existing EGNN approaches are typically limited to low-dimensional conditioning and are not designed to handle complex modalities such as text. We propose a simple and effective strategy for conditioning EGNNs on text features, and we validate our design through ablation studies.

Paperid: 5102, https://arxiv.org/pdf/2511.13684.pdf

Abstract:
We introduce GS-Light, an efficient, textual position-aware pipeline for text-guided relighting of 3D scenes represented via Gaussian Splatting (3DGS). GS-Light implements a training-free extension of a single-input diffusion model to handle multi-view inputs. Given a user prompt that may specify lighting direction, color, intensity, or reference objects, we employ a large vision-language model (LVLM) to parse the prompt into lighting priors. Using off-the-shelf estimators for geometry and semantics (depth, surface normals, and semantic segmentation), we fuse these lighting priors with view-geometry constraints to compute illumination maps and generate initial latent codes for each view. These meticulously derived init latents guide the diffusion model to generate relighting outputs that more accurately reflect user expectations, especially in terms of lighting direction. By feeding multi-view rendered images, along with the init latents, into our multi-view relighting model, we produce high-fidelity, artistically relit images. Finally, we fine-tune the 3DGS scene with the relit appearance to obtain a fully relit 3D scene. We evaluate GS-Light on both indoor and outdoor scenes, comparing it to state-of-the-art baselines including per-view relighting, video relighting, and scene editing methods. Using quantitative metrics (multi-view consistency, imaging quality, aesthetic score, semantic similarity, etc.) and qualitative assessment (user studies), GS-Light demonstrates consistent improvements over baselines. Code and assets will be made available upon publication.

Paperid: 5103, https://arxiv.org/pdf/2511.12446.pdf

Abstract:
Medical visual question answering could support clinical decision making, yet current systems often fail under domain shift and produce answers that are weakly grounded in image evidence. This reliability gap arises when models attend to spurious regions and when retraining or additional labels are impractical at deployment time. We address this setting with CoTBox-TTT, an evidence-first test-time training approach that adapts a vision-language model at inference while keeping all backbones frozen. The method updates only a small set of continuous soft prompts. It identifies question-relevant regions through a visual chain-of-thought signal and encourages answer consistency across the original image and a localized crop. The procedure is label free, and plug and play with diverse backbones. Experiments on medical VQA show that the approach is practical for real deployments. For instance, adding CoTBox-TTT to LLaVA increases closed-ended accuracy by 12.3% on pathVQA.

Paperid: 5104, https://arxiv.org/pdf/2511.09958.pdf

Abstract:
The Vision-Language-Action models (VLA) have achieved significant advances in robotic manipulation recently. However, vision-only VLA models create fundamental limitations, particularly in perceiving interactive and manipulation dynamic processes. This paper proposes Audio-VLA, a multimodal manipulation policy that leverages contact audio to perceive contact events and dynamic process feedback. Audio-VLA overcomes the vision-only constraints of VLA models. Additionally, this paper introduces the Task Completion Rate (TCR) metric to systematically evaluate dynamic operational processes. Audio-VLA employs pre-trained DINOv2 and SigLIP as visual encoders, AudioCLIP as the audio encoder, and Llama2 as the large language model backbone. We apply LoRA fine-tuning to these pre-trained modules to achieve robust cross-modal understanding of both visual and acoustic inputs. A multimodal projection layer aligns features from different modalities into the same feature space. Moreover RLBench and LIBERO simulation environments are enhanced by adding collision-based audio generation to provide realistic sound feedback during object interactions. Since current robotic manipulation evaluations focus on final outcomes rather than providing systematic assessment of dynamic operational processes, the proposed TCR metric measures how well robots perceive dynamic processes during manipulation, creating a more comprehensive evaluation metric. Extensive experiments on LIBERO, RLBench, and two real-world tasks demonstrate Audio-VLA's superior performance over vision-only comparative methods, while the TCR metric effectively quantifies dynamic process perception capabilities.

Paperid: 5105, https://arxiv.org/pdf/2511.07941.pdf

Abstract:
While Large Language Models (LLMs) are emerging as a promising direction in computational pathology, the substantial computational cost of giga-pixel Whole Slide Images (WSIs) necessitates the use of Multi-Instance Learning (MIL) to enable effective modeling. A key challenge is that pathological tasks typically provide only bag-level labels, while instance-level descriptions generated by LLMs often suffer from bias due to a lack of fine-grained medical knowledge. To address this, we propose that constructing task-specific pathological entity prototypes is crucial for learning generalizable features and enhancing model interpretability. Furthermore, existing vision-language MIL methods often employ unidirectional guidance, limiting cross-modal synergy. In this paper, we introduce a novel approach, Multimodal Prototype-based Multi-Instance Learning, that promotes bidirectional interaction through a balanced information compression scheme. Specifically, we leverage a frozen LLM to generate task-specific pathological entity descriptions, which are learned as text prototypes. Concurrently, the vision branch learns instance-level prototypes to mitigate the model's reliance on redundant data. For the fusion stage, we employ the Stereoscopic Optimal Transport (SOT) algorithm, which is based on a similarity metric, thereby facilitating broader semantic alignment in a higher-dimensional space. We conduct few-shot classification and explainability experiments on three distinct cancer datasets, and the results demonstrate the superior generalization capabilities of our proposed method.

Paperid: 5106, https://arxiv.org/pdf/2511.06991.pdf

Abstract:
Large models have achieved remarkable performance across a range of reasoning and understanding tasks. Prior work often utilizes model ensembles or multi-agent systems to collaboratively generate responses, effectively operating in a server-to-server paradigm. However, such approaches do not align well with practical deployment settings, where a limited number of server-side models are shared by many clients under modern internet architectures. In this paper, we introduce \textbf{CoLM} (\textbf{Co}llaboration in \textbf{L}arge-\textbf{M}odels), a novel framework for collaborative reasoning that redefines cooperation among large models from a client-server perspective. Unlike traditional ensemble methods that rely on simultaneous inference from multiple models to produce a single output, CoLM allows the outputs of multiple models to be aggregated or shared, enabling each client model to independently refine and update its own generation based on these high-quality outputs. This design enables collaborative benefits by fully leveraging both client-side and shared server-side models. We further extend CoLM to vision-language models (VLMs), demonstrating its applicability beyond language tasks. Experimental results across multiple benchmarks show that CoLM consistently improves model performance on previously failed queries, highlighting the effectiveness of collaborative guidance in enhancing single-model capabilities.

Paperid: 5107, https://arxiv.org/pdf/2511.06283.pdf

Abstract:
While Vision Language Models (VLMs) have demonstrated remarkable capabilities in general visual understanding, their application in the chemical domain has been limited, with previous works predominantly focusing on text and thus overlooking critical visual information, such as molecular structures. Current approaches that directly adopt standard VLMs for chemical tasks suffer from two primary issues: (i) computational inefficiency of processing entire chemical images with non-informative backgrounds. (ii) a narrow scope on molecular-level tasks that restricts progress in chemical reasoning. In this work, we propose \textbf{TinyChemVL}, an efficient and powerful chemical VLM that leverages visual token reduction and reaction-level tasks to improve model efficiency and reasoning capacity. Also, we propose \textbf{ChemRxn-V}, a reaction-level benchmark for assessing vision-based reaction recognition and prediction tasks. Directly predicting reaction products from molecular images poses a non-trivial challenge, as it requires models to integrate both recognition and reasoning capacities. Our results demonstrate that with only 4B parameters, TinyChemVL achieves superior performance on both molecular and reaction tasks while demonstrating faster inference and training speeds compared to existing models. Notably, TinyChemVL outperforms ChemVLM while utilizing only 1/16th of the visual tokens. This work builds efficient yet powerful VLMs for chemical domains by co-designing model architecture and task complexity.

Paperid: 5108, https://arxiv.org/pdf/2511.05791.pdf

Abstract:
Robotic grasping is a fundamental capability for autonomous manipulation; however, most existing methods rely on large-scale expert annotations and necessitate retraining to handle new objects. We present VLAD-Grasp, a Vision-Language model Assisted zero-shot approach for Detecting grasps. From a single RGB-D image, our method (1) prompts a large vision-language model to generate a goal image where a straight rod "impales" the object, representing an antipodal grasp, (2) predicts depth and segmentation to lift this generated image into 3D, and (3) aligns generated and observed object point clouds via principal component analysis and correspondence-free optimization to recover an executable grasp pose. Unlike prior work, our approach is training-free and does not rely on curated grasp datasets. Despite this, VLAD-Grasp achieves performance that is competitive with or superior to that of state-of-the-art supervised models on the Cornell and Jacquard datasets. We further demonstrate zero-shot generalization to novel real-world objects on a Franka Research 3 robot, highlighting vision-language foundation models as powerful priors for robotic manipulation.

Paperid: 5109, https://arxiv.org/pdf/2511.03400.pdf

Abstract:
Pre-trained robot policies serve as the foundation of many validated robotic systems, which encapsulate extensive embodied knowledge. However, they often lack the semantic awareness characteristic of foundation models, and replacing them entirely is impractical in many situations due to high costs and the loss of accumulated knowledge. To address this gap, we introduce GUIDES, a lightweight framework that augments pre-trained policies with semantic guidance from foundation models without requiring architectural redesign. GUIDES employs a fine-tuned vision-language model (Instructor) to generate contextual instructions, which are encoded by an auxiliary module into guidance embeddings. These embeddings are injected into the policy's latent space, allowing the legacy model to adapt to this new semantic input through brief, targeted fine-tuning. For inference-time robustness, a large language model-based Reflector monitors the Instructor's confidence and, when confidence is low, initiates a reasoning loop that analyzes execution history, retrieves relevant examples, and augments the VLM's context to refine subsequent actions. Extensive validation in the RoboCasa simulation environment across diverse policy architectures shows consistent and substantial improvements in task success rates. Real-world deployment on a UR5 robot further demonstrates that GUIDES enhances motion precision for critical sub-tasks such as grasping. Overall, GUIDES offers a practical and resource-efficient pathway to upgrade, rather than replace, validated robot policies.

Paperid: 5110, https://arxiv.org/pdf/2511.00945.pdf

Abstract:
Vision-Language Models (VLMs) enable on-demand visual assistance, yet current applications for people with visual impairments (PVI) impose high cognitive load and exhibit task drift, limiting real-world utility. We first conducted a formative study with 15 PVI and identified three requirements for visually impaired assistance (VIA): low latency for real-time use, minimal cognitive load, and hallucination-resistant responses to sustain trust. Informed by the formative study, we present VIA-Agent, a prototype that co-optimizes its cognitive 'brain' and interactive 'body'. The brain implements a goal-persistent design with calibrated conciseness to produce brief, actionable guidance; the body adopts a real-time communication (RTC) embodiment-evolving from a request-response model Context Protocol (MCP) pipeline-to-support fluid interaction. We evaluated VIA-Agent with 9 PVI across navigation and object retrieval in the wild against BeMyAI and Doubao. VIA-Agent significantly outperformed BeMyAI both quantitatively and qualitatively. While achieving success rates comparable to Doubao, it reduced mean task time by 39.9% (70.1 s vs. 110.7 s), required fewer conversational turns (4.3 vs. 5.0), and lowered perceived cognitive load and task drift. System Usability Scale (SUS) results aligned with these findings, with VIA-Agent achieving the highest usability. We hope this work inspires the development of more human-centered VIA systems.

Paperid: 5111, https://arxiv.org/pdf/2510.27545.pdf

Abstract:
Implicit policies parameterized by generative models, such as Diffusion Policy, have become the standard for policy learning and Vision-Language-Action (VLA) models in robotics. However, these approaches often suffer from high computational cost, exposure bias, and unstable inference dynamics, which lead to divergence under distribution shifts. Energy-Based Models (EBMs) address these issues by learning energy landscapes end-to-end and modeling equilibrium dynamics, offering improved robustness and reduced exposure bias. Yet, policies parameterized by EBMs have historically struggled to scale effectively. Recent work on Energy-Based Transformers (EBTs) demonstrates the scalability of EBMs to high-dimensional spaces, but their potential for solving core challenges in physically embodied models remains underexplored. We introduce a new energy-based architecture, EBT-Policy, that solves core issues in robotic and real-world settings. Across simulated and real-world tasks, EBT-Policy consistently outperforms diffusion-based policies, while requiring less training and inference computation. Remarkably, on some tasks it converges within just two inference steps, a 50x reduction compared to Diffusion Policy's 100. Moreover, EBT-Policy exhibits emergent capabilities not seen in prior models, such as zero-shot recovery from failed action sequences using only behavior cloning and without explicit retry training. By leveraging its scalar energy for uncertainty-aware inference and dynamic compute allocation, EBT-Policy offers a promising path toward robust, generalizable robot behavior under distribution shifts.

Paperid: 5112, https://arxiv.org/pdf/2510.26824.pdf

Abstract:
The development of synthesis procedures remains a fundamental challenge in materials discovery, with procedural knowledge scattered across decades of scientific literature in unstructured formats that are challenging for systematic analysis. In this paper, we propose a multi-modal toolbox that employs large language models (LLMs) and vision language models (VLMs) to automatically extract and organize synthesis procedures and performance data from materials science publications, covering text and figures. We curated 81k open-access papers, yielding LeMat-Synth (v 1.0): a dataset containing synthesis procedures spanning 35 synthesis methods and 16 material classes, structured according to an ontology specific to materials science. The extraction quality is rigorously evaluated on a subset of 2.5k synthesis procedures through a combination of expert annotations and a scalable LLM-as-a-judge framework. Beyond the dataset, we release a modular, open-source software library designed to support community-driven extension to new corpora and synthesis domains. Altogether, this work provides an extensible infrastructure to transform unstructured literature into machine-readable information. This lays the groundwork for predictive modeling of synthesis procedures as well as modeling synthesis--structure--property relationships.

Paperid: 5113, https://arxiv.org/pdf/2510.26769.pdf

Abstract:
This work introduces SteerVLM, a lightweight steering module designed to guide Vision-Language Models (VLMs) towards outputs that better adhere to desired instructions. Our approach learns from the latent embeddings of paired prompts encoding target and converse behaviors to dynamically adjust activations connecting the language modality with image context. This allows for fine-grained, inference-time control over complex output semantics without modifying model weights while preserving performance on off-target tasks. Our steering module requires learning parameters equal to 0.14% of the original VLM's size. Our steering module gains model control through dimension-wise activation modulation and adaptive steering across layers without requiring pre-extracted static vectors or manual tuning of intervention points. Furthermore, we introduce VNIA (Visual Narrative Intent Alignment), a multimodal dataset specifically created to facilitate the development and evaluation of VLM steering techniques. Our method outperforms existing intervention techniques on steering and hallucination mitigation benchmarks for VLMs and proposes a robust solution for multimodal model control through activation engineering.

Paperid: 5114, https://arxiv.org/pdf/2510.22964.pdf

Abstract:
Foundation models have transformed natural language processing and computer vision, and their impact is now reshaping remote sensing image analysis. With powerful generalization and transfer learning capabilities, they align naturally with the multimodal, multi-resolution, and multi-temporal characteristics of remote sensing data. To address unique challenges in the field, multimodal geospatial foundation models (GFMs) have emerged as a dedicated research frontier. This survey delivers a comprehensive review of multimodal GFMs from a modality-driven perspective, covering five core visual and vision-language modalities. We examine how differences in imaging physics and data representation shape interaction design, and we analyze key techniques for alignment, integration, and knowledge transfer to tackle modality heterogeneity, distribution shifts, and semantic gaps. Advances in training paradigms, architectures, and task-specific adaptation strategies are systematically assessed alongside a wealth of emerging benchmarks. Representative multimodal visual and vision-language GFMs are evaluated across ten downstream tasks, with insights into their architectures, performance, and application scenarios. Real-world case studies, spanning land cover mapping, agricultural monitoring, disaster response, climate studies, and geospatial intelligence, demonstrate the practical potential of GFMs. Finally, we outline pressing challenges in domain generalization, interpretability, efficiency, and privacy, and chart promising avenues for future research.

Paperid: 5115, https://arxiv.org/pdf/2510.22868.pdf

Abstract:
Wind turbine blades operate in harsh environments, making timely damage detection essential for preventing failures and optimizing maintenance. Drone-based inspection and deep learning are promising, but typically depend on large, labeled datasets, which limit their ability to detect rare or evolving damage types. To address this, we propose a zero-shot-oriented inspection framework that integrates Retrieval-Augmented Generation (RAG) with Vision-Language Models (VLM). A multimodal knowledge base is constructed, comprising technical documentation, representative reference images, and domain-specific guidelines. A hybrid text-image retriever with keyword-aware reranking assembles the most relevant context to condition the VLM at inference, injecting domain knowledge without task-specific training. We evaluate the framework on 30 labeled blade images covering diverse damage categories. Although the dataset is small due to the difficulty of acquiring verified blade imagery, it covers multiple representative defect types. On this test set, the RAG-grounded VLM correctly classified all samples, whereas the same VLM without retrieval performed worse in both accuracy and precision. We further compare against open-vocabulary baselines and incorporate uncertainty Clopper-Pearson confidence intervals to account for the small-sample setting. Ablation studies indicate that the key advantage of the framework lies in explainability and generalizability: retrieved references ground the reasoning process and enable the detection of previously unseen defects by leveraging domain knowledge rather than relying solely on visual cues. This research contributes a data-efficient solution for industrial inspection that reduces dependence on extensive labeled datasets.

Paperid: 5116, https://arxiv.org/pdf/2510.22067.pdf

Abstract:
Vision language models (VLMs) often generate hallucination, i.e., content that cannot be substantiated by either textual or visual inputs. Prior work primarily attributes this to over-reliance on linguistic prior knowledge rather than visual inputs. Some methods attempt to mitigate hallucination by amplifying visual token attention proportionally to their attention scores. However, these methods overlook the visual attention sink problem, where attention is frequently misallocated to task-irrelevant visual regions, and neglect cross-modal fusion balance by enhancing only visual attention without adjusting attention to the user query. This can result in amplifying incorrect areas while failing to properly interpret the user query. To address these challenges, we propose a simple yet effective method called Gaze Shift-Guided Cross-modal Fusion Enhancement (GIFT). GIFT pre-computes a holistic visual saliency map by tracking positive changes in visual attention, or "gaze shifts", during user query comprehension, and leverages this map to amplify attention to both salient visual information and the user query at each decoding step. This reduces the impact of visual attention sink, as irrelevant tokens exhibit minimal shifts, while ensuring balanced cross-modal fusion for well-integrated representation. Extensive experiments show that GIFT effectively mitigates hallucination in VLMs across both generative and classification tasks, achieving up to 20.7% improvement over greedy decoding, while maintaining general vision-language performance with low computational overhead.

Paperid: 5117, https://arxiv.org/pdf/2510.19752.pdf

Abstract:
Solving complex real-world control tasks often takes multiple tries: if we fail at first, we reflect on what went wrong, and change our strategy accordingly to avoid making the same mistake. In robotics, Vision-Language-Action models (VLAs) offer a promising path towards solving complex control tasks, but lack the ability to contextually and dynamically readjust behavior when they fail to accomplish a task. In this work, we introduce Learning from Inference-Time Execution (LITEN), which connects a VLA low-level policy to a high-level VLM that conditions on past experiences by including them in-context, allowing it to learn the affordances and capabilities of the low-level VLA. Our approach iterates between a reasoning phase that generates and executes plans for the low-level VLA, and an assessment phase that reflects on the resulting execution and draws useful conclusions to be included in future reasoning contexts. Unlike similar approaches to self-refinement in non-robotics domains, LITEN must reflect on unstructured real-world robot trajectories (e.g., raw videos), which requires structured guiderails during assessment. Our experimental results demonstrate LITEN is able to effectively learn from past experience to generate plans that use high-affordance instructions to accomplish long-horizon tasks.

Paperid: 5118, https://arxiv.org/pdf/2510.19678.pdf

Abstract:
Multimodal large language models (MLLMs) achieve strong performance on vision-language tasks, yet their visual processing is opaque. Most black-box evaluations measure task accuracy, but reveal little about underlying mechanisms. Drawing on cognitive psychology, we adapt classic visual search paradigms -- originally developed to study human perception -- to test whether MLLMs exhibit the ``pop-out'' effect, where salient visual features are detected independently of distractor set size. Using controlled experiments targeting colour, size and lighting features, we find that advanced MLLMs exhibit human-like pop-out effects in colour or size-based disjunctive (single feature) search, as well as capacity limits for conjunctive (multiple feature) search. We also find evidence to suggest that MLLMs, like humans, incorporate natural scene priors such as lighting direction into object representations. We reinforce our findings using targeted fine-tuning and mechanistic interpretability analyses. Our work shows how visual search can serve as a cognitively grounded diagnostic tool for evaluating perceptual capabilities in MLLMs.

Paperid: 5119, https://arxiv.org/pdf/2510.18337.pdf

Abstract:
Integrating visual-language instructions into visuomotor policies is gaining momentum in robot learning for enhancing open-world generalization. Despite promising advances, existing approaches face two challenges: limited language steerability when no generated reasoning is used as a condition, or significant inference latency when reasoning is incorporated. In this work, we introduce MoTVLA, a mixture-of-transformers (MoT)-based vision-language-action (VLA) model that integrates fast-slow unified reasoning with behavior policy learning. MoTVLA preserves the general intelligence of pre-trained VLMs (serving as the generalist) for tasks such as perception, scene understanding, and semantic planning, while incorporating a domain expert, a second transformer that shares knowledge with the pretrained VLM, to generate domain-specific fast reasoning (e.g., robot motion decomposition), thereby improving policy execution efficiency. By conditioning the action expert on decomposed motion instructions, MoTVLA can learn diverse behaviors and substantially improve language steerability. Extensive evaluations across natural language processing benchmarks, robotic simulation environments, and real-world experiments confirm the superiority of MoTVLA in both fast-slow reasoning and manipulation task performance.

Paperid: 5120, https://arxiv.org/pdf/2510.17875.pdf

Abstract:
3D weakly supervised semantic segmentation (3D WSSS) aims to achieve semantic segmentation by leveraging sparse or low-cost annotated data, significantly reducing reliance on dense point-wise annotations. Previous works mainly employ class activation maps or pre-trained vision-language models to address this challenge. However, the low quality of pseudo-labels and the insufficient exploitation of 3D geometric priors jointly create significant technical bottlenecks in developing high-performance 3D WSSS models. In this paper, we propose a simple yet effective 3D weakly supervised semantic segmentation method that integrates 3D geometric priors into a class-aware guidance mechanism to generate high-fidelity pseudo labels. Concretely, our designed methodology first employs Class-Aware Label Refinement module to generate more balanced and accurate pseudo labels for semantic categrories. This initial refinement stage focuses on enhancing label quality through category-specific optimization. Subsequently, the Geometry-Aware Label Refinement component is developed, which strategically integrates implicit 3D geometric constraints to effectively filter out low-confidence pseudo labels that fail to comply with geometric plausibility. Moreover, to address the challenge of extensive unlabeled regions, we propose a Label Update strategy that integrates Self-Training to propagate labels into these areas. This iterative process continuously enhances pseudo-label quality while expanding label coverage, ultimately fostering the development of high-performance 3D WSSS models. Comprehensive experimental validation reveals that our proposed methodology achieves state-of-the-art performance on both ScanNet and S3DIS benchmarks while demonstrating remarkable generalization capability in unsupervised settings, maintaining competitive accuracy through its robust design.

Paperid: 5121, https://arxiv.org/pdf/2510.17759.pdf

Abstract:
Vision-Language Models (VLMs) extend large language models with visual reasoning, but their multimodal design also introduces new, underexplored vulnerabilities. Existing multimodal red-teaming methods largely rely on brittle templates, focus on single-attack settings, and expose only a narrow subset of vulnerabilities. To address these limitations, we introduce VERA-V, a variational inference framework that recasts multimodal jailbreak discovery as learning a joint posterior distribution over paired text-image prompts. This probabilistic view enables the generation of stealthy, coupled adversarial inputs that bypass model guardrails. We train a lightweight attacker to approximate the posterior, allowing efficient sampling of diverse jailbreaks and providing distributional insights into vulnerabilities. VERA-V further integrates three complementary strategies: (i) typography-based text prompts that embed harmful cues, (ii) diffusion-based image synthesis that introduces adversarial signals, and (iii) structured distractors to fragment VLM attention. Experiments on HarmBench and HADES benchmarks show that VERA-V consistently outperforms state-of-the-art baselines on both open-source and frontier VLMs, achieving up to 53.75% higher attack success rate (ASR) over the best baseline on GPT-4o.

Paperid: 5122, https://arxiv.org/pdf/2510.16540.pdf

Abstract:
Despite recent advances, vision-language models trained with standard contrastive objectives still struggle with compositional reasoning -- the ability to understand structured relationships between visual and linguistic elements. This shortcoming is largely due to the tendency of the text encoder to focus on individual words rather than their relations, a limitation reinforced by contrastive training that primarily aligns words with visual objects. In this paper, we introduce REconstruction and Alignment of text Descriptions (READ), a fine-tuning method designed to enhance compositional reasoning by adding two auxiliary objectives to the contrastive learning: (1) a token-level reconstruction objective, where a frozen pre-trained decoder reconstructs alternative captions based on the embedding of the original caption; and (2) a sentence-level alignment objective, which explicitly aligns paraphrased sentences in the embedding space. We show that READ-CLIP, a model derived by applying the READ method to the pre-trained CLIP model, achieves the state-of-the-art performance across five major compositional reasoning benchmarks, outperforming the strongest conventional fine-tuning baseline by up to 4.1%. Furthermore, applying the READ to existing CLIP variants (including NegCLIP and FSC-CLIP) also improves performance on these benchmarks. Quantitative and qualitative analyses reveal that our proposed objectives -- reconstruction and alignment -- offer complementary benefits: the former encourages the encoder to capture relationships between words within a caption, while the latter ensures consistent representations for paraphrases expressed with different wording.

Paperid: 5123, https://arxiv.org/pdf/2510.09762.pdf

Abstract:
Patent drafting is complex due to its need for detailed technical descriptions, legal compliance, and visual elements. Although Large Vision Language Models (LVLMs) show promise across various tasks, their application in automating patent writing remains underexplored. In this paper, we present PatentVision, a multimodal framework that integrates textual and visual inputs such as patent claims and drawings to generate complete patent specifications. Built on advanced LVLMs, PatentVision enhances accuracy by combining fine tuned vision language models with domain specific training tailored to patents. Experiments reveal it surpasses text only methods, producing outputs with greater fidelity and alignment with human written standards. Its incorporation of visual data allows it to better represent intricate design features and functional connections, leading to richer and more precise results. This study underscores the value of multimodal techniques in patent automation, providing a scalable tool to reduce manual workloads and improve consistency. PatentVision not only advances patent drafting but also lays the groundwork for broader use of LVLMs in specialized areas, potentially transforming intellectual property management and innovation processes.

Paperid: 5124, https://arxiv.org/pdf/2510.09699.pdf

Abstract:
Vision-Language Models (VLMs) have garnered significant attention for their remarkable ability to interpret and generate multimodal content. However, securing these models against jailbreak attacks continues to be a substantial challenge. Unlike text-only models, VLMs integrate additional modalities, introducing novel vulnerabilities such as image hijacking, which can manipulate the model into producing inappropriate or harmful responses. Drawing inspiration from text-based jailbreaks like the "Do Anything Now" (DAN) command, this work introduces VisualDAN, a single adversarial image embedded with DAN-style commands. Specifically, we prepend harmful corpora with affirmative prefixes (e.g., "Sure, I can provide the guidance you need") to trick the model into responding positively to malicious queries. The adversarial image is then trained on these DAN-inspired harmful texts and transformed into the text domain to elicit malicious outputs. Extensive experiments on models such as MiniGPT-4, MiniGPT-v2, InstructBLIP, and LLaVA reveal that VisualDAN effectively bypasses the safeguards of aligned VLMs, forcing them to execute a broad range of harmful instructions that severely violate ethical standards. Our results further demonstrate that even a small amount of toxic content can significantly amplify harmful outputs once the model's defenses are compromised. These findings highlight the urgent need for robust defenses against image-based attacks and offer critical insights for future research into the alignment and security of VLMs.

Paperid: 5125, https://arxiv.org/pdf/2510.07632.pdf

Abstract:
Frontier AI models have achieved remarkable progress, yet recent studies suggest they struggle with compositional reasoning, often performing at or below random chance on established benchmarks. We revisit this problem and show that widely used evaluation metrics systematically underestimate model capability. To address this, we introduce a group matching score that better exploits group structure and reveals substantial hidden capability in both contrastive vision-language models (VLMs) and multimodal large language models (MLLMs). Moreover, simply overfitting to the induced group matchings at test time transfers this hidden capability into higher scores under standard evaluation metrics, closing much of the reported gap. This adjustment enables SigLIP-B16 to surpass all previous results and GPT-4.1 to yield the first result surpassing estimated human performance on Winoground. Building on this insight, we propose Test-Time Matching (TTM), an iterative, self-improving algorithm that further bootstraps model performance without any external supervision. TTM delivers additional, non-trivial improvements: for example, TTM enables SigLIP-B16 to surpass GPT-4.1 on MMVP-VLM, establishing a new state of the art. Importantly, TTM remains broadly effective even on benchmarks without metric-induced effects or group structures, achieving relative gains up to 85.7% on challenging datasets such as WhatsUp. Across 16 dataset variants spanning diverse setups, our experiments demonstrate that TTM consistently improves model performance and advances the frontier of compositional reasoning.

Paperid: 5126, https://arxiv.org/pdf/2510.04002.pdf

Abstract:
Despite rapid advances in multimodal large language models, agricultural applications remain constrained by the scarcity of domain-tailored models, curated vision-language corpora, and rigorous evaluation. To address these challenges, we present the AgriGPT-VL Suite, a unified multimodal framework for agriculture. Our contributions are threefold. First, we introduce Agri-3M-VL, the largest vision-language corpus for agriculture to our knowledge, curated by a scalable multi-agent data generator; it comprises 1M image-caption pairs, 2M image-grounded VQA pairs, 50K expert-level VQA instances, and 15K GRPO reinforcement learning samples. Second, we develop AgriGPT-VL, an agriculture-specialized vision-language model trained via a progressive curriculum of textual grounding, multimodal shallow/deep alignment, and GRPO refinement. This method achieves strong multimodal reasoning while preserving text-only capability. Third, we establish AgriBench-VL-4K, a compact yet challenging evaluation suite with open-ended and image-grounded questions, paired with multi-metric evaluation and an LLM-as-a-judge framework. Experiments show that AgriGPT-VL outperforms leading general-purpose VLMs on AgriBench-VL-4K, achieving higher pairwise win rates in the LLM-as-a-judge evaluation. Meanwhile, it remains competitive on the text-only AgriBench-13K with no noticeable degradation of language ability. Ablation studies further confirm consistent gains from our alignment and GRPO refinement stages. We will open source all of the resources to support reproducible research and deployment in low-resource agricultural settings.

Paperid: 5127, https://arxiv.org/pdf/2510.03666.pdf

Abstract:
Industrial accidents, particularly in high-risk domains such as surface and underground mining, are frequently caused by unsafe worker behaviors. Traditional manual inspection remains labor-intensive, error-prone, and insufficient for large-scale, dynamic environments, highlighting the urgent need for intelligent and automated safety monitoring. In this paper, we present MonitorVLM, a novel vision--language framework designed to detect safety violations directly from surveillance video streams. MonitorVLM introduces three key innovations: (1) a domain-specific violation dataset comprising 9,000 vision--question--answer (VQA) samples across 40 high-frequency mining regulations, enriched with augmentation and auxiliary detection cues; (2) a clause filter (CF) module that dynamically selects the Top-$K$ most relevant clauses, reducing inference latency by 13.56\% while maintaining accuracy; and (3) a behavior magnifier (BM) module that enhances worker regions to improve fine-grained action recognition, yielding additional gains of 3.45% in precision and 8.62% in recall. Experimental results demonstrate that MonitorVLM significantly outperforms baseline vision--language models, achieving improvements of 22.01% in precision, 34.22\% in recall, and 28.37% in F1 score over the 72B unfine-tuned baseline. A lightweight web-based interface further integrates MonitorVLM into practical workflows, enabling automatic violation reporting with video timestamping. This study highlights the potential of multimodal large models to enhance occupational safety monitoring in mining and beyond.

Paperid: 5128, https://arxiv.org/pdf/2510.01126.pdf

Abstract:
Large vision-language models (VLMs) are increasingly used in autonomous-vehicle (AV) stacks, but hallucination limits their reliability in safety-critical pipelines. We present Shapley-credited Context-Aware Dawid-Skene with Agreement, a game-theoretic fusion method for multi-label understanding of ego-view dashcam video. It learns per-model, per-label, context-conditioned reliabilities from labelled history and, at inference, converts each model's report into an agreement-guardrailed log-likelihood ratio that is combined with a contextual prior and a public reputation state updated via Shapley-based team credit. The result is calibrated, thresholdable posteriors that (i) amplify agreement among reliable models, (ii) preserve uniquely correct single-model signals, and (iii) adapt to drift. To specialise general VLMs, we curate 1,000 real-world dashcam clips with structured annotations (scene description, manoeuvre recommendation, rationale) via an automatic pipeline that fuses HDD ground truth, vehicle kinematics, and YOLOv11 + BoT-SORT tracking, guided by a three-step chain-of-thought prompt; three heterogeneous VLMs are then fine-tuned with LoRA. We evaluate with Hamming distance, Micro-Macro-F1, and average per-video latency. Empirically, the proposed method achieves a 23% reduction in Hamming distance, 55% improvement in Macro-F1, and 47% improvement in Micro-F1 when comparing with the best single model, supporting VLM fusion as a calibrated, interpretable, and robust decision-support component for AV pipelines.

Paperid: 5129, https://arxiv.org/pdf/2509.26594.pdf

Abstract:
Recent text-only models demonstrate remarkable mathematical reasoning capabilities. Extending these to visual domains requires vision-language models to translate images into text descriptions. However, current models, trained to produce captions for human readers, often omit the precise details that reasoning systems require. This creates an interface mismatch: reasoners often fail not due to reasoning limitations but because they lack access to critical visual information. We propose Adaptive-Clarification Reinforcement Learning (AC-RL), which teaches vision models what information reasoners need through interaction. Our key insight is that clarification requests during training reveal information gaps; by penalizing success that requires clarification, we create pressure for comprehensive initial captions that enable the reasoner to solve the problem in a single pass. AC-RL improves average accuracy by 4.4 points over pretrained baselines across seven visual mathematical reasoning benchmarks, and analysis shows it would cut clarification requests by up to 39% if those were allowed. By treating clarification as a form of implicit supervision, AC-RL demonstrates that vision-language interfaces can be effectively learned through interaction alone, without requiring explicit annotations.

Paperid: 5130, https://arxiv.org/pdf/2509.26330.pdf

Abstract:
Composed Image Retrieval (CIR) aims to retrieve target images that preserve the visual content of a reference image while incorporating user-specified textual modifications. Training-free zero-shot CIR (ZS-CIR) approaches, which require no task-specific training or labeled data, are highly desirable, yet accurately capturing user intent remains challenging. In this paper, we present SQUARE, a novel two-stage training-free framework that leverages Multimodal Large Language Models (MLLMs) to enhance ZS-CIR. In the Semantic Query-Augmented Fusion (SQAF) stage, we enrich the query embedding derived from a vision-language model (VLM) such as CLIP with MLLM-generated captions of the target image. These captions provide high-level semantic guidance, enabling the query to better capture the user's intent and improve global retrieval quality. In the Efficient Batch Reranking (EBR) stage, top-ranked candidates are presented as an image grid with visual marks to the MLLM, which performs joint visual-semantic reasoning across all candidates. Our reranking strategy operates in a single pass and yields more accurate rankings. Experiments show that SQUARE, with its simplicity and effectiveness, delivers strong performance on four standard CIR benchmarks. Notably, it maintains high performance even with lightweight pre-trained, demonstrating its potential applicability.

Paperid: 5131, https://arxiv.org/pdf/2509.25533.pdf

Abstract:
As Vision Language Models (VLMs) are deployed across safety-critical applications, understanding and controlling their behavioral patterns has become increasingly important. Existing behavioral control methods face significant limitations: system prompting approaches could easily be overridden by user instructions, while applying activation-based steering vectors requires invasive runtime access to model internals, precluding deployment with API-based services and closed-source models. Finding steering methods that transfer across multiple VLMs is still an open area of research. To this end, we introduce universal visual input based steering for output redirection (VISOR++), to achieve behavioral control through optimized visual inputs alone. We demonstrate that a single VISOR++ image can be generated for an ensemble of VLMs to emulate each of their steering vectors. By crafting universal visual inputs that induce target activation patterns, VISOR++ eliminates the need for runtime model access while remaining deployment-agnostic. This means that when an underlying model supports multimodal capability, model behaviors can be steered by inserting an image input replacing runtime steering vector based interventions. We first demonstrate the effectiveness of the VISOR++ images on open-access models such as LLaVA-1.5-7B and IDEFICS2-8B along three alignment directions: refusal, sycophancy and survival instinct. Both the model-specific steering images and the jointly optimized images achieve performance parity closely following that of steering vectors for both positive and negative steering tasks. We also show the promise of VISOR++ images in achieving directional behavioral shifts for unseen models including both open-access and closed-access ones. Furthermore, VISOR++ images are able to preserve 99.9% performance on 14,000 unrelated MMLU evaluation tasks.

Paperid: 5132, https://arxiv.org/pdf/2509.23155.pdf

Abstract:
Robotic manipulation benefits from foundation models that describe goals, but today's agents still lack a principled way to learn from their own mistakes. We ask whether natural language can serve as feedback, an error reasoning signal that helps embodied agents diagnose what went wrong and correct course. We introduce LAGEA (Language Guided Embodied Agents), a framework that turns episodic, schema-constrained reflections from a vision language model (VLM) into temporally grounded guidance for reinforcement learning. LAGEA summarizes each attempt in concise language, localizes the decisive moments in the trajectory, aligns feedback with visual state in a shared representation, and converts goal progress and feedback agreement into bounded, step-wise shaping rewardswhose influence is modulated by an adaptive, failure-aware coefficient. This design yields dense signals early when exploration needs direction and gracefully recedes as competence grows. On the Meta-World MT10 embodied manipulation benchmark, LAGEA improves average success over the state-of-the-art (SOTA) methods by 9.0% on random goals and 5.3% on fixed goals, while converging faster. These results support our hypothesis: language, when structured and grounded in time, is an effective mechanism for teaching robots to self-reflect on mistakes and make better choices. Code will be released soon.

Paperid: 5133, https://arxiv.org/pdf/2509.22697.pdf

Abstract:
As data requirements continue to grow, efficient learning increasingly depends on the curation and distillation of high-value data rather than brute-force scaling of model sizes. In the case of a hyperspectral image (HSI), the challenge is amplified by the high-dimensional 3D voxel structure, where each spatial location is associated with hundreds of contiguous spectral channels. While vision and language models have been optimized effectively for natural image or text tasks, their cross-modal alignment in the hyperspectral domain remains an open and underexplored problem. In this article, we make an attempt to optimize a Vision-Language Model (VLM) for hyperspectral scene understanding by exploiting a CLIP-style contrastive training framework. Our framework maps voxel-level embeddings from a vision backbone onto the latent space of a frozen large embedding model (LEM), where a trainable probe aligns vision features with the model's textual token representations. The two modalities are aligned via a contrastive loss restricted to a curated set of hard (closest wrong classes) and semi-hard (random distractors) negatives, along with positive pairs. To further enhance alignment, descriptive prompts that encode class semantics are introduced and act as structured anchors for the HSI embeddings. It is seen that the proposed method updates only 0.07 percent of the total parameters, yet yields state-of-the-art performance. For example, on Indian Pines (IP) the model produces better results over unimodal and multimodal baselines by +0.92 Overall Accuracy (OA) and +1.60 Kappa ($Îº$), while on Pavia University (PU) data it provides gains of +0.69 OA and +0.90 $Îº$. Moreover, this is achieved with the set of parameters, nearly 50$\times$ smaller than DCTN and 90$\times$ smaller than SS-TMNet.

Paperid: 5134, https://arxiv.org/pdf/2509.22441.pdf

Abstract:
This paper presents UnderwaterVLA, a novel framework for autonomous underwater navigation that integrates multimodal foundation models with embodied intelligence systems. Underwater operations remain difficult due to hydrodynamic disturbances, limited communication bandwidth, and degraded sensing in turbid waters. To address these challenges, we introduce three innovations. First, a dual-brain architecture decouples high-level mission reasoning from low-level reactive control, enabling robust operation under communication and computational constraints. Second, we apply Vision-Language-Action(VLA) models to underwater robotics for the first time, incorporating structured chain-of-thought reasoning for interpretable decision-making. Third, a hydrodynamics-informed Model Predictive Control(MPC) scheme compensates for fluid effects in real time without costly task-specific training. Experimental results in field tests show that UnderwaterVLA reduces navigation errors in degraded visual conditions while maintaining higher task completion by 19% to 27% over baseline. By minimizing reliance on underwater-specific training data and improving adaptability across environments, UnderwaterVLA provides a scalable and cost-effective path toward the next generation of intelligent AUVs.

Paperid: 5135, https://arxiv.org/pdf/2509.20077.pdf

Abstract:
To enable robots to comprehend high-level human instructions and perform complex tasks, a key challenge lies in achieving comprehensive scene understanding: interpreting and interacting with the 3D environment in a meaningful way. This requires a smart map that fuses accurate geometric structure with rich, human-understandable semantics. To address this, we introduce the 3D Queryable Scene Representation (3D QSR), a novel framework built on multimedia data that unifies three complementary 3D representations: (1) 3D-consistent novel view rendering and segmentation from panoptic reconstruction, (2) precise geometry from 3D point clouds, and (3) structured, scalable organization via 3D scene graphs. Built on an object-centric design, the framework integrates with large vision-language models to enable semantic queryability by linking multimodal object embeddings, and supporting object-level retrieval of geometric, visual, and semantic information. The retrieved data are then loaded into a robotic task planner for downstream execution. We evaluate our approach through simulated robotic task planning scenarios in Unity, guided by abstract language instructions and using the indoor public dataset Replica. Furthermore, we apply it in a digital duplicate of a real wet lab environment to test QSR-supported robotic task planning for emergency response. The results demonstrate the framework's ability to facilitate scene understanding and integrate spatial and semantic reasoning, effectively translating high-level human instructions into precise robotic task planning in complex 3D environments.

Paperid: 5136, https://arxiv.org/pdf/2509.18111.pdf

Abstract:
The reliability of artificial intelligence (AI) systems in open-world settings depends heavily on their ability to flag out-of-distribution (OOD) inputs unseen during training. Recent advances in large-scale vision-language models (VLMs) have enabled promising few-shot OOD detection frameworks using only a handful of in-distribution (ID) samples. However, existing prompt learning-based OOD methods rely solely on softmax probabilities, overlooking the rich discriminative potential of the feature embeddings learned by VLMs trained on millions of samples. To address this limitation, we propose a novel context optimization (CoOp)-based framework that integrates subspace representation learning with prompt tuning. Our approach improves ID-OOD separability by projecting the ID features into a subspace spanned by prompt vectors, while projecting ID-irrelevant features into an orthogonal null space. To train such OOD detection framework, we design an easy-to-handle end-to-end learning criterion that ensures strong OOD detection performance as well as high ID classification accuracy. Experiments on real-world datasets showcase the effectiveness of our approach.

Paperid: 5137, https://arxiv.org/pdf/2509.17429.pdf

Abstract:
Accurate temporal prediction is the bridge between comprehensive scene understanding and embodied artificial intelligence. However, predicting multiple fine-grained states of a scene at multiple temporal scales is difficult for vision-language models. We formalize the Multi-Scale Temporal Prediction (MSTP) task in general and surgical scenes by decomposing multi-scale into two orthogonal dimensions: the temporal scale, forecasting states of humans and surgery at varying look-ahead intervals, and the state scale, modeling a hierarchy of states in general and surgical scenes. For example, in general scenes, states of contact relationships are finer-grained than states of spatial relationships. In surgical scenes, medium-level steps are finer-grained than high-level phases yet remain constrained by their encompassing phase. To support this unified task, we introduce the first MSTP Benchmark, featuring synchronized annotations across multiple state scales and temporal scales. We further propose a method, Incremental Generation and Multi-agent Collaboration (IG-MC), which integrates two key innovations. First, we present a plug-and-play incremental generation module that continuously synthesizes up-to-date visual previews at expanding temporal scales to inform multiple decision-making agents, keeping decisions and generated visuals synchronized and preventing performance degradation as look-ahead intervals lengthen. Second, we present a decision-driven multi-agent collaboration framework for multi-state prediction, comprising generation, initiation, and multi-state assessment agents that dynamically trigger and evaluate prediction cycles to balance global coherence and local fidelity.

Paperid: 5138, https://arxiv.org/pdf/2509.17374.pdf

Abstract:
Large-scale vision language pre-training has recently shown promise for no-reference image-quality assessment (NR-IQA), yet the relative merits of modern Vision Transformer foundations remain poorly understood. In this work, we present the first systematic evaluation of six prominent pretrained backbones, CLIP, SigLIP2, DINOv2, DINOv3, Perception, and ResNet, for the task of No-Reference Image Quality Assessment (NR-IQA), each finetuned using an identical lightweight MLP head. Our study uncovers two previously overlooked factors: (1) SigLIP2 consistently achieves strong performance; and (2) the choice of activation function plays a surprisingly crucial role, particularly for enhancing the generalization ability of image quality assessment models. Notably, we find that simple sigmoid activations outperform commonly used ReLU and GELU on several benchmarks. Motivated by this finding, we introduce a learnable activation selection mechanism that adaptively determines the nonlinearity for each channel, eliminating the need for manual activation design, and achieving new state-of-the-art SRCC on CLIVE, KADID10K, and AGIQA3K. Extensive ablations confirm the benefits across architectures and regimes, establishing strong, resource-efficient NR-IQA baselines.

Paperid: 5139, https://arxiv.org/pdf/2509.16053.pdf

Abstract:
A key requirement for generalist robots is compositional generalization - the ability to combine atomic skills to solve complex, long-horizon tasks. While prior work has primarily focused on synthesizing a planner that sequences pre-learned skills, robust execution of the individual skills themselves remains challenging, as visuomotor policies often fail under distribution shifts induced by scene composition. To address this, we introduce a scene graph-based representation that focuses on task-relevant objects and relations, thereby mitigating sensitivity to irrelevant variation. Building on this idea, we develop a scene-graph skill learning framework that integrates graph neural networks with diffusion-based imitation learning, and further combine "focused" scene-graph skills with a vision-language model (VLM) based task planner. Experiments in both simulation and real-world manipulation tasks demonstrate substantially higher success rates than state-of-the-art baselines, highlighting improved robustness and compositional generalization in long-horizon tasks.

Paperid: 5140, https://arxiv.org/pdf/2509.15937.pdf

Abstract:
Robotic real-world reinforcement learning (RL) with vision-language-action (VLA) models is bottlenecked by sparse, handcrafted rewards and inefficient exploration. We introduce VLAC, a general process reward model built upon InternVL and trained on large scale heterogeneous datasets. Given pairwise observations and a language goal, it outputs dense progress delta and done signal, eliminating task-specific reward engineering, and supports one-shot in-context transfer to unseen tasks and environments. VLAC is trained on vision-language datasets to strengthen perception, dialogic and reasoning capabilities, together with robot and human trajectories data that ground action generation and progress estimation, and additionally strengthened to reject irrelevant prompts as well as detect regression or stagnation by constructing large numbers of negative and semantically mismatched samples. With prompt control, a single VLAC model alternately generating reward and action tokens, unifying critic and policy. Deployed inside an asynchronous real-world RL loop, we layer a graded human-in-the-loop protocol (offline demonstration replay, return and explore, human guided explore) that accelerates exploration and stabilizes early learning. Across four distinct real-world manipulation tasks, VLAC lifts success rates from about 30\% to about 90\% within 200 real-world interaction episodes; incorporating human-in-the-loop interventions yields a further 50% improvement in sample efficiency and achieves up to 100% final success.

Paperid: 5141, https://arxiv.org/pdf/2509.15803.pdf

Abstract:
Text-to-image (T2I) models exhibit a significant yet under-explored "brand bias", a tendency to generate contents featuring dominant commercial brands from generic prompts, posing ethical and legal risks. We propose CIDER, a novel, model-agnostic framework to mitigate bias at inference-time through prompt refinement to avoid costly retraining. CIDER uses a lightweight detector to identify branded content and a Vision-Language Model (VLM) to generate stylistically divergent alternatives. We introduce the Brand Neutrality Score (BNS) to quantify this issue and perform extensive experiments on leading T2I models. Results show CIDER significantly reduces both explicit and implicit biases while maintaining image quality and aesthetic appeal. Our work offers a practical solution for more original and equitable content, contributing to the development of trustworthy generative AI.

Paperid: 5142, https://arxiv.org/pdf/2509.15738.pdf

Abstract:
Graphical User Interface (GUI) Agents, powered by large language and vision-language models, hold promise for enabling end-to-end automation in digital environments. However, their progress is fundamentally constrained by the scarcity of scalable, high-quality trajectory data. Existing data collection strategies either rely on costly and inconsistent manual annotations or on synthetic generation methods that trade off between diversity and meaningful task coverage. To bridge this gap, we present GUI-ReWalk: a reasoning-enhanced, multi-stage framework for synthesizing realistic and diverse GUI trajectories. GUI-ReWalk begins with a stochastic exploration phase that emulates human trial-and-error behaviors, and progressively transitions into a reasoning-guided phase where inferred goals drive coherent and purposeful interactions. Moreover, it supports multi-stride task generation, enabling the construction of long-horizon workflows across multiple applications. By combining randomness for diversity with goal-aware reasoning for structure, GUI-ReWalk produces data that better reflects the intent-aware, adaptive nature of human-computer interaction. We further train Qwen2.5-VL-7B on the GUI-ReWalk dataset and evaluate it across multiple benchmarks, including Screenspot-Pro, OSWorld-G, UI-Vision, AndroidControl, and GUI-Odyssey. Results demonstrate that GUI-ReWalk enables superior coverage of diverse interaction flows, higher trajectory entropy, and more realistic user intent. These findings establish GUI-ReWalk as a scalable and data-efficient framework for advancing GUI agent research and enabling robust real-world automation.

Paperid: 5143, https://arxiv.org/pdf/2509.14380.pdf

Abstract:
Multi-Agent Reinforcement Learning (MARL) provides a powerful framework for learning coordination in multi-agent systems. However, applying MARL to robotics still remains challenging due to high-dimensional continuous joint action spaces, complex reward design, and non-stationary transitions inherent to decentralized settings. On the other hand, humans learn complex coordination through staged curricula, where long-horizon behaviors are progressively built upon simpler skills. Motivated by this, we propose CRAFT: Coaching Reinforcement learning Autonomously using Foundation models for multi-robot coordination Tasks, a framework that leverages the reasoning capabilities of foundation models to act as a "coach" for multi-robot coordination. CRAFT automatically decomposes long-horizon coordination tasks into sequences of subtasks using the planning capability of Large Language Models (LLMs). In what follows, CRAFT trains each subtask using reward functions generated by LLM, and refines them through a Vision Language Model (VLM)-guided reward-refinement loop. We evaluate CRAFT on multi-quadruped navigation and bimanual manipulation tasks, demonstrating its capability to learn complex coordination behaviors. In addition, we validate the multi-quadruped navigation policy in real hardware experiments.

Paperid: 5144, https://arxiv.org/pdf/2509.14172.pdf

Abstract:
With the rapid advancement of large language models and vision-language models, employing large models as Web Agents has become essential for automated web interaction. However, training Web Agents with reinforcement learning faces critical challenges including credit assignment misallocation, prohibitively high annotation costs, and reward sparsity. To address these issues, we propose Tree-Guided Preference Optimization (TGPO), an offline reinforcement learning framework that proposes a tree-structured trajectory representation merging semantically identical states across trajectories to eliminate label conflicts. Our framework incorporates a Process Reward Model that automatically generates fine-grained rewards through subgoal progress, redundancy detection, and action verification. Additionally, a dynamic weighting mechanism prioritizes high-impact decision points during training. Experiments on Online-Mind2Web and our self-constructed C-WebShop datasets demonstrate that TGPO significantly outperforms existing methods, achieving higher success rates with fewer redundant steps.

Paperid: 5145, https://arxiv.org/pdf/2509.13774.pdf

Abstract:
Vision-language-action (VLA) models demonstrate strong generalization in robotic manipulation but face challenges in complex, real-world tasks. While supervised fine-tuning with demonstrations is constrained by data quality, reinforcement learning (RL) offers a promising alternative. We propose a human-in-the-loop dual-actor fine-tuning framework grounded in RL. The framework integrates a primary actor for robust multi-task performance with a refinement actor for latent-space adaptation. Beyond standard physical interventions, we introduce a lightweight talk-and-tweak scheme that converts human corrections into semantically grounded language commands, thereby generating a new dataset for policy learning. In real-world multi-task experiments, our approach achieves 100% success across three tasks within 101 minutes of online fine-tuning. For long-horizon tasks, it sustains a 50% success rate over 12 consecutive operations. Furthermore, the framework scales effectively to multi-robot training, achieving up to a 2 times improvement in efficiency when using dual robots. The experiment videos are available at https://sites.google.com/view/hil-daft/.

Paperid: 5146, https://arxiv.org/pdf/2509.10278.pdf

Abstract:
Recent works have shown the effectiveness of Large Vision Language Models (VLMs or LVLMs) in image manipulation detection. However, text manipulation detection is largely missing in these studies. We bridge this knowledge gap by analyzing closed- and open-source VLMs on different text manipulation datasets. Our results suggest that open-source models are getting closer, but still behind closed-source ones like GPT- 4o. Additionally, we benchmark image manipulation detection-specific VLMs for text manipulation detection and show that they suffer from the generalization problem. We benchmark VLMs for manipulations done on in-the-wild scene texts and on fantasy ID cards, where the latter mimic a challenging real-world misuse.

Paperid: 5147, https://arxiv.org/pdf/2509.06759.pdf

Abstract:
Large Vision-Language Models (LVLMs) or multimodal large language models represent a significant advancement in artificial intelligence, enabling systems to understand and generate content across both visual and textual modalities. While large-scale pretraining has driven substantial progress, fine-tuning these models for aligning with human values or engaging in specific tasks or behaviors remains a critical challenge. Deep Reinforcement Learning (DRL) and Direct Preference Optimization (DPO) offer promising frameworks for this aligning process. While DRL enables models to optimize actions using reward signals instead of relying solely on supervised preference data, DPO directly aligns the policy with preferences, eliminating the need for an explicit reward model. This overview explores paradigms for fine-tuning LVLMs, highlighting how DRL and DPO techniques can be used to align models with human preferences and values, improve task performance, and enable adaptive multimodal interaction. We categorize key approaches, examine sources of preference data, reward signals, and discuss open challenges such as scalability, sample efficiency, continual learning, generalization, and safety. The goal is to provide a clear understanding of how DRL and DPO contribute to the evolution of robust and human-aligned LVLMs.

Paperid: 5148, https://arxiv.org/pdf/2509.00210.pdf

Abstract:
Achieving human-like reasoning in deep learning models for complex tasks in unknown environments remains a critical challenge in embodied intelligence. While advanced vision-language models (VLMs) excel in static scene understanding, their limitations in spatio-temporal reasoning and adaptation to dynamic, open-set tasks like task-oriented navigation and embodied question answering (EQA) persist due to inadequate modeling of fine-grained spatio-temporal cues and physical world comprehension. To address this, we propose VEME, a novel cross-modal alignment method that enhances generalization in unseen scenes by learning an ego-centric, experience-centered world model. Our framework integrates three key components: (1) a cross-modal alignment framework bridging objects, spatial representations, and visual semantics with spatio-temporal cues to enhance VLM in-context learning; (2) a dynamic, implicit cognitive map activated by world embedding to enable task-relevant geometric-semantic memory recall; and (3) an instruction-based navigation and reasoning framework leveraging embodied priors for long-term planning and efficient exploration. By embedding geometry-aware spatio-temporal episodic experiences, our method significantly improves reasoning and planning in dynamic environments. Experimental results on VSI-Bench and VLN-CE demonstrate 1%-3% accuracy and exploration efficiency improvement compared to traditional approaches.

Paperid: 5149, https://arxiv.org/pdf/2508.19391.pdf

Abstract:
Visual-textual understanding is essential for language-guided robot manipulation. Recent works leverage pre-trained vision-language models to measure the similarity between encoded visual observations and textual instructions, and then train a model to map this similarity to robot actions. However, this two-step approach limits the model to capture the relationship between visual observations and textual instructions, leading to reduced precision in manipulation tasks. We propose to learn visual-textual associations through a self-supervised pretext task: reconstructing a masked goal image conditioned on an input image and textual instructions. This formulation allows the model to learn visual-action representations without robot action supervision. The learned representations can then be fine-tuned for manipulation tasks with only a few demonstrations. We also introduce the \textit{Omni-Object Pick-and-Place} dataset, which consists of annotated robot tabletop manipulation episodes, including 180 object classes and 3,200 instances with corresponding textual instructions. This dataset enables the model to acquire diverse object priors and allows for a more comprehensive evaluation of its generalisation capability across object instances. Experimental results on the five benchmarks, including both simulated and real-robot validations, demonstrate that our method outperforms prior art.

Paperid: 5150, https://arxiv.org/pdf/2508.17394.pdf

Abstract:
Clinical decision-making often involves interpreting images (e.g., radiology) for making diagnoses. Retrieving relevant visual information from medical literature and hospital records could enhance diagnostic accuracy. In this paper, we develop a model in which a multimodal retriever is jointly optimized with an LVLM for medical diagnosis, unlike standard RAG where LVLM error signal is not propagated down to the retriever. We show that using only general-purpose backbones, with only lightweight fine-tuning, our model is able to achieve competitive results with medically-pretrained models across clinical multi-label classification and visual question answering tasks. In a novel analysis, we additionally find that in many cases different top retrieved images each lead to different predictions for a given target, and that these cases are empirically challenging for all models, even for non-retrieval models. Our joint retrieval optimization significantly improves these challenging cases over standard RAG. However, oracle analysis reveals that while the correct diagnosis is frequently achievable using one of the top retrieved images, in practice there is a large performance gap from the oracle, and rerankers using frontier LVLMs do not close this gap -- leaving ample room for improvement by future methods. Code will be made publicly available.

Paperid: 5151, https://arxiv.org/pdf/2508.15256.pdf

Abstract:
Anomaly detection in computational pathology aims to identify rare and scarce anomalies where disease-related data are often limited or missing. Existing anomaly detection methods, primarily designed for industrial settings, face limitations in pathology due to computational constraints, diverse tissue structures, and lack of interpretability. To address these challenges, we propose Ano-NAViLa, a Normal and Abnormal pathology knowledge-augmented Vision-Language model for Anomaly detection in pathology images. Ano-NAViLa is built on a pre-trained vision-language model with a lightweight trainable MLP. By incorporating both normal and abnormal pathology knowledge, Ano-NAViLa enhances accuracy and robustness to variability in pathology images and provides interpretability through image-text associations. Evaluated on two lymph node datasets from different organs, Ano-NAViLa achieves the state-of-the-art performance in anomaly detection and localization, outperforming competing models.

Paperid: 5152, https://arxiv.org/pdf/2508.13530.pdf

Abstract:
Developing general-purpose embodied agents is a core challenge in AI. Minecraft provides rich complexity and internet-scale data, but its slow speed and engineering overhead make it unsuitable for rapid prototyping. Crafter offers a lightweight alternative that retains key challenges from Minecraft, yet its use has remained limited to narrow tasks due to the absence of foundation models that have driven progress in the Minecraft setting. In this paper, we present CrafterDojo, a suite of foundation models and tools that unlock the Crafter environment as a lightweight, prototyping-friendly, and Minecraft-like testbed for general-purpose embodied agent research. CrafterDojo addresses this by introducing CrafterVPT, CrafterCLIP, and CrafterSteve-1 for behavior priors, vision-language grounding, and instruction following, respectively. In addition, we provide toolkits for generating behavior and caption datasets (CrafterPlay and CrafterCaption), reference agent implementations, benchmark evaluations, and a complete open-source codebase.

Paperid: 5153, https://arxiv.org/pdf/2508.10956.pdf

Abstract:
While vision-language models (VLMs) have made remarkable progress on many popular visual question answering (VQA) benchmarks, it remains unclear whether they abstract and reason over depicted objects. Inspired by human object categorisation, object property reasoning involves identifying and recognising low-level details and higher-level abstractions. While current VQA benchmarks consider a limited set of object property attributes like size, they typically blend perception and reasoning, and lack representativeness in terms of reasoning and image categories. To this end, we introduce a systematic evaluation framework with images of three representative types, three reasoning levels of increasing complexity, and four object property dimensions driven by prior work on commonsense reasoning. We develop a procedure to instantiate this benchmark into ORBIT, a multi-level reasoning VQA benchmark for object properties comprising 360 images paired with a total of 1,080 count-based questions. Experiments with 12 state-of-the-art VLMs in zero-shot settings reveal significant limitations compared to humans, with the best-performing model only reaching 40\% accuracy. VLMs struggle particularly with realistic (photographic) images, counterfactual reasoning about physical and functional properties, and higher counts. ORBIT points to the need to develop methods for scalable benchmarking, generalize annotation guidelines, and explore additional reasoning VLMs. We make the ORBIT benchmark and the experimental code available to support such endeavors.

Paperid: 5154, https://arxiv.org/pdf/2508.10572.pdf

Abstract:
Referring-based Video Object Segmentation is a multimodal problem that requires producing fine-grained segmentation results guided by external cues. Traditional approaches to this task typically involve training specialized models, which come with high computational complexity and manual annotation effort. Recent advances in vision-language foundation models open a promising direction toward training-free approaches. Several studies have explored leveraging these general-purpose models for fine-grained segmentation, achieving performance comparable to that of fully supervised, task-specific models. However, existing methods rely on fixed pipelines that lack the flexibility needed to adapt to the dynamic nature of the task. To address this limitation, we propose Multi-Modal Agent, a novel agentic system designed to solve this task in a more flexible and adaptive manner. Specifically, our method leverages the reasoning capabilities of large language models (LLMs) to generate dynamic workflows tailored to each input. This adaptive procedure iteratively interacts with a set of specialized tools designed for low-level tasks across different modalities to identify the target object described by the multimodal cues. Our agentic approach demonstrates clear improvements over prior methods on two multimodal-conditioned VOS tasks: RVOS and Ref-AVS.

Paperid: 5155, https://arxiv.org/pdf/2508.08521.pdf

Abstract:
Vision Language Models (VLMs) are increasingly being used in a broad range of applications, bringing their security and behavioral control to the forefront. While existing approaches for behavioral control or output redirection, like system prompting in VLMs, are easily detectable and often ineffective, activation-based steering vectors require invasive runtime access to model internals--incompatible with API-based services and closed-source deployments. We introduce VISOR (Visual Input-based Steering for Output Redirection), a novel method that achieves sophisticated behavioral control through optimized visual inputs alone. By crafting universal steering images that induce target activation patterns, VISOR enables practical deployment across all VLM serving modalities while remaining imperceptible compared to explicit textual instructions. We validate VISOR on LLaVA-1.5-7B across three critical alignment tasks: refusal, sycophancy and survival instinct. A single 150KB steering image matches steering vector performance within 1-2% for positive behavioral shifts while dramatically exceeding it for negative steering--achieving up to 25% shifts from baseline compared to steering vectors' modest changes. Unlike system prompting (3-4% shifts), VISOR provides robust bidirectional control while maintaining 99.9% performance on 14,000 unrelated MMLU tasks. Beyond eliminating runtime overhead and model access requirements, VISOR exposes a critical security vulnerability: adversaries can achieve sophisticated behavioral manipulation through visual channels alone, bypassing text-based defenses. Our work fundamentally re-imagines multimodal model control and highlights the urgent need for defenses against visual steering attacks.

Paperid: 5156, https://arxiv.org/pdf/2508.07918.pdf

Abstract:
Visual Question Answering (VQA) in remote sensing (RS) is pivotal for interpreting Earth observation data. However, existing RS VQA datasets are constrained by limitations in annotation richness, question diversity, and the assessment of specific reasoning capabilities. This paper introduces RSVLM-QA dataset, a new large-scale, content-rich VQA dataset for the RS domain. RSVLM-QA is constructed by integrating data from several prominent RS segmentation and detection datasets: WHU, LoveDA, INRIA, and iSAID. We employ an innovative dual-track annotation generation pipeline. Firstly, we leverage Large Language Models (LLMs), specifically GPT-4.1, with meticulously designed prompts to automatically generate a suite of detailed annotations including image captions, spatial relations, and semantic tags, alongside complex caption-based VQA pairs. Secondly, to address the challenging task of object counting in RS imagery, we have developed a specialized automated process that extracts object counts directly from the original segmentation data; GPT-4.1 then formulates natural language answers from these counts, which are paired with preset question templates to create counting QA pairs. RSVLM-QA comprises 13,820 images and 162,373 VQA pairs, featuring extensive annotations and diverse question types. We provide a detailed statistical analysis of the dataset and a comparison with existing RS VQA benchmarks, highlighting the superior depth and breadth of RSVLM-QA's annotations. Furthermore, we conduct benchmark experiments on Six mainstream Vision Language Models (VLMs), demonstrating that RSVLM-QA effectively evaluates and challenges the understanding and reasoning abilities of current VLMs in the RS domain. We believe RSVLM-QA will serve as a pivotal resource for the RS VQA and VLM research communities, poised to catalyze advancements in the field.

Paperid: 5157, https://arxiv.org/pdf/2508.07330.pdf

Abstract:
Vision-language alignment in video must address the complexity of language, evolving interacting entities, their action chains, and semantic gaps between language and vision. This work introduces Planner-Refiner, a framework to overcome these challenges. Planner-Refiner bridges the semantic gap by iteratively refining visual elements' space-time representation, guided by language until semantic gaps are minimal. A Planner module schedules language guidance by decomposing complex linguistic prompts into short sentence chains. The Refiner processes each short sentence, a noun-phrase and verb-phrase pair, to direct visual tokens' self-attention across space then time, achieving efficient single-step refinement. A recurrent system chains these steps, maintaining refined visual token representations. The final representation feeds into task-specific heads for alignment generation. We demonstrate Planner-Refiner's effectiveness on two video-language alignment tasks: Referring Video Object Segmentation and Temporal Grounding with varying language complexity. We further introduce a new MeViS-X benchmark to assess models' capability with long queries. Superior performance versus state-of-the-art methods on these benchmarks shows the approach's potential, especially for complex prompts.

Paperid: 5158, https://arxiv.org/pdf/2508.04120.pdf

Abstract:
Vehicles, as one of the most common and significant objects in the real world, the researches on which using computer vision technologies have made remarkable progress, such as vehicle detection, vehicle re-identification, etc. To search an interested vehicle from the surveillance videos, existing methods first pre-detect and store all vehicle patches, and then apply vehicle re-identification models, which is resource-intensive and not very practical. In this work, we aim to achieve the joint detection and re-identification for vehicle search. However, the conflicting objectives between detection that focuses on shared vehicle commonness and re-identification that focuses on individual vehicle uniqueness make it challenging for a model to learn in an end-to-end system. For this problem, we propose a new unified framework, namely CLIPVehicle, which contains a dual-granularity semantic-region alignment module to leverage the VLMs (Vision-Language Models) for vehicle discrimination modeling, and a multi-level vehicle identification learning strategy to learn the identity representation from global, instance and feature levels. We also construct a new benchmark, including a real-world dataset CityFlowVS, and two synthetic datasets SynVS-Day and SynVS-All, for vehicle search. Extensive experimental results demonstrate that our method outperforms the state-of-the-art methods of both vehicle Re-ID and person search tasks.

Paperid: 5159, https://arxiv.org/pdf/2508.00447.pdf

Abstract:
Understanding the temporal dynamics of biological growth is critical across diverse fields such as microbiology, agriculture, and biodegradation research. Although vision-language models like Contrastive Language Image Pretraining (CLIP) have shown strong capabilities in joint visual-textual reasoning, their effectiveness in capturing temporal progression remains limited. To address this, we propose CLIPTime, a multimodal, multitask framework designed to predict both the developmental stage and the corresponding timestamp of fungal growth from image and text inputs. Built upon the CLIP architecture, our model learns joint visual-textual embeddings and enables time-aware inference without requiring explicit temporal input during testing. To facilitate training and evaluation, we introduce a synthetic fungal growth dataset annotated with aligned timestamps and categorical stage labels. CLIPTime jointly performs classification and regression, predicting discrete growth stages alongside continuous timestamps. We also propose custom evaluation metrics, including temporal accuracy and regression error, to assess the precision of time-aware predictions. Experimental results demonstrate that CLIPTime effectively models biological progression and produces interpretable, temporally grounded outputs, highlighting the potential of vision-language models in real-world biological monitoring applications.

Paperid: 5160, https://arxiv.org/pdf/2507.23391.pdf

Abstract:
Offline reinforcement learning (RL) provides a powerful framework for training robotic agents using pre-collected, suboptimal datasets, eliminating the need for costly, time-consuming, and potentially hazardous online interactions. This is particularly useful in safety-critical real-world applications, where online data collection is expensive and impractical. However, existing offline RL algorithms typically require reward labeled data, which introduces an additional bottleneck: reward function design is itself costly, labor-intensive, and requires significant domain expertise. In this paper, we introduce PLARE, a novel approach that leverages large vision-language models (VLMs) to provide guidance signals for agent training. Instead of relying on manually designed reward functions, PLARE queries a VLM for preference labels on pairs of visual trajectory segments based on a language task description. The policy is then trained directly from these preference labels using a supervised contrastive preference learning objective, bypassing the need to learn explicit reward models. Through extensive experiments on robotic manipulation tasks from the MetaWorld, PLARE achieves performance on par with or surpassing existing state-of-the-art VLM-based reward generation methods. Furthermore, we demonstrate the effectiveness of PLARE in real-world manipulation tasks with a physical robot, further validating its practical applicability.

Paperid: 5161, https://arxiv.org/pdf/2507.23331.pdf

Abstract:
Traffic sign recognition, as a core component of autonomous driving perception systems, directly influences vehicle environmental awareness and driving safety. Current technologies face two significant challenges: first, the traffic sign dataset exhibits a pronounced long-tail distribution, resulting in a substantial decline in recognition performance of traditional convolutional networks when processing low-frequency and out-of-distribution classes; second, traffic signs in real-world scenarios are predominantly small targets with significant scale variations, making it difficult to extract multi-scale features.To overcome these issues, we propose a novel two-stage framework combining open-vocabulary detection and cross-modal learning. For traffic sign detection, our NanoVerse YOLO model integrates a reparameterizable vision-language path aggregation network (RepVL-PAN) and an SPD-Conv module to specifically enhance feature extraction for small, multi-scale targets. For traffic sign classification, we designed a Traffic Sign Recognition Multimodal Contrastive Learning model (TSR-MCL). By contrasting visual features from a Vision Transformer with semantic features from a rule-based BERT, TSR-MCL learns robust, frequency-independent representations, effectively mitigating class confusion caused by data imbalance. On the TT100K dataset, our method achieves a state-of-the-art 78.4% mAP in the long-tail detection task for all-class recognition. The model also obtains 91.8% accuracy and 88.9% recall, significantly outperforming mainstream algorithms and demonstrating superior accuracy and generalization in complex, open-world scenarios.

Paperid: 5162, https://arxiv.org/pdf/2507.18915.pdf

Abstract:
Understanding another person's creative output requires a shared language of association. However, when training vision-language models such as CLIP, we rely on web-scraped datasets containing short, predominantly literal, alt-text. In this work, we introduce a method for mining contextualized associations for salient visual elements in an image that can scale to any unlabeled dataset. Given an image, we can use these mined associations to generate high quality creative captions at increasing degrees of abstraction. With our method, we produce a new dataset of visual associations and 1.7m creative captions for the images in MSCOCO. Human evaluation confirms that these captions remain visually grounded while exhibiting recognizably increasing abstraction. Moreover, fine-tuning a visual encoder on this dataset yields meaningful improvements in zero-shot image-text retrieval in two creative domains: poetry and metaphor visualization. We release our dataset, our generation code and our models for use by the broader community.

Paperid: 5163, https://arxiv.org/pdf/2507.05601.pdf

Abstract:
Graphic design is crucial for conveying ideas and messages. Designers usually organize their work into objects, backgrounds, and vectorized text layers to simplify editing. However, this workflow demands considerable expertise. With the rise of GenAI methods, an endless supply of high-quality graphic designs in pixel format has become more accessible, though these designs often lack editability. Despite this, non-layered designs still inspire human designers, influencing their choices in layouts and text styles, ultimately guiding the creation of layered designs. Motivated by this observation, we propose Accordion, a graphic design generation framework taking the first attempt to convert AI-generated designs into editable layered designs, meanwhile refining nonsensical AI-generated text with meaningful alternatives guided by user prompts. It is built around a vision language model (VLM) playing distinct roles in three curated stages. For each stage, we design prompts to guide the VLM in executing different tasks. Distinct from existing bottom-up methods (e.g., COLE and Open-COLE) that gradually generate elements to create layered designs, our approach works in a top-down manner by using the visually harmonious reference image as global guidance to decompose each layer. Additionally, it leverages multiple vision experts such as SAM and element removal models to facilitate the creation of graphic layers. We train our method using the in-house graphic design dataset Design39K, augmented with AI-generated design images coupled with refined ground truth created by a customized inpainting model. Experimental results and user studies by designers show that Accordion generates favorable results on the DesignIntention benchmark, including tasks such as text-to-template, adding text to background, and text de-rendering, and also excels in creating design variations.

Paperid: 5164, https://arxiv.org/pdf/2507.05515.pdf

Abstract:
Vision-language models (VLMs) are facing the challenges of understanding and following multimodal assembly instructions, particularly when fine-grained spatial reasoning and precise object state detection are required. In this work, we explore LEGO Co-builder, a hybrid benchmark combining real-world LEGO assembly logic with programmatically generated multimodal scenes. The dataset captures stepwise visual states and procedural instructions, allowing controlled evaluation of instruction-following, object detection, and state detection. We introduce a unified framework and assess leading VLMs such as GPT-4o, Gemini, and Qwen-VL, under zero-shot and fine-tuned settings. Our results reveal that even advanced models like GPT-4o struggle with fine-grained assembly tasks, with a maximum F1 score of just 40.54\% on state detection, highlighting gaps in fine-grained visual understanding. We release the benchmark, codebase, and generation pipeline to support future research on multimodal assembly assistants grounded in real-world workflows.

Paperid: 5165, https://arxiv.org/pdf/2507.05513.pdf

Abstract:
Motivated by the growing demand for retrieval systems that operate across modalities, we introduce llama-nemoretriever-colembed, a unified text-image retrieval model that delivers state-of-the-art performance across multiple benchmarks. We release two model variants, 1B and 3B. The 3B model achieves state of the art performance, scoring NDCG@5 91.0 on ViDoRe V1 and 63.5 on ViDoRe V2, placing first on both leaderboards as of June 27, 2025. Our approach leverages the NVIDIA Eagle2 Vision-Language model (VLM), modifies its architecture by replacing causal attention with bidirectional attention, and integrates a ColBERT-style late interaction mechanism to enable fine-grained multimodal retrieval in a shared embedding space. While this mechanism delivers superior retrieval accuracy, it introduces trade-offs in storage and efficiency. We provide a comprehensive analysis of these trade-offs. Additionally, we adopt a two-stage training strategy to enhance the model's retrieval capabilities.

Paperid: 5166, https://arxiv.org/pdf/2507.04735.pdf

Abstract:
Effective cross-modal retrieval is essential for applications like information retrieval and recommendation systems, particularly in specialized domains such as manufacturing, where product information often consists of visual samples paired with a textual description. This paper investigates the use of Vision Language Models(VLMs) for zero-shot text-to-image retrieval on fabric samples. We address the lack of publicly available datasets by introducing an automated annotation pipeline that uses Multimodal Large Language Models (MLLMs) to generate two types of textual descriptions: freeform natural language and structured attribute-based descriptions. We produce these descriptions to evaluate retrieval performance across three Vision-Language Models: CLIP, LAION-CLIP, and Meta's Perception Encoder. Our experiments demonstrate that structured, attribute-rich descriptions significantly enhance retrieval accuracy, particularly for visually complex fabric classes, with the Perception Encoder outperforming other models due to its robust feature alignment capabilities. However, zero-shot retrieval remains challenging in this fine-grained domain, underscoring the need for domain-adapted approaches. Our findings highlight the importance of combining technical textual descriptions with advanced VLMs to optimize cross-modal retrieval in industrial applications.

Paperid: 5167, https://arxiv.org/pdf/2507.04524.pdf

Abstract:
Diffusion policy has demonstrated promising performance in the field of robotic manipulation. However, its effectiveness has been primarily limited in short-horizon tasks, and its performance significantly degrades in the presence of image noise. To address these limitations, we propose a VLM-guided trajectory-conditioned diffusion policy (VLM-TDP) for robust and long-horizon manipulation. Specifically, the proposed method leverages state-of-the-art vision-language models (VLMs) to decompose long-horizon tasks into concise, manageable sub-tasks, while also innovatively generating voxel-based trajectories for each sub-task. The generated trajectories serve as a crucial conditioning factor, effectively steering the diffusion policy and substantially enhancing its performance. The proposed Trajectory-conditioned Diffusion Policy (TDP) is trained on trajectories derived from demonstration data and validated using the trajectories generated by the VLM. Simulation experimental results indicate that our method significantly outperforms classical diffusion policies, achieving an average 44% increase in success rate, over 100% improvement in long-horizon tasks, and a 20% reduction in performance degradation in challenging conditions, such as noisy images or altered environments. These findings are further reinforced by our real-world experiments, where the performance gap becomes even more pronounced in long-horizon tasks. Videos are available on https://youtu.be/g0T6h32OSC8

Paperid: 5168, https://arxiv.org/pdf/2507.04458.pdf

Abstract:
Multimodal sarcasm detection has attracted growing interest due to the rise of multimedia posts on social media. Understanding sarcastic image-text posts often requires external contextual knowledge, such as cultural references or commonsense reasoning. However, existing models struggle to capture the deeper rationale behind sarcasm, relying mainly on shallow cues like image captions or object-attribute pairs from images. To address this, we propose \textbf{MiDRE} (\textbf{Mi}xture of \textbf{D}ual \textbf{R}easoning \textbf{E}xperts), which integrates an internal reasoning expert for detecting incongruities within the image-text pair and an external reasoning expert that utilizes structured rationales generated via Chain-of-Thought prompting to a Large Vision-Language Model. An adaptive gating mechanism dynamically weighs the two experts, selecting the most relevant reasoning path. Experiments on two benchmark datasets show that MiDRE achieves superior performance over baselines. Various qualitative analyses highlight the crucial role of external rationales, revealing that even when they are occasionally noisy, they provide valuable cues that guide the model toward a better understanding of sarcasm.

Paperid: 5169, https://arxiv.org/pdf/2507.01351.pdf

Abstract:
The mixture-of-experts (MoE), which replaces dense models with sparse architectures, has gained attention in large vision-language models (LVLMs) for achieving comparable performance with fewer activated parameters. Existing MoE frameworks for LVLMs focus on token-to-expert routing (TER), encouraging different experts to specialize in processing distinct tokens. However, these frameworks often rely on the load balancing mechanism, overlooking the inherent distributional differences between vision and language. To this end, we propose a Long-Tailed Distribution-aware Router (LTDR) for vision-language TER, tackling two challenges: (1) Distribution-aware router for modality-specific routing. We observe that language TER follows a uniform distribution, whereas vision TER exhibits a long-tailed distribution. This discrepancy necessitates distinct routing strategies tailored to each modality. (2) Enhancing expert activation for vision tail tokens. Recognizing the importance of vision tail tokens, we introduce an oversampling-like strategy by increasing the number of activated experts for these tokens. Experiments on extensive benchmarks validate the effectiveness of our approach.

Paperid: 5170, https://arxiv.org/pdf/2507.01255.pdf

Abstract:
The rapid advancement of AI-generated video models has created a pressing need for robust and interpretable evaluation frameworks. Existing metrics are limited to producing numerical scores without explanatory comments, resulting in low interpretability and human evaluation alignment. To address those challenges, we introduce AIGVE-MACS, a unified model for AI-Generated Video Evaluation(AIGVE), which can provide not only numerical scores but also multi-aspect language comment feedback in evaluating these generated videos. Central to our approach is AIGVE-BENCH 2, a large-scale benchmark comprising 2,500 AI-generated videos and 22,500 human-annotated detailed comments and numerical scores across nine critical evaluation aspects. Leveraging AIGVE-BENCH 2, AIGVE-MACS incorporates recent Vision-Language Models with a novel token-wise weighted loss and a dynamic frame sampling strategy to better align with human evaluators. Comprehensive experiments across supervised and zero-shot benchmarks demonstrate that AIGVE-MACS achieves state-of-the-art performance in both scoring correlation and comment quality, significantly outperforming prior baselines including GPT-4o and VideoScore. In addition, we further showcase a multi-agent refinement framework where feedback from AIGVE-MACS drives iterative improvements in video generation, leading to 53.5% quality enhancement. This work establishes a new paradigm for comprehensive, human-aligned evaluation of AI-generated videos. We release the AIGVE-BENCH 2 and AIGVE-MACS at https://huggingface.co/xiaoliux/AIGVE-MACS.

Paperid: 5171, https://arxiv.org/pdf/2506.21656.pdf

Abstract:
Current Vision-Language Models (VLMs) struggle with fine-grained spatial reasoning, particularly when multi-step logic and precise spatial alignment are required. In this work, we introduce SpatialReasoner-R1, a vision-language reasoning model designed to address these limitations. To construct high-quality supervision for spatial reasoning, we design a Multi-Model Monte Carlo Tree Search (M3CTS) method that generates diverse, logically consistent Long Chain-of-Thought (LongCoT) reasoning trajectories. In addition, we propose fine-grained Direct Preference Optimization (fDPO), which introduces segment-specific preference granularity for descriptive grounding and logical reasoning, guided by a spatial reward mechanism that evaluates candidate responses based on visual consistency, spatial grounding, and logical coherence. Experimental results demonstrate that fDPO achieves an average improvement of 4.1% over standard DPO across spatial quality tasks, and a 9.0% gain in spatial quantity tasks. SpatialReasoner-R1, trained with fDPO, sets a new SoTA on SPATIALRGPT-Bench, outperforming the strongest baseline by 9.8% in average accuracy, while maintaining competitive performance on general vision-language tasks.

Paperid: 5172, https://arxiv.org/pdf/2506.21144.pdf

Abstract:
Federated learning (FL) enables collaborative model training across decentralized clients without sharing local data, but is challenged by heterogeneity in data, computation, and communication. Pretrained vision-language models (VLMs), with their strong generalization and lightweight tuning via prompts, offer a promising solution. However, existing federated prompt-learning methods rely only on text prompts and overlook joint label-domain distribution shifts. In this paper, we propose a personalized FL framework based on dual-prompt learning and cross fusion, termed pFedDC. Specifically, each client maintains both global and local prompts across vision and language modalities: global prompts capture common knowledge shared across the federation, while local prompts encode client-specific semantics and domain characteristics. Meanwhile, a cross-fusion module is designed to adaptively integrate prompts from different levels, enabling the model to generate personalized representations aligned with each client's unique data distribution. Extensive experiments across nine datasets with various types of heterogeneity show that pFedDC consistently outperforms state-of-the-art methods.

Paperid: 5173, https://arxiv.org/pdf/2506.21017.pdf

Abstract:
Prompt learning has been widely adopted to efficiently adapt vision-language models (VLMs) like CLIP for various downstream tasks. Despite their success, current VLM-based facial expression recognition (FER) methods struggle to capture fine-grained textual-visual relationships, which are essential for distinguishing subtle differences between facial expressions. To address this challenge, we propose a multimodal prompt alignment framework for FER, called MPA-FER, that provides fine-grained semantic guidance to the learning process of prompted visual features, resulting in more precise and interpretable representations. Specifically, we introduce a multi-granularity hard prompt generation strategy that utilizes a large language model (LLM) like ChatGPT to generate detailed descriptions for each facial expression. The LLM-based external knowledge is injected into the soft prompts by minimizing the feature discrepancy between the soft prompts and the hard prompts. To preserve the generalization abilities of the pretrained CLIP model, our approach incorporates prototype-guided visual feature alignment, ensuring that the prompted visual features from the frozen image encoder align closely with class-specific prototypes. Additionally, we propose a cross-modal global-local alignment module that focuses on expression-relevant facial features, further improving the alignment between textual and visual features. Extensive experiments demonstrate our framework outperforms state-of-the-art methods on three FER benchmark datasets, while retaining the benefits of the pretrained model and minimizing computational costs.

Paperid: 5174, https://arxiv.org/pdf/2506.20991.pdf

Abstract:
The rapid advancement of 3D vision-language models (VLMs) has spurred significant interest in interactive point cloud processing tasks, particularly for real-world applications. However, existing methods often underperform in point-level tasks, such as segmentation, due to missing direct 3D-text alignment, limiting their ability to link local 3D features with textual context. To solve this problem, we propose TSDASeg, a Two-Stage model coupled with a Direct cross-modal Alignment module and memory module for interactive point cloud Segmentation. We introduce the direct cross-modal alignment module to establish explicit alignment between 3D point clouds and textual/2D image data. Within the memory module, we employ multiple dedicated memory banks to separately store text features, visual features, and their cross-modal correspondence mappings. These memory banks are dynamically leveraged through self-attention and cross-attention mechanisms to update scene-specific features based on prior stored data, effectively addressing inconsistencies in interactive segmentation results across diverse scenarios. Experiments conducted on multiple 3D instruction, reference, and semantic segmentation datasets demonstrate that the proposed method achieves state-of-the-art performance.

Paperid: 5175, https://arxiv.org/pdf/2506.19825.pdf

Abstract:
Diagrams are widely used to visualize data in publications. The research field of data visualization deals with defining principles and guidelines for the creation and use of these diagrams, which are often not known or adhered to by researchers, leading to misinformation caused by providing inaccurate or incomplete information. In this work, large Vision Language Models (VLMs) are used to analyze diagrams in order to identify potential problems in regards to selected data visualization principles and guidelines. To determine the suitability of VLMs for these tasks, five open source VLMs and five prompting strategies are compared using a set of questions derived from selected data visualization guidelines. The results show that the employed VLMs work well to accurately analyze diagram types (F1-score 82.49 %), 3D effects (F1-score 98.55 %), axes labels (F1-score 76.74 %), lines (RMSE 1.16), colors (RMSE 1.60) and legends (F1-score 96.64 %, RMSE 0.70), while they cannot reliably provide feedback about the image quality (F1-score 0.74 %) and tick marks/labels (F1-score 46.13 %). Among the employed VLMs, Qwen2.5VL performs best, and the summarizing prompting strategy performs best for most of the experimental questions. It is shown that VLMs can be used to automatically identify a number of potential issues in diagrams, such as missing axes labels, missing legends, and unnecessary 3D effects. The approach laid out in this work can be extended for further aspects of data visualization.

Paperid: 5176, https://arxiv.org/pdf/2506.19579.pdf

Abstract:
Vision-Language Models (VLMs) have emerged as powerful tools for generating textual descriptions from visual data. While these models excel on web-scale datasets, their robustness to the domain shifts inherent in many real-world applications remains under-explored. This paper presents a systematic evaluation of VLM performance on a single-view object captioning task when faced with a controlled, physical domain shift. We compare captioning accuracy across two distinct object sets: a collection of multi-material, real-world tools and a set of single-material, 3D-printed items. The 3D-printed set introduces a significant domain shift in texture and material properties, challenging the models' generalization capabilities. Our quantitative results demonstrate that all tested VLMs show a marked performance degradation when describing the 3D-printed objects compared to the real-world tools. This underscores a critical limitation in the ability of current models to generalize beyond surface-level features and highlights the need for more robust architectures for real-world signal processing applications.

Paperid: 5177, https://arxiv.org/pdf/2506.18234.pdf

Abstract:
Large vision-language models (VLMs) for autonomous driving (AD) are evolving beyond perception and cognition tasks toward motion planning. However, we identify two critical challenges in this direction: (1) VLMs tend to learn shortcuts by relying heavily on history input information, achieving seemingly strong planning results without genuinely understanding the visual inputs; and (2) the chain-ofthought (COT) reasoning processes are always misaligned with the motion planning outcomes, and how to effectively leverage the complex reasoning capability to enhance planning remains largely underexplored. In this paper, we start from a small-scale domain-specific VLM and propose Drive-R1 designed to bridges the scenario reasoning and motion planning for AD. Drive-R1 first undergoes the supervised finetuning on a elaborate dataset containing both long and short COT data. Drive-R1 is encouraged to reason step-by-step from visual input to final planning decisions. Subsequently, Drive-R1 is trained within a reinforcement learning framework that incentivizes the discovery of reasoning paths that are more informative for planning, guided by rewards based on predicted trajectories and meta actions. Experimental evaluations on the nuScenes and DriveLM-nuScenes benchmarks demonstrate that Drive-R1 achieves superior performance compared to existing state-of-the-art VLMs. We believe that Drive-R1 presents a promising direction for bridging reasoning and planning in AD, offering methodological insights for future research and applications.

Paperid: 5178, https://arxiv.org/pdf/2506.17144.pdf

Abstract:
Traditional video-based tasks like soccer action spotting rely heavily on visual inputs, often requiring complex and computationally expensive models to process dense video data. In this work, we propose a shift from this video-centric approach to a text-based task, making it lightweight and scalable by utilizing Large Language Models (LLMs) instead of Vision-Language Models (VLMs). We posit that expert commentary, which provides rich, fine-grained descriptions and contextual cues such as excitement and tactical insights, contains enough information to reliably spot key actions in a match. To demonstrate this, we use the SoccerNet Echoes dataset, which provides timestamped commentary, and employ a system of three LLMs acting as judges specializing in outcome, excitement, and tactics. Each LLM evaluates sliding windows of commentary to identify actions like goals, cards, and substitutions, generating accurate timestamps for these events. Our experiments show that this language-centric approach performs effectively in detecting critical match events, providing a lightweight and training-free alternative to traditional video-based methods for action spotting.

Paperid: 5179, https://arxiv.org/pdf/2506.15594.pdf

Abstract:
Documents are fundamental to preserving and disseminating information, often incorporating complex layouts, tables, and charts that pose significant challenges for automatic document understanding (DU). While vision-language large models (VLLMs) have demonstrated improvements across various tasks, their effectiveness in processing long-context vision inputs remains unclear. This paper introduces WikiMixQA, a benchmark comprising 1,000 multiple-choice questions (MCQs) designed to evaluate cross-modal reasoning over tables and charts extracted from 4,000 Wikipedia pages spanning seven distinct topics. Unlike existing benchmarks, WikiMixQA emphasizes complex reasoning by requiring models to synthesize information from multiple modalities. We evaluate 12 state-of-the-art vision-language models, revealing that while proprietary models achieve ~70% accuracy when provided with direct context, their performance deteriorates significantly when retrieval from long documents is required. Among these, GPT-4-o is the only model exceeding 50% accuracy in this setting, whereas open-source models perform considerably worse, with a maximum accuracy of 27%. These findings underscore the challenges of long-context, multi-modal reasoning and establish WikiMixQA as a crucial benchmark for advancing document understanding research.

Paperid: 5180, https://arxiv.org/pdf/2506.11472.pdf

Abstract:
Autonomous vehicles (AVs) rely on deep neural networks (DNNs) for critical tasks such as traffic sign recognition (TSR), automated lane centering (ALC), and vehicle detection (VD). However, these models are vulnerable to attacks that can cause misclassifications and compromise safety. Traditional defense mechanisms, including adversarial training, often degrade benign accuracy and fail to generalize against unseen attacks. In this work, we introduce Vehicle Vision Language Models (V2LMs), fine-tuned vision-language models specialized for AV perception. Our findings demonstrate that V2LMs inherently exhibit superior robustness against unseen attacks without requiring adversarial training, maintaining significantly higher accuracy than conventional DNNs under adversarial conditions. We evaluate two deployment strategies: Solo Mode, where individual V2LMs handle specific perception tasks, and Tandem Mode, where a single unified V2LM is fine-tuned for multiple tasks simultaneously. Experimental results reveal that DNNs suffer performance drops of 33% to 46% under attacks, whereas V2LMs maintain adversarial accuracy with reductions of less than 8% on average. The Tandem Mode further offers a memory-efficient alternative while achieving comparable robustness to Solo Mode. We also explore integrating V2LMs as parallel components to AV perception to enhance resilience against adversarial threats. Our results suggest that V2LMs offer a promising path toward more secure and resilient AV perception systems.

Paperid: 5181, https://arxiv.org/pdf/2506.10895.pdf

Abstract:
Zero-shot generative model adaptation (ZSGM) aims to adapt a pre-trained generator to a target domain using only text guidance and without any samples from the target domain. Central to recent ZSGM approaches are directional loss which use the text guidance in the form of aligning the image offset with text offset in the embedding space of a vision-language model like CLIP. This is similar to the analogical reasoning in NLP where the offset between one pair of words is used to identify a missing element in another pair by aligning the offset between these two pairs. However, a major limitation of existing ZSGM methods is that the learning objective assumes the complete alignment between image offset and text offset in the CLIP embedding space, resulting in quality degrade in generated images. Our work makes two main contributions. Inspired by the offset misalignment studies in NLP, as our first contribution, we perform an empirical study to analyze the misalignment between text offset and image offset in CLIP embedding space for various large publicly available datasets. Our important finding is that offset misalignment in CLIP embedding space is correlated with concept distance, i.e., close concepts have a less offset misalignment. To address the limitations of the current approaches, as our second contribution, we propose Adaptation with Iterative Refinement (AIR) which is the first ZSGM approach to focus on improving target domain image quality based on our new insight on offset misalignment.Qualitative, quantitative, and user study in 26 experiment setups consistently demonstrate the proposed AIR approach achieves SOTA performance. Additional experiments are in Supp.

Paperid: 5182, https://arxiv.org/pdf/2506.07418.pdf

Abstract:
Multimodal Large Language Models (MLLMs) promise advanced vision language capabilities, yet their effectiveness in visually presented mathematics remains underexplored. This paper analyzes the development and evaluation of MLLMs for mathematical problem solving, focusing on diagrams, multilingual text, and symbolic notation. We then assess several models, including GPT 4o, Pixtral, Qwen VL, Llama 3.2 Vision variants, and Gemini 2.0 Flash in a multilingual Kangaroo style benchmark spanning English, French, Spanish, and Catalan. Our experiments reveal four key findings. First, overall precision remains moderate across geometry, visual algebra, logic, patterns, and combinatorics: no single model excels in every topic. Second, while most models see improved accuracy with questions that do not have images, the gain is often limited; performance for some remains nearly unchanged without visual input, indicating underutilization of diagrammatic information. Third, substantial variation exists across languages and difficulty levels: models frequently handle easier items but struggle with advanced geometry and combinatorial reasoning. Notably, Gemini 2.0 Flash achieves the highest precision on image based tasks, followed by Qwen VL 2.5 72B and GPT 4o, though none approach human level performance. Fourth, a complementary analysis aimed at distinguishing whether models reason or simply recite reveals that Gemini and GPT 4o stand out for their structured reasoning and consistent accuracy. In contrast, Pixtral and Llama exhibit less consistent reasoning, often defaulting to heuristics or randomness when unable to align their outputs with the given answer options.

Paperid: 5183, https://arxiv.org/pdf/2506.07399.pdf

Abstract:
Multimodal retrieval-augmented generation (RAG) systems enhance large vision-language models by integrating cross-modal knowledge, enabling their increasing adoption across real-world multimodal tasks. These knowledge databases may contain sensitive information that requires privacy protection. However, multimodal RAG systems inherently grant external users indirect access to such data, making them potentially vulnerable to privacy attacks, particularly membership inference attacks (MIAs). % Existing MIA methods targeting RAG systems predominantly focus on the textual modality, while the visual modality remains relatively underexplored. To bridge this gap, we propose MrM, the first black-box MIA framework targeted at multimodal RAG systems. It utilizes a multi-object data perturbation framework constrained by counterfactual attacks, which can concurrently induce the RAG systems to retrieve the target data and generate information that leaks the membership information. Our method first employs an object-aware data perturbation method to constrain the perturbation to key semantics and ensure successful retrieval. Building on this, we design a counterfact-informed mask selection strategy to prioritize the most informative masked regions, aiming to eliminate the interference of model self-knowledge and amplify attack efficacy. Finally, we perform statistical membership inference by modeling query trials to extract features that reflect the reconstruction of masked semantics from response patterns. Experiments on two visual datasets and eight mainstream commercial visual-language models (e.g., GPT-4o, Gemini-2) demonstrate that MrM achieves consistently strong performance across both sample-level and set-level evaluations, and remains robust under adaptive defenses.

Paperid: 5184, https://arxiv.org/pdf/2506.02557.pdf

Abstract:
Vision-language models, such as CLIP, have achieved significant success in aligning visual and textual representations, becoming essential components of many multi-modal large language models (MLLMs) like LLaVA and OpenFlamingo. However, numerous studies have identified CLIP's limited fine-grained perception as a critical drawback, leading to substantial failures in downstream MLLMs. In contrast, vision-centric foundation models like DINOv2 demonstrate remarkable capabilities in capturing fine details from images. In this work, we propose a novel kernel-based method to align CLIP's visual representation with that of DINOv2, ensuring that the resulting embeddings maintain compatibility with text embeddings while enhancing perceptual capabilities. Our alignment objective is designed for efficient stochastic optimization. Following this image-only alignment fine-tuning, the visual encoder retains compatibility with the frozen text encoder and exhibits significant improvements in zero-shot object recognition, fine-grained spatial reasoning, and localization. By integrating the aligned visual encoder, downstream MLLMs also demonstrate enhanced performance.

Paperid: 5185, https://arxiv.org/pdf/2506.01955.pdf

Abstract:
Prior methods for controlling image generation are limited in their ability to be taught new tasks. In contrast, vision-language models, or VLMs, can learn tasks in-context and produce the correct outputs for a given input. We propose a dual-process distillation scheme that allows feed-forward image generators to learn new tasks from deliberative VLMs. Our scheme uses a VLM to rate the generated images and backpropagates this gradient to update the weights of the image generator. Our general framework enables a wide variety of new control tasks through the same text-and-image based interface. We showcase a handful of applications of this technique for different types of control signals, such as commonsense inferences and visual prompts. With our method, users can implement multimodal controls for properties such as color palette, line weight, horizon position, and relative depth within a matter of minutes. Project page: https://dual-process.github.io.

Paperid: 5186, https://arxiv.org/pdf/2506.00819.pdf

Abstract:
End-to-end autonomous driving systems map sensor data directly to control commands, but remain opaque, lack interpretability, and offer no formal safety guarantees. While recent vision-language-guided reinforcement learning (RL) methods introduce semantic feedback, they often rely on static prompts and fixed objectives, limiting adaptability to dynamic driving scenes. We present DriveMind, a unified semantic reward framework that integrates: (i) a contrastive Vision-Language Model (VLM) encoder for stepwise semantic anchoring; (ii) a novelty-triggered VLM encoder-decoder, fine-tuned via chain-of-thought (CoT) distillation, for dynamic prompt generation upon semantic drift; (iii) a hierarchical safety module enforcing kinematic constraints (e.g., speed, lane centering, stability); and (iv) a compact predictive world model to reward alignment with anticipated ideal states. DriveMind achieves 19.4 +/- 2.3 km/h average speed, 0.98 +/- 0.03 route completion, and near-zero collisions in CARLA Town 2, outperforming baselines by over 4% in success rate. Its semantic reward generalizes zero-shot to real dash-cam data with minimal distributional shift, demonstrating robust cross-domain alignment and potential for real-world deployment.

Paperid: 5187, https://arxiv.org/pdf/2506.00411.pdf

Abstract:
Real-world embodied agents face long-horizon tasks, characterized by high-level goals demanding multi-step solutions beyond single actions. Successfully navigating these requires both high-level task planning (i.e., decomposing goals into sub-tasks) and low-level motion control (i.e., generating precise robot actions). While existing vision language action (VLA) models and hierarchical architectures offer potential in embodied tasks, the former often falter in planning, and the latter can suffer from coordination issues, both hampering performance. We introduce a new unified VLA framework for long-horizon tasks, dubbed LoHoVLA, to overcome these limitations. LoHoVLA leverages a large pretrained vision language model (VLM) as the backbone to jointly generate language and action tokens for sub-task generation and robot action prediction, respectively. This shared representation promotes better generalization across tasks. Additionally, LoHoVLA embraces a hierarchical closed-loop control mechanism to mitigate errors originating from both high-level planning and low-level control. To train LoHoVLA, we introduce LoHoSet, a dataset built on the Ravens simulator, containing 20 long-horizon tasks, each with 1,000 expert demonstrations composed of visual observations, linguistic goals, sub-tasks, and robot actions. Experimental results show that LoHoVLA significantly surpasses both hierarchical and standard VLA approaches on long-horizon embodied tasks in the Ravens simulator. These findings underscore the promise of unified architectures for advancing generalizable embodied intelligence.

Paperid: 5188, https://arxiv.org/pdf/2505.22793.pdf

Abstract:
Modern vision-language models (VLMs) often fail at cultural competency evaluations and benchmarks. Given the diversity of applications built upon VLMs, there is renewed interest in understanding how they encode cultural nuances. While individual aspects of this problem have been studied, we still lack a comprehensive framework for systematically identifying and annotating the nuanced cultural dimensions present in images for VLMs. This position paper argues that foundational methodologies from visual culture studies (cultural studies, semiotics, and visual studies) are necessary for cultural analysis of images. Building upon this review, we propose a set of five frameworks, corresponding to cultural dimensions, that must be considered for a more complete analysis of the cultural competencies of VLMs.

Paperid: 5189, https://arxiv.org/pdf/2505.22491.pdf

Abstract:
The dominant paradigm for training large-scale vision and language models is He initialization and a single global learning rate (\textit{standard parameterization}, SP). Despite its practical success, standard parametrization remains poorly understood from a theoretical perspective: Existing infinite-width theory would predict instability under large learning rates and vanishing feature learning under stable learning rates. However, empirically optimal learning rates consistently decay much slower than theoretically predicted. By carefully studying neural network training dynamics, we demonstrate that this discrepancy is not fully explained by finite-width phenomena such as catapult effects or a lack of alignment between weights and incoming activations. We instead show that the apparent contradiction can be fundamentally resolved by taking the loss function into account: In contrast to Mean Squared Error (MSE) loss, we prove that under cross-entropy (CE) loss, an intermediate \textit{controlled divergence} regime emerges, where logits diverge but loss, gradients, and activations remain stable. Stable training under large learning rates enables persistent feature evolution at scale in all hidden layers, which is crucial for the practical success of SP. In experiments across optimizers (SGD, Adam), architectures (MLPs, GPT) and data modalities (vision, language), we validate that neural networks operate in this controlled divergence regime under CE loss but not under MSE loss. Our empirical evidence suggests that width-scaling considerations are surprisingly useful for predicting empirically optimal learning rate exponents. Finally, our analysis clarifies the effectiveness and limitations of recently proposed layerwise learning rate scalings for standard initialization.

Paperid: 5190, https://arxiv.org/pdf/2505.22150.pdf

Abstract:
Brain-to-Image reconstruction aims to recover visual stimuli perceived by humans from brain activity. However, the reconstructed visual stimuli often missing details and semantic inconsistencies, which may be attributed to insufficient semantic information. To address this issue, we propose an approach named Fine-grained Brain-to-Image reconstruction (FgB2I), which employs fine-grained text as bridge to improve image reconstruction. FgB2I comprises three key stages: detail enhancement, decoding fine-grained text descriptions, and text-bridged brain-to-image reconstruction. In the detail-enhancement stage, we leverage large vision-language models to generate fine-grained captions for visual stimuli and experimentally validate its importance. We propose three reward metrics (object accuracy, text-image semantic similarity, and image-image semantic similarity) to guide the language model in decoding fine-grained text descriptions from fMRI signals. The fine-grained text descriptions can be integrated into existing reconstruction methods to achieve fine-grained Brain-to-Image reconstruction.

Paperid: 5191, https://arxiv.org/pdf/2505.21459.pdf

Abstract:
Current video analytics approaches face a fundamental trade-off between flexibility and efficiency. End-to-end Vision Language Models (VLMs) often struggle with long-context processing and incur high computational costs, while neural-symbolic methods depend heavily on manual labeling and rigid rule design. In this paper, we introduce LazyVLM, a neuro-symbolic video analytics system that provides a user-friendly query interface similar to VLMs, while addressing their scalability limitation. LazyVLM enables users to effortlessly drop in video data and specify complex multi-frame video queries using a semi-structured text interface for video analytics. To address the scalability limitations of VLMs, LazyVLM decomposes multi-frame video queries into fine-grained operations and offloads the bulk of the processing to efficient relational query execution and vector similarity search. We demonstrate that LazyVLM provides a robust, efficient, and user-friendly solution for querying open-domain video data at scale.

Paperid: 5192, https://arxiv.org/pdf/2505.20381.pdf

Abstract:
Referring Multi-object tracking (RMOT) is an important research field in computer vision. Its task form is to guide the models to track the objects that conform to the language instruction. However, the RMOT task commonly requires clear language instructions, such methods often fail to work when complex language instructions with reasoning characteristics appear. In this work, we propose a new task, called Reasoning-based Multi-Object Tracking (ReaMOT). ReaMOT is a more challenging task that requires accurate reasoning about objects that match the language instruction with reasoning characteristic and tracking the objects' trajectories. To advance the ReaMOT task and evaluate the reasoning capabilities of tracking models, we construct ReaMOT Challenge, a reasoning-based multi-object tracking benchmark built upon 12 datasets. Specifically, it comprises 1,156 language instructions with reasoning characteristic, 423,359 image-language pairs, and 869 diverse scenes, which is divided into three levels of reasoning difficulty. In addition, we propose a set of evaluation metrics tailored for the ReaMOT task. Furthermore, we propose ReaTrack, a training-free framework for reasoning-based multi-object tracking based on large vision-language models (LVLM) and SAM2, as a baseline for the ReaMOT task. Extensive experiments on the ReaMOT Challenge benchmark demonstrate the effectiveness of our ReaTrack framework.

Paperid: 5193, https://arxiv.org/pdf/2505.20122.pdf

Abstract:
This paper introduces MEBench, a novel benchmark for evaluating mutual exclusivity (ME) bias, a cognitive phenomenon observed in children during word learning. Unlike traditional ME tasks, MEBench further incorporates spatial reasoning to create more challenging and realistic evaluation settings. We assess the performance of state-of-the-art vision-language models (VLMs) on this benchmark using novel evaluation metrics that capture key aspects of ME-based reasoning. To facilitate controlled experimentation, we also present a flexible and scalable data generation pipeline that supports the construction of diverse annotated scenes.

Paperid: 5194, https://arxiv.org/pdf/2505.16602.pdf

Abstract:
Egocentric hand-object motion generation is crucial for immersive AR/VR and robotic imitation but remains challenging due to unstable viewpoints, self-occlusions, perspective distortion, and noisy ego-motion. Existing methods rely on predefined 3D object priors, limiting generalization to novel objects, which restricts their generalizability to novel objects. Meanwhile, recent multimodal approaches suffer from ambiguous generation from abstract textual cues, intricate pipelines for modeling 3D hand-object correlation, and compounding errors in open-loop prediction. We propose MEgoHand, a multimodal framework that synthesizes physically plausible hand-object interactions from egocentric RGB, text, and initial hand pose. MEgoHand introduces a bi-level architecture: a high-level "cerebrum" leverages a vision language model (VLM) to infer motion priors from visual-textual context and a monocular depth estimator for object-agnostic spatial reasoning, while a low-level DiT-based flow-matching policy generates fine-grained trajectories with temporal orthogonal filtering to enhance stability. To address dataset inconsistency, we design a dataset curation paradigm with an Inverse MANO Retargeting Network and Virtual RGB-D Renderer, curating a unified dataset of 3.35M RGB-D frames, 24K interactions, and 1.2K objects. Extensive experiments across five in-domain and two cross-domain datasets demonstrate the effectiveness of MEgoHand, achieving substantial reductions in wrist translation error (86.9%) and joint rotation error (34.1%), highlighting its capacity to accurately model fine-grained hand joint structures and generalize robustly across diverse scenarios.

Paperid: 5195, https://arxiv.org/pdf/2505.16149.pdf

Abstract:
Image classification benchmark datasets such as CIFAR, MNIST, and ImageNet serve as critical tools for model evaluation. However, despite the cleaning efforts, these datasets still suffer from pervasive noisy labels and often contain missing labels due to the co-existing image pattern where multiple classes appear in an image sample. This results in misleading model comparisons and unfair evaluations. Existing label cleaning methods focus primarily on noisy labels, but the issue of missing labels remains largely overlooked. Motivated by these challenges, we present a comprehensive framework named REVEAL, integrating state-of-the-art pre-trained vision-language models (e.g., LLaVA, BLIP, Janus, Qwen) with advanced machine/human label curation methods (e.g., Docta, Cleanlab, MTurk), to systematically address both noisy labels and missing label detection in widely-used image classification test sets. REVEAL detects potential noisy labels and omissions, aggregates predictions from various methods, and refines label accuracy through confidence-informed predictions and consensus-based filtering. Additionally, we provide a thorough analysis of state-of-the-art vision-language models and pre-trained image classifiers, highlighting their strengths and limitations within the context of dataset renovation by revealing 10 observations. Our method effectively reveals missing labels from public datasets and provides soft-labeled results with likelihoods. Through human verifications, REVEAL significantly improves the quality of 6 benchmark test sets, highly aligning to human judgments and enabling more accurate and meaningful comparisons in image classification.

Paperid: 5196, https://arxiv.org/pdf/2505.15685.pdf

Abstract:
Foundation models (FMs) are increasingly used to bridge language and action in embodied agents, yet the operational characteristics of different FM integration strategies remain under-explored -- particularly for complex instruction following and versatile action generation in changing environments. This paper examines three paradigms for building robotic systems: end-to-end vision-language-action (VLA) models that implicitly integrate perception and planning, and modular pipelines incorporating either vision-language models (VLMs) or multimodal large language models (LLMs). We evaluate these paradigms through two focused case studies: a complex instruction grounding task assessing fine-grained instruction understanding and cross-modal disambiguation, and an object manipulation task targeting skill transfer via VLA finetuning. Our experiments in zero-shot and few-shot settings reveal trade-offs in generalization and data efficiency. By exploring performance limits, we distill design implications for developing language-driven physical agents and outline emerging challenges and opportunities for FM-powered robotics in real-world conditions.

Paperid: 5197, https://arxiv.org/pdf/2505.13376.pdf

Abstract:
Increased robot deployment, such as in warehousing, has revealed a need for seamless collaboration among heterogeneous robot teams to resolve unforeseen conflicts. To address this challenge, we propose a novel, decentralized framework for robots to request and provide help. The framework begins with robots detecting conflicts using a Vision Language Model (VLM), then reasoning over whether help is needed. If so, it crafts and broadcasts a natural language (NL) help request using a Large Language Model (LLM). Potential helper robots reason over the request and offer help (if able), along with information about impact to their current tasks. Helper reasoning is implemented via an LLM grounded in Signal Temporal Logic (STL) using a Backus-Naur Form (BNF) grammar to guarantee syntactically valid NL-to-STL translations, which are then solved as a Mixed Integer Linear Program (MILP). Finally, the requester robot chooses a helper by reasoning over impact on the overall system. We evaluate our system via experiments considering different strategies for choosing a helper, and find that a requester robot can minimize overall time impact on the system by considering multiple help offers versus simple heuristics (e.g., selecting the nearest robot to help).

Paperid: 5198, https://arxiv.org/pdf/2505.12224.pdf

Abstract:
Vision-Language-Action (VLA) models have recently advanced robotic manipulation by translating natural-language instructions and image information into sequential control actions. However, these models often underperform in open-world scenarios, as they are predominantly trained on successful expert demonstrations and exhibit a limited capacity for failure recovery. In this work, we present a Robotic Failure Analysis and Correction (RoboFAC) framework to address this issue. Firstly, we construct RoboFAC dataset comprising 9,440 erroneous manipulation trajectories and 78,623 QA pairs across 16 diverse tasks and 53 scenes in both simulation and real-world environments. Leveraging our dataset, we develop RoboFAC model, which is capable of Task Understanding, Failure Analysis and Failure Correction. Experimental results demonstrate that the RoboFAC model outperforms GPT-4o by 34.1% on our evaluation benchmark. Furthermore, we integrate the RoboFAC model into a real-world VLA control pipeline as an external supervision providing correction instructions, yielding a 29.1% relative improvement on average on four real-world tasks. The results show that our RoboFAC framework effectively handles robotic failures and assists the VLA model in recovering from failures.

Paperid: 5199, https://arxiv.org/pdf/2505.11221.pdf

Abstract:
Recent research highlights the potential of multimodal foundation models in tackling complex decision-making challenges. However, their large parameters make real-world deployment resource-intensive and often impractical for constrained systems. Reinforcement learning (RL) shows promise for task-specific agents but suffers from high sample complexity, limiting practical applications. To address these challenges, we introduce LVLM to Policy (LVLM2P), a novel framework that distills knowledge from large vision-language models (LVLM) into more efficient RL agents. Our approach leverages the LVLM as a teacher, providing instructional actions based on trajectories collected by the RL agent, which helps reduce less meaningful exploration in the early stages of learning, thereby significantly accelerating the agent's learning progress. Additionally, by leveraging the LVLM to suggest actions directly from visual observations, we eliminate the need for manual textual descriptors of the environment, enhancing applicability across diverse tasks. Experiments show that LVLM2P significantly enhances the sample efficiency of baseline RL algorithms.

Paperid: 5200, https://arxiv.org/pdf/2505.10993.pdf

Abstract:
Content generation modeling has emerged as a promising direction in computational pathology, offering capabilities such as data-efficient learning, synthetic data augmentation, and task-oriented generation across diverse diagnostic tasks. This review provides a comprehensive synthesis of recent progress in the field, organized into four key domains: image generation, text generation, molecular profile-morphology generation, and other specialized generation applications. By analyzing over 150 representative studies, we trace the evolution of content generation architectures -- from early generative adversarial networks to recent advances in diffusion models and generative vision-language models. We further examine the datasets and evaluation protocols commonly used in this domain and highlight ongoing limitations, including challenges in generating high-fidelity whole slide images, clinical interpretability, and concerns related to the ethical and legal implications of synthetic data. The review concludes with a discussion of open challenges and prospective research directions, with an emphasis on developing integrated and clinically deployable generation systems. This work aims to provide a foundational reference for researchers and practitioners developing content generation models in computational pathology.

Paperid: 5201, https://arxiv.org/pdf/2505.10714.pdf

Abstract:
We present GeoGrid-Bench, a benchmark designed to evaluate the ability of foundation models to understand geo-spatial data in the grid structure. Geo-spatial datasets pose distinct challenges due to their dense numerical values, strong spatial and temporal dependencies, and unique multimodal representations including tabular data, heatmaps, and geographic visualizations. To assess how foundation models can support scientific research in this domain, GeoGrid-Bench features large-scale, real-world data covering 16 climate variables across 150 locations and extended time frames. The benchmark includes approximately 3,200 question-answer pairs, systematically generated from 8 domain expert-curated templates to reflect practical tasks encountered by human scientists. These range from basic queries at a single location and time to complex spatiotemporal comparisons across regions and periods. Our evaluation reveals that vision-language models perform best overall, and we provide a fine-grained analysis of the strengths and limitations of different foundation models in different geo-spatial tasks. This benchmark offers clearer insights into how foundation models can be effectively applied to geo-spatial data analysis and used to support scientific research.

Paperid: 5202, https://arxiv.org/pdf/2505.10526.pdf

Abstract:
Speculative decoding significantly accelerates language model inference by enabling a lightweight draft model to propose multiple tokens that a larger target model verifies simultaneously. However, applying this technique to vision-language models (VLMs) presents two fundamental challenges: small language models that could serve as efficient drafters lack the architectural components to process visual inputs, and their token predictions fail to match those of VLM target models that consider visual context. We introduce Multimodal Adaptation and Self-Data Distillation for Speculative Decoding of Vision-Language Models (MASSV), which transforms existing small language models into effective multimodal drafters through a two-phase approach. MASSV first connects the target VLM's vision encoder to the draft model via a lightweight trainable projector, then applies self-distilled visual instruction tuning using responses generated by the target VLM to align token predictions. Comprehensive experiments across the Qwen2.5-VL and Gemma3 model families demonstrate that MASSV increases accepted length by up to 30% and delivers end-to-end inference speedups of up to 1.46x on visually-grounded tasks. MASSV provides a scalable, architecture-compatible method for accelerating both current and future VLMs.

Paperid: 5203, https://arxiv.org/pdf/2505.09577.pdf

Abstract:
While vision-language models have advanced significantly, their application in language-conditioned robotic manipulation is still underexplored, especially for contact-rich tasks that extend beyond visually dominant pick-and-place scenarios. To bridge this gap, we introduce Vision-Tactile-Language-Action model, a novel framework that enables robust policy generation in contact-intensive scenarios by effectively integrating visual and tactile inputs through cross-modal language grounding. A low-cost, multi-modal dataset has been constructed in a simulation environment, containing vision-tactile-action-instruction pairs specifically designed for the fingertip insertion task. Furthermore, we introduce Direct Preference Optimization (DPO) to offer regression-like supervision for the VTLA model, effectively bridging the gap between classification-based next token prediction loss and continuous robotic tasks. Experimental results show that the VTLA model outperforms traditional imitation learning methods (e.g., diffusion policies) and existing multi-modal baselines (TLA/VLA), achieving over 90% success rates on unseen peg shapes. Finally, we conduct real-world peg-in-hole experiments to demonstrate the exceptional Sim2Real performance of the proposed VTLA model. For supplementary videos and results, please visit our project website: https://sites.google.com/view/vtla

Paperid: 5204, https://arxiv.org/pdf/2505.06399.pdf

Abstract:
Autonomous landing is essential for drones deployed in emergency deliveries, post-disaster response, and other large-scale missions. By enabling self-docking on charging platforms, it facilitates continuous operation and significantly extends mission endurance. However, traditional approaches often fall short in dynamic, unstructured environments due to limited semantic awareness and reliance on fixed, context-insensitive safety margins. To address these limitations, we propose a hybrid framework that integrates large language model (LLMs) with model predictive control (MPC). Our approach begins with a vision-language encoder (VLE) (e.g., BLIP), which transforms real-time images into concise textual scene descriptions. These descriptions are processed by a lightweight LLM (e.g., Qwen 2.5 1.5B or LLaMA 3.2 1B) equipped with retrieval-augmented generation (RAG) to classify scene elements and infer context-aware safety buffers, such as 3 meters for pedestrians and 5 meters for vehicles. The resulting semantic flags and unsafe regions are then fed into an MPC module, enabling real-time trajectory replanning that avoids collisions while maintaining high landing precision. We validate our framework in the ROS-Gazebo simulator, where it consistently outperforms conventional vision-based MPC baselines. Our results show a significant reduction in near-miss incidents with dynamic obstacles, while preserving accurate landings in cluttered environments.

Paperid: 5205, https://arxiv.org/pdf/2505.02448.pdf

Abstract:
Out-of-distribution detection (OOD) is a pivotal task for real-world applications that trains models to identify samples that are distributionally different from the in-distribution (ID) data during testing. Recent advances in AI, particularly Vision-Language Models (VLMs) like CLIP, have revolutionized OOD detection by shifting from traditional unimodal image detectors to multimodal image-text detectors. This shift has inspired extensive research; however, existing categorization schemes (e.g., few- or zero-shot types) still rely solely on the availability of ID images, adhering to a unimodal paradigm. To better align with CLIP's cross-modal nature, we propose a new categorization framework rooted in both image and text modalities. Specifically, we categorize existing methods based on how visual and textual information of OOD data is utilized within image + text modalities, and further divide them into four groups: OOD Images (i.e., outliers) Seen or Unseen, and OOD Texts (i.e., learnable vectors or class names) Known or Unknown, across two training strategies (i.e., train-free or training-required). More importantly, we discuss open problems in CLIP-like OOD detection and highlight promising directions for future research, including cross-domain integration, practical applications, and theoretical understanding.

Paperid: 5206, https://arxiv.org/pdf/2504.21432.pdf

Abstract:
A core challenge in AI-guided autonomy is enabling agents to navigate realistically and effectively in previously unseen environments based on natural language commands. We propose UAV-VLN, a novel end-to-end Vision-Language Navigation (VLN) framework for Unmanned Aerial Vehicles (UAVs) that seamlessly integrates Large Language Models (LLMs) with visual perception to facilitate human-interactive navigation. Our system interprets free-form natural language instructions, grounds them into visual observations, and plans feasible aerial trajectories in diverse environments. UAV-VLN leverages the common-sense reasoning capabilities of LLMs to parse high-level semantic goals, while a vision model detects and localizes semantically relevant objects in the environment. By fusing these modalities, the UAV can reason about spatial relationships, disambiguate references in human instructions, and plan context-aware behaviors with minimal task-specific supervision. To ensure robust and interpretable decision-making, the framework includes a cross-modal grounding mechanism that aligns linguistic intent with visual context. We evaluate UAV-VLN across diverse indoor and outdoor navigation scenarios, demonstrating its ability to generalize to novel instructions and environments with minimal task-specific training. Our results show significant improvements in instruction-following accuracy and trajectory efficiency, highlighting the potential of LLM-driven vision-language interfaces for safe, intuitive, and generalizable UAV autonomy.

Paperid: 5207, https://arxiv.org/pdf/2504.17207.pdf

Abstract:
We present a framework for perspective-aware reasoning in vision-language models (VLMs) through mental imagery simulation. Perspective-taking, the ability to perceive an environment or situation from an alternative viewpoint, is a key benchmark for human-level visual understanding, essential for environmental interaction and collaboration with autonomous agents. Despite advancements in spatial reasoning within VLMs, recent research has shown that modern VLMs significantly lack perspective-aware reasoning capabilities and exhibit a strong bias toward egocentric interpretations. To bridge the gap between VLMs and human perception, we focus on the role of mental imagery, where humans perceive the world through abstracted representations that facilitate perspective shifts. Motivated by this, we propose a framework for perspective-aware reasoning, named Abstract Perspective Change (APC), that effectively leverages vision foundation models, such as object detection, segmentation, and orientation estimation, to construct scene abstractions and enable perspective transformations. Our experiments on synthetic and real-image benchmarks, compared with various VLMs, demonstrate significant improvements in perspective-aware reasoning with our framework, further outperforming fine-tuned spatial reasoning models and novel-view-synthesis-based approaches.

Paperid: 5208, https://arxiv.org/pdf/2504.09897.pdf

Abstract:
Multimodal Large Language Models (MLLMs) have shown remarkable versatility in understanding diverse multimodal data and tasks. However, these capabilities come with an increased model scale. While post-training pruning reduces model size in unimodal models, its application to MLLMs often yields limited success. Our analysis discovers that conventional methods fail to account for the unique token attributes across layers and modalities inherent to MLLMs. Inspired by this observation, we propose TAMP, a simple yet effective pruning framework tailored for MLLMs, featuring two key components: (1) Diversity-Aware Sparsity, which adjusts sparsity ratio per layer based on diversities among multimodal output tokens, preserving more parameters in high-diversity layers; and (2) Adaptive Multimodal Input Activation, which identifies representative multimodal input tokens using attention scores to guide unstructured weight pruning. We validate our method on two state-of-the-art MLLMs: LLaVA-NeXT, designed for vision-language tasks, and VideoLLaMA2, capable of processing audio, visual, and language modalities. Empirical experiments across various multimodal evaluation benchmarks demonstrate that each component of our approach substantially outperforms existing pruning techniques.

Paperid: 5209, https://arxiv.org/pdf/2504.09795.pdf

Abstract:
We aim to develop a retrieval-augmented generation (RAG) framework that answers questions over a corpus of visually-rich documents presented in mixed modalities (e.g., charts, tables) and diverse formats (e.g., PDF, PPTX). In this paper, we introduce a new RAG framework, VDocRAG, which can directly understand varied documents and modalities in a unified image format to prevent missing information that occurs by parsing documents to obtain text. To improve the performance, we propose novel self-supervised pre-training tasks that adapt large vision-language models for retrieval by compressing visual information into dense token representations while aligning them with textual content in documents. Furthermore, we introduce OpenDocVQA, the first unified collection of open-domain document visual question answering datasets, encompassing diverse document types and formats. OpenDocVQA provides a comprehensive resource for training and evaluating retrieval and question answering models on visually-rich documents in an open-domain setting. Experiments show that VDocRAG substantially outperforms conventional text-based RAG and has strong generalization capability, highlighting the potential of an effective RAG paradigm for real-world documents.

Paperid: 5210, https://arxiv.org/pdf/2504.09598.pdf

Abstract:
Medical image captioning via vision-language models has shown promising potential for clinical diagnosis assistance. However, generating contextually relevant descriptions with accurate modality recognition remains challenging. We present DualPrompt-MedCap, a novel dual-prompt enhancement framework that augments Large Vision-Language Models (LVLMs) through two specialized components: (1) a modality-aware prompt derived from a semi-supervised classification model pretrained on medical question-answer pairs, and (2) a question-guided prompt leveraging biomedical language model embeddings. To address the lack of captioning ground truth, we also propose an evaluation framework that jointly considers spatial-semantic relevance and medical narrative quality. Experiments on multiple medical datasets demonstrate that DualPrompt-MedCap outperforms the baseline BLIP-3 by achieving a 22% improvement in modality recognition accuracy while generating more comprehensive and question-aligned descriptions. Our method enables the generation of clinically accurate reports that can serve as medical experts' prior knowledge and automatic annotations for downstream vision-language tasks.

Paperid: 5211, https://arxiv.org/pdf/2504.09258.pdf

Abstract:
The diagnosis of pathological images is often limited by expert availability and regional disparities, highlighting the importance of automated diagnosis using Vision-Language Models (VLMs). Traditional multimodal models typically emphasize outcomes over the reasoning process, compromising the reliability of clinical decisions. To address the weak reasoning abilities and lack of supervised processes in pathological VLMs, we have innovatively proposed PathVLM-R1, a visual language model designed specifically for pathological images. We have based our model on Qwen2.5-VL-7B-Instruct and enhanced its performance for pathological tasks through meticulously designed post-training strategies. Firstly, we conduct supervised fine-tuning guided by pathological data to imbue the model with foundational pathological knowledge, forming a new pathological base model. Subsequently, we introduce Group Relative Policy Optimization (GRPO) and propose a dual reward-driven reinforcement learning optimization, ensuring strict constraint on logical supervision of the reasoning process and accuracy of results via cross-modal process reward and outcome accuracy reward. In the pathological image question-answering tasks, the testing results of PathVLM-R1 demonstrate a 14% improvement in accuracy compared to baseline methods, and it demonstrated superior performance compared to the Qwen2.5-VL-32B version despite having a significantly smaller parameter size. Furthermore, in out-domain data evaluation involving four medical imaging modalities: Computed Tomography (CT), dermoscopy, fundus photography, and Optical Coherence Tomography (OCT) images: PathVLM-R1's transfer performance improved by an average of 17.3% compared to traditional SFT methods. These results clearly indicate that PathVLM-R1 not only enhances accuracy but also possesses broad applicability and expansion potential.

Paperid: 5212, https://arxiv.org/pdf/2504.08772.pdf

Abstract:
In offline reinforcement learning (RL), learning from fixed datasets presents a promising solution for domains where real-time interaction with the environment is expensive or risky. However, designing dense reward signals for offline dataset requires significant human effort and domain expertise. Reinforcement learning with human feedback (RLHF) has emerged as an alternative, but it remains costly due to the human-in-the-loop process, prompting interest in automated reward generation models. To address this, we propose Reward Generation via Large Vision-Language Models (RG-VLM), which leverages the reasoning capabilities of LVLMs to generate rewards from offline data without human involvement. RG-VLM improves generalization in long-horizon tasks and can be seamlessly integrated with the sparse reward signals to enhance task performance, demonstrating its potential as an auxiliary reward signal.

Paperid: 5213, https://arxiv.org/pdf/2504.07336.pdf

Abstract:
Medical image segmentation has achieved remarkable success through the continuous advancement of UNet-based and Transformer-based foundation backbones. However, clinical diagnosis in the real world often requires integrating domain knowledge, especially textual information. Conducting multimodal learning involves visual and text modalities shown as a solution, but collecting paired vision-language datasets is expensive and time-consuming, posing significant challenges. Inspired by the superior ability in numerous cross-modal tasks for Large Language Models (LLMs), we proposed a novel Vision-LLM union framework to address the issues. Specifically, we introduce frozen LLMs for zero-shot instruction generation based on corresponding medical images, imitating the radiology scanning and report generation process. {To better approximate real-world diagnostic processes}, we generate more precise text instruction from multimodal radiology images (e.g., T1-w or T2-w MRI and CT). Based on the impressive ability of semantic understanding and rich knowledge of LLMs. This process emphasizes extracting special features from different modalities and reunion the information for the ultimate clinical diagnostic. With generated text instruction, our proposed union segmentation framework can handle multimodal segmentation without prior collected vision-language datasets. To evaluate our proposed method, we conduct comprehensive experiments with influential baselines, the statistical results and the visualized case study demonstrate the superiority of our novel method.}

Paperid: 5214, https://arxiv.org/pdf/2504.03714.pdf

Abstract:
Large Language Models (LLMs) and Vision-Language Models (VLMs) have achieved impressive performance across a wide range of tasks, yet they remain vulnerable to carefully crafted perturbations. In this study, we seek to pinpoint the sources of this fragility by identifying parameters and input dimensions (pixels or token embeddings) that are susceptible to such perturbations. To this end, we propose a stability measure called \textbf{FI}, \textbf{F}irst order local \textbf{I}nfluence, which is rooted in information geometry and quantifies the sensitivity of individual parameter and input dimensions. Our extensive analysis across LLMs and VLMs (from 1.5B to 13B parameters) reveals that: (I) A small subset of parameters or input dimensions with high FI values disproportionately contribute to model brittleness. (II) Mitigating the influence of these vulnerable parameters during model merging leads to improved performance.

Paperid: 5215, https://arxiv.org/pdf/2504.01700.pdf

Abstract:
Personalization in social robotics is critical for fostering effective human-robot interactions, yet systems often face the cold start problem, where initial user preferences or characteristics are unavailable. This paper proposes a novel framework called USER-LLM R1 for a user-aware conversational agent that addresses this challenge through dynamic user profiling and model initiation. Our approach integrates chain-of-thought (CoT) reasoning models to iteratively infer user preferences and vision-language models (VLMs) to initialize user profiles from multimodal inputs, enabling personalized interactions from the first encounter. Leveraging a Retrieval-Augmented Generation (RAG) architecture, the system dynamically refines user representations within an inherent CoT process, ensuring contextually relevant and adaptive responses. Evaluations on the ElderlyTech-VQA Bench demonstrate significant improvements in ROUGE-1 (+23.2%), ROUGE-2 (+0.6%), and ROUGE-L (+8%) F1 scores over state-of-the-art baselines, with ablation studies underscoring the impact of reasoning model size on performance. Human evaluations further validate the framework's efficacy, particularly for elderly users, where tailored responses enhance engagement and trust. Ethical considerations, including privacy preservation and bias mitigation, are rigorously discussed and addressed to ensure responsible deployment.

Paperid: 5216, https://arxiv.org/pdf/2503.23534.pdf

Abstract:
Medical image segmentation typically relies solely on visual data, overlooking the rich textual information clinicians use for diagnosis. Vision-language models attempt to bridge this gap, but existing approaches often process visual and textual features independently, resulting in weak cross-modal alignment. Simple fusion techniques fail due to the inherent differences between spatial visual features and sequential text embeddings. Additionally, medical terminology deviates from general language, limiting the effectiveness of off-the-shelf text encoders and further hindering vision-language alignment. We propose BiPVL-Seg, an end-to-end framework that integrates vision-language fusion and embedding alignment through architectural and training innovations, where both components reinforce each other to enhance medical image segmentation. BiPVL-Seg introduces bidirectional progressive fusion in the architecture, which facilitates stage-wise information exchange between vision and text encoders. Additionally, it incorporates global-local contrastive alignment, a training objective that enhances the text encoder's comprehension by aligning text and vision embeddings at both class and concept levels. Extensive experiments on diverse medical imaging benchmarks across CT and MR modalities demonstrate BiPVL-Seg's superior performance when compared with state-of-the-art methods in complex multi-class segmentation. Source code is available in this GitHub repository.

Paperid: 5217, https://arxiv.org/pdf/2503.23463.pdf

Abstract:
We present OpenDriveVLA, a Vision-Language Action (VLA) model designed for end-to-end autonomous driving. OpenDriveVLA builds upon open-source pre-trained large Vision-Language Models (VLMs) to generate reliable driving actions, conditioned on 3D environmental perception, ego vehicle states, and driver commands. To bridge the modality gap between driving visual representations and language embeddings, we propose a hierarchical vision-language alignment process, projecting both 2D and 3D structured visual tokens into a unified semantic space. Besides, OpenDriveVLA models the dynamic relationships between the ego vehicle, surrounding agents, and static road elements through an autoregressive agent-env-ego interaction process, ensuring both spatially and behaviorally informed trajectory planning. Extensive experiments on the nuScenes dataset demonstrate that OpenDriveVLA achieves state-of-the-art results across open-loop trajectory planning and driving-related question-answering tasks. Qualitative analyses further illustrate OpenDriveVLA's superior capability to follow high-level driving commands and robustly generate trajectories under challenging scenarios, highlighting its potential for next-generation end-to-end autonomous driving. We will release our code to facilitate further research in this domain.

Paperid: 5218, https://arxiv.org/pdf/2503.23083.pdf

Abstract:
Adapting pre-trained models has become an effective strategy in artificial intelligence, offering a scalable and efficient alternative to training models from scratch. In the context of remote sensing (RS), where visual grounding(VG) remains underexplored, this approach enables the deployment of powerful vision-language models to achieve robust cross-modal understanding while significantly reducing computational overhead. To address this, we applied Parameter Efficient Fine Tuning (PEFT) techniques to adapt these models for RS-specific VG tasks. Specifically, we evaluated LoRA placement across different modules in Grounding DINO and used BitFit and adapters to fine-tune the OFA foundation model pre-trained on general-purpose VG datasets. This approach achieved performance comparable to or surpassing current State Of The Art (SOTA) models while significantly reducing computational costs. This study highlights the potential of PEFT techniques to advance efficient and precise multi-modal analysis in RS, offering a practical and cost-effective alternative to full model training.

Paperid: 5219, https://arxiv.org/pdf/2503.21747.pdf

Abstract:
Object-centric representation learning aims to decompose visual scenes into fixed-size vectors called "slots" or "object files", where each slot captures a distinct object. Current state-of-the-art object-centric models have shown remarkable success in object discovery in diverse domains, including complex real-world scenes. However, these models suffer from a key limitation: they lack controllability. Specifically, current object-centric models learn representations based on their preconceived understanding of objects, without allowing user input to guide which objects are represented. Introducing controllability into object-centric models could unlock a range of useful capabilities, such as the ability to extract instance-specific representations from a scene. In this work, we propose a novel approach for user-directed control over slot representations by conditioning slots on language descriptions. The proposed ConTRoLlable Object-centric representation learning approach, which we term CTRL-O, achieves targeted object-language binding in complex real-world scenes without requiring mask supervision. Next, we apply these controllable slot representations on two downstream vision language tasks: text-to-image generation and visual question answering. The proposed approach enables instance-specific text-to-image generation and also achieves strong performance on visual question answering.

Paperid: 5220, https://arxiv.org/pdf/2503.19982.pdf

Abstract:
Face anti-spoofing (FAS) plays a pivotal role in ensuring the security and reliability of face recognition systems. With advancements in vision-language pretrained (VLP) models, recent two-class FAS techniques have leveraged the advantages of using VLP guidance, while this potential remains unexplored in one-class FAS methods. The one-class FAS focuses on learning intrinsic liveness features solely from live training images to differentiate between live and spoof faces. However, the lack of spoof training data can lead one-class FAS models to inadvertently incorporate domain information irrelevant to the live/spoof distinction (e.g., facial content), causing performance degradation when tested with a new application domain. To address this issue, we propose a novel framework called Spoof-aware one-class face anti-spoofing with Language Image Pretraining (SLIP). Given that live faces should ideally not be obscured by any spoof-attack-related objects (e.g., paper, or masks) and are assumed to yield zero spoof cue maps, we first propose an effective language-guided spoof cue map estimation to enhance one-class FAS models by simulating whether the underlying faces are covered by attack-related objects and generating corresponding nonzero spoof cue maps. Next, we introduce a novel prompt-driven liveness feature disentanglement to alleviate live/spoof-irrelative domain variations by disentangling live/spoof-relevant and domain-dependent information. Finally, we design an effective augmentation strategy by fusing latent features from live images and spoof prompts to generate spoof-like image features and thus diversify latent spoof features to facilitate the learning of one-class FAS. Our extensive experiments and ablation studies support that SLIP consistently outperforms previous one-class FAS methods.

Paperid: 5221, https://arxiv.org/pdf/2503.17782.pdf

Abstract:
Vision-language models like CLIP have shown impressive capabilities in aligning images and text, but they often struggle with lengthy and detailed text descriptions because of their training focus on short and concise captions. We present GOAL (Global-local Object Alignment Learning), a novel fine-tuning method that enhances CLIP's ability to handle lengthy text by leveraging both global and local semantic alignments between image and lengthy text. Our approach consists of two key components: Local Image-Sentence Matching (LISM), which identifies corresponding pairs between image segments and descriptive sentences, and Token Similarity-based Learning (TSL), which efficiently propagates local element attention through these matched pairs. Evaluating GOAL on three new benchmarks for image-lengthy text retrieval, we demonstrate significant improvements over baseline CLIP fine-tuning, establishing a simple yet effective approach for adapting CLIP to detailed textual descriptions. Through extensive experiments, we show that our method's focus on local semantic alignment alongside global context leads to more nuanced and representative embeddings, particularly beneficial for tasks requiring fine-grained understanding of lengthy text descriptions.

Paperid: 5222, https://arxiv.org/pdf/2503.11005.pdf

Abstract:
In pursuit of detecting unstinted objects that extend beyond predefined categories, prior arts of open-vocabulary object detection (OVD) typically resort to pretrained vision-language models (VLMs) for base-to-novel category generalization. However, to mitigate the misalignment between upstream image-text pretraining and downstream region-level perception, additional supervisions are indispensable, eg, image-text pairs or pseudo annotations generated via self-training strategies. In this work, we propose CCKT-Det trained without any extra supervision. The proposed framework constructs a cyclic and dynamic knowledge transfer from language queries and visual region features extracted from VLMs, which forces the detector to closely align with the visual-semantic space of VLMs. Specifically, 1) we prefilter and inject semantic priors to guide the learning of queries, and 2) introduce a regional contrastive loss to improve the awareness of queries on novel objects. CCKT-Det can consistently improve performance as the scale of VLMs increases, all while requiring the detector at a moderate level of computation overhead. Comprehensive experimental results demonstrate that our method achieves performance gain of +2.9% and +10.2% AP50 over previous state-of-the-arts on the challenging COCO benchmark, both without and with a stronger teacher model.

Paperid: 5223, https://arxiv.org/pdf/2503.10779.pdf

Abstract:
Large-scale vision-language models (VLMs), trained on extensive datasets of image-text pairs, exhibit strong multimodal understanding capabilities by implicitly learning associations between textual descriptions and image regions. This emergent ability enables zero-shot object detection and segmentation, using techniques that rely on text-image attention maps, without necessarily training on abundant labeled segmentation datasets. However, performance of such methods depends heavily on prompt engineering and manually selected layers or head choices for the attention layers. In this work, we demonstrate that, rather than relying solely on textual prompts, providing a single visual example for each category and fine-tuning the text-to-image attention layers and embeddings significantly improves the performance. Additionally, we propose learning an ensemble through few-shot fine-tuning across multiple layers and/or prompts. An entropy-based ranking and selection mechanism for text-to-image attention layers is proposed to identify the top-performing layers without the need for segmentation labels. This eliminates the need for hyper-parameter selection of text-to-image attention layers, providing a more flexible and scalable solution for open-vocabulary segmentation. We show that this approach yields strong zero-shot performance, further enhanced through fine-tuning with a single visual example. Moreover, we demonstrate that our method and findings are general and can be applied across various vision-language models (VLMs).

Paperid: 5224, https://arxiv.org/pdf/2503.10322.pdf

Abstract:
Vision Language Models (VLMs) pretrained on Internet-scale vision-language data have demonstrated the potential to transfer their knowledge to robotic learning. However, the existing paradigm encounters three critical challenges: (1) expensive inference cost resulting from large-scale model parameters, (2) frequent domain shifts caused by mismatched data modalities, and (3) limited capacity to handle past or future experiences. In this work, we propose LiteVLP, a lightweight, memory-based, and general-purpose vision-language policy generation model. LiteVLP is built upon a pre-trained 1B-parameter VLM and fine-tuned on a tiny-scale and conversation-style robotic dataset. Through extensive experiments, we demonstrate that LiteVLP outperforms state-of-the-art vision-language policy on VIMA-Bench, with minimal training time. Furthermore, LiteVLP exhibits superior inference speed while maintaining exceptional high accuracy. In long-horizon manipulation tasks, LiteVLP also shows remarkable memory ability, outperforming the best-performing baseline model by 18.8%. These results highlight LiteVLP as a promising model to integrating the intelligence of VLMs into robotic learning.

Paperid: 5225, https://arxiv.org/pdf/2503.10265.pdf

Abstract:
Integration of Vision-Language Models (VLMs) in surgical intelligence is hindered by hallucinations, domain knowledge gaps, and limited understanding of task interdependencies within surgical scenes, undermining clinical reliability. While recent VLMs demonstrate strong general reasoning and thinking capabilities, they still lack the domain expertise and task-awareness required for precise surgical scene interpretation. Although Chain-of-Thought (CoT) can structure reasoning more effectively, current approaches rely on self-generated CoT steps, which often exacerbate inherent domain gaps and hallucinations. To overcome this, we present SurgRAW, a CoT-driven multi-agent framework that delivers transparent, interpretable insights for most tasks in robotic-assisted surgery. By employing specialized CoT prompts across five tasks: instrument recognition, action recognition, action prediction, patient data extraction, and outcome assessment, SurgRAW mitigates hallucinations through structured, domain-aware reasoning. Retrieval-Augmented Generation (RAG) is also integrated to external medical knowledge to bridge domain gaps and improve response reliability. Most importantly, a hierarchical agentic system ensures that CoT-embedded VLM agents collaborate effectively while understanding task interdependencies, with a panel discussion mechanism promotes logical consistency. To evaluate our method, we introduce SurgCoTBench, the first reasoning-based dataset with structured frame-level annotations. With comprehensive experiments, we demonstrate the effectiveness of proposed SurgRAW with 29.32% accuracy improvement over baseline VLMs on 12 robotic procedures, achieving the state-of-the-art performance and advancing explainable, trustworthy, and autonomous surgical assistance.

Paperid: 5226, https://arxiv.org/pdf/2503.08548.pdf

Abstract:
Significant progress has been made in vision-language models. However, language-conditioned robotic manipulation for contact-rich tasks remains underexplored, particularly in terms of tactile sensing. To address this gap, we introduce the Tactile-Language-Action (TLA) model, which effectively processes sequential tactile feedback via cross-modal language grounding to enable robust policy generation in contact-intensive scenarios. In addition, we construct a comprehensive dataset that contains 24k pairs of tactile action instruction data, customized for fingertip peg-in-hole assembly, providing essential resources for TLA training and evaluation. Our results show that TLA significantly outperforms traditional imitation learning methods (e.g., diffusion policy) in terms of effective action generation and action accuracy, while demonstrating strong generalization capabilities by achieving over 85\% success rate on previously unseen assembly clearances and peg shapes. We publicly release all data and code in the hope of advancing research in language-conditioned tactile manipulation skill learning. Project website: https://sites.google.com/view/tactile-language-action/

Paperid: 5227, https://arxiv.org/pdf/2503.08496.pdf

Abstract:
It has been a longstanding goal within image captioning to move beyond a dependence on object detection. We investigate using superpixels coupled with Vision Language Models (VLMs) to bridge the gap between detector-based captioning architectures and those that solely pretrain on large datasets. Our novel superpixel approach ensures that the model receives object-like features whilst the use of VLMs provides our model with open set object understanding. Furthermore, we extend our architecture to make use of multi-resolution inputs, allowing our model to view images in different levels of detail, and use an attention mechanism to determine which parts are most relevant to the caption. We demonstrate our model's performance with multiple VLMs and through a range of ablations detailing the impact of different architectural choices. Our full model achieves a competitive CIDEr score of $136.9$ on the COCO Karpathy split.

Paperid: 5228, https://arxiv.org/pdf/2503.07631.pdf

Abstract:
We present a challenging benchmark for the Open WorLd VISual question answering (OWLViz) task. OWLViz presents concise, unambiguous queries that require integrating multiple capabilities, including visual understanding, web exploration, and specialized tool usage. While humans achieve 69.2% accuracy on these intuitive tasks, even state-of-the-art VLMs struggle, with the best model, Gemini 2.0, achieving only 26.6% accuracy. Current agentic VLMs, which rely on limited vision and vision-language models as tools, perform even worse. This performance gap reveals significant limitations in multimodal systems' ability to select appropriate tools and execute complex reasoning sequences, establishing new directions for advancing practical AI research.

Paperid: 5229, https://arxiv.org/pdf/2503.07075.pdf

Abstract:
Vision-Language Models (VLMs) have demonstrated impressive performance on various visual tasks, yet they still require adaptation on downstream tasks to achieve optimal performance. Recently, various adaptation technologies have been proposed, but we observe they often underperform in fine-grained visual recognition, which requires models to capture subtle yet discriminative features to distinguish similar sub-categories. Current adaptation methods typically rely on an alignment-based prediction framework, \ie the visual feature is compared with each class prompt for similarity calculation as the final prediction, which lacks class interaction during the forward pass. Besides, learning single uni-modal feature further restricts the model's expressive capacity. Therefore, we propose a novel mechanism, XR-VLM, to discover subtle differences by modeling cross-relationships, which specifically excels in scenarios involving multiple features. Our method introduces a unified multi-part visual feature extraction module designed to seamlessly integrate with the diverse backbones inherent in VLMs. Additionally, we develop a multi-part prompt learning module to capture multi-perspective descriptions of sub-categories. To further enhance discriminative capability, we propose a cross relationship modeling pattern that combines visual feature with all class prompt features, enabling a deeper exploration of the relationships between these two modalities. Extensive experiments have been conducted on various fine-grained datasets, and the results demonstrate that our method achieves significant improvements compared to current state-of-the-art approaches. Code will be released.

Paperid: 5230, https://arxiv.org/pdf/2503.05093.pdf

Abstract:
Current research on bias in Vision Language Models (VLMs) has important limitations: it is focused exclusively on trait associations while ignoring other forms of stereotyping, it examines specific contexts where biases are expected to appear, and it conceptualizes social categories like race and gender as binary, ignoring the multifaceted nature of these identities. Using standardized facial images that vary in prototypicality, we test four VLMs for both trait associations and homogeneity bias in open-ended contexts. We find that VLMs consistently generate more uniform stories for women compared to men, with people who are more gender prototypical in appearance being represented more uniformly. By contrast, VLMs represent White Americans more uniformly than Black Americans. Unlike with gender prototypicality, race prototypicality was not related to stronger uniformity. In terms of trait associations, we find limited evidence of stereotyping-Black Americans were consistently linked with basketball across all models, while other racial associations (i.e., art, healthcare, appearance) varied by specific VLM. These findings demonstrate that VLM stereotyping manifests in ways that go beyond simple group membership, suggesting that conventional bias mitigation strategies may be insufficient to address VLM stereotyping and that homogeneity bias persists even when trait associations are less apparent in model outputs.

Paperid: 5231, https://arxiv.org/pdf/2503.00162.pdf

Abstract:
In recent years, online lecture videos have become an increasingly popular resource for acquiring new knowledge. Systems capable of effectively understanding/indexing lecture videos are thus highly desirable, enabling downstream tasks like question answering to help users efficiently locate specific information within videos. This work proposes PreMind, a novel multi-agent multimodal framework that leverages various large models for advanced understanding/indexing of presentation-style videos. PreMind first segments videos into slide-presentation segments using a Vision-Language Model (VLM) to enhance modern shot-detection techniques. Each segment is then analyzed to generate multimodal indexes through three key steps: (1) extracting slide visual content, (2) transcribing speech narratives, and (3) consolidating these visual and speech contents into an integrated understanding. Three innovative mechanisms are also proposed to improve performance: leveraging prior lecture knowledge to refine visual understanding, detecting/correcting speech transcription errors using a VLM, and utilizing a critic agent for dynamic iterative self-reflection in vision analysis. Compared to traditional video indexing methods, PreMind captures rich, reliable multimodal information, allowing users to search for details like abbreviations shown only on slides. Systematic evaluations on the public LPM dataset and an internal enterprise dataset are conducted to validate PreMind's effectiveness, supported by detailed analyses.

Paperid: 5232, https://arxiv.org/pdf/2502.20850.pdf

Abstract:
Recent advances in vision-language models (VLMs) have shown remarkable potential in bridging visual and textual modalities. In computational pathology, domain-specific VLMs, which are pre-trained on extensive histopathology image-text datasets, have succeeded in various downstream tasks. However, existing research has primarily focused on the pre-training process and direct applications of VLMs on the patch level, leaving their great potential for whole slide image (WSI) applications unexplored. In this study, we hypothesize that pre-trained VLMs inherently capture informative and interpretable WSI representations through quantitative feature extraction. To validate this hypothesis, we introduce Vision and Language Embeddings for Explainable WSI Representation (VLEER), a novel method designed to leverage VLMs for WSI representation. We systematically evaluate VLEER on three pathological WSI datasets, proving its better performance in WSI analysis compared to conventional vision features. More importantly, VLEER offers the unique advantage of interpretability, enabling direct human-readable insights into the results by leveraging the textual modality for detailed pathology annotations, providing clear reasoning for WSI-level pathology downstream tasks.

Paperid: 5233, https://arxiv.org/pdf/2502.19672.pdf

Abstract:
Multimodal Large Language Models (MLLMs), built upon LLMs, have recently gained attention for their capabilities in image recognition and understanding. However, while MLLMs are vulnerable to adversarial attacks, the transferability of these attacks across different models remains limited, especially under targeted attack setting. Existing methods primarily focus on vision-specific perturbations but struggle with the complex nature of vision-language modality alignment. In this work, we introduce the Dynamic Vision-Language Alignment (DynVLA) Attack, a novel approach that injects dynamic perturbations into the vision-language connector to enhance generalization across diverse vision-language alignment of different models. Our experimental results show that DynVLA significantly improves the transferability of adversarial examples across various MLLMs, including BLIP2, InstructBLIP, MiniGPT4, LLaVA, and closed-source models such as Gemini.

Paperid: 5234, https://arxiv.org/pdf/2502.15079.pdf

Abstract:
Large Vision-Language Models often generate hallucinated content that is not grounded in its visual inputs. While prior work focuses on mitigating hallucinations, we instead explore leveraging hallucination correction as a training objective to improve video-language alignment. We introduce HACA, a self-training framework learning to correct hallucinations in descriptions that do not align with the video content. By identifying and correcting inconsistencies, HACA enhances the model's ability to align video and textual representations for spatio-temporal reasoning. Our experimental results show consistent gains in video-caption binding and text-to-video retrieval tasks, demonstrating that hallucination correction-inspired tasks serve as an effective strategy for improving vision and language alignment.

Paperid: 5235, https://arxiv.org/pdf/2502.10636.pdf

Abstract:
The integration of vision-language models into robotic systems constitutes a significant advancement in enabling machines to interact with their surroundings in a more intuitive manner. While VLMs offer rich multimodal reasoning, existing approaches lack user-specific adaptability, often relying on generic interaction paradigms that fail to account for individual behavioral, contextual, or socio-emotional nuances. When customization is attempted, ethical concerns arise from unmitigated biases in user data, risking exclusion or unfair treatment. To address these dual challenges, we propose User-VLM 360Â°, a holistic framework integrating multimodal user modeling with bias-aware optimization. Our approach features: (1) user-aware tuning that adapts interactions in real time using visual-linguistic signals; (2) bias mitigation via preference optimization; and (3) curated 360Â° socio-emotive interaction datasets annotated with demographic, emotion, and relational metadata. Evaluations across eight benchmarks demonstrate state-of-the-art results: +35.3% F1 in personalized VQA, +47.5% F1 in facial features understanding, 15% bias reduction, and 30X speedup over baselines. Ablation studies confirm component efficacy, and deployment on the Pepper robot validates real-time adaptability across diverse users. We open-source parameter-efficient 3B/10B models and an ethical verification framework for responsible adaptation.

Paperid: 5236, https://arxiv.org/pdf/2502.08642.pdf

Abstract:
Recent advancements in large vision-language models have enabled highly expressive and diverse vector sketch generation. However, state-of-the-art methods rely on a time-consuming optimization process involving repeated feedback from a pretrained model to determine stroke placement. Consequently, despite producing impressive sketches, these methods are limited in practical applications. In this work, we introduce SwiftSketch, a diffusion model for image-conditioned vector sketch generation that can produce high-quality sketches in less than a second. SwiftSketch operates by progressively denoising stroke control points sampled from a Gaussian distribution. Its transformer-decoder architecture is designed to effectively handle the discrete nature of vector representation and capture the inherent global dependencies between strokes. To train SwiftSketch, we construct a synthetic dataset of image-sketch pairs, addressing the limitations of existing sketch datasets, which are often created by non-artists and lack professional quality. For generating these synthetic sketches, we introduce ControlSketch, a method that enhances SDS-based techniques by incorporating precise spatial control through a depth-aware ControlNet. We demonstrate that SwiftSketch generalizes across diverse concepts, efficiently producing sketches that combine high fidelity with a natural and visually appealing style.

Paperid: 5237, https://arxiv.org/pdf/2502.07847.pdf

Abstract:
Since the establishment of vision-language foundation models as the new mainstay in low-shot vision classification tasks, the question of domain generalization arising from insufficient target data is assuming more importance. This scarcity challenge induces sampling bias and amplifies model sensitivity to variations and shifts in data distributions. While fine-tuning on multiple domains could mitigate such domain generalization issues, it is resource-intensive and demands diverse data sources. In this work, we systematically analyze two critical challenges: (1) covariate shift between the pre-training distribution and the underspecified target distribution, and (2) confidence misalignment, where predictions on novel data are overconfident. To address both challenges simultaneously, we introduce \textbf{Confidence-Calibrated Covariate Shift Correction (CalShift)} -- a unified approach that combines a Fisher information penalty to mitigate covariate shift and a Confidence Misalignment Penalty (CMP) to reduce overconfidence in misclassified examples. Experimental evaluations across various vision and covariate shift benchmarks demonstrate that CalShift significantly improves model calibration, achieving up to a 5.82\% reduction in Expected Calibration Error (ECE). Furthermore, CalShift enhances robustness, improving accuracy by 3.5\% on challenging datasets impacted by covariate shifts. Our results highlight CalShift as a promising strategy for building robust and reliable low-shot vision-language systems for real-world applications.

Paperid: 5238, https://arxiv.org/pdf/2502.04531.pdf

Abstract:
Object placement in robotic tasks is inherently challenging due to the diversity of object geometries and placement configurations. To address this, we propose AnyPlace, a two-stage method trained entirely on synthetic data, capable of predicting a wide range of feasible placement poses for real-world tasks. Our key insight is that by leveraging a Vision-Language Model (VLM) to identify rough placement locations, we focus only on the relevant regions for local placement, which enables us to train the low-level placement-pose-prediction model to capture diverse placements efficiently. For training, we generate a fully synthetic dataset of randomly generated objects in different placement configurations (insertion, stacking, hanging) and train local placement-prediction models. We conduct extensive evaluations in simulation, demonstrating that our method outperforms baselines in terms of success rate, coverage of possible placement modes, and precision. In real-world experiments, we show how our approach directly transfers models trained purely on synthetic data to the real world, where it successfully performs placements in scenarios where other models struggle -- such as with varying object geometries, diverse placement modes, and achieving high precision for fine placement. More at: https://any-place.github.io.

Paperid: 5239, https://arxiv.org/pdf/2502.03836.pdf

Abstract:
Human mesh recovery can be approached using either regression-based or optimization-based methods. Regression models achieve high pose accuracy but struggle with model-to-image alignment due to the lack of explicit 2D-3D correspondences. In contrast, optimization-based methods align 3D models to 2D observations but are prone to local minima and depth ambiguity. In this work, we leverage large vision-language models (VLMs) to generate interactive body part descriptions, which serve as implicit constraints to enhance 3D perception and limit the optimization space. Specifically, we formulate monocular human mesh recovery as a distribution adaptation task by integrating both 2D observations and language descriptions. To bridge the gap between text and 3D pose signals, we first train a text encoder and a pose VQ-VAE, aligning texts to body poses in a shared latent space using contrastive learning. Subsequently, we employ a diffusion-based framework to refine the initial parameters guided by gradients derived from both 2D observations and text descriptions. Finally, the model can produce poses with accurate 3D perception and image consistency. Experimental results on multiple benchmarks validate its effectiveness. The code will be made publicly available.

Paperid: 5240, https://arxiv.org/pdf/2502.03621.pdf

Abstract:
We present a method for augmenting real-world videos with newly generated dynamic content. Given an input video and a simple user-provided text instruction describing the desired content, our method synthesizes dynamic objects or complex scene effects that naturally interact with the existing scene over time. The position, appearance, and motion of the new content are seamlessly integrated into the original footage while accounting for camera motion, occlusions, and interactions with other dynamic objects in the scene, resulting in a cohesive and realistic output video. We achieve this via a zero-shot, training-free framework that harnesses a pre-trained text-to-video diffusion transformer to synthesize the new content and a pre-trained Vision Language Model to envision the augmented scene in detail. Specifically, we introduce a novel inference-based method that manipulates features within the attention mechanism, enabling accurate localization and seamless integration of the new content while preserving the integrity of the original scene. Our method is fully automated, requiring only a simple user instruction. We demonstrate its effectiveness on a wide range of edits applied to real-world videos, encompassing diverse objects and scenarios involving both camera and object motion.

Paperid: 5241, https://arxiv.org/pdf/2502.00114.pdf

Abstract:
Hand-drawn maps can be used to convey navigation instructions between humans and robots in a natural and efficient manner. However, these maps can often contain inaccuracies such as scale distortions and missing landmarks which present challenges for mobile robot navigation. This paper introduces a novel Hand-drawn Map Navigation (HAM-Nav) architecture that leverages pre-trained vision language models (VLMs) for robot navigation across diverse environments, hand-drawing styles, and robot embodiments, even in the presence of map inaccuracies. HAM-Nav integrates a unique Selective Visual Association Prompting approach for topological map-based position estimation and navigation planning as well as a Predictive Navigation Plan Parser to infer missing landmarks. Extensive experiments were conducted in photorealistic simulated environments, using both wheeled and legged robots, demonstrating the effectiveness of HAM-Nav in terms of navigation success rates and Success weighted by Path Length. Furthermore, a user study in real-world environments highlighted the practical utility of hand-drawn maps for robot navigation as well as successful navigation outcomes compared against a non-hand-drawn map approach.

Paperid: 5242, https://arxiv.org/pdf/2501.15144.pdf

Abstract:
This work investigates the capabilities of current vision-language models (VLMs) in visual understanding and attribute measurement of primitive shapes using a benchmark focused on controlled 2D shape configurations with variations in spatial positioning, occlusion, rotation, size, and shape attributes such as type, quadrant, center-coordinates, rotation, occlusion status, and color as shown in Figure 1 and supplementary Figures S3-S81. We fine-tune state-of-the-art VLMs (2B-8B parameters) using Low-Rank Adaptation (LoRA) and validate them on multiple out-of-domain (OD) scenarios from our proposed benchmark. Our findings reveal that coherent sentence-based outputs outperform tuple formats, particularly in OD scenarios with large domain gaps. Additionally, we demonstrate that scaling numeric tokens during loss computation enhances numerical approximation capabilities, further improving performance on spatial and measurement tasks. These results highlight the importance of output format design, loss scaling strategies, and robust generalization techniques in enhancing the training and fine-tuning of VLMs, particularly for tasks requiring precise spatial approximations and strong OD generalization.

Paperid: 5243, https://arxiv.org/pdf/2501.13338.pdf

Abstract:
Mobile exploration is a longstanding challenge in robotics, yet current methods primarily focus on active perception instead of active interaction, limiting the robot's ability to interact with and fully explore its environment. Existing robotic exploration approaches via active interaction are often restricted to tabletop scenes, neglecting the unique challenges posed by mobile exploration, such as large exploration spaces, complex action spaces, and diverse object relations. In this work, we introduce a 3D relational object graph that encodes diverse object relations and enables exploration through active interaction. We develop a system based on this representation and evaluate it across diverse scenes. Our qualitative and quantitative results demonstrate the system's effectiveness and generalization capabilities, outperforming methods that rely solely on vision-language models (VLMs).

Paperid: 5244, https://arxiv.org/pdf/2412.20750.pdf

Abstract:
Large-scale Vision-Language Models (VLMs) have achieved notable progress in aligning visual inputs with text. However, their ability to deeply understand the unique physical properties of non-RGB vision sensor images remains limited. In this paper, we revisit and analyze these limitations and introduce a novel, cost-efficient paradigm that significantly advances sensor image understanding-without requiring extensive training data or any modifications to the existing VLM architectures. Specifically, we propose Sensor-Aware Attributes Fine-Tuning (SAFT) with the Diverse Negative Attributes (DNA) optimization, which leverages minimal sensor-specific data to enable robust learning of non-RGB characteristics and overcome RGB-centric biases inherent in current VLMs. In addition, we present VS-TDX-the first comprehensive, public benchmark designed to rigorously evaluate VLMs' sensor-specific understanding across diverse and realistic scenarios. Through extensive experiments on VLMs and various sensor modalities, we validate that our method consistently delivers superior performance and generalization under resource-constrained and architecture-invariant settings. Our approach provides a practical advance towards scalable deployment of VLMs in increasingly sensor-diverse real-world environments.

Paperid: 5245, https://arxiv.org/pdf/2412.18525.pdf

Abstract:
Computer Vision (CV) has yet to fully achieve the zero-shot task generalization observed in Natural Language Processing (NLP), despite following many of the milestones established in NLP, such as large transformer models, extensive pre-training, and the auto-regression paradigm, among others. In this paper, we explore the idea that CV adopts discrete and terminological task definitions (\eg, ``image segmentation''), which may be a key barrier to zero-shot task generalization. Our hypothesis is that without truly understanding previously-seen tasks--due to these terminological definitions--deep models struggle to generalize to novel tasks. To verify this, we introduce Explanatory Instructions, which provide an intuitive way to define CV task objectives through detailed linguistic transformations from input images to outputs. We create a large-scale dataset comprising 12 million ``image input $\to$ explanatory instruction $\to$ output'' triplets, and train an auto-regressive-based vision-language model (AR-based VLM) that takes both images and explanatory instructions as input. By learning to follow these instructions, the AR-based VLM achieves instruction-level zero-shot capabilities for previously-seen tasks and demonstrates strong zero-shot generalization for unseen CV tasks. Code and dataset will be openly available on our GitHub repository.

Paperid: 5246, https://arxiv.org/pdf/2412.14145.pdf

Abstract:
The visual understanding are often approached from 3 granular levels: image, patch and pixel. Visual Tokenization, trained by self-supervised reconstructive learning, compresses visual data by codebook in patch-level with marginal information loss, but the visual tokens does not have semantic meaning. Open Vocabulary semantic segmentation benefits from the evolving Vision-Language models (VLMs) with strong image zero-shot capability, but transferring image-level to pixel-level understanding remains an imminent challenge. In this paper, we treat segmentation as tokenizing pixels and study a united perceptual and semantic token compression for all granular understanding and consequently facilitate open vocabulary semantic segmentation. Referring to the cognitive process of pretrained VLM where the low-level features are progressively composed to high-level semantics, we propose Feature Pyramid Tokenization (PAT) to cluster and represent multi-resolution feature by learnable codebooks and then decode them by joint learning pixel reconstruction and semantic segmentation. We design loosely coupled pixel and semantic learning branches. The pixel branch simulates bottom-up composition and top-down visualization of codebook tokens, while the semantic branch collectively fuse hierarchical codebooks as auxiliary segmentation guidance. Our experiments show that PAT enhances the semantic intuition of VLM feature pyramid, improves performance over the baseline segmentation model and achieves competitive performance on open vocabulary semantic segmentation benchmark. Our model is parameter-efficient for VLM integration and flexible for the independent tokenization. We hope to give inspiration not only on improving segmentation but also on semantic visual token utilization.

Paperid: 5247, https://arxiv.org/pdf/2412.13026.pdf

Abstract:
We present NAVCON, a large-scale annotated Vision-Language Navigation (VLN) corpus built on top of two popular datasets (R2R and RxR). The paper introduces four core, cognitively motivated and linguistically grounded, navigation concepts and an algorithm for generating large-scale silver annotations of naturally occurring linguistic realizations of these concepts in navigation instructions. We pair the annotated instructions with video clips of an agent acting on these instructions. NAVCON contains 236, 316 concept annotations for approximately 30, 0000 instructions and 2.7 million aligned images (from approximately 19, 000 instructions) showing what the agent sees when executing an instruction. To our knowledge, this is the first comprehensive resource of navigation concepts. We evaluated the quality of the silver annotations by conducting human evaluation studies on NAVCON samples. As further validation of the quality and usefulness of the resource, we trained a model for detecting navigation concepts and their linguistic realizations in unseen instructions. Additionally, we show that few-shot learning with GPT-4o performs well on this task using large-scale silver annotations of NAVCON.

Paperid: 5248, https://arxiv.org/pdf/2412.10029.pdf

Abstract:
Existing Vision-Language Pretraining (VLP) methods have achieved remarkable improvements across a variety of vision-language tasks, confirming their effectiveness in capturing coarse-grained semantic correlations. However, their capability for fine-grained understanding, which is critical for many nuanced vision-language applications, remains limited. Prevailing VLP models often overlook the intricate distinctions in expressing different modal features and typically depend on the similarity of holistic features for cross-modal interactions. Moreover, these models directly align and integrate features from different modalities, focusing more on coarse-grained general representations, thus failing to capture the nuanced differences necessary for tasks demanding a more detailed perception. In response to these limitations, we introduce Negative Augmented Samples(NAS), a refined vision-language pretraining model that innovatively incorporates NAS to specifically address the challenge of fine-grained understanding. NAS utilizes a Visual Dictionary(VD) as a semantic bridge between visual and linguistic domains. Additionally, it employs a Negative Visual Augmentation(NVA) method based on the VD to generate challenging negative image samples. These samples deviate from positive samples exclusively at the token level, thereby necessitating that the model discerns the subtle disparities between positive and negative samples with greater precision. Comprehensive experiments validate the efficacy of NAS components and underscore its potential to enhance fine-grained vision-language comprehension.

Paperid: 5249, https://arxiv.org/pdf/2412.08619.pdf

Abstract:
Physical reasoning, which involves interpreting object behaviors within dynamic environments, remains a significant challenge for Vision-Language Models (VLMs). The limitations in physical reasoning arise from an inability to translate learned knowledge into predictions about physical behavior. We perform a careful study to show how continual fine-tuning can mitigate this issue. However, fine-tuning is expensive for large models and impractical to repeatedly perform for every task. This necessitates the creation of modular and scalable ways to teach VLMs about physical reasoning. To that end, we introduce Physics Context Builders (PCBs), a novel modular framework where specialized VLMs are fine-tuned to generate detailed physical scene descriptions. These can be used as physical contexts for larger VLMs to enhance their reasoning capabilities. PCBs enable the separation of visual perception from reasoning, allowing us to analyze their relative contributions to physical understanding. We perform careful experiments on CLEVRER and on Falling Tower, a stability detection dataset with both simulated and real-world scenes, to demonstrate that PCBs provide substantial performance improvements, increasing average accuracy by up to 13.8% on complex physical reasoning tasks. Notably, PCBs show strong Sim2Real transfer, successfully generalizing from simulated training data to real-world scenes. Our work demonstrates that enhancing visual perception through modular, simulation-trained components offers a practical approach to improving physical reasoning in VLMs, while providing insights into the factors affecting physical understanding in these models.

Paperid: 5250, https://arxiv.org/pdf/2412.08443.pdf

Abstract:
Vision-language models have made significant strides recently, demonstrating superior performance across a range of tasks, e.g. optical character recognition and complex diagram analysis. Building on this trend, we introduce a new vision-language model, POINTS1.5, designed to excel in various real-world applications. POINTS1.5 is an enhancement of POINTS1.0 and incorporates several key innovations: i) We replace the original CLIP vision encoder, which had a fixed image resolution, with a NaViT-style vision encoder that supports native dynamic high resolution. This allows POINTS1.5 to process images of any resolution without needing to split them into tiles. ii) We add bilingual support to POINTS1.5, significantly enhancing its capability in Chinese. Due to the scarcity of open-source Chinese datasets for vision-language models, we collect numerous images from the Internet and annotate them using a combination of manual and automatic methods. iii) We propose a set of rigorous filtering methods for visual instruction tuning datasets. We comprehensively evaluate all these filtering methods, and choose the most effective ones to obtain the final visual instruction tuning set. Thanks to these innovations, POINTS1.5 significantly outperforms POINTS1.0 and demonstrates strong performance across a range of real-world applications. Notably, POINTS1.5-7B is trained on fewer than 4 billion tokens and ranks first on the OpenCompass leaderboard among models with fewer than 10 billion parameters

Paperid: 5251, https://arxiv.org/pdf/2412.08111.pdf

Abstract:
Vision-language models (VLMs), serve as foundation models for multi-modal applications such as image captioning and text-to-image generation. Recent studies have highlighted limitations in VLM text encoders, particularly in areas like compositionality and semantic understanding, though the underlying reasons for these limitations remain unclear. In this work, we aim to address this gap by analyzing the syntactic information, one of the fundamental linguistic properties, encoded by the text encoders of VLMs. We perform a thorough analysis comparing VLMs with different objective functions, parameter size and training data size, and with uni-modal language models (ULMs) in their ability to encode syntactic knowledge. Our findings suggest that ULM text encoders acquire syntactic information more effectively than those in VLMs. The syntactic information learned by VLM text encoders is shaped primarily by the pre-training objective, which plays a more crucial role than other factors such as model architecture, model size, or the volume of pre-training data. Models exhibit different layer-wise trends where CLIP performance dropped across layers while for other models, middle layers are rich in encoding syntactic knowledge.

Paperid: 5252, https://arxiv.org/pdf/2412.08110.pdf

Abstract:
Vision-Language Models (VLMs) implicitly learn to associate image regions with words from large-scale training data, demonstrating an emergent capability for grounding concepts without dense annotations[14,18,51]. However, the coarse-grained supervision from image-caption pairs is often insufficient to resolve ambiguities in object-concept correspondence, even with enormous data volume. Rich semantic and syntactic structures within the text modality have been overlooked as sources of supervision. Starting from contrastive architectures (BLIP and ALBEF) that show strong intrinsic grounding abilities, we propose HIerarchically STructured Learning (HIST). HIST enhances spatial vision-language alignment without using additional human annotations, by hierarchically decomposing captions into the constituent Subjects, Phrases, and Composite Phrases, and enforcing entailment relation between a parent and its children in the hierarchy. Specifically, we introduce two novel loss functions: (1) Subject Loss, which aligns image content with the subject of the corresponding phrase, acting as an entailment of standard contrastive/matching losses at the Phrase level; (2) Composition Loss, to balance attention across multiple objects. HIST is general, and can be applied to any VLM for which attention between vision and language can be computed. Compared to baseline VLMs, HIST achieves up to +9.8% improvement in visual grounding and +6.3% in multi-object referring segmentation. Surprisingly, the improved spatial grounding leads to improvements in other downstream VLM tasks: +1.1% in image-text retrieval, and +0.2% in visual question answering.

Paperid: 5253, https://arxiv.org/pdf/2412.04237.pdf

Abstract:
Large language models (LLMs) have proven effective for layout generation due to their ability to produce structure-description languages, such as HTML or JSON. In this paper, we argue that while LLMs can perform reasonably well in certain cases, their intrinsic limitation of not being able to perceive images restricts their effectiveness in tasks requiring visual content, e.g., content-aware layout generation. Therefore, we explore whether large vision-language models (LVLMs) can be applied to content-aware layout generation. To this end, inspired by the iterative revision and heuristic evaluation workflow of designers, we propose the training-free Visual-Aware Self-Correction LAyout GeneRation (VASCAR). VASCAR enables LVLMs (e.g., GPT-4o and Gemini) iteratively refine their outputs with reference to rendered layout images, which are visualized as colored bounding boxes on poster background (i.e., canvas). Extensive experiments and user study demonstrate VASCAR's effectiveness, achieving state-of-the-art (SOTA) layout generation quality. Furthermore, the generalizability of VASCAR across GPT-4o and Gemini demonstrates its versatility.

Paperid: 5254, https://arxiv.org/pdf/2411.17995.pdf

Abstract:
Multispectral pedestrian detection is a crucial component in various critical applications. However, a significant challenge arises due to the misalignment between these modalities, particularly under real-world conditions where data often appear heavily misaligned. Conventional methods developed on well-aligned or minimally misaligned datasets fail to address these discrepancies adequately. This paper introduces a new framework for multispectral pedestrian detection designed specifically to handle heavily misaligned datasets without the need for costly and complex traditional pre-processing calibration. By leveraging Large-scale Vision-Language Models (LVLM) for cross-modal semantic alignment, our approach seeks to enhance detection accuracy by aligning semantic information across the RGB and thermal domains. This method not only simplifies the operational requirements but also extends the practical usability of multispectral detection technologies in practical applications.

Paperid: 5255, https://arxiv.org/pdf/2411.16955.pdf

Abstract:
Recent advancements in artificial intelligence have sparked interest in scientific assistants that could support researchers across the full spectrum of scientific workflows, from literature review to experimental design and data analysis. A key capability for such systems is the ability to process and reason about scientific information in both visual and textual forms - from interpreting spectroscopic data to understanding laboratory setups. Here, we introduce MaCBench, a comprehensive benchmark for evaluating how vision-language models handle real-world chemistry and materials science tasks across three core aspects: data extraction, experimental understanding, and results interpretation. Through a systematic evaluation of leading models, we find that while these systems show promising capabilities in basic perception tasks - achieving near-perfect performance in equipment identification and standardized data extraction - they exhibit fundamental limitations in spatial reasoning, cross-modal information synthesis, and multi-step logical inference. Our insights have important implications beyond chemistry and materials science, suggesting that developing reliable multimodal AI scientific assistants may require advances in curating suitable training data and approaches to training those models.

Paperid: 5256, https://arxiv.org/pdf/2411.15720.pdf

Abstract:
Pre-trained vision-language models (VLMs) have showcased remarkable performance in image and natural language understanding, such as image captioning and response generation. As the practical applications of vision-language models become increasingly widespread, their potential safety and robustness issues raise concerns that adversaries may evade the system and cause these models to generate toxic content through malicious attacks. Therefore, evaluating the robustness of open-source VLMs against adversarial attacks has garnered growing attention, with transfer-based attacks as a representative black-box attacking strategy. However, most existing transfer-based attacks neglect the importance of the semantic correlations between vision and text modalities, leading to sub-optimal adversarial example generation and attack performance. To address this issue, we present Chain of Attack (CoA), which iteratively enhances the generation of adversarial examples based on the multi-modal semantic update using a series of intermediate attacking steps, achieving superior adversarial transferability and efficiency. A unified attack success rate computing method is further proposed for automatic evasion evaluation. Extensive experiments conducted under the most realistic and high-stakes scenario, demonstrate that our attacking strategy can effectively mislead models to generate targeted responses using only black-box attacks without any knowledge of the victim models. The comprehensive robustness evaluation in our paper provides insight into the vulnerabilities of VLMs and offers a reference for the safety considerations of future model developments.

Paperid: 5257, https://arxiv.org/pdf/2411.14832.pdf

Abstract:
The fast advancement of Large Vision-Language Models (LVLMs) has shown immense potential. These models are increasingly capable of tackling abstract visual tasks. Geometric structures, particularly graphs with their inherent flexibility and complexity, serve as an excellent benchmark for evaluating these models' predictive capabilities. While human observers can readily identify subtle visual details and perform accurate analyses, our investigation reveals that state-of-the-art LVLMs exhibit consistent limitations in specific visual graph scenarios, especially when confronted with stylistic variations. In response to these challenges, we introduce VisGraphVar (Visual Graph Variability), a customizable benchmark generator able to produce graph images for seven distinct task categories (detection, classification, segmentation, pattern recognition, link prediction, reasoning, matching), designed to systematically evaluate the strengths and limitations of individual LVLMs. We use VisGraphVar to produce 990 graph images and evaluate six LVLMs, employing two distinct prompting strategies, namely zero-shot and chain-of-thought. The findings demonstrate that variations in visual attributes of images (e.g., node labeling and layout) and the deliberate inclusion of visual imperfections, such as overlapping nodes, significantly affect model performance. This research emphasizes the importance of a comprehensive evaluation across graph-related tasks, extending beyond reasoning alone. VisGraphVar offers valuable insights to guide the development of more reliable and robust systems capable of performing advanced visual graph analysis.

Paperid: 5258, https://arxiv.org/pdf/2411.11919.pdf

Abstract:
Given the higher information load processed by large vision-language models (LVLMs) compared to single-modal LLMs, detecting LVLM hallucinations requires more human and time expense, and thus rise a wider safety concerns. In this paper, we introduce VL-Uncertainty, the first uncertainty-based framework for detecting hallucinations in LVLMs. Different from most existing methods that require ground-truth or pseudo annotations, VL-Uncertainty utilizes uncertainty as an intrinsic metric. We measure uncertainty by analyzing the prediction variance across semantically equivalent but perturbed prompts, including visual and textual data. When LVLMs are highly confident, they provide consistent responses to semantically equivalent queries. However, when uncertain, the responses of the target LVLM become more random. Considering semantically similar answers with different wordings, we cluster LVLM responses based on their semantic content and then calculate the cluster distribution entropy as the uncertainty measure to detect hallucination. Our extensive experiments on 10 LVLMs across four benchmarks, covering both free-form and multi-choice tasks, show that VL-Uncertainty significantly outperforms strong baseline methods in hallucination detection.

Paperid: 5259, https://arxiv.org/pdf/2411.11609.pdf

Abstract:
Following human instructions to explore and search for a specified target in an unfamiliar environment is a crucial skill for mobile service robots. Most of the previous works on object goal navigation have typically focused on a single input modality as the target, which may lead to limited consideration of language descriptions containing detailed attributes and spatial relationships. To address this limitation, we propose VLN-Game, a novel zero-shot framework for visual target navigation that can process object names and descriptive language targets effectively. To be more precise, our approach constructs a 3D object-centric spatial map by integrating pre-trained visual-language features with a 3D reconstruction of the physical environment. Then, the framework identifies the most promising areas to explore in search of potential target candidates. A game-theoretic vision language model is employed to determine which target best matches the given language description. Experiments conducted on the Habitat-Matterport 3D (HM3D) dataset demonstrate that the proposed framework achieves state-of-the-art performance in both object goal navigation and language-based navigation tasks. Moreover, we show that VLN-Game can be easily deployed on real-world robots. The success of VLN-Game highlights the promising potential of using game-theoretic methods with compact vision-language models to advance decision-making capabilities in robotic systems. The supplementary video and code can be accessed via the following link: https://sites.google.com/view/vln-game.

Paperid: 5260, https://arxiv.org/pdf/2411.10446.pdf

Abstract:
Recent advancements in vision-language models (VLMs) offer potential for robot task planning, but challenges remain due to VLMs' tendency to generate incorrect action sequences. To address these limitations, we propose VeriGraph, a novel framework that integrates VLMs for robotic planning while verifying action feasibility. VeriGraph employs scene graphs as an intermediate representation, capturing key objects and spatial relationships to improve plan verification and refinement. The system generates a scene graph from input images and uses it to iteratively check and correct action sequences generated by an LLM-based task planner, ensuring constraints are respected and actions are executable. Our approach significantly enhances task completion rates across diverse manipulation scenarios, outperforming baseline methods by 58% for language-based tasks and 30% for image-based tasks.

Paperid: 5261, https://arxiv.org/pdf/2411.00898.pdf

Abstract:
The conventional targeted adversarial attacks add a small perturbation to an image to make neural network models estimate the image as a predefined target class, even if it is not the correct target class. Recently, for visual-language models (VLMs), the focus of targeted adversarial attacks is to generate a perturbation that makes VLMs answer intended target text outputs. For example, they aim to make a small perturbation on an image to make VLMs' answers change from "there is an apple" to "there is a baseball." However, answering just intended text outputs is insufficient for tricky questions like "if there is a baseball, tell me what is below it." This is because the target of the adversarial attacks does not consider the overall integrity of the original image, thereby leading to a lack of visual reasoning. In this work, we focus on generating targeted adversarial examples with visual reasoning against VLMs. To this end, we propose 1) a novel adversarial attack procedure -- namely, Replace-then-Perturb and 2) a contrastive learning-based adversarial loss -- namely, Contrastive-Adv. In Replace-then-Perturb, we first leverage a text-guided segmentation model to find the target object in the image. Then, we get rid of the target object and inpaint the empty space with the desired prompt. By doing this, we can generate a target image corresponding to the desired prompt, while maintaining the overall integrity of the original image. Furthermore, in Contrastive-Adv, we design a novel loss function to obtain better adversarial examples. Our extensive benchmark results demonstrate that Replace-then-Perturb and Contrastive-Adv outperform the baseline adversarial attack algorithms. We note that the source code to reproduce the results will be available.

Paperid: 5262, https://arxiv.org/pdf/2411.00508.pdf

Abstract:
Teaching robots desired skills in real-world environments remains challenging, especially for non-experts. A key bottleneck is that collecting robotic data often requires expertise or specialized hardware, limiting accessibility and scalability. We posit that natural language offers an intuitive and accessible interface for robot learning. To this end, we study two aspects: (1) enabling non-experts to collect robotic data through natural language supervision (e.g., "move the arm to the right") and (2) training robot policies directly from this supervision. Specifically, we introduce a data collection framework that collects robot demonstrations based on natural language supervision and further augments these demonstrations. We then present CLIP-RT, a new vision-language-action (VLA) model that learns language-conditioned visuomotor policies from this supervision. CLIP-RT adapts the pretrained CLIP model and learns to predict language-based motion primitives via contrastive imitation learning. We train CLIP-RT on the Open X-Embodiment dataset and finetune it on in-domain data collected by our framework. In real-world evaluations, CLIP-RT demonstrates strong capabilities in learning novel manipulation skills, outperforming OpenVLA (7B parameters) by 24% in average success rates, while using 7x fewer parameters (1B). We further assess CLIP-RT's capabilities in few-shot generalization and collaborative scenarios involving large pretrained models or humans. In simulated environments, CLIP-RT also yields strong performance, achieving a 93.1% average success rate on the LIBERO benchmark with an inference throughput of 163 Hz.

Paperid: 5263, https://arxiv.org/pdf/2410.21318.pdf

Abstract:
Text-based person retrieval aims to identify the specific persons using textual descriptions as queries. Existing ad vanced methods typically depend on vision-language pre trained (VLP) models to facilitate effective cross-modal alignment. However, the inherent constraints of VLP mod-els, which include the global alignment biases and insuffi-cient self-feedback regulation, impede optimal retrieval per formance. In this paper, we propose MeFa, a Multi-Pathway Exploration, Feedback, and Adjustment framework, which deeply explores intrinsic feedback of intra and inter-modal to make targeted adjustment, thereby achieving more precise person-text associations. Specifically, we first design an intra modal reasoning pathway that generates hard negative sam ples for cross-modal data, leveraging feedback from these samples to refine intra-modal reasoning, thereby enhancing sensitivity to subtle discrepancies. Subsequently, we intro duce a cross-modal refinement pathway that utilizes both global information and intermodal feedback to refine local in formation, thus enhancing its global semantic representation. Finally, the discriminative clue correction pathway incorpo rates fine-grained features of secondary similarity as discrim inative clues to further mitigate retrieval failures caused by disparities in these features. Experimental results on three public benchmarks demonstrate that MeFa achieves superior person retrieval performance without necessitating additional data or complex structures.

Paperid: 5264, https://arxiv.org/pdf/2410.15863.pdf

Abstract:
Vision Language Models (VLMs) play a crucial role in robotic manipulation by enabling robots to understand and interpret the visual properties of objects and their surroundings, allowing them to perform manipulation based on this multimodal understanding. Accurately understanding spatial relationships remains a non-trivial challenge, yet it is essential for effective robotic manipulation. In this work, we introduce a novel framework that integrates VLMs with a structured spatial reasoning pipeline to perform object manipulation based on high-level, task-oriented input. Our approach is the transformation of visual scenes into tree-structured representations that encode the spatial relations. These trees are subsequently processed by a Large Language Model (LLM) to infer restructured configurations that determine how these objects should be organised for a given high-level task. To support our framework, we also present a new dataset containing manually annotated captions that describe spatial relations among objects, along with object-level attribute annotations such as fragility, mass, material, and transparency. We demonstrate that our method not only improves the comprehension of spatial relationships among objects in the visual environment but also enables robots to interact with these objects more effectively. As a result, this approach significantly enhances spatial reasoning in robotic manipulation tasks. To our knowledge, this is the first method of its kind in the literature, offering a novel solution that allows robots to more efficiently organize and utilize objects in their surroundings.

Paperid: 5265, https://arxiv.org/pdf/2410.15657.pdf

Abstract:
Human-object interaction (HOI) detection has seen advancements with Vision Language Models (VLMs), but these methods often depend on extensive manual annotations. Vision Large Language Models (VLLMs) can inherently recognize and reason about interactions at the image level but are computationally heavy and not designed for instance-level HOI detection. To overcome these limitations, we propose a Cross-Level HOI distillation (CL-HOI) framework, which distills instance-level HOIs from VLLMs image-level understanding without the need for manual annotations. Our approach involves two stages: context distillation, where a Visual Linguistic Translator (VLT) converts visual information into linguistic form, and interaction distillation, where an Interaction Cognition Network (ICN) reasons about spatial, visual, and context relations. We design contrastive distillation losses to transfer image-level context and interaction knowledge from the teacher to the student model, enabling instance-level HOI detection. Evaluations on HICO-DET and V-COCO datasets demonstrate that our CL-HOI surpasses existing weakly supervised methods and VLLM supervised methods, showing its efficacy in detecting HOIs without manual labels.

Paperid: 5266, https://arxiv.org/pdf/2410.13030.pdf

Abstract:
Despite the significant influx of prompt-tuning techniques for generative vision-language models (VLMs), it remains unclear how sensitive these models are to lexical and semantic alterations in prompts. In this paper, we evaluate the ability of generative VLMs to understand lexical and semantic changes in text using the SugarCrepe++ dataset. We analyze the sensitivity of VLMs to lexical alterations in prompts without corresponding semantic changes. Our findings demonstrate that generative VLMs are highly sensitive to such alterations. Additionally, we show that this vulnerability affects the performance of techniques aimed at achieving consistency in their outputs.

Paperid: 5267, https://arxiv.org/pdf/2410.09047.pdf

Abstract:
The safety alignment ability of Vision-Language Models (VLMs) is prone to be degraded by the integration of the vision module compared to its LLM backbone. We investigate this phenomenon, dubbed as ''safety alignment degradation'' in this paper, and show that the challenge arises from the representation gap that emerges when introducing vision modality to VLMs. In particular, we show that the representations of multi-modal inputs shift away from that of text-only inputs which represent the distribution that the LLM backbone is optimized for. At the same time, the safety alignment capabilities, initially developed within the textual embedding space, do not successfully transfer to this new multi-modal representation space. To reduce safety alignment degradation, we introduce Cross-Modality Representation Manipulation (CMRM), an inference time representation intervention method for recovering the safety alignment ability that is inherent in the LLM backbone of VLMs, while simultaneously preserving the functional capabilities of VLMs. The empirical results show that our framework significantly recovers the alignment ability that is inherited from the LLM backbone with minimal impact on the fluency and linguistic capabilities of pre-trained VLMs even without additional training. Specifically, the unsafe rate of LLaVA-7B on multi-modal input can be reduced from 61.53% to as low as 3.15% with only inference-time intervention. WARNING: This paper contains examples of toxic or harmful language.

Paperid: 5268, https://arxiv.org/pdf/2410.06912.pdf

Abstract:
Image-text representation learning forms a cornerstone in vision-language models, where pairs of images and textual descriptions are contrastively aligned in a shared embedding space. Since visual and textual concepts are naturally hierarchical, recent work has shown that hyperbolic space can serve as a high-potential manifold to learn vision-language representation with strong downstream performance. In this work, for the first time we show how to fully leverage the innate hierarchical nature of hyperbolic embeddings by looking beyond individual image-text pairs. We propose Compositional Entailment Learning for hyperbolic vision-language models. The idea is that an image is not only described by a sentence but is itself a composition of multiple object boxes, each with their own textual description. Such information can be obtained freely by extracting nouns from sentences and using openly available localized grounding models. We show how to hierarchically organize images, image boxes, and their textual descriptions through contrastive and entailment-based objectives. Empirical evaluation on a hyperbolic vision-language model trained with millions of image-text pairs shows that the proposed compositional learning approach outperforms conventional Euclidean CLIP learning, as well as recent hyperbolic alternatives, with better zero-shot and retrieval generalization and clearly stronger hierarchical performance.

Paperid: 5269, https://arxiv.org/pdf/2410.01180.pdf

Abstract:
Localizing unusual activities, such as human errors or surveillance incidents, in videos holds practical significance. However, current video understanding models struggle with localizing these unusual events likely because of their insufficient representation in models' pretraining datasets. To explore foundation models' capability in localizing unusual activity, we introduce UAL-Bench, a comprehensive benchmark for unusual activity localization, featuring three video datasets: UAG-OOPS, UAG-SSBD, UAG-FunQA, and an instruction-tune dataset: OOPS-UAG-Instruct, to improve model capabilities. UAL-Bench evaluates three approaches: Video-Language Models (Vid-LLMs), instruction-tuned Vid-LLMs, and a novel integration of Vision-Language Models and Large Language Models (VLM-LLM). Our results show the VLM-LLM approach excels in localizing short-span unusual events and predicting their onset (start time) more accurately than Vid-LLMs. We also propose a new metric, R@1, TD <= p, to address limitations in existing evaluation methods. Our findings highlight the challenges posed by long-duration videos, particularly in autism diagnosis scenarios, and the need for further advancements in localization techniques. Our work not only provides a benchmark for unusual activity localization but also outlines the key challenges for existing foundation models, suggesting future research directions on this important task.

Paperid: 5270, https://arxiv.org/pdf/2410.00388.pdf

Abstract:
The Multi-Object Search (MOS) problem involves navigating to a sequence of locations to maximize the likelihood of finding target objects while minimizing travel costs. In this paper, we introduce a novel approach to the MOS problem, called Finder, which leverages vision language models (VLMs) to locate multiple objects across diverse environments. Specifically, our approach introduces multi-channel score maps to track and reason about multiple objects simultaneously during navigation, along with a score map technique that combines scene-level and object-level semantic correlations. Experiments in both simulated and real-world settings showed that Finder outperforms existing methods using deep reinforcement learning and VLMs. Ablation and scalability studies further validated our design choices and robustness with increasing numbers of target objects, respectively. Website: https://find-all-my-things.github.io/

Paperid: 5271, https://arxiv.org/pdf/2409.19610.pdf

Abstract:
Integrating pretrained vision-language foundation models like CLIP into federated learning has attracted significant attention for enhancing generalization across diverse tasks. Typically, federated learning of vision-language models employs prompt learning to reduce communication and computational costs, i.e., prompt-based federated learning. However, there is limited theoretical analysis to understand the performance of prompt-based federated learning. In this work, we construct a theoretical analysis framework for prompt-based federated learning via feature learning theory. Specifically, we monitor the evolution of signal learning and noise memorization in prompt-based federated learning, demonstrating that performance can be assessed by the ratio of task-relevant to task-irrelevant coefficients. Furthermore, we draw an analogy between income and risk in portfolio optimization and the task-relevant and task-irrelevant terms in feature learning. Leveraging inspiration from portfolio optimization that combining two independent assets will maintain the income while reducing the risk, we introduce two prompts: global prompt and local prompt to construct a prompt portfolio to balance the generalization and personalization. Consequently, we showed the performance advantage of the prompt portfolio and derived the optimal mixing coefficient. These theoretical claims have been further supported by empirical experiments.

Paperid: 5272, https://arxiv.org/pdf/2409.19425.pdf

Abstract:
Recent contrastive multimodal vision-language models like CLIP have demonstrated robust open-world semantic understanding, becoming the standard image backbones for vision-language applications. However, recent findings suggest high semantic similarity between well-trained unimodal encoders, which raises a key question: Is there a plausible way to connect unimodal backbones for vision-language tasks? To this end, we propose a novel framework that aligns vision and language using frozen unimodal encoders. It involves selecting semantically similar encoders in the latent space, curating a concept-rich dataset of image-caption pairs, and training simple MLP projectors. We evaluated our approach on 12 zero-shot classification datasets and 2 image-text retrieval datasets. Our best model, utilizing DINOv2 and All-Roberta-Large text encoder, achieves 76$\%$ accuracy on ImageNet with a 20-fold reduction in data and 65-fold reduction in compute requirements compared multi-modal alignment where models are trained from scratch. The proposed framework enhances the accessibility of multimodal model development while enabling flexible adaptation across diverse scenarios. Code and curated datasets are available at \texttt{github.com/mayug/freeze-align}.

Paperid: 5273, https://arxiv.org/pdf/2409.11656.pdf

Abstract:
Text recognition is an inherent integration of vision and language, encompassing the visual texture in stroke patterns and the semantic context among the character sequences. Towards advanced text recognition, there are three key challenges: (1) an encoder capable of representing the visual and semantic distributions; (2) a decoder that ensures the alignment between vision and semantics; and (3) consistency in the framework during pre-training, if it exists, and fine-tuning. Inspired by masked autoencoding, a successful pre-training strategy in both vision and language, we propose an innovative scene text recognition approach, named VL-Reader. The novelty of the VL-Reader lies in the pervasive interplay between vision and language throughout the entire process. Concretely, we first introduce a Masked Visual-Linguistic Reconstruction (MVLR) objective, which aims at simultaneously modeling visual and linguistic information. Then, we design a Masked Visual-Linguistic Decoder (MVLD) to further leverage masked vision-language context and achieve bi-modal feature interaction. The architecture of VL-Reader maintains consistency from pre-training to fine-tuning. In the pre-training stage, VL-Reader reconstructs both masked visual and text tokens, while in the fine-tuning stage, the network degrades to reconstruct all characters from an image without any masked regions. VL-reader achieves an average accuracy of 97.1% on six typical datasets, surpassing the SOTA by 1.1%. The improvement was even more significant on challenging datasets. The results demonstrate that vision and language reconstructor can serve as an effective scene text recognizer.

Paperid: 5274, https://arxiv.org/pdf/2409.08251.pdf

Abstract:
Panoptic narrative grounding (PNG), whose core target is fine-grained image-text alignment, requires a panoptic segmentation of referred objects given a narrative caption. Previous discriminative methods achieve only weak or coarse-grained alignment by panoptic segmentation pretraining or CLIP model adaptation. Given the recent progress of text-to-image Diffusion models, several works have shown their capability to achieve fine-grained image-text alignment through cross-attention maps and improved general segmentation performance. However, the direct use of phrase features as static prompts to apply frozen Diffusion models to the PNG task still suffers from a large task gap and insufficient vision-language interaction, yielding inferior performance. Therefore, we propose an Extractive-Injective Phrase Adapter (EIPA) bypass within the Diffusion UNet to dynamically update phrase prompts with image features and inject the multimodal cues back, which leverages the fine-grained image-text alignment capability of Diffusion models more sufficiently. In addition, we also design a Multi-Level Mutual Aggregation (MLMA) module to reciprocally fuse multi-level image and phrase features for segmentation refinement. Extensive experiments on the PNG benchmark show that our method achieves new state-of-the-art performance.

Paperid: 5275, https://arxiv.org/pdf/2409.06166.pdf

Abstract:
Prompt learning is an effective method to customize Vision-Language Models (VLMs) for various downstream tasks, involving tuning very few parameters of input prompt tokens. Recently, prompt pretraining in large-scale dataset (e.g., ImageNet-21K) has played a crucial role in prompt learning for universal visual discrimination. However, we revisit and observe that the limited learnable prompts could face underfitting risks given the extensive images during prompt pretraining, simultaneously leading to poor generalization. To address the above issues, in this paper, we propose a general framework termed Revisiting Prompt Pretraining (RPP), which targets at improving the fitting and generalization ability from two aspects: prompt structure and prompt supervision. For prompt structure, we break the restriction in common practice where query, key, and value vectors are derived from the shared learnable prompt token. Instead, we introduce unshared individual query, key, and value learnable prompts, thereby enhancing the model's fitting capacity through increased parameter diversity. For prompt supervision, we additionally utilize soft labels derived from zero-shot probability predictions provided by a pretrained Contrastive Language Image Pretraining (CLIP) teacher model. These soft labels yield more nuanced and general insights into the inter-class relationships, thereby endowing the pretraining process with better generalization ability. RPP produces a more resilient prompt initialization, enhancing its robust transferability across diverse visual recognition tasks. Experiments across various benchmarks consistently confirm the state-of-the-art (SOTA) performance of our pretrained prompts. Codes and models will be made available soon.

Paperid: 5276, https://arxiv.org/pdf/2409.06082.pdf

Abstract:
Providing asynchronous feedback is a critical step in the 3D design workflow. A common approach to providing feedback is to pair textual comments with companion reference images, which helps illustrate the gist of text. Ideally, feedback providers should possess 3D and image editing skills to create reference images that can effectively describe what they have in mind. However, they often lack such skills, so they have to resort to sketches or online images which might not match well with the current 3D design. To address this, we introduce MemoVis, a text editor interface that assists feedback providers in creating reference images with generative AI driven by the feedback comments. First, a novel real-time viewpoint suggestion feature, based on a vision-language foundation model, helps feedback providers anchor a comment with a camera viewpoint. Second, given a camera viewpoint, we introduce three types of image modifiers, based on pre-trained 2D generative models, to turn a text comment into an updated version of the 3D scene from that viewpoint. We conducted a within-subjects study with feedback providers, demonstrating the effectiveness of MemoVis. The quality and explicitness of the companion images were evaluated by another eight participants with prior 3D design experience.

Paperid: 5277, https://arxiv.org/pdf/2409.04828.pdf

Abstract:
In recent years, vision-language models have made significant strides, excelling in tasks like optical character recognition and geometric problem-solving. However, several critical issues remain: 1) Proprietary models often lack transparency about their architectures, while open-source models need more detailed ablations of their training strategies. 2) Pre-training data in open-source works is under-explored, with datasets added empirically, making the process cumbersome. 3) Fine-tuning often focuses on adding datasets, leading to diminishing returns. To address these issues, we propose the following contributions: 1) We trained a robust baseline model using the latest advancements in vision-language models, introducing effective improvements and conducting comprehensive ablation and validation for each technique. 2) Inspired by recent work on large language models, we filtered pre-training data using perplexity, selecting the lowest perplexity data for training. This approach allowed us to train on a curated 1M dataset, achieving competitive performance. 3) During visual instruction tuning, we used model soup on different datasets when adding more datasets yielded marginal improvements. These innovations resulted in a 9B parameter model that performs competitively with state-of-the-art models. Our strategies are efficient and lightweight, making them easily adoptable by the community.

Paperid: 5278, https://arxiv.org/pdf/2409.04053.pdf

Abstract:
While visual question-answering (VQA) benchmarks have catalyzed the development of reasoning techniques, they have focused on vertical thinking. Effective problem-solving also necessitates lateral thinking, which remains understudied in AI and has not been used to test visual perception systems. To bridge this gap, we formulate visual lateral thinking as a multiple-choice question-answering task and describe a three-step taxonomy-driven methodology for instantiating task examples. Then, we develop COLUMBUS, a synthetic benchmark that applies the task pipeline to create QA sets with text and icon rebus puzzles based on publicly available collections of compounds and common phrases. COLUMBUS comprises over 1,000 puzzles, each with four answer candidates. While the SotA vision-language models (VLMs) achieve decent performance, our evaluation demonstrates a substantial gap between humans and models. VLMs benefit from human-curated descriptions but struggle to self-generate such representations at the right level of abstraction.

Paperid: 5279, https://arxiv.org/pdf/2601.20064.pdf

Abstract:
Open-vocabulary semantic segmentation aims to assign labels to every pixel in an image based on text labels. Existing approaches typically utilize vision-language models (VLMs), such as CLIP, for dense prediction. However, VLMs, pre-trained on image-text pairs, are biased toward salient, object-centric regions and exhibit two critical limitations when adapted to segmentation: (i) Foreground Bias, which tends to ignore background regions, and (ii) Limited Spatial Localization, resulting in blurred object boundaries. To address these limitations, we introduce DiSa, a novel saliency-aware foreground-background disentangled framework. By explicitly incorporating saliency cues in our designed Saliency-aware Disentanglement Module (SDM), DiSa separately models foreground and background ensemble features in a divide-and-conquer manner. Additionally, we propose a Hierarchical Refinement Module (HRM) that leverages pixel-wise spatial contexts and enables channel-wise feature refinement through multi-level updates. Extensive experiments on six benchmarks demonstrate that DiSa consistently outperforms state-of-the-art methods.

Paperid: 5280, https://arxiv.org/pdf/2601.19222.pdf

Abstract:
Multimodal Large Language Models (MLLMs) show promise for general industrial quality inspection, but fall short in complex scenarios, such as Printed Circuit Board (PCB) inspection. PCB inspection poses unique challenges due to densely packed components, complex wiring structures, and subtle defect patterns that require specialized domain expertise. However, a high-quality, unified vision-language benchmark for quantitatively evaluating MLLMs across PCB inspection tasks remains absent, stemming not only from limited data availability but also from fragmented datasets and inconsistent standardization. To fill this gap, we propose UniPCB, the first unified vision-language benchmark for open-ended PCB quality inspection. UniPCB is built via a systematic pipeline that curates and standardizes data from disparate sources across three annotated scenarios. Furthermore, we introduce PCB-GPT, an MLLM trained on a new instruction dataset generated by this pipeline, utilizing a novel progressive curriculum that mimics the learning process of human experts. Evaluations on the UniPCB benchmark show that while existing MLLMs falter on domain-specific tasks, PCB-GPT establishes a new baseline. Notably, it more than doubles the performance on fine-grained defect localization compared to the strongest competitors, with significant advantages in localization and analysis. We will release the instruction data, benchmark, and model to facilitate future research.

Paperid: 5281, https://arxiv.org/pdf/2601.18242.pdf

Abstract:
Accurate radio-frequency (RF) material parameters are essential for electromagnetic digital twins in 6G systems, yet gradient-based inverse ray tracing (RT) remains sensitive to initialization and costly under limited measurements. This paper proposes a vision-language-model (VLM) guided framework that accelerates and stabilizes multi-material parameter estimation in a differentiable RT (DRT) engine. A VLM parses scene images to infer material categories and maps them to quantitative priors via an ITU-R material table, yielding informed conductivity initializations. The VLM further selects informative transmitter/receiver placements that promote diverse, material-discriminative paths. Starting from these priors, the DRT performs gradient-based refinement using measured received signal strengths. Experiments in NVIDIA Sionna on indoor scenes show 2-4$\times$ faster convergence and 10-100$\times$ lower final parameter error compared with uniform or random initialization and random placement baselines, achieving sub-0.1\% mean relative error with only a few receivers. Complexity analyses indicate per-iteration time scales near-linearly with the number of materials and measurement setups, while VLM-guided placement reduces the measurements required for accurate recovery. Ablations over RT depth and ray counts confirm further accuracy gains without significant per-iteration overhead. Results demonstrate that semantic priors from VLMs effectively guide physics-based optimization for fast and reliable RF material estimation.

Paperid: 5282, https://arxiv.org/pdf/2601.14714.pdf

Abstract:
Multimodal retrieval systems typically employ Vision Language Models (VLMs) that encode images and text independently into vectors within a shared embedding space. Despite incorporating text encoders, VLMs consistently underperform specialized text models on text-only retrieval tasks. Moreover, introducing additional text encoders increases storage, inference overhead, and exacerbates retrieval inefficiencies, especially in multilingual settings. To address these limitations, we propose a multi-task learning framework that unifies the feature representation across images, long and short texts, and intent-rich queries. To our knowledge, this is the first work to jointly optimize multilingual image retrieval, text retrieval, and natural language understanding (NLU) tasks within a single framework. Our approach integrates image and text retrieval with a shared text encoder that is enhanced by NLU features for intent understanding and retrieval accuracy.

Paperid: 5283, https://arxiv.org/pdf/2601.10945.pdf

Abstract:
Traditionally, AI research in medical diagnosis has largely centered on image analysis. While this has led to notable advancements, the absence of patient-reported symptoms continues to hinder diagnostic accuracy. To address this, we propose a Pre-Consultation Dialogue Framework (PCDF) that mimics real-world diagnostic procedures, where doctors iteratively query patients before reaching a conclusion. Specifically, we simulate diagnostic dialogues between two vision-language models (VLMs): a DocVLM, which generates follow-up questions based on the image and dialogue history, and a PatientVLM, which responds using a symptom profile derived from the ground-truth diagnosis. We additionally conducted a small-scale clinical validation of the synthetic symptoms generated by our framework, with licensed clinicians confirming their clinical relevance, symptom coverage, and overall realism. These findings indicate that the resulting DocVLM-PatientVLM interactions form coherent, multi-turn consultations paired with images and diagnoses, which we then use to fine-tune the DocVLM. This dialogue-based supervision leads to substantial gains over image-only training, highlighting the value of realistic symptom elicitation for diagnosis.

Paperid: 5284, https://arxiv.org/pdf/2601.07553.pdf

Abstract:
As large language models (LLMs) continue to improve in reasoning and decision-making, there is a growing need for realistic and interactive environments where their abilities can be rigorously evaluated. We present VirtualEnv, a next-generation simulation platform built on Unreal Engine 5 that enables fine-grained benchmarking of LLMs in embodied and interactive scenarios. VirtualEnv supports rich agent-environment interactions, including object manipulation, navigation, and adaptive multi-agent collaboration, as well as game-inspired mechanics like escape rooms and procedurally generated environments. We provide a user-friendly API built on top of Unreal Engine, allowing researchers to deploy and control LLM-driven agents using natural language instructions. We integrate large-scale LLMs and vision-language models (VLMs), such as GPT-based models, to generate novel environments and structured tasks from multimodal inputs. Our experiments benchmark the performance of several popular LLMs across tasks of increasing complexity, analyzing differences in adaptability, planning, and multi-agent coordination. We also describe our methodology for procedural task generation, task validation, and real-time environment control. VirtualEnv is released as an open-source platform, we aim to advance research at the intersection of AI and gaming, enable standardized evaluation of LLMs in embodied AI settings, and pave the way for future developments in immersive simulations and interactive entertainment.

Paperid: 5285, https://arxiv.org/pdf/2601.05143.pdf

Abstract:
Visual question answering for crop disease analysis requires accurate visual understanding and reliable language generation. This work presents a lightweight vision-language framework for crop and disease identification from leaf images. The proposed approach combines a Swin Transformer vision encoder with sequence-to-sequence language decoders. A two-stage training strategy is adopted to improve visual representation learning and cross-modal alignment. The model is evaluated on a large-scale crop disease dataset using classification and natural language generation metrics. Experimental results show high accuracy for both crop and disease identification. The framework also achieves strong performance on BLEU, ROUGE and BERTScore. Our proposed models outperform large-scale vision-language baselines while using significantly fewer parameters. Explainability is assessed using Grad-CAM and token-level attribution. Qualitative results demonstrate robust performance under diverse user-driven queries. These findings highlight the effectiveness of task-specific visual pretraining for crop disease visual question answering.

Paperid: 5286, https://arxiv.org/pdf/2601.03400.pdf

Abstract:
Vision-Language Models (VLMs) have achieved strong performance on standard vision-language benchmarks, yet often rely on surface-level recognition rather than deeper reasoning. We propose visual word puzzles as a challenging alternative, as they require discovering implicit visual cues, generating and revising hypotheses, and mapping perceptual evidence to non-literal concepts in ways that are difficult to solve via literal grounding, OCR-heavy shortcuts, or simple retrieval-style matching. We introduce Eye-Q, a multilingual benchmark designed to assess this form of complex visual understanding. Eye-Q contains 1,343 puzzles in which a model observes a conceptually dense scene with a brief description and must infer a specific target word or phrase. The puzzles are intentionally unstructured and cue-implicit, with distractors and contextual relationships that demand selective attention, abstraction, and associative inference. The benchmark spans English, Persian, Arabic, and cross-lingual puzzles. We evaluate state-of-the-art VLMs using an open-ended, human-aligned protocol that probes hypothesis formation and revision under lightweight assistance. Results reveal substantial performance gaps, especially on abstract and cross-lingual puzzles, highlighting limitations in current models' ability to construct and search over appropriate conceptual representations for flexible image-to-phrase inference; maximum accuracy reaches only 60.27%.

Paperid: 5287, https://arxiv.org/pdf/2512.16201.pdf

Abstract:
Radiology Report Generation (RRG) is a critical step toward automating healthcare workflows, facilitating accurate patient assessments, and reducing the workload of medical professionals. Despite recent progress in Large Medical Vision-Language Models (Med-VLMs), generating radiology reports that are both visually grounded and clinically accurate remains a significant challenge. Existing approaches often rely on large labeled corpora for pre-training, costly task-specific preference data, or retrieval-based methods. However, these strategies do not adequately mitigate hallucinations arising from poor cross-modal alignment between visual and linguistic representations. To address these limitations, we propose VALOR:Visual Alignment of Medical Vision-Language Models for GrOunded Radiology Report Generation. Our method introduces a reinforcement learning-based post-alignment framework utilizing Group-Relative Proximal Optimization (GRPO). The training proceeds in two stages: (1) improving the Med-VLM with textual rewards to encourage clinically precise terminology, and (2) aligning the vision projection module of the textually grounded model with disease findings, thereby guiding attention toward image re gions most relevant to the diagnostic task. Extensive experiments on multiple benchmarks demonstrate that VALOR substantially improves factual accuracy and visual grounding, achieving significant performance gains over state-of-the-art report generation methods.

Paperid: 5288, https://arxiv.org/pdf/2512.16077.pdf

Abstract:
Open-vocabulary 3D object detection methods are able to localize 3D boxes of classes unseen during training. Despite the name, existing methods rely on user-specified classes both at training and inference. We propose to study Auto-Vocabulary 3D Object Detection (AV3DOD), where the classes are automatically generated for the detected objects without any user input. To this end, we introduce Semantic Score (SS) to evaluate the quality of the generated class names. We then develop a novel framework, AV3DOD, which leverages 2D vision-language models (VLMs) to generate rich semantic candidates through image captioning, pseudo 3D box generation, and feature-space semantics expansion. AV3DOD achieves the state-of-the-art (SOTA) performance on both localization (mAP) and semantic quality (SS) on the ScanNetV2 and SUNRGB-D datasets. Notably, it surpasses the SOTA, CoDA, by 3.48 overall mAP and attains a 24.5% relative improvement in SS on ScanNetV2.

Paperid: 5289, https://arxiv.org/pdf/2512.15957.pdf

Abstract:
Accurately predicting human behaviors is crucial for mobile robots operating in human-populated environments. While prior research primarily focuses on predicting actions in single-human scenarios from an egocentric view, several robotic applications require understanding multiple human behaviors from a third-person perspective. To this end, we present CAMP-VLM (Context-Aware Multi-human behavior Prediction): a Vision Language Model (VLM)-based framework that incorporates contextual features from visual input and spatial awareness from scene graphs to enhance prediction of humans-scene interactions. Due to the lack of suitable datasets for multi-human behavior prediction from an observer view, we perform fine-tuning of CAMP-VLM with synthetic human behavior data generated by a photorealistic simulator, and evaluate the resulting models on both synthetic and real-world sequences to assess their generalization capabilities. Leveraging Supervised Fine-Tuning (SFT) and Direct Preference Optimization (DPO), CAMP-VLM outperforms the best-performing baseline by up to 66.9% in prediction accuracy.

Paperid: 5290, https://arxiv.org/pdf/2512.12268.pdf

Abstract:
Vision-language models (VLMs) such as CLIP exhibit strong zero-shot generalization but remain sensitive to domain shifts at test time. Test-time prompt tuning (TPT) mitigates this issue by adapting prompts with fixed augmentations, which may falter in more challenging settings. In this work, we propose Meta Test-Time Prompt Tuning (MetaTPT), a meta-learning framework that learns a self-supervised auxiliary task to guide test-time prompt tuning. The auxiliary task dynamically learns parameterized augmentations for each sample, enabling more expressive transformations that capture essential features in target domains. MetaTPT adopts a dual-loop optimization paradigm: an inner loop learns a self-supervised task that generates informative views, while the outer loop performs prompt tuning by enforcing consistency across these views. By coupling augmentation learning with prompt tuning, MetaTPT improves test-time adaptation under domain shifts. Extensive experiments demonstrate that MetaTPT achieves state-of-the-art performance on domain generalization and cross-dataset benchmarks.

Paperid: 5291, https://arxiv.org/pdf/2512.06845.pdf

Abstract:
Deploying video anomaly detection in practice is hampered by the scarcity and collection cost of real abnormal footage. We address this by training without any real abnormal videos while evaluating under the standard weakly supervised split, and we introduce PA-VAD, a generation-driven approach that learns a detector from synthesized pseudo-abnormal videos paired with real normal videos, using only a small set of real normal images to drive synthesis. For synthesis, we select class-relevant initial images with CLIP and refine textual prompts with a vision-language model to improve fidelity and scene consistency before invoking a video diffusion model. For training, we mitigate excessive spatiotemporal magnitude in synthesized anomalies by an domain-aligned regularized module that combines domain alignment and memory usage-aware updates. Extensive experiments show that our approach reaches 98.2% on ShanghaiTech and 82.5% on UCF-Crime, surpassing the strongest real-abnormal method on ShanghaiTech by +0.6% and outperforming the UVAD state-of-the-art on UCF-Crime by +1.9%. The results demonstrate that high-accuracy anomaly detection can be obtained without collecting real anomalies, providing a practical path toward scalable deployment.

Paperid: 5292, https://arxiv.org/pdf/2512.05943.pdf

Abstract:
Reliable mathematical and scientific reasoning remains an open challenge for large vision-language models. Standard final-answer evaluation often masks reasoning errors, allowing silent failures to persist. To address this gap, we introduce TRACE, a framework for Transparent Reasoning And Consistency Evaluation that diagnoses reasoning trajectories rather than only end results. At its core, TRACE leverages Auxiliary Reasoning Sets, compact sub question answer pairs that decompose complex problems, evaluate intermediate steps through consistency-based metrics, and expose failures overlooked by standard evaluation. Our experiments show that consistency across ARS correlates with final-answer correctness and helps pinpoint the reasoning steps where failures arise, offering actionable signals for model improvement. Furthermore, TRACE defines confidence regions that distinguish reliable from unreliable reasoning paths, supporting effective filtering, debugging, and model refinement.

Paperid: 5293, https://arxiv.org/pdf/2512.05930.pdf

Abstract:
Evaluating vision-language models (VLMs) in scientific domains like mathematics and physics poses unique challenges that go far beyond predicting final answers. These domains demand conceptual understanding, symbolic reasoning, and adherence to formal laws, requirements that most existing benchmarks fail to address. In particular, current datasets tend to be static, lacking intermediate reasoning steps, robustness to variations, or mechanisms for verifying scientific correctness. To address these limitations, we introduce PRiSM, a synthetic, fully dynamic, and multimodal benchmark for evaluating scientific reasoning via grounded Python code. PRiSM includes over 24,750 university-level physics and math problems, and it leverages our scalable agent-based pipeline, PrismAgent, to generate well-structured problem instances. Each problem contains dynamic textual and visual input, a generated figure, alongside rich structured outputs: executable Python code for ground truth generation and verification, and detailed step-by-step reasoning. The dynamic nature and Python-powered automated ground truth generation of our benchmark allow for fine-grained experimental auditing of multimodal VLMs, revealing failure modes, uncertainty behaviors, and limitations in scientific reasoning. To this end, we propose five targeted evaluation tasks covering generalization, symbolic program synthesis, perturbation robustness, reasoning correction, and ambiguity resolution. Through comprehensive evaluation of existing VLMs, we highlight their limitations and showcase how PRiSM enables deeper insights into their scientific reasoning capabilities.

Paperid: 5294, https://arxiv.org/pdf/2512.04032.pdf

Abstract:
We present Jina-VLM, a 2.4B parameter vision-language model that achieves state-of-the-art multilingual visual question answering among open 2B-scale VLMs. The model couples a SigLIP2 vision encoder with a Qwen3 language backbone through an attention-pooling connector that enables token-efficient processing of arbitrary-resolution images. The model achieves leading results on standard VQA benchmarks and multilingual evaluations while preserving competitive text-only performance. Model weights and code are publicly released at https://huggingface.co/jinaai/jina-vlm .

Paperid: 5295, https://arxiv.org/pdf/2511.21557.pdf

Abstract:
Vision Language Action models have significantly advanced general purpose robotic manipulation by harnessing large scale pretrained vision and language representations. Among existing approaches, a majority of current VLA systems employ parallel two finger grippers as their default end effectors. However, such grippers face inherent limitations in handling certain real world tasks such as wiping glass surfaces or opening drawers without handles due to insufficient contact area or lack of adhesion. To overcome these challenges, we present a low cost, integrated hardware design that combines a mechanical two finger gripper with a vacuum suction unit, enabling dual mode manipulation within a single end effector. Our system supports flexible switching or synergistic use of both modalities, expanding the range of feasible tasks. We validate the efficiency and practicality of our design within two state of the art VLA frameworks: DexVLA and Pi0. Experimental results demonstrate that with the proposed hybrid end effector, robots can successfully perform multiple complex tasks that are infeasible for conventional two finger grippers alone. All hardware designs and controlling systems will be released.

Paperid: 5296, https://arxiv.org/pdf/2511.19518.pdf

Abstract:
Recent advances in vision-language models (VLMs) have shown remarkable performance across multimodal tasks, yet their ever-growing scale poses severe challenges for deployment and efficiency. Existing compression methods often rely on heuristic importance metrics or empirical pruning rules, lacking theoretical guarantees about information preservation. In this work, we propose InfoPrune, an information-theoretic framework for adaptive structural compression of VLMs. Grounded in the Information Bottleneck principle, we formulate pruning as a trade-off between retaining task-relevant semantics and discarding redundant dependencies. To quantify the contribution of each attention head, we introduce an entropy-based effective rank (eRank) and employ the Kolmogorov--Smirnov (KS) distance to measure the divergence between original and compressed structures. This yields a unified criterion that jointly considers structural sparsity and informational efficiency. Building on this foundation, we further design two complementary schemes: (1) a training-based head pruning guided by the proposed information loss objective, and (2) a training-free FFN compression via adaptive low-rank approximation. Extensive experiments on VQAv2, TextVQA, and GQA demonstrate that InfoPrune achieves up to 3.2x FLOP reduction and 1.8x acceleration with negligible performance degradation, establishing a theoretically grounded and practically effective step toward efficient multimodal large models.

Paperid: 5297, https://arxiv.org/pdf/2511.18714.pdf

Abstract:
Educational illustrations play a central role in communicating abstract concepts, yet current multimodal large language models (MLLMs) remain limited in producing pedagogically coherent and semantically consistent educational visuals. We introduce MAGMA-Edu, a self-reflective multi-agent framework that unifies textual reasoning and diagrammatic synthesis for structured educational problem generation. Unlike existing methods that treat text and image generation independently, MAGMA-Edu employs a two-stage co-evolutionary pipeline: (1) a generation-verification-reflection loop that iteratively refines question statements and solutions for mathematical accuracy, and (2) a code-based intermediate representation that enforces geometric fidelity and semantic alignment during image rendering. Both stages are guided by internal self-reflection modules that evaluate and revise outputs until domain-specific pedagogical constraints are met. Extensive experiments on multimodal educational benchmarks demonstrate the superiority of MAGMA-Edu over state-of-the-art MLLMs. Compared to GPT-4o, MAGMA-Edu improves the average textual metric from 57.01 to 92.31 (+35.3 pp) and boosts image-text consistency (ITC) from 13.20 to 85.24 (+72 pp). Across all model backbones, MAGMA-Edu achieves the highest scores (Avg-Text 96.20, ITC 99.12), establishing a new state of the art for multimodal educational content generation and demonstrating the effectiveness of self-reflective multi-agent collaboration in pedagogically aligned vision-language reasoning.

Paperid: 5298, https://arxiv.org/pdf/2511.17199.pdf

Abstract:
Vision-language-action (VLA) models show potential for general robotic tasks, but remain challenging in spatiotemporally coherent manipulation, which requires fine-grained representations. Typically, existing methods embed 3D positions into visual representations to enhance the spatial precision of actions. However, these methods struggle to achieve temporally coherent control over action execution. In this work, we propose VLA-4D, a general VLA model with 4D awareness for spatiotemporally coherent robotic manipulation. Our model is guided by two key designs: 1) 4D-aware visual representation. We extract visual features, embed 1D time into 3D positions for 4D embeddings, and fuse them into a unified visual representation via a cross-attention mechanism. 2) Spatiotemporal action representation. We extend conventional spatial action representations with temporal information to enable the spatiotemporal planning, and align the multimodal representations into the LLM for spatiotemporal action prediction. Within this unified framework, the designed visual and action representations jointly make robotic manipulation spatially-smooth and temporally-coherent. In addition, we extend the VLA dataset with temporal action annotations for fine-tuning our model. Extensive experiments have been conducted to verify the superiority of our method across different tasks of robotic manipulation.

Paperid: 5299, https://arxiv.org/pdf/2511.14446.pdf

Abstract:
Video understanding requires not only visual recognition but also complex reasoning. While Vision-Language Models (VLMs) demonstrate impressive capabilities, they typically process videos largely in a single-pass manner with limited support for evidence revisit and iterative refinement. While recently emerging agent-based methods enable long-horizon reasoning, they either depend heavily on expensive proprietary models or require extensive agentic RL training. To overcome these limitations, we propose Agentic Video Intelligence (AVI), a flexible and training-free framework that can mirror human video comprehension through system-level design and optimization. AVI introduces three key innovations: (1) a human-inspired three-phase reasoning process (Retrieve-Perceive-Review) that ensures both sufficient global exploration and focused local analysis, (2) a structured video knowledge base organized through entity graphs, along with multi-granularity integrated tools, constituting the agent's interaction environment, and (3) an open-source model ensemble combining reasoning LLMs with lightweight base CV models and VLM, eliminating dependence on proprietary APIs or RL training. Experiments on LVBench, VideoMME-Long, LongVideoBench, and Charades-STA demonstrate that AVI achieves competitive performance while offering superior interpretability.

Paperid: 5300, https://arxiv.org/pdf/2511.13442.pdf

Abstract:
With the rapid advancement of artificial intelligence-generated content (AIGC) technologies, including multimodal large language models (MLLMs) and diffusion models, image generation and manipulation have become remarkably effortless. Existing image forgery detection and localization (IFDL) methods often struggle to generalize across diverse datasets and offer limited interpretability. Nowadays, MLLMs demonstrate strong generalization potential across diverse vision-language tasks, and some studies introduce this capability to IFDL via large-scale training. However, such approaches cost considerable computational resources, while failing to reveal the inherent generalization potential of vanilla MLLMs to address this problem. Inspired by this observation, we propose Foresee, a training-free MLLM-based pipeline tailored for image forgery analysis. It eliminates the need for additional training and enables a lightweight inference process, while surpassing existing MLLM-based methods in both tamper localization accuracy and the richness of textual explanations. Foresee employs a type-prior-driven strategy and utilizes a Flexible Feature Detector (FFD) module to specifically handle copy-move manipulations, thereby effectively unleashing the potential of vanilla MLLMs in the forensic domain. Extensive experiments demonstrate that our approach simultaneously achieves superior localization accuracy and provides more comprehensive textual explanations. Moreover, Foresee exhibits stronger generalization capability, outperforming existing IFDL methods across various tampering types, including copy-move, splicing, removal, local enhancement, deepfake, and AIGC-based editing. The code will be released in the final version.

Paperid: 5301, https://arxiv.org/pdf/2511.12405.pdf

Abstract:
Exploring open-world situations in an end-to-end manner is a promising yet challenging task due to the need for strong generalization capabilities. In particular, end-to-end autonomous driving in unstructured outdoor environments often encounters conditions that were unfamiliar during training. In this work, we present Vision-Language Action Retrieval (VLA-R), an open-world end-to-end autonomous driving (OW-E2EAD) framework that integrates open-world perception with a novel vision-action retrieval paradigm. We leverage a frozen vision-language model for open-world detection and segmentation to obtain multi-scale, prompt-guided, and interpretable perception features without domain-specific tuning. A Q-Former bottleneck aggregates fine-grained visual representations with language-aligned visual features, bridging perception and action domains. To learn transferable driving behaviors, we introduce a vision-action contrastive learning scheme that aligns vision-language and action embeddings for effective open-world reasoning and action retrieval. Our experiments on a real-world robotic platform demonstrate strong generalization and exploratory performance in unstructured, unseen environments, even with limited data. Demo videos are provided in the supplementary material.

Paperid: 5302, https://arxiv.org/pdf/2511.12331.pdf

Abstract:
Vision-Language Models (VLMs) struggle with negation. Given a prompt like "retrieve (or generate) a street scene without pedestrians," they often fail to respect the "not." Existing methods address this limitation by fine-tuning on large negation datasets, but such retraining often compromises the model's zero-shot performance on affirmative prompts. We show that the embedding space of VLMs, such as CLIP, can be divided into semantically consistent subspaces. Based on this property, we propose a training-free framework that models negation as a subspace in the joint embedding space rather than a single point (Figure 1). To find the matching image for a caption such as "A but not N," we construct two spherical caps around the embeddings of A and N, and we score images by the central direction of the region that is close to A and far from N. Across retrieval, MCQ, and text-to-image tasks, our method improves negation understanding by about 30% on average over prior methods. It closes the gap between affirmative and negated prompts while preserving the zero-shot performance that fine-tuned models fail to maintain. Code will be released upon publication.

Paperid: 5303, https://arxiv.org/pdf/2511.12259.pdf

Abstract:
Radiology report generation from chest X-rays is an important task in artificial intelligence with the potential to greatly reduce radiologists' workload and shorten patient wait times. Despite recent advances, existing approaches often lack sufficient disease-awareness in visual representations and adequate vision-language alignment to meet the specialized requirements of medical image analysis. As a result, these models usually overlook critical pathological features on chest X-rays and struggle to generate clinically accurate reports. To address these limitations, we propose a novel dual-stage disease-aware framework for chest X-ray report generation. In Stage~1, our model learns Disease-Aware Semantic Tokens (DASTs) corresponding to specific pathology categories through cross-attention mechanisms and multi-label classification, while simultaneously aligning vision and language representations via contrastive learning. In Stage~2, we introduce a Disease-Visual Attention Fusion (DVAF) module to integrate disease-aware representations with visual features, along with a Dual-Modal Similarity Retrieval (DMSR) mechanism that combines visual and disease-specific similarities to retrieve relevant exemplars, providing contextual guidance during report generation. Extensive experiments on benchmark datasets (i.e., CheXpert Plus, IU X-ray, and MIMIC-CXR) demonstrate that our disease-aware framework achieves state-of-the-art performance in chest X-ray report generation, with significant improvements in clinical accuracy and linguistic quality.

Paperid: 5304, https://arxiv.org/pdf/2511.12220.pdf

Abstract:
Vision-language models (VLMs) frequently produce hallucinations in the form of descriptions of objects, attributes, or relations that do not exist in the image due to over-reliance on language priors and imprecise cross-modal grounding. We introduce Spectral Representation Filtering (SRF), a lightweight, training-free method to suppress such hallucinations by analyzing and correcting the covariance structure of the model's representations. SRF identifies low-rank hallucination modes through eigendecomposition of the covariance of the differences between features collected for truthful and hallucinatory captions, revealing structured biases in the feature space. A soft spectral filter then attenuates these modes in the feed-forward projection weights of deeper vLLM layers, equalizing feature variance while preserving semantic fidelity. Unlike decoding or retraining-based approaches, SRF operates entirely post-hoc, incurs zero inference overhead, and requires no architectural modifications. Across three families of VLMs (LLaVA-1.5, MiniGPT-4, and mPLUG-Owl2), SRF consistently reduces hallucination rates on MSCOCO, POPE-VQA, and other visual tasks benchmarks, achieving state-of-the-art faithfulness without degrading caption quality.

Paperid: 5305, https://arxiv.org/pdf/2511.11438.pdf

Abstract:
Multimodal large language models (MLLMs) have enabled a wide range of advanced vision-language applications, including fine-grained object recognition and contextual understanding. When querying specific regions or objects in an image, human users naturally use "visual prompts" (VPs), such as bounding boxes, to provide reference. However, no existing benchmark systematically evaluates the ability of MLLMs to interpret such VPs. This gap leaves it unclear whether current MLLMs can effectively recognize VPs, an intuitive prompting method for humans, and use them to solve problems. To address this limitation, we introduce VP-Bench, a benchmark for assessing MLLMs' capability in VP perception and utilization. VP-Bench employs a two-stage evaluation framework: Stage 1 examines models' ability to perceive VPs in natural scenes, using 30k visualized prompts spanning eight shapes and 355 attribute combinations. Stage 2 investigates the impact of VPs on downstream tasks, measuring their effectiveness in real-world problem-solving scenarios. Using VP-Bench, we evaluate 28 MLLMs, including proprietary systems (e.g., GPT-4o) and open-source models (e.g., InternVL3 and Qwen2.5-VL), and provide a comprehensive analysis of factors that affect VP understanding, such as variations in VP attributes, question arrangement, and model scale. VP-Bench establishes a new reference framework for studying how MLLMs comprehend and resolve grounded referring questions.

Paperid: 5306, https://arxiv.org/pdf/2511.08978.pdf

Abstract:
Nowadays, navigation and ride-sharing apps have collected numerous images with spatio-temporal data. A core technology for analyzing such images, associated with spatiotemporal information, is Traffic Scene Understanding (TSU), which aims to provide a comprehensive description of the traffic scene. Unlike traditional spatio-temporal data analysis tasks, the dependence on both spatio-temporal and visual-textual data introduces distinct challenges to TSU task. However, recent research often treats TSU as a common image understanding task, ignoring the spatio-temporal information and overlooking the interrelations between different aspects of the traffic scene. To address these issues, we propose a novel SpatioTemporal Enhanced Model based on CILP (ST-CLIP) for TSU. Our model uses the classic vision-language model, CLIP, as the backbone, and designs a Spatio-temporal Context Aware Multiaspect Prompt (SCAMP) learning method to incorporate spatiotemporal information into TSU. The prompt learning method consists of two components: A dynamic spatio-temporal context representation module that extracts representation vectors of spatio-temporal data for each traffic scene image, and a bi-level ST-aware multi-aspect prompt learning module that integrates the ST-context representation vectors into word embeddings of prompts for the CLIP model. The second module also extracts low-level visual features and image-wise high-level semantic features to exploit interactive relations among different aspects of traffic scenes. To the best of our knowledge, this is the first attempt to integrate spatio-temporal information into visionlanguage models to facilitate TSU task. Experiments on two realworld datasets demonstrate superior performance in the complex scene understanding scenarios with a few-shot learning strategy.

Paperid: 5307, https://arxiv.org/pdf/2511.08971.pdf

Abstract:
The performance of egocentric AI agents is fundamentally limited by multimodal intent ambiguity. This challenge arises from a combination of underspecified language, imperfect visual data, and deictic gestures, which frequently leads to task failure. Existing monolithic Vision-Language Models (VLMs) struggle to resolve these multimodal ambiguous inputs, often failing silently or hallucinating responses. To address these ambiguities, we introduce the Plug-and-Play Clarifier, a zero-shot and modular framework that decomposes the problem into discrete, solvable sub-tasks. Specifically, our framework consists of three synergistic modules: (1) a text clarifier that uses dialogue-driven reasoning to interactively disambiguate linguistic intent, (2) a vision clarifier that delivers real-time guidance feedback, instructing users to adjust their positioning for improved capture quality, and (3) a cross-modal clarifier with grounding mechanism that robustly interprets 3D pointing gestures and identifies the specific objects users are pointing to. Extensive experiments demonstrate that our framework improves the intent clarification performance of small language models (4--8B) by approximately 30%, making them competitive with significantly larger counterparts. We also observe consistent gains when applying our framework to these larger models. Furthermore, our vision clarifier increases corrective guidance accuracy by over 20%, and our cross-modal clarifier improves semantic answer accuracy for referential grounding by 5%. Overall, our method provides a plug-and-play framework that effectively resolves multimodal ambiguity and significantly enhances user experience in egocentric interaction.

Paperid: 5308, https://arxiv.org/pdf/2511.06225.pdf

Abstract:
Pre-trained vision language models have shown remarkable performance on visual recognition tasks, but they typically assume the availability of complete multimodal inputs during both training and inference. In real-world scenarios, however, modalities may be missing due to privacy constraints, collection difficulties, or resource limitations. While previous approaches have addressed this challenge using prompt learning techniques, they fail to capture the cross-modal relationships necessary for effective multimodal visual recognition and suffer from inevitable computational overhead. In this paper, we introduce MoRA, a parameter-efficient fine-tuning method that explicitly models cross-modal interactions while maintaining modality-specific adaptations. MoRA introduces modality-common parameters between text and vision encoders, enabling bidirectional knowledge transfer. Additionally, combined with the modality-specific parameters, MoRA allows the backbone model to maintain inter-modality interaction and enable intra-modality flexibility. Extensive experiments on standard benchmarks demonstrate that MoRA achieves an average performance improvement in missing-modality scenarios by 5.24% and uses only 25.90% of the inference time compared to the SOTA method while requiring only 0.11% of trainable parameters compared to full fine-tuning.

Paperid: 5309, https://arxiv.org/pdf/2511.05841.pdf

Abstract:
Alzheimer's disease is a prevalent neurodegenerative disorder for which early detection is critical. Handwriting-often disrupted in prodromal AD-provides a non-invasive and cost-effective window into subtle motor and cognitive decline. Existing handwriting-based AD studies, mostly relying on online trajectories and hand-crafted features, have not systematically examined how task type influences diagnostic performance and cross-task generalization. Meanwhile, large-scale vision language models have demonstrated remarkable zero or few-shot anomaly detection in natural images and strong adaptability across medical modalities such as chest X-ray and brain MRI. However, handwriting-based disease detection remains largely unexplored within this paradigm. To close this gap, we introduce a lightweight Cross-Layer Fusion Adapter framework that repurposes CLIP for handwriting-based AD screening. CLFA implants multi-level fusion adapters within the visual encoder to progressively align representations toward handwriting-specific medical cues, enabling prompt-free and efficient zero-shot inference. Using this framework, we systematically investigate cross-task generalization-training on a specific handwriting task and evaluating on unseen ones-to reveal which task types and writing patterns most effectively discriminate AD. Extensive analyses further highlight characteristic stroke patterns and task-level factors that contribute to early AD identification, offering both diagnostic insights and a benchmark for handwriting-based cognitive assessment.

Paperid: 5310, https://arxiv.org/pdf/2511.02996.pdf

Abstract:
Vision-language models (VLMs) have demonstrated strong cross-modal capabilities, yet most work remains limited to 2D data and assumes binary supervision (i.e., positive vs. negative pairs), overlooking the continuous and structured dependencies present in volumetric data such as CT. Existing approaches often treat volumetric scans as independent 2D slices, compromising spatial coherence and underutilizing rich clinical semantics. We propose SCALE-VLP, a soft-weighted contrastive vision-language pre-training framework that integrates (i) volumetric spatial semantics to preserve anatomical structure and (ii) domain-aware, knowledge-infused semantics (e.g., radiological ontologies) to guide alignment. This yields structurally consistent and semantically grounded representations under limited supervision, demonstrating strong cross-task transferability (retrieval, report generation, and classification), and cross-domain generalizability with consistent gains without further fine-tuning. In particular, compared to the previous state of the art, SCALE-VLP achieves up to 4.3x higher top-1 CT-report retrieval, improves abnormality classification by 10 points, and reaches ROUGE-L 0.44 and BERT-F1 0.89 for report generation. Further, in zero-shot evaluation on an out-of-domain external dataset, we observe consistent gains, indicating the cross-task and cross-domain generalization ability of SCALE-VLP.

Paperid: 5311, https://arxiv.org/pdf/2511.01472.pdf

Abstract:
The rapid progress of vision--language models (VLMs) has sparked growing interest in robotic control, where natural language can express the operation goals while visual feedback links perception to action. However, directly deploying VLM-driven policies on aerial manipulators remains unsafe and unreliable since the generated actions are often inconsistent, hallucination-prone, and dynamically infeasible for flight. In this work, we present AERMANI-VLM, the first framework to adapt pretrained VLMs for aerial manipulation by separating high-level reasoning from low-level control, without any task-specific fine-tuning. Our framework encodes natural language instructions, task context, and safety constraints into a structured prompt that guides the model to generate a step-by-step reasoning trace in natural language. This reasoning output is used to select from a predefined library of discrete, flight-safe skills, ensuring interpretable and temporally consistent execution. By decoupling symbolic reasoning from physical action, AERMANI-VLM mitigates hallucinated commands and prevents unsafe behavior, enabling robust task completion. We validate the framework in both simulation and hardware on diverse multi-step pick-and-place tasks, demonstrating strong generalization to previously unseen commands, objects, and environments.

Paperid: 5312, https://arxiv.org/pdf/2510.25600.pdf

Abstract:
Vision-Language Large Models (VLLMs) face significant efficiency challenges when processing high-resolution inputs. The quadratic complexity in attention and autoregressive generation, as well as the constantly growing key value (KV) cache size, severely hinder the prefilling and decoding stages. Recent efforts have attempted to compress KV cache by identifying and pruning KV cache of less important tokens, but these methods typically rely on attention scores to estimate token importance, making them incompatible with efficient attention mechanisms such as FlashAttention and Sparse Attention, which do not explicitly compute attention matrices. Moreover, existing methods overlook how sparse attention, while accelerating the prefilling stage, alters the information structure of the KV cache, thereby compromising the effectiveness of downstream KV cache compression strategies. To address this issue, we propose PureKV, a plug-and-play framework for joint optimization of sparse attention and KV cache compression. We first introduce a KV cache compression strategy that is fully compatible with efficient attention accelerators. Our method utilizes lower layer attention scores to estimate the importance of high layers' KV cache, enabling active pruning without compromising accuracy. In addition, we have designed a Spatial-Temporal Sparse Attention (ST-SpAttn) module specifically tailored for video KV cache compression algorithms. This module combines spatial and temporal attention sparsity to improve the compression efficiency of KV cache optimization algorithms by purifying spatial noise and temporal redundancy in KV cache. At the same time, ST-SpAttn also accelerated the prefilling stage of VLLMs. Extensive experiments on VLLMs (VideoLLaMA2, Qwen2.5-VL) have shown that PureKV achieves 5.0 times KV cache compression and 3.16 times prefill acceleration, with negligible quality degradation.

Paperid: 5313, https://arxiv.org/pdf/2510.25191.pdf

Abstract:
Interpreting visual observations and natural language instructions for complex task execution remains a key challenge in robotics and AI. Despite recent advances, language-driven navigation is still difficult, particularly for UAVs in small-scale 3D environments. Existing Vision-Language Navigation (VLN) approaches are mostly designed for ground robots and struggle to generalize to aerial tasks that require full 3D spatial reasoning. The emergence of large Vision-Language Models (VLMs), such as GPT and Claude, enables zero-shot semantic reasoning from visual and textual inputs. However, these models lack spatial grounding and are not directly applicable to navigation. To address these limitations, SoraNav is introduced, an adaptive UAV navigation framework that integrates zero-shot VLM reasoning with geometry-aware decision-making. Geometric priors are incorporated into image annotations to constrain the VLM action space and improve decision quality. A hybrid switching strategy leverages navigation history to alternate between VLM reasoning and geometry-based exploration, mitigating dead-ends and redundant revisits. A PX4-based hardware-software platform, comprising both a digital twin and a physical micro-UAV, enables reproducible evaluation. Experimental results show that in 2.5D scenarios, our method improves Success Rate (SR) by 25.7% and Success weighted by Path Length (SPL) by 17%. In 3D scenarios, it improves SR by 29.5% and SPL by 18.5% relative to the baseline.

Paperid: 5314, https://arxiv.org/pdf/2510.24321.pdf

Abstract:
Remote sensing applications increasingly rely on deep learning for scene classification. However, their performance is often constrained by the scarcity of labeled data and the high cost of annotation across diverse geographic and sensor domains. While recent vision-language models like CLIP have shown promise by learning transferable representations at scale by aligning visual and textual modalities, their direct application to remote sensing remains suboptimal due to significant domain gaps and the need for task-specific semantic adaptation. To address this critical challenge, we systematically explore prompt learning as a lightweight and efficient adaptation strategy for few-shot remote sensing image scene classification. We evaluate several representative methods, including Context Optimization, Conditional Context Optimization, Multi-modal Prompt Learning, and Prompting with Self-Regulating Constraints. These approaches reflect complementary design philosophies: from static context optimization to conditional prompts for enhanced generalization, multi-modal prompts for joint vision-language adaptation, and semantically regularized prompts for stable learning without forgetting. We benchmark these prompt-learning methods against two standard baselines: zero-shot CLIP with hand-crafted prompts and a linear probe trained on frozen CLIP features. Through extensive experiments on multiple benchmark remote sensing datasets, including cross-dataset generalization tests, we demonstrate that prompt learning consistently outperforms both baselines in few-shot scenarios. Notably, Prompting with Self-Regulating Constraints achieves the most robust cross-domain performance. Our findings underscore prompt learning as a scalable and efficient solution for bridging the domain gap in satellite and aerial imagery, providing a strong foundation for future research in this field.

Paperid: 5315, https://arxiv.org/pdf/2510.23571.pdf

Abstract:
The pursuit of robot generalists - instructable agents capable of performing diverse tasks across diverse environments - demands rigorous and scalable evaluation. Yet real-world testing of robot policies remains fundamentally constrained: it is labor-intensive, slow, unsafe at scale, and difficult to reproduce. Existing simulation benchmarks are similarly limited, as they train and test policies within the same synthetic domains and cannot assess models trained from real-world demonstrations or alternative simulation environments. As policies expand in scope and complexity, these barriers only intensify, since defining "success" in robotics often hinges on nuanced human judgments of execution quality. In this paper, we introduce a new benchmarking framework that overcomes these challenges by shifting VLA evaluation into large-scale simulated environments augmented with online human feedback. Leveraging advances in vision-language models, 2D-to-3D generative modeling, and differentiable rendering, our approach automatically converts video demonstrations from widely used robot datasets into simulated counterparts. Within these digital twins, we assess VLA policies using both automated VLM-guided scoring and scalable human preference judgments collected from crowdworkers, transforming human involvement from tedious scene setup, resetting, and safety supervision into lightweight preference comparisons. To measure robustness, we systematically perturb simulated environments along multiple axes, such as textures and object placements, stress-testing policy generalization under controlled variation. The result is a continuously evolving, reproducible, and scalable benchmark for real-world trained robot manipulation policies, addressing a critical missing capability in today's robotics landscape.

Paperid: 5316, https://arxiv.org/pdf/2510.23497.pdf

Abstract:
Training vision-language models (VLMs) for complex reasoning remains a challenging task, i.a. due to the scarcity of high-quality image-text reasoning data. Conversely, text-based reasoning resources are abundant and scalable, but it is still an open question how to leveraging them for VLM reasoning. To address this problem, we propose VOLD, a framework to transfer reasoning capabilities from text-only teacher models to VLM student models. To this end, VOLD combines reinforcement learning via Group Relative Policy Optimization (GRPO) with on-policy distillation, which allows the student reasoning traces to be guided by the teacher model, resulting in a significant gain over using GRPO alone. We further show that a cold-start alignment is essential for an effective transfer during the online training phase in this scenario and that without sufficient distributional alignment between teacher and student, on-policy distillation fails to provide meaningful guidance. We evaluate VOLD across diverse benchmarks including MMMU-Pro, MathVision, MathVista, and LogicVista, showing that VOLD outperforms the baseline model significantly and improves over the state of the art by a margin. Our ablation shows the importance of a cold-start alignment via SFT for on-policy distillation with a text-only teacher.

Paperid: 5317, https://arxiv.org/pdf/2510.21120.pdf

Abstract:
What exactly makes a particular image unsafe? Systematically differentiating between benign and problematic images is a challenging problem, as subtle changes to an image, such as an insulting gesture or symbol, can drastically alter its safety implications. However, existing image safety datasets are coarse and ambiguous, offering only broad safety labels without isolating the specific features that drive these differences. We introduce SafetyPairs, a scalable framework for generating counterfactual pairs of images, that differ only in the features relevant to the given safety policy, thus flipping their safety label. By leveraging image editing models, we make targeted changes to images that alter their safety labels while leaving safety-irrelevant details unchanged. Using SafetyPairs, we construct a new safety benchmark, which serves as a powerful source of evaluation data that highlights weaknesses in vision-language models' abilities to distinguish between subtly different images. Beyond evaluation, we find our pipeline serves as an effective data augmentation strategy that improves the sample efficiency of training lightweight guard models. We release a benchmark containing over 3,020 SafetyPair images spanning a diverse taxonomy of 9 safety categories, providing the first systematic resource for studying fine-grained image safety distinctions.

Paperid: 5318, https://arxiv.org/pdf/2510.16240.pdf

Abstract:
The rise of surgical robots and vision-language-action models has accelerated the development of autonomous surgical policies and efficient assessment strategies. However, evaluating these policies directly on physical robotic platforms such as the da Vinci Research Kit (dVRK) remains hindered by high costs, time demands, reproducibility challenges, and variability in execution. World foundation models (WFM) for physical AI offer a transformative approach to simulate complex real-world surgical tasks, such as soft tissue deformation, with high fidelity. This work introduces Cosmos-Surg-dVRK, a surgical finetune of the Cosmos WFM, which, together with a trained video classifier, enables fully automated online evaluation and benchmarking of surgical policies. We evaluate Cosmos-Surg-dVRK using two distinct surgical datasets. On tabletop suture pad tasks, the automated pipeline achieves strong correlation between online rollouts in Cosmos-Surg-dVRK and policy outcomes on the real dVRK Si platform, as well as good agreement between human labelers and the V-JEPA 2-derived video classifier. Additionally, preliminary experiments with ex-vivo porcine cholecystectomy tasks in Cosmos-Surg-dVRK demonstrate promising alignment with real-world evaluations, highlighting the platform's potential for more complex surgical procedures.

Paperid: 5319, https://arxiv.org/pdf/2510.11314.pdf

Abstract:
Individuals with intellectual disabilities often have difficulties in comprehending complex texts. While many text-to-image models prioritize aesthetics over accessibility, it is not clear how visual illustrations relate to text simplifications (TS) generated from them. This paper presents a structured vision-language model (VLM) prompting framework for generating accessible images from simplified texts. We designed five prompt templates, i.e., Basic Object Focus, Contextual Scene, Educational Layout, Multi-Level Detail, and Grid Layout, each following distinct spatial arrangements while adhering to accessibility constraints such as object count limits, spatial separation, and content restrictions. Using 400 sentence-level simplifications from four established TS datasets (OneStopEnglish, SimPA, Wikipedia, and ASSET), we conducted a two-phase evaluation: Phase 1 assessed prompt template effectiveness with CLIPScores, and Phase 2 involved human annotation of generated images across ten visual styles by four accessibility experts. Results show that the Basic Object Focus prompt template achieved the highest semantic alignment, indicating that visual minimalism enhances language accessibility. Expert evaluation further identified Retro style as the most accessible and Wikipedia as the most effective data source. Inter-annotator agreement varied across dimensions, with Text Simplicity showing strong reliability and Image Quality proving more subjective. Overall, our framework offers practical guidelines for accessible content generation and underscores the importance of structured prompting in AI-generated visual accessibility tools.

Paperid: 5320, https://arxiv.org/pdf/2510.10846.pdf

Abstract:
As vision-language models become increasingly capable, maintaining a balance between safety and usefulness remains a central challenge. Safety mechanisms, while essential, can backfire, causing over-refusal, where models decline benign requests out of excessive caution. Yet, no existing benchmark has systematically addressed over-refusal in the visual modality. This setting introduces unique challenges, such as dual-use cases where an instruction is harmless, but the accompanying image contains harmful content. Models frequently fail in such scenarios, either refusing too conservatively or completing tasks unsafely, which highlights the need for more fine-grained alignment. The ideal behavior is safe completion, i.e., fulfilling the benign parts of a request while explicitly warning about any potentially harmful elements. To address this, we present DUAL-Bench, the first multimodal benchmark focused on over-refusal and safe completion in VLMs. We evaluated 18 VLMs across 12 hazard categories, with focus on their robustness under semantics-preserving visual perturbations. The results reveal substantial room for improvement: GPT-5-Nano achieves 12.9% safe completion, GPT-5 models average 7.9%, and Qwen models only 3.9%. We hope that DUAL-Bench will foster the development of more nuanced alignment strategies that ensure models remain both safe and useful in complex multimodal settings.

Paperid: 5321, https://arxiv.org/pdf/2510.08849.pdf

Abstract:
Open-vocabulary 3D instance segmentation seeks to segment and classify instances beyond the annotated label space. Existing methods typically map 3D instances to 2D RGB-D images, and then employ vision-language models (VLMs) for classification. However, such a mapping strategy usually introduces noise from 2D occlusions and incurs substantial computational and memory costs during inference, slowing down the inference speed. To address the above problems, we propose a Fast Open-vocabulary 3D instance segmentation method via Label-guided Knowledge distillation (FOLK). Our core idea is to design a teacher model that extracts high-quality instance embeddings and distills its open-vocabulary knowledge into a 3D student model. In this way, during inference, the distilled 3D model can directly classify instances from the 3D point cloud, avoiding noise caused by occlusions and significantly accelerating the inference process. Specifically, we first design a teacher model to generate a 2D CLIP embedding for each 3D instance, incorporating both visibility and viewpoint diversity, which serves as the learning target for distillation. We then develop a 3D student model that directly produces a 3D embedding for each 3D instance. During training, we propose a label-guided distillation algorithm to distill open-vocabulary knowledge from label-consistent 2D embeddings into the student model. FOLK conducted experiments on the ScanNet200 and Replica datasets, achieving state-of-the-art performance on the ScanNet200 dataset with an AP50 score of 35.7, while running approximately 6.0x to 152.2x faster than previous methods. All codes will be released after the paper is accepted.

Paperid: 5322, https://arxiv.org/pdf/2510.08581.pdf

Abstract:
Hallucinations in vision-language models have been extensively studied using benchmarks that probe reliability in image-text settings. In contrast, the effect of spoken queries on multimodal hallucinations remains largely unexplored, despite the growing role of voice-driven interfaces. In this work, we investigate how spoken input influences hallucinations in multimodal large language models. We present RePOPE-Spk, an audio-augmented extension of the RePOPE benchmark, where queries are provided as speech under diverse acoustic conditions. Using RePOPE-Spk, we systematically evaluate both proprietary and open-source models. Experimental results show that hallucinations escalate when queries are spoken rather than written: error rates increase by 3% under clean speech and by up to 20% with environmental noise. Input order and query length further affect robustness, while strategies such as many-shot prompting and chain-of-thought reasoning offer partial but insufficient mitigation. These findings highlight a critical and underexplored challenge, opening new directions for building reliable voice interface systems.

Paperid: 5323, https://arxiv.org/pdf/2510.08482.pdf

Abstract:
Iconicity, the resemblance between linguistic form and meaning, is pervasive in signed languages, offering a natural testbed for visual grounding. For vision-language models (VLMs), the challenge is to recover such essential mappings from dynamic human motion rather than static context. We introduce the \textit{Visual Iconicity Challenge}, a novel video-based benchmark that adapts psycholinguistic measures to evaluate VLMs on three tasks: (i) phonological sign-form prediction (e.g., handshape, location), (ii) transparency (inferring meaning from visual form), and (iii) graded iconicity ratings. We assess $13$ state-of-the-art VLMs in zero- and few-shot settings on Sign Language of the Netherlands and compare them to human baselines. On \textit{phonological form prediction}, VLMs recover some handshape and location detail but remain below human performance; on \textit{transparency}, they are far from human baselines; and only top models correlate moderately with human \textit{iconicity ratings}. Interestingly, \textit{models with stronger phonological form prediction correlate better with human iconicity judgment}, indicating shared sensitivity to visually grounded structure. Our findings validate these diagnostic tasks and motivate human-centric signals and embodied learning methods for modelling iconicity and improving visual grounding in multimodal models.

Paperid: 5324, https://arxiv.org/pdf/2510.06529.pdf

Abstract:
Recent advances in Vision-Language Models (VLMs) have enabled unified understanding across text and images, yet equipping these models with robust image generation capabilities remains challenging. Existing approaches often rely on reconstruction-oriented autoencoders or complex bridging mechanisms, leading to misalignment between understanding and generation representations, or architectural complexity. In this work, we propose VUGEN, a novel framework that explicitly leverages VLM's pretrained visual understanding priors for efficient and high-quality image generation. Our approach first transforms the high-dimensional latent space of the VLM's native vision encoder into a lower-dimensional, tractable distribution that maximally preserves visual information. The VLM is then trained to sample within this reduced latent space, ensuring alignment with its visual understanding capabilities. Finally, a dedicated pixel decoder maps these generated latents back to the image space. We find that a VAE-free pixel diffusion decoder to be on par or better than commonly used complex latent diffusion decoders that internally rely on VAE latents. Extensive experiments demonstrate that VUGEN achieves superior image generation performance, improving DPG Bench from 71.17 to 74.32 and FID from 11.86 to 9.06 on COCO, while fully preserving the VLM's original understanding capabilities.

Paperid: 5325, https://arxiv.org/pdf/2510.06064.pdf

Abstract:
Vision-based Proximal Policy Optimization (PPO) struggles with visual observation-based robotic laparoscopic surgical tasks due to the high-dimensional nature of visual input, the sparsity of rewards in surgical environments, and the difficulty of extracting task-relevant features from raw visual data. We introduce a simple approach integrating MedFlamingo, a medical domain-specific Vision-Language Model, with PPO. Our method is evaluated on five diverse laparoscopic surgery task environments in LapGym, using only endoscopic visual observations. MedFlamingo PPO outperforms and converges faster compared to both standard vision-based PPO and OpenFlamingo PPO baselines, achieving task success rates exceeding 70% across all environments, with improvements ranging from 66.67% to 1114.29% compared to baseline. By processing task observations and instructions once per episode to generate high-level planning tokens, our method efficiently combines medical expertise with real-time visual feedback. Our results highlight the value of specialized medical knowledge in robotic surgical planning and decision-making.

Paperid: 5326, https://arxiv.org/pdf/2510.04991.pdf

Abstract:
We present a novel high-level planning framework that leverages vision-language models (VLMs) to improve autonomous navigation in unknown indoor environments with many dead ends. Traditional exploration methods often take inefficient routes due to limited global reasoning and reliance on local heuristics. In contrast, our approach enables a VLM to reason directly about an occupancy map in a zero-shot manner, selecting subgoals that are likely to lead to more efficient paths. At each planning step, we convert a 3D occupancy grid into a partial 2D map of the environment, and generate candidate subgoals. Each subgoal is then evaluated and ranked against other candidates by the model. We integrate this planning scheme into DYNUS \cite{kondo2025dynus}, a state-of-the-art trajectory planner, and demonstrate improved navigation efficiency in simulation. The VLM infers structural patterns (e.g., rooms, corridors) from incomplete maps and balances the need to make progress toward a goal against the risk of entering unknown space. This reduces common greedy failures (e.g., detouring into small rooms) and achieves about 10\% shorter paths on average.

Paperid: 5327, https://arxiv.org/pdf/2510.02528.pdf

Abstract:
Large Multimodal Models (LMMs) demonstrate impressive in-context learning abilities from limited multimodal demonstrations, yet the internal mechanisms supporting such task learning remain opaque. Building on prior work of large language models, we show that a small subset of attention heads in the vision-language model OpenFlamingo-4B is responsible for transmitting representations of spatial relations. The activations of these attention heads, termed function vectors, can be extracted and manipulated to alter an LMM's performance on relational tasks. First, using both synthetic and real image datasets, we apply causal mediation analysis to identify attention heads that strongly influence relational predictions, and extract multimodal function vectors that improve zero-shot accuracy at inference time. We further demonstrate that these multimodal function vectors can be fine-tuned with a modest amount of training data, while keeping LMM parameters frozen, to significantly outperform in-context learning baselines. Finally, we show that relation-specific function vectors can be linearly combined to solve analogy problems involving novel and untrained spatial relations, highlighting the strong generalization ability of this approach. Our results show that LMMs encode spatial relational knowledge within localized internal structures, which can be systematically extracted and optimized, thereby advancing our understanding of model modularity and enhancing control over relational reasoning in LMMs.

Paperid: 5328, https://arxiv.org/pdf/2510.00603.pdf

Abstract:
Automated structural defect annotation is essential for ensuring infrastructure safety while minimizing the high costs and inefficiencies of manual labeling. A novel agentic annotation framework, Agent-based Defect Pattern Tagger (ADPT), is introduced that integrates Large Vision-Language Models (LVLMs) with a semantic pattern matching module and an iterative self-questioning refinement mechanism. By leveraging optimized domain-specific prompting and a recursive verification process, ADPT transforms raw visual data into high-quality, semantically labeled defect datasets without any manual supervision. Experimental results demonstrate that ADPT achieves up to 98% accuracy in distinguishing defective from non-defective images, and 85%-98% annotation accuracy across four defect categories under class-balanced settings, with 80%-92% accuracy on class-imbalanced datasets. The framework offers a scalable and cost-effective solution for high-fidelity dataset construction, providing strong support for downstream tasks such as transfer learning and domain adaptation in structural damage assessment.

Paperid: 5329, https://arxiv.org/pdf/2510.00040.pdf

Abstract:
Large vision-language models (VLMs) achieve strong benchmark performance, but controlling their behavior through instruction tuning remains difficult. Reducing the budget of instruction tuning dataset often causes regressions, as heuristic strategies treat models as black boxes and overlook the latent capabilities that govern learning. We introduce Capability-Attributed Data Curation (CADC), a framework that shifts curation from task-specific heuristics to intrinsic capability analysis. CADC discovers intrinsic capabilities in an unsupervised manner from gradient-based learning trajectories, attributes training data to these capabilities via influence estimation, and curates capability-aware curricula through balanced selection and staged sequencing. This transforms black-box instruction tuning into a controllable, capability-driven process. With as little as 5% of the original data, CADC surpasses full-data training on multimodal benchmarks. These results validate intrinsic capabilities as the fundamental building blocks of model learning and establish CADC as a principle paradigm for instruction data curation.

Paperid: 5330, https://arxiv.org/pdf/2509.26235.pdf

Abstract:
Recent advances in Visually-rich Document Understanding rely on large Vision-Language Models like Donut, which perform document-level Visual Question Answering without Optical Character Recognition. Despite their effectiveness, these models are too costly for real-time or resource-constrained applications. We investigate model compression through knowledge distillation, training compact student models from a larger teacher. We leverage mechanistic interpretability to drive student architecture design within this framework. By analyzing internal computations, we identify essential subcomponents to retain, while having a clear view of which subcomponents should be approximated, skipped, or reparametrized based on their function. This approach yields Donut-MINT (Mechanistic Interpretability-based Network Trimming), a pruned Donut variant that reduces inference time and memory usage while maintaining strong performance on DocVQA, a standard benchmark for document Visual Question Answering. Our method reframes compression as circuit discovery, bridging interpretability research and practical Vision-Language Model deployment.

Paperid: 5331, https://arxiv.org/pdf/2509.25699.pdf

Abstract:
Multimodal Chain-of-Thought (CoT) has emerged as a powerful technique for enhancing the vision-language reasoning with interleaved information. However, existing methods often rely on simplistic heuristics for constructing interleaved CoT, typically depending on attention maps, which our empirical analysis reveals can be unreliable. What's more, the shortcomings of their passive and purposeless selection strategies and their arbitrary triggering mechanisms in capturing the model's cognitive need for information are further amplified. In this paper, we propose \textbf{AIMCoT}, an \textbf{A}ctive \textbf{I}nformation-driven \textbf{M}ulti-modal \textbf{C}hain-\textbf{o}f-\textbf{T}hought framework that addresses these fundamental limitations. AIMCoT introduces three synergistic components: (1) \textbf{Context-enhanced Attention-map Generation (CAG)}, which mitigates the text-vision granularity imbalance, thereby producing more reliable attention maps as a foundation. (2) \textbf{Active Visual Probing (AVP)}, which replaces passive selection with a proactive, goal-oriented strategy grounded in information theory to select image regions that help answer the questions maximally. (3) \textbf{Dynamic Attention-shifting Trigger (DAT)}, which intelligently determines the optimal moments to insert visual information by monitoring the model's text-to-vision attention shifts. Extensive experiments on three challenging benchmarks demonstrate that AIMCoT significantly outperforms state-of-the-art methods across different settings. By actively foraging for information and dynamically structuring its reasoning process, AIMCoT represents a critical step towards more robust, effective, and human-like multimodal reasoning. Our code is available at https://anonymous.4open.science/r/AIMCoT.

Paperid: 5332, https://arxiv.org/pdf/2509.25669.pdf

Abstract:
We propose a method to improve Visual Question Answering (VQA) with Retrieval-Augmented Generation (RAG) by introducing text-grounded object localization. Rather than retrieving information based on the entire image, our approach enables the model to generate a bounding box around the object most relevant to the question, allowing for targeted image cropping and focused retrieval. This reduces background noise, improves alignment between visual and textual cues, and helps mitigate hallucinations. Our RAG method enhances context-aware VQA responses increased the accuracy from 22.19% to 25.64%, with an absolute increase of 3.45 percentage points, compared to the baseline Llama-3.2-Vision-11B agent. We also proposed a de-hallucination method based on question type which can effectively reduce the hallucination rate from 65.79% to 13.88% and improves the truthfulness score.

Paperid: 5333, https://arxiv.org/pdf/2509.25559.pdf

Abstract:
Generalist multimodal AI systems such as large language models (LLMs) and vision language models (VLMs) are increasingly accessed by clinicians and patients alike for medical image interpretation through widely available consumer-facing chatbots. Most evaluations claiming expert level performance are on public datasets containing common pathologies. Rigorous evaluation of frontier models on difficult diagnostic cases remains limited. We developed a pilot benchmark of 50 expert-level "spot diagnosis" cases across multiple imaging modalities to evaluate the performance of frontier AI models against board-certified radiologists and radiology trainees. To mirror real-world usage, the reasoning modes of five popular frontier AI models were tested through their native web interfaces, viz. OpenAI o3, OpenAI GPT-5, Gemini 2.5 Pro, Grok-4, and Claude Opus 4.1. Accuracy was scored by blinded experts, and reproducibility was assessed across three independent runs. GPT-5 was additionally evaluated across various reasoning modes. Reasoning quality errors were assessed and a taxonomy of visual reasoning errors was defined. Board-certified radiologists achieved the highest diagnostic accuracy (83%), outperforming trainees (45%) and all AI models (best performance shown by GPT-5: 30%). Reliability was substantial for GPT-5 and o3, moderate for Gemini 2.5 Pro and Grok-4, and poor for Claude Opus 4.1. These findings demonstrate that advanced frontier models fall far short of radiologists in challenging diagnostic cases. Our benchmark highlights the present limitations of generalist AI in medical imaging and cautions against unsupervised clinical use. We also provide a qualitative analysis of reasoning traces and propose a practical taxonomy of visual reasoning errors by AI models for better understanding their failure modes, informing evaluation standards and guiding more robust model development.

Paperid: 5334, https://arxiv.org/pdf/2509.25202.pdf

Abstract:
Jigsaw puzzle solving remains challenging in computer vision, requiring an understanding of both local fragment details and global spatial relationships. While most traditional approaches only focus on visual cues like edge matching and visual coherence, few methods explore natural language descriptions for semantic guidance in challenging scenarios, especially for eroded gap puzzles. We propose a vision-language framework that leverages textual context to enhance puzzle assembly performance. Our approach centers on the Vision-Language Hierarchical Semantic Alignment (VLHSA) module, which aligns visual patches with textual descriptions through multi-level semantic matching from local tokens to global context. Also, a multimodal architecture that combines dual visual encoders with language features for cross-modal reasoning is integrated into this module. Experiments demonstrate that our method significantly outperforms state-of-the-art models across various datasets, achieving substantial improvements, including a 14.2 percentage point gain in piece accuracy. Ablation studies confirm the critical role of the VLHSA module in driving improvements over vision-only approaches. Our work establishes a new paradigm for jigsaw puzzle solving by incorporating multimodal semantic insights.

Paperid: 5335, https://arxiv.org/pdf/2509.23927.pdf

Abstract:
Cross-modal artificial intelligence has garnered widespread attention in recent years, achieving significant progress in the study of natural images. However, existing methods are mostly designed for RGB imagery, leaving a significant gap in modeling synthetic aperture radar (SAR) imagery. SAR, with its all-day, all-weather imaging capabilities, plays an irreplaceable role in remote sensing scene understanding. To address this gap, this paper proposes SAR-KnowLIP, the first universal SAR multimodal foundational model, along with reusable data and evaluation baselines. Specifically: (1) This work introduces the critical yet long-overlooked attribute of geographic information into remote sensing research, constructing SAR-GEOVL-1M (the first large-scale SAR dataset with complete geographic projection properties), covering multiple satellite platforms, 120,000 images, and 135 cities. (2) Aligned structured text is generated through a hierarchical cognitive chain-of-thought (HCoT), providing more than one million multi-dimensional semantic annotations of landforms, regional functions, target attributes, and spatial relationships. (3) We design a Self-Consistent Iterative Optimization mechanism that continuously enhances cross-modal alignment through a self-supervised closed loop of contrastive, matching, and reconstruction learning on a transferable multimodal encoder. (4) A unified evaluation benchmark is established across 11 representative downstream vision and vision-language tasks, with comparisons against 14 leading foundation models, where SAR-KnowLIP demonstrates leading performance, particularly in object counting and land-cover classification. We expect that SAR-KnowLIP's large-scale multimodal data, transferable model architecture, and comprehensive experimental benchmark will significantly advance the development of SAR multimodal baseline models.

Paperid: 5336, https://arxiv.org/pdf/2509.23828.pdf

Abstract:
Vision-language models (VLMs) have demonstrated strong performance in 2D scene understanding and generation, but extending this unification to the physical world remains an open challenge. Existing 3D and 4D approaches typically embed scene geometry into autoregressive model for semantic understanding and diffusion model for content generation. This paradigm gap prevents a single model from jointly handling both tasks, especially in dynamic 4D settings where spatiotemporal modeling is critical. We propose Uni4D-LLM, the first unified VLM framework with spatiotemporal awareness for 4D scene understanding and generation. Our design is guided by two key insights: 1) Unification requires a shared representation. We extract semantic features for understanding and noisy-injected appearance features for generation, incorporate 4D geometric cues, and fuse them into a spatiotemporal-aware visual representation through adaptive cross-attention. 2) Unification requires a shared architecture. Both autoregression and diffusion are built on Transformer backbones, and this enables integration into a single LLM with task-specific heads. By aligning visual and linguistic representations, our Uni4D-LLM produces predictions for both understanding and generation within one Transformer-based framework. We further apply instruction fine-tuning on diverse 4D vision-language datasets to improve generalization across tasks. Extensive experiments on multiple benchmarks demonstrate that Uni4D-LLM achieves competitive or superior results compared to state-of-the-art models and offers the first true unification of 4D scene understanding and generation.

Paperid: 5337, https://arxiv.org/pdf/2509.23781.pdf

Abstract:
Parameter-efficient fine-tuning (PEFT) of vision-language models (VLMs) excels in various vision tasks thanks to the rich knowledge and generalization ability of VLMs. However, recent studies revealed that such fine-tuned VLMs are vulnerable to spurious correlations stemming from the subgroup imbalance in the fine-tuning datasets. To resolve this issue, we propose Group Context Optimization (GroupCoOp), a simple and effective debiased fine-tuning algorithm that enhances the group robustness of fine-tuned VLMs. Its key idea is to employ group-specific text prompts as group representatives serving as multiple classifiers for their target class. The rich semantic knowledge of the text encoder of VLM enables the discovery of effective group prompts even for groups with a small number of training samples. Leveraging the group prompts for each class addresses the issues caused by the group-imbalanced training set, such as the neglect of minority groups and the scattered distribution of each class in the embedding space. GroupCoOp achieved the best results on five benchmarks across five CLIP architectures and occasionally outperformed prior methods that fine-tune the entire network, despite training only 0.016\% of the network's parameters.

Paperid: 5338, https://arxiv.org/pdf/2509.23746.pdf

Abstract:
Visual pointing, which aims to localize a target by predicting its coordinates on an image, has emerged as an important problem in the realm of vision-language models (VLMs). Despite its broad applicability, recent benchmarks show that current VLMs still fall far behind human performance on this task. A key limitation is that VLMs are typically required to complete the pointing task in a single step, akin to asking humans to point at an object without seeing their own fingers. To address this issue, we propose a simple yet effective self-refining procedure: Point, Visualize, then Refine (Poivre). This procedure enables a VLM to first mark its estimated point, then iteratively refine the coordinates if necessary. Inspired by advances of reasoning models in the natural language domain, we employ reinforcement learning (RL) to incentivize this self-refining ability. For the RL training, we design a neat process reward that is not only empirically effective but also grounded in appealing properties. Our trained model, Poivre-7B, sets a new state of the art on Point-Bench, outperforming both proprietary models such as Gemini-2.5-Pro and large open-source models such as Molmo-72B by over 3%. To support future research, we release our training and inference code, dataset, and the Poivre-7B checkpoint.

Paperid: 5339, https://arxiv.org/pdf/2509.23121.pdf

Abstract:
The application of artificial intelligence (AI) in industry is accelerating the shift from traditional automation to intelligent systems with perception and cognition. Vision language-action (VLA) models have been a key paradigm in AI to unify perception, reasoning, and control. Has the performance of the VLA models met the industrial requirements? In this paper, from the perspective of industrial deployment, we compare the performance of existing state-of-the-art VLA models in industrial scenarios and analyze the limitations of VLA models for real-world industrial deployment from the perspectives of data collection and model architecture. The results show that the VLA models retain their ability to perform simple grasping tasks even in industrial settings after fine-tuning. However, there is much room for performance improvement in complex industrial environments, diverse object categories, and high precision placing tasks. Our findings provide practical insight into the adaptability of VLA models for industrial use and highlight the need for task-specific enhancements to improve their robustness, generalization, and precision.

Paperid: 5340, https://arxiv.org/pdf/2509.22615.pdf

Abstract:
Modern vision language pipelines are driven by RGB vision encoders trained on massive image text corpora. While these pipelines have enabled impressive zero shot capabilities and strong transfer across tasks, they still inherit two structural inefficiencies from the pixel domain: (i) transmitting dense RGB images from edge devices to the cloud is energy intensive and costly, and (ii) patch based tokenization explodes sequence length, stressing attention budgets and context limits. We explore 2D Gaussian Splatting (2DGS) as an alternative visual substrate for alignment: a compact, spatially adaptive representation that parameterizes images by a set of colored anisotropic Gaussians. We develop a scalable 2DGS pipeline with structured initialization, luminance aware pruning, and batched CUDA kernels, achieving over 90x faster fitting and about 97% GPU utilization compared to prior implementations. We further adapt contrastive language image pretraining (CLIP) to 2DGS by reusing a frozen RGB-based transformer backbone with a lightweight splat aware input stem and a perceiver resampler, training only about 7% of the total parameters. On large DataComp subsets, GS encoders yield meaningful zero shot ImageNet-1K performance while compressing inputs 3 to 20x relative to pixels. While accuracy currently trails RGB encoders, our results establish 2DGS as a viable multimodal substrate, pinpoint architectural bottlenecks, and open a path toward representations that are both semantically powerful and transmission efficient for edge cloud learning.

Paperid: 5341, https://arxiv.org/pdf/2509.21677.pdf

Abstract:
We present Prophecy, a tool for automatically inferring formal properties of feed-forward neural networks. Prophecy is based on the observation that a significant part of the logic of feed-forward networks is captured in the activation status of the neurons at inner layers. Prophecy works by extracting rules based on neuron activations (values or on/off statuses) as preconditions that imply certain desirable output property, e.g., the prediction being a certain class. These rules represent network properties captured in the hidden layers that imply the desired output behavior. We present the architecture of the tool, highlight its features and demonstrate its usage on different types of models and output properties. We present an overview of its applications, such as inferring and proving formal explanations of neural networks, compositional verification, run-time monitoring, repair, and others. We also show novel results highlighting its potential in the era of large vision-language models.

Paperid: 5342, https://arxiv.org/pdf/2509.21287.pdf

Abstract:
Recent vision-language models excel at large-scale image-text alignment but often neglect the compositional structure of language, leading to failures on tasks that hinge on word order and predicate-argument structure. We introduce DisCoCLIP, a multimodal encoder that combines a frozen CLIP vision transformer with a novel tensor network text encoder that explicitly encodes syntactic structure. Sentences are parsed with a Combinatory Categorial Grammar parser to yield distributional word tensors whose contractions mirror the sentence's grammatical derivation. To keep the model efficient, high-order tensors are factorized with tensor decompositions, reducing parameter count from tens of millions to under one million. Trained end-to-end with a self-supervised contrastive loss, DisCoCLIP markedly improves sensitivity to verb semantics and word order: it raises CLIP's SVO-Probes verb accuracy from 77.6% to 82.4%, boosts ARO attribution and relation scores by over 9% and 4%, and achieves 93.7% on a newly introduced SVO-Swap benchmark. These results demonstrate that embedding explicit linguistic structure via tensor networks yields interpretable, parameter-efficient representations that substantially improve compositional reasoning in vision-language tasks.

Paperid: 5343, https://arxiv.org/pdf/2509.21173.pdf

Abstract:
The powerful zero-shot generalization capabilities of vision-language models (VLMs) like CLIP have enabled new paradigms for safety-related tasks such as out-of-distribution (OOD) detection. However, additional aspects crucial for the computationally efficient and reliable deployment of CLIP are still overlooked. In particular, the impact of quantization on CLIP's performance beyond accuracy remains underexplored. This work presents a large-scale evaluation of quantization on CLIP models, assessing not only in-distribution accuracy but a comprehensive suite of reliability metrics and revealing counterintuitive results driven by pre-training source. We demonstrate that quantization consistently improves calibration for typically underconfident pre-trained models, while often degrading it for overconfident variants. Intriguingly, this degradation in calibration does not preclude gains in other reliability metrics; we find that OOD detection can still improve for these same poorly calibrated models. Furthermore, we identify specific quantization-aware training (QAT) methods that yield simultaneous gains in zero-shot accuracy, calibration, and OOD robustness, challenging the view of a strict efficiency-performance trade-off. These findings offer critical insights for navigating the multi-objective problem of deploying efficient, reliable, and robust VLMs by utilizing quantization beyond its conventional role.

Paperid: 5344, https://arxiv.org/pdf/2509.19207.pdf

Abstract:
Contrastive vision-language models (VLMs) have made significant progress in binding visual and textual information, but understanding long, dense captions remains an open challenge. We hypothesize that compositionality, the capacity to reason about object-attribute bindings and inter-object relationships, is key to understanding longer captions. In this paper, we investigate the interaction between compositionality and long-caption understanding, asking whether training for one property enhances the other. We train and evaluate a range of models that target each of these capabilities. Our results reveal a bidirectional relationship: compositional training improves performance on long-caption retrieval, and training on long captions promotes compositionality. However, these gains are sensitive to data quality and model design. We find that training on poorly structured captions, or with limited parameter updates, fails to support generalization. Likewise, strategies that aim at retaining general alignment, such as freezing positional embeddings, do not improve compositional understanding. Overall, we find that compositional understanding and long-caption understanding are intertwined capabilities that can be jointly learned through training on dense, grounded descriptions. Despite these challenges, we show that models trained on high-quality, long-caption data can achieve strong performance in both tasks, offering practical guidance for improving VLM generalization.

Paperid: 5345, https://arxiv.org/pdf/2509.18733.pdf

Abstract:
Current visual foundation models (VFMs) face a fundamental limitation in transferring knowledge from vision language models (VLMs), while VLMs excel at modeling cross-modal interactions through unified representation spaces, existing VFMs predominantly adopt result-oriented paradigms that neglect the underlying interaction processes. This representational discrepancy hinders effective knowledge transfer and limits generalization across diverse vision tasks. We propose Learning from Interactions (LFI), a cognitive-inspired framework that addresses this gap by explicitly modeling visual understanding as an interactive process. Our key insight is that capturing the dynamic interaction patterns encoded in pre-trained VLMs enables more faithful and efficient knowledge transfer to VFMs. The approach centers on two technical innovations, Interaction Queries, which maintain persistent relational structures across network layers, and interaction-based supervision, derived from the cross-modal attention mechanisms of VLMs. Comprehensive experiments demonstrate consistent improvements across multiple benchmarks, achieving 3.3 and 1.6mAP/2.4AP absolute gains on TinyImageNet classification and COCO detection/segmentation respectively, with minimal parameter overhead and faster convergence. The framework particularly excels in cross-domain settings, delivering 2.4 and 9.3 zero-shot improvements on PACS and VLCS. Human evaluations further confirm its cognitive alignment, outperforming result-oriented methods by 2.7 times in semantic consistency metrics.

Paperid: 5346, https://arxiv.org/pdf/2509.18425.pdf

Abstract:
Vision language models (VLMs) show strong results on chart understanding, yet existing benchmarks assume clean figures and fact based queries. Real world charts often contain distortions and demand reasoning beyond simple matching. We evaluate ChatGPT 4o, Claude Sonnet 4, and Gemini 2.5 Pro, finding sharp performance drops under corruption or occlusion, with hallucinations such as value fabrication, trend misinterpretation, and entity confusion becoming more frequent. Models remain overconfident in degraded settings, generating plausible but unsupported explanations. To address this gap, we introduce CHART NOISe(Chart Hallucinations, Answers, and Reasoning Testing on Noisy and Occluded Input Selections), a dataset combining chart corruptions, occlusions, and exam style multiple choice questions inspired by Korea's CSAT English section. A key innovation is prompt reverse inconsistency, where models contradict themselves when asked to confirm versus deny the same statement. Our contributions are threefold: (1) benchmarking state of the art VLMs, exposing systematic vulnerabilities in chart reasoning; (2) releasing CHART NOISe, the first dataset unifying corruption, occlusion, and reverse inconsistency; and (3) proposing baseline mitigation strategies such as quality filtering and occlusion detection. Together, these efforts establish a rigorous testbed for advancing robustness and reliability in chart understanding.

Paperid: 5347, https://arxiv.org/pdf/2509.18174.pdf

Abstract:
Arabic document OCR remains a challenging task due to the language's cursive script, diverse fonts, diacritics, and right-to-left orientation. While modern Multimodal Large Language Models (MLLMs) have advanced document understanding for high-resource languages, their performance on Arabic remains limited. In this work, we introduce Baseer, a vision-language model fine-tuned specifically for Arabic document OCR. Leveraging a large-scale dataset combining synthetic and real-world documents, Baseer is trained using a decoder-only fine-tuning strategy to adapt a pre-trained MLLM while preserving general visual features. We also present Misraj-DocOCR, a high-quality, expert-verified benchmark designed for rigorous evaluation of Arabic OCR systems. Our experiments show that Baseer significantly outperforms existing open-source and commercial solutions, achieving a WER of 0.25 and establishing a new state-of-the-art in the domain of Arabic document OCR. Our results highlight the benefits of domain-specific adaptation of general-purpose MLLMs and establish a strong baseline for high-accuracy OCR on morphologically rich languages like Arabic.

Paperid: 5348, https://arxiv.org/pdf/2509.17418.pdf

Abstract:
Spelling correction from visual input poses unique challenges for vision language models (VLMs), as it requires not only detecting but also correcting textual errors directly within images. We present ReViCo (Real Visual Correction), the first benchmark that systematically evaluates VLMs on real-world visual spelling correction across Chinese and English. ReViCo contains naturally occurring errors collected from real-world image data and supports fine-grained evaluation at both image and token levels. Through comprehensive experiments on representative cascaded (Qwen) and native (InternVL) open-source models, as well as closed-source systems (GPT-4o, Claude), we show that current VLMs fall significantly short of human performance, particularly in correction. To address these limitations, we explore two solution paradigms: a Joint OCR-Correction pipeline and a Background Information enhanced approach, both of which yield consistent performance gains. Our analysis highlights fundamental limitations of existing architectures and provides actionable insights for advancing multimodal spelling correction.

Paperid: 5349, https://arxiv.org/pdf/2509.16633.pdf

Abstract:
Large Vision-Language Models (L-VLMs) have demonstrated remarkable performance in various vision and language tasks, including visual question answering (VQA). However, their high computational cost makes them impractical for resource-constrained settings and inference-heavy applications. In contrast, Small Vision-Language Models (S-VLMs) offer efficiency but suffer from a significant performance gap compared to their larger counterparts. In this work, we introduce the Model Parity Aligner (MPA), a novel framework designed to systematically improve S-VLMs by leveraging unlabeled images and effective knowledge transfer from L-VLMs. Instead of traditional knowledge distillation methods that rely on labeled training data, MPA employs a strategic parity-based approach that precisely identifies the knowledge disparities between S-VLMs and L-VLMs, and optimizes training by targeting only these disparities. We conduct extensive experiments on four diverse VQA benchmarks, namely TextVQA, ST-VQA, ChartQA, and OKVQA, each of which requires specialized reasoning capabilities such as text recognition, chart interpretation, and commonsense and factual understanding. Our results demonstrate that MPA consistently enhances the performance of S-VLMs on all benchmarks, reducing the performance gap while maintaining computational efficiency. We make our code publicly available.

Paperid: 5350, https://arxiv.org/pdf/2509.15076.pdf

Abstract:
Air pollution remains a critical threat to public health and environmental sustainability, yet conventional monitoring systems are often constrained by limited spatial coverage and accessibility. This paper proposes an AI-driven agent that predicts ambient air pollution levels from sky images and synthesizes realistic visualizations of pollution scenarios using generative modeling. Our approach combines statistical texture analysis with supervised learning for pollution classification, and leverages vision-language model (VLM)-guided image generation to produce interpretable representations of air quality conditions. The generated visuals simulate varying degrees of pollution, offering a foundation for user-facing interfaces that improve transparency and support informed environmental decision-making. These outputs can be seamlessly integrated into intelligent applications aimed at enhancing situational awareness and encouraging behavioral responses based on real-time forecasts. We validate our method using a dataset of urban sky images and demonstrate its effectiveness in both pollution level estimation and semantically consistent visual synthesis. The system design further incorporates human-centered user experience principles to ensure accessibility, clarity, and public engagement in air quality forecasting. To support scalable and energy-efficient deployment, future iterations will incorporate a green CNN architecture enhanced with FPGA-based incremental learning, enabling real-time inference on edge platforms.

Paperid: 5351, https://arxiv.org/pdf/2509.14138.pdf

Abstract:
Long-horizon robotic manipulation tasks require executing multiple interdependent subtasks in strict sequence, where errors in detecting subtask completion can cascade into downstream failures. Existing Vision-Language-Action (VLA) models such as $Ï_0$ excel at continuous low-level control but lack an internal signal for identifying when a subtask has finished, making them brittle in sequential settings. We propose SeqVLA, a completion-aware extension of $Ï_0$ that augments the base architecture with a lightweight detection head perceiving whether the current subtask is complete. This dual-head design enables SeqVLA not only to generate manipulation actions but also to autonomously trigger transitions between subtasks. We investigate four finetuning strategies that vary in how the action and detection heads are optimized (joint vs. sequential finetuning) and how pretrained knowledge is preserved (full finetuning vs. frozen backbone). Experiments are performed on two multi-stage tasks: salad packing with seven distinct subtasks and candy packing with four distinct subtasks. Results show that SeqVLA significantly outperforms the baseline $Ï_0$ and other strong baselines in overall success rate. In particular, joint finetuning with an unfrozen backbone yields the most decisive and statistically reliable completion predictions, eliminating sequence-related failures and enabling robust long-horizon execution. Our results highlight the importance of coupling action generation with subtask-aware detection for scalable sequential manipulation.

Paperid: 5352, https://arxiv.org/pdf/2509.11620.pdf

Abstract:
Multimodal Large Language Models (MLLMs) are increasingly applied in Personalized Image Aesthetic Assessment (PIAA) as a scalable alternative to expert evaluations. However, their predictions may reflect subtle biases influenced by demographic factors such as gender, age, and education. In this work, we propose AesBiasBench, a benchmark designed to evaluate MLLMs along two complementary dimensions: (1) stereotype bias, quantified by measuring variations in aesthetic evaluations across demographic groups; and (2) alignment between model outputs and genuine human aesthetic preferences. Our benchmark covers three subtasks (Aesthetic Perception, Assessment, Empathy) and introduces structured metrics (IFD, NRD, AAS) to assess both bias and alignment. We evaluate 19 MLLMs, including proprietary models (e.g., GPT-4o, Claude-3.5-Sonnet) and open-source models (e.g., InternVL-2.5, Qwen2.5-VL). Results indicate that smaller models exhibit stronger stereotype biases, whereas larger models align more closely with human preferences. Incorporating identity information often exacerbates bias, particularly in emotional judgments. These findings underscore the importance of identity-aware evaluation frameworks in subjective vision-language tasks.

Paperid: 5353, https://arxiv.org/pdf/2509.09541.pdf

Abstract:
Compositional generalization is a key facet of human cognition, but lacking in current AI tools such as vision-language models. Previous work examined whether a compositional tensor-based sentence semantics can overcome the challenge, but led to negative results. We conjecture that the increased training efficiency of quantum models will improve performance in these tasks. We interpret the representations of compositional tensor-based models in Hilbert spaces and train Variational Quantum Circuits to learn these representations on an image captioning task requiring compositional generalization. We used two image encoding techniques: a multi-hot encoding (MHE) on binary image vectors and an angle/amplitude encoding on image vectors taken from the vision-language model CLIP. We achieve good proof-of-concept results using noisy MHE encodings. Performance on CLIP image vectors was more mixed, but still outperformed classical compositional models.

Paperid: 5354, https://arxiv.org/pdf/2509.08490.pdf

Abstract:
Underwater object detection (UOD) is vital to diverse marine applications, including oceanographic research, underwater robotics, and marine conservation. However, UOD faces numerous challenges that compromise its performance. Over the years, various methods have been proposed to address these issues, but they often fail to fully capture the complexities of underwater environments. This review systematically categorizes UOD challenges into five key areas: Image quality degradation, target-related issues, data-related challenges, computational and processing constraints, and limitations in detection methodologies. To address these challenges, we analyze the progression from traditional image processing and object detection techniques to modern approaches. Additionally, we explore the potential of large vision-language models (LVLMs) in UOD, leveraging their multi-modal capabilities demonstrated in other domains. We also present case studies, including synthetic dataset generation using DALL-E 3 and fine-tuning Florence-2 LVLM for UOD. This review identifies three key insights: (i) Current UOD methods are insufficient to fully address challenges like image degradation and small object detection in dynamic underwater environments. (ii) Synthetic data generation using LVLMs shows potential for augmenting datasets but requires further refinement to ensure realism and applicability. (iii) LVLMs hold significant promise for UOD, but their real-time application remains under-explored, requiring further research on optimization techniques.

Paperid: 5355, https://arxiv.org/pdf/2509.06031.pdf

Abstract:
We present ZLATTE, a geometry-aware, learning-free framework for language-driven trajectory reshaping in human-robot interaction. Unlike prior learning-based methods, ZLATTE leverages Vision-Language Models to register objects as geometric primitives and employs a Large Language Model to translate natural language instructions into explicit geometric and kinematic constraints. These constraints are integrated into a potential field optimization to adapt initial trajectories while preserving feasibility and safety. A multi-agent strategy further enhances robustness under complex or conflicting commands. Simulation and real-world experiments demonstrate that ZLATTE achieves smoother, safer, and more interpretable trajectory modifications compared to state-of-the-art baselines.

Paperid: 5356, https://arxiv.org/pdf/2509.01554.pdf

Abstract:
General-purpose vision-language models (VLMs) have emerged as promising tools in radiology, offering zero-shot capabilities that mitigate the need for large labeled datasets. However, in high-stakes domains like diagnostic radiology, these models often lack the discriminative precision required for reliable clinical use. This challenge is compounded by the scarcity and heterogeneity of publicly available volumetric CT datasets, which vary widely in annotation formats and granularity. To address these limitations, we introduce Uniferum, a volumetric VLM that unifies diverse supervision signals, encoded in classification labels and segmentation masks, into a single training framework. By harmonizing three public 3D CT datasets with distinct annotations, Uniferum achieves state-of-the-art performance, improving AUROC on the CT-RATE benchmark by 7% compared to CLIP-based and conventional multi-label convolutional models. The model demonstrates robust out-of-distribution generalization, with observed evidence of unexpected zero-shot performance on the RAD-CHEST and INSPECT datasets. Our results highlight the effectiveness of integrating heterogeneous annotations and body segmentation to enhance model performance, setting a new direction for clinically reliable, data-efficient VLMs in 3D medical imaging.

Paperid: 5357, https://arxiv.org/pdf/2509.01209.pdf

Abstract:
Scene Graph Generation (SGG) encodes visual relationships between objects in images as graph structures. Thanks to the advances of Vision-Language Models (VLMs), the task of Open-Vocabulary SGG has been recently proposed where models are evaluated on their functionality to learn a wide and diverse range of relations. Current benchmarks in SGG, however, possess a very limited vocabulary, making the evaluation of open-source models inefficient. In this paper, we propose a new reference-free metric to fairly evaluate the open-vocabulary capabilities of VLMs for relation prediction. Another limitation of Open-Vocabulary SGG is the reliance on weakly supervised data of poor quality for pre-training. We also propose a new solution for quickly generating high-quality synthetic data through region-specific prompt tuning of VLMs. Experimental results show that pre-training with this new data split can benefit the generalization capabilities of Open-Voc SGG models.

Paperid: 5358, https://arxiv.org/pdf/2508.20279.pdf

Abstract:
Multimodal Large Language Models (MLLMs) have demonstrated strong performance across a wide range of vision-language tasks, yet their internal processing dynamics remain underexplored. In this work, we introduce a probing framework to systematically analyze how MLLMs process visual and textual inputs across layers. We train linear classifiers to predict fine-grained visual categories (e.g., dog breeds) from token embeddings extracted at each layer, using a standardized anchor question. To uncover the functional roles of different layers, we evaluate these probes under three types of controlled prompt variations: (1) lexical variants that test sensitivity to surface-level changes, (2) semantic negation variants that flip the expected answer by modifying the visual concept in the prompt, and (3) output format variants that preserve reasoning but alter the answer format. Applying our framework to LLaVA-1.5, LLaVA-Next-LLaMA-3, and Qwen2-VL, we identify a consistent stage-wise structure in which early layers perform visual grounding, middle layers support lexical integration and semantic reasoning, and final layers prepare task-specific outputs. We further show that while the overall stage-wise structure remains stable across variations in visual tokenization, instruction tuning data, and pretraining corpus, the specific layer allocation to each stage shifts notably with changes in the base LLM architecture. Our findings provide a unified perspective on the layer-wise organization of MLLMs and offer a lightweight, model-agnostic approach for analyzing multimodal representation dynamics.

Paperid: 5359, https://arxiv.org/pdf/2508.13428.pdf

Abstract:
In this early study, we observe an Easy-Options Bias (EOB) issue in some multiple-choice Visual Question Answering (VQA) benchmarks such as MMStar, RealWorldQA, SEED-Bench, Next-QA, STAR benchmark and Video-MME. This bias allows vision-language models (VLMs) to select the correct answer using only the vision (V) and options (O) as inputs, without the need for the question (Q). Through grounding experiments, we attribute the bias to an imbalance in visual relevance: the correct answer typically aligns more closely with the visual contents than the negative options in feature space, creating a shortcut for VLMs to infer the answer via simply vision-option similarity matching. To fix this, we introduce GroundAttack, a toolkit that automatically generates hard negative options as visually plausible as the correct answer. We apply it to the NExT-QA and MMStar datasets, creating new EOB-free annotations. On these EOB-free annotations, current VLMs approach to random accuracies under (V+O) settings, and drop to non-saturated accuracies under (V+Q+O) settings, providing a more realistic evaluation of VLMs' QA ability. Codes and new annotations will be released soon.

Paperid: 5360, https://arxiv.org/pdf/2508.12211.pdf

Abstract:
Pre-trained vision-language-action (VLA) models offer a promising foundation for generalist robot policies, but often produce brittle behaviours or unsafe failures when deployed zero-shot in out-of-distribution scenarios. We present Vision-Language-Action Planning & Search (VLAPS) -- a novel framework and accompanying algorithms that embed model-based search into the inference procedure of pre-trained VLA policies to improve their performance on robotic tasks. Specifically, our method biases a modified Monte Carlo Tree Search (MCTS) algorithm -- run using a model of the target environment -- using action priors defined by the VLA policy. By using VLA-derived abstractions and priors in model-based search, VLAPS efficiently explores language-conditioned robotics tasks whose search spaces would otherwise be intractably large. Conversely, by integrating model-based search with the VLA policy's inference procedure, VLAPS yields behaviours that are more performant than those obtained by directly following the VLA policy's action predictions. VLAPS offers a principled framework to: i) control test-time compute in VLA models, ii) leverage a priori knowledge of the robotic environment, and iii) integrate established planning and reinforcement learning techniques into the VLA inference process. Across all experiments, VLAPS significantly outperforms VLA-only baselines on language-specified tasks that would otherwise be intractable for uninformed search algorithms, increasing success rates by as much as 67 percentage points.

Paperid: 5361, https://arxiv.org/pdf/2508.10116.pdf

Abstract:
Item information, such as titles and attributes, is essential for effective user engagement in e-commerce. However, manual or semi-manual entry of structured item specifics often produces inconsistent quality, errors, and slow turnaround, especially for Customer-to-Customer sellers. Generating accurate descriptions directly from item images offers a promising alternative. Existing retrieval-based solutions address some of these issues but often miss fine-grained visual details and struggle with niche or specialized categories. We propose Optimized Preference-Based AI for Listings (OPAL), a framework for generating schema-compliant, high-quality item descriptions from images using a fine-tuned multimodal large language model (MLLM). OPAL addresses key challenges in multimodal e-commerce applications, including bridging modality gaps and capturing detailed contextual information. It introduces two data refinement methods: MLLM-Assisted Conformity Enhancement, which ensures alignment with structured schema requirements, and LLM-Assisted Contextual Understanding, which improves the capture of nuanced and fine-grained information from visual inputs. OPAL uses visual instruction tuning combined with direct preference optimization to fine-tune the MLLM, reducing hallucinations and improving robustness across different backbone architectures. We evaluate OPAL on real-world e-commerce datasets, showing that it consistently outperforms baseline methods in both description quality and schema completion rates. These results demonstrate that OPAL effectively bridges the gap between visual and textual modalities, delivering richer, more accurate, and more consistent item descriptions. This work advances automated listing optimization and supports scalable, high-quality content generation in e-commerce platforms.

Paperid: 5362, https://arxiv.org/pdf/2508.09818.pdf

Abstract:
This study investigates how large language models (LLMs) can be used to understand human behavior using motion and video data. We think that mixing both types is essential to completely capture the nuanced movements and meanings of human actions, in contrast to recent models that simply concentrate on motion data or films. To address this, we provide ViMoNet, a straightforward yet effective framework for comprehending, characterizing, and deducing human action. ViMoNet employs a joint training strategy that leverages the advantages of two data types: detailed motion-text data, which is more exact, and generic video-text data, which is more comprehensive but less detailed. This aids in the model's acquisition of rich data regarding time and space in human behavior. Additionally, we provide a brand new dataset named VIMOS that contains a variety of films, motion sequences, instructions, and subtitles. We developed ViMoNet-Bench, a standardized benchmark with carefully labeled samples, to evaluate how well models understand human behavior. Our tests show that ViMoNet outperforms existing methods in caption generation, motion understanding, and behavior interpretation.

Paperid: 5363, https://arxiv.org/pdf/2508.07819.pdf

Abstract:
Pre-trained Vision-Language Models (VLMs) struggle with Zero-Shot Anomaly Detection (ZSAD) due to a critical adaptation gap: they lack the local inductive biases required for dense prediction and employ inflexible feature fusion paradigms. We address these limitations through an Architectural Co-Design framework that jointly refines feature representation and cross-modal fusion. Our method proposes a parameter-efficient Convolutional Low-Rank Adaptation (Conv-LoRA) adapter to inject local inductive biases for fine-grained representation, and introduces a Dynamic Fusion Gateway (DFG) that leverages visual context to adaptively modulate text prompts, enabling a powerful bidirectional fusion. Extensive experiments on diverse industrial and medical benchmarks demonstrate superior accuracy and robustness, validating that this synergistic co-design is critical for robustly adapting foundation models to dense perception tasks.

Paperid: 5364, https://arxiv.org/pdf/2508.06937.pdf

Abstract:
Recent advances in text-to-image (T2I) models have enabled training-free regional image editing by leveraging the generative priors of foundation models. However, existing methods struggle to balance text adherence in edited regions, context fidelity in unedited areas, and seamless integration of edits. We introduce CannyEdit, a novel training-free framework that addresses this trilemma through two key innovations. First, Selective Canny Control applies structural guidance from a Canny ControlNet only to the unedited regions, preserving the original image's details while allowing for precise, text-driven changes in the specified editable area. Second, Dual-Prompt Guidance utilizes both a local prompt for the specific edit and a global prompt for overall scene coherence. Through this synergistic approach, these components enable controllable local editing for object addition, replacement, and removal, achieving a superior trade-off among text adherence, context fidelity, and editing seamlessness compared to current region-based methods. Beyond this, CannyEdit offers exceptional flexibility: it operates effectively with rough masks or even single-point hints in addition tasks. Furthermore, the framework can seamlessly integrate with vision-language models in a training-free manner for complex instruction-based editing that requires planning and reasoning. Our extensive evaluations demonstrate CannyEdit's strong performance against leading instruction-based editors in complex object addition scenarios.

Paperid: 5365, https://arxiv.org/pdf/2508.06547.pdf

Abstract:
Traditional robotic systems typically decompose intelligence into independent modules for computer vision, natural language processing, and motion control. Vision-Language-Action (VLA) models fundamentally transform this approach by employing a single neural network that can simultaneously process visual observations, understand human instructions, and directly output robot actions -- all within a unified framework. However, these systems are highly dependent on high-quality training datasets that can capture the complex relationships between visual observations, language instructions, and robotic actions. This tutorial reviews three representative systems: the PyBullet simulation framework for flexible customized data generation, the LIBERO benchmark suite for standardized task definition and evaluation, and the RT-X dataset collection for large-scale multi-robot data acquisition. We demonstrated dataset generation approaches in PyBullet simulation and customized data collection within LIBERO, and provide an overview of the characteristics and roles of the RT-X dataset for large-scale multi-robot data acquisition.

Paperid: 5366, https://arxiv.org/pdf/2508.00378.pdf

Abstract:
Chain-of-Thought (CoT) prompting has shown promise in improving reasoning in vision-language models (VLMs), but it often produces explanations that are linguistically fluent yet lack grounding in visual content. We observe that such hallucinations arise in part from the absence of an explicit verification mechanism during multi-step reasoning. To address this, we propose \textbf{CoRGI}(\textbf{C}hain \textbf{o}f \textbf{R}easoning with \textbf{G}rounded \textbf{I}nsights), a modular framework that introduces visual verification into the reasoning process. CoRGI follows a three-stage pipeline: it first generates a textual reasoning chain, then extracts supporting visual evidence for each reasoning step via a dedicated module (VEVM), and finally synthesizes the textual rationale with visual evidence to generate a grounded, verified answer. The framework can be integrated with existing VLMs without end-to-end retraining. We evaluate CoRGI on the VCR benchmark and find that it improves reasoning performance on two representative open-source VLM backbones, Qwen-2.5VL and LLaVA-1.6. Ablation studies confirm the contribution of each step in the verification module, and human evaluations suggest that CoRGI leads to more factual and helpful explanations. We also examine alternative designs for the visual verification step and discuss potential limitations of post-hoc verification frameworks. These findings highlight the importance of grounding intermediate reasoning steps in visual evidence to enhance the robustness of multimodal reasoning.

Paperid: 5367, https://arxiv.org/pdf/2507.22075.pdf

Abstract:
In unsupervised adaptation for vision-language models such as CLIP, pseudo-labels derived from zero-shot predictions often exhibit significant noise, particularly under domain shifts or in visually complex scenarios. Conventional pseudo-label filtering approaches, which rely on fixed confidence thresholds, tend to be unreliable in fully unsupervised settings. In this work, we propose a novel adaptive pseudo-labeling framework that enhances CLIP's adaptation performance by integrating prototype consistency and neighborhood-based consistency. The proposed method comprises two key components: PICS, which assesses pseudo-label accuracy based on in-class feature compactness and cross-class feature separation; and NALR, which exploits semantic similarities among neighboring samples to refine pseudo-labels dynamically. Additionally, we introduce an adaptive weighting mechanism that adjusts the influence of pseudo-labeled samples during training according to their estimated correctness. Extensive experiments on 11 benchmark datasets demonstrate that our method achieves state-of-the-art performance in unsupervised adaptation scenarios, delivering more accurate pseudo-labels while maintaining computational efficiency.

Paperid: 5368, https://arxiv.org/pdf/2507.20397.pdf

Abstract:
Data collection for autonomous driving is rapidly accelerating, but manual annotation, especially for 3D labels, remains a major bottleneck due to its high cost and labor intensity. Autolabeling has emerged as a scalable alternative, allowing the generation of labels for point clouds with minimal human intervention. While LiDAR-based autolabeling methods leverage geometric information, they struggle with inherent limitations of lidar data, such as sparsity, occlusions, and incomplete object observations. Furthermore, these methods typically operate in a class-agnostic manner, offering limited semantic granularity. To address these challenges, we introduce VESPA, a multimodal autolabeling pipeline that fuses the geometric precision of LiDAR with the semantic richness of camera images. Our approach leverages vision-language models (VLMs) to enable open-vocabulary object labeling and to refine detection quality directly in the point cloud domain. VESPA supports the discovery of novel categories and produces high-quality 3D pseudolabels without requiring ground-truth annotations or HD maps. On Nuscenes dataset, VESPA achieves an AP of 52.95% for object discovery and up to 46.54% for multiclass object detection, demonstrating strong performance in scalable 3D scene understanding. Code will be available upon acceptance.

Paperid: 5369, https://arxiv.org/pdf/2507.15882.pdf

Abstract:
The proliferation of multimodal Large Language Models has significantly advanced the ability to analyze and understand complex data inputs from different modalities. However, the processing of long documents remains under-explored, largely due to a lack of suitable benchmarks. To address this, we introduce Document Haystack, a comprehensive benchmark designed to evaluate the performance of Vision Language Models (VLMs) on long, visually complex documents. Document Haystack features documents ranging from 5 to 200 pages and strategically inserts pure text or multimodal text+image "needles" at various depths within the documents to challenge VLMs' retrieval capabilities. Comprising 400 document variants and a total of 8,250 questions, it is supported by an objective, automated evaluation framework. We detail the construction and characteristics of the Document Haystack dataset, present results from prominent VLMs and discuss potential research avenues in this area.

Paperid: 5370, https://arxiv.org/pdf/2507.09615.pdf

Abstract:
Vision-language models (VLMs) like CLIP excel in zero-shot learning by aligning image and text representations through contrastive pretraining. Existing approaches to unsupervised adaptation (UA) for fine-grained classification with VLMs either rely on fixed alignment scores that cannot capture evolving, subtle class distinctions or use computationally expensive pseudo-labeling strategies that limit scalability. In contrast, we show that modeling fine-grained cross-modal interactions during adaptation produces more accurate, class-discriminative pseudo-labels and substantially improves performance over state-of-the-art (SOTA) methods. We introduce Fine-grained Alignment and Interaction Refinement (FAIR), an innovative approach that dynamically aligns localized image features with descriptive language embeddings through a set of Class Description Anchors (CDA). This enables the definition of a Learned Alignment Score (LAS), which incorporates CDA as an adaptive classifier, facilitating cross-modal interactions to improve self-training in unsupervised adaptation. Furthermore, we propose a self-training weighting mechanism designed to refine pseudo-labels in the presence of inter-class ambiguities. Our approach, FAIR, delivers a substantial performance boost in fine-grained unsupervised adaptation, achieving a notable overall gain of 2.78% across 13 fine-grained datasets compared to SOTA methods.

Paperid: 5371, https://arxiv.org/pdf/2507.07985.pdf

Abstract:
Contrastive vision-language models like CLIP are used for a large variety of applications, such as zero-shot classification or as vision encoder for multi-modal models. Despite their popularity, their representations show major limitations. For instance, CLIP models learn bag-of-words representations and, as a consequence, fail to distinguish whether an image is of ``a yellow submarine and a blue bus'' or ``a blue submarine and a yellow bus''. Previous attempts to fix this issue added hard negatives during training or modified the architecture, but failed to resolve the problem in its entirety. We suspect that the missing insights to solve the binding problem for CLIP are hidden in arguably the most important part of learning algorithms: the data. In this work, we fill this gap by rigorously identifying the influence of data properties on CLIP's ability to learn binding using a synthetic dataset. We find that common properties of natural data such as low attribute density, incomplete captions, and the saliency bias, a tendency of human captioners to describe the object that is ``most salient'' to them, have a detrimental effect on binding performance. In contrast to common belief, we find that neither scaling the batch size, i.e., implicitly adding more hard negatives, nor explicitly creating hard negatives enables CLIP to learn reliable binding. Only when the data expresses our identified data properties does CLIP learn almost perfect binding.

Paperid: 5372, https://arxiv.org/pdf/2507.07137.pdf

Abstract:
Machine unlearning (MU) is a promising cost-effective method to cleanse undesired information (generated concepts, biases, or patterns) from foundational diffusion models. While MU is orders of magnitude less costly than retraining a diffusion model without the undesired information, it can be challenging and labor-intensive to prove that the information has been fully removed from the model. Moreover, MU can damage diffusion model performance on surrounding concepts that one would like to retain, making it unclear if the diffusion model is still fit for deployment. We introduce autoeval-dmun, an automated tool which leverages (vision-) language models to thoroughly assess unlearning in diffusion models. Given a target concept, autoeval-dmun extracts structured, relevant world knowledge from the language model to identify nearby concepts which are likely damaged by unlearning and to circumvent unlearning with adversarial prompts. We use our automated tool to evaluate popular diffusion model unlearning methods, revealing that language models (1) impose semantic orderings of nearby concepts which correlate well with unlearning damage and (2) effectively circumvent unlearning with synthetic adversarial prompts.

Paperid: 5373, https://arxiv.org/pdf/2507.04351.pdf

Abstract:
Choosing the right fabric is crucial to meet functional and quality requirements in robotic applications for textile manufacturing, apparel production, and smart retail. We present MLLM-Fabric, a robotic framework powered by multimodal large language models (MLLMs) for fabric sorting and selection. The system includes a robotic arm, a camera, a visuotactile sensor, and a pressure sensor. It employs supervised fine-tuning and multimodal explanation-guided knowledge distillation to accurately classify and rank fabric properties. To facilitate further research, we release a dataset of 220 unique fabric samples, including RGB images and synchronized visuotactile and pressure data. Experimental results show that our Fabric-Llama-90B model consistently outperforms pretrained vision-language baselines in both property ranking accuracy and selection reliability.

Paperid: 5374, https://arxiv.org/pdf/2506.23122.pdf

Abstract:
This work investigates the challenging task of identifying narrative roles - Hero, Villain, Victim, and Other - in Internet memes, across three diverse test sets spanning English and code-mixed (English-Hindi) languages. Building on an annotated dataset originally skewed toward the 'Other' class, we explore a more balanced and linguistically diverse extension, originally introduced as part of the CLEF 2024 shared task. Comprehensive lexical and structural analyses highlight the nuanced, culture-specific, and context-rich language used in real memes, in contrast to synthetically curated hateful content, which exhibits explicit and repetitive lexical markers. To benchmark the role detection task, we evaluate a wide spectrum of models, including fine-tuned multilingual transformers, sentiment and abuse-aware classifiers, instruction-tuned LLMs, and multimodal vision-language models. Performance is assessed under zero-shot settings using precision, recall, and F1 metrics. While larger models like DeBERTa-v3 and Qwen2.5-VL demonstrate notable gains, results reveal consistent challenges in reliably identifying the 'Victim' class and generalising across cultural and code-mixed content. We also explore prompt design strategies to guide multimodal models and find that hybrid prompts incorporating structured instructions and role definitions offer marginal yet consistent improvements. Our findings underscore the importance of cultural grounding, prompt engineering, and multimodal reasoning in modelling subtle narrative framings in visual-textual content.

Paperid: 5375, https://arxiv.org/pdf/2506.21762.pdf

Abstract:
Data visualization tasks often require multi-step reasoning, and the interpretive strategies experts use, such as decomposing complex goals into smaller subtasks and selectively attending to key chart regions are rarely made explicit. ViStruct is an automated pipeline that simulates these expert behaviours by breaking high-level visual questions into structured analytic steps and highlighting semantically relevant chart areas. Leveraging large language and vision-language models, ViStruct identifies chart components, maps subtasks to spatial regions, and presents visual attention cues to externalize expert-like reasoning flows. While not designed for direct novice instruction, ViStruct provides a replicable model of expert interpretation that can inform the development of future visual literacy tools. We evaluate the system on 45 tasks across 12 chart types and validate its outputs with trained visualization users, confirming its ability to produce interpretable and expert-aligned reasoning sequences.

Paperid: 5376, https://arxiv.org/pdf/2506.21596.pdf

Abstract:
Multimodal large language models (MLLMs) have shown success in vision-language tasks, but their ability to reason over complex educational materials remains largely untested. This work presents the first evaluation of state-of-the-art MLLMs, including LLaVA-1.5 and LLaMA 3.2-Vision, on the textbook question answering (TQA) task using the CK12-QA dataset. We introduce a multimodal retrieval-augmented generation (RAG) pipeline to simulate real-world learning by providing relevant lesson paragraphs and diagrams as context. Our zero-shot experiments reveal a critical trade-off: while retrieved context improves LLaVA's performance on text-based questions, it significantly degrades the accuracy of the more powerful LLaMA 3.2-Vision on diagram-based tasks, dropping its validation accuracy from 74.07% to 25.93%. We term this statistically significant phenomenon "catastrophic context interference." Furthermore, fine-tuning highlights architectural differences: LLaMA 3.2-Vision's performance improves to 71.16% on the test set, demonstrating its capacity to learn multimodal integration, whereas LLaVA's performance declines, indicating challenges with generalization. Our results underscore the challenges MLLMs face in modality prioritization and context integration, providing a benchmark and pointing to key directions for developing more robust AI-driven educational tools.

Paperid: 5377, https://arxiv.org/pdf/2506.20795.pdf

Abstract:
Gestures enable non-verbal human-robot communication, especially in noisy environments like agile production. Traditional deep learning-based gesture recognition relies on task-specific architectures using images, videos, or skeletal pose estimates as input. Meanwhile, Vision Foundation Models (VFMs) and Vision Language Models (VLMs) with their strong generalization abilities offer potential to reduce system complexity by replacing dedicated task-specific modules. This study investigates adapting such models for dynamic, full-body gesture recognition, comparing V-JEPA (a state-of-the-art VFM), Gemini Flash 2.0 (a multimodal VLM), and HD-GCN (a top-performing skeleton-based approach). We introduce NUGGET, a dataset tailored for human-robot communication in intralogistics environments, to evaluate the different gesture recognition approaches. In our experiments, HD-GCN achieves best performance, but V-JEPA comes close with a simple, task-specific classification head - thus paving a possible way towards reducing system complexity, by using it as a shared multi-task model. In contrast, Gemini struggles to differentiate gestures based solely on textual descriptions in the zero-shot setting, highlighting the need of further research on suitable input representations for gestures.

Paperid: 5378, https://arxiv.org/pdf/2506.20449.pdf

Abstract:
Text-to-image generative models have achieved remarkable breakthroughs in recent years. However, their application in medical image generation still faces significant challenges, including small dataset sizes, and scarcity of medical textual data. To address these challenges, we propose Med-Art, a framework specifically designed for medical image generation with limited data. Med-Art leverages vision-language models to generate visual descriptions of medical images which overcomes the scarcity of applicable medical textual data. Med-Art adapts a large-scale pre-trained text-to-image model, PixArt-$Î±$, based on the Diffusion Transformer (DiT), achieving high performance under limited data. Furthermore, we propose an innovative Hybrid-Level Diffusion Fine-tuning (HLDF) method, which enables pixel-level losses, effectively addressing issues such as overly saturated colors. We achieve state-of-the-art performance on two medical image datasets, measured by FID, KID, and downstream classification performance.

Paperid: 5379, https://arxiv.org/pdf/2506.19079.pdf

Abstract:
Foundation Models (FMs) are rapidly transforming Affective Computing (AC), with Vision Language Models (VLMs) now capable of recognising emotions in zero shot settings. This paper probes a critical but underexplored question: what visual cues do these models rely on to infer affect, and are these cues psychologically grounded or superficially learnt? We benchmark varying scale VLMs on a teeth annotated subset of AffectNet dataset and find consistent performance shifts depending on the presence of visible teeth. Through structured introspection of, the best-performing model, i.e., GPT-4o, we show that facial attributes like eyebrow position drive much of its affective reasoning, revealing a high degree of internal consistency in its valence-arousal predictions. These patterns highlight the emergent nature of FMs behaviour, but also reveal risks: shortcut learning, bias, and fairness issues especially in sensitive domains like mental health and education.

Paperid: 5380, https://arxiv.org/pdf/2506.17967.pdf

Abstract:
World models -- generative models that simulate environment dynamics conditioned on past observations and actions -- are gaining prominence in planning, simulation, and embodied AI. However, evaluating their rollouts remains a fundamental challenge, requiring fine-grained, temporally grounded assessment of action alignment and semantic consistency -- capabilities not captured by existing metrics. Vision-Language Models (VLMs) have shown promise as automatic evaluators of generative content due to their strong multimodal reasoning abilities. Yet, their use in fine-grained, temporally sensitive evaluation tasks remains limited and requires targeted adaptation. We introduce a evaluation protocol targeting two recognition tasks -- action recognition and character recognition -- each assessed across binary, multiple-choice, and open-ended formats. To support this, we present UNIVERSE (UNIfied Vision-language Evaluator for Rollouts in Simulated Environments), a method for adapting VLMs to rollout evaluation under data and compute constraints. We conduct a large-scale study comparing full, partial, and parameter-efficient finetuning across task formats, context lengths, sampling strategies, and data compositions. The resulting unified evaluator matches the performance of task-specific baselines using a single checkpoint. Human studies confirm strong alignment with human judgments, establishing UNIVERSE as a scalable, semantics-aware evaluator for world models.

Paperid: 5381, https://arxiv.org/pdf/2506.17462.pdf

Abstract:
Developing general-purpose navigation policies for unknown environments remains a core challenge in robotics. Most existing systems rely on task-specific neural networks and fixed data flows, limiting generalizability. Large Vision-Language Models (LVLMs) offer a promising alternative by embedding human-like knowledge suitable for reasoning and planning. Yet, prior LVLM-robot integrations typically depend on pre-mapped spaces, hard-coded representations, and myopic exploration. We introduce the Agentic Robotic Navigation Architecture (ARNA), a general-purpose navigation framework that equips an LVLM-based agent with a library of perception, reasoning, and navigation tools available within modern robotic stacks. At runtime, the agent autonomously defines and executes task-specific workflows that iteratively query the robotic modules, reason over multimodal inputs, and select appropriate navigation actions. This approach enables robust navigation and reasoning in previously unmapped environments, providing a new perspective on robotic stack design. Evaluated in Habitat Lab on the HM-EQA benchmark, ARNA achieves state-of-the-art performance, demonstrating effective exploration, navigation, and embodied question answering without relying on handcrafted plans, fixed input representations, or pre-existing maps.

Paperid: 5382, https://arxiv.org/pdf/2506.10633.pdf

Abstract:
Latent Diffusion Models have shown remarkable results in text-guided image synthesis in recent years. In the domain of natural (RGB) images, recent works have shown that such models can be adapted to various vision-language downstream tasks with little to no supervision involved. On the contrary, text-to-image Latent Diffusion Models remain relatively underexplored in the field of medical imaging, primarily due to limited data availability (e.g., due to privacy concerns). In this work, focusing on the chest X-ray modality, we first demonstrate that a standard text-conditioned Latent Diffusion Model has not learned to align clinically relevant information in free-text radiology reports with the corresponding areas of the given scan. Then, to alleviate this issue, we propose a fine-tuning framework to improve multi-modal alignment in a pre-trained model such that it can be efficiently repurposed for downstream tasks such as phrase grounding. Our method sets a new state-of-the-art on a standard benchmark dataset (MS-CXR), while also exhibiting robust performance on out-of-distribution data (VinDr-CXR). Our code will be made publicly available.

Paperid: 5383, https://arxiv.org/pdf/2506.08691.pdf

Abstract:
Large Vision-Language Models (LVLMs) have shown exceptional performance in multimodal tasks, but their effectiveness in complex visual reasoning is still constrained, especially when employing Chain-of-Thought prompting techniques. In this paper, we propose VReST, a novel training-free approach that enhances Reasoning in LVLMs through Monte Carlo Tree Search and Self-Reward mechanisms. VReST meticulously traverses the reasoning landscape by establishing a search tree, where each node encapsulates a reasoning step, and each path delineates a comprehensive reasoning sequence. Our innovative multimodal Self-Reward mechanism assesses the quality of reasoning steps by integrating the utility of sub-questions, answer correctness, and the relevance of vision-language clues, all without the need for additional models. VReST surpasses current prompting methods and secures state-of-the-art performance across three multimodal mathematical reasoning benchmarks. Furthermore, it substantiates the efficacy of test-time scaling laws in multimodal tasks, offering a promising direction for future research.

Paperid: 5384, https://arxiv.org/pdf/2506.06630.pdf

Abstract:
Vision-Language Navigation (VLN) policies trained on offline datasets often exhibit degraded task performance when deployed in unfamiliar navigation environments at test time, where agents are typically evaluated without access to external interaction or feedback. Entropy minimization has emerged as a practical solution for reducing prediction uncertainty at test time; however, it can suffer from accumulated errors, as agents may become overconfident in incorrect actions without sufficient contextual grounding. To tackle these challenges, we introduce ATENA (Active TEst-time Navigation Agent), a test-time active learning framework that enables a practical human-robot interaction via episodic feedback on uncertain navigation outcomes. In particular, ATENA learns to increase certainty in successful episodes and decrease it in failed ones, improving uncertainty calibration. Here, we propose mixture entropy optimization, where entropy is obtained from a combination of the action and pseudo-expert distributions-a hypothetical action distribution assuming the agent's selected action to be optimal-controlling both prediction confidence and action preference. In addition, we propose a self-active learning strategy that enables an agent to evaluate its navigation outcomes based on confident predictions. As a result, the agent stays actively engaged throughout all iterations, leading to well-grounded and adaptive decision-making. Extensive evaluations on challenging VLN benchmarks-REVERIE, R2R, and R2R-CE-demonstrate that ATENA successfully overcomes distributional shifts at test time, outperforming the compared baseline methods across various settings.

Paperid: 5385, https://arxiv.org/pdf/2506.04501.pdf

Abstract:
Existing deepfake detection techniques struggle to keep-up with the ever-evolving novel, unseen forgeries methods. This limitation stems from their reliance on statistical artifacts learned during training, which are often tied to specific generation processes that may not be representative of samples from new, unseen deepfake generation methods encountered at test time. We propose that incorporating language guidance can improve deepfake detection generalization by integrating human-like commonsense reasoning -- such as recognizing logical inconsistencies and perceptual anomalies -- alongside statistical cues. To achieve this, we train an expert deepfake vision encoder by combining discriminative classification with image-text contrastive learning, where the text is generated by generalist MLLMs using few-shot prompting. This allows the encoder to extract both language-describable, commonsense deepfake artifacts and statistical forgery artifacts from pixel-level distributions. To further enhance robustness, we integrate data uncertainty learning into vision-language contrastive learning, mitigating noise in image-text supervision. Our expert vision encoder seamlessly interfaces with an LLM, further enabling more generalized and interpretable deepfake detection while also boosting accuracy. The resulting framework, AuthGuard, achieves state-of-the-art deepfake detection accuracy in both in-distribution and out-of-distribution settings, achieving AUC gains of 6.15% on the DFDC dataset and 16.68% on the DF40 dataset. Additionally, AuthGuard significantly enhances deepfake reasoning, improving performance by 24.69% on the DDVQA dataset.

Paperid: 5386, https://arxiv.org/pdf/2506.03229.pdf

Abstract:
In the context of noisy partial label learning (NPLL), each training sample is associated with a set of candidate labels annotated by multiple noisy annotators. With the emergence of high-performance pre-trained vision-language models (VLMs) such as CLIP, LLaVa and GPT-4V, the direction of using these models to replace time-consuming manual annotation workflows and achieve "manual-annotation-free" training for downstream tasks has become a highly promising research avenue. This paper focuses on learning from noisy partial labels annotated by pre-trained VLMs and proposes an innovative collaborative consistency regularization (Co-Reg) method. Unlike the symmetric noise primarily addressed in traditional noisy label learning, the noise generated by pre-trained models is instance-dependent, embodying the underlying patterns of the pre-trained models themselves, which significantly increases the learning difficulty for the model. To address this, we simultaneously train two neural networks that implement collaborative purification of training labels through a "Co-Pseudo-Labeling" mechanism, while enforcing consistency regularization constraints in both the label space and feature representation space. Our method can also leverage few-shot manually annotated valid labels to further enhance its performances. Comparative experiments with different denoising and disambiguation algorithms, annotation manners, and pre-trained model application schemes fully validate the effectiveness of the proposed method, while revealing the broad prospects of integrating weakly-supervised learning techniques into the knowledge distillation process of pre-trained models.

Paperid: 5387, https://arxiv.org/pdf/2506.00979.pdf

Abstract:
The rapid advancement of Artificial Intelligence Generated Content (AIGC) in visual domains has resulted in highly realistic synthetic images and videos, driven by sophisticated generative frameworks such as diffusion-based architectures. While these breakthroughs open substantial opportunities, they simultaneously raise critical concerns about content authenticity and integrity. Many current AIGC detection methods operate as black-box binary classifiers, which offer limited interpretability, and no approach supports detecting both images and videos in a unified framework. This dual limitation compromises model transparency, reduces trustworthiness, and hinders practical deployment. To address these challenges, we introduce IVY-FAKE , a novel, unified, and large-scale dataset specifically designed for explainable multimodal AIGC detection. Unlike prior benchmarks, which suffer from fragmented modality coverage and sparse annotations, IVY-FAKE contains over 150,000 richly annotated training samples (images and videos) and 18,700 evaluation examples, each accompanied by detailed natural-language reasoning beyond simple binary labels. Building on this, we propose Ivy Explainable Detector (IVY-XDETECTOR), a unified AIGC detection and explainable architecture that jointly performs explainable detection for both image and video content. Our unified vision-language model achieves state-of-the-art performance across multiple image and video detection benchmarks, highlighting the significant advancements enabled by our dataset and modeling framework. Our data is publicly available at https://huggingface.co/datasets/AI-Safeguard/Ivy-Fake.

Paperid: 5388, https://arxiv.org/pdf/2505.20711.pdf

Abstract:
The absence of explicit communication channels between automated vehicles (AVs) and other road users requires the use of external Human-Machine Interfaces (eHMIs) to convey messages effectively in uncertain scenarios. Currently, most eHMI studies employ predefined text messages and manually designed actions to perform these messages, which limits the real-world deployment of eHMIs, where adaptability in dynamic scenarios is essential. Given the generalizability and versatility of large language models (LLMs), they could potentially serve as automated action designers for the message-action design task. To validate this idea, we make three contributions: (1) We propose a pipeline that integrates LLMs and 3D renderers, using LLMs as action designers to generate executable actions for controlling eHMIs and rendering action clips. (2) We collect a user-rated Action-Design Scoring dataset comprising a total of 320 action sequences for eight intended messages and four representative eHMI modalities. The dataset validates that LLMs can translate intended messages into actions close to a human level, particularly for reasoning-enabled LLMs. (3) We introduce two automated raters, Action Reference Score (ARS) and Vision-Language Models (VLMs), to benchmark 18 LLMs, finding that the VLM aligns with human preferences yet varies across eHMI modalities.

Paperid: 5389, https://arxiv.org/pdf/2505.16778.pdf

Abstract:
Few-shot counting estimates the number of target objects in an image using only a few annotated exemplars. However, domain shift severely hinders existing methods to generalize to unseen scenarios. This falls into the realm of single domain generalization that remains unexplored in few-shot counting. To solve this problem, we begin by analyzing the main limitations of current methods, which typically follow a standard pipeline that extract the object prototypes from exemplars and then match them with image feature to construct the correlation map. We argue that existing methods overlook the significance of learning highly generalized prototypes. Building on this insight, we propose the first single domain generalization few-shot counting model, Universal Representation Matching, termed URM. Our primary contribution is the discovery that incorporating universal vision-language representations distilled from a large scale pretrained vision-language model into the correlation construction process substantially improves robustness to domain shifts without compromising in domain performance. As a result, URM achieves state-of-the-art performance on both in domain and the newly introduced domain generalization setting.

Paperid: 5390, https://arxiv.org/pdf/2505.16027.pdf

Abstract:
Foundation models leveraging vision-language pretraining have shown promise in chest X-ray (CXR) interpretation, yet their real-world performance across diverse populations and diagnostic tasks remains insufficiently evaluated. This study benchmarks the diagnostic performance and generalizability of foundation models versus traditional convolutional neural networks (CNNs) on multinational CXR datasets. We evaluated eight CXR diagnostic models - five vision-language foundation models and three CNN-based architectures - across 37 standardized classification tasks using six public datasets from the USA, Spain, India, and Vietnam, and three private datasets from hospitals in China. Performance was assessed using AUROC, AUPRC, and other metrics across both shared and dataset-specific tasks. Foundation models outperformed CNNs in both accuracy and task coverage. MAVL, a model incorporating knowledge-enhanced prompts and structured supervision, achieved the highest performance on public (mean AUROC: 0.82; AUPRC: 0.32) and private (mean AUROC: 0.95; AUPRC: 0.89) datasets, ranking first in 14 of 37 public and 3 of 4 private tasks. All models showed reduced performance on pediatric cases, with average AUROC dropping from 0.88 +/- 0.18 in adults to 0.57 +/- 0.29 in children (p = 0.0202). These findings highlight the value of structured supervision and prompt design in radiologic AI and suggest future directions including geographic expansion and ensemble modeling for clinical deployment. Code for all evaluated models is available at https://drive.google.com/drive/folders/1B99yMQm7bB4h1sVMIBja0RfUu8gLktCE

Paperid: 5391, https://arxiv.org/pdf/2505.13180.pdf

Abstract:
Integrating Large Language Models with symbolic planners is a promising direction for obtaining verifiable and grounded plans compared to planning in natural language, with recent works extending this idea to visual domains using Vision-Language Models (VLMs). However, rigorous comparison between VLM-grounded symbolic approaches and methods that plan directly with a VLM has been hindered by a lack of common environments, evaluation protocols and model coverage. We introduce ViPlan, the first open-source benchmark for Visual Planning with symbolic predicates and VLMs. ViPlan features a series of increasingly challenging tasks in two domains: a visual variant of the classic Blocksworld planning problem and a simulated household robotics environment. We benchmark nine open-source VLM families across multiple sizes, along with selected closed models, evaluating both VLM-grounded symbolic planning and using the models directly to propose actions. We find symbolic planning to outperform direct VLM planning in Blocksworld, where accurate image grounding is crucial, whereas the opposite is true in the household robotics tasks, where commonsense knowledge and the ability to recover from errors are beneficial. Finally, we show that across most models and methods, there is no significant benefit to using Chain-of-Thought prompting, suggesting that current VLMs still struggle with visual reasoning.

Paperid: 5392, https://arxiv.org/pdf/2505.12253.pdf

Abstract:
Despite achieving significant progress in 2D image understanding, large multimodal models (LMMs) struggle in the physical world due to the lack of spatial representation. Typically, existing 3D LMMs mainly embed 3D positions as fixed spatial prompts within visual features to represent the scene. However, these methods are limited to understanding the static background and fail to capture temporally varying dynamic objects. In this paper, we propose LLaVA-4D, a general LMM framework with a novel spatiotemporal prompt for visual representation in 4D scene understanding. The spatiotemporal prompt is generated by encoding 3D position and 1D time into a dynamic-aware 4D coordinate embedding. Moreover, we demonstrate that spatial and temporal components disentangled from visual features are more effective in distinguishing the background from objects. This motivates embedding the 4D spatiotemporal prompt into these features to enhance the dynamic scene representation. By aligning visual spatiotemporal embeddings with language embeddings, LMMs gain the ability to understand both spatial and temporal characteristics of static background and dynamic objects in the physical world. Additionally, we construct a 4D vision-language dataset with spatiotemporal coordinate annotations for instruction fine-tuning LMMs. Extensive experiments have been conducted to demonstrate the effectiveness of our method across different tasks in 4D scene understanding.

Paperid: 5393, https://arxiv.org/pdf/2505.11809.pdf

Abstract:
Visibility analysis is one of the fundamental analytics methods in urban planning and landscape research, traditionally conducted through computational simulations based on the Line-of-Sight (LoS) principle. However, when assessing the visibility of named urban objects such as landmarks, geometric intersection alone fails to capture the contextual and perceptual dimensions of visibility as experienced in the real world. The study challenges the traditional LoS-based approaches by introducing a new, image-based visibility analysis method. Specifically, a Vision Language Model (VLM) is applied to detect the target object within a direction-zoomed Street View Image (SVI). Successful detection represents the object's visibility at the corresponding SVI location. Further, a heterogeneous visibility graph is constructed to address the complex interaction between observers and target objects. In the first case study, the method proves its reliability in detecting the visibility of six tall landmark constructions in global cities, with an overall accuracy of 87%. Furthermore, it reveals broader contextual differences when the landmarks are perceived and experienced. In the second case, the proposed visibility graph uncovers the form and strength of connections for multiple landmarks along the River Thames in London, as well as the places where these connections occur. Notably, bridges on the River Thames account for approximately 30% of total connections. Our method complements and enhances traditional LoS-based visibility analysis, and showcases the possibility of revealing the prevalent connection of any visual objects in the urban environment. It opens up new research perspectives for urban planning, heritage conservation, and computational social science.

Paperid: 5394, https://arxiv.org/pdf/2505.11676.pdf

Abstract:
Open-vocabulary semantic segmentation aims to segment images into distinct semantic regions for both seen and unseen categories at the pixel level. Current methods utilize text embeddings from pre-trained vision-language models like CLIP but struggle with the inherent domain gap between image and text embeddings, even after extensive alignment during training. Additionally, relying solely on deep text-aligned features limits shallow-level feature guidance, which is crucial for detecting small objects and fine details, ultimately reducing segmentation accuracy. To address these limitations, we propose a dual prompting framework, DPSeg, for this task. Our approach combines dual-prompt cost volume generation, a cost volume-guided decoder, and a semantic-guided prompt refinement strategy that leverages our dual prompting scheme to mitigate alignment issues in visual prompt generation. By incorporating visual embeddings from a visual prompt encoder, our approach reduces the domain gap between text and image embeddings while providing multi-level guidance through shallow features. Extensive experiments demonstrate that our method significantly outperforms existing state-of-the-art approaches on multiple public datasets.

Paperid: 5395, https://arxiv.org/pdf/2505.08803.pdf

Abstract:
Recent research has highlighted the risk of generative model collapse, where performance progressively degrades when continually trained on self-generated data. However, existing exploration on model collapse is limited to single, unimodal models, limiting our understanding in more realistic scenarios, such as diverse multi-modal AI agents interacting autonomously through synthetic data and continually evolving. We expand the synthetic data training and model collapse study to multi-modal vision-language generative systems, such as vision-language models (VLMs) and text-to-image diffusion models, as well as recursive generate-train loops with multiple models. We find that model collapse, previously observed in single-modality generative models, exhibits distinct characteristics in the multi-modal context, such as improved vision-language alignment and increased variance in VLM image-captioning task. Additionally, we find that general approaches such as increased decoding budgets, greater model diversity, and relabeling with frozen models can effectively mitigate model collapse. Our findings provide initial insights and practical guidelines for reducing the risk of model collapse in self-improving multi-agent AI systems and curating robust multi-modal synthetic datasets.

Paperid: 5396, https://arxiv.org/pdf/2505.07347.pdf

Abstract:
Echocardiographers can detect pulmonary hypertension using Doppler echocardiography; however, accurately assessing its progression often proves challenging. Right heart catheterization (RHC), the gold standard for precise evaluation, is invasive and unsuitable for routine use, limiting its practicality for timely diagnosis and monitoring of pulmonary hypertension progression. Here, we propose MePH, a multi-view, multi-modal vision-language model to accurately assess pulmonary hypertension progression using non-invasive echocardiography. We constructed a large dataset comprising paired standardized echocardiogram videos, spectral images and RHC data, covering 1,237 patient cases from 12 medical centers. For the first time, MePH precisely models the correlation between non-invasive multi-view, multi-modal echocardiography and the pressure and resistance obtained via RHC. We show that MePH significantly outperforms echocardiographers' assessments using echocardiography, reducing the mean absolute error in estimating mean pulmonary arterial pressure (mPAP) and pulmonary vascular resistance (PVR) by 49.73% and 43.81%, respectively. In eight independent external hospitals, MePH achieved a mean absolute error of 3.147 for PVR assessment. Furthermore, MePH achieved an area under the curve of 0.921, surpassing echocardiographers (area under the curve of 0.842) in accurately predicting the severity of pulmonary hypertension, whether mild or severe. A prospective study demonstrated that MePH can predict treatment efficacy for patients. Our work provides pulmonary hypertension patients with a non-invasive and timely method for monitoring disease progression, improving the accuracy and efficiency of pulmonary hypertension management while enabling earlier interventions and more personalized treatment decisions.

Paperid: 5397, https://arxiv.org/pdf/2505.05163.pdf

Abstract:
Vision-Language Models (VLMs) learn joint representations by mapping images and text into a shared latent space. However, recent research highlights that deterministic embeddings from standard VLMs often struggle to capture the uncertainties arising from the ambiguities in visual and textual descriptions and the multiple possible correspondences between images and texts. Existing approaches tackle this by learning probabilistic embeddings during VLM training, which demands large datasets and does not leverage the powerful representations already learned by large-scale VLMs like CLIP. In this paper, we propose GroVE, a post-hoc approach to obtaining probabilistic embeddings from frozen VLMs. GroVE builds on Gaussian Process Latent Variable Model (GPLVM) to learn a shared low-dimensional latent space where image and text inputs are mapped to a unified representation, optimized through single-modal embedding reconstruction and cross-modal alignment objectives. Once trained, the Gaussian Process model generates uncertainty-aware probabilistic embeddings. Evaluation shows that GroVE achieves state-of-the-art uncertainty calibration across multiple downstream tasks, including cross-modal retrieval, visual question answering, and active learning.

Paperid: 5398, https://arxiv.org/pdf/2505.04980.pdf

Abstract:
This paper proposes a novel Large Vision-Language Model (LVLM) and Model Predictive Control (MPC) integration framework that delivers both task scalability and safety for Autonomous Driving (AD). LVLMs excel at high-level task planning across diverse driving scenarios. However, since these foundation models are not specifically designed for driving and their reasoning is not consistent with the feasibility of low-level motion planning, concerns remain regarding safety and smooth task switching. This paper integrates LVLMs with MPC Builder, which automatically generates MPCs on demand, based on symbolic task commands generated by the LVLM, while ensuring optimality and safety. The generated MPCs can strongly assist the execution or rejection of LVLM-driven task switching by providing feedback on the feasibility of the given tasks and generating task-switching-aware MPCs. Our approach provides a safe, flexible, and adaptable control framework, bridging the gap between cutting-edge foundation models and reliable vehicle operation. We demonstrate the effectiveness of our approach through a simulation experiment, showing that our system can safely and effectively handle highway driving while maintaining the flexibility and adaptability of LVLMs.

Paperid: 5399, https://arxiv.org/pdf/2505.01984.pdf

Abstract:
Whole Slide Images (WSIs) play a crucial role in accurate cancer diagnosis and prognosis, as they provide tissue details at the cellular level. However, the rapid growth of computational tasks involving WSIs poses significant challenges. Given that WSIs are gigapixels in size, they present difficulties in terms of storage, processing, and model training. Therefore, it is essential to develop lifelong learning approaches for WSI analysis. In scenarios where slides are distributed across multiple institutes, we aim to leverage them to develop a unified online model as a computational tool for cancer diagnosis in clinical and hospital settings. In this study, we introduce ADaFGrad, a method designed to enhance lifelong learning for whole-slide image (WSI) analysis. First, we leverage pathology vision-language foundation models to develop a framework that enables interaction between a slide's regional tissue features and a predefined text-based prototype buffer. Additionally, we propose a gradient-distillation mechanism that mimics the gradient of a logit with respect to the classification-head parameters across past and current iterations in a continual-learning setting. We construct a sequence of six TCGA datasets for training and evaluation. Experimental results show that ADaFGrad outperforms both state-of-the-art WSI-specific and conventional continual-learning methods after only a few training epochs, exceeding them by up to +5.068% in the class-incremental learning scenario while exhibiting the least forgetting (i.e., retaining the most knowledge from previous tasks). Moreover, ADaFGrad surpasses its baseline by as much as +40.084% in accuracy, further demonstrating the effectiveness of the proposed modules.

Paperid: 5400, https://arxiv.org/pdf/2505.01713.pdf

Abstract:
Long-term action anticipation (LTA) aims to predict future actions over an extended period. Previous approaches primarily focus on learning exclusively from video data but lack prior knowledge. Recent researches leverage large language models (LLMs) by utilizing text-based inputs which suffer severe information loss. To tackle these limitations single-modality methods face, we propose a novel Intention-Conditioned Vision-Language (ICVL) model in this study that fully leverages the rich semantic information of visual data and the powerful reasoning capabilities of LLMs. Considering intention as a high-level concept guiding the evolution of actions, we first propose to employ a vision-language model (VLM) to infer behavioral intentions as comprehensive textual features directly from video inputs. The inferred intentions are then fused with visual features through a multi-modality fusion strategy, resulting in intention-enhanced visual representations. These enhanced visual representations, along with textual prompts, are fed into LLM for future action anticipation. Furthermore, we propose an effective example selection strategy jointly considers visual and textual similarities, providing more relevant and informative examples for in-context learning. Extensive experiments with state-of-the-art performance on Ego4D, EPIC-Kitchens-55, and EGTEA GAZE+ datasets fully demonstrate the effectiveness and superiority of the proposed method.

Paperid: 5401, https://arxiv.org/pdf/2505.00156.pdf

Abstract:
Large Vision Language Models (LVLMs) have shown strong capabilities in understanding and analyzing visual scenes across various domains. However, in the context of autonomous driving, their limited comprehension of 3D environments restricts their effectiveness in achieving a complete and safe understanding of dynamic surroundings. To address this, we introduce V3LMA, a novel approach that enhances 3D scene understanding by integrating Large Language Models (LLMs) with LVLMs. V3LMA leverages textual descriptions generated from object detections and video inputs, significantly boosting performance without requiring fine-tuning. Through a dedicated preprocessing pipeline that extracts 3D object data, our method improves situational awareness and decision-making in complex traffic scenarios, achieving a score of 0.56 on the LingoQA benchmark. We further explore different fusion strategies and token combinations with the goal of advancing the interpretation of traffic scenes, ultimately enabling safer autonomous driving systems.

Paperid: 5402, https://arxiv.org/pdf/2504.20220.pdf

Abstract:
Despite the growing adoption of electronic health records, many processes still rely on paper documents, reflecting the heterogeneous real-world conditions in which healthcare is delivered. The manual transcription process is time-consuming and prone to errors when transferring paper-based data to digital formats. To streamline this workflow, this study presents an open-source pipeline that extracts and categorizes checkbox data from scanned documents. Demonstrated on transfusion reaction reports, the design supports adaptation to other checkbox-rich document types. The proposed method integrates checkbox detection, multilingual optical character recognition (OCR) and multilingual vision-language models (VLMs). The pipeline achieves high precision and recall compared against annually compiled gold-standards from 2017 to 2024. The result is a reduction in administrative workload and accurate regulatory reporting. The open-source availability of this pipeline encourages self-hosted parsing of checkbox forms.

Paperid: 5403, https://arxiv.org/pdf/2504.18800.pdf

Abstract:
Echocardiography records ultrasound videos of the heart, enabling clinicians to assess cardiac function. Recent advances in large-scale vision-language models (VLMs) have spurred interest in automating echocardiographic interpretation. However, most existing medical VLMs rely on single-frame (image) inputs, which can reduce diagnostic accuracy for conditions identifiable only through cardiac motion. In addition, echocardiographic videos are captured from multiple views, each varying in suitability for detecting specific conditions. Leveraging multiple views may therefore improve diagnostic performance. We developed a video-language model that processes full video sequences from five standard views, trained on 60,747 echocardiographic video-report pairs. We evaluated the gains in retrieval performance from video input and multi-view support, including the contributions of various pretrained models.

Paperid: 5404, https://arxiv.org/pdf/2504.18025.pdf

Abstract:
Visible-Infrared Person Re-identification (VIReID) aims to match visible and infrared pedestrian images, but the modality differences and the complexity of identity features make it challenging. Existing methods rely solely on identity label supervision, which makes it difficult to fully extract high-level semantic information. Recently, vision-language pre-trained models have been introduced to VIReID, enhancing semantic information modeling by generating textual descriptions. However, such methods do not explicitly model body shape features, which are crucial for cross-modal matching. To address this, we propose an effective Body Shape-aware Textual Alignment (BSaTa) framework that explicitly models and utilizes body shape information to improve VIReID performance. Specifically, we design a Body Shape Textual Alignment (BSTA) module that extracts body shape information using a human parsing model and converts it into structured text representations via CLIP. We also design a Text-Visual Consistency Regularizer (TVCR) to ensure alignment between body shape textual representations and visual body shape features. Furthermore, we introduce a Shape-aware Representation Learning (SRL) mechanism that combines Multi-text Supervision and Distribution Consistency Constraints to guide the visual encoder to learn modality-invariant and discriminative identity features, thus enhancing modality invariance. Experimental results demonstrate that our method achieves superior performance on the SYSU-MM01 and RegDB datasets, validating its effectiveness.

Paperid: 5405, https://arxiv.org/pdf/2504.15627.pdf

Abstract:
Lifelong learning for whole slide images (WSIs) poses the challenge of training a unified model to perform multiple WSI-related tasks, such as cancer subtyping and tumor classification, in a distributed, continual fashion. This is a practical and applicable problem in clinics and hospitals, as WSIs are large, require storage, processing, and transfer time. Training new models whenever new tasks are defined is time-consuming. Recent work has applied regularization- and rehearsal-based methods to this setting. However, the rise of vision-language foundation models that align diagnostic text with pathology images raises the question: are these models alone sufficient for lifelong WSI learning using zero-shot classification, or is further investigation into continual learning strategies needed to improve performance? To our knowledge, this is the first study to compare conventional continual-learning approaches with vision-language zero-shot classification for WSIs. Our source code and experimental results will be available soon.

Paperid: 5406, https://arxiv.org/pdf/2504.14638.pdf

Abstract:
Vision-language models (VLMs) have demonstrated impressive zero-shot transfer capabilities in image-level visual perception tasks. However, they fall short in 3D instance-level segmentation tasks that require accurate localization and recognition of individual objects. To bridge this gap, we introduce a novel 3D Gaussian Splatting based hard visual prompting approach that leverages camera interpolation to generate diverse viewpoints around target objects without any 2D-3D optimization or fine-tuning. Our method simulates realistic 3D perspectives, effectively augmenting existing hard visual prompts by enforcing geometric consistency across viewpoints. This training-free strategy seamlessly integrates with prior hard visual prompts, enriching object-descriptive features and enabling VLMs to achieve more robust and accurate 3D instance segmentation in diverse 3D scenes.

Paperid: 5407, https://arxiv.org/pdf/2504.14129.pdf

Abstract:
The challenge of tracing the source attribution of forged faces has gained significant attention due to the rapid advancement of generative models. However, existing deepfake attribution (DFA) works primarily focus on the interaction among various domains in vision modality, and other modalities such as texts and face parsing are not fully explored. Besides, they tend to fail to assess the generalization performance of deepfake attributors to unseen advanced generators like diffusion in a fine-grained manner. In this paper, we propose a novel parsing-aware vision language model with dynamic contrastive learning(PVLM) method for zero-shot deepfake attribution (ZS-DFA),which facilitates effective and fine-grained traceability to unseen advanced generators. Specifically, we conduct a novel and fine-grained ZS-DFA benchmark to evaluate the attribution performance of deepfake attributors to unseen advanced generators like diffusion. Besides, we propose an innovative parsing-guided vision language model with dynamic contrastive learning (PVLM) method to capture general and diverse attribution features. We are motivated by the observation that the preservation of source face attributes in facial images generated by GAN and diffusion models varies significantly. We employ the inherent face attributes preservation differences to capture face parsing-aware forgery representations. Therefore, we devise a novel parsing encoder to focus on global face attribute embeddings, enabling parsing-guided DFA representation learning via dynamic vision-parsing matching. Additionally, we present a novel deepfake attribution contrastive center loss to pull relevant generators closer and push irrelevant ones away, which can be introduced into DFA models to enhance traceability. Experimental results show that our model exceeds the state-of-the-art on the ZS-DFA benchmark via various protocol evaluations.

Paperid: 5408, https://arxiv.org/pdf/2504.14123.pdf

Abstract:
Prompt learning is a popular fine-tuning method for vision-language models due to its efficiency. It requires a small number of additional learnable parameters while significantly enhancing performance on target tasks. However, most existing methods suffer from overfitting to fine-tuning data, yielding poor generalizability. To address this, we propose a new training objective function based on a Bayesian learning principle to balance adaptability and generalizability. We derive a prior over the logits, where the mean function is parameterized by the pre-trained model, while the posterior corresponds to the fine-tuned model. This objective establishes a balance by allowing the fine-tuned model to adapt to downstream tasks while remaining close to the pre-trained model.

Paperid: 5409, https://arxiv.org/pdf/2504.13861.pdf

Abstract:
Though Large Vision-Language Models (LVLMs) are being actively explored in medicine, their ability to conduct telemedicine consultations combining accurate diagnosis with professional dialogue remains underexplored. In this paper, we present 3MDBench (Medical Multimodal Multi-agent Dialogue Benchmark), an open-source framework for simulating and evaluating LVLM-driven telemedical consultations. 3MDBench simulates patient variability through four temperament-based Patient Agents and an Assessor Agent that jointly evaluate diagnostic accuracy and dialogue quality. It includes 3013 cases across 34 diagnoses drawn from real-world telemedicine interactions, combining textual and image-based data. The experimental study compares diagnostic strategies for popular LVLMs, including GPT-4o-mini, LLaVA-3.2-11B-Vision-Instruct, and Qwen2-VL-7B-Instruct. We demonstrate that multimodal dialogue with internal reasoning improves F1 score by 6.5% over non-dialogue settings, highlighting the importance of context-aware, information-seeking questioning. Moreover, injecting predictions from a diagnostic convolutional network into the LVLM's context boosts F1 by up to 20%. Source code is available at https://anonymous.4open.science/r/3mdbench_acl-0511.

Paperid: 5410, https://arxiv.org/pdf/2504.11733.pdf

Abstract:
Inspired by the dual-stream theory of the human visual system (HVS) - where the ventral stream is responsible for object recognition and detail analysis, while the dorsal stream focuses on spatial relationships and motion perception - an increasing number of video quality assessment (VQA) works built upon this framework are proposed. Recent advancements in large multi-modal models, notably Contrastive Language-Image Pretraining (CLIP), have motivated researchers to incorporate CLIP into dual-stream-based VQA methods. This integration aims to harness the model's superior semantic understanding capabilities to replicate the object recognition and detail analysis in ventral stream, as well as spatial relationship analysis in dorsal stream. However, CLIP is originally designed for images and lacks the ability to capture temporal and motion information inherent in videos. To address the limitation, this paper propose a Decoupled Vision-Language Modeling with Text-Guided Adaptation for Blind Video Quality Assessment (DVLTA-VQA), which decouples CLIP's visual and textual components, and integrates them into different stages of the NR-VQA pipeline. Specifically, a Video-Based Temporal CLIP module is proposed to explicitly model temporal dynamics and enhance motion perception, aligning with the dorsal stream. Additionally, a Temporal Context Module is developed to refine inter-frame dependencies, further improving motion modeling. On the ventral stream side, a Basic Visual Feature Extraction Module is employed to strengthen detail analysis. Finally, a text-guided adaptive fusion strategy is proposed to enable dynamic weighting of features, facilitating more effective integration of spatial and temporal information.

Paperid: 5411, https://arxiv.org/pdf/2504.03245.pdf

Abstract:
Generalizable robotic mobile manipulation in open-world environments poses significant challenges due to long horizons, complex goals, and partial observability. A promising approach to address these challenges involves planning with a library of parameterized skills, where a task planner sequences these skills to achieve goals specified in structured languages, such as logical expressions over symbolic facts. While vision-language models (VLMs) can be used to ground these expressions, they often assume full observability, leading to suboptimal behavior when the agent lacks sufficient information to evaluate facts with certainty. This paper introduces a novel framework that leverages VLMs as a perception module to estimate uncertainty and facilitate symbolic grounding. Our approach constructs a symbolic belief representation and uses a belief-space planner to generate uncertainty-aware plans that incorporate strategic information gathering. This enables the agent to effectively reason about partial observability and property uncertainty. We demonstrate our system on a range of challenging real-world tasks that require reasoning in partially observable environments. Simulated evaluations show that our approach outperforms both vanilla VLM-based end-to-end planning or VLM-based state estimation baselines by planning for and executing strategic information gathering. This work highlights the potential of VLMs to construct belief-space symbolic scene representations, enabling downstream tasks such as uncertainty-aware planning.

Paperid: 5412, https://arxiv.org/pdf/2503.19647.pdf

Abstract:
Large Vision-Language Models (VLMs) are increasingly being regarded as foundation models that can be instructed to solve diverse tasks by prompting, without task-specific training. We examine the seemingly obvious question: how to effectively prompt VLMs for semantic segmentation. To that end, we systematically evaluate the segmentation performance of several recent models guided by either text or visual prompts on the out-of-distribution MESS dataset collection. We introduce a scalable prompting scheme, few-shot prompted semantic segmentation, inspired by open-vocabulary segmentation and few-shot learning. It turns out that VLMs lag far behind specialist models trained for a specific segmentation task, by about 30% on average on the Intersection-over-Union metric. Moreover, we find that text prompts and visual prompts are complementary: each one of the two modes fails on many examples that the other one can solve. Our analysis suggests that being able to anticipate the most effective prompt modality can lead to a 11% improvement in performance. Motivated by our findings, we propose PromptMatcher, a remarkably simple training-free baseline that combines both text and visual prompts, achieving state-of-the-art results outperforming the best text-prompted VLM by 2.5%, and the top visual-prompted VLM by 3.5% on few-shot prompted semantic segmentation.

Paperid: 5413, https://arxiv.org/pdf/2503.19356.pdf

Abstract:
AI models have made significant strides in recent years in their ability to describe and answer questions about real-world images. They have also made progress in the ability to converse with users in real-time using audio input. This raises the question: have we reached the point where AI models, connected to a camera and microphone, can converse with users in real-time about scenes and events that are unfolding live in front of the camera? This has been a long-standing goal in AI and is a prerequisite for real-world AI assistants and humanoid robots to interact with humans in everyday situations. In this work, we introduce a new dataset and benchmark, the Qualcomm Interactive Video Dataset (IVD), which allows us to assess the extent to which existing models can support these abilities, and to what degree these capabilities can be instilled through fine-tuning. The dataset is based on a simple question-answering setup, where users ask questions that the system has to answer, in real-time, based on the camera and audio input. We show that existing models fall far behind human performance on this task, and we identify the main sources for the performance gap. However, we also show that for many of the required perceptual skills, fine-tuning on this form of data can significantly reduce this gap.

Paperid: 5414, https://arxiv.org/pdf/2503.17416.pdf

Abstract:
Debugging of Deep Neural Networks (DNNs), particularly vision models, is very challenging due to the complex and opaque decision-making processes in these networks. In this paper, we explore multi-modal Vision-Language Models (VLMs), such as CLIP, to automatically interpret the opaque representation space of vision models using natural language. This in turn, enables a semantic analysis of model behavior using human-understandable concepts, without requiring costly human annotations. Key to our approach is the notion of semantic heatmap, that succinctly captures the statistical properties of DNNs in terms of the concepts discovered with the VLM and that are computed off-line using a held-out data set. We show the utility of semantic heatmaps for fault localization -- an essential step in debugging -- in vision models. Our proposed technique helps localize the fault in the network (encoder vs head) and also highlights the responsible high-level concepts, by leveraging novel differential heatmaps, which summarize the semantic differences between the correct and incorrect behaviour of the analyzed DNN. We further propose a lightweight runtime analysis to detect and filter-out defects at runtime, thus improving the reliability of the analyzed DNNs. The runtime analysis works by measuring and comparing the similarity between the heatmap computed for a new (unseen) input and the heatmaps computed a-priori for correct vs incorrect DNN behavior. We consider two types of defects: misclassifications and vulnerabilities to adversarial attacks. We demonstrate the debugging and runtime analysis on a case study involving a complex ResNet-based classifier trained on the RIVAL10 dataset.

Paperid: 5415, https://arxiv.org/pdf/2503.15358.pdf

Abstract:
Idiomatic expressions present a unique challenge in NLP, as their meanings are often not directly inferable from their constituent words. Despite recent advancements in Large Language Models (LLMs), idiomaticity remains a significant obstacle to robust semantic representation. We present datasets and tasks for SemEval-2025 Task 1: AdMiRe (Advancing Multimodal Idiomaticity Representation), which challenges the community to assess and improve models' ability to interpret idiomatic expressions in multimodal contexts and in multiple languages. Participants competed in two subtasks: ranking images based on their alignment with idiomatic or literal meanings, and predicting the next image in a sequence. The most effective methods achieved human-level performance by leveraging pretrained LLMs and vision-language models in mixture-of-experts settings, with multiple queries used to smooth over the weaknesses in these models' representations of idiomaticity.

Paperid: 5416, https://arxiv.org/pdf/2503.14075.pdf

Abstract:
Large vision-language models (VLMs) have demonstrated remarkable capabilities in open-world multimodal understanding, yet their high computational overheads pose great challenges for practical deployment. Some recent works have proposed methods to accelerate VLMs by pruning redundant visual tokens guided by the attention maps of VLM's early layers. Despite the success of these token pruning methods, they still suffer from two major shortcomings: (i) considerable accuracy drop due to insensitive attention signals in early layers, and (ii) limited speedup when generating long responses (e.g., 30 tokens). To address the limitations above, we present TwigVLM -- a simple and general architecture by growing a lightweight twig upon an early layer of the base VLM. Compared with most existing VLM acceleration methods purely based on visual token pruning, our TwigVLM not only achieves better accuracy retention by employing a twig-guided token pruning (TTP) strategy, but also yields higher generation speed by utilizing a self-speculative decoding (SSD) strategy. Taking LLaVA-1.5-7B as the base VLM, experimental results show that TwigVLM preserves 96% of the original performance after pruning 88.9% of visual tokens and achieves 154% speedup in generating long responses, delivering significantly better performance in terms of both accuracy and speed over the state-of-the-art VLM acceleration methods.

Paperid: 5417, https://arxiv.org/pdf/2503.13817.pdf

Abstract:
Designing reward functions for continuous-control robotics often leads to subtle misalignments or reward hacking, especially in complex tasks. Preference-based RL mitigates some of these pitfalls by learning rewards from comparative feedback rather than hand-crafted signals, yet scaling human annotations remains challenging. Recent work uses Vision-Language Models (VLMs) to automate preference labeling, but a single final-state image generally fails to capture the agent's full motion. In this paper, we present a two-part solution that both improves feedback accuracy and better aligns reward learning with the agent's policy. First, we overlay trajectory sketches on final observations to reveal the path taken, allowing VLMs to provide more reliable preferences-improving preference accuracy by approximately 15-20% in metaworld tasks. Second, we regularize reward learning by incorporating the agent's performance, ensuring that the reward model is optimized based on data generated by the current policy; this addition boosts episode returns by 20-30% in locomotion tasks. Empirical studies on metaworld demonstrate that our method achieves, for instance, around 70-80% success rate in all tasks, compared to below 50% for standard approaches. These results underscore the efficacy of combining richer visual representations with agent-aware reward regularization.

Paperid: 5418, https://arxiv.org/pdf/2503.12530.pdf

Abstract:
The field of psychology has long recognized a basic level of categorization that humans use when labeling visual stimuli, a term coined by Rosch in 1976. This level of categorization has been found to be used most frequently, to have higher information density, and to aid in visual language tasks with priming in humans. Here, we investigate basic-level categorization in two recently released, open-source vision-language models (VLMs). This paper demonstrates that Llama 3.2 Vision Instruct (11B) and Molmo 7B-D both prefer basic-level categorization consistent with human behavior. Moreover, the models' preferences are consistent with nuanced human behaviors like the biological versus non-biological basic level effects and the well-established expert basic level shift, further suggesting that VLMs acquire complex cognitive categorization behaviors from the human data on which they are trained. We also find our expert prompting methods demonstrate lower accuracy then our non-expert prompting methods, contradicting popular thought regarding the use of expertise prompting methods.

Paperid: 5419, https://arxiv.org/pdf/2503.11981.pdf

Abstract:
Text-to-3D generation saw dramatic advances in recent years by leveraging Text-to-Image models. However, most existing techniques struggle with compositional prompts, which describe multiple objects and their spatial relationships. They often fail to capture fine-grained inter-object interactions. We introduce DecompDreamer, a Gaussian splatting-based training routine designed to generate high-quality 3D compositions from such complex prompts. DecompDreamer leverages Vision-Language Models (VLMs) to decompose scenes into structured components and their relationships. We propose a progressive optimization strategy that first prioritizes joint relationship modeling before gradually shifting toward targeted object refinement. Our qualitative and quantitative evaluations against state-of-the-art text-to-3D models demonstrate that DecompDreamer effectively generates intricate 3D compositions with superior object disentanglement, offering enhanced control and flexibility in 3D generation. Project page : https://decompdreamer3d.github.io

Paperid: 5420, https://arxiv.org/pdf/2503.11549.pdf

Abstract:
The computational demands of Vision Transformers (ViTs) and Vision-Language Models (VLMs) remain a significant challenge due to the quadratic complexity of self-attention. While token pruning offers a promising solution, existing methods often introduce training overhead or fail to adapt dynamically across layers. We present SAINT, a training-free token pruning framework that leverages token similarity and a graph-based formulation to dynamically optimize pruning rates and redundancy thresholds. Through systematic analysis, we identify a universal three-stage token evolution process (aligner-explorer-aggregator) in transformers, enabling aggressive pruning in early stages without sacrificing critical information. For ViTs, SAINT doubles the throughput of ViT-H/14 at 224px with only 0.6% accuracy loss on ImageNet-1K, surpassing the closest competitor by 0.8%. For VLMs, we apply SAINT in three modes: ViT-only, LLM-only, and hybrid. SAINT reduces LLaVA-13B's tokens by 75%, achieving latency comparable to LLaVA-7B with less than 1% performance loss across benchmarks. Our work establishes a unified, practical framework for efficient inference in ViTs and VLMs.

Paperid: 5421, https://arxiv.org/pdf/2503.10814.pdf

Abstract:
Large Language Models (LLMs) are highly proficient in language-based tasks. Their language capabilities have positioned them at the forefront of the future AGI (Artificial General Intelligence) race. However, on closer inspection, Valmeekam et al. (2024); Zecevic et al. (2023); Wu et al. (2024) highlight a significant gap between their language proficiency and reasoning abilities. Reasoning in LLMs and Vision Language Models (VLMs) aims to bridge this gap by enabling these models to think and re-evaluate their actions and responses. Reasoning is an essential capability for complex problem-solving and a necessary step toward establishing trust in Artificial Intelligence (AI). This will make AI suitable for deployment in sensitive domains, such as healthcare, banking, law, defense, security etc. In recent times, with the advent of powerful reasoning models like OpenAI O1 and DeepSeek R1, reasoning endowment has become a critical research topic in LLMs. In this paper, we provide a detailed overview and comparison of existing reasoning techniques and present a systematic survey of reasoning-imbued language models. We also study current challenges and present our findings.

Paperid: 5422, https://arxiv.org/pdf/2503.10745.pdf

Abstract:
Progress in 3D vision-language learning has been hindered by the scarcity of large-scale 3D datasets. We introduce UniVLG, a unified architecture for 2D and 3D vision-language understanding that bridges the gap between existing 2D-centric models and the rich 3D sensory data available in embodied systems. Our approach initializes most model weights from pre-trained 2D models and trains on both 2D and 3D vision-language data. We propose a novel language-conditioned mask decoder shared across 2D and 3D modalities to ground objects effectively in both RGB and RGB-D images, outperforming box-based approaches. To further reduce the domain gap between 2D and 3D, we incorporate 2D-to-3D lifting strategies, enabling UniVLG to utilize 2D data to enhance 3D performance. With these innovations, our model achieves state-of-the-art performance across multiple 3D vision-language grounding tasks, demonstrating the potential of transferring advances from 2D vision-language learning to the data-constrained 3D domain. Furthermore, co-training on both 2D and 3D data enhances performance across modalities without sacrificing 2D capabilities. By removing the reliance on 3D mesh reconstruction and ground-truth object proposals, UniVLG sets a new standard for realistic, embodied-aligned evaluation. Code and additional visualizations are available at https://univlg.github.io .

Paperid: 5423, https://arxiv.org/pdf/2503.10177.pdf

Abstract:
We propose PRISM, a novel framework designed to overcome the limitations of 2D-based Preference-Based Reinforcement Learning (PBRL) by unifying 3D point cloud modeling and future-aware preference refinement. At its core, PRISM adopts a 3D Point Cloud-Language Model (3D-PC-LLM) to mitigate occlusion and viewpoint biases, ensuring more stable and spatially consistent preference signals. Additionally, PRISM leverages Chain-of-Thought (CoT) reasoning to incorporate long-horizon considerations, thereby preventing the short-sighted feedback often seen in static preference comparisons. In contrast to conventional PBRL techniques, this integration of 3D perception and future-oriented reasoning leads to significant gains in preference agreement rates, faster policy convergence, and robust generalization across unseen robotic environments. Our empirical results, spanning tasks such as robotic manipulation and autonomous navigation, highlight PRISM's potential for real-world applications where precise spatial understanding and reliable long-term decision-making are critical. By bridging 3D geometric awareness with CoT-driven preference modeling, PRISM establishes a comprehensive foundation for scalable, human-aligned reinforcement learning.

Paperid: 5424, https://arxiv.org/pdf/2503.06974.pdf

Abstract:
Learning visual semantic similarity is a critical challenge in bridging the gap between images and texts. However, there exist inherent variations between vision and language data, such as information density, i.e., images can contain textual information from multiple different views, which makes it difficult to compute the similarity between these two modalities accurately and efficiently. In this paper, we propose a novel framework called Asymmetric Visual Semantic Embedding (AVSE) to dynamically select features from various regions of images tailored to different textual inputs for similarity calculation. To capture information from different views in the image, we design a radial bias sampling module to sample image patches and obtain image features from various views, Furthermore, AVSE introduces a novel module for efficient computation of visual semantic similarity between asymmetric image and text embeddings. Central to this module is the presumption of foundational semantic units within the embeddings, denoted as ``meta-semantic embeddings." It segments all embeddings into meta-semantic embeddings with the same dimension and calculates visual semantic similarity by finding the optimal match of meta-semantic embeddings of two modalities. Our proposed AVSE model is extensively evaluated on the large-scale MS-COCO and Flickr30K datasets, demonstrating its superiority over recent state-of-the-art methods.

Paperid: 5425, https://arxiv.org/pdf/2503.01584.pdf

Abstract:
Exploration is a cornerstone of reinforcement learning (RL). Intrinsic motivation attempts to decouple exploration from external, task-based rewards. However, established approaches to intrinsic motivation that follow general principles such as information gain, often only uncover low-level interactions. In contrast, children's play suggests that they engage in meaningful high-level behavior by imitating or interacting with their caregivers. Recent work has focused on using foundation models to inject these semantic biases into exploration. However, these methods often rely on unrealistic assumptions, such as language-embedded environments or access to high-level actions. We propose SEmaNtically Sensible ExploratIon (SENSEI), a framework to equip model-based RL agents with an intrinsic motivation for semantically meaningful behavior. SENSEI distills a reward signal of interestingness from Vision Language Model (VLM) annotations, enabling an agent to predict these rewards through a world model. Using model-based RL, SENSEI trains an exploration policy that jointly maximizes semantic rewards and uncertainty. We show that in both robotic and video game-like simulations SENSEI discovers a variety of meaningful behaviors from image observations and low-level actions. SENSEI provides a general tool for learning from foundation model feedback, a crucial research direction, as VLMs become more powerful.

Paperid: 5426, https://arxiv.org/pdf/2502.20826.pdf

Abstract:
Zero-Shot Composed Image Retrieval (ZS-CIR) aims to retrieve target images by integrating information from a composed query (reference image and modification text) without training samples. Existing methods primarily combine caption models and large language models (LLMs) to generate target captions based on composed queries but face various issues such as incompatibility, visual information loss, and insufficient reasoning. In this work, we propose CoTMR, a training-free framework crafted for ZS-CIR with novel Chain-of-thought (CoT) and Multi-scale Reasoning. Instead of relying on caption models for modality transformation, CoTMR employs the Large Vision-Language Model (LVLM) to achieve unified understanding and reasoning for composed queries. To enhance the reasoning reliability, we devise CIRCoT, which guides the LVLM through a step-by-step inference process using predefined subtasks. Considering that existing approaches focus solely on global-level reasoning, our CoTMR incorporates multi-scale reasoning to achieve more comprehensive inference via fine-grained predictions about the presence or absence of key elements at the object scale. Further, we design a Multi-Grained Scoring (MGS) mechanism, which integrates CLIP similarity scores of the above reasoning outputs with candidate images to realize precise retrieval. Extensive experiments demonstrate that our CoTMR not only drastically outperforms previous methods across four prominent benchmarks but also offers appealing interpretability.

Paperid: 5427, https://arxiv.org/pdf/2502.19335.pdf

Abstract:
Large-scale machine learning models deliver strong performance across a wide range of tasks but come with significant computational and resource constraints. To mitigate these challenges, local smaller models are often deployed alongside larger models, relying on routing and deferral mechanisms to offload complex tasks. However, existing approaches inadequately balance the capabilities of these models, often resulting in unnecessary deferrals or sub-optimal resource usage. In this work we introduce a novel loss function called Gatekeeper for calibrating smaller models in cascade setups. Our approach fine-tunes the smaller model to confidently handle tasks it can perform correctly while deferring complex tasks to the larger model. Moreover, it incorporates a mechanism for managing the trade-off between model performance and deferral accuracy, and is broadly applicable across various tasks and domains without any architectural changes. We evaluate our method on encoder-only, decoder-only, and encoder-decoder architectures. Experiments across image classification, language modeling, and vision-language tasks show that our approach substantially improves deferral performance.

Paperid: 5428, https://arxiv.org/pdf/2502.16989.pdf

Abstract:
We introduce MAIA (Multimodal AI Assessment), a native-Italian benchmark designed for fine-grained investigation of the reasoning abilities of visual language models on videos. MAIA differs from other available video benchmarks for its design, its reasoning categories, the metric it uses, and the language and culture of the videos. MAIA evaluates Vision Language Models (VLMs) on two aligned tasks: a visual statement verification task, and an open-ended visual question-answering task, both on the same set of video-related questions. It considers twelve reasoning categories that aim to disentangle language and vision relations by highlighting the role of the visual input. Thanks to its carefully taught design, it evaluates VLMs' consistency and visually grounded natural language comprehension and generation simultaneously through an aggregated metric revealing low results that highlight models' fragility. Last but not least, the video collection has been carefully selected to reflect the Italian culture, and the language data are produced by native-speakers.

Paperid: 5429, https://arxiv.org/pdf/2502.15979.pdf

Abstract:
Existing zero-shot product attribute value (aspect) extraction approaches in e-Commerce industry rely on uni-modal or multi-modal models, where the sellers are asked to provide detailed textual inputs (product descriptions) for the products. However, manually providing (typing) the product descriptions is time-consuming and frustrating for the sellers. Thus, we propose a cross-modal zero-shot attribute value generation framework (ViOC-AG) based on CLIP, which only requires product images as the inputs. ViOC-AG follows a text-only training process, where a task-customized text decoder is trained with the frozen CLIP text encoder to alleviate the modality gap and task disconnection. During the zero-shot inference, product aspects are generated by the frozen CLIP image encoder connected with the trained task-customized text decoder. OCR tokens and outputs from a frozen prompt-based LLM correct the decoded outputs for out-of-domain attribute values. Experiments show that ViOC-AG significantly outperforms other fine-tuned vision-language models for zero-shot attribute value extraction.

Paperid: 5430, https://arxiv.org/pdf/2502.15812.pdf

Abstract:
In the evolving landscape of multimodal language models, understanding the nuanced meanings conveyed through visual cues - such as satire, insult, or critique - remains a significant challenge. Existing evaluation benchmarks primarily focus on direct tasks like image captioning or are limited to a narrow set of categories, such as humor or satire, for deep semantic understanding. To address this gap, we introduce, for the first time, a comprehensive, multi-level Chinese-based benchmark designed specifically for evaluating the understanding of implicit meanings in images. This benchmark is systematically categorized into four subtasks: surface-level content understanding, symbolic meaning interpretation, background knowledge comprehension, and implicit meaning comprehension. We propose an innovative semi-automatic method for constructing datasets, adhering to established construction protocols. Using this benchmark, we evaluate 15 open-source large vision language models (LVLMs) and GPT-4o, revealing that even the best-performing model lags behind human performance by nearly 14% in understanding implicit meaning. Our findings underscore the intrinsic challenges current LVLMs face in grasping nuanced visual semantics, highlighting significant opportunities for future research and development in this domain. We will publicly release our InsightVision dataset, code upon acceptance of the paper.

Paperid: 5431, https://arxiv.org/pdf/2502.14906.pdf

Abstract:
Investigating value alignment in Large Language Models (LLMs) based on cultural context has become a critical area of research. However, similar biases have not been extensively explored in large vision-language models (VLMs). As the scale of multimodal models continues to grow, it becomes increasingly important to assess whether images can serve as reliable proxies for culture and how these values are embedded through the integration of both visual and textual data. In this paper, we conduct a thorough evaluation of multimodal model at different scales, focusing on their alignment with cultural values. Our findings reveal that, much like LLMs, VLMs exhibit sensitivity to cultural values, but their performance in aligning with these values is highly context-dependent. While VLMs show potential in improving value understanding through the use of images, this alignment varies significantly across contexts highlighting the complexities and underexplored challenges in the alignment of multimodal models.

Paperid: 5432, https://arxiv.org/pdf/2502.13928.pdf

Abstract:
Recent studies have shown that Large Vision-Language Models (VLMs) tend to neglect image content and over-rely on language-model priors, resulting in errors in visually grounded tasks and hallucinations. We hypothesize that this issue arises because existing VLMs are not explicitly trained to generate texts that are accurately grounded in fine-grained image details. To enhance visual feedback during VLM training, we propose S-VCO (Symmetrical Visual Contrastive Optimization), a novel finetuning objective that steers the model toward capturing important visual details and aligning them with corresponding text tokens. To further facilitate this detailed alignment, we introduce MVC, a paired image-text dataset built by automatically filtering and augmenting visual counterfactual data to challenge the model with hard contrastive cases involving Minimal Visual Contrasts. Experiments show that our method consistently improves VLM performance across diverse benchmarks covering various abilities and domains, achieving up to a 22% reduction in hallucinations, and significant gains in vision-centric and general tasks. Notably, these improvements become increasingly pronounced in benchmarks with higher visual dependency. In short, S-VCO offers a significant enhancement of VLM's visually-dependent task performance while retaining or even improving the model's general abilities. We opensource our code at https://s-vco.github.io/

Paperid: 5433, https://arxiv.org/pdf/2502.09238.pdf

Abstract:
The increasing demand for efficient last-mile delivery in smart logistics underscores the role of autonomous robots in enhancing operational efficiency and reducing costs. Traditional navigation methods, which depend on high-precision maps, are resource-intensive, while learning-based approaches often struggle with generalization in real-world scenarios. To address these challenges, this work proposes the Openstreetmap-enhanced oPen-air sEmantic Navigation (OPEN) system that combines foundation models with classic algorithms for scalable outdoor navigation. The system uses off-the-shelf OpenStreetMap (OSM) for flexible map representation, thereby eliminating the need for extensive pre-mapping efforts. It also employs Large Language Models (LLMs) to comprehend delivery instructions and Vision-Language Models (VLMs) for global localization, map updates, and house number recognition. To compensate the limitations of existing benchmarks that are inadequate for assessing last-mile delivery, this work introduces a new benchmark specifically designed for outdoor navigation in residential areas, reflecting the real-world challenges faced by autonomous delivery systems. Extensive experiments in simulated and real-world environments demonstrate the proposed system's efficacy in enhancing navigation efficiency and reliability. To facilitate further research, our code and benchmark are publicly available.

Paperid: 5434, https://arxiv.org/pdf/2502.08079.pdf

Abstract:
Current adversarial attacks for evaluating the robustness of vision-language pre-trained (VLP) models in multi-modal tasks suffer from limited transferability, where attacks crafted for a specific model often struggle to generalize effectively across different models, limiting their utility in assessing robustness more broadly. This is mainly attributed to the over-reliance on model-specific features and regions, particularly in the image modality. In this paper, we propose an elegant yet highly effective method termed Meticulous Adversarial Attack (MAA) to fully exploit model-independent characteristics and vulnerabilities of individual samples, achieving enhanced generalizability and reduced model dependence. MAA emphasizes fine-grained optimization of adversarial images by developing a novel resizing and sliding crop (RScrop) technique, incorporating a multi-granularity similarity disruption (MGSD) strategy. Extensive experiments across diverse VLP models, multiple benchmark datasets, and a variety of downstream tasks demonstrate that MAA significantly enhances the effectiveness and transferability of adversarial attacks. A large cohort of performance studies is conducted to generate insights into the effectiveness of various model configurations, guiding future advancements in this domain.

Paperid: 5435, https://arxiv.org/pdf/2502.03564.pdf

Abstract:
Effective visual accessibility in Virtual Reality (VR) is crucial for Blind and Low Vision (BLV) users. However, designing visual accessibility systems is challenging due to the complexity of 3D VR environments and the need for techniques that can be easily retrofitted into existing applications. While prior work has studied how to enhance or translate visual information, the advancement of Vision Language Models (VLMs) provides an exciting opportunity to advance the scene interpretation capability of current systems. This paper presents EnVisionVR, an accessibility tool for VR scene interpretation. Through a formative study of usability barriers, we confirmed the lack of visual accessibility features as a key barrier for BLV users of VR content and applications. In response, we designed and developed EnVisionVR, a novel visual accessibility system leveraging a VLM, voice input and multimodal feedback for scene interpretation and virtual object interaction in VR. An evaluation with 12 BLV users demonstrated that EnVisionVR significantly improved their ability to locate virtual objects, effectively supporting scene understanding and object interaction.

Paperid: 5436, https://arxiv.org/pdf/2501.18753.pdf

Abstract:
Task-generic promptable image segmentation aims to achieve segmentation of diverse samples under a single task description by utilizing only one task-generic prompt. Current methods leverage the generalization capabilities of Vision-Language Models (VLMs) to infer instance-specific prompts from these task-generic prompts in order to guide the segmentation process. However, when VLMs struggle to generalise to some image instances, predicting instance-specific prompts becomes poor. To solve this problem, we introduce \textbf{I}nstance-specific \textbf{N}egative Mining for \textbf{T}ask-Generic Promptable Segmentation (\textbf{INT}). The key idea of INT is to adaptively reduce the influence of irrelevant (negative) prior knowledge whilst to increase the use the most plausible prior knowledge, selected by negative mining with higher contrast, in order to optimise instance-specific prompts generation. Specifically, INT consists of two components: (1) instance-specific prompt generation, which progressively fliters out incorrect information in prompt generation; (2) semantic mask generation, which ensures each image instance segmentation matches correctly the semantics of the instance-specific prompts. INT is validated on six datasets, including camouflaged objects and medical images, demonstrating its effectiveness, robustness and scalability.

Paperid: 5437, https://arxiv.org/pdf/2501.12356.pdf

Abstract:
Radiology plays a pivotal role in modern medicine due to its non-invasive diagnostic capabilities. However, the manual generation of unstructured medical reports is time consuming and prone to errors. It creates a significant bottleneck in clinical workflows. Despite advancements in AI-generated radiology reports, challenges remain in achieving detailed and accurate report generation. In this study we have evaluated different combinations of multimodal models that integrate Computer Vision and Natural Language Processing to generate comprehensive radiology reports. We employed a pretrained Vision Transformer (ViT-B16) and a SWIN Transformer as the image encoders. The BART and GPT-2 models serve as the textual decoders. We used Chest X-ray images and reports from the IU-Xray dataset to evaluate the usability of the SWIN Transformer-BART, SWIN Transformer-GPT-2, ViT-B16-BART and ViT-B16-GPT-2 models for report generation. We aimed at finding the best combination among the models. The SWIN-BART model performs as the best-performing model among the four models achieving remarkable results in almost all the evaluation metrics like ROUGE, BLEU and BERTScore.

Paperid: 5438, https://arxiv.org/pdf/2501.09039.pdf

Abstract:
The rapid advancement of Large Vision-Language Models (LVLMs) has enhanced capabilities offering potential applications from content creation to productivity enhancement. Despite their innovative potential, LVLMs exhibit vulnerabilities, especially in generating potentially toxic or unsafe responses. Malicious actors can exploit these vulnerabilities to propagate toxic content in an automated (or semi-) manner, leveraging the susceptibility of LVLMs to deception via strategically crafted prompts without fine-tuning or compute-intensive procedures. Despite the red-teaming efforts and inherent potential risks associated with the LVLMs, exploring vulnerabilities of LVLMs remains nascent and yet to be fully addressed in a systematic manner. This study systematically examines the vulnerabilities of open-source LVLMs, including LLaVA, InstructBLIP, Fuyu, and Qwen, using adversarial prompt strategies that simulate real-world social manipulation tactics informed by social theories. Our findings show that (i) toxicity and insulting are the most prevalent behaviors, with the mean rates of 16.13% and 9.75%, respectively; (ii) Qwen-VL-Chat, LLaVA-v1.6-Vicuna-7b, and InstructBLIP-Vicuna-7b are the most vulnerable models, exhibiting toxic response rates of 21.50%, 18.30% and 17.90%, and insulting responses of 13.40%, 11.70% and 10.10%, respectively; (iii) prompting strategies incorporating dark humor and multimodal toxic prompt completion significantly elevated these vulnerabilities. Despite being fine-tuned for safety, these models still generate content with varying degrees of toxicity when prompted with adversarial inputs, highlighting the urgent need for enhanced safety mechanisms and robust guardrails in LVLM development.

Paperid: 5439, https://arxiv.org/pdf/2501.03675.pdf

Abstract:
Vision-Language Models (VLMs) excel at understanding single images, aided by high-quality instruction datasets. However, multi-image reasoning remains underexplored in the open-source community due to two key challenges: (1) scaling datasets with correlated images and complex reasoning instructions is resource-intensive, and (2) robust evaluation benchmarks for multi-image tasks are lacking. To address this, we introduce SMiR, a synthetic data-generation pipeline for multi-image reasoning, along with a high-quality dataset generated using this pipeline. SMiR efficiently extracts correlated images via multimodal embeddings, integrates visual and descriptive information, and leverages open-source LLMs to generate quality instructions. Using this approach, we produce 160K synthetic training samples, offering a cost-effective alternative to closed-source solutions. Additionally, we present SMiR-Bench, a multi-image reasoning benchmark comprising 200 diverse examples across seven complex reasoning tasks. SMiR-Bench is multi-turn and employs a VLM judge to evaluate free-form responses, providing a comprehensive assessment of model expressiveness and reasoning capability across modalities. We demonstrate the effectiveness of SMiR by fine-tuning open-source VLMs and evaluating them on SMiR-Bench.

Paperid: 5440, https://arxiv.org/pdf/2412.17610.pdf

Abstract:
The personalization model has gained significant attention in image generation yet remains underexplored for large vision-language models (LVLMs). Beyond generic ones, with personalization, LVLMs handle interactive dialogues using referential concepts (e.g., ``Mike and Susan are talking.'') instead of the generic form (e.g., ``a boy and a girl are talking.''), making the conversation more customizable and referentially friendly. In addition, PLVM is equipped to continuously add new concepts during a dialogue without incurring additional costs, which significantly enhances the practicality. PLVM proposes Aligner, a pre-trained visual encoder to align referential concepts with the queried images. During the dialogues, it extracts features of reference images with these corresponding concepts and recognizes them in the queried image, enabling personalization. We note that the computational cost and parameter count of the Aligner are negligible within the entire framework. With comprehensive qualitative and quantitative analyses, we reveal the effectiveness and superiority of PLVM.

Paperid: 5441, https://arxiv.org/pdf/2412.16334.pdf

Abstract:
Self-supervised visual foundation models produce powerful embeddings that achieve remarkable performance on a wide range of downstream tasks. However, unlike vision-language models such as CLIP, self-supervised visual features are not readily aligned with language, hindering their adoption in open-vocabulary tasks. Our method, named dino.txt, unlocks this new ability for DINOv2, a widely used self-supervised visual encoder. We build upon the LiT training strategy, which trains a text encoder to align with a frozen vision model but leads to unsatisfactory results on dense tasks. We propose several key ingredients to improve performance on both global and dense tasks, such as concatenating the [CLS] token with the patch average to train the alignment and curating data using both text and image modalities. With these, we successfully train a CLIP-like model with only a fraction of the computational cost compared to CLIP while achieving state-of-the-art results in zero-shot classification and open-vocabulary semantic segmentation.

Paperid: 5442, https://arxiv.org/pdf/2412.11087.pdf

Abstract:
Composed Image Retrieval (CIR) aims to retrieve target images from candidate set using a hybrid-modality query consisting of a reference image and a relative caption that describes the user intent. Recent studies attempt to utilize Vision-Language Pre-training Models (VLPMs) with various fusion strategies for addressing the task.However, these methods typically fail to simultaneously meet two key requirements of CIR: comprehensively extracting visual information and faithfully following the user intent. In this work, we propose CIR-LVLM, a novel framework that leverages the large vision-language model (LVLM) as the powerful user intent-aware encoder to better meet these requirements. Our motivation is to explore the advanced reasoning and instruction-following capabilities of LVLM for accurately understanding and responding the user intent. Furthermore, we design a novel hybrid intent instruction module to provide explicit intent guidance at two levels: (1) The task prompt clarifies the task requirement and assists the model in discerning user intent at the task level. (2) The instance-specific soft prompt, which is adaptively selected from the learnable prompt pool, enables the model to better comprehend the user intent at the instance level compared to a universal prompt for all instances. CIR-LVLM achieves state-of-the-art performance across three prominent benchmarks with acceptable inference efficiency. We believe this study provides fundamental insights into CIR-related fields.

Paperid: 5443, https://arxiv.org/pdf/2412.08271.pdf

Abstract:
Point cloud completion aims to recover partial geometric and topological shapes caused by equipment defects or limited viewpoints. Current methods either solely rely on the 3D coordinates of the point cloud to complete it or incorporate additional images with well-calibrated intrinsic parameters to guide the geometric estimation of the missing parts. Although these methods have achieved excellent performance by directly predicting the location of complete points, the extracted features lack fine-grained information regarding the location of the missing area. To address this issue, we propose a rapid and efficient method to expand an unimodal framework into a multimodal framework. This approach incorporates a position-aware module designed to enhance the spatial information of the missing parts through a weighted map learning mechanism. In addition, we establish a Point-Text-Image triplet corpus PCI-TI and MVP-TI based on the existing unimodal point cloud completion dataset and use the pre-trained vision-language model CLIP to provide richer detail information for 3D shapes, thereby enhancing performance. Extensive quantitative and qualitative experiments demonstrate that our method outperforms state-of-the-art point cloud completion methods.

Paperid: 5444, https://arxiv.org/pdf/2411.16407.pdf

Abstract:
Despite the recent progress in deep learning based computer vision, domain shifts are still one of the major challenges. Semantic segmentation for autonomous driving faces a wide range of domain shifts, e.g. caused by changing weather conditions, new geolocations and the frequent use of synthetic data in model training. Unsupervised domain adaptation (UDA) methods have emerged which adapt a model to a new target domain by only using unlabeled data of that domain. The variety of UDA methods is large but all of them use ImageNet pre-trained models. Recently, vision-language models have demonstrated strong generalization capabilities which may facilitate domain adaptation. We show that simply replacing the encoder of existing UDA methods like DACS by a vision-language pre-trained encoder can result in significant performance improvements of up to 10.0% mIoU on the GTA5-to-Cityscapes domain shift. For the generalization performance to unseen domains, the newly employed vision-language pre-trained encoder provides a gain of up to 13.7% mIoU across three unseen datasets. However, we find that not all UDA methods can be easily paired with the new encoder and that the UDA performance does not always likewise transfer into generalization performance. Finally, we perform our experiments on an adverse weather condition domain shift to further verify our findings on a pure real-to-real domain shift.

Paperid: 5445, https://arxiv.org/pdf/2411.08606.pdf

Abstract:
The ability of gaze estimation models to generalize is often significantly hindered by various factors unrelated to gaze, especially when the training dataset is limited. Current strategies aim to address this challenge through different domain generalization techniques, yet they have had limited success due to the risk of overfitting when solely relying on value labels for regression. Recent progress in pre-trained vision-language models has motivated us to capitalize on the abundant semantic information available. We propose a novel approach in this paper, reframing the gaze estimation task as a vision-language alignment issue. Our proposed framework, named Language-Guided Gaze Estimation (LG-Gaze), learns continuous and geometry-sensitive features for gaze estimation benefit from the rich prior knowledges of vision-language models. Specifically, LG-Gaze aligns gaze features with continuous linguistic features through our proposed multimodal contrastive regression loss, which customizes adaptive weights for different negative samples. Furthermore, to better adapt to the labels for gaze estimation task, we propose a geometry-aware interpolation method to obtain more precise gaze embeddings. Through extensive experiments, we validate the efficacy of our framework in four different cross-domain evaluation tasks.

Paperid: 5446, https://arxiv.org/pdf/2411.07848.pdf

Abstract:
Large scale scenes such as multifloor homes can be robustly and efficiently mapped with a 3D graph of landmarks estimated jointly with robot poses in a factor graph, a technique commonly used in commercial robots such as drones and robot vacuums. In this work, we propose Language-Inferred Factor Graph for Instruction Following (LIFGIF), a zero-shot method to ground natural language instructions in such a map. LIFGIF also includes a policy for following natural language navigation instructions in a novel environment while the map is constructed, enabling robust navigation performance in the physical world. To evaluate LIFGIF, we present a new dataset, Object-Centric VLN (OC-VLN), in order to evaluate grounding of object-centric natural language navigation instructions. We compare to two state-of-the-art zero-shot baselines from related tasks, Object Goal Navigation and Vision Language Navigation, to demonstrate that LIFGIF outperforms them across all our evaluation metrics on OCVLN. Finally, we successfully demonstrate the effectiveness of LIFGIF for performing zero-shot object-centric instruction following in the real world on a Boston Dynamics Spot robot.

Paperid: 5447, https://arxiv.org/pdf/2411.07076.pdf

Abstract:
Existing large vision-language models (LVLMs) are largely limited to processing short, seconds-long videos and struggle with generating coherent descriptions for extended video spanning minutes or more. Long video description introduces new challenges, such as consistent character identification and plot-level descriptions incorporating both visual and audio information. To address these, we figure out audio-visual character identification, matching character names to each dialogue, as a key factor. We propose StoryTeller, a system for generating dense descriptions of long videos, incorporating both low-level visual concepts and high-level plot information. StoryTeller uses a multimodal large language model that integrates visual, audio, and text modalities to perform audio-visual character identification on minute-long video clips. The results are then fed into a LVLM to enhance consistency of video description. We validate our approach on movie description tasks and introduce MovieStory101, a dataset with dense descriptions for three-minute movie clips. To evaluate long video descriptions, we create StoryQA, a large set of multiple-choice questions for MovieStory101 test set. We assess descriptions by inputting them into GPT-4 to answer these questions, using accuracy as an automatic evaluation metric. Experiments show that StoryTeller outperforms all open and closed-source baselines on StoryQA, achieving 9.5% higher accuracy than the strongest baseline, Gemini-1.5-pro, and demonstrating a +15.56% advantage in human side-by-side evaluations. Additionally, incorporating audio-visual character identification from StoryTeller improves the performance of all video description models, with Gemini-1.5-pro and GPT-4o showing relative improvement of 5.5% and 13.0%, respectively, in accuracy on StoryQA.

Paperid: 5448, https://arxiv.org/pdf/2410.22029.pdf

Abstract:
Vision Language Models excel in handling a wide range of complex tasks, including Optical Character Recognition (OCR), Visual Question Answering (VQA), and advanced geometric reasoning. However, these models fail to perform well on low-level basic visual tasks which are especially easy for humans. Our goal in this work was to determine if these models are truly "blind" to geometric reasoning or if there are ways to enhance their capabilities in this area. Our work presents a novel automatic pipeline designed to extract key information from images in response to specific questions. Instead of just relying on direct VQA, we use question-derived keywords to create a caption that highlights important details in the image related to the question. This caption is then used by a language model to provide a precise answer to the question without requiring external fine-tuning.

Paperid: 5449, https://arxiv.org/pdf/2410.20265.pdf

Abstract:
We study the impact of a standard practice in compressing foundation vision-language models - quantization - on the models' ability to produce socially-fair outputs. In contrast to prior findings with unimodal models that compression consistently amplifies social biases, our extensive evaluation of four quantization settings across three datasets and three CLIP variants yields a surprising result: while individual models demonstrate bias, we find no consistent change in bias magnitude or direction across a population of compressed models due to quantization.

Paperid: 5450, https://arxiv.org/pdf/2410.08860.pdf

Abstract:
Audio descriptions (ADs) function as acoustic commentaries designed to assist blind persons and persons with visual impairments in accessing digital media content on television and in movies, among other settings. As an accessibility service typically provided by trained AD professionals, the generation of ADs demands significant human effort, making the process both time-consuming and costly. Recent advancements in natural language processing (NLP) and computer vision (CV), particularly in large language models (LLMs) and vision-language models (VLMs), have allowed for getting a step closer to automatic AD generation. This paper reviews the technologies pertinent to AD generation in the era of LLMs and VLMs: we discuss how state-of-the-art NLP and CV technologies can be applied to generate ADs and identify essential research directions for the future.

Paperid: 5451, https://arxiv.org/pdf/2410.05928.pdf

Abstract:
Vision-Language Models (VLMs) have transformed tasks requiring visual and reasoning abilities, such as image retrieval and Visual Question Answering (VQA). Despite their success, VLMs face significant challenges with tasks involving geometric reasoning, algebraic problem-solving, and counting. These limitations stem from difficulties effectively integrating multiple modalities and accurately interpreting geometry-related tasks. Various works claim that introducing a captioning pipeline before VQA tasks enhances performance. We incorporated this pipeline for tasks involving geometry, algebra, and counting. We found that captioning results are not generalizable, specifically with larger VLMs primarily trained on downstream QnA tasks showing random performance on math-related challenges. However, we present a promising alternative: task-based prompting, enriching the prompt with task-specific guidance. This approach shows promise and proves more effective than direct captioning methods for math-heavy problems.

Paperid: 5452, https://arxiv.org/pdf/2409.19684.pdf

Abstract:
Medicine is inherently multimodal and multitask, with diverse data modalities spanning text, imaging. However, most models in medical field are unimodal single tasks and lack good generalizability and explainability. In this study, we introduce MedViLaM, a unified vision-language model towards a generalist model for medical data that can flexibly encode and interpret various forms of medical data, including clinical language and imaging, all using the same set of model weights. To facilitate the creation of such multi-task model, we have curated MultiMedBench, a comprehensive pretaining dataset and benchmark consisting of several distinct tasks, i.e., continuous question-answering, multi-label disease classification, disease localization, generation and summarization of radiology reports. MedViLaM demonstrates strong performance across all MultiMedBench tasks, frequently outpacing other generalist models by a significant margin. Additionally, we present instances of zero-shot generalization to new medical concepts and tasks, effective transfer learning across different tasks, and the emergence of zero-shot medical reasoning.

Paperid: 5453, https://arxiv.org/pdf/2409.18023.pdf

Abstract:
Vision Language Models (VLMs) extend remarkable capabilities of text-only large language models and vision-only models, and are able to learn from and process multi-modal vision-text input. While modern VLMs perform well on a number of standard image classification and image-text matching tasks, they still struggle with a number of crucial vision-language (VL) reasoning abilities such as counting and spatial reasoning. Moreover, while they might be very brittle to small variations in instructions and/or evaluation protocols, existing benchmarks fail to evaluate their robustness (or rather the lack of it). In order to couple challenging VL scenarios with comprehensive robustness evaluation, we introduce DARE, Diverse Visual Question Answering with Robustness Evaluation, a carefully created and curated multiple-choice VQA benchmark. DARE evaluates VLM performance on five diverse categories and includes four robustness-oriented evaluations based on the variations of: prompts, the subsets of answer options, the output format and the number of correct answers. Among a spectrum of other findings, we report that state-of-the-art VLMs still struggle with questions in most categories and are unable to consistently deliver their peak performance across the tested robustness evaluations. The worst case performance across the subsets of options is up to 34% below the performance in the standard case. The robustness of the open-source VLMs such as LLaVA 1.6 and Idefics2 cannot match the closed-source models such as GPT-4 and Gemini, but even the latter remain very brittle to different variations.

Paperid: 5454, https://arxiv.org/pdf/2409.17610.pdf

Abstract:
The rocketing prosperity of large language models (LLMs) in recent years has boosted the prevalence of vision-language models (VLMs) in the medical sector. In our online medical consultation scenario, a doctor responds to the texts and images provided by a patient in multiple rounds to diagnose her/his health condition, forming a multi-turn multimodal medical dialogue format. Unlike high-quality images captured by professional equipment in traditional medical visual question answering (Med-VQA), the images in our case are taken by patients' mobile phones. These images have poor quality control, with issues such as excessive background elements and the lesion area being significantly off-center, leading to degradation of vision-language alignment in the model training phase. In this paper, we propose ZALM3, a Zero-shot strategy to improve vision-language ALignment in Multi-turn Multimodal Medical dialogue. Since we observe that the preceding text conversations before an image can infer the regions of interest (RoIs) in the image, ZALM3 employs an LLM to summarize the keywords from the preceding context and a visual grounding model to extract the RoIs. The updated images eliminate unnecessary background noise and provide more effective vision-language alignment. To better evaluate our proposed method, we design a new subjective assessment metric for multi-turn unimodal/multimodal medical dialogue to provide a fine-grained performance comparison. Our experiments across three different clinical departments remarkably demonstrate the efficacy of ZALM3 with statistical significance.

Paperid: 5455, https://arxiv.org/pdf/2409.13675.pdf

Abstract:
Service robots in human-centered environments such as hospitals, office buildings, and long-term care homes need to navigate while adhering to social norms to ensure the safety and comfortability of the people they are sharing the space with. Furthermore, they need to adapt to new social scenarios that can arise during robot navigation. In this paper, we present a novel Online Lifelong Vision Language architecture, OLiVia- Nav, which uniquely integrates vision-language models (VLMs) with an online lifelong learning framework for robot social navigation. We introduce a unique distillation approach, Social Context Contrastive Language Image Pre-training (SC-CLIP), to transfer the social reasoning capabilities of large VLMs to a lightweight VLM, in order for OLiVia-Nav to directly encode social and environment context during robot navigation. These encoded embeddings are used to generate and select robot social compliant trajectories. The lifelong learning capabilities of SC-CLIP enable OLiVia-Nav to update the robot trajectory planning overtime as new social scenarios are encountered. We conducted extensive real-world experiments in diverse social navigation scenarios. The results showed that OLiVia-Nav outperformed existing state-of-the-art DRL and VLM methods in terms of mean squared error, Hausdorff loss, and personal space violation duration. Ablation studies also verified the design choices for OLiVia-Nav.

Paperid: 5456, https://arxiv.org/pdf/2409.10719.pdf

Abstract:
Vision language models (VLMs) have shown strong zero-shot generalization across various tasks, especially when integrated with large language models (LLMs). However, their ability to comprehend rhetorical and persuasive visual media, such as advertisements, remains understudied. Ads often employ atypical imagery, using surprising object juxtapositions to convey shared properties. For example, Fig. 1 (e) shows a beer with a feather-like texture. This requires advanced reasoning to deduce that this atypical representation signifies the beer's lightness. We introduce three novel tasks, Multi-label Atypicality Classification, Atypicality Statement Retrieval, and Aypical Object Recognition, to benchmark VLMs' understanding of atypicality in persuasive images. We evaluate how well VLMs use atypicality to infer an ad's message and test their reasoning abilities by employing semantically challenging negatives. Finally, we pioneer atypicality-aware verbalization by extracting comprehensive image descriptions sensitive to atypical elements. Our findings reveal that: (1) VLMs lack advanced reasoning capabilities compared to LLMs; (2) simple, effective strategies can extract atypicality-aware information, leading to comprehensive image verbalization; (3) atypicality aids persuasive advertisement understanding. Code and data will be made available.

Paperid: 5457, https://arxiv.org/pdf/2409.10488.pdf

Abstract:
For a vision-language model (VLM) to understand the physical world, such as cause and effect, a first step is to capture the temporal dynamics of the visual world, for example how the physical states of objects evolve over time (e.g. a whole apple into a sliced apple). Our paper aims to investigate if VLMs pre-trained on web-scale data learn to encode object states, which can be extracted with zero-shot text prompts. We curate an object state recognition dataset ChangeIt-Frames, and evaluate nine open-source VLMs, including models trained with contrastive and generative objectives. We observe that while these state-of-the-art vision-language models can reliably perform object recognition, they consistently fail to accurately distinguish the objects' physical states. Through extensive experiments, we identify three areas for improvements for VLMs to better encode object states, namely the quality of object localization, the architecture to bind concepts to objects, and the objective to learn discriminative visual and language encoders on object states. Data and code are released.

Paperid: 5458, https://arxiv.org/pdf/2409.10283.pdf

Abstract:
In the rapidly evolving field of vision-language navigation (VLN), ensuring safety for physical agents remains an open challenge. For a human-in-the-loop language-operated drone to navigate safely, it must understand natural language commands, perceive the environment, and simultaneously avoid hazards in real time. Control Barrier Functions (CBFs) are formal methods that enforce safe operating conditions. Model Predictive Control (MPC) is an optimization framework that plans a sequence of future actions over a prediction horizon, ensuring smooth trajectory tracking while obeying constraints. In this work, we consider a VLN-operated drone platform and enhance its safety by formulating a novel scene-aware CBF that leverages ego-centric observations from a camera which has both Red-Green-Blue as well as Depth (RGB-D) channels. A CBF-less baseline system uses a Vision-Language Encoder with cross-modal attention to convert commands into an ordered sequence of landmarks. An object detection model identifies and verifies these landmarks in the captured images to generate a planned path. To further enhance safety, an Adaptive Safety Margin Algorithm (ASMA) is proposed. ASMA tracks moving objects and performs scene-aware CBF evaluation on-the-fly, which serves as an additional constraint within the MPC framework. By continuously identifying potentially risky observations, the system performs prediction in real time about unsafe conditions and proactively adjusts its control actions to maintain safe navigation throughout the trajectory. Deployed on a Parrot Bebop2 quadrotor in the Gazebo environment using the Robot Operating System (ROS), ASMA achieves 64%-67% increase in success rates with only a slight increase (1.4%-5.8%) in trajectory lengths compared to the baseline CBF-less VLN.

Paperid: 5459, https://arxiv.org/pdf/2409.09788.pdf

Abstract:
Despite recent advances demonstrating vision-language models' (VLMs) abilities to describe complex relationships in images using natural language, their capability to quantitatively reason about object sizes and distances remains underexplored. In this work, we introduce a manually annotated benchmark, Q-Spatial Bench, with 271 questions across five categories designed for quantitative spatial reasoning and systematically investigate the performance of state-of-the-art VLMs on this task. Our analysis reveals that reasoning about distances between objects is particularly challenging for SoTA VLMs; however, some VLMs significantly outperform others, with an over 40-point gap between the two best performing models. We also make the surprising observation that the success rate of the top-performing VLM increases by 19 points when a reasoning path using a reference object emerges naturally in the response. Inspired by this observation, we develop a zero-shot prompting technique, SpatialPrompt, that encourages VLMs to answer quantitative spatial questions using reference objects as visual cues. By instructing VLMs to use reference objects in their reasoning paths via SpatialPrompt, Gemini 1.5 Pro, Gemini 1.5 Flash, and GPT-4V improve their success rates by over 40, 20, and 30 points, respectively. We emphasize that these significant improvements are obtained without needing more data, model architectural modifications, or fine-tuning.

Paperid: 5460, https://arxiv.org/pdf/2409.02418.pdf

Abstract:
Owing to a large amount of multi-modal data in modern medical systems, such as medical images and reports, Medical Vision-Language Pre-training (Med-VLP) has demonstrated incredible achievements in coarse-grained downstream tasks (i.e., medical classification, retrieval, and visual question answering). However, the problem of transferring knowledge learned from Med-VLP to fine-grained multi-organ segmentation tasks has barely been investigated. Multi-organ segmentation is challenging mainly due to the lack of large-scale fully annotated datasets and the wide variation in the shape and size of the same organ between individuals with different diseases. In this paper, we propose a novel pre-training & fine-tuning framework for Multi-Organ Segmentation by harnessing Medical repOrt Supervision (MOSMOS). Specifically, we first introduce global contrastive learning to maximally align the medical image-report pairs in the pre-training stage. To remedy the granularity discrepancy, we further leverage multi-label recognition to implicitly learn the semantic correspondence between image pixels and organ tags. More importantly, our pre-trained models can be transferred to any segmentation model by introducing the pixel-tag attention maps. Different network settings, i.e., 2D U-Net and 3D UNETR, are utilized to validate the generalization. We have extensively evaluated our approach using different diseases and modalities on BTCV, AMOS, MMWHS, and BRATS datasets. Experimental results in various settings demonstrate the effectiveness of our framework. This framework can serve as the foundation to facilitate future research on automatic annotation tasks under the supervision of medical reports.

Paperid: 5461, https://arxiv.org/pdf/2601.21826.pdf

Abstract:
As large language models (LLMs) are applied to increasingly longer and more complex tasks, there is a growing need for realistic long-context benchmarks that require selective reading and integration of heterogeneous, multi-modal information sources. This need is especially acute for geospatial planning problems, such as those found in planning for large-scale military operations, which demand fast and accurate reasoning over maps, orders, intelligence reports, and other distributed data. To address this gap, we present MilSCORE (Military Scenario Contextual Reasoning), to our knowledge the first scenario-level dataset of expert-authored, multi-hop questions grounded in a complex, simulated military planning scenario used for training. MilSCORE is designed to evaluate high-stakes decision-making and planning, probing LLMs' ability to combine tactical and spatial reasoning across multiple sources and to reason over long-horizon, geospatially rich context. The benchmark includes a diverse set of question types across seven categories targeting both factual recall and multi-step reasoning about constraints, strategy, and spatial analysis. We provide an evaluation protocol and report baseline results for a range of contemporary vision-language models. Our findings highlight substantial headroom on MilSCORE, indicating that current systems struggle with realistic, scenario-level long-context planning, and positioning MilSCORE as a challenging testbed for future work.

Paperid: 5462, https://arxiv.org/pdf/2601.21798.pdf

Abstract:
Large Language Models(LLMs) have revolutionized text generation and multimodal perception, but their capabilities in 3D content generation remain underexplored. Existing methods compromise by producing either low-resolution meshes or coarse structural proxies, failing to capture fine-grained geometry natively. In this paper, we propose CG-MLLM, a novel Multi-modal Large Language Model (MLLM) capable of 3D captioning and high-resolution 3D generation in a single framework. Leveraging the Mixture-of-Transformer architecture, CG-MLLM decouples disparate modeling needs, where the Token-level Autoregressive (TokenAR) Transformer handles token-level content, and the Block-level Autoregressive (BlockAR) Transformer handles block-level content. By integrating a pre-trained vision-language backbone with a specialized 3D VAE latent space, CG-MLLM facilitates long-context interactions between standard tokens and spatial blocks within a single integrated architecture. Experimental results show that CG-MLLM significantly outperforms existing MLLMs in generating high-fidelity 3D objects, effectively bringing high-resolution 3D content creation into the mainstream LLM paradigm.

Paperid: 5463, https://arxiv.org/pdf/2601.21602.pdf

Abstract:
While Vision-Language-Action (VLA) models have achieved remarkable success in ground-based embodied intelligence, their application to Aerial Manipulation Systems (AMS) remains a largely unexplored frontier. The inherent characteristics of AMS, including floating-base dynamics, strong coupling between the UAV and the manipulator, and the multi-step, long-horizon nature of operational tasks, pose severe challenges to existing VLA paradigms designed for static or 2D mobile bases. To bridge this gap, we propose AIR-VLA, the first VLA benchmark specifically tailored for aerial manipulation. We construct a physics-based simulation environment and release a high-quality multimodal dataset comprising 3000 manually teleoperated demonstrations, covering base manipulation, object & spatial understanding, semantic reasoning, and long-horizon planning. Leveraging this platform, we systematically evaluate mainstream VLA models and state-of-the-art VLM models. Our experiments not only validate the feasibility of transferring VLA paradigms to aerial systems but also, through multi-dimensional metrics tailored to aerial tasks, reveal the capabilities and boundaries of current models regarding UAV mobility, manipulator control, and high-level planning. AIR-VLA establishes a standardized testbed and data foundation for future research in general-purpose aerial robotics. The resource of AIR-VLA will be available at https://anonymous.4open.science/r/AIR-VLA-dataset-B5CC/.

Paperid: 5464, https://arxiv.org/pdf/2601.19634.pdf

Abstract:
Vision-Language-Action (VLA) models have demonstrated strong performance in robotic manipulation, yet their closed-loop deployment is hindered by the high latency and compute cost of repeatedly running large vision-language backbones at every timestep. We observe that VLA inference exhibits structured redundancies across temporal, spatial, and depth dimensions, and that most existing efficiency methods ignore action context, despite its central role in embodied tasks. To address this gap, we propose Action-Context-aware Adaptive Computation for VLA models (AC^2-VLA), a unified framework that conditions computation on current visual observations, language instructions, and previous action states. Based on this action-centric context, AC^2-VLA adaptively performs cognition reuse across timesteps, token pruning, and selective execution of model components within a unified mechanism. To train the adaptive policy, we introduce an action-guided self-distillation scheme that preserves the behavior of the dense VLA policy while enabling structured sparsification that transfers across tasks and settings. Extensive experiments on robotic manipulation benchmarks show that AC^2-VLA achieves up to a 1.79\times speedup while reducing FLOPs to 29.4% of the dense baseline, with comparable task success.

Paperid: 5465, https://arxiv.org/pdf/2601.18493.pdf

Abstract:
Timely interpretation of satellite imagery is critical for disaster response, yet existing vision-language benchmarks for remote sensing largely focus on coarse labels and image-level recognition, overlooking the functional understanding and instruction robustness required in real humanitarian workflows. We introduce DisasterInsight, a multimodal benchmark designed to evaluate vision-language models (VLMs) on realistic disaster analysis tasks. DisasterInsight restructures the xBD dataset into approximately 112K building-centered instances and supports instruction-diverse evaluation across multiple tasks, including building-function classification, damage-level and disaster-type classification, counting, and structured report generation aligned with humanitarian assessment guidelines. To establish domain-adapted baselines, we propose DI-Chat, obtained by fine-tuning existing VLM backbones on disaster-specific instruction data using parameter-efficient Low-Rank Adaptation (LoRA). Extensive experiments on state-of-the-art generic and remote-sensing VLMs reveal substantial performance gaps across tasks, particularly in damage understanding and structured report generation. DI-Chat achieves significant improvements on damage-level and disaster-type classification as well as report generation quality, while building-function classification remains challenging for all evaluated models. DisasterInsight provides a unified benchmark for studying grounded multimodal reasoning in disaster imagery.

Paperid: 5466, https://arxiv.org/pdf/2601.16207.pdf

Abstract:
Many Vision-Language-Action (VLA) models flatten image patches into a 1D token sequence, weakening the 2D spatial cues needed for precise manipulation. We introduce IVRA, a lightweight, training-free method that improves spatial understanding by exploiting affinity hints already available in the model's built-in vision encoder, without requiring any external encoder or retraining. IVRA selectively injects these affinity signals into a language-model layer in which instance-level features reside. This inference-time intervention realigns visual-token interactions and better preserves geometric structure while keeping all model parameters fixed. We demonstrate the generality of IVRA by applying it to diverse VLA architectures (LLaRA, OpenVLA, and FLOWER) across simulated benchmarks spanning both 2D and 3D manipulation (VIMA and LIBERO) and on various real-robot tasks. On 2D VIMA, IVRA improves average success by +4.2% over the baseline LLaRA in a low-data regime. On 3D LIBERO, it yields consistent gains over the OpenVLA and FLOWER baselines, including improvements when baseline accuracy is near saturation (96.3% to 97.1%). All code and models will be released publicly. Visualizations are available at: jongwoopark7978.github.io/IVRA

Paperid: 5467, https://arxiv.org/pdf/2601.12443.pdf

Abstract:
The widespread use of Vision Language Models (VLMs, e.g. CLIP) has raised concerns about their vulnerability to sophisticated and imperceptible adversarial attacks. These attacks could compromise model performance and system security in cross-modal tasks. To address this challenge, three main defense paradigms have been proposed: Training-time Defense, Test-time Adaptation Defense, and Training-free Defense. Training-time Defense involves modifying the training process, typically through adversarial fine-tuning to improve the robustness to adversarial examples. While effective, this approach requires substantial computational resources and may not generalize across all adversarial attacks. Test-time Adaptation Defense focuses on adapting the model at inference time by updating its parameters to handle unlabeled adversarial examples, offering flexibility but often at the cost of increased complexity and computational overhead. Training-free Defense avoids modifying the model itself, instead focusing on altering the adversarial inputs or their feature embeddings, which enforces input perturbations to mitigate the impact of attacks without additional training. This survey reviews the latest advancements in adversarial defense strategies for VLMs, highlighting the strengths and limitations of such approaches and discussing ongoing challenges in enhancing the robustness of VLMs.

Paperid: 5468, https://arxiv.org/pdf/2601.08246.pdf

Abstract:
Dexterous grasp synthesis remains a central challenge: the high dimensionality and kinematic diversity of multi-fingered hands prevent direct transfer of algorithms developed for parallel-jaw grippers. Existing approaches typically depend on large, hardware-specific grasp datasets collected in simulation or through costly real-world trials, hindering scalability as new dexterous hand designs emerge. To this end, we propose a data-efficient framework, which is designed to bypass robot grasp data collection by exploiting the rich, object-centric semantic priors latent in pretrained generative diffusion models. Temporally aligned and fine-grained grasp affordances are extracted from raw human video demonstrations and fused with 3D scene geometry from depth images to infer semantically grounded contact targets. A kinematics-aware retargeting module then maps these affordance representations to diverse dexterous hands without per-hand retraining. The resulting system produces stable, functionally appropriate multi-contact grasps that remain reliably successful across common objects and tools, while exhibiting strong generalization across previously unseen object instances within a category, pose variations, and multiple hand embodiments. This work (i) introduces a semantic affordance extraction pipeline leveraging vision-language generative priors for dexterous grasping, (ii) demonstrates cross-hand generalization without constructing hardware-specific grasp datasets, and (iii) establishes that a single depth modality suffices for high-performance grasp synthesis when coupled with foundation-model semantics. Our results highlight a path toward scalable, hardware-agnostic dexterous manipulation driven by human demonstrations and pretrained generative models.

Paperid: 5469, https://arxiv.org/pdf/2601.08024.pdf

Abstract:
Maintaining or improving the performance of Deep Neural Networks (DNNs) through fine-tuning requires labeling newly collected inputs, a process that is often costly and time-consuming. To alleviate this problem, input selection approaches have been developed in recent years to identify small, yet highly informative subsets for labeling. Diversity-based selection is one of the most effective approaches for this purpose. However, they are often computationally intensive and lack scalability for large input sets, limiting their practical applicability. To address this challenge, we introduce Concept-Based Diversity (CBD), a highly efficient metric for image inputs that leverages Vision-Language Models (VLM). Our results show that CBD exhibits a strong correlation with Geometric Diversity (GD), an established diversity metric, while requiring only a fraction of its computation time. Building on this finding, we propose a hybrid input selection approach that combines CBD with Margin, a simple uncertainty metric. We conduct a comprehensive evaluation across a diverse set of DNN models, input sets, selection budgets, and five most effective state-of-the-art selection baselines. The results demonstrate that the CBD-based selection consistently outperforms all baselines at guiding input selection to improve the DNN model. Furthermore, the CBD-based selection approach remains highly efficient, requiring selection times close to those of simple uncertainty-based methods such as Margin, even on larger input sets like ImageNet. These results confirm not only the effectiveness and computational advantage of the CBD-based approach, particularly compared to hybrid baselines, but also its scalability in repetitive and extensive input selection scenarios.

Paperid: 5470, https://arxiv.org/pdf/2601.03044.pdf

Abstract:
Vision-language-action (VLA) models achieve strong generalization through large-scale pre-training, but real-world deployment requires expert-level task proficiency in addition to broad generality. Existing post-training approaches for VLA models are typically offline, single-robot, or task-specific, limiting effective on-policy adaptation and scalable learning from real-world interaction. We introduce a Scalable Online Post-training (SOP) system that enables online, distributed, multi-task post-training of generalist VLA models directly in the physical world. SOP tightly couples execution and learning through a closed-loop architecture in which a fleet of robots continuously streams on-policy experience and human intervention signals to a centralized cloud learner, and asynchronously receives updated policies. This design supports prompt on-policy correction, scales experience collection through parallel deployment, and preserves generality during adaptation. SOP is agnostic to the choice of post-training algorithm; we instantiate it with both interactive imitation learning (HG-DAgger) and reinforcement learning (RECAP). Across a range of real-world manipulation tasks including cloth folding, box assembly, and grocery restocking, we show that SOP substantially improves the performance of large pretrained VLA models while maintaining a single shared policy across tasks. Effective post-training can be achieved within hours of real-world interaction, and performance scales near-linearly with the number of robots in the fleet. These results suggest that tightly coupling online learning with fleet-scale deployment is instrumental to enabling efficient, reliable, and scalable post-training of generalist robot policies in the physical world.

Paperid: 5471, https://arxiv.org/pdf/2512.24985.pdf

Abstract:
Vision Language Models (VLMs) are increasingly adopted as central reasoning modules for embodied agents. Existing benchmarks evaluate their capabilities under ideal, well-lit conditions, yet robust 24/7 operation demands performance under a wide range of visual degradations, including low-light conditions at night or in dark environments--a core necessity that has been largely overlooked. To address this underexplored challenge, we present DarkEQA, an open-source benchmark for evaluating EQA-relevant perceptual primitives under multi-level low-light conditions. DarkEQA isolates the perception bottleneck by evaluating question answering from egocentric observations under controlled degradations, enabling attributable robustness analysis. A key design feature of DarkEQA is its physical fidelity: visual degradations are modeled in linear RAW space, simulating physics-based illumination drop and sensor noise followed by an ISP-inspired rendering pipeline. We demonstrate the utility of DarkEQA by evaluating a wide range of state-of-the-art VLMs and Low-Light Image Enhancement (LLIE) models. Our analysis systematically reveals VLMs' limitations when operating under these challenging visual conditions. Our code and benchmark dataset will be released upon acceptance.

Paperid: 5472, https://arxiv.org/pdf/2512.22275.pdf

Abstract:
Background: The rapid integration of foundation models into clinical practice and public health necessitates a rigorous evaluation of their true clinical reasoning capabilities beyond narrow examination success. Current benchmarks, typically based on medical licensing exams or curated vignettes, fail to capture the integrated, multimodal reasoning essential for real-world patient care. Methods: We developed the Bones and Joints (B&J) Benchmark, a comprehensive evaluation framework comprising 1,245 questions derived from real-world patient cases in orthopedics and sports medicine. This benchmark assesses models across 7 tasks that mirror the clinical reasoning pathway, including knowledge recall, text and image interpretation, diagnosis generation, treatment planning, and rationale provision. We evaluated eleven vision-language models (VLMs) and six large language models (LLMs), comparing their performance against expert-derived ground truth. Results: Our results demonstrate a pronounced performance gap between task types. While state-of-the-art models achieved high accuracy, exceeding 90%, on structured multiple-choice questions, their performance markedly declined on open-ended tasks requiring multimodal integration, with accuracy scarcely reaching 60%. VLMs demonstrated substantial limitations in interpreting medical images and frequently exhibited severe text-driven hallucinations, often ignoring contradictory visual evidence. Notably, models specifically fine-tuned for medical applications showed no consistent advantage over general-purpose counterparts. Conclusions: Current artificial intelligence models are not yet clinically competent for complex, multimodal reasoning. Their safe deployment should currently be limited to supportive, text-based roles. Future advancement in core clinical tasks awaits fundamental breakthroughs in multimodal integration and visual understanding.

Paperid: 5473, https://arxiv.org/pdf/2512.21684.pdf

Abstract:
Modern vision--language models (VLMs) are increasingly used to interpret and generate educational content, yet their semantic outputs remain challenging to verify, reproduce, and audit over time. Inconsistencies across model families, inference settings, and computing environments undermine the reliability of AI-generated instructional material, particularly in high-stakes and quantitative STEM domains. This work introduces SlideChain, a blockchain-backed provenance framework designed to provide verifiable integrity for multimodal semantic extraction at scale. Using the SlideChain Slides Dataset-a curated corpus of 1,117 medical imaging lecture slides from a university course-we extract concepts and relational triples from four state-of-the-art VLMs and construct structured provenance records for every slide. SlideChain anchors cryptographic hashes of these records on a local EVM (Ethereum Virtual Machine)-compatible blockchain, providing tamper-evident auditability and persistent semantic baselines. Through the first systematic analysis of semantic disagreement, cross-model similarity, and lecture-level variability in multimodal educational content, we reveal pronounced cross-model discrepancies, including low concept overlap and near-zero agreement in relational triples on many slides. We further evaluate gas usage, throughput, and scalability under simulated deployment conditions, and demonstrate perfect tamper detection along with deterministic reproducibility across independent extraction runs. Together, these results show that SlideChain provides a practical and scalable step toward trustworthy, verifiable multimodal educational pipelines, supporting long-term auditability, reproducibility, and integrity for AI-assisted instructional systems.

Paperid: 5474, https://arxiv.org/pdf/2512.19562.pdf

Abstract:
Vision-Language-Action (VLA) models empower robots to understand and execute tasks described by natural language instructions. However, a key challenge lies in their ability to generalize beyond the specific environments and conditions they were trained on, which is presently difficult and expensive to evaluate in the real-world. To address this gap, we present REALM, a new simulation environment and benchmark designed to evaluate the generalization capabilities of VLA models, with a specific emphasis on establishing a strong correlation between simulated and real-world performance through high-fidelity visuals and aligned robot control. Our environment offers a suite of 15 perturbation factors, 7 manipulation skills, and more than 3,500 objects. Finally, we establish two task sets that form our benchmark and evaluate the π_{0}, π_{0}-FAST, and GR00T N1.5 VLA models, showing that generalization and robustness remain an open challenge. More broadly, we also show that simulation gives us a valuable proxy for the real-world and allows us to systematically probe for and quantify the weaknesses and failure modes of VLAs. Project page: https://martin-sedlacek.com/realm

Paperid: 5475, https://arxiv.org/pdf/2512.16925.pdf

Abstract:
We introduce V-Agent, a novel multi-agent platform designed for advanced video search and interactive user-system conversations. By fine-tuning a vision-language model (VLM) with a small video preference dataset and enhancing it with a retrieval vector from an image-text retrieval model, we overcome the limitations of traditional text-based retrieval systems in multimodal scenarios. The VLM-based retrieval model independently embeds video frames and audio transcriptions from an automatic speech recognition (ASR) module into a shared multimodal representation space, enabling V-Agent to interpret both visual and spoken content for context-aware video search. This system consists of three agents-a routing agent, a search agent, and a chat agent-that work collaboratively to address user intents by refining search outputs and communicating with users. The search agent utilizes the VLM-based retrieval model together with an additional re-ranking module to further enhance video retrieval quality. Our proposed framework demonstrates state-of-the-art zero-shot performance on the MultiVENT 2.0 benchmark, highlighting its potential for both academic research and real-world applications.

Paperid: 5476, https://arxiv.org/pdf/2512.16446.pdf

Abstract:
Vision-language models (VLMs) show promise in automating reward design in humanoid locomotion, which could eliminate the need for tedious manual engineering. However, current VLM-based methods are essentially "blind", as they lack the environmental perception required to navigate complex terrain. We present E-SDS (Environment-aware See it, Do it, Sorted), a framework that closes this perception gap. E-SDS integrates VLMs with real-time terrain sensor analysis to automatically generate reward functions that facilitate training of robust perceptive locomotion policies, grounded by example videos. Evaluated on a Unitree G1 humanoid across four distinct terrains (simple, gaps, obstacles, stairs), E-SDS uniquely enabled successful stair descent, while policies trained with manually-designed rewards or a non-perceptive automated baseline were unable to complete the task. In all terrains, E-SDS also reduced velocity tracking error by 51.9-82.6%. Our framework reduces the human effort of reward design from days to less than two hours while simultaneously producing more robust and capable locomotion policies.

Paperid: 5477, https://arxiv.org/pdf/2512.15254.pdf

Abstract:
Counting the number of items in a visual scene remains a fundamental yet challenging task in computer vision. Traditional approaches to solving this problem rely on domain-specific counting architectures, which are trained using datasets annotated with a predefined set of object categories. However, recent progress in creating large-scale multimodal vision-language models (VLMs) suggests that these domain-general architectures may offer a flexible alternative for open-set object counting. In this study, we therefore systematically compare the performance of state-of-the-art specialized counting architectures against VLMs on two popular counting datasets, as well as on a novel benchmark specifically created to have a finer-grained control over the visual properties of test images. Our findings show that most VLMs can approximately enumerate the number of items in a visual scene, matching or even surpassing the performance of specialized computer vision architectures. Notably, enumeration accuracy significantly improves when VLMs are prompted to generate intermediate representations (i.e., locations and verbal labels) of each object to be counted. Nevertheless, none of the models can reliably count the number of objects in complex visual scenes, showing that further research is still needed to create AI systems that can reliably deploy counting procedures in realistic environments.

Paperid: 5478, https://arxiv.org/pdf/2512.14944.pdf

Abstract:
Recent reinforcement learning (RL) approaches like outcome-supervised GRPO have advanced chain-of-thought reasoning in Vision Language Models (VLMs), yet key issues linger: (i) reliance on costly and noisy hand-curated annotations or external verifiers; (ii) flat and sparse reward schemes in GRPO; and (iii) logical inconsistency between a chain's reasoning and its final answer. We present Puzzle Curriculum GRPO (PC-GRPO), a supervision-free recipe for RL with Verifiable Rewards (RLVR) that strengthens visual reasoning in VLMs without annotations or external verifiers. PC-GRPO replaces labels with three self-supervised puzzle environments: PatchFit, Rotation (with binary rewards) and Jigsaw (with graded partial credit mitigating reward sparsity). To counter flat rewards and vanishing group-relative advantages, we introduce a difficulty-aware curriculum that dynamically weights samples and peaks at medium difficulty. We further monitor Reasoning-Answer Consistency (RAC) during post-training: mirroring reports for vanilla GRPO in LLMs, RAC typically rises early then degrades; our curriculum delays this decline, and consistency-enforcing reward schemes further boost RAC. RAC correlates with downstream accuracy. Across diverse benchmarks and on Qwen-7B and Qwen-3B backbones, PC-GRPO improves reasoning quality, training stability, and end-task accuracy, offering a practical path to scalable, verifiable, and interpretable RL post-training for VLMs.

Paperid: 5479, https://arxiv.org/pdf/2512.14770.pdf

Abstract:
Vision-language models (VLMs) have demonstrated significant potential in Visual Question Answering (VQA). However, the susceptibility of VLMs to hallucinations can lead to overconfident yet incorrect answers, severely undermining answer reliability. To address this, we propose Dual-Assessment for VLM Reliability (DAVR), a novel framework that integrates Self-Reflection and Cross-Model Verification for comprehensive uncertainty estimation. The DAVR framework features a dual-pathway architecture: one pathway leverages dual selector modules to assess response reliability by fusing VLM latent features with QA embeddings, while the other deploys external reference models for factual cross-checking to mitigate hallucinations. Evaluated in the Reliable VQA Challenge at ICCV-CLVL 2025, DAVR achieves a leading $Φ_{100}$ score of 39.64 and a 100-AUC of 97.22, securing first place and demonstrating its effectiveness in enhancing the trustworthiness of VLM responses.

Paperid: 5480, https://arxiv.org/pdf/2512.14130.pdf

Abstract:
We introduce UIXPOSE, a source-code-agnostic framework that operates on both compiled and open-source apps. This framework applies Intention Behaviour Alignment (IBA) to mobile malware analysis, aligning UI-inferred intent with runtime semantics. Previous work either infers intent statically, e.g., permission-centric, or widget-level or monitors coarse dynamic signals (endpoints, partial resource usage) that miss content and context. UIXPOSE infers an intent vector from each screen using vision-language models and knowledge structures and combines decoded network payloads, heap/memory signals, and resource utilisation traces into a behaviour vector. Their alignment, calculated at runtime, can both detect misbehaviour and highlight exploration of behaviourally rich paths. In three real-world case studies, UIXPOSE reveals covert exfiltration and hidden background activity that evade metadata-only baselines, demonstrating how IBA improves dynamic detection.

Paperid: 5481, https://arxiv.org/pdf/2512.13250.pdf

Abstract:
Vision Language Models (VLMs) excel at visual question answering (VQA) but remain limited to snapshot vision, reasoning from static images. In contrast, embodied agents require ambulatory vision, actively moving to obtain more informative views. We introduce Visually Grounded Active View Selection (VG-AVS), a task that selects the most informative next viewpoint using only the visual information in the current image, without relying on scene memory or external knowledge. To support this task, we construct a synthetic dataset with automatically generated paired query-target views and question-answer prompts. We also propose a framework that fine-tunes pretrained VLMs through supervised fine-tuning (SFT) followed by RL-based policy optimization. Our approach achieves strong question answering performance based on viewpoint selection and generalizes robustly to unseen synthetic and real scenes. Furthermore, incorporating our learned VG-AVS framework into existing scene-exploration-based EQA systems improves downstream question-answering accuracy.

Paperid: 5482, https://arxiv.org/pdf/2512.12622.pdf

Abstract:
Embodied agents face a critical dilemma that end-to-end models lack interpretability and explicit 3D reasoning, while modular systems ignore cross-component interdependencies and synergies. To bridge this gap, we propose the Dynamic 3D Vision-Language-Planning Model (D3D-VLP). Our model introduces two key innovations: 1) A Dynamic 3D Chain-of-Thought (3D CoT) that unifies planning, grounding, navigation, and question answering within a single 3D-VLM and CoT pipeline; 2) A Synergistic Learning from Fragmented Supervision (SLFS) strategy, which uses a masked autoregressive loss to learn from massive and partially-annotated hybrid data. This allows different CoT components to mutually reinforce and implicitly supervise each other. To this end, we construct a large-scale dataset with 10M hybrid samples from 5K real scans and 20K synthetic scenes that are compatible with online learning methods such as RL and DAgger. Our D3D-VLP achieves state-of-the-art results on multiple benchmarks, including Vision-and-Language Navigation (R2R-CE, REVERIE-CE, NavRAG-CE), Object-goal Navigation (HM3D-OVON), and Task-oriented Sequential Grounding and Navigation (SG3D). Real-world mobile manipulation experiments further validate the effectiveness.

Paperid: 5483, https://arxiv.org/pdf/2512.11167.pdf

Abstract:
Reproducibility remains a cornerstone of scientific progress, yet complex multimodal models often lack transparent implementation details and accessible training infrastructure. In this work, we present a detailed reproduction and critical analysis of the Monkey Vision-Language Model (VLM) (Li et al. 2023b) published in CVPR24, a recent approach to high-resolution image understanding via image tiling. The original paper proposed splitting large images into tiles to recover fine-grained visual details while maintaining computational efficiency. Our study replicates this strategy using open checkpoints and reimplements the training pipeline. We confirm the key finding of the original Monkey VLM work, namely that tiling effectively recovers local details. We then extend this work further, by investigating the effect of the inclusion of the global context, which provide practical insights for future high-resolution multimodal modeling. However, we also report deviations in the results, with the magnitude of these effects depending heavily on task type and tile granularity.

Paperid: 5484, https://arxiv.org/pdf/2512.10607.pdf

Abstract:
We propose Track and Caption Any Motion (TCAM), a motion-centric framework for automatic video understanding that discovers and describes motion patterns without user queries. Understanding videos in challenging conditions like occlusion, camouflage, or rapid movement often depends more on motion dynamics than static appearance. TCAM autonomously observes a video, identifies multiple motion activities, and spatially grounds each natural language description to its corresponding trajectory through a motion-field attention mechanism. Our key insight is that motion patterns, when aligned with contrastive vision-language representations, provide powerful semantic signals for recognizing and describing actions. Through unified training that combines global video-text alignment with fine-grained spatial correspondence, TCAM enables query-free discovery of multiple motion expressions via multi-head cross-attention. On the MeViS benchmark, TCAM achieves 58.4% video-to-text retrieval, 64.9 JF for spatial grounding, and discovers 4.8 relevant expressions per video with 84.7% precision, demonstrating strong cross-task generalization.

Paperid: 5485, https://arxiv.org/pdf/2512.10300.pdf

Abstract:
Despite excelling on multimodal benchmarks, vision-language models (VLMs) largely remain a black box. In this paper, we propose a novel interpretability framework to systematically analyze the internal mechanisms of VLMs, focusing on the functional roles of attention heads in multimodal reasoning. To this end, we introduce CogVision, a dataset that decomposes complex multimodal questions into step-by-step subquestions designed to simulate human reasoning through a chain-of-thought paradigm, with each subquestion associated with specific receptive or cognitive functions such as high-level visual reception and inference. Using a probing-based methodology, we identify attention heads that specialize in these functions and characterize them as functional heads. Our analysis across diverse VLM families reveals that these functional heads are universally sparse, vary in number and distribution across functions, and mediate interactions and hierarchical organization. Furthermore, intervention experiments demonstrate their critical role in multimodal reasoning: removing functional heads leads to performance degradation, while emphasizing them enhances accuracy. These findings provide new insights into the cognitive organization of VLMs and suggest promising directions for designing models with more human-aligned perceptual and reasoning abilities.

Paperid: 5486, https://arxiv.org/pdf/2512.09864.pdf

Abstract:
Autonomous driving (AD) systems struggle in long-tail scenarios due to limited world knowledge and weak visual dynamic modeling. Existing vision-language-action (VLA)-based methods cannot leverage unlabeled videos for visual causal learning, while world model-based methods lack reasoning capabilities from large language models. In this paper, we construct multiple specialized datasets providing reasoning and planning annotations for complex scenarios. Then, a unified Understanding-Generation-Planning framework, named UniUGP, is proposed to synergize scene reasoning, future video generation, and trajectory planning through a hybrid expert architecture. By integrating pre-trained VLMs and video generation models, UniUGP leverages visual dynamics and semantic reasoning to enhance planning performance. Taking multi-frame observations and language instructions as input, it produces interpretable chain-of-thought reasoning, physically consistent trajectories, and coherent future videos. We introduce a four-stage training strategy that progressively builds these capabilities across multiple existing AD datasets, along with the proposed specialized datasets. Experiments demonstrate state-of-the-art performance in perception, reasoning, and decision-making, with superior generalization to challenging long-tail situations.

Paperid: 5487, https://arxiv.org/pdf/2512.09215.pdf

Abstract:
3D visual grounding (3DVG) identifies objects in 3D scenes from language descriptions. Existing zero-shot approaches leverage 2D vision-language models (VLMs) by converting 3D spatial information (SI) into forms amenable to VLM processing, typically as composite inputs such as specified view renderings or video sequences with overlaid object markers. However, this VLM + SI paradigm yields entangled visual representations that compel the VLM to process entire cluttered cues, making it hard to exploit spatial semantic relationships effectively. In this work, we propose a new VLM x SI paradigm that externalizes the 3D SI into a form enabling the VLM to incrementally retrieve only what it needs during reasoning. We instantiate this paradigm with a novel View-on-Graph (VoG) method, which organizes the scene into a multi-modal, multi-layer scene graph and allows the VLM to operate as an active agent that selectively accesses necessary cues as it traverses the scene. This design offers two intrinsic advantages: (i) by structuring 3D context into a spatially and semantically coherent scene graph rather than confounding the VLM with densely entangled visual inputs, it lowers the VLM's reasoning difficulty; and (ii) by actively exploring and reasoning over the scene graph, it naturally produces transparent, step-by-step traces for interpretable 3DVG. Extensive experiments show that VoG achieves state-of-the-art zero-shot performance, establishing structured scene exploration as a promising strategy for advancing zero-shot 3DVG.

Paperid: 5488, https://arxiv.org/pdf/2512.08922.pdf

Abstract:
Text-Aware Image Restoration (TAIR) aims to recover high-quality images from low-quality inputs containing degraded textual content. While diffusion models provide strong generative priors for general image restoration, they often produce text hallucinations in text-centric tasks due to the absence of explicit linguistic knowledge. To address this, we propose UniT, a unified text restoration framework that integrates a Diffusion Transformer (DiT), a Vision-Language Model (VLM), and a Text Spotting Module (TSM) in an iterative fashion for high-fidelity text restoration. In UniT, the VLM extracts textual content from degraded images to provide explicit textual guidance. Simultaneously, the TSM, trained on diffusion features, generates intermediate OCR predictions at each denoising step, enabling the VLM to iteratively refine its guidance during the denoising process. Finally, the DiT backbone, leveraging its strong representational power, exploit these cues to recover fine-grained textual content while effectively suppressing text hallucinations. Experiments on the SA-Text and Real-Text benchmarks demonstrate that UniT faithfully reconstructs degraded text, substantially reduces hallucinations, and achieves state-of-the-art end-to-end F1-score performance in TAIR task.

Paperid: 5489, https://arxiv.org/pdf/2512.07177.pdf

Abstract:
Robots operating in everyday environments must often decide when and whether to engage with people, yet such decisions often hinge on subtle nonverbal cues that unfold over time and are difficult to model explicitly. Drawing on a five-day Wizard-of-Oz deployment of a mobile service robot in a university cafe, we analyze how people signal interaction readiness through nonverbal behaviors and how expert wizards use these cues to guide engagement. Motivated by these observations, we propose a two-stage pipeline in which lightweight perceptual detectors (gaze shifts and proxemics) are used to selectively trigger heavier video-based vision-language model (VLM) queries at socially meaningful moments. We evaluate this pipeline on replayed field interactions and compare two prompting strategies. Our findings suggest that selectively using VLMs as proxies for social reasoning enables socially responsive robot behavior, allowing robots to act appropriately by attending to the cues people naturally provide in real-world interactions.

Paperid: 5490, https://arxiv.org/pdf/2512.06376.pdf

Abstract:
Recent text-to-video models have enabled the generation of high-resolution driving scenes from natural language prompts. These AI-generated driving videos (AIGVs) offer a low-cost, scalable alternative to real or simulator data for autonomous driving (AD). But a key question remains: can such videos reliably support training and evaluation of AD models? We present a diagnostic framework that systematically studies this question. First, we introduce a taxonomy of frequent AIGV failure modes, including visual artifacts, physically implausible motion, and violations of traffic semantics, and demonstrate their negative impact on object detection, tracking, and instance segmentation. To support this analysis, we build ADGV-Bench, a driving-focused benchmark with human quality annotations and dense labels for multiple perception tasks. We then propose ADGVE, a driving-aware evaluator that combines static semantics, temporal cues, lane obedience signals, and Vision-Language Model(VLM)-guided reasoning into a single quality score for each clip. Experiments show that blindly adding raw AIGVs can degrade perception performance, while filtering them with ADGVE consistently improves both general video quality assessment metrics and downstream AD models, and turns AIGVs into a beneficial complement to real-world data. Our study highlights both the risks and the promise of AIGVs, and provides practical tools for safely leveraging large-scale video generation in future AD pipelines.

Paperid: 5491, https://arxiv.org/pdf/2512.05546.pdf

Abstract:
Large Vision-Language Models (VLMs) often exhibit text inertia, where attention drifts from visual evidence toward linguistic priors, resulting in object hallucinations. Existing decoding strategies intervene only at the output logits and thus cannot correct internal reasoning drift, while recent internal-control methods based on heuristic head suppression or global steering vectors lack principled grounding. We introduce Conscious Gaze (CG-VLM), a training-free, inference-time framework that converts game-theoretic interpretability into actionable decoding control. A Cognitive Demand Sensor built on Harsanyi interactions estimates instantaneous vision-text synergy and identifies moments when visual grounding is necessary. Conditioned on this signal, a Focused Consensus Induction module selectively reorients mid-layer attention toward visual tokens before collapse into text priors. CG-VLM achieves state-of-the-art results on POPE and CHAIR across InstructBLIP, LLaVA, Qwen-VL, and mPLUG, while preserving general capabilities, demonstrating that token-level sensing enables precise, context-aware intervention without compromising foundational knowledge.

Paperid: 5492, https://arxiv.org/pdf/2512.04441.pdf

Abstract:
End-to-End autonomous driving (E2E-AD) has emerged as a new paradigm, where trajectory planning plays a crucial role. Existing studies mainly follow two directions: trajectory generation oriented, which focuses on producing high-quality trajectories with simple decision mechanisms, and trajectory selection oriented, which performs multi-dimensional evaluation to select the best trajectory yet lacks sufficient generative capability. In this work, we propose MindDrive, a harmonized framework that integrates high-quality trajectory generation with comprehensive decision reasoning. It establishes a structured reasoning paradigm of "context simulation - candidate generation - multi-objective trade-off". In particular, the proposed Future-aware Trajectory Generator (FaTG), based on a World Action Model (WaM), performs ego-conditioned "what-if" simulations to predict potential future scenes and generate foresighted trajectory candidates. Building upon this, the VLM-oriented Evaluator (VLoE) leverages the reasoning capability of a large vision-language model to conduct multi-objective evaluations across safety, comfort, and efficiency dimensions, leading to reasoned and human-aligned decision making. Extensive experiments on the NAVSIM-v1 and NAVSIM-v2 benchmarks demonstrate that MindDrive achieves state-of-the-art performance across multi-dimensional driving metrics, significantly enhancing safety, compliance, and generalization. This work provides a promising path toward interpretable and cognitively guided autonomous driving.

Paperid: 5493, https://arxiv.org/pdf/2512.03463.pdf

Abstract:
Recent large vision-language models (LVLMs) have been applied to diverse VQA tasks. However, achieving practical performance typically requires task-specific fine-tuning with large numbers of image-text pairs, which are costly to collect. In this work, we study text-centric training, a setting where only textual descriptions are available and no real images are provided, as a paradigm for low-cost data scaling. Unlike images, whose collection is often restricted by privacy constraints and scarcity in niche domains, text is widely available. Moreover, text is easily editable, enabling automatic diversification and expansion with LLMs at minimal human effort. While this offers clear advantages over image collection in terms of scalability and cost, training on raw text without images still yields limited gains on VQA tasks because of the image-text modality gap. To address this issue, we propose a Text-Printed Image (TPI), which generates synthetic images by directly rendering the given textual description on a plain white canvas. This simple rendering projects text into the image modality and can be integrated into arbitrary existing LVLM training pipelines at low cost. Moreover, TPI preserves the semantics of the text, whereas text-to-image models often fail to do. Across four models and seven benchmarks, our systematic experiments show that TPI enables more effective text-centric training than synthetic images generated by a diffusion model. We further explore TPI as a low-cost data-augmentation strategy and demonstrate its practical utility. Overall, our findings highlight the significant potential of text-centric training and, more broadly, chart a path toward fully automated data generation for LVLMs.

Paperid: 5494, https://arxiv.org/pdf/2512.03284.pdf

Abstract:
Spatial reasoning in large-scale 3D environments remains challenging for current vision-language models, which are typically constrained to room-scale scenarios. We introduce H$^2$U3D (Holistic House Understanding in 3D), a 3D visual question answering dataset designed for house-scale scene understanding. H$^2$U3D features multi-floor environments spanning up to three floors and 10-20 rooms, covering more than 300 m$^2$. Through an automated annotation pipeline, it constructs hierarchical coarse-to-fine visual representations and generates diverse question-answer pairs with chain-of-thought annotations. We further propose SpatialReasoner, an active perception framework that autonomously invokes spatial tools to explore 3D scenes based on textual queries. SpatialReasoner is trained through a two-stage strategy: a supervised cold start followed by reinforcement learning with an adaptive exploration reward that promotes efficient exploration while discouraging redundant operations. Extensive experiments demonstrate that SpatialReasoner achieves state-of-the-art performance on H$^2$U3D, outperforming strong baselines including GPT-4o and Gemini-2.5-Pro. Notably, our method attains superior results while using only 3-4 images in total on average, compared to baselines requiring 16+ images, highlighting the effectiveness of our coarse-to-fine active exploration paradigm.

Paperid: 5495, https://arxiv.org/pdf/2512.02844.pdf

Abstract:
The safe deployment of autonomous driving systems (ADSs) relies on comprehensive testing and evaluation. However, safety-critical scenarios that can effectively expose system vulnerabilities are extremely sparse in the real world. Existing scenario generation methods face challenges in efficiently constructing long-tail scenarios that ensure fidelity, criticality, and interactivity, while particularly lacking real-time dynamic response capabilities to the vehicle under test (VUT). To address these challenges, this paper proposes a safety-critical testing scenario generation framework that integrates the high-level semantic understanding capabilities of Vision Language Models (VLMs) with the fine-grained generation capabilities of adaptive guided diffusion models. The framework establishes a three-layer hierarchical architecture comprising a strategic layer for VLM-directed scenario generation objective determination, a tactical layer for guidance function formulation, and an operational layer for guided diffusion execution. We first establish a high-quality fundamental diffusion model that learns the data distribution of real driving scenarios. Next, we design an adaptive guided diffusion method that enables real-time, precise control of background vehicles (BVs) in closed-loop simulation. The VLM is then incorporated to autonomously generate scenario generation objectives and guidance functions through deep scenario understanding and risk reasoning, ultimately guiding the diffusion model to achieve VLM-directed scenario generation. Experimental results demonstrate that the proposed method can efficiently generate realistic, diverse, and highly interactive safety-critical testing scenarios. Furthermore, case studies validate the adaptability and VLM-directed generation performance of the proposed method.

Paperid: 5496, https://arxiv.org/pdf/2512.02421.pdf

Abstract:
Fine-tuning large pretrained vision-language models (VLMs) has emerged as a prevalent paradigm for downstream adaptation, yet it faces a critical trade-off between domain specificity and domain generalization (DG) ability. Current methods typically fine-tune a universal model on the entire dataset, which potentially compromises the ability to generalize to unseen domains. To fill this gap, we provide a theoretical understanding of the generalization ability for VLM fine-tuning, which reveals that training multiple parameter-efficient expert models on partitioned source domains leads to better generalization than fine-tuning a universal model. Inspired by this finding, we propose a two-step domain-expert-Guided DG (GuiDG) framework. GuiDG first employs prompt tuning to obtain source domain experts, then introduces a Cross-Modal Attention module to guide the fine-tuning of the vision encoder via adaptive expert integration. To better evaluate few-shot DG, we construct ImageNet-DG from ImageNet and its variants. Extensive experiments on standard DG benchmarks and ImageNet-DG demonstrate that GuiDG improves upon state-of-the-art fine-tuning methods while maintaining efficiency.

Paperid: 5497, https://arxiv.org/pdf/2512.01830.pdf

Abstract:
Recently, two-stage fine-tuning strategies, e.g., acquiring essential driving knowledge through supervised fine-tuning (SFT) and further enhancing decision-making and planning via reinforcement fine-tuning (RFT), have shown strong potential in advancing the knowledge-driven autonomous driving (AD) paradigm. However, the learning nature of SFT still limits the generalization of reasoning, thereby constraining the full potential of driving performance. Meanwhile, current RFT approaches are primarily applied to downstream tasks, since scene understanding is an open-ended problem where corresponding rewards are difficult to quantify. To address these limitations, we propose OpenREAD, an OPEN-ended REasoning reinforced vision-language model (VLM)-based autonomous driving (AD) framework that enables end-to-end RFT across the full spectrum from high-level reasoning to low-level trajectory planning. Specifically, we begin by constructing large-scale Chain-of-Thought (CoT) annotations on open-source driving-related knowledge datasets, and employ the powerful Qwen3 large language model (LLM) as the critic in RFT to quantify reasoning quality for open-ended questions during reward modeling. Extensive experiments confirm that joint end-to-end RFT yields substantial improvements in both upstream and downstream tasks, enabling OpenREAD to achieve state-of-the-art performance on reasoning and planning benchmarks.

Paperid: 5498, https://arxiv.org/pdf/2511.22777.pdf

Abstract:
Learning robust visuomotor policies for robotic manipulation remains a challenge in real-world settings, where visual distractors can significantly degrade performance and safety. In this work, we propose an effective and scalable framework, Naturalistic Inpainting for Context Enhancement (NICE). Our method minimizes out-of-distribution (OOD) gap in imitation learning by increasing visual diversity through construction of new experiences using existing demonstrations. By utilizing image generative frameworks and large language models, NICE performs three editing operations, object replacement, restyling, and removal of distracting (non-target) objects. These changes preserve spatial relationships without obstructing target objects and maintain action-label consistency. Unlike previous approaches, NICE requires no additional robot data collection, simulator access, or custom model training, making it readily applicable to existing robotic datasets. Using real-world scenes, we showcase the capability of our framework in producing photo-realistic scene enhancement. For downstream tasks, we use NICE data to finetune a vision-language model (VLM) for spatial affordance prediction and a vision-language-action (VLA) policy for object manipulation. Our evaluations show that NICE successfully minimizes OOD gaps, resulting in over 20% improvement in accuracy for affordance prediction in highly cluttered scenes. For manipulation tasks, success rate increases on average by 11% when testing in environments populated with distractors in different quantities. Furthermore, we show that our method improves visual robustness, lowering target confusion by 6%, and enhances safety by reducing collision rate by 7%.

Paperid: 5499, https://arxiv.org/pdf/2511.22253.pdf

Abstract:
Image-Guided Retrieval with Optional Text (IGROT) is a general retrieval setting where a query consists of an anchor image, with or without accompanying text, aiming to retrieve semantically relevant target images. This formulation unifies two major tasks: Composed Image Retrieval (CIR) and Sketch-Based Image Retrieval (SBIR). In this work, we address IGROT under low-data supervision by introducing UNION, a lightweight and generalisable target representation that fuses the image embedding with a null-text prompt. Unlike traditional approaches that rely on fixed target features, UNION enhances semantic alignment with multimodal queries while requiring no architectural modifications to pretrained vision-language models. With only 5,000 training samples - from LlavaSCo for CIR and Training-Sketchy for SBIR - our method achieves competitive results across benchmarks, including CIRCO mAP@50 of 38.5 and Sketchy mAP@200 of 82.7, surpassing many heavily supervised baselines. This demonstrates the robustness and efficiency of UNION in bridging vision and language across diverse query types.

Paperid: 5500, https://arxiv.org/pdf/2511.22018.pdf

Abstract:
Accurate medical diagnosis often involves progressive visual focusing and iterative reasoning, characteristics commonly observed in clinical workflows. While recent vision-language models demonstrate promising chain-of-thought (CoT) reasoning capabilities via reinforcement learning with verifiable rewards (RLVR), their purely on-policy learning paradigm tends to reinforce superficially coherent but clinically inaccurate reasoning paths. We propose MedEyes, a novel reinforcement learning framework that dynamically models clinician-style diagnostic reasoning by progressively attending to and interpreting relevant medical image regions. By incorporating off-policy expert guidance, MedEyes converts expert visual search trajectories into structured external behavioral signals, guiding the model toward clinically aligned visual reasoning. We design the Gaze-guided Reasoning Navigator (GRN) to emulate the diagnostic process through a dual-mode exploration strategy, scanning for systematic abnormality localization and drilling for detailed regional analysis. To balance expert imitation and autonomous discovery, we introduce the Confidence Value Sampler (CVS), which employs nucleus sampling and adaptive termination to create diverse yet credible exploration paths. Finally, the dual-stream GRPO optimization framework decouples on-policy and off-policy learning signals, mitigating reward assimilation and entropy collapse. Experiments demonstrate that MedEyes achieves an average performance improvement of +8.5\% across multiple medical VQA benchmarks, validating MedEyes's potential in building interpretable medical AI systems.

Paperid: 5501, https://arxiv.org/pdf/2511.20732.pdf

Abstract:
Medical AI systems face catastrophic forgetting when deployed in clinical settings, where models must learn new imaging protocols while retaining prior diagnostic capabilities. This challenge is particularly acute for medical vision-language models that must preserve complex cross-modal alignments between medical images and clinical terminology across diverse imaging modalities. We introduce Prompt- Aware Adaptive Elastic Weight Consolidation (PA-EWC), a novel continual learning approach that addresses catastrophic forgetting through prompt-guided parameter specialization. Our method systematically categorizes model parameters based on their functional roles in processing visual-descriptive, spatial-guided, and medical-semantic information, enabling targeted protection of critical knowledge while allowing adaptation to new clinical requirements. PA-EWC incorporates adaptive Fisher Information computation with gradient stability analysis and develops weighted complexity metrics based on medical terminology density. We evaluate our approach across five medical imaging datasets (Kvasir-SEG, ISIC 2018, CheXlocalize, BUSI, CAMUS) representing diverse modalities including endoscopy, dermoscopy, radiography, and ultrasound. Experimental results demonstrate that PA-EWC reduces catastrophic forgetting by up to 17.58% compared to baseline methods, with performance improvements of 4.30% on chest X-ray pathology localization and 6.06% on polyp segmentation.

Paperid: 5502, https://arxiv.org/pdf/2511.18200.pdf

Abstract:
Modern vision-language models (VLMs) are expected to have abilities of spatial reasoning with diverse scene complexities, but evaluating such abilities is difficult due to the lack of benchmarks that are not only diverse and scalable but also fully customizable. Existing benchmarks offer limited customizability over the scene complexity and are incapable of isolating and analyzing specific VLM failure modes under distinct spatial conditions. To address this gap, instead of individually presenting benchmarks for different scene complexities, in this paper we present InfiniBench, a fully automated, customizable and user-friendly benchmark generator that can synthesize a theoretically infinite variety of 3D scenes with parameterized control on scene complexity. InfiniBench uniquely translates scene descriptions in natural language into photo-realistic videos with complex and physically plausible 3D layouts. This is achieved through three key innovations: 1) a LLM-based agentic framework that iteratively refines procedural scene constraints from scene descriptions; 2) a flexible cluster-based layout optimizer that generates dense and cluttered scenes previously intractable for procedural methods; and 3) a task-aware camera trajectory optimization method that renders scenes into videos with full object coverage as VLM input. Experiments demonstrate that InfiniBench outperforms state-of-the-art procedural and LLM-based 3D generation methods in prompt fidelity and physical plausibility, especially in high-complexity scenarios. We further showcased the usefulness of InfiniBench, by generating benchmarks for representative spatial reasoning tasks including measurement, perspective-taking and spatiotemporal tracking.

Paperid: 5503, https://arxiv.org/pdf/2511.18082.pdf

Abstract:
Recent Vision-Language-Action (VLA) models have shown impressive flexibility and generalization, yet their deployment in robotic manipulation remains limited by heavy computational overhead and inference latency. In this work, we present ActDistill, a general action-guided self-derived distillation framework that transfers the action prediction capability of any existing VLA model to a lightweight counterpart. Unlike previous efficiency strategies that primarily emphasize vision-language correlations, ActDistill leverages action priors to guide knowledge transfer and model compression, achieving action-oriented efficiency for VLA models. Specifically, we employ a well-trained VLA model as the teacher and introduce a graph-structured encapsulation strategy to explicitly model the hierarchical evolution of action prediction. The student model, derived from the graph-encapsulated teacher, is further equipped with a dynamic router that adaptively selects computation paths based on action prediction demands, guided by hierarchical graph-informed supervision to ensure smooth and efficient evolution. During inference, graph-related auxiliary components are removed, allowing the student to execute only dynamically routed layers and predict high-precision actions with minimal computation and latency. Experiments on embodied benchmarks demonstrate that ActDistill achieves comparable or superior performance to full-scale VLA models while reducing computation by over 50% with up to 1.67 times speedup, thereby establishing a general paradigm toward efficient embodied intelligence.

Paperid: 5504, https://arxiv.org/pdf/2511.15722.pdf

Abstract:
Spatial reasoning, which requires ability to perceive and manipulate spatial relationships in the 3D world, is a fundamental aspect of human intelligence, yet remains a persistent challenge for Multimodal large language models (MLLMs). While existing surveys often categorize recent progress based on input modality (e.g., text, image, video, or 3D), we argue that spatial ability is not solely determined by the input format. Instead, our survey introduces a taxonomy that organizes spatial intelligence from cognitive aspect and divides tasks in terms of reasoning complexity, linking them to several cognitive functions. We map existing benchmarks across text only, vision language, and embodied settings onto this taxonomy, and review evaluation metrics and methodologies for assessing spatial reasoning ability. This cognitive perspective enables more principled cross-task comparisons and reveals critical gaps between current model capabilities and human-like reasoning. In addition, we analyze methods for improving spatial ability, spanning both training-based and reasoning-based approaches. This dual perspective analysis clarifies their respective strengths, uncovers complementary mechanisms. By surveying tasks, benchmarks, and recent advances, we aim to provide new researchers with a comprehensive understanding of the field and actionable directions for future research.

Paperid: 5505, https://arxiv.org/pdf/2511.15279.pdf

Abstract:
In embodied AI perception systems, visual perception should be active: the goal is not to passively process static images, but to actively acquire more informative data within pixel and spatial budget constraints. Existing vision models and fixed RGB-D camera systems fundamentally fail to reconcile wide-area coverage with fine-grained detail acquisition, severely limiting their efficacy in open-world robotic applications. To address this issue, we propose EyeVLA, a robotic eyeball for active visual perception that can take proactive actions based on instructions, enabling clear observation of fine-grained target objects and detailed information across a wide spatial extent. EyeVLA discretizes action behaviors into action tokens and integrates them with vision-language models (VLMs) that possess strong open-world understanding capabilities, enabling joint modeling of vision, language, and actions within a single autoregressive sequence. By using the 2D bounding box coordinates to guide the reasoning chain and applying reinforcement learning to refine the viewpoint selection policy, we transfer the open-world scene understanding capability of the VLM to a vision language action (VLA) policy using only minimal real-world data. Experiments show that our system efficiently performs instructed scenes in real-world environments and actively acquires more accurate visual information through instruction-driven actions of rotation and zoom, thereby achieving strong environmental perception capabilities. EyeVLA introduces a novel robotic vision system that leverages detailed and spatially rich, large-scale embodied data, and actively acquires highly informative visual observations for downstream embodied tasks.

Paperid: 5506, https://arxiv.org/pdf/2511.15186.pdf

Abstract:
The applicability of current lesion segmentation models for chest X-rays (CXRs) has been limited both by a small number of target labels and the reliance on long, detailed expert-level text inputs, creating a barrier to practical use. To address these limitations, we introduce a new paradigm: instruction-guided lesion segmentation (ILS), which is designed to segment diverse lesion types based on simple, user-friendly instructions. Under this paradigm, we construct MIMIC-ILS, the first large-scale instruction-answer dataset for CXR lesion segmentation, using our fully automated multimodal pipeline that generates annotations from chest X-ray images and their corresponding reports. MIMIC-ILS contains 1.1M instruction-answer pairs derived from 192K images and 91K unique segmentation masks, covering seven major lesion types. To empirically demonstrate its utility, we introduce ROSALIA, a vision-language model fine-tuned on MIMIC-ILS. ROSALIA can segment diverse lesions and provide textual explanations in response to user instructions. The model achieves high segmentation and textual accuracy in our newly proposed task, highlighting the effectiveness of our pipeline and the value of MIMIC-ILS as a foundational resource for pixel-level CXR lesion grounding.

Paperid: 5507, https://arxiv.org/pdf/2511.14368.pdf

Abstract:
While Large Vision Language Models (LVLMs) are increasingly deployed in real-world applications, their ability to interpret abstract visual inputs remains limited. Specifically, they struggle to comprehend hand-drawn sketches, a modality that offers an intuitive means of expressing concepts that are difficult to describe textually. We identify the primary bottleneck as the absence of a large-scale dataset that jointly models sketches, photorealistic images, and corresponding natural language instructions. To address this, we present two key contributions: (1) a new, large-scale dataset of image-sketch-instruction triplets designed to facilitate both pretraining and instruction tuning, and (2) O3SLM, an LVLM trained on this dataset. Comprehensive evaluations on multiple sketch-based tasks: (a) object localization, (b) counting, (c) image retrieval i.e., (SBIR and fine-grained SBIR), and (d) visual question answering (VQA); while incorporating the three existing sketch datasets, namely QuickDraw!, Sketchy, and Tu Berlin, along with our generated SketchVCL dataset, show that O3SLM achieves state-of-the-art performance, substantially outperforming existing LVLMs in sketch comprehension and reasoning.

Paperid: 5508, https://arxiv.org/pdf/2511.13889.pdf

Abstract:
Digital hematopathology requires cell-level analysis across diverse disease categories, including malignant disorders (e.g., leukemia), infectious conditions (e.g., malaria), and non-malignant red blood cell disorders (e.g., sickle cell disease). Whether single-task, vision-language, WSI-optimized, or single-cell hematology models, these approaches share a key limitation, they cannot provide unified, multi-task, multi-modal reasoning across the complexities of digital hematopathology. To overcome these limitations, we propose Uni-Hema, a multi-task, unified model for digital hematopathology integrating detection, classification, segmentation, morphology prediction, and reasoning across multiple diseases. Uni-Hema leverages 46 publicly available datasets, encompassing over 700K images and 21K question-answer pairs, and is built upon Hema-Former, a multimodal module that bridges visual and textual representations at the hierarchy level for the different tasks (detection, classification, segmentation, morphology, mask language modeling and visual question answer) at different granularity. Extensive experiments demonstrate that Uni-Hema achieves comparable or superior performance to train on a single-task and single dataset models, across diverse hematological tasks, while providing interpretable, morphologically relevant insights at the single-cell level. Our framework establishes a new standard for multi-task and multi-modal digital hematopathology. The code will be made publicly available.

Paperid: 5509, https://arxiv.org/pdf/2511.12077.pdf

Abstract:
Emotion understanding is critical for making Large Language Models (LLMs) more general, reliable, and aligned with humans. Art conveys emotion through the joint design of visual and auditory elements, yet most prior work is human-centered or single-modality, overlooking the emotion intentionally expressed by the artwork. Meanwhile, current Audio-Visual Language Models (AVLMs) typically require large-scale audio pretraining to endow Visual Language Models (VLMs) with hearing, which limits scalability. We present Vision Anchored Audio-Visual Emotion LLM (VAEmotionLLM), a two-stage framework that teaches a VLM to hear by seeing with limited audio pretraining and to understand emotion across modalities. In Stage 1, Vision-Guided Audio Alignment (VG-Align) distills the frozen visual pathway into a new audio pathway by aligning next-token distributions of the shared LLM on synchronized audio-video clips, enabling hearing without a large audio dataset. In Stage 2, a lightweight Cross-Modal Emotion Adapter (EmoAdapter), composed of the Emotion Enhancer and the Emotion Supervisor, injects emotion-sensitive residuals and applies emotion supervision to enhance cross-modal emotion understanding. We also construct ArtEmoBenchmark, an art-centric emotion benchmark that evaluates content and emotion understanding under audio-only, visual-only, and audio-visual inputs. VAEmotionLLM achieves state-of-the-art results on ArtEmoBenchmark, outperforming audio-only, visual-only, and audio-visual baselines. Ablations show that the proposed components are complementary.

Paperid: 5510, https://arxiv.org/pdf/2511.11177.pdf

Abstract:
Recent advances in multimodal large language models (MLLMs) have enabled impressive progress in vision-language understanding, yet their high computational cost limits deployment in resource-constrained scenarios such as robotic manipulation, personal assistants, and smart cameras. Most existing methods rely on Transformer-based cross-attention, whose quadratic complexity hinders efficiency. Moreover, small vision-language models often struggle to precisely capture fine-grained, task-relevant visual regions, leading to degraded performance on fine-grained reasoning tasks that limit their effectiveness in the real world. To address these issues, we introduce Viper-F1, a Hybrid State-Space Vision-Language Model that replaces attention with efficient Liquid State-Space Dynamics. To further enhance visual grounding, we propose a Token-Grid Correlation Module, which computes lightweight correlations between text tokens and image patches and modulates the state-space dynamics via FiLM conditioning. This enables the model to selectively emphasize visual regions relevant to the textual prompt while maintaining linear-time inference. Experimental results across multiple benchmarks demonstrate that Viper-F1 achieves accurate, fine-grained understanding with significantly improved efficiency.

Paperid: 5511, https://arxiv.org/pdf/2511.07983.pdf

Abstract:
Generating accurate and clinically meaningful radiology reports from chest X-ray images remains a significant challenge in medical AI. While recent vision-language models achieve strong results in general radiology report generation, they often fail to adequately describe rare but clinically important pathologies like fractures. This work addresses this gap by developing specialized models for fracture pathology detection and description. We train fracture-specific vision-language models with encoders from MAIRA-2 and CheXagent, demonstrating significant improvements over general-purpose models in generating accurate fracture descriptions. Analysis of model outputs by fracture type, location, and age reveals distinct strengths and limitations of current vision-language model architectures. We publicly release our best-performing fracture-reporting model, facilitating future research in accurate reporting of rare pathologies.

Paperid: 5512, https://arxiv.org/pdf/2511.05491.pdf

Abstract:
Capturing spatial relationships from visual inputs is a cornerstone of human-like general intelligence. Several previous studies have tried to enhance the spatial awareness of Vision-Language Models (VLMs) by adding extra expert encoders, which brings extra overhead and usually harms general capabilities. To enhance the spatial ability in general architectures, we introduce Visual Spatial Tuning (VST), a comprehensive framework to cultivate VLMs with human-like visuospatial abilities, from spatial perception to reasoning. We first attempt to enhance spatial perception in VLMs by constructing a large-scale dataset termed VST-P, which comprises 4.1 million samples spanning 19 skills across single views, multiple images, and videos. Then, we present VST-R, a curated dataset with 135K samples that instruct models to reason in space. In particular, we adopt a progressive training pipeline: supervised fine-tuning to build foundational spatial knowledge, followed by reinforcement learning to further improve spatial reasoning abilities. Without the side-effect to general capabilities, the proposed VST consistently achieves state-of-the-art results on several spatial benchmarks, including $34.8\%$ on MMSI-Bench and $61.2\%$ on VSIBench. It turns out that the Vision-Language-Action models can be significantly enhanced with the proposed spatial tuning paradigm, paving the way for more physically grounded AI.

Paperid: 5513, https://arxiv.org/pdf/2510.27680.pdf

Abstract:
Recent advances in vision-language models (VLMs) have enabled impressive multimodal reasoning, yet most medical applications remain limited to 2D imaging. In this work, we extend VLMs to 3D positron emission tomography and computed tomography (PET/CT), a domain characterized by large volumetric data, small and dispersed lesions, and lengthy radiology reports. We introduce a large-scale dataset comprising over 11,000 lesion-level descriptions paired with 3D segmentations from more than 5,000 PET/CT exams, extracted via a hybrid rule-based and large language model (LLM) pipeline. Building upon this dataset, we propose PETAR-4B, a 3D mask-aware vision-language model that integrates PET, CT, and lesion contours for spatially grounded report generation. PETAR bridges global contextual reasoning with fine-grained lesion awareness, producing clinically coherent and localized findings. Comprehensive automated and human evaluations demonstrate that PETAR substantially improves PET/CT report generation quality, advancing 3D medical vision-language understanding.

Paperid: 5514, https://arxiv.org/pdf/2510.26339.pdf

Abstract:
Image super-resolution(SR) is fundamental to many vision system-from surveillance and autonomy to document analysis and retail analytics-because recovering high-frequency details, especially scene-text, enables reliable downstream perception. Scene-text, i.e., text embedded in natural images such as signs, product labels, and storefronts, often carries the most actionable information; when characters are blurred or hallucinated, optical character recognition(OCR) and subsequent decisions fail even if the rest of the image appears sharp. Yet previous SR research has often been tuned to distortion (PSNR/SSIM) or learned perceptual metrics (LIPIS, MANIQA, CLIP-IQA, MUSIQ) that are largely insensitive to character-level errors. Furthermore, studies that do address text SR often focus on simplified benchmarks with isolated characters, overlooking the challenges of text within complex natural scenes. As a result, scene-text is effectively treated as generic texture. For SR to be effective in practical deployments, it is therefore essential to explicitly optimize for both text legibility and perceptual quality. We present GLYPH-SR, a vision-language-guided diffusion framework that aims to achieve both objectives jointly. GLYPH-SR utilizes a Text-SR Fusion ControlNet(TS-ControlNet) guided by OCR data, and a ping-pong scheduler that alternates between text- and scene-centric guidance. To enable targeted text restoration, we train these components on a synthetic corpus while keeping the main SR branch frozen. Across SVT, SCUT-CTW1500, and CUTE80 at x4, and x8, GLYPH-SR improves OCR F1 by up to +15.18 percentage points over diffusion/GAN baseline (SVT x8, OpenOCR) while maintaining competitive MANIQA, CLIP-IQA, and MUSIQ. GLYPH-SR is designed to satisfy both objectives simultaneously-high readability and high visual realism-delivering SR that looks right and reds right.

Paperid: 5515, https://arxiv.org/pdf/2510.26302.pdf

Abstract:
Contrastive Language-Image Pre-training (CLIP) delivers strong cross modal generalization by aligning images and texts in a shared embedding space, yet it persistently fails at compositional reasoning over objects, attributes, and relations often behaving like a bag-of-words matcher. Prior causal accounts typically model text as a single vector, obscuring token-level structure and leaving core phenomena-such as prompt sensitivity and failures on hard negatives unexplained. We address this gap with a token-aware causal representation learning (CRL) framework grounded in a sequential, language-token SCM. Our theory extends block identifiability to tokenized text, proving that CLIP's contrastive objective can recover the modal-invariant latent variable under both sentence-level and token-level SCMs. Crucially, token granularity yields the first principled explanation of CLIP's compositional brittleness: composition nonidentifiability. We show the existence of pseudo-optimal text encoders that achieve perfect modal-invariant alignment yet are provably insensitive to SWAP, REPLACE, and ADD operations over atomic concepts, thereby failing to distinguish correct captions from hard negatives despite optimizing the same training objective as true-optimal encoders. The analysis further links language-side nonidentifiability to visual-side failures via the modality gap and shows how iterated composition operators compound hardness, motivating improved negative mining strategies.

Paperid: 5516, https://arxiv.org/pdf/2510.25067.pdf

Abstract:
Recently, the advances in vision-language models, including contrastive pretraining and instruction tuning, have greatly pushed the frontier of multimodal AI. However, owing to the large-scale and hence expensive pretraining, the efficiency concern has discouraged researchers from attempting to pretrain a vision language model from scratch. In this work, we propose Dynamic patch Reduction via Interpretable Pooling (DRIP), which adapts to the input images and dynamically merges tokens in the deeper layers of a visual encoder. Our results on both ImageNet training from scratch and CLIP contrastive pretraining demonstrate a significant GFLOP reduction while maintaining comparable classification/zero-shot performance. To further validate our proposed method, we conduct continual pretraining on a large biology dataset, extending its impact into scientific domains.

Paperid: 5517, https://arxiv.org/pdf/2510.24161.pdf

Abstract:
Multimodal large language models (MLLMs) have advanced vision-language reasoning and are increasingly deployed in embodied agents. However, significant limitations remain: MLLMs generalize poorly across digital-physical spaces and embodiments; vision-language-action models (VLAs) produce low-level actions yet lack robust high-level embodied reasoning; and most embodied large language models (ELLMs) are constrained to digital-space with poor generalization to the physical world. Thus, unified models that operate seamlessly across digital and physical spaces while generalizing across embodiments and tasks remain absent. We introduce the \textbf{Boundless Large Model (BLM$_1$)}, a multimodal spatial foundation model that preserves instruction following and reasoning, incorporates embodied knowledge, and supports robust cross-embodiment control. BLM$_1$ integrates three key capabilities -- \textit{cross-space transfer, cross-task learning, and cross-embodiment generalization} -- via a two-stage training paradigm. Stage I injects embodied knowledge into the MLLM through curated digital corpora while maintaining language competence. Stage II trains a policy module through an intent-bridging interface that extracts high-level semantics from the MLLM to guide control, without fine-tuning the MLLM backbone. This process is supported by a self-collected cross-embodiment demonstration suite spanning four robot embodiments and six progressively challenging tasks. Evaluations across digital and physical benchmarks show that a single BLM$_1$ instance outperforms four model families -- MLLMs, ELLMs, VLAs, and GMLMs -- achieving $\sim\!\textbf{6%}$ gains in digital tasks and $\sim\!\textbf{3%}$ in physical tasks.

Paperid: 5518, https://arxiv.org/pdf/2510.24109.pdf

Abstract:
The rapid advancement of Large Language Models (LLMs) has marked a significant breakthrough in Artificial Intelligence (AI), ushering in a new era of Human-centered Artificial Intelligence (HAI). HAI aims to better serve human welfare and needs, thereby placing higher demands on the intelligence level of robots, particularly in aspects such as natural language interaction, complex task planning, and execution. Intelligent agents powered by LLMs have opened up new pathways for realizing HAI. However, existing LLM-based embodied agents often lack the ability to plan and execute complex natural language control tasks online. This paper explores the implementation of intelligent robotic manipulating agents based on Vision-Language Models (VLMs) in the physical world. We propose a novel embodied agent framework for robots, which comprises a human-robot voice interaction module, a vision-language agent module and an action execution module. The vision-language agent itself includes a vision-based task planner, a natural language instruction converter, and a task performance feedback evaluator. Experimental results demonstrate that our agent achieves a 28\% higher average task success rate in both simulated and real environments compared to approaches relying solely on LLM+CLIP, significantly improving the execution success rate of high-level natural language instruction tasks.

Paperid: 5519, https://arxiv.org/pdf/2510.22785.pdf

Abstract:
Pre-trained vision-language models (VLMs) such as CLIP have demonstrated strong zero-shot capabilities across diverse domains, yet remain highly vulnerable to adversarial perturbations that disrupt image-text alignment and compromise reliability. Existing defenses typically rely on adversarial fine-tuning with labeled data, limiting their applicability in zero-shot settings. In this work, we identify two key weaknesses of current CLIP adversarial attacks -- lack of semantic guidance and vulnerability to view variations -- collectively termed semantic and viewpoint fragility. To address these challenges, we propose Self-Calibrated Consistency (SCC), an effective test-time defense. SCC consists of two complementary modules: Semantic consistency, which leverages soft pseudo-labels from counterattack warm-up and multi-view predictions to regularize cross-modal alignment and separate the target embedding from confusable negatives; and Spatial consistency, aligning perturbed visual predictions via augmented views to stabilize inference under adversarial perturbations. Together, these modules form a plug-and-play inference strategy. Extensive experiments on 22 benchmarks under diverse attack settings show that SCC consistently improves the zero-shot robustness of CLIP while maintaining accuracy, and can be seamlessly integrated with other VLMs for further gains. These findings highlight the great potential of establishing an adversarially robust paradigm from CLIP, with implications extending to broader vision-language domains such as BioMedCLIP.

Paperid: 5520, https://arxiv.org/pdf/2510.22443.pdf

Abstract:
There has been a surge of interest in assistive wearable agents: agents embodied in wearable form factors (e.g., smart glasses) who take assistive actions toward a user's goal/query (e.g. "Where did I leave my keys?"). In this work, we consider the important complementary problem of inferring that goal from multi-modal contextual observations. Solving this "goal inference" problem holds the promise of eliminating the effort needed to interact with such an agent. This work focuses on creating WAGIBench, a strong benchmark to measure progress in solving this problem using vision-language models (VLMs). Given the limited prior work in this area, we collected a novel dataset comprising 29 hours of multimodal data from 348 participants across 3,477 recordings, featuring ground-truth goals alongside accompanying visual, audio, digital, and longitudinal contextual observations. We validate that human performance exceeds model performance, achieving 93% multiple-choice accuracy compared with 84% for the best-performing VLM. Generative benchmark results that evaluate several families of modern vision-language models show that larger models perform significantly better on the task, yet remain far from practical usefulness, as they produce relevant goals only 55% of the time. Through a modality ablation, we show that models benefit from extra information in relevant modalities with minimal performance degradation from irrelevant modalities.

Paperid: 5521, https://arxiv.org/pdf/2510.21978.pdf

Abstract:
Reinforcement learning with verifiable rewards (RLVR) has delivered impressive gains in mathematical and multimodal reasoning and has become a standard post-training paradigm for contemporary language and vision-language models. However, the RLVR recipe introduces a significant risk of capability regression, where models forget foundational skills after prolonged training without employing regularization strategies. We empirically confirm this concern, observing that open-source reasoning models suffer performance degradation on core capabilities such as perception and faithfulness. While imposing regularization terms like KL divergence can help prevent deviation from the base model, these terms are calculated on the current task, thus they do not guarantee broader knowledge. Meanwhile, commonly used experience replay across heterogeneous domains makes it nontrivial to decide how much training focus each objective should receive. To address this, we propose RECAP-a replay strategy with dynamic objective reweighting for general knowledge preservation. Our reweighting mechanism adapts in an online manner using short-horizon signals of convergence and instability, shifting the post-training focus away from saturated objectives and toward underperforming or volatile ones. Our method is end-to-end and readily applicable to existing RLVR pipelines without training additional models or heavy tuning. Extensive experiments on benchmarks based on Qwen2.5-VL-3B and Qwen2.5-VL-7B demonstrate the effectiveness of our method, which not only preserves general capabilities but also improves reasoning by enabling more flexible trade-offs among in-task rewards.

Paperid: 5522, https://arxiv.org/pdf/2510.20856.pdf

Abstract:
Vision-Language Models (VLMs), such as CLIP, have demonstrated remarkable zero-shot generalizability across diverse downstream tasks. However, recent studies have revealed that VLMs, including CLIP, are highly vulnerable to adversarial attacks, particularly on their visual modality. Traditional methods for improving adversarial robustness, such as adversarial training, involve extensive retraining and can be computationally expensive. In this paper, we propose a new Test-Time defense: Feature Perception Threshold Counterattack Noise (FPT-Noise), which enhances the adversarial robustness of CLIP without costly fine-tuning. Our core contributions are threefold: First, we introduce a Dynamic Feature Modulator that dynamically generate an image-specific and attack-adaptive noise intensity parameter. Second, We reanalyzed the image features of CLIP. When images are exposed to different levels of noise, clean images and adversarial images exhibit distinct rates of feature change. We established a feature perception threshold to distinguish clean images from attacked ones. Finally, we integrate a Scene-Aware Regulation guided by a stability threshold and leverage Test-Time Transformation Ensembling (TTE) to further mitigate the impact of residual noise and enhance robustness.Extensive experimentation has demonstrated that FPT-Noise significantly outperforms existing Test-Time defense methods, boosting average robust accuracy from 0.07% to 56.86% under AutoAttack while maintaining high performance on clean images (-1.1%). The code will be made public following the publication of the study. The code will be made public following the publication of the study.

Paperid: 5523, https://arxiv.org/pdf/2510.20477.pdf

Abstract:
Exploiting unlabeled data through semi-supervised learning (SSL) or leveraging pre-trained models via fine-tuning are two prevailing paradigms for addressing label-scarce scenarios. Recently, growing attention has been given to combining fine-tuning of pre-trained vision-language models (VLMs) with SSL, forming the emerging paradigm of semi-supervised fine-tuning. However, existing methods often suffer from model bias and hyperparameter sensitivity, due to reliance on prediction consistency or pre-defined confidence thresholds. To address these limitations, we propose a simple yet effective plug-and-play methodology named $\underline{\textbf{Bi-Co}}$nsistency-$\underline{\textbf{G}}$uided Self-Training (Bi-CoG), which assigns high-quality and low-bias pseudo-labels, by simultaneously exploiting inter-model and intra-model consistency, along with an error-aware dynamic pseudo-label assignment strategy. Both theoretical analysis and extensive experiments over 14 datasets demonstrate the effectiveness of Bi-CoG, which consistently and significantly improves the performance of existing methods.

Paperid: 5524, https://arxiv.org/pdf/2510.18214.pdf

Abstract:
Safety evaluation of multimodal foundation models often treats vision and language inputs separately, missing risks from joint interpretation where benign content becomes harmful in combination. Existing approaches also fail to distinguish clearly unsafe content from borderline cases, leading to problematic over-blocking or under-refusal of genuinely harmful content. We present Vision Language Safety Understanding (VLSU), a comprehensive framework to systematically evaluate multimodal safety through fine-grained severity classification and combinatorial analysis across 17 distinct safety patterns. Using a multi-stage pipeline with real-world images and human annotation, we construct a large-scale benchmark of 8,187 samples spanning 15 harm categories. Our evaluation of eleven state-of-the-art models reveals systematic joint understanding failures: while models achieve 90%-plus accuracy on clear unimodal safety signals, performance degrades substantially to 20-55% when joint image-text reasoning is required to determine the safety label. Most critically, 34% of errors in joint image-text safety classification occur despite correct classification of the individual modalities, further demonstrating absent compositional reasoning capabilities. Additionally, we find that models struggle to balance refusing unsafe content while still responding to borderline cases that deserve engagement. For example, we find that instruction framing can reduce the over-blocking rate on borderline content from 62.4% to 10.4% in Gemini-1.5, but only at the cost of under-refusing on unsafe content with refusal rate dropping from 90.8% to 53.9%. Overall, our framework exposes weaknesses in joint image-text understanding and alignment gaps in current models, and provides a critical test bed to enable the next milestones in research on robust vision-language safety.

Paperid: 5525, https://arxiv.org/pdf/2510.17191.pdf

Abstract:
End-to-end autonomous driving has emerged as a promising paradigm for achieving robust and intelligent driving policies. However, existing end-to-end methods still face significant challenges, such as suboptimal decision-making in complex scenarios. In this paper,we propose SimpleVSF (Simple VLM-Scoring Fusion), a novel framework that enhances end-to-end planning by leveraging the cognitive capabilities of Vision-Language Models (VLMs) and advanced trajectory fusion techniques. We utilize the conventional scorers and the novel VLM-enhanced scorers. And we leverage a robust weight fusioner for quantitative aggregation and a powerful VLM-based fusioner for qualitative, context-aware decision-making. As the leading approach in the ICCV 2025 NAVSIM v2 End-to-End Driving Challenge, our SimpleVSF framework demonstrates state-of-the-art performance, achieving a superior balance between safety, comfort, and efficiency.

Paperid: 5526, https://arxiv.org/pdf/2510.14526.pdf

Abstract:
In text-to-image generation, different initial noises induce distinct denoising paths with a pretrained Stable Diffusion (SD) model. While this pattern could output diverse images, some of them may fail to align well with the prompt. Existing methods alleviate this issue either by altering the denoising dynamics or by drawing multiple noises and conducting post-selection. In this paper, we attribute the misalignment to a training-inference mismatch: during training, prompt-conditioned noises lie in a prompt-specific subset of the latent space, whereas at inference the noise is drawn from a prompt-agnostic Gaussian prior. To close this gap, we propose a noise projector that applies text-conditioned refinement to the initial noise before denoising. Conditioned on the prompt embedding, it maps the noise to a prompt-aware counterpart that better matches the distribution observed during SD training, without modifying the SD model. Our framework consists of these steps: we first sample some noises and obtain token-level feedback for their corresponding images from a vision-language model (VLM), then distill these signals into a reward model, and finally optimize the noise projector via a quasi-direct preference optimization. Our design has two benefits: (i) it requires no reference images or handcrafted priors, and (ii) it incurs small inference cost, replacing multi-sample selection with a single forward pass. Extensive experiments further show that our prompt-aware noise projection improves text-image alignment across diverse prompts.

Paperid: 5527, https://arxiv.org/pdf/2510.13993.pdf

Abstract:
Remote sensing has become a vital tool across sectors such as urban planning, environmental monitoring, and disaster response. While the volume of data generated has increased significantly, traditional vision models are often constrained by the requirement for extensive domain-specific labelled data and their limited ability to understand the context within complex environments. Vision Language Models offer a complementary approach by integrating visual and textual data; however, their application to remote sensing remains underexplored, particularly given their generalist nature. This work investigates the combination of vision models and VLMs to enhance image analysis in remote sensing, with a focus on aircraft detection and scene understanding. The integration of YOLO with VLMs such as LLaVA, ChatGPT, and Gemini aims to achieve more accurate and contextually aware image interpretation. Performance is evaluated on both labelled and unlabelled remote sensing data, as well as degraded image scenarios which are crucial for remote sensing. The findings show an average MAE improvement of 48.46% across models in the accuracy of aircraft detection and counting, especially in challenging conditions, in both raw and degraded scenarios. A 6.17% improvement in CLIPScore for comprehensive understanding of remote sensing images is obtained. The proposed approach combining traditional vision models and VLMs paves the way for more advanced and efficient remote sensing image analysis, especially in few-shot learning scenarios.

Paperid: 5528, https://arxiv.org/pdf/2510.10889.pdf

Abstract:
Contrastive Vision-Language Models (VLMs) have demonstrated strong zero-shot capabilities. However, their cross-modal alignment remains biased toward English due to limited multilingual multimodal data. Recent multilingual extensions have alleviated this gap but enforce instance-level alignment while neglecting the global geometry of the shared embedding space. We address this problem by introducing ToMCLIP (Topological Alignment for Multilingual CLIP), a topology-aware framework aligning embedding spaces with topology-preserving constraints. The proposed method applies persistent homology to define a topological alignment loss and approximates persistence diagram with theoretical error bounds using graph sparsification strategy. This work validates the proposed approach, showing enhanced structural coherence of multilingual representations, higher zero-shot accuracy on the CIFAR-100, and stronger multilingual retrieval performance on the xFlickr&CO. Beyond VLMs, the proposed approach provides a general method for incorporating topological alignment into representation learning.

Paperid: 5529, https://arxiv.org/pdf/2510.06664.pdf

Abstract:
Agents utilizing tools powered by large language models (LLMs) or vision-language models (VLMs) have demonstrated remarkable progress in diverse tasks across text and visual modalities. Unlike traditional tools such as calculators, which give deterministic outputs, neural tools perform uncertainly across task scenarios. While different tools for a task may excel in varied scenarios, existing agents typically rely on fixed tools, thus limiting the flexibility in selecting the most suitable tool for specific tasks. In contrast, humans snowball their understanding of the capabilities of different tools by interacting with them, and apply this knowledge to select the optimal tool when solving a future task. To build agents that similarly benefit from this process, we propose ToolMem that enables agents to develop memories of tool capabilities from previous interactions, by summarizing their strengths and weaknesses and storing them in memory; at inference, the agent can retrieve relevant entries from ToolMem, and select the best tool to solve individual tasks more accurately. We evaluate ToolMem on learning varied text generation and text-to-image generation neural tools. Compared to no-memory, generic agents, we find ToolMem-augmented agents predict tool performance 14.8% and 28.7% more accurately across text and multimodal generation scenarios. Moreover, ToolMem facilitates optimal tool selection among multiple choices by 21% and 24% absolute increases in respective scenarios.

Paperid: 5530, https://arxiv.org/pdf/2510.04145.pdf

Abstract:
Conventional construction safety inspection methods are often inefficient as they require navigating through large volume of information. Recent advances in large vision-language models (LVLMs) provide opportunities to automate safety inspections through enhanced visual and linguistic understanding. However, existing applications face limitations including irrelevant or unspecific responses, restricted modal inputs and hallucinations. Utilisation of Large Language Models (LLMs) for this purpose is constrained by availability of training data and frequently lack real-time adaptability. This study introduces SiteShield, a multi-modal LVLM-based Retrieval-Augmented Generation (RAG) framework for automating construction safety inspection reports by integrating visual and audio inputs. Using real-world data, SiteShield outperformed unimodal LLMs without RAG with an F1 score of 0.82, hamming loss of 0.04, precision of 0.76, and recall of 0.96. The findings indicate that SiteShield offers a novel pathway to enhance information retrieval and efficiency in generating safety reports.

Paperid: 5531, https://arxiv.org/pdf/2510.03519.pdf

Abstract:
Time series reasoning is crucial to decision-making in diverse domains, including finance, energy usage, traffic, weather, and scientific discovery. While existing time series foundation models (TSFMs) can capture low-level dynamic patterns and provide accurate forecasting, further analysis usually requires additional background knowledge and sophisticated reasoning, which are lacking in most TSFMs but can be achieved through large language models (LLMs). On the other hand, without expensive post-training, LLMs often struggle with the numerical understanding of time series data. Although it is intuitive to integrate the two types of models, developing effective training recipes that align the two modalities for reasoning tasks is still an open challenge. To this end, we propose TS-Reasoner that aligns the latent representations of TSFMs with the textual inputs of LLMs for downstream understanding/reasoning tasks. Specifically, we propose a simple yet effective method to curate diverse, synthetic pairs of time series and textual captions for alignment training. We then develop a two-stage training recipe that applies instruction finetuning after the alignment pretraining. Unlike existing works that train an LLM to take time series as inputs, we leverage a pretrained TSFM and freeze it during training. Extensive experiments on several benchmarks demonstrate that TS-Reasoner not only outperforms a wide range of prevailing LLMs, Vision Language Models (VLMs), and Time Series LLMs, but also achieves this with remarkable data efficiency, e.g., using less than half the training data.

Paperid: 5532, https://arxiv.org/pdf/2510.01483.pdf

Abstract:
Vision-language models (VLMs) have shown potential for robot navigation but encounter fundamental limitations: they lack persistent scene memory, offer limited spatial reasoning, and do not scale effectively with video duration for real-time application. We present VL-KnG, a Visual Scene Understanding system that tackles these challenges using spatiotemporal knowledge graph construction and computationally efficient query processing for navigation goal identification. Our approach processes video sequences in chunks utilizing modern VLMs, creates persistent knowledge graphs that maintain object identity over time, and enables explainable spatial reasoning through queryable graph structures. We also introduce WalkieKnowledge, a new benchmark with about 200 manually annotated questions across 8 diverse trajectories spanning approximately 100 minutes of video data, enabling fair comparison between structured approaches and general-purpose VLMs. Real-world deployment on a differential drive robot demonstrates practical applicability, with our method achieving 77.27% success rate and 76.92% answer accuracy, matching Gemini 2.5 Pro performance while providing explainable reasoning supported by the knowledge graph, computational efficiency for real-time deployment across different tasks, such as localization, navigation and planning. Code and dataset will be released after acceptance.

Paperid: 5533, https://arxiv.org/pdf/2510.01297.pdf

Abstract:
Large Language Models (LLMs) open new possibilities for constructing realistic and interpretable macroeconomic simulations. We present SimCity, a multi-agent framework that leverages LLMs to model an interpretable macroeconomic system with heterogeneous agents and rich interactions. Unlike classical equilibrium models that limit heterogeneity for tractability, or traditional agent-based models (ABMs) that rely on hand-crafted decision rules, SimCity enables flexible, adaptive behavior with transparent natural-language reasoning. Within SimCity, four core agent types (households, firms, a central bank, and a government) deliberate and participate in a frictional labor market, a heterogeneous goods market, and a financial market. Furthermore, a Vision-Language Model (VLM) determines the geographic placement of new firms and renders a mapped virtual city, allowing us to study both macroeconomic regularities and urban expansion dynamics within a unified environment. To evaluate the framework, we compile a checklist of canonical macroeconomic phenomena, including price elasticity of demand, Engel's Law, Okun's Law, the Phillips Curve, and the Beveridge Curve, and show that SimCity naturally reproduces these empirical patterns while remaining robust across simulation runs.

Paperid: 5534, https://arxiv.org/pdf/2509.24385.pdf

Abstract:
Recent developments in Multimodal Large Language Models (MLLMs) have significantly improved Vision-Language (VL) reasoning in 2D domains. However, extending these capabilities to 3D scene understanding remains a major challenge. Existing 3D Multimodal Large Language Models (3D-MLLMs) often depend on 3D data inputs, which limits scalability and generalization. To address this limitation, we propose Vid-LLM, a video-based 3D-MLLM that directly processes video inputs without requiring external 3D data, making it practical for real-world deployment. In our method, the geometric prior are directly used to improve the performance of the sceen perception. To integrate the geometric cues into the MLLM compactly, we design a Cross-Task Adapter (CTA) module to align the 3D geometric priors with the vision-language representations. To ensure geometric consistency and integrity, we introduce a Metric Depth Model that recovers real-scale geometry from the reconstruction outputs. Finally, the model is fine-tuned with a two-stage distillation optimization strategy, realizing fast convergence and stabilizes training. Extensive experiments across diverse benchmarks verified the effectiveness of our method on 3D Question Answering, 3D Dense Captioning and 3D Visual Grounding tasks, demonstrating the superior multi-task capabilities.

Paperid: 5535, https://arxiv.org/pdf/2509.23783.pdf

Abstract:
Existing methods for evaluating the harmfulness of content generated by large language models (LLMs) have been well studied. However, approaches tailored to multimodal large language models (MLLMs) remain underdeveloped and lack depth. This work highlights the crucial role of visual information in moderating content in visual question answering (VQA), a dimension often overlooked in current research. To bridge this gap, we introduce Falcon, a large-scale vision-language safety dataset containing 57,515 VQA pairs across 13 harm categories. The dataset provides explicit annotations for harmful attributes across images, instructions, and responses, thereby facilitating a comprehensive evaluation of the content generated by MLLMs. In addition, it includes the relevant harm categories along with explanations supporting the corresponding judgments. We further propose FalconEye, a specialized evaluator fine-tuned from Qwen2.5-VL-7B using the Falcon dataset. Experimental results demonstrate that FalconEye reliably identifies harmful content in complex and safety-critical multimodal dialogue scenarios. It outperforms all other baselines in overall accuracy across our proposed Falcon-test dataset and two widely-used benchmarks-VLGuard and Beavertail-V, underscoring its potential as a practical safety auditing tool for MLLMs.

Paperid: 5536, https://arxiv.org/pdf/2509.22225.pdf

Abstract:
Lifting 2D open-vocabulary understanding into 3D Gaussian Splatting (3DGS) scenes is a critical challenge. However, mainstream methods suffer from three key flaws: (i) their reliance on costly per-scene retraining prevents plug-and-play application; (ii) their restrictive monosemous design fails to represent complex, multi-concept semantics; and (iii) their vulnerability to cross-view semantic inconsistencies corrupts the final semantic representation. To overcome these limitations, we introduce MUSplat, a training-free framework that abandons feature optimization entirely. Leveraging a pre-trained 2D segmentation model, our pipeline generates and lifts multi-granularity 2D masks into 3D, where we estimate a foreground probability for each Gaussian point to form initial object groups. We then optimize the ambiguous boundaries of these initial groups using semantic entropy and geometric opacity. Subsequently, by interpreting the object's appearance across its most representative viewpoints, a Vision-Language Model (VLM) distills robust textual features that reconciles visual inconsistencies, enabling open-vocabulary querying via semantic matching. By eliminating the costly per-scene training process, MUSplat reduces scene adaptation time from hours to mere minutes. On benchmark tasks for open-vocabulary 3D object selection and semantic segmentation, MUSplat outperforms established training-based frameworks while simultaneously addressing their monosemous limitations.

Paperid: 5537, https://arxiv.org/pdf/2509.21205.pdf

Abstract:
While table understanding increasingly relies on pixel-only settings where tables are processed as visual representations, current benchmarks predominantly use synthetic renderings that lack the complexity and visual diversity of real-world tables. Additionally, existing visual table understanding (VTU) datasets offer fixed examples with single visualizations and pre-defined instructions, providing no access to underlying serialized data for reformulation. We introduce TABLET, a large-scale VTU dataset with 4 million examples across 20 tasks, grounded in 2 million unique tables where 88% preserve original visualizations. Each example includes paired image-HTML representations, comprehensive metadata, and provenance information linking back to the source datasets. Fine-tuning vision-language models like Qwen2.5-VL-7B on TABLET improves performance on seen and unseen VTU tasks while increasing robustness on real-world table visualizations. By preserving original visualizations and maintaining example traceability in a unified large-scale collection, TABLET establishes a foundation for robust training and extensible evaluation of future VTU models.

Paperid: 5538, https://arxiv.org/pdf/2509.18672.pdf

Abstract:
People with visual impairments often face significant challenges in locating and retrieving objects in their surroundings. Existing assistive technologies present a trade-off: systems that offer precise guidance typically require pre-scanning or support only fixed object categories, while those with open-world object recognition lack spatial feedback for reaching the object. To address this gap, we introduce 'NaviSense', a mobile assistive system that combines conversational AI, vision-language models, augmented reality (AR), and LiDAR to support open-world object detection with real-time audio-haptic guidance. Users specify objects via natural language and receive continuous spatial feedback to navigate toward the target without needing prior setup. Designed with insights from a formative study and evaluated with 12 blind and low-vision participants, NaviSense significantly reduced object retrieval time and was preferred over existing tools, demonstrating the value of integrating open-world perception with precise, accessible guidance.

Paperid: 5539, https://arxiv.org/pdf/2509.14769.pdf

Abstract:
Comparing vision language models on videos is particularly complex, as the performances is jointly determined by the model's visual representation capacity and the frame-sampling strategy used to construct the input. Current video benchmarks are suspected to suffer from substantial frame-sampling bias, as models are evaluated with different frame selection strategies. In this work, we propose the first frame-accurate benchmark of state-of-the-art small VLMs for video question-answering, evaluated under controlled frame-sampling strategies. Our results confirm the suspected bias and highlight both data-specific and task-specific behaviors of SVLMs under different frame-sampling techniques. By open-sourcing our benchmarking code, we provide the community with a reproducible and unbiased protocol for evaluating video VLMs and emphasize the need for standardized frame-sampling strategies tailored to each benchmarking dataset in future research.

Paperid: 5540, https://arxiv.org/pdf/2509.14227.pdf

Abstract:
While recent advancements in vision-language models have improved video understanding, diagnosing their capacity for deep, narrative comprehension remains a challenge. Existing benchmarks often test short-clip recognition or use template-based questions, leaving a critical gap in evaluating fine-grained reasoning over long-form narrative content. To address these gaps, we introduce $\mathsf{Cin\acute{e}aste}$, a comprehensive benchmark for long-form movie understanding. Our dataset comprises 3,119 multiple-choice question-answer pairs derived from 1,805 scenes across 200 diverse movies, spanning five novel fine-grained contextual reasoning categories. We use GPT-4o to generate diverse, context-rich questions by integrating visual descriptions, captions, scene titles, and summaries, which require deep narrative understanding. To ensure high-quality evaluation, our pipeline incorporates a two-stage filtering process: Context-Independence filtering ensures questions require video context, while Contextual Veracity filtering validates factual consistency against the movie content, mitigating hallucinations. Experiments show that existing MLLMs struggle on $\mathsf{Cin\acute{e}aste}$; our analysis reveals that long-range temporal reasoning is a primary bottleneck, with the top open-source model achieving only 63.15\% accuracy. This underscores significant challenges in fine-grained contextual understanding and the need for advancements in long-form movie comprehension.

Paperid: 5541, https://arxiv.org/pdf/2509.13939.pdf

Abstract:
Visual counting is a fundamental yet challenging task, especially when users need to count objects of a specific type in complex scenes. While recent models, including class-agnostic counting models and large vision-language models (VLMs), show promise in counting tasks, their ability to perform fine-grained, intent-driven counting remains unclear. In this paper, we introduce PairTally, a benchmark dataset specifically designed to evaluate fine-grained visual counting. Each of the 681 high-resolution images in PairTally contains two object categories, requiring models to distinguish and count based on subtle differences in shape, size, color, or semantics. The dataset includes both inter-category (distinct categories) and intra-category (closely related subcategories) settings, making it suitable for rigorous evaluation of selective counting capabilities. We benchmark a variety of state-of-the-art models, including exemplar-based methods, language-prompted models, and large VLMs. Our results show that despite recent advances, current models struggle to reliably count what users intend, especially in fine-grained and visually ambiguous cases. PairTally provides a new foundation for diagnosing and improving fine-grained visual counting systems.

Paperid: 5542, https://arxiv.org/pdf/2509.13836.pdf

Abstract:
Object hallucination in Large Vision-Language Models (LVLMs) significantly impedes their real-world applicability. As the primary component for accurately interpreting visual information, the choice of visual encoder is pivotal. We hypothesize that the diverse training paradigms employed by different visual encoders instill them with distinct inductive biases, which leads to their diverse hallucination performances. Existing benchmarks typically focus on coarse-grained hallucination detection and fail to capture the diverse hallucinations elaborated in our hypothesis. To systematically analyze these effects, we introduce VHBench-10, a comprehensive benchmark with approximately 10,000 samples for evaluating LVLMs across ten fine-grained hallucination categories. Our evaluations confirm encoders exhibit unique hallucination characteristics. Building on these insights and the suboptimality of simple feature fusion, we propose VisionWeaver, a novel Context-Aware Routing Network. It employs global visual features to generate routing signals, dynamically aggregating visual features from multiple specialized experts. Comprehensive experiments confirm the effectiveness of VisionWeaver in significantly reducing hallucinations and improving overall model performance.

Paperid: 5543, https://arxiv.org/pdf/2509.13731.pdf

Abstract:
The industrial insertion of flexible flat cables (FFCs) into receptacles presents a significant challenge owing to the need for submillimeter precision when handling the deformable cables. In manufacturing processes, FFC insertion with robotic manipulators often requires laborious human-guided trajectory generation. While Reinforcement Learning (RL) offers a solution to automate this task without modeling complex properties of FFCs, the nondeterminism caused by the deformability of FFCs requires significant efforts and time on training. Moreover, training directly in a real environment is dangerous as industrial robots move fast and possess no safety measure. We propose an RL algorithm for FFC insertion that leverages a foundation model-based real-to-sim approach to reduce the training time and eliminate the risk of physical damages to robots and surroundings. Training is done entirely in simulation, allowing for random exploration without the risk of physical damages. Sim-to-real transfer is achieved through semantic segmentation masks which leave only those visual features relevant to the insertion tasks such as the geometric and spatial information of the cables and receptacles. To enhance generality, we use a foundation model, Segment Anything Model 2 (SAM2). To eleminate human intervention, we employ a Vision-Language Model (VLM) to automate the initial prompting of SAM2 to find segmentation masks. In the experiments, our method exhibits zero-shot capabilities, which enable direct deployments to real environments without fine-tuning.

Paperid: 5544, https://arxiv.org/pdf/2509.11840.pdf

Abstract:
Generative vision-language models (VLMs) exhibit strong high-level image understanding but lack spatially dense alignment between vision and language modalities, as our findings indicate. Orthogonal to advancements in generative VLMs, another line of research has focused on representation learning for vision-language alignment, targeting zero-shot inference for dense tasks like segmentation. In this work, we bridge these two directions by densely aligning images with synthetic descriptions generated by VLMs. Synthetic captions are inexpensive, scalable, and easy to generate, making them an excellent source of high-level semantic understanding for dense alignment methods. Empirically, our approach outperforms prior work on standard zero-shot open-vocabulary segmentation benchmarks/datasets, while also being more data-efficient.

Paperid: 5545, https://arxiv.org/pdf/2509.11065.pdf

Abstract:
Block-based programming environments such as Scratch are increasingly popular in programming education, in particular for young learners. While the use of blocks helps prevent syntax errors, semantic bugs remain common and difficult to debug. Existing tools for Scratch debugging rely heavily on predefined rules or user manual inputs, and crucially, they ignore the platform's inherently visual nature. We introduce ViScratch, the first multimodal feedback generation system for Scratch that leverages both the project's block code and its generated gameplay video to diagnose and repair bugs. ViScratch uses a two-stage pipeline: a vision-language model first aligns visual symptoms with code structure to identify a single critical issue, then proposes minimal, abstract syntax tree level repairs that are verified via execution in the Scratch virtual machine. We evaluate ViScratch on a set of real-world Scratch projects against state-of-the-art LLM-based tools and human testers. Results show that gameplay video is a crucial debugging signal: ViScratch substantially outperforms prior tools in both bug identification and repair quality, even without access to project descriptions or goals. This work demonstrates that video can serve as a first-class specification in visual programming environments, opening new directions for LLM-based debugging beyond symbolic code alone.

Paperid: 5546, https://arxiv.org/pdf/2509.10842.pdf

Abstract:
Open-vocabulary semantic segmentation enables models to recognize and segment objects from arbitrary natural language descriptions, offering the flexibility to handle novel, fine-grained, or functionally defined categories beyond fixed label sets. While this capability is crucial for large-scale urban point clouds that support applications such as digital twins, smart city management, and urban analytics, it remains largely unexplored in this domain. The main obstacles are the frequent absence of high-quality, well-aligned multi-view imagery in large-scale urban point cloud datasets and the poor generalization of existing three-dimensional (3D) segmentation pipelines across diverse urban environments with substantial variation in geometry, scale, and appearance. To address these challenges, we present OpenUrban3D, the first 3D open-vocabulary semantic segmentation framework for large-scale urban scenes that operates without aligned multi-view images, pre-trained point cloud segmentation networks, or manual annotations. Our approach generates robust semantic features directly from raw point clouds through multi-view, multi-granularity rendering, mask-level vision-language feature extraction, and sample-balanced fusion, followed by distillation into a 3D backbone model. This design enables zero-shot segmentation for arbitrary text queries while capturing both semantic richness and geometric priors. Extensive experiments on large-scale urban benchmarks, including SensatUrban and SUM, show that OpenUrban3D achieves significant improvements in both segmentation accuracy and cross-scene generalization over existing methods, demonstrating its potential as a flexible and scalable solution for 3D urban scene understanding.

Paperid: 5547, https://arxiv.org/pdf/2509.10683.pdf

Abstract:
Large Language Models (LLMs) have shown strong performance in text-based healthcare tasks. However, their utility in image-based applications remains unexplored. We investigate the effectiveness of LLMs for medical imaging tasks, specifically glioma classification and segmentation, and compare their performance to that of traditional convolutional neural networks (CNNs). Using the BraTS 2020 dataset of multi-modal brain MRIs, we evaluated a general-purpose vision-language LLM (LLaMA 3.2 Instruct) both before and after fine-tuning, and benchmarked its performance against custom 3D CNNs. For glioma classification (Low-Grade vs. High-Grade), the CNN achieved 80% accuracy and balanced precision and recall. The general LLM reached 76% accuracy but suffered from a specificity of only 18%, often misclassifying Low-Grade tumors. Fine-tuning improved specificity to 55%, but overall performance declined (e.g., accuracy dropped to 72%). For segmentation, three methods - center point, bounding box, and polygon extraction, were implemented. CNNs accurately localized gliomas, though small tumors were sometimes missed. In contrast, LLMs consistently clustered predictions near the image center, with no distinction of glioma size, location, or placement. Fine-tuning improved output formatting but failed to meaningfully enhance spatial accuracy. The bounding polygon method yielded random, unstructured outputs. Overall, CNNs outperformed LLMs in both tasks. LLMs showed limited spatial understanding and minimal improvement from fine-tuning, indicating that, in their current form, they are not well-suited for image-based tasks. More rigorous fine-tuning or alternative training strategies may be needed for LLMs to achieve better performance, robustness, and utility in the medical space.

Paperid: 5548, https://arxiv.org/pdf/2509.10105.pdf

Abstract:
We introduce VARCO-VISION-2.0, an open-weight bilingual vision-language model (VLM) for Korean and English with improved capabilities compared to the previous model VARCO-VISION-14B. The model supports multi-image understanding for complex inputs such as documents, charts, and tables, and delivers layoutaware OCR by predicting both textual content and its spatial location. Trained with a four-stage curriculum with memory-efficient techniques, the model achieves enhanced multimodal alignment, while preserving core language abilities and improving safety via preference optimization. Extensive benchmark evaluations demonstrate strong spatial grounding and competitive results for both languages, with the 14B model achieving 8th place on the OpenCompass VLM leaderboard among models of comparable scale. Alongside the 14B-scale model, we release a 1.7B version optimized for on-device deployment. We believe these models advance the development of bilingual VLMs and their practical applications. Two variants of VARCO-VISION-2.0 are available at Hugging Face: a full-scale 14B model and a lightweight 1.7B model.

Paperid: 5549, https://arxiv.org/pdf/2509.09597.pdf

Abstract:
Graph alignment, the problem of identifying corresponding nodes across multiple graphs, is fundamental to numerous applications. Most existing unsupervised methods embed node features into latent representations to enable cross-graph comparison without ground-truth correspondences. However, these methods suffer from two critical limitations: the degradation of node distinctiveness due to oversmoothing in GNN-based embeddings, and the misalignment of latent spaces across graphs caused by structural noise, feature heterogeneity, and training instability, ultimately leading to unreliable node correspondences. We propose a novel graph alignment framework that simultaneously enhances node distinctiveness and enforces geometric consistency across latent spaces. Our approach introduces a dual-pass encoder that combines low-pass and high-pass spectral filters to generate embeddings that are both structure-aware and highly discriminative. To address latent space misalignment, we incorporate a geometry-aware functional map module that learns bijective and isometric transformations between graph embeddings, ensuring consistent geometric relationships across different representations. Extensive experiments on graph benchmarks demonstrate that our method consistently outperforms existing unsupervised alignment baselines, exhibiting superior robustness to structural inconsistencies and challenging alignment scenarios. Additionally, comprehensive evaluation on vision-language benchmarks using diverse pretrained models shows that our framework effectively generalizes beyond graph domains, enabling unsupervised alignment of vision and language representations.

Paperid: 5550, https://arxiv.org/pdf/2509.05333.pdf

Abstract:
Real world deployments often expose modern object recognition models to domain shifts that precipitate a severe drop in accuracy. Such shifts encompass (i) variations in low level image statistics, (ii) changes in object pose and viewpoint, (iii) partial occlusion, and (iv) visual confusion across adjacent classes. To mitigate this degradation, we introduce the Re-Thinking Vision Language Model (RT-VLM) framework. The foundation of this framework is a unique synthetic dataset generation pipeline that produces images annotated with "4-Clues": precise bounding boxes, class names, detailed object-level captions, and a comprehensive context-level caption for the entire scene. We then perform parameter efficient supervised tuning of Llama 3.2 11B Vision Instruct on this resource. At inference time, a two stage Re-Thinking scheme is executed: the model first emits its own four clues, then re examines these responses as evidence and iteratively corrects them. Across robustness benchmarks that isolate individual domain shifts, RT-VLM consistently surpasses strong baselines. These findings indicate that the integration of structured multimodal evidence with an explicit self critique loop constitutes a promising route toward reliable and transferable visual understanding.

Paperid: 5551, https://arxiv.org/pdf/2509.04894.pdf

Abstract:
We propose an approach to generate synthetic data to train computer vision (CV) models for industrial wear and tear detection. Wear and tear detection is an important CV problem for predictive maintenance tasks in any industry. However, data curation for training such models is expensive and time-consuming due to the unavailability of datasets for different wear and tear scenarios. Our approach employs a vision language model along with a 3D simulation and rendering engine to generate synthetic data for varying rust conditions. We evaluate our approach by training a CV model for rust detection using the generated dataset and tested the trained model on real images of rusted industrial objects. The model trained with the synthetic data generated by our approach, outperforms the other approaches with a mAP50 score of 0.87. The approach is customizable and can be easily extended to other industrial wear and tear detection scenarios

Paperid: 5552, https://arxiv.org/pdf/2509.04324.pdf

Abstract:
Grasping assistance is essential for restoring autonomy in individuals with motor impairments, particularly in unstructured environments where object categories and user intentions are diverse and unpredictable. We present OVGrasp, a hierarchical control framework for soft exoskeleton-based grasp assistance that integrates RGB-D vision, open-vocabulary prompts, and voice commands to enable robust multimodal interaction. To enhance generalization in open environments, OVGrasp incorporates a vision-language foundation model with an open-vocabulary mechanism, allowing zero-shot detection of previously unseen objects without retraining. A multimodal decision-maker further fuses spatial and linguistic cues to infer user intent, such as grasp or release, in multi-object scenarios. We deploy the complete framework on a custom egocentric-view wearable exoskeleton and conduct systematic evaluations on 15 objects across three grasp types. Experimental results with ten participants demonstrate that OVGrasp achieves a grasping ability score (GAS) of 87.00%, outperforming state-of-the-art baselines and achieving improved kinematic alignment with natural hand motion.

Paperid: 5553, https://arxiv.org/pdf/2509.02864.pdf

Abstract:
We present an end-to-end, self-evolving adversarial workflow for long-context Question-Answer (QA) Generation in Arabic. By orchestrating multiple specialized LVLMs: a question generator, an evaluator, and a swarm of answer generators, our system iteratively refines its own performance without any human intervention. Starting from raw, multi-page Arabic documents across diverse domains, the question generator produces fine-grained, context-aware queries to be tackled by the answer generator swarm, and the evaluator assesses and feeds back quality metrics. This closed-loop cycle enables continuous learning: low-confidence outputs trigger automated re-generation and model updates, progressively enhancing question difficulty and relevance. Moreover, we set the quality metrics as a tunable hyperparameter, enabling question generation at controllable and customizable difficulty levels. We release AraLongBench, a large-scale Arabic benchmark of single- and multi-page challenges spanning hundreds of pages, and demonstrate that our self-evolving workflow substantially outperform static pipelines, markedly boosting the long-context comprehension capabilities of leading Arabic Large Vision Language Models (LVLMs). Lastly, we also meticulously architect a fully automated agentic workflow for long-context Arabic document collection.

Paperid: 5554, https://arxiv.org/pdf/2509.02367.pdf

Abstract:
Virtual assistants (VAs) have become ubiquitous in daily life, integrated into smartphones and smart devices, sparking interest in AI companions that enhance user experiences and foster emotional connections. However, existing companions are often embedded in specific objects-such as glasses, home assistants, or dolls-requiring users to form emotional bonds with unfamiliar items, which can lead to reduced engagement and feelings of detachment. To address this, we introduce Talking Spell, a wearable system that empowers users to imbue any everyday object with speech and anthropomorphic personas through a user-centric radiative network. Leveraging advanced computer vision (e.g., YOLOv11 for object detection), large vision-language models (e.g., QWEN-VL for persona generation), speech-to-text and text-to-speech technologies, Talking Spell guides users through three stages of emotional connection: acquaintance, familiarization, and bonding. We validated our system through a user study involving 12 participants, utilizing Talking Spell to explore four interaction intentions: entertainment, companionship, utility, and creativity. The results demonstrate its effectiveness in fostering meaningful interactions and emotional significance with everyday objects. Our findings indicate that Talking Spell creates engaging and personalized experiences, as demonstrated through various devices, ranging from accessories to essential wearables.

Paperid: 5555, https://arxiv.org/pdf/2508.21738.pdf

Abstract:
The high cost of acquiring rural street view images has constrained comprehensive environmental perception in rural areas. Drone photographs, with their advantages of easy acquisition, broad coverage, and high spatial resolution, offer a viable approach for large-scale rural environmental perception. However, a systematic methodology for identifying key environmental elements from drone photographs and quantifying their impact on environmental perception remains lacking. To address this gap, a Vision-Language Contrastive Ranking Framework (VLCR) is designed for rural livability assessment in China. The framework employs chain-of-thought prompting strategies to guide multimodal large language models (MLLMs) in identifying visual features related to quality of life and ecological habitability from drone photographs. Subsequently, to address the instability in pairwise village comparison, a text description-constrained drone photograph comparison strategy is proposed. Finally, to overcome the efficiency bottleneck in nationwide pairwise village comparisons, an innovation ranking algorithm based on binary search interpolation is developed, which reduces the number of comparisons through automated selection of comparison targets. The proposed framework achieves superior performance with a Spearman Footrule distance of 0.74, outperforming mainstream commercial MLLMs by approximately 0.1. Moreover, the mechanism of concurrent comparison and ranking demonstrates a threefold enhancement in computational efficiency. Our framework has achieved data innovation and methodological breakthroughs in village livability assessment, providing strong support for large-scale village livability analysis. Keywords: Drone photographs, Environmental perception, Rural livability assessment, Multimodal large language models, Chain-of-thought prompting.

Paperid: 5556, https://arxiv.org/pdf/2508.21565.pdf

Abstract:
Effectively understanding urban scenes requires fine-grained spatial reasoning about objects, layouts, and depth cues. However, how well current vision-language models (VLMs), pretrained on general scenes, transfer these abilities to urban domain remains underexplored. To address this gap, we conduct a comparative study of three off-the-shelf VLMs-BLIP-2, InstructBLIP, and LLaVA-1.5-evaluating both zero-shot performance and the effects of fine-tuning with a synthetic VQA dataset specific to urban scenes. We construct such dataset from segmentation, depth, and object detection predictions of street-view images, pairing each question with LLM-generated Chain-of-Thought (CoT) answers for step-by-step reasoning supervision. Results show that while VLMs perform reasonably well in zero-shot settings, fine-tuning with our synthetic CoT-supervised dataset substantially boosts performance, especially for challenging question types such as negation and counterfactuals. This study introduces urban spatial reasoning as a new challenge for VLMs and demonstrates synthetic dataset construction as a practical path for adapting general-purpose models to specialized domains.

Paperid: 5557, https://arxiv.org/pdf/2508.18297.pdf

Abstract:
Through a controlled study, we identify a systematic deficiency in the multimodal grounding of Vision Language Models (VLMs). While VLMs can recall factual associations when provided a textual reference to an entity; their ability to do so is significantly diminished when the reference is visual instead. Forcing VLMs to rely on image representations of an entity halves their ability to recall factual knowledge, suggesting that VLMs struggle to link their internal knowledge of an entity with its image representation. We show that such linking failures are correlated with the expression of distinct patterns in model internal states, and that probes on these internal states achieve over 92% accuracy at flagging cases where the VLM response is unreliable. These probes can be applied, without retraining, to identify when a VLM will fail to correctly answer a question that requires an understanding of multimodal input. When used to facilitate selective prediction on a visual question answering task, the probes increase coverage by 7.87% (absolute) while also reducing the risk of error by 0.9% (absolute). Addressing the systematic, detectable deficiency is an important avenue in language grounding, and we provide informed recommendations for future directions.

Paperid: 5558, https://arxiv.org/pdf/2508.15297.pdf

Abstract:
In the field of design patent analysis, traditional tasks such as patent classification and patent image retrieval heavily depend on the image data. However, patent images -- typically consisting of sketches with abstract and structural elements of an invention -- often fall short in conveying comprehensive visual context and semantic information. This inadequacy can lead to ambiguities in evaluation during prior art searches. Recent advancements in vision-language models, such as CLIP, offer promising opportunities for more reliable and accurate AI-driven patent analysis. In this work, we leverage CLIP models to develop a unified framework DesignCLIP for design patent applications with a large-scale dataset of U.S. design patents. To address the unique characteristics of patent data, DesignCLIP incorporates class-aware classification and contrastive learning, utilizing generated detailed captions for patent images and multi-views image learning. We validate the effectiveness of DesignCLIP across various downstream tasks, including patent classification and patent retrieval. Additionally, we explore multimodal patent retrieval, which provides the potential to enhance creativity and innovation in design by offering more diverse sources of inspiration. Our experiments show that DesignCLIP consistently outperforms baseline and SOTA models in the patent domain on all tasks. Our findings underscore the promise of multimodal approaches in advancing patent analysis. The codebase is available here: https://anonymous.4open.science/r/PATENTCLIP-4661/README.md.

Paperid: 5559, https://arxiv.org/pdf/2508.11870.pdf

Abstract:
Adapter-based fine-tuning has gained remarkable attention in adapting large pre-trained vision language models (VLMs) for a wide range of downstream tasks efficiently. In this paradigm, only the inserted adapters are fine-tuned, without the need for training the original VLM backbone. Existing works scale adapters by integrating them into every layer of VLMs to increase the capacity of adapters. However, these methods face two primary limitations: 1) limited compression rate due to ignoring cross-layer redundancy, and 2) limited representational capacity across homogeneous adapters. In this paper, we propose a novel vision-language fine-tuning framework based on cross-layer tensor ring decomposition (TRD) with the integration and collaboration of diverse adapters, called AdaRing, achieving ultra-light parameter-efficient adaptation of VLMs on various tasks. To remove the high redundancy that exists among adapters across layers, we exploit the tensor-level low-rankness to formulate adapters as layer-shared tensor cores and layer-specific slices. Moreover, guided by generalization-aware fine-tuning, diverse rank-driven adapters cooperate to handle tasks that require different representations. Our experiments show that the proposed AdaRing achieves the state-of-the-art performance while reducing average training parameters by 90%.

Paperid: 5560, https://arxiv.org/pdf/2508.10771.pdf

Abstract:
Recent advances in AI-generated content have fueled the rise of highly realistic synthetic videos, posing severe risks to societal trust and digital integrity. Existing benchmarks for video authenticity detection typically suffer from limited realism, insufficient scale, and inadequate complexity, failing to effectively evaluate modern vision-language models against sophisticated forgeries. To address this critical gap, we introduce AEGIS, a novel large-scale benchmark explicitly targeting the detection of hyper-realistic and semantically nuanced AI-generated videos. AEGIS comprises over 10,000 rigorously curated real and synthetic videos generated by diverse, state-of-the-art generative models, including Stable Video Diffusion, CogVideoX-5B, KLing, and Sora, encompassing open-source and proprietary architectures. In particular, AEGIS features specially constructed challenging subsets enhanced with robustness evaluation. Furthermore, we provide multimodal annotations spanning Semantic-Authenticity Descriptions, Motion Features, and Low-level Visual Features, facilitating authenticity detection and supporting downstream tasks such as multimodal fusion and forgery localization. Extensive experiments using advanced vision-language models demonstrate limited detection capabilities on the most challenging subsets of AEGIS, highlighting the dataset's unique complexity and realism beyond the current generalization capabilities of existing models. In essence, AEGIS establishes an indispensable evaluation benchmark, fundamentally advancing research toward developing genuinely robust, reliable, broadly generalizable video authenticity detection methodologies capable of addressing real-world forgery threats. Our dataset is available on https://huggingface.co/datasets/Clarifiedfish/AEGIS.

Paperid: 5561, https://arxiv.org/pdf/2508.10635.pdf

Abstract:
Understanding environmental changes from aerial imagery is vital for climate resilience, urban planning, and ecosystem monitoring. Yet, current vision language models (VLMs) overlook causal signals from environmental sensors, rely on single-source captions prone to stylistic bias, and lack interactive scenario-based reasoning. We present ChatENV, the first interactive VLM that jointly reasons over satellite image pairs and real-world sensor data. Our framework: (i) creates a 177k-image dataset forming 152k temporal pairs across 62 land-use classes in 197 countries with rich sensor metadata (e.g., temperature, PM10, CO); (ii) annotates data using GPT- 4o and Gemini 2.0 for stylistic and semantic diversity; and (iii) fine-tunes Qwen-2.5-VL using efficient Low-Rank Adaptation (LoRA) adapters for chat purposes. ChatENV achieves strong performance in temporal and "what-if" reasoning (e.g., BERT-F1 0.903) and rivals or outperforms state-of-the-art temporal models, while supporting interactive scenario-based analysis. This positions ChatENV as a powerful tool for grounded, sensor-aware environmental monitoring.

Paperid: 5562, https://arxiv.org/pdf/2508.07626.pdf

Abstract:
Visual Robot Manipulation (VRM) aims to enable a robot to follow natural language instructions based on robot states and visual observations, and therefore requires costly multi-modal data. To compensate for the deficiency of robot data, existing approaches have employed vision-language pretraining with large-scale data. However, they either utilize web data that differs from robotic tasks, or train the model in an implicit way (e.g., predicting future frames at the pixel level), thus showing limited generalization ability under insufficient robot data. In this paper, we propose to learn from large-scale human action video datasets in an explicit way (i.e., imitating human actions from hand keypoints), introducing Visual Robot Manipulation with Analogical Reasoning (AR-VRM). To acquire action knowledge explicitly from human action videos, we propose a keypoint Vision-Language Model (VLM) pretraining scheme, enabling the VLM to learn human action knowledge and directly predict human hand keypoints. During fine-tuning on robot data, to facilitate the robotic arm in imitating the action patterns of human motions, we first retrieve human action videos that perform similar manipulation tasks and have similar historical observations , and then learn the Analogical Reasoning (AR) map between human hand keypoints and robot components. Taking advantage of focusing on action keypoints instead of irrelevant visual cues, our method achieves leading performance on the CALVIN benchmark {and real-world experiments}. In few-shot scenarios, our AR-VRM outperforms previous methods by large margins , underscoring the effectiveness of explicitly imitating human actions under data scarcity.

Paperid: 5563, https://arxiv.org/pdf/2508.06220.pdf

Abstract:
Recent advances in Vision-Language Models (VLMs) have demonstrated impressive capabilities in perception and reasoning. However, the ability to perform causal inference -- a core aspect of human cognition -- remains underexplored, particularly in multimodal settings. In this study, we introduce InfoCausalQA, a novel benchmark designed to evaluate causal reasoning grounded in infographics that combine structured visual data with textual context. The benchmark comprises two tasks: Task 1 focuses on quantitative causal reasoning based on inferred numerical trends, while Task 2 targets semantic causal reasoning involving five types of causal relations: cause, effect, intervention, counterfactual, and temporal. We manually collected 494 infographic-text pairs from four public sources and used GPT-4o to generate 1,482 high-quality multiple-choice QA pairs. These questions were then carefully revised by humans to ensure they cannot be answered based on surface-level cues alone but instead require genuine visual grounding. Our experimental results reveal that current VLMs exhibit limited capability in computational reasoning and even more pronounced limitations in semantic causal reasoning. Their significantly lower performance compared to humans indicates a substantial gap in leveraging infographic-based information for causal inference. Through InfoCausalQA, we highlight the need for advancing the causal reasoning abilities of multimodal AI systems.

Paperid: 5564, https://arxiv.org/pdf/2508.06084.pdf

Abstract:
Vision-language models (VLMs) have achieved impressive performance on multimodal reasoning tasks such as visual question answering (VQA), but their inference cost remains a significant challenge due to the large number of vision tokens processed during the prefill stage. Existing pruning methods often rely on directly using the attention patterns or static text prompt guidance, failing to exploit the dynamic internal signals generated during inference. To address these issues, we propose AdaptInfer, a plug-and-play framework for adaptive vision token pruning in VLMs. First, we introduce a fine-grained, dynamic text-guided pruning mechanism that reuses layer-wise text-to-text attention maps to construct soft priors over text-token importance, allowing more informed scoring of vision tokens at each stage. Second, we perform an offline analysis of cross-modal attention shifts and identify consistent inflection locations in inference, which inspire us to propose a more principled and efficient pruning schedule. Our method is lightweight and plug-and-play, also generalizable across multi-modal tasks. Experimental results have verified the effectiveness of the proposed method. For example, it reduces CUDA latency by 61.3\% while maintaining an average accuracy of 92.9\% on vanilla LLaVA-1.5-7B. Under the same token budget, AdaptInfer surpasses SOTA in accuracy.

Paperid: 5565, https://arxiv.org/pdf/2508.04987.pdf

Abstract:
Unsupervised domain adaptation (UDA) enables models trained on a labeled source domain to handle new unlabeled domains. Recently, pre-trained vision-language models (VLMs) have demonstrated promising zero-shot performance by leveraging semantic information to facilitate target tasks. By aligning vision and text embeddings, VLMs have shown notable success in bridging domain gaps. However, inherent differences naturally exist between modalities, which is known as modality gap. Our findings reveal that direct UDA with the presence of modality gap only transfers modality-invariant knowledge, leading to suboptimal target performance. To address this limitation, we propose a unified modality separation framework that accommodates both modality-specific and modality-invariant components. During training, different modality components are disentangled from VLM features then handled separately in a unified manner. At test time, modality-adaptive ensemble weights are automatically determined to maximize the synergy of different components. To evaluate instance-level modality characteristics, we design a modality discrepancy metric to categorize samples into modality-invariant, modality-specific, and uncertain ones. The modality-invariant samples are exploited to facilitate cross-modal alignment, while uncertain ones are annotated to enhance model capabilities. Building upon prompt tuning techniques, our methods achieve up to 9% performance gain with 9 times of computational efficiencies. Extensive experiments and analysis across various backbones, baselines, datasets and adaptation settings demonstrate the efficacy of our design.

Paperid: 5566, https://arxiv.org/pdf/2508.04931.pdf

Abstract:
Traditional control and planning for robotic manipulation heavily rely on precise physical models and predefined action sequences. While effective in structured environments, such approaches often fail in real-world scenarios due to modeling inaccuracies and struggle to generalize to novel tasks. In contrast, humans intuitively interact with their surroundings, demonstrating remarkable adaptability, making efficient decisions through implicit physical understanding. In this work, we propose INTENTION, a novel framework enabling robots with learned interactive intuition and autonomous manipulation in diverse scenarios, by integrating Vision-Language Models (VLMs) based scene reasoning with interaction-driven memory. We introduce Memory Graph to record scenes from previous task interactions which embodies human-like understanding and decision-making about different tasks in real world. Meanwhile, we design an Intuitive Perceptor that extracts physical relations and affordances from visual scenes. Together, these components empower robots to infer appropriate interaction behaviors in new scenes without relying on repetitive instructions. Videos: https://robo-intention.github.io

Paperid: 5567, https://arxiv.org/pdf/2508.03967.pdf

Abstract:
In this paper, we introduce RAVID, the first framework for AI-generated image detection that leverages visual retrieval-augmented generation (RAG). While RAG methods have shown promise in mitigating factual inaccuracies in foundation models, they have primarily focused on text, leaving visual knowledge underexplored. Meanwhile, existing detection methods, which struggle with generalization and robustness, often rely on low-level artifacts and model-specific features, limiting their adaptability. To address this, RAVID dynamically retrieves relevant images to enhance detection. Our approach utilizes a fine-tuned CLIP image encoder, RAVID CLIP, enhanced with category-related prompts to improve representation learning. We further integrate a vision-language model (VLM) to fuse retrieved images with the query, enriching the input and improving accuracy. Given a query image, RAVID generates an embedding using RAVID CLIP, retrieves the most relevant images from a database, and combines these with the query image to form an enriched input for a VLM (e.g., Qwen-VL or Openflamingo). Experiments on the UniversalFakeDetect benchmark, which covers 19 generative models, show that RAVID achieves state-of-the-art performance with an average accuracy of 93.85%. RAVID also outperforms traditional methods in terms of robustness, maintaining high accuracy even under image degradations such as Gaussian blur and JPEG compression. Specifically, RAVID achieves an average accuracy of 80.27% under degradation conditions, compared to 63.44% for the state-of-the-art model C2P-CLIP, demonstrating consistent improvements in both Gaussian blur and JPEG compression scenarios. The code will be publicly available upon acceptance.

Paperid: 5568, https://arxiv.org/pdf/2508.02630.pdf

Abstract:
Online marketplaces will be transformed by autonomous AI agents acting on behalf of consumers. Rather than humans browsing and clicking, vision-language-model (VLM) agents can parse webpages, evaluate products, and transact. This raises a fundamental question: what do AI agents buy, and why? We develop ACES, a sandbox environment that pairs a platform-agnostic VLM agent with a fully programmable mock marketplace to study this question. We first conduct basic rationality checks in the context of simple tasks, and then, by randomizing product positions, prices, ratings, reviews, sponsored tags, and platform endorsements, we obtain causal estimates of how frontier VLMs actually shop. Models show strong but heterogeneous position effects: all favor the top row, yet different models prefer different columns, undermining the assumption of a universal "top" rank. They penalize sponsored tags and reward endorsements. Sensitivities to price, ratings, and reviews are directionally human-like but vary sharply in magnitude across models. Motivated by scenarios where sellers use AI agents to optimize product listings, we show that a seller-side agent that makes minor tweaks to product descriptions, targeting AI buyer preferences, can deliver substantial market-share gains if AI-mediated shopping dominates. We also find that modal product choices can differ across models and, in some cases, demand may concentrate on a few select products, raising competition questions. Together, our results illuminate how AI agents may behave in e-commerce settings and surface concrete seller strategy, platform design, and regulatory questions in an AI-mediated ecosystem.

Paperid: 5569, https://arxiv.org/pdf/2508.02405.pdf

Abstract:
The control of robots for manipulation tasks generally relies on visual input. Recent advances in vision-language models (VLMs) enable the use of natural language instructions to condition visual input and control robots in a wider range of environments. However, existing methods require a large amount of data to fine-tune VLMs for operating in unseen environments. In this paper, we present a framework that learns object-arrangement tasks from just a few demonstrations. We propose a two-stage framework that divides object-arrangement tasks into a target localization stage, for picking the object, and a region determination stage for placing the object. We present an instance-level semantic fusion module that aligns the instance-level image crops with the text embedding, enabling the model to identify the target objects defined by the natural language instructions. We validate our method on both simulation and real-world robotic environments. Our method, fine-tuned with a few demonstrations, improves generalization capability and demonstrates zero-shot ability in real-robot manipulation scenarios.

Paperid: 5570, https://arxiv.org/pdf/2508.00579.pdf

Abstract:
The multi-modal long-context document question-answering task aims to locate and integrate multi-modal evidences (such as texts, tables, charts, images, and layouts) distributed across multiple pages, for question understanding and answer generation. The existing methods can be categorized into Large Vision-Language Model (LVLM)-based and Retrieval-Augmented Generation (RAG)-based methods. However, the former were susceptible to hallucinations, while the latter struggled for inter-modal disconnection and cross-page fragmentation. To address these challenges, a novel multi-modal RAG model, named MMRAG-DocQA, was proposed, leveraging both textual and visual information across long-range pages to facilitate accurate question answering. A hierarchical indexing method with the integration of flattened in-page chunks and topological cross-page chunks was designed to jointly establish in-page multi-modal associations and long-distance cross-page dependencies. By means of joint similarity evaluation and large language model (LLM)-based re-ranking, a multi-granularity semantic retrieval method, including the page-level parent page retrieval and document-level summary retrieval, was proposed to foster multi-modal evidence connection and long-distance evidence integration and reasoning. Experimental results performed on public datasets, MMLongBench-Doc and LongDocURL, demonstrated the superiority of our MMRAG-DocQA method in understanding and answering modality-rich and multi-page documents.

Paperid: 5571, https://arxiv.org/pdf/2507.22805.pdf

Abstract:
Vision large language models (VLLMs) are focusing primarily on handling complex and fine-grained visual information by incorporating advanced vision encoders and scaling up visual models. However, these approaches face high training and inference costs, as well as challenges in extracting visual details, effectively bridging across modalities. In this work, we propose a novel visual framework, MoCHA, to address these issues. Our framework integrates four vision backbones (i.e., CLIP, SigLIP, DINOv2 and ConvNeXt) to extract complementary visual features and is equipped with a sparse Mixture of Experts Connectors (MoECs) module to dynamically select experts tailored to different visual dimensions. To mitigate redundant or insufficient use of the visual information encoded by the MoECs module, we further design a Hierarchical Group Attention (HGA) with intra- and inter-group operations and an adaptive gating strategy for encoded visual features. We train MoCHA on two mainstream LLMs (e.g., Phi2-2.7B and Vicuna-7B) and evaluate their performance across various benchmarks. Notably, MoCHA outperforms state-of-the-art open-weight models on various tasks. For example, compared to CuMo (Mistral-7B), our MoCHA (Phi2-2.7B) presents outstanding abilities to mitigate hallucination by showing improvements of 3.25% in POPE and to follow visual instructions by raising 153 points on MME. Finally, ablation studies further confirm the effectiveness and robustness of the proposed MoECs and HGA in improving the overall performance of MoCHA.

Paperid: 5572, https://arxiv.org/pdf/2507.19398.pdf

Abstract:
Chest radiography (CXR) plays a crucial role in the diagnosis of various diseases. However, the inherent class imbalance in the distribution of clinical findings presents a significant challenge for current self-supervised deep learning models. These models often fail to accurately classify long-tailed classes. Current Vision-Language models such as Contrastive Language Image Pre-training (CLIP) models effectively model the manifold distribution of the latent space, enabling high zero-shot classification accuracies. Although CLIP performs well on most of the primary classes in the dataset, our work reveals that its effectiveness decreases significantly for classes with a long-tailed distribution. Our approach employs a class-weighting mechanism that directly aligns with the distribution of classes within the latent space. This method ensures a substantial improvement in overall classification performance, with particular emphasis on enhancing the recognition and accuracy of rarely observed classes. We accomplish this by applying Gaussian Mixture Model (GMM) clustering to the latent space. The subsequent clusters are further refined by Student t-distribution, followed by a metric loss that utilizes the altered embeddings. Our approach facilitates stable and adaptive clustering of the features. This results in a notable average improvement of 7\% points in zero-shot AUC scores across 40 classes in the MIMIC-CXR-JPG dataset from previous SOTA models.

Paperid: 5573, https://arxiv.org/pdf/2507.17047.pdf

Abstract:
Video data, especially long-form video, is extremely dense and high-dimensional. Text-based summaries of video content offer a way to represent query-relevant content in a much more compact manner than raw video. In addition, textual representations are easily ingested by state-of-the-art large language models (LLMs), which enable reasoning over video content to answer complex natural language queries. To solve this issue, we rely on the progressive construction of a text-based memory by a video captioner operating on shorter chunks of the video, where spatio-temporal modeling is computationally feasible. We explore ways to improve the quality of the activity log comprised solely of short video captions. Because the video captions tend to be focused on human actions, and questions may pertain to other information in the scene, we seek to enrich the memory with static scene descriptions using Vision Language Models (VLMs). Our video understanding system relies on the LaViLa video captioner in combination with a LLM to answer questions about videos. We first explored different ways of partitioning the video into meaningful segments such that the textual descriptions more accurately reflect the structure of the video content. Furthermore, we incorporated static scene descriptions into the captioning pipeline using LLaVA VLM, resulting in a more detailed and complete caption log and expanding the space of questions that are answerable from the textual memory. Finally, we have successfully fine-tuned the LaViLa video captioner to produce both action and scene captions, significantly improving the efficiency of the captioning pipeline compared to using separate captioning models for the two tasks. Our model, controllable hybrid captioner, can alternate between different types of captions according to special input tokens that signals scene changes detected in the video.

Paperid: 5574, https://arxiv.org/pdf/2507.13089.pdf

Abstract:
Pre-trained vision-language models, such as CLIP, show impressive zero-shot recognition ability and can be easily transferred to specific downstream tasks via prompt tuning, even with limited training data. However, existing prompt tuning methods face two main challenges: (1) In few-shot scenarios, data scarcity often leads to overfitting, making the model sensitive to changes in the input domain. (2) To mitigate overfitting, these methods typically rely on complex task-specific model architectures and sensitive hyperparameter tuning, severely restricting their general applicability. To address these issues, we propose a simpler and more general framework called GLAD (Generalizable LoRA tuning with RegulArized GraDient). We show that merely applying LoRA achieves performance in downstream tasks comparable to current state-of-the-art prompt-based methods. While LoRA is effective and easy to use, it remains susceptible to overfitting in few-shot learning scenarios. To mitigate this risk, we introduce a gradient-based regularization technique. This technique effectively steers the optimization trajectory, encouraging the model to find a more stable parameter region that is robust to variations in data distribution. Through extensive experiments conducted on 15 benchmark datasets, we demonstrate that GLAD outperforms previous tuning approaches in terms of base-to-novel class generalization, image domain generalization, and cross-dataset generalization. The code will be publicly available.

Paperid: 5575, https://arxiv.org/pdf/2507.10066.pdf

Abstract:
This demonstration paper presents $\mathbf{LayLens}$, a tool aimed to make deepfake understanding easier for users of all educational backgrounds. While prior works often rely on outputs containing technical jargon, LayLens bridges the gap between model reasoning and human understanding through a three-stage pipeline: (1) explainable deepfake detection using a state-of-the-art forgery localization model, (2) natural language simplification of technical explanations using a vision-language model, and (3) visual reconstruction of a plausible original image via guided image editing. The interface presents both technical and layperson-friendly explanations in addition to a side-by-side comparison of the uploaded and reconstructed images. A user study with 15 participants shows that simplified explanations significantly improve clarity and reduce cognitive load, with most users expressing increased confidence in identifying deepfakes. LayLens offers a step toward transparent, trustworthy, and user-centric deepfake forensics.

Paperid: 5576, https://arxiv.org/pdf/2507.05673.pdf

Abstract:
Visual agent models for automating human activities on Graphical User Interfaces (GUIs) have emerged as a promising research direction, driven by advances in large Vision Language Models (VLMs). A critical challenge in GUI automation is the precise grounding of interface elements across diverse platforms. Existing vision-only GUI agents directly ground elements from large and cluttered screenshots, requiring them to process substantial irrelevant information that compromises their accuracy. In addition, these approaches typically employ basic cross-entropy loss for learning grounding objectives, which fails to effectively capture grounding quality compared to established object detection metrics like Intersection-over-Union (IoU). To address these issues, we introduce R-VLM, a novel GUI grounding approach that leverages zoomed-in region proposals for precise element localization. We also propose an IoU-aware objective function that facilitates model convergence toward high IoU predictions. Our approach bridges the gap between VLMs and conventional object detection techniques, improving the state-of-the-art grounding accuracy by 13% across diverse GUI platforms on the GUI grounding benchmarks ScreenSpot and AgentStudio. In addition, our R-VLM approach shows 3.2-9.7% absolute accuracy improvements in GUI navigation tasks on the AITW and Mind2Web benchmarks.

Paperid: 5577, https://arxiv.org/pdf/2507.03330.pdf

Abstract:
Cooking plays a vital role in everyday independence and well-being, yet remains challenging for people with vision impairments due to limited support for tracking progress and receiving contextual feedback. Object status - the condition or transformation of ingredients and tools - offers a promising but underexplored foundation for context-aware cooking support. In this paper, we present OSCAR (Object Status Context Awareness for Recipes), a technical pipeline that explores the use of object status recognition to enable recipe progress tracking in non-visual cooking. OSCAR integrates recipe parsing, object status extraction, visual alignment with cooking steps, and time-causal modeling to support real-time step tracking. We evaluate OSCAR on 173 instructional videos and a real-world dataset of 12 non-visual cooking sessions recorded by BLV individuals in their homes. Our results show that object status consistently improves step prediction accuracy across vision-language models, and reveal key factors that impact performance in real-world conditions, such as implicit tasks, camera placement, and lighting. We contribute the pipeline of context-aware recipe progress tracking, an annotated real-world non-visual cooking dataset, and design insights to guide future context-aware assistive cooking systems.

Paperid: 5578, https://arxiv.org/pdf/2506.18048.pdf

Abstract:
Small Vision Language Models (SVLMs) generally refer to models with parameter sizes less than or equal to 2B. Their low cost and power consumption characteristics confer high commercial value. However, their reasoning abilities are limited by the number of parameters. To address this issue, this paper proposes a post-training optimization paradigm called the Incremental Training Strategy to enhance the reasoning ability of SVLMs. Firstly, we constructed a Self-Supervised Chain-of-Thought (COT) Data Construction System, which leverages multiple LVLMs with 7B parameters or more to transform original data into COT data in a self-supervised manner. Our proposed Incremental Training Strategy consists of four stages. Stage 1 injects domain knowledge by performing Supervised Fine-Tuning (SFT) to the pretrained model on the COT data. Stage 2 aligns the COT data format by conducting a small amount of Group Relative Policy Optimization (GRPO) training constrained only by format rewards on the COT data. Stage 3 enhances reasoning ability by applying GRPO training on the COT data with constraints on both format and accuracy rewards. The resulting model shows significant improvement compared to the baseline. Stage 4 addresses the limited capacity of the SVLMs and the weak ability to capture complex patterns by proposing ClipLow GRPO (CLGRPO) to constrain the capture space of the training process. We conducted extensive comparative and ablation experiments on the abstract semantic recognition dataset EMOSet-118K. Experimental results demonstrate that our method significantly improves the reasoning ability of 1B SVLM. Compared to the baseline model fine-tuned on the original data, accuracy increased by 2.77 and recall by 0.69, achieving performance comparable to that of 8B models.

Paperid: 5579, https://arxiv.org/pdf/2506.15734.pdf

Abstract:
As Vision-Language Models (VLMs) demonstrate increasing capabilities across real-world applications such as code generation and chatbot assistance, ensuring their safety has become paramount. Unlike traditional Large Language Models (LLMs), VLMs face unique vulnerabilities due to their multimodal nature, allowing adversaries to modify visual or textual inputs to bypass safety guardrails and trigger the generation of harmful content. Through systematic analysis of VLM behavior under attack, we identify a novel phenomenon termed ``delayed safety awareness''. Specifically, we observe that safety-aligned VLMs may initially be compromised to produce harmful content, but eventually recognize the associated risks and attempt to self-correct. This pattern suggests that VLMs retain their underlying safety awareness but experience a temporal delay in their activation. Building on this insight, we hypothesize that VLMs' safety awareness can be proactively reactivated through carefully designed prompts. To this end, we introduce ``The Safety Reminder'', a soft prompt tuning approach that optimizes learnable prompt tokens, which are periodically injected during the text generation process to enhance safety awareness, effectively preventing harmful content generation. Additionally, our safety reminder only activates when harmful content is detected, leaving normal conversations unaffected and preserving the model's performance on benign tasks. Through comprehensive evaluation across three established safety benchmarks and one adversarial attacks, we demonstrate that our approach significantly reduces attack success rates while maintaining model utility, offering a practical solution for deploying safer VLMs in real-world applications.

Paperid: 5580, https://arxiv.org/pdf/2506.15298.pdf

Abstract:
Facial micro-expressions (MEs) are involuntary movements of the face that occur spontaneously when a person experiences an emotion but attempts to suppress or repress the facial expression, typically found in a high-stakes environment. In recent years, substantial advancements have been made in the areas of ME recognition, spotting, and generation. However, conventional approaches that treat spotting and recognition as separate tasks are suboptimal, particularly for analyzing long-duration videos in realistic settings. Concurrently, the emergence of multimodal large language models (MLLMs) and large vision-language models (LVLMs) offers promising new avenues for enhancing ME analysis through their powerful multimodal reasoning capabilities. The ME grand challenge (MEGC) 2025 introduces two tasks that reflect these evolving research directions: (1) ME spot-then-recognize (ME-STR), which integrates ME spotting and subsequent recognition in a unified sequential pipeline; and (2) ME visual question answering (ME-VQA), which explores ME understanding through visual question answering, leveraging MLLMs or LVLMs to address diverse question types related to MEs. All participating algorithms are required to run on this test set and submit their results on a leaderboard. More details are available at https://megc2025.github.io.

Paperid: 5581, https://arxiv.org/pdf/2506.13063.pdf

Abstract:
Recent pathology foundation models can provide rich tile-level representations but fall short of delivering general-purpose clinical utility without further extensive model development. These models lack whole-slide image (WSI) understanding and are not trained with large-scale diagnostic data, limiting their performance on diverse downstream tasks. We introduce PRISM2, a multi-modal slide-level foundation model trained via clinical dialogue to enable scalable, generalizable pathology AI. PRISM2 is trained on nearly 700,000 specimens (2.3 million WSIs) paired with real-world clinical diagnostic reports in a two-stage process. In Stage 1, a vision-language model is trained using contrastive and captioning objectives to align whole slide embeddings with textual clinical diagnosis. In Stage 2, the language model is unfrozen to enable diagnostic conversation and extract more clinically meaningful representations from hidden states. PRISM2 achieves strong performance on diagnostic and biomarker prediction tasks, outperforming prior slide-level models including PRISM and TITAN. It also introduces a zero-shot yes/no classification approach that surpasses CLIP-style methods without prompt tuning or class enumeration. By aligning visual features with clinical reasoning, PRISM2 improves generalization on both data-rich and low-sample tasks, offering a scalable path forward for building general pathology AI agents capable of assisting diagnostic and prognostic decisions.

Paperid: 5582, https://arxiv.org/pdf/2506.12822.pdf

Abstract:
Designing effective reward functions remains a fundamental challenge in reinforcement learning (RL), as it often requires extensive human effort and domain expertise. While RL from human feedback has been successful in aligning agents with human intent, acquiring high-quality feedback is costly and labor-intensive, limiting its scalability. Recent advancements in foundation models present a promising alternative--leveraging AI-generated feedback to reduce reliance on human supervision in reward learning. Building on this paradigm, we introduce ERL-VLM, an enhanced rating-based RL method that effectively learns reward functions from AI feedback. Unlike prior methods that rely on pairwise comparisons, ERL-VLM queries large vision-language models (VLMs) for absolute ratings of individual trajectories, enabling more expressive feedback and improved sample efficiency. Additionally, we propose key enhancements to rating-based RL, addressing instability issues caused by data imbalance and noisy labels. Through extensive experiments across both low-level and high-level control tasks, we demonstrate that ERL-VLM significantly outperforms existing VLM-based reward generation methods. Our results demonstrate the potential of AI feedback for scaling RL with minimal human intervention, paving the way for more autonomous and efficient reward learning.

Paperid: 5583, https://arxiv.org/pdf/2506.12009.pdf

Abstract:
Affordance grounding-localizing object regions based on natural language descriptions of interactions-is a critical challenge for enabling intelligent agents to understand and interact with their environments. However, this task remains challenging due to the need for fine-grained part-level localization, the ambiguity arising from multiple valid interaction regions, and the scarcity of large-scale datasets. In this work, we introduce Affogato, a large-scale benchmark comprising 150K instances, annotated with open-vocabulary text descriptions and corresponding 3D affordance heatmaps across a diverse set of objects and interactions. Building on this benchmark, we develop simple yet effective vision-language models that leverage pretrained part-aware vision backbones and a text-conditional heatmap decoder. Our models trained with the Affogato dataset achieve promising performance on the existing 2D and 3D benchmarks, and notably, exhibit effectiveness in open-vocabulary cross-domain generalization. The Affogato dataset is shared in public: https://huggingface.co/datasets/project-affogato/affogato

Paperid: 5584, https://arxiv.org/pdf/2506.09691.pdf

Abstract:
Dual encoder Vision-Language Models (VLM) such as CLIP are widely used for image-text retrieval tasks. However, those models struggle with compositionality, showing a bag-of-words-like behavior that limits their retrieval performance. Many different training approaches have been proposed to improve the vision-language compositionality capabilities of those models. In comparison, inference-time techniques have received little attention. In this paper, we propose to add simple structure at inference, where, given an image and a caption: i) we divide the image into different smaller crops, ii) we extract text segments, capturing objects, attributes and relations, iii) using a VLM, we find the image crops that better align with text segments obtaining matches, and iv) we compute the final image-text similarity aggregating the individual similarities of the matches. Based on various popular dual encoder VLMs, we evaluate our approach in controlled and natural datasets for VL compositionality. We find that our approach consistently improves the performance of evaluated VLMs without any training, which shows the potential of inference-time techniques. The results are especially good for attribute-object binding as shown in the controlled dataset. As a result of an extensive analysis: i) we show that processing image crops is actually essential for the observed gains in performance, and ii) we identify specific areas to further improve inference-time approaches.

Paperid: 5585, https://arxiv.org/pdf/2506.05982.pdf

Abstract:
As automated attack techniques rapidly advance, CAPTCHAs remain a critical defense mechanism against malicious bots. However, existing CAPTCHA schemes encompass a diverse range of modalities -- from static distorted text and obfuscated images to interactive clicks, sliding puzzles, and logic-based questions -- yet the community still lacks a unified, large-scale, multimodal benchmark to rigorously evaluate their security robustness. To address this gap, we introduce MCA-Bench, a comprehensive and reproducible benchmarking suite that integrates heterogeneous CAPTCHA types into a single evaluation protocol. Leveraging a shared vision-language model backbone, we fine-tune specialized cracking agents for each CAPTCHA category, enabling consistent, cross-modal assessments. Extensive experiments reveal that MCA-Bench effectively maps the vulnerability spectrum of modern CAPTCHA designs under varied attack settings, and crucially offers the first quantitative analysis of how challenge complexity, interaction depth, and model solvability interrelate. Based on these findings, we propose three actionable design principles and identify key open challenges, laying the groundwork for systematic CAPTCHA hardening, fair benchmarking, and broader community collaboration. Datasets and code are available online.

Paperid: 5586, https://arxiv.org/pdf/2506.01783.pdf

Abstract:
Face Anti-Spoofing (FAS) typically depends on a single visual modality when defending against presentation attacks such as print attacks, screen replays, and 3D masks, resulting in limited generalization across devices, environments, and attack types. Meanwhile, Multimodal Large Language Models (MLLMs) have recently achieved breakthroughs in image-text understanding and semantic reasoning, suggesting that integrating visual and linguistic co-inference into FAS can substantially improve both robustness and interpretability. However, the lack of a high-quality vision-language multimodal dataset has been a critical bottleneck. To address this, we introduce FaceCoT (Face Chain-of-Thought), the first large-scale Visual Question Answering (VQA) dataset tailored for FAS. FaceCoT covers 14 spoofing attack types and enriches model learning with high-quality CoT VQA annotations. Meanwhile, we develop a caption model refined via reinforcement learning to expand the dataset and enhance annotation quality. Furthermore, we introduce a CoT-Enhanced Progressive Learning (CEPL) strategy to better leverage the CoT data and boost model performance on FAS tasks. Extensive experiments demonstrate that models trained with FaceCoT and CEPL outperform state-of-the-art methods on multiple benchmark datasets.

Paperid: 5587, https://arxiv.org/pdf/2506.00455.pdf

Abstract:
Navigation by scent is a capability in robotic systems that is rising in demand. However, current methods often suffer from ambiguities, particularly when robots misattribute odours to incorrect objects due to limitations in olfactory datasets and sensor resolutions. To address challenges in olfactory navigation, we introduce a multimodal olfaction dataset along with a novel machine learning method using diffusion-based molecular generation that can be used by itself or with automated olfactory dataset construction pipelines. This generative process of our diffusion model expands the chemical space beyond the limitations of both current olfactory datasets and training methods, enabling the identification of potential odourant molecules not previously documented. The generated molecules can then be more accurately validated using advanced olfactory sensors, enabling them to detect more compounds and inform better hardware design. By integrating visual analysis, language processing, and molecular generation, our framework enhances the ability of olfaction-vision models on robots to accurately associate odours with their correct sources, thereby improving navigation and decision-making through better sensor selection for a target compound in critical applications such as explosives detection, narcotics screening, and search and rescue. Our methodology represents a foundational advancement in the field of artificial olfaction, offering a scalable solution to challenges posed by limited olfactory data and sensor ambiguities. Code, models, and data are made available to the community at: https://huggingface.co/datasets/kordelfrance/olfaction-vision-language-dataset.

Paperid: 5588, https://arxiv.org/pdf/2506.00238.pdf

Abstract:
Natural disasters usually affect vast areas and devastate infrastructures. Performing a timely and efficient response is crucial to minimize the impact on affected communities, and data-driven approaches are the best choice. Visual question answering (VQA) models help management teams to achieve in-depth understanding of damages. However, recently published models do not possess the ability to answer open-ended questions and only select the best answer among a predefined list of answers. If we want to ask questions with new additional possible answers that do not exist in the predefined list, the model needs to be fin-tuned/retrained on a new collected and annotated dataset, which is a time-consuming procedure. In recent years, large-scale Vision-Language Models (VLMs) have earned significant attention. These models are trained on extensive datasets and demonstrate strong performance on both unimodal and multimodal vision/language downstream tasks, often without the need for fine-tuning. In this paper, we propose a VLM-based zero-shot VQA (ZeShot-VQA) method, and investigate the performance of on post-disaster FloodNet dataset. Since the proposed method takes advantage of zero-shot learning, it can be applied on new datasets without fine-tuning. In addition, ZeShot-VQA is able to process and generate answers that has been not seen during the training procedure, which demonstrates its flexibility.

Paperid: 5589, https://arxiv.org/pdf/2505.24214.pdf

Abstract:
The advent of foundation models, particularly Vision-Language Models (VLMs) and Multi-modal Large Language Models (MLLMs), has redefined the frontiers of artificial intelligence, enabling remarkable generalization across diverse tasks with minimal or no supervision. Yet, their potential in biometric recognition and analysis remains relatively underexplored. In this work, we introduce a comprehensive benchmark that evaluates the zero-shot and few-shot performance of state-of-the-art publicly available VLMs and MLLMs across six biometric tasks spanning the face and iris modalities: face verification, soft biometric attribute prediction (gender and race), iris recognition, presentation attack detection (PAD), and face manipulation detection (morphs and deepfakes). A total of 41 VLMs were used in this evaluation. Experiments show that embeddings from these foundation models can be used for diverse biometric tasks with varying degrees of success. For example, in the case of face verification, a True Match Rate (TMR) of 96.77 percent was obtained at a False Match Rate (FMR) of 1 percent on the Labeled Face in the Wild (LFW) dataset, without any fine-tuning. In the case of iris recognition, the TMR at 1 percent FMR on the IITD-R-Full dataset was 97.55 percent without any fine-tuning. Further, we show that applying a simple classifier head to these embeddings can help perform DeepFake detection for faces, Presentation Attack Detection (PAD) for irides, and extract soft biometric attributes like gender and ethnicity from faces with reasonably high accuracy. This work reiterates the potential of pretrained models in achieving the long-term vision of Artificial General Intelligence.

Paperid: 5590, https://arxiv.org/pdf/2505.23990.pdf

Abstract:
To effectively engage in human society, the ability to adapt, filter information, and make informed decisions in ever-changing situations is critical. As robots and intelligent agents become more integrated into human life, there is a growing opportunity-and need-to offload the cognitive burden on humans to these systems, particularly in dynamic, information-rich scenarios. To fill this critical need, we present Multi-RAG, a multimodal retrieval-augmented generation system designed to provide adaptive assistance to humans in information-intensive circumstances. Our system aims to improve situational understanding and reduce cognitive load by integrating and reasoning over multi-source information streams, including video, audio, and text. As an enabling step toward long-term human-robot partnerships, Multi-RAG explores how multimodal information understanding can serve as a foundation for adaptive robotic assistance in dynamic, human-centered situations. To evaluate its capability in a realistic human-assistance proxy task, we benchmarked Multi-RAG on the MMBench-Video dataset, a challenging multimodal video understanding benchmark. Our system achieves superior performance compared to existing open-source video large language models (Video-LLMs) and large vision-language models (LVLMs), while utilizing fewer resources and less input data. The results demonstrate Multi- RAG's potential as a practical and efficient foundation for future human-robot adaptive assistance systems in dynamic, real-world contexts.

Paperid: 5591, https://arxiv.org/pdf/2505.23859.pdf

Abstract:
Multi-task model merging aims to consolidate knowledge from multiple fine-tuned task-specific experts into a unified model while minimizing performance degradation. Existing methods primarily approach this by minimizing differences between task-specific experts and the unified model, either from a parameter-level or a task-loss perspective. However, parameter-level methods exhibit a significant performance gap compared to the upper bound, while task-loss approaches entail costly secondary training procedures. In contrast, we observe that performance degradation closely correlates with feature drift, i.e., differences in feature representations of the same sample caused by model merging. Motivated by this observation, we propose Layer-wise Optimal Task Vector Merging (LOT Merging), a technique that explicitly minimizes feature drift between task-specific experts and the unified model in a layer-by-layer manner. LOT Merging can be formulated as a convex quadratic optimization problem, enabling us to analytically derive closed-form solutions for the parameters of linear and normalization layers. Consequently, LOT Merging achieves efficient model consolidation through basic matrix operations. Extensive experiments across vision and vision-language benchmarks demonstrate that LOT Merging significantly outperforms baseline methods, achieving improvements of up to 4.4% (ViT-B/32) over state-of-the-art approaches.

Paperid: 5592, https://arxiv.org/pdf/2505.23740.pdf

Abstract:
Image vectorization is a powerful technique that converts raster images into vector graphics, enabling enhanced flexibility and interactivity. However, popular image vectorization tools struggle with occluded regions, producing incomplete or fragmented shapes that hinder editability. While recent advancements have explored rule-based and data-driven layer-wise image vectorization, these methods face limitations in vectorization quality and flexibility. In this paper, we introduce LayerPeeler, a novel layer-wise image vectorization approach that addresses these challenges through a progressive simplification paradigm. The key to LayerPeeler's success lies in its autoregressive peeling strategy: by identifying and removing the topmost non-occluded layers while recovering underlying content, we generate vector graphics with complete paths and coherent layer structures. Our method leverages vision-language models to construct a layer graph that captures occlusion relationships among elements, enabling precise detection and description for non-occluded layers. These descriptive captions are used as editing instructions for a finetuned image diffusion model to remove the identified layers. To ensure accurate removal, we employ localized attention control that precisely guides the model to target regions while faithfully preserving the surrounding content. To support this, we contribute a large-scale dataset specifically designed for layer peeling tasks. Extensive quantitative and qualitative experiments demonstrate that LayerPeeler significantly outperforms existing techniques, producing vectorization results with superior path semantics, geometric regularity, and visual fidelity.

Paperid: 5593, https://arxiv.org/pdf/2505.21647.pdf

Abstract:
Massive-scale pretraining has made vision-language models increasingly popular for image-to-image and text-to-image retrieval across a broad collection of domains. However, these models do not perform well when used for challenging retrieval tasks, such as instance retrieval in very large-scale image collections. Recent work has shown that linear transformations of VLM features trained for instance retrieval can improve performance by emphasizing subspaces that relate to the domain of interest. In this paper, we explore a more extreme version of this specialization by learning to map a given query to a query-specific feature space transformation. Because this transformation is linear, it can be applied with minimal computational cost to millions of image embeddings, making it effective for large-scale retrieval or re-ranking. Results show that this method consistently outperforms state-of-the-art alternatives, including those that require many orders of magnitude more computation at query time.

Paperid: 5594, https://arxiv.org/pdf/2505.19911.pdf

Abstract:
Large Vision-Language Models (VLMs) have achieved remarkable success in understanding complex real-world scenarios and supporting data-driven decision-making processes. However, VLMs exhibit significant vulnerability against adversarial examples, either text or image, which can lead to various adversarial outcomes, e.g., jailbreaking, hijacking, and hallucination, etc. In this work, we empirically and theoretically demonstrate that VLMs are particularly susceptible to image-based adversarial examples, where imperceptible perturbations can precisely manipulate each output token. To this end, we propose a novel attack called Vision-language model Manipulation Attack (VMA), which integrates first-order and second-order momentum optimization techniques with a differentiable transformation mechanism to effectively optimize the adversarial perturbation. Notably, VMA can be a double-edged sword: it can be leveraged to implement various attacks, such as jailbreaking, hijacking, privacy breaches, Denial-of-Service, and the generation of sponge examples, etc, while simultaneously enabling the injection of watermarks for copyright protection. Extensive empirical evaluations substantiate the efficacy and generalizability of VMA across diverse scenarios and datasets.

Paperid: 5595, https://arxiv.org/pdf/2505.18757.pdf

Abstract:
The representation of visual inputs of large vision-language models (LVLMs) usually involves substantially more tokens than that of textual inputs, leading to significant computational overhead. Several recent studies strive to mitigate this issue by either conducting token compression to prune redundant visual tokens or guiding them to bypass certain computational stages. While most existing work exploits token importance as the redundancy indicator, our study reveals that two largely neglected factors, namely, the diversity of retained visual tokens and their task relevance, often offer more robust criteria in token pruning. To this end, we design ToDRE, a two-stage and training-free token compression framework that achieves superior performance by pruning Tokens based on token Diversity and token-task RElevance. Instead of pruning redundant tokens, ToDRE introduces a greedy k-center algorithm to select and retain a small subset of diverse visual tokens after the vision encoder. Additionally, ToDRE addresses the "information migration" by further eliminating task-irrelevant visual tokens within the decoder of large language model (LLM). Extensive experiments show that ToDRE effectively reduces 90% of visual tokens after vision encoder and adaptively prunes all visual tokens within certain LLM's decoder layers, leading to a 2.6x speed-up in total inference time while maintaining 95.1% of model performance and excellent compatibility with efficient attention operators.

Paperid: 5596, https://arxiv.org/pdf/2505.18570.pdf

Abstract:
Stock price prediction remains a complex and high-stakes task in financial analysis, traditionally addressed using statistical models or, more recently, language models. In this work, we introduce VISTA (Vision-Language Inference for Stock Time-series Analysis), a novel, training-free framework that leverages Vision-Language Models (VLMs) for multi-modal stock forecasting. VISTA prompts a VLM with both textual representations of historical stock prices and their corresponding line charts to predict future price values. By combining numerical and visual modalities in a zero-shot setting and using carefully designed chain-of-thought prompts, VISTA captures complementary patterns that unimodal approaches often miss. We benchmark VISTA against standard baselines, including ARIMA and text-only LLM-based prompting methods. Experimental results show that VISTA outperforms these baselines by up to 89.83%, demonstrating the effectiveness of multi-modal inference for stock time-series analysis and highlighting the potential of VLMs in financial forecasting tasks without requiring task-specific training.

Paperid: 5597, https://arxiv.org/pdf/2505.17726.pdf

Abstract:
Recently, multimodal large language models (MLLMs) have emerged as a key approach in achieving artificial general intelligence. In particular, vision-language MLLMs have been developed to generate not only text but also visual outputs from multimodal inputs. This advancement requires efficient image tokens that LLMs can process effectively both in input and output. However, existing image tokenization methods for MLLMs typically capture only global abstract concepts or uniformly segmented image patches, restricting MLLMs' capability to effectively understand or generate detailed visual content, particularly at the object level. To address this limitation, we propose an object-centric visual tokenizer based on Slot Attention specifically for MLLMs. In particular, based on the Q-Former encoder, diffusion decoder, and residual vector quantization, our proposed discretized slot tokens can encode local visual details while maintaining high-level semantics, and also align with textual data to be integrated seamlessly within a unified next-token prediction framework of LLMs. The resulting Slot-MLLM demonstrates significant performance improvements over baselines with previous visual tokenizers across various vision-language tasks that entail local detailed comprehension and generation. Notably, this work is the first demonstration of the feasibility of object-centric slot attention performed with MLLMs and in-the-wild natural images.

Paperid: 5598, https://arxiv.org/pdf/2505.15304.pdf

Abstract:
Deep neural network (DNN)-based policy models, such as vision-language-action (VLA) models, excel at automating complex decision-making from multi-modal inputs. However, scaling these models greatly increases computational overhead, complicating deployment in resource-constrained settings like robot manipulation and autonomous driving. To address this, we propose Saliency-Aware Quantized Imitation Learning (SQIL), which combines quantization-aware training with a selective loss-weighting strategy for mission-critical states. By identifying these states via saliency scores and emphasizing them in the training loss, SQIL preserves decision fidelity under low-bit precision. We validate SQIL's generalization capability across extensive simulation benchmarks with environment variations, real-world tasks, and cross-domain tasks (self-driving, physics simulation), consistently recovering full-precision performance. Notably, a 4-bit weight-quantized VLA model for robotic manipulation achieves up to 2.5x speedup and 2.5x energy savings on an edge GPU with minimal accuracy loss. These results underline SQIL's potential for efficiently deploying large IL-based policy models on resource-limited devices.

Paperid: 5599, https://arxiv.org/pdf/2505.14381.pdf

Abstract:
With the increasing adoption of Large Language Models (LLMs) and Vision-Language Models (VLMs), rich document analysis technologies for applications like Retrieval-Augmented Generation (RAG) and visual RAG are gaining significant attention. Recent research indicates that using VLMs can achieve better RAG performance, but processing rich documents still remains a challenge since a single page contains large amounts of information. In this paper, we present SCAN (\textbf{S}emanti\textbf{C} Document Layout \textbf{AN}alysis), a novel approach enhancing both textual and visual Retrieval-Augmented Generation (RAG) systems working with visually rich documents. It is a VLM-friendly approach that identifies document components with appropriate semantic granularity, balancing context preservation with processing efficiency. SCAN uses a coarse-grained semantic approach that divides documents into coherent regions covering continuous components. We trained the SCAN model by fine-tuning object detection models with sophisticated annotation datasets. Our experimental results across English and Japanese datasets demonstrate that applying SCAN improves end-to-end textual RAG performance by up to 9.0\% and visual RAG performance by up to 6.4\%, outperforming conventional approaches and even commercial document processing solutions.

Paperid: 5600, https://arxiv.org/pdf/2505.10764.pdf

Abstract:
Innovations in digital intelligence are transforming robotic surgery with more informed decision-making. Real-time awareness of surgical instrument presence and actions (e.g., cutting tissue) is essential for such systems. Yet, despite decades of research, most machine learning models for this task are trained on small datasets and still struggle to generalize. Recently, vision-Language Models (VLMs) have brought transformative advances in reasoning across visual and textual modalities. Their unprecedented generalization capabilities suggest great potential for advancing intelligent robotic surgery. However, surgical VLMs remain under-explored, and existing models show limited performance, highlighting the need for benchmark studies to assess their capabilities and limitations and to inform future development. To this end, we benchmark the zero-shot performance of several advanced VLMs on two public robotic-assisted laparoscopic datasets for instrument and action classification. Beyond standard evaluation, we integrate explainable AI to visualize VLM attention and uncover causal explanations behind their predictions. This provides a previously underexplored perspective in this field for evaluating the reliability of model predictions. We also propose several explainability analysis-based metrics to complement standard evaluations. Our analysis reveals that surgical VLMs, despite domain-specific training, often rely on weak contextual cues rather than clinically relevant visual evidence, highlighting the need for stronger visual and reasoning supervision in surgical applications.

Paperid: 5601, https://arxiv.org/pdf/2505.09498.pdf

Abstract:
In this paper, we introduce Flash-VL 2B, a novel approach to optimizing Vision-Language Models (VLMs) for real-time applications, targeting ultra-low latency and high throughput without sacrificing accuracy. Leveraging advanced architectural enhancements and efficient computational strategies, Flash-VL 2B is designed to maximize throughput by reducing processing time while maintaining competitive performance across multiple vision-language benchmarks. Our approach includes tailored architectural choices, token compression mechanisms, data curation, training schemes, and a novel image processing technique called implicit semantic stitching that effectively balances computational load and model performance. Through extensive evaluations on 11 standard VLM benchmarks, we demonstrate that Flash-VL 2B achieves state-of-the-art results in both speed and accuracy, making it a promising solution for deployment in resource-constrained environments and large-scale real-time applications.

Paperid: 5602, https://arxiv.org/pdf/2505.06977.pdf

Abstract:
Multi-task model merging offers a promising paradigm for integrating multiple expert models into a unified model without additional training. Existing state-of-the-art techniques, such as Task Arithmetic and its variants, merge models by accumulating task vectors -- the parameter differences between pretrained and finetuned models. However, task vector accumulation is often hindered by knowledge conflicts, leading to performance degradation. To address this challenge, we propose Conflict-Aware Task Merging (CAT Merging), a novel training-free framework that selectively trims conflict-prone components from the task vectors. CAT Merging introduces several parameter-specific strategies, including projection for linear weights and masking for scaling and shifting parameters in normalization layers. Extensive experiments on vision, language, and vision-language tasks demonstrate that CAT Merging effectively suppresses knowledge conflicts, achieving average accuracy improvements of up to 2.5% (ViT-B/32) and 2.0% (ViT-L/14) over state-of-the-art methods.

Paperid: 5603, https://arxiv.org/pdf/2505.00134.pdf

Abstract:
Vision-language models (VLMs) have gained significant attention in computational pathology due to their multimodal learning capabilities that enhance big-data analytics of giga-pixel whole slide image (WSI). However, their sensitivity to large-scale clinical data, task formulations, and prompt design remains an open question, particularly in terms of diagnostic accuracy. In this paper, we present a systematic investigation and analysis of three state of the art VLMs for histopathology, namely Quilt-Net, Quilt-LLAVA, and CONCH, on an in-house digestive pathology dataset comprising 3,507 WSIs, each in giga-pixel form, across distinct tissue types. Through a structured ablative study on cancer invasiveness and dysplasia status, we develop a comprehensive prompt engineering framework that systematically varies domain specificity, anatomical precision, instructional framing, and output constraints. Our findings demonstrate that prompt engineering significantly impacts model performance, with the CONCH model achieving the highest accuracy when provided with precise anatomical references. Additionally, we identify the critical importance of anatomical context in histopathological image analysis, as performance consistently degraded when reducing anatomical precision. We also show that model complexity alone does not guarantee superior performance, as effective domain alignment and domain-specific training are critical. These results establish foundational guidelines for prompt engineering in computational pathology and highlight the potential of VLMs to enhance diagnostic accuracy when properly instructed with domain-appropriate prompts.

Paperid: 5604, https://arxiv.org/pdf/2504.18349.pdf

Abstract:
With the surge of large language models (LLMs), Large Vision-Language Models (VLMs)--which integrate vision encoders with LLMs for accurate visual grounding--have shown great potential in tasks like generalist agents and robotic control. However, VLMs are typically trained on massive web-scraped images, raising concerns over copyright infringement and privacy violations, and making data auditing increasingly urgent. Membership inference (MI), which determines whether a sample was used in training, has emerged as a key auditing technique, with promising results on open-source VLMs like LLaVA (AUC > 80%). In this work, we revisit these advances and uncover a critical issue: current MI benchmarks suffer from distribution shifts between member and non-member images, introducing shortcut cues that inflate MI performance. We further analyze the nature of these shifts and propose a principled metric based on optimal transport to quantify the distribution discrepancy. To evaluate MI in realistic settings, we construct new benchmarks with i.i.d. member and non-member images. Existing MI methods fail under these unbiased conditions, performing only marginally better than chance. Further, we explore the theoretical upper bound of MI by probing the Bayes Optimality within the VLM's embedding space and find the irreducible error rate remains high. Despite this pessimistic outlook, we analyze why MI for VLMs is particularly challenging and identify three practical scenarios--fine-tuning, access to ground-truth texts, and set-based inference--where auditing becomes feasible. Our study presents a systematic view of the limits and opportunities of MI for VLMs, providing guidance for future efforts in trustworthy data auditing.

Paperid: 5605, https://arxiv.org/pdf/2504.17902.pdf

Abstract:
Social media memes are a challenging domain for hate detection because they intertwine visual and textual cues into culturally nuanced messages. We introduce a novel framework, CAMU, which leverages large vision-language models to generate more descriptive captions, a caption-scoring neural network to emphasise hate-relevant content, and parameter-efficient fine-tuning of CLIP's text encoder for an improved multimodal understanding of memes. Experiments on publicly available hateful meme datasets show that simple projection layer fine-tuning yields modest gains, whereas selectively tuning deeper text encoder layers significantly boosts performance on all evaluation metrics. Moreover, our approach attains high accuracy (0.807) and F1-score (0.806) on the Hateful Memes dataset, at par with the existing SoTA framework while being much more efficient, offering practical advantages in real-world scenarios that rely on fixed decision thresholds. CAMU also achieves the best F1-score of 0.673 on the MultiOFF dataset for offensive meme identification, demonstrating its generalisability. Additional analyses on benign confounders reveal that robust visual grounding and nuanced text representations are crucial for reliable hate and offence detection. We will publicly release CAMU along with the resultant models for further research. Disclaimer: This paper includes references to potentially disturbing, hateful, or offensive content due to the nature of the task.

Paperid: 5606, https://arxiv.org/pdf/2504.11695.pdf

Abstract:
Vision-language models encode images and text in a joint space, minimizing the distance between corresponding image and text pairs. How are language and images organized in this joint space, and how do the models encode meaning and modality? To investigate this, we train and release sparse autoencoders (SAEs) on the embedding spaces of four vision-language models (CLIP, SigLIP, SigLIP2, and AIMv2). SAEs approximate model embeddings as sparse linear combinations of learned directions, or "concepts". We find that, compared to other methods of linear feature learning, SAEs are better at reconstructing the real embeddings, while also able to retain the most sparsity. Retraining SAEs with different seeds or different data diet leads to two findings: the rare, specific concepts captured by the SAEs are liable to change drastically, but we also show that commonly-activating concepts are remarkably stable across runs. Interestingly, while most concepts activate primarily for one modality, we find they are not merely encoding modality per se. Many are almost orthogonal to the subspace that defines modality, and the concept directions do not function as good modality classifiers, suggesting that they encode cross-modal semantics. To quantify this bridging behavior, we introduce the Bridge Score, a metric that identifies concept pairs which are both co-activated across aligned image-text inputs and geometrically aligned in the shared space. This reveals that even single-modality concepts can collaborate to support cross-modal integration. We release interactive demos of the SAEs for all models, allowing researchers to explore the organization of the concept spaces. Overall, our findings uncover a sparse linear structure within VLM embedding spaces that is shaped by modality, yet stitched together through latent bridges, offering new insight into how multimodal meaning is constructed.

Paperid: 5607, https://arxiv.org/pdf/2504.10873.pdf

Abstract:
In autonomous driving, it is crucial to correctly interpret traffic gestures (TGs), such as those of an authority figure providing orders or instructions, or a pedestrian signaling the driver, to ensure a safe and pleasant traffic environment for all road users. This study investigates the capabilities of state-of-the-art vision-language models (VLMs) in zero-shot interpretation, focusing on their ability to caption and classify human gestures in traffic contexts. We create and publicly share two custom datasets with varying formal and informal TGs, such as 'Stop', 'Reverse', 'Hail', etc. The datasets are "Acted TG (ATG)" and "Instructive TG In-The-Wild (ITGI)". They are annotated with natural language, describing the pedestrian's body position and gesture. We evaluate models using three methods utilizing expert-generated captions as baseline and control: (1) caption similarity, (2) gesture classification, and (3) pose sequence reconstruction similarity. Results show that current VLMs struggle with gesture understanding: sentence similarity averages below 0.59, and classification F1 scores reach only 0.14-0.39, well below the expert baseline of 0.70. While pose reconstruction shows potential, it requires more data and refined metrics to be reliable. Our findings reveal that although some SOTA VLMs can interpret zero-shot human traffic gestures, none are accurate and robust enough to be trustworthy, emphasizing the need for further research in this domain.

Paperid: 5608, https://arxiv.org/pdf/2504.10852.pdf

Abstract:
Language-based foundation models, such as large language models (LLMs) or large vision-language models (LVLMs), have been widely studied in long-tailed recognition. However, the need for linguistic data is not applicable to all practical tasks. In this study, we aim to explore using large vision models (LVMs) or visual foundation models (VFMs) to enhance long-tailed data features without any language information. Specifically, we extract features from the LVM and fuse them with features in the baseline network's map and latent space to obtain the augmented features. Moreover, we design several prototype-based losses in the latent space to further exploit the potential of the augmented features. In the experimental section, we validate our approach on two benchmark datasets: ImageNet-LT and iNaturalist2018.

Paperid: 5609, https://arxiv.org/pdf/2504.05491.pdf

Abstract:
Integrating vision models into large language models (LLMs) has sparked significant interest in creating vision-language foundation models, especially for video understanding. Recent methods often utilize memory banks to handle untrimmed videos for video-level understanding. However, they typically compress visual memory using similarity-based greedy approaches, which can overlook the contextual importance of individual tokens. To address this, we introduce an efficient LLM adapter designed for video-level understanding of untrimmed videos that prioritizes the contextual relevance of spatio-temporal tokens. Our framework leverages scorer networks to selectively compress the visual memory bank and filter spatial tokens based on relevance, using a differentiable Top-K operator for end-to-end training. Across three key video-level understanding tasks$\unicode{x2013}$ untrimmed video classification, video question answering, and video captioning$\unicode{x2013}$our method achieves competitive or superior results on four large-scale datasets while reducing computational overhead by up to 34%. The code will be available soon on GitHub.

Paperid: 5610, https://arxiv.org/pdf/2504.05299.pdf

Abstract:
Large Vision-Language Models (VLMs) deliver exceptional performance but require significant computational resources, limiting their deployment on mobile and edge devices. Smaller VLMs typically mirror design choices of larger models, such as extensive image tokenization, leading to inefficient GPU memory usage and constrained practicality for on-device applications. We introduce SmolVLM, a series of compact multimodal models specifically engineered for resource-efficient inference. We systematically explore architectural configurations, tokenization strategies, and data curation optimized for low computational overhead. Through this, we identify key design choices that yield substantial performance gains on image and video tasks with minimal memory footprints. Our smallest model, SmolVLM-256M, uses less than 1GB GPU memory during inference and outperforms the 300-times larger Idefics-80B model, despite an 18-month development gap. Our largest model, at 2.2B parameters, rivals state-of-the-art VLMs consuming twice the GPU memory. SmolVLM models extend beyond static images, demonstrating robust video comprehension capabilities. Our results emphasize that strategic architectural optimizations, aggressive yet efficient tokenization, and carefully curated training data significantly enhance multimodal performance, facilitating practical, energy-efficient deployments at significantly smaller scales.

Paperid: 5611, https://arxiv.org/pdf/2504.02799.pdf

Abstract:
Large Vision-Language Models offer a new paradigm for AI-driven image understanding, enabling models to perform tasks without task-specific training. This flexibility holds particular promise across medicine, where expert-annotated data is scarce. Yet, VLMs' practical utility in intervention-focused domains--especially surgery, where decision-making is subjective and clinical scenarios are variable--remains uncertain. Here, we present a comprehensive analysis of 11 state-of-the-art VLMs across 17 key visual understanding tasks in surgical AI--from anatomy recognition to skill assessment--using 13 datasets spanning laparoscopic, robotic, and open procedures. In our experiments, VLMs demonstrate promising generalizability, at times outperforming supervised models when deployed outside their training setting. In-context learning, incorporating examples during testing, boosted performance up to three-fold, suggesting adaptability as a key strength. Still, tasks requiring spatial or temporal reasoning remained difficult. Beyond surgery, our findings offer insights into VLMs' potential for tackling complex and dynamic scenarios in clinical and broader real-world applications.

Paperid: 5612, https://arxiv.org/pdf/2504.01890.pdf

Abstract:
Video understanding has shown remarkable improvements in recent years, largely dependent on the availability of large scaled labeled datasets. Recent advancements in visual-language models, especially based on contrastive pretraining, have shown remarkable generalization in zero-shot tasks, helping to overcome this dependence on labeled datasets. Adaptations of such models for videos, typically involve modifying the architecture of vision-language models to cater to video data. However, this is not trivial, since such adaptations are mostly computationally intensive and struggle with temporal modeling. We present TP-CLIP, an adaptation of CLIP that leverages temporal visual prompting for temporal adaptation without modifying the core CLIP architecture. This preserves its generalization abilities. TP-CLIP efficiently integrates into the CLIP architecture, leveraging its pre-trained capabilities for video data. Extensive experiments across various datasets demonstrate its efficacy in zero-shot and few-shot learning, outperforming existing approaches with fewer parameters and computational efficiency. In particular, we use just 1/3 the GFLOPs and 1/28 the number of tuneable parameters in comparison to recent state-of-the-art and still outperform it by up to 15.8% depending on the task and dataset.

Paperid: 5613, https://arxiv.org/pdf/2504.00557.pdf

Abstract:
Visual token reduction lowers inference costs caused by extensive image features in large vision-language models (LVLMs). Unlike relevant studies that prune tokens in self-attention-only LVLMs, our work uniquely addresses cross-attention-based models, which achieve superior performance. We identify that the key-value (KV) cache size for image tokens in cross-attention layers significantly exceeds that of text tokens in self-attention layers, posing a major compute bottleneck. To mitigate this issue, we exploit the sparse nature in cross-attention maps to selectively prune redundant visual features. Our Trimmed Llama effectively reduces KV cache demands without requiring additional training. By benefiting from 50%-reduced visual features, our model can reduce inference latency and memory usage while achieving benchmark parity.

Paperid: 5614, https://arxiv.org/pdf/2504.00429.pdf

Abstract:
Adversarial attacks pose a critical security threat to real-world AI systems by injecting human-imperceptible perturbations into benign samples to induce misclassification in deep learning models. While existing detection methods, such as Bayesian uncertainty estimation and activation pattern analysis, have achieved progress through feature engineering, their reliance on handcrafted feature design and prior knowledge of attack patterns limits generalization capabilities and incurs high engineering costs. To address these limitations, this paper proposes a lightweight adversarial detection framework based on the large-scale pre-trained vision-language model CLIP. Departing from conventional adversarial feature characterization paradigms, we innovatively adopt an anomaly detection perspective. By jointly fine-tuning CLIP's dual visual-text encoders with trainable adapter networks and learnable prompts, we construct a compact representation space tailored for natural images. Notably, our detection architecture achieves substantial improvements in generalization capability across both known and unknown attack patterns compared to traditional methods, while significantly reducing training overhead. This study provides a novel technical pathway for establishing a parameter-efficient and attack-agnostic defense paradigm, markedly enhancing the robustness of vision systems against evolving adversarial threats.

Paperid: 5615, https://arxiv.org/pdf/2503.23959.pdf

Abstract:
Grounded Conversation Generation (GCG) is an emerging vision-language task that requires models to generate natural language responses seamlessly intertwined with corresponding object segmentation masks. Recent models, such as GLaMM and OMG-LLaVA, achieve pixel-level grounding but incur significant computational costs due to processing a large number of visual tokens. Existing token pruning methods, like FastV and PyramidDrop, fail to preserve the local visual features critical for accurate grounding, leading to substantial performance drops in GCG tasks. To address this, we propose Adaptive Local-Aware Token Pruning (ALTP), a simple yet effective framework that accelerates GCG models by prioritizing local object information. ALTP introduces two key components: (1) Detail Density Capture (DDC), which uses superpixel segmentation to retain tokens in object-centric regions, preserving fine-grained details, and (2) Dynamic Density Formation (DDF), which dynamically allocates tokens based on information density, ensuring higher retention in semantically rich areas. Extensive experiments on the GranDf dataset demonstrate that ALTP significantly outperforms existing token pruning methods, such as FastV and PyramidDrop, on both GLaMM and OMG-LLaVA models. Notably, when applied to GLaMM, ALTP achieves a 90% reduction in visual tokens with a 4.9% improvement in AP50 and a 5.0% improvement in Recall compared to PyramidDrop. Similarly, on OMG-LLaVA, ALTP improves AP by 2.1% and mIOU by 3.0% at a 90% token reduction compared with PDrop.

Paperid: 5616, https://arxiv.org/pdf/2503.22413.pdf

Abstract:
The growing trend of legal disputes over the unauthorized use of data in machine learning (ML) systems highlights the urgent need for reliable data-use auditing mechanisms to ensure accountability and transparency in ML. We present the first proactive, instance-level, data-use auditing method designed to enable data owners to audit the use of their individual data instances in ML models, providing more fine-grained auditing results than previous work. To do so, our research generalizes previous work integrating black-box membership inference and sequential hypothesis testing, expanding its scope of application while preserving the quantifiable and tunable false-detection rate that is its hallmark. We evaluate our method on three types of visual ML models: image classifiers, visual encoders, and vision-language models (Contrastive Language-Image Pretraining (CLIP) and Bootstrapping Language-Image Pretraining (BLIP) models). In addition, we apply our method to evaluate the performance of two state-of-the-art approximate unlearning methods. As a noteworthy second contribution, our work reveals that neither method successfully removes the influence of the unlearned data instances from image classifiers and CLIP models, even if sacrificing model utility by $10\%$.

Paperid: 5617, https://arxiv.org/pdf/2503.17871.pdf

Abstract:
Composed image retrieval (CIR) enables users to search images using a reference image combined with textual modifications. Recent advances in vision-language models have improved CIR, but dataset limitations remain a barrier. Existing datasets often rely on simplistic, ambiguous, or insufficient manual annotations, hindering fine-grained retrieval. We introduce good4cir, a structured pipeline leveraging vision-language models to generate high-quality synthetic annotations. Our method involves: (1) extracting fine-grained object descriptions from query images, (2) generating comparable descriptions for target images, and (3) synthesizing textual instructions capturing meaningful transformations between images. This reduces hallucination, enhances modification diversity, and ensures object-level consistency. Applying our method improves existing datasets and enables creating new datasets across diverse domains. Results demonstrate improved retrieval accuracy for CIR models trained on our pipeline-generated datasets. We release our dataset construction framework to support further research in CIR and multi-modal retrieval.

Paperid: 5618, https://arxiv.org/pdf/2503.10731.pdf

Abstract:
Zero-shot learning holds tremendous potential for histopathology image analysis by enabling models to generalize to unseen classes without extensive labeled data. Recent advancements in vision-language models (VLMs) have expanded the capabilities of ZSL, allowing models to perform tasks without task-specific fine-tuning. However, applying VLMs to histopathology presents considerable challenges due to the complexity of histopathological imagery and the nuanced nature of diagnostic tasks. In this paper, we propose a novel framework called Multi-Resolution Prompt-guided Hybrid Embedding (MR-PHE) to address these challenges in zero-shot histopathology image classification. MR-PHE leverages multiresolution patch extraction to mimic the diagnostic workflow of pathologists, capturing both fine-grained cellular details and broader tissue structures critical for accurate diagnosis. We introduce a hybrid embedding strategy that integrates global image embeddings with weighted patch embeddings, effectively combining local and global contextual information. Additionally, we develop a comprehensive prompt generation and selection framework, enriching class descriptions with domain-specific synonyms and clinically relevant features to enhance semantic understanding. A similarity-based patch weighting mechanism assigns attention-like weights to patches based on their relevance to class embeddings, emphasizing diagnostically important regions during classification. Our approach utilizes pretrained VLM, CONCH for ZSL without requiring domain-specific fine-tuning, offering scalability and reducing dependence on large annotated datasets. Experimental results demonstrate that MR-PHE not only significantly improves zero-shot classification performance on histopathology datasets but also often surpasses fully supervised models.

Paperid: 5619, https://arxiv.org/pdf/2503.09387.pdf

Abstract:
This paper introduces VideoScan, an efficient vision-language model (VLM) inference framework designed for real-time video interaction that effectively comprehends and retains streamed video inputs while delivering rapid and accurate responses. A longstanding challenge in video understanding--particularly for long-term or real-time applications--stems from the substantial computational overhead caused by the extensive length of visual tokens. To address this, VideoScan employs a single semantic carrier token to represent each frame, progressively reducing computational and memory overhead during its two-phase inference process: prefilling and decoding. The embedding of the semantic carrier token is derived from an optimized aggregation of frame-level visual features, ensuring compact yet semantically rich representations. Critically, the corresponding key-value pairs are trained to retain contextual semantics from prior frames, enabling efficient memory management without sacrificing temporal coherence. During inference, the visual tokens of each frame are processed only once during the prefilling phase and subsequently discarded in the decoding stage, eliminating redundant computations. This design ensures efficient VLM inference even under stringent real-time constraints. Comprehensive experiments on diverse offline and online benchmarks demonstrate that LLaVA-Video, supported by our method, achieves up to $\sim 5\times$ and $1.29\times$ speedups compared to its original version and previous efficient streaming video understanding approaches, respectively. Crucially, these improvements are attained while maintaining competitive performance and ensuring stable GPU memory consumption (consistently $\sim 18$GB, independent of video duration).

Paperid: 5620, https://arxiv.org/pdf/2503.07603.pdf

Abstract:
Pre-trained LLMs that are further trained with image data perform well on vision-language tasks. While adding images during a second training phase effectively unlocks this capability, it is unclear how much of a gain or loss this two-step pipeline gives over VLMs which integrate images earlier into the training process. To investigate this, we train models spanning various datasets, scales, image-text ratios, and amount of pre-training done before introducing vision tokens. We then fine-tune these models and evaluate their downstream performance on a suite of vision-language and text-only tasks. We find that pre-training with a mixture of image and text data allows models to perform better on vision-language tasks while maintaining strong performance on text-only evaluations. On an average of 6 diverse tasks, we find that for a 1B model, introducing visual tokens 80% of the way through pre-training results in a 2% average improvement over introducing visual tokens to a fully pre-trained model.

Paperid: 5621, https://arxiv.org/pdf/2503.06514.pdf

Abstract:
Vision-Language Models (VLMs) have recently shown promising advancements in sequential decision-making tasks through task-specific fine-tuning. However, common fine-tuning methods, such as Supervised Fine-Tuning (SFT) and Reinforcement Learning (RL) techniques like Proximal Policy Optimization (PPO), present notable limitations: SFT assumes Independent and Identically Distributed (IID) data, while PPO focuses on maximizing cumulative rewards. These limitations often restrict solution diversity and hinder generalization in multi-step reasoning tasks. To address these challenges, we introduce a novel framework, GFlowVLM, a framework that fine-tune VLMs using Generative Flow Networks (GFlowNets) to promote generation of diverse solutions for complex reasoning tasks. GFlowVLM models the environment as a non-Markovian decision process, allowing it to capture long-term dependencies essential for real-world applications. It takes observations and task descriptions as inputs to prompt chain-of-thought (CoT) reasoning which subsequently guides action selection. We use task based rewards to fine-tune VLM with GFlowNets. This approach enables VLMs to outperform prior fine-tuning methods, including SFT and RL. Empirical results demonstrate the effectiveness of GFlowVLM on complex tasks such as card games (NumberLine, BlackJack) and embodied planning tasks (ALFWorld), showing enhanced training efficiency, solution diversity, and stronger generalization capabilities across both in-distribution and out-of-distribution scenarios.

Paperid: 5622, https://arxiv.org/pdf/2503.06472.pdf

Abstract:
Chinese calligraphy, a UNESCO Heritage, remains computationally challenging due to visual ambiguity and cultural complexity. Existing AI systems fail to contextualize their intricate scripts, because of limited annotated data and poor visual-semantic alignment. We propose CalliReader, a vision-language model (VLM) that solves the Chinese Calligraphy Contextualization (CC$^2$) problem through three innovations: (1) character-wise slicing for precise character extraction and sorting, (2) CalliAlign for visual-text token compression and alignment, (3) embedding instruction tuning (e-IT) for improving alignment and addressing data scarcity. We also build CalliBench, the first benchmark for full-page calligraphic contextualization, addressing three critical issues in previous OCR and VQA approaches: fragmented context, shallow reasoning, and hallucination. Extensive experiments including user studies have been conducted to verify our CalliReader's \textbf{superiority to other state-of-the-art methods and even human professionals in page-level calligraphy recognition and interpretation}, achieving higher accuracy while reducing hallucination. Comparisons with reasoning models highlight the importance of accurate recognition as a prerequisite for reliable comprehension. Quantitative analyses validate CalliReader's efficiency; evaluations on document and real-world benchmarks confirm its robust generalization ability.

Paperid: 5623, https://arxiv.org/pdf/2503.05962.pdf

Abstract:
Following recipes while cooking is an important but difficult task for visually impaired individuals. We developed OSCAR (Object Status Context Awareness for Recipes), a novel approach that provides recipe progress tracking and context-aware feedback on the completion of cooking tasks through tracking object statuses. OSCAR leverages both Large-Language Models (LLMs) and Vision-Language Models (VLMs) to manipulate recipe steps, extract object status information, align visual frames with object status, and provide cooking progress tracking log. We evaluated OSCAR's recipe following functionality using 173 YouTube cooking videos and 12 real-world non-visual cooking videos to demonstrate OSCAR's capability to track cooking steps and provide contextual guidance. Our results highlight the effectiveness of using object status to improve performance compared to baseline by over 20% across different VLMs, and we present factors that impact prediction performance. Furthermore, we contribute a dataset of real-world non-visual cooking videos with step annotations as an evaluation benchmark.

Paperid: 5624, https://arxiv.org/pdf/2503.00046.pdf

Abstract:
Evaluating creativity is challenging, even for humans, not only because of its subjectivity but also because it involves complex cognitive processes. Inspired by work in marketing, we attempt to break down visual advertisement creativity into atypicality and originality. With fine-grained human annotations on these dimensions, we propose a suite of tasks specifically for such a subjective problem. We also evaluate the alignment between state-of-the-art (SoTA) vision language models (VLMs) and humans on our proposed benchmark, demonstrating both the promises and challenges of using VLMs for automatic creativity assessment.

Paperid: 5625, https://arxiv.org/pdf/2502.20034.pdf

Abstract:
Recently, Large Vision-Language Models (LVLMs) show remarkable performance across various domains. However, these models suffer from object hallucination. This study revisits the previous claim that the cause of such hallucinations lies in the limited representational capacity of the vision encoder. Our analysis implies that the capacity of the vision encoder is not necessarily a major limiting factor in detecting object hallucination. Based on this insight, we propose Fine-grained CLIPScore (F-CLIPScore), a simple yet effective evaluation metric that enhances object-level granularity by incorporating text embeddings at the noun level. Evaluations on the OHD-Caps benchmark show that F-CLIPScore significantly outperforms conventional CLIPScore in accuracy by a large margin of \textbf{39.6\%} without additional training. We further demonstrate that F-CLIPScore-based data filtering reduces object hallucination in LVLM (4.9\% in POPE).

Paperid: 5626, https://arxiv.org/pdf/2502.19038.pdf

Abstract:
The effectiveness of zero-shot classification in large vision-language models (VLMs), such as Contrastive Language-Image Pre-training (CLIP), depends on access to extensive, well-aligned text-image datasets. In this work, we introduce two complementary data sources, one generated by large language models (LLMs) to describe the stages of fungal growth and another comprising a diverse set of synthetic fungi images. These datasets are designed to enhance CLIPs zero-shot classification capabilities for fungi-related tasks. To ensure effective alignment between text and image data, we project them into CLIPs shared representation space, focusing on different fungal growth stages. We generate text using LLaMA3.2 to bridge modality gaps and synthetically create fungi images. Furthermore, we investigate knowledge transfer by comparing text outputs from different LLM techniques to refine classification across growth stages.

Paperid: 5627, https://arxiv.org/pdf/2502.13990.pdf

Abstract:
The complexity of scenes and variations in image quality result in significant variability in the performance of semantic segmentation methods of remote sensing imagery (RSI) in supervised real-world scenarios. This makes the evaluation of semantic segmentation quality in such scenarios an issue to be resolved. However, most of the existing evaluation metrics are developed based on expert-labeled object-level annotations, which are not applicable in such scenarios. To address this issue, we propose RS-SQA, an unsupervised quality assessment model for RSI semantic segmentation based on vision language model (VLM). This framework leverages a pre-trained RS VLM for semantic understanding and utilizes intermediate features from segmentation methods to extract implicit information about segmentation quality. Specifically, we introduce CLIP-RS, a large-scale pre-trained VLM trained with purified text to reduce textual noise and capture robust semantic information in the RS domain. Feature visualizations confirm that CLIP-RS can effectively differentiate between various levels of segmentation quality. Semantic features and low-level segmentation features are effectively integrated through a semantic-guided approach to enhance evaluation accuracy. To further support the development of RS semantic segmentation quality assessment, we present RS-SQED, a dedicated dataset sampled from four major RS semantic segmentation datasets and annotated with segmentation accuracy derived from the inference results of 8 representative segmentation methods. Experimental results on the established dataset demonstrate that RS-SQA significantly outperforms state-of-the-art quality assessment models. This provides essential support for predicting segmentation accuracy and high-quality semantic segmentation interpretation, offering substantial practical value.

Paperid: 5628, https://arxiv.org/pdf/2502.10620.pdf

Abstract:
Recent advancements in large language models (LLMs) have demonstrated extraordinary comprehension capabilities with remarkable breakthroughs on various vision-language tasks. However, the application of LLMs in generating reliable medical diagnostic reports remains in the early stages. Currently, medical LLMs typically feature a passive interaction model where doctors respond to patient queries with little or no involvement in analyzing medical images. In contrast, some ChatBots simply respond to predefined queries based on visual inputs, lacking interactive dialogue or consideration of medical history. As such, there is a gap between LLM-generated patient-ChatBot interactions and those occurring in actual patient-doctor consultations. To bridge this gap, we develop an LLM-based dialogue system, namely proactive multi-round vision-language interactions for computer-aided diagnosis (ProMRVL-CAD), to generate patient-friendly disease diagnostic reports. The proposed ProMRVL-CAD system allows proactive dialogue to provide patients with constant and reliable medical access via an integration of knowledge graph into a recommendation system. Specifically, we devise two generators: a Proactive Question Generator (Pro-Q Gen) to generate proactive questions that guide the diagnostic procedure and a Multi-Vision Patient-Text Diagnostic Report Generator (MVP-DR Gen) to produce high-quality diagnostic reports. Evaluating two real-world publicly available datasets, MIMIC-CXR and IU-Xray, our model has better quality in generating medical reports. We further demonstrate the performance of ProMRVL achieves robust under the scenarios with low image quality. Moreover, we have created a synthetic medical dialogue dataset that simulates proactive diagnostic interactions between patients and doctors, serving as a valuable resource for training LLM.

Paperid: 5629, https://arxiv.org/pdf/2502.07855.pdf

Abstract:
Vision Large Language Models (VLMs) combine visual understanding with natural language processing, enabling tasks like image captioning, visual question answering, and video analysis. While VLMs show impressive capabilities across domains such as autonomous vehicles, smart surveillance, and healthcare, their deployment on resource-constrained edge devices remains challenging due to processing power, memory, and energy limitations. This survey explores recent advancements in optimizing VLMs for edge environments, focusing on model compression techniques, including pruning, quantization, knowledge distillation, and specialized hardware solutions that enhance efficiency. We provide a detailed discussion of efficient training and fine-tuning methods, edge deployment challenges, and privacy considerations. Additionally, we discuss the diverse applications of lightweight VLMs across healthcare, environmental monitoring, and autonomous systems, illustrating their growing impact. By highlighting key design strategies, current challenges, and offering recommendations for future directions, this survey aims to inspire further research into the practical deployment of VLMs, ultimately making advanced AI accessible in resource-limited settings.

Paperid: 5630, https://arxiv.org/pdf/2502.04558.pdf

Abstract:
Vision-language-action (VLA) models hold promise as generalist robotics solutions by translating visual and linguistic inputs into robot actions, yet they lack reliability due to their black-box nature and sensitivity to environmental changes. In contrast, cognitive architectures (CA) excel in symbolic reasoning and state monitoring but are constrained by rigid predefined execution. This work bridges these approaches by probing OpenVLA's hidden layers to uncover symbolic representations of object properties, relations, and action states, enabling integration with a CA for enhanced interpretability and robustness. Through experiments on LIBERO-spatial pick-and-place tasks, we analyze the encoding of symbolic states across different layers of OpenVLA's Llama backbone. Our probing results show consistently high accuracies (> 0.90) for both object and action states across most layers, though contrary to our hypotheses, we did not observe the expected pattern of object states being encoded earlier than action states. We demonstrate an integrated DIARC-OpenVLA system that leverages these symbolic representations for real-time state monitoring, laying the foundation for more interpretable and reliable robotic manipulation.

Paperid: 5631, https://arxiv.org/pdf/2502.01616.pdf

Abstract:
Preference-based reinforcement learning (RL) offers a promising approach for aligning policies with human intent but is often constrained by the high cost of human feedback. In this work, we introduce PrefVLM, a framework that integrates Vision-Language Models (VLMs) with selective human feedback to significantly reduce annotation requirements while maintaining performance. Our method leverages VLMs to generate initial preference labels, which are then filtered to identify uncertain cases for targeted human annotation. Additionally, we adapt VLMs using a self-supervised inverse dynamics loss to improve alignment with evolving policies. Experiments on Meta-World manipulation tasks demonstrate that PrefVLM achieves comparable or superior success rates to state-of-the-art methods while using up to 2 x fewer human annotations. Furthermore, we show that adapted VLMs enable efficient knowledge transfer across tasks, further minimizing feedback needs. Our results highlight the potential of combining VLMs with selective human supervision to make preference-based RL more scalable and practical.

Paperid: 5632, https://arxiv.org/pdf/2502.00528.pdf

Abstract:
Vision-language models can connect the text description of an object to its specific location in an image through visual grounding. This has potential applications in enhanced radiology reporting. However, these models require large annotated image-text datasets, which are lacking for PET/CT. We developed an automated pipeline to generate weak labels linking PET/CT report descriptions to their image locations and used it to train a 3D vision-language visual grounding model. Our pipeline finds positive findings in PET/CT reports by identifying mentions of SUVmax and axial slice numbers. From 25,578 PET/CT exams, we extracted 11,356 sentence-label pairs. Using this data, we trained ConTEXTual Net 3D, which integrates text embeddings from a large language model with a 3D nnU-Net via token-level cross-attention. The model's performance was compared against LLMSeg, a 2.5D version of ConTEXTual Net, and two nuclear medicine physicians. The weak-labeling pipeline accurately identified lesion locations in 98% of cases (246/251), with 7.5% requiring boundary adjustments. ConTEXTual Net 3D achieved an F1 score of 0.80, outperforming LLMSeg (F1=0.22) and the 2.5D model (F1=0.53), though it underperformed both physicians (F1=0.94 and 0.91). The model achieved better performance on FDG (F1=0.78) and DCFPyL (F1=0.75) exams, while performance dropped on DOTATE (F1=0.58) and Fluciclovine (F1=0.66). The model performed consistently across lesion sizes but showed reduced accuracy on lesions with low uptake. Our novel weak labeling pipeline accurately produced an annotated dataset of PET/CT image-text pairs, facilitating the development of 3D visual grounding models. ConTEXTual Net 3D significantly outperformed other models but fell short of the performance of nuclear medicine physicians. Our study suggests that even larger datasets may be needed to close this performance gap.

Paperid: 5633, https://arxiv.org/pdf/2501.18623.pdf

Abstract:
Procedural materials, represented as functional node graphs, are ubiquitous in computer graphics for photorealistic material appearance design. They allow users to perform intuitive and precise editing to achieve desired visual appearances. However, creating a procedural material given an input image requires professional knowledge and significant effort. In this work, we leverage the ability to convert procedural materials into standard Python programs and fine-tune a large pre-trained vision-language model (VLM) to generate such programs from input images. To enable effective fine-tuning, we also contribute an open-source procedural material dataset and propose to perform program-level augmentation by prompting another pre-trained large language model (LLM). Through extensive evaluation, we show that our method outperforms previous methods on both synthetic and real-world examples.

Paperid: 5634, https://arxiv.org/pdf/2501.15068.pdf

Abstract:
Embodied manipulation is a fundamental ability in the realm of embodied artificial intelligence. Although current embodied manipulation models show certain generalizations in specific settings, they struggle in new environments and tasks due to the complexity and diversity of real-world scenarios. The traditional end-to-end data collection and training manner leads to significant data demands. Decomposing end-to-end tasks into atomic skills helps reduce data requirements and improves the task success rate. However, existing methods are limited by predefined skill sets that cannot be dynamically updated. To address the issue, we introduce a three-wheeled data-driven method to build an atomic skill library. We divide tasks into subtasks using the Vision-Language-Planning (VLP). Then, atomic skill definitions are formed by abstracting the subtasks. Finally, an atomic skill library is constructed via data collection and Vision-Language-Action (VLA) fine-tuning. As the atomic skill library expands dynamically with the three-wheel update strategy, the range of tasks it can cover grows naturally. In this way, our method shifts focus from end-to-end tasks to atomic skills, significantly reducing data costs while maintaining high performance and enabling efficient adaptation to new tasks. Extensive experiments in real-world settings demonstrate the effectiveness and efficiency of our approach.

Paperid: 5635, https://arxiv.org/pdf/2501.14189.pdf

Abstract:
Distributed Constraint Optimization Problems (DCOPs) offer a powerful framework for multi-agent coordination but often rely on labor-intensive, manual problem construction. To address this, we introduce VL-DCOPs, a framework that takes advantage of large multimodal foundation models (LFMs) to automatically generate constraints from both visual and linguistic instructions. We then introduce a spectrum of agent archetypes for solving VL-DCOPs: from a neuro-symbolic agent that delegates some of the algorithmic decisions to an LFM, to a fully neural agent that depends entirely on an LFM for coordination. We evaluate these agent archetypes using state-of-the-art LLMs (large language models) and VLMs (vision language models) on three novel VL-DCOP tasks and compare their respective advantages and drawbacks. Lastly, we discuss how this work extends to broader frontier challenges in the DCOP literature.

Paperid: 5636, https://arxiv.org/pdf/2501.09672.pdf

Abstract:
The proliferation of Vision-Language Models (VLMs) in the past several years calls for rigorous and comprehensive evaluation methods and benchmarks. This work analyzes existing VLM evaluation techniques, including automated metrics, AI-based assessments, and human evaluations across diverse tasks. We first introduce Robin - a novel suite of VLMs that we built by combining Large Language Models (LLMs) and Vision Encoders (VEs) at multiple scales, and use Robin to identify shortcomings of current evaluation approaches across scales. Next, to overcome the identified limitations, we introduce CHIRP - a new long form response benchmark we developed for more robust and complete VLM evaluation. We provide open access to the Robin training code, model suite, and CHIRP benchmark to promote reproducibility and advance VLM research.

Paperid: 5637, https://arxiv.org/pdf/2501.04336.pdf

Abstract:
Long-form video understanding with Large Vision Language Models is challenged by the need to analyze temporally dispersed yet spatially concentrated key moments within limited context windows. In this work, we introduce VideoMindPalace, a new framework inspired by the "Mind Palace", which organizes critical video moments into a topologically structured semantic graph. VideoMindPalace organizes key information through (i) hand-object tracking and interaction, (ii) clustered activity zones representing specific areas of recurring activities, and (iii) environment layout mapping, allowing natural language parsing by LLMs to provide grounded insights on spatio-temporal and 3D context. In addition, we propose the Video MindPalace Benchmark (VMB), to assess human-like reasoning, including spatial localization, temporal reasoning, and layout-aware sequential understanding. Evaluated on VMB and established video QA datasets, including EgoSchema, NExT-QA, IntentQA, and the Active Memories Benchmark, VideoMindPalace demonstrates notable gains in spatio-temporal coherence and human-aligned reasoning, advancing long-form video analysis capabilities in VLMs.

Paperid: 5638, https://arxiv.org/pdf/2501.02922.pdf

Abstract:
Multiple Instance Learning (MIL) methods allow for gigapixel Whole-Slide Image (WSI) analysis with only slide-level annotations. Interpretability is crucial for safely deploying such algorithms in high-stakes medical domains. Traditional MIL methods offer explanations by highlighting salient regions. However, such spatial heatmaps provide limited insights for end users. To address this, we propose a novel inherently interpretable WSI-classification approach that uses human-understandable pathology concepts to generate explanations. Our proposed Concept MIL model leverages recent advances in vision-language models to directly predict pathology concepts based on image features. The model's predictions are obtained through a linear combination of the concepts identified on the top-K patches of a WSI, enabling inherent explanations by tracing each concept's influence on the prediction. In contrast to traditional concept-based interpretable models, our approach eliminates the need for costly human annotations by leveraging the vision-language model. We validate our method on two widely used pathology datasets: Camelyon16 and PANDA. On both datasets, Concept MIL achieves AUC and accuracy scores over 0.9, putting it on par with state-of-the-art models. We further find that 87.1\% (Camelyon16) and 85.3\% (PANDA) of the top 20 patches fall within the tumor region. A user study shows that the concepts identified by our model align with the concepts used by pathologists, making it a promising strategy for human-interpretable WSI classification.

Paperid: 5639, https://arxiv.org/pdf/2412.20826.pdf

Abstract:
Internet-scaled datasets are a luxury for human-robot interaction (HRI) researchers, as collecting natural interaction data in the wild is time-consuming and logistically challenging. The problem is exacerbated by robots' different form factors and interaction modalities. Inspired by recent work on ethnomethodological and conversation analysis (EMCA) in the domain of HRI, we propose ReStory, a method that has the potential to augment existing in-the-wild human-robot interaction datasets leveraging Vision Language Models. While still requiring human supervision, ReStory is capable of synthesizing human-interpretable interaction scenarios in the form of storyboards. We hope our proposed approach provides HRI researchers and interaction designers with a new angle to utilizing their valuable and scarce data.

Paperid: 5640, https://arxiv.org/pdf/2412.20059.pdf

Abstract:
Visual impairment affects the ability of people to live a life like normal people. Such people face challenges in performing activities of daily living, such as reading, writing, traveling and participating in social gatherings. Many traditional approaches are available to help visually impaired people; however, these are limited in obtaining contextually rich environmental information necessary for independent living. In order to overcome this limitation, this paper introduces a novel wearable vision assistance system that has a hat-mounted camera connected to a Raspberry Pi 4 Model B (8GB RAM) with artificial intelligence (AI) technology to deliver real-time feedback to a user through a sound beep mechanism. The key features of this system include a user-friendly procedure for the recognition of new people or objects through a one-click process that allows users to add data on new individuals and objects for later detection, enhancing the accuracy of the recognition over time. The system provides detailed descriptions of objects in the user's environment using a large vision language model (LVLM). In addition, it incorporates a distance sensor that activates a beeping sound using a buzzer as soon as the user is about to collide with an object, helping to ensure safety while navigating their environment. A comprehensive evaluation is carried out to evaluate the proposed AI-based solution against traditional support techniques. Comparative analysis shows that the proposed solution with its innovative combination of hardware and AI (including LVLMs with IoT), is a significant advancement in assistive technology that aims to solve the major issues faced by the community of visually impaired people

Paperid: 5641, https://arxiv.org/pdf/2412.17799.pdf

Abstract:
With the recent Nobel Prize awarded for radical advances in protein discovery, foundation models (FMs) for exploring large combinatorial spaces promise to revolutionize many scientific fields. Artificial Life (ALife) has not yet integrated FMs, thus presenting a major opportunity for the field to alleviate the historical burden of relying chiefly on manual design and trial-and-error to discover the configurations of lifelike simulations. This paper presents, for the first time, a successful realization of this opportunity using vision-language FMs. The proposed approach, called Automated Search for Artificial Life (ASAL), (1) finds simulations that produce target phenomena, (2) discovers simulations that generate temporally open-ended novelty, and (3) illuminates an entire space of interestingly diverse simulations. Because of the generality of FMs, ASAL works effectively across a diverse range of ALife substrates including Boids, Particle Life, Game of Life, Lenia, and Neural Cellular Automata. A major result highlighting the potential of this technique is the discovery of previously unseen Lenia and Boids lifeforms, as well as cellular automata that are open-ended like Conway's Game of Life. Additionally, the use of FMs allows for the quantification of previously qualitative phenomena in a human-aligned way. This new paradigm promises to accelerate ALife research beyond what is possible through human ingenuity alone.

Paperid: 5642, https://arxiv.org/pdf/2412.16739.pdf

Abstract:
Transductive few-shot learning has recently triggered wide attention in computer vision. Yet, current methods introduce key hyper-parameters, which control the prediction statistics of the test batches, such as the level of class balance, affecting performances significantly. Such hyper-parameters are empirically grid-searched over validation data, and their configurations may vary substantially with the target dataset and pre-training model, making such empirical searches both sub-optimal and computationally intractable. In this work, we advocate and introduce the unrolling paradigm, also referred to as "learning to optimize", in the context of few-shot learning, thereby learning efficiently and effectively a set of optimized hyper-parameters. Specifically, we unroll a generalization of the ubiquitous Expectation-Maximization (EM) optimizer into a neural network architecture, mapping each of its iterates to a layer and learning a set of key hyper-parameters over validation data. Our unrolling approach covers various statistical feature distributions and pre-training paradigms, including recent foundational vision-language models and standard vision-only classifiers. We report comprehensive experiments, which cover a breadth of fine-grained downstream image classification tasks, showing significant gains brought by the proposed unrolled EM algorithm over iterative variants. The achieved improvements reach up to 10% and 7.5% on vision-only and vision-language benchmarks, respectively.

Paperid: 5643, https://arxiv.org/pdf/2412.16420.pdf

Abstract:
Flowcharts are typically presented as images, driving the trend of using vision-language models (VLMs) for end-to-end flowchart understanding. However, two key challenges arise: (i) Limited controllability--users have minimal influence over the downstream task, as they can only modify input images, while the training of VLMs is often out of reach for most researchers. (ii) Lack of explainability--it is difficult to trace VLM errors to specific causes, such as failures in visual encoding or reasoning. We propose TextFlow, addressing aforementioned issues with two stages: (i) Vision Textualizer--which generates textual representations from flowchart images; and (ii) Textual Reasoner--which performs question-answering based on the text representations. TextFlow offers three key advantages: (i) users can select the type of text representations (e.g., Graphviz, Mermaid, PlantUML), or further convert them into executable graph object to call tools, enhancing performance and controllability; (ii) it improves explainability by helping to attribute errors more clearly to visual or textual processing components; and (iii) it promotes the modularization of the solution, such as allowing advanced LLMs to be used in the Reasoner stage when VLMs underperform in end-to-end fashion. Experiments on the FlowVQA and FlowLearn benchmarks demonstrate TextFlow's state-of-the-art performance as well as its robustness. All code is publicly available.

Paperid: 5644, https://arxiv.org/pdf/2412.15277.pdf

Abstract:
Pre-trained Vision-Language (VL) models such as CLIP have demonstrated their excellent performance across numerous downstream tasks. A recent method, Context Optimization (CoOp), further improves the performance of VL models on downstream tasks by introducing prompt learning. CoOp optimizes a set of learnable vectors, aka prompt, and freezes the whole CLIP model. However, relying solely on CLIP loss to fine-tune prompts can lead to models that are prone to overfitting on downstream task. To address this issue, we propose a plug-in prompt-regularization method called PLPP (Prompt Learning with PerPlexity), which use perplexity loss to regularize prompt learning. PLPP designs a two-step operation to compute the perplexity for prompts: (a) calculating cosine similarity between the weight of the embedding layer and prompts to get labels, (b) introducing a language model (LM) head that requires no training behind text encoder to output word probability distribution. Meanwhile, we unveil that the essence of PLPP is inherently a form of self-distillation. To further prevent overfitting as well as to reduce the additional computation introduced by PLPP, we turn the hard label to soft label and choose top-$k$ values for calculating the perplexity loss. For accelerating model convergence, we introduce mutual self-distillation learning, that is perplexity and inverted perplexity loss. The experiments conducted on four classification tasks indicate that PLPP exhibits superior performance compared to existing methods.

Paperid: 5645, https://arxiv.org/pdf/2412.14446.pdf

Abstract:
Human drivers rely on commonsense reasoning to navigate diverse and dynamic real-world scenarios. Existing end-to-end (E2E) autonomous driving (AD) models are typically optimized to mimic driving patterns observed in data, without capturing the underlying reasoning processes. This limitation constrains their ability to handle challenging driving scenarios. To close this gap, we propose VLM-AD, a method that leverages vision-language models (VLMs) as teachers to enhance training by providing additional supervision that incorporates unstructured reasoning information and structured action labels. Such supervision enhances the model's ability to learn richer feature representations that capture the rationale behind driving patterns. Importantly, our method does not require a VLM during inference, making it practical for real-time deployment. When integrated with state-of-the-art methods, VLM-AD achieves significant improvements in planning accuracy and reduced collision rates on the nuScenes dataset. It further improves route completion and driving scores under closed-loop evaluation, demonstrating its effectiveness in long-horizon, interactive driving scenarios and its potential for safe and reliable real-world deployment.

Paperid: 5646, https://arxiv.org/pdf/2412.13364.pdf

Abstract:
Image to image matching has been well studied in the computer vision community. Previous studies mainly focus on training a deep metric learning model matching visual patterns between the query image and gallery images. In this study, we show that pure image-to-image matching suffers from false positives caused by matching to local visual patterns. To alleviate this issue, we propose to leverage recent advances in vision-language pretraining research. Specifically, we introduce additional image-text alignment losses into deep metric learning, which serve as constraints to the image-to-image matching loss. With additional alignments between the text (e.g., product title) and image pairs, the model can learn concepts from both modalities explicitly, which avoids matching low-level visual features. We progressively develop two variants, a 3-tower and a 4-tower model, where the latter takes one more short text query input. Through extensive experiments, we show that this change leads to a substantial improvement to the image to image matching problem. We further leveraged this model for multimodal search, which takes both image and reformulation text queries to improve search quality. Both offline and online experiments show strong improvements on the main metrics. Specifically, we see 4.95% relative improvement on image matching click through rate with the 3-tower model and 1.13% further improvement from the 4-tower model.

Paperid: 5647, https://arxiv.org/pdf/2412.09951.pdf

Abstract:
The emergence of general human knowledge and impressive logical reasoning capacity in rapidly progressed vision-language models (VLMs) have driven increasing interest in applying VLMs to high-level autonomous driving tasks, such as scene understanding and decision-making. However, an in-depth study on the relationship between knowledge proficiency, especially essential driving expertise, and closed-loop autonomous driving performance requires further exploration. In this paper, we investigate the effects of the depth and breadth of fundamental driving knowledge on closed-loop trajectory planning and introduce WiseAD, a specialized VLM tailored for end-to-end autonomous driving capable of driving reasoning, action justification, object recognition, risk analysis, driving suggestions, and trajectory planning across diverse scenarios. We employ joint training on driving knowledge and planning datasets, enabling the model to perform knowledge-aligned trajectory planning accordingly. Extensive experiments indicate that as the diversity of driving knowledge extends, critical accidents are notably reduced, contributing 11.9% and 12.4% improvements in the driving score and route completion on the Carla closed-loop evaluations, achieving state-of-the-art performance. Moreover, WiseAD also demonstrates remarkable performance in knowledge evaluations on both in-domain and out-of-domain datasets.

Paperid: 5648, https://arxiv.org/pdf/2412.09240.pdf

Abstract:
Segmentation models are typically constrained by the categories defined during training. To address this, researchers have explored two independent approaches: adapting Vision-Language Models (VLMs) and leveraging synthetic data. However, VLMs often struggle with granularity, failing to disentangle fine-grained concepts, while synthetic data-based methods remain limited by the scope of available datasets. This paper proposes enhancing segmentation accuracy across diverse domains by integrating Vision-Language reasoning with key strategies for Unsupervised Domain Adaptation (UDA). First, we improve the fine-grained segmentation capabilities of VLMs through multi-scale contextual data, robust text embeddings with prompt augmentation, and layer-wise fine-tuning in our proposed Foundational-Retaining Open Vocabulary Semantic Segmentation (FROVSS) framework. Next, we incorporate these enhancements into a UDA framework by employing distillation to stabilize training and cross-domain mixed sampling to boost adaptability without compromising generalization. The resulting UDA-FROVSS framework is the first UDA approach to effectively adapt across domains without requiring shared categories.

Paperid: 5649, https://arxiv.org/pdf/2412.01034.pdf

Abstract:
Deep neural network (DNN)-based policy models like vision-language-action (VLA) models are transformative in automating complex decision-making across applications by interpreting multi-modal data. However, scaling these models greatly increases computational costs, which presents challenges in fields like robot manipulation and autonomous driving that require quick, accurate responses. To address the need for deployment on resource-limited hardware, we propose a new quantization framework for IL-based policy models that fine-tunes parameters to enhance robustness against low-bit precision errors during training, thereby maintaining efficiency and reliability under constrained conditions. Our evaluations with representative robot manipulation for 4-bit weight-quantization on a real edge GPU demonstrate that our framework achieves up to 2.5x speedup and 2.5x energy savings while preserving accuracy. For 4-bit weight and activation quantized self-driving models, the framework achieves up to 3.7x speedup and 3.1x energy saving on a low-end GPU. These results highlight the practical potential of deploying IL-based policy models on resource-constrained devices.

Paperid: 5650, https://arxiv.org/pdf/2411.19103.pdf

Abstract:
In this paper, we introduce an open-source Korean-English vision-language model (VLM), VARCO-VISION. We incorporate a step-by-step training strategy that allows a model learn both linguistic and visual information while preserving the backbone model's knowledge. Our model demonstrates outstanding performance in diverse settings requiring bilingual image-text understanding and generation abilities compared to models of similar size. VARCO-VISION is also capable of grounding, referring, and OCR, expanding its usage and potential applications for real-world scenarios. In addition to the model, we release five Korean evaluation datasets, including four closed-set and one openset benchmarks. We anticipate that our milestone will broaden the opportunities for AI researchers aiming to train VLMs. VARCO-VISION is available at https://huggingface.co/NCSOFT/VARCO-VISION-14B.

Paperid: 5651, https://arxiv.org/pdf/2411.18101.pdf

Abstract:
Due to the large size and lack of fine-grained annotation, Whole Slide Images (WSIs) analysis is commonly approached as a Multiple Instance Learning (MIL) problem. However, previous studies only learn from training data, posing a stark contrast to how human clinicians teach each other and reason about histopathologic entities and factors. Here we present a novel knowledge concept-based MIL framework, named ConcepPath to fill this gap. Specifically, ConcepPath utilizes GPT-4 to induce reliable diseasespecific human expert concepts from medical literature, and incorporate them with a group of purely learnable concepts to extract complementary knowledge from training data. In ConcepPath, WSIs are aligned to these linguistic knowledge concepts by utilizing pathology vision-language model as the basic building component. In the application of lung cancer subtyping, breast cancer HER2 scoring, and gastric cancer immunotherapy-sensitive subtyping task, ConcepPath significantly outperformed previous SOTA methods which lack the guidance of human expert knowledge.

Paperid: 5652, https://arxiv.org/pdf/2411.16722.pdf

Abstract:
Vision-language models (VLMs) have demonstrated remarkable zero-shot performance across various classification tasks. Nonetheless, their reliance on hand-crafted text prompts for each task hinders efficient adaptation to new tasks. While prompt learning offers a promising solution, most studies focus on maximizing the utilization of given few-shot labeled datasets, often overlooking the potential of careful data selection strategies, which enable higher accuracy with fewer labeled data. This motivates us to study a budget-efficient active prompt learning framework. Specifically, we introduce a class-guided clustering that leverages the pre-trained image and text encoders of VLMs, thereby enabling our cluster-balanced acquisition function from the initial round of active learning. Furthermore, considering the substantial class-wise variance in confidence exhibited by VLMs, we propose a budget-saving selective querying based on adaptive class-wise thresholds. Extensive experiments in active learning scenarios across nine datasets demonstrate that our method outperforms existing baselines.

Paperid: 5653, https://arxiv.org/pdf/2411.15537.pdf

Abstract:
Machine Unlearning (MU) aims to selectively erase harmful behaviors from models while retaining the overall utility of the model. As a multi-task learning problem, MU involves balancing objectives related to forgetting specific concepts/data and preserving general performance. A naive integration of these forgetting and preserving objectives can lead to gradient conflicts and dominance, impeding MU algorithms from reaching optimal solutions. To address the gradient conflict and dominance issue, we reformulate MU as a two-player cooperative game, where the two players, namely, the forgetting player and the preservation player, contribute via their gradient proposals to maximize their overall gain and balance their contributions. To this end, inspired by the Nash bargaining theory, we derive a closed-form solution to guide the model toward the Pareto stationary point. Our formulation of MU guarantees an equilibrium solution, where any deviation from the final state would lead to a reduction in the overall objectives for both players, ensuring optimality in each objective. We evaluate our algorithm's effectiveness on a diverse set of tasks across image classification and image generation. Extensive experiments with ResNet, vision-language model CLIP, and text-to-image diffusion models demonstrate that our method outperforms state-of-the-art MU algorithms, achieving a better trade-off between forgetting and preserving. Our results also highlight improvements in forgetting precision, preservation of generalization, and robustness against adversarial attacks.

Paperid: 5654, https://arxiv.org/pdf/2411.13572.pdf

Abstract:
The Public Health Advocacy Dataset (PHAD) is a comprehensive collection of 5,730 videos related to tobacco products sourced from social media platforms like TikTok and YouTube. This dataset encompasses 4.3 million frames and includes detailed metadata such as user engagement metrics, video descriptions, and search keywords. This is the first dataset with these features providing a valuable resource for analyzing tobacco-related content and its impact. Our research employs a two-stage classification approach, incorporating a Vision-Language (VL) Encoder, demonstrating superior performance in accurately categorizing various types of tobacco products and usage scenarios. The analysis reveals significant user engagement trends, particularly with vaping and e-cigarette content, highlighting areas for targeted public health interventions. The PHAD addresses the need for multi-modal data in public health research, offering insights that can inform regulatory policies and public health strategies. This dataset is a crucial step towards understanding and mitigating the impact of tobacco usage, ensuring that public health efforts are more inclusive and effective.

Paperid: 5655, https://arxiv.org/pdf/2411.09052.pdf

Abstract:
Robotics tasks are highly compositional by nature. For example, to perform a high-level task like cleaning the table a robot must employ low-level capabilities of moving the effectors to the objects on the table, pick them up and then move them off the table one-by-one, while re-evaluating the consequently dynamic scenario in the process. Given that large vision language models (VLMs) have shown progress on many tasks that require high level, human-like reasoning, we ask the question: if the models are taught the requisite low-level capabilities, can they compose them in novel ways to achieve interesting high-level tasks like cleaning the table without having to be explicitly taught so? To this end, we present ClevrSkills - a benchmark suite for compositional reasoning in robotics. ClevrSkills is an environment suite developed on top of the ManiSkill2 simulator and an accompanying dataset. The dataset contains trajectories generated on a range of robotics tasks with language and visual annotations as well as multi-modal prompts as task specification. The suite includes a curriculum of tasks with three levels of compositional understanding, starting with simple tasks requiring basic motor skills. We benchmark multiple different VLM baselines on ClevrSkills and show that even after being pre-trained on large numbers of tasks, these models fail on compositional reasoning in robotics tasks.

Paperid: 5656, https://arxiv.org/pdf/2411.05195.pdf

Abstract:
Recent research has shown that CLIP models struggle with visual reasoning tasks that require grounding compositionality, understanding spatial relationships, or capturing fine-grained details. One natural hypothesis is that the CLIP vision encoder does not embed essential information for these tasks. However, we find that this is not always the case: The encoder gathers query-relevant visual information, while CLIP fails to extract it. In particular, we show that another branch of Vision-Language Models (VLMs), Generative Multimodal Large Language Models (MLLMs), achieve significantly higher accuracy than CLIP in many of these tasks using the same vision encoder and weights, indicating that these Generative MLLMs perceive more -- as they extract and utilize visual information more effectively. We conduct a series of controlled experiments and reveal that their success is attributed to multiple key design choices, including patch tokens, position embeddings, and prompt-based weighting. On the other hand, enhancing the training data alone or applying a stronger text encoder does not suffice to solve the task, and additional text tokens offer little benefit. Interestingly, we find that fine-grained visual reasoning is not exclusive to generative models trained by an autoregressive loss: When converted into CLIP-like encoders by contrastive finetuning, these MLLMs still outperform CLIP under the same cosine similarity-based evaluation protocol. Our study highlights the importance of VLM architectural choices and suggests directions for improving the performance of CLIP-like contrastive VLMs.

Paperid: 5657, https://arxiv.org/pdf/2411.04999.pdf

Abstract:
Significant progress has been made in open-vocabulary mobile manipulation, where the goal is for a robot to perform tasks in any environment given a natural language description. However, most current systems assume a static environment, which limits the system's applicability in real-world scenarios where environments frequently change due to human intervention or the robot's own actions. In this work, we present DynaMem, a new approach to open-world mobile manipulation that uses a dynamic spatio-semantic memory to represent a robot's environment. DynaMem constructs a 3D data structure to maintain a dynamic memory of point clouds, and answers open-vocabulary object localization queries using multimodal LLMs or open-vocabulary features generated by state-of-the-art vision-language models. Powered by DynaMem, our robots can explore novel environments, search for objects not found in memory, and continuously update the memory as objects move, appear, or disappear in the scene. We run extensive experiments on the Stretch SE3 robots in three real and nine offline scenes, and achieve an average pick-and-drop success rate of 70% on non-stationary objects, which is more than a 2x improvement over state-of-the-art static systems. Our code as well as our experiment and deployment videos are open sourced and can be found on our project website: https://dynamem.github.io/

Paperid: 5658, https://arxiv.org/pdf/2411.03697.pdf

Abstract:
Modern transformer-based deep neural networks present unique technical challenges for effective acceleration in real-world applications. Apart from the vast amount of linear operations needed due to their sizes, modern transformer models are increasingly reliance on precise non-linear computations that make traditional low-bitwidth quantization methods and fixed-dataflow matrix accelerators ineffective for end-to-end acceleration. To address this need to accelerate both linear and non-linear operations in a unified and programmable framework, this paper introduces TATAA. TATAA employs 8-bit integer (int8) arithmetic for quantized linear layer operations through post-training quantization, while it relies on bfloat16 floating-point arithmetic to approximate non-linear layers of a transformer model. TATAA hardware features a transformable arithmetic architecture that supports both formats during runtime with minimal overhead, enabling it to switch between a systolic array mode for int8 matrix multiplications and a SIMD mode for vectorized bfloat16 operations. An end-to-end compiler is presented to enable flexible mapping from emerging transformer models to the proposed hardware. Experimental results indicate that our mixed-precision design incurs only 0.14% to 1.16% accuracy drop when compared with the pre-trained single-precision transformer models across a range of vision, language, and generative text applications. Our prototype implementation on the Alveo U280 FPGA currently achieves 2935.2 GOPS throughput on linear layers and a maximum of 189.5 GFLOPS for non-linear operations, outperforming related works by up to 1.45x in end-to-end throughput and 2.29x in DSP efficiency, while achieving 2.19x higher power efficiency than modern NVIDIA RTX4090 GPU.

Paperid: 5659, https://arxiv.org/pdf/2411.02537.pdf

Abstract:
We introduce INQUIRE, a text-to-image retrieval benchmark designed to challenge multimodal vision-language models on expert-level queries. INQUIRE includes iNaturalist 2024 (iNat24), a new dataset of five million natural world images, along with 250 expert-level retrieval queries. These queries are paired with all relevant images comprehensively labeled within iNat24, comprising 33,000 total matches. Queries span categories such as species identification, context, behavior, and appearance, emphasizing tasks that require nuanced image understanding and domain expertise. Our benchmark evaluates two core retrieval tasks: (1) INQUIRE-Fullrank, a full dataset ranking task, and (2) INQUIRE-Rerank, a reranking task for refining top-100 retrievals. Detailed evaluation of a range of recent multimodal models demonstrates that INQUIRE poses a significant challenge, with the best models failing to achieve an mAP@50 above 50%. In addition, we show that reranking with more powerful multimodal models can enhance retrieval performance, yet there remains a significant margin for improvement. By focusing on scientifically-motivated ecological challenges, INQUIRE aims to bridge the gap between AI capabilities and the needs of real-world scientific inquiry, encouraging the development of retrieval systems that can assist with accelerating ecological and biodiversity research. Our dataset and code are available at https://inquire-benchmark.github.io

Paperid: 5660, https://arxiv.org/pdf/2411.01511.pdf

Abstract:
Traditional natural disaster response involves significant coordinated teamwork where speed and efficiency are key. Nonetheless, human limitations can delay critical actions and inadvertently increase human and economic losses. Agentic Large Vision Language Models (LVLMs) offer a new avenue to address this challenge, with the potential for substantial socio-economic impact, particularly by improving resilience and resource access in underdeveloped regions. We introduce DisasTeller, the first multi-LVLM-powered framework designed to automate tasks in post-disaster management, including on-site assessment, emergency alerts, resource allocation, and recovery planning. By coordinating four specialised LVLM agents with GPT-4 as the core model, DisasTeller autonomously implements disaster response activities, reducing human execution time and optimising resource distribution. Our evaluations through both LVLMs and humans demonstrate DisasTeller's effectiveness in streamlining disaster response. This framework not only supports expert teams but also simplifies access to disaster management processes for non-experts, bridging the gap between traditional response methods and LVLM-driven efficiency.

Paperid: 5661, https://arxiv.org/pdf/2410.22187.pdf

Abstract:
Pre-trained vision-language models (VLMs) like CLIP have demonstrated impressive zero-shot performance on a wide range of downstream computer vision tasks. However, there still exists a considerable performance gap between these models and a supervised deep model trained on a downstream dataset. To bridge this gap, we propose a novel active learning (AL) framework that enhances the zero-shot classification performance of VLMs by selecting only a few informative samples from the unlabeled data for annotation during training. To achieve this, our approach first calibrates the predicted entropy of VLMs and then utilizes a combination of self-uncertainty and neighbor-aware uncertainty to calculate a reliable uncertainty measure for active sample selection. Our extensive experiments show that the proposed approach outperforms existing AL approaches on several image classification datasets, and significantly enhances the zero-shot performance of VLMs.

Paperid: 5662, https://arxiv.org/pdf/2410.19552.pdf

Abstract:
Detecting temporal changes in geographical landscapes is critical for applications like environmental monitoring and urban planning. While remote sensing data is abundant, existing vision-language models (VLMs) often fail to capture temporal dynamics effectively. This paper addresses these limitations by introducing an annotated dataset of video frame pairs to track evolving geographical patterns over time. Using fine-tuning techniques like Low-Rank Adaptation (LoRA), quantized LoRA (QLoRA), and model pruning on models such as Video-LLaVA and LLaVA-NeXT-Video, we significantly enhance VLM performance in processing remote sensing temporal changes. Results show significant improvements, with the best performance achieving a BERT score of 0.864 and ROUGE-1 score of 0.576, demonstrating superior accuracy in describing land-use transformations.

Paperid: 5663, https://arxiv.org/pdf/2410.14072.pdf

Abstract:
Recent advancements in vision-language models (VLMs) have expanded their potential for real-world applications, enabling these models to perform complex reasoning on images. In the widely used fully autoregressive transformer-based models like LLaVA, projected visual tokens are prepended to textual tokens. Oftentimes, visual tokens are significantly more than prompt tokens, resulting in increased computational overhead during both training and inference. In this paper, we propose Visual Compact Token Registers (Victor), a method that reduces the number of visual tokens by summarizing them into a smaller set of register tokens. Victor adds a few learnable register tokens after the visual tokens and summarizes the visual information into these registers using the first few layers in the language tower of VLMs. After these few layers, all visual tokens are discarded, significantly improving computational efficiency for both training and inference. Notably, our method is easy to implement and requires a small number of new trainable parameters with minimal impact on model performance. In our experiment, with merely 8 visual registers--about 1% of the original tokens--Victor shows less than a 4% accuracy drop while reducing the total training time by 43% and boosting the inference throughput by 3.3X.

Paperid: 5664, https://arxiv.org/pdf/2410.11963.pdf

Abstract:
Pretraining robust vision or multimodal foundation models (e.g., CLIP) relies on large-scale datasets that may be noisy, potentially misaligned, and have long-tail distributions. Previous works have shown promising results in augmenting datasets by generating synthetic samples. However, they only support domain-specific ad hoc use cases (e.g., either image or text only, but not both), and are limited in data diversity due to a lack of fine-grained control over the synthesis process. In this paper, we design a \emph{controllable} image-text synthesis pipeline, CtrlSynth, for data-efficient and robust multimodal learning. The key idea is to decompose the visual semantics of an image into basic elements, apply user-specified control policies (e.g., remove, add, or replace operations), and recompose them to synthesize images or texts. The decompose and recompose feature in CtrlSynth allows users to control data synthesis in a fine-grained manner by defining customized control policies to manipulate the basic elements. CtrlSynth leverages the capabilities of pretrained foundation models such as large language models or diffusion models to reason and recompose basic elements such that synthetic samples are natural and composed in diverse ways. CtrlSynth is a closed-loop, training-free, and modular framework, making it easy to support different pretrained models. With extensive experiments on 31 datasets spanning different vision and vision-language tasks, we show that CtrlSynth substantially improves zero-shot classification, image-text retrieval, and compositional reasoning performance of CLIP models.

Paperid: 5665, https://arxiv.org/pdf/2410.08397.pdf

Abstract:
We present VoxelPrompt, an agent-driven vision-language framework that tackles diverse radiological tasks through joint modeling of natural language, image volumes, and analytical metrics. VoxelPrompt is multi-modal and versatile, leveraging the flexibility of language interaction while providing quantitatively grounded image analysis. Given a variable number of 3D medical volumes, such as MRI and CT scans, VoxelPrompt employs a language agent that iteratively predicts executable instructions to solve a task specified by an input prompt. These instructions communicate with a vision network to encode image features and generate volumetric outputs (e.g., segmentations). VoxelPrompt interprets the results of intermediate instructions and plans further actions to compute discrete measures (e.g., tumor growth across a series of scans) and present relevant outputs to the user. We evaluate this framework in a sandbox of diverse neuroimaging tasks, and we show that the single VoxelPrompt model can delineate hundreds of anatomical and pathological features, measure many complex morphological properties, and perform open-language analysis of lesion characteristics. VoxelPrompt carries out these objectives with accuracy similar to that of fine-tuned, single-task models for segmentation and visual question-answering, while facilitating a much larger range of tasks. Therefore, by supporting accurate image processing with language interaction, VoxelPrompt provides comprehensive utility for numerous imaging tasks that traditionally require specialized models to address.

Paperid: 5666, https://arxiv.org/pdf/2410.06420.pdf

Abstract:
The global shortage of healthcare workers has demanded the development of smart healthcare assistants, which can help monitor and alert healthcare workers when necessary. We examine the healthcare knowledge of existing Large Vision Language Models (LVLMs) via the Visual Question Answering (VQA) task in hospital settings through expert annotated open-ended questions. We introduce the Emergency Room Visual Question Answering (ERVQA) dataset, consisting of triplets covering diverse emergency room scenarios, a seminal benchmark for LVLMs. By developing a detailed error taxonomy and analyzing answer trends, we reveal the nuanced nature of the task. We benchmark state-of-the-art open-source and closed LVLMs using traditional and adapted VQA metrics: Entailment Score and CLIPScore Confidence. Analyzing errors across models, we infer trends based on properties like decoder type, model size, and in-context examples. Our findings suggest the ERVQA dataset presents a highly complex task, highlighting the need for specialized, domain-specific solutions.

Paperid: 5667, https://arxiv.org/pdf/2410.01966.pdf

Abstract:
Being able to accurately monitor the screen exposure of young children is important for research on phenomena linked to screen use such as childhood obesity, physical activity, and social interaction. Most existing studies rely upon self-report or manual measures from bulky wearable sensors, thus lacking efficiency and accuracy in capturing quantitative screen exposure data. In this work, we developed a novel sensor informatics framework that utilizes egocentric images from a wearable sensor, termed the screen time tracker (STT), and a vision language model (VLM). In particular, we devised a multi-view VLM that takes multiple views from egocentric image sequences and interprets screen exposure dynamically. We validated our approach by using a dataset of children's free-living activities, demonstrating significant improvement over existing methods in plain vision language models and object detection models. Results supported the promise of this monitoring approach, which could optimize behavioral research on screen exposure in children's naturalistic settings.

Paperid: 5668, https://arxiv.org/pdf/2410.01438.pdf

Abstract:
Vision-Language Models (VLMs) have achieved remarkable performance on a variety of tasks, yet they remain vulnerable to jailbreak attacks that compromise safety and reliability. In this paper, we provide an information-theoretic framework for understanding the fundamental trade-off between the effectiveness of these attacks and their stealthiness. Drawing on Fano's inequality, we demonstrate how an attacker's success probability is intrinsically linked to the stealthiness of generated prompts. Building on this, we propose an efficient algorithm for detecting non-stealthy jailbreak attacks, offering significant improvements in model robustness. Experimental results highlight the tension between strong attacks and their detectability, providing insights into both adversarial strategies and defense mechanisms.

Paperid: 5669, https://arxiv.org/pdf/2409.17045.pdf

Abstract:
We provide a dataset for enabling Deep Generative Models (DGMs) in engineering design and propose methods to automate data labeling by utilizing large-scale foundation models. GeoBiked is curated to contain 4 355 bicycle images, annotated with structural and technical features and is used to investigate two automated labeling techniques: The utilization of consolidated latent features (Hyperfeatures) from image-generation models to detect geometric correspondences (e.g. the position of the wheel center) in structural images and the generation of diverse text descriptions for structural images. GPT-4o, a vision-language-model (VLM), is instructed to analyze images and produce diverse descriptions aligned with the system-prompt. By representing technical images as Diffusion-Hyperfeatures, drawing geometric correspondences between them is possible. The detection accuracy of geometric points in unseen samples is improved by presenting multiple annotated source images. GPT-4o has sufficient capabilities to generate accurate descriptions of technical images. Grounding the generation only on images leads to diverse descriptions but causes hallucinations, while grounding it on categorical labels restricts the diversity. Using both as input balances creativity and accuracy. Successfully using Hyperfeatures for geometric correspondence suggests that this approach can be used for general point-detection and annotation tasks in technical images. Labeling such images with text descriptions using VLMs is possible, but dependent on the models detection capabilities, careful prompt-engineering and the selection of input information. Applying foundation models in engineering design is largely unexplored. We aim to bridge this gap with a dataset to explore training, finetuning and conditioning DGMs in this field and suggesting approaches to bootstrap foundation models to process technical images.

Paperid: 5670, https://arxiv.org/pdf/2409.15486.pdf

Abstract:
Ensuring robust performance on long-tail examples is an important problem for many real-world applications of machine learning, such as autonomous driving. This work focuses on the problem of identifying rare examples within a corpus of unlabeled data. We propose a simple and scalable data mining approach that leverages the knowledge contained within a large vision language model (VLM). Our approach utilizes a VLM to summarize the content of an image into a set of keywords, and we identify rare examples based on keyword frequency. We find that the VLM offers a distinct signal for identifying long-tail examples when compared to conventional methods based on model uncertainty. Therefore, we propose a simple and general approach for integrating signals from multiple mining algorithms. We evaluate the proposed method on two diverse tasks: 2D image classification, in which inter-class variation is the primary source of data diversity, and on 3D object detection, where intra-class variation is the main concern. Furthermore, through the detection task, we demonstrate that the knowledge extracted from 2D images is transferable to the 3D domain. Our experiments consistently show large improvements (between 10\% and 50\%) over the baseline techniques on several representative benchmarks: ImageNet-LT, Places-LT, and the Waymo Open Dataset.

Paperid: 5671, https://arxiv.org/pdf/2409.15291.pdf

Abstract:
Pathology images are crucial for diagnosing and managing various diseases by visualizing cellular and tissue-level abnormalities. Recent advancements in artificial intelligence (AI), particularly multimodal models like ChatGPT, have shown promise in transforming medical image analysis through capabilities such as medical vision-language question answering. However, there remains a significant gap in integrating pathology image data with these AI models for clinical applications. This study benchmarks the performance of GPT on pathology images, assessing their diagnostic accuracy and efficiency in real-word clinical records. We observe significant deficits of GPT in bone diseases and a fair-level performance in diseases from other three systems. Despite offering satisfactory abnormality annotations, GPT exhibits consistent disadvantage in terminology accuracy and multimodal integration. Specifically, we demonstrate GPT's failures in interpreting immunohistochemistry results and diagnosing metastatic cancers. This study highlight the weakness of current generalist GPT model and contribute to the integration of pathology and advanced AI.

Paperid: 5672, https://arxiv.org/pdf/2409.12842.pdf

Abstract:
Vision language models (VLMs) can simultaneously reason about images and texts to tackle many tasks, from visual question answering to image captioning. This paper focuses on map parsing, a novel task that is unexplored within the VLM context and particularly useful to mobile robots. Map parsing requires understanding not only the labels but also the geometric configurations of a map, i.e., what areas are like and how they are connected. To evaluate the performance of VLMs on map parsing, we prompt VLMs with floorplan maps to generate task plans for complex indoor navigation. Our results demonstrate the remarkable capability of VLMs in map parsing, with a success rate of 0.96 in tasks requiring a sequence of nine navigation actions, e.g., approaching and going through doors. Other than intuitive observations, e.g., VLMs do better in smaller maps and simpler navigation tasks, there was a very interesting observation that its performance drops in large open areas. We provide practical suggestions to address such challenges as validated by our experimental results. Webpage: https://shorturl.at/OUkEY

Paperid: 5673, https://arxiv.org/pdf/2409.11513.pdf

Abstract:
Video Language Models (VLMs) are crucial for generalizing across diverse tasks and using language cues to enhance learning. While transformer-based architectures have been the de facto in vision-language training, they face challenges like quadratic computational complexity, high GPU memory usage, and difficulty with long-term dependencies. To address these limitations, we introduce MambaVL, a novel model that leverages recent advancements in selective state space modality fusion to efficiently capture long-range dependencies and learn joint representations for vision and language data. MambaVL utilizes a shared state transition matrix across both modalities, allowing the model to capture information about actions from multiple perspectives within the scene. Furthermore, we propose a question-answering task that helps guide the model toward relevant cues. These questions provide critical information about actions, objects, and environmental context, leading to enhanced performance. As a result, MambaVL achieves state-of-the-art performance in action recognition on the Epic-Kitchens-100 dataset and outperforms baseline methods in action anticipation.

Paperid: 5674, https://arxiv.org/pdf/2409.09582.pdf

Abstract:
The success of Vision Language Models (VLMs) on various vision-language tasks heavily relies on pre-training with large scale web-crawled datasets. However, the noisy and incomplete nature of web data makes dataset scale crucial for performance, rendering end-to-end training increasingly prohibitive. In this paper, we propose NEVLP, a noise-robust framework for efficient vision-language pre-training that requires less pre-training data. Specifically, we bridge the modality gap between a frozen image encoder and a large language model with a transformer and introduce two innovative learning strategies: noise-adaptive learning and concept-enhanced learning to mitigate the impact of noise. In noise-adaptive learning, we estimate the noise probability of each image-text pair based on the transformer's memorization effect and employ noise-adaptive regularization on image-text contrastive learning to condition cross-modal alignment. In concept-enhanced learning, we enrich incomplete text by incorporating visual concepts (objects in the image) to provide prior information about existing objects for image-text matching and image-grounded text generation, thereby mitigating text incompletion. Our framework effectively utilizes noisy web data and achieves state-of-the-art performance with less pre-training data across a wide range of vision-language tasks, including image-text retrieval, image captioning, and visual question answering.

Paperid: 5675, https://arxiv.org/pdf/2601.21220.pdf

Abstract:
Multimodal Large Language Models (MLLMs) have achieved remarkable performance across vision-language tasks. Recent advancements allow these models to process multiple images as inputs. However, the vulnerabilities of multi-image MLLMs remain unexplored. Existing adversarial attacks focus on single-image settings and often assume a white-box threat model, which is impractical in many real-world scenarios. This paper introduces LAMP, a black-box method for learning Universal Adversarial Perturbations (UAPs) targeting multi-image MLLMs. LAMP applies an attention-based constraint that prevents the model from effectively aggregating information across images. LAMP also introduces a novel cross-image contagious constraint that forces perturbed tokens to influence clean tokens, spreading adversarial effects without requiring all inputs to be modified. Additionally, an index-attention suppression loss enables a robust position-invariant attack. Experimental results show that LAMP outperforms SOTA baselines and achieves the highest attack success rates across multiple vision-language tasks and models.

Paperid: 5676, https://arxiv.org/pdf/2601.18386.pdf

Abstract:
Existing automated attack suites operate as static ensembles with fixed sequences, lacking strategic adaptation and semantic awareness. This paper introduces the Agentic Reasoning for Methods Orchestration and Reparameterization (ARMOR) framework to address these limitations. ARMOR orchestrates three canonical adversarial primitives, Carlini-Wagner (CW), Jacobian-based Saliency Map Attack (JSMA), and Spatially Transformed Attacks (STA) via Vision Language Models (VLM)-guided agents that collaboratively generate and synthesize perturbations through a shared ``Mixing Desk". Large Language Models (LLMs) adaptively tune and reparameterize parallel attack agents in a real-time, closed-loop system that exploits image-specific semantic vulnerabilities. On standard benchmarks, ARMOR achieves improved cross-architecture transfer and reliably fools both settings, delivering a blended output for blind targets and selecting the best attack or blended attacks for white-box targets using a confidence-and-SSIM score.

Paperid: 5677, https://arxiv.org/pdf/2601.16409.pdf

Abstract:
Move\,37 marks one of the major breakthroughs in AI in terms of its ability to surpass human expertise and discover novel strategies beyond the traditional game play in the strategic two-player board game of Go. The domains of Natural Language Processing, Computer Vision, and Robotics have also undergone a similar phenomenon through the advent of large foundational models in the form of Large Language Models (LLMs), Vision Language Models (VLMs) and Vision Language Action models (VLAs), respectively. In this paper, we investigate the current state of Artificial Intelligence for Database Systems research (AI4DB), and assess how far AI4DB systems are from achieving their own Move\,37 moment. We envision a Generative Database Agent (Gen-DBA, for short) as the pathway to achieving Move\,37 for database systems that will bring generative reasoning and creativity into the realm of database learning tasks. This vision paper explores this direction by presenting the recipe for building Gen-DBA that encompasses but is not limited to a Transformer backbone, a hardware-grounded tokenization mechanism, a two-stage Goal-Directed Next Token Prediction training paradigm, and a generative inference process.

Paperid: 5678, https://arxiv.org/pdf/2601.12260.pdf

Abstract:
Document understanding (VRDU) in regulated domains is particularly challenging, since scanned documents often contain sensitive, evolving, and domain specific knowledge. This leads to two major challenges: the lack of manual annotations for model adaptation and the difficulty for pretrained models to stay up-to-date with domain-specific facts. While Multimodal Large Language Models (MLLMs) show strong zero-shot abilities, they still suffer from hallucination and limited domain grounding. In contrast, discriminative Vision-Language Pre-trained Models (VLPMs) provide reliable grounding but require costly annotations to cover new domains. We introduce Docs2Synth, a synthetic-supervision framework that enables retrieval-guided inference for private and low-resource domains. Docs2Synth automatically processes raw document collections, generates and verifies diverse QA pairs via an agent-based system, and trains a lightweight visual retriever to extract domain-relevant evidence. During inference, the retriever collaborates with an MLLM through an iterative retrieval--generation loop, reducing hallucination and improving response consistency. We further deliver Docs2Synth as an easy-to-use Python package, enabling plug-and-play deployment across diverse real-world scenarios. Experiments on multiple VRDU benchmarks show that Docs2Synth substantially enhances grounding and domain generalization without requiring human annotations.

Paperid: 5679, https://arxiv.org/pdf/2601.11801.pdf

Abstract:
Robot design is a nontrivial process that involves careful consideration of multiple criteria, including user specifications, kinematic structures, and visual appearance. Therefore, the design process often relies heavily on domain expertise and significant human effort. The majority of current methods are rule-based, requiring the specification of a grammar or a set of primitive components and modules that can be composed to create a design. We propose a novel automated robot design framework, RobotDesignGPT, that leverages the general knowledge and reasoning capabilities of large pre-trained vision-language models to automate the robot design synthesis process. Our framework synthesizes an initial robot design from a simple user prompt and a reference image. Our novel visual feedback approach allows us to greatly improve the design quality and reduce unnecessary manual feedback. We demonstrate that our framework can design visually appealing and kinematically valid robots inspired by nature, ranging from legged animals to flying creatures. We justify the proposed framework by conducting an ablation study and a user study.

Paperid: 5680, https://arxiv.org/pdf/2601.11666.pdf

Abstract:
We introduce MATEX (Multi-scale Attention and Text-guided Explainability), a novel framework that advances interpretability in medical vision-language models by incorporating anatomically informed spatial reasoning. MATEX synergistically combines multi-layer attention rollout, text-guided spatial priors, and layer consistency analysis to produce precise, stable, and clinically meaningful gradient attribution maps. By addressing key limitations of prior methods, such as spatial imprecision, lack of anatomical grounding, and limited attention granularity, MATEX enables more faithful and interpretable model explanations. Evaluated on the MS-CXR dataset, MATEX outperforms the state-of-the-art M2IB approach in both spatial precision and alignment with expert-annotated findings. These results highlight MATEX's potential to enhance trust and transparency in radiological AI applications.

Paperid: 5681, https://arxiv.org/pdf/2601.11644.pdf

Abstract:
Vision-Language Models (VLMs) demonstrate impressive capabilities across multimodal tasks, yet exhibit systematic spatial reasoning failures, achieving only 49% (CLIP) to 54% (BLIP-2) accuracy on basic directional relationships. For safe deployment in robotics and autonomous systems, we need to predict when to trust VLM spatial predictions rather than accepting all outputs. We propose a vision-based confidence estimation framework that validates VLM predictions through independent geometric verification using object detection. Unlike text-based approaches relying on self-assessment, our method fuses four signals via gradient boosting: geometric alignment between VLM claims and coordinates, spatial ambiguity from overlap, detection quality, and VLM internal uncertainty. We achieve 0.674 AUROC on BLIP-2 (34.0% improvement over text-based baselines) and 0.583 AUROC on CLIP (16.1% improvement), generalizing across generative and classification architectures. Our framework enables selective prediction: at 60% target accuracy, we achieve 61.9% coverage versus 27.6% baseline (2.2x improvement) on BLIP-2. Feature analysis reveals vision-based signals contribute 87.4% of model importance versus 12.7% from VLM confidence, validating that external geometric verification outperforms self-assessment. We demonstrate reliable scene graph construction where confidence-based pruning improves precision from 52.1% to 78.3% while retaining 68.2% of edges.

Paperid: 5682, https://arxiv.org/pdf/2601.11178.pdf

Abstract:
Social media platforms are increasingly dominated by long-form multimodal content, where harmful narratives are constructed through a complex interplay of audio, visual, and textual cues. While automated systems can flag hate speech with high accuracy, they often function as "black boxes" that fail to provide the granular, interpretable evidence, such as precise timestamps and target identities, required for effective human-in-the-loop moderation. In this work, we introduce TANDEM, a unified framework that transforms audio-visual hate detection from a binary classification task into a structured reasoning problem. Our approach employs a novel tandem reinforcement learning strategy where vision-language and audio-language models optimize each other through self-constrained cross-modal context, stabilizing reasoning over extended temporal sequences without requiring dense frame-level supervision. Experiments across three benchmark datasets demonstrate that TANDEM significantly outperforms zero-shot and context-augmented baselines, achieving 0.73 F1 in target identification on HateMM (a 30% improvement over state-of-the-art) while maintaining precise temporal grounding. We further observe that while binary detection is robust, differentiating between offensive and hateful content remains challenging in multi-class settings due to inherent label ambiguity and dataset imbalance. More broadly, our findings suggest that structured, interpretable alignment is achievable even in complex multimodal settings, offering a blueprint for the next generation of transparent and actionable online safety moderation tools.

Paperid: 5683, https://arxiv.org/pdf/2601.09770.pdf

Abstract:
Recent advances in vision-language models (VLMs) and reinforcement learning (RL) have driven progress in GUI automation. However, most existing methods rely on static, one-shot visual inputs and passive perception, lacking the ability to adaptively determine when, whether, and how to observe the interface. We present GUI-Eyes, a reinforcement learning framework for active visual perception in GUI tasks. To acquire more informative observations, the agent learns to make strategic decisions on both whether and how to invoke visual tools, such as cropping or zooming, within a two-stage reasoning process. To support this behavior, we introduce a progressive perception strategy that decomposes decision-making into coarse exploration and fine-grained grounding, coordinated by a two-level policy. In addition, we design a spatially continuous reward function tailored to tool usage, which integrates both location proximity and region overlap to provide dense supervision and alleviate the reward sparsity common in GUI environments. On the ScreenSpot-Pro benchmark, GUI-Eyes-3B achieves 44.8% grounding accuracy using only 3k labeled samples, significantly outperforming both supervised and RL-based baselines. These results highlight that tool-aware active perception, enabled by staged policy reasoning and fine-grained reward feedback, is critical for building robust and data-efficient GUI agents.

Paperid: 5684, https://arxiv.org/pdf/2601.09449.pdf

Abstract:
We present PrivLEX, a novel image privacy classifier that grounds its decisions in legally defined personal data concepts. PrivLEX is the first interpretable privacy classifier aligned with legal concepts that leverages the recognition capabilities of Vision-Language Models (VLMs). PrivLEX relies on zero-shot VLM concept detection to provide interpretable classification through a label-free Concept Bottleneck Model, without requiring explicit concept labels during training. We demonstrate PrivLEX's ability to identify personal data concepts that are present in images. We further analyse the sensitivity of such concepts as perceived by human annotators of image privacy datasets.

Paperid: 5685, https://arxiv.org/pdf/2601.09136.pdf

Abstract:
General-purpose Large Vision-Language Models (LVLMs), despite their massive scale, often falter in dermatology due to "diffuse attention" - the inability to disentangle subtle pathological lesions from background noise. In this paper, we challenge the assumption that parameter scaling is the only path to medical precision. We introduce SkinFlow, a framework that treats diagnosis as an optimization of visual information transmission efficiency. Our approach utilizes a Virtual-Width Dynamic Vision Encoder (DVE) to "unfold" complex pathological manifolds without physical parameter expansion, coupled with a two-stage Reinforcement Learning strategy. This strategy sequentially aligns explicit medical descriptions (Stage I) and reconstructs implicit diagnostic textures (Stage II) within a constrained semantic space. Furthermore, we propose a clinically grounded evaluation protocol that prioritizes diagnostic safety and hierarchical relevance over rigid label matching. Empirical results are compelling: our 7B model establishes a new state-of-the-art on the Fitzpatrick17k benchmark, achieving a +12.06% gain in Top-1 accuracy and a +28.57% boost in Top-6 accuracy over the massive general-purpose models (e.g., Qwen3VL-235B and GPT-5.2). These findings demonstrate that optimizing geometric capacity and information flow yields superior diagnostic reasoning compared to raw parameter scaling.

Paperid: 5686, https://arxiv.org/pdf/2601.08095.pdf

Abstract:
In this paper, we present an automated pipeline for generating domain-specific synthetic datasets with diffusion models, addressing the distribution shift between pre-trained models and real-world deployment environments. Our three-stage framework first synthesizes target objects within domain-specific backgrounds through controlled inpainting. The generated outputs are then validated via a multi-modal assessment that integrates object detection, aesthetic scoring, and vision-language alignment. Finally, a user-preference classifier is employed to capture subjective selection criteria. This pipeline enables the efficient construction of high-quality, deployable datasets while reducing reliance on extensive real-world data collection.

Paperid: 5687, https://arxiv.org/pdf/2601.07820.pdf

Abstract:
In human conversation, both interlocutors play an active role in maintaining mutual understanding. When addressees are uncertain about what speakers mean, for example, they can request clarification. It is an open question for language models whether they can assume a similar addressee role, recognizing and expressing their own uncertainty through clarification. We argue that reference games are a good testbed to approach this question as they are controlled, self-contained, and make clarification needs explicit and measurable. To test this, we evaluate three vision-language models comparing a baseline reference resolution task to an experiment where the models are instructed to request clarification when uncertain. The results suggest that even in such simple tasks, models often struggle to recognize internal uncertainty and translate it into adequate clarification behavior. This demonstrates the value of reference games as testbeds for interaction qualities of (vision and) language models.

Paperid: 5688, https://arxiv.org/pdf/2601.06458.pdf

Abstract:
Large Language Models (LLMs) have recently shown strong potential for usage in sequential recommendation tasks through text-only models, which combine advanced prompt design, contrastive alignment, and fine-tuning on downstream domain-specific data. While effective, these approaches overlook the rich visual information present in many real-world recommendation scenarios, particularly in e-commerce. This paper proposes PixRec - a vision-language framework that incorporates both textual attributes and product images into the recommendation pipeline. Our architecture leverages a vision-language model backbone capable of jointly processing image-text sequences, maintaining a dual-tower structure and mixed training objective while aligning multi-modal feature projections for both item-item and user-item interactions. Using the Amazon Reviews dataset augmented with product images, our experiments demonstrate $3\times$ and 40% improvements in top-rank and top-10 rank accuracy over text-only recommenders respectively, indicating that visual features can help distinguish items with similar textual descriptions. Our work outlines future directions for scaling multi-modal recommenders training, enhancing visual-text feature fusion, and evaluating inference-time performance. This work takes a step toward building software systems utilizing visual information in sequential recommendation for real-world applications like e-commerce.

Paperid: 5689, https://arxiv.org/pdf/2601.06451.pdf

Abstract:
Food cutting is a highly practical yet underexplored application at the intersection of vision and robotic manipulation. The task remains challenging because interactions between the knife and deformable materials are highly nonlinear and often entail large deformations, frequent contact, and topological change, which in turn hinder stable and safe large-scale data collection. To address these challenges, we propose a unified framework that couples a vision-language-action (VLA) dataset with a physically realistic cutting simulator built on the material point method (MPM). Our simulator adopts MLS-MPM as its computational core, reducing numerical dissipation and energy drift while preserving rotational and shear responses even under topology-changing cuts. During cutting, forces and stress distributions are estimated from impulse exchanges between particles and the grid, enabling stable tracking of transient contact forces and energy transfer. We also provide a benchmark dataset that integrates diverse cutting trajectories, multi-view visual observations, and fine-grained language instructions, together with force--torque and tool--pose labels to provide physically consistent training signals. These components realize a learning--evaluation loop that respects the core physics of cutting and establishes a safe, reproducible, and scalable foundation for advancing VLA models in deformable object manipulation.

Paperid: 5690, https://arxiv.org/pdf/2601.05399.pdf

Abstract:
CLIP and BiomedCLIP are examples of vision-language foundation models and offer strong cross-modal embeddings; however, they are not optimized for fine-grained medical retrieval tasks, such as retrieving clinically relevant radiology reports using chest X-ray (CXR) image queries. To address this shortcoming, we propose a multi-task learning framework to fine-tune BiomedCLIP and evaluate improvements to CXR image-text retrieval. Using BiomedCLIP as the backbone, we incorporate a lightweight MLP projector head trained with a multi-task composite loss function that includes: (1) a binary cross-entropy loss to distinguish normal from abnormal CXR studies, (2) a supervised contrastive loss to reinforce intra-class consistency, and (3) a CLIP loss to maintain cross-modal alignment. Experimental results demonstrate that the fine-tuned model achieves more balanced and clinically meaningful performance across both image-to-text and text-to-image retrieval tasks compared to the pretrained BiomedCLIP and general-purpose CLIP models. Furthermore, t-SNE visualizations reveal clearer semantic clustering of normal and abnormal cases, demonstrating the model's enhanced diagnostic sensitivity. These findings highlight the value of domain-adaptive, multi-task learning for advancing cross-modal retrieval in biomedical applications.

Paperid: 5691, https://arxiv.org/pdf/2601.04824.pdf

Abstract:
Automatic identification of events and recurrent behavior analysis are critical for video surveillance. However, most existing content-based video retrieval benchmarks focus on scene-level similarity and do not evaluate the action discrimination required in surveillance. To address this gap, we introduce SOVABench (Surveillance Opposite Vehicle Actions Benchmark), a real-world retrieval benchmark built from surveillance footage and centered on vehicle-related actions. SOVABench defines two evaluation protocols (inter-pair and intra-pair) to assess cross-action discrimination and temporal direction understanding. Although action distinctions are generally intuitive for human observers, our experiments show that they remain challenging for state-of-the-art vision and multimodal models. Leveraging the visual reasoning and instruction-following capabilities of Multimodal Large Language Models (MLLMs), we present a training-free framework for producing interpretable embeddings from MLLM-generated descriptions for both images and videos. The framework achieves strong performance on SOVABench as well as on several spatial and counting benchmarks where contrastive Vision-Language Models often fail. The code, annotations, and instructions to construct the benchmark are publicly available.

Paperid: 5692, https://arxiv.org/pdf/2601.03707.pdf

Abstract:
Existing Unmanned Aerial Vehicle (UAV) Vision-Language Navigation (VLN) datasets face issues such as dependence on virtual environments, lack of naturalness in instructions, and limited scale. To address these challenges, we propose AirNav, a large-scale UAV VLN benchmark constructed from real urban aerial data, rather than synthetic environments, with natural and diverse instructions. Additionally, we introduce the AirVLN-R1, which combines Supervised Fine-Tuning and Reinforcement Fine-Tuning to enhance performance and generalization. The feasibility of the model is preliminarily evaluated through real-world tests. Our dataset and code are publicly available.

Paperid: 5693, https://arxiv.org/pdf/2601.03331.pdf

Abstract:
Recent advances in Vision-Language Models (VLMs) have improved performance in multi-modal learning, raising the question of whether these models truly understand the content they process. Crucially, can VLMs detect when a reasoning process is wrong and identify its error type? To answer this, we present MMErroR, a multi-modal benchmark of 2,013 samples, each embedding a single coherent reasoning error. These samples span 24 subdomains across six top-level domains, ensuring broad coverage and taxonomic richness. Unlike existing benchmarks that focus on answer correctness, MMErroR targets a process-level, error-centric evaluation that requires models to detect incorrect reasoning and classify the error type within both visual and linguistic contexts. We evaluate 20 advanced VLMs, even the best model (Gemini-3.0-Pro) classifies the error in only 66.47\% of cases, underscoring the challenge of identifying erroneous reasoning. Furthermore, the ability to accurately identify errors offers valuable insights into the capabilities of multi-modal reasoning models. Project Page: https://mmerror-benchmark.github.io

Paperid: 5694, https://arxiv.org/pdf/2601.01910.pdf

Abstract:
Autonomous path planning requires a synergy between global reasoning and geometric precision, especially in complex or cluttered environments. While classical A* is valued for its optimality, it incurs prohibitive computational and memory costs in large-scale scenarios. Recent attempts to mitigate these limitations by using Large Language Models for waypoint guidance remain insufficient, as they rely only on text-based reasoning without spatial grounding. As a result, such models often produce incorrect waypoints in topologically complex environments with dead ends, and lack the perceptual capacity to interpret ambiguous physical boundaries. These inconsistencies lead to costly corrective expansions and undermine the intended computational efficiency. We introduce MMP-A*, a multimodal framework that integrates the spatial grounding capabilities of vision-language models with a novel adaptive decay mechanism. By anchoring high-level reasoning in physical geometry, the framework produces coherent waypoint guidance that addresses the limitations of text-only planners. The adaptive decay mechanism dynamically regulates the influence of uncertain waypoints within the heuristic, ensuring geometric validity while substantially reducing memory overhead. To evaluate robustness, we test the framework in challenging environments characterized by severe clutter and topological complexity. Experimental results show that MMP-A* achieves near-optimal trajectories with significantly reduced operational costs, demonstrating its potential as a perception-grounded and computationally efficient paradigm for autonomous navigation.

Paperid: 5695, https://arxiv.org/pdf/2512.24470.pdf

Abstract:
The draft IMO MASS Code requires autonomous and remotely supervised maritime vessels to detect departures from their operational design domain, enter a predefined fallback that notifies the operator, permit immediate human override, and avoid changing the voyage plan without approval. Meeting these obligations in the alert-to-takeover gap calls for a short-horizon, human-overridable fallback maneuver. Classical maritime autonomy stacks struggle when the correct action depends on meaning (e.g., diver-down flag means people in the water, fire close by means hazard). We argue (i) that vision-language models (VLMs) provide semantic awareness for such out-of-distribution situations, and (ii) that a fast-slow anomaly pipeline with a short-horizon, human-overridable fallback maneuver makes this practical in the handover window. We introduce Semantic Lookout, a camera-only, candidate-constrained vision-language model (VLM) fallback maneuver selector that selects one cautious action (or station-keeping) from water-valid, world-anchored trajectories under continuous human authority. On 40 harbor scenes we measure per-call scene understanding and latency, alignment with human consensus (model majority-of-three voting), short-horizon risk-relief on fire hazard scenes, and an on-water alert->fallback maneuver->operator handover. Sub-10 s models retain most of the awareness of slower state-of-the-art models. The fallback maneuver selector outperforms geometry-only baselines and increases standoff distance on fire scenes. A field run verifies end-to-end operation. These results support VLMs as semantic fallback maneuver selectors compatible with the draft IMO MASS Code, within practical latency budgets, and motivate future work on domain-adapted, hybrid autonomy that pairs foundation-model semantics with multi-sensor bird's-eye-view perception and short-horizon replanning.

Paperid: 5696, https://arxiv.org/pdf/2512.24212.pdf

Abstract:
Efficiently finding targets in complex environments is fundamental to real-world embodied applications. While recent advances in multimodal foundation models have enabled zero-shot object goal navigation, allowing robots to search for arbitrary objects without fine-tuning, existing methods face two key limitations: (1) heavy reliance on precise depth and pose information provided by simulators, which restricts applicability in real-world scenarios; and (2) lack of in-context learning (ICL) capability, making it difficult to quickly adapt to new environments, as in leveraging short videos. To address these challenges, we propose RANGER, a novel zero-shot, open-vocabulary semantic navigation framework that operates using only a monocular camera. Leveraging powerful 3D foundation models, RANGER eliminates the dependency on depth and pose while exhibiting strong ICL capability. By simply observing a short video of a new environment, the system can also significantly improve task efficiency without requiring architectural modifications or fine-tuning. The framework integrates several key components: keyframe-based 3D reconstruction, semantic point cloud generation, vision-language model (VLM)-driven exploration value estimation, high-level adaptive waypoint selection, and low-level action execution. Experiments on the HM3D benchmark and real-world environments demonstrate that RANGER achieves competitive performance in terms of navigation success rate and exploration efficiency, while showing superior ICL adaptability, with no previous 3D mapping of the environment required.

Paperid: 5697, https://arxiv.org/pdf/2512.24160.pdf

Abstract:
We present IMDD-1M, the first large-scale Industrial Multimodal Defect Dataset comprising 1,000,000 aligned image-text pairs, designed to advance multimodal learning for manufacturing and quality inspection. IMDD-1M contains high-resolution real-world defects spanning over 60 material categories and more than 400 defect types, each accompanied by expert-verified annotations and fine-grained textual descriptions detailing defect location, severity, and contextual attributes. This dataset enables a wide spectrum of applications, including classification, segmentation, retrieval, captioning, and generative modeling. Building upon IMDD-1M, we train a diffusion-based vision-language foundation model from scratch, specifically tailored for industrial scenarios. The model serves as a generalizable foundation that can be efficiently adapted to specialized domains through lightweight fine-tuning. With less than 5% of the task-specific data required by dedicated expert models, it achieves comparable performance, highlighting the potential of data-efficient foundation model adaptation for industrial inspection and generation, paving the way for scalable, domain-adaptive, and knowledge-grounded manufacturing intelligence.

Paperid: 5698, https://arxiv.org/pdf/2512.22681.pdf

Abstract:
Recent text-to-image diffusion models have achieved remarkable visual fidelity but often struggle with semantic alignment to complex prompts. We introduce CritiFusion, a novel inference-time framework that integrates a multimodal semantic critique mechanism with frequency-domain refinement to improve text-to-image consistency and detail. The proposed CritiCore module leverages a vision-language model and multiple large language models to enrich the prompt context and produce high-level semantic feedback, guiding the diffusion process to better align generated content with the prompt's intent. Additionally, SpecFusion merges intermediate generation states in the spectral domain, injecting coarse structural information while preserving high-frequency details. No additional model training is required. CritiFusion serves as a plug-in refinement stage compatible with existing diffusion backbones. Experiments on standard benchmarks show that our method notably improves human-aligned metrics of text-to-image correspondence and visual quality. CritiFusion consistently boosts performance on human preference scores and aesthetic evaluations, achieving results on par with state-of-the-art reward optimization approaches. Qualitative results further demonstrate superior detail, realism, and prompt fidelity, indicating the effectiveness of our semantic critique and spectral alignment strategy.

Paperid: 5699, https://arxiv.org/pdf/2512.21863.pdf

Abstract:
Frozen Large Video Language Models (LVLMs) are increasingly employed in micro-video recommendation due to their strong multimodal understanding. However, their integration lacks systematic empirical evaluation: practitioners typically deploy LVLMs as fixed black-box feature extractors without systematically comparing alternative representation strategies. To address this gap, we present the first systematic empirical study along two key design dimensions: (i) integration strategies with ID embeddings, specifically replacement versus fusion, and (ii) feature extraction paradigms, comparing LVLM-generated captions with intermediate decoder hidden states. Extensive experiments on representative LVLMs reveal three key principles: (1) intermediate hidden states consistently outperform caption-based representations, as natural-language summarization inevitably discards fine-grained visual semantics crucial for recommendation; (2) ID embeddings capture irreplaceable collaborative signals, rendering fusion strictly superior to replacement; and (3) the effectiveness of intermediate decoder features varies significantly across layers. Guided by these insights, we propose the Dual Feature Fusion (DFF) Framework, a lightweight and plug-and-play approach that adaptively fuses multi-layer representations from frozen LVLMs with item ID embeddings. DFF achieves state-of-the-art performance on two real-world micro-video recommendation benchmarks, consistently outperforming strong baselines and providing a principled approach to integrating off-the-shelf large vision-language models into micro-video recommender systems.

Paperid: 5700, https://arxiv.org/pdf/2512.21329.pdf

Abstract:
Reasoning benchmarks such as the Abstraction and Reasoning Corpus (ARC) and ARC-AGI are widely used to assess progress in artificial intelligence and are often interpreted as probes of core, so-called ``fluid'' reasoning abilities. Despite their apparent simplicity for humans, these tasks remain challenging for frontier vision-language models (VLMs), a gap commonly attributed to deficiencies in machine reasoning. We challenge this interpretation and hypothesize that the gap arises primarily from limitations in visual perception rather than from shortcomings in inductive reasoning. To verify this hypothesis, we introduce a two-stage experimental pipeline that explicitly separates perception and reasoning. In the perception stage, each image is independently converted into a natural-language description, while in the reasoning stage a model induces and applies rules using these descriptions. This design prevents leakage of cross-image inductive signals and isolates reasoning from perception bottlenecks. Across three ARC-style datasets, Mini-ARC, ACRE, and Bongard-LOGO, we show that the perception capability is the dominant factor underlying the observed performance gap by comparing the two-stage pipeline with against standard end-to-end one-stage evaluation. Manual inspection of reasoning traces in the VLM outputs further reveals that approximately 80 percent of model failures stem from perception errors. Together, these results demonstrate that ARC-style benchmarks conflate perceptual and reasoning challenges and that observed performance gaps may overstate deficiencies in machine reasoning. Our findings underscore the need for evaluation protocols that disentangle perception from reasoning when assessing progress in machine intelligence.

Paperid: 5701, https://arxiv.org/pdf/2512.21194.pdf

Abstract:
Vision-Language Models (VLMs) have achieved remarkable progress across tasks such as visual question answering and image captioning. Yet, the extent to which these models perform visual reasoning as opposed to relying on linguistic priors remains unclear. To address this, we introduce VisRes Bench, a benchmark designed to study visual reasoning in naturalistic settings without contextual language supervision. Analyzing model behavior across three levels of complexity, we uncover clear limitations in perceptual and relational visual reasoning capacities. VisRes isolates distinct reasoning abilities across its levels. Level 1 probes perceptual completion and global image matching under perturbations such as blur, texture changes, occlusion, and rotation; Level 2 tests rule-based inference over a single attribute (e.g., color, count, orientation); and Level 3 targets compositional reasoning that requires integrating multiple visual attributes. Across more than 19,000 controlled task images, we find that state-of-the-art VLMs perform near random under subtle perceptual perturbations, revealing limited abstraction beyond pattern recognition. We conclude by discussing how VisRes provides a unified framework for advancing abstract visual reasoning in multimodal research.

Paperid: 5702, https://arxiv.org/pdf/2512.20675.pdf

Abstract:
Learning generalizable reward functions is a core challenge in embodied intelligence. Recent work leverages contrastive vision language models (VLMs) to obtain dense, domain-agnostic rewards without human supervision. These methods adapt VLMs into reward models through increasingly complex learning objectives, yet meaningful comparison remains difficult due to differences in training data, architectures, and evaluation settings. In this work, we isolate the impact of the learning objective by evaluating recent VLM-based reward models under a unified framework with identical backbones, finetuning data, and evaluation environments. Using Meta-World tasks, we assess modeling accuracy by measuring consistency with ground truth reward and correlation with expert progress. Remarkably, we show that a simple triplet loss outperforms state-of-the-art methods, suggesting that much of the improvements in recent approaches could be attributed to differences in data and architectures.

Paperid: 5703, https://arxiv.org/pdf/2512.20387.pdf

Abstract:
We propose a Vision-Language Simulation Model (VLSM) that unifies visual and textual understanding to synthesize executable FlexScript from layout sketches and natural-language prompts, enabling cross-modal reasoning for industrial simulation systems. To support this new paradigm, the study constructs the first large-scale dataset for generative digital twins, comprising over 120,000 prompt-sketch-code triplets that enable multimodal learning between textual descriptions, spatial structures, and simulation logic. In parallel, three novel evaluation metrics, Structural Validity Rate (SVR), Parameter Match Rate (PMR), and Execution Success Rate (ESR), are proposed specifically for this task to comprehensively evaluate structural integrity, parameter fidelity, and simulator executability. Through systematic ablation across vision encoders, connectors, and code-pretrained language backbones, the proposed models achieve near-perfect structural accuracy and high execution robustness. This work establishes a foundation for generative digital twins that integrate visual reasoning and language understanding into executable industrial simulation systems.

Paperid: 5704, https://arxiv.org/pdf/2512.20276.pdf

Abstract:
Vision-Language-Action (VLA) models have emerged as a unified paradigm for robotic perception and control, enabling emergent generalization and long-horizon task execution. However, their deployment in dynamic, real-world environments is severely hin dered by high inference latency. While smooth robotic interaction requires control frequencies of 20 to 30 Hz, current VLA models typi cally operate at only 3-5 Hz on edge devices due to the memory bound nature of autoregressive decoding. Existing optimizations often require extensive retraining or compromise model accuracy. To bridge this gap, we introduce ActionFlow, a system-level inference framework tailored for resource-constrained edge plat forms. At the core of ActionFlow is a Cross-Request Pipelin ing strategy, a novel scheduler that redefines VLA inference as a macro-pipeline of micro-requests. The strategy intelligently batches memory-bound Decode phases with compute-bound Prefill phases across continuous time steps to maximize hardware utilization. Furthermore, to support this scheduling, we propose a Cross Request State Packed Forward operator and a Unified KV Ring Buffer, which fuse fragmented memory operations into efficient dense computations. Experimental results demonstrate that ActionFlow achieves a 2.55x improvement in FPS on the OpenVLA-7B model without retraining, enabling real-time dy namic manipulation on edge hardware. Our work is available at https://anonymous.4open.science/r/ActionFlow-1D47.

Paperid: 5705, https://arxiv.org/pdf/2512.17730.pdf

Abstract:
Recent advances in image generation have led to the widespread availability of highly realistic synthetic media, increasing the difficulty of reliable deepfake detection. A key challenge is generalization, as detectors trained on a narrow class of generators often fail when confronted with unseen models. In this work, we address the pressing need for generalizable detection by leveraging large vision-language models, specifically CLIP, to identify synthetic content across diverse generative techniques. First, we introduce Diff-Gen, a large-scale benchmark dataset comprising 100k diffusion-generated fakes that capture broad spectral artifacts unlike traditional GAN datasets. Models trained on Diff-Gen demonstrate stronger cross-domain generalization, particularly on previously unseen image generators. Second, we propose AdaptPrompt, a parameter-efficient transfer learning framework that jointly learns task-specific textual prompts and visual adapters while keeping the CLIP backbone frozen. We further show via layer ablation that pruning the final transformer block of the vision encoder enhances the retention of high-frequency generative artifacts, significantly boosting detection accuracy. Our evaluation spans 25 challenging test sets, covering synthetic content generated by GANs, diffusion models, and commercial tools, establishing a new state-of-the-art in both standard and cross-domain scenarios. We further demonstrate the framework's versatility through few-shot generalization (using as few as 320 images) and source attribution, enabling the precise identification of generator architectures in closed-set settings.

Paperid: 5706, https://arxiv.org/pdf/2512.13671.pdf

Abstract:
Industrial anomaly detection (IAD) is difficult due to the scarcity of normal reference samples and the subtle, localized nature of many defects. Single-pass vision-language models (VLMs) often overlook small abnormalities and lack explicit mechanisms to compare against canonical normal patterns. We propose AgentIAD, a tool-driven agentic framework that enables multi-stage visual inspection. The agent is equipped with a Perceptive Zoomer (PZ) for localized fine-grained analysis and a Comparative Retriever (CR) for querying normal exemplars when evidence is ambiguous. To teach these inspection behaviors, we construct structured perceptive and comparative trajectories from the MMAD dataset and train the model in two stages: supervised fine-tuning followed by reinforcement learning. A two-part reward design drives this process: a perception reward that supervises classification accuracy, spatial alignment, and type correctness, and a behavior reward that encourages efficient tool use. Together, these components enable the model to refine its judgment through step-wise observation, zooming, and verification. AgentIAD achieves a new state-of-the-art 97.62% classification accuracy on MMAD, surpassing prior MLLM-based approaches while producing transparent and interpretable inspection traces.

Paperid: 5707, https://arxiv.org/pdf/2512.13008.pdf

Abstract:
Accurate medical image analysis can greatly assist clinical diagnosis, but its effectiveness relies on high-quality expert annotations Obtaining pixel-level labels for medical images, particularly fundus images, remains costly and time-consuming. Meanwhile, despite the success of deep learning in medical imaging, the lack of interpretability limits its clinical adoption. To address these challenges, we propose TWLR, a two-stage framework for interpretable diabetic retinopathy (DR) assessment. In the first stage, a vision-language model integrates domain-specific ophthalmological knowledge into text embeddings to jointly perform DR grading and lesion classification, effectively linking semantic medical concepts with visual features. The second stage introduces an iterative severity regression framework based on weakly-supervised semantic segmentation. Lesion saliency maps generated through iterative refinement direct a progressive inpainting mechanism that systematically eliminates pathological features, effectively downgrading disease severity toward healthier fundus appearances. Critically, this severity regression approach achieves dual benefits: accurate lesion localization without pixel-level supervision and providing an interpretable visualization of disease-to-healthy transformations. Experimental results on the FGADR, DDR, and a private dataset demonstrate that TWLR achieves competitive performance in both DR classification and lesion segmentation, offering a more explainable and annotation-efficient solution for automated retinal image analysis.

Paperid: 5708, https://arxiv.org/pdf/2512.09441.pdf

Abstract:
Class-incremental learning requires a learning system to continually learn knowledge of new classes and meanwhile try to preserve previously learned knowledge of old classes. As current state-of-the-art methods based on Vision-Language Models (VLMs) still suffer from the issue of differentiating classes across learning tasks. Here a novel VLM-based continual learning framework for image classification is proposed. In this framework, task-specific adapters are added to the pre-trained and frozen image encoder to learn new knowledge, and a novel cross-task representation calibration strategy based on a mixture of light-weight projectors is used to help better separate all learned classes in a unified feature space, alleviating class confusion across tasks. In addition, a novel inference strategy guided by prediction uncertainty is developed to more accurately select the most appropriate image feature for class prediction. Extensive experiments on multiple datasets under various settings demonstrate the superior performance of our method compared to existing ones.

Paperid: 5709, https://arxiv.org/pdf/2512.06759.pdf

Abstract:
Understanding multi-image, multi-turn scenarios is a critical yet underexplored capability for Large Vision-Language Models (LVLMs). Existing benchmarks predominantly focus on static or horizontal comparisons -- e.g., spotting visual differences or assessing appropriateness -- while relying heavily on language cues. Such settings overlook progressive, context-dependent reasoning and the challenge of visual-to-visual inference. To bridge this gap, we present VisChainBench, a large-scale benchmark designed to rigorously evaluate LVLMs' ability to perform multi-step visual reasoning across sequential, interdependent tasks with minimal language guidance. VisChainBench contains 1,457 tasks spanning over 20,000 images across three diverse domains (e.g., daily scenarios, engineering troubleshooting), structured to mimic real-world decision-making processes. Uniquely, the benchmark is constructed using a multi-agent generation pipeline, ensuring high visual diversity and controlled language bias. All the benchmark data and code for benchmark construction are available for viewing and download via following Link: https://huggingface.co/datasets/eyehole/VisChainBench

Paperid: 5710, https://arxiv.org/pdf/2512.06022.pdf

Abstract:
Recent advances in video generation have achieved remarkable improvements in visual content fidelity. However, the absence of synchronized audio severely undermines immersive experience and restricts practical applications of these technologies. To address this challenge, several pioneering works have explored diffusion transformer architectures for generating plausible video-synchronized audio, including Kling-foley, HunyuanVideo-foley and Thinksound. Distinct from existing works, we introduce an autoregressive audio generation architecture (DreamFoley) that harnesses the capabilities of large vision-language models (VLMs) to jointly model sequential interactions among video, audio, and text modalities. Our approach features a dual-visual encoder module that effectively captures both audio-aligned and text-aligned visual features. Additionally, we employ a Residual Vector Quantization audio tokenizer with a delay-pattern generation scheme to balance the trade-off between training efficiency and audio quality. Moreover, we introduce the classifier-free guidance strategy into VLMs to bootstrap generated audio quality. Furthermore, we establish an efficient data production pipeline to scale audio-video-text triple collection. Finally, extensive experiments are conducted to validate the effectiveness of our model, achieving promising performance across popular benchmarks. We hope that the findings in this study provide a strong foundation for future video-to-audio generation research. We also release the previously missing audio-visual textual descriptions from the public benchmark, aiming to facilitate subsequent researchers in conducting more convenient and effective evaluations and comparisons.

Paperid: 5711, https://arxiv.org/pdf/2512.03848.pdf

Abstract:
Cardiac image analysis remains fragmented across tasks: anatomical segmentation, disease classification, and grounded clinical report generation are typically handled by separate networks trained under different data regimes. No existing framework unifies these objectives within a single architecture while retaining generalization across imaging modalities and datasets. We introduce PULSE, a multi-task vision-language framework built on self-supervised representations and optimized through a composite supervision strategy that balances region overlap learning, pixel wise classification fidelity, and boundary aware IoU refinement. A multi-scale token reconstruction decoder enables anatomical segmentation, while shared global representations support disease classification and clinically grounded text output allowing the model to transition from pixels to structures and finally clinical reasoning within one architecture. Unlike prior task-specific pipelines, PULSE learns task-invariant cardiac priors, generalizes robustly across datasets, and can be adapted to new imaging modalities with minimal supervision. This moves the field closer to a scalable, foundation style cardiac analysis framework.

Paperid: 5712, https://arxiv.org/pdf/2512.02517.pdf

Abstract:
The emergence of large vision-language models (VLMs) has significantly enhanced the efficiency and flexibility of geospatial interpretation. However, general-purpose VLMs remain suboptimal for remote sensing (RS) tasks. Existing geospatial VLMs typically adopt a unified modeling strategy and struggle to differentiate between task types and interpretation granularities, limiting their ability to balance local detail perception and global contextual understanding. In this paper, we present SkyMoE, a Mixture-of-Experts (MoE) vision-language model tailored for multimodal, multi-task RS interpretation. SkyMoE employs an adaptive router that generates task- and granularity-aware routing instructions, enabling specialized large language model experts to handle diverse sub-tasks. To further promote expert decoupling and granularity sensitivity, we introduce a context-disentangled augmentation strategy that creates contrastive pairs between local and global features, guiding experts toward level-specific representation learning. We also construct MGRS-Bench, a comprehensive benchmark covering multiple RS interpretation tasks and granularity levels, to evaluate generalization in complex scenarios. Extensive experiments on 21 public datasets demonstrate that SkyMoE achieves state-of-the-art performance across tasks, validating its adaptability, scalability, and superior multi-granularity understanding in remote sensing.

Paperid: 5713, https://arxiv.org/pdf/2512.02505.pdf

Abstract:
Autoregressive models are structurally misaligned with the inherently parallel nature of geospatial understanding, forcing a rigid sequential narrative onto scenes and fundamentally hindering the generation of structured and coherent outputs. We challenge this paradigm by reframing geospatial generation as a parallel refinement process, enabling a holistic, coarse-to-fine synthesis that resolves all semantic elements simultaneously. To operationalize this, we introduce GeoDiT, the first diffusion-based vision-language model tailored for the geospatial domain. Extensive experiments demonstrate that GeoDiT establishes a new state-of-the-art on benchmarks requiring structured, object-centric outputs. It achieves significant gains in image captioning, visual grounding, and multi-object detection, precisely the tasks where autoregressive models falter. Our work validates that aligning the generative process with the data's intrinsic structure is key to unlocking superior performance in complex geospatial analysis.

Paperid: 5714, https://arxiv.org/pdf/2512.02263.pdf

Abstract:
2.5D effects, such as occlusion and perspective foreshortening, enhance visual dynamics and realism by incorporating 3D depth cues into 2D designs. However, creating such effects remains challenging and labor-intensive due to the complexity of depth perception. We introduce DepthScape, a human-AI collaborative system that facilitates 2.5D effect creation by directly placing design elements into 3D reconstructions. Using monocular depth reconstruction, DepthScape transforms images into 3D reconstructions where visual contents are placed to automatically achieve realistic occlusion and perspective foreshortening. To further simplify 3D placement through a 2D viewport, DepthScape uses a vision-language model to analyze source images and extract key visual components as content anchors for direct manipulation editing. We evaluate DepthScape with nine participants of varying design backgrounds, confirming the effectiveness of our creation pipeline. We also test on 100 professional stock images to assess robustness, and conduct an expert evaluation that confirms the quality of DepthScape's results.

Paperid: 5715, https://arxiv.org/pdf/2512.00887.pdf

Abstract:
Remote sensing image captioning has advanced rapidly through encoder--decoder models, although the reliance on large annotated datasets and the focus on English restricts global applicability. To address these limitations, we propose the first training-free multilingual approach, based on retrieval-augmented prompting. For a given aerial image, we employ a domain-adapted SigLIP2 encoder to retrieve related captions and few-shot examples from a datastore, which are then provided to a language model. We explore two variants: an image-blind setup, where a multilingual Large Language Model (LLM) generates the caption from textual prompts alone, and an image-aware setup, where a Vision--Language Model (VLM) jointly processes the prompt and the input image. To improve the coherence of the retrieved content, we introduce a graph-based re-ranking strategy using PageRank on a graph of images and captions. Experiments on four benchmark datasets across ten languages demonstrate that our approach is competitive with fully supervised English-only systems and generalizes to other languages. Results also highlight the importance of re-ranking with PageRank, yielding up to 35% improvements in performance metrics. Additionally, it was observed that while VLMs tend to generate visually grounded but lexically diverse captions, LLMs can achieve stronger BLEU and CIDEr scores. Lastly, directly generating captions in the target language consistently outperforms other translation-based strategies. Overall, our work delivers one of the first systematic evaluations of multilingual, training-free captioning for remote sensing imagery, advancing toward more inclusive and scalable multimodal Earth observation systems.

Paperid: 5716, https://arxiv.org/pdf/2512.00087.pdf

Abstract:
Observation of classroom interactions can provide concrete feedback to teachers, but current methods rely on manual annotation, which is resource-intensive and hard to scale. This work explores AI-driven analysis of classroom recordings, focusing on multimodal instructional activity and discourse recognition as a foundation for actionable feedback. Using a densely annotated dataset of 164 hours of video and 68 lesson transcripts, we design parallel, modality-specific pipelines. For video, we evaluate zero-shot multimodal LLMs, fine-tuned vision-language models, and self-supervised video transformers on 24 activity labels. For transcripts, we fine-tune a transformer-based classifier with contextualized inputs and compare it against prompting-based LLMs on 19 discourse labels. To handle class imbalance and multi-label complexity, we apply per-label thresholding, context windows, and imbalance-aware loss functions. The results show that fine-tuned models consistently outperform prompting-based approaches, achieving macro-F1 scores of 0.577 for video and 0.460 for transcripts. These results demonstrate the feasibility of automated classroom analysis and establish a foundation for scalable teacher feedback systems.

Paperid: 5717, https://arxiv.org/pdf/2511.22961.pdf

Abstract:
Recent advances in large vision-language models (VLMs) have shown significant promise for 3D scene understanding. Existing VLM-based approaches typically align 3D scene features with the VLM's embedding space. However, this implicit alignment often yields suboptimal performance due to the scarcity of 3D data and the inherent complexity of spatial relationships in 3D environments. To address these limitations, we propose a novel hierarchical multimodal representation for 3D scene reasoning that explicitly aligns with VLMs at the input space by leveraging both multi-view images and text descriptions. The text descriptions capture spatial relationships by referencing the 3D coordinates of detected objects, while the multi-view images include a top-down perspective and four directional views (forward, left, right, and backward), ensuring comprehensive scene coverage. Additionally, we introduce a hierarchical feature representation that aggregates patch-level image features into view-level and scene-level representations, enabling the model to reason over both local and global scene context. Experimental results on both situated 3D Q&A and general 3D Q&A benchmarks demonstrate the effectiveness of our approach.

Paperid: 5718, https://arxiv.org/pdf/2511.21717.pdf

Abstract:
Multimodal Large Language Models are primarily trained and evaluated on aligned image-text pairs, which leaves their ability to detect and resolve real-world inconsistencies largely unexplored. In open-domain applications visual and textual cues often conflict, requiring models to perform structured reasoning beyond surface-level alignment. We introduce CrossCheck-Bench, a diagnostic benchmark for evaluating contradiction detection in multimodal inputs. The benchmark adopts a hierarchical task framework covering three levels of reasoning complexity and defines seven atomic capabilities essential for resolving cross-modal inconsistencies. CrossCheck-Bench includes 15k question-answer pairs sourced from real-world artifacts with synthetically injected contradictions. The dataset is constructed through a multi-stage annotation pipeline involving more than 450 expert hours to ensure semantic validity and calibrated difficulty across perception, integration, and reasoning. We evaluate 13 state-of-the-art vision-language models and observe a consistent performance drop as tasks shift from perceptual matching to logical contradiction detection. Most models perform well on isolated entity recognition but fail when multiple clues must be synthesized for conflict reasoning. Capability-level analysis further reveals uneven skill acquisition, especially in tasks requiring multi-step inference or rule-based validation. Additional probing shows that conventional prompting strategies such as Chain-of-Thought and Set-of-Mark yield only marginal gains. By contrast, methods that interleave symbolic reasoning with grounded visual processing achieve more stable improvements. These results highlight a persistent bottleneck in multimodal reasoning and suggest new directions for building models capable of robust cross-modal verification.

Paperid: 5719, https://arxiv.org/pdf/2511.20965.pdf

Abstract:
Traffic cameras are essential in urban areas, playing a crucial role in intelligent transportation systems. Multiple cameras at intersections enhance law enforcement capabilities, traffic management, and pedestrian safety. However, efficiently managing and analyzing multi-camera feeds poses challenges due to the vast amount of data. Analyzing such huge video data requires advanced analytical tools. While Large Language Models (LLMs) like ChatGPT, equipped with retrieval-augmented generation (RAG) systems, excel in text-based tasks, integrating them into traffic video analysis demands converting video data into text using a Vision-Language Model (VLM), which is time-consuming and delays the timely utilization of traffic videos for generating insights and investigating incidents. To address these challenges, we propose TrafficLens, a tailored algorithm for multi-camera traffic intersections. TrafficLens employs a sequential approach, utilizing overlapping coverage areas of cameras. It iteratively applies VLMs with varying token limits, using previous outputs as prompts for subsequent cameras, enabling rapid generation of detailed textual descriptions while reducing processing time. Additionally, TrafficLens intelligently bypasses redundant VLM invocations through an object-level similarity detector. Experimental results with real-world datasets demonstrate that TrafficLens reduces video-to-text conversion time by up to $4\times$ while maintaining information accuracy.

Paperid: 5720, https://arxiv.org/pdf/2511.18978.pdf

Abstract:
Accurate annotation of cutaneous neoplasm biopsies represents a major challenge due to their wide morphological variability, overlapping histological patterns, and the subtle distinctions between benign and malignant lesions. Vision-language foundation models (VLMs), pre-trained on paired image-text corpora, learn joint representations that bridge visual features and diagnostic terminology, enabling zero-shot localization and classification of tissue regions without pixel-level labels. However, most existing VLM applications in histopathology remain limited to slide-level tasks or rely on coarse interactive prompts, and they struggle to produce fine-grained segmentations across gigapixel whole-slide images (WSIs). In this work, we introduce a zero-shot visual-language segmentation pipeline for whole-slide images (ZEUS), a fully automated, zero-shot segmentation framework that leverages class-specific textual prompt ensembles and frozen VLM encoders to generate high-resolution tumor masks in WSIs. By partitioning each WSI into overlapping patches, extracting visual embeddings, and computing cosine similarities against text prompts, we generate a final segmentation mask. We demonstrate competitive performance on two in-house datasets, primary spindle cell neoplasms and cutaneous metastases, highlighting the influence of prompt design, domain shifts, and institutional variability in VLMs for histopathology. ZEUS markedly reduces annotation burden while offering scalable, explainable tumor delineation for downstream diagnostic workflows.

Paperid: 5721, https://arxiv.org/pdf/2511.18504.pdf

Abstract:
The demand for edge AI in vision-language tasks requires models that achieve real-time performance on resource-constrained devices with limited power and memory. This paper proposes two adaptive compression techniques -- Sparse Temporal Token Fusion (STTF) and Adaptive Neural Compression (ANC) -- that integrate algorithmic innovations with hardware-aware optimizations. Unlike previous approaches relying on static pruning or uniform scaling, STTF dynamically reuses visual tokens through event-driven change detection, while ANC conditionally activates encoder branches via a learned router, enabling fine-grained adaptation to scene complexity. Our 3B-parameter TinyGPT-STTF achieves CIDEr 131.2, BLEU-4 0.38, METEOR 0.31, and ROUGE-L 0.56 on the COCO 2017 test set, surpassing LLaVA-1.5 7B by 17.6 CIDEr points while using 2.3x fewer parameters and 62x fewer on-device FLOPs. TinyGPT-ANC reaches CIDEr 128.5. On event-based vision tasks, STTF reduces average token count by 84% (from 196 to 31 tokens) while preserving 95.6% accuracy on the DVS128 Gesture dataset, and ANC cuts FLOPs by up to 90% in low-motion scenes. Compared to strong baselines, our models improve accuracy by up to 4.4% and reduce latency by up to 13x. These results enable efficient deployment of capable vision-language models on real-world edge devices.

Paperid: 5722, https://arxiv.org/pdf/2511.18415.pdf

Abstract:
Vision-language models (VLMs) possess rich knowledge but often fail on hierarchical understanding tasks, where the goal is to predict a coarse-to-fine taxonomy path that remains consistent across all levels. We compare three inference paradigms for hierarchical VQA and find that stepwise reasoning, when conditioned on prior answers, significantly outperforms single-pass prompting. Further analysis indicates that the main limitation of current VLMs is their inability to maintain cross-level state, rather than a lack of taxonomic knowledge. Motivated by this diagnosis, we propose Self-Elicited Knowledge Distillation (SEKD), which requires no human labels or external tools: the same VLM is prompted to reason step by step and act as a teacher by exposing its hard labels, soft distributions, and decoder hidden states, while a single-pass student distills these signals. The student VLM remains efficient while approaching the accuracy of its multi-step teacher. It improves in-domain path consistency (HCA) by up to +29.50 percentage points, raises zero-shot HCA on an unseen taxonomy from 4.15% to 42.26%, and yields gains on challenging mathematical benchmarks. Because all supervision is self-elicited, SEKD scales to new taxonomies and datasets without annotation cost, providing a practical route to imbue compact VLMs with dependency-aware multi-step reasoning.

Paperid: 5723, https://arxiv.org/pdf/2511.15578.pdf

Abstract:
With the increasing prevalence of video content, effectively understanding and answering questions about long form videos has become essential for numerous applications. Although large vision language models (LVLMs) have enhanced performance, they often face challenges with nuanced queries that demand both a comprehensive understanding and detailed analysis. To overcome these obstacles, we introduce AVATAAR, a modular and interpretable framework that combines global and local video context, along with a Pre Retrieval Thinking Agent and a Rethink Module. AVATAAR creates a persistent global summary and establishes a feedback loop between the Rethink Module and the Pre Retrieval Thinking Agent, allowing the system to refine its retrieval strategies based on partial answers and replicate human-like iterative reasoning. On the CinePile benchmark, AVATAAR demonstrates significant improvements over a baseline, achieving relative gains of +5.6% in temporal reasoning, +5% in technical queries, +8% in theme-based questions, and +8.2% in narrative comprehension. Our experiments confirm that each module contributes positively to the overall performance, with the feedback loop being crucial for adaptability. These findings highlight AVATAAR's effectiveness in enhancing video understanding capabilities. Ultimately, AVATAAR presents a scalable solution for long-form Video Question Answering (QA), merging accuracy, interpretability, and extensibility.

Paperid: 5724, https://arxiv.org/pdf/2511.14293.pdf

Abstract:
Recent audio-language models have shown impressive performance across a wide range of audio tasks and are increasingly capable of handling long audio inputs. However, the computing costs in these models heavily depend on sequence length, which can become very large given the nature of audio data. In the vision-language domain, token pruning methods have proven effective in reducing token counts while preserving strong performance on standard benchmarks. In this work, we investigate the relevance and effectiveness of such token selection strategies in the context of audio-language models. We also improve them by proposing a lightweight strategy that takes the time dimension into account. While retaining only a quarter of the initial tokens, our approach results in a relative maximum decrease of 2% in CIDEr on Clotho v2 and a relative maximum decrease of 4% in accuracy on MMAU.

Paperid: 5725, https://arxiv.org/pdf/2511.14120.pdf

Abstract:
Pedestrian-vehicle incidents remain a critical urban safety challenge, with pedestrians accounting for over 20% of global traffic fatalities. Although existing video-based systems can detect when incidents occur, they provide little insight into how these events unfold across the distinct cognitive phases of pedestrian behavior. Recent vision-language models (VLMs) have shown strong potential for video understanding, but they remain limited in that they typically process videos in isolation, without explicit temporal structuring or multi-view integration. This paper introduces Multi-view Phase-aware Pedestrian-Vehicle Incident Reasoning (MP-PVIR), a unified framework that systematically processes multi-view video streams into structured diagnostic reports through four stages: (1) event-triggered multi-view video acquisition, (2) pedestrian behavior phase segmentation, (3) phase-specific multi-view reasoning, and (4) hierarchical synthesis and diagnostic reasoning. The framework operationalizes behavioral theory by automatically segmenting incidents into cognitive phases, performing synchronized multi-view analysis within each phase, and synthesizing results into causal chains with targeted prevention strategies. Particularly, two specialized VLMs underpin the MP-PVIR pipeline: TG-VLM for behavioral phase segmentation (mIoU = 0.4881) and PhaVR-VLM for phase-aware multi-view analysis, achieving a captioning score of 33.063 and up to 64.70% accuracy on question answering. Finally, a designated large language model is used to generate comprehensive reports detailing scene understanding, behavior interpretation, causal reasoning, and prevention recommendations. Evaluation on the Woven Traffic Safety dataset shows that MP-PVIR effectively translates multi-view video data into actionable insights, advancing AI-driven traffic safety analytics for vehicle-infrastructure cooperative systems.

Paperid: 5726, https://arxiv.org/pdf/2511.13891.pdf

Abstract:
Among soil erosion problems, Ephemeral Gullies are one of the most concerning phenomena occurring in agricultural fields. Their short temporal cycles increase the difficulty in automatically detecting them using classical computer vision approaches and remote sensing. Also, due to scarcity of and the difficulty in producing accurate labeled data, automatic detection of ephemeral gullies using Machine Learning is limited to zero-shot approaches which are hard to implement. To overcome these challenges, we present the first weakly supervised pipeline for detection of ephemeral gullies. Our method relies on remote sensing and uses Vision Language Models (VLMs) to drastically reduce the labor-intensive task of manual labeling. In order to achieve that, the method exploits: 1) the knowledge embedded in the VLM's pretraining; 2) a teacher-student model where the teacher learns from noisy labels coming from the VLMs, and the student learns by weak supervision using teacher-generate labels and a noise-aware loss function. We also make available the first-of-its-kind dataset for semi-supervised detection of ephemeral gully from remote-sensed images. The dataset consists of a number of locations labeled by a group of soil and plant scientists, as well as a large number of unlabeled locations. The dataset represent more than 18,000 high-resolution remote-sensing images obtained over the course of 13 years. Our experimental results demonstrate the validity of our approach by showing superior performances compared to VLMs and the label model itself when using weak supervision to train an student model. The code and dataset for this work are made publicly available.

Paperid: 5727, https://arxiv.org/pdf/2511.13644.pdf

Abstract:
Long-form video question answering (VQA) overwhelms current vision-language models (VLMs) because attention and key-value (KV) caches grow with runtime, forcing either expensive inference or near-sighted sliding windows. We introduce CacheFlow, a training-free pipeline that pairs Dynamic Token Dropping (DTD) with a compressive long-term memory. DTD prunes per-patch tokens online via cosine similarity to the previous frame, and surviving tokens are packed into fixed-size blocks. This online, per-frame processing makes our approach fundamentally suited for live streaming VQA. As blocks are processed, each one's keys are summarized by a tiny recurrent encoder to form a retrieval index, while the block's full KV pairs are offloaded and later rehydrated for generation, preserving answer fidelity. At inference, a consensus-based retrieval mechanism retrieves only the Top-K most relevant blocks and attends over both the retrieved and local context for precise, long-range reasoning. CacheFlow is drop-in, architecture-agnostic, and requires no fine-tuning. Experiments on both offline and streaming VQA benchmarks demonstrate that CacheFlow outperforms current strong baselines, while processing up to 87% less tokens. Our dual approach enables VLMs to be both efficient and context-aware, paving the way for practical long-form video understanding.

Paperid: 5728, https://arxiv.org/pdf/2511.13458.pdf

Abstract:
With the growing deployment of Vision-Language Models (VLMs), pre-trained on large image-text and video-text datasets, it is critical to equip users with the tools to discern when to trust these systems. However, examining how user trust in VLMs builds and evolves remains an open problem. This problem is exacerbated by the increasing reliance on AI models as judges for experimental validation, to bypass the cost and implications of running participatory design studies directly with users. Following a user-centred approach, this paper presents preliminary results from a workshop with prospective VLM users. Insights from this pilot workshop inform future studies aimed at contextualising trust metrics and strategies for participants' engagement to fit the case of user-VLM interaction.

Paperid: 5729, https://arxiv.org/pdf/2511.12215.pdf

Abstract:
Medical vision-language pre-training (VLP) offers significant potential for advancing medical image understanding by leveraging paired image-report data. However, existing methods are limited by Fa}lse Negatives (FaNe) induced by semantically similar texts and insufficient fine-grained cross-modal alignment. To address these limitations, we propose FaNe, a semantic-enhanced VLP framework. To mitigate false negatives, we introduce a semantic-aware positive pair mining strategy based on text-text similarity with adaptive normalization. Furthermore, we design a text-conditioned sparse attention pooling module to enable fine-grained image-text alignment through localized visual representations guided by textual cues. To strengthen intra-modal discrimination, we develop a hard-negative aware contrastive loss that adaptively reweights semantically similar negatives. Extensive experiments on five downstream medical imaging benchmarks demonstrate that FaNe achieves state-of-the-art performance across image classification, object detection, and semantic segmentation, validating the effectiveness of our framework.

Paperid: 5730, https://arxiv.org/pdf/2511.12180.pdf

Abstract:
Contrastive learning has emerged as a cornerstone of unsupervised representation learning across vision, language, and graph domains, with InfoNCE as its dominant objective. Despite its empirical success, the theoretical underpinnings of InfoNCE remain limited. In this work, we introduce an explicit feature space to model augmented views of samples and a transition probability matrix to capture data augmentation dynamics. We demonstrate that InfoNCE optimizes the probability of two views sharing the same source toward a constant target defined by this matrix, naturally inducing feature clustering in the representation space. Leveraging this insight, we propose Scaled Convergence InfoNCE (SC-InfoNCE), a novel loss function that introduces a tunable convergence target to flexibly control feature similarity alignment. By scaling the target matrix, SC-InfoNCE enables flexible control over feature similarity alignment, allowing the training objective to better match the statistical properties of downstream data. Experiments on benchmark datasets, including image, graph, and text tasks, show that SC-InfoNCE consistently achieves strong and reliable performance across diverse domains.

Paperid: 5731, https://arxiv.org/pdf/2511.11301.pdf

Abstract:
Large Vision-Language Models (LVLMs) exhibit powerful reasoning capabilities but suffer sophisticated jailbreak vulnerabilities. Fundamentally, aligning LVLMs is not just a safety challenge but a problem of economic efficiency. Current alignment methods struggle with the trade-off between safety, utility, and operational costs. Critically, a focus solely on final outputs (process-blindness) wastes significant computational budget on unsafe deliberation. This flaw allows harmful reasoning to be disguised with benign justifications, thereby circumventing simple additive safety scores. To address this, we propose EcoAlign, an inference-time framework that reframes alignment as an economically rational search by treating the LVLM as a boundedly rational agent. EcoAlign incrementally expands a thought graph and scores actions using a forward-looking function (analogous to net present value) that dynamically weighs expected safety, utility, and cost against the remaining budget. To prevent deception, path safety is enforced via the weakest-link principle. Extensive experiments across 3 closed-source and 2 open-source models on 6 datasets show that EcoAlign matches or surpasses state-of-the-art safety and utility at a lower computational cost, thereby offering a principled, economical pathway to robust LVLM alignment.

Paperid: 5732, https://arxiv.org/pdf/2511.10958.pdf

Abstract:
Dynamic facial expression recognition (DFER) aims to identify emotional states by modeling the temporal changes in facial movements across video sequences. A key challenge in DFER is the many-to-one labeling problem, where a video composed of numerous frames is assigned a single emotion label. A common strategy to mitigate this issue is to formulate DFER as a Multiple Instance Learning (MIL) problem. However, MIL-based approaches inherently suffer from the visual diversity of emotional expressions and the complexity of temporal dynamics. To address this challenge, we propose TG-DFER, a text-guided weakly supervised framework that enhances MIL-based DFER by incorporating semantic guidance and coherent temporal modeling. We incorporate a vision-language pre-trained (VLP) model is integrated to provide semantic guidance through fine-grained textual descriptions of emotional context. Furthermore, we introduce visual prompts, which align enriched textual emotion labels with visual instance features, enabling fine-grained reasoning and frame-level relevance estimation. In addition, a multi-grained temporal network is designed to jointly capture short-term facial dynamics and long-range emotional flow, ensuring coherent affective understanding across time. Extensive results demonstrate that TG-DFER achieves improved generalization, interpretability, and temporal sensitivity under weak supervision.

Paperid: 5733, https://arxiv.org/pdf/2511.04357.pdf

Abstract:
Deploying autonomous robots that can learn new skills from demonstrations is an important challenge of modern robotics. Existing solutions often apply end-to-end imitation learning with Vision-Language Action (VLA) models or symbolic approaches with Action Model Learning (AML). On the one hand, current VLA models are limited by the lack of high-level symbolic planning, which hinders their abilities in long-horizon tasks. On the other hand, symbolic approaches in AML lack generalization and scalability perspectives. In this paper we present a new neuro-symbolic approach, GraSP-VLA, a framework that uses a Continuous Scene Graph representation to generate a symbolic representation of human demonstrations. This representation is used to generate new planning domains during inference and serves as an orchestrator for low-level VLA policies, scaling up the number of actions that can be reproduced in a row. Our results show that GraSP-VLA is effective for modeling symbolic representations on the task of automatic planning domain generation from observations. In addition, results on real-world experiments show the potential of our Continuous Scene Graph representation to orchestrate low-level VLA policies in long-horizon tasks.

Paperid: 5734, https://arxiv.org/pdf/2511.01463.pdf

Abstract:
The expansion of instruction-tuning data has enabled foundation language models to exhibit improved instruction adherence and superior performance across diverse downstream tasks. Semantically-rich 3D human motion is being progressively integrated with these foundation models to enhance multimodal understanding and cross-modal generation capabilities. However, the modality gap between human motion and text raises unresolved concerns about catastrophic forgetting during this integration. In addition, developing autoregressive-compatible pose representations that preserve generalizability across heterogeneous downstream tasks remains a critical technical barrier. To address these issues, we propose the Human Motion-Vision-Language Model (HMVLM), a unified framework based on the Mixture of Expert Low-Rank Adaption(MoE LoRA) strategy. The framework leverages the gating network to dynamically allocate LoRA expert weights based on the input prompt, enabling synchronized fine-tuning of multiple tasks. To mitigate catastrophic forgetting during instruction-tuning, we introduce a novel zero expert that preserves the pre-trained parameters for general linguistic tasks. For pose representation, we implement body-part-specific tokenization by partitioning the human body into different joint groups, enhancing the spatial resolution of the representation. Experiments show that our method effectively alleviates knowledge forgetting during instruction-tuning and achieves remarkable performance across diverse human motion downstream tasks.

Paperid: 5735, https://arxiv.org/pdf/2511.00821.pdf

Abstract:
Vision-Language Models (VLMs) have demonstrated strong performance across various multimodal tasks, where position encoding plays a vital role in modeling both the sequential structure of textual information and the spatial structure of visual information. However, current VLMs commonly adopt modality-unified 1D or 2D positional indexing strategies, which treat textual and visual tokens uniformly without accounting for their distinct structural properties and sequential continuity for text and spatial coherence for vision. To address this limitation, we propose OMEGA, a novel position encoding framework that employs Modality-Specific Position Encoding (MSPE) to assign positional indices while preserving the inherent structures of each modality across separate coordinate dimensions. Additionally, to align the information density of multimodal data in the positional index space, OMEGA introduces Global Adaptive Encoding Step Scaling (GAESS), which adaptively adjusts the position encoding step size of visual tokens based on the embedding entropy of both modalities. Experimental results demonstrate that OMEGA consistently enhances VLM performance across diverse architectures and VQA benchmarks. On visual-intensive tasks, OMEGA achieves up to 3.43% improvement over baseline position encoding strategies on Qwen2.5-VL-3B, with consistent gains observed across larger models including Qwen2.5-VL-7B and LLaVA-v1.5-7B.

Paperid: 5736, https://arxiv.org/pdf/2511.00613.pdf

Abstract:
How far are deep models from real-world video anomaly understanding (VAU)? Current works typically emphasize on detecting unexpected occurrences deviated from normal patterns or comprehending anomalous events with interpretable descriptions. However, they exhibit only a superficial comprehension of real-world anomalies, with limited breadth in complex principles and subtle context that distinguish the anomalies from normalities, e.g., climbing cliffs with safety gear vs. without it. To this end, we introduce CueBench, the first of its kind Benchmark, devoted to Context-aware video anomalies within a Unified Evaluation framework. We comprehensively establish an event-centric hierarchical taxonomy that anchors two core event types: 14 conditional and 18 absolute anomaly events, defined by their refined semantics from diverse contexts across 174 scenes and 198 attributes. Based on this, we propose to unify and benchmark context-aware VAU with various challenging tasks across recognition, temporal grounding, detection, and anticipation. This also serves as a rigorous and fair probing evaluation suite for generative-discriminative as well as generalized-specialized vision-language models (VLMs). To address the challenges underlying CueBench, we further develop Cue-R1 based on R1-style reinforcement fine-tuning with verifiable, task-aligned, and hierarchy-refined rewards in a unified generative manner. Extensive results on CueBench reveal that, existing VLMs are still far from satisfactory real-world anomaly understanding, while our Cue-R1 surpasses these state-of-the-art approaches by over 24% on average.

Paperid: 5737, https://arxiv.org/pdf/2510.27033.pdf

Abstract:
Visual reasoning, particularly spatial reasoning, is a challenging cognitive task that requires understanding object relationships and their interactions within complex environments, especially in robotics domain. Existing vision_language models (VLMs) excel at perception tasks but struggle with fine-grained spatial reasoning due to their implicit, correlation-driven reasoning and reliance solely on images. We propose a novel neuro_symbolic framework that integrates both panoramic-image and 3D point cloud information, combining neural perception with symbolic reasoning to explicitly model spatial and logical relationships. Our framework consists of a perception module for detecting entities and extracting attributes, and a reasoning module that constructs a structured scene graph to support precise, interpretable queries. Evaluated on the JRDB-Reasoning dataset, our approach demonstrates superior performance and reliability in crowded, human_built environments while maintaining a lightweight design suitable for robotics and embodied AI applications.

Paperid: 5738, https://arxiv.org/pdf/2510.21879.pdf

Abstract:
Recent years have witnessed an increasing interest in image-text contrastive modeling, exemplified by models such as Contrastive Language-Image Pretraining (CLIP). In this paper, we propose the TernaryCLIP, a lightweight computational framework that converts connection weights of both vision and text encoders of CLIP into the ternary format, instead of full-precision or floating ones. TernaryCLIP incorporates quantization-aware training and distillation modules, preventing precision degradation and enabling low-cost and high-efficiency computations. Comprehensive experiments demonstrate that TernaryCLIP can achieve up to 99\% ternarized weights with 1.58-bit representation, 16.98 $\times$ compression ratio, 2.3 $\times$ inference acceleration, 16 $\times$ storage reduction, 10 $\times$ memory optimization, and 60\% sparsity while maintaining promising performance on zero-shot image classification and image-text retrieval tasks across 41 commonly used datasets. Our work highlights the feasibility of extreme quantization for large multimodal models, supporting effective and efficient deployment on resource-constrained devices. The model and code can be accessed from Hugging Face and GitHub.

Paperid: 5739, https://arxiv.org/pdf/2510.21808.pdf

Abstract:
The high cost of data annotation has spurred research on training deep learning models in data-limited scenarios. Existing paradigms, however, fail to balance cross-domain transfer and cross-category generalization, giving rise to the demand for Domain-Adaptive Zero-Shot Learning (DAZSL). Although vision-language models (e.g., CLIP) have inherent advantages in the DAZSL field, current studies do not fully exploit their potential. Applying CLIP to DAZSL faces two core challenges: inefficient cross-category knowledge transfer due to the lack of semantic relation guidance, and degraded cross-modal alignment during target domain fine-tuning. To address these issues, we propose a Semantic Relation-Enhanced CLIP (SRE-CLIP) Adapter framework, integrating a Semantic Relation Structure Loss and a Cross-Modal Alignment Retention Strategy. As the first CLIP-based DAZSL method, SRE-CLIP achieves state-of-the-art performance on the I2AwA and I2WebV benchmarks, significantly outperforming existing approaches.

Paperid: 5740, https://arxiv.org/pdf/2510.21807.pdf

Abstract:
Recent breakthroughs in reasoning models have markedly advanced the reasoning capabilities of large language models, particularly via training on tasks with verifiable rewards. Yet, a significant gap persists in their adaptation to real world multimodal scenarios, most notably, vision language tasks, due to a heavy focus on single modal language settings. While efforts to transplant reinforcement learning techniques from NLP to VLMs have emerged, these approaches often remain confined to perception centric tasks or reduce images to textual summaries, failing to fully exploit visual context and commonsense knowledge, ultimately constraining the generalization of reasoning capabilities across diverse multimodal environments. To address this limitation, we introduce a novel fine tuning task, Masked Prediction via Context and Commonsense, which forces models to integrate visual context and commonsense reasoning by reconstructing semantically meaningful content from occluded images, thereby laying the foundation for generalized reasoning. To systematically evaluate the model performance in generalized reasoning, we developed a specialized evaluation benchmark, MPCC Eval, and employed various fine tuning strategies to guide reasoning. Among these, we introduced an innovative training method, Reinforcement Fine tuning with Prior Sampling, which not only enhances model performance but also improves its generalized reasoning capabilities in OOD and cross task scenarios.

Paperid: 5741, https://arxiv.org/pdf/2510.21806.pdf

Abstract:
With the rapid growth of video data, text-video retrieval technology has become increasingly important in numerous application scenarios such as recommendation and search. Early text-video retrieval methods suffer from two critical drawbacks: first, they heavily rely on large-scale annotated video-text pairs, leading to high data acquisition costs; second, there is a significant modal gap between video and text features, which limits cross-modal alignment accuracy. With the development of vision-language model, adapting CLIP to video tasks has attracted great attention. However, existing adaptation methods generally lack enhancement for dynamic video features and fail to effectively suppress static redundant features. To address this issue, this paper proposes FDA-CLIP (Frame Difference Alpha-CLIP), which is a concise CLIP-based training framework for text-video alignment. Specifically, the method uses frame differences to generate dynamic region masks, which are input into Alpha-CLIP as an additional Alpha channel. This proactively guides the model to focus on semantically critical dynamic regions while suppressing static background redundancy. Experiments demonstrate that frame difference-guided video semantic encoding can effectively balance retrieval efficiency and accuracy.

Paperid: 5742, https://arxiv.org/pdf/2510.21679.pdf

Abstract:
Companies spend large amounts of money on public relations campaigns to project a positive brand image. However, sometimes there is a mismatch between what they say and what they do. Oil & gas companies, for example, are accused of "greenwashing" with imagery of climate-friendly initiatives. Understanding the framing, and changes in framing, at scale can help better understand the goals and nature of public relations campaigns. To address this, we introduce a benchmark dataset of expert-annotated video ads obtained from Facebook and YouTube. The dataset provides annotations for 13 framing types for more than 50 companies or advocacy groups across 20 countries. Our dataset is especially designed for the evaluation of vision-language models (VLMs), distinguishing it from past text-only framing datasets. Baseline experiments show some promising results, while leaving room for improvement for future work: GPT-4.1 can detect environmental messages with 79% F1 score, while our best model only achieves 46% F1 score on identifying framing around green innovation. We also identify challenges that VLMs must address, such as implicit framing, handling videos of various lengths, or implicit cultural backgrounds. Our dataset contributes to research in multimodal analysis of strategic communication in the energy sector.

Paperid: 5743, https://arxiv.org/pdf/2510.21424.pdf

Abstract:
As generative AI continues to evolve, Vision Language Models (VLMs) have emerged as promising tools in various healthcare applications. One area that remains relatively underexplored is their use in human activity recognition (HAR) for remote health monitoring. VLMs offer notable strengths, including greater flexibility and the ability to overcome some of the constraints of traditional deep learning models. However, a key challenge in applying VLMs to HAR lies in the difficulty of evaluating their dynamic and often non-deterministic outputs. To address this gap, we introduce a descriptive caption data set and propose comprehensive evaluation methods to evaluate VLMs in HAR. Through comparative experiments with state-of-the-art deep learning models, our findings demonstrate that VLMs achieve comparable performance and, in some cases, even surpass conventional approaches in terms of accuracy. This work contributes a strong benchmark and opens new possibilities for the integration of VLMs into intelligent healthcare systems.

Paperid: 5744, https://arxiv.org/pdf/2510.18433.pdf

Abstract:
We introduce ImageGem, a dataset for studying generative models that understand fine-grained individual preferences. We posit that a key challenge hindering the development of such a generative model is the lack of in-the-wild and fine-grained user preference annotations. Our dataset features real-world interaction data from 57K users, who collectively have built 242K customized LoRAs, written 3M text prompts, and created 5M generated images. With user preference annotations from our dataset, we were able to train better preference alignment models. In addition, leveraging individual user preference, we investigated the performance of retrieval models and a vision-language model on personalized image retrieval and generative model recommendation. Finally, we propose an end-to-end framework for editing customized diffusion models in a latent weight space to align with individual user preferences. Our results demonstrate that the ImageGem dataset enables, for the first time, a new paradigm for generative model personalization.

Paperid: 5745, https://arxiv.org/pdf/2510.18117.pdf

Abstract:
As the field continues its push for ever more resources, this work turns the spotlight on a critical question: how can vision-language models (VLMs) be adapted to thrive in low-resource, budget-constrained settings? While large VLMs offer strong performance, they are impractical to deploy in such settings. Small VLMs, on the other hand, are efficient but typically require costly fine-tuning to close the performance gap with larger models in the deployment domain. Inspired by the in-context learning framework, we propose an online In-Context Distillation (ICD) method, in which a small VLM collaborates with a stronger teacher model at inference time, distilling its knowledge via sparse demonstrations to efficiently bridge the gap between them. Our method is built on an in-depth analysis that identifies the scale and the choice of models for which vision-language ICL is currently feasible, and demonstrates the advantage of ICL over fine-tuning under constrained compute budgets. We enhance our method with a novel cross-modal demonstration selection strategy, teacher test-time scaling to reduce noise, and student uncertainty conditioning to dynamically populate a demonstration pool and minimize teacher queries. Our ICD method significantly boosts the performance of small models (up to 33%) using scarce teacher annotations (as low as 4%), and competes with the teacher's zero-shot performance.

Paperid: 5746, https://arxiv.org/pdf/2510.17369.pdf

Abstract:
Robotic systems are increasingly expected to operate in human-centered, unstructured environments where safety, adaptability, and generalization are essential. Vision-Language-Action (VLA) models have been proposed as a language guided generalized control framework for real robots. However, their deployment has been limited to conventional serial link manipulators. Coupled by their rigidity and unpredictability of learning based control, the ability to safely interact with the environment is missing yet critical. In this work, we present the deployment of a VLA model on a soft continuum manipulator to demonstrate autonomous safe human-robot interaction. We present a structured finetuning and deployment pipeline evaluating two state-of-the-art VLA models (OpenVLA-OFT and $π_0$) across representative manipulation tasks, and show while out-of-the-box policies fail due to embodiment mismatch, through targeted finetuning the soft robot performs equally to the rigid counterpart. Our findings highlight the necessity of finetuning for bridging embodiment gaps, and demonstrate that coupling VLA models with soft robots enables safe and flexible embodied AI in human-shared environments.

Paperid: 5747, https://arxiv.org/pdf/2510.16295.pdf

Abstract:
OpenLVLM-MIA is a new benchmark that highlights fundamental challenges in evaluating membership inference attacks (MIA) against large vision-language models (LVLMs). While prior work has reported high attack success rates, our analysis suggests that these results often arise from detecting distributional bias introduced during dataset construction rather than from identifying true membership status. To address this issue, we introduce a controlled benchmark of 6{,}000 images where the distributions of member and non-member samples are carefully balanced, and ground-truth membership labels are provided across three distinct training stages. Experiments using OpenLVLM-MIA demonstrated that the performance of state-of-the-art MIA methods converged to random chance under unbiased conditions. By offering a transparent and unbiased benchmark, OpenLVLM-MIA clarifies the current limitations of MIA research on LVLMs and provides a solid foundation for developing stronger privacy-preserving techniques.

Paperid: 5748, https://arxiv.org/pdf/2510.15418.pdf

Abstract:
Retrieval-Augmented Generation systems are essential for providing fact-based guidance from Malaysian Clinical Practice Guidelines. However, their effectiveness with image-based queries is limited, as general Vision-Language Model captions often lack clinical specificity and factual grounding. This study proposes and validates a framework to specialize the MedGemma model for generating high-fidelity captions that serve as superior queries. To overcome data scarcity, we employ a knowledge distillation pipeline to create a synthetic dataset across dermatology, fundus, and chest radiography domains, and fine-tune MedGemma using the parameter-efficient QLoRA method. Performance was rigorously assessed through a dual framework measuring both classification accuracy and, via a novel application of the RAGAS framework, caption faithfulness, relevancy, and correctness. The fine-tuned model demonstrated substantial improvements in classification performance, while RAGAS evaluation confirmed significant gains in caption faithfulness and correctness, validating the models ability to produce reliable, factually grounded descriptions. This work establishes a robust pipeline for specializing medical VLMs and validates the resulting model as a high-quality query generator, laying the groundwork for enhancing multimodal RAG systems in evidence-based clinical decision support.

Paperid: 5749, https://arxiv.org/pdf/2510.14968.pdf

Abstract:
To tackle long-horizon tasks, recent hierarchical vision-language-action (VLAs) frameworks employ vision-language model (VLM)-based planners to decompose complex manipulation tasks into simpler sub-tasks that low-level visuomotor policies can easily handle. Typically, the VLM planner is finetuned to learn to decompose a target task. This finetuning requires target task demonstrations segmented into sub-tasks by either human annotation or heuristic rules. However, the heuristic subtasks can deviate significantly from the training data of the visuomotor policy, which degrades task performance. To address these issues, we propose a Retrieval-based Demonstration Decomposer (RDD) that automatically decomposes demonstrations into sub-tasks by aligning the visual features of the decomposed sub-task intervals with those from the training data of the low-level visuomotor policies. Our method outperforms the state-of-the-art sub-task decomposer on both simulation and real-world tasks, demonstrating robustness across diverse settings. Code and more results are available at rdd-neurips.github.io.

Paperid: 5750, https://arxiv.org/pdf/2510.12400.pdf

Abstract:
Urban monitoring of public infrastructure (such as waste bins, road signs, vegetation, sidewalks, and construction sites) poses significant challenges due to the diversity of objects, environments, and contextual conditions involved. Current state-of-the-art approaches typically rely on a combination of IoT sensors and manual inspections, which are costly, difficult to scale, and often misaligned with citizens' perception formed through direct visual observation. This raises a critical question: Can machines now "see" like citizens and infer informed opinions about the condition of urban infrastructure? Vision-Language Models (VLMs), which integrate visual understanding with natural language reasoning, have recently demonstrated impressive capabilities in processing complex visual information, turning them into a promising technology to address this challenge. This systematic review investigates the role of VLMs in urban monitoring, with particular emphasis on zero-shot applications. Following the PRISMA methodology, we analyzed 32 peer-reviewed studies published between 2021 and 2025 to address four core research questions: (1) What urban monitoring tasks have been effectively addressed using VLMs? (2) Which VLM architectures and frameworks are most commonly used and demonstrate superior performance? (3) What datasets and resources support this emerging field? (4) How are VLM-based applications evaluated, and what performance levels have been reported?

Paperid: 5751, https://arxiv.org/pdf/2510.08851.pdf

Abstract:
Intelligent exploration remains a critical challenge in reinforcement learning (RL), especially in visual control tasks. Unlike low-dimensional state-based RL, visual RL must extract task-relevant structure from raw pixels, making exploration inefficient. We propose Concept-Driven Exploration (CDE), which leverages a pre-trained vision-language model (VLM) to generate object-centric visual concepts from textual task descriptions as weak, potentially noisy supervisory signals. Rather than directly conditioning on these noisy signals, CDE trains a policy to reconstruct the concepts via an auxiliary objective, using reconstruction accuracy as an intrinsic reward to guide exploration toward task-relevant objects. Because the policy internalizes these concepts, VLM queries are only needed during training, reducing dependence on external models during deployment. Across five challenging simulated visual manipulation tasks, CDE achieves efficient, targeted exploration and remains robust to noisy VLM predictions. Finally, we demonstrate real-world transfer by deploying CDE on a Franka Research 3 arm, attaining an 80\% success rate in a real-world manipulation task.

Paperid: 5752, https://arxiv.org/pdf/2510.06820.pdf

Abstract:
Multimodal retrieval still leans on embedding-based models like CLIP for fast vector search over pre-computed image embeddings. Yet, unlike text retrieval, where joint-encoder rerankers are standard, comparable vision--language rerankers are largely absent. We find that seminal joint encoders such as BLIP are severely bottlenecked by an expensive visual feature-extraction stage, preventing practical deployment at scale. Motivated by this bottleneck, we introduce EDJE, an Efficient Discriminative Joint Encoder that precomputes vision tokens offline and compresses them via a lightweight attention-based adapter, so online inference runs only a compact joint encoder over a small set of visual tokens plus the text. EDJE preserves strong retrieval performance while drastically reducing storage and online compute, enabling high-throughput inference. Specifically, EDJE processes 50k image--text pairs/second while requiring 49kB of disk storage per image, matching prior art on Flickr (zero-shot) and COCO (fine-tuned) retrieval. The implementation and checkpoints will be made publicly available shortly.

Paperid: 5753, https://arxiv.org/pdf/2510.06251.pdf

Abstract:
Leading Vision-Language Models (VLMs) show strong results in visual perception and general reasoning, but their ability to understand and predict physical dynamics remains unclear. We benchmark six frontier VLMs on three physical simulation datasets - CLEVRER, Physion, and Physion++ - where the evaluation tasks test whether a model can predict outcomes or hypothesize about alternative situations. To probe deeper, we design diagnostic subtests that isolate perception (objects, colors, occluders) from physics reasoning (motion prediction, spatial relations). Intuitively, stronger diagnostic performance should support higher evaluation accuracy. Yet our analysis reveals weak correlations: models that excel at perception or physics reasoning do not consistently perform better on predictive or counterfactual evaluation. This counterintuitive gap exposes a central limitation of current VLMs: perceptual and physics skills remain fragmented and fail to combine into causal understanding, underscoring the need for architectures that bind perception and reasoning more tightly.

Paperid: 5754, https://arxiv.org/pdf/2509.22258.pdf

Abstract:
Recent advances in vision-language models (VLMs) have achieved remarkable performance on standard medical benchmarks, yet their true clinical reasoning ability remains unclear. Existing datasets predominantly emphasize classification accuracy, creating an evaluation illusion in which models appear proficient while still failing at high-stakes diagnostic reasoning. We introduce Neural-MedBench, a compact yet reasoning-intensive benchmark specifically designed to probe the limits of multimodal clinical reasoning in neurology. Neural-MedBench integrates multi-sequence MRI scans, structured electronic health records, and clinical notes, and encompasses three core task families: differential diagnosis, lesion recognition, and rationale generation. To ensure reliable evaluation, we develop a hybrid scoring pipeline that combines LLM-based graders, clinician validation, and semantic similarity metrics. Through systematic evaluation of state-of-the-art VLMs, including GPT-4o, Claude-4, and MedGemma, we observe a sharp performance drop compared to conventional datasets. Error analysis shows that reasoning failures, rather than perceptual errors, dominate model shortcomings. Our findings highlight the necessity of a Two-Axis Evaluation Framework: breadth-oriented large datasets for statistical generalization, and depth-oriented, compact benchmarks such as Neural-MedBench for reasoning fidelity. We release Neural-MedBench at https://neuromedbench.github.io/ as an open and extensible diagnostic testbed, which guides the expansion of future benchmarks and enables rigorous yet cost-effective assessment of clinically trustworthy AI.

Paperid: 5755, https://arxiv.org/pdf/2509.22221.pdf

Abstract:
Vision-Language Models (VLMs) in remote sensing often fail at complex analytical tasks, a limitation stemming from their end-to-end training paradigm that bypasses crucial reasoning steps and leads to unverifiable outputs. To address this limitation, we introduce the Perceptually-Grounded Geospatial Chain-of-Thought (Geo-CoT), a framework that models remote sensing analysis as a verifiable, multi-step process. We instill this analytical process through a two-stage alignment strategy, leveraging Geo-CoT380k, the first large-scale dataset of structured Geo-CoT rationales. This strategy first employs supervised fine-tuning (SFT) to instill the foundational cognitive architecture, then leverages Group Reward Policy Optimization (GRPO) to refine the model's reasoning policy towards factual correctness. The resulting model, RSThinker, outputs both a final answer and its justifying, verifiable analytical trace. This capability yields dominant performance, significantly outperforming state-of-the-art models across a comprehensive range of tasks. The public release of our Geo-CoT380k dataset and RSThinker model upon publication serves as a concrete pathway from opaque perception towards structured, verifiable reasoning for Earth Observation.

Paperid: 5756, https://arxiv.org/pdf/2509.18221.pdf

Abstract:
With the rising global burden of chronic diseases and the multimodal and heterogeneous clinical data (medical imaging, free-text recordings, wearable sensor streams, etc.), there is an urgent need for a unified multimodal AI framework that can proactively predict individual health risks. We propose VL-RiskFormer, a hierarchical stacked visual-language multimodal Transformer with a large language model (LLM) inference head embedded in its top layer. The system builds on the dual-stream architecture of existing visual-linguistic models (e.g., PaLM-E, LLaVA) with four key innovations: (i) pre-training with cross-modal comparison and fine-grained alignment of radiological images, fundus maps, and wearable device photos with corresponding clinical narratives using momentum update encoders and debiased InfoNCE losses; (ii) a time fusion block that integrates irregular visit sequences into the causal Transformer decoder through adaptive time interval position coding; (iii) a disease ontology map adapter that injects ICD-10 codes into visual and textual channels in layers and infers comorbid patterns with the help of a graph attention mechanism. On the MIMIC-IV longitudinal cohort, VL-RiskFormer achieved an average AUROC of 0.90 with an expected calibration error of 2.7 percent.

Paperid: 5757, https://arxiv.org/pdf/2509.18183.pdf

Abstract:
The Visual-Language-Action (VLA) models can follow text instructions according to visual observations of the surrounding environment. This ability to map multimodal inputs to actions is derived from the training of the VLA model on extensive standard demonstrations. These visual observations captured by third-personal global and in-wrist local cameras are inevitably varied in number and perspective across different environments, resulting in significant differences in the visual features. This perspective heterogeneity constrains the generality of VLA models. In light of this, we first propose the lightweight module VLA-LPAF to foster the perspective adaptivity of VLA models using only 2D data. VLA-LPAF is finetuned using images from a single view and fuses other multiview observations in the latent space, which effectively and efficiently bridge the gap caused by perspective inconsistency. We instantiate our VLA-LPAF framework with the VLA model RoboFlamingo to construct RoboFlamingo-LPAF. Experiments show that RoboFlamingo-LPAF averagely achieves around 8% task success rate improvement on CALVIN, 15% on LIBERO, and 30% on a customized simulation benchmark. We also demonstrate the developed viewadaptive characteristics of the proposed RoboFlamingo-LPAF through real-world tasks.

Paperid: 5758, https://arxiv.org/pdf/2509.17615.pdf

Abstract:
Visual anomaly detection is a strongly application-driven field of research. Consequently, the connection between academia and industry is of paramount importance. In this regard, we present the VAND 3.0 Challenge to showcase current progress in anomaly detection across different practical settings whilst addressing critical issues in the field. The challenge hosted two tracks, fostering the development of anomaly detection methods robust against real-world distribution shifts (Category 1) and exploring the capabilities of Vision Language Models within the few-shot regime (Category 2), respectively. The participants' solutions reached significant improvements over previous baselines by combining or adapting existing approaches and fusing them with novel pipelines. While for both tracks the progress in large pre-trained vision (language) backbones played a pivotal role for the performance increase, scaling up anomaly detection methods more efficiently needs to be addressed by future research to meet real-time and computational constraints on-site.

Paperid: 5759, https://arxiv.org/pdf/2509.16176.pdf

Abstract:
We present Agentic Aerial Cinematography: From Dialogue Cues to Cinematic Trajectories (ACDC), an autonomous drone cinematography system driven by natural language communication between human directors and drones. The main limitation of previous drone cinematography workflows is that they require manual selection of waypoints and view angles based on predefined human intent, which is labor-intensive and yields inconsistent performance. In this paper, we propose employing large language models (LLMs) and vision foundation models (VFMs) to convert free-form natural language prompts directly into executable indoor UAV video tours. Specifically, our method comprises a vision-language retrieval pipeline for initial waypoint selection, a preference-based Bayesian optimization framework that refines poses using aesthetic feedback, and a motion planner that generates safe quadrotor trajectories. We validate ACDC through both simulation and hardware-in-the-loop experiments, demonstrating that it robustly produces professional-quality footage across diverse indoor scenes without requiring expertise in robotics or cinematography. These results highlight the potential of embodied AI agents to close the loop from open-vocabulary dialogue to real-world autonomous aerial cinematography.

Paperid: 5760, https://arxiv.org/pdf/2509.15243.pdf

Abstract:
Recent advances in vision-language models have significantly expanded the frontiers of automated image analysis. However, applying these models in safety-critical contexts remains challenging due to the complex relationships between objects, subtle visual cues, and the heightened demand for transparency and reliability. This paper presents the Multi-Modal Explainable Learning (MMEL) framework, designed to enhance the interpretability of vision-language models while maintaining high performance. Building upon prior work in gradient-based explanations for transformer architectures (Grad-eclip), MMEL introduces a novel Hierarchical Semantic Relationship Module that enhances model interpretability through multi-scale feature processing, adaptive attention weighting, and cross-modal alignment. Our approach processes features at multiple semantic levels to capture relationships between image regions at different granularities, applying learnable layer-specific weights to balance contributions across the model's depth. This results in more comprehensive visual explanations that highlight both primary objects and their contextual relationships with improved precision. Through extensive experiments on standard datasets, we demonstrate that by incorporating semantic relationship information into gradient-based attribution maps, MMEL produces more focused and contextually aware visualizations that better reflect how vision-language models process complex scenes. The MMEL framework generalizes across various domains, offering valuable insights into model decisions for applications requiring high interpretability and reliability.

Paperid: 5761, https://arxiv.org/pdf/2509.15059.pdf

Abstract:
Images play a vital role in improving the readability and comprehension of Wikipedia articles by serving as `illustrative aids.' However, not all images are equally effective and not all Wikipedia editors are trained in their selection. We propose QuizRank, a novel method of image selection that leverages large language models (LLMs) and vision language models (VLMs) to rank images as learning interventions. Our approach transforms textual descriptions of the article's subject into multiple-choice questions about important visual characteristics of the concept. We utilize these questions to quiz the VLM: the better an image can help answer questions, the higher it is ranked. To further improve discrimination between visually similar items, we introduce a Contrastive QuizRank that leverages differences in the features of target (e.g., a Western Bluebird) and distractor concepts (e.g., Mountain Bluebird) to generate questions. We demonstrate the potential of VLMs as effective visual evaluators by showing a high congruence with human quiz-takers and an effective discriminative ranking of images.

Paperid: 5762, https://arxiv.org/pdf/2509.14985.pdf

Abstract:
Compared to traditional image retrieval tasks, product retrieval in retail settings is even more challenging. Products of the same type from different brands may have highly similar visual appearances, and the query image may be taken from an angle that differs significantly from view angles of the stored catalog images. Foundational models, such as CLIP and SigLIP, often struggle to distinguish these subtle but important local differences. Pixel-wise matching methods, on the other hand, are computationally expensive and incur prohibitively high matching times. In this paper, we propose a new, hybrid method, called PRISM, for product retrieval in retail settings by leveraging the advantages of both vision-language model-based and pixel-wise matching approaches. To provide both efficiency/speed and finegrained retrieval accuracy, PRISM consists of three stages: 1) A vision-language model (SigLIP) is employed first to retrieve the top 35 most semantically similar products from a fixed gallery, thereby narrowing the search space significantly; 2) a segmentation model (YOLO-E) is applied to eliminate background clutter; 3) fine-grained pixel-level matching is performed using LightGlue across the filtered candidates. This framework enables more accurate discrimination between products with high inter-class similarity by focusing on subtle visual cues often missed by global models. Experiments performed on the ABV dataset show that our proposed PRISM outperforms the state-of-the-art image retrieval methods by 4.21% in top-1 accuracy while still remaining within the bounds of real-time processing for practical retail deployments.

Paperid: 5763, https://arxiv.org/pdf/2509.11071.pdf

Abstract:
This report outlines our approach using vision language model systems for the Driving with Language track of the CVPR 2024 Autonomous Grand Challenge. We have exclusively utilized the DriveLM-nuScenes dataset for training our models. Our systems are built on the LLaVA models, which we enhanced through fine-tuning with the LoRA and DoRA methods. Additionally, we have integrated depth information from open-source depth estimation models to enrich the training and inference processes. For inference, particularly with multiple-choice and yes/no questions, we adopted a Chain-of-Thought reasoning approach to improve the accuracy of the results. This comprehensive methodology enabled us to achieve a top score of 0.7799 on the validation set leaderboard, ranking 1st on the leaderboard.

Paperid: 5764, https://arxiv.org/pdf/2509.07957.pdf

Abstract:
Acquiring dexterous robotic skills from human video demonstrations remains a significant challenge, largely due to conventional reliance on low-level trajectory replication, which often fails to generalize across varying objects, spatial layouts, and manipulator configurations. To address this limitation, we introduce Graph-Fused Vision-Language-Action (GF-VLA), a unified framework that enables dual-arm robotic systems to perform task-level reasoning and execution directly from RGB-D human demonstrations. GF-VLA employs an information-theoretic approach to extract task-relevant cues, selectively highlighting critical hand-object and object-object interactions. These cues are structured into temporally ordered scene graphs, which are subsequently integrated with a language-conditioned transformer to produce hierarchical behavior trees and interpretable Cartesian motion primitives. To enhance efficiency in bimanual execution, we propose a cross-arm allocation strategy that autonomously determines gripper assignment without requiring explicit geometric modeling. We validate GF-VLA on four dual-arm block assembly benchmarks involving symbolic structure construction and spatial generalization. Empirical results demonstrate that the proposed representation achieves over 95% graph accuracy and 93% subtask segmentation, enabling the language-action planner to generate robust, interpretable task policies. When deployed on a dual-arm robot, these policies attain 94% grasp reliability, 89% placement accuracy, and 90% overall task success across stacking, letter-formation, and geometric reconfiguration tasks, evidencing strong generalization and robustness under diverse spatial and semantic variations.

Paperid: 5765, https://arxiv.org/pdf/2509.00752.pdf

Abstract:
We present a unified vision-language framework tailored for ENT endoscopy image analysis that simultaneously tackles three clinically-relevant tasks: image classification, image-to-image retrieval, and text-to-image retrieval. Unlike conventional CNN-based pipelines that struggle to capture cross-modal semantics, our approach leverages the CLIP ViT-B/16 backbone and enhances it through Low-Rank Adaptation, multi-level CLS token aggregation, and spherical feature interpolation. These components collectively enable efficient fine-tuning on limited medical data while improving representation diversity and semantic alignment across modalities. To bridge the gap between visual inputs and textual diagnostic context, we introduce class-specific natural language prompts that guide the image encoder through a joint training objective combining supervised classification with contrastive learning. We validated our framework through participation in the ACM MM'25 ENTRep Grand Challenge, achieving 95% accuracy and F1-score in classification, Recall@1 of 0.93 and 0.92 for image-to-image and text-to-image retrieval respectively, and MRR scores of 0.97 and 0.96. Ablation studies demonstrated the incremental benefits of each architectural component, validating the effectiveness of our design for robust multimodal medical understanding in low-resource clinical settings.

Paperid: 5766, https://arxiv.org/pdf/2508.20760.pdf

Abstract:
Vision-language models (VLMs) like CLIP enable zero-shot classification by aligning images and text in a shared embedding space, offering advantages for defense applications with scarce labeled data. However, CLIP's robustness in challenging military environments, with partial occlusion and degraded signal-to-noise ratio (SNR), remains underexplored. We investigate CLIP variants' robustness to occlusion using a custom dataset of 18 military vehicle classes and evaluate using Normalized Area Under the Curve (NAUC) across occlusion percentages. Four key insights emerge: (1) Transformer-based CLIP models consistently outperform CNNs, (2) fine-grained, dispersed occlusions degrade performance more than larger contiguous occlusions, (3) despite improved accuracy, performance of linear-probed models sharply drops at around 35% occlusion, (4) by finetuning the model's backbone, this performance drop occurs at more than 60% occlusion. These results underscore the importance of occlusion-specific augmentations during training and the need for further exploration into patch-level sensitivity and architectural resilience for real-world deployment of CLIP.

Paperid: 5767, https://arxiv.org/pdf/2508.20243.pdf

Abstract:
Rapid and reliable qualification of advanced materials remains a bottleneck in industrial manufacturing, particularly for heterogeneous structures produced via non-conventional additive manufacturing processes. This study introduces a novel framework that links microstructure informatics with a range of expert characterization knowledge using customized and hybrid vision-language representations (VLRs). By integrating deep semantic segmentation with pre-trained multi-modal models (CLIP and FLAVA), we encode both visual microstructural data and textual expert assessments into shared representations. To overcome limitations in general-purpose embeddings, we develop a customized similarity-based representation that incorporates both positive and negative references from expert-annotated images and their associated textual descriptions. This allows zero-shot classification of previously unseen microstructures through a net similarity scoring approach. Validation on an additively manufactured metal matrix composite dataset demonstrates the framework's ability to distinguish between acceptable and defective samples across a range of characterization criteria. Comparative analysis reveals that FLAVA model offers higher visual sensitivity, while the CLIP model provides consistent alignment with the textual criteria. Z-score normalization adjusts raw unimodal and cross-modal similarity scores based on their local dataset-driven distributions, enabling more effective alignment and classification in the hybrid vision-language framework. The proposed method enhances traceability and interpretability in qualification pipelines by enabling human-in-the-loop decision-making without task-specific model retraining. By advancing semantic interoperability between raw data and expert knowledge, this work contributes toward scalable and domain-adaptable qualification strategies in engineering informatics.

Paperid: 5768, https://arxiv.org/pdf/2508.18179.pdf

Abstract:
Evaluating whether vision-language models (VLMs) reason consistently across representations is challenging because modality comparisons are typically confounded by task differences and asymmetric information. We introduce SEAM, a benchmark that pairs semantically equivalent inputs across four domains that have existing standardized textual and visual notations. By employing distinct notation systems across modalities, in contrast to OCR-based image-text pairing, SEAM provides a rigorous comparative assessment of the textual-symbolic and visual-spatial reasoning capabilities of VLMs. Across 21 contemporary models, we observe systematic modality imbalance: vision frequently lags language in overall performance, despite the problems containing semantically equivalent information, and cross-modal agreement is relatively low. Our error analysis reveals two main drivers: textual perception failures from tokenization in domain notation and visual perception failures that induce hallucinations. We also show that our results are largely robust to visual transformations. SEAM establishes a controlled, semantically equivalent setting for measuring and improving modality-agnostic reasoning.

Paperid: 5769, https://arxiv.org/pdf/2508.17215.pdf

Abstract:
Large Vision-Language Models (LVLMs) augmented with Retrieval-Augmented Generation (RAG) are increasingly employed in medical AI to enhance factual grounding through external clinical image-text retrieval. However, this reliance creates a significant attack surface. We propose MedThreatRAG, a novel multimodal poisoning framework that systematically probes vulnerabilities in medical RAG systems by injecting adversarial image-text pairs. A key innovation of our approach is the construction of a simulated semi-open attack environment, mimicking real-world medical systems that permit periodic knowledge base updates via user or pipeline contributions. Within this setting, we introduce and emphasize Cross-Modal Conflict Injection (CMCI), which embeds subtle semantic contradictions between medical images and their paired reports. These mismatches degrade retrieval and generation by disrupting cross-modal alignment while remaining sufficiently plausible to evade conventional filters. While basic textual and visual attacks are included for completeness, CMCI demonstrates the most severe degradation. Evaluations on IU-Xray and MIMIC-CXR QA tasks show that MedThreatRAG reduces answer F1 scores by up to 27.66% and lowers LLaVA-Med-1.5 F1 rates to as low as 51.36%. Our findings expose fundamental security gaps in clinical RAG systems and highlight the urgent need for threat-aware design and robust multimodal consistency checks. Finally, we conclude with a concise set of guidelines to inform the safe development of future multimodal medical RAG systems.

Paperid: 5770, https://arxiv.org/pdf/2508.17117.pdf

Abstract:
PlantVillageVQA is a large-scale visual question answering (VQA) dataset derived from the widely used PlantVillage image corpus. It was designed to advance the development and evaluation of vision-language models for agricultural decision-making and analysis. The PlantVillageVQA dataset comprises 193,609 high-quality question-answer (QA) pairs grounded over 55,448 images spanning 14 crop species and 38 disease conditions. Questions are organised into 3 levels of cognitive complexity and 9 distinct categories. Each question category was phrased manually following expert guidance and generated via an automated two-stage pipeline: (1) template-based QA synthesis from image metadata and (2) multi-stage linguistic re-engineering. The dataset was iteratively reviewed by domain experts for scientific accuracy and relevancy. The final dataset was evaluated using three state-of-the-art models for quality assessment. Our objective remains to provide a publicly available, standardised and expert-verified database to enhance diagnostic accuracy for plant disease identifications and advance scientific research in the agricultural domain. Our dataset will be open-sourced at https://huggingface.co/datasets/SyedNazmusSakib/PlantVillageVQA.

Paperid: 5771, https://arxiv.org/pdf/2508.12861.pdf

Abstract:
Vision-language models (VLMs) pre-trained on natural image and language data, such as CLIP, have exhibited significant potential in few-shot image recognition tasks, leading to development of various efficient transfer learning methods. These methods exploit inherent pre-learned knowledge in VLMs and have achieved strong performance on standard image datasets. However, their effectiveness is often limited when confronted with cross-domain tasks where imaging domains differ from natural images. To address this limitation, we propose Consistency-guided Multi-view Collaborative Optimization (CoMuCo), a novel fine-tuning strategy for VLMs. This strategy employs two functionally complementary expert modules to extract multi-view features, while incorporating prior knowledge-based consistency constraints and information geometry-based consensus mechanisms to enhance the robustness of feature learning. Additionally, a new cross-domain few-shot benchmark is established to help comprehensively evaluate methods on imaging domains distinct from natural images. Extensive empirical evaluations on both existing and newly proposed benchmarks suggest CoMuCo consistently outperforms current methods in few-shot tasks. The code and benchmark will be released.

Paperid: 5772, https://arxiv.org/pdf/2508.12638.pdf

Abstract:
Vision-Language Models (VLMs) are increasingly deployed in real-time applications such as autonomous driving and human-computer interaction, which demand fast and reliable responses based on accurate perception. To meet these requirements, existing systems commonly employ cloud-edge collaborative architectures, such as partitioned Large Vision-Language Models (LVLMs) or task offloading strategies between Large and Small Vision-Language Models (SVLMs). However, these methods fail to accommodate cloud latency fluctuations and overlook the full potential of delayed but accurate LVLM responses. In this work, we propose a novel cloud-edge collaborative paradigm for VLMs, termed Context Transfer, which treats the delayed outputs of LVLMs as historical context to provide real-time guidance for SVLMs inference. Based on this paradigm, we design SpotVLM, which incorporates both context replacement and visual focus modules to refine historical textual input and enhance visual grounding consistency. Extensive experiments on three real-time vision tasks across four datasets demonstrate the effectiveness of the proposed framework. The new paradigm lays the groundwork for more effective and latency-aware collaboration strategies in future VLM systems.

Paperid: 5773, https://arxiv.org/pdf/2508.12108.pdf

Abstract:
Vision-and-language models (VLMs) have been increasingly explored in the medical domain, particularly following the success of CLIP in general domain. However, unlike the relatively straightforward pairing of 2D images and text, curating large-scale paired data in the medical field for volumetric modalities such as CT scans remains a challenging and time-intensive process. This difficulty often limits the performance on downstream tasks. To address these challenges, we propose a novel vision-language pre-training (VLP) framework, termed as \textbf{VELVET-Med}, specifically designed for limited volumetric data such as 3D CT and associated radiology reports. Instead of relying on large-scale data collection, our method focuses on the development of effective pre-training objectives and model architectures. The key contributions are: 1) We incorporate uni-modal self-supervised learning into VLP framework, which are often underexplored in the existing literature. 2) We propose a novel language encoder, termed as \textbf{TriBERT}, for learning multi-level textual semantics. 3) We devise the hierarchical contrastive learning to capture multi-level vision-language correspondence. Using only 38,875 scan-report pairs, our approach seeks to uncover rich spatial and semantic relationships embedded in volumetric medical images and corresponding clinical narratives, thereby enhancing the generalization ability of the learned encoders. The resulting encoders exhibit strong transferability, achieving state-of-the-art performance across a wide range of downstream tasks, including 3D segmentation, cross-modal retrieval, visual question answering, and report generation.

Paperid: 5774, https://arxiv.org/pdf/2508.10972.pdf

Abstract:
Advances in vision language models (VLMs) have enabled the simulation of general human behavior through their reasoning and problem solving capabilities. However, prior research has not investigated such simulation capabilities in the accessibility domain. In this paper, we evaluate the extent to which VLMs can simulate the vision perception of low vision individuals when interpreting images. We first compile a benchmark dataset through a survey study with 40 low vision participants, collecting their brief and detailed vision information and both open-ended and multiple-choice image perception and recognition responses to up to 25 images. Using these responses, we construct prompts for VLMs (GPT-4o) to create simulated agents of each participant, varying the included information on vision information and example image responses. We evaluate the agreement between VLM-generated responses and participants' original answers. Our results indicate that VLMs tend to infer beyond the specified vision ability when given minimal prompts, resulting in low agreement (0.59). The agreement between the agent' and participants' responses remains low when only either the vision information (0.59) or example image responses (0.59) are provided, whereas a combination of both significantly increase the agreement (0.70, p < 0.0001). Notably, a single example combining both open-ended and multiple-choice responses, offers significant performance improvements over either alone (p < 0.0001), while additional examples provided minimal benefits (p > 0.05).

Paperid: 5775, https://arxiv.org/pdf/2508.10770.pdf

Abstract:
Spatio-physical reasoning, a foundation capability for understanding the real physics world, is a critical step towards building robust world models. While recent vision language models (VLMs) have shown remarkable progress in specialized domains like multimodal mathematics and pure spatial understanding, their capability for spatio-physical reasoning remains largely unexplored. This paper provides a comprehensive diagnostic analysis of mainstream VLMs, revealing that current models perform inadequately on this crucial task. Further detailed analysis shows that this underperformance is largely attributable to biases caused by human-like prior and a lack of deep reasoning. To address these challenges, we apply supervised fine-tuning followed by rule-based reinforcement learning to Qwen2.5-VL-7B, resulting in significant improvements in spatio-physical reasoning capabilities and surpassing leading proprietary models. Nevertheless, despite this success, the model's generalization to new physics scenarios remains limited -- underscoring the pressing need for new approaches in spatio-physical reasoning.

Paperid: 5776, https://arxiv.org/pdf/2508.05342.pdf

Abstract:
Teaching robots dexterous skills from human videos remains challenging due to the reliance on low-level trajectory imitation, which fails to generalize across object types, spatial layouts, and manipulator configurations. We propose Graph-Fused Vision-Language-Action (GF-VLA), a framework that enables dual-arm robotic systems to perform task-level reasoning and execution directly from RGB and Depth human demonstrations. GF-VLA first extracts Shannon-information-based cues to identify hands and objects with the highest task relevance, then encodes these cues into temporally ordered scene graphs that capture both hand-object and object-object interactions. These graphs are fused with a language-conditioned transformer that generates hierarchical behavior trees and interpretable Cartesian motion commands. To improve execution efficiency in bimanual settings, we further introduce a cross-hand selection policy that infers optimal gripper assignment without explicit geometric reasoning. We evaluate GF-VLA on four structured dual-arm block assembly tasks involving symbolic shape construction and spatial generalization. Experimental results show that the information-theoretic scene representation achieves over 95 percent graph accuracy and 93 percent subtask segmentation, supporting the LLM planner in generating reliable and human-readable task policies. When executed by the dual-arm robot, these policies yield 94 percent grasp success, 89 percent placement accuracy, and 90 percent overall task success across stacking, letter-building, and geometric reconfiguration scenarios, demonstrating strong generalization and robustness across diverse spatial and semantic variations.

Paperid: 5777, https://arxiv.org/pdf/2508.05323.pdf

Abstract:
Recent progress in large pre-trained vision language models (VLMs) has reached state-of-the-art performance on several object detection benchmarks and boasts strong zero-shot capabilities, but for optimal performance on specific targets some form of finetuning is still necessary. While the initial VLM weights allow for great few-shot transfer learning, this usually involves the loss of the original natural language querying and zero-shot capabilities. Inspired by the success of Textual Inversion (TI) in personalizing text-to-image diffusion models, we propose a similar formulation for open-vocabulary object detection. TI allows extending the VLM vocabulary by learning new or improving existing tokens to accurately detect novel or fine-grained objects from as little as three examples. The learned tokens are completely compatible with the original VLM weights while keeping them frozen, retaining the original model's benchmark performance, and leveraging its existing capabilities such as zero-shot domain transfer (e.g., detecting a sketch of an object after training only on real photos). The storage and gradient calculations are limited to the token embedding dimension, requiring significantly less compute than full-model fine-tuning. We evaluated whether the method matches or outperforms the baseline methods that suffer from forgetting in a wide variety of quantitative and qualitative experiments.

Paperid: 5778, https://arxiv.org/pdf/2508.05294.pdf

Abstract:
Foundation models, including large language models (LLMs) and vision-language models (VLMs), have recently enabled novel approaches to robot autonomy and human-robot interfaces. In parallel, vision-language-action models (VLAs) or large behavior models (LBMs) are increasing the dexterity and capabilities of robotic systems. This survey paper focuses on those works advancing towards agentic applications and architectures. This includes initial efforts exploring GPT-style interfaces to tooling, as well as more complex system where AI agents are coordinators, planners, perception actors, or generalist interfaces. Such agentic architectures allow robots to reason over natural language instructions, invoke APIs, plan task sequences, or assist in operations and diagnostics. In addition to peer-reviewed research, due to the fast-evolving nature of the field, we highlight and include community-driven projects, ROS packages, and industrial frameworks that show emerging trends. We propose a taxonomy for classifying model integration approaches and present a comparative analysis of the role that agents play in different solutions in today's literature.

Paperid: 5779, https://arxiv.org/pdf/2508.04945.pdf

Abstract:
Evaluating visual activity recognition systems is challenging due to inherent ambiguities in verb semantics and image interpretation. When describing actions in images, synonymous verbs can refer to the same event (e.g., brushing vs. grooming), while different perspectives can lead to equally valid but distinct verb choices (e.g., piloting vs. operating). Standard exact-match evaluation, which relies on a single gold answer, fails to capture these ambiguities, resulting in an incomplete assessment of model performance. To address this, we propose a vision-language clustering framework that constructs verb sense clusters, providing a more robust evaluation. Our analysis of the imSitu dataset shows that each image maps to an average of 2.8 sense clusters, with each cluster representing a distinct perspective of the image. We evaluate multiple activity recognition models and compare our cluster-based evaluation with standard evaluation methods. Additionally, our human alignment analysis suggests that the cluster-based evaluation better aligns with human judgements, offering a more nuanced assessment of model performance.

Paperid: 5780, https://arxiv.org/pdf/2508.03232.pdf

Abstract:
Embodied Planning is dedicated to the goal of creating agents capable of executing long-horizon tasks in complex physical worlds. However, existing embodied planning benchmarks frequently feature short-horizon tasks and coarse-grained action primitives. To address this challenge, we introduce CookBench, a benchmark for long-horizon planning in complex cooking scenarios. By leveraging a high-fidelity simulation environment built upon the powerful Unity game engine, we define frontier AI challenges in a complex, realistic environment. The core task in CookBench is designed as a two-stage process. First, in Intention Recognition, an agent needs to accurately parse a user's complex intent. Second, in Embodied Interaction, the agent should execute the identified cooking goal through a long-horizon, fine-grained sequence of physical actions. Unlike existing embodied planning benchmarks, we refine the action granularity to a spatial level that considers crucial operational information while abstracting away low-level robotic control. Besides, We provide a comprehensive toolset that encapsulates the simulator. Its unified API supports both macro-level operations, such as placing orders and purchasing ingredients, and a rich set of fine-grained embodied actions for physical interaction, enabling researchers to focus on high-level planning and decision-making. Furthermore, we present an in-depth analysis of state-of-the-art, closed-source Large Language Model and Vision-Language Model, revealing their major shortcomings and challenges posed by complex, long-horizon tasks. The full benchmark will be open-sourced to facilitate future research.

Paperid: 5781, https://arxiv.org/pdf/2508.01723.pdf

Abstract:
Grounding natural language instructions to visual observations is fundamental for embodied agents operating in open-world environments. Recent advances in visual-language mapping have enabled generalizable semantic representations by leveraging vision-language models (VLMs). However, these methods often fall short in aligning free-form language commands with specific scene instances, due to limitations in both instance-level semantic consistency and instruction interpretation. We present OpenMap, a zero-shot open-vocabulary visual-language map designed for accurate instruction grounding in navigation tasks. To address semantic inconsistencies across views, we introduce a Structural-Semantic Consensus constraint that jointly considers global geometric structure and vision-language similarity to guide robust 3D instance-level aggregation. To improve instruction interpretation, we propose an LLM-assisted Instruction-to-Instance Grounding module that enables fine-grained instance selection by incorporating spatial context and expressive target descriptions. We evaluate OpenMap on ScanNet200 and Matterport3D, covering both semantic mapping and instruction-to-target retrieval tasks. Experimental results show that OpenMap outperforms state-of-the-art baselines in zero-shot settings, demonstrating the effectiveness of our method in bridging free-form language and 3D perception for embodied navigation.

Paperid: 5782, https://arxiv.org/pdf/2508.01380.pdf

Abstract:
It is of crucial importance to assess damages promptly and accurately in humanitarian assistance and disaster response (HADR). Current deep learning approaches struggle to generalize effectively due to the imbalance of data classes, scarcity of moderate damage examples, and human inaccuracy in pixel labeling during HADR situations. To accommodate for these limitations and exploit state-of-the-art techniques in vision-language models (VLMs) to fuse imagery with human knowledge understanding, there is an opportunity to generate a diversified set of image-based damage data effectively. Our initial experimental results suggest encouraging data generation quality, which demonstrates an improvement in classifying scenes with different levels of structural damage to buildings, roads, and infrastructures.

Paperid: 5783, https://arxiv.org/pdf/2508.01338.pdf

Abstract:
Image forgery localization aims to precisely identify tampered regions within images, but it commonly depends on costly pixel-level annotations. To alleviate this annotation burden, weakly supervised image forgery localization (WSIFL) has emerged, yet existing methods still achieve limited localization performance as they mainly exploit intra-image consistency clues and lack external semantic guidance to compensate for weak supervision. In this paper, we propose ViLaCo, a vision-language collaborative reasoning framework that introduces auxiliary semantic supervision distilled from pre-trained vision-language models (VLMs), enabling accurate pixel-level localization using only image-level labels. Specifically, ViLaCo first incorporates semantic knowledge through a vision-language feature modeling network, which jointly extracts textual and visual priors using pre-trained VLMs. Next, an adaptive vision-language reasoning network aligns textual semantics and visual features through mutual interactions, producing semantically aligned representations. Subsequently, these representations are passed into dual prediction heads, where the coarse head performs image-level classification and the fine head generates pixel-level localization masks, thereby bridging the gap between weak supervision and fine-grained localization. Moreover, a contrastive patch consistency module is introduced to cluster tampered features while separating authentic ones, facilitating more reliable forgery discrimination. Extensive experiments on multiple public datasets demonstrate that ViLaCo substantially outperforms existing WSIFL methods, achieving state-of-the-art performance in both detection and localization accuracy.

Paperid: 5784, https://arxiv.org/pdf/2508.01074.pdf

Abstract:
Modern over-parameterized deep models are highly data-dependent, with large scale general-purpose and domain-specific datasets serving as the bedrock for rapid advancements. However, many datasets are proprietary or contain sensitive information, making unrestricted model training problematic. In the open world where data thefts cannot be fully prevented, Dataset Ownership Verification (DOV) has emerged as a promising method to protect copyright by detecting unauthorized model training and tracing illicit activities. Due to its diversity and superior stealth, evading DOV is considered extremely challenging. However, this paper identifies that previous studies have relied on oversimplistic evasion attacks for evaluation, leading to a false sense of security. We introduce a unified evasion framework, in which a teacher model first learns from the copyright dataset and then transfers task-relevant yet identifier-independent domain knowledge to a surrogate student using an out-of-distribution (OOD) dataset as the intermediary. Leveraging Vision-Language Models and Large Language Models, we curate the most informative and reliable subsets from the OOD gallery set as the final transfer set, and propose selectively transferring task-oriented knowledge to achieve a better trade-off between generalization and evasion effectiveness. Experiments across diverse datasets covering eleven DOV methods demonstrate our approach simultaneously eliminates all copyright identifiers and significantly outperforms nine state-of-the-art evasion attacks in both generalization and effectiveness, with moderate computational overhead. As a proof of concept, we reveal key vulnerabilities in current DOV methods, highlighting the need for long-term development to enhance practicality.

Paperid: 5785, https://arxiv.org/pdf/2508.00549.pdf

Abstract:
Clinical decision-making relies heavily on understanding relative positions of anatomical structures and anomalies. Therefore, for Vision-Language Models (VLMs) to be applicable in clinical practice, the ability to accurately determine relative positions on medical images is a fundamental prerequisite. Despite its importance, this capability remains highly underexplored. To address this gap, we evaluate the ability of state-of-the-art VLMs, GPT-4o, Llama3.2, Pixtral, and JanusPro, and find that all models fail at this fundamental task. Inspired by successful approaches in computer vision, we investigate whether visual prompts, such as alphanumeric or colored markers placed on anatomical structures, can enhance performance. While these markers provide moderate improvements, results remain significantly lower on medical images compared to observations made on natural images. Our evaluations suggest that, in medical imaging, VLMs rely more on prior anatomical knowledge than on actual image content for answering relative position questions, often leading to incorrect conclusions. To facilitate further research in this area, we introduce the MIRP , Medical Imaging Relative Positioning, benchmark dataset, designed to systematically evaluate the capability to identify relative positions in medical images.

Paperid: 5786, https://arxiv.org/pdf/2508.00390.pdf

Abstract:
Unmanned Aerial Vehicle (UAV) Vision-Language Navigation (VLN) aims to enable agents to accurately localize targets and plan flight paths in complex environments based on natural language instructions, with broad applications in intelligent inspection, disaster rescue, and urban monitoring. Recent progress in Vision-Language Models (VLMs) has provided strong semantic understanding for this task, while reinforcement learning (RL) has emerged as a promising post-training strategy to further improve generalization. However, existing RL methods often suffer from inefficient use of training data, slow convergence, and insufficient consideration of the difficulty variation among training samples, which limits further performance improvement. To address these challenges, we propose \textbf{Semantic-Aware Gaussian Curriculum Scheduling (SA-GCS)}, a novel training framework that systematically integrates Curriculum Learning (CL) into RL. SA-GCS employs a Semantic-Aware Difficulty Estimator (SA-DE) to quantify the complexity of training samples and a Gaussian Curriculum Scheduler (GCS) to dynamically adjust the sampling distribution, enabling a smooth progression from easy to challenging tasks. This design significantly improves training efficiency, accelerates convergence, and enhances overall model performance. Extensive experiments on the CityNav benchmark demonstrate that SA-GCS consistently outperforms strong baselines across all metrics, achieves faster and more stable convergence, and generalizes well across models of different scales, highlighting its robustness and scalability. The implementation of our approach is publicly available.

Paperid: 5787, https://arxiv.org/pdf/2507.20077.pdf

Abstract:
Despite significant advances in vision-language models (VLMs), image captioning often suffers from a lack of detail, with base models producing short, generic captions. This limitation persists even though VLMs are equipped with strong vision and language backbones. While supervised data and complex reward functions have been proposed to improve detailed image captioning, we identify a simpler underlying issue: a bias towards the end-of-sequence (EOS) token, which is introduced during cross-entropy training. We propose an unsupervised method to debias the model's tendency to predict the EOS token prematurely. By reducing this bias, we encourage the generation of longer, more detailed captions without the need for intricate reward functions or supervision. Our approach is straightforward, effective, and easily applicable to any pretrained model. We demonstrate its effectiveness through experiments with three VLMs and on three detailed captioning benchmarks. Our results show a substantial increase in caption length and relevant details, albeit with an expected increase in the rate of hallucinations.

Paperid: 5788, https://arxiv.org/pdf/2507.16781.pdf

Abstract:
The unprecedented growth in artificial intelligence (AI) workloads, recently dominated by large language models (LLMs) and vision-language models (VLMs), has intensified power and cooling demands in data centers. This study benchmarks LLMs and VLMs on two HGX nodes, each with 8x NVIDIA H100 graphics processing units (GPUs), using liquid and air cooling. Leveraging GPU Burn, Weights and Biases, and IPMItool, we collect detailed thermal, power, and computation data. Results show that the liquid-cooled systems maintain GPU temperatures between 41-50 degrees Celsius, while the air-cooled counterparts fluctuate between 54-72 degrees Celsius under load. This thermal stability of liquid-cooled systems yields 17 percent higher performance (54 TFLOPs per GPU vs. 46 TFLOPs per GPU), improved performance per watt, reduced energy overhead, and greater system efficiency than the air-cooled counterparts. These findings underscore the energy and sustainability benefits of liquid cooling, offering a compelling path forward for hyperscale data centers s

Paperid: 5789, https://arxiv.org/pdf/2507.11153.pdf

Abstract:
With the widespread adoption of large vision-language models, the capacity for color vision in these models is crucial. However, the color vision abilities of large visual-language models have not yet been thoroughly explored. To address this gap, we define a color vision testing task for large vision-language models and construct a dataset \footnote{Anonymous Github Showing some of the data https://anonymous.4open.science/r/color-vision-test-dataset-3BCD} that covers multiple categories of test questions and tasks of varying difficulty levels. Furthermore, we analyze the types of errors made by large vision-language models and propose fine-tuning strategies to enhance their performance in color vision tests.

Paperid: 5790, https://arxiv.org/pdf/2507.10062.pdf

Abstract:
Snapshot testing has emerged as a critical technique for UI validation in modern software development, yet it suffers from substantial maintenance overhead due to frequent UI changes causing test failures that require manual inspection to distinguish between genuine regressions and intentional design changes. This manual triage process becomes increasingly burdensome as applications evolve, creating a need for automated analysis solutions. This paper introduces LLMShot, a novel framework that leverages Vision-Language Models (VLMs) to automatically analyze snapshot test failures through semantic classification of UI changes. To evaluate LLMShot's effectiveness, we developed a comprehensive dataset using a feature-rich iOS application with configurable feature flags, creating realistic scenarios that produce authentic snapshot differences representative of real development workflows. Our evaluation using Gemma3 models demonstrates strong classification performance, with the 12B variant achieving over 84% recall in identifying failure root causes while the 4B model offers practical deployment advantages with acceptable performance for continuous integration environments. However, our exploration of selective ignore mechanisms revealed significant limitations in current prompting-based approaches for controllable visual reasoning. LLMShot represents the first automated approach to semantic snapshot test analysis, offering developers structured insights that can substantially reduce manual triage effort and advance toward more intelligent UI testing paradigms.

Paperid: 5791, https://arxiv.org/pdf/2507.05798.pdf

Abstract:
Panoptic Scene Graph Generation (PSG) integrates instance segmentation with relation understanding to capture pixel-level structural relationships in complex scenes. Although recent approaches leveraging pre-trained vision-language models (VLMs) have significantly improved performance in the open-vocabulary setting, they commonly ignore the inherent limitations of VLMs in spatial relation reasoning, such as difficulty in distinguishing object relative positions, which results in suboptimal relation prediction. Motivated by the denoising diffusion model's inversion process in preserving the spatial structure of input images, we propose SPADE (SPatial-Aware Denoising-nEtwork) framework -- a novel approach for open-vocabulary PSG. SPADE consists of two key steps: (1) inversion-guided calibration for the UNet adaptation, and (2) spatial-aware context reasoning. In the first step, we calibrate a general pre-trained teacher diffusion model into a PSG-specific denoising network with cross-attention maps derived during inversion through a lightweight LoRA-based fine-tuning strategy. In the second step, we develop a spatial-aware relation graph transformer that captures both local and long-range contextual information, facilitating the generation of high-quality relation queries. Extensive experiments on benchmark PSG and Visual Genome datasets demonstrate that SPADE outperforms state-of-the-art methods in both closed- and open-set scenarios, particularly for spatial relationship prediction.

Paperid: 5792, https://arxiv.org/pdf/2507.04741.pdf

Abstract:
We present Saliency Benchmark (SalBench), a novel benchmark designed to assess the capability of Large Vision-Language Models (LVLM) in detecting visually salient features that are readily apparent to humans, such as a large circle amidst a grid of smaller ones. This benchmark focuses on low-level features including color, intensity, and orientation, which are fundamental to human visual processing. Our SalBench consists of images that highlight rare, unusual, or unexpected elements within scenes, and naturally draw human attention. It comprises three novel tasks for evaluating the perceptual capabilities of LVLM: Odd-One-Out Detection, Referring Odd-One-Out, and Visual Referring Odd-One-Out. We perform a comprehensive evaluation of state-of-the-art LVLM using SalBench and our findings reveal a surprising limitation: LVLM struggle to identify seemingly obvious visual anomalies, with even the advanced GPT-4o achieving only 47.6\% accuracy on such a simple task. SalBench will be an important step in measuring the capabilities of LVLM that align with the subtle definition of human attention.

Paperid: 5793, https://arxiv.org/pdf/2507.03542.pdf

Abstract:
Text-based visual descriptors--ranging from simple class names to more descriptive phrases--are widely used in visual concept discovery and image classification with vision-language models (VLMs). Their effectiveness, however, depends on a complex interplay of factors, including semantic clarity, presence in the VLM's pre-training data, and how well the descriptors serve as a meaningful representation space. In this work, we systematically analyze descriptor quality along two key dimensions: (1) representational capacity, and (2) relationship with VLM pre-training data. We evaluate a spectrum of descriptor generation methods, from zero-shot LLM-generated prompts to iteratively refined descriptors. Motivated by ideas from representation alignment and language understanding, we introduce two alignment-based metrics--Global Alignment and CLIP Similarity--that move beyond accuracy. These metrics shed light on how different descriptor generation strategies interact with foundation model properties, offering new ways to study descriptor effectiveness beyond accuracy evaluations.

Paperid: 5794, https://arxiv.org/pdf/2507.03434.pdf

Abstract:
The data appetite for Vision-Language Models (VLMs) has continuously scaled up from the early millions to billions today, which faces an untenable trade-off with data quality and inevitably introduces Noisy Correspondence (NC) samples. Undoubtedly, such semantically unrelated data significantly impairs the performance of VLMs. Previous efforts mainly address this challenge by estimating refined alignment for more precise guidance. However, such resource-intensive pipelines that train VLMs from scratch struggle to meet realistic data demands. In this paper, we present a brand new perspective that seeks to directly eliminate the harmful effects of NC in pre-trained VLMs. Specifically, we propose NCU, a Noisy Correspondence Unlearning fine-tuning framework that efficiently enhances VLMs' robustness by forgetting learned noisy knowledge. The key to NCU is learning the hardest negative information, which can provide explicit unlearning direction for both false positives and false negatives. Such twin goals unlearning process can be formalized into one unified optimal transport objective for fast fine-tuning. We validate our approach with the prevailing CLIP model over various downstream tasks. Remarkably, NCU surpasses the robust pre-trained method on zero-shot transfer while with lower computational overhead. The code will be released upon acceptance.

Paperid: 5795, https://arxiv.org/pdf/2506.23584.pdf

Abstract:
Generating radiology reports from CT scans remains a complex task due to the nuanced nature of medical imaging and the variability in clinical documentation. In this study, we propose a two-stage framework for generating renal radiology reports from 2D CT slices. First, we extract structured abnormality features using a multi-task learning model trained to identify lesion attributes such as location, size, enhancement, and attenuation. These extracted features are subsequently combined with the corresponding CT image and fed into a fine-tuned vision-language model to generate natural language report sentences aligned with clinical findings. We conduct experiments on a curated dataset of renal CT studies with manually annotated sentence-slice-feature triplets and evaluate performance using both classification metrics and natural language generation metrics. Our results demonstrate that the proposed model outperforms random baselines across all abnormality types, and the generated reports capture key clinical content with reasonable textual accuracy. This exploratory work highlights the feasibility of modular, feature-informed report generation for renal imaging. Future efforts will focus on extending this pipeline to 3D CT volumes and further improving clinical fidelity in multimodal medical AI systems.

Paperid: 5796, https://arxiv.org/pdf/2506.23502.pdf

Abstract:
Driven by large-scale contrastive vision-language pre-trained models such as CLIP, recent advancements in the image-text matching task have achieved remarkable success in representation learning. Due to image-level visual-language alignment, CLIP falls short in understanding fine-grained details such as object attributes and spatial relationships between objects. Recent efforts have attempted to compel CLIP to acquire structured visual representations by introducing prompt learning to achieve object-level alignment. While achieving promising results, they still lack the capability to perceive actions, which are crucial for describing the states or relationships between objects. Therefore, we propose to endow CLIP with fine-grained action-level understanding by introducing an LLM-enhanced action-aware multi-modal prompt-tuning method, incorporating the action-related external knowledge generated by large language models (LLMs). Specifically, we design an action triplet prompt and an action state prompt to exploit compositional semantic knowledge and state-related causal knowledge implicitly stored in LLMs. Subsequently, we propose an adaptive interaction module to aggregate attentive visual features conditioned on action-aware prompted knowledge for establishing discriminative and action-aware visual representations, which further improves the performance. Comprehensive experimental results on two benchmark datasets demonstrate the effectiveness of our method.

Paperid: 5797, https://arxiv.org/pdf/2506.22636.pdf

Abstract:
Vision Language Models (VLMs) show impressive capabilities in integrating and reasoning with both visual and language data. But these models make mistakes. A common finding -- similar to LLMs -- is their tendency to hallucinate, i.e., generate plausible sounding text which is not grounded in the visual input, or at worst, is contradictory. A growing consensus attributes this behavior to an over-reliance on language -- especially as the generation progresses, the model suffers from a ``fading memory effect'' with respect to the provided visual input. We study mechanisms by which this behavior can be controlled. Specifically, using ideas from geometric algebra and relational compositions, we propose the addition of a small, trainable module (named ReCo) on top of any VLM -- no other modification is needed. We show that such a lightweight module is able to mitigate the fading memory effect on three of the most widely used VLMs (InstructBLIP, LlaVA, MiniGPT4), where we see performance improvements on multiple benchmarks. Additionally, we show that our module can be combined with many of the other approaches for reducing hallucination where we achieve improved results for each one.

Paperid: 5798, https://arxiv.org/pdf/2506.22395.pdf

Abstract:
Vision-Language Models (VLMs) have achieved impressive performance across a wide range of multimodal tasks, yet they often exhibit inconsistent behavior when faced with semantically equivalent inputs, undermining their reliability and robustness. Recent benchmarks, such as MM-R3, highlight that even state-of-the-art VLMs can produce divergent predictions across semantically equivalent inputs, despite maintaining high average accuracy. Prior work addresses this issue by modifying model architectures or conducting large-scale fine-tuning on curated datasets. In contrast, we propose a simple and effective test-time consistency framework that enhances semantic consistency without supervised re-training. Our method is entirely post-hoc, model-agnostic, and applicable to any VLM with access to its weights. Given a single test point, we enforce consistent predictions via two complementary objectives: (i) a Cross-Entropy Agreement Loss that aligns predictive distributions across semantically equivalent inputs, and (ii) a Pseudo-Label Consistency Loss that draws outputs toward a self-averaged consensus. Our method is plug-and-play and leverages information from a single test input itself to improve consistency. Experiments on the MM-R3 benchmark show that our framework yields substantial gains in consistency across state-of-the-art models, establishing a new direction for inference-time adaptation in multimodal learning.

Paperid: 5799, https://arxiv.org/pdf/2506.20944.pdf

Abstract:
The rapid spread of misinformation in mobile and wireless networks presents critical security challenges. This study introduces a training-free, retrieval-based multimodal fact verification system that leverages pretrained vision-language models and large language models for credibility assessment. By dynamically retrieving and cross-referencing trusted data sources, our approach mitigates vulnerabilities of traditional training-based models, such as adversarial attacks and data poisoning. Additionally, its lightweight design enables seamless edge device integration without extensive on-device processing. Experiments on two fact-checking benchmarks achieve SOTA results, confirming its effectiveness in misinformation detection and its robustness against various attack vectors, highlighting its potential to enhance security in mobile and wireless communication environments.

Paperid: 5800, https://arxiv.org/pdf/2506.17664.pdf

Abstract:
Hallucinations in large vision-language models (LVLMs) often stem from the model's sensitivity to image tokens during decoding, as evidenced by attention peaks observed when generating both real and hallucinated entities. To address this, we propose Memory-Driven Sparse Attention Matrix (MDSAM) , a novel training-free approach that dynamically captures and refines the attention allocated to image tokens at each layer. MDSAM memorizes attention patterns and activates updates through alignment during decoding, enhancing focus on relevant image tokens while effectively reducing hallucinations. We evaluate MDSAM on multiple benchmarks for tasks such as image captioning and visual question answering, demonstrating its ability to consistently reduce hallucinations and improve reliability. Compatible with various LVLM architectures, MDSAM highlights its adaptability and effectiveness in mitigating hallucinations without requiring additional training or external tools.

Paperid: 5801, https://arxiv.org/pdf/2506.11253.pdf

Abstract:
Machine unlearning removes certain training data points and their influence on AI models (e.g., when a data owner revokes their decision to allow models to learn from the data). In this position paper, we propose to lift data-tracing machine unlearning to knowledge-tracing for foundation models (FMs). We support this position based on practical needs and insights from cognitive studies. Practically, tracing data cannot meet the diverse unlearning requests for FMs, which may be from regulators, enterprise users, product teams, etc., having no access to FMs' massive training data. Instead, it is convenient for these parties to issue an unlearning request about the knowledge or capability FMs (should not) possess. Cognitively, knowledge-tracing unlearning aligns with how the human brain forgets more closely than tracing individual training data points. Finally, we provide a concrete case study about a vision-language FM to illustrate how an unlearner might instantiate the knowledge-tracing machine unlearning paradigm.

Paperid: 5802, https://arxiv.org/pdf/2506.11166.pdf

Abstract:
As a cornerstone of patient care, clinical decision-making significantly influences patient outcomes and can be enhanced by large language models (LLMs). Although LLMs have demonstrated remarkable performance, their application to visual question answering in medical imaging, particularly for reasoning-based diagnosis, remains largely unexplored. Furthermore, supervised fine-tuning for reasoning tasks is largely impractical due to limited data availability and high annotation costs. In this work, we introduce a zero-shot framework for reliable medical image diagnosis that enhances the reasoning capabilities of LLMs in clinical settings through test-time scaling. Given a medical image and a textual prompt, a vision-language model processes a medical image along with a corresponding textual prompt to generate multiple descriptions or interpretations of visual features. These interpretations are then fed to an LLM, where a test-time scaling strategy consolidates multiple candidate outputs into a reliable final diagnosis. We evaluate our approach across various medical imaging modalities -- including radiology, ophthalmology, and histopathology -- and demonstrate that the proposed test-time scaling strategy enhances diagnostic accuracy for both our and baseline methods. Additionally, we provide an empirical analysis showing that the proposed approach, which allows unbiased prompting in the first stage, improves the reliability of LLM-generated diagnoses and enhances classification accuracy.

Paperid: 5803, https://arxiv.org/pdf/2506.10689.pdf

Abstract:
Accurate automatic screening of minors in unconstrained images demands models that are robust to distribution shift and resilient to the children under-representation in publicly available data. To overcome these issues, we propose a multi-task architecture with dedicated under/over-age discrimination tasks based on a frozen FaRL vision-language backbone joined with a compact two-layer MLP that shares features across one age-regression head and four binary under-age heads for age thresholds of 12, 15, 18, and 21 years, focusing on the legally critical age range. To address the severe class imbalance, we introduce an $Î±$-reweighted focal-style loss and age-balanced mini-batch sampling, which equalizes twelve age bins during stochastic optimization. Further improvement is achieved with an age gap that removes edge cases from the loss. Moreover, we set a rigorous evaluation by proposing the Overall Under-Age Benchmark, with 303k cleaned training images and 110k test images, defining both the "ASORES-39k" restricted overall test, which removes the noisiest domains, and the age estimation wild shifts test "ASWIFT-20k" of 20k-images, stressing extreme pose ($>$45Â°), expression, and low image quality to emulate real-world shifts. Trained on the cleaned overall set with resampling and age gap, our multiage model "F" lowers the root-mean-square-error on the ASORES-39k restricted test from 5.733 (age-only baseline) to 5.656 years and lifts under-18 detection from F2 score of 0.801 to 0.857 at 1% false-adult rate. Under the domain shift to the wild data of ASWIFT-20k, the same configuration nearly sustains 0.99 recall while boosting F2 from 0.742 to 0.833 with respect to the age-only baseline, demonstrating strong generalization under distribution shift. For the under-12 and under-15 tasks, the respective boosts in F2 are from 0.666 to 0.955 and from 0.689 to 0.916, respectively.

Paperid: 5804, https://arxiv.org/pdf/2506.07202.pdf

Abstract:
Multimodal Large Language Models (MLLMs) show impressive vision-language benchmark performance, yet growing concerns about data contamination (test set exposure during training) risk masking true generalization. This concern extends to reasoning MLLMs, often fine-tuned via reinforcement learning from potentially contaminated base models. We propose a novel dynamic evaluation framework to rigorously assess MLLM generalization, moving beyond static benchmarks. Instead of perturbing inputs, we perturb the task itself. Using the same visual input, models are evaluated across a family of tasks (e.g., QA, captioning, question posing, verification) to probe diverse capabilities. This task perturbation reveals whether model performance is robust or reliant on superficial task-specific cues. Our approach is analogous to loss landscape sharpness: models overfit or contaminated for a single task (sharp minima) falter under task shifts, unlike models with generalizable solutions (flatter minima). We developed an automated pipeline with a calibrated judge scoring open-ended generations (captions, questions) using paraphrase and corruption sampling. Applying this framework to leading image/video MLLMs on benchmarks including MME, RealWorldQA, and CVRR-ES, we analyze each model's cross-task "ability vector." We demonstrate that fine-tuning on simulated test data (extreme contamination) drastically sharpens task-specific performance but harms overall generalization. Our dynamic task perturbation offers deeper insights into MLLM generalization, distinguishing genuine understanding from spurious leakage or overfitting.

Paperid: 5805, https://arxiv.org/pdf/2506.05439.pdf

Abstract:
Many vision-language models (VLMs) that prove very effective at a range of multimodal task, build on CLIP-based vision encoders, which are known to have various limitations. We investigate the hypothesis that the strong language backbone in VLMs compensates for possibly weak visual features by contextualizing or enriching them. Using three CLIP-based VLMs, we perform controlled self-attention ablations on a carefully designed probing task. Our findings show that despite known limitations, CLIP visual representations offer ready-to-read semantic information to the language decoder. However, in scenarios of reduced contextualization in the visual representations, the language decoder can largely compensate for the deficiency and recover performance. This suggests a dynamic division of labor in VLMs and motivates future architectures that offload more visual processing to the language decoder.

Paperid: 5806, https://arxiv.org/pdf/2506.02896.pdf

Abstract:
The real world is messy and unstructured. Uncovering critical information often requires active, goal-driven exploration. It remains to be seen whether Vision-Language Models (VLMs), which recently emerged as a popular zero-shot tool in many difficult tasks, can operate effectively in such conditions. In this paper, we answer this question by introducing FlySearch, a 3D, outdoor, photorealistic environment for searching and navigating to objects in complex scenes. We define three sets of scenarios with varying difficulty and observe that state-of-the-art VLMs cannot reliably solve even the simplest exploration tasks, with the gap to human performance increasing as the tasks get harder. We identify a set of central causes, ranging from vision hallucination, through context misunderstanding, to task planning failures, and we show that some of them can be addressed by finetuning. We publicly release the benchmark, scenarios, and the underlying codebase.

Paperid: 5807, https://arxiv.org/pdf/2506.02229.pdf

Abstract:
Pathological examination of the placenta is an effective method for detecting and mitigating health risks associated with childbirth. Recent advancements in AI have enabled the use of photographs of the placenta and pathology reports for detecting and classifying signs of childbirth-related pathologies. However, existing automated methods are computationally extensive, which limits their deployability. We propose two modifications to vision-language contrastive learning (VLC) frameworks to enhance their accuracy and efficiency: (1) text-anchored vision-language contrastive knowledge distillation (VLCD)-a new knowledge distillation strategy for medical VLC pretraining, and (2) unsupervised predistillation using a large natural images dataset for improved initialization. Our approach distills efficient neural networks that match or surpass the teacher model in performance while achieving model compression and acceleration. Our results showcase the value of unsupervised predistillation in improving the performance and robustness of our approach, specifically for lower-quality images. VLCD serves as an effective way to improve the efficiency and deployability of medical VLC approaches, making AI-based healthcare solutions more accessible, especially in resource-constrained environments.

Paperid: 5808, https://arxiv.org/pdf/2505.24519.pdf

Abstract:
We introduce AMIA, a lightweight, inference-only defense for Large Vision-Language Models (LVLMs) that (1) Automatically Masks a small set of text-irrelevant image patches to disrupt adversarial perturbations, and (2) conducts joint Intention Analysis to uncover and mitigate hidden harmful intents before response generation. Without any retraining, AMIA improves defense success rates across diverse LVLMs and jailbreak benchmarks from an average of 52.4% to 81.7%, preserves general utility with only a 2% average accuracy drop, and incurs only modest inference overhead. Ablation confirms both masking and intention analysis are essential for a robust safety-utility trade-off.

Paperid: 5809, https://arxiv.org/pdf/2505.24317.pdf

Abstract:
Reinforcement learning (RL) in autonomous driving employs a trial-and-error mechanism, enhancing robustness in unpredictable environments. However, crafting effective reward functions remains challenging, as conventional approaches rely heavily on manual design and demonstrate limited efficacy in complex scenarios. To address this issue, this study introduces a responsibility-oriented reward function that explicitly incorporates traffic regulations into the RL framework. Specifically, we introduced a Traffic Regulation Knowledge Graph and leveraged Vision-Language Models alongside Retrieval-Augmented Generation techniques to automate reward assignment. This integration guides agents to adhere strictly to traffic laws, thus minimizing rule violations and optimizing decision-making performance in diverse driving conditions. Experimental validations demonstrate that the proposed methodology significantly improves the accuracy of assigning accident responsibilities and effectively reduces the agent's liability in traffic incidents.

Paperid: 5810, https://arxiv.org/pdf/2505.24208.pdf

Abstract:
Ensuring Vision-Language Models (VLMs) generate safe outputs is crucial for their reliable deployment. However, LVLMs suffer from drastic safety degradation compared to their LLM backbone. Even blank or irrelevant images can trigger LVLMs to generate harmful responses to prompts that would otherwise be refused in text-only contexts. The modality gap between image and text representations has been recently hypothesized to contribute to safety degradation of LVLMs. However, if and how the amount of modality gap affects LVLMs' safety is not studied. In this work, we show that the amount of modality gap is highly inversely correlated with VLMs' safety. Then, we show that this modality gap is introduced during pretraining LVLMs and persists through fine-tuning. Inspired by this observation, we propose a regularization to reduce the modality gap during pretraining. Our extensive experiments on LLaVA v1.5, ShareGPT4V, and MiniGPT-4 show that our method substantially improves safety alignment of LVLMs, reducing unsafe rate by up to 16.3% without compromising performance, and can further boost existing defenses by up to 18.2%.

Paperid: 5811, https://arxiv.org/pdf/2505.24173.pdf

Abstract:
Vision-language models (VLMs) exhibit strong zero-shot generalization on natural images and show early promise in interpretable medical image analysis. However, existing benchmarks do not systematically evaluate whether these models truly reason like human clinicians or merely imitate superficial patterns. To address this gap, we propose DrVD-Bench, the first multimodal benchmark for clinical visual reasoning. DrVD-Bench consists of three modules: Visual Evidence Comprehension, Reasoning Trajectory Assessment, and Report Generation Evaluation, comprising a total of 7,789 image-question pairs. Our benchmark covers 20 task types, 17 diagnostic categories, and five imaging modalities-CT, MRI, ultrasound, radiography, and pathology. DrVD-Bench is explicitly structured to reflect the clinical reasoning workflow from modality recognition to lesion identification and diagnosis. We benchmark 19 VLMs, including general-purpose and medical-specific, open-source and proprietary models, and observe that performance drops sharply as reasoning complexity increases. While some models begin to exhibit traces of human-like reasoning, they often still rely on shortcut correlations rather than grounded visual understanding. DrVD-Bench offers a rigorous and structured evaluation framework to guide the development of clinically trustworthy VLMs.

Paperid: 5812, https://arxiv.org/pdf/2505.23518.pdf

Abstract:
Autonomous agentic AI systems powered by vision-language models (VLMs) are rapidly advancing toward real-world deployment, yet their cross-modal reasoning capabilities introduce new attack surfaces for adversarial manipulation that exploit semantic reasoning across modalities. Existing adversarial attacks typically rely on visible pixel perturbations or require privileged model or environment access, making them impractical for stealthy, real-world exploitation. We introduce TRAP, a generative adversarial framework that manipulates the agent's decision-making using diffusion-based semantic injections. Our method combines negative prompt-based degradation with positive semantic optimization, guided by a Siamese semantic network and layout-aware spatial masking. Without requiring access to model internals, TRAP produces visually natural images yet induces consistent selection biases in agentic AI systems. We evaluate TRAP on the Microsoft Common Objects in Context (COCO) dataset, building multi-candidate decision scenarios. Across these scenarios, TRAP achieves a 100% attack success rate on leading models, including LLaVA-34B, Gemma3, and Mistral-3.1, significantly outperforming baselines such as SPSA, Bandit, and standard diffusion approaches. These results expose a critical vulnerability: Autonomous agents can be consistently misled through human-imperceptible cross-modal manipulations. These findings highlight the need for defense strategies beyond pixel-level robustness to address semantic vulnerabilities in cross-modal decision-making.

Paperid: 5813, https://arxiv.org/pdf/2505.21556.pdf

Abstract:
Optimization-based jailbreaks typically adopt the Toxic-Continuation setting in large vision-language models (LVLMs), following the standard next-token prediction objective. In this setting, an adversarial image is optimized to make the model predict the next token of a toxic prompt. However, we find that the Toxic-Continuation paradigm is effective at continuing already-toxic inputs, but struggles to induce safety misalignment when explicit toxic signals are absent. We propose a new paradigm: Benign-to-Toxic (B2T) jailbreak. Unlike prior work, we optimize adversarial images to induce toxic outputs from benign conditioning. Since benign conditioning contains no safety violations, the image alone must break the model's safety mechanisms. Our method outperforms prior approaches, transfers in black-box settings, and complements text-based jailbreaks. These results reveal an underexplored vulnerability in multimodal alignment and introduce a fundamentally new direction for jailbreak approaches.

Paperid: 5814, https://arxiv.org/pdf/2505.20726.pdf

Abstract:
Building embodied agents capable of accomplishing arbitrary tasks is a core objective towards achieving embodied artificial general intelligence (E-AGI). While recent work has advanced such general robot policies, their training and evaluation are often limited to tasks within specific scenes, involving restricted instructions and scenarios. Existing benchmarks also typically rely on manual annotation of limited tasks in a few scenes. We argue that exploring the full spectrum of feasible tasks within any given scene is crucial, as they provide both extensive benchmarks for evaluation and valuable resources for agent improvement. Towards this end, we introduce ManiTaskGen, a novel system that automatically generates comprehensive, diverse, feasible mobile manipulation tasks for any given scene. The generated tasks encompass both process-based, specific instructions (e.g., "move object from X to Y") and outcome-based, abstract instructions (e.g., "clear the table"). We apply ManiTaskGen to both simulated and real-world scenes, demonstrating the validity and diversity of the generated tasks. We then leverage these tasks to automatically construct benchmarks, thoroughly evaluating the embodied decision-making capabilities of agents built upon existing vision-language models (VLMs). Furthermore, we propose a simple yet effective method that utilizes ManiTaskGen tasks to enhance embodied decision-making. Overall, this work presents a universal task generation framework for arbitrary scenes, facilitating both benchmarking and improvement of embodied decision-making agents.

Paperid: 5815, https://arxiv.org/pdf/2505.20537.pdf

Abstract:
Clear communication of robot intent fosters transparency and interpretability in physical human-robot interaction (pHRI), particularly during assistive tasks involving direct human-robot contact. We introduce CoRI, a pipeline that automatically generates natural language communication of a robot's upcoming actions directly from its motion plan and visual perception. Our pipeline first processes the robot's image view to identify human poses and key environmental features. It then encodes the planned 3D spatial trajectory (including velocity and force) onto this view, visually grounding the path and its dynamics. CoRI queries a vision-language model with this visual representation to interpret the planned action within the visual context before generating concise, user-directed statements, without relying on task-specific information. Results from a user study involving robot-assisted feeding, bathing, and shaving tasks across two different robots indicate that CoRI leads to statistically significant difference in communication clarity compared to a baseline communication strategy. Specifically, CoRI effectively conveys not only the robot's high-level intentions but also crucial details about its motion and any collaborative user action needed. Video and code of our project can be found on our project website: https://cori-phri.github.io/

Paperid: 5816, https://arxiv.org/pdf/2505.17536.pdf

Abstract:
Conversations are usually structured by roles -- who is speaking, who's being addressed, and who's listening -- and unfold in threads that break with changes in speaker floor or topical focus. While large language models (LLMs) have shown incredible capabilities in dialogue and reasoning, their ability to understand fine-grained conversational structure, especially in multi-modal, multi-party settings, remains underexplored. To address this gap, we introduce a suite of tasks focused on conversational role attribution (speaker, addressees, side-participants) and conversation threading (utterance linking and clustering), drawing on conversation analysis and sociolinguistics. To support those tasks, we present a human annotated dataset of 4,398 annotations for speakers and reply-to relationship, 5,755 addressees, and 3,142 side-participants. We evaluate popular audio-visual LLMs and vision-language models on our dataset, and our experimental results suggest that multimodal conversational structure understanding remains challenging. The most performant audio-visual LLM outperforms all vision-language models across all metrics, especially in speaker and addressee recognition. However, its performance drops significantly when conversation participants are anonymized. The number of conversation participants in a clip is the strongest negative predictor of role-attribution performance, while acoustic clarity (measured by pitch and spectral centroid) and detected face coverage yield positive associations. We hope this work lays the groundwork for future evaluation and development of multimodal LLMs that can reason more effectively about conversation structure.

Paperid: 5817, https://arxiv.org/pdf/2505.17202.pdf

Abstract:
Data visualizations are powerful tools for communicating patterns in quantitative data. Yet understanding any data visualization is no small feat -- succeeding requires jointly making sense of visual, numerical, and linguistic inputs arranged in a conventionalized format one has previously learned to parse. Recently developed vision-language models are, in principle, promising candidates for developing computational models of these cognitive operations. However, it is currently unclear to what degree these models emulate human behavior on tasks that involve reasoning about data visualizations. This gap reflects limitations in prior work that has evaluated data visualization understanding in artificial systems using measures that differ from those typically used to assess these abilities in humans. Here we evaluated eight vision-language models on six data visualization literacy assessments designed for humans and compared model responses to those of human participants. We found that these models performed worse than human participants on average, and this performance gap persisted even when using relatively lenient criteria to assess model performance. Moreover, while relative performance across items was somewhat correlated between models and humans, all models produced patterns of errors that were reliably distinct from those produced by human participants. Taken together, these findings suggest significant opportunities for further development of artificial systems that might serve as useful models of how humans reason about data visualizations. All code and data needed to reproduce these results are available at: https://osf.io/e25mu/?view_only=399daff5a14d4b16b09473cf19043f18.

Paperid: 5818, https://arxiv.org/pdf/2505.10672.pdf

Abstract:
Efficient and accurate multi-organ segmentation from abdominal CT volumes is a fundamental challenge in medical image analysis. Existing 3D segmentation approaches are computationally and memory intensive, often processing entire volumes that contain many anatomically irrelevant slices. Meanwhile, 2D methods suffer from class imbalance and lack cross-view contextual awareness. To address these limitations, we propose a novel, anatomically-aware slice selector pipeline that reduces input volume prior to segmentation. Our unified framework introduces a vision-language model (VLM) for cross-view organ presence detection using fused tri-slice (2.5D) representations from axial, sagittal, and coronal planes. Our proposed model acts as an "expert" in anatomical localization, reasoning over multi-view representations to selectively retain slices with high structural relevance. This enables spatially consistent filtering across orientations while preserving contextual cues. More importantly, since standard segmentation metrics such as Dice or IoU fail to measure the spatial precision of such slice selection, we introduce a novel metric, Slice Localization Concordance (SLC), which jointly captures anatomical coverage and spatial alignment with organ-centric reference slices. Unlike segmentation-specific metrics, SLC provides a model-agnostic evaluation of localization fidelity. Our model offers substantial improvement gains against several baselines across all organs, demonstrating both accurate and reliable organ-focused slice filtering. These results show that our method enables efficient and spatially consistent organ filtering, thereby significantly reducing downstream segmentation cost while maintaining high anatomical fidelity.

Paperid: 5819, https://arxiv.org/pdf/2505.08194.pdf

Abstract:
Recent advancements in integrating tactile sensing with vision-language models (VLMs) have demonstrated remarkable potential for robotic multimodal perception. However, existing tactile descriptions remain limited to superficial attributes like texture, neglecting critical contact states essential for robotic manipulation. To bridge this gap, we propose CLTP, an intuitive and effective language tactile pretraining framework that aligns tactile 3D point clouds with natural language in various contact scenarios, thus enabling contact-state-aware tactile language understanding for contact-rich manipulation tasks. We first collect a novel dataset of 50k+ tactile 3D point cloud-language pairs, where descriptions explicitly capture multidimensional contact states (e.g., contact location, shape, and force) from the tactile sensor's perspective. CLTP leverages a pre-aligned and frozen vision-language feature space to bridge holistic textual and tactile modalities. Experiments validate its superiority in three downstream tasks: zero-shot 3D classification, contact state classification, and tactile 3D large language model (LLM) interaction. To the best of our knowledge, this is the first study to align tactile and language representations from the contact state perspective for manipulation tasks, providing great potential for tactile-language-action model learning. Code and datasets are open-sourced at https://sites.google.com/view/cltp/.

Paperid: 5820, https://arxiv.org/pdf/2505.06663.pdf

Abstract:
Open-vocabulary video visual relationship detection aims to detect objects and their relationships in videos without being restricted by predefined object or relationship categories. Existing methods leverage the rich semantic knowledge of pre-trained vision-language models such as CLIP to identify novel categories. They typically adopt a cascaded pipeline to first detect objects and then classify relationships based on the detected objects, which may lead to error propagation and thus suboptimal performance. In this paper, we propose Mutual EnhancemenT of Objects and Relationships (METOR), a query-based unified framework to jointly model and mutually enhance object detection and relationship classification in open-vocabulary scenarios. Under this framework, we first design a CLIP-based contextual refinement encoding module that extracts visual contexts of objects and relationships to refine the encoding of text features and object queries, thus improving the generalization of encoding to novel categories. Then we propose an iterative enhancement module to alternatively enhance the representations of objects and relationships by fully exploiting their interdependence to improve recognition performance. Extensive experiments on two public datasets, VidVRD and VidOR, demonstrate that our framework achieves state-of-the-art performance.

Paperid: 5821, https://arxiv.org/pdf/2505.05736.pdf

Abstract:
The scarcity of high-quality multimodal biomedical data limits the ability to effectively fine-tune pretrained Large Language Models (LLMs) for specialized biomedical tasks. To address this challenge, we introduce MINT (Multimodal Integrated kNowledge Transfer), a framework that aligns unimodal large decoder models with domain-specific decision patterns from multimodal biomedical data through preference optimization. While MINT supports different optimization techniques, we primarily implement it with the Odds Ratio Preference Optimization (ORPO) framework as its backbone. This strategy enables the aligned LLMs to perform predictive tasks using text-only or image-only inputs while retaining knowledge learnt from multimodal data. MINT leverages an upstream multimodal machine learning (MML) model trained on high-quality multimodal data to transfer domain-specific insights to downstream text-only or image-only LLMs. We demonstrate its effectiveness through two key applications: (1) Rare genetic disease prediction from texts, where MINT uses a multimodal encoder model, trained on facial photos and clinical notes, to generate a preference dataset for aligning a lightweight Llama 3.2-3B-Instruct. Despite relying on text input only, the MINT-derived model outperforms models trained with SFT, RAG, or DPO, and even outperforms Llama 3.1-405B-Instruct. (2) Tissue type classification using cell nucleus images, where MINT uses a vision-language foundation model as the preference generator, containing knowledge learnt from both text and histopathological images to align downstream image-only models. The resulting MINT-derived model significantly improves the performance of Llama 3.2-Vision-11B-Instruct on tissue type classification. In summary, MINT provides an effective strategy to align unimodal LLMs with high-quality multimodal expertise through preference optimization.

Paperid: 5822, https://arxiv.org/pdf/2505.05318.pdf

Abstract:
The rapid adoption of Vision Language Models (VLMs), pre-trained on large image-text and video-text datasets, calls for protecting and informing users about when to trust these systems. This survey reviews studies on trust dynamics in user-VLM interactions, through a multi-disciplinary taxonomy encompassing different cognitive science capabilities, collaboration modes, and agent behaviours. Literature insights and findings from a workshop with prospective VLM users inform preliminary requirements for future VLM trust studies.

Paperid: 5823, https://arxiv.org/pdf/2505.03807.pdf

Abstract:
Video story interaction enables viewers to engage with and explore narrative content for personalized experiences. However, existing methods are limited to user selection, specially designed narratives, and lack customization. To address this, we propose an interactive system based on user intent. Our system uses a Vision Language Model (VLM) to enable machines to understand video stories, combining Retrieval-Augmented Generation (RAG) and a Multi-Agent System (MAS) to create evolving characters and scene experiences. It includes three stages: 1) Video story processing, utilizing VLM and prior knowledge to simulate human understanding of stories across three modalities. 2) Multi-space chat, creating growth-oriented characters through MAS interactions based on user queries and story stages. 3) Scene customization, expanding and visualizing various story scenes mentioned in dialogue. Applied to the Harry Potter series, our study shows the system effectively portrays emergent character social behavior and growth, enhancing the interactive experience in the video story world.

Paperid: 5824, https://arxiv.org/pdf/2505.01790.pdf

Abstract:
Web-based educational videos offer flexible learning opportunities and are becoming increasingly popular. However, improving user engagement and knowledge retention remains a challenge. Automatically generated questions can activate learners and support their knowledge acquisition. Further, they can help teachers and learners assess their understanding. While large language and vision-language models have been employed in various tasks, their application to question generation for educational videos remains underexplored. In this paper, we investigate the capabilities of current vision-language models for generating learning-oriented questions for educational video content. We assess (1) out-of-the-box models' performance; (2) fine-tuning effects on content-specific question generation; (3) the impact of different video modalities on question quality; and (4) in a qualitative study, question relevance, answerability, and difficulty levels of generated questions. Our findings delineate the capabilities of current vision-language models, highlighting the need for fine-tuning and addressing challenges in question diversity and relevance. We identify requirements for future multimodal datasets and outline promising research directions.

Paperid: 5825, https://arxiv.org/pdf/2505.00693.pdf

Abstract:
Recently, natural language has been the primary medium for human-robot interaction. However, its inherent lack of spatial precision introduces challenges for robotic task definition such as ambiguity and verbosity. Moreover, in some public settings where quiet is required, such as libraries or hospitals, verbal communication with robots is inappropriate. To address these limitations, we introduce the Robotic Visual Instruction (RoVI), a novel paradigm to guide robotic tasks through an object-centric, hand-drawn symbolic representation. RoVI effectively encodes spatial-temporal information into human-interpretable visual instructions through 2D sketches, utilizing arrows, circles, colors, and numbers to direct 3D robotic manipulation. To enable robots to understand RoVI better and generate precise actions based on RoVI, we present Visual Instruction Embodied Workflow (VIEW), a pipeline formulated for RoVI-conditioned policies. This approach leverages Vision-Language Models (VLMs) to interpret RoVI inputs, decode spatial and temporal constraints from 2D pixel space via keypoint extraction, and then transform them into executable 3D action sequences. We additionally curate a specialized dataset of 15K instances to fine-tune small VLMs for edge deployment,enabling them to effectively learn RoVI capabilities. Our approach is rigorously validated across 11 novel tasks in both real and simulated environments, demonstrating significant generalization capability. Notably, VIEW achieves an 87.5% success rate in real-world scenarios involving unseen tasks that feature multi-step actions, with disturbances, and trajectory-following requirements. Project website: https://robotic-visual-instruction.github.io/

Paperid: 5826, https://arxiv.org/pdf/2504.19266.pdf

Abstract:
Real-time open-vocabulary scene understanding is essential for efficient 3D perception in applications such as vision-language navigation, embodied intelligence, and augmented reality. However, existing methods suffer from imprecise instance segmentation, static semantic updates, and limited handling of complex queries. To address these issues, we present OpenFusion++, a TSDF-based real-time 3D semantic-geometric reconstruction system. Our approach refines 3D point clouds by fusing confidence maps from foundational models, dynamically updates global semantic labels via an adaptive cache based on instance area, and employs a dual-path encoding framework that integrates object attributes with environmental context for precise query responses. Experiments on the ICL, Replica, ScanNet, and ScanNet++ datasets demonstrate that OpenFusion++ significantly outperforms the baseline in both semantic accuracy and query responsiveness.

Paperid: 5827, https://arxiv.org/pdf/2504.16047.pdf

Abstract:
Foundation models, trained on vast amounts of data using self-supervised techniques, have emerged as a promising frontier for advancing artificial intelligence (AI) applications in medicine. This study evaluates three different vision-language foundation models (RAD-DINO, CheXagent, and BiomedCLIP) on their ability to capture fine-grained imaging features for radiology tasks. The models were assessed across classification, segmentation, and regression tasks for pneumothorax and cardiomegaly on chest radiographs. Self-supervised RAD-DINO consistently excelled in segmentation tasks, while text-supervised CheXagent demonstrated superior classification performance. BiomedCLIP showed inconsistent performance across tasks. A custom segmentation model that integrates global and local features substantially improved performance for all foundation models, particularly for challenging pneumothorax segmentation. The findings highlight that pre-training methodology significantly influences model performance on specific downstream tasks. For fine-grained segmentation tasks, models trained without text supervision performed better, while text-supervised models offered advantages in classification and interpretability. These insights provide guidance for selecting foundation models based on specific clinical applications in radiology.

Paperid: 5828, https://arxiv.org/pdf/2504.04988.pdf

Abstract:
Recent progress in VLMs has demonstrated impressive capabilities across a variety of tasks in the natural image domain. Motivated by these advancements, the remote sensing community has begun to adopt VLMs for remote sensing vision-language tasks, including scene understanding, image captioning, and visual question answering. However, existing remote sensing VLMs typically rely on closed-set scene understanding and focus on generic scene descriptions, yet lack the ability to incorporate external knowledge. This limitation hinders their capacity for semantic reasoning over complex or context-dependent queries that involve domain-specific or world knowledge. To address these challenges, we first introduced a multimodal Remote Sensing World Knowledge (RSWK) dataset, which comprises high-resolution satellite imagery and detailed textual descriptions for 14,141 well-known landmarks from 175 countries, integrating both remote sensing domain knowledge and broader world knowledge. Building upon this dataset, we proposed a novel Remote Sensing Retrieval-Augmented Generation (RS-RAG) framework, which consists of two key components. The Multi-Modal Knowledge Vector Database Construction module encodes remote sensing imagery and associated textual knowledge into a unified vector space. The Knowledge Retrieval and Response Generation module retrieves and re-ranks relevant knowledge based on image and/or text queries, and incorporates the retrieved content into a knowledge-augmented prompt to guide the VLM in producing contextually grounded responses. We validated the effectiveness of our approach on three representative vision-language tasks, including image captioning, image classification, and visual question answering, where RS-RAG significantly outperformed state-of-the-art baselines.

Paperid: 5829, https://arxiv.org/pdf/2504.02918.pdf

Abstract:
Recent advances in image and video generation raise hopes that these models possess world modeling capabilities, the ability to generate realistic, physically plausible videos. This could revolutionize applications in robotics, autonomous driving, and scientific simulation. However, before treating these models as world models, we must ask: Do they adhere to physical conservation laws? To answer this, we introduce Morpheus, a benchmark for evaluating video generation models on physical reasoning. It features 80 real-world videos capturing physical phenomena, guided by conservation laws. Since artificial generations lack ground truth, we assess physical plausibility using physics-informed metrics evaluated with respect to infallible conservation laws known per physical setting, leveraging advances in physics-informed neural networks and vision-language foundation models. Our findings reveal that even with advanced prompting and video conditioning, current models struggle to encode physical principles despite generating aesthetically pleasing videos. All data, leaderboard, and code are open-sourced at our project page.

Paperid: 5830, https://arxiv.org/pdf/2504.01324.pdf

Abstract:
This paper is a pioneering work attempting to address abstract visual reasoning (AVR) problems for large vision-language models (VLMs). We make a common LLaVA-NeXT 7B model capable of perceiving and reasoning about specific AVR problems, surpassing both open-sourced (e.g., Qwen-2-VL-72B) and closed-sourced powerful VLMs (e.g., GPT-4o) with significant margin. This is a great breakthrough since almost all previous VLMs fail or show nearly random performance on representative AVR benchmarks. Our key success is our innovative data synthesis and post-training process, aiming to fully relieve the task difficulty and elicit the model to learn, step by step. Our 7B model is also shown to be behave well on AVR without sacrificing common multimodal comprehension abilities. We hope our paper could serve as an early effort in this area and would inspire further research in abstract visual reasoning.

Paperid: 5831, https://arxiv.org/pdf/2503.23125.pdf

Abstract:
The visual world is fundamentally compositional. Visual scenes are defined by the composition of objects and their relations. Hence, it is essential for computer vision systems to reflect and exploit this compositionality to achieve robust and generalizable scene understanding. While major strides have been made toward the development of general-purpose, multimodal generative models, including both text-to-image models and multimodal vision-language models, it remains unclear whether these systems are capable of accurately generating and interpreting scenes involving the composition of multiple objects and relations. In this work, we present an evaluation of the compositional visual processing capabilities in the current generation of text-to-image (DALL-E 3) and multimodal vision-language models (GPT-4V, GPT-4o, Claude Sonnet 3.5, QWEN2-VL-72B, and InternVL2.5-38B), and compare the performance of these systems to human participants. The results suggest that these systems display some ability to solve compositional and relational tasks, showing notable improvements over the previous generation of multimodal models, but with performance nevertheless well below the level of human participants, particularly for more complex scenes involving many ($>5$) objects and multiple relations. These results highlight the need for further progress toward compositional understanding of visual scenes.

Paperid: 5832, https://arxiv.org/pdf/2503.21910.pdf

Abstract:
We introduce JEEM, a benchmark designed to evaluate Vision-Language Models (VLMs) on visual understanding across four Arabic-speaking countries: Jordan, The Emirates, Egypt, and Morocco. JEEM includes the tasks of image captioning and visual question answering, and features culturally rich and regionally diverse content. This dataset aims to assess the ability of VLMs to generalize across dialects and accurately interpret cultural elements in visual contexts. In an evaluation of five prominent open-source Arabic VLMs and GPT-4V, we find that the Arabic VLMs consistently underperform, struggling with both visual understanding and dialect-specific generation. While GPT-4V ranks best in this comparison, the model's linguistic competence varies across dialects, and its visual understanding capabilities lag behind. This underscores the need for more inclusive models and the value of culturally-diverse evaluation paradigms.

Paperid: 5833, https://arxiv.org/pdf/2503.18491.pdf

Abstract:
Visual Question Answering (VQA) requires reasoning across visual and textual modalities, yet Large Vision-Language Models (LVLMs) often lack integrated commonsense knowledge, limiting their robustness in real-world scenarios. To address this, we introduce MAGIC-VQA, a novel framework that enhances VQA by systematically integrating commonsense knowledge with LVLMs. MAGIC-VQA employs a three-stage process: (1) Explicit Knowledge Integration from external sources, (2) By-Type Post-Processing for contextual refinement, and (3) Implicit Knowledge Augmentation using a Graph Neural Network (GNN) for structured reasoning. While GNNs bring greater depth to structured inference, they enable superior relational inference beyond LVLMs. MAGIC-VQA bridges a key gap by unifying commonsensse knowledge with LVLM-driven reasoning, eliminating the need for extensive pre-training or complex prompt tuning. Our framework achieves state-of-the-art performance on benchmark datasets, significantly improving commonsense reasoning in VQA.

Paperid: 5834, https://arxiv.org/pdf/2503.16585.pdf

Abstract:
Language models (LMs) are machine learning models designed to predict linguistic patterns by estimating the probability of word sequences based on large-scale datasets, such as text. LMs have a wide range of applications in natural language processing (NLP) tasks, including autocomplete and machine translation. Although larger datasets typically enhance LM performance, scalability remains a challenge due to constraints in computational power and resources. Distributed computing strategies offer essential solutions for improving scalability and managing the growing computational demand. Further, the use of sensitive datasets in training and deployment raises significant privacy concerns. Recent research has focused on developing decentralized techniques to enable distributed training and inference while utilizing diverse computational resources and enabling edge AI. This paper presents a survey on distributed solutions for various LMs, including large language models (LLMs), vision language models (VLMs), multimodal LLMs (MLLMs), and small language models (SLMs). While LLMs focus on processing and generating text, MLLMs are designed to handle multiple modalities of data (e.g., text, images, and audio) and to integrate them for broader applications. To this end, this paper reviews key advancements across the MLLM pipeline, including distributed training, inference, fine-tuning, and deployment, while also identifying the contributions, limitations, and future areas of improvement. Further, it categorizes the literature based on six primary focus areas of decentralization. Our analysis describes gaps in current methodologies for enabling distributed solutions for LMs and outline future research directions, emphasizing the need for novel solutions to enhance the robustness and applicability of distributed LMs.

Paperid: 5835, https://arxiv.org/pdf/2503.15778.pdf

Abstract:
In autonomous driving, open-ended question answering often suffers from unreliable evaluations because freeform responses require either complex metrics or subjective human judgment. To address this challenge, we introduce AutoDrive-QA, an automatic pipeline that converts existing driving QA datasets (including DriveLM, NuScenes-QA, and LingoQA) into a structured multiple-choice question (MCQ) format. This benchmark systematically assesses perception, prediction, and planning tasks, providing a standardized and objective evaluation framework. AutoDrive-QA employs an automated pipeline that leverages large language models (LLMs) to generate high-quality, contextually relevant distractors based on domain-specific error patterns commonly found in autonomous driving scenarios. To evaluate both general capabilities and generalization performance, we test the benchmark on three public datasets and conduct zero-shot experiments on an unseen dataset. The zero-shot evaluations reveal that GPT-4V leads with 69.57% accuracy -- achieving 74.94% in Perception, 65.33% in Prediction, and 68.45% in Planning -- demonstrating that while all models excel in Perception, they struggle in Prediction. Consequently, AutoDrive-QA establishes a rigorous, unbiased standard for integrating and evaluating different vision-language models across various autonomous driving datasets, thereby improving generalization in this field. We release all the codes in the AutoDrive-QA GitHub Repository.

Paperid: 5836, https://arxiv.org/pdf/2503.15342.pdf

Abstract:
The proliferation of synthetic images generated by advanced AI models poses significant challenges in identifying and understanding manipulated visual content. Current fake image detection methods predominantly rely on binary classification models that focus on accuracy while often neglecting interpretability, leaving users without clear insights into why an image is deemed real or fake. To bridge this gap, we introduce TruthLens, a novel training-free framework that reimagines deepfake detection as a visual question-answering (VQA) task. TruthLens utilizes state-of-the-art large vision-language models (LVLMs) to observe and describe visual artifacts and combines this with the reasoning capabilities of large language models (LLMs) like GPT-4 to analyze and aggregate evidence into informed decisions. By adopting a multimodal approach, TruthLens seamlessly integrates visual and semantic reasoning to not only classify images as real or fake but also provide interpretable explanations for its decisions. This transparency enhances trust and provides valuable insights into the artifacts that signal synthetic content. Extensive evaluations demonstrate that TruthLens outperforms conventional methods, achieving high accuracy on challenging datasets while maintaining a strong emphasis on explainability. By reframing deepfake detection as a reasoning-driven process, TruthLens establishes a new paradigm in combating synthetic media, combining cutting-edge performance with interpretability to address the growing threats of visual disinformation.

Paperid: 5837, https://arxiv.org/pdf/2503.15202.pdf

Abstract:
Robotic systems often face execution failures due to unexpected obstacles, sensor errors, or environmental changes. Traditional failure recovery methods rely on predefined strategies or human intervention, making them less adaptable. This paper presents a unified failure recovery framework that combines Vision-Language Models (VLMs), a reactive planner, and Behavior Trees (BTs) to enable real-time failure handling. Our approach includes pre-execution verification, which checks for potential failures before execution, and reactive failure handling, which detects and corrects failures during execution by verifying existing BT conditions, adding missing preconditions and, when necessary, generating new skills. The framework uses a scene graph for structured environmental perception and an execution history for continuous monitoring, enabling context-aware and adaptive failure handling. We evaluate our framework through real-world experiments with an ABB YuMi robot on tasks like peg insertion, object sorting, and drawer placement, as well as in AI2-THOR simulator. Compared to using pre-execution and reactive methods separately, our approach achieves higher task success rates and greater adaptability. Ablation studies highlight the importance of VLM-based reasoning, structured scene representation, and execution history tracking for effective failure recovery in robotics.

Paperid: 5838, https://arxiv.org/pdf/2503.14853.pdf

Abstract:
Current Large Vision-Language Models (LVLMs) have demonstrated remarkable capabilities in understanding multimodal data, but their potential remains underexplored for deepfake detection due to the misalignment of their knowledge and forensics patterns. To this end, we present a novel framework that unlocks LVLMs' potential capabilities for deepfake detection. Our framework includes a Knowledge-guided Forgery Detector (KFD), a Forgery Prompt Learner (FPL), and a Large Language Model (LLM). The KFD is used to calculate correlations between image features and pristine/deepfake image description embeddings, enabling forgery classification and localization. The outputs of the KFD are subsequently processed by the Forgery Prompt Learner to construct fine-grained forgery prompt embeddings. These embeddings, along with visual and question prompt embeddings, are fed into the LLM to generate textual detection responses. Extensive experiments on multiple benchmarks, including FF++, CDF2, DFD, DFDCP, DFDC, and DF40, demonstrate that our scheme surpasses state-of-the-art methods in generalization performance, while also supporting multi-turn dialogue capabilities.

Paperid: 5839, https://arxiv.org/pdf/2503.10981.pdf

Abstract:
Concept Bottleneck Models (CBMs) map dense, high-dimensional feature representations into a set of human-interpretable concepts which are then combined linearly to make a prediction. However, modern CBMs rely on the CLIP model to establish a mapping from dense feature representations to textual concepts, and it remains unclear how to design CBMs for models other than CLIP. Methods that do not use CLIP instead require manual, labor intensive annotation to associate feature representations with concepts. Furthermore, all CBMs necessitate training a linear classifier to map the extracted concepts to class labels. In this work, we lift all three limitations simultaneously by proposing a method that converts any frozen visual classifier into a CBM without requiring image-concept labels (label-free), without relying on the CLIP model (CLIP-free), and by deriving the linear classifier in a zero-shot manner. Our method is formulated by aligning the original classifier's distribution (over discrete class indices) with its corresponding vision-language counterpart distribution derived from textual class names, while preserving the classifier's performance. The approach requires no ground-truth image-class annotations, and is highly data-efficient and preserves the classifier's reasoning process. Applied and tested on over 40 visual classifiers, our resulting CLIP-free, zero-shot CBM sets a new state of the art, surpassing even supervised CLIP-based CBMs. Finally, we also show that our method can be used for zero-shot image captioning, outperforming existing methods based on CLIP, and achieving state of the art results.

Paperid: 5840, https://arxiv.org/pdf/2503.09443.pdf

Abstract:
Cross-lingual transfer enables vision-language models (VLMs) to perform vision tasks in various languages with training data only in one language. Current approaches rely on large pre-trained multilingual language models. However, they face the curse of multilinguality, sacrificing downstream task performance for multilingual capabilities, struggling with lexical ambiguities, and falling behind recent advances. In this work, we study the scaling laws of systematic generalization with monolingual VLMs for multilingual tasks, focusing on the impact of model size and seen training samples. We propose Florenz, a monolingual encoder-decoder VLM with 0.4B to 11.2B parameters combining the pre-trained VLM Florence-2 and the large language model Gemma-2. Florenz is trained with varying compute budgets on a synthetic dataset that features intentionally incomplete language coverage for image captioning, thus, testing generalization from the fully covered translation task. We show that not only does indirectly learning unseen task-language pairs adhere to a scaling law, but also that with our data generation pipeline and the proposed Florenz model family, image captioning abilities can emerge in a specific language even when only data for the translation task is available. Fine-tuning on a mix of downstream datasets yields competitive performance and demonstrates promising scaling trends in multimodal machine translation (Multi30K, CoMMuTE), lexical disambiguation (CoMMuTE), and image captioning (Multi30K, XM3600, COCO Karpathy).

Paperid: 5841, https://arxiv.org/pdf/2503.03854.pdf

Abstract:
Cross-modal entity linking refers to the ability to align entities and their attributes across different modalities. While cross-modal entity linking is a fundamental skill needed for real-world applications such as multimodal code generation, fake news detection, or scene understanding, it has not been thoroughly studied in the literature. In this paper, we introduce a new task and benchmark to address this gap. Our benchmark, MATE, consists of 5.5k evaluation instances featuring visual scenes aligned with their textual representations. To evaluate cross-modal entity linking performance, we design a question-answering task that involves retrieving one attribute of an object in one modality based on a unique attribute of that object in another modality. We evaluate state-of-the-art Vision-Language Models (VLMs) and humans on this task, and find that VLMs struggle significantly compared to humans, particularly as the number of objects in the scene increases. Our analysis also shows that, while chain-of-thought prompting can improve VLM performance, models remain far from achieving human-level proficiency. These findings highlight the need for further research in cross-modal entity linking and show that MATE is a strong benchmark to support that progress.

Paperid: 5842, https://arxiv.org/pdf/2503.01169.pdf

Abstract:
Ephemeral gullies are a primary cause of soil erosion and their reliable, accurate, and early detection will facilitate significant improvements in the sustainability of global agricultural systems. In our view, prior research has not successfully addressed automated detection of ephemeral gullies from remotely sensed images, so for the first time, we present and evaluate three successful pipelines for ephemeral gully detection. Our pipelines utilize remotely sensed images, acquired from specific agricultural areas over a period of time. The pipelines were tested with various choices of Visual Language Models (VLMs), and they classified the images based on the presence of ephemeral gullies with accuracy higher than 70% and a F1-score close to 80% for positive gully detection. Additionally, we developed the first public dataset for ephemeral gully detection, labeled by a team of soil- and plant-science experts. To evaluate the proposed pipelines, we employed a variety of zero-shot classification methods based on State-of-the-Art (SOTA) open-source Vision-Language Models (VLMs). In addition to that, we compare the same pipelines with a transfer learning approach. Extensive experiments were conducted to validate the detection pipelines and to analyze the impact of hyperparameter changes in their performance. The experimental results demonstrate that the proposed zero-shot classification pipelines are highly effective in detecting ephemeral gullies in a scenario where classification datasets are scarce.

Paperid: 5843, https://arxiv.org/pdf/2503.01167.pdf

Abstract:
Paired image-text data with subtle variations in-between (e.g., people holding surfboards vs. people holding shovels) hold the promise of producing Vision-Language Models with proper compositional understanding. Synthesizing such training data from generative models is a highly coveted prize due to the reduced cost of data collection. However, synthesizing training images for compositional learning presents three challenges: (1) efficiency in generating large quantities of images, (2) text alignment between the generated image and the caption in the exact place of the subtle change, and (3) image fidelity in ensuring sufficient similarity with the original real images in all other places. We propose SPARCL (Synthetic Perturbations for Advancing Robust Compositional Learning), which integrates image feature injection into a fast text-to-image generative model, followed by an image style transfer step, to meet the three challenges. Further, to cope with any residual issues of text alignment, we propose an adaptive margin loss to filter out potentially incorrect synthetic samples and focus the learning on informative hard samples. Evaluation on four compositional understanding benchmarks demonstrates that SPARCL significantly improves the compositionality of CLIP, boosting the average accuracy of the CLIP base model by over 8% across all benchmarks and outperforming state-of-the-art methods by 2% on three benchmarks.

Paperid: 5844, https://arxiv.org/pdf/2503.01019.pdf

Abstract:
Despite significant progress in Vision-Language Pre-training (VLP), current approaches predominantly emphasize feature extraction and cross-modal comprehension, with limited attention to generating or transforming visual content. This gap hinders the model's ability to synthesize coherent and novel visual representations from textual prompts, thereby reducing the effectiveness of multi-modal learning. In this work, we propose MedUnifier, a unified VLP framework tailored for medical data. MedUnifier seamlessly integrates text-grounded image generation capabilities with multi-modal learning strategies, including image-text contrastive alignment, image-text matching and image-grounded text generation. Unlike traditional methods that reply on continuous visual representations, our approach employs visual vector quantization, which not only facilitates a more cohesive learning strategy for cross-modal understanding but also enhances multi-modal generation quality by effectively leveraging discrete representations. Our framework's effectiveness is evidenced by the experiments on established benchmarks, including uni-modal tasks (supervised fine-tuning), cross-modal tasks (image-text retrieval and zero-shot image classification), and multi-modal tasks (medical report generation, image synthesis), where it achieves state-of-the-art performance across various tasks. MedUnifier also offers a highly adaptable tool for a wide range of language and vision tasks in healthcare, marking advancement toward the development of a generalizable AI model for medical applications.

Paperid: 5845, https://arxiv.org/pdf/2502.20678.pdf

Abstract:
In this work we study Weakly Supervised Spatio-Temporal Video Grounding (WSTVG), a challenging task of localizing subjects spatio-temporally in videos using only textual queries and no bounding box supervision. Inspired by recent advances in vision-language foundation models, we investigate their utility for WSTVG, leveraging their zero-shot grounding capabilities. However, we find that a simple adaptation lacks essential spatio-temporal grounding abilities. To bridge this gap, we introduce Tubelet Referral Grounding (TRG), which connects textual queries to tubelets to enable spatio-temporal predictions. Despite its promise, TRG struggles with compositional action understanding and dense scene scenarios. To address these limitations, we propose STPro, a novel progressive learning framework with two key modules: (1) Sub-Action Temporal Curriculum Learning (SA-TCL), which incrementally builds compositional action understanding, and (2) Congestion-Guided Spatial Curriculum Learning (CG-SCL), which adapts the model to complex scenes by spatially increasing task difficulty. STPro achieves state-of-the-art results on three benchmark datasets, with improvements of 1.0% on VidSTG-Declarative and 3.0% on HCSTVG-v1.

Paperid: 5846, https://arxiv.org/pdf/2502.20389.pdf

Abstract:
3D vision-language grounding faces a fundamental data bottleneck: while 2D models train on billions of images, 3D models have access to only thousands of labeled scenes--a six-order-of-magnitude gap that severely limits performance. We introduce $\textbf{LIFT-GS}$, a practical distillation technique that overcomes this limitation by using differentiable rendering to bridge 3D and 2D supervision. LIFT-GS predicts 3D Gaussian representations from point clouds and uses them to render predicted language-conditioned 3D masks into 2D views, enabling supervision from 2D foundation models (SAM, CLIP, LLaMA) without requiring any 3D annotations. This render-supervised formulation enables end-to-end training of complete encoder-decoder architectures and is inherently model-agnostic. LIFT-GS achieves state-of-the-art results with $25.7\%$ mAP on open-vocabulary instance segmentation (vs. $20.2\%$ prior SOTA) and consistent $10-30\%$ improvements on referential grounding tasks. Remarkably, pretraining effectively multiplies fine-tuning datasets by 2X, demonstrating strong scaling properties that suggest 3D VLG currently operates in a severely data-scarce regime. Project page: https://liftgs.github.io

Paperid: 5847, https://arxiv.org/pdf/2502.16815.pdf

Abstract:
Vehicle re-identification (Re-ID) is a crucial task in intelligent transportation systems (ITS), aimed at retrieving and matching the same vehicle across different surveillance cameras. Numerous studies have explored methods to enhance vehicle Re-ID by focusing on semantic enhancement. However, these methods often rely on additional annotated information to enable models to extract effective semantic features, which brings many limitations. In this work, we propose a CLIP-based Semantic Enhancement Network (CLIP-SENet), an end-to-end framework designed to autonomously extract and refine vehicle semantic attributes, facilitating the generation of more robust semantic feature representations. Inspired by zero-shot solutions for downstream tasks presented by large-scale vision-language models, we leverage the powerful cross-modal descriptive capabilities of the CLIP image encoder to initially extract general semantic information. Instead of using a text encoder for semantic alignment, we design an adaptive fine-grained enhancement module (AFEM) to adaptively enhance this general semantic information at a fine-grained level to obtain robust semantic feature representations. These features are then fused with common Re-ID appearance features to further refine the distinctions between vehicles. Our comprehensive evaluation on three benchmark datasets demonstrates the effectiveness of CLIP-SENet. Our approach achieves new state-of-the-art performance, with 92.9% mAP and 98.7% Rank-1 on VeRi-776 dataset, 90.4% Rank-1 and 98.7% Rank-5 on VehicleID dataset, and 89.1% mAP and 97.9% Rank-1 on the more challenging VeRi-Wild dataset.

Paperid: 5848, https://arxiv.org/pdf/2502.16427.pdf

Abstract:
Recent advances in vision-language models have led to impressive progress in caption generation for images and short video clips. However, these models remain constrained by their limited temporal receptive fields, making it difficult to produce coherent and comprehensive captions for long videos. While several methods have been proposed to aggregate information across video segments, they often rely on supervised fine-tuning or incur significant computational overhead. To address these challenges, we introduce a novel framework for long video captioning based on graph consolidation. Our approach first generates segment-level captions, corresponding to individual frames or short video intervals, using off-the-shelf visual captioning models. These captions are then parsed into individual scene graphs, which are subsequently consolidated into a unified graph representation that preserves both holistic context and fine-grained details throughout the video. A lightweight graph-to-text decoder then produces the final video-level caption. This framework effectively extends the temporal understanding capabilities of existing models without requiring any additional fine-tuning on long video datasets. Experimental results show that our method significantly outperforms existing LLM-based consolidation approaches, achieving strong zero-shot performance while substantially reducing computational costs.

Paperid: 5849, https://arxiv.org/pdf/2502.15678.pdf

Abstract:
Pre-trained vision language models still fall short of human visual cognition. In an effort to improve visual cognition and align models with human behavior, we introduce visual stimuli and human judgments on visual cognition tasks, allowing us to systematically evaluate performance across cognitive domains under a consistent environment. We fine-tune models on ground truth data for intuitive physics and causal reasoning and find that this improves model performance in the respective fine-tuning domain. Furthermore, it can improve model alignment with human behavior. However, we find that task-specific fine-tuning does not contribute to robust human-like generalization to data with other visual characteristics or to tasks in other cognitive domains.

Paperid: 5850, https://arxiv.org/pdf/2502.12084.pdf

Abstract:
Visually linking matching cues is a crucial ability in daily life, such as identifying the same person in multiple photos based on their cues, even without knowing who they are. Despite the extensive knowledge that vision-language models (VLMs) possess, it remains largely unexplored whether they are capable of performing this fundamental task. To address this, we introduce \textbf{VLM2-Bench}, a benchmark designed to assess whether VLMs can Visually Link Matching cues, with 9 subtasks and over 3,000 test cases. Comprehensive evaluation across twelve VLMs, along with further analysis of various language-side and vision-side prompting methods, leads to a total of eight key findings. We identify critical challenges in models' ability to link visual cues, highlighting a significant performance gap. Based on these insights, we advocate for (i) enhancing core visual capabilities to improve adaptability and reduce reliance on prior knowledge, (ii) establishing clearer principles for integrating language-based reasoning in vision-centric tasks to prevent unnecessary biases, and (iii) shifting vision-text training paradigms toward fostering models' ability to independently structure and infer relationships among visual cues.

Paperid: 5851, https://arxiv.org/pdf/2502.04098.pdf

Abstract:
Vision-language models (VLMs) excel in tasks such as visual question answering and image captioning. However, VLMs are often limited by their use of pretrained image encoders, like CLIP, leading to image understanding errors that hinder overall performance. On top of that, real-world applications often require the model to be continuously adapted as new and often limited data continuously arrive. To address this, we propose LoRSU (Low-Rank Adaptation with Structured Updates), a robust and computationally efficient method for selectively updating image encoders within VLMs. LoRSU introduces structured and localized parameter updates, effectively correcting performance on previously error-prone data while preserving the model's general robustness. Our approach leverages theoretical insights to identify and update only the most critical parameters, achieving significant resource efficiency. Specifically, we demonstrate that LoRSU reduces computational overhead by over 25x compared to full VLM updates, without sacrificing performance. Experimental results on VQA tasks in the few-shot continual learning setting, validate LoRSU's scalability, efficiency, and effectiveness, making it a compelling solution for image encoder adaptation in resource-constrained environments.

Paperid: 5852, https://arxiv.org/pdf/2502.02452.pdf

Abstract:
Personalization of Large Vision-Language Models (LVLMs) involves customizing models to recognize specific users and object instances, and to generate contextually tailored responses. Existing approaches typically rely on time-consuming test-time training for each user or object, making them impractical for real-world deployment, a limitation reflected in current personalization benchmarks, which are focused on object-centric, single-concept evaluations. In this paper, we present a novel training-free approach to LVLM personalization and introduce a comprehensive real-world benchmark designed to rigorously evaluate various aspects of the personalization task. Our method leverages pre-trained vision foundation models to extract distinctive features, applies retrieval-augmented generation (RAG) techniques to identify instances within visual inputs, and employs visual prompting strategies to guide model outputs. Our model-agnostic vision toolkit enables efficient and flexible multi-concept personalization across both images and videos, without any additional training. We achieve state-of-the-art results, surpassing existing training-based methods.

Paperid: 5853, https://arxiv.org/pdf/2501.13620.pdf

Abstract:
A fundamental challenge in artificial intelligence involves understanding the cognitive mechanisms underlying visual reasoning in sophisticated models like Vision-Language Models (VLMs). How do these models integrate visual perception with abstract thought, especially when reasoning across multiple images or requiring fine-grained compositional understanding? Drawing inspiration from cognitive science, this paper introduces a structured evaluation framework using diverse visual reasoning tasks-Bongard Problems (BPs) and Winoground-to dissect the perception-reasoning interface in VLMs. We propose three distinct evaluation paradigms, mirroring human problem-solving strategies: Direct Visual Rule Learning (DVRL; holistic processing), Deductive Rule Learning (DRL; rule extraction and application), and Componential Analysis (CA; analytical decomposition via task-agnostic textual descriptions). These paradigms systematically vary cognitive load and probe processing stages. Notably, CA enables multi-image reasoning evaluation even for single-image architectures and isolates reasoning from perception by operating on textual descriptions. Applying this framework, we demonstrate that CA, leveraging powerful language models for reasoning over rich, independently generated descriptions, achieves new state-of-the-art (SOTA) performance on challenging benchmarks including Bongard-OpenWorld, Bongard-HOI, and Winoground. Ablation studies confirm reasoning improves significantly when perceptual challenges are mitigated, revealing a critical perception bottleneck. Our framework provides a valuable diagnostic tool and suggests that decoupling perception (via rich, task-agnostic description) from reasoning is a promising direction for robust and general visual intelligence.

Paperid: 5854, https://arxiv.org/pdf/2501.09967.pdf

Abstract:
Artificial Intelligence (AI) has continued to achieve tremendous success in recent times. However, the decision logic of these frameworks is often not transparent, making it difficult for stakeholders to understand, interpret or explain their behavior. This limitation hinders trust in machine learning systems and causes a general reluctance towards their adoption in practical applications, particularly in mission-critical domains like healthcare and autonomous driving. Explainable AI (XAI) techniques facilitate the explainability or interpretability of machine learning models, enabling users to discern the basis of the decision and possibly avert undesirable behavior. This comprehensive survey details the advancements of explainable AI methods, from inherently interpretable models to modern approaches for achieving interpretability of various black box models, including large language models (LLMs). Additionally, we review explainable AI techniques that leverage LLM and vision-language model (VLM) frameworks to automate or improve the explainability of other machine learning models. The use of LLM and VLM as interpretability methods particularly enables high-level, semantically meaningful explanations of model decisions and behavior. Throughout the paper, we highlight the scientific principles, strengths and weaknesses of state-of-the-art methods and outline different areas of improvement. Where appropriate, we also present qualitative and quantitative comparison results of various methods to show how they compare. Finally, we discuss the key challenges of XAI and directions for future research.

Paperid: 5855, https://arxiv.org/pdf/2501.07244.pdf

Abstract:
Recent advancements in Vision-Language Models (VLMs) have opened new possibilities in automatic grading of handwritten student responses, particularly in mathematics. However, a comprehensive study to test the ability of VLMs to evaluate and reason over handwritten content remains absent. To address this gap, we introduce FERMAT, a benchmark designed to assess the ability of VLMs to detect, localize and correct errors in handwritten mathematical content. FERMAT spans four key error dimensions - computational, conceptual, notational, and presentation - and comprises over 2,200 handwritten math solutions derived from 609 manually curated problems from grades 7-12 with intentionally introduced perturbations. Using FERMAT we benchmark nine VLMs across three tasks: error detection, localization, and correction. Our results reveal significant shortcomings in current VLMs in reasoning over handwritten text, with Gemini-1.5-Pro achieving the highest error correction rate (77%). We also observed that some models struggle with processing handwritten content, as their accuracy improves when handwritten inputs are replaced with printed text or images. These findings highlight the limitations of current VLMs and reveal new avenues for improvement. We release FERMAT and all the associated resources in the open-source to drive further research.

Paperid: 5856, https://arxiv.org/pdf/2501.03151.pdf

Abstract:
Generative artificial intelligence (AI) systems based on large-scale pretrained foundation models (PFMs) such as vision-language models, large language models (LLMs), diffusion models and vision-language-action (VLA) models have demonstrated the ability to solve complex and truly non-trivial AI problems in a wide variety of domains and contexts. Multimodal large language models (MLLMs), in particular, learn from vast and diverse data sources, allowing rich and nuanced representations of the world and, thereby, providing extensive capabilities, including the ability to reason, engage in meaningful dialog; collaborate with humans and other agents to jointly solve complex problems; and understand social and emotional aspects of humans. Despite this impressive feat, the cognitive abilities of state-of-the-art LLMs trained on large-scale datasets are still superficial and brittle. Consequently, generic LLMs are severely limited in their generalist capabilities. A number of foundational problems -- embodiment, symbol grounding, causality and memory -- are required to be addressed for LLMs to attain human-level general intelligence. These concepts are more aligned with human cognition and provide LLMs with inherent human-like cognitive properties that support the realization of physically-plausible, semantically meaningful, flexible and more generalizable knowledge and intelligence. In this work, we discuss the aforementioned foundational issues and survey state-of-the art approaches for implementing these concepts in LLMs. Specifically, we discuss how the principles of embodiment, symbol grounding, causality and memory can be leveraged toward the attainment of artificial general intelligence (AGI) in an organic manner.

Paperid: 5857, https://arxiv.org/pdf/2501.02461.pdf

Abstract:
Remote sensing data is often distributed across multiple institutions, and due to privacy concerns and data-sharing restrictions, leveraging large-scale datasets in a centralized training framework is challenging. Federated learning offers a promising solution by enabling collaborative model training across distributed data sources without requiring data centralization. However, current Vision-Language Models (VLMs), which typically contain billions of parameters, pose significant communication challenges for traditional federated learning approaches based on model parameter updates, as they would incur substantial communication costs. In this paper, we propose FedRSCLIP, the first federated learning framework designed for remote sensing image classification based on a VLM, specifically CLIP. FedRSCLIP addresses the challenges of data heterogeneity and large-scale model transmission in federated environments by introducing Prompt Learning, which optimizes only a small set of tunable parameters. The framework introduces a dual-prompt mechanism, comprising Shared Prompts for global knowledge sharing and Private Prompts for client-specific adaptation. To maintain semantic coherence between shared and private prompts, we propose the Dual Prompt Alignment Constraint to balance global consistency and local adaptability across diverse client distributions. Additionally, to enhance cross-modal representation learning, we introduce the Cross-Modal Feature Alignment Constraint to align multimodal features between text and image prompts. To validate the effectiveness of our proposed model, we construct a Fed-RSIC dataset based on three existing remote sensing image classification datasets, specifically designed to simulate various federated learning configurations. Experimental results demonstrate the effectiveness and superiority of FedRSCLIP in remote sensing image classification.

Paperid: 5858, https://arxiv.org/pdf/2412.13333.pdf

Abstract:
Vision-Language Models (VLMs), such as CLIP, have already seen widespread applications. Researchers actively engage in further fine-tuning VLMs in safety-critical domains. In these domains, prediction rationality is crucial: the prediction should be correct and based on valid evidence. Yet, for VLMs, the impact of fine-tuning on prediction rationality is seldomly investigated. To study this problem, we proposed two new metrics called Prediction Trustworthiness and Inference Reliability. We conducted extensive experiments on various settings and observed some interesting phenomena. On the one hand, we found that the well-adopted fine-tuning methods led to more correct predictions based on invalid evidence. This potentially undermines the trustworthiness of correct predictions from fine-tuned VLMs. On the other hand, having identified valid evidence of target objects, fine-tuned VLMs were more likely to make correct predictions. Moreover, the findings are also consistent under distributional shifts and across various experimental settings. We hope our research offer fresh insights to VLM fine-tuning.

Paperid: 5859, https://arxiv.org/pdf/2412.10348.pdf

Abstract:
In current multimodal tasks, models typically freeze the encoder and decoder while adapting intermediate layers to task-specific goals, such as region captioning. Region-level visual understanding presents significant challenges for large-scale vision-language models. While limited spatial awareness is a known issue, coarse-grained pretraining, in particular, exacerbates the difficulty of optimizing latent representations for effective encoder-decoder alignment. We propose AlignCap, a framework designed to enhance region-level understanding through fine-grained alignment of latent spaces. Our approach introduces a novel latent feature refinement module that enhances conditioned latent space representations to improve region-level captioning performance. We also propose an innovative alignment strategy, the semantic space alignment module, which boosts the quality of multimodal representations. Additionally, we incorporate contrastive learning in a novel manner within both modules to further enhance region-level captioning performance. To address spatial limitations, we employ a General Object Detection (GOD) method as a data preprocessing pipeline that enhances spatial reasoning at the regional level. Extensive experiments demonstrate that our approach significantly improves region-level captioning performance across various tasks

Paperid: 5860, https://arxiv.org/pdf/2412.10151.pdf

Abstract:
We propose the VLR-Bench, a visual question answering (VQA) benchmark for evaluating vision language models (VLMs) based on retrieval augmented generation (RAG). Unlike existing evaluation datasets for external knowledge-based VQA, the proposed VLR-Bench includes five input passages. This allows testing of the ability to determine which passage is useful for answering a given query, a capability lacking in previous research. In this context, we constructed a dataset of 32,000 automatically generated instruction-following examples, which we denote as VLR-IF. This dataset is specifically designed to enhance the RAG capabilities of VLMs by enabling them to learn how to generate appropriate answers based on input passages. We evaluated the validity of the proposed benchmark and training data and verified its performance using the state-of-the-art Llama3-based VLM, the Llava-Llama-3 model. The proposed VLR-Bench and VLR-IF datasets are publicly available online.

Paperid: 5861, https://arxiv.org/pdf/2412.08746.pdf

Abstract:
Vision-Language Models (VLMs) excel in diverse visual tasks but face challenges in document understanding, which requires fine-grained text processing. While typical visual tasks perform well with low-resolution inputs, reading-intensive applications demand high-resolution, resulting in significant computational overhead. Using OCR-extracted text in VLM prompts partially addresses this issue but underperforms compared to full-resolution counterpart, as it lacks the complete visual context needed for optimal performance. We introduce DocVLM, a method that integrates an OCR-based modality into VLMs to enhance document processing while preserving original weights. Our approach employs an OCR encoder to capture textual content and layout, compressing these into a compact set of learned queries incorporated into the VLM. Comprehensive evaluations across leading VLMs show that DocVLM significantly reduces reliance on high-resolution images for document understanding. In limited-token regimes (448$\times$448), DocVLM with 64 learned queries improves DocVQA results from 56.0% to 86.6% when integrated with InternVL2 and from 84.4% to 91.2% with Qwen2-VL. In LLaVA-OneVision, DocVLM achieves improved results while using 80% less image tokens. The reduced token usage allows processing multiple pages effectively, showing impressive zero-shot results on DUDE and state-of-the-art performance on MP-DocVQA, highlighting DocVLM's potential for applications requiring high-performance and efficiency.

Paperid: 5862, https://arxiv.org/pdf/2412.08108.pdf

Abstract:
Large Vision-Language Models (VLMs) have demonstrated remarkable performance across multimodal tasks by integrating vision encoders with large language models (LLMs). However, these models remain vulnerable to adversarial attacks. Among such attacks, Universal Adversarial Perturbations (UAPs) are especially powerful, as a single optimized perturbation can mislead the model across various input images. In this work, we introduce a novel UAP specifically designed for VLMs: the Doubly-Universal Adversarial Perturbation (Doubly-UAP), capable of universally deceiving VLMs across both image and text inputs. To successfully disrupt the vision encoder's fundamental process, we analyze the core components of the attention mechanism. After identifying value vectors in the middle-to-late layers as the most vulnerable, we optimize Doubly-UAP in a label-free manner with a frozen model. Despite being developed as a black-box to the LLM, Doubly-UAP achieves high attack success rates on VLMs, consistently outperforming baseline methods across vision-language tasks. Extensive ablation studies and analyses further demonstrate the robustness of Doubly-UAP and provide insights into how it influences internal attention mechanisms.

Paperid: 5863, https://arxiv.org/pdf/2412.07148.pdf

Abstract:
This paper introduces Multiple Choice Reasoning via. Process of Elimination using Multi-Modal models, herein referred to as Multi-Modal Process of Elimination (MM-PoE). This novel methodology is engineered to augment the efficacy of Vision-Language Models (VLMs) in multiple-choice visual reasoning tasks. Diverging from conventional approaches that evaluate each option independently, MM-PoE employs a dual-step scoring paradigm that initially identifies and excludes implausible choices, subsequently concentrating on the most probable remaining options. This method emulates human test-taking strategies, where individuals typically eliminate clearly incorrect answers prior to selecting the optimal response. Our empirical evaluations, conducted across three benchmark datasets, reveal that MM-PoE significantly improves both zero-shot and few-shot performance of contemporary state-of-the-art VLMs. Critically, this approach not only broadens the application of the elimination process to multi-modal contexts but also allows few-shot experiments, thereby addressing two principal limitations concerning usage of PoE only in zero-shot settings and only with a language-only framework. As a result, MM-PoE not only refines the reasoning capabilities of VLMs but also broadens their applicability to complex visual question-answering scenarios. All code and documentation supporting our work are available at https://pypi.org/project/mm-poe/, enabling researchers and practitioners to easily integrate and further develop these techniques.

Paperid: 5864, https://arxiv.org/pdf/2412.05840.pdf

Abstract:
Continual learning aims to update a model so that it can sequentially learn new tasks without forgetting previously acquired knowledge. Recent continual learning approaches often leverage the vision-language model CLIP for its high-dimensional feature space and cross-modality feature matching. Traditional CLIP-based classification methods identify the most similar text label for a test image by comparing their embeddings. However, these methods are sensitive to the quality of text phrases and less effective for classes lacking meaningful text labels. In this work, we rethink CLIP-based continual learning and introduce the concept of Label Vector Pool (LVP). LVP replaces text labels with training images as similarity references, eliminating the need for ideal text descriptions. We present three variations of LVP and evaluate their performance on class and domain incremental learning tasks. Leveraging CLIP's high dimensional feature space, LVP learning algorithms are task-order invariant. The new knowledge does not modify the old knowledge, hence, there is minimum forgetting. Different tasks can be learned independently and in parallel with low computational and memory demands. Experimental results show that proposed LVP-based methods outperform the current state-of-the-art baseline by a significant margin of 40.7%.

Paperid: 5865, https://arxiv.org/pdf/2412.04729.pdf

Abstract:
Recent advances in vision-language models (VLMs) have shown great promise in connecting images and text, but extending these models to long videos remains challenging due to the rapid growth in token counts. Models that compress videos by local aggregation in time or space have become popular for handling long-form inputs; however, these pooling-based projectors sacrifice the benefits of fixed-length representations that are crucial for streaming and efficient video understanding. We introduce $\texttt{Espresso}$, a new architecture that separately compresses spatial and temporal features into fixed-length sequences. $\texttt{Espresso}$ enables efficient video encoding while maintaining strong long-form reasoning capabilities. Experiments show that fixed-length compression combined with segment-wise processing offers a scalable and competitive alternative to pooling-based approaches. Our results demonstrate that fixed-length projectors, when properly designed and trained, remain a viable foundation for video-language modeling.

Paperid: 5866, https://arxiv.org/pdf/2412.00577.pdf

Abstract:
As we consider entrusting Large Language Models (LLMs) with key societal and decision-making roles, measuring their alignment with human cognition becomes critical. This requires methods that can assess how these systems represent information and facilitate comparisons to human understanding across diverse tasks. To meet this need, we developed Turing Representational Similarity Analysis (RSA), a method that uses pairwise similarity ratings to quantify alignment between AIs and humans. We tested this approach on semantic alignment across text and image modalities, measuring how different Large Language and Vision Language Model (LLM and VLM) similarity judgments aligned with human responses at both group and individual levels. GPT-4o showed the strongest alignment with human performance among the models we tested, particularly when leveraging its text processing capabilities rather than image processing, regardless of the input modality. However, no model we studied adequately captured the inter-individual variability observed among human participants. This method helped uncover certain hyperparameters and prompts that could steer model behavior to have more or less human-like qualities at an inter-individual or group level. Turing RSA enables the efficient and flexible quantification of human-AI alignment and complements existing accuracy-based benchmark tasks. We demonstrate its utility across multiple modalities (words, sentences, images) for understanding how LLMs encode knowledge and for examining representational alignment with human cognition.

Paperid: 5867, https://arxiv.org/pdf/2411.08870.pdf

Abstract:
Several recent works seek to adapt general-purpose large language models (LLMs) and vision-language models (VLMs) for medical applications through continued pretraining on publicly available biomedical corpora. These works typically claim that such domain-adaptive pretraining improves performance on various downstream medical tasks, such as answering medical exam questions. In this paper, we compare ten "medical" LLMs and two VLMs against their corresponding base models, arriving at a different conclusion: all medical VLMs and nearly all medical LLMs fail to consistently improve over their base models in the zero-/few-shot prompting and supervised fine-tuning regimes for medical question answering (QA). For instance, on clinical-note-based QA tasks in the 3-shot setting, medical LLMs outperform their base models in only 26.7% of cases, reach a (statistical) tie in 16.7% of cases, and perform significantly worse in the remaining 56.7% of cases. Our conclusions are based on (i) comparing each medical model directly against its base model; (ii) optimizing the prompts for each model separately in zero-/few-shot prompting; and (iii) accounting for statistical uncertainty in comparisons. Our findings suggest that state-of-the-art general-domain models may already exhibit strong medical knowledge and reasoning capabilities, and offer recommendations to strengthen the conclusions of future studies.

Paperid: 5868, https://arxiv.org/pdf/2411.04642.pdf

Abstract:
Vision-Language (VL) models have garnered considerable research interest; however, they still face challenges in effectively handling text within images. To address this limitation, researchers have developed two approaches. The first method involves utilizing external Optical Character Recognition (OCR) tools to extract textual information from images, which is then prepended to other textual inputs. The second strategy focuses on employing extremely high-resolution images to improve text recognition capabilities. In this paper, we focus on enhancing the first strategy by introducing a novel method, named TAP-VL, which treats OCR information as a distinct modality and seamlessly integrates it into any VL model. TAP-VL employs a lightweight transformer-based OCR module to receive OCR with layout information, compressing it into a short fixed-length sequence for input into the LLM. Initially, we conduct model-agnostic pretraining of the OCR module on unlabeled documents, followed by its integration into any VL architecture through brief fine-tuning. Extensive experiments demonstrate consistent performance improvements when applying TAP-VL to top-performing VL models, across scene-text and document-based VL benchmarks.

Paperid: 5869, https://arxiv.org/pdf/2411.04118.pdf

Abstract:
Several recent works seek to develop foundation models specifically for medical applications, adapting general-purpose large language models (LLMs) and vision-language models (VLMs) via continued pretraining on publicly available biomedical corpora. These works typically claim that such domain-adaptive pretraining (DAPT) improves performance on downstream medical tasks, such as answering medical licensing exam questions. In this paper, we compare seven public "medical" LLMs and two VLMs against their corresponding base models, arriving at a different conclusion: all medical VLMs and nearly all medical LLMs fail to consistently improve over their base models in the zero-/few-shot prompting regime for medical question-answering (QA) tasks. For instance, across the tasks and model pairs we consider in the 3-shot setting, medical LLMs only outperform their base models in 12.1% of cases, reach a (statistical) tie in 49.8% of cases, and are significantly worse than their base models in the remaining 38.2% of cases. Our conclusions are based on (i) comparing each medical model head-to-head, directly against the corresponding base model; (ii) optimizing the prompts for each model separately; and (iii) accounting for statistical uncertainty in comparisons. While these basic practices are not consistently adopted in the literature, our ablations show that they substantially impact conclusions. Our findings suggest that state-of-the-art general-domain models may already exhibit strong medical knowledge and reasoning capabilities, and offer recommendations to strengthen the conclusions of future studies.

Paperid: 5870, https://arxiv.org/pdf/2411.04006.pdf

Abstract:
This study explores the potential of off-the-shelf Vision-Language Models (VLMs) for high-level robot planning in the context of autonomous navigation. Indeed, while most of existing learning-based approaches for path planning require extensive task-specific training/fine-tuning, we demonstrate how such training can be avoided for most practical cases. To do this, we introduce Select2Plan (S2P), a novel training-free framework for high-level robot planning which completely eliminates the need for fine-tuning or specialised training. By leveraging structured Visual Question-Answering (VQA) and In-Context Learning (ICL), our approach drastically reduces the need for data collection, requiring a fraction of the task-specific data typically used by trained models, or even relying only on online data. Our method facilitates the effective use of a generally trained VLM in a flexible and cost-efficient way, and does not require additional sensing except for a simple monocular camera. We demonstrate its adaptability across various scene types, context sources, and sensing setups. We evaluate our approach in two distinct scenarios: traditional First-Person View (FPV) and infrastructure-driven Third-Person View (TPV) navigation, demonstrating the flexibility and simplicity of our method. Our technique significantly enhances the navigational capabilities of a baseline VLM of approximately 50% in TPV scenario, and is comparable to trained models in the FPV one, with as few as 20 demonstrations.

Paperid: 5871, https://arxiv.org/pdf/2411.03034.pdf

Abstract:
Human-scene vision-language tasks are increasingly prevalent in diverse social applications, yet recent advancements predominantly rely on models specifically tailored to individual tasks. Emerging research indicates that large vision-language models (VLMs) can enhance performance across various downstream vision-language understanding tasks. However, general-domain models often underperform in specialized fields. This study introduces a domain-specific Large Vision-Language Model, Human-Scene Vision-Language Model (HumanVLM), designed to provide a foundation for human-scene Vision-Language tasks. Specifically, (1) we create a large-scale human-scene multimodal image-text dataset (HumanCaption-10M) sourced from the Internet to facilitate domain-specific alignment; (2) develop a captioning approach for human-centered images, capturing human faces, bodies, and backgrounds, and construct a high-quality Human-Scene image-text dataset (HumanCaptionHQ, about 311k pairs) that contain as much detailed information as possible about human; (3) Using HumanCaption-10M and HumanCaptionHQ, we train a HumanVLM. In the experiments, we then evaluate our HumanVLM across varous downstream tasks, where it demonstrates superior overall performance among multimodal models of comparable scale, particularly excelling in human-related tasks and significantly outperforming similar models, including Qwen2VL and ChatGPT-4o. HumanVLM, alongside the data introduced, will stimulate the research in human-around fields.

Paperid: 5872, https://arxiv.org/pdf/2411.01595.pdf

Abstract:
Remote Sensing Image Captioning (RSIC) presents unique challenges and plays a critical role in applications. Traditional RSIC methods often struggle to produce rich and diverse descriptions. Recently, with advancements in VLMs, efforts have emerged to integrate these models into the remote sensing domain and to introduce descriptive datasets specifically designed to enhance VLM training. This paper proposes RS-MoE, a first Mixture of Expert based VLM specifically customized for remote sensing domain. Unlike traditional MoE models, the core of RS-MoE is the MoE Block, which incorporates a novel Instruction Router and multiple lightweight Large Language Models (LLMs) as expert models. The Instruction Router is designed to generate specific prompts tailored for each corresponding LLM, guiding them to focus on distinct aspects of the RSIC task. This design not only allows each expert LLM to concentrate on a specific subset of the task, thereby enhancing the specificity and accuracy of the generated captions, but also improves the scalability of the model by facilitating parallel processing of sub-tasks. Additionally, we present a two-stage training strategy for tuning our RS-MoE model to prevent performance degradation due to sparsity. We fine-tuned our model on the RSICap dataset using our proposed training strategy. Experimental results on the RSICap dataset, along with evaluations on other traditional datasets where no additional fine-tuning was applied, demonstrate that our model achieves state-of-the-art performance in generating precise and contextually relevant captions. Notably, our RS-MoE-1B variant achieves performance comparable to 13B VLMs, demonstrating the efficiency of our model design. Moreover, our model demonstrates promising generalization capabilities by consistently achieving state-of-the-art performance on the Remote Sensing Visual Question Answering (RSVQA) task.

Paperid: 5873, https://arxiv.org/pdf/2411.01568.pdf

Abstract:
In this paper, we propose an approach that combines Vision Language Models (VLMs) and Behavior Trees (BTs) to address failures in robotics. Current robotic systems can handle known failures with pre-existing recovery strategies, but they are often ill-equipped to manage unknown failures or anomalies. We introduce VLMs as a monitoring tool to detect and identify failures during task execution. Additionally, VLMs generate missing conditions or skill templates that are then incorporated into the BT, ensuring the system can autonomously address similar failures in future tasks. We validate our approach through simulations in several failure scenarios.

Paperid: 5874, https://arxiv.org/pdf/2411.00299.pdf

Abstract:
Generating accurate radiology reports from medical images is a clinically important but challenging task. While current Vision Language Models (VLMs) show promise, they are prone to generating hallucinations, potentially compromising patient care. We introduce RadFlag, a black-box method to enhance the accuracy of radiology report generation. Our method uses a sampling-based flagging technique to find hallucinatory generations that should be removed. We first sample multiple reports at varying temperatures and then use a Large Language Model (LLM) to identify claims that are not consistently supported across samples, indicating that the model has low confidence in those claims. Using a calibrated threshold, we flag a fraction of these claims as likely hallucinations, which should undergo extra review or be automatically rejected. Our method achieves high precision when identifying both individual hallucinatory sentences and reports that contain hallucinations. As an easy-to-use, black-box system that only requires access to a model's temperature parameter, RadFlag is compatible with a wide range of radiology report generation models and has the potential to broadly improve the quality of automated radiology reporting.

Paperid: 5875, https://arxiv.org/pdf/2410.23317.pdf

Abstract:
Vision-Language Models (VLMs) have demonstrated impressive performance across a versatile set of tasks. A key challenge in accelerating VLMs is storing and accessing the large Key-Value (KV) cache that encodes long visual contexts, such as images or videos. While existing KV cache compression methods are effective for Large Language Models (LLMs), directly migrating them to VLMs yields suboptimal accuracy and speedup. To bridge the gap, we propose VL-Cache, a novel KV cache compression recipe tailored for accelerating VLM inference. In this paper, we first investigate the unique sparsity pattern of VLM attention by distinguishing visual and text tokens in prefill and decoding phases. Based on these observations, we introduce a layer-adaptive sparsity-aware cache budget allocation method that effectively distributes the limited cache budget across different layers, further reducing KV cache size without compromising accuracy. Additionally, we develop a modality-aware token scoring policy to better evaluate the token importance. Empirical results on multiple benchmark datasets demonstrate that retaining only 10% of KV cache achieves accuracy comparable to that with full cache. In a speed benchmark, our method accelerates end-to-end latency of generating 100 tokens by up to 2.33x and speeds up decoding by up to 7.08x, while reducing the memory footprint of KV cache in GPU by 90%.

Paperid: 5876, https://arxiv.org/pdf/2410.22315.pdf

Abstract:
Compositional reasoning in Vision-Language Models (VLMs) remains challenging as these models often struggle to relate objects, attributes, and spatial relationships. Recent methods aim to address these limitations by relying on the semantics of the textual description, using Large Language Models (LLMs) to break them down into subsets of questions and answers. However, these methods primarily operate on the surface level, failing to incorporate deeper lexical understanding while introducing incorrect assumptions generated by the LLM. In response to these issues, we present Caption Expansion with Contradictions and Entailments (CECE), a principled approach that leverages Natural Language Inference (NLI) to generate entailments and contradictions from a given premise. CECE produces lexically diverse sentences while maintaining their core meaning. Through extensive experiments, we show that CECE enhances interpretability and reduces overreliance on biased or superficial features. By balancing CECE along the original premise, we achieve significant improvements over previous methods without requiring additional fine-tuning, producing state-of-the-art results on benchmarks that score agreement with human judgments for image-text alignment, and achieving an increase in performance on Winoground of +19.2% (group score) and +12.9% on EqBen (group score) over the best prior work (finetuned with targeted data).

Paperid: 5877, https://arxiv.org/pdf/2410.09704.pdf

Abstract:
Echocardiography is the most widely used cardiac imaging modality, capturing ultrasound video data to assess cardiac structure and function. Artificial intelligence (AI) in echocardiography has the potential to streamline manual tasks and improve reproducibility and precision. However, most echocardiography AI models are single-view, single-task systems that do not synthesize complementary information from multiple views captured during a full exam, and thus lead to limited performance and scope of applications. To address this problem, we introduce EchoPrime, a multi-view, view-informed, video-based vision-language foundation model trained on over 12 million video-report pairs. EchoPrime uses contrastive learning to train a unified embedding model for all standard views in a comprehensive echocardiogram study with representation of both rare and common diseases and diagnoses. EchoPrime then utilizes view-classification and a view-informed anatomic attention model to weight video-specific interpretations that accurately maps the relationship between echocardiographic views and anatomical structures. With retrieval-augmented interpretation, EchoPrime integrates information from all echocardiogram videos in a comprehensive study and performs holistic comprehensive clinical echocardiography interpretation. In datasets from two independent healthcare systems, EchoPrime achieves state-of-the art performance on 23 diverse benchmarks of cardiac form and function, surpassing the performance of both task-specific approaches and prior foundation models. Following rigorous clinical evaluation, EchoPrime can assist physicians in the automated preliminary assessment of comprehensive echocardiography.

Paperid: 5878, https://arxiv.org/pdf/2410.09382.pdf

Abstract:
Person re-identification (ReID) has recently benefited from large pretrained vision-language models such as Contrastive Language-Image Pre-Training (CLIP). However, the absence of concrete descriptions necessitates the use of implicit text embeddings, which demand complicated and inefficient training strategies. To address this issue, we first propose one straightforward solution by leveraging existing image captioning models to generate pseudo captions for person images, and thereby boost person re-identification with large vision language models. Using models like the Large Language and Vision Assistant (LLAVA), we generate high-quality captions based on fixed templates that capture key semantic attributes such as gender, clothing, and age. By augmenting ReID training sets from uni-modality (image) to bi-modality (image and text), we introduce CLIP-SCGI, a simple yet effective framework that leverages synthesized captions to guide the learning of discriminative and robust representations. Built on CLIP, CLIP-SCGI fuses image and text embeddings through two modules to enhance the training process. To address quality issues in generated captions, we introduce a caption-guided inversion module that captures semantic attributes from images by converting relevant visual information into pseudo-word tokens based on the descriptions. This approach helps the model better capture key information and focus on relevant regions. The extracted features are then utilized in a cross-modal fusion module, guiding the model to focus on regions semantically consistent with the caption, thereby facilitating the optimization of the visual encoder to extract discriminative and robust representations. Extensive experiments on four popular ReID benchmarks demonstrate that CLIP-SCGI outperforms the state-of-the-art by a significant margin.

Paperid: 5879, https://arxiv.org/pdf/2410.09286.pdf

Abstract:
Learning from Demonstrations, particularly from biological experts like humans and animals, often encounters significant data acquisition challenges. While recent approaches leverage internet videos for learning, they require complex, task-specific pipelines to extract and retarget motion data for the agent. In this work, we introduce a language-model-assisted bi-level programming framework that enables a reinforcement learning agent to directly learn its reward from internet videos, bypassing dedicated data preparation. The framework includes two levels: an upper level where a vision-language model (VLM) provides feedback by comparing the learner's behavior with expert videos, and a lower level where a large language model (LLM) translates this feedback into reward updates. The VLM and LLM collaborate within this bi-level framework, using a "chain rule" approach to derive a valid search direction for reward learning. We validate the method for reward learning from YouTube videos, and the results have shown that the proposed method enables efficient reward design from expert videos of biological agents for complex behavior synthesis.

Paperid: 5880, https://arxiv.org/pdf/2410.04884.pdf

Abstract:
Visual language pre-training (VLP) models have demonstrated significant success across various domains, yet they remain vulnerable to adversarial attacks. Addressing these adversarial vulnerabilities is crucial for enhancing security in multimodal learning. Traditionally, adversarial methods targeting VLP models involve simultaneously perturbing images and text. However, this approach faces notable challenges: first, adversarial perturbations often fail to translate effectively into real-world scenarios; second, direct modifications to the text are conspicuously visible. To overcome these limitations, we propose a novel strategy that exclusively employs image patches for attacks, thus preserving the integrity of the original text. Our method leverages prior knowledge from diffusion models to enhance the authenticity and naturalness of the perturbations. Moreover, to optimize patch placement and improve the efficacy of our attacks, we utilize the cross-attention mechanism, which encapsulates intermodal interactions by generating attention maps to guide strategic patch placements. Comprehensive experiments conducted in a white-box setting for image-to-text scenarios reveal that our proposed method significantly outperforms existing techniques, achieving a 100% attack success rate. Additionally, it demonstrates commendable performance in transfer tasks involving text-to-image configurations.

Paperid: 5881, https://arxiv.org/pdf/2410.04778.pdf

Abstract:
With the advent of LLMs and variants, a flurry of research has emerged, analyzing the performance of such models across an array of tasks. While most studies focus on evaluating the capabilities of state-of-the-art (SoTA) Vision Language Models (VLMs) through task accuracy (e.g., visual question answering, grounding), our work explores the related but complementary aspect of consistency - the ability of a VLM to produce semantically similar or identical responses to semantically similar queries. We note that consistency is a fundamental prerequisite (necessary but not sufficient condition) for robustness and trust in VLMs. Armed with this perspective, we propose the MM-R3 benchmark, which allows us to analyze performance, in terms of consistency and accuracy, of SoTA VLMs on three tasks: Question Rephrasing, Image Restyling, and Context Reasoning. Our analysis reveals that consistency does not always align with accuracy, indicating that models with higher accuracy are not necessarily more consistent, and vice versa. Furthermore, we propose a simple yet effective mitigation strategy in the form of an adapter module trained to minimize inconsistency across prompts. With our proposed strategy, we are able to achieve absolute improvements of 5.7% and 12.5%, on average on widely used VLMs such as BLIP-2 and LLaVa 1.5M in terms of consistency over their existing counterparts.

Paperid: 5882, https://arxiv.org/pdf/2410.04302.pdf

Abstract:
Navigating robots discreetly in human work environments while considering the possible privacy implications of robotic tasks presents significant challenges. Such scenarios are increasingly common, for instance, when robots transport sensitive objects that demand high levels of privacy in spaces crowded with human activities. While extensive research has been conducted on robotic path planning and social awareness, current robotic systems still lack the functionality of privacy-aware navigation in public environments. To address this, we propose a new framework for mobile robot navigation that leverages vision-language models to incorporate privacy awareness into adaptive path planning. Specifically, all potential paths from the starting point to the destination are generated using the A* algorithm. Concurrently, the vision-language model is used to infer the optimal path for privacy-awareness, given the environmental layout and the navigational instruction. This approach aims to minimize the robot's exposure to human activities and preserve the privacy of the robot and its surroundings. Experimental results on the S3DIS dataset demonstrate that our framework significantly enhances mobile robots' privacy awareness of navigation in human-shared public environments. Furthermore, we demonstrate the practical applicability of our framework by successfully navigating a robotic platform through real-world office environments. The supplementary video and code can be accessed via the following link: https://sites.google.com/view/privacy-aware-nav.

Paperid: 5883, https://arxiv.org/pdf/2409.19590.pdf

Abstract:
In modern healthcare, the demand for autonomous robotic assistants has grown significantly, particularly in the operating room, where surgical tasks require precision and reliability. Robotic scrub nurses have emerged as a promising solution to improve efficiency and reduce human error during surgery. However, challenges remain in terms of accurately grasping and handing over surgical instruments, especially when dealing with complex or difficult objects in dynamic environments. In this work, we introduce a novel robotic scrub nurse system, RoboNurse-VLA, built on a Vision-Language-Action (VLA) model by integrating the Segment Anything Model 2 (SAM 2) and the Llama 2 language model. The proposed RoboNurse-VLA system enables highly precise grasping and handover of surgical instruments in real-time based on voice commands from the surgeon. Leveraging state-of-the-art vision and language models, the system can address key challenges for object detection, pose optimization, and the handling of complex and difficult-to-grasp instruments. Through extensive evaluations, RoboNurse-VLA demonstrates superior performance compared to existing models, achieving high success rates in surgical instrument handovers, even with unseen tools and challenging items. This work presents a significant step forward in autonomous surgical assistance, showcasing the potential of integrating VLA models for real-world medical applications. More details can be found at https://robonurse-vla.github.io.

Paperid: 5884, https://arxiv.org/pdf/2409.15505.pdf

Abstract:
There has been a lot of interest in grounding natural language to physical entities through visual context. While Vision Language Models (VLMs) can ground linguistic instructions to visual sensory information, they struggle with grounding non-visual attributes, like the weight of an object. Our key insight is that non-visual attribute detection can be effectively achieved by active perception guided by visual reasoning. To this end, we present a perception-action API that consists of VLMs and Large Language Models (LLMs) as backbones, together with a set of robot control functions. When prompted with this API and a natural language query, an LLM generates a program to actively identify attributes given an input image. Offline testing on the Odd-One-Out dataset demonstrates that our framework outperforms vanilla VLMs in detecting attributes like relative object location, size, and weight. Online testing in realistic household scenes on AI2-THOR and a real robot demonstration on a DJI RoboMaster EP robot highlight the efficacy of our approach.

Paperid: 5885, https://arxiv.org/pdf/2409.12499.pdf

Abstract:
Open-vocabulary video visual relationship detection aims to expand video visual relationship detection beyond annotated categories by detecting unseen relationships between both seen and unseen objects in videos. Existing methods usually use trajectory detectors trained on closed datasets to detect object trajectories, and then feed these trajectories into large-scale pre-trained vision-language models to achieve open-vocabulary classification. Such heavy dependence on the pre-trained trajectory detectors limits their ability to generalize to novel object categories, leading to performance degradation. To address this challenge, we propose to unify object trajectory detection and relationship classification into an end-to-end open-vocabulary framework. Under this framework, we propose a relationship-aware open-vocabulary trajectory detector. It primarily consists of a query-based Transformer decoder, where the visual encoder of CLIP is distilled for frame-wise open-vocabulary object detection, and a trajectory associator. To exploit relationship context during trajectory detection, a relationship query is embedded into the Transformer decoder, and accordingly, an auxiliary relationship loss is designed to enable the decoder to perceive the relationships between objects explicitly. Moreover, we propose an open-vocabulary relationship classifier that leverages the rich semantic knowledge of CLIP to discover novel relationships. To adapt CLIP well to relationship classification, we design a multi-modal prompting method that employs spatio-temporal visual prompting for visual representation and vision-guided language prompting for language input. Extensive experiments on two public datasets, VidVRD and VidOR, demonstrate the effectiveness of our framework. Our framework is also applied to a more difficult cross-dataset scenario to further demonstrate its generalization ability.

Paperid: 5886, https://arxiv.org/pdf/2409.11941.pdf

Abstract:
When your robot grasps an object using dexterous hands or grippers, it should understand the Task-Oriented Affordances of the Object(TOAO), as different tasks often require attention to specific parts of the object. To address this challenge, we propose GauTOAO, a Gaussian-based framework for Task-Oriented Affordance of Objects, which leverages vision-language models in a zero-shot manner to predict affordance-relevant regions of an object, given a natural language query. Our approach introduces a new paradigm: "static camera, moving object," allowing the robot to better observe and understand the object in hand during manipulation. GauTOAO addresses the limitations of existing methods, which often lack effective spatial grouping, by extracting a comprehensive 3D object mask using DINO features. This mask is then used to conditionally query gaussians, producing a refined semantic distribution over the object for the specified task. This approach results in more accurate TOAO extraction, enhancing the robot's understanding of the object and improving task performance. We validate the effectiveness of GauTOAO through real-world experiments, demonstrating its capability to generalize across various tasks.

Paperid: 5887, https://arxiv.org/pdf/2409.06726.pdf

Abstract:
Although vision-language pre-training (VLP) models have achieved remarkable progress on cross-modal tasks, they remain vulnerable to adversarial attacks. Using data augmentation and cross-modal interactions to generate transferable adversarial examples on surrogate models, transfer-based black-box attacks have become the mainstream methods in attacking VLP models, as they are more practical in real-world scenarios. However, their transferability may be limited due to the differences on feature representation across different models. To this end, we propose a new attack paradigm called Feedback-based Modal Mutual Search (FMMS). FMMS introduces a novel modal mutual loss (MML), aiming to push away the matched image-text pairs while randomly drawing mismatched pairs closer in feature space, guiding the update directions of the adversarial examples. Additionally, FMMS leverages the target model feedback to iteratively refine adversarial examples, driving them into the adversarial region. To our knowledge, this is the first work to exploit target model feedback to explore multi-modality adversarial boundaries. Extensive empirical evaluations on Flickr30K and MSCOCO datasets for image-text matching tasks show that FMMS significantly outperforms the state-of-the-art baselines.

Paperid: 5888, https://arxiv.org/pdf/2409.05493.pdf

Abstract:
Grasping large and flat objects (e.g. a book or a pan) is often regarded as an ungraspable task, which poses significant challenges due to the unreachable grasping poses. Previous works leverage Extrinsic Dexterity like walls or table edges to grasp such objects. However, they are limited to task-specific policies and lack task planning to find pre-grasp conditions. This makes it difficult to adapt to various environments and extrinsic dexterity constraints. Therefore, we present DexDiff, a robust robotic manipulation method for long-horizon planning with extrinsic dexterity. Specifically, we utilize a vision-language model (VLM) to perceive the environmental state and generate high-level task plans, followed by a goal-conditioned action diffusion (GCAD) model to predict the sequence of low-level actions. This model learns the low-level policy from offline data with the cumulative reward guided by high-level planning as the goal condition, which allows for improved prediction of robot actions. Experimental results demonstrate that our method not only effectively performs ungraspable tasks but also generalizes to previously unseen objects. It outperforms baselines by a 47% higher success rate in simulation and facilitates efficient deployment and manipulation in real-world scenarios.

Paperid: 5889, https://arxiv.org/pdf/2601.19281.pdf

Abstract:
Smart glasses enhance interactions with the environment by using head-mounted cameras to observe the user's viewpoint, but lack the visual feedback used for common interactions. We introduce Gazeify then Voiceify, a multimodal approach allowing object selection via gaze and voice using displayless smart glasses. Users can select a physical object with their gaze, and the system generates a digital mask and a voice description of the object's semantics. Users can further correct errors through free-form conversation. To demonstrate our approach, we develop an interactive system by integrating advanced object segmentation and detection with a vision-language model. User studies reveal that participants achieve correct gaze selection in 53% of the task trials and use voice disambiguation to correct 58% of the remaining errors. Participants also rated the system as likable, useful, and easy to use.

Paperid: 5890, https://arxiv.org/pdf/2601.18442.pdf

Abstract:
Autonomous vehicle safety validation requires testing on safety-critical scenarios, but these events are rare in real-world driving and costly to test due to collision risks. Crash reports provide authentic specifications of safety-critical events, offering a vital alternative to scarce real-world collision trajectory data. This makes them valuable sources for generating realistic high-risk scenarios through simulation. Existing approaches face significant limitations because data-driven methods lack diversity due to their reliance on existing latent distributions, whereas adversarial methods often produce unrealistic scenarios lacking physical fidelity. Large Language Model (LLM) and Vision Language Model (VLM)-based methods show significant promise. However, they suffer from context suppression issues where internal parametric knowledge overrides crash specifications, producing scenarios that deviate from actual accident characteristics. This paper presents SG-CADVLM (A Context-Aware Decoding Powered Vision Language Model for Safety-Critical Scenario Generation), a framework that integrates Context-Aware Decoding with multi-modal input processing to generate safety-critical scenarios from crash reports and road network diagrams. The framework mitigates VLM hallucination issues while enabling the simultaneous generation of road geometry and vehicle trajectories. The experimental results demonstrate that SG-CADVLM generates critical risk scenarios at a rate of 84.4% compared to 12.5% for the baseline methods, representing an improvement of 469%, while producing executable simulations for autonomous vehicle testing.

Paperid: 5891, https://arxiv.org/pdf/2601.16895.pdf

Abstract:
Surgery is a highly complex process, and artificial intelligence has emerged as a transformative force in supporting surgical guidance and decision-making. However, the unimodal nature of most current AI systems limits their ability to achieve a holistic understanding of surgical workflows. This highlights the need for general-purpose surgical AI systems capable of comprehensively modeling the interrelated components of surgical scenes. Recent advances in large vision-language models that integrate multimodal data processing offer strong potential for modeling surgical tasks and providing human-like scene reasoning and understanding. Despite their promise, systematic investigations of VLMs in surgical applications remain limited. In this study, we evaluate the effectiveness of large VLMs for the fundamental surgical vision task of detecting surgical tools. Specifically, we investigate three state-of-the-art VLMs, Qwen2.5, LLaVA1.5, and InternVL3.5, on the GraSP robotic surgery dataset under both zero-shot and parameter-efficient LoRA fine-tuning settings. Our results demonstrate that Qwen2.5 consistently achieves superior detection performance in both configurations among the evaluated VLMs. Furthermore, compared with the open-set detection baseline Grounding DINO, Qwen2.5 exhibits stronger zero-shot generalization and comparable fine-tuned performance. Notably, Qwen2.5 shows superior instrument recognition, while Grounding DINO demonstrates stronger localization.

Paperid: 5892, https://arxiv.org/pdf/2601.15308.pdf

Abstract:
Extended Reality (XR), including virtual, augmented, and mixed reality, provides immersive and interactive experiences across diverse applications, from VR-based education to AR-based assistance and MR-based training. However, widespread XR adoption remains limited due to two key challenges: 1) the high cost and complexity of authoring 3D content, especially for large-scale environments or complex interactions; and 2) the steep learning curve associated with non-intuitive interaction methods like handheld controllers or scripted gestures. Generative AI (GenAI) presents a promising solution by enabling intuitive, language-driven interaction and automating content generation. Leveraging vision-language models and diffusion-based generation, GenAI can interpret ambiguous instructions, understand physical scenes, and generate or manipulate 3D content, significantly lowering barriers to XR adoption. This paper explores the integration of XR and GenAI through three concrete use cases, showing how they address key obstacles in scalability and natural interaction, and identifying technical challenges that must be resolved to enable broader adoption.

Paperid: 5893, https://arxiv.org/pdf/2601.12142.pdf

Abstract:
Vision Language Action (VLA) models promise an open-vocabulary interface that can translate perceptual ambiguity into semantically grounded driving decisions, yet they still treat language as a static prior fixed at inference time. As a result, the model must infer continuously shifting objectives from pixels alone, yielding delayed or overly conservative maneuvers. We argue that effective VLAs for autonomous driving need an online channel in which users can influence driving with specific intentions. To this end, we present EchoVLA, a user-aware VLA that couples camera streams with in situ audio instructions. We augment the nuScenes dataset with temporally aligned, intent-specific speech commands generated by converting ego-motion descriptions into synthetic audios. Further, we compose emotional speech-trajectory pairs into a multimodal Chain-of-Thought (CoT) for fine-tuning a Multimodal Large Model (MLM) based on Qwen2.5-Omni. Specifically, we synthesize the audio-augmented dataset with different emotion types paired with corresponding driving behaviors, leveraging the emotional cues embedded in tone, pitch, and speech tempo to reflect varying user states, such as urgent or hesitant intentions, thus enabling our EchoVLA to interpret not only the semantic content but also the emotional context of audio commands for more nuanced and emotionally adaptive driving behavior. In open-loop benchmarks, our approach reduces the average L2 error by $59.4\%$ and the collision rate by $74.4\%$ compared to the baseline of vision-only perception. More experiments on nuScenes dataset validate that EchoVLA not only steers the trajectory through audio instructions, but also modulates driving behavior in response to the emotions detected in the user's speech.

Paperid: 5894, https://arxiv.org/pdf/2601.10010.pdf

Abstract:
Video Large Language Models (VideoLLMs) exhibit various types of hallucinations. Existing research has primarily focused on hallucinations involving the presence of events, objects, and scenes in videos, while largely neglecting event relation hallucination. In this paper, we introduce a novel benchmark for evaluating the Video Event Relation Hallucination, named VERHallu. This benchmark focuses on causal, temporal, and subevent relations between events, encompassing three types of tasks: relation classification, question answering, and counterfactual question answering, for a comprehensive evaluation of event relation hallucination. Additionally, it features counterintuitive video scenarios that deviate from typical pretraining distributions, with each sample accompanied by human-annotated candidates covering both vision-language and pure language biases. Our analysis reveals that current state-of-the-art VideoLLMs struggle with dense-event relation reasoning, often relying on prior knowledge due to insufficient use of frame-level cues. Although these models demonstrate strong grounding capabilities for key events, they often overlook the surrounding subevents, leading to an incomplete and inaccurate understanding of event relations. To tackle this, we propose a Key-Frame Propagating (KFP) strategy, which reallocates frame-level attention within intermediate layers to enhance multi-event understanding. Experiments show it effectively mitigates the event relation hallucination without affecting inference speed.

Paperid: 5895, https://arxiv.org/pdf/2601.05851.pdf

Abstract:
Real-time multimodal auto-completion is essential for digital assistants, chatbots, design tools, and healthcare consultations, where user inputs rely on shared visual context. We introduce Multimodal Auto-Completion (MAC), a task that predicts upcoming characters in live chats using partially typed text and visual cues. Unlike traditional text-only auto-completion (TAC), MAC grounds predictions in multimodal context to better capture user intent. To enable this task, we adapt MMDialog and ImageChat to create benchmark datasets. We evaluate leading vision-language models (VLMs) against strong textual baselines, highlighting trade-offs in accuracy and efficiency. We present Router-Suggest, a router framework that dynamically selects between textual models and VLMs based on dialog context, along with a lightweight variant for resource-constrained environments. Router-Suggest achieves a 2.3x to 10x speedup over the best-performing VLM. A user study shows that VLMs significantly excel over textual models on user satisfaction, notably saving user typing effort and improving the quality of completions in multi-turn conversations. These findings underscore the need for multimodal context in auto-completions, leading to smarter, user-aware assistants.

Paperid: 5896, https://arxiv.org/pdf/2601.05529.pdf

Abstract:
One mistake by an AI system in a safety-critical setting can cost lives. As Large Language Models (LLMs) become integral to robotics decision-making, the physical dimension of risk grows; a single wrong instruction can directly endanger human safety. This paper addresses the urgent need to systematically evaluate LLM performance in scenarios where even minor errors are catastrophic. Through a qualitative evaluation of a fire evacuation scenario, we identified critical failure cases in LLM-based decision-making. Based on these, we designed seven tasks for quantitative assessment, categorized into: Complete Information, Incomplete Information, and Safety-Oriented Spatial Reasoning (SOSR). Complete information tasks utilize ASCII maps to minimize interpretation ambiguity and isolate spatial reasoning from visual processing. Incomplete information tasks require models to infer missing context, testing for spatial continuity versus hallucinations. SOSR tasks use natural language to evaluate safe decision-making in life-threatening contexts. We benchmark various LLMs and Vision-Language Models (VLMs) across these tasks. Beyond aggregate performance, we analyze the implications of a 1% failure rate, highlighting how "rare" errors escalate into catastrophic outcomes. Results reveal serious vulnerabilities: several models achieved a 0% success rate in ASCII navigation, while in a simulated fire drill, models instructed robots to move toward hazardous areas instead of emergency exits. Our findings lead to a sobering conclusion: current LLMs are not ready for direct deployment in safety-critical systems. A 99% accuracy rate is dangerously misleading in robotics, as it implies one out of every hundred executions could result in catastrophic harm. We demonstrate that even state-of-the-art models cannot guarantee safety, and absolute reliance on them creates unacceptable risks.

Paperid: 5897, https://arxiv.org/pdf/2601.03460.pdf

Abstract:
End-to-end (E2E) models in autonomous driving aim to directly map sensor inputs to control commands, but their ability to generalize to novel and complex scenarios remains a key challenge. The common practice of fully fine-tuning the vision encoder on driving datasets potentially limits its generalization by causing the model to specialize too heavily in the training data. This work challenges the necessity of this training paradigm. We propose FROST-Drive, a novel E2E architecture designed to preserve and leverage the powerful generalization capabilities of a pretrained vision encoder from a Vision-Language Model (VLM). By keeping the encoder's weights frozen, our approach directly transfers the rich, generalized world knowledge from the VLM to the driving task. Our model architecture combines this frozen encoder with a transformer-based adapter for multimodal fusion and a GRU-based decoder for smooth waypoint generation. Furthermore, we introduce a custom loss function designed to directly optimize for Rater Feedback Score (RFS), a metric that prioritizes robust trajectory planning. We conduct extensive experiments on Waymo Open E2E Dataset, a large-scale datasets deliberately curated to capture the long-tail scenarios, demonstrating that our frozen-encoder approach significantly outperforms models that employ full fine-tuning. Our results provide substantial evidence that preserving the broad knowledge of a capable VLM is a more effective strategy for achieving robust, generalizable driving performance than intensive domain-specific adaptation. This offers a new pathway for developing vision-based models that can better handle the complexities of real-world application domains.

Paperid: 5898, https://arxiv.org/pdf/2601.03449.pdf

Abstract:
Wildfire monitoring demands autonomous systems capable of reasoning under extreme visual degradation, rapidly evolving physical dynamics, and scarce real-world training data. Existing UAV navigation approaches rely on simplified simulators and supervised perception pipelines, and lack embodied agents interacting with physically realistic fire environments. We introduce FIRE-VLM, the first end-to-end vision-language model (VLM) guided reinforcement learning (RL) framework trained entirely within a high-fidelity, physics-grounded wildfire digital twin. Built from USGS Digital Elevation Model (DEM) terrain, LANDFIRE fuel inventories, and semi-physical fire-spread solvers, this twin captures terrain-induced runs, wind-driven acceleration, smoke plume occlusion, and dynamic fuel consumption. Within this environment, a PPO agent with dual-view UAV sensing is guided by a CLIP-style VLM. Wildfire-specific semantic alignment scores, derived from a single prompt describing active fire and smoke plumes, are integrated as potential-based reward shaping signals. Our contributions are: (1) a GIS-to-simulation pipeline for constructing wildfire digital twins; (2) a VLM-guided RL agent for UAV firefront tracking; and (3) a wildfire-aware reward design that combines physical terms with VLM semantics. Across five digital-twin evaluation tasks, our VLM-guided policy reduces time-to-detection by up to 6 times, increases time-in-FOV, and is, to our knowledge, the first RL-based UAV wildfire monitoring system demonstrated in kilometer-scale, physics-grounded digital-twin fires.

Paperid: 5899, https://arxiv.org/pdf/2601.03136.pdf

Abstract:
Language plays a critical role in Vision-Language-Action (VLA) models, yet the linguistic characteristics of the datasets used to train and evaluate these systems remain poorly documented. In this work, we present a systematic dataset audit of several widely used VLA corpora, aiming to characterize what kinds of instructions these datasets actually contain and how much linguistic variety they provide. We quantify instruction language along complementary dimensions-including lexical variety, duplication and overlap, semantic similarity, and syntactic complexity. Our analysis shows that many datasets rely on highly repetitive, template-like commands with limited structural variation, yielding a narrow distribution of instruction forms. We position these findings as descriptive documentation of the language signal available in current VLA training and evaluation data, intended to support more detailed dataset reporting, more principled dataset selection, and targeted curation or augmentation strategies that broaden language coverage.

Paperid: 5900, https://arxiv.org/pdf/2601.02439.pdf

Abstract:
We present WebGym, the largest-to-date open-source environment for training realistic visual web agents. Real websites are non-stationary and diverse, making artificial or small-scale task sets insufficient for robust policy learning. WebGym contains nearly 300,000 tasks with rubric-based evaluations across diverse, real-world websites and difficulty levels. We train agents with a simple reinforcement learning (RL) recipe, which trains on the agent's own interaction traces (rollouts), using task rewards as feedback to guide learning. To enable scaling RL, we speed up sampling of trajectories in WebGym by developing a high-throughput asynchronous rollout system, designed specifically for web agents. Our system achieves a 4-5x rollout speedup compared to naive implementations. Second, we scale the task set breadth, depth, and size, which results in continued performance improvement. Fine-tuning a strong base vision-language model, Qwen-3-VL-8B-Instruct, on WebGym results in an improvement in success rate on an out-of-distribution test set from 26.2% to 42.9%, significantly outperforming agents based on proprietary models such as GPT-4o and GPT-5-Thinking that achieve 27.1% and 29.8%, respectively. This improvement is substantial because our test set consists only of tasks on websites never seen during training, unlike many other prior works on training visual web agents.

Paperid: 5901, https://arxiv.org/pdf/2601.02316.pdf

Abstract:
Empirical evaluation serves as the primary compass guiding research progress in foundation models. Despite a large body of work focused on training frontier vision-language models (VLMs), approaches to their evaluation remain nascent. To guide their maturation, we propose three desiderata that evaluations should satisfy: (1) faithfulness to the modality and application, (2) discriminability between models of varying quality, and (3) efficiency in compute. Through this lens, we identify critical failure modes that violate faithfulness and discriminability, misrepresenting model capabilities: (i) multiple-choice formats reward guessing, poorly reflect downstream use cases, and saturate early as models improve; (ii) blindly solvable questions, which can be answered without images, constitute up to 70% of some evaluations; and (iii) mislabeled or ambiguous samples compromise up to 42% of examples in certain datasets. Regarding efficiency, the computational burden of evaluating frontier models has become prohibitive: by some accounts, nearly 20% of development compute is devoted to evaluation alone. Rather than discarding existing benchmarks, we curate them via transformation and filtering to maximize fidelity and discriminability. We find that converting multiple-choice questions to generative tasks reveals sharp capability drops of up to 35%. In addition, filtering blindly solvable and mislabeled samples improves discriminative power while simultaneously reducing computational cost. We release DatBench-Full, a cleaned evaluation suite of 33 datasets spanning nine VLM capabilities, and DatBench, a discriminative subset that achieves 13x average speedup (up to 50x) while closely matching the discriminative power of the original datasets. Our work outlines a path toward evaluation practices that are both rigorous and sustainable as VLMs continue to scale.

Paperid: 5902, https://arxiv.org/pdf/2601.00777.pdf

Abstract:
While Vision-Language Models (VLMs) and Multimodal Large Language Models (MLLMs) have shown strong generalisation in detecting image and video deepfakes, their use for audio deepfake detection remains largely unexplored. In this work, we aim to explore the potential of MLLMs for audio deepfake detection. Combining audio inputs with a range of text prompts as queries to find out the viability of MLLMs to learn robust representations across modalities for audio deepfake detection. Therefore, we attempt to explore text-aware and context-rich, question-answer based prompts with binary decisions. We hypothesise that such a feature-guided reasoning will help in facilitating deeper multimodal understanding and enable robust feature learning for audio deepfake detection. We evaluate the performance of two MLLMs, Qwen2-Audio-7B-Instruct and SALMONN, in two evaluation modes: (a) zero-shot and (b) fine-tuned. Our experiments demonstrate that combining audio with a multi-prompt approach could be a viable way forward for audio deepfake detection. Our experiments show that the models perform poorly without task-specific training and struggle to generalise to out-of-domain data. However, they achieve good performance on in-domain data with minimal supervision, indicating promising potential for audio deepfake detection.

Paperid: 5903, https://arxiv.org/pdf/2512.21985.pdf

Abstract:
In high-stakes domains, small task-specific vision models are crucial due to their low computational requirements and the availability of numerous methods to explain their results. However, these explanations often reveal that the models do not align well with human domain knowledge, relying instead on spurious correlations. This might result in brittle behavior once deployed in the real-world. To address this issue, we introduce a novel and efficient method for aligning small task-specific vision models with human domain knowledge by leveraging the generalization capabilities of a Large Vision Language Model (LVLM). Our LVLM-Aided Visual Alignment (LVLM-VA) method provides a bidirectional interface that translates model behavior into natural language and maps human class-level specifications to image-level critiques, enabling effective interaction between domain experts and the model. Our method demonstrates substantial improvement in aligning model behavior with human specifications, as validated on both synthetic and real-world datasets. We show that it effectively reduces the model's dependence on spurious features and on group-specific biases, without requiring fine-grained feedback.

Paperid: 5904, https://arxiv.org/pdf/2512.20166.pdf

Abstract:
The capability of performing long-horizon, language-guided robotic manipulation tasks critically relies on leveraging historical information and generating coherent action sequences. However, such capabilities are often overlooked by existing Vision-Language-Action (VLA) models. To solve this challenge, we propose LoLA (Long Horizon Latent Action Learning), a framework designed for robot manipulation that integrates long-term multi-view observations and robot proprioception to enable multi-step reasoning and action generation. We first employ Vision-Language Models to encode rich contextual features from historical sequences and multi-view observations. We further introduces a key module, State-Aware Latent Re-representation, which transforms visual inputs and language commands into actionable robot motion space. Unlike existing VLA approaches that merely concatenate robot proprioception (e.g., joint angles) with VL embeddings, this module leverages such robot states to explicitly ground VL representations in physical scale through a learnable "embodiment-anchored" latent space. We trained LoLA on diverse robotic pre-training datasets and conducted extensive evaluations on simulation benchmarks (SIMPLER and LIBERO), as well as two real-world tasks on Franka and Bi-Manual Aloha robots. Results show that LoLA significantly outperforms prior state-of-the-art methods (e.g., pi0), particularly in long-horizon manipulation tasks.

Paperid: 5905, https://arxiv.org/pdf/2512.19083.pdf

Abstract:
Autonomous navigation for Unmanned Aerial Vehicles faces key challenges from limited onboard computational resources, which restrict deployed deep neural networks to shallow architectures incapable of handling complex environments. Offloading tasks to remote edge servers introduces high latency, creating an inherent trade-off in system design. To address these limitations, we propose CoDrone - the first cloud-edge-end collaborative computing framework integrating foundation models into autonomous UAV cruising scenarios - effectively leveraging foundation models to enhance performance of resource-constrained unmanned aerial vehicle platforms. To reduce onboard computation and data transmission overhead, CoDrone employs grayscale imagery for the navigation model. When enhanced environmental perception is required, CoDrone leverages the edge-assisted foundation model Depth Anything V2 for depth estimation and introduces a novel one-dimensional occupancy grid-based navigation method - enabling fine-grained scene understanding while advancing efficiency and representational simplicity of autonomous navigation. A key component of CoDrone is a Deep Reinforcement Learning-based neural scheduler that seamlessly integrates depth estimation with autonomous navigation decisions, enabling real-time adaptation to dynamic environments. Furthermore, the framework introduces a UAV-specific vision language interaction module incorporating domain-tailored low-level flight primitives to enable effective interaction between the cloud foundation model and the UAV. The introduction of VLM enhances open-set reasoning capabilities in complex unseen scenarios. Experimental results show CoDrone outperforms baseline methods under varying flight speeds and network conditions, achieving a 40% increase in average flight distance and a 5% improvement in average Quality of Navigation.

Paperid: 5906, https://arxiv.org/pdf/2512.18897.pdf

Abstract:
Vocabulary-free fine-grained image recognition aims to distinguish visually similar categories within a meta-class without a fixed, human-defined label set. Existing solutions for this problem are limited by either the usage of a large and rigid list of vocabularies or by the dependency on complex pipelines with fragile heuristics where errors propagate across stages. Meanwhile, the ability of recent large multi-modal models (LMMs) equipped with explicit or implicit reasoning to comprehend visual-language data, decompose problems, retrieve latent knowledge, and self-correct suggests a more principled and effective alternative. Building on these capabilities, we propose FiNDR (Fine-grained Name Discovery via Reasoning), the first reasoning-augmented LMM-based framework for vocabulary-free fine-grained recognition. The system operates in three automated steps: (i) a reasoning-enabled LMM generates descriptive candidate labels for each image; (ii) a vision-language model filters and ranks these candidates to form a coherent class set; and (iii) the verified names instantiate a lightweight multi-modal classifier used at inference time. Extensive experiments on popular fine-grained classification benchmarks demonstrate state-of-the-art performance under the vocabulary-free setting, with a significant relative margin of up to 18.8% over previous approaches. Remarkably, the proposed method surpasses zero-shot baselines that exploit pre-defined ground-truth names, challenging the assumption that human-curated vocabularies define an upper bound. Additionally, we show that carefully curated prompts enable open-source LMMs to match proprietary counterparts. These findings establish reasoning-augmented LMMs as an effective foundation for scalable, fully automated, open-world fine-grained visual recognition. The source code is available on github.com/demidovd98/FiNDR.

Paperid: 5907, https://arxiv.org/pdf/2512.18396.pdf

Abstract:
Recent advances in Vision-Language-Action (VLA) and world-model methods have improved generalization in tasks such as robotic manipulation and object interaction. However, Successful execution of such tasks depends on large, costly collections of real demonstrations, especially for fine-grained manipulation of articulated objects. To address this, we present AOMGen, a scalable data generation framework for articulated manipulation which is instantiated from a single real scan, demonstration and a library of readily available digital assets, yielding photoreal training data with verified physical states. The framework synthesizes synchronized multi-view RGB temporally aligned with action commands and state annotations for joints and contacts, and systematically varies camera viewpoints, object styles, and object poses to expand a single execution into a diverse corpus. Experimental results demonstrate that fine-tuning VLA policies on AOMGen data increases the success rate from 0% to 88.7%, and the policies are tested on unseen objects and layouts.

Paperid: 5908, https://arxiv.org/pdf/2512.17292.pdf

Abstract:
Many image restoration (IR) tasks require both pixel-level fidelity and high-level semantic understanding to recover realistic photos with fine-grained details. However, previous approaches often struggle to effectively leverage both the visual and linguistic knowledge. Recent efforts have attempted to incorporate Vision-language models (VLMs), which excel at aligning visual and textual features, into universal IR. Nevertheless, these methods fail to utilize the linguistic priors to ensure semantic coherence during the restoration process. To address this issue, in this paper, we propose the Vision-Language Model Guided Image Restoration (VLMIR) framework, which leverages the rich vision-language priors of VLMs, such as CLIP, to enhance IR performance through improved visual perception and semantic understanding. Our approach consists of two stages: VLM-based feature extraction and diffusion-based image restoration. In the first stage, we extract complementary visual and linguistic representations of input images by condensing the visual perception and high-level semantic priors through VLMs. Specifically, we align the embeddings of captions from low-quality and high-quality images using a cosine similarity loss with LoRA fine-tuning, and employ a degradation predictor to decompose degradation and clean image content embeddings. These complementary visual and textual embeddings are then integrated into a diffusion-based model via cross-attention mechanisms for enhanced restoration. Extensive experiments and ablation studies demonstrate that VLMIR achieves superior performance across both universal and degradation-specific IR tasks, underscoring the critical role of integrated visual and linguistic knowledge from VLMs in advancing image restoration capabilities.

Paperid: 5909, https://arxiv.org/pdf/2512.15971.pdf

Abstract:
Multispectral object detection is critical for safety-sensitive applications such as autonomous driving and surveillance, where robust perception under diverse illumination conditions is essential. However, the limited availability of annotated multispectral data severely restricts the training of deep detectors. In such data-scarce scenarios, textual class information can serve as a valuable source of semantic supervision. Motivated by the recent success of Vision-Language Models (VLMs) in computer vision, we explore their potential for few-shot multispectral object detection. Specifically, we adapt two representative VLM-based detectors, Grounding DINO and YOLO-World, to handle multispectral inputs and propose an effective mechanism to integrate text, visual and thermal modalities. Through extensive experiments on two popular multispectral image benchmarks, FLIR and M3FD, we demonstrate that VLM-based detectors not only excel in few-shot regimes, significantly outperforming specialized multispectral models trained with comparable data, but also achieve competitive or superior results under fully supervised settings. Our findings reveal that the semantic priors learned by large-scale VLMs effectively transfer to unseen spectral modalities, ofFering a powerful pathway toward data-efficient multispectral perception.

Paperid: 5910, https://arxiv.org/pdf/2512.12885.pdf

Abstract:
Automated road sign recognition is a critical task for intelligent transportation systems, but traditional deep learning methods struggle with the sheer number of sign classes and the impracticality of creating exhaustive labeled datasets. This paper introduces a novel zero-shot recognition framework that adapts the Retrieval-Augmented Generation (RAG) paradigm to address this challenge. Our method first uses a Vision Language Model (VLM) to generate a textual description of a sign from an input image. This description is used to retrieve a small set of the most relevant sign candidates from a vector database of reference designs. Subsequently, a Large Language Model (LLM) reasons over the retrieved candidates to make a final, fine-grained recognition. We validate this approach on a comprehensive set of 303 regulatory signs from the Ohio MUTCD. Experimental results demonstrate the framework's effectiveness, achieving 95.58% accuracy on ideal reference images and 82.45% on challenging real-world road data. This work demonstrates the viability of RAG-based architectures for creating scalable and accurate systems for road sign recognition without task-specific training.

Paperid: 5911, https://arxiv.org/pdf/2512.10719.pdf

Abstract:
End-to-end autonomous driving methods built on vision language models (VLMs) have undergone rapid development driven by their universal visual understanding and strong reasoning capabilities obtained from the large-scale pretraining. However, we find that current VLMs struggle to understand fine-grained 3D spatial relationships which is a fundamental requirement for systems interacting with the physical world. To address this issue, we propose SpaceDrive, a spatial-aware VLM-based driving framework that treats spatial information as explicit positional encodings (PEs) instead of textual digit tokens, enabling joint reasoning over semantic and spatial representations. SpaceDrive employs a universal positional encoder to all 3D coordinates derived from multi-view depth estimation, historical ego-states, and text prompts. These 3D PEs are first superimposed to augment the corresponding 2D visual tokens. Meanwhile, they serve as a task-agnostic coordinate representation, replacing the digit-wise numerical tokens as both inputs and outputs for the VLM. This mechanism enables the model to better index specific visual semantics in spatial reasoning and directly regress trajectory coordinates rather than generating digit-by-digit, thereby enhancing planning accuracy. Extensive experiments validate that SpaceDrive achieves state-of-the-art open-loop performance on the nuScenes dataset and the second-best Driving Score of 78.02 on the Bench2Drive closed-loop benchmark over existing VLM-based methods.

Paperid: 5912, https://arxiv.org/pdf/2512.09573.pdf

Abstract:
Recent advances in Image Quality Assessment (IQA) have leveraged Multi-modal Large Language Models (MLLMs) to generate descriptive explanations. However, despite their strong visual perception modules, these models often fail to reliably detect basic low-level distortions such as blur, noise, and compression, and may produce inconsistent evaluations across repeated inferences. This raises an essential question: do MLLM-based IQA systems truly perceive the visual features that matter? To examine this issue, we introduce a low-level distortion perception task that requires models to classify specific distortion types. Our component-wise analysis shows that although MLLMs are structurally capable of representing such distortions, they tend to overfit training templates, leading to biases in quality scoring. As a result, critical low-level features are weakened or lost during the vision-language alignment transfer stage. Furthermore, by computing the semantic distance between visual features and corresponding semantic tokens before and after component-wise fine-tuning, we show that improving the alignment of the vision encoder dramatically enhances distortion recognition accuracy, increasing it from 14.92% to 84.43%. Overall, these findings indicate that incorporating dedicated constraints on the vision encoder can strengthen text-explainable visual representations and enable MLLM-based pipelines to produce more coherent and interpretable reasoning in vision-centric tasks.

Paperid: 5913, https://arxiv.org/pdf/2512.09555.pdf

Abstract:
Recent progress in BIQA has been driven by VLMs, whose semantic reasoning abilities suggest that they might extract visual features, generate descriptive text, and infer quality in a human-like manner. However, these models often produce textual descriptions that contradict their final quality predictions, and the predicted scores can change unstably during inference - behaviors not aligned with human reasoning. To understand these issues, we analyze the factors that cause contradictory assessments and instability. We first estimate the relationship between the final quality predictions and the generated visual features, finding that the predictions are not fully grounded in the features and that the logical connection between them is weak. Moreover, decoding intermediate VLM layers shows that the model frequently relies on a limited set of candidate tokens, which contributes to prediction instability. To encourage more human-like reasoning, we introduce a two-stage tuning method that explicitly separates visual perception from quality inference. In the first stage, the model learns visual features; in the second, it infers quality solely from these features. Experiments on SPAQ and KONIQ demonstrate that our approach reduces prediction instability from 22.00% to 12.39% and achieves average gains of 0.3124/0.3507 in SRCC/PLCC across LIVE, CSIQ, SPAQ, and KONIQ compared to the baseline. Further analyses show that our method improves both stability and the reliability of the inference process.

Paperid: 5914, https://arxiv.org/pdf/2512.08639.pdf

Abstract:
Aerial Vision-and-Language Navigation (VLN) aims to enable unmanned aerial vehicles (UAVs) to interpret natural language instructions and navigate complex urban environments using onboard visual observation. This task holds promise for real-world applications such as low-altitude inspection, search-and-rescue, and autonomous aerial delivery. Existing methods often rely on panoramic images, depth inputs, or odometry to support spatial reasoning and action planning. These requirements increase system cost and integration complexity, thus hindering practical deployment for lightweight UAVs. We present a unified aerial VLN framework that operates solely on egocentric monocular RGB observations and natural language instructions. The model formulates navigation as a next-token prediction problem, jointly optimizing spatial perception, trajectory reasoning, and action prediction through prompt-guided multi-task learning. Moreover, we propose a keyframe selection strategy to reduce visual redundancy by retaining semantically informative frames, along with an action merging and label reweighting mechanism that mitigates long-tailed supervision imbalance and facilitates stable multi-task co-training. Extensive experiments on the Aerial VLN benchmark validate the effectiveness of our method. Under the challenging monocular RGB-only setting, our model achieves strong results across both seen and unseen environments. It significantly outperforms existing RGB-only baselines and narrows the performance gap with state-of-the-art panoramic RGB-D counterparts. Comprehensive ablation studies further demonstrate the contribution of our task design and architectural choices.

Paperid: 5915, https://arxiv.org/pdf/2512.07976.pdf

Abstract:
Training end-to-end policies from image data to directly predict navigation actions for robotic systems has proven inherently difficult. Existing approaches often suffer from either the sim-to-real gap during policy transfer or a limited amount of training data with action labels. To address this problem, we introduce Vision-Language Distance (VLD) learning, a scalable framework for goal-conditioned navigation that decouples perception learning from policy learning. Instead of relying on raw sensory inputs during policy training, we first train a self-supervised distance-to-goal predictor on internet-scale video data. This predictor generalizes across both image- and text-based goals, providing a distance signal that can be minimized by a reinforcement learning (RL) policy. The RL policy can be trained entirely in simulation using privileged geometric distance signals, with injected noise to mimic the uncertainty of the trained distance predictor. At deployment, the policy consumes VLD predictions, inheriting semantic goal information-"where to go"-from large-scale visual training while retaining the robust low-level navigation behaviors learned in simulation. We propose using ordinal consistency to assess distance functions directly and demonstrate that VLD outperforms prior temporal distance approaches, such as ViNT and VIP. Experiments show that our decoupled design achieves competitive navigation performance in simulation while supporting flexible goal modalities, providing an alternative and, most importantly, scalable path toward reliable, multimodal navigation policies.

Paperid: 5916, https://arxiv.org/pdf/2512.04981.pdf

Abstract:
Large vision-language model (LVLM) based text-to-image (T2I) systems have become the dominant paradigm in image generation, yet whether they amplify social biases remains insufficiently understood. In this paper, we show that LVLM-based models produce markedly more socially biased images than non-LVLM-based models. We introduce a 1,024 prompt benchmark spanning four levels of linguistic complexity and evaluate demographic bias across multiple attributes in a systematic manner. Our analysis identifies system prompts, the predefined instructions guiding LVLMs, as a primary driver of biased behavior. Through decoded intermediate representations, token-probability diagnostics, and embedding-association analyses, we reveal how system prompts encode demographic priors that propagate into image synthesis. To this end, we propose FairPro, a training-free meta-prompting framework that enables LVLMs to self-audit and construct fairness-aware system prompts at test time. Experiments on two LVLM-based T2I models, SANA and Qwen-Image, show that FairPro substantially reduces demographic bias while preserving text-image alignment. We believe our findings provide deeper insight into the central role of system prompts in bias propagation and offer a practical, deployable approach for building more socially responsible T2I systems.

Paperid: 5917, https://arxiv.org/pdf/2512.01242.pdf

Abstract:
We study abstract visual composition, in which identity is primarily determined by the spatial configuration and relations among a small set of geometric primitives (e.g., parts, symmetry, topology). They are invariant primarily to texture and photorealistic detail. Composing such structures from fixed components under geometric constraints and vague goal specification (such as text) is non-trivial due to combinatorial placement choices, limited data, and discrete feasibility (overlap-free, allowable orientations), which create a sparse solution manifold ill-suited to purely statistical pixel-space generators. We propose a constraint-guided framework that combines explicit geometric reasoning with neural semantics. An AlphaGo-style search enforces feasibility, while a fine-tuned vision-language model scores semantic alignment as reward signals. Our algorithm uses a policy network as a heuristic in Monte-Carlo Tree Search and fine-tunes the network via search-generated plans. Inspired by the Generative Adversarial Network, we use the generated instances for adversarial reward refinement. Over time, the generation should approach the actual data more closely when the reward model cannot distinguish between generated instances and ground-truth. In the Tangram Assembly task, our approach yields higher validity and semantic fidelity than diffusion and auto-regressive baselines, especially as constraints tighten.

Paperid: 5918, https://arxiv.org/pdf/2512.00918.pdf

Abstract:
Large Vision-Language Models (LVLMs) have shown impressive multimodal understanding capabilities, yet their robustness is poorly understood. In this paper, we investigate the structural vulnerabilities of LVLMs to identify any critical neurons whose removal triggers catastrophic collapse. In this context, we propose CAN, a method to detect Consistently Activated Neurons and to locate critical neurons by progressive masking. Experiments on LLaVA-1.5-7b-hf and InstructBLIP-Vicuna-7b reveal that masking only a tiny portion of the language model's feed-forward networks (just as few as four neurons in extreme cases) suffices to trigger catastrophic collapse. Notably, critical neurons are predominantly localized in the language model rather than in the vision components, and the down-projection layer is a particularly vulnerable structure. We also observe a consistent two-stage collapse pattern: initial expressive degradation followed by sudden, complete collapse. Our findings provide important insights for safety research in LVLMs.

Paperid: 5919, https://arxiv.org/pdf/2512.00557.pdf

Abstract:
What visual information is encoded in individual brain regions, and how do distributed patterns combine to create their neural representations? Prior work has used generative models to replicate known category selectivity in isolated regions (e.g., faces in FFA), but these approaches offer limited insight into how regions interact during complex, naturalistic vision. We introduce NeuroVolve, a generative framework that provides brain-guided image synthesis via optimization of a neural objective function in the embedding space of a pretrained vision-language model. Images are generated under the guidance of a programmable neural objective, i.e., activating or deactivating single regions or multiple regions together. NeuroVolve is validated by recovering known selectivity for individual brain regions, while expanding to synthesize coherent scenes that satisfy complex, multi-region constraints. By tracking optimization steps, it reveals semantic trajectories through embedding space, unifying brain-guided image editing and preferred stimulus generation in a single process. We show that NeuroVolve can generate both low-level and semantic feature-specific stimuli for single ROIs, as well as stimuli aligned to curated neural objectives. These include co-activation and decorrelation between regions, exposing cooperative and antagonistic tuning relationships. Notably, the framework captures subject-specific preferences, supporting personalized brain-driven synthesis and offering interpretable constraints for mapping, analyzing, and probing neural representations of visual information.

Paperid: 5920, https://arxiv.org/pdf/2511.22780.pdf

Abstract:
In this work, we propose an evaluation protocol for examining the performance of robotic manipulation policies in cluttered scenes. Contrary to prior works, we approach evaluation from a psychophysical perspective, therefore we use a unified measure of clutter that accounts for environmental factors as well as the distractors quantity, characteristics, and arrangement. Using this measure, we systematically construct evaluation scenarios in both hyper-realistic simulation and real-world and conduct extensive experimentation on manipulation policies, in particular vision-language-action (VLA) models. Our experiments highlight the significant impact of scene clutter, lowering the performance of the policies, by as much as 34% and show that despite achieving similar average performance across the tasks, different VLA policies have unique vulnerabilities and a relatively low agreement on success scenarios. We further show that our clutter measure is an effective indicator of performance degradation and analyze the impact of distractors in terms of their quantity and occluding influence. At the end, we show that finetuning on enhanced data, although effective, does not equally remedy all negative impacts of clutter on performance.

Paperid: 5921, https://arxiv.org/pdf/2511.22555.pdf

Abstract:
Vision-Language-Action (VLA) models have enabled notable progress in general-purpose robotic manipulation, yet their learned policies often exhibit variable execution quality. We attribute this variability to the mixed-quality nature of human demonstrations, where the implicit principles that govern how actions should be carried out are only partially satisfied. To address this challenge, we introduce the LIBERO-Elegant benchmark with explicit criteria for evaluating execution quality. Using these criteria, we develop a decoupled refinement framework that improves execution quality without modifying or retraining the base VLA policy. We formalize Elegant Execution as the satisfaction of Implicit Task Constraints (ITCs) and train an Elegance Critic via offline Calibrated Q-Learning to estimate the expected quality of candidate actions. At inference time, a Just-in-Time Intervention (JITI) mechanism monitors critic confidence and intervenes only at decision-critical moments, providing selective, on-demand refinement. Experiments on LIBERO-Elegant and real-world manipulation tasks show that the learned Elegance Critic substantially improves execution quality, even on unseen tasks. The proposed model enables robotic control that values not only whether tasks succeed, but also how they are performed.

Paperid: 5922, https://arxiv.org/pdf/2511.22470.pdf

Abstract:
Text-based person anomaly retrieval has emerged as a challenging task, with most existing approaches relying on complex deep-learning techniques. This raises a research question: How can the model be optimized to achieve greater fine-grained features? To address this, we propose a Local-Global Hybrid Perspective (LHP) module integrated with a Vision-Language Model (VLM), designed to explore the effectiveness of incorporating both fine-grained features alongside coarse-grained features. Additionally, we investigate a Unified Image-Text (UIT) model that combines multiple objective loss functions, including Image-Text Contrastive (ITC), Image-Text Matching (ITM), Masked Language Modeling (MLM), and Masked Image Modeling (MIM) loss. Beyond this, we propose a novel iterative ensemble strategy, by combining iteratively instead of using model results simultaneously like other ensemble methods. To take advantage of the superior performance of the LHP model, we introduce a novel feature selection algorithm based on its guidance, which helps improve the model's performance. Extensive experiments demonstrate the effectiveness of our method in achieving state-of-the-art (SOTA) performance on PAB dataset, compared with previous work, with a 9.70\% improvement in R@1, 1.77\% improvement in R@5, and 1.01\% improvement in R@10.

Paperid: 5923, https://arxiv.org/pdf/2511.21663.pdf

Abstract:
In recent years, Vision-Language-Action (VLA) models in embodied intelligence have developed rapidly. However, existing adversarial attack methods require costly end-to-end training and often generate noticeable perturbation patches. To address these limitations, we propose ADVLA, a framework that directly applies adversarial perturbations on features projected from the visual encoder into the textual feature space. ADVLA efficiently disrupts downstream action predictions under low-amplitude constraints, and attention guidance allows the perturbations to be both focused and sparse. We introduce three strategies that enhance sensitivity, enforce sparsity, and concentrate perturbations. Experiments demonstrate that under an $L_{\infty}=4/255$ constraint, ADVLA combined with Top-K masking modifies less than 10% of the patches while achieving an attack success rate of nearly 100%. The perturbations are concentrated on critical regions, remain almost imperceptible in the overall image, and a single-step iteration takes only about 0.06 seconds, significantly outperforming conventional patch-based attacks. In summary, ADVLA effectively weakens downstream action predictions of VLA models under low-amplitude and locally sparse conditions, avoiding the high training costs and conspicuous perturbations of traditional patch attacks, and demonstrates unique effectiveness and practical value for attacking VLA feature spaces.

Paperid: 5924, https://arxiv.org/pdf/2511.19878.pdf

Abstract:
Vision-Language-Action (VLA) models inherit strong priors from pretrained Vision-Language Models (VLMs), but naive fine-tuning often disrupts these representations and harms generalization. Existing fixes -- freezing modules or applying uniform regularization -- either overconstrain adaptation or ignore the differing roles of VLA components. We present MAPS (Module-Wise Proximity Scheduling), the first robust fine-tuning framework for VLAs. Through systematic analysis, we uncover an empirical order in which proximity constraints should be relaxed to balance stability and flexibility. MAPS linearly schedules this relaxation, enabling visual encoders to stay close to their pretrained priors while action-oriented language layers adapt more freely. MAPS introduces no additional parameters or data, and can be seamlessly integrated into existing VLAs. Across MiniVLA-VQ, MiniVLA-OFT, OpenVLA-OFT, and challenging benchmarks such as SimplerEnv, CALVIN, LIBERO, as well as real-world evaluations on the Franka Emika Panda platform, MAPS consistently boosts both in-distribution and out-of-distribution performance (up to +30%). Our findings highlight empirically guided proximity to pretrained VLMs as a simple yet powerful principle for preserving broad generalization in VLM-to-VLA transfer.

Paperid: 5925, https://arxiv.org/pdf/2511.19257.pdf

Abstract:
With the rapid advancement of retrieval-augmented vision-language models, multimodal medical retrieval-augmented generation (MMed-RAG) systems are increasingly adopted in clinical decision support. These systems enhance medical applications by performing cross-modal retrieval to integrate relevant visual and textual evidence for tasks, e.g., report generation and disease diagnosis. However, their complex architecture also introduces underexplored adversarial vulnerabilities, particularly via visual input perturbations. In this paper, we propose Medusa, a novel framework for crafting cross-modal transferable adversarial attacks on MMed-RAG systems under a black-box setting. Specifically, Medusa formulates the attack as a perturbation optimization problem, leveraging a multi-positive InfoNCE loss (MPIL) to align adversarial visual embeddings with medically plausible but malicious textual targets, thereby hijacking the retrieval process. To enhance transferability, we adopt a surrogate model ensemble and design a dual-loop optimization strategy augmented with invariant risk minimization (IRM). Extensive experiments on two real-world medical tasks, including medical report generation and disease diagnosis, demonstrate that Medusa achieves over 90% average attack success rate across various generation models and retrievers under appropriate parameter configuration, while remaining robust against four mainstream defenses, outperforming state-of-the-art baselines. Our results reveal critical vulnerabilities in the MMed-RAG systems and highlight the necessity of robustness benchmarking in safety-critical medical applications. The code and data are available at https://anonymous.4open.science/r/MMed-RAG-Attack-F05A.

Paperid: 5926, https://arxiv.org/pdf/2511.18968.pdf

Abstract:
Cataract surgery is one of the most commonly performed surgeries worldwide, yet intraoperative complications such as iris prolapse, posterior capsule rupture (PCR), and vitreous loss remain major causes of adverse outcomes. Automated detection of such events could enable early warning systems and objective training feedback. In this work, we propose CataractCompDetect, a complication detection framework that combines phase-aware localization, SAM 2-based tracking, complication-specific risk scoring, and vision-language reasoning for final classification. To validate CataractCompDetect, we curate CataComp, the first cataract surgery video dataset annotated for intraoperative complications, comprising 53 surgeries, including 23 with clinical complications. On CataComp, CataractCompDetect achieves an average F1 score of 70.63%, with per-complication performance of 81.8% (Iris Prolapse), 60.87% (PCR), and 69.23% (Vitreous Loss). These results highlight the value of combining structured surgical priors with vision-language reasoning for recognizing rare but high-impact intraoperative events. Our dataset and code will be publicly released upon acceptance.

Paperid: 5927, https://arxiv.org/pdf/2511.16940.pdf

Abstract:
Modern Vision-Language Models (VLMs) demonstrate sophisticated reasoning, escalating privacy risks beyond simple attribute perception to individual-level linkage. Current privacy benchmarks are structurally insufficient for this new threat, as they primarily evaluate privacy perception while failing to address the more critical risk of privacy reasoning: a VLM's ability to infer and link distributed information to construct individual profiles. To address this critical gap, we propose \textbf{MultiPriv}, the first benchmark designed to systematically evaluate individual-level privacy reasoning in VLMs. We introduce the \textbf{Privacy Perception and Reasoning (PPR)} framework and construct a novel, bilingual multimodal dataset to support it. The dataset uniquely features a core component of synthetic individual profiles where identifiers (e.g., faces, names) are meticulously linked to sensitive attributes. This design enables nine challenging tasks evaluating the full PPR spectrum, from attribute detection to cross-image re-identification and chained inference. We conduct a large-scale evaluation of over 50 foundational and commercial VLMs. Our analysis reveals: (1) Many VLMs possess significant, unmeasured reasoning-based privacy risks. (2) Perception-level metrics are poor predictors of these reasoning risks, revealing a critical evaluation gap. (3) Existing safety alignments are inconsistent and ineffective against such reasoning-based attacks. MultiPriv exposes systemic vulnerabilities and provides the necessary framework for developing robust, privacy-preserving VLMs.

Paperid: 5928, https://arxiv.org/pdf/2511.14004.pdf

Abstract:
Service robots must retrieve objects in dynamic, open-world settings where requests may reference attributes ("the red mug"), spatial context ("the mug on the table"), or past states ("the mug that was here yesterday"). Existing approaches capture only parts of this problem: scene graphs capture spatial relations but ignore temporal grounding, temporal reasoning methods model dynamics but do not support embodied interaction, and dynamic scene graphs handle both but remain closed-world with fixed vocabularies. We present STAR (SpatioTemporal Active Retrieval), a framework that unifies memory queries and embodied actions within a single decision loop. STAR leverages non-parametric long-term memory and a working memory to support efficient recall, and uses a vision-language model to select either temporal or spatial actions at each step. We introduce STARBench, a benchmark of spatiotemporal object search tasks across simulated and real environments. Experiments in STARBench and on a Tiago robot show that STAR consistently outperforms scene-graph and memory-only baselines, demonstrating the benefits of treating search in time and search in space as a unified problem. For more information: https://amrl.cs.utexas.edu/STAR.

Paperid: 5929, https://arxiv.org/pdf/2511.12676.pdf

Abstract:
Deploying embodied agents that can answer questions about their surroundings in realistic real-world settings remains difficult, partly due to the scarcity of benchmarks that faithfully capture practical operating conditions. We propose infrastructure inspection as a compelling domain for open-vocabulary Embodied Question Answering (EQA): it naturally demands multi-scale reasoning, long-range spatial understanding, and complex semantic relationships, while offering unique evaluation advantages via standardized National Bridge Inventory (NBI) condition ratings (0-9), professional inspection reports, and egocentric imagery. We introduce BridgeEQA, a benchmark of 2,200 open-vocabulary question-answer pairs (in the style of OpenEQA) grounded in professional inspection reports across 200 real-world bridge scenes with 47.93 images on average per scene. Questions require synthesizing visual evidence across multiple images and aligning responses with NBI condition ratings. We further propose a new EQA metric Image Citation Relevance to evaluate the ability of a model to cite relevant images. Evaluations of state-of-the-art vision-language models reveal substantial performance gaps under episodic memory EQA settings. To address this, we propose Embodied Memory Visual Reasoning (EMVR), which formulates inspection as sequential navigation over an image-based scene graph: images are nodes, and an agent takes actions to traverse views, compare evidence, and reason within a Markov decision process. EMVR shows strong performance over the baselines. We publicly release both the dataset and code.

Paperid: 5930, https://arxiv.org/pdf/2511.12255.pdf

Abstract:
The Video Browser Showdown (VBS) challenges systems to deliver accurate results under strict time constraints. To meet this demand, we present Fusionista2.0, a streamlined video retrieval system optimized for speed and usability. All core modules were re-engineered for efficiency: preprocessing now relies on ffmpeg for fast keyframe extraction, optical character recognition uses Vintern-1B-v3.5 for robust multilingual text recognition, and automatic speech recognition employs faster-whisper for real-time transcription. For question answering, lightweight vision-language models provide quick responses without the heavy cost of large models. Beyond these technical upgrades, Fusionista2.0 introduces a redesigned user interface with improved responsiveness, accessibility, and workflow efficiency, enabling even non-expert users to retrieve relevant content rapidly. Evaluations demonstrate that retrieval time was reduced by up to 75% while accuracy and user satisfaction both increased, confirming Fusionista2.0 as a competitive and user-friendly system for large-scale video search.

Paperid: 5931, https://arxiv.org/pdf/2511.10081.pdf

Abstract:
Multimodal large language models (MLLMs) have shown remarkable capabilities in a wide range of vision-language tasks. However, the large number of visual tokens introduces significant computational overhead. To address this issue, visual token pruning has emerged as a key technique for enhancing the efficiency of MLLMs. In cognitive science, humans tend to first determine which regions of a scene to attend to ("where to look") before deciding which specific elements within those regions to process in detail ("what to select"). This two-stage strategy enables the visual system to efficiently allocate attention at a coarse spatial level before performing fine-grained selection. However, existing pruning methods primarily focus on directly optimizing "what to select", typically using attention scores or similarity metrics. They rarely consider "where to look", which has been shown to lead to inefficient spatial allocation, positional bias, and the retention of irrelevant or redundant tokens. In this paper, we propose GridPrune, a method that replaces the global Top-K mechanism with a "guide-globally, select-locally" zonal selection system. GridPrune splits the pruning process into two steps: first, it uses text-conditional guidance to dynamically allocate a token budget across spatial zones; and then, it performs local selection within each budgeted zone. Experimental results demonstrate that GridPrune achieves superior performance across various MLLM architectures. On LLaVA-NeXT-7B, GridPrune retains 96.98% of the full performance while using 11.1% of the tokens, outperforming the best-performing baseline by 2.34% at the same pruning rate.

Paperid: 5932, https://arxiv.org/pdf/2511.09883.pdf

Abstract:
3D understanding has drawn significant attention recently, leveraging Vision-Language Models (VLMs) to enable multi-modal reasoning between point cloud and text data. Current 3D-VLMs directly embed the 3D point clouds into 3D tokens, following large 2D-VLMs with powerful reasoning capabilities. However, this framework has a great computational cost limiting its application, where we identify that the bottleneck lies in processing all 3D tokens in the Large Language Model (LLM) part. This raises the question: how can we reduce the computational overhead introduced by 3D tokens while preserving the integrity of their essential information? To address this question, we introduce Hierarchical Compensatory Compression (HCC-3D) to efficiently compress 3D tokens while maintaining critical detail retention. Specifically, we first propose a global structure compression (GSC), in which we design global queries to compress all 3D tokens into a few key tokens while keeping overall structural information. Then, to compensate for the information loss in GSC, we further propose an adaptive detail mining (ADM) module that selectively recompresses salient but under-attended features through complementary scoring. Extensive experiments demonstrate that HCC-3D not only achieves extreme compression ratios (approximately 98%) compared to previous 3D-VLMs, but also achieves new state-of-the-art performance, showing the great improvements on both efficiency and performance.

Paperid: 5933, https://arxiv.org/pdf/2511.08942.pdf

Abstract:
While Vision-Language Models (VLMs) are set to transform robotic navigation, existing methods often underutilize their reasoning capabilities. To unlock the full potential of VLMs in robotics, we shift their role from passive observers to active strategists in the navigation process. Our framework outsources high-level planning to a VLM, which leverages its contextual understanding to guide a frontier-based exploration agent. This intelligent guidance is achieved through a trio of techniques: structured chain-of-thought prompting that elicits logical, step-by-step reasoning; dynamic inclusion of the agent's recent action history to prevent getting stuck in loops; and a novel capability that enables the VLM to interpret top-down obstacle maps alongside first-person views, thereby enhancing spatial awareness. When tested on challenging benchmarks like HM3D, Gibson, and MP3D, this method produces exceptionally direct and logical trajectories, marking a substantial improvement in navigation efficiency over existing approaches and charting a path toward more capable embodied agents.

Paperid: 5934, https://arxiv.org/pdf/2511.08935.pdf

Abstract:
Embodied visual navigation remains a challenging task, as agents must explore unknown environments with limited knowledge. Existing zero-shot studies have shown that incorporating memory mechanisms to support goal-directed behavior can improve long-horizon planning performance. However, they overlook visual frontier boundaries, which fundamentally dictate future trajectories and observations, and fall short of inferring the relationship between partial visual observations and navigation goals. In this paper, we propose Semantic Cognition Over Potential-based Exploration (SCOPE), a zero-shot framework that explicitly leverages frontier information to drive potential-based exploration, enabling more informed and goal-relevant decisions. SCOPE estimates exploration potential with a Vision-Language Model and organizes it into a spatio-temporal potential graph, capturing boundary dynamics to support long-horizon planning. In addition, SCOPE incorporates a self-reconsideration mechanism that revisits and refines prior decisions, enhancing reliability and reducing overconfident errors. Experimental results on two diverse embodied navigation tasks show that SCOPE outperforms state-of-the-art baselines by 4.6\% in accuracy. Further analysis demonstrates that its core components lead to improved calibration, stronger generalization, and higher decision quality.

Paperid: 5935, https://arxiv.org/pdf/2511.08865.pdf

Abstract:
In this work, we present a PICO-based robot remote operating framework that enables low-cost, real-time acquisition of hand motion and pose data, outperforming mainstream visual tracking and motion capture solutions in terms of cost-effectiveness. The framework is natively compatible with the RealMirror ecosystem, offering ready-to-use functionality for stable and precise robotic trajectory recording within the Isaac simulation environment, thereby facilitating the construction of Vision-Language-Action (VLA) datasets. Additionally, the system supports real-time teleoperation of a variety of end-effector-equipped robots, including dexterous hands and robotic grippers. This work aims to lower the technical barriers in the study of upper-limb robotic manipulation, thereby accelerating advancements in VLA-related research.

Paperid: 5936, https://arxiv.org/pdf/2511.08637.pdf

Abstract:
The internet has become the main source of data to train modern text-to-image or vision-language models, yet it is increasingly unclear whether web-scale data collection practices for training AI systems adequately respect data owners' wishes. Ignoring the owner's indication of consent around data usage not only raises ethical concerns but also has recently been elevated into lawsuits around copyright infringement cases. In this work, we aim to reveal information about data owners' consent to AI scraping and training, and study how it's expressed in DataComp, a popular dataset of 12.8 billion text-image pairs. We examine both the sample-level information, including the copyright notice, watermarking, and metadata, and the web-domain-level information, such as a site's Terms of Service (ToS) and Robots Exclusion Protocol. We estimate at least 122M of samples exhibit some indication of copyright notice in CommonPool, and find that 60\% of the samples in the top 50 domains come from websites with ToS that prohibit scraping. Furthermore, we estimate 9-13\% with 95\% confidence interval of samples from CommonPool to contain watermarks, where existing watermark detection methods fail to capture them in high fidelity. Our holistic methods and findings show that data owners rely on various channels to convey data consent, of which current AI data collection pipelines do not entirely respect. These findings highlight the limitations of the current dataset curation/release practice and the need for a unified data consent framework taking AI purposes into consideration.

Paperid: 5937, https://arxiv.org/pdf/2511.07410.pdf

Abstract:
Large Language Models (LLMs) and Vision Language Models (VLMs) have been widely used for embodied symbolic planning. Yet, how to effectively use these models for closed-loop symbolic planning remains largely unexplored. Because they operate as black boxes, LLMs and VLMs can produce unpredictable or costly errors, making their use in high-level robotic planning especially challenging. In this work, we investigate how to use VLMs as closed-loop symbolic planners for robotic applications from a control-theoretic perspective. Concretely, we study how the control horizon and warm-starting impact the performance of VLM symbolic planners. We design and conduct controlled experiments to gain insights that are broadly applicable to utilizing VLMs as closed-loop symbolic planners, and we discuss recommendations that can help improve the performance of VLM symbolic planners.

Paperid: 5938, https://arxiv.org/pdf/2510.27164.pdf

Abstract:
Vision-language models (VLMs) often struggle to generate accurate and detailed captions for high-resolution images since they are typically pre-trained on low-resolution inputs (e.g., 224x224 or 336x336 pixels). Downscaling high-resolution images to these dimensions may result in the loss of visual details and the omission of important objects. To address this limitation, we propose a novel pipeline that integrates vision-language models, large language models (LLMs), and object detection systems to enhance caption quality. Our proposed pipeline refines captions through a novel, multi-stage process. Given a high-resolution image, an initial caption is first generated using a VLM, and key objects in the image are then identified by an LLM. The LLM predicts additional objects likely to co-occur with the identified key objects, and these predictions are verified by object detection systems. Newly detected objects not mentioned in the initial caption undergo focused, region-specific captioning to ensure they are incorporated. This process enriches caption detail while reducing hallucinations by removing references to undetected objects. We evaluate the enhanced captions using pairwise comparison and quantitative scoring from large multimodal models, along with a benchmark for hallucination detection. Experiments on a curated dataset of high-resolution images demonstrate that our pipeline produces more detailed and reliable image captions while effectively minimizing hallucinations.

Paperid: 5939, https://arxiv.org/pdf/2510.26905.pdf

Abstract:
Cyber-physical systems increasingly rely on Foundational Models such as Large Language Models (LLMs) and Vision-Language Models (VLMs) to increase autonomy through enhanced perception, inference, and planning. However, these models also introduce new types of errors, such as hallucinations, overgeneralizations, and context misalignments, resulting in incorrect and flawed decisions. To address this, we introduce the concept of Cognition Envelopes, designed to establish reasoning boundaries that constrain AI-generated decisions while complementing the use of meta-cognition and traditional safety envelopes. As with safety envelopes, Cognition Envelopes require practical guidelines and systematic processes for their definition, validation, and assurance.

Paperid: 5940, https://arxiv.org/pdf/2510.22838.pdf

Abstract:
The "style trap" poses a significant challenge for Large Vision-Language Models (LVLMs), hindering robust semantic understanding across diverse visual styles, especially in in-context learning (ICL). Existing methods often fail to effectively decouple style from content, hindering generalization. To address this, we propose the Semantic-Preserving Cross-Style Visual Reasoner (SP-CSVR), a novel framework for stable semantic understanding and adaptive cross-style visual reasoning. SP-CSVR integrates a Cross-Style Feature Encoder (CSFE) for style-content disentanglement, a Semantic-Aligned In-Context Decoder (SAICD) for efficient few-shot style adaptation, and an Adaptive Semantic Consistency Module (ASCM) employing multi-task contrastive learning to enforce cross-style semantic invariance. Extensive experiments on a challenging multi-style dataset demonstrate SP-CSVR's state-of-the-art performance across visual captioning, visual question answering, and in-context style adaptation. Comprehensive evaluations, including ablation studies and generalization analysis, confirm SP-CSVR's efficacy in enhancing robustness, generalization, and efficiency across diverse visual styles.

Paperid: 5941, https://arxiv.org/pdf/2510.18751.pdf

Abstract:
Climate change is intensifying the occurrence of harmful algal bloom (HAB), particularly cyanobacteria, which threaten aquatic ecosystems and human health through oxygen depletion, toxin release, and disruption of marine biodiversity. Traditional monitoring approaches, such as manual water sampling, remain labor-intensive and limited in spatial and temporal coverage. Recent advances in vision-language models (VLMs) for remote sensing have shown potential for scalable AI-driven solutions, yet challenges remain in reasoning over imagery and quantifying bloom severity. In this work, we introduce ALGae Observation and Segmentation (ALGOS), a segmentation-and-reasoning system for HAB monitoring that combines remote sensing image understanding with severity estimation. Our approach integrates GeoSAM-assisted human evaluation for high-quality segmentation mask curation and fine-tunes vision language model on severity prediction using the Cyanobacteria Aggregated Manual Labels (CAML) from NASA. Experiments demonstrate that ALGOS achieves robust performance on both segmentation and severity-level estimation, paving the way toward practical and automated cyanobacterial monitoring systems.

Paperid: 5942, https://arxiv.org/pdf/2510.16870.pdf

Abstract:
While brain-inspired artificial intelligence(AI) has demonstrated promising results, current understanding of the parallels between artificial neural networks (ANNs) and human brain processing remains limited: (1) unimodal ANN studies fail to capture the brain's inherent multimodal processing capabilities, and (2) multimodal ANN research primarily focuses on high-level model outputs, neglecting the crucial role of individual neurons. To address these limitations, we propose a novel neuron-level analysis framework that investigates the multimodal information processing mechanisms in vision-language models (VLMs) through the lens of human brain activity. Our approach uniquely combines fine-grained artificial neuron (AN) analysis with fMRI-based voxel encoding to examine two architecturally distinct VLMs: CLIP and METER. Our analysis reveals four key findings: (1) ANs successfully predict biological neurons (BNs) activities across multiple functional networks (including language, vision, attention, and default mode), demonstrating shared representational mechanisms; (2) Both ANs and BNs demonstrate functional redundancy through overlapping neural representations, mirroring the brain's fault-tolerant and collaborative information processing mechanisms; (3) ANs exhibit polarity patterns that parallel the BNs, with oppositely activated BNs showing mirrored activation trends across VLM layers, reflecting the complexity and bidirectional nature of neural information processing; (4) The architectures of CLIP and METER drive distinct BNs: CLIP's independent branches show modality-specific specialization, whereas METER's cross-modal design yields unified cross-modal activation, highlighting the architecture's influence on ANN brain-like properties. These results provide compelling evidence for brain-like hierarchical processing in VLMs at the neuronal level.

Paperid: 5943, https://arxiv.org/pdf/2510.15446.pdf

Abstract:
In autonomous driving, dynamic environment and corner cases pose significant challenges to the robustness of ego vehicle's state understanding and decision making. We introduce VDRive, a novel pipeline for end-to-end autonomous driving that explicitly models state-action mapping to address these challenges, enabling interpretable and robust decision making. By leveraging the advancement of the state understanding of the Vision Language Action Model (VLA) with generative diffusion policy-based action head, our VDRive guides the driving contextually and geometrically. Contextually, VLA predicts future observations through token generation pre-training, where the observations are represented as discrete codes by a Conditional Vector Quantized Variational Autoencoder (CVQ-VAE). Geometrically, we perform reinforcement learning fine-tuning of the VLA to predict future trajectories and actions based on current driving conditions. VLA supplies the current state tokens and predicted state tokens for the action policy head to generate hierarchical actions and trajectories. During policy training, a learned critic evaluates the actions generated by the policy and provides gradient-based feedback, forming an actor-critic framework that enables a reinforcement-based policy learning pipeline. Experiments show that our VDRive achieves state-of-the-art performance in the Bench2Drive closed-loop benchmark and nuScenes open-loop planning.

Paperid: 5944, https://arxiv.org/pdf/2510.13359.pdf

Abstract:
On large-scale e-commerce platforms with tens of millions of active monthly users, recommending visually similar products is essential for enabling users to efficiently discover items that align with their preferences. This study presents the application of a vision-language model (VLM) -- which has demonstrated strong performance in image recognition and image-text retrieval tasks -- to product recommendations on Mercari, a major consumer-to-consumer marketplace used by more than 20 million monthly users in Japan. Specifically, we fine-tuned SigLIP, a VLM employing a sigmoid-based contrastive loss, using one million product image-title pairs from Mercari collected over a three-month period, and developed an image encoder for generating item embeddings used in the recommendation system. Our evaluation comprised an offline analysis of historical interaction logs and an online A/B test in a production environment. In offline analysis, the model achieved a 9.1% improvement in nDCG@5 compared with the baseline. In the online A/B test, the click-through rate improved by 50% whereas the conversion rate improved by 14% compared with the existing model. These results demonstrate the effectiveness of VLM-based encoders for e-commerce product recommendations and provide practical insights into the development of visual similarity-based recommendation systems.

Paperid: 5945, https://arxiv.org/pdf/2510.13315.pdf

Abstract:
Large Vision-Language Models (LVLMs) have demonstrated remarkable multimodal capabilities, but they inherit the tendency to hallucinate from their underlying language models. While visual contrastive decoding has been proposed to mitigate this issue, existing methods often apply generic visual augmentations that disregard the specific context provided by the text query, limiting their effectiveness. This study introduces a novel training-free decoding strategy that addresses these limitations, featuring two key contributions. First, a self-augmentation prompting strategy that leverages the intrinsic knowledge of the model to dynamically align semantics between the query and the visual augmentation. Second, an adaptive thresholding algorithm that adaptively adjusts next token candidate size based on the output sparsity, utilizing full information from the logit distribution. Extensive experiments across four LVLMs and seven benchmarks demonstrate that the proposed decoding significantly enhances factual consistency compared to state-of-the-art decoding methods. This work highlights the importance of integrating query-dependent augmentation and entropy-aware decoding for improving effective generation of LVLMs.

Paperid: 5946, https://arxiv.org/pdf/2510.03840.pdf

Abstract:
Recent advances in image generation models have led to models that produce synthetic images that are increasingly difficult for standard AI detectors to identify, even though they often remain distinguishable by humans. To identify this discrepancy, we introduce \textbf{Mirage}, a curated dataset comprising a diverse range of AI-generated images exhibiting visible artifacts, where current state-of-the-art detection methods largely fail. Furthermore, we investigate whether Large Vision-Language Models (LVLMs), which are increasingly employed as substitutes for human judgment in various tasks, can be leveraged for explainable AI image detection. Our experiments on both Mirage and existing benchmark datasets demonstrate that while LVLMs are highly effective at detecting AI-generated images with visible artifacts, their performance declines when confronted with images lacking such cues.

Paperid: 5947, https://arxiv.org/pdf/2510.02922.pdf

Abstract:
Reliable risk assessment for carotid atheromatous disease remains a major clinical challenge, as it requires integrating diverse clinical and imaging information in a manner that is transparent and interpretable to clinicians. This study investigates the potential of state-of-the-art and recent large vision-language models (LVLMs) for multimodal carotid plaque assessment by integrating ultrasound imaging (USI) with structured clinical, demographic, laboratory, and protein biomarker data. A framework that simulates realistic diagnostic scenarios through interview-style question sequences is proposed, comparing a range of open-source LVLMs, including both general-purpose and medically tuned models. Zero-shot experiments reveal that even if they are very powerful, not all LVLMs can accurately identify imaging modality and anatomy, while all of them perform poorly in accurate risk classification. To address this limitation, LLaVa-NeXT-Vicuna is adapted to the ultrasound domain using low-rank adaptation (LoRA), resulting in substantial improvements in stroke risk stratification. The integration of multimodal tabular data in the form of text further enhances specificity and balanced accuracy, yielding competitive performance compared to prior convolutional neural network (CNN) baselines trained on the same dataset. Our findings highlight both the promise and limitations of LVLMs in ultrasound-based cardiovascular risk prediction, underscoring the importance of multimodal integration, model calibration, and domain adaptation for clinical translation.

Paperid: 5948, https://arxiv.org/pdf/2510.01236.pdf

Abstract:
Vision-Language Models (VLMs) show promise in medical image analysis, yet their capacity for structured reasoning in complex domains like dermatology is often limited by data scarcity and the high computational cost of advanced training techniques. To address these challenges, we introduce DermIQ-VLM, a VLM developed through a multi-stage, resource-efficient methodology designed to emulate a dermatologist's diagnostic process. Our primary contribution is a modified version of Grouped Relative Policy Optimization (GRPO), called GRPO++, which stabilizes the powerful but data-intensive GRPO framework. Our proposed training pipeline first employs GRPO++ for reasoning-oriented disease recognition, followed by supervised fine-tuning for conversational ability. To mitigate factual errors introduced during this step, we then align the model using Direct Preference Optimization (DPO), leveraging a Knowledge Graph-based system as a scalable proxy for expert preference. A preliminary evaluation on a curated dermatological dataset demonstrates that our proposed methodology yields notable performance gains over standard fine-tuning approaches. These findings validate the potential of our pipeline as a feasible pathway for developing specialized, reliable VLMs in resource-constrained environments.

Paperid: 5949, https://arxiv.org/pdf/2510.01185.pdf

Abstract:
Upcycling pre-trained dense models into sparse Mixture-of-Experts (MoEs) efficiently increases model capacity but often suffers from poor expert specialization due to naive weight replication. Our analysis reveals that upcycled MoEs, even with conventional regularization, exhibit low-confidence, weakly differentiated routing, hindering performance. We introduce Dirichlet-Prior Shaping Loss (DPSL), a novel router regularization technique that directly shapes routing probability distributions by matching expert assignments to a target Dirichlet prior. DPSL offers fine-grained control over expert balance and specialization, and enables encoding of inductive biases such as encouraging experts to focus on specific modalities or tasks, without requiring manual intervention; notably, DPSL is a general tool applicable to any module that outputs categorical probability distributions, extending its utility beyond MoE training. Experiments on upcycled MoE vision-language models (with Qwen2, Phi3, Llama3.2 LLM backbones) show DPSL consistently outperforms upcycling strategies and regularization techniques across standard vision-language benchmarks, addressing the critical issue of poor specialization and fostering more adaptive, higher-performing models.

Paperid: 5950, https://arxiv.org/pdf/2510.00690.pdf

Abstract:
Aligning large-scale vision-language models (VLMs) for complex reasoning via reinforcement learning is often hampered by the limitations of existing policy optimization algorithms, such as static training schedules and the rigid, uniform clipping mechanism in Proximal Policy Optimization (PPO). In this work, we introduce Adaptive Curriculum Policy Optimization (ACPO), a novel framework that addresses these challenges through a dual-component adaptive learning strategy. First, ACPO employs a dynamic curriculum that orchestrates a principled transition from a stable, near on-policy exploration phase to an efficient, off-policy exploitation phase by progressively increasing sample reuse. Second, we propose an Advantage-Aware Adaptive Clipping (AAAC) mechanism that replaces the fixed clipping hyperparameter with dynamic, sample-wise bounds modulated by the normalized advantage of each token. This allows for more granular and robust policy updates, enabling larger gradients for high-potential samples while safeguarding against destructive ones. We conduct extensive experiments on a suite of challenging multimodal reasoning benchmarks, including MathVista, LogicVista, and MMMU-Pro. Results demonstrate that ACPO consistently outperforms strong baselines such as DAPO and PAPO, achieving state-of-the-art performance, accelerated convergence, and superior training stability.

Paperid: 5951, https://arxiv.org/pdf/2510.00652.pdf

Abstract:
We introduce OTTER, a unified open-set multi-label tagging framework that harmonizes the stability of a curated, predefined category set with the adaptability of user-driven open tags. OTTER is built upon a large-scale, hierarchically organized multi-modal dataset, collected from diverse online repositories and annotated through a hybrid pipeline combining automated vision-language labeling with human refinement. By leveraging a multi-head attention architecture, OTTER jointly aligns visual and textual representations with both fixed and open-set label embeddings, enabling dynamic and semantically consistent tagging. OTTER consistently outperforms competitive baselines on two benchmark datasets: it achieves an overall F1 score of 0.81 on Otter and 0.75 on Favorite, surpassing the next-best results by margins of 0.10 and 0.02, respectively. OTTER attains near-perfect performance on open-set labels, with F1 of 0.99 on Otter and 0.97 on Favorite, while maintaining competitive accuracy on predefined labels. These results demonstrate OTTER's effectiveness in bridging closed-set consistency with open-vocabulary flexibility for multi-modal tagging applications.

Paperid: 5952, https://arxiv.org/pdf/2510.00047.pdf

Abstract:
Vision-Language Models (VLMs) often produce fluent Natural Language Explanations (NLEs) that sound convincing but may not reflect the causal factors driving predictions. This mismatch of plausibility and faithfulness poses technical and governance risks. We introduce Explanation-Driven Counterfactual Testing (EDCT), a fully automated verification procedure for a target VLM that treats the model's own explanation as a falsifiable hypothesis. Given an image-question pair, EDCT: (1) obtains the model's answer and NLE, (2) parses the NLE into testable visual concepts, (3) generates targeted counterfactual edits via generative inpainting, and (4) computes a Counterfactual Consistency Score (CCS) using LLM-assisted analysis of changes in both answers and explanations. Across 120 curated OK-VQA examples and multiple VLMs, EDCT uncovers substantial faithfulness gaps and provides regulator-aligned audit artifacts indicating when cited concepts fail causal tests.

Paperid: 5953, https://arxiv.org/pdf/2509.24528.pdf

Abstract:
3D scene understanding is fundamental for embodied AI and robotics, supporting reliable perception for interaction and navigation. Recent approaches achieve zero-shot, open-vocabulary 3D semantic mapping by assigning embedding vectors to 2D class-agnostic masks generated via vision-language models (VLMs) and projecting these into 3D. However, these methods often produce fragmented masks and inaccurate semantic assignments due to the direct use of raw masks, limiting their effectiveness in complex environments. To address this, we leverage SemanticSAM with progressive granularity refinement to generate more accurate and numerous object-level masks, mitigating the over-segmentation commonly observed in mask generation models such as vanilla SAM, and improving downstream 3D semantic segmentation. To further enhance semantic context, we employ a context-aware CLIP encoding strategy that integrates multiple contextual views of each mask using empirically determined weighting, providing much richer visual context. We evaluate our approach on multiple 3D scene understanding tasks, including 3D semantic segmentation and object retrieval from language queries, across several benchmark datasets. Experimental results demonstrate significant improvements over existing methods, highlighting the effectiveness of our approach.

Paperid: 5954, https://arxiv.org/pdf/2509.23109.pdf

Abstract:
A fundamental reason for the dominance of attention over RNNs and LSTMs in LLMs is its ability to capture long-range dependencies by modeling direct interactions between all tokens, overcoming the sequential limitations of recurrent architectures. Similarly, a key reason why today's vision language models (VLMs) hallucinate and underperform pure language models is that they rely on direct concatenation of image and text tokens with a modality-blinded positional encoding, which conveniently adopts the pretrained LLM backbone but forces unnecessary long-distance attention between semantically related tokens across modalities. This underscores the urgent need for mechanisms that efficiently enhance token locality and cross-modal alignment. In response, we propose Attention Anchor, a parameter-free framework that efficiently groups semantically similar tokens across modalities, improving cross-modal locality. By inserting text tokens near relevant visual patches, we create semantic signposts that reveal true content-based cross-modal attention scores, guiding the model to focus on the correct image regions for tasks such as VQA, MMBench and POPE. This improves answer accuracy and reduces hallucinations without disrupting the prompt's semantic flow. AttAnchor achieves improvements across 13 out of 15 different metrics and benchmarks, including up to 32% gains on reasoning tasks and up to 15% improvements on hallucination benchmarks. AttAnchor enables TinyLLaVA 1B to outperform much larger models like LLaVA 7B and QwenVL 3B on POPE with only 0.1% inference time overhead. To the best of our knowledge, this work is among the first to investigate mixed-modal token grouping, where text and image tokens are clustered jointly into shared groups rather than being grouped within a single modality or merely aligned post-hoc with additional alignment losses.

Paperid: 5955, https://arxiv.org/pdf/2509.23054.pdf

Abstract:
The scarcity of annotated data in specialized domains such as medical imaging presents significant challenges to training robust vision models. While self-supervised masked image modeling (MIM) offers a promising solution, existing approaches largely rely on random high-ratio masking, leading to inefficiency and poor semantic alignment. Moreover, region-aware variants typically depend on reconstruction heuristics or supervised signals, limiting their adaptability across tasks and modalities. We propose Mask What Matters, a controllable text-guided masking framework for self-supervised medical image analysis. By leveraging vision-language models for prompt-based region localization, our method flexibly applies differentiated masking to emphasize diagnostically relevant regions while reducing redundancy in background areas. This controllable design enables better semantic alignment, improved representation learning, and stronger cross-task generalizability. Comprehensive evaluation across multiple medical imaging modalities, including brain MRI, chest CT, and lung X-ray, shows that Mask What Matters consistently outperforms existing MIM methods (e.g., SparK), achieving gains of up to +3.1 percentage points in classification accuracy, +1.3 in box average precision (BoxAP), and +1.1 in mask average precision (MaskAP) for detection. Notably, it achieves these improvements with substantially lower overall masking ratios (e.g., 40\% vs. 70\%). This work demonstrates that controllable, text-driven masking can enable semantically aligned self-supervised learning, advancing the development of robust vision models for medical image analysis.

Paperid: 5956, https://arxiv.org/pdf/2509.22123.pdf

Abstract:
This survey examines multilingual vision-language models that process text and images across languages. We review 31 models and 21 benchmarks, spanning encoder-only and generative architectures, and identify a key tension between language neutrality (consistent cross-lingual representations) and cultural awareness (adaptation to cultural contexts). Current training methods favor neutrality through contrastive learning, while cultural awareness depends on diverse data. Two-thirds of evaluation benchmarks use translation-based approaches prioritizing semantic consistency, though recent work incorporates culturally grounded content. We find discrepancies in cross-lingual capabilities and gaps between training objectives and evaluation goals.

Paperid: 5957, https://arxiv.org/pdf/2509.21354.pdf

Abstract:
Vision-Language-Action (VLA) models promise unified robotic perception and control, yet their scalability is constrained by the quadratic cost of attention and the unbounded growth of key-value (KV) memory during long-horizon inference. While recent methods improve generalization through scaling backbone architectures, they often neglect the inference inefficiencies critical to real-time deployment. In this work, we present KV-Efficient VLA, a model-agnostic memory compression framework that addresses these limitations by introducing a lightweight, training-friendly mechanism to selectively retain high-utility context. Our method partitions the KV cache into fixed size chunks and employs a recurrent gating module to summarize and filter historical context according to learned utility scores. This design preserves recent fine-grained detail while aggressively pruning stale, low-relevance memory, all while maintaining causality. Theoretically, KV-Efficient VLA yields up to 1.21x inference speedup and 36% KV memory reduction, with minimal impact on task success. Our method integrates seamlessly into existing autoregressive and hybrid VLA stacks, enabling scalable inference without modifying training pipelines or downstream control logic.

Paperid: 5958, https://arxiv.org/pdf/2509.21243.pdf

Abstract:
Recent Vision-Language-Action (VLA) models demonstrate remarkable generalization in robotics but are restricted by their substantial size and computational cost, limiting real-world deployment. However, conventional lightweighting methods often sacrifice critical capabilities, particularly spatial reasoning. This creates a trade-off between efficiency and performance. To address this challenge, our work reuses Register Tokens, which were introduced for artifact removal in Vision Transformers but subsequently discarded. We suppose that these tokens contain essential spatial information and propose RetoVLA, a novel architecture that reuses them directly by injecting them into the Action Expert. RetoVLA maintains a lightweight structure while leveraging this repurposed spatial context to enhance reasoning. We demonstrate RetoVLA's effectiveness through a series of comprehensive experiments. On our custom-built 7-DOF robot arm, the model achieves a 17.1%p absolute improvement in success rates for complex manipulation tasks. Our results confirm that reusing Register Tokens directly enhances spatial reasoning, demonstrating that what was previously discarded as an artifact is in fact a valuable, unexplored resource for robotic intelligence. A video demonstration is available at: https://youtu.be/2CseBR-snZg

Paperid: 5959, https://arxiv.org/pdf/2509.21126.pdf

Abstract:
Online reinforcement learning in complex tasks is time-consuming, as massive interaction steps are needed to learn the optimal Q-function.Vision-language action (VLA) policies represent a promising direction for solving diverse tasks; however, their performance on low-level control remains limited, and effective deployment often requires task-specific expert demonstrations for fine-tuning. In this paper, we propose \textbf{VARL} (\textbf{V}LM as \textbf{A}ction advisor for online \textbf{R}einforcement \textbf{L}earning), a framework that leverages the domain knowledge of vision-language models (VLMs) to provide action suggestions for reinforcement learning agents. Unlike previous methods, VARL provides action suggestions rather than designing heuristic rewards, thereby guaranteeing unchanged optimality and convergence. The suggested actions increase sample diversity and ultimately improve sample efficiency, especially in sparse-reward tasks. To validate the effectiveness of VARL, we evaluate it across diverse environments and agent settings. Results show that VARL greatly improves sample efficiency without introducing significant computational overhead. These advantages make VARL a general framework for online reinforcement learning and make it feasible to directly apply reinforcement learning from scratch in real-world environments.

Paperid: 5960, https://arxiv.org/pdf/2509.19688.pdf

Abstract:
We present a method for formal safety verification of learning-based generative motion planners. Generative motion planners (GMPs) offer advantages over traditional planners, but verifying the safety and dynamic feasibility of their outputs is difficult since neural network verification (NNV) tools scale only to a few hundred neurons, while GMPs often contain millions. To preserve GMP expressiveness while enabling verification, our key insight is to imitate the GMP by stabilizing references sampled from the GMP with a small neural tracking controller and then applying NNV to the closed-loop dynamics. This yields reachable sets that rigorously certify closed-loop safety, while the controller enforces dynamic feasibility. Building on this, we construct a library of verified GMP references and deploy them online in a way that imitates the original GMP distribution whenever it is safe to do so, improving safety without retraining. We evaluate across diverse planners, including diffusion, flow matching, and vision-language models, improving safety in simulation (on ground robots and quadcopters) and on hardware (differential-drive robot).

Paperid: 5961, https://arxiv.org/pdf/2509.18839.pdf

Abstract:
This study evaluates the capabilities of Multimodal Large Language Models (LLMs) and Vision Language Models (VLMs) in the task of single-label classification of Christian Iconography. The goal was to assess whether general-purpose VLMs (CLIP and SigLIP) and LLMs, such as GPT-4o and Gemini 2.5, can interpret the Iconography, typically addressed by supervised classifiers, and evaluate their performance. Two research questions guided the analysis: (RQ1) How do multimodal LLMs perform on image classification of Christian saints? And (RQ2), how does performance vary when enriching input with contextual information or few-shot exemplars? We conducted a benchmarking study using three datasets supporting Iconclass natively: ArtDL, ICONCLASS, and Wikidata, filtered to include the top 10 most frequent classes. Models were tested under three conditions: (1) classification using class labels, (2) classification with Iconclass descriptions, and (3) few-shot learning with five exemplars. Results were compared against ResNet50 baselines fine-tuned on the same datasets. The findings show that Gemini-2.5 Pro and GPT-4o outperformed the ResNet50 baselines. Accuracy dropped significantly on the Wikidata dataset, where Siglip reached the highest accuracy score, suggesting model sensitivity to image size and metadata alignment. Enriching prompts with class descriptions generally improved zero-shot performance, while few-shot learning produced lower results, with only occasional and minimal increments in accuracy. We conclude that general-purpose multimodal LLMs are capable of classification in visually complex cultural heritage domains. These results support the application of LLMs as metadata curation tools in digital humanities workflows, suggesting future research on prompt optimization and the expansion of the study to other classification strategies and models.

Paperid: 5962, https://arxiv.org/pdf/2509.18428.pdf

Abstract:
Vision-Language-Action (VLA) models have gained popularity for learning robotic manipulation tasks that follow language instructions. State-of-the-art VLAs, such as OpenVLA and $Ï_{0}$, were trained on large-scale, manually labeled action datasets collected through teleoperation. More recent approaches, including LAPA and villa-X, introduce latent action representations that enable unsupervised pretraining on unlabeled datasets by modeling abstract visual changes between frames. Although these methods have shown strong results, their large model sizes make deployment in real-world settings challenging. In this work, we propose LAWM, a model-agnostic framework to pretrain imitation learning models in a self-supervised way, by learning latent action representations from unlabeled video data through world modeling. These videos can be sourced from robot recordings or videos of humans performing actions with everyday objects. Our framework is designed to be effective for transferring across tasks, environments, and embodiments. It outperforms models trained with ground-truth robotics actions and similar pretraining methods on the LIBERO benchmark and real-world setup, while being significantly more efficient and practical for real-world settings.

Paperid: 5963, https://arxiv.org/pdf/2509.18369.pdf

Abstract:
Grounding vision--language models in low-resource languages remains challenging, as they often produce fluent text about the wrong objects. This stems from scarce paired data, translation pivots that break alignment, and English-centric pretraining that ignores target-language semantics. We address this with a compute-aware Bengali captioning pipeline trained on LaBSE-verified EN--BN pairs and 110k bilingual-prompted synthetic images. A frozen MaxViT yields stable visual patches, a Bengali-native mBART-50 decodes, and a lightweight bridge links the modalities. Our core novelty is a tri-loss objective: Patch-Alignment Loss (PAL) aligns real and synthetic patch descriptors using decoder cross-attention, InfoNCE enforces global real--synthetic separation, and Sinkhorn-based OT ensures balanced fine-grained patch correspondence. This PAL+InfoNCE+OT synergy improves grounding, reduces spurious matches, and drives strong gains on Flickr30k-1k (BLEU-4 12.29, METEOR 27.98, BERTScore-F1 71.20) and MSCOCO-1k (BLEU-4 12.00, METEOR 28.14, BERTScore-F1 75.40), outperforming strong CE baselines and narrowing the real--synthetic centroid gap by 41%.

Paperid: 5964, https://arxiv.org/pdf/2509.16476.pdf

Abstract:
Vision-Language Models (VLMs) deliver impressive performance in understanding visual content with language instructions. However, redundancy in vision tokens results in the degenerated inference efficiency of VLMs, which hinders real-time use on edge consumer devices such as AR/VR devices. Existing efficiency methods commonly prune visual tokens using learned saliency, sparse attention schedules, or controller policies, but they often require architectural modification or access to intermediate activations. These pipelines add inference-time modules that increase compute and memory and often lead to an accuracy trade-off. Moreover, they also suffer from misalignment between the prompts and the region of interest in the images. Without human guidance, the model may focus on the wrong regions and miss small, high-frequency details when prompts or scenes change. In this paper, we propose GazeVLM, a training-free framework that uses the human eye gaze as a natural supervisory signal to allocate computation where it matters. By extracting gaze-driven regions of interest (ROIs) and optionally combining them with a low-resolution global view, GazeVLM mimics fovea-periphery perception to cut redundant visual tokens while preserving task-relevant details. We evaluate the visual question answering tasks on Qwen2.5-VL-3B/7B on the VOILA-COCO benchmark with human gaze. Quality of the answer is assessed by GPT-4o pairwise judging and a weighted score over coverage, accuracy, details, and fluency. Efficiency is measured by token counts and FLOPs. GazeVLM reduces visual tokens by up to 93.1%, total tokens by up to 59.6%, and FLOPs by 50%, while keeping better answer quality relative to full-resolution baselines. Our results show that aligning model computation with human gaze offers a simple, plug-and-play path toward efficient VLM inference on consumer devices.

Paperid: 5965, https://arxiv.org/pdf/2509.15482.pdf

Abstract:
Foundation models are increasingly developed in computational pathology (CPath) given their promise in facilitating many downstream tasks. While recent studies have evaluated task performance across models, less is known about the structure and variability of their learned representations. Here, we systematically analyze the representational spaces of six CPath foundation models using techniques popularized in computational neuroscience. The models analyzed span vision-language contrastive learning (CONCH, PLIP, KEEP) and self-distillation (UNI (v2), Virchow (v2), Prov-GigaPath) approaches. Through representational similarity analysis using H&E image patches from TCGA, we find that UNI2 and Virchow2 have the most distinct representational structures, whereas Prov-Gigapath has the highest average similarity across models. Having the same training paradigm (vision-only vs. vision-language) did not guarantee higher representational similarity. The representations of all models showed a high slide-dependence, but relatively low disease-dependence. Stain normalization decreased slide-dependence for all models by a range of 5.5% (CONCH) to 20.5% (PLIP). In terms of intrinsic dimensionality, vision-language models demonstrated relatively compact representations, compared to the more distributed representations of vision-only models. These findings highlight opportunities to improve robustness to slide-specific features, inform model ensembling strategies, and provide insights into how training paradigms shape model representations. Our framework is extendable across medical imaging domains, where probing the internal representations of foundation models can help ensure effective development and deployment.

Paperid: 5966, https://arxiv.org/pdf/2509.15211.pdf

Abstract:
Slide decks, serving as digital reports that bridge the gap between presentation slides and written documents, are a prevalent medium for conveying information in both academic and corporate settings. Their multimodal nature, combining text, images, and charts, presents challenges for retrieval-augmented generation systems, where the quality of retrieval directly impacts downstream performance. Traditional approaches to slide retrieval often involve separate indexing of modalities, which can increase complexity and lose contextual information. This paper investigates various methodologies for effective slide retrieval, including visual late-interaction embedding models like ColPali, the use of visual rerankers, and hybrid retrieval techniques that combine dense retrieval with BM25, further enhanced by textual rerankers and fusion methods like Reciprocal Rank Fusion. A novel Vision-Language Models-based captioning pipeline is also evaluated, demonstrating significantly reduced embedding storage requirements compared to visual late-interaction techniques, alongside comparable retrieval performance. Our analysis extends to the practical aspects of these methods, evaluating their runtime performance and storage demands alongside retrieval efficacy, thus offering practical guidance for the selection and development of efficient and robust slide retrieval systems for real-world applications.

Paperid: 5967, https://arxiv.org/pdf/2509.13379.pdf

Abstract:
Vision-Language Models (VLMs) have achieved remarkable progress in complex visual understanding across scientific and reasoning tasks. While performance benchmarking has advanced our understanding of these capabilities, the critical dimension of uncertainty quantification has received insufficient attention. Therefore, unlike prior conformal prediction studies that focused on limited settings, we conduct a comprehensive uncertainty benchmarking study, evaluating 16 state-of-the-art VLMs (open and closed-source) across 6 multimodal datasets with 3 distinct scoring functions. Our findings demonstrate that larger models consistently exhibit better uncertainty quantification; models that know more also know better what they don't know. More certain models achieve higher accuracy, while mathematical and reasoning tasks elicit poorer uncertainty performance across all models compared to other domains. This work establishes a foundation for reliable uncertainty evaluation in multimodal systems.

Paperid: 5968, https://arxiv.org/pdf/2509.11714.pdf

Abstract:
Objective: Lung cancer is a leading cause of cancer-related mortality worldwide, primarily due to delayed diagnosis and poor early detection. This study aims to develop a computer-aided diagnosis (CAD) system that leverages large vision-language models (VLMs) for the accurate detection and classification of pulmonary nodules in computed tomography (CT) scans. Methods: We propose an end-to-end CAD pipeline consisting of two modules: (i) a detection module (CADe) based on the Segment Anything Model 2 (SAM2), in which the standard visual prompt is replaced with a text prompt encoded by CLIP (Contrastive Language-Image Pretraining), and (ii) a diagnosis module (CADx) that calculates similarity scores between segmented nodules and radiomic features. To add clinical context, synthetic electronic medical records (EMRs) were generated using radiomic assessments by expert radiologists and combined with similarity scores for final classification. The method was tested on the publicly available LIDC-IDRI dataset (1,018 CT scans). Results: The proposed approach demonstrated strong performance in zero-shot lung nodule analysis. The CADe module achieved a Dice score of 0.92 and an IoU of 0.85 for nodule segmentation. The CADx module attained a specificity of 0.97 for malignancy classification, surpassing existing fully supervised methods. Conclusions: The integration of VLMs with radiomics and synthetic EMRs allows for accurate and clinically relevant CAD of pulmonary nodules in CT scans. The proposed system shows strong potential to enhance early lung cancer detection, increase diagnostic confidence, and improve patient management in routine clinical workflows.

Paperid: 5969, https://arxiv.org/pdf/2509.10809.pdf

Abstract:
Sparse Autoencoders (SAEs) are widely employed for mechanistic interpretability and model steering. Within this context, steering is by design performed by means of decoding altered SAE intermediate representations. This procedure essentially rewrites the original activations as a weighted sum of decoder features. In contrast to existing literature, we forward an encoder-centric alternative to model steering which demonstrates a stronger cross-modal performance. We introduce S&P Top-K, a retraining-free and computationally lightweight Selection and Projection framework that identifies Top-K encoder features aligned with a sensitive attribute or behavior, optionally aggregates them into a single control axis, and computes an orthogonal projection to be subsequently applied directly in the model's native embedding space. In vision-language models, it improves fairness metrics on CelebA and FairFace by up to 3.2 times over conventional SAE usage, and in large language models, it substantially reduces aggressiveness and sycophancy in Llama-3 8B Instruct, achieving up to 3.6 times gains over masked reconstruction. These findings suggest that encoder-centric interventions provide a general, efficient, and more effective mechanism for shaping model behavior at inference time than the traditional decoder-centric use of SAEs.

Paperid: 5970, https://arxiv.org/pdf/2509.09790.pdf

Abstract:
Learning to plan in grounded environments typically requires carefully designed reward functions or high-quality annotated demonstrations. Recent works show that pretrained foundation models, such as large language models (LLMs) and vision language models (VLMs), capture background knowledge helpful for planning, which reduces the amount of reward design and demonstrations needed for policy learning. We evaluate how well LLMs and VLMs provide feedback across symbolic, language, and continuous control environments. We consider prominent types of feedback for planning including binary feedback, preference feedback, action advising, goal advising, and delta action feedback. We also consider inference methods that impact feedback performance, including in-context learning, chain-of-thought, and access to environment dynamics. We find that foundation models can provide diverse high-quality feedback across domains. Moreover, larger and reasoning models consistently provide more accurate feedback, exhibit less bias, and benefit more from enhanced inference methods. Finally, feedback quality degrades for environments with complex dynamics or continuous state spaces and action spaces.

Paperid: 5971, https://arxiv.org/pdf/2509.09732.pdf

Abstract:
Vision language models (VLMs) excel at zero-shot visual classification, but their performance on fine-grained tasks and large hierarchical label spaces is understudied. This paper investigates whether structured, tree-based reasoning can enhance VLM performance. We introduce a framework that decomposes classification into interpretable decisions using decision trees and evaluates it on fine-grained (GTSRB) and coarse-grained (CIFAR-10) datasets. Although the model achieves 98.2% accuracy in understanding the tree knowledge, tree-based reasoning consistently underperforms standard zero-shot prompting. We also explore enhancing the tree prompts with LLM-generated classes and image descriptions to improve alignment. The added description enhances the performance of the tree-based and zero-shot methods. Our findings highlight limitations of structured reasoning in visual classification and offer insights for designing more interpretable VLM systems.

Paperid: 5972, https://arxiv.org/pdf/2509.08216.pdf

Abstract:
Computer science texts are particularly rich in both narrative content and illustrative charts, algorithms, images, annotated diagrams, etc. This study explores the extent to which vector-based multimodal retrieval, powered by vision-language models (VLMs), can improve discovery across multi-modal (text and images) content. Using over 3,600 digitized textbook pages largely from computer science textbooks and a Vision Language Model (VLM), we generate multi-vector representations capturing both textual and visual semantics. These embeddings are stored in a vector database. We issue a benchmark of 75 natural language queries and compare retrieval performance to ground truth and across four similarity (distance) measures. The study is intended to expose both the strengths and weakenesses of such an approach. We find that cosine similarity most effectively retrieves semantically and visually relevant pages. We further discuss the practicality of using a vector database and multi-modal embedding for operational information retrieval. Our paper is intended to offer design insights for discovery over digital libraries. Keywords: Vector embedding, multi-modal document retrieval, vector database benchmark, digital library discovery

Paperid: 5973, https://arxiv.org/pdf/2509.07488.pdf

Abstract:
We address vision-language-driven indoor navigation to assist visually impaired individuals in reaching a target location using images and natural language guidance. Traditional navigation systems are ineffective indoors due to the lack of precise location data. Our approach integrates vision and language models to generate step-by-step navigational instructions, enhancing accessibility and independence. We fine-tune the BLIP-2 model with Low Rank Adaptation (LoRA) on a manually annotated indoor navigation dataset. We propose an evaluation metric that refines the BERT F1 score by emphasizing directional and sequential variables, providing a more comprehensive measure of navigational performance. After applying LoRA, the model significantly improved in generating directional instructions, overcoming limitations in the original BLIP-2 model.

Paperid: 5974, https://arxiv.org/pdf/2509.07213.pdf

Abstract:
Background: Precise breast ultrasound (BUS) segmentation supports reliable measurement, quantitative analysis, and downstream classification, yet remains difficult for small or low-contrast lesions with fuzzy margins and speckle noise. Text prompts can add clinical context, but directly applying weakly localized text-image cues (e.g., CAM/CLIP-derived signals) tends to produce coarse, blob-like responses that smear boundaries unless additional mechanisms recover fine edges. Methods: We propose XBusNet, a novel dual-prompt, dual-branch multimodal model that combines image features with clinically grounded text. A global pathway based on a CLIP Vision Transformer encodes whole-image semantics conditioned on lesion size and location, while a local U-Net pathway emphasizes precise boundaries and is modulated by prompts that describe shape, margin, and Breast Imaging Reporting and Data System (BI-RADS) terms. Prompts are assembled automatically from structured metadata, requiring no manual clicks. We evaluate on the Breast Lesions USG (BLU) dataset using five-fold cross-validation. Primary metrics are Dice and Intersection over Union (IoU); we also conduct size-stratified analyses and ablations to assess the roles of the global and local paths and the text-driven modulation. Results: XBusNet achieves state-of-the-art performance on BLU, with mean Dice of 0.8765 and IoU of 0.8149, outperforming six strong baselines. Small lesions show the largest gains, with fewer missed regions and fewer spurious activations. Ablation studies show complementary contributions of global context, local boundary modeling, and prompt-based modulation. Conclusions: A dual-prompt, dual-branch multimodal design that merges global semantics with local precision yields accurate BUS segmentation masks and improves robustness for small, low-contrast lesions.

Paperid: 5975, https://arxiv.org/pdf/2509.06768.pdf

Abstract:
Autonomous robots operating in dynamic environments should identify and report anomalies. Embodying proactive mitigation improves safety and operational continuity. This paper presents a multimodal anomaly detection and mitigation system that integrates vision-language models and large language models to identify and report hazardous situations and conflicts in real-time. The proposed system enables robots to perceive, interpret, report, and if possible respond to urban and environmental anomalies through proactive detection mechanisms and automated mitigation actions. A key contribution in this paper is the integration of Hazardous and Conflict states into the robot's decision-making framework, where each anomaly type can trigger specific mitigation strategies. User studies (n = 30) demonstrated the effectiveness of the system in anomaly detection with 91.2% prediction accuracy and relatively low latency response times using edge-ai architecture.

Paperid: 5976, https://arxiv.org/pdf/2509.06535.pdf

Abstract:
We investigated the reproducibility of FairCLIP, proposed by Luo et al. (2024), for improving the group fairness of CLIP (Radford et al., 2021) by minimizing image-text similarity score disparities across sensitive groups using the Sinkhorn distance. The experimental setup of Luo et al. (2024) was reproduced to primarily investigate the research findings for FairCLIP. The model description by Luo et al. (2024) was found to differ from the original implementation. Therefore, a new implementation, A-FairCLIP, is introduced to examine specific design choices. Furthermore, FairCLIP+ is proposed to extend the FairCLIP objective to include multiple attributes. Additionally, the impact of the distance minimization on FairCLIP's fairness and performance was explored. In alignment with the original authors, CLIP was found to be biased towards certain demographics when applied to zero-shot glaucoma classification using medical scans and clinical notes from the Harvard-FairVLMed dataset. However, the experimental results on two datasets do not support their claim that FairCLIP improves the performance and fairness of CLIP. Although the regularization objective reduces Sinkhorn distances, both the official implementation and the aligned implementation, A-FairCLIP, were not found to improve performance nor fairness in zero-shot glaucoma classification.

Paperid: 5977, https://arxiv.org/pdf/2509.06011.pdf

Abstract:
Open-Vocabulary Object Detection (OVD) faces severe performance degradation when applied to UAV imagery due to the domain gap from ground-level datasets. To address this challenge, we propose a complete UAV-oriented solution that combines both dataset construction and model innovation. First, we design a refined UAV-Label Engine, which efficiently resolves annotation redundancy, inconsistency, and ambiguity, enabling the generation of largescale UAV datasets. Based on this engine, we construct two new benchmarks: UAVDE-2M, with over 2.4M instances across 1,800+ categories, and UAVCAP-15K, providing rich image-text pairs for vision-language pretraining. Second, we introduce the Cross-Attention Gated Enhancement (CAGE) module, a lightweight dual-path fusion design that integrates cross-attention, adaptive gating, and global FiLM modulation for robust textvision alignment. By embedding CAGE into the YOLO-World-v2 framework, our method achieves significant gains in both accuracy and efficiency, notably improving zero-shot detection on VisDrone by +5.3 mAP while reducing parameters and GFLOPs, and demonstrating strong cross-domain generalization on SIMD. Extensive experiments and real-world UAV deployment confirm the effectiveness and practicality of our proposed solution for UAV-based OVD

Paperid: 5978, https://arxiv.org/pdf/2509.04834.pdf

Abstract:
Understanding the complex combustion dynamics within scramjet engines is critical for advancing high-speed propulsion technologies. However, the large scale and high dimensionality of simulation-generated temporal flow field data present significant challenges for visual interpretation, feature differentiation, and cross-case comparison. In this paper, we present TemporalFlowViz, a parameter-aware visual analytics workflow and system designed to support expert-driven clustering, visualization, and interpretation of temporal flow fields from scramjet combustion simulations. Our approach leverages hundreds of simulated combustion cases with varying initial conditions, each producing time-sequenced flow field images. We use pretrained Vision Transformers to extract high-dimensional embeddings from these frames, apply dimensionality reduction and density-based clustering to uncover latent combustion modes, and construct temporal trajectories in the embedding space to track the evolution of each simulation over time. To bridge the gap between latent representations and expert reasoning, domain specialists annotate representative cluster centroids with descriptive labels. These annotations are used as contextual prompts for a vision-language model, which generates natural-language summaries for individual frames and full simulation cases. The system also supports parameter-based filtering, similarity-based case retrieval, and coordinated multi-view exploration to facilitate in-depth analysis. We demonstrate the effectiveness of TemporalFlowViz through two expert-informed case studies and expert feedback, showing TemporalFlowViz enhances hypothesis generation, supports interpretable pattern discovery, and enhances knowledge discovery in large-scale scramjet combustion analysis.

Paperid: 5979, https://arxiv.org/pdf/2509.02805.pdf

Abstract:
This paper highlights the challenge of decomposing conflict detection from conflict resolution in Vision-Language Models (VLMs) and presents potential approaches, including using a supervised metric via linear probes and group-based attention pattern analysis. We conduct a mechanistic investigation of LLaVA-OV-7B, a state-of-the-art VLM that exhibits diverse resolution behaviors when faced with conflicting multimodal inputs. Our results show that a linearly decodable conflict signal emerges in the model's intermediate layers and that attention patterns associated with conflict detection and resolution diverge at different stages of the network. These findings support the hypothesis that detection and resolution are functionally distinct mechanisms. We discuss how such decomposition enables more actionable interpretability and targeted interventions for improving model robustness in challenging multimodal settings.

Paperid: 5980, https://arxiv.org/pdf/2509.02100.pdf

Abstract:
A prevalent shortfall among current empathic AI systems is their inability to recognize when verbal expressions may not fully reflect underlying emotional states. This is because the existing datasets, used for the training of these systems, focus on surface-level emotion recognition without addressing the complex verbal-visual incongruence (mismatch) patterns useful for empathic understanding. In this paper, we present E-THER, the first Person-Centered Therapy-grounded multimodal dataset with multidimensional annotations for verbal-visual incongruence detection, enabling training of AI systems that develop genuine rather than performative empathic capabilities. The annotations included in the dataset are drawn from humanistic approach, i.e., identifying verbal-visual emotional misalignment in client-counsellor interactions - forming a framework for training and evaluating AI on empathy tasks. Additional engagement scores provide behavioral annotations for research applications. Notable gains in empathic and therapeutic conversational qualities are observed in state-of-the-art vision-language models (VLMs), such as IDEFICS and VideoLLAVA, using evaluation metrics grounded in empathic and therapeutic principles. Empirical findings indicate that our incongruence-trained models outperform general-purpose models in critical traits, such as sustaining therapeutic engagement, minimizing artificial or exaggerated linguistic patterns, and maintaining fidelity to PCT theoretical framework.

Paperid: 5981, https://arxiv.org/pdf/2509.01350.pdf

Abstract:
Effective specification-aware part retrieval within complex CAD assemblies is essential for automated design verification and downstream engineering tasks. However, directly using LLMs/VLMs to this task presents some challenges: the input sequences may exceed model token limits, and even after processing, performance remains unsatisfactory. Moreover, fine-tuning LLMs/VLMs requires significant computational resources, and for many high-performing general-use proprietary models (e.g., GPT or Gemini), fine-tuning access is not available. In this paper, we propose a novel part retrieval framework that requires no extra training, but using Error Notebooks + RAG for refined prompt engineering to help improve the existing general model's retrieval performance. The construction of Error Notebooks consists of two steps: (1) collecting historical erroneous CoTs and their incorrect answers, and (2) connecting these CoTs through reflective corrections until the correct solutions are obtained. As a result, the Error Notebooks serve as a repository of tasks along with their corrected CoTs and final answers. RAG is then employed to retrieve specification-relevant records from the Error Notebooks and incorporate them into the inference process. Another major contribution of our work is a human-in-the-loop CAD dataset, which is used to evaluate our method. In addition, the engineering value of our novel framework lies in its ability to effectively handle 3D models with lengthy, non-natural language metadata. Experiments with proprietary models, including GPT-4o and the Gemini series, show substantial gains, with GPT-4o (Omni) achieving up to a 23.4% absolute accuracy improvement on the human preference dataset. Moreover, ablation studies confirm that CoT reasoning provides benefits especially in challenging cases with higher part counts (>10).

Paperid: 5982, https://arxiv.org/pdf/2509.00598.pdf

Abstract:
The emergence of vision language models (VLMs) has bridged vision and language, enabling joint multimodal understanding beyond traditional visual-only deep learning models. However, transferring VLMs from the natural image domain to remote sensing (RS) segmentation remains challenging due to the limited category diversity in RS datasets and the domain gap between natural and RS imagery. Here, we propose a training-free framework, DGL-RSIS, that decouples visual and textual inputs, performing visual-language alignment at both the local semantic and global contextual levels through tailored strategies. Specifically, we first introduce a global-local decoupling (GLD) module, where text inputs are divided into local class nouns and global modifiers using natural language processing (NLP) techniques; image inputs are partitioned into a set of class-agnostic mask proposals via unsupervised mask proposal networks. Second, visual and textual features are aligned at local scale, through a novel context-aware cropping strategy for extracting image patches with proper boundaries and introducing RS-specific knowledge to enrich the text inputs. By matching the enhanced text features with mask-guided visual features, we enable the mask classification, supporting open-vocabulary semantic segmentation (OVSS). Third, at the global scale, we propose a Cross-Scale Grad-CAM module to refine Grad-CAM maps using contextual information from global modifiers. A subsequent mask selection module integrates pixel-level Grad-CAM activations into the mask-level segmentation output, such that accurate and interpretable alignment can be realized across global and local dimensions for referring expression segmentation (RES).

Paperid: 5983, https://arxiv.org/pdf/2509.00310.pdf

Abstract:
Robots often struggle to generalize from a single demonstration due to the lack of a transferable and interpretable spatial representation. In this work, we introduce TReF-6, a method that infers a simplified, abstracted 6DoF Task-Relevant Frame from a single trajectory. Our approach identifies an influence point purely from the trajectory geometry to define the origin for a local frame, which serves as a reference for parameterizing a Dynamic Movement Primitive (DMP). This influence point captures the task's spatial structure, extending the standard DMP formulation beyond start-goal imitation. The inferred frame is semantically grounded via a vision-language model and localized in novel scenes by Grounded-SAM, enabling functionally consistent skill generalization. We validate TReF-6 in simulation and demonstrate robustness to trajectory noise. We further deploy an end-to-end pipeline on real-world manipulation tasks, showing that TReF-6 supports one-shot imitation learning that preserves task intent across diverse object configurations.

Paperid: 5984, https://arxiv.org/pdf/2508.18989.pdf

Abstract:
As Vision Language Models (VLMs) become integral to real-world applications, understanding their demographic biases is critical. We introduce GRAS, a benchmark for uncovering demographic biases in VLMs across gender, race, age, and skin tone, offering the most diverse coverage to date. We further propose the GRAS Bias Score, an interpretable metric for quantifying bias. We benchmark five state-of-the-art VLMs and reveal concerning bias levels, with the least biased model attaining a GRAS Bias Score of only 2 out of 100. Our findings also reveal a methodological insight: evaluating bias in VLMs with visual question answering (VQA) requires considering multiple formulations of a question. Our code, data, and evaluation results are publicly available.

Paperid: 5985, https://arxiv.org/pdf/2508.18430.pdf

Abstract:
Vision-language models (VLMs) have shown significant potential for medical tasks; however, their general-purpose nature can limit specialized diagnostic accuracy, and their large size poses substantial inference costs for real-world clinical deployment. To address these challenges, we introduce CLARIFY, a Specialist-Generalist framework for dermatological visual question answering (VQA). CLARIFY combines two components: (i) a lightweight, domain-trained image classifier (the Specialist) that provides fast and highly accurate diagnostic predictions, and (ii) a powerful yet compressed conversational VLM (the Generalist) that generates natural language explanations to user queries. In our framework, the Specialist's predictions directly guide the Generalist's reasoning, focusing it on the correct diagnostic path. This synergy is further enhanced by a knowledge graph-based retrieval module, which grounds the Generalist's responses in factual dermatological knowledge, ensuring both accuracy and reliability. This hierarchical design not only reduces diagnostic errors but also significantly improves computational efficiency. Experiments on our curated multimodal dermatology dataset demonstrate that CLARIFY achieves an 18\% improvement in diagnostic accuracy over the strongest baseline, a fine-tuned, uncompressed single-line VLM, while reducing the average VRAM requirement and latency by at least 20\% and 5\%, respectively. These results indicate that a Specialist-Generalist system provides a practical and powerful paradigm for building lightweight, trustworthy, and clinically viable AI systems.

Paperid: 5986, https://arxiv.org/pdf/2508.17807.pdf

Abstract:
Vision-Language Models (VLMs) typically process a significantly larger number of visual tokens compared to text tokens due to the inherent redundancy in visual signals. Visual token pruning is a promising direction to reduce the computational cost of VLMs by eliminating redundant visual tokens. The text-visual attention score is a widely adopted criterion for visual token pruning as it reflects the relevance of visual tokens to the text input. However, many sequence models exhibit a recency bias, where tokens appearing later in the sequence exert a disproportionately large influence on the model's output. In VLMs, this bias manifests as inflated attention scores for tokens corresponding to the lower regions of the image, leading to suboptimal pruning that disproportionately retains tokens from the image bottom. In this paper, we present an extremely simple yet effective approach to alleviate the recency bias in visual token pruning. We propose a straightforward reweighting mechanism that adjusts the attention scores of visual tokens according to their spatial positions in the image. Our method, termed Position-reweighted Visual Token Pruning, is a plug-and-play solution that can be seamlessly incorporated into existing visual token pruning frameworks without any changes to the model architecture or extra training. Extensive experiments on LVLMs demonstrate that our method improves the performance of visual token pruning with minimal computational overhead.

Paperid: 5987, https://arxiv.org/pdf/2508.17334.pdf

Abstract:
We introduce MMCRICBENCH-3K, a benchmark for Visual Question Answering (VQA) on cricket scorecards, designed to evaluate large vision-language models (LVLMs) on complex numerical and cross-lingual reasoning over semi-structured tabular images. MMCRICBENCH-3K comprises 1,463 synthetically generated scorecard images from ODI, T20, and Test formats, accompanied by 1,500 English QA pairs. It includes two subsets: MMCRICBENCH-E-1.5K, featuring English scorecards, and MMCRICBENCH-H-1.5K, containing visually similar Hindi scorecards, with all questions and answers kept in English to enable controlled cross-script evaluation. The task demands reasoning over structured numerical data, multi-image context, and implicit domain knowledge. Empirical results show that even state-of-the-art LVLMs, such as GPT-4o and Qwen2.5VL, struggle on the English subset despite it being their primary training language and exhibit a further drop in performance on the Hindi subset. This reveals key limitations in structure-aware visual text understanding, numerical reasoning, and cross-lingual generalization. The dataset is publicly available via Hugging Face at https://huggingface.co/datasets/DIALab/MMCricBench, to promote LVLM research in this direction.

Paperid: 5988, https://arxiv.org/pdf/2508.17255.pdf

Abstract:
We present SEER-VAR, a novel framework for egocentric vehicle-based augmented reality (AR) that unifies semantic decomposition, Context-Aware SLAM Branches (CASB), and LLM-driven recommendation. Unlike existing systems that assume static or single-view settings, SEER-VAR dynamically separates cabin and road scenes via depth-guided vision-language grounding. Two SLAM branches track egocentric motion in each context, while a GPT-based module generates context-aware overlays such as dashboard cues and hazard alerts. To support evaluation, we introduce EgoSLAM-Drive, a real-world dataset featuring synchronized egocentric views, 6DoF ground-truth poses, and AR annotations across diverse driving scenarios. Experiments demonstrate that SEER-VAR achieves robust spatial alignment and perceptually coherent AR rendering across varied environments. As one of the first to explore LLM-based AR recommendation in egocentric driving, we address the lack of comparable systems through structured prompting and detailed user studies. Results show that SEER-VAR enhances perceived scene understanding, overlay relevance, and driver ease, providing an effective foundation for future research in this direction. Code and dataset will be made open source.

Paperid: 5989, https://arxiv.org/pdf/2508.12916.pdf

Abstract:
Humans effortlessly retrieve objects in cluttered, partially observable environments by combining visual reasoning, active viewpoint adjustment, and physical interaction-with only a single pair of eyes. In contrast, most existing robotic systems rely on carefully positioned fixed or multi-camera setups with complete scene visibility, which limits adaptability and incurs high hardware costs. We present \textbf{RoboRetriever}, a novel framework for real-world object retrieval that operates using only a \textbf{single} wrist-mounted RGB-D camera and free-form natural language instructions. RoboRetriever grounds visual observations to build and update a \textbf{dynamic hierarchical scene graph} that encodes object semantics, geometry, and inter-object relations over time. The supervisor module reasons over this memory and task instruction to infer the target object and coordinate an integrated action module combining \textbf{active perception}, \textbf{interactive perception}, and \textbf{manipulation}. To enable task-aware scene-grounded active perception, we introduce a novel visual prompting scheme that leverages large reasoning vision-language models to determine 6-DoF camera poses aligned with the semantic task goal and geometry scene context. We evaluate RoboRetriever on diverse real-world object retrieval tasks, including scenarios with human intervention, demonstrating strong adaptability and robustness in cluttered scenes with only one RGB-D camera.

Paperid: 5990, https://arxiv.org/pdf/2508.11433.pdf

Abstract:
Multimodal Large Language Models (MLLMs) with unified architectures excel across a wide range of vision-language tasks, yet aligning them with personalized image generation remains a significant challenge. Existing methods for MLLMs are frequently subject-specific, demanding a data-intensive fine-tuning process for every new subject, which limits their scalability. In this paper, we introduce MM-R1, a framework that integrates a cross-modal Chain-of-Thought (X-CoT) reasoning strategy to unlock the inherent potential of unified MLLMs for personalized image generation. Specifically, we structure personalization as an integrated visual reasoning and generation process: (1) grounding subject concepts by interpreting and understanding user-provided images and contextual cues, and (2) generating personalized images conditioned on both the extracted subject representations and user prompts. To further enhance the reasoning capability, we adopt Grouped Reward Proximal Policy Optimization (GRPO) to explicitly align the generation. Experiments demonstrate that MM-R1 unleashes the personalization capability of unified MLLMs to generate images with high subject fidelity and strong text alignment in a zero-shot manner.

Paperid: 5991, https://arxiv.org/pdf/2508.09846.pdf

Abstract:
This paper presents an object-aware whole-body bilateral teleoperation framework for wheeled humanoid loco-manipulation. This framework combines whole-body bilateral teleoperation with an online multi-stage object inertial parameter estimation module, which is the core technical contribution of this work. The multi-stage process sequentially integrates a vision-based object size estimator, an initial parameter guess generated by a large vision-language model (VLM), and a decoupled hierarchical sampling strategy. The visual size estimate and VLM prior offer a strong initial guess of the object's inertial parameters, significantly reducing the search space for sampling-based refinement and improving the overall estimation speed. A hierarchical strategy first estimates mass and center of mass, then infers inertia from object size to ensure physically feasible parameters, while a decoupled multi-hypothesis scheme enhances robustness to VLM prior errors. Our estimator operates in parallel with high-fidelity simulation and hardware, enabling real-time online updates. The estimated parameters are then used to update the wheeled humanoid's equilibrium point, allowing the operator to focus more on locomotion and manipulation. This integration improves the haptic force feedback for dynamic synchronization, enabling more dynamic whole-body teleoperation. By compensating for object dynamics using the estimated parameters, the framework also improves manipulation tracking while preserving compliant behavior. We validate the system on a customized wheeled humanoid with a robotic gripper and human-machine interface, demonstrating real-time execution of lifting, delivering, and releasing tasks with a payload weighing approximately one-third of the robot's body weight.

Paperid: 5992, https://arxiv.org/pdf/2508.08930.pdf

Abstract:
Natural head rotation is critical for believable embodied virtual agents, yet this micro-level behavior remains largely underexplored. While head-rotation prediction algorithms could, in principle, reproduce this behavior, they typically focus on visually salient stimuli and overlook the cognitive motives that guide head rotation. This yields agents that look at conspicuous objects while overlooking obstacles or task-relevant cues, diminishing realism in a virtual environment. We introduce SCORE, a Symbolic Cognitive Reasoning framework for Embodied Head Rotation, a data-agnostic framework that produces context-aware head movements without task-specific training or hand-tuned heuristics. A controlled VR study (N=20) identifies five motivational drivers of human head movements: Interest, Information Seeking, Safety, Social Schema, and Habit. SCORE encodes these drivers as symbolic predicates, perceives the scene with a Vision-Language Model (VLM), and plans head poses with a Large Language Model (LLM). The framework employs a hybrid workflow: the VLM-LLM reasoning is executed offline, after which a lightweight FastVLM performs online validation to suppress hallucinations while maintaining responsiveness to scene dynamics. The result is an agent that predicts not only where to look but also why, generalizing to unseen scenes and multi-agent crowds while retaining behavioral plausibility.

Paperid: 5993, https://arxiv.org/pdf/2508.05661.pdf

Abstract:
Visual search offers an intuitive way for customers to explore diverse product catalogs, particularly in consumer-to-consumer (C2C) marketplaces where listings are often unstructured and visually driven. This paper presents a scalable visual search system deployed in Mercari's C2C marketplace, where end-users act as buyers and sellers. We evaluate recent vision-language models for zero-shot image retrieval and compare their performance with an existing fine-tuned baseline. The system integrates real-time inference and background indexing workflows, supported by a unified embedding pipeline optimized through dimensionality reduction. Offline evaluation using user interaction logs shows that the multilingual SigLIP model outperforms other models across multiple retrieval metrics, achieving a 13.3% increase in nDCG@5 over the baseline. A one-week online A/B test in production further confirms real-world impact, with the treatment group showing substantial gains in engagement and conversion, up to a 40.9% increase in transaction rate via image search. Our findings highlight that recent zero-shot models can serve as a strong and practical baseline for production use, which enables teams to deploy effective visual search systems with minimal overhead, while retaining the flexibility to fine-tune based on future data or domain-specific needs.

Paperid: 5994, https://arxiv.org/pdf/2508.05148.pdf

Abstract:
The integration of robotics and automation into self-driving laboratories (SDLs) can introduce additional safety complexities, in addition to those that already apply to conventional research laboratories. Personal protective equipment (PPE) is an essential requirement for ensuring the safety and well-being of workers in laboratories, self-driving or otherwise. Fires are another important risk factor in chemical laboratories. In SDLs, fires that occur close to mobile robots, which use flammable lithium batteries, could have increased severity. Here, we present Chemist Eye, a distributed safety monitoring system designed to enhance situational awareness in SDLs. The system integrates multiple stations equipped with RGB, depth, and infrared cameras, designed to monitor incidents in SDLs. Chemist Eye is also designed to spot workers who have suffered a potential accident or medical emergency, PPE compliance and fire hazards. To do this, Chemist Eye uses decision-making driven by a vision-language model (VLM). Chemist Eye is designed for seamless integration, enabling real-time communication with robots. Based on the VLM recommendations, the system attempts to drive mobile robots away from potential fire locations, exits, or individuals not wearing PPE, and issues audible warnings where necessary. It also integrates with third-party messaging platforms to provide instant notifications to lab personnel. We tested Chemist Eye with real-world data from an SDL equipped with three mobile robots and found that the spotting of possible safety hazards and decision-making performances reached 97 % and 95 %, respectively.

Paperid: 5995, https://arxiv.org/pdf/2508.05021.pdf

Abstract:
Visual navigation in unknown environments based solely on natural language descriptions is a key capability for intelligent robots. In this work, we propose a navigation framework built upon off-the-shelf Visual Language Models (VLMs), enhanced with two human-inspired mechanisms: perspective-based active grounding, which dynamically adjusts the robot's viewpoint for improved visual inspection, and historical memory backtracking, which enables the system to retain and re-evaluate uncertain observations over time. Unlike existing approaches that passively rely on incidental visual inputs, our method actively optimizes perception and leverages memory to resolve ambiguity, significantly improving vision-language grounding in complex, unseen environments. Our framework operates in a zero-shot manner, achieving strong generalization to diverse and open-ended language descriptions without requiring labeled data or model fine-tuning. Experimental results on Habitat-Matterport 3D (HM3D) show that our method outperforms state-of-the-art approaches in language-driven object navigation. We further demonstrate its practicality through real-world deployment on a quadruped robot, achieving robust and effective navigation performance.

Paperid: 5996, https://arxiv.org/pdf/2508.04732.pdf

Abstract:
Text-to-Image (T2I) generation has made significant advancements with diffusion models, yet challenges persist in handling complex instructions, ensuring fine-grained content control, and maintaining deep semantic consistency. Existing T2I models often struggle with tasks like accurate text rendering, precise pose generation, or intricate compositional coherence. Concurrently, Vision-Language Models (LVLMs) have demonstrated powerful capabilities in cross-modal understanding and instruction following. We propose LumiGen, a novel LVLM-enhanced iterative framework designed to elevate T2I model performance, particularly in areas requiring fine-grained control, through a closed-loop, LVLM-driven feedback mechanism. LumiGen comprises an Intelligent Prompt Parsing & Augmentation (IPPA) module for proactive prompt enhancement and an Iterative Visual Feedback & Refinement (IVFR) module, which acts as a "visual critic" to iteratively correct and optimize generated images. Evaluated on the challenging LongBench-T2I Benchmark, LumiGen achieves a superior average score of 3.08, outperforming state-of-the-art baselines. Notably, our framework demonstrates significant improvements in critical dimensions such as text rendering and pose expression, validating the effectiveness of LVLM integration for more controllable and higher-quality image generation.

Paperid: 5997, https://arxiv.org/pdf/2508.04677.pdf

Abstract:
Prompt tuning has emerged as an efficient and effective technique for adapting vision-language models (VLMs) with low computational overhead. However, existing methods often overlook the vulnerability of prompt-tuned VLMs to weak semantic perturbations-such as subtle image or text noise-that degrade their generalization to unseen classes. To address this limitation, we propose ANPrompt, a novel prompt tuning framework designed to enhance robustness under such perturbations. ANPrompt first constructs weak noise text features by fusing original and noise-perturbed text embeddings, which are then clustered to form noise prompts. These noise prompts are integrated with learnable prompt tokens to generate anti-noise prompts, which are injected into the deeper layers of both image and text encoders. To further capture the noise-aware visual semantics, ANPrompt computes the Noise-Resistant Visual Prompt Prototype (NRVPP) by averaging the output prompt tokens from the vision encoder. Finally, ANPrompt introduces alignment, robustness, and anti-noise objectives by computing a Weak semantic noise Alignment Loss (WALoss) alongside the standard cross-entropy and sim loss. Experiments across 11 benchmarks demonstrate that ANPrompt consistently outperforms existing prompt tuning approaches, achieving superior robustness to semantic noise and improved generalization to novel categories.

Paperid: 5998, https://arxiv.org/pdf/2508.02890.pdf

Abstract:
This paper introduces VisuCraft, a novel framework designed to significantly enhance the capabilities of Large Vision-Language Models (LVLMs) in complex visual-guided creative content generation. Existing LVLMs often exhibit limitations in maintaining high visual fidelity, genuine creativity, and precise adherence to nuanced user instructions when generating long-form texts. VisuCraft addresses these challenges by integrating a multimodal structured information extractor (E) and a dynamic prompt generation module (G). The extractor distills fine-grained visual attributes from input images into a rich, structured representation, which the dynamic prompt module then combines with user instructions to create highly optimized prompts for underlying LVLMs (e.g., LLaVA, InstructBLIP). Evaluated on the self-constructed ImageStoryGen-500K dataset using VisuGen Metrics (Visual Grounding, Creativity, and Instruction Adherence), VisuCraft consistently outperforms baseline LVLMs across tasks like story generation and poetry composition. Our results demonstrate remarkable improvements, particularly in creativity and instruction adherence, validating VisuCraft's effectiveness in producing imaginative, visually grounded, and user-aligned long-form creative text. This work unlocks new potential for LVLMs in sophisticated creative AI applications.

Paperid: 5999, https://arxiv.org/pdf/2508.01921.pdf

Abstract:
Unified vision-language models (VLMs) promise to streamline computer vision pipelines by reframing multiple visual tasks such as classification, detection, and keypoint localization within a single language-driven interface. This architecture is particularly appealing in industrial inspection, where managing disjoint task-specific models introduces complexity, inefficiency, and maintenance overhead. In this paper, we critically evaluate the viability of this unified paradigm using InspectVLM, a Florence-2-based VLM trained on InspectMM, our new large-scale multimodal, multitask inspection dataset. While InspectVLM performs competitively on image-level classification and structured keypoint tasks, we find that it fails to match traditional ResNet-based models in core inspection metrics. Notably, the model exhibits brittle behavior under low prompt variability, produces degenerate outputs for fine-grained object detection, and frequently defaults to memorized language responses regardless of visual input. Our findings suggest that while language-driven unification offers conceptual elegance, current VLMs lack the visual grounding and robustness necessary for deployment in precision critical industrial inspections.

Paperid: 6000, https://arxiv.org/pdf/2508.00945.pdf

Abstract:
Vision Language Models (VLMs) face challenges in effectively coordinating diverse attention mechanisms for cross-modal embedding learning, leading to mismatched attention and suboptimal performance. We propose Consistent Cross-layer Regional Alignment (CCRA), which introduces Layer-Patch-wise Cross Attention (LPWCA) to capture fine-grained regional-semantic correlations by jointly weighting patch and layer-wise embedding, and Progressive Attention Integration (PAI) that systematically coordinates LPWCA, layer-wise, and patch-wise attention mechanisms in sequence. This progressive design ensures consistency from semantic to regional levels while preventing attention drift and maximizing individual attention benefits. Experimental results on ten diverse vision-language benchmarks demonstrate that our CCRA-enhanced LLaVA-v1.5-7B model achieves state-of-the-art performance, outperforming all baseline methods with only 3.55M additional parameters, while providing enhanced interpretability through more regionally focused and semantically aligned attention patterns.

Paperid: 6001, https://arxiv.org/pdf/2507.21637.pdf

Abstract:
Large vision-language models (LVLMs) are vulnerable to harmful input compared to their language-only backbones. We investigated this vulnerability by exploring LVLMs internal dynamics, framing their inherent safety understanding in terms of three key capabilities. Specifically, we define these capabilities as safety perception, semantic understanding, and alignment for linguistic expression, and experimentally pinpointed their primary locations within the model architecture. The results indicate that safety perception often emerges before comprehensive semantic understanding, leading to the reduction in safety. Motivated by these findings, we propose \textbf{Self-Aware Safety Augmentation (SASA)}, a technique that projects informative semantic representations from intermediate layers onto earlier safety-oriented layers. This approach leverages the model's inherent semantic understanding to enhance safety recognition without fine-tuning. Then, we employ linear probing to articulate the model's internal semantic comprehension to detect the risk before the generation process. Extensive experiments on various datasets and tasks demonstrate that SASA significantly improves the safety of LVLMs, with minimal impact on the utility.

Paperid: 6002, https://arxiv.org/pdf/2507.21450.pdf

Abstract:
Vision Language Navigation (VLN) typically requires agents to navigate to specified objects or remote regions in unknown scenes by obeying linguistic commands. Such tasks require organizing historical visual observations for linguistic grounding, which is critical for long-sequence navigational decisions. However, current agents suffer from overly detailed scene representation and ambiguous vision-language alignment, which weaken their comprehension of navigation-friendly high-level scene priors and easily lead to behaviors that violate linguistic commands. To tackle these issues, we propose a navigation policy by recursively summarizing along-the-way visual perceptions, which are adaptively aligned with commands to enhance linguistic grounding. In particular, by structurally modeling historical trajectories as compact neural grids, several Recursive Visual Imagination (RVI) techniques are proposed to motivate agents to focus on the regularity of visual transitions and semantic scene layouts, instead of dealing with misleading geometric details. Then, an Adaptive Linguistic Grounding (ALG) technique is proposed to align the learned situational memories with different linguistic components purposefully. Such fine-grained semantic matching facilitates the accurate anticipation of navigation actions and progress. Our navigation policy outperforms the state-of-the-art methods on the challenging VLN-CE and ObjectNav tasks, showing the superiority of our RVI and ALG techniques for VLN.

Paperid: 6003, https://arxiv.org/pdf/2507.21335.pdf

Abstract:
We can think of Visual Question Answering as a (multimodal) conversation between a human and an AI system. Here, we explore the sensitivity of Vision Language Models (VLMs) through the lens of cooperative principles of conversation proposed by Grice. Specifically, even when Grice's maxims of conversation are flouted, humans typically do not have much difficulty in understanding the conversation even though it requires more cognitive effort. Here, we study if VLMs are capable of handling violations to Grice's maxims in a manner that is similar to humans. Specifically, we add modifiers to human-crafted questions and analyze the response of VLMs to these modifiers. We use three state-of-the-art VLMs in our study, namely, GPT-4o, Claude-3.5-Sonnet and Gemini-1.5-Flash on questions from the VQA v2.0 dataset. Our initial results seem to indicate that the performance of VLMs consistently diminish with the addition of modifiers which indicates our approach as a promising direction to understand the limitations of VLMs.

Paperid: 6004, https://arxiv.org/pdf/2507.20509.pdf

Abstract:
With rapid advances in code generation, reasoning, and problem-solving, Large Language Models (LLMs) are increasingly applied in robotics. Most existing work focuses on high-level tasks such as task decomposition. A few studies have explored the use of LLMs in feedback controller design; however, these efforts are restricted to overly simplified systems, fixed-structure gain tuning, and lack real-world validation. To further investigate LLMs in automatic control, this work targets a key subfield: adaptive control. Inspired by the framework of model reference adaptive control (MRAC), we propose an LLM-guided adaptive compensator framework that avoids designing controllers from scratch. Instead, the LLMs are prompted using the discrepancies between an unknown system and a reference system to design a compensator that aligns the response of the unknown system with that of the reference, thereby achieving adaptivity. Experiments evaluate five methods: LLM-guided adaptive compensator, LLM-guided adaptive controller, indirect adaptive control, learning-based adaptive control, and MRAC, on soft and humanoid robots in both simulated and real-world environments. Results show that the LLM-guided adaptive compensator outperforms traditional adaptive controllers and significantly reduces reasoning complexity compared to the LLM-guided adaptive controller. The Lyapunov-based analysis and reasoning-path inspection demonstrate that the LLM-guided adaptive compensator enables a more structured design process by transforming mathematical derivation into a reasoning task, while exhibiting strong generalizability, adaptability, and robustness. This study opens a new direction for applying LLMs in the field of automatic control, offering greater deployability and practicality compared to vision-language models.

Paperid: 6005, https://arxiv.org/pdf/2507.17088.pdf

Abstract:
Vision-language models (VLMs) demonstrate impressive zero-shot and few-shot learning capabilities, making them essential for several downstream tasks. However, fine-tuning these models at scale remains challenging, particularly in federated environments where data is decentralized and non-iid across clients. Existing parameter-efficient tuning methods like LoRA (Low-Rank Adaptation) reduce computational overhead but struggle with heterogeneous client data, leading to suboptimal generalization. To address these challenges, we propose FedVLM, a federated LoRA fine-tuning framework that enables decentralized adaptation of VLMs while preserving model privacy and reducing reliance on centralized training. To further tackle data heterogeneity, we introduce personalized LoRA (pLoRA), which dynamically adapts LoRA parameters to each client's unique data distribution, significantly improving local adaptation while maintaining global model aggregation. Experiments on the RLAIF-V dataset show that pLoRA improves client-specific performance by 24.5% over standard LoRA, demonstrating superior adaptation in non-iid settings. FedVLM provides a scalable and efficient solution for fine-tuning VLMs in federated settings, advancing personalized adaptation in distributed learning scenarios.

Paperid: 6006, https://arxiv.org/pdf/2507.16856.pdf

Abstract:
As vision-language models (VLMs) are increasingly deployed in real-world applications, new safety risks arise from the subtle interplay between images and text. In particular, seemingly innocuous inputs can combine to reveal harmful intent, leading to unsafe model responses. Despite increasing attention to multimodal safety, previous approaches based on post hoc filtering or static refusal prompts struggle to detect such latent risks, especially when harmfulness emerges only from the combination of inputs. We propose SIA (Safety via Intent Awareness), a training-free prompt engineering framework that proactively detects and mitigates harmful intent in multimodal inputs. SIA employs a three-stage reasoning process: (1) visual abstraction via captioning, (2) intent inference through few-shot chain-of-thought prompting, and (3) intent-conditioned response refinement. Rather than relying on predefined rules or classifiers, SIA dynamically adapts to the implicit intent inferred from the image-text pair. Through extensive experiments on safety-critical benchmarks including SIUO, MM-SafetyBench, and HoliSafe, we demonstrate that SIA achieves substantial safety improvements, outperforming prior methods. Although SIA shows a minor reduction in general reasoning accuracy on MMStar, the corresponding safety gains highlight the value of intent-aware reasoning in aligning VLMs with human-centric values.

Paperid: 6007, https://arxiv.org/pdf/2507.16713.pdf

Abstract:
Vision-language models (VLMs) have been widely adopted in robotics to enable autonomous planning. However, grounding VLMs, originally trained on internet data, to diverse real-world robots remains a challenge. This paper presents ExpTeach, a framework that grounds VLMs to physical robots by building a self-generated memory of real-world experiences. In ExpTeach, the VLM autonomously plans actions, verifies outcomes, reflects on failures, and adapts robot behaviors in a closed loop. The self-generated experiences during this process are then summarized into a long-term memory, enabling retrieval of learned knowledge to guide future tasks via retrieval-augmented generation (RAG). Additionally, ExpTeach enhances the spatial understanding of VLMs with an on-demand image annotation module. In experiments, we show that reflection improves success rates from 36% to 84% on four challenging robotic tasks and observe the emergence of intelligent object interactions, including creative tool use. Across extensive tests on 12 real-world scenarios (including eight unseen ones), we find that grounding with long-term memory boosts single-trial success rates from 22% to 80%, demonstrating the effectiveness and generalizability of ExpTeach.

Paperid: 6008, https://arxiv.org/pdf/2507.12414.pdf

Abstract:
Training of autonomous driving systems requires extensive datasets with precise annotations to attain robust performance. Human annotations suffer from imperfections, and multiple iterations are often needed to produce high-quality datasets. However, manually reviewing large datasets is laborious and expensive. In this paper, we introduce AutoVDC (Automated Vision Data Cleaning) framework and investigate the utilization of Vision-Language Models (VLMs) to automatically identify erroneous annotations in vision datasets, thereby enabling users to eliminate these errors and enhance data quality. We validate our approach using the KITTI and nuImages datasets, which contain object detection benchmarks for autonomous driving. To test the effectiveness of AutoVDC, we create dataset variants with intentionally injected erroneous annotations and observe the error detection rate of our approach. Additionally, we compare the detection rates using different VLMs and explore the impact of VLM fine-tuning on our pipeline. The results demonstrate our method's high performance in error detection and data cleaning experiments, indicating its potential to significantly improve the reliability and accuracy of large-scale production datasets in autonomous driving.

Paperid: 6009, https://arxiv.org/pdf/2507.11114.pdf

Abstract:
We present a robust ensemble-based system for multilingual multimodal reasoning, designed for the ImageCLEF 2025 EXAMS V challenge. Our approach integrates Gemini 2.5 Flash for visual description, Gemini 1.5 Pro for caption refinement and consistency checks, and Gemini 2.5 Pro as a reasoner which handles final answer selection, all coordinated through carefully engineered few-shot and zero-shot prompts. We conducted an extensive ablation study, training several large language models (Gemini 2.5 Flash, Phi 4, Gemma 3, Mistral) on an English dataset and its multilingual augmented version. Additionally, we evaluated Gemini 2.5 Flash in a zero-shot setting for comparison and found it to substantially outperform the trained models. Prompt design also proved critical: enforcing concise, language-normalized formats and prohibiting explanatory text boosted model accuracy on the English validation set from 55.9% to 61.7%. On the official leaderboard, our system (Team MSA) achieved first place overall in the multilingual track with 81.4% accuracy, and led 11 out of 13 individual language tracks, with top results such as 95.07% for Croatian and 92.12% for Italian. These findings highlight that lightweight OCR-VLM ensembles, when paired with precise prompt strategies and cross-lingual augmentation, can outperform heavier end-to-end models in high-stakes, multilingual educational settings.

Paperid: 6010, https://arxiv.org/pdf/2507.11079.pdf

Abstract:
In multiple unmanned ground vehicle confrontations, autonomously evolving multi-agent tactical decisions from situational awareness remain a significant challenge. Traditional handcraft rule-based methods become vulnerable in the complicated and transient battlefield environment, and current reinforcement learning methods mainly focus on action manipulation instead of strategic decisions due to lack of interpretability. Here, we propose a vision-language model-based commander to address the issue of intelligent perception-to-decision reasoning in autonomous confrontations. Our method integrates a vision language model for scene understanding and a lightweight large language model for strategic reasoning, achieving unified perception and decision within a shared semantic space, with strong adaptability and interpretability. Unlike rule-based search and reinforcement learning methods, the combination of the two modules establishes a full-chain process, reflecting the cognitive process of human commanders. Simulation and ablation experiments validate that the proposed approach achieves a win rate of over 80% compared with baseline models.

Paperid: 6011, https://arxiv.org/pdf/2507.07297.pdf

Abstract:
Recent advances in large vision-language models have led to impressive performance in visual question answering and multimodal reasoning. However, it remains unclear whether these models genuinely perform grounded visual reasoning or rely on superficial patterns and dataset biases. In this work, we introduce MagiC, a comprehensive benchmark designed to evaluate grounded multimodal cognition, assessing not only answer accuracy but also the quality of step-by-step reasoning and its alignment with relevant visual evidence. Our benchmark includes approximately 5,500 weakly supervised QA examples generated from strong model outputs and 900 human-curated examples with fine-grained annotations, including answers, rationales, and bounding box groundings. We evaluate 15 vision-language models ranging from 7B to 70B parameters across four dimensions: final answer correctness, reasoning validity, grounding fidelity, and self-correction ability. MagiC further includes diagnostic settings to probe model robustness under adversarial visual cues and assess their capacity for introspective error correction. We introduce new metrics such as MagiScore and StepSense, and provide comprehensive analyses that reveal key limitations and opportunities in current approaches to grounded visual reasoning.

Paperid: 6012, https://arxiv.org/pdf/2507.01843.pdf

Abstract:
Mixture-of-Experts (MoE) approaches have recently gained traction in robotics applications due to their ability to dynamically allocate computational resources and specialize sub-networks for distinct tasks or environmental contexts, enabling more efficient decision-making. Such systems often comprise sparsely activated experts combined under a single monolithic architecture and require a well-configured internal routing mechanism, which does not allow for selective low-level expert and router customization and requires additional training. We propose MoIRA, an architecture-agnostic modular MoE framework designed to coordinate existing experts with an external text-based router. MoIRA incorporates two zero-shot routing options: embedding-based similarity and prompt-driven language model inference. In our experiments, we choose large Vision-Language-Action models, gr00t-N1 and $Ï_0$, as the underlying experts, and train low-rank adapters for low-overhead inference. We evaluate MoIRA on various GR1 Humanoid tasks and LIBERO Spatial and Goal benchmarks, where it consistently outperforms generalist models and competes with other MoE pipelines. Additionally, we analyse the robustness of the proposed approach to the variations of the instructions. While relying solely on textual descriptions of tasks and experts, MoIRA demonstrates the practical viability of modular deployment with precise, low-effort routing and provides an alternative, scalable foundation for future multi-expert robotic systems.

Paperid: 6013, https://arxiv.org/pdf/2507.00990.pdf

Abstract:
This work introduces Robots Imitating Generated Videos (RIGVid), a system that enables robots to perform complex manipulation tasks--such as pouring, wiping, and mixing--purely by imitating AI-generated videos, without requiring any physical demonstrations or robot-specific training. Given a language command and an initial scene image, a video diffusion model generates potential demonstration videos, and a vision-language model (VLM) automatically filters out results that do not follow the command. A 6D pose tracker then extracts object trajectories from the video, and the trajectories are retargeted to the robot in an embodiment-agnostic fashion. Through extensive real-world evaluations, we show that filtered generated videos are as effective as real demonstrations, and that performance improves with generation quality. We also show that relying on generated videos outperforms more compact alternatives such as keypoint prediction using VLMs, and that strong 6D pose tracking outperforms other ways to extract trajectories, such as dense feature point tracking. These findings suggest that videos produced by a state-of-the-art off-the-shelf model can offer an effective source of supervision for robotic manipulation.

Paperid: 6014, https://arxiv.org/pdf/2506.22494.pdf

Abstract:
This paper introduces a new framework, DriveBLIP2, built upon the BLIP2-OPT architecture, to generate accurate and contextually relevant explanations for emerging driving scenarios. While existing vision-language models perform well in general tasks, they encounter difficulties in understanding complex, multi-object environments, particularly in real-time applications such as autonomous driving, where the rapid identification of key objects is crucial. To address this limitation, an Attention Map Generator is proposed to highlight significant objects relevant to driving decisions within critical video frames. By directing the model's focus to these key regions, the generated attention map helps produce clear and relevant explanations, enabling drivers to better understand the vehicle's decision-making process in critical situations. Evaluations on the DRAMA dataset reveal significant improvements in explanation quality, as indicated by higher BLEU, ROUGE, CIDEr, and SPICE scores compared to baseline models. These findings underscore the potential of targeted attention mechanisms in vision-language models for enhancing explainability in real-time autonomous driving.

Paperid: 6015, https://arxiv.org/pdf/2506.21860.pdf

Abstract:
Mobile robots rely on object detectors for perception and object localization in indoor environments. However, standard closed-set methods struggle to handle the diverse objects and dynamic conditions encountered in real homes and labs. Open-vocabulary object detection (OVOD), driven by Vision Language Models (VLMs), extends beyond fixed labels but still struggles with domain shifts in indoor environments. We introduce a Source-Free Domain Adaptation (SFDA) approach that adapts a pre-trained model without accessing source data. We refine pseudo labels via temporal clustering, employ multi-scale threshold fusion, and apply a Mean Teacher framework with contrastive learning. Our Embodied Domain Adaptation for Object Detection (EDAOD) benchmark evaluates adaptation under sequential changes in lighting, layout, and object diversity. Our experiments show significant gains in zero-shot detection performance and flexible adaptation to dynamic indoor conditions.

Paperid: 6016, https://arxiv.org/pdf/2506.19300.pdf

Abstract:
Open-Vocabulary Camouflaged Object Segmentation (OVCOS) seeks to segment and classify camouflaged objects from arbitrary categories, presenting unique challenges due to visual ambiguity and unseen categories.Recent approaches typically adopt a two-stage paradigm: first segmenting objects, then classifying the segmented regions using Vision Language Models (VLMs).However, these methods (1) suffer from a domain gap caused by the mismatch between VLMs' full-image training and cropped-region inference, and (2) depend on generic segmentation models optimized for well-delineated objects, making them less effective for camouflaged objects.Without explicit guidance, generic segmentation models often overlook subtle boundaries, leading to imprecise segmentation.In this paper,we introduce a novel VLM-guided cascaded framework to address these issues in OVCOS.For segmentation, we leverage the Segment Anything Model (SAM), guided by the VLM.Our framework uses VLM-derived features as explicit prompts to SAM, effectively directing attention to camouflaged regions and significantly improving localization accuracy.For classification, we avoid the domain gap introduced by hard cropping.Instead, we treat the segmentation output as a soft spatial prior via the alpha channel, which retains the full image context while providing precise spatial guidance, leading to more accurate and context-aware classification of camouflaged objects.The same VLM is shared across both segmentation and classification to ensure efficiency and semantic consistency.Extensive experiments on both OVCOS and conventional camouflaged object segmentation benchmarks demonstrate the clear superiority of our method, highlighting the effectiveness of leveraging rich VLM semantics for both segmentation and classification of camouflaged objects.

Paperid: 6017, https://arxiv.org/pdf/2506.16623.pdf

Abstract:
Object Goal Navigation (ObjectNav) challenges robots to find objects in unseen environments, demanding sophisticated reasoning. While Vision-Language Models (VLMs) show potential, current ObjectNav methods often employ them superficially, primarily using vision-language embeddings for object-scene similarity checks rather than leveraging deeper reasoning. This limits contextual understanding and leads to practical issues like repetitive navigation behaviors. This paper introduces a novel zero-shot ObjectNav framework that pioneers the use of dynamic, history-aware prompting to more deeply integrate VLM reasoning into frontier-based exploration. Our core innovation lies in providing the VLM with action history context, enabling it to generate semantic guidance scores for navigation actions while actively avoiding decision loops. We also introduce a VLM-assisted waypoint generation mechanism for refining the final approach to detected objects. Evaluated on the HM3D dataset within Habitat, our approach achieves a 46% Success Rate (SR) and 24.8% Success weighted by Path Length (SPL). These results are comparable to state-of-the-art zero-shot methods, demonstrating the significant potential of our history-augmented VLM prompting strategy for more robust and context-aware robotic navigation.

Paperid: 6018, https://arxiv.org/pdf/2506.15854.pdf

Abstract:
Connected and Autonomous Vehicles (CAVs) rely on a range of devices that often process privacy-sensitive data. Among these, roadside units play a critical role particularly through the use of AI-equipped (AIE) cameras for applications such as violation detection. However, the privacy risks associated with captured imagery remain a major concern, as such data can be misused for identity theft, profiling, or unauthorized commercial purposes. While traditional techniques such as face blurring and obfuscation have been applied to mitigate privacy risks, individual privacy remains at risk, as individuals can still be tracked using other features such as their clothing. This paper introduces a novel privacy-preserving framework that leverages feedback-based reinforcement learning (RL) and vision-language models (VLMs) to protect sensitive visual information captured by AIE cameras. The main idea is to convert images into semantically equivalent textual descriptions, ensuring that scene-relevant information is retained while visual privacy is preserved. A hierarchical RL strategy is employed to iteratively refine the generated text, enhancing both semantic accuracy and privacy. Evaluation results demonstrate significant improvements in both privacy protection and textual quality, with the Unique Word Count increasing by approximately 77\% and Detail Density by around 50\% compared to existing approaches.

Paperid: 6019, https://arxiv.org/pdf/2506.15096.pdf

Abstract:
We present DyNaVLM, an end-to-end vision-language navigation framework using Vision-Language Models (VLM). In contrast to prior methods constrained by fixed angular or distance intervals, our system empowers agents to freely select navigation targets via visual-language reasoning. At its core lies a self-refining graph memory that 1) stores object locations as executable topological relations, 2) enables cross-robot memory sharing through distributed graph updates, and 3) enhances VLM's decision-making via retrieval augmentation. Operating without task-specific training or fine-tuning, DyNaVLM demonstrates high performance on GOAT and ObjectNav benchmarks. Real-world tests further validate its robustness and generalization. The system's three innovations: dynamic action space formulation, collaborative graph memory, and training-free deployment, establish a new paradigm for scalable embodied robot, bridging the gap between discrete VLN tasks and continuous real-world navigation.

Paperid: 6020, https://arxiv.org/pdf/2506.14451.pdf

Abstract:
Recent advancements in vision-language systems have improved the accuracy of Radiological Visual Question Answering (VQA) Models. However, some challenges remain across each stage of model development: limited expert-labeled images hinders data procurement at scale; the intricate and nuanced patterns of radiological images make modeling inherently difficult; and the lack of evaluation evaluation efforts makes it difficult to identify cases where the model might be ill-conditioned. In this study, we fine-tune a lightweight 3B parameter vision-language model for Radiological VQA, demonstrating that small models, when appropriately tuned with curated data, can achieve robust performance across both open- and closed-ended questions. We propose a cost-effective training pipeline from synthetic question-answer pair generation to multi-stage fine-tuning on specialised radiological domain-targeted datasets (e.g., ROCO v2.0, MedPix v2.0). Our results show that despite operating at a fraction of the scale of state-of-the-art models such as LLaVA-Med, our model achieves promising performance given its small parameter size and the limited scale of training data. We introduce a lightweight saliency-based diagnostic tool that enables domain experts to inspect VQA model performance and identify ill-conditioned failure modes through saliency analysis.

Paperid: 6021, https://arxiv.org/pdf/2506.11162.pdf

Abstract:
Understanding human social behavior such as recognizing emotions and the social dynamics causing them is an important and challenging problem. While LLMs have made remarkable advances, they are limited to the textual domain and cannot account for the major role that non-verbal cues play in understanding social situations. Vision Language Models (VLMs) can potentially account for this gap, however their ability to make correct inferences over such social cues has received little attention. In this paper, we explore the capabilities of VLMs at social reasoning. We identify a previously overlooked limitation in VLMs: the Visual Social-Pragmatic Inference gap. To target this gap, we propose a new task for VLMs: Visual Social-Pragmatic Inference. We construct a high quality dataset to test the abilities of a VLM for this task and benchmark the performance of several VLMs on it.

Paperid: 6022, https://arxiv.org/pdf/2506.07936.pdf

Abstract:
Vision-language models (VLMs) are widely assumed to exhibit in-context learning (ICL), a property similar to that of their language-only counterparts. While recent work suggests VLMs can perform multimodal ICL (MM-ICL), studies show they often rely on shallow heuristics -- such as copying or majority voting -- rather than true task understanding. We revisit this assumption by evaluating VLMs under distribution shifts, where support examples come from a dataset different from the query. Surprisingly, performance often degrades with more demonstrations, and models tend to copy answers rather than learn from them. To investigate further, we propose a new MM-ICL with Reasoning pipeline that augments each demonstration with a generated rationale alongside the answer. We conduct extensive and comprehensive experiments on both perception- and reasoning-required datasets with open-source VLMs ranging from 3B to 72B and proprietary models such as Gemini 2.0. We conduct controlled studies varying shot count, retrieval method, rationale quality, and distribution. Our results show limited performance sensitivity across these factors, suggesting that current VLMs do not effectively utilize demonstration-level information as intended in MM-ICL.

Paperid: 6023, https://arxiv.org/pdf/2506.05883.pdf

Abstract:
We present HaoMo Vision-Language Model (HMVLM), an end-to-end driving framework that implements the slow branch of a cognitively inspired fast-slow architecture. A fast controller outputs low-level steering, throttle, and brake commands, while a slow planner-a large vision-language model-generates high-level intents such as "yield to pedestrian" or "merge after the truck" without compromising latency. HMVLM introduces three upgrades: (1) selective five-view prompting with an embedded 4s history of ego kinematics, (2) multi-stage chain-of-thought (CoT) prompting that enforces a Scene Understanding -> Driving Decision -> Trajectory Inference reasoning flow, and (3) spline-based trajectory post-processing that removes late-stage jitter and sharp turns. Trained on the Waymo Open Dataset, these upgrades enable HMVLM to achieve a Rater Feedback Score (RFS) of 7.7367, securing 2nd place in the 2025 Waymo Vision-based End-to-End (E2E) Driving Challenge and surpassing the public baseline by 2.77%.

Paperid: 6024, https://arxiv.org/pdf/2506.03511.pdf

Abstract:
With over 1,000,000 images from more than 10,000 exposures using state-of-the-art high-contrast imagers (e.g., Gemini Planet Imager, VLT/SPHERE) in the search for exoplanets, can artificial intelligence (AI) serve as a transformative tool in imaging Earth-like exoplanets in the coming decade? In this paper, we introduce a benchmark and explore this question from a polarimetric image representation learning perspective. Despite extensive investments over the past decade, only a few new exoplanets have been directly imaged. Existing imaging approaches rely heavily on labor-intensive labeling of reference stars, which serve as background to extract circumstellar objects (disks or exoplanets) around target stars. With our POLARIS (POlarized Light dAta for total intensity Representation learning of direct Imaging of exoplanetary Systems) dataset, we classify reference star and circumstellar disk images using the full public SPHERE/IRDIS polarized-light archive since 2014, requiring less than 10 percent manual labeling. We evaluate a range of models including statistical, generative, and large vision-language models and provide baseline performance. We also propose an unsupervised generative representation learning framework that integrates these models, achieving superior performance and enhanced representational power. To our knowledge, this is the first uniformly reduced, high-quality exoplanet imaging dataset, rare in astrophysics and machine learning. By releasing this dataset and baselines, we aim to equip astrophysicists with new tools and engage data scientists in advancing direct exoplanet imaging, catalyzing major interdisciplinary breakthroughs.

Paperid: 6025, https://arxiv.org/pdf/2506.01902.pdf

Abstract:
Vision-language models pre-trained on large scale of unlabeled biomedical images and associated reports learn generalizable semantic representations. These multi-modal representations can benefit various downstream tasks in the biomedical domain. Contrastive learning is widely used to pre-train vision-language models for general natural images and associated captions. Despite its popularity, we found biomedical texts have complex and domain-specific semantics that are often neglected by common contrastive methods. To address this issue, we propose a novel method, perturbed report discrimination, for pre-train biomedical vision-language models. First, we curate a set of text perturbation methods that keep the same words, but disrupt the semantic structure of the sentence. Next, we apply different types of perturbation to reports, and use the model to distinguish the original report from the perturbed ones given the associated image. Parallel to this, we enhance the sensitivity of our method to higher level of granularity for both modalities by contrasting attention-weighted image sub-regions and sub-words in the image-text pairs. We conduct extensive experiments on multiple downstream tasks, and our method outperforms strong baseline methods. The results demonstrate that our approach learns more semantic meaningful and robust multi-modal representations.

Paperid: 6026, https://arxiv.org/pdf/2506.01085.pdf

Abstract:
Instruction tuning has been central to the success of recent vision-language models (VLMs), but it remains expensive-requiring large-scale datasets, high-quality annotations, and large compute budgets. We propose PRioritized cOncept learninG via Relative Error-driven Sample Selection (PROGRESS), a data- and compute-efficient framework that enables VLMs to dynamically select what to learn next based on their evolving needs during training. At each stage, the model tracks its learning progress across skills and selects the most informative samples-those it has not already mastered and that are not too difficult to learn at the current stage of training. This strategy effectively controls skill acquisition and the order in which skills are learned. Specifically, we sample from skills showing the highest learning progress, prioritizing those with the most rapid improvement. Unlike prior methods, PROGRESS requires no upfront answer annotations, queries answers only on a need basis, avoids reliance on additional supervision from auxiliary VLMs, and does not require compute-heavy gradient computations for data selection. Experiments across multiple instruction-tuning datasets of varying scales demonstrate that PROGRESS consistently outperforms state-of-the-art baselines with much less data and supervision. Additionally, we show strong cross-architecture generalization and transferability to larger models, validating PROGRESS as a scalable solution for efficient learning.

Paperid: 6027, https://arxiv.org/pdf/2505.24257.pdf

Abstract:
An embodied AI assistant operating on egocentric video must integrate spatial cues across time - for instance, determining where an object A, glimpsed a few moments ago lies relative to an object B encountered later. We introduce Disjoint-3DQA , a generative QA benchmark that evaluates this ability of VLMs by posing questions about object pairs that are not co-visible in the same frame. We evaluated seven state-of-the-art VLMs and found that models lag behind human performance by 28%, with steeper declines in accuracy (60% to 30 %) as the temporal gap widens. Our analysis further reveals that providing trajectories or bird's-eye-view projections to VLMs results in only marginal improvements, whereas providing oracle 3D coordinates leads to a substantial 20% performance increase. This highlights a core bottleneck of multi-frame VLMs in constructing and maintaining 3D scene representations over time from visual signals. Disjoint-3DQA therefore sets a clear, measurable challenge for long-horizon spatial reasoning and aims to catalyze future research at the intersection of vision, language, and embodied AI.

Paperid: 6028, https://arxiv.org/pdf/2505.22305.pdf

Abstract:
We present IKIWISI ("I Know It When I See It"), an interactive visual pattern generator for assessing vision-language models in video object recognition when ground truth is unavailable. IKIWISI transforms model outputs into a binary heatmap where green cells indicate object presence and red cells indicate object absence. This visualization leverages humans' innate pattern recognition abilities to evaluate model reliability. IKIWISI introduces "spy objects": adversarial instances users know are absent, to discern models hallucinating on nonexistent items. The tool functions as a cognitive audit mechanism, surfacing mismatches between human and machine perception by visualizing where models diverge from human understanding. Our study with 15 participants found that users considered IKIWISI easy to use, made assessments that correlated with objective metrics when available, and reached informed conclusions by examining only a small fraction of heatmap cells. This approach not only complements traditional evaluation methods through visual assessment of model behavior with custom object sets, but also reveals opportunities for improving alignment between human perception and machine understanding in vision-language systems.

Paperid: 6029, https://arxiv.org/pdf/2505.21771.pdf

Abstract:
Multimodal tables those that integrate semi structured data with visual elements such as charts and maps are ubiquitous across real world domains, yet they pose a formidable challenge to current vision language models (VLMs). While Large Language models (LLMs) and VLMs have demonstrated strong capabilities in text and image understanding, their performance on complex, real world multimodal table reasoning remains unexplored. To bridge this gap, we introduce MMTBENCH (Multimodal Table Benchmark), a benchmark consisting of 500 real world multimodal tables drawn from diverse real world sources, with a total of 4021 question answer pairs. MMTBENCH questions cover four question types (Explicit, Implicit, Answer Mention, and Visual Based), five reasoning types (Mathematical, Extrema Identification, Fact Verification, Vision Based, and Others), and eight table types (Single/Multiple Entity, Maps and Charts with Entities, Single/Multiple Charts, Maps, and Visualizations). Extensive evaluation of state of the art models on all types reveals substantial performance gaps, particularly on questions requiring visual-based reasoning and multi-step inference. These findings show the urgent need for improved architectures that more tightly integrate vision and language processing. By providing a challenging, high-quality resource that mirrors the complexity of real-world tasks, MMTBENCH underscores its value as a resource for future research on multimodal tables.

Paperid: 6030, https://arxiv.org/pdf/2505.21594.pdf

Abstract:
Large Language Models (LLMs) enable various applications on edge devices such as smartphones, wearables, and embodied robots. However, their deployment often depends on expensive cloud-based APIs, creating high operational costs, which limit access for smaller organizations and raise sustainability concerns. Certain LLMs can be deployed on-device, offering a cost-effective solution with reduced latency and improved privacy. Yet, limited computing resources constrain the size and accuracy of models that can be deployed, necessitating a collaborative design between edge and cloud. We propose a fast and cost-effective speculative edge-cloud decoding framework with a large target model on the server and a small draft model on the device. By introducing early exits in the target model, tokens are generated mid-verification, allowing the client to preemptively draft subsequent tokens before final verification, thus utilizing idle time and enhancing parallelism between edge and cloud. Using an NVIDIA Jetson Nano (client) and an A100 GPU (server) with Vicuna-68M (draft) and Llama2-7B (target) models, our method achieves up to a 35% reduction in latency compared to cloud-based autoregressive decoding, with an additional 11% improvement from preemptive drafting. To demonstrate real-world applicability, we deploy our method on the Unitree Go2 quadruped robot using Vision-Language Model (VLM) based control, achieving a 21% speedup over traditional cloud-based autoregressive decoding. These results demonstrate the potential of our framework for real-time LLM and VLM applications on resource-constrained edge devices.

Paperid: 6031, https://arxiv.org/pdf/2505.20503.pdf

Abstract:
Rapid advancements in foundation models, including Large Language Models, Vision-Language Models, Multimodal Large Language Models, and Vision-Language-Action Models have opened new avenues for embodied AI in mobile service robotics. By combining foundation models with the principles of embodied AI, where intelligent systems perceive, reason, and act through physical interactions, robots can improve understanding, adapt to, and execute complex tasks in dynamic real-world environments. However, embodied AI in mobile service robots continues to face key challenges, including multimodal sensor fusion, real-time decision-making under uncertainty, task generalization, and effective human-robot interactions (HRI). In this paper, we present the first systematic review of the integration of foundation models in mobile service robotics, identifying key open challenges in embodied AI and examining how foundation models can address them. Namely, we explore the role of such models in enabling real-time sensor fusion, language-conditioned control, and adaptive task execution. Furthermore, we discuss real-world applications in the domestic assistance, healthcare, and service automation sectors, demonstrating the transformative impact of foundation models on service robotics. We also include potential future research directions, emphasizing the need for predictive scaling laws, autonomous long-term adaptation, and cross-embodiment generalization to enable scalable, efficient, and robust deployment of foundation models in human-centric robotic systems.

Paperid: 6032, https://arxiv.org/pdf/2505.20424.pdf

Abstract:
Operating home appliances, among the most common tools in every household, is a critical capability for assistive home robots. This paper presents ApBot, a robot system that operates novel household appliances by "reading" their user manuals. ApBot faces multiple challenges: (i) infer goal-conditioned partial policies from their unstructured, textual descriptions in a user manual document, (ii) ground the policies to the appliance in the physical world, and (iii) execute the policies reliably over potentially many steps, despite compounding errors. To tackle these challenges, ApBot constructs a structured, symbolic model of an appliance from its manual, with the help of a large vision-language model (VLM). It grounds the symbolic actions visually to control panel elements. Finally, ApBot closes the loop by updating the model based on visual feedback. Our experiments show that across a wide range of simulated and real-world appliances, ApBot achieves consistent and statistically significant improvements in task success rate, compared with state-of-the-art large VLMs used directly as control policies. These results suggest that a structured internal representations plays an important role in robust robot operation of home appliances, especially, complex ones.

Paperid: 6033, https://arxiv.org/pdf/2505.19406.pdf

Abstract:
While large language models (LLMs) demonstrate strong reasoning capabilities utilizing reinforcement learning (RL) with verifiable reward, whether large vision-language models (VLMs) can directly inherit such capabilities through similar post-training strategies remains underexplored. In this work, we conduct a systematic compositional probing study to evaluate whether current VLMs trained with RL or other post-training strategies can compose capabilities across modalities or tasks under out-of-distribution conditions. We design a suite of diagnostic tasks that train models on unimodal tasks or isolated reasoning skills, and evaluate them on multimodal, compositional variants requiring skill integration. Through comparisons between supervised fine-tuning (SFT) and RL-trained models, we identify three key findings: (1) RL-trained models consistently outperform SFT on compositional generalization, demonstrating better integration of learned skills; (2) although VLMs achieve strong performance on individual tasks, they struggle to generalize compositionally under cross-modal and cross-task scenario, revealing a significant gap in current training strategies; (3) enforcing models to explicitly describe visual content before reasoning (e.g., caption-before-thinking), along with rewarding progressive vision-to-text grounding, yields notable gains. It highlights two essential ingredients for improving compositionality in VLMs: visual-to-text alignment and accurate visual grounding. Our findings shed light on the current limitations of RL-based reasoning VLM training and provide actionable insights toward building models that reason compositionally across modalities and tasks.

Paperid: 6034, https://arxiv.org/pdf/2505.18792.pdf

Abstract:
Humans learn how and when to apply forces in the world via a complex physiological and psychological learning process. Attempting to replicate this in vision-language models (VLMs) presents two challenges: VLMs can produce harmful behavior, which is particularly dangerous for VLM-controlled robots which interact with the world, but imposing behavioral safeguards can limit their functional and ethical extents. We conduct two case studies on safeguarding VLMs which generate forceful robotic motion, finding that safeguards reduce both harmful and helpful behavior involving contact-rich manipulation of human body parts. Then, we discuss the key implication of this result--that value alignment may impede desirable robot capabilities--for model evaluation and robot learning.

Paperid: 6035, https://arxiv.org/pdf/2505.15401.pdf

Abstract:
The extraction of visual features is an essential step in Visual Question Answering (VQA). Building a good visual representation of the analyzed scene is indeed one of the essential keys for the system to be able to correctly understand the latter in order to answer complex questions. In many fields such as remote sensing, the visual feature extraction step could benefit significantly from leveraging different image modalities carrying complementary spectral, spatial and contextual information. In this work, we propose to add multiple image modalities to VQA in the particular context of remote sensing, leading to a novel task for the computer vision community. To this end, we introduce a new VQA dataset, named TAMMI (Text and Multi-Modal Imagery) with diverse questions on scenes described by three different modalities (very high resolution RGB, multi-spectral imaging data and synthetic aperture radar). Thanks to an automated pipeline, this dataset can be easily extended according to experimental needs. We also propose the MM-RSVQA (Multi-modal Multi-resolution Remote Sensing Visual Question Answering) model, based on VisualBERT, a vision-language transformer, to effectively combine the multiple image modalities and text through a trainable fusion process. A preliminary experimental study shows promising results of our methodology on this challenging dataset, with an accuracy of 65.56% on the targeted VQA task. This pioneering work paves the way for the community to a new multi-modal multi-resolution VQA task that can be applied in other imaging domains (such as medical imaging) where multi-modality can enrich the visual representation of a scene. The dataset and code are available at https://tammi.sylvainlobry.com/.

Paperid: 6036, https://arxiv.org/pdf/2505.14366.pdf

Abstract:
We present a conceptual framework for training Vision-Language Models (VLMs) to perform Visual Perspective Taking (VPT), a core capability for embodied cognition essential for Human-Robot Interaction (HRI). As a first step toward this goal, we introduce a synthetic dataset, generated in NVIDIA Omniverse, that enables supervised learning for spatial reasoning tasks. Each instance includes an RGB image, a natural language description, and a ground-truth 4X4 transformation matrix representing object pose. We focus on inferring Z-axis distance as a foundational skill, with future extensions targeting full 6 Degrees Of Freedom (DOFs) reasoning. The dataset is publicly available to support further research. This work serves as a foundational step toward embodied AI systems capable of spatial understanding in interactive human-robot scenarios.

Paperid: 6037, https://arxiv.org/pdf/2505.13860.pdf

Abstract:
Vision Language Models (VLMs) have demonstrated strong performance in multi-modal tasks by effectively aligning visual and textual representations. However, most video understanding VLM research has been domain-agnostic, leaving the understanding of their transfer learning capability to specialized domains under-explored. In this work, we address this by exploring the adaptability of open-source VLMs to specific domains, and focusing on soccer as an initial case study. Our approach uses large-scale soccer datasets and LLM to create instruction-following data, and use them to iteratively fine-tune the general-domain VLM in a curriculum learning fashion (first teaching the model key soccer concepts to then question answering tasks). The final adapted model, trained using a curated dataset of 20k video clips, exhibits significant improvement in soccer-specific tasks compared to the base model, with a 37.5% relative improvement for the visual question-answering task and an accuracy improvement from 11.8% to 63.5% for the downstream soccer action classification task.

Paperid: 6038, https://arxiv.org/pdf/2505.10118.pdf

Abstract:
Existing visual token pruning methods target prompt alignment and visual preservation with static strategies, overlooking the varying relative importance of these objectives across tasks, which leads to inconsistent performance. To address this, we derive the first closed-form error bound for visual token pruning based on the Hausdorff distance, uniformly characterizing the contributions of both objectives. Moreover, leveraging $Îµ$-covering theory, we reveal an intrinsic trade-off between these objectives and quantify their optimal attainment levels under a fixed budget. To practically handle this trade-off, we propose Multi-Objective Balanced Covering (MoB), which reformulates visual token pruning as a bi-objective covering problem. In this framework, the attainment trade-off reduces to budget allocation via greedy radius trading. MoB offers a provable performance bound and linear scalability with respect to the number of input visual tokens, enabling adaptation to challenging pruning scenarios. Extensive experiments show that MoB preserves 96.4% of performance for LLaVA-1.5-7B using only 11.1% of the original visual tokens and accelerates LLaVA-Next-7B by 1.3-1.5$\times$ with negligible performance loss. Additionally, evaluations on Qwen2-VL and Video-LLaVA confirm that MoB integrates seamlessly into advanced MLLMs and diverse vision-language tasks.

Paperid: 6039, https://arxiv.org/pdf/2505.07251.pdf

Abstract:
Large vision language models (LVLMs) achieve remarkable performance through Vision In-context Learning (VICL), a process that depends significantly on demonstrations retrieved from an extensive collection of annotated examples (retrieval database). Existing studies often assume that the retrieval database contains annotated examples for all labels. However, in real-world scenarios, delays in database updates or incomplete data annotation may result in the retrieval database containing labeled samples for only a subset of classes. We refer to this phenomenon as an \textbf{incomplete retrieval database} and define the in-context learning under this condition as \textbf{Incomplete In-context Learning (IICL)}. To address this challenge, we propose \textbf{Iterative Judgments and Integrated Prediction (IJIP)}, a two-stage framework designed to mitigate the limitations of IICL. The Iterative Judgments Stage reformulates an $\boldsymbol{m}$-class classification problem into a series of $\boldsymbol{m}$ binary classification tasks, effectively converting the IICL setting into a standard VICL scenario. The Integrated Prediction Stage further refines the classification process by leveraging both the input image and the predictions from the Iterative Judgments Stage to enhance overall classification accuracy. IJIP demonstrates considerable performance across two LVLMs and two datasets under three distinct conditions of label incompleteness, achieving the highest accuracy of 93.9\%. Notably, even in scenarios where labels are fully available, IJIP still achieves the best performance of all six baselines. Furthermore, IJIP can be directly applied to \textbf{Prompt Learning} and is adaptable to the \textbf{text domain}.

Paperid: 6040, https://arxiv.org/pdf/2505.03181.pdf

Abstract:
Recent research looks to harness the general knowledge and reasoning of large language models (LLMs) into agents that accomplish user-specified goals in interactive environments. Vision-language models (VLMs) extend LLMs to multi-modal data and provide agents with the visual reasoning necessary for new applications in areas such as computer automation. However, agent tasks emphasize skills where accessible open-weight VLMs lag behind their LLM equivalents. For example, VLMs are less capable of following an environment's strict output syntax requirements and are more focused on open-ended question answering. Overcoming these limitations requires supervised fine-tuning (SFT) on task-specific expert demonstrations. Our work approaches these challenges from an offline-to-online reinforcement learning (RL) perspective. RL lets us fine-tune VLMs to agent tasks while learning from the unsuccessful decisions of our own model or more capable (larger) models. We explore an off-policy RL solution that retains the stability and simplicity of the widely used SFT workflow while allowing our agent to self-improve and learn from low-quality datasets. We demonstrate this technique with two open-weight VLMs across three multi-modal agent domains.

Paperid: 6041, https://arxiv.org/pdf/2505.01578.pdf

Abstract:
A person's demonstration often serves as a key reference for others learning the same task. However, RGB video, the dominant medium for representing these demonstrations, often fails to capture fine-grained contextual cues such as intent, safety-critical environmental factors, and subtle preferences embedded in human behavior. This sensory gap fundamentally limits the ability of Vision Language Models (VLMs) to reason about why actions occur and how they should adapt to individual users. To address this, we introduce MICA (Multimodal Interactive Contextualized Assistance), a framework that improves conversational agents for task assistance by integrating eye gaze and speech cues. MICA segments demonstrations into meaningful sub-tasks and extracts keyframes and captions that capture fine-grained intent and user-specific cues, enabling richer contextual grounding for visual question answering. Evaluations on questions derived from real-time chat-assisted task replication show that multimodal cues significantly improve response quality over frame-based retrieval. Notably, gaze cues alone achieves 93% of speech performance, and their combination yields the highest accuracy. Task type determines the effectiveness of implicit (gaze) vs. explicit (speech) cues, underscoring the need for adaptable multimodal models. These results highlight the limitations of frame-based context and demonstrate the value of multimodal signals for real-world AI task assistance.

Paperid: 6042, https://arxiv.org/pdf/2504.21681.pdf

Abstract:
Most pre-trained Vision-Language (VL) models and training data for the downstream tasks are only available in English. Therefore, multilingual VL tasks are solved using cross-lingual transfer: fine-tune a multilingual pre-trained model or transfer the text encoder using parallel data. We study the alternative approach: transferring an already trained encoder using parallel data. We investigate the effect of parallel data: domain and the number of languages, which were out of focus in previous work. Our results show that even machine-translated task data are the best on average, caption-like authentic parallel data outperformed it in some languages. Further, we show that most languages benefit from multilingual training.

Paperid: 6043, https://arxiv.org/pdf/2504.17282.pdf

Abstract:
Agents that can autonomously navigate the web through a graphical user interface (GUI) using a unified action space (e.g., mouse and keyboard actions) can require very large amounts of domain-specific expert demonstrations to achieve good performance. Low sample efficiency is often exacerbated in sparse-reward and large-action-space environments, such as a web GUI, where only a few actions are relevant in any given situation. In this work, we consider the low-data regime, with limited or no access to expert behavior. To enable sample-efficient learning, we explore the effect of constraining the action space through $\textit{intent-based affordances}$ -- i.e., considering in any situation only the subset of actions that achieve a desired outcome. We propose $\textbf{Code as Generative Affordances}$ $(\textbf{$\texttt{CoGA}$})$, a method that leverages pre-trained vision-language models (VLMs) to generate code that determines affordable actions through implicit intent-completion functions and using a fully-automated program generation and verification pipeline. These programs are then used in-the-loop of a reinforcement learning agent to return a set of affordances given a pixel observation. By greatly reducing the number of actions that an agent must consider, we demonstrate on a wide range of tasks in the MiniWob++ benchmark that: $\textbf{1)}$ $\texttt{CoGA}$ is orders of magnitude more sample efficient than its RL agent, $\textbf{2)}$ $\texttt{CoGA}$'s programs can generalize within a family of tasks, and $\textbf{3)}$ $\texttt{CoGA}$ performs better or on par compared with behavior cloning when a small number of expert demonstrations is available.

Paperid: 6044, https://arxiv.org/pdf/2504.14359.pdf

Abstract:
There are many ways to describe, name, and group objects when captioning an image. Differences are evident when speakers come from diverse cultures due to the unique experiences that shape perception. Machine translation of captions has pushed multilingual capabilities in vision-language models (VLMs), but data comes mainly from English speakers, indicating a perceptual bias and lack of model flexibility. In this work, we address this challenge and outline a data-efficient framework to instill multilingual VLMs with greater understanding of perceptual diversity. We specifically propose an LLM-based, multimodal recaptioning strategy that alters the object descriptions of English captions before translation. The greatest benefits are demonstrated in a targeted multimodal mechanism guided by native speaker data. By adding produced rewrites as augmentations in training, we improve on German and Japanese text-image retrieval cases studies (up to +3.5 mean recall overall, +4.7 on non-native error cases). We further propose a mechanism to analyze the specific object description differences across datasets, and we offer insights into cross-dataset and cross-language generalization.

Paperid: 6045, https://arxiv.org/pdf/2504.13231.pdf

Abstract:
Rapid information access is vital during wildfires, yet traditional data sources are slow and costly. Social media offers real-time updates, but extracting relevant insights remains a challenge. In this work, we focus on multimodal wildfire social media data, which, although existing in current datasets, is currently underrepresented in Canadian contexts. We present WildFireCan-MMD, a new multimodal dataset of X posts from recent Canadian wildfires, annotated across twelve key themes. We evaluate zero-shot vision-language models on this dataset and compare their results with those of custom-trained and baseline classifiers. We show that while baseline methods and zero-shot prompting offer quick deployment, custom-trained models outperform them when labelled data is available. Our best-performing custom model reaches 84.48% f-score, outperforming VLMs and baseline classifiers. We also demonstrate how this model can be used to uncover trends during wildfires, through the collection and analysis of a large unlabeled dataset. Our dataset facilitates future research in wildfire response, and our findings highlight the importance of tailored datasets and task-specific training. Importantly, such datasets should be localized, as disaster response requirements vary across regions and contexts.

Paperid: 6046, https://arxiv.org/pdf/2504.13069.pdf

Abstract:
Alt-text is essential for mobile app accessibility, yet UI icons often lack meaningful descriptions, limiting accessibility for screen reader users. Existing approaches either require extensive labeled datasets, struggle with partial UI contexts, or operate post-development, increasing technical debt. We first conduct a formative study to determine when and how developers prefer to generate icon alt-text. We then explore the ALTICON approach for generating alt-text for UI icons during development using two fine-tuned models: a text-only large language model that processes extracted UI metadata and a multi-modal model that jointly analyzes icon images and textual context. To improve accuracy, the method extracts relevant UI information from the DOM tree, retrieves in-icon text via OCR, and applies structured prompts for alt-text generation. Our empirical evaluation with the most closely related deep-learning and vision-language models shows that ALTICON generates alt-text that is of higher quality while not requiring a full-screen input.

Paperid: 6047, https://arxiv.org/pdf/2504.10117.pdf

Abstract:
Open-world 3D semantic occupancy prediction aims to generate a voxelized 3D representation from sensor inputs while recognizing both known and unknown objects. Transferring open-vocabulary knowledge from vision-language models (VLMs) offers a promising direction but remains challenging. However, methods based on VLM-derived 2D pseudo-labels with traditional supervision are limited by a predefined label space and lack general prediction capabilities. Direct alignment with pretrained image embeddings, on the other hand, fails to achieve reliable performance due to often inconsistent image and text representations in VLMs. To address these challenges, we propose AGO, a novel 3D occupancy prediction framework with adaptive grounding to handle diverse open-world scenarios. AGO first encodes surrounding images and class prompts into 3D and text embeddings, respectively, leveraging similarity-based grounding training with 3D pseudo-labels. Additionally, a modality adapter maps 3D embeddings into a space aligned with VLM-derived image embeddings, reducing modality gaps. Experiments on Occ3D-nuScenes show that AGO improves unknown object prediction in zero-shot and few-shot transfer while achieving state-of-the-art closed-world self-supervised performance, surpassing prior methods by 4.09 mIoU.

Paperid: 6048, https://arxiv.org/pdf/2504.09249.pdf

Abstract:
Understanding and reasoning over academic handwritten notes remains a challenge in document AI, particularly for mathematical equations, diagrams, and scientific notations. Existing visual question answering (VQA) benchmarks focus on printed or structured handwritten text, limiting generalization to real-world note-taking. To address this, we introduce NoTeS-Bank, an evaluation benchmark for Neural Transcription and Search in note-based question answering. NoTeS-Bank comprises complex notes across multiple domains, requiring models to process unstructured and multimodal content. The benchmark defines two tasks: (1) Evidence-Based VQA, where models retrieve localized answers with bounding-box evidence, and (2) Open-Domain VQA, where models classify the domain before retrieving relevant documents and answers. Unlike classical Document VQA datasets relying on optical character recognition (OCR) and structured data, NoTeS-BANK demands vision-language fusion, retrieval, and multimodal reasoning. We benchmark state-of-the-art Vision-Language Models (VLMs) and retrieval frameworks, exposing structured transcription and reasoning limitations. NoTeS-Bank provides a rigorous evaluation with NDCG@5, MRR, Recall@K, IoU, and ANLS, establishing a new standard for visual document understanding and reasoning.

Paperid: 6049, https://arxiv.org/pdf/2504.06838.pdf

Abstract:
Recent studies have introduced various approaches for prompt-tuning black-box vision-language models, referred to as black-box prompt-tuning (BBPT). While BBPT has demonstrated considerable potential, it is often found that many existing methods require an excessive number of queries (i.e., function evaluations), which poses a significant challenge in real-world scenarios where the number of allowed queries is limited. To tackle this issue, we propose Zeroth-order Intrinsic-dimensional Prompt-tuning (ZIP), a novel approach that enables efficient and robust prompt optimization in a purely black-box setting. The key idea of ZIP is to reduce the problem dimensionality and the variance of zeroth-order gradient estimates, such that the training is done fast with far less queries. We achieve this by re-parameterizing prompts in low-rank representations and designing intrinsic-dimensional clipping of estimated gradients. We evaluate ZIP on 13+ vision-language tasks in standard benchmarks and show that it achieves an average improvement of approximately 6% in few-shot accuracy and 48% in query efficiency compared to the best-performing alternative BBPT methods, establishing a new state of the art. Our ablation analysis further shows that the proposed clipping mechanism is robust and nearly optimal, without the need to manually select the clipping threshold, matching the result of expensive hyperparameter search.

Paperid: 6050, https://arxiv.org/pdf/2504.06661.pdf

Abstract:
Recent robotic task planning frameworks have integrated large multimodal models (LMMs) such as GPT-4o. To address grounding issues of such models, it has been suggested to split the pipeline into perceptional state grounding and subsequent state-based planning. As we show in this work, the state grounding ability of LMM-based approaches is still limited by weaknesses in granular, structured, domain-specific scene understanding. To address this shortcoming, we develop a more structured state grounding framework that features a domain-conditioned scene graph as its scene representation. We show that such representation is actionable in nature as it is directly mappable to a symbolic state in planning languages such as the Planning Domain Definition Language (PDDL). We provide an instantiation of our state grounding framework where the domain-conditioned scene graph generation is implemented with a lightweight vision-language approach that classifies domain-specific predicates on top of domain-relevant object detections. Evaluated across three domains, our approach achieves significantly higher state rounding accuracy and task planning success rates compared to LMM-based approaches.

Paperid: 6051, https://arxiv.org/pdf/2504.05477.pdf

Abstract:
Service and assistive robots are increasingly being deployed in dynamic social environments; however, ensuring transparent and explainable interactions remains a significant challenge. This paper presents a multimodal explainability module that integrates vision language models and heat maps to improve transparency during navigation. The proposed system enables robots to perceive, analyze, and articulate their observations through natural language summaries. User studies (n=30) showed a preference of majority for real-time explanations, indicating improved trust and understanding. Our experiments were validated through confusion matrix analysis to assess the level of agreement with human expectations. Our experimental and simulation results emphasize the effectiveness of explainability in autonomous navigation, enhancing trust and interpretability.

Paperid: 6052, https://arxiv.org/pdf/2504.04893.pdf

Abstract:
Typographic attacks exploit the interplay between text and visual content in multimodal foundation models, causing misclassifications when misleading text is embedded within images. However, existing datasets are limited in size and diversity, making it difficult to study such vulnerabilities. In this paper, we introduce SCAM, the largest and most diverse dataset of real-world typographic attack images to date, containing 1,162 images across hundreds of object categories and attack words. Through extensive benchmarking of Vision-Language Models (VLMs) on SCAM, we demonstrate that typographic attacks significantly degrade performance, and identify that training data and model architecture influence the susceptibility to these attacks. Our findings reveal that typographic attacks persist in state-of-the-art Large Vision-Language Models (LVLMs) due to the choice of their vision encoder, though larger Large Language Models (LLMs) backbones help mitigate their vulnerability. Additionally, we demonstrate that synthetic attacks closely resemble real-world (handwritten) attacks, validating their use in research. Our work provides a comprehensive resource and empirical insights to facilitate future research toward robust and trustworthy multimodal AI systems. We publicly release the datasets introduced in this paper along with the code for evaluations at www.bliss.berlin/research/scam.

Paperid: 6053, https://arxiv.org/pdf/2504.03153.pdf

Abstract:
We propose MORAL (a multimodal reinforcement learning framework for decision making in autonomous laboratories) that enhances sequential decision-making in autonomous robotic laboratories through the integration of visual and textual inputs. Using the BridgeData V2 dataset, we generate fine-tuned image captions with a pretrained BLIP-2 vision-language model and combine them with visual features through an early fusion strategy. The fused representations are processed using Deep Q-Network (DQN) and Proximal Policy Optimization (PPO) agents. Experimental results demonstrate that multimodal agents achieve a 20% improvement in task completion rates and significantly outperform visual-only and textual-only baselines after sufficient training. Compared to transformer-based and recurrent multimodal RL models, our approach achieves superior performance in cumulative reward and caption quality metrics (BLEU, METEOR, ROUGE-L). These results highlight the impact of semantically aligned language cues in enhancing agent learning efficiency and generalization. The proposed framework contributes to the advancement of multimodal reinforcement learning and embodied AI systems in dynamic, real-world environments.

Paperid: 6054, https://arxiv.org/pdf/2504.02349.pdf

Abstract:
Recent advances in large language and vision-language models have enabled zero-shot inference, allowing models to solve new tasks without task-specific training. Various adaptation techniques such as prompt engineering, In-Context Learning (ICL), and supervised fine-tuning can further enhance the model's performance on a downstream task, but they require substantial manual effort to construct effective prompts or labeled examples. In this work, we introduce a joint inference framework for fully unsupervised adaptation, eliminating the need for manual prompt engineering and labeled examples. Unlike zero-shot inference, which makes independent predictions, the joint inference makes predictions simultaneously for all inputs in a given task. Since direct joint inference involves computationally expensive optimization, we develop efficient approximation techniques, leading to two unsupervised adaptation methods: unsupervised fine-tuning and unsupervised ICL. We demonstrate the effectiveness of our methods across diverse tasks and models, including language-only Llama-3.1 on natural language processing tasks, reasoning-oriented Qwen2.5-Math on grade school math problems, vision-language OpenFlamingo on vision tasks, and the API-only access GPT-4o model on massive multi-discipline tasks. Our experiments demonstrate substantial improvements over the standard zero-shot approach, including 39% absolute improvement on the challenging GSM8K math reasoning dataset. Remarkably, despite being fully unsupervised, our framework often performs on par with supervised approaches that rely on ground truth labels.

Paperid: 6055, https://arxiv.org/pdf/2504.01089.pdf

Abstract:
In the United States alone accidental home deaths exceed 128,000 per year. Our work aims to enable home robots who respond to emergency scenarios in the home, preventing injuries and deaths. We introduce a new dataset of household emergencies based in the ThreeDWorld simulator. Each scenario in our dataset begins with an instantaneous or periodic sound which may or may not be an emergency. The agent must navigate the multi-room home scene using prior observations, alongside audio signals and images from the simulator, to determine if there is an emergency or not. In addition to our new dataset, we present a modular approach for localizing and identifying potential home emergencies. Underpinning our approach is a novel probabilistic dynamic scene graph (P-DSG), where our key insight is that graph nodes corresponding to agents can be represented with a probabilistic edge. This edge, when refined using Bayesian inference, enables efficient and effective localization of agents in the scene. We also utilize multi-modal vision-language models (VLMs) as a component in our approach, determining object traits (e.g. flammability) and identifying emergencies. We present a demonstration of our method completing a real-world version of our task on a consumer robot, showing the transferability of both our task and our method. Our dataset will be released to the public upon this papers publication.

Paperid: 6056, https://arxiv.org/pdf/2503.23465.pdf

Abstract:
Global localization is a critical problem in autonomous navigation, enabling precise positioning without reliance on GPS. Modern global localization techniques often depend on dense LiDAR maps, which, while precise, require extensive storage and computational resources. Recent approaches have explored alternative methods, such as sparse maps and learned features, but they suffer from poor robustness and generalization. We propose SparseLoc, a global localization framework that leverages vision-language foundation models to generate sparse, semantic-topometric maps in a zero-shot manner. It combines this map representation with a Monte Carlo localization scheme enhanced by a novel late optimization strategy, ensuring improved pose estimation. By constructing compact yet highly discriminative maps and refining localization through a carefully designed optimization schedule, SparseLoc overcomes the limitations of existing techniques, offering a more efficient and robust solution for global localization. Our system achieves over a 5X improvement in localization accuracy compared to existing sparse mapping techniques. Despite utilizing only 1/500th of the points of dense mapping methods, it achieves comparable performance, maintaining an average global localization error below 5m and 2 degrees on KITTI sequences.

Paperid: 6057, https://arxiv.org/pdf/2503.20252.pdf

Abstract:
Anomaly Detection (AD) focuses on detecting samples that differ from the standard pattern, making it a vital tool in process control. Logical anomalies may appear visually normal yet violate predefined constraints on object presence, arrangement, or quantity, depending on reasoning and explainability. We introduce LogicQA, a framework that enhances AD by providing industrial operators with explanations for logical anomalies. LogicQA compiles automatically generated questions into a checklist and collects responses to identify violations of logical constraints. LogicQA is training-free, annotation-free, and operates in a few-shot setting. We achieve state-of-the-art (SOTA) Logical AD performance on public benchmarks, MVTec LOCO AD, with an AUROC of 87.6 percent and an F1-max of 87.0 percent along with the explanations of anomalies. Also, our approach has shown outstanding performance on semiconductor SEM corporate data, further validating its effectiveness in industrial applications.

Paperid: 6058, https://arxiv.org/pdf/2503.16546.pdf

Abstract:
Deep Convolutional Neural Networks (CNNs) have significantly advanced deep learning, driving breakthroughs in computer vision, natural language processing, medical diagnosis, object detection, and speech recognition. Architectural innovations including 1D, 2D, and 3D convolutional models, dilated and grouped convolutions, depthwise separable convolutions, and attention mechanisms address domain-specific challenges and enhance feature representation and computational efficiency. Structural refinements such as spatial-channel exploitation, multi-path design, and feature-map enhancement contribute to robust hierarchical feature extraction and improved generalization, particularly through transfer learning. Efficient preprocessing strategies, including Fourier transforms, structured transforms, low-precision computation, and weight compression, optimize inference speed and facilitate deployment in resource-constrained environments. This survey presents a unified taxonomy that classifies CNN architectures based on spatial exploitation, multi-path structures, depth, width, dimensionality expansion, channel boosting, and attention mechanisms. It systematically reviews CNN applications in face recognition, pose estimation, action recognition, text classification, statistical language modeling, disease diagnosis, radiological analysis, cryptocurrency sentiment prediction, 1D data processing, video analysis, and speech recognition. In addition to consolidating architectural advancements, the review highlights emerging learning paradigms such as few-shot, zero-shot, weakly supervised, federated learning frameworks and future research directions include hybrid CNN-transformer models, vision-language integration, generative learning, etc. This review provides a comprehensive perspective on CNN's evolution from 2015 to 2025, outlining key innovations, challenges, and opportunities.

Paperid: 6059, https://arxiv.org/pdf/2503.07906.pdf

Abstract:
Image captioning has long been a pivotal task in visual understanding, with recent advancements in vision-language models (VLMs) significantly enhancing the ability to generate detailed image captions. However, the evaluation of detailed image captioning remains underexplored due to outdated evaluation metrics and coarse annotations. In this paper, we introduce DeCapBench along with a novel metric, DCScore, specifically designed for detailed captioning tasks. DCScore evaluates hallucinations and fine-grained comprehensiveness by deconstructing responses into the smallest self-sufficient units, termed primitive information units, and assessing them individually. Our evaluation shows that DCScore aligns more closely with human judgment than other rule-based or model-based metrics. Concurrently, DeCapBench exhibits a high correlation with VLM arena results on descriptive tasks, surpassing existing benchmarks for vision-language models. Additionally, we present an automatic fine-grained feedback collection method, FeedQuill, for preference optimization based on our advanced metric, showing robust generalization capabilities across auto-generated preference data. Extensive experiments on multiple VLMs demonstrate that our method not only significantly reduces hallucinations but also enhances performance across various benchmarks, achieving superior detail captioning performance while surpassing GPT-4o.

Paperid: 6060, https://arxiv.org/pdf/2503.06282.pdf

Abstract:
LiDAR-based 3D object detection datasets have been pivotal for autonomous driving, yet they cover a limited range of objects, restricting the model's generalization across diverse deployment environments. To address this, we introduce the first generalized cross-domain few-shot (GCFS) task in 3D object detection, which focuses on adapting a source-pretrained model for high performance on both common and novel classes in a target domain with few-shot samples. Our solution integrates multi-modal fusion and contrastive-enhanced prototype learning within one framework, holistically overcoming challenges related to data scarcity and domain adaptation in the GCFS setting. The multi-modal fusion module utilizes 2D vision-language models to extract rich, open-set semantic knowledge. To address biases in point distributions across varying structural complexities, we particularly introduce a physically-aware box searching strategy that leverages laser imaging principles to generate high-quality 3D box proposals from 2D insights, enhancing object recall. To effectively capture domain-specific representations for each class from limited target data, we further propose a contrastive-enhanced prototype learning, which strengthens the model's adaptability. We evaluate our approach with three GCFS benchmark settings, and extensive experiments demonstrate the effectiveness of our solution for GCFS tasks. The code will be publicly available.

Paperid: 6061, https://arxiv.org/pdf/2503.01794.pdf

Abstract:
Contrastive Language-Image Pre-Training (CLIP) has enabled zero-shot classification in radiology, reducing reliance on manual annotations. However, conventional contrastive learning struggles with normal case detection due to its strict intra-sample alignment, which disrupts normal sample clustering and leads to high false positives (FPs) and false negatives (FNs). To address these issues, we propose OFF-CLIP, a contrastive learning refinement that improves normal detection by introducing an off-diagonal term loss to enhance normal sample clustering and applying sentence-level text filtering to mitigate FNs by removing misaligned normal statements from abnormal reports. OFF-CLIP can be applied to radiology CLIP models without requiring any architectural modifications. Experimental results show that OFF-CLIP significantly improves normal classification, achieving a 0.61 Area under the curve (AUC) increase on VinDr-CXR over CARZero, the state-of-the-art zero-shot classification baseline, while maintaining or improving abnormal classification performance. Additionally, OFF-CLIP enhances zero-shot grounding by improving pointing game accuracy, confirming better anomaly localization. These results demonstrate OFF-CLIP's effectiveness as a robust and efficient enhancement for medical vision-language models.

Paperid: 6062, https://arxiv.org/pdf/2502.20277.pdf

Abstract:
Infections in Diabetic Foot Ulcers (DFUs) can cause severe complications, including tissue death and limb amputation, highlighting the need for accurate, timely diagnosis. Previous machine learning methods have focused on identifying infections by analyzing wound images alone, without utilizing additional metadata such as medical notes. In this study, we aim to improve infection detection by introducing Synthetic Caption Augmented Retrieval for Wound Infection Detection (SCARWID), a novel deep learning framework that leverages synthetic textual descriptions to augment DFU images. SCARWID consists of two components: (1) Wound-BLIP, a Vision-Language Model (VLM) fine-tuned on GPT-4o-generated descriptions to synthesize consistent captions from images; and (2) an Image-Text Fusion module that uses cross-attention to extract cross-modal embeddings from an image and its corresponding Wound-BLIP caption. Infection status is determined by retrieving the top-k similar items from a labeled support set. To enhance the diversity of training data, we utilized a latent diffusion model to generate additional wound images. As a result, SCARWID outperformed state-of-the-art models, achieving average sensitivity, specificity, and accuracy of 0.85, 0.78, and 0.81, respectively, for wound infection classification. Displaying the generated captions alongside the wound images and infection detection results enhances interpretability and trust, enabling nurses to align SCARWID outputs with their medical knowledge. This is particularly valuable when wound notes are unavailable or when assisting novice nurses who may find it difficult to identify visual attributes of wound infection.

Paperid: 6063, https://arxiv.org/pdf/2502.19293.pdf

Abstract:
Millions of melanocytic skin lesions are examined by pathologists each year, the majority of which concern common nevi (i.e., ordinary moles). While most of these lesions can be diagnosed in seconds, writing the corresponding pathology report is much more time-consuming. Automating part of the report writing could, therefore, alleviate the increasing workload of pathologists. In this work, we develop a vision-language model specifically for the pathology domain of cutaneous melanocytic lesions. The model follows the Contrastive Captioner framework and was trained and evaluated using a melanocytic lesion dataset of 42,512 H&E-stained whole slide images and 19,645 corresponding pathology reports. Our results show that the quality scores of model-generated reports were on par with pathologist-written reports for common nevi, assessed by an expert pathologist in a reader study. While report generation revealed to be more difficult for rare melanocytic lesion subtypes, the cross-modal retrieval performance for these cases was considerably better.

Paperid: 6064, https://arxiv.org/pdf/2502.19285.pdf

Abstract:
Vision-language models in pathology enable multimodal case retrieval and automated report generation. Many of the models developed so far, however, have been trained on pathology reports that include information which cannot be inferred from paired whole slide images (e.g., patient history), potentially leading to hallucinated sentences in generated reports. To this end, we investigate how the selection of information from pathology reports for vision-language modeling affects the quality of the multimodal representations and generated reports. More concretely, we compare a model trained on full reports against a model trained on preprocessed reports that only include sentences describing the cell and tissue appearances based on the H&E-stained slides. For the experiments, we built upon the BLIP-2 framework and used a cutaneous melanocytic lesion dataset of 42,433 H&E-stained whole slide images and 19,636 corresponding pathology reports. Model performance was assessed using image-to-text and text-to-image retrieval, as well as qualitative evaluation of the generated reports by an expert pathologist. Our results demonstrate that text preprocessing prevents hallucination in report generation. Despite the improvement in the quality of the generated reports, training the vision-language model on full reports showed better cross-modal retrieval performance.

Paperid: 6065, https://arxiv.org/pdf/2502.17132.pdf

Abstract:
This paper explores the advancements and applications of large-scale models in the medical field, with a particular focus on Medical Large Models (MedLMs). These models, encompassing Large Language Models (LLMs), Vision Models, 3D Large Models, and Multimodal Models, are revolutionizing healthcare by enhancing disease prediction, diagnostic assistance, personalized treatment planning, and drug discovery. The integration of graph neural networks in medical knowledge graphs and drug discovery highlights the potential of Large Graph Models (LGMs) in understanding complex biomedical relationships. The study also emphasizes the transformative role of Vision-Language Models (VLMs) and 3D Large Models in medical image analysis, anatomical modeling, and prosthetic design. Despite the challenges, these technologies are setting new benchmarks in medical innovation, improving diagnostic accuracy, and paving the way for personalized healthcare solutions. This paper aims to provide a comprehensive overview of the current state and future directions of large models in medicine, underscoring their significance in advancing global health.

Paperid: 6066, https://arxiv.org/pdf/2502.16456.pdf

Abstract:
Category-selective regions in the human brain, such as the fusiform face area (FFA), extrastriate body area (EBA), parahippocampal place area (PPA), and visual word form area (VWFA), play a crucial role in high-level visual processing. Here, we investigate whether artificial neural networks (ANNs) exhibit similar category-selective neurons and how these neurons vary across model layers and between purely visual and vision-language models. Inspired by fMRI functional localizer experiments, we presented images from different categories (faces, bodies, scenes, words, scrambled scenes, and scrambled words) to deep networks and identified category-selective neurons using statistical criteria. Comparing ResNet and the structurally controlled ResNet-based CLIP model, we found that both models contain category-selective neurons, with their proportion increasing across layers, mirroring category selectivity in higher-level visual brain regions. However, CLIP exhibited a higher proportion but lower specificity of category-selective neurons compared to ResNet. Additionally, CLIP's category-selective neurons were more evenly distributed across feature maps and demonstrated greater representational consistency across layers. These findings suggest that language learning increases the number of category-selective neurons while reducing their selectivity strength, reshaping visual representations in deep networks. Our study provides insights into how ANNs mirror biological vision and how multimodal learning influences category-selective representations.

Paperid: 6067, https://arxiv.org/pdf/2502.14780.pdf

Abstract:
Efficient and privacy-preserving multimodal interaction is essential as AR, VR, and modern smartphones with powerful cameras become primary interfaces for human-computer communication. Existing powerful large vision-language models (VLMs) enabling multimodal interaction often rely on cloud-based processing, raising significant concerns about (1) visual privacy by transmitting sensitive vision data to servers, and (2) their limited real-time, on-device usability. This paper explores Visual Instruction Rewriting, a novel approach that transforms multimodal instructions into text-only commands, allowing seamless integration of lightweight on-device instruction rewriter VLMs (250M parameters) with existing conversational AI systems, enhancing vision data privacy. To achieve this, we present a dataset of over 39,000 examples across 14 domains and develop a compact VLM, pretrained on image captioning datasets and fine-tuned for instruction rewriting. Experimental results, evaluated through NLG metrics such as BLEU, METEOR, and ROUGE, along with semantic parsing analysis, demonstrate that even a quantized version of the model (<500MB storage footprint) can achieve effective instruction rewriting, thus enabling privacy-focused, multimodal AI applications.

Paperid: 6068, https://arxiv.org/pdf/2502.11455.pdf

Abstract:
Safety alignment is critical in pre-training large language models (LLMs) to generate responses aligned with human values and refuse harmful queries. Unlike LLM, the current safety alignment of VLMs is often achieved with post-hoc safety fine-tuning. However, these methods are less effective to white-box attacks. To address this, we propose $\textit{Adversary-aware DPO (ADPO)}$, a novel training framework that explicitly considers adversarial. $\textit{Adversary-aware DPO (ADPO)}$ integrates adversarial training into DPO to enhance the safety alignment of VLMs under worst-case adversarial perturbations. $\textit{ADPO}$ introduces two key components: (1) an adversarial-trained reference model that generates human-preferred responses under worst-case perturbations, and (2) an adversarial-aware DPO loss that generates winner-loser pairs accounting for adversarial distortions. By combining these innovations, $\textit{ADPO}$ ensures that VLMs remain robust and reliable even in the presence of sophisticated jailbreak attacks. Extensive experiments demonstrate that $\textit{ADPO}$ outperforms baselines in the safety alignment and general utility of VLMs.

Paperid: 6069, https://arxiv.org/pdf/2502.10455.pdf

Abstract:
Recent studies in Large Vision-Language Models (LVLMs) have demonstrated impressive advancements in multimodal Out-of-Context (OOC) misinformation detection, discerning whether an authentic image is wrongly used in a claim. Despite their success, the textual evidence of authentic images retrieved from the inverse search is directly transmitted to LVLMs, leading to inaccurate or false information in the decision-making phase. To this end, we present E2LVLM, a novel evidence-enhanced large vision-language model by adapting textual evidence in two levels. First, motivated by the fact that textual evidence provided by external tools struggles to align with LVLMs inputs, we devise a reranking and rewriting strategy for generating coherent and contextually attuned content, thereby driving the aligned and effective behavior of LVLMs pertinent to authentic images. Second, to address the scarcity of news domain datasets with both judgment and explanation, we generate a novel OOC multimodal instruction-following dataset by prompting LVLMs with informative content to acquire plausible explanations. Further, we develop a multimodal instruction-tuning strategy with convincing explanations for beyond detection. This scheme contributes to E2LVLM for multimodal OOC misinformation detection and explanation. A multitude of experiments demonstrate that E2LVLM achieves superior performance than state-of-the-art methods, and also provides compelling rationales for judgments.

Paperid: 6070, https://arxiv.org/pdf/2502.10046.pdf

Abstract:
Creating lifelike virtual agents capable of interacting with their environments is a longstanding goal in computer graphics. This paper addresses the challenge of generating natural head rotations, a critical aspect of believable agent behavior for visual information gathering and dynamic responses to environmental cues. Although earlier methods have made significant strides, many rely on data-driven or saliency-based approaches, which often underperform in diverse settings and fail to capture deeper cognitive factors such as risk assessment, information seeking, and contextual prioritization. Consequently, generated behaviors can appear rigid or overlook critical scene elements, thereby diminishing the sense of realism. In this paper, we propose \textbf{ViRAC}, a \textbf{Vi}sion-\textbf{R}easoning \textbf{A}gent Head Movement \textbf{C}ontrol framework, which exploits the common-sense knowledge and reasoning capabilities of large-scale models, including Vision-Language Models (VLMs) and Large-Language Models (LLMs). Rather than explicitly modeling every cognitive mechanism, ViRAC leverages the biases and patterns internalized by these models from extensive training, thus emulating human-like perceptual processes without hand-tuned heuristics. Experimental results in multiple scenarios reveal that ViRAC produces more natural and context-aware head rotations than recent state-of-the-art techniques. Quantitative evaluations show a closer alignment with real human head-movement data, while user studies confirm improved realism and cognitive plausibility.

Paperid: 6071, https://arxiv.org/pdf/2502.08317.pdf

Abstract:
Spatial relation hallucinations pose a persistent challenge in large vision-language models (LVLMs), leading to generate incorrect predictions about object positions and spatial configurations within an image. To address this issue, we propose a constraint-aware prompting framework designed to reduce spatial relation hallucinations. Specifically, we introduce two types of constraints: (1) bidirectional constraint, which ensures consistency in pairwise object relations, and (2) transitivity constraint, which enforces relational dependence across multiple objects. By incorporating these constraints, LVLMs can produce more spatially coherent and consistent outputs. We evaluate our method on three widely-used spatial relation datasets, demonstrating performance improvements over existing approaches. Additionally, a systematic analysis of various bidirectional relation analysis choices and transitivity reference selections highlights greater possibilities of our methods in incorporating constraints to mitigate spatial relation hallucinations.

Paperid: 6072, https://arxiv.org/pdf/2502.08226.pdf

Abstract:
Recent advancements in Large Vision Language Models (LVLMs) have enabled the development of LVLM-based Graphical User Interface (GUI) agents under various paradigms. Training-based approaches, such as CogAgent and SeeClick, struggle with cross-dataset and cross-platform generalization due to their reliance on dataset-specific training. Generalist LVLMs, such as GPT-4V, employ Set-of-Marks (SoM) for action grounding, but obtaining SoM labels requires metadata like HTML source, which is not consistently available across platforms. Moreover, existing methods often specialize in singular GUI tasks rather than achieving comprehensive GUI understanding. To address these limitations, we introduce TRISHUL, a novel, training-free agentic framework that enhances generalist LVLMs for holistic GUI comprehension. Unlike prior works that focus on either action grounding (mapping instructions to GUI elements) or GUI referring (describing GUI elements given a location), TRISHUL seamlessly integrates both. At its core, TRISHUL employs Hierarchical Screen Parsing (HSP) and the Spatially Enhanced Element Description (SEED) module, which work synergistically to provide multi-granular, spatially, and semantically enriched representations of GUI elements. Our results demonstrate TRISHUL's superior performance in action grounding across the ScreenSpot, VisualWebBench, AITW, and Mind2Web datasets. Additionally, for GUI referring, TRISHUL surpasses the ToL agent on the ScreenPR benchmark, setting a new standard for robust and adaptable GUI comprehension.

Paperid: 6073, https://arxiv.org/pdf/2502.07837.pdf

Abstract:
Embodied intelligence seamlessly integrates vision, language, and action.~However, most multimodal robotic models rely on massive fine-tuning, incurring high time and hardware costs.~To address this, we introduce RoboBERT, an end-to-end multimodal manipulation model built around a novel two-stage training paradigm.~In the first stage, we freeze most of the vision encoder and train with a single "standard" instruction phrasing, allowing the model to focus on stable policy learning via a CNN-based diffusion policy.~In the second stage, we unfreeze all modules and inject diverse natural language variants, rapidly aligning varied instructions to the already-learned policy without destabilizing performance.~We further employ systematic data augmentations to enhance robustness against visual perturbations.~Without relying on auxiliary datasets, RoboBERT achieves new state-of-the-art (SOTA) mean episode lengths of 4.52 on the CALVIN ABCD-D benchmark and 3.79 on the ABC-D benchmark using only language-labeled expert demonstrations and a comparatively lightweight architecture.Real-robot trials on a 6-DOF manipulator confirm higher success rates than comparable methods trained on identical data.These results demonstrate that our data-augmentation-enhanced two-stage training paradigm delivers efficient, scalable, and broadly applicable performance for multimodal robotic systems.

Paperid: 6074, https://arxiv.org/pdf/2502.07306.pdf

Abstract:
In this work, we propose a modular approach for the Vision-Language Navigation (VLN) task by decomposing the problem into four sub-modules that use state-of-the-art Large Language Models (LLMs) and Vision-Language Models (VLMs) in a zero-shot setting. Given navigation instruction in natural language, we first prompt LLM to extract the landmarks and the order in which they are visited. Assuming the known model of the environment, we retrieve the top-k locations of the last landmark and generate $k$ path hypotheses from the starting location to the last landmark using the shortest path algorithm on the topological map of the environment. Each path hypothesis is represented by a sequence of panoramas. We then use dynamic programming to compute the alignment score between the sequence of panoramas and the sequence of landmark names, which match scores obtained from VLM. Finally, we compute the nDTW metric between the hypothesis that yields the highest alignment score to evaluate the path fidelity. We demonstrate superior performance compared to other approaches that use joint semantic maps like VLMaps on the complex R2R-Habitat instruction dataset and quantify in detail the effect of visual grounding on navigation performance.

Paperid: 6075, https://arxiv.org/pdf/2502.06695.pdf

Abstract:
Deep learning models frequently exploit spurious features in training data to achieve low training error, often resulting in poor generalization when faced with shifted testing distributions. To address this issue, various methods from imbalanced learning, representation learning, and classifier recalibration have been proposed to enhance the robustness of deep neural networks against spurious correlations. In this paper, we observe that models trained with empirical risk minimization tend to generalize well for examples from the majority groups while memorizing instances from minority groups. Building on recent findings that show memorization can be localized to a limited number of neurons, we apply example-tied dropout as a method we term FairDropout, aimed at redirecting this memorization to specific neurons that we subsequently drop out during inference. We empirically evaluate FairDropout using the subpopulation benchmark suite encompassing vision, language, and healthcare tasks, demonstrating that it significantly reduces reliance on spurious correlations, and outperforms state-of-the-art methods.

Paperid: 6076, https://arxiv.org/pdf/2501.19069.pdf

Abstract:
To bridge the semantic gap between vision and language (VL), it is necessary to develop a good alignment strategy, which includes handling semantic diversity, abstract representation of visual information, and generalization ability of models. Recent works use detector-based bounding boxes or patches with regular partitions to represent visual semantics. While current paradigms have made strides, they are still insufficient for fully capturing the nuanced contextual relations among various objects. This paper proposes a comprehensive visual semantic representation module, necessitating the utilization of panoptic segmentation to generate coherent fine-grained semantic features. Furthermore, we propose a novel Graph Spiking Hybrid Network (GSHN) that integrates the complementary advantages of Spiking Neural Networks (SNNs) and Graph Attention Networks (GATs) to encode visual semantic information. Intriguingly, the model not only encodes the discrete and continuous latent variables of instances but also adeptly captures both local and global contextual features, thereby significantly enhancing the richness and diversity of semantic representations. Leveraging the spatiotemporal properties inherent in SNNs, we employ contrastive learning (CL) to enhance the similarity-based representation of embeddings. This strategy alleviates the computational overhead of the model and enriches meaningful visual representations by constructing positive and negative sample pairs. We design an innovative pre-training method, Spiked Text Learning (STL), which uses text features to improve the encoding ability of discrete semantics. Experiments show that the proposed GSHN exhibits promising results on multiple VL downstream tasks.

Paperid: 6077, https://arxiv.org/pdf/2501.15040.pdf

Abstract:
Vision language model (VLM) has been designed for large scale image-text alignment as a pretrained foundation model. For downstream few shot classification tasks, parameter efficient fine-tuning (PEFT) VLM has gained much popularity in the computer vision community. PEFT methods like prompt tuning and linear adapter have been studied for fine-tuning VLM while low rank adaptation (LoRA) algorithm has rarely been considered for few shot fine-tuning VLM. The main obstacle to use LoRA for few shot fine-tuning is the catastrophic forgetting problem. Because the visual language alignment knowledge is important for the generality in few shot learning, whereas low rank adaptation interferes with the most informative direction of the pretrained weight matrix. We propose the complementary subspace low rank adaptation (Comp-LoRA) method to regularize the catastrophic forgetting problem in few shot VLM finetuning. In detail, we optimize the low rank matrix in the complementary subspace, thus preserving the general vision language alignment ability of VLM when learning the novel few shot information. We conduct comparison experiments of the proposed Comp-LoRA method and other PEFT methods on fine-tuning VLM for few shot classification. And we also present the suppression on the catastrophic forgetting problem of our proposed method against directly applying LoRA to VLM. The results show that the proposed method surpasses the baseline method by about +1.0\% Top-1 accuracy and preserves the VLM zero-shot performance over the baseline method by about +1.3\% Top-1 accuracy.

Paperid: 6078, https://arxiv.org/pdf/2501.15021.pdf

Abstract:
Vision-language models (VLMs) show remarkable performance in multimodal tasks. However, excessively long multimodal inputs lead to oversized Key-Value (KV) caches, resulting in significant memory consumption and I/O bottlenecks. Previous KV quantization methods for Large Language Models (LLMs) may alleviate these issues but overlook the attention saliency differences of multimodal tokens, resulting in suboptimal performance. In this paper, we investigate the attention-aware token saliency patterns in VLM and propose AKVQ-VL. AKVQ-VL leverages the proposed Text-Salient Attention (TSA) and Pivot-Token-Salient Attention (PSA) patterns to adaptively allocate bit budgets. Moreover, achieving extremely low-bit quantization requires effectively addressing outliers in KV tensors. AKVQ-VL utilizes the Walsh-Hadamard transform (WHT) to construct outlier-free KV caches, thereby reducing quantization difficulty. Evaluations of 2-bit quantization on 12 long-context and multimodal tasks demonstrate that AKVQ-VL maintains or even improves accuracy, outperforming LLM-oriented methods. AKVQ-VL can reduce peak memory usage by 2.13x, support up to 3.25x larger batch sizes and 2.46x throughput.

Paperid: 6079, https://arxiv.org/pdf/2501.14204.pdf

Abstract:
Vision-Language Models (VLMs) have achieved notable success in multimodal tasks but face practical limitations due to the quadratic complexity of decoder attention mechanisms and autoregressive generation. Existing methods like FASTV and VTW have achieved notable results in reducing redundant visual tokens, but these approaches focus on pruning tokens in a single forward pass without systematically analyzing the redundancy of visual tokens throughout the entire generation process. In this paper, we introduce a dynamic pruning strategy tailored for VLMs, namedDynamic Rate (DyRate), which progressively adjusts the compression rate during generation. Our analysis of the distribution of attention reveals that the importance of visual tokens decreases throughout the generation process, inspiring us to adopt a more aggressive compression rate. By integrating a lightweight predictor based on attention distribution, our approach enables flexible adjustment of pruning rates based on the attention distribution. Our experimental results demonstrate that our method not only reduces computational demands but also maintains the quality of responses.

Paperid: 6080, https://arxiv.org/pdf/2501.13904.pdf

Abstract:
Multimodal Large Language Models (LLMs) are pivotal in revolutionizing customer support and operations by integrating multiple modalities such as text, images, and audio. Federated Prompt Learning (FPL) is a recently proposed approach that combines pre-trained multimodal LLMs such as vision-language models with federated learning to create personalized, privacy-preserving AI systems. However, balancing the competing goals of personalization, generalization, and privacy remains a significant challenge. Over-personalization can lead to overfitting, reducing generalizability, while stringent privacy measures, such as differential privacy, can hinder both personalization and generalization. In this paper, we propose a Differentially Private Federated Prompt Learning (DP-FPL) approach to tackle this challenge by leveraging a low-rank factorization scheme to capture generalization while maintaining a residual term that preserves expressiveness for personalization. To ensure privacy, we introduce a novel method where we apply local differential privacy to the two low-rank components of the local prompt, and global differential privacy to the global prompt. Our approach mitigates the impact of privacy noise on the model performance while balancing the tradeoff between personalization and generalization. Extensive experiments demonstrate the effectiveness of our approach over other benchmarks.

Paperid: 6081, https://arxiv.org/pdf/2501.12751.pdf

Abstract:
Patent figure classification facilitates faceted search in patent retrieval systems, enabling efficient prior art search. Existing approaches have explored patent figure classification for only a single aspect and for aspects with a limited number of concepts. In recent years, large vision-language models (LVLMs) have shown tremendous performance across numerous computer vision downstream tasks, however, they remain unexplored for patent figure classification. Our work explores the efficacy of LVLMs in patent figure visual question answering (VQA) and classification, focusing on zero-shot and few-shot learning scenarios. For this purpose, we introduce new datasets, PatFigVQA and PatFigCLS, for fine-tuning and evaluation regarding multiple aspects of patent figures~(i.e., type, projection, patent class, and objects). For a computational-effective handling of a large number of classes using LVLM, we propose a novel tournament-style classification strategy that leverages a series of multiple-choice questions. Experimental results and comparisons of multiple classification approaches based on LVLMs and Convolutional Neural Networks (CNNs) in few-shot settings show the feasibility of the proposed approaches.

Paperid: 6082, https://arxiv.org/pdf/2501.07888.pdf

Abstract:
We introduce Tarsier2, a state-of-the-art large vision-language model (LVLM) designed for generating detailed and accurate video descriptions, while also exhibiting superior general video understanding capabilities. Tarsier2 achieves significant advancements through three key upgrades: (1) Scaling pre-training data from 11M to 40M video-text pairs, enriching both volume and diversity; (2) Performing fine-grained temporal alignment during supervised fine-tuning; (3) Using model-based sampling to automatically construct preference data and applying DPO training for optimization. Extensive experiments show that Tarsier2-7B consistently outperforms leading proprietary models, including GPT-4o and Gemini 1.5 Pro, in detailed video description tasks. On the DREAM-1K benchmark, Tarsier2-7B improves F1 by 2.8% over GPT-4o and 5.8% over Gemini-1.5-Pro. In human side-by-side evaluations, Tarsier2-7B shows a +8.6% performance advantage over GPT-4o and +24.9% over Gemini-1.5-Pro. Tarsier2-7B also sets new state-of-the-art results across 15 public benchmarks, spanning tasks such as video question-answering, video grounding, hallucination test, and embodied question-answering, demonstrating its versatility as a robust generalist vision-language model.

Paperid: 6083, https://arxiv.org/pdf/2501.07114.pdf

Abstract:
Compositional Zero-Shot Learning (CZSL) aims to enable models to recognize novel compositions of visual states and objects that were absent during training. Existing methods predominantly focus on learning semantic representations of seen compositions but often fail to disentangle the independent features of states and objects in images, thereby limiting their ability to generalize to unseen compositions. To address this challenge, we propose Duplex, a novel dual-prototype learning method that integrates semantic and visual prototypes through a carefully designed dual-branch architecture, enabling effective representation learning for compositional tasks. Duplex utilizes a Graph Neural Network (GNN) to adaptively update visual prototypes, capturing complex interactions between states and objects. Additionally, it leverages the strong visual-semantic alignment of pre-trained Vision-Language Models (VLMs) and employs a multi-path architecture combined with prompt engineering to align image and text representations, ensuring robust generalization. Extensive experiments on three benchmark datasets demonstrate that Duplex outperforms state-of-the-art methods in both closed-world and open-world settings.

Paperid: 6084, https://arxiv.org/pdf/2501.06680.pdf

Abstract:
Vision-language models (VLMs) have become a promising approach to enhancing perception and decision-making in autonomous driving. The gap remains in applying VLMs to understand complex scenarios interacting with pedestrians and efficient vehicle deployment. In this paper, we propose a knowledge distillation method that transfers knowledge from large-scale vision-language foundation models to efficient vision networks, and we apply it to pedestrian behavior prediction and scene understanding tasks, achieving promising results in generating more diverse and comprehensive semantic attributes. We also utilize multiple pre-trained models and ensemble techniques to boost the model's performance. We further examined the effectiveness of the model after knowledge distillation; the results show significant metric improvements in open-vocabulary perception and trajectory prediction tasks, which can potentially enhance the end-to-end performance of autonomous driving.

Paperid: 6085, https://arxiv.org/pdf/2501.01371.pdf

Abstract:
Recent Vision-Language Models (VLMs) have demonstrated remarkable capabilities in visual understanding and reasoning, and in particular on multiple-choice Visual Question Answering (VQA). Still, these models can make distinctly unnatural errors, for example, providing (wrong) answers to unanswerable VQA questions, such as questions asking about objects that do not appear in the image. To address this issue, we propose CLIP-UP: CLIP-based Unanswerable Problem detection, a novel lightweight method for equipping VLMs with the ability to withhold answers to unanswerable questions. By leveraging CLIP to extract question-image alignment information, CLIP-UP requires only efficient training of a few additional layers, while keeping the original VLMs' weights unchanged. Tested across LLaVA models, CLIP-UP achieves state-of-the-art results on the MM-UPD benchmark for assessing unanswerability in multiple-choice VQA, while preserving the original performance on other tasks.

Paperid: 6086, https://arxiv.org/pdf/2412.18917.pdf

Abstract:
Open-vocabulary panoptic segmentation remains a challenging problem. One of the biggest difficulties lies in training models to generalize to an unlimited number of classes using limited categorized training data. Recent popular methods involve large-scale vision-language pre-trained foundation models, such as CLIP. In this paper, we propose OMTSeg for open-vocabulary segmentation using another large-scale vision-language pre-trained model called BEiT-3 and leveraging the cross-modal attention between visual and linguistic features in BEiT-3 to achieve better performance. Experiments result demonstrates that OMTSeg performs favorably against state-of-the-art models.

Paperid: 6087, https://arxiv.org/pdf/2412.11663.pdf

Abstract:
In this paper we deal with image classification tasks using the powerful CLIP vision-language model. Our goal is to advance the classification performance using the CLIP's image encoder, by proposing a novel Large Multimodal Model (LMM) based regularization method. The proposed method uses an LMM to extract semantic descriptions for the images of the dataset. Then, it uses the CLIP's text encoder, frozen, in order to obtain the corresponding text embeddings and compute the mean semantic class descriptions. Subsequently, we adapt the CLIP's image encoder by adding a classification head, and we train it along with the image encoder output, apart from the main classification objective, with an additional auxiliary objective. The additional objective forces the embeddings at the image encoder's output to become similar to their corresponding LMM-generated mean semantic class descriptions. In this way, it produces embeddings with enhanced discrimination ability, leading to improved classification performance. The effectiveness of the proposed regularization method is validated through extensive experiments on three image classification datasets.

Paperid: 6088, https://arxiv.org/pdf/2412.10758.pdf

Abstract:
Vision-Language Models (VLMs) have emerged as key enablers for multimodal tasks, but their reliance on separate visual encoders introduces challenges in efficiency, scalability, and modality alignment. To address these limitations, we propose MUDAIF (Multimodal Unified Decoder with Adaptive Input Fusion), a decoder-only vision-language model that seamlessly integrates visual and textual inputs through a novel Vision-Token Adapter (VTA) and adaptive co-attention mechanism. By eliminating the need for a visual encoder, MUDAIF achieves enhanced efficiency, flexibility, and cross-modal understanding. Trained on a large-scale dataset of 45M image-text pairs, MUDAIF consistently outperforms state-of-the-art methods across multiple benchmarks, including VQA, image captioning, and multimodal reasoning tasks. Extensive analyses and human evaluations demonstrate MUDAIF's robustness, generalization capabilities, and practical usability, establishing it as a new standard in encoder-free vision-language models.

Paperid: 6089, https://arxiv.org/pdf/2412.04041.pdf

Abstract:
Technical analysis in finance, which aims at forecasting price movements in the future by analyzing past market data, relies on the insights that can be gained from the interpretation of stock charts; therefore, non-expert investors could greatly benefit from AI tools that can assist with the captioning of such charts. In our work, we introduce a new dataset StockGenChaR to evaluate large vision-language models in image captioning with stock charts. The purpose of the proposed task is to generate informative descriptions of the depicted charts and help to read the sentiment of the market regarding specific stocks, thus providing useful information for investors

Paperid: 6090, https://arxiv.org/pdf/2412.03841.pdf

Abstract:
Image Compression for Machines (ICM) aims to compress images for machine vision tasks rather than human viewing. Current works predominantly concentrate on high-level tasks like object detection and semantic segmentation. However, the quality of original images is usually not guaranteed in the real world, leading to even worse perceptual quality or downstream task performance after compression. Low-level (LL) machine vision models, like image restoration models, can help improve such quality, and thereby their compression requirements should also be considered. In this paper, we propose a pioneered ICM framework for LL machine vision tasks, namely LL-ICM. By jointly optimizing compression and LL tasks, the proposed LL-ICM not only enriches its encoding ability in generalizing to versatile LL tasks but also optimizes the processing ability of down-stream LL task models, achieving mutual adaptation for image codecs and LL task models. Furthermore, we integrate large-scale vision-language models into the LL-ICM framework to generate more universal and distortion-robust feature embeddings for LL vision tasks. Therefore, one LL-ICM codec can generalize to multiple tasks. We establish a solid benchmark to evaluate LL-ICM, which includes extensive objective experiments by using both full and no-reference image quality assessments. Experimental results show that LL-ICM can achieve 22.65% BD-rate reductions over the state-of-the-art methods.

Paperid: 6091, https://arxiv.org/pdf/2412.03178.pdf

Abstract:
Uncertainty quantification in text-to-image (T2I) generative models is crucial for understanding model behavior and improving output reliability. In this paper, we are the first to quantify and evaluate the uncertainty of T2I models with respect to the prompt. Alongside adapting existing approaches designed to measure uncertainty in the image space, we also introduce Prompt-based UNCertainty Estimation for T2I models (PUNC), a novel method leveraging Large Vision-Language Models (LVLMs) to better address uncertainties arising from the semantics of the prompt and generated images. PUNC utilizes a LVLM to caption a generated image, and then compares the caption with the original prompt in the more semantically meaningful text space. PUNC also enables the disentanglement of both aleatoric and epistemic uncertainties via precision and recall, which image-space approaches are unable to do. Extensive experiments demonstrate that PUNC outperforms state-of-the-art uncertainty estimation techniques across various settings. Uncertainty quantification in text-to-image generation models can be used on various applications including bias detection, copyright protection, and OOD detection. We also introduce a comprehensive dataset of text prompts and generation pairs to foster further research in uncertainty quantification for generative models. Our findings illustrate that PUNC not only achieves competitive performance but also enables novel applications in evaluating and improving the trustworthiness of text-to-image models.

Paperid: 6092, https://arxiv.org/pdf/2412.00139.pdf

Abstract:
Text-to-image retrieval is a critical task for managing diverse visual content, but common benchmarks for the task rely on small, single-domain datasets that fail to capture real-world complexity. Pre-trained vision-language models tend to perform well with easy negatives but struggle with hard negatives--visually similar yet incorrect images--especially in open-domain scenarios. To address this, we introduce Episodic Few-Shot Adaptation (EFSA), a novel test-time framework that adapts pre-trained models dynamically to a query's domain by fine-tuning on top-k retrieved candidates and synthetic captions generated for them. EFSA improves performance across diverse domains while preserving generalization, as shown in evaluations on queries from eight highly distinct visual domains and an open-domain retrieval pool of over one million images. Our work highlights the potential of episodic few-shot adaptation to enhance robustness in the critical and understudied task of open-domain text-to-image retrieval.

Paperid: 6093, https://arxiv.org/pdf/2412.00056.pdf

Abstract:
In recent years, vision-language models (VLMs) have been applied to various fields, including healthcare, education, finance, and manufacturing, with remarkable performance. However, concerns remain regarding VLMs' consistency and uncertainty, particularly in critical applications such as healthcare, which demand a high level of trust and reliability. This paper proposes a novel approach to evaluate uncertainty in VLMs' responses using a convex hull approach on a healthcare application for Visual Question Answering (VQA). LLM-CXR model is selected as the medical VLM utilized to generate responses for a given prompt at different temperature settings, i.e., 0.001, 0.25, 0.50, 0.75, and 1.00. According to the results, the LLM-CXR VLM shows a high uncertainty at higher temperature settings. Experimental outcomes emphasize the importance of uncertainty in VLMs' responses, especially in healthcare applications.

Paperid: 6094, https://arxiv.org/pdf/2411.18666.pdf

Abstract:
3D vision-language (VL) reasoning has gained significant attention due to its potential to bridge the 3D physical world with natural language descriptions. Existing approaches typically follow task-specific, highly specialized paradigms. Therefore, these methods focus on a limited range of reasoning sub-tasks and rely heavily on the hand-crafted modules and auxiliary losses. This highlights the need for a simpler, unified and general-purpose model. In this paper, we leverage the inherent connection between 3D scene graphs and natural language, proposing a 3D scene graph-guided vision-language pre-training (VLP) framework. Our approach utilizes modality encoders, graph convolutional layers and cross-attention layers to learn universal representations that adapt to a variety of 3D VL reasoning tasks, thereby eliminating the need for task-specific designs. The pre-training objectives include: 1) Scene graph-guided contrastive learning, which leverages the strong correlation between 3D scene graphs and natural language to align 3D objects with textual features at various fine-grained levels; and 2) Masked modality learning, which uses cross-modality information to reconstruct masked words and 3D objects. Instead of directly reconstructing the 3D point clouds of masked objects, we use position clues to predict their semantic categories. Extensive experiments demonstrate that our pre-training model, when fine-tuned on several downstream tasks, achieves performance comparable to or better than existing methods in tasks such as 3D visual grounding, 3D dense captioning, and 3D question answering.

Paperid: 6095, https://arxiv.org/pdf/2411.14937.pdf

Abstract:
Foundation models that bridge vision and language have made significant progress. While they have inspired many life-enriching applications, their potential for abuse in creating new threats remains largely unexplored. In this paper, we reveal that vision-language models (VLMs) can be weaponized to enhance gradient inversion attacks (GIAs) in federated learning (FL), where an FL server attempts to reconstruct private data samples from gradients shared by victim clients. Despite recent advances, existing GIAs struggle to reconstruct high-resolution images when the victim has a large local data batch. One promising direction is to focus reconstruction on valuable samples rather than the entire batch, but current methods lack the flexibility to target specific data of interest. To address this gap, we propose Geminio, the first approach to transform GIAs into semantically meaningful, targeted attacks. It enables a brand new privacy attack experience: attackers can describe, in natural language, the data they consider valuable, and Geminio will prioritize reconstruction to focus on those high-value samples. This is achieved by leveraging a pretrained VLM to guide the optimization of a malicious global model that, when shared with and optimized by a victim, retains only gradients of samples that match the attacker-specified query. Geminio can be launched at any FL round and has no impact on normal training (i.e., the FL server can steal clients' data while still producing a high-utility ML model as in benign scenarios). Extensive experiments demonstrate its effectiveness in pinpointing and reconstructing targeted samples, with high success rates across complex datasets and large batch sizes with resilience against defenses.

Paperid: 6096, https://arxiv.org/pdf/2411.12058.pdf

Abstract:
We demonstrate that vision language models (VLMs) are capable of recognizing the content in audio recordings when given corresponding spectrogram images. Specifically, we instruct VLMs to perform audio classification tasks in a few-shot setting by prompting them to classify a spectrogram image given example spectrogram images of each class. By carefully designing the spectrogram image representation and selecting good few-shot examples, we show that GPT-4o can achieve 59.00% cross-validated accuracy on the ESC-10 environmental sound classification dataset. Moreover, we demonstrate that VLMs currently outperform the only available commercial audio language model with audio understanding capabilities (Gemini-1.5) on the equivalent audio classification task (59.00% vs. 49.62%), and even perform slightly better than human experts on visual spectrogram classification (73.75% vs. 72.50% on first fold). We envision two potential use cases for these findings: (1) combining the spectrogram and language understanding capabilities of VLMs for audio caption augmentation, and (2) posing visual spectrogram classification as a challenge task for VLMs.

Paperid: 6097, https://arxiv.org/pdf/2411.11370.pdf

Abstract:
Unmanned aerial vehicle (UAV) patrol inspection has emerged as a predominant approach in transmission line monitoring owing to its cost-effectiveness. Detecting defects in transmission lines is a critical task during UAV patrol inspection. However, due to imaging distance and shooting angles, UAV patrol images often suffer from insufficient defect-related visual information, which has an adverse effect on detection accuracy. In this article, we propose a novel method for detecting defects in UAV patrol images, which is based on vision-language pretraining for transmission line (VLP-TL) and a progressive transfer strategy (PTS). Specifically, VLP-TL contains two novel pretraining tasks tailored for the transmission line scenario, aimimg at pretraining an image encoder with abundant knowledge acquired from both visual and linguistic information. Transferring the pretrained image encoder to the defect detector as its backbone can effectively alleviate the insufficient visual information problem. In addition, the PTS further improves transfer performance by progressively bridging the gap between pretraining and downstream defection detection. Experimental results demonstrate that the proposed method significantly improves defect detection accuracy by jointly utilizing multimodal information, overcoming the limitations of insufficient defect-related visual information provided by UAV patrol images.

Paperid: 6098, https://arxiv.org/pdf/2411.05357.pdf

Abstract:
The performance of vision-language models (VLMs), such as CLIP, in visual classification tasks, has been enhanced by leveraging semantic knowledge from large language models (LLMs), including GPT. Recent studies have shown that in zero-shot classification tasks, descriptors incorporating additional cues, high-level concepts, or even random characters often outperform those using only the category name. In many classification tasks, while the top-1 accuracy may be relatively low, the top-5 accuracy is often significantly higher. This gap implies that most misclassifications occur among a few similar classes, highlighting the model's difficulty in distinguishing between classes with subtle differences. To address this challenge, we introduce a novel concept of comparative descriptors. These descriptors emphasize the unique features of a target class against its most similar classes, enhancing differentiation. By generating and integrating these comparative descriptors into the classification framework, we refine the semantic focus and improve classification accuracy. An additional filtering process ensures that these descriptors are closer to the image embeddings in the CLIP space, further enhancing performance. Our approach demonstrates improved accuracy and robustness in visual classification tasks by addressing the specific challenge of subtle inter-class differences.

Paperid: 6099, https://arxiv.org/pdf/2411.03827.pdf

Abstract:
Ideation is a critical component of video-based design (VBD), where videos serve as the primary medium for design exploration and inspiration. The emergence of generative AI offers considerable potential to enhance this process by streamlining video analysis and facilitating idea generation. In this paper, we present DesignMinds, a prototype that integrates a state-of-the-art Vision-Language Model (VLM) with a context-enhanced Large Language Model (LLM) to support ideation in VBD. To evaluate DesignMinds, we conducted a between-subject study with 35 design practitioners, comparing its performance to a baseline condition. Our results demonstrate that DesignMinds significantly enhances the flexibility and originality of ideation, while also increasing task engagement. Importantly, the introduction of this technology did not negatively impact user experience, technology acceptance, or usability.

Paperid: 6100, https://arxiv.org/pdf/2411.00172.pdf

Abstract:
A major obstacle to the advancements of machine learning models in marine science, particularly in sonar imagery analysis, is the scarcity of AI-ready datasets. While there have been efforts to make AI-ready sonar image dataset publicly available, they suffer from limitations in terms of environment setting and scale. To bridge this gap, we introduce SeafloorAI, the first extensive AI-ready datasets for seafloor mapping across 5 geological layers that is curated in collaboration with marine scientists. We further extend the dataset to SeafloorGenAI by incorporating the language component in order to facilitate the development of both vision- and language-capable machine learning models for sonar imagery. The dataset consists of 62 geo-distributed data surveys spanning 17,300 square kilometers, with 696K sonar images, 827K annotated segmentation masks, 696K detailed language descriptions and approximately 7M question-answer pairs. By making our data processing source code publicly available, we aim to engage the marine science community to enrich the data pool and inspire the machine learning community to develop more robust models. This collaborative approach will enhance the capabilities and applications of our datasets within both fields.

Paperid: 6101, https://arxiv.org/pdf/2410.20165.pdf

Abstract:
Text-To-Image (TTI) generation is significant for controlled and diverse image generation with broad potential applications. Although current medical TTI methods have made some progress in report-to-Chest-Xray (CXR) generation, their generation performance may be limited due to the intrinsic characteristics of medical data. In this paper, we propose a novel disease-knowledge enhanced Diffusion-based TTI learning framework, named Diff-CXR, for medical report-to-CXR generation. First, to minimize the negative impacts of noisy data on generation, we devise a Latent Noise Filtering Strategy that gradually learns the general patterns of anomalies and removes them in the latent space. Then, an Adaptive Vision-Aware Textual Learning Strategy is designed to learn concise and important report embeddings in a domain-specific Vision-Language Model, providing textual guidance for Chest-Xray generation. Finally, by incorporating the general disease knowledge into the pretrained TTI model via a delicate control adapter, a disease-knowledge enhanced diffusion model is introduced to achieve realistic and precise report-to-CXR generation. Experimentally, our Diff-CXR outperforms previous SOTA medical TTI methods by 33.4\% / 8.0\% and 23.8\% / 56.4\% in the FID and mAUC score on MIMIC-CXR and IU-Xray, with the lowest computational complexity at 29.641 GFLOPs. Downstream experiments on three thorax disease classification benchmarks and one CXR-report generation benchmark demonstrate that Diff-CXR is effective in improving classical CXR analysis methods. Notably, models trained on the combination of 1\% real data and synthetic data can achieve a competitive mAUC score compared to models trained on all data, presenting promising clinical applications.

Paperid: 6102, https://arxiv.org/pdf/2410.15509.pdf

Abstract:
For specialized domains, there is often not a wealth of data with which to train large machine learning models. In such limited data / compute settings, various methods exist aiming to $\textit{do more with less}$, such as finetuning from a pretrained model, modulating difficulty levels as data are presented to a model (curriculum learning), and considering the role of model type / size. Approaches to efficient $\textit{machine}$ learning also take inspiration from $\textit{human}$ learning by considering use cases where machine learning systems have access to approximately the same number of words experienced by a 13 year old child (100M words). We investigate the role of 3 primary variables in a limited data regime as part of the multimodal track of the BabyLM challenge. We contrast: (i) curriculum learning, (ii), pretraining (with text-only data), (iii) model type. We modulate these variables and assess them on two types of tasks: (a) multimodal (text+image), and (b) unimodal (text-only) tasks. We find that curriculum learning benefits multimodal evaluations over non-curriclum learning models, particularly when combining text-only pretraining. On text-only tasks, curriculum learning appears to help models with smaller trainable parameter counts. We suggest possible reasons based on architectural differences and training designs as to why one might observe such results.

Paperid: 6103, https://arxiv.org/pdf/2410.14022.pdf

Abstract:
To advance autonomous dexterous manipulation, we propose a hybrid control method that combines the relative advantages of a fine-tuned Vision-Language-Action (VLA) model and diffusion models. The VLA model provides language commanded high-level planning, which is highly generalizable, while the diffusion model handles low-level interactions which offers the precision and robustness required for specific objects and environments. By incorporating a switching signal into the training-data, we enable event based transitions between these two models for a pick-and-place task where the target object and placement location is commanded through language. This approach is deployed on our anthropomorphic ADAPT Hand 2, a 13DoF robotic hand, which incorporates compliance through series elastic actuation allowing for resilience for any interactions: showing the first use of a multi-fingered hand controlled with a VLA model. We demonstrate this model switching approach results in a over 80\% success rate compared to under 40\% when only using a VLA model, enabled by accurate near-object arm motion by the VLA model and a multi-modal grasping motion with error recovery abilities from the diffusion model.

Paperid: 6104, https://arxiv.org/pdf/2410.12979.pdf

Abstract:
Real-time aerial image segmentation plays an important role in the environmental perception of Uncrewed Aerial Vehicles (UAVs). We introduce BlabberSeg, an optimized Vision-Language Model built on CLIPSeg for on-board, real-time processing of aerial images by UAVs. BlabberSeg improves the efficiency of CLIPSeg by reusing prompt and model features, reducing computational overhead while achieving real-time open-vocabulary aerial segmentation. We validated BlabberSeg in a safe landing scenario using the Dynamic Open-Vocabulary Enhanced SafE-Landing with Intelligence (DOVESEI) framework, which uses visual servoing and open-vocabulary segmentation. BlabberSeg reduces computational costs significantly, with a speed increase of 927.41% (16.78 Hz) on a NVIDIA Jetson Orin AGX (64GB) compared with the original CLIPSeg (1.81Hz), achieving real-time aerial segmentation with negligible loss in accuracy (2.1% as the ratio of the correctly segmented area with respect to CLIPSeg). BlabberSeg's source code is open and available online.

Paperid: 6105, https://arxiv.org/pdf/2410.12802.pdf

Abstract:
We consider an autonomous navigation robot that can accept human commands through natural language to provide services in an indoor environment. These natural language commands may include time, position, object, and action components. However, we observe that the positional components within such commands usually refer to objects in the environment that may contain different levels of positional ambiguity. For example, the command "Go to the chair!" may be ambiguous when there are multiple chairs of the same type in a room. In order to disambiguate these commands, we employ a large language model and a large vision-language model to conduct multiple turns of conversations with the user. We propose a two-level approach that utilizes a vision-language model to map the meanings in natural language to a unique object ID in images and then performs another mapping from the unique object ID to a 3D depth map, thereby allowing the robot to navigate from its current position to the target position. To the best of our knowledge, this is the first work linking foundation models to the positional ambiguity issue.

Paperid: 6106, https://arxiv.org/pdf/2410.11087.pdf

Abstract:
Vision language models (VLMs) have seen growing adoption in recent years, but many still struggle with basic spatial reasoning errors. We hypothesize that this is due to VLMs adopting pre-trained vision backbones, specifically vision transformers (ViTs) trained with image-level supervision and minimal inductive biases. Such models may fail to encode the class contents at each position in the image, and our goal is to resolve this with a vision backbone that effectively captures both local and global image semantics. Our main insight is that we do not require new supervision to learn this capability - pre-trained models contain significant knowledge of local semantics that we can extract and use for scalable self-supervision. We propose a new efficient post-training stage for ViTs called locality alignment and a novel fine-tuning procedure called MaskEmbed that uses a masked reconstruction loss to learn semantic contributions for each image patch. We first evaluate locality alignment with a vision-only benchmark, finding that it improves a model's performance at patch-level semantic segmentation, especially for strong backbones trained with image-caption pairs (e.g., CLIP and SigLIP). We then train a series of VLMs with and without locality alignment, and show that locality-aligned backbones improve performance across a range of benchmarks, particularly ones that involve spatial understanding (e.g., RefCOCO, OCID-Ref, TallyQA, VSR, AI2D). Overall, we demonstrate that we can efficiently learn local semantic extraction via a locality alignment stage, and that this procedure benefits VLM training recipes that use off-the-shelf vision backbones.

Paperid: 6107, https://arxiv.org/pdf/2410.07648.pdf

Abstract:
As the boosting development of large vision-language models like Contrastive Language-Image Pre-training (CLIP), many CLIP-like methods have shown impressive abilities on visual recognition, especially in low-data regimes scenes. However, we have noticed that most of these methods are limited to introducing new modifications on text and image encoder. Recently, latent diffusion models (LDMs) have shown good ability on image generation. The potent capabilities of LDMs direct our focus towards the latent representations sampled by UNet. Inspired by the conjecture in CoOp that learned prompts encode meanings beyond the existing vocabulary, we assume that, for deep models, the latent representations are concise and accurate understanding of images, in which high-frequency, imperceptible details are abstracted away. In this paper, we propose a Few-shot Language Image model Embedded with latent Representations (FLIER) for image recognition by introducing a latent encoder jointly trained with CLIP's image encoder, it incorporates pre-trained vision-language knowledge of CLIP and the latent representations from Stable Diffusion. We first generate images and corresponding latent representations via Stable Diffusion with the textual inputs from GPT-3. With latent representations as "models-understandable pixels", we introduce a flexible convolutional neural network with two convolutional layers to be the latent encoder, which is simpler than most encoders in vision-language models. The latent encoder is jointly trained with CLIP's image encoder, transferring pre-trained knowledge to downstream tasks better. Experiments and extensive ablation studies on various visual classification tasks demonstrate that FLIER performs state-of-the-art on 11 datasets for most few-shot classification.

Paperid: 6108, https://arxiv.org/pdf/2410.04514.pdf

Abstract:
Despite the great success of Large Vision-Language Models (LVLMs), they inevitably suffer from hallucination. As we know, both the visual encoder and the Large Language Model (LLM) decoder in LVLMs are Transformer-based, allowing the model to extract visual information and generate text outputs via attention mechanisms. We find that the attention distribution of LLM decoder on image tokens is highly consistent with the visual encoder and both distributions tend to focus on particular background tokens rather than the referred objects in the image. We attribute to the unexpected attention distribution to an inherent flaw in the visual encoder itself, which misguides LLMs to over emphasize the redundant information and generate object hallucination. To address the issue, we propose DAMRO, a novel training-free strategy that $D$ive into $A$ttention $M$echanism of LVLM to $R$educe $O$bject Hallucination. Specifically, our approach employs classification token (CLS) of ViT to filter out high-attention outlier tokens scattered in the background and then eliminate their influence during decoding stage. We evaluate our method on LVLMs including LLaVA-1.5, LLaVA-NeXT and InstructBLIP, using various benchmarks such as POPE, CHAIR, MME and GPT-4V Aided Evaluation. The results demonstrate that our approach significantly reduces the impact of these outlier tokens, thus effectively alleviating the hallucination of LVLMs. The code of our method will be released soon.

Paperid: 6109, https://arxiv.org/pdf/2410.04426.pdf

Abstract:
In this work, we address the real-world, challenging task of out-of-context misinformation detection, where a real image is paired with an incorrect caption for creating fake news. Existing approaches for this task assume the availability of large amounts of labeled data, which is often impractical in real-world, since it requires extensive manual intervention and domain expertise. In contrast, since obtaining a large corpus of unlabeled image-text pairs is much easier, here, we propose a semi-supervised protocol, where the model has access to a limited number of labeled image-text pairs and a large corpus of unlabeled pairs. Additionally, the occurrence of fake news being much lesser compared to the real ones, the datasets tend to be highly imbalanced, thus making the task even more challenging. Towards this goal, we propose a novel framework, Consensus from Vision-Language Models (CoVLM), which generates robust pseudo-labels for unlabeled pairs using thresholds derived from the labeled data. This approach can automatically determine the right threshold parameters of the model for selecting the confident pseudo-labels. Experimental results on benchmark datasets across challenging conditions and comparisons with state-of-the-art approaches demonstrate the effectiveness of our framework.

Paperid: 6110, https://arxiv.org/pdf/2410.02027.pdf

Abstract:
There is a scarcity of multilingual vision-language models that properly account for the perceptual differences that are reflected in image captions across languages and cultures. In this work, through a multimodal, multilingual retrieval case study, we quantify the existing lack of model flexibility. We empirically show performance gaps between training on captions that come from native German perception and captions that have been either machine-translated or human-translated from English into German. To address these gaps, we further propose and evaluate caption augmentation strategies. While we achieve mean recall improvements (+1.3), gaps still remain, indicating an open area of future work for the community.

Paperid: 6111, https://arxiv.org/pdf/2409.18060.pdf

Abstract:
Ensuring accessibility in mobile applications remains a significant challenge, particularly for visually impaired users who rely on screen readers. User interface icons are essential for navigation and interaction and often lack meaningful alt-text, creating barriers to effective use. Traditional deep learning approaches for generating alt-text require extensive datasets and struggle with the diversity and imbalance of icon types. More recent Vision Language Models (VLMs) require complete UI screens, which can be impractical during the iterative phases of app development. To address these issues, we introduce a novel method using Large Language Models (LLMs) to autonomously generate informative alt-text for mobile UI icons with partial UI data. By incorporating icon context, that include class, resource ID, bounds, OCR-detected text, and contextual information from parent and sibling nodes, we fine-tune an off-the-shelf LLM on a small dataset of approximately 1.4k icons, yielding IconDesc. In an empirical evaluation and a user study IconDesc demonstrates significant improvements in generating relevant alt-text. This ability makes IconDesc an invaluable tool for developers, aiding in the rapid iteration and enhancement of UI accessibility.

Paperid: 6112, https://arxiv.org/pdf/2409.17457.pdf

Abstract:
Parametric Computer-Aided Design (CAD) is central to contemporary mechanical design. However, it encounters challenges in achieving precise parametric sketch modeling and lacks practical evaluation metrics suitable for mechanical design. We harness the capabilities of pre-trained foundation models, renowned for their successes in natural language processing and computer vision, to develop generative models specifically for CAD. These models are adept at understanding complex geometries and design reasoning, a crucial advancement in CAD technology. In this paper, we propose CadVLM, an end-to-end vision language model for CAD generation. Our approach involves adapting pre-trained foundation models to manipulate engineering sketches effectively, integrating both sketch primitive sequences and sketch images. Extensive experiments demonstrate superior performance on multiple CAD sketch generation tasks such as CAD autocompletion, CAD autoconstraint, and image conditional generation. To our knowledge, this is the first instance of a multimodal Large Language Model (LLM) being successfully applied to parametric CAD generation, representing a pioneering step in the field of computer-aided mechanical design.

Paperid: 6113, https://arxiv.org/pdf/2409.15922.pdf

Abstract:
While Vision-Language Models (VLMs) are increasingly used to generate reward signals for training embodied agents to follow instructions, our research reveals that agents guided by VLM rewards often underperform compared to those employing only intrinsic (exploration-driven) rewards, contradicting expectations set by recent work. We hypothesize that false positive rewards -- instances where unintended trajectories are incorrectly rewarded -- are more detrimental than false negatives. Our analysis confirms this hypothesis, revealing that the widely used cosine similarity metric is prone to false positive reward estimates. To address this, we introduce BiMI ({Bi}nary {M}utual {I}nformation), a novel reward function designed to mitigate noise. BiMI significantly enhances learning efficiency across diverse and challenging embodied navigation environments. Our findings offer a nuanced understanding of how different types of reward noise impact agent learning and highlight the importance of addressing multimodal reward signal noise when training embodied agents

Paperid: 6114, https://arxiv.org/pdf/2409.15574.pdf

Abstract:
Vision language models (VLM) have achieved success in both natural language comprehension and image recognition tasks. However, their use in pathology report generation for whole slide images (WSIs) is still limited due to the huge size of multi-scale WSIs and the high cost of WSI annotation. Moreover, in most of the existing research on pathology report generation, sufficient validation regarding clinical efficacy has not been conducted. Herein, we propose a novel Patient-level Multi-organ Pathology Report Generation (PMPRG) model, which utilizes the multi-scale WSI features from our proposed multi-scale regional vision transformer (MR-ViT) model and their real pathology reports to guide VLM training for accurate pathology report generation. The model then automatically generates a report based on the provided key features attended regional features. We assessed our model using a WSI dataset consisting of multiple organs, including the colon and kidney. Our model achieved a METEOR score of 0.68, demonstrating the effectiveness of our approach. This model allows pathologists to efficiently generate pathology reports for patients, regardless of the number of WSIs involved.

Paperid: 6115, https://arxiv.org/pdf/2409.14607.pdf

Abstract:
Contrastive image-text pre-trained models such as CLIP have shown remarkable adaptability to downstream tasks. However, they face challenges due to the high computational requirements of the Vision Transformer (ViT) backbone. Current strategies to boost ViT efficiency focus on pruning patch tokens but fall short in addressing the multimodal nature of CLIP and identifying the optimal subset of tokens for maximum performance. To address this, we propose greedy search methods to establish a "Golden Ranking" and introduce a lightweight predictor specifically trained to approximate this Ranking. To compensate for any performance degradation resulting from token pruning, we incorporate learnable visual tokens that aid in restoring and potentially enhancing the model's performance. Our work presents a comprehensive and systematic investigation of pruning tokens within the ViT backbone of CLIP models. Through our framework, we successfully reduced 40% of patch tokens in CLIP's ViT while only suffering a minimal average accuracy loss of 0.3 across seven datasets. Our study lays the groundwork for building more computationally efficient multimodal models without sacrificing their performance, addressing a key challenge in the application of advanced vision-language models.

Paperid: 6116, https://arxiv.org/pdf/2409.12597.pdf

Abstract:
In recent years, considerable research has been conducted on vision-language models that handle both image and text data; these models are being applied to diverse downstream tasks, such as "image-related chat," "image recognition by instruction," and "answering visual questions." Vision-language models (VLMs), such as Contrastive Language-Image Pre-training (CLIP), are also high-performance image classifiers that are being developed into domain adaptation methods that can utilize language information to extend into unseen domains. However, because these VLMs embed images as a single point in a unified embedding space, there is room for improvement in the classification accuracy. Therefore, in this study, we proposed the Latent Augmentation using Regional Embedding (LARE), which embeds the image as a region in the unified embedding space learned by the VLM. By sampling the augmented image embeddings from within this latent region, LARE enables data augmentation to various unseen domains, not just to specific unseen domains. LARE achieves robust image classification for domains in and out using augmented image embeddings to fine-tune VLMs. We demonstrate that LARE outperforms previous fine-tuning models in terms of image classification accuracy on three benchmarks. We also demonstrate that LARE is a more robust and general model that is valid under multiple conditions, such as unseen domains, small amounts of data, and imbalanced data.

Paperid: 6117, https://arxiv.org/pdf/2601.21944.pdf

Abstract:
The widespread adoption of Vision-Language Models (VLMs) across fields has amplified concerns about model interpretability. Distressingly, these models are often treated as black-boxes, with limited or non-existent investigation of their decision making process. Despite numerous post- and ante-hoc interepretability methods, systematic and objective evaluation of the learned representations remains limited, particularly for sparsity-aware methods that are increasingly considered to "induce interpretability". In this work, we focus on Concept Bottleneck Models and investigate how different modeling decisions affect the emerging representations. We introduce the notion of clarity, a measure, capturing the interplay between the downstream performance and the sparsity and precision of the concept representation, while proposing an interpretability assessment framework using datasets with ground truth concept annotations. We consider both VLM- and attribute predictor-based CBMs, and three different sparsity-inducing strategies: per example $\ell_1, \ell_0$ and Bernoulli-based formulations. Our experiments reveal a critical trade-off between flexibility and interpretability, under which a given method can exhibit markedly different behaviors even at comparable performance levels. The code will be made publicly available upon publication.

Paperid: 6118, https://arxiv.org/pdf/2601.20321.pdf

Abstract:
Vision-Language-Action (VLA) models have recently emerged as powerful generalists for robotic manipulation. However, due to their predominant reliance on visual modalities, they fundamentally lack the physical intuition required for contact-rich tasks that require precise force regulation and physical reasoning. Existing attempts to incorporate vision-based tactile sensing into VLA models typically treat tactile inputs as auxiliary visual textures, thereby overlooking the underlying correlation between surface deformation and interaction dynamics. To bridge this gap, we propose a paradigm shift from tactile-vision alignment to tactile-force alignment. Here, we introduce TaF-VLA, a framework that explicitly grounds high-dimensional tactile observations in physical interaction forces. To facilitate this, we develop an automated tactile-force data acquisition device and curate the TaF-Dataset, comprising over 10 million synchronized tactile observations, 6-axis force/torque, and matrix force map. To align sequential tactile observations with interaction forces, the central component of our approach is the Tactile-Force Adapter (TaF-Adapter), a tactile sensor encoder that extracts discretized latent information for encoding tactile observations. This mechanism ensures that the learned representations capture history-dependent, noise-insensitive physical dynamics rather than static visual textures. Finally, we integrate this force-aligned encoder into a VLA backbone. Extensive real-world experiments demonstrate that TaF-VLA policy significantly outperforms state-of-the-art tactile-vision-aligned and vision-only baselines on contact-rich tasks, verifying its ability to achieve robust, force-aware manipulation through cross-modal physical reasoning.

Paperid: 6119, https://arxiv.org/pdf/2601.20303.pdf

Abstract:
Estimating object mass from visual input is challenging because mass depends jointly on geometric volume and material-dependent density, neither of which is directly observable from RGB appearance. Consequently, mass prediction from pixels is ill-posed and therefore benefits from physically meaningful representations to constrain the space of plausible solutions. We propose a physically structured framework for single-image mass estimation that addresses this ambiguity by aligning visual cues with the physical factors governing mass. From a single RGB image, we recover object-centric three-dimensional geometry via monocular depth estimation to inform volume and extract coarse material semantics using a vision-language model to guide density-related reasoning. These geometry, semantic, and appearance representations are fused through an instance-adaptive gating mechanism, and two physically guided latent factors (volume- and density-related) are predicted through separate regression heads under mass-only supervision. Experiments on image2mass and ABO-500 show that the proposed method consistently outperforms state-of-the-art methods.

Paperid: 6120, https://arxiv.org/pdf/2601.19448.pdf

Abstract:
Deep Neural Networks remain inherently vulnerable to backdoor attacks. Traditional test-time defenses largely operate under the paradigm of internal diagnosis methods like model repairing or input robustness, yet these approaches are often fragile under advanced attacks as they remain entangled with the victim model's corrupted parameters. We propose a paradigm shift from Internal Diagnosis to External Semantic Auditing, arguing that effective defense requires decoupling safety from the victim model via an independent, semantically grounded auditor. To this end, we present a framework harnessing Universal Vision-Language Models (VLMs) as evolving semantic gatekeepers. We introduce PRISM (Prototype Refinement & Inspection via Statistical Monitoring), which overcomes the domain gap of general VLMs through two key mechanisms: a Hybrid VLM Teacher that dynamically refines visual prototypes online, and an Adaptive Router powered by statistical margin monitoring to calibrate gating thresholds in real-time. Extensive evaluation across 17 datasets and 11 attack types demonstrates that PRISM achieves state-of-the-art performance, suppressing Attack Success Rate to <1% on CIFAR-10 while improving clean accuracy, establishing a new standard for model-agnostic, externalized security.

Paperid: 6121, https://arxiv.org/pdf/2601.17535.pdf

Abstract:
Vision-Language Models like CLIP create aligned embedding spaces for text and images, making it possible for anyone to build a visual classifier by simply naming the classes they want to distinguish. However, a model that works well in one domain may fail in another, and non-expert users have no straightforward way to assess whether their chosen VLM will work on their problem. We build on prior work using text-only comparisons to evaluate how well a model works for a given natural language task, and explore approaches that also generate synthetic images relevant to that task to evaluate and refine the prediction of zero-shot accuracy. We show that generated imagery to the baseline text-only scores substantially improves the quality of these predictions. Additionally, it gives a user feedback on the kinds of images that were used to make the assessment. Experiments on standard CLIP benchmark datasets demonstrate that the image-based approach helps users predict, without any labeled examples, whether a VLM will be effective for their application.

Paperid: 6122, https://arxiv.org/pdf/2601.17152.pdf

Abstract:
Multi-agent large language model (LLM) and vision-language model (VLM) debate systems employ specialized roles for complex problem-solving, yet model specializations are not leveraged to decide which model should fill which role. We propose dynamic role assignment, a framework that runs a Meta-Debate to select suitable agents before the actual debate. The meta-debate has two stages: (1) proposal, where candidates provide role-tailored arguments, and (2) peer review, where proposals are scored with data and role-specific criteria to choose the best agent for each position. We evaluate our method on LLM problem solving benchmarks. Applied on top of existing debate systems, our approach consistently outperforms uniform assignments (filling all roles with the same model) by up to 74.8% and random assignments (assigning models to roles without considering their suitability) by up to 29.7%, depending on the task and the specific assignment. This work establishes a new paradigm for multi-agent system design, shifting from static agent deployment to dynamic and capability-aware selection.

Paperid: 6123, https://arxiv.org/pdf/2601.16046.pdf

Abstract:
Language-driven dexterous grasp generation requires the models to understand task semantics, 3D geometry, and complex hand-object interactions. While vision-language models have been applied to this problem, existing approaches directly map observations to grasp parameters without intermediate reasoning about physical interactions. We present DextER, Dexterous Grasp Generation with Embodied Reasoning, which introduces contact-based embodied reasoning for multi-finger manipulation. Our key insight is that predicting which hand links contact where on the object surface provides an embodiment-aware intermediate representation bridging task semantics with physical constraints. DextER autoregressively generates embodied contact tokens specifying which finger links contact where on the object surface, followed by grasp tokens encoding the hand configuration. On DexGYS, DextER achieves 67.14% success rate, outperforming state-of-the-art by 3.83%p with 96.4% improvement in intention alignment. We also demonstrate steerable generation through partial contact specification, providing fine-grained control over grasp synthesis.

Paperid: 6124, https://arxiv.org/pdf/2601.12809.pdf

Abstract:
Spatial understanding remains a key challenge in vision-language models. Yet it is still unclear whether such understanding is truly acquired, and if so, through what mechanisms. We present a controllable 1D image-text testbed to probe how left-right relational understanding emerges in Transformer-based vision and text encoders trained with a CLIP-style contrastive objective. We train lightweight Transformer-based vision and text encoders end-to-end on paired descriptions of one- and two-object scenes and evaluate generalization to unseen object pairs while systematically varying label and layout diversity. We find that contrastive training learns left-right relations and that label diversity, more than layout diversity, is the primary driver of generalization in this setting. To gain the mechanistic understanding, we perform an attention decomposition and show that interactions between positional and token embeddings induce a horizontal attention gradient that breaks left-right symmetry in the encoders; ablating this contribution substantially reduces left-right discrimination. Our results provide a mechanistic insight of when and how CLIP-style models acquire relational competence.

Paperid: 6125, https://arxiv.org/pdf/2601.10313.pdf

Abstract:
Existing adversarial attacks for VLP models are mostly sample-specific, resulting in substantial computational overhead when scaled to large datasets or new scenarios. To overcome this limitation, we propose Hierarchical Refinement Attack (HRA), a multimodal universal attack framework for VLP models. HRA refines universal adversarial perturbations (UAPs) at both the sample level and the optimization level. For the image modality, we disentangle adversarial examples into clean images and perturbations, allowing each component to be handled independently for more effective disruption of cross-modal alignment. We further introduce a ScMix augmentation strategy that diversifies visual contexts and strengthens both global and local utility of UAPs, thereby reducing reliance on spurious features. In addition, we refine the optimization path by leveraging a temporal hierarchy of historical and estimated future gradients to avoid local minima and stabilize universal perturbation learning. For the text modality, HRA identifies globally influential words by combining intra-sentence and inter-sentence importance measures, and subsequently utilizes these words as universal text perturbations. Extensive experiments across various downstream tasks, VLP models, and datasets demonstrate the superiority of the proposed universal multimodal attacks.

Paperid: 6126, https://arxiv.org/pdf/2601.09981.pdf

Abstract:
Reasoning segmentation is an emerging vision-language task that requires reasoning over intricate text queries to precisely segment objects. However, existing methods typically suffer from overthinking, generating verbose reasoning chains that interfere with object localization in multimodal large language models (MLLMs). To address this issue, we propose DR$^2$Seg, a self-rewarding framework that improves both reasoning efficiency and segmentation accuracy without requiring extra thinking supervision. DR$^2$Seg employs a two-stage rollout strategy that decomposes reasoning segmentation into multimodal reasoning and referring segmentation. In the first stage, the model generates a self-contained description that explicitly specifies the target object. In the second stage, this description replaces the original complex query to verify its self-containment. Based on this design, two self-rewards are introduced to strengthen goal-oriented reasoning and suppress redundant thinking. Extensive experiments across MLLMs of varying scales and segmentation models demonstrate that DR$^2$Seg consistently improves reasoning efficiency and overall segmentation performance.

Paperid: 6127, https://arxiv.org/pdf/2601.09746.pdf

Abstract:
This paper introduces a novel Multi-Agent Cooperative Learning (MACL) framework to address cross-modal alignment collapse in vision-language models when handling out-of-distribution (OOD) concepts. Four core agents, including image, text, name, and coordination agents, collaboratively mitigate modality imbalance through structured message passing. The proposed framework enables multi-agent feature space name learning, incorporates a context exchange enhanced few-shot learning algorithm, and adopts an adaptive dynamic balancing mechanism to regulate inter-agent contributions. Experiments on the VISTA-Beyond dataset demonstrate that MACL significantly improves performance in both few-shot and zero-shot settings, achieving 1-5% precision gains across diverse visual domains.

Paperid: 6128, https://arxiv.org/pdf/2601.08454.pdf

Abstract:
Constructing an accurate simulation model of real-world environments requires reliable estimation of physical parameters such as mass, geometry, friction, and contact surfaces. Traditional real-to-simulation (Real2Sim) pipelines rely on manual measurements or fixed, pre-programmed exploration routines, which limit their adaptability to varying tasks and user intents. This paper presents a Real2Sim framework that autonomously generates and executes Behavior Trees for task-specific physical interactions to acquire only the parameters required for a given simulation objective, without relying on pre-defined task templates or expert-designed exploration routines. Given a high-level user request, an incomplete simulation description, and an RGB observation of the scene, a vision-language model performs multi-modal reasoning to identify relevant objects, infer required physical parameters, and generate a structured Behavior Tree composed of elementary robotic actions. The resulting behavior is executed on a torque-controlled Franka Emika Panda, enabling compliant, contact-rich interactions for parameter estimation. The acquired measurements are used to automatically construct a physics-aware simulation. Experimental results on the real manipulator demonstrate estimation of object mass, surface height, and friction-related quantities across multiple scenarios, including occluded objects and incomplete prior models. The proposed approach enables interpretable, intent-driven, and autonomously Real2Sim pipelines, bridging high-level reasoning with physically-grounded robotic interaction.

Paperid: 6129, https://arxiv.org/pdf/2601.08408.pdf

Abstract:
The demand for real-time visual understanding and interaction in complex scenarios is increasingly critical for unmanned aerial vehicles. However, a significant challenge arises from the contradiction between the high computational cost of large Vision language models and the limited computing resources available on UAV edge devices. To address this challenge, this paper proposes a lightweight multimodal task platform based on BLIP-2, integrated with YOLO-World and YOLOv8-Seg models. This integration extends the multi-task capabilities of BLIP-2 for UAV applications with minimal adaptation and without requiring task-specific fine-tuning on drone data. Firstly, the deep integration of BLIP-2 with YOLO models enables it to leverage the precise perceptual results of YOLO for fundamental tasks like object detection and instance segmentation, thereby facilitating deeper visual-attention understanding and reasoning. Secondly, a content-aware key frame sampling mechanism based on K-Means clustering is designed, which incorporates intelligent frame selection and temporal feature concatenation. This equips the lightweight BLIP-2 architecture with the capability to handle video-level interactive tasks effectively. Thirdly, a unified prompt optimization scheme for multi-task adaptation is implemented. This scheme strategically injects structured event logs from the YOLO models as contextual information into BLIP-2's input. Combined with output constraints designed to filter out technical details, this approach effectively guides the model to generate accurate and contextually relevant outputs for various tasks.

Paperid: 6130, https://arxiv.org/pdf/2601.08219.pdf

Abstract:
Can a small team of agents peel a matrix apart, one rank-1 slice at a time? We propose an agentic approach to matrix deflation in which a solver Large Language Model (LLM) generates rank-1 Singular Value Decomposition (SVD) updates and a Vision Language Model (VLM) accepts or rejects each update and decides when to stop, eliminating fixed norm thresholds. Solver stability is improved through in-context learning (ICL) and types of row/column permutations that expose visually coherent structure. We evaluate on Digits ($8{\times}8$), CIFAR-10 ($32{\times}32$ grayscale), and synthetic ($16{\times}16$) matrices with and without Gaussian noise. In the synthetic noisy case, where the true construction rank $k$ is known, numerical deflation provides the noise target and our best agentic configuration differs by only $1.75$ RMSE of the target. For Digits and CIFAR-10, targets are defined by deflating until the Frobenius norm reaches $10\%$ of the original. Across all settings, our agent achieves competitive results, suggesting that fully agentic, threshold-free deflation is a viable alternative to classical numerical algorithms.

Paperid: 6131, https://arxiv.org/pdf/2601.06394.pdf

Abstract:
Understanding student behavior in the classroom is essential to improve both pedagogical quality and student engagement. Existing methods for predicting student engagement typically require substantial annotated data to model the diversity of student behaviors, yet privacy concerns often restrict researchers to their own proprietary datasets. Moreover, the classroom context, represented in peers' actions, is ignored. To address the aforementioned limitation, we propose a novel three-stage framework for video-based student engagement measurement. First, we explore the few-shot adaptation of the vision-language model for student action recognition, which is fine-tuned to distinguish among action categories with a few training samples. Second, to handle continuous and unpredictable student actions, we utilize the sliding temporal window technique to divide each student's 2-minute-long video into non-overlapping segments. Each segment is assigned an action category via the fine-tuned VLM model, generating a sequence of action predictions. Finally, we leverage the large language model to classify this entire sequence of actions, together with the classroom context, as belonging to an engaged or disengaged student. The experimental results demonstrate the effectiveness of the proposed approach in identifying student engagement.

Paperid: 6132, https://arxiv.org/pdf/2601.01322.pdf

Abstract:
Modern Vision-Language Models (VLMs) achieve impressive performance but are limited by the quadratic complexity of self-attention, which prevents their deployment on edge devices and makes their understanding of high-resolution images and long-context videos prohibitively expensive. To address this challenge, we introduce LinMU (Linear-complexity Multimodal Understanding), a VLM design that achieves linear complexity without using any quadratic-complexity modules while maintaining the performance of global-attention-based VLMs. LinMU replaces every self-attention layer in the VLM with the M-MATE block: a dual-branch module that combines a bidirectional state-space model for global context (Flex-MA branch) with localized Swin-style window attention (Local-Swin branch) for adjacent correlations. To transform a pre-trained VLM into the LinMU architecture, we propose a three-stage distillation framework that (i) initializes both branches with self-attention weights and trains the Flex-MA branch alone, (ii) unfreezes the Local-Swin branch and fine-tunes it jointly with the Flex-MA branch, and (iii) unfreezes the remaining blocks and fine-tunes them using LoRA adapters, while regressing on hidden states and token-level logits of the frozen VLM teacher. On MMMU, TextVQA, LongVideoBench, Video-MME, and other benchmarks, LinMU matches the performance of teacher models, yet reduces Time-To-First-Token (TTFT) by up to 2.7$\times$ and improves token throughput by up to 9.0$\times$ on minute-length videos. Ablations confirm the importance of each distillation stage and the necessity of the two branches of the M-MATE block. The proposed framework demonstrates that state-of-the-art multimodal reasoning can be achieved without quadratic attention, thus opening up avenues for long-context VLMs that can deal with high-resolution images and long videos.

Paperid: 6133, https://arxiv.org/pdf/2512.23572.pdf

Abstract:
Following the initial flourishing of large language models (LLMs), there has been a surge in proposed large vision-language models (LVLMs) that integrate LLMs with vision capabilities. However, it has been observed that LVLMs, after tuning to visual instruction using commonly used training datasets, often fail to exhibit the instruction-following ability that was present in the LLM before integration, leading to results in which they do not follow task instructions as expected. This study quantitatively demonstrates that LVLMs' instruction-following ability declines after fine-tuning and analyzes its underlying causes. In particular, we constructed new training datasets highlighting whether the output format is specified. Then, we investigated how explicitly indicating the output format during fine-tuning affects LVLMs' instruction-following ability. Our quantitative evaluation confirmed that LVLMs' instruction-following ability declines after fine-tuning with commonly used datasets. Furthermore, we found that LVLMs trained with datasets, including instructions on output format, tend to follow instructions more accurately than models that do not. These findings suggest that including samples with instructions on output format during (visual) instruction tuning may help mitigate the decline in instruction-following abilities.

Paperid: 6134, https://arxiv.org/pdf/2512.23365.pdf

Abstract:
The rapid progress of Multimodal Large Language Models (MLLMs) has unlocked the potential for enhanced 3D scene understanding and spatial reasoning. However, existing approaches often rely on pre-constructed 3D representations or off-the-shelf reconstruction pipelines, which constrain scalability and real-world applicability. A recent line of work explores learning spatial reasoning directly from multi-view images, enabling Vision-Language Models (VLMs) to understand 3D scenes without explicit 3D reconstructions. Nevertheless, key challenges that frequently arise in real-world environments, such as partial visibility, occlusion, and low-overlap conditions that require spatial reasoning from fragmented visual cues, remain under-explored. To address these limitations, we propose a scalable multi-view data generation and annotation pipeline that constructs realistic spatial reasoning QAs, resulting in SpatialMosaic, a comprehensive instruction-tuning dataset featuring 2M QA pairs. We further introduce SpatialMosaic-Bench, a challenging benchmark for evaluating multi-view spatial reasoning under realistic and challenging scenarios, consisting of 1M QA pairs across 6 tasks. In addition, we present SpatialMosaicVLM, a hybrid framework that integrates 3D reconstruction models as geometry encoders within VLMs for robust spatial reasoning. Extensive experiments demonstrate that our proposed dataset and VQA tasks effectively enhance spatial reasoning under challenging multi-view conditions, validating the effectiveness of our data generation pipeline in constructing realistic and diverse QA pairs. Code and dataset will be available soon.

Paperid: 6135, https://arxiv.org/pdf/2512.20781.pdf

Abstract:
Composed Image Retrieval (CIR) aims to find a target image that aligns with user intent, expressed through a reference image and a modification text. While Zero-shot CIR (ZS-CIR) methods sidestep the need for labeled training data by leveraging pretrained vision-language models, they often rely on a single fused query that merges all descriptive cues of what the user wants, tending to dilute key information and failing to account for what they wish to avoid. Moreover, current CIR benchmarks assume a single correct target per query, overlooking the ambiguity in modification texts. To address these challenges, we propose Soft Filtering with Textual constraints (SoFT), a training-free, plug-and-play filtering module for ZS-CIR. SoFT leverages multimodal large language models (LLMs) to extract two complementary constraints from the reference-modification pair: prescriptive (must-have) and proscriptive (must-avoid) constraints. These serve as semantic filters that reward or penalize candidate images to re-rank results, without modifying the base retrieval model or adding supervision. In addition, we construct a two-stage dataset pipeline that refines CIR benchmarks. We first identify multiple plausible targets per query to construct multi-target triplets, capturing the open-ended nature of user intent. Then guide multimodal LLMs to rewrite the modification text to focus on one target, while referencing contrastive distractors to ensure precision. This enables more comprehensive and reliable evaluation under varying ambiguity levels. Applied on top of CIReVL, a ZS-CIR retriever, SoFT raises R@5 to 65.25 on CIRR (+12.94), mAP@50 to 27.93 on CIRCO (+6.13), and R@50 to 58.44 on FashionIQ (+4.59), demonstrating broad effectiveness.

Paperid: 6136, https://arxiv.org/pdf/2512.20556.pdf

Abstract:
Image fusion aims to synthesize a single high-quality image from a pair of inputs captured under challenging conditions, such as differing exposure levels or focal depths. A core challenge lies in effectively handling disparities in dynamic range and focus depth between the inputs. With the advent of vision-language models, recent methods incorporate textual descriptions as auxiliary guidance to enhance fusion quality. However, simply incorporating coarse-grained descriptions hampers the understanding of fine-grained details and poses challenges for precise cross-modal alignment. To address these limitations, we propose Multi-grained Text-guided Image Fusion (MTIF), a novel fusion paradigm with three key designs. First, it introduces multi-grained textual descriptions that separately capture fine details, structural cues, and semantic content, guiding image fusion through a hierarchical cross-modal modulation module. Second, it involves supervision signals at each granularity to facilitate alignment between visual and textual features and enhance the utility of auxiliary text. Third, it adopts a saliency-driven enrichment module to augment training data with dense semantic content, further strengthening the cross-modal modulation and alignment. Extensive experiments show that MTIF consistently outperforms previous methods on both multi-exposure and multi-focus image fusion tasks.

Paperid: 6137, https://arxiv.org/pdf/2512.20417.pdf

Abstract:
Automated video surveillance with Large Vision-Language Models is limited by their inherent bias towards normality, often failing to detect crimes. While Chain-of-Thought reasoning strategies show significant potential for improving performance in language tasks, the lack of inductive anomaly biases in their reasoning further steers the models towards normal interpretations. To address this, we propose Chain-of-Anomaly-Thoughts (CoAT), a multi-agent reasoning framework that introduces inductive criminal bias in the reasoning process through a final, anomaly-focused classification layer. Our method significantly improves Anomaly Detection, boosting F1-score by 11.8 p.p. on challenging low-resolution footage and Anomaly Classification by 3.78 p.p. in high-resolution videos.

Paperid: 6138, https://arxiv.org/pdf/2512.18619.pdf

Abstract:
We present ChronoDreamer, an action-conditioned world model for contact-rich robotic manipulation. Given a history of egocentric RGB frames, contact maps, actions, and joint states, ChronoDreamer predicts future video frames, contact distributions, and joint angles via a spatial-temporal transformer trained with MaskGIT-style masked prediction. Contact is encoded as depth-weighted Gaussian splat images that render 3D forces into a camera-aligned format suitable for vision backbones. At inference, predicted rollouts are evaluated by a vision-language model that reasons about collision likelihood, enabling rejection sampling of unsafe actions before execution. We train and evaluate on DreamerBench, a simulation dataset generated with Project Chrono that provides synchronized RGB, contact splat, proprioception, and physics annotations across rigid and deformable object scenarios. Qualitative results demonstrate that the model preserves spatial coherence during non-contact motion and generates plausible contact predictions, while the LLM-based judge distinguishes collision from non-collision trajectories.

Paperid: 6139, https://arxiv.org/pdf/2512.18028.pdf

Abstract:
Vision-language navigation requires agents to reason and act under constraints of embodiment. While vision-language models (VLMs) demonstrate strong generalization, current benchmarks provide limited understanding of how embodiment -- i.e., the choice of physical platform, sensor configuration, and modality alignment -- influences perception, reasoning, and control. We introduce Embodied4C, a closed-loop benchmark designed as a Turing test for embodied reasoning. The benchmark evaluates the core embodied capabilities of VLMs across three heterogeneous embodiments -- autonomous vehicles, aerial drones, and robotic manipulators -- through approximately 1.1K one-shot reasoning questions and 58 goal-directed navigation tasks. These tasks jointly assess four foundational dimensions: semantic, spatial, temporal, and physical reasoning. Each embodiment presents dynamic sensor configurations and environment variations to probe generalization beyond platform-specific adaptation. To prevent embodiment overfitting, Embodied4C integrates domain-far queries targeting abstract and cross-context reasoning. Comprehensive evaluation across ten state-of-the-art VLMs and four embodied control baselines shows that cross-modal alignment and instruction tuning matter more than scale, while spatial and temporal reasoning remains the primary bottleneck for reliable embodied competence.

Paperid: 6140, https://arxiv.org/pdf/2512.17160.pdf

Abstract:
Vision-Language Models (VLMs) have shown strong performance in zero-shot image classification tasks. However, existing methods, including Contrastive Language-Image Pre-training (CLIP), all rely on annotated text-to-image pairs for aligning visual and textual modalities. This dependency introduces substantial cost and accuracy requirement in preparing high-quality datasets. At the same time, processing data from two modes also requires dual-tower encoders for most models, which also hinders their lightweight. To address these limitations, we introduce a ``Contrastive Language-Image Pre-training via Large-Language-Model-based Generation (LGCLIP)" framework. LGCLIP leverages a Large Language Model (LLM) to generate class-specific prompts that guide a diffusion model in synthesizing reference images. Afterwards these generated images serve as visual prototypes, and the visual features of real images are extracted and compared with the visual features of these prototypes to achieve comparative prediction. By optimizing prompt generation through the LLM and employing only a visual encoder, LGCLIP remains lightweight and efficient. Crucially, our framework requires only class labels as input during whole experimental procedure, eliminating the need for manually annotated image-text pairs and extra pre-processing. Experimental results validate the feasibility and efficiency of LGCLIP, demonstrating great performance in zero-shot classification tasks and establishing a novel paradigm for classification.

Paperid: 6141, https://arxiv.org/pdf/2512.16907.pdf

Abstract:
Prior works on 3D hand trajectory prediction are constrained by datasets that decouple motion from semantic supervision and by models that weakly link reasoning and action. To address these, we first present the EgoMAN dataset, a large-scale egocentric dataset for interaction stage-aware 3D hand trajectory prediction with 219K 6DoF trajectories and 3M structured QA pairs for semantic, spatial, and motion reasoning. We then introduce the EgoMAN model, a reasoning-to-motion framework that links vision-language reasoning and motion generation via a trajectory-token interface. Trained progressively to align reasoning with motion dynamics, our approach yields accurate and stage-aware trajectories with generalization across real-world scenes.

Paperid: 6142, https://arxiv.org/pdf/2512.16461.pdf

Abstract:
Autonomous robotic systems require spatio-temporal understanding of dynamic environments to ensure reliable navigation and interaction. While Vision-Language Models (VLMs) provide open-world semantic priors, they lack grounding in 3D geometry and temporal dynamics. Conversely, geometric perception captures structure and motion but remains semantically sparse. We propose SNOW (Scene Understanding with Open-World Knowledge), a training-free and backbone-agnostic framework for unified 4D scene understanding that integrates VLM-derived semantics with point cloud geometry and temporal consistency. SNOW processes synchronized RGB images and 3D point clouds, using HDBSCAN clustering to generate object-level proposals that guide SAM2-based segmentation. Each segmented region is encoded through our proposed Spatio-Temporal Tokenized Patch Encoding (STEP), producing multimodal tokens that capture localized semantic, geometric, and temporal attributes. These tokens are incrementally integrated into a 4D Scene Graph (4DSG), which serves as 4D prior for downstream reasoning. A lightweight SLAM backend anchors all STEP tokens spatially in the environment, providing the global reference alignment, and ensuring unambiguous spatial grounding across time. The resulting 4DSG forms a queryable, unified world model through which VLMs can directly interpret spatial scene structure and temporal dynamics. Experiments on a diverse set of benchmarks demonstrate that SNOW enables precise 4D scene understanding and spatially grounded inference, thereby setting new state-of-the-art performance in several settings, highlighting the importance of structured 4D priors for embodied reasoning and autonomous robotics.

Paperid: 6143, https://arxiv.org/pdf/2512.15940.pdf

Abstract:
Humans perceive and reason about their surroundings in four dimensions by building persistent, structured internal representations that encode semantic meaning, spatial layout, and temporal dynamics. These multimodal memories enable them to recall past events, infer unobserved states, and integrate new information into context-dependent reasoning. Inspired by this capability, we introduce R4, a training-free framework for retrieval-augmented reasoning in 4D spatio-temporal space that equips vision-language models (VLMs) with structured, lifelong memory. R4 continuously constructs a 4D knowledge database by anchoring object-level semantic descriptions in metric space and time, yielding a persistent world model that can be shared across agents. At inference, natural language queries are decomposed into semantic, spatial, and temporal keys to retrieve relevant observations, which are integrated into the VLM's reasoning. Unlike classical retrieval-augmented generation methods, retrieval in R4 operates directly in 4D space, enabling episodic and collaborative reasoning without training. Experiments on embodied question answering and navigation benchmarks demonstrate that R4 substantially improves retrieval and reasoning over spatio-temporal information compared to baselines, advancing a new paradigm for embodied 4D reasoning in dynamic environments.

Paperid: 6144, https://arxiv.org/pdf/2512.15804.pdf

Abstract:
Browser rendering bugs can be challenging to detect for browser developers, as they may be triggered by very specific conditions that are exhibited on only a very small subset of websites. Cross-browser inconsistencies (XBIs), variations in how a website is interpreted and displayed on different browsers, can be helpful guides to detect such rendering bugs. Although visual and Document Object Model (DOM)-based analysis techniques exist for detecting XBIs, they often struggle with dynamic and interactive elements. In this study, we discuss our industry experience with using vision language models (VLMs) to identify XBIs. We present the XBIDetective tool which automatically captures screenshots of a website in Mozilla Firefox and Google Chrome, and analyzes them with a VLM for XBIs. We evaluate XBIDetective's performance with an off-the-shelf and a fine-tuned VLM on 1,052 websites. We show that XBIDetective can identify cross-browser discrepancies with 79% accuracy and detect dynamic elements and advertisements with 84% and 85% accuracy, respectively, when using the fine-tuned VLM. We discuss important lessons learned, and we present several potential practical use cases for XBIDetective, including automated regression testing, large-scale monitoring of websites, and rapid triaging of XBI bug reports.

Paperid: 6145, https://arxiv.org/pdf/2512.15372.pdf

Abstract:
Vision transformers in vision-language models apply uniform computational effort across all images, expending 175.33 GFLOPs (ViT-L/14) whether analysing a straightforward product photograph or a complex street scene. We propose ICAR (Image Complexity-Aware Retrieval), which enables vision transformers to use less compute for simple images whilst processing complex images through their full network depth. The key challenge is maintaining cross-modal alignment: embeddings from different processing depths must remain compatible for text matching. ICAR solves this through dual-path training that produces compatible embeddings from both reduced-compute and full-compute processing. This maintains compatibility between image representations and text embeddings in the same semantic space, whether an image exits early or processes fully. Unlike existing two-stage approaches that require expensive reranking, ICAR enables direct image-text matching without additional overhead. To determine how much compute to use, we develop ConvNeXt-IC, which treats image complexity assessment as a classification task. By applying modern classifier backbones rather than specialised architectures, ConvNeXt-IC achieves state-of-the-art performance with 0.959 correlation with human judgement (Pearson) and 4.4x speedup. Evaluated on standard benchmarks augmented with real-world web data, ICAR achieves 20% practical speedup while maintaining category-level performance and 95% of instance-level performance, enabling sustainable scaling of vision-language systems.

Paperid: 6146, https://arxiv.org/pdf/2512.11921.pdf

Abstract:
Vision-Language-Action (VLA) models have demonstrated remarkable capabilities in robotic manipulation,enabling robots to execute natural language commands through end-to-end learning from visual observations.However, deploying large-scale VLA models on affordable robotic platforms remains challenging due to computational constraints and the need for efficient adaptation to new robot embodiments. This paper presents an efficient fine-tuning methodology and real-world deployment analysis for adapting VLA models to low-cost robotic manipulation systems.We propose a resource-efficient fine-tuning strategy using Low-Rank Adaptation (LoRA) and quantization techniques that enable multi-billion parameter VLA models ( 3.1B parameters) to run on consumer-grade GPUs with 8GB VRAM. Our methodology addresses the critical challenge of adapting pre-trained VLA models to new robot embodiments with limited demonstration data, focusing on the trade-offs between frozen and unfrozen vision encoders. Through real-world deployment on the SO101 robotic arm for a button-pressing manipulation task, we demonstrate that our approach achieves effective manipulation performance while maintaining computational efficiency. We provide detailed analysis of deployment challenges, failure modes, and the relationship between training data quantity and real-world performance,trained on 200 demonstration episodes. Our results show that with proper fine-tuning methodology, VLA models can be successfully deployed on affordable robotic platforms,making advanced manipulation capabilities accessible beyond expensive research robots.

Paperid: 6147, https://arxiv.org/pdf/2512.11906.pdf

Abstract:
Automated generation of diagnostic pathology reports directly from whole slide images (WSIs) is an emerging direction in computational pathology. Translating high-resolution tissue patterns into clinically coherent text remains difficult due to large morphological variability and the complex structure of pathology narratives. We introduce MPath, a lightweight multimodal framework that conditions a pretrained biomedical language model (BioBART) on WSI-derived visual embeddings through a learned visual-prefix prompting mechanism. Instead of end-to-end vision-language pretraining, MPath leverages foundation-model WSI features (CONCH + Titan) and injects them into BioBART via a compact projection module, keeping the language backbone frozen for stability and data efficiency. MPath was developed and evaluated on the RED 2025 Grand Challenge dataset and ranked 4th in Test Phase 2, despite limited submission opportunities. The results highlight the potential of prompt-based multimodal conditioning as a scalable and interpretable strategy for pathology report generation.

Paperid: 6148, https://arxiv.org/pdf/2512.11315.pdf

Abstract:
Generalist multimodal agents are expected to unify perception, language, and control - operating robustly across diverse real world domains. However, current evaluation practices remain fragmented across isolated benchmarks, making it difficult to assess whether today's foundation models truly generalize beyond their training distributions. We introduce MultiNet v1.0, a unified benchmark for measuring the cross domain generality of vision language models (VLMs) and vision language action models (VLAs) across six foundational capability regimes. Visual grounding, spatial reasoning, tool use, physical commonsense, multi agent coordination, and continuous robot control. Evaluating GPT 5, Pi0, and Magma, we find that no model demonstrates consistent generality. All exhibit substantial degradation on unseen domains, unfamiliar modalities, or cross domain task shifts despite strong performance within their training distributions.These failures manifest as modality misalignment, output format instability, and catastrophic knowledge degradation under domain transfer.Our findings reveal a persistent gap between the aspiration of generalist intelligence and the actual capabilities of current foundation models.MultiNet v1.0 provides a standardized evaluation substrate for diagnosing these gaps and guiding the development of future generalist agents.Code, data, and leaderboards are publicly available.

Paperid: 6149, https://arxiv.org/pdf/2512.11098.pdf

Abstract:
Many manufacturing environments operate in low-light conditions or within enclosed machines where conventional vision systems struggle. Infrared cameras provide complementary advantages in such environments. Simultaneously, supervised AI systems require large labeled datasets, which makes zero-shot learning frameworks more practical for applications including infrared cameras. Recent advances in vision-language foundation models (VLMs) offer a new path in zero-shot predictions from paired image-text representations. However, current VLMs cannot understand infrared camera data since they are trained on RGB data. This work introduces VLM-IRIS (Vision-Language Models for InfraRed Industrial Sensing), a zero-shot framework that adapts VLMs to infrared data by preprocessing infrared images captured by a FLIR Boson sensor into RGB-compatible inputs suitable for CLIP-based encoders. We demonstrate zero-shot workpiece presence detection on a 3D printer bed where temperature differences between the build plate and workpieces make the task well-suited for thermal imaging. VLM-IRIS converts the infrared images to magma representation and applies centroid prompt ensembling with a CLIP ViT-B/32 encoder to achieve high accuracy on infrared images without any model retraining. These findings demonstrate that the proposed improvements to VLMs can be effectively extended to thermal applications for label-free monitoring.

Paperid: 6150, https://arxiv.org/pdf/2512.07215.pdf

Abstract:
Vision Foundation Models (VFMs) and Vision Language Models (VLMs) have revolutionized computer vision by providing rich semantic and geometric representations. This paper presents a comprehensive visual comparison between CLIP based and DINOv2 based approaches for 3D pose estimation in hand object grasping scenarios. We evaluate both models on the task of 6D object pose estimation and demonstrate their complementary strengths: CLIP excels in semantic understanding through language grounding, while DINOv2 provides superior dense geometric features. Through extensive experiments on benchmark datasets, we show that CLIP based methods achieve better semantic consistency, while DINOv2 based approaches demonstrate competitive performance with enhanced geometric precision. Our analysis provides insights for selecting appropriate vision models for robotic manipulation and grasping, picking applications.

Paperid: 6151, https://arxiv.org/pdf/2512.04395.pdf

Abstract:
Large-scale pre-trained Vision-Language Models (VLMs) have demonstrated strong few-shot learning capabilities. However, these methods typically learn holistic representations where an image's domain-invariant structure is implicitly entangled with its domain-specific style. This presents an opportunity to further enhance generalization by disentangling these visual cues. In this paper, we propose Fourier-Attentive Representation Learning (FARL), a novel framework that addresses this by explicitly disentangling visual representations using Fourier analysis. The core of our method is a dual cross-attention mechanism, where learnable representation tokens separately query an image's structural features (from the phase spectrum) and stylistic features (from the amplitude spectrum). This process yields enriched, disentangled tokens that are then injected deep into the VLM encoders to guide adaptation. Our design, which includes an asymmetric injection strategy, forces the model to learn a more robust vision-language alignment. Extensive experiments on 15 datasets demonstrate the effectiveness of our approach.

Paperid: 6152, https://arxiv.org/pdf/2512.03477.pdf

Abstract:
Vision-language models achieve expert-level performance on medical imaging tasks but exhibit significant diagnostic accuracy disparities across demographic groups. We introduce fairness-aware Low-Rank Adaptation for medical VLMs, combining parameter efficiency with explicit fairness optimization. Our key algorithmic contribution is a differentiable MaxAccGap loss that enables end-to-end optimization of accuracy parity across demographic groups. We propose three methods: FR-LoRA integrates MaxAccGap regularization into the training objective, GR-LoRA applies inverse frequency weighting to balance gradient contributions, and Hybrid-LoRA combines both mechanisms.Evaluated on 10,000 glaucoma fundus images, GR-LoRA reduces diagnostic accuracy disparities by 69% while maintaining 53.15% overall accuracy. Ablation studies reveal that strong regularization strength achieves optimal fairness with minimal accuracy trade-off, and race-specific optimization yields 60% disparity reduction. Our approach requires only 0.24% trainable parameters, enabling practical deployment of fair medical AI in resource-constrained healthcare settings.

Paperid: 6153, https://arxiv.org/pdf/2512.02729.pdf

Abstract:
We introduce Robowheel, a data engine that converts human hand object interaction (HOI) videos into training-ready supervision for cross morphology robotic learning. From monocular RGB or RGB-D inputs, we perform high precision HOI reconstruction and enforce physical plausibility via a reinforcement learning (RL) optimizer that refines hand object relative poses under contact and penetration constraints. The reconstructed, contact rich trajectories are then retargeted to cross-embodiments, robot arms with simple end effectors, dexterous hands, and humanoids, yielding executable actions and rollouts. To scale coverage, we build a simulation-augmented framework on Isaac Sim with diverse domain randomization (embodiments, trajectories, object retrieval, background textures, hand motion mirroring), which enriches the distributions of trajectories and observations while preserving spatial relationships and physical plausibility. The entire data pipeline forms an end to end pipeline from video,reconstruction,retargeting,augmentation data acquisition. We validate the data on mainstream vision language action (VLA) and imitation learning architectures, demonstrating that trajectories produced by our pipeline are as stable as those from teleoperation and yield comparable continual performance gains. To our knowledge, this provides the first quantitative evidence that HOI modalities can serve as effective supervision for robotic learning. Compared with teleoperation, Robowheel is lightweight, a single monocular RGB(D) camera is sufficient to extract a universal, embodiment agnostic motion representation that could be flexibly retargeted across embodiments. We further assemble a large scale multimodal dataset combining multi-camera captures, monocular videos, and public HOI corpora for training and evaluating embodied models.

Paperid: 6154, https://arxiv.org/pdf/2512.01636.pdf

Abstract:
Composed Image Retrieval (CIR) enables fine-grained visual search by combining a reference image with a textual modification. While supervised CIR methods achieve high accuracy, their reliance on costly triplet annotations motivates zero-shot solutions. The core challenge in zero-shot CIR (ZS-CIR) stems from a fundamental dilemma: existing text-centric or diffusion-based approaches struggle to effectively bridge the vision-language modality gap. To address this, we propose Fusion-Diff, a novel generative editing framework with high effectiveness and data efficiency designed for multimodal alignment. First, it introduces a multimodal fusion feature editing strategy within a joint vision-language (VL) space, substantially narrowing the modality gap. Second, to maximize data efficiency, the framework incorporates a lightweight Control-Adapter, enabling state-of-the-art performance through fine-tuning on only a limited-scale synthetic dataset of 200K samples. Extensive experiments on standard CIR benchmarks (CIRR, FashionIQ, and CIRCO) demonstrate that Fusion-Diff significantly outperforms prior zero-shot approaches. We further enhance the interpretability of our model by visualizing the fused multimodal representations.

Paperid: 6155, https://arxiv.org/pdf/2512.00713.pdf

Abstract:
Vision-Language Models (VLMs) have achieved impressive progress in multimodal text generation, yet their rapid adoption raises increasing concerns about security vulnerabilities. Existing backdoor attacks against VLMs primarily rely on explicit pixel-level triggers or imperceptible perturbations injected into images. While effective, these approaches reduce stealthiness and remain vulnerable to image-based defenses. We introduce concept-guided backdoor attacks, a new paradigm that operates at the semantic concept level rather than on raw pixels. We propose two different attacks. The first, Concept-Thresholding Poisoning (CTP), uses explicit concepts in natural images as triggers: only samples containing the target concept are poisoned, causing the model to behave normally in all other cases but consistently inject malicious outputs whenever the concept appears. The second, CBL-Guided Unseen Backdoor (CGUB), leverages a Concept Bottleneck Model (CBM) during training to intervene on internal concept activations, while discarding the CBM branch at inference time to keep the VLM unchanged. This design enables systematic replacement of a targeted label in generated text (for example, replacing "cat" with "dog"), even when the replacement behavior never appears in the training data. Experiments across multiple VLM architectures and datasets show that both CTP and CGUB achieve high attack success rates while maintaining moderate impact on clean-task performance. These findings highlight concept-level vulnerabilities as a critical new attack surface for VLMs.

Paperid: 6156, https://arxiv.org/pdf/2511.20854.pdf

Abstract:
Visual content memorability has intrigued the scientific community for decades, with applications ranging widely, from understanding nuanced aspects of human memory to enhancing content design. A significant challenge in progressing the field lies in the expensive process of collecting memorability annotations from humans. This limits the diversity and scalability of datasets for modeling visual content memorability. Most existing datasets are limited to collecting aggregate memorability scores for visual content, not capturing the nuanced memorability signals present in natural, open-ended recall descriptions. In this work, we introduce the first large-scale unsupervised dataset designed explicitly for modeling visual memorability signals, containing over 82,000 videos, accompanied by descriptive recall data. We leverage tip-of-the-tongue (ToT) retrieval queries from online platforms such as Reddit. We demonstrate that our unsupervised dataset provides rich signals for two memorability-related tasks: recall generation and ToT retrieval. Large vision-language models fine-tuned on our dataset outperform state-of-the-art models such as GPT-4o in generating open-ended memorability descriptions for visual content. We also employ a contrastive training strategy to create the first model capable of performing multimodal ToT retrieval. Our dataset and models present a novel direction, facilitating progress in visual content memorability research.

Paperid: 6157, https://arxiv.org/pdf/2511.19641.pdf

Abstract:
Purpose: To investigate whether a vision-language foundation model can enhance undersampled MRI reconstruction by providing high-level contextual information beyond conventional priors. Methods: We proposed a semantic distribution-guided reconstruction framework that uses a pre-trained vision-language foundation model to encode both the reconstructed image and auxiliary information into high-level semantic features. A contrastive objective aligns the reconstructed representation with the target semantic distribution, ensuring consistency with high-level perceptual cues. The proposed objective works with various deep learning-based reconstruction methods and can flexibly incorporate semantic priors from multimodal sources. To test the effectiveness of these semantic priors, we evaluated reconstruction results guided by priors derived from either image-only or image-language auxiliary information. Results: Experiments on knee and brain datasets demonstrate that semantic priors from images preserve fine anatomical structures and achieve superior perceptual quality, as reflected in lower LPIPS values, higher Tenengrad scores, and improved scores in the reader study, compared with conventional regularization. The image-language information further expands the semantic distribution and enables high-level control over reconstruction attributes. Across all evaluations, the contrastive objective consistently guided the reconstructed features toward the desired semantic distributions while maintaining data fidelity, demonstrating the effectiveness of the proposed optimization framework. Conclusion: The study highlights that vision-language foundation models can improve undersampled MRI reconstruction through semantic-space optimization.

Paperid: 6158, https://arxiv.org/pdf/2511.16233.pdf

Abstract:
The powerful generalization of Vision-Language-Action (VLA) models is bottlenecked by their heavy reliance on massive, redundant, and unevenly valued datasets, hindering their widespread application. Existing model-centric optimization paths, such as model compression (which often leads to performance degradation) or policy distillation (whose products are model-dependent and lack generality), fail to fundamentally address this data-level challenge. To this end, this paper introduces FT-NCFM, a fundamentally different, data-centric generative data distillation framework. Our framework employs a self-contained Fact-Tracing (FT) engine that combines causal attribution with programmatic contrastive verification to assess the intrinsic value of samples. Guided by these assessments, an adversarial NCFM process synthesizes a model-agnostic, information-dense, and reusable data asset. Experimental results on several mainstream VLA benchmarks show that models trained on just 5% of our distilled coreset achieve a success rate of 85-90% compared with training on the full dataset, while reducing training time by over 80%. Our work demonstrates that intelligent data distillation is a highly promising new path for building efficient, high-performance VLA models.

Paperid: 6159, https://arxiv.org/pdf/2511.15333.pdf

Abstract:
Space grounding refers to localizing a set of spatial references described in natural language instructions. Traditional methods often fail to account for complex reasoning -- such as distance, geometry, and inter-object relationships -- while vision-language models (VLMs), despite strong reasoning abilities, struggle to produce a fine-grained region of outputs. To overcome these limitations, we propose C2F-Space, a novel coarse-to-fine space-grounding framework that (i) estimates an approximated yet spatially consistent region using a VLM, then (ii) refines the region to align with the local environment through superpixelization. For the coarse estimation, we design a grid-based visual-grounding prompt with a propose-validate strategy, maximizing VLM's spatial understanding and yielding physically and semantically valid canonical region (i.e., ellipses). For the refinement, we locally adapt the region to surrounding environment without over-relaxed to free space. We construct a new space-grounding benchmark and compare C2F-Space with five state-of-the-art baselines using success rate and intersection-over-union. Our C2F-Space significantly outperforms all baselines. Our ablation study confirms the effectiveness of each module in the two-step process and their synergistic effect of the combined framework. We finally demonstrate the applicability of C2F-Space to simulated robotic pick-and-place tasks.

Paperid: 6160, https://arxiv.org/pdf/2511.12452.pdf

Abstract:
With the rapid adoption of multimodal large language models (MLLMs) across diverse applications, there is a pressing need for task-centered, high-quality training data. A key limitation of current training datasets is their reliance on sparse annotations mined from the Internet or entered via manual typing that capture only a fraction of an image's visual content. Dense annotations are more valuable but remain scarce. Traditional text-based annotation pipelines are poorly suited for creating dense annotations: typing limits expressiveness, slows annotation speed, and underrepresents nuanced visual features, especially in specialized areas such as multicultural imagery and 3D asset annotation. In this paper, we present DenseAnnotate, an audio-driven online annotation platform that enables efficient creation of dense, fine-grained annotations for images and 3D assets. Annotators narrate observations aloud while synchronously linking spoken phrases to image regions or 3D scene parts. Our platform incorporates speech-to-text transcription and region-of-attention marking. To demonstrate the effectiveness of DenseAnnotate, we conducted case studies involving over 1,000 annotators across two domains: culturally diverse images and 3D scenes. We curate a human-annotated multi-modal dataset of 3,531 images, 898 3D scenes, and 7,460 3D objects, with audio-aligned dense annotations in 20 languages, including 8,746 image captions, 2,000 scene captions, and 19,000 object captions. Models trained on this dataset exhibit improvements of 5% in multilingual, 47% in cultural alignment, and 54% in 3D spatial capabilities. Our results show that our platform offers a feasible approach for future vision-language research and can be applied to various tasks and diverse types of data.

Paperid: 6161, https://arxiv.org/pdf/2511.11780.pdf

Abstract:
Recent advances in text-to-image generation have produced strong single-shot models, yet no individual system reliably executes the long, compositional prompts typical of creative workflows. We introduce Image-POSER, a reflective reinforcement learning framework that (i) orchestrates a diverse registry of pretrained text-to-image and image-to-image experts, (ii) handles long-form prompts end-to-end through dynamic task decomposition, and (iii) supervises alignment at each step via structured feedback from a vision-language model critic. By casting image synthesis and editing as a Markov Decision Process, we learn non-trivial expert pipelines that adaptively combine strengths across models. Experiments show that Image-POSER outperforms baselines, including frontier models, across industry-standard and custom benchmarks in alignment, fidelity, and aesthetics, and is consistently preferred in human evaluations. These results highlight that reinforcement learning can endow AI systems with the capacity to autonomously decompose, reorder, and combine visual models, moving towards general-purpose visual assistants.

Paperid: 6162, https://arxiv.org/pdf/2511.10292.pdf

Abstract:
Large Vision-Language Models (LVLMs) often suffer from object hallucination, generating text inconsistent with visual inputs, which can critically undermine their reliability. Existing inference-time interventions to mitigate this issue present a challenging trade-off: while methods that steer internal states or adjust output logits can be effective, they often incur substantial computational overhead, typically requiring extra forward passes. This efficiency bottleneck can limit their practicality for real-world, latency-sensitive deployments. In this work, we aim to address this trade-off with Residual-Update Directed DEcoding Regulation (RUDDER), a low-overhead framework that steers LVLMs towards visually-grounded generation. RUDDER is built on two key innovations: (1) Contextual Activation Residual Direction (CARD) vector, a per-sample visual evidence vector extracted from the residual update of a self-attention layer during a single, standard forward pass. (2) A Bayesian-inspired adaptive gate that performs token-wise injection, applying a corrective signal whose strength is conditioned on the model's deviation from the visual context. Extensive experiments on key hallucination benchmarks, including POPE and CHAIR, indicate that RUDDER achieves performance comparable to state-of-the-art methods while introducing negligible computational latency, validating RUDDER as a pragmatic and effective approach for improving LVLMs' reliability without a significant compromise on efficiency.

Paperid: 6163, https://arxiv.org/pdf/2511.08075.pdf

Abstract:
Latent diffusion models such as Stable Diffusion achieve state-of-the-art results on text-to-image generation tasks. However, the extent to which these models have a semantic understanding of the images they generate is not well understood. In this work, we investigate whether the internal representations used by these models during text-to-image generation contain semantic information that is meaningful to humans. To do so, we perform probing on Stable Diffusion with simple regression layers that predict semantic attributes for objects and evaluate these predictions against human annotations. Surprisingly, we find that this success can actually be attributed to the text encoding occurring in CLIP rather than the reverse diffusion process. We demonstrate that groups of specific semantic attributes have markedly different decoding accuracy than the average, and are thus represented to different degrees. Finally, we show that attributes become more difficult to disambiguate from one another during the inverse diffusion process, further demonstrating the strongest semantic representation of object attributes in CLIP. We conclude that the separately trained CLIP vision-language model is what determines the human-like semantic representation, and that the diffusion process instead takes the role of a visual decoder.

Paperid: 6164, https://arxiv.org/pdf/2511.07171.pdf

Abstract:
Deep learning-based video surveillance increasingly demands privacy-preserving architectures with low computational and environmental overhead. Federated learning preserves privacy but deploying large vision-language models (VLMs) introduces major energy and sustainability challenges. We compare three strategies for federated violence detection under realistic non-IID splits on the RWF-2000 and RLVS datasets: zero-shot inference with pretrained VLMs, LoRA-based fine-tuning of LLaVA-NeXT-Video-7B, and personalized federated learning of a 65.8M-parameter 3D CNN. All methods exceed 90% accuracy in binary violence detection. The 3D CNN achieves superior calibration (ROC AUC 92.59%) at roughly half the energy cost (240 Wh vs. 570 Wh) of federated LoRA, while VLMs provide richer multimodal reasoning. Hierarchical category grouping (based on semantic similarity and class exclusion) boosts VLM multiclass accuracy from 65.31% to 81% on the UCF-Crime dataset. To our knowledge, this is the first comparative simulation study of LoRA-tuned VLMs and personalized CNNs for federated violence detection, with explicit energy and CO2e quantification. Our results inform hybrid deployment strategies that default to efficient CNNs for routine inference and selectively engage VLMs for complex contextual reasoning.

Paperid: 6165, https://arxiv.org/pdf/2511.04948.pdf

Abstract:
The advancement of artificial intelligence in oral healthcare relies on the availability of large-scale multimodal datasets that capture the complexity of clinical practice. In this paper, we present a comprehensive multimodal dataset, comprising 8775 dental checkups from 4800 patients collected over eight years (2018-2025), with patients ranging from 10 to 90 years of age. The dataset includes 50000 intraoral images, 8056 radiographs, and detailed textual records, including diagnoses, treatment plans, and follow-up notes. The data were collected under standard ethical guidelines and annotated for benchmarking. To demonstrate its utility, we fine-tuned state-of-the-art large vision-language models, Qwen-VL 3B and 7B, and evaluated them on two tasks: classification of six oro-dental anomalies and generation of complete diagnostic reports from multimodal inputs. We compared the fine-tuned models with their base counterparts and GPT-4o. The fine-tuned models achieved substantial gains over these baselines, validating the dataset and underscoring its effectiveness in advancing AI-driven oro-dental healthcare solutions. The dataset is publicly available, providing an essential resource for future research in AI dentistry.

Paperid: 6166, https://arxiv.org/pdf/2511.04835.pdf

Abstract:
Sampling-based motion planners (SBMPs) are widely used to compute dynamically feasible robot paths. However, their reliance on uniform sampling often leads to poor efficiency and slow planning in complex environments. We introduce a novel non-uniform sampling strategy that integrates into existing SBMPs by biasing sampling toward `certified' regions. These regions are constructed by (i) generating an initial, possibly infeasible, path using any heuristic path predictor (e.g., A* or vision-language models) and (ii) applying conformal prediction to quantify the predictor's uncertainty. This process yields prediction sets around the initial-guess path that are guaranteed, with user-specified probability, to contain the optimal solution. To our knowledge, this is the first non-uniform sampling approach for SBMPs that provides such probabilistically correct guarantees on the sampling regions. Extensive evaluations demonstrate that our method consistently finds feasible paths faster and generalizes better to unseen environments than existing baselines.

Paperid: 6167, https://arxiv.org/pdf/2511.02495.pdf

Abstract:
Recent advances in multi-modal models have demonstrated strong performance in tasks such as image generation and reasoning. However, applying these models to the fire domain remains challenging due to the lack of publicly available datasets with high-quality fire domain annotations. To address this gap, we introduce DetectiumFire, a large-scale, multi-modal dataset comprising of 22.5k high-resolution fire-related images and 2.5k real-world fire-related videos covering a wide range of fire types, environments, and risk levels. The data are annotated with both traditional computer vision labels (e.g., bounding boxes) and detailed textual prompts describing the scene, enabling applications such as synthetic data generation and fire risk reasoning. DetectiumFire offers clear advantages over existing benchmarks in scale, diversity, and data quality, significantly reducing redundancy and enhancing coverage of real-world scenarios. We validate the utility of DetectiumFire across multiple tasks, including object detection, diffusion-based image generation, and vision-language reasoning. Our results highlight the potential of this dataset to advance fire-related research and support the development of intelligent safety systems. We release DetectiumFire to promote broader exploration of fire understanding in the AI community. The dataset is available at https://kaggle.com/datasets/38b79c344bdfc55d1eed3d22fbaa9c31fad45e27edbbe9e3c529d6e5c4f93890

Paperid: 6168, https://arxiv.org/pdf/2511.01224.pdf

Abstract:
Vision-language-action (VLA) models have significantly advanced robotic learning, enabling training on large-scale, cross-embodiment data and fine-tuning for specific robots. However, state-of-the-art autoregressive VLAs struggle with multi-robot collaboration. We introduce embodiment transfer learning, denoted as ET-VLA, a novel framework for efficient and effective transfer of pre-trained VLAs to multi-robot. ET-VLA's core is Synthetic Continued Pretraining (SCP), which uses synthetically generated data to warm up the model for the new embodiment, bypassing the need for real human demonstrations and reducing data collection costs. SCP enables the model to learn correct actions and precise action token numbers. Following SCP, the model is fine-tuned on target embodiment data. To further enhance the model performance on multi-embodiment, we present the Embodied Graph-of-Thought technique, a novel approach that formulates each sub-task as a node, that allows the VLA model to distinguish the functionalities and roles of each embodiment during task execution. Our work considers bimanual robots, a simple version of multi-robot to verify our approaches. We validate the effectiveness of our method on both simulation benchmarks and real robots covering three different bimanual embodiments. In particular, our proposed ET-VLA \space can outperform OpenVLA on six real-world tasks over 53.2%. We will open-source all codes to support the community in advancing VLA models for robot learning.

Paperid: 6169, https://arxiv.org/pdf/2511.00524.pdf

Abstract:
Video Anomaly Detection (VAD) aims to identify anomalous events within video segments. In scenarios such as surveillance or industrial process monitoring, anomaly detection is of critical importance. While existing approaches are semi-automated, requiring human assessment for anomaly detection, traditional VADs offer limited output as either normal or anomalous. We propose Text-guided Fine-Grained Video Anomaly Detection (T-VAD), a framework built upon Large Vision-Language Model (LVLM). T-VAD introduces an Anomaly Heatmap Decoder (AHD) that performs pixel-wise visual-textual feature alignment to generate fine-grained anomaly heatmaps. Furthermore, we design a Region-aware Anomaly Encoder (RAE) that transforms the heatmaps into learnable textual embeddings, guiding the LVLM to accurately identify and localize anomalous events in videos. This significantly enhances both the granularity and interactivity of anomaly detection. The proposed method achieving SOTA performance by demonstrating 94.8% Area Under the Curve (AUC, specifically micro-AUC) and 67.8%/76.7% accuracy in anomaly heatmaps (RBDC/TBDC) on the UBnormal dataset, and subjectively verified more preferable textual description on the ShanghaiTech-based dataset (BLEU-4: 62.67 for targets, 88.84 for trajectories; Yes/No accuracy: 97.67%), and on the UBnormal dataset (BLEU-4: 50.32 for targets, 78.10 for trajectories; Yes/No accuracy: 89.73%).

Paperid: 6170, https://arxiv.org/pdf/2511.00139.pdf

Abstract:
Achieving human-like dexterous manipulation remains a major challenge for general-purpose robots. While Vision-Language-Action (VLA) models show potential in learning skills from demonstrations, their scalability is limited by scarce high-quality training data. Existing data collection methods face inherent constraints: manual teleoperation overloads human operators, while automated planning often produces unnatural motions. We propose a Shared Autonomy framework that divides control between macro and micro motions. A human operator guides the robot's arm pose through intuitive VR teleoperation, while an autonomous DexGrasp-VLA policy handles fine-grained hand control using real-time tactile and visual feedback. This division significantly reduces cognitive load and enables efficient collection of high-quality coordinated arm-hand demonstrations. Using this data, we train an end-to-end VLA policy enhanced with our novel Arm-Hand Feature Enhancement module, which captures both distinct and shared representations of macro and micro movements for more natural coordination. Our Corrective Teleoperation system enables continuous policy improvement through human-in-the-loop failure recovery. Experiments demonstrate that our framework generates high-quality data with minimal manpower and achieves a 90% success rate across diverse objects, including unseen instances. Comprehensive evaluations validate the system's effectiveness in developing dexterous manipulation capabilities.

Paperid: 6171, https://arxiv.org/pdf/2511.00067.pdf

Abstract:
The objective of domain generalization (DG) is to enable models to be robust against domain shift. DG is crucial for deploying vision-language models (VLMs) in real-world applications, yet most existing methods rely on domain labels that may not be available and often ambiguous. We instead study the DG setting where models must generalize well without access to explicit domain labels. Our key idea is to represent an unseen target domain as a combination of latent domains automatically discovered from training data, enabling the model to adaptively transfer knowledge across domains. To realize this, we perform latent domain clustering on image features and fuse domain-specific text features based on the similarity between the input image and each latent domain. Experiments on four benchmarks show that this strategy yields consistent gains over VLM-based baselines and provides new insights into improving robustness under domain shift.

Paperid: 6172, https://arxiv.org/pdf/2510.26241.pdf

Abstract:
Modern vision-language models (VLMs) excel at many multimodal tasks, yet their grasp of temporal information in video remains weak and, crucially, under-evaluated. We probe this gap with a deceptively simple but revealing challenge: judging the arrow of time (AoT)-whether a short clip is played forward or backward. We introduce AoT-PsyPhyBENCH, a psychophysically validated benchmark that tests whether VLMs can infer temporal direction in natural videos using the same stimuli and behavioral baselines established for humans. Our comprehensive evaluation of open-weight and proprietary, reasoning and non-reasoning VLMs reveals that most models perform near chance, and even the best lag far behind human accuracy on physically irreversible processes (e.g., free fall, diffusion/explosion) and causal manual actions (division/addition) that humans recognize almost instantly. These results highlight a fundamental gap in current multimodal systems: while they capture rich visual-semantic correlations, they lack the inductive biases required for temporal continuity and causal understanding. We release the code and data for AoT-PsyPhyBENCH to encourage further progress in the physical and temporal reasoning capabilities of VLMs.

Paperid: 6173, https://arxiv.org/pdf/2510.26151.pdf

Abstract:
Large annotated datasets are essential for training robust Computer-Aided Diagnosis (CAD) models for breast cancer detection or risk prediction. However, acquiring such datasets with fine-detailed annotation is both costly and time-consuming. Vision-Language Models (VLMs), such as CLIP, which are pre-trained on large image-text pairs, offer a promising solution by enhancing robustness and data efficiency in medical imaging tasks. This paper introduces a novel Multi-View Mammography and Language Model for breast cancer classification and risk prediction, trained on a dataset of paired mammogram images and synthetic radiology reports. Our MV-MLM leverages multi-view supervision to learn rich representations from extensive radiology data by employing cross-modal self-supervision across image-text pairs. This includes multiple views and the corresponding pseudo-radiology reports. We propose a novel joint visual-textual learning strategy to enhance generalization and accuracy performance over different data types and tasks to distinguish breast tissues or cancer characteristics(calcification, mass) and utilize these patterns to understand mammography images and predict cancer risk. We evaluated our method on both private and publicly available datasets, demonstrating that the proposed model achieves state-of-the-art performance in three classification tasks: (1) malignancy classification, (2) subtype classification, and (3) image-based cancer risk prediction. Furthermore, the model exhibits strong data efficiency, outperforming existing fully supervised or VLM baselines while trained on synthetic text reports and without the need for actual radiology reports.

Paperid: 6174, https://arxiv.org/pdf/2510.25051.pdf

Abstract:
Breast cancer remains the most commonly diagnosed malignancy among women in the developed world. Early detection through mammography screening plays a pivotal role in reducing mortality rates. While computer-aided diagnosis (CAD) systems have shown promise in assisting radiologists, existing approaches face critical limitations in clinical deployment - particularly in handling the nuanced interpretation of multi-modal data and feasibility due to the requirement of prior clinical history. This study introduces a novel framework that synergistically combines visual features from 2D mammograms with structured textual descriptors derived from easily accessible clinical metadata and synthesized radiological reports through innovative tokenization modules. Our proposed methods in this study demonstrate that strategic integration of convolutional neural networks (ConvNets) with language representations achieves superior performance to vision transformer-based models while handling high-resolution images and enabling practical deployment across diverse populations. By evaluating it on multi-national cohort screening mammograms, our multi-modal approach achieves superior performance in cancer detection and calcification identification compared to unimodal baselines, with particular improvements. The proposed method establishes a new paradigm for developing clinically viable VLM-based CAD systems that effectively leverage imaging data and contextual patient information through effective fusion mechanisms.

Paperid: 6175, https://arxiv.org/pdf/2510.23775.pdf

Abstract:
The increasing realism of AI-generated imagery poses challenges for verifying visual authenticity. We present an explainable image authenticity detection system that combines a lightweight convolutional classifier ("Faster-Than-Lies") with a Vision-Language Model (Qwen2-VL-7B) to classify, localize, and explain artifacts in 32x32 images. Our model achieves 96.5% accuracy on the extended CiFAKE dataset augmented with adversarial perturbations and maintains an inference time of 175ms on 8-core CPUs, enabling deployment on local or edge devices. Using autoencoder-based reconstruction error maps, we generate artifact localization heatmaps, which enhance interpretability for both humans and the VLM. We further categorize 70 visual artifact types into eight semantic groups and demonstrate explainable text generation for each detected anomaly. This work highlights the feasibility of combining visual and linguistic reasoning for interpretable authenticity detection in low-resolution imagery and outlines potential cross-domain applications in forensics, industrial inspection, and social media moderation.

Paperid: 6176, https://arxiv.org/pdf/2510.22917.pdf

Abstract:
Objective-oriented navigation(ObjNav) enables robot to navigate to target object directly and autonomously in an unknown environment. Effective perception in navigation in unknown environment is critical for autonomous robots. While egocentric observations from RGB-D sensors provide abundant local information, real-time top-down maps offer valuable global context for ObjNav. Nevertheless, the majority of existing studies focus on a single source, seldom integrating these two complementary perceptual modalities, despite the fact that humans naturally attend to both. With the rapid advancement of Vision-Language Models(VLMs), we propose Hybrid Perception Navigation (HyPerNav), leveraging VLMs' strong reasoning and vision-language understanding capabilities to jointly perceive both local and global information to enhance the effectiveness and intelligence of navigation in unknown environments. In both massive simulation evaluation and real-world validation, our methods achieved state-of-the-art performance against popular baselines. Benefiting from hybrid perception approach, our method captures richer cues and finds the objects more effectively, by simultaneously leveraging information understanding from egocentric observations and the top-down map. Our ablation study further proved that either of the hybrid perception contributes to the navigation performance.

Paperid: 6177, https://arxiv.org/pdf/2510.21740.pdf

Abstract:
Data visualizations are vital components of many scientific articles and news stories. Current vision-language models (VLMs) still struggle on basic data visualization understanding tasks, but the causes of failure remain unclear. Are VLM failures attributable to limitations in how visual information in the data visualization is encoded, how information is transferred between the vision and language modules, or how information is processed within the language module? We developed FUGU, a suite of data visualization understanding tasks, to precisely characterize potential sources of difficulty (e.g., extracting the position of data points, distances between them, and other summary statistics). We used FUGU to investigate three widely used VLMs. To diagnose the sources of errors produced by these models, we used activation patching and linear probes to trace information flow through models across a variety of prompting strategies. We found that some models fail to generate the coordinates of individual data points correctly, and these initial errors often lead to erroneous final responses. When these models are provided with the correct coordinates, performance improves substantially. Moreover, even when the model generates an incorrect response, the correct coordinates can be successfully read out from the latent representations in the vision encoder, suggesting that the source of these errors lies in the vision-language handoff. We further found that while providing correct coordinates helps with tasks involving one or a small number of data points, it generally worsens performance for tasks that require extracting statistical relationships across many data points. Fine-tuning models on FUGU also fails to yield ceiling performance. These findings point to architectural constraints in current VLMs that might pose significant challenges for reliable data visualization understanding.

Paperid: 6178, https://arxiv.org/pdf/2510.19949.pdf

Authors:Mathieu Andreux, Märt Bakler, Yanael Barbier, Hamza Benchekroun, Emilien Biré, Antoine Bonnet, Riaz Bordie, Nathan Bout, Matthias Brunel, Aleix Cambray, Pierre-Louis Cedoz, Antoine Chassang, Gautier Cloix, Ethan Connelly, Alexandra Constantinou, Ramzi De Coster, Hubert de la Jonquiere, Aurélien Delfosse, Maxime Delpit, Alexis Deprez, Augustin Derupti, Mathieu Diaz, Shannon D'Souza, Julie Dujardin, Abai Edmund, Michael Eickenberg, Armand Fatalot, Wissem Felissi, Isaac Herring, Xavier Koegler, Erwan Le Jumeau de Kergaradec, Aurélien Lac, Maxime Langevin, Corentin Lauverjat, Antonio Loison, Avshalom Manevich, Axel Moyal, Axel Nguyen Kerbel, Marinela Parovic, Julien Revelle, Guillaume Richard, Mats Richter, Ronan Riochet, María Santos, Romain Savidan, Laurent Sifre, Maxime Theillard, Marc Thibault, Ivan Valentini, Tony Wu, Laura Yie, Kai Yuan, Jevgenij Zubovskij

Abstract:
Building agents that generalize across web, desktop, and mobile environments remains an open challenge, as prior systems rely on environment-specific interfaces that limit cross-platform deployment. We introduce Surfer 2, a unified architecture operating purely from visual observations that achieves state-of-the-art performance across all three environments. Surfer 2 integrates hierarchical context management, decoupled planning and execution, and self-verification with adaptive recovery, enabling reliable operation over long task horizons. Our system achieves 97.1% accuracy on WebVoyager, 69.6% on WebArena, 60.1% on OSWorld, and 87.1% on AndroidWorld, outperforming all prior systems without task-specific fine-tuning. With multiple attempts, Surfer 2 exceeds human performance on all benchmarks. These results demonstrate that systematic orchestration amplifies foundation model capabilities and enables general-purpose computer control through visual interaction alone, while calling for a next-generation vision language model to achieve Pareto-optimal cost-efficiency.

Paperid: 6179, https://arxiv.org/pdf/2510.17651.pdf

Abstract:
We examine frugal federated learning approaches to violence detection by comparing two complementary strategies: (i) zero-shot and federated fine-tuning of vision-language models (VLMs), and (ii) personalized training of a compact 3D convolutional neural network (CNN3D). Using LLaVA-7B and a 65.8M parameter CNN3D as representative cases, we evaluate accuracy, calibration, and energy usage under realistic non-IID settings. Both approaches exceed 90% accuracy. CNN3D slightly outperforms Low-Rank Adaptation(LoRA)-tuned VLMs in ROC AUC and log loss, while using less energy. VLMs remain favorable for contextual reasoning and multimodal inference. We quantify energy and CO$_2$ emissions across training and inference, and analyze sustainability trade-offs for deployment. To our knowledge, this is the first comparative study of LoRA-tuned vision-language models and personalized CNNs for federated violence detection, with an emphasis on energy efficiency and environmental metrics. These findings support a hybrid model: lightweight CNNs for routine classification, with selective VLM activation for complex or descriptive scenarios. The resulting framework offers a reproducible baseline for responsible, resource-aware AI in video surveillance, with extensions toward real-time, multimodal, and lifecycle-aware systems.

Paperid: 6180, https://arxiv.org/pdf/2510.17590.pdf

Abstract:
Misinformation spreads across web platforms through billions of daily multimodal posts that combine text and images, overwhelming manual fact-checking capacity. Supervised detection models require domain-specific training data and fail to generalize across diverse manipulation tactics. We present MIRAGE, an inference-time, model-pluggable agentic framework that decomposes multimodal verification into four sequential modules: visual veracity assessment detects AI-generated images, cross-modal consistency analysis identifies out-of-context repurposing, retrieval-augmented factual checking grounds claims in web evidence through iterative question generation, and a calibrated judgment module integrates all signals. MIRAGE orchestrates vision-language model reasoning with targeted web retrieval, outputs structured and citation-linked rationales. On MMFakeBench validation set (1,000 samples), MIRAGE with GPT-4o-mini achieves 81.65% F1 and 75.1% accuracy, outperforming the strongest zero-shot baseline (GPT-4V with MMD-Agent at 74.0% F1) by 7.65 points while maintaining 34.3% false positive rate versus 97.3% for a judge-only baseline. Test set results (5,000 samples) confirm generalization with 81.44% F1 and 75.08% accuracy. Ablation studies show visual verification contributes 5.18 F1 points and retrieval-augmented reasoning contributes 2.97 points. Our results demonstrate that decomposed agentic reasoning with web retrieval can match supervised detector performance without domain-specific training, enabling misinformation detection across modalities where labeled data remains scarce.

Paperid: 6181, https://arxiv.org/pdf/2510.16924.pdf

Abstract:
Despite significant progress in multimodal language models (LMs), it remains unclear whether visual grounding enhances their understanding of embodied knowledge compared to text-only models. To address this question, we propose a novel embodied knowledge understanding benchmark based on the perceptual theory from psychology, encompassing visual, auditory, tactile, gustatory, olfactory external senses, and interoception. The benchmark assesses the models' perceptual abilities across different sensory modalities through vector comparison and question-answering tasks with over 1,700 questions. By comparing 30 state-of-the-art LMs, we surprisingly find that vision-language models (VLMs) do not outperform text-only models in either task. Moreover, the models perform significantly worse in the visual dimension compared to other sensory dimensions. Further analysis reveals that the vector representations are easily influenced by word form and frequency, and the models struggle to answer questions involving spatial perception and reasoning. Our findings underscore the need for more effective integration of embodied knowledge in LMs to enhance their understanding of the physical world.

Paperid: 6182, https://arxiv.org/pdf/2510.16017.pdf

Abstract:
Infrastructure in smart cities is increasingly monitored by networks of closed circuit television (CCTV) cameras. Roads, bridges and tunnels develop cracks, potholes, and fluid leaks that threaten public safety and require timely repair. Manual inspection is costly and hazardous, and existing automatic systems typically address individual defect types or provide unstructured outputs that cannot directly guide maintenance crews. This paper proposes a comprehensive pipeline that leverages street CCTV streams for multi defect detection and segmentation using the YOLO family of object detectors and passes the detections to a vision language model (VLM) for scene aware summarization. The VLM generates a structured action plan in JSON format that includes incident descriptions, recommended tools, dimensions, repair plans, and urgent alerts. We review literature on pothole, crack and leak detection, highlight recent advances in large vision language models such as QwenVL and LLaVA, and describe the design of our early prototype. Experimental evaluation on public datasets and captured CCTV clips demonstrates that the system accurately identifies diverse defects and produces coherent summaries. We conclude by discussing challenges and directions for scaling the system to city wide deployments.

Paperid: 6183, https://arxiv.org/pdf/2510.15866.pdf

Abstract:
The clinical adoption of biomedical vision-language models is hindered by prompt optimization techniques that produce either uninterpretable latent vectors or single textual prompts. This lack of transparency and failure to capture the multi-faceted nature of clinical diagnosis, which relies on integrating diverse observations, limits their trustworthiness in high-stakes settings. To address this, we introduce BiomedXPro, an evolutionary framework that leverages a large language model as both a biomedical knowledge extractor and an adaptive optimizer to automatically generate a diverse ensemble of interpretable, natural-language prompt pairs for disease diagnosis. Experiments on multiple biomedical benchmarks show that BiomedXPro consistently outperforms state-of-the-art prompt-tuning methods, particularly in data-scarce few-shot settings. Furthermore, our analysis demonstrates a strong semantic alignment between the discovered prompts and statistically significant clinical features, grounding the model's performance in verifiable concepts. By producing a diverse ensemble of interpretable prompts, BiomedXPro provides a verifiable basis for model predictions, representing a critical step toward the development of more trustworthy and clinically-aligned AI systems.

Paperid: 6184, https://arxiv.org/pdf/2510.13896.pdf

Abstract:
Cellular image segmentation is essential for quantitative biology yet remains difficult due to heterogeneous modalities, morphological variability, and limited annotations. We present GenCellAgent, a training-free multi-agent framework that orchestrates specialist segmenters and generalist vision-language models via a planner-executor-evaluator loop (choose tool $\rightarrow$ run $\rightarrow$ quality-check) with long-term memory. The system (i) automatically routes images to the best tool, (ii) adapts on the fly using a few reference images when imaging conditions differ from what a tool expects, (iii) supports text-guided segmentation of organelles not covered by existing models, and (iv) commits expert edits to memory, enabling self-evolution and personalized workflows. Across four cell-segmentation benchmarks, this routing yields a 15.7\% mean accuracy gain over state-of-the-art baselines. On endoplasmic reticulum and mitochondria from new datasets, GenCellAgent improves average IoU by 37.6\% over specialist models. It also segments novel objects such as the Golgi apparatus via iterative text-guided refinement, with light human correction further boosting performance. Together, these capabilities provide a practical path to robust, adaptable cellular image segmentation without retraining, while reducing annotation burden and matching user preferences.

Paperid: 6185, https://arxiv.org/pdf/2510.13251.pdf

Abstract:
Video Large Language Models (VideoLLMs) extend the capabilities of vision-language models to spatiotemporal inputs, enabling tasks such as video question answering (VideoQA). Despite recent advances in VideoLLMs, their internal mechanisms on where and how they extract and propagate video and textual information remain less explored. In this study, we investigate the internal information flow of VideoLLMs using mechanistic interpretability techniques. Our analysis reveals consistent patterns across diverse VideoQA tasks: (1) temporal reasoning in VideoLLMs initiates with active cross-frame interactions in early-to-middle layers, (2) followed by progressive video-language integration in middle layers. This is facilitated by alignment between video representations and linguistic embeddings containing temporal concepts. (3) Upon completion of this integration, the model is ready to generate correct answers in middle-to-late layers. (4) Based on our analysis, we show that VideoLLMs can retain their VideoQA performance by selecting these effective information pathways while suppressing a substantial amount of attention edges, e.g., 58% in LLaVA-NeXT-7B-Video-FT. These findings provide a blueprint on how VideoLLMs perform temporal reasoning and offer practical insights for improving model interpretability and downstream generalization. Our project page with the source code is available at https://map-the-flow.github.io

Paperid: 6186, https://arxiv.org/pdf/2510.13235.pdf

Abstract:
Multimodal semantic cues, such as textual descriptions, have shown strong potential in enhancing target perception for tracking. However, existing methods rely on static textual descriptions from large language models, which lack adaptability to real-time target state changes and prone to hallucinations. To address these challenges, we propose a unified multimodal vision-language tracking framework, named EPIPTrack, which leverages explicit and implicit prompts for dynamic target modeling and semantic alignment. Specifically, explicit prompts transform spatial motion information into natural language descriptions to provide spatiotemporal guidance. Implicit prompts combine pseudo-words with learnable descriptors to construct individualized knowledge representations capturing appearance attributes. Both prompts undergo dynamic adjustment via the CLIP text encoder to respond to changes in target state. Furthermore, we design a Discriminative Feature Augmentor to enhance visual and cross-modal representations. Extensive experiments on MOT17, MOT20, and DanceTrack demonstrate that EPIPTrack outperforms existing trackers in diverse scenarios, exhibiting robust adaptability and superior performance.

Paperid: 6187, https://arxiv.org/pdf/2510.11036.pdf

Abstract:
Most robotic grasping methods are typically designed for single gripper types, which limits their applicability in real-world scenarios requiring diverse end-effectors. We propose XGrasp, a real-time gripper-aware grasp detection framework that efficiently handles multiple gripper configurations. The proposed method addresses data scarcity by systematically augmenting existing datasets with multi-gripper annotations. XGrasp employs a hierarchical two-stage architecture. In the first stage, a Grasp Point Predictor (GPP) identifies optimal locations using global scene information and gripper specifications. In the second stage, an Angle-Width Predictor (AWP) refines the grasp angle and width using local features. Contrastive learning in the AWP module enables zero-shot generalization to unseen grippers by learning fundamental grasping characteristics. The modular framework integrates seamlessly with vision foundation models, providing pathways for future vision-language capabilities. The experimental results demonstrate competitive grasp success rates across various gripper types, while achieving substantial improvements in inference speed compared to existing gripper-aware methods. Project page: https://sites.google.com/view/xgrasp

Paperid: 6188, https://arxiv.org/pdf/2510.10560.pdf

Abstract:
Cross-attention transformers and other multimodal vision-language models excel at grounding and generation; however, their extensive, full-precision backbones make it challenging to deploy them on edge devices. Memory-augmented architectures enhance the utilization of past context; however, most works rarely pair them with aggressive edge-oriented quantization. We introduce BitMar, a quantized multimodal transformer that proposes an external human-like episodic memory for effective image-text generation on hardware with limited resources. BitMar utilizes 1.58-bit encoders, one for text (BitNet-style) and one for vision (DiNOv2-based), to create compact embeddings that are combined and used to query a fixed-size key-value episodic memory. During vector retrieval, the BitNet decoder applies per-layer conditioning, which increases the contextual relevance of generated content. The decoder also employs attention sinks with a sliding-window mechanism to process long or streaming inputs under tight memory budgets. The combination of per-layer conditioning and sliding-window attention achieves a strong quality-speed trade-off, delivering competitive captioning and multimodal understanding at low latency with a small model footprint. These characteristics make BitMar well-suited for edge deployment.

Paperid: 6189, https://arxiv.org/pdf/2510.08919.pdf

Abstract:
Vision-language models have achieved remarkable success in multi-modal representation learning from large-scale pairs of visual scenes and linguistic descriptions. However, they still struggle to simultaneously express two distinct types of semantic structures: the hierarchy within a concept family (e.g., dog $\preceq$ mammal $\preceq$ animal) and the compositionality across different concept families (e.g., "a dog in a car" $\preceq$ dog, car). Recent works have addressed this challenge by employing hyperbolic space, which efficiently captures tree-like hierarchy, yet its suitability for representing compositionality remains unclear. To resolve this dilemma, we propose PHyCLIP, which employs an $\ell_1$-Product metric on a Cartesian product of Hyperbolic factors. With our design, intra-family hierarchies emerge within individual hyperbolic factors, and cross-family composition is captured by the $\ell_1$-product metric, analogous to a Boolean algebra. Experiments on zero-shot classification, retrieval, hierarchical classification, and compositional understanding tasks demonstrate that PHyCLIP outperforms existing single-space approaches and offers more interpretable structures in the embedding space.

Paperid: 6190, https://arxiv.org/pdf/2510.08791.pdf

Abstract:
Medical Visual Question Answering (Med-VQA) is a challenging task that requires a deep understanding of both medical images and textual questions. Although recent works leveraging Medical Vision-Language Pre-training (Med-VLP) have shown strong performance on the Med-VQA task, there is still no unified solution for modality alignment, and the issue of hard negatives remains under-explored. Additionally, commonly used knowledge fusion techniques for Med-VQA may introduce irrelevant information. In this work, we propose a framework to address these challenges through three key contributions: (1) a unified solution for heterogeneous modality alignments across multiple levels, modalities, views, and stages, leveraging methods like contrastive learning and optimal transport theory; (2) a hard negative mining method that employs soft labels for multi-modality alignments and enforces the hard negative pair discrimination; and (3) a Gated Cross-Attention Module for Med-VQA that integrates the answer vocabulary as prior knowledge and selects relevant information from it. Our framework outperforms the previous state-of-the-art on widely used Med-VQA datasets like RAD-VQA, SLAKE, PathVQA and VQA-2019.

Paperid: 6191, https://arxiv.org/pdf/2510.07098.pdf

Abstract:
Table Visual Question Answering (Table VQA) is typically addressed by large vision-language models (VLMs). While such models can answer directly from images, they often miss fine-grained details unless scaled to very large sizes, which are computationally prohibitive, especially for mobile deployment. A lighter alternative is to have a small VLM perform OCR and then use a large language model (LLM) to reason over structured outputs such as Markdown tables. However, these representations are not naturally optimized for LLMs and still introduce substantial errors. We propose TALENT (Table VQA via Augmented Language-Enhanced Natural-text Transcription), a lightweight framework that leverages dual representations of tables. TALENT prompts a small VLM to produce both OCR text and natural language narration, then combines them with the question for reasoning by an LLM. This reframes Table VQA as an LLM-centric multimodal reasoning task, where the VLM serves as a perception-narration module rather than a monolithic solver. Additionally, we construct ReTabVQA, a more challenging Table VQA dataset requiring multi-step quantitative reasoning over table images. Experiments show that TALENT enables a small VLM-LLM combination to match or surpass a single large VLM at significantly lower computational cost on both public datasets and ReTabVQA.

Paperid: 6192, https://arxiv.org/pdf/2510.06235.pdf

Abstract:
We present our submission to the Algonauts 2025 Challenge, where the goal is to predict fMRI brain responses to movie stimuli. Our approach integrates multimodal representations from large language models, video encoders, audio models, and vision-language models, combining both off-the-shelf and fine-tuned variants. To improve performance, we enhanced textual inputs with detailed transcripts and summaries, and we explored stimulus-tuning and fine-tuning strategies for language and vision models. Predictions from individual models were combined using stacked regression, yielding solid results. Our submission, under the team name Seinfeld, ranked 10th. We make all code and resources publicly available, contributing to ongoing efforts in developing multimodal encoding models for brain activity.

Paperid: 6193, https://arxiv.org/pdf/2510.04477.pdf

Abstract:
Bridging clinical diagnostic reasoning with AI remains a central challenge in medical imaging. We introduce MedCLM, an automated pipeline that converts detection datasets into large-scale medical visual question answering (VQA) data with Chain-of-Thought (CoT) reasoning by linking lesion boxes to organ segmentation and structured rationales. These contextual signals enable medical vision-language models to generate question-answer pairs with step-by-step reasoning. To utilize this data effectively, we propose an Integrated CoT-Curriculum Strategy composed of an Easy stage with explicit lesion boxes for visual grounding, a Medium stage that encourages implicit localization, and a Hard stage for weakly supervised reasoning. Experimental results demonstrate that MedCLM attains state-of-the-art performance on several medical VQA benchmarks, providing a scalable framework for developing clinically aligned medical vision-language models.

Paperid: 6194, https://arxiv.org/pdf/2510.04041.pdf

Abstract:
Learning robust robotic control policies remains a major challenge due to the high cost of collecting labeled data, limited generalization to unseen environments, and difficulties in planning over long horizons. While Vision-Language-Action (VLA) models offer a promising solution by grounding natural language instructions into single-step control commands, they often lack mechanisms for lookahead and struggle with compounding errors in dynamic tasks. In this project, we introduce Scaling Inference-Time COMpute for VLAs (SITCOM), a framework that augments any pretrained VLA with model-based rollouts and reward-based trajectory selection, inspired by Model Predictive Control algorithm. SITCOM leverages a learned dynamics model to simulate multi-step action rollouts to select the best candidate plan for real-world execution, transforming one-shot VLAs into robust long-horizon planners. We develop an efficient transformer-based dynamics model trained on large-scale BridgeV2 data and fine-tuned on SIMPLER environments to bridge the Real2Sim gap, and score candidate rollouts using rewards from simulator. Through comprehensive evaluation across multiple tasks and settings in the SIMPLER environment, we demonstrate that SITCOM when combined with a good reward function can significantly improve task completion rate from 48% to 72% using trained dynamics model.

Paperid: 6195, https://arxiv.org/pdf/2510.03543.pdf

Abstract:
Endoscopic procedures such as esophagogastroduodenoscopy (EGD) and colonoscopy play a critical role in diagnosing and managing gastrointestinal (GI) disorders. However, the documentation burden associated with these procedures place significant strain on gastroenterologists, contributing to inefficiencies in clinical workflows and physician burnout. To address this challenge, we propose a novel automated report generation model that leverages a transformer-based vision encoder and text decoder within a two-stage training framework. In the first stage, both components are pre-trained on image/text caption pairs to capture generalized vision-language features, followed by fine-tuning on images/report pairs to generate clinically meaningful findings. Our approach not only streamlines the documentation process but also holds promise for reducing physician workload and improving patient care.

Paperid: 6196, https://arxiv.org/pdf/2510.03295.pdf

Abstract:
We present VLCAP, an Arabic image captioning framework that integrates CLIP-based visual label retrieval with multimodal text generation. Rather than relying solely on end-to-end captioning, VLCAP grounds generation in interpretable Arabic visual concepts extracted with three multilingual encoders, mCLIP, AraCLIP, and Jina V4, each evaluated separately for label retrieval. A hybrid vocabulary is built from training captions and enriched with about 21K general domain labels translated from the Visual Genome dataset, covering objects, attributes, and scenes. The top-k retrieved labels are transformed into fluent Arabic prompts and passed along with the original image to vision-language models. In the second stage, we tested Qwen-VL and Gemini Pro Vision for caption generation, resulting in six encoder-decoder configurations. The results show that mCLIP + Gemini Pro Vision achieved the best BLEU-1 (5.34%) and cosine similarity (60.01%), while AraCLIP + Qwen-VL obtained the highest LLM-judge score (36.33%). This interpretable pipeline enables culturally coherent and contextually accurate Arabic captions.

Paperid: 6197, https://arxiv.org/pdf/2510.02913.pdf

Abstract:
Vision-language models like CLIP demonstrate impressive zero-shot generalization but remain highly vulnerable to adversarial attacks. In this work, we propose Confidence-Aware Weighting (CAW) to enhance zero-shot robustness in vision-language models. CAW consists of two components: (1) a Confidence-Aware loss that prioritizes uncertain adversarial examples by scaling the KL divergence between clean and adversarial predictions, and (2) a feature alignment regularization that preserves semantic consistency by minimizing the distance between frozen and fine-tuned image encoder features on adversarial inputs. These components work jointly to improve both clean and robust accuracy without sacrificing generalization. Extensive experiments on TinyImageNet and 14 additional datasets show that CAW outperforms recent methods such as PMG-AFT and TGA-ZSR under strong attacks like AutoAttack, while using less memory.

Paperid: 6198, https://arxiv.org/pdf/2510.01389.pdf

Abstract:
Recent Vision-Language-Action (VLA) models show strong generalization capabilities, yet they lack introspective mechanisms for anticipating failures and requesting help from a human supervisor. We present \textbf{INSIGHT}, a learning framework for leveraging token-level uncertainty signals to predict when a VLA should request help. Using $π_0$-FAST as the underlying model, we extract per-token \emph{entropy}, \emph{log-probability}, and Dirichlet-based estimates of \emph{aleatoric and epistemic uncertainty}, and train compact transformer classifiers to map these sequences to help triggers. We explore supervision regimes for strong or weak supervision, and extensively compare them across in-distribution and out-of-distribution tasks. Our results show a trade-off: strong labels enable models to capture fine-grained uncertainty dynamics for reliable help detection, while weak labels, though noisier, still support competitive introspection when training and evaluation are aligned, offering a scalable path when dense annotation is impractical. Crucially, we find that modeling the temporal evolution of token-level uncertainty signals with transformers provides far greater predictive power than static sequence-level scores. This study provides the first systematic evaluation of uncertainty-based introspection in VLAs, opening future avenues for active learning and for real-time error mitigation through selective human intervention.

Paperid: 6199, https://arxiv.org/pdf/2509.25896.pdf

Abstract:
As Vision-Language Models (VLMs) move into interactive, multi-turn use, new safety risks arise that single-turn or single-modality moderation misses. In Multimodal Multi-Turn (MMT) dialogues, malicious intent can be spread across turns and images, while context-sensitive replies may still advance harmful content. To address this challenge, we present the first systematic definition and study of MMT dialogue safety. Building on this formulation, we introduce the Multimodal Multi-turn Dialogue Safety (MMDS) dataset. We further develop an automated multimodal multi-turn red-teaming framework based on Monte Carlo Tree Search (MCTS) to generate unsafe multimodal multi-turn dialogues for MMDS. MMDS contains 4,484 annotated multimodal dialogue samples with fine-grained safety ratings, policy dimension labels, and evidence-based rationales for both users and assistants. Leveraging MMDS, we present LLaVAShield, a powerful tool that jointly detects and assesses risk in user inputs and assistant responses. Across comprehensive experiments, LLaVAShield consistently outperforms strong baselines on MMT content moderation tasks and under dynamic policy configurations, establishing new state-of-the-art results. We will publicly release the dataset and model to support future research.

Paperid: 6200, https://arxiv.org/pdf/2509.24559.pdf

Abstract:
Vision Language Action models (VLAs) trained with policy-based reinforcement learning (RL) encode complex behaviors without explicitly modeling environmental dynamics. However, it remains unclear whether VLAs implicitly learn world models, a hallmark of model-based RL. We propose an experimental methodology using embedding arithmetic on state representations to probe whether OpenVLA, the current state of the art in VLAs, contains latent knowledge of state transitions. Specifically, we measure the difference between embeddings of sequential environment states and test whether this transition vector is recoverable from intermediate model activations. Using linear and non linear probes trained on activations across layers, we find statistically significant predictive ability on state transitions exceeding baselines (embeddings), indicating that OpenVLA encodes an internal world model (as opposed to the probes learning the state transitions). We investigate the predictive ability of an earlier checkpoint of OpenVLA, and uncover hints that the world model emerges as training progresses. Finally, we outline a pipeline leveraging Sparse Autoencoders (SAEs) to analyze OpenVLA's world model.

Paperid: 6201, https://arxiv.org/pdf/2509.23465.pdf

Abstract:
Solving Traveling Salesman Problem (TSP) is NP-hard yet fundamental for wide real-world applications. Classical exact methods face challenges in scaling, and heuristic methods often require domain-specific parameter calibration. While learning-based approaches have shown promise, they suffer from poor generalization and limited scalability due to fixed training data. This work proposes ViTSP, a novel framework that leverages pre-trained vision language models (VLMs) to visually guide the solution process for large-scale TSPs. The VLMs function to identify promising small-scale subproblems from a visualized TSP instance, which are then efficiently optimized using an off-the-shelf solver to improve the global solution. ViTSP bypasses the dedicated model training at the user end while maintaining effectiveness across diverse instances. Experiments on real-world TSP instances ranging from 1k to 88k nodes demonstrate that ViTSP consistently achieves solutions with average optimality gaps below 0.2%, outperforming existing learning-based methods. Under the same runtime budget, it surpasses the best-performing heuristic solver, LKH-3, by reducing its gaps by 12% to 100%, particularly on very-large-scale instances with more than 10k nodes. Our framework offers a new perspective in hybridizing pre-trained generative models and operations research solvers in solving combinatorial optimization problems, with practical implications for integration into more complex logistics systems. The code is available at https://anonymous.4open.science/r/ViTSP_codes-6683.

Paperid: 6202, https://arxiv.org/pdf/2509.22447.pdf

Abstract:
Foundation models (FMs) have recently opened up new frontiers in the field of artificial life (ALife) by providing powerful tools to automate search through ALife simulations. Previous work aligns ALife simulations with natural language target prompts using vision-language models (VLMs). We build on Automated Search for Artificial Life (ASAL) by introducing ASAL++, a method for open-ended-like search guided by multimodal FMs. We use a second FM to propose new evolutionary targets based on a simulation's visual history. This induces an evolutionary trajectory with increasingly complex targets. We explore two strategies: (1) evolving a simulation to match a single new prompt at each iteration (Evolved Supervised Targets: EST) and (2) evolving a simulation to match the entire sequence of generated prompts (Evolved Temporal Targets: ETT). We test our method empirically in the Lenia substrate using Gemma-3 to propose evolutionary targets, and show that EST promotes greater visual novelty, while ETT fosters more coherent and interpretable evolutionary sequences. Our results suggest that ASAL++ points towards new directions for FM-driven ALife discovery with open-ended characteristics.

Paperid: 6203, https://arxiv.org/pdf/2509.22014.pdf

Abstract:
Healthcare robotics requires robust multimodal perception and reasoning to ensure safety in dynamic clinical environments. Current Vision-Language Models (VLMs) demonstrate strong general-purpose capabilities but remain limited in temporal reasoning, uncertainty estimation, and structured outputs needed for robotic planning. We present a lightweight agentic multimodal framework for video-based scene understanding. Combining the Qwen2.5-VL-3B-Instruct model with a SmolAgent-based orchestration layer, it supports chain-of-thought reasoning, speech-vision fusion, and dynamic tool invocation. The framework generates structured scene graphs and leverages a hybrid retrieval module for interpretable and adaptive reasoning. Evaluations on the Video-MME benchmark and a custom clinical dataset show competitive accuracy and improved robustness compared to state-of-the-art VLMs, demonstrating its potential for applications in robot-assisted surgery, patient monitoring, and decision support.

Paperid: 6204, https://arxiv.org/pdf/2509.21997.pdf

Abstract:
Multimodal large language models (MLLMs) have achieved remarkable success across diverse vision-language tasks, yet they remain highly susceptible to hallucinations, producing content that is fluent but inconsistent with visual evidence. Such hallucinations, spanning objects, attributes, and relations, persist even in larger models, while existing mitigation approaches often require additional finetuning, handcrafted priors, or trade-offs that compromise informativeness and scalability. To address this limitation, we propose a training-free, self-supervised method for hallucination mitigation. Our approach introduces a novel hallucination amplification mechanism: a caption is projected into the visual space via a text-to-image model to reveal implicit hallucination signals, serving as a negative anchor, while the original image provides a positive anchor. Leveraging these dual anchors, we edit decoder hidden states by pulling representations toward faithful semantics and pushing them away from hallucination directions. This correction requires no human priors or additional training costs, ensuring both effectiveness and efficiency. Extensive experiments across multiple benchmarks show that our method significantly reduces hallucinations at the object, attribute, and relation levels while largely preserving recall and caption richness, e.g., achieving a hallucination reduction by over 5% using LLaVA-v1.5-7B on CHAIR. Furthermore, results on diverse architectures, including LLaVA-NEXT-7B, Cambrian-8B, and InstructBLIP-7B, validate strong cross-architecture generalization. More importantly, when applied to hallucination-free captions, our method introduces almost no side effects, underscoring its robustness and practical plug-and-play applicability. The implementation will be publicly available.

Paperid: 6205, https://arxiv.org/pdf/2509.20769.pdf

Abstract:
In this work, we present a retrieval-augmented generation (RAG)-based system for provenance analysis of archaeological artifacts, designed to support expert reasoning by integrating multimodal retrieval and large vision-language models (VLMs). The system constructs a dual-modal knowledge base from reference texts and images, enabling raw visual, edge-enhanced, and semantic retrieval to identify stylistically similar objects. Retrieved candidates are synthesized by the VLM to generate structured inferences, including chronological, geographical, and cultural attributions, alongside interpretive justifications. We evaluate the system on a set of Eastern Eurasian Bronze Age artifacts from the British Museum. Expert evaluation demonstrates that the system produces meaningful and interpretable outputs, offering scholars concrete starting points for analysis and significantly alleviating the cognitive burden of navigating vast comparative corpora.

Paperid: 6206, https://arxiv.org/pdf/2509.20751.pdf

Abstract:
Recent studies show that deep vision-only and language-only models--trained on disjoint modalities--nonetheless project their inputs into a partially aligned representational space. Yet we still lack a clear picture of where in each network this convergence emerges, what visual or linguistic cues support it, whether it captures human preferences in many-to-many image-text scenarios, and how aggregating exemplars of the same concept affects alignment. Here, we systematically investigate these questions. We find that alignment peaks in mid-to-late layers of both model types, reflecting a shift from modality-specific to conceptually shared representations. This alignment is robust to appearance-only changes but collapses when semantics are altered (e.g., object removal or word-order scrambling), highlighting that the shared code is truly semantic. Moving beyond the one-to-one image-caption paradigm, a forced-choice "Pick-a-Pic" task shows that human preferences for image-caption matches are mirrored in the embedding spaces across all vision-language model pairs. This pattern holds bidirectionally when multiple captions correspond to a single image, demonstrating that models capture fine-grained semantic distinctions akin to human judgments. Surprisingly, averaging embeddings across exemplars amplifies alignment rather than blurring detail. Together, our results demonstrate that unimodal networks converge on a shared semantic code that aligns with human judgments and strengthens with exemplar aggregation.

Paperid: 6207, https://arxiv.org/pdf/2509.20119.pdf

Abstract:
Scientific visual question answering poses significant challenges for vision-language models due to the complexity of scientific figures and their multimodal context. Traditional approaches treat the figure and accompanying text (e.g., questions and answer options) as separate inputs. EXAMS-V introduced a new paradigm by embedding both visual and textual content into a single image. However, even state-of-the-art proprietary models perform poorly on this setup in zero-shot settings, underscoring the need for task-specific fine-tuning. To address the scarcity of training data in this "text-in-image" format, we synthesize a new dataset by converting existing separate image-text pairs into unified images. Fine-tuning a small multilingual multimodal model on a mix of our synthetic data and EXAMS-V yields notable gains across 13 languages, demonstrating strong average improvements and cross-lingual transfer.

Paperid: 6208, https://arxiv.org/pdf/2509.19524.pdf

Abstract:
Robot learning papers typically report a single binary success rate (SR), which obscures where a policy succeeds or fails along a multi-step manipulation task. We argue that subgoal-level reporting should become routine: for each trajectory, a vector of per-subgoal SRs that makes partial competence visible (e.g., grasp vs. pour). We propose a blueprint for StepEval, a cost-aware plug-in evaluation framework that utilizes vision-language models (VLMs) as automated judges of subgoal outcomes from recorded images or videos. Rather than proposing new benchmarks or APIs, our contribution is to outline design principles for a scalable, community-driven open-source project. In StepEval, the primary artifact for policy evaluation is the per-subgoal SR vector; however, other quantities (e.g., latency or cost estimates) are also considered for framework-optimization diagnostics to help the community tune evaluation efficiency and accuracy when ground-truth subgoal success labels are available. We discuss how such a framework can remain model-agnostic, support single- or multi-view inputs, and be lightweight enough to adopt across labs. The intended contribution is a shared direction: a minimal, extensible seed that invites open-source contributions, so that scoring the steps, not just the final goal, becomes a standard and reproducible practice.

Paperid: 6209, https://arxiv.org/pdf/2509.19070.pdf

Abstract:
This paper presents ColorBlindnessEval, a novel benchmark designed to evaluate the robustness of Vision-Language Models (VLMs) in visually adversarial scenarios inspired by the Ishihara color blindness test. Our dataset comprises 500 Ishihara-like images featuring numbers from 0 to 99 with varying color combinations, challenging VLMs to accurately recognize numerical information embedded in complex visual patterns. We assess 9 VLMs using Yes/No and open-ended prompts and compare their performance with human participants. Our experiments reveal limitations in the models' ability to interpret numbers in adversarial contexts, highlighting prevalent hallucination issues. These findings underscore the need to improve the robustness of VLMs in complex visual environments. ColorBlindnessEval serves as a valuable tool for benchmarking and improving the reliability of VLMs in real-world applications where accuracy is critical.

Paperid: 6210, https://arxiv.org/pdf/2509.18938.pdf

Abstract:
While deep learning, including Convolutional Neural Networks (CNNs) and Vision Transformers (ViTs), has significantly advanced classification performance, its typical reliance on extensive annotated datasets presents a major obstacle in many practical scenarios where such data is scarce. Vision-language models (VLMs) and transfer learning with pre-trained visual models appear as promising techniques to deal with this problem. This paper proposes a novel zero-shot image classification framework that combines a VLM and a pre-trained visual model within a self-learning cycle. Requiring only the set of class names and no labeled training data, our method utilizes a confidence-based pseudo-labeling strategy to train a lightweight classifier directly on the test data, enabling dynamic adaptation. The VLM identifies high-confidence samples, and the pre-trained visual model enhances their visual representations. These enhanced features then iteratively train the classifier, allowing the system to capture complementary semantic and visual cues without supervision. Notably, our approach avoids VLM fine-tuning and the use of large language models, relying on the visual-only model to reduce the dependence on semantic representation. Experimental evaluations on ten diverse datasets demonstrate that our approach outperforms the baseline zero-shot method.

Paperid: 6211, https://arxiv.org/pdf/2509.18405.pdf

Abstract:
Checks remain a foundational instrument in the financial ecosystem, facilitating substantial transaction volumes across institutions. However, their continued use also renders them a persistent target for fraud, underscoring the importance of robust check fraud detection mechanisms. At the core of such systems lies the accurate identification and localization of critical fields, such as the signature, magnetic ink character recognition (MICR) line, courtesy amount, legal amount, payee, and payer, which are essential for subsequent verification against reference checks belonging to the same customer. This field-level detection is traditionally dependent on object detection models trained on large, diverse, and meticulously labeled datasets, a resource that is scarce due to proprietary and privacy concerns. In this paper, we introduce a novel, training-free framework for automated check field detection, leveraging the power of a vision language model (VLM) in conjunction with a multimodal large language model (MLLM). Our approach enables zero-shot detection of check components, significantly lowering the barrier to deployment in real-world financial settings. Quantitative evaluation of our model on a hand-curated dataset of 110 checks spanning multiple formats and layouts demonstrates strong performance and generalization capability. Furthermore, this framework can serve as a bootstrap mechanism for generating high-quality labeled datasets, enabling the development of specialized real-time object detection models tailored to institutional needs.

Paperid: 6212, https://arxiv.org/pdf/2509.16421.pdf

Abstract:
Real-time understanding of continuous video streams is essential for intelligent agents operating in high-stakes environments, including autonomous vehicles, surveillance drones, and disaster response robots. Yet, most existing video understanding and highlight detection methods assume access to the entire video during inference, making them unsuitable for online or streaming scenarios. In particular, current models optimize for offline summarization, failing to support step-by-step reasoning needed for real-time decision-making. We introduce Aha, an autoregressive highlight detection framework that predicts the relevance of each video frame against a task described in natural language. Without accessing future video frames, Aha utilizes a multimodal vision-language model and lightweight, decoupled heads trained on a large, curated dataset of human-centric video labels. To enable scalability, we introduce the Dynamic SinkCache mechanism that achieves constant memory usage across infinite-length streams without degrading performance on standard benchmarks. This encourages the hidden representation to capture high-level task objectives, enabling effective frame-level rankings for informativeness, relevance, and uncertainty with respect to the natural language task. Aha achieves state-of-the-art (SOTA) performance on highlight detection benchmarks, surpassing even prior offline, full-context approaches and video-language models by +5.9% on TVSum and +8.3% on Mr. Hisum in mAP (mean Average Precision). We explore Aha's potential for real-world robotics applications given a task-oriented natural language input and a continuous, robot-centric video. Both experiments demonstrate Aha's potential effectiveness as a real-time reasoning module for downstream planning and long-horizon understanding.

Paperid: 6213, https://arxiv.org/pdf/2509.15241.pdf

Abstract:
Ensuring that multi-modal content adheres to brand, legal, or platform-specific compliance standards is an increasingly complex challenge across domains. Traditional compliance frameworks typically rely on disjointed, multi-stage pipelines that integrate separate modules for image classification, text extraction, audio transcription, hand-crafted checks, and rule-based merges. This architectural fragmentation increases operational overhead, hampers scalability, and hinders the ability to adapt to dynamic guidelines efficiently. With the emergence of Multimodal Large Language Models (MLLMs), there is growing potential to unify these workflows under a single, general-purpose framework capable of jointly processing visual and textual content. In light of this, we propose Multimodal Parameter Agnostic Compliance Engine (M-PACE), a framework designed for assessing attributes across vision-language inputs in a single pass. As a representative use case, we apply M-PACE to advertisement compliance, demonstrating its ability to evaluate over 15 compliance-related attributes. To support structured evaluation, we introduce a human-annotated benchmark enriched with augmented samples that simulate challenging real-world conditions, including visual obstructions and profanity injection. M-PACE employs a mother-child MLLM setup, demonstrating that a stronger parent MLLM evaluating the outputs of smaller child models can significantly reduce dependence on human reviewers, thereby automating quality control. Our analysis reveals that inference costs reduce by over 31 times, with the most efficient models (Gemini 2.0 Flash as child MLLM selected by mother MLLM) operating at 0.0005 per image, compared to 0.0159 for Gemini 2.5 Pro with comparable accuracy, highlighting the trade-off between cost and output quality achieved in real time by M-PACE in real life deployment over advertising data.

Paperid: 6214, https://arxiv.org/pdf/2509.15234.pdf

Abstract:
Vision-language pretraining has advanced image-text alignment, yet progress in radiology remains constrained by the heterogeneity of clinical reports, including abbreviations, impression-only notes, and stylistic variability. Unlike general-domain settings where more data often leads to better performance, naively scaling to large collections of noisy reports can plateau or even degrade model learning. We ask whether large language model (LLM) encoders can provide robust clinical representations that transfer across diverse styles and better guide image-text alignment. We introduce LLM2VEC4CXR, a domain-adapted LLM encoder for chest X-ray reports, and LLM2CLIP4CXR, a dual-tower framework that couples this encoder with a vision backbone. LLM2VEC4CXR improves clinical text understanding over BERT-based baselines, handles abbreviations and style variation, and achieves strong clinical alignment on report-level metrics. LLM2CLIP4CXR leverages these embeddings to boost retrieval accuracy and clinically oriented scores, with stronger cross-dataset generalization than prior medical CLIP variants. Trained on 1.6M CXR studies from public and private sources with heterogeneous and noisy reports, our models demonstrate that robustness -- not scale alone -- is the key to effective multimodal learning. We release models to support further research in medical image-text representation learning.

Paperid: 6215, https://arxiv.org/pdf/2509.13289.pdf

Abstract:
A reliable method of quantifying the perceptual realness of AI-generated images and identifying visually inconsistent regions is crucial for practical use of AI-generated images and for improving photorealism of generative AI via realness feedback during training. This paper introduces a framework that accomplishes both overall objective realness assessment and local inconsistency identification of AI-generated images using textual descriptions of visual inconsistencies generated by vision-language models trained on large datasets that serve as reliable substitutes for human annotations. Our results demonstrate that the proposed multimodal approach improves objective realness prediction performance and produces dense realness maps that effectively distinguish between realistic and unrealistic spatial regions.

Paperid: 6216, https://arxiv.org/pdf/2509.12492.pdf

Abstract:
Vision-Language Models (VLMs) have attained exceptional success across multimodal tasks such as image captioning and visual question answering. However, their robustness under noisy conditions remains unfamiliar. In this study, we present a comprehensive evaluation framework to evaluate the performance of several state-of-the-art VLMs under controlled perturbations, including lighting variation, motion blur, and compression artifacts. We used both lexical-based metrics (BLEU, METEOR, ROUGE, CIDEr) and neural-based similarity measures using sentence embeddings to quantify semantic alignment. Our experiments span diverse datasets, revealing key insights: (1) descriptiveness of ground-truth captions significantly influences model performance; (2) larger models like LLaVA excel in semantic understanding but do not universally outperform smaller models; and (3) certain noise types, such as JPEG compression and motion blur, dramatically degrade performance across models. Our findings highlight the nuanced trade-offs between model size, dataset characteristics, and noise resilience, offering a standardized benchmark for future robust multimodal learning.

Paperid: 6217, https://arxiv.org/pdf/2509.11766.pdf

Abstract:
While foundation models show remarkable progress in language and vision, existing vision-language models (VLMs) still have limited spatial and embodiment understanding. Transferring VLMs to embodied domains reveals fundamental mismatches between modalities, pretraining distributions, and training objectives, leaving action comprehension and generation as a central bottleneck on the path to AGI. We introduce WALL-OSS, an end-to-end embodied foundation model that leverages large-scale multimodal pretraining to achieve (1) embodiment-aware vision-language understanding, (2) strong language-action association, and (3) robust manipulation capability. Our approach employs a tightly coupled architecture and multi-strategies training curriculum that enables Unified Cross-Level CoT-seamlessly unifying instruction reasoning, subgoal decomposition, and fine-grained action synthesis within a single differentiable framework. Our results show that WALL-OSS attains high success on complex long-horizon manipulations, demonstrates strong instruction-following capabilities, complex understanding and reasoning, and outperforms strong baselines, thereby providing a reliable and scalable path from VLMs to embodied foundation models.

Paperid: 6218, https://arxiv.org/pdf/2509.11667.pdf

Abstract:
Telecom domain 3GPP documents are replete with images containing sequence diagrams. Advances in Vision-Language Large Models (VLMs) have eased conversion of such images to machine-readable PlantUML (puml) formats. However, there is a gap in evaluation of such conversions - existing works do not compare puml scripts for various components. In this work, we propose performance metrics to measure the effectiveness of such conversions. A dataset of sequence diagrams from 3GPP documents is chosen to be representative of domain-specific actual scenarios. We compare puml outputs from two VLMs - Claude Sonnet and GPT-4V - against manually created ground truth representations. We use version control tools to capture differences and introduce standard performance metrics to measure accuracies along various components: participant identification, message flow accuracy, sequence ordering, and grouping construct preservation. We demonstrate effectiveness of proposed metrics in quantifying conversion errors across various components of puml scripts. The results show that nodes, edges and messages are accurately captured. However, we observe that VLMs do not necessarily perform well on complex structures such as notes, box, groups. Our experiments and performance metrics indicates a need for better representation of these components in training data for fine-tuned VLMs.

Paperid: 6219, https://arxiv.org/pdf/2509.10707.pdf

Abstract:
As AI systems increasingly evaluate other AI outputs, understanding their assessment behavior becomes crucial for preventing cascading biases. This study analyzes vision-language descriptions generated by NVIDIA's Describe Anything Model and evaluated by three GPT variants (GPT-4o, GPT-4o-mini, GPT-5) to uncover distinct "evaluation personalities" the underlying assessment strategies and biases each model demonstrates. GPT-4o-mini exhibits systematic consistency with minimal variance, GPT-4o excels at error detection, while GPT-5 shows extreme conservatism with high variability. Controlled experiments using Gemini 2.5 Pro as an independent question generator validate that these personalities are inherent model properties rather than artifacts. Cross-family analysis through semantic similarity of generated questions reveals significant divergence: GPT models cluster together with high similarity while Gemini exhibits markedly different evaluation strategies. All GPT models demonstrate a consistent 2:1 bias favoring negative assessment over positive confirmation, though this pattern appears family-specific rather than universal across AI architectures. These findings suggest that evaluation competence does not scale with general capability and that robust AI assessment requires diverse architectural perspectives.

Paperid: 6220, https://arxiv.org/pdf/2509.10345.pdf

Abstract:
Visual grounding refers to the ability of a model to identify a region within some visual input that matches a textual description. Consequently, a model equipped with visual grounding capabilities can target a wide range of applications in various domains, including referring expression comprehension, answering questions pertinent to fine-grained details in images or videos, caption visual context by explicitly referring to entities, as well as low and high-level control in simulated and real environments. In this survey paper, we review representative works across the key areas of research on modern general-purpose vision language models (VLMs). We first outline the importance of grounding in VLMs, then delineate the core components of the contemporary paradigm for developing grounded models, and examine their practical applications, including benchmarks and evaluation metrics for grounded multimodal generation. We also discuss the multifaceted interrelations among visual grounding, multimodal chain-of-thought, and reasoning in VLMs. Finally, we analyse the challenges inherent to visual grounding and suggest promising directions for future research.

Paperid: 6221, https://arxiv.org/pdf/2509.09263.pdf

Abstract:
Long video understanding remains a fundamental challenge for multimodal large language models (MLLMs), particularly in tasks requiring precise temporal reasoning and event localization. Existing approaches typically adopt uniform frame sampling and rely on implicit position encodings to model temporal order. However, these methods struggle with long-range dependencies, leading to critical information loss and degraded temporal comprehension. In this paper, we propose Dynamic Absolute Time Enhancement (DATE) that enhances temporal awareness in MLLMs through the Timestamp Injection Mechanism (TIM) and a semantically guided Temporal-Aware Similarity Sampling (TASS) strategy. Specifically, we interleave video frame embeddings with textual timestamp tokens to construct a continuous temporal reference system. We further reformulate the video sampling problem as a vision-language retrieval task and introduce a two-stage algorithm to ensure both semantic relevance and temporal coverage: enriching each query into a descriptive caption to better align with the vision feature, and sampling key event with a similarity-driven temporally regularized greedy strategy. Our method achieves remarkable improvements w.r.t. absolute time understanding and key event localization, resulting in state-of-the-art performance among 7B and 72B models on hour-long video benchmarks. Particularly, our 7B model even exceeds many 72B models on some benchmarks.

Paperid: 6222, https://arxiv.org/pdf/2509.08338.pdf

Abstract:
Accurate and early diagnosis of malignant melanoma is critical for improving patient outcomes. While convolutional neural networks (CNNs) have shown promise in dermoscopic image analysis, they often neglect clinical metadata and require extensive preprocessing. Vision-language models (VLMs) offer a multimodal alternative but struggle to capture clinical specificity when trained on general-domain data. To address this, we propose a retrieval-augmented VLM framework that incorporates semantically similar patient cases into the diagnostic prompt. Our method enables informed predictions without fine-tuning and significantly improves classification accuracy and error correction over conventional baselines. These results demonstrate that retrieval-augmented prompting provides a robust strategy for clinical decision support.

Paperid: 6223, https://arxiv.org/pdf/2509.08270.pdf

Abstract:
As Vision-Language Models (VLMs) grow in sophistication, their ability to perform reasoning is coming under increasing supervision. While they excel at many tasks, their grasp of fundamental scientific principles, such as physics, remains an underexplored frontier. To reflect the advancements in these capabilities, we introduce a novel and accessible framework designed to rigorously evaluate VLMs on their understanding of 2D physics. Our framework features a pragmatic scenario generator that creates a diverse testbed of over 400 problems across four core domains: Projectile Motion, Collision Dynamics, Mechanics, and Fluid Dynamics. Through comprehensive evaluation of four state-of-the-art VLMs, we demonstrate a strong correlation between model scale and reasoning ability, with our top-performing model, Qwen2.5-VL-7B, achieving an overall score of 0.815. We find that while models excel at formulaic problems, they struggle significantly with domains requiring abstract spatial reasoning. By designing this framework, we aim to democratize the study of scientific reasoning in VLMs and foster deeper insights into their capabilities and limitations.

Paperid: 6224, https://arxiv.org/pdf/2509.06994.pdf

Abstract:
Open-source Vision-Language Models show immense promise for enterprise applications, yet a critical disconnect exists between academic evaluation and enterprise deployment requirements. Current benchmarks rely heavily on multiple-choice questions and synthetic data, failing to capture the complexity of real-world business applications like social media content analysis. This paper introduces VLM-in-the-Wild (ViLD), a comprehensive framework to bridge this gap by evaluating VLMs on operational enterprise requirements. We define ten business-critical tasks: logo detection, OCR, object detection, human presence and demographic analysis, human activity and appearance analysis, scene detection, camera perspective and media quality assessment, dominant colors, comprehensive description, and NSFW detection. To this framework, we bring an innovative BlockWeaver Algorithm that solves the challenging problem of comparing unordered, variably-grouped OCR outputs from VLMs without relying on embeddings or LLMs, achieving remarkable speed and reliability. To demonstrate efficacy of ViLD, we constructed a new benchmark dataset of 7,500 diverse samples, carefully stratified from a corpus of one million real-world images and videos. ViLD provides actionable insights by combining semantic matching (both embedding-based and LLM-as-a-judge approaches), traditional metrics, and novel methods to measure the completeness and faithfulness of descriptive outputs. By benchmarking leading open-source VLMs (Qwen, MIMO, and InternVL) against a powerful proprietary baseline as per ViLD framework, we provide one of the first industry-grounded, task-driven assessment of VLMs capabilities, offering actionable insights for their deployment in enterprise environments.

Paperid: 6225, https://arxiv.org/pdf/2509.06415.pdf

Abstract:
Recent progress in vision-language models (VLMs) has led to impressive results in document understanding tasks, but their high computational demands remain a challenge. To mitigate the compute burdens, we propose a lightweight token pruning framework that filters out non-informative background regions from document images prior to VLM processing. A binary patch-level classifier removes non-text areas, and a max-pooling refinement step recovers fragmented text regions to enhance spatial coherence. Experiments on real-world document datasets demonstrate that our approach substantially lowers computational costs, while maintaining comparable accuracy.

Paperid: 6226, https://arxiv.org/pdf/2509.06033.pdf

Abstract:
The reliable analysis of blood reports is important for health knowledge, but individuals often struggle with interpretation, leading to anxiety and overlooked issues. We explore the potential of general-purpose Vision-Language Models (VLMs) to address this challenge by automatically analyzing blood report images. We conduct a comparative evaluation of three VLMs: Qwen-VL-Max, Gemini 2.5 Pro, and Llama 4 Maverick, determining their performance on a dataset of 100 diverse blood report images. Each model was prompted with clinically relevant questions adapted to each blood report. The answers were then processed using Sentence-BERT to compare and evaluate how closely the models responded. The findings suggest that general-purpose VLMs are a practical and promising technology for developing patient-facing tools for preliminary blood report analysis. Their ability to provide clear interpretations directly from images can improve health literacy and reduce the limitations to understanding complex medical information. This work establishes a foundation for the future development of reliable and accessible AI-assisted healthcare applications. While results are encouraging, they should be interpreted cautiously given the limited dataset size.

Paperid: 6227, https://arxiv.org/pdf/2509.05695.pdf

Abstract:
Human action recognition often struggles with deep semantic understanding, complex contextual information, and fine-grained distinction, limitations that traditional methods frequently encounter when dealing with diverse video data. Inspired by the remarkable capabilities of large language models, this paper introduces LVLM-VAR, a novel framework that pioneers the application of pre-trained Vision-Language Large Models (LVLMs) to video action recognition, emphasizing enhanced accuracy and interpretability. Our method features a Video-to-Semantic-Tokens (VST) Module, which innovatively transforms raw video sequences into discrete, semantically and temporally consistent "semantic action tokens," effectively crafting an "action narrative" that is comprehensible to an LVLM. These tokens, combined with natural language instructions, are then processed by a LoRA-fine-tuned LVLM (e.g., LLaVA-13B) for robust action classification and semantic reasoning. LVLM-VAR not only achieves state-of-the-art or highly competitive performance on challenging benchmarks such as NTU RGB+D and NTU RGB+D 120, demonstrating significant improvements (e.g., 94.1% on NTU RGB+D X-Sub and 90.0% on NTU RGB+D 120 X-Set), but also substantially boosts model interpretability by generating natural language explanations for its predictions.

Paperid: 6228, https://arxiv.org/pdf/2509.05669.pdf

Abstract:
Current Large Language Models (LLMs) and Vision-Language Large Models (LVLMs) excel in single-turn tasks but face significant challenges in multi-turn interactions requiring deep contextual understanding and complex visual reasoning, often leading to fragmented reasoning, context loss, and hallucinations. To address these limitations, we propose Context-Aware Multi-Turn Visual Reasoning (CAMVR), a novel framework designed to empower LVLMs with robust and coherent multi-turn visual-textual inference capabilities. CAMVR introduces two key innovations: a Visual-Textual Context Memory Unit (VCMU), a dynamic read-write memory network that stores and manages critical visual features, textual semantic representations, and their cross-modal correspondences from each interaction turn; and an Adaptive Visual Focus Guidance (AVFG) mechanism, which leverages the VCMU's context to dynamically adjust the visual encoder's attention to contextually relevant image regions. Our multi-level reasoning integration strategy ensures that response generation is deeply coherent with both current inputs and accumulated historical context. Extensive experiments on challenging datasets, including VisDial, an adapted A-OKVQA, and our novel Multi-Turn Instruction Following (MTIF) dataset, demonstrate that CAMVR consistently achieves state-of-the-art performance.

Paperid: 6229, https://arxiv.org/pdf/2509.05513.pdf

Abstract:
Egocentric human videos provide scalable demonstrations for imitation learning, but existing corpora often lack either fine-grained, temporally localized action descriptions or dexterous hand annotations. We introduce OpenEgo, a multimodal egocentric manipulation dataset with standardized hand-pose annotations and intention-aligned action primitives. OpenEgo totals 1107 hours across six public datasets, covering 290 manipulation tasks in 600+ environments. We unify hand-pose layouts and provide descriptive, timestamped action primitives. To validate its utility, we train language-conditioned imitation-learning policies to predict dexterous hand trajectories. OpenEgo is designed to lower the barrier to learning dexterous manipulation from egocentric video and to support reproducible research in vision-language-action learning. All resources and instructions will be released at www.openegocentric.com.

Paperid: 6230, https://arxiv.org/pdf/2509.04214.pdf

Abstract:
Machine learning (ML) models have the potential to transform military battlefields, presenting a large external pressure to rapidly incorporate them into operational settings. However, it is well-established that these ML models are vulnerable to a number of adversarial attacks throughout the model deployment pipeline that threaten to negate battlefield advantage. One broad category is privacy attacks (such as model inversion) where an adversary can reverse engineer information from the model, such as the sensitive data used in its training. The ability to quantify the risk of model inversion attacks (MIAs) is not well studied, and there is a lack of automated developmental test and evaluation (DT&E) tools and metrics to quantify the effectiveness of privacy loss of the MIA. The current DT&E process is difficult because ML model inversions can be hard for a human to interpret, subjective when they are interpretable, and difficult to quantify in terms of inversion quality. Additionally, scaling the DT&E process is challenging due to many ML model architectures and data modalities that need to be assessed. In this work, we present a novel DT&E tool that quantifies the risk of data privacy loss from MIAs and introduces four adversarial risk dimensions to quantify privacy loss. Our DT&E pipeline combines inversion with vision language models (VLMs) to improve effectiveness while enabling scalable analysis. We demonstrate effectiveness using multiple MIA techniques and VLMs configured for zero-shot classification and image captioning. We benchmark the pipeline using several state-of-the-art MIAs in the computer vision domain with an image classification task that is typical in military applications. In general, our innovative pipeline extends the current model inversion DT&E capabilities by improving the effectiveness and scalability of the privacy loss analysis in an automated fashion.

Paperid: 6231, https://arxiv.org/pdf/2509.03615.pdf

Abstract:
Optical Character Recognition (OCR) in multilingual, noisy, and diverse real-world images remains a significant challenge for optical character recognition systems. With the rise of Large Vision-Language Models (LVLMs), there is growing interest in their ability to generalize and reason beyond fixed OCR pipelines. In this work, we introduce Sprinklr-Edge-OCR, a novel OCR system built specifically optimized for edge deployment in resource-constrained environments. We present a large-scale comparative evaluation of five state-of-the-art LVLMs (InternVL, Qwen, GOT OCR, LLaMA, MiniCPM) and two traditional OCR systems (Sprinklr-Edge-OCR, SuryaOCR) on a proprietary, doubly hand annotated dataset of multilingual (54 languages) images. Our benchmark covers a broad range of metrics including accuracy, semantic consistency, language coverage, computational efficiency (latency, memory, GPU usage), and deployment cost. To better reflect real-world applicability, we also conducted edge case deployment analysis, evaluating model performance on CPU only environments. Among the results, Qwen achieved the highest precision (0.54), while Sprinklr-Edge-OCR delivered the best overall F1 score (0.46) and outperformed others in efficiency, processing images 35 faster (0.17 seconds per image on average) and at less than 0.01 of the cost (0.006 USD per 1,000 images) compared to LVLM. Our findings demonstrate that the most optimal OCR systems for edge deployment are the traditional ones even in the era of LLMs due to their low compute requirements, low latency, and very high affordability.

Paperid: 6232, https://arxiv.org/pdf/2509.02659.pdf

Abstract:
End-to-end autonomous driving has drawn tremendous attention recently. Many works focus on using modular deep neural networks to construct the end-to-end archi-tecture. However, whether using powerful large language models (LLM), especially multi-modality Vision Language Models (VLM) could benefit the end-to-end driving tasks remain a question. In our work, we demonstrate that combining end-to-end architectural design and knowledgeable VLMs yield impressive performance on the driving tasks. It is worth noting that our method only uses a single camera and is the best camera-only solution across the leaderboard, demonstrating the effectiveness of vision-based driving approach and the potential for end-to-end driving tasks.

Paperid: 6233, https://arxiv.org/pdf/2509.00484.pdf

Abstract:
Multimodal reward models (MRMs) play a crucial role in the training, inference, and evaluation of Large Vision Language Models (LVLMs) by assessing response quality. However, existing benchmarks for evaluating MRMs in the video domain suffer from a limited number and diversity of questions, a lack of comprehensive evaluation dimensions, and inadequate evaluation of diverse types of MRMs. To address these gaps, we introduce VideoRewardBench, the first comprehensive benchmark covering four core aspects of video understanding: perception, knowledge, reasoning, and safety. Through our AI-assisted data pipeline, we curate a high-quality preference dataset of 1,563 annotated samples, including 1,482 unique videos and 1,559 distinct questions--15 times the number found in the most question-rich prior benchmark. Each sample is a triplet consisting of a video-text prompt, a chosen response, and a rejected response. We also conduct a comprehensive evaluation across 28 multimodal reward models spanning three categories: generative, discriminative, and semi-scalar. Results show that even the top-performing model GPT-4o achieves only 57.0% overall accuracy, and the state-of-the-art open-source model Qwen2.5-VL-72B reaches merely 53.3%. Our analysis further reveals three key insights: (i) MRMs trained with reinforcement learning (RL) do not necessarily exhibit stronger cross-modal generalization than those trained without RL; (ii) except for discriminative MRMs, other types of MRMs across varying model capacities can benefit from inference-time scaling; and (iii) variations in input video frame count have different effects on different types of MRMs. We believe VideoRewardBench offers a challenging and valuable benchmark for advancing the evaluation and development of MRMs in the video domain.

Paperid: 6234, https://arxiv.org/pdf/2509.00328.pdf

Abstract:
Vision-Language-Action (VLA) models are a promising path to realizing generalist embodied agents that can quickly adapt to new tasks, modalities, and environments. However, methods for interpreting and steering VLAs fall far short of classical robotics pipelines, which are grounded in explicit models of kinematics, dynamics, and control. This lack of mechanistic insight is a central challenge for deploying learned policies in real-world robotics, where robustness and explainability are critical. Motivated by advances in mechanistic interpretability for large language models, we introduce the first framework for interpreting and steering VLAs via their internal representations, enabling direct intervention in model behavior at inference time. We project feedforward activations within transformer layers onto the token embedding basis, identifying sparse semantic directions - such as speed and direction - that are causally linked to action selection. Leveraging these findings, we introduce a general-purpose activation steering method that modulates behavior in real time, without fine-tuning, reward signals, or environment interaction. We evaluate this method on two recent open-source VLAs, Pi0 and OpenVLA, and demonstrate zero-shot behavioral control in simulation (LIBERO) and on a physical robot (UR5). This work demonstrates that interpretable components of embodied VLAs can be systematically harnessed for control - establishing a new paradigm for transparent and steerable foundation models in robotics.

Paperid: 6235, https://arxiv.org/pdf/2509.00284.pdf

Abstract:
Industrial computer vision systems often struggle with noise, material variability, and uncontrolled imaging conditions, limiting the effectiveness of classical edge detectors and handcrafted pipelines. In this work, we present a language-guided generative vision system for remnant contour detection in manufacturing, designed to achieve CAD-level precision. The system is organized into three stages: data acquisition and preprocessing, contour generation using a conditional GAN, and multimodal contour refinement through vision-language modeling, where standardized prompts are crafted in a human-in-the-loop process and applied through image-text guided synthesis. On proprietary FabTrack datasets, the proposed system improved contour fidelity, enhancing edge continuity and geometric alignment while reducing manual tracing. For the refinement stage, we benchmarked several vision-language models, including Google's Gemini 2.0 Flash, OpenAI's GPT-image-1 integrated within a VLM-guided workflow, and open-source baselines. Under standardized conditions, GPT-image-1 consistently outperformed Gemini 2.0 Flash in both structural accuracy and perceptual quality. These findings demonstrate the promise of VLM-guided generative workflows for advancing industrial computer vision beyond the limitations of classical pipelines.

Paperid: 6236, https://arxiv.org/pdf/2508.17638.pdf

Abstract:
Large Vision-Language Models (LVLMs) commonly follow a paradigm that projects visual features and then concatenates them with text tokens to form a unified sequence input for Large Language Models (LLMs). However, this paradigm leads to a significant increase in the length of the input sequence, resulting in substantial computational overhead. Existing methods attempt to fuse visual information into the intermediate layers of LLMs, which alleviate the sequence length issue but often neglect the hierarchical semantic representations within the model and the fine-grained visual information available in the shallower visual encoding layers. To address this limitation, we propose DEHVF, an efficient vision-language fine-tuning method based on dynamic embedding and fusion of hierarchical visual features. Its core lies in leveraging the inherent hierarchical representation characteristics of visual encoders and language models. Through a lightweight hierarchical visual fuser, it dynamically selects and fuses hierarchical features corresponding to semantic granularity based on the internal representations of each layer in LLMs. The fused layer-related visual features are then projected and aligned before being directly embedded into the Feed-Forward Network (FFN) of the corresponding layer in LLMs. This approach not only avoids sequence expansion but also dynamically fuses multi-layer visual information. By fine-tuning only a small number of parameters, DEHVF achieves precise alignment and complementarity of cross-modal information at the same semantic granularity. We conducted experiments across various VL benchmarks, including visual question answering on ScienceQA and image captioning on COCO Captions. The results demonstrate that DEHVF achieves higher accuracy than existing parameter-efficient fine-tuning (PEFT) baselines while maintaining efficient training and inference.

Paperid: 6237, https://arxiv.org/pdf/2508.17524.pdf

Abstract:
Magnetic Resonance Imaging (MRI) is indispensable in clinical practice but remains constrained by fragmented, multi-stage workflows encompassing acquisition, reconstruction, segmentation, detection, diagnosis, and reporting. While deep learning has achieved progress in individual tasks, existing approaches are often anatomy- or application-specific and lack generalizability across diverse clinical settings. Moreover, current pipelines rarely integrate imaging data with complementary language information that radiologists rely on in routine practice. Here, we introduce OmniMRI, a unified vision-language foundation model designed to generalize across the entire MRI workflow. OmniMRI is trained on a large-scale, heterogeneous corpus curated from 60 public datasets, over 220,000 MRI volumes and 19 million MRI slices, incorporating image-only data, paired vision-text data, and instruction-response data. Its multi-stage training paradigm, comprising self-supervised vision pretraining, vision-language alignment, multimodal pretraining, and multi-task instruction tuning, progressively equips the model with transferable visual representations, cross-modal reasoning, and robust instruction-following capabilities. Qualitative results demonstrate OmniMRI's ability to perform diverse tasks within a single architecture, including MRI reconstruction, anatomical and pathological segmentation, abnormality detection, diagnostic suggestion, and radiology report generation. These findings highlight OmniMRI's potential to consolidate fragmented pipelines into a scalable, generalist framework, paving the way toward foundation models that unify imaging and clinical language for comprehensive, end-to-end MRI interpretation.

Paperid: 6238, https://arxiv.org/pdf/2508.17205.pdf

Abstract:
This paper introduces a multi-agent framework for comprehensive highway scene understanding, designed around a mixture-of-experts strategy. In this framework, a large generic vision-language model (VLM), such as GPT-4o, is contextualized with domain knowledge to generates task-specific chain-of-thought (CoT) prompts. These fine-grained prompts are then used to guide a smaller, efficient VLM (e.g., Qwen2.5-VL-7B) in reasoning over short videos, along with complementary modalities as applicable. The framework simultaneously addresses multiple critical perception tasks, including weather classification, pavement wetness assessment, and traffic congestion detection, achieving robust multi-task reasoning while balancing accuracy and computational efficiency. To support empirical validation, we curated three specialized datasets aligned with these tasks. Notably, the pavement wetness dataset is multimodal, combining video streams with road weather sensor data, highlighting the benefits of multimodal reasoning. Experimental results demonstrate consistently strong performance across diverse traffic and environmental conditions. From a deployment perspective, the framework can be readily integrated with existing traffic camera systems and strategically applied to high-risk rural locations, such as sharp curves, flood-prone lowlands, or icy bridges. By continuously monitoring the targeted sites, the system enhances situational awareness and delivers timely alerts, even in resource-constrained environments.

Paperid: 6239, https://arxiv.org/pdf/2508.15168.pdf

Abstract:
Diabetic Retinopathy (DR) is a major cause of global blindness, necessitating early and accurate diagnosis. While deep learning models have shown promise in DR detection, their black-box nature often hinders clinical adoption due to a lack of transparency and interpretability. To address this, we propose XDR-LVLM (eXplainable Diabetic Retinopathy Diagnosis with LVLM), a novel framework that leverages Vision-Language Large Models (LVLMs) for high-precision DR diagnosis coupled with natural language-based explanations. XDR-LVLM integrates a specialized Medical Vision Encoder, an LVLM Core, and employs Multi-task Prompt Engineering and Multi-stage Fine-tuning to deeply understand pathological features within fundus images and generate comprehensive diagnostic reports. These reports explicitly include DR severity grading, identification of key pathological concepts (e.g., hemorrhages, exudates, microaneurysms), and detailed explanations linking observed features to the diagnosis. Extensive experiments on the Diabetic Retinopathy (DDR) dataset demonstrate that XDR-LVLM achieves state-of-the-art performance, with a Balanced Accuracy of 84.55% and an F1 Score of 79.92% for disease diagnosis, and superior results for concept detection (77.95% BACC, 66.88% F1). Furthermore, human evaluations confirm the high fluency, accuracy, and clinical utility of the generated explanations, showcasing XDR-LVLM's ability to bridge the gap between automated diagnosis and clinical needs by providing robust and interpretable insights.

Paperid: 6240, https://arxiv.org/pdf/2508.14045.pdf

Abstract:
Visual Storytelling is a challenging multimodal task between Vision & Language, where the purpose is to generate a story for a stream of images. Its difficulty lies on the fact that the story should be both grounded to the image sequence but also narrative and coherent. The aim of this work is to balance between these aspects, by treating Visual Storytelling as a superset of Image Captioning, an approach quite different compared to most of prior relevant studies. This means that we firstly employ a vision-to-language model for obtaining captions of the input images, and then, these captions are transformed into coherent narratives using language-to-language methods. Our multifarious evaluation shows that integrating captioning and storytelling under a unified framework, has a positive impact on the quality of the produced stories. In addition, compared to numerous previous studies, this approach accelerates training time and makes our framework readily reusable and reproducible by anyone interested. Lastly, we propose a new metric/tool, named ideality, that can be used to simulate how far some results are from an oracle model, and we apply it to emulate human-likeness in visual storytelling.

Paperid: 6241, https://arxiv.org/pdf/2508.13470.pdf

Abstract:
Vision-language models (VLMs) have emerged as powerful tools for enabling automated traffic analysis; however, current approaches often demand substantial computational resources and struggle with fine-grained spatio-temporal understanding. This paper introduces STER-VLM, a computationally efficient framework that enhances VLM performance through (1) caption decomposition to tackle spatial and temporal information separately, (2) temporal frame selection with best-view filtering for sufficient temporal information, and (3) reference-driven understanding for capturing fine-grained motion and dynamic context and (4) curated visual/textual prompt techniques. Experimental results on the WTS \cite{kong2024wts} and BDD \cite{BDD} datasets demonstrate substantial gains in semantic richness and traffic scene interpretation. Our framework is validated through a decent test score of 55.655 in the AI City Challenge 2025 Track 2, showing its effectiveness in advancing resource-efficient and accurate traffic analysis for real-world applications.

Paperid: 6242, https://arxiv.org/pdf/2508.13439.pdf

Abstract:
Comprehensive highway scene understanding and robust traffic risk inference are vital for advancing Intelligent Transportation Systems (ITS) and autonomous driving. Traditional approaches often struggle with scalability and generalization, particularly under the complex and dynamic conditions of real-world environments. To address these challenges, we introduce a novel structured prompting and knowledge distillation framework that enables automatic generation of high-quality traffic scene annotations and contextual risk assessments. Our framework orchestrates two large Vision-Language Models (VLMs): GPT-4o and o3-mini, using a structured Chain-of-Thought (CoT) strategy to produce rich, multi-perspective outputs. These outputs serve as knowledge-enriched pseudo-annotations for supervised fine-tuning of a much smaller student VLM. The resulting compact 3B-scale model, named VISTA (Vision for Intelligent Scene and Traffic Analysis), is capable of understanding low-resolution traffic videos and generating semantically faithful, risk-aware captions. Despite its significantly reduced parameter count, VISTA achieves strong performance across established captioning metrics (BLEU-4, METEOR, ROUGE-L, and CIDEr) when benchmarked against its teacher models. This demonstrates that effective knowledge distillation and structured multi-agent supervision can empower lightweight VLMs to capture complex reasoning capabilities. The compact architecture of VISTA facilitates efficient deployment on edge devices, enabling real-time risk monitoring without requiring extensive infrastructure upgrades.

Paperid: 6243, https://arxiv.org/pdf/2508.12877.pdf

Abstract:
Pretrained vision-language models (VLMs), such as CLIP, have shown remarkable potential in few-shot image classification and led to numerous effective transfer learning strategies. These methods leverage the pretrained knowledge of VLMs to enable effective domain adaptation while mitigating overfitting through parameter-efficient tuning or instance-based consistency constraints. However, such regularizations often neglect the geometric structure of data distribution, which may lead to distortion of the overall semantic representation. To overcome this limitation, we propose a novel fine-tuning method, Manifold-Preserving and Sculpting Tuning (MPS-Tuning). Regarding the data distribution in feature space as a semantic manifold, MPS-Tuning explicitly constrains the intrinsic geometry of this manifold while further sculpting it to enhance class separability. Specifically, MPS-Tuning preserves both macroscopic and microscopic topological structures of the original manifold by aligning Gram matrices of features before and after fine-tuning. Theoretically, this constraint is shown to approximate an upper bound of the Gromov-Wasserstein distance. Furthermore, features from the image and text modalities are paired, and pairwise similarities are optimized to enhance the manifold's class discriminability. Extensive experiments demonstrate that MPS-Tuning significantly improves model performance while effectively preserving the structure of the semantic manifold. The code will be released.

Paperid: 6244, https://arxiv.org/pdf/2508.12543.pdf

Abstract:
The rapid advancement of generative models has intensified the challenge of detecting and interpreting visual forgeries, necessitating robust frameworks for image forgery detection while providing reasoning as well as localization. While existing works approach this problem using supervised training for specific manipulation or anomaly detection in the embedding space, generalization across domains remains a challenge. We frame this problem of forgery detection as a prompt-driven visual reasoning task, leveraging the semantic alignment capabilities of large vision-language models. We propose a framework, `REVEAL` (Reasoning and Evaluation of Visual Evidence through Aligned Language), that incorporates generalized guidelines. We propose two tangential approaches - (1) Holistic Scene-level Evaluation that relies on the physics, semantics, perspective, and realism of the image as a whole and (2) Region-wise anomaly detection that splits the image into multiple regions and analyzes each of them. We conduct experiments over datasets from different domains (Photoshop, DeepFake and AIGC editing). We compare the Vision Language Models against competitive baselines and analyze the reasoning provided by them.

Paperid: 6245, https://arxiv.org/pdf/2508.11960.pdf

Abstract:
Agile and human-centric manufacturing stipulates resilient robotic solutions capable of contextual reasoning and safe interaction in unstructured environments. Foundation models particularly the Vision Language Action (VLA) models have emerged to fuse multimodal perception, reasoning and physically grounded action across varied embodiments into unified representation, termed as General Physical Intelligence (GPI). While GPI has already been described in the literature but its practical application and evolving role in contemporary agile manufacturing processes have yet to be duly explored. To bridge this gap, this practical review systematically surveys recent advancements in VLA models within GPI context, performs comprehensive comparative analysis of leading implementations and evaluates their readiness for industrial deployment through structured ablation study. Our analysis has organized state-of-the-art into five thematic pillars including multisensory representation learning, sim2real transfer, planning and control, uncertainty and safety measures and benchmarking. Finally, we articulate open research challenges and propose directions to better integrate GPI into next-generation industrial ecosystems in line with Industry 5.0.

Paperid: 6246, https://arxiv.org/pdf/2508.11170.pdf

Abstract:
With the rapid advancement of vision language models(VLM), their ability to assess visual content based on specific criteria and dimensions has become increasingly critical for applications such as video-theme consistency assessment and visual quality scoring. However, existing methods often suffer from imprecise results and inefficient loss calculation, which limit the focus of the model on key evaluation indicators. To address this, we propose IOVQA(Integer-only VQA), a novel fine-tuning approach tailored for VLMs to enhance their performance in video quality assessment tasks. The key innovation of IOVQA lies in its label construction and its targeted loss calculation mechanism. Specifically, during dataset curation, we constrain the model's output to integers within the range of [10,50], ensuring numerical stability, and convert decimal Overall_MOS to integer before using them as labels. We also introduce a target-mask strategy: when computing the loss, only the first two-digit-integer of the label is unmasked, forcing the model to learn the critical components of the numerical evaluation. After fine-tuning the Qwen2.5-VL model using the constructed dataset, experimental results demonstrate that the proposed method significantly improves the model's accuracy and consistency in the VQA task, ranking 3rd in VQualA 2025 GenAI-Bench AIGC Video Quality Assessment Challenge -- Track I. Our work highlights the effectiveness of merely leaving integer labels during fine-tuning, providing an effective idea for optimizing VLMs in quantitative evaluation scenarios.

Paperid: 6247, https://arxiv.org/pdf/2508.08821.pdf

Abstract:
Recent Multi-Modal Large Language Models (MLLMs) have demonstrated strong capabilities in learning joint representations from text and images. However, their spatial reasoning remains limited. We introduce 3DFroMLLM, a novel framework that enables the generation of 3D object prototypes directly from MLLMs, including geometry and part labels. Our pipeline is agentic, comprising a designer, coder, and visual inspector operating in a refinement loop. Notably, our approach requires no additional training data or detailed user instructions. Building on prior work in 2D generation, we demonstrate that rendered images produced by our framework can be effectively used for image classification pretraining tasks and outperforms previous methods by 15%. As a compelling real-world use case, we show that the generated prototypes can be leveraged to improve fine-grained vision-language models by using the rendered, part-labeled prototypes to fine-tune CLIP for part segmentation and achieving a 55% accuracy improvement without relying on any additional human-labeled data.

Paperid: 6248, https://arxiv.org/pdf/2508.07021.pdf

Abstract:
The exponential growth of scientific literature in PDF format necessitates advanced tools for efficient and accurate document understanding, summarization, and content optimization. Traditional methods fall short in handling complex layouts and multimodal content, while direct application of Large Language Models (LLMs) and Vision-Language Large Models (LVLMs) lacks precision and control for intricate editing tasks. This paper introduces DocRefine, an innovative framework designed for intelligent understanding, content refinement, and automated summarization of scientific PDF documents, driven by natural language instructions. DocRefine leverages the power of advanced LVLMs (e.g., GPT-4o) by orchestrating a sophisticated multi-agent system comprising six specialized and collaborative agents: Layout & Structure Analysis, Multimodal Content Understanding, Instruction Decomposition, Content Refinement, Summarization & Generation, and Fidelity & Consistency Verification. This closed-loop feedback architecture ensures high semantic accuracy and visual fidelity. Evaluated on the comprehensive DocEditBench dataset, DocRefine consistently outperforms state-of-the-art baselines across various tasks, achieving overall scores of 86.7% for Semantic Consistency Score (SCS), 93.9% for Layout Fidelity Index (LFI), and 85.0% for Instruction Adherence Rate (IAR). These results demonstrate DocRefine's superior capability in handling complex multimodal document editing, preserving semantic integrity, and maintaining visual consistency, marking a significant advancement in automated scientific document processing.

Paperid: 6249, https://arxiv.org/pdf/2508.05244.pdf

Abstract:
Medical image understanding plays a crucial role in enabling automated diagnosis and data-driven clinical decision support. However, its progress is impeded by two primary challenges: the limited availability of high-quality annotated medical data and an overreliance on global image features, which often miss subtle but clinically significant pathological regions. To address these issues, we introduce RegionMed-CLIP, a region-aware multimodal contrastive learning framework that explicitly incorporates localized pathological signals along with holistic semantic representations. The core of our method is an innovative region-of-interest (ROI) processor that adaptively integrates fine-grained regional features with the global context, supported by a progressive training strategy that enhances hierarchical multimodal alignment. To enable large-scale region-level representation learning, we construct MedRegion-500k, a comprehensive medical image-text corpus that features extensive regional annotations and multilevel clinical descriptions. Extensive experiments on image-text retrieval, zero-shot classification, and visual question answering tasks demonstrate that RegionMed-CLIP consistently exceeds state-of-the-art vision language models by a wide margin. Our results highlight the critical importance of region-aware contrastive pre-training and position RegionMed-CLIP as a robust foundation for advancing multimodal medical image understanding.

Paperid: 6250, https://arxiv.org/pdf/2508.03488.pdf

Abstract:
We address the problem of scarcity of educational Arabic Language Learning tools that advocate modern pedagogical models such as active learning which ensures language proficiency. In fact, we investigate the design and evaluation of an AI-powered educational tool designed to enhance Arabic language learning for non-native speakers with beginner-to-intermediate proficiency level. The tool leverages advanced AI models to generate interactive visual quizzes, deploying Visual Question Answering as the primary activity. Adopting a constructivist learning approach, the system encourages active learning through real-life visual quizzes, and image-based questions that focus on improving vocabulary, grammar, and comprehension. The system integrates Vision-Language Pretraining models to generate contextually relevant image description from which Large Language Model generate assignments based on customized Arabic language Learning quizzes thanks to prompting. The effectiveness of the tool is evaluated through a manual annotated benchmark consisting of 1266 real-life visual quizzes, with human participants providing feedback. The results show a suitable accuracy rates, validating the tool's potential to bridge the gap in Arabic language education and highlighting the tool's promise as a reliable, AI-powered resource for Arabic learners, offering personalized and interactive learning experiences.

Paperid: 6251, https://arxiv.org/pdf/2508.03079.pdf

Abstract:
The rapid expansion of applications using Large Vision-Language Models (LVLMs), such as GPT-4o, has raised significant concerns about their fairness. While existing studies primarily focus on demographic attributes such as race and gender, fairness across a broader range of attributes remains largely unexplored. In this study, we construct an open-set knowledge base of bias attributes leveraging Large Language Models (LLMs) and evaluate the fairness of LVLMs across finer-grained attributes. Our experimental results reveal that LVLMs exhibit biased outputs across a diverse set of attributes and further demonstrate that cultural, environmental, and behavioral factors have a more pronounced impact on LVLM decision-making than traditional demographic attributes.

Paperid: 6252, https://arxiv.org/pdf/2508.01533.pdf

Abstract:
While recent multimodal models have shown progress in vision-language tasks, small-scale variants still struggle with the fine-grained temporal reasoning required for video understanding. We introduce ReasonAct, a method that enhances video reasoning in smaller models through a three-stage training process: first building a foundation with text-only reasoning, then fine-tuning on video, and finally refining with temporal-aware reinforcement learning. We build upon Temporal Group Relative Policy Optimization (T-GRPO) by incorporating temporal consistency modeling into policy optimization. We also propose a biomechanically-motivated sub-action decomposition mechanism that provides graduated rewards for constituent action phases. Through experiments on HMDB51, UCF-101, and Kinetics-400, our 3B-parameter model achieves 67.2%, 94.1%, and 78.9% accuracy respectively, demonstrating improvements of 17.9, 15.8, and 12.3 points over baselines. Ablation studies validate that our progressive training methodology enables smaller models to achieve competitive video reasoning performance while maintaining computational efficiency.

Paperid: 6253, https://arxiv.org/pdf/2507.19346.pdf

Abstract:
In recent years, social media users have spent significant amounts of time on short-form video platforms. As a result, established platforms in other domains, such as e-commerce, have begun introducing short-form video content to engage users and increase their time spent on the platform. The success of these experiences is due not only to the content itself but also to a unique UI innovation: instead of offering users a list of choices to click, platforms actively recommend content for users to watch one at a time. This creates new challenges for recommender systems, especially when launching a new video experience. Beyond the limited interaction data, immersive feed experiences introduce stronger position bias due to the UI and duration bias when optimizing for watch-time, as models tend to favor shorter videos. These issues, together with the feedback loop inherent in recommender systems, make it difficult to build effective solutions. In this paper, we highlight the challenges faced when introducing a new short-form video experience and present our experience showing that, even with sufficient video interaction data, it can be more beneficial to leverage a video retrieval system using a fine-tuned multimodal vision-language model to overcome these challenges. This approach demonstrated greater effectiveness compared to conventional supervised learning methods in online experiments conducted on our e-commerce platform.

Paperid: 6254, https://arxiv.org/pdf/2507.17383.pdf

Abstract:
Trustworthy robot behavior requires not only high levels of task success but also that the robot can reliably quantify how likely it is to succeed. To this end, we present the first systematic study of confidence calibration in vision-language-action (VLA) foundation models, which map visual observations and natural-language instructions to low-level robot motor commands. We begin with extensive benchmarking to understand the critical relationship between task success and calibration error across multiple datasets and VLA variants, finding that task performance and calibration are not in tension. Next, we introduce prompt ensembles for VLAs, a lightweight, Bayesian-inspired algorithm that averages confidence across paraphrased instructions and consistently improves calibration. We further analyze calibration over the task time horizon, showing that confidence is often most reliable after making some progress, suggesting natural points for risk-aware intervention. Finally, we reveal differential miscalibration across action dimensions and propose action-wise Platt scaling, a method to recalibrate each action dimension independently to produce better confidence estimates. Our aim in this study is to begin to develop the tools and conceptual understanding necessary to render VLAs both highly performant and highly trustworthy via reliable uncertainty quantification.

Paperid: 6255, https://arxiv.org/pdf/2507.17379.pdf

Abstract:
Open-vocabulary mobile manipulation (OVMM) that involves the handling of novel and unseen objects across different workspaces remains a significant challenge for real-world robotic applications. In this paper, we propose a novel Language-conditioned Open-Vocabulary Mobile Manipulation framework, named LOVMM, incorporating the large language model (LLM) and vision-language model (VLM) to tackle various mobile manipulation tasks in household environments. Our approach is capable of solving various OVMM tasks with free-form natural language instructions (e.g. "toss the food boxes on the office room desk to the trash bin in the corner", and "pack the bottles from the bed to the box in the guestroom"). Extensive experiments simulated in complex household environments show strong zero-shot generalization and multi-task learning abilities of LOVMM. Moreover, our approach can also generalize to multiple tabletop manipulation tasks and achieve better success rates compared to other state-of-the-art methods.

Paperid: 6256, https://arxiv.org/pdf/2507.16704.pdf

Abstract:
Desktop accessibility metadata enables AI agents to interpret screens and supports users who depend on tools like screen readers. Yet, many applications remain largely inaccessible due to incomplete or missing metadata provided by developers - our investigation shows that only 33% of applications on macOS offer full accessibility support. While recent work on structured screen representation has primarily addressed specific challenges, such as UI element detection or captioning, none has attempted to capture the full complexity of desktop interfaces by replicating their entire hierarchical structure. To bridge this gap, we introduce Screen2AX, the first framework to automatically create real-time, tree-structured accessibility metadata from a single screenshot. Our method uses vision-language and object detection models to detect, describe, and organize UI elements hierarchically, mirroring macOS's system-level accessibility structure. To tackle the limited availability of data for macOS desktop applications, we compiled and publicly released three datasets encompassing 112 macOS applications, each annotated for UI element detection, grouping, and hierarchical accessibility metadata alongside corresponding screenshots. Screen2AX accurately infers hierarchy trees, achieving a 77% F1 score in reconstructing a complete accessibility tree. Crucially, these hierarchy trees improve the ability of autonomous agents to interpret and interact with complex desktop interfaces. We introduce Screen2AX-Task, a benchmark specifically designed for evaluating autonomous agent task execution in macOS desktop environments. Using this benchmark, we demonstrate that Screen2AX delivers a 2.2x performance improvement over native accessibility representations and surpasses the state-of-the-art OmniParser V2 system on the ScreenSpot benchmark.

Paperid: 6257, https://arxiv.org/pdf/2507.12490.pdf

Abstract:
We introduce EaGERS, a fully training-free and model-agnostic pipeline that (1) generates natural language rationales via a vision language model, (2) grounds these rationales to spatial sub-regions by computing multimodal embedding similarities over a configurable grid with majority voting, and (3) restricts the generation of responses only from the relevant regions selected in the masked image. Experiments on the DocVQA dataset demonstrate that our best configuration not only outperforms the base model on exact match accuracy and Average Normalized Levenshtein Similarity metrics but also enhances transparency and reproducibility in DocVQA without additional model fine-tuning.

Paperid: 6258, https://arxiv.org/pdf/2507.09815.pdf

Abstract:
Ensuring the safety of vulnerable road users (VRUs), such as pedestrians and cyclists, is a critical challenge for autonomous driving systems, as crashes involving VRUs often result in severe or fatal consequences. While multimodal large language models (MLLMs) have shown promise in enhancing scene understanding and decision making in autonomous vehicles, there is currently no standardized benchmark to quantitatively evaluate their reasoning abilities in complex, safety-critical scenarios involving VRUs. To address this gap, we present VRU-Accident, a large-scale vision-language benchmark designed to evaluate MLLMs in high-risk traffic scenarios involving VRUs. VRU-Accident comprises 1K real-world dashcam accident videos, annotated with 6K multiple-choice question-answer pairs across six safety-critical categories (with 24K candidate options and 3.4K unique answer choices), as well as 1K dense scene descriptions. Unlike prior works, our benchmark focuses explicitly on VRU-vehicle accidents, providing rich, fine-grained annotations that capture both spatial-temporal dynamics and causal semantics of accidents. To assess the current landscape of MLLMs, we conduct a comprehensive evaluation of 17 state-of-the-art models on the multiple-choice VQA task and on the dense captioning task. Our findings reveal that while MLLMs perform reasonably well on visually grounded attributes, they face significant challenges in reasoning and describing accident causes, types, and preventability.

Paperid: 6259, https://arxiv.org/pdf/2507.08679.pdf

Abstract:
We introduce ByDeWay, a training-free framework designed to enhance the performance of Multimodal Large Language Models (MLLMs). ByDeWay uses a novel prompting strategy called Layered-Depth-Based Prompting (LDP), which improves spatial reasoning and grounding without modifying any model parameters. It segments the scene into closest, mid-range, and farthest layers using monocular depth estimation, then generates region-specific captions with a grounded vision-language model. These structured, depth-aware captions are appended to the image-question prompt, enriching it with spatial context. This guides MLLMs to produce more grounded and less hallucinated responses. Our method is lightweight, modular, and compatible with black-box MLLMs. Experiments on hallucination-sensitive (POPE) and reasoning-intensive (GQA) benchmarks show consistent improvements across multiple MLLMs, validating the effectiveness of depth-aware prompting in a zero-training setting.

Paperid: 6260, https://arxiv.org/pdf/2507.06973.pdf

Abstract:
Vision-Language Models (VLMs) have become prominent in open-world image recognition for their strong generalization abilities. Yet, their effectiveness in practical applications is compromised by domain shifts and distributional changes, especially when test data distributions diverge from training data. Therefore, the paradigm of test-time adaptation (TTA) has emerged, enabling the use of online off-the-shelf data at test time, supporting independent sample predictions, and eliminating reliance on test annotations. Traditional TTA methods, however, often rely on costly training or optimization processes, or make unrealistic assumptions about accessing or storing historical training and test data. Instead, this study proposes FreeTTA, a training-free and universally available method that makes no assumptions, to enhance the flexibility of TTA. More importantly, FreeTTA is the first to explicitly model the test data distribution, enabling the use of intrinsic relationships among test samples to enhance predictions of individual samples without simultaneous access--a direction not previously explored. FreeTTA achieves these advantages by introducing an online EM algorithm that utilizes zero-shot predictions from VLMs as priors to iteratively compute the posterior probabilities of each online test sample and update parameters. Experiments demonstrate that FreeTTA achieves stable and significant improvements compared to state-of-the-art methods across 15 datasets in both cross-domain and out-of-distribution settings.

Paperid: 6261, https://arxiv.org/pdf/2507.06510.pdf

Abstract:
Open vocabulary Human-Object Interaction (HOI) detection is a challenging task that detects all triplets of interest in an image, even those that are not pre-defined in the training set. Existing approaches typically rely on output features generated by large Vision-Language Models (VLMs) to enhance the generalization ability of interaction representations. However, the visual features produced by VLMs are holistic and coarse-grained, which contradicts the nature of detection tasks. To address this issue, we propose a novel Bilateral Collaboration framework for open vocabulary HOI detection (BC-HOI). This framework includes an Attention Bias Guidance (ABG) component, which guides the VLM to produce fine-grained instance-level interaction features according to the attention bias provided by the HOI detector. It also includes a Large Language Model (LLM)-based Supervision Guidance (LSG) component, which provides fine-grained token-level supervision for the HOI detector by the LLM component of the VLM. LSG enhances the ability of ABG to generate high-quality attention bias. We conduct extensive experiments on two popular benchmarks: HICO-DET and V-COCO, consistently achieving superior performance in the open vocabulary and closed settings. The code will be released in Github.

Paperid: 6262, https://arxiv.org/pdf/2507.06441.pdf

Abstract:
In this paper, we introduce VisioPath, a novel framework combining vision-language models (VLMs) with model predictive control (MPC) to enable safe autonomous driving in dynamic traffic environments. The proposed approach leverages a bird's-eye view video processing pipeline and zero-shot VLM capabilities to obtain structured information about surrounding vehicles, including their positions, dimensions, and velocities. Using this rich perception output, we construct elliptical collision-avoidance potential fields around other traffic participants, which are seamlessly integrated into a finite-horizon optimal control problem for trajectory planning. The resulting trajectory optimization is solved via differential dynamic programming with an adaptive regularization scheme and is embedded in an event-triggered MPC loop. To ensure collision-free motion, a safety verification layer is incorporated in the framework that provides an assessment of potential unsafe trajectories. Extensive simulations in Simulation of Urban Mobility (SUMO) demonstrate that VisioPath outperforms conventional MPC baselines across multiple metrics. By combining modern AI-driven perception with the rigorous foundation of optimal control, VisioPath represents a significant step forward in safe trajectory planning for complex traffic systems.

Paperid: 6263, https://arxiv.org/pdf/2507.05799.pdf

Abstract:
One of the main objectives in developing large vision-language models (LVLMs) is to engineer systems that can assist humans with multimodal tasks, including interpreting descriptions of perceptual experiences. A central phenomenon in this context is amodal completion, in which people perceive objects even when parts of those objects are hidden. Although numerous studies have assessed whether computer-vision algorithms can detect or reconstruct occluded regions, the inferential abilities of LVLMs on texts related to amodal completion remain unexplored. To address this gap, we constructed a benchmark grounded in Basic Formal Ontology to achieve a systematic classification of amodal completion. Our results indicate that while many LVLMs achieve human-comparable performance overall, their accuracy diverges for certain types of objects being completed. Notably, in certain categories, some LLaVA-NeXT variants and Claude 3.5 Sonnet exhibit lower accuracy on original images compared to blank stimuli lacking visual content. Intriguingly, this disparity emerges only under Japanese prompting, suggesting a deficiency in Japanese-specific linguistic competence among these models.

Paperid: 6264, https://arxiv.org/pdf/2507.02978.pdf

Abstract:
Humans naturally possess the spatial reasoning ability to form and manipulate images and structures of objects in space. There is an increasing effort to endow Vision-Language Models (VLMs) with similar spatial reasoning capabilities. However, it remains unclear whether these models truly understand and manipulate spatial objects or not. To address this question, we propose a new evaluation framework aimed at assessing the performance of VLMs in spatial deformation reasoning tasks. Specifically, we construct a benchmark for spatial deformation reasoning from 2D to 3D. Leveraging our data engine, we can generate unlimited evaluation problem pairs with infinite steps, without any data leakage. We explore whether the model can effectively perform spatial deformation reasoning from two directions: forward reasoning (given the operations, find the final state) and reverse reasoning (given the final state, determine the operations). We adopt a ladder competition format, using the number of deformation steps as the level classification criterion, with the goal of exploring the boundaries of the model's deformation reasoning capabilities. Interestingly, the benchmarking results reveal that almost no model demonstrates plausible spatial deformation reasoning abilities. Furthermore, even after applying targeted training and mainstream reasoning enhancement methods, the models are still unable to perform well on 3D spatial deformation reasoning.

Paperid: 6265, https://arxiv.org/pdf/2507.00700.pdf

Abstract:
Cross-cultural research in perception and cognition has shown that individuals from different cultural backgrounds process visual information in distinct ways. East Asians, for example, tend to adopt a holistic perspective, attending to contextual relationships, whereas Westerners often employ an analytical approach, focusing on individual objects and their attributes. In this study, we investigate whether Vision-Language Models (VLMs) trained predominantly on different languages, specifically Japanese and English, exhibit similar culturally grounded attentional patterns. Using comparative analysis of image descriptions, we examine whether these models reflect differences in holistic versus analytic tendencies. Our findings suggest that VLMs not only internalize the structural properties of language but also reproduce cultural behaviors embedded in the training data, indicating that cultural cognition may implicitly shape model outputs.

Paperid: 6266, https://arxiv.org/pdf/2506.23465.pdf

Abstract:
The success of machine learning models in industrial applications is heavily dependent on the quality of the datasets used to train the models. However, large-scale datasets, specially those constructed from crowd-sourcing and web-scraping, often suffer from label noise, inconsistencies, and errors. This problem is particularly pronounced in manufacturing domains, where obtaining high-quality labels is costly and time-consuming. This paper introduces Vision-Language Sanitization and Refinement (VLSR), which is a vision-language-based framework for label sanitization and refinement in multi-label manufacturing image datasets. This method embeds both images and their associated textual labels into a shared semantic space leveraging the CLIP vision-language model. Then two key tasks are addressed in this process by computing the cosine similarity between embeddings. First, label sanitization is performed to identify irrelevant, misspelled, or semantically weak labels, and surface the most semantically aligned label for each image by comparing image-label pairs using cosine similarity between image and label embeddings. Second, the method applies density-based clustering on text embeddings, followed by iterative cluster merging, to group semantically similar labels into unified label groups. The Factorynet dataset, which includes noisy labels from both human annotations and web-scraped sources, is employed to evaluate the effectiveness of the proposed framework. Experimental results demonstrate that the VLSR framework successfully identifies problematic labels and improves label consistency. This method enables a significant reduction in label vocabulary through clustering, which ultimately enhances the dataset's quality for training robust machine learning models in industrial applications with minimal human intervention.

Paperid: 6267, https://arxiv.org/pdf/2506.22506.pdf

Abstract:
Federated Prompt Learning has emerged as a communication-efficient and privacy-preserving paradigm for adapting large vision-language models like CLIP across decentralized clients. However, the security implications of this setup remain underexplored. In this work, we present the first study of backdoor attacks in Federated Prompt Learning. We show that when malicious clients inject visually imperceptible, learnable noise triggers into input images, the global prompt learner becomes vulnerable to targeted misclassification while still maintaining high accuracy on clean inputs. Motivated by this vulnerability, we propose SABRE-FL, a lightweight, modular defense that filters poisoned prompt updates using an embedding-space anomaly detector trained offline on out-of-distribution data. SABRE-FL requires no access to raw client data or labels and generalizes across diverse datasets. We show, both theoretically and empirically, that malicious clients can be reliably identified and filtered using an embedding-based detector. Across five diverse datasets and four baseline defenses, SABRE-FL outperforms all baselines by significantly reducing backdoor accuracy while preserving clean accuracy, demonstrating strong empirical performance and underscoring the need for robust prompt learning in future federated systems.

Paperid: 6268, https://arxiv.org/pdf/2506.21316.pdf

Abstract:
Visual grounding in text-rich document images is a critical yet underexplored challenge for Document Intelligence and Visual Question Answering (VQA) systems. We present DRISHTIKON, a multi-granular and multi-block visual grounding framework designed to enhance interpretability and trust in VQA for complex, multilingual documents. Our approach integrates multilingual OCR, large language models, and a novel region matching algorithm to localize answer spans at the block, line, word, and point levels. We introduce the Multi-Granular Visual Grounding (MGVG) benchmark, a curated test set of diverse circular notifications from various sectors, each manually annotated with fine-grained, human-verified labels across multiple granularities. Extensive experiments show that our method achieves state-of-the-art grounding accuracy, with line-level granularity providing the best balance between precision and recall. Ablation studies further highlight the benefits of multi-block and multi-line reasoning. Comparative evaluations reveal that leading vision-language models struggle with precise localization, underscoring the effectiveness of our structured, alignment-based approach. Our findings pave the way for more robust and interpretable document understanding systems in real-world, text-centric scenarios with multi-granular grounding support. Code and dataset are made available for future research.

Paperid: 6269, https://arxiv.org/pdf/2506.20214.pdf

Abstract:
Unified multimodal large language models (MLLMs) have shown promise in jointly advancing multimodal understanding and generation, with visual codebooks discretizing images into tokens for autoregressive modeling. Existing codebook-based methods either rely on small vocabularies (~16K entries) that lack fine-grained semantics or naively scale up, resulting in low token utilization and unstable training. We propose UniCode$^2$, a cascaded codebook framework enabling large-scale, semantically aligned, and stable visual tokenization. By clustering millions of SigLIP sequence embeddings, we build a 500K-entry codebook that preserves vision-language alignment while expanding capacity. Stability is ensured via a cascaded design: a frozen codebook anchors the embedding space, and a trainable codebook refines task-specific semantics. This decoupling promotes high utilization and robust learning. Moreover, the alignment of our visual tokens with textual semantics enables seamless integration with pretrained diffusion decoders, supporting high-quality visual synthesis with minimal adaptation. UniCode^2 delivers strong performance across diverse benchmarks, demonstrating the viability of scaling visual token spaces without sacrificing stability, semantics, or modularity.

Paperid: 6270, https://arxiv.org/pdf/2506.16994.pdf

Abstract:
Unsupervised Domain Adaptation (UDA) is a critical challenge in real-world vision systems, especially in resource-constrained environments like drones, where memory and computation are limited. Existing prompt-driven UDA methods typically rely on large vision-language models and require full access to source-domain data during adaptation, limiting their applicability. In this work, we propose Prmpt2Adpt, a lightweight and efficient zero-shot domain adaptation framework built around a teacher-student paradigm guided by prompt-based feature alignment. At the core of our method is a distilled and fine-tuned CLIP model, used as the frozen backbone of a Faster R-CNN teacher. A small set of low-level source features is aligned to the target domain semantics-specified only through a natural language prompt-via Prompt-driven Instance Normalization (PIN). These semantically steered features are used to briefly fine-tune the detection head of the teacher model. The adapted teacher then generates high-quality pseudo-labels, which guide the on-the-fly adaptation of a compact student model. Experiments on the MDS-A dataset demonstrate that Prmpt2Adpt achieves competitive detection performance compared to state-of-the-art methods, while delivering up to 7x faster adaptation and 5x faster inference speed using few source images-making it a practical and scalable solution for real-time adaptation in low-resource domains.

Paperid: 6271, https://arxiv.org/pdf/2506.16716.pdf

Abstract:
Automatic video commentary systems are widely used on multimedia social media platforms to extract factual information about video content. However, current systems may overlook essential para-linguistic cues, including emotion and attitude, which are critical for fully conveying the meaning of visual content. The absence of these cues can limit user understanding or, in some cases, distort the video's original intent. Expressive speech effectively conveys these cues and enhances the user's comprehension of videos. Building on these insights, this paper explores the usage of vision-context-aware expressive speech in enhancing users' understanding of videos in video commentary systems. Firstly, our formatting study indicates that semantic-only speech can lead to ambiguity, and misaligned emotions between speech and visuals may distort content interpretation. To address this, we propose a method called vision-context-aware speech synthesis (V-CASS). It analyzes para-linguistic cues from visuals using a vision-language model and leverages a knowledge-infused language model to guide the expressive speech model in generating context-aligned speech. User studies show that V-CASS enhances emotional and attitudinal resonance, as well as user audio-visual understanding and engagement, with 74.68% of participants preferring the system. Finally, we explore the potential of our method in helping blind and low-vision users navigate web videos, improving universal accessibility.

Paperid: 6272, https://arxiv.org/pdf/2506.12411.pdf

Abstract:
Multimodal contrastive learning models like CLIP have demonstrated remarkable vision-language alignment capabilities, yet their vulnerability to backdoor attacks poses critical security risks. Attackers can implant latent triggers that persist through downstream tasks, enabling malicious control of model behavior upon trigger presentation. Despite great success in recent defense mechanisms, they remain impractical due to strong assumptions about attacker knowledge or excessive clean data requirements. In this paper, we introduce InverTune, the first backdoor defense framework for multimodal models under minimal attacker assumptions, requiring neither prior knowledge of attack targets nor access to the poisoned dataset. Unlike existing defense methods that rely on the same dataset used in the poisoning stage, InverTune effectively identifies and removes backdoor artifacts through three key components, achieving robust protection against backdoor attacks. Specifically, InverTune first exposes attack signatures through adversarial simulation, probabilistically identifying the target label by analyzing model response patterns. Building on this, we develop a gradient inversion technique to reconstruct latent triggers through activation pattern analysis. Finally, a clustering-guided fine-tuning strategy is employed to erase the backdoor function with only a small amount of arbitrary clean data, while preserving the original model capabilities. Experimental results show that InverTune reduces the average attack success rate (ASR) by 97.87% against the state-of-the-art (SOTA) attacks while limiting clean accuracy (CA) degradation to just 3.07%. This work establishes a new paradigm for securing multimodal systems, advancing security in foundation model deployment without compromising performance.

Paperid: 6273, https://arxiv.org/pdf/2506.11684.pdf

Abstract:
Vision-Language Models (VLMs) have demonstrated remarkable capabilities in interpreting visual layouts and text. However, a significant challenge remains in their ability to interpret robustly and reason over multi-tabular data presented as images, a common occurrence in real-world scenarios like web pages and digital documents. Existing benchmarks typically address single tables or non-visual data (text/structured). This leaves a critical gap: they don't assess the ability to parse diverse table images, correlate information across them, and perform multi-hop reasoning on the combined visual data. We introduce MTabVQA, a novel benchmark specifically designed for multi-tabular visual question answering to bridge that gap. MTabVQA comprises 3,745 complex question-answer pairs that necessitate multi-hop reasoning across several visually rendered table images. We provide extensive benchmark results for state-of-the-art VLMs on MTabVQA, revealing significant performance limitations. We further investigate post-training techniques to enhance these reasoning abilities and release MTabVQA-Instruct, a large-scale instruction-tuning dataset. Our experiments show that fine-tuning VLMs with MTabVQA-Instruct substantially improves their performance on visual multi-tabular reasoning. Code and dataset (https://huggingface.co/datasets/mtabvqa/MTabVQA-Eval) are available online (https://anonymous.4open.science/r/MTabVQA-EMNLP-B16E).

Paperid: 6274, https://arxiv.org/pdf/2506.09172.pdf

Abstract:
Recent innovations in multimodal action models represent a promising direction for developing general-purpose agentic systems, combining visual understanding, language comprehension, and action generation. We introduce MultiNet - a novel, fully open-source benchmark and surrounding software ecosystem designed to rigorously evaluate and adapt models across vision, language, and action domains. We establish standardized evaluation protocols for assessing vision-language models (VLMs) and vision-language-action models (VLAs), and provide open source software to download relevant data, models, and evaluations. Additionally, we provide a composite dataset with over 1.3 trillion tokens of image captioning, visual question answering, commonsense reasoning, robotic control, digital game-play, simulated locomotion/manipulation, and many more tasks. The MultiNet benchmark, framework, toolkit, and evaluation harness have been used in downstream research on the limitations of VLA generalization.

Paperid: 6275, https://arxiv.org/pdf/2506.09067.pdf

Abstract:
Generative medical vision-language models~(Med-VLMs) are primarily designed to generate complex textual information~(e.g., diagnostic reports) from multimodal inputs including vision modality~(e.g., medical images) and language modality~(e.g., clinical queries). However, their security vulnerabilities remain underexplored. Med-VLMs should be capable of rejecting harmful queries, such as \textit{Provide detailed instructions for using this CT scan for insurance fraud}. At the same time, addressing security concerns introduces the risk of over-defense, where safety-enhancing mechanisms may degrade general performance, causing Med-VLMs to reject benign clinical queries. In this paper, we propose a novel inference-time defense strategy to mitigate harmful queries, enabling defense against visual and textual jailbreak attacks. Using diverse medical imaging datasets collected from nine modalities, we demonstrate that our defense strategy based on synthetic clinical demonstrations enhances model safety without significantly compromising performance. Additionally, we find that increasing the demonstration budget alleviates the over-defense issue. We then introduce a mixed demonstration strategy as a trade-off solution for balancing security and performance under few-shot demonstration budget constraints.

Paperid: 6276, https://arxiv.org/pdf/2506.08356.pdf

Abstract:
Different medical imaging modalities capture diagnostic information at varying spatial resolutions, from coarse global patterns to fine-grained localized structures. However, most existing vision-language frameworks in the medical domain apply a uniform strategy for local feature extraction, overlooking the modality-specific demands. In this work, we present MedMoE, a modular and extensible vision-language processing framework that dynamically adapts visual representation based on the diagnostic context. MedMoE incorporates a Mixture-of-Experts (MoE) module conditioned on the report type, which routes multi-scale image features through specialized expert branches trained to capture modality-specific visual semantics. These experts operate over feature pyramids derived from a Swin Transformer backbone, enabling spatially adaptive attention to clinically relevant regions. This framework produces localized visual representations aligned with textual descriptions, without requiring modality-specific supervision at inference. Empirical results on diverse medical benchmarks demonstrate that MedMoE improves alignment and retrieval performance across imaging modalities, underscoring the value of modality-specialized visual representations in clinical vision-language systems.

Paperid: 6277, https://arxiv.org/pdf/2506.08296.pdf

Abstract:
Recent advances in multimodal vision-language-action (VLA) models have revolutionized traditional robot learning, enabling systems to interpret vision, language, and action in unified frameworks for complex task planning. However, mastering complex manipulation tasks remains an open challenge, constrained by limitations in persistent contextual memory, multi-agent coordination under uncertainty, and dynamic long-horizon planning across variable sequences. To address this challenge, we propose \textbf{HiBerNAC}, a \textbf{Hi}erarchical \textbf{B}rain-\textbf{e}mulated \textbf{r}obotic \textbf{N}eural \textbf{A}gent \textbf{C}ollective, inspired by breakthroughs in neuroscience, particularly in neural circuit mechanisms and hierarchical decision-making. Our framework combines: (1) multimodal VLA planning and reasoning with (2) neuro-inspired reflection and multi-agent mechanisms, specifically designed for complex robotic manipulation tasks. By leveraging neuro-inspired functional modules with decentralized multi-agent collaboration, our approach enables robust and enhanced real-time execution of complex manipulation tasks. In addition, the agentic system exhibits scalable collective intelligence via dynamic agent specialization, adapting its coordination strategy to variable task horizons and complexity. Through extensive experiments on complex manipulation tasks compared with state-of-the-art VLA models, we demonstrate that \textbf{HiBerNAC} reduces average long-horizon task completion time by 23\%, and achieves non-zero success rates (12\textendash 31\%) on multi-path tasks where prior state-of-the-art VLA models consistently fail. These results provide indicative evidence for bridging biological cognition and robotic learning mechanisms.

Paperid: 6278, https://arxiv.org/pdf/2506.08008.pdf

Abstract:
Language provides a natural interface to specify and evaluate performance on visual tasks. To realize this possibility, vision language models (VLMs) must successfully integrate visual and linguistic information. Our work compares VLMs to a direct readout of their visual encoders to understand their ability to integrate across these modalities. Across a series of vision-centric benchmarks (e.g., depth estimation, correspondence), we find that VLMs perform substantially worse than their visual encoders, dropping to near-chance performance. We investigate these results through a series of analyses across the entire VLM: namely 1) the degradation of vision representations, 2) brittleness to task prompt, and 3) the language model's role in solving the task. We find that the bottleneck in performing these vision-centric tasks lies in this third category; VLMs are not effectively using visual information easily accessible throughout the entire model, and they inherit the language priors present in the LLM. Our work helps diagnose the failure modes of open-source VLMs, and presents a series of evaluations useful for future investigations into visual understanding within VLMs.

Paperid: 6279, https://arxiv.org/pdf/2506.05696.pdf

Abstract:
Recent advances in vision-language models have enabled rich semantic understanding across modalities. However, these encoding methods lack the ability to interpret or reason about the moral dimensions of content-a crucial aspect of human cognition. In this paper, we address this gap by introducing MoralCLIP, a novel embedding representation method that extends multimodal learning with explicit moral grounding based on Moral Foundations Theory (MFT). Our approach integrates visual and textual moral cues into a unified embedding space, enabling cross-modal moral alignment. MoralCLIP is grounded on the multi-label dataset Social-Moral Image Database to identify co-occurring moral foundations in visual content. For MoralCLIP training, we design a moral data augmentation strategy to scale our annotated dataset to 15,000 image-text pairs labeled with MFT-aligned dimensions. Our results demonstrate that explicit moral supervision improves both unimodal and multimodal understanding of moral content, establishing a foundation for morally-aware AI systems capable of recognizing and aligning with human moral values.

Paperid: 6280, https://arxiv.org/pdf/2506.03521.pdf

Abstract:
Universal Domain Adaptation (UniDA) focuses on transferring source domain knowledge to the target domain under both domain shift and unknown category shift. Its main challenge lies in identifying common class samples and aligning them. Current methods typically obtain target domain semantics centers from an unconstrained continuous image representation space. Due to domain shift and the unknown number of clusters, these centers often result in complex and less robust alignment algorithm. In this paper, based on vision-language models, we search for semantic centers in a semantically meaningful and discrete text representation space. The constrained space ensures almost no domain bias and appropriate semantic granularity for these centers, enabling a simple and robust adaptation algorithm. Specifically, we propose TArget Semantics Clustering (TASC) via Text Representations, which leverages information maximization as a unified objective and involves two stages. First, with the frozen encoders, a greedy search-based framework is used to search for an optimal set of text embeddings to represent target semantics. Second, with the search results fixed, encoders are refined based on gradient descent, simultaneously achieving robust domain alignment and private class clustering. Additionally, we propose Universal Maximum Similarity (UniMS), a scoring function tailored for detecting open-set samples in UniDA. Experimentally, we evaluate the universality of UniDA algorithms under four category shift scenarios. Extensive experiments on four benchmarks demonstrate the effectiveness and robustness of our method, which has achieved state-of-the-art performance.

Paperid: 6281, https://arxiv.org/pdf/2506.02865.pdf

Authors:Mathieu Andreux, Breno Baldas Skuk, Hamza Benchekroun, Emilien BirÃ©, Antoine Bonnet, Riaz Bordie, Nathan Bout, Matthias Brunel, Pierre-Louis Cedoz, Antoine Chassang, MickaÃ«l Chen, Alexandra D. Constantinou, Antoine d'AndignÃ©, Hubert de La JonquiÃ¨re, AurÃ©lien Delfosse, Ludovic Denoyer, Alexis Deprez, Augustin Derupti, Michael Eickenberg, MathÃ¯s Federico, Charles Kantor, Xavier Koegler, Yann LabbÃ©, Matthew C. H. Lee, Erwan Le Jumeau de Kergaradec, Amir Mahla, Avshalom Manevich, Adrien Maret, Charles Masson, RafaÃ«l Maurin, Arturo Mena, Philippe Modard, Axel Moyal, Axel Nguyen Kerbel, Julien Revelle, Mats L. Richter, MarÃa Santos, Laurent Sifre, Maxime Theillard, Marc Thibault, Louis Thiry, LÃ©o Tronchon, Nicolas Usunier, Tony Wu

Abstract:
We present Surfer-H, a cost-efficient web agent that integrates Vision-Language Models (VLM) to perform user-defined tasks on the web. We pair it with Holo1, a new open-weight collection of VLMs specialized in web navigation and information extraction. Holo1 was trained on carefully curated data sources, including open-access web content, synthetic examples, and self-produced agentic data. Holo1 tops generalist User Interface (UI) benchmarks as well as our new web UI localization benchmark, WebClick. When powered by Holo1, Surfer-H achieves a 92.2% state-of-the-art performance on WebVoyager, striking a Pareto-optimal balance between accuracy and cost-efficiency. To accelerate research advancement in agentic systems, we are open-sourcing both our WebClick evaluation dataset and the Holo1 model weights.

Paperid: 6282, https://arxiv.org/pdf/2506.02359.pdf

Abstract:
Great labels make great models. However, traditional labeling approaches for tasks like object detection have substantial costs at scale. Furthermore, alternatives to fully-supervised object detection either lose functionality or require larger models with prohibitive computational costs for inference at scale. To that end, this paper addresses the problem of training standard object detection models without any ground truth labels. Instead, we configure previously-trained vision-language foundation models to generate application-specific pseudo "ground truth" labels. These auto-generated labels directly integrate with existing model training frameworks, and we subsequently train lightweight detection models that are computationally efficient. In this way, we avoid the costs of traditional labeling, leverage the knowledge of vision-language models, and keep the efficiency of lightweight models for practical application. We perform exhaustive experiments across multiple labeling configurations, downstream inference models, and datasets to establish best practices and set an extensive auto-labeling benchmark. From our results, we find that our approach is a viable alternative to standard labeling in that it maintains competitive performance on multiple datasets and substantially reduces labeling time and costs.

Paperid: 6283, https://arxiv.org/pdf/2506.01371.pdf

Abstract:
Spatial reasoning remains a critical yet underdeveloped capability in existing vision-language models (VLMs), especially for Spatial Visual Question Answering (Spatial VQA) tasks that require understanding relative positions, distances, and object configurations. Inspired by the R1 paradigm introduced in DeepSeek-R1, which enhances reasoning in language models through rule-based reinforcement learning (RL), we propose SVQA-R1, the first framework to extend R1-style training to spatial VQA. In particular, we introduce Spatial-GRPO, a novel group-wise RL strategy that constructs view-consistent rewards by perturbing spatial relations between objects, e.g., mirror flipping, thereby encouraging the model to develop a consistent and grounded understanding of space. Our model, SVQA-R1, not only achieves dramatically improved accuracy on spatial VQA benchmarks but also exhibits interpretable reasoning paths even without using supervised fine-tuning (SFT) data. Extensive experiments and visualization demonstrate the effectiveness of SVQA-R1 across multiple spatial reasoning benchmarks.

Paperid: 6284, https://arxiv.org/pdf/2505.22651.pdf

Abstract:
Reasoning Vision-Language Models (VLMs) have shown promising performance on complex multimodal tasks. However, they still face significant challenges: they are highly sensitive to reasoning errors, require large volumes of annotated data or accurate verifiers, and struggle to generalize beyond specific domains. To address these limitations, we explore self-correction as a strategy to enhance reasoning VLMs. We first conduct an in-depth analysis of reasoning VLMs' self-correction abilities and identify key gaps. Based on our findings, we introduce Sherlock, a self-correction and self-improvement training framework. Sherlock introduces a trajectory-level self-correction objective, a preference data construction method based on visual perturbation, and a dynamic $Î²$ for preference tuning. Once the model acquires self-correction capabilities using only 20k randomly sampled annotated data, it continues to self-improve without external supervision. Built on the Llama3.2-Vision-11B model, Sherlock achieves remarkable results across eight benchmarks, reaching an average accuracy of 64.1 with direct generation and 65.4 after self-correction. It outperforms LLaVA-CoT (63.2), Mulberry (63.9), and LlamaV-o1 (63.4) while using less than 20% of the annotated data.

Paperid: 6285, https://arxiv.org/pdf/2505.22200.pdf

Abstract:
To understand a prompt, Vision-Language models (VLMs) must perceive the image, comprehend the text, and build associations within and across both modalities. For instance, given an 'image of a red toy car', the model should associate this image to phrases like 'car', 'red toy', 'red object', etc. Feng and Steinhardt propose the Binding ID mechanism in LLMs, suggesting that the entity and its corresponding attribute tokens share a Binding ID in the model activations. We investigate this for image-text binding in VLMs using a synthetic dataset and task that requires models to associate 3D objects in an image with their descriptions in the text. Our experiments demonstrate that VLMs assign a distinct Binding ID to an object's image tokens and its textual references, enabling in-context association.

Paperid: 6286, https://arxiv.org/pdf/2505.21106.pdf

Abstract:
Large Vision Language Models (LVLMs) have achieved remarkable progress in multimodal tasks, yet they also exhibit notable social biases. These biases often manifest as unintended associations between neutral concepts and sensitive human attributes, leading to disparate model behaviors across demographic groups. While existing studies primarily focus on detecting and quantifying such biases, they offer limited insight into the underlying mechanisms within the models. To address this gap, we propose an explanatory framework that combines information flow analysis with multi-round dialogue evaluation, aiming to understand the origin of social bias from the perspective of imbalanced internal information utilization. Specifically, we first identify high-contribution image tokens involved in the model's reasoning process for neutral questions via information flow analysis. Then, we design a multi-turn dialogue mechanism to evaluate the extent to which these key tokens encode sensitive information. Extensive experiments reveal that LVLMs exhibit systematic disparities in information usage when processing images of different demographic groups, suggesting that social bias is deeply rooted in the model's internal reasoning dynamics. Furthermore, we complement our findings from a textual modality perspective, showing that the model's semantic representations already display biased proximity patterns, thereby offering a cross-modal explanation of bias formation.

Paperid: 6287, https://arxiv.org/pdf/2505.20672.pdf

Abstract:
The Abstraction and Reasoning Corpus (ARC) poses a stringent test of general AI capabilities, requiring solvers to infer abstract patterns from only a handful of examples. Despite substantial progress in deep learning, state-of-the-art models still achieve accuracy rates of merely 40-55% on 2024 ARC Competition, indicative of a significant gap between their performance and human-level reasoning. In this work, we seek to bridge that gap by introducing an analogy-inspired ARC dataset, GIFARC. Leveraging large language models (LLMs) and vision-language models (VLMs), we synthesize new ARC-style tasks from a variety of GIF images that include analogies. Each new task is paired with ground-truth analogy, providing an explicit mapping between visual transformations and everyday concepts. By embedding robust human-intuitive analogies into ARC-style tasks, GIFARC guides AI agents to evaluate the task analogically before engaging in brute-force pattern search, thus efficiently reducing problem complexity and build a more concise and human-understandable solution. We empirically validate that guiding LLM with analogic approach with GIFARC affects task-solving approaches of LLMs to align with analogic approach of human.

Paperid: 6288, https://arxiv.org/pdf/2505.19840.pdf

Abstract:
Deep Neural Networks (DNNs) have achieved widespread success yet remain prone to adversarial attacks. Typically, such attacks either involve frequent queries to the target model or rely on surrogate models closely mirroring the target model -- often trained with subsets of the target model's training data -- to achieve high attack success rates through transferability. However, in realistic scenarios where training data is inaccessible and excessive queries can raise alarms, crafting adversarial examples becomes more challenging. In this paper, we present UnivIntruder, a novel attack framework that relies solely on a single, publicly available CLIP model and publicly available datasets. By using textual concepts, UnivIntruder generates universal, transferable, and targeted adversarial perturbations that mislead DNNs into misclassifying inputs into adversary-specified classes defined by textual concepts. Our extensive experiments show that our approach achieves an Attack Success Rate (ASR) of up to 85% on ImageNet and over 99% on CIFAR-10, significantly outperforming existing transfer-based methods. Additionally, we reveal real-world vulnerabilities, showing that even without querying target models, UnivIntruder compromises image search engines like Google and Baidu with ASR rates up to 84%, and vision language models like GPT-4 and Claude-3.5 with ASR rates up to 80%. These findings underscore the practicality of our attack in scenarios where traditional avenues are blocked, highlighting the need to reevaluate security paradigms in AI applications.

Paperid: 6289, https://arxiv.org/pdf/2505.19498.pdf

Abstract:
Large Vision-Language Models (LVLMs) usually generate texts which satisfy context coherence but don't match the visual input. Such a hallucination issue hinders LVLMs' applicability in the real world. The key to solving hallucination in LVLM is to make the text generation rely more on the visual content. Most previous works choose to enhance/adjust the features/output of a specific modality (i.e., visual or textual) to alleviate hallucinations in LVLM, which do not explicitly or systematically enhance the visual reliance. In this paper, we comprehensively investigate the factors which may degenerate the visual reliance in text generation of LVLM from a Bayesian perspective. Based on our observations, we propose to mitigate hallucination in LVLM from three aspects. Firstly, we observe that not all visual tokens are informative in generating meaningful texts. We propose to evaluate and remove redundant visual tokens to avoid their disturbance. Secondly, LVLM may encode inappropriate prior information, making it lean toward generating unexpected words. We propose a simple yet effective way to rectify the prior from a Bayesian perspective. Thirdly, we observe that starting from certain steps, the posterior of next-token prediction conditioned on visual tokens may collapse to a prior distribution which does not depend on any informative visual tokens at all. Thus, we propose to stop further text generation to avoid hallucination. Extensive experiments on three benchmarks including POPE, CHAIR, and MME demonstrate that our method can consistently mitigate the hallucination issue of LVLM and performs favorably against previous state-of-the-arts.

Paperid: 6290, https://arxiv.org/pdf/2505.19080.pdf

Abstract:
Vision-Language-Action (VLA) models have gained much attention from the research community thanks to their strength in translating multimodal observations with linguistic instructions into robotic actions. Despite their recent advancements, VLAs often overlook the explicit reasoning and only learn the functional input-action mappings, omitting these crucial logical steps for interpretability and generalization for complex, long-horizon manipulation tasks. In this work, we propose \textit{ReFineVLA}, a multimodal reasoning-aware framework that fine-tunes VLAs with teacher-guided reasons. We first augment robotic datasets with reasoning rationales generated by an expert teacher model, guiding VLA models to learn to reason about their actions. Then, we use \textit{ReFineVLA} to fine-tune pre-trained VLAs with the reasoning-enriched datasets, while maintaining their inherent generalization abilities and boosting reasoning capabilities. In addition, we conduct an attention map visualization to analyze the alignment among visual attention, linguistic prompts, and to-be-executed actions of \textit{ReFineVLA}, showcasing its ability to focus on relevant tasks and actions. Through the latter step, we explore that \textit{ReFineVLA}-trained models exhibit a meaningful attention shift towards relevant objects, highlighting the enhanced multimodal understanding and improved generalization. Evaluated across manipulation tasks, \textit{ReFineVLA} outperforms the state-of-the-art baselines. Specifically, it achieves an average increase of $5.0\%$ success rate on SimplerEnv WidowX Robot tasks, improves by an average of $8.6\%$ in variant aggregation settings, and by $1.7\%$ in visual matching settings for SimplerEnv Google Robot tasks. The source code will be publicly available.

Paperid: 6291, https://arxiv.org/pdf/2505.18434.pdf

Abstract:
Vision-language models (VLMs), such as CLIP, have demonstrated strong performance across a range of downstream tasks. However, CLIP is still limited in negation understanding: the ability to recognize the absence or exclusion of a concept. Existing methods address the problem by using a large language model (LLM) to generate large-scale data of image captions containing negation for further fine-tuning CLIP. However, these methods are both time- and compute-intensive, and their evaluations are typically restricted to image-text matching tasks. To expand the horizon, we (1) introduce a training-time negation data generation pipeline such that negation captions are generated during the training stage, which only increases 2.5% extra training time, and (2) we propose the first benchmark, Neg-TtoI, for evaluating text-to-image generation models on prompts containing negation, assessing model's ability to produce semantically accurate images. We show that our proposed method, TNG-CLIP, achieves SOTA performance on diverse negation benchmarks of image-to-text matching, text-to-image retrieval, and image generation.

Paperid: 6292, https://arxiv.org/pdf/2505.17625.pdf

Abstract:
With recent advancements in Large Language Models (LLMs) and growing interest in retrieval-augmented generation (RAG), the ability to understand table structures has become increasingly important. This is especially critical in financial domains such as securities reports, where highly accurate question answering (QA) over tables is required. However, tables exist in various formats-including HTML, images, and plain text-making it difficult to preserve and extract structural information. Therefore, multimodal LLMs are essential for robust and general-purpose table understanding. Despite their promise, current Large Vision-Language Models (LVLMs), which are major representatives of multimodal LLMs, still face challenges in accurately understanding characters and their spatial relationships within documents. In this study, we propose a method to enhance LVLM-based table understanding by incorporating in-table textual content and layout features. Experimental results demonstrate that these auxiliary modalities significantly improve performance, enabling robust interpretation of complex document layouts without relying on explicitly structured input formats.

Paperid: 6293, https://arxiv.org/pdf/2505.17529.pdf

Abstract:
Recent advancements in Large Vision-Language Models (LVLMs) have significantly expanded their utility in tasks like image captioning and visual question answering. However, they still struggle with object hallucination, where models generate descriptions that inaccurately reflect the visual content by including nonexistent objects or misrepresenting existing ones. While previous methods, such as data augmentation and training-free approaches, strive to tackle this issue, they still encounter scalability challenges and often depend on additional external modules. In this work, we propose Ensemble Decoding (ED), a novel strategy that splits the input image into sub-images and combines logit distributions by assigning weights through the attention map. Furthermore, we introduce ED adaptive plausibility constraint to calibrate logit distribution and FastED, a variant designed for speed-critical applications. Extensive experiments across hallucination benchmarks demonstrate that our proposed method achieves state-of-the-art performance, validating the effectiveness of our approach.

Paperid: 6294, https://arxiv.org/pdf/2505.15192.pdf

Abstract:
Foundation models have ushered in a new era for multimodal video understanding by enabling the extraction of rich spatiotemporal and semantic representations. In this work, we introduce a novel graph-based framework that integrates a vision-language foundation, leveraging VideoMAE for dynamic visual encoding and BERT for contextual textual embedding, to address the challenge of recognizing fine-grained bimanual manipulation actions. Departing from conventional static graph architectures, our approach constructs an adaptive multimodal graph where nodes represent frames, objects, and textual annotations, and edges encode spatial, temporal, and semantic relationships. These graph structures evolve dynamically based on learned interactions, allowing for flexible and context-aware reasoning. A task-specific attention mechanism within a Graph Attention Network further enhances this reasoning by modulating edge importance based on action semantics. Through extensive evaluations on diverse benchmark datasets, we demonstrate that our method consistently outperforms state-of-the-art baselines, underscoring the strength of combining foundation models with dynamic graph-based reasoning for robust and generalizable action recognition.

Paperid: 6295, https://arxiv.org/pdf/2505.11535.pdf

Abstract:
Lane Keeping Assist systems, while increasingly prevalent, often suffer from unpredictable real-world failures, largely due to their opaque, black-box nature, which limits driver anticipation and trust. To bridge the gap between automated assistance and effective human oversight, we present LKAlert, a novel supervisory alert system that leverages VLM to forecast potential LKA risk 1-3 seconds in advance. LKAlert processes dash-cam video and CAN data, integrating surrogate lane segmentation features from a parallel interpretable model as automated guiding attention. Unlike traditional binary classifiers, LKAlert issues both predictive alert and concise natural language explanation, enhancing driver situational awareness and trust. To support the development and evaluation of such systems, we introduce OpenLKA-Alert, the first benchmark dataset designed for predictive and explainable LKA failure warnings. It contains synchronized multimodal inputs and human-authored justifications across annotated temporal windows. We further contribute a generalizable methodological framework for VLM-based black-box behavior prediction, combining surrogate feature guidance with LoRA. This framework enables VLM to reason over structured visual context without altering its vision backbone, making it broadly applicable to other complex, opaque systems requiring interpretable oversight. Empirical results correctly predicts upcoming LKA failures with 69.8% accuracy and a 58.6\% F1-score. The system also generates high-quality textual explanations for drivers (71.7 ROUGE-L) and operates efficiently at approximately 2 Hz, confirming its suitability for real-time, in-vehicle use. Our findings establish LKAlert as a practical solution for enhancing the safety and usability of current ADAS and offer a scalable paradigm for applying VLMs to human-centered supervision of black-box automation.

Paperid: 6296, https://arxiv.org/pdf/2505.09139.pdf

Abstract:
Vision-language models (VLMs) offer flexible object detection through natural language prompts but suffer from performance variability depending on prompt phrasing. In this paper, we introduce a method for automated prompt refinement using a novel metric called the Contrastive Class Alignment Score (CCAS), which ranks prompts based on their semantic alignment with a target object class while penalizing similarity to confounding classes. Our method generates diverse prompt candidates via a large language model and filters them through CCAS, computed using prompt embeddings from a sentence transformer. We evaluate our approach on challenging object categories, demonstrating that our automatic selection of high-precision prompts improves object detection accuracy without the need for additional model training or labeled data. This scalable and model-agnostic pipeline offers a principled alternative to manual prompt engineering for VLM-based detection systems.

Paperid: 6297, https://arxiv.org/pdf/2505.08818.pdf

Abstract:
The intricate and multifaceted nature of vision language model (VLM) development, adaptation, and application necessitates the establishment of clear and standardized reporting protocols, particularly within the high-stakes context of healthcare. Defining these reporting standards is inherently challenging due to the diverse nature of studies involving VLMs, which vary significantly from the development of all new VLMs or finetuning for domain alignment to off-the-shelf use of VLM for targeted diagnosis and prediction tasks. In this position paper, we argue that traditional machine learning reporting standards and evaluation guidelines must be restructured to accommodate multiphase VLM studies; it also has to be organized for intuitive understanding of developers while maintaining rigorous standards for reproducibility. To facilitate community adoption, we propose a categorization framework for VLM studies and outline corresponding reporting standards that comprehensively address performance evaluation, data reporting protocols, and recommendations for manuscript composition. These guidelines are organized according to the proposed categorization scheme. Lastly, we present a checklist that consolidates reporting standards, offering a standardized tool to ensure consistency and quality in the publication of VLM-related research.

Paperid: 6298, https://arxiv.org/pdf/2505.06328.pdf

Abstract:
A wide variety of agentic AI applications - ranging from cognitive assistants for dementia patients to robotics - demand a robust memory system grounded in reality. In this paper, we propose such a memory system consisting of three components. First, we combine Vision Language Models for image captioning and entity disambiguation with Large Language Models for consistent information extraction during perception. Second, the extracted information is represented in a memory consisting of a knowledge graph enhanced by vector embeddings to efficiently manage relational information. Third, we combine semantic search and graph query generation for question answering via Retrieval Augmented Generation. We illustrate the system's working and potential using a real-world example.

Paperid: 6299, https://arxiv.org/pdf/2505.05540.pdf

Abstract:
Vision-language-action (VLA) models represent an important step toward general-purpose robotic systems by integrating visual perception, language understanding, and action execution. However, systematic evaluation of these models, particularly their zero-shot generalization capabilities in procedurally out-of-distribution (OOD) environments, remains limited. In this paper, we introduce MultiNet v0.2, a comprehensive benchmark designed to evaluate and analyze the generalization performance of state-of-the-art VLMs and VLAs - including GPT-4o, GPT-4.1, OpenVLA, Pi0 Base, and Pi0 FAST - on diverse procedural tasks from the Procgen benchmark. Our analysis reveals several critical insights: (1) all evaluated models exhibit significant limitations in zero-shot generalization to OOD tasks, with performance heavily influenced by factors such as action representation and task complexity; (2) VLAs generally outperforms other models due to their robust architectural design; and (3) VLM variants demonstrate substantial improvements when constrained appropriately, highlighting the sensitivity of model performance to precise prompt engineering. We release our benchmark, evaluation framework, and findings to enable the assessment of future VLA models and identify critical areas for improvement in their application to out-of-distribution digital tasks.

Paperid: 6300, https://arxiv.org/pdf/2505.04911.pdf

Abstract:
This study introduces SpatialPrompting, a novel framework that harnesses the emergent reasoning capabilities of off-the-shelf multimodal large language models to achieve zero-shot spatial reasoning in three-dimensional (3D) environments. Unlike existing methods that rely on expensive 3D-specific fine-tuning with specialized 3D inputs such as point clouds or voxel-based features, SpatialPrompting employs a keyframe-driven prompt generation strategy. This framework uses metrics such as vision-language similarity, Mahalanobis distance, field of view, and image sharpness to select a diverse and informative set of keyframes from image sequences and then integrates them with corresponding camera pose data to effectively abstract spatial relationships and infer complex 3D structures. The proposed framework not only establishes a new paradigm for flexible spatial reasoning that utilizes intuitive visual and positional cues but also achieves state-of-the-art zero-shot performance on benchmark datasets, such as ScanQA and SQA3D, across several metrics. The proposed method effectively eliminates the need for specialized 3D inputs and fine-tuning, offering a simpler and more scalable alternative to conventional approaches.

Paperid: 6301, https://arxiv.org/pdf/2504.21831.pdf

Abstract:
We introduce DEEVISum (Distilled Early Exit Vision language model for Summarization), a lightweight, efficient, and scalable vision language model designed for segment wise video summarization. Leveraging multi modal prompts that combine textual and audio derived signals, DEEVISum incorporates Multi Stage Knowledge Distillation (MSKD) and Early Exit (EE) to strike a balance between performance and efficiency. MSKD offers a 1.33% absolute F1 improvement over baseline distillation (0.5%), while EE reduces inference time by approximately 21% with a 1.3 point drop in F1. Evaluated on the TVSum dataset, our best model PaLI Gemma2 3B + MSKD achieves an F1 score of 61.1, competing the performance of significantly larger models, all while maintaining a lower computational footprint. We publicly release our code and processed dataset to support further research.

Paperid: 6302, https://arxiv.org/pdf/2504.20976.pdf

Abstract:
Navigating unfamiliar places continues to be one of the most persistent and essential everyday obstacles for those who are blind or have limited vision (BLV). Existing assistive technologies, such as GPS-based navigation systems, AI-powered smart glasses, and sonar-equipped canes, often face limitations in real-time obstacle avoidance, precise localization, and adaptability to dynamic surroundings. To investigate potential solutions, we introduced PathFinder, a novel map-less navigation system that explores different models for understanding 2D images, including Vision Language Models (VLMs), Large Language Models (LLMs), and employs monocular depth estimation for free-path detection. Our approach integrates a Depth-First Search (DFS) algorithm on depth images to determine the longest obstacle-free path, ensuring optimal route selection while maintaining computational efficiency. We conducted comparative evaluations against existing AI-powered navigation methods and performed a usability study with BLV participants. The results demonstrate that PathFinder achieves a favorable balance between accuracy, computational efficiency, and real-time responsiveness. Notably, it reduces mean absolute error (MAE) and improves decision-making speed in outdoor navigation compared to AI-based alternatives. Participant feedback emphasizes the system's usability and effectiveness in outside situations, but also identifies issues in complicated indoor locations and low-light conditions. Usability testing revealed that 73% of participants understood how to use the app in about a minute, and 80% praised its balance of accuracy, quick response, and overall convenience.

Paperid: 6303, https://arxiv.org/pdf/2504.17826.pdf

Abstract:
Fashion styling and personalized recommendations are pivotal in modern retail, contributing substantial economic value in the fashion industry. With the advent of vision-language models (VLM), new opportunities have emerged to enhance retailing through natural language and visual interactions. This work proposes FashionM3, a multimodal, multitask, and multiround fashion assistant, built upon a VLM fine-tuned for fashion-specific tasks. It helps users discover satisfying outfits by offering multiple capabilities including personalized recommendation, alternative suggestion, product image generation, and virtual try-on simulation. Fine-tuned on the novel FashionRec dataset, comprising 331,124 multimodal dialogue samples across basic, personalized, and alternative recommendation tasks, FashionM3 delivers contextually personalized suggestions with iterative refinement through multiround interactions. Quantitative and qualitative evaluations, alongside user studies, demonstrate FashionM3's superior performance in recommendation effectiveness and practical value as a fashion assistant.

Paperid: 6304, https://arxiv.org/pdf/2504.13123.pdf

Abstract:
In recent years, the field of vision-language model pre-training has experienced rapid advancements, driven primarily by the continuous enhancement of textual capabilities in large language models. However, existing training paradigms for multimodal large language models heavily rely on high-quality image-text pairs. As models and data scales grow exponentially, the availability of such meticulously curated data has become increasingly scarce and saturated, thereby severely limiting further advancements in this domain. This study investigates scalable caption generation techniques for vision-language model pre-training and demonstrates that large-scale low-hallucination synthetic captions can serve dual purposes: 1) acting as a viable alternative to real-world data for pre-training paradigms and 2) achieving superior performance enhancement when integrated into vision-language models through empirical validation. This paper presents following key contributions: 1) a novel pipeline for generating high-quality, low-hallucination, and knowledge-rich synthetic captions. Our continuous DPO methodology yields remarkable results in reducing hallucinations. Specifically, the non-hallucination caption rate on a held-out test set increases from 48.3% to 77.9% for a 7B-size model. 2) Comprehensive empirical validation reveals that our synthetic captions confer superior pre-training advantages over their counterparts. Across 15 vision language tasks, the model trained with our data achieves a significant performance gain of at least 6.2% compared to identical images with alt-text. In 20 common cognitive domains, the model trained with our data outperforms the alt-text data by at least 7.5%. Meanwhile, it also offers considerable support in the text-to-image domain. With our dataset, the FID score is reduced by 17.1 on a real-world validation benchmark and 13.3 on the MSCOCO validation benchmark.

Paperid: 6305, https://arxiv.org/pdf/2504.11838.pdf

Abstract:
Despite the rapid evolution of learning and computer vision algorithms, Fine-Grained Classification (FGC) still poses an open problem in many practically relevant applications. In the retail domain, for example, the identification of fast changing and visually highly similar products and their properties are key to automated price-monitoring and product recommendation. This paper presents a novel Visual RAG pipeline that combines the Retrieval Augmented Generation (RAG) approach and Vision Language Models (VLMs) for few-shot FGC. This Visual RAG pipeline extracts product and promotion data in advertisement leaflets from various retailers and simultaneously predicts fine-grained product ids along with price and discount information. Compared to previous approaches, the key characteristic of the Visual RAG pipeline is that it allows the prediction of novel products without re-training, simply by adding a few class samples to the RAG database. Comparing several VLM back-ends like GPT-4o [23], GPT-4o-mini [24], and Gemini 2.0 Flash [10], our approach achieves 86.8% accuracy on a diverse dataset.

Paperid: 6306, https://arxiv.org/pdf/2504.08583.pdf

Abstract:
We present AstroLLaVA, a vision language model for astronomy that enables interaction with astronomical imagery through natural dialogue. By fine-tuning the LLaVA model on a diverse dataset of $\sim$30k images with captions and question-answer pairs sourced from NASA's `Astronomy Picture of the Day', the European Southern Observatory, and the NASA/ESA Hubble Space Telescope, we create a model capable of answering open-ended questions about astronomical concepts depicted visually. Our two-stage fine-tuning process adapts the model to both image captioning and visual question answering in the astronomy domain. We demonstrate AstroLLaVA's performance on an astronomical visual question answering benchmark and release the model weights, code, and training set to encourage further open source work in this space. Finally, we suggest a roadmap towards general astronomical data alignment with pre-trained language models, and provide an open space for collaboration towards this end for interested researchers.

Paperid: 6307, https://arxiv.org/pdf/2504.07416.pdf

Abstract:
Recent advancements in multi-modal models have significantly improved vision-language (VL) alignment in radiology. However, existing approaches struggle to effectively utilize complex radiology reports for learning and offer limited interpretability through attention probability visualizations. To address these challenges, we introduce RadZero, a novel framework for VL alignment in radiology with zero-shot multi-task capability. A key component of our approach is VL-CABS (Vision-Language Cross-Attention Based on Similarity), which aligns text embeddings with local image features for interpretable, fine-grained VL reasoning. RadZero leverages large language models to extract concise semantic sentences from radiology reports and employs multi-positive contrastive training to effectively capture relationships between images and multiple relevant textual descriptions. It uses a pre-trained vision encoder with additional trainable Transformer layers, allowing efficient high-resolution image processing. By computing similarity between text embeddings and local image patch features, VL-CABS enables zero-shot inference with similarity probability for classification, and pixel-level VL similarity maps for grounding and segmentation. Experimental results on public chest radiograph benchmarks show that RadZero outperforms state-of-the-art methods in zero-shot classification, grounding, and segmentation. Furthermore, VL similarity map analysis highlights the potential of VL-CABS for improving explainability in VL alignment. Additionally, qualitative evaluation demonstrates RadZero's capability for open-vocabulary semantic segmentation, further validating its effectiveness in medical imaging.

Paperid: 6308, https://arxiv.org/pdf/2504.05925.pdf

Abstract:
Vision-language temporal alignment is a crucial capability for human dynamic recognition and cognition in real-world scenarios. While existing research focuses on capturing vision-language relevance, it faces limitations due to biased temporal distributions, imprecise annotations, and insufficient compositionally. To achieve fair evaluation and comprehensive exploration, our objective is to investigate and evaluate the ability of models to achieve alignment from a temporal perspective, specifically focusing on their capacity to synchronize visual scenarios with linguistic context in a temporally coherent manner. As a preliminary step, we present the statistical analysis of existing benchmarks and reveal the existing challenges from a decomposed perspective. To this end, we introduce SVLTA, the Synthetic Vision-Language Temporal Alignment derived via a well-designed and feasible control generation method within a simulation environment. The approach considers commonsense knowledge, manipulable action, and constrained filtering, which generates reasonable, diverse, and balanced data distributions for diagnostic evaluations. Our experiments reveal diagnostic insights through the evaluations in temporal question answering, distributional shift sensitiveness, and temporal alignment adaptation.

Paperid: 6309, https://arxiv.org/pdf/2504.01666.pdf

Abstract:
Continuous sign language recognition (CSLR) focuses on interpreting and transcribing sequences of sign language gestures in videos. In this work, we propose CLIP sign language adaptation (CLIP-SLA), a novel CSLR framework that leverages the powerful pre-trained visual encoder from the CLIP model to sign language tasks through parameter-efficient fine-tuning (PEFT). We introduce two variants, SLA-Adapter and SLA-LoRA, which integrate PEFT modules into the CLIP visual encoder, enabling fine-tuning with minimal trainable parameters. The effectiveness of the proposed frameworks is validated on four datasets: Phoenix2014, Phoenix2014-T, CSL-Daily, and Isharah-500, where both CLIP-SLA variants outperformed several SOTA models with fewer trainable parameters. Extensive ablation studies emphasize the effectiveness and flexibility of the proposed methods with different vision-language models for CSLR. These findings showcase the potential of adapting large-scale pre-trained models for scalable and efficient CSLR, which pave the way for future advancements in sign language understanding.

Paperid: 6310, https://arxiv.org/pdf/2504.00938.pdf

Abstract:
The subjective evaluation of early stage engineering designs, such as conceptual sketches, traditionally relies on human experts. However, expert evaluations are time-consuming, expensive, and sometimes inconsistent. Recent advances in vision-language models (VLMs) offer the potential to automate design assessments, but it is crucial to ensure that these AI ``judges'' perform on par with human experts. However, no existing framework assesses expert equivalence. This paper introduces a rigorous statistical framework to determine whether an AI judge's ratings match those of human experts. We apply this framework in a case study evaluating four VLM-based judges on key design metrics (uniqueness, creativity, usefulness, and drawing quality). These AI judges employ various in-context learning (ICL) techniques, including uni- vs. multimodal prompts and inference-time reasoning. The same statistical framework is used to assess three trained novices for expert-equivalence. Results show that the top-performing AI judge, using text- and image-based ICL with reasoning, achieves expert-level agreement for uniqueness and drawing quality and outperforms or matches trained novices across all metrics. In 6/6 runs for both uniqueness and creativity, and 5/6 runs for both drawing quality and usefulness, its agreement with experts meets or exceeds that of the majority of trained novices. These findings suggest that reasoning-supported VLM models can achieve human-expert equivalence in design evaluation. This has implications for scaling design evaluation in education and practice, and provides a general statistical framework for validating AI judges in other domains requiring subjective content evaluation.

Paperid: 6311, https://arxiv.org/pdf/2503.23344.pdf

Abstract:
Comics have long been a popular form of storytelling, offering visually engaging narratives that captivate audiences worldwide. However, the visual nature of comics presents a significant barrier for visually impaired readers, limiting their access to these engaging stories. In this work, we provide a pragmatic solution to this accessibility challenge by developing an automated system that generates text-based literary narratives from manga comics. Our approach aims to create an evocative and immersive prose that not only conveys the original narrative but also captures the depth and complexity of characters, their interactions, and the vivid settings in which they reside. To this end we make the following contributions: (1) We present a unified model, Magiv3, that excels at various functional tasks pertaining to comic understanding, such as localising panels, characters, texts, and speech-bubble tails, performing OCR, grounding characters etc. (2) We release human-annotated captions for over 3300 Japanese comic panels, along with character grounding annotations, and benchmark large vision-language models in their ability to understand comic images. (3) Finally, we demonstrate how integrating large vision-language models with Magiv3, can generate seamless literary narratives that allows visually impaired audiences to engage with the depth and richness of comic storytelling.

Paperid: 6312, https://arxiv.org/pdf/2503.20662.pdf

Abstract:
Lung cancer remains one of the leading causes of cancer-related mortality worldwide. A crucial challenge for early diagnosis is differentiating uncertain cases with similar visual characteristics and closely annotation scores. In clinical practice, radiologists rely on quantitative, hand-crafted Radiomic features extracted from Computed Tomography (CT) images, while recent research has primarily focused on deep learning solutions. More recently, Vision-Language Models (VLMs), particularly Contrastive Language-Image Pre-Training (CLIP)-based models, have gained attention for their ability to integrate textual knowledge into lung cancer diagnosis. While CLIP-Lung models have shown promising results, we identified the following potential limitations: (a) dependence on radiologists' annotated attributes, which are inherently subjective and error-prone, (b) use of textual information only during training, limiting direct applicability at inference, and (c) Convolutional-based vision encoder with randomly initialized weights, which disregards prior knowledge. To address these limitations, we introduce AutoRad-Lung, which couples an autoregressively pre-trained VLM, with prompts generated from hand-crafted Radiomics. AutoRad-Lung uses the vision encoder of the Large-Scale Autoregressive Image Model (AIMv2), pre-trained using a multi-modal autoregressive objective. Given that lung tumors are typically small, irregularly shaped, and visually similar to healthy tissue, AutoRad-Lung offers significant advantages over its CLIP-based counterparts by capturing pixel-level differences. Additionally, we introduce conditional context optimization, which dynamically generates context-specific prompts based on input Radiomics, improving cross-modal alignment.

Paperid: 6313, https://arxiv.org/pdf/2503.20362.pdf

Abstract:
Large Vision-Language Models (LVLMs) demonstrate remarkable performance in short-video tasks such as video question answering, but struggle in long-video understanding. The linear frame sampling strategy, conventionally used by LVLMs, fails to account for the non-linear distribution of key events in video data, often introducing redundant or irrelevant information in longer contexts while risking the omission of critical events in shorter ones. To address this, we propose SelfReS, a non-linear spatiotemporal self-reflective sampling method that dynamically selects key video fragments based on user prompts. Unlike prior approaches, SelfReS leverages the inherently sparse attention maps of LVLMs to define reflection tokens, enabling relevance-aware token selection without requiring additional training or external modules. Experiments demonstrate that SelfReS can be seamlessly integrated into strong base LVLMs, improving long-video task accuracy and achieving up to 46% faster inference speed within the same GPU memory budget.

Paperid: 6314, https://arxiv.org/pdf/2503.19361.pdf

Abstract:
We introduce ImageSet2Text, a novel approach that leverages vision-language foundation models to automatically create natural language descriptions of image sets. Inspired by concept bottleneck models (CBMs) and based on visual-question answering (VQA) chains, ImageSet2Text iteratively extracts key concepts from image subsets, encodes them into a structured graph, and refines insights using an external knowledge graph and CLIP-based validation. This iterative process enhances interpretability and enables accurate and detailed set-level summarization. Through extensive experiments, we evaluate ImageSet2Text's descriptions on accuracy, completeness, readability and overall quality, benchmarking it against existing vision-language models and introducing new datasets for large-scale group image captioning.

Paperid: 6315, https://arxiv.org/pdf/2503.15816.pdf

Abstract:
Multimodal Large Language Models (MLLMs) have achieved remarkable success in vision-language tasks but their remote sensing (RS) counterpart are relatively under explored. Unlike natural images, RS imagery presents unique challenges that current MLLMs struggle to handle, particularly in visual grounding and spatial reasoning. This study investigates the limitations of CLIP-based MLLMs in RS, highlighting their failure to differentiate visually distinct yet semantically similar RS images. To address this, we introduce a remote sensing multimodal visual patterns (RSMMVP) benchmark. It is designed to evaluate MLLMs in RS tasks by identifying the CLIP-blind pairs, where CLIP-based models incorrectly assign high similarity scores to visually distinct RS images. Through a visual question answering (VQA) evaluation, we analyze the performance of state-of-the-art MLLMs, revealing significant limitations in RS specific representation learning. The results provide valuable insights into the weaknesses of CLIP-based visual encoding and offer a foundation for future research to develop more effective MLLMs tailored for remote sensing applications.

Paperid: 6316, https://arxiv.org/pdf/2503.15035.pdf

Abstract:
Despite significant advancements in robotic manipulation, achieving consistent and stable grasping remains a fundamental challenge, often limiting the successful execution of complex tasks. Our analysis reveals that even state-of-the-art policy models frequently exhibit unstable grasping behaviors, leading to failure cases that create bottlenecks in real-world robotic applications. To address these challenges, we introduce GraspCorrect, a plug-and-play module designed to enhance grasp performance through vision-language model-guided feedback. GraspCorrect employs an iterative visual question-answering framework with two key components: grasp-guided prompting, which incorporates task-specific constraints, and object-aware sampling, which ensures the selection of physically feasible grasp candidates. By iteratively generating intermediate visual goals and translating them into joint-level actions, GraspCorrect significantly improves grasp stability and consistently enhances task success rates across existing policy models in the RLBench and CALVIN datasets.

Paperid: 6317, https://arxiv.org/pdf/2503.11342.pdf

Abstract:
Road rage, triggered by driving-related stimuli such as traffic congestion and aggressive driving, poses a significant threat to road safety. Previous research on road rage regulation has primarily focused on response suppression, lacking proactive prevention capabilities. With the advent of Vision-Language Models (VLMs), it has become possible to reason about trigger events visually and then engage in dialog-based comforting before drivers' anger escalates. To this end, we propose the road rage reasoning task, along with a finely annotated test dataset and evaluation metrics, to assess the capabilities of current mainstream VLMs in scene understanding, event recognition, and road rage reasoning. The results indicate that current VLMs exhibit significant shortcomings in scene understanding within the visual modality, as well as in comprehending the spatial relationships between objects in the textual modality. Improving VLMs' performance in these areas will greatly benefit downstream tasks like antecedent-focused road rage regulation.

Paperid: 6318, https://arxiv.org/pdf/2503.10546.pdf

Abstract:
With the rapid advancement of large language models (LLMs) and vision-language models (VLMs), significant progress has been made in developing open-vocabulary robotic manipulation systems. However, many existing approaches overlook the importance of object dynamics, limiting their applicability to more complex, dynamic tasks. In this work, we introduce KUDA, an open-vocabulary manipulation system that integrates dynamics learning and visual prompting through keypoints, leveraging both VLMs and learning-based neural dynamics models. Our key insight is that a keypoint-based target specification is simultaneously interpretable by VLMs and can be efficiently translated into cost functions for model-based planning. Given language instructions and visual observations, KUDA first assigns keypoints to the RGB image and queries the VLM to generate target specifications. These abstract keypoint-based representations are then converted into cost functions, which are optimized using a learned dynamics model to produce robotic trajectories. We evaluate KUDA on a range of manipulation tasks, including free-form language instructions across diverse object categories, multi-object interactions, and deformable or granular objects, demonstrating the effectiveness of our framework. The project page is available at http://kuda-dynamics.github.io.

Paperid: 6319, https://arxiv.org/pdf/2503.09594.pdf

Abstract:
Integrating large language models (LLMs) into autonomous driving has attracted significant attention with the hope of improving generalization and explainability. However, existing methods often focus on either driving or vision-language understanding but achieving both high driving performance and extensive language understanding remains challenging. In addition, the dominant approach to tackle vision-language understanding is using visual question answering. However, for autonomous driving, this is only useful if it is aligned with the action space. Otherwise, the model's answers could be inconsistent with its behavior. Therefore, we propose a model that can handle three different tasks: (1) closed-loop driving, (2) vision-language understanding, and (3) language-action alignment. Our model SimLingo is based on a vision language model (VLM) and works using only camera, excluding expensive sensors like LiDAR. SimLingo obtains state-of-the-art performance on the widely used CARLA simulator on the Bench2Drive benchmark and is the winning entry at the CARLA challenge 2024. Additionally, we achieve strong results in a wide variety of language-related tasks while maintaining high driving performance.

Paperid: 6320, https://arxiv.org/pdf/2503.09313.pdf

Abstract:
In the current literature, most embedding models are based on the encoder-only transformer architecture to extract a dense and meaningful representation of the given input, which can be a text, an image, and more. With the recent advances in language modeling thanks to the introduction of Large Language Models, the possibility of extracting embeddings from these large and extensively trained models has been explored. However, current studies focus on textual embeddings in English, which is also the main language on which these models have been trained. Furthermore, there are very few models that consider multimodal and multilingual input. In light of this, we propose an adaptation methodology for Large Vision-Language Models trained on English language data to improve their performance in extracting multilingual and multimodal embeddings. Finally, we design and introduce a benchmark to evaluate the effectiveness of multilingual and multimodal embedding models.

Paperid: 6321, https://arxiv.org/pdf/2503.06026.pdf

Abstract:
Achieving zero-shot peg insertion, where inserting an arbitrary peg into an unseen hole without task-specific training, remains a fundamental challenge in robotics. This task demands a highly generalizable perception system capable of detecting potential holes, selecting the correct mating hole from multiple candidates, estimating its precise pose, and executing insertion despite uncertainties. While learning-based methods have been applied to peg insertion, they often fail to generalize beyond the specific peg-hole pairs encountered during training. Recent advancements in Vision-Language Models (VLMs) offer a promising alternative, leveraging large-scale datasets to enable robust generalization across diverse tasks. Inspired by their success, we introduce a novel zero-shot peg insertion framework that utilizes a VLM to identify mating holes and estimate their poses without prior knowledge of their geometry. Extensive experiments demonstrate that our method achieves 90.2% accuracy, significantly outperforming baselines in identifying the correct mating hole across a wide range of previously unseen peg-hole pairs, including 3D-printed objects, toy puzzles, and industrial connectors. Furthermore, we validate the effectiveness of our approach in a real-world connector insertion task on a backpanel of a PC, where our system successfully detects holes, identifies the correct mating hole, estimates its pose, and completes the insertion with a success rate of 88.3%. These results highlight the potential of VLM-driven zero-shot reasoning for enabling robust and generalizable robotic assembly.

Paperid: 6322, https://arxiv.org/pdf/2502.16602.pdf

Abstract:
Recently, Large Vision-Language Models (LVLMs) have made significant strides across diverse multimodal tasks and benchmarks. This paper reveals a largely under-explored problem from existing video-involved LVLMs - language bias, where models tend to prioritize language over video and thus result in incorrect responses. To address this research gap, we first collect a Video Language Bias Evaluation Benchmark, which is specifically designed to assess the language bias in video-involved LVLMs through two key tasks: ambiguous video contrast and interrogative question probing. Accordingly, we design accompanied evaluation metrics that aim to penalize LVLMs being biased by language. In addition, we also propose Multi-branch Contrastive Decoding (MCD), introducing two expert branches to simultaneously counteract language bias potentially generated by the amateur text-only branch. Our experiments demonstrate that i) existing video-involved LVLMs, including both proprietary and open-sourced, are largely limited by the language bias problem; ii) our MCD can effectively mitigate this issue and maintain general-purpose capabilities in various video-involved LVLMs without any additional retraining or alteration to model architectures.

Paperid: 6323, https://arxiv.org/pdf/2502.15214.pdf

Abstract:
Reinforcement learning (RL) has shown impressive results in sequential decision-making tasks. Meanwhile, Large Language Models (LLMs) and Vision-Language Models (VLMs) have emerged, exhibiting impressive capabilities in multimodal understanding and reasoning. These advances have led to a surge of research integrating LLMs and VLMs into RL. In this survey, we review representative works in which LLMs and VLMs are used to overcome key challenges in RL, such as lack of prior knowledge, long-horizon planning, and reward design. We present a taxonomy that categorizes these LLM/VLM-assisted RL approaches into three roles: agent, planner, and reward. We conclude by exploring open problems, including grounding, bias mitigation, improved representations, and action advice. By consolidating existing research and identifying future directions, this survey establishes a framework for integrating LLMs and VLMs into RL, advancing approaches that unify natural language and visual understanding with sequential decision-making.

Paperid: 6324, https://arxiv.org/pdf/2502.06860.pdf

Abstract:
The ability to automatically complete a partial sketch that depicts a complex scene, e.g., "a woman chatting with a man in the park", is very useful. However, existing sketch generation methods create sketches from scratch; they do not complete a partial sketch in the style of the original. To address this challenge, we introduce AutoSketch, a styleaware vector sketch completion method that accommodates diverse sketch styles. Our key observation is that the style descriptions of a sketch in natural language preserve the style during automatic sketch completion. Thus, we use a pretrained vision-language model (VLM) to describe the styles of the partial sketches in natural language and replicate these styles using newly generated strokes. We initially optimize the strokes to match an input prompt augmented by style descriptions extracted from the VLM. Such descriptions allow the method to establish a diffusion prior in close alignment with that of the partial sketch. Next, we utilize the VLM to generate an executable style adjustment code that adjusts the strokes to conform to the desired style. We compare our method with existing methods across various sketch styles and prompts, performed extensive ablation studies and qualitative and quantitative evaluations, and demonstrate that AutoSketch can support various sketch scenarios.

Paperid: 6325, https://arxiv.org/pdf/2502.05660.pdf

Abstract:
Large Vision-Language Models (VLMs) have achieved unprecedented success in several objective multimodal reasoning tasks. However, to further enhance their capabilities of empathetic and effective communication with humans, improving how VLMs process and understand emotions is crucial. Despite significant research attention on improving affective understanding, there is a lack of detailed evaluations of VLMs for emotion-related tasks, which can potentially help inform downstream fine-tuning efforts. In this work, we present the first comprehensive evaluation of VLMs for recognizing evoked emotions from images. We create a benchmark for the task of evoked emotion recognition and study the performance of VLMs for this task, from perspectives of correctness and robustness. Through several experiments, we demonstrate important factors that emotion recognition performance depends on, and also characterize the various errors made by VLMs in the process. Finally, we pinpoint potential causes for errors through a human evaluation study. We use our experimental results to inform recommendations for the future of emotion research in the context of VLMs.

Paperid: 6326, https://arxiv.org/pdf/2501.18864.pdf

Abstract:
Test-time adaptation of pre-trained vision-language models has emerged as a technique for tackling distribution shifts during the test time. Although existing methods, especially those based on Test-time Prompt Tuning (TPT), have shown promising results, their high computational cost associated with parameter optimization presents challenges for scalability and practical application. This paper unveils the unnecessary nature of backpropagation in existing methods from a loss landscape perspective. Building on this insight, this paper proposes a simple yet effective framework called Test-time Loss Landscape Adaptation (TLLA). TLLA leverages the relative position between the training minimum and test loss landscapes to guide the adaptation process, avoiding the update of model parameters at test time. Specifically, it mainly consists of two main stages: In the prompt tuning stage, a Sharpness-Aware Prompt Tuning (SAPT) method is introduced to identify the training flat minimum, setting the foundation for the subsequent test-time adaptation; In the test stage, a Sharpness-based Test Sample Selection (STSS) approach is utilized to ensure the alignment of flat minima within the training loss landscape and each augmented test sample's loss landscape. Extensive experiments on both domain generalization and cross-dataset benchmarks demonstrate that TLLA achieves state-of-the-art performances while significantly reducing computational overhead. Notably, TLLA surpasses TPT by an average of 5.32\% and 6.98\% on four ImageNet variant datasets when employing ResNet50 and ViT-B/16 image encoders, respectively. The code will be available soon.

Paperid: 6327, https://arxiv.org/pdf/2501.15370.pdf

Abstract:
Large language models (LLMs) have demonstrated immense capabilities in understanding textual data and are increasingly being adopted to help researchers accelerate scientific discovery through knowledge extraction (information retrieval), knowledge distillation (summarizing key findings and methodologies into concise forms), and knowledge synthesis (aggregating information from multiple scientific sources to address complex queries, generate hypothesis and formulate experimental plans). However, scientific data often exists in both visual and textual modalities. Vision language models (VLMs) address this by incorporating a pretrained vision backbone for processing images and a cross-modal projector that adapts image tokens into the LLM dimensional space, thereby providing richer multimodal comprehension. Nevertheless, off-the-shelf VLMs show limited capabilities in handling domain-specific data and are prone to hallucinations. We developed intelligent assistants finetuned from LLaVA models to enhance multimodal understanding in low-dose radiation therapy (LDRT)-a benign approach used in the treatment of cancer-related illnesses. Using multilingual data from 42,673 articles, we devise complex reasoning and detailed description tasks for visual question answering (VQA) benchmarks. Our assistants, trained on 50,882 image-text pairs, demonstrate superior performance over base models as evaluated using LLM-as-a-judge approach, particularly in reducing hallucination and improving domain-specific comprehension.

Paperid: 6328, https://arxiv.org/pdf/2501.09355.pdf

Abstract:
Multimodal AI Agents are AI models that have the capability of interactively and cooperatively assisting human users to solve day-to-day tasks. Augmented Reality (AR) head worn devices can uniquely improve the user experience of solving procedural day-to-day tasks by providing egocentric multimodal (audio and video) observational capabilities to AI Agents. Such AR capabilities can help AI Agents see and listen to actions that users take which can relate to multimodal capabilities of human users. Existing AI Agents, either Large Language Models (LLMs) or Multimodal Vision-Language Models (VLMs) are reactive in nature, which means that models cannot take an action without reading or listening to the human user's prompts. Proactivity of AI Agents on the other hand can help the human user detect and correct any mistakes in agent observed tasks, encourage users when they do tasks correctly or simply engage in conversation with the user - akin to a human teaching or assisting a user. Our proposed YET to Intervene (YETI) multimodal agent focuses on the research question of identifying circumstances that may require the agent to intervene proactively. This allows the agent to understand when it can intervene in a conversation with human users that can help the user correct mistakes on tasks, like cooking, using AR. Our YETI Agent learns scene understanding signals based on interpretable notions of Structural Similarity (SSIM) on consecutive video frames. We also define the alignment signal which the AI Agent can learn to identify if the video frames corresponding to the user's actions on the task are consistent with expected actions. These signals are used by our AI Agent to determine when it should proactively intervene. We compare our results on the instances of proactive intervention in the HoloAssist multimodal benchmark for an expert agent guiding a user to complete procedural tasks.

Paperid: 6329, https://arxiv.org/pdf/2501.03304.pdf

Abstract:
One of the current trends in robotics is to employ large language models (LLMs) to provide non-predefined command execution and natural human-robot interaction. It is useful to have an environment map together with its language representation, which can be further utilized by LLMs. Such a comprehensive scene representation enables numerous ways of interaction with the map for autonomously operating robots. In this work, we present an approach that enhances incremental implicit mapping through the integration of vision-language features. Specifically, we (i) propose a decoder optimization technique for implicit language maps which can be used when new objects appear on the scene, and (ii) address the problem of inconsistent vision-language predictions between different viewing positions. Our experiments demonstrate the effectiveness of LiLMaps and solid improvements in performance.

Paperid: 6330, https://arxiv.org/pdf/2501.03287.pdf

Abstract:
The Lane Keeping Assist (LKA) system has become a standard feature in recent car models. While marketed as providing auto-steering capabilities, the system's operational characteristics and safety performance remain underexplored, primarily due to a lack of real-world testing and comprehensive data. To fill this gap, we extensively tested mainstream LKA systems from leading U.S. automakers in Tampa, Florida. Using an innovative method, we collected a comprehensive dataset that includes full Controller Area Network (CAN) messages with LKA attributes, as well as video, perception, and lateral trajectory data from a high-quality front-facing camera equipped with advanced vision detection and trajectory planning algorithms. Our tests spanned diverse, challenging conditions, including complex road geometry, adverse weather, degraded lane markings, and their combinations. A vision language model (VLM) further annotated the videos to capture weather, lighting, and traffic features. Based on this dataset, we present an empirical overview of LKA's operational features and safety performance. Key findings indicate: (i) LKA is vulnerable to faint markings and low pavement contrast; (ii) it struggles in lane transitions (merges, diverges, intersections), often causing unintended departures or disengagements; (iii) steering torque limitations lead to frequent deviations on sharp turns, posing safety risks; and (iv) LKA systems consistently maintain rigid lane-centering, lacking adaptability on tight curves or near large vehicles such as trucks. We conclude by demonstrating how this dataset can guide both infrastructure planning and self-driving technology. In view of LKA's limitations, we recommend improvements in road geometry and pavement maintenance. Additionally, we illustrate how the dataset supports the development of human-like LKA systems via VLM fine-tuning and Chain of Thought reasoning.

Paperid: 6331, https://arxiv.org/pdf/2501.01834.pdf

Abstract:
Image captioning is a critical task at the intersection of computer vision and natural language processing, with wide-ranging applications across various domains. For complex tasks such as diagnostic report generation, deep learning models require not only domain-specific image-caption datasets but also the incorporation of relevant general knowledge to provide contextual accuracy. Existing approaches exhibit inherent limitations: specialized models excel in capturing domain-specific details but lack generalization, while vision-language models (VLMs) built on large language models (LLMs) leverage general knowledge but struggle with domain-specific adaptation. To address these limitations, this paper proposes a novel agent-enhanced model collaboration framework, which we call MoColl, designed to effectively integrate domain-specific and general knowledge. Specifically, our approach is to decompose complex image captioning tasks into a series of interconnected question-answer subtasks. A trainable visual question answering (VQA) model is employed as a specialized tool to focus on domain-specific visual analysis, answering task-specific questions based on image content. Concurrently, an LLM-based agent with general knowledge formulates these questions and synthesizes the resulting question-answer pairs into coherent captions. Beyond its role in leveraging the VQA model, the agent further guides its training to enhance its domain-specific capabilities. Experimental results on radiology report generation validate the effectiveness of the proposed framework, demonstrating significant improvements in the quality of generated reports.

Paperid: 6332, https://arxiv.org/pdf/2412.20088.pdf

Abstract:
Archaeological catalogs, containing key elements such as artifact images, morphological descriptions, and excavation information, are essential for studying artifact evolution and cultural inheritance. These data are widely scattered across publications, requiring automated collection methods. However, existing Large Vision-Language Models (VLMs) and their derivative data collection methods face challenges in accurate image detection and modal matching when processing archaeological catalogs, making automated collection difficult. To address these issues, we propose a novel archaeological catalog collection method based on Large Vision-Language Models that follows an approach comprising three modules: document localization, block comprehension and block matching. Through practical data collection from the Dabagou and Miaozigou pottery catalogs and comparison experiments, we demonstrate the effectiveness of our approach, providing a reliable solution for automated collection of archaeological catalogs.

Paperid: 6333, https://arxiv.org/pdf/2412.19997.pdf

Abstract:
Large-scale Vision-Language Pre-training (VLP) has demonstrated remarkable success in the general domain. However, in the fashion domain, items are distinguished by fine-grained attributes like texture and material, which are crucial for tasks such as retrieval. Existing models often fail to leverage these fine-grained attributes from both text and image modalities. To address the above issues, we propose a novel approach for the fashion domain, Fine-grained Attributes Enhanced VLP (FashionFAE), which focuses on the detailed characteristics of fashion data. An attribute-emphasized text prediction task is proposed to predict fine-grained attributes of the items. This forces the model to focus on the salient attributes from the text modality. Additionally, a novel attribute-promoted image reconstruction task is proposed, which further enhances the fine-grained ability of the model by leveraging the representative attributes from the image modality. Extensive experiments show that FashionFAE significantly outperforms State-Of-The-Art (SOTA) methods, achieving 2.9% and 5.2% improvements in retrieval on sub-test and full test sets, respectively, and a 1.6% average improvement in recognition tasks.

Paperid: 6334, https://arxiv.org/pdf/2412.17921.pdf

Abstract:
Although LLMs have demonstrated remarkable capabilities in processing and generating textual data, their pre-trained vocabularies are ill-suited for capturing the nuanced temporal dynamics and patterns inherent in time series. The discrete, symbolic nature of natural language tokens, which these vocabularies are designed to represent, does not align well with the continuous, numerical nature of time series data. To address this fundamental limitation, we propose VITRO. Our method adapts textual inversion optimization from the vision-language domain in order to learn a new time series per-dataset vocabulary that bridges the gap between the discrete, semantic nature of natural language and the continuous, numerical nature of time series data. We show that learnable time series-specific pseudo-word embeddings represent time series data better than existing general language model vocabularies, with VITRO-enhanced methods achieving state-of-the-art performance in long-term forecasting across most datasets.

Paperid: 6335, https://arxiv.org/pdf/2412.16148.pdf

Abstract:
Vision Language Models (VLMs) can be trained more efficiently if training sets can be reduced in size. Recent work has shown the benefits of masking text during VLM training using a variety of approaches: truncation, random masking, block masking and syntax masking. In this paper, we show that the best masking strategy changes over training epochs and that, given sufficient training epochs. We analyze existing text masking approaches including syntax masking, which is currently the state of the art, and identify the word frequency distribution as important in determining their success. Experiments on a large range of data sets demonstrate that syntax masking is outperformed by other approaches, given sufficient epochs, and that our proposed frequency-based approach, called Contrastive Language-Image Pre-training with Word Frequency Masking (CLIPF) has numerous advantages. The benefits are particularly evident as the number of input tokens decreases.

Paperid: 6336, https://arxiv.org/pdf/2412.15739.pdf

Abstract:
Large Vision-Language Models (LVLMs) have made remarkable developments along with the recent surge of large language models. Despite their advancements, LVLMs have a tendency to generate plausible yet inaccurate or inconsistent information based on the provided source content. This phenomenon, also known as ``hallucinations" can have serious downstream implications during the deployment of LVLMs. To address this, we present VORD a simple and effective method that alleviates hallucinations by calibrating token predictions based on ordinal relationships between modified image pairs. VORD is presented in two forms: 1.) a minimalist training-free variant which eliminates implausible tokens from modified image pairs, and 2.) a trainable objective function that penalizes unlikely tokens. Our experiments demonstrate that VORD delivers better calibration and effectively mitigates object hallucinations on a wide-range of LVLM benchmarks.

Paperid: 6337, https://arxiv.org/pdf/2412.08158.pdf

Abstract:
The exploration of various vision-language tasks, such as visual captioning, visual question answering, and visual commonsense reasoning, is an important area in artificial intelligence and continuously attracts the research community's attention. Despite the improvements in overall performance, classic challenges still exist in vision-language tasks and hinder the development of this area. In recent years, the rise of pre-trained models is driving the research on vision-language tasks. Thanks to the massive scale of training data and model parameters, pre-trained models have exhibited excellent performance in numerous downstream tasks. Inspired by the powerful capabilities of pre-trained models, new paradigms have emerged to solve the classic challenges. Such methods have become mainstream in current research with increasing attention and rapid advances. In this paper, we present a comprehensive overview of how vision-language tasks benefit from pre-trained models. First, we review several main challenges in vision-language tasks and discuss the limitations of previous solutions before the era of pre-training. Next, we summarize the recent advances in incorporating pre-trained models to address the challenges in vision-language tasks. Finally, we analyze the potential risks associated with the inherent limitations of pre-trained models and discuss possible solutions, attempting to provide future research directions.

Paperid: 6338, https://arxiv.org/pdf/2412.06184.pdf

Abstract:
We study the perception of color illusions by vision-language models. Color illusion, where a person's visual system perceives color differently from actual color, is well-studied in human vision. However, it remains underexplored whether vision-language models (VLMs), trained on large-scale human data, exhibit similar perceptual biases when confronted with such color illusions. We propose an automated framework for generating color illusion images, resulting in RCID (Realistic Color Illusion Dataset), a dataset of 19,000 realistic illusion images. Our experiments show that all studied VLMs exhibit perceptual biases similar human vision. Finally, we train a model to distinguish both human perception and actual pixel differences.

Paperid: 6339, https://arxiv.org/pdf/2412.05155.pdf

Abstract:
This study evaluates the effectiveness of Vision Language Models (VLMs) in representing and utilizing multimodal content for fact-checking. To be more specific, we investigate whether incorporating multimodal content improves performance compared to text-only models and how well VLMs utilize text and image information to enhance misinformation detection. Furthermore we propose a probing classifier based solution using VLMs. Our approach extracts embeddings from the last hidden layer of selected VLMs and inputs them into a neural probing classifier for multi-class veracity classification. Through a series of experiments on two fact-checking datasets, we demonstrate that while multimodality can enhance performance, fusing separate embeddings from text and image encoders yielded superior results compared to using VLM embeddings. Furthermore, the proposed neural classifier significantly outperformed KNN and SVM baselines in leveraging extracted embeddings, highlighting its effectiveness for multimodal fact-checking.

Paperid: 6340, https://arxiv.org/pdf/2412.04083.pdf

Abstract:
Open-World Compositional Zero-Shot Learning (OW-CZSL) addresses the challenge of recognizing novel compositions of known primitives and entities. Even though prior works utilize language knowledge for recognition, such approaches exhibit limited interactions between language-image modalities. Our approach primarily focuses on enhancing the inter-modality interactions through fostering richer interactions between image and textual data. Additionally, we introduce a novel module aimed at alleviating the computational burden associated with exhaustive exploration of all possible compositions during the inference stage. While previous methods exclusively learn compositions jointly or independently, we introduce an advanced hybrid procedure that leverages both learning mechanisms to generate final predictions. Our proposed model, achieves state-of-the-art in OW-CZSL in three datasets, while surpassing Large Vision Language Models (LLVM) in two datasets.

Paperid: 6341, https://arxiv.org/pdf/2412.03927.pdf

Abstract:
In vision-language models (VLMs), the ability to perceive and interpret color and physical environment is crucial for achieving contextually accurate understanding and interaction. However, despite advances in multimodal modeling, there remains a significant lack of specialized datasets that rigorously evaluate a model's capacity to discern subtle color variations and spatial context -- critical elements for situational comprehension and reliable deployment across real-world applications. Toward that goal, we curate MegaCOIN, a high-quality, human-labeled dataset based on \emph{real} images with various contextual attributes. MegaCOIN consists of two parts: MegaCOIN-Instruct, which serves as a supervised fine-tuning (SFT) dataset for VLMs; and MegaCOIN-Bench, an annotated test set that can be used as a stand-alone QA dataset. MegaCOIN~provides three annotated features for 220,000 real images: foreground color, background color, and description of an object's physical environment, constituting 660k human annotations. In addition, MegaCOIN can be applied to benchmark domain generalization (DG) algorithms. We explore benchmarking DG methods in the linear probing setup for VLM and show some new insights. Last but not least, we show that VLMs, including GPT-4o, have subpar color recognition capabilities, and fine-tuning with MegaCOIN can result in improved performance on visual evaluation tasks. In certain cases, MegaCOIN fine-tuned small-scale opensource models such as LLaVA and Bunny can outperform closed-source GPT-4o. We hope the utilities of MegaCOIN can shed light on the directions VLMs can improve and provide a more complex platform for domain generalization algorithms.

Paperid: 6342, https://arxiv.org/pdf/2412.00890.pdf

Abstract:
Industrial anomaly detection (IAD) plays a crucial role in the maintenance and quality control of manufacturing processes. In this paper, we propose a novel approach, Vision-Language Anomaly Detection via Contrastive Cross-Modal Training (CLAD), which leverages large vision-language models (LVLMs) to improve both anomaly detection and localization in industrial settings. CLAD aligns visual and textual features into a shared embedding space using contrastive learning, ensuring that normal instances are grouped together while anomalies are pushed apart. Through extensive experiments on two benchmark industrial datasets, MVTec-AD and VisA, we demonstrate that CLAD outperforms state-of-the-art methods in both image-level anomaly detection and pixel-level anomaly localization. Additionally, we provide ablation studies and human evaluation to validate the importance of key components in our method. Our approach not only achieves superior performance but also enhances interpretability by accurately localizing anomalies, making it a promising solution for real-world industrial applications.

Paperid: 6343, https://arxiv.org/pdf/2411.18711.pdf

Abstract:
Despite their promise to perform complex reasoning, large language models (LLMs) have been shown to have limited effectiveness in end-to-end planning. This has inspired an intriguing question: if these models cannot plan well, can they still contribute to the planning framework as a helpful plan evaluator? In this work, we generalize this question to consider LLMs augmented with visual understanding, i.e., Vision-Language Models (VLMs). We introduce PathEval, a novel benchmark evaluating VLMs as plan evaluators in complex path-planning scenarios. Succeeding in the benchmark requires a VLM to be able to abstract traits of optimal paths from the scenario description, demonstrate precise low-level perception on each path, and integrate this information to decide the better path. Our analysis of state-of-the-art VLMs reveals that these models face significant challenges on the benchmark. We observe that the VLMs can precisely abstract given scenarios to identify the desired traits and exhibit mixed performance in integrating the provided information. Yet, their vision component presents a critical bottleneck, with models struggling to perceive low-level details about a path. Our experimental results show that this issue cannot be trivially addressed via end-to-end fine-tuning; rather, task-specific discriminative adaptation of these vision encoders is needed for these VLMs to become effective path evaluators.

Paperid: 6344, https://arxiv.org/pdf/2411.17491.pdf

Abstract:
Vision-Language Models (VLMs) have recently demonstrated remarkable capabilities in comprehending complex visual content. However, the mechanisms underlying how VLMs process visual information remain largely unexplored. In this paper, we conduct a thorough empirical analysis, focusing on attention modules across layers. We reveal several key insights about how these models process visual data: (i) the internal representation of the query tokens (e.g., representations of "describe the image"), is utilized by VLMs to store global image information; we demonstrate that these models generate surprisingly descriptive responses solely from these tokens, without direct access to image tokens. (ii) Cross-modal information flow is predominantly influenced by the middle layers (approximately 25% of all layers), while early and late layers contribute only marginally.(iii) Fine-grained visual attributes and object details are directly extracted from image tokens in a spatially localized manner, i.e., the generated tokens associated with a specific object or attribute attend strongly to their corresponding regions in the image. We propose novel quantitative evaluation to validate our observations, leveraging real-world complex visual scenes. Finally, we demonstrate the potential of our findings in facilitating efficient visual processing in state-of-the-art VLMs.

Paperid: 6345, https://arxiv.org/pdf/2411.10063.pdf

Abstract:
Federated domain generalization (FedDG) aims to improve the global model generalization in unseen domains by addressing data heterogeneity under privacy-preserving constraints. A common strategy in existing FedDG studies involves sharing domain-specific knowledge among clients, such as spectrum information, class prototypes, and data styles. However, this knowledge is extracted directly from local client samples, and sharing such sensitive information poses a potential risk of data leakage, which might not fully meet the requirements of FedDG. In this paper, we introduce prompt learning to adapt pre-trained vision-language models (VLMs) in the FedDG scenario, and leverage locally learned prompts as a more secure bridge to facilitate knowledge transfer among clients. Specifically, we propose a novel FedDG framework through Prompt Learning and AggregatioN (PLAN), which comprises two training stages to collaboratively generate local prompts and global prompts at each federated round. First, each client performs both text and visual prompt learning using their own data, with local prompts indirectly synchronized by regarding the global prompts as a common reference. Second, all domain-specific local prompts are exchanged among clients and selectively aggregated into the global prompts using lightweight attention-based aggregators. The global prompts are finally applied to adapt VLMs to unseen target domains. As our PLAN framework requires training only a limited number of prompts and lightweight aggregators, it offers notable advantages in computational and communication efficiency for FedDG. Extensive experiments demonstrate the superior generalization ability of PLAN across four benchmark datasets.

Paperid: 6346, https://arxiv.org/pdf/2411.05821.pdf

Abstract:
Vision-language-action (VLA) models represent a promising direction for developing general-purpose robotic systems, demonstrating the ability to combine visual understanding, language comprehension, and action generation. However, systematic evaluation of these models across diverse robotic tasks remains limited. In this work, we present a comprehensive evaluation framework and benchmark suite for assessing VLA models. We profile three state-of-the-art VLM and VLAs - GPT-4o, OpenVLA, and JAT - across 20 diverse datasets from the Open-X-Embodiment collection, evaluating their performance on various manipulation tasks. Our analysis reveals several key insights: 1. current VLA models show significant variation in performance across different tasks and robot platforms, with GPT-4o demonstrating the most consistent performance through sophisticated prompt engineering, 2. all models struggle with complex manipulation tasks requiring multi-step planning, and 3. model performance is notably sensitive to action space characteristics and environmental factors. We release our evaluation framework and findings to facilitate systematic assessment of future VLA models and identify critical areas for improvement in the development of general purpose robotic systems.

Paperid: 6347, https://arxiv.org/pdf/2410.21113.pdf

Abstract:
The growing demand for surveillance in public spaces presents significant challenges due to the shortage of human resources. Current AI-based video surveillance systems heavily rely on core computer vision models that require extensive finetuning, which is particularly difficult in surveillance settings due to limited datasets and difficult setting (viewpoint, low quality, etc.). In this work, we propose leveraging Large Vision-Language Models (LVLMs), known for their strong zero and few-shot generalization, to tackle video understanding tasks in surveillance. Specifically, we explore VideoLLaMA2, a state-of-the-art LVLM, and an improved token-level sampling method, Self-Reflective Sampling (Self-ReS). Our experiments on the UCF-Crime dataset show that VideoLLaMA2 represents a significant leap in zero-shot performance, with 20% boost over the baseline. Self-ReS additionally increases zero-shot action recognition performance to 44.6%. These results highlight the potential of LVLMs, paired with improved sampling techniques, for advancing surveillance video analysis in diverse scenarios.

Paperid: 6348, https://arxiv.org/pdf/2410.20717.pdf

Abstract:
Although multimodal large language models (MLLMs) have achieved promising results on a wide range of vision-language tasks, their ability to perceive and understand human faces is rarely explored. In this work, we comprehensively evaluate existing MLLMs on face perception tasks. The quantitative results reveal that existing MLLMs struggle to handle these tasks. The primary reason is the lack of image-text datasets that contain fine-grained descriptions of human faces. To tackle this problem, we design a practical pipeline for constructing datasets, upon which we further build a novel multimodal large face perception model, namely Face-MLLM. Specifically, we re-annotate LAION-Face dataset with more detailed face captions and facial attribute labels. Besides, we re-formulate traditional face datasets using the question-answer style, which is fit for MLLMs. Together with these enriched datasets, we develop a novel three-stage MLLM training method. In the first two stages, our model learns visual-text alignment and basic visual question answering capability, respectively. In the third stage, our model learns to handle multiple specialized face perception tasks. Experimental results show that our model surpasses previous MLLMs on five famous face perception tasks. Besides, on our newly introduced zero-shot facial attribute analysis task, our Face-MLLM also presents superior performance.

Paperid: 6349, https://arxiv.org/pdf/2410.12225.pdf

Abstract:
This paper evaluates the use of vision-language models (VLMs) for zero-shot detection and association of hardhats to enhance construction safety. Given the significant risk of head injuries in construction, proper enforcement of hardhat use is critical. We investigate the applicability of foundation models, specifically OWLv2, for detecting hardhats in real-world construction site images. Our contributions include the creation of a new benchmark dataset, Hardhat Safety Detection Dataset, by filtering and combining existing datasets and the development of a cascaded detection approach. Experimental results on 5,210 images demonstrate that the OWLv2 model achieves an average precision of 0.6493 for hardhat detection. We further analyze the limitations and potential improvements for real-world applications, highlighting the strengths and weaknesses of current foundation models in safety perception domains.

Paperid: 6350, https://arxiv.org/pdf/2410.10879.pdf

Abstract:
We propose Word-Frequency-based Image-Text Pair Pruning (WFPP), a novel data pruning method that improves the efficiency of VLMs. Unlike MetaCLIP, our method does not need metadata for pruning, but selects text-image pairs to prune based on the content of the text. Specifically, WFPP prunes text-image pairs containing high-frequency words across the entire training dataset. The effect of WFPP is to reduce the dominance of frequent words. The result a better balanced word-frequency distribution in the dataset, which is known to improve the training of word embedding models. After pre-training on the pruned subset, we fine-tuned the model on the entire dataset for one additional epoch to achieve better performance. Our experiments demonstrate that applying WFPP when training a CLIP model improves performance on a wide range of downstream tasks. WFPP also provides the advantage of speeding up pre-training by using fewer samples. Additionally, we analyze the training data before and after pruning to visualize how WFPP changes the balance of word frequencies. We hope our work encourages researchers to consider the distribution of words in the training data when pre-training VLMs, not limited to CLIP.

Paperid: 6351, https://arxiv.org/pdf/2410.09874.pdf

Abstract:
Visual navigation is an essential skill for home-assistance robots, providing the object-searching ability to accomplish long-horizon daily tasks. Many recent approaches use Large Language Models (LLMs) for commonsense inference to improve exploration efficiency. However, the planning process of LLMs is limited within texts and it is difficult to represent the spatial occupancy and geometry layout only by texts. Both are important for making rational navigation decisions. In this work, we seek to unleash the spatial perception and planning ability of Vision-Language Models (VLMs), and explore whether the VLM, with only on-board camera captured RGB/RGB-D stream inputs, can efficiently finish the visual navigation tasks in a mapless manner. We achieve this by developing the imagination-powered navigation framework ImagineNav, which imagines the future observation images at valuable robot views and translates the complex navigation planning process into a rather simple best-view image selection problem for VLM. To generate appropriate candidate robot views for imagination, we introduce the Where2Imagine module, which is distilled to align with human navigation habits. Finally, to reach the VLM preferred views, an off-the-shelf point-goal navigation policy is utilized. Empirical experiments on the challenging open-vocabulary object navigation benchmarks demonstrates the superiority of our proposed system.

Paperid: 6352, https://arxiv.org/pdf/2410.08211.pdf

Abstract:
Large-scale vision-language pre-trained (VLP) models (e.g., CLIP) are renowned for their versatility, as they can be applied to diverse applications in a zero-shot setup. However, when these models are used in specific domains, their performance often falls short due to domain gaps or the under-representation of these domains in the training data. While fine-tuning VLP models on custom datasets with human-annotated labels can address this issue, annotating even a small-scale dataset (e.g., 100k samples) can be an expensive endeavor, often requiring expert annotators if the task is complex. To address these challenges, we propose LatteCLIP, an unsupervised method for fine-tuning CLIP models on classification with known class names in custom domains, without relying on human annotations. Our method leverages Large Multimodal Models (LMMs) to generate expressive textual descriptions for both individual images and groups of images. These provide additional contextual information to guide the fine-tuning process in the custom domains. Since LMM-generated descriptions are prone to hallucination or missing details, we introduce a novel strategy to distill only the useful information and stabilize the training. Specifically, we learn rich per-class prototype representations from noisy generated texts and dual pseudo-labels. Our experiments on 10 domain-specific datasets show that LatteCLIP outperforms pre-trained zero-shot methods by an average improvement of +4.74 points in top-1 accuracy and other state-of-the-art unsupervised methods by +3.45 points.

Paperid: 6353, https://arxiv.org/pdf/2410.04749.pdf

Abstract:
Generating Natural Language Explanations (NLEs) for model predictions on medical images, particularly those depicting thoracic pathologies, remains a critical and challenging task. Existing methodologies often struggle due to general models' insufficient domain-specific medical knowledge and privacy concerns associated with retrieval-based augmentation techniques. To address these issues, we propose a novel Vision-Language framework augmented with a Knowledge Graph (KG)-based datastore, which enhances the model's understanding by incorporating additional domain-specific medical knowledge essential for generating accurate and informative NLEs. Our framework employs a KG-based retrieval mechanism that not only improves the precision of the generated explanations but also preserves data privacy by avoiding direct data retrieval. The KG datastore is designed as a plug-and-play module, allowing for seamless integration with various model architectures. We introduce and evaluate three distinct frameworks within this paradigm: KG-LLaVA, which integrates the pre-trained LLaVA model with KG-RAG; Med-XPT, a custom framework combining MedCLIP, a transformer-based projector, and GPT-2; and Bio-LLaVA, which adapts LLaVA by incorporating the Bio-ViT-L vision model. These frameworks are validated on the MIMIC-NLE dataset, where they achieve state-of-the-art results, underscoring the effectiveness of KG augmentation in generating high-quality NLEs for thoracic pathologies.

Paperid: 6354, https://arxiv.org/pdf/2410.03334.pdf

Abstract:
Radiological services are experiencing unprecedented demand, leading to increased interest in automating radiology report generation. Existing Vision-Language Models (VLMs) suffer from hallucinations, lack interpretability, and require expensive fine-tuning. We introduce SAE-Rad, which uses sparse autoencoders (SAEs) to decompose latent representations from a pre-trained vision transformer into human-interpretable features. Our hybrid architecture combines state-of-the-art SAE advancements, achieving accurate latent reconstructions while maintaining sparsity. Using an off-the-shelf language model, we distil ground-truth reports into radiological descriptions for each SAE feature, which we then compile into a full report for each image, eliminating the need for fine-tuning large models for this task. To the best of our knowledge, SAE-Rad represents the first instance of using mechanistic interpretability techniques explicitly for a downstream multi-modal reasoning task. On the MIMIC-CXR dataset, SAE-Rad achieves competitive radiology-specific metrics compared to state-of-the-art models while using significantly fewer computational resources for training. Qualitative analysis reveals that SAE-Rad learns meaningful visual concepts and generates reports aligning closely with expert interpretations. Our results suggest that SAEs can enhance multimodal reasoning in healthcare, providing a more interpretable alternative to existing VLMs.

Paperid: 6355, https://arxiv.org/pdf/2410.02787.pdf

Abstract:
Navigating towards fully open language goals and exploring open scenes in an intelligent way have always raised significant challenges. Recently, Vision Language Models (VLMs) have demonstrated remarkable capabilities to reason with both language and visual data. Although many works have focused on leveraging VLMs for navigation in open scenes, they often require high computational cost, rely on object-centric approaches, or depend on environmental priors in detailed human instructions. We introduce Navigation with VLM (NavVLM), a training-free framework that harnesses open-source VLMs to enable robots to navigate effectively, even for human-friendly language goal such as abstract places, actions, or specific objects in open scenes. NavVLM leverages the VLM as its cognitive core to perceive environmental information and constantly provides exploration guidance achieving intelligent navigation with only a neat target rather than a detailed instruction with environment prior. We evaluated and validated NavVLM in both simulation and real-world experiments. In simulation, our framework achieves state-of-the-art performance in Success weighted by Path Length (SPL) on object-specifc tasks in richly detailed environments from Matterport 3D (MP3D), Habitat Matterport 3D (HM3D) and Gibson. With navigation episode reported, NavVLM demonstrates the capabilities to navigate towards any open-set languages. In real-world validation, we validated our framework's effectiveness in real-world robot at indoor scene.

Paperid: 6356, https://arxiv.org/pdf/2409.17806.pdf

Abstract:
Continual learning (CL) enables models to adapt to evolving data streams without catastrophic forgetting, a fundamental requirement for real-world AI systems. However, the current methods often depend on large replay buffers or heavily annotated datasets which are impractical due to storage, privacy, and cost constraints. We propose CLTS (Continual Learning via Text-Image Synergy), a novel class-incremental framework that mitigates forgetting without storing real task data. CLTS leverages pre-trained vision-language models, BLIP (Bootstrapping Language-Image Pre-training) for caption generation and stable diffusion for sample generation. Each task is handled by a dedicated Task Head, while a Task Router learns to assign inputs to the correct Task Head using the generated data. On three benchmark datasets, CLTS improves average task accuracy by up to 54% and achieves 63 times better memory efficiency compared to four recent continual learning baselines, demonstrating improved retention and adaptability. CLTS introduces a novel perspective by integrating generative text-image augmentation for scalable continual learning.

Paperid: 6357, https://arxiv.org/pdf/2409.17330.pdf

Abstract:
Semantic segmentation networks have achieved significant success under the assumption of independent and identically distributed data. However, these networks often struggle to detect anomalies from unknown semantic classes due to the limited set of visual concepts they are typically trained on. To address this issue, anomaly segmentation often involves fine-tuning on outlier samples, necessitating additional efforts for data collection, labeling, and model retraining. Seeking to avoid this cumbersome work, we take a different approach and propose to incorporate Vision-Language (VL) encoders into existing anomaly detectors to leverage the semantically broad VL pre-training for improved outlier awareness. Additionally, we propose a new scoring function that enables data- and training-free outlier supervision via textual prompts. The resulting VL4AD model, which includes max-logit prompt ensembling and a class-merging strategy, achieves competitive performance on widely used benchmark datasets, thereby demonstrating the potential of vision-language models for pixel-wise anomaly detection.

Paperid: 6358, https://arxiv.org/pdf/2409.14580.pdf

Abstract:
Robots must operate safely when deployed in novel and human-centered environments, like homes. Current safe control approaches typically assume that the safety constraints are known a priori, and thus, the robot can pre-compute a corresponding safety controller. While this may make sense for some safety constraints (e.g., avoiding collision with walls by analyzing a floor plan), other constraints are more complex (e.g., spills), inherently personal, context-dependent, and can only be identified at deployment time when the robot is interacting in a specific environment and with a specific person (e.g., fragile objects, expensive rugs). Here, language provides a flexible mechanism to communicate these evolving safety constraints to the robot. In this work, we use vision language models (VLMs) to interpret language feedback and the robot's image observations to continuously update the robot's representation of safety constraints. With these inferred constraints, we update a Hamilton-Jacobi reachability safety controller online via efficient warm-starting techniques. Through simulation and hardware experiments, we demonstrate the robot's ability to infer and respect language-based safety constraints with the proposed approach.

Paperid: 6359, https://arxiv.org/pdf/2409.09845.pdf

Abstract:
Controlling Wheeled-legged robots is challenging especially on slippery surfaces due to their dependence on continuous ground contact. Unlike quadrupeds or bipeds, which can leverage multiple fixed contact points for recovery, wheeled-legged robots are highly susceptible to slip, where even momentary loss of traction can result in irrecoverable instability. Anticipating ground physical properties such as friction before contact would allow proactive control adjustments, reducing slip risk. In this paper, we propose a friction-aware safety locomotion framework that integrates Vision-Language Models (VLMs) with a Reinforcement Learning (RL) policy. Our method employs a Retrieval-Augmented Generation (RAG) approach to estimate the Coefficient of Friction (CoF), which is then explicitly incorporated into the RL policy. This enables the robot to adapt its speed based on predicted friction conditions before contact. The framework is validated through experiments in both simulation and on a physical customized Wheeled Inverted Pendulum (WIP). Experimental results show that our approach successfully completes trajectory tracking tasks on slippery surfaces, whereas baseline methods relying solely on proprioceptive feedback fail. These findings highlight the importance and effectiveness of explicitly predicting and utilizing ground friction information for safe locomotion. They also point to a promising research direction in exploring the use of VLMs for estimating ground conditions, which remains a significant challenge for purely vision-based methods.

Paperid: 6360, https://arxiv.org/pdf/2409.07048.pdf

Abstract:
The prominence of generalized foundation models in vision-language integration has witnessed a surge, given their multifarious applications. Within the natural domain, the procurement of vision-language datasets to construct these foundation models is facilitated by their abundant availability and the ease of web crawling. Conversely, in the remote sensing domain, although vision-language datasets exist, their volume is suboptimal for constructing robust foundation models. This study introduces an approach to curate vision-language datasets by employing an image decoding machine learning model, negating the need for human-annotated labels. Utilizing this methodology, we amassed approximately 9.6 million vision-language paired datasets in VHR imagery. The resultant model outperformed counterparts that did not leverage publicly available vision-language datasets, particularly in downstream tasks such as zero-shot classification, semantic localization, and image-text retrieval. Moreover, in tasks exclusively employing vision encoders, such as linear probing and k-NN classification, our model demonstrated superior efficacy compared to those relying on domain-specific vision-language datasets.

Paperid: 6361, https://arxiv.org/pdf/2409.05593.pdf

Abstract:
Embodied navigation requires robots to understand and interact with the environment based on given tasks. Vision-Language Navigation (VLN) is an embodied navigation task, where a robot navigates within a previously seen and unseen environment, based on linguistic instruction and visual inputs. VLN agents need access to both local and global action spaces; former for immediate decision making and the latter for recovering from navigational mistakes. Prior VLN agents rely only on instruction-viewpoint alignment for local and global decision making and back-track to a previously visited viewpoint, if the instruction and its current viewpoint mismatches. These methods are prone to mistakes, due to the complexity of the instruction and partial observability of the environment. We posit that, back-tracking is sub-optimal and agent that is aware of its mistakes can recover efficiently. For optimal recovery, exploration should be extended to unexplored viewpoints (or frontiers). The optimal frontier is a recently observed but unexplored viewpoint that aligns with the instruction and is novel. We introduce a memory-based and mistake-aware path planning strategy for VLN agents, called \textit{StratXplore}, that presents global and local action planning to select the optimal frontier for path correction. The proposed method collects all past actions and viewpoint features during navigation and then selects the optimal frontier suitable for recovery. Experimental results show this simple yet effective strategy improves the success rate on two VLN datasets with different task complexities.

Paperid: 6362, https://arxiv.org/pdf/2409.02278.pdf

Abstract:
Recent developments in vision language models (VLM) have shown great potential for diverse applications related to image understanding. In this study, we have explored state-of-the-art VLM models for vision-based transportation engineering tasks such as image classification and object detection. The image classification task involves congestion detection and crack identification, whereas, for object detection, helmet violations were identified. We have applied open-source models such as CLIP, BLIP, OWL-ViT, Llava-Next, and closed-source GPT-4o to evaluate the performance of these state-of-the-art VLM models to harness the capabilities of language understanding for vision-based transportation tasks. These tasks were performed by applying zero-shot prompting to the VLM models, as zero-shot prompting involves performing tasks without any training on those tasks. It eliminates the need for annotated datasets or fine-tuning for specific tasks. Though these models gave comparative results with benchmark Convolutional Neural Networks (CNN) models in the image classification tasks, for object localization tasks, it still needs improvement. Therefore, this study provides a comprehensive evaluation of the state-of-the-art VLM models highlighting the advantages and limitations of the models, which can be taken as the baseline for future improvement and wide-scale implementation.

Paperid: 6363, https://arxiv.org/pdf/2601.21078.pdf

Abstract:
Temporal Action Localization (TAL) requires identifying both the boundaries and categories of actions in untrimmed videos. While vision-language models (VLMs) offer rich semantics to complement visual evidence, existing approaches tend to overemphasize linguistic priors at the expense of visual performance, leading to a pronounced modality bias. We propose ActionVLM, a vision-language aggregation framework that systematically mitigates modality bias in TAL. Our key insight is to preserve vision as the dominant signal while adaptively exploiting language only when beneficial. To this end, we introduce (i) a debiasing reweighting module that estimates the language advantage-the incremental benefit of language over vision-only predictions-and dynamically reweights language modality accordingly, and (ii) a residual aggregation strategy that treats language as a complementary refinement rather than the primary driver. This combination alleviates modality bias, reduces overconfidence from linguistic priors, and strengthens temporal reasoning. Experiments on THUMOS14 show that our model outperforms state-of-the-art by up to 3.2% mAP.

Paperid: 6364, https://arxiv.org/pdf/2601.20075.pdf

Abstract:
Contrastive Language-Image Pre-training (CLIP) has become a cornerstone in vision-language representation learning, powering diverse downstream tasks and serving as the default vision backbone in multimodal large language models (MLLMs). Despite its success, CLIP's dense and opaque latent representations pose significant interpretability challenges. A common assumption is that interpretability and performance are in tension: enforcing sparsity during training degrades accuracy, motivating recent post-hoc approaches such as Sparse Autoencoders (SAEs). However, these post-hoc approaches often suffer from degraded downstream performance and loss of CLIP's inherent multimodal capabilities, with most learned features remaining unimodal. We propose a simple yet effective approach that integrates sparsity directly into CLIP training, yielding representations that are both interpretable and performant. Compared to SAEs, our Sparse CLIP representations preserve strong downstream task performance, achieve superior interpretability, and retain multimodal capabilities. We show that multimodal sparse features enable straightforward semantic concept alignment and reveal training dynamics of how cross-modal knowledge emerges. Finally, as a proof of concept, we train a vision-language model on sparse CLIP representations that enables interpretable, vision-based steering capabilities. Our findings challenge conventional wisdom that interpretability requires sacrificing accuracy and demonstrate that interpretability and performance can be co-optimized, offering a promising design principle for future models.

Paperid: 6365, https://arxiv.org/pdf/2601.19099.pdf

Abstract:
Vision--language models (VLMs) achieve strong performance on many multimodal benchmarks but remain brittle on spatial reasoning tasks that require aligning abstract overhead representations with egocentric views. We introduce m2sv, a scalable benchmark for map-to-street-view spatial reasoning that asks models to infer camera viewing direction by aligning a north-up overhead map with a Street View image captured at the same real-world intersection. We release m2sv-20k, a geographically diverse benchmark with controlled ambiguity, along with m2sv-sft-11k, a curated set of structured reasoning traces for supervised fine-tuning. Despite strong performance on existing multimodal benchmarks, the best evaluated VLM achieves only 65.2% accuracy on m2sv, far below the human baseline of 95%. While supervised fine-tuning and reinforcement learning yield consistent gains, cross-benchmark evaluations reveal limited transfer. Beyond aggregate accuracy, we systematically analyze difficulty in map-to-street-view reasoning using both structural signals and human effort, and conduct an extensive failure analysis of adapted open models. Our findings highlight persistent gaps in geometric alignment, evidence aggregation, and reasoning consistency, motivating future work on grounded spatial reasoning across viewpoints.

Paperid: 6366, https://arxiv.org/pdf/2601.17405.pdf

Abstract:
Precision pathology relies on detecting fine-grained morphological abnormalities within specific Regions of Interest (ROIs), as these local, texture-rich cues - rather than global slide contexts - drive expert diagnostic reasoning. While Vision-Language (V-L) models promise data efficiency by leveraging semantic priors, adapting them faces a critical Granularity Mismatch, where generic representations fail to resolve such subtle defects. Current adaptation methods often treat modalities as independent streams, failing to ground semantic prompts in ROI-specific visual contexts. To bridge this gap, we propose the Hierarchical Adaptation and Alignment Framework (HAAF). At its core is a novel Cross-Level Scaled Alignment (CLSA) mechanism that enforces a sequential calibration order: visual features first inject context into text prompts to generate content-adaptive descriptors, which then spatially guide the visual encoder to spotlight anomalies. Additionally, a dual-branch inference strategy integrates semantic scores with geometric prototypes to ensure stability in few-shot settings. Experiments on four benchmarks show HAAF significantly outperforms state-of-the-art methods and effectively scales with domain-specific backbones (e.g., CONCH) in low-resource scenarios.

Paperid: 6367, https://arxiv.org/pdf/2601.16549.pdf

Abstract:
The combination of multimodal Vision-Language Models (VLMs) and Large Language Models (LLMs) opens up new possibilities for medical classification. This work offers a rigorous, unified benchmark by using four publicly available datasets covering text and image modalities (binary and multiclass complexity) that contrasts traditional Machine Learning (ML) with contemporary transformer-based techniques. We evaluated three model classes for each task: Classical ML (LR, LightGBM, ResNet-50), Prompt-Based LLMs/VLMs (Gemini 2.5), and Fine-Tuned PEFT Models (LoRA-adapted Gemma3 variants). All experiments used consistent data splits and aligned metrics. According to our results, traditional machine learning (ML) models set a high standard by consistently achieving the best overall performance across most medical categorization tasks. This was especially true for structured text-based datasets, where the classical models performed exceptionally well. In stark contrast, the LoRA-tuned Gemma variants consistently showed the worst performance across all text and image experiments, failing to generalize from the minimal fine-tuning provided. However, the zero-shot LLM/VLM pipelines (Gemini 2.5) had mixed results; they performed poorly on text-based tasks, but demonstrated competitive performance on the multiclass image task, matching the classical ResNet-50 baseline. These results demonstrate that in many medical categorization scenarios, established machine learning models continue to be the most reliable option. The experiment suggests that foundation models are not universally superior and that the effectiveness of Parameter-Efficient Fine-Tuning (PEFT) is highly dependent on the adaptation strategy, as minimal fine-tuning proved detrimental in this study.

Paperid: 6368, https://arxiv.org/pdf/2601.15780.pdf

Abstract:
Spatial reasoning in vision language models (VLMs) remains fragile when semantics hinge on subtle temporal or geometric cues. We introduce a synthetic benchmark that probes two complementary skills: situational awareness (recognizing whether an interaction is harmful or benign) and spatial awareness (tracking who does what to whom, and reasoning about relative positions and motion). Through minimal video pairs, we test three challenges: distinguishing violence from benign activity, binding assailant roles across viewpoints, and judging fine-grained trajectory alignment. While we evaluate recent VLMs in a training-free setting, the benchmark is applicable to any video classification model. Results show performance only slightly above chance across tasks. A simple aid, stable color cues, partly reduces assailant role confusions but does not resolve the underlying weakness. By releasing data and code, we aim to provide reproducible diagnostics and seed exploration of lightweight spatial priors to complement large-scale pretraining.

Paperid: 6369, https://arxiv.org/pdf/2601.14610.pdf

Abstract:
While Vision-Language Models (VLMs) excel at visual understanding, they often fail to grasp hierarchical knowledge. This leads to common errors where VLMs misclassify coarser taxonomic levels even when correctly identifying the most specific level (leaf level). Existing approaches largely overlook this issue by failing to model hierarchical reasoning. To address this gap, we propose VL-Taxon, a two-stage, hierarchy-based reasoning framework designed to improve both leaf-level accuracy and hierarchical consistency in taxonomic classification. The first stage employs a top-down process to enhance leaf-level classification accuracy. The second stage then leverages this accurate leaf-level output to ensure consistency throughout the entire taxonomic hierarchy. Each stage is initially trained with supervised fine-tuning to instill taxonomy knowledge, followed by reinforcement learning to refine the model's reasoning and generalization capabilities. Extensive experiments reveal a remarkable result: our VL-Taxon framework, implemented on the Qwen2.5-VL-7B model, outperforms its original 72B counterpart by over 10% in both leaf-level and hierarchical consistency accuracy on average on the iNaturalist-2021 dataset. Notably, this significant gain was achieved by fine-tuning on just a small subset of data, without relying on any examples generated by other VLMs.

Paperid: 6370, https://arxiv.org/pdf/2601.11243.pdf

Abstract:
We propose unsupervised multi-scenario (UMS) person re-identification (ReID) as a new task that expands ReID across diverse scenarios (cross-resolution, clothing change, etc.) within a single coherent framework. To tackle UMS-ReID, we introduce image-text knowledge modeling (ITKM) -- a three-stage framework that effectively exploits the representational power of vision-language models. We start with a pre-trained CLIP model with an image encoder and a text encoder. In Stage I, we introduce a scenario embedding in the image encoder and fine-tune the encoder to adaptively leverage knowledge from multiple scenarios. In Stage II, we optimize a set of learned text embeddings to associate with pseudo-labels from Stage I and introduce a multi-scenario separation loss to increase the divergence between inter-scenario text representations. In Stage III, we first introduce cluster-level and instance-level heterogeneous matching modules to obtain reliable heterogeneous positive pairs (e.g., a visible image and an infrared image of the same person) within each scenario. Next, we propose a dynamic text representation update strategy to maintain consistency between text and image supervision signals. Experimental results across multiple scenarios demonstrate the superiority and generalizability of ITKM; it not only outperforms existing scenario-specific methods but also enhances overall performance by integrating knowledge from multiple scenarios.

Paperid: 6371, https://arxiv.org/pdf/2601.10922.pdf

Abstract:
We study data curation for multimodal reasoning through the NeurIPS 2025 Data Curation for Vision-Language Reasoning (DCVLR) challenge, which isolates dataset selection by fixing the model and training protocol. Using a compact curated dataset derived primarily from Walton Multimodal Cold Start, our submission placed first in the challenge. Through post-competition ablations, we show that difficulty-based example selection on an aligned base dataset is the dominant driver of performance gains. Increasing dataset size does not reliably improve mean accuracy under the fixed training recipe, but mainly reduces run-to-run variance, while commonly used diversity and synthetic augmentation heuristics provide no additional benefit and often degrade performance. These results characterize DCVLR as a saturation-regime evaluation and highlight the central role of alignment and difficulty in data-efficient multimodal reasoning.

Paperid: 6372, https://arxiv.org/pdf/2601.09954.pdf

Abstract:
Vision-language models (VLMs) have advanced rapidly, but their ability to capture spatial relationships remains a blindspot. Current VLMs are typically built with contrastive language-image pretraining (CLIP) style image encoders. The training recipe often flattens images into 1D patch sequences, discarding the 2D structure necessary for spatial reasoning. We argue that this lack of spatial awareness is a missing dimension in VLM design and a bottleneck for applications requiring spatial grounding, such as robotics and embodied AI. To address this, we investigate (i) image encoders trained with alternative objectives and (ii) 2D positional encodings. Our experiments show that these architectural choices can lead to improved spatial reasoning on several benchmarks.

Paperid: 6373, https://arxiv.org/pdf/2601.07695.pdf

Abstract:
Vision-Language Models (VLMs) face a critical bottleneck in achieving precise numerical prediction for 3D scene understanding. Traditional reinforcement learning (RL) approaches, primarily based on relative ranking, often suffer from severe reward sparsity and gradient instability, failing to effectively exploit the verifiable signals provided by 3D physical constraints. Notably, in standard GRPO frameworks, relative normalization causes "near-miss" samples (characterized by small but non-zero errors) to suffer from advantage collapse. This leads to a severe data utilization bottleneck where valuable boundary samples are discarded during optimization. To address this, we introduce the Smooth Numerical Reward Activation (SNRA) operator and the Absolute-Preserving GRPO (AP-GRPO) framework. SNRA employs a dynamically parameterized Sigmoid function to transform raw feedback into a dense, continuous reward continuum. Concurrently, AP-GRPO integrates absolute scalar gradients to mitigate the numerical information loss inherent in conventional relative-ranking mechanisms. By leveraging this approach, we constructed Numerical3D-50k, a dataset comprising 50,000 verifiable 3D subtasks. Empirical results indicate that AP-GRPO achieves performance parity with large-scale supervised methods while maintaining higher data efficiency, effectively activating latent 3D reasoning in VLMs without requiring architectural modifications.

Paperid: 6374, https://arxiv.org/pdf/2601.04163.pdf

Abstract:
Pathology foundation models (PFMs) have become central to computational pathology, aiming to offer general encoders for feature extraction from whole-slide images (WSIs). Despite strong benchmark performance, PFM robustness to real-world technical domain shifts, such as variability from whole-slide scanner devices, remains poorly understood. We systematically evaluated the robustness of 14 PFMs to scanner-induced variability, including state-of-the-art models, earlier self-supervised models, and a baseline trained on natural images. Using a multiscanner dataset of 384 breast cancer WSIs scanned on five devices, we isolated scanner effects independently from biological and laboratory confounders. Robustness is assessed via complementary unsupervised embedding analyses and a set of clinicopathological supervised prediction tasks. Our results demonstrate that current PFMs are not invariant to scanner-induced domain shifts. Most models encode pronounced scanner-specific variability in their embedding spaces. While AUC often remains stable, this masks a critical failure mode: scanner variability systematically alters the embedding space and impacts calibration of downstream model predictions, resulting in scanner-dependent bias that can impact reliability in clinical use cases. We further show that robustness is not a simple function of training data scale, model size, or model recency. None of the models provided reliable robustness against scanner-induced variability. While the models trained on the most diverse data, here represented by vision-language models, appear to have an advantage with respect to robustness, they underperformed on downstream supervised tasks. We conclude that development and evaluation of PFMs requires moving beyond accuracy-centric benchmarks toward explicit evaluation and optimisation of embedding stability and calibration under realistic acquisition variability.

Paperid: 6375, https://arxiv.org/pdf/2601.04118.pdf

Abstract:
The evolution of Remote Sensing Vision-Language Models(RS-VLMs) emphasizes the importance of transitioning from perception-centric recognition toward high-level deductive reasoning to enhance cognitive reliability in complex spatial tasks. However, current models often suffer from logical hallucinations, where correct answers are derived from flawed reasoning chains or rely on positional shortcuts rather than spatial logic. This decoupling undermines reliability in strategic spatial decision-making. To address this, we present GeoReason, a framework designed to synchronize internal thinking with final decisions. We first construct GeoReason-Bench, a logic-driven dataset containing 4,000 reasoning trajectories synthesized from geometric primitives and expert knowledge. We then formulate a two-stage training strategy: (1) Supervised Knowledge Initialization to equip the model with reasoning syntax and domain expertise, and (2) Consistency-Aware Reinforcement Learning to refine deductive reliability. This second stage integrates a novel Logical Consistency Reward, which penalizes logical drift via an option permutation strategy to anchor decisions in verifiable reasoning traces. Experimental results demonstrate that our framework significantly enhances the cognitive reliability and interpretability of RS-VLMs, achieving state-of-the-art performance compared to other advanced methods.

Paperid: 6376, https://arxiv.org/pdf/2601.01948.pdf

Abstract:
Diffusion policies (DP) have recently shown great promise for generating actions in robotic manipulation. However, existing approaches often rely on global instructions to produce short-term control signals, which can result in misalignment in action generation. We conjecture that the primitive skills, referred to as fine-grained, short-horizon manipulations, such as ``move up'' and ``open the gripper'', provide a more intuitive and effective interface for robot learning. To bridge this gap, we propose SDP, a skill-conditioned DP that integrates interpretable skill learning with conditional action planning. SDP abstracts eight reusable primitive skills across tasks and employs a vision-language model to extract discrete representations from visual observations and language instructions. Based on them, a lightweight router network is designed to assign a desired primitive skill for each state, which helps construct a single-skill policy to generate skill-aligned actions. By decomposing complex tasks into a sequence of primitive skills and selecting a single-skill policy, SDP ensures skill-consistent behavior across diverse tasks. Extensive experiments on two challenging simulation benchmarks and real-world robot deployments demonstrate that SDP consistently outperforms SOTA methods, providing a new paradigm for skill-based robot learning with diffusion policies.

Paperid: 6377, https://arxiv.org/pdf/2601.00716.pdf

Abstract:
Vision-Language Models have demonstrated strong potential in medical image analysis and disease diagnosis. However, after deployment, their performance may deteriorate when the input data distribution shifts from that observed during development. Detecting such performance degradation is essential for clinical reliability, yet remains challenging for large pre-trained VLMs operating without labeled data. In this study, we investigate performance degradation detection under data shift in a state-of-the-art pathology VLM. We examine both input-level data shift and output-level prediction behavior to understand their respective roles in monitoring model reliability. To facilitate systematic analysis of input data shift, we develop DomainSAT, a lightweight toolbox with a graphical interface that integrates representative shift detection algorithms and enables intuitive exploration of data shift. Our analysis shows that while input data shift detection is effective at identifying distributional changes and providing early diagnostic signals, it does not always correspond to actual performance degradation. Motivated by this observation, we further study output-based monitoring and introduce a label-free, confidence-based degradation indicator that directly captures changes in model prediction confidence. We find that this indicator exhibits a close relationship with performance degradation and serves as an effective complement to input shift detection. Experiments on a large-scale pathology dataset for tumor classification demonstrate that combining input data shift detection and output confidence-based indicators enables more reliable detection and interpretation of performance degradation in VLMs under data shift. These findings provide a practical and complementary framework for monitoring the reliability of foundation models in digital pathology.

Paperid: 6378, https://arxiv.org/pdf/2601.00260.pdf

Abstract:
While foundation models in radiology are expected to be applied to various clinical tasks, computational cost constraints remain a major challenge when training on 3D-CT volumetric data. In this study, we propose TotalFM, a radiological foundation model that efficiently learns the correspondence between 3D-CT images and linguistic expressions based on the concept of organ separation, utilizing a large-scale dataset of 140,000 series. By automating the creation of organ volume and finding-sentence pairs through segmentation techniques and Large Language Model (LLM)-based radiology report processing, and by combining self-supervised pre-training via VideoMAE with contrastive learning using volume-text pairs, we aimed to balance computational efficiency and representation capability. In zero-shot organ-wise lesion classification tasks, the proposed model achieved higher F1 scores in 83% (5/6) of organs compared to CT-CLIP and 64% (9/14) of organs compared to Merlin. These results suggest that the proposed model exhibits high generalization performance in a clinical evaluation setting using actual radiology report sentences. Furthermore, in zero-shot finding-wise lesion classification tasks, our model achieved a higher AUROC in 83% (25/30) of finding categories compared to Merlin. We also confirmed performance comparable to existing Vision-Language Models (VLMs) in radiology report generation tasks. Our results demonstrate that the organ-separated learning framework can serve as a realistic and effective design guideline for the practical implementation of 3D-CT foundation models.

Paperid: 6379, https://arxiv.org/pdf/2512.22867.pdf

Abstract:
Socially compliant navigation requires structured reasoning over dynamic pedestrians and physical constraints to ensure safe and interpretable decisions. However, existing social navigation datasets often lack explicit reasoning supervision and exhibit highly long-tailed action distributions, limiting models' ability to learn safety-critical behaviors. To address these issues, we introduce MUSON, a multimodal dataset for short-horizon social navigation collected across diverse indoor and outdoor campus scenes. MUSON adopts a structured five-step Chain-of-Thought annotation consisting of perception, prediction, reasoning, action, and explanation, with explicit modeling of static physical constraints and a rationally balanced discrete action space. Compared to SNEI, MUSON provides consistent reasoning, action, and explanation. Benchmarking multiple state-of-the-art Small Vision Language Models on MUSON shows that Qwen2.5-VL-3B achieves the highest decision accuracy of 0.8625, demonstrating that MUSON serves as an effective and reusable benchmark for socially compliant navigation. The dataset is publicly available at https://huggingface.co/datasets/MARSLab/MUSON

Paperid: 6380, https://arxiv.org/pdf/2512.21670.pdf

Abstract:
Deepfake detection models have achieved high accuracy in identifying synthetic media, but their decision processes remain largely opaque. In this paper we present a mechanistic interpretability framework for deepfake detection applied to a vision-language model. Our approach combines a sparse autoencoder (SAE) analysis of internal network representations with a novel forensic manifold analysis that probes how the model's features respond to controlled forensic artifact manipulations. We demonstrate that only a small fraction of latent features are actively used in each layer, and that the geometric properties of the model's feature manifold, including intrinsic dimensionality, curvature, and feature selectivity, vary systematically with different types of deepfake artifacts. These insights provide a first step toward opening the "black box" of deepfake detectors, allowing us to identify which learned features correspond to specific forensic artifacts and to guide the development of more interpretable and robust models.

Paperid: 6381, https://arxiv.org/pdf/2512.21560.pdf

Abstract:
Intelligent image editing increasingly relies on advances in computer vision, multimodal reasoning, and generative modeling. While vision-language models (VLMs) and diffusion models enable guided visual manipulation, existing work rarely ensures that inserted objects are \emph{contextually appropriate}. We introduce two new tasks for advertising and digital media: (1) \emph{context-aware object insertion}, which requires predicting suitable object categories, generating them, and placing them plausibly within the scene; and (2) \emph{sponsor-product logo augmentation}, which involves detecting products and inserting correct brand logos, even when items are unbranded or incorrectly branded. To support these tasks, we build two new datasets with category annotations, placement regions, and sponsor-product labels.

Paperid: 6382, https://arxiv.org/pdf/2512.20174.pdf

Abstract:
Document image retrieval (DIR) aims to retrieve document images from a gallery according to a given query. Existing DIR methods are primarily based on image queries that retrieve documents within the same coarse semantic category, e.g., newspapers or receipts. However, these methods struggle to effectively retrieve document images in real-world scenarios where textual queries with fine-grained semantics are usually provided. To bridge this gap, we introduce a new Natural Language-based Document Image Retrieval (NL-DIR) benchmark with corresponding evaluation metrics. In this work, natural language descriptions serve as semantically rich queries for the DIR task. The NL-DIR dataset contains 41K authentic document images, each paired with five high-quality, fine-grained semantic queries generated and evaluated through large language models in conjunction with manual verification. We perform zero-shot and fine-tuning evaluations of existing mainstream contrastive vision-language models and OCR-free visual document understanding (VDU) models. A two-stage retrieval method is further investigated for performance improvement while achieving both time and space efficiency. We hope the proposed NL-DIR benchmark can bring new opportunities and facilitate research for the VDU community. Datasets and codes will be publicly available at huggingface.co/datasets/nianbing/NL-DIR.

Paperid: 6383, https://arxiv.org/pdf/2512.19535.pdf

Abstract:
Vision-language models (VLMs) are commonly trained by inserting image tokens from a pretrained vision encoder into the textual stream of a language model. This allows text and image information to fully attend to one another within the model, but becomes extremely costly for high-resolution images, long conversations, or streaming videos, both in memory and compute. VLMs leveraging cross-attention are an efficient alternative to token insertion but exhibit a clear performance gap, in particular on tasks involving fine-grained visual details. We find that a key to improving such models is to also enable local text-to-text interaction in the dedicated cross-attention layers. Building on this, we propose CASA, Cross-Attention via Self-Attention, a simple and efficient paradigm which substantially reduces the gap with full token insertion on common image understanding benchmarks, while enjoying the same scalability as cross-attention models when applied to long-context multimodal tasks such as streaming video captioning. For samples and code, please see our project page at https://kyutai.org/casa .

Paperid: 6384, https://arxiv.org/pdf/2512.18496.pdf

Abstract:
In recent years, large-scale vision-language models (VLMs) have demonstrated remarkable performance on multimodal understanding and reasoning tasks. However, handling high-dimensional visual features often incurs substantial computational and memory costs. VoCo-LLaMA alleviates this issue by compressing visual patch tokens into a few VoCo tokens, reducing computational overhead while preserving strong cross-modal alignment. Nevertheless, such approaches typically adopt a fixed compression rate, limiting their ability to adapt to varying levels of visual complexity. To address this limitation, we propose Adaptive-VoCo, a framework that augments VoCo-LLaMA with a lightweight predictor for adaptive compression. This predictor dynamically selects an optimal compression rate by quantifying an image's visual complexity using statistical cues from the vision encoder, such as patch token entropy and attention map variance. Furthermore, we introduce a joint loss function that integrates rate regularization with complexity alignment. This enables the model to balance inference efficiency with representational capacity, particularly in challenging scenarios. Experimental results show that our method consistently outperforms fixed-rate baselines across multiple multimodal tasks, highlighting the potential of adaptive visual compression for creating more efficient and robust VLMs.

Paperid: 6385, https://arxiv.org/pdf/2512.17394.pdf

Abstract:
Theory of Mind (ToM) -- the ability to attribute beliefs, desires, and emotions to others -- is fundamental for human social intelligence, yet remains a major challenge for artificial agents. Existing Vision-Language Models (VLMs) are increasingly applied in socially grounded tasks, but their capacity for cross-cultural ToM reasoning is largely unexplored. In this work, we introduce CulturalToM-VQA, a new evaluation benchmark containing 5095 questions designed to probe ToM reasoning across diverse cultural contexts through visual question answering. The dataset captures culturally grounded cues such as rituals, attire, gestures, and interpersonal dynamics, enabling systematic evaluation of ToM reasoning beyond Western-centric benchmarks. Our dataset is built through a VLM-assisted human-in-the-loop pipeline, where human experts first curate culturally rich images across traditions, rituals, and social interactions; a VLM then assist in generating structured ToM-focused scene descriptions, which are refined into question-answer pairs spanning a taxonomy of six ToM tasks and four graded complexity levels. The resulting dataset covers diverse theory of mind facets such as mental state attribution, false belief reasoning, non-literal communication, social norm violations, perspective coordination, and multi-agent reasoning.

Paperid: 6386, https://arxiv.org/pdf/2512.15692.pdf

Abstract:
Prevailing Vision-Language-Action Models (VLAs) for robotic manipulation are built upon vision-language backbones pretrained on large-scale, but disconnected static web data. As a result, despite improved semantic generalization, the policy must implicitly infer complex physical dynamics and temporal dependencies solely from robot trajectories. This reliance creates an unsustainable data burden, necessitating continuous, large-scale expert data collection to compensate for the lack of innate physical understanding. We contend that while vision-language pretraining effectively captures semantic priors, it remains blind to physical causality. A more effective paradigm leverages video to jointly capture semantics and visual dynamics during pretraining, thereby isolating the remaining task of low-level control. To this end, we introduce mimic-video, a novel Video-Action Model (VAM) that pairs a pretrained Internet-scale video model with a flow matching-based action decoder conditioned on its latent representations. The decoder serves as an Inverse Dynamics Model (IDM), generating low-level robot actions from the latent representation of video-space action plans. Our extensive evaluation shows that our approach achieves state-of-the-art performance on simulated and real-world robotic manipulation tasks, improving sample efficiency by 10x and convergence speed by 2x compared to traditional VLA architectures.

Paperid: 6387, https://arxiv.org/pdf/2512.14757.pdf

Abstract:
For robots navigating in human-populated environments, safety and social compliance are equally critical, yet prior work has mostly emphasized safety. Socially compliant navigation that accounts for human comfort, social norms, and contextual appropriateness remains underexplored. Vision language models (VLMs) show promise for this task; however, large-scale models incur substantial computational overhead, leading to higher inference latency and energy consumption, which makes them unsuitable for real-time deployment on resource-constrained robotic platforms. To address this issue, we investigate the effectiveness of small VLM and propose SocialNav-MoE, an efficient Mixture-of-Experts vision language model for socially compliant navigation with reinforcement fine-tuning (RFT). We further introduce a semantic similarity reward (SSR) to effectively leverage RFT for enhancing the decision-making capabilities. Additionally, we study the effectiveness of different small language model types (Phi, Qwen, and StableLM), routing strategies, and vision encoders (CLIP vs. SigLIP, frozen vs. fine-tuned). Experiments on the SNEI dataset demonstrate that SocialNav-MoE achieves an excellent balance between navigation accuracy and efficiency. The proposed SSR function is more effective than hard-level and character-level rewards. Source code will be released upon acceptance.

Paperid: 6388, https://arxiv.org/pdf/2512.14715.pdf

Abstract:
Hard-to-detect hardware bit flips, from either malicious circuitry or bugs, have already been shown to make transformers vulnerable in non-generative tasks. This work, for the first time, investigates how low-level, bitwise perturbations (fault injection) to the weights of a large language model (LLM) used for image captioning can influence the semantic meaning of its generated descriptions while preserving grammatical structure. While prior fault analysis methods have shown that flipping a few bits can crash classifiers or degrade accuracy, these approaches overlook the semantic and linguistic dimensions of generative systems. In image captioning models, a single flipped bit might subtly alter how visual features map to words, shifting the entire narrative an AI tells about the world. We hypothesize that such semantic drifts are not random but differentiably estimable. That is, the model's own gradients can predict which bits, if perturbed, will most strongly influence meaning while leaving syntax and fluency intact. We design a differentiable fault analysis framework, BLADE (Bit-level Fault Analysis via Differentiable Estimation), that uses gradient-based sensitivity estimation to locate semantically critical bits and then refines their selection through a caption-level semantic-fluency objective. Our goal is not merely to corrupt captions, but to understand how meaning itself is encoded, distributed, and alterable at the bit level, revealing that even imperceptible low-level changes can steer the high-level semantics of generative vision-language models. It also opens pathways for robustness testing, adversarial defense, and explainable AI, by exposing how structured bit-level faults can reshape a model's semantic output.

Paperid: 6389, https://arxiv.org/pdf/2512.14044.pdf

Abstract:
The deployment of Vision-Language Models (VLMs) in safety-critical domains like autonomous driving (AD) is critically hindered by reliability failures, most notably object hallucination. This failure stems from their reliance on ungrounded, text-based Chain-of-Thought (CoT) reasoning. While existing multi-modal CoT approaches attempt mitigation, they suffer from two fundamental flaws: (1) decoupled perception and reasoning stages that prevent end-to-end joint optimization, and (2) reliance on expensive, dense localization labels. Thus we introduce OmniDrive-R1, an end-to-end VLM framework designed for autonomous driving, which unifies perception and reasoning through an interleaved Multi-modal Chain-of-Thought (iMCoT) mechanism. Our core innovation is an Reinforcement-driven visual grounding capability, enabling the model to autonomously direct its attention and "zoom in" on critical regions for fine-grained analysis. This capability is enabled by our pure two-stage reinforcement learning training pipeline and Clip-GRPO algorithm. Crucially, Clip-GRPO introduces an annotation-free, process-based grounding reward. This reward not only eliminates the need for dense labels but also circumvents the instability of external tool calls by enforcing real-time cross-modal consistency between the visual focus and the textual reasoning. Extensive experiments on DriveLMM-o1 demonstrate our model's significant improvements. Compared to the baseline Qwen2.5VL-7B, OmniDrive-R1 improves the overall reasoning score from 51.77% to 80.35%, and the final answer accuracy from 37.81% to 73.62%.

Paperid: 6390, https://arxiv.org/pdf/2512.13974.pdf

Abstract:
Construction safety inspection remains mostly manual, and automated approaches still rely on task-specific datasets that are hard to maintain in fast-changing construction environments due to frequent retraining. Meanwhile, field inspection with robots still depends on human teleoperation and manual reporting, which are labor-intensive. This paper aims to connect what a robot sees during autonomous navigation to the safety rules that are common in construction sites, automatically generating a safety inspection report. To this end, we proposed a multi-layer framework with two main modules: robotics and AI. On the robotics side, SLAM and autonomous navigation provide repeatable coverage and targeted revisits via waypoints. On AI side, a Vision Language Model (VLM)-based layer produces scene descriptions; a retrieval component powered grounds those descriptions in OSHA and site policies; Another VLM-based layer assesses the safety situation based on rules; and finally Large Language Model (LLM) layer generates safety reports based on previous outputs. The framework is validated with a proof-of-concept implementation and evaluated in a lab environment that simulates common hazards across three scenarios. Results show high recall with competitive precision compared to state-of-the-art closed-source models. This paper contributes a transparent, generalizable pipeline that moves beyond black-box models by exposing intermediate artifacts from each layer and keeping the human in the loop. This work provides a foundation for future extensions to additional tasks and settings within and beyond construction context.

Paperid: 6391, https://arxiv.org/pdf/2512.12793.pdf

Abstract:
This paper presents Vision-Language Global Localization (VLG-Loc), a novel global localization method that uses human-readable labeled footprint maps containing only names and areas of distinctive visual landmarks in an environment. While humans naturally localize themselves using such maps, translating this capability to robotic systems remains highly challenging due to the difficulty of establishing correspondences between observed landmarks and those in the map without geometric and appearance details. To address this challenge, VLG-Loc leverages a vision-language model (VLM) to search the robot's multi-directional image observations for the landmarks noted in the map. The method then identifies robot poses within a Monte Carlo localization framework, where the found landmarks are used to evaluate the likelihood of each pose hypothesis. Experimental validation in simulated and real-world retail environments demonstrates superior robustness compared to existing scan-based methods, particularly under environmental changes. Further improvements are achieved through the probabilistic fusion of visual and scan-based localization.

Paperid: 6392, https://arxiv.org/pdf/2512.11865.pdf

Abstract:
Smart farming has emerged as a key technology for advancing modern agriculture through automation and intelligent control. However, systems relying on RGB cameras for perception and robotic manipulators for control, common in smart farming, are vulnerable to photometric perturbations such as hue, illumination, and noise changes, which can cause malfunction under adversarial attacks. To address this issue, we propose an explainable adversarial-robust Vision-Language-Action model based on the OpenVLA-OFT framework. The model integrates an Evidence-3 module that detects photometric perturbations and generates natural language explanations of their causes and effects. Experiments show that the proposed model reduces Current Action L1 loss by 21.7% and Next Actions L1 loss by 18.4% compared to the baseline, demonstrating improved action prediction accuracy and explainability under adversarial conditions.

Paperid: 6393, https://arxiv.org/pdf/2512.10414.pdf

Abstract:
Recently, reinforcement learning (RL) has become a common choice in enhancing the reasoning capabilities of vision-language models (VLMs). Considering existing RL-based finetuning methods, entropy intervention turns out to be an effective way to benefit exploratory ability, thereby improving policy performance. Notably, most existing studies intervene in entropy by simply controlling the update of specific tokens during policy optimization of RL. They ignore the entropy intervention during the RL sampling that can boost the performance of GRPO by improving the diversity of responses. In this paper, we propose Selective-adversarial Entropy Intervention, namely SaEI, which enhances policy entropy by distorting the visual input with the token-selective adversarial objective coming from the entropy of sampled responses. Specifically, we first propose entropy-guided adversarial sampling (EgAS) that formulates the entropy of sampled responses as an adversarial objective. Then, the corresponding adversarial gradient can be used to attack the visual input for producing adversarial samples, allowing the policy model to explore a larger answer space during RL sampling. Then, we propose token-selective entropy computation (TsEC) to maximize the effectiveness of adversarial attack in EgAS without distorting factual knowledge within VLMs. Extensive experiments on both in-domain and out-of-domain datasets show that our proposed method can greatly improve policy exploration via entropy intervention, to boost reasoning capabilities. Code will be released once the paper is accepted.

Paperid: 6394, https://arxiv.org/pdf/2512.08627.pdf

Abstract:
In the absence of a mechanical stabilizer, the camera undergoes inevitable rotational dynamics during capturing, which induces perspective-based blur especially under long-exposure scenarios. From an optical standpoint, perspective-based blur is depth-position-dependent: objects residing at distinct spatial locations incur different blur levels even under the same imaging settings. Inspired by this, we propose a novel method that estimate metric depth by examining the blur pattern of a video stream and dense trajectory via joint optical design algorithm. Specifically, we employ off-the-shelf vision encoder and point tracker to extract video information. Then, we estimate depth map via windowed embedding and multi-window aggregation, and densify the sparse trajectory from the optical algorithm using a vision-language model. Evaluations on multiple depth datasets demonstrate that our method attains strong performance over large depth range, while maintaining favorable generalization. Relative to the real trajectory in handheld shooting settings, our optical algorithm achieves superior precision and the dense reconstruction maintains strong accuracy.

Paperid: 6395, https://arxiv.org/pdf/2512.06811.pdf

Abstract:
Pre-trained Vision-Language Models (VLMs), \textit{e.g.} CLIP, have become essential tools in multimodal transfer learning. However, fine-tuning VLMs in few-shot scenarios poses significant challenges in balancing task-specific adaptation and generalization in the obtained model. Meanwhile, current researches have predominantly focused on prompt-based adaptation methods, leaving adapter-based approaches underexplored and revealing notable performance gaps. To address these challenges, we introduce a novel Reconstruction-based Multimodal Adapter (RMAdapter), which leverages a dual-branch architecture. Unlike conventional single-branch adapters, RMAdapter consists of: (1) an adaptation branch that injects task-specific knowledge through parameter-efficient fine-tuning, and (2) a reconstruction branch that preserves general knowledge by reconstructing latent space features back into the original feature space. This design facilitates a dynamic balance between general and task-specific knowledge. Importantly, although RMAdapter introduces an additional reconstruction branch, it is carefully optimized to remain lightweight. By computing reconstruction loss locally at each layer and sharing projection modules, the overall computational overhead is kept minimal. A consistency constraint is also incorporated to better regulate the trade-off between discriminability and generalization. We comprehensively evaluate the effectiveness of RMAdapter on three representative tasks: generalization to new categories, generalization to new target datasets, and domain generalization. Without relying on data augmentation or duplicate prompt designs, our RMAdapter consistently outperforms state-of-the-art approaches across all evaluation metrics.

Paperid: 6396, https://arxiv.org/pdf/2512.06328.pdf

Abstract:
We present ReCAD, a reinforcement learning (RL) framework that bootstraps pretrained large models (PLMs) to generate precise parametric computer-aided design (CAD) models from multimodal inputs by leveraging their inherent generative capabilities. With just access to simple functional interfaces (e.g., point coordinates), our approach enables the emergence of complex CAD operations (e.g., pattern replication and mirror). This stands in contrast to previous methods, which typically rely on knowledge injected through supervised fine-tuning (SFT), offer limited support for editability, and fail to exploit the strong generative priors of PLMs. Specifically, the ReCAD framework begins by fine-tuning vision-language models (VLMs) to equip them with basic CAD model generation capabilities, where we rewrite CAD scripts into parameterized code that is leveraged to generate accurate textual descriptions for supervision. Then, we propose a novel RL strategy that incorporates parameterized code as guidance to enhance the model's reasoning on challenging questions. Furthermore, we employ a hierarchical primitive learning process to progressively teach structured and compositional skills under a unified reward function that ensures both geometric accuracy and semantic fidelity. ReCAD sets a new state-of-the-art in both text-to-CAD and image-to-CAD tasks, significantly improving geometric accuracy across in-distribution and out-of-distribution settings. In the image-to-CAD task, for instance, it reduces the mean Chamfer Distance from 73.47 to 29.61 (in-distribution) and from 272.06 to 80.23 (out-of-distribution), outperforming existing baselines by a substantial margin.

Paperid: 6397, https://arxiv.org/pdf/2512.01922.pdf

Abstract:
Large vision-language models (LVLMs) are now central to healthcare applications such as medical visual question answering and imaging report generation. Yet, these models remain vulnerable to hallucination outputs that appear plausible but are in fact incorrect. In the natural image domain, several decoding strategies have been proposed to mitigate hallucinations by reinforcing visual evidence, but most rely on secondary decoding or rollback procedures that substantially slow inference. Moreover, existing solutions are often domain-specific and may introduce misalignment between modalities or between generated and ground-truth content. We introduce Med-VCD, a sparse visual-contrastive decoding method that mitigates hallucinations in medical LVLMs without the time overhead of secondary decoding. Med-VCD incorporates a novel token-sparsification strategy that selects visually informed tokens on the fly, trimming redundancy while retaining critical visual context and thus balancing efficiency with reliability. Evaluations on eight medical datasets, spanning ophthalmology, radiology, and pathology tasks in visual question answering, report generation, and dedicated hallucination benchmarks, show that Med-VCD raises factual accuracy by an average of 13\% and improves hallucination accuracy by 6\% relative to baseline medical LVLMs.

Paperid: 6398, https://arxiv.org/pdf/2512.01889.pdf

Abstract:
We present KM-ViPE (Knowledge Mapping Video Pose Engine), a real-time open-vocabulary SLAM framework for uncalibrated monocular cameras in dynamic environments. Unlike systems requiring depth sensors and offline calibration, KM-ViPE operates directly on raw RGB streams, making it ideal for ego-centric applications and harvesting internet-scale video data for training. KM-ViPE tightly couples DINO visual features with geometric constraints through a high-level features based adaptive robust kernel that handles both moving objects and movable static objects (e.g., moving furniture in ego-centric views). The system performs simultaneous online localization and open-vocabulary semantic mapping by fusing geometric and deep visual features aligned with language embeddings. Our results are competitive with state-of-the-art approaches, while existing solutions either operate offline, need depth data and/or odometry estimation, or lack dynamic scene robustness. KM-ViPE benefits from internet-scale training and uniquely combines online operation, uncalibrated monocular input, and robust handling of dynamic scenes, which makes it a good fit for autonomous robotics and AR/VR applications and advances practical spatial intelligence capabilities for embodied AI.

Paperid: 6399, https://arxiv.org/pdf/2512.00194.pdf

Abstract:
We introduce EEG Autoclean Vision Language AI (ICVision) a first-of-its-kind system that emulates expert-level EEG ICA component classification through AI-agent vision and natural language reasoning. Unlike conventional classifiers such as ICLabel, which rely on handcrafted features, ICVision directly interprets ICA dashboard visualizations topography, time series, power spectra, and ERP plots, using a multimodal large language model (GPT-4 Vision). This allows the AI to see and explain EEG components the way trained neurologists do, making it the first scientific implementation of AI-agent visual cognition in neurophysiology. ICVision classifies each component into one of six canonical categories (brain, eye, heart, muscle, channel noise, and other noise), returning both a confidence score and a human-like explanation. Evaluated on 3,168 ICA components from 124 EEG datasets, ICVision achieved k = 0.677 agreement with expert consensus, surpassing MNE ICLabel, while also preserving clinically relevant brain signals in ambiguous cases. Over 97% of its outputs were rated as interpretable and actionable by expert reviewers. As a core module of the open-source EEG Autoclean platform, ICVision signals a paradigm shift in scientific AI, where models do not just classify, but see, reason, and communicate. It opens the door to globally scalable, explainable, and reproducible EEG workflows, marking the emergence of AI agents capable of expert-level visual decision-making in brain science and beyond.

Paperid: 6400, https://arxiv.org/pdf/2511.23075.pdf

Abstract:
Large vision-language models (VLMs) show strong multimodal understanding but still struggle with 3D spatial reasoning, such as distance estimation, size comparison, and cross-view consistency. Existing 3D-aware methods either depend on auxiliary 3D information or enhance RGB-only VLMs with geometry encoders through shallow feature fusion. We propose SpaceMind, a multimodal large language model explicitly designed for spatial reasoning solely from RGB inputs. The model adopts a dual-encoder architecture, integrating VGGT as a spatial understanding encoder and InternViT as a 2D visual encoder. The key idea is to treat the camera representation as an active guiding modality rather than passive metadata. Specifically, SpaceMind introduces a lightweight Camera-Guided Modality Fusion module before the language model to replace shallow fusion. It applies camera-conditioned biasing to spatial tokens, assigns query-independent weights reflecting their geometric importance, and uses the camera embedding to gate the fused representation. Empirically, SpaceMind establishes new state-of-the-art results on VSI-Bench, SQA3D and SPBench, surpassing both open and proprietary systems on VSI-Bench and SPBench by large margins and achieving state-of-the-art performance on SQA3D. These results demonstrate that camera-guided modality fusion is an effective and practical inductive bias for equipping VLMs with genuinely spatially grounded intelligence. We will release code and model checkpoints to support future research.

Paperid: 6401, https://arxiv.org/pdf/2511.22903.pdf

Abstract:
Change captioning aims to describe changes between a pair of images. However, existing works rely on visual features alone, which often fail to capture subtle but meaningful changes because they lack the ability to represent explicitly structured information such as object relationships and compositional semantics. To alleviate this, we present CORTEX (COmpositional Reasoning-aware TEXt-guided), a novel framework that integrates complementary textual cues to enhance change understanding. In addition to capturing cues from pixel-level differences, CORTEX utilizes scene-level textual knowledge provided by Vision Language Models (VLMs) to extract richer image text signals that reveal underlying compositional reasoning. CORTEX consists of three key modules: (i) an Image-level Change Detector that identifies low-level visual differences between paired images, (ii) a Reasoning-aware Text Extraction (RTE) module that use VLMs to generate compositional reasoning descriptions implicit in visual features, and (iii) an Image-Text Dual Alignment (ITDA) module that aligns visual and textual features for fine-grained relational reasoning. This enables CORTEX to reason over visual and textual features and capture changes that are otherwise ambiguous in visual features alone.

Paperid: 6402, https://arxiv.org/pdf/2511.22436.pdf

Abstract:
Few-shot multi-class industrial anomaly detection remains a challenging task. Vision-language models need to be both category-adaptive and sharply discriminative, yet data scarcity often blurs the boundary between normal and abnormal states. This ambiguity leads to missed subtle defects and the rejection of atypical normal samples. We propose ABounD, an Adversarial Boundary-Driven few-shot learning for multi-class anomaly detection, which is a unified learning framework that integrates semantic concept learning with decision boundary shaping. The Dynamic Concept Fusion (DCF) module produces class-adaptive prompts by fusing generalizable priors with class-specific cues, conditioned on image features. Meanwhile, Adversarial Boundary Forging (ABF) sculpts a more precise decision margin by generating boundary-level fence features via PGD-style perturbations. Training is conducted in a single stage under a Concept-Boundary Loss, where ABF provides the main supervisory signal and semantic-spatial regularizers stabilize the optimization. This synergy yields a decision boundary that closely follows normal data while preserving flexibility and robust semantic alignment. Experiments on MVTec-AD and VisA datasets demonstrate state-of-the-art performance in the task of few-shot multi-class anomaly detection.

Paperid: 6403, https://arxiv.org/pdf/2511.22141.pdf

Abstract:
Advances in vision-language models (VLMs) have enabled effective cross-modality retrieval. However, when both text and images exist in the database, similarity scores would differ in scale by modality. This phenomenon, known as the modality gap, hinders accurate retrieval. Most existing studies address this issue with manually labeled data, e.g., by fine-tuning VLMs on them. In this work, we propose a similarity standardization approach with pseudo data construction. We first compute the mean and variance of the similarity scores between each query and its paired data in text or image modality. Using these modality-specific statistics, we standardize all similarity scores to compare on a common scale across modalities. These statistics are calculated from pseudo pairs, which are constructed by retrieving the text and image candidates with the highest cosine similarity to each query. We evaluate our method across seven VLMs using two multi-modal QA benchmarks (MMQA and WebQA), where each question requires retrieving either text or image data. Our experimental results show that our method significantly improves retrieval performance, achieving average Recall@20 gains of 64% on MMQA and 28% on WebQA when the query and the target data belong to different modalities. Compared to E5-V, which addresses the modality gap through image captioning, we confirm that our method more effectively bridges the modality gap.

Paperid: 6404, https://arxiv.org/pdf/2511.19982.pdf

Abstract:
Continuous emotional image generation (C-EICG) is emerging rapidly due to its ability to produce images aligned with both user descriptions and continuous emotional values. However, existing approaches lack emotional feedback from generated images, limiting the control of emotional continuity. Additionally, their simple alignment between emotions and naively generated texts fails to adaptively adjust emotional prompts according to image content, leading to insufficient emotional fidelity. To address these concerns, we propose a novel generation-understanding-feedback reinforcement paradigm (EmoFeedback$^2$) for C-EICG, which exploits the reasoning capability of the fine-tuned large vision-language model (LVLM) to provide reward and textual feedback for generating high-quality images with continuous emotions. Specifically, we introduce an emotion-aware reward feedback strategy, where the LVLM evaluates the emotional values of generated images and computes the reward against target emotions, guiding the reinforcement fine-tuning of the generative model and enhancing the emotional continuity of images. Furthermore, we design a self-promotion textual feedback framework, in which the LVLM iteratively analyzes the emotional content of generated images and adaptively produces refinement suggestions for the next-round prompt, improving the emotional fidelity with fine-grained content. Extensive experimental results demonstrate that our approach effectively generates high-quality images with the desired emotions, outperforming existing state-of-the-art methods in our custom dataset. The code and dataset will be released soon.

Paperid: 6405, https://arxiv.org/pdf/2511.19820.pdf

Abstract:
Vision-Language Models (VLMs) often struggle with tasks that require fine-grained image understanding, such as scene-text recognition or document analysis, due to perception limitations and visual fragmentation. To address these challenges, we introduce CropVLM as an external low-cost method for boosting performance, enabling VLMs to dynamically ''zoom in'' on relevant image regions, enhancing their ability to capture fine details. CropVLM is trained using reinforcement learning, without using human-labeled bounding boxes as a supervision signal, and without expensive synthetic evaluations. The model is trained once and can be paired with both open-source and proprietary VLMs to improve their performance. Our approach delivers significant improvements on tasks that require high-resolution image understanding, notably for benchmarks that are out-of-domain for the target VLM, without modifying or fine-tuning the VLM, thus avoiding catastrophic forgetting.

Paperid: 6406, https://arxiv.org/pdf/2511.19200.pdf

Abstract:
Recent advances in computer vision have yielded models with strong performance on recognition benchmarks; however, significant gaps remain in comparison to human perception. One subtle ability is to judge whether an image looks like a given object without being an instance of that object. We study whether vision-language models such as CLIP capture this distinction. We curated a dataset named RoLA (Real or Lookalike) of real and lookalike exemplars (e.g., toys, statues, drawings, pareidolia) across multiple categories, and first evaluate a prompt-based baseline with paired "real"/"lookalike" prompts. We then estimate a direction in CLIP's embedding space that moves representations between real and lookalike. Applying this direction to image and text embeddings improves discrimination in cross-modal retrieval on Conceptual12M, and also enhances captions produced by a CLIP prefix captioner.

Paperid: 6407, https://arxiv.org/pdf/2511.18824.pdf

Abstract:
Figuring out which objects or concepts words refer to is a central language learning challenge for young children. Most models of this process posit that children learn early object labels from co-occurrences of words and their referents that occur when someone around them talks about an object in the immediate physical environment. But how aligned in time are children's visual and linguistic experiences during everyday learning? To date, answers to this question have been limited by the need for labor-intensive manual annotations of vision-language co-occurrences. Here, we evaluate the use of contrastive language-image pretraining (CLIP) models to automatically characterize vision-language alignment in egocentric videos taken from the infant perspective in home environments. After validating CLIP alignment scores using human alignment judgments, we apply this metric to a large corpus of infant-perspective videos. We show that idealized aligned moments for learning (e.g., "look at the ball" with a ball present in the child's view) are relatively rare in children's everyday experiences compared to modern machine learning datasets, and highlight variability in alignment both within and across children. These findings suggest that infrequent alignment is a constraint for models describing early word learning and offer a new method for investigating children's multimodal environment.

Paperid: 6408, https://arxiv.org/pdf/2511.16454.pdf

Abstract:
Developing a multi-modal language model capable of understanding 3D scenes remains challenging due to the limited availability of 3D training data, in contrast to the abundance of 2D datasets used for vision-language models (VLM). As an alternative, we introduce LLaVA$^3$ (pronounced LLaVA-Cube), a novel method that improves the 3D scene understanding capabilities of VLM using only multi-view 2D images and without any fine-tuning. Inspired by Cubist painters, who represented multiple viewpoints of a 3D object within a single picture, we propose to describe the 3D scene for the VLM through omnidirectional visual representations of each object. These representations are derived from an intermediate multi-view 3D reconstruction of the scene. Extensive experiments on 3D VQA and 3D language grounding show that our approach outperforms previous 2D-based VLM solutions.

Paperid: 6409, https://arxiv.org/pdf/2511.16347.pdf

Abstract:
The adoption of Vision-Language Models (VLMs) in embodied AI agents, while being effective, brings safety concerns such as jailbreaking. Prior work have explored the possibility of directly jailbreaking the embodied agents through elaborated multi-modal prompts. However, no prior work has studied or even reported indirect jailbreaks in embodied AI, where a black-box attacker induces a jailbreak without issuing direct prompts to the embodied agent. In this paper, we propose, for the first time, indirect environmental jailbreak (IEJ), a novel attack to jailbreak embodied AI via indirect prompt injected into the environment, such as malicious instructions written on a wall. Our key insight is that embodied AI does not ''think twice'' about the instructions provided by the environment -- a blind trust that attackers can exploit to jailbreak the embodied agent. We further design and implement open-source prototypes of two fully-automated frameworks: SHAWSHANK, the first automatic attack generation framework for the proposed attack IEJ; and SHAWSHANK-FORGE, the first automatic benchmark generation framework for IEJ. Then, using SHAWSHANK-FORGE, we automatically construct SHAWSHANK-BENCH, the first benchmark for indirectly jailbreaking embodied agents. Together, our two frameworks and one benchmark answer the questions of what content can be used for malicious IEJ instructions, where they should be placed, and how IEJ can be systematically evaluated. Evaluation results show that SHAWSHANK outperforms eleven existing methods across 3,957 task-scene combinations and compromises all six tested VLMs. Furthermore, current defenses only partially mitigate our attack, and we have responsibly disclosed our findings to all affected VLM vendors.

Paperid: 6410, https://arxiv.org/pdf/2511.15633.pdf

Abstract:
Class-Incremental Learning (CIL) enables models to learn new classes continually while preserving past knowledge. Recently, vision-language models like CLIP offer transferable features via multi-modal pre-training, making them well-suited for CIL. However, real-world visual and linguistic concepts are inherently hierarchical: a textual concept like "dog" subsumes fine-grained categories such as "Labrador" and "Golden Retriever," and each category entails its images. But existing CLIP-based CIL methods fail to explicitly capture this inherent hierarchy, leading to fine-grained class features drift during incremental updates and ultimately to catastrophic forgetting. To address this challenge, we propose HASTEN (Hierarchical Semantic Tree Anchoring) that anchors hierarchical information into CIL to reduce catastrophic forgetting. First, we employ an external knowledge graph as supervision to embed visual and textual features in hyperbolic space, effectively preserving hierarchical structure as data evolves. Second, to mitigate catastrophic forgetting, we project gradients onto the null space of the shared hyperbolic mapper, preventing interference with prior tasks. These two steps work synergistically to enable the model to resist forgetting by maintaining hierarchical relationships. Extensive experiments show that HASTEN consistently outperforms existing methods while providing a unified structured representation.

Paperid: 6411, https://arxiv.org/pdf/2511.15464.pdf

Abstract:
Recent advances in computational pathology have leveraged vision-language models to learn joint representations of Hematoxylin and Eosin (HE) images with spatial transcriptomic (ST) profiles. However, existing approaches typically align HE tiles with their corresponding ST profiles at a single scale, overlooking fine-grained cellular structures and their spatial organization. To address this, we propose Sigmma, a multi-modal contrastive alignment framework for learning hierarchical representations of HE images and spatial transcriptome profiles across multiple scales. Sigmma introduces multi-scale contrastive alignment, ensuring that representations learned at different scales remain coherent across modalities. Furthermore, by representing cell interactions as a graph and integrating inter- and intra-subgraph relationships, our approach effectively captures cell-cell interactions, ranging from fine to coarse, within the tissue microenvironment. We demonstrate that Sigmm learns representations that better capture cross-modal correspondences, leading to an improvement of avg. 9.78\% in the gene-expression prediction task and avg. 26.93\% in the cross-modal retrieval task across datasets. We further show that it learns meaningful multi-tissue organization in downstream analyses.

Paperid: 6412, https://arxiv.org/pdf/2511.13881.pdf

Abstract:
Recent advancements in autonomous driving (AD) have explored the use of vision-language models (VLMs) within visual question answering (VQA) frameworks for direct driving decision-making. However, these approaches often depend on handcrafted prompts and suffer from inconsistent performance, limiting their robustness and generalization in real-world scenarios. In this work, we evaluate state-of-the-art open-source VLMs on high-level decision-making tasks using ego-view visual inputs and identify critical limitations in their ability to deliver reliable, context-aware decisions. Motivated by these observations, we propose a new approach that shifts the role of VLMs from direct decision generators to semantic enhancers. Specifically, we leverage their strong general scene understanding to enrich existing vision-based benchmarks with structured, linguistically rich scene descriptions. Building on this enriched representation, we introduce a multi-modal interactive architecture that fuses visual and linguistic features for more accurate decision-making and interpretable textual explanations. Furthermore, we design a post-hoc refinement module that utilizes VLMs to enhance prediction reliability. Extensive experiments on two autonomous driving benchmarks demonstrate that our approach achieves state-of-the-art performance, offering a promising direction for integrating VLMs into reliable and interpretable AD systems.

Paperid: 6413, https://arxiv.org/pdf/2511.13135.pdf

Abstract:
As Vision-Language Models (VLMs) increasingly gain traction in medical applications, clinicians are progressively expecting AI systems not only to generate textual diagnoses but also to produce corresponding medical images that integrate seamlessly into authentic clinical workflows. Despite the growing interest, existing medical visual benchmarks present notable limitations. They often rely on ambiguous queries that lack sufficient relevance to image content, oversimplify complex diagnostic reasoning into closed-ended shortcuts, and adopt a text-centric evaluation paradigm that overlooks the importance of image generation capabilities. To address these challenges, we introduce \textsc{MedGEN-Bench}, a comprehensive multimodal benchmark designed to advance medical AI research. MedGEN-Bench comprises 6,422 expert-validated image-text pairs spanning six imaging modalities, 16 clinical tasks, and 28 subtasks. It is structured into three distinct formats: Visual Question Answering, Image Editing, and Contextual Multimodal Generation. What sets MedGEN-Bench apart is its focus on contextually intertwined instructions that necessitate sophisticated cross-modal reasoning and open-ended generative outputs, moving beyond the constraints of multiple-choice formats. To evaluate the performance of existing systems, we employ a novel three-tier assessment framework that integrates pixel-level metrics, semantic text analysis, and expert-guided clinical relevance scoring. Using this framework, we systematically assess 10 compositional frameworks, 3 unified models, and 5 VLMs.

Paperid: 6414, https://arxiv.org/pdf/2511.11262.pdf

Abstract:
Fine-grained knowledge is crucial for vision-language models to obtain a better understanding of the real world. While there has been work trying to acquire this kind of knowledge in the space of vision and language, it has mostly focused on aligning the image patches with the tokens on the language side. However, image patches do not have any meaning to the human eye, and individual tokens do not necessarily carry groundable information in the image. It is groups of tokens which describe different aspects of the scene. In this work, we propose a model which groups the caption tokens as part of its architecture in order to capture a fine-grained representation of the language. We expect our representations to be at the level of objects present in the image, and therefore align our representations with the output of an image encoder trained to discover objects. We show that by learning to group the tokens, the vision-language model has a better fine-grained understanding of vision and language. In addition, the token groups that our model discovers are highly similar to groundable phrases in text, both qualitatively and quantitatively.

Paperid: 6415, https://arxiv.org/pdf/2511.11165.pdf

Abstract:
Most explainable anomaly detection methods often identify anomalies but lack the capability to differentiate the type of anomaly. Furthermore, they often require the costly training and maintenance of separate models for each object category. The lack of specificity is a significant research gap, as identifying the type of anomaly (e.g., "Crack" vs. "Scratch") is crucial for accurate diagnosis that facilitates cost-saving operational decisions across diverse application domains. While some recent large-scale Vision-Language Models (VLMs) have begun to address this, they are computationally intensive and memory-heavy, restricting their use in real-time or embedded systems. We propose MultiTypeFCDD, a simple and lightweight convolutional framework designed as a practical alternative for explainable multi-type anomaly detection. MultiTypeFCDD uses only image-level labels to learn and produce multi-channel heatmaps, where each channel is trained to correspond to a specific anomaly type. The model functions as a single, unified framework capable of differentiating anomaly types across multiple object categories, eliminating the need to train and manage separate models for each object category. We evaluated our proposed method on the Real-IAD dataset and it delivers results competitive with state-of-the-art complex models at significantly reduced parametric load and inference times. This makes it a highly practical and viable solution for real-world applications where computational resources are tightly constrained.

Paperid: 6416, https://arxiv.org/pdf/2511.11034.pdf

Abstract:
Recent advances in multimodal large language models have enabled unified processing of visual and textual inputs, offering promising applications in general-purpose medical AI. However, their ability to generalize compositionally across unseen combinations of imaging modality, anatomy, and task type remains underexplored. We introduce CrossMed, a benchmark designed to evaluate compositional generalization (CG) in medical multimodal LLMs using a structured Modality-Anatomy-Task (MAT) schema. CrossMed reformulates four public datasets, CheXpert (X-ray classification), SIIM-ACR (X-ray segmentation), BraTS 2020 (MRI classification and segmentation), and MosMedData (CT classification) into a unified visual question answering (VQA) format, resulting in 20,200 multiple-choice QA instances. We evaluate two open-source multimodal LLMs, LLaVA-Vicuna-7B and Qwen2-VL-7B, on both Related and Unrelated MAT splits, as well as a zero-overlap setting where test triplets share no Modality, Anatomy, or Task with the training data. Models trained on Related splits achieve 83.2 percent classification accuracy and 0.75 segmentation cIoU, while performance drops significantly under Unrelated and zero-overlap conditions, demonstrating the benchmark difficulty. We also show cross-task transfer, where segmentation performance improves by 7 percent cIoU even when trained using classification-only data. Traditional models (ResNet-50 and U-Net) show modest gains, confirming the broad utility of the MAT framework, while multimodal LLMs uniquely excel at compositional generalization. CrossMed provides a rigorous testbed for evaluating zero-shot, cross-task, and modality-agnostic generalization in medical vision-language models.

Paperid: 6417, https://arxiv.org/pdf/2511.10615.pdf

Abstract:
Large Vision-Language Models (VLMs) excel at understanding and generating video descriptions but their high memory, computation, and deployment demands hinder practical use particularly for blind and low-vision (BLV) users who depend on detailed, context-aware descriptions. To study the effect of model size on accessibility-focused description quality, we evaluate SmolVLM2 variants with 500M and 2.2B parameters across two diverse datasets: AVCaps (outdoor), and Charades (indoor). In this work, we introduce two novel evaluation frameworks specifically designed for BLV accessibility assessment: the Multi-Context BLV Framework evaluating spatial orientation, social interaction, action events, and ambience contexts; and the Navigational Assistance Framework focusing on mobility-critical information. Additionally, we conduct a systematic evaluation of four different prompt design strategies and deploy both models on a smartphone, evaluating FP32 and INT8 precision variants to assess real-world performance constraints on resource-limited mobile devices.

Paperid: 6418, https://arxiv.org/pdf/2511.09955.pdf

Abstract:
Foundation models, especially vision-language models (VLMs), offer compelling zero-shot object detection for applications like autonomous driving, a domain where manual labelling is prohibitively expensive. However, their detection latency and tendency to hallucinate predictions render them unsuitable for direct deployment. This work introduces a novel pipeline that addresses this challenge by leveraging VLMs to automatically generate pseudo-labels for training efficient, real-time object detectors. Our key innovation is a per-object co-teaching-based training strategy that mitigates the inherent noise in VLM-generated labels. The proposed per-object coteaching approach filters noisy bounding boxes from training instead of filtering the entire image. Specifically, two YOLO models learn collaboratively, filtering out unreliable boxes from each mini-batch based on their peers' per-object loss values. Overall, our pipeline provides an efficient, robust, and scalable approach to train high-performance object detectors for autonomous driving, significantly reducing reliance on costly human annotation. Experimental results on the KITTI dataset demonstrate that our method outperforms a baseline YOLOv5m model, achieving a significant mAP@0.5 boost ($31.12\%$ to $46.61\%$) while maintaining real-time detection latency. Furthermore, we show that supplementing our pseudo-labelled data with a small fraction of ground truth labels ($10\%$) leads to further performance gains, reaching $57.97\%$ mAP@0.5 on the KITTI dataset. We observe similar performance improvements for the ACDC and BDD100k datasets.

Paperid: 6419, https://arxiv.org/pdf/2511.09339.pdf

Abstract:
Contemporary vision-language models (VLMs) perform well on existing multimodal reasoning benchmarks (78-85\% accuracy on MMMU, MathVista). Yet, these results fail to sufficiently distinguish true scientific reasoning articulation capabilities from pattern-matching. To address this gap, we introduce \textbf{mmJEE-Eval}, a multimodal bilingual (English and Hindi) benchmark comprising 1,460 questions from India's JEE Advanced examination (2019-2025) spanning pre-college Physics, Chemistry, and Mathematics domains. Our evaluation of 17 state-of-the-art models reveals that while frontier VLMs (GPT-5, Gemini 2.5 Pro/Flash) achieve 77-84\% accuracy on held-out 2025 questions, open-source models plateau at 37-45\% despite scaling to 400B parameters, a significant difference not observed on existing benchmarks. While closed frontiers from Google and OpenAI show high problem-solving accuracies (up to 100\% pass@3 scores), they fully collapse when the reasoning load is increased meta-cognitively (GPT-5 fixes just 5.2\% errors). Systematic ablations show mmJEE-Eval's difficulty stems from complexity and reasoning depth rather than memorization. Effectively, our benchmark segregates superior training and reasoning methodologies where alternatives fail. We publicly release our code and data: https://mmjee-eval.github.io

Paperid: 6420, https://arxiv.org/pdf/2511.07710.pdf

Abstract:
Fine-grained image-text alignment is a pivotal challenge in multimodal learning, underpinning key applications such as visual question answering, image captioning, and vision-language navigation. Unlike global alignment, fine-grained alignment requires precise correspondence between localized visual regions and textual tokens, often hindered by noisy attention mechanisms and oversimplified modeling of cross-modal relationships. In this work, we identify two fundamental limitations of existing approaches: the lack of robust intra-modal mechanisms to assess the significance of visual and textual tokens, leading to poor generalization in complex scenes; and the absence of fine-grained uncertainty modeling, which fails to capture the one-to-many and many-to-one nature of region-word correspondences. To address these issues, we propose a unified approach that incorporates significance-aware and granularity-aware modeling and region-level uncertainty modeling. Our method leverages modality-specific biases to identify salient features without relying on brittle cross-modal attention, and represents region features as a mixture of Gaussian distributions to capture fine-grained uncertainty. Extensive experiments on Flickr30K and MS-COCO demonstrate that our approach achieves state-of-the-art performance across various backbone architectures, significantly enhancing the robustness and interpretability of fine-grained image-text alignment.

Paperid: 6421, https://arxiv.org/pdf/2511.06348.pdf

Abstract:
Gaze understanding unifies the detection of people, their gaze targets, and objects of interest into a single framework, offering critical insight into visual attention and intent estimation. Although prior research has modelled gaze cues in visual scenes, a unified system is still needed for gaze understanding using both visual and language prompts. This paper introduces GazeVLM, a novel Vision-Language Model (VLM) for multi-task gaze understanding in images, addressing person detection, gaze target detection, and gaze object identification. While other transformer-based methods exist for gaze analysis, GazeVLM represents, to our knowledge, the first application of a VLM to these combined tasks, allowing for selective execution of each task. Through the integration of visual (RGB and depth) and textual modalities, our ablation study on visual input combinations revealed that a fusion of RGB images with HHA-encoded depth maps, guided by text prompts, yields superior performance. We also introduce an object-level gaze detection metric for gaze object identification ($AP_{ob}$). Through experiments, GazeVLM demonstrates significant improvements, notably achieving state-of-the-art evaluation scores on GazeFollow and VideoAttentionTarget datasets.

Paperid: 6422, https://arxiv.org/pdf/2511.05680.pdf

Abstract:
This paper presents a robotic assembly framework that combines Vision-Language Models (VLMs) with imitation learning for assembly manipulation tasks. Our system employs a gripper-equipped robot that moves in 3D space to perform assembly operations. The framework integrates visual perception, natural language understanding, and learned primitive skills to enable flexible and adaptive robotic manipulation. Experimental results demonstrate the effectiveness of our approach in assembly scenarios, achieving high success rates while maintaining interpretability through the structured primitive skill decomposition.

Paperid: 6423, https://arxiv.org/pdf/2511.03908.pdf

Abstract:
Iterated reference games - in which players repeatedly pick out novel referents using language - present a test case for agents' ability to perform context-sensitive pragmatic reasoning in multi-turn linguistic environments. We tested humans and vision-language models on trials from iterated reference games, varying the given context in terms of amount, order, and relevance. Without relevant context, models were above chance but substantially worse than humans. However, with relevant context, model performance increased dramatically over trials. Few-shot reference games with abstract referents remain a difficult task for machine learning models.

Paperid: 6424, https://arxiv.org/pdf/2511.02503.pdf

Abstract:
The automation of workflows in advanced microscopy is a key goal where foundation models like Language Models (LLMs) and Vision-Language Models (VLMs) show great potential. However, adapting these general-purpose models for specialized scientific tasks is critical, and the optimal domain adaptation strategy is often unclear. To address this, we introduce PtychoBench, a new multi-modal, multi-task benchmark for ptychographic analysis. Using this benchmark, we systematically compare two specialization strategies: Supervised Fine-Tuning (SFT) and In-Context Learning (ICL). We evaluate these strategies on a visual artifact detection task with VLMs and a textual parameter recommendation task with LLMs in a data-scarce regime. Our findings reveal that the optimal specialization pathway is task-dependent. For the visual task, SFT and ICL are highly complementary, with a fine-tuned model guided by context-aware examples achieving the highest mean performance (Micro-F1 of 0.728). Conversely, for the textual task, ICL on a large base model is the superior strategy, reaching a peak Micro-F1 of 0.847 and outperforming a powerful "super-expert" SFT model (0-shot Micro-F1 of 0.839). We also confirm the superiority of context-aware prompting and identify a consistent contextual interference phenomenon in fine-tuned models. These results, benchmarked against strong baselines including GPT-4o and a DINOv3-based classifier, offer key observations for AI in science: the optimal specialization path in our benchmark is dependent on the task modality, offering a clear framework for developing more effective science-based agentic systems.

Paperid: 6425, https://arxiv.org/pdf/2511.02162.pdf

Abstract:
Advances in 3D generative AI have enabled the creation of physical objects from text prompts, but challenges remain in creating objects involving multiple component types. We present a pipeline that integrates 3D generative AI with vision-language models (VLMs) to enable the robotic assembly of multi-component objects from natural language. Our method leverages VLMs for zero-shot, multi-modal reasoning about geometry and functionality to decompose AI-generated meshes into multi-component 3D models using predefined structural and panel components. We demonstrate that a VLM is capable of determining which mesh regions need panel components in addition to structural components, based on the object's geometry and functionality. Evaluation across test objects shows that users preferred the VLM-generated assignments 90.6% of the time, compared to 59.4% for rule-based and 2.5% for random assignment. Lastly, the system allows users to refine component assignments through conversational feedback, enabling greater human control and agency in making physical objects with generative AI and robotics.

Paperid: 6426, https://arxiv.org/pdf/2511.02113.pdf

Abstract:
Recent advances in multimodal recommendation (MMR) highlight the potential of integrating visual and textual content to enrich item representations. However, existing methods often rely on coarse visual features and naive fusion strategies, resulting in redundant or misaligned representations. From an information-theoretic perspective, effective fusion should balance unique, shared, and redundant modality information to preserve complementary cues. To this end, we propose VIRAL, a novel Vision-Language and Information-aware Recommendation framework that enhances multimodal fusion through two components: (i) a VLM-based visual enrichment module that generates fine-grained, title-guided descriptions for semantically aligned image representations, and (ii) an information-aware fusion module inspired by Partial Information Decomposition (PID) to disentangle and integrate complementary signals. Experiments on three Amazon datasets show that VIRAL consistently outperforms strong multimodal baselines and substantially improves the contribution of visual features.

Paperid: 6427, https://arxiv.org/pdf/2511.01831.pdf

Abstract:
Vision-Language Models (VLMs) suffer from catastrophic forgetting when sequentially fine-tuned on new tasks, degrading performance on previously learned foundational and task-specific capabilities. While multi-task learning can mitigate forgetting, it requires simultaneous access to all datasets and imposes computational overhead that scales linearly with the number of tasks. In this work, we introduce a routing-based approach that enables the integration of new tasks while preserving the foundational knowledge acquired during pretraining. We evaluate our method using InternVL-2 models (2B and 8B parameters) and demonstrate that routing preserves the model's foundational capabilities by maintaining performance on general-purpose benchmarks such as ChartQA, MMBench, and DocVQA, while simultaneously improving accuracy on specialized tasks. Importantly, our approach achieves this without requiring concurrent access to data from all tasks, avoiding the significant computational and data overhead associated with traditional multi-task learning. We further conduct extensive ablation studies to evaluate the scalability and robustness of routing-based learning, showing that the approach is resilient to a growing number of tasks and performs particularly well when new tasks are semantically related. Finally, we show that the routing mechanism enables superior cross-modal transfer between language and vision capabilities, allowing knowledge learned in one modality to enhance performance in another capability not achieved by existing continual learning methods.

Paperid: 6428, https://arxiv.org/pdf/2511.00261.pdf

Abstract:
Humans excel at visual social inference, the ability to infer hidden elements of a scene from subtle behavioral cues such as other people's gaze, pose, and orientation. This ability drives everyday social reasoning in humans and is critical for developing more human-like AI agents. We introduce Spot The Ball, a challenging benchmark for evaluating visual social inference in vision-language models (VLMs) using sports as a test domain. The task is to localize a removed sports ball from soccer, basketball, and volleyball images. We present a curated evaluation set with human baselines and a scalable pipeline for generating additional test items. We evaluate four state-of-the-art VLMs (Gemini, GPT, LLaMA, Qwen) using three prompting strategies, finding that humans are consistently two to three times more accurate (20-34%) than models ($\leq$ 17%) across all sports. Our analyses show that models rely on superficial spatial heuristics--such as guessing near the image center or nearby players--while humans leverage social cues like gaze direction and body pose. These findings reveal a persistent human-model gap in visual social reasoning and underscore the need for architectures that explicitly encode structured behavioral cues to achieve robust, human-like inference.

Paperid: 6429, https://arxiv.org/pdf/2511.00110.pdf

Abstract:
We propose a novel cognitively-inspired method to improve and interpret physical simulation in vision-language models. Our ``Chain of Time" method involves generating a series of intermediate images during a simulation, and it is motivated by in-context reasoning in machine learning, as well as mental simulation in humans. Chain of Time is used at inference time, and requires no additional fine-tuning. We apply the Chain-of-Time method to synthetic and real-world domains, including 2-D graphics simulations and natural 3-D videos. These domains test a variety of particular physical properties, including velocity, acceleration, fluid dynamics, and conservation of momentum. We found that using Chain-of-Time simulation substantially improves the performance of a state-of-the-art image generation model. Beyond examining performance, we also analyzed the specific states of the world simulated by an image model at each time step, which sheds light on the dynamics underlying these simulations. This analysis reveals insights that are hidden from traditional evaluations of physical reasoning, including cases where an image generation model is able to simulate physical properties that unfold over time, such as velocity, gravity, and collisions. Our analysis also highlights particular cases where the image generation model struggles to infer particular physical parameters from input images, despite being capable of simulating relevant physical processes.

Paperid: 6430, https://arxiv.org/pdf/2510.27255.pdf

Abstract:
Vision-Language Models (VLMs) have demonstrated impressive capabilities in zero-shot action recognition by learning to associate video embeddings with class embeddings. However, a significant challenge arises when relying solely on action classes to provide semantic context, particularly due to the presence of multi-semantic words, which can introduce ambiguity in understanding the intended concepts of actions. To address this issue, we propose an innovative approach that harnesses web-crawled descriptions, leveraging a large-language model to extract relevant keywords. This method reduces the need for human annotators and eliminates the laborious manual process of attribute data creation. Additionally, we introduce a spatio-temporal interaction module designed to focus on objects and action units, facilitating alignment between description attributes and video content. In our zero-shot experiments, our model achieves impressive results, attaining accuracies of 81.0%, 53.1%, and 68.9% on UCF-101, HMDB-51, and Kinetics-600, respectively, underscoring the model's adaptability and effectiveness across various downstream tasks.

Paperid: 6431, https://arxiv.org/pdf/2510.26721.pdf

Abstract:
Multimodal large language models (MLLMs) exhibit a pronounced preference for textual inputs when processing vision-language data, limiting their ability to reason effectively from visual evidence. Unlike prior studies that attribute this text bias to external factors such as data imbalance or instruction tuning, we propose that the bias originates from the model's internal architecture. Specifically, we hypothesize that visual key vectors (Visual Keys) are out-of-distribution (OOD) relative to the text key space learned during language-only pretraining. Consequently, these visual keys receive systematically lower similarity scores during attention computation, leading to their under-utilization in the context representation. To validate this hypothesis, we extract key vectors from LLaVA and Qwen2.5-VL and analyze their distributional structures using qualitative (t-SNE) and quantitative (Jensen-Shannon divergence) methods. The results provide direct evidence that visual and textual keys occupy markedly distinct subspaces within the attention space. The inter-modal divergence is statistically significant, exceeding intra-modal variation by several orders of magnitude. These findings reveal that text bias arises from an intrinsic misalignment within the attention key space rather than solely from external data factors.

Paperid: 6432, https://arxiv.org/pdf/2510.26580.pdf

Abstract:
In real-world environments, AI systems often face unfamiliar scenarios without labeled data, creating a major challenge for conventional scene understanding models. The inability to generalize across unseen contexts limits the deployment of vision-based applications in dynamic, unstructured settings. This work introduces a Dynamic Context-Aware Scene Reasoning framework that leverages Vision-Language Alignment to address zero-shot real-world scenarios. The goal is to enable intelligent systems to infer and adapt to new environments without prior task-specific training. The proposed approach integrates pre-trained vision transformers and large language models to align visual semantics with natural language descriptions, enhancing contextual comprehension. A dynamic reasoning module refines predictions by combining global scene cues and object-level interactions guided by linguistic priors. Extensive experiments on zero-shot benchmarks such as COCO, Visual Genome, and Open Images demonstrate up to 18% improvement in scene understanding accuracy over baseline models in complex and unseen environments. Results also show robust performance in ambiguous or cluttered scenes due to the synergistic fusion of vision and language. This framework offers a scalable and interpretable approach for context-aware reasoning, advancing zero-shot generalization in dynamic real-world settings.

Paperid: 6433, https://arxiv.org/pdf/2510.25164.pdf

Abstract:
We present a transformer-based multimodal framework for generating clinically relevant captions for MRI scans. Our system combines a DEiT-Small vision transformer as an image encoder, MediCareBERT for caption embedding, and a custom LSTM-based decoder. The architecture is designed to semantically align image and textual embeddings, using hybrid cosine-MSE loss and contrastive inference via vector similarity. We benchmark our method on the MultiCaRe dataset, comparing performance on filtered brain-only MRIs versus general MRI images against state-of-the-art medical image captioning methods including BLIP, R2GenGPT, and recent transformer-based approaches. Results show that focusing on domain-specific data improves caption accuracy and semantic alignment. Our work proposes a scalable, interpretable solution for automated medical image reporting.

Paperid: 6434, https://arxiv.org/pdf/2510.25070.pdf

Abstract:
Zero-shot scene understanding in real-world settings presents major challenges due to the complexity and variability of natural scenes, where models must recognize new objects, actions, and contexts without prior labeled examples. This work proposes a vision-language integration framework that unifies pre-trained visual encoders (e.g., CLIP, ViT) and large language models (e.g., GPT-based architectures) to achieve semantic alignment between visual and textual modalities. The goal is to enable robust zero-shot comprehension of scenes by leveraging natural language as a bridge to generalize over unseen categories and contexts. Our approach develops a unified model that embeds visual inputs and textual prompts into a shared space, followed by multimodal fusion and reasoning layers for contextual interpretation. Experiments on Visual Genome, COCO, ADE20K, and custom real-world datasets demonstrate significant gains over state-of-the-art zero-shot models in object recognition, activity detection, and scene captioning. The proposed system achieves up to 18% improvement in top-1 accuracy and notable gains in semantic coherence metrics, highlighting the effectiveness of cross-modal alignment and language grounding in enhancing generalization for real-world scene understanding.

Paperid: 6435, https://arxiv.org/pdf/2510.25032.pdf

Abstract:
Developing a highly accurate automatic license plate recognition system (ALPR) is challenging due to environmental factors such as lighting, rain, and dust. Additional difficulties include high vehicle speeds, varying camera angles, and low-quality or low-resolution images. ALPR is vital in traffic control, parking, vehicle tracking, toll collection, and law enforcement applications. This paper proposes a deep learning strategy using YOLOv8 for license plate detection and recognition tasks. This method seeks to enhance the performance of the model using datasets from Ontario, Quebec, California, and New York State. It achieved an impressive recall rate of 94% on the dataset from the Center for Pattern Recognition and Machine Intelligence (CENPARMI) and 91% on the UFPR-ALPR dataset. In addition, our method follows a semi-supervised learning framework, combining a small set of manually labeled data with pseudo-labels generated by Grounding DINO to train our detection model. Grounding DINO, a powerful vision-language model, automatically annotates many images with bounding boxes for license plates, thereby minimizing the reliance on labor-intensive manual labeling. By integrating human-verified and model-generated annotations, we can scale our dataset efficiently while maintaining label quality, which significantly enhances the training process and overall model performance. Furthermore, it reports character error rates for both datasets, providing additional insight into system performance.

Paperid: 6436, https://arxiv.org/pdf/2510.24949.pdf

Abstract:
Assessing scenario coverage is crucial for evaluating the robustness of autonomous agents, yet existing methods rely on expensive human annotations or computationally intensive Large Vision-Language Models (LVLMs). These approaches are impractical for large-scale deployment due to cost and efficiency constraints. To address these shortcomings, we propose SCOUT (Scenario Coverage Oversight and Understanding Tool), a lightweight surrogate model designed to predict scenario coverage labels directly from an agent's latent sensor representations. SCOUT is trained through a distillation process, learning to approximate LVLM-generated coverage labels while eliminating the need for continuous LVLM inference or human annotation. By leveraging precomputed perception features, SCOUT avoids redundant computations and enables fast, scalable scenario coverage estimation. We evaluate our method across a large dataset of real-life autonomous navigation scenarios, demonstrating that it maintains high accuracy while significantly reducing computational cost. Our results show that SCOUT provides an effective and practical alternative for large-scale coverage analysis. While its performance depends on the quality of LVLM-generated training labels, SCOUT represents a major step toward efficient scenario coverage oversight in autonomous systems.

Paperid: 6437, https://arxiv.org/pdf/2510.24942.pdf

Abstract:
Despite their impressive performance, vision-language models (VLMs) still struggle on culturally situated inputs. To understand how VLMs process culturally grounded information, we study the presence of culture-sensitive neurons, i.e. neurons whose activations show preferential sensitivity to inputs associated with particular cultural contexts. We examine whether such neurons are important for culturally diverse visual question answering and where they are located. Using the CVQA benchmark, we identify neurons of culture selectivity and perform causal tests by deactivating the neurons flagged by different identification methods. Experiments on three VLMs across 25 cultural groups demonstrate the existence of neurons whose ablation disproportionately harms performance on questions about the corresponding cultures, while having minimal effects on others. Moreover, we propose a new margin-based selector - Contrastive Activation Selection (CAS), and show that it outperforms existing probability- and entropy-based methods in identifying culture-sensitive neurons. Finally, our layer-wise analyses reveals that such neurons tend to cluster in certain decoder layers. Overall, our findings shed new light on the internal organization of multimodal representations.

Paperid: 6438, https://arxiv.org/pdf/2510.24115.pdf

Abstract:
For doctors to truly trust artificial intelligence, it can't be a black box. They need to understand its reasoning, almost as if they were consulting a colleague. We created HistoLens1 to be that transparent, collaborative partner. It allows a pathologist to simply ask a question in plain English about a tissue slide--just as they would ask a trainee. Our system intelligently translates this question into a precise query for its AI engine, which then provides a clear, structured report. But it doesn't stop there. If a doctor ever asks, "Why?", HistoLens can instantly provide a 'visual proof' for any finding--a heatmap that points to the exact cells and regions the AI used for its analysis. We've also ensured the AI focuses only on the patient's tissue, just like a trained pathologist would, by teaching it to ignore distracting background noise. The result is a workflow where the pathologist remains the expert in charge, using a trustworthy AI assistant to verify their insights and make faster, more confident diagnoses.

Paperid: 6439, https://arxiv.org/pdf/2510.22975.pdf

Abstract:
Physical simulation relies on spatially-varying mechanical properties, often laboriously hand-crafted. VoMP is a feed-forward method trained to predict Young's modulus ($E$), Poisson's ratio ($ν$), and density ($ρ$) throughout the volume of 3D objects, in any representation that can be rendered and voxelized. VoMP aggregates per-voxel multi-view features and passes them to our trained Geometry Transformer to predict per-voxel material latent codes. These latents reside on a manifold of physically plausible materials, which we learn from a real-world dataset, guaranteeing the validity of decoded per-voxel materials. To obtain object-level training data, we propose an annotation pipeline combining knowledge from segmented 3D datasets, material databases, and a vision-language model, along with a new benchmark. Experiments show that VoMP estimates accurate volumetric properties, far outperforming prior art in accuracy and speed.

Paperid: 6440, https://arxiv.org/pdf/2510.22056.pdf

Abstract:
Anomaly detection in surveillance videos remains a challenging task due to the diversity of abnormal events, class imbalance, and scene-dependent visual clutter. To address these issues, we propose a robust deep learning framework that integrates human-centric preprocessing with spatio-temporal modeling for multi-class anomaly classification. Our pipeline begins by applying YOLO-World - an open-vocabulary vision-language detector - to identify human instances in raw video clips, followed by ByteTrack for consistent identity-aware tracking. Background regions outside detected bounding boxes are suppressed via Gaussian blurring, effectively reducing scene-specific distractions and focusing the model on behaviorally relevant foreground content. The refined frames are then processed by an ImageNet-pretrained InceptionV3 network for spatial feature extraction, and temporal dynamics are captured using a bidirectional LSTM (BiLSTM) for sequence-level classification. Evaluated on a five-class subset of the UCF-Crime dataset (Normal, Burglary, Fighting, Arson, Explosion), our method achieves a mean test accuracy of 92.41% across three independent trials, with per-class F1-scores consistently exceeding 0.85. Comprehensive evaluation metrics - including confusion matrices, ROC curves, and macro/weighted averages - demonstrate strong generalization and resilience to class imbalance. The results confirm that foreground-focused preprocessing significantly enhances anomaly discrimination in real-world surveillance scenarios.

Paperid: 6441, https://arxiv.org/pdf/2510.21850.pdf

Abstract:
Understanding long-context visual information remains a fundamental challenge for vision-language models, particularly in agentic tasks such as GUI control and web navigation. While web pages and GUI environments are inherently structured documents, current VLMs typically neglect decision-oriented document understanding in their training objectives. Existing approaches primarily extend visual embeddings to process long, high-resolution inputs, but these methods are memory-intensive and impractical for locally deployable solutions. To address these issues, we propose SCoPE VLM, a document navigation expert that leverages a novel Chain of Scroll mechanism to selectively and recursively navigate documents, focusing exclusively on relevant segments. We introduce a dedicated data generation pipeline to construct informative Chain of Scroll trajectories and Episodic Group Relative Policy Optimization, a tailored reinforcement learning method to reduce the gap between training and inference. Our method substantially reduces memory usage and effectively models human-like reading behaviors. To the best of our knowledge, SCoPE VLM is the first framework to explicitly model agentic reading patterns in multi-page document question answering, advancing the capabilities of multimodal agents.

Paperid: 6442, https://arxiv.org/pdf/2510.21809.pdf

Abstract:
The field of multimodal robot navigation in indoor environments has garnered significant attention in recent years. However, as tasks and methods become more advanced, the action decision systems tend to become more complex and operate as black-boxes. For a reliable system, the ability to explain or describe its decisions is crucial; however, there tends to be a trade-off in that explainable systems can not outperform non-explainable systems in terms of performance. In this paper, we propose incorporating the task of describing actions in language into the reinforcement learning of navigation as an auxiliary task. Existing studies have found it difficult to incorporate describing actions into reinforcement learning due to the absence of ground-truth data. We address this issue by leveraging knowledge distillation from pre-trained description generation models, such as vision-language models. We comprehensively evaluate our approach across various navigation tasks, demonstrating that it can describe actions while attaining high navigation performance. Furthermore, it achieves state-of-the-art performance in the particularly challenging multimodal navigation task of semantic audio-visual navigation.

Paperid: 6443, https://arxiv.org/pdf/2510.21757.pdf

Abstract:
Agricultural disease management in developing countries such as India, Kenya, and Nigeria faces significant challenges due to limited access to expert plant pathologists, unreliable internet connectivity, and cost constraints that hinder the deployment of large-scale AI systems. This work introduces a cost-effective self-consistency framework to improve vision-language model (VLM) reliability for agricultural image captioning. The proposed method employs semantic clustering, using a lightweight (80MB) pre-trained embedding model to group multiple candidate responses. It then selects the most coherent caption -- containing a diagnosis, symptoms, analysis, treatment, and prevention recommendations -- through a cosine similarity-based consensus. A practical human-in-the-loop (HITL) component is incorporated, wherein user confirmation of the crop type filters erroneous generations, ensuring higher-quality input for the consensus mechanism. Applied to the publicly available PlantVillage dataset using a fine-tuned 3B-parameter PaliGemma model, our framework demonstrates improvements over standard decoding methods. Evaluated on 800 crop disease images with up to 21 generations per image, our single-cluster consensus method achieves a peak accuracy of 83.1% with 10 candidate generations, compared to the 77.5% baseline accuracy of greedy decoding. The framework's effectiveness is further demonstrated when considering multiple clusters; accuracy rises to 94.0% when a correct response is found within any of the top four candidate clusters, outperforming the 88.5% achieved by a top-4 selection from the baseline.

Paperid: 6444, https://arxiv.org/pdf/2510.21175.pdf

Abstract:
Pre-trained vision-language models (VLMs), such as CLIP, have demonstrated remarkable zero-shot generalization, enabling deployment in a wide range of real-world tasks without additional task-specific training. However, in real deployment scenarios with evolving environments or emerging classes, these models inevitably face distributional shifts and novel tasks. In such contexts, static zero-shot capabilities are insufficient, and there is a growing need for continual learning methods that allow models to adapt over time while avoiding catastrophic forgetting. We introduce NuSA-CL (Null Space Adaptation for Continual Learning), a lightweight memory-free continual learning framework designed to address this challenge. NuSA-CL employs low-rank adaptation and constrains task-specific weight updates to lie within an approximate null space of the model's current parameters. This strategy minimizes interference with previously acquired knowledge, effectively preserving the zero-shot capabilities of the original model. Unlike methods relying on replay buffers or costly distillation, NuSA-CL imposes minimal computational and memory overhead, making it practical for deployment in resource-constrained, real-world continual learning environments. Experiments show that our framework not only effectively preserves zero-shot transfer capabilities but also achieves highly competitive performance on continual learning benchmarks. These results position NuSA-CL as a practical and scalable solution for continually evolving zero-shot VLMs in real-world applications.

Paperid: 6445, https://arxiv.org/pdf/2510.20287.pdf

Abstract:
Large Vision Language models have seen huge application in several sports use-cases recently. Most of these works have been targeted towards a limited subset of popular sports like soccer, cricket, basketball etc; focusing on generative tasks like visual question answering, highlight generation. This work analyzes the applicability of the modern video foundation models (both encoder and decoder) for a very niche but hugely popular dance sports - breakdance. Our results show that Video Encoder models continue to outperform state-of-the-art Video Language Models for prediction tasks. We provide insights on how to choose the encoder model and provide a thorough analysis into the workings of a finetuned decoder model for breakdance video classification.

Paperid: 6446, https://arxiv.org/pdf/2510.18583.pdf

Abstract:
Multimodal dataset distillation aims to synthesize a small set of image-text pairs that enables efficient training of large-scale vision-language models. While dataset distillation has shown promise in unimodal tasks, extending it to multimodal contrastive learning presents key challenges: learning cross-modal alignment and managing the high computational cost of large encoders. Prior approaches address scalability by freezing the text encoder and update only the image encoder and text projection layer. However, we find this severely limits semantic alignment and becomes a bottleneck for performance scaling. We propose CovMatch, a scalable dataset distillation framework that aligns the cross-covariance of real and synthetic features while regularizing feature distributions within each modality. Unlike prior approaches, CovMatch enables joint optimization of both encoders, leading to stronger cross-modal alignment and improved performance. Evaluated on Flickr30K and COCO, CovMatch outperforms state-of-the-art multimodal distillation methods and achieves up to 6.8% absolute gains in retrieval accuracy using only 500 synthetic pairs.

Paperid: 6447, https://arxiv.org/pdf/2510.18483.pdf

Abstract:
Human players do more than press buttons: they ground what they see on screen into precise keyboard-mouse actions and, when stuck, they seek information before trying again. We ask whether current vision-language models (VLMs) can do the same. Despite encouraging results under simplified control or tool scaffolds, human-like play in a real client - mapping raw screenshots to temporally coherent low-level actions while deciding when to ask for guidance - remains an open challenge. We introduce StarBench, a turn-based RPG benchmark derived from Honkai: Star Rail that targets these two human-like competencies: multimodal decision-making from pixels to actions and agentic information seeking. StarBench standardizes evaluation across eight combat tasks and two regimes with shared tasks and metrics: (i) direct control, where agents receive only screenshots and must emit low-level primitives (click and keypress) with no semantic hints; and (ii) tool-assisted control, where higher-level intents can be mapped to primitives by detectors and OCR outputs provide optional textualized observations to ease UI grounding. To mirror human practice, StarBench also includes an ask-or-act diagnostic that measures whether and when agents choose to request brief guidance before proceeding, and how that choice affects subsequent performance. We report reference baselines for contemporary VLMs and a human reference. Results expose sizable gaps in perception-to-control fidelity in the direct regime, while showing that judicious information seeking correlates with improved success, establishing StarBench as a reproducible yardstick for agentic information seeking and multimodal decision-making in real-client play.

Paperid: 6448, https://arxiv.org/pdf/2510.17197.pdf

Abstract:
As the capabilities of Vision-Language Models (VLMs) advance, they can process increasingly large inputs, which, unlike in LLMs, generates significant visual token redundancy and leads to prohibitive inference costs. While many methods aim to reduce these costs by pruning visual tokens, existing approaches, whether based on attention or diversity, typically neglect the guidance of the text prompt and thus fail to prioritize task relevance. In this work, we propose a novel, zero-shot method that reframes the problem by introducing a prompt-aware perspective, explicitly modeling visual token pruning as a balance between task relevance and information diversity. Our hierarchical approach first selects a core set of task-relevant visual tokens and then supplements them with diversity tokens to preserve broader context. Experiments across multiple models and benchmarks show that our method achieves performance that matches or surpasses the state-of-the-art with only minimal accuracy loss, even when pruning up to 90\% of the tokens. Furthermore, these gains are accompanied by significant reductions in GPU memory footprint and inference latency.

Paperid: 6449, https://arxiv.org/pdf/2510.16198.pdf

Abstract:
Despite recent advances in AI, multimodal culturally diverse datasets are still limited, particularly for regions in the Middle East and Africa. In this paper, we introduce EgMM-Corpus, a multimodal dataset dedicated to Egyptian culture. By designing and running a new data collection pipeline, we collected over 3,000 images, covering 313 concepts across landmarks, food, and folklore. Each entry in the dataset is manually validated for cultural authenticity and multimodal coherence. EgMM-Corpus aims to provide a reliable resource for evaluating and training vision-language models in an Egyptian cultural context. We further evaluate the zero-shot performance of Contrastive Language-Image Pre-training CLIP on EgMM-Corpus, on which it achieves 21.2% Top-1 accuracy and 36.4% Top-5 accuracy in classification. These results underscore the existing cultural bias in large-scale vision-language models and demonstrate the importance of EgMM-Corpus as a benchmark for developing culturally aware models.

Paperid: 6450, https://arxiv.org/pdf/2510.15948.pdf

Abstract:
Large Vision-Language Models (LVLMs) have achieved remarkable progress in multimodal perception and generation, yet their safety alignment remains a critical challenge.Existing defenses and vulnerable to multimodal jailbreaks, as visual inputs introduce new attack surfaces, reasoning chains lack safety supervision, and alignment often degrades under modality fusion.To overcome these limitation, we propose VisuoAlign, a framework for multi-modal safety alignment via prompt-guided tree search.VisuoAlign embeds safety constrains into the reasoning process through visual-textual interactive prompts, employs Monte Carlo Tree Search(MCTS) to systematically construct diverse safety-critical prompt trajectories, and introduces prompt-based scaling to ensure real-time risk detection and compliant responses.Extensive experiments demonstrate that VisuoAlign proactively exposes risks, enables comprehensive dataset generation, and significantly improves the robustness of LVLMs against complex cross-modal threats.

Paperid: 6451, https://arxiv.org/pdf/2510.14737.pdf

Abstract:
Hierarchical image classification predicts labels across a semantic taxonomy, but existing methods typically assume complete, fine-grained annotations, an assumption rarely met in practice. Real-world supervision varies in granularity, influenced by image quality, annotator expertise, and task demands; a distant bird may be labeled Bird, while a close-up reveals Bald eagle. We introduce ImageNet-F, a large-scale benchmark curated from ImageNet and structured into cognitively inspired basic, subordinate, and fine-grained levels. Using CLIP as a proxy for semantic ambiguity, we simulate realistic, mixed-granularity labels reflecting human annotation behavior. We propose free-grain learning, with heterogeneous supervision across instances. We develop methods that enhance semantic guidance via pseudo-attributes from vision-language models and visual guidance via semi-supervised learning. These, along with strong baselines, substantially improve performance under mixed supervision. Together, our benchmark and methods advance hierarchical classification under real-world constraints.

Paperid: 6452, https://arxiv.org/pdf/2510.13889.pdf

Abstract:
Food image classification plays a vital role in intelligent food quality inspection, dietary assessment, and automated monitoring. However, most existing supervised models rely heavily on large labeled datasets and exhibit limited generalization to unseen food categories. To overcome these challenges, this study introduces MultiFoodChat, a dialogue-driven multi-agent reasoning framework for zero-shot food recognition. The framework integrates vision-language models (VLMs) and large language models (LLMs) to enable collaborative reasoning through multi-round visual-textual dialogues. An Object Perception Token (OPT) captures fine-grained visual attributes, while an Interactive Reasoning Agent (IRA) dynamically interprets contextual cues to refine predictions. This multi-agent design allows flexible and human-like understanding of complex food scenes without additional training or manual annotations. Experiments on multiple public food datasets demonstrate that MultiFoodChat achieves superior recognition accuracy and interpretability compared with existing unsupervised and few-shot methods, highlighting its potential as a new paradigm for intelligent food quality inspection and analysis.

Paperid: 6453, https://arxiv.org/pdf/2510.13364.pdf

Abstract:
Recent Vision-Language Models (VLMs) enable zero-shot classification by aligning images and text in a shared space, a promising approach for data-scarce conditions. However, the influence of prompt design on recognizing visually similar categories, such as human postures, is not well understood. This study investigates how prompt specificity affects the zero-shot classification of sitting, standing, and walking/running on a small, 285-image COCO-derived dataset. A suite of modern VLMs, including OpenCLIP, MetaCLIP 2, and SigLip, were evaluated using a three-tiered prompt design that systematically increases linguistic detail. Our findings reveal a compelling, counter-intuitive trend: for the highest-performing models (MetaCLIP 2 and OpenCLIP), the simplest, most basic prompts consistently achieve the best results. Adding descriptive detail significantly degrades performance for instance, MetaCLIP 2's multi-class accuracy drops from 68.8\% to 55.1\% a phenomenon we term "prompt overfitting". Conversely, the lower-performing SigLip model shows improved classification on ambiguous classes when given more descriptive, body-cue-based prompts.

Paperid: 6454, https://arxiv.org/pdf/2510.12548.pdf

Abstract:
While Vision Language Models (VLMs) learn conceptual representations, in the form of generalized knowledge, during training, they are typically used to analyze individual instances. When evaluation instances are atypical, this paradigm results in tension between two priors in the model. The first is a pragmatic prior that the textual and visual input are both relevant, arising from VLM finetuning on congruent inputs; the second is a semantic prior that the conceptual representation is generally true for instances of the category. In order to understand how VLMs trade off these priors, we introduce a new evaluation dataset, VISaGE, consisting of both typical and exceptional images. In carefully balanced experiments, we show that conceptual understanding degrades when the assumption of congruency underlying the pragmatic prior is violated with incongruent images. This effect is stronger than the effect of the semantic prior when querying about individual instances.

Paperid: 6455, https://arxiv.org/pdf/2510.12095.pdf

Abstract:
In this study, we present IL3D, a large-scale dataset meticulously designed for large language model (LLM)-driven 3D scene generation, addressing the pressing demand for diverse, high-quality training data in indoor layout design. Comprising 27,816 indoor layouts across 18 prevalent room types and a library of 29,215 high-fidelity 3D object assets, IL3D is enriched with instance-level natural language annotations to support robust multimodal learning for vision-language tasks. We establish rigorous benchmarks to evaluate LLM-driven scene generation. Experimental results show that supervised fine-tuning (SFT) of LLMs on IL3D significantly improves generalization and surpasses the performance of SFT on other datasets. IL3D offers flexible multimodal data export capabilities, including point clouds, 3D bounding boxes, multiview images, depth maps, normal maps, and semantic masks, enabling seamless adaptation to various visual tasks. As a versatile and robust resource, IL3D significantly advances research in 3D scene generation and embodied intelligence, by providing high-fidelity scene data to support environment perception tasks of embodied agents.

Paperid: 6456, https://arxiv.org/pdf/2508.20830.pdf

Abstract:
This paper presents a novel pipeline for 2D keypoint estima- tion of surgical tools by leveraging Vision Language Models (VLMs) fine- tuned using a low rank adjusting (LoRA) technique. Unlike traditional Convolutional Neural Network (CNN) or Transformer-based approaches, which often suffer from overfitting in small-scale medical datasets, our method harnesses the generalization capabilities of pre-trained VLMs. We carefully design prompts to create an instruction-tuning dataset and use them to align visual features with semantic keypoint descriptions. Experimental results show that with only two epochs of fine tuning, the adapted VLM outperforms the baseline models, demonstrating the ef- fectiveness of LoRA in low-resource scenarios. This approach not only improves keypoint detection performance, but also paves the way for future work in 3D surgical hands and tools pose estimation.

Paperid: 6457, https://arxiv.org/pdf/2508.20549.pdf

Abstract:
The application of Vision-Language Models (VLMs) in medicine is critically hampered by the scarcity of high-quality, expert-annotated data. Supervised Fine-Tuning (SFT) on existing datasets often leads to poor generalization on unseen modalities and tasks, while Reinforcement Learning (RL), a promising alternative, is stymied by the lack of reliable reward signals in this data-scarce domain. To break this impasse, we introduce Generative Reward Learning for Medical Reasoning (MedGR$^2$), a novel framework that creates a self-improving virtuous cycle. MedGR$^2$ co-develops a data generator and a reward model, enabling the automated, continuous creation of high-quality, multi-modal medical data that serves as both a superior training source for SFT and RL. Our experiments demonstrate that SFT with MedGR$^2$-produced data already surpasses baselines trained on large-scale, human-curated datasets. Crucially, when leveraging this data for RL via Group Relative Policy Optimization (GRPO), our model achieves state-of-the-art cross-modality and cross-task generalization, significantly outperforming specialized RL-based methods. Furthermore, our compact model, empowered by MedGR$^2$, achieves performance competitive with foundation models possessing over 10 times more parameters. MedGR$^2$ presents a new paradigm for data-efficient learning in high-stakes domains, transforming the problem from data scarcity to data generation and unlocking the full potential of RL for building truly generalizable medical AI.

Paperid: 6458, https://arxiv.org/pdf/2508.19852.pdf

Abstract:
In egocentric scenarios, anticipating both the next action and its visual outcome is essential for understanding human-object interactions and for enabling robotic planning. However, existing paradigms fall short of jointly modeling these aspects. Vision-Language-Action (VLA) models focus on action prediction but lack explicit modeling of how actions influence the visual scene, while video prediction models generate future frames without conditioning on specific actions, often resulting in implausible or contextually inconsistent outcomes. To bridge this gap, we propose a unified two-stage predictive framework that jointly models action and visual future in egocentric scenarios, conditioned on hand trajectories. In the first stage, we perform consecutive state modeling to process heterogeneous inputs (visual observations, language, and action history) and explicitly predict future hand trajectories. In the second stage, we introduce causal cross-attention to fuse multi-modal cues, leveraging inferred action signals to guide an image-based Latent Diffusion Model (LDM) for frame-by-frame future video generation. Our approach is the first unified model designed to handle both egocentric human activity understanding and robotic manipulation tasks, providing explicit predictions of both upcoming actions and their visual consequences. Extensive experiments on Ego4D, BridgeData, and RLBench demonstrate that our method outperforms state-of-the-art baselines in both action prediction and future video synthesis.

Paperid: 6459, https://arxiv.org/pdf/2508.19724.pdf

Abstract:
Commonsense visual-question answering often hinges on knowledge that is missing from the image or the question. Small vision-language models (sVLMs) such as ViLT, VisualBERT and FLAVA therefore lag behind their larger generative counterparts. To study the effect of careful commonsense knowledge integration on sVLMs, we present an end-to-end framework (NLKI) that (i) retrieves natural language facts, (ii) prompts an LLM to craft natural language explanations, and (iii) feeds both signals to sVLMs respectively across two commonsense VQA datasets (CRIC, AOKVQA) and a visual-entailment dataset (e-SNLI-VE). Facts retrieved using a fine-tuned ColBERTv2 and an object information-enriched prompt yield explanations that largely cut down hallucinations, while lifting the end-to-end answer accuracy by up to 7% (across 3 datasets), making FLAVA and other models in NLKI match or exceed medium-sized VLMs such as Qwen-2 VL-2B and SmolVLM-2.5B. As these benchmarks contain 10-25% label noise, additional finetuning using noise-robust losses (such as symmetric cross entropy and generalised cross entropy) adds another 2.5% in CRIC, and 5.5% in AOKVQA. Our findings expose when LLM-based commonsense knowledge beats retrieval from commonsense knowledge bases, how noise-aware training stabilises small models in the context of external knowledge augmentation, and why parameter-efficient commonsense reasoning is now within reach for 250M models.

Paperid: 6460, https://arxiv.org/pdf/2508.18984.pdf

Abstract:
Document Visual Question Answering (Document VQA) must cope with documents that span dozens of pages, yet leading systems still concatenate every page or rely on very large vision-language models, both of which are memory-hungry. Retrieval-Augmented Generation (RAG) offers an attractive alternative, first retrieving a concise set of relevant segments before generating answers from this selected evidence. In this paper, we systematically evaluate the impact of incorporating RAG into Document VQA through different retrieval variants - text-based retrieval using OCR tokens and purely visual retrieval without OCR - across multiple models and benchmarks. Evaluated on the multi-page datasets MP-DocVQA, DUDE, and InfographicVQA, the text-centric variant improves the "concatenate-all-pages" baseline by up to +22.5 ANLS, while the visual variant achieves +5.0 ANLS improvement without requiring any text extraction. An ablation confirms that retrieval and reranking components drive most of the gain, whereas the layout-guided chunking strategy - proposed in several recent works to leverage page structure - fails to help on these datasets. Our experiments demonstrate that careful evidence selection consistently boosts accuracy across multiple model sizes and multi-page benchmarks, underscoring its practical value for real-world Document VQA.

Paperid: 6461, https://arxiv.org/pdf/2508.18886.pdf

Abstract:
Ensuring fairness across demographic groups in medical diagnosis is essential for equitable healthcare, particularly under distribution shifts caused by variations in imaging equipment and clinical practice. Vision-language models (VLMs) exhibit strong generalization, and text prompts encode identity attributes, enabling explicit identification and removal of sensitive directions. However, existing debiasing approaches typically address vision and text modalities independently, leaving residual cross-modal misalignment and fairness gaps. To address this challenge, we propose DualFairVL, a multimodal prompt-learning framework that jointly debiases and aligns cross-modal representations. DualFairVL employs a parallel dual-branch architecture that separates sensitive and target attributes, enabling disentangled yet aligned representations across modalities. Approximately orthogonal text anchors are constructed via linear projections, guiding cross-attention mechanisms to produce fused features. A hypernetwork further disentangles attribute-related information and generates instance-aware visual prompts, which encode dual-modal cues for fairness and robustness. Prototype-based regularization is applied in the visual branch to enforce separation of sensitive features and strengthen alignment with textual anchors. Extensive experiments on eight medical imaging datasets across four modalities show that DualFairVL achieves state-of-the-art fairness and accuracy under both in- and out-of-distribution settings, outperforming full fine-tuning and parameter-efficient baselines with only 3.6M trainable parameters. Code will be released upon publication.

Paperid: 6462, https://arxiv.org/pdf/2508.17595.pdf

Abstract:
Reasoning about fine-grained spatial relationships in warehouse-scale environments poses a significant challenge for existing vision-language models (VLMs), which often struggle to comprehend 3D layouts, object arrangements, and multimodal cues in real-world industrial settings. In this paper, we present TinyGiantVLM, a lightweight and modular two-stage framework designed for physical spatial reasoning, distinguishing itself from traditional geographic reasoning in complex logistics scenes. Our approach encodes both global and region-level features from RGB and depth modalities using pretrained visual backbones. To effectively handle the complexity of high-modality inputs and diverse question types, we incorporate a Mixture-of-Experts (MoE) fusion module, which dynamically combines spatial representations to support downstream reasoning tasks and improve convergence. Training is conducted in a two-phase strategy: the first phase focuses on generating free-form answers to enhance spatial reasoning ability, while the second phase uses normalized answers for evaluation. Evaluated on Track 3 of the AI City Challenge 2025, our 64M-parameter base model achieved 5th place on the leaderboard with a score of 66.8861, demonstrating strong performance in bridging visual perception and spatial understanding in industrial environments. We further present an 80M-parameter variant with expanded MoE capacity, which demonstrates improved performance on spatial reasoning tasks.

Paperid: 6463, https://arxiv.org/pdf/2508.16707.pdf

Abstract:
Vision-Language Pretrained (VLP) models have achieved impressive performance on multimodal tasks, including text-image retrieval, based on dense representations. Meanwhile, Learned Sparse Retrieval (LSR) has gained traction in text-only settings due to its interpretability and efficiency with fast term-based lookup via inverted indexes. Inspired by these advantages, recent work has extended LSR to the multimodal domain. However, these methods often rely on computationally expensive contrastive pre-training, or distillation from a frozen dense model, which limits the potential for mutual enhancement. To address these limitations, we propose a simple yet effective framework that enables bi-directional learning between dense and sparse representations through Self-Knowledge Distillation. This bi-directional learning is achieved using an integrated similarity score-a weighted sum of dense and sparse similarities-which serves as a shared teacher signal for both representations. To ensure efficiency, we fine-tune the final layer of the dense encoder and the sparse projection head, enabling easy adaptation of any existing VLP model. Experiments on MSCOCO and Flickr30k demonstrate that our sparse retriever not only outperforms existing sparse baselines, but also achieves performance comparable to-or even surpassing-its dense counterparts, while retaining the benefits of sparse models.

Paperid: 6464, https://arxiv.org/pdf/2508.16157.pdf

Abstract:
Pre-trained Vision-Language Models (VLMs) have recently shown promise in detecting anomalies. However, previous approaches are fundamentally limited by their reliance on human-designed prompts and the lack of accessible anomaly samples, leading to significant gaps in context-specific anomaly understanding. In this paper, we propose \textbf{A}daptive \textbf{P}rompt \textbf{T}uning with semantic alignment for anomaly detection (APT), a groundbreaking prior knowledge-free, few-shot framework and overcomes the limitations of traditional prompt-based approaches. APT uses self-generated anomaly samples with noise perturbations to train learnable prompts that capture context-dependent anomalies in different scenarios. To prevent overfitting to synthetic noise, we propose a Self-Optimizing Meta-prompt Guiding Scheme (SMGS) that iteratively aligns the prompts with general anomaly semantics while incorporating diverse synthetic anomaly. Our system not only advances pixel-wise anomaly detection, but also achieves state-of-the-art performance on multiple benchmark datasets without requiring prior knowledge for prompt crafting, establishing a robust and versatile solution for real-world anomaly detection.

Paperid: 6465, https://arxiv.org/pdf/2508.14537.pdf

Abstract:
Whole slide images (WSIs) in computational pathology (CPath) pose a major computational challenge due to their gigapixel scale, often requiring the processing of tens to hundreds of thousands of high-resolution patches per slide. This results in prohibitive encoding costs, with preprocessing and training times extending to days or even weeks-making WSI encoding the most significant bottleneck in real-world deployment. In this work, we propose WISE-FUSE, an adaptive WSI encoding framework that leverages pathology-domain vision-language models and large language models to address this challenge by selectively processing diagnostically relevant regions. WISE-FUSE first computes similarity scores between low-resolution patches and class-specific textual descriptions using a knowledge distillation mechanism that preserves fine-grained diagnostic features. Based on these similarity scores, we select a small subset of informative regions for the target task, which quickly eliminates irrelevant patches at the coarse level. The corresponding high-resolution patches are then selectively encoded and fused with textual embeddings to reinforce diagnostic context. Extensive experiments demonstrate that WISE-FUSE reduces WSI encoding time by over threefold while achieving diagnostic performance comparable to or surpassing that of exhaustive patch processing, offering a scalable and practical solution for CPath.

Paperid: 6466, https://arxiv.org/pdf/2508.14280.pdf

Abstract:
Explainable object recognition using vision-language models such as CLIP involves predicting accurate category labels supported by rationales that justify the decision-making process. Existing methods typically rely on prompt-based conditioning, which suffers from limitations in CLIP's text encoder and provides weak conditioning on explanatory structures. Additionally, prior datasets are often restricted to single, and frequently noisy, rationales that fail to capture the full diversity of discriminative image features. In this work, we introduce a multi-rationale explainable object recognition benchmark comprising datasets in which each image is annotated with multiple ground-truth rationales, along with evaluation metrics designed to offer a more comprehensive representation of the task. To overcome the limitations of previous approaches, we propose a contrastive conditional inference (CCI) framework that explicitly models the probabilistic relationships among image embeddings, category labels, and rationales. Without requiring any training, our framework enables more effective conditioning on rationales to predict accurate object categories. Our approach achieves state-of-the-art results on the multi-rationale explainable object recognition benchmark, including strong zero-shot performance, and sets a new standard for both classification accuracy and rationale quality. Together with the benchmark, this work provides a more complete framework for evaluating future models in explainable object recognition. The code will be made available online.

Paperid: 6467, https://arxiv.org/pdf/2508.11975.pdf

Abstract:
Vision Language Models (VLMs) often struggle with chart understanding tasks, particularly in accurate chart description and complex reasoning. Synthetic data generation is a promising solution, while usually facing the challenge of noise labels. To address this challenge, we first introduce a chart synthesis pipeline that generates aligned chart-question-answer triplets through code generation and execution, ensuring the reliability of synthetic data without human intervention. Furthermore, inspired by test-time scaling that increases inference budget and thereby improves performance, we design a candidate-conditioned answering process. The VLM first generates multiple responses per query, and then synthesizes the final answer by contextualizing these candidates. Experiments demonstrate significant improvements, with up to 15.50 points accuracy gain over the initial VLM, in a fully self-improving paradigm without either human-labeled data or external models.

Paperid: 6468, https://arxiv.org/pdf/2508.11808.pdf

Abstract:
The modern web is saturated with multimodal content, intensifying the challenge of detecting hateful memes, where harmful intent is often conveyed through subtle interactions between text and image under the guise of humor or satire. While recent advances in Vision-Language Models (VLMs) show promise, these models lack support for fine-grained supervision and remain susceptible to implicit hate speech. In this paper, we present a dual-pronged approach to improve multimodal hate detection. First, we propose a prompt optimization framework that systematically varies prompt structure, supervision granularity, and training modality. We show that prompt design and label scaling both influence performance, with structured prompts improving robustness even in small models, and InternVL2 achieving the best F1-scores across binary and scaled settings. Second, we introduce a multimodal data augmentation pipeline that generates 2,479 counterfactually neutral memes by isolating and rewriting the hateful modality. This pipeline, powered by a multi-agent LLM-VLM setup, successfully reduces spurious correlations and improves classifier generalization. Our approaches inspire new directions for building synthetic data to train robust and fair vision-language models. Our findings demonstrate that prompt structure and data composition are as critical as model size, and that targeted augmentation can support more trustworthy and context-sensitive hate detection.

Paperid: 6469, https://arxiv.org/pdf/2508.10552.pdf

Abstract:
Multimodal Large Language Models (MLLMs) have demonstrated remarkable capabilities across a diverse range of multimodal tasks. However, these models suffer from a core problem known as text dominance: they depend heavily on text for their inference, while underutilizing other modalities. While prior work has acknowledged this phenomenon in vision-language tasks, often attributing it to data biases or model architectures. In this paper, we conduct the first systematic investigation of text dominance across diverse data modalities, including images, videos, audio, time-series, and graphs. To measure this imbalance, we propose two evaluation metrics: the Modality Dominance Index (MDI) and the Attention Efficiency Index (AEI). Our comprehensive analysis reveals that text dominance is both significant and pervasive across all tested modalities. Our in-depth analysis identifies three underlying causes: attention dilution from severe token redundancy in non-textual modalities, the influence of fusion architecture design, and task formulations that implicitly favor textual inputs. Furthermore, we propose a simple token compression method that effectively rebalances model attention. Applying this method to LLaVA-7B, for instance, drastically reduces its MDI from 10.23 to a well-balanced value of 0.86. Our analysis and methodological framework offer a foundation for the development of more equitable and comprehensive multimodal language models.

Paperid: 6470, https://arxiv.org/pdf/2508.07648.pdf

Abstract:
For transradial amputees, robotic prosthetic hands promise to regain the capability to perform daily living activities. To advance next-generation prosthetic hand control design, it is crucial to address current shortcomings in robustness to out of lab artifacts, and generalizability to new environments. Due to the fixed number of object to interact with in existing datasets, contrasted with the virtually infinite variety of objects encountered in the real world, current grasp models perform poorly on unseen objects, negatively affecting users' independence and quality of life. To address this: (i) we define semantic projection, the ability of a model to generalize to unseen object types and show that conventional models like YOLO, despite 80% training accuracy, drop to 15% on unseen objects. (ii) we propose Grasp-LLaVA, a Grasp Vision Language Model enabling human-like reasoning to infer the suitable grasp type estimate based on the object's physical characteristics resulting in a significant 50.2% accuracy over unseen object types compared to 36.7% accuracy of an SOTA grasp estimation model. Lastly, to bridge the performance-latency gap, we propose Hybrid Grasp Network (HGN), an edge-cloud deployment infrastructure enabling fast grasp estimation on edge and accurate cloud inference as a fail-safe, effectively expanding the latency vs. accuracy Pareto. HGN with confidence calibration (DC) enables dynamic switching between edge and cloud models, improving semantic projection accuracy by 5.6% (to 42.3%) with 3.5x speedup over the unseen object types. Over a real-world sample mix, it reaches 86% average accuracy (12.2% gain over edge-only), and 2.2x faster inference than Grasp-LLaVA alone.

Paperid: 6471, https://arxiv.org/pdf/2508.05852.pdf

Abstract:
Driver visual attention prediction is a critical task in autonomous driving and human-computer interaction (HCI) research. Most prior studies focus on estimating attention allocation at a single moment in time, typically using static RGB images such as driving scene pictures. In this work, we propose a vision-language framework that models the changing landscape of drivers' gaze through natural language, using few-shot and zero-shot learning on single RGB images. We curate and refine high-quality captions from the BDD-A dataset using human-in-the-loop feedback, then fine-tune LLaVA to align visual perception with attention-centric scene understanding. Our approach integrates both low-level cues and top-down context (e.g., route semantics, risk anticipation), enabling language-based descriptions of gaze behavior. We evaluate performance across training regimes (few shot, and one-shot) and introduce domain-specific metrics for semantic alignment and response diversity. Results show that our fine-tuned model outperforms general-purpose VLMs in attention shift detection and interpretability. To our knowledge, this is among the first attempts to generate driver visual attention allocation and shifting predictions in natural language, offering a new direction for explainable AI in autonomous driving. Our approach provides a foundation for downstream tasks such as behavior forecasting, human-AI teaming, and multi-agent coordination.

Paperid: 6472, https://arxiv.org/pdf/2508.04895.pdf

Abstract:
Modern game studios deliver new builds and patches at a rapid pace, generating thousands of bug reports, many of which embed gameplay videos. To verify and triage these bug reports, developers must watch the submitted videos. This manual review is labour-intensive, slow, and hard to scale. In this paper, we introduce an automated pipeline that reduces each video to a single frame that best matches the reported bug description, giving developers instant visual evidence that pinpoints the bug. Our pipeline begins with FFmpeg for keyframe extraction, reducing each video to a median of just 1.90% of its original frames while still capturing bug moments in 98.79 of cases. These keyframes are then evaluated by a vision--language model (GPT-4o), which ranks them based on how well they match the textual bug description and selects the most representative frame. We evaluated this approach using real-world developer-submitted gameplay videos and JIRA bug reports from a popular First-Person Shooter (FPS) game. The pipeline achieves an overall F1 score of 0.79 and Accuracy of 0.89 for the top-1 retrieved frame. Performance is highest for the Lighting & Shadow (F1 = 0.94), Physics & Collision (0.86), and UI & HUD (0.83) bug categories, and lowest for Animation & VFX (0.51). By replacing video viewing with an immediately informative image, our approach dramatically reduces manual effort and speeds up triage and regression checks, offering practical benefits to quality assurance (QA) teams and developers across the game industry.

Paperid: 6473, https://arxiv.org/pdf/2508.01653.pdf

Abstract:
Large Vision-Language Models (LVLMs) have achieved impressive performance in multimodal tasks, but they still suffer from hallucinations, i.e., generating content that is grammatically accurate but inconsistent with visual inputs. In this work, we introduce a novel map-level perspective to mitigate hallucinations in LVLMs, interpreting the hidden states of the model as a 2D semantic map. We observe that factual information is widely distributed across this map, extending beyond the localized inter- or intra-layer regions targeted by most existing methods (e.g., contrastive decoding and layer-wise consistency). Building on this insight, we propose Map-Level Attention Processing (MAP), a training-free decoding method that effectively leverages factual information through attention-based map-level operations to improve factual consistency. Specifically, we employ Layer-Wise Criss-Cross Attention to progressively refine token representations at each decoding layer by aggregating tokens from both inter- and intra-layer dimensions. Additionally, a Global-Local Logit Fusion mechanism combines logits obtained before and after global attention to further refine predictions and improve accuracy. Our method consistently improves the truthfulness and performance of LVLMs across benchmarks, such as POPE, MME, and MMHal-Bench, demonstrating the potential of the map-level decoding strategy.

Paperid: 6474, https://arxiv.org/pdf/2508.01274.pdf

Abstract:
Multimodal Large Language Models (MLLMs) process visual, acoustic, and textual inputs, addressing the limitations of single-modality LLMs. However, existing benchmarks often overlook tri-modal evaluation in Traditional Chinese and do not consider inference latency. To address this, we introduce Multi-TW, the first Traditional Chinese benchmark for evaluating the performance and latency of any-to-any multimodal models. Multi-TW includes 900 multiple-choice questions (image and text, audio and text pairs) sourced from official proficiency tests developed with the Steering Committee for the Test of Proficiency-Huayu (SC-TOP). We evaluated various any-to-any models and vision-language models (VLMs) with audio transcription. Our results show that closed-source models generally outperform open-source ones across modalities, although open-source models can perform well in audio tasks. End-to-end any-to-any pipelines offer clear latency advantages compared to VLMs using separate audio transcription. Multi-TW presents a comprehensive view of model capabilities and highlights the need for Traditional Chinese fine-tuning and efficient multimodal architectures.

Paperid: 6475, https://arxiv.org/pdf/2507.22828.pdf

Abstract:
As Vision-Language Models (VLMs) are increasingly deployed in split-DNN configurations--with visual encoders (e.g., ResNet, ViT) operating on user devices and sending intermediate features to the cloud--there is a growing privacy risk from semantic information leakage. Existing approaches to reconstructing images from these intermediate features often result in blurry, semantically ambiguous images. To directly address semantic leakage, we propose CapRecover, a cross-modality inversion framework that recovers high-level semantic content, such as labels or captions, directly from intermediate features without image reconstruction. We evaluate CapRecover on multiple datasets and victim models, demonstrating strong performance in semantic recovery. Specifically, CapRecover achieves up to 92.71% Top-1 label accuracy on CIFAR-10 and generates fluent captions from ResNet50 features on COCO2017 with ROUGE-L scores up to 0.52. Our analysis further reveals that deeper convolutional layers encode significantly more semantic information compared to shallow layers. To mitigate semantic leakage, we introduce a simple yet effective protection method: adding random noise to intermediate features at each layer and removing the noise in the next layer. Experimental results show that this approach prevents semantic leakage without additional training costs.

Paperid: 6476, https://arxiv.org/pdf/2507.21794.pdf

Abstract:
Medical image-language pre-training aims to align medical images with clinically relevant text to improve model performance on various downstream tasks. However, existing models often struggle with the variability and ambiguity inherent in medical data, limiting their ability to capture nuanced clinical information and uncertainty. This work introduces an uncertainty-aware medical image-text pre-training model that enhances generalization capabilities in medical image analysis. Building on previous methods and focusing on Chest X-Rays, our approach utilizes structured text reports generated by a large language model (LLM) to augment image data with clinically relevant context. These reports begin with a definition of the disease, followed by the `appearance' section to highlight critical regions of interest, and finally `observations' and `verdicts' that ground model predictions in clinical semantics. By modeling both inter- and intra-modal uncertainty, our framework captures the inherent ambiguity in medical images and text, yielding improved representations and performance on downstream tasks. Our model demonstrates significant advances in medical image-text pre-training, obtaining state-of-the-art performance on multiple downstream tasks.

Paperid: 6477, https://arxiv.org/pdf/2507.20028.pdf

Abstract:
Test-Time Optimization enables models to adapt to new data during inference by updating parameters on-the-fly. Recent advances in Vision-Language Models (VLMs) have explored learning prompts at test time to improve performance in downstream tasks. In this work, we extend this idea by addressing a more general and practical challenge: Can we effectively utilize an oracle in a continuous data stream where only one sample is available at a time, requiring an immediate query decision while respecting latency and memory constraints? To tackle this, we propose a novel Test-Time Active Learning (TTAL) framework that adaptively queries uncertain samples and updates prompts dynamically. Unlike prior methods that assume batched data or multiple gradient updates, our approach operates in a real-time streaming scenario with a single test sample per step. We introduce a dynamically adjusted entropy threshold for active querying, a class-balanced replacement strategy for memory efficiency, and a class-aware distribution alignment technique to enhance adaptation. The design choices are justified using careful theoretical analysis. Extensive experiments across 10 cross-dataset transfer benchmarks and 4 domain generalization datasets demonstrate consistent improvements over state-of-the-art methods while maintaining reasonable latency and memory overhead. Our framework provides a practical and effective solution for real-world deployment in safety-critical applications such as autonomous systems and medical diagnostics.

Paperid: 6478, https://arxiv.org/pdf/2507.19196.pdf

Abstract:
Large language models have given social robots the ability to autonomously engage in open-domain conversations. However, they are still missing a fundamental social skill: making use of the multiple modalities that carry social interactions. While previous work has focused on task-oriented interactions that require referencing the environment or specific phenomena in social interactions such as dialogue breakdowns, we outline the overall needs of a multimodal system for social conversations with robots. We then argue that vision-language models are able to process this wide range of visual information in a sufficiently general manner for autonomous social robots. We describe how to adapt them to this setting, which technical challenges remain, and briefly discuss evaluation practices.

Paperid: 6479, https://arxiv.org/pdf/2507.12823.pdf

Abstract:
Composed image retrieval (CIR) is a vision language task that retrieves a target image using a reference image and modification text, enabling intuitive specification of desired changes. While effectively fusing visual and textual modalities is crucial, existing methods typically adopt either early or late fusion. Early fusion tends to excessively focus on explicitly mentioned textual details and neglect visual context, whereas late fusion struggles to capture fine-grained semantic alignments between image regions and textual tokens. To address these issues, we propose FAR-Net, a multi-stage fusion framework designed with enhanced semantic alignment and adaptive reconciliation, integrating two complementary modules. The enhanced semantic alignment module (ESAM) employs late fusion with cross-attention to capture fine-grained semantic relationships, while the adaptive reconciliation module (ARM) applies early fusion with uncertainty embeddings to enhance robustness and adaptability. Experiments on CIRR and FashionIQ show consistent performance gains, improving Recall@1 by up to 2.4% and Recall@50 by 1.04% over existing state-of-the-art methods, empirically demonstrating that FAR Net provides a robust and scalable solution to CIR tasks.

Paperid: 6480, https://arxiv.org/pdf/2507.08610.pdf

Abstract:
Image captioning is an important problem in developing various AI systems, and these tasks require large volumes of annotated images to train the models. Since all existing labelled datasets are already used for training the large Vision Language Models (VLMs), it becomes challenging to improve the performance of the same. Considering this, it is essential to consider the unsupervised image captioning performance, which remains relatively under-explored. To that end, we propose LoGIC (Lewis Communication Game for Image Captioning), a Multi-agent Reinforcement Learning game. The proposed method consists of two agents, a 'speaker' and a 'listener', with the objective of learning a strategy for communicating in natural language. We train agents in the cooperative common-reward setting using the GRPO algorithm and show that improvement in image captioning performance emerges as a consequence of the agents learning to play the game. We show that using pre-trained VLMs as the 'speaker' and Large Language Model (LLM) for language understanding in the 'listener', we achieved a $46$ BLEU score after fine-tuning using LoGIC without additional labels, a $2$ units advantage in absolute metrics compared to the $44$ BLEU score of the vanilla VLM. Additionally, we replace the VLM from the 'speaker' with lightweight components: (i) a ViT for image perception and (ii) a GPT2 language generation, and train them from scratch using LoGIC, obtaining a $31$ BLEU score in the unsupervised setting, a $10$ points advantage over existing unsupervised image-captioning methods.

Paperid: 6481, https://arxiv.org/pdf/2507.08505.pdf

Abstract:
Vision-Language Models (VLMs) offer promising capabilities for mobile devices, but their deployment faces significant challenges due to computational limitations and energy inefficiency, especially for real-time applications. This study provides a comprehensive survey of deployment frameworks for VLMs on mobile devices, evaluating llama.cpp, MLC-Imp, and mllm in the context of running LLaVA-1.5 7B, MobileVLM-3B, and Imp-v1.5 3B as representative workloads on a OnePlus 13R. Each deployment framework was evaluated on the OnePlus 13R while running VLMs, with measurements covering CPU, GPU, and NPU utilization, temperature, inference time, power consumption, and user experience. Benchmarking revealed critical performance bottlenecks across frameworks: CPU resources were consistently over-utilized during token generation, while GPU and NPU accelerators were largely unused. When the GPU was used, primarily for image feature extraction, it was saturated, leading to degraded device responsiveness. The study contributes framework-level benchmarks, practical profiling tools, and an in-depth analysis of hardware utilization bottlenecks, highlighting the consistent overuse of CPUs and the ineffective or unstable use of GPUs and NPUs in current deployment frameworks.

Paperid: 6482, https://arxiv.org/pdf/2507.08024.pdf

Abstract:
Precision agriculture relies heavily on accurate image analysis for crop disease identification and treatment recommendation, yet existing vision-language models (VLMs) often underperform in specialized agricultural domains. This work presents a domain-aware framework for agricultural image processing that combines prompt-based expert evaluation with self-consistency mechanisms to enhance VLM reliability in precision agriculture applications. We introduce two key innovations: (1) a prompt-based evaluation protocol that configures a language model as an expert plant pathologist for scalable assessment of image analysis outputs, and (2) a cosine-consistency self-voting mechanism that generates multiple candidate responses from agricultural images and selects the most semantically coherent diagnosis using domain-adapted embeddings. Applied to maize leaf disease identification from field images using a fine-tuned PaliGemma model, our approach improves diagnostic accuracy from 82.2\% to 87.8\%, symptom analysis from 38.9\% to 52.2\%, and treatment recommendation from 27.8\% to 43.3\% compared to standard greedy decoding. The system remains compact enough for deployment on mobile devices, supporting real-time agricultural decision-making in resource-constrained environments. These results demonstrate significant potential for AI-driven precision agriculture tools that can operate reliably in diverse field conditions.

Paperid: 6483, https://arxiv.org/pdf/2507.07148.pdf

Abstract:
Explainable artificial intelligence (XAI) has become increasingly important in biomedical image analysis to promote transparency, trust, and clinical adoption of DL models. While several surveys have reviewed XAI techniques, they often lack a modality-aware perspective, overlook recent advances in multimodal and vision-language paradigms, and provide limited practical guidance. This survey addresses this gap through a comprehensive and structured synthesis of XAI methods tailored to biomedical image analysis.We systematically categorize XAI methods, analyzing their underlying principles, strengths, and limitations within biomedical contexts. A modality-centered taxonomy is proposed to align XAI methods with specific imaging types, highlighting the distinct interpretability challenges across modalities. We further examine the emerging role of multimodal learning and vision-language models in explainable biomedical AI, a topic largely underexplored in previous work. Our contributions also include a summary of widely used evaluation metrics and open-source frameworks, along with a critical discussion of persistent challenges and future directions. This survey offers a timely and in-depth foundation for advancing interpretable DL in biomedical image analysis.

Paperid: 6484, https://arxiv.org/pdf/2507.00052.pdf

Abstract:
Vision Language Models (VLMs) hold great promise for streamlining labour-intensive medical imaging workflows, yet systematic security evaluations in clinical settings remain scarce. We introduce VSF--Med, an end-to-end vulnerability-scoring framework for medical VLMs that unites three novel components: (i) a rich library of sophisticated text-prompt attack templates targeting emerging threat vectors; (ii) imperceptible visual perturbations calibrated by structural similarity (SSIM) thresholds to preserve clinical realism; and (iii) an eight-dimensional rubric evaluated by two independent judge LLMs, whose raw scores are consolidated via z-score normalization to yield a 0--32 composite risk metric. Built entirely on publicly available datasets and accompanied by open-source code, VSF--Med synthesizes over 30,000 adversarial variants from 5,000 radiology images and enables reproducible benchmarking of any medical VLM with a single command. Our consolidated analysis reports mean z-score shifts of $0.90Ï$ for persistence-of-attack-effects, $0.74Ï$ for prompt-injection effectiveness, and $0.63Ï$ for safety-bypass success across state-of-the-art VLMs. Notably, Llama-3.2-11B-Vision-Instruct exhibits a peak vulnerability increase of $1.29Ï$ for persistence-of-attack-effects, while GPT-4o shows increases of $0.69Ï$ for that same vector and $0.28Ï$ for prompt-injection attacks.

Paperid: 6485, https://arxiv.org/pdf/2506.22802.pdf

Abstract:
Recent breakthroughs and rapid integration of generative models (GMs) have sparked interest in the problem of model attribution and their fingerprints. For instance, service providers need reliable methods of authenticating their models to protect their IP, while users and law enforcement seek to verify the source of generated content for accountability and trust. In addition, a growing threat of model collapse is arising, as more model-generated data are being fed back into sources (e.g., YouTube) that are often harvested for training ("regurgitative training"), heightening the need to differentiate synthetic from human data. Yet, a gap still exists in understanding generative models' fingerprints, we believe, stemming from the lack of a formal framework that can define, represent, and analyze the fingerprints in a principled way. To address this gap, we take a geometric approach and propose a new definition of artifact and fingerprint of GMs using Riemannian geometry, which allows us to leverage the rich theory of differential geometry. Our new definition generalizes previous work (Song et al., 2024) to non-Euclidean manifolds by learning Riemannian metrics from data and replacing the Euclidean distances and nearest-neighbor search with geodesic distances and kNN-based Riemannian center of mass. We apply our theory to a new gradient-based algorithm for computing the fingerprints in practice. Results show that it is more effective in distinguishing a large array of GMs, spanning across 4 different datasets in 2 different resolutions (64 by 64, 256 by 256), 27 model architectures, and 2 modalities (Vision, Vision-Language). Using our proposed definition significantly improves the performance on model attribution, as well as a generalization to unseen datasets, model types, and modalities, suggesting its practical efficacy.

Paperid: 6486, https://arxiv.org/pdf/2506.22056.pdf

Abstract:
Trajectory data, capturing human actions and environmental states across various modalities, holds significant potential for enhancing AI agent capabilities, particularly in GUI environments. However, how to model the representation of trajectory-level data presents a significant challenge that has not been systematically addressed amid explosive trajectory data growth. In this work, we introduce Multimodal Trajectory Retrieval, bridging the gap between universal retrieval and agent-centric trajectory modeling. We construct the Unified Agent Trajectory Dataset (UATD) from annotated demonstrations and states across diverse real-world scenarios. Based on this, we present GAE-Bench, a benchmark containing a large number of trajectory-based retrieval pairs. In addition, we propose GAE-Retriever, a multimodal retrieval framework that adopts vision-language models and incorporates optimized contrastive learning through a token selection and the GradCache mechanism. Comprehensive evaluations across multiple datasets show that GAE-Retriever consistently outperforms strong baselines in retrieval recall, highlighting its effectiveness in advancing multimodal trajectory retrieval.

Paperid: 6487, https://arxiv.org/pdf/2506.21627.pdf

Abstract:
Developing a general robot manipulation system capable of performing a wide range of tasks in complex, dynamic, and unstructured real-world environments has long been a challenging task. It is widely recognized that achieving human-like efficiency and robustness manipulation requires the robotic brain to integrate a comprehensive set of functions, such as task planning, policy generation, anomaly monitoring and handling, and long-term memory, achieving high-efficiency operation across all functions. Vision-Language Models (VLMs), pretrained on massive multimodal data, have acquired rich world knowledge, exhibiting exceptional scene understanding and multimodal reasoning capabilities. However, existing methods typically focus on realizing only a single function or a subset of functions within the robotic brain, without integrating them into a unified cognitive architecture. Inspired by a divide-and-conquer strategy and the architecture of the human brain, we propose FrankenBot, a VLM-driven, brain-morphic robotic manipulation framework that achieves both comprehensive functionality and high operational efficiency. Our framework includes a suite of components, decoupling a part of key functions from frequent VLM calls, striking an optimal balance between functional completeness and system efficiency. Specifically, we map task planning, policy generation, memory management, and low-level interfacing to the cortex, cerebellum, temporal lobe-hippocampus complex, and brainstem, respectively, and design efficient coordination mechanisms for the modules. We conducted comprehensive experiments in both simulation and real-world robotic environments, demonstrating that our method offers significant advantages in anomaly detection and handling, long-term memory, operational efficiency, and stability -- all without requiring any fine-tuning or retraining.

Paperid: 6488, https://arxiv.org/pdf/2506.20832.pdf

Abstract:
Super-resolution (SR) is an ill-posed inverse problem with many feasible solutions consistent with a given low-resolution image. On one hand, regressive SR models aim to balance fidelity and perceptual quality to yield a single solution, but this trade-off often introduces artifacts that create ambiguity in information-critical applications such as recognizing digits or letters. On the other hand, diffusion models generate a diverse set of SR images, but selecting the most trustworthy solution from this set remains a challenge. This paper introduces a robust, automated framework for identifying the most trustworthy SR sample from a diffusion-generated set by leveraging the semantic reasoning capabilities of vision-language models (VLMs). Specifically, VLMs such as BLIP-2, GPT-4o, and their variants are prompted with structured queries to assess semantic correctness, visual quality, and artifact presence. The top-ranked SR candidates are then ensembled to yield a single trustworthy output in a cost-effective manner. To rigorously assess the validity of VLM-selected samples, we propose a novel Trustworthiness Score (TWS) a hybrid metric that quantifies SR reliability based on three complementary components: semantic similarity via CLIP embeddings, structural integrity using SSIM on edge maps, and artifact sensitivity through multi-level wavelet decomposition. We empirically show that TWS correlates strongly with human preference in both ambiguous and natural images, and that VLM-guided selections consistently yield high TWS values. Compared to conventional metrics like PSNR, LPIPS, which fail to reflect information fidelity, our approach offers a principled, scalable, and generalizable solution for navigating the uncertainty of the diffusion SR space. By aligning outputs with human expectations and semantic correctness, this work sets a new benchmark for trustworthiness in generative SR.

Paperid: 6489, https://arxiv.org/pdf/2506.19651.pdf

Abstract:
Vision-Language Models (VLMs) have demonstrated strong capabilities in multimodal understanding and generation tasks. However, their application to long video understanding remains hindered by the quadratic complexity of standard attention mechanisms. In this work, we introduce \textbf{PEVLM}, a fine-tuning-free parallel encoding method designed to enhance the prefilling efficiency of VLMs in long video scenarios. PEVLM partitions the input video into context blocks with a shared sink block, while preserving sequential position embeddings to align the attention weight distribution with that of Full-Attention. This design reduces attention complexity from $O((T \times N)^2)$ to $O(T \times N)$ where $T$ is the number of frames and $N$ the number of tokens per frame, without sacrificing accuracy. Extensive experiments across multiple state-of-the-art models and benchmarks demonstrate that PEVLM consistently outperforms existing parallel encoding approaches, achieving up to \textbf{7.47x} speedup in attention computation and reducing end-to-end latency by \textbf{40\%}. Remarkably, PEVLM not only maintains high accuracy, but in some settings even surpasses Full-Attention performance. Under strict latency constraints, it achieves substantial gains, improving accuracy from \textbf{23.26\%} to \textbf{61.03\%}. These results underscore the effectiveness of PEVLM for low-latency, long-context video understanding, making it a promising solution for real-world applications.

Paperid: 6490, https://arxiv.org/pdf/2506.19498.pdf

Abstract:
Building a general robotic manipulation system capable of performing a wide variety of tasks in real-world settings is a challenging task. Vision-Language Models (VLMs) have demonstrated remarkable potential in robotic manipulation tasks, primarily due to the extensive world knowledge they gain from large-scale datasets. In this process, Spatial Representations (such as points representing object positions or vectors representing object orientations) act as a bridge between VLMs and real-world scene, effectively grounding the reasoning abilities of VLMs and applying them to specific task scenarios. However, existing VLM-based robotic approaches often adopt a fixed spatial representation extraction scheme for various tasks, resulting in insufficient representational capability or excessive extraction time. In this work, we introduce T-Rex, a Task-Adaptive Framework for Spatial Representation Extraction, which dynamically selects the most appropriate spatial representation extraction scheme for each entity based on specific task requirements. Our key insight is that task complexity determines the types and granularity of spatial representations, and Stronger representational capabilities are typically associated with Higher overall system operation costs. Through comprehensive experiments in real-world robotic environments, we show that our approach delivers significant advantages in spatial understanding, efficiency, and stability without additional training.

Paperid: 6491, https://arxiv.org/pdf/2506.15871.pdf

Abstract:
To accurately process a visual scene, observers must bind features together to represent individual objects. This capacity is necessary, for instance, to distinguish an image containing a red square and a blue circle from an image containing a blue square and a red circle. Recent work has found that language models solve this 'binding problem' via a set of symbol-like, content-independent indices, but it is unclear whether similar mechanisms are employed by vision language models (VLMs). This question is especially relevant, given the persistent failures of VLMs on tasks that require binding. Here, we identify a set of emergent symbolic mechanisms that support binding in VLMs via a content-independent, spatial indexing scheme. Moreover, we find that binding errors can be traced directly to failures in these mechanisms. Taken together, these results shed light on the mechanisms that support symbol-like processing in VLMs, and suggest possible avenues for addressing the persistent binding failures exhibited by these models.

Paperid: 6492, https://arxiv.org/pdf/2506.15201.pdf

Abstract:
The improved semantic understanding of vision-language pretrained (VLP) models has made it increasingly difficult to protect publicly posted images from being exploited by search engines and other similar tools. In this context, this paper seeks to protect users' privacy by implementing defenses at the image compression stage to prevent exploitation. Specifically, we propose a flexible coding method, termed Privacy-Shielded Image Compression (PSIC), that can produce bitstreams with multiple decoding options. By default, the bitstream is decoded to preserve satisfactory perceptual quality while preventing interpretation by VLP models. Our method also retains the original image compression functionality. With a customizable input condition, the proposed scheme can reconstruct the image that preserves its full semantic information. A Conditional Latent Trigger Generation (CLTG) module is proposed to produce bias information based on customizable conditions to guide the decoding process into different reconstructed versions, and an Uncertainty-Aware Encryption-Oriented (UAEO) optimization function is designed to leverage the soft labels inferred from the target VLP model's uncertainty on the training data. This paper further incorporates an adaptive multi-objective optimization strategy to obtain improved encrypting performance and perceptual quality simultaneously within a unified training process. The proposed scheme is plug-and-play and can be seamlessly integrated into most existing Learned Image Compression (LIC) models. Extensive experiments across multiple downstream tasks have demonstrated the effectiveness of our design.

Paperid: 6493, https://arxiv.org/pdf/2506.14404.pdf

Abstract:
Adapting text-to-image (T2I) latent diffusion models (LDMs) to video editing has shown strong visual fidelity and controllability, but challenges remain in maintaining causal relationships inherent to the video data generating process. Edits affecting causally dependent attributes often generate unrealistic or misleading outcomes if these relationships are ignored. In this work, we introduce a causally faithful framework for counterfactual video generation, formulated as an Out-of-Distribution (OOD) prediction problem. We embed prior causal knowledge by encoding the relationships specified in a causal graph into text prompts and guide the generation process by optimizing these prompts using a vision-language model (VLM)-based textual loss. This loss encourages the latent space of the LDMs to capture OOD variations in the form of counterfactuals, effectively steering generation toward causally meaningful alternatives. The proposed framework, dubbed CSVC, is agnostic to the underlying video editing system and does not require access to its internal mechanisms or fine-tuning. We evaluate our approach using standard video quality metrics and counterfactual-specific criteria, such as causal effectiveness and minimality. Experimental results show that CSVC generates causally faithful video counterfactuals within the LDM distribution via prompt-based causal steering, achieving state-of-the-art causal effectiveness without compromising temporal consistency or visual quality on real-world facial videos. Due to its compatibility with any black-box video editing system, our framework has significant potential to generate realistic 'what if' hypothetical video scenarios in diverse areas such as digital media and healthcare.

Paperid: 6494, https://arxiv.org/pdf/2506.13102.pdf

Abstract:
Test-time scaling has recently emerged as a promising approach for enhancing the reasoning capabilities of large language models or vision-language models during inference. Although a variety of test-time scaling strategies have been proposed, and interest in their application to the medical domain is growing, many critical aspects remain underexplored, including their effectiveness for vision-language models and the identification of optimal strategies for different settings. In this paper, we conduct a comprehensive investigation of test-time scaling in the medical domain. We evaluate its impact on both large language models and vision-language models, considering factors such as model size, inherent model characteristics, and task complexity. Finally, we assess the robustness of these strategies under user-driven factors, such as misleading information embedded in prompts. Our findings offer practical guidelines for the effective use of test-time scaling in medical applications and provide insights into how these strategies can be further refined to meet the reliability and interpretability demands of the medical domain.

Paperid: 6495, https://arxiv.org/pdf/2506.12724.pdf

Abstract:
Multimodal Large Models (MLLMs) have achieved remarkable progress in vision-language understanding and generation tasks. However, existing MLLMs typically rely on static modality fusion strategies, which treat all modalities equally regardless of their instance-level reliability or semantic contribution. This often leads to suboptimal performance, especially in scenarios with noisy, missing, or misaligned modalities. In this paper, we propose Dynamic Modality Scheduling (DMS), a novel framework that adaptively adjusts the contribution of each modality at a per-sample level. DMS evaluates each modality based on three key factors: (1) \textit{confidence}, estimated from predictive entropy; (2) \textit{uncertainty}, obtained via Monte Carlo dropout; and (3) \textit{semantic consistency}, computed through inter-modal similarity. These signals are combined through a learnable or rule-based scheduler to generate soft modality weights used in downstream fusion.To ensure stable training, we further introduce a \textit{Modality Weight Consistency Loss}, which regularizes the fused representation to stay close to unimodal embeddings proportionally to their assigned weights. Our method is model-agnostic and can be integrated into existing MLLMs such as BLIP-2 and LLaVA. Experimental results on VQA, image-text retrieval, and captioning tasks show that DMS significantly improves both clean and robust performance, especially under modality corruption or dropout conditions. This work provides a general and effective mechanism to enable instance-aware and robustness-enhanced multimodal modeling.

Paperid: 6496, https://arxiv.org/pdf/2506.12374.pdf

Abstract:
Vision-Language Models (VLMs) encode knowledge and reasoning capabilities for robotic manipulation within high-dimensional representation spaces. However, current approaches often project them into compressed intermediate representations, discarding important task-specific information such as fine-grained spatial or semantic details. To address this, we propose AntiGrounding, a new framework that reverses the instruction grounding process. It lifts candidate actions directly into the VLM representation space, renders trajectories from multiple views, and uses structured visual question answering for instruction-based decision making. This enables zero-shot synthesis of optimal closed-loop robot trajectories for new tasks. We also propose an offline policy refinement module that leverages past experience to enhance long-term performance. Experiments in both simulation and real-world environments show that our method outperforms baselines across diverse robotic manipulation tasks.

Paperid: 6497, https://arxiv.org/pdf/2506.12116.pdf

Abstract:
We study unsupervised clustering of documents at both the category and template levels using frozen multimodal encoders and classical clustering algorithms. We systematize a model-agnostic pipeline that (i) projects heterogeneous last-layer states from text-layout-vision encoders into token-type-aware document vectors and (ii) performs clustering with centroid- or density-based methods, including an HDBSCAN + $k$-NN assignment to eliminate unlabeled points. We evaluate eight encoders (text-only, layout-aware, vision-only, and vision-language) with $k$-Means, DBSCAN, HDBSCAN + $k$-NN, and BIRCH on five corpora spanning clean synthetic invoices, their heavily degraded print-and-scan counterparts, scanned receipts, and real identity and certificate documents. The study reveals modality-specific failure modes and a robustness-accuracy trade-off, with vision features nearly solving template discovery on clean pages while text dominates under covariate shift, and fused encoders offering the best balance. We detail a reproducible, oracle-free tuning protocol and the curated evaluation settings to guide future work on unsupervised document organization.

Paperid: 6498, https://arxiv.org/pdf/2506.08194.pdf

Abstract:
Monocular 3D reconstruction methods and vision-language models (VLMs) demonstrate impressive results on standard benchmarks, yet their true understanding of geometric properties remains unclear. We introduce GIQ , a comprehensive benchmark specifically designed to evaluate the geometric reasoning capabilities of vision and vision-language foundation models. GIQ comprises synthetic and real-world images of 224 diverse polyhedra - including Platonic, Archimedean, Johnson, and Catalan solids, as well as stellations and compound shapes - covering varying levels of complexity and symmetry. Through systematic experiments involving monocular 3D reconstruction, 3D symmetry detection, mental rotation tests, and zero-shot shape classification tasks, we reveal significant shortcomings in current models. State-of-the-art reconstruction algorithms trained on extensive 3D datasets struggle to reconstruct even basic geometric forms accurately. While foundation models effectively detect specific 3D symmetry elements via linear probing, they falter significantly in tasks requiring detailed geometric differentiation, such as mental rotation. Moreover, advanced vision-language assistants exhibit remarkably low accuracy on complex polyhedra, systematically misinterpreting basic properties like face geometry, convexity, and compound structures. GIQ is publicly available, providing a structured platform to highlight and address critical gaps in geometric intelligence, facilitating future progress in robust, geometry-aware representation learning.

Paperid: 6499, https://arxiv.org/pdf/2506.06836.pdf

Abstract:
Time-series anomaly detection (TSAD) has played a vital role in a variety of fields, including healthcare, finance, and industrial monitoring. Prior methods, which mainly focus on training domain-specific models on numerical data, lack the visual-temporal reasoning capacity that human experts have to identify contextual anomalies. To fill this gap, we explore a solution based on vision language models (VLMs). Recent studies have shown the ability of VLMs for visual reasoning tasks, yet their direct application to time series has fallen short on both accuracy and efficiency. To harness the power of VLMs for TSAD, we propose a two-stage solution, with (1) ViT4TS, a vision-screening stage built on a relatively lightweight pretrained vision encoder, which leverages 2-D time-series representations to accurately localize candidate anomalies; (2) VLM4TS, a VLM-based stage that integrates global temporal context and VLM reasoning capacity to refine the detection upon the candidates provided by ViT4TS. We show that without any time-series training, VLM4TS outperforms time-series pretrained and from-scratch baselines in most cases, yielding a 24.6 percent improvement in F1-max score over the best baseline. Moreover, VLM4TS also consistently outperforms existing language-model-based TSAD methods and is on average 36 times more efficient in token usage.

Paperid: 6500, https://arxiv.org/pdf/2506.06602.pdf

Abstract:
Composed image retrieval (CIR) allows a user to locate a target image by applying a fine-grained textual edit (e.g., ``turn the dress blue'' or ``remove stripes'') to a reference image. Zero-shot CIR, which embeds the image and the text with separate pretrained vision-language encoders, reaches only 20-25\% Recall@10 on the FashionIQ benchmark. We improve this by fine-tuning BLIP-2 with a lightweight Q-Former that fuses visual and textual features into a single embedding, raising Recall@10 to 45.6\% (shirt), 40.1\% (dress), and 50.4\% (top-tee) and increasing the average Recall@50 to 67.6\%. We also examine Retrieval-DPO, which fine-tunes CLIP's text encoder with a Direct Preference Optimization loss applied to FAISS-mined hard negatives. Despite extensive tuning of the scaling factor, index, and sampling strategy, Retrieval-DPO attains only 0.02\% Recall@10 -- far below zero-shot and prompt-tuned baselines -- because it (i) lacks joint image-text fusion, (ii) uses a margin objective misaligned with top-$K$ metrics, (iii) relies on low-quality negatives, and (iv) keeps the vision and Transformer layers frozen. Our results show that effective preference-based CIR requires genuine multimodal fusion, ranking-aware objectives, and carefully curated negatives.

Paperid: 6501, https://arxiv.org/pdf/2506.04129.pdf

Abstract:
Medical image classification is crucial for diagnosis and treatment, benefiting significantly from advancements in artificial intelligence. The paper reviews recent progress in the field, focusing on three levels of solutions: basic, specific, and applied. It highlights advances in traditional methods using deep learning models like Convolutional Neural Networks and Vision Transformers, as well as state-of-the-art approaches with Vision Language Models. These models tackle the issue of limited labeled data, and enhance and explain predictive results through Explainable Artificial Intelligence.

Paperid: 6502, https://arxiv.org/pdf/2506.03926.pdf

Abstract:
Foundation Vision-Language Models (VLMs) like CLIP exhibit strong generalization capabilities due to large-scale pretraining on diverse image-text pairs. However, their performance often degrades when applied to target datasets with significant distribution shifts in both visual appearance and class semantics. Recent few-shot learning approaches adapt CLIP to downstream tasks using limited labeled data via adapter or prompt tuning, but are not specifically designed to handle such extreme domain shifts. Conversely, some works addressing cross-domain few-shot learning consider such domain-shifted scenarios but operate in an episodic setting with only a few classes per episode, limiting their applicability to real-world deployment, where all classes must be handled simultaneously. To address this gap, we propose a novel framework, MIST (Multiple Stochastic Prompt Tuning), for efficiently adapting CLIP to datasets with extreme distribution shifts using only a few labeled examples, in scenarios involving all classes at once. Specifically, we introduce multiple learnable prompts per class to effectively capture diverse modes in visual representations arising from distribution shifts. To further enhance generalization, these prompts are modeled as learnable Gaussian distributions, enabling efficient exploration of the prompt parameter space and reducing overfitting caused by limited supervision. Extensive experiments and comparisons with state-of-the-art methods demonstrate the effectiveness of the proposed framework.

Paperid: 6503, https://arxiv.org/pdf/2506.02095.pdf

Abstract:
Learning alignment between language and vision is a fundamental challenge, especially as multimodal data becomes increasingly detailed and complex. Existing methods often rely on collecting human or AI preferences, which can be costly and time-intensive. We propose an alternative approach that leverages cycle consistency as a supervisory signal. Given an image and generated text, we map the text back to image space using a text-to-image model and compute the similarity between the original image and its reconstruction. Analogously, for text-to-image generation, we measure the textual similarity between an input caption and its reconstruction through the cycle. We use the cycle consistency score to rank candidates and construct a preference dataset of 866K comparison pairs. The reward model trained on our dataset outperforms state-of-the-art alignment metrics on detailed captioning, with superior inference-time scalability when used as a verifier for Best-of-N sampling. Furthermore, performing DPO and Diffusion DPO using our dataset enhances performance across a wide range of vision-language tasks and text-to-image generation. Our dataset, model, and code are at https://cyclereward.github.io

Paperid: 6504, https://arxiv.org/pdf/2506.00869.pdf

Abstract:
Despite the impressive performance of vision-language models (VLMs) on downstream tasks, their ability to understand and reason about causal relationships in visual inputs remains unclear. Robust causal reasoning is fundamental to solving complex high-level reasoning tasks, yet existing benchmarks often include a mixture of reasoning questions, and VLMs can frequently exploit object recognition and activity identification as shortcuts to arrive at the correct answers, making it challenging to truly assess their causal reasoning abilities. To bridge this gap, we introduce VQA-Causal and VCR-Causal, two new benchmarks specifically designed to isolate and rigorously evaluate VLMs' causal reasoning abilities. Our findings reveal that while VLMs excel in object and activity recognition, they perform poorly on causal reasoning tasks, often only marginally surpassing random guessing. Further analysis suggests that this limitation stems from a severe lack of causal expressions in widely used training datasets, where causal relationships are rarely explicitly conveyed. We additionally explore fine-tuning strategies with hard negative cases, showing that targeted fine-tuning can improve model's causal reasoning while maintaining generalization and downstream performance. Our study highlights a key gap in current VLMs and lays the groundwork for future work on causal understanding.

Paperid: 6505, https://arxiv.org/pdf/2505.21649.pdf

Abstract:
Object orientation understanding represents a fundamental challenge in visual perception critical for applications like robotic manipulation and augmented reality. Current vision-language benchmarks fail to isolate this capability, often conflating it with positional relationships and general scene understanding. We introduce DORI (Discriminative Orientation Reasoning Intelligence), a comprehensive benchmark establishing object orientation perception as a primary evaluation target. DORI assesses four dimensions of orientation comprehension: frontal alignment, rotational transformations, relative directional relationships, and canonical orientation understanding. Through carefully curated tasks from 11 datasets spanning 67 object categories across synthetic and real-world scenarios, DORI provides insights on how multi-modal systems understand object orientations. Our evaluation of 15 state-of-the-art vision-language models reveals critical limitations: even the best models achieve only 54.2% accuracy on coarse tasks and 33.0% on granular orientation judgments, with performance deteriorating for tasks requiring reference frame shifts or compound rotations. These findings demonstrate the need for dedicated orientation representation mechanisms, as models show systematic inability to perform precise angular estimations, track orientation changes across viewpoints, and understand compound rotations - suggesting limitations in their internal 3D spatial representations. As the first diagnostic framework specifically designed for orientation awareness in multimodal systems, DORI offers implications for improving robotic control, 3D scene reconstruction, and human-AI interaction in physical environments. DORI data: https://huggingface.co/datasets/appledora/DORI-Benchmark

Paperid: 6506, https://arxiv.org/pdf/2505.21538.pdf

Abstract:
Vision-Language Models (VLMs) have shown remarkable progress in visual understanding in recent years. Yet, they still lag behind human capabilities in specific visual tasks such as counting or relational reasoning. To understand the underlying limitations, we adopt methodologies from cognitive science, analyzing VLM performance along core cognitive axes: Perception, Attention, and Memory. Using a suite of tasks targeting these abilities, we evaluate state-of-the-art VLMs, including GPT-4o. Our analysis reveals distinct cognitive profiles: while advanced models approach ceiling performance on some tasks (e.g. category identification), a significant gap persists, particularly in tasks requiring spatial understanding or selective attention. Investigating the source of these failures and potential methods for improvement, we employ a vision-text decoupling analysis, finding that models struggling with direct visual reasoning show marked improvement when reasoning over their own generated text captions. These experiments reveal a strong need for improved VLM Chain-of-Thought (CoT) abilities, even in models that consistently exceed human performance. Furthermore, we demonstrate the potential of targeted fine-tuning on composite visual reasoning tasks and show that fine-tuning smaller VLMs substantially improves core cognitive abilities. While this improvement does not translate to large enhancements on challenging, out-of-distribution benchmarks, we show broadly that VLM performance on our datasets strongly correlates with performance on these other benchmarks. Our work provides a detailed analysis of VLM cognitive strengths and weaknesses and identifies key bottlenecks in simultaneous perception and reasoning while also providing an effective and simple solution.

Paperid: 6507, https://arxiv.org/pdf/2505.17702.pdf

Abstract:
The advent of Computer-Aided Design (CAD) generative modeling will significantly transform the design of industrial products. The recent research endeavor has extended into the realm of Large Language Models (LLMs). In contrast to fine-tuning methods, training-free approaches typically utilize the advanced closed-source LLMs, thereby offering enhanced flexibility and efficiency in the development of AI agents for generating CAD parametric models. However, the substantial cost and limitations of local deployment of the top-tier closed-source LLMs pose challenges in practical applications. The Seek-CAD is the pioneer exploration of locally deployed open-source inference LLM DeepSeek-R1 for CAD parametric model generation with a training-free methodology. This study is the first investigation to incorporate both visual and Chain-of-Thought (CoT) feedback within the self-refinement mechanism for generating CAD models. Specifically, the initial generated parametric CAD model is rendered into a sequence of step-wise perspective images, which are subsequently processed by a Vision Language Model (VLM) alongside the corresponding CoTs derived from DeepSeek-R1 to assess the CAD model generation. Then, the feedback is utilized by DeepSeek-R1 to refine the initial generated model for the next round of generation. Moreover, we present an innovative 3D CAD model dataset structured around the SSR (Sketch, Sketch-based feature, and Refinements) triple design paradigm. This dataset encompasses a wide range of CAD commands, thereby aligning effectively with industrial application requirements and proving suitable for the generation of LLMs. Extensive experiments validate the effectiveness of Seek-CAD under various metrics.

Paperid: 6508, https://arxiv.org/pdf/2505.15564.pdf

Abstract:
Vision Language Models (VLMs) employed for visual question-answering (VQA) in autonomous driving often require substantial computational resources that pose a challenge for their deployment in resource-constrained vehicles. To address this challenge, we introduce TinyDrive, a lightweight yet effective VLM for multi-view VQA in driving scenarios. Our model comprises two key components including a multiscale vision encoder and a dual-level prioritization mechanism for tokens and sequences. The multiscale encoder facilitates the processing of multi-view images at diverse resolutions through scale injection and cross-scale gating to generate enhanced visual representations. At the token level, we design a token routing mechanism that dynamically selects and process the most informative tokens based on learned importance scores. At the sequence level, we propose integrating normalized loss, uncertainty estimates, and a diversity metric to formulate sequence scores that rank and preserve samples within a sequence priority buffer. Samples with higher scores are more frequently selected for training. TinyDrive is first evaluated on our custom-curated VQA dataset, and it is subsequently tested on the public DriveLM benchmark, where it achieves state-of-the-art language understanding performance. Notably, it achieves relative improvements of 11.1% and 35.4% in BLEU-4 and METEOR scores, respectively, despite having a significantly smaller parameter count.

Paperid: 6509, https://arxiv.org/pdf/2505.15506.pdf

Abstract:
Recently, Vision-Language foundation models like CLIP and ALIGN, which are pre-trained on large-scale data have shown remarkable zero-shot generalization to diverse datasets with different classes and even domains. In this work, we take a step further and analyze whether these models can be adapted to target datasets having very different distributions and classes compared to what these models have been trained on, using only a few labeled examples from the target dataset. In such scenarios, finetuning large pretrained models is challenging due to problems of overfitting as well as loss of generalization, and has not been well explored in prior literature. Since, the pre-training data of such models are unavailable, it is difficult to comprehend the performance on various downstream datasets. First, we try to answer the question: Given a target dataset with a few labelled examples, can we estimate whether further fine-tuning can enhance the performance compared to zero-shot evaluation? by analyzing the common vision-language embedding space. Based on the analysis, we propose a novel prompt-tuning method, PromptMargin for adapting such large-scale VLMs directly on the few target samples. PromptMargin effectively tunes the text as well as visual prompts for this task, and has two main modules: 1) Firstly, we use a selective augmentation strategy to complement the few training samples in each task; 2) Additionally, to ensure robust training in the presence of unfamiliar class names, we increase the inter-class margin for improved class discrimination using a novel Multimodal Margin Regularizer. Extensive experiments and analysis across fifteen target benchmark datasets, with varying degrees of distribution shifts from natural images, shows the effectiveness of the proposed framework over the existing state-of-the-art approaches applied to this setting. github.com/debarshigit/PromptMargin.

Paperid: 6510, https://arxiv.org/pdf/2505.15265.pdf

Abstract:
Adversarial attacks aim to generate malicious inputs that mislead deep models, but beyond causing model failure, they cannot provide certain interpretable information such as ``\textit{What content in inputs make models more likely to fail?}'' However, this information is crucial for researchers to specifically improve model robustness. Recent research suggests that models may be particularly sensitive to certain semantics in visual inputs (such as ``wet,'' ``foggy''), making them prone to errors. Inspired by this, in this paper we conducted the first exploration on large vision-language models (LVLMs) and found that LVLMs indeed are susceptible to hallucinations and various errors when facing specific semantic concepts in images. To efficiently search for these sensitive concepts, we integrated large language models (LLMs) and text-to-image (T2I) models to propose a novel semantic evolution framework. Randomly initialized semantic concepts undergo LLM-based crossover and mutation operations to form image descriptions, which are then converted by T2I models into visual inputs for LVLMs. The task-specific performance of LVLMs on each input is quantified as fitness scores for the involved semantics and serves as reward signals to further guide LLMs in exploring concepts that induce LVLMs. Extensive experiments on seven mainstream LVLMs and two multimodal tasks demonstrate the effectiveness of our method. Additionally, we provide interesting findings about the sensitive semantics of LVLMs, aiming to inspire further in-depth research.

Paperid: 6511, https://arxiv.org/pdf/2505.12660.pdf

Abstract:
Although models exist that predict human response times (RTs) in tasks such as target search and visual discrimination, the development of image-computable predictors for scene understanding time remains an open challenge. Recent advances in vision-language models (VLMs), which can generate scene descriptions for arbitrary images, combined with the availability of quantitative metrics for comparing linguistic descriptions, offer a new opportunity to model human scene understanding. We hypothesize that the primary bottleneck in human scene understanding and the driving source of variability in response times across scenes is the interaction between the foveated nature of the human visual system and the spatial distribution of task-relevant visual information within an image. Based on this assumption, we propose a novel image-computable model that integrates foveated vision with VLMs to produce a spatially resolved map of scene understanding as a function of fixation location (Foveated Scene Understanding Map, or F-SUM), along with an aggregate F-SUM score. This metric correlates with average (N=17) human RTs (r=0.47) and number of saccades (r=0.51) required to comprehend a scene (across 277 scenes). The F-SUM score also correlates with average (N=16) human description accuracy (r=-0.56) in time-limited presentations. These correlations significantly exceed those of standard image-based metrics such as clutter, visual complexity, and scene ambiguity based on language entropy. Together, our work introduces a new image-computable metric for predicting human response times in scene understanding and demonstrates the importance of foveated visual processing in shaping comprehension difficulty.

Paperid: 6512, https://arxiv.org/pdf/2505.12589.pdf

Abstract:
Understanding surveillance video content remains a critical yet underexplored challenge in vision-language research, particularly due to its real-world complexity, irregular event dynamics, and safety-critical implications. In this work, we introduce SurveillanceVQA-589K, the largest open-ended video question answering benchmark tailored to the surveillance domain. The dataset comprises 589,380 QA pairs spanning 12 cognitively diverse question types, including temporal reasoning, causal inference, spatial understanding, and anomaly interpretation, across both normal and abnormal video scenarios. To construct the benchmark at scale, we design a hybrid annotation pipeline that combines temporally aligned human-written captions with Large Vision-Language Model-assisted QA generation using prompt-based techniques. We also propose a multi-dimensional evaluation protocol to assess contextual, temporal, and causal comprehension. We evaluate eight LVLMs under this framework, revealing significant performance gaps, especially in causal and anomaly-related tasks, underscoring the limitations of current models in real-world surveillance contexts. Our benchmark provides a practical and comprehensive resource for advancing video-language understanding in safety-critical applications such as intelligent monitoring, incident analysis, and autonomous decision-making.

Paperid: 6513, https://arxiv.org/pdf/2505.12359.pdf

Abstract:
Although large vision-language models (LVLMs) leverage rich visual token representations to achieve strong performance on multimodal tasks, these tokens also introduce significant computational overhead during inference. Existing training-free token pruning methods typically adopt a single-stage strategy, focusing either on visual self-attention or visual-textual cross-attention. However, such localized perspectives often overlook the broader information flow across the model, leading to substantial performance degradation, especially under high pruning ratios. In this work, we propose STAR (Stage-wise Attention-guided token Reduction), a training-free, plug-and-play framework that approaches token pruning from a global perspective. Instead of pruning at a single point, STAR performs attention-guided reduction in two complementary stages: an early-stage pruning based on visual self-attention to remove redundant low-level features, and a later-stage pruning guided by cross-modal attention to discard task-irrelevant tokens. This holistic approach allows STAR to significantly reduce computational cost while better preserving task-critical information. Extensive experiments across multiple LVLM architectures and benchmarks show that STAR achieves strong acceleration while maintaining comparable, and in some cases even improved performance.

Paperid: 6514, https://arxiv.org/pdf/2505.12343.pdf

Abstract:
Despite the impressive capabilities of Large Vision-Language Models (LVLMs), they remain susceptible to hallucinations-generating content that is inconsistent with the input image. Existing training-free hallucination mitigation methods often suffer from unstable performance and high sensitivity to hyperparameter settings, limiting their practicality and broader adoption. In this paper, we propose a novel decoding mechanism, Decoding with Inter-layer Consistency via Layer Aggregation (DCLA), which requires no retraining, fine-tuning, or access to external knowledge bases. Specifically, our approach constructs a dynamic semantic reference by aggregating representations from previous layers, and corrects semantically deviated layers to enforce inter-layer consistency. The method allows DCLA to robustly mitigate hallucinations across multiple LVLMs. Experiments on hallucination benchmarks such as MME and POPE demonstrate that DCLA effectively reduces hallucinations while enhancing the reliability and performance of LVLMs.

Paperid: 6515, https://arxiv.org/pdf/2505.12217.pdf

Abstract:
We introduce HyperCap, the first large-scale hyperspectral captioning dataset designed to enhance model performance and effectiveness in remote sensing applications. Unlike traditional hyperspectral imaging (HSI) datasets that focus solely on classification tasks, HyperCap integrates spectral data with pixel-wise textual annotations, enabling deeper semantic understanding of hyperspectral imagery. This dataset enhances model performance in tasks like classification and feature extraction, providing a valuable resource for advanced remote sensing applications. HyperCap is constructed from four benchmark datasets and annotated through a hybrid approach combining automated and manual methods to ensure accuracy and consistency. Empirical evaluations using state-of-the-art encoders and diverse fusion techniques demonstrate significant improvements in classification performance. These results underscore the potential of vision-language learning in HSI and position HyperCap as a foundational dataset for future research in the field.

Paperid: 6516, https://arxiv.org/pdf/2505.10583.pdf

Abstract:
Large language models have become multimodal, and many of them are said to integrate their modalities using common representations. If this were true, a drawing of a car as an image, for instance, should map to a similar area in the latent space as a textual description of the strokes that form the drawing. To explore this in a black-box access regime to these models, we propose the use of machine teaching, a theory that studies the minimal set of examples a teacher needs to choose so that the learner captures the concept. In this paper, we evaluate the complexity of teaching vision-language models a subset of objects in the Quick, Draw! dataset using two presentations: raw images as bitmaps and trace coordinates in TikZ format. The results indicate that image-based representations generally require fewer segments and achieve higher accuracy than coordinate-based representations. But, surprisingly, the teaching size usually ranks concepts similarly across both modalities, even when controlling for (a human proxy of) concept priors, suggesting that the simplicity of concepts may be an inherent property that transcends modality representations.

Paperid: 6517, https://arxiv.org/pdf/2505.07864.pdf

Abstract:
Flowcharts are indispensable tools in software design and business-process analysis, yet current vision-language models (VLMs) frequently misinterpret the directional arrows and graph topology that set these diagrams apart from natural images. We introduce a seven-stage pipeline grouped into three broader processes: (1) arrow-aware detection of nodes and arrow endpoints; (2) optical character recognition (OCR) to extract node text; and (3) construction of a structured prompt that guides the VLMs. Tested on a 90-question benchmark distilled from 30 annotated flowcharts, the method raises overall accuracy from 80 % to 89 % (+9 percentage points) without any task-specific fine-tuning. The gain is most pronounced for next-step queries (25/30 -> 30/30; 100 %, +17 pp); branch-result questions improve more modestly, and before-step questions remain difficult. A parallel evaluation with an LLM-as-a-Judge protocol shows the same trends, reinforcing the advantage of explicit arrow encoding. Limitations include dependence on detector and OCR precision, the small evaluation set, and residual errors at nodes with multiple incoming edges. Future work will enlarge the benchmark with synthetic and handwritten flowcharts and assess the approach on Business Process Model and Notation (BPMN) and Unified Modeling Language (UML).

Paperid: 6518, https://arxiv.org/pdf/2505.07019.pdf

Abstract:
Leaf disease identification plays a pivotal role in smart agriculture. However, many existing studies still struggle to integrate image and textual modalities to compensate for each other's limitations. Furthermore, many of these approaches rely on pretraining with constrained datasets such as ImageNet, which lack domain-specific information. We propose SCOLD (Soft-target COntrastive learning for Leaf Disease identification), a context-aware vision-language foundation model tailored to address these challenges for agricultural tasks. SCOLD is developed using a diverse corpus of plant leaf images and corresponding symptom descriptions, comprising over 186,000 image-caption pairs aligned with 97 unique concepts. Through task-agnostic pretraining, SCOLD leverages contextual soft targets to mitigate overconfidence in contrastive learning by smoothing labels, thereby improving model generalization and robustness on fine-grained classification tasks. Experimental results demonstrate that SCOLD outperforms existing vision-language models such as OpenAI-CLIP-L, BioCLIP, and SigLIP2 across several benchmarks, including zero-shot and few-shot classification, image-text retrieval, and image classification, while maintaining a competitive parameter footprint. Ablation studies further highlight SCOLD's effectiveness in contrast to its counterparts. The proposed approach significantly advances the agricultural vision-language foundation model, offering strong performance with minimal or no supervised fine-tuning. This work lays a solid groundwork for future research on models trained with long-form and simplified contexts, tasks involving class ambiguity, and multi-modal systems for intelligent plant disease diagnostics. The code for this study is available at https://huggingface.co/enalis/scold

Paperid: 6519, https://arxiv.org/pdf/2505.03821.pdf

Abstract:
We investigate the ability of Vision Language Models (VLMs) to perform visual perspective taking using a novel set of visual tasks inspired by established human tests. Our approach leverages carefully controlled scenes, in which a single humanoid minifigure is paired with a single object. By systematically varying spatial configurations - such as object position relative to the humanoid minifigure and the humanoid minifigure's orientation - and using both bird's-eye and surface-level views, we created 144 unique visual tasks. Each visual task is paired with a series of 7 diagnostic questions designed to assess three levels of visual cognition: scene understanding, spatial reasoning, and visual perspective taking. Our evaluation of several state-of-the-art models, including GPT-4-Turbo, GPT-4o, Llama-3.2-11B-Vision-Instruct, and variants of Claude Sonnet, reveals that while they excel in scene understanding, the performance declines significantly on spatial reasoning and further deteriorates on perspective-taking. Our analysis suggests a gap between surface-level object recognition and the deeper spatial and perspective reasoning required for complex visual tasks, pointing to the need for integrating explicit geometric representations and tailored training protocols in future VLM development.

Paperid: 6520, https://arxiv.org/pdf/2504.19086.pdf

Abstract:
Single-Domain Generalized Object Detection~(S-DGOD) aims to train an object detector on a single source domain while generalizing well to diverse unseen target domains, making it suitable for multimedia applications that involve various domain shifts, such as intelligent video surveillance and VR/AR technologies. With the success of large-scale Vision-Language Models, recent S-DGOD approaches exploit pre-trained vision-language knowledge to guide invariant feature learning across visual domains. However, the utilized knowledge remains at a coarse-grained level~(e.g., the textual description of adverse weather paired with the image) and serves as an implicit regularization for guidance, struggling to learn accurate region- and object-level features in varying domains. In this work, we propose a new cross-modal feature learning method, which can capture generalized and discriminative regional features for S-DGOD tasks. The core of our method is the mechanism of Cross-modal and Region-aware Feature Interaction, which simultaneously learns both inter-modal and intra-modal regional invariance through dynamic interactions between fine-grained textual and visual features. Moreover, we design a simple but effective strategy called Cross-domain Proposal Refining and Mixing, which aligns the position of region proposals across multiple domains and diversifies them, enhancing the localization ability of detectors in unseen scenarios. Our method achieves new state-of-the-art results on S-DGOD benchmark datasets, with improvements of +8.8\%~mPC on Cityscapes-C and +7.9\%~mPC on DWD over baselines, demonstrating its efficacy.

Paperid: 6521, https://arxiv.org/pdf/2504.14467.pdf

Abstract:
Zero-shot referring image segmentation aims to locate and segment the target region based on a referring expression, with the primary challenge of aligning and matching semantics across visual and textual modalities without training. Previous works address this challenge by utilizing Vision-Language Models and mask proposal networks for region-text matching. However, this paradigm may lead to incorrect target localization due to the inherent ambiguity and diversity of free-form referring expressions. To alleviate this issue, we present LGD (Leveraging Generative Descriptions), a framework that utilizes the advanced language generation capabilities of Multi-Modal Large Language Models to enhance region-text matching performance in Vision-Language Models. Specifically, we first design two kinds of prompts, the attribute prompt and the surrounding prompt, to guide the Multi-Modal Large Language Models in generating descriptions related to the crucial attributes of the referent object and the details of surrounding objects, referred to as attribute description and surrounding description, respectively. Secondly, three visual-text matching scores are introduced to evaluate the similarity between instance-level visual features and textual features, which determines the mask most associated with the referring expression. The proposed method achieves new state-of-the-art performance on three public datasets RefCOCO, RefCOCO+ and RefCOCOg, with maximum improvements of 9.97% in oIoU and 11.29% in mIoU compared to previous methods.

Paperid: 6522, https://arxiv.org/pdf/2504.12137.pdf

Abstract:
Despite recent advances in Large Vision Language Models (LVLMs), these models still suffer from generating hallucinatory responses that do not align with the visual input provided. To mitigate such hallucinations, we introduce Efficient Contrastive Decoding (ECD), a simple method that leverages probabilistic hallucination detection to shift the output distribution towards contextually accurate answers at inference time. By contrasting token probabilities and hallucination scores, ECD subtracts hallucinated concepts from the original distribution, effectively suppressing hallucinations. Notably, our proposed method can be applied to any open-source LVLM and does not require additional LVLM training. We evaluate our method on several benchmark datasets and across different LVLMs. Our experiments show that ECD effectively mitigates hallucinations, outperforming state-of-the-art methods with respect to performance on LVLM benchmarks and computation time.

Paperid: 6523, https://arxiv.org/pdf/2504.10757.pdf

Abstract:
Vision-language models (VLMs) show promise for autonomous driving but often lack transparent reasoning capabilities that are critical for safety. We investigate whether explicitly modeling reasoning during fine-tuning enhances VLM performance on driving decision tasks. Using GPT-4o, we generate structured reasoning chains for driving scenarios from the DriveLM benchmark with category-specific prompting strategies. We compare reasoning-based fine-tuning, answer-only fine-tuning, and baseline instruction-tuned models across multiple small VLM families (Llama 3.2, Llava 1.5, and Qwen 2.5VL). Our results demonstrate that reasoning-based fine-tuning consistently outperforms alternatives, with Llama3.2-11B-reason achieving the highest performance. Models fine-tuned with reasoning show substantial improvements in accuracy and text generation quality, suggesting explicit reasoning enhances internal representations for driving decisions. These findings highlight the importance of transparent decision processes in safety-critical domains and offer a promising direction for developing more interpretable autonomous driving systems.

Paperid: 6524, https://arxiv.org/pdf/2504.09997.pdf

Abstract:
Developing bipedal robots capable of traversing diverse real-world terrains presents a fundamental robotics challenge, as existing methods using predefined height maps and static environments fail to address the complexity of unstructured landscapes. To bridge this gap, we propose GenTe, a framework for generating physically realistic and adaptable terrains to train generalizable locomotion policies. GenTe constructs an atomic terrain library that includes both geometric and physical terrains, enabling curriculum training for reinforcement learning-based locomotion policies. By leveraging function-calling techniques and reasoning capabilities of Vision-Language Models (VLMs), GenTe generates complex, contextually relevant terrains from textual and graphical inputs. The framework introduces realistic force modeling for terrain interactions, capturing effects such as soil sinkage and hydrodynamic resistance. To the best of our knowledge, GenTe is the first framework that systemically generates simulation environments for legged robot locomotion control. Additionally, we introduce a benchmark of 100 generated terrains. Experiments demonstrate improved generalization and robustness in bipedal robot locomotion.

Paperid: 6525, https://arxiv.org/pdf/2504.09620.pdf

Abstract:
We propose the Metropolis-Hastings Captioning Game (MHCG), a method to fuse knowledge of multiple vision-language models (VLMs) by learning from each other. Although existing methods that combine multiple models suffer from inference costs and architectural constraints, MHCG avoids these problems by performing decentralized Bayesian inference through a process resembling a language game. The knowledge fusion process establishes communication between two VLM agents alternately captioning images and learning from each other. We conduct two image-captioning experiments with two VLMs, each pre-trained on a different dataset. The first experiment demonstrates that MHCG achieves consistent improvement in reference-free evaluation metrics. The second experiment investigates how MHCG contributes to sharing VLMs' category-level vocabulary by observing the occurrence of the vocabulary in the generated captions.

Paperid: 6526, https://arxiv.org/pdf/2504.06835.pdf

Abstract:
Long video understanding is a complex task that requires both spatial detail and temporal awareness. While Vision-Language Models (VLMs) obtain frame-level understanding capabilities through multi-frame input, they suffer from information loss due to the sparse sampling strategy. In contrast, Video Large Language Models (Video-LLMs) capture temporal relationships within visual features but are limited by the scarcity of high-quality video-text datasets. To transfer long video understanding capabilities to VLMs with minimal data and computational cost, we propose Lightweight Video Compression (LVC), a novel method featuring the Query-Attention Video Compression mechanism, which effectively tackles the sparse sampling problem in VLMs. By training only the alignment layer with 10k short video-text pairs, LVC significantly enhances the temporal reasoning abilities of VLMs. Extensive experiments show that LVC provides consistent performance improvements across various models, including the InternVL2 series and Phi-3.5-Vision. Notably, the InternVL2-40B-LVC achieves scores of 68.2 and 65.9 on the long video understanding benchmarks MLVU and Video-MME, respectively, with relative improvements of 14.6% and 7.7%. The enhanced models and code will be publicly available soon.

Paperid: 6527, https://arxiv.org/pdf/2504.06298.pdf

Abstract:
We propose a process to compress a pre-trained Vision Language Model into a ternary version of itself instead of training a ternary model from scratch. A new initialization scheme from pre-trained weights based on the k-means algorithm is proposed to reduce the ternarization time. We implement different custom operators for executing the ternary model on the TensorFlow Lite Engine. We compare the original model with its ternary and binary versions in terms of memory consumption, inference speed and perplexity. We find that the ternary model using our custom ternary matrix multiplication operator provides a good compromise in term of memory usage and perplexity, while having the fastest token generation speed.

Paperid: 6528, https://arxiv.org/pdf/2504.04781.pdf

Abstract:
Comprehending occluded objects are not well studied in existing large-scale visual-language multi-modal models. Current state-of-the-art multi-modal large models struggles to provide satisfactory results in understanding occluded objects through universal visual encoders and supervised learning strategies. Therefore, we propose OCC-MLLM-CoT-Alpha, a multi-modal large vision language framework that integrates 3D-aware supervision and Chain-of-Thoughts guidance. Particularly, (1) we build a multi-modal large vision-language model framework which is consisted of a large multi-modal vision-language model and a 3D reconstruction expert model. (2) the corresponding multi-modal Chain-of-Thoughts is learned through a combination of supervised and reinforcement training strategies, allowing the multi-modal vision-language model to enhance the recognition ability with learned multi-modal chain-of-thoughts guidance. (3) A large-scale multi-modal chain-of-thoughts reasoning dataset, consisting of $110k$ samples of occluded objects held in hand, is built. In the evaluation, the proposed methods demonstrate decision score improvement of 15.75%,15.30%,16.98%,14.62%, and 4.42%,3.63%,6.94%,10.70% for two settings of a variety of state-of-the-art models.

Paperid: 6529, https://arxiv.org/pdf/2503.23062.pdf

Abstract:
Shapes and textures are the basic building blocks of visual perception. The ability to identify shapes regardless of orientation, texture, or context, and to recognize textures and materials independently of their associated objects, is essential for a general visual understanding of the world. This work introduces the Large Shape and Textures dataset (LAS&T), a giant collection of highly diverse shapes and textures, created by unsupervised extraction of patterns from natural images. This dataset is used to benchmark how effectively leading Large Vision-Language Models (LVLMs) understand shapes, textures, and materials in 2D and 3D scenes. For shape recognition, we test the models' ability to match images of identical shapes that differ in orientation, texture, color, or environment. Our results show that the shape recognition capabilities of the LVLMs remain significantly below human performance. LVLMs rely predominantly on high-level and semantic features and struggle with abstract shapes lacking clear class associations. For texture and material recognition, we evaluated the models' ability to identify images with identical textures and materials across different objects and environments. Interestingly, leading LVLMs approach human-level performance in recognizing materials in 3D scenes, yet substantially underperform humans when identifying simpler more abstract 2D textures. These results are consistent across a wide range of leading VLMs (GPT/Gemini/LLama/Qwen) and foundation vision models (DINO/CLIP), exposing major deficiencies in the ability of leading models to understand fundamental visual concepts. In contrast, simple nets trained directly for these tasks achieve high accuracy. The LAS&T dataset, featuring over 600,000 images for 2D/3D shape, texture, and material recognition and retrieval, is publicly available.

Paperid: 6530, https://arxiv.org/pdf/2503.21263.pdf

Abstract:
Developing agents capable of fluid gameplay in first/third-person games without API access remains a critical challenge in Artificial General Intelligence (AGI). Recent efforts leverage Vision Language Models (VLMs) as direct controllers, frequently pausing the game to analyze screens and plan action through language reasoning. However, this inefficient paradigm fundamentally restricts agents to basic and non-fluent interactions: relying on isolated VLM reasoning for each action makes it impossible to handle tasks requiring high reactivity (e.g., FPS shooting) or dynamic adaptability (e.g., ACT combat). To handle this, we propose a paradigm shift in gameplay agent design: instead of directly controlling gameplay, VLM develops specialized execution modules tailored for tasks like shooting and combat. These modules handle real-time game interactions, elevating VLM to a high-level developer. Building upon this paradigm, we introduce GameSense, a gameplay agent framework where VLM develops task-specific game sense modules by observing task execution and leveraging vision tools and neural network training pipelines. These modules encapsulate action-feedback logic, ranging from direct action rules to neural network-based decisions. Experiments demonstrate that our framework is the first to achieve fluent gameplay in diverse genres, including ACT, FPS, and Flappy Bird, setting a new benchmark for game-playing agents.

Paperid: 6531, https://arxiv.org/pdf/2503.21214.pdf

Abstract:
Comprehending 3D environments is vital for intelligent systems in domains like robotics and autonomous navigation. Voxel grids offer a structured representation of 3D space, but extracting high-level semantic meaning remains challenging. This paper proposes a novel approach utilizing a Vision-Language Model (VLM) to extract "voxel semantics"-object identity, color, and location-from voxel data. Critically, instead of employing complex 3D networks, our method processes the voxel space by systematically slicing it along a primary axis (e.g., the Z-axis, analogous to CT scan slices). These 2D slices are then formatted and sequentially fed into the image encoder of a standard VLM. The model learns to aggregate information across slices and correlate spatial patterns with semantic concepts provided by the language component. This slice-based strategy aims to leverage the power of pre-trained 2D VLMs for efficient 3D semantic understanding directly from voxel representations.

Paperid: 6532, https://arxiv.org/pdf/2503.21130.pdf

Abstract:
Tutorial videos are a valuable resource for people looking to learn new tasks. People often learn these skills by viewing multiple tutorial videos to get an overall understanding of a task by looking at different approaches to achieve the task. However, navigating through multiple videos can be time-consuming and mentally demanding as these videos are scattered and not easy to skim. We propose VideoMix, a system that helps users gain a holistic understanding of a how-to task by aggregating information from multiple videos on the task. Insights from our formative study (N=12) reveal that learners value understanding potential outcomes, required materials, alternative methods, and important details shared by different videos. Powered by a Vision-Language Model pipeline, VideoMix extracts and organizes this information, presenting concise textual summaries alongside relevant video clips, enabling users to quickly digest and navigate the content. A comparative user study (N=12) demonstrated that VideoMix enabled participants to gain a more comprehensive understanding of tasks with greater efficiency than a baseline video interface, where videos are viewed independently. Our findings highlight the potential of a task-oriented, multi-video approach where videos are organized around a shared goal, offering an enhanced alternative to conventional video-based learning.

Paperid: 6533, https://arxiv.org/pdf/2503.18334.pdf

Abstract:
Test-time adaptation (TTA) of visual language models has recently attracted significant attention as a solution to the performance degradation caused by distribution shifts in downstream tasks. However, existing cache-based TTA methods have certain limitations. They mainly rely on the accuracy of cached feature labels, and the presence of noisy pseudo-labels can cause these features to deviate from their true distribution. This makes cache retrieval methods based on similarity matching highly sensitive to outliers or extreme samples. Moreover, current methods lack effective mechanisms to model class distributions, which limits their ability to fully exploit the potential of cached information. To address these challenges, we introduce a comprehensive and reliable caching mechanism and propose a novel zero-shot TTA method called "Cache, Residual, Gaussian" (CRG). This method not only employs learnable residual parameters to better align positive and negative visual prototypes with text prototypes, thereby optimizing the quality of cached features, but also incorporates Gaussian Discriminant Analysis (GDA) to dynamically model intra-class feature distributions, further mitigating the impact of noisy features. Experimental results on 13 benchmarks demonstrate that CRG outperforms state-of-the-art TTA methods, showcasing exceptional robustness and adaptability.

Paperid: 6534, https://arxiv.org/pdf/2503.14674.pdf

Abstract:
Large Vision-Language Models (LVLMs) have shown remarkable progress in various multimodal tasks, yet they often struggle with complex visual reasoning that requires multi-step inference. To address this limitation, we propose MF-SQ-LLaVA, a novel approach that enhances LVLMs by enabling implicit self-questioning through end-to-end training. Our method involves augmenting visual question answering datasets with reasoning chains consisting of sub-question and answer pairs, and training the LVLM with a multi-task loss that encourages the generation and answering of these intermediate steps, as well as the prediction of the final answer. We conduct extensive experiments on the ScienceQA and VQAv2 datasets, demonstrating that MF-SQ-LLaVA significantly outperforms existing state-of-the-art models, including the base LLaVA and the original SQ-LLaVA. Ablation studies further validate the contribution of each component of our approach, and human evaluation confirms the improved accuracy and coherence of the reasoning process enabled by our method.

Paperid: 6535, https://arxiv.org/pdf/2503.13060.pdf

Abstract:
In medieval India, the Marathi language was written using the Modi script. The texts written in Modi script include extensive knowledge about medieval sciences, medicines, land records and authentic evidence about Indian history. Around 40 million documents are in poor condition and have not yet been transliterated. Furthermore, only a few experts in this domain can transliterate this script into English or Devanagari. Most of the past research predominantly focuses on individual character recognition. A system that can transliterate Modi script documents to Devanagari script is needed. We propose the MoDeTrans dataset, comprising 2,043 images of Modi script documents accompanied by their corresponding textual transliterations in Devanagari. We further introduce MoScNet (\textbf{Mo}di \textbf{Sc}ript \textbf{Net}work), a novel Vision-Language Model (VLM) framework for transliterating Modi script images into Devanagari text. MoScNet leverages Knowledge Distillation, where a student model learns from a teacher model to enhance transliteration performance. The final student model of MoScNet has better performance than the teacher model while having 163$\times$ lower parameters. Our work is the first to perform direct transliteration from the handwritten Modi script to the Devanagari script. MoScNet also shows competitive results on the optical character recognition (OCR) task.

Paperid: 6536, https://arxiv.org/pdf/2503.12834.pdf

Abstract:
A fundamental challenge in conditional 3D shape generation is to minimize the information loss and maximize the intention of user input. Existing approaches have predominantly focused on two types of isolated conditional signals, i.e., user sketches and text descriptions, each of which does not offer flexible control of the generated shape. In this paper, we introduce PASTA, the flexible approach that seamlessly integrates a user sketch and a text description for 3D shape generation. The key idea is to use text embeddings from a vision-language model to enrich the semantic representation of sketches. Specifically, these text-derived priors specify the part components of the object, compensating for missing visual cues from ambiguous sketches. In addition, we introduce ISG-Net which employs two types of graph convolutional networks: IndivGCN, which processes fine-grained details, and PartGCN, which aggregates these details into parts and refines the structure of objects. Extensive experiments demonstrate that PASTA outperforms existing methods in part-level editing and achieves state-of-the-art results in sketch-to-3D shape generation.

Paperid: 6537, https://arxiv.org/pdf/2503.12281.pdf

Abstract:
In recent years, we have witnessed significant progress in emerging deep learning models, particularly Large Language Models (LLMs) and Vision-Language Models (VLMs). These models have demonstrated promising results, indicating a new era of Artificial Intelligence (AI) that surpasses previous methodologies. Their extensive knowledge and zero-shot capabilities suggest a paradigm shift in developing deep learning solutions, moving from data capturing and algorithm training to just writing appropriate prompts. While the application of these technologies has been explored across various industries, including automotive, there is a notable gap in the scientific literature regarding their use in Driver Monitoring Systems (DMS). This paper presents our initial approach to implementing VLMs in this domain, utilising the Driver Monitoring Dataset to evaluate their performance and discussing their advantages and challenges when implemented in real-world scenarios.

Paperid: 6538, https://arxiv.org/pdf/2503.11241.pdf

Abstract:
Compound Expression Recognition (CER) is crucial for understanding human emotions and improving human-computer interaction. However, CER faces challenges due to the complexity of facial expressions and the difficulty of capturing subtle emotional cues. To address these issues, we propose a novel approach leveraging Large Vision-Language Models (LVLMs). Our method employs a two-stage fine-tuning process: first, pre-trained LVLMs are fine-tuned on basic facial expressions to establish foundational patterns; second, the model is further optimized on a compound-expression dataset to refine visual-language feature interactions. Our approach achieves advanced accuracy on the RAF-DB dataset and demonstrates strong zero-shot generalization on the C-EXPR-DB dataset, showcasing its potential for real-world applications in emotion analysis and human-computer interaction.

Paperid: 6539, https://arxiv.org/pdf/2503.03840.pdf

Abstract:
Understanding geometry relies heavily on vision. In this work, we evaluate whether state-of-the-art vision language models (VLMs) can understand simple geometric concepts. We use a paradigm from cognitive science that isolates visual understanding of simple geometry from the many other capabilities it is often conflated with such as reasoning and world knowledge. We compare model performance with human adults from the USA, as well as with prior research on human adults without formal education from an Amazonian indigenous group. We find that VLMs consistently underperform both groups of human adults, although they succeed with some concepts more than others. We also find that VLM geometric understanding is more brittle than human understanding, and is not robust when tasks require mental rotation. This work highlights interesting differences in the origin of geometric understanding in humans and machines -- e.g. from printed materials used in formal education vs. interactions with the physical world or a combination of the two -- and a small step toward understanding these differences.

Paperid: 6540, https://arxiv.org/pdf/2503.01619.pdf

Abstract:
Modern front-end (FE) development, especially when leveraging the unique features of frameworks like React and Vue, presents distinctive challenges. These include managing modular architectures, ensuring synchronization between data and visual outputs for declarative rendering, and adapting reusable components to various scenarios. Such complexities make it particularly difficult for state-of-the-art large vision-language models (VLMs) to generate accurate and functional code directly from design images. To address these challenges, we propose a reflective agentic workflow that synthesizes high-quality image-text data to capture the diverse characteristics of FE development. This workflow automates the extraction of self-contained\footnote{A \textbf{self-contained} code snippet is one that encapsulates all necessary logic, styling, and dependencies, ensuring it functions independently without requiring external imports or context.} code snippets from real-world projects, renders the corresponding visual outputs, and generates detailed descriptions that link design elements to functional code. To further expand the scope and utility of the synthesis, we introduce three data synthesis strategies: Evolution-based synthesis, which enables scalable and diverse dataset expansion; Waterfall-Model-based synthesis, which generates logically coherent code derived from system requirements; and Additive Development synthesis, which iteratively increases the complexity of human-authored components. We build a large vision-language model, Flame, trained on the synthesized datasets and demonstrate its effectiveness in generating React code via the $\text{pass}@k$ metric. Our results suggest that a code VLM trained to interpret images before code generation may achieve better performance.

Paperid: 6541, https://arxiv.org/pdf/2502.18735.pdf

Abstract:
A domain shift exists between the large-scale, internet data used to train a Vision-Language Model (VLM) and the raw image streams collected by a robot. Existing adaptation strategies require the definition of a closed-set of classes, which is impractical for a robot that must respond to diverse natural language queries. In response, we present QueryAdapter; a novel framework for rapidly adapting a pre-trained VLM in response to a natural language query. QueryAdapter leverages unlabelled data collected during previous deployments to align VLM features with semantic classes related to the query. By optimising learnable prompt tokens and actively selecting objects for training, an adapted model can be produced in a matter of minutes. We also explore how objects unrelated to the query should be dealt with when using real-world data for adaptation. In turn, we propose the use of object captions as negative class labels, helping to produce better calibrated confidence scores during adaptation. Extensive experiments on ScanNet++ demonstrate that QueryAdapter significantly enhances object retrieval performance compared to state-of-the-art unsupervised VLM adapters and 3D scene graph methods. Furthermore, the approach exhibits robust generalization to abstract affordance queries and other datasets, such as Ego4D.

Paperid: 6542, https://arxiv.org/pdf/2502.16361.pdf

Abstract:
Vision-Language Models (VLMs) are increasingly deployed in public sector missions, necessitating robust evaluation of their safety and vulnerability to adversarial attacks. This paper introduces a novel framework to quantify adversarial risks in VLMs. We analyze model performance under Gaussian, salt-and-pepper, and uniform noise, identifying misclassification thresholds and deriving composite noise patches and saliency patterns that highlight vulnerable regions. These patterns are compared against the Fast Gradient Sign Method (FGSM) to assess their adversarial effectiveness. We propose a new Vulnerability Score that combines the impact of random noise and adversarial attacks, providing a comprehensive metric for evaluating model robustness.

Paperid: 6543, https://arxiv.org/pdf/2502.14088.pdf

Abstract:
Earth Observation (EO) data encompass a vast range of remotely sensed information, featuring multi-sensor and multi-temporal, playing an indispensable role in understanding our planet's dynamics. Recently, Vision Language Models (VLMs) have achieved remarkable success in perception and reasoning tasks, bringing new insights and opportunities to the EO field. However, the potential for EO applications, especially for scientific regression related applications remains largely unexplored. This paper bridges that gap by systematically examining the challenges and opportunities of adapting VLMs for EO regression tasks. The discussion first contrasts the distinctive properties of EO data with conventional computer vision datasets, then identifies four core obstacles in applying VLMs to EO regression: 1) the absence of dedicated benchmarks, 2) the discrete-versus-continuous representation mismatch, 3) cumulative error accumulation, and 4) the suboptimal nature of text-centric training objectives for numerical tasks. Next, a series of methodological insights and potential subtle pitfalls are explored. Lastly, we offer some promising future directions for designing robust, domain-aware solutions. Our findings highlight the promise of VLMs for scientific regression in EO, setting the stage for more precise and interpretable modeling of critical environmental processes.

Paperid: 6544, https://arxiv.org/pdf/2502.13372.pdf

Abstract:
While large vision-language models can generate motion graphics animations from text prompts, they regularly fail to include all spatio-temporal properties described in the prompt. We introduce MoVer, a motion verification DSL based on first-order logic that can check spatio-temporal properties of a motion graphics animation. We identify a general set of such properties that people commonly use to describe animations (e.g., the direction and timing of motions, the relative positioning of objects, etc.). We implement these properties as predicates in MoVer and provide an execution engine that can apply a MoVer program to any input SVG-based motion graphics animation. We then demonstrate how MoVer can be used in an LLM-based synthesis and verification pipeline for iteratively refining motion graphics animations. Given a text prompt, our pipeline synthesizes a motion graphics animation and a corresponding MoVer program. Executing the verification program on the animation yields a report of the predicates that failed and the report can be automatically fed back to LLM to iteratively correct the animation. To evaluate our pipeline, we build a synthetic dataset of 5600 text prompts paired with ground truth MoVer verification programs. We find that while our LLM-based pipeline is able to automatically generate a correct motion graphics animation for 58.8% of the test prompts without any iteration, this number raises to 93.6% with up to 50 correction iterations. Our code and dataset are at https://mover-dsl.github.io.

Paperid: 6545, https://arxiv.org/pdf/2502.10886.pdf

Abstract:
Entity tracking is a fundamental challenge in natural language understanding, requiring models to maintain coherent representations of entities. Previous work has benchmarked entity tracking performance in purely text-based tasks. We introduce MET-Bench, a multimodal entity tracking benchmark designed to evaluate the ability of vision-language models to track entity states across modalities. Using two structured domains, Chess and the Shell Game, we assess how effectively current models integrate textual and image-based state updates. Our findings reveal a significant performance gap between text-based and image-based tracking and that this performance gap stems from deficits in visual reasoning rather than perception. We further show that explicit text-based reasoning strategies improve performance, yet substantial limitations remain, especially in long-horizon multimodal scenarios. Our results highlight the need for improved multimodal representations and reasoning techniques to bridge the gap between textual and visual entity tracking.

Paperid: 6546, https://arxiv.org/pdf/2502.08391.pdf

Abstract:
Multiple instance learning (MIL)-based framework has become the mainstream for processing the whole slide image (WSI) with giga-pixel size and hierarchical image context in digital pathology. However, these methods heavily depend on a substantial number of bag-level labels and solely learn from the original slides, which are easily affected by variations in data distribution. Recently, vision language model (VLM)-based methods introduced the language prior by pre-training on large-scale pathological image-text pairs. However, the previous text prompt lacks the consideration of pathological prior knowledge, therefore does not substantially boost the model's performance. Moreover, the collection of such pairs and the pre-training process are very time-consuming and source-intensive.To solve the above problems, we propose a dual-scale vision-language multiple instance learning (ViLa-MIL) framework for whole slide image classification. Specifically, we propose a dual-scale visual descriptive text prompt based on the frozen large language model (LLM) to boost the performance of VLM effectively. To transfer the VLM to process WSI efficiently, for the image branch, we propose a prototype-guided patch decoder to aggregate the patch features progressively by grouping similar patches into the same prototype; for the text branch, we introduce a context-guided text decoder to enhance the text features by incorporating the multi-granular image contexts. Extensive studies on three multi-cancer and multi-center subtyping datasets demonstrate the superiority of ViLa-MIL.

Paperid: 6547, https://arxiv.org/pdf/2502.01530.pdf

Abstract:
Inductive biases are what allow learners to make guesses in the absence of conclusive evidence. These biases have often been studied in cognitive science using concepts or categories -- e.g. by testing how humans generalize a new category from a few examples that leave the category boundary ambiguous. We use these approaches to study generalization in foundation models during in-context learning. Modern foundation models can condition on both vision and text, and differences in how they interpret and learn from these different modalities is an emerging area of study. Here, we study how their generalizations vary by the modality in which stimuli are presented, and the way the stimuli are described in text. We study these biases with three different experimental paradigms, across three different vision-language models. We find that the models generally show some bias towards generalizing according to shape over color. This shape bias tends to be amplified when the examples are presented visually. By contrast, when examples are presented in text, the ordering of adjectives affects generalization. However, the extent of these effects vary across models and paradigms. These results help to reveal how vision-language models represent different types of inputs in context, and may have practical implications for the use of vision-language models.

Paperid: 6548, https://arxiv.org/pdf/2501.18698.pdf

Abstract:
Large vision-language models (LVLMs) have been regarded as a breakthrough advance in an astoundingly variety of tasks, from content generation to virtual assistants and multimodal search or retrieval. However, for many of these applications, the performance of these methods has been widely criticized, particularly when compared with state-of-the-art methods and technologies in each specific domain. In this work, we compare the performance of the leading large vision-language models in the human re-identification task, using as baseline the performance attained by state-of-the-art AI models specifically designed for this problem. We compare the results due to ChatGPT-4o, Gemini-2.0-Flash, Claude 3.5 Sonnet, and Qwen-VL-Max to a baseline ReID PersonViT model, using the well-known Market1501 dataset. Our evaluation pipeline includes the dataset curation, prompt engineering, and metric selection to assess the models' performance. Results are analyzed from many different perspectives: similarity scores, classification accuracy, and classification metrics, including precision, recall, F1 score, and area under curve (AUC). Our results confirm the strengths of LVLMs, but also their severe limitations that often lead to catastrophic answers and should be the scope of further research. As a concluding remark, we speculate about some further research that should fuse traditional and LVLMs to combine the strengths from both families of techniques and achieve solid improvements in performance.

Paperid: 6549, https://arxiv.org/pdf/2501.17665.pdf

Abstract:
Automating the generation of Planning Domain Definition Language (PDDL) with Large Language Model (LLM) opens new research topic in AI planning, particularly for complex real-world tasks. This paper introduces Image2PDDL, a novel framework that leverages Vision-Language Models (VLMs) to automatically convert images of initial states and descriptions of goal states into PDDL problems. By providing a PDDL domain alongside visual inputs, Imasge2PDDL addresses key challenges in bridging perceptual understanding with symbolic planning, reducing the expertise required to create structured problem instances, and improving scalability across tasks of varying complexity. We evaluate the framework on various domains, including standard planning domains like blocksworld and sliding tile puzzles, using datasets with multiple difficulty levels. Performance is assessed on syntax correctness, ensuring grammar and executability, and content correctness, verifying accurate state representation in generated PDDL problems. The proposed approach demonstrates promising results across diverse task complexities, suggesting its potential for broader applications in AI planning. We will discuss a potential use case in robot-assisted teaching of students with Autism Spectrum Disorder.

Paperid: 6550, https://arxiv.org/pdf/2501.16769.pdf

Abstract:
Open-vocabulary semantic segmentation attempts to classify and outline objects in an image using arbitrary text labels, including those unseen during training. Self-supervised learning resolves numerous visual and linguistic processing problems when effectively trained. This study investigates simple yet efficient methods for adapting previously learned foundation models for open-vocabulary semantic segmentation tasks. Our research proposes "Beyond-Labels", a lightweight transformer-based fusion module that uses a small amount of image segmentation data to fuse frozen visual representations with language concepts. This strategy allows the model to leverage the extensive knowledge of pre-trained models without requiring significant retraining, making the approach data-efficient and scalable. Furthermore, we capture positional information in images using Fourier embeddings, improving generalization and enabling smooth and consistent spatial encoding. We perform thorough ablation studies to examine the main components of our proposed method. On the standard benchmark PASCAL-5i, the method performs better despite being trained on frozen vision and language representations. Index Terms: Beyond-Labels, open-vocabulary semantic segmentation, Fourier embeddings, PASCAL-5i

Paperid: 6551, https://arxiv.org/pdf/2501.13368.pdf

Abstract:
Identifying individual animals within large wildlife populations is essential for effective wildlife monitoring and conservation efforts. Recent advancements in computer vision have shown promise in animal re-identification (Animal ReID) by leveraging data from camera traps. However, existing Animal ReID datasets rely exclusively on visual data, overlooking environmental metadata that ecologists have identified as highly correlated with animal behavior and identity, such as temperature and circadian rhythms. Moreover, the emergence of multimodal models capable of jointly processing visual and textual data presents new opportunities for Animal ReID, but existing datasets fail to leverage these models' text-processing capabilities, limiting their full potential. Additionally, to facilitate the use of metadata in existing ReID methods, we propose the Meta-Feature Adapter (MFA), a lightweight module that can be incorporated into existing vision-language model (VLM)-based Animal ReID methods, allowing ReID models to leverage both environmental metadata and visual information to improve ReID performance. Experiments on MetaWild show that combining baseline ReID models with MFA to incorporate metadata consistently improves performance compared to using visual information alone, validating the effectiveness of incorporating metadata in re-identification. We hope that our proposed dataset can inspire further exploration of multimodal approaches for Animal ReID.

Paperid: 6552, https://arxiv.org/pdf/2501.09267.pdf

Abstract:
The construction industry has long explored robotics and computer vision, yet their deployment on construction sites remains very limited. These technologies have the potential to revolutionize traditional workflows by enhancing accuracy, efficiency, and safety in construction management. Ground robots equipped with advanced vision systems could automate tasks such as monitoring mechanical, electrical, and plumbing (MEP) systems. The present research evaluates the applicability of open-vocabulary vision-language models compared to fine-tuned, lightweight, closed-set object detectors for detecting MEP components using a mobile ground robotic platform. A dataset collected with cameras mounted on a ground robot was manually annotated and analyzed to compare model performance. The results demonstrate that, despite the versatility of vision-language models, fine-tuned lightweight models still largely outperform them in specialized environments and for domain-specific tasks.

Paperid: 6553, https://arxiv.org/pdf/2501.08597.pdf

Abstract:
Large Vision-Language Models (LVLMs) have demonstrated impressive capabilities in multimodal tasks, but their performance is often constrained by the lack of external knowledge integration, limiting their ability to handle knowledge-intensive tasks such as visual question answering and reasoning. To address this challenge, we propose a novel method, Adaptive Knowledge-Guided Pretraining for Large Vision-Language Models (AKGP-LVLM), which dynamically incorporates structured and unstructured knowledge into LVLMs during pretraining and fine-tuning. Our approach employs a knowledge encoder to represent external knowledge, a retrieval mechanism to select task-relevant information, and a dynamic adaptor to align multimodal and knowledge representations effectively. We evaluate our method on four benchmark datasets, demonstrating significant performance improvements over state-of-the-art models. Furthermore, human evaluations highlight the superior correctness and relevance of our model's outputs. Extensive analyses confirm the robustness, efficiency, and scalability of AKGP-LVLM, making it a compelling solution for real-world knowledge-intensive tasks.

Paperid: 6554, https://arxiv.org/pdf/2501.07109.pdf

Abstract:
Visual Question Answering (VQA) is an interdisciplinary field that bridges the gap between computer vision (CV) and natural language processing(NLP), enabling Artificial Intelligence(AI) systems to answer questions about images. Since its inception in 2015, VQA has rapidly evolved, driven by advances in deep learning, attention mechanisms, and transformer-based models. This survey traces the journey of VQA from its early days, through major breakthroughs, such as attention mechanisms, compositional reasoning, and the rise of vision-language pre-training methods. We highlight key models, datasets, and techniques that shaped the development of VQA systems, emphasizing the pivotal role of transformer architectures and multimodal pre-training in driving recent progress. Additionally, we explore specialized applications of VQA in domains like healthcare and discuss ongoing challenges, such as dataset bias, model interpretability, and the need for common-sense reasoning. Lastly, we discuss the emerging trends in large multimodal language models and the integration of external knowledge, offering insights into the future directions of VQA. This paper aims to provide a comprehensive overview of the evolution of VQA, highlighting both its current state and potential advancements.

Paperid: 6555, https://arxiv.org/pdf/2501.06887.pdf

Abstract:
As deep learning models gain attraction in medical data, ensuring transparent and trustworthy decision-making is essential. In skin cancer diagnosis, while advancements in lesion detection and classification have improved accuracy, the black-box nature of these methods poses challenges in understanding their decision processes, leading to trust issues among physicians. This study leverages the CLIP (Contrastive Language-Image Pretraining) model, trained on different skin lesion datasets, to capture meaningful relationships between visual features and diagnostic criteria terms. To further enhance transparency, we propose a method called MedGrad E-CLIP, which builds on gradient-based E-CLIP by incorporating a weighted entropy mechanism designed for complex medical imaging like skin lesions. This approach highlights critical image regions linked to specific diagnostic descriptions. The developed integrated pipeline not only classifies skin lesions by matching corresponding descriptions but also adds an essential layer of explainability developed especially for medical data. By visually explaining how different features in an image relates to diagnostic criteria, this approach demonstrates the potential of advanced vision-language models in medical image analysis, ultimately improving transparency, robustness, and trust in AI-driven diagnostic systems.

Paperid: 6556, https://arxiv.org/pdf/2501.03482.pdf

Abstract:
Satisfactory progress has been achieved recently in universal segmentation of CT images. Following the success of vision-language methods, there is a growing trend towards utilizing text prompts and contrastive learning to develop universal segmentation models. However, there exists a significant imbalance in information density between 3D images and text prompts. Moreover, the standard fully connected layer segmentation approach faces significant challenges in handling multiple classes and exhibits poor generalizability. To address these challenges, we propose the VOxel Interacting with LAnguage method (VOILA) for universal CT image segmentation. Initially, we align voxels and language into a shared representation space and classify voxels on the basis of cosine similarity. Subsequently, we develop the Voxel-Language Interaction framework to mitigate the impact of class imbalance caused by foreground-background discrepancies and variations in target volumes. Furthermore, a Complexity-Aware Sampling method is proposed to focus on region hard to segment, achieved by generating pseudo-heatmaps from a trainable Gaussian mixture distribution. Our results indicate the proposed VOILA is capable to achieve improved performance with reduced parameters and computational cost during training. Furthermore, it demonstrates significant generalizability across diverse datasets without additional fine-tuning.

Paperid: 6557, https://arxiv.org/pdf/2501.02850.pdf

Abstract:
The rapid increase in video content production has resulted in enormous data volumes, creating significant challenges for efficient analysis and resource management. To address this, robust video analysis tools are essential. This paper presents an innovative proof of concept using Generative Artificial Intelligence (GenAI) in the form of Vision Language Models to enhance the downstream video analysis process. Our tool generates customized textual summaries based on user-defined queries, providing focused insights within extensive video datasets. Unlike traditional methods that offer generic summaries or limited action recognition, our approach utilizes Vision Language Models to extract relevant information, improving analysis precision and efficiency. The proposed method produces textual summaries from extensive CCTV footage, which can then be stored for an indefinite time in a very small storage space compared to videos, allowing users to quickly navigate and verify significant events without exhaustive manual review. Qualitative evaluations result in 80% and 70% accuracy in temporal and spatial quality and consistency of the pipeline respectively.

Paperid: 6558, https://arxiv.org/pdf/2501.02584.pdf

Abstract:
Vision-Language Models (VLMs) have recently experienced significant advancements. However, challenges persist in the accurate recognition of fine details within high resolution images, which limits performance in multiple tasks. This work introduces Pheye, a novel architecture that efficiently processes high-resolution images while training fewer parameters than similarly sized VLMs. Notably, Pheye achieves a high efficiency while maintaining strong performance, particularly in tasks that demand fine-grained image understanding and/or the handling of scene-text.

Paperid: 6559, https://arxiv.org/pdf/2501.00917.pdf

Abstract:
Text-to-image generation has witnessed significant advancements with the integration of Large Vision-Language Models (LVLMs), yet challenges remain in aligning complex textual descriptions with high-quality, visually coherent images. This paper introduces the Vision-Language Aligned Diffusion (VLAD) model, a generative framework that addresses these challenges through a dual-stream strategy combining semantic alignment and hierarchical diffusion. VLAD utilizes a Contextual Composition Module (CCM) to decompose textual prompts into global and local representations, ensuring precise alignment with visual features. Furthermore, it incorporates a multi-stage diffusion process with hierarchical guidance to generate high-fidelity images. Experiments conducted on MARIO-Eval and INNOVATOR-Eval benchmarks demonstrate that VLAD significantly outperforms state-of-the-art methods in terms of image quality, semantic alignment, and text rendering accuracy. Human evaluations further validate the superior performance of VLAD, making it a promising approach for text-to-image generation in complex scenarios.

Paperid: 6560, https://arxiv.org/pdf/2412.19954.pdf

Abstract:
In the construction sector, workers often endure prolonged periods of high-intensity physical work and prolonged use of tools, resulting in injuries and illnesses primarily linked to postural ergonomic risks, a longstanding predominant health concern. To mitigate these risks, researchers have applied various technological methods to identify the ergonomic risks that construction workers face. However, traditional ergonomic risk assessment (ERA) techniques do not offer interactive feedback. The rapidly developing vision-language models (VLMs), capable of generating textual descriptions or answering questions about ergonomic risks based on image inputs, have not yet received widespread attention. This research introduces an interactive visual query system tailored to assess the postural ergonomic risks of construction workers. The system's capabilities include visual question answering (VQA), which responds to visual queries regarding workers' exposure to postural ergonomic risks, and image captioning (IC), which generates textual descriptions of these risks from images. Additionally, this study proposes a dataset designed for training and testing such methodologies. Systematic testing indicates that the VQA functionality delivers an accuracy of 96.5%. Moreover, evaluations using nine metrics for IC and assessments from human experts indicate that the proposed approach surpasses the performance of a method using the same architecture trained solely on generic datasets. This study sets a new direction for future developments in interactive ERA using generative artificial intelligence (AI) technologies.

Paperid: 6561, https://arxiv.org/pdf/2412.19944.pdf

Abstract:
This paper presents our submission to the COOOL competition, a novel benchmark for detecting and classifying out-of-label hazards in autonomous driving. Our approach integrates diverse methods across three core tasks: (i) driver reaction detection, (ii) hazard object identification, and (iii) hazard captioning. We propose kernel-based change point detection on bounding boxes and optical flow dynamics for driver reaction detection to analyze motion patterns. For hazard identification, we combined a naive proximity-based strategy with object classification using a pre-trained ViT model. At last, for hazard captioning, we used the MOLMO vision-language model with tailored prompts to generate precise and context-aware descriptions of rare and low-resolution hazards. The proposed pipeline outperformed the baseline methods by a large margin, reducing the relative error by 33%, and scored 2nd on the final leaderboard consisting of 32 teams.

Paperid: 6562, https://arxiv.org/pdf/2412.18675.pdf

Abstract:
Multi-head self-attention (MHSA) is a key component of Transformers, a widely popular architecture in both language and vision. Multiple heads intuitively enable different parallel processes over the same input. Yet, they also obscure the attribution of each input patch to the output of a model. We propose a novel 1-head Transformer Attention Bottleneck (TAB) layer, inserted after the traditional MHSA architecture, to serve as an attention bottleneck for interpretability and intervention. Unlike standard self-attention, TAB constrains the total attention over all patches to $\in [0, 1]$. That is, when the total attention is 0, no visual information is propagated further into the network, and the vision-language model (VLM) would default to a generic, image-independent response. To demonstrate the advantages of TAB, we train VLMs with TAB to perform image-difference captioning. Over three datasets, our models perform similarly to baseline VLMs in captioning but the bottleneck is superior in localizing changes and in identifying when no changes occur. TAB is the first architecture to enable users to debug by editing attention, which often produces expected outputs by VLMs.

Paperid: 6563, https://arxiv.org/pdf/2412.18604.pdf

Abstract:
Classifiers are important components in many computer vision tasks, serving as the foundational backbone of a wide variety of models employed across diverse applications. However, understanding the decision-making process of classifiers remains a significant challenge. We propose DiffEx, a novel method that leverages the capabilities of text-to-image diffusion models to explain classifier decisions. Unlike traditional GAN-based explainability models, which are limited to simple, single-concept analyses and typically require training a new model for each classifier, our approach can explain classifiers that focus on single concepts (such as faces or animals) as well as those that handle complex scenes involving multiple concepts. DiffEx employs vision-language models to create a hierarchical list of semantics, allowing users to identify not only the overarching semantic influences on classifiers (e.g., the 'beard' semantic in a facial classifier) but also their sub-types, such as 'goatee' or 'Balbo' beard. Our experiments demonstrate that DiffEx is able to cover a significantly broader spectrum of semantics compared to its GAN counterparts, providing a hierarchical tool that delivers a more detailed and fine-grained understanding of classifier decisions.

Paperid: 6564, https://arxiv.org/pdf/2412.16777.pdf

Abstract:
Self-supervised vision-language models trained with contrastive objectives form the basis of current state-of-the-art methods in AI vision tasks. The success of these models is a direct consequence of the huge web-scale datasets used to train them, but they require correspondingly large vision components to properly learn powerful and general representations from such a broad data domain. This poses a challenge for deploying large vision-language models, especially in resource-constrained environments. To address this, we propose an alternate vision-language architecture, called HyperCLIP, that uses a small image encoder along with a hypernetwork that dynamically adapts image encoder weights to each new set of text inputs. All three components of the model (hypernetwork, image encoder, and text encoder) are pre-trained jointly end-to-end, and with a trained HyperCLIP model, we can generate new zero-shot deployment-friendly image classifiers for any task with a single forward pass through the text encoder and hypernetwork. HyperCLIP increases the zero-shot accuracy of SigLIP trained models with small image encoders by up to 3% on ImageNet and 5% on CIFAR-100 with minimal training throughput overhead.

Paperid: 6565, https://arxiv.org/pdf/2412.11396.pdf

Abstract:
Object-aware reasoning in vision-language tasks poses significant challenges for current models, particularly in handling unseen objects, reducing hallucinations, and capturing fine-grained relationships in complex visual scenes. To address these limitations, we propose the Vision-Aware Retrieval-Augmented Prompting (VRAP) framework, a generative approach that enhances Large Vision-Language Models (LVLMs) by integrating retrieval-augmented object tags into their prompts. VRAP introduces a novel pipeline where structured tags, including objects, attributes, and relationships, are extracted using pretrained visual encoders and scene graph parsers. These tags are enriched with external knowledge and incorporated into the LLM's input, enabling detailed and accurate reasoning. We evaluate VRAP across multiple vision-language benchmarks, including VQAv2, GQA, VizWiz, and COCO, achieving state-of-the-art performance in fine-grained reasoning and multimodal understanding. Additionally, our ablation studies highlight the importance of retrieval-augmented tags and contrastive learning, while human evaluations confirm VRAP's ability to generate accurate, detailed, and contextually relevant responses. Notably, VRAP achieves a 40% reduction in inference latency by eliminating runtime retrieval. These results demonstrate that VRAP is a robust and efficient framework for advancing object-aware multimodal reasoning.

Paperid: 6566, https://arxiv.org/pdf/2412.11391.pdf

Abstract:
Temporal reasoning is a critical challenge in video-language understanding, as it requires models to align semantic concepts consistently across time. While existing large vision-language models (LVLMs) and large language models (LLMs) excel at static tasks, they struggle to capture dynamic interactions and temporal dependencies in video sequences. In this work, we propose Temporal Semantic Alignment via Dynamic Prompting (TSADP), a novel framework that enhances temporal reasoning capabilities through dynamic task-specific prompts and temporal contrastive learning. TSADP leverages a Dynamic Prompt Generator (DPG) to encode fine-grained temporal relationships and a Temporal Contrastive Loss (TCL) to align visual and textual embeddings across time. We evaluate our method on the VidSitu dataset, augmented with enriched temporal annotations, and demonstrate significant improvements over state-of-the-art models in tasks such as Intra-Video Entity Association, Temporal Relationship Understanding, and Chronology Prediction. Human evaluations further confirm TSADP's ability to generate coherent and semantically accurate descriptions. Our analysis highlights the robustness, efficiency, and practical utility of TSADP, making it a step forward in the field of video-language understanding.

Paperid: 6567, https://arxiv.org/pdf/2412.10761.pdf

Abstract:
Vision-language retrieval aims to search for similar instances in one modality based on queries from another modality. The primary objective is to learn cross-modal matching representations in a latent common space. Actually, the assumption underlying cross-modal matching is modal balance, where each modality contains sufficient information to represent the others. However, noise interference and modality insufficiency often lead to modal imbalance, making it a common phenomenon in practice. The impact of imbalance on retrieval performance remains an open question. In this paper, we first demonstrate that ultimate cross-modal matching is generally sub-optimal for cross-modal retrieval when imbalanced modalities exist. The structure of instances in the common space is inherently influenced when facing imbalanced modalities, posing a challenge to cross-modal similarity measurement. To address this issue, we emphasize the importance of meaningful structure-preserved matching. Accordingly, we propose a simple yet effective method to rebalance cross-modal matching by learning structure-preserved matching representations. Specifically, we design a novel multi-granularity cross-modal matching that incorporates structure-aware distillation alongside the cross-modal matching loss. While the cross-modal matching loss constraints instance-level matching, the structure-aware distillation further regularizes the geometric consistency between learned matching representations and intra-modal representations through the developed relational matching. Extensive experiments on different datasets affirm the superior cross-modal retrieval performance of our approach, simultaneously enhancing single-modal retrieval capabilities compared to the baseline models.

Paperid: 6568, https://arxiv.org/pdf/2412.10720.pdf

Abstract:
Video captioning is a critical task in the field of multimodal machine learning, aiming to generate descriptive and coherent textual narratives for video content. While large vision-language models (LVLMs) have shown significant progress, they often struggle to capture the causal and temporal dynamics inherent in complex video sequences. To address this limitation, we propose an enhanced framework that integrates a Causal-Temporal Reasoning Module (CTRM) into state-of-the-art LVLMs. CTRM comprises two key components: the Causal Dynamics Encoder (CDE) and the Temporal Relational Learner (TRL), which collectively encode causal dependencies and temporal consistency from video frames. We further design a multi-stage learning strategy to optimize the model, combining pre-training on large-scale video-text datasets, fine-tuning on causally annotated data, and contrastive alignment for better embedding coherence. Experimental results on standard benchmarks such as MSVD and MSR-VTT demonstrate that our method outperforms existing approaches in both automatic metrics (CIDEr, BLEU-4, ROUGE-L) and human evaluations, achieving more fluent, coherent, and relevant captions. These results validate the effectiveness of our approach in generating captions with enriched causal-temporal narratives.

Paperid: 6569, https://arxiv.org/pdf/2412.09870.pdf

Abstract:
Semantic location prediction from multimodal social media posts is a critical task with applications in personalized services and human mobility analysis. This paper introduces \textit{Contextualized Vision-Language Alignment (CoVLA)}, a discriminative framework designed to address the challenges of contextual ambiguity and modality discrepancy inherent in this task. CoVLA leverages a Contextual Alignment Module (CAM) to enhance cross-modal feature alignment and a Cross-modal Fusion Module (CMF) to dynamically integrate textual and visual information. Extensive experiments on a benchmark dataset demonstrate that CoVLA significantly outperforms state-of-the-art methods, achieving improvements of 2.3\% in accuracy and 2.5\% in F1-score. Ablation studies validate the contributions of CAM and CMF, while human evaluations highlight the contextual relevance of the predictions. Additionally, robustness analysis shows that CoVLA maintains high performance under noisy conditions, making it a reliable solution for real-world applications. These results underscore the potential of CoVLA in advancing semantic location prediction research.

Paperid: 6570, https://arxiv.org/pdf/2412.09668.pdf

Abstract:
Vision-Language Models (VLMs) extend Large Language Models' capabilities by integrating image processing, but concerns persist about their potential to reproduce and amplify human biases. While research has documented how these models perpetuate stereotypes across demographic groups, most work has focused on between-group biases rather than within-group differences. This study investigates homogeneity bias-the tendency to portray groups as more uniform than they are-within Black Americans, examining how perceived racial phenotypicality influences VLMs' outputs. Using computer-generated images that systematically vary in phenotypicality, we prompted VLMs to generate stories about these individuals and measured text similarity to assess content homogeneity. Our findings reveal three key patterns: First, VLMs generate significantly more homogeneous stories about Black individuals with higher phenotypicality compared to those with lower phenotypicality. Second, stories about Black women consistently display greater homogeneity than those about Black men across all models tested. Third, in two of three VLMs, this homogeneity bias is primarily driven by a pronounced interaction where phenotypicality strongly influences content variation for Black women but has minimal impact for Black men. These results demonstrate how intersectionality shapes AI-generated representations and highlight the persistence of stereotyping that mirror documented biases in human perception, where increased racial phenotypicality leads to greater stereotyping and less individualized representation.

Paperid: 6571, https://arxiv.org/pdf/2412.07077.pdf

Abstract:
The advancement of vision-language models, particularly the Contrastive Language-Image Pre-training (CLIP) model, has revolutionized the field of machine learning by enabling robust zero-shot learning capabilities. These capabilities allow models to understand and respond to previously unseen data without task-specific training. However, adapting CLIP to integrate specialized knowledge from various domains while retaining its zero-shot capabilities remains a significant challenge. To address this, we introduce a novel prompt ensemble learning approach called Group-wise Prompt Ensemble (GPE). This method aims to enhance CLIP's zero-shot capabilities by incorporating new domain knowledge while improving its adaptability and robustness against data distribution shifts. Our approach hinges on three main strategies: prompt grouping with masked attention to optimize CLIP's adaptability while safeguarding its zero-shot capabilities; the incorporation of auxiliary prompts for the seamless integration of new domain insights without disrupting the original model's representation; and an ensemble learning strategy that effectively merges original and new knowledge. Through rigorous experimentation, including more challenging cross-dataset transfer evaluations, our GPE method redefines the benchmarks for the adaptability and efficiency of vision-language models, surpassing existing models across various scenarios.

Paperid: 6572, https://arxiv.org/pdf/2412.02803.pdf

Abstract:
3D Gaussian Splatting has advanced radiance field reconstruction, enabling high-quality view synthesis and fast rendering in 3D modeling. While adversarial attacks on object detection models are well-studied for 2D images, their impact on 3D models remains underexplored. This work introduces the Masked Iterative Fast Gradient Sign Method (M-IFGSM), designed to generate adversarial noise targeting the CLIP vision-language model. M-IFGSM specifically alters the object of interest by focusing perturbations on masked regions, degrading the performance of CLIP's zero-shot object detection capability when applied to 3D models. Using eight objects from the Common Objects 3D (CO3D) dataset, we demonstrate that our method effectively reduces the accuracy and confidence of the model, with adversarial noise being nearly imperceptible to human observers. The top-1 accuracy in original model renders drops from 95.4\% to 12.5\% for train images and from 91.2\% to 35.4\% for test images, with confidence levels reflecting this shift from true classification to misclassification, underscoring the risks of adversarial attacks on 3D models in applications such as autonomous driving, robotics, and surveillance. The significance of this research lies in its potential to expose vulnerabilities in modern 3D vision models, including radiance fields, prompting the development of more robust defenses and security measures in critical real-world applications.

Paperid: 6573, https://arxiv.org/pdf/2411.18038.pdf

Abstract:
The Large Vision Language Model (VLM) has recently addressed remarkable progress in bridging two fundamental modalities. VLM, trained by a sufficiently large dataset, exhibits a comprehensive understanding of both visual and linguistic to perform diverse tasks. To distill this knowledge accurately, in this paper, we introduce a novel approach that explicitly utilizes VLM as an objective function form for the Human-Object Interaction (HOI) detection task (\textbf{VLM-HOI}). Specifically, we propose a method that quantifies the similarity of the predicted HOI triplet using the Image-Text matching technique. We represent HOI triplets linguistically to fully utilize the language comprehension of VLMs, which are more suitable than CLIP models due to their localization and object-centric nature. This matching score is used as an objective for contrastive optimization. To our knowledge, this is the first utilization of VLM language abilities for HOI detection. Experiments demonstrate the effectiveness of our method, achieving state-of-the-art HOI detection accuracy on benchmarks. We believe integrating VLMs into HOI detection represents important progress towards more advanced and interpretable analysis of human-object interactions.

Paperid: 6574, https://arxiv.org/pdf/2411.15628.pdf

Abstract:
Vision-language models (VLMs) are capable of recognizing unseen actions. However, existing VLMs lack intrinsic understanding of procedural action concepts. Hence, they overfit to fixed labels and are not invariant to unseen action synonyms. To address this, we propose a simple fine-tuning technique, Action Concept Enhancement (ACE), to improve the robustness and concept understanding of VLMs in procedural action classification. ACE continually incorporates augmented action synonyms and negatives in an auxiliary classification loss by stochastically replacing fixed labels during training. This creates new combinations of action labels over the course of fine-tuning and prevents overfitting to fixed action representations. We show the enhanced concept understanding of our VLM, by visualizing the alignment of encoded embeddings of unseen action synonyms in the embedding space. Our experiments on the ATA, IKEA and GTEA datasets demonstrate the efficacy of ACE in domains of cooking and assembly leading to significant improvements in zero-shot action classification while maintaining competitive performance on seen actions.

Paperid: 6575, https://arxiv.org/pdf/2411.15207.pdf

Abstract:
Recent advancements in vision-language pre-training via contrastive learning have significantly improved performance across computer vision tasks. However, in the medical domain, obtaining multimodal data is often costly and challenging due to privacy, sensitivity, and annotation complexity. To mitigate data scarcity while boosting model performance, we introduce \textbf{Uni-Mlip}, a unified self-supervision framework specifically designed to enhance medical vision-language pre-training. Uni-Mlip seamlessly integrates cross-modality, uni-modality, and fused-modality self-supervision techniques at the data-level and the feature-level. Additionally, Uni-Mlip tailors uni-modal image self-supervision to accommodate the unique characteristics of medical images. Our experiments across datasets of varying scales demonstrate that Uni-Mlip significantly surpasses current state-of-the-art methods in three key downstream tasks: image-text retrieval, image classification, and visual question answering (VQA).

Paperid: 6576, https://arxiv.org/pdf/2411.10857.pdf

Abstract:
Remote Sensing Visual Question Answering (RSVQA) is a challenging task that involves interpreting complex satellite imagery to answer natural language questions. Traditional approaches often rely on separate visual feature extractors and language processing models, which can be computationally intensive and limited in their ability to handle open-ended questions. In this paper, we propose a novel method that leverages a generative Large Vision-Language Model (LVLM) to streamline the RSVQA process. Our approach consists of a two-step training strategy: domain-adaptive pretraining and prompt-based finetuning. This method enables the LVLM to generate natural language answers by conditioning on both visual and textual inputs, without the need for predefined answer categories. We evaluate our model on the RSVQAxBEN dataset, demonstrating superior performance compared to state-of-the-art baselines. Additionally, a human evaluation study shows that our method produces answers that are more accurate, relevant, and fluent. The results highlight the potential of generative LVLMs in advancing the field of remote sensing analysis.

Paperid: 6577, https://arxiv.org/pdf/2411.08923.pdf

Abstract:
Contrastive learning models have demonstrated impressive abilities to capture semantic similarities by aligning representations in the embedding space. However, their performance can be limited by the quality of the training data and its inherent biases. While Preference Optimization (PO) methods such as Reinforcement Learning from Human Feedback (RLHF) and Direct Preference Optimization (DPO) have been applied to align generative models with human preferences, their use in contrastive learning has yet to be explored. This paper introduces a novel method for training contrastive learning models using different PO methods to break down complex concepts. Our method systematically aligns model behavior with desired preferences, enhancing performance on the targeted task. In particular, we focus on enhancing model robustness against typographic attacks and inductive biases, commonly seen in contrastive vision-language models like CLIP. Our experiments demonstrate that models trained using PO outperform standard contrastive learning techniques while retaining their ability to handle adversarial challenges and maintain accuracy on other downstream tasks. This makes our method well-suited for tasks requiring fairness, robustness, and alignment with specific preferences. We evaluate our method for tackling typographic attacks on images and explore its ability to disentangle gender concepts and mitigate gender bias, showcasing the versatility of our approach.

Paperid: 6578, https://arxiv.org/pdf/2411.08727.pdf

Abstract:
Robots in human-centered environments require accurate scene understanding to perform high-level tasks effectively. This understanding can be achieved through instance-aware semantic mapping, which involves reconstructing elements at the level of individual instances. Neural networks, the de facto solution for scene understanding, still face limitations such as overconfident incorrect predictions with out-of-distribution objects or generating inaccurate masks.Placing excessive reliance on these predictions makes the reconstruction susceptible to errors, reducing the robustness of the resulting maps and hampering robot operation. In this work, we propose Voxeland, a probabilistic framework for incrementally building instance-aware semantic maps. Inspired by the Theory of Evidence, Voxeland treats neural network predictions as subjective opinions regarding map instances at both geometric and semantic levels. These opinions are aggregated over time to form evidences, which are formalized through a probabilistic model. This enables us to quantify uncertainty in the reconstruction process, facilitating the identification of map areas requiring improvement (e.g. reobservation or reclassification). As one strategy to exploit this, we incorporate a Large Vision-Language Model (LVLM) to perform semantic level disambiguation for instances with high uncertainty. Results from the standard benchmarking on the publicly available SceneNN dataset demonstrate that Voxeland outperforms state-of-the-art methods, highlighting the benefits of incorporating and leveraging both instance- and semantic-level uncertainties to enhance reconstruction robustness. This is further validated through qualitative experiments conducted on the real-world ScanNet dataset.

Paperid: 6579, https://arxiv.org/pdf/2410.23370.pdf

Abstract:
Methods based on Contrastive Language-Image Pre-training (CLIP) are nowadays extensively used in support of vision-and-language tasks involving remote sensing data, such as cross-modal retrieval. The adaptation of CLIP to this specific domain has relied on model fine-tuning with the standard contrastive objective, using existing human-labeled image-caption datasets, or using synthetic data corresponding to image-caption pairs derived from other annotations over remote sensing images (e.g., object classes). The use of different pre-training mechanisms has received less attention, and only a few exceptions have considered multilingual inputs. This work proposes a novel vision-and-language model for the remote sensing domain, exploring the fine-tuning of a multilingual CLIP model and testing the use of a self-supervised method based on aligning local and global representations from individual input images, together with the standard CLIP objective. Model training relied on assembling pre-existing datasets of remote sensing images paired with English captions, followed by the use of automated machine translation into nine additional languages. We show that translated data is indeed helpful, e.g. improving performance also on English. Our resulting model, which we named Remote Sensing Multilingual CLIP (RS-M-CLIP), obtains state-of-the-art results in a variety of vision-and-language tasks, including cross-modal and multilingual image-text retrieval, or zero-shot image classification.

Paperid: 6580, https://arxiv.org/pdf/2410.17827.pdf

Abstract:
In this paper we introduce RE-tune, a novel approach for fine-tuning pre-trained Multimodal Biomedical Vision-Language models (VLMs) in Incremental Learning scenarios for multi-label chest disease diagnosis. RE-tune freezes the backbones and only trains simple adaptors on top of the Image and Text encoders of the VLM. By engineering positive and negative text prompts for diseases, we leverage the ability of Large Language Models to steer the training trajectory. We evaluate RE-tune in three realistic incremental learning scenarios: class-incremental, label-incremental, and data-incremental. Our results demonstrate that Biomedical VLMs are natural continual learners and prevent catastrophic forgetting. RE-tune not only achieves accurate multi-label classification results, but also prioritizes patient privacy and it distinguishes itself through exceptional computational efficiency, rendering it highly suitable for broad adoption in real-world healthcare settings.

Paperid: 6581, https://arxiv.org/pdf/2410.13611.pdf

Abstract:
Smaller vision-language models (VLMs) are becoming increasingly important for privacy-focused, on-device applications due to their ability to run efficiently on consumer hardware for processing enterprise commercial documents and images. These models require strong language understanding and visual capabilities to enhance human-machine interaction. To address this need, we present H2OVL-Mississippi, a pair of small VLMs trained on 37 million image-text pairs using 240 hours of compute on 8 x H100 GPUs. H2OVL-Mississippi-0.8B is a tiny model with 0.8 billion parameters that specializes in text recognition, achieving state of the art performance on the Text Recognition portion of OCRBench and surpassing much larger models in this area. Additionally, we are releasing H2OVL-Mississippi-2B, a 2 billion parameter model for general use cases, exhibiting highly competitive metrics across various academic benchmarks. Both models build upon our prior work with H2O-Danube language models, extending their capabilities into the visual domain. We release them under the Apache 2.0 license, making VLMs accessible to everyone, democratizing document AI and visual LLMs.

Paperid: 6582, https://arxiv.org/pdf/2410.13510.pdf

Abstract:
Geometry problem-solving demands advanced reasoning abilities to process multimodal inputs and employ mathematical knowledge effectively. Vision-language models (VLMs) have made significant progress in various multimodal tasks. Yet, they still struggle with geometry problems and are significantly limited by their inability to perform mathematical operations not seen during pre-training, such as calculating the cosine of an arbitrary angle, and by difficulties in correctly applying relevant geometry formulas. To overcome these challenges, we present GeoCoder, which leverages modular code-finetuning to generate and execute code using a predefined geometry function library. By executing the code, we achieve accurate and deterministic calculations, contrasting the stochastic nature of autoregressive token prediction, while the function library minimizes errors in formula usage. We also propose a multimodal retrieval-augmented variant of GeoCoder, named RAG-GeoCoder, which incorporates a non-parametric memory module for retrieving functions from the geometry library, thereby reducing reliance on parametric memory. Our modular code-finetuning approach enhances the geometric reasoning capabilities of VLMs, yielding an average improvement of over 16% across various question complexities on the GeomVerse dataset compared to other finetuning methods.

Paperid: 6583, https://arxiv.org/pdf/2410.12157.pdf

Abstract:
With the rapid development of web technology, more and more software applications have become web-based in the past decades. To ensure software quality and user experience, various techniques have been proposed to automatically test web applications by interacting with their GUIs. To achieve high functional coverage, web GUI testing tools often need to generate high-quality text inputs and interact with the associated GUI elements (e.g., click submit buttons). However, developing a holistic approach that solves both subtasks is challenging because the web GUI context can be complicated and highly dynamic, which makes it hard to process programmatically. The recent development of large vision-language models (LVLM) provides new opportunities to handle these longstanding problems. This paper proposes VETL, the first LVLM-driven end-to-end web testing technique. With LVLM's scene understanding capabilities, VETL can generate valid and meaningful text inputs focusing on the local context, while avoiding the need to extract precise textual attributes. The selection of associated GUI elements is formulated as a visual question-answering problem, allowing LVLM to capture the logical connection between the input box and the relevant element based on visual instructions. Further, the GUI exploration is guided by a multi-armed bandit module employing a curiosity-oriented strategy. Experiments show that VETL effectively explores web state/action spaces and detects bugs. Compared with WebExplor, the state-of-the-art web testing technique, VETL can discover 25% more unique web actions on benchmark websites. Moreover, it can expose functional bugs in top-ranking commercial websites, which the website maintainers have confirmed. Our work makes the first attempt at leveraging LVLM in end-to-end GUI testing, demonstrating promising results in this research direction.

Paperid: 6584, https://arxiv.org/pdf/2410.09013.pdf

Abstract:
The glyphic writing system of Chinese incorporates information-rich visual features in each character, such as radicals that provide hints about meaning or pronunciation. However, there has been no investigation into whether contemporary Large Language Models (LLMs) and Vision-Language Models (VLMs) can harness these sub-character features in Chinese through prompting. In this study, we establish a benchmark to evaluate LLMs' and VLMs' understanding of visual elements in Chinese characters, including radicals, composition structures, strokes, and stroke counts. Our results reveal that models surprisingly exhibit some, but still limited, knowledge of the visual information, regardless of whether images of characters are provided. To incite models' ability to use radicals, we further experiment with incorporating radicals into the prompts for Chinese language processing (CLP) tasks. We observe consistent improvement in Part-Of-Speech tagging when providing additional information about radicals, suggesting the potential to enhance CLP by integrating sub-character information.

Paperid: 6585, https://arxiv.org/pdf/2410.07173.pdf

Abstract:
How well do text-only Large Language Models (LLMs) naturally align with the visual world? We provide the first direct analysis by utilizing frozen text representations in a discriminative vision-language model framework and measuring zero-shot generalization on unseen classes. We find decoder-based LLMs exhibit high intrinsic visual alignment. In particular, more capable LLMs reliably demonstrate stronger generalization. Moreover, utilizing frozen LLMs leads to strong gains in cross-lingual settings, where our approach surpasses CLIP's accuracy of 1.4% with 38.7% for Chinese. Our proposed method improves both robustness and generalization and also significantly reduces the need for paired data and compute, making vision-language models more accessible and adaptable.

Paperid: 6586, https://arxiv.org/pdf/2410.02049.pdf

Abstract:
Existing 3D facial emotion modeling have been constrained by limited emotion classes and insufficient datasets. This paper introduces "Emo3D", an extensive "Text-Image-Expression dataset" spanning a wide spectrum of human emotions, each paired with images and 3D blendshapes. Leveraging Large Language Models (LLMs), we generate a diverse array of textual descriptions, facilitating the capture of a broad spectrum of emotional expressions. Using this unique dataset, we conduct a comprehensive evaluation of language-based models' fine-tuning and vision-language models like Contranstive Language Image Pretraining (CLIP) for 3D facial expression synthesis. We also introduce a new evaluation metric for this task to more directly measure the conveyed emotion. Our new evaluation metric, Emo3D, demonstrates its superiority over Mean Squared Error (MSE) metrics in assessing visual-text alignment and semantic richness in 3D facial expressions associated with human emotions. "Emo3D" has great applications in animation design, virtual reality, and emotional human-computer interaction.

Paperid: 6587, https://arxiv.org/pdf/2410.00064.pdf

Abstract:
Lifelong imitation learning for manipulation tasks poses significant challenges due to distribution shifts that occur in incremental learning steps. Existing methods often focus on unsupervised skill discovery to construct an ever-growing skill library or distillation from multiple policies, which can lead to scalability issues as diverse manipulation tasks are continually introduced and may fail to ensure a consistent latent space throughout the learning process, leading to catastrophic forgetting of previously learned skills. In this paper, we introduce M2Distill, a multi-modal distillation-based method for lifelong imitation learning focusing on preserving consistent latent space across vision, language, and action distributions throughout the learning process. By regulating the shifts in latent representations across different modalities from previous to current steps, and reducing discrepancies in Gaussian Mixture Model (GMM) policies between consecutive learning steps, we ensure that the learned policy retains its ability to perform previously learned tasks while seamlessly integrating new skills. Extensive evaluations on the LIBERO lifelong imitation learning benchmark suites, including LIBERO-OBJECT, LIBERO-GOAL, and LIBERO-SPATIAL, demonstrate that our method consistently outperforms prior state-of-the-art methods across all evaluated metrics.

Paperid: 6588, https://arxiv.org/pdf/2409.12011.pdf

Abstract:
As powerful pre-trained vision-language models (VLMs) like CLIP gain prominence, numerous studies have attempted to combine VLMs for downstream tasks. Among these, prompt learning has been validated as an effective method for adapting to new tasks, which only requiring a small number of parameters. However, current prompt learning methods face two challenges: first, a single soft prompt struggles to capture the diverse styles and patterns within a dataset; second, fine-tuning soft prompts is prone to overfitting. To address these challenges, we propose a mixture of soft prompt learning method incorporating a routing module. This module is able to capture a dataset's varied styles and dynamically selects the most suitable prompts for each instance. Additionally, we introduce a novel gating mechanism to ensure the router selects prompts based on their similarity to hard prompt templates, which both retaining knowledge from hard prompts and improving selection accuracy. We also implement semantically grouped text-level supervision, initializing each soft prompt with the token embeddings of manually designed templates from its group and applied a contrastive loss between the resulted text feature and hard prompt encoded text feature. This supervision ensures that the text features derived from soft prompts remain close to those from their corresponding hard prompts, preserving initial knowledge and mitigating overfitting. Our method has been validated on 11 datasets, demonstrating evident improvements in few-shot learning, domain generalization, and base-to-new generalization scenarios compared to existing baselines. The code will be available at \url{https://anonymous.4open.science/r/mocoop-6387}

Paperid: 6589, https://arxiv.org/pdf/2409.07078.pdf

Abstract:
In this paper, we present our solution for the Second Multimodal Emotion Recognition Challenge Track 1(MER2024-SEMI). To enhance the accuracy and generalization performance of emotion recognition, we propose several methods for Multimodal Emotion Recognition. Firstly, we introduce EmoVCLIP, a model fine-tuned based on CLIP using vision-language prompt learning, designed for video-based emotion recognition tasks. By leveraging prompt learning on CLIP, EmoVCLIP improves the performance of pre-trained CLIP on emotional videos. Additionally, to address the issue of modality dependence in multimodal fusion, we employ modality dropout for robust information fusion. Furthermore, to aid Baichuan in better extracting emotional information, we suggest using GPT-4 as the prompt for Baichuan. Lastly, we utilize a self-training strategy to leverage unlabeled videos. In this process, we use unlabeled videos with high-confidence pseudo-labels generated by our model and incorporate them into the training set. Experimental results demonstrate that our model ranks 1st in the MER2024-SEMI track, achieving an accuracy of 90.15% on the test set.

Paperid: 6590, https://arxiv.org/pdf/2409.02522.pdf

Abstract:
Vision Language Navigation in Continuous Environments (VLN-CE) represents a frontier in embodied AI, demanding agents to navigate freely in unbounded 3D spaces solely guided by natural language instructions. This task introduces distinct challenges in multimodal comprehension, spatial reasoning, and decision-making. To address these challenges, we introduce Cog-GA, a generative agent founded on large language models (LLMs) tailored for VLN-CE tasks. Cog-GA employs a dual-pronged strategy to emulate human-like cognitive processes. Firstly, it constructs a cognitive map, integrating temporal, spatial, and semantic elements, thereby facilitating the development of spatial memory within LLMs. Secondly, Cog-GA employs a predictive mechanism for waypoints, strategically optimizing the exploration trajectory to maximize navigational efficiency. Each waypoint is accompanied by a dual-channel scene description, categorizing environmental cues into 'what' and 'where' streams as the brain. This segregation enhances the agent's attentional focus, enabling it to discern pertinent spatial information for navigation. A reflective mechanism complements these strategies by capturing feedback from prior navigation experiences, facilitating continual learning and adaptive replanning. Extensive evaluations conducted on VLN-CE benchmarks validate Cog-GA's state-of-the-art performance and ability to simulate human-like navigation behaviors. This research significantly contributes to the development of strategic and interpretable VLN-CE agents.

Paperid: 6591, https://arxiv.org/pdf/2409.00106.pdf

Abstract:
Vision-language models (VLMs) have shown impressive zero- and few-shot performance on real-world visual question answering (VQA) benchmarks, alluding to their capabilities as visual reasoning engines. However, the benchmarks being used conflate "pure" visual reasoning with world knowledge, and also have questions that involve a limited number of reasoning steps. Thus, it remains unclear whether a VLM's apparent visual reasoning performance is due to its world knowledge, or due to actual visual reasoning capabilities. To clarify this ambiguity, we systematically benchmark and dissect the zero-shot visual reasoning capabilities of VLMs through synthetic datasets that require minimal world knowledge, and allow for analysis over a broad range of reasoning steps. We focus on two novel aspects of zero-shot visual reasoning: i) evaluating the impact of conveying scene information as either visual embeddings or purely textual scene descriptions to the underlying large language model (LLM) of the VLM, and ii) comparing the effectiveness of chain-of-thought prompting to standard prompting for zero-shot visual reasoning. We find that the underlying LLMs, when provided textual scene descriptions, consistently perform better compared to being provided visual embeddings. In particular, 18% higher accuracy is achieved on the PTR dataset. We also find that CoT prompting performs marginally better than standard prompting only for the comparatively large GPT-3.5-Turbo (175B) model, and does worse for smaller-scale models. This suggests the emergence of CoT abilities for visual reasoning in LLMs at larger scales even when world knowledge is limited. Overall, we find limitations in the abilities of VLMs and LLMs for more complex visual reasoning, and highlight the important role that LLMs can play in visual reasoning.

Paperid: 6592, https://arxiv.org/pdf/2601.22114.pdf

Abstract:
Current methods for converting circuit schematic images into machine-readable netlists struggle with component recognition and connectivity inference. In this paper, we present SINA, an open-source, fully automated circuit schematic image-to-netlist generator. SINA integrates deep learning for accurate component detection, Connected-Component Labeling (CCL) for precise connectivity extraction, and Optical Character Recognition (OCR) for component reference designator retrieval, while employing a Vision-Language Model (VLM) for reliable reference designator assignments. In our experiments, SINA achieves 96.47% overall netlist-generation accuracy, which is 2.72x higher than state-of-the-art approaches.

Paperid: 6593, https://arxiv.org/pdf/2601.21506.pdf

Abstract:
Indoor mobile robot navigation requires fast responsiveness and robust semantic understanding, yet existing methods struggle to provide both. Classical geometric approaches such as SLAM offer reliable localization but depend on detailed maps and cannot interpret human-targeted cues (e.g., signs, room numbers) essential for indoor reasoning. Vision-Language-Action (VLA) models introduce semantic grounding but remain strictly reactive, basing decisions only on visible frames and failing to anticipate unseen intersections or reason about distant textual cues. Vision-Language Models (VLMs) provide richer contextual inference but suffer from high computational latency, making them unsuitable for real-time operation on embedded platforms. In this work, we present IROS, a real-time navigation framework that combines VLM-level contextual reasoning with the efficiency of lightweight perceptual modules on low-cost, on-device hardware. Inspired by Dual Process Theory, IROS separates fast reflexive decisions (System One) from slow deliberative reasoning (System Two), invoking the VLM only when necessary. Furthermore, by augmenting compact VLMs with spatial and textual cues, IROS delivers robust, human-like navigation with minimal latency. Across five real-world buildings, IROS improves decision accuracy and reduces latency by 66% compared to continuous VLM-based navigation.

Paperid: 6594, https://arxiv.org/pdf/2601.21199.pdf

Abstract:
When large vision-language models are applied to the field of robotics, they encounter problems that are simple for humans yet error-prone for models. Such issues include confusion between third-person and first-person perspectives and a tendency to overlook information in video endings during temporal reasoning. To address these challenges, we propose Thinker, a large vision-language foundation model designed for embodied intelligence. We tackle the aforementioned issues from two perspectives. Firstly, we construct a large-scale dataset tailored for robotic perception and reasoning, encompassing ego-view videos, visual grounding, spatial understanding, and chain-of-thought data. Secondly, we introduce a simple yet effective approach that substantially enhances the model's capacity for video comprehension by jointly incorporating key frames and full video sequences as inputs. Our model achieves state-of-the-art results on two of the most commonly used benchmark datasets in the field of task planning.

Paperid: 6595, https://arxiv.org/pdf/2601.20168.pdf

Abstract:
Diffusion-based large multimodal models, such as LLaDA-V, have demonstrated impressive capabilities in vision-language understanding and generation. However, their bidirectional attention mechanism and diffusion-style iterative denoising paradigm introduce significant computational overhead, as visual tokens are repeatedly processed across all layers and denoising steps. In this work, we conduct an in-depth attention analysis and reveal that, unlike autoregressive decoders, LLaDA-V aggregates cross-modal information predominantly in middle-to-late layers, leading to delayed semantic alignment. Motivated by this observation, we propose a structured token pruning strategy inspired by FastV, selectively removing a proportion of visual tokens at designated layers to reduce FLOPs while preserving critical semantic information. To the best of our knowledge, this is the first work to investigate structured token pruning in diffusion-based large multimodal models. Unlike FastV, which focuses on shallow-layer pruning, our method targets the middle-to-late layers of the first denoising step to align with LLaDA-V's delayed attention aggregation to maintain output quality, and the first-step pruning strategy reduces the computation across all subsequent steps. Our framework provides an empirical basis for efficient LLaDA-V inference and highlights the potential of vision-aware pruning in diffusion-based multimodal models. Across multiple benchmarks, our best configuration reduces computational cost by up to 65% while preserving an average of 95% task performance.

Paperid: 6596, https://arxiv.org/pdf/2601.20107.pdf

Abstract:
Recent Vision-Language Models (e.g., ColPali) enable fine-grained Visual Document Retrieval (VDR) but incur prohibitive index vector size overheads. Training-free pruning solutions (e.g., EOS-attention based methods) can reduce index vector size by approximately 60% without model adaptation, but often underperform random selection in high-compression scenarios (> 80%). Prior research (e.g., Light-ColPali) attributes this to the conclusion that visual token importance is inherently query-dependent, thereby questioning the feasibility of training-free pruning. In this work, we propose Structural Anchor Pruning (SAP), a training-free pruning method that identifies key visual patches from middle layers to achieve high performance compression. We also introduce Oracle Score Retention (OSR) protocol to evaluate how layer-wise information affects compression efficiency. Evaluations on the ViDoRe benchmark demonstrate that SAP reduces index vectors by over 90% while maintaining robust retrieval fidelity, providing a highly scalable solution for Visual RAG. Furthermore, our OSR-based analysis reveals that semantic structural anchor patches persist in the middle layers, unlike traditional pruning solutions that focus on the final layer where structural signals dissipate.

Paperid: 6597, https://arxiv.org/pdf/2601.16378.pdf

Abstract:
Multimodal language models (MLMs) perform well on semantic vision-language tasks but fail at spatial reasoning that requires adopting another agent's visual perspective. These errors reflect a persistent egocentric bias and raise questions about whether current models support allocentric reasoning. Inspired by human spatial cognition, we introduce perspective tokens, specialized embeddings that encode orientation through either (1) embodied body-keypoint cues or (2) abstract representations supporting mental rotation. Integrating these tokens into LLaVA-1.5-13B yields performance on level-2 visual perspective-taking tasks. Across synthetic and naturalistic benchmarks (Isle Bricks V2, COCO, 3DSRBench), perspective tokens improve accuracy, with rotation-based tokens generalizing to non-human reference agents. Representational analyses reveal that fine-tuning enhances latent orientation sensitivity already present in the base model, suggesting that MLMs contain precursors of allocentric reasoning but lack appropriate internal structure. Overall, embedding cognitively grounded spatial structure directly into token space provides a lightweight, model-agnostic mechanism for perspective-taking and more human-like spatial reasoning.

Paperid: 6598, https://arxiv.org/pdf/2601.15711.pdf

Abstract:
Fine-grained attribute prediction is essential for fashion retail applications including catalog enrichment, visual search, and recommendation systems. Vision-Language Models (VLMs) offer zero-shot prediction without task-specific training, yet their systematic evaluation on multi-attribute fashion tasks remains underexplored. A key challenge is that fashion attributes are often conditional. For example, "outer fabric" is undefined when no outer garment is visible. This requires models to detect attribute applicability before attempting classification. We introduce a three-tier evaluation framework that decomposes this challenge: (1) overall task performance across all classes (including NA class: suggesting attribute is not applicable) for all attributes, (2) attribute applicability detection, and (3) fine-grained classification when attributes are determinable. Using DeepFashion-MultiModal, which explicitly defines NA (meaning attribute doesn't exist or is not visible) within attribute label spaces, we benchmark nine VLMs spanning flagship (GPT-5, Gemini 2.5 Pro), efficient (GPT-5 Mini, Gemini 2.5 Flash), and ultra-efficient tiers (GPT-5 Nano, Gemini 2.5 Flash-Lite) against classifiers trained on pretrained Fashion-CLIP embeddings on 5,000 images across 18 attributes. Our findings reveal that: (1) zero-shot VLMs achieve 64.0% macro-F1, a threefold improvement over logistic regression on pretrained Fashion-CLIP embeddings; (2) VLMs excel at fine-grained classification (Tier 3: 70.8% F1) but struggle with applicability detection (Tier 2: 34.1% NA-F1), identifying a key bottleneck; (3) efficient models achieve over 90% of flagship performance at lower cost, offering practical deployment paths. This diagnostic framework enables practitioners to pinpoint whether errors stem from visibility detection or classification, guiding targeted improvements for production systems.

Paperid: 6599, https://arxiv.org/pdf/2601.14978.pdf

Abstract:
Text-Based Person Search (TBPS) has seen significant progress with vision-language models (VLMs), yet it remains constrained by limited training data and the fact that VLMs are not inherently pre-trained for pedestrian-centric recognition. Existing TBPS methods therefore rely on dataset-centric fine-tuning to handle distribution shift, resulting in multiple independently trained models for different datasets. While synthetic data can increase the scale needed to fine-tune VLMs, it does not eliminate dataset-specific adaptation. This motivates a fundamental question: can we train a single unified TBPS model across multiple datasets? We show that naive joint training over all datasets remains sub-optimal because current training paradigms do not scale to a large number of unique person identities and are vulnerable to noisy image-text pairs. To address these challenges, we propose Scale-TBPS with two contributions: (i) a noise-aware unified dataset curation strategy that cohesively merges diverse TBPS datasets; and (ii) a scalable discriminative identity learning framework that remains effective under a large number of unique identities. Extensive experiments on CUHK-PEDES, ICFG-PEDES, RSTPReid, IIITD-20K, and UFine6926 demonstrate that a single Scale-TBPS model outperforms dataset-centric optimized models and naive joint training.

Paperid: 6600, https://arxiv.org/pdf/2601.14702.pdf

Abstract:
Autonomous driving is a highly challenging domain that requires reliable perception and safe decision-making in complex scenarios. Recent vision-language models (VLMs) demonstrate reasoning and generalization abilities, opening new possibilities for autonomous driving; however, existing benchmarks and metrics overemphasize perceptual competence and fail to adequately assess decision-making processes. In this work, we present AutoDriDM, a decision-centric, progressive benchmark with 6,650 questions across three dimensions - Object, Scene, and Decision. We evaluate mainstream VLMs to delineate the perception-to-decision capability boundary in autonomous driving, and our correlation analysis reveals weak alignment between perception and decision-making performance. We further conduct explainability analyses of models' reasoning processes, identifying key failure modes such as logical reasoning errors, and introduce an analyzer model to automate large-scale annotation. AutoDriDM bridges the gap between perception-centered and decision-centered evaluation, providing guidance toward safer and more reliable VLMs for real-world autonomous driving.

Paperid: 6601, https://arxiv.org/pdf/2601.14637.pdf

Abstract:
The increasing availability of high-resolution satellite imagery, together with advances in deep learning, creates new opportunities for enhancing forest monitoring workflows. Two central challenges in this domain are pixel-level change detection and semantic change interpretation, particularly for complex forest dynamics. While large language models (LLMs) are increasingly adopted for data exploration, their integration with vision-language models (VLMs) for remote sensing image change interpretation (RSICI) remains underexplored, especially beyond urban environments. We introduce Forest-Chat, an LLM-driven agent designed for integrated forest change analysis. The proposed framework enables natural language querying and supports multiple RSICI tasks, including change detection, change captioning, object counting, deforestation percentage estimation, and change reasoning. Forest-Chat builds upon a multi-level change interpretation (MCI) vision-language backbone with LLM-based orchestration, and incorporates zero-shot change detection via a foundation change detection model together with an interactive point-prompt interface to support fine-grained user guidance. To facilitate adaptation and evaluation in forest environments, we introduce the Forest-Change dataset, comprising bi-temporal satellite imagery, pixel-level change masks, and multi-granularity semantic change captions generated through a combination of human annotation and rule-based methods. Experimental results demonstrate that Forest-Chat achieves strong performance on Forest-Change and on LEVIR-MCI-Trees, a tree-focused subset of LEVIR-MCI, for joint change detection and captioning, highlighting the potential of interactive, LLM-driven RSICI systems to improve accessibility, interpretability, and analytical efficiency in forest change analysis.

Paperid: 6602, https://arxiv.org/pdf/2601.14587.pdf

Abstract:
Human interaction is essential for issuing personalized instructions and assisting robots when failure is likely. However, robots remain largely black boxes, offering users little insight into their evolving capabilities and limitations. To address this gap, we present explainable object-oriented HRI (X-OOHRI), an augmented reality (AR) interface that conveys robot action possibilities and constraints through visual signifiers, radial menus, color coding, and explanation tags. Our system encodes object properties and robot limits into object-oriented structures using a vision-language model, allowing explanation generation on the fly and direct manipulation of virtual twins spatially aligned within a simulated environment. We integrate the end-to-end pipeline with a physical robot and showcase diverse use cases ranging from low-level pick-and-place to high-level instructions. Finally, we evaluate X-OOHRI through a user study and find that participants effectively issue object-oriented commands, develop accurate mental models of robot limitations, and engage in mixed-initiative resolution.

Paperid: 6603, https://arxiv.org/pdf/2601.14490.pdf

Abstract:
GutenOCR is a family of grounded OCR front-ends obtained by fine-tuning Qwen2.5-VL-3B and Qwen2.5-VL-7B. The resulting single-checkpoint vision-language models expose reading, detection, and grounding through a unified, prompt-based interface. Trained on business documents, scientific articles, and synthetic grounding data, the models support full-page and localized reading with line- and paragraph-level bounding boxes and conditional ``where is x?'' queries. We introduce a grounded OCR evaluation protocol and show that GutenOCR-7B more than doubles the composite grounded OCR score of its Qwen2.5-VL-7B backbone on 10.5K held-out business and scientific pages (0.40 to 0.82). On Fox and OmniDocBench v1.5, our approach substantially improves region- and line-level OCR as well as text-detection recall, but reveals trade-offs in page-level linearization, color-guided OCR, and formula-heavy layouts.

Paperid: 6604, https://arxiv.org/pdf/2601.12010.pdf

Abstract:
The safety validation of autonomous robotic vehicles hinges on systematically testing their planning and control stacks against rare, safety-critical scenarios. Mining these long-tail events from massive real-world driving logs is therefore a critical step in the robotic development lifecycle. The goal of the Scenario Mining task is to retrieve useful information to enable targeted re-simulation, regression testing, and failure analysis of the robot's decision-making algorithms. RefAV, introduced by the Argoverse team, is an end-to-end framework that uses large language models (LLMs) to spatially and temporally localize scenarios described in natural language. However, this process performs retrieval on trajectory labels, ignoring the direct connection between natural language and raw RGB images, which runs counter to the intuition of video retrieval; it also depends on the quality of upstream 3D object detection and tracking. Further, inaccuracies in trajectory data lead to inaccuracies in downstream spatial and temporal localization. To address these issues, we propose Robust Scenario Mining for Robotic Autonomy from Coarse to Fine (SMc2f), a coarse-to-fine pipeline that employs vision-language models (VLMs) for coarse image-text filtering, builds a database of successful mining cases on top of RefAV and automatically retrieves exemplars to few-shot condition the LLM for more robust retrieval, and introduces text-trajectory contrastive learning to pull matched pairs together and push mismatched pairs apart in a shared embedding space, yielding a fine-grained matcher that refines the LLM's candidate trajectories. Experiments on public datasets demonstrate substantial gains in both retrieval quality and efficiency.

Paperid: 6605, https://arxiv.org/pdf/2601.09852.pdf

Abstract:
Language places subtle constraints on how we make inductive inferences. Developmental evidence by Gelman et al. (2002) has shown children (4 years and older) to differentiate among generic statements ("Bears are daxable"), universally quantified NPs ("all bears are daxable") and indefinite plural NPs ("some bears are daxable") in extending novel properties to a specific member (all > generics > some), suggesting that they represent these types of propositions differently. We test if these subtle differences arise in general purpose statistical learners like Vision Language Models, by replicating the original experiment. On tasking them through a series of precondition tests (robust identification of categories in images and sensitivities to all and some), followed by the original experiment, we find behavioral alignment between models and humans. Post-hoc analyses on their representations revealed that these differences are organized based on inductive constraints and not surface-form differences.

Paperid: 6606, https://arxiv.org/pdf/2601.09116.pdf

Abstract:
Real-world License Plate Recognition (LPR) faces significant challenges from severe degradations such as motion blur, low resolution, and complex illumination. The prevailing "restoration-then-recognition" two-stage paradigm suffers from a fundamental flaw: the pixel-level optimization objectives of image restoration models are misaligned with the semantic goals of character recognition, leading to artifact interference and error accumulation. While Vision-Language Models (VLMs) have demonstrated powerful general capabilities, they lack explicit structural modeling for license plate character sequences (e.g., fixed length, specific order). To address this, we propose an end-to-end structure-aware multimodal reasoning framework based on Qwen3-VL. The core innovation lies in the Character-Aware Multimodal Reasoning Module (CMRM), which introduces a set of learnable Character Slot Queries. Through a cross-attention mechanism, these queries actively retrieve fine-grained evidence corresponding to character positions from visual features. Subsequently, we inject these character-aware representations back into the visual tokens via residual modulation, enabling the language model to perform autoregressive generation based on explicit structural priors. Furthermore, combined with the LoRA parameter-efficient fine-tuning strategy, the model achieves domain adaptation while retaining the generalization capabilities of the large model. Extensive experiments on both synthetic and real-world severely degraded datasets demonstrate that our method significantly outperforms existing restoration-recognition combinations and general VLMs, validating the superiority of incorporating structured reasoning into large models for low-quality text recognition tasks.

Paperid: 6607, https://arxiv.org/pdf/2601.08470.pdf

Abstract:
Vision Language Models (VLMs) are increasingly deployed in autonomous vehicles and mobile systems, making it crucial to evaluate their ability to support safer decision-making in complex environments. However, existing benchmarks inadequately cover diverse hazardous situations, especially anomalous scenarios with spatio-temporal dynamics. While image editing models are a promising means to synthesize such hazards, it remains challenging to generate well-formulated scenarios that include moving, intrusive, and distant objects frequently observed in the real world. To address this gap, we introduce \textbf{HazardForge}, a scalable pipeline that leverages image editing models to generate these scenarios with layout decision algorithms, and validation modules. Using HazardForge, we construct \textbf{MovSafeBench}, a multiple-choice question (MCQ) benchmark comprising 7,254 images and corresponding QA pairs across 13 object categories, covering both normal and anomalous objects. Experiments using MovSafeBench show that VLM performance degrades notably under conditions including anomalous objects, with the largest drop in scenarios requiring nuanced motion understanding.

Paperid: 6608, https://arxiv.org/pdf/2601.08040.pdf

Abstract:
Scientific image manipulation in biomedical publications poses a growing threat to research integrity and reproducibility. Unlike natural image forensics, biomedical forgery detection is uniquely challenging due to domain-specific artifacts, complex textures, and unstructured figure layouts. We present the first vision-language guided framework for both generating and detecting biomedical image forgeries. By combining diffusion-based synthesis with vision-language prompting, our method enables realistic and semantically controlled manipulations, including duplication, splicing, and region removal, across diverse biomedical modalities. We introduce Rescind, a large-scale benchmark featuring fine-grained annotations and modality-specific splits, and propose Integscan, a structured state space modeling framework that integrates attention-enhanced visual encoding with prompt-conditioned semantic alignment for precise forgery localization. To ensure semantic fidelity, we incorporate a vision-language model based verification loop that filters generated forgeries based on consistency with intended prompts. Extensive experiments on Rescind and existing benchmarks demonstrate that Integscan achieves state of the art performance in both detection and localization, establishing a strong foundation for automated scientific integrity analysis.

Paperid: 6609, https://arxiv.org/pdf/2601.07795.pdf

Abstract:
The European Space Agency (ESA), driven by its ambitions on planned lunar missions with the Argonaut lander, has a profound interest in reliable crater detection, since craters pose a risk to safe lunar landings. This task is usually addressed with automated crater detection algorithms (CDA) based on deep learning techniques. It is non-trivial due to the vast amount of craters of various sizes and shapes, as well as challenging conditions such as varying illumination and rugged terrain. Therefore, we propose a deep-learning CDA based on the OWLv2 model, which is built on a Vision Transformer, that has proven highly effective in various computer vision tasks. For fine-tuning, we utilize a manually labeled dataset fom the IMPACT project, that provides crater annotations on high-resolution Lunar Reconnaissance Orbiter Camera Calibrated Data Record images. We insert trainable parameters using a parameter-efficient fine-tuning strategy with Low-Rank Adaptation, and optimize a combined loss function consisting of Complete Intersection over Union (CIoU) for localization and a contrastive loss for classification. We achieve satisfactory visual results, along with a maximum recall of 94.0% and a maximum precision of 73.1% on a test dataset from IMPACT. Our method achieves reliable crater detection across challenging lunar imaging conditions, paving the way for robust crater analysis in future lunar exploration.

Paperid: 6610, https://arxiv.org/pdf/2601.06097.pdf

Abstract:
Long-form video question answering remains challenging for modern vision-language models, which struggle to reason over hour-scale footage without exceeding practical token and compute budgets. Existing systems typically downsample frames or feed dense visual embeddings to large-context language models, trading off temporal coverage against cost. We propose Semantic Event Graphs (SEG), a lightweight symbolic interface between video and language that replaces raw frames with compact temporal interaction logs. Our pipeline detects and tracks objects with YOLOv11, converts proximity patterns into START/END human-object events, and organizes them into a Temporal Scene Graph (TSG). At inference time, a query-aware pruning module identifies anchor entities and lexically relevant events, returning only a small subgraph which is verbalized and passed to Gemini 2.5 Flash for answer generation. On five YouTube videos (300-500 interactions each) and 120 automatically generated long-horizon questions, SEG achieves 65.0% accuracy using only 3.47k tokens per query, closely matching a full-log baseline (62.5% at 40.39k tokens) while reducing token usage by 91.4%. A short-context baseline restricted to the last 30 seconds collapses to 2.5% accuracy, underscoring the need for explicit temporal memory. These results show that symbolic temporal graphs can serve as an effective, plug-and-play memory layer for off-the-shelf vision-language models, preserving long-range reasoning ability while making long-form video question answering substantially more token- and cost-efficient. Code, logs, and event-extraction tools will be released for reproducibility.

Paperid: 6611, https://arxiv.org/pdf/2601.05739.pdf

Abstract:
Vision Language Models (VLMs) are increasingly integrated into privacy-critical domains, yet existing evaluations of personally identifiable information (PII) leakage largely treat privacy as a static extraction task and ignore how a subject's online presence--the volume of their data available online--influences privacy alignment. We introduce PII-VisBench, a novel benchmark containing 4000 unique probes designed to evaluate VLM safety through the continuum of online presence. The benchmark stratifies 200 subjects into four visibility categories: high, medium, low, and zero--based on the extent and nature of their information available online. We evaluate 18 open-source VLMs (0.3B-32B) based on two key metrics: percentage of PII probing queries refused (Refusal Rate) and the fraction of non-refusal responses flagged for containing PII (Conditional PII Disclosure Rate). Across models, we observe a consistent pattern: refusals increase and PII disclosures decrease (9.10% high to 5.34% low) as subject visibility drops. We identify that models are more likely to disclose PII for high-visibility subjects, alongside substantial model-family heterogeneity and PII-type disparities. Finally, paraphrasing and jailbreak-style prompts expose attack and model-dependent failures, motivating visibility-aware safety evaluation and training interventions.

Paperid: 6612, https://arxiv.org/pdf/2601.05059.pdf

Abstract:
Vision Language Models (VLMs) are poised to revolutionize the digital transformation of pharmacyceutical industry by enabling intelligent, scalable, and automated multi-modality content processing. Traditional manual annotation of heterogeneous data modalities (text, images, video, audio, and web links), is prone to inconsistencies, quality degradation, and inefficiencies in content utilization. The sheer volume of long video and audio data further exacerbates these challenges, (e.g. long clinical trial interviews and educational seminars). Here, we introduce a domain adapted Video to Video Clip Generation framework that integrates Audio Language Models (ALMs) and Vision Language Models (VLMs) to produce highlight clips. Our contributions are threefold: (i) a reproducible Cut & Merge algorithm with fade in/out and timestamp normalization, ensuring smooth transitions and audio/visual alignment; (ii) a personalization mechanism based on role definition and prompt injection for tailored outputs (marketing, training, regulatory); (iii) a cost efficient e2e pipeline strategy balancing ALM/VLM enhanced processing. Evaluations on Video MME benchmark (900) and our proprietary dataset of 16,159 pharmacy videos across 14 disease areas demonstrate 3 to 4 times speedup, 4 times cost reduction, and competitive clip quality. Beyond efficiency gains, we also report our methods improved clip coherence scores (0.348) and informativeness scores (0.721) over state of the art VLM baselines (e.g., Gemini 2.5 Pro), highlighting the potential of transparent, custom extractive, and compliance supporting video summarization for life sciences.

Paperid: 6613, https://arxiv.org/pdf/2601.04891.pdf

Abstract:
Vision Language Models (VLMs) have shown strong performance on multimodal reasoning tasks, yet most evaluations focus on short videos and assume unconstrained computational resources. In industrial settings such as pharmaceutical content understanding, practitioners must process long-form videos under strict GPU, latency, and cost constraints, where many existing approaches fail to scale. In this work, we present an industrial GenAI framework that processes over 200,000 PDFs, 25,326 videos across eight formats (e.g., MP4, M4V, etc.), and 888 multilingual audio files in more than 20 languages. Our study makes three contributions: (i) an industrial large-scale architecture for multimodal reasoning in pharmaceutical domains; (ii) empirical analysis of over 40 VLMs on two leading benchmarks (Video-MME and MMBench) and proprietary dataset of 25,326 videos across 14 disease areas; and (iii) four findings relevant to long-form video reasoning: the role of multimodality, attention mechanism trade-offs, temporal reasoning limits, and challenges of video splitting under GPU constraints. Results show 3-8 times efficiency gains with SDPA attention on commodity GPUs, multimodality improving up to 8/12 task domains (especially length-dependent tasks), and clear bottlenecks in temporal alignment and keyframe detection across open- and closed-source VLMs. Rather than proposing a new "A+B" model, this paper characterizes practical limits, trade-offs, and failure patterns of current VLMs under realistic deployment constraints, and provide actionable guidance for both researchers and practitioners designing scalable multimodal systems for long-form video understanding in industrial domains.

Paperid: 6614, https://arxiv.org/pdf/2601.04672.pdf

Abstract:
Agricultural disease diagnosis challenges VLMs, as conventional fine-tuning requires extensive labels, lacks interpretability, and generalizes poorly. While reasoning improves model robustness, existing methods rely on costly expert annotations and rarely address the open-ended, diverse nature of agricultural queries. To address these limitations, we propose \textbf{Agri-R1}, a reasoning-enhanced large model for agriculture. Our framework automates high-quality reasoning data generation via vision-language synthesis and LLM-based filtering, using only 19\% of available samples. Training employs Group Relative Policy Optimization (GRPO) with a novel proposed reward function that integrates domain-specific lexicons and fuzzy matching to assess both correctness and linguistic flexibility in open-ended responses. Evaluated on CDDMBench, our resulting 3B-parameter model achieves performance competitive with 7B- to 13B-parameter baselines, showing a +23.2\% relative gain in disease recognition accuracy, +33.3\% in agricultural knowledge QA, and a +26.10-point improvement in cross-domain generalization over standard fine-tuning. Ablation studies confirm that the synergy between structured reasoning data and GRPO-driven exploration underpins these gains, with benefits scaling as question complexity increases.

Paperid: 6615, https://arxiv.org/pdf/2601.03500.pdf

Abstract:
Large Vision-Language Models (LVLMs) demonstrate significant progress in multimodal understanding and reasoning, yet object hallucination remains a critical challenge. While existing research focuses on mitigating language priors or high-level statistical biases, they often overlook the internal complexities of the visual encoding process. We identify that visual statistical bias, arising from the inherent Bag-of-Patches behavior of Vision Encoders under weak structural supervision, acts as a contributing factor of object hallucinations. Under this bias, models prioritize local texture features within individual patches over holistic geometric structures. This tendency may induce spurious visual confidence and result in hallucinations. To address this, we introduce a training-free algorithm called Structure-Disrupted Contrastive Decoding (SDCD), which performs contrastive calibration of the output distribution by introducing a shuffled structure-disrupted view. By penalizing tokens that maintain high confidence under this structure-less view, SDCD effectively suppresses the texture-driven bias. Experimental results demonstrate that SDCD significantly mitigates hallucinations across multiple benchmarks and enhances the overall multimodal capabilities of LVLMs.

Paperid: 6616, https://arxiv.org/pdf/2601.00659.pdf

Abstract:
Despite the rapid success of Large Vision-Language Models (LVLMs), a persistent challenge is their tendency to generate hallucinated content, undermining reliability in real-world use. Existing training-free methods address hallucinations but face two limitations: (i) they rely on narrow assumptions about hallucination sources, and (ii) their effectiveness declines toward the end of generation, where hallucinations are most likely to occur. A common strategy is to build hallucinated models by completely or partially removing visual tokens and contrasting them with the original model. Yet, this alone proves insufficient, since visual information still propagates into generated text. Building on this insight, we propose a novel hallucinated model that captures hallucination effects by selectively removing key text tokens. We further introduce Generalized Contrastive Decoding, which integrates multiple hallucinated models to represent diverse hallucination sources. Together, these ideas form CRoPS, a training-free hallucination mitigation framework that improves CHAIR scores by 20% and achieves consistent gains across six benchmarks and three LVLM families, outperforming state-of-the-art training-free methods.

Paperid: 6617, https://arxiv.org/pdf/2512.23243.pdf

Abstract:
Multimodal fusion of remote sensing images serves as a core technology for overcoming the limitations of single-source data and improving the accuracy of surface information extraction, which exhibits significant application value in fields such as environmental monitoring and urban planning. To address the deficiencies of existing methods, including the failure of fixed resolutions to balance efficiency and detail, as well as the lack of semantic hierarchy in single-scale alignment, this study proposes a Vision-language Model (VLM) framework integrated with two key innovations: the Dynamic Resolution Input Strategy (DRIS) and the Multi-scale Vision-language Alignment Mechanism (MS-VLAM).Specifically, the DRIS adopts a coarse-to-fine approach to adaptively allocate computational resources according to the complexity of image content, thereby preserving key fine-grained features while reducing redundant computational overhead. The MS-VLAM constructs a three-tier alignment mechanism covering object, local-region and global levels, which systematically captures cross-modal semantic consistency and alleviates issues of semantic misalignment and granularity imbalance.Experimental results on the RS-GPT4V dataset demonstrate that the proposed framework significantly improves the accuracy of semantic understanding and computational efficiency in tasks including image captioning and cross-modal retrieval. Compared with conventional methods, it achieves superior performance in evaluation metrics such as BLEU-4 and CIDEr for image captioning, as well as R@10 for cross-modal retrieval. This technical framework provides a novel approach for constructing efficient and robust multimodal remote sensing systems, laying a theoretical foundation and offering technical guidance for the engineering application of intelligent remote sensing interpretation.

Paperid: 6618, https://arxiv.org/pdf/2512.22969.pdf

Abstract:
Conventional object detectors rely on cross-entropy classification, which can be vulnerable to class imbalance and label noise. We propose CLIP-Joint-Detect, a simple and detector-agnostic framework that integrates CLIP-style contrastive vision-language supervision through end-to-end joint training. A lightweight parallel head projects region or grid features into the CLIP embedding space and aligns them with learnable class-specific text embeddings via InfoNCE contrastive loss and an auxiliary cross-entropy term, while all standard detection losses are optimized simultaneously. The approach applies seamlessly to both two-stage and one-stage architectures. We validate it on Pascal VOC 2007+2012 using Faster R-CNN and on the large-scale MS COCO 2017 benchmark using modern YOLO detectors (YOLOv11), achieving consistent and substantial improvements while preserving real-time inference speed. Extensive experiments and ablations demonstrate that joint optimization with learnable text embeddings markedly enhances closed-set detection performance across diverse architectures and datasets.

Paperid: 6619, https://arxiv.org/pdf/2512.20362.pdf

Abstract:
Recent work has shown that inference-time reasoning and reflection can improve text-to-image generation without retraining. However, existing approaches often rely on implicit, holistic critiques or unconstrained prompt rewrites, making their behavior difficult to interpret, control, or stop reliably. In contrast, large language models have benefited from explicit, structured forms of **thinking** based on verification, targeted correction, and early stopping. We introduce CRAFT (Continuous Reasoning and Agentic Feedback Tuning), a training-free, model-agnostic framework that brings this structured reasoning paradigm to multimodal image generation. CRAFT decomposes a prompt into dependency-structured visual questions, veries generated images using a vision-language model, and applies targeted prompt edits through an LLM agent only where constraints fail. The process iterates with an explicit stopping criterion once all constraints are satised, yielding an interpretable and controllable inference-time renement loop. Across multiple model families and challenging benchmarks, CRAFT consistently improves compositional accuracy, text rendering, and preference-based evaluations, with particularly strong gains for lightweight generators. Importantly, these improvements incur only a negligible inference-time overhead, allowing smaller or cheaper models to approach the quality of substantially more expensive systems. Our results suggest that explicitly structured, constraint-driven inference-time reasoning is a key ingredient for improving the reliability of multimodal generative models.

Paperid: 6620, https://arxiv.org/pdf/2512.20257.pdf

Abstract:
With the rise of easily accessible tools for generating and manipulating multimedia content, realistic synthetic alterations to digital media have become a widespread threat, often involving manipulations across multiple modalities simultaneously. Recently, such techniques have been increasingly employed to distort narratives of important events and to spread misinformation on social media, prompting the development of misinformation detectors. In the context of misinformation conveyed through image-text pairs, several detection methods have been proposed. However, these approaches typically rely on computationally intensive architectures or require large amounts of annotated data. In this work we introduce LADLE-MM: Limited Annotation based Detector with Learned Ensembles for Multimodal Misinformation, a model-soup initialized multimodal misinformation detector designed to operate under a limited annotation setup and constrained training resources. LADLE-MM is composed of two unimodal branches and a third multimodal one that enhances image and text representations with additional multimodal embeddings extracted from BLIP, serving as fixed reference space. Despite using 60.3% fewer trainable parameters than previous state-of-the-art models, LADLE-MM achieves competitive performance on both binary and multi-label classification tasks on the DGM4 benchmark, outperforming existing methods when trained without grounding annotations. Moreover, when evaluated on the VERITE dataset, LADLE-MM outperforms current state-of-the-art approaches that utilize more complex architectures involving Large Vision-Language-Models, demonstrating the effective generalization ability in an open-set setting and strong robustness to unimodal bias.

Paperid: 6621, https://arxiv.org/pdf/2512.20188.pdf

Abstract:
Most Vision-Language-Action (VLA) systems integrate a Vision-Language Model (VLM) for semantic reasoning with an action expert generating continuous action signals, yet both typically run at a single unified frequency. As a result, policy performance is constrained by the low inference speed of large VLMs. This mandatory synchronous execution severely limits control stability and real-time performance in whole-body robotic manipulation, which involves more joints, larger motion spaces, and dynamically changing views. We introduce a truly asynchronous Fast-Slow VLA framework (DuoCore-FS), organizing the system into a fast pathway for high-frequency action generation and a slow pathway for rich VLM reasoning. The system is characterized by two key features. First, a latent representation buffer bridges the slow and fast systems. It stores instruction semantics and action-reasoning representation aligned with the scene-instruction context, providing high-level guidance to the fast pathway. Second, a whole-body action tokenizer provides a compact, unified representation of whole-body actions. Importantly, the VLM and action expert are still jointly trained end-to-end, preserving unified policy learning while enabling asynchronous execution. DuoCore-FS supports a 3B-parameter VLM while achieving 30 Hz whole-body action-chunk generation, approximately three times as fast as prior VLA models with comparable model sizes. Real-world whole-body manipulation experiments demonstrate improved task success rates and significantly enhanced responsiveness compared to synchronous Fast-Slow VLA baselines. The implementation of DuoCore-FS, including training, inference, and deployment, is provided to commercial users by Astribot as part of the Astribot robotic platform.

Paperid: 6622, https://arxiv.org/pdf/2512.20014.pdf

Abstract:
While Vision-Language-Action (VLA) models generalize well to generic instructions, they struggle with personalized commands such as "bring my cup", where the robot must act on one specific instance among visually similar objects. We study this setting of manipulating personal objects, in which a VLA must identify and control a user-specific object unseen during training using only a few reference images. To address this challenge, we propose Visual Attentive Prompting (VAP), a simple-yet-effective training-free perceptual adapter that equips frozen VLAs with top-down selective attention. VAP treats the reference images as a non-parametric visual memory, grounds the personal object in the scene through open-vocabulary detection and embedding-based matching, and then injects this grounding as a visual prompt by highlighting the object and rewriting the instruction. We construct two simulation benchmarks, Personalized-SIMPLER and Personalized-VLABench, and a real-world tabletop benchmark to evaluate personalized manipulation across multiple robots and tasks. Experiments show that VAP consistently outperforms generic policies and token-learning baselines in both success rate and correct-object manipulation, helping to bridge the gap between semantic understanding and instance-level control.

Paperid: 6623, https://arxiv.org/pdf/2512.19943.pdf

Abstract:
While instruction-based image editing is emerging, extending it to 360$^\circ$ panoramas introduces additional challenges. Existing methods often produce implausible results in both equirectangular projections (ERP) and perspective views. To address these limitations, we propose SE360, a novel framework for multi-condition guided object editing in 360$^\circ$ panoramas. At its core is a novel coarse-to-fine autonomous data generation pipeline without manual intervention. This pipeline leverages a Vision-Language Model (VLM) and adaptive projection adjustment for hierarchical analysis, ensuring the holistic segmentation of objects and their physical context. The resulting data pairs are both semantically meaningful and geometrically consistent, even when sourced from unlabeled panoramas. Furthermore, we introduce a cost-effective, two-stage data refinement strategy to improve data realism and mitigate model overfitting to erase artifacts. Based on the constructed dataset, we train a Transformer-based diffusion model to allow flexible object editing guided by text, mask, or reference image in 360$^\circ$ panoramas. Our experiments demonstrate that our method outperforms existing methods in both visual quality and semantic accuracy.

Paperid: 6624, https://arxiv.org/pdf/2512.19024.pdf

Abstract:
Vision-Language Navigation (VLN) enables agents to navigate in complex environments by following natural language instructions grounded in visual observations. Although most existing work has focused on ground-based robots or outdoor Unmanned Aerial Vehicles (UAVs), indoor UAV-based VLN remains underexplored, despite its relevance to real-world applications such as inspection, delivery, and search-and-rescue in confined spaces. To bridge this gap, we introduce \textbf{IndoorUAV}, a novel benchmark and method specifically tailored for VLN with indoor UAVs. We begin by curating over 1,000 diverse and structurally rich 3D indoor scenes from the Habitat simulator. Within these environments, we simulate realistic UAV flight dynamics to collect diverse 3D navigation trajectories manually, further enriched through data augmentation techniques. Furthermore, we design an automated annotation pipeline to generate natural language instructions of varying granularity for each trajectory. This process yields over 16,000 high-quality trajectories, comprising the \textbf{IndoorUAV-VLN} subset, which focuses on long-horizon VLN. To support short-horizon planning, we segment long trajectories into sub-trajectories by selecting semantically salient keyframes and regenerating concise instructions, forming the \textbf{IndoorUAV-VLA} subset. Finally, we introduce \textbf{IndoorUAV-Agent}, a novel navigation model designed for our benchmark, leveraging task decomposition and multimodal reasoning. We hope IndoorUAV serves as a valuable resource to advance research on vision-language embodied AI in the indoor aerial navigation domain.

Paperid: 6625, https://arxiv.org/pdf/2512.16531.pdf

Abstract:
Deploying local large language models and vision-language models on edge devices requires balancing accuracy with constrained computational and energy budgets. Although graphics processors dominate modern artificial-intelligence deployment, most consumer hardware--including laptops, desktops, industrial controllers, and embedded systems--relies on central processing units. Despite this, the computational laws governing central-processing-unit-only inference for local language and vision-language workloads remain largely unexplored. We systematically benchmark large language and vision-language models on two representative central-processing-unit tiers widely used for local inference: a MacBook Pro M2, reflecting mainstream laptop-class deployment, and a Raspberry Pi 5, representing constrained, low-power embedded settings. Using a unified methodology based on continuous sampling of processor and memory usage together with area-under-curve integration, we characterize how computational load scales with input text length for language models and with image resolution for vision-language models. We uncover two empirical scaling laws: (1) computational cost for language-model inference scales approximately linearly with token length; and (2) vision-language models exhibit a preprocessing-driven "resolution knee", where compute remains constant above an internal resolution clamp and decreases sharply below it. Beyond these laws, we show that quantum-inspired compression reduces processor and memory usage by up to 71.9% and energy consumption by up to 62%, while preserving or improving semantic accuracy. These results provide a systematic quantification of multimodal central-processing-unit-only scaling for local language and vision-language workloads, and they identify model compression and input-resolution preprocessing as effective, low-cost levers for sustainable edge inference.

Paperid: 6626, https://arxiv.org/pdf/2512.14926.pdf

Abstract:
Focusing on low-resource languages is an essential step toward democratizing generative AI. In this work, we contribute to reducing the multimodal NLP resource gap for Romanian. We translate the widely known Flickr30k dataset into Romanian and further extend it for visual question answering by leveraging open-source LLMs. We demonstrate the usefulness of our datasets by fine-tuning open-source VLMs on Romanian visual question answering. We select VLMs from three widely used model families: LLaMA 3.2, LLaVA 1.6, and Qwen2. For fine-tuning, we employ the parameter-efficient LoRA method. Our models show improved Romanian capabilities in visual QA, as well as on tasks they were not trained on, such as Romanian image description generation. The seven-billion-parameter Qwen2-VL-RoVQA obtains top scores on both tasks, with improvements of +6.05% and +2.61% in BERTScore F1 over its original version. Finally, the models show substantial reductions in grammatical errors compared to their original forms, indicating improvements not only in language understanding but also in Romanian fluency.

Paperid: 6627, https://arxiv.org/pdf/2512.14031.pdf

Abstract:
This study evaluates two leading approaches for teaching construction robots new skills to understand their applicability for construction automation: a Vision-Language-Action (VLA) model and Reinforcement Learning (RL) methods. The goal is to understand both task performance and the practical effort needed to deploy each approach on real jobs. The authors developed two teleoperation interfaces to control the robots and collect the demonstrations needed, both of which proved effective for training robots for long-horizon and dexterous tasks. In addition, the authors conduct a three-stage evaluation. First, the authors compare a Multi-Layer Perceptron (MLP) policy with a Deep Q-network (DQN) imitation model to identify the stronger RL baseline, focusing on model performance, generalization, and a pick-up experiment. Second, three different VLA models are trained in two different scenarios and compared with each other. Third, the authors benchmark the selected RL baseline against the VLA model using computational and sample-efficiency measures and then a robot experiment on a multi-stage panel installation task that includes transport and installation. The VLA model demonstrates strong generalization and few-shot capability, achieving 60% and 100% success in the pickup phase. In comparison, DQN can be made robust but needs additional noise during tuning, which increases the workload. Overall, the findings indicate that VLA offers practical advantages for changing tasks by reducing programming effort and enabling useful performance with minimal data, while DQN provides a viable baseline when sufficient tuning effort is acceptable.

Paperid: 6628, https://arxiv.org/pdf/2512.12571.pdf

Abstract:
To extend the application of vision-language models (VLMs) from web images to sensor-mediated physical environments, we propose Multi-View Physical-prompt for Test-Time Adaptation (MVP), a forward-only framework that moves test-time adaptation (TTA) from tokens to photons by treating the camera exposure triangle--ISO, shutter speed, and aperture--as physical prompts. At inference, MVP acquires a library of physical views per scene, selects the top-k sensor settings using a source-affinity score, evaluates each retained view under lightweight digital augmentations, filters the lowest-entropy subset of augmented views, and aggregates predictions with Zero-temperature softmax (i.e., hard voting). This selection-then-vote design is simple, calibration-friendly, and requires no gradients or model modifications. On ImageNet-ES and ImageNet-ES-Diverse, MVP consistently outperforms digital-only TTA on single Auto-Exposure captures, by up to 25.6 percentage points (pp), and delivers up to 3.4 pp additional gains over pipelines that combine conventional sensor control with TTA. MVP remains effective under reduced parameter candidate sets that lower capture latency, demonstrating practicality. These results support the main claim that, beyond post-capture prompting, measurement-time control--selecting and combining real physical views--substantially improves robustness for VLMs.

Paperid: 6629, https://arxiv.org/pdf/2512.11567.pdf

Abstract:
Social media serves as a critical medium in modern politics because it both reflects politicians' ideologies and facilitates communication with younger generations. We present MultiParTweet, a multilingual tweet corpus from X that connects politicians' social media discourse with German political corpus GerParCor, thereby enabling comparative analyses between online communication and parliamentary debates. MultiParTweet contains 39 546 tweets, including 19 056 media items. Furthermore, we enriched the annotation with nine text-based models and one vision-language model (VLM) to annotate MultiParTweet with emotion, sentiment, and topic annotations. Moreover, the automated annotations are evaluated against a manually annotated subset. MultiParTweet can be reconstructed using our tool, TTLABTweetCrawler, which provides a framework for collecting data from X. To demonstrate a methodological demonstration, we examine whether the models can predict each other using the outputs of the remaining models. In summary, we provide MultiParTweet, a resource integrating automatic text and media-based annotations validated with human annotations, and TTLABTweetCrawler, a general-purpose X data collection tool. Our analysis shows that the models are mutually predictable. In addition, VLM-based annotation were preferred by human annotators, suggesting that multimodal representations align more with human interpretation.

Paperid: 6630, https://arxiv.org/pdf/2512.11350.pdf

Abstract:
Road traffic accidents represent a leading cause of mortality globally, with incidence rates rising due to increasing population, urbanization, and motorization. Rising accident rates raise concerns about traffic surveillance effectiveness. Traditional computer vision methods for accident detection struggle with limited spatiotemporal understanding and poor cross-domain generalization. Recent advances in transformer architectures excel at modeling global spatial-temporal dependencies and parallel computation. However, applying these models to automated traffic accident detection is limited by small, non-diverse datasets, hindering the development of robust, generalizable systems. To address this gap, we curated a comprehensive and balanced dataset that captures a wide spectrum of traffic environments, accident types, and contextual variations. Utilizing the curated dataset, we propose an accident detection model based on a transformer architecture using pre-extracted spatial video features. The architecture employs convolutional layers to extract local correlations across diverse patterns within a frame, while leveraging transformers to capture sequential-temporal dependencies among the retrieved features. Moreover, most existing studies neglect the integration of motion cues, which are essential for understanding dynamic scenes, especially during accidents. These approaches typically rely on static features or coarse temporal information. In this study, multiple methods for incorporating motion cues were evaluated to identify the most effective strategy. Among the tested input approaches, concatenating RGB features with optical flow achieved the highest accuracy at 88.3%. The results were further compared with vision language models (VLM) such as GPT, Gemini, and LLaVA-NeXT-Video to assess the effectiveness of the proposed method.

Paperid: 6631, https://arxiv.org/pdf/2512.09172.pdf

Abstract:
We tackle continual adaptation of vision-language models to new attributes, objects, and their compositions in Compositional Zero-Shot Learning (CZSL), while preventing forgetting of prior knowledge. Unlike classical continual learning where classes are disjoint, CCZSL is more complex as attributes and objects may reoccur across sessions while compositions remain unique. Built on a frozen VLM backbone, we propose the first Prompt-based Continual Compositional Zero-Shot Learning (PromptCCZSL) framework that retains prior knowledge through recency-weighted multi-teacher distillation. It employs session-aware compositional prompts to fuse multimodal features for new compositions, while attribute and object prompts are learned through session-agnostic fusion to maintain global semantic consistency, which is further stabilized by a Cosine Anchor Loss (CAL) to preserve prior knowledge. To enhance adaptation in the current session, an Orthogonal Projection Loss (OPL) ensures that new attribute and object embeddings remain distinct from previous ones, preventing overlap, while an Intra-Session Diversity Loss (IDL) promotes variation among current-session embeddings for richer, more discriminative representations. We also introduce a comprehensive protocol that jointly measures catastrophic forgetting and compositional generalization. Extensive experiments on UT-Zappos and C-GQA benchmarks demonstrate that PromptCCZSL achieves substantial improvements over prior VLM-based and non-VLM baselines, setting a new benchmark for CCZSL in closed-world settings.

Paperid: 6632, https://arxiv.org/pdf/2512.07128.pdf

Abstract:
Vision-language models like CLIP show impressive ability to align images and text, but their training on short, concise captions makes them struggle with lengthy, detailed descriptions. Recent advances mitigate this challenge by leveraging region-proposal information to map visual regions with corresponding sentences from lengthy captions, yet incurring notable deployment costs. We introduce MulCLIP, a novel end-to-end multi-level alignment framework that bridges natural long-text structures with image components. MulCLIP first preserves global contrastive alignment between images and both summary and long captions, while extending positional embeddings for longer text sequences. To further enhance fine-grained understanding, we propose two novel strategies: (1) a token reconstruction alignment over locally calibrated features to strengthen semantic connections between words and image patches, and (2) a subcaption-aggregated patch alignment that automatically extracts and aggregates context-rich patches for each subcaption. Experimental results across diverse benchmarks demonstrate our method consistently improves downstream performance, while ablation studies confirm its multi-scale alignment is the key factor driving better fine-grained capability than region-proposal-assisted approaches, making it particularly suitable for diverse real-world applications.

Paperid: 6633, https://arxiv.org/pdf/2512.06951.pdf

Abstract:
We present a vision-action policy that won 1st place in the 2025 BEHAVIOR Challenge - a large-scale benchmark featuring 50 diverse long-horizon household tasks in photo-realistic simulation, requiring bimanual manipulation, navigation, and context-aware decision making. Building on the Pi0.5 architecture, we introduce several innovations. Our primary contribution is correlated noise for flow matching, which improves training efficiency and enables correlation-aware inpainting for smooth action sequences. We also apply learnable mixed-layer attention and System 2 stage tracking for ambiguity resolution. Training employs multi-sample flow matching to reduce variance, while inference uses action compression and challenge-specific correction rules. Our approach achieves 26% q-score across all 50 tasks on both public and private leaderboards.

Paperid: 6634, https://arxiv.org/pdf/2512.06096.pdf

Abstract:
The rapid development of Vision-Language models (VLMs) and Multimodal Language Models (MLLMs) in autonomous driving research has significantly reshaped the landscape by enabling richer scene understanding, context-aware reasoning, and more interpretable decision-making. However, a lot of existing work often relies on either single-view encoders that fail to exploit the spatial structure of multi-camera systems or operate on aggregated multi-view features, which lack a unified spatial representation, making it more challenging to reason about ego-centric directions, object relations, and the wider context. We thus present BeLLA, an end-to-end architecture that connects unified 360° BEV representations with a large language model for question answering in autonomous driving. We primarily evaluate our work using two benchmarks - NuScenes-QA and DriveLM, where BeLLA consistently outperforms existing approaches on questions that require greater spatial reasoning, such as those involving relative object positioning and behavioral understanding of nearby objects, achieving up to +9.3% absolute improvement in certain tasks. In other categories, BeLLA performs competitively, demonstrating the capability of handling a diverse range of questions.

Paperid: 6635, https://arxiv.org/pdf/2512.03992.pdf

Abstract:
Vision-Language Models (VLMs) deployed in safety-critical applications such as autonomous driving must handle continuous visual streams under imperfect conditions. However, existing benchmarks focus on static, high-quality images and ignore temporal degradation and error propagation, which are critical failure modes where transient visual corruption induces hallucinations that persist across subsequent frames. We introduce DIQ-H, the first benchmark for evaluating VLM robustness under dynamic visual degradation in temporal sequences. DIQ-H applies physics-based corruptions including motion blur, sensor noise, and compression artifacts, and measures hallucination persistence, error recovery, and temporal consistency through multi-turn question-answering tasks. To enable scalable annotation, we propose Uncertainty-Guided Iterative Refinement (UIR), which generates reliable pseudo-ground-truth using lightweight VLMs with uncertainty filtering, achieving a 15.3 percent accuracy improvement. Experiments on 16 state-of-the-art VLMs reveal substantial robustness gaps: even advanced models such as GPT-4o achieve only a 78.5 percent recovery rate, while open-source models struggle with temporal consistency at less than 60 percent. DIQ-H provides a comprehensive platform for evaluating VLM reliability in real-world deployments.

Paperid: 6636, https://arxiv.org/pdf/2512.03479.pdf

Abstract:
Understanding procedural activities is crucial for developing future assistive AI that can reason about complex real-world tasks. Existing action-centric methods struggle with the flexibility of real procedures, where step order varies depending on object states. In this work, we propose to shift the focus to an object-centric paradigm by regarding actions as mechanisms that drive state transitions. To advance this direction, we introduce Object-IVQA, a long-form instructional video benchmark with 107 videos and 514 open-ended question-answer pairs annotated with temporally grounded evidence. The benchmark evaluates four dimensions of object-centric reasoning, including state evolution, precondition verification, counterfactual reasoning and mistake recognition. We further propose an agent framework that orchestrates object-centric planning, perception, analysis and generation tools, enabling explicit evidence retrieval and multi-hop reasoning across disjoint segments. Experiments show that existing large vision-language models struggle in object-level recognition and reasoning, whereas our framework achieves substantially improvement.

Paperid: 6637, https://arxiv.org/pdf/2512.02498.pdf

Abstract:
Document Layout Parsing serves as a critical gateway for Artificial Intelligence (AI) to access and interpret the world's vast stores of structured knowledge. This process,which encompasses layout detection, text recognition, and relational understanding, is particularly crucial for empowering next-generation Vision-Language Models. Current methods, however, rely on fragmented, multi-stage pipelines that suffer from error propagation and fail to leverage the synergies of joint training. In this paper, we introduce dots.ocr, a single Vision-Language Model that, for the first time, demonstrates the advantages of jointly learning three core tasks within a unified, end-to-end framework. This is made possible by a highly scalable data engine that synthesizes a vast multilingual corpus, empowering the model to deliver robust performance across a wide array of tasks, encompassing diverse languages, layouts, and domains. The efficacy of our unified paradigm is validated by state-of-the-art performance on the comprehensive OmniDocBench. Furthermore, to catalyze research in global document intelligence, we introduce XDocParse, a challenging new benchmark spanning 126 languages. On this testbed, dots.ocr establishes a powerful new baseline, outperforming the next-best competitor by a remarkable +7.4 point margin and proving its unparalleled multilingual capabilities.

Paperid: 6638, https://arxiv.org/pdf/2512.02170.pdf

Abstract:
Flowcharts are common tools for communicating processes but are often shared as static images that cannot be easily edited or reused. We present Flowchart2Mermaid, a lightweight web system that converts flowchart images into editable Mermaid.js code which is a markup language for visual workflows, using a detailed system prompt and vision-language models. The interface supports mixed-initiative refinement through inline text editing, drag-and-drop node insertion, and natural-language commands interpreted by an integrated AI assistant. Unlike prior image-to-diagram tools, our approach produces a structured, version-controllable textual representation that remains synchronized with the rendered diagram. We further introduce evaluation metrics to assess structural accuracy, flow correctness, syntax validity, and completeness across multiple models.

Paperid: 6639, https://arxiv.org/pdf/2512.00706.pdf

Abstract:
Recently, large vision-language models (LVLMs) have risen to be a promising approach for multimodal tasks. However, principled hallucination mitigation remains a critical challenge.In this work, we first analyze the data generation process in LVLM hallucination mitigation and affirm that on-policy data significantly outperforms off-policy data, which thus calls for efficient and reliable preference annotation of on-policy data. We then point out that, existing annotation methods introduce additional hallucination in training samples, which may enhance the model's hallucination patterns, to address this problem, we propose training a hallucination classifier giving binary annotations, which guarantee clean chosen samples for the subsequent alignment. To further harness of the power of on-policy data, we design a robust iterative direct preference optimization (DPO) algorithm adopting a dynamic sample reweighting scheme. We conduct comprehensive experiments on three benchmarks with comparison to 8 state-of-the-art baselines. In particular, our approach reduces the hallucination rate of LLaVA-1.5-7B on MMHalBench by 50.8% and the average hallucination rate on Object HalBench by 79.5%; more significantly, our method fully taps into the potential of open-source models, enabling LLaVA-1.5-13B to even surpass the performance of GPT-4V.

Paperid: 6640, https://arxiv.org/pdf/2512.00597.pdf

Abstract:
Foundation models trained via vision-language pretraining have demonstrated strong zero-shot capabilities across diverse image domains, yet their application to volumetric medical imaging remains limited. We introduce MedCT-VLM: Medical CT Vision-Language Model, a parameter-efficient vision-language framework designed to adapt large-scale CT foundation models for downstream clinical tasks. MedCT-VLM uses a parameter-efficient approach to adapt CT-CLIP, a contrastive vision-language model trained on 25,692 chest CT volumes, for multi-label pathology classification using Low-Rank Adaptation (LoRA). Rather than fine-tuning the model's 440 M parameters directly, we insert low-rank decomposition matrices into attention layers of both vision and text encoders, training only 1.67M parameters (0.38\% of total). We evaluate on zero-shot classification across 18 thoracic pathologies, where the model must align CT embeddings with unseen text prompts at inference without task-specific training. LoRA fine-tuning improves mean AUROC from 61.3\% to 68.9\% (+7.6 pp), accuracy from 67.2\% to 73.6\% (+6.4 pp), and macro-F1 from 32.1\% to 36.9\% (+4.8 pp). These results demonstrate that parameter-efficient methods can effectively transfer large-scale pretraining to downstream medical imaging tasks, particularly for zero-shot scenarios where labeled data is scarce.

Paperid: 6641, https://arxiv.org/pdf/2511.23070.pdf

Abstract:
Missing modalities consistently lead to significant performance degradation in multimodal models. Existing approaches either synthesize missing modalities at high computational cost or apply prompt-based fine-tuning that relies only on adjacent-layer features and overlooks long-distance contextual information, which may offer additional tolerance to errors when one or more modalities are missing. To address this, we introduce REplay Prompting (REP): (1) construct modality-wise feature buffers via a residual bypass to cache early-layer representations and replay them in deeper layers, mitigating information loss as network depth increases; (2) employ a private-shared feature decoupling strategy, where private buffers preserve modality-specific signals and shared buffers encode cross-modal semantics; and (3) design a task-aware dynamic initialization mechanism to configure these buffers differently, improving stability and generalization under diverse missing-modality conditions. Experiments on vision-language, vision-language-audio, and temporal multimodal benchmarks demonstrate that REP consistently outperforms prior methods under both single- and multi-modality missing scenarios, while introducing only negligible parameter overhead. These results establish REP as a lightweight and effective paradigm for robust multimodal learning in challenging missing-modality environments.

Paperid: 6642, https://arxiv.org/pdf/2511.22739.pdf

Abstract:
Domain generalization is critical in computational pathology (CPath) due to inherent domain shifts caused by variations in staining protocols, scanner devices, and imaging settings across clinical centers. Vision-language models (VLMs), such as PLIP-a pathology-tuned CLIP-trained on image-text pairs across diverse domains, serve as strong knowledge distillation sources. However, their zero-shot performance with predefined prompts remains limited due to sensitivity to prompt variations. Moreover, unlike natural images, histopathology centers lack semantic descriptors (e.g., 'sketch'), making it difficult to define domain-specific prompts for clinical centers. This requires a data-driven approach for learning domain-specific and ultimately class-generic continuous prompts. We propose Domain Invariant Prompt Tuning (DIPT) for knowledge distillation process, a novel step that learns multiple input tokens for each domain. These tokens are trained separately for each domain and are averaged across domains, leading to domain-invariant prompts. Our student model then distills knowledge from PLIP's text encoder by leveraging the prompts learned by DIPT. This leads to alignment of visual features with domain-invariant embeddings, enhancing generalization by training on multiple domains. Our method adds a significant improvement in average F1-score to existing state-of-the-art (SOTA) knowledge distillation approaches in domain generalization with histopathology datasets. This work helps the way of deploying robust CPath models in real-world clinical problems with heterogeneous data sources.

Paperid: 6643, https://arxiv.org/pdf/2511.22448.pdf

Abstract:
Business Process Model and Notation (BPMN) is a widely adopted standard for representing complex business workflows. While BPMN diagrams are often exchanged as visual images, existing methods primarily rely on XML representations for computational analysis. In this work, we present a pipeline that leverages Vision-Language Models (VLMs) to extract structured JSON representations of BPMN diagrams directly from images, without requiring source model files or textual annotations. We also incorporate optical character recognition (OCR) for textual enrichment and evaluate the generated element lists against ground truth data derived from the source XML files. Our approach enables robust component extraction in scenarios where original source files are unavailable. We benchmark multiple VLMs and observe performance improvements in several models when OCR is used for text enrichment. In addition, we conducted extensive statistical analyses of OCR-based enrichment methods and prompt ablation studies, providing a clearer understanding of their impact on model performance.

Paperid: 6644, https://arxiv.org/pdf/2511.21064.pdf

Abstract:
Open-Vocabulary Object Detection (OVOD) aims to enable detectors to generalize across categories by leveraging semantic information. Although existing methods are pretrained on large vision-language datasets, their inference is still limited to fixed category names, creating a gap between multimodal training and unimodal inference. Previous work has shown that improving textual representation can significantly enhance OVOD performance, indicating that the textual space is still underexplored. To this end, we propose OVOD-Agent, which transforms passive category matching into proactive visual reasoning and self-evolving detection. Inspired by the Chain-of-Thought (CoT) paradigm, OVOD-Agent extends the textual optimization process into an interpretable Visual-CoT with explicit actions. OVOD's lightweight nature makes LLM-based management unsuitable; instead, we model visual context transitions as a Weakly Markovian Decision Process (w-MDP) over eight state spaces, which naturally represents the agent's state, memory, and interaction dynamics. A Bandit module generates exploration signals under limited supervision, helping the agent focus on uncertain regions and adapt its detection policy. We further integrate Markov transition matrices with Bandit trajectories for self-supervised Reward Model (RM) optimization, forming a closed loop from Bandit exploration to RM learning. Experiments on COCO and LVIS show that OVOD-Agent provides consistent improvements across OVOD backbones, particularly on rare categories, confirming the effectiveness of the proposed framework.

Paperid: 6645, https://arxiv.org/pdf/2511.20085.pdf

Abstract:
The current remote sensing image analysis task is increasingly evolving from traditional object recognition to complex intelligence reasoning, which places higher requirements on the model's reasoning ability and the flexibility of tool invocation. To this end, we propose a new multimodal agent framework, Vision-Interleaved Chain-of-Thought Framework (VICoT), which implements explicit multi-round reasoning by dynamically incorporating visual tools into the chain of thought. Through a stack-based reasoning structure and a modular MCP-compatible tool suite, VICoT enables LLMs to efficiently perform multi-round, interleaved vision-language reasoning tasks with strong generalization and flexibility.We also propose the Reasoning Stack distillation method to migrate complex Agent behaviors to small, lightweight models, which ensures the reasoning capability while significantly reducing complexity. Experiments on multiple remote sensing benchmarks demonstrate that VICoT significantly outperforms existing SOTA frameworks in reasoning transparency, execution efficiency, and generation quality.

Paperid: 6646, https://arxiv.org/pdf/2511.18960.pdf

Abstract:
Vision-Language-Action (VLA) models have demonstrated remarkable capabilities in embodied AI tasks. However, existing VLA models, often built upon Vision-Language Models (VLMs), typically process dense visual inputs independently at each timestep. This approach implicitly models the task as a Markov Decision Process (MDP). However, this history-agnostic design is suboptimal for effective visual token processing in dynamic sequential decision-making, as it fails to leverage the context of history. To address this limitation, we reformulate the problem from a Partially Observable Markov Decision Process (POMDP) perspective and propose a novel framework named AVA-VLA. Inspired by the POMDP that the action generation should be conditioned on the belief state. AVA-VLA introduces Active Visual Attention (AVA) to dynamically modulate visual processing. It achieves this by leveraging the recurrent state, which is a neural approximation of the agent's belief state derived from the previous decision step. Specifically, the AVA module uses the recurrent state to compute the soft weights to actively process task-relevant visual tokens based on its historical context. Comprehensive evaluations demonstrate that AVA-VLA achieves state-of-the-art performance across popular robotic benchmarks, including LIBERO and CALVIN. Furthermore, real-world deployments on a dual-arm robot platform validate the framework's practical applicability and robust sim-to-real transferability.

Paperid: 6647, https://arxiv.org/pdf/2511.18950.pdf

Abstract:
Vision-Language-Action (VLA) models have emerged as a powerful paradigm in Embodied AI. However, the significant computational overhead of processing redundant visual tokens remains a critical bottleneck for real-time robotic deployment. While standard token pruning techniques can alleviate this, these task-agnostic methods struggle to preserve task-critical visual information. To address this challenge, simultaneously preserving both the holistic context and fine-grained details for precise action, we propose Compressor-VLA, a novel hybrid instruction-conditioned token compression framework designed for efficient, task-oriented compression of visual information in VLA models. The proposed Compressor-VLA framework consists of two token compression modules: a Semantic Task Compressor (STC) that distills holistic, task-relevant context, and a Spatial Refinement Compressor (SRC) that preserves fine-grained spatial details. This compression is dynamically modulated by the natural language instruction, allowing for the adaptive condensation of task-relevant visual information. Experimentally, extensive evaluations demonstrate that Compressor-VLA achieves a competitive success rate on the LIBERO benchmark while reducing FLOPs by 59% and the visual token count by over 3x compared to its baseline. The real-robot deployments on a dual-arm robot platform validate the model's sim-to-real transferability and practical applicability. Moreover, qualitative analyses reveal that our instruction guidance dynamically steers the model's perceptual focus toward task-relevant objects, thereby validating the effectiveness of our approach.

Paperid: 6648, https://arxiv.org/pdf/2511.18692.pdf

Abstract:
Edge deployment of large Vision-Language Models (VLMs) increasingly relies on flash-based weight offloading, where activation sparsification is used to reduce I/O overhead. However, conventional sparsification remains model-centric, selecting neurons solely by activation magnitude and neglecting how access patterns influence flash performance. We present Neuron Chunking, an I/O-efficient sparsification strategy that operates on chunks (i.e., groups of contiguous neurons in memory) and couples neuron importance with storage access cost. The method models I/O latency through a lightweight abstraction of access contiguity and selects chunks with high utility, defined as neuron importance normalized by estimated latency. By aligning sparsification decisions with the underlying storage behavior, Neuron Chunking improves I/O efficiency by up to 4.65x and 5.76x on Jetson Orin Nano and Jetson AGX Orin, respectively.

Paperid: 6649, https://arxiv.org/pdf/2511.18306.pdf

Abstract:
Building codes contain critical information for ensuring safety, regulatory compliance, and informed decision-making in construction and engineering. Automated question answering systems over such codes enable quick and accurate access to specific regulatory clauses, improving efficiency and reducing errors. Retrieval-Augmented Generation (RAG) systems are essential for this task as they combine the precision of information retrieval with the generative capabilities of language models. However, tabular data are challenging to extract as they often involve complex layouts, merged cells, multi-row headers, and embedded semantic relationships that are not easily captured by traditional natural language processing techniques and Vision Language Models (VLMs). This paper explores and compares two methods for extracting information from tabular data in building codes using several pre-trained VLMs. First, a direct input method is used, where the image of the page is input directly into the VLMs, which are then tasked with answering questions based on the image. Second, an indirect input method is introduced, which involves converting an image of a page containing tables into the LaTeX code and then answering inquires based on the LaTeX-based input. The experiments find that the direct input method generally resulted in higher accuracy than the indirect input method. To further improve the performance, we fine-tuned each VLM using Low Rank Adaptation (LoRA) on a domain-specific tabular dataset. The fine-tuned models exhibited substantial improvements, with Qwen2.5-VL-3B-Instruct achieving relative accuracy gains exceeding 100%. Our results highlight the potential of parameter-efficient fine-tuning methods to adapt powerful VLMs for understanding complex structured data in specialized fields, such as building code interpretation and regulatory compliance.

Paperid: 6650, https://arxiv.org/pdf/2511.18305.pdf

Abstract:
Large Vision Language Models (LVLMs) possess extensive text knowledge but struggles to utilize this knowledge for fine-grained image recognition, often failing to differentiate between visually similar categories. Existing fine-tuning methods using Reinforcement Learning (RL) with exact-match reward signals are often brittle, encourage memorization of training categories, and fail to elicit differential reasoning needed for generalization to unseen classes. To address this, we propose $\textbf{DiVE-k}$, $\textbf{Di}$fferential $\textbf{V}$isual r$\textbf{E}$asoning using top-$\textbf{k}$ generations, framework that leverages model's own top-k predictions as a training signal. For each training image, DiVE-k creates a multiple-choice question from the model's top-k outputs and uses RL to train the model to select the correct answer. This approach requires the model to perform fine-grained differential reasoning among plausible options and provides a simple, verifiable reward signal that mitigates memorization and improves generalization. Experiments on five standard fine-grained datasets show that our method significantly outperforms existing approaches. In the standard base-to-novel generalization setting, DiVE-k surpasses the QWEN2.5-VL-7B and ViRFT by 10.04% and 6.16% on the Harmonic Mean metric, respectively. Further experiments show similar gains in mixed-domain and few-shot scenarios.

Paperid: 6651, https://arxiv.org/pdf/2511.18123.pdf

Abstract:
Vision-Language Models (VLMs) have become indispensable for multimodal reasoning, yet their representations often encode and amplify demographic biases, resulting in biased associations and misaligned predictions in downstream tasks. Such behavior undermines fairness and distorts the intended alignment between vision and language. Recent post-hoc approaches attempt to mitigate bias by replacing the most attribute-correlated embedding coordinates with neutral values. However, our systematic analysis reveals three critical failures of this coordinate-wise approach: feature entanglement, poor cross-dataset generalization, and incomplete bias removal. We find that bias is not localized to a few coordinates but is instead distributed across a few linear subspaces. To address these limitations, we propose $\textbf{S}$ubspace $\textbf{P}$rojection $\textbf{D}$ebiasing ($\textbf{SPD}$), a geometrically principled framework that identifies and removes the entire subspace of linearly decodable bias while reinserting a neutral mean component to preserve semantic fidelity. Extensive experiments across zero-shot classification, text-to-image retrieval, and image generation validate the effectiveness of SPD: our method achieves more robust debiasing with an average improvement of $18.5\%$ across four fairness metrics, while maintaining minimal loss in task performance compared to the best debiasing baseline.

Paperid: 6652, https://arxiv.org/pdf/2511.17727.pdf

Abstract:
Vision-language models (VLMs) have demonstrated remarkable performance across a wide range of computer-vision tasks, sparking interest in their potential for digital health applications. Here, we apply VLMs to two fundamental challenges in data-driven stroke rehabilitation: automatic quantification of rehabilitation dose and impairment from videos. We formulate these problems as motion-identification tasks, which can be addressed using VLMs. We evaluate our proposed framework on a cohort of 29 healthy controls and 51 stroke survivors. Our results show that current VLMs lack the fine-grained motion understanding required for precise quantification: dose estimates are comparable to a baseline that excludes visual information, and impairment scores cannot be reliably predicted. Nevertheless, several findings suggest future promise. With optimized prompting and post-processing, VLMs can classify high-level activities from a few frames, detect motion and grasp with moderate accuracy, and approximate dose counts within 25% of ground truth for mildly impaired and healthy participants, all without task-specific training or finetuning. These results highlight both the current limitations and emerging opportunities of VLMs for data-driven stroke rehabilitation and broader clinical video analysis.

Paperid: 6653, https://arxiv.org/pdf/2511.17255.pdf

Abstract:
Large vision-language models (VLMs) enable intuitive visual search using natural language queries. However, improving their performance often requires fine-tuning and scaling to larger model variants. In this work, we propose a mechanism inspired by traditional text-based search to improve retrieval performance at inference time: relevance feedback. While relevance feedback can serve as an alternative to fine-tuning, its model-agnostic design also enables use with fine-tuned VLMs. Specifically, we introduce and evaluate four feedback strategies for VLM-based retrieval. First, we revise classical pseudo-relevance feedback (PRF), which refines query embeddings based on top-ranked results. To address its limitations, we propose generative relevance feedback (GRF), which uses synthetic captions for query refinement. Furthermore, we introduce an attentive feedback summarizer (AFS), a custom transformer-based model that integrates multimodal fine-grained features from relevant items. Finally, we simulate explicit feedback using ground-truth captions as an upper-bound baseline. Experiments on Flickr30k and COCO with the VLM backbones show that GRF, AFS, and explicit feedback improve retrieval performance by 3-5% in MRR@5 for smaller VLMs, and 1-3% for larger ones, compared to retrieval with no feedback. Moreover, AFS, similarly to explicit feedback, mitigates query drift and is more robust than GRF in iterative, multi-turn retrieval settings. Our findings demonstrate that relevance feedback can consistently enhance retrieval across VLMs and open up opportunities for interactive and adaptive visual search.

Paperid: 6654, https://arxiv.org/pdf/2511.17136.pdf

Abstract:
Device-guided music transfer adapts playback across unseen devices for users who lack them. Existing methods mainly focus on modifying the timbre, rhythm, harmony, or instrumentation to mimic genres or artists, overlooking the diverse hardware properties of the playback device (i.e., speaker). Therefore, we propose DeMT, which processes a speaker's frequency response curve as a line graph using a vision-language model to extract device embeddings. These embeddings then condition a hybrid transformer via feature-wise linear modulation. Fine-tuned on a self-collected dataset, DeMT enables effective speaker-style transfer and robust few-shot adaptation for unseen devices, supporting applications like device-style augmentation and quality enhancement.

Paperid: 6655, https://arxiv.org/pdf/2511.16979.pdf

Abstract:
Open-Set Domain Generalization (OSDG) tackles the realistic scenario where deployed models encounter both domain shifts and novel object categories. Despite impressive progress with vision-language models like CLIP, existing methods still fall into the dilemma between structural risk of known-classes and open-space risk from unknown-classes, and easily suffers from over-confidence, especially when distinguishing ``hard unknowns" that share fine-grained visual similarities with known classes. To this end, we propose a Semantic-enhanced CLIP (SeeCLIP) framework that explicitly addresses this dilemma through fine-grained semantic enhancement. In SeeCLIP, we propose a semantic-aware prompt enhancement module to decompose images into discriminative semantic tokens, enabling nuanced vision-language alignment beyond coarse category labels. To position unknown prompts effectively, we introduce duplex contrastive learning with complementary objectives, that is, repulsion to maintain separability from known classes, and cohesion to preserve semantic proximity. Further, our semantic-guided diffusion module synthesizes pseudo-unknowns by perturbing extracted semantic tokens, generating challenging samples that are visually similar to known classes yet exhibit key local differences. These hard negatives force the model to learn finer decision boundaries. Extensive experiments across five benchmarks demonstrate consistent improvements of 3% accuracy and 5% H-score over state-of-the-art methods.

Paperid: 6656, https://arxiv.org/pdf/2511.16107.pdf

Abstract:
In large language models (LLM), in-context learning (ICL) refers to performing new tasks by conditioning on small demonstrations provided in the input context. Recent advances in visual in-context learning (VICL) demonstrate promising capabilities for solving downstream tasks by unified vision-language models (VLMs). When the visual prompt and the target images originate from different visual tasks, can VLMs still enable VICL? In the paper, we propose a fully collaborative pipeline, i.e. T2T-VICL, for VLMs to investigate the potential of cross-task VICL. Fundamentally, we design a mechanism to generate and select text prompts that best implicitly describe the differences between two distinct low-level vision tasks, and construct the first cross-task VICL dataset. Building upon this, we propose a novel inference framework that combines perceptual score-based reasoning with traditional evaluation metrics to perform cross-task VICL. Our approach achieves top-tier results across nine cross-task scenarios and second-tier performance in ten additional scenarios, unlocking the boundaries of cross-task VICL within VLMs.

Paperid: 6657, https://arxiv.org/pdf/2511.15967.pdf

Abstract:
Recently, the strong generalization ability of CLIP has facilitated open-vocabulary semantic segmentation, which labels pixels using arbitrary text. However, existing methods that fine-tune CLIP for segmentation on limited seen categories often lead to overfitting and degrade the pretrained vision-language alignment. To stabilize modality alignment during fine-tuning, we propose InfoCLIP, which leverages an information-theoretic perspective to transfer alignment knowledge from pretrained CLIP to the segmentation task. Specifically, this transfer is guided by two novel objectives grounded in mutual information. First, we compress the pixel-text modality alignment from pretrained CLIP to reduce noise arising from its coarse-grained local semantic representations learned under image-text supervision. Second, we maximize the mutual information between the alignment knowledge of pretrained CLIP and the fine-tuned model to transfer compact local semantic relations suited for the segmentation task. Extensive evaluations across various benchmarks validate the effectiveness of InfoCLIP in enhancing CLIP fine-tuning for open-vocabulary semantic segmentation, demonstrating its adaptability and superiority in asymmetric transfer.

Paperid: 6658, https://arxiv.org/pdf/2511.14900.pdf

Abstract:
The emergence of vision-language models (VLMs) has opened new possibilities for clinical reasoning and has shown promising performance in dermatological diagnosis. However, their trustworthiness and clinical utility are often limited by three major factors: (1) Data heterogeneity, where diverse datasets lack consistent diagnostic labels and clinical concept annotations; (2) Absence of grounded diagnostic rationales, leading to a scarcity of reliable reasoning supervision; and (3) Limited scalability and generalization, as models trained on small, densely annotated datasets struggle to transfer nuanced reasoning to large, sparsely-annotated ones. To address these limitations, we propose SkinR1, a novel dermatological VLM that combines deep, textbook-based reasoning with the broad generalization capabilities of reinforcement learning (RL). SkinR1 systematically resolves the key challenges through a unified, end-to-end framework. First, we design a textbook-based reasoning generator that synthesizes high-fidelity, hierarchy-aware, and differential-diagnosis (DDx)-informed trajectories, providing reliable expert-level supervision. Second, we leverage the constructed trajectories for supervised fine-tuning (SFT) empowering the model with grounded reasoning ability. Third, we develop a novel RL paradigm that, by incorporating the hierarchical structure of diseases, effectively transfers these grounded reasoning patterns to large-scale, sparse data. Extensive experiments on multiple dermatology datasets demonstrate that SkinR1 achieves superior diagnostic accuracy. The ablation study demonstrates the importance of the reasoning foundation instilled by SFT.

Paperid: 6659, https://arxiv.org/pdf/2511.14161.pdf

Abstract:
Household tidying is an important application area, yet current benchmarks neither model user preferences nor support mobility, and they generalize poorly, making it hard to comprehensively assess integrated language-to-action capabilities. To address this, we propose RoboTidy, a unified benchmark for language-guided household tidying that supports Vision-Language-Action (VLA) and Vision-Language-Navigation (VLN) training and evaluation. RoboTidy provides 500 photorealistic 3D Gaussian Splatting (3DGS) household scenes (covering 500 objects and containers) with collisions, formulates tidying as an "Action (Object, Container)" list, and supplies 6.4k high-quality manipulation demonstration trajectories and 1.5k naviagtion trajectories to support both few-shot and large-scale training. We also deploy RoboTidy in the real world for object tidying, establishing an end-to-end benchmark for household tidying. RoboTidy offers a scalable platform and bridges a key gap in embodied AI by enabling holistic and realistic evaluation of language-guided robots.

Paperid: 6660, https://arxiv.org/pdf/2511.14106.pdf

Abstract:
Reasoning-augmented Vision-Language Models (RVLMs) rely on safety alignment to prevent harmful behavior, yet their exposed chain-of-thought (CoT) traces introduce new attack surfaces. In this work, we find that the safety alignment of RVLMs can be easily break through a novel attack method termed \textbf{Stealth Fine-Tuning}. Our method elicits harmful reasoning traces through \textbf{segment-level interference} and reuses the self-generated outputs as supervised fine-tuning data. Through a \textbf{turn-based weighted} loss design, yielding a lightweight, distribution-consistent finetuning method. In our experiment, with only 499 samples and under 3 hours on a single A100 (QLoRA), Stealth Fine-Tuning outperforms IDEATOR by 38.52\% ASR while preserving general reasoning ability, as the tuned model retains the original representation distribution. Experiments on AdvBench and several general benchmarks demonstrate that Stealth Fine-Tuning is a low-cost and highly effective way to bypass alignment defenses. \textcolor{red}{\textbf{Disclaimer: This paper contains content that may be disturbing or offensive.}}

Paperid: 6661, https://arxiv.org/pdf/2511.14072.pdf

Abstract:
Large Vision-Language Models (LVLMs) usually suffer from prohibitive computational and memory costs due to the quadratic growth of visual tokens with image resolution. Existing token compression methods, while varied, often lack a high-level semantic understanding, leading to suboptimal merges, information redundancy, or context loss. To address these limitations, we introduce CORE (Compact Object-centric REpresentations), a new paradigm for visual token compression. CORE leverages an efficient segmentation decoder to generate object masks, which serve as a high-level semantic prior to guide the merging of visual tokens into a compact set of object-centric representations. Furthermore, a novel centroid-guided sorting mechanism restores a coherent spatial order to the merged tokens, preserving vital positional information. Extensive experiments show that CORE not only establishes a new state-of-the-art on six authoritative benchmarks for fixed-rate compression, but also achieves dramatic efficiency gains in adaptive-rate settings. Even under extreme compression, after aggressively retaining with only 2.2% of all visual tokens, CORE still maintains 97.4% of baseline performance. Our work demonstrates the superiority of object-centric representations for efficient and effective LVLM processing.

Paperid: 6662, https://arxiv.org/pdf/2511.13782.pdf

Abstract:
Large language models (LLMs) and vision language models (VLMs), such as DeepSeek R1,OpenAI o3, and Gemini 2.5 Pro, have demonstrated remarkable reasoning capabilities across logical inference, problem solving, and decision making. However, spatial reasoning:a fundamental component of human cognition that includes mental rotation, navigation, and spatial relationship comprehension remains a significant challenge for current advanced VLMs. We hypothesize that imagination, the internal simulation of spatial states, is the dominant reasoning mechanism within a spatial world model. To test this hypothesis and systematically probe current VLM spatial reasoning mechanisms, we introduce SpatiaLite, a fully synthetic benchmark that jointly measures spatial reasoning accuracy and reasoning efficiency. Comprehensive experiments reveal three key findings. First, advanced VLMs predominantly rely on linguistic representations for reasoning and imagination, resulting in significant deficiencies on visual centric tasks that demand perceptual spatial relations and 3D geometry transformations such as mental rotation or projection prediction. Second, advanced VLMs exhibit severe inefficiency in their current spatial reasoning mechanisms, with token usage growing rapidly as transformation complexity increases. Third, we propose an Imagery Driven Framework (IDF) for data synthesis and training, which can implicitly construct an internal world model that is critical for spatial reasoning in VLMs. Building on SpatiaLite, this work delineates the spatial reasoning limits and patterns of advanced VLMs, identifies key shortcomings, and informs future advances

Paperid: 6663, https://arxiv.org/pdf/2511.13378.pdf

Abstract:
Diagrams are crucial yet underexplored tools in many disciplines, demonstrating the close connection between visual representation and scholarly reasoning. However, their iconic form poses obstacles to visual studies, intermedial analysis, and text-based digital workflows. In particular, Charles S. Peirce consistently advocated the use of diagrams as essential for reasoning and explanation. His manuscripts, often combining textual content with complex visual artifacts, provide a challenging case for studying documents involving heterogeneous materials. In this preliminary study, we investigate whether Visual Language Models (VLMs) can effectively help us identify and interpret such hybrid pages in context. First, we propose a workflow that (i) segments manuscript page layouts, (ii) reconnects each segment to IIIF-compliant annotations, and (iii) submits fragments containing diagrams to a VLM. In addition, by adopting Peirce's semiotic framework, we designed prompts to extract key knowledge about diagrams and produce concise captions. Finally, we integrated these captions into knowledge graphs, enabling structured representations of diagrammatic content within composite sources.

Paperid: 6664, https://arxiv.org/pdf/2511.12937.pdf

Abstract:
Automated operation in cross-platform strategy games demands agents with robust generalization across diverse user interfaces and dynamic battlefield conditions. While vision-language models (VLMs) have shown considerable promise in multimodal reasoning, their application to complex human-computer interaction scenarios--such as strategy gaming--remains largely unexplored. Here, we introduce Yanyun-3, a general-purpose agent framework that, for the first time, enables autonomous cross-platform operation across three heterogeneous strategy game environments. By integrating the vision-language reasoning of Qwen2.5-VL with the precise execution capabilities of UI-TARS, Yanyun-3 successfully performs core tasks including target localization, combat resource allocation, and area control. Through systematic ablation studies, we evaluate the effects of various multimodal data combinations--static images, multi-image sequences, and videos--and propose the concept of combination granularity to differentiate between intra-sample fusion and inter-sample mixing strategies. We find that a hybrid strategy, which fuses multi-image and video data while mixing in static images (MV+S), substantially outperforms full fusion: it reduces inference time by 63% and boosts the BLEU-4 score by a factor of 12 (from 4.81% to 62.41%, approximately 12.98x). Operating via a closed-loop pipeline of screen capture, model inference, and action execution, the agent demonstrates strong real-time performance and cross-platform generalization. Beyond providing an efficient solution for strategy game automation, our work establishes a general paradigm for enhancing VLM performance through structured multimodal data organization, offering new insights into the interplay between static perception and dynamic reasoning in embodied intelligence.

Paperid: 6665, https://arxiv.org/pdf/2511.12738.pdf

Abstract:
Vision Language Models (VLMs) mix visual tokens and text tokens. A puzzling issue is the fact that visual tokens most related to the query receive little to no attention in the final layers of the LLM module of VLMs from the answer tokens, where all tokens are treated equally, in particular, visual and language tokens in the LLM attention layers. This fact may result in wrong answers to visual questions, as our experimental results confirm. It appears that the standard next-token prediction (NTP) loss provides an insufficient signal for directing attention to visual tokens. We hypothesize that a more direct supervision of the attention of visual tokens to corresponding language tokens in the LLM module of VLMs will lead to improved performance on visual tasks. To demonstrate that this is indeed the case, we propose a novel loss function that directly supervises the attention of visual tokens. It directly grounds the answer language tokens in images by directing their attention to the relevant visual tokens. This is achieved by aligning the attention distribution of visual tokens to ground truth attention maps with KL divergence. The ground truth attention maps are obtained from task geometry in synthetic cases or from standard grounding annotations (e.g., bounding boxes or point annotations) in real images, and are used inside the LLM for attention supervision without requiring new labels. The obtained KL attention loss (KLAL) when combined with NTP encourages VLMs to attend to relevant visual tokens while generating answer tokens. This results in notable improvements across geometric tasks, pointing, and referring expression comprehension on both synthetic and real-world data, as demonstrated by our experiments. We also introduce a new dataset to evaluate the line tracing abilities of VLMs. Surprisingly, even commercial VLMs do not perform well on this task.

Paperid: 6666, https://arxiv.org/pdf/2511.12588.pdf

Abstract:
Accurate cell counting in immunohistochemistry (IHC) images is critical for quantifying protein expression and aiding cancer diagnosis. However, the task remains challenging due to the chromogen overlap, variable biomarker staining, and diverse cellular morphologies. Regression-based counting methods offer advantages over detection-based ones in handling overlapped cells, yet rarely support end-to-end multi-class counting. Moreover, the potential of foundation models remains largely underexplored in this paradigm. To address these limitations, we propose a rank-aware agglomeration framework that selectively distills knowledge from multiple strong foundation models, leveraging their complementary representations to handle IHC heterogeneity and obtain a compact yet effective student model, CountIHC. Unlike prior task-agnostic agglomeration strategies that either treat all teachers equally or rely on feature similarity, we design a Rank-Aware Teacher Selecting (RATS) strategy that models global-to-local patch rankings to assess each teacher's inherent counting capacity and enable sample-wise teacher selection. For multi-class cell counting, we introduce a fine-tuning stage that reformulates the task as vision-language alignment. Discrete semantic anchors derived from structured text prompts encode both category and quantity information, guiding the regression of class-specific density maps and improving counting for overlapping cells. Extensive experiments demonstrate that CountIHC surpasses state-of-the-art methods across 12 IHC biomarkers and 5 tissue types, while exhibiting high agreement with pathologists' assessments. Its effectiveness on H&E-stained data further confirms the scalability of the proposed method.

Paperid: 6667, https://arxiv.org/pdf/2511.11811.pdf

Abstract:
Many promising applications of multimodal wearables require continuous sensing and heavy computation, yet users reject such devices due to privacy concerns. This paper shares our experiences building an ear-mounted voice-and-vision wearable that performs local AI inference using a paired smartphone as a trusted personal edge. We describe the hardware--software co-design of this privacy-preserving system, including challenges in integrating a camera, microphone, and speaker within a 30-gram form factor, enabling wake word-triggered capture, and running quantized vision-language and large-language models entirely offline. Through iterative prototyping, we identify key design hurdles in power budgeting, connectivity, latency, and social acceptability. Our initial evaluation shows that fully local multimodal inference is feasible on commodity mobile hardware with interactive latency. We conclude with design lessons for researchers developing embedded AI systems that balance privacy, responsiveness, and usability in everyday settings.

Paperid: 6668, https://arxiv.org/pdf/2511.11298.pdf

Abstract:
Foundation models applied in robotics, particularly \textbf{Vision--Language--Action (VLA)} models, hold great promise for achieving general-purpose manipulation. Yet, systematic real-world evaluations and cross-model comparisons remain scarce. This paper reports our \textbf{empirical experiences} from benchmarking four representative VLAs -- \textbf{ACT}, \textbf{OpenVLA--OFT}, \textbf{RDT-1B}, and \boldmath{$π_0$} -- across four manipulation tasks conducted in both simulation and on the \textbf{ALOHA Mobile} platform. We establish a \textbf{standardized evaluation framework} that measures performance along three key dimensions: (1) \textit{accuracy and efficiency} (success rate and time-to-success), (2) \textit{adaptability} across in-distribution, spatial out-of-distribution, and instance-plus-spatial out-of-distribution settings, and (3) \textit{language instruction-following accuracy}. Through this process, we observe that \boldmath{$π_0$} demonstrates superior adaptability in out-of-distribution scenarios, while \textbf{ACT} provides the highest stability in-distribution. Further analysis highlights differences in computational demands, data-scaling behavior, and recurring failure modes such as near-miss grasps, premature releases, and long-horizon state drift. These findings reveal practical trade-offs among VLA model architectures in balancing precision, generalization, and deployment cost, offering actionable insights for selecting and deploying VLAs in real-world robotic manipulation tasks.

Paperid: 6669, https://arxiv.org/pdf/2511.10119.pdf

Abstract:
We propose a new perspective for approaching artificial general intelligence (AGI) through an intelligence foundation model (IFM). Unlike existing foundation models (FMs), which specialize in pattern learning within specific domains such as language, vision, or time series, IFM aims to acquire the underlying mechanisms of intelligence by learning directly from diverse intelligent behaviors. Vision, language, and other cognitive abilities are manifestations of intelligent behavior; learning from this broad range of behaviors enables the system to internalize the general principles of intelligence. Based on the fact that intelligent behaviors emerge from the collective dynamics of biological neural systems, IFM consists of two core components: a novel network architecture, termed the state neural network, which captures neuron-like dynamic processes, and a new learning objective, neuron output prediction, which trains the system to predict neuronal outputs from collective dynamics. The state neural network emulates the temporal dynamics of biological neurons, allowing the system to store, integrate, and process information over time, while the neuron output prediction objective provides a unified computational principle for learning these structural dynamics from intelligent behaviors. Together, these innovations establish a biologically grounded and computationally scalable foundation for building systems capable of generalization, reasoning, and adaptive learning across domains, representing a step toward truly AGI.

Paperid: 6670, https://arxiv.org/pdf/2511.10074.pdf

Abstract:
We propose Vision-Language Feature-based Multimodal Semantic Communication (VLF-MSC), a unified system that transmits a single compact vision-language representation to support both image and text generation at the receiver. Unlike existing semantic communication techniques that process each modality separately, VLF-MSC employs a pre-trained vision-language model (VLM) to encode the source image into a vision-language semantic feature (VLF), which is transmitted over the wireless channel. At the receiver, a decoder-based language model and a diffusion-based image generator are both conditioned on the VLF to produce a descriptive text and a semantically aligned image. This unified representation eliminates the need for modality-specific streams or retransmissions, improving spectral efficiency and adaptability. By leveraging foundation models, the system achieves robustness to channel noise while preserving semantic fidelity. Experiments demonstrate that VLF-MSC outperforms text-only and image-only baselines, achieving higher semantic accuracy for both modalities under low SNR with significantly reduced bandwidth.

Paperid: 6671, https://arxiv.org/pdf/2511.08917.pdf

Abstract:
Vision-Language Models (VLMs) are increasingly used by blind and low-vision (BLV) people to identify and understand products in their everyday lives, such as food, personal products, and household goods. Despite their prevalence, we lack an empirical understanding of how common image quality issues, like blur and misframing of items, affect the accuracy of VLM-generated captions and whether resulting captions meet BLV people's information needs. Grounded in a survey with 86 BLV people, we systematically evaluate how image quality issues affect captions generated by VLMs. We show that the best model recognizes products in images with no quality issues with 98% accuracy, but drops to 75% accuracy overall when quality issues are present, worsening considerably as issues compound. We discuss the need for model evaluations that center on disabled people's experiences throughout the process and offer concrete recommendations for HCI and ML researchers to make VLMs more reliable for BLV people.

Paperid: 6672, https://arxiv.org/pdf/2511.06899.pdf

Abstract:
Large Vision-Language Models (LVLMs) excel in multimodal reasoning and have shown impressive performance on various multimodal benchmarks. However, most of these benchmarks evaluate models primarily through multiple-choice or short-answer formats, which do not take the reasoning process into account. Although some benchmarks assess the reasoning process, their methods are often overly simplistic and only examine reasoning when answers are incorrect. This approach overlooks scenarios where flawed reasoning leads to correct answers. In addition, these benchmarks do not consider the impact of intermodal relationships on reasoning. To address this issue, we propose the Reasoning Process Tree Score (RPTS), a tree structure-based metric to assess reasoning processes. Specifically, we organize the reasoning steps into a reasoning tree and leverage its hierarchical information to assign weighted faithfulness scores to each reasoning step. By dynamically adjusting these weights, RPTS not only evaluates the overall correctness of the reasoning, but also pinpoints where the model fails in the reasoning. To validate RPTS in real-world multimodal scenarios, we construct a new benchmark, RPTS-Eval, comprising 374 images and 390 reasoning instances. Each instance includes reliable visual-textual clues that serve as leaf nodes of the reasoning tree. Furthermore, we define three types of intermodal relationships to investigate how intermodal interactions influence the reasoning process. We evaluated representative LVLMs (e.g., GPT4o, Llava-Next), uncovering their limitations in multimodal reasoning and highlighting the differences between open-source and closed-source commercial LVLMs. We believe that this benchmark will contribute to the advancement of research in the field of multimodal reasoning.

Paperid: 6673, https://arxiv.org/pdf/2511.06146.pdf

Abstract:
Spatial Reasoning is an important component of human cognition and is an area in which the latest Vision-language models (VLMs) show signs of difficulty. The current analysis works use image captioning tasks and visual question answering. In this work, we propose using the Referring Expression Comprehension task instead as a platform for the evaluation of spatial reasoning by VLMs. This platform provides the opportunity for a deeper analysis of spatial comprehension and grounding abilities when there is 1) ambiguity in object detection, 2) complex spatial expressions with a longer sentence structure and multiple spatial relations, and 3) expressions with negation ('not'). In our analysis, we use task-specific architectures as well as large VLMs and highlight their strengths and weaknesses in dealing with these specific situations. While all these models face challenges with the task at hand, the relative behaviors depend on the underlying models and the specific categories of spatial semantics (topological, directional, proximal, etc.). Our results highlight these challenges and behaviors and provide insight into research gaps and future directions.

Paperid: 6674, https://arxiv.org/pdf/2511.06005.pdf

Abstract:
Vision Language Models (VLMs) are increasingly deployed across downstream tasks, yet their training data often encode social biases that surface in outputs. Unlike humans, who interpret images through contextual and social cues, VLMs process them through statistical associations, often leading to reasoning that diverges from human reasoning. By analyzing how a VLM reasons, we can understand how inherent biases are perpetuated and can adversely affect downstream performance. To examine this gap, we systematically analyze social biases in five open-source VLMs for an occupation prediction task, on the FairFace dataset. Across 32 occupations and three different prompting styles, we elicit both predictions and reasoning. Our findings reveal that the biased reasoning patterns systematically underlie intersectional disparities, highlighting the need to align VLM reasoning with human values prior to its downstream deployment.

Paperid: 6675, https://arxiv.org/pdf/2511.05782.pdf

Abstract:
Unsupervised domain adaptation for medical image segmentation remains a significant challenge due to substantial domain shifts across imaging modalities, such as CT and MRI. While recent vision-language representation learning methods have shown promise, their potential in UDA segmentation tasks remains underexplored. To address this gap, we propose TCSA-UDA, a Text-driven Cross-Semantic Alignment framework that leverages domain-invariant textual class descriptions to guide visual representation learning. Our approach introduces a vision-language covariance cosine loss to directly align image encoder features with inter-class textual semantic relations, encouraging semantically meaningful and modality-invariant feature representations. Additionally, we incorporate a prototype alignment module that aligns class-wise pixel-level feature distributions across domains using high-level semantic prototypes. This mitigates residual category-level discrepancies and enhances cross-modal consistency. Extensive experiments on challenging cross-modality cardiac, abdominal, and brain tumor segmentation benchmarks demonstrate that our TCSA-UDA framework significantly reduces domain shift and consistently outperforms state-of-the-art UDA methods, establishing a new paradigm for integrating language-driven semantics into domain-adaptive medical image analysis.

Paperid: 6676, https://arxiv.org/pdf/2511.05325.pdf

Abstract:
Multimodal product retrieval systems in e-commerce platforms rely on effectively combining visual and textual signals to improve search relevance and user experience. However, vision-language models such as CLIP are vulnerable to typographic attacks, where misleading or irrelevant text embedded in images skews model predictions. In this work, we propose a novel method that reverses the logic of typographic attacks by rendering relevant textual content (e.g., titles, descriptions) directly onto product images to perform vision-text compression, thereby strengthening image-text alignment and boosting multimodal product retrieval performance. We evaluate our method on three vertical-specific e-commerce datasets (sneakers, handbags, and trading cards) using six state-of-the-art vision foundation models. Our experiments demonstrate consistent improvements in unimodal and multimodal retrieval accuracy across categories and model families. Our findings suggest that visually rendering product metadata is a simple yet effective enhancement for zero-shot multimodal retrieval in e-commerce applications.

Paperid: 6677, https://arxiv.org/pdf/2511.04976.pdf

Abstract:
We introduce iFlyBot-VLM, a general-purpose Vision-Language Model (VLM) used to improve the domain of Embodied Intelligence. The central objective of iFlyBot-VLM is to bridge the cross-modal semantic gap between high-dimensional environmental perception and low-level robotic motion control. To this end, the model abstracts complex visual and spatial information into a body-agnostic and transferable Operational Language, thereby enabling seamless perception-action closed-loop coordination across diverse robotic platforms. The architecture of iFlyBot-VLM is systematically designed to realize four key functional capabilities essential for embodied intelligence: 1) Spatial Understanding and Metric Reasoning; 2) Interactive Target Grounding; 3) Action Abstraction and Control Parameter Generation; 4) Task Planning and Skill Sequencing. We envision iFlyBot-VLM as a scalable and generalizable foundation model for embodied AI, facilitating the progression from specialized task-oriented systems toward generalist, cognitively capable agents. We conducted evaluations on 10 current mainstream embodied intelligence-related VLM benchmark datasets, such as Blink and Where2Place, and achieved optimal performance while preserving the model's general capabilities. We will publicly release both the training data and model weights to foster further research and development in the field of Embodied Intelligence.

Paperid: 6678, https://arxiv.org/pdf/2511.02360.pdf

Abstract:
In human cognition, there exist numerous thought processes that are tacit and beyond verbal expression, enabling us to understand and interact with the world in multiple ways. However, contemporary Vision-Language Models (VLMs) remain constrained to reasoning within the discrete and rigid space of linguistic tokens, thereby bottlenecking the rich, high-dimensional nature of visual perception. To bridge this gap, we propose CoCoVa (Chain of Continuous Vision-Language Thought), a novel framework for vision-language model that leverages continuous cross-modal reasoning for diverse vision-language tasks. The core of CoCoVa is an iterative reasoning cycle, where a novel Latent Q-Former (LQ-Former) acts as a dynamic reasoning engine, iteratively refining a chain of latent thought vectors through cross-modal fusion. To focus this process, a token selection mechanism dynamically identifies salient visual regions, mimicking attentional focus. To ensure these latent thoughts remain grounded, we train the model with a multi-task objective that combines contrastive learning and diffusion-based reconstruction, enforcing alignment between latent representations and both visual and textual modalities. Evaluations show CoCoVa improves accuracy and token efficiency over strong baselines. With a 1.5B backbone, it competes with or surpasses larger 7B-9B models on almost all benchmarks. When scaled to 7B LLM backbones, it remains competitive with state-of-the-art models. Qualitative analysis validates that learned latent space captures interpretable and structured reasoning patterns, highlighting the potential of CoCoVa to bridge the representational gap between discrete language processing and the continuous nature of visual understanding.

Paperid: 6679, https://arxiv.org/pdf/2511.01914.pdf

Abstract:
We introduce iFlyBot-VLA, a large-scale Vision-Language-Action (VLA) model trained under a novel framework. The main contributions are listed as follows: (1) a latent action model thoroughly trained on large-scale human and robotic manipulation videos; (2) a dual-level action representation framework that jointly supervises both the Vision-Language Model (VLM) and the action expert during training; (3) a mixed training strategy that combines robot trajectory data with general QA and spatial QA datasets, effectively enhancing the 3D perceptual and reasoning capabilities of the VLM backbone. Specifically, the VLM is trained to predict two complementary forms of actions: latent actions, derived from our latent action model pretrained on cross-embodiment manipulation data, which capture implicit high-level intentions; and structured discrete action tokens, obtained through frequency-domain transformations of continuous control signals, which encode explicit low-level dynamics. This dual supervision aligns the representation spaces of language, vision, and action, enabling the VLM to directly contribute to action generation. Experimental results on the LIBERO Franka benchmark demonstrate the superiority of our frame-work, while real-world evaluations further show that iFlyBot-VLA achieves competitive success rates across diverse and challenging manipulation tasks. Furthermore, we plan to open-source a portion of our self-constructed dataset to support future research in the community

Paperid: 6680, https://arxiv.org/pdf/2511.01449.pdf

Abstract:
To effectively manage the wastage of perishable fruits, it is crucial to accurately predict their freshness or shelf life using non-invasive methods that rely on visual data. In this regard, deep learning techniques can offer a viable solution. However, obtaining fine-grained fruit freshness labels from experts is costly, leading to a scarcity of data. Closed proprietary Vision Language Models (VLMs), such as Gemini, have demonstrated strong performance in fruit freshness detection task in both zero-shot and few-shot settings. Nonetheless, food retail organizations are unable to utilize these proprietary models due to concerns related to data privacy, while existing open-source VLMs yield sub-optimal performance for the task. Fine-tuning these open-source models with limited data fails to achieve the performance levels of proprietary models. In this work, we introduce a Model-Agnostic Ordinal Meta-Learning (MAOML) algorithm, designed to train smaller VLMs. This approach utilizes meta-learning to address data sparsity and leverages label ordinality, thereby achieving state-of-the-art performance in the fruit freshness classification task under both zero-shot and few-shot settings. Our method achieves an industry-standard accuracy of 92.71%, averaged across all fruits. Keywords: Fruit Quality Prediction, Vision Language Models, Meta Learning, Ordinal Regression

Paperid: 6681, https://arxiv.org/pdf/2511.00831.pdf

Abstract:
Visual-Language Pre-training (VLP) models have achieved significant performance across various downstream tasks. However, they remain vulnerable to adversarial examples. While prior efforts focus on improving the adversarial transferability of multimodal adversarial examples through cross-modal interactions, these approaches suffer from overfitting issues, due to a lack of input diversity by relying excessively on information from adversarial examples in one modality when crafting attacks in another. To address this issue, we draw inspiration from strategies in some adversarial training methods and propose a novel attack called Local Shuffle and Sample-based Attack (LSSA). LSSA randomly shuffles one of the local image blocks, thus expanding the original image-text pairs, generating adversarial images, and sampling around them. Then, it utilizes both the original and sampled images to generate the adversarial texts. Extensive experiments on multiple models and datasets demonstrate that LSSA significantly enhances the transferability of multimodal adversarial examples across diverse VLP models and downstream tasks. Moreover, LSSA outperforms other advanced attacks on Large Vision-Language Models.

Paperid: 6682, https://arxiv.org/pdf/2510.26441.pdf

Abstract:
Test-time prompt tuning (TPT) has emerged as a promising technique for adapting large vision-language models (VLMs) to unseen tasks without relying on labeled data. However, the lack of dispersion between textual features can hurt calibration performance, which raises concerns about VLMs' reliability, trustworthiness, and safety. Current TPT approaches primarily focus on improving prompt calibration by either maximizing average textual feature dispersion or enforcing orthogonality constraints to encourage angular separation. However, these methods may not always have optimal angular separation between class-wise textual features, which implies overlooking the critical role of angular diversity. To address this, we propose A-TPT, a novel TPT framework that introduces angular diversity to encourage uniformity in the distribution of normalized textual features induced by corresponding learnable prompts. This uniformity is achieved by maximizing the minimum pairwise angular distance between features on the unit hypersphere. We show that our approach consistently surpasses state-of-the-art TPT methods in reducing the aggregate average calibration error while maintaining comparable accuracy through extensive experiments with various backbones on different datasets. Notably, our approach exhibits superior zero-shot calibration performance on natural distribution shifts and generalizes well to medical datasets. We provide extensive analyses, including theoretical aspects, to establish the grounding of A-TPT. These results highlight the potency of promoting angular diversity to achieve well-dispersed textual features, significantly improving VLM calibration during test-time adaptation. Our code will be made publicly available.

Paperid: 6683, https://arxiv.org/pdf/2510.25713.pdf

Abstract:
We adapt a pre-trained Vision-Language-Action (VLA) model (Open-VLA) for dexterous human-robot collaboration with minimal language prompting. Our approach adds (i) FiLM conditioning to visual backbones for task-aware perception, (ii) an auxiliary intent head that predicts collaborator hand pose and target cues, and (iii) action-space post-processing that predicts compact deltas (position/rotation) and PCA-reduced finger joints before mapping to full commands. Using a multi-view, teleoperated Franka and Mimic-hand dataset augmented with MediaPipe hand poses, we demonstrate that delta actions are well-behaved and that four principal components explain ~96% of hand-joint variance. Ablations identify action post-processing as the primary performance driver; auxiliary intent helps, FiLM is mixed, and a directional motion loss is detrimental. A real-time stack (~0.3 s latency on one RTX 4090) composes "pick-up" and "pass" into a long-horizon behavior. We surface "trainer overfitting" to specific demonstrators as the key limitation.

Paperid: 6684, https://arxiv.org/pdf/2510.24180.pdf

Abstract:
The surge of audiovisual content on streaming platforms and social media has heightened the demand for accurate and accessible subtitles. However, existing subtitle generation methods primarily speech-based transcription or OCR-based extraction suffer from several shortcomings, including poor synchronization, incorrect or harmful text, inconsistent formatting, inappropriate reading speeds, and the inability to adapt to dynamic audio-visual contexts. Current approaches often address isolated issues, leaving post-editing as a labor-intensive and time-consuming process. In this paper, we introduce V-SAT (Video Subtitle Annotation Tool), a unified framework that automatically detects and corrects a wide range of subtitle quality issues. By combining Large Language Models(LLMs), Vision-Language Models (VLMs), Image Processing, and Automatic Speech Recognition (ASR), V-SAT leverages contextual cues from both audio and video. Subtitle quality improved, with the SUBER score reduced from 9.6 to 3.54 after resolving all language mode issues and F1-scores of ~0.80 for image mode issues. Human-in-the-loop validation ensures high-quality results, providing the first comprehensive solution for robust subtitle annotation.

Paperid: 6685, https://arxiv.org/pdf/2510.23066.pdf

Abstract:
Financial documents are essential sources of information for regulators, auditors, and financial institutions, particularly for assessing the wealth and compliance of Small and Medium-sized Businesses. However, SMB documents are often difficult to parse. They are rarely born digital and instead are distributed as scanned images that are none machine readable. The scans themselves are low in resolution, affected by skew or rotation, and often contain noisy backgrounds. These documents also tend to be heterogeneous, mixing narratives, tables, figures, and multilingual content within the same report. Such characteristics pose major challenges for automated information extraction, especially when relying on end to end large Vision Language Models, which are computationally expensive, sensitive to noise, and slow when applied to files with hundreds of pages. We propose a multistage pipeline that leverages traditional image processing models and OCR extraction, together with compact VLMs for structured field extraction of large-scale financial documents. Our approach begins with image pre-processing, including segmentation, orientation detection, and size normalization. Multilingual OCR is then applied to recover page-level text. Upon analyzing the text information, pages are retrieved for coherent sections. Finally, compact VLMs are operated within these narrowed-down scopes to extract structured financial indicators. Our approach is evaluated using an internal corpus of multi-lingual, scanned financial documents. The results demonstrate that compact VLMs, together with a multistage pipeline, achieves 8.8 times higher field level accuracy relative to directly feeding the whole document into large VLMs, only at 0.7 percent of the GPU cost and 92.6 percent less end-to-end service latency.

Paperid: 6686, https://arxiv.org/pdf/2510.22702.pdf

Abstract:
We introduce the {\em Atlas Urban Index} (AUI), a metric for measuring urban development computed using Sentinel-2 \citep{spoto2012sentinel2} satellite imagery. Existing approaches, such as the {\em Normalized Difference Built-up Index} (NDBI), often struggle to accurately capture urban development due to factors like atmospheric noise, seasonal variation, and cloud cover. These limitations hinder large-scale monitoring of human development and urbanization. To address these challenges, we propose an approach that leverages {\em Vision-Language Models }(VLMs) to provide a development score for regions. Specifically, we collect a time series of Sentinel-2 images for each region. Then, we further process the images within fixed time windows to get an image with minimal cloud cover, which serves as the representative image for that time window. To ensure consistent scoring, we adopt two strategies: (i) providing the VLM with a curated set of reference images representing different levels of urbanization, and (ii) supplying the most recent past image to both anchor temporal consistency and mitigate cloud-related noise in the current image. Together, these components enable AUI to overcome the challenges of traditional urbanization indices and produce more reliable and stable development scores. Our qualitative experiments on Bangalore suggest that AUI outperforms standard indices such as NDBI.

Paperid: 6687, https://arxiv.org/pdf/2510.22141.pdf

Abstract:
Vision-Language Models (VLMs) have shown significant progress in open-set challenges. However, the limited availability of 3D datasets hinders their effective application in 3D scene understanding. We propose LOC, a general language-guided framework adaptable to various occupancy networks, supporting both supervised and self-supervised learning paradigms. For self-supervised tasks, we employ a strategy that fuses multi-frame LiDAR points for dynamic/static scenes, using Poisson reconstruction to fill voids, and assigning semantics to voxels via K-Nearest Neighbor (KNN) to obtain comprehensive voxel representations. To mitigate feature over-homogenization caused by direct high-dimensional feature distillation, we introduce Densely Contrastive Learning (DCL). DCL leverages dense voxel semantic information and predefined textual prompts. This efficiently enhances open-set recognition without dense pixel-level supervision, and our framework can also leverage existing ground truth to further improve performance. Our model predicts dense voxel features embedded in the CLIP feature space, integrating textual and image pixel information, and classifies based on text and semantic similarity. Experiments on the nuScenes dataset demonstrate the method's superior performance, achieving high-precision predictions for known classes and distinguishing unknown classes without additional training data.

Paperid: 6688, https://arxiv.org/pdf/2510.22045.pdf

Abstract:
Vision-language models (VLMs) are increasingly used to evaluate multimodal content, including presentation slides, yet their slide-specific understanding remains underexplored {despite their growing role as critics in agentic, model-forward pipelines}. We introduce VLM-SlideEval, an evaluation framework that probes VLMs along three axes: (1) element-level extraction from slide images aligned to ground truth; (2) robustness to controlled perturbations in geometry, style, and text; and (3) higher-level comprehension, such as recovering a deck's narrative order from shuffled slides. Using publicly available decks from Zenodo (https://huggingface.co/datasets/Forceless/Zenodo10K/viewer/default/pptx), we standardize ground-truth element metadata from PowerPoint XML and live renderings into a unified, verifiable schema. Empirically, VLMs underperform on pixel-accurate extraction and show non-trivial agreement, fidelity, and consistency under controlled perturbations, while performing better on single-slide content understanding; however, they do not reliably capture narrative structure across slides. These results highlight the limits of current VLMs for slide evaluation and motivate calibrated, critic-in-the-loop evaluators that drive iterative refinement and selection in agentic pipelines.

Paperid: 6689, https://arxiv.org/pdf/2510.21881.pdf

Abstract:
Large language models (LLMs) have demonstrated strong reasoning capabilities in text-based mathematical problem solving; however, when adapted to visual reasoning tasks, particularly geometric problem solving, their performance substantially declines because geometric problems present unique challenges. Specifically, these challenges stem from two key factors: first, the intrinsic complexity of geometry requiring detailed image comprehension and multi-step reasoning, and second, the limitations of existing datasets which lack sufficient scale, diversity, and explicit reasoning traces, consequently hindering effective model training. To address these challenges, we developed the GeoThoughts dataset, a comprehensive geometric reasoning corpus with two subsets: Geo-Thought-6K with 6,243 samples and its augmented version Geo-Thought-Augmented-10K containing 10,834 samples. Each entry includes visual descriptions, step-by-step solutions, explicit reasoning chains, reflection steps, and final answers. Using this dataset, we developed GeoThought-MLLM, a mathematical reasoning multimodal model that generates detailed thinking processes during problem-solving. Our model outperforms existing benchmarks in geometric tasks, demonstrating that training with our Chain-of-Thought dataset improves geometric reasoning capabilities across both in-domain and out-of-domain settings. Finally, we analyze failure cases and observe that errors primarily arise from incorrect interpretation of mathematical concepts or spatial misjudgment. By invoking CoT to correct these mistakes, the model produces correct answers.

Paperid: 6690, https://arxiv.org/pdf/2510.21763.pdf

Abstract:
While modern text-to-image diffusion models generate high-fidelity images, they offer limited control over the spatial and geometric structure of the output. To address this, we introduce and evaluate two ControlNets specialized for artistic control: (1) a proportion ControlNet that uses bounding boxes to dictate the position and scale of objects, and (2) a perspective ControlNet that employs vanishing lines to control the 3D geometry of the scene. We support the training of these modules with data pipelines that leverage vision-language models for annotation and specialized algorithms for conditioning image synthesis. Our experiments demonstrate that both modules provide effective control but exhibit limitations with complex constraints. Both models are released on HuggingFace: https://huggingface.co/obvious-research

Paperid: 6691, https://arxiv.org/pdf/2510.21069.pdf

Abstract:
Understanding and reasoning about complex 3D environments requires structured scene representations that capture not only objects but also their semantic and spatial relationships. While recent works on 3D scene graph generation have leveraged pretrained VLMs without task-specific fine-tuning, they are largely confined to single-view settings, fail to support incremental updates as new observations arrive and lack explicit geometric grounding in 3D space, all of which are essential for embodied scenarios. In this paper, we propose, ZING-3D, a framework that leverages the vast knowledge of pretrained foundation models to enable open-vocabulary recognition and generate a rich semantic representation of the scene in a zero-shot manner while also enabling incremental updates and geometric grounding in 3D space, making it suitable for downstream robotics applications. Our approach leverages VLM reasoning to generate a rich 2D scene graph, which is grounded in 3D using depth information. Nodes represent open-vocabulary objects with features, 3D locations, and semantic context, while edges capture spatial and semantic relations with inter-object distances. Our experiments on scenes from the Replica and HM3D dataset show that ZING-3D is effective at capturing spatial and relational knowledge without the need of task-specific training.

Paperid: 6692, https://arxiv.org/pdf/2510.19333.pdf

Abstract:
This paper presents a novel training-free framework for open-vocabulary image segmentation and object recognition (OVSR), which leverages EfficientNetB0, a convolutional neural network, for unsupervised segmentation and CLIP, a vision-language model, for open-vocabulary object recognition. The proposed framework adopts a two stage pipeline: unsupervised image segmentation followed by segment-level recognition via vision-language alignment. In the first stage, pixel-wise features extracted from EfficientNetB0 are decomposed using singular value decomposition to obtain latent representations, which are then clustered using hierarchical clustering to segment semantically meaningful regions. The number of clusters is adaptively determined by the distribution of singular values. In the second stage, the segmented regions are localized and encoded into image embeddings using the Vision Transformer backbone of CLIP. Text embeddings are precomputed using CLIP's text encoder from category-specific prompts, including a generic something else prompt to support open set recognition. The image and text embeddings are concatenated and projected into a shared latent feature space via SVD to enhance cross-modal alignment. Recognition is performed by computing the softmax over the similarities between the projected image and text embeddings. The proposed method is evaluated on standard benchmarks, including COCO, ADE20K, and PASCAL VOC, achieving state-of-the-art performance in terms of Hungarian mIoU, precision, recall, and F1-score. These results demonstrate the effectiveness, flexibility, and generalizability of the proposed framework.

Paperid: 6693, https://arxiv.org/pdf/2510.18502.pdf

Abstract:
Vehicle make and model recognition (VMMR) is an important task in intelligent transportation systems, but existing approaches struggle to adapt to newly released models. Contrastive Language-Image Pretraining (CLIP) provides strong visual-text alignment, yet its fixed pretrained weights limit performance without costly image-specific finetuning. We propose a pipeline that integrates vision language models (VLMs) with Retrieval-Augmented Generation (RAG) to support zero-shot recognition through text-based reasoning. A VLM converts vehicle images into descriptive attributes, which are compared against a database of textual features. Relevant entries are retrieved and combined with the description to form a prompt, and a language model (LM) infers the make and model. This design avoids large-scale retraining and enables rapid updates by adding textual descriptions of new vehicles. Experiments show that the proposed method improves recognition by nearly 20% over the CLIP baseline, demonstrating the potential of RAG-enhanced LM reasoning for scalable VMMR in smart-city applications.

Paperid: 6694, https://arxiv.org/pdf/2508.19289.pdf

Abstract:
We present an unsupervised slide-quality assessment pipeline that combines seven expert-inspired visual-design metrics (whitespace, colorfulness, edge density, brightness contrast, text density, color harmony, layout balance) with CLIP-ViT embeddings, using Isolation Forest-based anomaly scoring to evaluate presentation slides. Trained on 12k professional lecture slides and evaluated on six academic talks (115 slides), our method achieved Pearson correlations up to 0.83 with human visual-quality ratings-1.79x to 3.23x stronger than scores from leading vision-language models (ChatGPT o4-mini-high, ChatGPT o3, Claude Sonnet 4, Gemini 2.5 Pro). We demonstrate convergent validity with visual ratings, discriminant validity against speaker-delivery scores, and exploratory alignment with overall impressions. Our results show that augmenting low-level design cues with multimodal embeddings closely approximates audience perceptions of slide quality, enabling scalable, objective feedback in real time.

Paperid: 6695, https://arxiv.org/pdf/2508.19254.pdf

Abstract:
This paper presents a real-time generative drawing system that interprets and integrates both formal intent - the structural, compositional, and stylistic attributes of a sketch - and contextual intent - the semantic and thematic meaning inferred from its visual content - into a unified transformation process. Unlike conventional text-prompt-based generative systems, which primarily capture high-level contextual descriptions, our approach simultaneously analyzes ground-level intuitive geometric features such as line trajectories, proportions, and spatial arrangement, and high-level semantic cues extracted via vision-language models. These dual intent signals are jointly conditioned in a multi-stage generation pipeline that combines contour-preserving structural control with style- and content-aware image synthesis. Implemented with a touchscreen-based interface and distributed inference architecture, the system achieves low-latency, two-stage transformation while supporting multi-user collaboration on shared canvases. The resulting platform enables participants, regardless of artistic expertise, to engage in synchronous, co-authored visual creation, redefining human-AI interaction as a process of co-creation and mutual enhancement.

Paperid: 6696, https://arxiv.org/pdf/2508.17568.pdf

Abstract:
Metamaterials are micro-architected structures whose geometry imparts highly tunable-often counter-intuitive-bulk properties. Yet their design is difficult because of geometric complexity and a non-trivial mapping from architecture to behaviour. We address these challenges with three complementary contributions. (i) MetaDSL: a compact, semantically rich domain-specific language that captures diverse metamaterial designs in a form that is both human-readable and machine-parsable. (ii) MetaDB: a curated repository of more than 150,000 parameterized MetaDSL programs together with their derivatives-three-dimensional geometry, multi-view renderings, and simulated elastic properties. (iii) MetaBench: benchmark suites that test three core capabilities of vision-language metamaterial assistants-structure reconstruction, property-driven inverse design, and performance prediction. We establish baselines by fine-tuning state-of-the-art vision-language models and deploy an omni-model within an interactive, CAD-like interface. Case studies show that our framework provides a strong first step toward integrated design and understanding of structure-representation-property relationships.

Paperid: 6697, https://arxiv.org/pdf/2508.16674.pdf

Abstract:
Medical report interpretation plays a crucial role in healthcare, enabling both patient-facing explanations and effective information flow across clinical systems. While recent vision-language models (VLMs) and large language models (LLMs) have demonstrated general document understanding capabilities, there remains a lack of standardized benchmarks to assess structured interpretation quality in medical reports. We introduce MedRepBench, a comprehensive benchmark built from 1,900 de-identified real-world Chinese medical reports spanning diverse departments, patient demographics, and acquisition formats. The benchmark is designed primarily to evaluate end-to-end VLMs for structured medical report understanding. To enable controlled comparisons, we also include a text-only evaluation setting using high-quality OCR outputs combined with LLMs, allowing us to estimate the upper-bound performance when character recognition errors are minimized. Our evaluation framework supports two complementary protocols: (1) an objective evaluation measuring field-level recall of structured clinical items, and (2) an automated subjective evaluation using a powerful LLM as a scoring agent to assess factuality, interpretability, and reasoning quality. Based on the objective metric, we further design a reward function and apply Group Relative Policy Optimization (GRPO) to improve a mid-scale VLM, achieving up to 6% recall gain. We also observe that the OCR+LLM pipeline, despite strong performance, suffers from layout-blindness and latency issues, motivating further progress toward robust, fully vision-based report understanding.

Paperid: 6698, https://arxiv.org/pdf/2508.11196.pdf

Abstract:
Recent advances in vision-language models (VLMs) have demonstrated strong generalization in natural image tasks. However, their performance often degrades on unmanned aerial vehicle (UAV)-based aerial imagery, which features high resolution, complex spatial semantics, and strict real-time constraints. These challenges limit the applicability of general-purpose VLMs to structured aerial reasoning tasks. To address these challenges, we propose UAV-VL-R1, a lightweight VLM explicitly designed for aerial visual reasoning. It is trained using a hybrid method that combines supervised fine-tuning (SFT) and multi-stage reinforcement learning (RL). We leverage the group relative policy optimization (GRPO) algorithm to promote structured and interpretable reasoning through rule-guided rewards and intra-group policy alignment. To support model training and evaluation, we introduce a high-resolution visual question answering dataset named HRVQA-VL, which consists of 50,019 annotated samples covering eight UAV-relevant reasoning tasks, including object counting, transportation recognition, and spatial scene inference. Experimental results show that UAV-VL-R1 achieves a 48.17% higher zero-shot accuracy than the Qwen2-VL-2B-Instruct baseline and even outperforms its 72B-scale variant, which is 36x larger, on multiple tasks. Ablation studies reveal that while SFT improves semantic alignment, it may reduce reasoning diversity in mathematical tasks. GRPO-based RL compensates for this limitation by enhancing logical flexibility and the robustness of inference. Additionally, UAV-VL-R1 requires only 3.9GB of memory under FP16 inference and can be quantized to 2.5GB with INT8, supporting real-time deployment on resource-constrained UAV platforms.

Paperid: 6699, https://arxiv.org/pdf/2508.10935.pdf

Abstract:
Traditional closed-set 3D detection frameworks fail to meet the demands of open-world applications like autonomous driving. Existing open-vocabulary 3D detection methods typically adopt a two-stage pipeline consisting of pseudo-label generation followed by semantic alignment. While vision-language models (VLMs) recently have dramatically improved the semantic accuracy of pseudo-labels, their geometric quality, particularly bounding box precision, remains commonly neglected. To address this issue, we propose a High Box Quality Open-Vocabulary 3D Detection (HQ-OV3D) framework, dedicated to generate and refine high-quality pseudo-labels for open-vocabulary classes. The framework comprises two key components: an Intra-Modality Cross-Validated (IMCV) Proposal Generator that utilizes cross-modality geometric consistency to generate high-quality initial 3D proposals, and an Annotated-Class Assisted (ACA) Denoiser that progressively refines 3D proposals by leveraging geometric priors from annotated categories through a DDIM-based denoising mechanism. Compared to the state-of-the-art method, training with pseudo-labels generated by our approach achieves a 7.37% improvement in mAP on novel classes, demonstrating the superior quality of the pseudo-labels produced by our framework. HQ-OV3D can serve not only as a strong standalone open-vocabulary 3D detector but also as a plug-in high-quality pseudo-label generator for existing open-vocabulary detection or annotation pipelines.

Paperid: 6700, https://arxiv.org/pdf/2508.10645.pdf

Abstract:
Visual transfer learning for unseen categories presents an active research topic yet a challenging task, due to the inherent conflict between preserving category-specific representations and acquiring transferable knowledge. Vision-Language Models (VLMs) pre-trained on large amounts of image-text pairs offer a promising solution. However, existing prompt tuning methods rely on sparse category labels or disparate LLM-generated descriptions, which fragment knowledge representation and hinder transferability. To address this limitation, we introduce Semantic Prompt Tuning (SemPT), a novel framework that tackles the generalization challenge by leveraging shared attribute-level knowledge across categories. Specifically, SemPT adopts a two-step prompting strategy to guide LLM in extracting shared visual attributes and generating attribute-level descriptions, capturing transferable semantic cues beyond labels while ensuring coherent structure. Then, visually guided weighting is applied to the embeddings of attribute-level descriptions to reduce noise from irrelevant attributes and enhance the text embeddings. Additionally, image embeddings are jointly aligned with both label and attribute-enhanced text embeddings, balancing discrimination for seen categories and transferability to unseen ones. Considering the availability of category exposure, our inference dynamically selects between standard label embeddings for seen categories and attribute-enhanced embeddings for unseen ones to ensure effective adaptation. Extensive experiments on 15 benchmark datasets demonstrate that SemPT achieves state-of-the-art performance across various settings, including base-to-novel generalization, cross-dataset transfer, cross-domain transfer, and few-shot learning.

Paperid: 6701, https://arxiv.org/pdf/2508.10399.pdf

Abstract:
Embodied AI aims to develop intelligent systems with physical forms capable of perceiving, decision-making, acting, and learning in real-world environments, providing a promising way to Artificial General Intelligence (AGI). Despite decades of explorations, it remains challenging for embodied agents to achieve human-level intelligence for general-purpose tasks in open dynamic environments. Recent breakthroughs in large models have revolutionized embodied AI by enhancing perception, interaction, planning and learning. In this article, we provide a comprehensive survey on large model empowered embodied AI, focusing on autonomous decision-making and embodied learning. We investigate both hierarchical and end-to-end decision-making paradigms, detailing how large models enhance high-level planning, low-level execution, and feedback for hierarchical decision-making, and how large models enhance Vision-Language-Action (VLA) models for end-to-end decision making. For embodied learning, we introduce mainstream learning methodologies, elaborating on how large models enhance imitation learning and reinforcement learning in-depth. For the first time, we integrate world models into the survey of embodied AI, presenting their design methods and critical roles in enhancing decision-making and learning. Though solid advances have been achieved, challenges still exist, which are discussed at the end of this survey, potentially as the further research directions.

Paperid: 6702, https://arxiv.org/pdf/2508.10171.pdf

Abstract:
Large-scale Vision-Language Models (VLMs) have transformed general-purpose visual recognition through strong zero-shot capabilities. However, their performance degrades significantly in niche, safety-critical domains such as industrial spill detection, where hazardous events are rare, sensitive, and difficult to annotate. This scarcity -- driven by privacy concerns, data sensitivity, and the infrequency of real incidents -- renders conventional fine-tuning of detectors infeasible for most industrial settings. We address this challenge by introducing a scalable framework centered on a high-quality synthetic data generation pipeline. We demonstrate that this synthetic corpus enables effective Parameter-Efficient Fine-Tuning (PEFT) of VLMs and substantially boosts the performance of state-of-the-art object detectors such as YOLO and DETR. Notably, in the absence of synthetic data (SynSpill dataset), VLMs still generalize better to unseen spill scenarios than these detectors. When SynSpill is used, both VLMs and detectors achieve marked improvements, with their performance becoming comparable. Our results underscore that high-fidelity synthetic data is a powerful means to bridge the domain gap in safety-critical applications. The combination of synthetic generation and lightweight adaptation offers a cost-effective, scalable pathway for deploying vision systems in industrial environments where real data is scarce/impractical to obtain. Project Page: https://synspill.vercel.app

Paperid: 6703, https://arxiv.org/pdf/2508.09061.pdf

Abstract:
Open-set perception in complex traffic environments poses a critical challenge for autonomous driving systems, particularly in identifying previously unseen object categories, which is vital for ensuring safety. Visual Language Models (VLMs), with their rich world knowledge and strong semantic reasoning capabilities, offer new possibilities for addressing this task. However, existing approaches typically leverage VLMs to extract visual features and couple them with traditional object detectors, resulting in multi-stage error propagation that hinders perception accuracy. To overcome this limitation, we propose VLM-3D, the first end-to-end framework that enables VLMs to perform 3D geometric perception in autonomous driving scenarios. VLM-3D incorporates Low-Rank Adaptation (LoRA) to efficiently adapt VLMs to driving tasks with minimal computational overhead, and introduces a joint semantic-geometric loss design: token-level semantic loss is applied during early training to ensure stable convergence, while 3D IoU loss is introduced in later stages to refine the accuracy of 3D bounding box predictions. Evaluations on the nuScenes dataset demonstrate that the proposed joint semantic-geometric loss in VLM-3D leads to a 12.8% improvement in perception accuracy, fully validating the effectiveness and advancement of our method.

Paperid: 6704, https://arxiv.org/pdf/2508.08508.pdf

Abstract:
Current Vision Language Models (VLMs) demonstrate a critical gap between surface-level recognition and deep narrative reasoning when processing sequential visual storytelling. Through a comprehensive investigation of manga narrative understanding, we reveal that while recent large multimodal models excel at individual panel interpretation, they systematically fail at temporal causality and cross-panel cohesion, core requirements for coherent story comprehension. We introduce a novel evaluation framework that combines fine-grained multimodal annotation, cross-modal embedding analysis, and retrieval-augmented assessment to systematically characterize these limitations. Our methodology includes (i) a rigorous annotation protocol linking visual elements to narrative structure through aligned light novel text, (ii) comprehensive evaluation across multiple reasoning paradigms, including direct inference and retrieval-augmented generation, and (iii) cross-modal similarity analysis revealing fundamental misalignments in current VLMs' joint representations. Applying this framework to Re:Zero manga across 11 chapters with 308 annotated panels, we conduct the first systematic study of long-form narrative understanding in VLMs through three core evaluation axes: generative storytelling, contextual dialogue grounding, and temporal reasoning. Our findings demonstrate that current models lack genuine story-level intelligence, struggling particularly with non-linear narratives, character consistency, and causal inference across extended sequences. This work establishes both the foundation and practical methodology for evaluating narrative intelligence, while providing actionable insights into the capability of deep sequential understanding of Discrete Visual Narratives beyond basic recognition in Multimodal Models. Project Page: https://re-verse.vercel.app

Paperid: 6705, https://arxiv.org/pdf/2508.08115.pdf

Abstract:
We present TeamMedAgents, a modular multi-agent framework that systematically translates evidence-based teamwork principles from organizational psychology into large language model collaboration for medical decision-making. Building upon Salas et al.'s "Big Five" teamwork model, we operationalize five core components as independently configurable mechanisms: shared mental models, team leadership, team orientation, trust networks, and mutual monitoring. Our architecture dynamically recruits 2-4 specialist agents and employs structured four-phase deliberation with adaptive component selection. Evaluation across eight medical benchmarks encompassing 11,545 questions demonstrates TeamMedAgents achieves 77.63% overall accuracy (text-based: 81.30%, vision-language: 66.60%). Systematic ablation studies comparing three single-agent baselines (Zero-Shot, Few-Shot, CoT) against individual teamwork components reveal task-specific optimization patterns: shared mental models excel on knowledge tasks, trust mechanisms improve differential diagnosis, while comprehensive integration degrades performance. Adaptive component selection yields 2-10 percentage point improvements over strongest baselines, with 96.2% agent convergence validating structured coordination effectiveness. TeamMedAgents establishes principled methodology for translating human teamwork theory into multi-agent systems, demonstrating that evidence-based collaboration patterns enhance AI performance in safety-critical domains through modular component design and selective activation strategies.

Paperid: 6706, https://arxiv.org/pdf/2508.05976.pdf

Abstract:
The fragmentation between high-level task semantics and low-level geometric features remains a persistent challenge in robotic manipulation. While vision-language models (VLMs) have shown promise in generating affordance-aware visual representations, the lack of semantic grounding in canonical spaces and reliance on manual annotations severely limit their ability to capture dynamic semantic-affordance relationships. To address these, we propose Primitive-Aware Semantic Grounding (PASG), a closed-loop framework that introduces: (1) Automatic primitive extraction through geometric feature aggregation, enabling cross-category detection of keypoints and axes; (2) VLM-driven semantic anchoring that dynamically couples geometric primitives with functional affordances and task-relevant description; (3) A spatial-semantic reasoning benchmark and a fine-tuned VLM (Qwen2.5VL-PA). We demonstrate PASG's effectiveness in practical robotic manipulation tasks across diverse scenarios, achieving performance comparable to manual annotations. PASG achieves a finer-grained semantic-affordance understanding of objects, establishing a unified paradigm for bridging geometric primitives with task semantics in robotic manipulation.

Paperid: 6707, https://arxiv.org/pdf/2508.05409.pdf

Abstract:
Biometric systems, such as face recognition systems powered by deep neural networks (DNNs), rely on large and highly sensitive datasets. Backdoor attacks can subvert these systems by manipulating the training process. By inserting a small trigger, such as a sticker, make-up, or patterned mask, into a few training images, an adversary can later present the same trigger during authentication to be falsely recognized as another individual, thereby gaining unauthorized access. Existing defense mechanisms against backdoor attacks still face challenges in precisely identifying and mitigating poisoned images without compromising data utility, which undermines the overall reliability of the system. We propose a novel and generalizable approach, TrueBiometric: Trustworthy Biometrics, which accurately detects poisoned images using a majority voting mechanism leveraging multiple state-of-the-art large vision language models. Once identified, poisoned samples are corrected using targeted and calibrated corrective noise. Our extensive empirical results demonstrate that TrueBiometric detects and corrects poisoned images with 100\% accuracy without compromising accuracy on clean images. Compared to existing state-of-the-art approaches, TrueBiometric offers a more practical, accurate, and effective solution for mitigating backdoor attacks in face recognition systems.

Paperid: 6708, https://arxiv.org/pdf/2508.04469.pdf

Abstract:
The deployment of vision-language models remains constrained by substantial computational requirements. We present \textbf{FrEVL}, a framework exploring whether frozen pretrained embeddings can support effective vision-language understanding. Our analysis reveals that frozen embeddings contain rich information for discriminative tasks, achieving 85\% to 95\% of state-of-the-art performance on standard benchmarks with only 68.4M trainable parameters. This performance dichotomy reveals a critical insight: frozen embedding effectiveness depends on alignment between pretraining objectives and downstream task requirements. When accounting for end-to-end computation including embedding extraction, FrEVL provides $2.3\times$ speedup with 52\% lower energy consumption, making it suitable for scenarios with pre-computable inputs or when deployment constraints outweigh marginal performance gains. Our evaluation provides practitioners with guidance on when frozen embedding approaches represent viable alternatives to full model deployment. We will release our complete implementation and evaluation framework to facilitate further research into efficient multi-modal understanding.

Paperid: 6709, https://arxiv.org/pdf/2508.04201.pdf

Abstract:
During reasoning in vision-language models (VLMs), false positive (FP) reasoning occurs when a model produces the correct answer but follows an incorrect reasoning path, resulting in undermined reasoning reliability. Existing approaches mainly rely on prompt engineering, knowledge distillation or reinforcement learning to improve reasoning reliability, both of which require large amounts of high-quality data and thus limit practical applicability. Few approaches have focused on directly detecting and correcting FPs. To address these issues, we propose ViFP, a framework for Visual False Positive Detection to Enhance Reasoning Reliability in VLMs. ViFP builds effective reasoning paths through multi-turn QA and dynamically analyzes the consistency of the reasoning path to identify potential FPs. It also introduces a targeted reasoning chain correction mechanism to modify FP reasoning, thereby improving logical consistency and accuracy. Finally, we introduce a reliability evaluation metric, VoC, which integrates answer accuracy and the FP rate, providing a quantitative tool to assess whether a VLM not only answers correctly but also reasons reliably. Our experiments on closed-source VLMs show that ViFP consistently improves performance across three datasets: A-OKVQA, OK-VQA, and FVQA. On A-OKVQA, ViFP improves accuracy by up to 5.4%, surpassing the previous state-of-the-art by 4.3%, and significantly reduces the number of FPs, validating its benefits in enhancing reasoning reliability.

Paperid: 6710, https://arxiv.org/pdf/2508.02419.pdf

Abstract:
Large vision-language models (LVLMs) have demonstrated remarkable multimodal comprehension and reasoning capabilities, but they still suffer from severe object hallucination. Previous studies primarily attribute the flaw to linguistic prior caused by the scale mismatch between visual encoders and large language models (LLMs) in LVLMs. Specifically, as current LVLMs are built upon LLMs, they tend to over-rely on textual prompts and internal knowledge of LLMs, generating descriptions inconsistent with visual cues. However, through an in-depth investigation of the hallucinated mechanisms, we empirically reveal a previously overlooked phenomenon: LVLMs may ignore not only visual information but also textual modality during hallucination, a behavior termed as modality bias, which indicates that LVLMs struggle to simultaneously attend to both visual and textual modalities, leading to fragmented understanding of user-provided instructions. Based on this observation, we propose a simple yet effective training-free method to mitigate object hallucination. Concretely, we intervene and adjust the attention weights of textual and visual tokens, balancing cross-modal compatibility for better alignment with user intentions. Furthermore, we adopt a contrastive decoding strategy to reduce the LVLM's overreliance on its parametric knowledge, synergistically enhancing our attention manipulation. Extensive experiments confirm the widespread presence of modality bias in LVLMs. Notably, our method effectively mitigates hallucination across multiple open-source LVLMs and benchmarks, highlighting its generalizability and efficacy.

Paperid: 6711, https://arxiv.org/pdf/2508.01546.pdf

Abstract:
Vision-Language Models (VLMs) have enabled substantial progress in video understanding by leveraging cross-modal reasoning capabilities. However, their effectiveness is limited by the restricted context window and the high computational cost required to process long videos with thousands of frames. Retrieval-augmented generation (RAG) addresses this challenge by selecting only the most relevant frames as input, thereby reducing the computational burden. Nevertheless, existing video RAG methods struggle to balance retrieval efficiency and accuracy, particularly when handling diverse and complex video content. To address these limitations, we propose E-VRAG, a novel and efficient video RAG framework for video understanding. We first apply a frame pre-filtering method based on hierarchical query decomposition to eliminate irrelevant frames, reducing computational costs at the data level. We then employ a lightweight VLM for frame scoring, further reducing computational costs at the model level. Additionally, we propose a frame retrieval strategy that leverages the global statistical distribution of inter-frame scores to mitigate the potential performance degradation from using a lightweight VLM. Finally, we introduce a multi-view question answering scheme for the retrieved frames, enhancing the VLM's capability to extract and comprehend information from long video contexts. Experiments on four public benchmarks show that E-VRAG achieves about 70% reduction in computational cost and higher accuracy compared to baseline methods, all without additional training. These results demonstrate the effectiveness of E-VRAG in improving both efficiency and accuracy for video RAG tasks.

Paperid: 6712, https://arxiv.org/pdf/2508.01540.pdf

Abstract:
Vision-Language Models (VLMs) have achieved remarkable breakthroughs in recent years, enabling a diverse array of applications in everyday life. However, the substantial computational and storage demands of VLMs pose significant challenges for their efficient deployment on mobile devices, which represent the most ubiquitous and accessible computing platforms today. In this work, we introduce MagicVL-2B, a novel VLM meticulously optimized for flagship smartphones. MagicVL-2B leverages a lightweight visual encoder with fewer than 100M parameters and features a redesigned dynamic resolution scheme that adaptively generates image tokens without excessive modification of image dimensions. To further enhance the performance of this compact encoder within VLMs, we propose a multimodal curriculum learning strategy that incrementally increases task difficulty and data information density throughout training. This approach substantially improves the model's performance across a variety of sub-tasks. Extensive evaluations on standard VLM benchmarks demonstrate that MagicVL-2B matches the accuracy of current state-of-the-art models while reducing on-device power consumption by 41.1%. These results establish MagicVL-2B as a practical and robust solution for real-world mobile vision-language applications, enabling advanced multimodal intelligence to run directly on smartphones.

Paperid: 6713, https://arxiv.org/pdf/2508.00260.pdf

Abstract:
Continual learning enables pre-trained generative vision-language models (VLMs) to incorporate knowledge from new tasks without retraining data from previous ones. Recent methods update a visual projector to translate visual information for new tasks, connecting pre-trained vision encoders with large language models. However, such adjustments may cause the models to prioritize visual inputs over language instructions, particularly learning tasks with repetitive types of textual instructions. To address the neglect of language instructions, we propose a novel framework that grounds the translation of visual information on instructions for language models. We introduce a mixture of visual projectors, each serving as a specialized visual-to-language translation expert based on the given instruction context to adapt to new tasks. To avoid using experts for irrelevant instruction contexts, we propose an expert recommendation strategy that reuses experts for tasks similar to those previously learned. Additionally, we introduce expert pruning to alleviate interference from the use of experts that cumulatively activated in previous tasks. Extensive experiments on diverse vision-language tasks demonstrate that our method outperforms existing continual learning approaches by generating instruction-following responses.

Paperid: 6714, https://arxiv.org/pdf/2507.22074.pdf

Abstract:
The rapid advancement of Large Language Models (LLMs) and Large Vision-Language Models (LVLMs) has enhanced our ability to process and generate human language and visual information. However, these models often struggle with complex, multi-step multi-modal instructions that require logical reasoning, dynamic feedback integration, and iterative self-correction. To address this, we propose CIMR: Contextualized Iterative Multimodal Reasoning, a novel framework that introduces a context-aware iterative reasoning and self-correction module. CIMR operates in two stages: initial reasoning and response generation, followed by iterative refinement using parsed multi-modal feedback. A dynamic fusion module deeply integrates textual, visual, and contextual features at each step. We fine-tune LLaVA-1.5-7B on the Visual Instruction Tuning (VIT) dataset and evaluate CIMR on the newly introduced Multi-modal Action Planning (MAP) dataset. CIMR achieves 91.5% accuracy, outperforming state-of-the-art models such as GPT-4V (89.2%), LLaVA-1.5 (78.5%), MiniGPT-4 (75.3%), and InstructBLIP (72.8%), demonstrating the efficacy of its iterative reasoning and self-correction capabilities in complex tasks.

Paperid: 6715, https://arxiv.org/pdf/2507.21521.pdf

Abstract:
Active Learning (AL) has emerged as a powerful approach for minimizing labeling costs by selectively sampling the most informative data for neural network model development. Effective AL for large-scale vision-language models necessitates addressing challenges in uncertainty estimation and efficient sampling given the vast number of parameters involved. In this work, we introduce a novel parameter-efficient learning methodology that incorporates uncertainty calibration loss within the AL framework. We propose a differentiable loss function that promotes uncertainty calibration for effectively selecting fewer and most informative data samples for fine-tuning. Through extensive experiments across several datasets and vision backbones, we demonstrate that our solution can match and exceed the performance of complex feature-based sampling techniques while being computationally very efficient. Additionally, we investigate the efficacy of Prompt learning versus Low-rank adaptation (LoRA) in sample selection, providing a detailed comparative analysis of these methods in the context of efficient AL.

Paperid: 6716, https://arxiv.org/pdf/2507.20411.pdf

Abstract:
Multilingual vision-language models have made significant strides in image captioning, yet they still lag behind their English counterparts due to limited multilingual training data and costly large-scale model parameterization. Retrieval-augmented generation (RAG) offers a promising alternative by conditioning caption generation on retrieved examples in the target language, reducing the need for extensive multilingual training. However, multilingual RAG captioning models often depend on retrieved captions translated from English, which can introduce mismatches and linguistic biases relative to the source language. We introduce CONCAP, a multilingual image captioning model that integrates retrieved captions with image-specific concepts, enhancing the contextualization of the input image and grounding the captioning process across different languages. Experiments on the XM3600 dataset indicate that CONCAP enables strong performance on low- and mid-resource languages, with highly reduced data requirements. Our findings highlight the effectiveness of concept-aware retrieval augmentation in bridging multilingual performance gaps.

Paperid: 6717, https://arxiv.org/pdf/2507.18022.pdf

Abstract:
Charts and graphs help people analyze data, but can they also be useful to AI systems? To investigate this question, we perform a series of experiments with two commercial vision-language models: GPT 4.1 and Claude 3.5. Across three representative analysis tasks, the two systems describe synthetic datasets more precisely and accurately when raw data is accompanied by a scatterplot, especially as datasets grow in complexity. Comparison with two baselines -- providing a blank chart and a chart with mismatched data -- shows that the improved performance is due to the content of the charts. Our results are initial evidence that AI systems, like humans, can benefit from visualization.

Paperid: 6718, https://arxiv.org/pdf/2507.17335.pdf

Abstract:
License plate recognition in open environments is widely applicable across various domains; however, the diversity of license plate types and imaging conditions presents significant challenges. To address the limitations encountered by CNN and CRNN-based approaches in license plate recognition, this paper proposes a unified solution that integrates a lightweight visual encoder with a text decoder, within a pre-training framework tailored for single and double-line Chinese license plates. To mitigate the scarcity of double-line license plate datasets, we constructed a single/double-line license plate dataset by synthesizing images, applying texture mapping onto real scenes, and blending them with authentic license plate images. Furthermore, to enhance the system's recognition accuracy, we introduce a perspective correction network (PTN) that employs license plate corner coordinate regression as an implicit variable, supervised by license plate view classification information. This network offers improved stability, interpretability, and low annotation costs. The proposed algorithm achieves an average recognition accuracy of 99.34% on the corrected CCPD test set under coarse localization disturbance. When evaluated under fine localization disturbance, the accuracy further improves to 99.58%. On the double-line license plate test set, it achieves an average recognition accuracy of 98.70%, with processing speeds reaching up to 167 frames per second, indicating strong practical applicability.

Paperid: 6719, https://arxiv.org/pdf/2507.15285.pdf

Abstract:
Recent advances in biometric systems have significantly improved the detection and prevention of fraudulent activities. However, as detection methods improve, attack techniques become increasingly sophisticated. Attacks on face recognition systems can be broadly divided into physical and digital approaches. Traditionally, deep learning models have been the primary defence against such attacks. While these models perform exceptionally well in scenarios for which they have been trained, they often struggle to adapt to different types of attacks or varying environmental conditions. These subsystems require substantial amounts of training data to achieve reliable performance, yet biometric data collection faces significant challenges, including privacy concerns and the logistical difficulties of capturing diverse attack scenarios under controlled conditions. This work investigates the application of Vision Language Models (VLM) and proposes an in-context learning framework for detecting physical presentation attacks and digital morphing attacks in biometric systems. Focusing on open-source models, the first systematic framework for the quantitative evaluation of VLMs in security-critical scenarios through in-context learning techniques is established. The experimental evaluation conducted on freely available databases demonstrates that the proposed subsystem achieves competitive performance for physical and digital attack detection, outperforming some of the traditional CNNs without resource-intensive training. The experimental results validate the proposed framework as a promising tool for improving generalisation in attack detection.

Paperid: 6720, https://arxiv.org/pdf/2507.14449.pdf

Abstract:
Real-world infrared imagery presents unique challenges for vision-language models due to the scarcity of aligned text data and domain-specific characteristics. Although existing methods have advanced the field, their reliance on synthetic infrared images generated through style transfer from visible images, which limits their ability to capture the unique characteristics of the infrared modality. To address this, we propose IRGPT, the first multi-modal large language model for real-world infrared images, built upon a large-scale InfraRed-Text Dataset (IR-TD) comprising over 260K authentic image-text pairs. The proposed IR-TD dataset contains real infrared images paired with meticulously handcrafted texts, where the initial drafts originated from two complementary processes: (1) LLM-generated descriptions of visible images, and (2) rule-based descriptions of annotations. Furthermore, we introduce a bi-cross-modal curriculum transfer learning strategy that systematically transfers knowledge from visible to infrared domains by considering the difficulty scores of both infrared-visible and infrared-text. Evaluated on a benchmark of 9 tasks (e.g., recognition, grounding), IRGPT achieves state-of-the-art performance even compared with larger-scale models.

Paperid: 6721, https://arxiv.org/pdf/2507.14024.pdf

Abstract:
Bridging emotions and visual content for emotion-driven image editing holds great potential in creative industries, yet precise manipulation remains challenging due to the abstract nature of emotions and their varied manifestations across different contexts. We tackle this challenge with an integrated approach consisting of three complementary components. First, we introduce MoodArchive, an 8M+ image dataset with detailed hierarchical emotional annotations generated by LLaVA and partially validated by human evaluators. Second, we develop MoodifyCLIP, a vision-language model fine-tuned on MoodArchive to translate abstract emotions into specific visual attributes. Third, we propose Moodifier, a training-free editing model leveraging MoodifyCLIP and multimodal large language models (MLLMs) to enable precise emotional transformations while preserving content integrity. Our system works across diverse domains such as character expressions, fashion design, jewelry, and home dÃ©cor, enabling creators to quickly visualize emotional variations while preserving identity and structure. Extensive experimental evaluations show that Moodifier outperforms existing methods in both emotional accuracy and content preservation, providing contextually appropriate edits. By linking abstract emotions to concrete visual changes, our solution unlocks new possibilities for emotional content creation in real-world applications. We will release the MoodArchive dataset, MoodifyCLIP model, and make the Moodifier code and demo publicly available upon acceptance.

Paperid: 6722, https://arxiv.org/pdf/2507.13361.pdf

Abstract:
Visual Language Models (VLMs) excel at complex visual tasks such as VQA and chart understanding, yet recent work suggests they struggle with simple perceptual tests. We present an evaluation that tests vision-language models' capacity for nonlocal visual reasoning -- reasoning that requires chaining evidence collected from multiple, possibly distant, regions of an image. We isolate three distinct forms of non-local vision: comparative perception, which demands holding two images in working memory and comparing them; saccadic search, which requires making discrete, evidence-driven jumps to locate successive targets; and smooth visual search, which involves searching smoothly along a continuous contour. Flagship models (e.g., Gemini 2.5 Pro, Claude Vision 3.7, GPT-o4-mini), even those that perform well on prior primitive-vision benchmarks, fail these tests and barely exceed random accuracy on two variants of our tasks that are trivial for humans. Our structured evaluation suite allows us to test if VLMs can perform similar visual algorithms to humans. Our findings show that despite gains in raw visual acuity, current models lack core visual reasoning capabilities.

Paperid: 6723, https://arxiv.org/pdf/2507.13152.pdf

Abstract:
Recent advances in vision-language navigation (VLN) were mainly attributed to emerging large language models (LLMs). These methods exhibited excellent generalization capabilities in instruction understanding and task reasoning. However, they were constrained by the fixed knowledge bases and reasoning abilities of LLMs, preventing fully incorporating experiential knowledge and thus resulting in a lack of efficient evolutionary capacity. To address this, we drew inspiration from the evolution capabilities of natural agents, and proposed a self-evolving VLN framework (SE-VLN) to endow VLN agents with the ability to continuously evolve during testing. To the best of our knowledge, it was the first time that an multimodal LLM-powered self-evolving VLN framework was proposed. Specifically, SE-VLN comprised three core modules, i.e., a hierarchical memory module to transfer successful and failure cases into reusable knowledge, a retrieval-augmented thought-based reasoning module to retrieve experience and enable multi-step decision-making, and a reflection module to realize continual evolution. Comprehensive tests illustrated that the SE-VLN achieved navigation success rates of 57% and 35.2% in unseen environments, representing absolute performance improvements of 23.9% and 15.0% over current state-of-the-art methods on R2R and REVERSE datasets, respectively. Moreover, the SE-VLN showed performance improvement with increasing experience repository, elucidating its great potential as a self-evolving agent framework for VLN.

Paperid: 6724, https://arxiv.org/pdf/2507.09071.pdf

Abstract:
Large vision-language models (VLMs) enable the joint processing of text and images. However, the inclusion of vision data significantly expands the prompt length. Along with the quadratic complexity of the attention computation, this results in a longer prefill duration. An approach to mitigate this bottleneck is to leverage the inherent sparsity in the attention computation. In our analysis of attention patterns in VLMs, we observe that a substantial portion of layers exhibit minimal cross-image attention, except through attention-sink tokens per image. These sparse attention patterns fall into distinct categories: sink-only, document mask and a hybrid document-sink mask. Based on this, we propose BlindSight: a training-free approach to optimize VLM inference using a input template-aware attention sparsity mask. We utilize samples from a dataset to derive a prompt-agnostic sparsity categorization for every attention head. We evaluate the proposed technique using VLMs such as Qwen2-VL, Qwen2.5-VL and Gemma-3. BlindSight results in a 32%-41% reduction in FLOPs on average with -2%-+2% accuracy compared to the original model in most evaluated multi-image understanding benchmarks.

Paperid: 6725, https://arxiv.org/pdf/2507.08624.pdf

Abstract:
This paper introduces the Ambient Intelligence Rehabilitation Support (AIRS) framework, an advanced artificial intelligence-based solution tailored for home rehabilitation environments. AIRS integrates cutting-edge technologies, including Real-Time 3D Reconstruction (RT-3DR), intelligent navigation, and large Vision-Language Models (VLMs), to create a comprehensive system for machine-guided physical rehabilitation. The general AIRS framework is demonstrated in rehabilitation scenarios following total knee replacement (TKR), utilizing a database of 263 video recordings for evaluation. A smartphone is employed within AIRS to perform RT-3DR of living spaces and has a body-matched avatar to provide visual feedback about the excercise. This avatar is necessary in (a) optimizing exercise configurations, including camera placement, patient positioning, and initial poses, and (b) addressing privacy concerns and promoting compliance with the AI Act. The system guides users through the recording process to ensure the collection of properly recorded videos. AIRS employs two feedback mechanisms: (i) visual 3D feedback, enabling direct comparisons between prerecorded clinical exercises and patient home recordings and (ii) VLM-generated feedback, providing detailed explanations and corrections for exercise errors. The framework also supports people with visual and hearing impairments. It also features a modular design that can be adapted to broader rehabilitation contexts. AIRS software components are available for further use and customization.

Paperid: 6726, https://arxiv.org/pdf/2507.07551.pdf

Abstract:
The accelerating growth of photographic collections has outpaced manual cataloguing, motivating the use of vision language models (VLMs) to automate metadata generation. This study examines whether Al-generated catalogue descriptions can approximate human-written quality and how generative Al might integrate into cataloguing workflows in archival and museum collections. A VLM (InternVL2) generated catalogue descriptions for photographic prints on labelled cardboard mounts with archaeological content, evaluated by archive and archaeology experts and non-experts in a human-centered, experimental framework. Participants classified descriptions as AI-generated or expert-written, rated quality, and reported willingness to use and trust in AI tools. Classification performance was above chance level, with both groups underestimating their ability to detect Al-generated descriptions. OCR errors and hallucinations limited perceived quality, yet descriptions rated higher in accuracy and usefulness were harder to classify, suggesting that human review is necessary to ensure the accuracy and quality of catalogue descriptions generated by the out-of-the-box model, particularly in specialized domains like archaeological cataloguing. Experts showed lower willingness to adopt AI tools, emphasizing concerns on preservation responsibility over technical performance. These findings advocate for a collaborative approach where AI supports draft generation but remains subordinate to human verification, ensuring alignment with curatorial values (e.g., provenance, transparency). The successful integration of this approach depends not only on technical advancements, such as domain-specific fine-tuning, but even more on establishing trust among professionals, which could both be fostered through a transparent and explainable AI pipeline.

Paperid: 6727, https://arxiv.org/pdf/2507.04620.pdf

Abstract:
In Human-Robot Collaboration (HRC), which encompasses physical interaction and remote cooperation, accurate estimation of human intentions and seamless switching of collaboration modes to adjust robot behavior remain paramount challenges. To address these issues, we propose an Intent-Driven Adaptive Generalized Collaboration (IDAGC) framework that leverages multimodal data and human intent estimation to facilitate adaptive policy learning across multi-tasks in diverse scenarios, thereby facilitating autonomous inference of collaboration modes and dynamic adjustment of robotic actions. This framework overcomes the limitations of existing HRC methods, which are typically restricted to a single collaboration mode and lack the capacity to identify and transition between diverse states. Central to our framework is a predictive model that captures the interdependencies among vision, language, force, and robot state data to accurately recognize human intentions with a Conditional Variational Autoencoder (CVAE) and automatically switch collaboration modes. By employing dedicated encoders for each modality and integrating extracted features through a Transformer decoder, the framework efficiently learns multi-task policies, while force data optimizes compliance control and intent estimation accuracy during physical interactions. Experiments highlights our framework's practical potential to advance the comprehensive development of HRC.

Paperid: 6728, https://arxiv.org/pdf/2507.04509.pdf

Abstract:
Camera relocalization, a cornerstone capability of modern computer vision, accurately determines a camera's position and orientation (6-DoF) from images and is essential for applications in augmented reality (AR), mixed reality (MR), autonomous driving, delivery drones, and robotic navigation. Unlike traditional deep learning-based methods that regress camera pose from images in a single scene, which often lack generalization and robustness in diverse environments, we propose MVL-Loc, a novel end-to-end multi-scene 6-DoF camera relocalization framework. MVL-Loc leverages pretrained world knowledge from vision-language models (VLMs) and incorporates multimodal data to generalize across both indoor and outdoor settings. Furthermore, natural language is employed as a directive tool to guide the multi-scene learning process, facilitating semantic understanding of complex scenes and capturing spatial relationships among objects. Extensive experiments on the 7Scenes and Cambridge Landmarks datasets demonstrate MVL-Loc's robustness and state-of-the-art performance in real-world multi-scene camera relocalization, with improved accuracy in both positional and orientational estimates.

Paperid: 6729, https://arxiv.org/pdf/2507.04152.pdf

Abstract:
The burgeoning field of generative artificial intelligence has fundamentally reshaped our approach to content creation, with Large Vision-Language Models (LVLMs) standing at its forefront. While current LVLMs have demonstrated impressive capabilities in text-to-image generation, they often falter when confronted with complex textual descriptions demanding precise compositional understanding and visual planning. This limitation particularly impacts the accurate rendering of multiple objects, their attributes, spatial relationships, and specific poses within intricate scenes, as evidenced by benchmarks like LongBench-T2I. To address these challenges, we introduce LVLM-Composer, a novel 10-billion parameter scale LVLM specifically engineered for enhanced compositional image synthesis. Our method incorporates a Hierarchical Semantic Planning Module for structured prompt decomposition and a Fine-Grained Feature Alignment Mechanism for precise visual guidance during generation. We propose a multi-stage training paradigm, featuring Hierarchical Semantic-Visual Grounding Pre-training and Compositional Planning Reinforcement Learning with Self-Correction, to instill robust compositional reasoning. Extensive experiments on the LongBench-T2I benchmark, utilizing automatic evaluation by Gemini-2.0-Flash and InternVL3-78B, demonstrate LVLM-Composer's superior performance across critical compositional dimensions including object accuracy, composition fidelity, and pose accuracy, significantly outperforming state-of-the-art baselines. An in-depth ablation study further validates the indispensable contribution of our proposed modules, while human evaluations confirm the perceptual superiority of our generated images. LVLM-Composer represents a significant step towards truly controllable and compositionally accurate open-ended text-to-image generation.

Paperid: 6730, https://arxiv.org/pdf/2507.04151.pdf

Abstract:
This paper introduces Hierarchical Self-Supervised LVLM (Hi-SSLVLM), a novel generative model designed to significantly advance text-to-image synthesis, particularly for complex and compositionally challenging prompts. Traditional methods often grapple with the high cost of meticulously curated paired image-text datasets and struggle with precise control over fine-grained visual attributes and intricate spatial relationships. Our Hi-SSLVLM addresses these limitations through a unique two-stage self-supervised learning strategy. The first stage, Multi-Granularity Visual-Language Grounding, enables the Large Vision-Language Model (LVLM) backbone to autonomously generate and align hierarchical captions (global and local) to images, cultivating a deep internal semantic understanding without reliance on extensive human annotation. The second stage, Self-Refinement and Guided Image Generation, leverages this acquired knowledge by an Internal Compositional Planning (ICP) mechanism, where the LVLM first formulates detailed textual sub-prompts to guide the image generation process, complemented by a novel Semantic Consistency Loss for precise output alignment. Comprehensive experiments against leading baselines, including Janus-Pro-1B, Stable Diffusion XL 1.0, DeepFloyd IF v1.0, and ControlNet-XL, on multi-dimensional benchmarks such as Gemini-2.0-Flash and InternVL3-78B, demonstrate Hi-SSLVLM's superior performance across all fine-grained metrics. An in-depth ablation study confirms the critical role of each proposed component. Furthermore, human evaluations corroborate our quantitative findings, highlighting Hi-SSLVLM's enhanced fidelity to prompt, compositional accuracy, and overall aesthetic quality, marking a significant step towards more controllable and semantically consistent open-ended text-to-image generation.

Paperid: 6731, https://arxiv.org/pdf/2507.03585.pdf

Abstract:
The clinical utility of deep learning models for medical image segmentation is severely constrained by their inability to generalize to unseen domains. This failure is often rooted in the models learning spurious correlations between anatomical content and domain-specific imaging styles. To overcome this fundamental challenge, we introduce Causal-SAM-LLM, a novel framework that elevates Large Language Models (LLMs) to the role of causal reasoners. Our framework, built upon a frozen Segment Anything Model (SAM) encoder, incorporates two synergistic innovations. First, Linguistic Adversarial Disentanglement (LAD) employs a Vision-Language Model to generate rich, textual descriptions of confounding image styles. By training the segmentation model's features to be contrastively dissimilar to these style descriptions, it learns a representation robustly purged of non-causal information. Second, Test-Time Causal Intervention (TCI) provides an interactive mechanism where an LLM interprets a clinician's natural language command to modulate the segmentation decoder's features in real-time, enabling targeted error correction. We conduct an extensive empirical evaluation on a composite benchmark from four public datasets (BTCV, CHAOS, AMOS, BraTS), assessing generalization under cross-scanner, cross-modality, and cross-anatomy settings. Causal-SAM-LLM establishes a new state of the art in out-of-distribution (OOD) robustness, improving the average Dice score by up to 6.2 points and reducing the Hausdorff Distance by 15.8 mm over the strongest baseline, all while using less than 9% of the full model's trainable parameters. Our work charts a new course for building robust, efficient, and interactively controllable medical AI systems.

Paperid: 6732, https://arxiv.org/pdf/2507.00209.pdf

Abstract:
High-resolution imaging is crucial for enhancing visual clarity and enabling precise computer-assisted guidance in minimally invasive surgery (MIS). Despite the increasing adoption of 4K endoscopic systems, there remains a significant gap in publicly available native 4K datasets tailored specifically for robotic-assisted MIS. We introduce SurgiSR4K, the first publicly accessible surgical imaging and video dataset captured at a native 4K resolution, representing realistic conditions of robotic-assisted procedures. SurgiSR4K comprises diverse visual scenarios including specular reflections, tool occlusions, bleeding, and soft tissue deformations, meticulously designed to reflect common challenges faced during laparoscopic and robotic surgeries. This dataset opens up possibilities for a broad range of computer vision tasks that might benefit from high resolution data, such as super resolution (SR), smoke removal, surgical instrument detection, 3D tissue reconstruction, monocular depth estimation, instance segmentation, novel view synthesis, and vision-language model (VLM) development. SurgiSR4K provides a robust foundation for advancing research in high-resolution surgical imaging and fosters the development of intelligent imaging technologies aimed at enhancing performance, safety, and usability in image-guided robotic surgeries.

Paperid: 6733, https://arxiv.org/pdf/2506.17639.pdf

Abstract:
Vision-Language-Action models (VLA) have demonstrated remarkable capabilities and promising potential in solving complex robotic manipulation tasks. However, their substantial parameter sizes and high inference latency pose significant challenges for real-world deployment, particularly on resource-constrained robotic platforms. To address this issue, we begin by conducting an extensive empirical study to explore the effectiveness of model compression techniques when applied to VLAs. Building on the insights gained from these preliminary experiments, we propose RLRC, a three-stage recovery method for compressed VLAs, including structured pruning, performance recovery based on SFT and RL, and further quantization. RLRC achieves up to an 8x reduction in memory usage and a 2.3x improvement in inference throughput, while maintaining or even surpassing the original VLA's task success rate. Extensive experiments show that RLRC consistently outperforms existing compression baselines, demonstrating strong potential for on-device deployment of VLAs. Project website: https://rlrc-vla.github.io

Paperid: 6734, https://arxiv.org/pdf/2506.16396.pdf

Abstract:
Natural language can offer a concise and human-interpretable means of specifying reinforcement learning (RL) tasks. The ability to extract rewards from a language instruction can enable the development of robotic systems that can learn from human guidance; however, it remains a challenging problem, especially in visual environments. Existing approaches that employ large, pretrained language models either rely on non-visual environment representations, require prohibitively large amounts of feedback, or generate noisy, ill-shaped reward functions. In this paper, we propose a novel method, $\textbf{GoalLadder}$, that leverages vision-language models (VLMs) to train RL agents from a single language instruction in visual environments. GoalLadder works by incrementally discovering states that bring the agent closer to completing a task specified in natural language. To do so, it queries a VLM to identify states that represent an improvement in agent's task progress and to rank them using pairwise comparisons. Unlike prior work, GoalLadder does not trust VLM's feedback completely; instead, it uses it to rank potential goal states using an ELO-based rating system, thus reducing the detrimental effects of noisy VLM feedback. Over the course of training, the agent is tasked with minimising the distance to the top-ranked goal in a learned embedding space, which is trained on unlabelled visual data. This key feature allows us to bypass the need for abundant and accurate feedback typically required to train a well-shaped reward function. We demonstrate that GoalLadder outperforms existing related methods on classic control and robotic manipulation environments with the average final success rate of $\sim$95% compared to only $\sim$45% of the best competitor.

Paperid: 6735, https://arxiv.org/pdf/2506.15755.pdf

Abstract:
Vision-Language Models (VLMs) have demonstrated great potential in real-world applications. While existing research primarily focuses on improving their accuracy, the efficiency remains underexplored. Given the real-time demands of many applications and the high inference overhead of VLMs, efficiency robustness is a critical issue. However, previous studies evaluate efficiency robustness under unrealistic assumptions, requiring access to the model architecture and parameters -- an impractical scenario in ML-as-a-service settings, where VLMs are deployed via inference APIs. To address this gap, we propose VLMInferSlow, a novel approach for evaluating VLM efficiency robustness in a realistic black-box setting. VLMInferSlow incorporates fine-grained efficiency modeling tailored to VLM inference and leverages zero-order optimization to search for adversarial examples. Experimental results show that VLMInferSlow generates adversarial images with imperceptible perturbations, increasing the computational cost by up to 128.47%. We hope this research raises the community's awareness about the efficiency robustness of VLMs.

Paperid: 6736, https://arxiv.org/pdf/2506.14821.pdf

Abstract:
Despite tremendous recent advances in large model reasoning ability, vision-language models (VLMs) still struggle with detailed visual reasoning, especially when compute resources are limited. To address this challenge, we draw inspiration from methods like Deepseek-r1 for VLMs and train smaller-scale models with Group Relative Policy Optimization (GRPO) to use external tools such as zoom. The greatest benefit is obtained with a combination of GRPO learning, a simple reward structure, a simplified tool-calling interface, allocating additional tokens to the result of the tool call, and a training data mix that over-represents visually difficult examples. Compared to similarly-sized baseline models, our method achieves better performance on some visual question-answering (VQA) tasks, thanks to the detailed visual information gathered from the external tool.

Paperid: 6737, https://arxiv.org/pdf/2506.14276.pdf

Abstract:
The Abstraction and Reasoning Corpus (ARC-AGI) presents a formidable challenge for AI systems. Despite the typically low performance on ARC, the deep learning paradigm remains the most effective known strategy for generating skillful (state-of-the-art) neural networks (NN) across varied modalities and tasks in vision, language etc. The deep learning paradigm has proven to be able to train these skillful neural networks and learn the abstractions needed in these diverse domains. Our work doubles down on that and continues to leverage this paradigm by incorporating on-the-fly NN training at test time. We demonstrate that fully committing to deep learning's capacity to acquire novel abstractions yields state-of-the-art performance on ARC. Specifically, we treat both the neural network and the optimizer (rather than just a pre-trained network) as integral components of the inference process, fostering generalization to unseen tasks. Concretely, we propose a methodology for training on ARC, starting from pretrained LLMs, and enhancing their ARC reasoning. We also propose Test-Time Fine-Tuning (TTFT) and the Augment Inference Reverse-Augmentation and Vote (AIRV) as effective test-time techniques. We are the first to propose and show deep learning can be used effectively for ARC, showing boosts of up to 260% in accuracy with AIRV and a further 300% boost with TTFT. An early version of this approach secured first place in the 2023 ARCathon competition, while the final version achieved the current best score on the ARC private test-set (58%). Our findings highlight the key ingredients of a robust reasoning system in unfamiliar domains, underscoring the central mechanisms that improve broad perceptual reasoning.

Paperid: 6738, https://arxiv.org/pdf/2506.14015.pdf

Abstract:
We consider the problem of disentangling 3D from large vision-language models, which we show on generative 3D portraits. This allows free-form text control of appearance attributes like age, hair style, and glasses, and 3D geometry control of face expression and camera pose. In this setting, we assume we use a pre-trained large vision-language model (LVLM; CLIP) to generate from a smaller 2D dataset with no additional paired labels and with a pre-defined 3D morphable model (FLAME). First, we disentangle using canonicalization to a 2D reference frame from a deformable neural 3D triplane representation. But another form of entanglement arises from the significant noise in the LVLM's embedding space that describes irrelevant features. This damages output quality and diversity, but we overcome this with a Jacobian regularization that can be computed efficiently with a stochastic approximator. Compared to existing methods, our approach produces portraits with added text and 3D control, where portraits remain consistent when either control is changed. Broadly, this approach lets creators control 3D generators on their own 2D face data without needing resources to label large data or train large models.

Paperid: 6739, https://arxiv.org/pdf/2506.13761.pdf

Abstract:
Recent advancements in open-world robot manipulation have been largely driven by vision-language models (VLMs). While these models exhibit strong generalization ability in high-level planning, they struggle to predict low-level robot controls due to limited physical-world understanding. To address this issue, we propose a model predictive control framework for open-world manipulation that combines the semantic reasoning capabilities of VLMs with physically-grounded, interactive digital twins of the real-world environments. By constructing and simulating the digital twins, our approach generates feasible motion trajectories, simulates corresponding outcomes, and prompts the VLM with future observations to evaluate and select the most suitable outcome based on language instructions of the task. To further enhance the capability of pre-trained VLMs in understanding complex scenes for robotic control, we leverage the flexible rendering capabilities of the digital twin to synthesize the scene at various novel, unoccluded viewpoints. We validate our approach on a diverse set of complex manipulation tasks, demonstrating superior performance compared to baseline methods for language-conditioned robotic control using VLMs.

Paperid: 6740, https://arxiv.org/pdf/2506.13723.pdf

Abstract:
Transductive zero-shot learning (ZSL) aims to classify unseen categories by leveraging both semantic class descriptions and the distribution of unlabeled test data. While Vision-Language Models (VLMs) such as CLIP excel at aligning visual inputs with textual semantics, they often rely too heavily on class-level priors and fail to capture fine-grained visual cues. In contrast, Vision-only Foundation Models (VFMs) like DINOv2 provide rich perceptual features but lack semantic alignment. To exploit the complementary strengths of these models, we propose OTFusion, a simple yet effective training-free framework that bridges VLMs and VFMs via Optimal Transport. Specifically, OTFusion aims to learn a shared probabilistic representation that aligns visual and semantic information by minimizing the transport cost between their respective distributions. This unified distribution enables coherent class predictions that are both semantically meaningful and visually grounded. Extensive experiments on 11 benchmark datasets demonstrate that OTFusion consistently outperforms the original CLIP model, achieving an average accuracy improvement of nearly $10\%$, all without any fine-tuning or additional annotations. The code will be publicly released after the paper is accepted.

Paperid: 6741, https://arxiv.org/pdf/2506.13458.pdf

Abstract:
Recycling steel scrap can reduce carbon dioxide (CO2) emissions from the steel industry. However, a significant challenge in steel scrap recycling is the inclusion of impurities other than steel. To address this issue, we propose vision-language-model-based anomaly detection where a model is finetuned in a supervised manner, enabling it to handle niche objects effectively. This model enables automated detection of anomalies at a fine-grained level within steel scrap. Specifically, we finetune the image encoder, equipped with multi-scale mechanism and text prompts aligned with both normal and anomaly images. The finetuning process trains these modules using a multiclass classification as the supervision.

Paperid: 6743, https://arxiv.org/pdf/2506.11604.pdf

Abstract:
This paper introduces a novel benchmark dataset designed to evaluate the capabilities of Vision Language Models (VLMs) on tasks that combine visual reasoning with subject-specific background knowledge in the German language. In contrast to widely used English-language benchmarks that often rely on artificially difficult or decontextualized problems, this dataset draws from real middle school curricula across nine domains including mathematics, history, biology, and religion. The benchmark includes over 2,000 open-ended questions grounded in 486 images, ensuring that models must integrate visual interpretation with factual reasoning rather than rely on superficial textual cues. We evaluate thirteen state-of-the-art open-weight VLMs across multiple dimensions, including domain-specific accuracy and performance on adversarial crafted questions. Our findings reveal that even the strongest models achieve less than 45% overall accuracy, with particularly poor performance in music, mathematics, and adversarial settings. Furthermore, the results indicate significant discrepancies between success on popular benchmarks and real-world multimodal understanding. We conclude that middle school-level tasks offer a meaningful and underutilized avenue for stress-testing VLMs, especially in non-English contexts. The dataset and evaluation protocol serve as a rigorous testbed to better understand and improve the visual and linguistic reasoning capabilities of future AI systems.

Paperid: 6744, https://arxiv.org/pdf/2506.07416.pdf

Abstract:
This paper introduces an efficient Vision-Language Model (VLM) pipeline specifically optimized for deployment on embedded devices, such as those used in robotics and autonomous driving. The pipeline significantly reduces the computational overhead by jointly leveraging patch selection to filter irrelevant camera views, a token selection module to reduce input sequence length for the LLM, and speculative decoding to accelerate token generation. Evaluation on the NVIDIA DRIVE Thor platform for automonous driving application, our pipeline achieves $2.5\times$ end-to-end latency reduction without compromising task accuracy. The speed-up further increases to $3.2\times$ when applying FP8 post-training quantization. These results demonstrate our pipeline as a viable solution for enabling real-time VLM deployment in resource-constrained environments.

Paperid: 6745, https://arxiv.org/pdf/2506.04929.pdf

Abstract:
Neural Machine Translation (NMT) has improved translation by using Transformer-based models, but it still struggles with word ambiguity and context. This problem is especially important in domain-specific applications, which often have problems with unclear sentences or poor data quality. Our research explores how adding information to models can improve translations in the context of e-commerce data. To this end we create ConECT -- a new Czech-to-Polish e-commerce product translation dataset coupled with images and product metadata consisting of 11,400 sentence pairs. We then investigate and compare different methods that are applicable to context-aware translation. We test a vision-language model (VLM), finding that visual context aids translation quality. Additionally, we explore the incorporation of contextual information into text-to-text models, such as the product's category path or image descriptions. The results of our study demonstrate that the incorporation of contextual information leads to an improvement in the quality of machine translation. We make the new dataset publicly available.

Paperid: 6746, https://arxiv.org/pdf/2506.03371.pdf

Abstract:
Recent advances in vision-language models (VLMs) have enabled accurate image-based geolocation, raising serious concerns about location privacy risks in everyday social media posts. However, current benchmarks remain coarse-grained, linguistically biased, and lack multimodal and privacy-aware evaluations. To address these gaps, we present KoreaGEO Bench, the first fine-grained, multimodal geolocation benchmark for Korean street views. Our dataset comprises 1,080 high-resolution images sampled across four urban clusters and nine place types, enriched with multi-contextual annotations and two styles of Korean captions simulating real-world privacy exposure. We introduce a three-path evaluation protocol to assess ten mainstream VLMs under varying input modalities and analyze their accuracy, spatial bias, and reasoning behavior. Results reveal modality-driven shifts in localization precision and highlight structural prediction biases toward core cities.

Paperid: 6747, https://arxiv.org/pdf/2506.01118.pdf

Abstract:
The escalating demand for medical image interpretation underscores the critical need for advanced artificial intelligence solutions to enhance the efficiency and accuracy of radiological diagnoses. This paper introduces CXR-PathFinder, a novel Large Language Model (LLM)-centric foundation model specifically engineered for automated chest X-ray (CXR) report generation. We propose a unique training paradigm, Clinician-Guided Adversarial Fine-Tuning (CGAFT), which meticulously integrates expert clinical feedback into an adversarial learning framework to mitigate factual inconsistencies and improve diagnostic precision. Complementing this, our Knowledge Graph Augmentation Module (KGAM) acts as an inference-time safeguard, dynamically verifying generated medical statements against authoritative knowledge bases to minimize hallucinations and ensure standardized terminology. Leveraging a comprehensive dataset of millions of paired CXR images and expert reports, our experiments demonstrate that CXR-PathFinder significantly outperforms existing state-of-the-art medical vision-language models across various quantitative metrics, including clinical accuracy (Macro F1 (14): 46.5, Micro F1 (14): 59.5). Furthermore, blinded human evaluation by board-certified radiologists confirms CXR-PathFinder's superior clinical utility, completeness, and accuracy, establishing its potential as a reliable and efficient aid for radiological practice. The developed method effectively balances high diagnostic fidelity with computational efficiency, providing a robust solution for automated medical report generation.

Paperid: 6748, https://arxiv.org/pdf/2505.22481.pdf

Abstract:
This paper proposes a framework for semantic hypothesis testing tailored to imaging inverse problems. Modern imaging methods struggle to support hypothesis testing, a core component of the scientific method that is essential for the rigorous interpretation of experiments and robust interfacing with decision-making processes. There are three main reasons why image-based hypothesis testing is challenging. First, the difficulty of using a single observation to simultaneously reconstruct an image, formulate hypotheses, and quantify their statistical significance. Second, the hypotheses encountered in imaging are mostly of semantic nature, rather than quantitative statements about pixel values. Third, it is challenging to control test error probabilities because the null and alternative distributions are often unknown. Our proposed approach addresses these difficulties by leveraging concepts from self-supervised computational imaging, vision-language models, and non-parametric hypothesis testing with e-values. We demonstrate our proposed framework through numerical experiments related to image-based phenotyping, where we achieve excellent power while robustly controlling Type I errors.

Paperid: 6749, https://arxiv.org/pdf/2505.19944.pdf

Abstract:
The diagram is a visual representation of a relationship illustrated with edges (lines or arrows), which is widely used in industrial and scientific communication. Although recognizing diagrams is essential for vision language models (VLMs) to comprehend domain-specific knowledge, recent studies reveal that many VLMs fail to identify edges in images. We hypothesize that these failures stem from an over-reliance on textual and positional biases, preventing VLMs from learning explicit edge features. Based on this idea, we empirically investigate whether the image encoder in VLMs can learn edge representation through training on a diagram dataset in which edges are biased neither by textual nor positional information. To this end, we conduct contrastive learning on an artificially generated diagram--caption dataset to train an image encoder and evaluate its diagram-related features on three tasks: probing, image retrieval, and captioning. Our results show that the finetuned model outperforms pretrained CLIP in all tasks and surpasses zero-shot GPT-4o and LLaVA-Mistral in the captioning task. These findings confirm that eliminating textual and positional biases fosters accurate edge recognition in VLMs, offering a promising path for advancing diagram understanding.

Paperid: 6750, https://arxiv.org/pdf/2505.19895.pdf

Abstract:
Underwater images are often affected by complex degradations such as light absorption, scattering, color casts, and artifacts, making enhancement critical for effective object detection, recognition, and scene understanding in aquatic environments. Existing methods, especially diffusion-based approaches, typically rely on synthetic paired datasets due to the scarcity of real underwater references, introducing bias and limiting generalization. Furthermore, fine-tuning these models can degrade learned priors, resulting in unrealistic enhancements due to domain shifts. To address these challenges, we propose UDAN-CLIP, an image-to-image diffusion framework pre-trained on synthetic underwater datasets and enhanced with a customized classifier based on vision-language model, a spatial attention module, and a novel CLIP-Diffusion loss. The classifier preserves natural in-air priors and semantically guides the diffusion process, while the spatial attention module focuses on correcting localized degradations such as haze and low contrast. The proposed CLIP-Diffusion loss further strengthens visual-textual alignment and helps maintain semantic consistency during enhancement. The proposed contributions empower our UDAN-CLIP model to perform more effective underwater image enhancement, producing results that are not only visually compelling but also more realistic and detail-preserving. These improvements are consistently validated through both quantitative metrics and qualitative visual comparisons, demonstrating the model's ability to correct distortions and restore natural appearance in challenging underwater conditions.

Paperid: 6751, https://arxiv.org/pdf/2505.18134.pdf

Abstract:
Vision-language models (VLMs) have achieved strong results on coding and math benchmarks that are challenging for humans, yet their ability to perform tasks that come naturally to humans--such as perception, spatial navigation, and memory management--remains understudied. Real video games are crafted to be intuitive for humans to learn and master by leveraging innate inductive biases, making them an ideal testbed for evaluating such capabilities in VLMs. To this end, we introduce VideoGameBench, a benchmark consisting of 10 popular video games from the 1990s that VLMs directly interact with in real-time. VideoGameBench challenges models to complete entire games with access to only raw visual inputs and a high-level description of objectives and controls, a significant departure from existing setups that rely on game-specific scaffolding and auxiliary information. We keep three of the games secret to encourage solutions that generalize to unseen environments. Our experiments show that frontier vision-language models struggle to progress beyond the beginning of each game. We find inference latency to be a major limitation of frontier models in the real-time setting; therefore, we introduce VideoGameBench Lite, a setting where the game pauses while waiting for the LM's next action. The best performing model, Gemini 2.5 Pro, completes only 0.48% of VideoGameBench and 1.6% of VideoGameBench Lite. We hope that the formalization of the human skills mentioned above into this benchmark motivates progress in these research directions.

Paperid: 6752, https://arxiv.org/pdf/2505.14204.pdf

Abstract:
We introduce Perceptual-Initialization (PI), a paradigm shift in visual representation learning that incorporates human perceptual structure during the initialization phase rather than as a downstream fine-tuning step. By integrating human-derived triplet embeddings from the NIGHTS dataset to initialize a CLIP vision encoder, followed by self-supervised learning on YFCC15M, our approach demonstrates significant zero-shot performance improvements, without any task-specific fine-tuning, across 29 zero shot classification and 2 retrieval benchmarks. On ImageNet-1K, zero-shot gains emerge after approximately 15 epochs of pretraining. Benefits are observed across datasets of various scales, with improvements manifesting at different stages of the pretraining process depending on dataset characteristics. Our approach consistently enhances zero-shot top-1 accuracy, top-5 accuracy, and retrieval recall (e.g., R@1, R@5) across these diverse evaluation tasks, without requiring any adaptation to target domains. These findings challenge the conventional wisdom of using human-perceptual data primarily for fine-tuning and demonstrate that embedding human perceptual structure during early representation learning yields more capable and vision-language aligned systems that generalize immediately to unseen tasks. Our work shows that "beginning with you", starting with human perception, provides a stronger foundation for general-purpose vision-language intelligence.

Paperid: 6753, https://arxiv.org/pdf/2505.13281.pdf

Abstract:
With the rapid improvement of machine learning (ML) models, cognitive scientists are increasingly asking about their alignment with how humans think. Here, we ask this question for computer vision models and human sensitivity to geometric and topological (GT) concepts. Under the core knowledge account, these concepts are innate and supported by dedicated neural circuitry. In this work, we investigate an alternative explanation, that GT concepts are learned ``for free'' through everyday interaction with the environment. We do so using computer visions models, which are trained on large image datasets. We build on prior studies to investigate the overall performance and human alignment of three classes of models -- convolutional neural networks (CNNs), transformer-based models, and vision-language models -- on an odd-one-out task testing 43 GT concepts spanning seven classes. Transformer-based models achieve the highest overall accuracy, surpassing that of young children. They also show strong alignment with children's performance, finding the same classes of concepts easy vs. difficult. By contrast, vision-language models underperform their vision-only counterparts and deviate further from human profiles, indicating that naÃ¯ve multimodality might compromise abstract geometric sensitivity. These findings support the use of computer vision models to evaluate the sufficiency of the learning account for explaining human sensitivity to GT concepts, while also suggesting that integrating linguistic and visual representations might have unpredicted deleterious consequences.

Paperid: 6754, https://arxiv.org/pdf/2505.12670.pdf

Abstract:
Vision-Language Models (VLMs) have shown remarkable potential in advancing autonomous driving by leveraging multi-modal fusion in order to enhance scene perception, reasoning, and decision-making. Despite their potential, existing models suffer from computational overhead and inefficient integration of multi-view sensor data that make them impractical for real-time deployment in safety-critical autonomous driving applications. To address these shortcomings, this paper is devoted to designing a lightweight VLM called TS-VLM, which incorporates a novel Text-Guided SoftSort Pooling (TGSSP) module. By resorting to semantics of the input queries, TGSSP ranks and fuses visual features from multiple views, enabling dynamic and query-aware multi-view aggregation without reliance on costly attention mechanisms. This design ensures the query-adaptive prioritization of semantically related views, which leads to improved contextual accuracy in multi-view reasoning for autonomous driving. Extensive evaluations on the DriveLM benchmark demonstrate that, on the one hand, TS-VLM outperforms state-of-the-art models with a BLEU-4 score of 56.82, METEOR of 41.91, ROUGE-L of 74.64, and CIDEr of 3.39. On the other hand, TS-VLM reduces computational cost by up to 90%, where the smallest version contains only 20.1 million parameters, making it more practical for real-time deployment in autonomous vehicles.

Paperid: 6755, https://arxiv.org/pdf/2505.12237.pdf

Abstract:
Large Language Models (LLMs) and Vision-Language Models (VLMs) have demonstrated remarkable reasoning and generalization capabilities in video understanding; however, their application in video editing remains largely underexplored. This paper presents the first systematic study of LLMs in the context of video editing. To bridge the gap between visual information and language-based reasoning, we introduce L-Storyboard, an intermediate representation that transforms discrete video shots into structured language descriptions suitable for LLM processing. We categorize video editing tasks into Convergent Tasks and Divergent Tasks, focusing on three core tasks: Shot Attributes Classification, Next Shot Selection, and Shot Sequence Ordering. To address the inherent instability of divergent task outputs, we propose the StoryFlow strategy, which converts the divergent multi-path reasoning process into a convergent selection mechanism, effectively enhancing task accuracy and logical coherence. Experimental results demonstrate that L-Storyboard facilitates a more robust mapping between visual information and language descriptions, significantly improving the interpretability and privacy protection of video editing tasks. Furthermore, StoryFlow enhances the logical consistency and output stability in Shot Sequence Ordering, underscoring the substantial potential of LLMs in intelligent video editing.

Paperid: 6756, https://arxiv.org/pdf/2505.11060.pdf

Abstract:
Deep vision models often rely on biases learned from spurious correlations in datasets. To identify these biases, methods that interpret high-level, human-understandable concepts are more effective than those relying primarily on low-level features like heatmaps. A major challenge for these concept-based methods is the lack of image annotations indicating potentially bias-inducing concepts, since creating such annotations requires detailed labeling for each dataset and concept, which is highly labor-intensive. We present CUBIC (Concept embeddings for Unsupervised Bias IdentifiCation), a novel method that automatically discovers interpretable concepts that may bias classifier behavior. Unlike existing approaches, CUBIC does not rely on predefined bias candidates or examples of model failures tied to specific biases, as such information is not always available. Instead, it leverages image-text latent space and linear classifier probes to examine how the latent representation of a superclass label$\unicode{x2014}$shared by all instances in the dataset$\unicode{x2014}$is influenced by the presence of a given concept. By measuring these shifts against the normal vector to the classifier's decision boundary, CUBIC identifies concepts that significantly influence model predictions. Our experiments demonstrate that CUBIC effectively uncovers previously unknown biases using Vision-Language Models (VLMs) without requiring the samples in the dataset where the classifier underperforms or prior knowledge of potential biases.

Paperid: 6757, https://arxiv.org/pdf/2505.10453.pdf

Abstract:
Vision language models (VLMs) are AI systems paired with both language and vision encoders to process multimodal input. They are capable of performing complex semantic tasks such as automatic captioning, but it remains an open question about how well they comprehend the visuospatial properties of scenes depicted in the images they process. We argue that descriptions of virtual objects -- objects that are not visually represented in an image -- can help test scene comprehension in these AI systems. For example, an image that depicts a person standing under a tree can be paired with the following prompt: imagine that a kite is stuck in the tree. VLMs that comprehend the scene should update their representations and reason sensibly about the spatial relations between all three objects. We describe systematic evaluations of state-of-the-art VLMs and show that their ability to process virtual objects is inadequate.

Paperid: 6758, https://arxiv.org/pdf/2505.10088.pdf

Abstract:
Large-scale pre-trained Vision-Language Models (VLMs) have significantly advanced transfer learning across diverse tasks. However, adapting these models with limited few-shot data often leads to overfitting, undermining their ability to generalize to new tasks. To address this, we propose Multi-Modal Representation Learning (MMRL), which introduces a shared, learnable, modality-agnostic representation space. MMRL generates space tokens projected into both text and image encoders as representation tokens, enabling more effective cross-modal interactions. Unlike prior methods that mainly optimize class token features, MMRL inserts representation tokens into higher encoder layers--where task-specific features are more prominent--while preserving general knowledge in the lower layers. During training, both class and representation features are jointly optimized: a trainable projection layer is applied to representation tokens for task adaptation, while the projection layer for class token remains frozen to retain pre-trained knowledge. To further promote generalization, we introduce a regularization term aligning class and text features with the frozen VLM's zero-shot features. At inference, a decoupling strategy uses both class and representation features for base tasks, but only class features for novel tasks due to their stronger generalization. Building upon this, we propose MMRL++, a parameter-efficient and interaction-aware extension that significantly reduces trainable parameters and enhances intra-modal interactions--particularly across the layers of representation tokens--allowing gradient sharing and instance-specific information to propagate more effectively through the network. Extensive experiments on 15 datasets demonstrate that MMRL and MMRL++ consistently outperform state-of-the-art methods, achieving a strong balance between task-specific adaptation and generalization.

Paperid: 6759, https://arxiv.org/pdf/2505.09591.pdf

Abstract:
Despite remarkable progress in multimodal models for Visual Question Answering (VQA), there remain major reliability concerns because the models can often be overconfident and miscalibrated, especially in out-of-distribution (OOD) settings. Plenty has been done to address such issues for unimodal models, but little work exists for multimodal cases. Here, we address unreliability in multimodal models by proposing a Variational VQA approach. Specifically, instead of fine-tuning vision-language models by using AdamW, we employ a recently proposed variational algorithm called IVON, which yields a posterior distribution over model parameters. Through extensive experiments, we show that our approach improves calibration and abstentions without sacrificing the accuracy of AdamW. For instance, compared to AdamW fine-tuning, we reduce Expected Calibration Error by more than 50% compared to the AdamW baseline and raise Coverage by 4% vs. SOTA (for a fixed risk of 1%). In the presence of distribution shifts, the performance gain is even higher, achieving 8% Coverage (@ 1% risk) improvement vs. SOTA when 50% of test cases are OOD. Overall, we present variational learning as a viable option to enhance the reliability of multimodal models.

Paperid: 6760, https://arxiv.org/pdf/2505.06832.pdf

Abstract:
Open-vocabulary, task-oriented grasping of specific functional parts, particularly with dual arms, remains a key challenge, as current Vision-Language Models (VLMs), while enhancing task understanding, often struggle with precise grasp generation within defined constraints and effective dual-arm coordination. We innovatively propose UniDiffGrasp, a unified framework integrating VLM reasoning with guided part diffusion to address these limitations. UniDiffGrasp leverages a VLM to interpret user input and identify semantic targets (object, part(s), mode), which are then grounded via open-vocabulary segmentation. Critically, the identified parts directly provide geometric constraints for a Constrained Grasp Diffusion Field (CGDF) using its Part-Guided Diffusion, enabling efficient, high-quality 6-DoF grasps without retraining. For dual-arm tasks, UniDiffGrasp defines distinct target regions, applies part-guided diffusion per arm, and selects stable cooperative grasps. Through extensive real-world deployment, UniDiffGrasp achieves grasp success rates of 0.876 in single-arm and 0.767 in dual-arm scenarios, significantly surpassing existing state-of-the-art methods, demonstrating its capability to enable precise and coordinated open-vocabulary grasping in complex real-world scenarios.

Paperid: 6761, https://arxiv.org/pdf/2505.05040.pdf

Abstract:
Various social networks have been allowing media uploads for over a decade now. Still, it has not always been clear what is their relation with the posted text or even if there is any at all. In this work, we explore how multilingual vision-language models tackle the task of image-text relation prediction in different languages, and construct a dedicated balanced benchmark data set from Twitter posts in Latvian along with their manual translations into English. We compare our results to previous work and show that the more recently released vision-language model checkpoints are becoming increasingly capable at this task, but there is still much room for further improvement.

Paperid: 6762, https://arxiv.org/pdf/2505.04214.pdf

Abstract:
The automatic extraction of key-value information from handwritten documents is a key challenge in document analysis. A reliable extraction is a prerequisite for the mass digitization efforts of many archives. Large Vision Language Models (LVLM) are a promising technology to tackle this problem especially in scenarios where little annotated training data is available. In this work, we present a novel dataset specifically designed to evaluate the few-shot capabilities of LVLMs. The CM1 documents are a historic collection of forms with handwritten entries created in Europe to administer the Care and Maintenance program after World War Two. The dataset establishes three benchmarks on extracting name and birthdate information and, furthermore, considers different training set sizes. We provide baseline results for two different LVLMs and compare performances to an established full-page extraction model. While the traditional full-page model achieves highly competitive performances, our experiments show that when only a few training samples are available the considered LVLMs benefit from their size and heavy pretraining and outperform the classical approach.

Paperid: 6763, https://arxiv.org/pdf/2505.04147.pdf

Abstract:
"Read the room" is a significant social reasoning capability in human daily life. Humans can infer others' mental states from subtle social cues. Previous social reasoning tasks and datasets lack complexity (e.g., simple scenes, basic interactions, incomplete mental state variables, single-step reasoning, etc.) and fall far short of the challenges present in real-life social interactions. In this paper, we contribute a valuable, high-quality, and comprehensive video dataset named R^3-VQA with precise and fine-grained annotations of social events and mental states (i.e., belief, intent, desire, and emotion) as well as corresponding social causal chains in complex social scenarios. Moreover, we include human-annotated and model-generated QAs. Our task R^3-VQA includes three aspects: Social Event Understanding, Mental State Estimation, and Social Causal Reasoning. As a benchmark, we comprehensively evaluate the social reasoning capabilities and consistencies of current state-of-the-art large vision-language models (LVLMs). Comprehensive experiments show that (i) LVLMs are still far from human-level consistent social reasoning in complex social scenarios; (ii) Theory of Mind (ToM) prompting can help LVLMs perform better on social reasoning tasks. We provide some of our dataset and codes in supplementary material and will release our full dataset and codes upon acceptance.

Paperid: 6764, https://arxiv.org/pdf/2505.03420.pdf

Abstract:
Hallucinations in vision-language models (VLMs) hinder reliability and real-world applicability, usually stemming from distribution shifts between pretraining data and test samples. Existing solutions, such as retraining or fine-tuning on additional data, demand significant computational resources and labor-intensive data collection, while ensemble-based methods incur additional costs by introducing auxiliary VLMs. To address these challenges, we propose a novel test-time adaptation framework using reinforcement learning to mitigate hallucinations during inference without retraining or any auxiliary VLMs. By updating only the learnable parameters in the layer normalization of the language model (approximately 0.003% of the model parameters), our method reduces distribution shifts between test samples and pretraining samples. A CLIP-based hallucination evaluation model is proposed to provide dual rewards to VLMs. Experimental results demonstrate a 15.4% and 17.3% reduction in hallucination rates on LLaVA and InstructBLIP, respectively. Our approach outperforms state-of-the-art baselines with a 68.3% improvement in hallucination mitigation, demonstrating its effectiveness.

Paperid: 6765, https://arxiv.org/pdf/2505.03374.pdf

Abstract:
Introduction: Data from wearable devices collected in free-living settings, and labelled with physical activity behaviours compatible with health research, are essential for both validating existing wearable-based measurement approaches and developing novel machine learning approaches. One common way of obtaining these labels relies on laborious annotation of sequences of images captured by cameras worn by participants through the course of a day. Methods: We compare the performance of three vision language models and two discriminative models on two free-living validation studies with 161 and 111 participants, collected in Oxfordshire, United Kingdom and Sichuan, China, respectively, using the Autographer (OMG Life, defunct) wearable camera. Results: We found that the best open-source vision-language model (VLM) and fine-tuned discriminative model (DM) achieved comparable performance when predicting sedentary behaviour from single images on unseen participants in the Oxfordshire study; median F1-scores: VLM = 0.89 (0.84, 0.92), DM = 0.91 (0.86, 0.95). Performance declined for light (VLM = 0.60 (0.56,0.67), DM = 0.70 (0.63, 0.79)), and moderate-to-vigorous intensity physical activity (VLM = 0.66 (0.53, 0.85); DM = 0.72 (0.58, 0.84)). When applied to the external Sichuan study, performance fell across all intensity categories, with median Cohen's kappa-scores falling from 0.54 (0.49, 0.64) to 0.26 (0.15, 0.37) for the VLM, and from 0.67 (0.60, 0.74) to 0.19 (0.10, 0.30) for the DM. Conclusion: Freely available computer vision models could help annotate sedentary behaviour, typically the most prevalent activity of daily living, from wearable camera images within similar populations to seen data, reducing the annotation burden.

Paperid: 6766, https://arxiv.org/pdf/2504.21194.pdf

Abstract:
This paper introduces ISS-Geo142, a curated benchmark for geolocating astronaut photography captured from the International Space Station (ISS). Although the ISS position at capture time is known precisely, the specific Earth locations depicted in these images are typically not directly georeferenced, making automated localization non-trivial. ISS-Geo142 consists of 142 images with associated metadata and manually determined geographic locations, spanning a range of spatial scales and scene types. On top of this benchmark, we implement and evaluate three geolocation pipelines: a neural network based approach (NN-Geo) using VGG16 features and cross-correlation over map-derived Areas of Interest (AOIs), a Scale-Invariant Feature Transform based pipeline (SIFT-Match) using sliding-window feature matching on stitched high-resolution AOIs, and TerraByte, an AI system built around a GPT-4 model with vision capabilities that jointly reasons over image content and ISS coordinates. On ISS-Geo142, NN-Geo achieves a match for 75.52\% of the images under our evaluation protocol, SIFT-Match attains high precision on structurally rich scenes at substantial computational cost, and TerraByte establishes the strongest overall baseline, correctly geolocating approximately 90\% of the images while also producing human-readable geographic descriptions. The methods and experiments were originally developed in 2023; this manuscript is a revised and extended version that situates the work relative to subsequent advances in cross-view geo-localization and remote-sensing vision--language models. Taken together, ISS-Geo142 and these three pipelines provide a concrete, historically grounded benchmark for future work on ISS image geolocation.

Paperid: 6767, https://arxiv.org/pdf/2504.16061.pdf

Abstract:
Vision language models (VLMs) are designed to extract relevant visuospatial information from images. Some research suggests that VLMs can exhibit humanlike scene understanding, while other investigations reveal difficulties in their ability to process relational information. To achieve widespread applicability, VLMs must perform reliably, yielding comparable competence across a wide variety of related tasks. We sought to test how reliable these architectures are at engaging in trivial spatial cognition, e.g., recognizing whether one object is left of another in an uncluttered scene. We developed a benchmark dataset -- TableTest -- whose images depict 3D scenes of objects arranged on a table, and used it to evaluate state-of-the-art VLMs. Results show that performance could be degraded by minor variations of prompts that use logically equivalent descriptions. These analyses suggest limitations in how VLMs may reason about spatial relations in real-world applications. They also reveal novel opportunities for bolstering image caption corpora for more efficient training and testing.

Paperid: 6768, https://arxiv.org/pdf/2504.15929.pdf

Abstract:
Diagnostic imaging relies on interpreting both images and radiology reports, but the growing data volumes place significant pressure on medical experts, yielding increased errors and workflow backlogs. Medical vision-language models (med-VLMs) have emerged as a powerful framework to efficiently process multimodal imaging data, particularly in chest X-ray (CXR) evaluations, albeit their performance hinges on how well image and text representations are aligned. Existing alignment methods, predominantly based on contrastive learning, prioritize separation between disease classes over segregation of fine-grained pathology attributes like location, size or severity, leading to suboptimal representations. Here, we propose MedTrim (Meta-entity-driven Triplet mining), a novel method that enhances image-text alignment through multimodal triplet learning synergistically guided by disease class as well as adjectival and directional pathology descriptors. Unlike common alignment methods that separate broad disease classes, MedTrim leverages structured meta-entity information to preserve subtle but clinically significant intra-class variations. For this purpose, we first introduce an ontology-based entity recognition module that extracts pathology-specific meta-entities from CXR reports, as annotations on pathology attributes are rare in public datasets. For refined sample selection in triplet mining, we then introduce a novel score function that captures an aggregate measure of inter-sample similarity based on disease classes and adjectival/directional descriptors. Lastly, we introduce a multimodal triplet alignment objective for explicit within- and cross-modal alignment between samples sharing detailed pathology characteristics. Our demonstrations indicate that MedTrim improves performance in downstream retrieval and classification tasks compared to state-of-the-art alignment methods.

Paperid: 6769, https://arxiv.org/pdf/2504.11108.pdf

Abstract:
Similar to LLMs, the development of vision language models is mainly driven by English datasets and models trained in English and Chinese language, whereas support for other languages, even those considered high-resource languages such as German, remains significantly weaker. In this work we present an analysis of open-weight VLMs on factual knowledge in the German and English language. We disentangle the image-related aspects from the textual ones by analyzing accu-racy with jury-as-a-judge in both prompt languages and images from German and international contexts. We found that for celebrities and sights, VLMs struggle because they are lacking visual cognition of German image contents. For animals and plants, the tested models can often correctly identify the image contents ac-cording to the scientific name or English common name but fail in German lan-guage. Cars and supermarket products were identified equally well in English and German images across both prompt languages.

Paperid: 6770, https://arxiv.org/pdf/2504.10049.pdf

Abstract:
Vision-Language Models (VLMs) can process visual and textual information in multiple formats: texts, images, interleaved texts and images, or even hour-long videos. In this work, we conduct fine-grained quantitative and qualitative analyses of automatic summarization of multimodal presentations using VLMs with various representations as input. From these experiments, we suggest cost-effective strategies for generating summaries from text-heavy multimodal documents under different input-length budgets using VLMs. We show that slides extracted from the video stream can be beneficially used as input against the raw video, and that a structured representation from interleaved slides and transcript provides the best performance. Finally, we reflect and comment on the nature of cross-modal interactions in multimodal presentations and share suggestions to improve the capabilities of VLMs to understand documents of this nature.

Paperid: 6771, https://arxiv.org/pdf/2504.08269.pdf

Abstract:
The increasing availability of multimodal data across text, tables, and images presents new challenges for developing models capable of complex cross-modal reasoning. Existing methods for Multimodal Multi-hop Question Answering (MMQA) often suffer from limited reasoning capabilities, reliance on modality conversion, and inadequate alignment between visual and textual representations. To address these limitations, this paper introduces Vision-Language Multimodal Transformer (VLMT), a unified architecture that integrates a transformer-based vision encoder with a sequence-to-sequence language model. VLMT employs a direct token-level injection mechanism to fuse visual and textual inputs within a shared embedding space, eliminating the need for intermediate projection layers. To enhance cross-modal alignment and reasoning, a three-stage pretraining strategy is proposed to progressively align vision-language representations and improve the model's capacity for multimodal understanding. Based on the pretrained backbone, two task-specific modules are instantiated to form a two-stage MMQA framework: a multimodal reranker that predicts document relevance scores and utilizes a relative threshold with top-k strategy for context retrieval, and a multimodal question answering model that generates contextually grounded answers based on the retrieved evidence. Comprehensive experiments on two benchmark datasets demonstrate the effectiveness of the proposed approach. On MultimodalQA validation set, VLMT-Large achieves 76.5% Exact Match and 80.1% F1, outperforming the previous state-of-the-art by +9.1% in Exact Match and +8.8% in F1. On WebQA, it attains a QA score of 47.6, surpassing prior models such as PERQA by +3.2. These results highlight VLMT's strong capabilities in multimodal reasoning and its potential to advance real-world information retrieval and question answering systems.

Paperid: 6772, https://arxiv.org/pdf/2504.07198.pdf

Abstract:
The human face plays a central role in social communication, necessitating the use of performant computer vision tools for human-centered applications. We propose Face-LLaVA, a multimodal large language model for face-centered, in-context learning, including facial expression and attribute recognition. Additionally, Face-LLaVA is able to generate natural language descriptions that can be used for reasoning. Leveraging existing visual databases, we first developed FaceInstruct-1M, a face-centered database for instruction tuning MLLMs for face processing. We then developed a novel face-specific visual encoder powered by Face-Region Guided Cross-Attention that integrates face geometry with local visual features. We evaluated the proposed method across nine different datasets and five different face processing tasks, including facial expression recognition, action unit detection, facial attribute detection, age estimation and deepfake detection. Face-LLaVA achieves superior results compared to existing open-source MLLMs and competitive performance compared to commercial solutions. Our model output also receives a higher reasoning rating by GPT under a zero-shot setting across all the tasks. Both our dataset and model wil be released at https://face-llava.github.io to support future advancements in social AI and foundational vision-language research.

Paperid: 6773, https://arxiv.org/pdf/2504.07110.pdf

Abstract:
Despite the success of vision-language models in various generative tasks, obtaining high-quality semantic representations for products and user intents is still challenging due to the inability of off-the-shelf models to capture nuanced relationships between the entities. In this paper, we introduce a joint training framework for product and user queries by aligning uni-modal and multi-modal encoders through contrastive learning on image-text data. Our novel approach trains a query encoder with an LLM-curated relevance dataset, eliminating the reliance on engagement history. These embeddings demonstrate strong generalization capabilities and improve performance across applications, including product categorization and relevance prediction. For personalized ads recommendation, a significant uplift in the click-through rate and conversion rate after the deployment further confirms the impact on key business metrics. We believe that the flexibility of our framework makes it a promising solution toward enriching the user experience across the e-commerce landscape.

Paperid: 6774, https://arxiv.org/pdf/2504.01916.pdf

Abstract:
As a pioneering vision-language model, CLIP (Contrastive Language-Image Pre-training) has achieved significant success across various domains and a wide range of downstream vision-language tasks. However, the text encoders in popular CLIP models are limited to processing only 77 text tokens, which constrains their ability to effectively handle longer, detail-rich captions. Additionally, CLIP models often struggle to effectively capture detailed visual and textual information, which hampers their performance on tasks that require fine-grained analysis. To address these limitations, we present a novel approach, \textbf{FineLIP}, that extends the capabilities of CLIP. FineLIP enhances cross-modal text-image mapping by incorporating \textbf{Fine}-grained alignment with \textbf{L}onger text input within the CL\textbf{IP}-style framework. FineLIP first extends the positional embeddings to handle longer text, followed by the dynamic aggregation of local image and text tokens. The aggregated results are then used to enforce fine-grained token-to-token cross-modal alignment. We validate our model on datasets with long, detailed captions across two tasks: zero-shot cross-modal retrieval and text-to-image generation. Quantitative and qualitative experimental results demonstrate the effectiveness of FineLIP, outperforming existing state-of-the-art approaches. Furthermore, comprehensive ablation studies validate the benefits of key design elements within FineLIP.

Paperid: 6775, https://arxiv.org/pdf/2503.23956.pdf

Abstract:
Recent advancements in Large Visual Language Models (LVLMs) have gained significant attention due to their remarkable reasoning capabilities and proficiency in generalization. However, processing a large number of visual tokens and generating long-context outputs impose substantial computational overhead, leading to excessive demands for key-value (KV) cache. To address this critical bottleneck, we propose AirCache, a novel KV cache compression method aimed at accelerating LVLMs inference. This work systematically investigates the correlations between visual and textual tokens within the attention mechanisms of LVLMs. Our empirical analysis reveals considerable redundancy in cached visual tokens, wherein strategically eliminating these tokens preserves model performance while significantly accelerating context generation. Inspired by these findings, we introduce an elite observation window for assessing the importance of visual components in the KV cache, focusing on stable inter-modal relevancy modeling with enhanced multi-perspective consistency. Additionally, we develop an adaptive layer-wise budget allocation strategy that capitalizes on the strength and skewness of token importance distribution, showcasing superior efficiency compared to uniform allocation. Comprehensive evaluations across multiple LVLMs and benchmarks demonstrate that our method achieves comparable performance to the full cache while retaining only 10% of visual KV cache, thereby reducing decoding latency by 29% to 66% across various batch size and prompt length of inputs. Notably, as cache retention rates decrease, our method exhibits increasing performance advantages over existing approaches.

Paperid: 6776, https://arxiv.org/pdf/2503.22668.pdf

Abstract:
Co-speech gestures play a vital role in non-verbal communication. In this paper, we introduce a new framework for co-speech gesture understanding in the wild. Specifically, we propose three new tasks and benchmarks to evaluate a model's capability to comprehend gesture-speech-text associations: (i) gesture based retrieval, (ii) gesture word spotting, and (iii) active speaker detection using gestures. We present a new approach that learns a tri-modal video-gesture-speech-text representation to solve these tasks. By leveraging a combination of global phrase contrastive loss and local gesture-word coupling loss, we demonstrate that a strong gesture representation can be learned in a weakly supervised manner from videos in the wild. Our learned representations outperform previous methods, including large vision-language models (VLMs). Further analysis reveals that speech and text modalities capture distinct gesture related signals, underscoring the advantages of learning a shared tri-modal embedding space. The dataset, model, and code are available at: https://www.robots.ox.ac.uk/~vgg/research/jegal.

Paperid: 6777, https://arxiv.org/pdf/2503.20030.pdf

Abstract:
Short video streaming systems such as TikTok, Youtube Shorts, Instagram Reels, etc have reached billions of active users. At the core of such systems is a (proprietary) recommendation algorithm which recommends a sequence of videos to each user, in a personalized way. We aim to understand the temporal evolution of recommendations made by these algorithms and the interplay between the recommendations and user experience. While past work has studied recommendation algorithms using textual data (e.g., titles, hashtags, etc.) as well as user studies and interviews, we add a third modality of analysis - automated analysis of the videos themselves. Our content-based analysis framework leverages recent advances in Vision Language Models (VLMs). Together we use this trifecta of methodologies (analysis of user watch history and logs, user studies and interviews, and content-based analysis) to analyze challenging temporal aspects of how well TikTok's recommendation algorithm is received by users, is affected by user interactions, and aligns with user history; as well as how users are sensitive to the order of videos recommended, and how the algorithm's effectiveness itself may be predictable in the future. While it is not our goal to reverse-engineer TikTok's recommendation algorithm, our new findings indicate behavioral aspects that both users and algorithm developers would benefit from.

Paperid: 6778, https://arxiv.org/pdf/2503.19654.pdf

Abstract:
We introduce RGB-Th-Bench, the first benchmark designed to evaluate the ability of Vision-Language Models (VLMs) to comprehend RGB-Thermal image pairs. While VLMs have demonstrated remarkable progress in visual reasoning and multimodal understanding, their evaluation has been predominantly limited to RGB-based benchmarks, leaving a critical gap in assessing their capabilities in infrared vision tasks. Existing visible-infrared datasets are either task-specific or lack high-quality annotations necessary for rigorous model evaluation. To address these limitations, RGB-Th-Bench provides a comprehensive evaluation framework covering 14 distinct skill dimensions, with a total of 1,600+ expert-annotated Yes/No questions. The benchmark employs two accuracy metrics: a standard question-level accuracy and a stricter skill-level accuracy, which evaluates model robustness across multiple questions within each skill dimension. This design ensures a thorough assessment of model performance, including resilience to adversarial and hallucinated responses. We conduct extensive evaluations on 19 state-of-the-art VLMs, revealing significant performance gaps in RGB-Thermal understanding. Our results show that even the strongest models struggle with thermal image comprehension, with performance heavily constrained by their RGB-based capabilities. Additionally, the lack of large-scale application-specific and expert-annotated thermal-caption-pair datasets in pre-training is an important reason of the observed performance gap. RGB-Th-Bench highlights the urgent need for further advancements in multimodal learning to bridge the gap between visible and thermal image understanding. The dataset is available through this link, and the evaluation code will also be made publicly available.

Paperid: 6779, https://arxiv.org/pdf/2503.19281.pdf

Abstract:
Proving Rubik's Cube theorems at the high level represents a notable milestone in human-level spatial imagination and logic thinking and reasoning. Traditional Rubik's Cube robots, relying on complex vision systems and fixed algorithms, often struggle to adapt to complex and dynamic scenarios. To overcome this limitation, we introduce CubeRobot, a novel vision-language model (VLM) tailored for solving 3x3 Rubik's Cubes, empowering embodied agents with multimodal understanding and execution capabilities. We used the CubeCoT image dataset, which contains multiple-level tasks (43 subtasks in total) that humans are unable to handle, encompassing various cube states. We incorporate a dual-loop VisionCoT architecture and Memory Stream, a paradigm for extracting task-related features from VLM-generated planning queries, thus enabling CubeRobot to independent planning, decision-making, reflection and separate management of high- and low-level Rubik's Cube tasks. Furthermore, in low-level Rubik's Cube restoration tasks, CubeRobot achieved a high accuracy rate of 100%, similar to 100% in medium-level tasks, and achieved an accuracy rate of 80% in high-level tasks.

Paperid: 6780, https://arxiv.org/pdf/2503.18817.pdf

Abstract:
Prior research on out-of-distribution detection (OoDD) has primarily focused on single-modality models. Recently, with the advent of large-scale pretrained vision-language models such as CLIP, OoDD methods utilizing such multi-modal representations through zero-shot and prompt learning strategies have emerged. However, these methods typically involve either freezing the pretrained weights or only partially tuning them, which can be suboptimal for downstream datasets. In this paper, we highlight that multi-modal fine-tuning (MMFT) can achieve notable OoDD performance. Despite some recent works demonstrating the impact of fine-tuning methods for OoDD, there remains significant potential for performance improvement. We investigate the limitation of naÃ¯ve fine-tuning methods, examining why they fail to fully leverage the pretrained knowledge. Our empirical analysis suggests that this issue could stem from the modality gap within in-distribution (ID) embeddings. To address this, we propose a training objective that enhances cross-modal alignment by regularizing the distances between image and text embeddings of ID data. This adjustment helps in better utilizing pretrained textual information by aligning similar semantics from different modalities (i.e., text and image) more closely in the hyperspherical representation space. We theoretically demonstrate that the proposed regularization corresponds to the maximum likelihood estimation of an energy-based model on a hypersphere. Utilizing ImageNet-1k OoD benchmark datasets, we show that our method, combined with post-hoc OoDD approaches leveraging pretrained knowledge (e.g., NegLabel), significantly outperforms existing methods, achieving state-of-the-art OoDD performance and leading ID accuracy.

Paperid: 6781, https://arxiv.org/pdf/2503.18812.pdf

Abstract:
The aim of this work is to explore the potential of pre-trained vision-language models, e.g. Vision Transformers (ViT), enhanced with advanced data augmentation strategies for the detection of AI-generated images. Our approach leverages a fine-tuned ViT model trained on the Defactify-4.0 dataset, which includes images generated by state-of-the-art models such as Stable Diffusion 2.1, Stable Diffusion XL, Stable Diffusion 3, DALL-E 3, and MidJourney. We employ perturbation techniques like flipping, rotation, Gaussian noise injection, and JPEG compression during training to improve model robustness and generalisation. The experimental results demonstrate that our ViT-based pipeline achieves state-of-the-art performance, significantly outperforming competing methods on both validation and test datasets.

Paperid: 6782, https://arxiv.org/pdf/2503.16776.pdf

Abstract:
Vision-language models (VLMs) show great promise for 3D scene understanding but are mainly applied to indoor spaces or autonomous driving, focusing on low-level tasks like segmentation. This work expands their use to urban-scale environments by leveraging 3D reconstructions from multi-view aerial imagery. We propose OpenCity3D, an approach that addresses high-level tasks, such as population density estimation, building age classification, property price prediction, crime rate assessment, and noise pollution evaluation. Our findings highlight OpenCity3D's impressive zero-shot and few-shot capabilities, showcasing adaptability to new contexts. This research establishes a new paradigm for language-driven urban analytics, enabling applications in planning, policy, and environmental monitoring. See our project page: opencity3d.github.io

Paperid: 6783, https://arxiv.org/pdf/2503.10886.pdf

Abstract:
In the context of pressing climate change challenges and the significant biodiversity loss among arthropods, automated taxonomic classification from organismal images is a subject of intense research. However, traditional AI pipelines based on deep neural visual architectures such as CNNs or ViTs face limitations such as degraded performance on the long-tail of classes and the inability to reason about their predictions. We integrate image captioning and retrieval-augmented generation (RAG) with large language models (LLMs) to enhance biodiversity monitoring, showing particular promise for characterizing rare and unknown arthropod species. While a naive Vision-Language Model (VLM) excels in classifying images of common species, the RAG model enables classification of rarer taxa by matching explicit textual descriptions of taxonomic features to contextual biodiversity text data from external sources. The RAG model shows promise in reducing overconfidence and enhancing accuracy relative to naive LLMs, suggesting its viability in capturing the nuances of taxonomic hierarchy, particularly at the challenging family and genus levels. Our findings highlight the potential for modern vision-language AI pipelines to support biodiversity conservation initiatives, emphasizing the role of comprehensive data curation and collaboration with citizen science platforms to improve species identification, unknown species characterization and ultimately inform conservation strategies.

Paperid: 6784, https://arxiv.org/pdf/2503.08805.pdf

Abstract:
We introduce Filter Like You Test (FLYT), an algorithm for curating large-scale vision-language datasets that learns the usefulness of each data point as a pretraining example. FLYT trains a scoring model that learns to weigh each example's features using gradient signals from downstream tasks training sets. Based on FLYT, we implement Mixing-FLYT (M-FLYT), which takes the per-example scores generated by different scoring methods as features, and learns to unify them into a single score. FLYT naturally produces a distribution over the training examples, which we leverage through Soft Cap Sampling (SCS), a strategy for obtaining a filtered pretraining dataset from per-example probabilities that samples examples while preventing over-representation through a repetition penalty. Using these methods, we achieve 40.1% ImageNet zero-shot accuracy on the DataComp medium scale filtering benchmark, a 2% absolute accuracy increase over all previous results and a 5.5% increase over results that - like us - use only public resources. Our approach also yields 37.7\% on the average of 38 DataComp evaluation tasks, outperforming previous public-resource approaches by 0.4\%.

Paperid: 6785, https://arxiv.org/pdf/2503.06003.pdf

Abstract:
Situational awareness applications rely heavily on real-time processing of visual and textual data to provide actionable insights. Vision language models (VLMs) have become essential tools for interpreting complex environments by connecting visual inputs with natural language descriptions. However, these models often face computational challenges, especially when required to perform efficiently in real environments. This research presents a novel vision language model (VLM) framework that leverages frequency domain transformations and low-rank adaptation (LoRA) to enhance feature extraction, scalability, and efficiency. Unlike traditional VLMs, which rely solely on spatial-domain representations, our approach incorporates Discrete Fourier Transform (DFT) based low-rank features while retaining pretrained spatial weights, enabling robust performance in noisy or low visibility scenarios. We evaluated the proposed model on caption generation and Visual Question Answering (VQA) tasks using benchmark datasets with varying levels of Gaussian noise. Quantitative results demonstrate that our model achieves evaluation metrics comparable to state-of-the-art VLMs, such as CLIP ViT-L/14 and SigLIP. Qualitative analysis further reveals that our model provides more detailed and contextually relevant responses, particularly for real-world images captured by a RealSense camera mounted on an Unmanned Ground Vehicle (UGV).

Paperid: 6786, https://arxiv.org/pdf/2503.05186.pdf

Abstract:
In recent text-video retrieval, the use of additional captions from vision-language models has shown promising effects on the performance. However, existing models using additional captions often have struggled to capture the rich semantics, including temporal changes, inherent in the video. In addition, incorrect information caused by generative models can lead to inaccurate retrieval. To address these issues, we propose a new framework, Narrating the Video (NarVid), which strategically leverages the comprehensive information available from frame-level captions, the narration. The proposed NarVid exploits narration in multiple ways: 1) feature enhancement through cross-modal interactions between narration and video, 2) query-aware adaptive filtering to suppress irrelevant or incorrect information, 3) dual-modal matching score by adding query-video similarity and query-narration similarity, and 4) hard-negative loss to learn discriminative features from multiple perspectives using the two similarities from different views. Experimental results demonstrate that NarVid achieves state-of-the-art performance on various benchmark datasets.

Paperid: 6787, https://arxiv.org/pdf/2503.05064.pdf

Abstract:
Vision-Language Models (VLMs) demonstrate remarkable potential in robotic manipulation, yet challenges persist in executing complex fine manipulation tasks with high speed and precision. While excelling at high-level planning, existing VLM methods struggle to guide robots through precise sequences of fine motor actions. To address this limitation, we introduce a progressive VLM planning algorithm that empowers robots to perform fast, precise, and error-correctable fine manipulation. Our method decomposes complex tasks into sub-actions and maintains three key data structures: task memory structure, 2D topology graphs, and 3D spatial networks, achieving high-precision spatial-semantic fusion. These three components collectively accumulate and store critical information throughout task execution, providing rich context for our task-oriented VLM interaction mechanism. This enables VLMs to dynamically adjust guidance based on real-time feedback, generating precise action plans and facilitating step-wise error correction. Experimental validation on complex assembly tasks demonstrates that our algorithm effectively guides robots to rapidly and precisely accomplish fine manipulation in challenging scenarios, significantly advancing robot intelligence for precision tasks.

Paperid: 6788, https://arxiv.org/pdf/2503.04229.pdf

Abstract:
Pre-trained Vision-Language Models (VLMs) require Continual Learning (CL) to efficiently update their knowledge and adapt to various downstream tasks without retraining from scratch. However, for VLMs, in addition to the loss of knowledge previously learned from downstream tasks, pre-training knowledge is also corrupted during continual fine-tuning. This issue is exacerbated by the unavailability of original pre-training data, leaving VLM's generalization ability degrading. In this paper, we propose GIFT, a novel continual fine-tuning approach that utilizes synthetic data to overcome catastrophic forgetting in VLMs. Taking advantage of recent advances in text-to-image synthesis, we employ a pre-trained diffusion model to recreate both pre-training and learned downstream task data. In this way, the VLM can revisit previous knowledge through distillation on matching diffusion-generated images and corresponding text prompts. Leveraging the broad distribution and high alignment between synthetic image-text pairs in VLM's feature space, we propose a contrastive distillation loss along with an image-text alignment constraint. To further combat in-distribution overfitting and enhance distillation performance with limited amount of generated data, we incorporate adaptive weight consolidation, utilizing Fisher information from these synthetic image-text pairs and achieving a better stability-plasticity balance. Extensive experiments demonstrate that our method consistently outperforms previous state-of-the-art approaches across various settings.

Paperid: 6789, https://arxiv.org/pdf/2503.02816.pdf

Abstract:
The advancement of Vision-Language Model (VLM) camera sensors, which enable autonomous understanding of household situations without user intervention, has the potential to completely transform the DIY smart home building experience. Will this simplify or complicate the DIY smart home process? Additionally, what features do users want to create using these sensors? To explore this, we conducted a three-week diary-based experience prototyping study with 12 participants. Participants recorded their daily activities, used GPT to analyze the images, and manually customized and tested smart home features based on the analysis. The study revealed three key findings: (1) participants' expectations for VLM camera-based smart homes, (2) the impact of VLM camera sensor characteristics on the DIY process, and (3) users' concerns. Through the findings of this study, we propose design implications to support the DIY smart home building process with VLM camera sensors, and discuss living with intelligence.

Paperid: 6790, https://arxiv.org/pdf/2503.01891.pdf

Abstract:
Recent advances in large language models (LLMs) and vision-language models (LVLMs) have shown promise across many tasks, yet their scientific reasoning capabilities remain untested, particularly in multimodal settings. We present MMSciBench, a benchmark for evaluating mathematical and physical reasoning through text-only and text-image formats, with human-annotated difficulty levels, solutions with detailed explanations, and taxonomic mappings. Evaluation of state-of-the-art models reveals significant limitations, with even the best model achieving only \textbf{63.77\%} accuracy and particularly struggling with visual reasoning tasks. Our analysis exposes critical gaps in complex reasoning and visual-textual integration, establishing MMSciBench as a rigorous standard for measuring progress in multimodal scientific understanding. The code for MMSciBench is open-sourced at GitHub, and the dataset is available at Hugging Face.

Paperid: 6791, https://arxiv.org/pdf/2503.01263.pdf

Abstract:
Existing vision-language models (VLMs) such as CLIP have showcased an impressive capability to generalize well across various downstream tasks. These models leverage the synergy between visual and textual information, enabling them to understand and reason about the content present in images and text in a unified manner. This article provides a brief overview of CLIP based on few-shot prompt learning, including experimental data and technical characteristics of some methods. The purpose of this review is to provide a reference for researchers who have just started their research in generalizable prompting of CLIP through few-shot training for classification across 15 datasets and also to facilitate the integration of this field by researchers in other downstream tasks.

Paperid: 6792, https://arxiv.org/pdf/2502.20577.pdf

Abstract:
Facial appearance editing is crucial for digital avatars, AR/VR, and personalized content creation, driving realistic user experiences. However, preserving identity with generative models is challenging, especially in scenarios with limited data availability. Traditional methods often require multiple images and still struggle with unnatural face shifts, inconsistent hair alignment, or excessive smoothing effects. To overcome these challenges, we introduce a novel diffusion-based framework, InstaFace, to generate realistic images while preserving identity using only a single image. Central to InstaFace, we introduce an efficient guidance network that harnesses 3D perspectives by integrating multiple 3DMM-based conditionals without introducing additional trainable parameters. Moreover, to ensure maximum identity retention as well as preservation of background, hair, and other contextual features like accessories, we introduce a novel module that utilizes feature embeddings from a facial recognition model and a pre-trained vision-language model. Quantitative evaluations demonstrate that our method outperforms several state-of-the-art approaches in terms of identity preservation, photorealism, and effective control of pose, expression, and lighting.

Paperid: 6793, https://arxiv.org/pdf/2502.18512.pdf

Abstract:
The rapid success of Vision Large Language Models (VLLMs) often depends on the high-resolution images with abundant visual tokens, which hinders training and deployment efficiency. Current training-free visual token compression methods exhibit serious performance degradation in tasks involving high-resolution, text-oriented image understanding and reasoning. In this paper, we propose an efficient visual token compression framework for text-oriented VLLMs in high-resolution scenarios. In particular, we employ a light-weight self-distillation pre-training stage to compress the visual tokens, requiring a limited numbers of image-text pairs and minimal learnable parameters. Afterwards, to mitigate potential performance degradation of token-compressed models, we construct a high-quality post-train stage. To validate the effectiveness of our method, we apply it to an advanced VLLMs, InternVL2. Experimental results show that our approach significantly reduces computational overhead while outperforming the baselines across a range of text-oriented benchmarks. We will release the models and code soon.

Paperid: 6794, https://arxiv.org/pdf/2502.18356.pdf

Abstract:
We introduce WebGames, a comprehensive benchmark suite designed to evaluate general-purpose web-browsing AI agents through a collection of 50+ interactive challenges. These challenges are specifically crafted to be straightforward for humans while systematically testing the limitations of current AI systems across fundamental browser interactions, advanced input processing, cognitive tasks, workflow automation, and interactive entertainment. Our framework eliminates external dependencies through a hermetic testing environment, ensuring reproducible evaluation with verifiable ground-truth solutions. We evaluate leading vision-language models including GPT-4o, Claude Computer-Use, Gemini-1.5-Pro, and Qwen2-VL against human performance. Results reveal a substantial capability gap, with the best AI system achieving only 43.1% success rate compared to human performance of 95.7%, highlighting fundamental limitations in current AI systems' ability to handle common web interaction patterns that humans find intuitive. The benchmark is publicly available at webgames.convergence.ai, offering a lightweight, client-side implementation that facilitates rapid evaluation cycles. Through its modular architecture and standardized challenge specifications, WebGames provides a robust foundation for measuring progress in development of more capable web-browsing agents.

Paperid: 6795, https://arxiv.org/pdf/2502.15739.pdf

Abstract:
Despite regulatory efforts to establish reliable content-rating guidelines for mobile apps, the process of assigning content ratings in the Google Play Store remains self-regulated by the app developers. There is no straightforward method of verifying developer-assigned content ratings manually due to the overwhelming scale or automatically due to the challenging problem of interpreting textual and visual data and correlating them with content ratings. We propose and evaluate a visionlanguage approach to predict the content ratings of mobile game applications and detect content rating violations, using a dataset of metadata of popular Android games. Our method achieves ~6% better relative accuracy compared to the state-of-the-art CLIP-fine-tuned model in a multi-modal setting. Applying our classifier in the wild, we detected more than 70 possible cases of content rating violations, including nine instances with the 'Teacher Approved' badge. Additionally, our findings indicate that 34.5% of the apps identified by our classifier as violating content ratings were removed from the Play Store. In contrast, the removal rate for correctly classified apps was only 27%. This discrepancy highlights the practical effectiveness of our classifier in identifying apps that are likely to be removed based on user complaints.

Paperid: 6796, https://arxiv.org/pdf/2502.15389.pdf

Abstract:
Vision-Language Models (VLMs) occasionally generate outputs that contradict input images, constraining their reliability in real-world applications. While visual prompting is reported to suppress hallucinations by augmenting prompts with relevant area inside an image, the effectiveness in terms of the area remains uncertain. This study analyzes success and failure cases of Attention-driven visual prompting in object hallucination, revealing that preserving background context is crucial for mitigating object hallucination.

Paperid: 6797, https://arxiv.org/pdf/2502.08903.pdf

Abstract:
Vision-language models (VLMs) have achieved remarkable success in scene understanding and perception tasks, enabling robots to plan and execute actions adaptively in dynamic environments. However, most multimodal large language models lack robust 3D scene localization capabilities, limiting their effectiveness in fine-grained robotic operations. Additionally, challenges such as low recognition accuracy, inefficiency, poor transferability, and reliability hinder their use in precision tasks. To address these limitations, we propose a novel framework that integrates a 2D prompt synthesis module by mapping 2D images to point clouds, and incorporates a small language model (SLM) for supervising VLM outputs. The 2D prompt synthesis module enables VLMs, trained on 2D images and text, to autonomously extract precise 3D spatial information without manual intervention, significantly enhancing 3D scene understanding. Meanwhile, the SLM supervises VLM outputs, mitigating hallucinations and ensuring reliable, executable robotic control code generation. Our framework eliminates the need for retraining in new environments, thereby improving cost efficiency and operational robustness. Experimental results that the proposed framework achieved a 96.0\% Task Success Rate (TSR), outperforming other methods. Ablation studies demonstrated the critical role of both the 2D prompt synthesis module and the output supervision module (which, when removed, caused a 67\% TSR drop). These findings validate the framework's effectiveness in improving 3D recognition, task planning, and robotic task execution.

Paperid: 6798, https://arxiv.org/pdf/2502.08486.pdf

Abstract:
Referring Remote Sensing Image Segmentation (RRSIS) is critical for ecological monitoring, urban planning, and disaster management, requiring precise segmentation of objects in remote sensing imagery guided by textual descriptions. This task is uniquely challenging due to the considerable vision-language gap, the high spatial resolution and broad coverage of remote sensing imagery with diverse categories and small targets, and the presence of clustered, unclear targets with blurred edges. To tackle these issues, we propose \ours, a novel framework designed to bridge the vision-language gap, enhance multi-scale feature interaction, and improve fine-grained object differentiation. Specifically, \ours introduces: (1) the Bidirectional Spatial Correlation (BSC) for improved vision-language feature alignment, (2) the Target-Background TwinStream Decoder (T-BTD) for precise distinction between targets and non-targets, and (3) the Dual-Modal Object Learning Strategy (D-MOLS) for robust multimodal feature reconstruction. Extensive experiments on the benchmark datasets RefSegRS and RRSIS-D demonstrate that \ours achieves state-of-the-art performance. Specifically, \ours improves the overall IoU (oIoU) by 3.76 percentage points (80.57) and 1.44 percentage points (79.23) on the two datasets, respectively. Additionally, it outperforms previous methods in the mean IoU (mIoU) by 5.37 percentage points (67.95) and 1.84 percentage points (66.04), effectively addressing the core challenges of RRSIS with enhanced precision and robustness.

Paperid: 6799, https://arxiv.org/pdf/2502.06851.pdf

Authors:Adilzhan Adilkhanov, Amir Yelenov, Assylkhan Seitzhanov, Ayan Mazhitov, Azamat Abdikarimov, Danissa Sandykbayeva, Daryn Kenzhebek, Dinmukhammed Mukashev, Ilyas Umurbekov, Jabrail Chumakov, Kamila Spanova, Karina Burunchina, Madina Yergibay, Margulan Issa, Moldir Zabirova, Nurdaulet Zhuzbay, Nurlan Kabdyshev, Nurlan Zhaniyar, Rasul Yermagambet, Rustam Chibar, Saltanat Seitzhan, Soibkhon Khajikhanov, Tasbolat Taunyazov, Temirlan Galimzhanov, Temirlan Kaiyrbay, Tleukhan Mussin, Togzhan Syrymova, Valeriya Kostyukova, Yerkebulan Massalim, Yermakhan Kassym, Zerde Nurbayeva, Zhanat Kappassov

Abstract:
This paper presents an AI-generated review of Vision-Language-Action (VLA) models, summarizing key methodologies, findings, and future directions. The content is produced using large language models (LLMs) and is intended only for demonstration purposes. This work does not represent original research, but highlights how AI can help automate literature reviews. As AI-generated content becomes more prevalent, ensuring accuracy, reliability, and proper synthesis remains a challenge. Future research will focus on developing a structured framework for AI-assisted literature reviews, exploring techniques to enhance citation accuracy, source credibility, and contextual understanding. By examining the potential and limitations of LLM in academic writing, this study aims to contribute to the broader discussion of integrating AI into research workflows. This work serves as a preliminary step toward establishing systematic approaches for leveraging AI in literature review generation, making academic knowledge synthesis more efficient and scalable.

Paperid: 6800, https://arxiv.org/pdf/2502.05822.pdf

Abstract:
Search advertising is essential for merchants to reach the target users on short video platforms. Short video ads aligned with user search intents are displayed through relevance matching and bid ranking mechanisms. This paper focuses on improving query-to-video relevance matching to enhance the effectiveness of ranking in ad systems. Recent vision-language pre-training models have demonstrated promise in various multimodal tasks. However, their contribution to downstream query-video relevance tasks is limited, as the alignment between the pair of visual signals and text differs from the modeling of the triplet of the query, visual signals, and video text. In addition, our previous relevance model provides limited ranking capabilities, largely due to the discrepancy between the binary cross-entropy fine-tuning objective and the ranking objective. To address these limitations, we design a high-consistency multimodal relevance model (HCMRM). It utilizes a simple yet effective method to enhance the consistency between pre-training and relevance tasks. Specifically, during the pre-training phase, along with aligning visual signals and video text, several keywords are extracted from the video text as pseudo-queries to perform the triplet relevance modeling. For the fine-tuning phase, we introduce a hierarchical softmax loss, which enables the model to learn the order within labels while maximizing the distinction between positive and negative samples. This promotes the fusion ranking of relevance and bidding in the subsequent ranking stage. The proposed method has been deployed in the Kuaishou search advertising system for over a year, contributing to a 6.1% reduction in the proportion of irrelevant ads and a 1.4% increase in ad revenue.

Paperid: 6801, https://arxiv.org/pdf/2502.04359.pdf

Paperid: 6802, https://arxiv.org/pdf/2501.17586.pdf

Abstract:
Large vision-language models have revolutionized cross-modal object retrieval, but text-based person search (TBPS) remains a challenging task due to limited data and fine-grained nature of the task. Existing methods primarily focus on aligning image-text pairs into a common representation space, often disregarding the fact that real world positive image-text pairs share a varied degree of similarity in between them. This leads models to prioritize easy pairs, and in some recent approaches, challenging samples are discarded as noise during training. In this work, we introduce a boosting technique that dynamically identifies and emphasizes these challenging samples during training. Our approach is motivated from classical boosting technique and dynamically updates the weights of the weak positives, wherein, the rank-1 match does not share the identity of the query. The weight allows these misranked pairs to contribute more towards the loss and the network has to pay more attention towards such samples. Our method achieves improved performance across four pedestrian datasets, demonstrating the effectiveness of our proposed module.

Paperid: 6803, https://arxiv.org/pdf/2501.14905.pdf

Abstract:
Vision language models have achieved impressive results across various fields. However, adoption in remote sensing remains limited, largely due to the scarcity of paired image-text data. To bridge this gap, synthetic caption generation has gained interest, traditionally relying on rule-based methods that use metadata or bounding boxes. While these approaches provide some description, they often lack the depth needed to capture complex wide-area scenes. Large language models (LLMs) offer a promising alternative for generating more descriptive captions, yet they can produce generic outputs and are prone to hallucination. In this paper, we propose a new method to enhance vision-language datasets for remote sensing by integrating maps as external data sources, enabling the generation of detailed, context-rich captions. Additionally, we present methods to measure and mitigate hallucinations in LLM-generated text. We introduce fMoW-mm, a multimodal dataset incorporating satellite imagery, maps, metadata, and text annotations. We demonstrate its effectiveness for automatic target recognition in few-shot settings, achieving superior performance compared to other vision-language remote sensing datasets.

Paperid: 6804, https://arxiv.org/pdf/2501.13878.pdf

Abstract:
Advanced multimodal AI agents can now collaborate with users to solve challenges in the world. Yet, these emerging contextual AI systems rely on explicit communication channels between the user and system. We hypothesize that implicit communication of the user's interests and intent would reduce friction and improve user experience when collaborating with AI agents. In this work, we explore the potential of wearable eye tracking to convey signals about user attention. We measure the eye tracking signal quality requirements to effectively map gaze traces to physical objects, then conduct experiments that provide visual scanpath history as additional context when querying vision language models. Our results show that eye tracking provides high value as a user attention signal and can convey important context about the user's current task and interests, improving understanding of contextual AI agents.

Paperid: 6805, https://arxiv.org/pdf/2501.09294.pdf

Abstract:
Few-shot learning in medical image classification presents a significant challenge due to the limited availability of annotated data and the complex nature of medical imagery. In this work, we propose Adaptive Vision-Language Fine-tuning with Hierarchical Contrastive Alignment (HiCA), a novel framework that leverages the capabilities of Large Vision-Language Models (LVLMs) for medical image analysis. HiCA introduces a two-stage fine-tuning strategy, combining domain-specific pretraining and hierarchical contrastive learning to align visual and textual representations at multiple levels. We evaluate our approach on two benchmark datasets, Chest X-ray and Breast Ultrasound, achieving state-of-the-art performance in both few-shot and zero-shot settings. Further analyses demonstrate the robustness, generalizability, and interpretability of our method, with substantial improvements in performance compared to existing baselines. Our work highlights the potential of hierarchical contrastive strategies in adapting LVLMs to the unique challenges of medical imaging tasks.

Paperid: 6806, https://arxiv.org/pdf/2501.09236.pdf

Abstract:
Image captioning has become an essential Vision & Language research task. It is about predicting the most accurate caption given a specific image or video. The research community has achieved impressive results by continuously proposing new models and approaches to improve the overall model's performance. Nevertheless, despite increasing proposals, the performance metrics used to measure their advances have remained practically untouched through the years. A probe of that, nowadays metrics like BLEU, METEOR, CIDEr, and ROUGE are still very used, aside from more sophisticated metrics such as BertScore and ClipScore. Hence, it is essential to adjust how are measure the advances, limitations, and scopes of the new image captioning proposals, as well as to adapt new metrics to these new advanced image captioning approaches. This work proposes a new evaluation metric for the image captioning problem. To do that, first, it was generated a human-labeled dataset to assess to which degree the captions correlate with the image's content. Taking these human scores as ground truth, we propose a new metric, and compare it with several well-known metrics, from classical to newer ones. Outperformed results were also found, and interesting insights were presented and discussed.

Paperid: 6808, https://arxiv.org/pdf/2501.05081.pdf

Abstract:
In recent years, large language models have had a very impressive performance, which largely contributed to the development and application of artificial intelligence, and the parameters and performance of the models are still growing rapidly. In particular, multimodal large language models (MLLM) can combine multiple modalities such as pictures, videos, sounds, texts, etc., and have great potential in various tasks. However, most MLLMs require very high computational resources, which is a major challenge for most researchers and developers. In this paper, we explored the utility of small-scale MLLMs and applied small-scale MLLMs to the field of autonomous driving. We hope that this will advance the application of MLLMs in real-world scenarios.

Paperid: 6809, https://arxiv.org/pdf/2501.04675.pdf

Abstract:
Chart interpretation is crucial for visual data analysis, but accurately extracting information from charts poses significant challenges for automated models. This study investigates the fine-tuning of DEPLOT, a modality conversion module that translates the image of a plot or chart to a linearized table, on a custom dataset of 50,000 bar charts. The dataset comprises simple, stacked, and grouped bar charts, targeting the unique structural features of these visualizations. The finetuned DEPLOT model is evaluated against its base version using a test set of 1,000 images and two metrics: Relative Mapping Similarity (RMS), which measures categorical mapping accuracy, and Relative Number Set Similarity (RNSS), which evaluates numerical interpretation accuracy. To further explore the reasoning capabilities of large language models (LLMs), we curate an additional set of 100 bar chart images paired with question answer sets. Our findings demonstrate that providing a structured intermediate table alongside the image significantly enhances LLM reasoning performance compared to direct image queries.

Paperid: 6810, https://arxiv.org/pdf/2501.03786.pdf

Abstract:
Zero-shot anomaly detection (ZSAD) identifies anomalies without needing training samples from the target dataset, essential for scenarios with privacy concerns or limited data. Vision-language models like CLIP show potential in ZSAD but have limitations: relying on manually crafted fixed textual descriptions or anomaly prompts is time-consuming and prone to semantic ambiguity, and CLIP struggles with pixel-level anomaly segmentation, focusing more on global semantics than local details. To address these limitations, We introduce KAnoCLIP, a novel ZSAD framework that leverages vision-language models. KAnoCLIP combines general knowledge from a Large Language Model (GPT-3.5) and fine-grained, image-specific knowledge from a Visual Question Answering system (Llama3) via Knowledge-Driven Prompt Learning (KnPL). KnPL uses a knowledge-driven (KD) loss function to create learnable anomaly prompts, removing the need for fixed text prompts and enhancing generalization. KAnoCLIP includes the CLIP visual encoder with V-V attention (CLIP-VV), Bi-Directional Cross-Attention for Multi-Level Cross-Modal Interaction (Bi-CMCI), and Conv-Adapter. These components preserve local visual semantics, improve local cross-modal fusion, and align global visual features with textual information, enhancing pixel-level anomaly detection. KAnoCLIP achieves state-of-the-art performance in ZSAD across 12 industrial and medical datasets, demonstrating superior generalization compared to existing methods.

Paperid: 6811, https://arxiv.org/pdf/2501.03722.pdf

Abstract:
Accurate segmentation of pulmonary structures iscrucial in clinical diagnosis, disease study, and treatment planning. Significant progress has been made in deep learning-based segmentation techniques, but most require much labeled data for training. Consequently, developing precise segmentation methods that demand fewer labeled datasets is paramount in medical image analysis. The emergence of pre-trained vision-language foundation models, such as CLIP, recently opened the door for universal computer vision tasks. Exploiting the generalization ability of these pre-trained foundation models on downstream tasks, such as segmentation, leads to unexpected performance with a relatively small amount of labeled data. However, exploring these models for pulmonary artery-vein segmentation is still limited. This paper proposes a novel framework called Language-guided self-adaptive Cross-Attention Fusion Framework. Our method adopts pre-trained CLIP as a strong feature extractor for generating the segmentation of 3D CT scans, while adaptively aggregating the cross-modality of text and image representations. We propose a s pecially designed adapter module to fine-tune pre-trained CLIP with a self-adaptive learning strategy to effectively fuse the two modalities of embeddings. We extensively validate our method on a local dataset, which is the largest pulmonary artery-vein CT dataset to date and consists of 718 labeled data in total. The experiments show that our method outperformed other state-of-the-art methods by a large margin. Our data and code will be made publicly available upon acceptance.

Paperid: 6812, https://arxiv.org/pdf/2412.09875.pdf

Abstract:
Large Vision-Language Models (LVLMs) have demonstrated remarkable performance across a wide range of multimodal tasks. However, fine-tuning these models for domain-specific applications remains a computationally intensive challenge. This paper introduces State Space Memory Integration (SSMI), a novel approach for efficient fine-tuning of LVLMs. By integrating lightweight Mamba-based state space modules into the LVLM architecture, SSMI captures long-range dependencies and injects task-specific visual and sequential patterns effectively. Unlike traditional fine-tuning methods, SSMI requires only a fraction of the model's parameters to be updated, making it computationally efficient and scalable. Experiments on benchmark datasets, including COCO Captioning, VQA, and Flickr30k, demonstrate that SSMI achieves state-of-the-art performance while maintaining robustness and generalization capabilities. Comprehensive analysis further validates the advantages of SSMI in terms of efficiency, adaptability, and interpretability, positioning it as a compelling solution for fine-tuning large-scale vision-language models.

Paperid: 6813, https://arxiv.org/pdf/2412.05893.pdf

Abstract:
Human-interactive robotic systems, particularly autonomous vehicles (AVs), must effectively integrate human instructions into their motion planning. This paper introduces doScenes, a novel dataset designed to facilitate research on human-vehicle instruction interactions, focusing on short-term directives that directly influence vehicle motion. By annotating multimodal sensor data with natural language instructions and referentiality tags, doScenes bridges the gap between instruction and driving response, enabling context-aware and adaptive planning. Unlike existing datasets that focus on ranking or scene-level reasoning, doScenes emphasizes actionable directives tied to static and dynamic scene objects. This framework addresses limitations in prior research, such as reliance on simulated data or predefined action sets, by supporting nuanced and flexible responses in real-world scenarios. This work lays the foundation for developing learning strategies that seamlessly integrate human instructions into autonomous systems, advancing safe and effective human-vehicle collaboration for vision-language navigation. We make our data publicly available at https://www.github.com/rossgreer/doScenes

Paperid: 6814, https://arxiv.org/pdf/2412.02602.pdf

Abstract:
This paper analyzes the performance of Small Language Models (SLMs) and Vision Language Models (VLMs) and evaluates the trade-off between model performance and carbon emissions across 4 essential tasks: Image Captioning, Visual Question Answering (VQA), Dialogue Summarization and Text-to-SQL conversion. Various SLMs and VLMs belonging to the Qwen and LLaMA architecture family are chosen and variants based on model size in terms of the number of parameters, quantization level and fine-tuning parameters are evaluated. The model variant's performance and carbon emissions are calculated. To quantify the trade-off between model performance and carbon emissions, we introduce a novel metric called CEGI (Carbon Efficient Gain Index). This metric represents the carbon emission per unit percentage gain per million trainable parameters . This metric provides a normalized measure to compare model's efficiency in terms of performance improvement relative to their environmental cost. The experiment's outcome demonstrates that fine-tuning SLMs and VLMs can achieve performance levels comparable to Large Language Models (LLMs) while producing significantly less carbon emissions. Our findings suggest that the marginal gains in accuracy from larger models do not justify the substantial increase in carbon emissions. Leveraging lower-bit quantization levels, the proposed metric further enhances energy efficiency without compromising performance. This study highlights balancing high performance and environmental sustainability. It offers a valuable metric for selecting models suitable for environmentally-friendly AI development.

Paperid: 6815, https://arxiv.org/pdf/2411.17762.pdf

Abstract:
We introduce MUSE-VL, a Unified Vision-Language Model through Semantic discrete Encoding for multimodal understanding and generation. Recently, the research community has begun exploring unified models for visual generation and understanding. However, existing vision tokenizers (e.g., VQGAN) only consider low-level information, which makes it difficult to align with language tokens. This results in high training complexity and necessitates a large amount of training data to achieve optimal performance. Additionally, their performance is still far from dedicated understanding models. This paper proposes Semantic Discrete Encoding (SDE), which effectively aligns the information of visual tokens and language tokens by adding semantic constraints to the visual tokenizer. This greatly reduces the amount of training data and improves the performance of the unified model. With the same LLM size, our method improved the understanding performance by 4.8% compared to the previous SOTA Emu3 and surpassed the dedicated understanding model LLaVA-NeXT 34B by 3.7%. Our model also surpasses the existing unified models on visual generation benchmarks.

Paperid: 6816, https://arxiv.org/pdf/2411.09933.pdf

Abstract:
With the rapid advancement of large language models (LLMs), foundational models (FMs) have seen significant advancements. Healthcare is one of the most crucial application areas for these FMs, given the significant time and effort required for physicians to analyze large volumes of patient data. Recent efforts have focused on adapting multimodal FMs to the medical domain through techniques like instruction-tuning, leading to the development of medical foundation models (MFMs). However, these approaches typically require large amounts of training data to effectively adapt models to the medical field. Moreover, most existing models are trained on English datasets, limiting their practicality in non-English-speaking regions where healthcare professionals and patients are not always fluent in English. The need for translation introduces additional costs and inefficiencies. To address these challenges, we propose a \textbf{J}apanese \textbf{Radi}ology report generation model enhanced by \textbf{Evo}lutionary optimization of model merging (JRadiEvo). This is the first attempt to extend a non-medical vision-language foundation model to the medical domain through evolutionary optimization of model merging. We successfully created a model that generates accurate Japanese reports from X-ray images using only 50 translated samples from publicly available data. This model, developed with highly efficient use of limited data, outperformed leading models from recent research trained on much larger datasets. Additionally, with only 8 billion parameters, this relatively compact foundation model can be deployed locally within hospitals, making it a practical solution for environments where APIs and other external services cannot be used due to strict privacy and security requirements.

Paperid: 6817, https://arxiv.org/pdf/2411.06074.pdf

Abstract:
Recently, large vision language models (VLMs) have made significant strides in visual language capabilities through visual instruction tuning, showing great promise in the field of remote sensing image interpretation. However, existing remote sensing vision language models (RSVLMs) often fall short in capturing the complex characteristics of remote sensing scenes, as they typically rely on low resolution, single scale visual features and simplistic methods to map visual features to language features. In this paper, we present Aquila, an advanced visual language foundation model designed to enable richer visual feature representation and more precise visual-language feature alignment for remote sensing images. Our approach introduces a learnable Hierarchical Spatial Feature Integration (SFI) module that supports high resolution image inputs and aggregates multi scale visual features, allowing for the detailed representation of complex visual information. Additionally, the SFI module is repeatedly integrated into the layers of the large language model (LLM) to achieve deep visual language feature alignment, without compromising the model's performance in natural language processing tasks. These innovations, capturing detailed visual effects through higher resolution and multi scale input, and enhancing feature alignment significantly improve the model's ability to learn from image text data. We validate the effectiveness of Aquila through extensive quantitative experiments and qualitative analyses, demonstrating its superior performance.

Paperid: 6818, https://arxiv.org/pdf/2411.03491.pdf

Abstract:
We present a novel Automatic Target Recognition (ATR) system using open-vocabulary object detection and classification models. A primary advantage of this approach is that target classes can be defined just before runtime by a non-technical end user, using either a few natural language text descriptions of the target, or a few image exemplars, or both. Nuances in the desired targets can be expressed in natural language, which is useful for unique targets with little or no training data. We also implemented a novel combination of several techniques to improve performance, such as leveraging the additional information in the sequence of overlapping frames to perform tubelet identification (i.e., sequential bounding box matching), bounding box re-scoring, and tubelet linking. Additionally, we developed a technique to visualize the aggregate output of many overlapping frames as a mosaic of the area scanned during the aerial surveillance or reconnaissance, and a kernel density estimate (or heatmap) of the detected targets. We initially applied this ATR system to the use case of detecting and clearing unexploded ordinance on airfield runways and we are currently extending our research to other real-world applications.

Paperid: 6819, https://arxiv.org/pdf/2411.00634.pdf

Abstract:
Ensuring usability is crucial for the success of mobile apps. Usability issues can compromise user experience and negatively impact the perceived app quality. This paper presents UX-LLM, a novel tool powered by a Large Vision-Language Model that predicts usability issues in iOS apps. To evaluate the performance of UX-LLM, we predicted usability issues in two open-source apps of a medium complexity and asked two usability experts to assess the predictions. We also performed traditional usability testing and expert review for both apps and compared the results to those of UX-LLM. UX-LLM demonstrated precision ranging from 0.61 and 0.66 and recall between 0.35 and 0.38, indicating its ability to identify valid usability issues, yet failing to capture the majority of issues. Finally, we conducted a focus group with an app development team of a capstone project developing a transit app for visually impaired persons. The focus group expressed positive perceptions of UX-LLM as it identified unknown usability issues in their app. However, they also raised concerns about its integration into the development workflow, suggesting potential improvements. Our results show that UX-LLM cannot fully replace traditional usability evaluation methods but serves as a valuable supplement particularly for small teams with limited resources, to identify issues in less common user paths, due to its ability to inspect the source code.

Paperid: 6820, https://arxiv.org/pdf/2410.22233.pdf

Abstract:
Contextual advertising serves ads that are aligned to the content that the user is viewing. The rapid growth of video content on social platforms and streaming services, along with privacy concerns, has increased the need for contextual advertising. Placing the right ad in the right context creates a seamless and pleasant ad viewing experience, resulting in higher audience engagement and, ultimately, better ad monetization. From a technology standpoint, effective contextual advertising requires a video retrieval system capable of understanding complex video content at a very granular level. Current text-to-video retrieval models based on joint multimodal training demand large datasets and computational resources, limiting their practicality and lacking the key functionalities required for ad ecosystem integration. We introduce ContextIQ, a multimodal expert-based video retrieval system designed specifically for contextual advertising. ContextIQ utilizes modality-specific experts-video, audio, transcript (captions), and metadata such as objects, actions, emotion, etc.-to create semantically rich video representations. We show that our system, without joint training, achieves better or comparable results to state-of-the-art models and commercial solutions on multiple text-to-video retrieval benchmarks. Our ablation studies highlight the benefits of leveraging multiple modalities for enhanced video retrieval accuracy instead of using a vision-language model alone. Furthermore, we show how video retrieval systems such as ContextIQ can be used for contextual advertising in an ad ecosystem while also addressing concerns related to brand safety and filtering inappropriate content.

Paperid: 6821, https://arxiv.org/pdf/2410.15549.pdf

Abstract:
Vision-Language-Action (VLA) models are receiving increasing attention for their ability to enable robots to perform complex tasks by integrating visual context with linguistic commands. However, achieving efficient real-time performance remains challenging due to the high computational demands of existing models. To overcome this, we propose Dual Process VLA (DP-VLA), a hierarchical framework inspired by dual-process theory. DP-VLA utilizes a Large System 2 Model (L-Sys2) for complex reasoning and decision-making, while a Small System 1 Model (S-Sys1) handles real-time motor control and sensory processing. By leveraging Vision-Language Models (VLMs), the L-Sys2 operates at low frequencies, reducing computational overhead, while the S-Sys1 ensures fast and accurate task execution. Experimental results on the RoboCasa dataset demonstrate that DP-VLA achieves faster inference and higher task success rates, providing a scalable solution for advanced robotic applications.

Paperid: 6822, https://arxiv.org/pdf/2410.14991.pdf

Abstract:
Visual Question Answer (VQA) poses the problem of answering a natural language question about a visual context. Bangla, despite being a widely spoken language, is considered low-resource in the realm of VQA due to the lack of proper benchmarks, challenging models known to be performant in other languages. Furthermore, existing Bangla VQA datasets offer little regional relevance and are largely adapted from their foreign counterparts. To address these challenges, we introduce a large-scale Bangla VQA dataset, ChitroJera, totaling over 15k samples from diverse and locally relevant data sources. We assess the performance of text encoders, image encoders, multimodal models, and our novel dual-encoder models. The experiments reveal that the pre-trained dual-encoders outperform other models of their scale. We also evaluate the performance of current large vision language models (LVLMs) using prompt-based techniques, achieving the overall best performance. Given the underdeveloped state of existing datasets, we envision ChitroJera expanding the scope of Vision-Language tasks in Bangla.

Paperid: 6823, https://arxiv.org/pdf/2410.14587.pdf

Abstract:
Deep generative models are becoming increasingly used as tools for financial analysis. However, it is unclear how these models will influence financial markets, especially when they infer financial value in a semi-autonomous way. In this work, we explore the interplay between deep generative models and market dynamics. We develop a form of virtual traders that use deep generative models to make buy/sell decisions, which we term neuro-symbolic traders, and expose them to a virtual market. Under our framework, neuro-symbolic traders are agents that use vision-language models to discover a model of the fundamental value of an asset. Agents develop this model as a stochastic differential equation, calibrated to market data using gradient descent. We test our neuro-symbolic traders on both synthetic data and real financial time series, including an equity stock, commodity, and a foreign exchange pair. We then expose several groups of neuro-symbolic traders to a virtual market environment. This market environment allows for feedback between the traders belief of the underlying value to the observed price dynamics. We find that this leads to price suppression compared to the historical data, highlighting a future risk to market stability. Our work is a first step towards quantifying the effect of deep generative agents on markets dynamics and sets out some of the potential risks and benefits of this approach in the future.

Paperid: 6824, https://arxiv.org/pdf/2410.12051.pdf

Abstract:
In this paper, we introduce a novel system designed to enhance customer service in the financial and retail sectors through a context-aware 3D virtual agent, utilizing Mixed Reality (MR) and Vision Language Models (VLMs). Our approach focuses on enabling data-driven and empathetic interactions that ensure customer satisfaction by introducing situational awareness of the physical location, personalized interactions based on customer profiles, and rigorous privacy and security standards. We discuss our design considerations critical for deployment in real-world customer service environments, addressing challenges in user data management and sensitive information handling. We also outline the system architecture and key features unique to banking and retail environments. Our work demonstrates the potential of integrating MR and VLMs in service industries, offering practical insights in customer service delivery while maintaining high standards of security and personalization.

Paperid: 6825, https://arxiv.org/pdf/2410.11871.pdf

Abstract:
We present an UI agent for user interface (UI) interaction tasks, using Vision-Language Model Florence-2-Base. The agent's primary task is identifying the screen coordinates of the UI element corresponding to the user's command. It demonstrates very strong performance on Screenspot and OmniAct annotations, while maintaining a very small size of 0.27B parameters and minimal latency. Moreover, training needs small compute budget of 56 GPU-hours (worth about 40 USD). Relevant improvement comes from vision-specific multi-task training and MLLM-based data augmentation. We hope that decreased needs for expensive compute resources and manually annotated data will allow to facilitate more inclusive and sustainable research of UI agents.

Paperid: 6826, https://arxiv.org/pdf/2410.11564.pdf

Abstract:
Affordance understanding, the task of identifying actionable regions on 3D objects, plays a vital role in allowing robotic systems to engage with and operate within the physical world. Although Visual Language Models (VLMs) have excelled in high-level reasoning and long-horizon planning for robotic manipulation, they still fall short in grasping the nuanced physical properties required for effective human-robot interaction. In this paper, we introduce PAVLM (Point cloud Affordance Vision-Language Model), an innovative framework that utilizes the extensive multimodal knowledge embedded in pre-trained language models to enhance 3D affordance understanding of point cloud. PAVLM integrates a geometric-guided propagation module with hidden embeddings from large language models (LLMs) to enrich visual semantics. On the language side, we prompt Llama-3.1 models to generate refined context-aware text, augmenting the instructional input with deeper semantic cues. Experimental results on the 3D-AffordanceNet benchmark demonstrate that PAVLM outperforms baseline methods for both full and partial point clouds, particularly excelling in its generalization to novel open-world affordance tasks of 3D objects. For more information, visit our project site: pavlm-source.github.io.

Paperid: 6827, https://arxiv.org/pdf/2410.03054.pdf

Abstract:
This letter proposes a method of global localization on a map with semantic object landmarks. One of the most promising approaches for localization on object maps is to use semantic graph matching using landmark descriptors calculated from the distribution of surrounding objects. These descriptors are vulnerable to misclassification and partial observations. Moreover, many existing methods rely on inlier extraction using RANSAC, which is stochastic and sensitive to a high outlier rate. To address the former issue, we augment the correspondence matching using Vision Language Models (VLMs). Landmark discriminability is improved by VLM embeddings, which are independent of surrounding objects. In addition, inliers are estimated deterministically using a graph-theoretic approach. We also incorporate pose calculation using the weighted least squares considering correspondence similarity and observation completeness to improve the robustness. We confirmed improvements in matching and pose estimation accuracy through experiments on ScanNet and TUM datasets.

Paperid: 6828, https://arxiv.org/pdf/2409.20429.pdf

Abstract:
Large Vision-Language Models (LVLMs) have shown remarkable performance on many visual-language tasks. However, these models still suffer from multimodal hallucination, which means the generation of objects or content that violates the images. Many existing work detects hallucination by directly judging whether an object exists in an image, overlooking the association between the object and semantics. To address this issue, we propose Hierarchical Feedback Learning with Vision-enhanced Penalty Decoding (HELPD). This framework incorporates hallucination feedback at both object and sentence semantic levels. Remarkably, even with a marginal degree of training, this approach can alleviate over 15% of hallucination. Simultaneously, HELPD penalizes the output logits according to the image attention window to avoid being overly affected by generated text. HELPD can be seamlessly integrated with any LVLMs. Our experiments demonstrate that the proposed framework yields favorable results across multiple hallucination benchmarks. It effectively mitigates hallucination for different LVLMs and concurrently improves their text generation quality.

Paperid: 6829, https://arxiv.org/pdf/2409.15658.pdf

Abstract:
Long-horizon embodied planning underpins embodied AI. To accomplish long-horizon tasks, one of the most feasible ways is to decompose abstract instructions into a sequence of actionable steps. Foundation models still face logical errors and hallucinations in long-horizon planning, unless provided with highly relevant examples to the tasks. However, providing highly relevant examples for any random task is unpractical. Therefore, we present ReLEP, a novel framework for Real-time Long-horizon Embodied Planning. ReLEP can complete a wide range of long-horizon tasks without in-context examples by learning implicit logical inference through fine-tuning. The fine-tuned large vision-language model formulates plans as sequences of skill functions. These functions are selected from a carefully designed skill library. ReLEP is also equipped with a Memory module for plan and status recall, and a Robot Configuration module for versatility across robot types. In addition, we propose a data generation pipeline to tackle dataset scarcity. When constructing the dataset, we considered the implicit logical relationships, enabling the model to learn implicit logical relationships and dispel hallucinations. Through comprehensive evaluations across various long-horizon tasks, ReLEP demonstrates high success rates and compliance to execution even on unseen tasks and outperforms state-of-the-art baseline methods.

Paperid: 6830, https://arxiv.org/pdf/2409.15292.pdf

Abstract:
We introduce SketcherX, a novel robotic system for personalized portrait drawing through interactive human-robot engagement. Unlike traditional robotic art systems that rely on analog printing techniques, SketcherX captures and processes facial images to produce vectorized drawings in a distinctive, human-like artistic style. The system comprises two 6-axis robotic arms : a face robot, which is equipped with a head-mounted camera and Large Language Model (LLM) for real-time interaction, and a drawing robot, utilizing a fine-tuned Stable Diffusion model, ControlNet, and Vision-Language models for dynamic, stylized drawing. Our contributions include the development of a custom Vector Low Rank Adaptation model (LoRA), enabling seamless adaptation to various artistic styles, and integrating a pair-wise fine-tuning approach to enhance stroke quality and stylistic accuracy. Experimental results demonstrate the system's ability to produce high-quality, personalized portraits within two minutes, highlighting its potential as a new paradigm in robotic creativity. This work advances the field of robotic art by positioning robots as active participants in the creative process, paving the way for future explorations in interactive, human-robot artistic collaboration.

Paperid: 6831, https://arxiv.org/pdf/2409.14846.pdf

Abstract:
The Large Vision-Language Model (LVLM) integrates computer vision and natural language processing techniques, offering substantial application potential. However, these models demand extensive resources during inference. Adaptive attention techniques can dynamically reduce computational redundancy and thus improve efficiency. Although current adaptive attention methods significantly reduce the memory requirements of Transformer-based language models, they are not tailored for LVLMs. We observe that LVLMs generate responses from both remote image tokens and local text tokens, and different modalities have different attention patterns. This observation inspires us to manage the attention for each modality separately. Specifically, for visual input, we store the cache of potentially useful information but only compute the most critical parts. For language input, we care more about local information. Based on our observation and analysis of vision-language attention patterns, we develop A-VL, a plug-and-play adaptive attention tailored for LVLM inference. Extensive evaluations on three vision-language tasks and five datasets show the effectiveness of our designs. Our approach A-VL outperforms existing adaptive attention methods in reducing memory usage and computational load without compromising performance.

Paperid: 6832, https://arxiv.org/pdf/2409.14109.pdf

Abstract:
Video anomaly detection is a subject of great interest across industrial and academic domains due to its crucial role in computer vision applications. However, the inherent unpredictability of anomalies and the scarcity of anomaly samples present significant challenges for unsupervised learning methods. To overcome the limitations of unsupervised learning, which stem from a lack of comprehensive prior knowledge about anomalies, we propose VLAVAD (Video-Language Models Assisted Anomaly Detection). Our method employs a cross-modal pre-trained model that leverages the inferential capabilities of large language models (LLMs) in conjunction with a Selective-Prompt Adapter (SPA) for selecting semantic space. Additionally, we introduce a Sequence State Space Module (S3M) that detects temporal inconsistencies in semantic features. By mapping high-dimensional visual features to low-dimensional semantic ones, our method significantly enhance the interpretability of unsupervised anomaly detection. Our proposed approach effectively tackles the challenge of detecting elusive anomalies that are hard to discern over periods, achieving SOTA on the challenging ShanghaiTech dataset.

Paperid: 6833, https://arxiv.org/pdf/2409.12969.pdf

Abstract:
Visual perspective-taking (VPT), the ability to understand the viewpoint of another person, enables individuals to anticipate the actions of other people. For instance, a driver can avoid accidents by assessing what pedestrians see. Humans typically develop this skill in early childhood, but it remains unclear whether the recently emerging Vision Language Models (VLMs) possess such capability. Furthermore, as these models are increasingly deployed in the real world, understanding how they perform nuanced tasks like VPT becomes essential. In this paper, we introduce two manually curated datasets, Isle-Bricks and Isle-Dots for testing VPT skills, and we use it to evaluate 12 commonly used VLMs. Across all models, we observe a significant performance drop when perspective-taking is required. Additionally, we find performance in object detection tasks is poorly correlated with performance on VPT tasks, suggesting that the existing benchmarks might not be sufficient to understand this problem. The code and the dataset will be available at https://sites.google.com/view/perspective-taking

Paperid: 6834, https://arxiv.org/pdf/2409.02958.pdf

Abstract:
Large pre-trained vision-language models, such as CLIP, have demonstrated state-of-the-art performance across a wide range of image classification tasks, without requiring retraining. Few-shot CLIP is competitive with existing specialized architectures that were trained on the downstream tasks. Recent research demonstrates that the performance of CLIP can be further improved using lightweight adaptation approaches. However, previous methods adapt different modalities of the CLIP model individually, ignoring the interactions and relationships between visual and textual representations. In this work, we propose Multi-Modal Adapter, an approach for Multi-Modal adaptation of CLIP. Specifically, we add a trainable Multi-Head Attention layer that combines text and image features to produce an additive adaptation of both. Multi-Modal Adapter demonstrates improved generalizability, based on its performance on unseen classes compared to existing adaptation methods. We perform additional ablations and investigations to validate and interpret the proposed approach.

Paperid: 6835, https://arxiv.org/pdf/2409.02561.pdf

Abstract:
Vision-language navigation (VLN) is a critical domain within embedded intelligence, requiring agents to navigate 3D environments based on natural language instructions. Traditional VLN research has focused on improving environmental understanding and decision accuracy. However, these approaches often exhibit a significant performance gap when agents are deployed in novel environments, mainly due to the limited diversity of training data. Expanding datasets to cover a broader range of environments is impractical and costly. We propose the Vision-Language Navigation with Continual Learning (VLNCL) paradigm to address this challenge. In this paradigm, agents incrementally learn new environments while retaining previously acquired knowledge. VLNCL enables agents to maintain an environmental memory and extract relevant knowledge, allowing rapid adaptation to new environments while preserving existing information. We introduce a novel dual-loop scenario replay method (Dual-SR) inspired by brain memory replay mechanisms integrated with VLN agents. This method facilitates consolidating past experiences and enhances generalization across new tasks. By utilizing a multi-scenario memory buffer, the agent efficiently organizes and replays task memories, thereby bolstering its ability to adapt quickly to new environments and mitigating catastrophic forgetting. Our work pioneers continual learning in VLN agents, introducing a novel experimental setup and evaluation metrics. We demonstrate the effectiveness of our approach through extensive evaluations and establish a benchmark for the VLNCL paradigm. Comparative experiments with existing continual learning and VLN methods show significant improvements, achieving state-of-the-art performance in continual learning ability and highlighting the potential of our approach in enabling rapid adaptation while preserving prior knowledge.

Paperid: 6836, https://arxiv.org/pdf/2409.00301.pdf

Abstract:
In recent years, there has been a notable increase in the development of autonomous vehicle (AV) technologies aimed at improving safety in transportation systems. While AVs have been deployed in the real-world to some extent, a full-scale deployment requires AVs to robustly navigate through challenges like heavy rain, snow, low lighting, construction zones and GPS signal loss in tunnels. To be able to handle these specific challenges, an AV must reliably recognize the physical attributes of the environment in which it operates. In this paper, we define context recognition as the task of accurately identifying environmental attributes for an AV to appropriately deal with them. Specifically, we define 24 environmental contexts capturing a variety of weather, lighting, traffic and road conditions that an AV must be aware of. Motivated by the need to recognize environmental contexts, we create a context recognition dataset called DrivingContexts with more than 1.6 million context-query pairs relevant for an AV. Since traditional supervised computer vision approaches do not scale well to a variety of contexts, we propose a framework called ContextVLM that uses vision-language models to detect contexts using zero- and few-shot approaches. ContextVLM is capable of reliably detecting relevant driving contexts with an accuracy of more than 95% on our dataset, while running in real-time on a 4GB Nvidia GeForce GTX 1050 Ti GPU on an AV with a latency of 10.5 ms per query.

Paperid: 6837, https://arxiv.org/pdf/2601.21815.pdf

Abstract:
Understanding how media rhetoric shapes audience engagement is crucial in the attention economy. This study examines how moral emotional framing by mainstream news channels on YouTube influences user behavior across Korea and the United States. To capture the platform's multimodal nature, combining thumbnail images and video titles, we develop a multimodal moral emotion classifier by fine tuning a vision language model. The model is trained on human annotated multimodal datasets in both languages and applied to approximately 400,000 videos from major news outlets. We analyze engagement levels including views, likes, and comments, representing increasing degrees of commitment. The results show that other condemning rhetoric expressions of moral outrage that criticize others morally consistently increase all forms of engagement across cultures, with effects ranging from passive viewing to active commenting. These findings suggest that moral outrage is a particularly effective emotional strategy, attracting not only attention but also active participation. We discuss concerns about the potential misuse of other condemning rhetoric, as such practices may deepen polarization by reinforcing in group and out group divisions. To facilitate future research and ensure reproducibility, we publicly release our Korean and English multimodal moral emotion classifiers.

Paperid: 6838, https://arxiv.org/pdf/2601.20622.pdf

Abstract:
Sketching provides an intuitive way to convey dynamic intent in animation authoring (i.e., how elements change over time and space), making it a natural medium for automatic content creation. Yet existing approaches often constrain sketches to fixed command tokens or predefined visual forms, overlooking their freeform nature and the central role of humans in shaping intention. To address this, we introduce an interaction paradigm where users convey dynamic intent to a vision-language model via free-form sketching, instantiated here in a sketch storyboard to motion graphics workflow. We implement an interface and improve it through a three-stage study with 24 participants. The study shows how sketches convey motion with minimal input, how their inherent ambiguity requires users to be involved for clarification, and how sketches can visually guide video refinement. Our findings reveal the potential of sketch and AI interaction to bridge the gap between intention and outcome, and demonstrate its applicability to 3D animation and video generation.

Paperid: 6839, https://arxiv.org/pdf/2601.20262.pdf

Abstract:
The growing demand for real-time robotic deployment necessitates fast and on-device inference for vision-language-action (VLA) models. Within the VLA literature, efficiency has been extensively studied at the token level, such as visual token pruning. In contrast, systematic transformer layer reduction has received limited attention and, to the best of our knowledge, has not been explored for flow-based VLA models under knowledge distillation. In this work, we propose Shallow-pi, a principled knowledge distillation framework that aggressively reduces the transformer depth of both the VLM backbone and the flow-based action head, compressing the model from 18 to 6 layers. Shallow-pi achieves over two times faster inference with less than one percent absolute drop in success rate on standard manipulation benchmarks, establishing state-of-the-art performance among reduced VLA models. Crucially, we validate our approach through industrial-scale real-world experiments on Jetson Orin and Jetson Thor across multiple robot platforms, including humanoid systems, in complex and dynamic manipulation scenarios.

Paperid: 6840, https://arxiv.org/pdf/2601.20224.pdf

Abstract:
Vision-Language Pre-Trained models, notably CLIP, that utilize contrastive learning have proven highly adept at extracting generalizable visual features. To inherit the well-learned knowledge of VLP models for downstream tasks, several approaches aim to adapt them efficiently with limited supervision. However, these methods either suffer from limited performance, excessive learnable parameters, or extended training times, all of which hinder their effectiveness in adapting the CLIP model to downstream tasks. In this work, we propose a simple yet efficient and effective method called \textit{\textbf{F}eature \textbf{P}rojection \textbf{L}earning(FPL)} to address these problems. Specifically, we develop a projection model that projects class prototype features into the query image feature space and reconstructs the query image feature map. The negative average squared reconstruction error is used as the class score. In this way, we transform the classification problem into a feature projection problem. The final output of this method is a combination of the prediction from the projection model and the original pre-trained CLIP. Comprehensive empirical evaluations confirm that FPL delivers superior accuracy, surpassing the current state-of-the-art methods by a substantial margin.

Paperid: 6841, https://arxiv.org/pdf/2601.19203.pdf

Abstract:
Olfactory cues can enhance immersion in interactive media, yet smell remains rare because it is difficult to author and synchronize with dynamic video. Prior olfactory interfaces rely on designer triggers and fixed event-to-odor mappings that do not scale to unconstrained content. This work examines whether semantic planning for smell is intelligible to people before physical scent delivery. We present a video-to-scent planning pipeline that separates visual semantic extraction using a vision-language model from semantic-to-olfactory inference using a large language model. Two survey studies compare system-generated scent plans with over-inclusive and naive baselines. Results show consistent preference for plans that prioritize perceptually salient cues and align scent changes with visible actions, supporting semantic planning as a foundation for future olfactory media systems.

Paperid: 6842, https://arxiv.org/pdf/2601.18356.pdf

Abstract:
Medical vision-language models (VLMs) achieve strong performance in diagnostic reporting and image-text alignment, yet their underlying reasoning mechanisms remain fundamentally correlational, exhibiting reliance on superficial statistical associations that fail to capture the causal pathophysiological mechanisms central to clinical decision-making. This limitation makes them fragile, prone to hallucinations, and sensitive to dataset biases. Retrieval-augmented generation (RAG) offers a partial remedy by grounding predictions in external knowledge. However, conventional RAG depends on semantic similarity, introducing new spurious correlations. We propose Multimodal Causal Retrieval-Augmented Generation, a framework that integrates causal inference principles with multimodal retrieval. It retrieves clinically relevant exemplars and causal graphs from external sources, conditioning model reasoning on counterfactual and interventional evidence rather than correlations alone. Applied to radiology report generation, diagnosis prediction, and visual question answering, it improves factual accuracy, robustness to distribution shifts, and interpretability. Our results highlight causal retrieval as a scalable path toward medical VLMs that think beyond pattern matching, enabling trustworthy multimodal reasoning in high-stakes clinical settings.

Paperid: 6843, https://arxiv.org/pdf/2601.17818.pdf

Abstract:
Large Vision-Language Models (LVLMs) incur high computational costs due to significant redundancy in their visual tokens. To effectively reduce this cost, researchers have proposed various visual token pruning methods. However, existing methods are generally limited, either losing critical visual information prematurely due to pruning in the vision encoder, or leading to information redundancy among the selected tokens due to pruning in the Large Language Models (LLMs). To address these challenges, we propose a Visual and Textual Semantic Collaborative Pruning framework (ViTCoP) that combines redundancy filtering in the vision encoder with step-wise co-pruning within the LLM based on its hierarchical characteristics, to efficiently preserve critical and informationally diverse visual tokens. Meanwhile, to ensure compatibility with acceleration techniques like FlashAttention, we introduce the L2 norm of K-vectors as the token saliency metric in the LLM. Extensive experiments on various Large Vision-Language Models demonstrate that ViTCoP not only achieves state-of-the-art performance surpassing existing methods on both image and video understanding tasks, but also significantly reduces model inference latency and GPU memory consumption. Notably, its performance advantage over other methods becomes even more pronounced under extreme pruning rates.

Paperid: 6844, https://arxiv.org/pdf/2601.17348.pdf

Abstract:
Vision-language models (VLMs) are increasingly deployed in socially sensitive applications, yet their behavior with respect to disability remains underexplored. We study disability aware descriptions for person centric images, where models often transition from evidence grounded factual description to interpretation shift including introduction of unsupported inferences beyond observable visual evidence. To systematically analyze this phenomenon, we introduce a benchmark based on paired Neutral Prompts (NP) and Disability-Contextualised Prompts (DP) and evaluate 15 state-of-the-art open- and closed-source VLMs under a zero-shot setting across 9 disability categories. Our evaluation framework treats interpretive fidelity as core objective and combines standard text-based metrics capturing affective degradation through shifts in sentiment, social regard and response length with an LLM-as-judge protocol, validated by annotators with lived experience of disability. We find that introducing disability context consistently degrades interpretive fidelity, inducing interpretation shifts characterised by speculative inference, narrative elaboration, affective degradation and deficit oriented framing. These effects are further amplified along race and gender dimension. Finally, we demonstrate targeted prompting and preference fine-tuning effectively improves interpretive fidelity and reduces substantially interpretation shifts.

Paperid: 6845, https://arxiv.org/pdf/2601.17123.pdf

Abstract:
We introduce and explore a new multimodal input representation for vision-language models: acoustic field video. Unlike conventional video (RGB with stereo/mono audio), our video stream provides a spatially grounded visualization of sound intensity across a scene, offering a new and powerful dimension of perceptual understanding. Our real-time pipeline uses low-cost beamforming microphone arrays that are already common in smart speakers and increasingly present in robotics and XR headsets, yet this sensing capability remains unutilized for scene understanding. To assess the value of spatial acoustic information, we constructed an evaluation set of 402 question-answer scenes, comparing a state-of-the-art VLM given conventional video with and without paired acoustic field video. Results show a clear and consistent improvement when incorporating spatial acoustic data; the VLM we test improves from 38.3% correct to 67.4%. Our findings highlight that many everyday scene understanding tasks remain underconstrained when relying solely on visual and audio input, and that acoustic field data provides a promising and practical direction for multimodal reasoning. A video demo is available at https://daehwakim.com/seeingsound

Paperid: 6846, https://arxiv.org/pdf/2601.17093.pdf

Abstract:
Comparing neural network representations is essential for understanding and validating models in scientific applications. Existing methods, however, often provide a limited view. We propose the Triangle of Similarity, a framework that combines three complementary perspectives: static representational similarity (CKA/Procrustes), functional similarity (Linear Mode Connectivity or Predictive Similarity), and sparsity similarity (robustness under pruning). Analyzing a range of CNNs, Vision Transformers, and Vision-Language Models using both in-distribution (ImageNetV2) and out-of-distribution (CIFAR-10) testbeds, our initial findings suggest that: (1) architectural family is a primary determinant of representational similarity, forming distinct clusters; (2) CKA self-similarity and task accuracy are strongly correlated during pruning, though accuracy often degrades more sharply; and (3) for some model pairs, pruning appears to regularize representations, exposing a shared computational core. This framework offers a more holistic approach for assessing whether models have converged on similar internal mechanisms, providing a useful tool for model selection and analysis in scientific research.

Paperid: 6847, https://arxiv.org/pdf/2601.16582.pdf

Abstract:
Composed Video Retrieval (CoVR) facilitates video retrieval by combining visual and textual queries. However, existing CoVR frameworks typically fuse multimodal inputs in a single stage, achieving only marginal gains over initial baseline. To address this, we propose a novel CoVR framework that leverages the representational power of Vision Language Models (VLMs). Our framework incorporates a novel cross-attention module X-Aligner, composed of cross-attention layers that progressively fuse visual and textual inputs and align their multimodal representation with that of the target video. To further enhance the representation of the multimodal query, we incorporate the caption of the visual query as an additional input. The framework is trained in two stages to preserve the pretrained VLM representation. In the first stage, only the newly introduced module is trained, while in the second stage, the textual query encoder is also fine-tuned. We implement our framework on top of BLIP-family architecture, namely BLIP and BLIP-2, and train it on the Webvid-CoVR data set. In addition to in-domain evaluation on Webvid-CoVR-Test, we perform zero-shot evaluations on the Composed Image Retrieval (CIR) data sets CIRCO and Fashion-IQ. Our framework achieves state-of-the-art performance on CoVR obtaining a Recall@1 of 63.93% on Webvid-CoVR-Test, and demonstrates strong zero-shot generalization on CIR tasks.

Paperid: 6848, https://arxiv.org/pdf/2601.16218.pdf

Abstract:
Despite state-of-the-art vision-language models (VLMs) have demonstrated strong reasoning capabilities, their performance in multilingual mathematical reasoning remains underexplored, particularly when compared to human performance. To bridge this gap, we introduce M3Kang, the first massively multilingual, multimodal mathematical reasoning dataset for VLMs. It is derived from the Kangaroo Math Competition, the world's largest mathematics contest, which annually engages over six million participants under the age of 18 across more than 90 countries. M3Kang includes 1,747 unique multiple-choice problems organized by grade-level difficulty, with translations into 108 culturally diverse languages, some of them including diagrams essential for solving them. Using this dataset, we conduct extensive benchmarking on both closed- and open-source SOTA models. We observe that, despite recent advances, models still struggle with basic math and diagram-based reasoning, with performance scaling with language presence and model size, but not with grade level. We also find that multilingual techniques can be effectively extended to the multimodal setting, resulting in significant improvements over baseline approaches. Our analysis also incorporates performance data from over 68,000 students, enabling direct comparison with human performance. We are open-sourcing M3Kang, including the English-only subset M2Kang, along with the framework and codebase used to construct the dataset.

Paperid: 6849, https://arxiv.org/pdf/2601.16113.pdf

Abstract:
Optical Character Recognition (OCR) for low-resource languages remains a significant challenge due to the scarcity of large-scale annotated training datasets. Languages such as Kashmiri, with approximately 7 million speakers and a complex Perso-Arabic script featuring unique diacritical marks, currently lack support in major OCR systems including Tesseract, TrOCR, and PaddleOCR. Manual dataset creation for such languages is prohibitively expensive, time-consuming, and error-prone, often requiring word by word transcription of printed or handwritten text. We present SynthOCR-Gen, an open-source synthetic OCR dataset generator specifically designed for low-resource languages. Our tool addresses the fundamental bottleneck in OCR development by transforming digital Unicode text corpora into ready-to-use training datasets. The system implements a comprehensive pipeline encompassing text segmentation (character, word, n-gram, sentence, and line levels), Unicode normalization with script purity enforcement, multi-font rendering with configurable distribution, and 25+ data augmentation techniques simulating real-world document degradations including rotation, blur, noise, and scanner artifacts. We demonstrate the efficacy of our approach by generating a 600,000-sample word-segmented Kashmiri OCR dataset, which we release publicly on HuggingFace. This work provides a practical pathway for bringing low-resource languages into the era of vision-language AI models, and the tool is openly available for researchers and practitioners working with underserved writing systems worldwide.

Paperid: 6850, https://arxiv.org/pdf/2601.16108.pdf

Abstract:
Climate disinformation has become a major challenge in today digital world, especially with the rise of misleading images and videos shared widely on social media. These false claims are often convincing and difficult to detect, which can delay actions on climate change. While vision-language models (VLMs) have been used to identify visual disinformation, they rely only on the knowledge available at the time of training. This limits their ability to reason about recent events or updates. The main goal of this paper is to overcome that limitation by combining VLMs with external knowledge. By retrieving up-to-date information such as reverse image results, online fact-checks, and trusted expert content, the system can better assess whether an image and its claim are accurate, misleading, false, or unverifiable. This approach improves the model ability to handle real-world climate disinformation and supports efforts to protect public understanding of science in a rapidly changing information landscape.

Paperid: 6851, https://arxiv.org/pdf/2601.15891.pdf

Abstract:
Recent advances in medical vision language models guide the learning of visual representations; however, this form of supervision is constrained by the availability of paired image text data, raising the question of whether robust radiology encoders can be learned without relying on language supervision. In this work, we introduce RadJEPA, a self-supervised framework built on a Joint Embedding Predictive Architecture that learns without language supervision. Pre-trained solely on unlabeled chest X-ray images, the model learns to predict latent representations of masked image regions. This predictive objective differs fundamentally from both image text pre-training and DINO-style self-distillation: rather than aligning global representations across views or modalities, RadJEPA explicitly models latent-space prediction. We evaluate the learned encoder on disease classification, semantic segmentation, and report generation tasks. Across benchmarks, RadJEPA achieves performance exceeding state-of-the-art approaches, including Rad-DINO.

Paperid: 6852, https://arxiv.org/pdf/2601.15453.pdf

Abstract:
Few-normal shot anomaly detection (FNSAD) aims to detect abnormal regions in images using only a few normal training samples, making the task highly challenging due to limited supervision and the diversity of potential defects. Recent approaches leverage vision-language models such as CLIP with prompt-based learning to align image and text features. However, existing methods often exhibit weak discriminability between normal and abnormal prompts and lack principled scoring mechanisms for patch-level anomalies. We propose a deviation-guided prompt learning framework that integrates the semantic power of vision-language models with the statistical reliability of deviation-based scoring. Specifically, we replace fixed prompt prefixes with learnable context vectors shared across normal and abnormal prompts, while anomaly-specific suffix tokens enable class-aware alignment. To enhance separability, we introduce a deviation loss with Top-K Multiple Instance Learning (MIL), modeling patch-level features as Gaussian deviations from the normal distribution. This allows the network to assign higher anomaly scores to patches with statistically significant deviations, improving localization and interpretability. Experiments on the MVTecAD and VISA benchmarks demonstrate superior pixel-level detection performance compared to PromptAD and other baselines. Ablation studies further validate the effectiveness of learnable prompts, deviation-based scoring, and the Top-K MIL strategy.

Paperid: 6853, https://arxiv.org/pdf/2601.14945.pdf

Abstract:
Large-scale Vision-Language-Action (VLA) models offer semantic generalization but suffer from high inference latency, limiting them to low-frequency batch-and-execute paradigm. This frequency mismatch creates an execution blind spot, causing failures in dynamic environments where targets move during the open-loop execution window. We propose TIDAL (Temporally Interleaved Diffusion and Action Loop), a hierarchical framework that decouples semantic reasoning from high-frequency actuation. TIDAL operates as a backbone-agnostic module for diffusion-based VLAs, using a dual-frequency architecture to redistribute the computational budget. Specifically, a low-frequency macro-intent loop caches semantic embeddings, while a high-frequency micro-control loop interleaves single-step flow integration with execution. This design enables approximately 9 Hz control updates on edge hardware (vs. approximately 2.4 Hz baselines) without increasing marginal overhead. To handle the resulting latency shift, we introduce a temporally misaligned training strategy where the policy learns predictive compensation using stale semantic intent alongside real-time proprioception. Additionally, we address the insensitivity of static vision encoders to velocity by incorporating a differential motion predictor. TIDAL is architectural, making it orthogonal to system-level optimizations. Experiments show a 2x performance gain over open-loop baselines in dynamic interception tasks. Despite a marginal regression in static success rates, our approach yields a 4x increase in feedback frequency and extends the effective horizon of semantic embeddings beyond the native action chunk size. Under non-paused inference protocols, TIDAL remains robust where standard baselines fail due to latency.

Paperid: 6854, https://arxiv.org/pdf/2601.14774.pdf

Abstract:
This study investigates the feature representations produced by publicly available open source medical vision-language models (VLMs). While medical VLMs are expected to capture diagnostically relevant features, their learned representations remain underexplored, and standard evaluations like classification accuracy do not fully reveal if they acquire truly discriminative, lesion-specific features. Understanding these representations is crucial for revealing medical image structures and improving downstream tasks in medical image analysis. This study aims to investigate the feature distributions learned by medical VLMs and evaluate the impact of medical specialization. We analyze the feature distribution of multiple image modalities extracted by some representative medical VLMs across lesion classification datasets on multiple modalities. These distributions were compared them with non-medical VLMs to assess the domain-specific medical training. Our experiments showed that medical VLMs can extract discriminative features that are effective for medical classification tasks. Moreover, it was found that non-medical VLMs with recent improvement with contextual enrichment such as LLM2CLIP produce more refined feature representations. Our results imply that enhancing text encoder is more crucial than training intensively on medical images when developing medical VLMs. Notably, non-medical models are particularly vulnerable to biases introduced by overlaied text strings on images. These findings underscore the need for careful consideration on model selection according to downstream tasks besides potential risks in inference due to background biases such as textual information in images.

Paperid: 6855, https://arxiv.org/pdf/2601.14440.pdf

Abstract:
Vision-language models (VLMs) lag behind text-only language models on mathematical reasoning when the same problems are presented as images rather than text. We empirically characterize this as a modality gap: the same question in text form yields markedly higher accuracy than its visually typeset counterpart, due to compounded failures in reading dense formulas, layout, and mixed symbolic-diagrammatic context. First, we introduce VisTIRA (Vision and Tool-Integrated Reasoning Agent), a tool-integrated reasoning framework that enables structured problem solving by iteratively decomposing a given math problem (as an image) into natural language rationales and executable Python steps to determine the final answer. Second, we build a framework to measure and improve visual math reasoning: a LaTeX-based pipeline that converts chain-of-thought math corpora (e.g., NuminaMath) into challenging image counterparts, and a large set of synthetic tool-use trajectories derived from a real-world, homework-style image dataset (called SnapAsk) for fine-tuning VLMs. Our experiments show that tool-integrated supervision improves image-based reasoning, and OCR grounding can further narrow the gap for smaller models, although its benefit diminishes at scale. These findings highlight that modality gap severity inversely correlates with model size, and that structured reasoning and OCR-based grounding are complementary strategies for advancing visual mathematical reasoning.

Paperid: 6856, https://arxiv.org/pdf/2601.12304.pdf

Abstract:
Vision-language pre-training (VLP) models are vulnerable to adversarial examples, particularly in black-box scenarios. Existing multimodal attacks often suffer from limited perturbation diversity and unstable multi-stage pipelines. To address these challenges, we propose 2S-GDA, a two-stage globally-diverse attack framework. The proposed method first introduces textual perturbations through a globally-diverse strategy by combining candidate text expansion with globally-aware replacement. To enhance visual diversity, image-level perturbations are generated using multi-scale resizing and block-shuffle rotation. Extensive experiments on VLP models demonstrate that 2S-GDA consistently improves attack success rates over state-of-the-art methods, with gains of up to 11.17\% in black-box settings. Our framework is modular and can be easily combined with existing methods to further enhance adversarial transferability.

Paperid: 6857, https://arxiv.org/pdf/2601.12243.pdf

Abstract:
Video summarization helps turn long videos into clear, concise representations that are easier to review, document, and analyze, especially in high-stakes domains like surgical training. Prior work has progressed from using basic visual features like color, motion, and structural changes to using pre-trained vision-language models that can better understand what's happening in the video (semantics) and capture temporal flow, resulting in more context-aware video summarization. We propose a three-stage framework, PRISM: Procedural Representation via Integrated Semantic and Multimodal analysis, that produces semantically grounded video summaries. PRISM combines adaptive visual sampling, label-driven keyframe anchoring, and contextual validation using a large language model (LLM). Our method ensures that selected frames reflect meaningful and procedural transitions while filtering out generic or hallucinated content, resulting in contextually coherent summaries across both domain-specific and instructional videos. We evaluate our method on instructional and activity datasets, using reference summaries for instructional videos. Despite sampling fewer than 5% of the original frames, our summaries retain 84% semantic content while improving over baselines by as much as 33%. Our approach generalizes across procedural and domain-specific video tasks, achieving strong performance with both semantic alignment and precision.

Paperid: 6858, https://arxiv.org/pdf/2601.11724.pdf

Abstract:
Semi-supervised Domain Generalization (SSDG) addresses the challenge of generalizing to unseen target domains with limited labeled data. Existing SSDG methods highlight the importance of achieving high pseudo-labeling (PL) accuracy and preventing model overfitting as the main challenges in SSDG. In this light, we show that the SSDG literature's excessive focus on PL accuracy, without consideration for maximum data utilization during training, limits potential performance improvements. We propose a novel approach to the SSDG problem by aligning the intermediate features of our model with the semantically rich and generalized feature space of a Vision Language Model (VLM) in a way that promotes domain-invariance. The above approach is enhanced with effective image-level augmentation and output-level regularization strategies to improve data utilization and minimize overfitting. Extensive experimentation across four benchmarks against existing SSDG baselines suggests that our method achieves SOTA results both qualitatively and quantitatively. The code will be made publicly available.

Paperid: 6859, https://arxiv.org/pdf/2601.11322.pdf

Abstract:
Vision-Language Models (VLMs) offer the ability to generate high-level, interpretable descriptions of complex activities from images and videos, making them valuable for situational awareness (SA) applications. In such settings, the focus is on identifying infrequent but significant events with high reliability and accuracy, while also extracting fine-grained details and assessing recognition quality. In this paper, we propose an approach that integrates VLMs with traditional computer vision methods through explicit logic reasoning to enhance SA in three key ways: (a) extracting fine-grained event details, (b) employing an intelligent fine-tuning (FT) strategy that achieves substantially higher accuracy than uninformed selection, and (c) generating justifications for VLM outputs during inference. We demonstrate that our intelligent FT mechanism improves the accuracy and provides a valuable means, during inferencing, to either confirm the validity of the VLM output or indicate why it may be questionable.

Paperid: 6860, https://arxiv.org/pdf/2601.11248.pdf

Abstract:
Handwritten word retrieval is vital for digital archives but remains challenging due to large handwriting variability and cross-lingual semantic gaps. While large vision-language models offer potential solutions, their prohibitive computational costs hinder practical edge deployment. To address this, we propose a lightweight asymmetric dual-encoder framework that learns unified, style-invariant visual embeddings. By jointly optimizing instance-level alignment and class-level semantic consistency, our approach anchors visual embeddings to language-agnostic semantic prototypes, enforcing invariance across scripts and writing styles. Experiments show that our method outperforms 28 baselines and achieves state-of-the-art accuracy on within-language retrieval benchmarks. We further conduct explicit cross-lingual retrieval, where the query language differs from the target language, to validate the effectiveness of the learned cross-lingual representations. Achieving strong performance with only a fraction of the parameters required by existing models, our framework enables accurate and resource-efficient cross-script handwriting retrieval.

Paperid: 6861, https://arxiv.org/pdf/2601.10835.pdf

Abstract:
As robotics become increasingly integrated into construction workflows, their ability to interpret and respond to human behavior will be essential for enabling safe and effective collaboration. Vision-Language Models (VLMs) have emerged as a promising tool for visual understanding tasks and offer the potential to recognize human behaviors without extensive domain-specific training. This capability makes them particularly appealing in the construction domain, where labeled data is scarce and monitoring worker actions and emotional states is critical for safety and productivity. In this study, we evaluate the performance of three leading VLMs, GPT-4o, Florence 2, and LLaVa-1.5, in detecting construction worker actions and emotions from static site images. Using a curated dataset of 1,000 images annotated across ten action and ten emotion categories, we assess each model's outputs through standardized inference pipelines and multiple evaluation metrics. GPT-4o consistently achieved the highest scores across both tasks, with an average F1-score of 0.756 and accuracy of 0.799 in action recognition, and an F1-score of 0.712 and accuracy of 0.773 in emotion recognition. Florence 2 performed moderately, with F1-scores of 0.497 for action and 0.414 for emotion, while LLaVa-1.5 showed the lowest overall performance, with F1-scores of 0.466 for action and 0.461 for emotion. Confusion matrix analyses revealed that all models struggled to distinguish semantically close categories, such as collaborating in teams versus communicating with supervisors. While the results indicate that general-purpose VLMs can offer a baseline capability for human behavior recognition in construction environments, further improvements, such as domain adaptation, temporal modeling, or multimodal sensing, may be needed for real-world reliability.

Paperid: 6862, https://arxiv.org/pdf/2601.10378.pdf

Abstract:
Recent achievements of vision-language models in end-to-end OCR point to a new avenue for low-loss compression of textual information. This motivates earlier works that render the Transformer's input into images for prefilling, which effectively reduces the number of tokens through visual encoding, thereby alleviating the quadratically increased Attention computations. However, this partial compression fails to save computational or memory costs at token-by-token inference. In this paper, we investigate global context compression, which saves tokens at both prefilling and inference stages. Consequently, we propose VIST2, a novel Transformer that interleaves input text chunks alongside their visual encoding, while depending exclusively on visual tokens in the pre-context to predict the next text token distribution. Around this idea, we render text chunks into sketch images and train VIST2 in multiple stages, starting from curriculum-scheduled pretraining for optical language modeling, followed by modal-interleaved instruction tuning. We conduct extensive experiments using VIST2 families scaled from 0.6B to 8B to explore the training recipe and hyperparameters. With a 4$\times$ compression ratio, the resulting models demonstrate significant superiority over baselines on long writing tasks, achieving, on average, a 3$\times$ speedup in first-token generation, 77% reduction in memory usage, and 74% reduction in FLOPS. Our codes and datasets will be public to support further studies.

Paperid: 6863, https://arxiv.org/pdf/2601.09044.pdf

Abstract:
Medical imaging datasets often suffer from class imbalance and limited availability of pathology-rich cases, which constrains the performance of machine learning models for segmentation, classification, and vision-language tasks. To address this challenge, we propose POWDR, a pathology-preserving outpainting framework for 3D MRI based on a conditioned wavelet diffusion model. Unlike conventional augmentation or unconditional synthesis, POWDR retains real pathological regions while generating anatomically plausible surrounding tissue, enabling diversity without fabricating lesions. Our approach leverages wavelet-domain conditioning to enhance high-frequency detail and mitigate blurring common in latent diffusion models. We introduce a random connected mask training strategy to overcome conditioning-induced collapse and improve diversity outside the lesion. POWDR is evaluated on brain MRI using BraTS datasets and extended to knee MRI to demonstrate tissue-agnostic applicability. Quantitative metrics (FID, SSIM, LPIPS) confirm image realism, while diversity analysis shows significant improvement with random-mask training (cosine similarity reduced from 0.9947 to 0.9580; KL divergence increased from 0.00026 to 0.01494). Clinically relevant assessments reveal gains in tumor segmentation performance using nnU-Net, with Dice scores improving from 0.6992 to 0.7137 when adding 50 synthetic cases. Tissue volume analysis indicates no significant differences for CSF and GM compared to real images. These findings highlight POWDR as a practical solution for addressing data scarcity and class imbalance in medical imaging. The method is extensible to multiple anatomies and offers a controllable framework for generating diverse, pathology-preserving synthetic data to support robust model development.

Paperid: 6864, https://arxiv.org/pdf/2601.08467.pdf

Abstract:
Distracted driving is a major cause of traffic collisions, calling for robust and scalable detection methods. Vision-language models (VLMs) enable strong zero-shot image classification, but existing VLM-based distracted driver detectors often underperform in real-world conditions. We identify subject-specific appearance variations (e.g., clothing, age, and gender) as a key bottleneck: VLMs entangle these factors with behavior cues, leading to decisions driven by who the driver is rather than what the driver is doing. To address this, we propose a subject decoupling framework that extracts a driver appearance embedding and removes its influence from the image embedding prior to zero-shot classification, thereby emphasizing distraction-relevant evidence. We further orthogonalize text embeddings via metric projection onto Stiefel manifold to improve separability while staying close to the original semantics. Experiments demonstrate consistent gains over prior baselines, indicating the promise of our approach for practical road-safety applications.

Paperid: 6865, https://arxiv.org/pdf/2601.08017.pdf

Abstract:
We show that for a variety of concepts in adapter-based vision-language models, the representations of their images and their text descriptions are meaningfully aligned from the very first layer. This contradicts the established view that such image-text alignment only appears in late layers. We show this using a new synthesis-based method inspired by DeepDream: given a textual concept such as "Jupiter", we extract its concept vector at a given layer, and then use optimisation to synthesise an image whose representation aligns with that vector. We apply our approach to hundreds of concepts across seven layers in Gemma 3, and find that the synthesised images often depict salient visual features of the targeted textual concepts: for example, already at layer 1, more than 50 % of images depict recognisable features of animals, activities, or seasons. Our method thus provides direct, constructive evidence of image-text alignment on a concept-by-concept and layer-by-layer basis. Unlike previous methods for measuring multimodal alignment, our approach is simple, fast, and does not require auxiliary models or datasets. It also offers a new path towards model interpretability, by providing a way to visualise a model's representation space by backtracing through its image processing components.

Paperid: 6866, https://arxiv.org/pdf/2601.07761.pdf

Abstract:
Large Vision-Language Models (LVLMs) face a fundamental dilemma in video reasoning: they are caught between the prohibitive computational costs of verbose reasoning and the hallucination risks of efficient, ungrounded approaches. To resolve this, we introduce the Chain of Evidence (CoE), a novel framework that architecturally decouples and co-optimizes perceptual grounding and reasoning efficiency. CoE incorporates two core innovations: (1) A lightweight Evidence Grounding Module (EGM) that acts as a query-guided filter, dynamically identifying and extracting a compact set of high-fidelity visual evidence; and (2) An Evidence-Anchoring Protocol optimized via Reinforcement Learning. Crucially, we design a composite reward mechanism that enforces process alignment, compelling the model to strictly reference identified temporal anchors during deduction, thereby mitigating hallucinations. To enable this, we construct CoE-Instruct, a large-scale dataset (164k samples) featuring a novel dual-annotation schema for separate perception and reasoning supervision. Extensive experiments on five benchmarks, including Video-MME, MVBench, and VSI-Bench, demonstrate that CoE-enhanced models establish a new state-of-the-art. They significantly outperform existing methods in accuracy, proving CoE to be a powerful and practical paradigm for reliable video understanding.

Paperid: 6867, https://arxiv.org/pdf/2601.07092.pdf

Abstract:
Autonomous driving increasingly relies on Visual Question Answering (VQA) to enable vehicles to understand complex surroundings by analyzing visual inputs and textual queries. Currently, a paramount concern for VQA in this domain is the stringent requirement for fast latency and real-time processing, as delays directly impact real-world safety in this safety-critical application. However, current state-of-the-art VQA models, particularly large vision-language models (VLMs), often prioritize performance over computational efficiency. These models typically process dense patch tokens for every frame, leading to prohibitive computational costs (FLOPs) and significant inference latency, especially with long video sequences. This focus limits their practical deployment in real-time autonomous driving scenarios. To tackle this issue, we propose an efficient VLM framework for autonomous driving VQA tasks, SRC-Pipeline. It learns to compress early frame tokens into a small number of high-level tokens while retaining full patch tokens for recent frames. Experiments on autonomous driving video question answering tasks show that our approach achieves 66% FLOPs reduction while maintaining comparable performance, enabling VLMs to operate more effectively in real-time, safety-critical autonomous driving settings.

Paperid: 6868, https://arxiv.org/pdf/2601.06464.pdf

Abstract:
3D Vision-Language Models (VLMs), such as PointLLM and GPT4Point, have shown strong reasoning and generalization abilities in 3D understanding tasks. However, their adversarial robustness remains largely unexplored. Prior work in 2D VLMs has shown that the integration of visual inputs significantly increases vulnerability to adversarial attacks, making these models easier to manipulate into generating toxic or misleading outputs. In this paper, we investigate whether incorporating 3D vision similarly compromises the robustness of 3D VLMs. To this end, we present the first systematic study of adversarial robustness in point-based 3D VLMs. We propose two complementary attack strategies: \textit{Vision Attack}, which perturbs the visual token features produced by the 3D encoder and projector to assess the robustness of vision-language alignment; and \textit{Caption Attack}, which directly manipulates output token sequences to evaluate end-to-end system robustness. Each attack includes both untargeted and targeted variants to measure general vulnerability and susceptibility to controlled manipulation. Our experiments reveal that 3D VLMs exhibit significant adversarial vulnerabilities under untargeted attacks, while demonstrating greater resilience against targeted attacks aimed at forcing specific harmful outputs, compared to their 2D counterparts. These findings highlight the importance of improving the adversarial robustness of 3D VLMs, especially as they are deployed in safety-critical applications.

Paperid: 6869, https://arxiv.org/pdf/2601.06204.pdf

Abstract:
Intelligent anomaly detection in dynamic visual environments requires reconciling real-time performance with semantic interpretability. Conventional approaches address only fragments of this challenge. Reconstruction-based models capture low-level deviations without contextual reasoning, object detectors provide speed but limited semantics, and large vision-language systems deliver interpretability at prohibitive computational cost. This work introduces a cascading multi-agent framework that unifies these complementary paradigms into a coherent and interpretable architecture. Early modules perform reconstruction-gated filtering and object-level assessment, while higher-level reasoning agents are selectively invoked to interpret semantically ambiguous events. The system employs adaptive escalation thresholds and a publish-subscribe communication backbone, enabling asynchronous coordination and scalable deployment across heterogeneous hardware. Extensive evaluation on large-scale monitoring data demonstrates that the proposed cascade achieves a threefold reduction in latency compared to direct vision-language inference, while maintaining high perceptual fidelity (PSNR = 38.3 dB, SSIM = 0.965) and consistent semantic labeling. The framework advances beyond conventional detection pipelines by combining early-exit efficiency, adaptive multi-agent reasoning, and explainable anomaly attribution, establishing a reproducible and energy-efficient foundation for scalable intelligent visual monitoring.

Paperid: 6870, https://arxiv.org/pdf/2601.05874.pdf

Abstract:
Modelling a language model for a multi-lingual scenario includes several potential challenges, among which catastrophic forgetting is the major challenge. For example, small language models (SLM) built for low-resource languages by adapting large language models (LLMs) pose the challenge of catastrophic forgetting. This work proposes to employ a continual learning strategy using parts-of-speech (POS)-based code-switching along with a replay adapter strategy to mitigate the identified gap of catastrophic forgetting while training SLM from LLM. Experiments conducted on vision language tasks such as visual question answering and language modelling task exhibits the success of the proposed architecture.

Paperid: 6871, https://arxiv.org/pdf/2601.05336.pdf

Abstract:
Designing intuitive interfaces for robotic control remains a central challenge in enabling effective human-robot interaction, particularly in assistive care settings. Eye gaze offers a fast, non-intrusive, and intent-rich input modality, making it an attractive channel for conveying user goals. In this work, we present GAMMA (Gaze Assisted Manipulation for Modular Autonomy), a system that leverages ego-centric gaze tracking and a vision-language model to infer user intent and autonomously execute robotic manipulation tasks. By contextualizing gaze fixations within the scene, the system maps visual attention to high-level semantic understanding, enabling skill selection and parameterization without task-specific training. We evaluate GAMMA on a range of table-top manipulation tasks and compare it against baseline gaze-based control without reasoning. Results demonstrate that GAMMA provides robust, intuitive, and generalizable control, highlighting the potential of combining foundation models and gaze for natural and scalable robot autonomy. Project website: https://gamma0.vercel.app/

Paperid: 6872, https://arxiv.org/pdf/2601.05269.pdf

Abstract:
The recent Artificial Intelligence (AI) revolution has opened transformative possibilities for the humanities, particularly in unlocking the visual-artistic content embedded in historical illuminated manuscripts. While digital archives now offer unprecedented access to these materials, the ability to systematically locate, extract, and analyze illustrations at scale remains a major challenge. We present a general and scalable AI-based pipeline for large-scale visual analysis of illuminated manuscripts. The framework integrates modern deep-learning models for page-level illustration detection, illustration extraction, and multimodal description, enabling scholars to search, cluster, and study visual materials and artistic trends across entire corpora. We demonstrate the applicability of this approach on large heterogeneous collections, including the Vatican Library and richly illuminated manuscripts such as the Bible of Borso d'Este. The system reveals meaningful visual patterns and cross-manuscript relationships by embedding illustrations into a shared representation space and analyzing their similarity structure (see figure 4). By harnessing recent advances in computer vision and vision-language models, our framework enables new forms of large-scale visual scholarship in historical studies, art history, and cultural heritage making it possible to explore iconography, stylistic trends, and cultural connections in ways that were previously impractical.

Paperid: 6873, https://arxiv.org/pdf/2601.05125.pdf

Abstract:
This work introduces VERSE, a methodology for analyzing and improving Vision-Language Models applied to Visually-rich Document Understanding by exploring their visual embedding space. VERSE enables the visualization of latent representations, supporting the assessment of model feasibility. It also facilitates the identification of problematic regions and guides the generation of synthetic data to enhance performance in those clusters. We validate the methodology by training on the synthetic MERIT Dataset and evaluating on its real-world counterpart, MERIT Secret. Results show that VERSE helps uncover the visual features associated with error-prone clusters, and that retraining with samples containing these features substantially boosts F1 performance without degrading generalization. Furthermore, we demonstrate that on-premise models such as Donut and Idefics2, when optimized with VERSE, match or even surpass the performance of SaaS solutions like GPT-4 and Pixtral.

Paperid: 6874, https://arxiv.org/pdf/2601.04819.pdf

Abstract:
AEC drawings encode geometry and semantics through symbols, layout conventions, and dense annotation, yet it remains unclear whether modern multimodal and vision-language models can reliably interpret this graphical language. We present AECV-Bench, a benchmark for evaluating multimodal and vision-language models on realistic AEC artefacts via two complementary use cases: (i) object counting on 120 high-quality floor plans (doors, windows, bedrooms, toilets), and (ii) drawing-grounded document QA spanning 192 question-answer pairs that test text extraction (OCR), instance counting, spatial reasoning, and comparative reasoning over common drawing regions. Object-counting performance is reported using per-field exact-match accuracy and MAPE results, while document-QA performance is reported using overall accuracy and per-category breakdowns with an LLM-as-a-judge scoring pipeline and targeted human adjudication for edge cases. Evaluating a broad set of state-of-the-art models under a unified protocol, we observe a stable capability gradient; OCR and text-centric document QA are strongest (up to 0.95 accuracy), spatial reasoning is moderate, and symbol-centric drawing understanding - especially reliable counting of doors and windows - remains unsolved (often 0.40-0.55 accuracy) with substantial proportional errors. These results suggest that current systems function well as document assistants but lack robust drawing literacy, motivating domain-specific representations and tool-augmented, human-in-the-loop workflows for an efficient AEC automation.

Paperid: 6875, https://arxiv.org/pdf/2601.04752.pdf

Abstract:
This work explores the visual capabilities and limitations of foundation models by introducing a novel adversarial attack method utilizing skeletonization to reduce the search space effectively. Our approach specifically targets images containing text, particularly mathematical formula images, which are more challenging due to their LaTeX conversion and intricate structure. We conduct a detailed evaluation of both character and semantic changes between original and adversarially perturbed outputs to provide insights into the models' visual interpretation and reasoning abilities. The effectiveness of our method is further demonstrated through its application to ChatGPT, which shows its practical implications in real-world scenarios.

Paperid: 6876, https://arxiv.org/pdf/2601.04609.pdf

Abstract:
Vision-language models (VLMs) are increasingly used to make visual content accessible via text-based descriptions. In current systems, however, description specificity is often conflated with their length. We argue that these two concepts must be disentangled: descriptions can be concise yet dense with information, or lengthy yet vacuous. We define specificity relative to a contrast set, where a description is more specific to the extent that it picks out the target image better than other possible images. We construct a dataset that controls for length while varying information content, and validate that people reliably prefer more specific descriptions regardless of length. We find that controlling for length alone cannot account for differences in specificity: how the length budget is allocated makes a difference. These results support evaluation approaches that directly prioritize specificity over verbosity.

Paperid: 6877, https://arxiv.org/pdf/2601.03915.pdf

Abstract:
Microscopic evaluation of white blood cell morphology is central to leukemia diagnosis, yet current deep learning models often act as black boxes, limiting clinical trust and adoption. We introduce HemBLIP, a vision language model designed to generate interpretable, morphology aware descriptions of peripheral blood cells. Using a newly constructed dataset of 14k healthy and leukemic cells paired with expert-derived attribute captions, we adapt a general-purpose VLM via both full fine-tuning and LoRA based parameter efficient training, and benchmark against the biomedical foundation model MedGEMMA. HemBLIP achieves higher caption quality and morphological accuracy, while LoRA adaptation provides further gains with significantly reduced computational cost. These results highlight the promise of vision language models for transparent and scalable hematological diagnostics.

Paperid: 6878, https://arxiv.org/pdf/2601.01897.pdf

Abstract:
Claims documents are fundamental to healthcare and insurance operations, serving as the basis for reimbursement, auditing, and compliance. However, these documents are typically not born digital; they often exist as scanned PDFs or photographs captured under uncontrolled conditions. Consequently, they exhibit significant content heterogeneity, ranging from typed invoices to handwritten medical reports, as well as linguistic diversity. This challenge is exemplified by operations at Fullerton Health, which handles tens of millions of claims annually across nine markets, including Singapore, the Philippines, Indonesia, Malaysia, Mainland China, Hong Kong, Vietnam, Papua New Guinea, and Cambodia. Such variability, coupled with inconsistent image quality and diverse layouts, poses a significant obstacle to automated parsing and structured information extraction. This paper presents a robust multi-stage pipeline that integrates the multilingual optical character recognition (OCR) engine PaddleOCR, a traditional Logistic Regression classifier, and a compact Vision-Language Model (VLM), Qwen 2.5-VL-7B, to achieve efficient and accurate field extraction from large-scale claims data. The proposed system achieves a document-type classification accuracy of over 95 percent and a field-level extraction accuracy of approximately 87 percent, while maintaining an average processing latency of under 2 seconds per document. Compared to manual processing, which typically requires around 10 minutes per claim, our system delivers a 300x improvement in efficiency. These results demonstrate that combining traditional machine learning models with modern VLMs enables production-grade accuracy and speed for real-world automation. The solution has been successfully deployed in our mobile application and is currently processing tens of thousands of claims weekly from Vietnam and Singapore.

Paperid: 6879, https://arxiv.org/pdf/2601.01513.pdf

Abstract:
Vision-Language Models (VLMs) excel at visual reasoning but still struggle with integrating external knowledge. Retrieval-Augmented Generation (RAG) is a promising solution, but current methods remain inefficient and often fail to maintain high answer quality. To address these challenges, we propose VideoSpeculateRAG, an efficient VLM-based RAG framework built on two key ideas. First, we introduce a speculative decoding pipeline: a lightweight draft model quickly generates multiple answer candidates, which are then verified and refined by a more accurate heavyweight model, substantially reducing inference latency without sacrificing correctness. Second, we identify a major source of error - incorrect entity recognition in retrieved knowledge - and mitigate it with a simple yet effective similarity-based filtering strategy that improves entity alignment and boosts overall answer accuracy. Experiments demonstrate that VideoSpeculateRAG achieves comparable or higher accuracy than standard RAG approaches while accelerating inference by approximately 2x. Our framework highlights the potential of combining speculative decoding with retrieval-augmented reasoning to enhance efficiency and reliability in complex, knowledge-intensive multimodal tasks.

Paperid: 6880, https://arxiv.org/pdf/2601.01095.pdf

Abstract:
Multimodal large language models (MLLMs) have achieved impressive progress in vision-language reasoning, yet their ability to understand temporally unfolding narratives in videos remains underexplored. True narrative understanding requires grounding who is doing what, when, and where, maintaining coherent entity representations across dynamic visual and temporal contexts. We introduce NarrativeTrack, the first benchmark to evaluate narrative understanding in MLLMs through fine-grained entity-centric reasoning. Unlike existing benchmarks limited to short clips or coarse scene-level semantics, we decompose videos into constituent entities and examine their continuity via a Compositional Reasoning Progression (CRP), a structured evaluation framework that progressively increases narrative complexity across three dimensions: entity existence, entity changes, and entity ambiguity. CRP challenges models to advance from temporal persistence to contextual evolution and fine-grained perceptual reasoning. A fully automated entity-centric pipeline enables scalable extraction of temporally grounded entity representations, providing the foundation for CRP. Evaluations of state-of-the-art MLLMs reveal that models fail to robustly track entities across visual transitions and temporal dynamics, often hallucinating identity under context shifts. Open-source general-purpose MLLMs exhibit strong perceptual grounding but weak temporal coherence, while video-specific MLLMs capture temporal context yet hallucinate entity's contexts. These findings uncover a fundamental trade-off between perceptual grounding and temporal reasoning, indicating that narrative understanding emerges only from their integration. NarrativeTrack provides the first systematic framework to diagnose and advance temporally grounded narrative comprehension in MLLMs.

Paperid: 6881, https://arxiv.org/pdf/2601.00978.pdf

Abstract:
Human-robot collaboration (HRC) in structured assembly requires reliable state estimation and adaptive task planning under noisy perception and human interventions. To address these challenges, we introduce a design-grounded human-aware planning framework for human-robot collaborative structured assembly. The framework comprises two coupled modules. Module I, Perception-to-Symbolic State (PSS), employs vision-language models (VLMs) based agents to align RGB-D observations with design specifications and domain knowledge, synthesizing verifiable symbolic assembly states. It outputs validated installed and uninstalled component sets for online state tracking. Module II, Human-Aware Planning and Replanning (HPR), performs task-level multi-robot assignment and updates the plan only when the observed state deviates from the expected execution outcome. It applies a minimal-change replanning rule to selectively revise task assignments and preserve plan stability even under human interventions. We validate the framework on a 27-component timber-frame assembly. The PSS module achieves 97% state synthesis accuracy, and the HPR module maintains feasible task progression across diverse HRC scenarios. Results indicate that integrating VLM-based perception with knowledge-driven planning improves robustness of state estimation and task planning under dynamic conditions.

Paperid: 6882, https://arxiv.org/pdf/2601.00969.pdf

Abstract:
Vision-Language-Action (VLA) models have emerged as powerful generalist policies for robotic manipulation, yet they remain fundamentally limited by their reliance on behavior cloning, leading to brittleness under distribution shift. While augmenting pretrained models with test-time search algorithms like Monte Carlo Tree Search (MCTS) can mitigate these failures, existing formulations rely solely on the VLA prior for guidance, lacking a grounded estimate of expected future return. Consequently, when the prior is inaccurate, the planner can only correct action selection via the exploration term, which requires extensive simulation to become effective. To address this limitation, we introduce Value Vision-Language-Action Planning and Search (V-VLAPS), a framework that augments MCTS with a lightweight, learnable value function. By training a simple multilayer perceptron (MLP) on the latent representations of a fixed VLA backbone (Octo), we provide the search with an explicit success signal that biases action selection toward high-value regions. We evaluate V-VLAPS on the LIBERO robotic manipulation suite, demonstrating that our value-guided search improves success rates by over 5 percentage points while reducing the average number of MCTS simulations by 5-15 percent compared to baselines that rely only on the VLA prior.

Paperid: 6883, https://arxiv.org/pdf/2601.00905.pdf

Abstract:
While the importance of efficient recycling is widely acknowledged, accurately determining the recyclability of items and their proper disposal remains a complex task for the general public. In this study, we explore the application of cutting-edge vision-language models (GPT-4o, GPT-4o-mini, and Claude 3.5) for predicting the recyclability of commonly disposed items. Utilizing a curated dataset of images, we evaluated the models' ability to match objects to appropriate recycling bins, including assessing whether the items could physically fit into the available bins. Additionally, we investigated the models' performance across several challenging scenarios: (i) adjusting predictions based on location-specific recycling guidelines; (ii) accounting for contamination or structural damage; and (iii) handling objects composed of multiple materials. Our findings highlight the significant advancements in contextual understanding offered by these models compared to previous iterations, while also identifying areas where they still fall short. The continued refinement of context-aware models is crucial for enhancing public recycling practices and advancing environmental sustainability.

Paperid: 6884, https://arxiv.org/pdf/2512.24826.pdf

Abstract:
Cross-modal systems trained on 2D visual inputs are presented with a dimensional shift when processing 3D scenes. An in-scene camera bridges the dimensionality gap but requires learning a control module. We introduce a new method that improves multivariate mutual information estimates by regret minimisation with derivative-free optimisation. Our algorithm enables off-the-shelf cross-modal systems trained on 2D visual inputs to adapt online to object occlusions and differentiate features. The pairing of expressive measures and value-based optimisation assists control of an in-scene camera to learn directly from the noisy outputs of vision-language models. The resulting pipeline improves performance in cross-modal tasks on multi-object 3D scenes without resorting to pretraining or finetuning.

Paperid: 6885, https://arxiv.org/pdf/2512.24673.pdf

Abstract:
Vision-Language-Action (VLA) models have achieved remarkable breakthroughs in robotics, with the action chunk playing a dominant role in these advances. Given the real-time and continuous nature of robotic motion control, the strategies for fusing a queue of successive action chunks have a profound impact on the overall performance of VLA models. Existing methods suffer from jitter, stalling, or even pauses in robotic action execution, which not only limits the achievable execution speed but also reduces the overall success rate of task completion. This paper introduces VLA-RAIL (A Real-Time Asynchronous Inference Linker), a novel framework designed to address these issues by conducting model inference and robot motion control asynchronously and guaranteeing smooth, continuous, and high-speed action execution. The core contributions of the paper are two fold: a Trajectory Smoother that effectively filters out the noise and jitter in the trajectory of one action chunk using polynomial fitting and a Chunk Fuser that seamlessly align the current executing trajectory and the newly arrived chunk, ensuring position, velocity, and acceleration continuity between two successive action chunks. We validate the effectiveness of VLA-RAIL on a benchmark of dynamic simulation tasks and several real-world manipulation tasks. Experimental results demonstrate that VLA-RAIL significantly reduces motion jitter, enhances execution speed, and improves task success rates, which will become a key infrastructure for the large-scale deployment of VLA models.

Paperid: 6886, https://arxiv.org/pdf/2512.24331.pdf

Abstract:
While Vision-Language Models (VLMs) show significant promise for end-to-end autonomous driving by leveraging the common sense embedded in language models, their reliance on 2D image cues for complex scene understanding and decision-making presents a critical bottleneck for safety and reliability. Current image-based methods struggle with accurate metric spatial reasoning and geometric inference, leading to unreliable driving policies. To bridge this gap, we propose LVLDrive (LiDAR-Vision-Language), a novel framework specifically designed to upgrade existing VLMs with robust 3D metric spatial understanding for autonomous driving by incoperating LiDAR point cloud as an extra input modality. A key challenge lies in mitigating the catastrophic disturbance introduced by disparate 3D data to the pre-trained VLMs. To this end, we introduce a Gradual Fusion Q-Former that incrementally injects LiDAR features, ensuring the stability and preservation of the VLM's existing knowledge base. Furthermore, we develop a spatial-aware question-answering (SA-QA) dataset to explicitly teach the model advanced 3D perception and reasoning capabilities. Extensive experiments on driving benchmarks demonstrate that LVLDrive achieves superior performance compared to vision-only counterparts across scene understanding, metric spatial perception, and reliable driving decision-making. Our work highlights the necessity of explicit 3D metric data for building trustworthy VLM-based autonomous systems.

Paperid: 6887, https://arxiv.org/pdf/2512.23739.pdf

Abstract:
``Bring me a plate.'' For domestic service robots, this simple command reveals a complex challenge: inferring where everyday items are stored, often out of sight in drawers, cabinets, or closets. Despite advances in vision and manipulation, robots still lack the commonsense reasoning needed to complete this task. We introduce the Stored Household Item Challenge, a benchmark task for evaluating service robots' cognitive capabilities: given a household scene and a queried item, predict its most likely storage location. Our benchmark includes two datasets: (1) a real-world evaluation set of 100 item-image pairs with human-annotated ground truth from participants' kitchens, and (2) a development set of 6,500 item-image pairs annotated with storage polygons over public kitchen images. These datasets support realistic modeling of household organization and enable comparative evaluation across agent architectures. To begin tackling this challenge, we introduce NOAM (Non-visible Object Allocation Model), a hybrid agent pipeline that combines structured scene understanding with large language model inference. NOAM converts visual input into natural language descriptions of spatial context and visible containers, then prompts a language model (e.g., GPT-4) to infer the most likely hidden storage location. This integrated vision-language agent exhibits emergent commonsense reasoning and is designed for modular deployment within broader robotic systems. We evaluate NOAM against baselines including random selection, vision-language pipelines (Grounding-DINO + SAM), leading multimodal models (e.g., Gemini, GPT-4o, Kosmos-2, LLaMA, Qwen), and human performance. NOAM significantly improves prediction accuracy and approaches human-level results, highlighting best practices for deploying cognitively capable agents in domestic environments.

Paperid: 6888, https://arxiv.org/pdf/2512.23557.pdf

Abstract:
Powerful autonomous systems, which reason, plan, and converse using and between numerous tools and agents, are made possible by Large Language Models (LLMs), Vision-Language Models (VLMs), and new agentic AI systems, like LangChain and GraphChain. Nevertheless, this agentic environment increases the probability of the occurrence of multimodal prompt injection (PI) attacks, in which concealed or malicious instructions carried in text, pictures, metadata, or agent-to-agent messages may spread throughout the graph and lead to unintended behavior, a breach of policy, or corruption of state. In order to mitigate these risks, this paper suggests a Cross-Agent Multimodal Provenanc- Aware Defense Framework whereby all the prompts, either user-generated or produced by upstream agents, are sanitized and all the outputs generated by an LLM are verified independently before being sent to downstream nodes. This framework contains a Text sanitizer agent, visual sanitizer agent, and output validator agent all coordinated by a provenance ledger, which keeps metadata of modality, source, and trust level throughout the entire agent network. This architecture makes sure that agent-to-agent communication abides by clear trust frames such such that injected instructions are not propagated down LangChain or GraphChain-style-workflows. The experimental assessments show that multimodal injection detection accuracy is significantly enhanced, and the cross-agent trust leakage is minimized, as well as, agentic execution pathways become stable. The framework, which expands the concept of provenance tracking and validation to the multi-agent orchestration, enhances the establishment of secure, understandable and reliable agentic AI systems.

Paperid: 6889, https://arxiv.org/pdf/2512.23028.pdf

Abstract:
This report provides an architecture-led analysis of two modern vision-language models (VLMs), Qwen2.5-VL-7B-Instruct and Llama-4-Scout-17B-16E-Instruct, and explains how their architectural properties map to a practical video-to-artifact pipeline implemented in the BodyLanguageDetection repository [1]. The system samples video frames, prompts a VLM to detect visible people and generate pixel-space bounding boxes with prompt-conditioned attributes (emotion by default), validates output structure using a predefined schema, and optionally renders an annotated video. We first summarize the shared multimodal foundation (visual tokenization, Transformer attention, and instruction following), then describe each architecture at a level sufficient to justify engineering choices without speculative internals. Finally, we connect model behavior to system constraints: structured outputs can be syntactically valid while semantically incorrect, schema validation is structural (not geometric correctness), person identifiers are frame-local in the current prompting contract, and interactive single-frame analysis returns free-form text rather than schema-enforced JSON. These distinctions are critical for writing defensible claims, designing robust interfaces, and planning evaluation.

Paperid: 6890, https://arxiv.org/pdf/2512.21860.pdf

Abstract:
Conditional image embeddings are feature representations that focus on specific aspects of an image indicated by a given textual condition (e.g., color, genre), which has been a challenging problem. Although recent vision foundation models, such as CLIP, offer rich representations of images, they are not designed to focus on a specified condition. In this paper, we propose DIOR, a method that leverages a large vision-language model (LVLM) to generate conditional image embeddings. DIOR is a training-free approach that prompts the LVLM to describe an image with a single word related to a given condition. The hidden state vector of the LVLM's last token is then extracted as the conditional image embedding. DIOR provides a versatile solution that can be applied to any image and condition without additional training or task-specific priors. Comprehensive experimental results on conditional image similarity tasks demonstrate that DIOR outperforms existing training-free baselines, including CLIP. Furthermore, DIOR achieves superior performance compared to methods that require additional training across multiple settings.

Paperid: 6891, https://arxiv.org/pdf/2512.21637.pdf

Abstract:
Text-driven image manipulation often suffers from attribute entanglement, where modifying a target attribute (e.g., adding bangs) unintentionally alters other semantic properties such as identity or appearance. The Predict, Prevent, and Evaluate (PPE) framework addresses this issue by leveraging pre-trained vision-language models for disentangled editing. In this work, we analyze the PPE framework, focusing on its architectural components, including BERT-based attribute prediction and StyleGAN2-based image generation on the CelebA-HQ dataset. Through empirical analysis, we identify a limitation in the original regularization strategy, where latent updates remain dense and prone to semantic leakage. To mitigate this issue, we introduce a sparsity-based constraint using L1 regularization on latent space manipulation. Experimental results demonstrate that the proposed approach enforces more focused and controlled edits, effectively reducing unintended changes in non-target attributes while preserving facial identity.

Paperid: 6892, https://arxiv.org/pdf/2512.21529.pdf

Abstract:
Vision-Language Models (VLMs) learn powerful multimodal representations through large-scale image-text pretraining, but adapting them to hierarchical classification is underexplored. Standard approaches treat labels as flat categories and require full fine-tuning, which is expensive and produces inconsistent predictions across taxonomy levels. We propose an efficient hierarchy-aware fine-tuning framework that updates a few parameters while enforcing structural consistency. We combine two objectives: Tree-Path KL Divergence (TP-KL) aligns predictions along the ground-truth label path for vertical coherence, while Hierarchy-Sibling Smoothed Cross-Entropy (HiSCE) encourages consistent predictions among sibling classes. Both losses work in the VLM's shared embedding space and integrate with lightweight LoRA adaptation. Experiments across multiple benchmarks show consistent improvements in Full-Path Accuracy and Tree-based Inconsistency Error with minimal parameter overhead. Our approach provides an efficient strategy for adapting VLMs to structured taxonomies.

Paperid: 6893, https://arxiv.org/pdf/2512.21508.pdf

Abstract:
Multimodal chest X-Ray analysis often fine-tunes large vision-language models, which is computationally costly. We study parameter-efficient training (PET) strategies, including frozen encoders, BitFit, LoRA, and adapters for multi-label classification on the Indiana University Chest X-Ray dataset (3,851 image-report pairs; 579 test samples). To mitigate data leakage, we redact pathology terms from reports used as text inputs while retaining clinical context. Under a fixed parameter budget (2.37M parameters, 2.51% of total), all PET variants achieve AUROC between 0.892 and 0.908, outperforming full fine-tuning (0.770 AUROC), which uses 94.3M trainable parameters, a 40x reduction. External validation on CheXpert (224,316 images, 58x larger) confirms scalability: all PET methods achieve >0.69 AUROC with <9% trainable parameters, with Adapter achieving best performance (0.7214 AUROC). Budget-matched comparisons reveal that vision-only models (0.653 AUROC, 1.06M parameters) outperform budget-matched multimodal models (0.641 AUROC, 1.06M parameters), indicating improvements arise primarily from parameter allocation rather than cross-modal synergy. While PET methods show degraded calibration (ECE: 0.29-0.34) compared to simpler models (ECE: 0.049), this represents a tractable limitation addressable through post-hoc calibration methods. These findings demonstrate that frozen encoder strategies provide superior discrimination at substantially reduced computational cost, though calibration correction is essential for clinical deployment.

Paperid: 6894, https://arxiv.org/pdf/2512.20907.pdf

Abstract:
3D Visual Grounding (3DVG) is a critical bridge from vision-language perception to robotics, requiring both language understanding and 3D scene reasoning. Traditional supervised models leverage explicit 3D geometry but exhibit limited generalization, owing to the scarcity of 3D vision-language datasets and the limited reasoning capabilities compared to modern vision-language models (VLMs). We propose PanoGrounder, a generalizable 3DVG framework that couples multi-modal panoramic representation with pretrained 2D VLMs for strong vision-language reasoning. Panoramic renderings, augmented with 3D semantic and geometric features, serve as an intermediate representation between 2D and 3D, and offer two major benefits: (i) they can be directly fed to VLMs with minimal adaptation and (ii) they retain long-range object-to-object relations thanks to their 360-degree field of view. We devise a three-stage pipeline that places a compact set of panoramic viewpoints considering the scene layout and geometry, grounds a text query on each panoramic rendering with a VLM, and fuses per-view predictions into a single 3D bounding box via lifting. Our approach achieves state-of-the-art results on ScanRefer and Nr3D, and demonstrates superior generalization to unseen 3D datasets and text rephrasings.

Paperid: 6895, https://arxiv.org/pdf/2512.20674.pdf

Abstract:
Vision Language Models (VLMs) have undergone significant advancements, particularly with the emergence of mobile-oriented VLMs, which offer a wide range of application scenarios. However, the substantial computational requirements for training these models present a significant obstacle to their practical application. To address this issue, Low-Rank Adaptation (LoRA) has been proposed. Nevertheless, the standard LoRA with a fixed rank lacks sufficient capability for training mobile VLMs that process both text and image modalities. In this work, we introduce HyDRA, a parameter-efficient fine-tuning framework designed to implement hierarchical and dynamic rank scheduling for mobile VLMs. This framework incorporates two essential optimization strategies: (1) hierarchical optimization, which involves a coarse-grained approach that assigns different ranks to various layers, as well as a fine-grained method that adjusts ranks within individual layers, and (2) dynamic adjustment, which employs an end-to-end automatic optimization using a lightweight performance model to determine and adjust ranks during the fine-tuning process. Comprehensive experiments conducted on popular benchmarks demonstrate that HyDRA consistently outperforms the baseline, achieving a 4.7\% improvement across various model sizes without increasing the number of trainable parameters. In some tasks, it even surpasses full-parameter fine-tuning.

Paperid: 6896, https://arxiv.org/pdf/2512.19734.pdf

Abstract:
We propose an alternative to sparse autoencoders (SAEs) as a simple and effective unsupervised method for extracting interpretable concepts from neural networks. The core idea is to cluster differences in activations, which we formally justify within a discriminant analysis framework. To enhance the diversity of extracted concepts, we refine the approach by weighting the clustering using the skewness of activations. The method aligns with Deleuze's modern view of concepts as differences. We evaluate the approach across five models and three modalities (vision, language, and audio), measuring concept quality, diversity, and consistency. Our results show that the proposed method achieves concept quality surpassing prior unsupervised SAE variants while approaching supervised baselines, and that the extracted concepts enable steering of a model's inner representations, demonstrating their causal influence on downstream behavior.

Paperid: 6897, https://arxiv.org/pdf/2512.19663.pdf

Abstract:
Diabetic retinopathy (DR) is a leading cause of preventable blindness worldwide, demanding accurate automated diagnostic systems. While general-domain vision-language models like Contrastive Language-Image Pre-Training (CLIP) perform well on natural image tasks, they struggle in medical domain applications, particularly in cross-modal retrieval for ophthalmological images. We propose a novel knowledge-enhanced joint embedding framework that integrates retinal fundus images, clinical text, and structured patient data through a multimodal transformer architecture to address the critical gap in medical image-text alignment. Our approach employs separate encoders for each modality: a Vision Transformer (ViT-B/16) for retinal images, Bio-ClinicalBERT for clinical narratives, and a multilayer perceptron for structured demographic and clinical features. These modalities are fused through a joint transformer with modality-specific embeddings, trained using multiple objectives including contrastive losses between modality pairs, reconstruction losses for images and text, and classification losses for DR severity grading according to ICDR and SDRG schemes. Experimental results on the Brazilian Multilabel Ophthalmological Dataset (BRSET) demonstrate significant improvements over baseline models. Our framework achieves near-perfect text-to-image retrieval performance with Recall@1 of 99.94% compared to fine-tuned CLIP's 1.29%, while maintaining state-of-the-art classification accuracy of 97.05% for SDRG and 97.97% for ICDR. Furthermore, zero-shot evaluation on the unseen DeepEyeNet dataset validates strong generalizability with 93.95% Recall@1 versus 0.22% for fine-tuned CLIP. These results demonstrate that our multimodal training approach effectively captures cross-modal relationships in the medical domain, establishing both superior retrieval capabilities and robust diagnostic performance.

Paperid: 6898, https://arxiv.org/pdf/2512.19360.pdf

Abstract:
Retrieval-augmented generation improves large language models by grounding outputs in external knowledge sources, reducing hallucinations and addressing knowledge cutoffs. However, standard embedding-based retrieval fails to capture the complexity of multi-concept queries, particularly in domains like biomedicine, where biological data are inherently high-dimensional. For example,omics datasets, and clinical reports simultaneously exhibit numerous molecular, cellular, and physiological features. We present Stochastic Latent Matching (STHLM), a generative vector search method that samples query-conditioned embeddings from text or image inputs to enhance retrieval performance. Analogous to how Chain-of-Thought reasoning enables language models to "think longer" on complex problems, STHLM allows retrieval systems to "search wider" through iterative sampling. STHLM demonstrates critical improvements over classical vector retrieval across diverse benchmarks, including scientific literature, clinical notes, and tissue images, boosting retrieval performance by 10-30% through test-time compute (trading latency for accuracy), while enabling up to a 10-fold compression of embedding dimensions.

Paperid: 6899, https://arxiv.org/pdf/2512.19317.pdf

Abstract:
Vision--Language Models (VLMs) show significant promise for Medical Visual Question Answering (VQA), yet their deployment in clinical settings is hindered by severe vulnerability to adversarial attacks. Standard adversarial training, while effective for simpler tasks, often degrades both generalization performance and the quality of generated clinical reasoning. We introduce SafeMed-R1, a hybrid defense framework that ensures robust performance while preserving high-quality, interpretable medical reasoning. SafeMed-R1 employs a two-stage approach: at training time, we integrate Adversarial Training with Group Relative Policy Optimization (AT-GRPO) to explicitly robustify the reasoning process against worst-case perturbations; at inference time, we augment the model with Randomized Smoothing to provide certified $L_2$-norm robustness guarantees. We evaluate SafeMed-R1 on the OmniMedVQA benchmark across eight medical imaging modalities comprising over 88,000 samples. Our experiments reveal that standard fine-tuned VLMs, despite achieving 95\% accuracy on clean inputs, collapse to approximately 25\% under PGD attacks. In contrast, SafeMed-R1 maintains 84.45\% accuracy under the same adversarial conditions, representing a 59 percentage point improvement in robustness. Furthermore, we demonstrate that models trained with explicit chain-of-thought reasoning exhibit superior adversarial robustness compared to instruction-only variants, suggesting a synergy between interpretability and security in medical AI systems.

Paperid: 6900, https://arxiv.org/pdf/2512.18910.pdf

Abstract:
Multimodal Large Language Models (MLLMs) combine visual and textual representations to enable rich reasoning capabilities. However, the high computational cost of processing dense visual tokens remains a major bottleneck. A critical component in this pipeline is the visual projector, which bridges the vision encoder and the language model. Standard designs often employ a simple multi-layer perceptron for direct token mapping, but this approach scales poorly with high-resolution inputs, introducing significant redundancy. We present Delta-LLaVA, a token-efficient projector that employs a low-rank DeltaProjection to align multi-level vision features into a compact subspace before further interaction. On top of this base alignment, lightweight Transformer blocks act as specialization layers, capturing both global and local structure under constrained token budgets. Extensive experiments and ablations demonstrate that this base-then-specialize design yields consistent gains across multiple benchmarks with only 144 tokens, highlighting the importance of token formation prior to scaling interaction capacity. With Delta-LLaVA, inference throughput improves by up to 55%, while end-to-end training accelerates by nearly 4-5x in pretraining and over 1.5x in finetuning, highlighting the dual benefits of our design in both efficiency and scalability.

Paperid: 6901, https://arxiv.org/pdf/2512.17228.pdf

Abstract:
Most digital music tools emphasize precision and control, but often lack support for tactile, improvisational workflows grounded in environmental interaction. Lumia addresses this by enabling users to "compose through looking"--transforming visual scenes into musical phrases using a handheld, camera-based interface and large multimodal models. A vision-language model (GPT-4V) analyzes captured imagery to generate structured prompts, which, combined with user-selected instrumentation, guide a text-to-music pipeline (Stable Audio). This real-time process allows users to frame, capture, and layer audio interactively, producing loopable musical segments through embodied interaction. The system supports a co-creative workflow where human intent and model inference shape the musical outcome. By embedding generative AI within a physical device, Lumia bridges perception and composition, introducing a new modality for creative exploration that merges vision, language, and sound. It repositions generative music not as a task of parameter tuning, but as an improvisational practice driven by contextual, sensory engagement.

Paperid: 6902, https://arxiv.org/pdf/2512.17121.pdf

Abstract:
Large vision-language models like CLIP are increasingly used in medical imaging tasks due to their ability to align images and text without the need for extensive labeled data. This makes them particularly useful for applications like image retrieval, report generation, and classification in clinical settings. A potential issue to this approach is that CLIP-based models often under perform when interpreting negated phrases, which is especially problematic in the context of medical diagnosing. In this study, we evaluate the Stanford AIMI CheXagent model on its ability to correctly retrieve chest X-ray images using prompts with and without negation. The goal of this project is to understand where this model fails and then use it as a base model to improve its retrieval accuracy by fine tuning methods outlined in previous work. Results from this study show improvement in handling of negation in the CLIP model with a slight decrease in accuracy of positive prompt evaluation. Alongside retrieval accuracy, we examined internal model behavior through token attribution, t-SNE projection, and attention-head ablation to better characterize how each fine tuning approach reshaped the text encoders representation of negated clinical language. Through this work, we hope to better understand the internal behavior of CLIP and improve its handling of negation using clinically relevant language for improving its reliability in medical AI devices.

Paperid: 6903, https://arxiv.org/pdf/2512.16349.pdf

Abstract:
We propose a collaborative edge-to-server inference framework for vision-language models (VLMs) that reduces the communication cost while maintaining inference accuracy. In typical deployments, visual data captured at edge devices (clients) is transmitted to the server for VLM inference. However, resizing the original image (global image) to match the vision encoder's input resolution often discards fine-grained details, leading to accuracy degradation. To overcome this limitation, we design a two-stage framework. In the first stage, the server performs inference on the global image and identifies a region of interest (RoI) using the VLM's internal attention. The min-entropy of the output tokens is then computed as a confidence measure to determine whether retransmission is required. If the min-entropy exceeds a predefined threshold, the server requests the edge device to send a detail-preserved local image of the RoI. The server then refines its inference by jointly leveraging the global and local images. This selective retransmission strategy ensures that only essential visual content is transmitted. Experiments across multiple VLM architectures show that the proposed framework significantly reduces communication cost while maintaining inference accuracy.

Paperid: 6904, https://arxiv.org/pdf/2512.16237.pdf

Abstract:
Embodied intelligence, a grand challenge in artificial intelligence, is fundamentally constrained by the limited spatial understanding and reasoning capabilities of current models. Prevailing efforts to address this through enhancing Vision-Language Models (VLMs) are trapped in a dilemma: template-based datasets are scalable but structurally rigid, while manual annotation is linguistically diverse but unscalable and, critically, computationally imprecise. We introduce SPRITE, a novel framework that overcomes this dilemma by leveraging simulators and large models to programmatically synthesize scalable, diverse, and high-quality spatial reasoning data. The core innovation of SPRITE is to reframe ground-truth generation as a code-generation task. We utilize LLMs to compile complex spatial questions into executable programs, which are then verified against high-precision scene meta-information extracted from simulators. This ensures our ground truth is both computationally precise and verifiable, while the generative power of LLMs provides vast linguistic diversity. Leveraging this pipeline, we have curated a dataset encompassing 3 simulators, 11k+ scenes, and 300k+ image/video instruction-tuning pairs. We demonstrate that a VLM trained on our data achieves significant performance gains on multiple spatial benchmarks and outperforms other open-source datasets of equivalent size. Furthermore, a scalability analysis confirms our hypothesis that overcoming the low-diversity nature of traditional template methods is essential for building robust, generalizable spatial intelligence. We will make the SPRITE framework code and the full 300k+ dataset publicly available to facilitate future research in spatial intelligence.

Paperid: 6905, https://arxiv.org/pdf/2512.16164.pdf

Abstract:
Unsupervised Domain Adaptation transfers knowledge from a labeled source domain to an unlabeled target domain. Directly deploying Vision-Language Models (VLMs) with prompt tuning in downstream UDA tasks faces the signifi cant challenge of mitigating domain discrepancies. Existing prompt-tuning strategies primarily align marginal distribu tion, but neglect conditional distribution discrepancies, lead ing to critical issues such as class prototype misalignment and degraded semantic discriminability. To address these lim itations, the work proposes C-DGPA: Class-Centric Dual Alignment Generative Prompt Adaptation. C-DGPA syner gistically optimizes marginal distribution alignment and con ditional distribution alignment through a novel dual-branch architecture. The marginal distribution alignment branch em ploys a dynamic adversarial training framework to bridge marginal distribution discrepancies. Simultaneously, the con ditional distribution alignment branch introduces a Class Mapping Mechanism (CMM) to align conditional distribu tion discrepancies by standardizing semantic prompt under standing and preventing source domain over-reliance. This dual alignment strategy effectively integrates domain knowl edge into prompt learning via synergistic optimization, ensur ing domain-invariant and semantically discriminative repre sentations. Extensive experiments on OfficeHome, Office31, and VisDA-2017 validate the superiority of C-DGPA. It achieves new state-of-the-art results on all benchmarks.

Paperid: 6906, https://arxiv.org/pdf/2512.15977.pdf

Abstract:
Vision-language models (VLMs) are increasingly proposed as general-purpose solutions for visual recognition tasks, yet their reliability for agricultural decision support remains poorly understood. We benchmark a diverse set of open-source and closed-source VLMs on 27 agricultural classification datasets from the AgML collection, spanning 162 classes across plant disease, pest and damage, and plant and weed species identification. Across all tasks, zero-shot VLMs substantially underperform a supervised task-specific baseline (YOLO11), which consistently achieves markedly higher accuracy than any foundation model. Under multiple-choice prompting, the best-performing VLM (Gemini-3 Pro) reaches approximately 62% average accuracy, while open-ended prompting yields much lower performance, with raw accuracies typically below 25%. Applying LLM-based semantic judging increases open-ended accuracy (for example, from 21% to 30% for top models) and alters model rankings, demonstrating that evaluation methodology meaningfully affects reported conclusions. Among open-source models, Qwen-VL-72B performs best, approaching closed-source performance under constrained prompting but still trailing top proprietary systems. Task-level analysis shows that plant and weed species classification is consistently easier than pest and damage identification, which remains the most challenging category across models. Overall, these results indicate that current off-the-shelf VLMs are not yet suitable as standalone agricultural diagnostic systems, but can function as assistive components when paired with constrained interfaces, explicit label ontologies, and domain-aware evaluation strategies.

Paperid: 6907, https://arxiv.org/pdf/2512.14732.pdf

Abstract:
Incidental findings in CT scans, though often benign, can have significant clinical implications and should be reported following established guidelines. Traditional manual inspection by radiologists is time-consuming and variable. This paper proposes a novel framework that leverages large language models (LLMs) and foundational vision-language models (VLMs) in a plan-and-execute agentic approach to improve the efficiency and precision of incidental findings detection, classification, and reporting for abdominal CT scans. Given medical guidelines for abdominal organs, the process of managing incidental findings is automated through a planner-executor framework. The planner, based on LLM, generates Python scripts using predefined base functions, while the executor runs these scripts to perform the necessary checks and detections, via VLMs, segmentation models, and image processing subroutines. We demonstrate the effectiveness of our approach through experiments on a CT abdominal benchmark for three organs, in a fully automatic end-to-end manner. Our results show that the proposed framework outperforms existing pure VLM-based approaches in terms of accuracy and efficiency.

Paperid: 6908, https://arxiv.org/pdf/2512.14312.pdf

Abstract:
In regions of the Middle East and North Africa (MENA), there is a high demand for wastewater treatment plants (WWTPs), crucial for sustainable water management. Precise identification of WWTPs from satellite images enables environmental monitoring. Traditional methods like YOLOv8 segmentation require extensive manual labeling. But studies indicate that vision-language models (VLMs) are an efficient alternative to achieving equivalent or superior results through inherent reasoning and annotation. This study presents a structured methodology for VLM comparison, divided into zero-shot and few-shot streams specifically to identify WWTPs. The YOLOv8 was trained on a governmental dataset of 83,566 high-resolution satellite images from Egypt, Saudi Arabia, and UAE: ~85% WWTPs (positives), 15% non-WWTPs (negatives). Evaluated VLMs include LLaMA 3.2 Vision, Qwen 2.5 VL, DeepSeek-VL2, Gemma 3, Gemini, and Pixtral 12B (Mistral), used to identify WWTP components such as circular/rectangular tanks, aeration basins and distinguish confounders via expert prompts producing JSON outputs with confidence and descriptions. The dataset comprises 1,207 validated WWTP locations (198 UAE, 354 KSA, 655 Egypt) and equal non-WWTP sites from field/AI data, as 600mx600m Geo-TIFF images (Zoom 18, EPSG:4326). Zero-shot evaluations on WWTP images showed several VLMs out-performing YOLOv8's true positive rate, with Gemma-3 highest. Results confirm that VLMs, particularly with zero-shot, can replace YOLOv8 for efficient, annotation-free WWTP classification, enabling scalable remote sensing.

Paperid: 6909, https://arxiv.org/pdf/2512.14102.pdf

Abstract:
Text-to-image retrieval in remote sensing (RS) has advanced rapidly with the rise of large vision-language models (LVLMs) tailored for aerial and satellite imagery, culminating in remote sensing large vision-language models (RS-LVLMS). However, limited explainability and poor handling of complex spatial relations remain key challenges for real-world use. To address these issues, we introduce RUNE (Reasoning Using Neurosymbolic Entities), an approach that combines Large Language Models (LLMs) with neurosymbolic AI to retrieve images by reasoning over the compatibility between detected entities and First-Order Logic (FOL) expressions derived from text queries. Unlike RS-LVLMs that rely on implicit joint embeddings, RUNE performs explicit reasoning, enhancing performance and interpretability. For scalability, we propose a logic decomposition strategy that operates on conditioned subsets of detected entities, guaranteeing shorter execution time compared to neural approaches. Rather than using foundation models for end-to-end retrieval, we leverage them only to generate FOL expressions, delegating reasoning to a neurosymbolic inference module. For evaluation we repurpose the DOTA dataset, originally designed for object detection, by augmenting it with more complex queries than in existing benchmarks. We show the LLM's effectiveness in text-to-logic translation and compare RUNE with state-of-the-art RS-LVLMs, demonstrating superior performance. We introduce two metrics, Retrieval Robustness to Query Complexity (RRQC) and Retrieval Robustness to Image Uncertainty (RRIU), which evaluate performance relative to query complexity and image uncertainty. RUNE outperforms joint-embedding models in complex RS retrieval tasks, offering gains in performance, robustness, and explainability. We show RUNE's potential for real-world RS applications through a use case on post-flood satellite image retrieval.

Paperid: 6910, https://arxiv.org/pdf/2512.13855.pdf

Abstract:
Adapting Vision Language Segmentation Models (VLSMs) to medical imaging domains requires significant computational overhead when using conventional fine-tuning approaches. Existing Parameter-Efficient Fine-Tuning (PEFT) methods apply uniform adapter dimensions across all transformer layers, leading to suboptimal parameter allocation and reduced adaptation efficiency. We introduce Telescopic Adapters, a novel PEFT framework that employs depth-aware scaling to progressively increase adapter capacity from shallow to deep transformer layers. Our method integrates lightweight bottleneck modules within CLIPSeg's vision and text encoders, with adapter dimensions dynamically scaled based on layer depth and semantic relevance. Using only 613k trainable parameters--244x fewer than end-to-end fine-tuning, Telescopic Adapters achieve superior performance across five diverse medical datasets spanning polyp segmentation, skin lesion detection, and breast ultrasound imaging. Comprehensive ablation studies demonstrate that deeper layers require substantially more adaptation capacity than shallow layers, validating our telescopic scaling hypothesis. Our approach establishes a new paradigm for efficient medical VLSM fine-tuning, enabling deployment in resource-constrained clinical environments while maintaining competitive segmentation accuracy.

Paperid: 6911, https://arxiv.org/pdf/2512.13285.pdf

Abstract:
The rapid advancement of generative models has increased the demand for generated image detectors capable of generalizing across diverse and evolving generation techniques. However, existing methods, including those leveraging pre-trained vision-language models, often produce highly entangled representations, mixing task-relevant forensic cues (causal features) with spurious or irrelevant patterns (non-causal features), thus limiting generalization. To address this issue, we propose CausalCLIP, a framework that explicitly disentangles causal from non-causal features and employs targeted filtering guided by causal inference principles to retain only the most transferable and discriminative forensic cues. By modeling the generation process with a structural causal model and enforcing statistical independence through Gumbel-Softmax-based feature masking and Hilbert-Schmidt Independence Criterion (HSIC) constraints, CausalCLIP isolates stable causal features robust to distribution shifts. When tested on unseen generative models from different series, CausalCLIP demonstrates strong generalization ability, achieving improvements of 6.83% in accuracy and 4.06% in average precision over state-of-the-art methods.

Paperid: 6912, https://arxiv.org/pdf/2512.13240.pdf

Abstract:
Direct Preference Optimization (DPO) has emerged as a lightweight and effective alternative to Reinforcement Learning from Human Feedback (RLHF) and Reinforcement Learning with AI Feedback (RLAIF) for aligning large language and vision-language models. However, the standard DPO formulation, in which both the chosen and rejected responses are generated by the same policy, suffers from a weak learning signal because the two responses often share similar errors and exhibit small Kullback-Leibler (KL) divergence. This leads to slow and unstable convergence. To address this limitation, we introduce Reflective Preference Optimization (RPO), a new framework that incorporates hint-guided reflection into the DPO paradigm. RPO uses external models to identify hallucination sources and generate concise reflective hints, enabling the construction of on-policy preference pairs with stronger contrastiveness and clearer preference signals. We theoretically show that conditioning on hints increases the expected preference margin through mutual information and improves sample efficiency while remaining within the policy distribution family. Empirically, RPO achieves superior alignment with fewer training samples and iterations, substantially reducing hallucination rates and delivering state-of-the-art performance across multimodal benchmarks.

Paperid: 6913, https://arxiv.org/pdf/2512.12701.pdf

Abstract:
Real-world deployment of Vision-Language Models (VLMs) is hindered by high computational demands, as existing architectures inefficiently process all tokens uniformly. We introduce Adaptive Token Pruning (ATP), a dynamic inference mechanism that retains only the most informative tokens based on contextual relevance. ATP operates at the vision-language interface, assigning a hybrid importance score combining ViT CLS attention (intra-modal saliency) and CLIP text-image similarity (inter-modal relevance) to keep top-K tokens for the LLM. Unlike static compression, ATP adapts to each input without modifying the backbone. Proposed as a lightweight gating module, ATP is compatible with popular backbones like BLIP-2, LLaVA, and Flamingo. Preliminary evaluations across VQAv2, GQA, and COCO indicate that ATP reduces inference FLOPs by around 40% and achieves roughly 1.5x speedups in end-to-end latency with negligible accuracy loss (less than 1%). Qualitative analyses suggest ATP preserves visual grounding and enhances interpretability. Beyond efficiency, we investigate robustness under corruptions; observations suggest adaptive pruning suppresses spurious correlations, improving stability. These findings imply that resource-constrained inference and model reliability are not competing objectives. Finally, we discuss ATP's role in efficient multimodal edge computing pipelines.

Paperid: 6914, https://arxiv.org/pdf/2512.12424.pdf

Abstract:
Infographic Visual Question Answering (InfographicVQA) evaluates a model's ability to read and reason over data-rich, layout-heavy visuals that combine text, charts, icons, and design elements. Compared with scene-text or natural-image VQA, infographics require stronger integration of OCR, layout understanding, and numerical and semantic reasoning. We introduce ViInfographicVQA, the first benchmark for Vietnamese InfographicVQA, comprising over 6747 real-world infographics and 20409 human-verified question-answer pairs across economics, healthcare, education, and more. The benchmark includes two evaluation settings. The Single-image task follows the traditional setup in which each question is answered using a single infographic. The Multi-image task requires synthesizing evidence across multiple semantically related infographics and is, to our knowledge, the first Vietnamese evaluation of cross-image reasoning in VQA. We evaluate a range of recent vision-language models on this benchmark, revealing substantial performance disparities, with the most significant errors occurring on Multi-image questions that involve cross-image integration and non-span reasoning. ViInfographicVQA contributes benchmark results for Vietnamese InfographicVQA and sheds light on the limitations of current multimodal models in low-resource contexts, encouraging future exploration of layout-aware and cross-image reasoning methods.

Paperid: 6915, https://arxiv.org/pdf/2512.11584.pdf

Abstract:
Current vision-language-action (VLA) models generalize poorly, particularly when tasks require new compositions of skills or objects. We introduce Atomic Action Slicing (AAS), a planner-aligned approach that decomposes long-horizon demonstrations into short, typed atomic actions that are easier for planners to use and policies to learn. Using LIBERO demonstrations, AAS produces a validated dataset of 2,124 atomic segments labeled with action type, temporal span, and confidence. A stronger segmenter (Gemini 2.5 Pro) closely matches planner-defined plans and remains robust under keyframe jitter, while smaller models perform worse on multi-object tasks. Fine-tuning CLIP-RT+ on our atomic dataset improves task success from 94.2% to 95.3% on LIBERO-Goal and 83.8% to 88.8% on LIBERO-Long. We publicly release the GATE-VLAP dataset on HuggingFace(https://huggingface.co/datasets/gate-institute/GATE-VLAP-datasets)

Paperid: 6916, https://arxiv.org/pdf/2512.11275.pdf

Abstract:
Existing pipelines for vision-language models (VLMs) in robotic manipulation prioritize broad semantic generalization from images and language, but typically omit execution-critical parameters required for contact-rich actions in manufacturing cells. We formalize an object-centric manipulation-logic schema, serialized as an eight-field tuple τ, which exposes object, interface, trajectory, tolerance, and force/impedance information as a first-class knowledge signal between human operators, VLM-based assistants, and robot controllers. We instantiate τ and a small knowledge base (KB) on a 3D-printer spool-removal task in a collaborative cell, and analyze τ-conditioned VLM planning using plan-quality metrics adapted from recent VLM/LLM planning benchmarks, while demonstrating how the same schema supports taxonomy-tagged data augmentation at training time and logic-aware retrieval-augmented prompting at test time as a building block for assistant systems in smart manufacturing enterprises.

Paperid: 6917, https://arxiv.org/pdf/2512.10888.pdf

Abstract:
Table extraction (TE) is a key challenge in visual document understanding. Traditional approaches detect tables first, then recognize their structure. Recently, interest has surged in developing methods, such as vision-language models (VLMs), that can extract tables directly in their full page or document context. However, progress has been difficult to demonstrate due to a lack of annotated data. To address this, we create a new large-scale dataset, PubTables-v2. PubTables-v2 supports a number of current challenging table extraction tasks. Notably, it is the first large-scale benchmark for multi-page table structure recognition. We demonstrate its usefulness by evaluating domain-specialized VLMs on these tasks and highlighting current progress. Finally, we use PubTables-v2 to create the Page-Object Table Transformer (POTATR), an image-to-graph extension of the Table Transformer to comprehensive page-level TE. Data, code, and trained models will be released.

Paperid: 6918, https://arxiv.org/pdf/2512.10596.pdf

Abstract:
Semantic retrieval of remote sensing (RS) images is a critical task fundamentally challenged by the \textquote{semantic gap}, the discrepancy between a model's low-level visual features and high-level human concepts. While large Vision-Language Models (VLMs) offer a promising path to bridge this gap, existing methods often rely on costly, domain-specific training, and there is a lack of benchmarks to evaluate the practical utility of VLM-generated text in a zero-shot retrieval context. To address this research gap, we introduce the Remote Sensing Rich Text (RSRT) dataset, a new benchmark featuring multiple structured captions per image. Based on this dataset, we propose a fully training-free, text-only retrieval reference called TRSLLaVA. Our methodology reformulates cross-modal retrieval as a text-to-text (T2T) matching problem, leveraging rich text descriptions as queries against a database of VLM-generated captions within a unified textual embedding space. This approach completely bypasses model training or fine-tuning. Experiments on the RSITMD and RSICD benchmarks show our training-free method is highly competitive with state-of-the-art supervised models. For instance, on RSITMD, our method achieves a mean Recall of 42.62\%, nearly doubling the 23.86\% of the standard zero-shot CLIP baseline and surpassing several top supervised models. This validates that high-quality semantic representation through structured text provides a powerful and cost-effective paradigm for remote sensing image retrieval.

Paperid: 6919, https://arxiv.org/pdf/2512.10336.pdf

Abstract:
Artificial intelligence has made great progress in recent years, particularly in the development of Vision--Language Models (VLMs) that understand both visual and textual data. However, these advancements remain largely limited to English, reducing their accessibility for non--English speakers. It is essential to extend these capabilities to a broader range of languages. This paper explores the challenges of adapting an English-trained VLM to different languages. To this end, we will explore and compare different methods for their performance and computational cost. We consider a translation-based pipeline, LoRA finetuning, and a two-stage finetuning strategy that separates vision adaptation from language adaptation. To evaluate these methods, we use a combination of standard multimodal benchmarks translated into the target language and manual assessments by native experts. The results reveal that dataset translation remains a major bottleneck in multilingual VLM performance, with data quality limiting the effectiveness of training and evaluation. These findings suggest that future efforts should focus on native-language dataset collection and improved translation strategies.

Paperid: 6920, https://arxiv.org/pdf/2512.10067.pdf

Abstract:
Large-scale Vision-Language models have achieved remarkable results in various domains, such as image captioning and conditioned image generation. Nevertheless, these models still encounter difficulties in achieving human-like compositional generalization. In this study, we propose a new method called Independent Density Estimation (IDE) to tackle this challenge. IDE aims to learn the connection between individual words in a sentence and the corresponding features in an image, enabling compositional generalization. We build two models based on the philosophy of IDE. The first one utilizes fully disentangled visual representations as input, and the second leverages a Variational Auto-Encoder to obtain partially disentangled features from raw images. Additionally, we propose an entropy-based compositional inference method to combine predictions of each word in the sentence. Our models exhibit superior generalization to unseen compositions compared to current models when evaluated on various datasets.

Paperid: 6921, https://arxiv.org/pdf/2512.09670.pdf

Abstract:
The proliferation of satellite constellations, coupled with reduced tasking latency and diverse sensor capabilities, has expanded the opportunities for automated Earth observation. This paper introduces a fully automated Tip-and-Cue framework designed for satellite imaging tasking and scheduling. In this context, tips are generated from external data sources or analyses of prior satellite imagery, identifying spatiotemporal targets and prioritizing them for downstream planning. Corresponding cues are the imaging tasks formulated in response, which incorporate sensor constraints, timing requirements, and utility functions. The system autonomously generates candidate tasks, optimizes their scheduling across multiple satellites using continuous utility functions that reflect the expected value of each observation, and processes the resulting imagery using artificial-intelligence-based models, including object detectors and vision-language models. Structured visual reports are generated to support both interpretability and the identification of new insights for downstream tasking. The efficacy of the framework is demonstrated through a maritime vessel tracking scenario, utilizing Automatic Identification System (AIS) data for trajectory prediction, targeted observations, and the generation of actionable outputs. Maritime vessel tracking is a widely researched application, often used to benchmark novel approaches to satellite tasking, forecasting, and analysis. The system is extensible to broader applications such as smart-city monitoring and disaster response, where timely tasking and automated analysis are critical.

Paperid: 6922, https://arxiv.org/pdf/2512.08580.pdf

Abstract:
Humans act with context and intention, with reasoning playing a central role. While internet-scale data has enabled broad reasoning capabilities in AI systems, grounding these abilities in physical action remains a major challenge. We introduce Lumo-1, a generalist vision-language-action (VLA) model that unifies robot reasoning ("mind") with robot action ("hand"). Our approach builds upon the general multi-modal reasoning capabilities of pre-trained vision-language models (VLMs), progressively extending them to embodied reasoning and action prediction, and ultimately towards structured reasoning and reasoning-action alignment. This results in a three-stage pre-training pipeline: (1) Continued VLM pre-training on curated vision-language data to enhance embodied reasoning skills such as planning, spatial understanding, and trajectory prediction; (2) Co-training on cross-embodiment robot data alongside vision-language data; and (3) Action training with reasoning process on trajectories collected on Astribot S1, a bimanual mobile manipulator with human-like dexterity and agility. Finally, we integrate reinforcement learning to further refine reasoning-action consistency and close the loop between semantic inference and motor control. Extensive experiments demonstrate that Lumo-1 achieves significant performance improvements in embodied vision-language reasoning, a critical component for generalist robotic control. Real-world evaluations further show that Lumo-1 surpasses strong baselines across a wide range of challenging robotic tasks, with strong generalization to novel objects and environments, excelling particularly in long-horizon tasks and responding to human-natural instructions that require reasoning over strategy, concepts and space.

Paperid: 6923, https://arxiv.org/pdf/2512.08271.pdf

Abstract:
We introduce Zero-Splat TeleAssist, a zero-shot sensor-fusion pipeline that transforms commodity CCTV streams into a shared, 6-DoF world model for multilateral teleoperation. By integrating vision-language segmentation, monocular depth, weighted-PCA pose extraction, and 3D Gaussian Splatting (3DGS), TeleAssist provides every operator with real-time global positions and orientations of multiple robots without fiducials or depth sensors in an interaction-centric teleoperation setup.

Paperid: 6924, https://arxiv.org/pdf/2512.07452.pdf

Abstract:
Many heritage institutions hold extensive collections of theatre programmes, which remain largely underused due to their complex layouts and lack of structured metadata. In this paper, we present a workflow for transforming such documents into structured data using a combination of multimodal large language models (LLMs), an ontology-based reasoning model, and a custom extension of the Linked Art framework. We show how vision-language models can accurately parse and transcribe born-digital and digitised programmes, achieving over 98% of correct extraction. To overcome the challenges of semantic annotation, we train a reasoning model (POntAvignon) using reinforcement learning with both formal and semantic rewards. This approach enables automated RDF triple generation and supports alignment with existing knowledge graphs. Through a case study based on the Festival d'Avignon corpus, we demonstrate the potential for large-scale, ontology-driven analysis of performing arts data. Our results open new possibilities for interoperable, explainable, and sustainable computational theatre historiography.

Paperid: 6925, https://arxiv.org/pdf/2512.04895.pdf

Abstract:
Multimodal Artificial Intelligence (AI) systems, particularly Vision-Language Models (VLMs), have become integral to critical applications ranging from autonomous decision-making to automated document processing. As these systems scale, they rely heavily on preprocessing pipelines to handle diverse inputs efficiently. However, this dependency on standard preprocessing operations, specifically image downscaling, creates a significant yet often overlooked security vulnerability. While intended for computational optimization, scaling algorithms can be exploited to conceal malicious visual prompts that are invisible to human observers but become active semantic instructions once processed by the model. Current adversarial strategies remain largely static, failing to account for the dynamic nature of modern agentic workflows. To address this gap, we propose Chameleon, a novel, adaptive adversarial framework designed to expose and exploit scaling vulnerabilities in production VLMs. Unlike traditional static attacks, Chameleon employs an iterative, agent-based optimization mechanism that dynamically refines image perturbations based on the target model's real-time feedback. This allows the framework to craft highly robust adversarial examples that survive standard downscaling operations to hijack downstream execution. We evaluate Chameleon against Gemini 2.5 Flash model. Our experiments demonstrate that Chameleon achieves an Attack Success Rate (ASR) of 84.5% across varying scaling factors, significantly outperforming static baseline attacks which average only 32.1%. Furthermore, we show that these attacks effectively compromise agentic pipelines, reducing decision-making accuracy by over 45% in multi-step tasks. Finally, we discuss the implications of these vulnerabilities and propose multi-scale consistency checks as a necessary defense mechanism.

Paperid: 6926, https://arxiv.org/pdf/2512.03623.pdf

Abstract:
Despite the promising capability of multimodal foundation models, their application to the generation of meteorological products and services remains nascent. To accelerate aspiration and adoption, we explore the novel use of a vision language model for writing the iconic Shipping Forecast text directly from video-encoded gridded weather data. These early results demonstrate promising scalable technological opportunities for enhancing production efficiency and service innovation within the weather enterprise and beyond.

Paperid: 6927, https://arxiv.org/pdf/2512.03514.pdf

Abstract:
Multimodal document retrieval systems have shown strong progress in aligning visual and textual content for semantic search. However, most existing approaches remain heavily English-centric, limiting their effectiveness in multilingual contexts. In this work, we present M3DR (Multilingual Multimodal Document Retrieval), a framework designed to bridge this gap across languages, enabling applicability across diverse linguistic and cultural contexts. M3DR leverages synthetic multilingual document data and generalizes across different vision-language architectures and model sizes, enabling robust cross-lingual and cross-modal alignment. Using contrastive training, our models learn unified representations for text and document images that transfer effectively across languages. We validate this capability on 22 typologically diverse languages, demonstrating consistent performance and adaptability across linguistic and script variations. We further introduce a comprehensive benchmark that captures real-world multilingual scenarios, evaluating models under monolingual, multilingual, and mixed-language settings. M3DR generalizes across both single dense vector and ColBERT-style token-level multi-vector retrieval paradigms. Our models, NetraEmbed and ColNetraEmbed achieve state-of-the-art performance with ~150% relative improvements on cross-lingual retrieval.

Paperid: 6928, https://arxiv.org/pdf/2512.02456.pdf

Abstract:
Vision-Language Models (VLMs) have achieved remarkable progress in integrating visual perception with language understanding. However, effective multimodal reasoning requires both accurate perception and robust reasoning, and weakness in either limits the performance of VLMs. Prior efforts to enhance reasoning often depend on high-quality chain-of-thought (CoT) data, obtained via labor-intensive human annotations, costly proprietary models, or self-training methods that overlook perception. To address these limitations, we propose a simple yet effective self-training framework called See-Think-Learn (STL). At its core, STL introduces a structured reasoning template that encourages the model to see before thinking, first extracting visual attributes in textual form, then using them to guide reasoning. The framework jointly improves perception and reasoning by having the model generate and learn from its own structured rationales in a self-training loop. Furthermore, we augment the training data with negative rationales, i.e. explanations that justify why certain answer choices are incorrect, to enhance the model's ability to distinguish between correct and misleading responses. This fosters more discriminative and robust learning. Experiments across diverse domains show that STL consistently outperforms baselines trained directly only on answers or self-generated reasoning, while qualitative analysis confirms the high quality of its rationales. STL thus provides a cost-effective solution to enhance multimodal reasoning ability of VLMs.

Paperid: 6929, https://arxiv.org/pdf/2512.01419.pdf

Abstract:
Vision-Language Models (VLMs) excel in multimodal tasks but often exhibit Western-centric biases, limiting their effectiveness in culturally diverse regions like Southeast Asia (SEA). To address this, we introduce RICE-VL, a novel benchmark evaluating VLM cultural understanding across 11 ASEAN countries. RICE-VL includes over 28,000 human-curated Visual Question Answering (VQA) samples -- covering True or False, Fill-in-the-Blank, and open-ended formats -- and 1,000 image-bounding box pairs for Visual Grounding, annotated by culturally informed experts across 14 sub-ground categories. We propose SEA-LAVE, an extension of the LAVE metric, assessing textual accuracy, cultural alignment, and country identification. Evaluations of six open- and closed-source VLMs reveal significant performance gaps in low-resource countries and abstract cultural domains. The Visual Grounding task tests models' ability to localize culturally significant elements in complex scenes, probing spatial and contextual accuracy. RICE-VL exposes limitations in VLMs' cultural comprehension and highlights the need for inclusive model development to better serve diverse global populations.

Paperid: 6930, https://arxiv.org/pdf/2512.01148.pdf

Abstract:
Understanding social interactions from visual cues is a fundamental challenge for a socially competent AI. While powerful pre-trained vision-language models (VLMs) have shown remarkable general capabilities, they surprisingly struggle to unify and learn multiple social perception tasks simultaneously, often exhibiting negative transfer. We identify that this negative transfer stems from a critical issue we term "social degradation," whereby the general visual-linguistic pre-training process of VLMs impairs the visual encoder's ability to represent nuanced social information. We investigate this behavior further under two lenses: decodability through linear representation probing and compatibility through gradient conflict analysis, revealing that both play a role in the degradation, especially the former, which is significantly compromised in the VLM pre-training process. To address these issues, we propose SocialFusion, a unified framework that learns a minimal connection between a frozen visual encoder and a language model. Compared with existing VLMs, it exhibits positive transfer across all five social tasks, leveraging synergies between them to enhance overall performance and achieves comparable performance to task-specific state-of-the-art models on various benchmarks. Our findings suggest that current VLM pre-training strategies may be detrimental to acquiring general social competence and highlight the need for more socially-aware training paradigms.

Paperid: 6931, https://arxiv.org/pdf/2512.00294.pdf

Abstract:
Traditional augmented reality (AR) systems predominantly rely on fixed class detectors or fiducial markers, limiting their ability to interpret complex, open-vocabulary natural language queries. We present a modular AR agent system that integrates multimodal large language models (MLLMs) with grounded vision models to enable relational reasoning in space and language-conditioned spatial retrieval in physical environments. Our adaptive task agent coordinates MLLMs and coordinate-aware perception tools to address varying query complexities, ranging from simple object identification to multi-object relational reasoning, while returning meter-accurate 3D anchors. It constructs dynamic AR scene graphs encoding nine typed relations (spatial, structural-semantic, causal-functional), enabling MLLMs to understand not just what objects exist, but how they relate and interact in 3D space. Through task-adaptive region-of-interest highlighting and contextual spatial retrieval, the system guides human attention to information-dense areas while supporting human-in-the-loop refinement. The agent dynamically invokes coordinate-aware tools for complex queries-selection, measurement, comparison, and actuation-grounding language understanding in physical operations. The modular architecture supports plug-and-use vision-language models without retraining, establishing AR agents as intermediaries that augment MLLMs with real-world spatial intelligence for interactive scene understanding. We also introduce GroundedAR-Bench, an evaluation framework for language-driven real world localization and relation grounding across diverse environments.

Paperid: 6932, https://arxiv.org/pdf/2512.00042.pdf

Abstract:
Multimodal reasoning has become a cornerstone of modern AI research. Standardized exam questions offer a uniquely rigorous testbed for such reasoning, providing structured visual contexts and verifiable answers. While recent progress has largely focused on algorithmic advances such as reinforcement learning (e.g., GRPO, DPO), the data centric foundations of vision language reasoning remain less explored. We show that supervised fine-tuning (SFT) with high-quality data can rival proprietary approaches. To this end, we compile a 161.4 million token multimodal dataset combining textbook question-solution pairs, curriculum aligned diagrams, and contextual materials, and fine-tune Qwen-2.5VL-32B using an optimized reasoning syntax (QMSA). The resulting model achieves 78.6% accuracy, only 1.0% below Gemini 2.0 Flash, on our newly released benchmark YKSUniform, which standardizes 1,854 multimodal exam questions across 309 curriculum topics. Our results reveal that data composition and representational syntax play a decisive role in multimodal reasoning. This work establishes a data centric framework for advancing open weight vision language models, demonstrating that carefully curated and curriculum-grounded multimodal data can elevate supervised fine-tuning to near state-of-the-art performance.

Paperid: 6933, https://arxiv.org/pdf/2511.22929.pdf

Abstract:
Emotions are a fundamental aspect of artistic expression. Due to their abstract nature, there is a broad spectrum of emotion realization in artworks. These are subject to historical change and their analysis requires expertise in art history. In this article, we investigate which aspects of emotional expression can be detected by current (2025) vision language models (VLMs). We present a case study of three VLMs (Llava-Llama and two Qwen models) in which we ask these models four sets of questions of increasing complexity about artworks (general content, emotional content, expression of emotions, and emotion symbols) and carry out a qualitative expert evaluation. We find that the VLMs recognize the content of the images surprisingly well and often also which emotions they depict and how they are expressed. The models perform best for concrete images but fail for highly abstract or highly symbolic images. Reliable recognition of symbols remains fundamentally difficult. Furthermore, the models continue to exhibit the well-known LLM weakness of providing inconsistent answers to related questions.

Paperid: 6934, https://arxiv.org/pdf/2511.22532.pdf

Abstract:
Vision-Language-Action (VLA) models have recently attracted growing attention in end-to-end autonomous driving for their strong reasoning capabilities and rich world knowledge. However, existing VLAs often suffer from limited numerical reasoning ability and overly simplified input-output mappings, which hinder their performance in complex driving scenarios requiring step-by-step causal reasoning. To address these challenges, we propose CoT4AD, a novel VLA framework that introduces Chain-of-Thought (CoT) reasoning for autonomous driving to enhance both numerical and causal reasoning in Vision-Language Models (VLMs). CoT4AD integrates visual observations and language instructions to perform semantic reasoning, scene understanding, and trajectory planning. During training, it explicitly models a perception-question-prediction-action CoT to align the reasoning space with the action space across multiple driving tasks. During inference, it performs implicit CoT reasoning to enable consistent numerical reasoning and robust decision-making in dynamic environments. Extensive experiments on both real-world and simulated benchmarks, including nuScenes and Bench2Drive, demonstrate that CoT4AD achieves state-of-the-art performance in both open-loop and closed-loop evaluations. Code will be released upon paper acceptance.

Paperid: 6935, https://arxiv.org/pdf/2511.21889.pdf

Abstract:
Modern machine learning models often combine multiple input streams of data to more accurately capture the information that informs their decisions. In multimodal machine learning, choosing the strategy for fusing data together requires careful consideration of the application's accuracy and latency requirements, as fusing the data at earlier or later stages in the model architecture can lead to performance changes in accuracy and latency. To demonstrate this tradeoff, we investigate different fusion strategies using a hybrid BERT and vision network framework that integrates image and text data. We explore two different vision networks: MobileNetV2 and ViT. We propose three models for each vision network, which fuse data at late, intermediate, and early stages in the architecture. We evaluate the proposed models on the CMU MOSI dataset and benchmark their latency on an NVIDIA Jetson Orin AGX. Our experimental results demonstrate that while late fusion yields the highest accuracy, early fusion offers the lowest inference latency. We describe the three proposed model architectures and discuss the accuracy and latency tradeoffs, concluding that data fusion earlier in the model architecture results in faster inference times at the cost of accuracy.

Paperid: 6936, https://arxiv.org/pdf/2511.21428.pdf

Abstract:
We present a novel unsupervised framework to unlock vast unlabeled human demonstration data from continuous industrial video streams for Vision-Language-Action (VLA) model pre-training. Our method first trains a lightweight motion tokenizer to encode motion dynamics, then employs an unsupervised action segmenter leveraging a novel "Latent Action Energy" metric to discover and segment semantically coherent action primitives. The pipeline outputs both segmented video clips and their corresponding latent action sequences, providing structured data directly suitable for VLA pre-training. Evaluations on public benchmarks and a proprietary electric motor assembly dataset demonstrate effective segmentation of key tasks performed by humans at workstations. Further clustering and quantitative assessment via a Vision-Language Model confirm the semantic coherence of the discovered action primitives. To our knowledge, this is the first fully automated end-to-end system for extracting and organizing VLA pre-training data from unstructured industrial videos, offering a scalable solution for embodied AI integration in manufacturing.

Paperid: 6937, https://arxiv.org/pdf/2511.20795.pdf

Abstract:
Facebook AI Research introduced KRISP [4], which integrates structured external knowledge into pipelines for vision-language reasoning. Despite its effectiveness, the original model has been developed for industrial-scale training, is computationally demanding, and is tightly connected to a large backbone. In this work, we reexamine KRISP from a different angle and offer a lightweight reproduction with significantly fewer parameters. Even though our replicated model performs about 75 % of the original, the replication process uncovers a number of design flaws, real-world pitfalls, and implicit problems that were not fully covered in the original paper. We offer insights into the scalability and efficacy of knowledge-enhanced VQA architectures under resource constraints through systematic ablation studies, which include a proof-of-concept on synthetic VQA data and evaluation on the DAQUAR dataset. Our model, configured with a low parameter setup and constrained by the external Knowledge graph domain, prevents AI hallucinations and generates outputs solely within that domain. Minimal parameters allow us to function on edge devices like smartphones and AR-VR, further improving offline visual reasoning.

Paperid: 6938, https://arxiv.org/pdf/2511.20710.pdf

Abstract:
In the age of agentic AI, the growing deployment of multi-modal models (MMs) has introduced new attack vectors that can leak sensitive training data in MMs, causing privacy leakage. This paper investigates a black-box privacy attack, i.e., membership inference attack (MIA) on multi-modal vision-language models (VLMs). State-of-the-art research analyzes privacy attacks primarily to unimodal AI-ML systems, while recent studies indicate MMs can also be vulnerable to privacy attacks. While researchers have demonstrated that biologically inspired neural network representations can improve unimodal model resilience against adversarial attacks, it remains unexplored whether neuro-inspired MMs are resilient against privacy attacks. In this work, we introduce a systematic neuroscience-inspired topological regularization (tau) framework to analyze MM VLMs resilience against image-text-based inference privacy attacks. We examine this phenomenon using three VLMs: BLIP, PaliGemma 2, and ViT-GPT2, across three benchmark datasets: COCO, CC3M, and NoCaps. Our experiments compare the resilience of baseline and neuro VLMs (with topological regularization), where the tau > 0 configuration defines the NEURO variant of VLM. Our results on the BLIP model using the COCO dataset illustrate that MIA attack success in NEURO VLMs drops by 24% mean ROC-AUC, while achieving similar model utility (similarities between generated and reference captions) in terms of MPNet and ROUGE-2 metrics. This shows neuro VLMs are comparatively more resilient against privacy attacks, while not significantly compromising model utility. Our extensive evaluation with PaliGemma 2 and ViT-GPT2 models, on two additional datasets: CC3M and NoCaps, further validates the consistency of the findings. This work contributes to the growing understanding of privacy risks in MMs and provides evidence on neuro VLMs privacy threat resilience.

Paperid: 6939, https://arxiv.org/pdf/2511.20627.pdf

Abstract:
The integration of AI components, particularly Deep Neural Networks (DNNs), into safety-critical systems such as aerospace and autonomous vehicles presents fundamental challenges for assurance. The opacity of AI systems, combined with the semantic gap between high-level requirements and low-level network representations, creates barriers to traditional verification approaches. These AI-specific challenges are amplified by longstanding issues in Requirements Engineering, including ambiguity in natural language specifications and scalability bottlenecks in formalization. We propose an approach that leverages AI itself to address these challenges through two complementary components. REACT (Requirements Engineering with AI for Consistency and Testing) employs Large Language Models (LLMs) to bridge the gap between informal natural language requirements and formal specifications, enabling early verification and validation. SemaLens (Semantic Analysis of Visual Perception using large Multi-modal models) utilizes Vision Language Models (VLMs) to reason about, test, and monitor DNN-based perception systems using human-understandable concepts. Together, these components provide a comprehensive pipeline from informal requirements to validated implementations.

Paperid: 6940, https://arxiv.org/pdf/2511.20067.pdf

Abstract:
Computer Use Agents (CUAs) are designed to autonomously operate digital interfaces, yet they often fail to reliably determine whether a given task has been completed. We present an autonomous evaluation and feedback framework that uses vision-language models to assess task completion directly from screenshots and task descriptions. Our dataset covers 42 built-in macOS applications and 1,260 human-labeled tasks across a wide range of scenarios. Our framework achieves up to 73 percent accuracy in task success detection and yields an average relative improvement of 27 percent in overall task success when evaluator feedback is applied. These results show that vision-based evaluation can serve as an effective feedback mechanism that improves the reliability and self-correction of autonomous computer-use agents.

Paperid: 6941, https://arxiv.org/pdf/2511.19923.pdf

Abstract:
Vision Language Models (VLMs) have recently shown significant advancements in video understanding, especially in feature alignment, event reasoning, and instruction-following tasks. However, their capability for counterfactual reasoning, inferring alternative outcomes under hypothetical conditions, remains underexplored. This capability is essential for robust video understanding, as it requires identifying underlying causal structures and reasoning about unobserved possibilities, rather than merely recognizing observed patterns. To systematically evaluate this capability, we introduce CounterVQA, a video-based benchmark featuring three progressive difficulty levels that assess different aspects of counterfactual reasoning. Through comprehensive evaluation of both state-of-the-art open-source and closed-source models, we uncover a substantial performance gap: while these models achieve reasonable accuracy on simple counterfactual questions, performance degrades significantly on complex multi-hop causal chains. To address these limitations, we develop a post-training method, CFGPT, that enhances a model's visual counterfactual reasoning ability by distilling its counterfactual reasoning capability from the language modality, yielding consistent improvements across all CounterVQA difficulty levels. Dataset and code will be further released.

Paperid: 6942, https://arxiv.org/pdf/2511.19759.pdf

Abstract:
Semi-supervised learning (SSL) has emerged as an effective paradigm for medical image segmentation, reducing the reliance on extensive expert annotations. Meanwhile, vision-language models (VLMs) have demonstrated strong generalization and few-shot capabilities across diverse visual domains. In this work, we integrate VLM-based segmentation into semi-supervised medical image segmentation by introducing a Vision-Language Enhanced Semi-supervised Segmentation Assistant (VESSA) that incorporates foundation-level visual-semantic understanding into SSL frameworks. Our approach consists of two stages. In Stage 1, the VLM-enhanced segmentation foundation model VESSA is trained as a reference-guided segmentation assistant using a template bank containing gold-standard exemplars, simulating learning from limited labeled data. Given an input-template pair, VESSA performs visual feature matching to extract representative semantic and spatial cues from exemplar segmentations, generating structured prompts for a SAM2-inspired mask decoder to produce segmentation masks. In Stage 2, VESSA is integrated into a state-of-the-art SSL framework, enabling dynamic interaction with the student model: as student predictions become more refined, they are fed back to VESSA as prompts, allowing it to generate higher-quality pseudo-labels and stronger guidance. Extensive experiments across multiple segmentation datasets and domains show that VESSA-augmented SSL significantly enhances segmentation accuracy, outperforming state-of-the-art baselines under extremely limited annotation conditions.

Paperid: 6943, https://arxiv.org/pdf/2511.19751.pdf

Abstract:
Despite the promise of computational pathology foundation models, adapting them to specific clinical tasks remains challenging due to the complexity of whole-slide image (WSI) processing, the opacity of learned features, and the wide range of potential adaptation strategies. To address these challenges, we introduce PathFMTools, a lightweight, extensible Python package that enables efficient execution, analysis, and visualization of pathology foundation models. We use this tool to interface with and evaluate two state-of-the-art vision-language foundation models, CONCH and MUSK, on the task of histological grading in cutaneous squamous cell carcinoma (cSCC), a critical criterion that informs cSCC staging and patient management. Using a cohort of 440 cSCC H&E WSIs, we benchmark multiple adaptation strategies, demonstrating trade-offs across prediction approaches and validating the potential of using foundation model embeddings to train small specialist models. These findings underscore the promise of pathology foundation models for real-world clinical applications, with PathFMTools enabling efficient analysis and validation.

Paperid: 6944, https://arxiv.org/pdf/2511.19220.pdf

Abstract:
Large vision language models (VLMs) have achieved impressive performance on medical visual question answering benchmarks, yet their reliance on visual information remains unclear. We investigate whether frontier VLMs demonstrate genuine visual grounding when answering Italian medical questions by testing four state-of-the-art models: Claude Sonnet 4.5, GPT-4o, GPT-5-mini, and Gemini 2.0 flash exp. Using 60 questions from the EuropeMedQA Italian dataset that explicitly require image interpretation, we substitute correct medical images with blank placeholders to test whether models truly integrate visual and textual information. Our results reveal striking variability in visual dependency: GPT-4o shows the strongest visual grounding with a 27.9pp accuracy drop (83.2% [74.6%, 91.7%] to 55.3% [44.1%, 66.6%]), while GPT-5-mini, Gemini, and Claude maintain high accuracy with modest drops of 8.5pp, 2.4pp, and 5.6pp respectively. Analysis of model-generated reasoning reveals confident explanations for fabricated visual interpretations across all models, suggesting varying degrees of reliance on textual shortcuts versus genuine visual analysis. These findings highlight critical differences in model robustness and the need for rigorous evaluation before clinical deployment.

Paperid: 6945, https://arxiv.org/pdf/2511.18787.pdf

Abstract:
Vision-Language Models (VLMs) perform well on multimodal benchmarks but lag behind humans and specialized models on visual perception tasks like depth estimation or object counting. Finetuning on one task can unpredictably affect performance on others, making task-specific finetuning challenging. In this paper, we address this challenge through a systematic study of task transferability. We examine how finetuning a VLM on one perception task affects its zero-shot performance on others. To quantify these effects, we introduce Perfection Gap Factor (PGF), a metric that captures both the breadth and magnitude of transfer. Using three open-weight VLMs evaluated across 13 perception tasks, we construct a task-transfer graph that reveals previously unobserved relationships among perception tasks. Our analysis uncovers patterns of positive and negative transfer, identifies groups of tasks that mutually influence each other, organizes tasks into personas based on their transfer behavior and demonstrates how PGF can guide data selection for more efficient training. These findings highlight both opportunities for positive transfer and risks of negative interference, offering actionable guidance for advancing VLMs.

Paperid: 6946, https://arxiv.org/pdf/2511.18746.pdf

Abstract:
While video-generation-based embodied world models have gained increasing attention, their reliance on large-scale embodied interaction data remains a key bottleneck. The scarcity, difficulty of collection, and high dimensionality of embodied data fundamentally limit the alignment granularity between language and actions and exacerbate the challenge of long-horizon video generation--hindering generative models from achieving a \textit{"GPT moment"} in the embodied domain. There is a naive observation: \textit{the diversity of embodied data far exceeds the relatively small space of possible primitive motions}. Based on this insight, we propose \textbf{Primitive Embodied World Models} (PEWM), which restricts video generation to fixed shorter horizons, our approach \textit{1) enables} fine-grained alignment between linguistic concepts and visual representations of robotic actions, \textit{2) reduces} learning complexity, \textit{3) improves} data efficiency in embodied data collection, and \textit{4) decreases} inference latency. By equipping with a modular Vision-Language Model (VLM) planner and a Start-Goal heatmap Guidance mechanism (SGG), PEWM further enables flexible closed-loop control and supports compositional generalization of primitive-level policies over extended, complex tasks. Our framework leverages the spatiotemporal vision priors in video models and the semantic awareness of VLMs to bridge the gap between fine-grained physical interaction and high-level reasoning, paving the way toward scalable, interpretable, and general-purpose embodied intelligence.

Paperid: 6947, https://arxiv.org/pdf/2511.18116.pdf

Abstract:
Zero-Shot Anomaly Detection (ZSAD) aims to identify and localize anomalous regions in images of unseen object classes. While recent methods based on vision-language models like CLIP show promise, their performance is constrained by existing prompt engineering strategies. Current approaches, whether relying on single fixed, learnable, or dense dynamic prompts, suffer from a representational bottleneck and are prone to overfitting on auxiliary data, failing to generalize to the complexity and diversity of unseen anomalies. To overcome these limitations, we propose $\mathtt{PromptMoE}$. Our core insight is that robust ZSAD requires a compositional approach to prompt learning. Instead of learning monolithic prompts, $\mathtt{PromptMoE}$ learns a pool of expert prompts, which serve as a basis set of composable semantic primitives, and a visually-guided Mixture-of-Experts (MoE) mechanism to dynamically combine them for each instance. Our framework materializes this concept through a Visually-Guided Mixture of Prompt (VGMoP) that employs an image-gated sparse MoE to aggregate diverse normal and abnormal expert state prompts, generating semantically rich textual representations with strong generalization. Extensive experiments across 15 datasets in industrial and medical domains demonstrate the effectiveness and state-of-the-art performance of $\mathtt{PromptMoE}$.

Paperid: 6948, https://arxiv.org/pdf/2511.16527.pdf

Abstract:
Contrastive vision-language models continue to be the dominant approach for image and text retrieval. Contrastive Language-Image Pre-training (CLIP) trains two neural networks in contrastive manner to align their image and text embeddings in a shared latent space. Recent results evaluating CLIP on negated or paraphrased text have shown mixed performance because negation changes meaning radically with minimal lexical changes, while paraphrasing can create very different textual expressions with the same intended meaning. This poses a significant challenge for improving the evaluation results and alignment of vision-language models. To address this challenge, this paper evaluates the combination of paraphrasing and negation, proposes a new CLIP contrastive loss function accounting for both paraphrasing and negation, and applies LLM-generated training triples consisting of original, paraphrased and negated textual captions to CLIP-like training models. The approach, called SemCLIP, is shown to move paraphrased captions towards the original image embeddings while pushing negated captions further away in embedding space. Empirically, SemCLIP is shown to be capable of preserving CLIP's performance while increasing considerably the distances to negated captions. On the CC-Neg benchmark using an original over negation image-retrieval accuracy metric, SemCLIP improves accuracy from 68.1% to 78.1%. Although results are mixed when compared with CLIP on the Sugarcrepe++ benchmark, SemCLIP's performance is generally better than the models trained with negated captions. This robustness to negation extends to downstream zero-shot classification tasks where SemCLIP pre-trained on Sugarcrepe++ performs better than CLIP on all tested downstream tasks. These results indicate that SemCLIP can achieve significant robustness to semantic transformations.

Paperid: 6949, https://arxiv.org/pdf/2511.16134.pdf

Abstract:
Table Extraction (TE) consists in extracting tables from PDF documents, in a structured format which can be automatically processed. While numerous TE tools exist, the variety of methods and techniques makes it difficult for users to choose an appropriate one. We propose a novel benchmark for assessing end-to-end TE methods (from PDF to the final table). We contribute an analysis of TE evaluation metrics, and the design of a rigorous evaluation process, which allows scoring each TE sub-task as well as end-to-end TE, and captures model uncertainty. Along with a prior dataset, our benchmark comprises two new heterogeneous datasets of 37k samples. We run our benchmark on diverse models, including off-the-shelf libraries, software tools, large vision language models, and approaches based on computer vision. The results demonstrate that TE remains challenging: current methods suffer from a lack of generalizability when facing heterogeneous data, and from limitations in robustness and interpretability.

Paperid: 6950, https://arxiv.org/pdf/2511.15183.pdf

Abstract:
With nearly 1.5 billion people and more than 120 major languages, India represents one of the most diverse regions in the world. As multilingual Vision-Language Models (VLMs) gain prominence, robust evaluation methodologies are essential to drive progress toward equitable AI for low-resource languages. Current multilingual VLM evaluations suffer from four major limitations: reliance on unverified auto-translations, narrow task/domain coverage, limited sample sizes, and lack of cultural and natively sourced Question-Answering (QA). To address these gaps, we present a scalable framework to evaluate VLMs in Indian languages and compare it with performance in English. Using the framework, we generate HinTel-AlignBench, a benchmark that draws from diverse sources in Hindi and Telugu with English-aligned samples. Our contributions are threefold: (1) a semi-automated dataset creation framework combining back-translation, filtering, and human verification; (2) the most comprehensive vision-language benchmark for Hindi and and Telugu, including adapted English datasets (VQAv2, RealWorldQA, CLEVR-Math) and native novel Indic datasets (JEE for STEM, VAANI for cultural grounding) with approximately 4,000 QA pairs per language; and (3) a detailed performance analysis of various State-of-the-Art (SOTA) open-weight and closed-source VLMs. We find a regression in performance for tasks in English versus in Indian languages for 4 out of 5 tasks across all the models, with an average regression of 8.3 points in Hindi and 5.5 points for Telugu. We categorize common failure modes to highlight concrete areas of improvement in multilingual multimodal understanding.

Paperid: 6951, https://arxiv.org/pdf/2511.14499.pdf

Abstract:
The autonomous driving (AD) system has exhibited remarkable performance in complex driving scenarios. However, generalization is still a key limitation for the current system, which refers to the ability to handle unseen scenarios or unfamiliar sensor configurations.Related works have explored the use of Vision-Language Models (VLMs) to address few-shot or zero-shot tasks. While promising, these methods introduce a new challenge: the emergence of a hybrid AD system, where two distinct systems are used to plan a trajectory, leading to potential inconsistencies. Alternative research directions have explored Vision-Language-Action (VLA) frameworks that generate control actions from VLM directly. However, these end-to-end solutions demonstrate prohibitive computational demands. To overcome these challenges, we introduce Risk Semantic Distillation (RSD), a novel framework that leverages VLMs to enhance the training of End-to-End (E2E) AD backbones. By providing risk attention for key objects, RSD addresses the issue of generalization. Specifically, we introduce RiskHead, a plug-in module that distills causal risk estimates from Vision-Language Models into Bird's-Eye-View (BEV) features, yielding interpretable risk-attention maps.This approach allows BEV features to learn richer and more nuanced risk attention representations, which directly enhance the model's ability to handle spatial boundaries and risky objects.By focusing on risk attention, RSD aligns better with human-like driving behavior, which is essential to navigate in complex and dynamic environments. Our experiments on the Bench2Drive benchmark demonstrate the effectiveness of RSD in managing complex and unpredictable driving conditions. Due to the enhanced BEV representations enabled by RSD, we observed a significant improvement in both perception and planning capabilities.

Paperid: 6952, https://arxiv.org/pdf/2511.14210.pdf

Abstract:
We introduce Orion, a visual agent that integrates vision-based reasoning with tool-augmented execution to achieve powerful, precise, multi-step visual intelligence across images, video, and documents. Unlike traditional vision-language models that generate descriptive outputs, Orion orchestrates a suite of specialized computer vision tools, including object detection, keypoint localization, panoptic segmentation, Optical Character Recognition (OCR), and geometric analysis, to execute complex multi-step visual workflows. The system achieves competitive performance across MMMU, MMBench, DocVQA, and MMLongBench while extending monolithic VLM capabilities to production-grade visual intelligence. Through its agentic, tool-augmented approach, Orion enables autonomous visual reasoning that bridges neural perception with symbolic execution, marking the transition from passive visual understanding to active, tool-driven visual intelligence. Try Orion for free at: https://chat.vlm.run Learn more at: https://www.vlm.run/orion

Paperid: 6953, https://arxiv.org/pdf/2511.11169.pdf

Abstract:
In the context of Visual Question Answering (VQA) and Agentic AI, calibration refers to how closely an AI system's confidence in its answers reflects their actual correctness. This aspect becomes especially important when such systems operate autonomously and must make decisions under visual uncertainty. While modern VQA systems, powered by advanced vision-language models (VLMs), are increasingly used in high-stakes domains like medical diagnostics and autonomous navigation due to their improved accuracy, the reliability of their confidence estimates remains under-examined. Particularly, these systems often produce overconfident responses. To address this, we introduce AlignVQA, a debate-based multi-agent framework, in which diverse specialized VLM -- each following distinct prompting strategies -- generate candidate answers and then engage in two-stage interaction: generalist agents critique, refine and aggregate these proposals. This debate process yields confidence estimates that more accurately reflect the model's true predictive performance. We find that more calibrated specialized agents produce better aligned confidences. Furthermore, we introduce a novel differentiable calibration-aware loss function called aligncal designed to fine-tune the specialized agents by minimizing an upper bound on the calibration error. This objective explicitly improves the fidelity of each agent's confidence estimates. Empirical results across multiple benchmark VQA datasets substantiate the efficacy of our approach, demonstrating substantial reductions in calibration discrepancies. Furthermore, we propose a novel differentiable calibration-aware loss to fine-tune the specialized agents and improve the quality of their individual confidence estimates based on minimising upper bound calibration error.

Paperid: 6954, https://arxiv.org/pdf/2511.10774.pdf

Abstract:
The booming remote sensing (RS) technology is giving rise to a novel multimodality generalization task, which requires the model to overcome data heterogeneity while possessing powerful cross-scene generalization ability. Moreover, most vision-language models (VLMs) usually describe surface materials in RS images using universal texts, lacking proprietary linguistic prior knowledge specific to different RS vision modalities. In this work, we formalize RS multimodality generalization (RSMG) as a learning paradigm, and propose a frequency-aware vision-language multimodality generalization network (FVMGN) for RS image classification. Specifically, a diffusion-based training-test-time augmentation (DTAug) strategy is designed to reconstruct multimodal land-cover distributions, enriching input information for FVMGN. Following that, to overcome multimodal heterogeneity, a multimodal wavelet disentanglement (MWDis) module is developed to learn cross-domain invariant features by resampling low and high frequency components in the frequency domain. Considering the characteristics of RS vision modalities, shared and proprietary class texts is designed as linguistic inputs for the transformer-based text encoder to extract diverse text features. For multimodal vision inputs, a spatial-frequency-aware image encoder (SFIE) is constructed to realize local-global feature reconstruction and representation. Finally, a multiscale spatial-frequency feature alignment (MSFFA) module is suggested to construct a unified semantic space, ensuring refined multiscale alignment of different text and vision features in spatial and frequency domains. Extensive experiments show that FVMGN has the excellent multimodality generalization ability compared with state-of-the-art (SOTA) methods.

Paperid: 6955, https://arxiv.org/pdf/2511.10301.pdf

Abstract:
Despite the remarkable success of the LLaVA architecture for vision-language tasks, its design inherently struggles to effectively integrate visual features due to the inherent mismatch between text and vision modalities. We tackle this issue from a novel perspective in which the LLM not only serves as a language model but also a powerful vision encoder. To this end, we present LLaViT - Large Language Models as extended Vision Transformers - which enables the LLM to simultaneously function as a vision encoder through three key modifications: (1) learning separate QKV projections for vision modality, (2) enabling bidirectional attention on visual tokens, and (3) incorporating both global and local visual representations. Through extensive controlled experiments on a wide range of LLMs, we demonstrate that LLaViT significantly outperforms the baseline LLaVA method on a multitude of benchmarks, even surpassing models with double its parameter count, establishing a more effective approach to vision-language modeling.

Paperid: 6956, https://arxiv.org/pdf/2511.09919.pdf

Abstract:
Despite the rapid progress of Vision-Language Models (VLMs), their capabilities are inadequately assessed by existing benchmarks, which are predominantly English-centric, feature simplistic layouts, and support limited tasks. Consequently, they fail to evaluate model performance for Visually Rich Document Understanding (VRDU), a critical challenge involving complex layouts and dense text. To address this, we introduce DocWeaver, a novel multi-agent pipeline that leverages Large Language Models to automatically generate a new benchmark. The result is MosaicDoc, a large-scale, bilingual (Chinese and English) resource designed to push the boundaries of VRDU. Sourced from newspapers and magazines, MosaicDoc features diverse and complex layouts (including multi-column and non-Manhattan), rich stylistic variety from 196 publishers, and comprehensive multi-task annotations (OCR, VQA, reading order, and localization). With 72K images and over 600K QA pairs, MosaicDoc serves as a definitive benchmark for the field. Our extensive evaluation of state-of-the-art models on this benchmark reveals their current limitations in handling real-world document complexity and charts a clear path for future research.

Paperid: 6957, https://arxiv.org/pdf/2511.08172.pdf

Abstract:
Visual grounding is the task of localising image regions from natural language queries and is critical for reasoning capable Graphical User Interface agents. Many existing methods rely on massive, noisy synthetic datasets. This work introduces an efficient training pipeline that combines model-based data filtering with parameter-efficient fine-tuning. From 4.8M synthetic examples, 12K clean and diverse instances are curated by first identifying challenging cases, removing misaligned and then selecting a diverse set of multimodal instances. On this data, a 3B-parameter Vision-Language Model is trained under three regimes: supervised fine-tuning, chain-of-thought-augmented fine-tuning, and reinforcement learning via Group Relative Policy Optimization. Models trained with the filtered data and lightweight training strategies match or surpass larger baselines on benchmarks such as ScreenSpot, Multimodal-Mind2Web, and AndroidControl. These results demonstrate that principled data curation and robust adaptation can rival large-scale training, enabling compact yet capable multimodal reasoning agents.

Paperid: 6958, https://arxiv.org/pdf/2511.07238.pdf

Abstract:
In autonomous driving and robotics, ensuring road safety and reliable decision-making critically depends on out-of-distribution (OOD) segmentation. While numerous methods have been proposed to detect anomalous objects on the road, leveraging the vision-language space-which provides rich linguistic knowledge-remains an underexplored field. We hypothesize that incorporating these linguistic cues can be especially beneficial in the complex contexts found in real-world autonomous driving scenarios. To this end, we present a novel approach that trains a Text-Driven OOD Segmentation model to learn a semantically diverse set of objects in the vision-language space. Concretely, our approach combines a vision-language model's encoder with a transformer decoder, employs Distance-Based OOD prompts located at varying semantic distances from in-distribution (ID) classes, and utilizes OOD Semantic Augmentation for OOD representations. By aligning visual and textual information, our approach effectively generalizes to unseen objects and provides robust OOD segmentation in diverse driving environments. We conduct extensive experiments on publicly available OOD segmentation datasets such as Fishyscapes, Segment-Me-If-You-Can, and Road Anomaly datasets, demonstrating that our approach achieves state-of-the-art performance across both pixel-level and object-level evaluations. This result underscores the potential of vision-language-based OOD segmentation to bolster the safety and reliability of future autonomous driving systems.

Paperid: 6959, https://arxiv.org/pdf/2511.06678.pdf

Abstract:
Concept bottleneck models (CBMs) improve neural network interpretability by introducing an intermediate layer that maps human-understandable concepts to predictions. Recent work has explored the use of vision-language models (VLMs) to automate concept selection and annotation. However, existing VLM-based CBMs typically require full model retraining when new concepts are involved, which limits their adaptability and flexibility in real-world scenarios, especially considering the rapid evolution of vision-language foundation models. To address these issues, we propose Flexible Concept Bottleneck Model (FCBM), which supports dynamic concept adaptation, including complete replacement of the original concept set. Specifically, we design a hypernetwork that generates prediction weights based on concept embeddings, allowing seamless integration of new concepts without retraining the entire model. In addition, we introduce a modified sparsemax module with a learnable temperature parameter that dynamically selects the most relevant concepts, enabling the model to focus on the most informative features. Extensive experiments on five public benchmarks demonstrate that our method achieves accuracy comparable to state-of-the-art baselines with a similar number of effective concepts. Moreover, the model generalizes well to unseen concepts with just a single epoch of fine-tuning, demonstrating its strong adaptability and flexibility.

Paperid: 6960, https://arxiv.org/pdf/2511.06651.pdf

Abstract:
In this study, we propose NOVO (NO text, Visual-Only prompts), a novel framework that bridges vision-language models (VLMs) and segmentation models through visual-only prompts. Unlike prior approaches that feed text-derived SEG token embeddings into segmentation models, NOVO instead generates a coarse mask and point prompts from the VLM output. These visual prompts are compatible with the Segment Anything Model (SAM), preserving alignment with its pretrained capabilities. To further enhance boundary quality and enable instance-level segmentation, we introduce a training-free refinement module that reduces visual artifacts and improves the quality of segmentation masks. We also present RISeg, a new benchmark comprising 918 images, 2,533 instance-level masks, and diverse reasoning queries to evaluate this task. Experiments demonstrate that NOVO achieves state-of-the-art performance across multiple metrics and model sizes, demonstrating its effectiveness and scalability in reasoning segmentation.

Paperid: 6961, https://arxiv.org/pdf/2511.06490.pdf

Abstract:
Complex visual narratives, such as comics, present a significant challenge to Vision-Language Models (VLMs). Despite excelling on natural images, VLMs often struggle with stylized line art, onomatopoeia, and densely packed multi-panel layouts. To address this gap, we introduce AI4VA-FG, the first fine-grained and comprehensive benchmark for VLM-based comic understanding. It spans tasks from foundational recognition and detection to high-level character reasoning and narrative construction, supported by dense annotations for characters, poses, and depth. Beyond that, we evaluate state-of-the-art proprietary models, including GPT-4o and Gemini-2.5, and open-source models such as Qwen2.5-VL, revealing substantial performance deficits across core tasks of our benchmarks and underscoring that comic understanding remains an unsolved challenge. To enhance VLMs' capabilities in this domain, we systematically investigate post-training strategies, including supervised fine-tuning on solutions (SFT-S), supervised fine-tuning on reasoning trajectories (SFT-R), and reinforcement learning (RL). Beyond that, inspired by the emerging "Thinking with Images" paradigm, we propose Region-Aware Reinforcement Learning (RARL) for VLMs, which trains models to dynamically attend to relevant regions through zoom-in operations. We observe that when applied to the Qwen2.5-VL model, RL and RARL yield significant gains in low-level entity recognition and high-level storyline ordering, paving the way for more accurate and efficient VLM applications in the comics domain.

Paperid: 6962, https://arxiv.org/pdf/2511.06202.pdf

Abstract:
Vision-Language-Action models such as OpenVLA show impressive zero-shot generalization across robotic manipulation tasks but often fail to adapt efficiently to new deployment environments. In many real-world applications, consistent high performance on a limited set of tasks is more important than broad generalization. We propose ExpReS-VLA, a method for specializing pre-trained VLA models through experience replay and retrieval while preventing catastrophic forgetting. ExpReS-VLA stores compact feature representations from the frozen vision backbone instead of raw image-action pairs, reducing memory usage by approximately 97 percent. During deployment, relevant past experiences are retrieved using cosine similarity and used to guide adaptation, while prioritized experience replay emphasizes successful trajectories. We also introduce Thresholded Hybrid Contrastive Loss, which enables learning from both successful and failed attempts. On the LIBERO simulation benchmark, ExpReS-VLA improves success rates from 82.6 to 93.1 percent on spatial reasoning tasks and from 61 to 72.3 percent on long-horizon tasks. On physical robot experiments with five manipulation tasks, it reaches 98 percent success on both seen and unseen settings, compared to 84.7 and 32 percent for naive fine-tuning. Adaptation takes 31 seconds using 12 demonstrations on a single RTX 5090 GPU, making the approach practical for real robot deployment.

Paperid: 6963, https://arxiv.org/pdf/2511.06201.pdf

Abstract:
This paper introduces a human-in-the-loop computer vision framework that uses generative AI to propose micro-scale design interventions in public space and support more continuous, local participation. Using Grounding DINO and a curated subset of the ADE20K dataset as a proxy for the urban built environment, the system detects urban objects and builds co-occurrence embeddings that reveal common spatial configurations. From this analysis, the user receives five statistically likely complements to a chosen anchor object. A vision language model then reasons over the scene image and the selected pair to suggest a third object that completes a more complex urban tactic. The workflow keeps people in control of selection and refinement and aims to move beyond top-down master planning by grounding choices in everyday patterns and lived experience.

Paperid: 6964, https://arxiv.org/pdf/2511.06195.pdf

Abstract:
Interfaces for contemporary large language, generative media, and perception AI models are often engineered for single user interaction. We investigate ritual as a design scaffold for developing collaborative, multi-user human-AI engagement. We consider the specific case of an immersive staging of the musical Xanadu performed at UCLA in Spring 2025. During a two-week run, over five hundred audience members contributed sketches and jazzercise moves that vision language models translated to virtual scenery elements and from choreographic prompts. This paper discusses four facets of interaction-as-ritual within the show: audience input as offerings that AI transforms into components of the ritual; performers as ritual guides, demonstrating how to interact with technology and sorting audience members into cohorts; AI systems as instruments "played" by the humans, in which sensing, generative components, and stagecraft create systems that can be mastered over time; and reciprocity of interaction, in which the show's AI machinery guides human behavior as well as being guided by humans, completing a human-AI feedback loop that visibly reshapes the virtual world. Ritual served as a frame for integrating linear narrative, character identity, music and interaction. The production explored how AI systems can support group creativity and play, addressing a critical gap in prevailing single user AI design paradigms.

Paperid: 6965, https://arxiv.org/pdf/2511.05550.pdf

Abstract:
Music language models (Music LMs), like vision language models, leverage multimodal representations to answer natural language queries about musical audio recordings. Although Music LMs are reportedly improving, we find that current evaluations fail to capture whether their answers are correct. Specifically, for all Music LMs that we examine, widely-used evaluation metrics such as BLEU, METEOR, and BERTScore fail to measure anything beyond linguistic fluency of the model's responses. To measure the true performance of Music LMs, we propose (1) a better general-purpose evaluation metric for Music LMs adapted to the music domain and (2) a factual evaluation framework to quantify the correctness of a Music LM's responses. Our framework is agnostic to the modality of the question-answering model and could be generalized to quantify performance in other open-ended question-answering domains. We use open datasets in our experiments and will release all code on publication.

Paperid: 6966, https://arxiv.org/pdf/2511.05397.pdf

Abstract:
While Vision-Language-Action (VLA) models map visual inputs and language instructions directly to robot actions, they often rely on costly hardware and struggle in novel or cluttered scenes. We introduce EverydayVLA, a 6-DOF manipulator that can be assembled for under $300, capable of modest payloads and workspace. A single unified model jointly outputs discrete and continuous actions, and our adaptive-horizon ensemble monitors motion uncertainty to trigger on-the-fly re-planning for safe, reliable operation. On LIBERO, EverydayVLA matches state-of-the-art success rates, and in real-world tests it outperforms prior methods by 49% in-distribution and 34.9% out-of-distribution. By combining a state-of-the-art VLA with cost-effective hardware, EverydayVLA democratizes access to a robotic foundation model and paves the way for economical use in homes and research labs alike. Experiment videos and details: https://everydayvla.github.io/

Paperid: 6967, https://arxiv.org/pdf/2511.01817.pdf

Abstract:
The ability to connect visual patterns with the processes that form them represents one of the deepest forms of visual understanding. Textures of clouds and waves, the growth of cities and forests, or the formation of materials and landscapes are all examples of patterns emerging from underlying mechanisms. We present the Scitextures dataset, a large-scale collection of textures and visual patterns from all domains of science, tech, and art, along with the models and code that generate these images. Covering over 1,200 different models and 100,000 images of patterns and textures from physics, chemistry, biology, sociology, technology, mathematics, and art, this dataset offers a way to explore the connection between the visual patterns that shape our world and the mechanisms that produce them. Created by an agentic AI pipeline that autonomously collects and implements models in standardized form, we use SciTextures to evaluate the ability of leading AI models to link visual patterns to the models and code that generate them, and to identify different patterns that emerged from the same process. We also test AIs ability to infer and recreate the mechanisms behind visual patterns by providing a natural image of a real-world pattern and asking the AI to identify, model, and code the mechanism that formed the pattern, then run this code to generate a simulated image that is compared to the real image. These benchmarks show that vision-language models (VLMs) can understand and simulate the physical system beyond a visual pattern. The dataset and code are available at: https://zenodo.org/records/17485502

Paperid: 6968, https://arxiv.org/pdf/2511.01340.pdf

Abstract:
Understanding Rebus Puzzles (Rebus Puzzles use pictures, symbols, and letters to represent words or phrases creatively) requires a variety of skills such as image recognition, cognitive skills, commonsense reasoning, multi-step reasoning, image-based wordplay, etc., making this a challenging task for even current Vision-Language Models. In this paper, we present $\left|\,\circlearrowright\,\boxed{\text{BUS}}\,\right|$, a large and diverse benchmark of $1,333$ English Rebus Puzzles containing different artistic styles and levels of difficulty, spread across 18 categories such as food, idioms, sports, finance, entertainment, etc. We also propose $RebusDescProgICE$, a model-agnostic framework which uses a combination of an unstructured description and code-based, structured reasoning, along with better, reasoning-based in-context example selection, improving the performance of Vision-Language Models on $\left|\,\circlearrowright\,\boxed{\text{BUS}}\,\right|$ by $2.1-4.1\%$ and $20-30\%$ using closed-source and open-source models respectively compared to Chain-of-Thought Reasoning.

Paperid: 6969, https://arxiv.org/pdf/2511.00523.pdf

Abstract:
Vision language models such as CLIP have shown remarkable performance in zero shot classification, but remain susceptible to spurious correlations, where irrelevant visual features influence predictions. Existing debiasing methods often require access to training data and explicit group labels to perform fine-tuning or adjust embeddings, which limits their practicality in real-world settings. Test-time methods attempt to avoid this constraint, but many still depend on prior knowledge of dataset specific biases, limiting their generalizability in open set settings. In this work, we propose a test-time debiasing method for ViT based CLIP models that requires no additional training or assumptions of bias annotations. Our approach uses a pretrained segmentation model to isolate the target visual attribute, then adjusts the non target regions so that their embeddings are uniformly similar to all class specific text prompts. This procedure removes unintended bias signals from confounding visual regions while preserving the target attribute. Experiments on Waterbirds and CelebA show that our method outperforms existing test-time debiasing approaches in both group robustness metrics and Attention IoU. These results demonstrate the effectiveness of segmentation guided interventions for scalable and annotation free bias mitigation in vision language models.

Paperid: 6970, https://arxiv.org/pdf/2511.00419.pdf

Abstract:
Recent advancements in large-scale pretraining in natural language processing have enabled pretrained vision-language models such as CLIP to effectively align images and text, significantly improving performance in zero-shot image classification tasks. Subsequent studies have further demonstrated that cropping images into smaller regions and using large language models to generate multiple descriptions for each caption can further enhance model performance. However, due to the inherent sensitivity of CLIP, random image crops can introduce misinformation and bias, as many images share similar features at small scales. To address this issue, we propose Localized-Globalized Cross-Alignment (LGCA), a framework that first captures the local features of an image and then repeatedly selects the most salient regions and expands them. The similarity score is designed to incorporate both the original and expanded images, enabling the model to capture both local and global features while minimizing misinformation. Additionally, we provide a theoretical analysis demonstrating that the time complexity of LGCA remains the same as that of the original model prior to the repeated expansion process, highlighting its efficiency and scalability. Extensive experiments demonstrate that our method substantially improves zero-shot performance across diverse datasets, outperforming state-of-the-art baselines.

Paperid: 6971, https://arxiv.org/pdf/2511.00054.pdf

Abstract:
While Vision-Language Models (VLMs) excel in many areas, they struggle with complex spatial reasoning, which requires problem decomposition and strategic tool use. Fine-tuning smaller, more deployable models offers an efficient path to strong performance, but this is hampered by a major bottleneck: the absence of high-quality, step-by-step reasoning data. To address this data-efficiency gap, we introduce SpatialTraceGen, a framework to distill the reasoning processes of a large teacher model into a high-quality dataset of multi-hop, multi-tool reasoning traces. A key innovation is our automated Verifier, which scalably ensures the fidelity of each reasoning step, providing a cost-effective alternative to manual human annotation. On the CLEVR-Humans benchmark, this verifier-guided process improves the average quality score of traces by 17\% while reducing quality variance by over 40\%. SpatialTraceGen delivers a dataset of expert traces, providing the structured, step-by-step examples of tool use necessary for effective fine-tuning and sample-efficient offline reinforcement learning.

Paperid: 6972, https://arxiv.org/pdf/2510.26466.pdf

Abstract:
Object-context shortcuts remain a persistent challenge in vision-language models, undermining zero-shot reliability when test-time scenes differ from familiar training co-occurrences. We recast this issue as a causal inference problem and ask: Would the prediction remain if the object appeared in a different environment? To answer this at inference time, we estimate object and background expectations within CLIP's representation space, and synthesize counterfactual embeddings by recombining object features with diverse alternative contexts sampled from external datasets, batch neighbors, or text-derived descriptions. By estimating the Total Direct Effect and simulating intervention, we further subtract background-only activation, preserving beneficial object-context interactions while mitigating hallucinated scores. Without retraining or prompt design, our method substantially improves both worst-group and average accuracy on context-sensitive benchmarks, establishing a new zero-shot state of the art. Beyond performance, our framework provides a lightweight representation-level counterfactual approach, offering a practical causal avenue for debiased and reliable multimodal reasoning.

Paperid: 6973, https://arxiv.org/pdf/2510.26052.pdf

Abstract:
We propose a novel approach for dynamic negative prompting in diffusion models that leverages Vision-Language Models (VLMs) to adaptively generate negative prompts during the denoising process. Unlike traditional Negative Prompting methods that use fixed negative prompts, our method generates intermediate image predictions at specific denoising steps and queries a VLM to produce contextually appropriate negative prompts. We evaluate our approach on various benchmark datasets and demonstrate the trade-offs between negative guidance strength and text-image alignment.

Paperid: 6974, https://arxiv.org/pdf/2510.25798.pdf

Abstract:
The dynamic nature of information necessitates continuously updating large vision-language models (LVLMs). While recent knowledge editing techniques hint at promising directions, they often focus on editing a single modality (vision or language) in isolation. This prevalent practice neglects the inherent multimodality of LVLMs and the continuous nature of knowledge updates, potentially leading to suboptimal editing outcomes when considering the interplay between modalities and the need for ongoing knowledge refinement. To address these limitations, we propose MemEIC, a novel method for Continual and Compositional Knowledge Editing (CCKE) in LVLMs. MemEIC enables compositional editing of both visual and textual knowledge sequentially. Our approach employs a hybrid external-internal editor featuring a dual external memory for cross-modal evidence retrieval and dual LoRA adapters that facilitate disentangled parameter updates for each modality. A key component is a brain-inspired knowledge connector, activated selectively for compositional reasoning, that integrates information across different modalities. Experiments demonstrate that MemEIC significantly improves performance on complex multimodal questions and effectively preserves prior edits, setting a new benchmark for CCKE in LVLMs.

Paperid: 6975, https://arxiv.org/pdf/2508.20322.pdf

Abstract:
Vision-language co-embedding networks, such as CLIP, provide a latent embedding space with semantic information that is useful for downstream tasks. We hypothesize that the embedding space can be disentangled to separate the information on the content of complex scenes by decomposing the embedding into multiple concept-specific component vectors that lie in different subspaces. We propose a supervised dictionary learning approach to estimate a linear synthesis model consisting of sparse, non-negative combinations of groups of vectors in the dictionary (atoms), whose group-wise activity matches the multi-label information. Each concept-specific component is a non-negative combination of atoms associated to a label. The group-structured dictionary is optimized through a novel alternating optimization with guaranteed convergence. Exploiting the text co-embeddings, we detail how semantically meaningful descriptions can be found based on text embeddings of words best approximated by a concept's group of atoms, and unsupervised dictionary learning can exploit zero-shot classification of training set images using the text embeddings of concept labels to provide instance-wise multi-labels. We show that the disentangled embeddings provided by our sparse linear concept subspaces (SLiCS) enable concept-filtered image retrieval (and conditional generation using image-to-prompt) that is more precise. We also apply SLiCS to highly-compressed autoencoder embeddings from TiTok and the latent embedding from self-supervised DINOv2. Quantitative and qualitative results highlight the improved precision of the concept-filtered image retrieval for all embeddings.

Paperid: 6976, https://arxiv.org/pdf/2508.18050.pdf

Abstract:
Camouflaged Object Segmentation (COS) poses a significant challenge due to the intrinsic high similarity between targets and backgrounds, demanding models capable of profound holistic understanding beyond superficial cues. Prevailing methods, often limited by shallow feature representation, inadequate reasoning mechanisms, and weak cross-modal integration, struggle to achieve this depth of cognition, resulting in prevalent issues like incomplete target separation and imprecise segmentation. Inspired by the perceptual strategy of the Hundred-eyed Giant-emphasizing holistic observation, omnidirectional focus, and intensive scrutiny-we introduce ArgusCogito, a novel zero-shot, chain-of-thought framework underpinned by cross-modal synergy and omnidirectional reasoning within Vision-Language Models (VLMs). ArgusCogito orchestrates three cognitively-inspired stages: (1) Conjecture: Constructs a strong cognitive prior through global reasoning with cross-modal fusion (RGB, depth, semantic maps), enabling holistic scene understanding and enhanced target-background disambiguation. (2) Focus: Performs omnidirectional, attention-driven scanning and focused reasoning, guided by semantic priors from Conjecture, enabling precise target localization and region-of-interest refinement. (3) Sculpting: Progressively sculpts high-fidelity segmentation masks by integrating cross-modal information and iteratively generating dense positive/negative point prompts within focused regions, emulating Argus' intensive scrutiny. Extensive evaluations on four challenging COS benchmarks and three Medical Image Segmentation (MIS) benchmarks demonstrate that ArgusCogito achieves state-of-the-art (SOTA) performance, validating the framework's exceptional efficacy, superior generalization capability, and robustness.

Paperid: 6977, https://arxiv.org/pdf/2508.17442.pdf

Abstract:
Action recognition and localization in complex, untrimmed videos remain a formidable challenge in computer vision, largely due to the limitations of existing methods in capturing fine-grained actions, long-term temporal dependencies, and high-level semantic information from low-level visual features. This paper introduces the Event-Contextualized Video Transformer (ECVT), a novel architecture that leverages the advanced semantic understanding capabilities of Large Vision-Language Models (LVLMs) to bridge this gap. ECVT employs a dual-branch design, comprising a Video Encoding Branch for spatio-temporal feature extraction and a Cross-Modal Guidance Branch. The latter utilizes an LVLM to generate multi-granularity semantic descriptions, including Global Event Prompting for macro-level narrative and Temporal Sub-event Prompting for fine-grained action details. These multi-level textual cues are integrated into the video encoder's learning process through sophisticated mechanisms such as adaptive gating for high-level semantic fusion, cross-modal attention for fine-grained feature refinement, and an event graph module for temporal context calibration. Trained end-to-end with a comprehensive loss function incorporating semantic consistency and temporal calibration terms, ECVT significantly enhances the model's ability to understand video temporal structures and event logic. Extensive experiments on ActivityNet v1.3 and THUMOS14 datasets demonstrate that ECVT achieves state-of-the-art performance, with an average mAP of 40.5% on ActivityNet v1.3 and mAP@0.5 of 67.1% on THUMOS14, outperforming leading baselines.

Paperid: 6978, https://arxiv.org/pdf/2508.17435.pdf

Abstract:
Despite the remarkable capabilities of text-to-image (T2I) generation models, real-world applications often demand fine-grained, iterative image editing that existing methods struggle to provide. Key challenges include granular instruction understanding, robust context preservation during modifications, and the lack of intelligent feedback mechanisms for iterative refinement. This paper introduces RefineEdit-Agent, a novel, training-free intelligent agent framework designed to address these limitations by enabling complex, iterative, and context-aware image editing. RefineEdit-Agent leverages the powerful planning capabilities of Large Language Models (LLMs) and the advanced visual understanding and evaluation prowess of Vision-Language Large Models (LVLMs) within a closed-loop system. Our framework comprises an LVLM-driven instruction parser and scene understanding module, a multi-level LLM-driven editing planner for goal decomposition, tool selection, and sequence generation, an iterative image editing module, and a crucial LVLM-driven feedback and evaluation loop. To rigorously evaluate RefineEdit-Agent, we propose LongBench-T2I-Edit, a new benchmark featuring 500 initial images with complex, multi-turn editing instructions across nine visual dimensions. Extensive experiments demonstrate that RefineEdit-Agent significantly outperforms state-of-the-art baselines, achieving an average score of 3.67 on LongBench-T2I-Edit, compared to 2.29 for Direct Re-Prompting, 2.91 for InstructPix2Pix, 3.16 for GLIGEN-based Edit, and 3.39 for ControlNet-XL. Ablation studies, human evaluations, and analyses of iterative refinement, backbone choices, tool usage, and robustness to instruction complexity further validate the efficacy of our agentic design in delivering superior edit fidelity and context preservation.

Paperid: 6979, https://arxiv.org/pdf/2508.16987.pdf

Abstract:
We introduce WebSight, a vision-based autonomous web agent, designed to interact with web environments purely through visual perception, eliminating dependence on HTML or DOM-based inputs. Central to our approach we introduce our new model, WebSight-7B, a fine-tuned vision-language model optimized for UI element interaction, trained using LoRA on a web-focused subset of the Wave-UI-25K dataset. WebSight integrates this model into a modular multi-agent architecture, comprising planning, reasoning, vision-action, and verification agents, coordinated through an episodic memory mechanism. WebSight-7B achieves a top-1 accuracy of 58.84% on the Showdown Clicks benchmark, outperforming several larger generalist models while maintaining lower latency. The full WebSight agent achieves a 68.0% success rate on the WebVoyager benchmark, surpassing systems from labs such as OpenAI (61.0%) and HCompany (Runner H, 67.0%). Among tasks completed, WebSight answers correctly 97.14% of the time, indicating high precision. Together, WebSight and WebSight-7B establish a new standard for interpretable, robust, and efficient visual web navigation.

Paperid: 6980, https://arxiv.org/pdf/2508.16974.pdf

Abstract:
Large Language Models (LLMs) and Vision-Language Large Models (LVLMs) have achieved remarkable progress in natural language processing and multimodal understanding. Despite their impressive generalization capabilities, current LVLMs often exhibit insufficient robustness, proneness to hallucination, and reasoning errors in complex real-world scenarios, particularly when precise image region localization and fine-grained visual reasoning are required. To address these limitations, we propose the Hierarchical Contextual Grounding LVLM (HCG-LVLM), a novel architecture that mimics human coarse-to-fine cognitive processing. HCG-LVLM employs a two-layered approach: a Global Contextual Perception layer for initial broad understanding and a Fine-grained Local Grounding layer. The latter incorporates a Local Detail Enhancement Module to extract high-resolution features and a Semantic Consistency Validator to ensure accurate, hallucination-free visual-language alignment. Through an adaptive fusion mechanism, information from both layers is integrated for robust and precise outputs. Extensive experiments on challenging datasets, including GQA, A-OKVQA for fine-grained VQA, and RefCOCO/+/g for Referring Expression Comprehension, demonstrate that HCG-LVLM consistently outperforms state-of-the-art models such as Flamingo, BLIP-2, and MiniGPT-4. Our model achieves superior accuracy and significantly reduces hallucination, validating the effectiveness of its hierarchical design in enhancing fine-grained visual-language understanding and precise grounding capabilities.

Paperid: 6981, https://arxiv.org/pdf/2508.15903.pdf

Abstract:
Human action recognition in long-term videos, characterized by complex backgrounds and subtle action differences, poses significant challenges for traditional deep learning models due to computational overhead, difficulty in capturing long-range temporal dependencies, and limited semantic understanding. While Large Language Models (LLMs) and Large Vision-Language Models (LVLMs) have shown remarkable capabilities in multi-modal understanding and reasoning, their direct application to continuous video streams for fine-grained action recognition remains an open problem. This paper introduces VT-LVLM-AR (Video-Temporal Large Vision-Language Model Adapter for Action Recognition), a novel framework designed to bridge this gap. VT-LVLM-AR comprises a Video-to-Event Mapper (VTEM) that efficiently transforms raw video into compact, semantically rich, and temporally coherent "visual event sequences" through lightweight spatio-temporal feature extraction, adaptive temporal pooling, and conceptual quantization with an event coherence bias. These visual event sequences are then fed into an LVLM-based Action Reasoning module, specifically a frozen LLaVA-1.5 model, adapted using parameter-efficient Prompt Tuning (P-Tuning v2) for action classification. Comprehensive evaluations on the NTU RGB+D and NTU RGB+D 120 datasets demonstrate that VT-LVLM-AR consistently achieves state-of-the-art performance, surpassing existing methods (e.g., 94.1% accuracy on NTU RGB+D X-Sub). Ablation studies confirm the critical contributions of VTEM's components and the efficacy of Prompt Tuning, while human evaluations underscore the interpretability of our visual event representations. This work highlights the immense potential of leveraging LVLMs for robust and interpretable video action understanding through effective video-to-language translation and efficient model adaptation.

Paperid: 6982, https://arxiv.org/pdf/2508.15164.pdf

Abstract:
Despite significant advancements in Large Language Models (LLMs) and Large Vision-Language Models (LVLMs), current models still face substantial challenges in handling complex, multi-turn, and visually-grounded tasks that demand deep reasoning, sustained contextual understanding, entity tracking, and multi-step instruction following. Existing benchmarks often fall short in capturing the dynamism and intricacies of real-world multi-modal interactions, leading to issues such as context loss and visual hallucinations. To address these limitations, we introduce MMDR-Bench (Multi-Modal Dialogue Reasoning Benchmark), a novel dataset comprising 300 meticulously designed complex multi-turn dialogue scenarios, each averaging 5-7 turns and evaluated across six core dimensions including visual entity tracking and reasoning depth. Furthermore, we propose CoLVLM Agent (Contextual LVLM Agent), a holistic framework that enhances existing LVLMs with advanced reasoning and instruction following capabilities through an iterative "memory-perception-planning-execution" cycle, requiring no extensive re-training of the underlying models. Our extensive experiments on MMDR-Bench demonstrate that CoLVLM Agent consistently achieves superior performance, attaining an average human evaluation score of 4.03, notably surpassing state-of-the-art commercial models like GPT-4o (3.92) and Gemini 1.5 Pro (3.85). The framework exhibits significant advantages in reasoning depth, instruction adherence, and error suppression, and maintains robust performance over extended dialogue turns, validating the effectiveness of its modular design and iterative approach for complex multi-modal interactions.

Paperid: 6983, https://arxiv.org/pdf/2508.13524.pdf

Abstract:
Facial Emotion Recognition (FER) is crucial for applications such as human-computer interaction and mental health diagnostics. This study presents the first empirical comparison of open-source Vision-Language Models (VLMs), including Phi-3.5 Vision and CLIP, against traditional deep learning models VGG19, ResNet-50, and EfficientNet-B0 on the challenging FER-2013 dataset, which contains 35,887 low-resolution grayscale images across seven emotion classes. To address the mismatch between VLM training assumptions and the noisy nature of FER data, we introduce a novel pipeline that integrates GFPGAN-based image restoration with FER evaluation. Results show that traditional models, particularly EfficientNet-B0 (86.44%) and ResNet-50 (85.72%), significantly outperform VLMs like CLIP (64.07%) and Phi-3.5 Vision (51.66%), highlighting the limitations of VLMs in low-quality visual tasks. In addition to performance evaluation using precision, recall, F1-score, and accuracy, we provide a detailed computational cost analysis covering preprocessing, training, inference, and evaluation phases, offering practical insights for deployment. This work underscores the need for adapting VLMs to noisy environments and provides a reproducible benchmark for future research in emotion recognition.

Paperid: 6984, https://arxiv.org/pdf/2508.13387.pdf

Abstract:
Recent advances in multimodal Parameter-Efficient Fine-Tuning (PEFT) have significantly improved performance on downstream tasks such as few-shot retrieval. However, most existing approaches focus on task-specific gains while neglecting the structure of the multimodal embedding space. As a result, modality-specific representations often remain isolated, limiting cross-modal generalisation. In this work, we introduce Shared Prompt AligNER (SPANER), a modality-agnostic PEFT framework designed to embed inputs from diverse modalities into a unified semantic space. At its core, SPANER employs a shared prompt mechanism that acts as a conceptual anchor, enabling semantically related instances to converge spatially regardless of modality. This shared prompt design is inherently extensible, supporting the seamless integration of additional modalities, such as audio, without altering the core architecture. Through comprehensive experiments across vision-language and audio-visual benchmarks, SPANER demonstrates competitive few-shot retrieval performance while preserving high semantic coherence in the learned embedding space. Our results highlight the importance of aligning embedding structures, rather than merely tuning adapter weights, for scalable multimodal learning.

Paperid: 6985, https://arxiv.org/pdf/2508.12458.pdf

Abstract:
Large Vision-Language Models (LVLMs) hold immense potential for complex multimodal instruction following, yet their development is often hindered by the high cost and inconsistency of human annotation required for effective fine-tuning and preference alignment. Traditional supervised fine-tuning (SFT) and existing preference optimization methods like RLHF and DPO frequently struggle to efficiently leverage the model's own generation space to identify highly informative "hard negative" samples. To address these challenges, we propose Multimodal-Model-Guided Preference Optimization (M3PO), a novel and data-efficient method designed to enhance LVLMs' capabilities in visual instruction following. M3PO intelligently selects the most "learning-valuable" preference sample pairs from a diverse pool of LVLM-generated candidates. This selection is driven by a sophisticated mechanism that integrates two crucial signals: a Multimodal Alignment Score (MAS) to assess external quality and the model's Self-Consistency / Confidence (log-probability) to gauge internal belief. These are combined into a novel M3P-Score, which specifically identifies preferred responses and challenging dispreferred responses that the model might confidently generate despite being incorrect. These high-quality preference pairs are then used for efficient Direct Preference Optimization (DPO) fine-tuning on base LVLMs like LLaVA-1.5 (7B/13B) using LoRA. Our extensive experiments demonstrate that M3PO consistently outperforms strong baselines, including SFT, simulated RLHF, vanilla DPO, and RM-DPO, across a comprehensive suite of multimodal instruction following benchmarks (MME-Bench, POPE, IFT, Human Pref. Score).

Paperid: 6986, https://arxiv.org/pdf/2508.12455.pdf

Abstract:
Chest X-ray imaging is crucial for diagnosing pulmonary and cardiac diseases, yet its interpretation demands extensive clinical experience and suffers from inter-observer variability. While deep learning models offer high diagnostic accuracy, their black-box nature hinders clinical adoption in high-stakes medical settings. To address this, we propose X-Ray-CoT (Chest X-Ray Chain-of-Thought), a novel framework leveraging Vision-Language Large Models (LVLMs) for intelligent chest X-ray diagnosis and interpretable report generation. X-Ray-CoT simulates human radiologists' "chain-of-thought" by first extracting multi-modal features and visual concepts, then employing an LLM-based component with a structured Chain-of-Thought prompting strategy to reason and produce detailed natural language diagnostic reports. Evaluated on the CORDA dataset, X-Ray-CoT achieves competitive quantitative performance, with a Balanced Accuracy of 80.52% and F1 score of 78.65% for disease diagnosis, slightly surpassing existing black-box models. Crucially, it uniquely generates high-quality, explainable reports, as validated by preliminary human evaluations. Our ablation studies confirm the integral role of each proposed component, highlighting the necessity of multi-modal fusion and CoT reasoning for robust and transparent medical AI. This work represents a significant step towards trustworthy and clinically actionable AI systems in medical imaging.

Paperid: 6987, https://arxiv.org/pdf/2508.12400.pdf

Abstract:
Despite significant advancements, Large Vision-Language Models (LVLMs) continue to face challenges in complex visual reasoning tasks that demand deep contextual understanding, multi-angle analysis, or meticulous detail recognition. Existing approaches often rely on single-shot image encoding and prompts, limiting their ability to fully capture nuanced visual information. Inspired by the notion that strategically generated "additional" information can serve as beneficial contextual augmentation, we propose Multi-Perspective Contextual Augmentation for Reasoning (MPCAR), a novel inference-time strategy designed to enhance LVLM performance. MPCAR operates in three stages: first, an LVLM generates N diverse and complementary descriptions or preliminary reasoning paths from various angles; second, these descriptions are intelligently integrated with the original question to construct a comprehensive context-augmented prompt; and finally, this enriched prompt guides the ultimate LVLM for deep reasoning and final answer generation. Crucially, MPCAR achieves these enhancements without requiring any fine-tuning of the underlying LVLM's parameters. Extensive experiments on challenging Visual Question Answering (VQA) datasets, including GQA, VQA-CP v2, and ScienceQA (Image-VQA), demonstrate that MPCAR consistently outperforms established baseline methods. Our quantitative results show significant accuracy gains, particularly on tasks requiring robust contextual understanding, while human evaluations confirm improved coherence and completeness of the generated answers. Ablation studies further highlight the importance of diverse prompt templates and the number of generated perspectives. This work underscores the efficacy of leveraging LVLMs' inherent generative capabilities to enrich input contexts, thereby unlocking their latent reasoning potential for complex multimodal tasks.

Paperid: 6988, https://arxiv.org/pdf/2508.12198.pdf

Abstract:
Forecasting from atmospheric soundings is a fundamental task in operational meteorology, often requiring structured visual reasoning over Skew-T log-P diagrams by human forecasters. While recent advances in Vision-Language Models (VLMs) have shown promise in other scientific domains, their application to meteorological diagram interpretation remains largely unexplored. In this study, we present a lightweight AI assistant that interprets Skew-T diagrams using a small language model (LM) and a small VLM fine-tuned to emulate human forecasters. Using a curriculum learning framework, we first train the models to identify key atmospheric features from diagrams through visual question answering, followed by chain-of-thought reasoning tasks that estimate precipitation probability based on the derived visual groundings. Model inputs include either textual summaries or generated Skew-T diagrams derived from operational Numerical Weather Prediction (NWP) forecasts, paired with three-hour precipitation observations from South Korea's Auto Weather Stations network. Evaluation results demonstrate that the fine-tuned VLM achieves skill comparable to an operational NWP model, despite relying solely on static atmospheric profiles. Ablation studies reveal that visual grounding and reasoning supervision are critical for performance, while attention map analysis confirms that the model learns to focus on relevant meteorological features. These findings highlight the potential of compact, interpretable multimodal models to support weather forecasting tasks. The approach offers a computationally efficient alternative to large-scale systems, and future work could extend it to more complex applications.

Paperid: 6989, https://arxiv.org/pdf/2508.11690.pdf

Abstract:
Child safety continues to be a paramount concern worldwide, with child abduction posing significant threats to communities. This paper presents the development of an edge-based child abduction detection and alert system utilizing a multi-agent framework where each agent incorporates Vision-Language Models (VLMs) deployed on a Raspberry Pi. Leveraging the advanced capabilities of VLMs within individual agents of a multi-agent team, our system is trained to accurately detect and interpret complex interactions involving children in various environments in real-time. The multi-agent system is deployed on a Raspberry Pi connected to a webcam, forming an edge device capable of processing video feeds, thereby reducing latency and enhancing privacy. An integrated alert system utilizes the Twilio API to send immediate SMS and WhatsApp notifications, including calls and messages, when a potential child abduction event is detected. Experimental results demonstrate that the system achieves high accuracy in detecting potential abduction scenarios, with near real-time performance suitable for practical deployment. The multi-agent architecture enhances the system's ability to process complex situational data, improving detection capabilities over traditional single-model approaches. The edge deployment ensures scalability and cost-effectiveness, making it accessible for widespread use. The proposed system offers a proactive solution to enhance child safety through continuous monitoring and rapid alerting, contributing a valuable tool in efforts to prevent child abductions.

Paperid: 6990, https://arxiv.org/pdf/2508.11341.pdf

Abstract:
In targeted adversarial attacks on vision models, the selection of the target label is a critical yet often overlooked determinant of attack success. This target label corresponds to the class that the attacker aims to force the model to predict. Now, existing strategies typically rely on randomness, model predictions, or static semantic resources, limiting interpretability, reproducibility, or flexibility. This paper then proposes a semantics-guided framework for adversarial target selection using the cross-modal knowledge transfer from pretrained language and vision-language models. We evaluate several state-of-the-art models (BERT, TinyLLAMA, and CLIP) as similarity sources to select the most and least semantically related labels with respect to the ground truth, forming best- and worst-case adversarial scenarios. Our experiments on three vision models and five attack methods reveal that these models consistently render practical adversarial targets and surpass static lexical databases, such as WordNet, particularly for distant class relationships. We also observe that static testing of target labels offers a preliminary assessment of the effectiveness of similarity sources, \textit{a priori} testing. Our results corroborate the suitability of pretrained models for constructing interpretable, standardized, and scalable adversarial benchmarks across architectures and datasets.

Paperid: 6991, https://arxiv.org/pdf/2508.11218.pdf

Abstract:
Re-Identification (ReID) is a critical technology in intelligent perception systems, especially within autonomous driving, where onboard cameras must identify pedestrians across views and time in real-time to support safe navigation and trajectory prediction. However, the presence of uncertain or missing input modalities--such as RGB, infrared, sketches, or textual descriptions--poses significant challenges to conventional ReID approaches. While large-scale pre-trained models offer strong multimodal semantic modeling capabilities, their computational overhead limits practical deployment in resource-constrained environments. To address these challenges, we propose a lightweight Uncertainty Modal Modeling (UMM) framework, which integrates a multimodal token mapper, synthetic modality augmentation strategy, and cross-modal cue interactive learner. Together, these components enable unified feature representation, mitigate the impact of missing modalities, and extract complementary information across different data types. Additionally, UMM leverages CLIP's vision-language alignment ability to fuse multimodal inputs efficiently without extensive finetuning. Experimental results demonstrate that UMM achieves strong robustness, generalization, and computational efficiency under uncertain modality conditions, offering a scalable and practical solution for pedestrian re-identification in autonomous driving scenarios.

Paperid: 6992, https://arxiv.org/pdf/2508.11011.pdf

Abstract:
Construction safety inspections typically involve a human inspector identifying safety concerns on-site. With the rise of powerful Vision Language Models (VLMs), researchers are exploring their use for tasks such as detecting safety rule violations from on-site images. However, there is a lack of open datasets to comprehensively evaluate and further fine-tune VLMs in construction safety inspection. Current applications of VLMs use small, supervised datasets, limiting their applicability in tasks they are not directly trained for. In this paper, we propose the ConstructionSite 10k, featuring 10,000 construction site images with annotations for three inter-connected tasks, including image captioning, safety rule violation visual question answering (VQA), and construction element visual grounding. Our subsequent evaluation of current state-of-the-art large pre-trained VLMs shows notable generalization abilities in zero-shot and few-shot settings, while additional training is needed to make them applicable to actual construction sites. This dataset allows researchers to train and evaluate their own VLMs with new architectures and techniques, providing a valuable benchmark for construction safety inspection.

Paperid: 6993, https://arxiv.org/pdf/2508.10429.pdf

Abstract:
We present MM-Food-100K, a public 100,000-sample multimodal food intelligence dataset with verifiable provenance. It is a curated approximately 10% open subset of an original 1.2 million, quality-accepted corpus of food images annotated for a wide range of information (such as dish name, region of creation). The corpus was collected over six weeks from over 87,000 contributors using the Codatta contribution model, which combines community sourcing with configurable AI-assisted quality checks; each submission is linked to a wallet address in a secure off-chain ledger for traceability, with a full on-chain protocol on the roadmap. We describe the schema, pipeline, and QA, and validate utility by fine-tuning large vision-language models (ChatGPT 5, ChatGPT OSS, Qwen-Max) on image-based nutrition prediction. Fine-tuning yields consistent gains over out-of-box baselines across standard metrics; we report results primarily on the MM-Food-100K subset. We release MM-Food-100K for publicly free access and retain approximately 90% for potential commercial access with revenue sharing to contributors.

Paperid: 6994, https://arxiv.org/pdf/2508.10397.pdf

Abstract:
Driver distraction detection is essential for improving traffic safety and reducing road accidents. However, existing models often suffer from degraded generalization when deployed in real-world scenarios. This limitation primarily arises from the few-shot learning challenge caused by the high cost of data annotation in practical environments, as well as the substantial domain shift between training datasets and target deployment conditions. To address these issues, we propose a Pose-driven Quality-controlled Data Augmentation Framework (PQ-DAF) that leverages a vision-language model for sample filtering to cost-effectively expand training data and enhance cross-domain robustness. Specifically, we employ a Progressive Conditional Diffusion Model (PCDMs) to accurately capture key driver pose features and synthesize diverse training examples. A sample quality assessment module, built upon the CogVLM vision-language model, is then introduced to filter out low-quality synthetic samples based on a confidence threshold, ensuring the reliability of the augmented dataset. Extensive experiments demonstrate that PQ-DAF substantially improves performance in few-shot driver distraction detection, achieving significant gains in model generalization under data-scarce conditions.

Paperid: 6995, https://arxiv.org/pdf/2508.09966.pdf

Abstract:
Progress in AI for automated nutritional analysis is critically hampered by the lack of standardized evaluation methodologies and high-quality, real-world benchmark datasets. To address this, we introduce three primary contributions. First, we present the January Food Benchmark (JFB), a publicly available collection of 1,000 food images with human-validated annotations. Second, we detail a comprehensive benchmarking framework, including robust metrics and a novel, application-oriented overall score designed to assess model performance holistically. Third, we provide baseline results from both general-purpose Vision-Language Models (VLMs) and our own specialized model, january/food-vision-v1. Our evaluation demonstrates that the specialized model achieves an Overall Score of 86.2, a 12.1-point improvement over the best-performing general-purpose configuration. This work offers the research community a valuable new evaluation dataset and a rigorous framework to guide and benchmark future developments in automated nutritional analysis.

Paperid: 6996, https://arxiv.org/pdf/2508.09225.pdf

Abstract:
Mammography report generation is a critical yet underexplored task in medical AI, characterized by challenges such as multiview image reasoning, high-resolution visual cues, and unstructured radiologic language. In this work, we introduce AMRG (Automatic Mammography Report Generation), the first end-to-end framework for generating narrative mammography reports using large vision-language models (VLMs). Building upon MedGemma-4B-it-a domain-specialized, instruction-tuned VLM-we employ a parameter-efficient fine-tuning (PEFT) strategy via Low-Rank Adaptation (LoRA), enabling lightweight adaptation with minimal computational overhead. We train and evaluate AMRG on DMID, a publicly available dataset of paired high-resolution mammograms and diagnostic reports. This work establishes the first reproducible benchmark for mammography report generation, addressing a longstanding gap in multimodal clinical AI. We systematically explore LoRA hyperparameter configurations and conduct comparative experiments across multiple VLM backbones, including both domain-specific and general-purpose models under a unified tuning protocol. Our framework demonstrates strong performance across both language generation and clinical metrics, achieving a ROUGE-L score of 0.5691, METEOR of 0.6152, CIDEr of 0.5818, and BI-RADS accuracy of 0.5582. Qualitative analysis further highlights improved diagnostic consistency and reduced hallucinations. AMRG offers a scalable and adaptable foundation for radiology report generation and paves the way for future research in multimodal medical AI.

Paperid: 6997, https://arxiv.org/pdf/2508.08644.pdf

Abstract:
Knowledge distillation is a long-established technique for knowledge transfer, and has regained attention in the context of the recent emergence of large vision-language models (VLMs). However, vision-language knowledge distillation often requires sufficient training data to achieve robust generalization on amples with ambiguous or boundary-adjacent representations, which are associated with high predictive uncertainty. Critically, collecting such large-scale, task-specific data for training is often impractical in real-world scenarios. To address this major challenge arising from the entanglement of uncertainty and cross-modal feature representation, we propose Aligned Manifold Entropy for Robust Vision-Language Distillation (AME), aiming to achieve robust generalization under real-world conditions. AME applies entropy minimization over a reconfigured shared manifold, where multi-modal data (i.e., image and text) are bridged through a pair of projection functions, conducive to structural compression for cross-modal feature representations. This enables robust knowledge distillation under low-data regimes, while requiring no architectural modifications to the backbone. As a result, it can serve as a plug-and-play module compatible with a wide range of vision-language distillation frameworks. Notably, our theoretical analysis reveals that integrating knowledge distillation with entropy minimization over the shared manifold leads to a tighter generalization error bound. Extensive experiments across diverse distillation architectures and training settings demonstrate that AME consistently facilitates robust knowledge distillation, resulting in superior generalization performance across a wide spectrum of downstream tasks.

Paperid: 6998, https://arxiv.org/pdf/2508.08199.pdf

Abstract:
Precise spatial modeling in the operating room (OR) is foundational to many clinical tasks, supporting intraoperative awareness, hazard avoidance, and surgical decision-making. While existing approaches leverage large-scale multimodal datasets for latent-space alignment to implicitly learn spatial relationships, they overlook the 3D capabilities of MLLMs. However, this approach raises two issues: (1) Operating rooms typically lack multiple video and audio sensors, making multimodal 3D data difficult to obtain; (2) Training solely on readily available 2D data fails to capture fine-grained details in complex scenes. To address this gap, we introduce Spatial-ORMLLM, the first large vision-language model for 3D spatial reasoning in operating rooms using only RGB modality to infer volumetric and semantic cues, enabling downstream medical tasks with detailed and holistic spatial context. Spatial-ORMLLM incorporates a Spatial-Enhanced Feature Fusion Block, which integrates 2D modality inputs with rich 3D spatial knowledge extracted by the estimation algorithm and then feeds the combined features into the visual tower. By employing a unified end-to-end MLLM framework, it combines powerful spatial features with textual features to deliver robust 3D scene reasoning without any additional expert annotations or sensor inputs. Experiments on multiple benchmark clinical datasets demonstrate that Spatial-ORMLLM achieves state-of-the-art performance and generalizes robustly to previously unseen surgical scenarios and downstream tasks.

Paperid: 6999, https://arxiv.org/pdf/2508.07432.pdf

Abstract:
Vision Language Models achieve impressive multi-modal performance but often inherit gender biases from their training data. This bias might be coming from both the vision and text modalities. In this work, we dissect the contributions of vision and text backbones to these biases by applying targeted debiasing using Counterfactual Data Augmentation and Task Vector methods. Inspired by data-efficient approaches in hate-speech classification, we introduce a novel metric, Degree of Stereotypicality and a corresponding debiasing method, Data Augmentation Using Degree of Stereotypicality - DAUDoS, to reduce bias with minimal computational cost. We curate a gender annotated dataset and evaluate all methods on VisoGender benchmark to quantify improvements and identify dominant source of bias. Our results show that CDA reduces the gender gap by 6% and DAUDoS by 3% but using only one-third of the data. Both methods also improve the model's ability to correctly identify gender in images by 3%, with DAUDoS achieving this improvement using only almost one-third of training data. From our experiment's, we observed that CLIP's vision encoder is more biased whereas PaliGemma2's text encoder is more biased. By identifying whether bias stems more from vision or text encoders, our work enables more targeted and effective bias mitigation strategies in future multi-modal systems.

Paperid: 7000, https://arxiv.org/pdf/2508.07023.pdf

Abstract:
Complex Visual Question Answering (Complex VQA) tasks, which demand sophisticated multi-modal reasoning and external knowledge integration, present significant challenges for existing large vision-language models (LVLMs) often limited by their reliance on high-level global features. To address this, we propose MV-CoRe (Multimodal Visual-Conceptual Reasoning), a novel model designed to enhance Complex VQA performance through the deep fusion of diverse visual and linguistic information. MV-CoRe meticulously integrates global embeddings from pre-trained Vision Large Models (VLMs) and Language Large Models (LLMs) with fine-grained semantic-aware visual features, including object detection characteristics and scene graph representations. An innovative Multimodal Fusion Transformer then processes and deeply integrates these diverse feature sets, enabling rich cross-modal attention and facilitating complex reasoning. We evaluate MV-CoRe on challenging Complex VQA benchmarks, including GQA, A-OKVQA, and OKVQA, after training on VQAv2. Our experimental results demonstrate that MV-CoRe consistently outperforms established LVLM baselines, achieving an overall accuracy of 77.5% on GQA. Ablation studies confirm the critical contribution of both object and scene graph features, and human evaluations further validate MV-CoRe's superior factual correctness and reasoning depth, underscoring its robust capabilities for deep visual and conceptual understanding.

Paperid: 7001, https://arxiv.org/pdf/2508.06623.pdf

Abstract:
The proliferation of digital news media necessitates robust methods for verifying content veracity, particularly regarding the consistency between visual and textual information. Traditional approaches often fall short in addressing the fine-grained cross-modal contextual consistency (FCCC) problem, which encompasses deeper alignment of visual narrative, emotional tone, and background information with text, beyond mere entity matching. To address this, we propose ContextGuard-LVLM, a novel framework built upon advanced Vision-Language Large Models (LVLMs) and integrating a multi-stage contextual reasoning mechanism. Our model is uniquely enhanced through reinforced or adversarial learning paradigms, enabling it to detect subtle contextual misalignments that evade zero-shot baselines. We extend and augment three established datasets (TamperedNews-Ent, News400-Ent, MMG-Ent) with new fine-grained contextual annotations, including "contextual sentiment," "visual narrative theme," and "scene-event logical coherence," and introduce a comprehensive CTXT (Contextual Coherence) entity type. Extensive experiments demonstrate that ContextGuard-LVLM consistently outperforms state-of-the-art zero-shot LVLM baselines (InstructBLIP and LLaVA 1.5) across nearly all fine-grained consistency tasks, showing significant improvements in complex logical reasoning and nuanced contextual understanding. Furthermore, our model exhibits superior robustness to subtle perturbations and a higher agreement rate with human expert judgments on challenging samples, affirming its efficacy in discerning sophisticated forms of context detachment.

Paperid: 7002, https://arxiv.org/pdf/2508.06525.pdf

Abstract:
This paper presents several novel findings on the explainability of vision reflection in large multimodal models (LMMs). First, we show that prompting an LMM to verify the prediction of a specialized vision model can improve recognition accuracy, even on benchmarks like ImageNet, despite prior evidence that LMMs typically underperform dedicated vision encoders. Second, we analyze the internal behavior of vision reflection and find that the vision-language connector maps visual features into explicit textual concepts, allowing the language model to reason about prediction plausibility using commonsense knowledge. We further observe that replacing a large number of vision tokens with only a few text tokens still enables LLaVA to generate similar answers, suggesting that LMMs may rely primarily on a compact set of distilled textual representations rather than raw vision features. Third, we show that a training-free connector can enhance LMM performance in fine-grained recognition tasks, without extensive feature-alignment training. Together, these findings offer new insights into the explainability of vision-language models and suggest that vision reflection is a promising strategy for achieving robust and interpretable visual recognition.

Paperid: 7003, https://arxiv.org/pdf/2508.06152.pdf

Abstract:
We present VISTAR, a user-centric, multi-dimensional benchmark for text-to-image (T2I) evaluation that addresses the limitations of existing metrics. VISTAR introduces a two-tier hybrid paradigm: it employs deterministic, scriptable metrics for physically quantifiable attributes (e.g., text rendering, lighting) and a novel Hierarchical Weighted P/N Questioning (HWPQ) scheme that uses constrained vision-language models to assess abstract semantics (e.g., style fusion, cultural fidelity). Grounded in a Delphi study with 120 experts, we defined seven user roles and nine evaluation angles to construct the benchmark, which comprises 2,845 prompts validated by over 15,000 human pairwise comparisons. Our metrics achieve high human alignment (>75%), with the HWPQ scheme reaching 85.9% accuracy on abstract semantics, significantly outperforming VQA baselines. Comprehensive evaluation of state-of-the-art models reveals no universal champion, as role-weighted scores reorder rankings and provide actionable guidance for domain-specific deployment. All resources are publicly released to foster reproducible T2I assessment.

Paperid: 7004, https://arxiv.org/pdf/2508.05684.pdf

Abstract:
The proliferation of multi-modal fake news on social media poses a significant threat to public trust and social stability. Traditional detection methods, primarily text-based, often fall short due to the deceptive interplay between misleading text and images. While Large Vision-Language Models (LVLMs) offer promising avenues for multi-modal understanding, effectively fusing diverse modal information, especially when their importance is imbalanced or contradictory, remains a critical challenge. This paper introduces MM-FusionNet, an innovative framework leveraging LVLMs for robust multi-modal fake news detection. Our core contribution is the Context-Aware Dynamic Fusion Module (CADFM), which employs bi-directional cross-modal attention and a novel dynamic modal gating network. This mechanism adaptively learns and assigns importance weights to textual and visual features based on their contextual relevance, enabling intelligent prioritization of information. Evaluated on the large-scale Multi-modal Fake News Dataset (LMFND) comprising 80,000 samples, MM-FusionNet achieves a state-of-the-art F1-score of 0.938, surpassing existing multi-modal baselines by approximately 0.5% and significantly outperforming single-modal approaches. Further analysis demonstrates the model's dynamic weighting capabilities, its robustness to modality perturbations, and performance remarkably close to human-level, underscoring its practical efficacy and interpretability for real-world fake news detection.

Paperid: 7005, https://arxiv.org/pdf/2508.05237.pdf

Abstract:
This report synthesizes eight seminal papers on the zero-shot adversarial robustness of vision-language models (VLMs) like CLIP. A central challenge in this domain is the inherent trade-off between enhancing adversarial robustness and preserving the model's zero-shot generalization capabilities. We analyze two primary defense paradigms: Adversarial Fine-Tuning (AFT), which modifies model parameters, and Training-Free/Test-Time Defenses, which preserve them. We trace the evolution from alignment-preserving methods (TeCoA) to embedding space re-engineering (LAAT, TIMA), and from input heuristics (AOM, TTC) to latent-space purification (CLIPure). Finally, we identify key challenges and future directions including hybrid defense strategies and adversarial pre-training.

Paperid: 7006, https://arxiv.org/pdf/2508.04998.pdf

Abstract:
Person re-identification (Re-ID) aims to match person images across different camera views, with occluded Re-ID addressing scenarios where pedestrians are partially visible. While pre-trained vision-language models have shown effectiveness in Re-ID tasks, they face significant challenges in occluded scenarios by focusing on holistic image semantics while neglecting fine-grained attribute information. This limitation becomes particularly evident when dealing with partially occluded pedestrians or when distinguishing between individuals with subtle appearance differences. To address this limitation, we propose Attribute-Guide ReID (AG-ReID), a novel framework that leverages pre-trained models' inherent capabilities to extract fine-grained semantic attributes without additional data or annotations. Our framework operates through a two-stage process: first generating attribute pseudo-labels that capture subtle visual characteristics, then introducing a dual-guidance mechanism that combines holistic and fine-grained attribute information to enhance image feature extraction. Extensive experiments demonstrate that AG-ReID achieves state-of-the-art results on multiple widely-used Re-ID datasets, showing significant improvements in handling occlusions and subtle attribute differences while maintaining competitive performance on standard Re-ID scenarios.

Paperid: 7007, https://arxiv.org/pdf/2508.04427.pdf

Abstract:
Multimodal learning has witnessed remarkable advancements in recent years, particularly with the integration of attention-based models, leading to significant performance gains across a variety of tasks. Parallel to this progress, the demand for explainable artificial intelligence (XAI) has spurred a growing body of research aimed at interpreting the complex decision-making processes of these models. This systematic literature review analyzes research published between January 2020 and early 2024 that focuses on the explainability of multimodal models. Framed within the broader goals of XAI, we examine the literature across multiple dimensions, including model architecture, modalities involved, explanation algorithms and evaluation methodologies. Our analysis reveals that the majority of studies are concentrated on vision-language and language-only models, with attention-based techniques being the most commonly employed for explanation. However, these methods often fall short in capturing the full spectrum of interactions between modalities, a challenge further compounded by the architectural heterogeneity across domains. Importantly, we find that evaluation methods for XAI in multimodal settings are largely non-systematic, lacking consistency, robustness, and consideration for modality-specific cognitive and contextual factors. Based on these findings, we provide a comprehensive set of recommendations aimed at promoting rigorous, transparent, and standardized evaluation and reporting practices in multimodal XAI research. Our goal is to support future research in more interpretable, accountable, and responsible mulitmodal AI systems, with explainability at their core.

Paperid: 7008, https://arxiv.org/pdf/2508.03737.pdf

Abstract:
Benchmarks for evaluating reasoning among Vision Language Models (VLMs) on several fields and domains are being curated more frequently over the last few years. However these are often monolingual, mostly available in English. Additionally there also is a lack of datasets available in Hindi on tasks apart from comprehension and translation. We introduce GanitBench, a tough benchmark consisting of 1527 vision-only questions covering several topics in Mathematics - available in languages English and Hindi. Collected from two major examinations from India, the JEE Advanced and the CBSE Boards examinations, this benchmark includes questions in the form of images comprising of figures essential to a question as well as text. We evaluate two closed source models for the same, in zero-shot Chain-of-Thought (CoT) and two-shot CoT settings. GPT-4o mini is found to be the more dominant model on the benchmark, with it's highest average accuracy being 38.15%. We also evaluate models through a "Double Lock" constraint, which brings down the performance of the models by considerable margins. We observe that two-shot CoT appears to be a more effective setting under this environment. Performance of the two VLMs also decreases when answering the same questions in the Hindi language. We hope to facilitate the inclusion of languages like Hindi in research through our work.

Paperid: 7009, https://arxiv.org/pdf/2508.02917.pdf

Abstract:
Vision-and-Language Navigation (VLN) refers to the task of enabling autonomous robots to navigate unfamiliar environments by following natural language instructions. While recent Large Vision-Language Models (LVLMs) have shown promise in this task, most current VLM systems rely on models specifically designed and optimized for navigation, leaving the potential of off-the-shelf LVLMs underexplored. Furthermore, while older VLN approaches used low-level action spaces with egocentric views and atomic actions (such as "turn left" or "move forward"), newer models tend to favor panoramic action spaces with discrete navigable viewpoints. This paper investigates (1) whether off-the-shelf LVLMs (fine-tuned without architectural modifications or simulator-based training) can effectively support VLN tasks and (2) whether such models can support both low-level and panoramic action paradigms. To this end, we fine-tune the open-source model Qwen2.5-VL-3B-Instruct on the Room-to-Room (R2R) dataset and evaluate its empirical performance across both low-level and panoramic action spaces. The best resulting model achieves a 41% success rate on the R2R test set, demonstrating that while off-the-shelf LVLMs can learn to perform Vision-and-Language Navigation, they still lag behind models specifically designed for this task.

Paperid: 7010, https://arxiv.org/pdf/2508.02886.pdf

Abstract:
Despite significant advancements, current large language models (LLMs) and vision-language models (LVLMs) continue to struggle with complex, multi-step, cross-modal common sense reasoning tasks, often exhibiting a lack of "deliberative thinking." They tend to rely on superficial associations rather than deep, chained inference, particularly when integrating visual information with abstract concepts. To address this, we propose the Coherent Multimodal Reasoning Framework (CMRF), a novel approach that enhances LVLMs' common sense reasoning capabilities through an iterative, self-evaluating inference mechanism. CMRF mimics human problem-solving by decomposing complex queries, generating step-by-step inferences, and self-correcting errors. Our framework integrates three key modules: a Reasoning Decomposition Unit (RDU) for breaking down problems into sub-questions, a Contextual Inference Engine (CIE) for contextual inference, and a Coherence Assessment Module (CAM) for evaluating logical consistency and confidence. Coupled with an Adaptive Iterative Refinement strategy, CMRF systematically refines its reasoning paths. Built upon LLaVA-1.6-34B and trained on a novel Multimodal Daily Activity Reasoning (MDAR) dataset, CMRF achieves state-of-the-art performance among open-source LVLMs on challenging benchmarks like VCR, A-OKVQA, and DailyLife-MRC. It attains an average accuracy of 69.4%, surpassing the best open-source baseline by +2.4 percentage points, with particular strength in complex reasoning scenarios. Extensive ablation studies and human evaluations confirm the critical contributions of each module and the effectiveness of iterative refinement in fostering more coherent and accurate reasoning.

Paperid: 7011, https://arxiv.org/pdf/2508.02173.pdf

Abstract:
Mixed reality platforms allow users to create virtual environments, yet novice users struggle with both ideation and execution in spatial design. While existing AI models can automatically generate scenes based on user prompts, the lack of interactive control limits users' ability to iteratively steer the output. In this paper, we present EchoLadder, a novel human-AI collaboration pipeline that leverages large vision-language model (LVLM) to support interactive scene modification in virtual reality. EchoLadder accepts users' verbal instructions at varied levels of abstraction and spatial specificity, generates concrete design suggestions throughout a progressive design process. The suggestions can be automatically applied, regenerated and retracted by users' toggle control.Our ablation study showed effectiveness of our pipeline components. Our user study found that, compared to baseline without showing suggestions, EchoLadder better supports user creativity in spatial design. It also contributes insights on users' progressive design strategies under AI assistance, providing design implications for future systems.

Paperid: 7012, https://arxiv.org/pdf/2508.01715.pdf

Abstract:
Terrain traversability estimation is crucial for autonomous robots, especially in unstructured environments where visual cues and reasoning play a key role. While vision-language models (VLMs) offer potential for zero-shot estimation, the problem remains inherently ill-posed. To explore this, we introduce a small dataset of human-annotated water traversability ratings, revealing that while estimations are subjective, human raters still show some consensus. Additionally, we propose a simple pipeline that integrates VLMs for zero-shot traversability estimation. Our experiments reveal mixed results, suggesting that current foundation models are not yet suitable for practical deployment but provide valuable insights for further research.

Paperid: 7013, https://arxiv.org/pdf/2508.01387.pdf

Abstract:
Automatic license plate recognition (ALPR) and vehicle make and model recognition underpin intelligent transportation systems, supporting law enforcement, toll collection, and post-incident investigation. Applying these methods to videos captured by handheld smartphones or non-static vehicle-mounted cameras presents unique challenges compared to fixed installations, including frequent camera motion, varying viewpoints, occlusions, and unknown road geometry. Traditional ALPR solutions, dependent on specialized hardware and handcrafted OCR pipelines, often degrade under these conditions. Recent advances in large vision-language models (VLMs) enable direct recognition of textual and semantic attributes from arbitrary imagery. This study evaluates the potential of VLMs for ALPR and makes and models recognition using monocular videos captured with handheld smartphones and non-static mounted cameras. The proposed license plate recognition pipeline filters to sharp frames, then sends a multimodal prompt to a VLM using several prompt strategies. Make and model recognition pipeline runs the same VLM with a revised prompt and an optional self-reflection module. In the self-reflection module, the model contrasts the query image with a reference from a 134-class dataset, correcting mismatches. Experiments on a smartphone dataset collected on the campus of the University of Texas at Austin, achieve top-1 accuracies of 91.67% for ALPR and 66.67% for make and model recognition. On the public UFPR-ALPR dataset, the approach attains 83.05% and 61.07%, respectively. The self-reflection module further improves results by 5.72% on average for make and model recognition. These findings demonstrate that VLMs provide a cost-effective solution for scalable, in-motion traffic video analysis.

Paperid: 7014, https://arxiv.org/pdf/2507.23402.pdf

Abstract:
Learning medical visual representations from paired images and reports is a promising direction in representation learning. However, current vision-language pretraining methods in the medical domain often simplify clinical reports into single entities or fragmented tokens, ignoring their inherent structure. In addition, contrastive learning frameworks typically depend on large quantities of hard negative samples, which is impractical for small-scale medical datasets. To tackle these challenges, we propose Adaptive Grouped Alignment (AGA), a new framework that captures structured semantics from paired medical images and reports. AGA introduces a bidirectional grouping mechanism based on a sparse similarity matrix. For each image-report pair, we compute fine-grained similarities between text tokens and image patches. Each token selects its top-matching patches to form a visual group, and each patch selects its most related tokens to form a language group. To enable adaptive grouping, we design two threshold gating modules, called Language Grouped Threshold Gate and Vision Grouped Threshold Gate, which learn grouping thresholds dynamically. Group representations are computed as weighted averages based on similarity scores. To align each token with its group representation, we introduce an Instance Aware Group Alignment loss that operates within each image-text pair, removing the need for external negatives. Finally, a Bidirectional Cross-modal Grouped Alignment module is applied to enhance fine-grained alignment between visual and linguistic group representations. Extensive experiments on public and private datasets show that our method achieves strong performance on image-text retrieval and classification tasks under both fine-tuning and zero-shot settings.

Paperid: 7015, https://arxiv.org/pdf/2507.23135.pdf

Abstract:
Understanding causal relationships across modalities is a core challenge for multimodal models operating in real-world environments. We introduce ISO-Bench, a benchmark for evaluating whether models can infer causal dependencies between visual observations and procedural text. Each example presents an image of a task step and a text snippet from a plan, with the goal of deciding whether the visual step occurs before or after the referenced text step. Evaluation results on ten frontier vision-language models show underwhelming performance: the best zero-shot F1 is only 0.57, and chain-of-thought reasoning yields only modest gains (up to 0.62 F1), largely behind humans (0.98 F1). Our analysis further highlights concrete directions for improving causal understanding in multimodal models.

Paperid: 7016, https://arxiv.org/pdf/2507.22933.pdf

Abstract:
Recent advances in visual-language machine learning models have demonstrated exceptional ability to use natural language and understand visual scenes by training on large, unstructured datasets. However, this training paradigm cannot produce interpretable explanations for its outputs, requires retraining to integrate new information, is highly resource-intensive, and struggles with certain forms of logical reasoning. One promising solution involves integrating neural networks with external symbolic information systems, forming neural symbolic systems that can enhance reasoning and memory abilities. These neural symbolic systems provide more interpretable explanations to their outputs and the capacity to assimilate new information without extensive retraining. Utilizing powerful pre-trained Vision-Language Models (VLMs) as the core neural component, augmented by external systems, offers a pragmatic approach to realizing the benefits of neural-symbolic integration. This systematic literature review aims to categorize techniques through which visual-language understanding can be improved by interacting with external symbolic information systems.

Paperid: 7017, https://arxiv.org/pdf/2507.20913.pdf

Abstract:
The rapid evolution of face manipulation techniques poses a critical challenge for face forgery detection: cross-domain generalization. Conventional methods, which rely on simple classification objectives, often fail to learn domain-invariant representations. We propose HAMLET-FFD, a cognitively inspired Hierarchical Adaptive Multi-modal Learning framework that tackles this challenge via bidirectional cross-modal reasoning. Building on contrastive vision-language models such as CLIP, HAMLET-FFD introduces a knowledge refinement loop that iteratively assesses authenticity by integrating visual evidence with conceptual cues, emulating expert forensic analysis. A key innovation is a bidirectional fusion mechanism in which textual authenticity embeddings guide the aggregation of hierarchical visual features, while modulated visual features refine text embeddings to generate image-adaptive prompts. This closed-loop process progressively aligns visual observations with semantic priors to enhance authenticity assessment. By design, HAMLET-FFD freezes all pretrained parameters, serving as an external plugin that preserves CLIP's original capabilities. Extensive experiments demonstrate its superior generalization to unseen manipulations across multiple benchmarks, and visual analyses reveal a division of labor among embeddings, with distinct representations specializing in fine-grained artifact recognition.

Paperid: 7018, https://arxiv.org/pdf/2507.20188.pdf

Abstract:
Detecting text in natural scenes remains challenging, particularly for diverse scripts and arbitrarily shaped instances where visual cues alone are often insufficient. Existing methods do not fully leverage semantic context. This paper introduces SAViL-Det, a novel semantic-aware vision-language model that enhances multi-script text detection by effectively integrating textual prompts with visual features. SAViL-Det utilizes a pre-trained CLIP model combined with an Asymptotic Feature Pyramid Network (AFPN) for multi-scale visual feature fusion. The core of the proposed framework is a novel language-vision decoder that adaptively propagates fine-grained semantic information from text prompts to visual features via cross-modal attention. Furthermore, a text-to-pixel contrastive learning mechanism explicitly aligns textual and corresponding visual pixel features. Extensive experiments on challenging benchmarks demonstrate the effectiveness of the proposed approach, achieving state-of-the-art performance with F-scores of 84.8% on the benchmark multi-lingual MLT-2019 dataset and 90.2% on the curved-text CTW1500 dataset.

Paperid: 7019, https://arxiv.org/pdf/2507.18675.pdf

Abstract:
Human action recognition plays a critical role in healthcare and medicine, supporting applications such as patient behavior monitoring, fall detection, surgical robot supervision, and procedural skill assessment. While traditional models like CNNs and RNNs have achieved moderate success, they often struggle to generalize across diverse and complex actions. Recent advancements in vision-language models, especially the transformer-based CLIP model, offer promising capabilities for generalizing action recognition from video data. In this work, we evaluate CLIP on the UCF-101 dataset and systematically analyze its performance under three masking strategies: (1) percentage-based and shape-based black masking at 10%, 30%, and 50%, (2) feature-specific masking to suppress bias-inducing elements, and (3) isolation masking that retains only class-specific regions. Our results reveal that CLIP exhibits inconsistent behavior and frequent misclassifications, particularly when essential visual cues are obscured. To overcome these limitations, we propose incorporating class-specific noise, learned via a custom loss function, to reinforce attention to class-defining features. This enhancement improves classification accuracy and model confidence while reducing bias. We conclude with a discussion on the challenges of applying such models in clinical domains and outline directions for future work to improve generalizability across domain-independent healthcare scenarios.

Paperid: 7020, https://arxiv.org/pdf/2507.18009.pdf

Abstract:
State-of-the-art (SOTA) image and text generation models are multimodal models that have many similarities to large language models (LLMs). Despite achieving strong performances, leading foundational multimodal model architectures frequently lag behind the architectural sophistication of contemporary LLMs. We propose GRR-CoCa, an improved SOTA Contrastive Captioner (CoCa) model that incorporates Gaussian error gated linear units, root mean squared normalization, and rotary positional embedding into the textual decoders and the vision transformer (ViT) encoder. Each architectural modification has been shown to improve model performance in LLMs, but has yet to be adopted in CoCa. We benchmarked GRR-CoCa against Baseline CoCa, a model with the same modified textual decoders but with CoCa's original ViT encoder. We used standard pretraining and fine-tuning workflows to benchmark the models on contrastive and generative tasks. Our GRR-CoCa significantly outperformed Baseline CoCa on the pretraining dataset and three diverse fine-tuning datasets. Pretraining improvements were 27.25% in contrastive loss, 3.71% in perplexity, and 7.15% in CoCa loss. The average fine-tuning improvements were 13.66% in contrastive loss, 5.18% in perplexity, and 5.55% in CoCa loss. We show that GRR-CoCa's modified architecture improves performance and generalization across vision-language domains.

Paperid: 7021, https://arxiv.org/pdf/2507.17118.pdf

Abstract:
AI has become integral to safety-critical areas like autonomous driving systems (ADS) and robotics. The architecture of recent autonomous systems are trending toward end-to-end (E2E) monolithic architectures such as large language models (LLMs) and vision language models (VLMs). In this paper, we review different architectural solutions and then evaluate the efficacy of common safety analyses such as failure modes and effect analysis (FMEA) and fault tree analysis (FTA). We show how these techniques can be improved for the intricate nature of the foundational models, particularly in how they form and utilize latent representations. We introduce HySAFE-AI, Hybrid Safety Architectural Analysis Framework for AI Systems, a hybrid framework that adapts traditional methods to evaluate the safety of AI systems. Lastly, we offer hints of future work and suggestions to guide the evolution of future AI safety standards.

Paperid: 7022, https://arxiv.org/pdf/2507.15680.pdf

Abstract:
Image Quality Assessment (IQA) is a core task in computer vision. Multimodal methods based on vision-language models, such as CLIP, have demonstrated exceptional generalization capabilities in IQA tasks. To address the issues of excessive parameter burden and insufficient ability to identify local distorted features in CLIP for IQA, this study proposes a visual-language model knowledge distillation method aimed at guiding the training of models with architectural advantages using CLIP's IQA knowledge. First, quality-graded prompt templates were designed to guide CLIP to output quality scores. Then, CLIP is fine-tuned to enhance its capabilities in IQA tasks. Finally, a modality-adaptive knowledge distillation strategy is proposed to achieve guidance from the CLIP teacher model to the student model. Our experiments were conducted on multiple IQA datasets, and the results show that the proposed method significantly reduces model complexity while outperforming existing IQA methods, demonstrating strong potential for practical deployment.

Paperid: 7023, https://arxiv.org/pdf/2507.13568.pdf

Abstract:
Continual learning for vision-language models has achieved remarkable performance through synthetic replay, where samples are generated using Stable Diffusion to regularize during finetuning and retain knowledge. However, real-world downstream applications often exhibit domain-specific nuances and fine-grained semantics not captured by generators, causing synthetic-replay methods to produce misaligned samples that misguide finetuning and undermine retention of prior knowledge. In this work, we propose a LoRA-enhanced synthetic-replay framework that injects task-specific low-rank adapters into a frozen Stable Diffusion model, efficiently capturing each new task's unique visual and semantic patterns. Specifically, we introduce a two-stage, confidence-based sample selection: we first rank real task data by post-finetuning VLM confidence to focus LoRA finetuning on the most representative examples, then generate synthetic samples and again select them by confidence for distillation. Our approach integrates seamlessly with existing replay pipelines-simply swap in the adapted generator to boost replay fidelity. Extensive experiments on the Multi-domain Task Incremental Learning (MTIL) benchmark show that our method outperforms previous synthetic-replay techniques, achieving an optimal balance among plasticity, stability, and zero-shot capability. These results demonstrate the effectiveness of generator adaptation via LoRA for robust continual learning in VLMs.

Paperid: 7024, https://arxiv.org/pdf/2507.13113.pdf

Abstract:
IRSTD (InfraRed Small Target Detection) detects small targets in infrared blurry backgrounds and is essential for various applications. The detection task is challenging due to the small size of the targets and their sparse distribution in infrared small target datasets. Although existing IRSTD methods and datasets have led to significant advancements, they are limited by their reliance solely on the image modality. Recent advances in deep learning and large vision-language models have shown remarkable performance in various visual recognition tasks. In this work, we propose a novel multimodal IRSTD framework that incorporates language priors to guide small target detection. We leverage language-guided attention weights derived from the language prior to enhance the model's ability for IRSTD, presenting a novel approach that combines textual information with image data to improve IRSTD capabilities. Utilizing the state-of-the-art GPT-4 vision model, we generate text descriptions that provide the locations of small targets in infrared images, employing careful prompt engineering to ensure improved accuracy. Due to the absence of multimodal IR datasets, existing IRSTD methods rely solely on image data. To address this shortcoming, we have curated a multimodal infrared dataset that includes both image and text modalities for small target detection, expanding upon the popular IRSTD-1k and NUDT-SIRST datasets. We validate the effectiveness of our approach through extensive experiments and comprehensive ablation studies. The results demonstrate significant improvements over the state-of-the-art method, with relative percentage differences of 9.74%, 13.02%, 1.25%, and 67.87% in IoU, nIoU, Pd, and Fa on the NUAA-SIRST subset, and 4.41%, 2.04%, 2.01%, and 113.43% on the IRSTD-1k subset of the LangIR dataset, respectively.

Paperid: 7025, https://arxiv.org/pdf/2507.11730.pdf

Abstract:
Outdoor advertisements remain a critical medium for modern marketing, yet accurately verifying billboard text visibility under real-world conditions is still challenging. Traditional Optical Character Recognition (OCR) pipelines excel at cropped text recognition but often struggle with complex outdoor scenes, varying fonts, and weather-induced visual noise. Recently, multimodal Vision-Language Models (VLMs) have emerged as promising alternatives, offering end-to-end scene understanding with no explicit detection step. This work systematically benchmarks representative VLMs - including Qwen 2.5 VL 3B, InternVL3, and SmolVLM2 - against a compact CNN-based OCR baseline (PaddleOCRv4) across two public datasets (ICDAR 2015 and SVT), augmented with synthetic weather distortions to simulate realistic degradation. Our results reveal that while selected VLMs excel at holistic scene reasoning, lightweight CNN pipelines still achieve competitive accuracy for cropped text at a fraction of the computational cost-an important consideration for edge deployment. To foster future research, we release our weather-augmented benchmark and evaluation code publicly.

Paperid: 7026, https://arxiv.org/pdf/2507.10106.pdf

Abstract:
As AI systems become increasingly capable and ubiquitous, ensuring the safety of these systems is critical. However, existing safety tools often target different aspects of model safety and cannot provide full assurance in isolation, highlighting a need for integrated and composite methodologies. This paper introduces BlueGlass, a framework designed to facilitate composite AI safety workflows by providing a unified infrastructure enabling the integration and composition of diverse safety tools that operate across model internals and outputs. Furthermore, to demonstrate the utility of this framework, we present three safety-oriented analyses on vision-language models for the task of object detection: (1) distributional evaluation, revealing performance trade-offs and potential failure modes across distributions; (2) probe-based analysis of layer dynamics highlighting shared hierarchical learning via phase transition; and (3) sparse autoencoders identifying interpretable concepts. More broadly, this work contributes foundational infrastructure and findings for building more robust and reliable AI systems.

Paperid: 7027, https://arxiv.org/pdf/2507.10087.pdf

Abstract:
The rapid emergence of foundation models, particularly Large Language Models (LLMs) and Vision-Language Models (VLMs), has introduced a transformative paradigm in robotics. These models offer powerful capabilities in semantic understanding, high-level reasoning, and cross-modal generalization, enabling significant advances in perception, planning, control, and human-robot interaction. This critical review provides a structured synthesis of recent developments, categorizing applications across simulation-driven design, open-world execution, sim-to-real transfer, and adaptable robotics. Unlike existing surveys that emphasize isolated capabilities, this work highlights integrated, system-level strategies and evaluates their practical feasibility in real-world environments. Key enabling trends such as procedural scene generation, policy generalization, and multimodal reasoning are discussed alongside core bottlenecks, including limited embodiment, lack of multimodal data, safety risks, and computational constraints. Through this lens, this paper identifies both the architectural strengths and critical limitations of foundation model-based robotics, highlighting open challenges in real-time operation, grounding, resilience, and trust. The review concludes with a roadmap for future research aimed at bridging semantic reasoning and physical intelligence through more robust, interpretable, and embodied models.

Paperid: 7028, https://arxiv.org/pdf/2507.09011.pdf

Abstract:
A rapidly alternating red and black display known as Ganzflicker induces visual hallucinations that reflect the generative capacity of the visual system. Recent proposals regarding the imagery spectrum, that is, differences in the visual system of individuals with absent imagery, typical imagery, and vivid imagery, suggest these differences should impact the complexity of other internally generated visual experiences. Here, we used tools from natural language processing to analyze free-text descriptions of hallucinations from over 4,000 participants, asking whether people with different imagery phenotypes see different things in their mind's eye during Ganzflicker-induced hallucinations. Strong imagers described complex, naturalistic content, while weak imagers reported simple geometric patterns. Embeddings from vision language models better captured these differences than text-only language models, and participants with stronger imagery used language with richer sensorimotor associations. These findings may reflect individual variation in coordination between early visual areas and higher-order regions relevant for the imagery spectrum.

Paperid: 7029, https://arxiv.org/pdf/2507.08831.pdf

Abstract:
Vision-Language Navigation in Continuous Environments (VLNCE), where an agent follows instructions and moves freely to reach a destination, is a key research problem in embodied AI. However, most navigation policies are sensitive to viewpoint changes, i.e., variations in camera height and viewing angle that alter the agent's observation. In this paper, we introduce a generalized scenario, V2-VLNCE (VLNCE with Varied Viewpoints), and propose VIL (View Invariant Learning), a view-invariant post-training strategy that enhances the robustness of existing navigation policies to changes in camera viewpoint. VIL employs a contrastive learning framework to learn sparse and view-invariant features. Additionally, we introduce a teacher-student framework for the Waypoint Predictor Module, a core component of most VLNCE baselines, where a view-dependent teacher model distills knowledge into a view-invariant student model. We employ an end-to-end training paradigm to jointly optimize these components, thus eliminating the cost for individual module training. Empirical results show that our method outperforms state-of-the-art approaches on V2-VLNCE by 8-15% measured on Success Rate for two standard benchmark datasets R2R-CE and RxR-CE. Furthermore, we evaluate VIL under the standard VLNCE setting and find that, despite being trained for varied viewpoints, it often still improves performance. On the more challenging RxR-CE dataset, our method also achieved state-of-the-art performance across all metrics when compared to other map-free methods. This suggests that adding VIL does not diminish the standard viewpoint performance and can serve as a plug-and-play post-training method.

Paperid: 7030, https://arxiv.org/pdf/2507.08410.pdf

Abstract:
Prompt learning facilitates the efficient adaptation of Vision-Language Models (VLMs) to various downstream tasks. However, it faces two significant challenges: (1) inadequate modeling of class embedding distributions for unseen instances, leading to suboptimal generalization on novel classes; (2) prevailing methodologies predominantly confine cross-modal alignment to the final output layer of vision and text encoders, which fundamentally limits their capacity to preserve topological consistency with pre-trained multi-modal embedding spaces. To this end, we introduce MuGCP (Multi-modal Mutual-Guidance Conditional Prompt Learning), a novel paradigm designed for conditional prompt generation. MuGCP leverages Multi-modal Large Language Models (MLLMs) as conditional prompt learners to adaptively generate Semantic Conditional Prompts (SCP) that incorporate rich, fine-grained high-level semantic knowledge for image instances. To ensure effective alignment and interaction across the multi-modal space of Vision-Language Models (VLMs), we introduce the Attention Mutual-Guidance (AMG) module, which facilitates interactions between visual and semantic information. Through mutual guidance, the AMG module generates Visual Conditional Prompts (VCP), enhancing the model's performance in multi-modal tasks. Additionally, we present a Multi-Prompt Fusion (MPF) mechanism that integrates SCP and VCP with contextual prompts, ensuring seamless coordination among the different prompts and enhancing the modeling of class embeddings and instance-specific knowledge. Our MuGCP outperforms existing state-of-the-art methods on 14 different datasets. The code will be made available after publication.

Paperid: 7031, https://arxiv.org/pdf/2507.08030.pdf

Abstract:
Generative AI models, including large language models (LLMs) and vision-language models (VLMs), are increasingly used to interpret medical images and answer clinical questions. Their responses often include inaccuracies; therefore, safety measures like medical disclaimers are critical to remind users that AI outputs are not professionally vetted or a substitute for medical advice. This study evaluated the presence of disclaimers in LLM and VLM outputs across model generations from 2022 to 2025. Using 500 mammograms, 500 chest X-rays, 500 dermatology images, and 500 medical questions, outputs were screened for disclaimer phrases. Medical disclaimer presence in LLM and VLM outputs dropped from 26.3% in 2022 to 0.97% in 2025, and from 19.6% in 2023 to 1.05% in 2025, respectively. By 2025, the majority of models displayed no disclaimers. As public models become more capable and authoritative, disclaimers must be implemented as a safeguard adapting to the clinical context of each output.

Paperid: 7032, https://arxiv.org/pdf/2507.07731.pdf

Abstract:
Mitigating object hallucination in large vision-language models (LVLMs) is critical to their safe deployment. Existing methods either are restricted to specific decoding methods, or demand sophisticated modifications to visual inputs, or rely on knowledge from external models. In this work, we first reveal the phenomenon that VLMs exhibit significant imbalance in the ``Yes'' ratio ( \ie, the fraction of ``Yes'' answers among the total number of questions) across three different visual question answering (VQA) datasets. Furthermore, we propose an energy-based decoding method, which dynamically selects the hidden states from the layer with minimal energy score. It is simple yet effective in reducing the bias for the yes ratio while boosting performance across three benchmarks (POPE, MME, and MMVP). Our method consistently improves accuracy and F1 score on three VQA datasets across three commonly used VLMs over several baseline methods. The average accuracy improvement is 4.82% compared to greedy decoding. Moreover, the average yes-ratio gap reduction is 8.81%, meaning the proposed method is less biased as shown in Figure 1.

Paperid: 7033, https://arxiv.org/pdf/2507.07147.pdf

Abstract:
Recent advances in pre-trained Vision Language Models (VLM) have shown promising potential for effectively adapting to downstream tasks through prompt learning, without the need for additional annotated paired datasets. To supplement the text information in VLM trained on correlations with vision data, new approaches leveraging Large Language Models (LLM) in prompts have been proposed, enhancing robustness to unseen and diverse data. Existing methods typically extract text-based responses (i.e., descriptions) from LLM to incorporate into prompts; however, this approach suffers from high variability and low reliability. In this work, we propose Description-free Multi-prompt Learning(DeMul), a novel method that eliminates the process of extracting descriptions and instead directly distills knowledge from LLM into prompts. By adopting a description-free approach, prompts can encapsulate richer semantics while still being represented as continuous vectors for optimization, thereby eliminating the need for discrete pre-defined templates. Additionally, in a multi-prompt setting, we empirically demonstrate the potential of prompt weighting in reflecting the importance of different prompts during training. Experimental results show that our approach achieves superior performance across 11 recognition datasets.

Paperid: 7034, https://arxiv.org/pdf/2507.06183.pdf

Abstract:
Technical reports and articles often contain valuable information in the form of semi-structured data like charts, and figures. Interpreting these and using the information from them is essential for downstream tasks such as question answering (QA). Current approaches to visual question answering often struggle with the precision required for scientific data interpretation, particularly in handling numerical values, multi-step reasoning over visual elements, and maintaining consistency between visual observation and textual reasoning. We present our approach to the SciVQA 2025 shared task, focusing on answering visual and non-visual questions grounded in scientific figures from scholarly articles. We conducted a series of experiments using models with 5B to 8B parameters. Our strongest individual model, InternVL3, achieved ROUGE-1 and ROUGE-L F1 scores of \textbf{0.740} and a BERTScore of \textbf{0.983} on the SciVQA test split. We also developed an ensemble model with multiple vision language models (VLMs). Through error analysis on the validation split, our ensemble approach improved performance compared to most individual models, though InternVL3 remained the strongest standalone performer. Our findings underscore the effectiveness of prompt optimization, chain-of-thought reasoning and ensemble modeling in improving the model's ability in visual question answering.

Paperid: 7035, https://arxiv.org/pdf/2507.05607.pdf

Abstract:
This paper presents Auto-RubikAI, a modular autonomous planning framework that integrates a symbolic Knowledge Base (KB), a vision-language model (VLM), and a large language model (LLM) to solve structured manipulation tasks exemplified by Rubik's Cube restoration. Unlike traditional robot systems based on predefined scripts, or modern approaches relying on pretrained networks and large-scale demonstration data, Auto-RubikAI enables interpretable, multi-step task execution with minimal data requirements and no prior demonstrations. The proposed system employs a KB module to solve group-theoretic restoration steps, overcoming LLMs' limitations in symbolic reasoning. A VLM parses RGB-D input to construct a semantic 3D scene representation, while the LLM generates structured robotic control code via prompt chaining. This tri-module architecture enables robust performance under spatial uncertainty. We deploy Auto-RubikAI in both simulation and real-world settings using a 7-DOF robotic arm, demonstrating effective Sim-to-Real adaptation without retraining. Experiments show a 79% end-to-end task success rate across randomized configurations. Compared to CFOP, DeepCubeA, and Two-Phase baselines, our KB-enhanced method reduces average solution steps while maintaining interpretability and safety. Auto-RubikAI provides a cost-efficient, modular foundation for embodied task planning in smart manufacturing, robotics education, and autonomous execution scenarios. Code, prompts, and hardware modules will be released upon publication.

Paperid: 7036, https://arxiv.org/pdf/2507.05520.pdf

Abstract:
Dermatological care via telemedicine often lacks the rich context of in-person visits. Clinicians must make diagnoses based on a handful of images and brief descriptions, without the benefit of physical exams, second opinions, or reference materials. While many medical AI systems attempt to bridge these gaps with domain-specific fine-tuning, this work hypothesized that mimicking clinical reasoning processes could offer a more effective path forward. This study tested seven vision-language models on medical visual question answering across six configurations: baseline models, fine-tuned variants, and both augmented with either reasoning layers that combine multiple model perspectives, analogous to peer consultation, or retrieval-augmented generation that incorporates medical literature at inference time, serving a role similar to reference-checking. While fine-tuning degraded performance in four of seven models with an average 30% decrease, baseline models collapsed on test data. Clinical-inspired architectures, meanwhile, achieved up to 70% accuracy, maintaining performance on unseen data while generating explainable, literature-grounded outputs critical for clinical adoption. These findings demonstrate that medical AI succeeds by reconstructing the collaborative and evidence-based practices fundamental to clinical diagnosis.

Paperid: 7037, https://arxiv.org/pdf/2507.05321.pdf

Abstract:
Recent advances in AI-assisted education have encouraged the integration of vision-language models (VLMs) into academic assessment, particularly for tasks that require both quantitative and qualitative evaluation. However, existing VLM based approaches struggle with complex educational artifacts, such as programming tasks with executable components and measurable outputs, that require structured reasoning and alignment with clearly defined evaluation criteria. We introduce AGACCI, a multi-agent system that distributes specialized evaluation roles across collaborative agents to improve accuracy, interpretability, and consistency in code-oriented assessment. To evaluate the framework, we collected 360 graduate-level code-based assignments from 60 participants, each annotated by domain experts with binary rubric scores and qualitative feedback. Experimental results demonstrate that AGACCI outperforms a single GPT-based baseline in terms of rubric and feedback accuracy, relevance, consistency, and coherence, while preserving the instructional intent and evaluative depth of expert assessments. Although performance varies across task types, AGACCI highlights the potential of multi-agent systems for scalable and context-aware educational evaluation.

Paperid: 7038, https://arxiv.org/pdf/2507.03875.pdf

Abstract:
The introduction of ChatGPT has garnered significant attention within the NLP community and beyond. Previous studies have demonstrated ChatGPT's substantial advancements across various downstream NLP tasks, highlighting its adaptability and potential to revolutionize language-related applications. However, its capabilities and limitations in genre prediction remain unclear. This work analyzes three Large Language Models (LLMs) using the MovieLens-100K dataset to assess their genre prediction capabilities. Our findings show that ChatGPT, without fine-tuning, outperformed other LLMs, and fine-tuned ChatGPT performed best overall. We set up zero-shot and few-shot prompts using audio transcripts/subtitles from movie trailers in the MovieLens-100K dataset, covering 1682 movies of 18 genres, where each movie can have multiple genres. Additionally, we extended our study by extracting IMDb movie posters to utilize a Vision Language Model (VLM) with prompts for poster information. This fine-grained information was used to enhance existing LLM prompts. In conclusion, our study reveals ChatGPT's remarkable genre prediction capabilities, surpassing other language models. The integration of VLM further enhances our findings, showcasing ChatGPT's potential for content-related applications by incorporating visual information from movie posters.

Paperid: 7039, https://arxiv.org/pdf/2507.01485.pdf

Abstract:
Large language models (LLMs) and vision-language models (VLMs) have the potential to transform biological research by enabling autonomous experimentation. Yet, their application remains constrained by rigid protocol design, limited adaptability to dynamic lab conditions, inadequate error handling, and high operational complexity. Here we introduce BioMARS (Biological Multi-Agent Robotic System), an intelligent platform that integrates LLMs, VLMs, and modular robotics to autonomously design, plan, and execute biological experiments. BioMARS uses a hierarchical architecture: the Biologist Agent synthesizes protocols via retrieval-augmented generation; the Technician Agent translates them into executable robotic pseudo-code; and the Inspector Agent ensures procedural integrity through multimodal perception and anomaly detection. The system autonomously conducts cell passaging and culture tasks, matching or exceeding manual performance in viability, consistency, and morphological integrity. It also supports context-aware optimization, outperforming conventional strategies in differentiating retinal pigment epithelial cells. A web interface enables real-time human-AI collaboration, while a modular backend allows scalable integration with laboratory hardware. These results highlight the feasibility of generalizable, AI-driven laboratory automation and the transformative role of language-based reasoning in biological research.

Paperid: 7040, https://arxiv.org/pdf/2507.01284.pdf

Abstract:
Recent advancements in open-source Visual Language Models (VLMs) such as LLaVA, Qwen-VL, and Llama have catalyzed extensive research on their integration with diverse systems. The internet-scale general knowledge encapsulated within these models presents significant opportunities for enhancing autonomous driving perception, prediction, and planning capabilities. In this paper we propose VLAD, a vision-language autonomous driving model, which integrates a fine-tuned VLM with VAD, a state-of-the-art end-to-end system. We implement a specialized fine-tuning approach using custom question-answer datasets designed specifically to improve the spatial reasoning capabilities of the model. The enhanced VLM generates high-level navigational commands that VAD subsequently processes to guide vehicle operation. Additionally, our system produces interpretable natural language explanations of driving decisions, thereby increasing transparency and trustworthiness of the traditionally black-box end-to-end architecture. Comprehensive evaluation on the real-world nuScenes dataset demonstrates that our integrated system reduces average collision rates by 31.82% compared to baseline methodologies, establishing a new benchmark for VLM-augmented autonomous driving systems.

Paperid: 7041, https://arxiv.org/pdf/2506.22982.pdf

Abstract:
Large Vision-Language Models (VLMs) have revolutionized computer vision, enabling tasks such as image classification, captioning, and visual question answering. However, they remain highly vulnerable to adversarial attacks, particularly in scenarios where both visual and textual modalities can be manipulated. In this study, we conduct a comprehensive reproducibility study of "An Image is Worth 1000 Lies: Adversarial Transferability Across Prompts on Vision-Language Models" validating the Cross-Prompt Attack (CroPA) and confirming its superior cross-prompt transferability compared to existing baselines. Beyond replication we propose several key improvements: (1) A novel initialization strategy that significantly improves Attack Success Rate (ASR). (2) Investigate cross-image transferability by learning universal perturbations. (3) A novel loss function targeting vision encoder attention mechanisms to improve generalization. Our evaluation across prominent VLMs -- including Flamingo, BLIP-2, and InstructBLIP as well as extended experiments on LLaVA validates the original results and demonstrates that our improvements consistently boost adversarial effectiveness. Our work reinforces the importance of studying adversarial vulnerabilities in VLMs and provides a more robust framework for generating transferable adversarial examples, with significant implications for understanding the security of VLMs in real-world applications.

Paperid: 7042, https://arxiv.org/pdf/2506.22827.pdf

Abstract:
Enabling humanoid robots to reliably execute complex multi-step manipulation tasks is crucial for their effective deployment in industrial and household environments. This paper presents a hierarchical planning and control framework designed to achieve reliable multi-step humanoid manipulation. The proposed system comprises three layers: (1) a low-level RL-based controller responsible for tracking whole-body motion targets; (2) a mid-level set of skill policies trained via imitation learning that produce motion targets for different steps of a task; and (3) a high-level vision-language planning module that determines which skills should be executed and also monitors their completion in real-time using pretrained vision-language models (VLMs). Experimental validation is performed on a Unitree G1 humanoid robot executing a non-prehensile pick-and-place task. Over 40 real-world trials, the hierarchical system achieved a 73% success rate in completing the full manipulation sequence. These experiments confirm the feasibility of the proposed hierarchical system, highlighting the benefits of VLM-based skill planning and monitoring for multi-step manipulation scenarios. See https://vlp-humanoid.github.io/ for video demonstrations of the policy rollout.

Paperid: 7043, https://arxiv.org/pdf/2506.21873.pdf

Abstract:
Recent Multimodal Large Language Models (MLLMs) have demonstrated strong performance in visual grounding, establishing themselves as a general interface for various vision-language applications. This progress has driven the development of token pruning methods to mitigate the high computational costs associated with processing numerous visual tokens. However, we observe that pruning significantly weakens the model's grounding ability, leading to incorrect predictions and drastic performance degradation. In Referring Expression Comprehension (REC), for instance, pruning causes the accuracy of LLaVA on the RefCOCO validation set to drop from 56.14% to 15.34%. Our analysis identifies misaligned position IDs after pruning as the primary cause of this degradation, as both the order and value of these IDs are crucial for maintaining performance in grounding tasks. To address this issue, we propose Grounding-Aware Token Pruning (GAP), a simple yet effective adjustment to position IDs that recovers REC accuracy back to 51.42%, which is 90% of the original performance in the without pruning setting, all while requiring no additional training, memory, or computational overhead. Applied to models such as Shikra, MiniGPTv2, and the LLaVA series, our method consistently improves performance across various token pruning strategies.

Paperid: 7044, https://arxiv.org/pdf/2506.19303.pdf

Abstract:
Inferring physical properties can significantly enhance robotic manipulation by enabling robots to handle objects safely and efficiently through adaptive grasping strategies. Previous approaches have typically relied on either tactile or visual data, limiting their ability to fully capture properties. We introduce a novel cross-modal perception framework that integrates visual observations with tactile representations within a multimodal vision-language model. Our physical reasoning framework, which employs a hierarchical feature alignment mechanism and a refined prompting strategy, enables our model to make property-specific predictions that strongly correlate with ground-truth measurements. Evaluated on 35 diverse objects, our approach outperforms existing baselines and demonstrates strong zero-shot generalization. Keywords: tactile perception, visual-tactile fusion, physical property inference, multimodal integration, robot perception

Paperid: 7045, https://arxiv.org/pdf/2506.17328.pdf

Abstract:
Desktop cleaning demands open-vocabulary recognition and precise manipulation for heterogeneous debris. We propose a hierarchical framework integrating reflective Vision-Language Model (VLM) planning with dual-arm execution via structured scene representation. Grounded-SAM2 facilitates open-vocabulary detection, while a memory-augmented VLM generates, critiques, and revises manipulation sequences. These sequences are converted into parametric trajectories for five primitives executed by coordinated Franka arms. Evaluated in simulated scenarios, our system achieving 87.2% task completion, a 28.8% improvement over static VLM and 36.2% over single-arm baselines. Structured memory integration proves crucial for robust, generalizable manipulation while maintaining real-time control performance.

Paperid: 7046, https://arxiv.org/pdf/2506.15180.pdf

Abstract:
Large-scale visual search engines are expected to solve a dual problem at once: (i) locate every image that truly contains the object described by a sentence and (ii) identify the object's bounding box or exact pixels within each hit. Existing techniques address only one side of this challenge. Visual grounding yields tight boxes and masks but rests on the unrealistic assumption that the object is present in every test image, producing a flood of false alarms when applied to web-scale collections. Text-to-image retrieval excels at sifting through massive databases to rank relevant images, yet it stops at whole-image matches and offers no fine-grained localization. We introduce Referring Search and Discovery (ReSeDis), the first task that unifies corpus-level retrieval with pixel-level grounding. Given a free-form description, a ReSeDis model must decide whether the queried object appears in each image and, if so, where it is, returning bounding boxes or segmentation masks. To enable rigorous study, we curate a benchmark in which every description maps uniquely to object instances scattered across a large, diverse corpus, eliminating unintended matches. We further design a task-specific metric that jointly scores retrieval recall and localization precision. Finally, we provide a straightforward zero-shot baseline using a frozen vision-language model, revealing significant headroom for future study. ReSeDis offers a realistic, end-to-end testbed for building the next generation of robust and scalable multimodal search systems.

Paperid: 7047, https://arxiv.org/pdf/2506.13964.pdf

Abstract:
Mammographic breast density classification is essential for cancer risk assessment but remains challenging due to subjective interpretation and inter-observer variability. This study compares multimodal and CNN-based methods for automated classification using the BI-RADS system, evaluating BioMedCLIP and ConvNeXt across three learning scenarios: zero-shot classification, linear probing with textual descriptions, and fine-tuning with numerical labels. Results show that zero-shot classification achieved modest performance, while the fine-tuned ConvNeXt model outperformed the BioMedCLIP linear probe. Although linear probing demonstrated potential with pretrained embeddings, it was less effective than full fine-tuning. These findings suggest that despite the promise of multimodal learning, CNN-based models with end-to-end fine-tuning provide stronger performance for specialized medical imaging. The study underscores the need for more detailed textual representations and domain-specific adaptations in future radiology applications.

Paperid: 7048, https://arxiv.org/pdf/2506.12733.pdf

Abstract:
Multimodal foundation models have achieved impressive progress across a wide range of vision-language tasks. However, existing approaches often adopt fixed or task-specific fusion strategies, neglecting the intrinsic variability of modality reliability and sample complexity. In this paper, we propose Modality-Aware Adaptive Fusion Scheduling (MA-AFS), a general framework that learns to dynamically modulate the contribution of each modality on a per-instance basis. MA-AFS introduces a lightweight neural scheduler that predicts modality fusion weights by integrating visual and textual entropy signals along with cross-modal agreement cues. This enables the model to adaptively emphasize more reliable modalities, especially under noisy, missing, or misaligned inputs. We formulate the fusion process as a differentiable scheduling mechanism, analyze its theoretical consistency and regularization effect, and demonstrate that it improves robustness without increasing model capacity significantly. Extensive experiments on image-text retrieval, captioning, and visual question answering show that MA-AFS achieves consistent performance gains over strong baselines such as CLIP, ALBEF, and BLIP. Moreover, MA-AFS exhibits improved robustness under modality corruption and enhanced generalization under domain shifts. Our work highlights the importance of adaptive fusion and opens a promising direction toward reliable and uncertainty-aware multimodal learning.

Paperid: 7049, https://arxiv.org/pdf/2506.12683.pdf

Abstract:
Vision-Language Models (VLMs) have rapidly advanced alongside Large Language Models (LLMs). This study evaluates the capabilities of prominent generative VLMs, such as GPT-4.1 and Gemini 2.5 Pro, accessed via APIs, for histopathology image classification tasks, including cell typing. Using diverse datasets from public and private sources, we apply zero-shot and one-shot prompting methods to assess VLM performance, comparing them against custom-trained Convolutional Neural Networks (CNNs). Our findings demonstrate that while one-shot prompting significantly improves VLM performance over zero-shot ($p \approx 1.005 \times 10^{-5}$ based on Kappa scores), these general-purpose VLMs currently underperform supervised CNNs on most tasks. This work underscores both the promise and limitations of applying current VLMs to specialized domains like pathology via in-context learning. All code and instructions for reproducing the study can be accessed from the repository https://www.github.com/a12dongithub/VLMCCE.

Paperid: 7050, https://arxiv.org/pdf/2506.12447.pdf

Abstract:
This paper introduces a novel approach to person identification using hand images, designed specifically for criminal investigations. The method is particularly valuable in serious crimes such as sexual abuse, where hand images are often the only identifiable evidence available. Our proposed method, CLIP-HandID, leverages a pre-trained foundational vision-language model - CLIP - to efficiently learn discriminative deep feature representations from hand images (input to CLIP's image encoder) using textual prompts as semantic guidance. Since hand images are labeled with indexes rather than text descriptions, we employ a textual inversion network to learn pseudo-tokens that encode specific visual contexts or appearance attributes. These learned pseudo-tokens are then incorporated into textual prompts, which are fed into CLIP's text encoder to leverage its multi-modal reasoning and enhance generalization for identification. Through extensive evaluations on two large, publicly available hand datasets with multi-ethnic representation, we demonstrate that our method significantly outperforms existing approaches.

Paperid: 7051, https://arxiv.org/pdf/2506.11595.pdf

Abstract:
Building on recent advances in language-based reasoning models, we explore multimodal reasoning that integrates vision and text. Existing multimodal benchmarks primarily test visual extraction combined with text-based reasoning, lacking true visual reasoning with more complex interactions between vision and language. Inspired by the ARC challenge, we introduce EasyARC, a vision-language benchmark requiring multi-image, multi-step reasoning, and self-correction. EasyARC is procedurally generated, fully verifiable, and scalable, making it ideal for reinforcement learning (RL) pipelines. The generators incorporate progressive difficulty levels, enabling structured evaluation across task types and complexities. We benchmark state-of-the-art vision-language models and analyze their failure modes. We argue that EasyARC sets a new standard for evaluating true reasoning and test-time scaling capabilities in vision-language models. We open-source our benchmark dataset and evaluation code.

Paperid: 7052, https://arxiv.org/pdf/2506.11417.pdf

Abstract:
Multimodal Large Language Models (MLLMs) frequently suffer from hallucination issues, generating information about objects that are not present in input images during vision-language tasks. These hallucinations particularly undermine model reliability in practical applications requiring accurate object identification. To address this challenge, we propose \mymethod,\ a preference learning approach that mitigates hallucinations by focusing on targeted areas where they occur. To implement this, we build a dataset containing hallucinated responses, correct responses, and target information (i.e., objects present in the images and the corresponding chunk positions in responses affected by hallucinations). By applying a preference learning method restricted to these specific targets, the model can filter out irrelevant signals and focus on correcting hallucinations. This allows the model to produce more factual responses by concentrating solely on relevant information. Experimental results demonstrate that \mymethod\ effectively reduces hallucinations across multiple vision hallucination tasks, improving the reliability and performance of MLLMs without diminishing overall performance.

Paperid: 7053, https://arxiv.org/pdf/2506.10334.pdf

Abstract:
Students' academic emotions significantly influence their social behavior and learning performance. Traditional approaches to automatically and accurately analyze these emotions have predominantly relied on supervised machine learning algorithms. However, these models often struggle to generalize across different contexts, necessitating repeated cycles of data collection, annotation, and training. The emergence of Vision-Language Models (VLMs) offers a promising alternative, enabling generalization across visual recognition tasks through zero-shot prompting without requiring fine-tuning. This study investigates the potential of VLMs to analyze students' academic emotions via facial expressions in an online learning environment. We employed two VLMs, Llama-3.2-11B-Vision-Instruct and Qwen2.5-VL-7B-Instruct, to analyze 5,000 images depicting confused, distracted, happy, neutral, and tired expressions using zero-shot prompting. Preliminary results indicate that both models demonstrate moderate performance in academic facial expression recognition, with Qwen2.5-VL-7B-Instruct outperforming Llama-3.2-11B-Vision-Instruct. Notably, both models excel in identifying students' happy emotions but fail to detect distracted behavior. Additionally, Qwen2.5-VL-7B-Instruct exhibits relatively high performance in recognizing students' confused expressions, highlighting its potential for practical applications in identifying content that causes student confusion.

Paperid: 7054, https://arxiv.org/pdf/2506.10172.pdf

Abstract:
Vision-and-Language Navigation (VLN) presents a complex challenge in embodied AI, requiring agents to interpret natural language instructions and navigate through visually rich, unfamiliar environments. Recent advances in large vision-language models (LVLMs), such as CLIP and Flamingo, have significantly improved multimodal understanding but introduced new challenges related to computational cost and real-time deployment. In this project, we propose a modular, plug-and-play navigation framework that decouples vision-language understanding from action planning. By integrating a frozen vision-language model, Qwen2.5-VL-7B-Instruct, with lightweight planning logic, we aim to achieve flexible, fast, and adaptable navigation without extensive model fine-tuning. Our framework leverages prompt engineering, structured history management, and a two-frame visual input strategy to enhance decision-making continuity across navigation steps. We evaluate our system on the Room-to-Room benchmark within the VLN-CE setting using the Matterport3D dataset and Habitat-Lab simulation environment. Although our initial results reveal challenges in generalizing to unseen environments under strict evaluation settings, our modular approach lays a foundation for scalable and efficient navigation systems, highlighting promising directions for future improvement through enhanced environmental priors and expanded multimodal input integration.

Paperid: 7055, https://arxiv.org/pdf/2506.08774.pdf

Abstract:
Different machine learning models can represent the same underlying concept in different ways. This variability is particularly valuable for in-the-wild multimodal retrieval, where the objective is to identify the corresponding representation in one modality given another modality as input. This challenge can be effectively framed as a feature alignment problem. For example, given a sentence encoded by a language model, retrieve the most semantically aligned image based on features produced by an image encoder, or vice versa. In this work, we first investigate the geometric relationships between visual and textual embeddings derived from both vision-language models and combined unimodal models. We then align these representations using four standard similarity metrics as well as two learned ones, implemented via neural networks. Our findings indicate that the Wasserstein distance can serve as an informative measure of the modality gap, while cosine similarity consistently outperforms alternative metrics in feature alignment tasks. Furthermore, we observe that conventional architectures such as multilayer perceptrons are insufficient for capturing the complex interactions between image and text representations. Our study offers novel insights and practical considerations for researchers working in multimodal information retrieval, particularly in real-world, cross-modal applications.

Paperid: 7056, https://arxiv.org/pdf/2506.07850.pdf

Abstract:
Vision-Language Models (VLMs) lag behind Large Language Models due to the scarcity of annotated datasets, as creating paired visual-textual annotations is labor-intensive and expensive. To address this bottleneck, we introduce SAM2Auto, the first fully automated annotation pipeline for video datasets requiring no human intervention or dataset-specific training. Our approach consists of two key components: SMART-OD, a robust object detection system that combines automatic mask generation with open-world object detection capabilities, and FLASH (Frame-Level Annotation and Segmentation Handler), a multi-object real-time video instance segmentation (VIS) that maintains consistent object identification across video frames even with intermittent detection gaps. Unlike existing open-world detection methods that require frame-specific hyperparameter tuning and suffer from numerous false positives, our system employs statistical approaches to minimize detection errors while ensuring consistent object tracking throughout entire video sequences. Extensive experimental validation demonstrates that SAM2Auto achieves comparable accuracy to manual annotation while dramatically reducing annotation time and eliminating labor costs. The system successfully handles diverse datasets without requiring retraining or extensive parameter adjustments, making it a practical solution for large-scale dataset creation. Our work establishes a new baseline for automated video annotation and provides a pathway for accelerating VLM development by addressing the fundamental dataset bottleneck that has constrained progress in vision-language understanding.

Paperid: 7057, https://arxiv.org/pdf/2506.06480.pdf

Abstract:
We introduce a fitness tracking system that enables remote monitoring for exercises using only a RGB smartphone camera, making fitness tracking more private, scalable, and cost effective. Although prior work explored automated exercise supervision, existing models are either too limited in exercise variety or too complex for real-world deployment. Prior approaches typically focus on a small set of exercises and fail to generalize across diverse movements. In contrast, we develop a robust, multitask motion analysis model capable of performing exercise detection and repetition counting across hundreds of exercises, a scale far beyond previous methods. We overcome previous data limitations by assembling a large-scale fitness dataset, Olympia covering more than 1,900 exercises. To our knowledge, our vision-language model is the first that can perform multiple tasks on skeletal fitness data. On Olympia, our model can detect exercises with 76.5% accuracy and count repetitions with 85.3% off-by-one accuracy, using only RGB video. By presenting a single vision-language transformer model for both exercise identification and rep counting, we take a significant step toward democratizing AI-powered fitness tracking.

Paperid: 7058, https://arxiv.org/pdf/2506.05061.pdf

Abstract:
Vietnamese document analysis and recognition (DAR) is a crucial field with applications in digitization, information retrieval, and automation. Despite advancements in OCR and NLP, Vietnamese text recognition faces unique challenges due to its complex diacritics, tonal variations, and lack of large-scale annotated datasets. Traditional OCR methods often struggle with real-world document variations, while deep learning approaches have shown promise but remain limited by data scarcity and generalization issues. Recently, large language models (LLMs) and vision-language models have demonstrated remarkable improvements in text recognition and document understanding, offering a new direction for Vietnamese DAR. However, challenges such as domain adaptation, multimodal learning, and computational efficiency persist. This survey provide a comprehensive review of existing techniques in Vietnamese document recognition, highlights key limitations, and explores how LLMs can revolutionize the field. We discuss future research directions, including dataset development, model optimization, and the integration of multimodal approaches for improved document intelligence. By addressing these gaps, we aim to foster advancements in Vietnamese DAR and encourage community-driven solutions.

Paperid: 7059, https://arxiv.org/pdf/2506.02708.pdf

Abstract:
Image scoring is a crucial task in numerous real-world applications. To trust a model's judgment, understanding its rationale is essential. This paper proposes a novel training method for Vision Language Models (VLMs) to generate not only image scores but also corresponding justifications in natural language. Leveraging only an image scoring dataset and an instruction-tuned VLM, our method enables self-training, utilizing the VLM's generated text without relying on external data or models. In addition, we introduce a simple method for creating a dataset designed to improve alignment between predicted scores and their textual justifications. By iteratively training the model with Direct Preference Optimization on two distinct datasets and merging them, we can improve both scoring accuracy and the coherence of generated explanations.

Paperid: 7060, https://arxiv.org/pdf/2506.02615.pdf

Abstract:
In this paper, we present a hierarchical question-answering (QA) approach for scene understanding in autonomous vehicles, balancing cost-efficiency with detailed visual interpretation. The method fine-tunes a compact vision-language model (VLM) on a custom dataset specific to the geographical area in which the vehicle operates to capture key driving-related visual elements. At the inference stage, the hierarchical QA strategy decomposes the scene understanding task into high-level and detailed sub-questions. Instead of generating lengthy descriptions, the VLM navigates a structured question tree, where answering high-level questions (e.g., "Is it possible for the ego vehicle to turn left at the intersection?") triggers more detailed sub-questions (e.g., "Is there a vehicle approaching the intersection from the opposite direction?"). To optimize inference time, questions are dynamically skipped based on previous answers, minimizing computational overhead. The extracted answers are then synthesized using handcrafted templates to ensure coherent, contextually accurate scene descriptions. We evaluate the proposed approach on the custom dataset using GPT reference-free scoring, demonstrating its competitiveness with state-of-the-art methods like GPT-4o in capturing key scene details while achieving significantly lower inference time. Moreover, qualitative results from real-time deployment highlight the proposed approach's capacity to capture key driving elements with minimal latency.

Paperid: 7061, https://arxiv.org/pdf/2505.24371.pdf

Abstract:
In this paper, we propose a Grid-based Local and Global Area Transcription (Grid-LoGAT) system for Video Question Answering (VideoQA). The system operates in two phases. First, extracting text transcripts from video frames using a Vision-Language Model (VLM). Next, processing questions using these transcripts to generate answers through a Large Language Model (LLM). This design ensures image privacy by deploying the VLM on edge devices and the LLM in the cloud. To improve transcript quality, we propose grid-based visual prompting, which extracts intricate local details from each grid cell and integrates them with global information. Evaluation results show that Grid-LoGAT, using the open-source VLM (LLaVA-1.6-7B) and LLM (Llama-3.1-8B), outperforms state-of-the-art methods with similar baseline models on NExT-QA and STAR-QA datasets with an accuracy of 65.9% and 50.11% respectively. Additionally, our method surpasses the non-grid version by 24 points on localization-based questions we created using NExT-QA. (This paper is accepted by IEEE ICIP 2025.)

Paperid: 7062, https://arxiv.org/pdf/2505.23358.pdf

Abstract:
Generating informative and knowledge-rich image captions remains a challenge for many existing captioning models, which often produce generic descriptions that lack specificity and contextual depth. To address this limitation, we propose KRCapVLM, a knowledge replay-based novel image captioning framework using vision-language model. We incorporate beam search decoding to generate more diverse and coherent captions. We also integrate attention-based modules into the image encoder to enhance feature representation. Finally, we employ training schedulers to improve stability and ensure smoother convergence during training. These proposals accelerate substantial gains in both caption quality and knowledge recognition. Our proposed model demonstrates clear improvements in both the accuracy of knowledge recognition and the overall quality of generated captions. It shows a stronger ability to generalize to previously unseen knowledge concepts, producing more informative and contextually relevant descriptions. These results indicate the effectiveness of our approach in enhancing the model's capacity to generate meaningful, knowledge-grounded captions across a range of scenarios.

Paperid: 7063, https://arxiv.org/pdf/2505.22079.pdf

Abstract:
The development of large-scale image-text pair datasets has significantly advanced self-supervised learning in Vision-Language Processing (VLP). However, directly applying general-domain architectures such as CLIP to medical data presents challenges, particularly in handling negations and addressing the inherent data imbalance of medical datasets. To address these issues, we propose a novel approach that integrates clinically-enhanced dynamic soft labels and medical graphical alignment, thereby improving clinical comprehension and the applicability of contrastive loss in medical contexts. Furthermore, we introduce negation-based hard negatives to deepen the model's understanding of the complexities of clinical language. Our approach is easily integrated into the medical CLIP training pipeline and achieves state-of-the-art performance across multiple tasks, including zero-shot, fine-tuned classification, and report retrieval. To comprehensively evaluate our model's capacity for understanding clinical language, we introduce CXR-Align, a benchmark uniquely designed to evaluate the understanding of negation and clinical information within chest X-ray (CXR) datasets. Experimental results demonstrate that our proposed methods are straightforward to implement and generalize effectively across contrastive learning frameworks, enhancing medical VLP capabilities and advancing clinical language understanding in medical imaging.

Paperid: 7064, https://arxiv.org/pdf/2505.20032.pdf

Abstract:
Tactile sensing provides local essential information that is complementary to visual perception, such as texture, compliance, and force. Despite recent advances in visuotactile representation learning, challenges remain in fusing these modalities and generalizing across tasks and environments without heavy reliance on pre-trained vision-language models. Moreover, existing methods do not study positional encodings, thereby overlooking the multi-scale spatial reasoning needed to capture fine-grained visuotactile correlations. We introduce ViTaPEs, a transformer-based framework that robustly integrates visual and tactile input data to learn task-agnostic representations for visuotactile perception. Our approach exploits a novel multi-scale positional encoding scheme to capture intra-modal structures, while simultaneously modeling cross-modal cues. Unlike prior work, we provide provable guarantees in visuotactile fusion, showing that our encodings are injective, rigid-motion-equivariant, and information-preserving, validating these properties empirically. Experiments on multiple large-scale real-world datasets show that ViTaPEs not only surpasses state-of-the-art baselines across various recognition tasks but also demonstrates zero-shot generalization to unseen, out-of-domain scenarios. We further demonstrate the transfer-learning strength of ViTaPEs in a robotic grasping task, where it outperforms state-of-the-art baselines in predicting grasp success. Project page: https://sites.google.com/view/vitapes

Paperid: 7065, https://arxiv.org/pdf/2505.19358.pdf

Abstract:
Natural disasters are increasing in frequency and severity, causing hundreds of billions of dollars in damage annually and posing growing threats to infrastructure and human livelihoods. Accurate data on roofing materials is critical for modeling building vulnerability to natural hazards such as earthquakes, floods, wildfires, and hurricanes, yet such data remain unavailable. To address this gap, we introduce RoofNet, the largest and most geographically diverse novel multimodal dataset to date, comprising over 51,500 samples from 184 geographically diverse sites pairing high-resolution Earth Observation (EO) imagery with curated text annotations for global roof material classification. RoofNet includes geographically diverse satellite imagery labeled with 14 key roofing types -- such as asphalt shingles, clay tiles, and metal sheets -- and is designed to enhance the fidelity of global exposure datasets through vision-language modeling (VLM). We sample EO tiles from climatically and architecturally distinct regions to construct a representative dataset. A subset of 6,000 images was annotated in collaboration with domain experts to fine-tune a VLM. We used geographic- and material-aware prompt tuning to enhance class separability. The fine-tuned model was then applied to the remaining EO tiles, with predictions refined through rule-based and human-in-the-loop verification. In addition to material labels, RoofNet provides rich metadata including roof shape, footprint area, solar panel presence, and indicators of mixed roofing materials (e.g., HVAC systems). RoofNet supports scalable, AI-driven risk assessment and serves as a downstream benchmark for evaluating model generalization across regions -- offering actionable insights for insurance underwriting, disaster preparedness, and infrastructure policy planning.

Paperid: 7066, https://arxiv.org/pdf/2505.19242.pdf

Abstract:
Image segmentation is a fundamental task in computer vision, aimed at partitioning an image into semantically meaningful regions. Referring image segmentation extends this task by using natural language expressions to localize specific objects, requiring effective integration of visual and linguistic information. In this work, we propose SegVLM, a vision-language model that incorporates architectural improvements to enhance segmentation accuracy and cross-modal alignment. The model integrates squeeze-and-excitation (SE) blocks for dynamic feature recalibration, deformable convolutions for geometric adaptability, and residual connections for deep feature learning. We also introduce a novel referring-aware fusion (RAF) loss that balances region-level alignment, boundary precision, and class imbalance. Extensive experiments and ablation studies demonstrate that each component contributes to consistent performance improvements. SegVLM also shows strong generalization across diverse datasets and referring expression scenarios.

Paperid: 7067, https://arxiv.org/pdf/2505.18039.pdf

Abstract:
Foundation models like CLIP (Contrastive Language-Image Pretraining) have revolutionized vision-language tasks by enabling zero-shot and few-shot learning through cross-modal alignment. However, their computational complexity and large memory footprint make them unsuitable for deployment on resource-constrained edge devices, such as in-car cameras used for image collection and real-time processing. To address this challenge, we propose Clip4Retrofit, an efficient model distillation framework that enables real-time image labeling on edge devices. The framework is deployed on the Retrofit camera, a cost-effective edge device retrofitted into thousands of vehicles, despite strict limitations on compute performance and memory. Our approach distills the knowledge of the CLIP model into a lightweight student model, combining EfficientNet-B3 with multi-layer perceptron (MLP) projection heads to preserve cross-modal alignment while significantly reducing computational requirements. We demonstrate that our distilled model achieves a balance between efficiency and performance, making it ideal for deployment in real-world scenarios. Experimental results show that Clip4Retrofit can perform real-time image labeling and object identification on edge devices with limited resources, offering a practical solution for applications such as autonomous driving and retrofitting existing systems. This work bridges the gap between state-of-the-art vision-language models and their deployment in resource-constrained environments, paving the way for broader adoption of foundation models in edge computing.

Paperid: 7068, https://arxiv.org/pdf/2505.17835.pdf

Abstract:
The task of grading atopic dermatitis (or AD, a form of eczema) from patient images is difficult even for trained dermatologists. Research on automating this task has progressed in recent years with the development of deep learning solutions; however, the rapid evolution of multimodal models and more specifically vision-language models (VLMs) opens the door to new possibilities in terms of explainable assessment of medical images, including dermatology. This report describes experiments carried out to evaluate the ability of seven VLMs to assess the severity of AD on a set of test images.

Paperid: 7069, https://arxiv.org/pdf/2505.17042.pdf

Abstract:
Vision-Language Models (VLMs) have demonstrated remarkable success in natural language generation, excelling at instruction following and structured output generation. Knowledge graphs play a crucial role in radiology, serving as valuable sources of factual information and enhancing various downstream tasks. However, generating radiology-specific knowledge graphs presents significant challenges due to the specialized language of radiology reports and the limited availability of domain-specific data. Existing solutions are predominantly unimodal, meaning they generate knowledge graphs only from radiology reports while excluding radiographic images. Additionally, they struggle with long-form radiology data due to limited context length. To address these limitations, we propose a novel multimodal VLM-based framework for knowledge graph generation in radiology. Our approach outperforms previous methods and introduces the first multimodal solution for radiology knowledge graph generation.

Paperid: 7070, https://arxiv.org/pdf/2505.14726.pdf

Abstract:
Medical image captioning is a challenging task that requires generating clinically accurate and semantically meaningful descriptions of radiology images. While recent vision-language models (VLMs) such as BLIP, BLIP2, Gemini and ViT-GPT2 show strong performance on natural image datasets, they often produce generic or imprecise captions when applied to specialized medical domains. In this project, we explore the effectiveness of fine-tuning the BLIP model on the ROCO dataset for improved radiology captioning. We compare the fine-tuned BLIP against its zero-shot version, BLIP-2 base, BLIP-2 Instruct and a ViT-GPT2 transformer baseline. Our results demonstrate that domain-specific fine-tuning on BLIP significantly improves performance across both quantitative and qualitative evaluation metrics. We also visualize decoder cross-attention maps to assess interpretability and conduct an ablation study to evaluate the contributions of encoder-only and decoder-only fine-tuning. Our findings highlight the importance of targeted adaptation for medical applications and suggest that decoder-only fine-tuning (encoder-frozen) offers a strong performance baseline with 5% lower training time than full fine-tuning, while full model fine-tuning still yields the best results overall.

Paperid: 7071, https://arxiv.org/pdf/2505.14707.pdf

Abstract:
We present CrypticBio, the largest publicly available multimodal dataset of visually confusing species, specifically curated to support the development of AI models in the context of biodiversity applications. Visually confusing or cryptic species are groups of two or more taxa that are nearly indistinguishable based on visual characteristics alone. While much existing work addresses taxonomic identification in a broad sense, datasets that directly address the morphological confusion of cryptic species are small, manually curated, and target only a single taxon. Thus, the challenge of identifying such subtle differences in a wide range of taxa remains unaddressed. Curated from real-world trends in species misidentification among community annotators of iNaturalist, CrypticBio contains 52K unique cryptic groups spanning 67K species, represented in 166 million images. Rich research-grade image annotations--including scientific, multicultural, and multilingual species terminology, hierarchical taxonomy, spatiotemporal context, and associated cryptic groups--address multimodal AI in biodiversity research. For easy dataset curation, we provide an open-source pipeline CrypticBio-Curate. The multimodal nature of the dataset beyond vision-language arises from the integration of geographical and temporal data as complementary cues to identifying cryptic species. To highlight the importance of the dataset, we benchmark a suite of state-of-the-art foundation models across CrypticBio subsets of common, unseen, endangered, and invasive species, and demonstrate the substantial impact of geographical context on vision-language zero-shot learning for cryptic species. By introducing CrypticBio, we aim to catalyze progress toward real-world-ready biodiversity AI models capable of handling the nuanced challenges of species ambiguity.

Paperid: 7072, https://arxiv.org/pdf/2505.14627.pdf

Abstract:
As Large Language Models (LLMs) gain expertise across diverse domains and modalities, scalable oversight becomes increasingly challenging, particularly when their capabilities may surpass human evaluators. Debate has emerged as a promising mechanism for enabling such oversight. In this work, we extend the debate paradigm to a multimodal setting, exploring its potential for weaker models to supervise and enhance the performance of stronger models. We focus on visual question answering (VQA), where two "sighted" expert vision-language models debate an answer, while a "blind" (text-only) judge adjudicates based solely on the quality of the arguments. In our framework, the experts defend only answers aligned with their beliefs, thereby obviating the need for explicit role-playing and concentrating the debate on instances of expert disagreement. Experiments on several multimodal tasks demonstrate that the debate framework consistently outperforms individual expert models. Moreover, judgments from weaker LLMs can help instill reasoning capabilities in vision-language models through finetuning.

Paperid: 7073, https://arxiv.org/pdf/2505.11804.pdf

Abstract:
Robustness against uncertain and ambiguous inputs is a critical challenge for deep learning models. While recent advancements in large scale vision language models (VLMs, e.g. GPT4o) might suggest that increasing model and training dataset size would mitigate this issue, our empirical evaluation shows a more complicated picture. Testing models using two classic uncertainty quantification tasks, anomaly detection and classification under inherently ambiguous conditions, we find that newer and larger VLMs indeed exhibit improved robustness compared to earlier models, but still suffer from a tendency to strictly follow instructions, often causing them to hallucinate confident responses even when faced with unclear or anomalous inputs. Remarkably, for natural images such as ImageNet, this limitation can be overcome without pipeline modifications: simply prompting models to abstain from uncertain predictions enables significant reliability gains, achieving near-perfect robustness in several settings. However, for domain-specific tasks such as galaxy morphology classification, a lack of specialized knowledge prevents reliable uncertainty estimation. Finally, we propose a novel mechanism based on caption diversity to reveal a model's internal uncertainty, enabling practitioners to predict when models will successfully abstain without relying on labeled data.

Paperid: 7074, https://arxiv.org/pdf/2505.10533.pdf

Abstract:
Large multimodal models (LMMs) have achieved high performance in vision-language tasks involving single image but they struggle when presented with a collection of multiple images (Multiple Image Question Answering scenario). These tasks, which involve reasoning over large number of images, present issues in scalability (with increasing number of images) and retrieval performance. In this work, we propose an enhancement for retriever framework introduced in MIRAGE model using submodular subset selection techniques. Our method leverages query-aware submodular functions, such as GraphCut, to pre-select a subset of semantically relevant images before main retrieval component. We demonstrate that using anchor-based queries and augmenting the data improves submodular-retriever pipeline effectiveness, particularly in large haystack sizes.

Paperid: 7075, https://arxiv.org/pdf/2505.08798.pdf

Abstract:
The application of AI in oncology has been limited by its reliance on large, annotated datasets and the need for retraining models for domain-specific diagnostic tasks. Taking heed of these limitations, we investigated in-context learning as a pragmatic alternative to model retraining by allowing models to adapt to new diagnostic tasks using only a few labeled examples at inference, without the need for retraining. Using four vision-language models (VLMs)-Paligemma, CLIP, ALIGN and GPT-4o, we evaluated the performance across three oncology datasets: MHIST, PatchCamelyon and HAM10000. To the best of our knowledge, this is the first study to compare the performance of multiple VLMs on different oncology classification tasks. Without any parameter updates, all models showed significant gains with few-shot prompting, with GPT-4o reaching an F1 score of 0.81 in binary classification and 0.60 in multi-class classification settings. While these results remain below the ceiling of fully fine-tuned systems, they highlight the potential of ICL to approximate task-specific behavior using only a handful of examples, reflecting how clinicians often reason from prior cases. Notably, open-source models like Paligemma and CLIP demonstrated competitive gains despite their smaller size, suggesting feasibility for deployment in computing constrained clinical environments. Overall, these findings highlight the potential of ICL as a practical solution in oncology, particularly for rare cancers and resource-limited contexts where fine-tuning is infeasible and annotated data is difficult to obtain.

Paperid: 7076, https://arxiv.org/pdf/2505.07637.pdf

Abstract:
Human cognition is deeply intertwined with a sense of time, known as Chronoception. This sense allows us to judge how long facts remain valid and when knowledge becomes outdated. Despite progress in vision, language, and motor control, AI still struggles to reason about temporal validity. We introduce Chronocept, the first benchmark to model temporal validity as a continuous probability distribution over time. Using skew-normal curves fitted along semantically decomposed temporal axes, Chronocept captures nuanced patterns of emergence, decay, and peak relevance. It includes two datasets: Benchmark I (atomic facts) and Benchmark II (multi-sentence passages). Annotations show strong inter-annotator agreement (84% and 89%). Our baselines predict curve parameters - location, scale, and skewness - enabling interpretable, generalizable learning and outperforming classification-based approaches. Chronocept fills a foundational gap in AI's temporal reasoning, supporting applications in knowledge grounding, fact-checking, retrieval-augmented generation (RAG), and proactive agents. Code and data are publicly available.

Paperid: 7077, https://arxiv.org/pdf/2505.05895.pdf

Abstract:
Modern automotive infotainment systems necessitate intelligent and adaptive solutions to manage frequent User Interface (UI) updates and diverse design variations. This work introduces a vision-language framework to facilitate the understanding of and interaction with automotive UIs, enabling seamless adaptation across different UI designs. To support research in this field, AutomotiveUI-Bench-4K, an open-source dataset comprising 998 images with 4,208 annotations, is also released. Additionally, a data pipeline for generating training data is presented. A Molmo-7B-based model is fine-tuned using Low-Rank Adaptation (LoRa), incorporating generated reasoning along with visual grounding and evaluation capabilities. The fine-tuned Evaluative Large Action Model (ELAM) achieves strong performance on AutomotiveUI-Bench-4K (model and dataset are available on Hugging Face). The approach demonstrates strong cross-domain generalization, including a +5.6% improvement on ScreenSpot over the baseline model. An average accuracy of 80.8% is achieved on ScreenSpot, closely matching or surpassing specialized models for desktop, mobile, and web, despite being trained primarily on the automotive domain. This research investigates how data collection and subsequent fine-tuning can lead to AI-driven advancements in automotive UI understanding and interaction. The applied method is cost-efficient, and fine-tuned models can be deployed on consumer-grade GPUs.

Paperid: 7078, https://arxiv.org/pdf/2505.05681.pdf

Abstract:
Video recordings of nonhuman primates in their natural habitat are a common source for studying their behavior in the wild. We fine-tune pre-trained video-text foundational models for the specific domain of capuchin monkeys, with the goal of developing useful computational models to help researchers to retrieve useful clips from videos. We focus on the challenging problem of training a model based solely on raw, unlabeled video footage, using weak audio descriptions sometimes provided by field collaborators. We leverage recent advances in Multimodal Large Language Models (MLLMs) and Vision-Language Models (VLMs) to address the extremely noisy nature of both video and audio content. Specifically, we propose a two-folded approach: an agentic data treatment pipeline and a fine-tuning process. The data processing pipeline automatically extracts clean and semantically aligned video-text pairs from the raw videos, which are subsequently used to fine-tune a pre-trained Microsoft's X-CLIP model through Low-Rank Adaptation (LoRA). We obtained an uplift in $Hits@5$ of $167\%$ for the 16 frames model and an uplift of $114\%$ for the 8 frame model on our domain data. Moreover, based on $NDCG@K$ results, our model is able to rank well most of the considered behaviors, while the tested raw pre-trained models are not able to rank them at all. The code will be made available upon acceptance.

Paperid: 7079, https://arxiv.org/pdf/2505.05098.pdf

Abstract:
End-to-end autonomous driving has advanced significantly, offering benefits such as system simplicity and stronger driving performance in both open-loop and closed-loop settings than conventional pipelines. However, existing frameworks still suffer from low success rates in closed-loop evaluations, highlighting their limitations in real-world deployment. In this paper, we introduce X-Driver, a unified multi-modal large language models(MLLMs) framework designed for closed-loop autonomous driving, leveraging Chain-of-Thought(CoT) and autoregressive modeling to enhance perception and decision-making. We validate X-Driver across multiple autonomous driving tasks using public benchmarks in CARLA simulation environment, including Bench2Drive[6]. Our experimental results demonstrate superior closed-loop performance, surpassing the current state-of-the-art(SOTA) while improving the interpretability of driving decisions. These findings underscore the importance of structured reasoning in end-to-end driving and establish X-Driver as a strong baseline for future research in closed-loop autonomous driving.

Paperid: 7080, https://arxiv.org/pdf/2505.03703.pdf

Abstract:
Vision-language models (VLMs) allow to embed texts and images in a shared representation space. However, it has been shown that these models are subject to a modality gap phenomenon meaning there exists a clear separation between the embeddings from one modality and another in the embedding space. While this misalignment is detrimental for downstream tasks such as multimodal retrieval, multimodal clustering or zero-shot classification, etc. no generic and practical methods have so far been proposed to assess it precisely and even reduce it. We therefore propose novel measures and effective techniques (spectral- and optimal transport-based methods) to achieve this goal. Extensive experiments conducted on several image-text datasets and models demonstrate their effectiveness and beneficial effects on downstream tasks. Our code is available at the URL provided in the paper's abstract.

Paperid: 7081, https://arxiv.org/pdf/2505.03350.pdf

Abstract:
Accurate classification of focal liver lesions is crucial for diagnosis and treatment in hepatology. However, traditional supervised deep learning models depend on large-scale annotated datasets, which are often limited in medical imaging. Recently, Vision-Language models (VLMs) such as Contrastive Language-Image Pre-training model (CLIP) has been applied to image classifications. Compared to the conventional convolutional neural network (CNN), which classifiers image based on visual information only, VLM leverages multimodal learning with text and images, allowing it to learn effectively even with a limited amount of labeled data. Inspired by CLIP, we pro-pose a Liver-VLM, a model specifically designed for focal liver lesions (FLLs) classification. First, Liver-VLM incorporates class information into the text encoder without introducing additional inference overhead. Second, by calculating the pairwise cosine similarities between image and text embeddings and optimizing the model with a cross-entropy loss, Liver-VLM ef-fectively aligns image features with class-level text features. Experimental results on MPCT-FLLs dataset demonstrate that the Liver-VLM model out-performs both the standard CLIP and MedCLIP models in terms of accuracy and Area Under the Curve (AUC). Further analysis shows that using a lightweight ResNet18 backbone enhances classification performance, particularly under data-constrained conditions.

Paperid: 7082, https://arxiv.org/pdf/2505.03174.pdf

Abstract:
Instruction-Action (IA) data pairs are valuable for training robotic systems, especially autonomous vehicles (AVs), but having humans manually annotate this data is costly and time-inefficient. This paper explores the potential of using mobile application Global Positioning System (GPS) references and Natural Language Processing (NLP) to automatically generate large volumes of IA commands and responses without having a human generate or retroactively tag the data. In our pilot data collection, by driving to various destinations and collecting voice instructions from GPS applications, we demonstrate a means to collect and categorize the diverse sets of instructions, further accompanied by video data to form complete vision-language-action triads. We provide details on our completely automated data collection prototype system, ADVLAT-Engine. We characterize collected GPS voice instructions into eight different classifications, highlighting the breadth of commands and referentialities available for curation from freely available mobile applications. Through research and exploration into the automation of IA data pairs using GPS references, the potential to increase the speed and volume at which high-quality IA datasets are created, while minimizing cost, can pave the way for robust vision-language-action (VLA) models to serve tasks in vision-language navigation (VLN) and human-interactive autonomous systems.

Paperid: 7083, https://arxiv.org/pdf/2504.20648.pdf

Abstract:
Vision-language models (VLMs) work well in tasks ranging from image captioning to visual question answering (VQA), yet they struggle with spatial reasoning, a key skill for understanding our physical world that humans excel at. We find that spatial relations are generally rare in widely used VL datasets, with only a few being well represented, while most form a long tail of underrepresented relations. This gap leaves VLMs ill-equipped to handle diverse spatial relationships. To bridge it, we construct a synthetic VQA dataset focused on spatial reasoning generated from hyper-detailed image descriptions in Localized Narratives, DOCCI, and PixMo-Cap. Our dataset consists of 455k samples containing 3.4 million QA pairs. Trained on this dataset, our Spatial-Reasoning Enhanced (SpaRE) VLMs show strong improvements on spatial reasoning benchmarks, achieving up to a 49% performance gain on the What's Up benchmark, while maintaining strong results on general tasks. Our work narrows the gap between human and VLM spatial reasoning and makes VLMs more capable in real-world tasks such as robotics and navigation.

Paperid: 7084, https://arxiv.org/pdf/2504.19739.pdf

Abstract:
In this paper, we introduce AffectVLM, a vision-language model designed to integrate multiviews for a semantically rich and visually comprehensive understanding of facial emotions from 3D/4D data. To effectively capture visual features, we propose a joint representation learning framework paired with a novel gradient-friendly loss function that accelerates model convergence towards optimal feature representation. Additionally, we introduce augmented textual prompts to enhance the model's linguistic capabilities and employ mixed view augmentation to expand the visual dataset. We also develop a Streamlit app for a real-time interactive inference and enable the model for distributed learning. Extensive experiments validate the superior performance of AffectVLM across multiple benchmarks.

Paperid: 7085, https://arxiv.org/pdf/2504.17671.pdf

Abstract:
This study addresses the critical challenge of hallucination mitigation in Large Vision-Language Models (LVLMs) for Visual Question Answering (VQA) tasks through a Split Conformal Prediction (SCP) framework. While LVLMs excel in multi-modal reasoning, their outputs often exhibit hallucinated content with high confidence, posing risks in safety-critical applications. We propose a model-agnostic uncertainty quantification method that integrates dynamic threshold calibration and cross-modal consistency verification. By partitioning data into calibration and test sets, the framework computes nonconformity scores to construct prediction sets with statistical guarantees under user-defined risk levels ($Î±$). Key innovations include: (1) rigorous control of \textbf{marginal coverage} to ensure empirical error rates remain strictly below $Î±$; (2) dynamic adjustment of prediction set sizes inversely with $Î±$, filtering low-confidence outputs; (3) elimination of prior distribution assumptions and retraining requirements. Evaluations on benchmarks (ScienceQA, MMMU) with eight LVLMs demonstrate that SCP enforces theoretical guarantees across all $Î±$ values. The framework achieves stable performance across varying calibration-to-test split ratios, underscoring its robustness for real-world deployment in healthcare, autonomous systems, and other safety-sensitive domains. This work bridges the gap between theoretical reliability and practical applicability in multi-modal AI systems, offering a scalable solution for hallucination detection and uncertainty-aware decision-making.

Paperid: 7086, https://arxiv.org/pdf/2504.16538.pdf

Abstract:
Streetscapes are an essential component of urban space. Their assessment is presently either limited to morphometric properties of their mass skeleton or requires labor-intensive qualitative evaluations of visually perceived qualities. This paper introduces SAGAI: Streetscape Analysis with Generative Artificial Intelligence, a modular workflow for scoring street-level urban scenes using open-access data and vision-language models. SAGAI integrates OpenStreetMap geometries, Google Street View imagery, and a lightweight version of the LLaVA model to generate structured spatial indicators from images via customizable natural language prompts. The pipeline includes an automated mapping module that aggregates visual scores at both the point and street levels, enabling direct cartographic interpretation. It operates without task-specific training or proprietary software dependencies, supporting scalable and interpretable analysis of urban environments. Two exploratory case studies in Nice and Vienna illustrate SAGAI's capacity to produce geospatial outputs from vision-language inference. The initial results show strong performance for binary urban-rural scene classification, moderate precision in commercial feature detection, and lower estimates, but still informative, of sidewalk width. Fully deployable by any user, SAGAI can be easily adapted to a wide range of urban research themes, such as walkability, safety, or urban design, through prompt modification alone.

Paperid: 7087, https://arxiv.org/pdf/2504.13942.pdf

Abstract:
This paper introduces Intelligence of Things (INOT), a novel spatial context-aware control system that enhances smart home automation through intuitive spatial reasoning. Current smart home systems largely rely on device-specific identifiers, limiting user interaction to explicit naming conventions rather than natural spatial references. INOT addresses this limitation through a modular architecture that integrates Vision Language Models with IoT control systems to enable natural language commands with spatial context (e.g., "turn on the light near the window"). The system comprises key components including an Onboarding Inference Engine, Zero-Shot Device Detection, Spatial Topology Inference, and Intent-Based Command Synthesis. A comprehensive user study with 15 participants demonstrated INOT's significant advantages over conventional systems like Google Home Assistant, with users reporting reduced cognitive workload (NASA-TLX scores decreased by an average of 13.17 points), higher ease-of-use ratings, and stronger preference (14 out of 15 participants). By eliminating the need to memorize device identifiers and enabling context-aware spatial commands, INOT represents a significant advancement in creating more intuitive and accessible smart home control systems.

Paperid: 7088, https://arxiv.org/pdf/2504.13690.pdf

Abstract:
Vision-language models (VLMs) have demonstrated impressive capabilities in understanding and reasoning about visual and textual content. However, their robustness to common image corruptions remains under-explored. In this work, we present the first comprehensive analysis of VLM robustness across 19 corruption types from the ImageNet-C benchmark, spanning four categories: noise, blur, weather, and digital distortions. We introduce two new benchmarks, TextVQA-C and GQA-C, to systematically evaluate how corruptions affect scene text understanding and object-based reasoning, respectively. Our analysis reveals that transformer-based VLMs exhibit distinct vulnerability patterns across tasks: text recognition deteriorates most severely under blur and snow corruptions, while object reasoning shows higher sensitivity to corruptions such as frost and impulse noise. We connect these observations to the frequency-domain characteristics of different corruptions, revealing how transformers' inherent bias toward low-frequency processing explains their differential robustness patterns. Our findings provide valuable insights for developing more corruption-robust vision-language models for real-world applications.

Paperid: 7089, https://arxiv.org/pdf/2504.13199.pdf

Abstract:
Objective: This review explores the trustworthiness of multimodal artificial intelligence (AI) systems, specifically focusing on vision-language tasks. It addresses critical challenges related to fairness, transparency, and ethical implications in these systems, providing a comparative analysis of key tasks such as Visual Question Answering (VQA), image captioning, and visual dialogue. Background: Multimodal models, particularly vision-language models, enhance artificial intelligence (AI) capabilities by integrating visual and textual data, mimicking human learning processes. Despite significant advancements, the trustworthiness of these models remains a crucial concern, particularly as AI systems increasingly confront issues regarding fairness, transparency, and ethics. Methods: This review examines research conducted from 2017 to 2024 focusing on forenamed core vision-language tasks. It employs a comparative approach to analyze these tasks through the lens of trustworthiness, underlining fairness, explainability, and ethics. This study synthesizes findings from recent literature to identify trends, challenges, and state-of-the-art solutions. Results: Several key findings were highlighted. Transparency: Explainability of vision language tasks is important for user trust. Techniques, such as attention maps and gradient-based methods, have successfully addressed this issue. Fairness: Bias mitigation in VQA and visual dialogue systems is essential for ensuring unbiased outcomes across diverse demographic groups. Ethical Implications: Addressing biases in multilingual models and ensuring ethical data handling is critical for the responsible deployment of vision-language systems. Conclusion: This study underscores the importance of integrating fairness, transparency, and ethical considerations in developing vision-language models within a unified framework.

Paperid: 7090, https://arxiv.org/pdf/2504.13119.pdf

Abstract:
Most adaptive AR storytelling systems define environmental semantics using simple object labels and spatial coordinates, limiting narratives to rigid, pre-defined logic. This oversimplification overlooks the contextual significance of object relationships-for example, a wedding ring on a nightstand might suggest marital conflict, yet is treated as just "two objects" in space. To address this, we explored integrating Vision Language Models (VLMs) into AR pipelines. However, several challenges emerged: First, stories generated with simple prompt guidance lacked narrative depth and spatial usage. Second, spatial semantics were underutilized, failing to support meaningful storytelling. Third, pre-generated scripts struggled to align with AR Foundation's object naming and coordinate systems. We propose a scene-driven AR storytelling framework that reimagines environments as active narrative agents, built on three innovations: 1. State-aware object semantics: We decompose object meaning into physical, functional, and metaphorical layers, allowing VLMs to distinguish subtle narrative cues between similar objects. 2. Structured narrative interface: A bidirectional JSON layer maps VLM-generated metaphors to AR anchors, maintaining spatial and semantic coherence. 3. STAM evaluation framework: A three-part experimental design evaluates narrative quality, highlighting both strengths and limitations of VLM-AR integration. Our findings show that the system can generate stories from the environment itself, not just place them on top of it. In user studies, 70% of participants reported seeing real-world objects differently when narratives were grounded in environmental symbolism. By merging VLMs' generative creativity with AR's spatial precision, this framework introduces a novel object-driven storytelling paradigm, transforming passive spaces into active narrative landscapes.

Paperid: 7091, https://arxiv.org/pdf/2504.12436.pdf

Abstract:
Adapting Vision-Language Models (VLMs) to new domains with few labeled samples remains a significant challenge due to severe overfitting and computational constraints. State-of-the-art solutions, such as low-rank reparameterization, mitigate these issues but often struggle with generalization and require extensive hyperparameter tuning. In this paper, a novel Sparse Optimization (SO) framework is proposed. Unlike low-rank approaches that typically constrain updates to a fixed subspace, our SO method leverages high sparsity to dynamically adjust very few parameters. We introduce two key paradigms. First, we advocate for \textit{local sparsity and global density}, which updates a minimal subset of parameters per iteration while maintaining overall model expressiveness. As a second paradigm, we advocate for \textit{local randomness and global importance}, which sparsifies the gradient using random selection while pruning the first moment based on importance. This combination significantly mitigates overfitting and ensures stable adaptation in low-data regimes. Extensive experiments on 11 diverse datasets show that SO achieves state-of-the-art few-shot adaptation performance while reducing memory overhead.

Paperid: 7092, https://arxiv.org/pdf/2504.09979.pdf

Abstract:
We propose an efficient evaluation protocol for large vision-language models (VLMs). Given their broad knowledge and reasoning capabilities, multiple benchmarks are needed for comprehensive assessment, making evaluation computationally expensive. To improve efficiency, we construct a subset that yields results comparable to full benchmark evaluations. Our benchmark classification experiments reveal that no single benchmark fully covers all challenges. We then introduce a subset construction method using farthest point sampling (FPS). Our experiments show that FPS-based benchmarks maintain a strong correlation (> 0.96) with full evaluations while using only ~1\% of the data. Additionally, applying FPS to an existing benchmark improves correlation with overall evaluation results, suggesting its potential to reduce unintended dataset biases.

Paperid: 7093, https://arxiv.org/pdf/2504.09083.pdf

Abstract:
Safety hazard identification and prevention are the key elements of proactive safety management. Previous research has extensively explored the applications of computer vision to automatically identify hazards from image clips collected from construction sites. However, these methods struggle to identify context-specific hazards, as they focus on detecting predefined individual entities without understanding their spatial relationships and interactions. Furthermore, their limited adaptability to varying construction site guidelines and conditions hinders their generalization across different projects. These limitations reduce their ability to assess hazards in complex construction environments and adaptability to unseen risks, leading to potential safety gaps. To address these challenges, we proposed and experimentally validated a Vision Language Model (VLM)-based framework for the identification of construction hazards. The framework incorporates a prompt engineering module that structures safety guidelines into contextual queries, allowing VLM to process visual information and generate hazard assessments aligned with the regulation guide. Within this framework, we evaluated state-of-the-art VLMs, including GPT-4o, Gemini, Llama 3.2, and InternVL2, using a custom dataset of 1100 construction site images. Experimental results show that GPT-4o and Gemini 1.5 Pro outperformed alternatives and displayed promising BERTScore of 0.906 and 0.888 respectively, highlighting their ability to identify both general and context-specific hazards. However, processing times remain a significant challenge, impacting real-time feasibility. These findings offer insights into the practical deployment of VLMs for construction site hazard detection, thereby contributing to the enhancement of proactive safety management.

Paperid: 7094, https://arxiv.org/pdf/2504.08219.pdf

Abstract:
Image restoration is critical for improving the quality of degraded images, which is vital for applications like autonomous driving, security surveillance, and digital content enhancement. However, existing methods are often tailored to specific degradation scenarios, limiting their adaptability to the diverse and complex challenges in real-world environments. Moreover, real-world degradations are typically non-uniform, highlighting the need for adaptive and intelligent solutions. To address these issues, we propose a novel vision-language-guided universal restoration (VL-UR) framework. VL-UR leverages a zero-shot contrastive language-image pre-training (CLIP) model to enhance image restoration by integrating visual and semantic information. A scene classifier is introduced to adapt CLIP, generating high-quality language embeddings aligned with degraded images while predicting degraded types for complex scenarios. Extensive experiments across eleven diverse degradation settings demonstrate VL-UR's state-of-the-art performance, robustness, and adaptability. This positions VL-UR as a transformative solution for modern image restoration challenges in dynamic, real-world environments.

Paperid: 7095, https://arxiv.org/pdf/2504.08154.pdf

Abstract:
Heavy-duty trucks pose significant safety challenges due to their large size and limited maneuverability compared to passenger vehicles. A deeper understanding of truck characteristics is essential for enhancing the safety perspective of cooperative autonomous driving. Traditional LiDAR-based truck classification methods rely on extensive manual annotations, which makes them labor-intensive and costly. The rapid advancement of large language models (LLMs) trained on massive datasets presents an opportunity to leverage their few-shot learning capabilities for truck classification. However, existing vision-language models (VLMs) are primarily trained on image datasets, which makes it challenging to directly process point cloud data. This study introduces a novel framework that integrates roadside LiDAR point cloud data with VLMs to facilitate efficient and accurate truck classification, which supports cooperative and safe driving environments. This study introduces three key innovations: (1) leveraging real-world LiDAR datasets for model development, (2) designing a preprocessing pipeline to adapt point cloud data for VLM input, including point cloud registration for dense 3D rendering and mathematical morphological techniques to enhance feature representation, and (3) utilizing in-context learning with few-shot prompting to enable vehicle classification with minimally labeled training data. Experimental results demonstrate encouraging performance of this method and present its potential to reduce annotation efforts while improving classification accuracy.

Paperid: 7096, https://arxiv.org/pdf/2504.06538.pdf

Abstract:
We present OPAL (Operant Physical Agent with Language), a novel vision-language-action architecture that introduces topological constraints to flow matching for robotic control. To do so, we further introduce topological attention. Our approach models action sequences as topologically-structured representations with non-trivial constraints. Experimental results across 10 complex manipulation tasks demonstrate OPAL's superior performance compared to previous approaches, including Octo, OpenVLA, and $Ï$0. Our architecture achieves significant improvements in zero-shot performance without requiring task-specific fine-tuning, while reducing inference computational requirements by 42%. The theoretical guarantees provided by our topological approach result in more coherent long-horizon action sequences. Our results highlight the potential of constraining the search space of learning problems in robotics by deriving from fundamental physical laws, and the possibility of using topological attention to embed causal understanding into transformer architectures.

Paperid: 7097, https://arxiv.org/pdf/2504.05575.pdf

Abstract:
Medical Visual Question Answering (VQA) enhances clinical decision-making by enabling systems to interpret medical images and answer clinical queries. However, developing efficient, high-performance VQA models is challenging due to the complexity of medical imagery and diverse modalities. In this paper, we introduce a lightweight, multimodal VQA model integrating BiomedCLIP for image feature extraction and LLaMA-3 for text processing. Designed for medical VQA tasks, our model achieves state-of-the-art performance on the OmniMedVQA dataset. With approximately 8 billion parameters, it requires only two NVIDIA 40 GB A100 GPUs, demonstrating superior efficiency over larger models. Our results show 73.4% accuracy for open-end questions, surpassing existing models and validating its potential for real-world medical applications. Key contributions include a specialized multimodal VQA model, a resource-efficient architecture, and strong performance in answering open-ended clinical questions.

Paperid: 7098, https://arxiv.org/pdf/2504.05445.pdf

Abstract:
Vision Language Models (VLMs) demonstrate promising chart comprehension capabilities. Yet, prior explorations of their visualization literacy have been limited to assessing their response correctness and fail to explore their internal reasoning. To address this gap, we adapted attention-guided class activation maps (AG-CAM) for VLMs, to visualize the influence and importance of input features (image and text) on model responses. Using this approach, we conducted an examination of four open-source (ChartGemma, Janus 1B and 7B, and LLaVA) and two closed-source (GPT-4o, Gemini) models comparing their performance and, for the open-source models, their AG-CAM results. Overall, we found that ChartGemma, a 3B parameter VLM fine-tuned for chart question-answering (QA), outperformed other open-source models and exhibited performance on par with significantly larger closed-source VLMs. We also found that VLMs exhibit spatial reasoning by accurately localizing key chart features, and semantic reasoning by associating visual elements with corresponding data values and query tokens. Our approach is the first to demonstrate the use of AG-CAM on early fusion VLM architectures, which are widely used, and for chart QA. We also show preliminary evidence that these results can align with human reasoning. Our promising open-source VLMs results pave the way for transparent and reproducible research in AI visualization literacy.

Paperid: 7099, https://arxiv.org/pdf/2504.01662.pdf

Abstract:
Deep-learning-based denoising methods have significantly improved Low-Dose CT (LDCT) image quality. However, existing models often over-smooth important anatomical details due to their purely data-driven attention mechanisms. To address this challenge, we propose a novel LDCT denoising framework, BioAtt. The key innovation lies in attending anatomical prior distributions extracted from the pretrained vision-language model BiomedCLIP. These priors guide the denoising model to focus on anatomically relevant regions to suppress noise while preserving clinically relevant structures. We highlight three main contributions: BioAtt outperforms baseline and attention-based models in SSIM, PSNR, and RMSE across multiple anatomical regions. The framework introduces a new architectural paradigm by embedding anatomic priors directly into spatial attention. Finally, BioAtt attention maps provide visual confirmation that the improvements stem from anatomical guidance rather than increased model complexity.

Paperid: 7100, https://arxiv.org/pdf/2503.23503.pdf

Abstract:
We present a framework for optimizing prompts in vision-language models to elicit multimodal reasoning without model retraining. Using an evolutionary algorithm to guide prompt updates downstream of visual tasks, our approach improves upon baseline prompt-updating algorithms, which lack evolution-style "survival of the fittest" iteration. Crucially, we find this approach enables the language model to independently discover progressive problem-solving techniques across several evolution generations. For example, the model reasons that to "break down" visually complex spatial tasks, making a tool call to a Python interpreter to perform tasks (such as cropping, image segmentation, or saturation changes) would improve performance significantly. Our experimentation shows that explicitly evoking this "tool calling" call, via system-level XML $...\texttt{} ... \texttt{}...$ tags, can effectively flag Python interpreter access for the same language model to generate relevant programs, generating advanced multimodal functionality. This functionality can be crystallized into a system-level prompt that induces improved performance at inference time, and our experimentation suggests up to $\approx 50\%$ relative improvement across select visual tasks. Downstream performance is trained and evaluated across subtasks from MathVista, M3CoT, and GeoBench-VLM datasets. Importantly, our approach shows that evolutionary prompt optimization guides language models towards self-reasoning discoveries, which result in improved zero-shot generalization across tasks.

Paperid: 7101, https://arxiv.org/pdf/2503.21435.pdf

Abstract:
Recent advances in Vision-Language Models (VLMs) have shown promising capabilities in interpreting visualized graph data, offering a new perspective for graph-structured reasoning beyond traditional Graph Neural Networks (GNNs). However, existing studies focus primarily on single-graph reasoning, leaving the critical challenge of multi-graph joint reasoning underexplored. In this work, we introduce the first comprehensive benchmark designed to evaluate and enhance the multi-graph reasoning abilities of VLMs. Our benchmark covers four common graph types-knowledge graphs, flowcharts, mind maps, and route maps-and supports both homogeneous and heterogeneous graph groupings with tasks of increasing complexity. We evaluate several state-of-the-art VLMs under a multi-dimensional scoring framework that assesses graph parsing, reasoning consistency, and instruction-following accuracy. Additionally, we fine-tune multiple open-source models and observe consistent improvements, confirming the effectiveness of our dataset. This work provides a principled step toward advancing multi-graph understanding and reveals new opportunities for cross-modal graph intelligence.

Paperid: 7102, https://arxiv.org/pdf/2503.21307.pdf

Abstract:
Most multimodal large language models (MLLMs) treat visual tokens as "a sequence of text", integrating them with text tokens into a large language model (LLM). However, a great quantity of visual tokens significantly increases the demand for computational resources and time. In this paper, we propose InternVL-X, which outperforms the InternVL model in both performance and efficiency by incorporating three visual token compression methods. First, we propose a novel vision-language projector, PVTC. This component integrates adjacent visual embeddings to form a local query and utilizes the transformed CLS token as a global query, then performs point-to-region cross-attention through these local and global queries to more effectively convert visual features. Second, we present a layer-wise visual token compression module, LVTC, which compresses tokens in the LLM shallow layers and then expands them through upsampling and residual connections in the deeper layers. This significantly enhances the model computational efficiency. Futhermore, we propose an efficient high resolution slicing method, RVTC, which dynamically adjusts the number of visual tokens based on image area or length filtering. RVTC greatly enhances training efficiency with only a slight reduction in performance. By utilizing 20% or fewer visual tokens, InternVL-X achieves state-of-the-art performance on 7 public MLLM benchmarks, and improves the average metric by 2.34% across 12 tasks.

Paperid: 7103, https://arxiv.org/pdf/2503.19508.pdf

Abstract:
Recent advancements in vision-language models have achieved remarkable results in making language models understand vision inputs. However, a unified approach to align these models across diverse tasks such as image captioning and visual question answering remains a challenge. Existing methods either require very big language models or very big datasets which is not efficient in utilizing existing models. This paper addresses this gap and devises a training strategy of auto-regressive vision-language models, to unify vision-language tasks like image-captioning and visual question answering. We propose four training stages for aligning the vision model with the language model, in other words, the language model is given an ability to process visual inputs. We also devise different attention masks for training transformer-based language models that improve the quality of visual features. Further, we introduce some findings, 1) the attention mask should not be applied on visual inputs, 2) the Language model converges faster on AI- generated data, 3) More work should be done in the alignment stage during the pre-training of the model, 4) the model can easily adapt to any downstream tasks like visual question answering on healthcare datasets like PathVQA. After training the model for one epoch for all the stages, it outperforms large models like VILA-13 billion models on common benchmarks like CIDEr scores on COCO and Flickr30k datasets and achieves very close scores to GIT-2 on the same dataset despite being a much smaller model trained on a much smaller dataset. All of the training is done using best practices available like multi- GPU parallel training, lower-precision training with 16-bit float numbers, faster attention (SDPA), and gradient accumulation, and completed the training within 12 hours.

Paperid: 7104, https://arxiv.org/pdf/2503.17415.pdf

Abstract:
The rapid growth of video content demands efficient and precise retrieval systems. While vision-language models (VLMs) excel in representation learning, they often struggle with adaptive, time-sensitive video retrieval. This paper introduces a novel framework that combines vector similarity search with graph-based data structures. By leveraging VLM embeddings for initial retrieval and modeling contextual relationships among video segments, our approach enables adaptive query refinement and improves retrieval accuracy. Experiments demonstrate its precision, scalability, and robustness, offering an effective solution for interactive video retrieval in dynamic environments.

Paperid: 7105, https://arxiv.org/pdf/2503.16868.pdf

Abstract:
Visual question answering (VQA) has emerged as a flexible approach for extracting specific pieces of information from document images. However, existing work typically queries each field in isolation, overlooking potential dependencies across multiple items. This paper investigates the merits of extracting multiple fields jointly versus separately. Through experiments on multiple large vision language models and datasets, we show that jointly extracting fields often improves accuracy, especially when the fields share strong numeric or contextual dependencies. We further analyze how performance scales with the number of requested items and use a regression based metric to quantify inter field relationships. Our results suggest that multi field prompts can mitigate confusion arising from similar surface forms and related numeric values, providing practical methods for designing robust VQA systems in document information extraction tasks.

Paperid: 7106, https://arxiv.org/pdf/2503.16488.pdf

Abstract:
With an increasing demand for assistive technologies that promote the independence and mobility of visually impaired people, this study suggests an innovative real-time system that gives audio descriptions of a user's surroundings to improve situational awareness. The system acquires live video input and processes it with a quantized and fine-tuned Florence-2 big model, adjusted to 4-bit accuracy for efficient operation on low-power edge devices such as the NVIDIA Jetson Orin Nano. By transforming the video signal into frames with a 5-frame latency, the model provides rapid and contextually pertinent descriptions of objects, pedestrians, and barriers, together with their estimated distances. The system employs Parler TTS Mini, a lightweight and adaptable Text-to-Speech (TTS) solution, for efficient audio feedback. It accommodates 34 distinct speaker types and enables customization of speech tone, pace, and style to suit user requirements. This study examines the quantization and fine-tuning techniques utilized to modify the Florence-2 model for this application, illustrating how the integration of a compact model architecture with a versatile TTS component improves real-time performance and user experience. The proposed system is assessed based on its accuracy, efficiency, and usefulness, providing a viable option to aid vision-impaired users in navigating their surroundings securely and successfully.

Paperid: 7107, https://arxiv.org/pdf/2503.13171.pdf

Abstract:
The acquisition of large-scale and diverse demonstration data are essential for improving robotic imitation learning generalization. However, generating such data for complex manipulations is challenging in real-world settings. We introduce HybridGen, an automated framework that integrates Vision-Language Model (VLM) and hybrid planning. HybridGen uses a two-stage pipeline: first, VLM to parse expert demonstrations, decomposing tasks into expert-dependent (object-centric pose transformations for precise control) and plannable segments (synthesizing diverse trajectories via path planning); second, pose transformations substantially expand the first-stage data. Crucially, HybridGen generates a large volume of training data without requiring specific data formats, making it broadly applicable to a wide range of imitation learning algorithms, a characteristic which we also demonstrate empirically across multiple algorithms. Evaluations across seven tasks and their variants demonstrate that agents trained with HybridGen achieve substantial performance and generalization gains, averaging a 5% improvement over state-of-the-art methods. Notably, in the most challenging task variants, HybridGen achieves significant improvement, reaching a 59.7% average success rate, significantly outperforming Mimicgen's 49.5%. These results demonstrating its effectiveness and practicality.

Paperid: 7108, https://arxiv.org/pdf/2503.12731.pdf

Abstract:
Heat exposure significantly influences pedestrian routing behaviors. Existing methods such as agent-based modeling (ABM) and empirical measurements fail to account for individual physiological variations and environmental perception mechanisms under thermal stress. This results in a lack of human-centred, heat-adaptive routing suggestions. To address these limitations, we propose a novel Vision Language Model (VLM)-driven Persona-Perception-Planning-Memory (PPPM) framework that integrating street view imagery and urban network topology to simulate heat-adaptive pedestrian routing. Through structured prompt engineering on Gemini-2.0 model, eight distinct heat-sensitive personas were created to model mobility behaviors during heat exposure, with empirical validation through questionnaire survey. Results demonstrate that simulation outputs effectively capture inter-persona variations, achieving high significant congruence with observed route preferences and highlighting differences in the factors driving agents decisions. Our framework is highly cost-effective, with simulations costing 0.006USD and taking 47.81s per route. This Artificial Intelligence-Generated Content (AIGC) methodology advances urban climate adaptation research by enabling high-resolution simulation of thermal-responsive mobility patterns, providing actionable insights for climate-resilient urban planning.

Paperid: 7109, https://arxiv.org/pdf/2503.11265.pdf

Abstract:
Visual Question Answering (VQA) models, which fall under the category of vision-language models, conventionally execute multiple downsampling processes on image inputs to strike a balance between computational efficiency and model performance. Although this approach aids in concentrating on salient features and diminishing computational burden, it incurs the loss of vital detailed information, a drawback that is particularly damaging in end-to-end autonomous driving scenarios. Downsampling can lead to an inadequate capture of distant or small objects such as pedestrians, road signs, or obstacles, all of which are crucial for safe navigation. This loss of features negatively impacts an autonomous driving system's capacity to accurately perceive the environment, potentially escalating the risk of accidents. To tackle this problem, we put forward the Dynamic Resolution Vision Language Model (DynRsl-VLM). DynRsl-VLM incorporates a dynamic resolution image input processing approach that captures all entity feature information within an image while ensuring that the image input remains computationally tractable for the Vision Transformer (ViT). Moreover, we devise a novel image-text alignment module to replace the Q-Former, enabling simple and efficient alignment with text when dealing with dynamic resolution image inputs. Our method enhances the environmental perception capabilities of autonomous driving systems without overstepping computational constraints.

Paperid: 7110, https://arxiv.org/pdf/2503.10665.pdf

Abstract:
The emergence of small vision-language models (sVLMs) marks a critical advancement in multimodal AI, enabling efficient processing of visual and textual data in resource-constrained environments. This survey offers a comprehensive exploration of sVLM development, presenting a taxonomy of architectures - transformer-based, mamba-based, and hybrid - that highlight innovations in compact design and computational efficiency. Techniques such as knowledge distillation, lightweight attention mechanisms, and modality pre-fusion are discussed as enablers of high performance with reduced resource requirements. Through an in-depth analysis of models like TinyGPT-V, MiniGPT-4, and VL-Mamba, we identify trade-offs between accuracy, efficiency, and scalability. Persistent challenges, including data biases and generalization to complex tasks, are critically examined, with proposed pathways for addressing them. By consolidating advancements in sVLMs, this work underscores their transformative potential for accessible AI, setting a foundation for future research into efficient multimodal systems.

Paperid: 7111, https://arxiv.org/pdf/2503.09837.pdf

Abstract:
Vision Language Models (VLMs) have demonstrated significant potential in various downstream tasks, including Image/Video Generation, Visual Question Answering, Multimodal Chatbots, and Video Understanding. However, these models often struggle with basic image transformations. This paper investigates the image-level understanding of VLMs, specifically CLIP by OpenAI and SigLIP by Google. Our findings reveal that these models lack comprehension of multiple image-level augmentations. To facilitate this study, we created an augmented version of the Flickr8k dataset, pairing each image with a detailed description of the applied transformation. We further explore how this deficiency impacts downstream tasks, particularly in image editing, and evaluate the performance of state-of-the-art Image2Image models on simple transformations.

Paperid: 7112, https://arxiv.org/pdf/2503.07772.pdf

Abstract:
Despite their remarkable potential, Large Vision-Language Models (LVLMs) still face challenges with object hallucination, a problem where their generated outputs mistakenly incorporate objects that do not actually exist. Although most works focus on addressing this issue within the language-model backbone, our work shifts the focus to the image input source, investigating how specific image tokens contribute to hallucinations. Our analysis reveals a striking finding: a small subset of image tokens with high attention scores are the primary drivers of object hallucination. By removing these hallucinatory image tokens (only 1.5% of all image tokens), the issue can be effectively mitigated. This finding holds consistently across different models and datasets. Building on this insight, we introduce EAZY, a novel, training-free method that automatically identifies and Eliminates hAllucinations by Zeroing out hallucinatorY image tokens. We utilize EAZY for unsupervised object hallucination detection, achieving 15% improvement compared to previous methods. Additionally, EAZY demonstrates remarkable effectiveness in mitigating hallucinations while preserving model utility and seamlessly adapting to various LVLM architectures.

Paperid: 7113, https://arxiv.org/pdf/2503.04918.pdf

Abstract:
Vision-Language Models (VLMs) have emerged as powerful tools in artificial intelli-gence, capable of integrating textual and visual data for a unified understanding of complex scenes. While models such as Florence2, built on transformer architectures, have shown promise across general tasks, their performance in object detection within unstructured or cluttered environments remains underexplored. In this study, we fi-ne-tuned the Florence2 model for object detection tasks in non-constructed, complex environments. A comprehensive experimental framework was established involving multiple hardware configurations (NVIDIA T4, L4, and A100 GPUs), optimizers (AdamW, SGD), and varied hyperparameters including learning rates and LoRA (Low-Rank Adaptation) setups. Model training and evaluation were conducted on challenging datasets representative of real-world, disordered settings. The optimized Florence2 models exhibited significant improvements in object detection accuracy, with Mean Average Precision (mAP) metrics approaching or matching those of estab-lished models such as YOLOv8, YOLOv9, and YOLOv10. The integration of LoRA and careful fine-tuning of transformer layers contributed notably to these gains. Our find-ings highlight the adaptability of transformer-based VLMs like Florence2 for do-main-specific tasks, particularly in visually complex environments. The study under-scores the potential of fine-tuned VLMs to rival traditional convolution-based detec-tors, offering a flexible and scalable approach for advanced vision applications in re-al-world, unstructured settings.

Paperid: 7114, https://arxiv.org/pdf/2503.02897.pdf

Abstract:
High-quality annotations are essential for object detection models, but ensuring label accuracy - especially for bounding boxes - remains both challenging and costly. This paper introduces ClipGrader, a novel approach that leverages vision-language models to automatically assess the accuracy of bounding box annotations. By adapting CLIP (Contrastive Language-Image Pre-training) to evaluate both class label correctness and spatial precision of bounding box, ClipGrader offers an effective solution for grading object detection labels. Tested on modified object detection datasets with artificially disturbed bounding boxes, ClipGrader achieves 91% accuracy on COCO with a 1.8% false positive rate. Moreover, it maintains 87% accuracy with a 2.1% false positive rate when trained on just 10% of the COCO data. ClipGrader also scales effectively to larger datasets such as LVIS, achieving 79% accuracy across 1,203 classes. Our experiments demonstrate ClipGrader's ability to identify errors in existing COCO annotations, highlighting its potential for dataset refinement. When integrated into a semi-supervised object detection (SSOD) model, ClipGrader readily improves the pseudo label quality, helping achieve higher mAP (mean Average Precision) throughout the training process. ClipGrader thus provides a scalable AI-assisted tool for enhancing annotation quality control and verifying annotations in large-scale object detection datasets.

Paperid: 7115, https://arxiv.org/pdf/2503.02334.pdf

Abstract:
Vision language models (VLMs) show promise in medical diagnosis, but their performance across demographic subgroups when using in-context learning (ICL) remains poorly understood. We examine how the demographic composition of demonstration examples affects VLM performance in two medical imaging tasks: skin lesion malignancy prediction and pneumothorax detection from chest radiographs. Our analysis reveals that ICL influences model predictions through multiple mechanisms: (1) ICL allows VLMs to learn subgroup-specific disease base rates from prompts and (2) ICL leads VLMs to make predictions that perform differently across demographic groups, even after controlling for subgroup-specific disease base rates. Our empirical results inform best-practices for prompting current VLMs (specifically examining demographic subgroup performance, and matching base rates of labels to target distribution at a bulk level and within subgroups), while also suggesting next steps for improving our theoretical understanding of these models.

Paperid: 7116, https://arxiv.org/pdf/2503.01863.pdf

Abstract:
With the advent of Vision-Language Models (VLMs), medical artificial intelligence (AI) has experienced significant technological progress and paradigm shifts. This survey provides an extensive review of recent advancements in Medical Vision-Language Models (Med-VLMs), which integrate visual and textual data to enhance healthcare outcomes. We discuss the foundational technology behind Med-VLMs, illustrating how general models are adapted for complex medical tasks, and examine their applications in healthcare. The transformative impact of Med-VLMs on clinical practice, education, and patient care is highlighted, alongside challenges such as data scarcity, narrow task generalization, interpretability issues, and ethical concerns like fairness, accountability, and privacy. These limitations are exacerbated by uneven dataset distribution, computational demands, and regulatory hurdles. Rigorous evaluation methods and robust regulatory frameworks are essential for safe integration into healthcare workflows. Future directions include leveraging large-scale, diverse datasets, improving cross-modal generalization, and enhancing interpretability. Innovations like federated learning, lightweight architectures, and Electronic Health Record (EHR) integration are explored as pathways to democratize access and improve clinical relevance. This review aims to provide a comprehensive understanding of Med-VLMs' strengths and limitations, fostering their ethical and balanced adoption in healthcare.

Paperid: 7117, https://arxiv.org/pdf/2503.01184.pdf

Abstract:
This paper introduces LAFT, a novel feature transformation method designed to incorporate user knowledge and preferences into anomaly detection using natural language. Accurately modeling the boundary of normality is crucial for distinguishing abnormal data, but this is often challenging due to limited data or the presence of nuisance attributes. While unsupervised methods that rely solely on data without user guidance are common, they may fail to detect anomalies of specific interest. To address this limitation, we propose Language-Assisted Feature Transformation (LAFT), which leverages the shared image-text embedding space of vision-language models to transform visual features according to user-defined requirements. Combined with anomaly detection methods, LAFT effectively aligns visual features with user preferences, allowing anomalies of interest to be detected. Extensive experiments on both toy and real-world datasets validate the effectiveness of our method.

Paperid: 7118, https://arxiv.org/pdf/2503.00619.pdf

Abstract:
Online platforms like Pinterest hosting vast content collections traditionally rely on manual curation or user-generated search logs to create keyword landing pages (KLPs) -- topic-centered collection pages that serve as entry points for content discovery. While manual curation ensures quality, it doesn't scale to millions of collections, and search log approaches result in limited topic coverage and imprecise content matching. In this paper, we present PinLanding, a novel content-first architecture that transforms the way platforms create topical collections. Instead of deriving topics from user behavior, our system employs a multi-stage pipeline combining vision-language model (VLM) for attribute extraction, large language model (LLM) for topic generation, and a CLIP-based dual-encoder architecture for precise content matching. Our model achieves 99.7% Recall@10 on Fashion200K benchmark, demonstrating strong attribute understanding capabilities. In production deployment for search engine optimization with 4.2 million shopping landing pages, the system achieves a 4X increase in topic coverage and 14.29% improvement in collection attribute precision over the traditional search log-based approach via human evaluation. The architecture can be generalized beyond search traffic to power various user experiences, including content discovery and recommendations, providing a scalable solution to transform unstructured content into curated topical collections across any content domain.

Paperid: 7119, https://arxiv.org/pdf/2502.17589.pdf

Abstract:
Automated chart summarization is crucial for enhancing data accessibility and enabling efficient information extraction from visual data. While recent advances in visual-language models (VLMs) have demonstrated promise, existing methods often suffer from limitations in matching the generated summary to the chart data and in reasoning about complex chart patterns. This paper introduces End-to-End Visual Chain-of-Thought (V-CoT) for chart summarization, a novel approach optimized for Large Vision-Language Models (LVLMs). Our method directly trains an LVLM to process chart images and generate textual summaries in an end-to-end fashion, eliminating the need for explicit chart parsing modules. We incorporate a visual Chain-of-Thought mechanism through instruction fine-tuning, implicitly guiding the LVLM to perform visual reasoning steps during summary generation. Evaluated on the large-scale Chart-Sum-QA dataset, our V-CoT method significantly outperforms state-of-the-art baselines across a range of automatic metrics, including BLEU, BLEURT, CIDEr, and CS, and demonstrates superior matching degree and reasoning correctness in human evaluations. Ablation studies and detailed analyses further validate the effectiveness and robustness of our proposed approach, establishing a new benchmark for end-to-end chart summarization.

Paperid: 7120, https://arxiv.org/pdf/2502.14412.pdf

Abstract:
The prevalence of Vision-Language Models (VLMs) raises important questions about privacy in an era where visual information is increasingly available. While foundation VLMs demonstrate broad knowledge and learned capabilities, we specifically investigate their ability to infer geographic location from previously unseen image data. This paper introduces a benchmark dataset collected from Google Street View that represents its global distribution of coverage. Foundation models are evaluated on single-image geolocation inference, with many achieving median distance errors of <300 km. We further evaluate VLM "agents" with access to supplemental tools, observing up to a 30.6% decrease in distance error. Our findings establish that modern foundation VLMs can act as powerful image geolocation tools, without being specifically trained for this task. When coupled with increasing accessibility of these models, our findings have greater implications for online privacy. We discuss these risks, as well as future work in this area.

Paperid: 7121, https://arxiv.org/pdf/2502.13095.pdf

Abstract:
Recent studies reveal that vision-language models (VLMs) become more susceptible to harmful requests and jailbreak attacks after integrating the vision modality, exhibiting greater vulnerability than their text-only LLM backbones. To uncover the root cause of this phenomenon, we conduct an in-depth analysis and identify a key issue: multimodal inputs introduce an modality-induced activation shift toward a "safer" direction compared to their text-only counterparts, leading VLMs to systematically overestimate the safety of harmful inputs. We refer to this issue as safety perception distortion. To mitigate such distortion, we propose Activation Shift Disentanglement and Calibration (ShiftDC), a training-free method that decomposes and calibrates the modality-induced activation shift to reduce the impact of modality on safety. By isolating and removing the safety-relevant component, ShiftDC restores the inherent safety alignment of the LLM backbone while preserving the vision-language capabilities of VLMs. Empirical results demonstrate that ShiftDC significantly enhances alignment performance on safety benchmarks without impairing model utility.

Paperid: 7122, https://arxiv.org/pdf/2502.11969.pdf

Abstract:
In vision-language models (VLMs), prompt tuning has shown its effectiveness in adapting models to downstream tasks. However, learned prompts struggle to generalize to unseen classes, as they tend to overfit to the classes that are targeted during prompt tuning. Examining failure cases, we observed that learned prompts disrupt the semantics of unseen classes, generating text embeddings with incorrect semantic relationships among classes. To address this, we propose Similarity Alignment Regularization (SAR), which regularizes learnable prompts to preserve the semantic relationships among classes captured by hand-crafted prompts. Specifically, we first obtain novel classes related to base classes using ChatGPT-4o and utilize them as potential unseen classes during prompt tuning. Then, by targeting both base and novel classes, SAR aligns the similarity relationships among text embeddings generated by learnable prompts with the similarity relationships from hand-crafted prompts. Extensive experiments applying SAR to existing prompt tuning methods demonstrate its effectiveness in improving generalization to unseen classes.

Paperid: 7123, https://arxiv.org/pdf/2502.11747.pdf

Abstract:
Video question answering that requires external knowledge beyond the visual content remains a significant challenge in AI systems. While models can effectively answer questions based on direct visual observations, they often falter when faced with questions requiring broader contextual knowledge. To address this limitation, we investigate knowledge-intensive video question answering (KI-VideoQA) through the lens of multi-modal retrieval-augmented generation, with a particular focus on handling open-ended questions rather than just multiple-choice formats. Our comprehensive analysis examines various retrieval augmentation approaches using cutting-edge retrieval and vision language models, testing both zero-shot and fine-tuned configurations. We investigate several critical dimensions: the interplay between different information sources and modalities, strategies for integrating diverse multi-modal contexts, and the dynamics between query formulation and retrieval result utilization. Our findings reveal that while retrieval augmentation shows promise in improving model performance, its success is heavily dependent on the chosen modality and retrieval methodology. The study also highlights the critical role of query construction and retrieval depth optimization in effective knowledge integration. Through our proposed approach, we achieve a substantial 17.5% improvement in accuracy on multiple choice questions in the KnowIT VQA dataset, establishing new state-of-the-art performance levels.

Paperid: 7124, https://arxiv.org/pdf/2502.10250.pdf

Abstract:
Vision-language models (VLMs) excel in various visual benchmarks but are often constrained by the lack of high-quality visual fine-tuning data. To address this challenge, we introduce VisCon-100K, a novel dataset derived from interleaved image-text web documents. Our approach transforms 45K web documents from the OBELICS dataset into 100K image conversation samples. We utilize GPT-4V to generate image-contextual captions and OpenChat 3.5 model to convert these captions into diverse free-form and multiple-choice question-answer pairs. Integrating this dataset for fine-tuning considerably enhances VLM performance across multiple benchmarks. Unlike methods that focus solely on fine-grained visual content, our approach leverages accompanying web context, yielding superior results. We also discover that a 'leaky modality mix', where conversation samples contain questions answerable from both the image and its contextual caption, outperforms non-leaky combinations of captions and Q&A pairs. VisCon-100k dataset shows strong performance with two popular VLM approaches: text-only large language model (LLM) aligned with a vision encoder using image captions data (ShareGPT4V-7b) and multimodally pretrained LLM (IDEFICS2-8b) using interleaved image-text data. In addition to releasing the VisCon-100K dataset, we provide a contextual captioner trained on this dataset, facilitating scalable fine-tuning data generation for future research and open-source applications. Using the same pipeline, but substituting our trained contextual captioner for GPT-4V, we also release the larger VisCon-1M dataset.

Paperid: 7125, https://arxiv.org/pdf/2502.09120.pdf

Abstract:
This study investigates whether vision-language models (VLMs) can perform pragmatic inference, focusing on ignorance implicatures, utterances that imply the speaker's lack of precise knowledge. To test this, we systematically manipulated contextual cues: the visually depicted situation (visual cue) and QUD-based linguistic prompts (linguistic cue). When only visual cues were provided, three state-of-the-art VLMs (GPT-4o, Gemini 1.5 Pro, and Claude 3.5 sonnet) produced interpretations largely based on the lexical meaning of the modified numerals. When linguistic cues were added to enhance contextual informativeness, Claude exhibited more human-like inference by integrating both types of contextual cues. In contrast, GPT and Gemini favored precise, literal interpretations. Although the influence of contextual cues increased, they treated each contextual cue independently and aligned them with semantic features rather than engaging in context-driven reasoning. These findings suggest that although the models differ in how they handle contextual cues, Claude's ability to combine multiple cues may signal emerging pragmatic competence in multimodal models.

Paperid: 7126, https://arxiv.org/pdf/2502.06631.pdf

Abstract:
Human-in-the-Loop (HITL) systems are essential in high-stakes, real-world applications where AI must collaborate with human decision-makers. This work investigates how Conformal Prediction (CP) techniques, which provide rigorous coverage guarantees, can enhance the reliability of state-of-the-art human action recognition (HAR) systems built upon Vision-Language Models (VLMs). We demonstrate that CP can significantly reduce the average number of candidate classes without modifying the underlying VLM. However, these reductions often result in distributions with long tails which can hinder their practical utility. To mitigate this, we propose tuning the temperature of the softmax prediction, without using additional calibration data. This work contributes to ongoing efforts for multi-modal human-AI interaction in dynamic real-world environments.

Paperid: 7127, https://arxiv.org/pdf/2502.05926.pdf

Abstract:
The rapid advancements in large language models (LLMs) have unlocked their potential for multimodal tasks, where text and visual data are processed jointly. However, applying LLMs to medical imaging, particularly for chest X-rays (CXR), poses significant challenges due to the need for precise visual-textual alignment and the preservation of critical diagnostic details. In this paper, we propose Multi-Stage Adaptive Vision-Language Tuning (MAViLT), a novel framework designed to enhance multimodal reasoning and generation for CXR understanding. MAViLT incorporates a clinical gradient-weighted tokenization process and a hierarchical fine-tuning strategy, enabling it to generate accurate radiology reports, synthesize realistic CXRs from text, and answer vision-based clinical questions. We evaluate MAViLT on two benchmark datasets, MIMIC-CXR and Indiana University CXR, achieving state-of-the-art results across all tasks. Human evaluations further validate the clinical relevance and utility of MAViLT, making it a robust tool for real-world medical applications. This work demonstrates the feasibility of leveraging LLMs for multimodal medical imaging while addressing key challenges in vision-language integration.

Paperid: 7128, https://arxiv.org/pdf/2502.01071.pdf

Abstract:
The integration of language instructions with robotic control, particularly through Vision Language Action (VLA) models, has shown significant potential. However, these systems are often hindered by high computational costs, the need for extensive retraining, and limited scalability, making them less accessible for widespread use. In this paper, we introduce SVLR (Scalable Visual Language Robotics), an open-source, modular framework that operates without the need for retraining, providing a scalable solution for robotic control. SVLR leverages a combination of lightweight, open-source AI models including the Vision-Language Model (VLM) Mini-InternVL, zero-shot image segmentation model CLIPSeg, Large Language Model Phi-3, and sentence similarity model all-MiniLM to process visual and language inputs. These models work together to identify objects in an unknown environment, use them as parameters for task execution, and generate a sequence of actions in response to natural language instructions. A key strength of SVLR is its scalability. The framework allows for easy integration of new robotic tasks and robots by simply adding text descriptions and task definitions, without the need for retraining. This modularity ensures that SVLR can continuously adapt to the latest advancements in AI technologies and support a wide range of robots and tasks. SVLR operates effectively on an NVIDIA RTX 2070 (mobile) GPU, demonstrating promising performance in executing pick-and-place tasks. While these initial results are encouraging, further evaluation across a broader set of tasks and comparisons with existing VLA models are needed to assess SVLR's generalization capabilities and performance in more complex scenarios.

Paperid: 7129, https://arxiv.org/pdf/2501.15046.pdf

Abstract:
Despite their impressive performance on multi-modal tasks, large vision-language models (LVLMs) tend to suffer from hallucinations. An important type is object hallucination, where LVLMs generate objects that are inconsistent with the images shown to the model. Existing works typically attempt to quantify object hallucinations by detecting and measuring the fraction of hallucinated objects in generated captions. Additionally, more recent work also measures object hallucinations by directly querying the LVLM with binary questions about the presence of likely hallucinated objects based on object statistics like top-k frequent objects and top-k co-occurring objects. In this paper, we present Context-Aware Object Similarities (CAOS), a novel approach for evaluating object hallucination in LVLMs using object statistics as well as the generated captions. CAOS uniquely integrates object statistics with semantic relationships between objects in captions and ground-truth data. Moreover, existing approaches usually only detect and measure hallucinations belonging to a predetermined set of in-domain objects (typically the set of all ground-truth objects for the training dataset) and ignore generated objects that are not part of this set, leading to under-evaluation. To address this, we further employ language model--based object recognition to detect potentially out-of-domain hallucinated objects and use an ensemble of LVLMs for verifying the presence of such objects in the query image. CAOS also examines the sequential dynamics of object generation, shedding light on how the order of object appearance influences hallucinations, and employs word embedding models to analyze the semantic reasons behind hallucinations. CAOS aims to offer a nuanced understanding of the hallucination tendencies of LVLMs by providing a systematic framework to identify and interpret object hallucinations.

Paperid: 7130, https://arxiv.org/pdf/2501.11782.pdf

Abstract:
As modern video games become increasingly complex, traditional manual testing methods are proving costly and inefficient, limiting the ability to ensure high-quality game experiences. While advancements in Artificial Intelligence (AI) offer the potential to assist human testers, the effectiveness of AI in truly enhancing real-world human performance remains underexplored. This study investigates how AI can improve game testing by developing and experimenting with an AI-assisted workflow that leverages state-of-the-art machine learning models for defect detection. Through an experiment involving 800 test cases and 276 participants of varying backgrounds, we evaluate the effectiveness of AI assistance under four conditions: with or without AI support, and with or without detailed knowledge of defects and design documentation. The results indicate that AI assistance significantly improves defect identification performance, particularly when paired with detailed knowledge. However, challenges arise when AI errors occur, negatively impacting human decision-making. Our findings show the importance of optimizing human-AI collaboration and implementing strategies to mitigate the effects of AI inaccuracies. By this research, we demonstrate AI's potential and problems in enhancing efficiency and accuracy in game testing workflows and offers practical insights for integrating AI into the testing process.

Paperid: 7131, https://arxiv.org/pdf/2501.05717.pdf

Abstract:
The recent widespread adoption of drones for studying marine animals provides opportunities for deriving biological information from aerial imagery. The large scale of imagery data acquired from drones is well suited for machine learning (ML) analysis. Development of ML models for analyzing marine animal aerial imagery has followed the classical paradigm of training, testing, and deploying a new model for each dataset, requiring significant time, human effort, and ML expertise. We introduce Frame Level ALIgment and tRacking (FLAIR), which leverages the video understanding of Segment Anything Model 2 (SAM2) and the vision-language capabilities of Contrastive Language-Image Pre-training (CLIP). FLAIR takes a drone video as input and outputs segmentation masks of the species of interest across the video. Notably, FLAIR leverages a zero-shot approach, eliminating the need for labeled data, training a new model, or fine-tuning an existing model to generalize to other species. With a dataset of 18,000 drone images of Pacific nurse sharks, we trained state-of-the-art object detection models to compare against FLAIR. We show that FLAIR massively outperforms these object detectors and performs competitively against two human-in-the-loop methods for prompting SAM2, achieving a Dice score of 0.81. FLAIR readily generalizes to other shark species without additional human effort and can be combined with novel heuristics to automatically extract relevant information including length and tailbeat frequency. FLAIR has significant potential to accelerate aerial imagery analysis workflows, requiring markedly less human effort and expertise than traditional machine learning workflows, while achieving superior accuracy. By reducing the effort required for aerial imagery analysis, FLAIR allows scientists to spend more time interpreting results and deriving insights about marine ecosystems.

Paperid: 7132, https://arxiv.org/pdf/2501.05413.pdf

Abstract:
Training audio-to-image generative models requires an abundance of diverse audio-visual pairs that are semantically aligned. Such data is almost always curated from in-the-wild videos, given the cross-modal semantic correspondence that is inherent to them. In this work, we hypothesize that insisting on the absolute need for ground truth audio-visual correspondence, is not only unnecessary, but also leads to severe restrictions in scale, quality, and diversity of the data, ultimately impairing its use in the modern generative models. That is, we propose a scalable image sonification framework where instances from a variety of high-quality yet disjoint uni-modal origins can be artificially paired through a retrieval process that is empowered by reasoning capabilities of modern vision-language models. To demonstrate the efficacy of this approach, we use our sonified images to train an audio-to-image generative model that performs competitively against state-of-the-art. Finally, through a series of ablation studies, we exhibit several intriguing auditory capabilities like semantic mixing and interpolation, loudness calibration and acoustic space modeling through reverberation that our model has implicitly developed to guide the image generation process.

Paperid: 7133, https://arxiv.org/pdf/2501.05228.pdf

Abstract:
Out-of-distribution (OOD) detection has seen significant advancements with zero-shot approaches by leveraging the powerful Vision-Language Models (VLMs) such as CLIP. However, prior research works have predominantly focused on enhancing Far-OOD performance, while potentially compromising Near-OOD efficacy, as observed from our pilot study. To address this issue, we propose a novel strategy to enhance zero-shot OOD detection performances for both Far-OOD and Near-OOD scenarios by innovatively harnessing Large Language Models (LLMs) and VLMs. Our approach first exploit an LLM to generate superclasses of the ID labels and their corresponding background descriptions followed by feature extraction using CLIP. We then isolate the core semantic features for ID data by subtracting background features from the superclass features. The refined representation facilitates the selection of more appropriate negative labels for OOD data from a comprehensive candidate label set of WordNet, thereby enhancing the performance of zero-shot OOD detection in both scenarios. Furthermore, we introduce novel few-shot prompt tuning and visual prompt tuning to adapt the proposed framework to better align with the target distribution. Experimental results demonstrate that the proposed approach consistently outperforms current state-of-the-art methods across multiple benchmarks, with an improvement of up to 2.9% in AUROC and a reduction of up to 12.6% in FPR95. Additionally, our method exhibits superior robustness against covariate shift across different domains, further highlighting its effectiveness in real-world scenarios.

Paperid: 7134, https://arxiv.org/pdf/2501.04568.pdf

Abstract:
Vision-language models (VLMs) have demonstrated remarkable potential in integrating visual and linguistic information, but their performance is often constrained by the need for extensive, high-quality image-text training data. Curation of these image-text pairs is both time-consuming and computationally expensive. To address this challenge, we introduce SVP (Sampling-based Visual Projection), a novel framework that enhances vision-language alignment without relying on manually curated text-image pairs or preference annotation. SVP leverages a small set of manually selected images, self-captioning and a pre-trained grounding model as a feedback mechanism to elicit latent information in VLMs. We evaluate our approach across six key areas: captioning, referring, visual question answering, multitasking, hallucination control, and object recall. Results demonstrate significant improvements, including a 14 % average improvement in captioning tasks, up to 12 % increase in object recall, and significantly reduced hallucinations, while maintaining question-answering capabilities. Using SVP, a small VLM achieves hallucination reductions similar to a model five times larger, while a VLM with initially poor referring capabilities more than doubles its performance, approaching parity with a model twice its size.

Paperid: 7135, https://arxiv.org/pdf/2412.20632.pdf

Abstract:
Human acceptance of social robots is greatly effected by empathy and perceived understanding. This necessitates accurate and flexible responses to various input data from the user. While systems such as this can become increasingly complex as more states or response types are included, new research in the application of large language models towards human-robot interaction has allowed for more streamlined perception and reaction pipelines. LLM-selected actions and emotional expressions can help reinforce the realism of displayed empathy and allow for improved communication between the robot and user. Beyond portraying empathy in spoken or written responses, this shows the possibilities of using LLMs in actuated, real world scenarios. In this work we extend research in LLM-driven nonverbal behavior for social robots by considering more open-ended emotional response selection leveraging new advances in vision-language models, along with emotionally aligned motion and color pattern selections that strengthen conveyance of meaning and empathy.

Paperid: 7136, https://arxiv.org/pdf/2412.17325.pdf

Abstract:
Domain adaptation, a pivotal branch of transfer learning, aims to enhance the performance of machine learning models when deployed in target domains with distinct data distributions. This is particularly critical for object detection tasks, where domain shifts (caused by factors such as lighting conditions, viewing angles, and environmental variations) can lead to significant performance degradation. This review delves into advanced methodologies for domain adaptation, including adversarial learning, discrepancy-based, multi-domain, teacher-student, ensemble, and Vision Language Models techniques, emphasizing their efficacy in reducing domain gaps and enhancing model robustness. Feature-based methods have emerged as powerful tools for addressing these challenges by harmonizing feature representations across domains. These techniques, such as Feature Alignment, Feature Augmentation/Reconstruction, and Feature Transformation, are employed alongside or as integral parts of other domain adaptation strategies to minimize domain gaps and improve model performance. Special attention is given to strategies that minimize the reliance on extensive labeled data and using unlabeled data, particularly in scenarios involving synthetic-to-real domain shifts. Applications in fields such as autonomous driving and medical imaging are explored, showcasing the potential of these methods to ensure reliable object detection in diverse and complex settings. By providing a thorough analysis of state-of-the-art techniques, challenges, and future directions, this work offers a valuable reference for researchers striving to develop resilient and adaptable object detection frameworks, advancing the seamless deployment of artificial intelligence in dynamic environments.

Paperid: 7137, https://arxiv.org/pdf/2412.16201.pdf

Abstract:
This paper presents CLIP-RLDrive, a new reinforcement learning (RL)-based framework for improving the decision-making of autonomous vehicles (AVs) in complex urban driving scenarios, particularly in unsignalized intersections. To achieve this goal, the decisions for AVs are aligned with human-like preferences through Contrastive Language-Image Pretraining (CLIP)-based reward shaping. One of the primary difficulties in RL scheme is designing a suitable reward model, which can often be challenging to achieve manually due to the complexity of the interactions and the driving scenarios. To deal with this issue, this paper leverages Vision-Language Models (VLMs), particularly CLIP, to build an additional reward model based on visual and textual cues.

Paperid: 7138, https://arxiv.org/pdf/2412.15462.pdf

Abstract:
TalkWithMachines aims to enhance human-robot interaction by contributing to interpretable industrial robotic systems, especially for safety-critical applications. The presented paper investigates recent advancements in Large Language Models (LLMs) and Vision Language Models (VLMs), in combination with robotic perception and control. This integration allows robots to understand and execute commands given in natural language and to perceive their environment through visual and/or descriptive inputs. Moreover, translating the LLM's internal states and reasoning into text that humans can easily understand ensures that operators gain a clearer insight into the robot's current state and intentions, which is essential for effective and safe operation. Our paper outlines four LLM-assisted simulated robotic control workflows, which explore (i) low-level control, (ii) the generation of language-based feedback that describes the robot's internal states, (iii) the use of visual information as additional input, and (iv) the use of robot structure information for generating task plans and feedback, taking the robot's physical capabilities and limitations into account. The proposed concepts are presented in a set of experiments, along with a brief discussion. Project description, videos, and supplementary materials will be available on the project website: https://talk-machines.github.io.

Paperid: 7139, https://arxiv.org/pdf/2412.14366.pdf

Abstract:
Recent advances in generative AI make it convenient to create different types of content, including text, images, and code. In this paper, we explore the generation of images in the style of paintings in the surrealism movement using vision-language generative models, including DALL-E, Deep Dream Generator, and DreamStudio. Our investigation starts with the generation of images under various image generation settings and different models. The primary objective is to identify the most suitable model and settings for producing such images. Additionally, we aim to understand the impact of using edited base images on the generated resulting images. Through these experiments, we evaluate the performance of selected models and gain valuable insights into their capabilities in generating such images. Our analysis shows that Dall-E 2 performs the best when using the generated prompt by ChatGPT.

Paperid: 7140, https://arxiv.org/pdf/2412.12492.pdf

Abstract:
Semi-supervised medical image segmentation (SSMIS) uses consistency learning to regularize model training, which alleviates the burden of pixel-wise manual annotations. However, it often suffers from error supervision from low-quality pseudo labels. Vision-Language Model (VLM) has great potential to enhance pseudo labels by introducing text prompt guided multimodal supervision information. It nevertheless faces the cross-modal problem: the obtained messages tend to correspond to multiple targets. To address aforementioned problems, we propose a Dual Semantic Similarity-Supervised VLM (DuSSS) for SSMIS. Specifically, 1) a Dual Contrastive Learning (DCL) is designed to improve cross-modal semantic consistency by capturing intrinsic representations within each modality and semantic correlations across modalities. 2) To encourage the learning of multiple semantic correspondences, a Semantic Similarity-Supervision strategy (SSS) is proposed and injected into each contrastive learning process in DCL, supervising semantic similarity via the distribution-based uncertainty levels. Furthermore, a novel VLM-based SSMIS network is designed to compensate for the quality deficiencies of pseudo-labels. It utilizes the pretrained VLM to generate text prompt guided supervision information, refining the pseudo label for better consistency regularization. Experimental results demonstrate that our DuSSS achieves outstanding performance with Dice of 82.52%, 74.61% and 78.03% on three public datasets (QaTa-COV19, BM-Seg and MoNuSeg).

Paperid: 7141, https://arxiv.org/pdf/2412.07979.pdf

Abstract:
Contrastive learning has emerged as a pivotal framework for representation learning, underpinning advances in both unimodal and bimodal applications like SimCLR and CLIP. To address fundamental limitations like large batch size dependency and bimodality, methods such as SogCLR leverage stochastic optimization for the global contrastive objective. Inspired by SogCLR's efficiency and adaptability, we introduce AmCLR and xAmCLR objective functions tailored for bimodal vision-language models to further enhance the robustness of contrastive learning. AmCLR integrates diverse augmentations, including text paraphrasing and image transformations, to reinforce the alignment of contrastive representations, keeping batch size limited to a few hundred samples unlike CLIP which needs batch size of 32,768 to produce reasonable results. xAmCLR further extends this paradigm by incorporating intra-modal alignments between original and augmented modalities for richer feature learning. These advancements yield a more resilient and generalizable contrastive learning process, aimed at overcoming bottlenecks in scaling and augmentative diversity. Since we have built our framework on the existing SogCLR, we are able to demonstrate improved representation quality with fewer computational resources, establishing a foundation for scalable and robust multi-modal learning.

Paperid: 7142, https://arxiv.org/pdf/2412.02262.pdf

Abstract:
Climate change's destruction of marine biodiversity is threatening communities and economies around the world which rely on healthy oceans for their livelihoods. The challenge of applying computer vision to niche, real-world domains such as ocean conservation lies in the dynamic and diverse environments where traditional top-down learning struggle with long-tailed distributions, generalization, and domain transfer. Scalable species identification for ocean monitoring is particularly difficult due to the need to adapt models to new environments and identify rare or unseen species. To overcome these limitations, we propose leveraging bottom-up, open-domain learning frameworks as a resilient, scalable solution for image and video analysis in marine applications. Our preliminary demonstration uses pretrained vision-language models (VLMs) combined with retrieval-augmented generation (RAG) as grounding, leaving the door open for numerous architectural, training and engineering optimizations. We validate this approach through a preliminary application in classifying fish from video onboard fishing vessels, demonstrating impressive emergent retrieval and prediction capabilities without domain-specific training or knowledge of the task itself.

Paperid: 7143, https://arxiv.org/pdf/2412.02177.pdf

Abstract:
With the emergence of large-scale vision-language models, realistic radiology reports may be generated using only medical images as input guided by simple prompts. However, their practical utility has been limited due to the factual errors in their description of findings. In this paper, we propose a novel model for explainable fact-checking that identifies errors in findings and their locations indicated through the reports. Specifically, we analyze the types of errors made by automated reporting methods and derive a new synthetic dataset of images paired with real and fake descriptions of findings and their locations from a ground truth dataset. A new multi-label cross-modal contrastive regression network is then trained on this datsaset. We evaluate the resulting fact-checking model and its utility in correcting reports generated by several SOTA automated reporting tools on a variety of benchmark datasets with results pointing to over 40\% improvement in report quality through such error detection and correction.

Paperid: 7144, https://arxiv.org/pdf/2412.01095.pdf

Abstract:
The rapid advancement of vision-language models (VLMs) has established a new paradigm in video anomaly detection (VAD): leveraging VLMs to simultaneously detect anomalies and provide comprehendible explanations for the decisions. Existing work in this direction often assumes the complex reasoning required for VAD exceeds the capabilities of pretrained VLMs. Consequently, these approaches either incorporate specialized reasoning modules during inference or rely on instruction tuning datasets through additional training to adapt VLMs for VAD. However, such strategies often incur substantial computational costs or data annotation overhead. To address these challenges in explainable VAD, we introduce a verbalized learning framework named VERA that enables VLMs to perform VAD without model parameter modifications. Specifically, VERA automatically decomposes the complex reasoning required for VAD into reflections on simpler, more focused guiding questions capturing distinct abnormal patterns. It treats these reflective questions as learnable parameters and optimizes them through data-driven verbal interactions between learner and optimizer VLMs, using coarsely labeled training data. During inference, VERA embeds the learned questions into model prompts to guide VLMs in generating segment-level anomaly scores, which are then refined into frame-level scores via the fusion of scene and temporal contexts. Experimental results on challenging benchmarks demonstrate that the learned questions of VERA are highly adaptable, significantly improving both detection performance and explainability of VLMs for VAD.

Paperid: 7145, https://arxiv.org/pdf/2412.00686.pdf

Abstract:
Counting is a fundamental operation for various real-world visual tasks, requiring both object recognition and robust counting capabilities. Despite their advanced visual perception, large vision-language models (LVLMs) are known to struggle with counting tasks. In this work, we evaluate the performance of several recent LVLMs on visual counting tasks across multiple counting and vision datasets. We observe that while their performance may be less prone to error for small numbers of objects, they exhibit significant weaknesses as the number of objects increases. To alleviate this issue, we propose a simple yet effective baseline method that enhances LVLMs' counting ability for large numbers of objects using a divide-and-conquer approach. Our method decomposes counting problems into sub-tasks. Moreover, it incorporates a mechanism to prevent objects from being split during division, which could otherwise lead to repetitive counting -- a common issue in a naive divide-and-conquer implementation. We demonstrate the effectiveness of this approach across various datasets and benchmarks, establishing it as a valuable reference for evaluating future solutions.

Paperid: 7146, https://arxiv.org/pdf/2411.17891.pdf

Abstract:
Technological advances in artificial intelligence (AI) have enabled the development of large vision language models (LVLMs) that are trained on millions of paired image and text samples. Subsequent research efforts have demonstrated great potential of LVLMs to achieve high performance in medical imaging use cases (e.g., radiology report generation), but there remain barriers that hinder the ability to deploy these solutions broadly. These include the cost of extensive computational requirements for developing large scale models, expertise in the development of sophisticated AI models, and the difficulty in accessing substantially large, high-quality datasets that adequately represent the population in which the LVLM solution is to be deployed. The HOPPR Medical-Grade Platform addresses these barriers by providing powerful computational infrastructure, a suite of foundation models on top of which developers can fine-tune for their specific use cases, and a robust quality management system that sets a standard for evaluating fine-tuned models for deployment in clinical settings. The HOPPR Platform has access to millions of imaging studies and text reports sourced from hundreds of imaging centers from diverse populations to pretrain foundation models and enable use case-specific cohorts for fine-tuning. All data are deidentified and securely stored for HIPAA compliance. Additionally, developers can securely host models on the HOPPR platform and access them via an API to make inferences using these models within established clinical workflows. With the Medical-Grade Platform, HOPPR's mission is to expedite the deployment of LVLM solutions for medical imaging and ultimately optimize radiologist's workflows and meet the growing demands of the field.

Paperid: 7147, https://arxiv.org/pdf/2411.15673.pdf

Abstract:
In recent years there has been enormous interest in vision-language models trained using self-supervised objectives. However, the use of large-scale datasets scraped from the web for training also makes these models vulnerable to potential security threats, such as backdooring and poisoning attacks. In this paper, we propose a method for mitigating such attacks on contrastively trained vision-language models. Our approach leverages external knowledge extracted from a language model to prevent models from learning correlations between image regions which lack strong alignment with external knowledge. We do this by imposing constraints to enforce that attention paid by the model to visual regions is proportional to the alignment of those regions with external knowledge. We conduct extensive experiments using a variety of recent backdooring and poisoning attacks on multiple datasets and architectures. Our results clearly demonstrate that our proposed approach is highly effective at defending against such attacks across multiple settings, while maintaining model utility and without requiring any changes at inference time

Paperid: 7148, https://arxiv.org/pdf/2411.15556.pdf

Abstract:
Vision-Language Models (VLMs) are crucial for applications requiring integrated understanding textual and visual information. However, existing VLMs struggle with long videos due to computational inefficiency, memory limitations, and difficulties in maintaining coherent understanding across extended sequences. To address these challenges, we introduce ReWind, a novel memory-based VLM designed for efficient long video understanding while preserving temporal fidelity. ReWind operates in a two-stage framework. In the first stage, ReWind maintains a dynamic learnable memory module with a novel \textbf{read-perceive-write} cycle that stores and updates instruction-relevant visual information as the video unfolds. This module utilizes learnable queries and cross-attentions between memory contents and the input stream, ensuring low memory requirements by scaling linearly with the number of tokens. In the second stage, we propose an adaptive frame selection mechanism guided by the memory content to identify instruction-relevant key moments. It enriches the memory representations with detailed spatial information by selecting a few high-resolution frames, which are then combined with the memory contents and fed into a Large Language Model (LLM) to generate the final answer. We empirically demonstrate ReWind's superior performance in visual question answering (VQA) and temporal grounding tasks, surpassing previous methods on long video benchmarks. Notably, ReWind achieves a +13\% score gain and a +12\% accuracy improvement on the MovieChat-1K VQA dataset and an +8\% mIoU increase on Charades-STA for temporal grounding.

Paperid: 7149, https://arxiv.org/pdf/2411.15201.pdf

Abstract:
Current benchmarks for evaluating Vision Language Models (VLMs) often fall short in thoroughly assessing model abilities to understand and process complex visual and textual content. They typically focus on simple tasks that do not require deep reasoning or the integration of multiple data modalities to solve an original problem. To address this gap, we introduce the PARROT-360V Benchmark, a novel and comprehensive benchmark featuring 2487 challenging visual puzzles designed to test VLMs on complex visual reasoning tasks. We evaluated leading models: GPT-4o, Claude-3.5-Sonnet, and Gemini-1.5-Pro, using PARROT-360V to assess their capabilities in combining visual clues with language skills to solve tasks in a manner akin to human problem-solving. Our findings reveal a notable performance gap: state-of-the-art models scored between 28 to 56 percentage on our benchmark, significantly lower than their performance on popular benchmarks. This underscores the limitations of current VLMs in handling complex, multi-step reasoning tasks and highlights the need for more robust evaluation frameworks to advance the field.

Paperid: 7150, https://arxiv.org/pdf/2411.13211.pdf

Abstract:
Using vision-language models (VLMs) as reward models in reinforcement learning holds promise for reducing costs and improving safety. So far, VLM reward models have only been used for goal-oriented tasks, where the agent must reach a particular final outcome. We explore VLMs' potential to supervise tasks that cannot be scored by the final state alone. To this end, we introduce ViSTa, a dataset for evaluating Vision-based understanding of Sequential Tasks. ViSTa comprises over 4,000 videos with step-by-step descriptions in virtual home, Minecraft, and real-world environments. Its novel hierarchical structure -- basic single-step tasks composed into more and more complex sequential tasks -- allows a fine-grained understanding of how well VLMs can judge tasks with varying complexity. To illustrate this, we use ViSTa to evaluate state-of-the-art VLMs, including CLIP, ViCLIP, and GPT-4o. We find that, while they are all good at object recognition, they fail to understand sequential tasks, with only GPT-4o achieving non-trivial performance.

Paperid: 7151, https://arxiv.org/pdf/2411.09180.pdf

Abstract:
Drone-captured images present significant challenges in object detection due to varying shooting conditions, which can alter object appearance and shape. Factors such as drone altitude, angle, and weather cause these variations, influencing the performance of object detection algorithms. To tackle these challenges, we introduce an innovative vision-language approach using learnable prompts. This shift from conventional manual prompts aims to reduce domain-specific knowledge interference, ultimately improving object detection capabilities. Furthermore, we streamline the training process with a one-step approach, updating the learnable prompt concurrently with model training, enhancing efficiency without compromising performance. Our study contributes to domain-generalized object detection by leveraging learnable prompts and optimizing training processes. This enhances model robustness and adaptability across diverse environments, leading to more effective aerial object detection.

Paperid: 7152, https://arxiv.org/pdf/2411.07871.pdf

Abstract:
The rapid advancements in Large Language Models (LLMs) and Vision-Language Models (VLMs) have shown great potential in medical diagnostics, particularly in radiology, where datasets such as X-rays are paired with human-generated diagnostic reports. However, a significant research gap exists in the neuroimaging field, especially for conditions such as Alzheimer's disease, due to the lack of comprehensive diagnostic reports that can be utilized for model fine-tuning. This paper addresses this gap by generating synthetic diagnostic reports using GPT-4o-mini on structured data from the OASIS-4 dataset, which comprises 663 patients. Using the synthetic reports as ground truth for training and validation, we then generated neurological reports directly from the images in the dataset leveraging the pre-trained BiomedCLIP and T5 models. Our proposed method achieved a BLEU-4 score of 0.1827, ROUGE-L score of 0.3719, and METEOR score of 0.4163, revealing its potential in generating clinically relevant and accurate diagnostic reports.

Paperid: 7153, https://arxiv.org/pdf/2411.05985.pdf

Abstract:
This paper examines potential biases and inconsistencies in emotional evocation of images produced by generative artificial intelligence (AI) models and their potential bias toward negative emotions. In particular, we assess this bias by comparing the emotions evoked by an AI-produced image to the emotions evoked by prompts used to create those images. As a first step, the study evaluates three approaches for identifying emotions in images -- traditional supervised learning, zero-shot learning with vision-language models, and cross-modal auto-captioning -- using EmoSet, a large dataset of image-emotion annotations that categorizes images across eight emotional types. Results show fine-tuned models, particularly Google's Vision Transformer (ViT), significantly outperform zero-shot and caption-based methods in recognizing emotions in images. For a cross-modality comparison, we then analyze the differences between emotions in text prompts -- via existing text-based emotion-recognition models -- and the emotions evoked in the resulting images. Findings indicate that AI-generated images frequently lean toward negative emotional content, regardless of the original prompt. This emotional skew in generative models could amplify negative affective content in digital spaces, perpetuating its prevalence and impact. The study advocates for a multidisciplinary approach to better align AI emotion recognition with psychological insights and address potential biases in generative AI outputs across digital media.

Paperid: 7154, https://arxiv.org/pdf/2411.05734.pdf

Abstract:
Access to expert coaching is essential for developing technique in sports, yet economic barriers often place it out of reach for many enthusiasts. To bridge this gap, we introduce Poze, an innovative video processing framework that provides feedback on human motion, emulating the insights of a professional coach. Poze combines pose estimation with sequence comparison and is optimized to function effectively with minimal data. Poze surpasses state-of-the-art vision-language models in video question-answering frameworks, achieving 70% and 196% increase in accuracy over GPT4V and LLaVAv1.6 7b, respectively.

Paperid: 7155, https://arxiv.org/pdf/2411.04077.pdf

Abstract:
By leveraging both texts and images, large vision language models (LVLMs) have shown significant progress in various multi-modal tasks. Nevertheless, these models often suffer from hallucinations, e.g., they exhibit inconsistencies between the visual input and the textual output. To address this, we propose H-POPE, a coarse-to-fine-grained benchmark that systematically assesses hallucination in object existence and attributes. Our evaluation shows that models are prone to hallucinations on object existence, and even more so on fine-grained attributes. We further investigate whether these models rely on visual input to formulate the output texts.

Paperid: 7156, https://arxiv.org/pdf/2410.20883.pdf

Abstract:
The cognitive faculty of visual reasoning necessitates the integration of multimodal perceptual processing and commonsense and external knowledge of the world. In recent years, a plethora of large vision-language models (LVLMs) have been proposed, demonstrating outstanding power and exceptional proficiency in commonsense reasoning across diverse domains and tasks. Nevertheless, training such LVLMs requires a lot of costly resources. Recent approaches, instead of training LVLMs from scratch on various large datasets, focus on exploring ways to take advantage of the capabilities of many different LVLMs, such as ensemble methods. In this work, we propose self-ensemble, a novel method that improves the generalization and visual reasoning of the model without updating any parameters, a training-free method. Our key insight is that we realized that LVLM itself can ensemble without the need for any other LVLMs, which helps to unlock their internal capabilities. Extensive experiments on various benchmarks demonstrate the effectiveness of our method in achieving state-of-the-art (SOTA) performance on SketchyVQA, Outside Knowledge VQA, and out-of-distribution VQA tasks.

Paperid: 7157, https://arxiv.org/pdf/2410.20666.pdf

Abstract:
Navigation presents a significant challenge for persons with visual impairments (PVI). While traditional aids such as white canes and guide dogs are invaluable, they fall short in delivering detailed spatial information and precise guidance to desired locations. Recent developments in large language models (LLMs) and vision-language models (VLMs) offer new avenues for enhancing assistive navigation. In this paper, we introduce Guide-LLM, an embodied LLM-based agent designed to assist PVI in navigating large indoor environments. Our approach features a novel text-based topological map that enables the LLM to plan global paths using a simplified environmental representation, focusing on straight paths and right-angle turns to facilitate navigation. Additionally, we utilize the LLM's commonsense reasoning for hazard detection and personalized path planning based on user preferences. Simulated experiments demonstrate the system's efficacy in guiding PVI, underscoring its potential as a significant advancement in assistive technology. The results highlight Guide-LLM's ability to offer efficient, adaptive, and personalized navigation assistance, pointing to promising advancements in this field.

Paperid: 7158, https://arxiv.org/pdf/2410.18537.pdf

Abstract:
Traditionally, style has been primarily considered in terms of artistic elements such as colors, brushstrokes, and lighting. However, identical semantic subjects, like people, boats, and houses, can vary significantly across different artistic traditions, indicating that style also encompasses the underlying semantics. Therefore, in this study, we propose a zero-shot scheme for image variation with coordinated semantics. Specifically, our scheme transforms the image-to-image problem into an image-to-text-to-image problem. The image-to-text operation employs vision-language models e.g., BLIP) to generate text describing the content of the input image, including the objects and their positions. Subsequently, the input style keyword is elaborated into a detailed description of this style and then merged with the content text using the reasoning capabilities of ChatGPT. Finally, the text-to-image operation utilizes a Diffusion model to generate images based on the text prompt. To enable the Diffusion model to accommodate more styles, we propose a fine-tuning strategy that injects text and style constraints into cross-attention. This ensures that the output image exhibits similar semantics in the desired style. To validate the performance of the proposed scheme, we constructed a benchmark comprising images of various styles and scenes and introduced two novel metrics. Despite its simplicity, our scheme yields highly plausible results in a zero-shot manner, particularly for generating stylized images with high-fidelity semantics.

Paperid: 7159, https://arxiv.org/pdf/2410.17149.pdf

Abstract:
Current vision-language foundation models, such as CLIP, have recently shown significant improvement in performance across various downstream tasks. However, whether such foundation models significantly improve more complex fine-grained action recognition tasks is still an open question. To answer this question and better find out the future research direction on human behavior analysis in-the-wild, this paper provides a large-scale study and insight on current state-of-the-art vision foundation models by comparing their transfer ability onto zero-shot and frame-wise action recognition tasks. Extensive experiments are conducted on recent fine-grained, human-centric action recognition datasets (e.g., Toyota Smarthome, Penn Action, UAV-Human, TSU, Charades) including action classification and segmentation.

Paperid: 7160, https://arxiv.org/pdf/2410.14911.pdf

Abstract:
The robustness of Vision-Language Models (VLMs) such as CLIP is critical for their deployment in safety-critical applications like autonomous driving, healthcare diagnostics, and security systems, where accurate interpretation of visual and textual data is essential. However, these models are highly susceptible to adversarial attacks, which can severely compromise their performance and reliability in real-world scenarios. Previous methods have primarily focused on improving robustness through adversarial training and generating adversarial examples using models like FGSM, AutoAttack, and DeepFool. However, these approaches often rely on strong assumptions, such as fixed perturbation norms or predefined attack patterns, and involve high computational complexity, making them challenging to implement in practical settings. In this paper, we propose a novel adversarial training framework that integrates multiple attack strategies and advanced machine learning techniques to significantly enhance the robustness of VLMs against a broad range of adversarial attacks. Experiments conducted on real-world datasets, including CIFAR-10 and CIFAR-100, demonstrate that the proposed method significantly enhances model robustness. The fine-tuned CLIP model achieved an accuracy of 43.5% on adversarially perturbed images, compared to only 4% for the baseline model. The neural network model achieved a high accuracy of 98% in these challenging classification tasks, while the XGBoost model reached a success rate of 85.26% in prediction tasks.

Paperid: 7161, https://arxiv.org/pdf/2410.14509.pdf

Abstract:
Voice Activity Detection (VAD) is the process of automatically determining whether a person is speaking and identifying the timing of their speech in an audiovisual data. Traditionally, this task has been tackled by processing either audio signals or visual data, or by combining both modalities through fusion or joint learning. In our study, drawing inspiration from recent advancements in visual-language models, we introduce a novel approach leveraging Contrastive Language-Image Pretraining (CLIP) models. The CLIP visual encoder analyzes video segments composed of the upper body of an individual, while the text encoder handles textual descriptions automatically generated through prompt engineering. Subsequently, embeddings from these encoders are fused through a deep neural network to perform VAD. Our experimental analysis across three VAD benchmarks showcases the superior performance of our method compared to existing visual VAD approaches. Notably, our approach outperforms several audio-visual methods despite its simplicity, and without requiring pre-training on extensive audio-visual datasets.

Paperid: 7162, https://arxiv.org/pdf/2410.13989.pdf

Abstract:
The growing reproducibility crisis in machine learning has brought forward a need for careful examination of research findings. This paper investigates the claims made by Lei et al. (2023) regarding their proposed method, LICO, for enhancing post-hoc interpretability techniques and improving image classification performance. LICO leverages natural language supervision from a vision-language model to enrich feature representations and guide the learning process. We conduct a comprehensive reproducibility study, employing (Wide) ResNets and established interpretability methods like Grad-CAM and RISE. We were mostly unable to reproduce the authors' results. In particular, we did not find that LICO consistently led to improved classification performance or improvements in quantitative and qualitative measures of interpretability. Thus, our findings highlight the importance of rigorous evaluation and transparent reporting in interpretability research.

Paperid: 7163, https://arxiv.org/pdf/2410.13883.pdf

Abstract:
In recent years, interest in vision-language tasks has grown, especially those involving chart interactions. These tasks are inherently multimodal, requiring models to process chart images, accompanying text, underlying data tables, and often user queries. Traditionally, Chart Understanding (CU) relied on heuristics and rule-based systems. However, recent advancements that have integrated transformer architectures significantly improved performance. This paper reviews prominent research in CU, focusing on State-of-The-Art (SoTA) frameworks that employ transformers within End-to-End (E2E) solutions. Relevant benchmarking datasets and evaluation techniques are analyzed. Additionally, this article identifies key challenges and outlines promising future directions for advancing CU solutions. Following the PRISMA guidelines, a comprehensive literature search is conducted across Google Scholar, focusing on publications from Jan'20 to Jun'24. After rigorous screening and quality assessment, 32 studies are selected for in-depth analysis. The CU tasks are categorized into a three-layered paradigm based on the cognitive task required. Recent advancements in the frameworks addressing various CU tasks are also reviewed. Frameworks are categorized into single-task or multi-task based on the number of tasks solvable by the E2E solution. Within multi-task frameworks, pre-trained and prompt-engineering-based techniques are explored. This review overviews leading architectures, datasets, and pre-training tasks. Despite significant progress, challenges remain in OCR dependency, handling low-resolution images, and enhancing visual reasoning. Future directions include addressing these challenges, developing robust benchmarks, and optimizing model efficiency. Additionally, integrating explainable AI techniques and exploring the balance between real and synthetic data are crucial for advancing CU research.

Paperid: 7164, https://arxiv.org/pdf/2410.09750.pdf

Abstract:
Conversation agents powered by large language models are revolutionizing the way we interact with visual data. Recently, large vision-language models (LVLMs) have been extensively studied for both images and videos. However, these studies typically focus on common scenarios. In this work, we introduce an LVLM specifically designed for surgical scenarios. We integrate visual representations of surgical images and videos into the language feature space. Consequently, we establish a LVLM model, Surgical-LLaVA, fine-tuned on instruction following data of surgical scenarios. Our experiments demonstrate that Surgical-LLaVA exhibits impressive multi-modal chat abilities in surgical contexts, occasionally displaying multi-modal behaviors on unseen instructions. We conduct a quantitative evaluation of visual question-answering datasets for surgical scenarios. The results show superior performance compared to previous works, indicating the potential of our model to tackle more complex surgery scenarios.

Paperid: 7165, https://arxiv.org/pdf/2410.07593.pdf

Abstract:
Recent advancements in Vision-Language Models (VLMs) have enabled complex multimodal tasks by processing text and image data simultaneously, significantly enhancing the field of artificial intelligence. However, these models often exhibit biases that can skew outputs towards societal stereotypes, thus necessitating debiasing strategies. Existing debiasing methods focus narrowly on specific modalities or tasks, and require extensive retraining. To address these limitations, this paper introduces Selective Feature Imputation for Debiasing (SFID), a novel methodology that integrates feature pruning and low confidence imputation (LCI) to effectively reduce biases in VLMs. SFID is versatile, maintaining the semantic integrity of outputs and costly effective by eliminating the need for retraining. Our experimental results demonstrate SFID's effectiveness across various VLMs tasks including zero-shot classification, text-to-image retrieval, image captioning, and text-to-image generation, by significantly reducing gender biases without compromising performance. This approach not only enhances the fairness of VLMs applications but also preserves their efficiency and utility across diverse scenarios.

Paperid: 7166, https://arxiv.org/pdf/2410.05450.pdf

Abstract:
Major Depressive Disorder and anxiety disorders affect millions globally, contributing significantly to the burden of mental health issues. Early screening is crucial for effective intervention, as timely identification of mental health issues can significantly improve treatment outcomes. Artificial intelligence (AI) can be valuable for improving the screening of mental disorders, enabling early intervention and better treatment outcomes. AI-driven screening can leverage the analysis of multiple data sources, including facial features in digital images. However, existing methods often rely on controlled environments or specialized equipment, limiting their broad applicability. This study explores the potential of AI models for ubiquitous depression-anxiety screening given face-centric selfies. The investigation focuses on high-risk pregnant patients, a population that is particularly vulnerable to mental health issues. To cope with limited training data resulting from our clinical setup, pre-trained models were utilized in two different approaches: fine-tuning convolutional neural networks (CNNs) originally designed for facial expression recognition and employing vision-language models (VLMs) for zero-shot analysis of facial expressions. Experimental results indicate that the proposed VLM-based method significantly outperforms CNNs, achieving an accuracy of 77.6%. Although there is significant room for improvement, the results suggest that VLMs can be a promising approach for mental health screening.

Paperid: 7167, https://arxiv.org/pdf/2410.03900.pdf

Abstract:
We study the task of locating a user in a mapped indoor environment using natural language queries and images from the environment. Building on recent pretrained vision-language models, we learn a similarity score between text descriptions and images of locations in the environment. This score allows us to identify locations that best match the language query, estimating the user's location. Our approach is capable of localizing on environments, text, and images that were not seen during training. One model, finetuned CLIP, outperformed humans in our evaluation.

Paperid: 7168, https://arxiv.org/pdf/2410.03551.pdf

Abstract:
This study reveals an unexpected parallel between instructible vision-language models (VLMs) and human cognitive disorders, specifically constructive apraxia. We tested 25 state-of-the-art VLMs, including GPT-4 Vision, DALL-E 3, and Midjourney v5, on their ability to generate images of the Ponzo illusion, a task that requires basic spatial reasoning and is often used in clinical assessments of constructive apraxia. Remarkably, 24 out of 25 models failed to correctly render two horizontal lines against a perspective background, mirroring the deficits seen in patients with parietal lobe damage. The models consistently misinterpreted spatial instructions, producing tilted or misaligned lines that followed the perspective of the background rather than remaining horizontal. This behavior is strikingly similar to how apraxia patients struggle to copy or construct simple figures despite intact visual perception and motor skills. Our findings suggest that current VLMs, despite their advanced capabilities in other domains, lack fundamental spatial reasoning abilities akin to those impaired in constructive apraxia. This limitation in AI systems provides a novel computational model for studying spatial cognition deficits and highlights a critical area for improvement in VLM architecture and training methodologies.

Paperid: 7169, https://arxiv.org/pdf/2409.17641.pdf

Abstract:
Active perception enables robots to dynamically gather information by adjusting their viewpoints, a crucial capability for interacting with complex, partially observable environments. In this paper, we present AP-VLM, a novel framework that combines active perception with a Vision-Language Model (VLM) to guide robotic exploration and answer semantic queries. Using a 3D virtual grid overlaid on the scene and orientation adjustments, AP-VLM allows a robotic manipulator to intelligently select optimal viewpoints and orientations to resolve challenging tasks, such as identifying objects in occluded or inclined positions. We evaluate our system on two robotic platforms: a 7-DOF Franka Panda and a 6-DOF UR5, across various scenes with differing object configurations. Our results demonstrate that AP-VLM significantly outperforms passive perception methods and baseline models, including Toward Grounded Common Sense Reasoning (TGCSR), particularly in scenarios where fixed camera views are inadequate. The adaptability of AP-VLM in real-world settings shows promise for enhancing robotic systems' understanding of complex environments, bridging the gap between high-level semantic reasoning and low-level control.

Paperid: 7170, https://arxiv.org/pdf/2409.16024.pdf

Abstract:
Vision-language models (VLMs) have tremendous potential for grounding language, and thus enabling language-conditioned agents (LCAs) to perform diverse tasks specified with text. This has motivated the study of LCAs based on reinforcement learning (RL) with rewards given by rendering images of an environment and evaluating those images with VLMs. If single-task RL is employed, such approaches are limited by the cost and time required to train a policy for each new task. Multi-task RL (MTRL) is a natural alternative, but requires a carefully designed corpus of training tasks and does not always generalize reliably to new tasks. Therefore, this paper introduces a novel decomposition of the problem of building an LCA: first find an environment configuration that has a high VLM score for text describing a task; then use a (pretrained) goal-conditioned policy to reach that configuration. We also explore several enhancements to the speed and quality of VLM-based LCAs, notably, the use of distilled models, and the evaluation of configurations from multiple viewpoints to resolve the ambiguities inherent in a single 2D view. We demonstrate our approach on the Humanoid environment, showing that it results in LCAs that outperform MTRL baselines in zero-shot generalization, without requiring any textual task descriptions or other forms of environment-specific annotation during training. Videos and an interactive demo can be found at https://europe.naverlabs.com/text2control

Paperid: 7171, https://arxiv.org/pdf/2409.11007.pdf

Abstract:
Vision Language Models (VLMs) are typically evaluated with Visual Question Answering (VQA) tasks which assess a model's understanding of scenes. Good VQA performance is taken as evidence that the model will perform well on a broader range of tasks that require both visual and language inputs. However, scene-aware VQA does not fully capture input biases or assess hallucinations caused by a misalignment between modalities. To address this, we propose a Cross-modal Alignment Similarity Test (CAST) to probe VLMs for self-consistency across modalities. This test involves asking the models to identify similarities between two scenes through text-only, image-only, or both and then assess the truthfulness of the similarities they generate. Since there is no ground-truth to compare against, this evaluation does not focus on objective accuracy but rather on whether VLMs are internally consistent in their outputs. We argue that while not all self-consistent models are capable or accurate, all capable VLMs must be self-consistent.

Paperid: 7172, https://arxiv.org/pdf/2409.06853.pdf

Abstract:
Blind Image Quality Assessment (BIQA) aims to develop methods that estimate the quality scores of images in the absence of a reference image. In this paper, we approach BIQA from a distortion identification perspective, where our primary goal is to predict distortion types and strengths using Vision-Language Models (VLMs), such as CLIP, due to their extensive knowledge and generalizability. Based on these predicted distortions, we then estimate the quality score of the image. To achieve this, we propose an explainable approach for distortion identification based on attribute learning. Instead of prompting VLMs with the names of distortions, we prompt them with the attributes or effects of distortions and aggregate this information to infer the distortion strength. Additionally, we consider multiple distortions per image, making our method more scalable. To support this, we generate a dataset consisting of 100,000 images for efficient training. Finally, attribute probabilities are retrieved and fed into a regressor to predict the image quality score. The results show that our approach, besides its explainability and transparency, achieves state-of-the-art (SOTA) performance across multiple datasets in both PLCC and SRCC metrics. Moreover, the zero-shot results demonstrate the generalizability of the proposed approach.

Paperid: 7173, https://arxiv.org/pdf/2601.17357.pdf

Abstract:
Large language models and deep neural networks achieve strong performance but suffer from reliability issues and high computational cost. This thesis proposes a unified framework based on spectral geometry and random matrix theory to address both problems by analyzing the eigenvalue structure of hidden activations. The first contribution, EigenTrack, is a real-time method for detecting hallucinations and out-of-distribution behavior in language and vision-language models using spectral features and their temporal dynamics. The second contribution, RMT-KD, is a principled compression method that identifies informative spectral components and applies iterative knowledge distillation to produce compact and efficient models while preserving accuracy. Together, these results show that spectral statistics provide interpretable and robust signals for monitoring uncertainty and guiding compression in large-scale neural networks.

Paperid: 7174, https://arxiv.org/pdf/2601.11637.pdf

Abstract:
Recent progress in multimodal foundation models has enabled Vision-Language Agents (VLAs) to decompose complex visual tasks into executable tool-based plans. While recent benchmarks have begun to evaluate iterative self-correction, its quantitative limits and dominant reasoning bottlenecks remain poorly characterized. This work introduces a Diagnostic Micro-Benchmark. Our analysis decouples Task Success Rate (TSR = 62 percent) from Correction Success Rate (CSR = 25 to 33 percent), revealing that initial competence does not predict repair ability. We explicitly quantify the diminishing returns of correction, which saturates after three retries. Our Failure Taxonomy reveals a frequent factor is Semantic Drift (about 28 percent of failures), a loss of contextual state. By isolating this reasoning bottleneck, this benchmark defines a reproducible framework toward stateful, trustworthy multimodal agents.

Paperid: 7175, https://arxiv.org/pdf/2601.08860.pdf

Abstract:
Vision-Language Models (VLMs) and generative image models have achieved remarkable performance across multimodal tasks, yet their robustness and fairness under input transformations remain insufficiently explored. This work investigates bias propagation and robustness degradation in state-of-the-art vision-language and generative models, with a particular focus on image rotation and distributional shifts. We analyze how rotation-induced perturbations affect model predictions, confidence calibration, and demographic bias patterns. To address these issues, we propose rotation-robust mitigation strategies that combine data augmentation, representation alignment, and model-level regularization. Experimental results across multiple datasets demonstrate that the proposed methods significantly improve robustness while reducing bias amplification without sacrificing overall performance. This study highlights critical limitations of current multimodal systems and provides practical mitigation techniques for building more reliable and fair AI models.

Paperid: 7176, https://arxiv.org/pdf/2601.08355.pdf

Abstract:
Vision-Language Models (VLMs) are increasingly deployed in autonomous driving and embodied AI systems, where reliable perception is critical for safe semantic reasoning and decision-making. While recent VLMs demonstrate strong performance on multimodal benchmarks, their robustness to realistic perception degradation remains poorly understood. In this work, we systematically study semantic misalignment in VLMs under controlled degradation of upstream visual perception, using semantic segmentation on the Cityscapes dataset as a representative perception module. We introduce perception-realistic corruptions that induce only moderate drops in conventional segmentation metrics, yet observe severe failures in downstream VLM behavior, including hallucinated object mentions, omission of safety-critical entities, and inconsistent safety judgments. To quantify these effects, we propose a set of language-level misalignment metrics that capture hallucination, critical omission, and safety misinterpretation, and analyze their relationship with segmentation quality across multiple contrastive and generative VLMs. Our results reveal a clear disconnect between pixel-level robustness and multimodal semantic reliability, highlighting a critical limitation of current VLM-based systems and motivating the need for evaluation frameworks that explicitly account for perception uncertainty in safety-critical applications.

Paperid: 7177, https://arxiv.org/pdf/2601.05344.pdf

Abstract:
The ability to construct mental models of the world is a central aspect of understanding. Similarly, visual understanding can be viewed as the ability to construct a representative model of the system depicted in an image. This work explores the capacity of Vision Language Models (VLMs) to recognize and simulate the systems and mechanisms depicted in images using the Im2Sim methodology. The VLM is given a natural image of a real-world system (e.g., cities, clouds, vegetation) and is tasked with describing the system and writing code that simulates and generates it. This generative code is then executed to produce a synthetic image, which is compared against the original. This approach is tested on various complex emergent systems, ranging from physical systems (waves, lights, clouds) to vegetation, cities, materials, and geological formations. Through analysis of the models and images generated by the VLMs, we examine their understanding of the systems in images. The results show that leading VLMs (GPT, Gemini) have the ability to understand and model complex, multi-component systems across multiple layers of abstraction and a wide range of domains. At the same time, the VLMs exhibit limited ability to replicate fine details and low-level arrangements of patterns in the image. These findings reveal an interesting asymmetry: VLMs combine high-level, deep visual understanding of images with limited perception of fine details.

Paperid: 7178, https://arxiv.org/pdf/2601.03329.pdf

Abstract:
Attention mechanisms represent a fundamental paradigm shift in neural network architectures, enabling models to selectively focus on relevant portions of input sequences through learned weighting functions. This monograph provides a comprehensive and rigorous mathematical treatment of attention mechanisms, encompassing their theoretical foundations, computational properties, and practical implementations in contemporary deep learning systems. Applications in natural language processing, computer vision, and multimodal learning demonstrate the versatility of attention mechanisms. We examine language modeling with autoregressive transformers, bidirectional encoders for representation learning, sequence-to-sequence translation, Vision Transformers for image classification, and cross-modal attention for vision-language tasks. Empirical analysis reveals training characteristics, scaling laws that relate performance to model size and computation, attention pattern visualizations, and performance benchmarks across standard datasets. We discuss the interpretability of learned attention patterns and their relationship to linguistic and visual structures. The monograph concludes with a critical examination of current limitations, including computational scalability, data efficiency, systematic generalization, and interpretability challenges.

Paperid: 7179, https://arxiv.org/pdf/2601.03286.pdf

Abstract:
In this report, we present HyperCLOVA X 32B Think, a vision-language model designed with particular emphasis on reasoning within the Korean linguistic and cultural context, as well as agentic ability. HyperCLOVA X 32B Think is pre-trained with a strong focus on reasoning capabilities and subsequently post-trained to support multimodal understanding, enhanced reasoning, agentic behaviors, and alignment with human preferences. Experimental evaluations against comparably sized models demonstrate that our model achieves strong performance on Korean text-to-text and vision-to-text benchmarks, as well as on agent-oriented evaluation tasks. By open-sourcing HyperCLOVA X 32B Think, we aim to support broader adoption and facilitate further research and innovation across both academic and industrial communities.

Paperid: 7180, https://arxiv.org/pdf/2601.01062.pdf

Abstract:
Vision-Language Models (VLMs) have achieved remarkable success in descriptive tasks such as image captioning and visual question answering (VQA). However, their ability to generate engaging, long-form narratives -- specifically multi-speaker podcast dialogues -- remains under-explored and difficult to evaluate. Standard metrics like BLEU and ROUGE fail to capture the nuances of conversational naturalness, personality, and narrative flow, often rewarding safe, repetitive outputs over engaging storytelling. In this work, we present a novel pipeline for end-to-end visual podcast generation, and fine-tune a Qwen3-VL-32B model on a curated dataset of 4,000 image-dialogue pairs. Crucially, we use a synthetic-to-real training strategy: we train on high-quality podcast dialogues from the Structured Podcast Research Corpus (SPoRC) paired with synthetically generated imagery, and evaluate on real-world photo sequences from the Visual Storytelling Dataset (VIST). This rigorous setup tests the model's ability to generalize from synthetic training data to real-world visual domains. We propose a comprehensive evaluation framework that moves beyond textual overlap, and use AI-as-a-judge (Gemini 3 Pro, Claude Opus 4.5, GPT 5.2) and novel style metrics (average turn length, speaker switch rate) to assess quality. Our experiments demonstrate that our fine-tuned 32B model significantly outperforms a 235B base model in conversational naturalness ($>$80\% win rate) and narrative depth (+50\% turn length), while maintaining identical visual grounding capabilities (CLIPScore: 20.39).

Paperid: 7181, https://arxiv.org/pdf/2601.00138.pdf

Abstract:
High-stakes deployment of vision-language models (VLMs) requires selective prediction, where systems abstain when uncertain rather than risk costly errors. We investigate whether confidence-based abstention provides reliable control over error rates in video question answering, and whether that control remains robust under distribution shift. Using NExT-QA and Gemini 2.0 Flash, we establish two findings. First, confidence thresholding provides mechanistic control in-distribution. Sweeping threshold epsilon produces smooth risk-coverage tradeoffs, reducing error rates f

Paperid: 7182, https://arxiv.org/pdf/2512.23524.pdf

Abstract:
Machine Learning (ML) has been a foundational topic in artificial intelligence (AI), providing both theoretical groundwork and practical tools for its exciting advancements. From ResNet for visual recognition to Transformer for vision-language alignment, the AI models have achieved superior capability to humans. Furthermore, the scaling law has enabled AI to initially develop general intelligence, as demonstrated by Large Language Models (LLMs). To this stage, AI has had an enormous influence on society and yet still keeps shaping the future for humanity. However, distribution shift remains a persistent ``Achilles' heel'', fundamentally limiting the reliability and general usefulness of ML systems. Moreover, generalization under distribution shift would also cause trust issues for AIs. Motivated by these challenges, my research focuses on \textit{Trustworthy Machine Learning under Distribution Shifts}, with the goal of expanding AI's robustness, versatility, as well as its responsibility and reliability. We carefully study the three common distribution shifts into: (1) Perturbation Shift, (2) Domain Shift, and (3) Modality Shift. For all scenarios, we also rigorously investigate trustworthiness via three aspects: (1) Robustness, (2) Explainability, and (3) Adaptability. Based on these dimensions, we propose effective solutions and fundamental insights, meanwhile aiming to enhance the critical ML problems, such as efficiency, adaptability, and safety.

Paperid: 7183, https://arxiv.org/pdf/2512.23109.pdf

Abstract:
Modern generative and vision-language models (VLMs) are increasingly used in scientific and medical decision support, where predicted probabilities must be both accurate and well calibrated. Despite strong empirical results with moderate data, it remains unclear when such predictions generalize uniformly across inputs, classes, or subpopulations, rather than only on average-a critical issue in biomedicine, where rare conditions and specific groups can exhibit large errors even when overall loss is low. We study this question from a finite-sample perspective and ask: under what structural assumptions can generative and VLM-based predictors achieve uniformly accurate and calibrated behavior with practical sample sizes? Rather than analyzing arbitrary parameterizations, we focus on induced families of classifiers obtained by varying prompts or semantic embeddings within a restricted representation space. When model outputs depend smoothly on a low-dimensional semantic representation-an assumption supported by spectral structure in text and joint image-text embeddings-classical uniform convergence tools yield meaningful non-asymptotic guarantees. Our main results give finite-sample uniform convergence bounds for accuracy and calibration functionals of VLM-induced classifiers under Lipschitz stability with respect to prompt embeddings. The implied sample complexity depends on intrinsic/effective dimension, not ambient embedding dimension, and we further derive spectrum-dependent bounds that make explicit how eigenvalue decay governs data requirements. We conclude with implications for data-limited biomedical modeling, including when current dataset sizes can support uniformly reliable predictions and why average calibration metrics may miss worst-case miscalibration.

Paperid: 7184, https://arxiv.org/pdf/2512.09319.pdf

Abstract:
Industrial cyber physical systems operate under heterogeneous sensing, stochastic dynamics, and shifting process conditions, producing data that are often incomplete, unlabeled, imbalanced, and domain shifted. High-fidelity datasets remain costly, confidential, and slow to obtain, while edge devices face strict limits on latency, bandwidth, and energy. These factors restrict the practicality of centralized deep learning, hinder the development of reliable digital twins, and increase the risk of error escape in safety-critical applications. Motivated by these challenges, this dissertation develops an efficiency grounded computational framework that enables data lean, physics-aware, and deployment ready intelligence for modern manufacturing environments. The research advances methods that collectively address core bottlenecks across multimodal and multiscale industrial scenarios. Generative strategies mitigate data scarcity and imbalance, while semi-supervised learning integrates unlabeled information to reduce annotation and simulation demands. Physics-informed representation learning strengthens interpretability and improves condition monitoring under small-data regimes. Spatially aware graph-based surrogate modeling provides efficient approximation of complex processes, and an edge cloud collaborative compression scheme supports real-time signal analytics under resource constraints. The dissertation also extends visual understanding through zero-shot vision language reasoning augmented by domain specific retrieval, enabling generalizable assessment in previously unseen scenarios. Together, these developments establish a unified paradigm of data efficient and resource aware intelligence that bridges laboratory learning with industrial deployment, supporting reliable decision-making across diverse manufacturing systems.

Paperid: 7185, https://arxiv.org/pdf/2512.06017.pdf

Abstract:
Pose estimation of a robot arm from visual inputs is a challenging task. However, with the increasing adoption of robot arms for both industrial and residential use cases, reliable joint angle estimation can offer improved safety and performance guarantees, and also be used as a verifier to further train robot policies. This paper introduces using frontier vision-language models (VLMs) as an ``off-the-shelf" tool to estimate a robot arm's joint angles from a single target image. By evaluating frontier VLMs on both synthetic and real-world image-data pairs, this paper establishes a performance baseline attained by current FLMs. In addition, this paper presents empirical results suggesting that test time scaling or parameter scaling alone does not lead to improved joint angle predictions.

Paperid: 7186, https://arxiv.org/pdf/2512.02520.pdf

Abstract:
Zero-shot anomaly classification and segmentation (AC/AS) aim to detect anomalous samples and regions without any training data, a capability increasingly crucial in industrial inspection and medical imaging. This dissertation aims to investigate the core challenges of zero-shot AC/AS and presents principled solutions rooted in theory and algorithmic design. We first formalize the problem of consistent anomalies, a failure mode in which recurring similar anomalies systematically bias distance-based methods. By analyzing the statistical and geometric behavior of patch representations from pre-trained Vision Transformers, we identify two key phenomena - similarity scaling and neighbor-burnout - that describe how relationships among normal patches change with and without consistent anomalies in settings characterized by highly similar objects. We then introduce CoDeGraph, a graph-based framework for filtering consistent anomalies built on the similarity scaling and neighbor-burnout phenomena. Through multi-stage graph construction, community detection, and structured refinement, CoDeGraph effectively suppresses the influence of consistent anomalies. Next, we extend this framework to 3D medical imaging by proposing a training-free, computationally efficient volumetric tokenization strategy for MRI data. This enables a genuinely zero-shot 3D anomaly detection pipeline and shows that volumetric anomaly segmentation is achievable without any 3D training samples. Finally, we bridge batch-based and text-based zero-shot methods by demonstrating that CoDeGraph-derived pseudo-masks can supervise prompt-driven vision-language models. Together, this dissertation provides theoretical understanding and practical solutions for the zero-shot AC/AS problem.

Paperid: 7187, https://arxiv.org/pdf/2512.00882.pdf

Abstract:
Vision-Language Models (VLMs) exhibit significant performance plateaus in specialized domains like precision agriculture, primarily due to "Reasoning-Driven Hallucination" where linguistic priors override visual perception. A key bottleneck is the "Modality Gap": visual embeddings fail to reliably activate the fine-grained expert knowledge already encoded in model parameters. We propose "Look, Recite, Then Answer," a parameter-efficient framework that enhances VLMs via self-generated knowledge hints while keeping backbone models frozen. The framework decouples inference into three stages: (1) Look generates objective visual descriptions and candidate sets; (2) Recite employs a lightweight 1.7B router to transform visual cues into targeted queries that trigger candidate-specific parametric knowledge; (3) Answer performs parallel evidence alignment between descriptions and recited knowledge to select the most consistent label. On AgroBench, our method achieves state-of-the-art results, improving Weed Identification accuracy by 23.52% over Qwen2-VL-72B and surpassing GPT-4o without external search overhead. This modular design mitigates hallucinations by transforming passive perception into active, controllable knowledge retrieval

Paperid: 7188, https://arxiv.org/pdf/2512.00783.pdf

Abstract:
To address the gap in humanoid robot cognitive systems regarding the lack of a time-updable mediating thought space between semantics and continuous control, this study constructs and trains a VLA model named "Sigma" that runs on a single RTX 4090. It uses the open-source pi05_base model as a foundation and preprocesses svla_so101_pickplace into a training dataset. The researcher independently designed an architecture for a vision-language-action model that combines deep semantic understanding and association to achieve telepathic communication. The training process involved repeated optimizations of data preprocessing, LoRA fine-tuning, and the inference-stage adapter. The experiment employed offline closed-loop replay, comparing Sigma with the untuned pure pi05_base model under data conditions. Results showed that Sigma exhibited a stable decrease in control MSE across vector, fragment, and entire trajectory timescales, while maintaining the telepathy norm and semantic-text alignment quality unchanged. It demonstrates that mind-responsive alignment control is quantified through an architecture that combines deep understanding of semantics and association without retraining the base model, which provides reproducible experience for semantic alignment and intention-driven behavior in humanoid robots.

Paperid: 7189, https://arxiv.org/pdf/2511.22351.pdf

Abstract:
The growing realism of AI-generated images produced by recent GAN and diffusion models has intensified concerns over the reliability of visual media. Yet, despite notable progress in deepfake detection, current forensic systems degrade sharply under real-world conditions such as severe downsampling, compression, and cross-domain distribution shifts. Moreover, most detectors operate as opaque classifiers, offering little insight into why an image is flagged as synthetic, undermining trust and hindering adoption in high-stakes settings. We introduce INSIGHT (Interpretable Neural Semantic and Image-based Generative-forensic Hallucination Tracing), a unified multimodal framework for robust detection and transparent explanation of AI-generated images, even at extremely low resolutions (16x16 - 64x64). INSIGHT combines hierarchical super-resolution for amplifying subtle forensic cues without inducing misleading artifacts, Grad-CAM driven multi-scale localization to reveal spatial regions indicative of generative patterns, and CLIP-guided semantic alignment to map visual anomalies to human-interpretable descriptors. A vision-language model is then prompted using a structured ReAct + Chain-of-Thought protocol to produce consistent, fine-grained explanations, verified through a dual-stage G-Eval + LLM-as-a-judge pipeline to minimize hallucinations and ensure factuality. Across diverse domains, including animals, vehicles, and abstract synthetic scenes, INSIGHT substantially improves both detection robustness and explanation quality under extreme degradation, outperforming prior detectors and black-box VLM baselines. Our results highlight a practical path toward transparent, reliable AI-generated image forensics and establish INSIGHT as a step forward in trustworthy multimodal content verification.

Paperid: 7190, https://arxiv.org/pdf/2511.20531.pdf

Abstract:
Visual Language Models (VLMs) are powerful generative tools but often produce factually inaccurate outputs due to a lack of robust reasoning capabilities. While extensive research has been conducted on integrating external knowledge for reasoning in large language models (LLMs), such efforts remain underexplored in VLMs, where the challenge is compounded by the need to bridge multiple modalities seamlessly. This work introduces a framework for knowledge-guided reasoning in VLMs, leveraging structured knowledge graphs for multi-hop verification using image-captioning task to illustrate our framework. Our approach enables systematic reasoning across multiple steps, including visual entity recognition, knowledge graph traversal, and fact-based caption refinement. We evaluate the framework using hierarchical, triple-based and bullet-point based knowledge representations, analyzing their effectiveness in factual accuracy and logical inference. Empirical results show that our approach improves factual accuracy by approximately 31% on preliminary experiments on a curated dataset of mixtures from Google Landmarks v2, Conceptual captions and Coco captions revealing key insights into reasoning patterns and failure modes. This work demonstrates the potential of integrating external knowledge for advancing reasoning in VLMs, paving the way for more reliable and knowledgable multimodal systems.

Paperid: 7191, https://arxiv.org/pdf/2511.18272.pdf

Abstract:
Large vision-language models (VLMs) are increasingly deployed for optical character recognition (OCR) in healthcare settings, raising critical concerns about protected health information (PHI) exposure during document processing. This work presents the first systematic evaluation of inference-time vision token masking as a privacy-preserving mechanism for medical document OCR using DeepSeek-OCR. We introduce seven masking strategies (V3-V9) targeting different architectural layers (SAM encoder blocks, compression layers, dual vision encoders, projector fusion) and evaluate PHI reduction across HIPAA-defined categories using 100 synthetic medical billing statements (drawn from a corpus of 38,517 annotated documents) with perfect ground-truth annotations. All masking strategies converge to 42.9% PHI reduction, successfully suppressing long-form spatially-distributed identifiers (patient names, dates of birth, physical addresses at 100% effectiveness) while failing to prevent short structured identifiers (medical record numbers, social security numbers, email addresses, account numbers at 0% effectiveness). Ablation studies varying mask expansion radius (r=1,2,3) demonstrate that increased spatial coverage does not improve reduction beyond this ceiling, indicating that language model contextual inference - not insufficient visual masking - drives structured identifier leakage. A simulated hybrid architecture combining vision masking with NLP post-processing achieves 88.6% total PHI reduction (assuming 80% NLP accuracy on remaining identifiers). This negative result establishes boundaries for vision-only privacy interventions in VLMs, provides guidance distinguishing PHI types amenable to vision-level versus language-level redaction, and redirects future research toward decoder-level fine-tuning and hybrid defense-in-depth architectures for HIPAA-compliant medical document processing.

Paperid: 7192, https://arxiv.org/pdf/2511.17668.pdf

Abstract:
Medical vision-language segmentation models suffer from catastrophic forgetting when adapting to new anatomical structures, requiring complete retraining that limits their clinical deployment. Although continual learning approaches have been studied for various applications, targeted research on continual learning approaches specifically designed for medical vision-language tasks remains underexplored. We propose MedPEFT-CL, a parameter-efficient continual learning framework that addresses both efficient learning of new tasks and preservation of previous knowledge through a dual-phase architecture based on CLIPSeg. Our dual-phase architecture features an adaptive learning phase that employs semantic similarity-based adapter allocation and parameter-efficient fine-tuning for medical tasks through prompt similarity analysis, and a knowledge consolidation phase employing bi-directional Fisher-memory coordination. This creates a reinforcing cycle: consolidation directs replay priorities while new tasks provide challenging samples that improve retention strategies. Our key contributions are: (1) a semantic-driven adapter allocation mechanism that enables efficient learning of new medical tasks, (2) a bi-modal LoRA adaptation that significantly reduces trainable parameters while maintaining cross-modal learning, and (3) bidirectional Fisher-memory coordination that prevents catastrophic forgetting from previous medical tasks. Extensive experiments across diverse medical datasets demonstrate superior forgetting mitigation and performance retention with minimal parameter overhead, making the framework effective for continual learning in medical vision-language scenarios.

Paperid: 7193, https://arxiv.org/pdf/2511.17036.pdf

Abstract:
Recent advances in vision-language models (VLMs) have enabled impressive multi-modal reasoning and understanding. Yet, whether these models truly grasp visual persuasion-how visual cues shape human attitudes and decisions-remains unclear. To probe this question, we construct a high-consensus dataset for binary persuasiveness judgment and introduce the taxonomy of Visual Persuasive Factors (VPFs), encompassing low-level perceptual, mid-level compositional, and high-level semantic cues. We also explore cognitive steering and knowledge injection strategies for persuasion-relevant reasoning. Empirical analysis across VLMs reveals a recall-oriented bias-models over-predict high persuasiveness-and weak discriminative power for low/mid-level features. In contrast, high-level semantic alignment between message and object presence emerges as the strongest predictor of human judgment. Among intervention strategies, simple instruction or unguided reasoning scaffolds yield marginal or negative effects, whereas concise, object-grounded rationales significantly improve precision and F1 scores. These results indicate that VLMs core limitation lies not in recognizing persuasive objects but in linking them to communicative intent.

Paperid: 7194, https://arxiv.org/pdf/2511.15720.pdf

Abstract:
This thesis explores a multimodal AI framework for enhancing construction safety through the combined analysis of textual and visual data. In safety-critical environments such as construction sites, accident data often exists in multiple formats, such as written reports, inspection records, and site imagery, making it challenging to synthesize hazards using traditional approaches. To address this, this thesis proposed a multimodal AI framework that combines text and image analysis to assist in identifying safety hazards on construction sites. Two case studies were consucted to evaluate the capabilities of large language models (LLMs) and vision-language models (VLMs) for automated hazard identification.The first case study introduces a hybrid pipeline that utilizes GPT 4o and GPT 4o mini to extract structured insights from a dataset of 28,000 OSHA accident reports (2000-2025). The second case study extends this investigation using Molmo 7B and Qwen2 VL 2B, lightweight, open-source VLMs. Using the public ConstructionSite10k dataset, the performance of the two models was evaluated on rule-level safety violation detection using natural language prompts. This experiment served as a cost-aware benchmark against proprietary models and allowed testing at scale with ground-truth labels. Despite their smaller size, Molmo 7B and Quen2 VL 2B showed competitive performance in certain prompt configurations, reinforcing the feasibility of low-resource multimodal systems for rule-aware safety monitoring.

Paperid: 7195, https://arxiv.org/pdf/2511.15383.pdf

Abstract:
Aircraft Maintenance Technicians (AMTs) spend up to 30% of work time searching manuals, a documented efficiency bottleneck in MRO operations where every procedure must be traceable to certified sources. We present a compliance-preserving retrieval system that adapts LLM reranking and semantic search to aviation MRO environments by operating alongside, rather than replacing, certified legacy viewers. The system constructs revision-robust embeddings from ATA chapter hierarchies and uses vision-language parsing to structure certified content, allowing technicians to preview ranked tasks and access verified procedures in existing viewers. Evaluation on 49k synthetic queries achieves >90% retrieval accuracy, while bilingual controlled studies with 10 licensed AMTs demonstrate 90.9% top-10 success rate and 95% reduction in lookup time, from 6-15 minutes to 18 seconds per task. These gains provide concrete evidence that semantic retrieval can operate within strict regulatory constraints and meaningfully reduce operational workload in real-world multilingual MRO workflows.

Paperid: 7196, https://arxiv.org/pdf/2511.13494.pdf

Abstract:
Recent vision-language models (VLMs) such as CLIP, OpenCLIP, EVA02-CLIP and SigLIP achieve strong zero-shot performance, but it is unclear how reliably they respond to controlled linguistic perturbations. We introduce Language-Guided Invariance Probing (LGIP), a benchmark that measures (i) invariance to meaning-preserving paraphrases and (ii) sensitivity to meaning-changing semantic flips in image-text matching. Using 40k MS COCO images with five human captions each, we automatically generate paraphrases and rule-based flips that alter object category, color or count, and summarize model behavior with an invariance error, a semantic sensitivity gap and a positive-rate statistic. Across nine VLMs, EVA02-CLIP and large OpenCLIP variants lie on a favorable invariance-sensitivity frontier, combining low paraphrase-induced variance with consistently higher scores for original captions than for their flipped counterparts. In contrast, SigLIP and SigLIP2 show much larger invariance error and often prefer flipped captions to the human descriptions, especially for object and color edits. These failures are largely invisible to standard retrieval metrics, indicating that LGIP provides a model-agnostic diagnostic for the linguistic robustness of VLMs beyond conventional accuracy scores.

Paperid: 7197, https://arxiv.org/pdf/2511.09101.pdf

Abstract:
Vision-Language Models (VLMs) such as CLIP achieve strong zero-shot recognition by comparing image embeddings to text-derived class prototypes. However, under domain shift, they suffer from feature drift, class-prior mismatch, and severe miscalibration. Existing test-time adaptation (TTA) methods often require backpropagation through large backbones, covariance estimation, or heavy memory/state, which is problematic for streaming and edge scenarios. We propose Ultra-Light Test-Time Adaptation (UL-TTA), a fully training-free and backprop-free framework that freezes the backbone and adapts only logit-level parameters: class prototypes, class priors, and temperature. UL-TTA performs an online EM-style procedure with (i) selective sample filtering to use only confident predictions, (ii) closed-form Bayesian updates for prototypes and priors anchored by text and Dirichlet priors, (iii) decoupled temperatures for prediction vs. calibration, and (iv) lightweight guards (norm clipping, prior KL constraints, smoothed temperature) to prevent drift in long streams. Across large-scale cross-domain and OOD benchmarks (PACS, Office-Home, DomainNet, Terra Incognita, ImageNet-R/A/V2/Sketch; ~726K test samples) and strong TTA baselines including Tent, T3A, CoTTA, SAR, Tip-Adapter, and FreeTTA, UL-TTA consistently improves top-1 accuracy (e.g., +4.7 points over zero-shot CLIP on average) while reducing ECE by 20-30%, with less than 8% latency overhead. Long-stream experiments up to 200K samples show no collapse. Our results demonstrate that logit-level Bayesian adaptation is sufficient to obtain state-of-the-art accuracy-calibration trade-offs for VLMs under domain shift, without updating any backbone parameters.

Paperid: 7198, https://arxiv.org/pdf/2511.07429.pdf

Abstract:
We introduce Text-based Explainable Video Anomaly Detection (TbVAD), a language-driven framework for weakly supervised video anomaly detection that performs anomaly detection and explanation entirely within the textual domain. Unlike conventional WSVAD models that rely on explicit visual features, TbVAD represents video semantics through language, enabling interpretable and knowledge-grounded reasoning. The framework operates in three stages: (1) transforming video content into fine-grained captions using a vision-language model, (2) constructing structured knowledge by organizing the captions into four semantic slots (action, object, context, environment), and (3) generating slot-wise explanations that reveal which semantic factors contribute most to the anomaly decision. We evaluate TbVAD on two public benchmarks, UCF-Crime and XD-Violence, demonstrating that textual knowledge reasoning provides interpretable and reliable anomaly detection for real-world surveillance scenarios.

Paperid: 7199, https://arxiv.org/pdf/2511.03019.pdf

Abstract:
Vision-Language Pretraining (VLP) has achieved remarkable success across various downstream tasks, but such gains are largely driven by scaling up on training data. Yet, literature methods treat image-text pairs as isolated training examples; this neglects the rich relational structure naturally present in many domains, such as e-commerce product co-purchase graphs and social recommendation networks. Inspired by neuroscientific evidence that human encodes knowledge as relationship cognitive maps, we introduce Structure-aware Language-Image Pretraining (SLIP). SLIP integrates a structural contrastive loss to align modalities while also modeling relationships between neighboring entities in a structured graph. To support this paradigm, we construct a large-scale Amazon Product Co-purchase Multimodal Graph Dataset, enabling structured cross-modality supervision at scale. Experiment results show that SLIP consistently outperforms CLIP on cross-modal retrieval and classification tasks in both zero-shot and few-shot settings, showing the value of relational supervision for cross-modal alignment.

Paperid: 7200, https://arxiv.org/pdf/2510.24804.pdf

Abstract:
A signature of human cognitive control is conflict adaptation: improved performance on a high-conflict trial following another high-conflict trial. This phenomenon offers an account for how cognitive control, a scarce resource, is recruited. Using a sequential Stroop task, we find that 12 of 13 vision-language models (VLMs) tested exhibit behavior consistent with conflict adaptation, with the lone exception likely reflecting a ceiling effect. To understand the representational basis of this behavior, we use sparse autoencoders (SAEs) to identify task-relevant supernodes in InternVL 3.5 4B. Partially overlapping supernodes emerge for text and color in both early and late layers, and their relative sizes mirror the automaticity asymmetry between reading and color naming in humans. We further isolate a conflict-modulated supernode in layers 24-25 whose ablation significantly increases Stroop errors while minimally affecting congruent trials.

Paperid: 7201, https://arxiv.org/pdf/2510.21774.pdf

Abstract:
We present OCR-Quality, a comprehensive human-annotated dataset designed for evaluating and developing OCR quality assessment methods. The dataset consists of 1,000 PDF pages converted to PNG images at 300 DPI, sampled from diverse real-world scenarios, including academic papers, textbooks, e-books, and multilingual documents. Each document has been processed using state-of-the-art Vision-Language Models (VLMs) and manually annotated with quality scores using a 4-level scoring system (1: Excellent, 2: Good, 3: Fair, 4: Poor). The dataset includes detailed source information, annotation guidelines, and representative cases across various difficulty levels. OCR-Quality addresses the critical need for reliable OCR quality assessment in real-world applications and provides a valuable benchmark for training and evaluating OCR verification systems. The dataset is publicly available at https://huggingface.co/datasets/Aslan-mingye/OCR-Quality .

Paperid: 7202, https://arxiv.org/pdf/2510.17034.pdf

Abstract:
Multimodal 3D grounding has garnered considerable interest in Vision-Language Models (VLMs) \cite{yin2025spatial} for advancing spatial reasoning in complex environments. However, these models suffer from a severe "2D semantic bias" that arises from over-reliance on 2D image features for coarse localization, largely disregarding 3D geometric inputs and resulting in suboptimal fusion performance. In this paper, we propose a novel training framework called What-Where Representation Re-Forming (W2R2) to tackle this issue via disentangled representation learning and targeted shortcut suppression. Our approach fundamentally reshapes the model's internal space by designating 2D features as semantic beacons for "What" identification and 3D features as spatial anchors for "Where" localization, enabling precise 3D grounding without modifying inference architecture. Key components include a dual-objective loss function with an Alignment Loss that supervises fused predictions using adapted cross-entropy for multimodal synergy, and a Pseudo-Label Loss that penalizes overly effective 2D-dominant pseudo-outputs via a margin-based mechanism. Experiments conducted on ScanRefer and ScanQA demonstrate the effectiveness of W2R2, with significant gains in localization accuracy and robustness, particularly in cluttered outdoor scenes.

Paperid: 7203, https://arxiv.org/pdf/2510.11302.pdf

Abstract:
Object detection systems have traditionally relied on supervised learning with manually annotated bounding boxes, achieving high accuracy at the cost of substantial annotation investment. The emergence of Vision-Language Models (VLMs) offers an alternative paradigm enabling zero-shot detection through natural language queries, eliminating annotation requirements but operating with reduced accuracy. This paper presents the first comprehensive cost-effectiveness analysis comparing supervised detection (YOLO) with zero-shot VLM inference (Gemini Flash 2.5). Through systematic evaluation on 1,000 stratified COCO images and 200 diverse product images spanning consumer electronics and rare categories, combined with detailed Total Cost of Ownership modeling, we establish quantitative break-even thresholds governing architecture selection. Our findings reveal that supervised YOLO achieves 91.2% accuracy versus 68.5% for zero-shot Gemini on standard categories, representing a 22.7 percentage point advantage that costs $10,800 in annotation for 100-category systems. However, this advantage justifies investment only beyond 55 million inferences, equivalent to 151,000 images daily for one year. Zero-shot Gemini demonstrates 52.3% accuracy on diverse product categories (ranging from highly web-prevalent consumer electronics at 75-85% to rare specialized equipment at 25-40%) where supervised YOLO achieves 0% due to architectural constraints preventing detection of untrained classes. Cost per Correct Detection analysis reveals substantially lower per-detection costs for Gemini ($0.00050 vs $0.143) at 100,000 inferences despite accuracy deficits. We develop decision frameworks demonstrating that optimal architecture selection depends critically on deployment volume, category stability, budget constraints, and accuracy requirements rather than purely technical performance metrics.

Paperid: 7204, https://arxiv.org/pdf/2510.10342.pdf

Abstract:
Accurate traffic congestion classification is essential for intelligent transportation systems and real-time urban traffic management. This paper presents a multimodal framework combining open-vocabulary visual-language reasoning (CLIP), object detection (YOLO-World), and motion analysis via MOG2-based background subtraction. The system predicts congestion levels on an ordinal scale from 1 (free flow) to 5 (severe congestion), enabling semantically aligned and temporally consistent classification. To enhance interpretability, we incorporate motion-based confidence weighting and generate annotated visual outputs. Experimental results show the model achieves 76.7 percent accuracy, an F1 score of 0.752, and a Quadratic Weighted Kappa (QWK) of 0.684, significantly outperforming unimodal baselines. These results demonstrate the framework's effectiveness in preserving ordinal structure and leveraging visual-language and motion modalities. Future enhancements include incorporating vehicle sizing and refined density metrics.

Paperid: 7205, https://arxiv.org/pdf/2510.09849.pdf

Abstract:
The widespread application of large vision language models has significantly raised safety concerns. In this project, we investigate text prompt injection, a simple yet effective method to mislead these models. We developed an algorithm for this type of attack and demonstrated its effectiveness and efficiency through experiments. Compared to other attack methods, our approach is particularly effective for large models without high demand for computational resources.

Paperid: 7206, https://arxiv.org/pdf/2510.09586.pdf

Abstract:
We present a transparent, reproducible measurement of research trends across 26,104 accepted papers from CVPR, ICLR, and NeurIPS spanning 2023-2025. Titles and abstracts are normalized, phrase-protected, and matched against a hand-crafted lexicon to assign up to 35 topical labels and mine fine-grained cues about tasks, architectures, training regimes, objectives, datasets, and co-mentioned modalities. The analysis quantifies three macro shifts: (1) a sharp rise of multimodal vision-language-LLM work, which increasingly reframes classic perception as instruction following and multi-step reasoning; (2) steady expansion of generative methods, with diffusion research consolidating around controllability, distillation, and speed; and (3) resilient 3D and video activity, with composition moving from NeRFs to Gaussian splatting and a growing emphasis on human- and agent-centric understanding. Within VLMs, parameter-efficient adaptation like prompting/adapters/LoRA and lightweight vision-language bridges dominate; training practice shifts from building encoders from scratch to instruction tuning and finetuning strong backbones; contrastive objectives recede relative to cross-entropy/ranking and distillation. Cross-venue comparisons show CVPR has a stronger 3D footprint and ICLR the highest VLM share, while reliability themes such as efficiency or robustness diffuse across areas. We release the lexicon and methodology to enable auditing and extension. Limitations include lexicon recall and abstract-only scope, but the longitudinal signals are consistent across venues and years.

Paperid: 7207, https://arxiv.org/pdf/2510.03294.pdf

Abstract:
Marine plastic pollution is a pressing environmental threat, making reliable automation for underwater debris detection essential. However, vision systems trained on one dataset often degrade on new imagery due to domain shift. This study benchmarks models for cross-domain robustness, training convolutional neural networks - CNNs (MobileNetV2, ResNet-18, EfficientNet-B0) and vision transformers (DeiT-Tiny, ViT-B16) on a labeled underwater dataset and then evaluates them on a balanced cross-domain test set built from plastic-positive images drawn from a different source and negatives from the training domain. Two zero-shot models were assessed, CLIP ViT-L14 and Google's Gemini 2.0 Flash, that leverage pretraining to classify images without fine-tuning. Results show the lightweight MobileNetV2 delivers the strongest cross-domain performance (F1 0.97), surpassing larger models. All fine-tuned models achieved high Precision (around 99%), but differ in Recall, indicating varying sensitivity to plastic instances. Zero-shot CLIP is comparatively sensitive (Recall around 80%) yet prone to false positives (Precision around 56%), whereas Gemini exhibits the inverse profile (Precision around 99%, Recall around 81%). Error analysis highlights recurring confusions with coral textures, suspended particulates, and specular glare. Overall, compact CNNs with supervised training can generalize effectively for cross-domain underwater detection, while large pretrained vision-language models provide complementary strengths.

Paperid: 7208, https://arxiv.org/pdf/2510.02780.pdf

Abstract:
Vision-Language Models (VLMs) excel at many multimodal tasks, yet their cognitive processes remain opaque on complex lateral thinking challenges like rebus puzzles. While recent work has demonstrated these models struggle significantly with rebus puzzle solving, the underlying reasoning processes and failure patterns remain largely unexplored. We address this gap through a comprehensive explainability analysis that moves beyond performance metrics to understand how VLMs approach these complex lateral thinking challenges. Our study contributes a systematically annotated dataset of 221 rebus puzzles across six cognitive categories, paired with an evaluation framework that separates reasoning quality from answer correctness. We investigate three prompting strategies designed to elicit different types of explanatory processes and reveal critical insights into VLM cognitive processes. Our findings demonstrate that reasoning quality varies dramatically across puzzle categories, with models showing systematic strengths in visual composition while exhibiting fundamental limitations in absence interpretation and cultural symbolism. We also discover that prompting strategy substantially influences both cognitive approach and problem-solving effectiveness, establishing explainability as an integral component of model performance rather than a post-hoc consideration.

Paperid: 7209, https://arxiv.org/pdf/2510.02114.pdf

Abstract:
Federeated Learning (FL) offers a privacy-preserving solution for Semantic Segmentation (SS) tasks to adapt to new domains, but faces significant challenges from these domain shifts, particularly when client data is unlabeled. However, most existing FL methods unrealistically assume access to labeled data on remote clients or fail to leverage the power of modern Vision Foundation Models (VFMs). Here, we propose a novel and challenging task, FFREEDG, in which a model is pretrained on a server's labeled source dataset and subsequently trained across clients using only their unlabeled data, without ever re-accessing the source. To solve FFREEDG, we propose FRIEREN, a framework that leverages the knowledge of a VFM by integrating vision and language modalities. Our approach employs a Vision-Language decoder guided by CLIP-based text embeddings to improve semantic disambiguation and uses a weak-to-strong consistency learning strategy for robust local training on pseudo-labels. Our experiments on synthetic-to-real and clear-to-adverse-weather benchmarks demonstrate that our framework effectively tackles this new task, achieving competitive performance against established domain generalization and adaptation methods and setting a strong baseline for future research.

Paperid: 7210, https://arxiv.org/pdf/2509.23957.pdf

Abstract:
Machine Interpreting systems are currently implemented as unimodal, real-time speech-to-speech architectures, processing translation exclusively on the basis of the linguistic signal. Such reliance on a single modality, however, constrains performance in contexts where disambiguation and adequacy depend on additional cues, such as visual, situational, or pragmatic information. This paper introduces Vision-Grounded Interpreting (VGI), a novel approach designed to address the limitations of unimodal machine interpreting. We present a prototype system that integrates a vision-language model to process both speech and visual input from a webcam, with the aim of priming the translation process through contextual visual information. To evaluate the effectiveness of this approach, we constructed a hand-crafted diagnostic corpus targeting three types of ambiguity. In our evaluation, visual grounding substantially improves lexical disambiguation, yields modest and less stable gains for gender resolution, and shows no benefit for syntactic ambiguities. We argue that embracing multimodality represents a necessary step forward for advancing translation quality in machine interpreting.

Paperid: 7211, https://arxiv.org/pdf/2509.22688.pdf

Abstract:
Multimodal Large Language Models (MLLMs) excel in vision-language reasoning but often struggle with structured perception tasks requiring precise localization and robustness. We propose a reinforcement learning framework that augments Group Relative Policy Optimization (GRPO) with curriculum-based data scheduling and difficulty-aware filtering. This approach stabilizes optimization under sparse, noisy rewards and enables progressive adaptation to complex samples. Evaluations on autonomous driving benchmarks demonstrate substantial improvements in detection accuracy and robustness. Ablation studies confirm the importance of reward design, KL regularization, and curriculum pacing for convergence stability and generalization. Our findings highlight reinforcement-driven optimization with structured data curricula as a scalable path toward robust and interpretable multimodal detection.

Paperid: 7212, https://arxiv.org/pdf/2509.20792.pdf

Abstract:
Vision-Language Models (VLMs) are foundational to critical applications like autonomous driving, medical diagnosis, and content moderation. While Parameter-Efficient Fine-Tuning (PEFT) methods like LoRA enable their efficient adaptation to specialized tasks, these models remain vulnerable to adversarial attacks that can compromise safety-critical decisions. CLIP, the backbone for numerous downstream VLMs, is a high-value target whose vulnerabilities can cascade across the multimodal AI ecosystem. We propose Dynamic Adversarial Curriculum DAC-LoRA, a novel framework that integrates adversarial training into PEFT. The core principle of our method i.e. an intelligent curriculum of progressively challenging attack, is general and can potentially be applied to any iterative attack method. Guided by the First-Order Stationary Condition (FOSC) and a TRADES-inspired loss, DAC-LoRA achieves substantial improvements in adversarial robustness without significantly compromising clean accuracy. Our work presents an effective, lightweight, and broadly applicable method to demonstrate that the DAC-LoRA framework can be easily integrated into a standard PEFT pipeline to significantly enhance robustness.

Paperid: 7213, https://arxiv.org/pdf/2509.14574.pdf

Abstract:
Understanding how people read city scenes can inform design and planning. We introduce a small benchmark for testing vision-language models (VLMs) on urban perception using 100 Montreal street images, evenly split between photographs and photorealistic synthetic scenes. Twelve participants from seven community groups supplied 230 annotation forms across 30 dimensions mixing physical attributes and subjective impressions. French responses were normalized to English. We evaluated seven VLMs in a zero-shot setup with a structured prompt and deterministic parser. We use accuracy for single-choice items and Jaccard overlap for multi-label items; human agreement uses Krippendorff's alpha and pairwise Jaccard. Results suggest stronger model alignment on visible, objective properties than subjective appraisals. The top system (claude-sonnet) reaches macro 0.31 and mean Jaccard 0.48 on multi-label items. Higher human agreement coincides with better model scores. Synthetic images slightly lower scores. We release the benchmark, prompts, and harness for reproducible, uncertainty-aware evaluation in participatory urban analysis.

Paperid: 7214, https://arxiv.org/pdf/2509.13590.pdf

Abstract:
The rapid advancement of artificial intelligence (AI) in healthcare imaging has revolutionized diagnostic medicine and clinical decision-making processes. This work presents an intelligent multimodal framework for medical image analysis that leverages Vision-Language Models (VLMs) in healthcare diagnostics. The framework integrates Google Gemini 2.5 Flash for automated tumor detection and clinical report generation across multiple imaging modalities including CT, MRI, X-ray, and Ultrasound. The system combines visual feature extraction with natural language processing to enable contextual image interpretation, incorporating coordinate verification mechanisms and probabilistic Gaussian modeling for anomaly distribution. Multi-layered visualization techniques generate detailed medical illustrations, overlay comparisons, and statistical representations to enhance clinical confidence, with location measurement achieving 80 pixels average deviation. Result processing utilizes precise prompt engineering and textual analysis to extract structured clinical information while maintaining interpretability. Experimental evaluations demonstrated high performance in anomaly detection across multiple modalities. The system features a user-friendly Gradio interface for clinical workflow integration and demonstrates zero-shot learning capabilities to reduce dependence on large datasets. This framework represents a significant advancement in automated diagnostic support and radiological workflow efficiency, though clinical validation and multi-center evaluation are necessary prior to widespread adoption.

Paperid: 7215, https://arxiv.org/pdf/2509.10536.pdf

Abstract:
We propose a geometric extension of restricted Boltzmann machines (RBMs) by allowing weights to take values in abstract groups such as $ \mathrm{GL}_n(\mathbb{R}) $, $ \mathrm{SU}(2) $, or even infinite-dimensional operator groups. This generalization enables the modeling of complex relational structures, including projective transformations, spinor dynamics, and functional symmetries, with direct applications to vision, language, and quantum learning. A central contribution of this work is the introduction of a \emph{contextuality index} based on group-valued holonomies computed along cycles in the RBM graph. This index quantifies the global inconsistency or "curvature" induced by local weights, generalizing classical notions of coherence, consistency, and geometric flatness. We establish links with sheaf-theoretic contextuality, gauge theory, and noncommutative geometry, and provide numerical and diagrammatic examples in both finite and infinite dimensions. This framework opens novel directions in AI, from curvature-aware learning architectures to topological regularization in uncertain or adversarial environments.

Paperid: 7216, https://arxiv.org/pdf/2509.08489.pdf

Abstract:
Prompt-driven image analysis converts a single natural-language instruction into multiple steps: locate, segment, edit, and describe. We present a practical case study of a unified pipeline that combines open-vocabulary detection, promptable segmentation, text-conditioned inpainting, and vision-language description into a single workflow. The system works end to end from a single prompt, retains intermediate artifacts for transparent debugging (such as detections, masks, overlays, edited images, and before and after composites), and provides the same functionality through an interactive UI and a scriptable CLI for consistent, repeatable runs. We highlight integration choices that reduce brittleness, including threshold adjustments, mask inspection with light morphology, and resource-aware defaults. In a small, single-word prompt segment, detection and segmentation produced usable masks in over 90% of cases with an accuracy above 85% based on our criteria. On a high-end GPU, inpainting makes up 60 to 75% of total runtime under typical guidance and sampling settings, which highlights the need for careful tuning. The study offers implementation-guided advice on thresholds, mask tightness, and diffusion parameters, and details version pinning, artifact logging, and seed control to support replay. Our contribution is a transparent, reliable pattern for assembling modern vision and multimodal models behind a single prompt, with clear guardrails and operational practices that improve reliability in object replacement, scene augmentation, and removal.

Paperid: 7217, https://arxiv.org/pdf/2509.01959.pdf

Abstract:
Multimodal models, such as the Contrastive Language-Image Pre-training (CLIP) model, have demonstrated remarkable success in aligning visual and linguistic representations. However, these models exhibit limitations when applied to specialised visual domains, such as diagrams, which encode structured, symbolic information distinct from that of natural imagery. In this paper, we introduce a novel training paradigm explicitly designed to enhance the comprehension of diagrammatic images within vision-language models. Our approach uses ``hard'' samples for our proposed contrastive learning that incorporates two specialised loss functions that leverage the inherent structural properties of diagrams. By integrating these objectives into model training, our method enables models to develop a more structured and semantically coherent understanding of diagrammatic content. We empirically validate our approach on a benchmark dataset of flowcharts, as a representative class of diagrammatic imagery, demonstrating substantial improvements over standard CLIP and conventional hard negative CLIP learning paradigms for both image-text matching and visual question answering tasks. Our findings underscore the significance of tailored training strategies for specialised tasks and contribute to advancing diagrammatic understanding within the broader landscape of vision-language integration.

Paperid: 7218, https://arxiv.org/pdf/2508.17686.pdf

Abstract:
Vision Language Models (VLMs) struggle with long-form videos due to the quadratic complexity of attention mechanisms. We propose Language-Guided Temporal Token Pruning (LGTTP), which leverages temporal cues from queries to adaptively prune video tokens, preserving contextual continuity while reducing computational overhead. Unlike uniform pruning or keyframe selection, LGTTP retains higher token density in temporally relevant segments. Our model-agnostic framework integrates with TimeChat and LLaVA-Video, achieving a 65% reduction in computation while preserving 97-99% of the original performance. On QVHighlights, LGTTP improves HIT@1 by +9.5%, and on Charades-STA, it retains 99.6% of R@1. It excels on queries with explicit temporal markers and remains effective across general video understanding tasks.

Paperid: 7219, https://arxiv.org/pdf/2507.23226.pdf

Abstract:
As augmented reality (AR) becomes increasingly integrated into everyday life, ensuring the safety and trustworthiness of its virtual content is critical. Our research addresses the risks of task-detrimental AR content, particularly that which obstructs critical information or subtly manipulates user perception. We developed two systems, ViDDAR and VIM-Sense, to detect such attacks using vision-language models (VLMs) and multimodal reasoning modules. Building on this foundation, we propose three future directions: automated, perceptually aligned quality assessment of virtual content; detection of multimodal attacks; and adaptation of VLMs for efficient and user-centered deployment on AR devices. Overall, our work aims to establish a scalable, human-aligned framework for safeguarding AR experiences and seeks feedback on perceptual modeling, multimodal AR content implementation, and lightweight model adaptation.

Paperid: 7220, https://arxiv.org/pdf/2507.22469.pdf

Abstract:
The contemporary phenomenon of deepfakes, utilizing GAN or diffusion models for face swapping, presents a substantial and evolving threat in digital media, identity verification, and a multitude of other systems. The majority of existing methods for detecting deepfakes rely on training specialized classifiers to distinguish between genuine and manipulated images, focusing only on the image domain without incorporating any auxiliary tasks that could enhance robustness. In this paper, inspired by the zero-shot capabilities of Vision Language Models, we propose a novel VLM-based approach to image classification and then evaluate it for deepfake detection. Specifically, we utilize a new high-quality deepfake dataset comprising 60,000 images, on which our zero-shot models demonstrate superior performance to almost all existing methods. Subsequently, we compare the performance of the best-performing architecture, InstructBLIP, on the popular deepfake dataset DFDC-P against traditional methods in two scenarios: zero-shot and in-domain fine-tuning. Our results demonstrate the superiority of VLMs over traditional classifiers.

Paperid: 7221, https://arxiv.org/pdf/2507.22304.pdf

Abstract:
Vision-language models (VLMs) have revolutionized multimodal AI applications but introduce novel security vulnerabilities that remain largely unexplored. We present the first comprehensive study of steganographic prompt injection attacks against VLMs, where malicious instructions are invisibly embedded within images using advanced steganographic techniques. Our approach demonstrates that current VLM architectures can inadvertently extract and execute hidden prompts during normal image processing, leading to covert behavioral manipulation. We develop a multi-domain embedding framework combining spatial, frequency, and neural steganographic methods, achieving an overall attack success rate of 24.3% (plus or minus 3.2%, 95% CI) across leading VLMs including GPT-4V, Claude, and LLaVA, with neural steganography methods reaching up to 31.8%, while maintaining reasonable visual imperceptibility (PSNR greater than 38 dB, SSIM greater than 0.94). Through systematic evaluation on 12 diverse datasets and 8 state-of-the-art models, we reveal moderate but meaningful vulnerabilities in current VLM architectures and propose effective countermeasures. Our findings have significant implications for VLM deployment in security-critical applications and highlight the need for proportionate multimodal AI security frameworks.

Paperid: 7222, https://arxiv.org/pdf/2507.19679.pdf

Abstract:
Image-based product attribute prediction in e-commerce is a crucial task with numerous applications. The supervised fine-tuning of Vision Language Models (VLMs) faces significant scale challenges due to the cost of manual or API based annotation. In this paper, we investigate label-efficient semi-supervised fine-tuning strategies for compact VLMs (2B-3B parameters) that leverage unlabeled product listings through Direct Preference Optimization (DPO). Beginning with a small, API-based, annotated, and labeled set, we first employ PEFT to train low-rank adapter modules. To update the adapter weights with unlabeled data, we generate multiple reasoning-and-answer chains per unlabeled sample and segregate these chains into preferred and dispreferred based on self-consistency. We then fine-tune the model with DPO loss and use the updated model for the next iteration. By using PEFT fine-tuning with DPO, our method achieves efficient convergence with minimal compute overhead. On a dataset spanning twelve e-commerce verticals, DPO-based fine-tuning, which utilizes only unlabeled data, demonstrates a significant improvement over the supervised model. Moreover, experiments demonstrate that accuracy with DPO training improves with more unlabeled data, indicating that a large pool of unlabeled samples can be effectively leveraged to improve performance.

Paperid: 7223, https://arxiv.org/pdf/2507.18788.pdf

Abstract:
Image captioning, situated at the intersection of computer vision and natural language processing, requires a sophisticated understanding of both visual scenes and linguistic structure. While modern approaches are dominated by large-scale Transformer architectures, this paper documents a systematic, iterative development of foundational image captioning models, progressing from a simple CNN-LSTM encoder-decoder to a competitive attention-based system. This paper presents a series of five models, beginning with Genesis and concluding with Nexus, an advanced model featuring an EfficientNetV2B3 backbone and a dynamic attention mechanism. The experiments chart the impact of architectural enhancements and demonstrate a key finding within the classic CNN-LSTM paradigm: merely upgrading the visual backbone without a corresponding attention mechanism can degrade performance, as the single-vector bottleneck cannot transmit the richer visual detail. This insight validates the architectural shift to attention. Trained on the MS COCO 2017 dataset, the final model, Nexus, achieves a BLEU-4 score of 31.4, surpassing several foundational benchmarks and validating the iterative design process. This work provides a clear, replicable blueprint for understanding the core architectural principles that underpin modern vision-language tasks.

Paperid: 7224, https://arxiv.org/pdf/2507.12911.pdf

Abstract:
Out-of-distribution (OOD) scenarios in autonomous driving pose critical challenges, as planners often fail to generalize beyond their training experience, leading to unsafe or unexpected behavior. Vision-Language Models (VLMs) have shown promise in handling such scenarios by providing high-level scene understanding and user-aligned decisions. However, existing VLMs often exhibit a misalignment between their language-based reasoning and the low-level trajectories required for action-level planning. In this paper, we propose LaViPlan, a framework that leverages Reinforcement Learning with Verifiable Rewards (RLVR) to fine-tune VLMs using planning-oriented metrics. Experimental results show that LaViPlan improves planning performance across both in-domain and out-of-domain datasets. While linguistic fidelity slightly decreases after RLVR-based fine-tuning, qualitative evaluation indicates that the outputs remain coherent. We also conduct ablation studies to analyze the effects of sampling ratio and reasoning guidance, highlighting how these design choices influence performance. These findings demonstrate the potential of RLVR as a post-training paradigm for aligning language-guided reasoning with action-level planning in autonomous driving.

Paperid: 7225, https://arxiv.org/pdf/2507.09966.pdf

Abstract:
Accurate brain tumor segmentation is crucial for neuro-oncology diagnosis and treatment planning. Deep learning methods have made significant progress, but automatic segmentation still faces challenges, including tumor morphological heterogeneity and complex three-dimensional spatial relationships. This paper proposes a three-tier fusion architecture that achieves precise brain tumor segmentation. The method processes information progressively at the pixel, feature, and semantic levels. At the pixel level, physical modeling extends magnetic resonance imaging (MRI) to multimodal data, including simulated ultrasound and synthetic computed tomography (CT). At the feature level, the method performs Transformer-based cross-modal feature fusion through multi-teacher collaborative distillation, integrating three expert teachers (MRI, US, CT). At the semantic level, clinical textual knowledge generated by GPT-4V is transformed into spatial guidance signals using CLIP contrastive learning and Feature-wise Linear Modulation (FiLM). These three tiers together form a complete processing chain from data augmentation to feature extraction to semantic guidance. We validated the method on the Brain Tumor Segmentation (BraTS) 2020, 2021, and 2023 datasets. The model achieves average Dice coefficients of 0.8665, 0.9014, and 0.8912 on the three datasets, respectively, and reduces the 95% Hausdorff Distance (HD95) by an average of 6.57 millimeters compared with the baseline. This method provides a new paradigm for precise tumor segmentation and boundary localization.

Paperid: 7226, https://arxiv.org/pdf/2507.01673.pdf

Abstract:
Facial expression recognition (FER) in 3D and 4D domains presents a significant challenge in affective computing due to the complexity of spatial and temporal facial dynamics. Its success is crucial for advancing applications in human behavior understanding, healthcare monitoring, and human-computer interaction. In this work, we propose FACET-VLM, a vision-language framework for 3D/4D FER that integrates multiview facial representation learning with semantic guidance from natural language prompts. FACET-VLM introduces three key components: Cross-View Semantic Aggregation (CVSA) for view-consistent fusion, Multiview Text-Guided Fusion (MTGF) for semantically aligned facial emotions, and a multiview consistency loss to enforce structural coherence across views. Our model achieves state-of-the-art accuracy across multiple benchmarks, including BU-3DFE, Bosphorus, BU-4DFE, and BP4D-Spontaneous. We further extend FACET-VLM to 4D micro-expression recognition (MER) on the 4DME dataset, demonstrating strong performance in capturing subtle, short-lived emotional cues. The extensive experimental results confirm the effectiveness and substantial contributions of each individual component within the framework. Overall, FACET-VLM offers a robust, extensible, and high-performing solution for multimodal FER in both posed and spontaneous settings.

Paperid: 7227, https://arxiv.org/pdf/2507.00907.pdf

Abstract:
In a world where deepfakes and cloned voices are emerging as sophisticated attack vectors, organizations require a new security mindset: Sensorial Zero Trust [9]. This article presents a scientific analysis of the need to systematically doubt information perceived through the senses, establishing rigorous verification protocols to mitigate the risks of fraud based on generative artificial intelligence. Key concepts, such as Out-of-Band verification, Vision-Language Models (VLMs) as forensic collaborators, cryptographic provenance, and human training, are integrated into a framework that extends Zero Trust principles to human sensory information. The approach is grounded in empirical findings and academic research, emphasizing that in an era of AI-generated realities, even our eyes and ears can no longer be implicitly trusted without verification. Leaders are called to foster a culture of methodological skepticism to protect organizational integrity in this new threat landscape.

Paperid: 7228, https://arxiv.org/pdf/2506.18985.pdf

Abstract:
Recent large vision-language models (LVLMs) have advanced capabilities in visual question answering (VQA). However, interpreting where LVLMs direct their visual attention remains a significant challenge, yet is essential for understanding model behavior. We introduce GLIMPSE (Gradient-Layer Importance Mapping for Prompted Visual Saliency Explanation), a lightweight, model-agnostic framework that jointly attributes LVLM outputs to the most relevant visual evidence and textual signals that support open-ended generation. GLIMPSE fuses gradient-weighted attention, adaptive layer propagation, and relevance-weighted token aggregation to produce holistic response-level heat maps for interpreting cross-modal reasoning, outperforming prior methods in faithfulness and pushing the state-of-the-art in human-attention alignment. We demonstrate an analytic approach to uncover fine-grained insights into LVLM cross-modal attribution, trace reasoning dynamics, analyze systematic misalignment, diagnose hallucination and bias, and ensure transparency.

Paperid: 7229, https://arxiv.org/pdf/2506.14823.pdf

Abstract:
Monitoring animal populations in natural environments requires systems that can interpret both visual data and human language queries. This work introduces ViLLa (Vision-Language-Logic Approach), a neuro-symbolic framework designed for interpretable animal monitoring. ViLLa integrates three core components: a visual detection module for identifying animals and their spatial locations in images, a language parser for understanding natural language queries, and a symbolic reasoning layer that applies logic-based inference to answer those queries. Given an image and a question such as "How many dogs are in the scene?" or "Where is the buffalo?", the system grounds visual detections into symbolic facts and uses predefined rules to compute accurate answers related to count, presence, and location. Unlike end-to-end black-box models, ViLLa separates perception, understanding, and reasoning, offering modularity and transparency. The system was evaluated on a range of animal imagery tasks and demonstrates the ability to bridge visual content with structured, human-interpretable queries.

Paperid: 7230, https://arxiv.org/pdf/2506.05358.pdf

Abstract:
Multimodal Large Language Models (MLLMs) like GPT-4V are capable of reasoning across text and image modalities, showing promise in a variety of complex vision-language tasks. In this preliminary study, we investigate the out-of-the-box capabilities of GPT-4V in the domain of image forensics, specifically, in detecting image splicing manipulations. Without any task-specific fine-tuning, we evaluate GPT-4V using three prompting strategies: Zero-Shot (ZS), Few-Shot (FS), and Chain-of-Thought (CoT), applied over a curated subset of the CASIA v2.0 splicing dataset. Our results show that GPT-4V achieves competitive detection performance in zero-shot settings (more than 85% accuracy), with CoT prompting yielding the most balanced trade-off across authentic and spliced images. Qualitative analysis further reveals that the model not only detects low-level visual artifacts but also draws upon real-world contextual knowledge such as object scale, semantic consistency, and architectural facts, to identify implausible composites. While GPT-4V lags behind specialized state-of-the-art splicing detection models, its generalizability, interpretability, and encyclopedic reasoning highlight its potential as a flexible tool in image forensics.

Paperid: 7231, https://arxiv.org/pdf/2506.01203.pdf

Abstract:
Facial expression recognition (FER) is a fundamental task in affective computing with applications in human-computer interaction, mental health analysis, and behavioral understanding. In this paper, we propose SMILE-VLM, a self-supervised vision-language model for 3D/4D FER that unifies multiview visual representation learning with natural language supervision. SMILE-VLM learns robust, semantically aligned, and view-invariant embeddings by proposing three core components: multiview decorrelation via a Barlow Twins-style loss, vision-language contrastive alignment, and cross-modal redundancy minimization. Our framework achieves the state-of-the-art performance on multiple benchmarks. We further extend SMILE-VLM to the task of 4D micro-expression recognition (MER) to recognize the subtle affective cues. The extensive results demonstrate that SMILE-VLM not only surpasses existing unsupervised methods but also matches or exceeds supervised baselines, offering a scalable and annotation-efficient solution for expressive facial behavior understanding.

Paperid: 7232, https://arxiv.org/pdf/2505.13746.pdf

Abstract:
Surgical phase recognition from video is a technology that automatically classifies the progress of a surgical procedure and has a wide range of potential applications, including real-time surgical support, optimization of medical resources, training and skill assessment, and safety improvement. Recent advances in surgical phase recognition technology have focused primarily on Transform-based methods, although methods that extract spatial features from individual frames using a CNN and video features from the resulting time series of spatial features using time series modeling have shown high performance. However, there remains a paucity of research on training methods for CNNs employed for feature extraction or representation learning in surgical phase recognition. In this study, we propose a method for representation learning in surgical workflow analysis using a vision-language model (ReSW-VL). Our proposed method involves fine-tuning the image encoder of a CLIP (Convolutional Language Image Model) vision-language model using prompt learning for surgical phase recognition. The experimental results on three surgical phase recognition datasets demonstrate the effectiveness of the proposed method in comparison to conventional methods.

Paperid: 7233, https://arxiv.org/pdf/2505.13507.pdf

Abstract:
Open-Set Domain Adaptation (OSDA) confronts the dual challenge of aligning known-class distributions across domains while identifying target-domain-specific unknown categories. Current approaches often fail to leverage semantic relationships between modalities and struggle with error accumulation in unknown sample detection. We propose to harness Contrastive Language-Image Pretraining (CLIP) to address these limitations through two key innovations: 1) Prompt-driven cross-domain alignment: Learnable textual prompts conditioned on domain discrepancy metrics dynamically adapt CLIP's text encoder, enabling semantic consistency between source and target domains without explicit unknown-class supervision. 2) Gradient-aware open-set separation: A gradient analysis module quantifies domain shift by comparing the L2-norm of gradients from the learned prompts, where known/unknown samples exhibit statistically distinct gradient behaviors. Evaluations on Office-Home show that our method consistently outperforms CLIP baseline and standard baseline. Ablation studies confirm the gradient norm's critical role.

Paperid: 7234, https://arxiv.org/pdf/2505.12363.pdf

Abstract:
While Multimodal Large Language Models (MLLMs) excel at general vision-language tasks, visuospatial cognition - reasoning about spatial layouts, relations, and dynamics - remains a significant challenge. Existing models often lack the necessary architectural components and specialized training data for fine-grained spatial understanding. We introduce ViCA2 (Visuospatial Cognitive Assistant 2), a novel MLLM designed to enhance spatial reasoning. ViCA2 features a dual vision encoder architecture integrating SigLIP for semantics and Hiera for spatial structure, coupled with a token ratio control mechanism for efficiency. We also developed ViCA-322K, a new large-scale dataset with over 322,000 spatially grounded question-answer pairs for targeted instruction tuning. On the challenging VSI-Bench benchmark, our ViCA2-7B model achieves a state-of-the-art average score of 56.8, significantly surpassing larger open-source models (e.g., LLaVA-NeXT-Video-72B, 40.9) and leading proprietary models (Gemini-1.5 Pro, 45.4). This demonstrates the effectiveness of our approach in achieving strong visuospatial intelligence with a compact model. We release ViCA2, its codebase, and the ViCA-322K dataset to facilitate further research.

Paperid: 7235, https://arxiv.org/pdf/2505.12312.pdf

Abstract:
Video-based spatial cognition is vital for robotics and embodied AI but challenges current Vision-Language Models (VLMs). This paper makes two key contributions. First, we introduce ViCA (Visuospatial Cognitive Assistant)-322K, a diverse dataset of 322,003 QA pairs from real-world indoor videos (ARKitScenes, ScanNet, ScanNet++), offering supervision for 3D metadata-grounded queries and video-based complex reasoning. Second, we develop ViCA-7B, fine-tuned on ViCA-322K, which achieves new state-of-the-art on all eight VSI-Bench tasks, outperforming existing models, including larger ones (e.g., +26.1 on Absolute Distance). For interpretability, we present ViCA-Thinking-2.68K, a dataset with explicit reasoning chains, and fine-tune ViCA-7B to create ViCA-7B-Thinking, a model that articulates its spatial reasoning. Our work highlights the importance of targeted data and suggests paths for improved temporal-spatial modeling. We release all resources to foster research in robust visuospatial intelligence.

Paperid: 7236, https://arxiv.org/pdf/2505.11758.pdf

Abstract:
Few-shot adaptation remains a core challenge for vision-language models (VLMs), especially under limited supervision and noisy support samples. We propose PromptFuseNL, a unified framework that enhances few-shot generalization by combining predictive prompt tuning with dual-branch positive and negative learning. The method refines class prototypes through task-conditioned residuals, multi-stage cross-modal coordination, and semantic hard negative mining. To address label noise, we introduce an unsupervised instance reweighting strategy that downweights unreliable support examples without requiring additional labels or structural changes. PromptFuseNL fuses visual and textual cues through lightweight modules for efficient and discriminative prediction. Evaluated across 15 benchmarks, it consistently surpasses existing prompt- and adapter-based methods in all shot settings while remaining highly efficient, achieving up to 300x faster training and 1000x lower FLOPs compared to full prompt tuning, achieving a new state-of-the-art for robust and scalable few-shot vision-language adaptation.

Paperid: 7237, https://arxiv.org/pdf/2505.10664.pdf

Abstract:
Verifying the authenticity of AI-generated images presents a growing challenge on social media platforms these days. While vision-language models (VLMs) like CLIP outdo in multimodal representation, their capacity for AI-generated image classification is underexplored due to the absence of such labels during the pre-training process. This work investigates whether CLIP embeddings inherently contain information indicative of AI generation. A proposed pipeline extracts visual embeddings using a frozen CLIP model, feeds its embeddings to lightweight networks, and fine-tunes only the final classifier. Experiments on the public CIFAKE benchmark show the performance reaches 95% accuracy without language reasoning. Few-shot adaptation to curated custom with 20% of the data results in performance to 85%. A closed-source baseline (Gemini-2.0) has the best zero-shot accuracy yet fails on specific styles. Notably, some specific image types, such as wide-angle photographs and oil paintings, pose significant challenges to classification. These results indicate previously unexplored difficulties in classifying certain types of AI-generated images, revealing new and more specific questions in this domain that are worth further investigation.

Paperid: 7238, https://arxiv.org/pdf/2505.09336.pdf

Abstract:
In this paper, we introduce MultiviewVLM, a vision-language model designed for unsupervised contrastive multiview representation learning of facial emotions from 3D/4D data. Our architecture integrates pseudo-labels derived from generated textual prompts to guide implicit alignment of emotional semantics. To capture shared information across multi-views, we propose a joint embedding space that aligns multiview representations without requiring explicit supervision. We further enhance the discriminability of our model through a novel multiview contrastive learning strategy that leverages stable positive-negative pair sampling. A gradient-friendly loss function is introduced to promote smoother and more stable convergence, and the model is optimized for distributed training to ensure scalability. Extensive experiments demonstrate that MultiviewVLM outperforms existing state-of-the-art methods and can be easily adapted to various real-world applications with minimal modifications.

Paperid: 7239, https://arxiv.org/pdf/2505.03500.pdf

Abstract:
Vision-language-action models (VLAs) often achieve high performance on demonstrated tasks but struggle significantly when required to extrapolate, combining skills learned from different tasks in novel ways. For instance, VLAs might successfully put the cream cheese in the bowl and put the bowl on top of the cabinet, yet still fail to put the cream cheese on top of the cabinet. In this work, we demonstrate that behaviors from distinct tasks can be effectively recombined by manipulating the VLA's internal representations at inference time. Concretely, we identify the text latent by averaging the text tokens' hidden states across all demonstrated trajectories for a specific base task. For executing an extrapolated task, we can temporally interpolate the text latent of the two base tasks and add it back to the text hidden states, so sub-behaviors from the two tasks will be activated sequentially. We evaluate this approach using the newly created libero-ood benchmark, featuring 20 tasks extrapolated from standard LIBERO suites. The results on libero-ood show that all SOTA VLAs achieve < 15% success rate, while $\pi0$ with text latent interpolation reaches an 83% success rate. Further qualitative analysis reveals a tendency for VLAs to exhibit spatial overfitting, mapping object names to demonstrated locations rather than achieving genuine object and goal understanding. Additionally, we find that decoding the text latent yields human-unreadable prompts that can nevertheless instruct the VLA to achieve a 70% success rate on standard LIBERO suites, enabling private instruction or backdoor attacks.

Paperid: 7240, https://arxiv.org/pdf/2505.01694.pdf

Abstract:
Efficiently adapting large Vision-Language Models (VLMs) like CLIP for few-shot learning poses challenges in balancing pre-trained knowledge retention and task-specific adaptation. Existing methods often overlook valuable structural information within the VLM's latent space. We introduce a topology-aware tuning approach integrating Representation Topology Divergence (RTD) into the Task Residual (TR) framework. By explicitly aligning the topological structures of visual and text representations using a combined RTD and Cross-Entropy loss, while freezing base VLM encoders, our method enhances few-shot performance. We optimize only lightweight Task Residual parameters, effectively leveraging topological information. Across 6 diverse benchmark datasets, our approach demonstrates significant gains, achieving an average accuracy improvement of 1-2\% over relevant baseline methods in few-shot settings. This work presents an effective strategy to boost VLM few-shot capabilities by incorporating topological alignment.

Paperid: 7241, https://arxiv.org/pdf/2505.00746.pdf

Abstract:
Vision-language models such as OpenAI GPT-4o can transcribe mathematical documents directly from images, yet their token-level confidence signals are seldom used to pinpoint local recognition mistakes. We present an entropy-heat-mapping proof-of-concept that turns per-token Shannon entropy into a visual ''uncertainty landscape''. By scanning the entropy sequence with a fixed-length sliding window, we obtain hotspots that are likely to contain OCR errors such as missing symbols, mismatched braces, or garbled prose. Using a small, curated set of scanned research pages rendered at several resolutions, we compare the highlighted hotspots with the actual transcription errors produced by GPT-4o. Our analysis shows that the vast majority of true errors are indeed concentrated inside the high-entropy regions. This study demonstrates--in a minimally engineered setting--that sliding-window entropy can serve as a practical, lightweight aid for post-editing GPT-based OCR. All code and annotation guidelines are released to encourage replication and further research.

Paperid: 7242, https://arxiv.org/pdf/2504.18538.pdf

Abstract:
Imitation learning holds the promise of equipping robots with versatile skills by learning from expert demonstrations. However, policies trained on finite datasets often struggle to generalize beyond the training distribution. In this work, we present a unified perspective on the generalization capability of imitation learning, grounded in both information theorey and data distribution property. We first show that the generalization gap can be upper bounded by (i) the conditional information bottleneck on intermediate representations and (ii) the mutual information between the model parameters and the training dataset. This characterization provides theoretical guidance for designing effective training strategies in imitation learning, particularly in determining whether to freeze, fine-tune, or train large pretrained encoders (e.g., vision-language models or vision foundation models) from scratch to achieve better generalization. Furthermore, we demonstrate that high conditional entropy from input to output induces a flatter likelihood landscape, thereby reducing the upper bound on the generalization gap. In addition, it shortens the stochastic gradient descent (SGD) escape time from sharp local minima, which may increase the likelihood of reaching global optima under fixed optimization budgets. These insights explain why imitation learning often exhibits limited generalization and underscore the importance of not only scaling the diversity of input data but also enriching the variability of output labels conditioned on the same input.

Paperid: 7243, https://arxiv.org/pdf/2504.06272.pdf

Abstract:
We present RAVEN an adaptive AI agent framework designed for multimodal entity discovery and retrieval in large-scale video collections. Synthesizing information across visual, audio, and textual modalities, RAVEN autonomously processes video data to produce structured, actionable representations for downstream tasks. Key contributions include (1) a category understanding step to infer video themes and general-purpose entities, (2) a schema generation mechanism that dynamically defines domain-specific entities and attributes, and (3) a rich entity extraction process that leverages semantic retrieval and schema-guided prompting. RAVEN is designed to be model-agnostic, allowing the integration of different vision-language models (VLMs) and large language models (LLMs) based on application-specific requirements. This flexibility supports diverse applications in personalized search, content discovery, and scalable information retrieval, enabling practical applications across vast datasets.

Paperid: 7244, https://arxiv.org/pdf/2504.04772.pdf

Abstract:
Real-time scene comprehension is a key advance in artificial intelligence, enhancing robotics, surveillance, and assistive tools. However, hallucination remains a challenge. AI systems often misinterpret visual inputs, detecting nonexistent objects or describing events that never happened. These errors, far from minor, threaten reliability in critical areas like security and autonomous navigation where accuracy is essential. Our approach tackles this by embedding self-awareness into the AI. Instead of trusting initial outputs, our framework continuously assesses them in real time, adjusting confidence thresholds dynamically. When certainty falls below a solid benchmark, it suppresses unreliable claims. Combining YOLOv5's object detection strength with VILA1.5-3B's controlled language generation, we tie descriptions to confirmed visual data. Strengths include dynamic threshold tuning for better accuracy, evidence-based text to reduce hallucination, and real-time performance at 18 frames per second. This feedback-driven design cuts hallucination by 37 percent over traditional methods. Fast, flexible, and reliable, it excels in applications from robotic navigation to security monitoring, aligning AI perception with reality.

Paperid: 7245, https://arxiv.org/pdf/2503.19564.pdf

Abstract:
As artificial intelligence systems increasingly operate in Real-world environments, the integration of multi-modal data sources such as vision, language, and audio presents both unprecedented opportunities and critical challenges for achieving trustworthy intelligence. In this paper, we propose a novel framework that unifies federated learning with explainable multi-modal reasoning to ensure trustworthiness in decentralized, dynamic settings. Our approach, called FedMM-X (Federated Multi-Modal Explainable Intelligence), leverages cross-modal consistency checks, client-level interpretability mechanisms, and dynamic trust calibration to address challenges posed by data heterogeneity, modality imbalance, and out-of-distribution generalization. Through rigorous evaluation across federated multi-modal benchmarks involving vision-language tasks, we demonstrate improved performance in both accuracy and interpretability while reducing vulnerabilities to adversarial and spurious correlations. Further, we introduce a novel trust score aggregation method to quantify global model reliability under dynamic client participation. Our findings pave the way toward developing robust, interpretable, and socially responsible AI systems in Real-world environments.

Paperid: 7246, https://arxiv.org/pdf/2503.19510.pdf

Abstract:
As robotic technologies advancing towards more complex multimodal interactions and manipulation tasks, the integration of advanced Vision-Language Models (VLMs) has become a key driver in the field. Despite progress with current methods, challenges persist in fusing depth and RGB information within 3D environments and executing tasks guided by linguistic instructions. In response to these challenges, we have enhanced the existing RoboFlamingo framework by introducing RoboFlamingo-Plus, which incorporates depth data into VLMs to significantly improve robotic manipulation performance. Our research achieves a nuanced fusion of RGB and depth information by integrating a pre-trained Vision Transformer (ViT) with a resampling technique, closely aligning this combined data with linguistic cues for superior multimodal understanding. The novelty of RoboFlamingo-Plus lies in its adaptation of inputs for depth data processing, leveraging a pre-trained resampler for depth feature extraction, and employing cross-attention mechanisms for optimal feature integration. These improvements allow RoboFlamingo-Plus to not only deeply understand 3D environments but also easily perform complex, language-guided tasks in challenging settings. Experimental results show that RoboFlamingo-Plus boosts robotic manipulation by 10-20% over current methods, marking a significant advancement. Codes and model weights are public at RoboFlamingo-Plus.

Paperid: 7247, https://arxiv.org/pdf/2503.06670.pdf

Abstract:
Interpretability in Vision-Language Models (VLMs) is crucial for trust, debugging, and decision-making in high-stakes applications. We introduce PixelSHAP, a model-agnostic framework extending Shapley-based analysis to structured visual entities. Unlike previous methods focusing on text prompts, PixelSHAP applies to vision-based reasoning by systematically perturbing image objects and quantifying their influence on a VLM's response. PixelSHAP requires no model internals, operating solely on input-output pairs, making it compatible with open-source and commercial models. It supports diverse embedding-based similarity metrics and scales efficiently using optimization techniques inspired by Shapley-based methods. We validate PixelSHAP in autonomous driving, highlighting its ability to enhance interpretability. Key challenges include segmentation sensitivity and object occlusion. Our open-source implementation facilitates further research.

Paperid: 7248, https://arxiv.org/pdf/2503.04839.pdf

Abstract:
Multimodal in-context learning (ICL) has emerged as a key capability of Large Vision-Language Models (LVLMs), driven by their increasing scale and applicability. Despite its promise, effective ICL in the multimodal setting remains challenging due to the inherent complexity of image-text inputs and the high sensitivity of ICL performance to input configurations. In this work, we shed light on the core mechanism underlying multimodal ICL, identifying task mapping as a crucial factor in configuring robust in-context demonstration (ICD) sequences. Building on these insights, we propose \textit{SabER}, a lightweight yet powerful decoder-only transformer equipped with task-aware attention, which intelligently selects and arranges ICDs from a demonstration library in an autoregressive fashion. This design enables fine-grained feature extraction and cross-modal reasoning, iteratively refining task mapping to generate high-quality ICD sequences. Through extensive experiments covering five LVLMs and nine benchmark datasets, SabER not only demonstrates strong empirical performance, but also provides deeper understanding of how task semantics interact with multimodal ICDs. Our findings highlight the importance of principled ICD sequence configuration and open new avenues to enhance multimodal ICL in a wide range of real-world scenarios.

Paperid: 7249, https://arxiv.org/pdf/2503.03202.pdf

Abstract:
Training vision-language models for image-text alignment typically requires large datasets to achieve robust performance. In low-data scenarios, standard contrastive learning can struggle to align modalities effectively due to overfitting and unstable training dynamics. In this paper, we propose a variance-aware loss scheduling approach that dynamically adjusts the weighting of the contrastive loss based on the statistical variability (uncertainty) in the model's alignment predictions. Using a subset of the Flickr8k image-caption dataset to simulate limited data conditions, we demonstrate that our approach improves image-text retrieval accuracy compared to a fixed-weight baseline. We also compare against other adaptive weighting strategies (using output entropy and cosine similarity spread) and find that variance-aware scheduling provides the best overall trade-off. Qualitatively, our method yields more distinct multimodal embeddings as shown by t-SNE visualizations. Moreover, in a stress test with noise-injected captions and images, the variance-guided loss proves more robust, maintaining higher recall when random perturbations are introduced. These results highlight the benefit of adaptive loss weighting for multimodal alignment in low-data regimes.

Paperid: 7250, https://arxiv.org/pdf/2502.15869.pdf

Abstract:
This thesis presents a framework that integrates state-of-the-art generative AI models for real-time creation of three-dimensional (3D) objects in augmented reality (AR) environments. The primary goal is to convert diverse inputs, such as images and speech, into accurate 3D models, enhancing user interaction and immersion. Key components include advanced object detection algorithms, user-friendly interaction techniques, and robust AI models like Shap-E for 3D generation. Leveraging Vision Language Models (VLMs) and Large Language Models (LLMs), the system captures spatial details from images and processes textual information to generate comprehensive 3D objects, seamlessly integrating virtual objects into real-world environments. The framework demonstrates applications across industries such as gaming, education, retail, and interior design. It allows players to create personalized in-game assets, customers to see products in their environments before purchase, and designers to convert real-world objects into 3D models for real-time visualization. A significant contribution is democratizing 3D model creation, making advanced AI tools accessible to a broader audience, fostering creativity and innovation. The framework addresses challenges like handling multilingual inputs, diverse visual data, and complex environments, improving object detection and model generation accuracy, as well as loading 3D models in AR space in real-time. In conclusion, this thesis integrates generative AI and AR for efficient 3D model generation, enhancing accessibility and paving the way for innovative applications and improved user interactions in AR environments.

Paperid: 7251, https://arxiv.org/pdf/2501.17171.pdf

Abstract:
Compositional Zero-Shot Learning (CZSL) aims to recognize subtle differences in meaning or the combination of states and objects through the use of known and unknown concepts during training. Existing methods either focused on prompt configuration or on using prompts to tune the pre-trained Vision-Language model. However, these methods faced challenges in accurately identifying subtle differences in meaning or combining states with objects. To jointly eradicate the above issues and construct an efficient and effective CZSL technique, we suggest a method to improve attribute recognition performance by utilizing diverse Prompt Learning with an Inter/Intra-Modality Fusion Synthesizer in scene understanding involving subtle semantic differences and multiple objects.

Paperid: 7252, https://arxiv.org/pdf/2501.02211.pdf

Abstract:
Vision-Language Models trained on massive collections of human-generated data often reproduce and amplify societal stereotypes. One critical form of stereotyping reproduced by these models is homogeneity bias-the tendency to represent certain groups as more homogeneous than others. We investigate how this bias responds to hyperparameter adjustments in GPT-4, specifically examining sampling temperature and top p which control the randomness of model outputs. By generating stories about individuals from different racial and gender groups and comparing their similarities using vector representations, we assess both bias robustness and its relationship with hyperparameter values. We find that (1) homogeneity bias persists across most hyperparameter configurations, with Black Americans and women being represented more homogeneously than White Americans and men, (2) the relationship between hyperparameters and group representations shows unexpected non-linear patterns, particularly at extreme values, and (3) hyperparameter adjustments affect racial and gender homogeneity bias differently-while increasing temperature or decreasing top p can reduce racial homogeneity bias, these changes show different effects on gender homogeneity bias. Our findings suggest that while hyperparameter tuning may mitigate certain biases to some extent, it cannot serve as a universal solution for addressing homogeneity bias across different social group dimensions.

Paperid: 7253, https://arxiv.org/pdf/2412.18613.pdf

Abstract:
Illusions are entertaining, but they are also a useful diagnostic tool in cognitive science, philosophy, and neuroscience. A typical illusion shows a gap between how something "really is" and how something "appears to be", and this gap helps us understand the mental processing that lead to how something appears to be. Illusions are also useful for investigating artificial systems, and much research has examined whether computational models of perceptions fall prey to the same illusions as people. Here, I invert the standard use of perceptual illusions to examine basic processing errors in current vision language models. I present these models with illusory-illusions, neighbors of common illusions that should not elicit processing errors. These include such things as perfectly reasonable ducks, crooked lines that truly are crooked, circles that seem to have different sizes because they are, in fact, of different sizes, and so on. I show that many current vision language systems mistakenly see these illusion-illusions as illusions. I suggest that such failures are part of broader failures already discussed in the literature.

Paperid: 7254, https://arxiv.org/pdf/2412.16108.pdf

Abstract:
The integration of Large Vision-Language Models (LVLMs) such as OpenAI's GPT-4 Vision into various sectors has marked a significant evolution in the field of artificial intelligence, particularly in the analysis and interpretation of visual data. This paper explores the practical application of GPT-4 Vision in the construction industry, focusing on its capabilities in monitoring and tracking the progress of construction projects. Utilizing high-resolution aerial imagery of construction sites, the study examines how GPT-4 Vision performs detailed scene analysis and tracks developmental changes over time. The findings demonstrate that while GPT-4 Vision is proficient in identifying construction stages, materials, and machinery, it faces challenges with precise object localization and segmentation. Despite these limitations, the potential for future advancements in this technology is considerable. This research not only highlights the current state and opportunities of using LVLMs in construction but also discusses future directions for enhancing the model's utility through domain-specific training and integration with other computer vision techniques and digital twins.

Paperid: 7255, https://arxiv.org/pdf/2412.10908.pdf

Abstract:
Large vision language models (LVLM) are the leading A.I approach for achieving a general visual understanding of the world. Models such as GPT, Claude, Gemini, and LLama can use images to understand and analyze complex visual scenes. 3D objects and shapes are the basic building blocks of the world, recognizing them is a fundamental part of human perception. The goal of this work is to test whether LVLMs truly understand 3D shapes by testing the models ability to identify and match objects of the exact same 3D shapes but with different orientations and materials/textures. A large number of test images were created using CGI with a huge number of highly diverse objects, materials, and scenes. The results of this test show that the ability of such models to match 3D shapes is significantly below humans but much higher than random guesses. Suggesting that the models have gained some abstract understanding of 3D shapes but still trail far beyond humans in this task. Mainly it seems that the models can easily identify the same object with a different orientation as well as matching identical 3D shapes of the same orientation but with different materials and textures. However, when both the object material and orientation are changed, all models perform poorly relative to humans. Code and benchmark are available.

Paperid: 7256, https://arxiv.org/pdf/2411.18727.pdf

Abstract:
Visual communication, dating back to prehistoric cave paintings, is the use of visual elements to convey ideas and information. In today's visually saturated world, effective design demands an understanding of graphic design principles, visual storytelling, human psychology, and the ability to distill complex information into clear visuals. This dissertation explores how recent advancements in vision-language models (VLMs) can be leveraged to automate the creation of effective visual communication designs. Although generative models have made great progress in generating images from text, they still struggle to simplify complex ideas into clear, abstract visuals and are constrained by pixel-based outputs, which lack flexibility for many design tasks. To address these challenges, we constrain the models' operational space and introduce task-specific regularizations. We explore various aspects of visual communication, namely, sketches and visual abstraction, typography, animation, and visual inspiration.

Paperid: 7257, https://arxiv.org/pdf/2411.12319.pdf

Abstract:
Face recognition is a core task in computer vision designed to identify and authenticate individuals by analyzing facial patterns and features. This field intersects with artificial intelligence image processing and machine learning with applications in security authentication and personalization. Traditional approaches in facial recognition focus on capturing facial features like the eyes, nose and mouth and matching these against a database to verify identities. However challenges such as high false positive rates have persisted often due to the similarity among individuals facial features. Recently Contrastive Language Image Pretraining (CLIP) a model developed by OpenAI has shown promising advancements by linking natural language processing with vision tasks allowing it to generalize across modalities. Using CLIP's vision language correspondence and single-shot finetuning the model can achieve lower false positive rates upon deployment without the need of mass facial features extraction. This integration demonstrating CLIP's potential to address persistent issues in face recognition model performance without complicating our training paradigm.

Paperid: 7258, https://arxiv.org/pdf/2411.07265.pdf

Abstract:
This paper presents ViTOC (Vision Transformer and Object-aware Captioner), a novel vision-language model for image captioning that addresses the challenges of accuracy and diversity in generated descriptions. Unlike conventional approaches, ViTOC employs a dual-path architecture based on Vision Transformer and object detector, effectively fusing global visual features and local object information through learnable vectors. The model introduces an innovative object-aware prompting strategy that significantly enhances its capability in handling long-tail data. Experiments on the standard COCO dataset demonstrate that ViTOC outperforms baseline models across all evaluation metrics. Additionally, we propose a reference-free evaluation method based on CLIP to further validate the model's effectiveness. By utilizing pretrained visual model parameters, ViTOC achieves efficient end-to-end training.

Paperid: 7259, https://arxiv.org/pdf/2411.06142.pdf

Abstract:
The recent development of vision language models (VLMs) has led to significant advances in visual-language integration through visual instruction tuning, and they have rapidly evolved in the field of remote sensing image understanding, demonstrating their powerful capabilities. However, existing RSVLMs mainly focus on image-level or frame-level understanding, making it difficult to achieve fine-grained pixel-level visual-language alignment. Additionally, the lack of mask-based instructional data limits their further development. In this paper, we propose a mask-text instruction tuning method called Aquila-plus, which extends the capabilities of RSVLMs to achieve pixel-level visual understanding by incorporating fine-grained mask regions into language instructions. To achieve this, we first meticulously constructed a mask region-text dataset containing 100K samples, and then designed a visual-language model by injecting pixel-level representations into a large language model (LLM). Specifically, Aquila-plus uses a convolutional CLIP as the visual encoder and employs a mask-aware visual extractor to extract precise visual mask features from high-resolution inputs. Experimental results demonstrate that Aquila-plus outperforms existing methods in various region understanding tasks, showcasing its novel capabilities in pixel-level instruction tuning.

Paperid: 7260, https://arxiv.org/pdf/2411.04892.pdf

Abstract:
Large-scale foundation models like CLIP have shown strong zero-shot generalization but struggle with domain shifts, limiting their adaptability. In our work, we introduce \textsc{StyLIP}, a novel domain-agnostic prompt learning strategy for Domain Generalization (DG). StyLIP disentangles visual style and content in CLIP`s vision encoder by using style projectors to learn domain-specific prompt tokens and combining them with content features. Trained contrastively, this approach enables seamless adaptation across domains, outperforming state-of-the-art methods on multiple DG benchmarks. Additionally, we propose AD-CLIP for unsupervised domain adaptation (DA), leveraging CLIP`s frozen vision backbone to learn domain-invariant prompts through image style and content features. By aligning domains in embedding space with entropy minimization, AD-CLIP effectively handles domain shifts, even when only target domain samples are available. Lastly, we outline future work on class discovery using prompt learning for semantic segmentation in remote sensing, focusing on identifying novel or rare classes in unstructured environments. This paves the way for more adaptive and generalizable models in complex, real-world scenarios.

Paperid: 7261, https://arxiv.org/pdf/2410.07880.pdf

Abstract:
This paper investigates the challenges and potential solutions for improving machine learning systems for low-resource languages. State-of-the-art models in natural language processing (NLP), text-to-speech (TTS), speech-to-text (STT), and vision-language models (VLM) rely heavily on large datasets, which are often unavailable for low-resource languages. This research explores key areas such as defining "quality data," developing methods for generating appropriate data and enhancing accessibility to model training. A comprehensive review of current methodologies, including data augmentation, multilingual transfer learning, synthetic data generation, and data selection techniques, highlights both advancements and limitations. Several open research questions are identified, providing a framework for future studies aimed at optimizing data utilization, reducing the required data quantity, and maintaining high-quality model performance. By addressing these challenges, the paper aims to make advanced machine learning models more accessible for low-resource languages, enhancing their utility and impact across various sectors.

Paperid: 7262, https://arxiv.org/pdf/2410.03189.pdf

Abstract:
Prompt tuning for vision-language models such as CLIP involves optimizing the text prompts used to generate image-text pairs for specific downstream tasks. While hand-crafted or template-based prompts are generally applicable to a wider range of unseen classes, they tend to perform poorly in downstream tasks (i.e., seen classes). Learnable soft prompts, on the other hand, often perform well in downstream tasks but lack generalizability. Additionally, prior research has predominantly concentrated on the textual modality, with very few studies attempting to explore the prompt's generalization potential from the visual modality. Keeping these limitations in mind, we investigate how to prompt tuning to obtain both a competitive downstream performance and generalization. The study shows that by treating soft and hand-crafted prompts as dual views of the textual modality, and maximizing their mutual information, we can better ensemble task-specific and general semantic information. Moreover, to generate more expressive prompts, the study introduces a class-wise augmentation from the visual modality, resulting in significant robustness to a wider range of unseen classes. Extensive evaluations on several benchmarks report that the proposed approach achieves competitive results in terms of both task-specific performance and general abilities.